From a9ed3aa9b87ee41e8ab3ff471b1331c154767e45 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Sat, 26 Sep 2026 07:57:59 +1000 Subject: Merge tag 'drm-misc-fixes-2026-09-24' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-fixes A number of fixes: - bridge: - samsung-dsim: fix GPIO lifetime - client: Null pointer dereference fix - imagination: error handling fix, page handling fix - nouveau: fix reference leaks, double-frees, out-of-bounds accesses, use-after-frees, don't reject config without SCDC, a number of workarounds - virtio: fix memory leak, reference leaks, null pointer dereference, add pixel blend mode, cache coherency fix Signed-off-by: Dave Airlie From: Maxime Ripard Link: https://patch.msgid.link/arU22zzqUGDEco1y@houat --- drivers/gpu/drm/msm/.gitignore | 1 + drivers/gpu/drm/msm/Kconfig | 205 + drivers/gpu/drm/msm/Makefile | 252 + drivers/gpu/drm/msm/NOTES | 87 + drivers/gpu/drm/msm/adreno/a2xx_catalog.c | 60 + drivers/gpu/drm/msm/adreno/a2xx_gpu.c | 558 +++ drivers/gpu/drm/msm/adreno/a2xx_gpu.h | 28 + drivers/gpu/drm/msm/adreno/a2xx_gpummu.c | 122 + drivers/gpu/drm/msm/adreno/a3xx_catalog.c | 88 + drivers/gpu/drm/msm/adreno/a3xx_gpu.c | 635 +++ drivers/gpu/drm/msm/adreno/a3xx_gpu.h | 28 + drivers/gpu/drm/msm/adreno/a4xx_catalog.c | 48 + drivers/gpu/drm/msm/adreno/a4xx_gpu.c | 725 +++ drivers/gpu/drm/msm/adreno/a4xx_gpu.h | 25 + drivers/gpu/drm/msm/adreno/a5xx_catalog.c | 153 + drivers/gpu/drm/msm/adreno/a5xx_debugfs.c | 159 + drivers/gpu/drm/msm/adreno/a5xx_gpu.c | 1806 +++++++ drivers/gpu/drm/msm/adreno/a5xx_gpu.h | 177 + drivers/gpu/drm/msm/adreno/a5xx_power.c | 390 ++ drivers/gpu/drm/msm/adreno/a5xx_preempt.c | 329 ++ drivers/gpu/drm/msm/adreno/a6xx_catalog.c | 2540 ++++++++++ drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 2499 ++++++++++ drivers/gpu/drm/msm/adreno/a6xx_gmu.h | 244 + drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 2934 +++++++++++ drivers/gpu/drm/msm/adreno/a6xx_gpu.h | 349 ++ drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c | 2135 ++++++++ drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h | 838 ++++ drivers/gpu/drm/msm/adreno/a6xx_hfi.c | 1102 +++++ drivers/gpu/drm/msm/adreno/a6xx_hfi.h | 338 ++ drivers/gpu/drm/msm/adreno/a6xx_preempt.c | 402 ++ drivers/gpu/drm/msm/adreno/a6xx_preempt.h | 82 + drivers/gpu/drm/msm/adreno/a8xx_gpu.c | 1372 ++++++ drivers/gpu/drm/msm/adreno/a8xx_preempt.c | 259 + drivers/gpu/drm/msm/adreno/adreno_device.c | 433 ++ .../gpu/drm/msm/adreno/adreno_gen7_0_0_snapshot.h | 935 ++++ .../gpu/drm/msm/adreno/adreno_gen7_17_0_snapshot.h | 428 ++ .../gpu/drm/msm/adreno/adreno_gen7_2_0_snapshot.h | 755 +++ .../gpu/drm/msm/adreno/adreno_gen7_9_0_snapshot.h | 1446 ++++++ drivers/gpu/drm/msm/adreno/adreno_gpu.c | 1270 +++++ drivers/gpu/drm/msm/adreno/adreno_gpu.h | 769 +++ .../drm/msm/disp/dpu1/catalog/dpu_10_0_sm8650.h | 451 ++ .../gpu/drm/msm/disp/dpu1/catalog/dpu_10_2_milos.h | 279 ++ .../drm/msm/disp/dpu1/catalog/dpu_12_0_sm8750.h | 492 ++ .../drm/msm/disp/dpu1/catalog/dpu_12_2_glymur.h | 539 +++ .../gpu/drm/msm/disp/dpu1/catalog/dpu_12_4_eliza.h | 363 ++ .../drm/msm/disp/dpu1/catalog/dpu_13_0_kaanapali.h | 490 ++ .../drm/msm/disp/dpu1/catalog/dpu_1_14_msm8937.h | 204 + .../drm/msm/disp/dpu1/catalog/dpu_1_15_msm8917.h | 183 + .../drm/msm/disp/dpu1/catalog/dpu_1_16_msm8953.h | 204 + .../drm/msm/disp/dpu1/catalog/dpu_1_7_msm8996.h | 327 ++ .../drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h | 312 ++ .../gpu/drm/msm/disp/dpu1/catalog/dpu_3_2_sdm660.h | 276 ++ .../gpu/drm/msm/disp/dpu1/catalog/dpu_3_3_sdm630.h | 214 + .../gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h | 332 ++ .../gpu/drm/msm/disp/dpu1/catalog/dpu_4_1_sdm670.h | 151 + .../gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h | 387 ++ .../drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h | 411 ++ .../gpu/drm/msm/disp/dpu1/catalog/dpu_5_2_sm7150.h | 315 ++ .../gpu/drm/msm/disp/dpu1/catalog/dpu_5_3_sm6150.h | 252 + .../gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h | 223 + .../gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h | 386 ++ .../gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h | 217 + .../gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h | 151 + .../gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h | 235 + .../drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h | 144 + .../gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h | 162 + .../gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h | 398 ++ .../gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h | 269 ++ .../drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h | 427 ++ .../gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h | 411 ++ .../drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h | 450 ++ .../gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h | 406 ++ .../drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h | 406 ++ .../drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h | 446 ++ drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.h | 34 + drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c | 519 ++ drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h | 76 + drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c | 1941 ++++++++ drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.h | 272 ++ drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c | 2946 +++++++++++ drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h | 101 + drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h | 356 ++ .../gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c | 784 +++ .../gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c | 791 +++ .../gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c | 694 +++ drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c | 364 ++ drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h | 41 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c | 801 +++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h | 801 +++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cdm.c | 257 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cdm.h | 134 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c | 874 ++++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h | 361 ++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cwb.c | 75 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cwb.h | 69 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c | 216 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h | 83 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c | 396 ++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c | 151 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.h | 112 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c | 815 ++++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h | 75 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c | 672 +++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h | 135 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c | 397 ++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h | 118 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h | 415 ++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c | 67 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.h | 51 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c | 332 ++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h | 126 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c | 757 +++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h | 392 ++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp_v13.c | 323 ++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c | 312 ++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.h | 164 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c | 621 +++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h | 385 ++ drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c | 245 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.h | 111 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c | 262 + drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h | 83 + drivers/gpu/drm/msm/disp/dpu1/dpu_hwio.h | 73 + drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c | 1540 ++++++ drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h | 172 + drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c | 1963 ++++++++ drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h | 78 + drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c | 1053 ++++ drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h | 129 + drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h | 1007 ++++ drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c | 314 ++ drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.h | 49 + drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.c | 145 + drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.h | 32 + drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c | 665 +++ drivers/gpu/drm/msm/disp/mdp4/mdp4_dsi_encoder.c | 153 + drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c | 190 + drivers/gpu/drm/msm/disp/mdp4/mdp4_irq.c | 110 + drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c | 575 +++ drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h | 207 + drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c | 396 ++ drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c | 183 + drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c | 436 ++ drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c | 1084 +++++ drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.h | 136 + drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c | 161 + drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c | 1359 ++++++ drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.c | 662 +++ drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.h | 76 + drivers/gpu/drm/msm/disp/mdp5/mdp5_encoder.c | 307 ++ drivers/gpu/drm/msm/disp/mdp5/mdp5_irq.c | 123 + drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c | 927 ++++ drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.h | 317 ++ drivers/gpu/drm/msm/disp/mdp5/mdp5_mixer.c | 164 + drivers/gpu/drm/msm/disp/mdp5/mdp5_mixer.h | 36 + drivers/gpu/drm/msm/disp/mdp5/mdp5_pipe.c | 171 + drivers/gpu/drm/msm/disp/mdp5/mdp5_pipe.h | 46 + drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c | 1040 ++++ drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.c | 383 ++ drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.h | 88 + drivers/gpu/drm/msm/disp/mdp_format.c | 659 +++ drivers/gpu/drm/msm/disp/mdp_format.h | 81 + drivers/gpu/drm/msm/disp/mdp_kms.c | 138 + drivers/gpu/drm/msm/disp/mdp_kms.h | 126 + drivers/gpu/drm/msm/disp/msm_disp_snapshot.c | 138 + drivers/gpu/drm/msm/disp/msm_disp_snapshot.h | 145 + drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c | 205 + drivers/gpu/drm/msm/dp/dp_audio.c | 377 ++ drivers/gpu/drm/msm/dp/dp_audio.h | 57 + drivers/gpu/drm/msm/dp/dp_aux.c | 729 +++ drivers/gpu/drm/msm/dp/dp_aux.h | 33 + drivers/gpu/drm/msm/dp/dp_ctrl.c | 2813 +++++++++++ drivers/gpu/drm/msm/dp/dp_ctrl.h | 58 + drivers/gpu/drm/msm/dp/dp_debug.c | 240 + drivers/gpu/drm/msm/dp/dp_debug.h | 48 + drivers/gpu/drm/msm/dp/dp_display.c | 1574 ++++++ drivers/gpu/drm/msm/dp/dp_display.h | 46 + drivers/gpu/drm/msm/dp/dp_drm.c | 375 ++ drivers/gpu/drm/msm/dp/dp_drm.h | 36 + drivers/gpu/drm/msm/dp/dp_link.c | 1350 ++++++ drivers/gpu/drm/msm/dp/dp_link.h | 133 + drivers/gpu/drm/msm/dp/dp_panel.c | 729 +++ drivers/gpu/drm/msm/dp/dp_panel.h | 95 + drivers/gpu/drm/msm/dp/dp_reg.h | 368 ++ drivers/gpu/drm/msm/dp/dp_utils.c | 88 + drivers/gpu/drm/msm/dp/dp_utils.h | 36 + drivers/gpu/drm/msm/dsi/dsi.c | 270 ++ drivers/gpu/drm/msm/dsi/dsi.h | 155 + drivers/gpu/drm/msm/dsi/dsi_cfg.c | 364 ++ drivers/gpu/drm/msm/dsi/dsi_cfg.h | 75 + drivers/gpu/drm/msm/dsi/dsi_host.c | 2711 +++++++++++ drivers/gpu/drm/msm/dsi/dsi_manager.c | 621 +++ drivers/gpu/drm/msm/dsi/phy/dsi_phy.c | 871 ++++ drivers/gpu/drm/msm/dsi/phy/dsi_phy.h | 142 + drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c | 1036 ++++ drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c | 1116 +++++ drivers/gpu/drm/msm/dsi/phy/dsi_phy_20nm.c | 147 + drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c | 956 ++++ drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c | 648 +++ drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c | 1566 ++++++ drivers/gpu/drm/msm/hdmi/hdmi.c | 484 ++ drivers/gpu/drm/msm/hdmi/hdmi.h | 269 ++ drivers/gpu/drm/msm/hdmi/hdmi_audio.c | 178 + drivers/gpu/drm/msm/hdmi/hdmi_bridge.c | 512 ++ drivers/gpu/drm/msm/hdmi/hdmi_hdcp.c | 1399 ++++++ drivers/gpu/drm/msm/hdmi/hdmi_hpd.c | 215 + drivers/gpu/drm/msm/hdmi/hdmi_i2c.c | 276 ++ drivers/gpu/drm/msm/hdmi/hdmi_phy.c | 230 + drivers/gpu/drm/msm/hdmi/hdmi_phy_8960.c | 51 + drivers/gpu/drm/msm/hdmi/hdmi_phy_8996.c | 759 +++ drivers/gpu/drm/msm/hdmi/hdmi_phy_8998.c | 763 +++ drivers/gpu/drm/msm/hdmi/hdmi_phy_8x60.c | 141 + drivers/gpu/drm/msm/hdmi/hdmi_phy_8x74.c | 44 + drivers/gpu/drm/msm/hdmi/hdmi_pll_8960.c | 460 ++ drivers/gpu/drm/msm/msm_atomic.c | 324 ++ drivers/gpu/drm/msm/msm_atomic_trace.h | 110 + drivers/gpu/drm/msm/msm_atomic_tracepoints.c | 3 + drivers/gpu/drm/msm/msm_debugfs.c | 414 ++ drivers/gpu/drm/msm/msm_debugfs.h | 14 + drivers/gpu/drm/msm/msm_drv.c | 1166 +++++ drivers/gpu/drm/msm/msm_drv.h | 539 +++ drivers/gpu/drm/msm/msm_dsc_helper.h | 27 + drivers/gpu/drm/msm/msm_fb.c | 249 + drivers/gpu/drm/msm/msm_fbdev.c | 192 + drivers/gpu/drm/msm/msm_fence.c | 201 + drivers/gpu/drm/msm/msm_fence.h | 101 + drivers/gpu/drm/msm/msm_gem.c | 1411 ++++++ drivers/gpu/drm/msm/msm_gem.h | 503 ++ drivers/gpu/drm/msm/msm_gem_prime.c | 134 + drivers/gpu/drm/msm/msm_gem_shrinker.c | 305 ++ drivers/gpu/drm/msm/msm_gem_submit.c | 853 ++++ drivers/gpu/drm/msm/msm_gem_vma.c | 1634 +++++++ drivers/gpu/drm/msm/msm_gpu.c | 1069 ++++ drivers/gpu/drm/msm/msm_gpu.h | 807 ++++ drivers/gpu/drm/msm/msm_gpu_devfreq.c | 375 ++ drivers/gpu/drm/msm/msm_gpu_trace.h | 238 + drivers/gpu/drm/msm/msm_gpu_tracepoints.c | 6 + drivers/gpu/drm/msm/msm_io_utils.c | 161 + drivers/gpu/drm/msm/msm_iommu.c | 825 ++++ drivers/gpu/drm/msm/msm_kms.c | 388 ++ drivers/gpu/drm/msm/msm_kms.h | 243 + drivers/gpu/drm/msm/msm_mdss.c | 581 +++ drivers/gpu/drm/msm/msm_mmu.h | 95 + drivers/gpu/drm/msm/msm_perfcntr.c | 670 +++ drivers/gpu/drm/msm/msm_perfcntr.h | 155 + drivers/gpu/drm/msm/msm_rd.c | 436 ++ drivers/gpu/drm/msm/msm_ringbuffer.c | 144 + drivers/gpu/drm/msm/msm_ringbuffer.h | 144 + drivers/gpu/drm/msm/msm_submitqueue.c | 353 ++ drivers/gpu/drm/msm/msm_syncobj.c | 172 + drivers/gpu/drm/msm/msm_syncobj.h | 37 + drivers/gpu/drm/msm/registers/.gitignore | 4 + drivers/gpu/drm/msm/registers/adreno/a2xx.xml | 1865 +++++++ .../drm/msm/registers/adreno/a2xx_perfcntrs.json | 109 + drivers/gpu/drm/msm/registers/adreno/a3xx.xml | 1749 +++++++ drivers/gpu/drm/msm/registers/adreno/a4xx.xml | 2409 +++++++++ drivers/gpu/drm/msm/registers/adreno/a5xx.xml | 2994 ++++++++++++ .../drm/msm/registers/adreno/a5xx_perfcntrs.json | 128 + drivers/gpu/drm/msm/registers/adreno/a6xx.xml | 5109 ++++++++++++++++++++ .../drm/msm/registers/adreno/a6xx_descriptors.xml | 173 + .../gpu/drm/msm/registers/adreno/a6xx_enums.xml | 432 ++ drivers/gpu/drm/msm/registers/adreno/a6xx_gmu.xml | 271 ++ .../drm/msm/registers/adreno/a6xx_perfcntrs.json | 112 + .../drm/msm/registers/adreno/a6xx_perfcntrs.xml | 600 +++ .../gpu/drm/msm/registers/adreno/a7xx_enums.xml | 216 + .../drm/msm/registers/adreno/a7xx_perfcntrs.json | 228 + .../drm/msm/registers/adreno/a7xx_perfcntrs.xml | 1030 ++++ .../drm/msm/registers/adreno/a8xx_descriptors.xml | 137 + .../gpu/drm/msm/registers/adreno/a8xx_enums.xml | 299 ++ .../drm/msm/registers/adreno/a8xx_perfcntrs.json | 241 + .../drm/msm/registers/adreno/a8xx_perfcntrs.xml | 1929 ++++++++ .../gpu/drm/msm/registers/adreno/adreno_common.xml | 454 ++ .../gpu/drm/msm/registers/adreno/adreno_pm4.xml | 2447 ++++++++++ drivers/gpu/drm/msm/registers/display/dsi.xml | 393 ++ .../gpu/drm/msm/registers/display/dsi_phy_10nm.xml | 102 + .../gpu/drm/msm/registers/display/dsi_phy_14nm.xml | 135 + .../gpu/drm/msm/registers/display/dsi_phy_20nm.xml | 100 + .../gpu/drm/msm/registers/display/dsi_phy_28nm.xml | 180 + .../msm/registers/display/dsi_phy_28nm_8960.xml | 134 + .../gpu/drm/msm/registers/display/dsi_phy_7nm.xml | 261 + drivers/gpu/drm/msm/registers/display/edp.xml | 239 + drivers/gpu/drm/msm/registers/display/hdmi.xml | 1104 +++++ drivers/gpu/drm/msm/registers/display/mdp4.xml | 504 ++ drivers/gpu/drm/msm/registers/display/mdp5.xml | 790 +++ .../gpu/drm/msm/registers/display/mdp_common.xml | 90 + drivers/gpu/drm/msm/registers/display/mdss.xml | 38 + drivers/gpu/drm/msm/registers/display/msm.xml | 32 + drivers/gpu/drm/msm/registers/display/sfpb.xml | 17 + .../gpu/drm/msm/registers/freedreno_copyright.xml | 40 + drivers/gpu/drm/msm/registers/gen_header.py | 1220 +++++ drivers/gpu/drm/msm/registers/rules-fd.xsd | 404 ++ 291 files changed, 145510 insertions(+) create mode 100644 drivers/gpu/drm/msm/.gitignore create mode 100644 drivers/gpu/drm/msm/Kconfig create mode 100644 drivers/gpu/drm/msm/Makefile create mode 100644 drivers/gpu/drm/msm/NOTES create mode 100644 drivers/gpu/drm/msm/adreno/a2xx_catalog.c create mode 100644 drivers/gpu/drm/msm/adreno/a2xx_gpu.c create mode 100644 drivers/gpu/drm/msm/adreno/a2xx_gpu.h create mode 100644 drivers/gpu/drm/msm/adreno/a2xx_gpummu.c create mode 100644 drivers/gpu/drm/msm/adreno/a3xx_catalog.c create mode 100644 drivers/gpu/drm/msm/adreno/a3xx_gpu.c create mode 100644 drivers/gpu/drm/msm/adreno/a3xx_gpu.h create mode 100644 drivers/gpu/drm/msm/adreno/a4xx_catalog.c create mode 100644 drivers/gpu/drm/msm/adreno/a4xx_gpu.c create mode 100644 drivers/gpu/drm/msm/adreno/a4xx_gpu.h create mode 100644 drivers/gpu/drm/msm/adreno/a5xx_catalog.c create mode 100644 drivers/gpu/drm/msm/adreno/a5xx_debugfs.c create mode 100644 drivers/gpu/drm/msm/adreno/a5xx_gpu.c create mode 100644 drivers/gpu/drm/msm/adreno/a5xx_gpu.h create mode 100644 drivers/gpu/drm/msm/adreno/a5xx_power.c create mode 100644 drivers/gpu/drm/msm/adreno/a5xx_preempt.c create mode 100644 drivers/gpu/drm/msm/adreno/a6xx_catalog.c create mode 100644 drivers/gpu/drm/msm/adreno/a6xx_gmu.c create mode 100644 drivers/gpu/drm/msm/adreno/a6xx_gmu.h create mode 100644 drivers/gpu/drm/msm/adreno/a6xx_gpu.c create mode 100644 drivers/gpu/drm/msm/adreno/a6xx_gpu.h create mode 100644 drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c create mode 100644 drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h create mode 100644 drivers/gpu/drm/msm/adreno/a6xx_hfi.c create mode 100644 drivers/gpu/drm/msm/adreno/a6xx_hfi.h create mode 100644 drivers/gpu/drm/msm/adreno/a6xx_preempt.c create mode 100644 drivers/gpu/drm/msm/adreno/a6xx_preempt.h create mode 100644 drivers/gpu/drm/msm/adreno/a8xx_gpu.c create mode 100644 drivers/gpu/drm/msm/adreno/a8xx_preempt.c create mode 100644 drivers/gpu/drm/msm/adreno/adreno_device.c create mode 100644 drivers/gpu/drm/msm/adreno/adreno_gen7_0_0_snapshot.h create mode 100644 drivers/gpu/drm/msm/adreno/adreno_gen7_17_0_snapshot.h create mode 100644 drivers/gpu/drm/msm/adreno/adreno_gen7_2_0_snapshot.h create mode 100644 drivers/gpu/drm/msm/adreno/adreno_gen7_9_0_snapshot.h create mode 100644 drivers/gpu/drm/msm/adreno/adreno_gpu.c create mode 100644 drivers/gpu/drm/msm/adreno/adreno_gpu.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_0_sm8650.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_2_milos.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_0_sm8750.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_2_glymur.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_4_eliza.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_13_0_kaanapali.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_14_msm8937.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_15_msm8917.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_16_msm8953.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_7_msm8996.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_2_sdm660.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_3_sdm630.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_1_sdm670.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_2_sm7150.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_3_sm6150.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cdm.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cdm.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cwb.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cwb.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp_v13.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hwio.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.h create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.c create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.h create mode 100644 drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c create mode 100644 drivers/gpu/drm/msm/disp/mdp4/mdp4_dsi_encoder.c create mode 100644 drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c create mode 100644 drivers/gpu/drm/msm/disp/mdp4/mdp4_irq.c create mode 100644 drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c create mode 100644 drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h create mode 100644 drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c create mode 100644 drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c create mode 100644 drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.h create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.c create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.h create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_encoder.c create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_irq.c create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.h create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_mixer.c create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_mixer.h create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_pipe.c create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_pipe.h create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.c create mode 100644 drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.h create mode 100644 drivers/gpu/drm/msm/disp/mdp_format.c create mode 100644 drivers/gpu/drm/msm/disp/mdp_format.h create mode 100644 drivers/gpu/drm/msm/disp/mdp_kms.c create mode 100644 drivers/gpu/drm/msm/disp/mdp_kms.h create mode 100644 drivers/gpu/drm/msm/disp/msm_disp_snapshot.c create mode 100644 drivers/gpu/drm/msm/disp/msm_disp_snapshot.h create mode 100644 drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c create mode 100644 drivers/gpu/drm/msm/dp/dp_audio.c create mode 100644 drivers/gpu/drm/msm/dp/dp_audio.h create mode 100644 drivers/gpu/drm/msm/dp/dp_aux.c create mode 100644 drivers/gpu/drm/msm/dp/dp_aux.h create mode 100644 drivers/gpu/drm/msm/dp/dp_ctrl.c create mode 100644 drivers/gpu/drm/msm/dp/dp_ctrl.h create mode 100644 drivers/gpu/drm/msm/dp/dp_debug.c create mode 100644 drivers/gpu/drm/msm/dp/dp_debug.h create mode 100644 drivers/gpu/drm/msm/dp/dp_display.c create mode 100644 drivers/gpu/drm/msm/dp/dp_display.h create mode 100644 drivers/gpu/drm/msm/dp/dp_drm.c create mode 100644 drivers/gpu/drm/msm/dp/dp_drm.h create mode 100644 drivers/gpu/drm/msm/dp/dp_link.c create mode 100644 drivers/gpu/drm/msm/dp/dp_link.h create mode 100644 drivers/gpu/drm/msm/dp/dp_panel.c create mode 100644 drivers/gpu/drm/msm/dp/dp_panel.h create mode 100644 drivers/gpu/drm/msm/dp/dp_reg.h create mode 100644 drivers/gpu/drm/msm/dp/dp_utils.c create mode 100644 drivers/gpu/drm/msm/dp/dp_utils.h create mode 100644 drivers/gpu/drm/msm/dsi/dsi.c create mode 100644 drivers/gpu/drm/msm/dsi/dsi.h create mode 100644 drivers/gpu/drm/msm/dsi/dsi_cfg.c create mode 100644 drivers/gpu/drm/msm/dsi/dsi_cfg.h create mode 100644 drivers/gpu/drm/msm/dsi/dsi_host.c create mode 100644 drivers/gpu/drm/msm/dsi/dsi_manager.c create mode 100644 drivers/gpu/drm/msm/dsi/phy/dsi_phy.c create mode 100644 drivers/gpu/drm/msm/dsi/phy/dsi_phy.h create mode 100644 drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c create mode 100644 drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c create mode 100644 drivers/gpu/drm/msm/dsi/phy/dsi_phy_20nm.c create mode 100644 drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c create mode 100644 drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c create mode 100644 drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c create mode 100644 drivers/gpu/drm/msm/hdmi/hdmi.c create mode 100644 drivers/gpu/drm/msm/hdmi/hdmi.h create mode 100644 drivers/gpu/drm/msm/hdmi/hdmi_audio.c create mode 100644 drivers/gpu/drm/msm/hdmi/hdmi_bridge.c create mode 100644 drivers/gpu/drm/msm/hdmi/hdmi_hdcp.c create mode 100644 drivers/gpu/drm/msm/hdmi/hdmi_hpd.c create mode 100644 drivers/gpu/drm/msm/hdmi/hdmi_i2c.c create mode 100644 drivers/gpu/drm/msm/hdmi/hdmi_phy.c create mode 100644 drivers/gpu/drm/msm/hdmi/hdmi_phy_8960.c create mode 100644 drivers/gpu/drm/msm/hdmi/hdmi_phy_8996.c create mode 100644 drivers/gpu/drm/msm/hdmi/hdmi_phy_8998.c create mode 100644 drivers/gpu/drm/msm/hdmi/hdmi_phy_8x60.c create mode 100644 drivers/gpu/drm/msm/hdmi/hdmi_phy_8x74.c create mode 100644 drivers/gpu/drm/msm/hdmi/hdmi_pll_8960.c create mode 100644 drivers/gpu/drm/msm/msm_atomic.c create mode 100644 drivers/gpu/drm/msm/msm_atomic_trace.h create mode 100644 drivers/gpu/drm/msm/msm_atomic_tracepoints.c create mode 100644 drivers/gpu/drm/msm/msm_debugfs.c create mode 100644 drivers/gpu/drm/msm/msm_debugfs.h create mode 100644 drivers/gpu/drm/msm/msm_drv.c create mode 100644 drivers/gpu/drm/msm/msm_drv.h create mode 100644 drivers/gpu/drm/msm/msm_dsc_helper.h create mode 100644 drivers/gpu/drm/msm/msm_fb.c create mode 100644 drivers/gpu/drm/msm/msm_fbdev.c create mode 100644 drivers/gpu/drm/msm/msm_fence.c create mode 100644 drivers/gpu/drm/msm/msm_fence.h create mode 100644 drivers/gpu/drm/msm/msm_gem.c create mode 100644 drivers/gpu/drm/msm/msm_gem.h create mode 100644 drivers/gpu/drm/msm/msm_gem_prime.c create mode 100644 drivers/gpu/drm/msm/msm_gem_shrinker.c create mode 100644 drivers/gpu/drm/msm/msm_gem_submit.c create mode 100644 drivers/gpu/drm/msm/msm_gem_vma.c create mode 100644 drivers/gpu/drm/msm/msm_gpu.c create mode 100644 drivers/gpu/drm/msm/msm_gpu.h create mode 100644 drivers/gpu/drm/msm/msm_gpu_devfreq.c create mode 100644 drivers/gpu/drm/msm/msm_gpu_trace.h create mode 100644 drivers/gpu/drm/msm/msm_gpu_tracepoints.c create mode 100644 drivers/gpu/drm/msm/msm_io_utils.c create mode 100644 drivers/gpu/drm/msm/msm_iommu.c create mode 100644 drivers/gpu/drm/msm/msm_kms.c create mode 100644 drivers/gpu/drm/msm/msm_kms.h create mode 100644 drivers/gpu/drm/msm/msm_mdss.c create mode 100644 drivers/gpu/drm/msm/msm_mmu.h create mode 100644 drivers/gpu/drm/msm/msm_perfcntr.c create mode 100644 drivers/gpu/drm/msm/msm_perfcntr.h create mode 100644 drivers/gpu/drm/msm/msm_rd.c create mode 100644 drivers/gpu/drm/msm/msm_ringbuffer.c create mode 100644 drivers/gpu/drm/msm/msm_ringbuffer.h create mode 100644 drivers/gpu/drm/msm/msm_submitqueue.c create mode 100644 drivers/gpu/drm/msm/msm_syncobj.c create mode 100644 drivers/gpu/drm/msm/msm_syncobj.h create mode 100644 drivers/gpu/drm/msm/registers/.gitignore create mode 100644 drivers/gpu/drm/msm/registers/adreno/a2xx.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/a2xx_perfcntrs.json create mode 100644 drivers/gpu/drm/msm/registers/adreno/a3xx.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/a4xx.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/a5xx.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/a5xx_perfcntrs.json create mode 100644 drivers/gpu/drm/msm/registers/adreno/a6xx.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/a6xx_descriptors.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/a6xx_enums.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/a6xx_gmu.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/a6xx_perfcntrs.json create mode 100644 drivers/gpu/drm/msm/registers/adreno/a6xx_perfcntrs.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/a7xx_enums.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/a7xx_perfcntrs.json create mode 100644 drivers/gpu/drm/msm/registers/adreno/a7xx_perfcntrs.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/a8xx_descriptors.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/a8xx_enums.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/a8xx_perfcntrs.json create mode 100644 drivers/gpu/drm/msm/registers/adreno/a8xx_perfcntrs.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/adreno_common.xml create mode 100644 drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml create mode 100644 drivers/gpu/drm/msm/registers/display/dsi.xml create mode 100644 drivers/gpu/drm/msm/registers/display/dsi_phy_10nm.xml create mode 100644 drivers/gpu/drm/msm/registers/display/dsi_phy_14nm.xml create mode 100644 drivers/gpu/drm/msm/registers/display/dsi_phy_20nm.xml create mode 100644 drivers/gpu/drm/msm/registers/display/dsi_phy_28nm.xml create mode 100644 drivers/gpu/drm/msm/registers/display/dsi_phy_28nm_8960.xml create mode 100644 drivers/gpu/drm/msm/registers/display/dsi_phy_7nm.xml create mode 100644 drivers/gpu/drm/msm/registers/display/edp.xml create mode 100644 drivers/gpu/drm/msm/registers/display/hdmi.xml create mode 100644 drivers/gpu/drm/msm/registers/display/mdp4.xml create mode 100644 drivers/gpu/drm/msm/registers/display/mdp5.xml create mode 100644 drivers/gpu/drm/msm/registers/display/mdp_common.xml create mode 100644 drivers/gpu/drm/msm/registers/display/mdss.xml create mode 100644 drivers/gpu/drm/msm/registers/display/msm.xml create mode 100644 drivers/gpu/drm/msm/registers/display/sfpb.xml create mode 100644 drivers/gpu/drm/msm/registers/freedreno_copyright.xml create mode 100644 drivers/gpu/drm/msm/registers/gen_header.py create mode 100644 drivers/gpu/drm/msm/registers/rules-fd.xsd (limited to 'drivers/gpu/drm/msm') diff --git a/drivers/gpu/drm/msm/.gitignore b/drivers/gpu/drm/msm/.gitignore new file mode 100644 index 000000000..9ab870da8 --- /dev/null +++ b/drivers/gpu/drm/msm/.gitignore @@ -0,0 +1 @@ +generated/ diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig new file mode 100644 index 000000000..ed1c4775d --- /dev/null +++ b/drivers/gpu/drm/msm/Kconfig @@ -0,0 +1,205 @@ +# SPDX-License-Identifier: GPL-2.0-only + +config DRM_MSM + tristate "MSM DRM" + depends on DRM + depends on ARCH_QCOM || SOC_IMX5 || COMPILE_TEST + depends on COMMON_CLK + depends on IOMMU_SUPPORT + depends on OF + depends on QCOM_AOSS_QMP || QCOM_AOSS_QMP=n + depends on QCOM_OCMEM || QCOM_OCMEM=n + depends on QCOM_LLCC || QCOM_LLCC=n + depends on QCOM_COMMAND_DB || QCOM_COMMAND_DB=n + depends on QCOM_SMEM || QCOM_SMEM=n + depends on PM + select IOMMU_IO_PGTABLE + select QCOM_MDT_LOADER if ARCH_QCOM + select REGULATOR + select DRM_EXEC + select DRM_GPUVM + select DRM_SCHED + select SHMEM + select TMPFS + select QCOM_SCM + select QCOM_PAS + select QCOM_UBWC_CONFIG if ARCH_QCOM && QCOM_SMEM + select WANT_DEV_COREDUMP + select SND_SOC_HDMI_CODEC if SND_SOC + select SYNC_FILE + select PM_OPP + select NVMEM + select PM_GENERIC_DOMAINS + select TRACE_GPU_MEM + help + DRM/KMS driver for MSM/snapdragon. + +config DRM_MSM_GPU_STATE + bool + depends on DRM_MSM && (DEBUG_FS || DEV_COREDUMP) + default y + +config DRM_MSM_GPU_SUDO + bool "Enable SUDO flag on submits" + depends on DRM_MSM && EXPERT + default n + help + Enable userspace that has CAP_SYS_RAWIO to submit GPU commands + that are run from RB instead of IB1. This essentially gives + userspace kernel level access, but is useful for firmware + debugging. + + Only use this if you are a driver developer. This should *not* + be enabled for production kernels. If unsure, say N. + +config DRM_MSM_VALIDATE_XML + bool "Validate XML register files against schema" + depends on DRM_MSM && EXPERT + depends on $(success,$(PYTHON3) -c "import lxml") + help + Validate XML files with register definitions against rules-fd schema. + This option is mostly targeting DRM MSM developers. If unsure, say N. + +config DRM_MSM_KMS + def_bool n + depends on DRM_MSM + select DRM_BRIDGE + select DRM_BRIDGE_CONNECTOR + select DRM_CLIENT_SELECTION + select DRM_DISPLAY_HELPER + select DRM_KMS_HELPER + select DRM_PANEL + select DRM_PANEL_BRIDGE + +config DRM_MSM_KMS_FBDEV + def_bool DRM_FBDEV_EMULATION + depends on DRM_MSM_KMS + select FB_SYSMEM_HELPERS + +config DRM_MSM_MDSS + bool + depends on DRM_MSM + default n + +config DRM_MSM_MDP4 + bool "Enable MDP4 support in MSM DRM driver" + depends on DRM_MSM + select DRM_MSM_KMS + default y + help + Compile in support for the Mobile Display Processor v4 (MDP4) in + the MSM DRM driver. It is the older display controller found in + devices using APQ8064/MSM8960/MSM8x60 platforms. + +config DRM_MSM_MDP5 + bool "Enable MDP5 support in MSM DRM driver" + depends on DRM_MSM + select DRM_MSM_MDSS + select DRM_MSM_KMS + default y + help + Compile in support for the Mobile Display Processor v5 (MDP5) in + the MSM DRM driver. It is the display controller found in devices + using e.g. APQ8016/MSM8916/APQ8096/MSM8996/MSM8974/SDM6x0 platforms. + +config DRM_MSM_DPU + bool "Enable DPU support in MSM DRM driver" + depends on DRM_MSM + select DRM_MSM_MDSS + select DRM_MSM_KMS + select DRM_DISPLAY_DSC_HELPER + default y + help + Compile in support for the Display Processing Unit in + the MSM DRM driver. It is the display controller found in devices + using e.g. SDM845 and newer platforms. + +config DRM_MSM_DP + bool "Enable DisplayPort support in MSM DRM driver" + depends on DRM_MSM + depends on DRM_MSM_KMS + select DRM_DISPLAY_HDMI_AUDIO_HELPER + select RATIONAL + select DRM_DISPLAY_DP_AUX_BUS + select DRM_DISPLAY_DP_HELPER + default y + help + Compile in support for DP driver in MSM DRM driver. DP external + display support is enabled through this config option. It can + be primary or secondary display on device. + +config DRM_MSM_DSI + bool "Enable DSI support in MSM DRM driver" + depends on DRM_MSM + depends on DRM_MSM_KMS + select DRM_PANEL + select DRM_MIPI_DSI + select DRM_DISPLAY_DSC_HELPER + default y + help + Choose this option if you have a need for MIPI DSI connector + support. + +config DRM_MSM_DSI_28NM_PHY + bool "Enable DSI 28nm PHY driver in MSM DRM" + depends on DRM_MSM_DSI + default y + help + Choose this option if the 28nm DSI PHY is used on the platform. + +config DRM_MSM_DSI_20NM_PHY + bool "Enable DSI 20nm PHY driver in MSM DRM" + depends on DRM_MSM_DSI + default y + help + Choose this option if the 20nm DSI PHY is used on the platform. + +config DRM_MSM_DSI_28NM_8960_PHY + bool "Enable DSI 28nm 8960 PHY driver in MSM DRM" + depends on DRM_MSM_DSI + default y + help + Choose this option if the 28nm DSI PHY 8960 variant is used on the + platform. + +config DRM_MSM_DSI_14NM_PHY + bool "Enable DSI 14nm PHY driver in MSM DRM (used by MSM8996/APQ8096)" + depends on DRM_MSM_DSI + default y + help + Choose this option if DSI PHY on 8996 is used on the platform. + +config DRM_MSM_DSI_10NM_PHY + bool "Enable DSI 10nm PHY driver in MSM DRM (used by SDM845)" + depends on DRM_MSM_DSI + default y + help + Choose this option if DSI PHY on SDM845 is used on the platform. + +config DRM_MSM_DSI_7NM_PHY + bool "Enable DSI 7nm/5nm/4nm PHY driver in MSM DRM" + depends on DRM_MSM_DSI + default y + help + Choose this option if DSI PHY on SM8150/SM8250/SM8350/SM8450/SM8550/SC7280 + is used on the platform. + +config DRM_MSM_HDMI + bool "Enable HDMI support in MSM DRM driver" + depends on DRM_MSM + depends on DRM_MSM_KMS + default y + select DRM_DISPLAY_HDMI_HELPER + select DRM_DISPLAY_HDMI_STATE_HELPER + help + Compile in support for the HDMI output MSM DRM driver. It can + be a primary or a secondary display on device. Note that this is used + only for the direct HDMI output. If the device outputs HDMI data + through some kind of DSI-to-HDMI bridge, this option can be disabled. + +config DRM_MSM_HDMI_HDCP + bool "Enable HDMI HDCP support in MSM DRM driver" + depends on DRM_MSM && DRM_MSM_HDMI + default y + help + Choose this option to enable HDCP state machine diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile new file mode 100644 index 000000000..0b8f2cafe --- /dev/null +++ b/drivers/gpu/drm/msm/Makefile @@ -0,0 +1,252 @@ +# SPDX-License-Identifier: GPL-2.0 +ccflags-y := -I $(src) +ccflags-y += -I $(obj)/generated +ccflags-y += -I $(src)/disp/dpu1 +ccflags-$(CONFIG_DRM_MSM_DSI) += -I $(src)/dsi +ccflags-$(CONFIG_DRM_MSM_DP) += -I $(src)/dp + +adreno-y := \ + adreno/adreno_device.o \ + adreno/adreno_gpu.o \ + adreno/a2xx_catalog.o \ + adreno/a2xx_gpu.o \ + adreno/a2xx_gpummu.o \ + adreno/a3xx_catalog.o \ + adreno/a3xx_gpu.o \ + adreno/a4xx_catalog.o \ + adreno/a4xx_gpu.o \ + adreno/a5xx_catalog.o \ + adreno/a5xx_gpu.o \ + adreno/a5xx_power.o \ + adreno/a5xx_preempt.o \ + adreno/a6xx_catalog.o \ + adreno/a6xx_gpu.o \ + adreno/a6xx_gmu.o \ + adreno/a6xx_hfi.o \ + adreno/a6xx_preempt.o \ + adreno/a8xx_gpu.o \ + adreno/a8xx_preempt.o \ + +adreno-$(CONFIG_DEBUG_FS) += adreno/a5xx_debugfs.o \ + +adreno-$(CONFIG_DRM_MSM_GPU_STATE) += adreno/a6xx_gpu_state.o + +msm-display-$(CONFIG_DRM_MSM_HDMI) += \ + hdmi/hdmi.o \ + hdmi/hdmi_audio.o \ + hdmi/hdmi_bridge.o \ + hdmi/hdmi_hpd.o \ + hdmi/hdmi_i2c.o \ + hdmi/hdmi_phy.o \ + hdmi/hdmi_phy_8960.o \ + hdmi/hdmi_phy_8996.o \ + hdmi/hdmi_phy_8998.o \ + hdmi/hdmi_phy_8x60.o \ + hdmi/hdmi_phy_8x74.o \ + hdmi/hdmi_pll_8960.o \ + +msm-display-$(CONFIG_DRM_MSM_MDP4) += \ + disp/mdp4/mdp4_crtc.o \ + disp/mdp4/mdp4_dsi_encoder.o \ + disp/mdp4/mdp4_dtv_encoder.o \ + disp/mdp4/mdp4_lcdc_encoder.o \ + disp/mdp4/mdp4_lvds_pll.o \ + disp/mdp4/mdp4_irq.o \ + disp/mdp4/mdp4_kms.o \ + disp/mdp4/mdp4_plane.o \ + +msm-display-$(CONFIG_DRM_MSM_MDP5) += \ + disp/mdp5/mdp5_cfg.o \ + disp/mdp5/mdp5_cmd_encoder.o \ + disp/mdp5/mdp5_ctl.o \ + disp/mdp5/mdp5_crtc.o \ + disp/mdp5/mdp5_encoder.o \ + disp/mdp5/mdp5_irq.o \ + disp/mdp5/mdp5_kms.o \ + disp/mdp5/mdp5_pipe.o \ + disp/mdp5/mdp5_mixer.o \ + disp/mdp5/mdp5_plane.o \ + disp/mdp5/mdp5_smp.o \ + +msm-display-$(CONFIG_DRM_MSM_DPU) += \ + disp/dpu1/dpu_core_perf.o \ + disp/dpu1/dpu_crtc.o \ + disp/dpu1/dpu_encoder.o \ + disp/dpu1/dpu_encoder_phys_cmd.o \ + disp/dpu1/dpu_encoder_phys_vid.o \ + disp/dpu1/dpu_encoder_phys_wb.o \ + disp/dpu1/dpu_formats.o \ + disp/dpu1/dpu_hw_catalog.o \ + disp/dpu1/dpu_hw_cdm.o \ + disp/dpu1/dpu_hw_ctl.o \ + disp/dpu1/dpu_hw_cwb.o \ + disp/dpu1/dpu_hw_dsc.o \ + disp/dpu1/dpu_hw_dsc_1_2.o \ + disp/dpu1/dpu_hw_interrupts.o \ + disp/dpu1/dpu_hw_intf.o \ + disp/dpu1/dpu_hw_lm.o \ + disp/dpu1/dpu_hw_pingpong.o \ + disp/dpu1/dpu_hw_sspp.o \ + disp/dpu1/dpu_hw_sspp_v13.o \ + disp/dpu1/dpu_hw_dspp.o \ + disp/dpu1/dpu_hw_merge3d.o \ + disp/dpu1/dpu_hw_top.o \ + disp/dpu1/dpu_hw_util.o \ + disp/dpu1/dpu_hw_vbif.o \ + disp/dpu1/dpu_hw_wb.o \ + disp/dpu1/dpu_kms.o \ + disp/dpu1/dpu_plane.o \ + disp/dpu1/dpu_rm.o \ + disp/dpu1/dpu_vbif.o \ + disp/dpu1/dpu_writeback.o + +msm-display-$(CONFIG_DRM_MSM_MDSS) += \ + msm_mdss.o \ + +msm-display-$(CONFIG_DRM_MSM_KMS) += \ + disp/mdp_format.o \ + disp/mdp_kms.o \ + disp/msm_disp_snapshot.o \ + disp/msm_disp_snapshot_util.o \ + +msm-y += \ + msm_debugfs.o \ + msm_drv.o \ + msm_fence.o \ + msm_gem.o \ + msm_gem_prime.o \ + msm_gem_shrinker.o \ + msm_gem_submit.o \ + msm_gem_vma.o \ + msm_gpu.o \ + msm_gpu_devfreq.o \ + msm_io_utils.o \ + msm_iommu.o \ + msm_perfcntr.o \ + msm_rd.o \ + msm_ringbuffer.o \ + msm_submitqueue.o \ + msm_syncobj.o \ + msm_gpu_tracepoints.o \ + +msm-$(CONFIG_DRM_MSM_KMS) += \ + msm_atomic.o \ + msm_atomic_tracepoints.o \ + msm_fb.o \ + msm_kms.o \ + +msm-$(CONFIG_DRM_MSM_KMS_FBDEV) += msm_fbdev.o + +msm-display-$(CONFIG_DRM_MSM_DP)+= dp/dp_aux.o \ + dp/dp_ctrl.o \ + dp/dp_debug.o \ + dp/dp_display.o \ + dp/dp_drm.o \ + dp/dp_link.o \ + dp/dp_panel.o \ + dp/dp_audio.o \ + dp/dp_utils.o + +msm-display-$(CONFIG_DRM_MSM_HDMI_HDCP) += hdmi/hdmi_hdcp.o + +msm-display-$(CONFIG_DRM_MSM_DSI) += dsi/dsi.o \ + dsi/dsi_cfg.o \ + dsi/dsi_host.o \ + dsi/dsi_manager.o \ + dsi/phy/dsi_phy.o + +msm-display-$(CONFIG_DRM_MSM_DSI_28NM_PHY) += dsi/phy/dsi_phy_28nm.o +msm-display-$(CONFIG_DRM_MSM_DSI_20NM_PHY) += dsi/phy/dsi_phy_20nm.o +msm-display-$(CONFIG_DRM_MSM_DSI_28NM_8960_PHY) += dsi/phy/dsi_phy_28nm_8960.o +msm-display-$(CONFIG_DRM_MSM_DSI_14NM_PHY) += dsi/phy/dsi_phy_14nm.o +msm-display-$(CONFIG_DRM_MSM_DSI_10NM_PHY) += dsi/phy/dsi_phy_10nm.o +msm-display-$(CONFIG_DRM_MSM_DSI_7NM_PHY) += dsi/phy/dsi_phy_7nm.o + +msm-y += $(adreno-y) +msm-$(CONFIG_DRM_MSM_KMS) += $(msm-display-y) + +obj-$(CONFIG_DRM_MSM) += msm.o + +ifeq (y,$(CONFIG_DRM_MSM_VALIDATE_XML)) + headergen-opts += --validate +else + headergen-opts += --no-validate +endif + +quiet_cmd_headergen = GENHDR $@ + cmd_headergen = mkdir -p $(obj)/generated && $(PYTHON3) $(src)/registers/gen_header.py \ + $(headergen-opts) --rnn $(src)/registers --xml $< c-defines > $@ + +quiet_cmd_headergen_json = GENHDRJSN $@ + cmd_headergen_json = mkdir -p $(obj)/generated && $(PYTHON3) $(src)/registers/gen_header.py \ + $(headergen-opts) --rnn $(src)/registers --xml $(filter %.xml,$^) perfcntrs --json $< > $@ + +$(obj)/generated/%.xml.h: $(src)/registers/adreno/%.xml \ + $(src)/registers/adreno/adreno_common.xml \ + $(src)/registers/adreno/adreno_pm4.xml \ + $(src)/registers/freedreno_copyright.xml \ + $(src)/registers/gen_header.py \ + $(src)/registers/rules-fd.xsd \ + FORCE + $(call if_changed,headergen) + +$(obj)/generated/%.xml.h: $(src)/registers/display/%.xml \ + $(src)/registers/freedreno_copyright.xml \ + $(src)/registers/gen_header.py \ + $(src)/registers/rules-fd.xsd \ + FORCE + $(call if_changed,headergen) + +ADRENO_PERFCNTRS = + +define adreno_perfcntrs +ADRENO_PERFCNTRS += generated/$(1)_perfcntrs.json.c +$$(obj)/generated/$(1)_perfcntrs.json.c: $$(src)/registers/adreno/$(1)_perfcntrs.json \ + $$(src)/registers/adreno/$(2).xml \ + FORCE + $$(call if_changed,headergen_json) +endef + +$(eval $(call adreno_perfcntrs,a2xx,a2xx)) +$(eval $(call adreno_perfcntrs,a5xx,a5xx)) +$(eval $(call adreno_perfcntrs,a6xx,a6xx)) +$(eval $(call adreno_perfcntrs,a7xx,a6xx)) +$(eval $(call adreno_perfcntrs,a8xx,a6xx)) + +adreno-y += $(ADRENO_PERFCNTRS:.c=.o) + +ADRENO_HEADERS = \ + generated/a2xx.xml.h \ + generated/a3xx.xml.h \ + generated/a4xx.xml.h \ + generated/a5xx.xml.h \ + generated/a6xx.xml.h \ + generated/a6xx_descriptors.xml.h \ + generated/a6xx_enums.xml.h \ + generated/a6xx_perfcntrs.xml.h \ + generated/a7xx_enums.xml.h \ + generated/a7xx_perfcntrs.xml.h \ + generated/a8xx_enums.xml.h \ + generated/a6xx_gmu.xml.h \ + generated/adreno_common.xml.h \ + generated/adreno_pm4.xml.h \ + +DISPLAY_HEADERS = \ + generated/dsi_phy_7nm.xml.h \ + generated/dsi_phy_10nm.xml.h \ + generated/dsi_phy_14nm.xml.h \ + generated/dsi_phy_20nm.xml.h \ + generated/dsi_phy_28nm_8960.xml.h \ + generated/dsi_phy_28nm.xml.h \ + generated/dsi.xml.h \ + generated/hdmi.xml.h \ + generated/mdp4.xml.h \ + generated/mdp5.xml.h \ + generated/mdp_common.xml.h \ + generated/mdss.xml.h \ + generated/sfpb.xml.h + +$(addprefix $(obj)/,$(msm-y)): $(addprefix $(obj)/,$(ADRENO_HEADERS)) +$(addprefix $(obj)/,$(msm-display-y)): $(addprefix $(obj)/,$(DISPLAY_HEADERS)) + +targets += $(ADRENO_HEADERS) $(DISPLAY_HEADERS) $(ADRENO_PERFCNTRS) diff --git a/drivers/gpu/drm/msm/NOTES b/drivers/gpu/drm/msm/NOTES new file mode 100644 index 000000000..9c4255b98 --- /dev/null +++ b/drivers/gpu/drm/msm/NOTES @@ -0,0 +1,87 @@ +NOTES about msm drm/kms driver: + +In the current snapdragon SoC's, we have (at least) 3 different +display controller blocks at play: + + MDP3 - ?? seems to be what is on geeksphone peak device + + MDP4 - S3 (APQ8060, touchpad), S4-pro (APQ8064, nexus4 & ifc6410) + + MDP5 - snapdragon 800 + +(I don't have a completely clear picture on which display controller +maps to which part #) + +Plus a handful of blocks around them for HDMI/DSI/etc output. + +And on gpu side of things: + + zero, one, or two 2d cores (z180) + + and either a2xx or a3xx 3d core. + +But, HDMI/DSI/etc blocks seem like they can be shared across multiple +display controller blocks. And I for sure don't want to have to deal +with N different kms devices from xf86-video-freedreno. Plus, it +seems like we can do some clever tricks like use GPU to trigger +pageflip after rendering completes (ie. have the kms/crtc code build +up gpu cmdstream to update scanout and write FLUSH register after). + +So, the approach is one drm driver, with some modularity. Different +'struct msm_kms' implementations, depending on display controller. +And one or more 'struct msm_gpu' for the various different gpu sub- +modules. + +(Second part is not implemented yet. So far this is just basic KMS +driver, and not exposing any custom ioctls to userspace for now.) + +The kms module provides the plane, crtc, and encoder objects, and +loads whatever connectors are appropriate. + +For MDP4, the mapping is: + + plane -> PIPE{RGBn,VGn} \ + crtc -> OVLP{n} + DMA{P,S,E} (??) |-> MDP "device" + encoder -> DTV/LCDC/DSI (within MDP4) / + connector -> HDMI/DSI/etc --> other device(s) + +Since the irq's that drm core mostly cares about are vblank/framedone, +we'll let msm_mdp4_kms provide the irq install/uninstall/etc functions +and treat the MDP4 block's irq as "the" irq. Even though the connectors +may have their own irqs which they install themselves. For this reason +the display controller is the "master" device. + +For MDP5, the mapping is: + + plane -> PIPE{RGBn,VIGn} \ + crtc -> LM (layer mixer) |-> MDP "device" + encoder -> INTF / + connector -> HDMI/DSI/eDP/etc --> other device(s) + +Unlike MDP4, it appears we can get by with a single encoder, rather +than needing a different implementation for DTV, DSI, etc. (Ie. the +register interface is same, just different bases.) + +Also unlike MDP4, with MDP5 all the IRQs for other blocks (HDMI, DSI, +etc) are routed through MDP. + +And finally, MDP5 has this "Shared Memory Pool" (called "SMP"), from +which blocks need to be allocated to the active pipes based on fetch +stride. + +Each connector probably ends up being a separate device, just for the +logistics of finding/mapping io region, irq, etc. Idealy we would +have a better way than just stashing the platform device in a global +(ie. like DT super-node.. but I don't have any snapdragon hw yet that +is using DT). + +Note that so far I've not been able to get any docs on the hw, and it +seems that access to such docs would prevent me from working on the +freedreno gallium driver. So there may be some mistakes in register +names (I had to invent a few, since no sufficient hint was given in +the downstream android fbdev driver), bitfield sizes, etc. My current +state of understanding the registers is given in the envytools rnndb +files at: + + https://github.com/freedreno/envytools/tree/master/rnndb + (the mdp4/hdmi/dsi directories) + +These files are used both for a parser tool (in the same tree) to +parse logged register reads/writes (both from downstream android fbdev +driver, and this driver with register logging enabled), as well as to +generate the register level headers. diff --git a/drivers/gpu/drm/msm/adreno/a2xx_catalog.c b/drivers/gpu/drm/msm/adreno/a2xx_catalog.c new file mode 100644 index 000000000..c1389578e --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a2xx_catalog.c @@ -0,0 +1,60 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013-2014 Red Hat + * Author: Rob Clark + * + * Copyright (c) 2014,2017 The Linux Foundation. All rights reserved. + */ + +#include "adreno_gpu.h" +#include "a2xx_gpu.h" + +static const struct adreno_info a2xx_gpus[] = { + { + .chip_ids = ADRENO_CHIP_IDS(0x02000000), + .family = ADRENO_2XX_GEN1, + .revn = 200, + .fw = { + [ADRENO_FW_PM4] = "yamato_pm4.fw", + [ADRENO_FW_PFP] = "yamato_pfp.fw", + }, + .gmem = SZ_256K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a2xx_gpu_funcs, + }, { /* a200 on i.mx51 has only 128kib gmem */ + .chip_ids = ADRENO_CHIP_IDS(0x02000001), + .family = ADRENO_2XX_GEN1, + .revn = 201, + .fw = { + [ADRENO_FW_PM4] = "yamato_pm4.fw", + [ADRENO_FW_PFP] = "yamato_pfp.fw", + }, + .gmem = SZ_128K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a2xx_gpu_funcs, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x02020000), + .family = ADRENO_2XX_GEN2, + .revn = 220, + .fw = { + [ADRENO_FW_PM4] = "leia_pm4_470.fw", + [ADRENO_FW_PFP] = "leia_pfp_470.fw", + }, + .gmem = SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a2xx_gpu_funcs, + }, { + /* Only for msm8960v3, v2 required a special firmware */ + .chip_ids = ADRENO_CHIP_IDS(0x02020506), + .family = ADRENO_2XX_GEN2, + .revn = 225, + .fw = { + [ADRENO_FW_PM4] = "a225_pm4.fw", + [ADRENO_FW_PFP] = "a225_pfp.fw", + }, + .gmem = SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a2xx_gpu_funcs, + } +}; +DECLARE_ADRENO_GPULIST(a2xx); diff --git a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c new file mode 100644 index 000000000..df4cded91 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c @@ -0,0 +1,558 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2018 The Linux Foundation. All rights reserved. */ + +#include "a2xx_gpu.h" +#include "msm_gem.h" +#include "msm_mmu.h" + +extern bool hang_debug; + +static void a2xx_dump(struct msm_gpu *gpu); +static bool a2xx_idle(struct msm_gpu *gpu); + +static void a2xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) +{ + struct msm_ringbuffer *ring = submit->ring; + unsigned int i; + + for (i = 0; i < submit->nr_cmds; i++) { + switch (submit->cmd[i].type) { + case MSM_SUBMIT_CMD_IB_TARGET_BUF: + /* ignore IB-targets */ + break; + case MSM_SUBMIT_CMD_CTX_RESTORE_BUF: + /* ignore if there has not been a ctx switch: */ + if (ring->cur_ctx_seqno == submit->queue->ctx->seqno) + break; + fallthrough; + case MSM_SUBMIT_CMD_BUF: + OUT_PKT3(ring, CP_INDIRECT_BUFFER_PFD, 2); + OUT_RING(ring, lower_32_bits(submit->cmd[i].iova)); + OUT_RING(ring, submit->cmd[i].size); + OUT_PKT2(ring); + break; + } + } + + OUT_PKT0(ring, REG_AXXX_CP_SCRATCH_REG2, 1); + OUT_RING(ring, submit->seqno); + + /* wait for idle before cache flush/interrupt */ + OUT_PKT3(ring, CP_WAIT_FOR_IDLE, 1); + OUT_RING(ring, 0x00000000); + + OUT_PKT3(ring, CP_EVENT_WRITE, 3); + OUT_RING(ring, CACHE_FLUSH_TS); + OUT_RING(ring, rbmemptr(ring, fence)); + OUT_RING(ring, submit->seqno); + OUT_PKT3(ring, CP_INTERRUPT, 1); + OUT_RING(ring, 0x80000000); + + adreno_flush(gpu, ring, REG_AXXX_CP_RB_WPTR); +} + +static bool a2xx_me_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a2xx_gpu *a2xx_gpu = to_a2xx_gpu(adreno_gpu); + struct msm_ringbuffer *ring = gpu->rb[0]; + + OUT_PKT3(ring, CP_ME_INIT, 18); + + /* All fields present (bits 9:0) */ + OUT_RING(ring, 0x000003ff); + /* Disable/Enable Real-Time Stream processing (present but ignored) */ + OUT_RING(ring, 0x00000000); + /* Enable (2D <-> 3D) implicit synchronization (present but ignored) */ + OUT_RING(ring, 0x00000000); + + OUT_RING(ring, REG_A2XX_RB_SURFACE_INFO - 0x2000); + OUT_RING(ring, REG_A2XX_PA_SC_WINDOW_OFFSET - 0x2000); + OUT_RING(ring, REG_A2XX_VGT_MAX_VTX_INDX - 0x2000); + OUT_RING(ring, REG_A2XX_SQ_PROGRAM_CNTL - 0x2000); + OUT_RING(ring, REG_A2XX_RB_DEPTHCONTROL - 0x2000); + OUT_RING(ring, REG_A2XX_PA_SU_POINT_SIZE - 0x2000); + OUT_RING(ring, REG_A2XX_PA_SC_LINE_CNTL - 0x2000); + OUT_RING(ring, REG_A2XX_PA_SU_POLY_OFFSET_FRONT_SCALE - 0x2000); + + /* Vertex and Pixel Shader Start Addresses in instructions + * (3 DWORDS per instruction) */ + if (adreno_is_a225(adreno_gpu)) + OUT_RING(ring, 0x80000300); + else + OUT_RING(ring, 0x80000180); + /* Maximum Contexts */ + OUT_RING(ring, 0x00000001); + /* Write Confirm Interval and The CP will wait the + * wait_interval * 16 clocks between polling */ + OUT_RING(ring, 0x00000000); + /* NQ and External Memory Swap */ + OUT_RING(ring, 0x00000000); + /* protected mode error checking (0x1f2 is REG_AXXX_CP_INT_CNTL) */ + if (a2xx_gpu->protection_disabled) + OUT_RING(ring, 0x00000000); + else + OUT_RING(ring, 0x200001f2); + /* Disable header dumping and Header dump address */ + OUT_RING(ring, 0x00000000); + /* Header dump size */ + OUT_RING(ring, 0x00000000); + + if (!a2xx_gpu->protection_disabled) { + /* enable protected mode */ + OUT_PKT3(ring, CP_SET_PROTECTED_MODE, 1); + OUT_RING(ring, 1); + } + + adreno_flush(gpu, ring, REG_AXXX_CP_RB_WPTR); + return a2xx_idle(gpu); +} + +static int a2xx_hw_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a2xx_gpu *a2xx_gpu = to_a2xx_gpu(adreno_gpu); + dma_addr_t pt_base, tran_error; + uint32_t *ptr, len; + int i, ret; + + a2xx_gpummu_params(to_msm_vm(gpu->vm)->mmu, &pt_base, &tran_error); + + DBG("%s", gpu->name); + + /* halt ME to avoid ucode upload issues on a20x */ + gpu_write(gpu, REG_AXXX_CP_ME_CNTL, AXXX_CP_ME_CNTL_HALT); + + gpu_write(gpu, REG_A2XX_RBBM_PM_OVERRIDE1, 0xfffffffe); + gpu_write(gpu, REG_A2XX_RBBM_PM_OVERRIDE2, 0xffffffff); + + /* note: kgsl uses 0x00000001 after first reset on a22x */ + gpu_write(gpu, REG_A2XX_RBBM_SOFT_RESET, 0xffffffff); + msleep(30); + gpu_write(gpu, REG_A2XX_RBBM_SOFT_RESET, 0x00000000); + + if (adreno_is_a225(adreno_gpu)) + gpu_write(gpu, REG_A2XX_SQ_FLOW_CONTROL, 0x18000000); + + /* note: kgsl uses 0x0000ffff for a20x */ + gpu_write(gpu, REG_A2XX_RBBM_CNTL, 0x00004442); + + /* MPU: physical range */ + gpu_write(gpu, REG_A2XX_MH_MMU_MPU_BASE, 0x00000000); + gpu_write(gpu, REG_A2XX_MH_MMU_MPU_END, 0xfffff000); + + gpu_write(gpu, REG_A2XX_MH_MMU_CONFIG, A2XX_MH_MMU_CONFIG_MMU_ENABLE | + A2XX_MH_MMU_CONFIG_RB_W_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_CP_W_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_CP_R0_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_CP_R1_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_CP_R2_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_CP_R3_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_CP_R4_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_VGT_R0_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_VGT_R1_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_TC_R_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_PA_W_CLNT_BEHAVIOR(BEH_TRAN_RNG)); + + /* same as parameters in adreno_gpu */ + gpu_write(gpu, REG_A2XX_MH_MMU_VA_RANGE, SZ_16M | + A2XX_MH_MMU_VA_RANGE_NUM_64KB_REGIONS(0xfff)); + + gpu_write(gpu, REG_A2XX_MH_MMU_PT_BASE, pt_base); + gpu_write(gpu, REG_A2XX_MH_MMU_TRAN_ERROR, tran_error); + + gpu_write(gpu, REG_A2XX_MH_MMU_INVALIDATE, + A2XX_MH_MMU_INVALIDATE_INVALIDATE_ALL | + A2XX_MH_MMU_INVALIDATE_INVALIDATE_TC); + + gpu_write(gpu, REG_A2XX_MH_ARBITER_CONFIG, + A2XX_MH_ARBITER_CONFIG_SAME_PAGE_LIMIT(16) | + A2XX_MH_ARBITER_CONFIG_L1_ARB_ENABLE | + A2XX_MH_ARBITER_CONFIG_L1_ARB_HOLD_ENABLE | + A2XX_MH_ARBITER_CONFIG_PAGE_SIZE(1) | + A2XX_MH_ARBITER_CONFIG_TC_REORDER_ENABLE | + A2XX_MH_ARBITER_CONFIG_TC_ARB_HOLD_ENABLE | + A2XX_MH_ARBITER_CONFIG_IN_FLIGHT_LIMIT_ENABLE | + A2XX_MH_ARBITER_CONFIG_IN_FLIGHT_LIMIT(8) | + A2XX_MH_ARBITER_CONFIG_CP_CLNT_ENABLE | + A2XX_MH_ARBITER_CONFIG_VGT_CLNT_ENABLE | + A2XX_MH_ARBITER_CONFIG_TC_CLNT_ENABLE | + A2XX_MH_ARBITER_CONFIG_RB_CLNT_ENABLE | + A2XX_MH_ARBITER_CONFIG_PA_CLNT_ENABLE); + if (!adreno_is_a20x(adreno_gpu)) + gpu_write(gpu, REG_A2XX_MH_CLNT_INTF_CTRL_CONFIG1, 0x00032f07); + + gpu_write(gpu, REG_A2XX_SQ_VS_PROGRAM, 0x00000000); + gpu_write(gpu, REG_A2XX_SQ_PS_PROGRAM, 0x00000000); + + gpu_write(gpu, REG_A2XX_RBBM_PM_OVERRIDE1, 0); /* 0x200 for msm8960? */ + gpu_write(gpu, REG_A2XX_RBBM_PM_OVERRIDE2, 0); /* 0x80/0x1a0 for a22x? */ + + /* note: gsl doesn't set this */ + gpu_write(gpu, REG_A2XX_RBBM_DEBUG, 0x00080000); + + gpu_write(gpu, REG_A2XX_RBBM_INT_CNTL, + A2XX_RBBM_INT_CNTL_RDERR_INT_MASK); + gpu_write(gpu, REG_AXXX_CP_INT_CNTL, + AXXX_CP_INT_CNTL_T0_PACKET_IN_IB_MASK | + AXXX_CP_INT_CNTL_OPCODE_ERROR_MASK | + AXXX_CP_INT_CNTL_PROTECTED_MODE_ERROR_MASK | + AXXX_CP_INT_CNTL_RESERVED_BIT_ERROR_MASK | + AXXX_CP_INT_CNTL_IB_ERROR_MASK | + AXXX_CP_INT_CNTL_IB1_INT_MASK | + AXXX_CP_INT_CNTL_RB_INT_MASK); + gpu_write(gpu, REG_A2XX_SQ_INT_CNTL, 0); + gpu_write(gpu, REG_A2XX_MH_INTERRUPT_MASK, + A2XX_MH_INTERRUPT_MASK_AXI_READ_ERROR | + A2XX_MH_INTERRUPT_MASK_AXI_WRITE_ERROR | + A2XX_MH_INTERRUPT_MASK_MMU_PAGE_FAULT); + + for (i = 3; i <= 5; i++) + if ((SZ_16K << i) == adreno_gpu->info->gmem) + break; + gpu_write(gpu, REG_A2XX_RB_EDRAM_INFO, i); + + ret = adreno_hw_init(gpu); + if (ret) + return ret; + + gpu_write(gpu, REG_AXXX_CP_RB_CNTL, + MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE); + + gpu_write(gpu, REG_AXXX_CP_RB_BASE, lower_32_bits(gpu->rb[0]->iova)); + + /* NOTE: PM4/micro-engine firmware registers look to be the same + * for a2xx and a3xx.. we could possibly push that part down to + * adreno_gpu base class. Or push both PM4 and PFP but + * parameterize the pfp ucode addr/data registers.. + */ + + /* Load PM4: */ + ptr = (uint32_t *)(adreno_gpu->fw[ADRENO_FW_PM4]->data); + len = adreno_gpu->fw[ADRENO_FW_PM4]->size / 4; + DBG("loading PM4 ucode version: %x", ptr[1]); + + /* + * New firmware files seem to have GPU and firmware version in this + * word (0x20xxxx for A200, 0x220xxx for A220, 0x225xxx for A225). + * Older firmware files, which lack protection support, have 0 instead. + */ + if (ptr[1] == 0 && !a2xx_gpu->protection_disabled) { + dev_warn(gpu->dev->dev, + "Legacy firmware detected, disabling protection support\n"); + a2xx_gpu->protection_disabled = true; + } + + gpu_write(gpu, REG_AXXX_CP_DEBUG, + AXXX_CP_DEBUG_MIU_128BIT_WRITE_ENABLE); + gpu_write(gpu, REG_AXXX_CP_ME_RAM_WADDR, 0); + for (i = 1; i < len; i++) + gpu_write(gpu, REG_AXXX_CP_ME_RAM_DATA, ptr[i]); + + /* Load PFP: */ + ptr = (uint32_t *)(adreno_gpu->fw[ADRENO_FW_PFP]->data); + len = adreno_gpu->fw[ADRENO_FW_PFP]->size / 4; + DBG("loading PFP ucode version: %x", ptr[5]); + + gpu_write(gpu, REG_A2XX_CP_PFP_UCODE_ADDR, 0); + for (i = 1; i < len; i++) + gpu_write(gpu, REG_A2XX_CP_PFP_UCODE_DATA, ptr[i]); + + gpu_write(gpu, REG_AXXX_CP_QUEUE_THRESHOLDS, 0x000C0804); + + /* clear ME_HALT to start micro engine */ + gpu_write(gpu, REG_AXXX_CP_ME_CNTL, 0); + + return a2xx_me_init(gpu) ? 0 : -EINVAL; +} + +static void a2xx_recover(struct msm_gpu *gpu) +{ + int i; + + adreno_dump_info(gpu); + + for (i = 0; i < 8; i++) { + printk("CP_SCRATCH_REG%d: %u\n", i, + gpu_read(gpu, REG_AXXX_CP_SCRATCH_REG0 + i)); + } + + /* dump registers before resetting gpu, if enabled: */ + if (hang_debug) + a2xx_dump(gpu); + + gpu_write(gpu, REG_A2XX_RBBM_SOFT_RESET, 1); + gpu_read(gpu, REG_A2XX_RBBM_SOFT_RESET); + gpu_write(gpu, REG_A2XX_RBBM_SOFT_RESET, 0); + adreno_recover(gpu); +} + +static void a2xx_destroy(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a2xx_gpu *a2xx_gpu = to_a2xx_gpu(adreno_gpu); + + DBG("%s", gpu->name); + + adreno_gpu_cleanup(adreno_gpu); + + kfree(a2xx_gpu); +} + +static bool a2xx_idle(struct msm_gpu *gpu) +{ + /* wait for ringbuffer to drain: */ + if (!adreno_idle(gpu, gpu->rb[0])) + return false; + + /* then wait for GPU to finish: */ + if (spin_until(!(gpu_read(gpu, REG_A2XX_RBBM_STATUS) & + A2XX_RBBM_STATUS_GUI_ACTIVE))) { + DRM_ERROR("%s: timeout waiting for GPU to idle!\n", gpu->name); + + /* TODO maybe we need to reset GPU here to recover from hang? */ + return false; + } + + return true; +} + +static irqreturn_t a2xx_irq(struct msm_gpu *gpu) +{ + uint32_t mstatus, status; + + mstatus = gpu_read(gpu, REG_A2XX_MASTER_INT_SIGNAL); + + if (mstatus & A2XX_MASTER_INT_SIGNAL_MH_INT_STAT) { + status = gpu_read(gpu, REG_A2XX_MH_INTERRUPT_STATUS); + + dev_warn(gpu->dev->dev, "MH_INT: %08X\n", status); + dev_warn(gpu->dev->dev, "MMU_PAGE_FAULT: %08X\n", + gpu_read(gpu, REG_A2XX_MH_MMU_PAGE_FAULT)); + + gpu_write(gpu, REG_A2XX_MH_INTERRUPT_CLEAR, status); + } + + if (mstatus & A2XX_MASTER_INT_SIGNAL_CP_INT_STAT) { + status = gpu_read(gpu, REG_AXXX_CP_INT_STATUS); + + /* only RB_INT is expected */ + if (status & ~AXXX_CP_INT_CNTL_RB_INT_MASK) + dev_warn(gpu->dev->dev, "CP_INT: %08X\n", status); + + gpu_write(gpu, REG_AXXX_CP_INT_ACK, status); + } + + if (mstatus & A2XX_MASTER_INT_SIGNAL_RBBM_INT_STAT) { + status = gpu_read(gpu, REG_A2XX_RBBM_INT_STATUS); + + dev_warn(gpu->dev->dev, "RBBM_INT: %08X\n", status); + + gpu_write(gpu, REG_A2XX_RBBM_INT_ACK, status); + } + + msm_gpu_retire(gpu); + + return IRQ_HANDLED; +} + +static const unsigned int a200_registers[] = { + 0x0000, 0x0002, 0x0004, 0x000B, 0x003B, 0x003D, 0x0040, 0x0044, + 0x0046, 0x0047, 0x01C0, 0x01C1, 0x01C3, 0x01C8, 0x01D5, 0x01D9, + 0x01DC, 0x01DD, 0x01EA, 0x01EA, 0x01EE, 0x01F3, 0x01F6, 0x01F7, + 0x01FC, 0x01FF, 0x0391, 0x0392, 0x039B, 0x039E, 0x03B2, 0x03B5, + 0x03B7, 0x03B7, 0x03F8, 0x03FB, 0x0440, 0x0440, 0x0443, 0x0444, + 0x044B, 0x044B, 0x044D, 0x044F, 0x0452, 0x0452, 0x0454, 0x045B, + 0x047F, 0x047F, 0x0578, 0x0587, 0x05C9, 0x05C9, 0x05D0, 0x05D0, + 0x0601, 0x0604, 0x0606, 0x0609, 0x060B, 0x060E, 0x0613, 0x0614, + 0x0A29, 0x0A2B, 0x0A2F, 0x0A31, 0x0A40, 0x0A43, 0x0A45, 0x0A45, + 0x0A4E, 0x0A4F, 0x0C2C, 0x0C2C, 0x0C30, 0x0C30, 0x0C38, 0x0C3C, + 0x0C40, 0x0C40, 0x0C44, 0x0C44, 0x0C80, 0x0C86, 0x0C88, 0x0C94, + 0x0C99, 0x0C9A, 0x0CA4, 0x0CA5, 0x0D00, 0x0D03, 0x0D06, 0x0D06, + 0x0D08, 0x0D0B, 0x0D34, 0x0D35, 0x0DAE, 0x0DC1, 0x0DC8, 0x0DD4, + 0x0DD8, 0x0DD9, 0x0E00, 0x0E00, 0x0E02, 0x0E04, 0x0E17, 0x0E1E, + 0x0EC0, 0x0EC9, 0x0ECB, 0x0ECC, 0x0ED0, 0x0ED0, 0x0ED4, 0x0ED7, + 0x0EE0, 0x0EE2, 0x0F01, 0x0F02, 0x0F0C, 0x0F0C, 0x0F0E, 0x0F12, + 0x0F26, 0x0F2A, 0x0F2C, 0x0F2C, 0x2000, 0x2002, 0x2006, 0x200F, + 0x2080, 0x2082, 0x2100, 0x2109, 0x210C, 0x2114, 0x2180, 0x2184, + 0x21F5, 0x21F7, 0x2200, 0x2208, 0x2280, 0x2283, 0x2293, 0x2294, + 0x2300, 0x2308, 0x2312, 0x2312, 0x2316, 0x231D, 0x2324, 0x2326, + 0x2380, 0x2383, 0x2400, 0x2402, 0x2406, 0x240F, 0x2480, 0x2482, + 0x2500, 0x2509, 0x250C, 0x2514, 0x2580, 0x2584, 0x25F5, 0x25F7, + 0x2600, 0x2608, 0x2680, 0x2683, 0x2693, 0x2694, 0x2700, 0x2708, + 0x2712, 0x2712, 0x2716, 0x271D, 0x2724, 0x2726, 0x2780, 0x2783, + 0x4000, 0x4003, 0x4800, 0x4805, 0x4900, 0x4900, 0x4908, 0x4908, + ~0 /* sentinel */ +}; + +static const unsigned int a220_registers[] = { + 0x0000, 0x0002, 0x0004, 0x000B, 0x003B, 0x003D, 0x0040, 0x0044, + 0x0046, 0x0047, 0x01C0, 0x01C1, 0x01C3, 0x01C8, 0x01D5, 0x01D9, + 0x01DC, 0x01DD, 0x01EA, 0x01EA, 0x01EE, 0x01F3, 0x01F6, 0x01F7, + 0x01FC, 0x01FF, 0x0391, 0x0392, 0x039B, 0x039E, 0x03B2, 0x03B5, + 0x03B7, 0x03B7, 0x03F8, 0x03FB, 0x0440, 0x0440, 0x0443, 0x0444, + 0x044B, 0x044B, 0x044D, 0x044F, 0x0452, 0x0452, 0x0454, 0x045B, + 0x047F, 0x047F, 0x0578, 0x0587, 0x05C9, 0x05C9, 0x05D0, 0x05D0, + 0x0601, 0x0604, 0x0606, 0x0609, 0x060B, 0x060E, 0x0613, 0x0614, + 0x0A29, 0x0A2B, 0x0A2F, 0x0A31, 0x0A40, 0x0A40, 0x0A42, 0x0A43, + 0x0A45, 0x0A45, 0x0A4E, 0x0A4F, 0x0C30, 0x0C30, 0x0C38, 0x0C39, + 0x0C3C, 0x0C3C, 0x0C80, 0x0C81, 0x0C88, 0x0C93, 0x0D00, 0x0D03, + 0x0D05, 0x0D06, 0x0D08, 0x0D0B, 0x0D34, 0x0D35, 0x0DAE, 0x0DC1, + 0x0DC8, 0x0DD4, 0x0DD8, 0x0DD9, 0x0E00, 0x0E00, 0x0E02, 0x0E04, + 0x0E17, 0x0E1E, 0x0EC0, 0x0EC9, 0x0ECB, 0x0ECC, 0x0ED0, 0x0ED0, + 0x0ED4, 0x0ED7, 0x0EE0, 0x0EE2, 0x0F01, 0x0F02, 0x2000, 0x2002, + 0x2006, 0x200F, 0x2080, 0x2082, 0x2100, 0x2102, 0x2104, 0x2109, + 0x210C, 0x2114, 0x2180, 0x2184, 0x21F5, 0x21F7, 0x2200, 0x2202, + 0x2204, 0x2204, 0x2208, 0x2208, 0x2280, 0x2282, 0x2294, 0x2294, + 0x2300, 0x2308, 0x2309, 0x230A, 0x2312, 0x2312, 0x2316, 0x2316, + 0x2318, 0x231D, 0x2324, 0x2326, 0x2380, 0x2383, 0x2400, 0x2402, + 0x2406, 0x240F, 0x2480, 0x2482, 0x2500, 0x2502, 0x2504, 0x2509, + 0x250C, 0x2514, 0x2580, 0x2584, 0x25F5, 0x25F7, 0x2600, 0x2602, + 0x2604, 0x2606, 0x2608, 0x2608, 0x2680, 0x2682, 0x2694, 0x2694, + 0x2700, 0x2708, 0x2712, 0x2712, 0x2716, 0x2716, 0x2718, 0x271D, + 0x2724, 0x2726, 0x2780, 0x2783, 0x4000, 0x4003, 0x4800, 0x4805, + 0x4900, 0x4900, 0x4908, 0x4908, + ~0 /* sentinel */ +}; + +static const unsigned int a225_registers[] = { + 0x0000, 0x0002, 0x0004, 0x000B, 0x003B, 0x003D, 0x0040, 0x0044, + 0x0046, 0x0047, 0x013C, 0x013C, 0x0140, 0x014F, 0x01C0, 0x01C1, + 0x01C3, 0x01C8, 0x01D5, 0x01D9, 0x01DC, 0x01DD, 0x01EA, 0x01EA, + 0x01EE, 0x01F3, 0x01F6, 0x01F7, 0x01FC, 0x01FF, 0x0391, 0x0392, + 0x039B, 0x039E, 0x03B2, 0x03B5, 0x03B7, 0x03B7, 0x03F8, 0x03FB, + 0x0440, 0x0440, 0x0443, 0x0444, 0x044B, 0x044B, 0x044D, 0x044F, + 0x0452, 0x0452, 0x0454, 0x045B, 0x047F, 0x047F, 0x0578, 0x0587, + 0x05C9, 0x05C9, 0x05D0, 0x05D0, 0x0601, 0x0604, 0x0606, 0x0609, + 0x060B, 0x060E, 0x0613, 0x0614, 0x0A29, 0x0A2B, 0x0A2F, 0x0A31, + 0x0A40, 0x0A40, 0x0A42, 0x0A43, 0x0A45, 0x0A45, 0x0A4E, 0x0A4F, + 0x0C01, 0x0C1D, 0x0C30, 0x0C30, 0x0C38, 0x0C39, 0x0C3C, 0x0C3C, + 0x0C80, 0x0C81, 0x0C88, 0x0C93, 0x0D00, 0x0D03, 0x0D05, 0x0D06, + 0x0D08, 0x0D0B, 0x0D34, 0x0D35, 0x0DAE, 0x0DC1, 0x0DC8, 0x0DD4, + 0x0DD8, 0x0DD9, 0x0E00, 0x0E00, 0x0E02, 0x0E04, 0x0E17, 0x0E1E, + 0x0EC0, 0x0EC9, 0x0ECB, 0x0ECC, 0x0ED0, 0x0ED0, 0x0ED4, 0x0ED7, + 0x0EE0, 0x0EE2, 0x0F01, 0x0F02, 0x2000, 0x200F, 0x2080, 0x2082, + 0x2100, 0x2109, 0x210C, 0x2114, 0x2180, 0x2184, 0x21F5, 0x21F7, + 0x2200, 0x2202, 0x2204, 0x2206, 0x2208, 0x2210, 0x2220, 0x2222, + 0x2280, 0x2282, 0x2294, 0x2294, 0x2297, 0x2297, 0x2300, 0x230A, + 0x2312, 0x2312, 0x2315, 0x2316, 0x2318, 0x231D, 0x2324, 0x2326, + 0x2340, 0x2357, 0x2360, 0x2360, 0x2380, 0x2383, 0x2400, 0x240F, + 0x2480, 0x2482, 0x2500, 0x2509, 0x250C, 0x2514, 0x2580, 0x2584, + 0x25F5, 0x25F7, 0x2600, 0x2602, 0x2604, 0x2606, 0x2608, 0x2610, + 0x2620, 0x2622, 0x2680, 0x2682, 0x2694, 0x2694, 0x2697, 0x2697, + 0x2700, 0x270A, 0x2712, 0x2712, 0x2715, 0x2716, 0x2718, 0x271D, + 0x2724, 0x2726, 0x2740, 0x2757, 0x2760, 0x2760, 0x2780, 0x2783, + 0x4000, 0x4003, 0x4800, 0x4806, 0x4808, 0x4808, 0x4900, 0x4900, + 0x4908, 0x4908, + ~0 /* sentinel */ +}; + +/* would be nice to not have to duplicate the _show() stuff with printk(): */ +static void a2xx_dump(struct msm_gpu *gpu) +{ + printk("status: %08x\n", + gpu_read(gpu, REG_A2XX_RBBM_STATUS)); + adreno_dump(gpu); +} + +static struct msm_gpu_state *a2xx_gpu_state_get(struct msm_gpu *gpu) +{ + struct msm_gpu_state *state = kzalloc_obj(*state); + + if (!state) + return ERR_PTR(-ENOMEM); + + adreno_gpu_state_get(gpu, state); + + state->rbbm_status = gpu_read(gpu, REG_A2XX_RBBM_STATUS); + + return state; +} + +static struct drm_gpuvm * +a2xx_create_vm(struct msm_gpu *gpu, struct platform_device *pdev) +{ + struct msm_mmu *mmu = a2xx_gpummu_new(&pdev->dev, gpu); + struct drm_gpuvm *vm; + + vm = msm_gem_vm_create(gpu->dev, mmu, "gpu", SZ_16M, 0xfff * SZ_64K, true); + + if (IS_ERR(vm) && !IS_ERR(mmu)) + mmu->funcs->destroy(mmu); + + return vm; +} + +static u32 a2xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + ring->memptrs->rptr = gpu_read(gpu, REG_AXXX_CP_RB_RPTR); + return ring->memptrs->rptr; +} + +static struct msm_gpu *a2xx_gpu_init(struct drm_device *dev) +{ + struct a2xx_gpu *a2xx_gpu = NULL; + struct adreno_gpu *adreno_gpu; + struct msm_gpu *gpu; + struct msm_drm_private *priv = dev->dev_private; + struct platform_device *pdev = priv->gpu_pdev; + struct adreno_platform_config *config = pdev->dev.platform_data; + int ret; + + if (!pdev) { + dev_err(dev->dev, "no a2xx device\n"); + ret = -ENXIO; + goto fail; + } + + a2xx_gpu = kzalloc_obj(*a2xx_gpu); + if (!a2xx_gpu) { + ret = -ENOMEM; + goto fail; + } + + adreno_gpu = &a2xx_gpu->base; + gpu = &adreno_gpu->base; + + ret = adreno_gpu_init(dev, pdev, adreno_gpu, config->info->funcs, 1); + if (ret) + goto fail; + + if (adreno_is_a20x(adreno_gpu)) + adreno_gpu->registers = a200_registers; + else if (adreno_is_a225(adreno_gpu)) + adreno_gpu->registers = a225_registers; + else + adreno_gpu->registers = a220_registers; + + return gpu; + +fail: + if (a2xx_gpu) + a2xx_destroy(&a2xx_gpu->base.base); + + return ERR_PTR(ret); +} + +const struct adreno_gpu_funcs a2xx_gpu_funcs = { + .base = { + .get_param = adreno_get_param, + .set_param = adreno_set_param, + .hw_init = a2xx_hw_init, + .pm_suspend = msm_gpu_pm_suspend, + .pm_resume = msm_gpu_pm_resume, + .recover = a2xx_recover, + .submit = a2xx_submit, + .active_ring = adreno_active_ring, + .irq = a2xx_irq, + .destroy = a2xx_destroy, +#if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP) + .show = adreno_show, +#endif + .gpu_state_get = a2xx_gpu_state_get, + .gpu_state_put = adreno_gpu_state_put, + .create_vm = a2xx_create_vm, + .get_rptr = a2xx_get_rptr, + }, + .init = a2xx_gpu_init, +}; diff --git a/drivers/gpu/drm/msm/adreno/a2xx_gpu.h b/drivers/gpu/drm/msm/adreno/a2xx_gpu.h new file mode 100644 index 000000000..162ef9895 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a2xx_gpu.h @@ -0,0 +1,28 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2018 The Linux Foundation. All rights reserved. */ + +#ifndef __A2XX_GPU_H__ +#define __A2XX_GPU_H__ + +#include "adreno_gpu.h" + +/* arrg, somehow fb.h is getting pulled in: */ +#undef ROP_COPY +#undef ROP_XOR + +#include "a2xx.xml.h" + +struct a2xx_gpu { + struct adreno_gpu base; + bool pm_enabled; + bool protection_disabled; +}; +#define to_a2xx_gpu(x) container_of(x, struct a2xx_gpu, base) + +extern const struct adreno_gpu_funcs a2xx_gpu_funcs; + +struct msm_mmu *a2xx_gpummu_new(struct device *dev, struct msm_gpu *gpu); +void a2xx_gpummu_params(struct msm_mmu *mmu, dma_addr_t *pt_base, + dma_addr_t *tran_error); + +#endif /* __A2XX_GPU_H__ */ diff --git a/drivers/gpu/drm/msm/adreno/a2xx_gpummu.c b/drivers/gpu/drm/msm/adreno/a2xx_gpummu.c new file mode 100644 index 000000000..e2225c5ba --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a2xx_gpummu.c @@ -0,0 +1,122 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2018 The Linux Foundation. All rights reserved. */ + +#include + +#include "msm_drv.h" +#include "msm_mmu.h" + +#include "adreno_gpu.h" +#include "a2xx_gpu.h" + +#include "a2xx.xml.h" + +struct a2xx_gpummu { + struct msm_mmu base; + struct msm_gpu *gpu; + dma_addr_t pt_base; + uint32_t *table; +}; +#define to_a2xx_gpummu(x) container_of(x, struct a2xx_gpummu, base) + +#define GPUMMU_VA_START SZ_16M +#define GPUMMU_VA_RANGE (0xfff * SZ_64K) +#define GPUMMU_PAGE_SIZE SZ_4K +#define TABLE_SIZE (sizeof(uint32_t) * GPUMMU_VA_RANGE / GPUMMU_PAGE_SIZE) + +static void a2xx_gpummu_detach(struct msm_mmu *mmu) +{ +} + +static int a2xx_gpummu_map(struct msm_mmu *mmu, uint64_t iova, + struct sg_table *sgt, size_t off, size_t len, + int prot) +{ + struct a2xx_gpummu *gpummu = to_a2xx_gpummu(mmu); + unsigned idx = (iova - GPUMMU_VA_START) / GPUMMU_PAGE_SIZE; + struct sg_dma_page_iter dma_iter; + unsigned prot_bits = 0; + + WARN_ON(off != 0); + + if (prot & IOMMU_WRITE) + prot_bits |= 1; + if (prot & IOMMU_READ) + prot_bits |= 2; + + for_each_sgtable_dma_page(sgt, &dma_iter, 0) { + dma_addr_t addr = sg_page_iter_dma_address(&dma_iter); + int i; + + for (i = 0; i < PAGE_SIZE; i += GPUMMU_PAGE_SIZE) + gpummu->table[idx++] = (addr + i) | prot_bits; + } + + /* we can improve by deferring flush for multiple map() */ + gpu_write(gpummu->gpu, REG_A2XX_MH_MMU_INVALIDATE, + A2XX_MH_MMU_INVALIDATE_INVALIDATE_ALL | + A2XX_MH_MMU_INVALIDATE_INVALIDATE_TC); + return 0; +} + +static int a2xx_gpummu_unmap(struct msm_mmu *mmu, uint64_t iova, size_t len) +{ + struct a2xx_gpummu *gpummu = to_a2xx_gpummu(mmu); + unsigned idx = (iova - GPUMMU_VA_START) / GPUMMU_PAGE_SIZE; + unsigned i; + + for (i = 0; i < len / GPUMMU_PAGE_SIZE; i++, idx++) + gpummu->table[idx] = 0; + + gpu_write(gpummu->gpu, REG_A2XX_MH_MMU_INVALIDATE, + A2XX_MH_MMU_INVALIDATE_INVALIDATE_ALL | + A2XX_MH_MMU_INVALIDATE_INVALIDATE_TC); + return 0; +} + +static void a2xx_gpummu_destroy(struct msm_mmu *mmu) +{ + struct a2xx_gpummu *gpummu = to_a2xx_gpummu(mmu); + + dma_free_attrs(mmu->dev, TABLE_SIZE + 32, gpummu->table, gpummu->pt_base, + DMA_ATTR_FORCE_CONTIGUOUS); + + kfree(gpummu); +} + +static const struct msm_mmu_funcs funcs = { + .detach = a2xx_gpummu_detach, + .map = a2xx_gpummu_map, + .unmap = a2xx_gpummu_unmap, + .destroy = a2xx_gpummu_destroy, +}; + +struct msm_mmu *a2xx_gpummu_new(struct device *dev, struct msm_gpu *gpu) +{ + struct a2xx_gpummu *gpummu; + + gpummu = kzalloc_obj(*gpummu); + if (!gpummu) + return ERR_PTR(-ENOMEM); + + gpummu->table = dma_alloc_attrs(dev, TABLE_SIZE + 32, &gpummu->pt_base, + GFP_KERNEL | __GFP_ZERO, DMA_ATTR_FORCE_CONTIGUOUS); + if (!gpummu->table) { + kfree(gpummu); + return ERR_PTR(-ENOMEM); + } + + gpummu->gpu = gpu; + msm_mmu_init(&gpummu->base, dev, &funcs, MSM_MMU_GPUMMU); + + return &gpummu->base; +} + +void a2xx_gpummu_params(struct msm_mmu *mmu, dma_addr_t *pt_base, + dma_addr_t *tran_error) +{ + dma_addr_t base = to_a2xx_gpummu(mmu)->pt_base; + + *pt_base = base; + *tran_error = base + TABLE_SIZE; /* 32-byte aligned */ +} diff --git a/drivers/gpu/drm/msm/adreno/a3xx_catalog.c b/drivers/gpu/drm/msm/adreno/a3xx_catalog.c new file mode 100644 index 000000000..6ae8716fc --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a3xx_catalog.c @@ -0,0 +1,88 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013-2014 Red Hat + * Author: Rob Clark + * + * Copyright (c) 2014,2017 The Linux Foundation. All rights reserved. + */ + +#include "adreno_gpu.h" +#include "a3xx_gpu.h" + +static const struct adreno_info a3xx_gpus[] = { + { + .chip_ids = ADRENO_CHIP_IDS(0x03000512), + .family = ADRENO_3XX, + .fw = { + [ADRENO_FW_PM4] = "a330_pm4.fw", + [ADRENO_FW_PFP] = "a330_pfp.fw", + }, + .gmem = SZ_128K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a3xx_gpu_funcs, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x03000520), + .family = ADRENO_3XX, + .revn = 305, + .fw = { + [ADRENO_FW_PM4] = "a300_pm4.fw", + [ADRENO_FW_PFP] = "a300_pfp.fw", + }, + .gmem = SZ_256K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a3xx_gpu_funcs, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x03000600), + .family = ADRENO_3XX, + .revn = 307, /* because a305c is revn==306 */ + .fw = { + [ADRENO_FW_PM4] = "a300_pm4.fw", + [ADRENO_FW_PFP] = "a300_pfp.fw", + }, + .gmem = SZ_128K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a3xx_gpu_funcs, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x03000620), + .family = ADRENO_3XX, + .revn = 308, + .fw = { + [ADRENO_FW_PM4] = "a300_pm4.fw", + [ADRENO_FW_PFP] = "a300_pfp.fw", + }, + .gmem = SZ_128K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a3xx_gpu_funcs, + }, { + .chip_ids = ADRENO_CHIP_IDS( + 0x03020000, + 0x03020001, + 0x03020002 + ), + .family = ADRENO_3XX, + .revn = 320, + .fw = { + [ADRENO_FW_PM4] = "a300_pm4.fw", + [ADRENO_FW_PFP] = "a300_pfp.fw", + }, + .gmem = SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a3xx_gpu_funcs, + }, { + .chip_ids = ADRENO_CHIP_IDS( + 0x03030000, + 0x03030001, + 0x03030002 + ), + .family = ADRENO_3XX, + .revn = 330, + .fw = { + [ADRENO_FW_PM4] = "a330_pm4.fw", + [ADRENO_FW_PFP] = "a330_pfp.fw", + }, + .gmem = SZ_1M, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a3xx_gpu_funcs, + } +}; +DECLARE_ADRENO_GPULIST(a3xx); diff --git a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c new file mode 100644 index 000000000..0259e60b8 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c @@ -0,0 +1,635 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + * + * Copyright (c) 2014 The Linux Foundation. All rights reserved. + */ + +#include "a3xx_gpu.h" + +#define A3XX_INT0_MASK \ + (A3XX_INT0_RBBM_AHB_ERROR | \ + A3XX_INT0_RBBM_ATB_BUS_OVERFLOW | \ + A3XX_INT0_CP_T0_PACKET_IN_IB | \ + A3XX_INT0_CP_OPCODE_ERROR | \ + A3XX_INT0_CP_RESERVED_BIT_ERROR | \ + A3XX_INT0_CP_HW_FAULT | \ + A3XX_INT0_CP_IB1_INT | \ + A3XX_INT0_CP_IB2_INT | \ + A3XX_INT0_CP_RB_INT | \ + A3XX_INT0_CP_REG_PROTECT_FAULT | \ + A3XX_INT0_CP_AHB_ERROR_HALT | \ + A3XX_INT0_CACHE_FLUSH_TS | \ + A3XX_INT0_UCHE_OOB_ACCESS) + +extern bool hang_debug; + +#define A3XX_VBIF_XIN_HALT_CTRL0_MASK GENMASK(5, 0) + +static void a3xx_dump(struct msm_gpu *gpu); +static bool a3xx_idle(struct msm_gpu *gpu); + +static void a3xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) +{ + struct msm_ringbuffer *ring = submit->ring; + unsigned int i; + + for (i = 0; i < submit->nr_cmds; i++) { + switch (submit->cmd[i].type) { + case MSM_SUBMIT_CMD_IB_TARGET_BUF: + /* ignore IB-targets */ + break; + case MSM_SUBMIT_CMD_CTX_RESTORE_BUF: + /* ignore if there has not been a ctx switch: */ + if (ring->cur_ctx_seqno == submit->queue->ctx->seqno) + break; + fallthrough; + case MSM_SUBMIT_CMD_BUF: + OUT_PKT3(ring, CP_INDIRECT_BUFFER_PFD, 2); + OUT_RING(ring, lower_32_bits(submit->cmd[i].iova)); + OUT_RING(ring, submit->cmd[i].size); + OUT_PKT2(ring); + break; + } + } + + OUT_PKT0(ring, REG_AXXX_CP_SCRATCH_REG2, 1); + OUT_RING(ring, submit->seqno); + + /* Flush HLSQ lazy updates to make sure there is nothing + * pending for indirect loads after the timestamp has + * passed: + */ + OUT_PKT3(ring, CP_EVENT_WRITE, 1); + OUT_RING(ring, HLSQ_FLUSH); + + /* wait for idle before cache flush/interrupt */ + OUT_PKT3(ring, CP_WAIT_FOR_IDLE, 1); + OUT_RING(ring, 0x00000000); + + /* BIT(31) of CACHE_FLUSH_TS triggers CACHE_FLUSH_TS IRQ from GPU */ + OUT_PKT3(ring, CP_EVENT_WRITE, 3); + OUT_RING(ring, CACHE_FLUSH_TS | CP_EVENT_WRITE_0_IRQ); + OUT_RING(ring, rbmemptr(ring, fence)); + OUT_RING(ring, submit->seqno); + +#if 0 + /* Dummy set-constant to trigger context rollover */ + OUT_PKT3(ring, CP_SET_CONSTANT, 2); + OUT_RING(ring, CP_REG(REG_A3XX_HLSQ_CL_KERNEL_GROUP_X_REG)); + OUT_RING(ring, 0x00000000); +#endif + + adreno_flush(gpu, ring, REG_AXXX_CP_RB_WPTR); +} + +static bool a3xx_me_init(struct msm_gpu *gpu) +{ + struct msm_ringbuffer *ring = gpu->rb[0]; + + OUT_PKT3(ring, CP_ME_INIT, 17); + OUT_RING(ring, 0x000003f7); + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000080); + OUT_RING(ring, 0x00000100); + OUT_RING(ring, 0x00000180); + OUT_RING(ring, 0x00006600); + OUT_RING(ring, 0x00000150); + OUT_RING(ring, 0x0000014e); + OUT_RING(ring, 0x00000154); + OUT_RING(ring, 0x00000001); + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + + adreno_flush(gpu, ring, REG_AXXX_CP_RB_WPTR); + return a3xx_idle(gpu); +} + +static int a3xx_hw_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a3xx_gpu *a3xx_gpu = to_a3xx_gpu(adreno_gpu); + uint32_t *ptr, len; + int i, ret; + + DBG("%s", gpu->name); + + if (adreno_is_a305(adreno_gpu)) { + /* Set up 16 deep read/write request queues: */ + gpu_write(gpu, REG_A3XX_VBIF_IN_RD_LIM_CONF0, 0x10101010); + gpu_write(gpu, REG_A3XX_VBIF_IN_RD_LIM_CONF1, 0x10101010); + gpu_write(gpu, REG_A3XX_VBIF_OUT_RD_LIM_CONF0, 0x10101010); + gpu_write(gpu, REG_A3XX_VBIF_OUT_WR_LIM_CONF0, 0x10101010); + gpu_write(gpu, REG_A3XX_VBIF_DDR_OUT_MAX_BURST, 0x0000303); + gpu_write(gpu, REG_A3XX_VBIF_IN_WR_LIM_CONF0, 0x10101010); + gpu_write(gpu, REG_A3XX_VBIF_IN_WR_LIM_CONF1, 0x10101010); + /* Enable WR-REQ: */ + gpu_write(gpu, REG_A3XX_VBIF_GATE_OFF_WRREQ_EN, 0x0000ff); + /* Set up round robin arbitration between both AXI ports: */ + gpu_write(gpu, REG_A3XX_VBIF_ARB_CTL, 0x00000030); + /* Set up AOOO: */ + gpu_write(gpu, REG_A3XX_VBIF_OUT_AXI_AOOO_EN, 0x0000003c); + gpu_write(gpu, REG_A3XX_VBIF_OUT_AXI_AOOO, 0x003c003c); + } else if (adreno_is_a305b(adreno_gpu)) { + gpu_write(gpu, REG_A3XX_VBIF_IN_RD_LIM_CONF0, 0x00181818); + gpu_write(gpu, REG_A3XX_VBIF_IN_WR_LIM_CONF0, 0x00181818); + gpu_write(gpu, REG_A3XX_VBIF_OUT_RD_LIM_CONF0, 0x00000018); + gpu_write(gpu, REG_A3XX_VBIF_OUT_WR_LIM_CONF0, 0x00000018); + gpu_write(gpu, REG_A3XX_VBIF_DDR_OUT_MAX_BURST, 0x00000303); + gpu_write(gpu, REG_A3XX_VBIF_ROUND_ROBIN_QOS_ARB, 0x0003); + } else if (adreno_is_a306(adreno_gpu)) { + gpu_write(gpu, REG_A3XX_VBIF_ROUND_ROBIN_QOS_ARB, 0x0003); + gpu_write(gpu, REG_A3XX_VBIF_OUT_RD_LIM_CONF0, 0x0000000a); + gpu_write(gpu, REG_A3XX_VBIF_OUT_WR_LIM_CONF0, 0x0000000a); + } else if (adreno_is_a306a(adreno_gpu)) { + gpu_write(gpu, REG_A3XX_VBIF_ROUND_ROBIN_QOS_ARB, 0x0003); + gpu_write(gpu, REG_A3XX_VBIF_OUT_RD_LIM_CONF0, 0x00000010); + gpu_write(gpu, REG_A3XX_VBIF_OUT_WR_LIM_CONF0, 0x00000010); + } else if (adreno_is_a320(adreno_gpu)) { + /* Set up 16 deep read/write request queues: */ + gpu_write(gpu, REG_A3XX_VBIF_IN_RD_LIM_CONF0, 0x10101010); + gpu_write(gpu, REG_A3XX_VBIF_IN_RD_LIM_CONF1, 0x10101010); + gpu_write(gpu, REG_A3XX_VBIF_OUT_RD_LIM_CONF0, 0x10101010); + gpu_write(gpu, REG_A3XX_VBIF_OUT_WR_LIM_CONF0, 0x10101010); + gpu_write(gpu, REG_A3XX_VBIF_DDR_OUT_MAX_BURST, 0x0000303); + gpu_write(gpu, REG_A3XX_VBIF_IN_WR_LIM_CONF0, 0x10101010); + gpu_write(gpu, REG_A3XX_VBIF_IN_WR_LIM_CONF1, 0x10101010); + /* Enable WR-REQ: */ + gpu_write(gpu, REG_A3XX_VBIF_GATE_OFF_WRREQ_EN, 0x0000ff); + /* Set up round robin arbitration between both AXI ports: */ + gpu_write(gpu, REG_A3XX_VBIF_ARB_CTL, 0x00000030); + /* Set up AOOO: */ + gpu_write(gpu, REG_A3XX_VBIF_OUT_AXI_AOOO_EN, 0x0000003c); + gpu_write(gpu, REG_A3XX_VBIF_OUT_AXI_AOOO, 0x003c003c); + /* Enable 1K sort: */ + gpu_write(gpu, REG_A3XX_VBIF_ABIT_SORT, 0x000000ff); + gpu_write(gpu, REG_A3XX_VBIF_ABIT_SORT_CONF, 0x000000a4); + + } else if (adreno_is_a330v2(adreno_gpu)) { + /* + * Most of the VBIF registers on 8974v2 have the correct + * values at power on, so we won't modify those if we don't + * need to + */ + /* Enable 1k sort: */ + gpu_write(gpu, REG_A3XX_VBIF_ABIT_SORT, 0x0001003f); + gpu_write(gpu, REG_A3XX_VBIF_ABIT_SORT_CONF, 0x000000a4); + /* Enable WR-REQ: */ + gpu_write(gpu, REG_A3XX_VBIF_GATE_OFF_WRREQ_EN, 0x00003f); + gpu_write(gpu, REG_A3XX_VBIF_DDR_OUT_MAX_BURST, 0x0000303); + /* Set up VBIF_ROUND_ROBIN_QOS_ARB: */ + gpu_write(gpu, REG_A3XX_VBIF_ROUND_ROBIN_QOS_ARB, 0x0003); + + } else if (adreno_is_a330(adreno_gpu)) { + /* Set up 16 deep read/write request queues: */ + gpu_write(gpu, REG_A3XX_VBIF_IN_RD_LIM_CONF0, 0x18181818); + gpu_write(gpu, REG_A3XX_VBIF_IN_RD_LIM_CONF1, 0x18181818); + gpu_write(gpu, REG_A3XX_VBIF_OUT_RD_LIM_CONF0, 0x18181818); + gpu_write(gpu, REG_A3XX_VBIF_OUT_WR_LIM_CONF0, 0x18181818); + gpu_write(gpu, REG_A3XX_VBIF_DDR_OUT_MAX_BURST, 0x0000303); + gpu_write(gpu, REG_A3XX_VBIF_IN_WR_LIM_CONF0, 0x18181818); + gpu_write(gpu, REG_A3XX_VBIF_IN_WR_LIM_CONF1, 0x18181818); + /* Enable WR-REQ: */ + gpu_write(gpu, REG_A3XX_VBIF_GATE_OFF_WRREQ_EN, 0x00003f); + /* Set up round robin arbitration between both AXI ports: */ + gpu_write(gpu, REG_A3XX_VBIF_ARB_CTL, 0x00000030); + /* Set up VBIF_ROUND_ROBIN_QOS_ARB: */ + gpu_write(gpu, REG_A3XX_VBIF_ROUND_ROBIN_QOS_ARB, 0x0001); + /* Set up AOOO: */ + gpu_write(gpu, REG_A3XX_VBIF_OUT_AXI_AOOO_EN, 0x0000003f); + gpu_write(gpu, REG_A3XX_VBIF_OUT_AXI_AOOO, 0x003f003f); + /* Enable 1K sort: */ + gpu_write(gpu, REG_A3XX_VBIF_ABIT_SORT, 0x0001003f); + gpu_write(gpu, REG_A3XX_VBIF_ABIT_SORT_CONF, 0x000000a4); + /* Disable VBIF clock gating. This is to enable AXI running + * higher frequency than GPU: + */ + gpu_write(gpu, REG_A3XX_VBIF_CLKON, 0x00000001); + + } else { + BUG(); + } + + /* Make all blocks contribute to the GPU BUSY perf counter: */ + gpu_write(gpu, REG_A3XX_RBBM_GPU_BUSY_MASKED, 0xffffffff); + + /* Tune the hystersis counters for SP and CP idle detection: */ + gpu_write(gpu, REG_A3XX_RBBM_SP_HYST_CNT, 0x10); + gpu_write(gpu, REG_A3XX_RBBM_WAIT_IDLE_CLOCKS_CTL, 0x10); + + /* Enable the RBBM error reporting bits. This lets us get + * useful information on failure: + */ + gpu_write(gpu, REG_A3XX_RBBM_AHB_CTL0, 0x00000001); + + /* Enable AHB error reporting: */ + gpu_write(gpu, REG_A3XX_RBBM_AHB_CTL1, 0xa6ffffff); + + /* Turn on the power counters: */ + gpu_write(gpu, REG_A3XX_RBBM_RBBM_CTL, 0x00030000); + + /* Turn on hang detection - this spews a lot of useful information + * into the RBBM registers on a hang: + */ + gpu_write(gpu, REG_A3XX_RBBM_INTERFACE_HANG_INT_CTL, 0x00010fff); + + /* Enable 64-byte cacheline size. HW Default is 32-byte (0x000000E0): */ + gpu_write(gpu, REG_A3XX_UCHE_CACHE_MODE_CONTROL_REG, 0x00000001); + + /* Enable Clock gating: */ + if (adreno_is_a305b(adreno_gpu) || + adreno_is_a306(adreno_gpu) || + adreno_is_a306a(adreno_gpu)) + gpu_write(gpu, REG_A3XX_RBBM_CLOCK_CTL, 0xaaaaaaaa); + else if (adreno_is_a320(adreno_gpu)) + gpu_write(gpu, REG_A3XX_RBBM_CLOCK_CTL, 0xbfffffff); + else if (adreno_is_a330v2(adreno_gpu)) + gpu_write(gpu, REG_A3XX_RBBM_CLOCK_CTL, 0xaaaaaaaa); + else if (adreno_is_a330(adreno_gpu)) + gpu_write(gpu, REG_A3XX_RBBM_CLOCK_CTL, 0xbffcffff); + + if (adreno_is_a330v2(adreno_gpu)) + gpu_write(gpu, REG_A3XX_RBBM_GPR0_CTL, 0x05515455); + else if (adreno_is_a330(adreno_gpu)) + gpu_write(gpu, REG_A3XX_RBBM_GPR0_CTL, 0x00000000); + + /* Set the OCMEM base address for A330, etc */ + if (a3xx_gpu->ocmem.hdl) { + gpu_write(gpu, REG_A3XX_RB_GMEM_BASE_ADDR, + (unsigned int)(a3xx_gpu->ocmem.base >> 14)); + } + + /* Turn on performance counters: */ + gpu_write(gpu, REG_A3XX_RBBM_PERFCTR_CTL, 0x01); + + gpu_write(gpu, REG_A3XX_RBBM_INT_0_MASK, A3XX_INT0_MASK); + + ret = adreno_hw_init(gpu); + if (ret) + return ret; + + /* + * Use the default ringbuffer size and block size but disable the RPTR + * shadow + */ + gpu_write(gpu, REG_AXXX_CP_RB_CNTL, + MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE); + + /* Set the ringbuffer address */ + gpu_write(gpu, REG_AXXX_CP_RB_BASE, lower_32_bits(gpu->rb[0]->iova)); + + /* setup access protection: */ + gpu_write(gpu, REG_A3XX_CP_PROTECT_CTRL, 0x00000007); + + /* RBBM registers */ + gpu_write(gpu, REG_A3XX_CP_PROTECT(0), 0x63000040); + gpu_write(gpu, REG_A3XX_CP_PROTECT(1), 0x62000080); + gpu_write(gpu, REG_A3XX_CP_PROTECT(2), 0x600000cc); + gpu_write(gpu, REG_A3XX_CP_PROTECT(3), 0x60000108); + gpu_write(gpu, REG_A3XX_CP_PROTECT(4), 0x64000140); + gpu_write(gpu, REG_A3XX_CP_PROTECT(5), 0x66000400); + + /* CP registers */ + gpu_write(gpu, REG_A3XX_CP_PROTECT(6), 0x65000700); + gpu_write(gpu, REG_A3XX_CP_PROTECT(7), 0x610007d8); + gpu_write(gpu, REG_A3XX_CP_PROTECT(8), 0x620007e0); + gpu_write(gpu, REG_A3XX_CP_PROTECT(9), 0x61001178); + gpu_write(gpu, REG_A3XX_CP_PROTECT(10), 0x64001180); + + /* RB registers */ + gpu_write(gpu, REG_A3XX_CP_PROTECT(11), 0x60003300); + + /* VBIF registers */ + gpu_write(gpu, REG_A3XX_CP_PROTECT(12), 0x6b00c000); + + /* NOTE: PM4/micro-engine firmware registers look to be the same + * for a2xx and a3xx.. we could possibly push that part down to + * adreno_gpu base class. Or push both PM4 and PFP but + * parameterize the pfp ucode addr/data registers.. + */ + + /* Load PM4: */ + ptr = (uint32_t *)(adreno_gpu->fw[ADRENO_FW_PM4]->data); + len = adreno_gpu->fw[ADRENO_FW_PM4]->size / 4; + DBG("loading PM4 ucode version: %x", ptr[1]); + + gpu_write(gpu, REG_AXXX_CP_DEBUG, + AXXX_CP_DEBUG_DYNAMIC_CLK_DISABLE | + AXXX_CP_DEBUG_MIU_128BIT_WRITE_ENABLE); + gpu_write(gpu, REG_AXXX_CP_ME_RAM_WADDR, 0); + for (i = 1; i < len; i++) + gpu_write(gpu, REG_AXXX_CP_ME_RAM_DATA, ptr[i]); + + /* Load PFP: */ + ptr = (uint32_t *)(adreno_gpu->fw[ADRENO_FW_PFP]->data); + len = adreno_gpu->fw[ADRENO_FW_PFP]->size / 4; + DBG("loading PFP ucode version: %x", ptr[5]); + + gpu_write(gpu, REG_A3XX_CP_PFP_UCODE_ADDR, 0); + for (i = 1; i < len; i++) + gpu_write(gpu, REG_A3XX_CP_PFP_UCODE_DATA, ptr[i]); + + /* CP ROQ queue sizes (bytes) - RB:16, ST:16, IB1:32, IB2:64 */ + if (adreno_is_a305(adreno_gpu) || + adreno_is_a306(adreno_gpu) || + adreno_is_a306a(adreno_gpu) || + adreno_is_a320(adreno_gpu)) { + gpu_write(gpu, REG_AXXX_CP_QUEUE_THRESHOLDS, + AXXX_CP_QUEUE_THRESHOLDS_CSQ_IB1_START(2) | + AXXX_CP_QUEUE_THRESHOLDS_CSQ_IB2_START(6) | + AXXX_CP_QUEUE_THRESHOLDS_CSQ_ST_START(14)); + } else if (adreno_is_a330(adreno_gpu) || adreno_is_a305b(adreno_gpu)) { + /* NOTE: this (value take from downstream android driver) + * includes some bits outside of the known bitfields. But + * A330 has this "MERCIU queue" thing too, which might + * explain a new bitfield or reshuffling: + */ + gpu_write(gpu, REG_AXXX_CP_QUEUE_THRESHOLDS, 0x003e2008); + } + + /* clear ME_HALT to start micro engine */ + gpu_write(gpu, REG_AXXX_CP_ME_CNTL, 0); + + return a3xx_me_init(gpu) ? 0 : -EINVAL; +} + +static void a3xx_recover(struct msm_gpu *gpu) +{ + int i; + + adreno_dump_info(gpu); + + for (i = 0; i < 8; i++) { + printk("CP_SCRATCH_REG%d: %u\n", i, + gpu_read(gpu, REG_AXXX_CP_SCRATCH_REG0 + i)); + } + + /* dump registers before resetting gpu, if enabled: */ + if (hang_debug) + a3xx_dump(gpu); + + gpu_write(gpu, REG_A3XX_RBBM_SW_RESET_CMD, 1); + gpu_read(gpu, REG_A3XX_RBBM_SW_RESET_CMD); + gpu_write(gpu, REG_A3XX_RBBM_SW_RESET_CMD, 0); + adreno_recover(gpu); +} + +static void a3xx_destroy(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a3xx_gpu *a3xx_gpu = to_a3xx_gpu(adreno_gpu); + + DBG("%s", gpu->name); + + adreno_gpu_cleanup(adreno_gpu); + + adreno_gpu_ocmem_cleanup(&a3xx_gpu->ocmem); + + kfree(a3xx_gpu); +} + +static bool a3xx_idle(struct msm_gpu *gpu) +{ + /* wait for ringbuffer to drain: */ + if (!adreno_idle(gpu, gpu->rb[0])) + return false; + + /* then wait for GPU to finish: */ + if (spin_until(!(gpu_read(gpu, REG_A3XX_RBBM_STATUS) & + A3XX_RBBM_STATUS_GPU_BUSY))) { + DRM_ERROR("%s: timeout waiting for GPU to idle!\n", gpu->name); + + /* TODO maybe we need to reset GPU here to recover from hang? */ + return false; + } + + return true; +} + +static irqreturn_t a3xx_irq(struct msm_gpu *gpu) +{ + uint32_t status; + + status = gpu_read(gpu, REG_A3XX_RBBM_INT_0_STATUS); + DBG("%s: %08x", gpu->name, status); + + // TODO + + gpu_write(gpu, REG_A3XX_RBBM_INT_CLEAR_CMD, status); + + msm_gpu_retire(gpu); + + return IRQ_HANDLED; +} + +static const unsigned int a3xx_registers[] = { + 0x0000, 0x0002, 0x0010, 0x0012, 0x0018, 0x0018, 0x0020, 0x0027, + 0x0029, 0x002b, 0x002e, 0x0033, 0x0040, 0x0042, 0x0050, 0x005c, + 0x0060, 0x006c, 0x0080, 0x0082, 0x0084, 0x0088, 0x0090, 0x00e5, + 0x00ea, 0x00ed, 0x0100, 0x0100, 0x0110, 0x0123, 0x01c0, 0x01c1, + 0x01c3, 0x01c5, 0x01c7, 0x01c7, 0x01d5, 0x01d9, 0x01dc, 0x01dd, + 0x01ea, 0x01ea, 0x01ee, 0x01f1, 0x01f5, 0x01f5, 0x01fc, 0x01ff, + 0x0440, 0x0440, 0x0443, 0x0443, 0x0445, 0x0445, 0x044d, 0x044f, + 0x0452, 0x0452, 0x0454, 0x046f, 0x047c, 0x047c, 0x047f, 0x047f, + 0x0578, 0x057f, 0x0600, 0x0602, 0x0605, 0x0607, 0x060a, 0x060e, + 0x0612, 0x0614, 0x0c01, 0x0c02, 0x0c06, 0x0c1d, 0x0c3d, 0x0c3f, + 0x0c48, 0x0c4b, 0x0c80, 0x0c80, 0x0c88, 0x0c8b, 0x0ca0, 0x0cb7, + 0x0cc0, 0x0cc1, 0x0cc6, 0x0cc7, 0x0ce4, 0x0ce5, 0x0e00, 0x0e05, + 0x0e0c, 0x0e0c, 0x0e22, 0x0e23, 0x0e41, 0x0e45, 0x0e64, 0x0e65, + 0x0e80, 0x0e82, 0x0e84, 0x0e89, 0x0ea0, 0x0ea1, 0x0ea4, 0x0ea7, + 0x0ec4, 0x0ecb, 0x0ee0, 0x0ee0, 0x0f00, 0x0f01, 0x0f03, 0x0f09, + 0x2040, 0x2040, 0x2044, 0x2044, 0x2048, 0x204d, 0x2068, 0x2069, + 0x206c, 0x206d, 0x2070, 0x2070, 0x2072, 0x2072, 0x2074, 0x2075, + 0x2079, 0x207a, 0x20c0, 0x20d3, 0x20e4, 0x20ef, 0x2100, 0x2109, + 0x210c, 0x210c, 0x210e, 0x210e, 0x2110, 0x2111, 0x2114, 0x2115, + 0x21e4, 0x21e4, 0x21ea, 0x21ea, 0x21ec, 0x21ed, 0x21f0, 0x21f0, + 0x2200, 0x2212, 0x2214, 0x2217, 0x221a, 0x221a, 0x2240, 0x227e, + 0x2280, 0x228b, 0x22c0, 0x22c0, 0x22c4, 0x22ce, 0x22d0, 0x22d8, + 0x22df, 0x22e6, 0x22e8, 0x22e9, 0x22ec, 0x22ec, 0x22f0, 0x22f7, + 0x22ff, 0x22ff, 0x2340, 0x2343, 0x2440, 0x2440, 0x2444, 0x2444, + 0x2448, 0x244d, 0x2468, 0x2469, 0x246c, 0x246d, 0x2470, 0x2470, + 0x2472, 0x2472, 0x2474, 0x2475, 0x2479, 0x247a, 0x24c0, 0x24d3, + 0x24e4, 0x24ef, 0x2500, 0x2509, 0x250c, 0x250c, 0x250e, 0x250e, + 0x2510, 0x2511, 0x2514, 0x2515, 0x25e4, 0x25e4, 0x25ea, 0x25ea, + 0x25ec, 0x25ed, 0x25f0, 0x25f0, 0x2600, 0x2612, 0x2614, 0x2617, + 0x261a, 0x261a, 0x2640, 0x267e, 0x2680, 0x268b, 0x26c0, 0x26c0, + 0x26c4, 0x26ce, 0x26d0, 0x26d8, 0x26df, 0x26e6, 0x26e8, 0x26e9, + 0x26ec, 0x26ec, 0x26f0, 0x26f7, 0x26ff, 0x26ff, 0x2740, 0x2743, + 0x300c, 0x300e, 0x301c, 0x301d, 0x302a, 0x302a, 0x302c, 0x302d, + 0x3030, 0x3031, 0x3034, 0x3036, 0x303c, 0x303c, 0x305e, 0x305f, + ~0 /* sentinel */ +}; + +/* would be nice to not have to duplicate the _show() stuff with printk(): */ +static void a3xx_dump(struct msm_gpu *gpu) +{ + printk("status: %08x\n", + gpu_read(gpu, REG_A3XX_RBBM_STATUS)); + adreno_dump(gpu); +} + +static struct msm_gpu_state *a3xx_gpu_state_get(struct msm_gpu *gpu) +{ + struct msm_gpu_state *state = kzalloc_obj(*state); + + if (!state) + return ERR_PTR(-ENOMEM); + + adreno_gpu_state_get(gpu, state); + + state->rbbm_status = gpu_read(gpu, REG_A3XX_RBBM_STATUS); + + return state; +} + +static u64 a3xx_gpu_busy(struct msm_gpu *gpu, unsigned long *out_sample_rate) +{ + u64 busy_cycles; + + busy_cycles = gpu_read64(gpu, REG_A3XX_RBBM_PERFCTR_RBBM_1_LO); + *out_sample_rate = clk_get_rate(gpu->core_clk); + + return busy_cycles; +} + +static int a3xx_vbif_halt(struct msm_gpu *gpu) +{ + u32 ack; + int ret; + + gpu_write(gpu, REG_A3XX_VBIF_XIN_HALT_CTRL0, + A3XX_VBIF_XIN_HALT_CTRL0_MASK); + ret = spin_until(((ack = gpu_read(gpu, REG_A3XX_VBIF_XIN_HALT_CTRL1)) & + A3XX_VBIF_XIN_HALT_CTRL0_MASK) == + A3XX_VBIF_XIN_HALT_CTRL0_MASK); + gpu_write(gpu, REG_A3XX_VBIF_XIN_HALT_CTRL0, 0); + + if (ret) + return -EBUSY; + + return 0; +} + +static int a3xx_pm_suspend(struct msm_gpu *gpu) +{ + int ret; + + if (!a3xx_idle(gpu)) + return -EBUSY; + + ret = a3xx_vbif_halt(gpu); + if (ret) + return ret; + + return msm_gpu_pm_suspend(gpu); +} + +static u32 a3xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + ring->memptrs->rptr = gpu_read(gpu, REG_AXXX_CP_RB_RPTR); + return ring->memptrs->rptr; +} + +static struct msm_gpu *a3xx_gpu_init(struct drm_device *dev) +{ + struct a3xx_gpu *a3xx_gpu = NULL; + struct adreno_gpu *adreno_gpu; + struct msm_gpu *gpu; + struct msm_drm_private *priv = dev->dev_private; + struct platform_device *pdev = priv->gpu_pdev; + struct adreno_platform_config *config = pdev->dev.platform_data; + struct icc_path *ocmem_icc_path; + struct icc_path *icc_path; + int ret; + + if (!pdev) { + DRM_DEV_ERROR(dev->dev, "no a3xx device\n"); + ret = -ENXIO; + goto fail; + } + + a3xx_gpu = kzalloc_obj(*a3xx_gpu); + if (!a3xx_gpu) { + ret = -ENOMEM; + goto fail; + } + + adreno_gpu = &a3xx_gpu->base; + gpu = &adreno_gpu->base; + + adreno_gpu->registers = a3xx_registers; + + ret = adreno_gpu_init(dev, pdev, adreno_gpu, config->info->funcs, 1); + if (ret) + goto fail; + + /* if needed, allocate gmem: */ + if (adreno_is_a330(adreno_gpu) || adreno_is_a305b(adreno_gpu)) { + ret = adreno_gpu_ocmem_init(&adreno_gpu->base.pdev->dev, + adreno_gpu, &a3xx_gpu->ocmem); + if (ret) + goto fail; + } + + icc_path = devm_of_icc_get(&pdev->dev, "gfx-mem"); + if (IS_ERR(icc_path)) { + ret = PTR_ERR(icc_path); + goto fail; + } + + ocmem_icc_path = devm_of_icc_get(&pdev->dev, "ocmem"); + if (IS_ERR(ocmem_icc_path)) { + ret = PTR_ERR(ocmem_icc_path); + /* allow -ENODATA, ocmem icc is optional */ + if (ret != -ENODATA) + goto fail; + ocmem_icc_path = NULL; + } + + + /* + * Set the ICC path to maximum speed for now by multiplying the fastest + * frequency by the bus width (8). We'll want to scale this later on to + * improve battery life. + */ + icc_set_bw(icc_path, 0, Bps_to_icc(gpu->fast_rate) * 8); + icc_set_bw(ocmem_icc_path, 0, Bps_to_icc(gpu->fast_rate) * 8); + + return gpu; + +fail: + if (a3xx_gpu) + a3xx_destroy(&a3xx_gpu->base.base); + + return ERR_PTR(ret); +} + +const struct adreno_gpu_funcs a3xx_gpu_funcs = { + .base = { + .get_param = adreno_get_param, + .set_param = adreno_set_param, + .hw_init = a3xx_hw_init, + .pm_suspend = a3xx_pm_suspend, + .pm_resume = msm_gpu_pm_resume, + .recover = a3xx_recover, + .submit = a3xx_submit, + .active_ring = adreno_active_ring, + .irq = a3xx_irq, + .destroy = a3xx_destroy, +#if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP) + .show = adreno_show, +#endif + .gpu_busy = a3xx_gpu_busy, + .gpu_state_get = a3xx_gpu_state_get, + .gpu_state_put = adreno_gpu_state_put, + .create_vm = adreno_create_vm, + .get_rptr = a3xx_get_rptr, + }, + .init = a3xx_gpu_init, +}; diff --git a/drivers/gpu/drm/msm/adreno/a3xx_gpu.h b/drivers/gpu/drm/msm/adreno/a3xx_gpu.h new file mode 100644 index 000000000..3d4ec9dbd --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a3xx_gpu.h @@ -0,0 +1,28 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#ifndef __A3XX_GPU_H__ +#define __A3XX_GPU_H__ + +#include "adreno_gpu.h" + +/* arrg, somehow fb.h is getting pulled in: */ +#undef ROP_COPY +#undef ROP_XOR + +#include "a3xx.xml.h" + +struct a3xx_gpu { + struct adreno_gpu base; + + /* if OCMEM is used for GMEM: */ + struct adreno_ocmem ocmem; +}; +#define to_a3xx_gpu(x) container_of(x, struct a3xx_gpu, base) + +extern const struct adreno_gpu_funcs a3xx_gpu_funcs; + +#endif /* __A3XX_GPU_H__ */ diff --git a/drivers/gpu/drm/msm/adreno/a4xx_catalog.c b/drivers/gpu/drm/msm/adreno/a4xx_catalog.c new file mode 100644 index 000000000..9192586f7 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a4xx_catalog.c @@ -0,0 +1,48 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013-2014 Red Hat + * Author: Rob Clark + * + * Copyright (c) 2014,2017 The Linux Foundation. All rights reserved. + */ + +#include "adreno_gpu.h" +#include "a4xx_gpu.h" + +static const struct adreno_info a4xx_gpus[] = { + { + .chip_ids = ADRENO_CHIP_IDS(0x04000500), + .family = ADRENO_4XX, + .revn = 405, + .fw = { + [ADRENO_FW_PM4] = "a420_pm4.fw", + [ADRENO_FW_PFP] = "a420_pfp.fw", + }, + .gmem = SZ_256K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a4xx_gpu_funcs, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x04020000), + .family = ADRENO_4XX, + .revn = 420, + .fw = { + [ADRENO_FW_PM4] = "a420_pm4.fw", + [ADRENO_FW_PFP] = "a420_pfp.fw", + }, + .gmem = (SZ_1M + SZ_512K), + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a4xx_gpu_funcs, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x04030002), + .family = ADRENO_4XX, + .revn = 430, + .fw = { + [ADRENO_FW_PM4] = "a420_pm4.fw", + [ADRENO_FW_PFP] = "a420_pfp.fw", + }, + .gmem = (SZ_1M + SZ_512K), + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a4xx_gpu_funcs, + } +}; +DECLARE_ADRENO_GPULIST(a4xx); diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c new file mode 100644 index 000000000..6392126f4 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c @@ -0,0 +1,725 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2014 The Linux Foundation. All rights reserved. + */ +#include "a4xx_gpu.h" + +#define A4XX_INT0_MASK \ + (A4XX_INT0_RBBM_AHB_ERROR | \ + A4XX_INT0_RBBM_ATB_BUS_OVERFLOW | \ + A4XX_INT0_CP_T0_PACKET_IN_IB | \ + A4XX_INT0_CP_OPCODE_ERROR | \ + A4XX_INT0_CP_RESERVED_BIT_ERROR | \ + A4XX_INT0_CP_HW_FAULT | \ + A4XX_INT0_CP_IB1_INT | \ + A4XX_INT0_CP_IB2_INT | \ + A4XX_INT0_CP_RB_INT | \ + A4XX_INT0_CP_REG_PROTECT_FAULT | \ + A4XX_INT0_CP_AHB_ERROR_HALT | \ + A4XX_INT0_CACHE_FLUSH_TS | \ + A4XX_INT0_UCHE_OOB_ACCESS) + +extern bool hang_debug; +static void a4xx_dump(struct msm_gpu *gpu); +static bool a4xx_idle(struct msm_gpu *gpu); + +static void a4xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) +{ + struct msm_ringbuffer *ring = submit->ring; + unsigned int i; + + for (i = 0; i < submit->nr_cmds; i++) { + switch (submit->cmd[i].type) { + case MSM_SUBMIT_CMD_IB_TARGET_BUF: + /* ignore IB-targets */ + break; + case MSM_SUBMIT_CMD_CTX_RESTORE_BUF: + /* ignore if there has not been a ctx switch: */ + if (ring->cur_ctx_seqno == submit->queue->ctx->seqno) + break; + fallthrough; + case MSM_SUBMIT_CMD_BUF: + OUT_PKT3(ring, CP_INDIRECT_BUFFER_PFE, 2); + OUT_RING(ring, lower_32_bits(submit->cmd[i].iova)); + OUT_RING(ring, submit->cmd[i].size); + OUT_PKT2(ring); + break; + } + } + + OUT_PKT0(ring, REG_AXXX_CP_SCRATCH_REG2, 1); + OUT_RING(ring, submit->seqno); + + /* Flush HLSQ lazy updates to make sure there is nothing + * pending for indirect loads after the timestamp has + * passed: + */ + OUT_PKT3(ring, CP_EVENT_WRITE, 1); + OUT_RING(ring, HLSQ_FLUSH); + + /* wait for idle before cache flush/interrupt */ + OUT_PKT3(ring, CP_WAIT_FOR_IDLE, 1); + OUT_RING(ring, 0x00000000); + + /* BIT(31) of CACHE_FLUSH_TS triggers CACHE_FLUSH_TS IRQ from GPU */ + OUT_PKT3(ring, CP_EVENT_WRITE, 3); + OUT_RING(ring, CACHE_FLUSH_TS | CP_EVENT_WRITE_0_IRQ); + OUT_RING(ring, rbmemptr(ring, fence)); + OUT_RING(ring, submit->seqno); + + adreno_flush(gpu, ring, REG_A4XX_CP_RB_WPTR); +} + +/* + * a4xx_enable_hwcg() - Program the clock control registers + * @device: The adreno device pointer + */ +static void a4xx_enable_hwcg(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + unsigned int i; + for (i = 0; i < 4; i++) + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL_TP(i), 0x02222202); + for (i = 0; i < 4; i++) + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL2_TP(i), 0x00002222); + for (i = 0; i < 4; i++) + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_HYST_TP(i), 0x0E739CE7); + for (i = 0; i < 4; i++) + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_DELAY_TP(i), 0x00111111); + for (i = 0; i < 4; i++) + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL_SP(i), 0x22222222); + for (i = 0; i < 4; i++) + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL2_SP(i), 0x00222222); + for (i = 0; i < 4; i++) + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_HYST_SP(i), 0x00000104); + for (i = 0; i < 4; i++) + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_DELAY_SP(i), 0x00000081); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL_UCHE, 0x22222222); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL2_UCHE, 0x02222222); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL3_UCHE, 0x00000000); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL4_UCHE, 0x00000000); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_HYST_UCHE, 0x00004444); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_DELAY_UCHE, 0x00001112); + for (i = 0; i < 4; i++) + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL_RB(i), 0x22222222); + + /* Disable L1 clocking in A420 due to CCU issues with it */ + for (i = 0; i < 4; i++) { + if (adreno_is_a420(adreno_gpu)) { + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL2_RB(i), + 0x00002020); + } else { + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL2_RB(i), + 0x00022020); + } + } + + /* No CCU for A405 */ + if (!adreno_is_a405(adreno_gpu)) { + for (i = 0; i < 4; i++) { + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL_MARB_CCU(i), + 0x00000922); + } + + for (i = 0; i < 4; i++) { + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_HYST_RB_MARB_CCU(i), + 0x00000000); + } + + for (i = 0; i < 4; i++) { + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_DELAY_RB_MARB_CCU_L1(i), + 0x00000001); + } + } + + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_MODE_GPC, 0x02222222); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_HYST_GPC, 0x04100104); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_DELAY_GPC, 0x00022222); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL_COM_DCOM, 0x00000022); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_HYST_COM_DCOM, 0x0000010F); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_DELAY_COM_DCOM, 0x00000022); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL_TSE_RAS_RBBM, 0x00222222); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00004104); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00000222); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL_HLSQ , 0x00000000); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_DELAY_HLSQ, 0x00220000); + /* Early A430's have a timing issue with SP/TP power collapse; + disabling HW clock gating prevents it. */ + if (adreno_is_a430(adreno_gpu) && adreno_patchid(adreno_gpu) < 2) + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL, 0); + else + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL, 0xAAAAAAAA); + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL2, 0); +} + + +static bool a4xx_me_init(struct msm_gpu *gpu) +{ + struct msm_ringbuffer *ring = gpu->rb[0]; + + OUT_PKT3(ring, CP_ME_INIT, 17); + OUT_RING(ring, 0x000003f7); + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000080); + OUT_RING(ring, 0x00000100); + OUT_RING(ring, 0x00000180); + OUT_RING(ring, 0x00006600); + OUT_RING(ring, 0x00000150); + OUT_RING(ring, 0x0000014e); + OUT_RING(ring, 0x00000154); + OUT_RING(ring, 0x00000001); + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + + adreno_flush(gpu, ring, REG_A4XX_CP_RB_WPTR); + return a4xx_idle(gpu); +} + +static int a4xx_hw_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a4xx_gpu *a4xx_gpu = to_a4xx_gpu(adreno_gpu); + uint32_t *ptr, len; + int i, ret; + + if (adreno_is_a405(adreno_gpu)) { + gpu_write(gpu, REG_A4XX_VBIF_ROUND_ROBIN_QOS_ARB, 0x00000003); + } else if (adreno_is_a420(adreno_gpu)) { + gpu_write(gpu, REG_A4XX_VBIF_ABIT_SORT, 0x0001001F); + gpu_write(gpu, REG_A4XX_VBIF_ABIT_SORT_CONF, 0x000000A4); + gpu_write(gpu, REG_A4XX_VBIF_GATE_OFF_WRREQ_EN, 0x00000001); + gpu_write(gpu, REG_A4XX_VBIF_IN_RD_LIM_CONF0, 0x18181818); + gpu_write(gpu, REG_A4XX_VBIF_IN_RD_LIM_CONF1, 0x00000018); + gpu_write(gpu, REG_A4XX_VBIF_IN_WR_LIM_CONF0, 0x18181818); + gpu_write(gpu, REG_A4XX_VBIF_IN_WR_LIM_CONF1, 0x00000018); + gpu_write(gpu, REG_A4XX_VBIF_ROUND_ROBIN_QOS_ARB, 0x00000003); + } else if (adreno_is_a430(adreno_gpu)) { + gpu_write(gpu, REG_A4XX_VBIF_GATE_OFF_WRREQ_EN, 0x00000001); + gpu_write(gpu, REG_A4XX_VBIF_IN_RD_LIM_CONF0, 0x18181818); + gpu_write(gpu, REG_A4XX_VBIF_IN_RD_LIM_CONF1, 0x00000018); + gpu_write(gpu, REG_A4XX_VBIF_IN_WR_LIM_CONF0, 0x18181818); + gpu_write(gpu, REG_A4XX_VBIF_IN_WR_LIM_CONF1, 0x00000018); + gpu_write(gpu, REG_A4XX_VBIF_ROUND_ROBIN_QOS_ARB, 0x00000003); + } else { + BUG(); + } + + /* Make all blocks contribute to the GPU BUSY perf counter */ + gpu_write(gpu, REG_A4XX_RBBM_GPU_BUSY_MASKED, 0xffffffff); + + /* Tune the hystersis counters for SP and CP idle detection */ + gpu_write(gpu, REG_A4XX_RBBM_SP_HYST_CNT, 0x10); + gpu_write(gpu, REG_A4XX_RBBM_WAIT_IDLE_CLOCKS_CTL, 0x10); + + if (adreno_is_a430(adreno_gpu)) { + gpu_write(gpu, REG_A4XX_RBBM_WAIT_IDLE_CLOCKS_CTL2, 0x30); + } + + /* Enable the RBBM error reporting bits */ + gpu_write(gpu, REG_A4XX_RBBM_AHB_CTL0, 0x00000001); + + /* Enable AHB error reporting*/ + gpu_write(gpu, REG_A4XX_RBBM_AHB_CTL1, 0xa6ffffff); + + /* Enable power counters*/ + gpu_write(gpu, REG_A4XX_RBBM_RBBM_CTL, 0x00000030); + + /* + * Turn on hang detection - this spews a lot of useful information + * into the RBBM registers on a hang: + */ + gpu_write(gpu, REG_A4XX_RBBM_INTERFACE_HANG_INT_CTL, + (1 << 30) | 0xFFFF); + + gpu_write(gpu, REG_A4XX_RB_GMEM_BASE_ADDR, + (unsigned int)(a4xx_gpu->ocmem.base >> 14)); + + /* Turn on performance counters: */ + gpu_write(gpu, REG_A4XX_RBBM_PERFCTR_CTL, 0x01); + + /* use the first CP counter for timestamp queries.. userspace may set + * this as well but it selects the same counter/countable: + */ + gpu_write(gpu, REG_A4XX_CP_PERFCTR_CP_SEL_0, CP_ALWAYS_COUNT); + + if (adreno_is_a430(adreno_gpu)) + gpu_write(gpu, REG_A4XX_UCHE_CACHE_WAYS_VFD, 0x07); + + /* Disable L2 bypass to avoid UCHE out of bounds errors */ + gpu_write(gpu, REG_A4XX_UCHE_TRAP_BASE_LO, lower_32_bits(adreno_gpu->uche_trap_base)); + gpu_write(gpu, REG_A4XX_UCHE_TRAP_BASE_HI, upper_32_bits(adreno_gpu->uche_trap_base)); + + gpu_write(gpu, REG_A4XX_CP_DEBUG, (1 << 25) | + (adreno_is_a420(adreno_gpu) ? (1 << 29) : 0)); + + /* On A430 enable SP regfile sleep for power savings */ + /* TODO downstream does this for !420, so maybe applies for 405 too? */ + if (!adreno_is_a420(adreno_gpu)) { + gpu_write(gpu, REG_A4XX_RBBM_SP_REGFILE_SLEEP_CNTL_0, + 0x00000441); + gpu_write(gpu, REG_A4XX_RBBM_SP_REGFILE_SLEEP_CNTL_1, + 0x00000441); + } + + a4xx_enable_hwcg(gpu); + + /* + * For A420 set RBBM_CLOCK_DELAY_HLSQ.CGC_HLSQ_TP_EARLY_CYC >= 2 + * due to timing issue with HLSQ_TP_CLK_EN + */ + if (adreno_is_a420(adreno_gpu)) { + unsigned int val; + val = gpu_read(gpu, REG_A4XX_RBBM_CLOCK_DELAY_HLSQ); + val &= ~A4XX_CGC_HLSQ_EARLY_CYC__MASK; + val |= 2 << A4XX_CGC_HLSQ_EARLY_CYC__SHIFT; + gpu_write(gpu, REG_A4XX_RBBM_CLOCK_DELAY_HLSQ, val); + } + + /* setup access protection: */ + gpu_write(gpu, REG_A4XX_CP_PROTECT_CTRL, 0x00000007); + + /* RBBM registers */ + gpu_write(gpu, REG_A4XX_CP_PROTECT(0), 0x62000010); + gpu_write(gpu, REG_A4XX_CP_PROTECT(1), 0x63000020); + gpu_write(gpu, REG_A4XX_CP_PROTECT(2), 0x64000040); + gpu_write(gpu, REG_A4XX_CP_PROTECT(3), 0x65000080); + gpu_write(gpu, REG_A4XX_CP_PROTECT(4), 0x66000100); + gpu_write(gpu, REG_A4XX_CP_PROTECT(5), 0x64000200); + + /* CP registers */ + gpu_write(gpu, REG_A4XX_CP_PROTECT(6), 0x67000800); + gpu_write(gpu, REG_A4XX_CP_PROTECT(7), 0x64001600); + + + /* RB registers */ + gpu_write(gpu, REG_A4XX_CP_PROTECT(8), 0x60003300); + + /* HLSQ registers */ + gpu_write(gpu, REG_A4XX_CP_PROTECT(9), 0x60003800); + + /* VPC registers */ + gpu_write(gpu, REG_A4XX_CP_PROTECT(10), 0x61003980); + + /* SMMU registers */ + gpu_write(gpu, REG_A4XX_CP_PROTECT(11), 0x6e010000); + + gpu_write(gpu, REG_A4XX_RBBM_INT_0_MASK, A4XX_INT0_MASK); + + ret = adreno_hw_init(gpu); + if (ret) + return ret; + + /* + * Use the default ringbuffer size and block size but disable the RPTR + * shadow + */ + gpu_write(gpu, REG_A4XX_CP_RB_CNTL, + MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE); + + /* Set the ringbuffer address */ + gpu_write(gpu, REG_A4XX_CP_RB_BASE, lower_32_bits(gpu->rb[0]->iova)); + + /* Load PM4: */ + ptr = (uint32_t *)(adreno_gpu->fw[ADRENO_FW_PM4]->data); + len = adreno_gpu->fw[ADRENO_FW_PM4]->size / 4; + DBG("loading PM4 ucode version: %u", ptr[0]); + gpu_write(gpu, REG_A4XX_CP_ME_RAM_WADDR, 0); + for (i = 1; i < len; i++) + gpu_write(gpu, REG_A4XX_CP_ME_RAM_DATA, ptr[i]); + + /* Load PFP: */ + ptr = (uint32_t *)(adreno_gpu->fw[ADRENO_FW_PFP]->data); + len = adreno_gpu->fw[ADRENO_FW_PFP]->size / 4; + DBG("loading PFP ucode version: %u", ptr[0]); + + gpu_write(gpu, REG_A4XX_CP_PFP_UCODE_ADDR, 0); + for (i = 1; i < len; i++) + gpu_write(gpu, REG_A4XX_CP_PFP_UCODE_DATA, ptr[i]); + + /* clear ME_HALT to start micro engine */ + gpu_write(gpu, REG_A4XX_CP_ME_CNTL, 0); + + return a4xx_me_init(gpu) ? 0 : -EINVAL; +} + +static void a4xx_recover(struct msm_gpu *gpu) +{ + int i; + + adreno_dump_info(gpu); + + for (i = 0; i < 8; i++) { + printk("CP_SCRATCH_REG%d: %u\n", i, + gpu_read(gpu, REG_AXXX_CP_SCRATCH_REG0 + i)); + } + + /* dump registers before resetting gpu, if enabled: */ + if (hang_debug) + a4xx_dump(gpu); + + gpu_write(gpu, REG_A4XX_RBBM_SW_RESET_CMD, 1); + gpu_read(gpu, REG_A4XX_RBBM_SW_RESET_CMD); + gpu_write(gpu, REG_A4XX_RBBM_SW_RESET_CMD, 0); + adreno_recover(gpu); +} + +static void a4xx_destroy(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a4xx_gpu *a4xx_gpu = to_a4xx_gpu(adreno_gpu); + + DBG("%s", gpu->name); + + adreno_gpu_cleanup(adreno_gpu); + + adreno_gpu_ocmem_cleanup(&a4xx_gpu->ocmem); + + kfree(a4xx_gpu); +} + +static bool a4xx_idle(struct msm_gpu *gpu) +{ + /* wait for ringbuffer to drain: */ + if (!adreno_idle(gpu, gpu->rb[0])) + return false; + + /* then wait for GPU to finish: */ + if (spin_until(!(gpu_read(gpu, REG_A4XX_RBBM_STATUS) & + A4XX_RBBM_STATUS_GPU_BUSY))) { + DRM_ERROR("%s: timeout waiting for GPU to idle!\n", gpu->name); + /* TODO maybe we need to reset GPU here to recover from hang? */ + return false; + } + + return true; +} + +static irqreturn_t a4xx_irq(struct msm_gpu *gpu) +{ + uint32_t status; + + status = gpu_read(gpu, REG_A4XX_RBBM_INT_0_STATUS); + DBG("%s: Int status %08x", gpu->name, status); + + if (status & A4XX_INT0_CP_REG_PROTECT_FAULT) { + uint32_t reg = gpu_read(gpu, REG_A4XX_CP_PROTECT_STATUS); + printk("CP | Protected mode error| %s | addr=%x\n", + reg & (1 << 24) ? "WRITE" : "READ", + (reg & 0xFFFFF) >> 2); + } + + gpu_write(gpu, REG_A4XX_RBBM_INT_CLEAR_CMD, status); + + msm_gpu_retire(gpu); + + return IRQ_HANDLED; +} + +static const unsigned int a4xx_registers[] = { + /* RBBM */ + 0x0000, 0x0002, 0x0004, 0x0021, 0x0023, 0x0024, 0x0026, 0x0026, + 0x0028, 0x002B, 0x002E, 0x0034, 0x0037, 0x0044, 0x0047, 0x0066, + 0x0068, 0x0095, 0x009C, 0x0170, 0x0174, 0x01AF, + /* CP */ + 0x0200, 0x0233, 0x0240, 0x0250, 0x04C0, 0x04DD, 0x0500, 0x050B, + 0x0578, 0x058F, + /* VSC */ + 0x0C00, 0x0C03, 0x0C08, 0x0C41, 0x0C50, 0x0C51, + /* GRAS */ + 0x0C80, 0x0C81, 0x0C88, 0x0C8F, + /* RB */ + 0x0CC0, 0x0CC0, 0x0CC4, 0x0CD2, + /* PC */ + 0x0D00, 0x0D0C, 0x0D10, 0x0D17, 0x0D20, 0x0D23, + /* VFD */ + 0x0E40, 0x0E4A, + /* VPC */ + 0x0E60, 0x0E61, 0x0E63, 0x0E68, + /* UCHE */ + 0x0E80, 0x0E84, 0x0E88, 0x0E95, + /* VMIDMT */ + 0x1000, 0x1000, 0x1002, 0x1002, 0x1004, 0x1004, 0x1008, 0x100A, + 0x100C, 0x100D, 0x100F, 0x1010, 0x1012, 0x1016, 0x1024, 0x1024, + 0x1027, 0x1027, 0x1100, 0x1100, 0x1102, 0x1102, 0x1104, 0x1104, + 0x1110, 0x1110, 0x1112, 0x1116, 0x1124, 0x1124, 0x1300, 0x1300, + 0x1380, 0x1380, + /* GRAS CTX 0 */ + 0x2000, 0x2004, 0x2008, 0x2067, 0x2070, 0x2078, 0x207B, 0x216E, + /* PC CTX 0 */ + 0x21C0, 0x21C6, 0x21D0, 0x21D0, 0x21D9, 0x21D9, 0x21E5, 0x21E7, + /* VFD CTX 0 */ + 0x2200, 0x2204, 0x2208, 0x22A9, + /* GRAS CTX 1 */ + 0x2400, 0x2404, 0x2408, 0x2467, 0x2470, 0x2478, 0x247B, 0x256E, + /* PC CTX 1 */ + 0x25C0, 0x25C6, 0x25D0, 0x25D0, 0x25D9, 0x25D9, 0x25E5, 0x25E7, + /* VFD CTX 1 */ + 0x2600, 0x2604, 0x2608, 0x26A9, + /* XPU */ + 0x2C00, 0x2C01, 0x2C10, 0x2C10, 0x2C12, 0x2C16, 0x2C1D, 0x2C20, + 0x2C28, 0x2C28, 0x2C30, 0x2C30, 0x2C32, 0x2C36, 0x2C40, 0x2C40, + 0x2C50, 0x2C50, 0x2C52, 0x2C56, 0x2C80, 0x2C80, 0x2C94, 0x2C95, + /* VBIF */ + 0x3000, 0x3007, 0x300C, 0x3014, 0x3018, 0x301D, 0x3020, 0x3022, + 0x3024, 0x3026, 0x3028, 0x302A, 0x302C, 0x302D, 0x3030, 0x3031, + 0x3034, 0x3036, 0x3038, 0x3038, 0x303C, 0x303D, 0x3040, 0x3040, + 0x3049, 0x3049, 0x3058, 0x3058, 0x305B, 0x3061, 0x3064, 0x3068, + 0x306C, 0x306D, 0x3080, 0x3088, 0x308B, 0x308C, 0x3090, 0x3094, + 0x3098, 0x3098, 0x309C, 0x309C, 0x30C0, 0x30C0, 0x30C8, 0x30C8, + 0x30D0, 0x30D0, 0x30D8, 0x30D8, 0x30E0, 0x30E0, 0x3100, 0x3100, + 0x3108, 0x3108, 0x3110, 0x3110, 0x3118, 0x3118, 0x3120, 0x3120, + 0x3124, 0x3125, 0x3129, 0x3129, 0x3131, 0x3131, 0x330C, 0x330C, + 0x3310, 0x3310, 0x3400, 0x3401, 0x3410, 0x3410, 0x3412, 0x3416, + 0x341D, 0x3420, 0x3428, 0x3428, 0x3430, 0x3430, 0x3432, 0x3436, + 0x3440, 0x3440, 0x3450, 0x3450, 0x3452, 0x3456, 0x3480, 0x3480, + 0x3494, 0x3495, 0x4000, 0x4000, 0x4002, 0x4002, 0x4004, 0x4004, + 0x4008, 0x400A, 0x400C, 0x400D, 0x400F, 0x4012, 0x4014, 0x4016, + 0x401D, 0x401D, 0x4020, 0x4027, 0x4060, 0x4062, 0x4200, 0x4200, + 0x4300, 0x4300, 0x4400, 0x4400, 0x4500, 0x4500, 0x4800, 0x4802, + 0x480F, 0x480F, 0x4811, 0x4811, 0x4813, 0x4813, 0x4815, 0x4816, + 0x482B, 0x482B, 0x4857, 0x4857, 0x4883, 0x4883, 0x48AF, 0x48AF, + 0x48C5, 0x48C5, 0x48E5, 0x48E5, 0x4905, 0x4905, 0x4925, 0x4925, + 0x4945, 0x4945, 0x4950, 0x4950, 0x495B, 0x495B, 0x4980, 0x498E, + 0x4B00, 0x4B00, 0x4C00, 0x4C00, 0x4D00, 0x4D00, 0x4E00, 0x4E00, + 0x4E80, 0x4E80, 0x4F00, 0x4F00, 0x4F08, 0x4F08, 0x4F10, 0x4F10, + 0x4F18, 0x4F18, 0x4F20, 0x4F20, 0x4F30, 0x4F30, 0x4F60, 0x4F60, + 0x4F80, 0x4F81, 0x4F88, 0x4F89, 0x4FEE, 0x4FEE, 0x4FF3, 0x4FF3, + 0x6000, 0x6001, 0x6008, 0x600F, 0x6014, 0x6016, 0x6018, 0x601B, + 0x61FD, 0x61FD, 0x623C, 0x623C, 0x6380, 0x6380, 0x63A0, 0x63A0, + 0x63C0, 0x63C1, 0x63C8, 0x63C9, 0x63D0, 0x63D4, 0x63D6, 0x63D6, + 0x63EE, 0x63EE, 0x6400, 0x6401, 0x6408, 0x640F, 0x6414, 0x6416, + 0x6418, 0x641B, 0x65FD, 0x65FD, 0x663C, 0x663C, 0x6780, 0x6780, + 0x67A0, 0x67A0, 0x67C0, 0x67C1, 0x67C8, 0x67C9, 0x67D0, 0x67D4, + 0x67D6, 0x67D6, 0x67EE, 0x67EE, 0x6800, 0x6801, 0x6808, 0x680F, + 0x6814, 0x6816, 0x6818, 0x681B, 0x69FD, 0x69FD, 0x6A3C, 0x6A3C, + 0x6B80, 0x6B80, 0x6BA0, 0x6BA0, 0x6BC0, 0x6BC1, 0x6BC8, 0x6BC9, + 0x6BD0, 0x6BD4, 0x6BD6, 0x6BD6, 0x6BEE, 0x6BEE, + ~0 /* sentinel */ +}; + +static const unsigned int a405_registers[] = { + /* RBBM */ + 0x0000, 0x0002, 0x0004, 0x0021, 0x0023, 0x0024, 0x0026, 0x0026, + 0x0028, 0x002B, 0x002E, 0x0034, 0x0037, 0x0044, 0x0047, 0x0066, + 0x0068, 0x0095, 0x009C, 0x0170, 0x0174, 0x01AF, + /* CP */ + 0x0200, 0x0233, 0x0240, 0x0250, 0x04C0, 0x04DD, 0x0500, 0x050B, + 0x0578, 0x058F, + /* VSC */ + 0x0C00, 0x0C03, 0x0C08, 0x0C41, 0x0C50, 0x0C51, + /* GRAS */ + 0x0C80, 0x0C81, 0x0C88, 0x0C8F, + /* RB */ + 0x0CC0, 0x0CC0, 0x0CC4, 0x0CD2, + /* PC */ + 0x0D00, 0x0D0C, 0x0D10, 0x0D17, 0x0D20, 0x0D23, + /* VFD */ + 0x0E40, 0x0E4A, + /* VPC */ + 0x0E60, 0x0E61, 0x0E63, 0x0E68, + /* UCHE */ + 0x0E80, 0x0E84, 0x0E88, 0x0E95, + /* GRAS CTX 0 */ + 0x2000, 0x2004, 0x2008, 0x2067, 0x2070, 0x2078, 0x207B, 0x216E, + /* PC CTX 0 */ + 0x21C0, 0x21C6, 0x21D0, 0x21D0, 0x21D9, 0x21D9, 0x21E5, 0x21E7, + /* VFD CTX 0 */ + 0x2200, 0x2204, 0x2208, 0x22A9, + /* GRAS CTX 1 */ + 0x2400, 0x2404, 0x2408, 0x2467, 0x2470, 0x2478, 0x247B, 0x256E, + /* PC CTX 1 */ + 0x25C0, 0x25C6, 0x25D0, 0x25D0, 0x25D9, 0x25D9, 0x25E5, 0x25E7, + /* VFD CTX 1 */ + 0x2600, 0x2604, 0x2608, 0x26A9, + /* VBIF version 0x20050000*/ + 0x3000, 0x3007, 0x302C, 0x302C, 0x3030, 0x3030, 0x3034, 0x3036, + 0x3038, 0x3038, 0x303C, 0x303D, 0x3040, 0x3040, 0x3049, 0x3049, + 0x3058, 0x3058, 0x305B, 0x3061, 0x3064, 0x3068, 0x306C, 0x306D, + 0x3080, 0x3088, 0x308B, 0x308C, 0x3090, 0x3094, 0x3098, 0x3098, + 0x309C, 0x309C, 0x30C0, 0x30C0, 0x30C8, 0x30C8, 0x30D0, 0x30D0, + 0x30D8, 0x30D8, 0x30E0, 0x30E0, 0x3100, 0x3100, 0x3108, 0x3108, + 0x3110, 0x3110, 0x3118, 0x3118, 0x3120, 0x3120, 0x3124, 0x3125, + 0x3129, 0x3129, 0x340C, 0x340C, 0x3410, 0x3410, + ~0 /* sentinel */ +}; + +static struct msm_gpu_state *a4xx_gpu_state_get(struct msm_gpu *gpu) +{ + struct msm_gpu_state *state = kzalloc_obj(*state); + + if (!state) + return ERR_PTR(-ENOMEM); + + adreno_gpu_state_get(gpu, state); + + state->rbbm_status = gpu_read(gpu, REG_A4XX_RBBM_STATUS); + + return state; +} + +static void a4xx_dump(struct msm_gpu *gpu) +{ + printk("status: %08x\n", + gpu_read(gpu, REG_A4XX_RBBM_STATUS)); + adreno_dump(gpu); +} + +static int a4xx_pm_resume(struct msm_gpu *gpu) { + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + int ret; + + ret = msm_gpu_pm_resume(gpu); + if (ret) + return ret; + + if (adreno_is_a430(adreno_gpu)) { + unsigned int reg; + /* Set the default register values; set SW_COLLAPSE to 0 */ + gpu_write(gpu, REG_A4XX_RBBM_POWER_CNTL_IP, 0x778000); + do { + udelay(5); + reg = gpu_read(gpu, REG_A4XX_RBBM_POWER_STATUS); + } while (!(reg & A4XX_RBBM_POWER_CNTL_IP_SP_TP_PWR_ON)); + } + return 0; +} + +static int a4xx_pm_suspend(struct msm_gpu *gpu) { + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + int ret; + + ret = msm_gpu_pm_suspend(gpu); + if (ret) + return ret; + + if (adreno_is_a430(adreno_gpu)) { + /* Set the default register values; set SW_COLLAPSE to 1 */ + gpu_write(gpu, REG_A4XX_RBBM_POWER_CNTL_IP, 0x778001); + } + return 0; +} + +static u64 a4xx_get_timestamp(struct msm_gpu *gpu) +{ + return gpu_read64(gpu, REG_A4XX_RBBM_PERFCTR_CP_0_LO); +} + +static u64 a4xx_gpu_busy(struct msm_gpu *gpu, unsigned long *out_sample_rate) +{ + u64 busy_cycles; + + busy_cycles = gpu_read64(gpu, REG_A4XX_RBBM_PERFCTR_RBBM_1_LO); + *out_sample_rate = clk_get_rate(gpu->core_clk); + + return busy_cycles; +} + +static u32 a4xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + ring->memptrs->rptr = gpu_read(gpu, REG_A4XX_CP_RB_RPTR); + return ring->memptrs->rptr; +} + +static struct msm_gpu *a4xx_gpu_init(struct drm_device *dev) +{ + struct a4xx_gpu *a4xx_gpu = NULL; + struct adreno_gpu *adreno_gpu; + struct msm_gpu *gpu; + struct msm_drm_private *priv = dev->dev_private; + struct platform_device *pdev = priv->gpu_pdev; + struct adreno_platform_config *config = pdev->dev.platform_data; + struct icc_path *ocmem_icc_path; + struct icc_path *icc_path; + int ret; + + if (!pdev) { + DRM_DEV_ERROR(dev->dev, "no a4xx device\n"); + ret = -ENXIO; + goto fail; + } + + a4xx_gpu = kzalloc_obj(*a4xx_gpu); + if (!a4xx_gpu) { + ret = -ENOMEM; + goto fail; + } + + adreno_gpu = &a4xx_gpu->base; + gpu = &adreno_gpu->base; + + ret = adreno_gpu_init(dev, pdev, adreno_gpu, config->info->funcs, 1); + if (ret) + goto fail; + + adreno_gpu->registers = adreno_is_a405(adreno_gpu) ? a405_registers : + a4xx_registers; + + /* if needed, allocate gmem: */ + ret = adreno_gpu_ocmem_init(dev->dev, adreno_gpu, + &a4xx_gpu->ocmem); + if (ret) + goto fail; + + adreno_gpu->uche_trap_base = 0xffff0000ffff0000ull; + + icc_path = devm_of_icc_get(&pdev->dev, "gfx-mem"); + if (IS_ERR(icc_path)) { + ret = PTR_ERR(icc_path); + goto fail; + } + + ocmem_icc_path = devm_of_icc_get(&pdev->dev, "ocmem"); + if (IS_ERR(ocmem_icc_path)) { + ret = PTR_ERR(ocmem_icc_path); + /* allow -ENODATA, ocmem icc is optional */ + if (ret != -ENODATA) + goto fail; + ocmem_icc_path = NULL; + } + + /* + * Set the ICC path to maximum speed for now by multiplying the fastest + * frequency by the bus width (8). We'll want to scale this later on to + * improve battery life. + */ + icc_set_bw(icc_path, 0, Bps_to_icc(gpu->fast_rate) * 8); + icc_set_bw(ocmem_icc_path, 0, Bps_to_icc(gpu->fast_rate) * 8); + + return gpu; + +fail: + if (a4xx_gpu) + a4xx_destroy(&a4xx_gpu->base.base); + + return ERR_PTR(ret); +} + +const struct adreno_gpu_funcs a4xx_gpu_funcs = { + .base = { + .get_param = adreno_get_param, + .set_param = adreno_set_param, + .hw_init = a4xx_hw_init, + .pm_suspend = a4xx_pm_suspend, + .pm_resume = a4xx_pm_resume, + .recover = a4xx_recover, + .submit = a4xx_submit, + .active_ring = adreno_active_ring, + .irq = a4xx_irq, + .destroy = a4xx_destroy, +#if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP) + .show = adreno_show, +#endif + .gpu_busy = a4xx_gpu_busy, + .gpu_state_get = a4xx_gpu_state_get, + .gpu_state_put = adreno_gpu_state_put, + .create_vm = adreno_create_vm, + .get_rptr = a4xx_get_rptr, + }, + .init = a4xx_gpu_init, + .get_timestamp = a4xx_get_timestamp, +}; diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.h b/drivers/gpu/drm/msm/adreno/a4xx_gpu.h new file mode 100644 index 000000000..71b164439 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.h @@ -0,0 +1,25 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2014 The Linux Foundation. All rights reserved. + */ +#ifndef __A4XX_GPU_H__ +#define __A4XX_GPU_H__ + +#include "adreno_gpu.h" + +/* arrg, somehow fb.h is getting pulled in: */ +#undef ROP_COPY +#undef ROP_XOR + +#include "a4xx.xml.h" + +struct a4xx_gpu { + struct adreno_gpu base; + + /* if OCMEM is used for GMEM: */ + struct adreno_ocmem ocmem; +}; +#define to_a4xx_gpu(x) container_of(x, struct a4xx_gpu, base) + +extern const struct adreno_gpu_funcs a4xx_gpu_funcs; + +#endif /* __A4XX_GPU_H__ */ diff --git a/drivers/gpu/drm/msm/adreno/a5xx_catalog.c b/drivers/gpu/drm/msm/adreno/a5xx_catalog.c new file mode 100644 index 000000000..babd320f3 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a5xx_catalog.c @@ -0,0 +1,153 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013-2014 Red Hat + * Author: Rob Clark + * + * Copyright (c) 2014,2017 The Linux Foundation. All rights reserved. + */ + +#include "adreno_gpu.h" +#include "a5xx_gpu.h" + +static const struct adreno_info a5xx_gpus[] = { + { + .chip_ids = ADRENO_CHIP_IDS(0x05000500), + .family = ADRENO_5XX, + .revn = 505, + .fw = { + [ADRENO_FW_PM4] = "a530_pm4.fw", + [ADRENO_FW_PFP] = "a530_pfp.fw", + }, + .gmem = (SZ_128K + SZ_8K), + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_TWO_PASS_USE_WFI | + ADRENO_QUIRK_LMLOADKILL_DISABLE, + .funcs = &a5xx_gpu_funcs, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x05000600), + .family = ADRENO_5XX, + .revn = 506, + .fw = { + [ADRENO_FW_PM4] = "a530_pm4.fw", + [ADRENO_FW_PFP] = "a530_pfp.fw", + }, + .gmem = (SZ_128K + SZ_8K), + /* + * Increase inactive period to 250 to avoid bouncing + * the GDSC which appears to make it grumpy + */ + .inactive_period = 250, + .quirks = ADRENO_QUIRK_TWO_PASS_USE_WFI | + ADRENO_QUIRK_LMLOADKILL_DISABLE, + .funcs = &a5xx_gpu_funcs, + .zapfw = "a506_zap.mdt", + }, { + .chip_ids = ADRENO_CHIP_IDS(0x05000800), + .family = ADRENO_5XX, + .revn = 508, + .fw = { + [ADRENO_FW_PM4] = "a530_pm4.fw", + [ADRENO_FW_PFP] = "a530_pfp.fw", + }, + .gmem = (SZ_128K + SZ_8K), + /* + * Increase inactive period to 250 to avoid bouncing + * the GDSC which appears to make it grumpy + */ + .inactive_period = 250, + .quirks = ADRENO_QUIRK_LMLOADKILL_DISABLE, + .funcs = &a5xx_gpu_funcs, + .zapfw = "a508_zap.mdt", + }, { + .chip_ids = ADRENO_CHIP_IDS(0x05000900), + .family = ADRENO_5XX, + .revn = 509, + .fw = { + [ADRENO_FW_PM4] = "a530_pm4.fw", + [ADRENO_FW_PFP] = "a530_pfp.fw", + }, + .gmem = (SZ_256K + SZ_16K), + /* + * Increase inactive period to 250 to avoid bouncing + * the GDSC which appears to make it grumpy + */ + .inactive_period = 250, + .quirks = ADRENO_QUIRK_LMLOADKILL_DISABLE, + .funcs = &a5xx_gpu_funcs, + /* Adreno 509 uses the same ZAP as 512 */ + .zapfw = "a512_zap.mdt", + }, { + .chip_ids = ADRENO_CHIP_IDS(0x05010000), + .family = ADRENO_5XX, + .revn = 510, + .fw = { + [ADRENO_FW_PM4] = "a530_pm4.fw", + [ADRENO_FW_PFP] = "a530_pfp.fw", + }, + .gmem = SZ_256K, + /* + * Increase inactive period to 250 to avoid bouncing + * the GDSC which appears to make it grumpy + */ + .inactive_period = 250, + .funcs = &a5xx_gpu_funcs, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x05010200), + .family = ADRENO_5XX, + .revn = 512, + .fw = { + [ADRENO_FW_PM4] = "a530_pm4.fw", + [ADRENO_FW_PFP] = "a530_pfp.fw", + }, + .gmem = (SZ_256K + SZ_16K), + /* + * Increase inactive period to 250 to avoid bouncing + * the GDSC which appears to make it grumpy + */ + .inactive_period = 250, + .quirks = ADRENO_QUIRK_LMLOADKILL_DISABLE, + .funcs = &a5xx_gpu_funcs, + .zapfw = "a512_zap.mdt", + }, { + .chip_ids = ADRENO_CHIP_IDS( + 0x05030002, + 0x05030004 + ), + .family = ADRENO_5XX, + .revn = 530, + .fw = { + [ADRENO_FW_PM4] = "a530_pm4.fw", + [ADRENO_FW_PFP] = "a530_pfp.fw", + [ADRENO_FW_GPMU] = "a530v3_gpmu.fw2", + }, + .gmem = SZ_1M, + /* + * Increase inactive period to 250 to avoid bouncing + * the GDSC which appears to make it grumpy + */ + .inactive_period = 250, + .quirks = ADRENO_QUIRK_TWO_PASS_USE_WFI | + ADRENO_QUIRK_FAULT_DETECT_MASK, + .funcs = &a5xx_gpu_funcs, + .zapfw = "a530_zap.mdt", + }, { + .chip_ids = ADRENO_CHIP_IDS(0x05040001), + .family = ADRENO_5XX, + .revn = 540, + .fw = { + [ADRENO_FW_PM4] = "a530_pm4.fw", + [ADRENO_FW_PFP] = "a530_pfp.fw", + [ADRENO_FW_GPMU] = "a540_gpmu.fw2", + }, + .gmem = SZ_1M, + /* + * Increase inactive period to 250 to avoid bouncing + * the GDSC which appears to make it grumpy + */ + .inactive_period = 250, + .quirks = ADRENO_QUIRK_LMLOADKILL_DISABLE, + .funcs = &a5xx_gpu_funcs, + .zapfw = "a540_zap.mdt", + } +}; +DECLARE_ADRENO_GPULIST(a5xx); diff --git a/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c b/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c new file mode 100644 index 000000000..625a4e787 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c @@ -0,0 +1,159 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2016-2017 The Linux Foundation. All rights reserved. + */ + +#include +#include + +#include +#include +#include + +#include "a5xx_gpu.h" + +static void pfp_print(struct msm_gpu *gpu, struct drm_printer *p) +{ + int i; + + drm_printf(p, "PFP state:\n"); + + for (i = 0; i < 36; i++) { + gpu_write(gpu, REG_A5XX_CP_PFP_STAT_ADDR, i); + drm_printf(p, " %02x: %08x\n", i, + gpu_read(gpu, REG_A5XX_CP_PFP_STAT_DATA)); + } +} + +static void me_print(struct msm_gpu *gpu, struct drm_printer *p) +{ + int i; + + drm_printf(p, "ME state:\n"); + + for (i = 0; i < 29; i++) { + gpu_write(gpu, REG_A5XX_CP_ME_STAT_ADDR, i); + drm_printf(p, " %02x: %08x\n", i, + gpu_read(gpu, REG_A5XX_CP_ME_STAT_DATA)); + } +} + +static void meq_print(struct msm_gpu *gpu, struct drm_printer *p) +{ + int i; + + drm_printf(p, "MEQ state:\n"); + gpu_write(gpu, REG_A5XX_CP_MEQ_DBG_ADDR, 0); + + for (i = 0; i < 64; i++) { + drm_printf(p, " %02x: %08x\n", i, + gpu_read(gpu, REG_A5XX_CP_MEQ_DBG_DATA)); + } +} + +static void roq_print(struct msm_gpu *gpu, struct drm_printer *p) +{ + int i; + + drm_printf(p, "ROQ state:\n"); + gpu_write(gpu, REG_A5XX_CP_ROQ_DBG_ADDR, 0); + + for (i = 0; i < 512 / 4; i++) { + uint32_t val[4]; + int j; + for (j = 0; j < 4; j++) + val[j] = gpu_read(gpu, REG_A5XX_CP_ROQ_DBG_DATA); + drm_printf(p, " %02x: %08x %08x %08x %08x\n", i, + val[0], val[1], val[2], val[3]); + } +} + +static int show(struct seq_file *m, void *arg) +{ + struct drm_info_node *node = m->private; + struct drm_device *dev = node->minor->dev; + struct msm_drm_private *priv = dev->dev_private; + struct drm_printer p = drm_seq_file_printer(m); + void (*show)(struct msm_gpu *gpu, struct drm_printer *p) = + node->info_ent->data; + + show(priv->gpu, &p); + return 0; +} + +#define ENT(n) { .name = #n, .show = show, .data = n ##_print } +static struct drm_info_list a5xx_debugfs_list[] = { + ENT(pfp), + ENT(me), + ENT(meq), + ENT(roq), +}; + +/* for debugfs files that can be written to, we can't use drm helper: */ +static int +reset_set(void *data, u64 val) +{ + struct drm_device *dev = data; + struct msm_drm_private *priv = dev->dev_private; + struct msm_gpu *gpu = priv->gpu; + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + + if (!capable(CAP_SYS_ADMIN)) + return -EINVAL; + + /* TODO do we care about trying to make sure the GPU is idle? + * Since this is just a debug feature limited to CAP_SYS_ADMIN, + * maybe it is fine to let the user keep both pieces if they + * try to reset an active GPU. + */ + + mutex_lock(&gpu->lock); + + release_firmware(adreno_gpu->fw[ADRENO_FW_PM4]); + adreno_gpu->fw[ADRENO_FW_PM4] = NULL; + + release_firmware(adreno_gpu->fw[ADRENO_FW_PFP]); + adreno_gpu->fw[ADRENO_FW_PFP] = NULL; + + if (a5xx_gpu->pm4_bo) { + msm_gem_unpin_iova(a5xx_gpu->pm4_bo, gpu->vm); + drm_gem_object_put(a5xx_gpu->pm4_bo); + a5xx_gpu->pm4_bo = NULL; + } + + if (a5xx_gpu->pfp_bo) { + msm_gem_unpin_iova(a5xx_gpu->pfp_bo, gpu->vm); + drm_gem_object_put(a5xx_gpu->pfp_bo); + a5xx_gpu->pfp_bo = NULL; + } + + gpu->needs_hw_init = true; + + pm_runtime_get_sync(&gpu->pdev->dev); + gpu->funcs->recover(gpu); + + pm_runtime_put_sync(&gpu->pdev->dev); + mutex_unlock(&gpu->lock); + + return 0; +} + +DEFINE_DEBUGFS_ATTRIBUTE(reset_fops, NULL, reset_set, "%llx\n"); + + +void a5xx_debugfs_init(struct msm_gpu *gpu, struct drm_minor *minor) +{ + struct drm_device *dev; + + if (!minor) + return; + + dev = minor->dev; + + drm_debugfs_create_files(a5xx_debugfs_list, + ARRAY_SIZE(a5xx_debugfs_list), + minor->debugfs_root, minor); + + debugfs_create_file_unsafe("reset", S_IWUGO, minor->debugfs_root, dev, + &reset_fops); +} diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c new file mode 100644 index 000000000..5d96b8f63 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c @@ -0,0 +1,1806 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2016-2017 The Linux Foundation. All rights reserved. + */ + +#include +#include +#include +#include +#include +#include +#include +#include "msm_gem.h" +#include "msm_mmu.h" +#include "a5xx_gpu.h" + +extern bool hang_debug; +static void a5xx_dump(struct msm_gpu *gpu); + +#define GPU_PAS_ID 13 + +static void update_shadow_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + + if (a5xx_gpu->has_whereami) { + OUT_PKT7(ring, CP_WHERE_AM_I, 2); + OUT_RING(ring, lower_32_bits(shadowptr(a5xx_gpu, ring))); + OUT_RING(ring, upper_32_bits(shadowptr(a5xx_gpu, ring))); + } +} + +void a5xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring, + bool sync) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + uint32_t wptr; + unsigned long flags; + + /* + * Most flush operations need to issue a WHERE_AM_I opcode to sync up + * the rptr shadow + */ + if (sync) + update_shadow_rptr(gpu, ring); + + spin_lock_irqsave(&ring->preempt_lock, flags); + + /* Copy the shadow to the actual register */ + ring->cur = ring->next; + + /* Make sure to wrap wptr if we need to */ + wptr = get_wptr(ring); + + spin_unlock_irqrestore(&ring->preempt_lock, flags); + + /* Make sure everything is posted before making a decision */ + mb(); + + /* Update HW if this is the current ring and we are not in preempt */ + if (a5xx_gpu->cur_ring == ring && !a5xx_in_preempt(a5xx_gpu)) + gpu_write(gpu, REG_A5XX_CP_RB_WPTR, wptr); +} + +static void a5xx_submit_in_rb(struct msm_gpu *gpu, struct msm_gem_submit *submit) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + struct msm_ringbuffer *ring = submit->ring; + struct drm_gem_object *obj; + uint32_t *ptr, dwords; + unsigned int i; + + for (i = 0; i < submit->nr_cmds; i++) { + switch (submit->cmd[i].type) { + case MSM_SUBMIT_CMD_IB_TARGET_BUF: + break; + case MSM_SUBMIT_CMD_CTX_RESTORE_BUF: + if (ring->cur_ctx_seqno == submit->queue->ctx->seqno) + break; + fallthrough; + case MSM_SUBMIT_CMD_BUF: + /* copy commands into RB: */ + obj = submit->bos[submit->cmd[i].idx].obj; + dwords = submit->cmd[i].size; + + ptr = msm_gem_get_vaddr(obj); + + /* _get_vaddr() shouldn't fail at this point, + * since we've already mapped it once in + * submit_reloc() + */ + if (WARN_ON(IS_ERR_OR_NULL(ptr))) + return; + + for (i = 0; i < dwords; i++) { + /* normally the OUT_PKTn() would wait + * for space for the packet. But since + * we just OUT_RING() the whole thing, + * need to call adreno_wait_ring() + * ourself: + */ + adreno_wait_ring(ring, 1); + OUT_RING(ring, ptr[i]); + } + + msm_gem_put_vaddr(obj); + + break; + } + } + + a5xx_gpu->last_seqno[ring->id] = submit->seqno; + a5xx_flush(gpu, ring, true); + a5xx_preempt_trigger(gpu); + + /* we might not necessarily have a cmd from userspace to + * trigger an event to know that submit has completed, so + * do this manually: + */ + a5xx_idle(gpu, ring); + ring->memptrs->fence = submit->seqno; + msm_gpu_retire(gpu); +} + +static void a5xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + struct msm_ringbuffer *ring = submit->ring; + unsigned int i, ibs = 0; + + adreno_check_and_reenable_stall(adreno_gpu); + + if (IS_ENABLED(CONFIG_DRM_MSM_GPU_SUDO) && submit->in_rb) { + ring->cur_ctx_seqno = 0; + a5xx_submit_in_rb(gpu, submit); + return; + } + + OUT_PKT7(ring, CP_PREEMPT_ENABLE_GLOBAL, 1); + OUT_RING(ring, 0x02); + + /* Turn off protected mode to write to special registers */ + OUT_PKT7(ring, CP_SET_PROTECTED_MODE, 1); + OUT_RING(ring, 0); + + /* Set the save preemption record for the ring/command */ + OUT_PKT4(ring, REG_A5XX_CP_CONTEXT_SWITCH_SAVE_ADDR_LO, 2); + OUT_RING(ring, lower_32_bits(a5xx_gpu->preempt_iova[submit->ring->id])); + OUT_RING(ring, upper_32_bits(a5xx_gpu->preempt_iova[submit->ring->id])); + + /* Turn back on protected mode */ + OUT_PKT7(ring, CP_SET_PROTECTED_MODE, 1); + OUT_RING(ring, 1); + + /* + * Disable local preemption by default because it requires + * user-space to be aware of it and provide additional handling + * to restore rendering state or do various flushes on switch. + */ + OUT_PKT7(ring, CP_PREEMPT_ENABLE_LOCAL, 1); + OUT_RING(ring, 0x0); + + /* Allow CP_CONTEXT_SWITCH_YIELD packets in the IB2 */ + OUT_PKT7(ring, CP_YIELD_ENABLE, 1); + OUT_RING(ring, 0x02); + + /* Submit the commands */ + for (i = 0; i < submit->nr_cmds; i++) { + switch (submit->cmd[i].type) { + case MSM_SUBMIT_CMD_IB_TARGET_BUF: + break; + case MSM_SUBMIT_CMD_CTX_RESTORE_BUF: + if (ring->cur_ctx_seqno == submit->queue->ctx->seqno) + break; + fallthrough; + case MSM_SUBMIT_CMD_BUF: + OUT_PKT7(ring, CP_INDIRECT_BUFFER_PFE, 3); + OUT_RING(ring, lower_32_bits(submit->cmd[i].iova)); + OUT_RING(ring, upper_32_bits(submit->cmd[i].iova)); + OUT_RING(ring, submit->cmd[i].size); + ibs++; + break; + } + + /* + * Periodically update shadow-wptr if needed, so that we + * can see partial progress of submits with large # of + * cmds.. otherwise we could needlessly stall waiting for + * ringbuffer state, simply due to looking at a shadow + * rptr value that has not been updated + */ + if ((ibs % 32) == 0) + update_shadow_rptr(gpu, ring); + } + + /* + * Write the render mode to NULL (0) to indicate to the CP that the IBs + * are done rendering - otherwise a lucky preemption would start + * replaying from the last checkpoint + */ + OUT_PKT7(ring, CP_SET_RENDER_MODE, 5); + OUT_RING(ring, 0); + OUT_RING(ring, 0); + OUT_RING(ring, 0); + OUT_RING(ring, 0); + OUT_RING(ring, 0); + + /* Turn off IB level preemptions */ + OUT_PKT7(ring, CP_YIELD_ENABLE, 1); + OUT_RING(ring, 0x01); + + /* Write the fence to the scratch register */ + OUT_PKT4(ring, REG_A5XX_CP_SCRATCH_REG(2), 1); + OUT_RING(ring, submit->seqno); + a5xx_gpu->last_seqno[ring->id] = submit->seqno; + + /* + * Execute a CACHE_FLUSH_TS event. This will ensure that the + * timestamp is written to the memory and then triggers the interrupt + */ + OUT_PKT7(ring, CP_EVENT_WRITE, 4); + OUT_RING(ring, CP_EVENT_WRITE_0_EVENT(CACHE_FLUSH_TS) | + CP_EVENT_WRITE_0_IRQ); + OUT_RING(ring, lower_32_bits(rbmemptr(ring, fence))); + OUT_RING(ring, upper_32_bits(rbmemptr(ring, fence))); + OUT_RING(ring, submit->seqno); + + /* Yield the floor on command completion */ + OUT_PKT7(ring, CP_CONTEXT_SWITCH_YIELD, 4); + /* + * If dword[2:1] are non zero, they specify an address for the CP to + * write the value of dword[3] to on preemption complete. Write 0 to + * skip the write + */ + OUT_RING(ring, 0x00); + OUT_RING(ring, 0x00); + /* Data value - not used if the address above is 0 */ + OUT_RING(ring, 0x01); + /* Set bit 0 to trigger an interrupt on preempt complete */ + OUT_RING(ring, 0x01); + + /* A WHERE_AM_I packet is not needed after a YIELD */ + a5xx_flush(gpu, ring, false); + + /* Check to see if we need to start preemption */ + a5xx_preempt_trigger(gpu); +} + +static const struct adreno_five_hwcg_regs { + u32 offset; + u32 value; +} a5xx_hwcg[] = { + {REG_A5XX_RBBM_CLOCK_CNTL_SP0, 0x02222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_SP1, 0x02222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_SP2, 0x02222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_SP3, 0x02222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220}, + {REG_A5XX_RBBM_CLOCK_CNTL2_SP1, 0x02222220}, + {REG_A5XX_RBBM_CLOCK_CNTL2_SP2, 0x02222220}, + {REG_A5XX_RBBM_CLOCK_CNTL2_SP3, 0x02222220}, + {REG_A5XX_RBBM_CLOCK_HYST_SP0, 0x0000F3CF}, + {REG_A5XX_RBBM_CLOCK_HYST_SP1, 0x0000F3CF}, + {REG_A5XX_RBBM_CLOCK_HYST_SP2, 0x0000F3CF}, + {REG_A5XX_RBBM_CLOCK_HYST_SP3, 0x0000F3CF}, + {REG_A5XX_RBBM_CLOCK_DELAY_SP0, 0x00000080}, + {REG_A5XX_RBBM_CLOCK_DELAY_SP1, 0x00000080}, + {REG_A5XX_RBBM_CLOCK_DELAY_SP2, 0x00000080}, + {REG_A5XX_RBBM_CLOCK_DELAY_SP3, 0x00000080}, + {REG_A5XX_RBBM_CLOCK_CNTL_TP0, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_TP1, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_TP2, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_TP3, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_TP1, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_TP2, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_TP3, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL3_TP0, 0x00002222}, + {REG_A5XX_RBBM_CLOCK_CNTL3_TP1, 0x00002222}, + {REG_A5XX_RBBM_CLOCK_CNTL3_TP2, 0x00002222}, + {REG_A5XX_RBBM_CLOCK_CNTL3_TP3, 0x00002222}, + {REG_A5XX_RBBM_CLOCK_HYST_TP0, 0x77777777}, + {REG_A5XX_RBBM_CLOCK_HYST_TP1, 0x77777777}, + {REG_A5XX_RBBM_CLOCK_HYST_TP2, 0x77777777}, + {REG_A5XX_RBBM_CLOCK_HYST_TP3, 0x77777777}, + {REG_A5XX_RBBM_CLOCK_HYST2_TP0, 0x77777777}, + {REG_A5XX_RBBM_CLOCK_HYST2_TP1, 0x77777777}, + {REG_A5XX_RBBM_CLOCK_HYST2_TP2, 0x77777777}, + {REG_A5XX_RBBM_CLOCK_HYST2_TP3, 0x77777777}, + {REG_A5XX_RBBM_CLOCK_HYST3_TP0, 0x00007777}, + {REG_A5XX_RBBM_CLOCK_HYST3_TP1, 0x00007777}, + {REG_A5XX_RBBM_CLOCK_HYST3_TP2, 0x00007777}, + {REG_A5XX_RBBM_CLOCK_HYST3_TP3, 0x00007777}, + {REG_A5XX_RBBM_CLOCK_DELAY_TP0, 0x11111111}, + {REG_A5XX_RBBM_CLOCK_DELAY_TP1, 0x11111111}, + {REG_A5XX_RBBM_CLOCK_DELAY_TP2, 0x11111111}, + {REG_A5XX_RBBM_CLOCK_DELAY_TP3, 0x11111111}, + {REG_A5XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111}, + {REG_A5XX_RBBM_CLOCK_DELAY2_TP1, 0x11111111}, + {REG_A5XX_RBBM_CLOCK_DELAY2_TP2, 0x11111111}, + {REG_A5XX_RBBM_CLOCK_DELAY2_TP3, 0x11111111}, + {REG_A5XX_RBBM_CLOCK_DELAY3_TP0, 0x00001111}, + {REG_A5XX_RBBM_CLOCK_DELAY3_TP1, 0x00001111}, + {REG_A5XX_RBBM_CLOCK_DELAY3_TP2, 0x00001111}, + {REG_A5XX_RBBM_CLOCK_DELAY3_TP3, 0x00001111}, + {REG_A5XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_UCHE, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL3_UCHE, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL4_UCHE, 0x00222222}, + {REG_A5XX_RBBM_CLOCK_HYST_UCHE, 0x00444444}, + {REG_A5XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002}, + {REG_A5XX_RBBM_CLOCK_CNTL_RB0, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_RB1, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_RB2, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_RB3, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_RB0, 0x00222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_RB1, 0x00222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_RB2, 0x00222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_RB3, 0x00222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_CCU0, 0x00022220}, + {REG_A5XX_RBBM_CLOCK_CNTL_CCU1, 0x00022220}, + {REG_A5XX_RBBM_CLOCK_CNTL_CCU2, 0x00022220}, + {REG_A5XX_RBBM_CLOCK_CNTL_CCU3, 0x00022220}, + {REG_A5XX_RBBM_CLOCK_CNTL_RAC, 0x05522222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_RAC, 0x00505555}, + {REG_A5XX_RBBM_CLOCK_HYST_RB_CCU0, 0x04040404}, + {REG_A5XX_RBBM_CLOCK_HYST_RB_CCU1, 0x04040404}, + {REG_A5XX_RBBM_CLOCK_HYST_RB_CCU2, 0x04040404}, + {REG_A5XX_RBBM_CLOCK_HYST_RB_CCU3, 0x04040404}, + {REG_A5XX_RBBM_CLOCK_HYST_RAC, 0x07444044}, + {REG_A5XX_RBBM_CLOCK_DELAY_RB_CCU_L1_0, 0x00000002}, + {REG_A5XX_RBBM_CLOCK_DELAY_RB_CCU_L1_1, 0x00000002}, + {REG_A5XX_RBBM_CLOCK_DELAY_RB_CCU_L1_2, 0x00000002}, + {REG_A5XX_RBBM_CLOCK_DELAY_RB_CCU_L1_3, 0x00000002}, + {REG_A5XX_RBBM_CLOCK_DELAY_RAC, 0x00010011}, + {REG_A5XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222}, + {REG_A5XX_RBBM_CLOCK_MODE_GPC, 0x02222222}, + {REG_A5XX_RBBM_CLOCK_MODE_VFD, 0x00002222}, + {REG_A5XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000}, + {REG_A5XX_RBBM_CLOCK_HYST_GPC, 0x04104004}, + {REG_A5XX_RBBM_CLOCK_HYST_VFD, 0x00000000}, + {REG_A5XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000}, + {REG_A5XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000}, + {REG_A5XX_RBBM_CLOCK_DELAY_GPC, 0x00000200}, + {REG_A5XX_RBBM_CLOCK_DELAY_VFD, 0x00002222} +}, a50x_hwcg[] = { + {REG_A5XX_RBBM_CLOCK_CNTL_SP0, 0x02222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220}, + {REG_A5XX_RBBM_CLOCK_HYST_SP0, 0x0000F3CF}, + {REG_A5XX_RBBM_CLOCK_DELAY_SP0, 0x00000080}, + {REG_A5XX_RBBM_CLOCK_CNTL_TP0, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL3_TP0, 0x00002222}, + {REG_A5XX_RBBM_CLOCK_HYST_TP0, 0x77777777}, + {REG_A5XX_RBBM_CLOCK_HYST2_TP0, 0x77777777}, + {REG_A5XX_RBBM_CLOCK_HYST3_TP0, 0x00007777}, + {REG_A5XX_RBBM_CLOCK_DELAY_TP0, 0x11111111}, + {REG_A5XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111}, + {REG_A5XX_RBBM_CLOCK_DELAY3_TP0, 0x00001111}, + {REG_A5XX_RBBM_CLOCK_CNTL2_UCHE, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL3_UCHE, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL4_UCHE, 0x00222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_HYST_UCHE, 0x00FFFFF4}, + {REG_A5XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002}, + {REG_A5XX_RBBM_CLOCK_CNTL_RB0, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_RB0, 0x00222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_CCU0, 0x00022220}, + {REG_A5XX_RBBM_CLOCK_CNTL_RAC, 0x05522222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_RAC, 0x00505555}, + {REG_A5XX_RBBM_CLOCK_HYST_RB_CCU0, 0x04040404}, + {REG_A5XX_RBBM_CLOCK_HYST_RAC, 0x07444044}, + {REG_A5XX_RBBM_CLOCK_DELAY_RB_CCU_L1_0, 0x00000002}, + {REG_A5XX_RBBM_CLOCK_DELAY_RAC, 0x00010011}, + {REG_A5XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222}, + {REG_A5XX_RBBM_CLOCK_MODE_GPC, 0x02222222}, + {REG_A5XX_RBBM_CLOCK_MODE_VFD, 0x00002222}, + {REG_A5XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000}, + {REG_A5XX_RBBM_CLOCK_HYST_GPC, 0x04104004}, + {REG_A5XX_RBBM_CLOCK_HYST_VFD, 0x00000000}, + {REG_A5XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000}, + {REG_A5XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000}, + {REG_A5XX_RBBM_CLOCK_DELAY_GPC, 0x00000200}, + {REG_A5XX_RBBM_CLOCK_DELAY_VFD, 0x00002222}, +}, a512_hwcg[] = { + {REG_A5XX_RBBM_CLOCK_CNTL_SP0, 0x02222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_SP1, 0x02222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220}, + {REG_A5XX_RBBM_CLOCK_CNTL2_SP1, 0x02222220}, + {REG_A5XX_RBBM_CLOCK_HYST_SP0, 0x0000F3CF}, + {REG_A5XX_RBBM_CLOCK_HYST_SP1, 0x0000F3CF}, + {REG_A5XX_RBBM_CLOCK_DELAY_SP0, 0x00000080}, + {REG_A5XX_RBBM_CLOCK_DELAY_SP1, 0x00000080}, + {REG_A5XX_RBBM_CLOCK_CNTL_TP0, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_TP1, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_TP1, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL3_TP0, 0x00002222}, + {REG_A5XX_RBBM_CLOCK_CNTL3_TP1, 0x00002222}, + {REG_A5XX_RBBM_CLOCK_HYST_TP0, 0x77777777}, + {REG_A5XX_RBBM_CLOCK_HYST_TP1, 0x77777777}, + {REG_A5XX_RBBM_CLOCK_HYST2_TP0, 0x77777777}, + {REG_A5XX_RBBM_CLOCK_HYST2_TP1, 0x77777777}, + {REG_A5XX_RBBM_CLOCK_HYST3_TP0, 0x00007777}, + {REG_A5XX_RBBM_CLOCK_HYST3_TP1, 0x00007777}, + {REG_A5XX_RBBM_CLOCK_DELAY_TP0, 0x11111111}, + {REG_A5XX_RBBM_CLOCK_DELAY_TP1, 0x11111111}, + {REG_A5XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111}, + {REG_A5XX_RBBM_CLOCK_DELAY2_TP1, 0x11111111}, + {REG_A5XX_RBBM_CLOCK_DELAY3_TP0, 0x00001111}, + {REG_A5XX_RBBM_CLOCK_DELAY3_TP1, 0x00001111}, + {REG_A5XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_UCHE, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL3_UCHE, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL4_UCHE, 0x00222222}, + {REG_A5XX_RBBM_CLOCK_HYST_UCHE, 0x00444444}, + {REG_A5XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002}, + {REG_A5XX_RBBM_CLOCK_CNTL_RB0, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_RB1, 0x22222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_RB0, 0x00222222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_RB1, 0x00222222}, + {REG_A5XX_RBBM_CLOCK_CNTL_CCU0, 0x00022220}, + {REG_A5XX_RBBM_CLOCK_CNTL_CCU1, 0x00022220}, + {REG_A5XX_RBBM_CLOCK_CNTL_RAC, 0x05522222}, + {REG_A5XX_RBBM_CLOCK_CNTL2_RAC, 0x00505555}, + {REG_A5XX_RBBM_CLOCK_HYST_RB_CCU0, 0x04040404}, + {REG_A5XX_RBBM_CLOCK_HYST_RB_CCU1, 0x04040404}, + {REG_A5XX_RBBM_CLOCK_HYST_RAC, 0x07444044}, + {REG_A5XX_RBBM_CLOCK_DELAY_RB_CCU_L1_0, 0x00000002}, + {REG_A5XX_RBBM_CLOCK_DELAY_RB_CCU_L1_1, 0x00000002}, + {REG_A5XX_RBBM_CLOCK_DELAY_RAC, 0x00010011}, + {REG_A5XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222}, + {REG_A5XX_RBBM_CLOCK_MODE_GPC, 0x02222222}, + {REG_A5XX_RBBM_CLOCK_MODE_VFD, 0x00002222}, + {REG_A5XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000}, + {REG_A5XX_RBBM_CLOCK_HYST_GPC, 0x04104004}, + {REG_A5XX_RBBM_CLOCK_HYST_VFD, 0x00000000}, + {REG_A5XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000}, + {REG_A5XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000}, + {REG_A5XX_RBBM_CLOCK_DELAY_GPC, 0x00000200}, + {REG_A5XX_RBBM_CLOCK_DELAY_VFD, 0x00002222}, +}; + +void a5xx_set_hwcg(struct msm_gpu *gpu, bool state) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + const struct adreno_five_hwcg_regs *regs; + unsigned int i, sz; + + if (adreno_is_a505(adreno_gpu) || adreno_is_a506(adreno_gpu) || + adreno_is_a508(adreno_gpu)) { + regs = a50x_hwcg; + sz = ARRAY_SIZE(a50x_hwcg); + } else if (adreno_is_a509(adreno_gpu) || adreno_is_a512(adreno_gpu)) { + regs = a512_hwcg; + sz = ARRAY_SIZE(a512_hwcg); + } else { + regs = a5xx_hwcg; + sz = ARRAY_SIZE(a5xx_hwcg); + } + + for (i = 0; i < sz; i++) + gpu_write(gpu, regs[i].offset, + state ? regs[i].value : 0); + + if (adreno_is_a540(adreno_gpu)) { + gpu_write(gpu, REG_A5XX_RBBM_CLOCK_DELAY_GPMU, state ? 0x00000770 : 0); + gpu_write(gpu, REG_A5XX_RBBM_CLOCK_HYST_GPMU, state ? 0x00000004 : 0); + } + + gpu_write(gpu, REG_A5XX_RBBM_CLOCK_CNTL, state ? 0xAAA8AA00 : 0); + gpu_write(gpu, REG_A5XX_RBBM_ISDB_CNT, state ? 0x182 : 0x180); +} + +static int a5xx_me_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct msm_ringbuffer *ring = gpu->rb[0]; + + OUT_PKT7(ring, CP_ME_INIT, 8); + + OUT_RING(ring, 0x0000002F); + + /* Enable multiple hardware contexts */ + OUT_RING(ring, 0x00000003); + + /* Enable error detection */ + OUT_RING(ring, 0x20000000); + + /* Don't enable header dump */ + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + + /* Specify workarounds for various microcode issues */ + if (adreno_is_a505(adreno_gpu) || adreno_is_a506(adreno_gpu) || + adreno_is_a530(adreno_gpu)) { + /* Workaround for token end syncs + * Force a WFI after every direct-render 3D mode draw and every + * 2D mode 3 draw + */ + OUT_RING(ring, 0x0000000B); + } else if (adreno_is_a510(adreno_gpu)) { + /* Workaround for token and syncs */ + OUT_RING(ring, 0x00000001); + } else { + /* No workarounds enabled */ + OUT_RING(ring, 0x00000000); + } + + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + + a5xx_flush(gpu, ring, true); + return a5xx_idle(gpu, ring) ? 0 : -EINVAL; +} + +static int a5xx_preempt_start(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + struct msm_ringbuffer *ring = gpu->rb[0]; + + if (gpu->nr_rings == 1) + return 0; + + /* Turn off protected mode to write to special registers */ + OUT_PKT7(ring, CP_SET_PROTECTED_MODE, 1); + OUT_RING(ring, 0); + + /* Set the save preemption record for the ring/command */ + OUT_PKT4(ring, REG_A5XX_CP_CONTEXT_SWITCH_SAVE_ADDR_LO, 2); + OUT_RING(ring, lower_32_bits(a5xx_gpu->preempt_iova[ring->id])); + OUT_RING(ring, upper_32_bits(a5xx_gpu->preempt_iova[ring->id])); + + /* Turn back on protected mode */ + OUT_PKT7(ring, CP_SET_PROTECTED_MODE, 1); + OUT_RING(ring, 1); + + OUT_PKT7(ring, CP_PREEMPT_ENABLE_GLOBAL, 1); + OUT_RING(ring, 0x00); + + OUT_PKT7(ring, CP_PREEMPT_ENABLE_LOCAL, 1); + OUT_RING(ring, 0x01); + + OUT_PKT7(ring, CP_YIELD_ENABLE, 1); + OUT_RING(ring, 0x01); + + /* Yield the floor on command completion */ + OUT_PKT7(ring, CP_CONTEXT_SWITCH_YIELD, 4); + OUT_RING(ring, 0x00); + OUT_RING(ring, 0x00); + OUT_RING(ring, 0x01); + OUT_RING(ring, 0x01); + + /* The WHERE_AMI_I packet is not needed after a YIELD is issued */ + a5xx_flush(gpu, ring, false); + + return a5xx_idle(gpu, ring) ? 0 : -EINVAL; +} + +static void a5xx_ucode_check_version(struct a5xx_gpu *a5xx_gpu, + struct drm_gem_object *obj) +{ + u32 *buf = msm_gem_get_vaddr(obj); + + if (IS_ERR(buf)) + return; + + /* + * If the lowest nibble is 0xa that is an indication that this microcode + * has been patched. The actual version is in dword [3] but we only care + * about the patchlevel which is the lowest nibble of dword [3] + */ + if (((buf[0] & 0xf) == 0xa) && (buf[2] & 0xf) >= 1) + a5xx_gpu->has_whereami = true; + + msm_gem_put_vaddr(obj); +} + +static int a5xx_ucode_load(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + int ret; + + if (!a5xx_gpu->pm4_bo) { + a5xx_gpu->pm4_bo = adreno_fw_create_bo(gpu, + adreno_gpu->fw[ADRENO_FW_PM4], &a5xx_gpu->pm4_iova); + + + if (IS_ERR(a5xx_gpu->pm4_bo)) { + ret = PTR_ERR(a5xx_gpu->pm4_bo); + a5xx_gpu->pm4_bo = NULL; + DRM_DEV_ERROR(gpu->dev->dev, "could not allocate PM4: %d\n", + ret); + return ret; + } + + msm_gem_object_set_name(a5xx_gpu->pm4_bo, "pm4fw"); + } + + if (!a5xx_gpu->pfp_bo) { + a5xx_gpu->pfp_bo = adreno_fw_create_bo(gpu, + adreno_gpu->fw[ADRENO_FW_PFP], &a5xx_gpu->pfp_iova); + + if (IS_ERR(a5xx_gpu->pfp_bo)) { + ret = PTR_ERR(a5xx_gpu->pfp_bo); + a5xx_gpu->pfp_bo = NULL; + DRM_DEV_ERROR(gpu->dev->dev, "could not allocate PFP: %d\n", + ret); + return ret; + } + + msm_gem_object_set_name(a5xx_gpu->pfp_bo, "pfpfw"); + a5xx_ucode_check_version(a5xx_gpu, a5xx_gpu->pfp_bo); + } + + if (a5xx_gpu->has_whereami) { + if (!a5xx_gpu->shadow_bo) { + a5xx_gpu->shadow = msm_gem_kernel_new(gpu->dev, + sizeof(u32) * gpu->nr_rings, + MSM_BO_WC | MSM_BO_MAP_PRIV, + gpu->vm, &a5xx_gpu->shadow_bo, + &a5xx_gpu->shadow_iova); + + if (IS_ERR(a5xx_gpu->shadow)) + return PTR_ERR(a5xx_gpu->shadow); + + msm_gem_object_set_name(a5xx_gpu->shadow_bo, "shadow"); + } + } else if (gpu->nr_rings > 1) { + /* Disable preemption if WHERE_AM_I isn't available */ + a5xx_preempt_fini(gpu); + gpu->nr_rings = 1; + } + + return 0; +} + +#define SCM_GPU_ZAP_SHADER_RESUME 0 + +static int a5xx_zap_shader_resume(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + int ret; + + /* + * Adreno 506 have CPZ Retention feature and doesn't require + * to resume zap shader + */ + if (adreno_is_a506(adreno_gpu)) + return 0; + + ret = qcom_pas_set_remote_state(SCM_GPU_ZAP_SHADER_RESUME, GPU_PAS_ID); + if (ret) + DRM_ERROR("%s: zap-shader resume failed: %d\n", + gpu->name, ret); + + return ret; +} + +static int a5xx_zap_shader_init(struct msm_gpu *gpu) +{ + static bool loaded; + int ret; + + /* + * If the zap shader is already loaded into memory we just need to kick + * the remote processor to reinitialize it + */ + if (loaded) + return a5xx_zap_shader_resume(gpu); + + ret = adreno_zap_shader_load(gpu, GPU_PAS_ID); + + loaded = !ret; + return ret; +} + +#define A5XX_INT_MASK (A5XX_RBBM_INT_0_MASK_RBBM_AHB_ERROR | \ + A5XX_RBBM_INT_0_MASK_RBBM_TRANSFER_TIMEOUT | \ + A5XX_RBBM_INT_0_MASK_RBBM_ME_MS_TIMEOUT | \ + A5XX_RBBM_INT_0_MASK_RBBM_PFP_MS_TIMEOUT | \ + A5XX_RBBM_INT_0_MASK_RBBM_ETS_MS_TIMEOUT | \ + A5XX_RBBM_INT_0_MASK_RBBM_ATB_ASYNC_OVERFLOW | \ + A5XX_RBBM_INT_0_MASK_CP_HW_ERROR | \ + A5XX_RBBM_INT_0_MASK_MISC_HANG_DETECT | \ + A5XX_RBBM_INT_0_MASK_CP_SW | \ + A5XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS | \ + A5XX_RBBM_INT_0_MASK_UCHE_OOB_ACCESS | \ + A5XX_RBBM_INT_0_MASK_GPMU_VOLTAGE_DROOP) + +static int a5xx_hw_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + u32 hbb; + int ret; + + gpu_write(gpu, REG_A5XX_VBIF_ROUND_ROBIN_QOS_ARB, 0x00000003); + + if (adreno_is_a509(adreno_gpu) || adreno_is_a512(adreno_gpu) || + adreno_is_a540(adreno_gpu)) + gpu_write(gpu, REG_A5XX_VBIF_GATE_OFF_WRREQ_EN, 0x00000009); + + /* Make all blocks contribute to the GPU BUSY perf counter */ + gpu_write(gpu, REG_A5XX_RBBM_PERFCTR_GPU_BUSY_MASKED, 0xFFFFFFFF); + + /* Enable RBBM error reporting bits */ + gpu_write(gpu, REG_A5XX_RBBM_AHB_CNTL0, 0x00000001); + + if (adreno_gpu->info->quirks & ADRENO_QUIRK_FAULT_DETECT_MASK) { + /* + * Mask out the activity signals from RB1-3 to avoid false + * positives + */ + + gpu_write(gpu, REG_A5XX_RBBM_INTERFACE_HANG_MASK_CNTL11, + 0xF0000000); + gpu_write(gpu, REG_A5XX_RBBM_INTERFACE_HANG_MASK_CNTL12, + 0xFFFFFFFF); + gpu_write(gpu, REG_A5XX_RBBM_INTERFACE_HANG_MASK_CNTL13, + 0xFFFFFFFF); + gpu_write(gpu, REG_A5XX_RBBM_INTERFACE_HANG_MASK_CNTL14, + 0xFFFFFFFF); + gpu_write(gpu, REG_A5XX_RBBM_INTERFACE_HANG_MASK_CNTL15, + 0xFFFFFFFF); + gpu_write(gpu, REG_A5XX_RBBM_INTERFACE_HANG_MASK_CNTL16, + 0xFFFFFFFF); + gpu_write(gpu, REG_A5XX_RBBM_INTERFACE_HANG_MASK_CNTL17, + 0xFFFFFFFF); + gpu_write(gpu, REG_A5XX_RBBM_INTERFACE_HANG_MASK_CNTL18, + 0xFFFFFFFF); + } + + /* Enable fault detection */ + gpu_write(gpu, REG_A5XX_RBBM_INTERFACE_HANG_INT_CNTL, + (1 << 30) | 0xFFFF); + + /* Turn on performance counters */ + gpu_write(gpu, REG_A5XX_RBBM_PERFCTR_CNTL, 0x01); + + /* Select CP0 to always count cycles */ + gpu_write(gpu, REG_A5XX_CP_PERFCTR_CP_SEL_0, PERF_CP_ALWAYS_COUNT); + + /* Select RBBM0 to countable 6 to get the busy status for devfreq */ + gpu_write(gpu, REG_A5XX_RBBM_PERFCTR_RBBM_SEL_0, 6); + + /* Increase VFD cache access so LRZ and other data gets evicted less */ + gpu_write(gpu, REG_A5XX_UCHE_CACHE_WAYS, 0x02); + + /* Disable L2 bypass in the UCHE */ + gpu_write64(gpu, REG_A5XX_UCHE_TRAP_BASE, adreno_gpu->uche_trap_base); + gpu_write64(gpu, REG_A5XX_UCHE_WRITE_THRU_BASE, adreno_gpu->uche_trap_base); + + /* Set the GMEM VA range (0 to gpu->gmem) */ + gpu_write64(gpu, REG_A5XX_UCHE_GMEM_RANGE_MIN, 0x00100000); + gpu_write64(gpu, REG_A5XX_UCHE_GMEM_RANGE_MAX, + 0x00100000 + adreno_gpu->info->gmem - 1); + + if (adreno_is_a505(adreno_gpu) || adreno_is_a506(adreno_gpu) || + adreno_is_a508(adreno_gpu) || adreno_is_a510(adreno_gpu)) { + gpu_write(gpu, REG_A5XX_CP_MEQ_THRESHOLDS, 0x20); + if (adreno_is_a505(adreno_gpu) || adreno_is_a506(adreno_gpu) || + adreno_is_a508(adreno_gpu)) + gpu_write(gpu, REG_A5XX_CP_MERCIU_SIZE, 0x400); + else + gpu_write(gpu, REG_A5XX_CP_MERCIU_SIZE, 0x20); + gpu_write(gpu, REG_A5XX_CP_ROQ_THRESHOLDS_2, 0x40000030); + gpu_write(gpu, REG_A5XX_CP_ROQ_THRESHOLDS_1, 0x20100D0A); + } else { + gpu_write(gpu, REG_A5XX_CP_MEQ_THRESHOLDS, 0x40); + if (adreno_is_a530(adreno_gpu)) + gpu_write(gpu, REG_A5XX_CP_MERCIU_SIZE, 0x40); + else + gpu_write(gpu, REG_A5XX_CP_MERCIU_SIZE, 0x400); + gpu_write(gpu, REG_A5XX_CP_ROQ_THRESHOLDS_2, 0x80000060); + gpu_write(gpu, REG_A5XX_CP_ROQ_THRESHOLDS_1, 0x40201B16); + } + + if (adreno_is_a505(adreno_gpu) || adreno_is_a506(adreno_gpu) || + adreno_is_a508(adreno_gpu)) + gpu_write(gpu, REG_A5XX_PC_DBG_ECO_CNTL, + (0x100 << 11 | 0x100 << 22)); + else if (adreno_is_a509(adreno_gpu) || adreno_is_a510(adreno_gpu) || + adreno_is_a512(adreno_gpu)) + gpu_write(gpu, REG_A5XX_PC_DBG_ECO_CNTL, + (0x200 << 11 | 0x200 << 22)); + else + gpu_write(gpu, REG_A5XX_PC_DBG_ECO_CNTL, + (0x400 << 11 | 0x300 << 22)); + + if (adreno_gpu->info->quirks & ADRENO_QUIRK_TWO_PASS_USE_WFI) + gpu_rmw(gpu, REG_A5XX_PC_DBG_ECO_CNTL, 0, (1 << 8)); + + /* + * Disable the RB sampler datapath DP2 clock gating optimization + * for 1-SP GPUs, as it is enabled by default. + */ + if (adreno_is_a505(adreno_gpu) || adreno_is_a506(adreno_gpu) || + adreno_is_a508(adreno_gpu) || adreno_is_a509(adreno_gpu) || + adreno_is_a512(adreno_gpu)) + gpu_rmw(gpu, REG_A5XX_RB_DBG_ECO_CNTL, 0, (1 << 9)); + + /* Disable UCHE global filter as SP can invalidate/flush independently */ + gpu_write(gpu, REG_A5XX_UCHE_MODE_CNTL, BIT(29)); + + /* Enable USE_RETENTION_FLOPS */ + gpu_write(gpu, REG_A5XX_CP_CHICKEN_DBG, 0x02000000); + + /* Enable ME/PFP split notification */ + gpu_write(gpu, REG_A5XX_RBBM_AHB_CNTL1, 0xA6FFFFFF); + + /* + * In A5x, CCU can send context_done event of a particular context to + * UCHE which ultimately reaches CP even when there is valid + * transaction of that context inside CCU. This can let CP to program + * config registers, which will make the "valid transaction" inside + * CCU to be interpreted differently. This can cause gpu fault. This + * bug is fixed in latest A510 revision. To enable this bug fix - + * bit[11] of RB_DBG_ECO_CNTL need to be set to 0, default is 1 + * (disable). For older A510 version this bit is unused. + */ + if (adreno_is_a510(adreno_gpu)) + gpu_rmw(gpu, REG_A5XX_RB_DBG_ECO_CNTL, (1 << 11), 0); + + /* Enable HWCG */ + a5xx_set_hwcg(gpu, true); + + gpu_write(gpu, REG_A5XX_RBBM_AHB_CNTL2, 0x0000003F); + + BUG_ON(adreno_gpu->ubwc_config->highest_bank_bit < 13); + hbb = adreno_gpu->ubwc_config->highest_bank_bit - 13; + + gpu_write(gpu, REG_A5XX_TPL1_MODE_CNTL, hbb << 7); + gpu_write(gpu, REG_A5XX_RB_MODE_CNTL, hbb << 1); + + if (adreno_is_a509(adreno_gpu) || adreno_is_a512(adreno_gpu) || + adreno_is_a540(adreno_gpu)) + gpu_write(gpu, REG_A5XX_UCHE_DBG_ECO_CNTL_2, hbb); + + /* Disable All flat shading optimization (ALLFLATOPTDIS) */ + gpu_rmw(gpu, REG_A5XX_VPC_DBG_ECO_CNTL, 0, (1 << 10)); + + /* Protect registers from the CP */ + gpu_write(gpu, REG_A5XX_CP_PROTECT_CNTL, 0x00000007); + + /* RBBM */ + gpu_write(gpu, REG_A5XX_CP_PROTECT(0), ADRENO_PROTECT_RW(0x04, 4)); + gpu_write(gpu, REG_A5XX_CP_PROTECT(1), ADRENO_PROTECT_RW(0x08, 8)); + gpu_write(gpu, REG_A5XX_CP_PROTECT(2), ADRENO_PROTECT_RW(0x10, 16)); + gpu_write(gpu, REG_A5XX_CP_PROTECT(3), ADRENO_PROTECT_RW(0x20, 32)); + gpu_write(gpu, REG_A5XX_CP_PROTECT(4), ADRENO_PROTECT_RW(0x40, 64)); + gpu_write(gpu, REG_A5XX_CP_PROTECT(5), ADRENO_PROTECT_RW(0x80, 64)); + + /* Content protect */ + gpu_write(gpu, REG_A5XX_CP_PROTECT(6), + ADRENO_PROTECT_RW(REG_A5XX_RBBM_SECVID_TSB_TRUSTED_BASE_LO, + 16)); + gpu_write(gpu, REG_A5XX_CP_PROTECT(7), + ADRENO_PROTECT_RW(REG_A5XX_RBBM_SECVID_TRUST_CNTL, 2)); + + /* CP */ + gpu_write(gpu, REG_A5XX_CP_PROTECT(8), ADRENO_PROTECT_RW(0x800, 64)); + gpu_write(gpu, REG_A5XX_CP_PROTECT(9), ADRENO_PROTECT_RW(0x840, 8)); + gpu_write(gpu, REG_A5XX_CP_PROTECT(10), ADRENO_PROTECT_RW(0x880, 32)); + gpu_write(gpu, REG_A5XX_CP_PROTECT(11), ADRENO_PROTECT_RW(0xAA0, 1)); + + /* RB */ + gpu_write(gpu, REG_A5XX_CP_PROTECT(12), ADRENO_PROTECT_RW(0xCC0, 1)); + gpu_write(gpu, REG_A5XX_CP_PROTECT(13), ADRENO_PROTECT_RW(0xCF0, 2)); + + /* VPC */ + gpu_write(gpu, REG_A5XX_CP_PROTECT(14), ADRENO_PROTECT_RW(0xE68, 8)); + gpu_write(gpu, REG_A5XX_CP_PROTECT(15), ADRENO_PROTECT_RW(0xE70, 16)); + + /* UCHE */ + gpu_write(gpu, REG_A5XX_CP_PROTECT(16), ADRENO_PROTECT_RW(0xE80, 16)); + + /* SMMU */ + gpu_write(gpu, REG_A5XX_CP_PROTECT(17), + ADRENO_PROTECT_RW(0x10000, 0x8000)); + + gpu_write(gpu, REG_A5XX_RBBM_SECVID_TSB_CNTL, 0); + /* + * Disable the trusted memory range - we don't actually supported secure + * memory rendering at this point in time and we don't want to block off + * part of the virtual memory space. + */ + gpu_write64(gpu, REG_A5XX_RBBM_SECVID_TSB_TRUSTED_BASE_LO, 0x00000000); + gpu_write(gpu, REG_A5XX_RBBM_SECVID_TSB_TRUSTED_SIZE, 0x00000000); + + /* Put the GPU into 64 bit by default */ + gpu_write(gpu, REG_A5XX_CP_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A5XX_VSC_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A5XX_GRAS_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A5XX_RB_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A5XX_PC_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A5XX_HLSQ_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A5XX_VFD_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A5XX_VPC_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A5XX_UCHE_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A5XX_SP_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A5XX_TPL1_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A5XX_RBBM_SECVID_TSB_ADDR_MODE_CNTL, 0x1); + + /* + * VPC corner case with local memory load kill leads to corrupt + * internal state. Normal Disable does not work for all a5x chips. + * So do the following setting to disable it. + */ + if (adreno_gpu->info->quirks & ADRENO_QUIRK_LMLOADKILL_DISABLE) { + gpu_rmw(gpu, REG_A5XX_VPC_DBG_ECO_CNTL, 0, BIT(23)); + gpu_rmw(gpu, REG_A5XX_HLSQ_DBG_ECO_CNTL, BIT(18), 0); + } + + ret = adreno_hw_init(gpu); + if (ret) + return ret; + + if (adreno_is_a530(adreno_gpu) || adreno_is_a540(adreno_gpu)) + a5xx_gpmu_ucode_init(gpu); + + gpu_write64(gpu, REG_A5XX_CP_ME_INSTR_BASE_LO, a5xx_gpu->pm4_iova); + gpu_write64(gpu, REG_A5XX_CP_PFP_INSTR_BASE_LO, a5xx_gpu->pfp_iova); + + /* Set the ringbuffer address */ + gpu_write64(gpu, REG_A5XX_CP_RB_BASE, gpu->rb[0]->iova); + + /* + * If the microcode supports the WHERE_AM_I opcode then we can use that + * in lieu of the RPTR shadow and enable preemption. Otherwise, we + * can't safely use the RPTR shadow or preemption. In either case, the + * RPTR shadow should be disabled in hardware. + */ + gpu_write(gpu, REG_A5XX_CP_RB_CNTL, + MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE); + + /* Configure the RPTR shadow if needed: */ + if (a5xx_gpu->shadow_bo) { + gpu_write64(gpu, REG_A5XX_CP_RB_RPTR_ADDR, + shadowptr(a5xx_gpu, gpu->rb[0])); + } + + a5xx_preempt_hw_init(gpu); + + /* Disable the interrupts through the initial bringup stage */ + gpu_write(gpu, REG_A5XX_RBBM_INT_0_MASK, A5XX_INT_MASK); + + /* Clear ME_HALT to start the micro engine */ + gpu_write(gpu, REG_A5XX_CP_PFP_ME_CNTL, 0); + ret = a5xx_me_init(gpu); + if (ret) + return ret; + + ret = a5xx_power_init(gpu); + if (ret) + return ret; + + /* + * Send a pipeline event stat to get misbehaving counters to start + * ticking correctly + */ + if (adreno_is_a530(adreno_gpu)) { + OUT_PKT7(gpu->rb[0], CP_EVENT_WRITE, 1); + OUT_RING(gpu->rb[0], CP_EVENT_WRITE_0_EVENT(STAT_EVENT)); + + a5xx_flush(gpu, gpu->rb[0], true); + if (!a5xx_idle(gpu, gpu->rb[0])) + return -EINVAL; + } + + /* + * If the chip that we are using does support loading one, then + * try to load a zap shader into the secure world. If successful + * we can use the CP to switch out of secure mode. If not then we + * have no resource but to try to switch ourselves out manually. If we + * guessed wrong then access to the RBBM_SECVID_TRUST_CNTL register will + * be blocked and a permissions violation will soon follow. + */ + ret = a5xx_zap_shader_init(gpu); + if (!ret) { + OUT_PKT7(gpu->rb[0], CP_SET_SECURE_MODE, 1); + OUT_RING(gpu->rb[0], 0x00000000); + + a5xx_flush(gpu, gpu->rb[0], true); + if (!a5xx_idle(gpu, gpu->rb[0])) + return -EINVAL; + } else if (ret == -ENODEV) { + /* + * This device does not use zap shader (but print a warning + * just in case someone got their dt wrong.. hopefully they + * have a debug UART to realize the error of their ways... + * if you mess this up you are about to crash horribly) + */ + dev_warn_once(gpu->dev->dev, + "Zap shader not enabled - using SECVID_TRUST_CNTL instead\n"); + gpu_write(gpu, REG_A5XX_RBBM_SECVID_TRUST_CNTL, 0x0); + } else { + return ret; + } + + /* Last step - yield the ringbuffer */ + a5xx_preempt_start(gpu); + + return 0; +} + +static void a5xx_recover(struct msm_gpu *gpu) +{ + int i; + + adreno_dump_info(gpu); + + for (i = 0; i < 8; i++) { + printk("CP_SCRATCH_REG%d: %u\n", i, + gpu_read(gpu, REG_A5XX_CP_SCRATCH_REG(i))); + } + + if (hang_debug) + a5xx_dump(gpu); + + gpu_write(gpu, REG_A5XX_RBBM_SW_RESET_CMD, 1); + gpu_read(gpu, REG_A5XX_RBBM_SW_RESET_CMD); + gpu_write(gpu, REG_A5XX_RBBM_SW_RESET_CMD, 0); + adreno_recover(gpu); +} + +static void a5xx_destroy(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + + DBG("%s", gpu->name); + + a5xx_preempt_fini(gpu); + + if (a5xx_gpu->pm4_bo) { + msm_gem_unpin_iova(a5xx_gpu->pm4_bo, gpu->vm); + drm_gem_object_put(a5xx_gpu->pm4_bo); + } + + if (a5xx_gpu->pfp_bo) { + msm_gem_unpin_iova(a5xx_gpu->pfp_bo, gpu->vm); + drm_gem_object_put(a5xx_gpu->pfp_bo); + } + + if (a5xx_gpu->gpmu_bo) { + msm_gem_unpin_iova(a5xx_gpu->gpmu_bo, gpu->vm); + drm_gem_object_put(a5xx_gpu->gpmu_bo); + } + + if (a5xx_gpu->shadow_bo) { + msm_gem_unpin_iova(a5xx_gpu->shadow_bo, gpu->vm); + drm_gem_object_put(a5xx_gpu->shadow_bo); + } + + adreno_gpu_cleanup(adreno_gpu); + kfree(a5xx_gpu); +} + +static inline bool _a5xx_check_idle(struct msm_gpu *gpu) +{ + if (gpu_read(gpu, REG_A5XX_RBBM_STATUS) & ~A5XX_RBBM_STATUS_HI_BUSY) + return false; + + /* + * Nearly every abnormality ends up pausing the GPU and triggering a + * fault so we can safely just watch for this one interrupt to fire + */ + return !(gpu_read(gpu, REG_A5XX_RBBM_INT_0_STATUS) & + A5XX_RBBM_INT_0_MASK_MISC_HANG_DETECT); +} + +bool a5xx_idle(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + + if (ring != a5xx_gpu->cur_ring) { + WARN(1, "Tried to idle a non-current ringbuffer\n"); + return false; + } + + /* wait for CP to drain ringbuffer: */ + if (!adreno_idle(gpu, ring)) + return false; + + if (spin_until(_a5xx_check_idle(gpu))) { + DRM_ERROR("%s: %ps: timeout waiting for GPU to idle: status %8.8X irq %8.8X rptr/wptr %d/%d\n", + gpu->name, __builtin_return_address(0), + gpu_read(gpu, REG_A5XX_RBBM_STATUS), + gpu_read(gpu, REG_A5XX_RBBM_INT_0_STATUS), + gpu_read(gpu, REG_A5XX_CP_RB_RPTR), + gpu_read(gpu, REG_A5XX_CP_RB_WPTR)); + return false; + } + + return true; +} + +static int a5xx_fault_handler(void *arg, unsigned long iova, int flags, void *data) +{ + struct msm_gpu *gpu = arg; + struct adreno_smmu_fault_info *info = data; + char block[12] = "unknown"; + u32 scratch[] = { + gpu_read(gpu, REG_A5XX_CP_SCRATCH_REG(4)), + gpu_read(gpu, REG_A5XX_CP_SCRATCH_REG(5)), + gpu_read(gpu, REG_A5XX_CP_SCRATCH_REG(6)), + gpu_read(gpu, REG_A5XX_CP_SCRATCH_REG(7)), + }; + + if (info) + snprintf(block, sizeof(block), "%x", info->fsynr1); + + return adreno_fault_handler(gpu, iova, flags, info, block, scratch); +} + +static void a5xx_cp_err_irq(struct msm_gpu *gpu) +{ + u32 status = gpu_read(gpu, REG_A5XX_CP_INTERRUPT_STATUS); + + if (status & A5XX_CP_INT_CP_OPCODE_ERROR) { + u32 val; + + gpu_write(gpu, REG_A5XX_CP_PFP_STAT_ADDR, 0); + + /* + * REG_A5XX_CP_PFP_STAT_DATA is indexed, and we want index 1 so + * read it twice + */ + + gpu_read(gpu, REG_A5XX_CP_PFP_STAT_DATA); + val = gpu_read(gpu, REG_A5XX_CP_PFP_STAT_DATA); + + dev_err_ratelimited(gpu->dev->dev, "CP | opcode error | possible opcode=0x%8.8X\n", + val); + } + + if (status & A5XX_CP_INT_CP_HW_FAULT_ERROR) + dev_err_ratelimited(gpu->dev->dev, "CP | HW fault | status=0x%8.8X\n", + gpu_read(gpu, REG_A5XX_CP_HW_FAULT)); + + if (status & A5XX_CP_INT_CP_DMA_ERROR) + dev_err_ratelimited(gpu->dev->dev, "CP | DMA error\n"); + + if (status & A5XX_CP_INT_CP_REGISTER_PROTECTION_ERROR) { + u32 val = gpu_read(gpu, REG_A5XX_CP_PROTECT_STATUS); + + dev_err_ratelimited(gpu->dev->dev, + "CP | protected mode error | %s | addr=0x%8.8X | status=0x%8.8X\n", + val & (1 << 24) ? "WRITE" : "READ", + (val & 0xFFFFF) >> 2, val); + } + + if (status & A5XX_CP_INT_CP_AHB_ERROR) { + u32 status = gpu_read(gpu, REG_A5XX_CP_AHB_FAULT); + const char *access[16] = { "reserved", "reserved", + "timestamp lo", "timestamp hi", "pfp read", "pfp write", + "", "", "me read", "me write", "", "", "crashdump read", + "crashdump write" }; + + dev_err_ratelimited(gpu->dev->dev, + "CP | AHB error | addr=%X access=%s error=%d | status=0x%8.8X\n", + status & 0xFFFFF, access[(status >> 24) & 0xF], + (status & (1 << 31)), status); + } +} + +static void a5xx_rbbm_err_irq(struct msm_gpu *gpu, u32 status) +{ + if (status & A5XX_RBBM_INT_0_MASK_RBBM_AHB_ERROR) { + u32 val = gpu_read(gpu, REG_A5XX_RBBM_AHB_ERROR_STATUS); + + dev_err_ratelimited(gpu->dev->dev, + "RBBM | AHB bus error | %s | addr=0x%X | ports=0x%X:0x%X\n", + val & (1 << 28) ? "WRITE" : "READ", + (val & 0xFFFFF) >> 2, (val >> 20) & 0x3, + (val >> 24) & 0xF); + + /* Clear the error */ + gpu_write(gpu, REG_A5XX_RBBM_AHB_CMD, (1 << 4)); + + /* Clear the interrupt */ + gpu_write(gpu, REG_A5XX_RBBM_INT_CLEAR_CMD, + A5XX_RBBM_INT_0_MASK_RBBM_AHB_ERROR); + } + + if (status & A5XX_RBBM_INT_0_MASK_RBBM_TRANSFER_TIMEOUT) + dev_err_ratelimited(gpu->dev->dev, "RBBM | AHB transfer timeout\n"); + + if (status & A5XX_RBBM_INT_0_MASK_RBBM_ME_MS_TIMEOUT) + dev_err_ratelimited(gpu->dev->dev, "RBBM | ME master split | status=0x%X\n", + gpu_read(gpu, REG_A5XX_RBBM_AHB_ME_SPLIT_STATUS)); + + if (status & A5XX_RBBM_INT_0_MASK_RBBM_PFP_MS_TIMEOUT) + dev_err_ratelimited(gpu->dev->dev, "RBBM | PFP master split | status=0x%X\n", + gpu_read(gpu, REG_A5XX_RBBM_AHB_PFP_SPLIT_STATUS)); + + if (status & A5XX_RBBM_INT_0_MASK_RBBM_ETS_MS_TIMEOUT) + dev_err_ratelimited(gpu->dev->dev, "RBBM | ETS master split | status=0x%X\n", + gpu_read(gpu, REG_A5XX_RBBM_AHB_ETS_SPLIT_STATUS)); + + if (status & A5XX_RBBM_INT_0_MASK_RBBM_ATB_ASYNC_OVERFLOW) + dev_err_ratelimited(gpu->dev->dev, "RBBM | ATB ASYNC overflow\n"); + + if (status & A5XX_RBBM_INT_0_MASK_RBBM_ATB_BUS_OVERFLOW) + dev_err_ratelimited(gpu->dev->dev, "RBBM | ATB bus overflow\n"); +} + +static void a5xx_uche_err_irq(struct msm_gpu *gpu) +{ + uint64_t addr = gpu_read64(gpu, REG_A5XX_UCHE_TRAP_LOG); + + dev_err_ratelimited(gpu->dev->dev, "UCHE | Out of bounds access | addr=0x%llX\n", + addr); +} + +static void a5xx_gpmu_err_irq(struct msm_gpu *gpu) +{ + dev_err_ratelimited(gpu->dev->dev, "GPMU | voltage droop\n"); +} + +static void a5xx_fault_detect_irq(struct msm_gpu *gpu) +{ + struct drm_device *dev = gpu->dev; + struct msm_ringbuffer *ring = gpu->funcs->active_ring(gpu); + + /* + * If stalled on SMMU fault, we could trip the GPU's hang detection, + * but the fault handler will trigger the devcore dump, and we want + * to otherwise resume normally rather than killing the submit, so + * just bail. + */ + if (gpu_read(gpu, REG_A5XX_RBBM_STATUS3) & BIT(24)) + return; + + DRM_DEV_ERROR(dev->dev, "gpu fault ring %d fence %x status %8.8X rb %4.4x/%4.4x ib1 %16.16llX/%4.4x ib2 %16.16llX/%4.4x\n", + ring ? ring->id : -1, ring ? ring->fctx->last_fence : 0, + gpu_read(gpu, REG_A5XX_RBBM_STATUS), + gpu_read(gpu, REG_A5XX_CP_RB_RPTR), + gpu_read(gpu, REG_A5XX_CP_RB_WPTR), + gpu_read64(gpu, REG_A5XX_CP_IB1_BASE), + gpu_read(gpu, REG_A5XX_CP_IB1_BUFSZ), + gpu_read64(gpu, REG_A5XX_CP_IB2_BASE), + gpu_read(gpu, REG_A5XX_CP_IB2_BUFSZ)); + + /* Turn off the hangcheck timer to keep it from bothering us */ + timer_delete(&gpu->hangcheck_timer); + + kthread_queue_work(gpu->worker, &gpu->recover_work); +} + +#define RBBM_ERROR_MASK \ + (A5XX_RBBM_INT_0_MASK_RBBM_AHB_ERROR | \ + A5XX_RBBM_INT_0_MASK_RBBM_TRANSFER_TIMEOUT | \ + A5XX_RBBM_INT_0_MASK_RBBM_ME_MS_TIMEOUT | \ + A5XX_RBBM_INT_0_MASK_RBBM_PFP_MS_TIMEOUT | \ + A5XX_RBBM_INT_0_MASK_RBBM_ETS_MS_TIMEOUT | \ + A5XX_RBBM_INT_0_MASK_RBBM_ATB_ASYNC_OVERFLOW) + +static irqreturn_t a5xx_irq(struct msm_gpu *gpu) +{ + struct msm_drm_private *priv = gpu->dev->dev_private; + u32 status = gpu_read(gpu, REG_A5XX_RBBM_INT_0_STATUS); + + /* + * Clear all the interrupts except RBBM_AHB_ERROR - if we clear it + * before the source is cleared the interrupt will storm. + */ + gpu_write(gpu, REG_A5XX_RBBM_INT_CLEAR_CMD, + status & ~A5XX_RBBM_INT_0_MASK_RBBM_AHB_ERROR); + + if (priv->disable_err_irq) { + /* Turn off interrupts to avoid interrupt storm */ + gpu_write(gpu, REG_A5XX_RBBM_INT_0_MASK, + A5XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS | + A5XX_RBBM_INT_0_MASK_CP_SW); + + status &= A5XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS | + A5XX_RBBM_INT_0_MASK_CP_SW; + } + + /* Pass status to a5xx_rbbm_err_irq because we've already cleared it */ + if (status & RBBM_ERROR_MASK) + a5xx_rbbm_err_irq(gpu, status); + + if (status & A5XX_RBBM_INT_0_MASK_CP_HW_ERROR) + a5xx_cp_err_irq(gpu); + + if (status & A5XX_RBBM_INT_0_MASK_MISC_HANG_DETECT) + a5xx_fault_detect_irq(gpu); + + if (status & A5XX_RBBM_INT_0_MASK_UCHE_OOB_ACCESS) + a5xx_uche_err_irq(gpu); + + if (status & A5XX_RBBM_INT_0_MASK_GPMU_VOLTAGE_DROOP) + a5xx_gpmu_err_irq(gpu); + + if (status & A5XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS) { + a5xx_preempt_trigger(gpu); + msm_gpu_retire(gpu); + } + + if (status & A5XX_RBBM_INT_0_MASK_CP_SW) + a5xx_preempt_irq(gpu); + + return IRQ_HANDLED; +} + +static const u32 a5xx_registers[] = { + 0x0000, 0x0002, 0x0004, 0x0020, 0x0022, 0x0026, 0x0029, 0x002B, + 0x002E, 0x0035, 0x0038, 0x0042, 0x0044, 0x0044, 0x0047, 0x0095, + 0x0097, 0x00BB, 0x03A0, 0x0464, 0x0469, 0x046F, 0x04D2, 0x04D3, + 0x04E0, 0x0533, 0x0540, 0x0555, 0x0800, 0x081A, 0x081F, 0x0841, + 0x0860, 0x0860, 0x0880, 0x08A0, 0x0B00, 0x0B12, 0x0B15, 0x0B28, + 0x0B78, 0x0B7F, 0x0BB0, 0x0BBD, 0x0BC0, 0x0BC6, 0x0BD0, 0x0C53, + 0x0C60, 0x0C61, 0x0C80, 0x0C82, 0x0C84, 0x0C85, 0x0C90, 0x0C98, + 0x0CA0, 0x0CA0, 0x0CB0, 0x0CB2, 0x2180, 0x2185, 0x2580, 0x2585, + 0x0CC1, 0x0CC1, 0x0CC4, 0x0CC7, 0x0CCC, 0x0CCC, 0x0CD0, 0x0CD8, + 0x0CE0, 0x0CE5, 0x0CE8, 0x0CE8, 0x0CEC, 0x0CF1, 0x0CFB, 0x0D0E, + 0x2100, 0x211E, 0x2140, 0x2145, 0x2500, 0x251E, 0x2540, 0x2545, + 0x0D10, 0x0D17, 0x0D20, 0x0D23, 0x0D30, 0x0D30, 0x20C0, 0x20C0, + 0x24C0, 0x24C0, 0x0E40, 0x0E43, 0x0E4A, 0x0E4A, 0x0E50, 0x0E57, + 0x0E60, 0x0E7C, 0x0E80, 0x0E8E, 0x0E90, 0x0E96, 0x0EA0, 0x0EA8, + 0x0EB0, 0x0EB2, 0xE140, 0xE147, 0xE150, 0xE187, 0xE1A0, 0xE1A9, + 0xE1B0, 0xE1B6, 0xE1C0, 0xE1C7, 0xE1D0, 0xE1D1, 0xE200, 0xE201, + 0xE210, 0xE21C, 0xE240, 0xE268, 0xE000, 0xE006, 0xE010, 0xE09A, + 0xE0A0, 0xE0A4, 0xE0AA, 0xE0EB, 0xE100, 0xE105, 0xE380, 0xE38F, + 0xE3B0, 0xE3B0, 0xE400, 0xE405, 0xE408, 0xE4E9, 0xE4F0, 0xE4F0, + 0xE280, 0xE280, 0xE282, 0xE2A3, 0xE2A5, 0xE2C2, 0xE940, 0xE947, + 0xE950, 0xE987, 0xE9A0, 0xE9A9, 0xE9B0, 0xE9B6, 0xE9C0, 0xE9C7, + 0xE9D0, 0xE9D1, 0xEA00, 0xEA01, 0xEA10, 0xEA1C, 0xEA40, 0xEA68, + 0xE800, 0xE806, 0xE810, 0xE89A, 0xE8A0, 0xE8A4, 0xE8AA, 0xE8EB, + 0xE900, 0xE905, 0xEB80, 0xEB8F, 0xEBB0, 0xEBB0, 0xEC00, 0xEC05, + 0xEC08, 0xECE9, 0xECF0, 0xECF0, 0xEA80, 0xEA80, 0xEA82, 0xEAA3, + 0xEAA5, 0xEAC2, 0xA800, 0xA800, 0xA820, 0xA828, 0xA840, 0xA87D, + 0XA880, 0xA88D, 0xA890, 0xA8A3, 0xA8D0, 0xA8D8, 0xA8E0, 0xA8F5, + 0xAC60, 0xAC60, ~0, +}; + +static void a5xx_dump(struct msm_gpu *gpu) +{ + DRM_DEV_INFO(gpu->dev->dev, "status: %08x\n", + gpu_read(gpu, REG_A5XX_RBBM_STATUS)); + adreno_dump(gpu); +} + +static int a5xx_pm_resume(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + int ret; + + /* Turn on the core power */ + ret = msm_gpu_pm_resume(gpu); + if (ret) + return ret; + + /* Adreno 505, 506, 508, 509, 510, 512 needs manual RBBM sus/res control */ + if (!(adreno_is_a530(adreno_gpu) || adreno_is_a540(adreno_gpu))) { + /* Halt the sp_input_clk at HM level */ + gpu_write(gpu, REG_A5XX_RBBM_CLOCK_CNTL, 0x00000055); + a5xx_set_hwcg(gpu, true); + /* Turn on sp_input_clk at HM level */ + gpu_rmw(gpu, REG_A5XX_RBBM_CLOCK_CNTL, 0xff, 0); + return 0; + } + + /* Turn the RBCCU domain first to limit the chances of voltage droop */ + gpu_write(gpu, REG_A5XX_GPMU_RBCCU_POWER_CNTL, 0x778000); + + /* Wait 3 usecs before polling */ + udelay(3); + + ret = spin_usecs(gpu, 20, REG_A5XX_GPMU_RBCCU_PWR_CLK_STATUS, + (1 << 20), (1 << 20)); + if (ret) { + DRM_ERROR("%s: timeout waiting for RBCCU GDSC enable: %X\n", + gpu->name, + gpu_read(gpu, REG_A5XX_GPMU_RBCCU_PWR_CLK_STATUS)); + return ret; + } + + /* Turn on the SP domain */ + gpu_write(gpu, REG_A5XX_GPMU_SP_POWER_CNTL, 0x778000); + ret = spin_usecs(gpu, 20, REG_A5XX_GPMU_SP_PWR_CLK_STATUS, + (1 << 20), (1 << 20)); + if (ret) + DRM_ERROR("%s: timeout waiting for SP GDSC enable\n", + gpu->name); + + return ret; +} + +static int a5xx_pm_suspend(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + u32 mask = 0xf; + int i, ret; + + /* A505, A506, A508, A510 have 3 XIN ports in VBIF */ + if (adreno_is_a505(adreno_gpu) || adreno_is_a506(adreno_gpu) || + adreno_is_a508(adreno_gpu) || adreno_is_a510(adreno_gpu)) + mask = 0x7; + + /* Clear the VBIF pipe before shutting down */ + gpu_write(gpu, REG_A5XX_VBIF_XIN_HALT_CTRL0, mask); + spin_until((gpu_read(gpu, REG_A5XX_VBIF_XIN_HALT_CTRL1) & + mask) == mask); + + gpu_write(gpu, REG_A5XX_VBIF_XIN_HALT_CTRL0, 0); + + /* + * Reset the VBIF before power collapse to avoid issue with FIFO + * entries on Adreno A510 and A530 (the others will tend to lock up) + */ + if (adreno_is_a510(adreno_gpu) || adreno_is_a530(adreno_gpu)) { + gpu_write(gpu, REG_A5XX_RBBM_BLOCK_SW_RESET_CMD, 0x003C0000); + gpu_write(gpu, REG_A5XX_RBBM_BLOCK_SW_RESET_CMD, 0x00000000); + } + + ret = msm_gpu_pm_suspend(gpu); + if (ret) + return ret; + + if (a5xx_gpu->has_whereami) + for (i = 0; i < gpu->nr_rings; i++) + a5xx_gpu->shadow[i] = 0; + + return 0; +} + +static u64 a5xx_get_timestamp(struct msm_gpu *gpu) +{ + return gpu_read64(gpu, REG_A5XX_RBBM_ALWAYSON_COUNTER_LO); +} + +struct a5xx_crashdumper { + void *ptr; + struct drm_gem_object *bo; + u64 iova; +}; + +struct a5xx_gpu_state { + struct msm_gpu_state base; + u32 *hlsqregs; +}; + +static int a5xx_crashdumper_init(struct msm_gpu *gpu, + struct a5xx_crashdumper *dumper) +{ + dumper->ptr = msm_gem_kernel_new(gpu->dev, + SZ_1M, MSM_BO_WC, gpu->vm, + &dumper->bo, &dumper->iova); + + if (!IS_ERR(dumper->ptr)) + msm_gem_object_set_name(dumper->bo, "crashdump"); + + return PTR_ERR_OR_ZERO(dumper->ptr); +} + +static int a5xx_crashdumper_run(struct msm_gpu *gpu, + struct a5xx_crashdumper *dumper) +{ + u32 val; + + if (IS_ERR_OR_NULL(dumper->ptr)) + return -EINVAL; + + gpu_write64(gpu, REG_A5XX_CP_CRASH_SCRIPT_BASE_LO, dumper->iova); + + gpu_write(gpu, REG_A5XX_CP_CRASH_DUMP_CNTL, 1); + + return gpu_poll_timeout(gpu, REG_A5XX_CP_CRASH_DUMP_CNTL, val, + val & 0x04, 100, 10000); +} + +/* + * These are a list of the registers that need to be read through the HLSQ + * aperture through the crashdumper. These are not nominally accessible from + * the CPU on a secure platform. + */ +static const struct { + u32 type; + u32 regoffset; + u32 count; +} a5xx_hlsq_aperture_regs[] = { + { 0x35, 0xe00, 0x32 }, /* HSLQ non-context */ + { 0x31, 0x2080, 0x1 }, /* HLSQ 2D context 0 */ + { 0x33, 0x2480, 0x1 }, /* HLSQ 2D context 1 */ + { 0x32, 0xe780, 0x62 }, /* HLSQ 3D context 0 */ + { 0x34, 0xef80, 0x62 }, /* HLSQ 3D context 1 */ + { 0x3f, 0x0ec0, 0x40 }, /* SP non-context */ + { 0x3d, 0x2040, 0x1 }, /* SP 2D context 0 */ + { 0x3b, 0x2440, 0x1 }, /* SP 2D context 1 */ + { 0x3e, 0xe580, 0x170 }, /* SP 3D context 0 */ + { 0x3c, 0xed80, 0x170 }, /* SP 3D context 1 */ + { 0x3a, 0x0f00, 0x1c }, /* TP non-context */ + { 0x38, 0x2000, 0xa }, /* TP 2D context 0 */ + { 0x36, 0x2400, 0xa }, /* TP 2D context 1 */ + { 0x39, 0xe700, 0x80 }, /* TP 3D context 0 */ + { 0x37, 0xef00, 0x80 }, /* TP 3D context 1 */ +}; + +static void a5xx_gpu_state_get_hlsq_regs(struct msm_gpu *gpu, + struct a5xx_gpu_state *a5xx_state) +{ + struct a5xx_crashdumper dumper = { 0 }; + u32 offset, count = 0; + u64 *ptr; + int i; + + if (a5xx_crashdumper_init(gpu, &dumper)) + return; + + /* The script will be written at offset 0 */ + ptr = dumper.ptr; + + /* Start writing the data at offset 256k */ + offset = dumper.iova + (256 * SZ_1K); + + /* Count how many additional registers to get from the HLSQ aperture */ + for (i = 0; i < ARRAY_SIZE(a5xx_hlsq_aperture_regs); i++) + count += a5xx_hlsq_aperture_regs[i].count; + + a5xx_state->hlsqregs = kcalloc(count, sizeof(u32), GFP_KERNEL); + if (!a5xx_state->hlsqregs) + return; + + /* Build the crashdump script */ + for (i = 0; i < ARRAY_SIZE(a5xx_hlsq_aperture_regs); i++) { + u32 type = a5xx_hlsq_aperture_regs[i].type; + u32 c = a5xx_hlsq_aperture_regs[i].count; + + /* Write the register to select the desired bank */ + *ptr++ = ((u64) type << 8); + *ptr++ = (((u64) REG_A5XX_HLSQ_DBG_READ_SEL) << 44) | + (1 << 21) | 1; + + *ptr++ = offset; + *ptr++ = (((u64) REG_A5XX_HLSQ_DBG_AHB_READ_APERTURE) << 44) + | c; + + offset += c * sizeof(u32); + } + + /* Write two zeros to close off the script */ + *ptr++ = 0; + *ptr++ = 0; + + if (a5xx_crashdumper_run(gpu, &dumper)) { + kfree(a5xx_state->hlsqregs); + msm_gem_kernel_put(dumper.bo, gpu->vm); + return; + } + + /* Copy the data from the crashdumper to the state */ + memcpy(a5xx_state->hlsqregs, dumper.ptr + (256 * SZ_1K), + count * sizeof(u32)); + + msm_gem_kernel_put(dumper.bo, gpu->vm); +} + +static struct msm_gpu_state *a5xx_gpu_state_get(struct msm_gpu *gpu) +{ + struct a5xx_gpu_state *a5xx_state = kzalloc_obj(*a5xx_state); + bool stalled = !!(gpu_read(gpu, REG_A5XX_RBBM_STATUS3) & BIT(24)); + + if (!a5xx_state) + return ERR_PTR(-ENOMEM); + + /* Temporarily disable hardware clock gating before reading the hw */ + a5xx_set_hwcg(gpu, false); + + /* First get the generic state from the adreno core */ + adreno_gpu_state_get(gpu, &(a5xx_state->base)); + + a5xx_state->base.rbbm_status = gpu_read(gpu, REG_A5XX_RBBM_STATUS); + + /* + * Get the HLSQ regs with the help of the crashdumper, but only if + * we are not stalled in an iommu fault (in which case the crashdumper + * would not have access to memory) + */ + if (!stalled) + a5xx_gpu_state_get_hlsq_regs(gpu, a5xx_state); + + a5xx_set_hwcg(gpu, true); + + return &a5xx_state->base; +} + +static void a5xx_gpu_state_destroy(struct kref *kref) +{ + struct msm_gpu_state *state = container_of(kref, + struct msm_gpu_state, ref); + struct a5xx_gpu_state *a5xx_state = container_of(state, + struct a5xx_gpu_state, base); + + kfree(a5xx_state->hlsqregs); + + adreno_gpu_state_destroy(state); + kfree(a5xx_state); +} + +static int a5xx_gpu_state_put(struct msm_gpu_state *state) +{ + if (IS_ERR_OR_NULL(state)) + return 1; + + return kref_put(&state->ref, a5xx_gpu_state_destroy); +} + + +#if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP) +static void a5xx_show(struct msm_gpu *gpu, struct msm_gpu_state *state, + struct drm_printer *p) +{ + int i, j; + u32 pos = 0; + struct a5xx_gpu_state *a5xx_state = container_of(state, + struct a5xx_gpu_state, base); + + if (IS_ERR_OR_NULL(state)) + return; + + adreno_show(gpu, state, p); + + /* Dump the additional a5xx HLSQ registers */ + if (!a5xx_state->hlsqregs) + return; + + drm_printf(p, "registers-hlsq:\n"); + + for (i = 0; i < ARRAY_SIZE(a5xx_hlsq_aperture_regs); i++) { + u32 o = a5xx_hlsq_aperture_regs[i].regoffset; + u32 c = a5xx_hlsq_aperture_regs[i].count; + + for (j = 0; j < c; j++, pos++, o++) { + /* + * To keep the crashdump simple we pull the entire range + * for each register type but not all of the registers + * in the range are valid. Fortunately invalid registers + * stick out like a sore thumb with a value of + * 0xdeadbeef + */ + if (a5xx_state->hlsqregs[pos] == 0xdeadbeef) + continue; + + drm_printf(p, " - { offset: 0x%04x, value: 0x%08x }\n", + o << 2, a5xx_state->hlsqregs[pos]); + } + } +} +#endif + +static struct msm_ringbuffer *a5xx_active_ring(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + + return a5xx_gpu->cur_ring; +} + +static u64 a5xx_gpu_busy(struct msm_gpu *gpu, unsigned long *out_sample_rate) +{ + u64 busy_cycles; + + busy_cycles = gpu_read64(gpu, REG_A5XX_RBBM_PERFCTR_RBBM_0_LO); + *out_sample_rate = clk_get_rate(gpu->core_clk); + + return busy_cycles; +} + +static uint32_t a5xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + + if (a5xx_gpu->has_whereami) + return a5xx_gpu->shadow[ring->id]; + + return ring->memptrs->rptr = gpu_read(gpu, REG_A5XX_CP_RB_RPTR); +} + +static void check_speed_bin(struct device *dev) +{ + struct nvmem_cell *cell; + u32 val; + + /* + * If the OPP table specifies a opp-supported-hw property then we have + * to set something with dev_pm_opp_set_supported_hw() or the table + * doesn't get populated so pick an arbitrary value that should + * ensure the default frequencies are selected but not conflict with any + * actual bins + */ + val = 0x80; + + cell = nvmem_cell_get(dev, "speed_bin"); + + if (!IS_ERR(cell)) { + void *buf = nvmem_cell_read(cell, NULL); + + if (!IS_ERR(buf)) { + u8 bin = *((u8 *) buf); + + val = (1 << bin); + kfree(buf); + } + + nvmem_cell_put(cell); + } + + devm_pm_opp_set_supported_hw(dev, &val, 1); +} + +static struct msm_gpu *a5xx_gpu_init(struct drm_device *dev) +{ + struct msm_drm_private *priv = dev->dev_private; + struct platform_device *pdev = priv->gpu_pdev; + struct adreno_platform_config *config = pdev->dev.platform_data; + struct a5xx_gpu *a5xx_gpu = NULL; + struct adreno_gpu *adreno_gpu; + struct msm_gpu *gpu; + unsigned int nr_rings; + u32 speedbin; + int ret; + + a5xx_gpu = kzalloc_obj(*a5xx_gpu); + if (!a5xx_gpu) + return ERR_PTR(-ENOMEM); + + adreno_gpu = &a5xx_gpu->base; + gpu = &adreno_gpu->base; + + adreno_gpu->registers = a5xx_registers; + + a5xx_gpu->lm_leakage = 0x4E001A; + + check_speed_bin(&pdev->dev); + + nr_rings = 4; + + if (config->info->revn == 510) + nr_rings = 1; + + ret = adreno_gpu_init(dev, pdev, adreno_gpu, config->info->funcs, nr_rings); + if (ret) { + a5xx_destroy(&(a5xx_gpu->base.base)); + return ERR_PTR(ret); + } + + /* Set the speedbin value that is passed to userspace */ + if (adreno_read_speedbin(&pdev->dev, &speedbin) || !speedbin) + speedbin = 0xffff; + adreno_gpu->speedbin = (uint16_t) (0xffff & speedbin); + + msm_mmu_set_fault_handler(to_msm_vm(gpu->vm)->mmu, gpu, + a5xx_fault_handler); + + /* Set up the preemption specific bits and pieces for each ringbuffer */ + a5xx_preempt_init(gpu); + + /* Inherit the common config and make some necessary fixups */ + adreno_gpu->ubwc_config = qcom_ubwc_config_get_data(); + if (IS_ERR(adreno_gpu->ubwc_config)) + return ERR_CAST(adreno_gpu->ubwc_config); + + adreno_gpu->uche_trap_base = 0x0001ffffffff0000ull; + + return gpu; +} + +const struct adreno_gpu_funcs a5xx_gpu_funcs = { + .base = { + .get_param = adreno_get_param, + .set_param = adreno_set_param, + .hw_init = a5xx_hw_init, + .ucode_load = a5xx_ucode_load, + .pm_suspend = a5xx_pm_suspend, + .pm_resume = a5xx_pm_resume, + .recover = a5xx_recover, + .submit = a5xx_submit, + .active_ring = a5xx_active_ring, + .irq = a5xx_irq, + .destroy = a5xx_destroy, +#if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP) + .show = a5xx_show, +#endif +#if defined(CONFIG_DEBUG_FS) + .debugfs_init = a5xx_debugfs_init, +#endif + .gpu_busy = a5xx_gpu_busy, + .gpu_state_get = a5xx_gpu_state_get, + .gpu_state_put = a5xx_gpu_state_put, + .create_vm = adreno_create_vm, + .get_rptr = a5xx_get_rptr, + }, + .init = a5xx_gpu_init, + .get_timestamp = a5xx_get_timestamp, +}; diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.h b/drivers/gpu/drm/msm/adreno/a5xx_gpu.h new file mode 100644 index 000000000..407bb950d --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.h @@ -0,0 +1,177 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2016-2017 The Linux Foundation. All rights reserved. + */ +#ifndef __A5XX_GPU_H__ +#define __A5XX_GPU_H__ + +#include "adreno_gpu.h" + +/* Bringing over the hack from the previous targets */ +#undef ROP_COPY +#undef ROP_XOR + +#include "a5xx.xml.h" + +struct a5xx_gpu { + struct adreno_gpu base; + + struct drm_gem_object *pm4_bo; + uint64_t pm4_iova; + + struct drm_gem_object *pfp_bo; + uint64_t pfp_iova; + + struct drm_gem_object *gpmu_bo; + uint64_t gpmu_iova; + uint32_t gpmu_dwords; + + uint32_t lm_leakage; + + struct msm_ringbuffer *cur_ring; + struct msm_ringbuffer *next_ring; + + struct drm_gem_object *preempt_bo[MSM_GPU_MAX_RINGS]; + struct drm_gem_object *preempt_counters_bo[MSM_GPU_MAX_RINGS]; + struct a5xx_preempt_record *preempt[MSM_GPU_MAX_RINGS]; + uint64_t preempt_iova[MSM_GPU_MAX_RINGS]; + uint32_t last_seqno[MSM_GPU_MAX_RINGS]; + + atomic_t preempt_state; + spinlock_t preempt_start_lock; + struct timer_list preempt_timer; + + struct drm_gem_object *shadow_bo; + uint64_t shadow_iova; + uint32_t *shadow; + + /* True if the microcode supports the WHERE_AM_I opcode */ + bool has_whereami; +}; + +#define to_a5xx_gpu(x) container_of(x, struct a5xx_gpu, base) + +#ifdef CONFIG_DEBUG_FS +void a5xx_debugfs_init(struct msm_gpu *gpu, struct drm_minor *minor); +#endif + +/* + * In order to do lockless preemption we use a simple state machine to progress + * through the process. + * + * PREEMPT_NONE - no preemption in progress. Next state START. + * PREEMPT_START - The trigger is evaulating if preemption is possible. Next + * states: TRIGGERED, NONE + * PREEMPT_ABORT - An intermediate state before moving back to NONE. Next + * state: NONE. + * PREEMPT_TRIGGERED: A preemption has been executed on the hardware. Next + * states: FAULTED, PENDING + * PREEMPT_FAULTED: A preemption timed out (never completed). This will trigger + * recovery. Next state: N/A + * PREEMPT_PENDING: Preemption complete interrupt fired - the callback is + * checking the success of the operation. Next state: FAULTED, NONE. + */ + +enum preempt_state { + PREEMPT_NONE = 0, + PREEMPT_START, + PREEMPT_ABORT, + PREEMPT_TRIGGERED, + PREEMPT_FAULTED, + PREEMPT_PENDING, +}; + +/* + * struct a5xx_preempt_record is a shared buffer between the microcode and the + * CPU to store the state for preemption. The record itself is much larger + * (64k) but most of that is used by the CP for storage. + * + * There is a preemption record assigned per ringbuffer. When the CPU triggers a + * preemption, it fills out the record with the useful information (wptr, ring + * base, etc) and the microcode uses that information to set up the CP following + * the preemption. When a ring is switched out, the CP will save the ringbuffer + * state back to the record. In this way, once the records are properly set up + * the CPU can quickly switch back and forth between ringbuffers by only + * updating a few registers (often only the wptr). + * + * These are the CPU aware registers in the record: + * @magic: Must always be 0x27C4BAFC + * @info: Type of the record - written 0 by the CPU, updated by the CP + * @data: Data field from SET_RENDER_MODE or a checkpoint. Written and used by + * the CP + * @cntl: Value of RB_CNTL written by CPU, save/restored by CP + * @rptr: Value of RB_RPTR written by CPU, save/restored by CP + * @wptr: Value of RB_WPTR written by CPU, save/restored by CP + * @rptr_addr: Value of RB_RPTR_ADDR written by CPU, save/restored by CP + * @rbase: Value of RB_BASE written by CPU, save/restored by CP + * @counter: GPU address of the storage area for the performance counters + */ +struct a5xx_preempt_record { + uint32_t magic; + uint32_t info; + uint32_t data; + uint32_t cntl; + uint32_t rptr; + uint32_t wptr; + uint64_t rptr_addr; + uint64_t rbase; + uint64_t counter; +}; + +/* Magic identifier for the preemption record */ +#define A5XX_PREEMPT_RECORD_MAGIC 0x27C4BAFCUL + +/* + * Even though the structure above is only a few bytes, we need a full 64k to + * store the entire preemption record from the CP + */ +#define A5XX_PREEMPT_RECORD_SIZE (64 * 1024) + +/* + * The preemption counter block is a storage area for the value of the + * preemption counters that are saved immediately before context switch. We + * append it on to the end of the allocation for the preemption record. + */ +#define A5XX_PREEMPT_COUNTER_SIZE (16 * 4) + +extern const struct adreno_gpu_funcs a5xx_gpu_funcs; + +int a5xx_power_init(struct msm_gpu *gpu); +void a5xx_gpmu_ucode_init(struct msm_gpu *gpu); + +static inline int spin_usecs(struct msm_gpu *gpu, uint32_t usecs, + uint32_t reg, uint32_t mask, uint32_t value) +{ + while (usecs--) { + udelay(1); + if ((gpu_read(gpu, reg) & mask) == value) + return 0; + cpu_relax(); + } + + return -ETIMEDOUT; +} + +#define shadowptr(a5xx_gpu, ring) ((a5xx_gpu)->shadow_iova + \ + ((ring)->id * sizeof(uint32_t))) + +bool a5xx_idle(struct msm_gpu *gpu, struct msm_ringbuffer *ring); +void a5xx_set_hwcg(struct msm_gpu *gpu, bool state); + +void a5xx_preempt_init(struct msm_gpu *gpu); +void a5xx_preempt_hw_init(struct msm_gpu *gpu); +void a5xx_preempt_trigger(struct msm_gpu *gpu); +void a5xx_preempt_irq(struct msm_gpu *gpu); +void a5xx_preempt_fini(struct msm_gpu *gpu); + +void a5xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring, bool sync); + +/* Return true if we are in a preempt state */ +static inline bool a5xx_in_preempt(struct a5xx_gpu *a5xx_gpu) +{ + int preempt_state = atomic_read(&a5xx_gpu->preempt_state); + + return !(preempt_state == PREEMPT_NONE || + preempt_state == PREEMPT_ABORT); +} + +#endif /* __A5XX_GPU_H__ */ diff --git a/drivers/gpu/drm/msm/adreno/a5xx_power.c b/drivers/gpu/drm/msm/adreno/a5xx_power.c new file mode 100644 index 000000000..d6da7351c --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a5xx_power.c @@ -0,0 +1,390 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2016 The Linux Foundation. All rights reserved. + */ + +#include +#include "a5xx_gpu.h" + +/* + * The GPMU data block is a block of shared registers that can be used to + * communicate back and forth. These "registers" are by convention with the GPMU + * firwmare and not bound to any specific hardware design + */ + +#define AGC_INIT_BASE REG_A5XX_GPMU_DATA_RAM_BASE +#define AGC_INIT_MSG_MAGIC (AGC_INIT_BASE + 5) +#define AGC_MSG_BASE (AGC_INIT_BASE + 7) + +#define AGC_MSG_STATE (AGC_MSG_BASE + 0) +#define AGC_MSG_COMMAND (AGC_MSG_BASE + 1) +#define AGC_MSG_PAYLOAD_SIZE (AGC_MSG_BASE + 3) +#define AGC_MSG_PAYLOAD(_o) ((AGC_MSG_BASE + 5) + (_o)) + +#define AGC_POWER_CONFIG_PRODUCTION_ID 1 +#define AGC_INIT_MSG_VALUE 0xBABEFACE + +/* AGC_LM_CONFIG (A540+) */ +#define AGC_LM_CONFIG (136/4) +#define AGC_LM_CONFIG_GPU_VERSION_SHIFT 17 +#define AGC_LM_CONFIG_ENABLE_GPMU_ADAPTIVE 1 +#define AGC_LM_CONFIG_THROTTLE_DISABLE (2 << 8) +#define AGC_LM_CONFIG_ISENSE_ENABLE (1 << 4) +#define AGC_LM_CONFIG_ENABLE_ERROR (3 << 4) +#define AGC_LM_CONFIG_LLM_ENABLED (1 << 16) +#define AGC_LM_CONFIG_BCL_DISABLED (1 << 24) + +#define AGC_LEVEL_CONFIG (140/4) + +static struct { + uint32_t reg; + uint32_t value; +} a5xx_sequence_regs[] = { + { 0xB9A1, 0x00010303 }, + { 0xB9A2, 0x13000000 }, + { 0xB9A3, 0x00460020 }, + { 0xB9A4, 0x10000000 }, + { 0xB9A5, 0x040A1707 }, + { 0xB9A6, 0x00010000 }, + { 0xB9A7, 0x0E000904 }, + { 0xB9A8, 0x10000000 }, + { 0xB9A9, 0x01165000 }, + { 0xB9AA, 0x000E0002 }, + { 0xB9AB, 0x03884141 }, + { 0xB9AC, 0x10000840 }, + { 0xB9AD, 0x572A5000 }, + { 0xB9AE, 0x00000003 }, + { 0xB9AF, 0x00000000 }, + { 0xB9B0, 0x10000000 }, + { 0xB828, 0x6C204010 }, + { 0xB829, 0x6C204011 }, + { 0xB82A, 0x6C204012 }, + { 0xB82B, 0x6C204013 }, + { 0xB82C, 0x6C204014 }, + { 0xB90F, 0x00000004 }, + { 0xB910, 0x00000002 }, + { 0xB911, 0x00000002 }, + { 0xB912, 0x00000002 }, + { 0xB913, 0x00000002 }, + { 0xB92F, 0x00000004 }, + { 0xB930, 0x00000005 }, + { 0xB931, 0x00000005 }, + { 0xB932, 0x00000005 }, + { 0xB933, 0x00000005 }, + { 0xB96F, 0x00000001 }, + { 0xB970, 0x00000003 }, + { 0xB94F, 0x00000004 }, + { 0xB950, 0x0000000B }, + { 0xB951, 0x0000000B }, + { 0xB952, 0x0000000B }, + { 0xB953, 0x0000000B }, + { 0xB907, 0x00000019 }, + { 0xB927, 0x00000019 }, + { 0xB947, 0x00000019 }, + { 0xB967, 0x00000019 }, + { 0xB987, 0x00000019 }, + { 0xB906, 0x00220001 }, + { 0xB926, 0x00220001 }, + { 0xB946, 0x00220001 }, + { 0xB966, 0x00220001 }, + { 0xB986, 0x00300000 }, + { 0xAC40, 0x0340FF41 }, + { 0xAC41, 0x03BEFED0 }, + { 0xAC42, 0x00331FED }, + { 0xAC43, 0x021FFDD3 }, + { 0xAC44, 0x5555AAAA }, + { 0xAC45, 0x5555AAAA }, + { 0xB9BA, 0x00000008 }, +}; + +/* + * Get the actual voltage value for the operating point at the specified + * frequency + */ +static inline uint32_t _get_mvolts(struct msm_gpu *gpu, uint32_t freq) +{ + struct drm_device *dev = gpu->dev; + struct msm_drm_private *priv = dev->dev_private; + struct platform_device *pdev = priv->gpu_pdev; + struct dev_pm_opp *opp; + u32 ret = 0; + + opp = dev_pm_opp_find_freq_exact(&pdev->dev, freq, true); + + if (!IS_ERR(opp)) { + ret = dev_pm_opp_get_voltage(opp) / 1000; + dev_pm_opp_put(opp); + } + + return ret; +} + +/* Setup thermal limit management */ +static void a530_lm_setup(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + unsigned int i; + + /* Write the block of sequence registers */ + for (i = 0; i < ARRAY_SIZE(a5xx_sequence_regs); i++) + gpu_write(gpu, a5xx_sequence_regs[i].reg, + a5xx_sequence_regs[i].value); + + /* Hard code the A530 GPU thermal sensor ID for the GPMU */ + gpu_write(gpu, REG_A5XX_GPMU_TEMP_SENSOR_ID, 0x60007); + gpu_write(gpu, REG_A5XX_GPMU_DELTA_TEMP_THRESHOLD, 0x01); + gpu_write(gpu, REG_A5XX_GPMU_TEMP_SENSOR_CONFIG, 0x01); + + /* Until we get clock scaling 0 is always the active power level */ + gpu_write(gpu, REG_A5XX_GPMU_GPMU_VOLTAGE, 0x80000000 | 0); + + gpu_write(gpu, REG_A5XX_GPMU_BASE_LEAKAGE, a5xx_gpu->lm_leakage); + + /* The threshold is fixed at 6000 for A530 */ + gpu_write(gpu, REG_A5XX_GPMU_GPMU_PWR_THRESHOLD, 0x80000000 | 6000); + + gpu_write(gpu, REG_A5XX_GPMU_BEC_ENABLE, 0x10001FFF); + gpu_write(gpu, REG_A5XX_GDPM_CONFIG1, 0x00201FF1); + + /* Write the voltage table */ + gpu_write(gpu, REG_A5XX_GPMU_BEC_ENABLE, 0x10001FFF); + gpu_write(gpu, REG_A5XX_GDPM_CONFIG1, 0x201FF1); + + gpu_write(gpu, AGC_MSG_STATE, 1); + gpu_write(gpu, AGC_MSG_COMMAND, AGC_POWER_CONFIG_PRODUCTION_ID); + + /* Write the max power - hard coded to 5448 for A530 */ + gpu_write(gpu, AGC_MSG_PAYLOAD(0), 5448); + gpu_write(gpu, AGC_MSG_PAYLOAD(1), 1); + + /* + * For now just write the one voltage level - we will do more when we + * can do scaling + */ + gpu_write(gpu, AGC_MSG_PAYLOAD(2), _get_mvolts(gpu, gpu->fast_rate)); + gpu_write(gpu, AGC_MSG_PAYLOAD(3), gpu->fast_rate / 1000000); + + gpu_write(gpu, AGC_MSG_PAYLOAD_SIZE, 4 * sizeof(uint32_t)); + gpu_write(gpu, AGC_INIT_MSG_MAGIC, AGC_INIT_MSG_VALUE); +} + +#define PAYLOAD_SIZE(_size) ((_size) * sizeof(u32)) +#define LM_DCVS_LIMIT 1 +#define LEVEL_CONFIG ~(0x303) + +static void a540_lm_setup(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + u32 config; + + /* The battery current limiter isn't enabled for A540 */ + config = AGC_LM_CONFIG_BCL_DISABLED; + config |= adreno_patchid(adreno_gpu) << AGC_LM_CONFIG_GPU_VERSION_SHIFT; + + /* For now disable GPMU side throttling */ + config |= AGC_LM_CONFIG_THROTTLE_DISABLE; + + /* Until we get clock scaling 0 is always the active power level */ + gpu_write(gpu, REG_A5XX_GPMU_GPMU_VOLTAGE, 0x80000000 | 0); + + /* Fixed at 6000 for now */ + gpu_write(gpu, REG_A5XX_GPMU_GPMU_PWR_THRESHOLD, 0x80000000 | 6000); + + gpu_write(gpu, AGC_MSG_STATE, 0x80000001); + gpu_write(gpu, AGC_MSG_COMMAND, AGC_POWER_CONFIG_PRODUCTION_ID); + + gpu_write(gpu, AGC_MSG_PAYLOAD(0), 5448); + gpu_write(gpu, AGC_MSG_PAYLOAD(1), 1); + + gpu_write(gpu, AGC_MSG_PAYLOAD(2), _get_mvolts(gpu, gpu->fast_rate)); + gpu_write(gpu, AGC_MSG_PAYLOAD(3), gpu->fast_rate / 1000000); + + gpu_write(gpu, AGC_MSG_PAYLOAD(AGC_LM_CONFIG), config); + gpu_write(gpu, AGC_MSG_PAYLOAD(AGC_LEVEL_CONFIG), LEVEL_CONFIG); + gpu_write(gpu, AGC_MSG_PAYLOAD_SIZE, + PAYLOAD_SIZE(AGC_LEVEL_CONFIG + 1)); + + gpu_write(gpu, AGC_INIT_MSG_MAGIC, AGC_INIT_MSG_VALUE); +} + +/* Enable SP/TP cpower collapse */ +static void a5xx_pc_init(struct msm_gpu *gpu) +{ + gpu_write(gpu, REG_A5XX_GPMU_PWR_COL_INTER_FRAME_CTRL, 0x7F); + gpu_write(gpu, REG_A5XX_GPMU_PWR_COL_BINNING_CTRL, 0); + gpu_write(gpu, REG_A5XX_GPMU_PWR_COL_INTER_FRAME_HYST, 0xA0080); + gpu_write(gpu, REG_A5XX_GPMU_PWR_COL_STAGGER_DELAY, 0x600040); +} + +/* Enable the GPMU microcontroller */ +static int a5xx_gpmu_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + struct msm_ringbuffer *ring = gpu->rb[0]; + + if (!a5xx_gpu->gpmu_dwords) + return 0; + + /* Turn off protected mode for this operation */ + OUT_PKT7(ring, CP_SET_PROTECTED_MODE, 1); + OUT_RING(ring, 0); + + /* Kick off the IB to load the GPMU microcode */ + OUT_PKT7(ring, CP_INDIRECT_BUFFER_PFE, 3); + OUT_RING(ring, lower_32_bits(a5xx_gpu->gpmu_iova)); + OUT_RING(ring, upper_32_bits(a5xx_gpu->gpmu_iova)); + OUT_RING(ring, a5xx_gpu->gpmu_dwords); + + /* Turn back on protected mode */ + OUT_PKT7(ring, CP_SET_PROTECTED_MODE, 1); + OUT_RING(ring, 1); + + a5xx_flush(gpu, ring, true); + + if (!a5xx_idle(gpu, ring)) { + DRM_ERROR("%s: Unable to load GPMU firmware. GPMU will not be active\n", + gpu->name); + return -EINVAL; + } + + if (adreno_is_a530(adreno_gpu)) + gpu_write(gpu, REG_A5XX_GPMU_WFI_CONFIG, 0x4014); + + /* Kick off the GPMU */ + gpu_write(gpu, REG_A5XX_GPMU_CM3_SYSRESET, 0x0); + + /* + * Wait for the GPMU to respond. It isn't fatal if it doesn't, we just + * won't have advanced power collapse. + */ + if (spin_usecs(gpu, 25, REG_A5XX_GPMU_GENERAL_0, 0xFFFFFFFF, + 0xBABEFACE)) + DRM_ERROR("%s: GPMU firmware initialization timed out\n", + gpu->name); + + if (!adreno_is_a530(adreno_gpu)) { + u32 val = gpu_read(gpu, REG_A5XX_GPMU_GENERAL_1); + + if (val) + DRM_ERROR("%s: GPMU firmware initialization failed: %d\n", + gpu->name, val); + } + + return 0; +} + +/* Enable limits management */ +static void a5xx_lm_enable(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + + /* This init sequence only applies to A530 */ + if (!adreno_is_a530(adreno_gpu)) + return; + + gpu_write(gpu, REG_A5XX_GDPM_INT_MASK, 0x0); + gpu_write(gpu, REG_A5XX_GDPM_INT_EN, 0x0A); + gpu_write(gpu, REG_A5XX_GPMU_GPMU_VOLTAGE_INTR_EN_MASK, 0x01); + gpu_write(gpu, REG_A5XX_GPMU_TEMP_THRESHOLD_INTR_EN_MASK, 0x50000); + gpu_write(gpu, REG_A5XX_GPMU_THROTTLE_UNMASK_FORCE_CTRL, 0x30000); + + gpu_write(gpu, REG_A5XX_GPMU_CLOCK_THROTTLE_CTRL, 0x011); +} + +int a5xx_power_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + int ret; + + /* Not all A5xx chips have a GPMU */ + if (!(adreno_is_a530(adreno_gpu) || adreno_is_a540(adreno_gpu))) + return 0; + + /* Set up the limits management */ + if (adreno_is_a530(adreno_gpu)) + a530_lm_setup(gpu); + else if (adreno_is_a540(adreno_gpu)) + a540_lm_setup(gpu); + + /* Set up SP/TP power collapse */ + a5xx_pc_init(gpu); + + /* Start the GPMU */ + ret = a5xx_gpmu_init(gpu); + if (ret) + return ret; + + /* Start the limits management */ + a5xx_lm_enable(gpu); + + return 0; +} + +void a5xx_gpmu_ucode_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + struct drm_device *drm = gpu->dev; + uint32_t dwords = 0, offset = 0, bosize; + unsigned int *data, *ptr, *cmds; + unsigned int cmds_size; + + if (!(adreno_is_a530(adreno_gpu) || adreno_is_a540(adreno_gpu))) + return; + + if (a5xx_gpu->gpmu_bo) + return; + + data = (unsigned int *) adreno_gpu->fw[ADRENO_FW_GPMU]->data; + + /* + * The first dword is the size of the remaining data in dwords. Use it + * as a checksum of sorts and make sure it matches the actual size of + * the firmware that we read + */ + + if (adreno_gpu->fw[ADRENO_FW_GPMU]->size < 8 || + (data[0] < 2) || (data[0] >= + (adreno_gpu->fw[ADRENO_FW_GPMU]->size >> 2))) + return; + + /* The second dword is an ID - look for 2 (GPMU_FIRMWARE_ID) */ + if (data[1] != 2) + return; + + cmds = data + data[2] + 3; + cmds_size = data[0] - data[2] - 2; + + /* + * A single type4 opcode can only have so many values attached so + * add enough opcodes to load the all the commands + */ + bosize = (cmds_size + (cmds_size / TYPE4_MAX_PAYLOAD) + 1) << 2; + + ptr = msm_gem_kernel_new(drm, bosize, + MSM_BO_WC | MSM_BO_GPU_READONLY, gpu->vm, + &a5xx_gpu->gpmu_bo, &a5xx_gpu->gpmu_iova); + if (IS_ERR(ptr)) + return; + + msm_gem_object_set_name(a5xx_gpu->gpmu_bo, "gpmufw"); + + while (cmds_size > 0) { + int i; + uint32_t _size = cmds_size > TYPE4_MAX_PAYLOAD ? + TYPE4_MAX_PAYLOAD : cmds_size; + + ptr[dwords++] = PKT4(REG_A5XX_GPMU_INST_RAM_BASE + offset, + _size); + + for (i = 0; i < _size; i++) + ptr[dwords++] = *cmds++; + + offset += _size; + cmds_size -= _size; + } + + msm_gem_put_vaddr(a5xx_gpu->gpmu_bo); + a5xx_gpu->gpmu_dwords = dwords; +} diff --git a/drivers/gpu/drm/msm/adreno/a5xx_preempt.c b/drivers/gpu/drm/msm/adreno/a5xx_preempt.c new file mode 100644 index 000000000..e4924b5e1 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a5xx_preempt.c @@ -0,0 +1,329 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2017 The Linux Foundation. All rights reserved. + */ + +#include "msm_gem.h" +#include "a5xx_gpu.h" + +/* + * Try to transition the preemption state from old to new. Return + * true on success or false if the original state wasn't 'old' + */ +static inline bool try_preempt_state(struct a5xx_gpu *a5xx_gpu, + enum preempt_state old, enum preempt_state new) +{ + enum preempt_state cur = atomic_cmpxchg(&a5xx_gpu->preempt_state, + old, new); + + return (cur == old); +} + +/* + * Force the preemption state to the specified state. This is used in cases + * where the current state is known and won't change + */ +static inline void set_preempt_state(struct a5xx_gpu *gpu, + enum preempt_state new) +{ + /* + * preempt_state may be read by other cores trying to trigger a + * preemption or in the interrupt handler so barriers are needed + * before... + */ + smp_mb__before_atomic(); + atomic_set(&gpu->preempt_state, new); + /* ... and after*/ + smp_mb__after_atomic(); +} + +/* Write the most recent wptr for the given ring into the hardware */ +static inline void update_wptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + unsigned long flags; + uint32_t wptr; + + if (!ring) + return; + + spin_lock_irqsave(&ring->preempt_lock, flags); + wptr = get_wptr(ring); + spin_unlock_irqrestore(&ring->preempt_lock, flags); + + gpu_write(gpu, REG_A5XX_CP_RB_WPTR, wptr); +} + +/* Return the highest priority ringbuffer with something in it */ +static struct msm_ringbuffer *get_next_ring(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + unsigned long flags; + int i; + + for (i = 0; i < gpu->nr_rings; i++) { + bool empty; + struct msm_ringbuffer *ring = gpu->rb[i]; + + spin_lock_irqsave(&ring->preempt_lock, flags); + empty = (get_wptr(ring) == gpu->funcs->get_rptr(gpu, ring)); + if (!empty && ring == a5xx_gpu->cur_ring) + empty = ring->memptrs->fence == a5xx_gpu->last_seqno[i]; + spin_unlock_irqrestore(&ring->preempt_lock, flags); + + if (!empty) + return ring; + } + + return NULL; +} + +static void a5xx_preempt_timer(struct timer_list *t) +{ + struct a5xx_gpu *a5xx_gpu = timer_container_of(a5xx_gpu, t, + preempt_timer); + struct msm_gpu *gpu = &a5xx_gpu->base.base; + struct drm_device *dev = gpu->dev; + + if (!try_preempt_state(a5xx_gpu, PREEMPT_TRIGGERED, PREEMPT_FAULTED)) + return; + + DRM_DEV_ERROR(dev->dev, "%s: preemption timed out\n", gpu->name); + kthread_queue_work(gpu->worker, &gpu->recover_work); +} + +/* Try to trigger a preemption switch */ +void a5xx_preempt_trigger(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + unsigned long flags; + struct msm_ringbuffer *ring; + + if (gpu->nr_rings == 1) + return; + + /* + * Serialize preemption start to ensure that we always make + * decision on latest state. Otherwise we can get stuck in + * lower priority or empty ring. + */ + spin_lock_irqsave(&a5xx_gpu->preempt_start_lock, flags); + + /* + * Try to start preemption by moving from NONE to START. If + * unsuccessful, a preemption is already in flight + */ + if (!try_preempt_state(a5xx_gpu, PREEMPT_NONE, PREEMPT_START)) + goto out; + + /* Get the next ring to preempt to */ + ring = get_next_ring(gpu); + + /* + * If no ring is populated or the highest priority ring is the current + * one do nothing except to update the wptr to the latest and greatest + */ + if (!ring || (a5xx_gpu->cur_ring == ring)) { + /* + * Its possible that while a preemption request is in progress + * from an irq context, a user context trying to submit might + * fail to update the write pointer, because it determines + * that the preempt state is not PREEMPT_NONE. + * + * Close the race by introducing an intermediate + * state PREEMPT_ABORT to let the submit path + * know that the ringbuffer is not going to change + * and can safely update the write pointer. + */ + + set_preempt_state(a5xx_gpu, PREEMPT_ABORT); + update_wptr(gpu, a5xx_gpu->cur_ring); + set_preempt_state(a5xx_gpu, PREEMPT_NONE); + goto out; + } + + spin_unlock_irqrestore(&a5xx_gpu->preempt_start_lock, flags); + + /* Make sure the wptr doesn't update while we're in motion */ + spin_lock_irqsave(&ring->preempt_lock, flags); + a5xx_gpu->preempt[ring->id]->wptr = get_wptr(ring); + spin_unlock_irqrestore(&ring->preempt_lock, flags); + + /* Set the address of the incoming preemption record */ + gpu_write64(gpu, REG_A5XX_CP_CONTEXT_SWITCH_RESTORE_ADDR_LO, + a5xx_gpu->preempt_iova[ring->id]); + + a5xx_gpu->next_ring = ring; + + /* Start a timer to catch a stuck preemption */ + mod_timer(&a5xx_gpu->preempt_timer, jiffies + msecs_to_jiffies(10000)); + + /* Set the preemption state to triggered */ + set_preempt_state(a5xx_gpu, PREEMPT_TRIGGERED); + + /* Make sure everything is written before hitting the button */ + wmb(); + + /* And actually start the preemption */ + gpu_write(gpu, REG_A5XX_CP_CONTEXT_SWITCH_CNTL, 1); + return; + +out: + spin_unlock_irqrestore(&a5xx_gpu->preempt_start_lock, flags); +} + +void a5xx_preempt_irq(struct msm_gpu *gpu) +{ + uint32_t status; + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + struct drm_device *dev = gpu->dev; + + if (!try_preempt_state(a5xx_gpu, PREEMPT_TRIGGERED, PREEMPT_PENDING)) + return; + + /* Delete the preemption watchdog timer */ + timer_delete(&a5xx_gpu->preempt_timer); + + /* + * The hardware should be setting CP_CONTEXT_SWITCH_CNTL to zero before + * firing the interrupt, but there is a non zero chance of a hardware + * condition or a software race that could set it again before we have a + * chance to finish. If that happens, log and go for recovery + */ + status = gpu_read(gpu, REG_A5XX_CP_CONTEXT_SWITCH_CNTL); + if (unlikely(status)) { + set_preempt_state(a5xx_gpu, PREEMPT_FAULTED); + DRM_DEV_ERROR(dev->dev, "%s: Preemption failed to complete\n", + gpu->name); + kthread_queue_work(gpu->worker, &gpu->recover_work); + return; + } + + a5xx_gpu->cur_ring = a5xx_gpu->next_ring; + a5xx_gpu->next_ring = NULL; + + update_wptr(gpu, a5xx_gpu->cur_ring); + + set_preempt_state(a5xx_gpu, PREEMPT_NONE); + + /* + * Try to trigger preemption again in case there was a submit or + * retire during ring switch + */ + a5xx_preempt_trigger(gpu); +} + +void a5xx_preempt_hw_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + int i; + + /* Always come up on rb 0 */ + a5xx_gpu->cur_ring = gpu->rb[0]; + + /* No preemption if we only have one ring */ + if (gpu->nr_rings == 1) + return; + + for (i = 0; i < gpu->nr_rings; i++) { + a5xx_gpu->preempt[i]->data = 0; + a5xx_gpu->preempt[i]->info = 0; + a5xx_gpu->preempt[i]->wptr = 0; + a5xx_gpu->preempt[i]->rptr = 0; + a5xx_gpu->preempt[i]->rbase = gpu->rb[i]->iova; + a5xx_gpu->preempt[i]->rptr_addr = shadowptr(a5xx_gpu, gpu->rb[i]); + } + + /* Write a 0 to signal that we aren't switching pagetables */ + gpu_write64(gpu, REG_A5XX_CP_CONTEXT_SWITCH_SMMU_INFO_LO, 0); + + /* Reset the preemption state */ + set_preempt_state(a5xx_gpu, PREEMPT_NONE); +} + +static int preempt_init_ring(struct a5xx_gpu *a5xx_gpu, + struct msm_ringbuffer *ring) +{ + struct adreno_gpu *adreno_gpu = &a5xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + struct a5xx_preempt_record *ptr; + void *counters; + struct drm_gem_object *bo = NULL, *counters_bo = NULL; + u64 iova = 0, counters_iova = 0; + + ptr = msm_gem_kernel_new(gpu->dev, + A5XX_PREEMPT_RECORD_SIZE + A5XX_PREEMPT_COUNTER_SIZE, + MSM_BO_WC | MSM_BO_MAP_PRIV, gpu->vm, &bo, &iova); + + if (IS_ERR(ptr)) + return PTR_ERR(ptr); + + /* The buffer to store counters needs to be unprivileged */ + counters = msm_gem_kernel_new(gpu->dev, + A5XX_PREEMPT_COUNTER_SIZE, + MSM_BO_WC, gpu->vm, &counters_bo, &counters_iova); + if (IS_ERR(counters)) { + msm_gem_kernel_put(bo, gpu->vm); + return PTR_ERR(counters); + } + + msm_gem_object_set_name(bo, "preempt"); + msm_gem_object_set_name(counters_bo, "preempt_counters"); + + a5xx_gpu->preempt_bo[ring->id] = bo; + a5xx_gpu->preempt_counters_bo[ring->id] = counters_bo; + a5xx_gpu->preempt_iova[ring->id] = iova; + a5xx_gpu->preempt[ring->id] = ptr; + + /* Set up the defaults on the preemption record */ + + ptr->magic = A5XX_PREEMPT_RECORD_MAGIC; + ptr->info = 0; + ptr->data = 0; + ptr->cntl = MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE; + + ptr->counter = counters_iova; + + return 0; +} + +void a5xx_preempt_fini(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + int i; + + for (i = 0; i < gpu->nr_rings; i++) { + msm_gem_kernel_put(a5xx_gpu->preempt_bo[i], gpu->vm); + msm_gem_kernel_put(a5xx_gpu->preempt_counters_bo[i], gpu->vm); + } +} + +void a5xx_preempt_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); + int i; + + /* No preemption if we only have one ring */ + if (gpu->nr_rings <= 1) + return; + + for (i = 0; i < gpu->nr_rings; i++) { + if (preempt_init_ring(a5xx_gpu, gpu->rb[i])) { + /* + * On any failure our adventure is over. Clean up and + * set nr_rings to 1 to force preemption off + */ + a5xx_preempt_fini(gpu); + gpu->nr_rings = 1; + + return; + } + } + + spin_lock_init(&a5xx_gpu->preempt_start_lock); + timer_setup(&a5xx_gpu->preempt_timer, a5xx_preempt_timer, 0); +} diff --git a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c new file mode 100644 index 000000000..90a0c7ffb --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c @@ -0,0 +1,2540 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013-2014 Red Hat + * Author: Rob Clark + * + * Copyright (c) 2014,2017 The Linux Foundation. All rights reserved. + */ + +#include "adreno_gpu.h" +#include "a6xx_gpu.h" +#include "a6xx.xml.h" +#include "a6xx_gmu.xml.h" + +static const struct adreno_reglist a612_hwcg[] = { + {REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220}, + {REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000081}, + {REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x0000f3cf}, + {REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00022222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111}, + {REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777}, + {REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x01202222}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x00040f00}, + {REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x05522022}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00005555}, + {REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011}, + {REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00445044}, + {REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222}, + {REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ_2, 0x00000002}, + {REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000}, + {REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004}, + {REG_A6XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000004}, + {REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002}, + {REG_A6XX_RBBM_ISDB_CNT, 0x00000182}, + {REG_A6XX_RBBM_RAC_THRESHOLD_CNT, 0x00000000}, + {REG_A6XX_RBBM_SP_HYST_CNT, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555}, + {}, +}; + +/* For a615 family (a615, a616, a618 and a619) */ +static const struct adreno_reglist a615_hwcg[] = { + {REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220}, + {REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000080}, + {REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x0000F3CF}, + {REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL_TP1, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_TP1, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_TP1, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00022222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_TP1, 0x00022222}, + {REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST_TP1, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST2_TP1, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST3_TP1, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777}, + {REG_A6XX_RBBM_CLOCK_HYST4_TP1, 0x00077777}, + {REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY_TP1, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY2_TP1, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY3_TP1, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111}, + {REG_A6XX_RBBM_CLOCK_DELAY4_TP1, 0x00011111}, + {REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_UCHE, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_UCHE, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_UCHE, 0x00222222}, + {REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000004}, + {REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002}, + {REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002020}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU1, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU2, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU3, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x00040F00}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU1, 0x00040F00}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU2, 0x00040F00}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU3, 0x00040F00}, + {REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x05022022}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00005555}, + {REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011}, + {REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00445044}, + {REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222}, + {REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x00222222}, + {REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004}, + {REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000}, + {REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200}, + {REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ_2, 0x00000002}, + {REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555}, + {}, +}; + +static const struct adreno_reglist a620_hwcg[] = { + {REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220}, + {REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000080}, + {REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x0000f3cf}, + {REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00022222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111}, + {REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777}, + {REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x01002222}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x00040f00}, + {REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x25222022}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00005555}, + {REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011}, + {REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00445044}, + {REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222}, + {REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x00222222}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ_2, 0x00000002}, + {REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000}, + {REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004}, + {REG_A6XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_TEX_FCHE, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TEX_FCHE, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_TEX_FCHE, 0x00000777}, + {REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000004}, + {REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002}, + {REG_A6XX_RBBM_ISDB_CNT, 0x00000182}, + {REG_A6XX_RBBM_RAC_THRESHOLD_CNT, 0x00000000}, + {REG_A6XX_RBBM_SP_HYST_CNT, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555}, + {}, +}; + +static const struct adreno_reglist a630_hwcg[] = { + {REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL_SP1, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL_SP2, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL_SP3, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02022220}, + {REG_A6XX_RBBM_CLOCK_CNTL2_SP1, 0x02022220}, + {REG_A6XX_RBBM_CLOCK_CNTL2_SP2, 0x02022220}, + {REG_A6XX_RBBM_CLOCK_CNTL2_SP3, 0x02022220}, + {REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000080}, + {REG_A6XX_RBBM_CLOCK_DELAY_SP1, 0x00000080}, + {REG_A6XX_RBBM_CLOCK_DELAY_SP2, 0x00000080}, + {REG_A6XX_RBBM_CLOCK_DELAY_SP3, 0x00000080}, + {REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x0000f3cf}, + {REG_A6XX_RBBM_CLOCK_HYST_SP1, 0x0000f3cf}, + {REG_A6XX_RBBM_CLOCK_HYST_SP2, 0x0000f3cf}, + {REG_A6XX_RBBM_CLOCK_HYST_SP3, 0x0000f3cf}, + {REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL_TP1, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL_TP2, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL_TP3, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_TP1, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_TP2, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_TP3, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_TP1, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_TP2, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_TP3, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00022222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_TP1, 0x00022222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_TP2, 0x00022222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_TP3, 0x00022222}, + {REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST_TP1, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST_TP2, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST_TP3, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST2_TP1, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST2_TP2, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST2_TP3, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST3_TP1, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST3_TP2, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST3_TP3, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777}, + {REG_A6XX_RBBM_CLOCK_HYST4_TP1, 0x00077777}, + {REG_A6XX_RBBM_CLOCK_HYST4_TP2, 0x00077777}, + {REG_A6XX_RBBM_CLOCK_HYST4_TP3, 0x00077777}, + {REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY_TP1, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY_TP2, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY_TP3, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY2_TP1, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY2_TP2, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY2_TP3, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY3_TP1, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY3_TP2, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY3_TP3, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111}, + {REG_A6XX_RBBM_CLOCK_DELAY4_TP1, 0x00011111}, + {REG_A6XX_RBBM_CLOCK_DELAY4_TP2, 0x00011111}, + {REG_A6XX_RBBM_CLOCK_DELAY4_TP3, 0x00011111}, + {REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_UCHE, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_UCHE, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_UCHE, 0x00222222}, + {REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000004}, + {REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002}, + {REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL_RB1, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL_RB2, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL_RB3, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RB1, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RB2, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RB3, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU1, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU2, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU3, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x00040f00}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU1, 0x00040f00}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU2, 0x00040f00}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU3, 0x00040f00}, + {REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x05022022}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00005555}, + {REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011}, + {REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00445044}, + {REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222}, + {REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x00222222}, + {REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004}, + {REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000}, + {REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200}, + {REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ_2, 0x00000002}, + {REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555}, + {}, +}; + +static const struct adreno_reglist a640_hwcg[] = { + {REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220}, + {REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000080}, + {REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x0000F3CF}, + {REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00022222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111}, + {REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777}, + {REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x01002222}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x00040F00}, + {REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x05222022}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00005555}, + {REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011}, + {REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00445044}, + {REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222}, + {REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x00222222}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ_2, 0x00000002}, + {REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000}, + {REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004}, + {REG_A6XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_TEX_FCHE, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TEX_FCHE, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_TEX_FCHE, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000004}, + {REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002}, + {REG_A6XX_RBBM_ISDB_CNT, 0x00000182}, + {REG_A6XX_RBBM_RAC_THRESHOLD_CNT, 0x00000000}, + {REG_A6XX_RBBM_SP_HYST_CNT, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555}, + {}, +}; + +static const struct adreno_reglist a650_hwcg[] = { + {REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220}, + {REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000080}, + {REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x0000F3CF}, + {REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00022222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111}, + {REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777}, + {REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x01002222}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x00040F00}, + {REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x25222022}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00005555}, + {REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011}, + {REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00445044}, + {REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222}, + {REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x00222222}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ_2, 0x00000002}, + {REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000}, + {REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004}, + {REG_A6XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_TEX_FCHE, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TEX_FCHE, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_TEX_FCHE, 0x00000777}, + {REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000004}, + {REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002}, + {REG_A6XX_RBBM_ISDB_CNT, 0x00000182}, + {REG_A6XX_RBBM_RAC_THRESHOLD_CNT, 0x00000000}, + {REG_A6XX_RBBM_SP_HYST_CNT, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555}, + {}, +}; + +static const struct adreno_reglist a660_hwcg[] = { + {REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220}, + {REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000080}, + {REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x0000F3CF}, + {REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00022222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111}, + {REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777}, + {REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x01002222}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x00040F00}, + {REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x25222022}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00005555}, + {REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011}, + {REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00445044}, + {REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222}, + {REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x00222222}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ_2, 0x00000002}, + {REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000}, + {REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004}, + {REG_A6XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_TEX_FCHE, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TEX_FCHE, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_TEX_FCHE, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000004}, + {REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002}, + {REG_A6XX_RBBM_ISDB_CNT, 0x00000182}, + {REG_A6XX_RBBM_RAC_THRESHOLD_CNT, 0x00000000}, + {REG_A6XX_RBBM_SP_HYST_CNT, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555}, + {}, +}; + +static const struct adreno_reglist a690_hwcg[] = { + {REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220}, + {REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000080}, + {REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x0000F3CF}, + {REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00022222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111}, + {REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777}, + {REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x01002222}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x00040F00}, + {REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x25222022}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00005555}, + {REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011}, + {REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00445044}, + {REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222}, + {REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x00222222}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ_2, 0x00000002}, + {REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000}, + {REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004}, + {REG_A6XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_TEX_FCHE, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TEX_FCHE, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_TEX_FCHE, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000004}, + {REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002}, + {REG_A6XX_RBBM_CLOCK_CNTL, 0x8AA8AA82}, + {REG_A6XX_RBBM_ISDB_CNT, 0x00000182}, + {REG_A6XX_RBBM_RAC_THRESHOLD_CNT, 0x00000000}, + {REG_A6XX_RBBM_SP_HYST_CNT, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555}, + {} +}; + +/* For a615, a616, a618, a619, a630, a640 and a680 */ +static const u32 a630_protect_regs[] = { + A6XX_PROTECT_RDONLY(0x00000, 0x04ff), + A6XX_PROTECT_RDONLY(0x00501, 0x0005), + A6XX_PROTECT_RDONLY(0x0050b, 0x02f4), + A6XX_PROTECT_NORDWR(0x0050e, 0x0000), + A6XX_PROTECT_NORDWR(0x00510, 0x0000), + A6XX_PROTECT_NORDWR(0x00534, 0x0000), + A6XX_PROTECT_NORDWR(0x00800, 0x0082), + A6XX_PROTECT_NORDWR(0x008a0, 0x0008), + A6XX_PROTECT_NORDWR(0x008ab, 0x0024), + A6XX_PROTECT_RDONLY(0x008de, 0x00ae), + A6XX_PROTECT_NORDWR(0x00900, 0x004d), + A6XX_PROTECT_NORDWR(0x0098d, 0x0272), + A6XX_PROTECT_NORDWR(0x00e00, 0x0001), + A6XX_PROTECT_NORDWR(0x00e03, 0x000c), + A6XX_PROTECT_NORDWR(0x03c00, 0x00c3), + A6XX_PROTECT_RDONLY(0x03cc4, 0x1fff), + A6XX_PROTECT_NORDWR(0x08630, 0x01cf), + A6XX_PROTECT_NORDWR(0x08e00, 0x0000), + A6XX_PROTECT_NORDWR(0x08e08, 0x0000), + A6XX_PROTECT_NORDWR(0x08e50, 0x001f), + A6XX_PROTECT_NORDWR(0x09624, 0x01db), + A6XX_PROTECT_NORDWR(0x09e70, 0x0001), + A6XX_PROTECT_NORDWR(0x09e78, 0x0187), + A6XX_PROTECT_NORDWR(0x0a630, 0x01cf), + A6XX_PROTECT_NORDWR(0x0ae02, 0x0000), + A6XX_PROTECT_NORDWR(0x0ae50, 0x032f), + A6XX_PROTECT_NORDWR(0x0b604, 0x0000), + A6XX_PROTECT_NORDWR(0x0be02, 0x0001), + A6XX_PROTECT_NORDWR(0x0be20, 0x17df), + A6XX_PROTECT_NORDWR(0x0f000, 0x0bff), + A6XX_PROTECT_RDONLY(0x0fc00, 0x1fff), + A6XX_PROTECT_NORDWR(0x11c00, 0x0000), /* note: infinite range */ +}; +DECLARE_ADRENO_PROTECT(a630_protect, 32); + +static const u32 a650_protect_regs[] = { + A6XX_PROTECT_RDONLY(0x00000, 0x04ff), + A6XX_PROTECT_RDONLY(0x00501, 0x0005), + A6XX_PROTECT_RDONLY(0x0050b, 0x02f4), + A6XX_PROTECT_NORDWR(0x0050e, 0x0000), + A6XX_PROTECT_NORDWR(0x00510, 0x0000), + A6XX_PROTECT_NORDWR(0x00534, 0x0000), + A6XX_PROTECT_NORDWR(0x00800, 0x0082), + A6XX_PROTECT_NORDWR(0x008a0, 0x0008), + A6XX_PROTECT_NORDWR(0x008ab, 0x0024), + A6XX_PROTECT_RDONLY(0x008de, 0x00ae), + A6XX_PROTECT_NORDWR(0x00900, 0x004d), + A6XX_PROTECT_NORDWR(0x0098d, 0x0272), + A6XX_PROTECT_NORDWR(0x00e00, 0x0001), + A6XX_PROTECT_NORDWR(0x00e03, 0x000c), + A6XX_PROTECT_NORDWR(0x03c00, 0x00c3), + A6XX_PROTECT_RDONLY(0x03cc4, 0x1fff), + A6XX_PROTECT_NORDWR(0x08630, 0x01cf), + A6XX_PROTECT_NORDWR(0x08e00, 0x0000), + A6XX_PROTECT_NORDWR(0x08e08, 0x0000), + A6XX_PROTECT_NORDWR(0x08e50, 0x001f), + A6XX_PROTECT_NORDWR(0x08e80, 0x027f), + A6XX_PROTECT_NORDWR(0x09624, 0x01db), + A6XX_PROTECT_NORDWR(0x09e60, 0x0011), + A6XX_PROTECT_NORDWR(0x09e78, 0x0187), + A6XX_PROTECT_NORDWR(0x0a630, 0x01cf), + A6XX_PROTECT_NORDWR(0x0ae02, 0x0000), + A6XX_PROTECT_NORDWR(0x0ae50, 0x032f), + A6XX_PROTECT_NORDWR(0x0b604, 0x0000), + A6XX_PROTECT_NORDWR(0x0b608, 0x0007), + A6XX_PROTECT_NORDWR(0x0be02, 0x0001), + A6XX_PROTECT_NORDWR(0x0be20, 0x17df), + A6XX_PROTECT_NORDWR(0x0f000, 0x0bff), + A6XX_PROTECT_RDONLY(0x0fc00, 0x1fff), + A6XX_PROTECT_NORDWR(0x18400, 0x1fff), + A6XX_PROTECT_NORDWR(0x1a800, 0x1fff), + A6XX_PROTECT_NORDWR(0x1f400, 0x0443), + A6XX_PROTECT_RDONLY(0x1f844, 0x007b), + A6XX_PROTECT_NORDWR(0x1f887, 0x001b), + A6XX_PROTECT_NORDWR(0x1f8c0, 0x0000), /* note: infinite range */ +}; +DECLARE_ADRENO_PROTECT(a650_protect, 48); + +/* These are for a635 and a660 */ +static const u32 a660_protect_regs[] = { + A6XX_PROTECT_RDONLY(0x00000, 0x04ff), + A6XX_PROTECT_RDONLY(0x00501, 0x0005), + A6XX_PROTECT_RDONLY(0x0050b, 0x02f4), + A6XX_PROTECT_NORDWR(0x0050e, 0x0000), + A6XX_PROTECT_NORDWR(0x00510, 0x0000), + A6XX_PROTECT_NORDWR(0x00534, 0x0000), + A6XX_PROTECT_NORDWR(0x00800, 0x0082), + A6XX_PROTECT_NORDWR(0x008a0, 0x0008), + A6XX_PROTECT_NORDWR(0x008ab, 0x0024), + A6XX_PROTECT_RDONLY(0x008de, 0x00ae), + A6XX_PROTECT_NORDWR(0x00900, 0x004d), + A6XX_PROTECT_NORDWR(0x0098d, 0x0272), + A6XX_PROTECT_NORDWR(0x00e00, 0x0001), + A6XX_PROTECT_NORDWR(0x00e03, 0x000c), + A6XX_PROTECT_NORDWR(0x03c00, 0x00c3), + A6XX_PROTECT_RDONLY(0x03cc4, 0x1fff), + A6XX_PROTECT_NORDWR(0x08630, 0x01cf), + A6XX_PROTECT_NORDWR(0x08e00, 0x0000), + A6XX_PROTECT_NORDWR(0x08e08, 0x0000), + A6XX_PROTECT_NORDWR(0x08e50, 0x001f), + A6XX_PROTECT_NORDWR(0x08e80, 0x027f), + A6XX_PROTECT_NORDWR(0x09624, 0x01db), + A6XX_PROTECT_NORDWR(0x09e60, 0x0011), + A6XX_PROTECT_NORDWR(0x09e78, 0x0187), + A6XX_PROTECT_NORDWR(0x0a630, 0x01cf), + A6XX_PROTECT_NORDWR(0x0ae02, 0x0000), + A6XX_PROTECT_NORDWR(0x0ae50, 0x012f), + A6XX_PROTECT_NORDWR(0x0b604, 0x0000), + A6XX_PROTECT_NORDWR(0x0b608, 0x0006), + A6XX_PROTECT_NORDWR(0x0be02, 0x0001), + A6XX_PROTECT_NORDWR(0x0be20, 0x015f), + A6XX_PROTECT_NORDWR(0x0d000, 0x05ff), + A6XX_PROTECT_NORDWR(0x0f000, 0x0bff), + A6XX_PROTECT_RDONLY(0x0fc00, 0x1fff), + A6XX_PROTECT_NORDWR(0x18400, 0x1fff), + A6XX_PROTECT_NORDWR(0x1a400, 0x1fff), + A6XX_PROTECT_NORDWR(0x1f400, 0x0443), + A6XX_PROTECT_RDONLY(0x1f844, 0x007b), + A6XX_PROTECT_NORDWR(0x1f860, 0x0000), + A6XX_PROTECT_NORDWR(0x1f887, 0x001b), + A6XX_PROTECT_NORDWR(0x1f8c0, 0x0000), /* note: infinite range */ +}; +DECLARE_ADRENO_PROTECT(a660_protect, 48); + +/* These are for a690 */ +static const u32 a690_protect_regs[] = { + A6XX_PROTECT_RDONLY(0x00000, 0x004ff), + A6XX_PROTECT_RDONLY(0x00501, 0x00001), + A6XX_PROTECT_RDONLY(0x0050b, 0x002f4), + A6XX_PROTECT_NORDWR(0x0050e, 0x00000), + A6XX_PROTECT_NORDWR(0x00510, 0x00000), + A6XX_PROTECT_NORDWR(0x00534, 0x00000), + A6XX_PROTECT_NORDWR(0x00800, 0x00082), + A6XX_PROTECT_NORDWR(0x008a0, 0x00008), + A6XX_PROTECT_NORDWR(0x008ab, 0x00024), + A6XX_PROTECT_RDONLY(0x008de, 0x000ae), + A6XX_PROTECT_NORDWR(0x00900, 0x0004d), + A6XX_PROTECT_NORDWR(0x0098d, 0x00272), + A6XX_PROTECT_NORDWR(0x00e00, 0x00001), + A6XX_PROTECT_NORDWR(0x00e03, 0x0000c), + A6XX_PROTECT_NORDWR(0x03c00, 0x000c3), + A6XX_PROTECT_RDONLY(0x03cc4, 0x01fff), + A6XX_PROTECT_NORDWR(0x08630, 0x001cf), + A6XX_PROTECT_NORDWR(0x08e00, 0x00000), + A6XX_PROTECT_NORDWR(0x08e08, 0x00007), + A6XX_PROTECT_NORDWR(0x08e50, 0x0001f), + A6XX_PROTECT_NORDWR(0x08e80, 0x0027f), + A6XX_PROTECT_NORDWR(0x09624, 0x001db), + A6XX_PROTECT_NORDWR(0x09e60, 0x00011), + A6XX_PROTECT_NORDWR(0x09e78, 0x00187), + A6XX_PROTECT_NORDWR(0x0a630, 0x001cf), + A6XX_PROTECT_NORDWR(0x0ae02, 0x00000), + A6XX_PROTECT_NORDWR(0x0ae50, 0x0012f), + A6XX_PROTECT_NORDWR(0x0b604, 0x00000), + A6XX_PROTECT_NORDWR(0x0b608, 0x00006), + A6XX_PROTECT_NORDWR(0x0be02, 0x00001), + A6XX_PROTECT_NORDWR(0x0be20, 0x0015f), + A6XX_PROTECT_NORDWR(0x0d000, 0x005ff), + A6XX_PROTECT_NORDWR(0x0f000, 0x00bff), + A6XX_PROTECT_RDONLY(0x0fc00, 0x01fff), + A6XX_PROTECT_NORDWR(0x11c00, 0x00000), /*note: infiite range */ +}; +DECLARE_ADRENO_PROTECT(a690_protect, 48); + +static const struct adreno_reglist a640_gbif[] = { + { REG_A6XX_GBIF_QSB_SIDE0, 0x00071620 }, + { REG_A6XX_GBIF_QSB_SIDE1, 0x00071620 }, + { REG_A6XX_GBIF_QSB_SIDE2, 0x00071620 }, + { REG_A6XX_GBIF_QSB_SIDE3, 0x00071620 }, + { }, +}; + +static const struct adreno_info a6xx_gpus[] = { + { + .chip_ids = ADRENO_CHIP_IDS(0x06010000), + .family = ADRENO_6XX_GEN1, + .revn = 610, + .fw = { + [ADRENO_FW_SQE] = "a630_sqe.fw", + }, + .gmem = (SZ_128K + SZ_4K), + .quirks = ADRENO_QUIRK_4GB_VA, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a6xx_gmuwrapper_funcs, + .zapfw = "a610_zap.mdt", + .a6xx = &(const struct a6xx_info) { + .hwcg = a612_hwcg, + .protect = &a630_protect, + .gbif_cx = a640_gbif, + .gmu_cgc_mode = 0x00020202, + .prim_fifo_threshold = 0x00080000, + }, + /* + * There are (at least) three SoCs implementing A610: SM6125 + * (trinket), SM6115 (bengal) and SM6225 (khaje). Trinket does + * not have speedbinning, as only a single SKU exists and we + * don't support khaje upstream yet. Hence, this matching + * table is only valid for bengal. + */ + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 206, 1 }, + { 200, 2 }, + { 157, 3 }, + { 127, 4 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS(0x06010200), + .family = ADRENO_6XX_GEN1, + .fw = { + [ADRENO_FW_SQE] = "a630_sqe.fw", + [ADRENO_FW_GMU] = "a612_rgmu.bin", + }, + .gmem = (SZ_128K + SZ_4K), + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a6xx_gmuwrapper_funcs, + .a6xx = &(const struct a6xx_info) { + .hwcg = a612_hwcg, + .protect = &a630_protect, + .gmu_cgc_mode = 0x00000022, + .prim_fifo_threshold = 0x00080000, + }, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x06010500), + .family = ADRENO_6XX_GEN1, + .revn = 615, + .fw = { + [ADRENO_FW_SQE] = "a630_sqe.fw", + [ADRENO_FW_GMU] = "a630_gmu.bin", + }, + .gmem = SZ_512K, + .quirks = ADRENO_QUIRK_4GB_VA, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a6xx_gpu_funcs, + .zapfw = "a615_zap.mdt", + .a6xx = &(const struct a6xx_info) { + .hwcg = a615_hwcg, + .protect = &a630_protect, + .gmu_cgc_mode = 0x00000222, + .prim_fifo_threshold = 0x0018000, + }, + .speedbins = ADRENO_SPEEDBINS( + /* + * The default speed bin (0) has the same values as + * speed bin 90 which goes up to 432 MHz. + */ + { 0, 0 }, + { 90, 0 }, + { 105, 1 }, + { 146, 2 }, + { 163, 3 }, + ), + }, { + .machine = "qcom,sm7150", + .chip_ids = ADRENO_CHIP_IDS(0x06010800), + .family = ADRENO_6XX_GEN1, + .fw = { + [ADRENO_FW_SQE] = "a630_sqe.fw", + [ADRENO_FW_GMU] = "a630_gmu.bin", + }, + .gmem = SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_4GB_VA, + .funcs = &a6xx_gpu_funcs, + .zapfw = "a615_zap.mbn", + .a6xx = &(const struct a6xx_info) { + .hwcg = a615_hwcg, + .protect = &a630_protect, + .gmu_cgc_mode = 0x00000222, + .prim_fifo_threshold = 0x00180000, + }, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 128, 1 }, + { 146, 2 }, + { 167, 3 }, + { 172, 4 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS(0x06010800), + .family = ADRENO_6XX_GEN1, + .revn = 618, + .fw = { + [ADRENO_FW_SQE] = "a630_sqe.fw", + [ADRENO_FW_GMU] = "a630_gmu.bin", + }, + .gmem = SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_4GB_VA, + .funcs = &a6xx_gpu_funcs, + .a6xx = &(const struct a6xx_info) { + .protect = &a630_protect, + .gmu_cgc_mode = 0x00000222, + .prim_fifo_threshold = 0x00180000, + }, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 169, 1 }, + { 174, 2 }, + ), + }, { + .machine = "qcom,sm4350", + .chip_ids = ADRENO_CHIP_IDS(0x06010900), + .family = ADRENO_6XX_GEN1, + .revn = 619, + .fw = { + [ADRENO_FW_SQE] = "a630_sqe.fw", + [ADRENO_FW_GMU] = "a619_gmu.bin", + }, + .gmem = SZ_512K, + .quirks = ADRENO_QUIRK_4GB_VA, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a6xx_gpu_funcs, + .zapfw = "a615_zap.mdt", + .a6xx = &(const struct a6xx_info) { + .hwcg = a615_hwcg, + .protect = &a630_protect, + .gmu_cgc_mode = 0x00000222, + .prim_fifo_threshold = 0x00018000, + }, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 138, 1 }, + { 92, 2 }, + ), + }, { + .machine = "qcom,sm6375", + .chip_ids = ADRENO_CHIP_IDS(0x06010901), + .family = ADRENO_6XX_GEN1, + .revn = 619, + .fw = { + [ADRENO_FW_SQE] = "a630_sqe.fw", + [ADRENO_FW_GMU] = "a619_gmu.bin", + }, + .gmem = SZ_512K, + .quirks = ADRENO_QUIRK_4GB_VA, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a6xx_gpu_funcs, + .zapfw = "a615_zap.mdt", + .a6xx = &(const struct a6xx_info) { + .hwcg = a615_hwcg, + .protect = &a630_protect, + .gmu_cgc_mode = 0x00000222, + .prim_fifo_threshold = 0x00018000, + }, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 190, 1 }, + { 177, 2 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS(0x06010900), + .family = ADRENO_6XX_GEN1, + .revn = 619, + .fw = { + [ADRENO_FW_SQE] = "a630_sqe.fw", + [ADRENO_FW_GMU] = "a619_gmu.bin", + }, + .gmem = SZ_512K, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_4GB_VA, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a6xx_gpu_funcs, + .zapfw = "a615_zap.mdt", + .a6xx = &(const struct a6xx_info) { + .hwcg = a615_hwcg, + .protect = &a630_protect, + .gmu_cgc_mode = 0x00000222, + .prim_fifo_threshold = 0x00018000, + }, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 120, 4 }, + { 138, 3 }, + { 169, 2 }, + { 180, 1 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS(0x06020100), + .family = ADRENO_6XX_GEN3, + .fw = { + [ADRENO_FW_SQE] = "a650_sqe.fw", + [ADRENO_FW_GMU] = "a621_gmu.bin", + }, + .gmem = SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV, + .funcs = &a6xx_gpu_funcs, + .zapfw = "a620_zap.mbn", + .a6xx = &(const struct a6xx_info) { + .hwcg = a620_hwcg, + .protect = &a650_protect, + .gbif_cx = a640_gbif, + .gmu_cgc_mode = 0x00020200, + .prim_fifo_threshold = 0x00010000, + }, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 137, 1 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS(0x06020300), + .family = ADRENO_6XX_GEN3, + .fw = { + [ADRENO_FW_SQE] = "a650_sqe.fw", + [ADRENO_FW_GMU] = "a623_gmu.bin", + }, + .gmem = SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV, + .funcs = &a6xx_gpu_funcs, + .a6xx = &(const struct a6xx_info) { + .hwcg = a690_hwcg, + .protect = &a650_protect, + .gbif_cx = a640_gbif, + .gmu_cgc_mode = 0x00020200, + .prim_fifo_threshold = 0x00010000, + .bcms = (const struct a6xx_bcm[]) { + { .name = "SH0", .buswidth = 16 }, + { .name = "MC0", .buswidth = 4 }, + { + .name = "ACV", + .fixed = true, + .perfmode = BIT(3), + }, + { /* sentinel */ }, + }, + }, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 185, 0 }, + { 127, 1 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS( + 0x06030001, + 0x06030002 + ), + .family = ADRENO_6XX_GEN1, + .revn = 630, + .fw = { + [ADRENO_FW_SQE] = "a630_sqe.fw", + [ADRENO_FW_GMU] = "a630_gmu.bin", + }, + .gmem = SZ_1M, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_4GB_VA, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a6xx_gpu_funcs, + .zapfw = "a630_zap.mdt", + .a6xx = &(const struct a6xx_info) { + .hwcg = a630_hwcg, + .protect = &a630_protect, + .gmu_cgc_mode = 0x00020202, + .prim_fifo_threshold = 0x00180000, + }, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x06040001), + .family = ADRENO_6XX_GEN2, + .revn = 640, + .fw = { + [ADRENO_FW_SQE] = "a630_sqe.fw", + [ADRENO_FW_GMU] = "a640_gmu.bin", + }, + .gmem = SZ_1M, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_4GB_VA, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a6xx_gpu_funcs, + .zapfw = "a640_zap.mdt", + .a6xx = &(const struct a6xx_info) { + .hwcg = a640_hwcg, + .protect = &a630_protect, + .gmu_cgc_mode = 0x00020202, + .prim_fifo_threshold = 0x00180000, + }, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 1, 1 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS(0x06050002), + .family = ADRENO_6XX_GEN3, + .revn = 650, + .fw = { + [ADRENO_FW_SQE] = "a650_sqe.fw", + [ADRENO_FW_GMU] = "a650_gmu.bin", + }, + .gmem = SZ_1M + SZ_128K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV, + .funcs = &a6xx_gpu_funcs, + .zapfw = "a650_zap.mdt", + .a6xx = &(const struct a6xx_info) { + .hwcg = a650_hwcg, + .protect = &a650_protect, + .gbif_cx = a640_gbif, + .gmu_cgc_mode = 0x00020202, + .prim_fifo_threshold = 0x00300200, + }, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 1, 1 }, + { 2, 3 }, /* Yep, 2 and 3 are swapped! :/ */ + { 3, 2 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS(0x06060001), + .family = ADRENO_6XX_GEN4, + .revn = 660, + .fw = { + [ADRENO_FW_SQE] = "a660_sqe.fw", + [ADRENO_FW_GMU] = "a660_gmu.bin", + }, + .gmem = SZ_1M + SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV, + .funcs = &a6xx_gpu_funcs, + .zapfw = "a660_zap.mdt", + .a6xx = &(const struct a6xx_info) { + .hwcg = a660_hwcg, + .protect = &a660_protect, + .gbif_cx = a640_gbif, + .gmu_cgc_mode = 0x00020000, + .prim_fifo_threshold = 0x00300200, + }, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x06060300), + .family = ADRENO_6XX_GEN4, + .fw = { + [ADRENO_FW_SQE] = "a660_sqe.fw", + [ADRENO_FW_GMU] = "a663_gmu.bin", + }, + .gmem = SZ_1M + SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV, + .funcs = &a6xx_gpu_funcs, + .a6xx = &(const struct a6xx_info) { + .hwcg = a690_hwcg, + .protect = &a660_protect, + .gbif_cx = a640_gbif, + .gmu_cgc_mode = 0x00020200, + .prim_fifo_threshold = 0x00300200, + }, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 169, 0 }, + { 113, 1 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS(0x06030500), + .family = ADRENO_6XX_GEN4, + .fw = { + [ADRENO_FW_SQE] = "a660_sqe.fw", + [ADRENO_FW_GMU] = "a660_gmu.bin", + }, + .gmem = SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV, + .funcs = &a6xx_gpu_funcs, + .zapfw = "a660_zap.mbn", + .a6xx = &(const struct a6xx_info) { + .hwcg = a660_hwcg, + .protect = &a660_protect, + .gbif_cx = a640_gbif, + .gmu_cgc_mode = 0x00020202, + .prim_fifo_threshold = 0x00200200, + }, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 117, 0 }, + { 129, 4 }, + { 172, 2 }, /* Called speedbin 1 downstream, but let's not break things! */ + { 190, 1 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS(0x06080001), + .family = ADRENO_6XX_GEN2, + .revn = 680, + .fw = { + [ADRENO_FW_SQE] = "a630_sqe.fw", + [ADRENO_FW_GMU] = "a640_gmu.bin", + }, + .gmem = SZ_2M, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_4GB_VA, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .funcs = &a6xx_gpu_funcs, + .zapfw = "a640_zap.mdt", + .a6xx = &(const struct a6xx_info) { + .hwcg = a640_hwcg, + .protect = &a630_protect, + .gmu_cgc_mode = 0x00020202, + .prim_fifo_threshold = 0x00200200, + }, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x06090000), + .family = ADRENO_6XX_GEN4, + .fw = { + [ADRENO_FW_SQE] = "a660_sqe.fw", + [ADRENO_FW_GMU] = "a660_gmu.bin", + }, + .gmem = SZ_4M, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV, + .funcs = &a6xx_gpu_funcs, + .zapfw = "a690_zap.mdt", + .a6xx = &(const struct a6xx_info) { + .hwcg = a690_hwcg, + .protect = &a690_protect, + .gbif_cx = a640_gbif, + .gmu_cgc_mode = 0x00020200, + .prim_fifo_threshold = 0x00800200, + }, + } +}; +DECLARE_ADRENO_GPULIST(a6xx); + +static const struct adreno_reglist a702_hwcg[] = { + { REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x22222222 }, + { REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220 }, + { REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000081 }, + { REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x0000f3cf }, + { REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x22222222 }, + { REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222 }, + { REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22222222 }, + { REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00022222 }, + { REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111 }, + { REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111 }, + { REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111 }, + { REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111 }, + { REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777 }, + { REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777 }, + { REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777 }, + { REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777 }, + { REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222 }, + { REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x01202222 }, + { REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002220 }, + { REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x00040f00 }, + { REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x05522022 }, + { REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00005555 }, + { REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011 }, + { REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00445044 }, + { REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222 }, + { REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00002222 }, + { REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x02222222 }, + { REG_A6XX_RBBM_CLOCK_DELAY_HLSQ_2, 0x00000002 }, + { REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222 }, + { REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000 }, + { REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00002222 }, + { REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200 }, + { REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004 }, + { REG_A6XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222 }, + { REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000004 }, + { REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002 }, + { REG_A6XX_RBBM_ISDB_CNT, 0x00000182 }, + { REG_A6XX_RBBM_RAC_THRESHOLD_CNT, 0x00000000 }, + { REG_A6XX_RBBM_SP_HYST_CNT, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222 }, + { REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111 }, + { REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555 }, + { REG_A6XX_RBBM_CLOCK_CNTL_FCHE, 0x00000222 }, + { REG_A6XX_RBBM_CLOCK_DELAY_FCHE, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_HYST_FCHE, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_CNTL_GLC, 0x00222222 }, + { REG_A6XX_RBBM_CLOCK_DELAY_GLC, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_HYST_GLC, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_CNTL_MHUB, 0x00000002 }, + { REG_A6XX_RBBM_CLOCK_DELAY_MHUB, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_HYST_MHUB, 0x00000000 }, + {} +}; + +static const struct adreno_reglist a730_hwcg[] = { + { REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x02222222 }, + { REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02022222 }, + { REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x0000f3cf }, + { REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000080 }, + { REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x22222220 }, + { REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222 }, + { REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22220222 }, + { REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00222222 }, + { REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777 }, + { REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777 }, + { REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777 }, + { REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777 }, + { REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111 }, + { REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111 }, + { REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111 }, + { REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111 }, + { REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222 }, + { REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000004 }, + { REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002 }, + { REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222 }, + { REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x01002222 }, + { REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002220 }, + { REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x44000f00 }, + { REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x25222022 }, + { REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00555555 }, + { REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011 }, + { REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00440044 }, + { REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222 }, + { REG_A7XX_RBBM_CLOCK_MODE2_GRAS, 0x00000222 }, + { REG_A7XX_RBBM_CLOCK_MODE_BV_GRAS, 0x00222222 }, + { REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x02222223 }, + { REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00002222 }, + { REG_A7XX_RBBM_CLOCK_MODE_BV_GPC, 0x00222222 }, + { REG_A7XX_RBBM_CLOCK_MODE_BV_VFD, 0x00002222 }, + { REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004 }, + { REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000 }, + { REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200 }, + { REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00002222 }, + { REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222 }, + { REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_DELAY_HLSQ_2, 0x00000002 }, + { REG_A7XX_RBBM_CLOCK_MODE_BV_LRZ, 0x55555552 }, + { REG_A7XX_RBBM_CLOCK_MODE_CP, 0x00000223 }, + { REG_A6XX_RBBM_CLOCK_CNTL, 0x8aa8aa82 }, + { REG_A6XX_RBBM_ISDB_CNT, 0x00000182 }, + { REG_A6XX_RBBM_RAC_THRESHOLD_CNT, 0x00000000 }, + { REG_A6XX_RBBM_SP_HYST_CNT, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222 }, + { REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111 }, + { REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555 }, + {}, +}; + +static const struct adreno_reglist a740_hwcg[] = { + { REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x02222222 }, + { REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x22022222 }, + { REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x003cf3cf }, + { REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000080 }, + { REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x22222220 }, + { REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222 }, + { REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22222222 }, + { REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00222222 }, + { REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777 }, + { REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777 }, + { REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777 }, + { REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777 }, + { REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111 }, + { REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111 }, + { REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111 }, + { REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111 }, + { REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222 }, + { REG_A6XX_RBBM_CLOCK_CNTL2_UCHE, 0x00222222 }, + { REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000444 }, + { REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000222 }, + { REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222 }, + { REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x01002222 }, + { REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002220 }, + { REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x44000f00 }, + { REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x25222022 }, + { REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00555555 }, + { REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011 }, + { REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00440044 }, + { REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222 }, + { REG_A7XX_RBBM_CLOCK_MODE2_GRAS, 0x00000222 }, + { REG_A7XX_RBBM_CLOCK_MODE_BV_GRAS, 0x00222222 }, + { REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x02222223 }, + { REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00222222 }, + { REG_A7XX_RBBM_CLOCK_MODE_BV_GPC, 0x00222222 }, + { REG_A7XX_RBBM_CLOCK_MODE_BV_VFD, 0x00002222 }, + { REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004 }, + { REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200 }, + { REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222 }, + { REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000 }, + { REG_A7XX_RBBM_CLOCK_MODE_BV_LRZ, 0x55555552 }, + { REG_A7XX_RBBM_CLOCK_HYST2_VFD, 0x00000000 }, + { REG_A7XX_RBBM_CLOCK_MODE_CP, 0x00000222 }, + { REG_A6XX_RBBM_CLOCK_CNTL, 0x8aa8aa82 }, + { REG_A6XX_RBBM_ISDB_CNT, 0x00000182 }, + { REG_A6XX_RBBM_RAC_THRESHOLD_CNT, 0x00000000 }, + { REG_A6XX_RBBM_SP_HYST_CNT, 0x00000000 }, + { REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222 }, + { REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111 }, + { REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555 }, + {}, +}; + +static const u32 a730_protect_regs[] = { + A6XX_PROTECT_RDONLY(0x00000, 0x04ff), + A6XX_PROTECT_RDONLY(0x0050b, 0x0058), + A6XX_PROTECT_NORDWR(0x0050e, 0x0000), + A6XX_PROTECT_NORDWR(0x00510, 0x0000), + A6XX_PROTECT_NORDWR(0x00534, 0x0000), + A6XX_PROTECT_RDONLY(0x005fb, 0x009d), + A6XX_PROTECT_NORDWR(0x00699, 0x01e9), + A6XX_PROTECT_NORDWR(0x008a0, 0x0008), + A6XX_PROTECT_NORDWR(0x008ab, 0x0024), + /* 0x008d0-0x008dd and 0x008e0-0x008e6 are unprotected on purpose for tools like perfetto */ + A6XX_PROTECT_NORDWR(0x008de, 0x0001), + A6XX_PROTECT_RDONLY(0x008e7, 0x014b), + A6XX_PROTECT_NORDWR(0x00900, 0x004d), + A6XX_PROTECT_NORDWR(0x0098d, 0x00b2), + A6XX_PROTECT_NORDWR(0x00a41, 0x01be), + A6XX_PROTECT_NORDWR(0x00df0, 0x0001), + A6XX_PROTECT_NORDWR(0x00e01, 0x0000), + A6XX_PROTECT_NORDWR(0x00e07, 0x0008), + A6XX_PROTECT_NORDWR(0x03c00, 0x00c3), + A6XX_PROTECT_RDONLY(0x03cc4, 0x1fff), + A6XX_PROTECT_NORDWR(0x08630, 0x01cf), + A6XX_PROTECT_NORDWR(0x08e00, 0x0000), + A6XX_PROTECT_NORDWR(0x08e08, 0x0000), + A6XX_PROTECT_NORDWR(0x08e50, 0x001f), + A6XX_PROTECT_NORDWR(0x08e80, 0x0280), + A6XX_PROTECT_NORDWR(0x09624, 0x01db), + A6XX_PROTECT_NORDWR(0x09e40, 0x0000), + A6XX_PROTECT_NORDWR(0x09e64, 0x000d), + A6XX_PROTECT_NORDWR(0x09e78, 0x0187), + A6XX_PROTECT_NORDWR(0x0a630, 0x01cf), + A6XX_PROTECT_NORDWR(0x0ae02, 0x0000), + A6XX_PROTECT_NORDWR(0x0ae50, 0x000f), + A6XX_PROTECT_NORDWR(0x0ae66, 0x0003), + A6XX_PROTECT_NORDWR(0x0ae6f, 0x0003), + A6XX_PROTECT_NORDWR(0x0b604, 0x0003), + A6XX_PROTECT_NORDWR(0x0ec00, 0x0fff), + A6XX_PROTECT_RDONLY(0x0fc00, 0x1fff), + A6XX_PROTECT_NORDWR(0x18400, 0x0053), + A6XX_PROTECT_RDONLY(0x18454, 0x0004), + A6XX_PROTECT_NORDWR(0x18459, 0x1fff), + A6XX_PROTECT_NORDWR(0x1a459, 0x1fff), + A6XX_PROTECT_NORDWR(0x1c459, 0x1fff), + A6XX_PROTECT_NORDWR(0x1f400, 0x0443), + A6XX_PROTECT_RDONLY(0x1f844, 0x007b), + A6XX_PROTECT_NORDWR(0x1f860, 0x0000), + A6XX_PROTECT_NORDWR(0x1f878, 0x002a), + /* CP_PROTECT_REG[45, 46] are left untouched! */ + 0, + 0, + A6XX_PROTECT_NORDWR(0x1f8c0, 0x00000), +}; +DECLARE_ADRENO_PROTECT(a730_protect, 48); + +static const uint32_t a7xx_pwrup_reglist_regs[] = { + REG_A6XX_UCHE_TRAP_BASE, + REG_A6XX_UCHE_TRAP_BASE + 1, + REG_A6XX_UCHE_WRITE_THRU_BASE, + REG_A6XX_UCHE_WRITE_THRU_BASE + 1, + REG_A6XX_UCHE_GMEM_RANGE_MIN, + REG_A6XX_UCHE_GMEM_RANGE_MIN + 1, + REG_A6XX_UCHE_GMEM_RANGE_MAX, + REG_A6XX_UCHE_GMEM_RANGE_MAX + 1, + REG_A6XX_UCHE_CACHE_WAYS, + REG_A6XX_UCHE_MODE_CNTL, + REG_A6XX_RB_NC_MODE_CNTL, + REG_A6XX_RB_CMP_DBG_ECO_CNTL, + REG_A6XX_RB_CONTEXT_SWITCH_GMEM_SAVE_RESTORE_ENABLE, + REG_A6XX_UCHE_GBIF_GX_CONFIG, + REG_A6XX_UCHE_CLIENT_PF, + REG_A6XX_TPL1_DBG_ECO_CNTL1, +}; + +DECLARE_ADRENO_REGLIST_LIST(a7xx_pwrup_reglist); + +/* Applicable for X185, A750 */ +static const u32 a750_ifpc_reglist_regs[] = { + REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(0), + REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(1), + REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(2), + REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(3), + REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(4), + REG_A6XX_RBBM_PERFCTR_CNTL, + REG_A6XX_TPL1_NC_MODE_CNTL, + REG_A6XX_SP_NC_MODE_CNTL, + REG_A6XX_CP_DBG_ECO_CNTL, + REG_A6XX_CP_PROTECT_CNTL, + REG_A6XX_CP_PROTECT(0), + REG_A6XX_CP_PROTECT(1), + REG_A6XX_CP_PROTECT(2), + REG_A6XX_CP_PROTECT(3), + REG_A6XX_CP_PROTECT(4), + REG_A6XX_CP_PROTECT(5), + REG_A6XX_CP_PROTECT(6), + REG_A6XX_CP_PROTECT(7), + REG_A6XX_CP_PROTECT(8), + REG_A6XX_CP_PROTECT(9), + REG_A6XX_CP_PROTECT(10), + REG_A6XX_CP_PROTECT(11), + REG_A6XX_CP_PROTECT(12), + REG_A6XX_CP_PROTECT(13), + REG_A6XX_CP_PROTECT(14), + REG_A6XX_CP_PROTECT(15), + REG_A6XX_CP_PROTECT(16), + REG_A6XX_CP_PROTECT(17), + REG_A6XX_CP_PROTECT(18), + REG_A6XX_CP_PROTECT(19), + REG_A6XX_CP_PROTECT(20), + REG_A6XX_CP_PROTECT(21), + REG_A6XX_CP_PROTECT(22), + REG_A6XX_CP_PROTECT(23), + REG_A6XX_CP_PROTECT(24), + REG_A6XX_CP_PROTECT(25), + REG_A6XX_CP_PROTECT(26), + REG_A6XX_CP_PROTECT(27), + REG_A6XX_CP_PROTECT(28), + REG_A6XX_CP_PROTECT(29), + REG_A6XX_CP_PROTECT(30), + REG_A6XX_CP_PROTECT(31), + REG_A6XX_CP_PROTECT(32), + REG_A6XX_CP_PROTECT(33), + REG_A6XX_CP_PROTECT(34), + REG_A6XX_CP_PROTECT(35), + REG_A6XX_CP_PROTECT(36), + REG_A6XX_CP_PROTECT(37), + REG_A6XX_CP_PROTECT(38), + REG_A6XX_CP_PROTECT(39), + REG_A6XX_CP_PROTECT(40), + REG_A6XX_CP_PROTECT(41), + REG_A6XX_CP_PROTECT(42), + REG_A6XX_CP_PROTECT(43), + REG_A6XX_CP_PROTECT(44), + REG_A6XX_CP_PROTECT(45), + REG_A6XX_CP_PROTECT(46), + REG_A6XX_CP_PROTECT(47), +}; + +DECLARE_ADRENO_REGLIST_LIST(a750_ifpc_reglist); + +static const struct adreno_reglist_pipe a7xx_dyn_pwrup_reglist_regs[] = { + { REG_A7XX_GRAS_NC_MODE_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, +}; + +DECLARE_ADRENO_REGLIST_PIPE_LIST(a7xx_dyn_pwrup_reglist); + +static const struct adreno_info a7xx_gpus[] = { + { + .chip_ids = ADRENO_CHIP_IDS(0x07000200, 0x07000400), + .family = ADRENO_6XX_GEN1, /* NOT a mistake! */ + .fw = { + [ADRENO_FW_SQE] = "a702_sqe.fw", + }, + .gmem = SZ_128K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_HW_APRIV, + .funcs = &a6xx_gmuwrapper_funcs, + .zapfw = "a702_zap.mbn", + .a6xx = &(const struct a6xx_info) { + .hwcg = a702_hwcg, + .protect = &a650_protect, + .gbif_cx = a640_gbif, + .gmu_cgc_mode = 0x00020202, + .prim_fifo_threshold = 0x0000c000, + }, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 236, 1 }, + { 178, 2 }, + { 142, 3 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS(0x07030001), + .family = ADRENO_7XX_GEN1, + .fw = { + [ADRENO_FW_SQE] = "a730_sqe.fw", + [ADRENO_FW_GMU] = "gmu_gen70000.bin", + }, + .gmem = SZ_2M, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV | + ADRENO_QUIRK_PREEMPTION, + .funcs = &a7xx_gpu_funcs, + .zapfw = "a730_zap.mdt", + .a6xx = &(const struct a6xx_info) { + .hwcg = a730_hwcg, + .protect = &a730_protect, + .pwrup_reglist = &a7xx_pwrup_reglist, + .dyn_pwrup_reglist = &a7xx_dyn_pwrup_reglist, + .gbif_cx = a640_gbif, + .gmu_cgc_mode = 0x00020000, + }, + .preempt_record_size = 2860 * SZ_1K, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x43020100), + .family = ADRENO_7XX_GEN1, + .fw = { + [ADRENO_FW_SQE] = "qcom/gen71700_sqe.fw", + [ADRENO_FW_GMU] = "qcom/gen71700_gmu.bin", + }, + .gmem = SZ_1M, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV | + ADRENO_QUIRK_PREEMPTION, + .funcs = &a7xx_gpu_funcs, + .a6xx = &(const struct a6xx_info) { + .hwcg = a730_hwcg, + .protect = &a730_protect, + .pwrup_reglist = &a7xx_pwrup_reglist, + .dyn_pwrup_reglist = &a7xx_dyn_pwrup_reglist, + .gbif_cx = a640_gbif, + .gmu_chipid = 0x07110000, + .gmu_cgc_mode = 0x00020000, + .bcms = (const struct a6xx_bcm[]) { + { .name = "SH0", .buswidth = 16 }, + { .name = "MC0", .buswidth = 4 }, + { + .name = "ACV", + .fixed = true, + .perfmode = BIT(3), + .perfmode_bw = 16500000, + }, + { /* sentinel */ }, + }, + }, + .preempt_record_size = 1536 * SZ_1K, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x43050a01), /* "C510v2" */ + .family = ADRENO_7XX_GEN2, + .fw = { + [ADRENO_FW_SQE] = "a740_sqe.fw", + [ADRENO_FW_GMU] = "gmu_gen70200.bin", + }, + .gmem = 3 * SZ_1M, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV | + ADRENO_QUIRK_PREEMPTION, + .funcs = &a7xx_gpu_funcs, + .zapfw = "a740_zap.mdt", + .a6xx = &(const struct a6xx_info) { + .hwcg = a740_hwcg, + .protect = &a730_protect, + .pwrup_reglist = &a7xx_pwrup_reglist, + .dyn_pwrup_reglist = &a7xx_dyn_pwrup_reglist, + .gbif_cx = a640_gbif, + .gmu_chipid = 0x7020100, + .gmu_cgc_mode = 0x00020202, + .bcms = (const struct a6xx_bcm[]) { + { .name = "SH0", .buswidth = 16 }, + { .name = "MC0", .buswidth = 4 }, + { + .name = "ACV", + .fixed = true, + .perfmode = BIT(3), + .perfmode_bw = 16500000, + }, + { /* sentinel */ }, + }, + }, + .preempt_record_size = 4192 * SZ_1K, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x43050c01), /* "C512v2" */ + .family = ADRENO_7XX_GEN2, + .fw = { + [ADRENO_FW_SQE] = "gen70500_sqe.fw", + [ADRENO_FW_GMU] = "gen70500_gmu.bin", + }, + .gmem = 3 * SZ_1M, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV | + ADRENO_QUIRK_PREEMPTION | + ADRENO_QUIRK_IFPC, + .funcs = &a7xx_gpu_funcs, + .a6xx = &(const struct a6xx_info) { + .hwcg = a740_hwcg, + .protect = &a730_protect, + .pwrup_reglist = &a7xx_pwrup_reglist, + .dyn_pwrup_reglist = &a7xx_dyn_pwrup_reglist, + .ifpc_reglist = &a750_ifpc_reglist, + .gbif_cx = a640_gbif, + .gmu_chipid = 0x7050001, + .gmu_cgc_mode = 0x00020202, + .bcms = (const struct a6xx_bcm[]) { + { .name = "SH0", .buswidth = 16 }, + { .name = "MC0", .buswidth = 4 }, + { + .name = "ACV", + .fixed = true, + .perfmode = BIT(3), + .perfmode_bw = 16500000, + }, + { /* sentinel */ }, + }, + }, + .preempt_record_size = 4192 * SZ_1K, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 59, 1 }, + { 7, 2 }, + { 232, 3 }, + { 146, 4 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS(0x43051401), /* "C520v2" */ + .family = ADRENO_7XX_GEN3, + .fw = { + [ADRENO_FW_SQE] = "gen70900_sqe.fw", + [ADRENO_FW_GMU] = "gmu_gen70900.bin", + }, + .gmem = 3 * SZ_1M, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV | + ADRENO_QUIRK_PREEMPTION | + ADRENO_QUIRK_IFPC, + .funcs = &a7xx_gpu_funcs, + .zapfw = "gen70900_zap.mbn", + .a6xx = &(const struct a6xx_info) { + .protect = &a730_protect, + .pwrup_reglist = &a7xx_pwrup_reglist, + .dyn_pwrup_reglist = &a7xx_dyn_pwrup_reglist, + .ifpc_reglist = &a750_ifpc_reglist, + .gbif_cx = a640_gbif, + .gmu_chipid = 0x7090100, + .gmu_cgc_mode = 0x00020202, + .bcms = (const struct a6xx_bcm[]) { + { .name = "SH0", .buswidth = 16 }, + { .name = "MC0", .buswidth = 4 }, + { + .name = "ACV", + .fixed = true, + .perfmode = BIT(2), + .perfmode_bw = 10687500, + }, + { /* sentinel */ }, + }, + }, + .preempt_record_size = 3572 * SZ_1K, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x43030c00), + .family = ADRENO_7XX_GEN2, + .fw = { + [ADRENO_FW_SQE] = "gen71500_sqe.fw", + [ADRENO_FW_GMU] = "gen71500_gmu.bin", + }, + .gmem = SZ_1M + SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV | + ADRENO_QUIRK_PREEMPTION, + .funcs = &a7xx_gpu_funcs, + .a6xx = &(const struct a6xx_info) { + .hwcg = a740_hwcg, + .protect = &a730_protect, + .pwrup_reglist = &a7xx_pwrup_reglist, + .dyn_pwrup_reglist = &a7xx_dyn_pwrup_reglist, + .gbif_cx = a640_gbif, + .gmu_chipid = 0x70f0000, + .gmu_cgc_mode = 0x00020222, + .bcms = (const struct a6xx_bcm[]) { + { .name = "SH0", .buswidth = 16 }, + { .name = "MC0", .buswidth = 4 }, + { + .name = "ACV", + .fixed = true, + .perfmode = BIT(3), + .perfmode_bw = 16500000, + }, + { /* sentinel */ }, + }, + }, + .preempt_record_size = 4192 * SZ_1K, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 294, 1 }, + { 263, 2 }, + { 233, 3 }, + { 141, 4 }, + ), + } +}; +DECLARE_ADRENO_GPULIST(a7xx); + +static const struct adreno_reglist_pipe x285_nonctxt_regs[] = { + { REG_A8XX_CP_SMMU_STREAM_ID_LPAC, 0x00000101, BIT(PIPE_NONE) }, + { REG_A8XX_GRAS_DBG_ECO_CNTL, 0x00000800, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_GRAS_TSEFE_DBG_ECO_CNTL, 0x00200000, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A6XX_PC_AUTO_VERTEX_STRIDE, 0x00000001, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_VIS_STREAM_CNTL, 0x10010000, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CONTEXT_SWITCH_STABILIZE_CNTL_1, 0x00000002, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_1, 0x00000003, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_2, 0x00000200, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_3, 0x00500000, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_4, 0x00500050, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_RB_GC_GMEM_PROTECT, 0x15000000, BIT(PIPE_BR) }, + { REG_A8XX_RB_RESOLVE_PREFETCH_CNTL, 0x00000007, BIT(PIPE_BR) }, + { REG_A8XX_RB_CMP_DBG_ECO_CNTL, 0x00004000, BIT(PIPE_BR) }, + { REG_A8XX_RBBM_NC_MODE_CNTL, 0x00000001, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_SLICE_NC_MODE_CNTL, 0x00000001, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_WAIT_IDLE_CLOCKS_CNTL, 0x00000030, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_WAIT_IDLE_CLOCKS_CNTL2, 0x00000030, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_INTERFACE_HANG_INT_CNTL, 0x0fffffff, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_GBIF_CLIENT_QOS_CNTL, 0x22122212, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_CGC_P2S_CNTL, 0x00000040, BIT(PIPE_NONE) }, + { REG_A7XX_SP_CHICKEN_BITS_2, 0x00820800, BIT(PIPE_NONE) }, + { REG_A7XX_SP_CHICKEN_BITS_3, 0x00300000, BIT(PIPE_NONE) }, + { REG_A6XX_SP_PERFCTR_SHADER_MASK, 0x0000003f, BIT(PIPE_NONE) }, + /* Disable CS dead batch merge */ + { REG_A7XX_SP_HLSQ_DBG_ECO_CNTL_2, BIT(31), BIT(PIPE_NONE) }, + { REG_A7XX_SP_HLSQ_TIMEOUT_THRESHOLD_DP, 0x00000080, BIT(PIPE_NONE) }, + { REG_A7XX_SP_READ_SEL, 0x0001ff00, BIT(PIPE_NONE) }, + { REG_A6XX_TPL1_DBG_ECO_CNTL, 0x10000000, BIT(PIPE_NONE) }, + /* BIT(26): Disable final clamp for bicubic filtering */ + { REG_A6XX_TPL1_DBG_ECO_CNTL1, 0x04000720, BIT(PIPE_NONE) }, + { REG_A6XX_UCHE_MODE_CNTL, 0x80080000, BIT(PIPE_NONE) }, + { REG_A8XX_UCHE_CCHE_MODE_CNTL, 0x00001000, BIT(PIPE_NONE) }, + { REG_A8XX_UCHE_CCHE_CACHE_WAYS, 0x00000800, BIT(PIPE_NONE) }, + { REG_A8XX_UCHE_GBIF_GX_CONFIG, 0x010240e0, BIT(PIPE_NONE) }, + { REG_A8XX_UCHE_VARB_IDLE_TIMEOUT, 0x00000020, BIT(PIPE_NONE) }, + { REG_A7XX_VFD_DBG_ECO_CNTL, 0x00008000, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BV_THRESHOLD, 0x00500050, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BR_THRESHOLD, 0x00600060, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BUSY_REQ_CNT, 0x00200020, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_LP_REQ_CNT, 0x00000020, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VPC_FLATSHADE_MODE_CNTL, 0x00000001, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { }, +}; + +static const u32 x285_protect_regs[] = { + A6XX_PROTECT_RDONLY(0x00008, 0x039b), + A6XX_PROTECT_RDONLY(0x003b4, 0x008b), + A6XX_PROTECT_NORDWR(0x00440, 0x001f), + A6XX_PROTECT_RDONLY(0x00580, 0x005f), + A6XX_PROTECT_NORDWR(0x005e0, 0x011f), + A6XX_PROTECT_RDONLY(0x0074a, 0x0005), + A6XX_PROTECT_RDONLY(0x00759, 0x0026), + A6XX_PROTECT_RDONLY(0x00789, 0x0000), + A6XX_PROTECT_RDONLY(0x0078c, 0x0013), + A6XX_PROTECT_NORDWR(0x00800, 0x0029), + A6XX_PROTECT_NORDWR(0x0082c, 0x0000), + A6XX_PROTECT_NORDWR(0x00837, 0x00af), + A6XX_PROTECT_RDONLY(0x008e7, 0x00c9), + A6XX_PROTECT_NORDWR(0x008ec, 0x00c3), + A6XX_PROTECT_NORDWR(0x009b1, 0x0250), + A6XX_PROTECT_RDONLY(0x00ce0, 0x0001), + A6XX_PROTECT_RDONLY(0x00df0, 0x0000), + A6XX_PROTECT_NORDWR(0x00df1, 0x0000), + A6XX_PROTECT_NORDWR(0x00e01, 0x0000), + A6XX_PROTECT_NORDWR(0x00e03, 0x1fff), + A6XX_PROTECT_NORDWR(0x03c00, 0x00c5), + A6XX_PROTECT_RDONLY(0x03cc6, 0x0039), + A6XX_PROTECT_NORDWR(0x03d00, 0x1fff), + A6XX_PROTECT_NORDWR(0x08600, 0x01ff), + A6XX_PROTECT_NORDWR(0x08e00, 0x00ff), + A6XX_PROTECT_RDONLY(0x08f00, 0x0000), + A6XX_PROTECT_NORDWR(0x08f01, 0x01be), + A6XX_PROTECT_NORDWR(0x09600, 0x01ff), + A6XX_PROTECT_RDONLY(0x0981a, 0x02e5), + A6XX_PROTECT_NORDWR(0x09e00, 0x01ff), + A6XX_PROTECT_NORDWR(0x0a600, 0x01ff), + A6XX_PROTECT_NORDWR(0x0a82e, 0x0000), + A6XX_PROTECT_NORDWR(0x0ae00, 0x0006), + A6XX_PROTECT_NORDWR(0x0ae08, 0x0006), + A6XX_PROTECT_NORDWR(0x0ae10, 0x00bf), + A6XX_PROTECT_RDONLY(0x0aed0, 0x002f), + A6XX_PROTECT_NORDWR(0x0af00, 0x027f), + A6XX_PROTECT_NORDWR(0x0b600, 0x1fff), + A6XX_PROTECT_NORDWR(0x0dc00, 0x1fff), + A6XX_PROTECT_RDONLY(0x0fc00, 0x1fff), + A6XX_PROTECT_NORDWR(0x18400, 0x003f), + A6XX_PROTECT_RDONLY(0x18440, 0x013f), + A6XX_PROTECT_NORDWR(0x18580, 0x1fff), + A6XX_PROTECT_NORDWR(0x1b400, 0x1fff), + A6XX_PROTECT_NORDWR(0x1f400, 0x0477), + A6XX_PROTECT_RDONLY(0x1f878, 0x0507), + A6XX_PROTECT_NORDWR(0x1f930, 0x0329), + A6XX_PROTECT_NORDWR(0x1fd80, 0x1fff), + A6XX_PROTECT_NORDWR(0x27800, 0x007f), + A6XX_PROTECT_RDONLY(0x27880, 0x0385), + A6XX_PROTECT_NORDWR(0x27882, 0x000a), + A6XX_PROTECT_NORDWR(0x27c06, 0x0000), +}; + +DECLARE_ADRENO_PROTECT(x285_protect, 15); + +static const struct adreno_reglist_pipe x285_dyn_pwrup_reglist_regs[] = { + { REG_A8XX_GRAS_TSEFE_DBG_ECO_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_GRAS_NC_MODE_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_GRAS_DBG_ECO_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A6XX_PC_AUTO_VERTEX_STRIDE, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_1, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_2, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_3, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_4, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CONTEXT_SWITCH_STABILIZE_CNTL_1, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_VIS_STREAM_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A7XX_RB_CCU_CNTL, 0, BIT(PIPE_BR) }, + { REG_A7XX_RB_CCU_DBG_ECO_CNTL, 0, BIT(PIPE_BR)}, + { REG_A8XX_RB_CCU_NC_MODE_CNTL, 0, BIT(PIPE_BR) }, + { REG_A8XX_RB_CMP_NC_MODE_CNTL, 0, BIT(PIPE_BR) }, + { REG_A6XX_RB_RBP_CNTL, 0, BIT(PIPE_BR) }, + { REG_A8XX_RB_RESOLVE_PREFETCH_CNTL, 0, BIT(PIPE_BR) }, + { REG_A8XX_RB_CMP_DBG_ECO_CNTL, 0, BIT(PIPE_BR) }, + { REG_A7XX_VFD_DBG_ECO_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BV_THRESHOLD, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BR_THRESHOLD, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BUSY_REQ_CNT, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_LP_REQ_CNT, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VPC_FLATSHADE_MODE_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_CP_HW_FAULT_STATUS_MASK_PIPE, 0, BIT(PIPE_BR) | + BIT(PIPE_BV) | BIT(PIPE_LPAC) | BIT(PIPE_AQE0) | + BIT(PIPE_AQE1) | BIT(PIPE_DDE_BR) | BIT(PIPE_DDE_BV) }, + { REG_A8XX_CP_INTERRUPT_STATUS_MASK_PIPE, 0, BIT(PIPE_BR) | + BIT(PIPE_BV) | BIT(PIPE_LPAC) | BIT(PIPE_AQE0) | + BIT(PIPE_AQE1) | BIT(PIPE_DDE_BR) | BIT(PIPE_DDE_BV) }, + { REG_A8XX_CP_PROTECT_CNTL_PIPE, 0, BIT(PIPE_BR) | BIT(PIPE_BV) | BIT(PIPE_LPAC)}, + { REG_A8XX_CP_PROTECT_PIPE(15), 0, BIT(PIPE_BR) | BIT(PIPE_BV) | BIT(PIPE_LPAC) }, + { REG_A8XX_RB_GC_GMEM_PROTECT, 0, BIT(PIPE_BR) }, + { REG_A8XX_RB_LPAC_GMEM_PROTECT, 0, BIT(PIPE_BR) }, + { REG_A6XX_RB_CONTEXT_SWITCH_GMEM_SAVE_RESTORE_ENABLE, 0, BIT(PIPE_BR) }, +}; +DECLARE_ADRENO_REGLIST_PIPE_LIST(x285_dyn_pwrup_reglist); + +static const struct adreno_reglist_pipe a810_nonctxt_regs[] = { + { REG_A8XX_CP_SMMU_STREAM_ID_LPAC, 0x00000101, BIT(PIPE_NONE) }, + { REG_A8XX_GRAS_DBG_ECO_CNTL, 0x00f80800, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A6XX_PC_AUTO_VERTEX_STRIDE, 0x00000001, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_VIS_STREAM_CNTL, 0x10010000, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CONTEXT_SWITCH_STABILIZE_CNTL_1, 0x00000002, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_1, 0x00000003, BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_1, 0x00000023, BIT(PIPE_BV) }, /* Avoid partial waves at VFD */ + { REG_A8XX_PC_CHICKEN_BITS_2, 0x00000200, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_3, 0x00500000, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_4, 0x00500050, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A7XX_RB_CCU_CNTL, 0x00000068, BIT(PIPE_BR) }, + /* Partially enable perf clear, Disable DINT to c/z be data forwarding */ + { REG_A7XX_RB_CCU_DBG_ECO_CNTL, 0x00002200, BIT(PIPE_BR) }, + { REG_A8XX_RB_RESOLVE_PREFETCH_CNTL, 0x00000007, BIT(PIPE_BR) }, + { REG_A8XX_RB_CMP_DBG_ECO_CNTL, 0x00004000, BIT(PIPE_BR) }, + { REG_A8XX_RBBM_NC_MODE_CNTL, 0x00000001, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_SLICE_NC_MODE_CNTL, 0x00000001, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_WAIT_IDLE_CLOCKS_CNTL, 0x00000030, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_WAIT_IDLE_CLOCKS_CNTL2, 0x00000030, BIT(PIPE_NONE) }, + { REG_A8XX_UCHE_GBIF_GX_CONFIG, 0x010240e0, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_GBIF_CLIENT_QOS_CNTL, 0x22122212, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_CGC_P2S_CNTL, 0x00000040, BIT(PIPE_NONE) }, + /* + * BIT(22): Disable PS out of order retire + * BIT(23): Enable half wave mode and MM instruction src&dst is half precision + */ + { REG_A7XX_SP_CHICKEN_BITS_2, BIT(22) | BIT(23), BIT(PIPE_NONE) }, + { REG_A7XX_SP_CHICKEN_BITS_3, 0x00300000, BIT(PIPE_NONE) }, + { REG_A6XX_SP_PERFCTR_SHADER_MASK, 0x0000003f, BIT(PIPE_NONE) }, + { REG_A7XX_SP_HLSQ_TIMEOUT_THRESHOLD_DP, 0x00000080, BIT(PIPE_NONE) }, + { REG_A6XX_TPL1_DBG_ECO_CNTL, 0x10100000, BIT(PIPE_NONE) }, + { REG_A6XX_TPL1_DBG_ECO_CNTL1, 0x04000724, BIT(PIPE_NONE) }, + { REG_A6XX_UCHE_MODE_CNTL, 0x00020000, BIT(PIPE_NONE) }, + { REG_A8XX_UCHE_CCHE_MODE_CNTL, 0x00001000, BIT(PIPE_NONE) }, + { REG_A8XX_UCHE_CCHE_CACHE_WAYS, 0x00000800, BIT(PIPE_NONE) }, + /* Disable write slow pointer in data phase queue */ + { REG_A8XX_UCHE_HW_DBG_CNTL, BIT(8), BIT(PIPE_NONE) }, + { REG_A8XX_UCHE_VARB_IDLE_TIMEOUT, 0x00000020, BIT(PIPE_NONE) }, + { REG_A7XX_VFD_DBG_ECO_CNTL, 0x00008000, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BV_THRESHOLD, 0x00500050, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BR_THRESHOLD, 0x00600060, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BUSY_REQ_CNT, 0x00200020, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_LP_REQ_CNT, 0x00100020, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VPC_FLATSHADE_MODE_CNTL, 0x00000001, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_RB_GC_GMEM_PROTECT, 0x00900000, BIT(PIPE_BR) }, + { }, +}; + +static const u32 a810_protect_regs[] = { + A6XX_PROTECT_RDONLY(0x00000, 0x03a3), + A6XX_PROTECT_RDONLY(0x003b4, 0x008b), + A6XX_PROTECT_NORDWR(0x00440, 0x001f), + A6XX_PROTECT_RDONLY(0x00580, 0x005f), + A6XX_PROTECT_NORDWR(0x005e0, 0x011f), + A6XX_PROTECT_RDONLY(0x0074a, 0x0005), + A6XX_PROTECT_RDONLY(0x00759, 0x0026), + A6XX_PROTECT_RDONLY(0x00789, 0x0000), + A6XX_PROTECT_RDONLY(0x0078c, 0x0013), + A6XX_PROTECT_NORDWR(0x00800, 0x0029), + A6XX_PROTECT_NORDWR(0x00837, 0x00af), + A6XX_PROTECT_RDONLY(0x008e7, 0x00c9), + A6XX_PROTECT_NORDWR(0x008ec, 0x00c3), + A6XX_PROTECT_NORDWR(0x009b1, 0x0250), + A6XX_PROTECT_RDONLY(0x00ce0, 0x0001), + A6XX_PROTECT_RDONLY(0x00df0, 0x0000), + A6XX_PROTECT_NORDWR(0x00df1, 0x0000), + A6XX_PROTECT_NORDWR(0x00e01, 0x0000), + A6XX_PROTECT_NORDWR(0x00e03, 0x1fff), + A6XX_PROTECT_NORDWR(0x03c00, 0x00c5), + A6XX_PROTECT_RDONLY(0x03cc6, 0x1fff), + A6XX_PROTECT_NORDWR(0x08600, 0x01ff), + A6XX_PROTECT_NORDWR(0x08e00, 0x00ff), + A6XX_PROTECT_RDONLY(0x08f00, 0x0000), + A6XX_PROTECT_NORDWR(0x08f01, 0x01be), + A6XX_PROTECT_NORDWR(0x09600, 0x01ff), + A6XX_PROTECT_RDONLY(0x0981a, 0x02e5), + A6XX_PROTECT_NORDWR(0x09e00, 0x01ff), + A6XX_PROTECT_NORDWR(0x0a600, 0x01ff), + A6XX_PROTECT_NORDWR(0x0ae00, 0x0006), + A6XX_PROTECT_NORDWR(0x0ae08, 0x0006), + A6XX_PROTECT_NORDWR(0x0ae10, 0x036f), + A6XX_PROTECT_NORDWR(0x0b600, 0x1fff), + A6XX_PROTECT_NORDWR(0x0dc00, 0x1fff), + A6XX_PROTECT_RDONLY(0x0fc00, 0x1fff), + A6XX_PROTECT_NORDWR(0x18400, 0x003f), + A6XX_PROTECT_RDONLY(0x18440, 0x013f), + A6XX_PROTECT_NORDWR(0x18580, 0x1fff), + A6XX_PROTECT_NORDWR(0x1b400, 0x1fff), + A6XX_PROTECT_NORDWR(0x1f400, 0x0477), + A6XX_PROTECT_RDONLY(0x1f878, 0x0787), + A6XX_PROTECT_NORDWR(0x1f930, 0x0329), + A6XX_PROTECT_NORDWR(0x20000, 0x1fff), + A6XX_PROTECT_NORDWR(0x27800, 0x007f), + A6XX_PROTECT_RDONLY(0x27880, 0x0381), + A6XX_PROTECT_NORDWR(0x27882, 0x0001), + A6XX_PROTECT_NORDWR(0x27c02, 0x0000), +}; +DECLARE_ADRENO_PROTECT(a810_protect, 64); + +static const uint32_t a810_pwrup_reglist_regs[] = { + REG_A6XX_UCHE_MODE_CNTL, + REG_A8XX_UCHE_VARB_IDLE_TIMEOUT, + REG_A8XX_UCHE_GBIF_GX_CONFIG, + REG_A8XX_UCHE_CACHE_WAYS, + REG_A8XX_UCHE_CCHE_MODE_CNTL, + REG_A8XX_UCHE_CCHE_CACHE_WAYS, + REG_A8XX_UCHE_CCHE_GC_GMEM_RANGE_MIN, + REG_A8XX_UCHE_CCHE_GC_GMEM_RANGE_MIN + 1, + REG_A8XX_UCHE_CCHE_TRAP_BASE, + REG_A8XX_UCHE_CCHE_TRAP_BASE + 1, + REG_A8XX_UCHE_CCHE_WRITE_THRU_BASE, + REG_A8XX_UCHE_CCHE_WRITE_THRU_BASE + 1, + REG_A8XX_UCHE_HW_DBG_CNTL, + REG_A8XX_UCHE_WRITE_THRU_BASE, + REG_A8XX_UCHE_WRITE_THRU_BASE + 1, + REG_A8XX_UCHE_TRAP_BASE, + REG_A8XX_UCHE_TRAP_BASE + 1, + REG_A8XX_UCHE_CLIENT_PF, + REG_A8XX_RB_CMP_NC_MODE_CNTL, + REG_A7XX_SP_HLSQ_TIMEOUT_THRESHOLD_DP, + REG_A8XX_SP_HLSQ_GC_GMEM_RANGE_MIN, + REG_A8XX_SP_HLSQ_GC_GMEM_RANGE_MIN + 1, + REG_A7XX_SP_READ_SEL, + REG_A6XX_TPL1_NC_MODE_CNTL, + REG_A6XX_TPL1_DBG_ECO_CNTL, + REG_A6XX_TPL1_DBG_ECO_CNTL1, + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(1), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(2), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(3), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(4), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(5), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(6), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(7), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(8), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(9), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(10), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(11), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(12), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(13), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(14), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(15), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(16), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(17), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(18), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(19), +}; +DECLARE_ADRENO_REGLIST_LIST(a810_pwrup_reglist); + +static const u32 a810_ifpc_reglist_regs[] = { + REG_A8XX_RBBM_NC_MODE_CNTL, + REG_A8XX_RBBM_PERFCTR_CNTL, + REG_A8XX_RBBM_SLICE_INTERFACE_HANG_INT_CNTL, + REG_A8XX_RBBM_SLICE_NC_MODE_CNTL, + REG_A6XX_SP_NC_MODE_CNTL, + REG_A7XX_SP_CHICKEN_BITS_2, + REG_A7XX_SP_CHICKEN_BITS_3, + REG_A6XX_SP_PERFCTR_SHADER_MASK, + REG_A8XX_CP_PROTECT_GLOBAL(0), + REG_A8XX_CP_PROTECT_GLOBAL(1), + REG_A8XX_CP_PROTECT_GLOBAL(2), + REG_A8XX_CP_PROTECT_GLOBAL(3), + REG_A8XX_CP_PROTECT_GLOBAL(4), + REG_A8XX_CP_PROTECT_GLOBAL(5), + REG_A8XX_CP_PROTECT_GLOBAL(6), + REG_A8XX_CP_PROTECT_GLOBAL(7), + REG_A8XX_CP_PROTECT_GLOBAL(8), + REG_A8XX_CP_PROTECT_GLOBAL(9), + REG_A8XX_CP_PROTECT_GLOBAL(10), + REG_A8XX_CP_PROTECT_GLOBAL(11), + REG_A8XX_CP_PROTECT_GLOBAL(12), + REG_A8XX_CP_PROTECT_GLOBAL(13), + REG_A8XX_CP_PROTECT_GLOBAL(14), + REG_A8XX_CP_PROTECT_GLOBAL(15), + REG_A8XX_CP_PROTECT_GLOBAL(16), + REG_A8XX_CP_PROTECT_GLOBAL(17), + REG_A8XX_CP_PROTECT_GLOBAL(18), + REG_A8XX_CP_PROTECT_GLOBAL(19), + REG_A8XX_CP_PROTECT_GLOBAL(20), + REG_A8XX_CP_PROTECT_GLOBAL(21), + REG_A8XX_CP_PROTECT_GLOBAL(22), + REG_A8XX_CP_PROTECT_GLOBAL(23), + REG_A8XX_CP_PROTECT_GLOBAL(24), + REG_A8XX_CP_PROTECT_GLOBAL(25), + REG_A8XX_CP_PROTECT_GLOBAL(26), + REG_A8XX_CP_PROTECT_GLOBAL(27), + REG_A8XX_CP_PROTECT_GLOBAL(28), + REG_A8XX_CP_PROTECT_GLOBAL(29), + REG_A8XX_CP_PROTECT_GLOBAL(30), + REG_A8XX_CP_PROTECT_GLOBAL(31), + REG_A8XX_CP_PROTECT_GLOBAL(32), + REG_A8XX_CP_PROTECT_GLOBAL(33), + REG_A8XX_CP_PROTECT_GLOBAL(34), + REG_A8XX_CP_PROTECT_GLOBAL(35), + REG_A8XX_CP_PROTECT_GLOBAL(36), + REG_A8XX_CP_PROTECT_GLOBAL(37), + REG_A8XX_CP_PROTECT_GLOBAL(38), + REG_A8XX_CP_PROTECT_GLOBAL(39), + REG_A8XX_CP_PROTECT_GLOBAL(40), + REG_A8XX_CP_PROTECT_GLOBAL(41), + REG_A8XX_CP_PROTECT_GLOBAL(42), + REG_A8XX_CP_PROTECT_GLOBAL(43), + REG_A8XX_CP_PROTECT_GLOBAL(44), + REG_A8XX_CP_PROTECT_GLOBAL(45), + REG_A8XX_CP_PROTECT_GLOBAL(46), + REG_A8XX_CP_PROTECT_GLOBAL(47), + REG_A8XX_CP_PROTECT_GLOBAL(48), + REG_A8XX_CP_PROTECT_GLOBAL(49), + REG_A8XX_CP_PROTECT_GLOBAL(50), + REG_A8XX_CP_PROTECT_GLOBAL(51), + REG_A8XX_CP_PROTECT_GLOBAL(52), + REG_A8XX_CP_PROTECT_GLOBAL(53), + REG_A8XX_CP_PROTECT_GLOBAL(54), + REG_A8XX_CP_PROTECT_GLOBAL(55), + REG_A8XX_CP_PROTECT_GLOBAL(56), + REG_A8XX_CP_PROTECT_GLOBAL(57), + REG_A8XX_CP_PROTECT_GLOBAL(58), + REG_A8XX_CP_PROTECT_GLOBAL(59), + REG_A8XX_CP_PROTECT_GLOBAL(60), + REG_A8XX_CP_PROTECT_GLOBAL(61), + REG_A8XX_CP_PROTECT_GLOBAL(62), + REG_A8XX_CP_PROTECT_GLOBAL(63), +}; +DECLARE_ADRENO_REGLIST_LIST(a810_ifpc_reglist); + +static const struct adreno_reglist_pipe a810_dyn_pwrup_reglist_regs[] = { + { REG_A8XX_CP_PROTECT_CNTL_PIPE, 0, BIT(PIPE_BR) | BIT(PIPE_BV) }, + { REG_A8XX_CP_PROTECT_PIPE(15), 0, BIT(PIPE_BR) | BIT(PIPE_BV) }, + { REG_A8XX_GRAS_TSEFE_DBG_ECO_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_GRAS_NC_MODE_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_GRAS_DBG_ECO_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A7XX_RB_CCU_CNTL, 0, BIT(PIPE_BR) }, + { REG_A7XX_RB_CCU_DBG_ECO_CNTL, 0, BIT(PIPE_BR) }, + { REG_A8XX_RB_CCU_NC_MODE_CNTL, 0, BIT(PIPE_BR) }, + { REG_A8XX_RB_CMP_NC_MODE_CNTL, 0, BIT(PIPE_BR) }, + { REG_A8XX_RB_RESOLVE_PREFETCH_CNTL, 0, BIT(PIPE_BR) }, + { REG_A8XX_RB_CMP_DBG_ECO_CNTL, 0, BIT(PIPE_BR) }, + { REG_A8XX_RB_GC_GMEM_PROTECT, 0, BIT(PIPE_BR) }, + { REG_A6XX_RB_CONTEXT_SWITCH_GMEM_SAVE_RESTORE_ENABLE, 0, BIT(PIPE_BR) }, + { REG_A8XX_VPC_FLATSHADE_MODE_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_1, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_2, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_3, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_4, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A6XX_PC_AUTO_VERTEX_STRIDE, 0, BIT(PIPE_BR) | BIT(PIPE_BV) }, + { REG_A8XX_PC_VIS_STREAM_CNTL, 0, BIT(PIPE_BR) | BIT(PIPE_BV) }, + { REG_A8XX_PC_CONTEXT_SWITCH_STABILIZE_CNTL_1, 0, BIT(PIPE_BR) | BIT(PIPE_BV) }, + { REG_A8XX_VFD_CB_BV_THRESHOLD, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BR_THRESHOLD, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BUSY_REQ_CNT, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_LP_REQ_CNT, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A7XX_VFD_DBG_ECO_CNTL, 0, BIT(PIPE_BR) | BIT(PIPE_BV) }, +}; +DECLARE_ADRENO_REGLIST_PIPE_LIST(a810_dyn_pwrup_reglist); + +static const struct adreno_reglist_pipe a840_nonctxt_regs[] = { + { REG_A8XX_CP_SMMU_STREAM_ID_LPAC, 0x00000101, BIT(PIPE_NONE) }, + { REG_A8XX_GRAS_DBG_ECO_CNTL, 0x00000800, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_GRAS_TSEFE_DBG_ECO_CNTL, 0x00200000, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A6XX_PC_AUTO_VERTEX_STRIDE, 0x00000001, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_VIS_STREAM_CNTL, 0x10010000, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CONTEXT_SWITCH_STABILIZE_CNTL_1, 0x00000002, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_1, 0x00000003, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_2, 0x00000200, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_3, 0x00500000, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_4, 0x00500050, BIT(PIPE_BV) | BIT(PIPE_BR) }, + /* Disable Dead Draw Merge scheme on RB-HLSQ */ + { REG_A6XX_RB_RBP_CNTL, BIT(5), BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A7XX_RB_CCU_CNTL, 0x00000068, BIT(PIPE_BR) }, + /* Partially enable perf clear, Disable DINT to c/z be data forwarding */ + { REG_A7XX_RB_CCU_DBG_ECO_CNTL, 0x00002200, BIT(PIPE_BR) }, + { REG_A8XX_RB_GC_GMEM_PROTECT, 0x12000000, BIT(PIPE_BR) }, + { REG_A8XX_RB_RESOLVE_PREFETCH_CNTL, 0x00000007, BIT(PIPE_BR) }, + { REG_A8XX_RB_CMP_DBG_ECO_CNTL, 0x00004000, BIT(PIPE_BR) }, + { REG_A8XX_RBBM_NC_MODE_CNTL, 0x00000001, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_SLICE_NC_MODE_CNTL, 0x00000001, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_POWER_UP_RESET_SW_OVERRIDE, 0x70809060, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_POWER_UP_RESET_SW_BV_OVERRIDE, 0x30000000, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_WAIT_IDLE_CLOCKS_CNTL, 0x00000030, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_WAIT_IDLE_CLOCKS_CNTL2, 0x00000030, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_INTERFACE_HANG_INT_CNTL, 0x0fffffff, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_GBIF_CLIENT_QOS_CNTL, 0x22122212, BIT(PIPE_NONE) }, + { REG_A8XX_RBBM_CGC_P2S_CNTL, 0x00000040, BIT(PIPE_NONE) }, + /* Disable mode_switch optimization in UMAS */ + { REG_A6XX_SP_CHICKEN_BITS, BIT(24) | BIT(26), BIT(PIPE_NONE) }, + /* Disable LPAC large-LM mode */ + { REG_A8XX_SP_SS_CHICKEN_BITS_0, BIT(3), BIT(PIPE_NONE) }, + /* Disable PS out of order retire */ + { REG_A7XX_SP_CHICKEN_BITS_2, 0x00c21800, BIT(PIPE_NONE) }, + { REG_A7XX_SP_CHICKEN_BITS_3, 0x00300000, BIT(PIPE_NONE) }, + /* Disable SP2TP info attribute */ + { REG_A8XX_SP_CHICKEN_BITS_4, 0x00000002, BIT(PIPE_NONE) }, + { REG_A6XX_SP_PERFCTR_SHADER_MASK, 0x0000003f, BIT(PIPE_NONE) }, + { REG_A7XX_SP_HLSQ_DBG_ECO_CNTL, BIT(14), BIT(PIPE_NONE) }, + /* Ignore HLSQ shared constant feedback from SP */ + { REG_A7XX_SP_HLSQ_DBG_ECO_CNTL_1, BIT(17), BIT(PIPE_NONE) }, + /* Disable CS dead batch merge */ + { REG_A7XX_SP_HLSQ_DBG_ECO_CNTL_2, BIT(24), BIT(PIPE_NONE) }, + { REG_A8XX_SP_HLSQ_DBG_ECO_CNTL_3, BIT(7), BIT(PIPE_NONE) }, + { REG_A7XX_SP_HLSQ_TIMEOUT_THRESHOLD_DP, 0x00000080, BIT(PIPE_NONE) }, + { REG_A7XX_SP_READ_SEL, 0x0001ff00, BIT(PIPE_NONE) }, + { REG_A6XX_TPL1_DBG_ECO_CNTL, 0x10100000, BIT(PIPE_NONE) }, + /* BIT(26): Disable final clamp for bicubic filtering */ + { REG_A6XX_TPL1_DBG_ECO_CNTL1, 0x04000720, BIT(PIPE_NONE) }, + { REG_A6XX_UCHE_MODE_CNTL, 0x80080000, BIT(PIPE_NONE) }, + { REG_A8XX_UCHE_CCHE_MODE_CNTL, 0x00001000, BIT(PIPE_NONE) }, + { REG_A8XX_UCHE_CCHE_CACHE_WAYS, 0x00000800, BIT(PIPE_NONE) }, + { REG_A8XX_UCHE_GBIF_GX_CONFIG, 0x010240e0, BIT(PIPE_NONE) }, + { REG_A8XX_UCHE_VARB_IDLE_TIMEOUT, 0x00000020, BIT(PIPE_NONE) }, + { REG_A7XX_VFD_DBG_ECO_CNTL, 0x00008000, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BV_THRESHOLD, 0x00500050, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BR_THRESHOLD, 0x00600060, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BUSY_REQ_CNT, 0x00200020, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_LP_REQ_CNT, 0x00000020, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VPC_FLATSHADE_MODE_CNTL, 0x00000001, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { }, +}; + +static const u32 a840_protect_regs[] = { + A6XX_PROTECT_RDONLY(0x00008, 0x039b), + A6XX_PROTECT_RDONLY(0x003b4, 0x008b), + A6XX_PROTECT_NORDWR(0x00440, 0x001f), + A6XX_PROTECT_RDONLY(0x00580, 0x005f), + A6XX_PROTECT_NORDWR(0x005e0, 0x011f), + A6XX_PROTECT_RDONLY(0x0074a, 0x0005), + A6XX_PROTECT_RDONLY(0x00759, 0x001b), + A6XX_PROTECT_NORDWR(0x00775, 0x000a), + A6XX_PROTECT_RDONLY(0x00789, 0x0000), + A6XX_PROTECT_RDONLY(0x0078c, 0x0013), + A6XX_PROTECT_NORDWR(0x00800, 0x0029), + A6XX_PROTECT_NORDWR(0x00837, 0x00af), + A6XX_PROTECT_RDONLY(0x008e7, 0x00c9), + A6XX_PROTECT_NORDWR(0x008ec, 0x00c3), + A6XX_PROTECT_NORDWR(0x009b1, 0x0250), + A6XX_PROTECT_NORDWR(0x00c07, 0x0008), + A6XX_PROTECT_RDONLY(0x00ce0, 0x0001), + A6XX_PROTECT_RDONLY(0x00df0, 0x0000), + A6XX_PROTECT_NORDWR(0x00df1, 0x0000), + A6XX_PROTECT_NORDWR(0x00e01, 0x0000), + A6XX_PROTECT_NORDWR(0x00e03, 0x1fff), + A6XX_PROTECT_NORDWR(0x03c00, 0x00c5), + A6XX_PROTECT_RDONLY(0x03cc6, 0x0039), + A6XX_PROTECT_NORDWR(0x03d00, 0x1fff), + A6XX_PROTECT_NORDWR(0x08600, 0x01ff), + A6XX_PROTECT_NORDWR(0x08e00, 0x00ff), + A6XX_PROTECT_RDONLY(0x08f00, 0x0000), + A6XX_PROTECT_NORDWR(0x08f01, 0x01be), + A6XX_PROTECT_NORDWR(0x09600, 0x01ff), + A6XX_PROTECT_RDONLY(0x0981a, 0x02e5), + A6XX_PROTECT_NORDWR(0x09e00, 0x01ff), + A6XX_PROTECT_NORDWR(0x0a600, 0x01ff), + A6XX_PROTECT_NORDWR(0x0a82e, 0x0000), + A6XX_PROTECT_NORDWR(0x0ae00, 0x0000), + A6XX_PROTECT_NORDWR(0x0ae02, 0x0004), + A6XX_PROTECT_NORDWR(0x0ae08, 0x0006), + A6XX_PROTECT_NORDWR(0x0ae10, 0x00bf), + A6XX_PROTECT_RDONLY(0x0aed0, 0x002f), + A6XX_PROTECT_NORDWR(0x0af00, 0x027f), + A6XX_PROTECT_NORDWR(0x0b600, 0x1fff), + A6XX_PROTECT_NORDWR(0x0dc00, 0x1fff), + A6XX_PROTECT_RDONLY(0x0fc00, 0x1fff), + A6XX_PROTECT_NORDWR(0x18400, 0x003f), + A6XX_PROTECT_RDONLY(0x18440, 0x013f), + A6XX_PROTECT_NORDWR(0x18580, 0x1fff), + A6XX_PROTECT_NORDWR(0x1b400, 0x1fff), + A6XX_PROTECT_NORDWR(0x1f400, 0x0477), + A6XX_PROTECT_RDONLY(0x1f878, 0x0507), + A6XX_PROTECT_NORDWR(0x1f930, 0x0329), + A6XX_PROTECT_NORDWR(0x1fd80, 0x1fff), + A6XX_PROTECT_NORDWR(0x27800, 0x007f), + A6XX_PROTECT_RDONLY(0x27880, 0x0385), + A6XX_PROTECT_NORDWR(0x27882, 0x0009), + A6XX_PROTECT_NORDWR(0x27c06, 0x0000), +}; +DECLARE_ADRENO_PROTECT(a840_protect, 15); + +static const struct adreno_reglist a840_gbif[] = { + { REG_A6XX_GBIF_QSB_SIDE0, 0x00071e20 }, + { REG_A6XX_GBIF_QSB_SIDE1, 0x00071e20 }, + { REG_A6XX_GBIF_QSB_SIDE2, 0x00071e20 }, + { REG_A6XX_GBIF_QSB_SIDE3, 0x00071e20 }, + { REG_A6XX_GBIF_CX_CONFIG, 0x20023000 }, + { }, +}; + +static const uint32_t a840_pwrup_reglist_regs[] = { + REG_A7XX_SP_HLSQ_TIMEOUT_THRESHOLD_DP, + REG_A7XX_SP_READ_SEL, + REG_A6XX_UCHE_MODE_CNTL, + REG_A8XX_UCHE_VARB_IDLE_TIMEOUT, + REG_A8XX_UCHE_GBIF_GX_CONFIG, + REG_A8XX_UCHE_CCHE_MODE_CNTL, + REG_A8XX_UCHE_CCHE_CACHE_WAYS, + REG_A8XX_UCHE_CACHE_WAYS, + REG_A8XX_UCHE_CCHE_GC_GMEM_RANGE_MIN, + REG_A8XX_UCHE_CCHE_GC_GMEM_RANGE_MIN + 1, + REG_A8XX_UCHE_CCHE_LPAC_GMEM_RANGE_MIN, + REG_A8XX_UCHE_CCHE_LPAC_GMEM_RANGE_MIN + 1, + REG_A8XX_UCHE_CCHE_TRAP_BASE, + REG_A8XX_UCHE_CCHE_TRAP_BASE + 1, + REG_A8XX_UCHE_CCHE_WRITE_THRU_BASE, + REG_A8XX_UCHE_CCHE_WRITE_THRU_BASE + 1, + REG_A8XX_UCHE_HW_DBG_CNTL, + REG_A8XX_UCHE_WRITE_THRU_BASE, + REG_A8XX_UCHE_WRITE_THRU_BASE + 1, + REG_A8XX_UCHE_TRAP_BASE, + REG_A8XX_UCHE_TRAP_BASE + 1, + REG_A8XX_UCHE_CLIENT_PF, + REG_A8XX_RB_CMP_NC_MODE_CNTL, + REG_A8XX_SP_HLSQ_GC_GMEM_RANGE_MIN, + REG_A8XX_SP_HLSQ_GC_GMEM_RANGE_MIN + 1, + REG_A6XX_TPL1_NC_MODE_CNTL, + REG_A6XX_TPL1_DBG_ECO_CNTL, + REG_A6XX_TPL1_DBG_ECO_CNTL1, + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(0), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(1), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(2), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(3), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(4), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(5), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(6), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(7), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(8), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(9), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(10), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(11), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(12), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(13), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(14), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(15), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(16), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(17), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(18), + REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(19), +}; +DECLARE_ADRENO_REGLIST_LIST(a840_pwrup_reglist); + +static const u32 a840_ifpc_reglist_regs[] = { + REG_A8XX_RBBM_NC_MODE_CNTL, + REG_A8XX_RBBM_SLICE_NC_MODE_CNTL, + REG_A6XX_SP_NC_MODE_CNTL, + REG_A6XX_SP_CHICKEN_BITS, + REG_A8XX_SP_SS_CHICKEN_BITS_0, + REG_A7XX_SP_CHICKEN_BITS_1, + REG_A7XX_SP_CHICKEN_BITS_2, + REG_A7XX_SP_CHICKEN_BITS_3, + REG_A8XX_SP_CHICKEN_BITS_4, + REG_A6XX_SP_PERFCTR_SHADER_MASK, + REG_A8XX_RBBM_SLICE_PERFCTR_CNTL, + REG_A8XX_RBBM_SLICE_INTERFACE_HANG_INT_CNTL, + REG_A7XX_SP_HLSQ_DBG_ECO_CNTL, + REG_A7XX_SP_HLSQ_DBG_ECO_CNTL_1, + REG_A7XX_SP_HLSQ_DBG_ECO_CNTL_2, + REG_A8XX_SP_HLSQ_DBG_ECO_CNTL_3, + REG_A8XX_SP_HLSQ_LPAC_GMEM_RANGE_MIN, + REG_A8XX_SP_HLSQ_LPAC_GMEM_RANGE_MIN + 1, + REG_A8XX_CP_INTERRUPT_STATUS_MASK_GLOBAL, + REG_A8XX_RBBM_PERFCTR_CNTL, + REG_A8XX_CP_PROTECT_GLOBAL(0), + REG_A8XX_CP_PROTECT_GLOBAL(1), + REG_A8XX_CP_PROTECT_GLOBAL(2), + REG_A8XX_CP_PROTECT_GLOBAL(3), + REG_A8XX_CP_PROTECT_GLOBAL(4), + REG_A8XX_CP_PROTECT_GLOBAL(5), + REG_A8XX_CP_PROTECT_GLOBAL(6), + REG_A8XX_CP_PROTECT_GLOBAL(7), + REG_A8XX_CP_PROTECT_GLOBAL(8), + REG_A8XX_CP_PROTECT_GLOBAL(9), + REG_A8XX_CP_PROTECT_GLOBAL(10), + REG_A8XX_CP_PROTECT_GLOBAL(11), + REG_A8XX_CP_PROTECT_GLOBAL(12), + REG_A8XX_CP_PROTECT_GLOBAL(13), + REG_A8XX_CP_PROTECT_GLOBAL(14), + REG_A8XX_CP_PROTECT_GLOBAL(15), + REG_A8XX_CP_PROTECT_GLOBAL(16), + REG_A8XX_CP_PROTECT_GLOBAL(17), + REG_A8XX_CP_PROTECT_GLOBAL(18), + REG_A8XX_CP_PROTECT_GLOBAL(19), + REG_A8XX_CP_PROTECT_GLOBAL(20), + REG_A8XX_CP_PROTECT_GLOBAL(21), + REG_A8XX_CP_PROTECT_GLOBAL(22), + REG_A8XX_CP_PROTECT_GLOBAL(23), + REG_A8XX_CP_PROTECT_GLOBAL(24), + REG_A8XX_CP_PROTECT_GLOBAL(25), + REG_A8XX_CP_PROTECT_GLOBAL(26), + REG_A8XX_CP_PROTECT_GLOBAL(27), + REG_A8XX_CP_PROTECT_GLOBAL(28), + REG_A8XX_CP_PROTECT_GLOBAL(29), + REG_A8XX_CP_PROTECT_GLOBAL(30), + REG_A8XX_CP_PROTECT_GLOBAL(31), + REG_A8XX_CP_PROTECT_GLOBAL(32), + REG_A8XX_CP_PROTECT_GLOBAL(33), + REG_A8XX_CP_PROTECT_GLOBAL(34), + REG_A8XX_CP_PROTECT_GLOBAL(35), + REG_A8XX_CP_PROTECT_GLOBAL(36), + REG_A8XX_CP_PROTECT_GLOBAL(37), + REG_A8XX_CP_PROTECT_GLOBAL(38), + REG_A8XX_CP_PROTECT_GLOBAL(39), + REG_A8XX_CP_PROTECT_GLOBAL(40), + REG_A8XX_CP_PROTECT_GLOBAL(41), + REG_A8XX_CP_PROTECT_GLOBAL(42), + REG_A8XX_CP_PROTECT_GLOBAL(43), + REG_A8XX_CP_PROTECT_GLOBAL(44), + REG_A8XX_CP_PROTECT_GLOBAL(45), + REG_A8XX_CP_PROTECT_GLOBAL(46), + REG_A8XX_CP_PROTECT_GLOBAL(47), + REG_A8XX_CP_PROTECT_GLOBAL(48), + REG_A8XX_CP_PROTECT_GLOBAL(49), + REG_A8XX_CP_PROTECT_GLOBAL(50), + REG_A8XX_CP_PROTECT_GLOBAL(51), + REG_A8XX_CP_PROTECT_GLOBAL(52), + REG_A8XX_CP_PROTECT_GLOBAL(53), + REG_A8XX_CP_PROTECT_GLOBAL(54), + REG_A8XX_CP_PROTECT_GLOBAL(55), + REG_A8XX_CP_PROTECT_GLOBAL(56), + REG_A8XX_CP_PROTECT_GLOBAL(57), + REG_A8XX_CP_PROTECT_GLOBAL(58), + REG_A8XX_CP_PROTECT_GLOBAL(59), + REG_A8XX_CP_PROTECT_GLOBAL(60), + REG_A8XX_CP_PROTECT_GLOBAL(61), + REG_A8XX_CP_PROTECT_GLOBAL(62), + REG_A8XX_CP_PROTECT_GLOBAL(63), +}; +DECLARE_ADRENO_REGLIST_LIST(a840_ifpc_reglist); + +static const struct adreno_reglist_pipe a840_dyn_pwrup_reglist_regs[] = { + { REG_A8XX_GRAS_TSEFE_DBG_ECO_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_GRAS_NC_MODE_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_GRAS_DBG_ECO_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A6XX_PC_AUTO_VERTEX_STRIDE, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_1, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_2, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_3, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CHICKEN_BITS_4, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_CONTEXT_SWITCH_STABILIZE_CNTL_1, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_PC_VIS_STREAM_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A7XX_RB_CCU_CNTL, 0, BIT(PIPE_BR) }, + { REG_A7XX_RB_CCU_DBG_ECO_CNTL, 0, BIT(PIPE_BR)}, + { REG_A8XX_RB_CCU_NC_MODE_CNTL, 0, BIT(PIPE_BR) }, + { REG_A8XX_RB_CMP_NC_MODE_CNTL, 0, BIT(PIPE_BR) }, + { REG_A6XX_RB_RBP_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_RB_RESOLVE_PREFETCH_CNTL, 0, BIT(PIPE_BR) }, + { REG_A6XX_RB_DBG_ECO_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_RB_CMP_DBG_ECO_CNTL, 0, BIT(PIPE_BR) }, + { REG_A7XX_VFD_DBG_ECO_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BV_THRESHOLD, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BR_THRESHOLD, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_BUSY_REQ_CNT, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VFD_CB_LP_REQ_CNT, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_VPC_FLATSHADE_MODE_CNTL, 0, BIT(PIPE_BV) | BIT(PIPE_BR) }, + { REG_A8XX_CP_HW_FAULT_STATUS_MASK_PIPE, 0, BIT(PIPE_BR) | + BIT(PIPE_BV) | BIT(PIPE_LPAC) | BIT(PIPE_AQE0) | + BIT(PIPE_AQE1) | BIT(PIPE_DDE_BR) | BIT(PIPE_DDE_BV) }, + { REG_A8XX_CP_INTERRUPT_STATUS_MASK_PIPE, 0, BIT(PIPE_BR) | + BIT(PIPE_BV) | BIT(PIPE_LPAC) | BIT(PIPE_AQE0) | + BIT(PIPE_AQE1) | BIT(PIPE_DDE_BR) | BIT(PIPE_DDE_BV) }, + { REG_A8XX_CP_PROTECT_CNTL_PIPE, 0, BIT(PIPE_BR) | BIT(PIPE_BV) | BIT(PIPE_LPAC)}, + { REG_A8XX_CP_PROTECT_PIPE(15), 0, BIT(PIPE_BR) | BIT(PIPE_BV) | BIT(PIPE_LPAC) }, + { REG_A8XX_RB_GC_GMEM_PROTECT, 0, BIT(PIPE_BR) }, + { REG_A8XX_RB_LPAC_GMEM_PROTECT, 0, BIT(PIPE_BR) }, + { REG_A6XX_RB_CONTEXT_SWITCH_GMEM_SAVE_RESTORE_ENABLE, 0, BIT(PIPE_BR) }, +}; +DECLARE_ADRENO_REGLIST_PIPE_LIST(a840_dyn_pwrup_reglist); + +static const struct adreno_info a8xx_gpus[] = { + { + .chip_ids = ADRENO_CHIP_IDS(0x44070001), + .family = ADRENO_8XX_GEN2, + .fw = { + [ADRENO_FW_SQE] = "gen80100_sqe.fw", + [ADRENO_FW_GMU] = "gen80100_gmu.bin", + }, + .gmem = 21 * SZ_1M, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV | + ADRENO_QUIRK_PREEMPTION | + ADRENO_QUIRK_SOFTFUSE, + .funcs = &a8xx_gpu_funcs, + .a6xx = &(const struct a6xx_info) { + .protect = &x285_protect, + .nonctxt_reglist = x285_nonctxt_regs, + .pwrup_reglist = &a840_pwrup_reglist, + .dyn_pwrup_reglist = &x285_dyn_pwrup_reglist, + .ifpc_reglist = &a840_ifpc_reglist, + .gbif_cx = a840_gbif, + .max_slices = 4, + .gmu_chipid = 0x8010100, + .bcms = (const struct a6xx_bcm[]) { + { .name = "SH0", .buswidth = 16 }, + { .name = "MC0", .buswidth = 4 }, + { + .name = "ACV", + .fixed = true, + .perfmode = BIT(2), + .perfmode_bw = 16500000, + }, + { /* sentinel */ }, + }, + }, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 388, 1 }, + { 357, 2 }, + { 284, 3 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS(0x44050a01), + .family = ADRENO_8XX_GEN2, + .fw = { + [ADRENO_FW_SQE] = "gen80200_sqe.fw", + [ADRENO_FW_GMU] = "gen80200_gmu.bin", + [ADRENO_FW_AQE] = "gen80200_aqe.fw", + }, + .gmem = 18 * SZ_1M, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV | + ADRENO_QUIRK_PREEMPTION | + ADRENO_QUIRK_IFPC, + .funcs = &a8xx_gpu_funcs, + .a6xx = &(const struct a6xx_info) { + .protect = &a840_protect, + .nonctxt_reglist = a840_nonctxt_regs, + .pwrup_reglist = &a840_pwrup_reglist, + .dyn_pwrup_reglist = &a840_dyn_pwrup_reglist, + .ifpc_reglist = &a840_ifpc_reglist, + .gbif_cx = a840_gbif, + .max_slices = 3, + .gmu_chipid = 0x8020100, + .bcms = (const struct a6xx_bcm[]) { + { .name = "SH0", .buswidth = 16 }, + { .name = "MC0", .buswidth = 4 }, + { + .name = "ACV", + .fixed = true, + .perfmode = BIT(2), + .perfmode_bw = 10687500, + }, + { /* sentinel */ }, + }, + }, + .preempt_record_size = 19708 * SZ_1K, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 273, 1 }, + { 252, 2 }, + { 221, 3 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS(0x44010000), + .family = ADRENO_8XX_GEN1, + .fw = { + [ADRENO_FW_SQE] = "gen80300_sqe.fw", + [ADRENO_FW_GMU] = "gen80300_gmu.bin", + }, + .gmem = SZ_512K + SZ_64K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV | + ADRENO_QUIRK_PREEMPTION | + ADRENO_QUIRK_IFPC, + .funcs = &a8xx_gpu_funcs, + .zapfw = "gen80300_zap.mbn", + .a6xx = &(const struct a6xx_info) { + .protect = &a810_protect, + .nonctxt_reglist = a810_nonctxt_regs, + .pwrup_reglist = &a810_pwrup_reglist, + .dyn_pwrup_reglist = &a810_dyn_pwrup_reglist, + .ifpc_reglist = &a810_ifpc_reglist, + .gbif_cx = a840_gbif, + .max_slices = 1, + .gmu_chipid = 0x8030000, + .bcms = (const struct a6xx_bcm[]) { + { .name = "SH0", .buswidth = 16 }, + { .name = "MC0", .buswidth = 4 }, + { + .name = "ACV", + .fixed = true, + .perfmode = BIT(2), + .perfmode_bw = 10687500, + }, + { /* sentinel */ }, + }, + }, + .preempt_record_size = 4558 * SZ_1K, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 242, 1 }, + { 221, 2 }, + ), + } +}; + +DECLARE_ADRENO_GPULIST(a8xx); + +static inline __always_unused void __build_asserts(void) +{ + BUILD_BUG_ON(a630_protect.count > a630_protect.count_max); + BUILD_BUG_ON(a650_protect.count > a650_protect.count_max); + BUILD_BUG_ON(a660_protect.count > a660_protect.count_max); + BUILD_BUG_ON(a690_protect.count > a690_protect.count_max); + BUILD_BUG_ON(a730_protect.count > a730_protect.count_max); + BUILD_BUG_ON(a810_protect.count > a810_protect.count_max); +} diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c new file mode 100644 index 000000000..6d49c51df --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c @@ -0,0 +1,2499 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2017-2019 The Linux Foundation. All rights reserved. */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "a6xx_gpu.h" +#include "a6xx_gmu.xml.h" +#include "msm_gem.h" +#include "msm_gpu_trace.h" +#include "msm_mmu.h" + +static void a6xx_gmu_fault(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + + /* FIXME: add a banner here */ + gmu->hung = true; + + /* Turn off the hangcheck timer while we are resetting */ + timer_delete(&gpu->hangcheck_timer); + + /* Queue the GPU handler because we need to treat this as a recovery */ + kthread_queue_work(gpu->worker, &gpu->recover_work); +} + +static irqreturn_t a6xx_gmu_irq(int irq, void *data) +{ + struct a6xx_gmu *gmu = data; + u32 status; + + status = gmu_read(gmu, REG_A6XX_GMU_AO_HOST_INTERRUPT_STATUS); + gmu_write(gmu, REG_A6XX_GMU_AO_HOST_INTERRUPT_CLR, status); + + if (status & A6XX_GMU_AO_HOST_INTERRUPT_STATUS_WDOG_BITE) { + dev_err_ratelimited(gmu->dev, "GMU watchdog expired\n"); + + a6xx_gmu_fault(gmu); + } + + if (status & A6XX_GMU_AO_HOST_INTERRUPT_STATUS_HOST_AHB_BUS_ERROR) + dev_err_ratelimited(gmu->dev, "GMU AHB bus error\n"); + + if (status & A6XX_GMU_AO_HOST_INTERRUPT_STATUS_FENCE_ERR) + dev_err_ratelimited(gmu->dev, "GMU fence error: 0x%x\n", + gmu_read(gmu, REG_A6XX_GMU_AHB_FENCE_STATUS)); + + return IRQ_HANDLED; +} + +static irqreturn_t a6xx_hfi_irq(int irq, void *data) +{ + struct a6xx_gmu *gmu = data; + u32 status; + + status = gmu_read(gmu, REG_A6XX_GMU_GMU2HOST_INTR_INFO); + gmu_write(gmu, REG_A6XX_GMU_GMU2HOST_INTR_CLR, status); + + if (status & A6XX_GMU_GMU2HOST_INTR_INFO_CM3_FAULT) { + dev_err_ratelimited(gmu->dev, "GMU firmware fault\n"); + + a6xx_gmu_fault(gmu); + } + + return IRQ_HANDLED; +} + +bool a6xx_gmu_sptprac_is_on(struct a6xx_gmu *gmu) +{ + u32 val; + + /* This can be called from gpu state code so make sure GMU is valid */ + if (!gmu->initialized) + return false; + + val = gmu_read(gmu, REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS); + + return !(val & + (A6XX_GMU_SPTPRAC_PWR_CLK_STATUS_SPTPRAC_GDSC_POWER_OFF | + A6XX_GMU_SPTPRAC_PWR_CLK_STATUS_SP_CLOCK_OFF)); +} + +/* Check to see if the GX rail is still powered */ +bool a6xx_gmu_gx_is_on(struct adreno_gpu *adreno_gpu) +{ + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + u32 val; + + /* This can be called from gpu state code so make sure GMU is valid */ + if (!gmu->initialized) + return false; + + /* If GMU is absent, then GX power domain is ON as long as GPU is in active state */ + if (adreno_has_gmu_wrapper(adreno_gpu)) + return true; + + val = gmu_read(gmu, REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS); + + if (adreno_is_a7xx(adreno_gpu)) + return !(val & + (A7XX_GMU_SPTPRAC_PWR_CLK_STATUS_GX_HM_GDSC_POWER_OFF | + A7XX_GMU_SPTPRAC_PWR_CLK_STATUS_GX_HM_CLK_OFF)); + + return !(val & + (A6XX_GMU_SPTPRAC_PWR_CLK_STATUS_GX_HM_GDSC_POWER_OFF | + A6XX_GMU_SPTPRAC_PWR_CLK_STATUS_GX_HM_CLK_OFF)); +} + +bool a7xx_gmu_gx_is_on(struct adreno_gpu *adreno_gpu) +{ + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + u32 val; + + /* This can be called from gpu state code so make sure GMU is valid */ + if (!gmu->initialized) + return false; + + val = gmu_read(gmu, REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS); + + return !(val & + (A7XX_GMU_SPTPRAC_PWR_CLK_STATUS_GX_HM_GDSC_POWER_OFF | + A7XX_GMU_SPTPRAC_PWR_CLK_STATUS_GX_HM_CLK_OFF)); +} + +bool a8xx_gmu_gx_is_on(struct adreno_gpu *adreno_gpu) +{ + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + u32 val; + + /* This can be called from gpu state code so make sure GMU is valid */ + if (!gmu->initialized) + return false; + + val = gmu_read(gmu, REG_A8XX_GMU_PWR_CLK_STATUS); + + return !(val & + (A8XX_GMU_PWR_CLK_STATUS_GX_HM_GDSC_POWER_OFF | + A8XX_GMU_PWR_CLK_STATUS_GX_HM_CLK_OFF)); +} + +void a6xx_gmu_set_freq(struct msm_gpu *gpu, struct dev_pm_opp *opp, + bool suspended) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + const struct a6xx_info *info = adreno_gpu->info->a6xx; + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + u32 perf_index; + u32 bw_index = 0; + unsigned long gpu_freq; + int ret = 0; + + gpu_freq = dev_pm_opp_get_freq(opp); + + if (gpu_freq == gmu->freq) + return; + + for (perf_index = 0; perf_index < gmu->nr_gpu_freqs - 1; perf_index++) + if (gpu_freq == gmu->gpu_freqs[perf_index]) + break; + + /* If enabled, find the corresponding DDR bandwidth index */ + if (info->bcms && gmu->nr_gpu_bws > 1) { + unsigned int bw = dev_pm_opp_get_bw(opp, true, 0); + + for (bw_index = 0; bw_index < gmu->nr_gpu_bws - 1; bw_index++) { + if (bw == gmu->gpu_bw_table[bw_index]) + break; + } + + /* Vote AB as a fraction of the max bandwidth, starting from A750 */ + if (bw && adreno_is_a750_family(adreno_gpu)) { + u64 tmp; + + /* For now, vote for 25% of the bandwidth */ + tmp = bw * 25; + do_div(tmp, 100); + + /* + * The AB vote consists of a 16 bit wide quantized level + * against the maximum supported bandwidth. + * Quantization can be calculated as below: + * vote = (bandwidth * 2^16) / max bandwidth + */ + tmp *= MAX_AB_VOTE; + do_div(tmp, gmu->gpu_bw_table[gmu->nr_gpu_bws - 1]); + + bw_index |= AB_VOTE(clamp(tmp, 1, MAX_AB_VOTE)); + bw_index |= AB_VOTE_ENABLE; + } + } + + gmu->current_perf_index = perf_index; + gmu->freq = gmu->gpu_freqs[perf_index]; + + trace_msm_gmu_freq_change(gmu->freq, perf_index); + + /* + * This can get called from devfreq while the hardware is idle. Don't + * bring up the power if it isn't already active. All we're doing here + * is updating the frequency so that when we come back online we're at + * the right rate. + */ + if (suspended) + return; + + if (!gmu->legacy) { + a6xx_hfi_set_freq(gmu, perf_index, bw_index); + /* With Bandwidth voting, we now vote for all resources, so skip OPP set */ + if (!bw_index) + dev_pm_opp_set_opp(&gpu->pdev->dev, opp); + return; + } + + gmu_write(gmu, REG_A6XX_GMU_DCVS_ACK_OPTION, 0); + + gmu_write(gmu, REG_A6XX_GMU_DCVS_PERF_SETTING, + ((3 & 0xf) << 28) | perf_index); + + /* + * Send an invalid index as a vote for the bus bandwidth and let the + * firmware decide on the right vote + */ + gmu_write(gmu, REG_A6XX_GMU_DCVS_BW_SETTING, 0xff); + + /* Set and clear the OOB for DCVS to trigger the GMU */ + a6xx_gmu_set_oob(gmu, GMU_OOB_DCVS_SET); + a6xx_gmu_clear_oob(gmu, GMU_OOB_DCVS_SET); + + ret = gmu_read(gmu, REG_A6XX_GMU_DCVS_RETURN); + if (ret) + dev_err(gmu->dev, "GMU set GPU frequency error: %d\n", ret); + + dev_pm_opp_set_opp(&gpu->pdev->dev, opp); +} + +unsigned long a6xx_gmu_get_freq(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + + return gmu->freq; +} + +static bool a6xx_gmu_check_idle_level(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + int local = gmu->idle_level; + u32 val; + + /* SPTP and IFPC both report as IFPC */ + if (gmu->idle_level == GMU_IDLE_STATE_SPTP) + local = GMU_IDLE_STATE_IFPC; + + if (adreno_is_a8xx(adreno_gpu)) + val = gmu_read(gmu, REG_A8XX_GPU_GMU_CX_GMU_RPMH_POWER_STATE); + else + val = gmu_read(gmu, REG_A6XX_GPU_GMU_CX_GMU_RPMH_POWER_STATE); + + if (val == local) { + if (gmu->idle_level != GMU_IDLE_STATE_IFPC || + !adreno_gpu->funcs->gx_is_on(adreno_gpu)) + return true; + } + + return false; +} + +/* Wait for the GMU to get to its most idle state */ +int a6xx_gmu_wait_for_idle(struct a6xx_gmu *gmu) +{ + return spin_until(a6xx_gmu_check_idle_level(gmu)); +} + +static int a6xx_gmu_start(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + u32 mask, reset_val, val; + int ret; + + val = gmu_read(gmu, REG_A6XX_GMU_CM3_DTCM_START + 0xff8); + if (val <= 0x20010004) { + mask = 0xffffffff; + reset_val = 0xbabeface; + } else { + mask = 0x1ff; + reset_val = 0x100; + } + + gmu_write(gmu, REG_A6XX_GMU_CM3_SYSRESET, 1); + + /* Set the log wptr index + * note: downstream saves the value in poweroff and restores it here + */ + if (adreno_is_a8xx(adreno_gpu)) + gmu_write(gmu, REG_A8XX_GMU_GENERAL_9, 0); + else if (adreno_is_a7xx(adreno_gpu)) + gmu_write(gmu, REG_A7XX_GMU_GENERAL_9, 0); + else + gmu_write(gmu, REG_A6XX_GPU_GMU_CX_GMU_PWR_COL_CP_RESP, 0); + + + gmu_write(gmu, REG_A6XX_GMU_CM3_SYSRESET, 0); + + ret = gmu_poll_timeout(gmu, REG_A6XX_GMU_CM3_FW_INIT_RESULT, val, + (val & mask) == reset_val, 100, 100000); + + if (ret) + DRM_DEV_ERROR(gmu->dev, "GMU firmware initialization timed out\n"); + + set_bit(GMU_STATUS_FW_START, &gmu->status); + + return ret; +} + +static int a6xx_gmu_hfi_start(struct a6xx_gmu *gmu) +{ + u32 val; + int ret; + + gmu_write(gmu, REG_A6XX_GMU_HFI_CTRL_INIT, 1); + + ret = gmu_poll_timeout(gmu, REG_A6XX_GMU_HFI_CTRL_STATUS, val, + val & 1, 100, 10000); + if (ret) + DRM_DEV_ERROR(gmu->dev, "Unable to start the HFI queues\n"); + + return ret; +} + +struct a6xx_gmu_oob_bits { + int set, ack, set_new, ack_new, clear, clear_new; + const char *name; +}; + +/* These are the interrupt / ack bits for each OOB request that are set + * in a6xx_gmu_set_oob and a6xx_clear_oob + */ +static const struct a6xx_gmu_oob_bits a6xx_gmu_oob_bits[] = { + [GMU_OOB_GPU_SET] = { + .name = "GPU_SET", + .set = 16, + .ack = 24, + .set_new = 30, + .ack_new = 31, + .clear = 24, + .clear_new = 31, + }, + + [GMU_OOB_PERFCOUNTER_SET] = { + .name = "PERFCOUNTER", + .set = 17, + .ack = 25, + .set_new = 28, + .ack_new = 30, + .clear = 25, + .clear_new = 29, + }, + + [GMU_OOB_BOOT_SLUMBER] = { + .name = "BOOT_SLUMBER", + .set = 22, + .ack = 30, + .clear = 30, + }, + + [GMU_OOB_DCVS_SET] = { + .name = "GPU_DCVS", + .set = 23, + .ack = 31, + .clear = 31, + }, +}; + +/* Trigger a OOB (out of band) request to the GMU */ +int a6xx_gmu_set_oob(struct a6xx_gmu *gmu, enum a6xx_gmu_oob_state state) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + int ret; + u32 val; + int request, ack; + + WARN_ON_ONCE(!mutex_is_locked(&gmu->lock)); + + /* Skip OOB calls since RGMU is not enabled */ + if (adreno_has_rgmu(adreno_gpu)) + return 0; + + if (state >= ARRAY_SIZE(a6xx_gmu_oob_bits)) + return -EINVAL; + + if (gmu->legacy) { + request = a6xx_gmu_oob_bits[state].set; + ack = a6xx_gmu_oob_bits[state].ack; + } else { + request = a6xx_gmu_oob_bits[state].set_new; + ack = a6xx_gmu_oob_bits[state].ack_new; + if (!request || !ack) { + DRM_DEV_ERROR(gmu->dev, + "Invalid non-legacy GMU request %s\n", + a6xx_gmu_oob_bits[state].name); + return -EINVAL; + } + } + + /* Trigger the equested OOB operation */ + gmu_write(gmu, REG_A6XX_GMU_HOST2GMU_INTR_SET, 1 << request); + + do { + /* Wait for the acknowledge interrupt */ + ret = gmu_poll_timeout(gmu, REG_A6XX_GMU_GMU2HOST_INTR_INFO, val, + val & (1 << ack), 100, 10000); + + if (!ret) + break; + + if (completion_done(&a6xx_gpu->base.fault_coredump_done)) + break; + + /* We may timeout because the GMU is temporarily wedged from + * pending faults from the GPU and we are taking a devcoredump. + * Wait until the MMU is resumed and try again. + */ + wait_for_completion(&a6xx_gpu->base.fault_coredump_done); + } while (true); + + if (ret) + DRM_DEV_ERROR(gmu->dev, + "Timeout waiting for GMU OOB set %s: 0x%x\n", + a6xx_gmu_oob_bits[state].name, + gmu_read(gmu, REG_A6XX_GMU_GMU2HOST_INTR_INFO)); + + /* Clear the acknowledge interrupt */ + gmu_write(gmu, REG_A6XX_GMU_GMU2HOST_INTR_CLR, 1 << ack); + + return ret; +} + +/* Clear a pending OOB state in the GMU */ +void a6xx_gmu_clear_oob(struct a6xx_gmu *gmu, enum a6xx_gmu_oob_state state) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + int bit; + + WARN_ON_ONCE(!mutex_is_locked(&gmu->lock)); + + /* Skip OOB calls since RGMU is not enabled */ + if (adreno_has_rgmu(adreno_gpu)) + return; + + if (state >= ARRAY_SIZE(a6xx_gmu_oob_bits)) + return; + + if (gmu->legacy) + bit = a6xx_gmu_oob_bits[state].clear; + else + bit = a6xx_gmu_oob_bits[state].clear_new; + + gmu_write(gmu, REG_A6XX_GMU_HOST2GMU_INTR_SET, 1 << bit); +} + +/* Enable CPU control of SPTP power power collapse */ +int a6xx_sptprac_enable(struct a6xx_gmu *gmu) +{ + int ret; + u32 val; + + WARN_ON(!gmu->legacy); + + /* Nothing to do if GMU does the power management */ + if (gmu->idle_level > GMU_IDLE_STATE_ACTIVE) + return 0; + + gmu_write(gmu, REG_A6XX_GMU_GX_SPTPRAC_POWER_CONTROL, 0x778000); + + ret = gmu_poll_timeout(gmu, REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS, val, + (val & 0x38) == 0x28, 1, 100); + + if (ret) { + DRM_DEV_ERROR(gmu->dev, "Unable to power on SPTPRAC: 0x%x\n", + gmu_read(gmu, REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS)); + } + + return 0; +} + +/* Disable CPU control of SPTP power power collapse */ +void a6xx_sptprac_disable(struct a6xx_gmu *gmu) +{ + u32 val; + int ret; + + if (!gmu->legacy) + return; + + /* Make sure retention is on */ + gmu_rmw(gmu, REG_A6XX_GPU_CC_GX_GDSCR, 0, (1 << 11)); + + gmu_write(gmu, REG_A6XX_GMU_GX_SPTPRAC_POWER_CONTROL, 0x778001); + + ret = gmu_poll_timeout(gmu, REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS, val, + (val & 0x04), 100, 10000); + + if (ret) + DRM_DEV_ERROR(gmu->dev, "failed to power off SPTPRAC: 0x%x\n", + gmu_read(gmu, REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS)); +} + +/* Let the GMU know we are starting a boot sequence */ +static int a6xx_gmu_gfx_rail_on(struct a6xx_gmu *gmu) +{ + u32 vote; + + /* Let the GMU know we are getting ready for boot */ + gmu_write(gmu, REG_A6XX_GMU_BOOT_SLUMBER_OPTION, 0); + + /* Choose the "default" power level as the highest available */ + vote = gmu->gx_arc_votes[gmu->nr_gpu_freqs - 1]; + + gmu_write(gmu, REG_A6XX_GMU_GX_VOTE_IDX, vote & 0xff); + gmu_write(gmu, REG_A6XX_GMU_MX_VOTE_IDX, (vote >> 8) & 0xff); + + /* Let the GMU know the boot sequence has started */ + return a6xx_gmu_set_oob(gmu, GMU_OOB_BOOT_SLUMBER); +} + +static void a6xx_gemnoc_workaround(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + + /* + * GEMNoC can power collapse whilst the GPU is being powered down, resulting + * in the power down sequence not being fully executed. That in turn can + * prevent CX_GDSC from collapsing. Assert Qactive to avoid this. + */ + if (adreno_is_a8xx(adreno_gpu)) + gmu_write(gmu, REG_A8XX_GPU_GMU_CX_GMU_CX_FALNEXT_INTF, BIT(0)); + else if (adreno_is_a7xx(adreno_gpu) || (adreno_is_a621(adreno_gpu) || + adreno_is_7c3(adreno_gpu))) + gmu_write(gmu, REG_A6XX_GPU_GMU_CX_GMU_CX_FALNEXT_INTF, BIT(0)); +} + +/* Let the GMU know that we are about to go into slumber */ +static int a6xx_gmu_notify_slumber(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + int ret; + + /* Disable the power counter so the GMU isn't busy */ + if (adreno_is_a8xx(adreno_gpu)) + gmu_write(gmu, REG_A8XX_GMU_CX_GMU_POWER_COUNTER_ENABLE, 0); + else + gmu_write(gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_ENABLE, 0); + + /* Disable SPTP_PC if the CPU is responsible for it */ + if (gmu->idle_level < GMU_IDLE_STATE_SPTP) + a6xx_sptprac_disable(gmu); + + if (!gmu->legacy) { + ret = a6xx_hfi_send_prep_slumber(gmu); + goto out; + } + + /* Tell the GMU to get ready to slumber */ + gmu_write(gmu, REG_A6XX_GMU_BOOT_SLUMBER_OPTION, 1); + + ret = a6xx_gmu_set_oob(gmu, GMU_OOB_BOOT_SLUMBER); + a6xx_gmu_clear_oob(gmu, GMU_OOB_BOOT_SLUMBER); + + if (!ret) { + /* Check to see if the GMU really did slumber */ + if (gmu_read(gmu, REG_A6XX_GPU_GMU_CX_GMU_RPMH_POWER_STATE) + != 0x0f) { + DRM_DEV_ERROR(gmu->dev, "The GMU did not go into slumber\n"); + ret = -ETIMEDOUT; + } + } + +out: + /* Put fence into allow mode */ + gmu_write(gmu, REG_A6XX_GMU_AO_AHB_FENCE_CTRL, 0); + a6xx_gemnoc_workaround(gmu); + return ret; +} + +static int a6xx_rpmh_start(struct a6xx_gmu *gmu) +{ + int ret; + u32 val; + + if (!test_and_clear_bit(GMU_STATUS_PDC_SLEEP, &gmu->status)) + return 0; + + gmu_write(gmu, REG_A6XX_GMU_RSCC_CONTROL_REQ, BIT(1)); + + ret = gmu_poll_timeout(gmu, REG_A6XX_GMU_RSCC_CONTROL_ACK, val, + val & (1 << 1), 100, 10000); + if (ret) { + DRM_DEV_ERROR(gmu->dev, "Unable to power on the GPU RSC\n"); + return ret; + } + + ret = gmu_poll_timeout_rscc(gmu, REG_A6XX_RSCC_SEQ_BUSY_DRV0, val, + !val, 100, 10000); + + if (ret) { + DRM_DEV_ERROR(gmu->dev, "GPU RSC sequence stuck while waking up the GPU\n"); + return ret; + } + + gmu_write(gmu, REG_A6XX_GMU_RSCC_CONTROL_REQ, 0); + + return 0; +} + +static void a6xx_rpmh_stop(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + u32 bitmask = BIT(16); + int ret; + u32 val; + + if (!test_and_clear_bit(GMU_STATUS_FW_START, &gmu->status)) + return; + + if (adreno_is_a840(adreno_gpu)) + bitmask = BIT(30); + + gmu_write(gmu, REG_A6XX_GMU_RSCC_CONTROL_REQ, 1); + + ret = gmu_poll_timeout_rscc(gmu, REG_A6XX_GPU_RSCC_RSC_STATUS0_DRV0, + val, val & bitmask, 100, 10000); + if (ret) + DRM_DEV_ERROR(gmu->dev, "Unable to power off the GPU RSC\n"); + + gmu_write(gmu, REG_A6XX_GMU_RSCC_CONTROL_REQ, 0); + + set_bit(GMU_STATUS_PDC_SLEEP, &gmu->status); +} + +static inline void pdc_write(void __iomem *ptr, u32 offset, u32 value) +{ + writel(value, ptr + (offset << 2)); +} + +static void a6xx_gmu_rpmh_init(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct platform_device *pdev = to_platform_device(gmu->dev); + u32 seqmem0_drv0_reg = REG_A6XX_RSCC_SEQ_MEM_0_DRV0; + void __iomem *seqptr = NULL; + uint32_t pdc_address_offset; + void __iomem *pdcptr; + bool pdc_in_aop = false; + + /* On A8x and above, RPMH/PDC configurations are entirely configured in AOP */ + if (adreno_is_a8xx(adreno_gpu)) + return; + + pdcptr = devm_platform_ioremap_resource_byname(pdev, "gmu_pdc"); + if (IS_ERR(pdcptr)) + return; + + if (adreno_is_a650_family(adreno_gpu) || + adreno_is_a7xx(adreno_gpu)) + pdc_in_aop = true; + else if (adreno_is_a618(adreno_gpu) || adreno_is_a640_family(adreno_gpu)) + pdc_address_offset = 0x30090; + else if (adreno_is_a619(adreno_gpu)) + pdc_address_offset = 0x300a0; + else + pdc_address_offset = 0x30080; + + if (!pdc_in_aop) { + seqptr = devm_platform_ioremap_resource_byname(pdev, "gmu_pdc_seq"); + if (IS_ERR(seqptr)) + return; + } + + /* Disable SDE clock gating */ + gmu_write_rscc(gmu, REG_A6XX_GPU_RSCC_RSC_STATUS0_DRV0, BIT(24)); + + /* Setup RSC PDC handshake for sleep and wakeup */ + gmu_write_rscc(gmu, REG_A6XX_RSCC_PDC_SLAVE_ID_DRV0, 1); + gmu_write_rscc(gmu, REG_A6XX_RSCC_HIDDEN_TCS_CMD0_DATA, 0); + gmu_write_rscc(gmu, REG_A6XX_RSCC_HIDDEN_TCS_CMD0_ADDR, 0); + gmu_write_rscc(gmu, REG_A6XX_RSCC_HIDDEN_TCS_CMD0_DATA + 2, 0); + gmu_write_rscc(gmu, REG_A6XX_RSCC_HIDDEN_TCS_CMD0_ADDR + 2, 0); + gmu_write_rscc(gmu, REG_A6XX_RSCC_HIDDEN_TCS_CMD0_DATA + 4, + (adreno_is_a740_family(adreno_gpu) || + adreno_is_a722(adreno_gpu)) ? 0x80000021 : 0x80000000); + gmu_write_rscc(gmu, REG_A6XX_RSCC_HIDDEN_TCS_CMD0_ADDR + 4, 0); + gmu_write_rscc(gmu, REG_A6XX_RSCC_OVERRIDE_START_ADDR, 0); + gmu_write_rscc(gmu, REG_A6XX_RSCC_PDC_SEQ_START_ADDR, 0x4520); + gmu_write_rscc(gmu, REG_A6XX_RSCC_PDC_MATCH_VALUE_LO, 0x4510); + gmu_write_rscc(gmu, REG_A6XX_RSCC_PDC_MATCH_VALUE_HI, 0x4514); + + /* The second spin of A7xx GPUs messed with some register offsets.. */ + if (adreno_is_a740_family(adreno_gpu) || adreno_is_a722(adreno_gpu)) + seqmem0_drv0_reg = REG_A7XX_RSCC_SEQ_MEM_0_DRV0_A740; + + /* Load RSC sequencer uCode for sleep and wakeup */ + if (adreno_is_a650_family(adreno_gpu) || + adreno_is_a7xx(adreno_gpu)) { + gmu_write_rscc(gmu, seqmem0_drv0_reg, 0xeaaae5a0); + gmu_write_rscc(gmu, seqmem0_drv0_reg + 1, 0xe1a1ebab); + gmu_write_rscc(gmu, seqmem0_drv0_reg + 2, 0xa2e0a581); + gmu_write_rscc(gmu, seqmem0_drv0_reg + 3, 0xecac82e2); + gmu_write_rscc(gmu, seqmem0_drv0_reg + 4, 0x0020edad); + } else { + gmu_write_rscc(gmu, REG_A6XX_RSCC_SEQ_MEM_0_DRV0, 0xa7a506a0); + gmu_write_rscc(gmu, REG_A6XX_RSCC_SEQ_MEM_0_DRV0 + 1, 0xa1e6a6e7); + gmu_write_rscc(gmu, REG_A6XX_RSCC_SEQ_MEM_0_DRV0 + 2, 0xa2e081e1); + gmu_write_rscc(gmu, REG_A6XX_RSCC_SEQ_MEM_0_DRV0 + 3, 0xe9a982e2); + gmu_write_rscc(gmu, REG_A6XX_RSCC_SEQ_MEM_0_DRV0 + 4, 0x0020e8a8); + } + + if (pdc_in_aop) + goto setup_pdc; + + /* Load PDC sequencer uCode for power up and power down sequence */ + pdc_write(seqptr, REG_A6XX_PDC_GPU_SEQ_MEM_0, 0xfebea1e1); + pdc_write(seqptr, REG_A6XX_PDC_GPU_SEQ_MEM_0 + 1, 0xa5a4a3a2); + pdc_write(seqptr, REG_A6XX_PDC_GPU_SEQ_MEM_0 + 2, 0x8382a6e0); + pdc_write(seqptr, REG_A6XX_PDC_GPU_SEQ_MEM_0 + 3, 0xbce3e284); + pdc_write(seqptr, REG_A6XX_PDC_GPU_SEQ_MEM_0 + 4, 0x002081fc); + + /* Set TCS commands used by PDC sequence for low power modes */ + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS1_CMD_ENABLE_BANK, 7); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS1_CMD_WAIT_FOR_CMPL_BANK, 0); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS1_CONTROL, 0); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS1_CMD0_MSGID, 0x10108); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS1_CMD0_ADDR, 0x30010); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS1_CMD0_DATA, 1); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS1_CMD0_MSGID + 4, 0x10108); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS1_CMD0_ADDR + 4, 0x30000); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS1_CMD0_DATA + 4, 0x0); + + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS1_CMD0_MSGID + 8, 0x10108); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS1_CMD0_ADDR + 8, pdc_address_offset); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS1_CMD0_DATA + 8, 0x0); + + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS3_CMD_ENABLE_BANK, 7); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS3_CMD_WAIT_FOR_CMPL_BANK, 0); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS3_CONTROL, 0); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS3_CMD0_MSGID, 0x10108); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS3_CMD0_ADDR, 0x30010); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS3_CMD0_DATA, 2); + + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS3_CMD0_MSGID + 4, 0x10108); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS3_CMD0_ADDR + 4, 0x30000); + if (adreno_is_a618(adreno_gpu) || adreno_is_a619(adreno_gpu) || + adreno_is_a650_family(adreno_gpu)) + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS3_CMD0_DATA + 4, 0x2); + else + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS3_CMD0_DATA + 4, 0x3); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS3_CMD0_MSGID + 8, 0x10108); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS3_CMD0_ADDR + 8, pdc_address_offset); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_TCS3_CMD0_DATA + 8, 0x3); + + /* Setup GPU PDC */ +setup_pdc: + pdc_write(pdcptr, REG_A6XX_PDC_GPU_SEQ_START_ADDR, 0); + pdc_write(pdcptr, REG_A6XX_PDC_GPU_ENABLE_PDC, 0x80000001); + + /* ensure no writes happen before the uCode is fully written */ + wmb(); +} + +/* + * The lowest 16 bits of this value are the number of XO clock cycles for main + * hysteresis which is set at 0x1680 cycles (300 us). The higher 16 bits are + * for the shorter hysteresis that happens after main - this is 0xa (.5 us) + */ + +#define GMU_PWR_COL_HYST 0x000a1680 + +/* Set up the idle state for the GMU */ +static void a6xx_gmu_power_config(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + + /* Disable GMU WB/RB buffer */ + gmu_write(gmu, REG_A6XX_GMU_SYS_BUS_CONFIG, 0x1); + gmu_write(gmu, REG_A6XX_GMU_ICACHE_CONFIG, 0x1); + gmu_write(gmu, REG_A6XX_GMU_DCACHE_CONFIG, 0x1); + + /* A7xx knows better by default! */ + if (adreno_is_a7xx(adreno_gpu) || adreno_is_a8xx(adreno_gpu)) + return; + + gmu_write(gmu, REG_A6XX_GMU_PWR_COL_INTER_FRAME_CTRL, 0x9c40400); + + switch (gmu->idle_level) { + case GMU_IDLE_STATE_IFPC: + gmu_write(gmu, REG_A6XX_GMU_PWR_COL_INTER_FRAME_HYST, + GMU_PWR_COL_HYST); + gmu_rmw(gmu, REG_A6XX_GMU_PWR_COL_INTER_FRAME_CTRL, 0, + A6XX_GMU_PWR_COL_INTER_FRAME_CTRL_IFPC_ENABLE | + A6XX_GMU_PWR_COL_INTER_FRAME_CTRL_HM_POWER_COLLAPSE_ENABLE); + fallthrough; + case GMU_IDLE_STATE_SPTP: + gmu_write(gmu, REG_A6XX_GMU_PWR_COL_SPTPRAC_HYST, + GMU_PWR_COL_HYST); + gmu_rmw(gmu, REG_A6XX_GMU_PWR_COL_INTER_FRAME_CTRL, 0, + A6XX_GMU_PWR_COL_INTER_FRAME_CTRL_IFPC_ENABLE | + A6XX_GMU_PWR_COL_INTER_FRAME_CTRL_SPTPRAC_POWER_CONTROL_ENABLE); + } + + /* Enable RPMh GPU client */ + gmu_rmw(gmu, REG_A6XX_GMU_RPMH_CTRL, 0, + A6XX_GMU_RPMH_CTRL_RPMH_INTERFACE_ENABLE | + A6XX_GMU_RPMH_CTRL_LLC_VOTE_ENABLE | + A6XX_GMU_RPMH_CTRL_DDR_VOTE_ENABLE | + A6XX_GMU_RPMH_CTRL_MX_VOTE_ENABLE | + A6XX_GMU_RPMH_CTRL_CX_VOTE_ENABLE | + A6XX_GMU_RPMH_CTRL_GFX_VOTE_ENABLE); +} + +struct block_header { + u32 addr; + u32 size; + u32 type; + u32 value; + u32 data[]; +}; + +static bool fw_block_mem(struct a6xx_gmu_bo *bo, const struct block_header *blk) +{ + if (!in_range(blk->addr, bo->iova, bo->size)) + return false; + + memcpy(bo->virt + blk->addr - bo->iova, blk->data, blk->size); + return true; +} + +#define NEXT_BLK(blk) \ + ((const struct block_header *)((const char *)(blk) + sizeof(*(blk)) + (blk)->size)) + +static int a6xx_gmu_fw_load(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + const struct firmware *fw_image = adreno_gpu->fw[ADRENO_FW_GMU]; + const struct block_header *blk; + u32 reg_offset; + u32 ver; + + u32 itcm_base = 0x00000000; + u32 dtcm_base = 0x00040000; + + if (adreno_is_a650_family(adreno_gpu) || + adreno_is_a7xx(adreno_gpu) || + adreno_is_a8xx(adreno_gpu)) + dtcm_base = 0x10004000; + + if (gmu->legacy) { + /* Sanity check the size of the firmware that was loaded */ + if (fw_image->size > 0x8000) { + DRM_DEV_ERROR(gmu->dev, + "GMU firmware is bigger than the available region\n"); + return -EINVAL; + } + + gmu_write_bulk(gmu, REG_A6XX_GMU_CM3_ITCM_START, + (u32*) fw_image->data, fw_image->size); + return 0; + } + + + for (blk = (const struct block_header *) fw_image->data; + (const u8*) blk < fw_image->data + fw_image->size; + blk = NEXT_BLK(blk)) { + if (blk->size == 0) + continue; + + if (in_range(blk->addr, itcm_base, SZ_16K)) { + reg_offset = (blk->addr - itcm_base) >> 2; + gmu_write_bulk(gmu, + REG_A6XX_GMU_CM3_ITCM_START + reg_offset, + blk->data, blk->size); + } else if (in_range(blk->addr, dtcm_base, SZ_16K)) { + reg_offset = (blk->addr - dtcm_base) >> 2; + gmu_write_bulk(gmu, + REG_A6XX_GMU_CM3_DTCM_START + reg_offset, + blk->data, blk->size); + } else if (!fw_block_mem(&gmu->icache, blk) && + !fw_block_mem(&gmu->dcache, blk) && + !fw_block_mem(&gmu->dummy, blk)) { + DRM_DEV_ERROR(gmu->dev, + "failed to match fw block (addr=%.8x size=%d data[0]=%.8x)\n", + blk->addr, blk->size, blk->data[0]); + } + } + + ver = gmu_read(gmu, REG_A6XX_GMU_CORE_FW_VERSION); + DRM_INFO_ONCE("Loaded GMU firmware v%u.%u.%u\n", + FIELD_GET(A6XX_GMU_CORE_FW_VERSION_MAJOR__MASK, ver), + FIELD_GET(A6XX_GMU_CORE_FW_VERSION_MINOR__MASK, ver), + FIELD_GET(A6XX_GMU_CORE_FW_VERSION_STEP__MASK, ver)); + + return 0; +} + +static int a6xx_gmu_fw_start(struct a6xx_gmu *gmu, unsigned int state) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + const struct a6xx_info *a6xx_info = adreno_gpu->info->a6xx; + const struct adreno_reglist *gbif_cx = a6xx_info->gbif_cx; + u32 fence_range_lower, fence_range_upper; + u32 chipid = 0; + int ret; + + /* Vote veto for FAL10 */ + if (adreno_is_a650_family(adreno_gpu) || adreno_is_a7xx(adreno_gpu)) { + gmu_write(gmu, REG_A6XX_GPU_GMU_CX_GMU_CX_FALNEXT_INTF, 1); + gmu_write(gmu, REG_A6XX_GPU_GMU_CX_GMU_CX_FAL_INTF, 1); + } else if (adreno_is_a8xx(adreno_gpu)) { + gmu_write(gmu, REG_A8XX_GPU_GMU_CX_GMU_CX_FALNEXT_INTF, 1); + gmu_write(gmu, REG_A8XX_GPU_GMU_CX_GMU_CX_FAL_INTF, 1); + } + + /* Turn on TCM (Tightly Coupled Memory) retention */ + if (adreno_is_a7xx(adreno_gpu)) + a6xx_cx_misc_write(a6xx_gpu, REG_A7XX_CX_MISC_TCM_RET_CNTL, 1); + else if (!adreno_is_a8xx(adreno_gpu)) + gmu_write(gmu, REG_A6XX_GMU_GENERAL_7, 1); + + ret = a6xx_rpmh_start(gmu); + if (ret) + return ret; + + if (state == GMU_COLD_BOOT) { + if (WARN(!adreno_gpu->fw[ADRENO_FW_GMU], + "GMU firmware is not loaded\n")) + return -ENOENT; + + ret = a6xx_gmu_fw_load(gmu); + if (ret) + return ret; + } + + /* Clear init result to make sure we are getting a fresh value */ + gmu_write(gmu, REG_A6XX_GMU_CM3_FW_INIT_RESULT, 0); + gmu_write(gmu, REG_A6XX_GMU_CM3_BOOT_CONFIG, 0x02); + + /* Write the iova of the HFI table */ + gmu_write(gmu, REG_A6XX_GMU_HFI_QTBL_ADDR, gmu->hfi.iova); + gmu_write(gmu, REG_A6XX_GMU_HFI_QTBL_INFO, 1); + + if (adreno_is_a8xx(adreno_gpu)) { + fence_range_upper = 0x32; + fence_range_lower = 0x8c0; + } else if (adreno_is_a7xx(adreno_gpu)) { + fence_range_upper = 0x32; + fence_range_lower = 0x8a0; + } else { + fence_range_upper = 0xa; + fence_range_lower = 0xa0; + } + + gmu_write(gmu, REG_A6XX_GMU_AHB_FENCE_RANGE_0, + BIT(31) | + FIELD_PREP(GENMASK(30, 18), fence_range_upper) | + FIELD_PREP(GENMASK(17, 0), fence_range_lower)); + + /* + * Snapshots toggle the NMI bit which will result in a jump to the NMI + * handler instead of __main. Set the M3 config value to avoid that. + */ + gmu_write(gmu, REG_A6XX_GMU_CM3_CFG, 0x4052); + + if (a6xx_info->gmu_chipid) { + chipid = a6xx_info->gmu_chipid; + } else { + /* + * Note that the GMU has a slightly different layout for + * chip_id, for whatever reason, so a bit of massaging + * is needed. The upper 16b are the same, but minor and + * patchid are packed in four bits each with the lower + * 8b unused: + */ + chipid = adreno_gpu->chip_id & 0xffff0000; + chipid |= (adreno_gpu->chip_id << 4) & 0xf000; /* minor */ + chipid |= (adreno_gpu->chip_id << 8) & 0x0f00; /* patchid */ + } + + if (adreno_is_a8xx(adreno_gpu)) { + gmu_write(gmu, REG_A8XX_GMU_GENERAL_10, chipid); + gmu_write(gmu, REG_A8XX_GMU_GENERAL_8, + (gmu->log.iova & GENMASK(31, 12)) | + ((gmu->log.size / SZ_4K - 1) & GENMASK(7, 0))); + } else if (adreno_is_a7xx(adreno_gpu)) { + gmu_write(gmu, REG_A7XX_GMU_GENERAL_10, chipid); + gmu_write(gmu, REG_A7XX_GMU_GENERAL_8, + (gmu->log.iova & GENMASK(31, 12)) | + ((gmu->log.size / SZ_4K - 1) & GENMASK(7, 0))); + } else { + gmu_write(gmu, REG_A6XX_GMU_HFI_SFR_ADDR, chipid); + + gmu_write(gmu, REG_A6XX_GPU_GMU_CX_GMU_PWR_COL_CP_MSG, + gmu->log.iova | (gmu->log.size / SZ_4K - 1)); + } + + /* For A7x and newer, do the CX GBIF configurations before GMU wake up */ + for (int i = 0; (gbif_cx && gbif_cx[i].offset); i++) + gpu_write(gpu, gbif_cx[i].offset, gbif_cx[i].value); + + if (adreno_is_a8xx(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_GBIF_CX_CONFIG, 0x20023000); + gmu_write(gmu, REG_A6XX_GMU_MRC_GBIF_QOS_CTRL, 0x33); + } else if (adreno_is_a722(adreno_gpu)) + gpu_rmw(gpu, REG_A6XX_GBIF_CX_CONFIG, GENMASK(31, 29), + FIELD_PREP(GENMASK(31, 29), 2)); + + /* Set up the lowest idle level on the GMU */ + a6xx_gmu_power_config(gmu); + + ret = a6xx_gmu_start(gmu); + if (ret) + return ret; + + if (gmu->legacy) { + ret = a6xx_gmu_gfx_rail_on(gmu); + if (ret) + return ret; + + ret = a6xx_sptprac_enable(gmu); + if (ret) + return ret; + } + + ret = a6xx_gmu_hfi_start(gmu); + if (ret) + return ret; + + /* FIXME: Do we need this wmb() here? */ + wmb(); + + return 0; +} + +#define A6XX_HFI_IRQ_MASK \ + (A6XX_GMU_GMU2HOST_INTR_INFO_CM3_FAULT) + +#define A6XX_GMU_IRQ_MASK \ + (A6XX_GMU_AO_HOST_INTERRUPT_STATUS_WDOG_BITE | \ + A6XX_GMU_AO_HOST_INTERRUPT_STATUS_HOST_AHB_BUS_ERROR | \ + A6XX_GMU_AO_HOST_INTERRUPT_STATUS_FENCE_ERR) + +static void a6xx_gmu_irq_disable(struct a6xx_gmu *gmu) +{ + disable_irq(gmu->gmu_irq); + disable_irq(gmu->hfi_irq); + + gmu_write(gmu, REG_A6XX_GMU_AO_HOST_INTERRUPT_MASK, ~0); + gmu_write(gmu, REG_A6XX_GMU_GMU2HOST_INTR_MASK, ~0); +} + +static void a6xx_gmu_rpmh_off(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + u32 val, seqmem_off = 0; + + /* The second spin of A7xx GPUs messed with some register offsets.. */ + if (adreno_is_a740_family(adreno_gpu) || adreno_is_a722(adreno_gpu) || + adreno_is_a8xx(adreno_gpu)) + seqmem_off = 4; + + /* Make sure there are no outstanding RPMh votes */ + gmu_poll_timeout_rscc(gmu, REG_A6XX_RSCC_TCS0_DRV0_STATUS + seqmem_off, + val, (val & 1), 100, 10000); + gmu_poll_timeout_rscc(gmu, REG_A6XX_RSCC_TCS1_DRV0_STATUS + seqmem_off, + val, (val & 1), 100, 10000); + gmu_poll_timeout_rscc(gmu, REG_A6XX_RSCC_TCS2_DRV0_STATUS + seqmem_off, + val, (val & 1), 100, 10000); + gmu_poll_timeout_rscc(gmu, REG_A6XX_RSCC_TCS3_DRV0_STATUS + seqmem_off, + val, (val & 1), 100, 1000); + + if (!adreno_is_a740_family(adreno_gpu) && !adreno_is_a722(adreno_gpu) && + !adreno_is_a8xx(adreno_gpu)) + return; + + gmu_poll_timeout_rscc(gmu, REG_A7XX_RSCC_TCS4_DRV0_STATUS + seqmem_off, + val, (val & 1), 100, 10000); + gmu_poll_timeout_rscc(gmu, REG_A7XX_RSCC_TCS5_DRV0_STATUS + seqmem_off, + val, (val & 1), 100, 10000); + gmu_poll_timeout_rscc(gmu, REG_A7XX_RSCC_TCS6_DRV0_STATUS + seqmem_off, + val, (val & 1), 100, 10000); + gmu_poll_timeout_rscc(gmu, REG_A7XX_RSCC_TCS7_DRV0_STATUS + seqmem_off, + val, (val & 1), 100, 1000); + gmu_poll_timeout_rscc(gmu, REG_A7XX_RSCC_TCS8_DRV0_STATUS + seqmem_off, + val, (val & 1), 100, 10000); + gmu_poll_timeout_rscc(gmu, REG_A7XX_RSCC_TCS9_DRV0_STATUS + seqmem_off, + val, (val & 1), 100, 1000); +} + +/* Force the GMU off in case it isn't responsive */ +static void a6xx_gmu_force_off(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + + /* + * Turn off keep alive that might have been enabled by the hang + * interrupt + */ + if (adreno_is_a8xx(adreno_gpu)) + gmu_write(&a6xx_gpu->gmu, REG_A8XX_GMU_GMU_PWR_COL_KEEPALIVE, 0); + else + gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_GMU_PWR_COL_KEEPALIVE, 0); + + /* Flush all the queues */ + a6xx_hfi_stop(gmu); + + /* Stop the interrupts */ + a6xx_gmu_irq_disable(gmu); + + /* Force off SPTP in case the GMU is managing it */ + a6xx_sptprac_disable(gmu); + + a6xx_gemnoc_workaround(gmu); + + /* Make sure there are no outstanding RPMh votes */ + a6xx_gmu_rpmh_off(gmu); + + /* Clear the WRITEDROPPED fields and put fence into allow mode */ + gmu_write(gmu, REG_A6XX_GMU_AHB_FENCE_STATUS_CLR, 0x7); + gmu_write(gmu, REG_A6XX_GMU_AO_AHB_FENCE_CTRL, 0); + + /* Make sure the above writes go through */ + wmb(); + + /* Halt the gmu cm3 core */ + gmu_write(gmu, REG_A6XX_GMU_CM3_SYSRESET, 1); + + adreno_gpu->funcs->bus_halt(adreno_gpu, true); + + /* Reset GPU core blocks */ + a6xx_gpu_sw_reset(gpu, true); + + a6xx_rpmh_stop(gmu); +} + +static void a6xx_gmu_set_initial_freq(struct msm_gpu *gpu, struct a6xx_gmu *gmu) +{ + struct dev_pm_opp *gpu_opp; + unsigned long gpu_freq = gmu->gpu_freqs[gmu->current_perf_index]; + + gpu_opp = dev_pm_opp_find_freq_exact(&gpu->pdev->dev, gpu_freq, true); + if (IS_ERR(gpu_opp)) + return; + + gmu->freq = 0; /* so a6xx_gmu_set_freq() doesn't exit early */ + a6xx_gmu_set_freq(gpu, gpu_opp, false); + dev_pm_opp_put(gpu_opp); +} + +static void a6xx_gmu_set_initial_bw(struct msm_gpu *gpu, struct a6xx_gmu *gmu) +{ + struct dev_pm_opp *gpu_opp; + unsigned long gpu_freq = gmu->gpu_freqs[gmu->current_perf_index]; + + gpu_opp = dev_pm_opp_find_freq_exact(&gpu->pdev->dev, gpu_freq, true); + if (IS_ERR(gpu_opp)) + return; + + dev_pm_opp_set_opp(&gpu->pdev->dev, gpu_opp); + dev_pm_opp_put(gpu_opp); +} + +static int a6xx_gmu_secure_init(struct a6xx_gpu *a6xx_gpu) +{ + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + u32 fuse_val; + int ret; + + if (test_bit(GMU_STATUS_SECURE_INIT, &gmu->status)) + return 0; + + if (adreno_is_a750(adreno_gpu) || adreno_is_a8xx(adreno_gpu)) { + /* + * Assume that if qcom scm isn't available, that whatever + * replacement allows writing the fuse register ourselves. + * Users of alternative firmware need to make sure this + * register is writeable or indicate that it's not somehow. + * Print a warning because if you mess this up you're about to + * crash horribly. + */ + if (!qcom_scm_is_available()) { + dev_warn_once(gpu->dev->dev, + "SCM is not available, poking fuse register\n"); + a6xx_cx_misc_write(a6xx_gpu, REG_A7XX_CX_MISC_SW_FUSE_VALUE, + A7XX_CX_MISC_SW_FUSE_VALUE_RAYTRACING | + A7XX_CX_MISC_SW_FUSE_VALUE_FASTBLEND | + A7XX_CX_MISC_SW_FUSE_VALUE_LPAC); + adreno_gpu->has_ray_tracing = true; + goto done; + } + + ret = qcom_scm_gpu_init_regs(QCOM_SCM_GPU_ALWAYS_EN_REQ | + QCOM_SCM_GPU_TSENSE_EN_REQ); + if (ret) { + dev_warn_once(gpu->dev->dev, + "SCM call failed\n"); + return ret; + } + + /* + * On A7XX_GEN3 and newer, raytracing may be disabled by the + * firmware, find out whether that's the case. The scm call + * above sets the fuse register. + */ + fuse_val = a6xx_cx_misc_read(a6xx_gpu, + REG_A7XX_CX_MISC_SW_FUSE_VALUE); + adreno_gpu->has_ray_tracing = + !!(fuse_val & A7XX_CX_MISC_SW_FUSE_VALUE_RAYTRACING); + } else if (adreno_is_a740(adreno_gpu)) { + /* Raytracing is always enabled on a740 */ + adreno_gpu->has_ray_tracing = true; + } + +done: + set_bit(GMU_STATUS_SECURE_INIT, &gmu->status); + return 0; +} + +static int a6xx_gmu_gxpd_get(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + + if (IS_ERR_OR_NULL(gmu->gxpd)) + return 0; + + /* + * On A8xx HW, GX GDSC is moved to a new clk controller block under GX + * power domain. The clock driver for this new block keeps the GX rail + * voted when gxpd is voted. So, use the gxpd only during gpu recovery. + */ + if (adreno_gpu->info->family >= ADRENO_8XX_GEN1) + return 0; + + /* + * On A6x/A7x, "enable" the GX power domain which won't actually do + * anything but it will make sure that the refcounting is correct in + * case we need to bring down the GX after a GMU failure + */ + return pm_runtime_get_sync(gmu->gxpd); +} + +static int a6xx_gmu_gxpd_put(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + + if (IS_ERR_OR_NULL(gmu->gxpd)) + return 0; + + if (adreno_gpu->info->family < ADRENO_8XX_GEN1) + return pm_runtime_put_sync(gmu->gxpd); + + /* + * On A8x, GX GDSC collapse should be triggered only when it is stuck ON + */ + if (adreno_gpu->funcs->gx_is_on(adreno_gpu)) { + pm_runtime_get_sync(gmu->gxpd); + /* + * Hint to gfxclkctl driver to do a hw collapse during the next + * RPM PUT. This is a special behavior in the gfxclkctl driver + */ + dev_pm_genpd_synced_poweroff(gmu->gxpd); + pm_runtime_put_sync(gmu->gxpd); + } + + return 0; +} + +int a6xx_gmu_resume(struct a6xx_gpu *a6xx_gpu) +{ + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + int status, ret; + + if (WARN(!gmu->initialized, "The GMU is not set up yet\n")) + return -EINVAL; + + gmu->hung = false; + + /* Turn on the resources */ + pm_runtime_get_sync(gmu->dev); + + a6xx_gmu_gxpd_get(gmu); + + /* Use a known rate to bring up the GMU */ + clk_set_rate(gmu->core_clk, 200000000); + clk_set_rate(gmu->hub_clk, adreno_is_a740_family(adreno_gpu) ? + 200000000 : 150000000); + ret = clk_bulk_prepare_enable(gmu->nr_clocks, gmu->clocks); + if (ret) + goto rpm_put; + + ret = a6xx_gmu_secure_init(a6xx_gpu); + if (ret) + goto disable_clk; + + /* Read the slice info on A8x GPUs */ + a8xx_gpu_get_slice_info(gpu); + + /* Set the bus quota to a reasonable value for boot */ + a6xx_gmu_set_initial_bw(gpu, gmu); + + /* Enable the GMU interrupt */ + gmu_write(gmu, REG_A6XX_GMU_AO_HOST_INTERRUPT_CLR, ~0); + gmu_write(gmu, REG_A6XX_GMU_AO_HOST_INTERRUPT_MASK, ~A6XX_GMU_IRQ_MASK); + enable_irq(gmu->gmu_irq); + + /* Check to see if we are doing a cold or warm boot */ + if (adreno_is_a7xx(adreno_gpu) || adreno_is_a8xx(adreno_gpu)) { + status = a6xx_cx_misc_read(a6xx_gpu, REG_A7XX_CX_MISC_TCM_RET_CNTL) == 1 ? + GMU_WARM_BOOT : GMU_COLD_BOOT; + } else if (gmu->legacy) { + status = gmu_read(gmu, REG_A6XX_GMU_GENERAL_7) == 1 ? + GMU_WARM_BOOT : GMU_COLD_BOOT; + } else { + /* + * Warm boot path does not work on newer A6xx GPUs + * Presumably this is because icache/dcache regions must be restored + */ + status = GMU_COLD_BOOT; + } + + ret = a6xx_gmu_fw_start(gmu, status); + if (ret) + goto disable_irq; + + ret = a6xx_hfi_start(gmu, status); + if (ret) + goto disable_irq; + + /* + * Turn on the GMU firmware fault interrupt after we know the boot + * sequence is successful + */ + gmu_write(gmu, REG_A6XX_GMU_GMU2HOST_INTR_CLR, ~0); + gmu_write(gmu, REG_A6XX_GMU_GMU2HOST_INTR_MASK, ~A6XX_HFI_IRQ_MASK); + enable_irq(gmu->hfi_irq); + + /* Set the GPU to the current freq */ + a6xx_gmu_set_initial_freq(gpu, gmu); + + return 0; + +disable_irq: + disable_irq(gmu->gmu_irq); + a6xx_rpmh_stop(gmu); +disable_clk: + clk_bulk_disable_unprepare(gmu->nr_clocks, gmu->clocks); +rpm_put: + a6xx_gmu_gxpd_put(gmu); + + pm_runtime_put(gmu->dev); + + return ret; +} + +bool a6xx_gmu_isidle(struct a6xx_gmu *gmu) +{ + u32 reg; + + if (!gmu->initialized) + return true; + + reg = gmu_read(gmu, REG_A6XX_GPU_GMU_AO_GPU_CX_BUSY_STATUS); + + if (reg & A6XX_GPU_GMU_AO_GPU_CX_BUSY_STATUS_GPUBUSYIGNAHB) + return false; + + return true; +} + +/* Gracefully try to shut down the GMU and by extension the GPU */ +static void a6xx_gmu_shutdown(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + u32 val; + int ret; + + /* + * GMU firmware's internal power state gets messed up if we send "prepare_slumber" hfi when + * oob_gpu handshake wasn't done after the last wake up. So do a dummy handshake here when + * required + */ + if (adreno_gpu->base.needs_hw_init) { + if (a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET)) + goto force_off; + + a6xx_gmu_clear_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET); + } + + if (test_and_clear_bit(GMU_STATUS_OOB_PERF_SET, &gmu->status)) + a6xx_gmu_clear_oob(gmu, GMU_OOB_PERFCOUNTER_SET); + + ret = a6xx_gmu_wait_for_idle(gmu); + + /* If the GMU isn't responding assume it is hung */ + if (ret) + goto force_off; + + adreno_gpu->funcs->bus_halt(adreno_gpu, a6xx_gpu->hung); + + /* tell the GMU we want to slumber */ + ret = a6xx_gmu_notify_slumber(gmu); + if (ret) + goto force_off; + + ret = gmu_poll_timeout(gmu, + REG_A6XX_GPU_GMU_AO_GPU_CX_BUSY_STATUS, val, + !(val & A6XX_GPU_GMU_AO_GPU_CX_BUSY_STATUS_GPUBUSYIGNAHB), + 100, 10000); + + /* + * Let the user know we failed to slumber but don't worry too + * much because we are powering down anyway + */ + + if (ret) + DRM_DEV_ERROR(gmu->dev, + "Unable to slumber GMU: status = 0%x/0%x\n", + gmu_read(gmu, + REG_A6XX_GPU_GMU_AO_GPU_CX_BUSY_STATUS), + gmu_read(gmu, + REG_A6XX_GPU_GMU_AO_GPU_CX_BUSY_STATUS2)); + + /* Turn off HFI */ + a6xx_hfi_stop(gmu); + + /* Stop the interrupts and mask the hardware */ + a6xx_gmu_irq_disable(gmu); + + /* Halt the gmu cm3 core */ + gmu_write(gmu, REG_A6XX_GMU_CM3_SYSRESET, 1); + + /* Tell RPMh to power off the GPU */ + a6xx_rpmh_stop(gmu); + + return; + +force_off: + a6xx_gmu_force_off(gmu); +} + + +int a6xx_gmu_stop(struct a6xx_gpu *a6xx_gpu) +{ + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + struct msm_gpu *gpu = &a6xx_gpu->base.base; + + if (!pm_runtime_active(gmu->dev)) + return 0; + + /* + * Force the GMU off if we detected a hang, otherwise try to shut it + * down gracefully + */ + if (gmu->hung) + a6xx_gmu_force_off(gmu); + else + a6xx_gmu_shutdown(gmu); + + /* Remove the bus vote */ + dev_pm_opp_set_opp(&gpu->pdev->dev, NULL); + + /* + * Make sure the GX domain is off before turning off the GMU (CX) + * domain. Usually the GMU does this but only if the shutdown sequence + * was successful + */ + a6xx_gmu_gxpd_put(gmu); + + clk_bulk_disable_unprepare(gmu->nr_clocks, gmu->clocks); + + pm_runtime_put_sync(gmu->dev); + + return 0; +} + +static void a6xx_gmu_memory_free(struct a6xx_gmu *gmu) +{ + struct msm_mmu *mmu = to_msm_vm(gmu->vm)->mmu; + + msm_gem_kernel_put(gmu->hfi.obj, gmu->vm); + msm_gem_kernel_put(gmu->debug.obj, gmu->vm); + msm_gem_kernel_put(gmu->icache.obj, gmu->vm); + msm_gem_kernel_put(gmu->dcache.obj, gmu->vm); + msm_gem_kernel_put(gmu->dummy.obj, gmu->vm); + msm_gem_kernel_put(gmu->log.obj, gmu->vm); + + mmu->funcs->detach(mmu); + drm_gpuvm_put(gmu->vm); +} + +static int a6xx_gmu_memory_alloc(struct a6xx_gmu *gmu, struct a6xx_gmu_bo *bo, + size_t size, u64 iova, const char *name) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct drm_device *dev = a6xx_gpu->base.base.dev; + uint32_t flags = MSM_BO_WC; + u64 range_start, range_end; + int ret; + + size = PAGE_ALIGN(size); + if (!iova) { + /* no fixed address - use GMU's uncached range */ + range_start = 0x60000000 + PAGE_SIZE; /* skip dummy page */ + range_end = 0x80000000; + } else { + /* range for fixed address */ + range_start = iova; + range_end = iova + size; + /* use IOMMU_PRIV for icache/dcache */ + flags |= MSM_BO_MAP_PRIV; + } + + bo->obj = msm_gem_new(dev, size, flags, NULL); + if (IS_ERR(bo->obj)) + return PTR_ERR(bo->obj); + + ret = msm_gem_get_and_pin_iova_range(bo->obj, gmu->vm, &bo->iova, + range_start, range_end); + if (ret) { + drm_gem_object_put(bo->obj); + return ret; + } + + bo->virt = msm_gem_get_vaddr(bo->obj); + bo->size = size; + + msm_gem_object_set_name(bo->obj, "%s", name); + + return 0; +} + +static int a6xx_gmu_memory_probe(struct drm_device *drm, struct a6xx_gmu *gmu) +{ + struct msm_mmu *mmu; + + mmu = msm_iommu_new(gmu->dev, 0); + if (IS_ERR(mmu)) + return PTR_ERR(mmu); + + gmu->vm = msm_gem_vm_create(drm, mmu, "gmu", 0x0, 0x80000000, true); + if (IS_ERR(gmu->vm)) + return PTR_ERR(gmu->vm); + + return 0; +} + +/** + * struct bcm_db - Auxiliary data pertaining to each Bus Clock Manager (BCM) + * @unit: divisor used to convert bytes/sec bw value to an RPMh msg + * @width: multiplier used to convert bytes/sec bw value to an RPMh msg + * @vcd: virtual clock domain that this bcm belongs to + * @reserved: reserved field + */ +struct bcm_db { + __le32 unit; + __le16 width; + u8 vcd; + u8 reserved; +}; + +static int a6xx_gmu_rpmh_bw_votes_init(struct adreno_gpu *adreno_gpu, + const struct a6xx_info *info, + struct a6xx_gmu *gmu) +{ + const struct bcm_db *bcm_data[GMU_MAX_BCMS] = { 0 }; + unsigned int bcm_index, bw_index, bcm_count = 0; + + /* Retrieve BCM data from cmd-db */ + for (bcm_index = 0; bcm_index < GMU_MAX_BCMS; bcm_index++) { + const struct a6xx_bcm *bcm = &info->bcms[bcm_index]; + size_t count; + + /* Stop at NULL terminated bcm entry */ + if (!bcm->name) + break; + + bcm_data[bcm_index] = cmd_db_read_aux_data(bcm->name, &count); + if (IS_ERR(bcm_data[bcm_index])) + return PTR_ERR(bcm_data[bcm_index]); + + if (!count) { + dev_err(gmu->dev, "invalid BCM '%s' aux data size\n", + bcm->name); + return -EINVAL; + } + + bcm_count++; + } + + /* Generate BCM votes values for each bandwidth & BCM */ + for (bw_index = 0; bw_index < gmu->nr_gpu_bws; bw_index++) { + u32 *data = gmu->gpu_ib_votes[bw_index]; + u32 bw = gmu->gpu_bw_table[bw_index]; + + /* Calculations loosely copied from bcm_aggregate() & tcs_cmd_gen() */ + for (bcm_index = 0; bcm_index < bcm_count; bcm_index++) { + const struct a6xx_bcm *bcm = &info->bcms[bcm_index]; + bool commit = false; + u64 peak; + u32 vote; + + if (bcm_index == bcm_count - 1 || + (bcm_data[bcm_index + 1] && + bcm_data[bcm_index]->vcd != bcm_data[bcm_index + 1]->vcd)) + commit = true; + + if (!bw) { + data[bcm_index] = BCM_TCS_CMD(commit, false, 0, 0); + continue; + } + + if (bcm->fixed) { + u32 perfmode = 0; + + /* GMU on A6xx votes perfmode on all valid bandwidth */ + if (!adreno_is_a7xx(adreno_gpu) || + (bcm->perfmode_bw && bw >= bcm->perfmode_bw)) + perfmode = bcm->perfmode; + + data[bcm_index] = BCM_TCS_CMD(commit, true, 0, perfmode); + continue; + } + + /* Multiply the bandwidth by the width of the connection */ + peak = (u64)bw * le16_to_cpu(bcm_data[bcm_index]->width); + do_div(peak, bcm->buswidth); + + /* Input bandwidth value is in KBps, scale the value to BCM unit */ + peak *= 1000; + do_div(peak, le32_to_cpu(bcm_data[bcm_index]->unit)); + + vote = clamp(peak, 1, BCM_TCS_CMD_VOTE_MASK); + + /* GMUs on A7xx votes on both x & y */ + if (adreno_is_a7xx(adreno_gpu) || adreno_is_a8xx(adreno_gpu)) + data[bcm_index] = BCM_TCS_CMD(commit, true, vote, vote); + else + data[bcm_index] = BCM_TCS_CMD(commit, true, 0, vote); + } + } + + return 0; +} + +/* Return the 'arc-level' for the given frequency */ +static unsigned int a6xx_gmu_get_arc_level(struct device *dev, + unsigned long freq) +{ + struct dev_pm_opp *opp; + unsigned int val; + + if (!freq) + return 0; + + opp = dev_pm_opp_find_freq_exact(dev, freq, true); + if (IS_ERR(opp)) + return 0; + + val = dev_pm_opp_get_level(opp); + + dev_pm_opp_put(opp); + + return val; +} + +static int a6xx_gmu_rpmh_arc_votes_init(struct device *dev, u32 *votes, + unsigned long *freqs, int freqs_count, + const char *pri_id, const char *sec_id) +{ + int i, j; + const u16 *pri, *sec; + size_t pri_count, sec_count; + + pri = cmd_db_read_aux_data(pri_id, &pri_count); + if (IS_ERR(pri)) + return PTR_ERR(pri); + /* + * The data comes back as an array of unsigned shorts so adjust the + * count accordingly + */ + pri_count >>= 1; + if (!pri_count) + return -EINVAL; + + sec = cmd_db_read_aux_data(sec_id, &sec_count); + if (IS_ERR(sec)) + return PTR_ERR(sec); + + sec_count >>= 1; + if (!sec_count) + return -EINVAL; + + /* Construct a vote for each frequency */ + for (i = 0; i < freqs_count; i++) { + u8 pindex = 0, sindex = 0; + unsigned int level = a6xx_gmu_get_arc_level(dev, freqs[i]); + + /* Get the primary index that matches the arc level */ + for (j = 0; j < pri_count; j++) { + if (pri[j] >= level) { + pindex = j; + break; + } + } + + if (j == pri_count) { + DRM_DEV_ERROR(dev, + "Level %u not found in the RPMh list\n", + level); + DRM_DEV_ERROR(dev, "Available levels:\n"); + for (j = 0; j < pri_count; j++) + DRM_DEV_ERROR(dev, " %u\n", pri[j]); + + return -EINVAL; + } + + /* + * Look for a level in in the secondary list that matches. If + * nothing fits, use the maximum non zero vote + */ + + for (j = 0; j < sec_count; j++) { + if (sec[j] >= level) { + sindex = j; + break; + } else if (sec[j]) { + sindex = j; + } + } + + /* Construct the vote */ + votes[i] = ((pri[pindex] & 0xffff) << 16) | + (sindex << 8) | pindex; + } + + return 0; +} + +static int a6xx_gmu_rpmh_dep_votes_init(struct device *dev, u32 *votes, + unsigned long *freqs, int freqs_count) +{ + const u16 *mx; + size_t count; + + mx = cmd_db_read_aux_data("mx.lvl", &count); + if (IS_ERR(mx)) + return PTR_ERR(mx); + /* + * The data comes back as an array of unsigned shorts so adjust the + * count accordingly + */ + count >>= 1; + if (!count) + return -EINVAL; + + /* Fix the vote for zero frequency */ + votes[0] = 0xffffffff; + + /* Construct a vote for rest of the corners */ + for (int i = 1; i < freqs_count; i++) { + unsigned int level = a6xx_gmu_get_arc_level(dev, freqs[i]); + u8 j, index = 0; + + /* Get the primary index that matches the arc level */ + for (j = 0; j < count; j++) { + if (mx[j] >= level) { + index = j; + break; + } + } + + if (j == count) { + DRM_DEV_ERROR(dev, + "Mx Level %u not found in the RPMh list\n", + level); + DRM_DEV_ERROR(dev, "Available levels:\n"); + for (j = 0; j < count; j++) + DRM_DEV_ERROR(dev, " %u\n", mx[j]); + + return -EINVAL; + } + + /* Construct the vote */ + votes[i] = (0x3fff << 14) | (index << 8) | (0xff); + } + + return 0; +} + +/* + * The GMU votes with the RPMh for itself and on behalf of the GPU but we need + * to construct the list of votes on the CPU and send it over. Query the RPMh + * voltage levels and build the votes + * The GMU can also vote for DDR interconnects, use the OPP bandwidth entries + * and BCM parameters to build the votes. + */ + +static int a6xx_gmu_rpmh_votes_init(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + const struct a6xx_info *info = adreno_gpu->info->a6xx; + struct msm_gpu *gpu = &adreno_gpu->base; + const char *sec_id; + const u16 *gmxc; + int ret; + + gmxc = cmd_db_read_aux_data("gmxc.lvl", NULL); + if (gmxc == ERR_PTR(-EPROBE_DEFER)) + return -EPROBE_DEFER; + + /* If GMxC is present, prefer that as secondary rail for GX votes */ + sec_id = IS_ERR_OR_NULL(gmxc) ? "mx.lvl" : "gmxc.lvl"; + + /* Build the GX votes */ + ret = a6xx_gmu_rpmh_arc_votes_init(&gpu->pdev->dev, gmu->gx_arc_votes, + gmu->gpu_freqs, gmu->nr_gpu_freqs, "gfx.lvl", sec_id); + + /* Build the CX votes */ + ret |= a6xx_gmu_rpmh_arc_votes_init(gmu->dev, gmu->cx_arc_votes, + gmu->gmu_freqs, gmu->nr_gmu_freqs, "cx.lvl", "mx.lvl"); + + ret |= a6xx_gmu_rpmh_dep_votes_init(gmu->dev, gmu->dep_arc_votes, + gmu->gpu_freqs, gmu->nr_gpu_freqs); + + /* Build the interconnect votes */ + if (info->bcms && gmu->nr_gpu_bws > 1) + ret |= a6xx_gmu_rpmh_bw_votes_init(adreno_gpu, info, gmu); + + return ret; +} + +static int a6xx_gmu_build_freq_table(struct device *dev, unsigned long *freqs, + u32 size) +{ + int count = dev_pm_opp_get_opp_count(dev); + struct dev_pm_opp *opp; + int i, index = 0; + unsigned long freq = 1; + + /* + * The OPP table doesn't contain the "off" frequency level so we need to + * add 1 to the table size to account for it + */ + + if (WARN(count + 1 > size, + "The GMU frequency table is being truncated\n")) + count = size - 1; + + /* Set the "off" frequency */ + freqs[index++] = 0; + + for (i = 0; i < count; i++) { + opp = dev_pm_opp_find_freq_ceil(dev, &freq); + if (IS_ERR(opp)) + break; + + dev_pm_opp_put(opp); + freqs[index++] = freq++; + } + + return index; +} + +static int a6xx_gmu_build_bw_table(struct device *dev, unsigned long *bandwidths, + u32 size) +{ + int count = dev_pm_opp_get_opp_count(dev); + struct dev_pm_opp *opp; + int i, index = 0; + unsigned int bandwidth = 1; + + /* + * The OPP table doesn't contain the "off" bandwidth level so we need to + * add 1 to the table size to account for it + */ + + if (WARN(count + 1 > size, + "The GMU bandwidth table is being truncated\n")) + count = size - 1; + + /* Set the "off" bandwidth */ + bandwidths[index++] = 0; + + for (i = 0; i < count; i++) { + opp = dev_pm_opp_find_bw_ceil(dev, &bandwidth, 0); + if (IS_ERR(opp)) + break; + + dev_pm_opp_put(opp); + bandwidths[index++] = bandwidth++; + } + + return index; +} + +static int a6xx_gmu_pwrlevels_probe(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + const struct a6xx_info *info = adreno_gpu->info->a6xx; + struct msm_gpu *gpu = &adreno_gpu->base; + + int ret = 0; + + /* + * The GMU handles its own frequency switching so build a list of + * available frequencies to send during initialization + */ + ret = devm_pm_opp_of_add_table(gmu->dev); + if (ret) { + DRM_DEV_ERROR(gmu->dev, "Unable to set the OPP table for the GMU\n"); + return ret; + } + + gmu->nr_gmu_freqs = a6xx_gmu_build_freq_table(gmu->dev, + gmu->gmu_freqs, ARRAY_SIZE(gmu->gmu_freqs)); + + /* + * The GMU also handles GPU frequency switching so build a list + * from the GPU OPP table + */ + gmu->nr_gpu_freqs = a6xx_gmu_build_freq_table(&gpu->pdev->dev, + gmu->gpu_freqs, ARRAY_SIZE(gmu->gpu_freqs)); + + gmu->current_perf_index = gmu->nr_gpu_freqs - 1; + + /* + * The GMU also handles GPU Interconnect Votes so build a list + * of DDR bandwidths from the GPU OPP table + */ + if (info->bcms) + gmu->nr_gpu_bws = a6xx_gmu_build_bw_table(&gpu->pdev->dev, + gmu->gpu_bw_table, ARRAY_SIZE(gmu->gpu_bw_table)); + + /* Build the list of RPMh votes that we'll send to the GMU */ + return a6xx_gmu_rpmh_votes_init(gmu); +} + +static int a6xx_gmu_acd_probe(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct a6xx_hfi_acd_table *cmd = &gmu->acd_table; + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + int ret, i, cmd_idx = 0; + extern bool disable_acd; + + /* Skip ACD probe if requested via module param */ + if (disable_acd) { + DRM_DEV_ERROR(gmu->dev, "Skipping GPU ACD probe\n"); + return 0; + } + + cmd->version = 1; + cmd->stride = 1; + cmd->enable_by_level = 0; + + /* Skip freq = 0 and parse acd-level for rest of the OPPs */ + for (i = 1; i < gmu->nr_gpu_freqs; i++) { + struct dev_pm_opp *opp; + struct device_node *np; + unsigned long freq; + u32 val; + + freq = gmu->gpu_freqs[i]; + /* This is unlikely to fail because we are passing back a known freq */ + opp = dev_pm_opp_find_freq_exact(&gpu->pdev->dev, freq, true); + np = dev_pm_opp_get_of_node(opp); + + ret = of_property_read_u32(np, "qcom,opp-acd-level", &val); + of_node_put(np); + dev_pm_opp_put(opp); + if (ret == -EINVAL) + continue; + else if (ret) { + DRM_DEV_ERROR(gmu->dev, "Unable to read acd level for freq %lu\n", freq); + return ret; + } + + cmd->enable_by_level |= BIT(i); + cmd->data[cmd_idx++] = val; + } + + cmd->num_levels = cmd_idx; + + /* It is a problem if qmp node is unavailable when ACD is required */ + if (cmd->enable_by_level && IS_ERR_OR_NULL(gmu->qmp)) { + DRM_DEV_ERROR(gmu->dev, "Unable to send ACD state to AOSS\n"); + return -EINVAL; + } + + /* Otherwise, nothing to do if qmp is unavailable */ + if (IS_ERR_OR_NULL(gmu->qmp)) + return 0; + + /* + * Notify AOSS about the ACD state. AOSS is supposed to assume that ACD is disabled on + * system reset. So it is harmless if we couldn't notify 'OFF' state + */ + ret = qmp_send(gmu->qmp, "{class: gpu, res: acd, val: %d}", !!cmd->enable_by_level); + if (ret && cmd->enable_by_level) { + DRM_DEV_ERROR(gmu->dev, "Failed to send ACD state to AOSS\n"); + return ret; + } + + return 0; +} + +static int a6xx_gmu_clocks_probe(struct a6xx_gmu *gmu) +{ + int ret = devm_clk_bulk_get_all(gmu->dev, &gmu->clocks); + + if (ret < 1) + return ret; + + gmu->nr_clocks = ret; + + gmu->core_clk = msm_clk_bulk_get_clock(gmu->clocks, + gmu->nr_clocks, "gmu"); + + gmu->hub_clk = msm_clk_bulk_get_clock(gmu->clocks, + gmu->nr_clocks, "hub"); + + return 0; +} + +static int a6xx_gmu_get_irq(struct a6xx_gmu *gmu, struct platform_device *pdev, + const char *name, irq_handler_t handler) +{ + int irq, ret; + + irq = platform_get_irq_byname(pdev, name); + + ret = request_irq(irq, handler, IRQF_TRIGGER_HIGH | IRQF_NO_AUTOEN, name, gmu); + if (ret) { + DRM_DEV_ERROR(&pdev->dev, "Unable to get interrupt %s %d\n", + name, ret); + return ret; + } + + return irq; +} + +void a6xx_gmu_sysprof_setup(struct msm_gpu *gpu, bool force_on) +{ + bool sysprof = msm_gpu_sysprof_no_ifpc(gpu) || force_on; + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + + /* Nothing to do if GPU is suspended. We will handle this during GMU resume */ + if (!pm_runtime_get_if_active(&gpu->pdev->dev)) + return; + + mutex_lock(&gmu->lock); + + /* + * 'Perfcounter select' register values are lost during IFPC collapse. To avoid that, + * use the currently unused perfcounter oob vote to block IFPC when sysprof is active + */ + if (sysprof && !test_and_set_bit(GMU_STATUS_OOB_PERF_SET, &gmu->status)) + a6xx_gmu_set_oob(gmu, GMU_OOB_PERFCOUNTER_SET); + else if (!sysprof && test_and_clear_bit(GMU_STATUS_OOB_PERF_SET, &gmu->status)) + a6xx_gmu_clear_oob(gmu, GMU_OOB_PERFCOUNTER_SET); + + mutex_unlock(&gmu->lock); + + pm_runtime_put(&gpu->pdev->dev); +} + +void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu) +{ + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + + mutex_lock(&gmu->lock); + if (!gmu->initialized) { + mutex_unlock(&gmu->lock); + return; + } + + gmu->initialized = false; + + mutex_unlock(&gmu->lock); + + pm_runtime_force_suspend(gmu->dev); + + /* + * Since cxpd is a virt device, the devlink with gmu-dev will be removed + * automatically when we do detach + */ + dev_pm_domain_detach(gmu->cxpd, false); + + if (!IS_ERR_OR_NULL(gmu->gxpd)) { + pm_runtime_disable(gmu->gxpd); + dev_pm_domain_detach(gmu->gxpd, false); + } + + if (!IS_ERR_OR_NULL(gmu->qmp)) + qmp_put(gmu->qmp); + + iounmap(gmu->mmio); + gmu->mmio = NULL; + gmu->rscc = NULL; + + if (!adreno_has_gmu_wrapper(adreno_gpu) && + !adreno_has_rgmu(adreno_gpu)) { + a6xx_gmu_memory_free(gmu); + + free_irq(gmu->gmu_irq, gmu); + free_irq(gmu->hfi_irq, gmu); + } + + /* Drop reference taken in of_find_device_by_node */ + put_device(gmu->dev); +} + +static int cxpd_notifier_cb(struct notifier_block *nb, + unsigned long action, void *data) +{ + struct a6xx_gmu *gmu = container_of(nb, struct a6xx_gmu, pd_nb); + + if (action == GENPD_NOTIFY_OFF) + complete_all(&gmu->pd_gate); + + return 0; +} + +static void __iomem *a6xx_gmu_get_mmio(struct platform_device *pdev, resource_size_t *start) +{ + struct resource *res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + void __iomem *ret; + + if (!res) { + DRM_DEV_ERROR(&pdev->dev, "Unable to find the gmu core registers\n"); + return ERR_PTR(-EINVAL); + } + + ret = ioremap(res->start, resource_size(res)); + if (!ret) { + DRM_DEV_ERROR(&pdev->dev, "Unable to map the gmu core registers\n"); + return ERR_PTR(-EINVAL); + } + + if (start) + *start = res->start; + + return ret; +} + +int a6xx_gmu_wrapper_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) +{ + struct platform_device *pdev = of_find_device_by_node(node); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + resource_size_t start; + struct resource *res; + int ret; + + if (!pdev) + return -ENODEV; + + gmu->dev = &pdev->dev; + + ret = of_dma_configure(gmu->dev, node, true); + if (ret) + return ret; + + pm_runtime_enable(gmu->dev); + + /* Mark legacy for manual SPTPRAC control */ + gmu->legacy = true; + + /* RGMU requires clocks */ + ret = devm_clk_bulk_get_all(gmu->dev, &gmu->clocks); + if (ret < 0) + goto err_clk; + + gmu->nr_clocks = ret; + + /* Map the GMU registers */ + gmu->mmio = a6xx_gmu_get_mmio(pdev, &start); + if (IS_ERR(gmu->mmio)) { + ret = PTR_ERR(gmu->mmio); + goto err_mmio; + } + + res = platform_get_resource_byname(gpu->pdev, IORESOURCE_MEM, "kgsl_3d0_reg_memory"); + if (!res) { + ret = -EINVAL; + goto err_mmio; + } + + /* Identify gmu base offset from gpu base address */ + gmu->mmio_offset = (u32)(start - res->start); + + gmu->cxpd = dev_pm_domain_attach_by_name(gmu->dev, "cx"); + if (IS_ERR(gmu->cxpd)) { + ret = PTR_ERR(gmu->cxpd); + goto err_mmio; + } + + if (!device_link_add(gmu->dev, gmu->cxpd, DL_FLAG_PM_RUNTIME)) { + ret = -ENODEV; + goto detach_cxpd; + } + + init_completion(&gmu->pd_gate); + complete_all(&gmu->pd_gate); + gmu->pd_nb.notifier_call = cxpd_notifier_cb; + + /* Get a link to the GX power domain to reset the GPU */ + gmu->gxpd = dev_pm_domain_attach_by_name(gmu->dev, "gx"); + if (IS_ERR(gmu->gxpd)) { + ret = PTR_ERR(gmu->gxpd); + goto err_mmio; + } + + gmu->initialized = true; + + return 0; + +detach_cxpd: + dev_pm_domain_detach(gmu->cxpd, false); + +err_mmio: + iounmap(gmu->mmio); + +err_clk: + /* Drop reference taken in of_find_device_by_node */ + put_device(gmu->dev); + + return ret; +} + +int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) +{ + struct platform_device *pdev = of_find_device_by_node(node); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + struct device_link *link; + resource_size_t start; + struct resource *res; + int ret; + + if (!pdev) + return -ENODEV; + + gmu->dev = &pdev->dev; + + ret = of_dma_configure(gmu->dev, node, true); + if (ret) + return ret; + + /* Set GMU idle level */ + gmu->idle_level = (adreno_gpu->info->quirks & ADRENO_QUIRK_IFPC) ? + GMU_IDLE_STATE_IFPC : GMU_IDLE_STATE_ACTIVE; + + pm_runtime_enable(gmu->dev); + + /* Get the list of clocks */ + ret = a6xx_gmu_clocks_probe(gmu); + if (ret) + goto err_put_device; + + ret = a6xx_gmu_memory_probe(adreno_gpu->base.dev, gmu); + if (ret) + goto err_put_device; + + + /* A660 now requires handling "prealloc requests" in GMU firmware + * For now just hardcode allocations based on the known firmware. + * note: there is no indication that these correspond to "dummy" or + * "debug" regions, but this "guess" allows reusing these BOs which + * are otherwise unused by a660. + */ + gmu->dummy.size = SZ_4K; + if (adreno_is_a660_family(adreno_gpu) || + adreno_is_a7xx(adreno_gpu) || + adreno_is_a8xx(adreno_gpu)) { + ret = a6xx_gmu_memory_alloc(gmu, &gmu->debug, SZ_4K * 7, + 0x60400000, "debug"); + if (ret) + goto err_memory; + + gmu->dummy.size = SZ_16K; + } + + /* Allocate memory for the GMU dummy page */ + ret = a6xx_gmu_memory_alloc(gmu, &gmu->dummy, gmu->dummy.size, + 0x60000000, "dummy"); + if (ret) + goto err_memory; + + /* Note that a650 family also includes a660 family: */ + if (adreno_is_a650_family(adreno_gpu) || + adreno_is_a7xx(adreno_gpu) || + adreno_is_a8xx(adreno_gpu)) { + ret = a6xx_gmu_memory_alloc(gmu, &gmu->icache, + SZ_16M - SZ_16K, 0x04000, "icache"); + if (ret) + goto err_memory; + /* + * NOTE: when porting legacy ("pre-650-family") GPUs you may be tempted to add a condition + * to allocate icache/dcache here, as per downstream code flow, but it may not actually be + * necessary. If you omit this step and you don't get random pagefaults, you are likely + * good to go without this! + */ + } else if (adreno_is_a640_family(adreno_gpu)) { + ret = a6xx_gmu_memory_alloc(gmu, &gmu->icache, + SZ_256K - SZ_16K, 0x04000, "icache"); + if (ret) + goto err_memory; + + ret = a6xx_gmu_memory_alloc(gmu, &gmu->dcache, + SZ_256K - SZ_16K, 0x44000, "dcache"); + if (ret) + goto err_memory; + } else if (adreno_is_a630_family(adreno_gpu)) { + /* HFI v1, has sptprac */ + gmu->legacy = true; + + /* Allocate memory for the GMU debug region */ + ret = a6xx_gmu_memory_alloc(gmu, &gmu->debug, SZ_16K, 0, "debug"); + if (ret) + goto err_memory; + } + + /* Allocate memory for the GMU log region */ + ret = a6xx_gmu_memory_alloc(gmu, &gmu->log, SZ_16K, 0, "log"); + if (ret) + goto err_memory; + + /* Allocate memory for for the HFI queues */ + ret = a6xx_gmu_memory_alloc(gmu, &gmu->hfi, SZ_16K, 0, "hfi"); + if (ret) + goto err_memory; + + /* Map the GMU registers */ + gmu->mmio = a6xx_gmu_get_mmio(pdev, &start); + if (IS_ERR(gmu->mmio)) { + ret = PTR_ERR(gmu->mmio); + goto err_memory; + } + + res = platform_get_resource_byname(gpu->pdev, IORESOURCE_MEM, "kgsl_3d0_reg_memory"); + if (!res) { + ret = -EINVAL; + goto err_mmio; + } + + /* Identify gmu base offset from gpu base address */ + gmu->mmio_offset = (u32)(start - res->start); + + if (adreno_is_a650_family(adreno_gpu) || + adreno_is_a7xx(adreno_gpu)) { + gmu->rscc = devm_platform_ioremap_resource_byname(pdev, "rscc"); + if (IS_ERR(gmu->rscc)) { + ret = -ENODEV; + goto err_mmio; + } + } else if (adreno_is_a8xx(adreno_gpu)) { + /* + * On a8xx , RSCC lives at GPU base + 0x50000, which falls + * inside the GPU's kgsl_3d0_reg_memory range rather than the + * GMU's. + */ + gmu->rscc = gpu->mmio + 0x50000; + } else { + gmu->rscc = gmu->mmio + 0x23000; + } + + /* Get the HFI and GMU interrupts */ + gmu->hfi_irq = a6xx_gmu_get_irq(gmu, pdev, "hfi", a6xx_hfi_irq); + gmu->gmu_irq = a6xx_gmu_get_irq(gmu, pdev, "gmu", a6xx_gmu_irq); + + if (gmu->hfi_irq < 0 || gmu->gmu_irq < 0) { + ret = -ENODEV; + goto err_mmio; + } + + gmu->cxpd = dev_pm_domain_attach_by_name(gmu->dev, "cx"); + if (IS_ERR(gmu->cxpd)) { + ret = PTR_ERR(gmu->cxpd); + goto err_mmio; + } + + link = device_link_add(gmu->dev, gmu->cxpd, DL_FLAG_PM_RUNTIME); + if (!link) { + ret = -ENODEV; + goto detach_cxpd; + } + + /* Other errors are handled during GPU ACD probe */ + gmu->qmp = qmp_get(gmu->dev); + if (PTR_ERR_OR_ZERO(gmu->qmp) == -EPROBE_DEFER) { + ret = -EPROBE_DEFER; + goto detach_gxpd; + } + + init_completion(&gmu->pd_gate); + complete_all(&gmu->pd_gate); + gmu->pd_nb.notifier_call = cxpd_notifier_cb; + + /* + * Get a link to the GX power domain to reset the GPU in case of GMU + * crash + */ + gmu->gxpd = dev_pm_domain_attach_by_name(gmu->dev, "gx"); + + /* Get the power levels for the GMU and GPU */ + a6xx_gmu_pwrlevels_probe(gmu); + + ret = a6xx_gmu_acd_probe(gmu); + if (ret) + goto detach_gxpd; + + /* Set up the HFI queues */ + a6xx_hfi_init(gmu); + + /* Initialize RPMh */ + a6xx_gmu_rpmh_init(gmu); + + gmu->initialized = true; + + return 0; + +detach_gxpd: + if (!IS_ERR_OR_NULL(gmu->gxpd)) + dev_pm_domain_detach(gmu->gxpd, false); + + if (!IS_ERR_OR_NULL(gmu->qmp)) + qmp_put(gmu->qmp); + + device_link_del(link); + +detach_cxpd: + dev_pm_domain_detach(gmu->cxpd, false); + +err_mmio: + iounmap(gmu->mmio); + free_irq(gmu->gmu_irq, gmu); + free_irq(gmu->hfi_irq, gmu); + +err_memory: + a6xx_gmu_memory_free(gmu); +err_put_device: + /* Drop reference taken in of_find_device_by_node */ + put_device(gmu->dev); + + return ret; +} diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h new file mode 100644 index 000000000..3f96b10b5 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h @@ -0,0 +1,244 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2017 The Linux Foundation. All rights reserved. */ + +#ifndef _A6XX_GMU_H_ +#define _A6XX_GMU_H_ + +#include +#include +#include +#include +#include +#include "msm_drv.h" +#include "adreno_gpu.h" +#include "a6xx_hfi.h" + +struct a6xx_gmu_bo { + struct drm_gem_object *obj; + void *virt; + size_t size; + u64 iova; +}; + +#define GMU_MAX_GX_FREQS 32 +#define GMU_MAX_CX_FREQS 6 +#define GMU_MAX_BCMS 3 + +struct a6xx_bcm { + char *name; + unsigned int buswidth; + bool fixed; + unsigned int perfmode; + unsigned int perfmode_bw; +}; + +/* + * These define the different GMU wake up options - these define how both the + * CPU and the GMU bring up the hardware + */ + +/* THe GMU has already been booted and the rentention registers are active */ +#define GMU_WARM_BOOT 0 + +/* the GMU is coming up for the first time or back from a power collapse */ +#define GMU_COLD_BOOT 1 + +/* + * These define the level of control that the GMU has - the higher the number + * the more things that the GMU hardware controls on its own. + */ + +/* The GMU does not do any idle state management */ +#define GMU_IDLE_STATE_ACTIVE 0 + +/* Unknown power state. Not exposed by the firmware. For documentation purpose only */ +#define GMU_IDLE_STATE_RESERVED 1 + +/* The GMU manages SPTP power collapse */ +#define GMU_IDLE_STATE_SPTP 2 + +/* The GMU does automatic IFPC (intra-frame power collapse) */ +#define GMU_IDLE_STATE_IFPC 3 + +struct a6xx_gmu { + struct device *dev; + + /* For serializing communication with the GMU: */ + struct mutex lock; + + struct drm_gpuvm *vm; + + void __iomem *mmio; + u32 mmio_offset; + void __iomem *rscc; + + int hfi_irq; + int gmu_irq; + + struct device *gxpd; + struct device *cxpd; + + int idle_level; + + struct a6xx_gmu_bo hfi; + struct a6xx_gmu_bo debug; + struct a6xx_gmu_bo icache; + struct a6xx_gmu_bo dcache; + struct a6xx_gmu_bo dummy; + struct a6xx_gmu_bo log; + + int nr_clocks; + struct clk_bulk_data *clocks; + struct clk *core_clk; + struct clk *hub_clk; + + /* current performance index set externally */ + int current_perf_index; + + int nr_gpu_freqs; + unsigned long gpu_freqs[GMU_MAX_GX_FREQS]; + u32 gx_arc_votes[GMU_MAX_GX_FREQS]; + u32 dep_arc_votes[GMU_MAX_GX_FREQS]; + struct a6xx_hfi_acd_table acd_table; + + int nr_gpu_bws; + unsigned long gpu_bw_table[GMU_MAX_GX_FREQS]; + u32 gpu_ib_votes[GMU_MAX_GX_FREQS][GMU_MAX_BCMS]; + + int nr_gmu_freqs; + unsigned long gmu_freqs[GMU_MAX_CX_FREQS]; + u32 cx_arc_votes[GMU_MAX_CX_FREQS]; + + unsigned long freq; + + struct a6xx_hfi_queue queues[HFI_MAX_QUEUES]; + + bool initialized; + bool hung; + bool legacy; /* a618 or a630 */ + + /* For power domain callback */ + struct notifier_block pd_nb; + struct completion pd_gate; + + struct qmp *qmp; + struct a6xx_hfi_msg_bw_table *bw_table; + +/* To check if we can trigger sleep seq at PDC. Cleared in a6xx_rpmh_stop() */ +#define GMU_STATUS_FW_START 0 +/* To track if PDC sleep seq was done */ +#define GMU_STATUS_PDC_SLEEP 1 +/* To track Perfcounter OOB set status */ +#define GMU_STATUS_OOB_PERF_SET 2 +/* To track whether secure world init was done */ +#define GMU_STATUS_SECURE_INIT 3 + unsigned long status; +}; + +#define GMU_BYTE_OFFSET(gmu, offset) (((offset) << 2) - (gmu)->mmio_offset) + +static inline u32 gmu_read(struct a6xx_gmu *gmu, u32 offset) +{ + /* The 'offset' is based on GPU's start address. Adjust it */ + return readl(gmu->mmio + GMU_BYTE_OFFSET(gmu, offset)); +} + +static inline void gmu_write(struct a6xx_gmu *gmu, u32 offset, u32 value) +{ + writel(value, gmu->mmio + GMU_BYTE_OFFSET(gmu, offset)); +} + +static inline void +gmu_write_bulk(struct a6xx_gmu *gmu, u32 offset, const u32 *data, u32 size) +{ + memcpy_toio(gmu->mmio + GMU_BYTE_OFFSET(gmu, offset), data, size); + wmb(); +} + +static inline void gmu_rmw(struct a6xx_gmu *gmu, u32 reg, u32 mask, u32 or) +{ + u32 val = gmu_read(gmu, reg); + + val &= ~mask; + + gmu_write(gmu, reg, val | or); +} + +static inline u64 gmu_read64(struct a6xx_gmu *gmu, u32 lo, u32 hi) +{ + u64 val; + + val = gmu_read(gmu, lo); + val |= ((u64) gmu_read(gmu, hi) << 32); + + return val; +} + +#define gmu_poll_timeout(gmu, addr, val, cond, interval, timeout) \ + readl_poll_timeout((gmu)->mmio + (GMU_BYTE_OFFSET(gmu, addr)), val, \ + cond, interval, timeout) +#define gmu_poll_timeout_atomic(gmu, addr, val, cond, interval, timeout) \ + readl_poll_timeout_atomic((gmu)->mmio + (GMU_BYTE_OFFSET(gmu, addr)), val, cond, \ + interval, timeout) + +static inline u32 gmu_read_rscc(struct a6xx_gmu *gmu, u32 offset) +{ + return readl(gmu->rscc + (offset << 2)); +} + +static inline void gmu_write_rscc(struct a6xx_gmu *gmu, u32 offset, u32 value) +{ + writel(value, gmu->rscc + (offset << 2)); +} + +#define gmu_poll_timeout_rscc(gmu, addr, val, cond, interval, timeout) \ + readl_poll_timeout((gmu)->rscc + ((addr) << 2), val, cond, \ + interval, timeout) + +/* + * These are the available OOB (out of band requests) to the GMU where "out of + * band" means that the CPU talks to the GMU directly and not through HFI. + * Normally this works by writing a ITCM/DTCM register and then triggering a + * interrupt (the "request" bit) and waiting for an acknowledgment (the "ack" + * bit). The state is cleared by writing the "clear' bit to the GMU interrupt. + * + * These are used to force the GMU/GPU to stay on during a critical sequence or + * for hardware workarounds. + */ + +enum a6xx_gmu_oob_state { + /* + * Let the GMU know that a boot or slumber operation has started. The value in + * REG_A6XX_GMU_BOOT_SLUMBER_OPTION lets the GMU know which operation we are + * doing + */ + GMU_OOB_BOOT_SLUMBER = 0, + /* + * Let the GMU know to not turn off any GPU registers while the CPU is in a + * critical section + */ + GMU_OOB_GPU_SET, + /* + * Set a new power level for the GPU when the CPU is doing frequency scaling + */ + GMU_OOB_DCVS_SET, + /* + * Used to keep the GPU on for CPU-side reads of performance counters. + */ + GMU_OOB_PERFCOUNTER_SET, +}; + +void a6xx_hfi_init(struct a6xx_gmu *gmu); +int a6xx_hfi_start(struct a6xx_gmu *gmu, int boot_state); +void a6xx_hfi_stop(struct a6xx_gmu *gmu); +int a6xx_hfi_send_prep_slumber(struct a6xx_gmu *gmu); +int a6xx_hfi_set_freq(struct a6xx_gmu *gmu, u32 perf_index, u32 bw_index); + +bool a6xx_gmu_gx_is_on(struct adreno_gpu *adreno_gpu); +bool a7xx_gmu_gx_is_on(struct adreno_gpu *adreno_gpu); +bool a8xx_gmu_gx_is_on(struct adreno_gpu *adreno_gpu); +bool a6xx_gmu_sptprac_is_on(struct a6xx_gmu *gmu); +void a6xx_sptprac_disable(struct a6xx_gmu *gmu); +int a6xx_sptprac_enable(struct a6xx_gmu *gmu); + +#endif diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c new file mode 100644 index 000000000..081e79ea4 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -0,0 +1,2934 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2017-2019 The Linux Foundation. All rights reserved. */ + + +#include "msm_gem.h" +#include "msm_mmu.h" +#include "msm_gpu_trace.h" +#include "msm_perfcntr.h" +#include "a6xx_gpu.h" +#include "a6xx_gmu.xml.h" + +#include +#include +#include +#include + +#define GPU_PAS_ID 13 + +static u64 a6xx_gmu_get_timestamp(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + u64 count_hi, count_lo, temp; + + do { + if (adreno_is_a750_family(adreno_gpu)) { + count_hi = gmu_read(&a6xx_gpu->gmu, REG_A7XX_GMU_CX_AO_COUNTER_H); + count_lo = gmu_read(&a6xx_gpu->gmu, REG_A7XX_GMU_CX_AO_COUNTER_L); + temp = gmu_read(&a6xx_gpu->gmu, REG_A7XX_GMU_CX_AO_COUNTER_H); + } else { + count_hi = gmu_read(&a6xx_gpu->gmu, REG_A6XX_GMU_ALWAYS_ON_COUNTER_H); + count_lo = gmu_read(&a6xx_gpu->gmu, REG_A6XX_GMU_ALWAYS_ON_COUNTER_L); + temp = gmu_read(&a6xx_gpu->gmu, REG_A6XX_GMU_ALWAYS_ON_COUNTER_H); + } + } while (unlikely(count_hi != temp)); + + return (count_hi << 32) | count_lo; +} + +static bool fence_status_check(struct msm_gpu *gpu, u32 offset, u32 value, u32 status, u32 mask) +{ + /* Success if !writedropped0/1 */ + if (!(status & mask)) + return true; + + udelay(10); + + /* Try to update fenced register again */ + gpu_write(gpu, offset, value); + + /* We can't do a posted write here because the power domain could be + * in collapse state. So use the heaviest barrier instead + */ + mb(); + return false; +} + +static int fenced_write(struct a6xx_gpu *a6xx_gpu, u32 offset, u32 value, u32 mask) +{ + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + u32 status; + + gpu_write(gpu, offset, value); + + /* Nothing else to be done in the case of no-GMU */ + if (adreno_has_gmu_wrapper(adreno_gpu)) + return 0; + + /* We can't do a posted write here because the power domain could be + * in collapse state. So use the heaviest barrier instead + */ + mb(); + + if (!gmu_poll_timeout(gmu, REG_A6XX_GMU_AHB_FENCE_STATUS, status, + fence_status_check(gpu, offset, value, status, mask), 0, 1000)) + return 0; + + /* Try again for another 1ms before failing */ + gpu_write(gpu, offset, value); + mb(); + + if (!gmu_poll_timeout(gmu, REG_A6XX_GMU_AHB_FENCE_STATUS, status, + fence_status_check(gpu, offset, value, status, mask), 0, 1000)) { + /* + * The 'delay' warning is here because the pause to print this + * warning will allow gpu to move to power collapse which + * defeats the purpose of continuous polling for 2 ms + */ + dev_err_ratelimited(gmu->dev, "delay in fenced register write (0x%x)\n", + offset); + return 0; + } + + dev_err_ratelimited(gmu->dev, "fenced register write (0x%x) fail\n", + offset); + + return -ETIMEDOUT; +} + +int a6xx_fenced_write(struct a6xx_gpu *a6xx_gpu, u32 offset, u64 value, u32 mask, bool is_64b) +{ + int ret; + + ret = fenced_write(a6xx_gpu, offset, lower_32_bits(value), mask); + if (ret) + return ret; + + if (!is_64b) + return 0; + + ret = fenced_write(a6xx_gpu, offset + 1, upper_32_bits(value), mask); + + return ret; +} + +static inline bool _a6xx_check_idle(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + /* Check that the GMU is idle */ + if (!adreno_has_gmu_wrapper(adreno_gpu) && !a6xx_gmu_isidle(&a6xx_gpu->gmu)) + return false; + + /* Check tha the CX master is idle */ + if (gpu_read(gpu, REG_A6XX_RBBM_STATUS) & + ~A6XX_RBBM_STATUS_CP_AHB_BUSY_CX_MASTER) + return false; + + return !(gpu_read(gpu, REG_A6XX_RBBM_INT_0_STATUS) & + A6XX_RBBM_INT_0_MASK_RBBM_HANG_DETECT); +} + +static bool a6xx_idle(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + /* wait for CP to drain ringbuffer: */ + if (!adreno_idle(gpu, ring)) + return false; + + if (spin_until(_a6xx_check_idle(gpu))) { + DRM_ERROR("%s: %ps: timeout waiting for GPU to idle: status %8.8X irq %8.8X rptr/wptr %d/%d\n", + gpu->name, __builtin_return_address(0), + gpu_read(gpu, REG_A6XX_RBBM_STATUS), + gpu_read(gpu, REG_A6XX_RBBM_INT_0_STATUS), + gpu_read(gpu, REG_A6XX_CP_RB_RPTR), + gpu_read(gpu, REG_A6XX_CP_RB_WPTR)); + return false; + } + + return true; +} + +static void update_shadow_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + /* Expanded APRIV doesn't need to issue the WHERE_AM_I opcode */ + if (a6xx_gpu->has_whereami && !adreno_gpu->base.hw_apriv) { + OUT_PKT7(ring, CP_WHERE_AM_I, 2); + OUT_RING(ring, lower_32_bits(shadowptr(a6xx_gpu, ring))); + OUT_RING(ring, upper_32_bits(shadowptr(a6xx_gpu, ring))); + } +} + +void a6xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + uint32_t wptr; + unsigned long flags; + + update_shadow_rptr(gpu, ring); + + spin_lock_irqsave(&ring->preempt_lock, flags); + + /* Copy the shadow to the actual register */ + ring->cur = ring->next; + + /* Make sure to wrap wptr if we need to */ + wptr = get_wptr(ring); + + /* Update HW if this is the current ring and we are not in preempt*/ + if (!a6xx_in_preempt(a6xx_gpu)) { + if (a6xx_gpu->cur_ring == ring) + a6xx_fenced_write(a6xx_gpu, REG_A6XX_CP_RB_WPTR, wptr, BIT(0), false); + else + ring->restore_wptr = true; + } else { + ring->restore_wptr = true; + } + + spin_unlock_irqrestore(&ring->preempt_lock, flags); +} + +void +a6xx_flush_yield(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + /* If preemption is enabled */ + if (gpu->nr_rings > 1) { + /* Yield the floor on command completion */ + OUT_PKT7(ring, CP_CONTEXT_SWITCH_YIELD, 4); + + /* + * If dword[2:1] are non zero, they specify an address for + * the CP to write the value of dword[3] to on preemption + * complete. Write 0 to skip the write + */ + OUT_RING(ring, 0x00); + OUT_RING(ring, 0x00); + /* Data value - not used if the address above is 0 */ + OUT_RING(ring, 0x01); + /* generate interrupt on preemption completion */ + OUT_RING(ring, 0x00); + } + + a6xx_flush(gpu, ring); +} + +static void get_stats_counter(struct msm_ringbuffer *ring, u32 counter, + u64 iova) +{ + OUT_PKT7(ring, CP_REG_TO_MEM, 3); + OUT_RING(ring, CP_REG_TO_MEM_0_REG(counter) | + CP_REG_TO_MEM_0_CNT(2) | + CP_REG_TO_MEM_0_64B); + OUT_RING(ring, lower_32_bits(iova)); + OUT_RING(ring, upper_32_bits(iova)); +} + +static void a6xx_set_pagetable(struct a6xx_gpu *a6xx_gpu, + struct msm_ringbuffer *ring, struct msm_gem_submit *submit) +{ + bool sysprof = msm_gpu_sysprof_no_perfcntr_zap(&a6xx_gpu->base.base); + struct msm_context *ctx = submit->queue->ctx; + struct drm_gpuvm *vm = ctx->vm; + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + phys_addr_t ttbr; + u32 asid; + u64 memptr = rbmemptr(ring, ttbr0); + + if (ctx->seqno == ring->cur_ctx_seqno) + return; + + if (msm_iommu_pagetable_params(to_msm_vm(vm)->mmu, &ttbr, &asid)) + return; + + if (adreno_gpu->info->family >= ADRENO_7XX_GEN1) { + /* Wait for previous submit to complete before continuing: */ + OUT_PKT7(ring, CP_WAIT_TIMESTAMP, 4); + OUT_RING(ring, 0); + OUT_RING(ring, lower_32_bits(rbmemptr(ring, fence))); + OUT_RING(ring, upper_32_bits(rbmemptr(ring, fence))); + OUT_RING(ring, submit->seqno - 1); + + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, CP_THREAD_CONTROL_0_SYNC_THREADS | CP_SET_THREAD_BOTH); + + /* Reset state used to synchronize BR and BV */ + OUT_PKT7(ring, CP_RESET_CONTEXT_STATE, 1); + OUT_RING(ring, + CP_RESET_CONTEXT_STATE_0_CLEAR_ON_CHIP_TS | + CP_RESET_CONTEXT_STATE_0_CLEAR_RESOURCE_TABLE | + CP_RESET_CONTEXT_STATE_0_CLEAR_BV_BR_COUNTER | + CP_RESET_CONTEXT_STATE_0_RESET_GLOBAL_LOCAL_TS); + + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, CP_THREAD_CONTROL_0_SYNC_THREADS | CP_SET_THREAD_BOTH); + + OUT_PKT7(ring, CP_EVENT_WRITE, 1); + OUT_RING(ring, LRZ_FLUSH_INVALIDATE); + + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, CP_THREAD_CONTROL_0_SYNC_THREADS | CP_SET_THREAD_BR); + } + + if (!sysprof) { + if (!(adreno_is_a7xx(adreno_gpu) || adreno_is_a8xx(adreno_gpu))) { + /* Turn off protected mode to write to special registers */ + OUT_PKT7(ring, CP_SET_PROTECTED_MODE, 1); + OUT_RING(ring, 0); + } + + if (adreno_is_a8xx(adreno_gpu)) { + OUT_PKT4(ring, REG_A8XX_RBBM_PERFCTR_SRAM_INIT_CMD, 1); + OUT_RING(ring, 1); + OUT_PKT4(ring, REG_A8XX_RBBM_SLICE_PERFCTR_SRAM_INIT_CMD, 1); + OUT_RING(ring, 1); + } else { + OUT_PKT4(ring, REG_A6XX_RBBM_PERFCTR_SRAM_INIT_CMD, 1); + OUT_RING(ring, 1); + } + } + + /* Execute the table update */ + OUT_PKT7(ring, CP_SMMU_TABLE_UPDATE, 4); + OUT_RING(ring, CP_SMMU_TABLE_UPDATE_0_TTBR0_LO(lower_32_bits(ttbr))); + + OUT_RING(ring, + CP_SMMU_TABLE_UPDATE_1_TTBR0_HI(upper_32_bits(ttbr)) | + CP_SMMU_TABLE_UPDATE_1_ASID(asid)); + OUT_RING(ring, CP_SMMU_TABLE_UPDATE_2_CONTEXTIDR(0)); + OUT_RING(ring, CP_SMMU_TABLE_UPDATE_3_CONTEXTBANK(0)); + + /* + * Write the new TTBR0 to the memstore. This is good for debugging. + * Needed for preemption + */ + OUT_PKT7(ring, CP_MEM_WRITE, 5); + OUT_RING(ring, A5XX_CP_MEM_WRITE_ADDR_LO(lower_32_bits(memptr))); + OUT_RING(ring, A5XX_CP_MEM_WRITE_ADDR_HI(upper_32_bits(memptr))); + OUT_RING(ring, lower_32_bits(ttbr)); + OUT_RING(ring, upper_32_bits(ttbr)); + OUT_RING(ring, ctx->seqno); + + /* + * Sync both threads after switching pagetables and enable BR only + * to make sure BV doesn't race ahead while BR is still switching + * pagetables. + */ + if (adreno_is_a7xx(&a6xx_gpu->base) || adreno_is_a8xx(&a6xx_gpu->base)) { + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, CP_THREAD_CONTROL_0_SYNC_THREADS | CP_SET_THREAD_BR); + } + + /* + * And finally, trigger a uche flush to be sure there isn't anything + * lingering in that part of the GPU + */ + + OUT_PKT7(ring, CP_EVENT_WRITE, 1); + OUT_RING(ring, CACHE_INVALIDATE); + + if (!sysprof) { + u32 reg_status = adreno_is_a8xx(adreno_gpu) ? + REG_A8XX_RBBM_PERFCTR_SRAM_INIT_STATUS : + REG_A6XX_RBBM_PERFCTR_SRAM_INIT_STATUS; + /* + * Wait for SRAM clear after the pgtable update, so the + * two can happen in parallel: + */ + OUT_PKT7(ring, CP_WAIT_REG_MEM, 6); + OUT_RING(ring, CP_WAIT_REG_MEM_0_FUNCTION(WRITE_EQ)); + OUT_RING(ring, CP_WAIT_REG_MEM_POLL_ADDR_LO(reg_status)); + OUT_RING(ring, CP_WAIT_REG_MEM_POLL_ADDR_HI(0)); + OUT_RING(ring, CP_WAIT_REG_MEM_3_REF(0x1)); + OUT_RING(ring, CP_WAIT_REG_MEM_4_MASK(0x1)); + OUT_RING(ring, CP_WAIT_REG_MEM_5_DELAY_LOOP_CYCLES(0)); + + if (!(adreno_is_a7xx(adreno_gpu) || adreno_is_a8xx(adreno_gpu))) { + /* Re-enable protected mode: */ + OUT_PKT7(ring, CP_SET_PROTECTED_MODE, 1); + OUT_RING(ring, 1); + } + } +} + +static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) +{ + unsigned int index = submit->seqno % MSM_GPU_SUBMIT_STATS_COUNT; + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct msm_ringbuffer *ring = submit->ring; + unsigned int i, ibs = 0; + + adreno_check_and_reenable_stall(adreno_gpu); + + a6xx_set_pagetable(a6xx_gpu, ring, submit); + + get_stats_counter(ring, REG_A6XX_RBBM_PERFCTR_CP(0), + rbmemptr_stats(ring, index, cpcycles_start)); + + /* + * For PM4 the GMU register offsets are calculated from the base of the + * GPU registers so we need to add 0x1a800 to the register value on A630 + * to get the right value from PM4. + */ + get_stats_counter(ring, REG_A6XX_CP_ALWAYS_ON_CONTEXT, + rbmemptr_stats(ring, index, alwayson_start)); + + /* Invalidate CCU depth and color */ + OUT_PKT7(ring, CP_EVENT_WRITE, 1); + OUT_RING(ring, CP_EVENT_WRITE_0_EVENT(PC_CCU_INVALIDATE_DEPTH)); + + OUT_PKT7(ring, CP_EVENT_WRITE, 1); + OUT_RING(ring, CP_EVENT_WRITE_0_EVENT(PC_CCU_INVALIDATE_COLOR)); + + /* Submit the commands */ + for (i = 0; i < submit->nr_cmds; i++) { + switch (submit->cmd[i].type) { + case MSM_SUBMIT_CMD_IB_TARGET_BUF: + break; + case MSM_SUBMIT_CMD_CTX_RESTORE_BUF: + if (ring->cur_ctx_seqno == submit->queue->ctx->seqno) + break; + fallthrough; + case MSM_SUBMIT_CMD_BUF: + OUT_PKT7(ring, CP_INDIRECT_BUFFER, 3); + OUT_RING(ring, lower_32_bits(submit->cmd[i].iova)); + OUT_RING(ring, upper_32_bits(submit->cmd[i].iova)); + OUT_RING(ring, A5XX_CP_INDIRECT_BUFFER_2_IB_SIZE(submit->cmd[i].size)); + ibs++; + break; + } + + /* + * Periodically update shadow-wptr if needed, so that we + * can see partial progress of submits with large # of + * cmds.. otherwise we could needlessly stall waiting for + * ringbuffer state, simply due to looking at a shadow + * rptr value that has not been updated + */ + if ((ibs % 32) == 0) + update_shadow_rptr(gpu, ring); + } + + get_stats_counter(ring, REG_A6XX_RBBM_PERFCTR_CP(0), + rbmemptr_stats(ring, index, cpcycles_end)); + get_stats_counter(ring, REG_A6XX_CP_ALWAYS_ON_CONTEXT, + rbmemptr_stats(ring, index, alwayson_end)); + + /* Write the fence to the scratch register */ + OUT_PKT4(ring, REG_A6XX_CP_SCRATCH(2), 1); + OUT_RING(ring, submit->seqno); + + /* + * Execute a CACHE_FLUSH_TS event. This will ensure that the + * timestamp is written to the memory and then triggers the interrupt + */ + OUT_PKT7(ring, CP_EVENT_WRITE, 4); + OUT_RING(ring, CP_EVENT_WRITE_0_EVENT(CACHE_FLUSH_TS) | + CP_EVENT_WRITE_0_IRQ); + OUT_RING(ring, lower_32_bits(rbmemptr(ring, fence))); + OUT_RING(ring, upper_32_bits(rbmemptr(ring, fence))); + OUT_RING(ring, submit->seqno); + + trace_msm_gpu_submit_flush(submit, adreno_gpu->funcs->get_timestamp(gpu)); + + a6xx_flush(gpu, ring); +} + +void a6xx_emit_set_pseudo_reg(struct msm_ringbuffer *ring, + struct a6xx_gpu *a6xx_gpu, struct msm_gpu_submitqueue *queue) +{ + u64 preempt_postamble; + + OUT_PKT7(ring, CP_SET_PSEUDO_REG, 12); + + OUT_RING(ring, SMMU_INFO); + /* don't save SMMU, we write the record from the kernel instead */ + OUT_RING(ring, 0); + OUT_RING(ring, 0); + + /* privileged and non secure buffer save */ + OUT_RING(ring, NON_SECURE_SAVE_ADDR); + OUT_RING(ring, lower_32_bits( + a6xx_gpu->preempt_iova[ring->id])); + OUT_RING(ring, upper_32_bits( + a6xx_gpu->preempt_iova[ring->id])); + + /* user context buffer save, seems to be unnused by fw */ + OUT_RING(ring, NON_PRIV_SAVE_ADDR); + OUT_RING(ring, 0); + OUT_RING(ring, 0); + + OUT_RING(ring, COUNTER); + /* seems OK to set to 0 to disable it */ + OUT_RING(ring, 0); + OUT_RING(ring, 0); + + /* Emit postamble to clear perfcounters */ + preempt_postamble = a6xx_gpu->preempt_postamble_iova; + + OUT_PKT7(ring, CP_SET_AMBLE, 3); + OUT_RING(ring, lower_32_bits(preempt_postamble)); + OUT_RING(ring, upper_32_bits(preempt_postamble)); + OUT_RING(ring, CP_SET_AMBLE_2_DWORDS( + a6xx_gpu->preempt_postamble_len) | + CP_SET_AMBLE_2_TYPE(KMD_AMBLE_TYPE)); +} + +static void a7xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) +{ + unsigned int index = submit->seqno % MSM_GPU_SUBMIT_STATS_COUNT; + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct msm_ringbuffer *ring = submit->ring; + u32 rbbm_perfctr_cp0, cp_always_on_context; + unsigned int i, ibs = 0; + + adreno_check_and_reenable_stall(adreno_gpu); + + /* + * Toggle concurrent binning for pagetable switch and set the thread to + * BR since only it can execute the pagetable switch packets. + */ + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, CP_THREAD_CONTROL_0_SYNC_THREADS | CP_SET_THREAD_BR); + + a6xx_set_pagetable(a6xx_gpu, ring, submit); + + /* + * If preemption is enabled, then set the pseudo register for the save + * sequence + */ + if (gpu->nr_rings > 1) + a6xx_emit_set_pseudo_reg(ring, a6xx_gpu, submit->queue); + + if (adreno_is_a8xx(adreno_gpu)) { + rbbm_perfctr_cp0 = REG_A8XX_RBBM_PERFCTR_CP(0); + cp_always_on_context = REG_A8XX_CP_ALWAYS_ON_CONTEXT; + } else { + rbbm_perfctr_cp0 = REG_A7XX_RBBM_PERFCTR_CP(0); + cp_always_on_context = REG_A6XX_CP_ALWAYS_ON_CONTEXT; + } + + get_stats_counter(ring, rbbm_perfctr_cp0, rbmemptr_stats(ring, index, cpcycles_start)); + get_stats_counter(ring, cp_always_on_context, rbmemptr_stats(ring, index, alwayson_start)); + + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, CP_SET_THREAD_BOTH); + + OUT_PKT7(ring, CP_SET_MARKER, 1); + OUT_RING(ring, 0x101); /* IFPC disable */ + + if (submit->queue->flags & MSM_SUBMITQUEUE_ALLOW_PREEMPT) { + OUT_PKT7(ring, CP_SET_MARKER, 1); + OUT_RING(ring, 0x00d); /* IB1LIST start */ + } + + /* Submit the commands */ + for (i = 0; i < submit->nr_cmds; i++) { + switch (submit->cmd[i].type) { + case MSM_SUBMIT_CMD_IB_TARGET_BUF: + break; + case MSM_SUBMIT_CMD_CTX_RESTORE_BUF: + if (ring->cur_ctx_seqno == submit->queue->ctx->seqno) + break; + fallthrough; + case MSM_SUBMIT_CMD_BUF: + OUT_PKT7(ring, CP_INDIRECT_BUFFER, 3); + OUT_RING(ring, lower_32_bits(submit->cmd[i].iova)); + OUT_RING(ring, upper_32_bits(submit->cmd[i].iova)); + OUT_RING(ring, A5XX_CP_INDIRECT_BUFFER_2_IB_SIZE(submit->cmd[i].size)); + ibs++; + break; + } + + /* + * Periodically update shadow-wptr if needed, so that we + * can see partial progress of submits with large # of + * cmds.. otherwise we could needlessly stall waiting for + * ringbuffer state, simply due to looking at a shadow + * rptr value that has not been updated + */ + if ((ibs % 32) == 0) + update_shadow_rptr(gpu, ring); + } + + if (submit->queue->flags & MSM_SUBMITQUEUE_ALLOW_PREEMPT) { + OUT_PKT7(ring, CP_SET_MARKER, 1); + OUT_RING(ring, 0x00e); /* IB1LIST end */ + } + + get_stats_counter(ring, rbbm_perfctr_cp0, rbmemptr_stats(ring, index, cpcycles_end)); + get_stats_counter(ring, cp_always_on_context, rbmemptr_stats(ring, index, alwayson_end)); + + /* Write the fence to the scratch register */ + if (adreno_is_a8xx(adreno_gpu)) { + OUT_PKT4(ring, REG_A8XX_CP_SCRATCH_GLOBAL(2), 1); + OUT_RING(ring, submit->seqno); + } else { + OUT_PKT4(ring, REG_A6XX_CP_SCRATCH(2), 1); + OUT_RING(ring, submit->seqno); + } + + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, CP_SET_THREAD_BR); + + OUT_PKT7(ring, CP_EVENT_WRITE, 1); + OUT_RING(ring, CCU_INVALIDATE_DEPTH); + + OUT_PKT7(ring, CP_EVENT_WRITE, 1); + OUT_RING(ring, CCU_INVALIDATE_COLOR); + + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, CP_SET_THREAD_BV); + + /* + * Make sure the timestamp is committed once BV pipe is + * completely done with this submission. + */ + OUT_PKT7(ring, CP_EVENT_WRITE, 4); + OUT_RING(ring, CACHE_CLEAN | BIT(27)); + OUT_RING(ring, lower_32_bits(rbmemptr(ring, bv_fence))); + OUT_RING(ring, upper_32_bits(rbmemptr(ring, bv_fence))); + OUT_RING(ring, submit->seqno); + + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, CP_SET_THREAD_BR); + + /* + * This makes sure that BR doesn't race ahead and commit + * timestamp to memstore while BV is still processing + * this submission. + */ + OUT_PKT7(ring, CP_WAIT_TIMESTAMP, 4); + OUT_RING(ring, 0); + OUT_RING(ring, lower_32_bits(rbmemptr(ring, bv_fence))); + OUT_RING(ring, upper_32_bits(rbmemptr(ring, bv_fence))); + OUT_RING(ring, submit->seqno); + + a6xx_gpu->last_seqno[ring->id] = submit->seqno; + + /* write the ringbuffer timestamp */ + OUT_PKT7(ring, CP_EVENT_WRITE, 4); + OUT_RING(ring, CACHE_CLEAN | CP_EVENT_WRITE_0_IRQ | BIT(27)); + OUT_RING(ring, lower_32_bits(rbmemptr(ring, fence))); + OUT_RING(ring, upper_32_bits(rbmemptr(ring, fence))); + OUT_RING(ring, submit->seqno); + + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, CP_SET_THREAD_BOTH); + + OUT_PKT7(ring, CP_SET_MARKER, 1); + OUT_RING(ring, 0x100); /* IFPC enable */ + + trace_msm_gpu_submit_flush(submit, adreno_gpu->funcs->get_timestamp(gpu)); + + a6xx_flush_yield(gpu, ring); + + /* Check to see if we need to start preemption */ + if (adreno_is_a8xx(adreno_gpu)) + a8xx_preempt_trigger(gpu); + else + a6xx_preempt_trigger(gpu); +} + +static void a6xx_set_hwcg(struct msm_gpu *gpu, bool state) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + const struct adreno_reglist *reg; + unsigned int i; + u32 cgc_delay, cgc_hyst; + u32 val, clock_cntl_on; + + if (!(adreno_gpu->info->a6xx->hwcg || adreno_is_a7xx(adreno_gpu))) + return; + + if (adreno_is_a630(adreno_gpu)) + clock_cntl_on = 0x8aa8aa02; + else if (adreno_is_a610(adreno_gpu) || adreno_is_a612(adreno_gpu)) + clock_cntl_on = 0xaaa8aa82; + else if (adreno_is_a702(adreno_gpu)) + clock_cntl_on = 0xaaaaaa82; + else + clock_cntl_on = 0x8aa8aa82; + + if (adreno_is_a612(adreno_gpu)) + cgc_delay = 0x11; + else if (adreno_is_a615_family(adreno_gpu)) + cgc_delay = 0x111; + else + cgc_delay = 0x10111; + + if (adreno_is_a612(adreno_gpu)) + cgc_hyst = 0x55; + else if (adreno_is_a615_family(adreno_gpu)) + cgc_hyst = 0x555; + else + cgc_hyst = 0x5555; + + gmu_write(&a6xx_gpu->gmu, REG_A6XX_GPU_GMU_AO_GMU_CGC_MODE_CNTL, + state ? adreno_gpu->info->a6xx->gmu_cgc_mode : 0); + gmu_write(&a6xx_gpu->gmu, REG_A6XX_GPU_GMU_AO_GMU_CGC_DELAY_CNTL, + state ? cgc_delay : 0); + gmu_write(&a6xx_gpu->gmu, REG_A6XX_GPU_GMU_AO_GMU_CGC_HYST_CNTL, + state ? cgc_hyst : 0); + + if (!adreno_gpu->info->a6xx->hwcg) { + gpu_write(gpu, REG_A7XX_RBBM_CLOCK_CNTL_GLOBAL, 1); + gpu_write(gpu, REG_A7XX_RBBM_CGC_GLOBAL_LOAD_CMD, state ? 1 : 0); + + if (state) { + gpu_write(gpu, REG_A7XX_RBBM_CGC_P2S_TRIG_CMD, 1); + + if (gpu_poll_timeout(gpu, REG_A7XX_RBBM_CGC_P2S_STATUS, val, + val & A7XX_RBBM_CGC_P2S_STATUS_TXDONE, 1, 10)) { + dev_err(&gpu->pdev->dev, "RBBM_CGC_P2S_STATUS TXDONE Poll failed\n"); + return; + } + + gpu_write(gpu, REG_A7XX_RBBM_CLOCK_CNTL_GLOBAL, 0); + } + + return; + } + + val = gpu_read(gpu, REG_A6XX_RBBM_CLOCK_CNTL); + + /* Don't re-program the registers if they are already correct */ + if ((!state && !val) || (state && (val == clock_cntl_on))) + return; + + /* Disable SP clock before programming HWCG registers */ + if (!adreno_is_a610_family(adreno_gpu) && !adreno_is_a7xx(adreno_gpu)) + gmu_rmw(gmu, REG_A6XX_GPU_GMU_GX_SPTPRAC_CLOCK_CONTROL, 1, 0); + + for (i = 0; (reg = &adreno_gpu->info->a6xx->hwcg[i], reg->offset); i++) + gpu_write(gpu, reg->offset, state ? reg->value : 0); + + /* Enable SP clock */ + if (!adreno_is_a610_family(adreno_gpu) && !adreno_is_a7xx(adreno_gpu)) + gmu_rmw(gmu, REG_A6XX_GPU_GMU_GX_SPTPRAC_CLOCK_CONTROL, 0, 1); + + gpu_write(gpu, REG_A6XX_RBBM_CLOCK_CNTL, state ? clock_cntl_on : 0); +} + +static void a6xx_set_cp_protect(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + const struct adreno_protect *protect = adreno_gpu->info->a6xx->protect; + unsigned i; + + /* + * Enable access protection to privileged registers, fault on an access + * protect violation and select the last span to protect from the start + * address all the way to the end of the register address space + */ + gpu_write(gpu, REG_A6XX_CP_PROTECT_CNTL, + A6XX_CP_PROTECT_CNTL_ACCESS_PROT_EN | + A6XX_CP_PROTECT_CNTL_ACCESS_FAULT_ON_VIOL_EN | + A6XX_CP_PROTECT_CNTL_LAST_SPAN_INF_RANGE); + + for (i = 0; i < protect->count - 1; i++) { + /* Intentionally skip writing to some registers */ + if (protect->regs[i]) + gpu_write(gpu, REG_A6XX_CP_PROTECT(i), protect->regs[i]); + } + /* last CP_PROTECT to have "infinite" length on the last entry */ + gpu_write(gpu, REG_A6XX_CP_PROTECT(protect->count_max - 1), protect->regs[i]); +} + +static void a6xx_set_ubwc_config(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + const struct qcom_ubwc_cfg_data *cfg = adreno_gpu->ubwc_config; + /* + * We subtract 13 from the highest bank bit (13 is the minimum value + * allowed by hw) and write the lowest two bits of the remaining value + * as hbb_lo and the one above it as hbb_hi to the hardware. + */ + BUG_ON(cfg->highest_bank_bit < 13); + u32 hbb = cfg->highest_bank_bit - 13; + bool rgb565_predicator = cfg->ubwc_enc_version >= UBWC_4_0; + u32 level2_swizzling_dis = !(qcom_ubwc_swizzle(cfg) & UBWC_SWIZZLE_ENABLE_LVL2); + bool ubwc_mode = qcom_ubwc_get_ubwc_mode(cfg); + bool amsbc = qcom_ubwc_enable_amsbc(cfg); + bool min_acc_len_64b; + u8 uavflagprd_inv = 0; + u32 hbb_hi = hbb >> 2; + u32 hbb_lo = hbb & 3; + + if (adreno_is_a650_family(adreno_gpu) || adreno_is_a7xx(adreno_gpu)) + uavflagprd_inv = 2; + + min_acc_len_64b = qcom_ubwc_min_acc_length_64b(cfg); + + gpu_write(gpu, REG_A6XX_RB_NC_MODE_CNTL, + level2_swizzling_dis << 12 | + rgb565_predicator << 11 | + hbb_hi << 10 | amsbc << 4 | + min_acc_len_64b << 3 | + hbb_lo << 1 | ubwc_mode); + + gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, + level2_swizzling_dis << 6 | hbb_hi << 4 | + min_acc_len_64b << 3 | + hbb_lo << 1 | ubwc_mode); + + gpu_write(gpu, REG_A6XX_SP_NC_MODE_CNTL, + level2_swizzling_dis << 12 | hbb_hi << 10 | + uavflagprd_inv << 4 | + min_acc_len_64b << 3 | + hbb_lo << 1 | ubwc_mode); + + if (adreno_is_a7xx(adreno_gpu)) { + for (u32 pipe_id = PIPE_BR; pipe_id <= PIPE_BV; pipe_id++) { + gpu_write(gpu, REG_A7XX_CP_APERTURE_CNTL_HOST, + A7XX_CP_APERTURE_CNTL_HOST_PIPE(pipe_id)); + gpu_write(gpu, REG_A7XX_GRAS_NC_MODE_CNTL, + FIELD_PREP(GENMASK(8, 5), hbb_lo)); + } + gpu_write(gpu, REG_A7XX_CP_APERTURE_CNTL_HOST, + A7XX_CP_APERTURE_CNTL_HOST_PIPE(PIPE_NONE)); + } + + gpu_write(gpu, REG_A6XX_UCHE_MODE_CNTL, + min_acc_len_64b << 23 | hbb_lo << 21); + + gpu_write(gpu, REG_A6XX_RBBM_NC_MODE_CNTL, + qcom_ubwc_macrotile_mode(cfg)); +} + +static void a7xx_patch_pwrup_reglist(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + const struct adreno_reglist_list *reglist; + const struct adreno_reglist_pipe_list *dyn_pwrup_reglist; + void *ptr = a6xx_gpu->pwrup_reglist_ptr; + struct cpu_gpu_lock *lock = ptr; + u32 *dest = (u32 *)&lock->regs[0]; + u32 dyn_pwrup_reglist_count = 0; + int i; + + lock->gpu_req = lock->cpu_req = lock->turn = 0; + + reglist = adreno_gpu->info->a6xx->ifpc_reglist; + if (reglist) { + lock->ifpc_list_len = reglist->count; + + /* + * For each entry in each of the lists, write the offset and the current + * register value into the GPU buffer + */ + for (i = 0; i < reglist->count; i++) { + *dest++ = reglist->regs[i]; + *dest++ = gpu_read(gpu, reglist->regs[i]); + } + } + + reglist = adreno_gpu->info->a6xx->pwrup_reglist; + lock->preemption_list_len = reglist->count; + + for (i = 0; i < reglist->count; i++) { + *dest++ = reglist->regs[i]; + *dest++ = gpu_read(gpu, reglist->regs[i]); + } + + /* + * The overall register list is composed of + * 1. Static IFPC-only registers + * 2. Static IFPC + preemption registers + * 3. Dynamic IFPC + preemption registers (ex: perfcounter selects) + * + * The first two lists are static. Size of these lists are stored as + * number of pairs in ifpc_list_len and preemption_list_len + * respectively. With concurrent binning, Some of the perfcounter + * registers being virtualized, CP needs to know the pipe id to program + * the aperture inorder to restore the same. Thus, third list is a + * dynamic list with triplets as + * (
), and the length is + * stored as number for triplets in dynamic_list_len. + */ + dyn_pwrup_reglist = adreno_gpu->info->a6xx->dyn_pwrup_reglist; + if (dyn_pwrup_reglist) { + for (u32 pipe_id = PIPE_BR; pipe_id <= PIPE_BV; pipe_id++) { + gpu_write(gpu, REG_A7XX_CP_APERTURE_CNTL_HOST, + A7XX_CP_APERTURE_CNTL_HOST_PIPE(pipe_id)); + for (i = 0; i < dyn_pwrup_reglist->count; i++) { + if ((dyn_pwrup_reglist->regs[i].pipe & BIT(pipe_id)) == 0) + continue; + *dest++ = A7XX_CP_APERTURE_CNTL_HOST_PIPE(pipe_id); + *dest++ = dyn_pwrup_reglist->regs[i].offset; + *dest++ = gpu_read(gpu, dyn_pwrup_reglist->regs[i].offset); + dyn_pwrup_reglist_count++; + } + } + gpu_write(gpu, REG_A7XX_CP_APERTURE_CNTL_HOST, + A7XX_CP_APERTURE_CNTL_HOST_PIPE(PIPE_NONE)); + } + lock->dynamic_list_len = dyn_pwrup_reglist_count; + a6xx_gpu->dynamic_sel_reglist_offset = dyn_pwrup_reglist_count; +} + +static int a7xx_preempt_start(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct msm_ringbuffer *ring = gpu->rb[0]; + + if (gpu->nr_rings <= 1) + return 0; + + /* Turn CP protection off */ + OUT_PKT7(ring, CP_SET_PROTECTED_MODE, 1); + OUT_RING(ring, 0); + + a6xx_emit_set_pseudo_reg(ring, a6xx_gpu, NULL); + + a6xx_flush_yield(gpu, ring); + + return a6xx_idle(gpu, ring) ? 0 : -EINVAL; +} + +static int a6xx_cp_init(struct msm_gpu *gpu) +{ + struct msm_ringbuffer *ring = gpu->rb[0]; + + OUT_PKT7(ring, CP_ME_INIT, 8); + + OUT_RING(ring, 0x0000002f); + + /* Enable multiple hardware contexts */ + OUT_RING(ring, 0x00000003); + + /* Enable error detection */ + OUT_RING(ring, 0x20000000); + + /* Don't enable header dump */ + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + + /* No workarounds enabled */ + OUT_RING(ring, 0x00000000); + + /* Pad rest of the cmds with 0's */ + OUT_RING(ring, 0x00000000); + OUT_RING(ring, 0x00000000); + + a6xx_flush(gpu, ring); + return a6xx_idle(gpu, ring) ? 0 : -EINVAL; +} + +static int a7xx_cp_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct msm_ringbuffer *ring = gpu->rb[0]; + u32 mask; + + /* Disable concurrent binning before sending CP init */ + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, BIT(27)); + + OUT_PKT7(ring, CP_ME_INIT, 7); + + /* Use multiple HW contexts */ + mask = BIT(0); + + /* Enable error detection */ + mask |= BIT(1); + + /* Set default reset state */ + mask |= BIT(3); + + /* Disable save/restore of performance counters across preemption */ + mask |= BIT(6); + + /* Enable the register init list with the spinlock */ + mask |= BIT(8); + + OUT_RING(ring, mask); + + /* Enable multiple hardware contexts */ + OUT_RING(ring, 0x00000003); + + /* Enable error detection */ + OUT_RING(ring, 0x20000000); + + /* Operation mode mask */ + OUT_RING(ring, 0x00000002); + + /* *Don't* send a power up reg list for concurrent binning (TODO) */ + /* Lo address */ + OUT_RING(ring, lower_32_bits(a6xx_gpu->pwrup_reglist_iova)); + /* Hi address */ + OUT_RING(ring, upper_32_bits(a6xx_gpu->pwrup_reglist_iova)); + /* BIT(31) set => read the regs from the list */ + OUT_RING(ring, BIT(31)); + + a6xx_flush(gpu, ring); + return a6xx_idle(gpu, ring) ? 0 : -EINVAL; +} + +/* + * Check that the microcode version is new enough to include several key + * security fixes. Return true if the ucode is safe. + */ +static bool a6xx_ucode_check_version(struct a6xx_gpu *a6xx_gpu, + struct drm_gem_object *obj) +{ + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + const char *sqe_name = adreno_gpu->info->fw[ADRENO_FW_SQE]; + u32 *buf = msm_gem_get_vaddr(obj); + bool ret = false; + + if (IS_ERR(buf)) + return false; + + /* A7xx is safe! */ + if (adreno_is_a7xx(adreno_gpu) || adreno_is_a702(adreno_gpu) || adreno_is_a8xx(adreno_gpu)) + return true; + + /* + * Targets up to a640 (a618, a630 and a640) need to check for a + * microcode version that is patched to support the whereami opcode or + * one that is new enough to include it by default. + * + * a650 tier targets don't need whereami but still need to be + * equal to or newer than 0.95 for other security fixes + * + * a660 targets have all the critical security fixes from the start + */ + if (!strcmp(sqe_name, "a630_sqe.fw")) { + /* + * If the lowest nibble is 0xa that is an indication that this + * microcode has been patched. The actual version is in dword + * [3] but we only care about the patchlevel which is the lowest + * nibble of dword [3] + * + * Otherwise check that the firmware is greater than or equal + * to 1.90 which was the first version that had this fix built + * in + */ + if ((((buf[0] & 0xf) == 0xa) && (buf[2] & 0xf) >= 1) || + (buf[0] & 0xfff) >= 0x190) { + a6xx_gpu->has_whereami = true; + ret = true; + goto out; + } + + DRM_DEV_ERROR(&gpu->pdev->dev, + "a630 SQE ucode is too old. Have version %x need at least %x\n", + buf[0] & 0xfff, 0x190); + } else if (!strcmp(sqe_name, "a650_sqe.fw")) { + if ((buf[0] & 0xfff) >= 0x095) { + ret = true; + goto out; + } + + DRM_DEV_ERROR(&gpu->pdev->dev, + "a650 SQE ucode is too old. Have version %x need at least %x\n", + buf[0] & 0xfff, 0x095); + } else if (!strcmp(sqe_name, "a660_sqe.fw")) { + ret = true; + } else { + DRM_DEV_ERROR(&gpu->pdev->dev, + "unknown GPU, add it to a6xx_ucode_check_version()!!\n"); + } +out: + msm_gem_put_vaddr(obj); + return ret; +} + +static int a6xx_ucode_load(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + if (!a6xx_gpu->sqe_bo) { + a6xx_gpu->sqe_bo = adreno_fw_create_bo(gpu, + adreno_gpu->fw[ADRENO_FW_SQE], &a6xx_gpu->sqe_iova); + + if (IS_ERR(a6xx_gpu->sqe_bo)) { + int ret = PTR_ERR(a6xx_gpu->sqe_bo); + + a6xx_gpu->sqe_bo = NULL; + DRM_DEV_ERROR(&gpu->pdev->dev, + "Could not allocate SQE ucode: %d\n", ret); + + return ret; + } + + msm_gem_object_set_name(a6xx_gpu->sqe_bo, "sqefw"); + if (!a6xx_ucode_check_version(a6xx_gpu, a6xx_gpu->sqe_bo)) { + msm_gem_unpin_iova(a6xx_gpu->sqe_bo, gpu->vm); + drm_gem_object_put(a6xx_gpu->sqe_bo); + + a6xx_gpu->sqe_bo = NULL; + return -EPERM; + } + } + + if (!a6xx_gpu->aqe_bo && adreno_gpu->fw[ADRENO_FW_AQE]) { + a6xx_gpu->aqe_bo = adreno_fw_create_bo(gpu, + adreno_gpu->fw[ADRENO_FW_AQE], &a6xx_gpu->aqe_iova); + + if (IS_ERR(a6xx_gpu->aqe_bo)) { + int ret = PTR_ERR(a6xx_gpu->aqe_bo); + + a6xx_gpu->aqe_bo = NULL; + DRM_DEV_ERROR(&gpu->pdev->dev, + "Could not allocate AQE ucode: %d\n", ret); + + return ret; + } + + msm_gem_object_set_name(a6xx_gpu->aqe_bo, "aqefw"); + } + + /* + * Expanded APRIV and targets that support WHERE_AM_I both need a + * privileged buffer to store the RPTR shadow + */ + if ((adreno_gpu->base.hw_apriv || a6xx_gpu->has_whereami) && + !a6xx_gpu->shadow_bo) { + a6xx_gpu->shadow = msm_gem_kernel_new(gpu->dev, + sizeof(u32) * gpu->nr_rings, + MSM_BO_WC | MSM_BO_MAP_PRIV, + gpu->vm, &a6xx_gpu->shadow_bo, + &a6xx_gpu->shadow_iova); + + if (IS_ERR(a6xx_gpu->shadow)) + return PTR_ERR(a6xx_gpu->shadow); + + msm_gem_object_set_name(a6xx_gpu->shadow_bo, "shadow"); + } + + a6xx_gpu->pwrup_reglist_ptr = msm_gem_kernel_new(gpu->dev, PWRUP_REGLIST_SIZE, + MSM_BO_WC | MSM_BO_MAP_PRIV, + gpu->vm, &a6xx_gpu->pwrup_reglist_bo, + &a6xx_gpu->pwrup_reglist_iova); + + if (IS_ERR(a6xx_gpu->pwrup_reglist_ptr)) + return PTR_ERR(a6xx_gpu->pwrup_reglist_ptr); + + msm_gem_object_set_name(a6xx_gpu->pwrup_reglist_bo, "pwrup_reglist"); + + return 0; +} + +int a6xx_zap_shader_init(struct msm_gpu *gpu) +{ + static bool loaded; + int ret; + + if (loaded) + return 0; + + ret = adreno_zap_shader_load(gpu, GPU_PAS_ID); + + loaded = !ret; + return ret; +} + +#define A6XX_INT_MASK (A6XX_RBBM_INT_0_MASK_CP_AHB_ERROR | \ + A6XX_RBBM_INT_0_MASK_RBBM_ATB_ASYNCFIFO_OVERFLOW | \ + A6XX_RBBM_INT_0_MASK_CP_HW_ERROR | \ + A6XX_RBBM_INT_0_MASK_CP_IB2 | \ + A6XX_RBBM_INT_0_MASK_CP_IB1 | \ + A6XX_RBBM_INT_0_MASK_CP_RB | \ + A6XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS | \ + A6XX_RBBM_INT_0_MASK_RBBM_ATB_BUS_OVERFLOW | \ + A6XX_RBBM_INT_0_MASK_RBBM_HANG_DETECT | \ + A6XX_RBBM_INT_0_MASK_UCHE_OOB_ACCESS | \ + A6XX_RBBM_INT_0_MASK_UCHE_TRAP_INTR) + +#define A7XX_INT_MASK (A6XX_RBBM_INT_0_MASK_CP_AHB_ERROR | \ + A6XX_RBBM_INT_0_MASK_RBBM_ATB_ASYNCFIFO_OVERFLOW | \ + A6XX_RBBM_INT_0_MASK_RBBM_GPC_ERROR | \ + A6XX_RBBM_INT_0_MASK_CP_SW | \ + A6XX_RBBM_INT_0_MASK_CP_HW_ERROR | \ + A6XX_RBBM_INT_0_MASK_PM4CPINTERRUPT | \ + A6XX_RBBM_INT_0_MASK_CP_RB_DONE_TS | \ + A6XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS | \ + A6XX_RBBM_INT_0_MASK_RBBM_ATB_BUS_OVERFLOW | \ + A6XX_RBBM_INT_0_MASK_RBBM_HANG_DETECT | \ + A6XX_RBBM_INT_0_MASK_UCHE_OOB_ACCESS | \ + A6XX_RBBM_INT_0_MASK_UCHE_TRAP_INTR | \ + A6XX_RBBM_INT_0_MASK_TSBWRITEERROR | \ + A6XX_RBBM_INT_0_MASK_SWFUSEVIOLATION) + +#define A7XX_APRIV_MASK (A6XX_CP_APRIV_CNTL_ICACHE | \ + A6XX_CP_APRIV_CNTL_RBFETCH | \ + A6XX_CP_APRIV_CNTL_RBPRIVLEVEL | \ + A6XX_CP_APRIV_CNTL_RBRPWB) + +#define A7XX_BR_APRIVMASK (A7XX_APRIV_MASK | \ + A6XX_CP_APRIV_CNTL_CDREAD | \ + A6XX_CP_APRIV_CNTL_CDWRITE) + +static int hw_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + u64 gmem_range_min; + unsigned int i; + int ret; + + if (!adreno_has_gmu_wrapper(adreno_gpu)) { + /* Make sure the GMU keeps the GPU on while we set it up */ + ret = a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET); + if (ret) + return ret; + } + + /* Clear GBIF halt in case GX domain was not collapsed */ + if (adreno_is_a619_holi(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_GBIF_HALT, 0); + gpu_read(gpu, REG_A6XX_GBIF_HALT); + + gpu_write(gpu, REG_A6XX_RBBM_GPR0_CNTL, 0); + gpu_read(gpu, REG_A6XX_RBBM_GPR0_CNTL); + } else if (a6xx_has_gbif(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_GBIF_HALT, 0); + gpu_read(gpu, REG_A6XX_GBIF_HALT); + + gpu_write(gpu, REG_A6XX_RBBM_GBIF_HALT, 0); + gpu_read(gpu, REG_A6XX_RBBM_GBIF_HALT); + } + + gpu_write(gpu, REG_A6XX_RBBM_SECVID_TSB_CNTL, 0); + + if (adreno_is_a619_holi(adreno_gpu)) + a6xx_sptprac_enable(gmu); + + /* + * Disable the trusted memory range - we don't actually supported secure + * memory rendering at this point in time and we don't want to block off + * part of the virtual memory space. + */ + gpu_write64(gpu, REG_A6XX_RBBM_SECVID_TSB_TRUSTED_BASE, 0x00000000); + gpu_write(gpu, REG_A6XX_RBBM_SECVID_TSB_TRUSTED_SIZE, 0x00000000); + + if (!adreno_is_a7xx(adreno_gpu)) { + /* Turn on 64 bit addressing for all blocks */ + gpu_write(gpu, REG_A6XX_CP_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A6XX_VSC_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A6XX_GRAS_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A6XX_RB_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A6XX_PC_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A6XX_HLSQ_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A6XX_VFD_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A6XX_VPC_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A6XX_UCHE_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A6XX_SP_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A6XX_TPL1_ADDR_MODE_CNTL, 0x1); + gpu_write(gpu, REG_A6XX_RBBM_SECVID_TSB_ADDR_MODE_CNTL, 0x1); + } + + /* enable hardware clockgating */ + a6xx_set_hwcg(gpu, true); + + /* For gmuwrapper implementations, do the VBIF/GBIF CX configuration here */ + if (adreno_is_a610_family(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_GBIF_QSB_SIDE0, 0x00071620); + gpu_write(gpu, REG_A6XX_GBIF_QSB_SIDE1, 0x00071620); + gpu_write(gpu, REG_A6XX_GBIF_QSB_SIDE2, 0x00071620); + gpu_write(gpu, REG_A6XX_GBIF_QSB_SIDE3, 0x00071620); + } + + if (adreno_is_a610_family(adreno_gpu) || + adreno_is_a640_family(adreno_gpu) || + adreno_is_a650_family(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_RBBM_GBIF_CLIENT_QOS_CNTL, 0x3); + } else if (adreno_is_a7xx(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_RBBM_GBIF_CLIENT_QOS_CNTL, 0x2120212); + } else { + gpu_write(gpu, REG_A6XX_RBBM_VBIF_CLIENT_QOS_CNTL, 0x3); + } + + if (adreno_is_a630(adreno_gpu)) + gpu_write(gpu, REG_A6XX_VBIF_GATE_OFF_WRREQ_EN, 0x00000009); + + if (adreno_is_a7xx(adreno_gpu)) + gpu_write(gpu, REG_A6XX_UCHE_GBIF_GX_CONFIG, 0x10240e0); + + /* Make all blocks contribute to the GPU BUSY perf counter */ + gpu_write(gpu, REG_A6XX_RBBM_PERFCTR_GPU_BUSY_MASKED, 0xffffffff); + + /* Disable L2 bypass in the UCHE */ + if (adreno_is_a7xx(adreno_gpu)) { + gpu_write64(gpu, REG_A6XX_UCHE_TRAP_BASE, adreno_gpu->uche_trap_base); + gpu_write64(gpu, REG_A6XX_UCHE_WRITE_THRU_BASE, adreno_gpu->uche_trap_base); + } else { + gpu_write64(gpu, REG_A6XX_UCHE_WRITE_RANGE_MAX, adreno_gpu->uche_trap_base + 0xfc0); + gpu_write64(gpu, REG_A6XX_UCHE_TRAP_BASE, adreno_gpu->uche_trap_base); + gpu_write64(gpu, REG_A6XX_UCHE_WRITE_THRU_BASE, adreno_gpu->uche_trap_base); + } + + if (!(adreno_is_a650_family(adreno_gpu) || + adreno_is_a702(adreno_gpu) || + adreno_is_a730(adreno_gpu))) { + gmem_range_min = (adreno_is_a740_family(adreno_gpu) || + adreno_is_a722(adreno_gpu)) ? SZ_16M : SZ_1M; + + /* Set the GMEM VA range [0x100000:0x100000 + gpu->gmem - 1] */ + gpu_write64(gpu, REG_A6XX_UCHE_GMEM_RANGE_MIN, gmem_range_min); + + gpu_write64(gpu, REG_A6XX_UCHE_GMEM_RANGE_MAX, + gmem_range_min + adreno_gpu->info->gmem - 1); + } + + if (adreno_is_a7xx(adreno_gpu)) + gpu_write(gpu, REG_A6XX_UCHE_CACHE_WAYS, BIT(23)); + else { + gpu_write(gpu, REG_A6XX_UCHE_FILTER_CNTL, 0x804); + gpu_write(gpu, REG_A6XX_UCHE_CACHE_WAYS, 0x4); + } + + if (adreno_is_a640_family(adreno_gpu) || adreno_is_a650_family(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_2, 0x02000140); + gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_1, 0x8040362c); + } else if (adreno_is_a610_family(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_2, 0x00800060); + gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_1, 0x40201b16); + } else if (!adreno_is_a7xx(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_2, 0x010000c0); + gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_1, 0x8040362c); + } + + if (adreno_is_a660_family(adreno_gpu)) + gpu_write(gpu, REG_A7XX_CP_LPAC_PROG_FIFO_SIZE, 0x00000020); + + /* Setting the mem pool size */ + if (adreno_is_a610(adreno_gpu) || adreno_is_a612(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_CP_MEM_POOL_SIZE, 48); + gpu_write(gpu, REG_A6XX_CP_MEM_POOL_DBG_ADDR, 47); + } else if (adreno_is_a702(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_CP_MEM_POOL_SIZE, 64); + gpu_write(gpu, REG_A6XX_CP_MEM_POOL_DBG_ADDR, 63); + } else if (!adreno_is_a7xx(adreno_gpu)) + gpu_write(gpu, REG_A6XX_CP_MEM_POOL_SIZE, 128); + + + /* Set the default primFifo threshold values */ + if (adreno_gpu->info->a6xx->prim_fifo_threshold) + gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, + adreno_gpu->info->a6xx->prim_fifo_threshold); + + /* Set the AHB default slave response to "ERROR" */ + gpu_write(gpu, REG_A6XX_CP_AHB_CNTL, 0x1); + + /* Turn on performance counters */ + gpu_write(gpu, REG_A6XX_RBBM_PERFCTR_CNTL, 0x1); + + if (adreno_is_a7xx(adreno_gpu)) { + /* Turn on the IFPC counter (countable 4 on XOCLK4) */ + gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_SELECT_1, + FIELD_PREP(GENMASK(7, 0), 0x4)); + } + + /* Select CP0 to always count cycles */ + gpu_write(gpu, REG_A6XX_CP_PERFCTR_CP_SEL(0), PERF_CP_ALWAYS_COUNT); + + a6xx_set_ubwc_config(gpu); + + /* Enable fault detection */ + if (adreno_is_a612(adreno_gpu) || + adreno_is_a722(adreno_gpu) || + adreno_is_a730(adreno_gpu) || + adreno_is_a740_family(adreno_gpu)) + gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, (1 << 30) | 0xcfffff); + else if (adreno_is_a690(adreno_gpu)) + gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, (1 << 30) | 0x4fffff); + else if (adreno_is_a619(adreno_gpu)) + gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, (1 << 30) | 0x3fffff); + else if (adreno_is_a610(adreno_gpu) || adreno_is_a702(adreno_gpu)) + gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, (1 << 30) | 0x3ffff); + else + gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, (1 << 30) | 0x1fffff); + + gpu_write(gpu, REG_A6XX_UCHE_CLIENT_PF, BIT(7) | 0x1); + + /* Set weights for bicubic filtering */ + if (adreno_is_a650_family(adreno_gpu) || adreno_is_x185(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(0), 0); + gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(1), + 0x3fe05ff4); + gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(2), + 0x3fa0ebee); + gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(3), + 0x3f5193ed); + gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(4), + 0x3f0243f0); + } + + /* Set up the CX GMU counter 0 to count busy ticks */ + gmu_write(gmu, REG_A6XX_GPU_GMU_AO_GPU_CX_BUSY_MASK, 0xff000000); + + /* Enable the power counter */ + gmu_rmw(gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_SELECT_0, 0xff, BIT(5)); + gmu_write(gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_ENABLE, 1); + + /* Protect registers from the CP */ + a6xx_set_cp_protect(gpu); + + if (adreno_is_a660_family(adreno_gpu)) { + if (adreno_is_a690(adreno_gpu)) + gpu_write(gpu, REG_A6XX_CP_CHICKEN_DBG, 0x00028801); + else + gpu_write(gpu, REG_A6XX_CP_CHICKEN_DBG, 0x1); + gpu_write(gpu, REG_A6XX_RBBM_GBIF_CLIENT_QOS_CNTL, 0x0); + } else if (adreno_is_a702(adreno_gpu)) { + /* Something to do with the HLSQ cluster */ + gpu_write(gpu, REG_A6XX_CP_CHICKEN_DBG, BIT(24)); + } + + if (adreno_is_a690(adreno_gpu)) + gpu_write(gpu, REG_A6XX_UCHE_CMDQ_CONFIG, 0x90); + /* Set dualQ + disable afull for A660 GPU */ + else if (adreno_is_a660(adreno_gpu) || adreno_is_a663(adreno_gpu)) + gpu_write(gpu, REG_A6XX_UCHE_CMDQ_CONFIG, 0x66906); + else if (adreno_is_a7xx(adreno_gpu)) + gpu_write(gpu, REG_A6XX_UCHE_CMDQ_CONFIG, + FIELD_PREP(GENMASK(19, 16), 6) | + FIELD_PREP(GENMASK(15, 12), 6) | + FIELD_PREP(GENMASK(11, 8), 9) | + BIT(3) | BIT(2) | + FIELD_PREP(GENMASK(1, 0), 2)); + + /* Enable expanded apriv for targets that support it */ + if (gpu->hw_apriv) { + if (adreno_is_a7xx(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_CP_APRIV_CNTL, + A7XX_BR_APRIVMASK); + gpu_write(gpu, REG_A7XX_CP_BV_APRIV_CNTL, + A7XX_APRIV_MASK); + gpu_write(gpu, REG_A7XX_CP_LPAC_APRIV_CNTL, + A7XX_APRIV_MASK); + } else + gpu_write(gpu, REG_A6XX_CP_APRIV_CNTL, + BIT(6) | BIT(5) | BIT(3) | BIT(2) | BIT(1)); + } + + if (adreno_is_a750(adreno_gpu)) { + /* Disable ubwc merged UFC request feature */ + gpu_rmw(gpu, REG_A6XX_RB_CMP_DBG_ECO_CNTL, BIT(19), BIT(19)); + + /* Enable TP flaghint and other performance settings */ + gpu_write(gpu, REG_A6XX_TPL1_DBG_ECO_CNTL1, 0xc0700); + } else if (adreno_is_a7xx(adreno_gpu)) { + /* Disable non-ubwc read reqs from passing write reqs */ + gpu_rmw(gpu, REG_A6XX_RB_CMP_DBG_ECO_CNTL, BIT(11), BIT(11)); + } + + /* Enable interrupts */ + gpu_write(gpu, REG_A6XX_RBBM_INT_0_MASK, + adreno_is_a7xx(adreno_gpu) ? A7XX_INT_MASK : A6XX_INT_MASK); + + ret = adreno_hw_init(gpu); + if (ret) + goto out; + + gpu_write64(gpu, REG_A6XX_CP_SQE_INSTR_BASE, a6xx_gpu->sqe_iova); + + /* Set the ringbuffer address */ + gpu_write64(gpu, REG_A6XX_CP_RB_BASE, gpu->rb[0]->iova); + + /* Targets that support extended APRIV can use the RPTR shadow from + * hardware but all the other ones need to disable the feature. Targets + * that support the WHERE_AM_I opcode can use that instead + */ + if (adreno_gpu->base.hw_apriv) + gpu_write(gpu, REG_A6XX_CP_RB_CNTL, MSM_GPU_RB_CNTL_DEFAULT); + else + gpu_write(gpu, REG_A6XX_CP_RB_CNTL, + MSM_GPU_RB_CNTL_DEFAULT | AXXX_CP_RB_CNTL_NO_UPDATE); + + /* Configure the RPTR shadow if needed: */ + if (a6xx_gpu->shadow_bo) { + gpu_write64(gpu, REG_A6XX_CP_RB_RPTR_ADDR, + shadowptr(a6xx_gpu, gpu->rb[0])); + for (unsigned int i = 0; i < gpu->nr_rings; i++) + a6xx_gpu->shadow[i] = 0; + } + + /* ..which means "always" on A7xx, also for BV shadow */ + if (adreno_is_a7xx(adreno_gpu)) { + gpu_write64(gpu, REG_A7XX_CP_BV_RB_RPTR_ADDR, + rbmemptr(gpu->rb[0], bv_rptr)); + } + + a6xx_preempt_hw_init(gpu); + + /* Always come up on rb 0 */ + a6xx_gpu->cur_ring = gpu->rb[0]; + + for (i = 0; i < gpu->nr_rings; i++) + gpu->rb[i]->cur_ctx_seqno = 0; + + /* Enable the SQE_to start the CP engine */ + gpu_write(gpu, REG_A6XX_CP_SQE_CNTL, 1); + + if (adreno_is_a7xx(adreno_gpu) && !a6xx_gpu->pwrup_reglist_emitted) { + a7xx_patch_pwrup_reglist(gpu); + a6xx_gpu->pwrup_reglist_emitted = true; + } + + ret = adreno_is_a7xx(adreno_gpu) ? a7xx_cp_init(gpu) : a6xx_cp_init(gpu); + if (ret) + goto out; + + /* + * Try to load a zap shader into the secure world. If successful + * we can use the CP to switch out of secure mode. If not then we + * have no resource but to try to switch ourselves out manually. If we + * guessed wrong then access to the RBBM_SECVID_TRUST_CNTL register will + * be blocked and a permissions violation will soon follow. + */ + ret = a6xx_zap_shader_init(gpu); + if (!ret) { + OUT_PKT7(gpu->rb[0], CP_SET_SECURE_MODE, 1); + OUT_RING(gpu->rb[0], 0x00000000); + + a6xx_flush(gpu, gpu->rb[0]); + if (!a6xx_idle(gpu, gpu->rb[0])) + return -EINVAL; + } else if (ret == -ENODEV) { + /* + * This device does not use zap shader (but print a warning + * just in case someone got their dt wrong.. hopefully they + * have a debug UART to realize the error of their ways... + * if you mess this up you are about to crash horribly) + */ + dev_warn_once(gpu->dev->dev, + "Zap shader not enabled - using SECVID_TRUST_CNTL instead\n"); + gpu_write(gpu, REG_A6XX_RBBM_SECVID_TRUST_CNTL, 0x0); + ret = 0; + } else { + return ret; + } + +out: + if (adreno_has_gmu_wrapper(adreno_gpu)) + return ret; + + /* Last step - yield the ringbuffer */ + a7xx_preempt_start(gpu); + + /* + * Tell the GMU that we are done touching the GPU and it can start power + * management + */ + a6xx_gmu_clear_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET); + + if (a6xx_gpu->gmu.legacy) { + /* Take the GMU out of its special boot mode */ + a6xx_gmu_clear_oob(&a6xx_gpu->gmu, GMU_OOB_BOOT_SLUMBER); + } + + if (!ret && msm_gpu_sysprof_no_ifpc(gpu)) { + ret = a6xx_gmu_set_oob(gmu, GMU_OOB_PERFCOUNTER_SET); + if (!ret) + set_bit(GMU_STATUS_OOB_PERF_SET, &gmu->status); + } + + return ret; +} + +static int a6xx_hw_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + int ret; + + mutex_lock(&a6xx_gpu->gmu.lock); + ret = hw_init(gpu); + mutex_unlock(&a6xx_gpu->gmu.lock); + + return ret; +} + +static void a6xx_dump(struct msm_gpu *gpu) +{ + DRM_DEV_INFO(&gpu->pdev->dev, "status: %08x\n", + gpu_read(gpu, REG_A6XX_RBBM_STATUS)); + adreno_dump(gpu); +} + +static void a6xx_recover(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + int active_submits; + + adreno_dump_info(gpu); + + if (adreno_gpu->funcs->gx_is_on(adreno_gpu)) { + /* Sometimes crashstate capture is skipped, so SQE should be halted here again */ + gpu_write(gpu, REG_A6XX_CP_SQE_CNTL, 3); + + if (hang_debug) + a6xx_dump(gpu); + + } + + /* + * To handle recovery specific sequences during the rpm suspend we are + * about to trigger + */ + + a6xx_gpu->hung = true; + + pm_runtime_dont_use_autosuspend(&gpu->pdev->dev); + + /* active_submit won't change until we make a submission */ + mutex_lock(&gpu->active_lock); + active_submits = gpu->active_submits; + + /* + * Temporarily clear active_submits count to silence a WARN() in the + * runtime suspend cb + */ + gpu->active_submits = 0; + + if (adreno_has_gmu_wrapper(adreno_gpu) || adreno_has_rgmu(adreno_gpu)) { + /* Drain the outstanding traffic on memory buses */ + adreno_gpu->funcs->bus_halt(adreno_gpu, true); + + /* Reset the GPU to a clean state */ + a6xx_gpu_sw_reset(gpu, true); + a6xx_gpu_sw_reset(gpu, false); + } + + reinit_completion(&gmu->pd_gate); + dev_pm_genpd_add_notifier(gmu->cxpd, &gmu->pd_nb); + dev_pm_genpd_synced_poweroff(gmu->cxpd); + + /* Drop the rpm refcount from active submits */ + if (active_submits) + pm_runtime_put(&gpu->pdev->dev); + + /* And the final one from recover worker */ + pm_runtime_put_sync(&gpu->pdev->dev); + + if (!wait_for_completion_timeout(&gmu->pd_gate, msecs_to_jiffies(1000))) + DRM_DEV_ERROR(&gpu->pdev->dev, "cx gdsc didn't collapse\n"); + + dev_pm_genpd_remove_notifier(gmu->cxpd); + + pm_runtime_use_autosuspend(&gpu->pdev->dev); + + if (active_submits) + pm_runtime_get(&gpu->pdev->dev); + + pm_runtime_get_sync(&gpu->pdev->dev); + + gpu->active_submits = active_submits; + mutex_unlock(&gpu->active_lock); + + msm_gpu_hw_init(gpu); + a6xx_gpu->hung = false; +} + +static const char *a6xx_uche_fault_block(struct msm_gpu *gpu, u32 mid) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + static const char *uche_clients[7] = { + "VFD", "SP", "VSC", "VPC", "HLSQ", "PC", "LRZ", + }; + u32 val; + + if (adreno_is_a7xx(adreno_gpu)) { + if (mid != 1 && mid != 2 && mid != 3 && mid != 8) + return "UNKNOWN"; + } else { + if (mid < 1 || mid > 3) + return "UNKNOWN"; + } + + /* + * The source of the data depends on the mid ID read from FSYNR1. + * and the client ID read from the UCHE block + */ + val = gpu_read(gpu, REG_A6XX_UCHE_CLIENT_PF); + + if (adreno_is_a7xx(adreno_gpu)) { + /* Bit 3 for mid=3 indicates BR or BV */ + static const char *uche_clients_a7xx[16] = { + "BR_VFD", "BR_SP", "BR_VSC", "BR_VPC", + "BR_HLSQ", "BR_PC", "BR_LRZ", "BR_TP", + "BV_VFD", "BV_SP", "BV_VSC", "BV_VPC", + "BV_HLSQ", "BV_PC", "BV_LRZ", "BV_TP", + }; + + /* LPAC has the same clients as BR and BV, but because it is + * compute-only some of them do not exist and there are holes + * in the array. + */ + static const char *uche_clients_lpac_a7xx[8] = { + "-", "LPAC_SP", "-", "-", + "LPAC_HLSQ", "-", "-", "LPAC_TP", + }; + + val &= GENMASK(6, 0); + + /* mid=3 refers to BR or BV */ + if (mid == 3) { + if (val < ARRAY_SIZE(uche_clients_a7xx)) + return uche_clients_a7xx[val]; + else + return "UCHE"; + } + + /* mid=8 refers to LPAC */ + if (mid == 8) { + if (val < ARRAY_SIZE(uche_clients_lpac_a7xx)) + return uche_clients_lpac_a7xx[val]; + else + return "UCHE_LPAC"; + } + + /* mid=2 is a catchall for everything else in LPAC */ + if (mid == 2) + return "UCHE_LPAC"; + + /* mid=1 is a catchall for everything else in BR/BV */ + return "UCHE"; + } else if (adreno_is_a660_family(adreno_gpu)) { + static const char *uche_clients_a660[8] = { + "VFD", "SP", "VSC", "VPC", "HLSQ", "PC", "LRZ", "TP", + }; + + static const char *uche_clients_a660_not[8] = { + "not VFD", "not SP", "not VSC", "not VPC", + "not HLSQ", "not PC", "not LRZ", "not TP", + }; + + val &= GENMASK(6, 0); + + if (mid == 3 && val < ARRAY_SIZE(uche_clients_a660)) + return uche_clients_a660[val]; + + if (mid == 1 && val < ARRAY_SIZE(uche_clients_a660_not)) + return uche_clients_a660_not[val]; + + return "UCHE"; + } else { + /* mid = 3 is most precise and refers to only one block per client */ + if (mid == 3) + return uche_clients[val & 7]; + + /* For mid=2 the source is TP or VFD except when the client id is 0 */ + if (mid == 2) + return ((val & 7) == 0) ? "TP" : "TP|VFD"; + + /* For mid=1 just return "UCHE" as a catchall for everything else */ + return "UCHE"; + } +} + +static const char *a6xx_fault_block(struct msm_gpu *gpu, u32 id) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + + if (id == 0) + return "CP"; + else if (id == 4) + return "CCU"; + else if (id == 6) + return "CDP Prefetch"; + else if (id == 7) + return "GMU"; + else if (id == 5 && adreno_is_a7xx(adreno_gpu)) + return "Flag cache"; + + return a6xx_uche_fault_block(gpu, id); +} + +static int a6xx_fault_handler(void *arg, unsigned long iova, int flags, void *data) +{ + struct msm_gpu *gpu = arg; + struct adreno_smmu_fault_info *info = data; + const char *block = "unknown"; + + u32 scratch[] = { + gpu_read(gpu, REG_A6XX_CP_SCRATCH(4)), + gpu_read(gpu, REG_A6XX_CP_SCRATCH(5)), + gpu_read(gpu, REG_A6XX_CP_SCRATCH(6)), + gpu_read(gpu, REG_A6XX_CP_SCRATCH(7)), + }; + + if (info) + block = a6xx_fault_block(gpu, info->fsynr1 & 0xff); + + return adreno_fault_handler(gpu, iova, flags, info, block, scratch); +} + +static void a6xx_cp_hw_err_irq(struct msm_gpu *gpu) +{ + u32 status = gpu_read(gpu, REG_A6XX_CP_INTERRUPT_STATUS); + + if (status & A6XX_CP_INT_CP_OPCODE_ERROR) { + u32 val; + + gpu_write(gpu, REG_A6XX_CP_SQE_STAT_ADDR, 1); + val = gpu_read(gpu, REG_A6XX_CP_SQE_STAT_DATA); + dev_err_ratelimited(&gpu->pdev->dev, + "CP | opcode error | possible opcode=0x%8.8X\n", + val); + } + + if (status & A6XX_CP_INT_CP_UCODE_ERROR) + dev_err_ratelimited(&gpu->pdev->dev, + "CP ucode error interrupt\n"); + + if (status & A6XX_CP_INT_CP_HW_FAULT_ERROR) + dev_err_ratelimited(&gpu->pdev->dev, "CP | HW fault | status=0x%8.8X\n", + gpu_read(gpu, REG_A6XX_CP_HW_FAULT)); + + if (status & A6XX_CP_INT_CP_REGISTER_PROTECTION_ERROR) { + u32 val = gpu_read(gpu, REG_A6XX_CP_PROTECT_STATUS); + + dev_err_ratelimited(&gpu->pdev->dev, + "CP | protected mode error | %s | addr=0x%8.8X | status=0x%8.8X\n", + val & (1 << 20) ? "READ" : "WRITE", + (val & 0x3ffff), val); + } + + if (status & A6XX_CP_INT_CP_AHB_ERROR && !adreno_is_a7xx(to_adreno_gpu(gpu))) + dev_err_ratelimited(&gpu->pdev->dev, "CP AHB error interrupt\n"); + + if (status & A6XX_CP_INT_CP_VSD_PARITY_ERROR) + dev_err_ratelimited(&gpu->pdev->dev, "CP VSD decoder parity error\n"); + + if (status & A6XX_CP_INT_CP_ILLEGAL_INSTR_ERROR) + dev_err_ratelimited(&gpu->pdev->dev, "CP illegal instruction error\n"); + +} + +static void a6xx_fault_detect_irq(struct msm_gpu *gpu) +{ + struct msm_ringbuffer *ring = gpu->funcs->active_ring(gpu); + + /* + * If stalled on SMMU fault, we could trip the GPU's hang detection, + * but the fault handler will trigger the devcore dump, and we want + * to otherwise resume normally rather than killing the submit, so + * just bail. + */ + if (gpu_read(gpu, REG_A6XX_RBBM_STATUS3) & A6XX_RBBM_STATUS3_SMMU_STALLED_ON_FAULT) + return; + + DRM_DEV_ERROR(&gpu->pdev->dev, + "gpu fault ring %d fence %x status %8.8X rb %4.4x/%4.4x ib1 %16.16llX/%4.4x ib2 %16.16llX/%4.4x\n", + ring ? ring->id : -1, ring ? ring->fctx->last_fence : 0, + gpu_read(gpu, REG_A6XX_RBBM_STATUS), + gpu_read(gpu, REG_A6XX_CP_RB_RPTR), + gpu_read(gpu, REG_A6XX_CP_RB_WPTR), + gpu_read64(gpu, REG_A6XX_CP_IB1_BASE), + gpu_read(gpu, REG_A6XX_CP_IB1_REM_SIZE), + gpu_read64(gpu, REG_A6XX_CP_IB2_BASE), + gpu_read(gpu, REG_A6XX_CP_IB2_REM_SIZE)); + + /* Turn off the hangcheck timer to keep it from bothering us */ + timer_delete(&gpu->hangcheck_timer); + + /* Turn off interrupts to avoid triggering recovery again */ + gpu_write(gpu, REG_A6XX_RBBM_INT_0_MASK, 0); + + kthread_queue_work(gpu->worker, &gpu->recover_work); +} + +static void a7xx_sw_fuse_violation_irq(struct msm_gpu *gpu) +{ + u32 status; + + status = gpu_read(gpu, REG_A7XX_RBBM_SW_FUSE_INT_STATUS); + gpu_write(gpu, REG_A7XX_RBBM_SW_FUSE_INT_MASK, 0); + + dev_err_ratelimited(&gpu->pdev->dev, "SW fuse violation status=%8.8x\n", status); + + /* + * Ignore FASTBLEND violations, because the HW will silently fall back + * to legacy blending. + */ + if (status & (A7XX_CX_MISC_SW_FUSE_VALUE_RAYTRACING | + A7XX_CX_MISC_SW_FUSE_VALUE_LPAC)) { + timer_delete(&gpu->hangcheck_timer); + + kthread_queue_work(gpu->worker, &gpu->recover_work); + } +} + +static void a6xx_gpu_keepalive_vote(struct msm_gpu *gpu, bool on) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + if (adreno_has_gmu_wrapper(adreno_gpu)) + return; + + gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_GMU_PWR_COL_KEEPALIVE, on); +} + +static int irq_poll_fence(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + u32 status; + + if (adreno_has_gmu_wrapper(adreno_gpu)) + return 0; + + if (gmu_poll_timeout_atomic(gmu, REG_A6XX_GMU_AO_AHB_FENCE_CTRL, status, !status, 1, 100)) { + u32 rbbm_unmasked = gmu_read(gmu, REG_A6XX_GMU_RBBM_INT_UNMASKED_STATUS); + + dev_err_ratelimited(&gpu->pdev->dev, + "irq fence poll timeout, fence_ctrl=0x%x, unmasked_status=0x%x\n", + status, rbbm_unmasked); + return -ETIMEDOUT; + } + + return 0; +} + +static irqreturn_t a6xx_irq(struct msm_gpu *gpu) +{ + struct msm_drm_private *priv = gpu->dev->dev_private; + + /* Set keepalive vote to avoid power collapse after RBBM_INT_0_STATUS is read */ + a6xx_gpu_keepalive_vote(gpu, true); + + if (irq_poll_fence(gpu)) + goto done; + + u32 status = gpu_read(gpu, REG_A6XX_RBBM_INT_0_STATUS); + + gpu_write(gpu, REG_A6XX_RBBM_INT_CLEAR_CMD, status); + + if (priv->disable_err_irq) { + /* Turn off interrupts to avoid interrupt storm */ + gpu_write(gpu, REG_A6XX_RBBM_INT_0_MASK, A6XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS); + status &= A6XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS; + } + + if (status & A6XX_RBBM_INT_0_MASK_RBBM_HANG_DETECT) + a6xx_fault_detect_irq(gpu); + + if (status & A6XX_RBBM_INT_0_MASK_CP_AHB_ERROR) + dev_err_ratelimited(&gpu->pdev->dev, "CP | AHB bus error\n"); + + if (status & A6XX_RBBM_INT_0_MASK_CP_HW_ERROR) + a6xx_cp_hw_err_irq(gpu); + + if (status & A6XX_RBBM_INT_0_MASK_RBBM_ATB_ASYNCFIFO_OVERFLOW) + dev_err_ratelimited(&gpu->pdev->dev, "RBBM | ATB ASYNC overflow\n"); + + if (status & A6XX_RBBM_INT_0_MASK_RBBM_ATB_BUS_OVERFLOW) + dev_err_ratelimited(&gpu->pdev->dev, "RBBM | ATB bus overflow\n"); + + if (status & A6XX_RBBM_INT_0_MASK_UCHE_OOB_ACCESS) + dev_err_ratelimited(&gpu->pdev->dev, "UCHE | Out of bounds access\n"); + + if (status & A6XX_RBBM_INT_0_MASK_SWFUSEVIOLATION) + a7xx_sw_fuse_violation_irq(gpu); + + if (status & A6XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS) { + msm_gpu_retire(gpu); + a6xx_preempt_trigger(gpu); + } + + if (status & A6XX_RBBM_INT_0_MASK_CP_SW) + a6xx_preempt_irq(gpu); + +done: + a6xx_gpu_keepalive_vote(gpu, false); + + return IRQ_HANDLED; +} + +static void a6xx_llc_deactivate(struct a6xx_gpu *a6xx_gpu) +{ + llcc_slice_deactivate(a6xx_gpu->llc_slice); + llcc_slice_deactivate(a6xx_gpu->htw_llc_slice); +} + +static void a6xx_llc_activate(struct a6xx_gpu *a6xx_gpu) +{ + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + u32 cntl1_regval = 0; + + if (IS_ERR_OR_NULL(a6xx_gpu->llc_slice) && IS_ERR_OR_NULL(a6xx_gpu->htw_llc_slice)) + return; + + if (!llcc_slice_activate(a6xx_gpu->llc_slice)) { + u32 gpu_scid = llcc_get_slice_id(a6xx_gpu->llc_slice); + + gpu_scid &= 0x1f; + cntl1_regval = (gpu_scid << 0) | (gpu_scid << 5) | (gpu_scid << 10) | + (gpu_scid << 15) | (gpu_scid << 20); + + /* On A660, the SCID programming for UCHE traffic is done in + * A6XX_GBIF_SCACHE_CNTL0[14:10] + */ + if (adreno_is_a660_family(adreno_gpu)) + gpu_rmw(gpu, REG_A6XX_GBIF_SCACHE_CNTL0, (0x1f << 10) | + (1 << 8), (gpu_scid << 10) | (1 << 8)); + } + + /* + * For targets with a MMU500, activate the slice but don't program the + * register. The XBL will take care of that. + */ + if (!llcc_slice_activate(a6xx_gpu->htw_llc_slice)) { + if (!a6xx_gpu->have_mmu500) { + u32 gpuhtw_scid = llcc_get_slice_id(a6xx_gpu->htw_llc_slice); + + gpuhtw_scid &= 0x1f; + cntl1_regval |= FIELD_PREP(GENMASK(29, 25), gpuhtw_scid); + } + } + + if (!cntl1_regval) + return; + + /* + * Program the slice IDs for the various GPU blocks and GPU MMU + * pagetables + */ + if (!a6xx_gpu->have_mmu500) { + a6xx_cx_misc_write(a6xx_gpu, + REG_A6XX_CX_MISC_SYSTEM_CACHE_CNTL_1, cntl1_regval); + + /* + * Program cacheability overrides to not allocate cache + * lines on a write miss + */ + a6xx_cx_misc_rmw(a6xx_gpu, + REG_A6XX_CX_MISC_SYSTEM_CACHE_CNTL_0, 0xF, 0x03); + return; + } + + gpu_rmw(gpu, REG_A6XX_GBIF_SCACHE_CNTL1, GENMASK(24, 0), cntl1_regval); +} + +static void a7xx_llc_activate(struct a6xx_gpu *a6xx_gpu) +{ + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + + if (IS_ERR_OR_NULL(a6xx_gpu->llc_slice) && IS_ERR_OR_NULL(a6xx_gpu->htw_llc_slice)) + return; + + if (!llcc_slice_activate(a6xx_gpu->llc_slice)) { + u32 gpu_scid = llcc_get_slice_id(a6xx_gpu->llc_slice); + + gpu_scid &= GENMASK(4, 0); + + gpu_write(gpu, REG_A6XX_GBIF_SCACHE_CNTL1, + FIELD_PREP(GENMASK(29, 25), gpu_scid) | + FIELD_PREP(GENMASK(24, 20), gpu_scid) | + FIELD_PREP(GENMASK(19, 15), gpu_scid) | + FIELD_PREP(GENMASK(14, 10), gpu_scid) | + FIELD_PREP(GENMASK(9, 5), gpu_scid) | + FIELD_PREP(GENMASK(4, 0), gpu_scid)); + + gpu_write(gpu, REG_A6XX_GBIF_SCACHE_CNTL0, + FIELD_PREP(GENMASK(14, 10), gpu_scid) | + BIT(8)); + } + + llcc_slice_activate(a6xx_gpu->htw_llc_slice); +} + +static void a6xx_llc_slices_destroy(struct a6xx_gpu *a6xx_gpu) +{ + /* No LLCC on non-RPMh (and by extension, non-GMU) SoCs */ + if (adreno_has_gmu_wrapper(&a6xx_gpu->base)) + return; + + llcc_slice_putd(a6xx_gpu->llc_slice); + llcc_slice_putd(a6xx_gpu->htw_llc_slice); +} + +static void a6xx_llc_slices_init(struct platform_device *pdev, + struct a6xx_gpu *a6xx_gpu, bool is_a7xx) +{ + /* No LLCC on non-RPMh (and by extension, non-GMU) SoCs */ + if (adreno_has_gmu_wrapper(&a6xx_gpu->base)) + return; + + a6xx_gpu->llc_slice = llcc_slice_getd(LLCC_GPU); + a6xx_gpu->htw_llc_slice = llcc_slice_getd(LLCC_GPUHTW); +} + +#define GBIF_CLIENT_HALT_MASK BIT(0) +#define GBIF_ARB_HALT_MASK BIT(1) +#define VBIF_XIN_HALT_CTRL0_MASK GENMASK(3, 0) +#define VBIF_RESET_ACK_MASK 0xF0 +#define GPR0_GBIF_HALT_REQUEST 0x1E0 + +void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, bool gx_off) +{ + struct msm_gpu *gpu = &adreno_gpu->base; + + if (adreno_is_a619_holi(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_RBBM_GPR0_CNTL, GPR0_GBIF_HALT_REQUEST); + spin_until((gpu_read(gpu, REG_A6XX_RBBM_VBIF_GX_RESET_STATUS) & + (VBIF_RESET_ACK_MASK)) == VBIF_RESET_ACK_MASK); + } else if (!a6xx_has_gbif(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, VBIF_XIN_HALT_CTRL0_MASK); + spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) & + (VBIF_XIN_HALT_CTRL0_MASK)) == VBIF_XIN_HALT_CTRL0_MASK); + gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0); + + return; + } + + if (gx_off) { + /* Halt the gx side of GBIF */ + gpu_write(gpu, REG_A6XX_RBBM_GBIF_HALT, 1); + spin_until(gpu_read(gpu, REG_A6XX_RBBM_GBIF_HALT_ACK) & 1); + } + + /* Halt new client requests on GBIF */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_CLIENT_HALT_MASK); + spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & + (GBIF_CLIENT_HALT_MASK)) == GBIF_CLIENT_HALT_MASK); + + /* Halt all AXI requests on GBIF */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_ARB_HALT_MASK); + spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & + (GBIF_ARB_HALT_MASK)) == GBIF_ARB_HALT_MASK); + + /* The GBIF halt needs to be explicitly cleared */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0); +} + +void a6xx_gpu_sw_reset(struct msm_gpu *gpu, bool assert) +{ + /* 11nm chips (e.g. ones with A610) have hw issues with the reset line! */ + if (adreno_is_a610(to_adreno_gpu(gpu)) || adreno_is_a8xx(to_adreno_gpu(gpu))) + return; + + gpu_write(gpu, REG_A6XX_RBBM_SW_RESET_CMD, assert); + /* Perform a bogus read and add a brief delay to ensure ordering. */ + gpu_read(gpu, REG_A6XX_RBBM_SW_RESET_CMD); + udelay(1); + + /* The reset line needs to be asserted for at least 100 us */ + if (assert) + udelay(100); +} + +static int a6xx_gmu_pm_resume(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + int ret; + + gpu->needs_hw_init = true; + + trace_msm_gpu_resume(0); + + mutex_lock(&a6xx_gpu->gmu.lock); + ret = a6xx_gmu_resume(a6xx_gpu); + mutex_unlock(&a6xx_gpu->gmu.lock); + if (ret) + return ret; + + msm_devfreq_resume(gpu); + + if (adreno_is_a8xx(adreno_gpu)) + a8xx_llc_activate(a6xx_gpu); + else if (adreno_is_a7xx(adreno_gpu)) + a7xx_llc_activate(a6xx_gpu); + else + a6xx_llc_activate(a6xx_gpu); + + return ret; +} + +static int a6xx_pm_resume(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + unsigned long freq = gpu->fast_rate; + struct dev_pm_opp *opp; + int ret; + + gpu->needs_hw_init = true; + + trace_msm_gpu_resume(0); + + mutex_lock(&a6xx_gpu->gmu.lock); + + opp = dev_pm_opp_find_freq_ceil(&gpu->pdev->dev, &freq); + if (IS_ERR(opp)) { + ret = PTR_ERR(opp); + goto err_set_opp; + } + dev_pm_opp_put(opp); + + /* Set the core clock and bus bw, having VDD scaling in mind */ + dev_pm_opp_set_opp(&gpu->pdev->dev, opp); + + pm_runtime_resume_and_get(gmu->dev); + pm_runtime_resume_and_get(gmu->gxpd); + + ret = clk_bulk_prepare_enable(gpu->nr_clocks, gpu->grp_clks); + if (ret) + goto err_bulk_clk; + + ret = clk_bulk_prepare_enable(gmu->nr_clocks, gmu->clocks); + if (ret) { + clk_bulk_disable_unprepare(gpu->nr_clocks, gpu->grp_clks); + goto err_bulk_clk; + } + + if (adreno_is_a619_holi(adreno_gpu)) + a6xx_sptprac_enable(gmu); + + /* If anything goes south, tear the GPU down piece by piece.. */ + if (ret) { +err_bulk_clk: + pm_runtime_put(gmu->gxpd); + pm_runtime_put(gmu->dev); + dev_pm_opp_set_opp(&gpu->pdev->dev, NULL); + } +err_set_opp: + mutex_unlock(&a6xx_gpu->gmu.lock); + + if (!ret) { + msm_devfreq_resume(gpu); + a6xx_llc_activate(a6xx_gpu); + } + + return ret; +} + +static int a6xx_gmu_pm_suspend(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + int i, ret; + + trace_msm_gpu_suspend(0); + + a6xx_llc_deactivate(a6xx_gpu); + + msm_devfreq_suspend(gpu); + + mutex_lock(&a6xx_gpu->gmu.lock); + ret = a6xx_gmu_stop(a6xx_gpu); + mutex_unlock(&a6xx_gpu->gmu.lock); + if (ret) + return ret; + + if (a6xx_gpu->shadow_bo) + for (i = 0; i < gpu->nr_rings; i++) + a6xx_gpu->shadow[i] = 0; + + gpu->suspend_count++; + + return 0; +} + +static int a6xx_pm_suspend(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + int i; + + trace_msm_gpu_suspend(0); + + a6xx_llc_deactivate(a6xx_gpu); + + msm_devfreq_suspend(gpu); + + mutex_lock(&a6xx_gpu->gmu.lock); + + /* Drain the outstanding traffic on memory buses */ + adreno_gpu->funcs->bus_halt(adreno_gpu, true); + + if (adreno_is_a619_holi(adreno_gpu)) + a6xx_sptprac_disable(gmu); + + clk_bulk_disable_unprepare(gpu->nr_clocks, gpu->grp_clks); + clk_bulk_disable_unprepare(gmu->nr_clocks, gmu->clocks); + + pm_runtime_put_sync(gmu->gxpd); + dev_pm_opp_set_opp(&gpu->pdev->dev, NULL); + pm_runtime_put_sync(gmu->dev); + + mutex_unlock(&a6xx_gpu->gmu.lock); + + if (a6xx_gpu->shadow_bo) + for (i = 0; i < gpu->nr_rings; i++) + a6xx_gpu->shadow[i] = 0; + + gpu->suspend_count++; + + return 0; +} + +static u64 a6xx_get_timestamp(struct msm_gpu *gpu) +{ + return gpu_read64(gpu, REG_A6XX_CP_ALWAYS_ON_COUNTER); +} + +static struct msm_ringbuffer *a6xx_active_ring(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + return a6xx_gpu->cur_ring; +} + +static void a6xx_destroy(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + if (a6xx_gpu->sqe_bo) { + msm_gem_unpin_iova(a6xx_gpu->sqe_bo, gpu->vm); + drm_gem_object_put(a6xx_gpu->sqe_bo); + } + + if (a6xx_gpu->aqe_bo) { + msm_gem_unpin_iova(a6xx_gpu->aqe_bo, gpu->vm); + drm_gem_object_put(a6xx_gpu->aqe_bo); + } + + if (a6xx_gpu->shadow_bo) { + msm_gem_unpin_iova(a6xx_gpu->shadow_bo, gpu->vm); + drm_gem_object_put(a6xx_gpu->shadow_bo); + } + + a6xx_llc_slices_destroy(a6xx_gpu); + + a6xx_gmu_remove(a6xx_gpu); + + adreno_gpu_cleanup(adreno_gpu); + + kfree(a6xx_gpu); +} + +static u64 a6xx_gpu_busy(struct msm_gpu *gpu, unsigned long *out_sample_rate) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + u64 busy_cycles; + + /* 19.2MHz */ + *out_sample_rate = 19200000; + + busy_cycles = gmu_read64(&a6xx_gpu->gmu, + REG_A6XX_GMU_CX_GMU_POWER_COUNTER_XOCLK_0_L, + REG_A6XX_GMU_CX_GMU_POWER_COUNTER_XOCLK_0_H); + + return busy_cycles; +} + +static void a6xx_gpu_set_freq(struct msm_gpu *gpu, struct dev_pm_opp *opp, + bool suspended) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + mutex_lock(&a6xx_gpu->gmu.lock); + a6xx_gmu_set_freq(gpu, opp, suspended); + mutex_unlock(&a6xx_gpu->gmu.lock); +} + +static struct drm_gpuvm * +a6xx_create_vm(struct msm_gpu *gpu, struct platform_device *pdev) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + unsigned long quirks = 0; + + /* + * This allows GPU to set the bus attributes required to use system + * cache on behalf of the iommu page table walker. + */ + if (!IS_ERR_OR_NULL(a6xx_gpu->htw_llc_slice) && + !device_iommu_capable(&pdev->dev, IOMMU_CAP_CACHE_COHERENCY)) + quirks |= IO_PGTABLE_QUIRK_ARM_OUTER_WBWA; + + return adreno_iommu_create_vm(gpu, pdev, quirks); +} + +static struct drm_gpuvm * +a6xx_create_private_vm(struct msm_gpu *gpu, bool kernel_managed) +{ + struct msm_mmu *mmu; + + mmu = msm_iommu_pagetable_create(to_msm_vm(gpu->vm)->mmu, kernel_managed); + + if (IS_ERR(mmu)) + return ERR_CAST(mmu); + + return msm_gem_vm_create(gpu->dev, mmu, "gpu", ADRENO_VM_START, + adreno_private_vm_size(gpu), kernel_managed); +} + +static uint32_t a6xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + if (adreno_gpu->base.hw_apriv || a6xx_gpu->has_whereami) + return a6xx_gpu->shadow[ring->id]; + + /* + * This is true only on an A6XX_GEN1 with GMU, has IFPC enabled and a super old SQE firmware + * without 'whereami' support + */ + WARN_ONCE((to_adreno_gpu(gpu)->info->quirks & ADRENO_QUIRK_IFPC), + "Can't read CP_RB_RPTR register reliably\n"); + + return ring->memptrs->rptr = gpu_read(gpu, REG_A6XX_CP_RB_RPTR); +} + +static bool a6xx_progress(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + struct msm_cp_state cp_state; + bool progress; + + /* + * With IFPC, KMD doesn't know whether GX power domain is collapsed + * or not. So, we can't blindly read the below registers in GX domain. + * Lets trust the hang detection in HW and lie to the caller that + * there was progress. + */ + if (to_adreno_gpu(gpu)->info->quirks & ADRENO_QUIRK_IFPC) + return true; + + cp_state = (struct msm_cp_state) { + .ib1_base = gpu_read64(gpu, REG_A6XX_CP_IB1_BASE), + .ib2_base = gpu_read64(gpu, REG_A6XX_CP_IB2_BASE), + .ib1_rem = gpu_read(gpu, REG_A6XX_CP_IB1_REM_SIZE), + .ib2_rem = gpu_read(gpu, REG_A6XX_CP_IB2_REM_SIZE), + }; + + /* + * Adjust the remaining data to account for what has already been + * fetched from memory, but not yet consumed by the SQE. + * + * This is not *technically* correct, the amount buffered could + * exceed the IB size due to hw prefetching ahead, but: + * + * (1) We aren't trying to find the exact position, just whether + * progress has been made + * (2) The CP_REG_TO_MEM at the end of a submit should be enough + * to prevent prefetching into an unrelated submit. (And + * either way, at some point the ROQ will be full.) + */ + cp_state.ib1_rem += gpu_read(gpu, REG_A6XX_CP_ROQ_AVAIL_IB1) >> 16; + cp_state.ib2_rem += gpu_read(gpu, REG_A6XX_CP_ROQ_AVAIL_IB2) >> 16; + + progress = !!memcmp(&cp_state, &ring->last_cp_state, sizeof(cp_state)); + + ring->last_cp_state = cp_state; + + return progress; +} + +static void +perfcntr_select(struct msm_ringbuffer *ring, enum adreno_pipe pipe, + uint32_t regidx, uint32_t *countables, uint32_t nr, + uint32_t **reglist) +{ + OUT_PKT4(ring, regidx, nr); + for (unsigned i = 0; i < nr; i++) + OUT_RING(ring, countables[i]); + + if (!*reglist) + return; + + for (unsigned i = 0; i < nr; i++) { + /* + * Bitfield is in same position on a7xx, but only 2 bits.. + * which is sufficient for NONE/BR/BV: + */ + *(*reglist)++ = A8XX_CP_APERTURE_CNTL_HOST_PIPEID(pipe); + *(*reglist)++ = regidx + i; + *(*reglist)++ = countables[i]; + } +} + +static void +a6xx_perfcntr_configure(struct msm_gpu *gpu, struct msm_ringbuffer *ring, + const struct msm_perfcntr_stream *stream) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + enum adreno_pipe pipe = PIPE_NONE; + uint32_t *reglist = NULL; + uint32_t *reglist_sel_start; + + if (to_adreno_gpu(gpu)->info->family >= ADRENO_7XX_GEN1) { + WARN_ON(!a6xx_gpu->pwrup_reglist_emitted); + + struct cpu_gpu_lock *lock = a6xx_gpu->pwrup_reglist_ptr; + int off = (2 * lock->ifpc_list_len) + + (2 * lock->preemption_list_len) + + (3 * a6xx_gpu->dynamic_sel_reglist_offset); + + reglist = (uint32_t *)&lock->regs[0]; + reglist += off; + reglist_sel_start = reglist; + + /* Clear any previously configured SEL reg entries: */ + lock->dynamic_list_len = a6xx_gpu->dynamic_sel_reglist_offset; + + /* + * Ensure CP sees the dynamic_list_len update before we + * start modifying the SEL entries: + */ + dma_wmb(); + } + + for (unsigned i = 0; i < stream->nr_groups; i++) { + unsigned group_idx = msm_perfcntr_group_idx(stream, i); + unsigned base = msm_perfcntr_counter_base(stream, group_idx); + + const struct msm_perfcntr_group *group = + &gpu->perfcntr_groups[group_idx]; + + struct msm_perfcntr_group_state *group_state = + gpu->perfcntrs->groups[group_idx]; + + if (group->pipe != pipe) { + pipe = group->pipe; + + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + + if (pipe == PIPE_BR) { + OUT_RING(ring, CP_SET_THREAD_BR); + } else if (pipe == PIPE_BV) { + OUT_RING(ring, CP_SET_THREAD_BV); + } else { + OUT_RING(ring, CP_SET_THREAD_BOTH); + } + } + + const struct msm_perfcntr_counter *counter = &group->counters[base]; + unsigned nr = group_state->allocated_counters; + perfcntr_select(ring, pipe, counter->select_reg, + group_state->countables, nr, ®list); + + for (unsigned s = 0; s < ARRAY_SIZE(counter->slice_select_regs); s++) { + if (!counter->slice_select_regs[s]) + break; + + perfcntr_select(ring, pipe, counter->slice_select_regs[s], + group_state->countables, nr, ®list); + } + } + + if (pipe != PIPE_NONE) { + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, CP_SET_THREAD_BOTH); + } + + OUT_PKT7(ring, CP_MEM_WRITE, 3); + OUT_RING(ring, lower_32_bits(rbmemptr(ring, perfcntr_fence))); + OUT_RING(ring, upper_32_bits(rbmemptr(ring, perfcntr_fence))); + OUT_RING(ring, stream->sel_fence); + + /* + * Update the pwrup reglist size before flushing. Kgsl does a shared- + * memory spinlock dance with SQE to avoid racing with IFPC exit. But + * we can skip that since the ringbuffer programming will be executed + * by SQE after dynamic reglist size is updated. So even if we lose + * the race, the register programming in the rb will overwrite/correct + * the SEL regs restored by SQE on IFPC exit, before sampling begins. + */ + if (reglist) { + struct cpu_gpu_lock *lock = a6xx_gpu->pwrup_reglist_ptr; + unsigned nr_regs = (reglist - reglist_sel_start) / 3; + + /* + * Ensure CP sees updates to the pwrup_reglist before it + * sees the new (increased) length: + */ + dma_wmb(); + + /* Update dynamic reglist len to include new SEL reg programming: */ + lock->dynamic_list_len = a6xx_gpu->dynamic_sel_reglist_offset + nr_regs; + + WARN_ON_ONCE(reglist > (uint32_t *)((uint8_t *)lock + PWRUP_REGLIST_SIZE)); + } + + a6xx_flush_yield(gpu, ring); + + /* Check to see if we need to start preemption */ + if (adreno_is_a8xx(to_adreno_gpu(gpu))) + a8xx_preempt_trigger(gpu); + else + a6xx_preempt_trigger(gpu); +} + +static u32 fuse_to_supp_hw(const struct adreno_info *info, u32 fuse) +{ + if (!info->speedbins) + return UINT_MAX; + + for (int i = 0; info->speedbins[i].fuse != SHRT_MAX; i++) + if (info->speedbins[i].fuse == fuse) + return BIT(info->speedbins[i].speedbin); + + return UINT_MAX; +} + +static int a6xx_read_speedbin(struct device *dev, struct a6xx_gpu *a6xx_gpu, + const struct adreno_info *info, u32 *speedbin) +{ + int ret; + + /* Use speedbin fuse if present. Otherwise, fallback to softfuse */ + ret = adreno_read_speedbin(dev, speedbin); + if (ret != -ENOENT) + return ret; + + if (info->quirks & ADRENO_QUIRK_SOFTFUSE) { + *speedbin = a6xx_cx_misc_read(a6xx_gpu, REG_A8XX_CX_MISC_SW_FUSE_FREQ_LIMIT_STATUS); + *speedbin = A8XX_CX_MISC_SW_FUSE_FREQ_LIMIT_STATUS_FINALFREQLIMIT(*speedbin); + return 0; + } + + return -ENOENT; +} + +static int a6xx_set_supported_hw(struct device *dev, struct a6xx_gpu *a6xx_gpu, + const struct adreno_info *info) +{ + u32 supp_hw; + u32 speedbin; + int ret; + + ret = a6xx_read_speedbin(dev, a6xx_gpu, info, &speedbin); + /* + * -ENOENT means that the platform doesn't support speedbin which is + * fine + */ + if (ret == -ENOENT) { + return 0; + } else if (ret) { + dev_err_probe(dev, ret, + "failed to read speed-bin. Some OPPs may not be supported by hardware\n"); + return ret; + } + + supp_hw = fuse_to_supp_hw(info, speedbin); + + if (supp_hw == UINT_MAX) { + DRM_DEV_ERROR(dev, + "missing support for speed-bin: %u. Some OPPs may not be supported by hardware\n", + speedbin); + supp_hw = BIT(0); /* Default */ + } + + ret = devm_pm_opp_set_supported_hw(dev, &supp_hw, 1); + if (ret) + return ret; + + return 0; +} + +static bool a6xx_aqe_is_enabled(struct adreno_gpu *adreno_gpu) +{ + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + /* + * AQE uses preemption context record as scratch pad, so check if + * preemption is enabled + */ + return (adreno_gpu->base.nr_rings > 1) && !!a6xx_gpu->aqe_bo; +} + +static struct msm_gpu *a6xx_gpu_init(struct drm_device *dev) +{ + struct msm_drm_private *priv = dev->dev_private; + struct platform_device *pdev = priv->gpu_pdev; + struct adreno_platform_config *config = pdev->dev.platform_data; + const struct adreno_info *info = config->info; + struct device_node *phandle; + struct a6xx_gpu *a6xx_gpu; + struct adreno_gpu *adreno_gpu; + struct msm_gpu *gpu; + extern int enable_preemption; + u32 speedbin; + bool is_a7xx; + int ret, nr_rings = 1; + + a6xx_gpu = kzalloc_obj(*a6xx_gpu); + if (!a6xx_gpu) + return ERR_PTR(-ENOMEM); + + adreno_gpu = &a6xx_gpu->base; + gpu = &adreno_gpu->base; + + if ((ADRENO_6XX_GEN1 <= config->info->family) && + (config->info->family <= ADRENO_6XX_GEN4)) { + gpu->perfcntr_groups = a6xx_perfcntr_groups; + gpu->num_perfcntr_groups = a6xx_num_perfcntr_groups; + } else if ((ADRENO_7XX_GEN1 <= config->info->family) && + (config->info->family <= ADRENO_7XX_GEN3)) { + gpu->perfcntr_groups = a7xx_perfcntr_groups; + gpu->num_perfcntr_groups = a7xx_num_perfcntr_groups; + } else if ((ADRENO_8XX_GEN1 <= config->info->family) && + (config->info->family <= ADRENO_8XX_GEN2)) { + gpu->perfcntr_groups = a8xx_perfcntr_groups; + gpu->num_perfcntr_groups = a8xx_num_perfcntr_groups; + } + + mutex_init(&a6xx_gpu->gmu.lock); + spin_lock_init(&a6xx_gpu->aperture_lock); + + adreno_gpu->registers = NULL; + + /* Check if there is a GMU phandle and set it up */ + struct device_node *node __free(device_node) = + of_parse_phandle(pdev->dev.of_node, "qcom,gmu", 0); + /* FIXME: How do we gracefully handle this? */ + BUG_ON(!node); + + adreno_gpu->gmu_is_wrapper = of_device_is_compatible(node, "qcom,adreno-gmu-wrapper"); + + adreno_gpu->base.hw_apriv = + !!(info->quirks & ADRENO_QUIRK_HAS_HW_APRIV); + + /* gpu->info only gets assigned in adreno_gpu_init(). A8x is included intentionally */ + is_a7xx = info->family >= ADRENO_7XX_GEN1; + + a6xx_llc_slices_init(pdev, a6xx_gpu, is_a7xx); + + /* + * There is a different programming path for A6xx targets with an + * mmu500 attached, so detect if that is the case + */ + phandle = of_parse_phandle(pdev->dev.of_node, "iommus", 0); + a6xx_gpu->have_mmu500 = (phandle && + of_device_is_compatible(phandle, "arm,mmu-500")); + of_node_put(phandle); + + if (is_a7xx || !a6xx_gpu->have_mmu500) + a6xx_gpu->cx_misc_mmio = msm_ioremap(pdev, "cx_mem"); + else + a6xx_gpu->cx_misc_mmio = NULL; + + ret = a6xx_set_supported_hw(&pdev->dev, a6xx_gpu, info); + if (ret) { + a6xx_llc_slices_destroy(a6xx_gpu); + kfree(a6xx_gpu); + return ERR_PTR(ret); + } + + if ((enable_preemption == 1) || (enable_preemption == -1 && + (info->quirks & ADRENO_QUIRK_PREEMPTION))) + nr_rings = 4; + + ret = adreno_gpu_init(dev, pdev, adreno_gpu, info->funcs, nr_rings); + if (ret) { + a6xx_destroy(&(a6xx_gpu->base.base)); + return ERR_PTR(ret); + } + + /* Set the speedbin value that is passed to userspace */ + if (a6xx_read_speedbin(&pdev->dev, a6xx_gpu, info, &speedbin) || !speedbin) + speedbin = 0xffff; + adreno_gpu->speedbin = (uint16_t) (0xffff & speedbin); + + /* + * For now only clamp to idle freq for devices where this is known not + * to cause power supply issues: + */ + if (adreno_is_a618(adreno_gpu) || adreno_is_7c3(adreno_gpu)) + priv->gpu_clamp_to_idle = true; + + if (adreno_has_gmu_wrapper(adreno_gpu) || adreno_has_rgmu(adreno_gpu)) + ret = a6xx_gmu_wrapper_init(a6xx_gpu, node); + else + ret = a6xx_gmu_init(a6xx_gpu, node); + if (ret) { + a6xx_destroy(&(a6xx_gpu->base.base)); + return ERR_PTR(ret); + } + + adreno_gpu->uche_trap_base = 0x1fffffffff000ull; + + msm_mmu_set_fault_handler(to_msm_vm(gpu->vm)->mmu, gpu, + adreno_gpu->funcs->mmu_fault_handler); + + adreno_gpu->ubwc_config = qcom_ubwc_config_get_data(); + if (IS_ERR(adreno_gpu->ubwc_config)) { + ret = PTR_ERR(adreno_gpu->ubwc_config); + a6xx_destroy(&(a6xx_gpu->base.base)); + return ERR_PTR(ret); + } + + /* Set up the preemption specific bits and pieces for each ringbuffer */ + a6xx_preempt_init(gpu); + + return gpu; +} + +const struct adreno_gpu_funcs a6xx_gpu_funcs = { + .base = { + .get_param = adreno_get_param, + .set_param = adreno_set_param, + .hw_init = a6xx_hw_init, + .ucode_load = a6xx_ucode_load, + .pm_suspend = a6xx_gmu_pm_suspend, + .pm_resume = a6xx_gmu_pm_resume, + .recover = a6xx_recover, + .submit = a6xx_submit, + .active_ring = a6xx_active_ring, + .irq = a6xx_irq, + .destroy = a6xx_destroy, +#if defined(CONFIG_DRM_MSM_GPU_STATE) + .show = a6xx_show, +#endif + .gpu_busy = a6xx_gpu_busy, + .gpu_get_freq = a6xx_gmu_get_freq, + .gpu_set_freq = a6xx_gpu_set_freq, +#if defined(CONFIG_DRM_MSM_GPU_STATE) + .gpu_state_get = a6xx_gpu_state_get, + .gpu_state_put = a6xx_gpu_state_put, +#endif + .create_vm = a6xx_create_vm, + .create_private_vm = a6xx_create_private_vm, + .get_rptr = a6xx_get_rptr, + .progress = a6xx_progress, + .sysprof_setup = a6xx_gmu_sysprof_setup, + .perfcntr_configure = a6xx_perfcntr_configure, + }, + .init = a6xx_gpu_init, + .get_timestamp = a6xx_gmu_get_timestamp, + .bus_halt = a6xx_bus_clear_pending_transactions, + .mmu_fault_handler = a6xx_fault_handler, + .gx_is_on = a6xx_gmu_gx_is_on, +}; + +const struct adreno_gpu_funcs a6xx_gmuwrapper_funcs = { + .base = { + .get_param = adreno_get_param, + .set_param = adreno_set_param, + .hw_init = a6xx_hw_init, + .ucode_load = a6xx_ucode_load, + .pm_suspend = a6xx_pm_suspend, + .pm_resume = a6xx_pm_resume, + .recover = a6xx_recover, + .submit = a6xx_submit, + .active_ring = a6xx_active_ring, + .irq = a6xx_irq, + .destroy = a6xx_destroy, +#if defined(CONFIG_DRM_MSM_GPU_STATE) + .show = a6xx_show, +#endif + .gpu_busy = a6xx_gpu_busy, +#if defined(CONFIG_DRM_MSM_GPU_STATE) + .gpu_state_get = a6xx_gpu_state_get, + .gpu_state_put = a6xx_gpu_state_put, +#endif + .create_vm = a6xx_create_vm, + .create_private_vm = a6xx_create_private_vm, + .get_rptr = a6xx_get_rptr, + .progress = a6xx_progress, + .perfcntr_configure = a6xx_perfcntr_configure, + }, + .init = a6xx_gpu_init, + .get_timestamp = a6xx_get_timestamp, + .bus_halt = a6xx_bus_clear_pending_transactions, + .mmu_fault_handler = a6xx_fault_handler, + .gx_is_on = a6xx_gmu_gx_is_on, +}; + +const struct adreno_gpu_funcs a7xx_gpu_funcs = { + .base = { + .get_param = adreno_get_param, + .set_param = adreno_set_param, + .hw_init = a6xx_hw_init, + .ucode_load = a6xx_ucode_load, + .pm_suspend = a6xx_gmu_pm_suspend, + .pm_resume = a6xx_gmu_pm_resume, + .recover = a6xx_recover, + .submit = a7xx_submit, + .active_ring = a6xx_active_ring, + .irq = a6xx_irq, + .destroy = a6xx_destroy, +#if defined(CONFIG_DRM_MSM_GPU_STATE) + .show = a6xx_show, +#endif + .gpu_busy = a6xx_gpu_busy, + .gpu_get_freq = a6xx_gmu_get_freq, + .gpu_set_freq = a6xx_gpu_set_freq, +#if defined(CONFIG_DRM_MSM_GPU_STATE) + .gpu_state_get = a6xx_gpu_state_get, + .gpu_state_put = a6xx_gpu_state_put, +#endif + .create_vm = a6xx_create_vm, + .create_private_vm = a6xx_create_private_vm, + .get_rptr = a6xx_get_rptr, + .progress = a6xx_progress, + .sysprof_setup = a6xx_gmu_sysprof_setup, + .perfcntr_configure = a6xx_perfcntr_configure, + }, + .init = a6xx_gpu_init, + .get_timestamp = a6xx_gmu_get_timestamp, + .bus_halt = a6xx_bus_clear_pending_transactions, + .mmu_fault_handler = a6xx_fault_handler, + .gx_is_on = a7xx_gmu_gx_is_on, + .aqe_is_enabled = a6xx_aqe_is_enabled, +}; + +const struct adreno_gpu_funcs a8xx_gpu_funcs = { + .base = { + .get_param = adreno_get_param, + .set_param = adreno_set_param, + .hw_init = a8xx_hw_init, + .ucode_load = a6xx_ucode_load, + .pm_suspend = a6xx_gmu_pm_suspend, + .pm_resume = a6xx_gmu_pm_resume, + .recover = a8xx_recover, + .submit = a7xx_submit, + .active_ring = a6xx_active_ring, + .irq = a8xx_irq, + .destroy = a6xx_destroy, + .gpu_busy = a8xx_gpu_busy, + .gpu_get_freq = a6xx_gmu_get_freq, + .gpu_set_freq = a6xx_gpu_set_freq, + .create_vm = a6xx_create_vm, + .create_private_vm = a6xx_create_private_vm, + .get_rptr = a6xx_get_rptr, + .progress = a8xx_progress, + .sysprof_setup = a6xx_gmu_sysprof_setup, + .perfcntr_configure = a6xx_perfcntr_configure, + .perfcntr_flush = a8xx_perfcntr_flush, + }, + .init = a6xx_gpu_init, + .get_timestamp = a8xx_gmu_get_timestamp, + .bus_halt = a8xx_bus_clear_pending_transactions, + .mmu_fault_handler = a8xx_fault_handler, + .gx_is_on = a8xx_gmu_gx_is_on, + .aqe_is_enabled = a6xx_aqe_is_enabled, +}; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h new file mode 100644 index 000000000..b50c57f42 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h @@ -0,0 +1,349 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2017, 2019 The Linux Foundation. All rights reserved. */ + +#ifndef __A6XX_GPU_H__ +#define __A6XX_GPU_H__ + + +#include "adreno_gpu.h" +#include "a6xx_enums.xml.h" +#include "a7xx_enums.xml.h" +#include "a6xx_perfcntrs.xml.h" +#include "a7xx_perfcntrs.xml.h" +#include "a6xx.xml.h" + +#include "a6xx_gmu.h" + +extern bool hang_debug; + +struct cpu_gpu_lock { + uint32_t gpu_req; + uint32_t cpu_req; + uint32_t turn; + union { + /* a6xx: */ + struct { + uint16_t list_length; + uint16_t list_offset; + }; + /* a7xx+: */ + struct { + uint8_t ifpc_list_len; + uint8_t preemption_list_len; + uint16_t dynamic_list_len; + }; + }; + uint64_t regs[]; +}; + +/** + * struct a6xx_info - a6xx specific information from device table + * + * @hwcg: hw clock gating register sequence + * @protect: CP_PROTECT settings + * @pwrup_reglist pwrup reglist for preemption + */ +struct a6xx_info { + const struct adreno_reglist *hwcg; + const struct adreno_protect *protect; + const struct adreno_reglist_list *pwrup_reglist; + const struct adreno_reglist_pipe_list *dyn_pwrup_reglist; + const struct adreno_reglist_list *ifpc_reglist; + const struct adreno_reglist *gbif_cx; + const struct adreno_reglist_pipe *nonctxt_reglist; + u32 max_slices; + u32 gmu_chipid; + u32 gmu_cgc_mode; + u32 prim_fifo_threshold; + const struct a6xx_bcm *bcms; +}; + +struct a6xx_gpu { + struct adreno_gpu base; + + struct drm_gem_object *sqe_bo; + uint64_t sqe_iova; + struct drm_gem_object *aqe_bo; + uint64_t aqe_iova; + + struct msm_ringbuffer *cur_ring; + struct msm_ringbuffer *next_ring; + + struct drm_gem_object *preempt_bo[MSM_GPU_MAX_RINGS]; + void *preempt[MSM_GPU_MAX_RINGS]; + uint64_t preempt_iova[MSM_GPU_MAX_RINGS]; + struct drm_gem_object *preempt_smmu_bo[MSM_GPU_MAX_RINGS]; + void *preempt_smmu[MSM_GPU_MAX_RINGS]; + uint64_t preempt_smmu_iova[MSM_GPU_MAX_RINGS]; + uint32_t last_seqno[MSM_GPU_MAX_RINGS]; + + atomic_t preempt_state; + spinlock_t eval_lock; + struct timer_list preempt_timer; + + unsigned int preempt_level; + bool uses_gmem; + bool skip_save_restore; + + struct drm_gem_object *preempt_postamble_bo; + void *preempt_postamble_ptr; + uint64_t preempt_postamble_iova; + uint64_t preempt_postamble_len; + bool postamble_enabled; + + struct a6xx_gmu gmu; + + struct drm_gem_object *shadow_bo; + uint64_t shadow_iova; + uint32_t *shadow; + + struct drm_gem_object *pwrup_reglist_bo; +#define PWRUP_REGLIST_SIZE (2 * PAGE_SIZE) + void *pwrup_reglist_ptr; + uint64_t pwrup_reglist_iova; + bool pwrup_reglist_emitted; + + /* + * Offset of start of SEL regs appended to pwrup_reglist. This + * is equal to lock->dynamic_list_len if no SEL regs are appended + * to the end of the dynamic reglist. + */ + uint16_t dynamic_sel_reglist_offset; + + bool has_whereami; + + void __iomem *cx_misc_mmio; + void *llc_slice; + void *htw_llc_slice; + bool have_mmu500; + bool hung; + + u32 cached_aperture; + spinlock_t aperture_lock; + + u32 slice_mask; +}; + +#define to_a6xx_gpu(x) container_of(x, struct a6xx_gpu, base) + +/* + * In order to do lockless preemption we use a simple state machine to progress + * through the process. + * + * PREEMPT_NONE - no preemption in progress. Next state START. + * PREEMPT_START - The trigger is evaluating if preemption is possible. Next + * states: TRIGGERED, NONE + * PREEMPT_FINISH - An intermediate state before moving back to NONE. Next + * state: NONE. + * PREEMPT_TRIGGERED: A preemption has been executed on the hardware. Next + * states: FAULTED, PENDING + * PREEMPT_FAULTED: A preemption timed out (never completed). This will trigger + * recovery. Next state: N/A + * PREEMPT_PENDING: Preemption complete interrupt fired - the callback is + * checking the success of the operation. Next state: FAULTED, NONE. + */ + +enum a6xx_preempt_state { + PREEMPT_NONE = 0, + PREEMPT_START, + PREEMPT_FINISH, + PREEMPT_TRIGGERED, + PREEMPT_FAULTED, + PREEMPT_PENDING, +}; + +/* + * struct a6xx_preempt_record is a shared buffer between the microcode and the + * CPU to store the state for preemption. The record itself is much larger + * (2112k) but most of that is used by the CP for storage. + * + * There is a preemption record assigned per ringbuffer. When the CPU triggers a + * preemption, it fills out the record with the useful information (wptr, ring + * base, etc) and the microcode uses that information to set up the CP following + * the preemption. When a ring is switched out, the CP will save the ringbuffer + * state back to the record. In this way, once the records are properly set up + * the CPU can quickly switch back and forth between ringbuffers by only + * updating a few registers (often only the wptr). + * + * These are the CPU aware registers in the record: + * @magic: Must always be 0xAE399D6EUL + * @info: Type of the record - written 0 by the CPU, updated by the CP + * @errno: preemption error record + * @data: Data field in YIELD and SET_MARKER packets, Written and used by CP + * @cntl: Value of RB_CNTL written by CPU, save/restored by CP + * @rptr: Value of RB_RPTR written by CPU, save/restored by CP + * @wptr: Value of RB_WPTR written by CPU, save/restored by CP + * @_pad: Reserved/padding + * @rptr_addr: Value of RB_RPTR_ADDR_LO|HI written by CPU, save/restored by CP + * @rbase: Value of RB_BASE written by CPU, save/restored by CP + * @counter: GPU address of the storage area for the preemption counters + * @bv_rptr_addr: Value of BV_RB_RPTR_ADDR_LO|HI written by CPU, save/restored by CP + */ +struct a6xx_preempt_record { + u32 magic; + u32 info; + u32 errno; + u32 data; + u32 cntl; + u32 rptr; + u32 wptr; + u32 _pad; + u64 rptr_addr; + u64 rbase; + u64 counter; + u64 bv_rptr_addr; +}; + +#define A6XX_PREEMPT_RECORD_MAGIC 0xAE399D6EUL + +#define PREEMPT_SMMU_INFO_SIZE 4096 + +#define PREEMPT_RECORD_SIZE(adreno_gpu) \ + ((adreno_gpu->info->preempt_record_size) == 0 ? \ + 4192 * SZ_1K : (adreno_gpu->info->preempt_record_size)) + +/* + * The preemption counter block is a storage area for the value of the + * preemption counters that are saved immediately before context switch. We + * append it on to the end of the allocation for the preemption record. + */ +#define A6XX_PREEMPT_COUNTER_SIZE (16 * 4) + +struct a7xx_cp_smmu_info { + u32 magic; + u32 _pad4; + u64 ttbr0; + u32 asid; + u32 context_idr; + u32 context_bank; +}; + +#define GEN7_CP_SMMU_INFO_MAGIC 0x241350d5UL + +/* + * Given a register and a count, return a value to program into + * REG_CP_PROTECT_REG(n) - this will block both reads and writes for + * _len + 1 registers starting at _reg. + */ +#define A6XX_PROTECT_NORDWR(_reg, _len) \ + ((1 << 31) | \ + (((_len) & 0x3FFF) << 18) | ((_reg) & 0x3FFFF)) + +/* + * Same as above, but allow reads over the range. For areas of mixed use (such + * as performance counters) this allows us to protect a much larger range with a + * single register + */ +#define A6XX_PROTECT_RDONLY(_reg, _len) \ + ((((_len) & 0x3FFF) << 18) | ((_reg) & 0x3FFFF)) + +extern const struct adreno_gpu_funcs a6xx_gpu_funcs; +extern const struct adreno_gpu_funcs a6xx_gmuwrapper_funcs; +extern const struct adreno_gpu_funcs a7xx_gpu_funcs; +extern const struct adreno_gpu_funcs a8xx_gpu_funcs; + +static inline bool a6xx_has_gbif(struct adreno_gpu *gpu) +{ + if(adreno_is_a630(gpu)) + return false; + + return true; +} + +static inline void a6xx_cx_misc_rmw(struct a6xx_gpu *a6xx_gpu, u32 reg, u32 mask, u32 or) +{ + return msm_rmw(a6xx_gpu->cx_misc_mmio + (reg << 2), mask, or); +} + +static inline u32 a6xx_cx_misc_read(struct a6xx_gpu *a6xx_gpu, u32 reg) +{ + return readl(a6xx_gpu->cx_misc_mmio + (reg << 2)); +} + +static inline void a6xx_cx_misc_write(struct a6xx_gpu *a6xx_gpu, u32 reg, u32 value) +{ + writel(value, a6xx_gpu->cx_misc_mmio + (reg << 2)); +} + +#define shadowptr(_a6xx_gpu, _ring) ((_a6xx_gpu)->shadow_iova + \ + ((_ring)->id * sizeof(uint32_t))) + +int a6xx_gmu_resume(struct a6xx_gpu *gpu); +int a6xx_gmu_stop(struct a6xx_gpu *gpu); + +int a6xx_gmu_wait_for_idle(struct a6xx_gmu *gmu); + +bool a6xx_gmu_isidle(struct a6xx_gmu *gmu); + +int a6xx_gmu_set_oob(struct a6xx_gmu *gmu, enum a6xx_gmu_oob_state state); +void a6xx_gmu_clear_oob(struct a6xx_gmu *gmu, enum a6xx_gmu_oob_state state); + +int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node); +int a6xx_gmu_wrapper_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node); +void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu); +void a6xx_gmu_sysprof_setup(struct msm_gpu *gpu, bool force_on); + +void a6xx_preempt_init(struct msm_gpu *gpu); +void a6xx_preempt_hw_init(struct msm_gpu *gpu); +void a6xx_preempt_trigger(struct msm_gpu *gpu); +void a6xx_preempt_irq(struct msm_gpu *gpu); +void a6xx_preempt_fini(struct msm_gpu *gpu); +void a6xx_emit_set_pseudo_reg(struct msm_ringbuffer *ring, + struct a6xx_gpu *a6xx_gpu, struct msm_gpu_submitqueue *queue); +int a6xx_preempt_submitqueue_setup(struct msm_gpu *gpu, + struct msm_gpu_submitqueue *queue); +void a6xx_preempt_submitqueue_close(struct msm_gpu *gpu, + struct msm_gpu_submitqueue *queue); + +/* Return true if we are in a preempt state */ +static inline bool a6xx_in_preempt(struct a6xx_gpu *a6xx_gpu) +{ + /* + * Make sure the read to preempt_state is ordered with respect to reads + * of other variables before ... + */ + smp_rmb(); + + int preempt_state = atomic_read(&a6xx_gpu->preempt_state); + + /* ... and after. */ + smp_rmb(); + + return !(preempt_state == PREEMPT_NONE || + preempt_state == PREEMPT_FINISH); +} + +void a6xx_gmu_set_freq(struct msm_gpu *gpu, struct dev_pm_opp *opp, + bool suspended); +unsigned long a6xx_gmu_get_freq(struct msm_gpu *gpu); + +void a6xx_show(struct msm_gpu *gpu, struct msm_gpu_state *state, + struct drm_printer *p); + +struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu); +int a6xx_gpu_state_put(struct msm_gpu_state *state); + +void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, bool gx_off); +void a6xx_gpu_sw_reset(struct msm_gpu *gpu, bool assert); +int a6xx_fenced_write(struct a6xx_gpu *gpu, u32 offset, u64 value, u32 mask, bool is_64b); +void a6xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring); +void a6xx_flush_yield(struct msm_gpu *gpu, struct msm_ringbuffer *ring); +int a6xx_zap_shader_init(struct msm_gpu *gpu); + +void a8xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, bool gx_off); +int a8xx_fault_handler(void *arg, unsigned long iova, int flags, void *data); +void a8xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring); +u64 a8xx_gmu_get_timestamp(struct msm_gpu *gpu); +u64 a8xx_gpu_busy(struct msm_gpu *gpu, unsigned long *out_sample_rate); +int a8xx_gpu_feature_probe(struct msm_gpu *gpu); +void a8xx_gpu_get_slice_info(struct msm_gpu *gpu); +int a8xx_hw_init(struct msm_gpu *gpu); +irqreturn_t a8xx_irq(struct msm_gpu *gpu); +void a8xx_llc_activate(struct a6xx_gpu *a6xx_gpu); +void a8xx_preempt_hw_init(struct msm_gpu *gpu); +void a8xx_preempt_trigger(struct msm_gpu *gpu); +void a8xx_preempt_irq(struct msm_gpu *gpu); +bool a8xx_progress(struct msm_gpu *gpu, struct msm_ringbuffer *ring); +void a8xx_perfcntr_flush(struct msm_gpu *gpu); +void a8xx_recover(struct msm_gpu *gpu); +#endif /* __A6XX_GPU_H__ */ diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c new file mode 100644 index 000000000..26c17861c --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c @@ -0,0 +1,2135 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2018-2019 The Linux Foundation. All rights reserved. */ + +#include +#include "msm_gem.h" +#include "a6xx_gpu.h" +#include "a6xx_gmu.h" +#include "a6xx_gpu_state.h" +#include "a6xx_gmu.xml.h" + +static const unsigned int *gen7_0_0_external_core_regs[] __always_unused; +static const unsigned int *gen7_2_0_external_core_regs[] __always_unused; +static const unsigned int *gen7_9_0_external_core_regs[] __always_unused; +static const unsigned int *gen7_17_0_external_core_regs[] __always_unused; +static const struct gen7_sptp_cluster_registers gen7_9_0_sptp_clusters[] __always_unused; +static const u32 gen7_9_0_cx_debugbus_blocks[] __always_unused; + +#include "adreno_gen7_0_0_snapshot.h" +#include "adreno_gen7_2_0_snapshot.h" +#include "adreno_gen7_9_0_snapshot.h" +#include "adreno_gen7_17_0_snapshot.h" + +struct a6xx_gpu_state_obj { + const void *handle; + u32 *data; + u32 count; /* optional, used when count potentially read from hw */ +}; + +struct a6xx_gpu_state { + struct msm_gpu_state base; + + struct a6xx_gpu_state_obj *gmu_registers; + int nr_gmu_registers; + + struct a6xx_gpu_state_obj *registers; + int nr_registers; + + struct a6xx_gpu_state_obj *shaders; + int nr_shaders; + + struct a6xx_gpu_state_obj *clusters; + int nr_clusters; + + struct a6xx_gpu_state_obj *dbgahb_clusters; + int nr_dbgahb_clusters; + + struct a6xx_gpu_state_obj *indexed_regs; + int nr_indexed_regs; + + struct a6xx_gpu_state_obj *debugbus; + int nr_debugbus; + + struct a6xx_gpu_state_obj *vbif_debugbus; + + struct a6xx_gpu_state_obj *cx_debugbus; + int nr_cx_debugbus; + + struct msm_gpu_state_bo *gmu_log; + struct msm_gpu_state_bo *gmu_hfi; + struct msm_gpu_state_bo *gmu_debug; + + s32 hfi_queue_history[HFI_MAX_QUEUES][HFI_HISTORY_SZ]; + + struct list_head objs; + + bool gpu_initialized; +}; + +static inline int CRASHDUMP_WRITE(u64 *in, u32 reg, u32 val) +{ + in[0] = val; + in[1] = (((u64) reg) << 44 | (1 << 21) | 1); + + return 2; +} + +static inline int CRASHDUMP_READ(u64 *in, u32 reg, u32 dwords, u64 target) +{ + in[0] = target; + in[1] = (((u64) reg) << 44 | dwords); + + return 2; +} + +static inline int CRASHDUMP_FINI(u64 *in) +{ + in[0] = 0; + in[1] = 0; + + return 2; +} + +struct a6xx_crashdumper { + void *ptr; + struct drm_gem_object *bo; + u64 iova; +}; + +struct a6xx_state_memobj { + struct list_head node; + unsigned long long data[]; +}; + +static void *state_kcalloc(struct a6xx_gpu_state *a6xx_state, int nr, size_t objsize) +{ + struct a6xx_state_memobj *obj = + kvzalloc((nr * objsize) + sizeof(*obj), GFP_KERNEL); + + if (!obj) + return NULL; + + list_add_tail(&obj->node, &a6xx_state->objs); + return &obj->data; +} + +static void *state_kmemdup(struct a6xx_gpu_state *a6xx_state, void *src, + size_t size) +{ + void *dst = state_kcalloc(a6xx_state, 1, size); + + if (dst) + memcpy(dst, src, size); + return dst; +} + +/* + * Allocate 1MB for the crashdumper scratch region - 8k for the script and + * the rest for the data + */ +#define A6XX_CD_DATA_OFFSET 8192 +#define A6XX_CD_DATA_SIZE (SZ_1M - 8192) + +static int a6xx_crashdumper_init(struct msm_gpu *gpu, + struct a6xx_crashdumper *dumper) +{ + dumper->ptr = msm_gem_kernel_new(gpu->dev, + SZ_1M, MSM_BO_WC, gpu->vm, + &dumper->bo, &dumper->iova); + + if (!IS_ERR(dumper->ptr)) + msm_gem_object_set_name(dumper->bo, "crashdump"); + + return PTR_ERR_OR_ZERO(dumper->ptr); +} + +static int a6xx_crashdumper_run(struct msm_gpu *gpu, + struct a6xx_crashdumper *dumper) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + u32 val; + int ret; + + if (IS_ERR_OR_NULL(dumper->ptr)) + return -EINVAL; + + if (!a6xx_gmu_sptprac_is_on(&a6xx_gpu->gmu)) + return -EINVAL; + + /* Make sure all pending memory writes are posted */ + wmb(); + + gpu_write64(gpu, REG_A6XX_CP_CRASH_DUMP_SCRIPT_BASE, dumper->iova); + + gpu_write(gpu, REG_A6XX_CP_CRASH_DUMP_CNTL, 1); + + ret = gpu_poll_timeout(gpu, REG_A6XX_CP_CRASH_DUMP_STATUS, val, + val & 0x02, 100, 10000); + + gpu_write(gpu, REG_A6XX_CP_CRASH_DUMP_CNTL, 0); + + return ret; +} + +/* read a value from the GX debug bus */ +static int debugbus_read(struct msm_gpu *gpu, u32 block, u32 offset, + u32 *data) +{ + u32 reg; + + if (to_adreno_gpu(gpu)->info->family >= ADRENO_7XX_GEN1) { + reg = A7XX_DBGC_CFG_DBGBUS_SEL_D_PING_INDEX(offset) | + A7XX_DBGC_CFG_DBGBUS_SEL_D_PING_BLK_SEL(block); + } else { + reg = A6XX_DBGC_CFG_DBGBUS_SEL_D_PING_INDEX(offset) | + A6XX_DBGC_CFG_DBGBUS_SEL_D_PING_BLK_SEL(block); + } + + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_SEL_A, reg); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_SEL_B, reg); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_SEL_C, reg); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_SEL_D, reg); + + /* Wait 1 us to make sure the data is flowing */ + udelay(1); + + data[0] = gpu_read(gpu, REG_A6XX_DBGC_CFG_DBGBUS_TRACE_BUF2); + data[1] = gpu_read(gpu, REG_A6XX_DBGC_CFG_DBGBUS_TRACE_BUF1); + + return 2; +} + +#define cxdbg_write(ptr, offset, val) \ + writel((val), (ptr) + ((offset) << 2)) + +#define cxdbg_read(ptr, offset) \ + readl((ptr) + ((offset) << 2)) + +/* read a value from the CX debug bus */ +static int cx_debugbus_read(struct msm_gpu *gpu, void __iomem *cxdbg, u32 block, u32 offset, + u32 *data) +{ + u32 reg; + + if (to_adreno_gpu(gpu)->info->family >= ADRENO_7XX_GEN1) { + reg = A7XX_CX_DBGC_CFG_DBGBUS_SEL_A_PING_INDEX(offset) | + A7XX_CX_DBGC_CFG_DBGBUS_SEL_A_PING_BLK_SEL(block); + } else { + reg = A6XX_CX_DBGC_CFG_DBGBUS_SEL_A_PING_INDEX(offset) | + A6XX_CX_DBGC_CFG_DBGBUS_SEL_A_PING_BLK_SEL(block); + } + + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_SEL_A, reg); + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_SEL_B, reg); + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_SEL_C, reg); + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_SEL_D, reg); + + /* Wait 1 us to make sure the data is flowing */ + udelay(1); + + data[0] = cxdbg_read(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_TRACE_BUF2); + data[1] = cxdbg_read(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_TRACE_BUF1); + + return 2; +} + +/* Read a chunk of data from the VBIF debug bus */ +static int vbif_debugbus_read(struct msm_gpu *gpu, u32 ctrl0, u32 ctrl1, + u32 reg, int count, u32 *data) +{ + int i; + + gpu_write(gpu, ctrl0, reg); + + for (i = 0; i < count; i++) { + gpu_write(gpu, ctrl1, i); + data[i] = gpu_read(gpu, REG_A6XX_VBIF_TEST_BUS_OUT); + } + + return count; +} + +#define AXI_ARB_BLOCKS 2 +#define XIN_AXI_BLOCKS 5 +#define XIN_CORE_BLOCKS 4 + +#define VBIF_DEBUGBUS_BLOCK_SIZE \ + ((16 * AXI_ARB_BLOCKS) + \ + (18 * XIN_AXI_BLOCKS) + \ + (12 * XIN_CORE_BLOCKS)) + +static void a6xx_get_vbif_debugbus_block(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + struct a6xx_gpu_state_obj *obj) +{ + u32 clk, *ptr; + int i; + + obj->data = state_kcalloc(a6xx_state, VBIF_DEBUGBUS_BLOCK_SIZE, + sizeof(u32)); + if (!obj->data) + return; + + obj->handle = NULL; + + /* Get the current clock setting */ + clk = gpu_read(gpu, REG_A6XX_VBIF_CLKON); + + /* Force on the bus so we can read it */ + gpu_write(gpu, REG_A6XX_VBIF_CLKON, + clk | A6XX_VBIF_CLKON_FORCE_ON_TESTBUS); + + /* We will read from BUS2 first, so disable BUS1 */ + gpu_write(gpu, REG_A6XX_VBIF_TEST_BUS1_CTRL0, 0); + + /* Enable the VBIF bus for reading */ + gpu_write(gpu, REG_A6XX_VBIF_TEST_BUS_OUT_CTRL, 1); + + ptr = obj->data; + + for (i = 0; i < AXI_ARB_BLOCKS; i++) + ptr += vbif_debugbus_read(gpu, + REG_A6XX_VBIF_TEST_BUS2_CTRL0, + REG_A6XX_VBIF_TEST_BUS2_CTRL1, + 1 << (i + 16), 16, ptr); + + for (i = 0; i < XIN_AXI_BLOCKS; i++) + ptr += vbif_debugbus_read(gpu, + REG_A6XX_VBIF_TEST_BUS2_CTRL0, + REG_A6XX_VBIF_TEST_BUS2_CTRL1, + 1 << i, 18, ptr); + + /* Stop BUS2 so we can turn on BUS1 */ + gpu_write(gpu, REG_A6XX_VBIF_TEST_BUS2_CTRL0, 0); + + for (i = 0; i < XIN_CORE_BLOCKS; i++) + ptr += vbif_debugbus_read(gpu, + REG_A6XX_VBIF_TEST_BUS1_CTRL0, + REG_A6XX_VBIF_TEST_BUS1_CTRL1, + 1 << i, 12, ptr); + + /* Restore the VBIF clock setting */ + gpu_write(gpu, REG_A6XX_VBIF_CLKON, clk); +} + +static void a6xx_get_debugbus_block(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_debugbus_block *block, + struct a6xx_gpu_state_obj *obj) +{ + int i; + u32 *ptr; + + obj->data = state_kcalloc(a6xx_state, block->count, sizeof(u64)); + if (!obj->data) + return; + + obj->handle = block; + + for (ptr = obj->data, i = 0; i < block->count; i++) + ptr += debugbus_read(gpu, block->id, i, ptr); +} + +static void a6xx_get_cx_debugbus_block(struct msm_gpu *gpu, + void __iomem *cxdbg, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_debugbus_block *block, + struct a6xx_gpu_state_obj *obj) +{ + int i; + u32 *ptr; + + obj->data = state_kcalloc(a6xx_state, block->count, sizeof(u64)); + if (!obj->data) + return; + + obj->handle = block; + + for (ptr = obj->data, i = 0; i < block->count; i++) + ptr += cx_debugbus_read(gpu, cxdbg, block->id, i, ptr); +} + +static void a6xx_get_debugbus_blocks(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state) +{ + int nr_debugbus_blocks = ARRAY_SIZE(a6xx_debugbus_blocks) + + (a6xx_has_gbif(to_adreno_gpu(gpu)) ? 1 : 0); + + if (adreno_is_a650_family(to_adreno_gpu(gpu))) + nr_debugbus_blocks += ARRAY_SIZE(a650_debugbus_blocks); + + a6xx_state->debugbus = state_kcalloc(a6xx_state, nr_debugbus_blocks, + sizeof(*a6xx_state->debugbus)); + + if (a6xx_state->debugbus) { + int i, j; + + for (i = 0; i < ARRAY_SIZE(a6xx_debugbus_blocks); i++) + a6xx_get_debugbus_block(gpu, + a6xx_state, + &a6xx_debugbus_blocks[i], + &a6xx_state->debugbus[i]); + + /* + * GBIF has same debugbus as of other GPU blocks, fall back to + * default path if GPU uses GBIF, also GBIF uses exactly same + * ID as of VBIF. + */ + if (a6xx_has_gbif(to_adreno_gpu(gpu))) { + a6xx_get_debugbus_block(gpu, a6xx_state, + &a6xx_gbif_debugbus_block, + &a6xx_state->debugbus[i]); + + i++; + } + + + if (adreno_is_a650_family(to_adreno_gpu(gpu))) { + for (j = 0; j < ARRAY_SIZE(a650_debugbus_blocks); i++, j++) + a6xx_get_debugbus_block(gpu, + a6xx_state, + &a650_debugbus_blocks[j], + &a6xx_state->debugbus[i]); + } + + a6xx_state->nr_debugbus = i; + } +} + +static void a7xx_get_debugbus_blocks(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + int debugbus_blocks_count, gbif_debugbus_blocks_count, total_debugbus_blocks; + const u32 *debugbus_blocks, *gbif_debugbus_blocks; + int i; + + if (adreno_gpu->info->family == ADRENO_7XX_GEN1) { + if (adreno_is_a722(adreno_gpu)) { + debugbus_blocks = gen7_17_0_debugbus_blocks; + debugbus_blocks_count = ARRAY_SIZE(gen7_17_0_debugbus_blocks); + } else { + debugbus_blocks = gen7_0_0_debugbus_blocks; + debugbus_blocks_count = ARRAY_SIZE(gen7_0_0_debugbus_blocks); + } + gbif_debugbus_blocks = a7xx_gbif_debugbus_blocks; + gbif_debugbus_blocks_count = ARRAY_SIZE(a7xx_gbif_debugbus_blocks); + } else if (adreno_gpu->info->family == ADRENO_7XX_GEN2) { + debugbus_blocks = gen7_2_0_debugbus_blocks; + debugbus_blocks_count = ARRAY_SIZE(gen7_2_0_debugbus_blocks); + gbif_debugbus_blocks = a7xx_gbif_debugbus_blocks; + gbif_debugbus_blocks_count = ARRAY_SIZE(a7xx_gbif_debugbus_blocks); + } else { + BUG_ON(adreno_gpu->info->family != ADRENO_7XX_GEN3); + debugbus_blocks = gen7_9_0_debugbus_blocks; + debugbus_blocks_count = ARRAY_SIZE(gen7_9_0_debugbus_blocks); + gbif_debugbus_blocks = gen7_9_0_gbif_debugbus_blocks; + gbif_debugbus_blocks_count = ARRAY_SIZE(gen7_9_0_gbif_debugbus_blocks); + } + + total_debugbus_blocks = debugbus_blocks_count + gbif_debugbus_blocks_count; + + a6xx_state->debugbus = state_kcalloc(a6xx_state, total_debugbus_blocks, + sizeof(*a6xx_state->debugbus)); + + if (a6xx_state->debugbus) { + for (i = 0; i < debugbus_blocks_count; i++) { + a6xx_get_debugbus_block(gpu, + a6xx_state, &a7xx_debugbus_blocks[debugbus_blocks[i]], + &a6xx_state->debugbus[i]); + } + + for (i = 0; i < gbif_debugbus_blocks_count; i++) { + a6xx_get_debugbus_block(gpu, + a6xx_state, &a7xx_debugbus_blocks[gbif_debugbus_blocks[i]], + &a6xx_state->debugbus[i + debugbus_blocks_count]); + } + + a6xx_state->nr_debugbus = total_debugbus_blocks; + } +} + +static void a6xx_get_debugbus(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct resource *res; + void __iomem *cxdbg = NULL; + + /* Set up the GX debug bus */ + + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_CNTLT, + A6XX_DBGC_CFG_DBGBUS_CNTLT_SEGT(0xf)); + + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_CNTLM, + A6XX_DBGC_CFG_DBGBUS_CNTLM_ENABLE(0xf)); + + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_IVTL_0, 0); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_IVTL_1, 0); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_IVTL_2, 0); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_IVTL_3, 0); + + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_BYTEL_0, 0x76543210); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_BYTEL_1, 0xFEDCBA98); + + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_MASKL_0, 0); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_MASKL_1, 0); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_MASKL_2, 0); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_MASKL_3, 0); + + /* Set up the CX debug bus - it lives elsewhere in the system so do a + * temporary ioremap for the registers + */ + res = platform_get_resource_byname(gpu->pdev, IORESOURCE_MEM, + "cx_dbgc"); + + if (res) + cxdbg = ioremap(res->start, resource_size(res)); + + if (cxdbg) { + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_CNTLT, + A6XX_DBGC_CFG_DBGBUS_CNTLT_SEGT(0xf)); + + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_CNTLM, + A6XX_DBGC_CFG_DBGBUS_CNTLM_ENABLE(0xf)); + + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_IVTL_0, 0); + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_IVTL_1, 0); + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_IVTL_2, 0); + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_IVTL_3, 0); + + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_BYTEL_0, + 0x76543210); + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_BYTEL_1, + 0xFEDCBA98); + + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_MASKL_0, 0); + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_MASKL_1, 0); + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_MASKL_2, 0); + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_MASKL_3, 0); + } + + if (adreno_is_a7xx(adreno_gpu)) { + a7xx_get_debugbus_blocks(gpu, a6xx_state); + } else { + a6xx_get_debugbus_blocks(gpu, a6xx_state); + } + + /* Dump the VBIF debugbus on applicable targets */ + if (!a6xx_has_gbif(adreno_gpu)) { + a6xx_state->vbif_debugbus = + state_kcalloc(a6xx_state, 1, + sizeof(*a6xx_state->vbif_debugbus)); + + if (a6xx_state->vbif_debugbus) + a6xx_get_vbif_debugbus_block(gpu, a6xx_state, + a6xx_state->vbif_debugbus); + } + + if (cxdbg) { + unsigned nr_cx_debugbus_blocks; + const struct a6xx_debugbus_block *cx_debugbus_blocks; + + if (adreno_is_a7xx(adreno_gpu)) { + BUG_ON(adreno_gpu->info->family > ADRENO_7XX_GEN3); + cx_debugbus_blocks = a7xx_cx_debugbus_blocks; + nr_cx_debugbus_blocks = ARRAY_SIZE(a7xx_cx_debugbus_blocks); + } else { + cx_debugbus_blocks = a6xx_cx_debugbus_blocks; + nr_cx_debugbus_blocks = ARRAY_SIZE(a6xx_cx_debugbus_blocks); + } + + a6xx_state->cx_debugbus = + state_kcalloc(a6xx_state, + nr_cx_debugbus_blocks, + sizeof(*a6xx_state->cx_debugbus)); + + if (a6xx_state->cx_debugbus) { + int i; + + for (i = 0; i < nr_cx_debugbus_blocks; i++) + a6xx_get_cx_debugbus_block(gpu, + cxdbg, + a6xx_state, + &cx_debugbus_blocks[i], + &a6xx_state->cx_debugbus[i]); + + a6xx_state->nr_cx_debugbus = + nr_cx_debugbus_blocks; + } + + iounmap(cxdbg); + } +} + +#define RANGE(reg, a) ((reg)[(a) + 1] - (reg)[(a)] + 1) + +/* Read a data cluster from behind the AHB aperture */ +static void a6xx_get_dbgahb_cluster(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_dbgahb_cluster *dbgahb, + struct a6xx_gpu_state_obj *obj, + struct a6xx_crashdumper *dumper) +{ + u64 *in = dumper->ptr; + u64 out = dumper->iova + A6XX_CD_DATA_OFFSET; + size_t datasize; + int i, regcount = 0; + + for (i = 0; i < A6XX_NUM_CONTEXTS; i++) { + int j; + + in += CRASHDUMP_WRITE(in, REG_A6XX_HLSQ_DBG_READ_SEL, + (dbgahb->statetype + i * 2) << 8); + + for (j = 0; j < dbgahb->count; j += 2) { + int count = RANGE(dbgahb->registers, j); + u32 offset = REG_A6XX_HLSQ_DBG_AHB_READ_APERTURE + + dbgahb->registers[j] - (dbgahb->base >> 2); + + in += CRASHDUMP_READ(in, offset, count, out); + + out += count * sizeof(u32); + + if (i == 0) + regcount += count; + } + } + + CRASHDUMP_FINI(in); + + datasize = regcount * A6XX_NUM_CONTEXTS * sizeof(u32); + + if (WARN_ON(datasize > A6XX_CD_DATA_SIZE)) + return; + + if (a6xx_crashdumper_run(gpu, dumper)) + return; + + obj->handle = dbgahb; + obj->data = state_kmemdup(a6xx_state, dumper->ptr + A6XX_CD_DATA_OFFSET, + datasize); +} + +static void a7xx_get_dbgahb_cluster(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct gen7_sptp_cluster_registers *dbgahb, + struct a6xx_gpu_state_obj *obj, + struct a6xx_crashdumper *dumper) +{ + u64 *in = dumper->ptr; + u64 out = dumper->iova + A6XX_CD_DATA_OFFSET; + size_t datasize; + int i, regcount = 0; + + in += CRASHDUMP_WRITE(in, REG_A7XX_SP_READ_SEL, + A7XX_SP_READ_SEL_LOCATION(dbgahb->location_id) | + A7XX_SP_READ_SEL_PIPE(dbgahb->pipe_id) | + A7XX_SP_READ_SEL_STATETYPE(dbgahb->statetype)); + + for (i = 0; dbgahb->regs[i] != UINT_MAX; i += 2) { + int count = RANGE(dbgahb->regs, i); + u32 offset = REG_A7XX_SP_AHB_READ_APERTURE + + dbgahb->regs[i] - dbgahb->regbase; + + in += CRASHDUMP_READ(in, offset, count, out); + + out += count * sizeof(u32); + regcount += count; + } + + CRASHDUMP_FINI(in); + + datasize = regcount * sizeof(u32); + + if (WARN_ON(datasize > A6XX_CD_DATA_SIZE)) + return; + + if (a6xx_crashdumper_run(gpu, dumper)) + return; + + obj->handle = dbgahb; + obj->data = state_kmemdup(a6xx_state, dumper->ptr + A6XX_CD_DATA_OFFSET, + datasize); +} + +static void a6xx_get_dbgahb_clusters(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + struct a6xx_crashdumper *dumper) +{ + int i; + + a6xx_state->dbgahb_clusters = state_kcalloc(a6xx_state, + ARRAY_SIZE(a6xx_dbgahb_clusters), + sizeof(*a6xx_state->dbgahb_clusters)); + + if (!a6xx_state->dbgahb_clusters) + return; + + a6xx_state->nr_dbgahb_clusters = ARRAY_SIZE(a6xx_dbgahb_clusters); + + for (i = 0; i < ARRAY_SIZE(a6xx_dbgahb_clusters); i++) + a6xx_get_dbgahb_cluster(gpu, a6xx_state, + &a6xx_dbgahb_clusters[i], + &a6xx_state->dbgahb_clusters[i], dumper); +} + +static void a7xx_get_dbgahb_clusters(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + struct a6xx_crashdumper *dumper) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + int i; + const struct gen7_sptp_cluster_registers *dbgahb_clusters; + unsigned dbgahb_clusters_size; + + if (adreno_gpu->info->family == ADRENO_7XX_GEN1) { + if (adreno_is_a722(adreno_gpu)) { + dbgahb_clusters = gen7_17_0_sptp_clusters; + dbgahb_clusters_size = ARRAY_SIZE(gen7_17_0_sptp_clusters); + } else { + dbgahb_clusters = gen7_0_0_sptp_clusters; + dbgahb_clusters_size = ARRAY_SIZE(gen7_0_0_sptp_clusters); + } + } else if (adreno_gpu->info->family == ADRENO_7XX_GEN2) { + dbgahb_clusters = gen7_2_0_sptp_clusters; + dbgahb_clusters_size = ARRAY_SIZE(gen7_2_0_sptp_clusters); + } else { + BUG_ON(adreno_gpu->info->family != ADRENO_7XX_GEN3); + dbgahb_clusters = gen7_9_0_sptp_clusters; + dbgahb_clusters_size = ARRAY_SIZE(gen7_9_0_sptp_clusters); + } + + a6xx_state->dbgahb_clusters = state_kcalloc(a6xx_state, + dbgahb_clusters_size, + sizeof(*a6xx_state->dbgahb_clusters)); + + if (!a6xx_state->dbgahb_clusters) + return; + + a6xx_state->nr_dbgahb_clusters = dbgahb_clusters_size; + + for (i = 0; i < dbgahb_clusters_size; i++) + a7xx_get_dbgahb_cluster(gpu, a6xx_state, + &dbgahb_clusters[i], + &a6xx_state->dbgahb_clusters[i], dumper); +} + +/* Read a data cluster from the CP aperture with the crashdumper */ +static void a6xx_get_cluster(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_cluster *cluster, + struct a6xx_gpu_state_obj *obj, + struct a6xx_crashdumper *dumper) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + u64 *in = dumper->ptr; + u64 out = dumper->iova + A6XX_CD_DATA_OFFSET; + size_t datasize; + int i, regcount = 0; + u32 id = cluster->id; + + /* Skip registers that are not present on older generation */ + if (!adreno_is_a660_family(adreno_gpu) && + cluster->registers == a660_fe_cluster) + return; + + if (adreno_is_a650_family(adreno_gpu) && + cluster->registers == a6xx_ps_cluster) + id = CLUSTER_VPC_PS; + + /* Some clusters need a selector register to be programmed too */ + if (cluster->sel_reg) + in += CRASHDUMP_WRITE(in, cluster->sel_reg, cluster->sel_val); + + for (i = 0; i < A6XX_NUM_CONTEXTS; i++) { + int j; + + in += CRASHDUMP_WRITE(in, REG_A6XX_CP_APERTURE_CNTL_CD, + (id << 8) | (i << 4) | i); + + for (j = 0; j < cluster->count; j += 2) { + int count = RANGE(cluster->registers, j); + + in += CRASHDUMP_READ(in, cluster->registers[j], + count, out); + + out += count * sizeof(u32); + + if (i == 0) + regcount += count; + } + } + + CRASHDUMP_FINI(in); + + datasize = regcount * A6XX_NUM_CONTEXTS * sizeof(u32); + + if (WARN_ON(datasize > A6XX_CD_DATA_SIZE)) + return; + + if (a6xx_crashdumper_run(gpu, dumper)) + return; + + obj->handle = cluster; + obj->data = state_kmemdup(a6xx_state, dumper->ptr + A6XX_CD_DATA_OFFSET, + datasize); +} + +static void a7xx_get_cluster(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct gen7_cluster_registers *cluster, + struct a6xx_gpu_state_obj *obj, + struct a6xx_crashdumper *dumper) +{ + u64 *in = dumper->ptr; + u64 out = dumper->iova + A6XX_CD_DATA_OFFSET; + size_t datasize; + int i, regcount = 0; + + in += CRASHDUMP_WRITE(in, REG_A7XX_CP_APERTURE_CNTL_CD, + A7XX_CP_APERTURE_CNTL_CD_PIPE(cluster->pipe_id) | + A7XX_CP_APERTURE_CNTL_CD_CLUSTER(cluster->cluster_id) | + A7XX_CP_APERTURE_CNTL_CD_CONTEXT(cluster->context_id)); + + /* Some clusters need a selector register to be programmed too */ + if (cluster->sel) + in += CRASHDUMP_WRITE(in, cluster->sel->cd_reg, cluster->sel->val); + + for (i = 0; cluster->regs[i] != UINT_MAX; i += 2) { + int count = RANGE(cluster->regs, i); + + in += CRASHDUMP_READ(in, cluster->regs[i], + count, out); + + out += count * sizeof(u32); + regcount += count; + } + + CRASHDUMP_FINI(in); + + datasize = regcount * sizeof(u32); + + if (WARN_ON(datasize > A6XX_CD_DATA_SIZE)) + return; + + if (a6xx_crashdumper_run(gpu, dumper)) + return; + + obj->handle = cluster; + obj->data = state_kmemdup(a6xx_state, dumper->ptr + A6XX_CD_DATA_OFFSET, + datasize); +} + +static void a6xx_get_clusters(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + struct a6xx_crashdumper *dumper) +{ + int i; + + a6xx_state->clusters = state_kcalloc(a6xx_state, + ARRAY_SIZE(a6xx_clusters), sizeof(*a6xx_state->clusters)); + + if (!a6xx_state->clusters) + return; + + a6xx_state->nr_clusters = ARRAY_SIZE(a6xx_clusters); + + for (i = 0; i < ARRAY_SIZE(a6xx_clusters); i++) + a6xx_get_cluster(gpu, a6xx_state, &a6xx_clusters[i], + &a6xx_state->clusters[i], dumper); +} + +static void a7xx_get_clusters(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + struct a6xx_crashdumper *dumper) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + int i; + const struct gen7_cluster_registers *clusters; + unsigned clusters_size; + + if (adreno_gpu->info->family == ADRENO_7XX_GEN1) { + if (adreno_is_a722(adreno_gpu)) { + clusters = gen7_17_0_clusters; + clusters_size = ARRAY_SIZE(gen7_17_0_clusters); + } else { + clusters = gen7_0_0_clusters; + clusters_size = ARRAY_SIZE(gen7_0_0_clusters); + } + } else if (adreno_gpu->info->family == ADRENO_7XX_GEN2) { + clusters = gen7_2_0_clusters; + clusters_size = ARRAY_SIZE(gen7_2_0_clusters); + } else { + BUG_ON(adreno_gpu->info->family != ADRENO_7XX_GEN3); + clusters = gen7_9_0_clusters; + clusters_size = ARRAY_SIZE(gen7_9_0_clusters); + } + + a6xx_state->clusters = state_kcalloc(a6xx_state, + clusters_size, sizeof(*a6xx_state->clusters)); + + if (!a6xx_state->clusters) + return; + + a6xx_state->nr_clusters = clusters_size; + + for (i = 0; i < clusters_size; i++) + a7xx_get_cluster(gpu, a6xx_state, &clusters[i], + &a6xx_state->clusters[i], dumper); +} + +/* Read a shader / debug block from the HLSQ aperture with the crashdumper */ +static void a6xx_get_shader_block(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_shader_block *block, + struct a6xx_gpu_state_obj *obj, + struct a6xx_crashdumper *dumper) +{ + u64 *in = dumper->ptr; + u64 out = dumper->iova + A6XX_CD_DATA_OFFSET; + size_t datasize = block->size * A6XX_NUM_SHADER_BANKS * sizeof(u32); + int i; + + if (WARN_ON(datasize > A6XX_CD_DATA_SIZE)) + return; + + for (i = 0; i < A6XX_NUM_SHADER_BANKS; i++) { + in += CRASHDUMP_WRITE(in, REG_A6XX_HLSQ_DBG_READ_SEL, + (block->type << 8) | i); + + in += CRASHDUMP_READ(in, REG_A6XX_HLSQ_DBG_AHB_READ_APERTURE, + block->size, out); + + out += block->size * sizeof(u32); + } + + CRASHDUMP_FINI(in); + + if (a6xx_crashdumper_run(gpu, dumper)) + return; + + obj->handle = block; + obj->data = state_kmemdup(a6xx_state, dumper->ptr + A6XX_CD_DATA_OFFSET, + datasize); +} + +static void a7xx_get_shader_block(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct gen7_shader_block *block, + struct a6xx_gpu_state_obj *obj, + struct a6xx_crashdumper *dumper) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + u64 *in = dumper->ptr; + u64 out = dumper->iova + A6XX_CD_DATA_OFFSET; + size_t datasize = block->size * block->num_sps * block->num_usptps * sizeof(u32); + int i, j; + + if (WARN_ON(datasize > A6XX_CD_DATA_SIZE)) + return; + + if (adreno_gpu->info->family == ADRENO_7XX_GEN1) { + gpu_rmw(gpu, REG_A7XX_SP_DBG_CNTL, GENMASK(1, 0), 3); + } + + for (i = 0; i < block->num_sps; i++) { + for (j = 0; j < block->num_usptps; j++) { + in += CRASHDUMP_WRITE(in, REG_A7XX_SP_READ_SEL, + A7XX_SP_READ_SEL_LOCATION(block->location) | + A7XX_SP_READ_SEL_PIPE(block->pipeid) | + A7XX_SP_READ_SEL_STATETYPE(block->statetype) | + A7XX_SP_READ_SEL_USPTP(j) | + A7XX_SP_READ_SEL_SPTP(i)); + + in += CRASHDUMP_READ(in, REG_A7XX_SP_AHB_READ_APERTURE, + block->size, out); + + out += block->size * sizeof(u32); + } + } + + CRASHDUMP_FINI(in); + + if (a6xx_crashdumper_run(gpu, dumper)) + goto out; + + obj->handle = block; + obj->data = state_kmemdup(a6xx_state, dumper->ptr + A6XX_CD_DATA_OFFSET, + datasize); + +out: + if (adreno_gpu->info->family == ADRENO_7XX_GEN1) { + gpu_rmw(gpu, REG_A7XX_SP_DBG_CNTL, GENMASK(1, 0), 0); + } +} + +static void a6xx_get_shaders(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + struct a6xx_crashdumper *dumper) +{ + int i; + + a6xx_state->shaders = state_kcalloc(a6xx_state, + ARRAY_SIZE(a6xx_shader_blocks), sizeof(*a6xx_state->shaders)); + + if (!a6xx_state->shaders) + return; + + a6xx_state->nr_shaders = ARRAY_SIZE(a6xx_shader_blocks); + + for (i = 0; i < ARRAY_SIZE(a6xx_shader_blocks); i++) + a6xx_get_shader_block(gpu, a6xx_state, &a6xx_shader_blocks[i], + &a6xx_state->shaders[i], dumper); +} + +static void a7xx_get_shaders(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + struct a6xx_crashdumper *dumper) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + const struct gen7_shader_block *shader_blocks; + unsigned num_shader_blocks; + int i; + + if (adreno_gpu->info->family == ADRENO_7XX_GEN1) { + if (adreno_is_a722(adreno_gpu)) { + shader_blocks = gen7_17_0_shader_blocks; + num_shader_blocks = ARRAY_SIZE(gen7_17_0_shader_blocks); + } else { + shader_blocks = gen7_0_0_shader_blocks; + num_shader_blocks = ARRAY_SIZE(gen7_0_0_shader_blocks); + } + } else if (adreno_gpu->info->family == ADRENO_7XX_GEN2) { + shader_blocks = gen7_2_0_shader_blocks; + num_shader_blocks = ARRAY_SIZE(gen7_2_0_shader_blocks); + } else { + BUG_ON(adreno_gpu->info->family != ADRENO_7XX_GEN3); + shader_blocks = gen7_9_0_shader_blocks; + num_shader_blocks = ARRAY_SIZE(gen7_9_0_shader_blocks); + } + + a6xx_state->shaders = state_kcalloc(a6xx_state, + num_shader_blocks, sizeof(*a6xx_state->shaders)); + + if (!a6xx_state->shaders) + return; + + a6xx_state->nr_shaders = num_shader_blocks; + + for (i = 0; i < num_shader_blocks; i++) + a7xx_get_shader_block(gpu, a6xx_state, &shader_blocks[i], + &a6xx_state->shaders[i], dumper); +} + +/* Read registers from behind the HLSQ aperture with the crashdumper */ +static void a6xx_get_crashdumper_hlsq_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_registers *regs, + struct a6xx_gpu_state_obj *obj, + struct a6xx_crashdumper *dumper) + +{ + u64 *in = dumper->ptr; + u64 out = dumper->iova + A6XX_CD_DATA_OFFSET; + int i, regcount = 0; + + in += CRASHDUMP_WRITE(in, REG_A6XX_HLSQ_DBG_READ_SEL, (regs->val1 & 0xff) << 8); + + for (i = 0; i < regs->count; i += 2) { + u32 count = RANGE(regs->registers, i); + u32 offset = REG_A6XX_HLSQ_DBG_AHB_READ_APERTURE + + regs->registers[i] - (regs->val0 >> 2); + + in += CRASHDUMP_READ(in, offset, count, out); + + out += count * sizeof(u32); + regcount += count; + } + + CRASHDUMP_FINI(in); + + if (WARN_ON((regcount * sizeof(u32)) > A6XX_CD_DATA_SIZE)) + return; + + if (a6xx_crashdumper_run(gpu, dumper)) + return; + + obj->handle = regs; + obj->data = state_kmemdup(a6xx_state, dumper->ptr + A6XX_CD_DATA_OFFSET, + regcount * sizeof(u32)); +} + +/* Read a block of registers using the crashdumper */ +static void a6xx_get_crashdumper_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_registers *regs, + struct a6xx_gpu_state_obj *obj, + struct a6xx_crashdumper *dumper) + +{ + u64 *in = dumper->ptr; + u64 out = dumper->iova + A6XX_CD_DATA_OFFSET; + int i, regcount = 0; + + /* Skip unsupported registers on older generations */ + if (!adreno_is_a660_family(to_adreno_gpu(gpu)) && + (regs->registers == a660_registers)) + return; + + /* Some blocks might need to program a selector register first */ + if (regs->val0) + in += CRASHDUMP_WRITE(in, regs->val0, regs->val1); + + for (i = 0; i < regs->count; i += 2) { + u32 count = RANGE(regs->registers, i); + + in += CRASHDUMP_READ(in, regs->registers[i], count, out); + + out += count * sizeof(u32); + regcount += count; + } + + CRASHDUMP_FINI(in); + + if (WARN_ON((regcount * sizeof(u32)) > A6XX_CD_DATA_SIZE)) + return; + + if (a6xx_crashdumper_run(gpu, dumper)) + return; + + obj->handle = regs; + obj->data = state_kmemdup(a6xx_state, dumper->ptr + A6XX_CD_DATA_OFFSET, + regcount * sizeof(u32)); +} + +static void a7xx_get_crashdumper_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct gen7_reg_list *regs, + struct a6xx_gpu_state_obj *obj, + struct a6xx_crashdumper *dumper) + +{ + u64 *in = dumper->ptr; + u64 out = dumper->iova + A6XX_CD_DATA_OFFSET; + int i, regcount = 0; + + /* Some blocks might need to program a selector register first */ + if (regs->sel) + in += CRASHDUMP_WRITE(in, regs->sel->cd_reg, regs->sel->val); + + for (i = 0; regs->regs[i] != UINT_MAX; i += 2) { + u32 count = RANGE(regs->regs, i); + + in += CRASHDUMP_READ(in, regs->regs[i], count, out); + + out += count * sizeof(u32); + regcount += count; + } + + CRASHDUMP_FINI(in); + + if (WARN_ON((regcount * sizeof(u32)) > A6XX_CD_DATA_SIZE)) + return; + + if (a6xx_crashdumper_run(gpu, dumper)) + return; + + obj->handle = regs->regs; + obj->data = state_kmemdup(a6xx_state, dumper->ptr + A6XX_CD_DATA_OFFSET, + regcount * sizeof(u32)); +} + + +/* Read a block of registers via AHB */ +static void a6xx_get_ahb_gpu_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_registers *regs, + struct a6xx_gpu_state_obj *obj) +{ + int i, regcount = 0, index = 0; + + /* Skip unsupported registers on older generations */ + if (!adreno_is_a660_family(to_adreno_gpu(gpu)) && + (regs->registers == a660_registers)) + return; + + for (i = 0; i < regs->count; i += 2) + regcount += RANGE(regs->registers, i); + + obj->handle = (const void *) regs; + obj->data = state_kcalloc(a6xx_state, regcount, sizeof(u32)); + if (!obj->data) + return; + + for (i = 0; i < regs->count; i += 2) { + u32 count = RANGE(regs->registers, i); + int j; + + for (j = 0; j < count; j++) + obj->data[index++] = gpu_read(gpu, + regs->registers[i] + j); + } +} + +static void a7xx_get_ahb_gpu_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const u32 *regs, + struct a6xx_gpu_state_obj *obj) +{ + int i, regcount = 0, index = 0; + + for (i = 0; regs[i] != UINT_MAX; i += 2) + regcount += RANGE(regs, i); + + obj->handle = (const void *) regs; + obj->data = state_kcalloc(a6xx_state, regcount, sizeof(u32)); + if (!obj->data) + return; + + for (i = 0; regs[i] != UINT_MAX; i += 2) { + u32 count = RANGE(regs, i); + int j; + + for (j = 0; j < count; j++) + obj->data[index++] = gpu_read(gpu, regs[i] + j); + } +} + +static void a7xx_get_ahb_gpu_reglist(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct gen7_reg_list *regs, + struct a6xx_gpu_state_obj *obj) +{ + if (regs->sel) + gpu_write(gpu, regs->sel->host_reg, regs->sel->val); + + a7xx_get_ahb_gpu_registers(gpu, a6xx_state, regs->regs, obj); +} + +/* Read a block of GMU registers */ +static void _a6xx_get_gmu_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_registers *regs, + struct a6xx_gpu_state_obj *obj, + bool rscc) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + int i, regcount = 0, index = 0; + + for (i = 0; i < regs->count; i += 2) + regcount += RANGE(regs->registers, i); + + obj->handle = (const void *) regs; + obj->data = state_kcalloc(a6xx_state, regcount, sizeof(u32)); + if (!obj->data) + return; + + for (i = 0; i < regs->count; i += 2) { + u32 count = RANGE(regs->registers, i); + int j; + + for (j = 0; j < count; j++) { + u32 offset = regs->registers[i] + j; + u32 val; + + if (rscc) + val = gmu_read_rscc(gmu, offset); + else + val = gmu_read(gmu, offset); + + obj->data[index++] = val; + } + } +} + +static void a6xx_get_gmu_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + a6xx_state->gmu_registers = state_kcalloc(a6xx_state, + 4, sizeof(*a6xx_state->gmu_registers)); + + if (!a6xx_state->gmu_registers) + return; + + a6xx_state->nr_gmu_registers = 4; + + /* Get the CX GMU registers from AHB */ + _a6xx_get_gmu_registers(gpu, a6xx_state, &a6xx_gmu_reglist[0], + &a6xx_state->gmu_registers[0], false); + _a6xx_get_gmu_registers(gpu, a6xx_state, &a6xx_gmu_reglist[1], + &a6xx_state->gmu_registers[1], true); + + if (adreno_is_a621(adreno_gpu)) + _a6xx_get_gmu_registers(gpu, a6xx_state, &a621_gpucc_reg, + &a6xx_state->gmu_registers[2], false); + else if (adreno_is_a623(adreno_gpu) || adreno_is_a663(adreno_gpu)) + _a6xx_get_gmu_registers(gpu, a6xx_state, &a623_gpucc_reg, + &a6xx_state->gmu_registers[2], false); + else + _a6xx_get_gmu_registers(gpu, a6xx_state, &a6xx_gpucc_reg, + &a6xx_state->gmu_registers[2], false); + + if (!adreno_gpu->funcs->gx_is_on(adreno_gpu)) + return; + + /* Set the fence to ALLOW mode so we can access the registers */ + gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_AO_AHB_FENCE_CTRL, 0); + + _a6xx_get_gmu_registers(gpu, a6xx_state, &a6xx_gmu_reglist[2], + &a6xx_state->gmu_registers[3], false); +} + +static struct msm_gpu_state_bo *a6xx_snapshot_gmu_bo( + struct a6xx_gpu_state *a6xx_state, struct a6xx_gmu_bo *bo) +{ + struct msm_gpu_state_bo *snapshot; + + if (!bo->size) + return NULL; + + snapshot = state_kcalloc(a6xx_state, 1, sizeof(*snapshot)); + if (!snapshot) + return NULL; + + snapshot->iova = bo->iova; + snapshot->size = bo->size; + snapshot->data = kvzalloc(snapshot->size, GFP_KERNEL); + if (!snapshot->data) + return NULL; + + memcpy(snapshot->data, bo->virt, bo->size); + + return snapshot; +} + +static void a6xx_snapshot_gmu_hfi_history(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + unsigned i, j; + + BUILD_BUG_ON(ARRAY_SIZE(gmu->queues) != ARRAY_SIZE(a6xx_state->hfi_queue_history)); + + for (i = 0; i < ARRAY_SIZE(gmu->queues); i++) { + struct a6xx_hfi_queue *queue = &gmu->queues[i]; + for (j = 0; j < HFI_HISTORY_SZ; j++) { + unsigned idx = (j + queue->history_idx) % HFI_HISTORY_SZ; + a6xx_state->hfi_queue_history[i][j] = queue->history[idx]; + } + } +} + +#define A6XX_REGLIST_SIZE 1 +#define A6XX_GBIF_REGLIST_SIZE 1 +static void a6xx_get_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + struct a6xx_crashdumper *dumper) +{ + int i, count = A6XX_REGLIST_SIZE + + ARRAY_SIZE(a6xx_reglist) + + ARRAY_SIZE(a6xx_hlsq_reglist) + A6XX_GBIF_REGLIST_SIZE; + int index = 0; + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + + a6xx_state->registers = state_kcalloc(a6xx_state, + count, sizeof(*a6xx_state->registers)); + + if (!a6xx_state->registers) + return; + + a6xx_state->nr_registers = count; + + a6xx_get_ahb_gpu_registers(gpu, + a6xx_state, &a6xx_ahb_reglist, + &a6xx_state->registers[index++]); + + if (a6xx_has_gbif(adreno_gpu)) + a6xx_get_ahb_gpu_registers(gpu, + a6xx_state, &a6xx_gbif_reglist, + &a6xx_state->registers[index++]); + else + a6xx_get_ahb_gpu_registers(gpu, + a6xx_state, &a6xx_vbif_reglist, + &a6xx_state->registers[index++]); + if (!dumper) { + /* + * We can't use the crashdumper when the SMMU is stalled, + * because the GPU has no memory access until we resume + * translation (but we don't want to do that until after + * we have captured as much useful GPU state as possible). + * So instead collect registers via the CPU: + */ + for (i = 0; i < ARRAY_SIZE(a6xx_reglist); i++) + a6xx_get_ahb_gpu_registers(gpu, + a6xx_state, &a6xx_reglist[i], + &a6xx_state->registers[index++]); + return; + } + + for (i = 0; i < ARRAY_SIZE(a6xx_reglist); i++) + a6xx_get_crashdumper_registers(gpu, + a6xx_state, &a6xx_reglist[i], + &a6xx_state->registers[index++], + dumper); + + for (i = 0; i < ARRAY_SIZE(a6xx_hlsq_reglist); i++) + a6xx_get_crashdumper_hlsq_registers(gpu, + a6xx_state, &a6xx_hlsq_reglist[i], + &a6xx_state->registers[index++], + dumper); +} + +#define A7XX_PRE_CRASHDUMPER_SIZE 1 +#define A7XX_POST_CRASHDUMPER_SIZE 1 +static void a7xx_get_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + struct a6xx_crashdumper *dumper) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + int i, count; + int index = 0; + const u32 *pre_crashdumper_regs; + const struct gen7_reg_list *reglist; + + if (adreno_gpu->info->family == ADRENO_7XX_GEN1) { + if (adreno_is_a722(adreno_gpu)) { + reglist = gen7_17_0_reg_list; + pre_crashdumper_regs = gen7_9_0_pre_crashdumper_gpu_registers; + } else { + reglist = gen7_0_0_reg_list; + pre_crashdumper_regs = gen7_0_0_pre_crashdumper_gpu_registers; + } + } else if (adreno_gpu->info->family == ADRENO_7XX_GEN2) { + reglist = gen7_2_0_reg_list; + pre_crashdumper_regs = gen7_0_0_pre_crashdumper_gpu_registers; + } else { + BUG_ON(adreno_gpu->info->family != ADRENO_7XX_GEN3); + reglist = gen7_9_0_reg_list; + pre_crashdumper_regs = gen7_9_0_pre_crashdumper_gpu_registers; + } + + count = A7XX_PRE_CRASHDUMPER_SIZE + A7XX_POST_CRASHDUMPER_SIZE; + + /* The downstream reglist contains registers in other memory regions + * (cx_misc/cx_mem and cx_dbgc) and we need to plumb through their + * offsets and map them to read them on the CPU. For now only read the + * first region which is the main one. + */ + if (dumper) { + for (i = 0; reglist[i].regs; i++) + count++; + } else { + count++; + } + + a6xx_state->registers = state_kcalloc(a6xx_state, + count, sizeof(*a6xx_state->registers)); + + if (!a6xx_state->registers) + return; + + a6xx_state->nr_registers = count; + + a7xx_get_ahb_gpu_registers(gpu, a6xx_state, pre_crashdumper_regs, + &a6xx_state->registers[index++]); + + if (!dumper) { + a7xx_get_ahb_gpu_reglist(gpu, + a6xx_state, ®list[0], + &a6xx_state->registers[index++]); + return; + } + + for (i = 0; reglist[i].regs; i++) + a7xx_get_crashdumper_registers(gpu, + a6xx_state, ®list[i], + &a6xx_state->registers[index++], + dumper); +} + +static void a7xx_get_post_crashdumper_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + const u32 *regs; + + BUG_ON(adreno_gpu->info->family > ADRENO_7XX_GEN3); + regs = adreno_is_a722(adreno_gpu) ? + gen7_17_0_post_crashdumper_registers : + gen7_0_0_post_crashdumper_registers; + + a7xx_get_ahb_gpu_registers(gpu, + a6xx_state, regs, + &a6xx_state->registers[a6xx_state->nr_registers - 1]); +} + +static u32 a6xx_get_cp_roq_size(struct msm_gpu *gpu) +{ + /* The value at [16:31] is in 4dword units. Convert it to dwords */ + return gpu_read(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_2) >> 14; +} + +static u32 a7xx_get_cp_roq_size(struct msm_gpu *gpu) +{ + /* + * The value at CP_ROQ_THRESHOLDS_2[20:31] is in 4dword units. + * That register however is not directly accessible from APSS on A7xx. + * Program the SQE_UCODE_DBG_ADDR with offset=0x70d3 and read the value. + */ + gpu_write(gpu, REG_A6XX_CP_SQE_UCODE_DBG_ADDR, 0x70d3); + + return 4 * (gpu_read(gpu, REG_A6XX_CP_SQE_UCODE_DBG_DATA) >> 20); +} + +/* Read a block of data from an indexed register pair */ +static void a6xx_get_indexed_regs(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_indexed_registers *indexed, + struct a6xx_gpu_state_obj *obj) +{ + u32 count = indexed->count; + int i; + + obj->handle = (const void *) indexed; + if (indexed->count_fn) + count = indexed->count_fn(gpu); + + obj->data = state_kcalloc(a6xx_state, count, sizeof(u32)); + obj->count = count; + if (!obj->data) + return; + + /* All the indexed banks start at address 0 */ + gpu_write(gpu, indexed->addr, 0); + + /* Read the data - each read increments the internal address by 1 */ + for (i = 0; i < count; i++) + obj->data[i] = gpu_read(gpu, indexed->data); +} + +static void a6xx_get_indexed_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state) +{ + u32 mempool_size; + int count = ARRAY_SIZE(a6xx_indexed_reglist) + 1; + int i; + + a6xx_state->indexed_regs = state_kcalloc(a6xx_state, count, + sizeof(*a6xx_state->indexed_regs)); + if (!a6xx_state->indexed_regs) + return; + + for (i = 0; i < ARRAY_SIZE(a6xx_indexed_reglist); i++) + a6xx_get_indexed_regs(gpu, a6xx_state, &a6xx_indexed_reglist[i], + &a6xx_state->indexed_regs[i]); + + if (adreno_is_a650_family(to_adreno_gpu(gpu))) { + u32 val; + + val = gpu_read(gpu, REG_A6XX_CP_CHICKEN_DBG); + gpu_write(gpu, REG_A6XX_CP_CHICKEN_DBG, val | 4); + + /* Get the contents of the CP mempool */ + a6xx_get_indexed_regs(gpu, a6xx_state, &a6xx_cp_mempool_indexed, + &a6xx_state->indexed_regs[i]); + + gpu_write(gpu, REG_A6XX_CP_CHICKEN_DBG, val); + a6xx_state->nr_indexed_regs = count; + return; + } + + /* Set the CP mempool size to 0 to stabilize it while dumping */ + mempool_size = gpu_read(gpu, REG_A6XX_CP_MEM_POOL_SIZE); + gpu_write(gpu, REG_A6XX_CP_MEM_POOL_SIZE, 0); + + /* Get the contents of the CP mempool */ + a6xx_get_indexed_regs(gpu, a6xx_state, &a6xx_cp_mempool_indexed, + &a6xx_state->indexed_regs[i]); + + /* + * Offset 0x2000 in the mempool is the size - copy the saved size over + * so the data is consistent + */ + a6xx_state->indexed_regs[i].data[0x2000] = mempool_size; + + /* Restore the size in the hardware */ + gpu_write(gpu, REG_A6XX_CP_MEM_POOL_SIZE, mempool_size); + + a6xx_state->nr_indexed_regs = count; +} + +static void a7xx_get_indexed_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + const struct a6xx_indexed_registers *indexed_regs; + const struct a6xx_indexed_registers *mempool_regs; + int i, indexed_count, mempool_count; + bool concurrent_binning; + + if (adreno_is_a722(adreno_gpu)) { + /* + * Eliza has no BV or LPAC SQE — skip the BV/LPAC indexed + * registers and the BV mempool + */ + indexed_regs = gen7_17_0_cp_indexed_reglist; + indexed_count = ARRAY_SIZE(gen7_17_0_cp_indexed_reglist); + mempool_regs = a7xx_cp_mempool_indexed; + mempool_count = ARRAY_SIZE(a7xx_cp_mempool_indexed); + concurrent_binning = false; + } else if (adreno_gpu->info->family <= ADRENO_7XX_GEN2) { + indexed_regs = a7xx_indexed_reglist; + indexed_count = ARRAY_SIZE(a7xx_indexed_reglist); + mempool_regs = a7xx_cp_bv_mempool_indexed; + mempool_count = ARRAY_SIZE(a7xx_cp_bv_mempool_indexed); + concurrent_binning = true; + } else { + BUG_ON(adreno_gpu->info->family != ADRENO_7XX_GEN3); + indexed_regs = gen7_9_0_cp_indexed_reg_list; + indexed_count = ARRAY_SIZE(gen7_9_0_cp_indexed_reg_list); + mempool_regs = a7xx_cp_bv_mempool_indexed; + mempool_count = ARRAY_SIZE(a7xx_cp_bv_mempool_indexed); + concurrent_binning = true; + } + + a6xx_state->indexed_regs = state_kcalloc(a6xx_state, + indexed_count + mempool_count, + sizeof(*a6xx_state->indexed_regs)); + if (!a6xx_state->indexed_regs) + return; + + a6xx_state->nr_indexed_regs = indexed_count + mempool_count; + + /* First read the common regs */ + for (i = 0; i < indexed_count; i++) + a6xx_get_indexed_regs(gpu, a6xx_state, &indexed_regs[i], + &a6xx_state->indexed_regs[i]); + + gpu_rmw(gpu, REG_A6XX_CP_CHICKEN_DBG, 0, BIT(2)); + if (concurrent_binning) + gpu_rmw(gpu, REG_A7XX_CP_BV_CHICKEN_DBG, 0, BIT(2)); + + /* Get the contents of the CP_BV mempool */ + for (i = 0; i < mempool_count; i++) + a6xx_get_indexed_regs(gpu, a6xx_state, &mempool_regs[i], + &a6xx_state->indexed_regs[indexed_count + i]); + + gpu_rmw(gpu, REG_A6XX_CP_CHICKEN_DBG, BIT(2), 0); + if (concurrent_binning) + gpu_rmw(gpu, REG_A7XX_CP_BV_CHICKEN_DBG, BIT(2), 0); + return; +} + +struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu) +{ + struct a6xx_crashdumper _dumper = { 0 }, *dumper = NULL; + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gpu_state *a6xx_state = kzalloc_obj(*a6xx_state); + bool stalled; + + if (!a6xx_state) + return ERR_PTR(-ENOMEM); + + INIT_LIST_HEAD(&a6xx_state->objs); + + /* Get the generic state from the adreno core */ + adreno_gpu_state_get(gpu, &a6xx_state->base); + + if (!adreno_has_gmu_wrapper(adreno_gpu) && + !adreno_has_rgmu(adreno_gpu)) { + a6xx_get_gmu_registers(gpu, a6xx_state); + + a6xx_state->gmu_log = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.log); + a6xx_state->gmu_hfi = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.hfi); + a6xx_state->gmu_debug = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.debug); + + a6xx_snapshot_gmu_hfi_history(gpu, a6xx_state); + } + + /* If GX isn't on the rest of the data isn't going to be accessible */ + if (!adreno_gpu->funcs->gx_is_on(adreno_gpu)) + return &a6xx_state->base; + + /* Halt SQE first */ + gpu_write(gpu, REG_A6XX_CP_SQE_CNTL, 3); + + /* Get the banks of indexed registers */ + if (adreno_is_a7xx(adreno_gpu)) + a7xx_get_indexed_registers(gpu, a6xx_state); + else + a6xx_get_indexed_registers(gpu, a6xx_state); + + stalled = !!(gpu_read(gpu, REG_A6XX_RBBM_STATUS3) & + A6XX_RBBM_STATUS3_SMMU_STALLED_ON_FAULT); + /* + * Try to initialize the crashdumper, if we are not dumping state + * with the SMMU stalled. The crashdumper needs memory access to + * write out GPU state, so we need to skip this when the SMMU is + * stalled in response to an iova fault + */ + if (!stalled && !gpu->needs_hw_init && + !a6xx_crashdumper_init(gpu, &_dumper)) { + dumper = &_dumper; + } + + if (adreno_is_a7xx(adreno_gpu)) { + a7xx_get_registers(gpu, a6xx_state, dumper); + + if (dumper) { + a7xx_get_shaders(gpu, a6xx_state, dumper); + a7xx_get_clusters(gpu, a6xx_state, dumper); + a7xx_get_dbgahb_clusters(gpu, a6xx_state, dumper); + + msm_gem_kernel_put(dumper->bo, gpu->vm); + } + + a7xx_get_post_crashdumper_registers(gpu, a6xx_state); + } else { + a6xx_get_registers(gpu, a6xx_state, dumper); + + if (dumper) { + a6xx_get_shaders(gpu, a6xx_state, dumper); + a6xx_get_clusters(gpu, a6xx_state, dumper); + a6xx_get_dbgahb_clusters(gpu, a6xx_state, dumper); + + msm_gem_kernel_put(dumper->bo, gpu->vm); + } + } + + if (snapshot_debugbus) + a6xx_get_debugbus(gpu, a6xx_state); + + a6xx_state->gpu_initialized = !gpu->needs_hw_init; + + return &a6xx_state->base; +} + +static void a6xx_gpu_state_destroy(struct kref *kref) +{ + struct a6xx_state_memobj *obj, *tmp; + struct msm_gpu_state *state = container_of(kref, + struct msm_gpu_state, ref); + struct a6xx_gpu_state *a6xx_state = container_of(state, + struct a6xx_gpu_state, base); + + if (a6xx_state->gmu_log) + kvfree(a6xx_state->gmu_log->data); + + if (a6xx_state->gmu_hfi) + kvfree(a6xx_state->gmu_hfi->data); + + if (a6xx_state->gmu_debug) + kvfree(a6xx_state->gmu_debug->data); + + list_for_each_entry_safe(obj, tmp, &a6xx_state->objs, node) { + list_del(&obj->node); + kvfree(obj); + } + + adreno_gpu_state_destroy(state); + kfree(a6xx_state); +} + +int a6xx_gpu_state_put(struct msm_gpu_state *state) +{ + if (IS_ERR_OR_NULL(state)) + return 1; + + return kref_put(&state->ref, a6xx_gpu_state_destroy); +} + +static void a6xx_show_registers(const u32 *registers, u32 *data, size_t count, + struct drm_printer *p) +{ + int i, index = 0; + + if (!data) + return; + + for (i = 0; i < count; i += 2) { + u32 count = RANGE(registers, i); + u32 offset = registers[i]; + int j; + + for (j = 0; j < count; index++, offset++, j++) { + if (data[index] == 0xdeafbead) + continue; + + drm_printf(p, " - { offset: 0x%06x, value: 0x%08x }\n", + offset << 2, data[index]); + } + } +} + +static void a7xx_show_registers_indented(const u32 *registers, u32 *data, + struct drm_printer *p, unsigned indent) +{ + int i, index = 0; + + for (i = 0; registers[i] != UINT_MAX; i += 2) { + u32 count = RANGE(registers, i); + u32 offset = registers[i]; + int j; + + for (j = 0; j < count; index++, offset++, j++) { + int k; + + if (data[index] == 0xdeafbead) + continue; + + for (k = 0; k < indent; k++) + drm_printf(p, " "); + drm_printf(p, "- { offset: 0x%06x, value: 0x%08x }\n", + offset << 2, data[index]); + } + } +} + +static void a7xx_show_registers(const u32 *registers, u32 *data, struct drm_printer *p) +{ + a7xx_show_registers_indented(registers, data, p, 1); +} + +static void print_ascii85(struct drm_printer *p, size_t len, u32 *data) +{ + char out[ASCII85_BUFSZ]; + long i, l, datalen = 0; + + for (i = 0; i < len >> 2; i++) { + if (data[i]) + datalen = (i + 1) << 2; + } + + if (datalen == 0) + return; + + drm_puts(p, " data: !!ascii85 |\n"); + drm_puts(p, " "); + + + l = ascii85_encode_len(datalen); + + for (i = 0; i < l; i++) + drm_puts(p, ascii85_encode(data[i], out)); + + drm_puts(p, "\n"); +} + +static void print_name(struct drm_printer *p, const char *fmt, const char *name) +{ + drm_puts(p, fmt); + drm_puts(p, name); + drm_puts(p, "\n"); +} + +static void a6xx_show_shader(struct a6xx_gpu_state_obj *obj, + struct drm_printer *p) +{ + const struct a6xx_shader_block *block = obj->handle; + int i; + + if (!obj->handle) + return; + + print_name(p, " - type: ", block->name); + + for (i = 0; i < A6XX_NUM_SHADER_BANKS; i++) { + drm_printf(p, " - bank: %d\n", i); + drm_printf(p, " size: %d\n", block->size); + + if (!obj->data) + continue; + + print_ascii85(p, block->size << 2, + obj->data + (block->size * i)); + } +} + +static void a7xx_show_shader(struct a6xx_gpu_state_obj *obj, + struct drm_printer *p) +{ + const struct gen7_shader_block *block = obj->handle; + int i, j; + u32 *data = obj->data; + + if (!obj->handle) + return; + + print_name(p, " - type: ", a7xx_statetype_names[block->statetype]); + print_name(p, " - pipe: ", a7xx_pipe_names[block->pipeid]); + drm_printf(p, " - location: %d\n", block->location); + + for (i = 0; i < block->num_sps; i++) { + drm_printf(p, " - sp: %d\n", i); + + for (j = 0; j < block->num_usptps; j++) { + drm_printf(p, " - usptp: %d\n", j); + drm_printf(p, " size: %d\n", block->size); + + if (!obj->data) + continue; + + print_ascii85(p, block->size << 2, data); + + data += block->size; + } + } +} + +static void a6xx_show_cluster_data(const u32 *registers, int size, u32 *data, + struct drm_printer *p) +{ + int ctx, index = 0; + + for (ctx = 0; ctx < A6XX_NUM_CONTEXTS; ctx++) { + int j; + + drm_printf(p, " - context: %d\n", ctx); + + for (j = 0; j < size; j += 2) { + u32 count = RANGE(registers, j); + u32 offset = registers[j]; + int k; + + for (k = 0; k < count; index++, offset++, k++) { + if (data[index] == 0xdeafbead) + continue; + + drm_printf(p, " - { offset: 0x%06x, value: 0x%08x }\n", + offset << 2, data[index]); + } + } + } +} + +static void a6xx_show_dbgahb_cluster(struct a6xx_gpu_state_obj *obj, + struct drm_printer *p) +{ + const struct a6xx_dbgahb_cluster *dbgahb = obj->handle; + + if (dbgahb) { + print_name(p, " - cluster-name: ", dbgahb->name); + a6xx_show_cluster_data(dbgahb->registers, dbgahb->count, + obj->data, p); + } +} + +static void a6xx_show_cluster(struct a6xx_gpu_state_obj *obj, + struct drm_printer *p) +{ + const struct a6xx_cluster *cluster = obj->handle; + + if (cluster) { + print_name(p, " - cluster-name: ", cluster->name); + a6xx_show_cluster_data(cluster->registers, cluster->count, + obj->data, p); + } +} + +static void a7xx_show_dbgahb_cluster(struct a6xx_gpu_state_obj *obj, + struct drm_printer *p) +{ + const struct gen7_sptp_cluster_registers *dbgahb = obj->handle; + + if (dbgahb) { + print_name(p, " - pipe: ", a7xx_pipe_names[dbgahb->pipe_id]); + print_name(p, " - cluster-name: ", a7xx_cluster_names[dbgahb->cluster_id]); + drm_printf(p, " - context: %d\n", dbgahb->context_id); + drm_printf(p, " - location: %d\n", dbgahb->location_id); + a7xx_show_registers_indented(dbgahb->regs, obj->data, p, 4); + } +} + +static void a7xx_show_cluster(struct a6xx_gpu_state_obj *obj, + struct drm_printer *p) +{ + const struct gen7_cluster_registers *cluster = obj->handle; + + if (cluster) { + int context = (cluster->context_id == STATE_FORCE_CTXT_1) ? 1 : 0; + + print_name(p, " - pipe: ", a7xx_pipe_names[cluster->pipe_id]); + print_name(p, " - cluster-name: ", a7xx_cluster_names[cluster->cluster_id]); + drm_printf(p, " - context: %d\n", context); + a7xx_show_registers_indented(cluster->regs, obj->data, p, 4); + } +} + +static void a6xx_show_indexed_regs(struct a6xx_gpu_state_obj *obj, + struct drm_printer *p) +{ + const struct a6xx_indexed_registers *indexed = obj->handle; + + if (!indexed) + return; + + print_name(p, " - regs-name: ", indexed->name); + drm_printf(p, " dwords: %d\n", obj->count); + + print_ascii85(p, obj->count << 2, obj->data); +} + +static void a6xx_show_debugbus_block(const struct a6xx_debugbus_block *block, + u32 *data, struct drm_printer *p) +{ + if (block) { + print_name(p, " - debugbus-block: ", block->name); + + /* + * count for regular debugbus data is in quadwords, + * but print the size in dwords for consistency + */ + drm_printf(p, " count: %d\n", block->count << 1); + + print_ascii85(p, block->count << 3, data); + } +} + +static void a6xx_show_debugbus(struct a6xx_gpu_state *a6xx_state, + struct drm_printer *p) +{ + int i; + + for (i = 0; i < a6xx_state->nr_debugbus; i++) { + struct a6xx_gpu_state_obj *obj = &a6xx_state->debugbus[i]; + + a6xx_show_debugbus_block(obj->handle, obj->data, p); + } + + if (a6xx_state->vbif_debugbus) { + struct a6xx_gpu_state_obj *obj = a6xx_state->vbif_debugbus; + + drm_puts(p, " - debugbus-block: A6XX_DBGBUS_VBIF\n"); + drm_printf(p, " count: %d\n", VBIF_DEBUGBUS_BLOCK_SIZE); + + /* vbif debugbus data is in dwords. Confusing, huh? */ + print_ascii85(p, VBIF_DEBUGBUS_BLOCK_SIZE << 2, obj->data); + } + + for (i = 0; i < a6xx_state->nr_cx_debugbus; i++) { + struct a6xx_gpu_state_obj *obj = &a6xx_state->cx_debugbus[i]; + + a6xx_show_debugbus_block(obj->handle, obj->data, p); + } +} + +void a6xx_show(struct msm_gpu *gpu, struct msm_gpu_state *state, + struct drm_printer *p) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu_state *a6xx_state = container_of(state, + struct a6xx_gpu_state, base); + int i; + + if (IS_ERR_OR_NULL(state)) + return; + + drm_printf(p, "gpu-initialized: %d\n", a6xx_state->gpu_initialized); + + adreno_show(gpu, state, p); + + drm_puts(p, "gmu-log:\n"); + if (a6xx_state->gmu_log) { + struct msm_gpu_state_bo *gmu_log = a6xx_state->gmu_log; + + drm_printf(p, " iova: 0x%016llx\n", gmu_log->iova); + drm_printf(p, " size: %zu\n", gmu_log->size); + adreno_show_object(p, &gmu_log->data, gmu_log->size, + &gmu_log->encoded); + } + + drm_puts(p, "gmu-hfi:\n"); + if (a6xx_state->gmu_hfi) { + struct msm_gpu_state_bo *gmu_hfi = a6xx_state->gmu_hfi; + unsigned i, j; + + drm_printf(p, " iova: 0x%016llx\n", gmu_hfi->iova); + drm_printf(p, " size: %zu\n", gmu_hfi->size); + for (i = 0; i < ARRAY_SIZE(a6xx_state->hfi_queue_history); i++) { + drm_printf(p, " queue-history[%u]:", i); + for (j = 0; j < HFI_HISTORY_SZ; j++) { + drm_printf(p, " %d", a6xx_state->hfi_queue_history[i][j]); + } + drm_printf(p, "\n"); + } + adreno_show_object(p, &gmu_hfi->data, gmu_hfi->size, + &gmu_hfi->encoded); + } + + drm_puts(p, "gmu-debug:\n"); + if (a6xx_state->gmu_debug) { + struct msm_gpu_state_bo *gmu_debug = a6xx_state->gmu_debug; + + drm_printf(p, " iova: 0x%016llx\n", gmu_debug->iova); + drm_printf(p, " size: %zu\n", gmu_debug->size); + adreno_show_object(p, &gmu_debug->data, gmu_debug->size, + &gmu_debug->encoded); + } + + drm_puts(p, "registers:\n"); + for (i = 0; i < a6xx_state->nr_registers; i++) { + struct a6xx_gpu_state_obj *obj = &a6xx_state->registers[i]; + + if (!obj->handle) + continue; + + if (adreno_is_a7xx(adreno_gpu)) { + a7xx_show_registers(obj->handle, obj->data, p); + } else { + const struct a6xx_registers *regs = obj->handle; + + a6xx_show_registers(regs->registers, obj->data, regs->count, p); + } + } + + drm_puts(p, "registers-gmu:\n"); + for (i = 0; i < a6xx_state->nr_gmu_registers; i++) { + struct a6xx_gpu_state_obj *obj = &a6xx_state->gmu_registers[i]; + const struct a6xx_registers *regs = obj->handle; + + if (!obj->handle) + continue; + + a6xx_show_registers(regs->registers, obj->data, regs->count, p); + } + + drm_puts(p, "indexed-registers:\n"); + for (i = 0; i < a6xx_state->nr_indexed_regs; i++) + a6xx_show_indexed_regs(&a6xx_state->indexed_regs[i], p); + + drm_puts(p, "shader-blocks:\n"); + for (i = 0; i < a6xx_state->nr_shaders; i++) { + if (adreno_is_a7xx(adreno_gpu)) + a7xx_show_shader(&a6xx_state->shaders[i], p); + else + a6xx_show_shader(&a6xx_state->shaders[i], p); + } + + drm_puts(p, "clusters:\n"); + for (i = 0; i < a6xx_state->nr_clusters; i++) { + if (adreno_is_a7xx(adreno_gpu)) + a7xx_show_cluster(&a6xx_state->clusters[i], p); + else + a6xx_show_cluster(&a6xx_state->clusters[i], p); + } + + for (i = 0; i < a6xx_state->nr_dbgahb_clusters; i++) { + if (adreno_is_a7xx(adreno_gpu)) + a7xx_show_dbgahb_cluster(&a6xx_state->dbgahb_clusters[i], p); + else + a6xx_show_dbgahb_cluster(&a6xx_state->dbgahb_clusters[i], p); + } + + drm_puts(p, "debugbus:\n"); + a6xx_show_debugbus(a6xx_state, p); +} diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h new file mode 100644 index 000000000..6f592394f --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h @@ -0,0 +1,838 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved. + * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#ifndef _A6XX_CRASH_DUMP_H_ +#define _A6XX_CRASH_DUMP_H_ + +#include "a6xx.xml.h" + +#define A6XX_NUM_CONTEXTS 2 +#define A6XX_NUM_SHADER_BANKS 3 + +static const u32 a6xx_gras_cluster[] = { + 0x8000, 0x8006, 0x8010, 0x8092, 0x8094, 0x809d, 0x80a0, 0x80a6, + 0x80af, 0x80f1, 0x8100, 0x8107, 0x8109, 0x8109, 0x8110, 0x8110, + 0x8400, 0x840b, +}; + +static const u32 a6xx_ps_cluster_rac[] = { + 0x8800, 0x8806, 0x8809, 0x8811, 0x8818, 0x881e, 0x8820, 0x8865, + 0x8870, 0x8879, 0x8880, 0x8889, 0x8890, 0x8891, 0x8898, 0x8898, + 0x88c0, 0x88c1, 0x88d0, 0x88e3, 0x8900, 0x890c, 0x890f, 0x891a, + 0x8c00, 0x8c01, 0x8c08, 0x8c10, 0x8c17, 0x8c1f, 0x8c26, 0x8c33, +}; + +static const u32 a6xx_ps_cluster_rbp[] = { + 0x88f0, 0x88f3, 0x890d, 0x890e, 0x8927, 0x8928, 0x8bf0, 0x8bf1, + 0x8c02, 0x8c07, 0x8c11, 0x8c16, 0x8c20, 0x8c25, +}; + +static const u32 a6xx_ps_cluster[] = { + 0x9200, 0x9216, 0x9218, 0x9236, 0x9300, 0x9306, +}; + +static const u32 a6xx_fe_cluster[] = { + 0x9300, 0x9306, 0x9800, 0x9806, 0x9b00, 0x9b07, 0xa000, 0xa009, + 0xa00e, 0xa0ef, 0xa0f8, 0xa0f8, +}; + +static const u32 a660_fe_cluster[] = { + 0x9807, 0x9807, +}; + +static const u32 a6xx_pc_vs_cluster[] = { + 0x9100, 0x9108, 0x9300, 0x9306, 0x9980, 0x9981, 0x9b00, 0x9b07, +}; + +#define CLUSTER_FE 0 +#define CLUSTER_SP_VS 1 +#define CLUSTER_PC_VS 2 +#define CLUSTER_GRAS 3 +#define CLUSTER_SP_PS 4 +#define CLUSTER_PS 5 +#define CLUSTER_VPC_PS 6 +#define CLUSTER_NONE 7 + +#define CLUSTER(_id, _reg, _sel_reg, _sel_val) \ + { .id = _id, .name = #_id,\ + .registers = _reg, \ + .count = ARRAY_SIZE(_reg), \ + .sel_reg = _sel_reg, .sel_val = _sel_val } + +static const struct a6xx_cluster { + u32 id; + const char *name; + const u32 *registers; + size_t count; + u32 sel_reg; + u32 sel_val; +} a6xx_clusters[] = { + CLUSTER(CLUSTER_GRAS, a6xx_gras_cluster, 0, 0), + CLUSTER(CLUSTER_PS, a6xx_ps_cluster_rac, REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_CD, 0x0), + CLUSTER(CLUSTER_PS, a6xx_ps_cluster_rbp, REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_CD, 0x9), + CLUSTER(CLUSTER_PS, a6xx_ps_cluster, 0, 0), + CLUSTER(CLUSTER_FE, a6xx_fe_cluster, 0, 0), + CLUSTER(CLUSTER_PC_VS, a6xx_pc_vs_cluster, 0, 0), + CLUSTER(CLUSTER_FE, a660_fe_cluster, 0, 0), +}; + +static const u32 a6xx_sp_vs_hlsq_cluster[] = { + 0xb800, 0xb803, 0xb820, 0xb822, +}; + +static const u32 a6xx_sp_vs_sp_cluster[] = { + 0xa800, 0xa824, 0xa830, 0xa83c, 0xa840, 0xa864, 0xa870, 0xa895, + 0xa8a0, 0xa8af, 0xa8c0, 0xa8c3, +}; + +static const u32 a6xx_hlsq_duplicate_cluster[] = { + 0xbb10, 0xbb11, 0xbb20, 0xbb29, +}; + +static const u32 a6xx_hlsq_2d_duplicate_cluster[] = { + 0xbd80, 0xbd80, +}; + +static const u32 a6xx_sp_duplicate_cluster[] = { + 0xab00, 0xab00, 0xab04, 0xab05, 0xab10, 0xab1b, 0xab20, 0xab20, +}; + +static const u32 a6xx_tp_duplicate_cluster[] = { + 0xb300, 0xb307, 0xb309, 0xb309, 0xb380, 0xb382, +}; + +static const u32 a6xx_sp_ps_hlsq_cluster[] = { + 0xb980, 0xb980, 0xb982, 0xb987, 0xb990, 0xb99b, 0xb9a0, 0xb9a2, + 0xb9c0, 0xb9c9, +}; + +static const u32 a6xx_sp_ps_hlsq_2d_cluster[] = { + 0xbd80, 0xbd80, +}; + +static const u32 a6xx_sp_ps_sp_cluster[] = { + 0xa980, 0xa9a8, 0xa9b0, 0xa9bc, 0xa9d0, 0xa9d3, 0xa9e0, 0xa9f3, + 0xaa00, 0xaa00, 0xaa30, 0xaa31, 0xaaf2, 0xaaf2, +}; + +static const u32 a6xx_sp_ps_sp_2d_cluster[] = { + 0xacc0, 0xacc0, +}; + +static const u32 a6xx_sp_ps_tp_cluster[] = { + 0xb180, 0xb183, 0xb190, 0xb191, +}; + +static const u32 a6xx_sp_ps_tp_2d_cluster[] = { + 0xb4c0, 0xb4d1, +}; + +#define CLUSTER_DBGAHB(_id, _base, _type, _reg) \ + { .name = #_id, .statetype = _type, .base = _base, \ + .registers = _reg, .count = ARRAY_SIZE(_reg) } + +static const struct a6xx_dbgahb_cluster { + const char *name; + u32 statetype; + u32 base; + const u32 *registers; + size_t count; +} a6xx_dbgahb_clusters[] = { + CLUSTER_DBGAHB(CLUSTER_SP_VS, 0x0002e000, 0x41, a6xx_sp_vs_hlsq_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_VS, 0x0002a000, 0x21, a6xx_sp_vs_sp_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_VS, 0x0002e000, 0x41, a6xx_hlsq_duplicate_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_VS, 0x0002f000, 0x45, a6xx_hlsq_2d_duplicate_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_VS, 0x0002a000, 0x21, a6xx_sp_duplicate_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_VS, 0x0002c000, 0x1, a6xx_tp_duplicate_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002e000, 0x42, a6xx_sp_ps_hlsq_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002f000, 0x46, a6xx_sp_ps_hlsq_2d_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002a000, 0x22, a6xx_sp_ps_sp_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002b000, 0x26, a6xx_sp_ps_sp_2d_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002c000, 0x2, a6xx_sp_ps_tp_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002d000, 0x6, a6xx_sp_ps_tp_2d_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002e000, 0x42, a6xx_hlsq_duplicate_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002a000, 0x22, a6xx_sp_duplicate_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002c000, 0x2, a6xx_tp_duplicate_cluster), +}; + +static const u32 a6xx_hlsq_registers[] = { + 0xbe00, 0xbe01, 0xbe04, 0xbe05, 0xbe08, 0xbe09, 0xbe10, 0xbe15, + 0xbe20, 0xbe23, +}; + +static const u32 a6xx_sp_registers[] = { + 0xae00, 0xae04, 0xae0c, 0xae0c, 0xae0f, 0xae2b, 0xae30, 0xae32, + 0xae35, 0xae35, 0xae3a, 0xae3f, 0xae50, 0xae52, +}; + +static const u32 a6xx_tp_registers[] = { + 0xb600, 0xb601, 0xb604, 0xb605, 0xb610, 0xb61b, 0xb620, 0xb623, +}; + +struct a6xx_registers { + const u32 *registers; + size_t count; + u32 val0; + u32 val1; +}; + +#define HLSQ_DBG_REGS(_base, _type, _array) \ + { .val0 = _base, .val1 = _type, .registers = _array, \ + .count = ARRAY_SIZE(_array), } + +static const struct a6xx_registers a6xx_hlsq_reglist[] = { + HLSQ_DBG_REGS(0x0002F800, 0x40, a6xx_hlsq_registers), + HLSQ_DBG_REGS(0x0002B800, 0x20, a6xx_sp_registers), + HLSQ_DBG_REGS(0x0002D800, 0x0, a6xx_tp_registers), +}; + +#define SHADER(_type, _size) \ + { .type = _type, .name = #_type, .size = _size } + +static const struct a6xx_shader_block { + const char *name; + u32 type; + u32 size; +} a6xx_shader_blocks[] = { + SHADER(A6XX_TP0_TMO_DATA, 0x200), + SHADER(A6XX_TP0_SMO_DATA, 0x80), + SHADER(A6XX_TP0_MIPMAP_BASE_DATA, 0x3c0), + SHADER(A6XX_TP1_TMO_DATA, 0x200), + SHADER(A6XX_TP1_SMO_DATA, 0x80), + SHADER(A6XX_TP1_MIPMAP_BASE_DATA, 0x3c0), + SHADER(A6XX_SP_INST_DATA, 0x800), + SHADER(A6XX_SP_LB_0_DATA, 0x800), + SHADER(A6XX_SP_LB_1_DATA, 0x800), + SHADER(A6XX_SP_LB_2_DATA, 0x800), + SHADER(A6XX_SP_LB_3_DATA, 0x800), + SHADER(A6XX_SP_LB_4_DATA, 0x800), + SHADER(A6XX_SP_LB_5_DATA, 0x200), + SHADER(A6XX_SP_CB_BINDLESS_DATA, 0x800), + SHADER(A6XX_SP_CB_LEGACY_DATA, 0x280), + SHADER(A6XX_SP_GFX_UAV_BASE_DATA, 0x80), + SHADER(A6XX_SP_INST_TAG, 0x80), + SHADER(A6XX_SP_CB_BINDLESS_TAG, 0x80), + SHADER(A6XX_SP_TMO_UMO_TAG, 0x80), + SHADER(A6XX_SP_SMO_TAG, 0x80), + SHADER(A6XX_SP_STATE_DATA, 0x3f), + SHADER(A6XX_HLSQ_CHUNK_CVS_RAM, 0x1c0), + SHADER(A6XX_HLSQ_CHUNK_CPS_RAM, 0x280), + SHADER(A6XX_HLSQ_CHUNK_CVS_RAM_TAG, 0x40), + SHADER(A6XX_HLSQ_CHUNK_CPS_RAM_TAG, 0x40), + SHADER(A6XX_HLSQ_ICB_CVS_CB_BASE_TAG, 0x4), + SHADER(A6XX_HLSQ_ICB_CPS_CB_BASE_TAG, 0x4), + SHADER(A6XX_HLSQ_CVS_MISC_RAM, 0x1c0), + SHADER(A6XX_HLSQ_CPS_MISC_RAM, 0x580), + SHADER(A6XX_HLSQ_INST_RAM, 0x800), + SHADER(A6XX_HLSQ_GFX_CVS_CONST_RAM, 0x800), + SHADER(A6XX_HLSQ_GFX_CPS_CONST_RAM, 0x800), + SHADER(A6XX_HLSQ_CVS_MISC_RAM_TAG, 0x8), + SHADER(A6XX_HLSQ_CPS_MISC_RAM_TAG, 0x4), + SHADER(A6XX_HLSQ_INST_RAM_TAG, 0x80), + SHADER(A6XX_HLSQ_GFX_CVS_CONST_RAM_TAG, 0xc), + SHADER(A6XX_HLSQ_GFX_CPS_CONST_RAM_TAG, 0x10), + SHADER(A6XX_HLSQ_PWR_REST_RAM, 0x28), + SHADER(A6XX_HLSQ_PWR_REST_TAG, 0x14), + SHADER(A6XX_HLSQ_DATAPATH_META, 0x40), + SHADER(A6XX_HLSQ_FRONTEND_META, 0x40), + SHADER(A6XX_HLSQ_INDIRECT_META, 0x40), + SHADER(A6XX_SP_LB_6_DATA, 0x200), + SHADER(A6XX_SP_LB_7_DATA, 0x200), + SHADER(A6XX_HLSQ_INST_RAM_1, 0x200), +}; + +static const u32 a6xx_rb_rac_registers[] = { + 0x8e04, 0x8e05, 0x8e07, 0x8e08, 0x8e10, 0x8e1c, 0x8e20, 0x8e25, + 0x8e28, 0x8e28, 0x8e2c, 0x8e2f, 0x8e50, 0x8e52, +}; + +static const u32 a6xx_rb_rbp_registers[] = { + 0x8e01, 0x8e01, 0x8e0c, 0x8e0c, 0x8e3b, 0x8e3e, 0x8e40, 0x8e43, + 0x8e53, 0x8e5f, 0x8e70, 0x8e77, +}; + +static const u32 a6xx_registers[] = { + /* RBBM */ + 0x0000, 0x0002, 0x0010, 0x0010, 0x0012, 0x0012, 0x0018, 0x001b, + 0x001e, 0x0032, 0x0038, 0x003c, 0x0042, 0x0042, 0x0044, 0x0044, + 0x0047, 0x0047, 0x0056, 0x0056, 0x00ad, 0x00ae, 0x00b0, 0x00fb, + 0x0100, 0x011d, 0x0200, 0x020d, 0x0218, 0x023d, 0x0400, 0x04f9, + 0x0500, 0x0500, 0x0505, 0x050b, 0x050e, 0x0511, 0x0533, 0x0533, + 0x0540, 0x0555, + /* CP */ + 0x0800, 0x0808, 0x0810, 0x0813, 0x0820, 0x0821, 0x0823, 0x0824, + 0x0826, 0x0827, 0x0830, 0x0833, 0x0840, 0x0845, 0x084f, 0x086f, + 0x0880, 0x088a, 0x08a0, 0x08ab, 0x08c0, 0x08c4, 0x08d0, 0x08dd, + 0x08f0, 0x08f3, 0x0900, 0x0903, 0x0908, 0x0911, 0x0928, 0x093e, + 0x0942, 0x094d, 0x0980, 0x0984, 0x098d, 0x0996, 0x0998, 0x099e, + 0x09a0, 0x09a6, 0x09a8, 0x09ae, 0x09b0, 0x09b1, 0x09c2, 0x09c8, + 0x0a00, 0x0a03, + /* VSC */ + 0x0c00, 0x0c04, 0x0c06, 0x0c06, 0x0c10, 0x0cd9, 0x0e00, 0x0e0e, + /* UCHE */ + 0x0e10, 0x0e13, 0x0e17, 0x0e19, 0x0e1c, 0x0e2b, 0x0e30, 0x0e32, + 0x0e38, 0x0e39, + /* GRAS */ + 0x8600, 0x8601, 0x8610, 0x861b, 0x8620, 0x8620, 0x8628, 0x862b, + 0x8630, 0x8637, + /* VPC */ + 0x9600, 0x9604, 0x9624, 0x9637, + /* PC */ + 0x9e00, 0x9e01, 0x9e03, 0x9e0e, 0x9e11, 0x9e16, 0x9e19, 0x9e19, + 0x9e1c, 0x9e1c, 0x9e20, 0x9e23, 0x9e30, 0x9e31, 0x9e34, 0x9e34, + 0x9e70, 0x9e72, 0x9e78, 0x9e79, 0x9e80, 0x9fff, + /* VFD */ + 0xa600, 0xa601, 0xa603, 0xa603, 0xa60a, 0xa60a, 0xa610, 0xa617, + 0xa630, 0xa630, + /* HLSQ */ + 0xd002, 0xd003, +}; + +static const u32 a660_registers[] = { + /* UCHE */ + 0x0e3c, 0x0e3c, +}; + +#define REGS(_array, _sel_reg, _sel_val) \ + { .registers = _array, .count = ARRAY_SIZE(_array), \ + .val0 = _sel_reg, .val1 = _sel_val } + +static const struct a6xx_registers a6xx_reglist[] = { + REGS(a6xx_registers, 0, 0), + REGS(a660_registers, 0, 0), + REGS(a6xx_rb_rac_registers, REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_CD, 0), + REGS(a6xx_rb_rbp_registers, REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_CD, 9), +}; + +static const u32 a6xx_ahb_registers[] = { + /* RBBM_STATUS - RBBM_STATUS3 */ + 0x210, 0x213, + /* CP_STATUS_1 */ + 0x825, 0x825, +}; + +static const u32 a6xx_vbif_registers[] = { + 0x3000, 0x3007, 0x300c, 0x3014, 0x3018, 0x302d, 0x3030, 0x3031, + 0x3034, 0x3036, 0x303c, 0x303d, 0x3040, 0x3040, 0x3042, 0x3042, + 0x3049, 0x3049, 0x3058, 0x3058, 0x305a, 0x3061, 0x3064, 0x3068, + 0x306c, 0x306d, 0x3080, 0x3088, 0x308b, 0x308c, 0x3090, 0x3094, + 0x3098, 0x3098, 0x309c, 0x309c, 0x30c0, 0x30c0, 0x30c8, 0x30c8, + 0x30d0, 0x30d0, 0x30d8, 0x30d8, 0x30e0, 0x30e0, 0x3100, 0x3100, + 0x3108, 0x3108, 0x3110, 0x3110, 0x3118, 0x3118, 0x3120, 0x3120, + 0x3124, 0x3125, 0x3129, 0x3129, 0x3131, 0x3131, 0x3154, 0x3154, + 0x3156, 0x3156, 0x3158, 0x3158, 0x315a, 0x315a, 0x315c, 0x315c, + 0x315e, 0x315e, 0x3160, 0x3160, 0x3162, 0x3162, 0x340c, 0x340c, + 0x3410, 0x3410, 0x3800, 0x3801, +}; + +static const u32 a6xx_gbif_registers[] = { + 0x3C00, 0X3C0B, 0X3C40, 0X3C47, 0X3CC0, 0X3CD1, 0xE3A, 0xE3A, +}; + +static const struct a6xx_registers a6xx_ahb_reglist = + REGS(a6xx_ahb_registers, 0, 0); + +static const struct a6xx_registers a6xx_vbif_reglist = + REGS(a6xx_vbif_registers, 0, 0); + +static const struct a6xx_registers a6xx_gbif_reglist = + REGS(a6xx_gbif_registers, 0, 0); + +static const u32 a6xx_gmu_gx_registers[] = { + /* GMU GX */ + 0x1a800, 0x1a800, 0x1a810, 0x1a813, 0x1a816, 0x1a816, 0x1a818, 0x1a81b, + 0x1a81e, 0x1a81e, 0x1a820, 0x1a823, 0x1a826, 0x1a826, 0x1a828, 0x1a82b, + 0x1a82e, 0x1a82e, 0x1a830, 0x1a833, 0x1a836, 0x1a836, 0x1a838, 0x1a83b, + 0x1a83e, 0x1a83e, 0x1a840, 0x1a843, 0x1a846, 0x1a846, 0x1a880, 0x1a884, + 0x1a900, 0x1a92b, 0x1a940, 0x1a940, +}; + +static const u32 a6xx_gmu_cx_registers[] = { + /* GMU CX */ + 0x1f400, 0x1f407, 0x1f410, 0x1f412, 0x1f500, 0x1f500, 0x1f507, 0x1f50a, + 0x1f800, 0x1f804, 0x1f807, 0x1f808, 0x1f80b, 0x1f80c, 0x1f80f, 0x1f81c, + 0x1f824, 0x1f82a, 0x1f82d, 0x1f830, 0x1f840, 0x1f853, 0x1f887, 0x1f889, + 0x1f8a0, 0x1f8a2, 0x1f8a4, 0x1f8af, 0x1f8c0, 0x1f8c3, 0x1f8d0, 0x1f8d0, + 0x1f8e4, 0x1f8e4, 0x1f8e8, 0x1f8ec, 0x1f900, 0x1f903, 0x1f940, 0x1f940, + 0x1f942, 0x1f944, 0x1f94c, 0x1f94d, 0x1f94f, 0x1f951, 0x1f954, 0x1f954, + 0x1f957, 0x1f958, 0x1f95d, 0x1f95d, 0x1f962, 0x1f962, 0x1f964, 0x1f965, + 0x1f980, 0x1f986, 0x1f990, 0x1f99e, 0x1f9c0, 0x1f9c0, 0x1f9c5, 0x1f9cc, + 0x1f9e0, 0x1f9e2, 0x1f9f0, 0x1f9f0, 0x1fa00, 0x1fa01, + /* GMU AO */ + 0x23b00, 0x23b16, 0x23c00, 0x23c00, +}; + +static const u32 a6xx_gmu_gpucc_registers[] = { + /* GPU CC */ + 0x24000, 0x24012, 0x24040, 0x24052, 0x24400, 0x24404, 0x24407, 0x2440b, + 0x24415, 0x2441c, 0x2441e, 0x2442d, 0x2443c, 0x2443d, 0x2443f, 0x24440, + 0x24442, 0x24449, 0x24458, 0x2445a, 0x24540, 0x2455e, 0x24800, 0x24802, + 0x24c00, 0x24c02, 0x25400, 0x25402, 0x25800, 0x25802, 0x25c00, 0x25c02, + 0x26000, 0x26002, + /* GPU CC ACD */ + 0x26400, 0x26416, 0x26420, 0x26427, +}; + +static const u32 a621_gmu_gpucc_registers[] = { + /* GPU CC */ + 0x24000, 0x2400e, 0x24400, 0x2440e, 0x24800, 0x24805, 0x24c00, 0x24cff, + 0x25800, 0x25804, 0x25c00, 0x25c04, 0x26000, 0x26004, 0x26400, 0x26405, + 0x26414, 0x2641d, 0x2642a, 0x26430, 0x26432, 0x26432, 0x26441, 0x26455, + 0x26466, 0x26468, 0x26478, 0x2647a, 0x26489, 0x2648a, 0x2649c, 0x2649e, + 0x264a0, 0x264a3, 0x264b3, 0x264b5, 0x264c5, 0x264c7, 0x264d6, 0x264d8, + 0x264e8, 0x264e9, 0x264f9, 0x264fc, 0x2650b, 0x2650c, 0x2651c, 0x2651e, + 0x26540, 0x26570, 0x26600, 0x26616, 0x26620, 0x2662d, +}; + +static const u32 a623_gmu_gpucc_registers[] = { + /* GPU CC */ + 0x24000, 0x2400e, 0x24400, 0x2440e, 0x25800, 0x25804, 0x25c00, 0x25c04, + 0x26000, 0x26004, 0x26400, 0x26405, 0x26414, 0x2641d, 0x2642a, 0x26430, + 0x26432, 0x26432, 0x26441, 0x26455, 0x26466, 0x26468, 0x26478, 0x2647a, + 0x26489, 0x2648a, 0x2649c, 0x2649e, 0x264a0, 0x264a3, 0x264b3, 0x264b5, + 0x264c5, 0x264c7, 0x264d6, 0x264d8, 0x264e8, 0x264e9, 0x264f9, 0x264fc, + 0x2650b, 0x2650c, 0x2651c, 0x2651e, 0x26540, 0x26570, 0x26600, 0x26616, + 0x26620, 0x2662d, +}; + +static const u32 a6xx_gmu_cx_rscc_registers[] = { + /* GPU RSCC */ + 0x008c, 0x008c, 0x0101, 0x0102, 0x0340, 0x0342, 0x0344, 0x0347, + 0x034c, 0x0387, 0x03ec, 0x03ef, 0x03f4, 0x042f, 0x0494, 0x0497, + 0x049c, 0x04d7, 0x053c, 0x053f, 0x0544, 0x057f, +}; + +static const struct a6xx_registers a6xx_gmu_reglist[] = { + REGS(a6xx_gmu_cx_registers, 0, 0), + REGS(a6xx_gmu_cx_rscc_registers, 0, 0), + REGS(a6xx_gmu_gx_registers, 0, 0), +}; + +static const struct a6xx_registers a6xx_gpucc_reg = REGS(a6xx_gmu_gpucc_registers, 0, 0); +static const struct a6xx_registers a621_gpucc_reg = REGS(a621_gmu_gpucc_registers, 0, 0); +static const struct a6xx_registers a623_gpucc_reg = REGS(a623_gmu_gpucc_registers, 0, 0); + +static u32 a6xx_get_cp_roq_size(struct msm_gpu *gpu); +static u32 a7xx_get_cp_roq_size(struct msm_gpu *gpu); + +struct a6xx_indexed_registers { + const char *name; + u32 addr; + u32 data; + u32 count; + u32 (*count_fn)(struct msm_gpu *gpu); +}; + +static const struct a6xx_indexed_registers a6xx_indexed_reglist[] = { + { "CP_SQE_STAT", REG_A6XX_CP_SQE_STAT_ADDR, + REG_A6XX_CP_SQE_STAT_DATA, 0x33, NULL }, + { "CP_DRAW_STATE", REG_A6XX_CP_DRAW_STATE_ADDR, + REG_A6XX_CP_DRAW_STATE_DATA, 0x100, NULL }, + { "CP_SQE_UCODE_DBG", REG_A6XX_CP_SQE_UCODE_DBG_ADDR, + REG_A6XX_CP_SQE_UCODE_DBG_DATA, 0x8000, NULL }, + { "CP_ROQ_DBG", REG_A6XX_CP_ROQ_DBG_ADDR, + REG_A6XX_CP_ROQ_DBG_DATA, 0, a6xx_get_cp_roq_size}, +}; + +static const struct a6xx_indexed_registers a7xx_indexed_reglist[] = { + { "CP_SQE_STAT", REG_A6XX_CP_SQE_STAT_ADDR, + REG_A6XX_CP_SQE_STAT_DATA, 0x40, NULL }, + { "CP_DRAW_STATE", REG_A6XX_CP_DRAW_STATE_ADDR, + REG_A6XX_CP_DRAW_STATE_DATA, 0x100, NULL }, + { "CP_SQE_UCODE_DBG", REG_A6XX_CP_SQE_UCODE_DBG_ADDR, + REG_A6XX_CP_SQE_UCODE_DBG_DATA, 0x8000, NULL }, + { "CP_BV_SQE_STAT", REG_A7XX_CP_BV_SQE_STAT_ADDR, + REG_A7XX_CP_BV_SQE_STAT_DATA, 0x40, NULL }, + { "CP_BV_DRAW_STATE", REG_A7XX_CP_BV_DRAW_STATE_ADDR, + REG_A7XX_CP_BV_DRAW_STATE_DATA, 0x100, NULL }, + { "CP_BV_SQE_UCODE_DBG", REG_A7XX_CP_BV_SQE_UCODE_DBG_ADDR, + REG_A7XX_CP_BV_SQE_UCODE_DBG_DATA, 0x8000, NULL }, + { "CP_SQE_AC_STAT", REG_A7XX_CP_SQE_AC_STAT_ADDR, + REG_A7XX_CP_SQE_AC_STAT_DATA, 0x40, NULL }, + { "CP_LPAC_DRAW_STATE", REG_A7XX_CP_LPAC_DRAW_STATE_ADDR, + REG_A7XX_CP_LPAC_DRAW_STATE_DATA, 0x100, NULL }, + { "CP_SQE_AC_UCODE_DBG", REG_A7XX_CP_SQE_AC_UCODE_DBG_ADDR, + REG_A7XX_CP_SQE_AC_UCODE_DBG_DATA, 0x8000, NULL }, + { "CP_LPAC_FIFO_DBG", REG_A7XX_CP_LPAC_FIFO_DBG_ADDR, + REG_A7XX_CP_LPAC_FIFO_DBG_DATA, 0x40, NULL }, + { "CP_ROQ_DBG", REG_A6XX_CP_ROQ_DBG_ADDR, + REG_A6XX_CP_ROQ_DBG_DATA, 0, a7xx_get_cp_roq_size }, +}; + +static const struct a6xx_indexed_registers a6xx_cp_mempool_indexed = { + "CP_MEM_POOL_DBG", REG_A6XX_CP_MEM_POOL_DBG_ADDR, + REG_A6XX_CP_MEM_POOL_DBG_DATA, 0x2060, NULL, +}; + +static const struct a6xx_indexed_registers a7xx_cp_bv_mempool_indexed[] = { + { "CP_MEM_POOL_DBG", REG_A6XX_CP_MEM_POOL_DBG_ADDR, + REG_A6XX_CP_MEM_POOL_DBG_DATA, 0x2200, NULL }, + { "CP_BV_MEM_POOL_DBG", REG_A7XX_CP_BV_MEM_POOL_DBG_ADDR, + REG_A7XX_CP_BV_MEM_POOL_DBG_DATA, 0x2200, NULL }, +}; + +static const struct a6xx_indexed_registers a7xx_cp_mempool_indexed[] = { + { "CP_MEM_POOL_DBG", REG_A6XX_CP_MEM_POOL_DBG_ADDR, + REG_A6XX_CP_MEM_POOL_DBG_DATA, 0x2200, NULL }, +}; + +#define DEBUGBUS(_id, _count) { .id = _id, .name = #_id, .count = _count } + +static const struct a6xx_debugbus_block { + const char *name; + u32 id; + u32 count; +} a6xx_debugbus_blocks[] = { + DEBUGBUS(A6XX_DBGBUS_CP, 0x100), + DEBUGBUS(A6XX_DBGBUS_RBBM, 0x100), + DEBUGBUS(A6XX_DBGBUS_HLSQ, 0x100), + DEBUGBUS(A6XX_DBGBUS_UCHE, 0x100), + DEBUGBUS(A6XX_DBGBUS_DPM, 0x100), + DEBUGBUS(A6XX_DBGBUS_TESS, 0x100), + DEBUGBUS(A6XX_DBGBUS_PC, 0x100), + DEBUGBUS(A6XX_DBGBUS_VFDP, 0x100), + DEBUGBUS(A6XX_DBGBUS_VPC, 0x100), + DEBUGBUS(A6XX_DBGBUS_TSE, 0x100), + DEBUGBUS(A6XX_DBGBUS_RAS, 0x100), + DEBUGBUS(A6XX_DBGBUS_VSC, 0x100), + DEBUGBUS(A6XX_DBGBUS_COM, 0x100), + DEBUGBUS(A6XX_DBGBUS_LRZ, 0x100), + DEBUGBUS(A6XX_DBGBUS_A2D, 0x100), + DEBUGBUS(A6XX_DBGBUS_CCUFCHE, 0x100), + DEBUGBUS(A6XX_DBGBUS_RBP, 0x100), + DEBUGBUS(A6XX_DBGBUS_DCS, 0x100), + DEBUGBUS(A6XX_DBGBUS_DBGC, 0x100), + DEBUGBUS(A6XX_DBGBUS_GMU_GX, 0x100), + DEBUGBUS(A6XX_DBGBUS_TPFCHE, 0x100), + DEBUGBUS(A6XX_DBGBUS_GPC, 0x100), + DEBUGBUS(A6XX_DBGBUS_LARC, 0x100), + DEBUGBUS(A6XX_DBGBUS_HLSQ_SPTP, 0x100), + DEBUGBUS(A6XX_DBGBUS_RB_0, 0x100), + DEBUGBUS(A6XX_DBGBUS_RB_1, 0x100), + DEBUGBUS(A6XX_DBGBUS_UCHE_WRAPPER, 0x100), + DEBUGBUS(A6XX_DBGBUS_CCU_0, 0x100), + DEBUGBUS(A6XX_DBGBUS_CCU_1, 0x100), + DEBUGBUS(A6XX_DBGBUS_VFD_0, 0x100), + DEBUGBUS(A6XX_DBGBUS_VFD_1, 0x100), + DEBUGBUS(A6XX_DBGBUS_VFD_2, 0x100), + DEBUGBUS(A6XX_DBGBUS_VFD_3, 0x100), + DEBUGBUS(A6XX_DBGBUS_SP_0, 0x100), + DEBUGBUS(A6XX_DBGBUS_SP_1, 0x100), + DEBUGBUS(A6XX_DBGBUS_TPL1_0, 0x100), + DEBUGBUS(A6XX_DBGBUS_TPL1_1, 0x100), + DEBUGBUS(A6XX_DBGBUS_TPL1_2, 0x100), + DEBUGBUS(A6XX_DBGBUS_TPL1_3, 0x100), +}; + +static const struct a6xx_debugbus_block a6xx_gbif_debugbus_block = + DEBUGBUS(A6XX_DBGBUS_VBIF, 0x100); + +static const struct a6xx_debugbus_block a6xx_cx_debugbus_blocks[] = { + DEBUGBUS(A6XX_DBGBUS_GMU_CX, 0x100), + DEBUGBUS(A6XX_DBGBUS_CX, 0x100), +}; + +static const struct a6xx_debugbus_block a650_debugbus_blocks[] = { + DEBUGBUS(A6XX_DBGBUS_RB_2, 0x100), + DEBUGBUS(A6XX_DBGBUS_CCU_2, 0x100), + DEBUGBUS(A6XX_DBGBUS_VFD_4, 0x100), + DEBUGBUS(A6XX_DBGBUS_VFD_5, 0x100), + DEBUGBUS(A6XX_DBGBUS_SP_2, 0x100), + DEBUGBUS(A6XX_DBGBUS_TPL1_4, 0x100), + DEBUGBUS(A6XX_DBGBUS_TPL1_5, 0x100), + DEBUGBUS(A6XX_DBGBUS_SPTP_0, 0x100), + DEBUGBUS(A6XX_DBGBUS_SPTP_1, 0x100), + DEBUGBUS(A6XX_DBGBUS_SPTP_2, 0x100), + DEBUGBUS(A6XX_DBGBUS_SPTP_3, 0x100), + DEBUGBUS(A6XX_DBGBUS_SPTP_4, 0x100), + DEBUGBUS(A6XX_DBGBUS_SPTP_5, 0x100), +}; + +static const u32 a7xx_gbif_debugbus_blocks[] = { + A7XX_DBGBUS_GBIF_CX, + A7XX_DBGBUS_GBIF_GX, +}; + +static const struct a6xx_debugbus_block a7xx_cx_debugbus_blocks[] = { + DEBUGBUS(A7XX_DBGBUS_GMU_CX, 0x100), + DEBUGBUS(A7XX_DBGBUS_CX, 0x100), + DEBUGBUS(A7XX_DBGBUS_GBIF_CX, 0x100), +}; + +#define STATE_NON_CONTEXT 0 +#define STATE_TOGGLE_CTXT 1 +#define STATE_FORCE_CTXT_0 2 +#define STATE_FORCE_CTXT_1 3 + +struct gen7_sel_reg { + unsigned int host_reg; + unsigned int cd_reg; + unsigned int val; +}; + +struct gen7_cluster_registers { + /* cluster_id: Cluster identifier */ + int cluster_id; + /* pipe_id: Pipe Identifier */ + int pipe_id; + /* context_id: one of STATE_ that identifies the context to dump */ + int context_id; + /* regs: Pointer to an array of register pairs */ + const u32 *regs; + /* sel: Pointer to a selector register to write before reading */ + const struct gen7_sel_reg *sel; +}; + +struct gen7_sptp_cluster_registers { + /* cluster_id: Cluster identifier */ + enum a7xx_cluster cluster_id; + /* statetype: SP block state type for the cluster */ + enum a7xx_statetype_id statetype; + /* pipe_id: Pipe identifier */ + enum adreno_pipe pipe_id; + /* context_id: Context identifier */ + int context_id; + /* location_id: Location identifier */ + enum a7xx_state_location location_id; + /* regs: Pointer to the list of register pairs to read */ + const u32 *regs; + /* regbase: Dword offset of the register block in the GPu register space */ + unsigned int regbase; +}; + +struct gen7_shader_block { + /* statetype: Type identifer for the block */ + u32 statetype; + /* size: Size of the block (in dwords) */ + u32 size; + /* num_sps: The SP id to dump */ + u32 num_sps; + /* num_usptps: The number of USPTPs to dump */; + u32 num_usptps; + /* pipe_id: Pipe identifier for the block data */ + u32 pipeid; + /* location: Location identifer for the block data */ + u32 location; +}; + +struct gen7_reg_list { + const u32 *regs; + const struct gen7_sel_reg *sel; +}; + +/* adreno_gen7_x_y_snapshot.h defines which debugbus blocks a given family has, but the + * list of debugbus blocks is global on a7xx. + */ + +#define A7XX_DEBUGBUS(_id, _count) [_id] = { .id = _id, .name = #_id, .count = _count }, +static const struct a6xx_debugbus_block a7xx_debugbus_blocks[] = { + A7XX_DEBUGBUS(A7XX_DBGBUS_CP_0_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_CP_0_1, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_RBBM, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_GBIF_GX, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_GBIF_CX, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_HLSQ, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_UCHE_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_UCHE_1, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TESS_BR, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TESS_BV, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_PC_BR, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_PC_BV, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VFDP_BR, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VFDP_BV, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VPC_BR, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VPC_BV, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TSE_BR, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TSE_BV, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_RAS_BR, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_RAS_BV, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VSC, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_COM_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_LRZ_BR, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_LRZ_BV, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_UFC_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_UFC_1, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_GMU_GX, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_DBGC, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_CX, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_GMU_CX, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_GPC_BR, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_GPC_BV, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_LARC, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_HLSQ_SPTP, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_RB_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_RB_1, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_RB_2, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_RB_3, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_RB_4, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_RB_5, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_UCHE_WRAPPER, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_CCU_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_CCU_1, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_CCU_2, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_CCU_3, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_CCU_4, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_CCU_5, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VFD_BR_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VFD_BR_1, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VFD_BR_2, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VFD_BR_3, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VFD_BR_4, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VFD_BR_5, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VFD_BR_6, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VFD_BR_7, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VFD_BV_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VFD_BV_1, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VFD_BV_2, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VFD_BV_3, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USP_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USP_1, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USP_2, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USP_3, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USP_4, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USP_5, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TP_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TP_1, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TP_2, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TP_3, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TP_4, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TP_5, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TP_6, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TP_7, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TP_8, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TP_9, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TP_10, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_TP_11, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USPTP_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USPTP_1, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USPTP_2, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USPTP_3, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USPTP_4, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USPTP_5, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USPTP_6, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USPTP_7, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USPTP_8, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USPTP_9, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USPTP_10, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_USPTP_11, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_CCHE_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_CCHE_1, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_CCHE_2, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VPC_DSTR_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VPC_DSTR_1, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_VPC_DSTR_2, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_HLSQ_DP_STR_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_HLSQ_DP_STR_1, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_HLSQ_DP_STR_2, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_HLSQ_DP_STR_3, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_HLSQ_DP_STR_4, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_HLSQ_DP_STR_5, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_UFC_DSTR_0, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_UFC_DSTR_1, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_UFC_DSTR_2, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_CGC_SUBCORE, 0x100) + A7XX_DEBUGBUS(A7XX_DBGBUS_CGC_CORE, 0x100) +}; + +#define A7XX_NAME(enumval) [enumval] = #enumval +static const char *a7xx_statetype_names[] = { + A7XX_NAME(A7XX_TP0_NCTX_REG), + A7XX_NAME(A7XX_TP0_CTX0_3D_CVS_REG), + A7XX_NAME(A7XX_TP0_CTX0_3D_CPS_REG), + A7XX_NAME(A7XX_TP0_CTX1_3D_CVS_REG), + A7XX_NAME(A7XX_TP0_CTX1_3D_CPS_REG), + A7XX_NAME(A7XX_TP0_CTX2_3D_CPS_REG), + A7XX_NAME(A7XX_TP0_CTX3_3D_CPS_REG), + A7XX_NAME(A7XX_TP0_TMO_DATA), + A7XX_NAME(A7XX_TP0_SMO_DATA), + A7XX_NAME(A7XX_TP0_MIPMAP_BASE_DATA), + A7XX_NAME(A7XX_SP_NCTX_REG), + A7XX_NAME(A7XX_SP_CTX0_3D_CVS_REG), + A7XX_NAME(A7XX_SP_CTX0_3D_CPS_REG), + A7XX_NAME(A7XX_SP_CTX1_3D_CVS_REG), + A7XX_NAME(A7XX_SP_CTX1_3D_CPS_REG), + A7XX_NAME(A7XX_SP_CTX2_3D_CPS_REG), + A7XX_NAME(A7XX_SP_CTX3_3D_CPS_REG), + A7XX_NAME(A7XX_SP_INST_DATA), + A7XX_NAME(A7XX_SP_INST_DATA_1), + A7XX_NAME(A7XX_SP_LB_0_DATA), + A7XX_NAME(A7XX_SP_LB_1_DATA), + A7XX_NAME(A7XX_SP_LB_2_DATA), + A7XX_NAME(A7XX_SP_LB_3_DATA), + A7XX_NAME(A7XX_SP_LB_4_DATA), + A7XX_NAME(A7XX_SP_LB_5_DATA), + A7XX_NAME(A7XX_SP_LB_6_DATA), + A7XX_NAME(A7XX_SP_LB_7_DATA), + A7XX_NAME(A7XX_SP_CB_RAM), + A7XX_NAME(A7XX_SP_LB_13_DATA), + A7XX_NAME(A7XX_SP_LB_14_DATA), + A7XX_NAME(A7XX_SP_INST_TAG), + A7XX_NAME(A7XX_SP_INST_DATA_2), + A7XX_NAME(A7XX_SP_TMO_TAG), + A7XX_NAME(A7XX_SP_SMO_TAG), + A7XX_NAME(A7XX_SP_STATE_DATA), + A7XX_NAME(A7XX_SP_HWAVE_RAM), + A7XX_NAME(A7XX_SP_L0_INST_BUF), + A7XX_NAME(A7XX_SP_LB_8_DATA), + A7XX_NAME(A7XX_SP_LB_9_DATA), + A7XX_NAME(A7XX_SP_LB_10_DATA), + A7XX_NAME(A7XX_SP_LB_11_DATA), + A7XX_NAME(A7XX_SP_LB_12_DATA), + A7XX_NAME(A7XX_HLSQ_DATAPATH_DSTR_META), + A7XX_NAME(A7XX_HLSQ_L2STC_TAG_RAM), + A7XX_NAME(A7XX_HLSQ_L2STC_INFO_CMD), + A7XX_NAME(A7XX_HLSQ_CVS_BE_CTXT_BUF_RAM_TAG), + A7XX_NAME(A7XX_HLSQ_CPS_BE_CTXT_BUF_RAM_TAG), + A7XX_NAME(A7XX_HLSQ_GFX_CVS_BE_CTXT_BUF_RAM), + A7XX_NAME(A7XX_HLSQ_GFX_CPS_BE_CTXT_BUF_RAM), + A7XX_NAME(A7XX_HLSQ_CHUNK_CVS_RAM), + A7XX_NAME(A7XX_HLSQ_CHUNK_CPS_RAM), + A7XX_NAME(A7XX_HLSQ_CHUNK_CVS_RAM_TAG), + A7XX_NAME(A7XX_HLSQ_CHUNK_CPS_RAM_TAG), + A7XX_NAME(A7XX_HLSQ_ICB_CVS_CB_BASE_TAG), + A7XX_NAME(A7XX_HLSQ_ICB_CPS_CB_BASE_TAG), + A7XX_NAME(A7XX_HLSQ_CVS_MISC_RAM), + A7XX_NAME(A7XX_HLSQ_CPS_MISC_RAM), + A7XX_NAME(A7XX_HLSQ_CPS_MISC_RAM_1), + A7XX_NAME(A7XX_HLSQ_INST_RAM), + A7XX_NAME(A7XX_HLSQ_GFX_CVS_CONST_RAM), + A7XX_NAME(A7XX_HLSQ_GFX_CPS_CONST_RAM), + A7XX_NAME(A7XX_HLSQ_CVS_MISC_RAM_TAG), + A7XX_NAME(A7XX_HLSQ_CPS_MISC_RAM_TAG), + A7XX_NAME(A7XX_HLSQ_INST_RAM_TAG), + A7XX_NAME(A7XX_HLSQ_GFX_CVS_CONST_RAM_TAG), + A7XX_NAME(A7XX_HLSQ_GFX_CPS_CONST_RAM_TAG), + A7XX_NAME(A7XX_HLSQ_GFX_LOCAL_MISC_RAM), + A7XX_NAME(A7XX_HLSQ_GFX_LOCAL_MISC_RAM_TAG), + A7XX_NAME(A7XX_HLSQ_INST_RAM_1), + A7XX_NAME(A7XX_HLSQ_STPROC_META), + A7XX_NAME(A7XX_HLSQ_BV_BE_META), + A7XX_NAME(A7XX_HLSQ_INST_RAM_2), + A7XX_NAME(A7XX_HLSQ_DATAPATH_META), + A7XX_NAME(A7XX_HLSQ_FRONTEND_META), + A7XX_NAME(A7XX_HLSQ_INDIRECT_META), + A7XX_NAME(A7XX_HLSQ_BACKEND_META), +}; + +static const char *a7xx_pipe_names[] = { + A7XX_NAME(PIPE_NONE), + A7XX_NAME(PIPE_BR), + A7XX_NAME(PIPE_BV), + A7XX_NAME(PIPE_LPAC), +}; + +static const char *a7xx_cluster_names[] = { + A7XX_NAME(A7XX_CLUSTER_NONE), + A7XX_NAME(A7XX_CLUSTER_FE), + A7XX_NAME(A7XX_CLUSTER_SP_VS), + A7XX_NAME(A7XX_CLUSTER_PC_VS), + A7XX_NAME(A7XX_CLUSTER_GRAS), + A7XX_NAME(A7XX_CLUSTER_SP_PS), + A7XX_NAME(A7XX_CLUSTER_VPC_PS), + A7XX_NAME(A7XX_CLUSTER_PS), +}; + +#endif diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c new file mode 100644 index 000000000..186a73c0b --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c @@ -0,0 +1,1102 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2017-2018 The Linux Foundation. All rights reserved. */ + +#include +#include +#include + +#include +#include + +#include "a6xx_gmu.h" +#include "a6xx_gmu.xml.h" +#include "a6xx_gpu.h" + +#define HFI_MSG_ID(val) [val] = #val + +static const char * const a6xx_hfi_msg_id[] = { + HFI_MSG_ID(HFI_H2F_MSG_INIT), + HFI_MSG_ID(HFI_H2F_MSG_FW_VERSION), + HFI_MSG_ID(HFI_H2F_MSG_BW_TABLE), + HFI_MSG_ID(HFI_H2F_MSG_PERF_TABLE), + HFI_MSG_ID(HFI_H2F_MSG_TEST), + HFI_MSG_ID(HFI_H2F_MSG_START), + HFI_MSG_ID(HFI_H2F_FEATURE_CTRL), + HFI_MSG_ID(HFI_H2F_MSG_CORE_FW_START), + HFI_MSG_ID(HFI_H2F_MSG_TABLE), + HFI_MSG_ID(HFI_H2F_MSG_GX_BW_PERF_VOTE), + HFI_MSG_ID(HFI_H2F_MSG_PREPARE_SLUMBER), +}; + +static int a6xx_hfi_queue_read(struct a6xx_gmu *gmu, + struct a6xx_hfi_queue *queue, u32 *data, u32 dwords) +{ + struct a6xx_hfi_queue_header *header = queue->header; + u32 i, hdr, index = header->read_index; + + if (header->read_index == READ_ONCE(header->write_index)) { + header->rx_request = 1; + return 0; + } + + hdr = queue->data[index]; + + queue->history[(queue->history_idx++) % HFI_HISTORY_SZ] = index; + + /* + * If we are to assume that the GMU firmware is in fact a rational actor + * and is programmed to not send us a larger response than we expect + * then we can also assume that if the header size is unexpectedly large + * that it is due to memory corruption and/or hardware failure. In this + * case the only reasonable course of action is to BUG() to help harden + * the failure. + */ + + BUG_ON(HFI_HEADER_SIZE(hdr) > dwords); + + for (i = 0; i < HFI_HEADER_SIZE(hdr); i++) { + data[i] = queue->data[index]; + index = (index + 1) % header->size; + } + + if (!gmu->legacy) + index = ALIGN(index, 4) % header->size; + + /* Ensure all memory operations are complete before updating the read index */ + dma_mb(); + + WRITE_ONCE(header->read_index, index); + return HFI_HEADER_SIZE(hdr); +} + +static int a6xx_hfi_queue_write(struct a6xx_gmu *gmu, + struct a6xx_hfi_queue *queue, u32 *data, u32 dwords) +{ + struct a6xx_hfi_queue_header *header = queue->header; + u32 i, space, index = header->write_index; + + spin_lock(&queue->lock); + + space = CIRC_SPACE(header->write_index, READ_ONCE(header->read_index), + header->size); + if (space < dwords) { + header->dropped++; + spin_unlock(&queue->lock); + return -ENOSPC; + } + + queue->history[(queue->history_idx++) % HFI_HISTORY_SZ] = index; + + for (i = 0; i < dwords; i++) { + queue->data[index] = data[i]; + index = (index + 1) % header->size; + } + + /* Cookify any non used data at the end of the write buffer */ + if (!gmu->legacy) { + for (; index % 4; index = (index + 1) % header->size) + queue->data[index] = 0xfafafafa; + } + + /* Ensure all memory operations are complete before updating the write index */ + dma_mb(); + + WRITE_ONCE(header->write_index, index); + spin_unlock(&queue->lock); + + gmu_write(gmu, REG_A6XX_GMU_HOST2GMU_INTR_SET, 0x01); + return 0; +} + +static int a6xx_hfi_wait_for_msg_interrupt(struct a6xx_gmu *gmu, u32 id, u32 seqnum) +{ + int ret; + u32 val; + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + + do { + /* Wait for a response */ + ret = gmu_poll_timeout(gmu, REG_A6XX_GMU_GMU2HOST_INTR_INFO, val, + val & A6XX_GMU_GMU2HOST_INTR_INFO_MSGQ, 100, 1000000); + + if (!ret) + break; + + if (completion_done(&a6xx_gpu->base.fault_coredump_done)) + break; + + /* We may timeout because the GMU is temporarily wedged from + * pending faults from the GPU and we are taking a devcoredump. + * Wait until the MMU is resumed and try again. + */ + wait_for_completion(&a6xx_gpu->base.fault_coredump_done); + } while (true); + + if (ret) { + DRM_DEV_ERROR(gmu->dev, + "Message %s id %d timed out waiting for response\n", + a6xx_hfi_msg_id[id], seqnum); + return -ETIMEDOUT; + } + + /* Clear the interrupt */ + gmu_write(gmu, REG_A6XX_GMU_GMU2HOST_INTR_CLR, + A6XX_GMU_GMU2HOST_INTR_INFO_MSGQ); + + return 0; +} + +static int a6xx_hfi_wait_for_ack(struct a6xx_gmu *gmu, u32 id, u32 seqnum, + u32 *payload, u32 payload_size) +{ + struct a6xx_hfi_queue *queue = &gmu->queues[HFI_RESPONSE_QUEUE]; + int ret; + + ret = a6xx_hfi_wait_for_msg_interrupt(gmu, id, seqnum); + if (ret) + return ret; + + for (;;) { + struct a6xx_hfi_msg_response resp; + + /* Get the next packet */ + ret = a6xx_hfi_queue_read(gmu, queue, (u32 *) &resp, + sizeof(resp) >> 2); + + /* If the queue is empty, there may have been previous missed + * responses that preceded the response to our packet. Wait + * further before we give up. + */ + if (!ret) { + ret = a6xx_hfi_wait_for_msg_interrupt(gmu, id, seqnum); + if (ret) { + DRM_DEV_ERROR(gmu->dev, + "The HFI response queue is unexpectedly empty\n"); + return ret; + } + continue; + } + + if (HFI_HEADER_ID(resp.header) == HFI_F2H_MSG_ERROR) { + struct a6xx_hfi_msg_error *error = + (struct a6xx_hfi_msg_error *) &resp; + + DRM_DEV_ERROR(gmu->dev, "GMU firmware error %d\n", + error->code); + continue; + } + + if (seqnum != HFI_HEADER_SEQNUM(resp.ret_header)) { + DRM_DEV_ERROR(gmu->dev, + "Unexpected message id %d on the response queue\n", + HFI_HEADER_SEQNUM(resp.ret_header)); + continue; + } + + if (resp.error) { + DRM_DEV_ERROR(gmu->dev, + "Message %s id %d returned error %d\n", + a6xx_hfi_msg_id[id], seqnum, resp.error); + return -EINVAL; + } + + /* All is well, copy over the buffer */ + if (payload && payload_size) + memcpy(payload, resp.payload, + min_t(u32, payload_size, sizeof(resp.payload))); + + return 0; + } +} + +static int a6xx_hfi_send_msg(struct a6xx_gmu *gmu, int id, + void *data, u32 size, u32 *payload, u32 payload_size) +{ + struct a6xx_hfi_queue *queue = &gmu->queues[HFI_COMMAND_QUEUE]; + int ret, dwords = size >> 2; + u32 seqnum; + + seqnum = atomic_inc_return(&queue->seqnum) % 0xfff; + + /* First dword of the message is the message header - fill it in */ + *((u32 *) data) = (seqnum << 20) | (HFI_MSG_CMD << 16) | + (dwords << 8) | id; + + ret = a6xx_hfi_queue_write(gmu, queue, data, dwords); + if (ret) { + DRM_DEV_ERROR(gmu->dev, "Unable to send message %s id %d\n", + a6xx_hfi_msg_id[id], seqnum); + return ret; + } + + return a6xx_hfi_wait_for_ack(gmu, id, seqnum, payload, payload_size); +} + +static int a6xx_hfi_send_gmu_init(struct a6xx_gmu *gmu, int boot_state) +{ + struct a6xx_hfi_msg_gmu_init_cmd msg = { 0 }; + + msg.dbg_buffer_addr = (u32) gmu->debug.iova; + msg.dbg_buffer_size = (u32) gmu->debug.size; + msg.boot_state = boot_state; + + return a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_INIT, &msg, sizeof(msg), + NULL, 0); +} + +static int a6xx_hfi_get_fw_version(struct a6xx_gmu *gmu, u32 *version) +{ + struct a6xx_hfi_msg_fw_version msg = { 0 }; + + /* Currently supporting version 1.10 */ + msg.supported_version = (1 << 28) | (1 << 19) | (1 << 17); + + return a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_FW_VERSION, &msg, sizeof(msg), + version, sizeof(*version)); +} + +static int a6xx_hfi_send_perf_table_v1(struct a6xx_gmu *gmu) +{ + struct a6xx_hfi_msg_perf_table_v1 msg = { 0 }; + int i; + + msg.num_gpu_levels = gmu->nr_gpu_freqs; + msg.num_gmu_levels = gmu->nr_gmu_freqs; + + for (i = 0; i < gmu->nr_gpu_freqs; i++) { + msg.gx_votes[i].vote = gmu->gx_arc_votes[i]; + msg.gx_votes[i].freq = gmu->gpu_freqs[i] / 1000; + } + + for (i = 0; i < gmu->nr_gmu_freqs; i++) { + msg.cx_votes[i].vote = gmu->cx_arc_votes[i]; + msg.cx_votes[i].freq = gmu->gmu_freqs[i] / 1000; + } + + return a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_PERF_TABLE, &msg, sizeof(msg), + NULL, 0); +} + +static int a8xx_hfi_send_perf_table(struct a6xx_gmu *gmu) +{ + unsigned int num_gx_votes = 3, num_cx_votes = 2; + struct a6xx_hfi_table_entry *entry; + struct a6xx_hfi_table *tbl; + int ret, i; + u32 size; + + size = sizeof(*tbl) + (2 * sizeof(tbl->entry[0])) + + (gmu->nr_gpu_freqs * num_gx_votes * sizeof(gmu->gx_arc_votes[0])) + + (gmu->nr_gmu_freqs * num_cx_votes * sizeof(gmu->cx_arc_votes[0])); + tbl = kzalloc(size, GFP_KERNEL); + if (!tbl) + return -ENOMEM; + tbl->type = HFI_TABLE_GPU_PERF; + + /* First fill GX votes */ + entry = &tbl->entry[0]; + entry->count = gmu->nr_gpu_freqs; + entry->stride = num_gx_votes; + + for (i = 0; i < gmu->nr_gpu_freqs; i++) { + unsigned int base = i * entry->stride; + + entry->data[base+0] = gmu->gx_arc_votes[i]; + entry->data[base+1] = gmu->dep_arc_votes[i]; + entry->data[base+2] = gmu->gpu_freqs[i] / 1000; + } + + /* Then fill CX votes */ + entry = (struct a6xx_hfi_table_entry *) + &tbl->entry[0].data[gmu->nr_gpu_freqs * num_gx_votes]; + + entry->count = gmu->nr_gmu_freqs; + entry->stride = num_cx_votes; + + for (i = 0; i < gmu->nr_gmu_freqs; i++) { + unsigned int base = i * entry->stride; + + entry->data[base] = gmu->cx_arc_votes[i]; + entry->data[base+1] = gmu->gmu_freqs[i] / 1000; + } + + ret = a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_TABLE, tbl, size, NULL, 0); + + kfree(tbl); + return ret; +} + +static int a6xx_hfi_send_perf_table(struct a6xx_gmu *gmu) +{ + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct a6xx_hfi_msg_perf_table msg = { 0 }; + int i; + + if (adreno_is_a8xx(adreno_gpu)) + return a8xx_hfi_send_perf_table(gmu); + + msg.num_gpu_levels = gmu->nr_gpu_freqs; + msg.num_gmu_levels = gmu->nr_gmu_freqs; + + for (i = 0; i < gmu->nr_gpu_freqs; i++) { + msg.gx_votes[i].vote = gmu->gx_arc_votes[i]; + msg.gx_votes[i].acd = 0xffffffff; + msg.gx_votes[i].freq = gmu->gpu_freqs[i] / 1000; + } + + for (i = 0; i < gmu->nr_gmu_freqs; i++) { + msg.cx_votes[i].vote = gmu->cx_arc_votes[i]; + msg.cx_votes[i].freq = gmu->gmu_freqs[i] / 1000; + } + + return a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_PERF_TABLE, &msg, sizeof(msg), + NULL, 0); +} + +static void a6xx_generate_bw_table(const struct a6xx_info *info, struct a6xx_gmu *gmu, + struct a6xx_hfi_msg_bw_table *msg) +{ + unsigned int i, j; + + for (i = 0; i < GMU_MAX_BCMS; i++) { + if (!info->bcms[i].name) + break; + msg->ddr_cmds_addrs[i] = cmd_db_read_addr(info->bcms[i].name); + } + msg->ddr_cmds_num = i; + + for (i = 0; i < gmu->nr_gpu_bws; ++i) + for (j = 0; j < msg->ddr_cmds_num; j++) + msg->ddr_cmds_data[i][j] = gmu->gpu_ib_votes[i][j]; + msg->bw_level_num = gmu->nr_gpu_bws; + + /* Compute the wait bitmask with each BCM having the commit bit */ + msg->ddr_wait_bitmask = 0; + for (j = 0; j < msg->ddr_cmds_num; j++) + if (msg->ddr_cmds_data[0][j] & BCM_TCS_CMD_COMMIT_MASK) + msg->ddr_wait_bitmask |= BIT(j); + + /* + * These are the CX (CNOC) votes - these are used by the GMU + * The 'CN0' BCM is used on all targets, and votes are basically + * 'off' and 'on' states with first bit to enable the path. + */ + + msg->cnoc_cmds_addrs[0] = cmd_db_read_addr("CN0"); + msg->cnoc_cmds_num = 1; + + msg->cnoc_cmds_data[0][0] = BCM_TCS_CMD(true, false, 0, 0); + msg->cnoc_cmds_data[1][0] = BCM_TCS_CMD(true, true, 0, BIT(0)); + + /* Compute the wait bitmask with each BCM having the commit bit */ + msg->cnoc_wait_bitmask = 0; + for (j = 0; j < msg->cnoc_cmds_num; j++) + if (msg->cnoc_cmds_data[0][j] & BCM_TCS_CMD_COMMIT_MASK) + msg->cnoc_wait_bitmask |= BIT(j); +} + +static void a618_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + /* Send a single "off" entry since the 618 GMU doesn't do bus scaling */ + msg->bw_level_num = 1; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x01; + + msg->ddr_cmds_addrs[0] = 0x50000; + msg->ddr_cmds_addrs[1] = 0x5003c; + msg->ddr_cmds_addrs[2] = 0x5000c; + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; + + /* + * These are the CX (CNOC) votes - these are used by the GMU but the + * votes are known and fixed for the target + */ + msg->cnoc_cmds_num = 1; + msg->cnoc_wait_bitmask = 0x01; + + msg->cnoc_cmds_addrs[0] = 0x5007c; + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[1][0] = 0x60000001; +} + +static void a619_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + msg->bw_level_num = 13; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x0; + + msg->ddr_cmds_addrs[0] = 0x50000; + msg->ddr_cmds_addrs[1] = 0x50004; + msg->ddr_cmds_addrs[2] = 0x50080; + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; + msg->ddr_cmds_data[1][0] = 0x6000030c; + msg->ddr_cmds_data[1][1] = 0x600000db; + msg->ddr_cmds_data[1][2] = 0x60000008; + msg->ddr_cmds_data[2][0] = 0x60000618; + msg->ddr_cmds_data[2][1] = 0x600001b6; + msg->ddr_cmds_data[2][2] = 0x60000008; + msg->ddr_cmds_data[3][0] = 0x60000925; + msg->ddr_cmds_data[3][1] = 0x60000291; + msg->ddr_cmds_data[3][2] = 0x60000008; + msg->ddr_cmds_data[4][0] = 0x60000dc1; + msg->ddr_cmds_data[4][1] = 0x600003dc; + msg->ddr_cmds_data[4][2] = 0x60000008; + msg->ddr_cmds_data[5][0] = 0x600010ad; + msg->ddr_cmds_data[5][1] = 0x600004ae; + msg->ddr_cmds_data[5][2] = 0x60000008; + msg->ddr_cmds_data[6][0] = 0x600014c3; + msg->ddr_cmds_data[6][1] = 0x600005d4; + msg->ddr_cmds_data[6][2] = 0x60000008; + msg->ddr_cmds_data[7][0] = 0x6000176a; + msg->ddr_cmds_data[7][1] = 0x60000693; + msg->ddr_cmds_data[7][2] = 0x60000008; + msg->ddr_cmds_data[8][0] = 0x60001f01; + msg->ddr_cmds_data[8][1] = 0x600008b5; + msg->ddr_cmds_data[8][2] = 0x60000008; + msg->ddr_cmds_data[9][0] = 0x60002940; + msg->ddr_cmds_data[9][1] = 0x60000b95; + msg->ddr_cmds_data[9][2] = 0x60000008; + msg->ddr_cmds_data[10][0] = 0x60002f68; + msg->ddr_cmds_data[10][1] = 0x60000d50; + msg->ddr_cmds_data[10][2] = 0x60000008; + msg->ddr_cmds_data[11][0] = 0x60003700; + msg->ddr_cmds_data[11][1] = 0x60000f71; + msg->ddr_cmds_data[11][2] = 0x60000008; + msg->ddr_cmds_data[12][0] = 0x60003fce; + msg->ddr_cmds_data[12][1] = 0x600011ea; + msg->ddr_cmds_data[12][2] = 0x60000008; + + msg->cnoc_cmds_num = 1; + msg->cnoc_wait_bitmask = 0x0; + + msg->cnoc_cmds_addrs[0] = 0x50054; + + msg->cnoc_cmds_data[0][0] = 0x40000000; +} + +static void a640_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + /* + * Send a single "off" entry just to get things running + * TODO: bus scaling + */ + msg->bw_level_num = 1; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x01; + + msg->ddr_cmds_addrs[0] = 0x50000; + msg->ddr_cmds_addrs[1] = 0x5003c; + msg->ddr_cmds_addrs[2] = 0x5000c; + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; + + /* + * These are the CX (CNOC) votes - these are used by the GMU but the + * votes are known and fixed for the target + */ + msg->cnoc_cmds_num = 3; + msg->cnoc_wait_bitmask = 0x01; + + msg->cnoc_cmds_addrs[0] = 0x50034; + msg->cnoc_cmds_addrs[1] = 0x5007c; + msg->cnoc_cmds_addrs[2] = 0x5004c; + + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[0][1] = 0x00000000; + msg->cnoc_cmds_data[0][2] = 0x40000000; + + msg->cnoc_cmds_data[1][0] = 0x60000001; + msg->cnoc_cmds_data[1][1] = 0x20000001; + msg->cnoc_cmds_data[1][2] = 0x60000001; +} + +static void a650_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + /* + * Send a single "off" entry just to get things running + * TODO: bus scaling + */ + msg->bw_level_num = 1; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x01; + + msg->ddr_cmds_addrs[0] = 0x50000; + msg->ddr_cmds_addrs[1] = 0x50004; + msg->ddr_cmds_addrs[2] = 0x5007c; + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; + + /* + * These are the CX (CNOC) votes - these are used by the GMU but the + * votes are known and fixed for the target + */ + msg->cnoc_cmds_num = 1; + msg->cnoc_wait_bitmask = 0x01; + + msg->cnoc_cmds_addrs[0] = 0x500a4; + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[1][0] = 0x60000001; +} + +static void a690_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + /* + * Send a single "off" entry just to get things running + * TODO: bus scaling + */ + msg->bw_level_num = 1; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x01; + + msg->ddr_cmds_addrs[0] = 0x50004; + msg->ddr_cmds_addrs[1] = 0x50000; + msg->ddr_cmds_addrs[2] = 0x500ac; + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; + + /* + * These are the CX (CNOC) votes - these are used by the GMU but the + * votes are known and fixed for the target + */ + msg->cnoc_cmds_num = 1; + msg->cnoc_wait_bitmask = 0x01; + + msg->cnoc_cmds_addrs[0] = 0x5003c; + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[1][0] = 0x60000001; +} + +static void a660_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + /* + * Send a single "off" entry just to get things running + * TODO: bus scaling + */ + msg->bw_level_num = 1; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x01; + + msg->ddr_cmds_addrs[0] = 0x50004; + msg->ddr_cmds_addrs[1] = 0x500a0; + msg->ddr_cmds_addrs[2] = 0x50000; + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; + + /* + * These are the CX (CNOC) votes - these are used by the GMU but the + * votes are known and fixed for the target + */ + msg->cnoc_cmds_num = 1; + msg->cnoc_wait_bitmask = 0x01; + + msg->cnoc_cmds_addrs[0] = 0x50070; + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[1][0] = 0x60000001; +} + +static void a663_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + /* + * Send a single "off" entry just to get things running + * TODO: bus scaling + */ + msg->bw_level_num = 1; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x07; + + msg->ddr_cmds_addrs[0] = 0x50004; + msg->ddr_cmds_addrs[1] = 0x50000; + msg->ddr_cmds_addrs[2] = 0x500b4; + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; + + /* + * These are the CX (CNOC) votes - these are used by the GMU but the + * votes are known and fixed for the target + */ + msg->cnoc_cmds_num = 1; + msg->cnoc_wait_bitmask = 0x01; + + msg->cnoc_cmds_addrs[0] = 0x50058; + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[1][0] = 0x60000001; +} + +static void adreno_7c3_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + /* + * Send a single "off" entry just to get things running + * TODO: bus scaling + */ + msg->bw_level_num = 1; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x07; + + msg->ddr_cmds_addrs[0] = 0x50004; + msg->ddr_cmds_addrs[1] = 0x50000; + msg->ddr_cmds_addrs[2] = 0x50088; + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; + + /* + * These are the CX (CNOC) votes - these are used by the GMU but the + * votes are known and fixed for the target + */ + msg->cnoc_cmds_num = 1; + msg->cnoc_wait_bitmask = 0x01; + + msg->cnoc_cmds_addrs[0] = 0x5006c; + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[1][0] = 0x60000001; +} + +static void a730_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + msg->bw_level_num = 12; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x7; + + msg->ddr_cmds_addrs[0] = cmd_db_read_addr("SH0"); + msg->ddr_cmds_addrs[1] = cmd_db_read_addr("MC0"); + msg->ddr_cmds_addrs[2] = cmd_db_read_addr("ACV"); + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; + msg->ddr_cmds_data[1][0] = 0x600002e8; + msg->ddr_cmds_data[1][1] = 0x600003d0; + msg->ddr_cmds_data[1][2] = 0x60000008; + msg->ddr_cmds_data[2][0] = 0x6000068d; + msg->ddr_cmds_data[2][1] = 0x6000089a; + msg->ddr_cmds_data[2][2] = 0x60000008; + msg->ddr_cmds_data[3][0] = 0x600007f2; + msg->ddr_cmds_data[3][1] = 0x60000a6e; + msg->ddr_cmds_data[3][2] = 0x60000008; + msg->ddr_cmds_data[4][0] = 0x600009e5; + msg->ddr_cmds_data[4][1] = 0x60000cfd; + msg->ddr_cmds_data[4][2] = 0x60000008; + msg->ddr_cmds_data[5][0] = 0x60000b29; + msg->ddr_cmds_data[5][1] = 0x60000ea6; + msg->ddr_cmds_data[5][2] = 0x60000008; + msg->ddr_cmds_data[6][0] = 0x60001698; + msg->ddr_cmds_data[6][1] = 0x60001da8; + msg->ddr_cmds_data[6][2] = 0x60000008; + msg->ddr_cmds_data[7][0] = 0x600018d2; + msg->ddr_cmds_data[7][1] = 0x60002093; + msg->ddr_cmds_data[7][2] = 0x60000008; + msg->ddr_cmds_data[8][0] = 0x60001e66; + msg->ddr_cmds_data[8][1] = 0x600027e6; + msg->ddr_cmds_data[8][2] = 0x60000008; + msg->ddr_cmds_data[9][0] = 0x600027c2; + msg->ddr_cmds_data[9][1] = 0x6000342f; + msg->ddr_cmds_data[9][2] = 0x60000008; + msg->ddr_cmds_data[10][0] = 0x60002e71; + msg->ddr_cmds_data[10][1] = 0x60003cf5; + msg->ddr_cmds_data[10][2] = 0x60000008; + msg->ddr_cmds_data[11][0] = 0x600030ae; + msg->ddr_cmds_data[11][1] = 0x60003fe5; + msg->ddr_cmds_data[11][2] = 0x60000008; + + msg->cnoc_cmds_num = 1; + msg->cnoc_wait_bitmask = 0x1; + + msg->cnoc_cmds_addrs[0] = cmd_db_read_addr("CN0"); + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[1][0] = 0x60000001; +} + +static void a740_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + msg->bw_level_num = 1; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x7; + + msg->ddr_cmds_addrs[0] = cmd_db_read_addr("SH0"); + msg->ddr_cmds_addrs[1] = cmd_db_read_addr("MC0"); + msg->ddr_cmds_addrs[2] = cmd_db_read_addr("ACV"); + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; + + /* TODO: add a proper dvfs table */ + + msg->cnoc_cmds_num = 1; + msg->cnoc_wait_bitmask = 0x1; + + msg->cnoc_cmds_addrs[0] = cmd_db_read_addr("CN0"); + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[1][0] = 0x60000001; +} + +static void a6xx_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + /* Send a single "off" entry since the 630 GMU doesn't do bus scaling */ + msg->bw_level_num = 1; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x07; + + msg->ddr_cmds_addrs[0] = 0x50000; + msg->ddr_cmds_addrs[1] = 0x5005c; + msg->ddr_cmds_addrs[2] = 0x5000c; + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; + + /* + * These are the CX (CNOC) votes. This is used but the values for the + * sdm845 GMU are known and fixed so we can hard code them. + */ + + msg->cnoc_cmds_num = 3; + msg->cnoc_wait_bitmask = 0x05; + + msg->cnoc_cmds_addrs[0] = 0x50034; + msg->cnoc_cmds_addrs[1] = 0x5007c; + msg->cnoc_cmds_addrs[2] = 0x5004c; + + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[0][1] = 0x00000000; + msg->cnoc_cmds_data[0][2] = 0x40000000; + + msg->cnoc_cmds_data[1][0] = 0x60000001; + msg->cnoc_cmds_data[1][1] = 0x20000001; + msg->cnoc_cmds_data[1][2] = 0x60000001; +} + + +static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu) +{ + struct a6xx_hfi_msg_bw_table *msg; + struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + const struct a6xx_info *info = adreno_gpu->info->a6xx; + + if (gmu->bw_table) + goto send; + + msg = devm_kzalloc(gmu->dev, sizeof(*msg), GFP_KERNEL); + if (!msg) + return -ENOMEM; + + if (info->bcms && gmu->nr_gpu_bws > 1) + a6xx_generate_bw_table(info, gmu, msg); + else if (adreno_is_a618(adreno_gpu)) + a618_build_bw_table(msg); + else if (adreno_is_a619(adreno_gpu)) + a619_build_bw_table(msg); + else if (adreno_is_a640_family(adreno_gpu)) + a640_build_bw_table(msg); + else if (adreno_is_a650(adreno_gpu)) + a650_build_bw_table(msg); + else if (adreno_is_7c3(adreno_gpu)) + adreno_7c3_build_bw_table(msg); + else if (adreno_is_a660(adreno_gpu)) + a660_build_bw_table(msg); + else if (adreno_is_a663(adreno_gpu)) + a663_build_bw_table(msg); + else if (adreno_is_a690(adreno_gpu)) + a690_build_bw_table(msg); + else if (adreno_is_a730(adreno_gpu)) + a730_build_bw_table(msg); + else if (adreno_is_a740_family(adreno_gpu)) + a740_build_bw_table(msg); + else + a6xx_build_bw_table(msg); + + gmu->bw_table = msg; + +send: + return a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_BW_TABLE, gmu->bw_table, sizeof(*(gmu->bw_table)), + NULL, 0); +} + +static int a6xx_hfi_feature_ctrl_msg(struct a6xx_gmu *gmu, u32 feature, u32 enable, u32 data) +{ + struct a6xx_hfi_msg_feature_ctrl msg = { + .feature = feature, + .enable = enable, + .data = data, + }; + + return a6xx_hfi_send_msg(gmu, HFI_H2F_FEATURE_CTRL, &msg, sizeof(msg), NULL, 0); +} + +#define IFPC_LONG_HYST 0x1680 + +static int a6xx_hfi_enable_ifpc(struct a6xx_gmu *gmu) +{ + if (gmu->idle_level != GMU_IDLE_STATE_IFPC) + return 0; + + return a6xx_hfi_feature_ctrl_msg(gmu, HFI_FEATURE_IFPC, 1, IFPC_LONG_HYST); +} + +static int a6xx_hfi_enable_acd(struct a6xx_gmu *gmu) +{ + struct a6xx_hfi_acd_table *acd_table = &gmu->acd_table; + int ret; + + if (!acd_table->enable_by_level) + return 0; + + /* Enable ACD feature at GMU */ + ret = a6xx_hfi_feature_ctrl_msg(gmu, HFI_FEATURE_ACD, 1, 0); + if (ret) { + DRM_DEV_ERROR(gmu->dev, "Unable to enable ACD (%d)\n", ret); + return ret; + } + + /* Send ACD table to GMU */ + ret = a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_ACD, acd_table, sizeof(*acd_table), NULL, 0); + if (ret) { + DRM_DEV_ERROR(gmu->dev, "Unable to ACD table (%d)\n", ret); + return ret; + } + + return 0; +} + +static int a6xx_hfi_send_test(struct a6xx_gmu *gmu) +{ + struct a6xx_hfi_msg_test msg = { 0 }; + + return a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_TEST, &msg, sizeof(msg), + NULL, 0); +} + +static int a6xx_hfi_send_start(struct a6xx_gmu *gmu) +{ + struct a6xx_hfi_msg_start msg = { 0 }; + + return a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_START, &msg, sizeof(msg), + NULL, 0); +} + +static int a6xx_hfi_send_core_fw_start(struct a6xx_gmu *gmu) +{ + struct a6xx_hfi_msg_core_fw_start msg = { 0 }; + + return a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_CORE_FW_START, &msg, + sizeof(msg), NULL, 0); +} + +int a6xx_hfi_set_freq(struct a6xx_gmu *gmu, u32 freq_index, u32 bw_index) +{ + struct a6xx_hfi_gx_bw_perf_vote_cmd msg = { 0 }; + + msg.ack_type = 1; /* blocking */ + msg.freq = freq_index; + msg.bw = bw_index; + + return a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_GX_BW_PERF_VOTE, &msg, + sizeof(msg), NULL, 0); +} + +int a6xx_hfi_send_prep_slumber(struct a6xx_gmu *gmu) +{ + struct a6xx_hfi_prep_slumber_cmd msg = { 0 }; + + /* TODO: should freq and bw fields be non-zero ? */ + + return a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_PREPARE_SLUMBER, &msg, + sizeof(msg), NULL, 0); +} + +static int a6xx_hfi_start_v1(struct a6xx_gmu *gmu, int boot_state) +{ + int ret; + + ret = a6xx_hfi_send_gmu_init(gmu, boot_state); + if (ret) + return ret; + + ret = a6xx_hfi_get_fw_version(gmu, NULL); + if (ret) + return ret; + + /* + * We have to get exchange version numbers per the sequence but at this + * point th kernel driver doesn't need to know the exact version of + * the GMU firmware + */ + + ret = a6xx_hfi_send_perf_table_v1(gmu); + if (ret) + return ret; + + ret = a6xx_hfi_send_bw_table(gmu); + if (ret) + return ret; + + /* + * Let the GMU know that there won't be any more HFI messages until next + * boot + */ + a6xx_hfi_send_test(gmu); + + return 0; +} + +int a6xx_hfi_start(struct a6xx_gmu *gmu, int boot_state) +{ + int ret; + + if (gmu->legacy) + return a6xx_hfi_start_v1(gmu, boot_state); + + + ret = a6xx_hfi_send_perf_table(gmu); + if (ret) + return ret; + + ret = a6xx_hfi_send_bw_table(gmu); + if (ret) + return ret; + + ret = a6xx_hfi_enable_acd(gmu); + if (ret) + return ret; + + ret = a6xx_hfi_enable_ifpc(gmu); + if (ret) + return ret; + + ret = a6xx_hfi_send_core_fw_start(gmu); + if (ret) + return ret; + + /* + * Downstream driver sends this in its "a6xx_hw_init" equivalent, + * but seems to be no harm in sending it here + */ + ret = a6xx_hfi_send_start(gmu); + if (ret) + return ret; + + return 0; +} + +void a6xx_hfi_stop(struct a6xx_gmu *gmu) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(gmu->queues); i++) { + struct a6xx_hfi_queue *queue = &gmu->queues[i]; + + if (!queue->header) + continue; + + if (queue->header->read_index != queue->header->write_index) + DRM_DEV_ERROR(gmu->dev, "HFI queue %d is not empty\n", i); + + queue->header->read_index = 0; + queue->header->write_index = 0; + + memset(&queue->history, 0xff, sizeof(queue->history)); + queue->history_idx = 0; + } +} + +static void a6xx_hfi_queue_init(struct a6xx_hfi_queue *queue, + struct a6xx_hfi_queue_header *header, void *virt, u64 iova, + u32 id) +{ + spin_lock_init(&queue->lock); + queue->header = header; + queue->data = virt; + atomic_set(&queue->seqnum, 0); + + memset(&queue->history, 0xff, sizeof(queue->history)); + queue->history_idx = 0; + + /* Set up the shared memory header */ + header->iova = iova; + header->type = 10 << 8 | id; + header->status = 1; + header->size = SZ_4K >> 2; + header->msg_size = 0; + header->dropped = 0; + header->rx_watermark = 1; + header->tx_watermark = 1; + header->rx_request = 1; + header->tx_request = 0; + header->read_index = 0; + header->write_index = 0; +} + +void a6xx_hfi_init(struct a6xx_gmu *gmu) +{ + struct a6xx_gmu_bo *hfi = &gmu->hfi; + struct a6xx_hfi_queue_table_header *table = hfi->virt; + struct a6xx_hfi_queue_header *headers = hfi->virt + sizeof(*table); + int table_size, idx; + u64 offset; + + /* + * The table size is the size of the table header plus all of the queue + * headers + */ + table_size = sizeof(*table); + table_size += (ARRAY_SIZE(gmu->queues) * + sizeof(struct a6xx_hfi_queue_header)); + + table->version = 0; + table->size = table_size; + /* First queue header is located immediately after the table header */ + table->qhdr0_offset = sizeof(*table) >> 2; + table->qhdr_size = sizeof(struct a6xx_hfi_queue_header) >> 2; + table->num_queues = ARRAY_SIZE(gmu->queues); + table->active_queues = ARRAY_SIZE(gmu->queues); + + /* Command queue */ + idx = 0; + offset = SZ_4K; + a6xx_hfi_queue_init(&gmu->queues[idx], &headers[idx], hfi->virt + offset, + hfi->iova + offset, 0); + + /* GMU response queue */ + idx++; + offset += SZ_4K; + a6xx_hfi_queue_init(&gmu->queues[idx], &headers[idx], hfi->virt + offset, + hfi->iova + offset, gmu->legacy ? 4 : 1); + + /* GMU Debug queue */ + idx++; + offset += SZ_4K; + a6xx_hfi_queue_init(&gmu->queues[idx], &headers[idx], hfi->virt + offset, + hfi->iova + offset, gmu->legacy ? 5 : 2); + + WARN_ON(idx >= HFI_MAX_QUEUES); +} diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.h b/drivers/gpu/drm/msm/adreno/a6xx_hfi.h new file mode 100644 index 000000000..e10d32ce9 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.h @@ -0,0 +1,338 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2017 The Linux Foundation. All rights reserved. */ + +#ifndef _A6XX_HFI_H_ +#define _A6XX_HFI_H_ + +#define HFI_MAX_QUEUES 3 + +struct a6xx_hfi_queue_table_header { + u32 version; + u32 size; /* Size of the queue table in dwords */ + u32 qhdr0_offset; /* Offset of the first queue header */ + u32 qhdr_size; /* Size of the queue headers */ + u32 num_queues; /* Number of total queues */ + u32 active_queues; /* Number of active queues */ +} __packed; + +struct a6xx_hfi_queue_header { + u32 status; + u32 iova; + u32 type; + u32 size; + u32 msg_size; + u32 dropped; + u32 rx_watermark; + u32 tx_watermark; + u32 rx_request; + u32 tx_request; + u32 read_index; + u32 write_index; +} __packed; + +struct a6xx_hfi_queue { + struct a6xx_hfi_queue_header *header; + spinlock_t lock; + u32 *data; + atomic_t seqnum; + + /* + * Tracking for the start index of the last N messages in the + * queue, for the benefit of devcore dump / crashdec (since + * parsing in the reverse direction to decode the last N + * messages is difficult to do and would rely on heuristics + * which are not guaranteed to be correct) + */ +#define HFI_HISTORY_SZ 8 + s32 history[HFI_HISTORY_SZ]; + u8 history_idx; +}; + +/* This is the outgoing queue to the GMU */ +#define HFI_COMMAND_QUEUE 0 + +/* THis is the incoming response queue from the GMU */ +#define HFI_RESPONSE_QUEUE 1 + +#define HFI_HEADER_ID(msg) ((msg) & 0xff) +#define HFI_HEADER_SIZE(msg) (((msg) >> 8) & 0xff) +#define HFI_HEADER_SEQNUM(msg) (((msg) >> 20) & 0xfff) + +/* FIXME: Do we need this or can we use ARRAY_SIZE? */ +#define HFI_RESPONSE_PAYLOAD_SIZE 16 + +/* HFI message types */ + +#define HFI_MSG_CMD 0 +#define HFI_MSG_ACK 1 +#define HFI_MSG_ACK_V1 2 + +#define HFI_F2H_MSG_ACK 126 + +struct a6xx_hfi_msg_response { + u32 header; + u32 ret_header; + u32 error; + u32 payload[HFI_RESPONSE_PAYLOAD_SIZE]; +} __packed; + +#define HFI_F2H_MSG_ERROR 100 + +struct a6xx_hfi_msg_error { + u32 header; + u32 code; + u32 payload[2]; +} __packed; + +#define HFI_H2F_MSG_INIT 0 + +struct a6xx_hfi_msg_gmu_init_cmd { + u32 header; + u32 seg_id; + u32 dbg_buffer_addr; + u32 dbg_buffer_size; + u32 boot_state; +} __packed; + +#define HFI_H2F_MSG_FW_VERSION 1 + +struct a6xx_hfi_msg_fw_version { + u32 header; + u32 supported_version; +} __packed; + +#define HFI_H2F_MSG_PERF_TABLE 4 + +struct perf_level { + u32 vote; + u32 freq; +} __packed; + +struct perf_gx_level { + u32 vote; + u32 acd; + u32 freq; +} __packed; + +struct a6xx_hfi_msg_perf_table_v1 { + u32 header; + u32 num_gpu_levels; + u32 num_gmu_levels; + + struct perf_level gx_votes[16]; + struct perf_level cx_votes[4]; +} __packed; + +struct a6xx_hfi_msg_perf_table { + u32 header; + u32 num_gpu_levels; + u32 num_gmu_levels; + + struct perf_gx_level gx_votes[16]; + struct perf_level cx_votes[4]; +} __packed; + +#define HFI_H2F_MSG_BW_TABLE 3 + +struct a6xx_hfi_msg_bw_table { + u32 header; + u32 bw_level_num; + u32 cnoc_cmds_num; + u32 ddr_cmds_num; + u32 cnoc_wait_bitmask; + u32 ddr_wait_bitmask; + u32 cnoc_cmds_addrs[6]; + u32 cnoc_cmds_data[2][6]; + u32 ddr_cmds_addrs[8]; + u32 ddr_cmds_data[16][8]; +} __packed; + +#define HFI_H2F_MSG_TEST 5 + +struct a6xx_hfi_msg_test { + u32 header; +} __packed; + +#define HFI_H2F_MSG_ACD 7 +#define MAX_ACD_STRIDE 2 + +struct a6xx_hfi_acd_table { + u32 header; + u32 version; + u32 enable_by_level; + u32 stride; + u32 num_levels; + u32 data[16 * MAX_ACD_STRIDE]; +} __packed; + +#define CLX_DATA(irated, num_phases, clx_path, extd_intf) \ + ((extd_intf << 29) | \ + (clx_path << 28) | \ + (num_phases << 22) | \ + (irated << 16)) + +struct a6xx_hfi_clx_domain_v2 { + /** + * @data: BITS[0:15] Migration time + * BITS[16:21] Current rating + * BITS[22:27] Phases for domain + * BITS[28:28] Path notification + * BITS[29:31] Extra features + */ + u32 data; + /** @clxt: CLX time in microseconds */ + u32 clxt; + /** @clxh: CLH time in microseconds */ + u32 clxh; + /** @urg_mode: Urgent HW throttle mode of operation */ + u32 urg_mode; + /** @lkg_en: Enable leakage current estimate */ + u32 lkg_en; + /** curr_budget: Current Budget */ + u32 curr_budget; +} __packed; + +#define HFI_H2F_MSG_CLX_TBL 8 + +#define MAX_CLX_DOMAINS 2 +struct a6xx_hfi_clx_table_v2_cmd { + u32 hdr; + u32 version; + struct a6xx_hfi_clx_domain_v2 domain[MAX_CLX_DOMAINS]; +} __packed; + +#define HFI_H2F_MSG_START 10 + +struct a6xx_hfi_msg_start { + u32 header; +} __packed; + +#define HFI_H2F_FEATURE_CTRL 11 + +struct a6xx_hfi_msg_feature_ctrl { + u32 header; + u32 feature; +#define HFI_FEATURE_DCVS 0 +#define HFI_FEATURE_HWSCHED 1 +#define HFI_FEATURE_PREEMPTION 2 +#define HFI_FEATURE_CLOCKS_ON 3 +#define HFI_FEATURE_BUS_ON 4 +#define HFI_FEATURE_RAIL_ON 5 +#define HFI_FEATURE_HWCG 6 +#define HFI_FEATURE_LM 7 +#define HFI_FEATURE_THROTTLE 8 +#define HFI_FEATURE_IFPC 9 +#define HFI_FEATURE_NAP 10 +#define HFI_FEATURE_BCL 11 +#define HFI_FEATURE_ACD 12 +#define HFI_FEATURE_DIDT 13 +#define HFI_FEATURE_DEPRECATED 14 +#define HFI_FEATURE_CB 15 +#define HFI_FEATURE_KPROF 16 +#define HFI_FEATURE_BAIL_OUT_TIMER 17 +#define HFI_FEATURE_GMU_STATS 18 +#define HFI_FEATURE_DBQ 19 +#define HFI_FEATURE_MINBW 20 +#define HFI_FEATURE_CLX 21 +#define HFI_FEATURE_LSR 23 +#define HFI_FEATURE_LPAC 24 +#define HFI_FEATURE_HW_FENCE 25 +#define HFI_FEATURE_PERF_NORETAIN 26 +#define HFI_FEATURE_DMS 27 +#define HFI_FEATURE_THERMAL 28 +#define HFI_FEATURE_AQE 29 +#define HFI_FEATURE_TDCVS 30 +#define HFI_FEATURE_DCE 31 +#define HFI_FEATURE_IFF_PCLX 32 +#define HFI_FEATURE_SOFT_RESET 0x10000001 +#define HFI_FEATURE_DCVS_PROFILE 0x10000002 +#define HFI_FEATURE_FAST_CTX_DESTROY 0x10000003 + u32 enable; + u32 data; +} __packed; + +#define HFI_H2F_MSG_CORE_FW_START 14 + +struct a6xx_hfi_msg_core_fw_start { + u32 header; + u32 handle; +} __packed; + +#define HFI_H2F_MSG_TABLE 15 + +struct a6xx_hfi_table_entry { + u32 count; + u32 stride; + u32 data[]; +} __packed; + +struct a6xx_hfi_table { + u32 header; + u32 version; + u32 type; +#define HFI_TABLE_BW_VOTE 0 +#define HFI_TABLE_GPU_PERF 1 +#define HFI_TABLE_DIDT 2 +#define HFI_TABLE_ACD 3 +#define HFI_TABLE_CLX_V1 4 /* Unused */ +#define HFI_TABLE_CLX_V2 5 +#define HFI_TABLE_THERM 6 +#define HFI_TABLE_DCVS 7 +#define HFI_TABLE_SYS_TIME 8 +#define HFI_TABLE_GMU_DCVS 9 +#define HFI_TABLE_LIMITS_MIT 10 + struct a6xx_hfi_table_entry entry[]; +} __packed; + +#define HFI_H2F_MSG_GX_BW_PERF_VOTE 30 + +struct a6xx_hfi_gx_bw_perf_vote_cmd { + u32 header; + u32 ack_type; + u32 freq; + u32 bw; +} __packed; + +#define AB_VOTE_MASK GENMASK(31, 16) +#define MAX_AB_VOTE (FIELD_MAX(AB_VOTE_MASK) - 1) +#define AB_VOTE(vote) FIELD_PREP(AB_VOTE_MASK, (vote)) +#define AB_VOTE_ENABLE BIT(8) + +#define HFI_H2F_MSG_PREPARE_SLUMBER 33 + +struct a6xx_hfi_prep_slumber_cmd { + u32 header; + u32 bw; + u32 freq; +} __packed; + +struct a6xx_hfi_limits_cfg { + u32 enable; + u32 msg_path; + u32 lkg_en; + /* + * BIT[0]: 0 = (static) throttle to fixed sid level + * 1 = (dynamic) throttle to sid level calculated by HW + * BIT[1]: 0 = Mx + * 1 = Bx + */ + u32 mode; + u32 sid; + /* Mitigation time in microseconds */ + u32 mit_time; + /* Max current in mA during mitigation */ + u32 curr_limit; +} __packed; + +struct a6xx_hfi_limits_tbl { + u8 feature_id; +#define GMU_MIT_IFF 0 +#define GMU_MIT_PCLX 1 + u8 domain; +#define GMU_GX_DOMAIN 0 +#define GMU_MX_DOMAIN 1 + u16 feature_rev; + struct a6xx_hfi_limits_cfg cfg; +} __packed; + +#endif diff --git a/drivers/gpu/drm/msm/adreno/a6xx_preempt.c b/drivers/gpu/drm/msm/adreno/a6xx_preempt.c new file mode 100644 index 000000000..1e599d4dd --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a6xx_preempt.c @@ -0,0 +1,402 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2018, The Linux Foundation. All rights reserved. */ +/* Copyright (c) 2023 Collabora, Ltd. */ +/* Copyright (c) 2024 Valve Corporation */ + +#include "msm_gem.h" +#include "a6xx_gpu.h" +#include "a6xx_gmu.xml.h" +#include "a6xx_preempt.h" +#include "msm_mmu.h" +#include "msm_gpu_trace.h" + +static void a6xx_preempt_timer(struct timer_list *t) +{ + struct a6xx_gpu *a6xx_gpu = timer_container_of(a6xx_gpu, t, + preempt_timer); + struct msm_gpu *gpu = &a6xx_gpu->base.base; + struct drm_device *dev = gpu->dev; + + if (!try_preempt_state(a6xx_gpu, PREEMPT_TRIGGERED, PREEMPT_FAULTED)) + return; + + dev_err(dev->dev, "%s: preemption timed out\n", gpu->name); + kthread_queue_work(gpu->worker, &gpu->recover_work); +} + +static void preempt_prepare_postamble(struct a6xx_gpu *a6xx_gpu) +{ + u32 *postamble = a6xx_gpu->preempt_postamble_ptr; + u32 count = 0; + + postamble[count++] = PKT7(CP_REG_RMW, 3); + postamble[count++] = REG_A6XX_RBBM_PERFCTR_SRAM_INIT_CMD; + postamble[count++] = 0; + postamble[count++] = 1; + + postamble[count++] = PKT7(CP_WAIT_REG_MEM, 6); + postamble[count++] = CP_WAIT_REG_MEM_0_FUNCTION(WRITE_EQ); + postamble[count++] = CP_WAIT_REG_MEM_POLL_ADDR_LO( + REG_A6XX_RBBM_PERFCTR_SRAM_INIT_STATUS); + postamble[count++] = CP_WAIT_REG_MEM_POLL_ADDR_HI(0); + postamble[count++] = CP_WAIT_REG_MEM_3_REF(0x1); + postamble[count++] = CP_WAIT_REG_MEM_4_MASK(0x1); + postamble[count++] = CP_WAIT_REG_MEM_5_DELAY_LOOP_CYCLES(0); + + a6xx_gpu->preempt_postamble_len = count; + + a6xx_gpu->postamble_enabled = true; +} + +static void preempt_disable_postamble(struct a6xx_gpu *a6xx_gpu) +{ + u32 *postamble = a6xx_gpu->preempt_postamble_ptr; + + /* + * Disable the postamble by replacing the first packet header with a NOP + * that covers the whole buffer. + */ + *postamble = PKT7(CP_NOP, (a6xx_gpu->preempt_postamble_len - 1)); + + a6xx_gpu->postamble_enabled = false; +} + +/* + * Set preemption keepalive vote. Please note that this vote is different from the one used in + * a6xx_irq() + */ +static void a6xx_preempt_keepalive_vote(struct msm_gpu *gpu, bool on) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + if (adreno_has_gmu_wrapper(adreno_gpu)) + return; + + gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_PWR_COL_PREEMPT_KEEPALIVE, on); +} + +void a6xx_preempt_irq(struct msm_gpu *gpu) +{ + uint32_t status; + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct drm_device *dev = gpu->dev; + + if (!try_preempt_state(a6xx_gpu, PREEMPT_TRIGGERED, PREEMPT_PENDING)) + return; + + /* Delete the preemption watchdog timer */ + timer_delete(&a6xx_gpu->preempt_timer); + + /* + * The hardware should be setting the stop bit of CP_CONTEXT_SWITCH_CNTL + * to zero before firing the interrupt, but there is a non zero chance + * of a hardware condition or a software race that could set it again + * before we have a chance to finish. If that happens, log and go for + * recovery + */ + status = gpu_read(gpu, REG_A6XX_CP_CONTEXT_SWITCH_CNTL); + if (unlikely(status & A6XX_CP_CONTEXT_SWITCH_CNTL_STOP)) { + DRM_DEV_ERROR(&gpu->pdev->dev, + "!!!!!!!!!!!!!!!! preemption faulted !!!!!!!!!!!!!! irq\n"); + set_preempt_state(a6xx_gpu, PREEMPT_FAULTED); + dev_err(dev->dev, "%s: Preemption failed to complete\n", + gpu->name); + kthread_queue_work(gpu->worker, &gpu->recover_work); + return; + } + + a6xx_gpu->cur_ring = a6xx_gpu->next_ring; + a6xx_gpu->next_ring = NULL; + + set_preempt_state(a6xx_gpu, PREEMPT_FINISH); + + update_wptr(a6xx_gpu, a6xx_gpu->cur_ring); + + set_preempt_state(a6xx_gpu, PREEMPT_NONE); + + a6xx_preempt_keepalive_vote(gpu, false); + + trace_msm_gpu_preemption_irq(a6xx_gpu->cur_ring->id); + + /* + * Retrigger preemption to avoid a deadlock that might occur when preemption + * is skipped due to it being already in flight when requested. + */ + a6xx_preempt_trigger(gpu); +} + +void a6xx_preempt_hw_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + int i; + + /* No preemption if we only have one ring */ + if (gpu->nr_rings == 1) + return; + + for (i = 0; i < gpu->nr_rings; i++) { + struct a6xx_preempt_record *record_ptr = a6xx_gpu->preempt[i]; + + record_ptr->wptr = 0; + record_ptr->rptr = 0; + record_ptr->rptr_addr = shadowptr(a6xx_gpu, gpu->rb[i]); + record_ptr->info = 0; + record_ptr->data = 0; + record_ptr->rbase = gpu->rb[i]->iova; + } + + /* Write a 0 to signal that we aren't switching pagetables */ + gpu_write64(gpu, REG_A6XX_CP_CONTEXT_SWITCH_SMMU_INFO, 0); + + /* Enable the GMEM save/restore feature for preemption */ + gpu_write(gpu, REG_A6XX_RB_CONTEXT_SWITCH_GMEM_SAVE_RESTORE_ENABLE, 0x1); + + /* Reset the preemption state */ + set_preempt_state(a6xx_gpu, PREEMPT_NONE); + + spin_lock_init(&a6xx_gpu->eval_lock); + + /* Always come up on rb 0 */ + a6xx_gpu->cur_ring = gpu->rb[0]; +} + +void a6xx_preempt_trigger(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + unsigned long flags; + struct msm_ringbuffer *ring; + unsigned int cntl; + bool sysprof; + + if (gpu->nr_rings == 1) + return; + + /* + * Lock to make sure another thread attempting preemption doesn't skip it + * while we are still evaluating the next ring. This makes sure the other + * thread does start preemption if we abort it and avoids a soft lock. + */ + spin_lock_irqsave(&a6xx_gpu->eval_lock, flags); + + /* + * Try to start preemption by moving from NONE to START. If + * unsuccessful, a preemption is already in flight + */ + if (!try_preempt_state(a6xx_gpu, PREEMPT_NONE, PREEMPT_START)) { + spin_unlock_irqrestore(&a6xx_gpu->eval_lock, flags); + return; + } + + cntl = A6XX_CP_CONTEXT_SWITCH_CNTL_LEVEL(a6xx_gpu->preempt_level); + + if (a6xx_gpu->skip_save_restore) + cntl |= A6XX_CP_CONTEXT_SWITCH_CNTL_SKIP_SAVE_RESTORE; + + if (a6xx_gpu->uses_gmem) + cntl |= A6XX_CP_CONTEXT_SWITCH_CNTL_USES_GMEM; + + cntl |= A6XX_CP_CONTEXT_SWITCH_CNTL_STOP; + + /* Get the next ring to preempt to */ + ring = get_next_ring(gpu); + + /* + * If no ring is populated or the highest priority ring is the current + * one do nothing except to update the wptr to the latest and greatest + */ + if (!ring || (a6xx_gpu->cur_ring == ring)) { + set_preempt_state(a6xx_gpu, PREEMPT_FINISH); + update_wptr(a6xx_gpu, a6xx_gpu->cur_ring); + set_preempt_state(a6xx_gpu, PREEMPT_NONE); + spin_unlock_irqrestore(&a6xx_gpu->eval_lock, flags); + return; + } + + spin_unlock_irqrestore(&a6xx_gpu->eval_lock, flags); + + spin_lock_irqsave(&ring->preempt_lock, flags); + + struct a7xx_cp_smmu_info *smmu_info_ptr = + a6xx_gpu->preempt_smmu[ring->id]; + struct a6xx_preempt_record *record_ptr = a6xx_gpu->preempt[ring->id]; + u64 ttbr0 = ring->memptrs->ttbr0; + u32 context_idr = ring->memptrs->context_idr; + + smmu_info_ptr->ttbr0 = ttbr0; + smmu_info_ptr->context_idr = context_idr; + record_ptr->wptr = get_wptr(ring); + + /* + * The GPU will write the wptr we set above when we preempt. Reset + * restore_wptr to make sure that we don't write WPTR to the same + * thing twice. It's still possible subsequent submissions will update + * wptr again, in which case they will set the flag to true. This has + * to be protected by the lock for setting the flag and updating wptr + * to be atomic. + */ + ring->restore_wptr = false; + + trace_msm_gpu_preemption_trigger(a6xx_gpu->cur_ring->id, ring->id); + + spin_unlock_irqrestore(&ring->preempt_lock, flags); + + /* Set the keepalive bit to keep the GPU ON until preemption is complete */ + a6xx_preempt_keepalive_vote(gpu, true); + + a6xx_fenced_write(a6xx_gpu, + REG_A6XX_CP_CONTEXT_SWITCH_SMMU_INFO, a6xx_gpu->preempt_smmu_iova[ring->id], + BIT(1), true); + + a6xx_fenced_write(a6xx_gpu, + REG_A6XX_CP_CONTEXT_SWITCH_PRIV_NON_SECURE_RESTORE_ADDR, + a6xx_gpu->preempt_iova[ring->id], BIT(1), true); + + a6xx_gpu->next_ring = ring; + + /* Start a timer to catch a stuck preemption */ + mod_timer(&a6xx_gpu->preempt_timer, jiffies + msecs_to_jiffies(10000)); + + /* Enable or disable postamble as needed */ + sysprof = msm_gpu_sysprof_no_perfcntr_zap(gpu); + + if (!sysprof && !a6xx_gpu->postamble_enabled) + preempt_prepare_postamble(a6xx_gpu); + + if (sysprof && a6xx_gpu->postamble_enabled) + preempt_disable_postamble(a6xx_gpu); + + /* Set the preemption state to triggered */ + set_preempt_state(a6xx_gpu, PREEMPT_TRIGGERED); + + /* Trigger the preemption */ + a6xx_fenced_write(a6xx_gpu, REG_A6XX_CP_CONTEXT_SWITCH_CNTL, cntl, BIT(1), false); +} + +static int preempt_init_ring(struct a6xx_gpu *a6xx_gpu, + struct msm_ringbuffer *ring) +{ + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + struct drm_gem_object *bo = NULL; + phys_addr_t ttbr; + u64 iova = 0; + void *ptr; + int asid; + + ptr = msm_gem_kernel_new(gpu->dev, + PREEMPT_RECORD_SIZE(adreno_gpu), + MSM_BO_WC | MSM_BO_MAP_PRIV, gpu->vm, &bo, &iova); + + if (IS_ERR(ptr)) + return PTR_ERR(ptr); + + memset(ptr, 0, PREEMPT_RECORD_SIZE(adreno_gpu)); + + msm_gem_object_set_name(bo, "preempt_record ring%d", ring->id); + + a6xx_gpu->preempt_bo[ring->id] = bo; + a6xx_gpu->preempt_iova[ring->id] = iova; + a6xx_gpu->preempt[ring->id] = ptr; + + struct a6xx_preempt_record *record_ptr = ptr; + + ptr = msm_gem_kernel_new(gpu->dev, + PREEMPT_SMMU_INFO_SIZE, + MSM_BO_WC | MSM_BO_MAP_PRIV | MSM_BO_GPU_READONLY, + gpu->vm, &bo, &iova); + + if (IS_ERR(ptr)) + return PTR_ERR(ptr); + + memset(ptr, 0, PREEMPT_SMMU_INFO_SIZE); + + msm_gem_object_set_name(bo, "preempt_smmu_info ring%d", ring->id); + + a6xx_gpu->preempt_smmu_bo[ring->id] = bo; + a6xx_gpu->preempt_smmu_iova[ring->id] = iova; + a6xx_gpu->preempt_smmu[ring->id] = ptr; + + struct a7xx_cp_smmu_info *smmu_info_ptr = ptr; + + msm_iommu_pagetable_params(to_msm_vm(gpu->vm)->mmu, &ttbr, &asid); + + smmu_info_ptr->magic = GEN7_CP_SMMU_INFO_MAGIC; + smmu_info_ptr->ttbr0 = ttbr; + smmu_info_ptr->asid = 0xdecafbad; + smmu_info_ptr->context_idr = 0; + + /* Set up the defaults on the preemption record */ + record_ptr->magic = A6XX_PREEMPT_RECORD_MAGIC; + record_ptr->info = 0; + record_ptr->data = 0; + record_ptr->rptr = 0; + record_ptr->wptr = 0; + record_ptr->cntl = MSM_GPU_RB_CNTL_DEFAULT; + record_ptr->rbase = ring->iova; + record_ptr->counter = 0; + record_ptr->bv_rptr_addr = rbmemptr(ring, bv_rptr); + + return 0; +} + +void a6xx_preempt_fini(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + int i; + + for (i = 0; i < gpu->nr_rings; i++) + msm_gem_kernel_put(a6xx_gpu->preempt_bo[i], gpu->vm); +} + +void a6xx_preempt_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + int i; + + /* No preemption if we only have one ring */ + if (gpu->nr_rings <= 1) + return; + + for (i = 0; i < gpu->nr_rings; i++) { + if (preempt_init_ring(a6xx_gpu, gpu->rb[i])) + goto fail; + } + + /* TODO: make this configurable? */ + a6xx_gpu->preempt_level = 1; + a6xx_gpu->uses_gmem = 1; + a6xx_gpu->skip_save_restore = 1; + + a6xx_gpu->preempt_postamble_ptr = msm_gem_kernel_new(gpu->dev, + PAGE_SIZE, + MSM_BO_WC | MSM_BO_MAP_PRIV | MSM_BO_GPU_READONLY, + gpu->vm, &a6xx_gpu->preempt_postamble_bo, + &a6xx_gpu->preempt_postamble_iova); + + if (IS_ERR(a6xx_gpu->preempt_postamble_ptr)) + goto fail; + + preempt_prepare_postamble(a6xx_gpu); + + timer_setup(&a6xx_gpu->preempt_timer, a6xx_preempt_timer, 0); + + return; +fail: + /* + * On any failure our adventure is over. Clean up and + * set nr_rings to 1 to force preemption off + */ + a6xx_preempt_fini(gpu); + gpu->nr_rings = 1; + + DRM_DEV_ERROR(&gpu->pdev->dev, + "preemption init failed, disabling preemption\n"); + + return; +} diff --git a/drivers/gpu/drm/msm/adreno/a6xx_preempt.h b/drivers/gpu/drm/msm/adreno/a6xx_preempt.h new file mode 100644 index 000000000..df36c945b --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a6xx_preempt.h @@ -0,0 +1,82 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2018, The Linux Foundation. All rights reserved. */ +/* Copyright (c) 2023 Collabora, Ltd. */ +/* Copyright (c) 2024 Valve Corporation */ +/* Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ + +/* + * Try to transition the preemption state from old to new. Return + * true on success or false if the original state wasn't 'old' + */ +static inline bool try_preempt_state(struct a6xx_gpu *a6xx_gpu, + enum a6xx_preempt_state old, enum a6xx_preempt_state new) +{ + enum a6xx_preempt_state cur = atomic_cmpxchg(&a6xx_gpu->preempt_state, + old, new); + + return (cur == old); +} + +/* + * Force the preemption state to the specified state. This is used in cases + * where the current state is known and won't change + */ +static inline void set_preempt_state(struct a6xx_gpu *gpu, + enum a6xx_preempt_state new) +{ + /* + * preempt_state may be read by other cores trying to trigger a + * preemption or in the interrupt handler so barriers are needed + * before... + */ + smp_mb__before_atomic(); + atomic_set(&gpu->preempt_state, new); + /* ... and after */ + smp_mb__after_atomic(); +} + +/* Write the most recent wptr for the given ring into the hardware */ +static inline void update_wptr(struct a6xx_gpu *a6xx_gpu, struct msm_ringbuffer *ring) +{ + unsigned long flags; + uint32_t wptr; + + spin_lock_irqsave(&ring->preempt_lock, flags); + + if (ring->restore_wptr) { + wptr = get_wptr(ring); + + a6xx_fenced_write(a6xx_gpu, REG_A6XX_CP_RB_WPTR, wptr, BIT(0), false); + + ring->restore_wptr = false; + } + + spin_unlock_irqrestore(&ring->preempt_lock, flags); +} + +/* Return the highest priority ringbuffer with something in it */ +static inline struct msm_ringbuffer *get_next_ring(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + unsigned long flags; + int i; + + for (i = 0; i < gpu->nr_rings; i++) { + bool empty; + struct msm_ringbuffer *ring = gpu->rb[i]; + + spin_lock_irqsave(&ring->preempt_lock, flags); + empty = (get_wptr(ring) == gpu->funcs->get_rptr(gpu, ring)); + if (!empty && ring == a6xx_gpu->cur_ring) + empty = ring->memptrs->fence == a6xx_gpu->last_seqno[i]; + spin_unlock_irqrestore(&ring->preempt_lock, flags); + + if (!empty) + return ring; + } + + return NULL; +} + diff --git a/drivers/gpu/drm/msm/adreno/a8xx_gpu.c b/drivers/gpu/drm/msm/adreno/a8xx_gpu.c new file mode 100644 index 000000000..add93c1fa --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a8xx_gpu.c @@ -0,0 +1,1372 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ + + +#include "msm_gem.h" +#include "msm_mmu.h" +#include "msm_gpu_trace.h" +#include "a6xx_gpu.h" +#include "a6xx_gmu.xml.h" + +#include +#include +#include +#include +#include + +#define GPU_PAS_ID 13 + +static void a8xx_aperture_slice_set(struct msm_gpu *gpu, enum adreno_pipe pipe, u32 slice) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + u32 val; + + val = A8XX_CP_APERTURE_CNTL_HOST_PIPEID(pipe) | A8XX_CP_APERTURE_CNTL_HOST_SLICEID(slice); + + if (a6xx_gpu->cached_aperture == val) + return; + + gpu_write(gpu, REG_A8XX_CP_APERTURE_CNTL_HOST, val); + + a6xx_gpu->cached_aperture = val; +} + +static void a8xx_aperture_acquire(struct msm_gpu *gpu, enum adreno_pipe pipe, unsigned long *flags) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + spin_lock_irqsave(&a6xx_gpu->aperture_lock, *flags); + + a8xx_aperture_slice_set(gpu, pipe, 0); +} + +static void a8xx_aperture_release(struct msm_gpu *gpu, unsigned long flags) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + spin_unlock_irqrestore(&a6xx_gpu->aperture_lock, flags); +} + +static void a8xx_aperture_clear(struct msm_gpu *gpu) +{ + unsigned long flags; + + a8xx_aperture_acquire(gpu, PIPE_NONE, &flags); + a8xx_aperture_release(gpu, flags); +} + +static void a8xx_write_pipe(struct msm_gpu *gpu, enum adreno_pipe pipe, u32 offset, u32 data) +{ + unsigned long flags; + + a8xx_aperture_acquire(gpu, pipe, &flags); + gpu_write(gpu, offset, data); + a8xx_aperture_release(gpu, flags); +} + +static u32 a8xx_read_pipe_slice(struct msm_gpu *gpu, enum adreno_pipe pipe, u32 slice, u32 offset) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + unsigned long flags; + u32 val; + + spin_lock_irqsave(&a6xx_gpu->aperture_lock, flags); + a8xx_aperture_slice_set(gpu, pipe, slice); + val = gpu_read(gpu, offset); + spin_unlock_irqrestore(&a6xx_gpu->aperture_lock, flags); + + return val; +} + +void a8xx_gpu_get_slice_info(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + const struct a6xx_info *info = adreno_gpu->info->a6xx; + struct device *dev = &gpu->pdev->dev; + u32 slice_mask; + + if (adreno_gpu->info->family < ADRENO_8XX_GEN1) + return; + + if (a6xx_gpu->slice_mask) + return; + + slice_mask = GENMASK(info->max_slices - 1, 0); + + /* GEN1 doesn't support partial slice configurations */ + if (adreno_gpu->info->family == ADRENO_8XX_GEN1) { + a6xx_gpu->slice_mask = slice_mask; + return; + } + + slice_mask &= a6xx_cx_misc_read(a6xx_gpu, + REG_A8XX_CX_MISC_SLICE_ENABLE_FINAL); + + a6xx_gpu->slice_mask = slice_mask; + + /* Chip ID depends on the number of slices available. So update it */ + adreno_gpu->chip_id |= FIELD_PREP(GENMASK(7, 4), hweight32(slice_mask)); + + /* Update the gpu-name to reflect the slice config: */ + const char *name = devm_kasprintf(dev, GFP_KERNEL, + "%"ADRENO_CHIPID_FMT, + ADRENO_CHIPID_ARGS(adreno_gpu->chip_id)); + if (name) { + devm_kfree(dev, adreno_gpu->base.name); + adreno_gpu->base.name = name; + } +} + +static u32 a8xx_get_first_slice(struct a6xx_gpu *a6xx_gpu) +{ + return ffs(a6xx_gpu->slice_mask) - 1; +} + +static inline bool _a8xx_check_idle(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + /* Check that the GMU is idle */ + if (!a6xx_gmu_isidle(&a6xx_gpu->gmu)) + return false; + + /* Check that the CX master is idle */ + if (gpu_read(gpu, REG_A8XX_RBBM_STATUS) & + ~A8XX_RBBM_STATUS_CP_AHB_BUSY_CX_MASTER) + return false; + + return !(gpu_read(gpu, REG_A8XX_RBBM_INT_0_STATUS) & + A6XX_RBBM_INT_0_MASK_RBBM_HANG_DETECT); +} + +static bool a8xx_idle(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + /* wait for CP to drain ringbuffer: */ + if (!adreno_idle(gpu, ring)) + return false; + + if (spin_until(_a8xx_check_idle(gpu))) { + DRM_ERROR( + "%s: %ps: timeout waiting for GPU to idle: status %8.8X irq %8.8X rptr/wptr %d/%d\n", + gpu->name, __builtin_return_address(0), + gpu_read(gpu, REG_A8XX_RBBM_STATUS), + gpu_read(gpu, REG_A8XX_RBBM_INT_0_STATUS), + gpu_read(gpu, REG_A6XX_CP_RB_RPTR), + gpu_read(gpu, REG_A6XX_CP_RB_WPTR)); + return false; + } + + return true; +} + +void a8xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + uint32_t wptr; + unsigned long flags; + + spin_lock_irqsave(&ring->preempt_lock, flags); + + /* Copy the shadow to the actual register */ + ring->cur = ring->next; + + /* Make sure to wrap wptr if we need to */ + wptr = get_wptr(ring); + + /* Update HW if this is the current ring and we are not in preempt*/ + if (!a6xx_in_preempt(a6xx_gpu)) { + if (a6xx_gpu->cur_ring == ring) + a6xx_fenced_write(a6xx_gpu, REG_A6XX_CP_RB_WPTR, wptr, BIT(0), false); + else + ring->restore_wptr = true; + } else { + ring->restore_wptr = true; + } + + spin_unlock_irqrestore(&ring->preempt_lock, flags); +} + +static void a8xx_set_hwcg(struct msm_gpu *gpu, bool state) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + u32 val; + + if (adreno_is_x285(adreno_gpu) && state) + gpu_write(gpu, REG_A8XX_RBBM_CGC_0_PC, 0x00000702); + + gmu_write(gmu, REG_A6XX_GPU_GMU_AO_GMU_CGC_MODE_CNTL, + state ? adreno_gpu->info->a6xx->gmu_cgc_mode : 0); + gmu_write(gmu, REG_A6XX_GPU_GMU_AO_GMU_CGC_DELAY_CNTL, + state ? 0x110111 : 0); + gmu_write(gmu, REG_A6XX_GPU_GMU_AO_GMU_CGC_HYST_CNTL, + state ? 0x55555 : 0); + + gpu_write(gpu, REG_A8XX_RBBM_CLOCK_CNTL_GLOBAL, 1); + gpu_write(gpu, REG_A8XX_RBBM_CGC_GLOBAL_LOAD_CMD, !!state); + + if (state) { + gpu_write(gpu, REG_A8XX_RBBM_CGC_P2S_TRIG_CMD, 1); + + if (gpu_poll_timeout(gpu, REG_A8XX_RBBM_CGC_P2S_STATUS, val, + val & A8XX_RBBM_CGC_P2S_STATUS_TXDONE, 1, 10)) { + dev_err(&gpu->pdev->dev, "RBBM_CGC_P2S_STATUS TXDONE Poll failed\n"); + return; + } + + gpu_write(gpu, REG_A8XX_RBBM_CLOCK_CNTL_GLOBAL, 0); + } else { + /* + * GMU enables clk gating in GBIF during boot up. So, + * override that here when hwcg feature is disabled + */ + gpu_rmw(gpu, REG_A6XX_GBIF_CX_CONFIG, BIT(0), 0); + } +} + +static void a8xx_set_cp_protect(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + const struct adreno_protect *protect = adreno_gpu->info->a6xx->protect; + u32 cntl, final_cfg; + unsigned int i; + + cntl = A8XX_CP_PROTECT_CNTL_PIPE_ACCESS_PROT_EN | + A8XX_CP_PROTECT_CNTL_PIPE_ACCESS_FAULT_ON_VIOL_EN | + A8XX_CP_PROTECT_CNTL_PIPE_LAST_SPAN_INF_RANGE | + A8XX_CP_PROTECT_CNTL_PIPE_HALT_SQE_RANGE__MASK; + /* + * Enable access protection to privileged registers, fault on an access + * protect violation and select the last span to protect from the start + * address all the way to the end of the register address space + */ + a8xx_write_pipe(gpu, PIPE_BR, REG_A8XX_CP_PROTECT_CNTL_PIPE, cntl); + a8xx_write_pipe(gpu, PIPE_BV, REG_A8XX_CP_PROTECT_CNTL_PIPE, cntl); + + a8xx_aperture_clear(gpu); + + for (i = 0; i < protect->count; i++) { + /* Intentionally skip writing to some registers */ + if (protect->regs[i]) { + gpu_write(gpu, REG_A8XX_CP_PROTECT_GLOBAL(i), protect->regs[i]); + final_cfg = protect->regs[i]; + } + } + + /* + * Last span feature is only supported on PIPE specific register. + * So update those here + */ + a8xx_write_pipe(gpu, PIPE_BR, REG_A8XX_CP_PROTECT_PIPE(15), final_cfg); + a8xx_write_pipe(gpu, PIPE_BV, REG_A8XX_CP_PROTECT_PIPE(15), final_cfg); + + a8xx_aperture_clear(gpu); +} + +static void a8xx_set_ubwc_config(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + const struct qcom_ubwc_cfg_data *cfg = adreno_gpu->ubwc_config; + u32 level2_swizzling_dis = !(qcom_ubwc_swizzle(cfg) & UBWC_SWIZZLE_ENABLE_LVL2); + u32 level3_swizzling_dis = !(qcom_ubwc_swizzle(cfg) & UBWC_SWIZZLE_ENABLE_LVL3); + bool rgba8888_lossless = false, fp16compoptdis = false; + bool yuvnotcomptofc = false, min_acc_len_64b = false; + bool rgb565_predicator = false; + bool amsbc = qcom_ubwc_enable_amsbc(cfg); + bool ubwc_mode = qcom_ubwc_get_ubwc_mode(cfg); + u32 ubwc_version = cfg->ubwc_enc_version; + u32 hbb, hbb_hi, hbb_lo, mode; + u8 uavflagprd_inv = 2; + + if (ubwc_version > UBWC_6_0) + dev_err(&gpu->pdev->dev, "Unknown UBWC version: 0x%x\n", ubwc_version); + + if (ubwc_version == UBWC_6_0) + yuvnotcomptofc = true; + + if (ubwc_version < UBWC_5_0 && + ubwc_version >= UBWC_4_0) + rgba8888_lossless = true; + + if (ubwc_version < UBWC_4_3) + fp16compoptdis = true; + + if (cfg->ubwc_enc_version >= UBWC_4_0) + rgb565_predicator = true; + + if (ubwc_version < UBWC_3_0) + dev_err(&gpu->pdev->dev, "Unsupported UBWC version: 0x%x\n", ubwc_version); + + mode = qcom_ubwc_version_tag(cfg); + + /* + * We subtract 13 from the highest bank bit (13 is the minimum value + * allowed by hw) and write the lowest two bits of the remaining value + * as hbb_lo and the one above it as hbb_hi to the hardware. + */ + WARN_ON(cfg->highest_bank_bit < 13); + hbb = cfg->highest_bank_bit - 13; + hbb_hi = hbb >> 2; + hbb_lo = hbb & 3; + + a8xx_write_pipe(gpu, PIPE_BV, REG_A8XX_GRAS_NC_MODE_CNTL, + hbb << 5 | + level3_swizzling_dis << 4 | + level2_swizzling_dis << 3); + + a8xx_write_pipe(gpu, PIPE_BR, REG_A8XX_GRAS_NC_MODE_CNTL, + hbb << 5 | + level3_swizzling_dis << 4 | + level2_swizzling_dis << 3); + + a8xx_write_pipe(gpu, PIPE_BR, REG_A8XX_RB_CCU_NC_MODE_CNTL, + yuvnotcomptofc << 6 | + level3_swizzling_dis << 5 | + level2_swizzling_dis << 4 | + hbb_hi << 3 | + hbb_lo << 1); + + a8xx_write_pipe(gpu, PIPE_BR, REG_A8XX_RB_CMP_NC_MODE_CNTL, + mode << 15 | + yuvnotcomptofc << 6 | + rgba8888_lossless << 4 | + fp16compoptdis << 3 | + rgb565_predicator << 2 | + amsbc << 1 | + min_acc_len_64b); + + a8xx_aperture_clear(gpu); + + gpu_write(gpu, REG_A6XX_SP_NC_MODE_CNTL, + level3_swizzling_dis << 13 | + level2_swizzling_dis << 12 | + hbb_hi << 10 | + uavflagprd_inv << 4 | + min_acc_len_64b << 3 | + hbb_lo << 1 | ubwc_mode); + + gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, + level3_swizzling_dis << 7 | + level2_swizzling_dis << 6 | + hbb_hi << 4 | + min_acc_len_64b << 3 | + hbb_lo << 1 | ubwc_mode); +} + +static void a8xx_nonctxt_config(struct msm_gpu *gpu, u32 *gmem_protect) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + const struct a6xx_info *info = adreno_gpu->info->a6xx; + const struct adreno_reglist_pipe *regs = info->nonctxt_reglist; + unsigned int pipe_id, i; + unsigned long flags; + + for (pipe_id = PIPE_NONE; pipe_id <= PIPE_DDE_BV; pipe_id++) { + /* We don't have support for LPAC yet */ + if (pipe_id == PIPE_LPAC) + continue; + + a8xx_aperture_acquire(gpu, pipe_id, &flags); + + for (i = 0; regs[i].offset; i++) { + if (!(BIT(pipe_id) & regs[i].pipe)) + continue; + + if (regs[i].offset == REG_A8XX_RB_GC_GMEM_PROTECT) + *gmem_protect = regs[i].value; + + gpu_write(gpu, regs[i].offset, regs[i].value); + } + + a8xx_aperture_release(gpu, flags); + } + + a8xx_aperture_clear(gpu); +} + +static void a8xx_patch_pwrup_reglist(struct msm_gpu *gpu) +{ + const struct adreno_reglist_pipe_list *dyn_pwrup_reglist; + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + const struct adreno_reglist_list *reglist; + void *ptr = a6xx_gpu->pwrup_reglist_ptr; + struct cpu_gpu_lock *lock = ptr; + u32 *dest = (u32 *)&lock->regs[0]; + u32 dyn_pwrup_reglist_count = 0; + int i; + + lock->gpu_req = lock->cpu_req = lock->turn = 0; + + reglist = adreno_gpu->info->a6xx->ifpc_reglist; + if (reglist) { + lock->ifpc_list_len = reglist->count; + + /* + * For each entry in each of the lists, write the offset and the current + * register value into the GPU buffer + */ + for (i = 0; i < reglist->count; i++) { + *dest++ = reglist->regs[i]; + *dest++ = gpu_read(gpu, reglist->regs[i]); + } + } + + reglist = adreno_gpu->info->a6xx->pwrup_reglist; + if (reglist) { + lock->preemption_list_len = reglist->count; + + for (i = 0; i < reglist->count; i++) { + *dest++ = reglist->regs[i]; + *dest++ = gpu_read(gpu, reglist->regs[i]); + } + } + + /* + * The overall register list is composed of + * 1. Static IFPC-only registers + * 2. Static IFPC + preemption registers + * 3. Dynamic IFPC + preemption registers (ex: perfcounter selects) + * + * The first two lists are static. Size of these lists are stored as + * number of pairs in ifpc_list_len and preemption_list_len + * respectively. With concurrent binning, Some of the perfcounter + * registers being virtualized, CP needs to know the pipe id to program + * the aperture inorder to restore the same. Thus, third list is a + * dynamic list with triplets as + * (
), and the length is + * stored as number for triplets in dynamic_list_len. + */ + dyn_pwrup_reglist = adreno_gpu->info->a6xx->dyn_pwrup_reglist; + if (!dyn_pwrup_reglist) + goto done; + + for (u32 pipe_id = PIPE_BR; pipe_id <= PIPE_DDE_BV; pipe_id++) { + for (i = 0; i < dyn_pwrup_reglist->count; i++) { + if (!(dyn_pwrup_reglist->regs[i].pipe & BIT(pipe_id))) + continue; + *dest++ = A8XX_CP_APERTURE_CNTL_HOST_PIPEID(pipe_id); + *dest++ = dyn_pwrup_reglist->regs[i].offset; + *dest++ = a8xx_read_pipe_slice(gpu, + pipe_id, + a8xx_get_first_slice(a6xx_gpu), + dyn_pwrup_reglist->regs[i].offset); + dyn_pwrup_reglist_count++; + } + } + + lock->dynamic_list_len = dyn_pwrup_reglist_count; + a6xx_gpu->dynamic_sel_reglist_offset = dyn_pwrup_reglist_count; + +done: + a8xx_aperture_clear(gpu); +} + +static int a8xx_preempt_start(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct msm_ringbuffer *ring = gpu->rb[0]; + + if (gpu->nr_rings <= 1) + return 0; + + /* Turn CP protection off */ + OUT_PKT7(ring, CP_SET_PROTECTED_MODE, 1); + OUT_RING(ring, 0); + + a6xx_emit_set_pseudo_reg(ring, a6xx_gpu, NULL); + + a6xx_flush_yield(gpu, ring); + + return a8xx_idle(gpu, ring) ? 0 : -EINVAL; +} + +static int a8xx_cp_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct msm_ringbuffer *ring = gpu->rb[0]; + u32 mask; + + /* Disable concurrent binning before sending CP init */ + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, BIT(27)); + + OUT_PKT7(ring, CP_ME_INIT, 7); + + /* Use multiple HW contexts */ + mask = BIT(0); + + /* Enable error detection */ + mask |= BIT(1); + + /* Set default reset state */ + mask |= BIT(3); + + /* Disable save/restore of performance counters across preemption */ + mask |= BIT(6); + + /* Enable the register init list with the spinlock */ + mask |= BIT(8); + + OUT_RING(ring, mask); + + /* Enable multiple hardware contexts */ + OUT_RING(ring, 0x00000003); + + /* Enable error detection */ + OUT_RING(ring, 0x20000000); + + /* Operation mode mask */ + OUT_RING(ring, 0x00000002); + + /* Lo address */ + OUT_RING(ring, lower_32_bits(a6xx_gpu->pwrup_reglist_iova)); + /* Hi address */ + OUT_RING(ring, upper_32_bits(a6xx_gpu->pwrup_reglist_iova)); + + /* Enable dyn pwrup list with triplets (offset, value, pipe) */ + OUT_RING(ring, BIT(31)); + + a6xx_flush(gpu, ring); + return a8xx_idle(gpu, ring) ? 0 : -EINVAL; +} + +#define A8XX_INT_MASK \ + (A6XX_RBBM_INT_0_MASK_CP_AHB_ERROR | \ + A6XX_RBBM_INT_0_MASK_RBBM_ATB_ASYNCFIFO_OVERFLOW | \ + A6XX_RBBM_INT_0_MASK_RBBM_GPC_ERROR | \ + A6XX_RBBM_INT_0_MASK_CP_SW | \ + A6XX_RBBM_INT_0_MASK_CP_HW_ERROR | \ + A6XX_RBBM_INT_0_MASK_PM4CPINTERRUPT | \ + A6XX_RBBM_INT_0_MASK_CP_RB_DONE_TS | \ + A6XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS | \ + A6XX_RBBM_INT_0_MASK_RBBM_ATB_BUS_OVERFLOW | \ + A6XX_RBBM_INT_0_MASK_RBBM_HANG_DETECT | \ + A6XX_RBBM_INT_0_MASK_UCHE_OOB_ACCESS | \ + A6XX_RBBM_INT_0_MASK_UCHE_TRAP_INTR | \ + A6XX_RBBM_INT_0_MASK_TSBWRITEERROR | \ + A6XX_RBBM_INT_0_MASK_SWFUSEVIOLATION) + +#define A8XX_APRIV_MASK \ + (A8XX_CP_APRIV_CNTL_PIPE_ICACHE | \ + A8XX_CP_APRIV_CNTL_PIPE_RBFETCH | \ + A8XX_CP_APRIV_CNTL_PIPE_RBPRIVLEVEL | \ + A8XX_CP_APRIV_CNTL_PIPE_RBRPWB) + +#define A8XX_BR_APRIV_MASK \ + (A8XX_APRIV_MASK | \ + A8XX_CP_APRIV_CNTL_PIPE_CDREAD | \ + A8XX_CP_APRIV_CNTL_PIPE_CDWRITE) + +#define A8XX_CP_GLOBAL_INT_MASK \ + (A8XX_CP_GLOBAL_INT_MASK_HWFAULTBR | \ + A8XX_CP_GLOBAL_INT_MASK_HWFAULTBV | \ + A8XX_CP_GLOBAL_INT_MASK_HWFAULTLPAC | \ + A8XX_CP_GLOBAL_INT_MASK_HWFAULTAQE0 | \ + A8XX_CP_GLOBAL_INT_MASK_HWFAULTAQE1 | \ + A8XX_CP_GLOBAL_INT_MASK_HWFAULTDDEBR | \ + A8XX_CP_GLOBAL_INT_MASK_HWFAULTDDEBV | \ + A8XX_CP_GLOBAL_INT_MASK_SWFAULTBR | \ + A8XX_CP_GLOBAL_INT_MASK_SWFAULTBV | \ + A8XX_CP_GLOBAL_INT_MASK_SWFAULTLPAC | \ + A8XX_CP_GLOBAL_INT_MASK_SWFAULTAQE0 | \ + A8XX_CP_GLOBAL_INT_MASK_SWFAULTAQE1 | \ + A8XX_CP_GLOBAL_INT_MASK_SWFAULTDDEBR | \ + A8XX_CP_GLOBAL_INT_MASK_SWFAULTDDEBV) + +#define A8XX_CP_INTERRUPT_STATUS_MASK_PIPE \ + (A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_CSFRBWRAP | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_CSFIB1WRAP | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_CSFIB2WRAP | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_CSFIB3WRAP | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_CSFSDSWRAP | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_CSFMRBWRAP | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_CSFVSDWRAP | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_OPCODEERROR | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_VSDPARITYERROR | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_REGISTERPROTECTIONERROR | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_ILLEGALINSTRUCTION | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_SMMUFAULT | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_VBIFRESPCLIENT| \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_VBIFRESPTYPE | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_VBIFRESPREAD | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_VBIFRESP | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_RTWROVF | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_LRZRTWROVF | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_LRZRTREFCNTOVF | \ + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE_LRZRTCLRRESMISS) + +#define A8XX_CP_HW_FAULT_STATUS_MASK_PIPE \ + (A8XX_CP_HW_FAULT_STATUS_MASK_PIPE_CSFRBFAULT | \ + A8XX_CP_HW_FAULT_STATUS_MASK_PIPE_CSFIB1FAULT | \ + A8XX_CP_HW_FAULT_STATUS_MASK_PIPE_CSFIB2FAULT | \ + A8XX_CP_HW_FAULT_STATUS_MASK_PIPE_CSFIB3FAULT | \ + A8XX_CP_HW_FAULT_STATUS_MASK_PIPE_CSFSDSFAULT | \ + A8XX_CP_HW_FAULT_STATUS_MASK_PIPE_CSFMRBFAULT | \ + A8XX_CP_HW_FAULT_STATUS_MASK_PIPE_CSFVSDFAULT | \ + A8XX_CP_HW_FAULT_STATUS_MASK_PIPE_SQEREADBURSTOVF | \ + A8XX_CP_HW_FAULT_STATUS_MASK_PIPE_EVENTENGINEOVF | \ + A8XX_CP_HW_FAULT_STATUS_MASK_PIPE_UCODEERROR) + +static int hw_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + unsigned int pipe_id, i; + u32 gmem_protect = 0; + u64 gmem_range_min; + int ret; + + ret = a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET); + if (ret) + return ret; + + /* Clear the cached value to force aperture configuration next time */ + a6xx_gpu->cached_aperture = UINT_MAX; + a8xx_aperture_clear(gpu); + + /* Clear GBIF halt in case GX domain was not collapsed */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, 0); + gpu_read(gpu, REG_A6XX_GBIF_HALT); + + gpu_write(gpu, REG_A8XX_RBBM_GBIF_HALT, 0); + gpu_read(gpu, REG_A8XX_RBBM_GBIF_HALT); + + gpu_write(gpu, REG_A6XX_RBBM_SECVID_TSB_CNTL, 0); + + /* + * Disable the trusted memory range - we don't actually supported secure + * memory rendering at this point in time and we don't want to block off + * part of the virtual memory space. + */ + gpu_write64(gpu, REG_A6XX_RBBM_SECVID_TSB_TRUSTED_BASE, 0x00000000); + gpu_write(gpu, REG_A6XX_RBBM_SECVID_TSB_TRUSTED_SIZE, 0x00000000); + + /* Make all blocks contribute to the GPU BUSY perf counter */ + gpu_write(gpu, REG_A8XX_RBBM_PERFCTR_GPU_BUSY_MASKED, 0xffffffff); + + /* Setup GMEM Range in UCHE */ + gmem_range_min = SZ_64M; + /* Set the GMEM VA range [0x100000:0x100000 + gpu->gmem - 1] */ + gpu_write64(gpu, REG_A8XX_UCHE_CCHE_GC_GMEM_RANGE_MIN, gmem_range_min); + gpu_write64(gpu, REG_A8XX_SP_HLSQ_GC_GMEM_RANGE_MIN, gmem_range_min); + + /* Setup UCHE Trap region */ + gpu_write64(gpu, REG_A8XX_UCHE_TRAP_BASE, adreno_gpu->uche_trap_base); + gpu_write64(gpu, REG_A8XX_UCHE_WRITE_THRU_BASE, adreno_gpu->uche_trap_base); + gpu_write64(gpu, REG_A8XX_UCHE_CCHE_TRAP_BASE, adreno_gpu->uche_trap_base); + gpu_write64(gpu, REG_A8XX_UCHE_CCHE_WRITE_THRU_BASE, adreno_gpu->uche_trap_base); + + /* Turn on performance counters */ + gpu_write(gpu, REG_A8XX_RBBM_PERFCTR_CNTL, 0x1); + gpu_write(gpu, REG_A8XX_RBBM_SLICE_PERFCTR_CNTL, 0x1); + + /* Turn on the IFPC counter (countable 4 on XOCLK1) */ + gmu_write(&a6xx_gpu->gmu, REG_A8XX_GMU_CX_GMU_POWER_COUNTER_SELECT_XOCLK_1, + FIELD_PREP(GENMASK(7, 0), 0x4)); + + /* Select CP0 to always count cycles */ + gpu_write(gpu, REG_A8XX_CP_PERFCTR_CP_SEL(0), 1); + + a8xx_set_ubwc_config(gpu); + + /* Set weights for bicubic filtering */ + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(0), 0); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(1), 0x3fe05ff4); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(2), 0x3fa0ebee); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(3), 0x3f5193ed); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(4), 0x3f0243f0); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(5), 0x00000000); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(6), 0x3fd093e8); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(7), 0x3f4133dc); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(8), 0x3ea1dfdb); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(9), 0x3e0283e0); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(10), 0x0000ac2b); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(11), 0x0000f01d); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(12), 0x00114412); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(13), 0x0021980a); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(14), 0x0051ec05); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(15), 0x0000380e); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(16), 0x3ff09001); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(17), 0x3fc10bfa); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(18), 0x3f9193f7); + gpu_write(gpu, REG_A8XX_TPL1_BICUBIC_WEIGHTS_TABLE(19), 0x3f7227f7); + + gpu_write(gpu, REG_A8XX_UCHE_CLIENT_PF, BIT(7) | 0x1); + + a8xx_nonctxt_config(gpu, &gmem_protect); + + /* Enable fault detection */ + gpu_write(gpu, REG_A8XX_RBBM_INTERFACE_HANG_INT_CNTL, BIT(30) | 0xcfffff); + gpu_write(gpu, REG_A8XX_RBBM_SLICE_INTERFACE_HANG_INT_CNTL, BIT(30)); + + /* Set up the CX GMU counter 0 to count busy ticks */ + gmu_write(gmu, REG_A6XX_GPU_GMU_AO_GPU_CX_BUSY_MASK, 0xff000000); + + /* Enable the power counter */ + gmu_rmw(gmu, REG_A8XX_GMU_CX_GMU_POWER_COUNTER_SELECT_XOCLK_0, 0xff, BIT(5)); + gmu_write(gmu, REG_A8XX_GMU_CX_GMU_POWER_COUNTER_ENABLE, 1); + + /* Protect registers from the CP */ + a8xx_set_cp_protect(gpu); + + /* Enable the GMEM save/restore feature for preemption */ + a8xx_write_pipe(gpu, PIPE_BR, REG_A6XX_RB_CONTEXT_SWITCH_GMEM_SAVE_RESTORE_ENABLE, 1); + + for (pipe_id = PIPE_BR; pipe_id <= PIPE_DDE_BV; pipe_id++) { + u32 apriv_mask = A8XX_APRIV_MASK; + unsigned long flags; + + if (pipe_id == PIPE_LPAC) + continue; + + if (pipe_id == PIPE_BR) + apriv_mask = A8XX_BR_APRIV_MASK; + + a8xx_aperture_acquire(gpu, pipe_id, &flags); + gpu_write(gpu, REG_A8XX_CP_APRIV_CNTL_PIPE, apriv_mask); + gpu_write(gpu, REG_A8XX_CP_INTERRUPT_STATUS_MASK_PIPE, + A8XX_CP_INTERRUPT_STATUS_MASK_PIPE); + gpu_write(gpu, REG_A8XX_CP_HW_FAULT_STATUS_MASK_PIPE, + A8XX_CP_HW_FAULT_STATUS_MASK_PIPE); + a8xx_aperture_release(gpu, flags); + } + + a8xx_aperture_clear(gpu); + + /* Enable interrupts */ + gpu_write(gpu, REG_A8XX_CP_INTERRUPT_STATUS_MASK_GLOBAL, A8XX_CP_GLOBAL_INT_MASK); + gpu_write(gpu, REG_A8XX_RBBM_INT_0_MASK, A8XX_INT_MASK); + + ret = adreno_hw_init(gpu); + if (ret) + goto out; + + gpu_write64(gpu, REG_A8XX_CP_SQE_INSTR_BASE, a6xx_gpu->sqe_iova); + if (a6xx_gpu->aqe_iova) + gpu_write64(gpu, REG_A8XX_CP_AQE_INSTR_BASE_0, a6xx_gpu->aqe_iova); + + /* Set the ringbuffer address */ + gpu_write64(gpu, REG_A6XX_CP_RB_BASE, gpu->rb[0]->iova); + gpu_write(gpu, REG_A6XX_CP_RB_CNTL, MSM_GPU_RB_CNTL_DEFAULT); + + /* Configure the RPTR shadow if needed: */ + gpu_write64(gpu, REG_A6XX_CP_RB_RPTR_ADDR, shadowptr(a6xx_gpu, gpu->rb[0])); + gpu_write64(gpu, REG_A8XX_CP_RB_RPTR_ADDR_BV, rbmemptr(gpu->rb[0], bv_rptr)); + + a8xx_preempt_hw_init(gpu); + + for (i = 0; i < gpu->nr_rings; i++) + a6xx_gpu->shadow[i] = 0; + + /* Always come up on rb 0 */ + a6xx_gpu->cur_ring = gpu->rb[0]; + + for (i = 0; i < gpu->nr_rings; i++) + gpu->rb[i]->cur_ctx_seqno = 0; + + /* Enable the SQE_to start the CP engine */ + gpu_write(gpu, REG_A8XX_CP_SQE_CNTL, 1); + + ret = a8xx_cp_init(gpu); + if (ret) + goto out; + + /* + * Try to load a zap shader into the secure world. If successful + * we can use the CP to switch out of secure mode. If not then we + * have no resource but to try to switch ourselves out manually. If we + * guessed wrong then access to the RBBM_SECVID_TRUST_CNTL register will + * be blocked and a permissions violation will soon follow. + */ + ret = a6xx_zap_shader_init(gpu); + if (!ret) { + OUT_PKT7(gpu->rb[0], CP_SET_SECURE_MODE, 1); + OUT_RING(gpu->rb[0], 0x00000000); + + a6xx_flush(gpu, gpu->rb[0]); + if (!a8xx_idle(gpu, gpu->rb[0])) + return -EINVAL; + } else if (ret == -ENODEV) { + /* + * This device does not use zap shader (but print a warning + * just in case someone got their dt wrong.. hopefully they + * have a debug UART to realize the error of their ways... + * if you mess this up you are about to crash horribly) + */ + dev_warn_once(gpu->dev->dev, + "Zap shader not enabled - using SECVID_TRUST_CNTL instead\n"); + gpu_write(gpu, REG_A6XX_RBBM_SECVID_TRUST_CNTL, 0x0); + ret = 0; + } else { + return ret; + } + + /* + * GMEM_PROTECT register should be programmed after GPU is transitioned to + * non-secure mode + */ + a8xx_write_pipe(gpu, PIPE_BR, REG_A8XX_RB_GC_GMEM_PROTECT, gmem_protect); + WARN_ON(!gmem_protect); + a8xx_aperture_clear(gpu); + + if (!a6xx_gpu->pwrup_reglist_emitted) { + a8xx_patch_pwrup_reglist(gpu); + a6xx_gpu->pwrup_reglist_emitted = true; + } + + /* Enable hardware clockgating */ + a8xx_set_hwcg(gpu, true); +out: + /* Last step - yield the ringbuffer */ + a8xx_preempt_start(gpu); + + /* + * Tell the GMU that we are done touching the GPU and it can start power + * management + */ + a6xx_gmu_clear_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET); + + if (!ret && msm_gpu_sysprof_no_perfcntr_zap(gpu)) { + ret = a6xx_gmu_set_oob(gmu, GMU_OOB_PERFCOUNTER_SET); + if (!ret) + set_bit(GMU_STATUS_OOB_PERF_SET, &gmu->status); + } + + return ret; +} + +int a8xx_hw_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + int ret; + + mutex_lock(&a6xx_gpu->gmu.lock); + ret = hw_init(gpu); + mutex_unlock(&a6xx_gpu->gmu.lock); + + return ret; +} + +static void a8xx_dump(struct msm_gpu *gpu) +{ + DRM_DEV_INFO(&gpu->pdev->dev, "status: %08x\n", gpu_read(gpu, REG_A8XX_RBBM_STATUS)); + adreno_dump(gpu); +} + +void a8xx_recover(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + int active_submits; + + adreno_dump_info(gpu); + + /* + * To handle recovery specific sequences during the rpm suspend we are + * about to trigger + */ + a6xx_gpu->hung = true; + + if (adreno_gpu->funcs->gx_is_on(adreno_gpu)) { + /* + * Sometimes crashstate capture is skipped, so SQE should be + * halted here again + */ + gpu_write(gpu, REG_A8XX_CP_SQE_CNTL, 3); + + if (hang_debug) + a8xx_dump(gpu); + } + + pm_runtime_dont_use_autosuspend(&gpu->pdev->dev); + + /* active_submit won't change until we make a submission */ + mutex_lock(&gpu->active_lock); + active_submits = gpu->active_submits; + + /* + * Temporarily clear active_submits count to silence a WARN() in the + * runtime suspend cb + */ + gpu->active_submits = 0; + + reinit_completion(&gmu->pd_gate); + dev_pm_genpd_add_notifier(gmu->cxpd, &gmu->pd_nb); + dev_pm_genpd_synced_poweroff(gmu->cxpd); + + /* Drop the rpm refcount from active submits */ + if (active_submits) + pm_runtime_put(&gpu->pdev->dev); + + /* And the final one from recover worker */ + pm_runtime_put_sync(&gpu->pdev->dev); + + if (!wait_for_completion_timeout(&gmu->pd_gate, msecs_to_jiffies(1000))) + DRM_DEV_ERROR(&gpu->pdev->dev, "cx gdsc didn't collapse\n"); + + dev_pm_genpd_remove_notifier(gmu->cxpd); + + pm_runtime_use_autosuspend(&gpu->pdev->dev); + + if (active_submits) + pm_runtime_get(&gpu->pdev->dev); + + pm_runtime_get_sync(&gpu->pdev->dev); + + gpu->active_submits = active_submits; + mutex_unlock(&gpu->active_lock); + + msm_gpu_hw_init(gpu); + a6xx_gpu->hung = false; +} + +static const char *a8xx_uche_fault_block(struct msm_gpu *gpu, u32 mid) +{ + static const char * const uche_clients[] = { + "BR_VFD", "BR_SP", "BR_VSC", "BR_VPC", "BR_HLSQ", "BR_PC", "BR_LRZ", "BR_TP", + "BV_VFD", "BV_SP", "BV_VSC", "BV_VPC", "BV_HLSQ", "BV_PC", "BV_LRZ", "BV_TP", + "STCHE", + }; + static const char * const uche_clients_lpac[] = { + "-", "SP_LPAC", "-", "-", "HLSQ_LPAC", "-", "-", "TP_LPAC", + }; + u32 val; + + /* + * The source of the data depends on the mid ID read from FSYNR1. + * and the client ID read from the UCHE block + */ + val = gpu_read(gpu, REG_A8XX_UCHE_CLIENT_PF); + + val &= GENMASK(6, 0); + + /* mid=3 refers to BR or BV */ + if (mid == 3) { + if (val < ARRAY_SIZE(uche_clients)) + return uche_clients[val]; + else + return "UCHE"; + } + + /* mid=8 refers to LPAC */ + if (mid == 8) { + if (val < ARRAY_SIZE(uche_clients_lpac)) + return uche_clients_lpac[val]; + else + return "UCHE_LPAC"; + } + + return "Unknown"; +} + +static const char *a8xx_fault_block(struct msm_gpu *gpu, u32 id) +{ + switch (id) { + case 0x0: + return "CP"; + case 0x1: + return "UCHE: Unknown"; + case 0x2: + return "UCHE_LPAC: Unknown"; + case 0x3: + case 0x8: + return a8xx_uche_fault_block(gpu, id); + case 0x4: + return "CCU"; + case 0x5: + return "Flag cache"; + case 0x6: + return "PREFETCH"; + case 0x7: + return "GMU"; + case 0x9: + return "UCHE_HPAC"; + } + + return "Unknown"; +} + +int a8xx_fault_handler(void *arg, unsigned long iova, int flags, void *data) +{ + struct msm_gpu *gpu = arg; + struct adreno_smmu_fault_info *info = data; + const char *block = "unknown"; + + u32 scratch[] = { + gpu_read(gpu, REG_A8XX_CP_SCRATCH_GLOBAL(0)), + gpu_read(gpu, REG_A8XX_CP_SCRATCH_GLOBAL(1)), + gpu_read(gpu, REG_A8XX_CP_SCRATCH_GLOBAL(2)), + gpu_read(gpu, REG_A8XX_CP_SCRATCH_GLOBAL(3)), + }; + + if (info) + block = a8xx_fault_block(gpu, info->fsynr1 & 0xff); + + return adreno_fault_handler(gpu, iova, flags, info, block, scratch); +} + +static void a8xx_cp_hw_err_irq(struct msm_gpu *gpu) +{ + u32 status = gpu_read(gpu, REG_A8XX_CP_INTERRUPT_STATUS_GLOBAL); + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + u32 slice = a8xx_get_first_slice(a6xx_gpu); + u32 hw_fault_mask = GENMASK(6, 0); + u32 sw_fault_mask = GENMASK(22, 16); + u32 pipe = 0; + + dev_err_ratelimited(&gpu->pdev->dev, "CP Fault Global INT status: 0x%x\n", status); + + if (status & (A8XX_CP_GLOBAL_INT_MASK_HWFAULTBR | + A8XX_CP_GLOBAL_INT_MASK_SWFAULTBR)) + pipe |= BIT(PIPE_BR); + + if (status & (A8XX_CP_GLOBAL_INT_MASK_HWFAULTBV | + A8XX_CP_GLOBAL_INT_MASK_SWFAULTBV)) + pipe |= BIT(PIPE_BV); + + if (!pipe) { + dev_err_ratelimited(&gpu->pdev->dev, "CP Fault Unknown pipe\n"); + goto out; + } + + for (unsigned int pipe_id = PIPE_NONE; pipe_id <= PIPE_DDE_BV; pipe_id++) { + if (!(BIT(pipe_id) & pipe)) + continue; + + if (hw_fault_mask & status) { + status = a8xx_read_pipe_slice(gpu, pipe_id, slice, + REG_A8XX_CP_HW_FAULT_STATUS_PIPE); + dev_err_ratelimited(&gpu->pdev->dev, + "CP HW FAULT pipe: %u status: 0x%x\n", pipe_id, status); + } + + if (sw_fault_mask & status) { + status = a8xx_read_pipe_slice(gpu, pipe_id, slice, + REG_A8XX_CP_INTERRUPT_STATUS_PIPE); + dev_err_ratelimited(&gpu->pdev->dev, + "CP SW FAULT pipe: %u status: 0x%x\n", pipe_id, status); + + if (status & BIT(8)) { + a8xx_write_pipe(gpu, pipe_id, REG_A8XX_CP_SQE_STAT_ADDR_PIPE, 1); + status = a8xx_read_pipe_slice(gpu, pipe_id, slice, + REG_A8XX_CP_SQE_STAT_DATA_PIPE); + dev_err_ratelimited(&gpu->pdev->dev, + "CP Opcode error, opcode=0x%x\n", status); + } + + if (status & BIT(10)) { + status = a8xx_read_pipe_slice(gpu, pipe_id, slice, + REG_A8XX_CP_PROTECT_STATUS_PIPE); + dev_err_ratelimited(&gpu->pdev->dev, + "CP REG PROTECT error, status=0x%x\n", status); + } + } + } + +out: + /* Turn off interrupts to avoid triggering recovery again */ + a8xx_aperture_clear(gpu); + gpu_write(gpu, REG_A8XX_CP_INTERRUPT_STATUS_MASK_GLOBAL, 0); + gpu_write(gpu, REG_A8XX_RBBM_INT_0_MASK, 0); + + kthread_queue_work(gpu->worker, &gpu->recover_work); +} + +static u32 gpu_periph_read(struct msm_gpu *gpu, u32 dbg_offset) +{ + gpu_write(gpu, REG_A8XX_CP_SQE_UCODE_DBG_ADDR_PIPE, dbg_offset); + + return gpu_read(gpu, REG_A8XX_CP_SQE_UCODE_DBG_DATA_PIPE); +} + +static u64 gpu_periph_read64(struct msm_gpu *gpu, u32 dbg_offset) +{ + u64 lo, hi; + + lo = gpu_periph_read(gpu, dbg_offset); + hi = gpu_periph_read(gpu, dbg_offset + 1); + + return (hi << 32) | lo; +} + +#define CP_PERIPH_IB1_BASE_LO 0x7005 +#define CP_PERIPH_IB1_BASE_HI 0x7006 +#define CP_PERIPH_IB1_SIZE 0x7007 +#define CP_PERIPH_IB1_OFFSET 0x7008 +#define CP_PERIPH_IB2_BASE_LO 0x7009 +#define CP_PERIPH_IB2_BASE_HI 0x700a +#define CP_PERIPH_IB2_SIZE 0x700b +#define CP_PERIPH_IB2_OFFSET 0x700c +#define CP_PERIPH_IB3_BASE_LO 0x700d +#define CP_PERIPH_IB3_BASE_HI 0x700e +#define CP_PERIPH_IB3_SIZE 0x700f +#define CP_PERIPH_IB3_OFFSET 0x7010 + +static void a8xx_fault_detect_irq(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct msm_ringbuffer *ring = gpu->funcs->active_ring(gpu); + unsigned long flags; + + /* + * If stalled on SMMU fault, we could trip the GPU's hang detection, + * but the fault handler will trigger the devcore dump, and we want + * to otherwise resume normally rather than killing the submit, so + * just bail. + */ + if (gpu_read(gpu, REG_A8XX_RBBM_MISC_STATUS) & A8XX_RBBM_MISC_STATUS_SMMU_STALLED_ON_FAULT) + return; + + /* + * Force the GPU to stay on until after we finish + * collecting information + */ + if (!adreno_has_gmu_wrapper(adreno_gpu)) + gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_GMU_PWR_COL_KEEPALIVE, 1); + + DRM_DEV_ERROR(&gpu->pdev->dev, + "gpu fault ring %d fence %x status %8.8X gfx_status %8.8X\n", + ring ? ring->id : -1, ring ? ring->fctx->last_fence : 0, + gpu_read(gpu, REG_A8XX_RBBM_STATUS), gpu_read(gpu, REG_A8XX_RBBM_GFX_STATUS)); + + a8xx_aperture_acquire(gpu, PIPE_BR, &flags); + + DRM_DEV_ERROR(&gpu->pdev->dev, + "BR: status %8.8X rb %4.4x/%4.4x ib1 %16.16llX/%4.4x ib2 %16.16llX/%4.4x ib3 %16.16llX/%4.4x\n", + gpu_read(gpu, REG_A8XX_RBBM_GFX_BR_STATUS), + gpu_read(gpu, REG_A6XX_CP_RB_RPTR), + gpu_read(gpu, REG_A6XX_CP_RB_WPTR), + gpu_periph_read64(gpu, CP_PERIPH_IB1_BASE_LO), + gpu_periph_read(gpu, CP_PERIPH_IB1_OFFSET), + gpu_periph_read64(gpu, CP_PERIPH_IB2_BASE_LO), + gpu_periph_read(gpu, CP_PERIPH_IB2_OFFSET), + gpu_periph_read64(gpu, CP_PERIPH_IB3_BASE_LO), + gpu_periph_read(gpu, CP_PERIPH_IB3_OFFSET)); + + a8xx_aperture_release(gpu, flags); + a8xx_aperture_acquire(gpu, PIPE_BV, &flags); + + DRM_DEV_ERROR(&gpu->pdev->dev, + "BV: status %8.8X rb %4.4x/%4.4x ib1 %16.16llX/%4.4x ib2 %16.16llX/%4.4x ib3 %16.16llX/%4.4x\n", + gpu_read(gpu, REG_A8XX_RBBM_GFX_BV_STATUS), + gpu_read(gpu, REG_A8XX_CP_RB_RPTR_BV), + gpu_read(gpu, REG_A6XX_CP_RB_WPTR), + gpu_periph_read64(gpu, CP_PERIPH_IB1_BASE_LO), + gpu_periph_read(gpu, CP_PERIPH_IB1_OFFSET), + gpu_periph_read64(gpu, CP_PERIPH_IB2_BASE_LO), + gpu_periph_read(gpu, CP_PERIPH_IB2_OFFSET), + gpu_periph_read64(gpu, CP_PERIPH_IB3_BASE_LO), + gpu_periph_read(gpu, CP_PERIPH_IB3_OFFSET)); + + a8xx_aperture_release(gpu, flags); + a8xx_aperture_clear(gpu); + + /* Turn off the hangcheck timer to keep it from bothering us */ + timer_delete(&gpu->hangcheck_timer); + + kthread_queue_work(gpu->worker, &gpu->recover_work); +} + +static void a8xx_sw_fuse_violation_irq(struct msm_gpu *gpu) +{ + u32 status; + + status = gpu_read(gpu, REG_A8XX_RBBM_SW_FUSE_INT_STATUS); + gpu_write(gpu, REG_A8XX_RBBM_SW_FUSE_INT_MASK, 0); + + dev_err_ratelimited(&gpu->pdev->dev, "SW fuse violation status=%8.8x\n", status); + + /* + * Ignore FASTBLEND violations, because the HW will silently fall back + * to legacy blending. + */ + if (status & (A7XX_CX_MISC_SW_FUSE_VALUE_RAYTRACING | + A7XX_CX_MISC_SW_FUSE_VALUE_LPAC)) { + timer_delete(&gpu->hangcheck_timer); + + kthread_queue_work(gpu->worker, &gpu->recover_work); + } +} + +irqreturn_t a8xx_irq(struct msm_gpu *gpu) +{ + struct msm_drm_private *priv = gpu->dev->dev_private; + u32 status = gpu_read(gpu, REG_A8XX_RBBM_INT_0_STATUS); + + gpu_write(gpu, REG_A8XX_RBBM_INT_CLEAR_CMD, status); + + if (priv->disable_err_irq) { + /* Turn off interrupts to avoid interrupt storm */ + gpu_write(gpu, REG_A8XX_RBBM_INT_0_MASK, A6XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS); + status &= A6XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS; + } + + if (status & A6XX_RBBM_INT_0_MASK_RBBM_HANG_DETECT) + a8xx_fault_detect_irq(gpu); + + if (status & A6XX_RBBM_INT_0_MASK_CP_AHB_ERROR) { + u32 rl0, rl1; + + rl0 = gpu_read(gpu, REG_A8XX_CP_RL_ERROR_DETAILS_0); + rl1 = gpu_read(gpu, REG_A8XX_CP_RL_ERROR_DETAILS_1); + dev_err_ratelimited(&gpu->pdev->dev, + "CP | AHB bus error RL_ERROR_0: %x, RL_ERROR_1: %x\n", rl0, rl1); + } + + if (status & A6XX_RBBM_INT_0_MASK_CP_HW_ERROR) + a8xx_cp_hw_err_irq(gpu); + + if (status & A6XX_RBBM_INT_0_MASK_RBBM_ATB_ASYNCFIFO_OVERFLOW) + dev_err_ratelimited(&gpu->pdev->dev, "RBBM | ATB ASYNC overflow\n"); + + if (status & A6XX_RBBM_INT_0_MASK_RBBM_ATB_BUS_OVERFLOW) + dev_err_ratelimited(&gpu->pdev->dev, "RBBM | ATB bus overflow\n"); + + if (status & A6XX_RBBM_INT_0_MASK_UCHE_OOB_ACCESS) + dev_err_ratelimited(&gpu->pdev->dev, "UCHE | Out of bounds access\n"); + + if (status & A6XX_RBBM_INT_0_MASK_UCHE_TRAP_INTR) + dev_err_ratelimited(&gpu->pdev->dev, "UCHE | Trap interrupt\n"); + + if (status & A6XX_RBBM_INT_0_MASK_SWFUSEVIOLATION) + a8xx_sw_fuse_violation_irq(gpu); + + if (status & A6XX_RBBM_INT_0_MASK_CP_CACHE_FLUSH_TS) { + msm_gpu_retire(gpu); + a8xx_preempt_trigger(gpu); + } + + if (status & A6XX_RBBM_INT_0_MASK_CP_SW) + a8xx_preempt_irq(gpu); + + return IRQ_HANDLED; +} + +void a8xx_llc_activate(struct a6xx_gpu *a6xx_gpu) +{ + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; + struct msm_gpu *gpu = &adreno_gpu->base; + + if (!llcc_slice_activate(a6xx_gpu->llc_slice)) { + u32 gpu_scid = llcc_get_slice_id(a6xx_gpu->llc_slice); + + gpu_scid &= GENMASK(5, 0); + + gpu_write(gpu, REG_A6XX_GBIF_SCACHE_CNTL1, + FIELD_PREP(GENMASK(29, 24), gpu_scid) | + FIELD_PREP(GENMASK(23, 18), gpu_scid) | + FIELD_PREP(GENMASK(17, 12), gpu_scid) | + FIELD_PREP(GENMASK(11, 6), gpu_scid) | + FIELD_PREP(GENMASK(5, 0), gpu_scid)); + + gpu_write(gpu, REG_A6XX_GBIF_SCACHE_CNTL0, + FIELD_PREP(GENMASK(27, 22), gpu_scid) | + FIELD_PREP(GENMASK(21, 16), gpu_scid) | + FIELD_PREP(GENMASK(15, 10), gpu_scid) | + BIT(8)); + } + + llcc_slice_activate(a6xx_gpu->htw_llc_slice); +} + +#define GBIF_CLIENT_HALT_MASK BIT(0) +#define GBIF_ARB_HALT_MASK BIT(1) +#define VBIF_XIN_HALT_CTRL0_MASK GENMASK(3, 0) +#define VBIF_RESET_ACK_MASK 0xF0 +#define GPR0_GBIF_HALT_REQUEST 0x1E0 + +void a8xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, bool gx_off) +{ + struct msm_gpu *gpu = &adreno_gpu->base; + + if (gx_off) { + /* Halt the gx side of GBIF */ + gpu_write(gpu, REG_A8XX_RBBM_GBIF_HALT, 1); + spin_until(gpu_read(gpu, REG_A8XX_RBBM_GBIF_HALT_ACK) & 1); + } + + /* Halt new client requests on GBIF */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_CLIENT_HALT_MASK); + spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & + (GBIF_CLIENT_HALT_MASK)) == GBIF_CLIENT_HALT_MASK); + + /* Halt all AXI requests on GBIF */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_ARB_HALT_MASK); + spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & + (GBIF_ARB_HALT_MASK)) == GBIF_ARB_HALT_MASK); + + /* The GBIF halt needs to be explicitly cleared */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0); +} + +u64 a8xx_gmu_get_timestamp(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + u64 count_hi, count_lo, temp; + + do { + count_hi = gmu_read(&a6xx_gpu->gmu, REG_A8XX_GMU_ALWAYS_ON_COUNTER_H); + count_lo = gmu_read(&a6xx_gpu->gmu, REG_A8XX_GMU_ALWAYS_ON_COUNTER_L); + temp = gmu_read(&a6xx_gpu->gmu, REG_A8XX_GMU_ALWAYS_ON_COUNTER_H); + } while (unlikely(count_hi != temp)); + + return (count_hi << 32) | count_lo; +} + +u64 a8xx_gpu_busy(struct msm_gpu *gpu, unsigned long *out_sample_rate) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + u64 busy_cycles; + + /* 19.2MHz */ + *out_sample_rate = 19200000; + + busy_cycles = gmu_read64(&a6xx_gpu->gmu, + REG_A8XX_GMU_CX_GMU_POWER_COUNTER_XOCLK_0_L, + REG_A8XX_GMU_CX_GMU_POWER_COUNTER_XOCLK_0_H); + + return busy_cycles; +} + +bool a8xx_progress(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + return true; +} + +void a8xx_perfcntr_flush(struct msm_gpu *gpu) +{ + u32 val; + + /* + * Flush delta counters (both perf counters and pipe stats) present in + * RBBM_S and RBBM_US to perf RAM logic to get the latest data. + */ + gpu_write(gpu, REG_A8XX_RBBM_PERFCTR_FLUSH_HOST_CMD, BIT(0)); + gpu_write(gpu, REG_A8XX_RBBM_SLICE_PERFCTR_FLUSH_HOST_CMD, BIT(0)); + + /* Ensure all writes are posted before polling status register */ + wmb(); + + if (gpu_poll_timeout(gpu, REG_A8XX_RBBM_PERFCTR_FLUSH_HOST_STATUS, val, + val & BIT(0), 100, 100 * 1000)) { + dev_err(&gpu->pdev->dev, "Perfcounter flush timed out: status=0x%08x\n", val); + } +} diff --git a/drivers/gpu/drm/msm/adreno/a8xx_preempt.c b/drivers/gpu/drm/msm/adreno/a8xx_preempt.c new file mode 100644 index 000000000..6cb53a071 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a8xx_preempt.c @@ -0,0 +1,259 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. */ + +#include "msm_gem.h" +#include "a6xx_gpu.h" +#include "a6xx_gmu.xml.h" +#include "a6xx_preempt.h" +#include "msm_mmu.h" +#include "msm_gpu_trace.h" + +static void preempt_prepare_postamble(struct a6xx_gpu *a6xx_gpu) +{ + u32 *postamble = a6xx_gpu->preempt_postamble_ptr; + u32 count = 0; + + postamble[count++] = PKT7(CP_REG_RMW, 3); + postamble[count++] = REG_A8XX_RBBM_PERFCTR_SRAM_INIT_CMD; + postamble[count++] = 0; + postamble[count++] = 1; + + postamble[count++] = PKT7(CP_WAIT_REG_MEM, 6); + postamble[count++] = CP_WAIT_REG_MEM_0_FUNCTION(WRITE_EQ); + postamble[count++] = CP_WAIT_REG_MEM_POLL_ADDR_LO( + REG_A8XX_RBBM_PERFCTR_SRAM_INIT_STATUS); + postamble[count++] = CP_WAIT_REG_MEM_POLL_ADDR_HI(0); + postamble[count++] = CP_WAIT_REG_MEM_3_REF(0x1); + postamble[count++] = CP_WAIT_REG_MEM_4_MASK(0x1); + postamble[count++] = CP_WAIT_REG_MEM_5_DELAY_LOOP_CYCLES(0); + + a6xx_gpu->preempt_postamble_len = count; + + a6xx_gpu->postamble_enabled = true; +} + +static void preempt_disable_postamble(struct a6xx_gpu *a6xx_gpu) +{ + u32 *postamble = a6xx_gpu->preempt_postamble_ptr; + + /* + * Disable the postamble by replacing the first packet header with a NOP + * that covers the whole buffer. + */ + *postamble = PKT7(CP_NOP, (a6xx_gpu->preempt_postamble_len - 1)); + + a6xx_gpu->postamble_enabled = false; +} + +/* + * Set preemption keepalive vote. Please note that this vote is different from the one used in + * a8xx_irq() + */ +static void a8xx_preempt_keepalive_vote(struct msm_gpu *gpu, bool on) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + gmu_write(&a6xx_gpu->gmu, REG_A8XX_GMU_PWR_COL_PREEMPT_KEEPALIVE, on); +} + +void a8xx_preempt_irq(struct msm_gpu *gpu) +{ + uint32_t status; + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct drm_device *dev = gpu->dev; + + if (!try_preempt_state(a6xx_gpu, PREEMPT_TRIGGERED, PREEMPT_PENDING)) + return; + + /* Delete the preemption watchdog timer */ + timer_delete(&a6xx_gpu->preempt_timer); + + /* + * The hardware should be setting the stop bit of CP_CONTEXT_SWITCH_CNTL + * to zero before firing the interrupt, but there is a non zero chance + * of a hardware condition or a software race that could set it again + * before we have a chance to finish. If that happens, log and go for + * recovery + */ + status = gpu_read(gpu, REG_A8XX_CP_CONTEXT_SWITCH_CNTL); + if (unlikely(status & A8XX_CP_CONTEXT_SWITCH_CNTL_STOP)) { + DRM_DEV_ERROR(&gpu->pdev->dev, + "!!!!!!!!!!!!!!!! preemption faulted !!!!!!!!!!!!!! irq\n"); + set_preempt_state(a6xx_gpu, PREEMPT_FAULTED); + dev_err(dev->dev, "%s: Preemption failed to complete\n", + gpu->name); + kthread_queue_work(gpu->worker, &gpu->recover_work); + return; + } + + a6xx_gpu->cur_ring = a6xx_gpu->next_ring; + a6xx_gpu->next_ring = NULL; + + set_preempt_state(a6xx_gpu, PREEMPT_FINISH); + + update_wptr(a6xx_gpu, a6xx_gpu->cur_ring); + + set_preempt_state(a6xx_gpu, PREEMPT_NONE); + + a8xx_preempt_keepalive_vote(gpu, false); + + trace_msm_gpu_preemption_irq(a6xx_gpu->cur_ring->id); + + /* + * Retrigger preemption to avoid a deadlock that might occur when preemption + * is skipped due to it being already in flight when requested. + */ + a8xx_preempt_trigger(gpu); +} + +void a8xx_preempt_hw_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + int i; + + /* No preemption if we only have one ring */ + if (gpu->nr_rings == 1) + return; + + for (i = 0; i < gpu->nr_rings; i++) { + struct a6xx_preempt_record *record_ptr = a6xx_gpu->preempt[i]; + + record_ptr->wptr = 0; + record_ptr->rptr = 0; + record_ptr->rptr_addr = shadowptr(a6xx_gpu, gpu->rb[i]); + record_ptr->info = 0; + record_ptr->data = 0; + record_ptr->rbase = gpu->rb[i]->iova; + } + + /* Write a 0 to signal that we aren't switching pagetables */ + gpu_write64(gpu, REG_A8XX_CP_CONTEXT_SWITCH_SMMU_INFO, 0); + + /* Enable the GMEM save/restore feature for preemption */ + gpu_write(gpu, REG_A6XX_RB_CONTEXT_SWITCH_GMEM_SAVE_RESTORE_ENABLE, 0x1); + + /* Reset the preemption state */ + set_preempt_state(a6xx_gpu, PREEMPT_NONE); + + spin_lock_init(&a6xx_gpu->eval_lock); + + /* Always come up on rb 0 */ + a6xx_gpu->cur_ring = gpu->rb[0]; +} + +void a8xx_preempt_trigger(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + unsigned long flags; + struct msm_ringbuffer *ring; + unsigned int cntl; + bool sysprof; + + if (gpu->nr_rings == 1) + return; + + /* + * Lock to make sure another thread attempting preemption doesn't skip it + * while we are still evaluating the next ring. This makes sure the other + * thread does start preemption if we abort it and avoids a soft lock. + */ + spin_lock_irqsave(&a6xx_gpu->eval_lock, flags); + + /* + * Try to start preemption by moving from NONE to START. If + * unsuccessful, a preemption is already in flight + */ + if (!try_preempt_state(a6xx_gpu, PREEMPT_NONE, PREEMPT_START)) { + spin_unlock_irqrestore(&a6xx_gpu->eval_lock, flags); + return; + } + + cntl = A8XX_CP_CONTEXT_SWITCH_CNTL_LEVEL(a6xx_gpu->preempt_level); + + if (a6xx_gpu->skip_save_restore) + cntl |= A8XX_CP_CONTEXT_SWITCH_CNTL_SKIP_SAVE_RESTORE; + + if (a6xx_gpu->uses_gmem) + cntl |= A8XX_CP_CONTEXT_SWITCH_CNTL_USES_GMEM; + + cntl |= A8XX_CP_CONTEXT_SWITCH_CNTL_STOP; + + /* Get the next ring to preempt to */ + ring = get_next_ring(gpu); + + /* + * If no ring is populated or the highest priority ring is the current + * one do nothing except to update the wptr to the latest and greatest + */ + if (!ring || (a6xx_gpu->cur_ring == ring)) { + set_preempt_state(a6xx_gpu, PREEMPT_FINISH); + update_wptr(a6xx_gpu, a6xx_gpu->cur_ring); + set_preempt_state(a6xx_gpu, PREEMPT_NONE); + spin_unlock_irqrestore(&a6xx_gpu->eval_lock, flags); + return; + } + + spin_unlock_irqrestore(&a6xx_gpu->eval_lock, flags); + + spin_lock_irqsave(&ring->preempt_lock, flags); + + struct a7xx_cp_smmu_info *smmu_info_ptr = + a6xx_gpu->preempt_smmu[ring->id]; + struct a6xx_preempt_record *record_ptr = a6xx_gpu->preempt[ring->id]; + u64 ttbr0 = ring->memptrs->ttbr0; + u32 context_idr = ring->memptrs->context_idr; + + smmu_info_ptr->ttbr0 = ttbr0; + smmu_info_ptr->context_idr = context_idr; + record_ptr->wptr = get_wptr(ring); + + /* + * The GPU will write the wptr we set above when we preempt. Reset + * restore_wptr to make sure that we don't write WPTR to the same + * thing twice. It's still possible subsequent submissions will update + * wptr again, in which case they will set the flag to true. This has + * to be protected by the lock for setting the flag and updating wptr + * to be atomic. + */ + ring->restore_wptr = false; + + trace_msm_gpu_preemption_trigger(a6xx_gpu->cur_ring->id, ring->id); + + spin_unlock_irqrestore(&ring->preempt_lock, flags); + + /* Set the keepalive bit to keep the GPU ON until preemption is complete */ + a8xx_preempt_keepalive_vote(gpu, true); + + a6xx_fenced_write(a6xx_gpu, + REG_A8XX_CP_CONTEXT_SWITCH_SMMU_INFO, a6xx_gpu->preempt_smmu_iova[ring->id], + BIT(1), true); + + a6xx_fenced_write(a6xx_gpu, + REG_A8XX_CP_CONTEXT_SWITCH_PRIV_NON_SECURE_RESTORE_ADDR, + a6xx_gpu->preempt_iova[ring->id], BIT(1), true); + + a6xx_gpu->next_ring = ring; + + /* Start a timer to catch a stuck preemption */ + mod_timer(&a6xx_gpu->preempt_timer, jiffies + msecs_to_jiffies(10000)); + + /* Enable or disable postamble as needed */ + sysprof = msm_gpu_sysprof_no_perfcntr_zap(gpu); + + if (!sysprof && !a6xx_gpu->postamble_enabled) + preempt_prepare_postamble(a6xx_gpu); + + if (sysprof && a6xx_gpu->postamble_enabled) + preempt_disable_postamble(a6xx_gpu); + + /* Set the preemption state to triggered */ + set_preempt_state(a6xx_gpu, PREEMPT_TRIGGERED); + + /* Trigger the preemption */ + a6xx_fenced_write(a6xx_gpu, REG_A8XX_CP_CONTEXT_SWITCH_CNTL, cntl, BIT(1), false); +} + diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c new file mode 100644 index 000000000..05c77fe27 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/adreno_device.c @@ -0,0 +1,433 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013-2014 Red Hat + * Author: Rob Clark + * + * Copyright (c) 2014,2017 The Linux Foundation. All rights reserved. + */ + +#include "adreno_gpu.h" + +bool hang_debug = false; +MODULE_PARM_DESC(hang_debug, "Dump registers when hang is detected (can be slow!)"); +module_param_named(hang_debug, hang_debug, bool, 0600); + +bool snapshot_debugbus = false; +MODULE_PARM_DESC(snapshot_debugbus, "Include debugbus sections in GPU devcoredump (if not fused off)"); +module_param_named(snapshot_debugbus, snapshot_debugbus, bool, 0600); + +int enable_preemption = -1; +MODULE_PARM_DESC(enable_preemption, "Enable preemption (A7xx+ only) (1=on , 0=disable, -1=auto (default))"); +module_param(enable_preemption, int, 0600); + +bool disable_acd; +MODULE_PARM_DESC(disable_acd, "Forcefully disable GPU ACD"); +module_param_unsafe(disable_acd, bool, 0400); + +static bool skip_gpu; +MODULE_PARM_DESC(skip_gpu, "Disable GPU driver register (0=enable GPU driver register (default), 1=skip GPU driver register"); +module_param(skip_gpu, bool, 0400); + +extern const struct adreno_gpulist a2xx_gpulist; +extern const struct adreno_gpulist a3xx_gpulist; +extern const struct adreno_gpulist a4xx_gpulist; +extern const struct adreno_gpulist a5xx_gpulist; +extern const struct adreno_gpulist a6xx_gpulist; +extern const struct adreno_gpulist a7xx_gpulist; +extern const struct adreno_gpulist a8xx_gpulist; + +static const struct adreno_gpulist *gpulists[] = { + &a2xx_gpulist, + &a3xx_gpulist, + &a4xx_gpulist, + &a5xx_gpulist, + &a6xx_gpulist, + &a7xx_gpulist, + &a8xx_gpulist, +}; + +static const struct adreno_info *adreno_info(uint32_t chip_id) +{ + /* identify gpu: */ + for (int i = 0; i < ARRAY_SIZE(gpulists); i++) { + for (int j = 0; j < gpulists[i]->gpus_count; j++) { + const struct adreno_info *info = &gpulists[i]->gpus[j]; + + if (info->machine && !of_machine_is_compatible(info->machine)) + continue; + + for (int k = 0; info->chip_ids[k]; k++) + if (info->chip_ids[k] == chip_id) + return info; + } + } + + return NULL; +} + +struct msm_gpu *adreno_load_gpu(struct drm_device *dev) +{ + struct msm_drm_private *priv = dev->dev_private; + struct platform_device *pdev = priv->gpu_pdev; + struct msm_gpu *gpu = NULL; + struct adreno_gpu *adreno_gpu; + int ret; + + if (pdev) + gpu = dev_to_gpu(&pdev->dev); + + if (!gpu) { + dev_err_once(dev->dev, "no GPU device was found\n"); + return NULL; + } + + adreno_gpu = to_adreno_gpu(gpu); + + /* + * The number one reason for HW init to fail is if the firmware isn't + * loaded yet. Try that first and don't bother continuing on + * otherwise + */ + + ret = adreno_load_fw(adreno_gpu); + if (ret) + return NULL; + + if (gpu->funcs->ucode_load) { + ret = gpu->funcs->ucode_load(gpu); + if (ret) + return NULL; + } + + /* + * Now that we have firmware loaded, and are ready to begin + * booting the gpu, go ahead and enable runpm: + */ + pm_runtime_enable(&pdev->dev); + + ret = pm_runtime_get_sync(&pdev->dev); + if (ret < 0) { + pm_runtime_put_noidle(&pdev->dev); + DRM_DEV_ERROR(dev->dev, "Couldn't power up the GPU: %d\n", ret); + goto err_disable_rpm; + } + + mutex_lock(&gpu->lock); + ret = msm_gpu_hw_init(gpu); + mutex_unlock(&gpu->lock); + if (ret) { + DRM_DEV_ERROR(dev->dev, "gpu hw init failed: %d\n", ret); + goto err_put_rpm; + } + + pm_runtime_put_autosuspend(&pdev->dev); + +#ifdef CONFIG_DEBUG_FS + if (gpu->funcs->debugfs_init) { + gpu->funcs->debugfs_init(gpu, dev->primary); + gpu->funcs->debugfs_init(gpu, dev->render); + } +#endif + + return gpu; + +err_put_rpm: + pm_runtime_put_sync_suspend(&pdev->dev); +err_disable_rpm: + pm_runtime_disable(&pdev->dev); + + return NULL; +} + +static int find_chipid(struct device_node *node, uint32_t *chipid) +{ + const char *compat; + int ret; + + /* first search the compat strings for qcom,adreno-XYZ.W: */ + ret = of_property_read_string_index(node, "compatible", 0, &compat); + if (ret == 0) { + unsigned int r, patch; + + if (sscanf(compat, "qcom,adreno-%u.%u", &r, &patch) == 2 || + sscanf(compat, "amd,imageon-%u.%u", &r, &patch) == 2) { + uint32_t core, major, minor; + + core = r / 100; + r %= 100; + major = r / 10; + r %= 10; + minor = r; + + *chipid = (core << 24) | + (major << 16) | + (minor << 8) | + patch; + + return 0; + } + + if (sscanf(compat, "qcom,adreno-%08x", chipid) == 1) + return 0; + } + + /* and if that fails, fall back to legacy "qcom,chipid" property: */ + ret = of_property_read_u32(node, "qcom,chipid", chipid); + if (ret) { + DRM_ERROR("%pOF: could not parse qcom,chipid: %d\n", + node, ret); + return ret; + } + + pr_warn("%pOF: Using legacy qcom,chipid binding!\n", node); + + return 0; +} + +bool adreno_has_gpu(struct device_node *node) +{ + const struct adreno_info *info; + uint32_t chip_id; + int ret; + + if (skip_gpu) + return false; + + ret = find_chipid(node, &chip_id); + if (ret) + return false; + + info = adreno_info(chip_id); + if (!info) { + pr_warn("%pOF: Unknown GPU revision: %"ADRENO_CHIPID_FMT"\n", + node, ADRENO_CHIPID_ARGS(chip_id)); + return false; + } + + return true; +} + +static int adreno_bind(struct device *dev, struct device *master, void *data) +{ + static struct adreno_platform_config config = {}; + const struct adreno_info *info; + struct msm_drm_private *priv = dev_get_drvdata(master); + struct drm_device *drm = priv->dev; + struct msm_gpu *gpu; + int ret; + + ret = find_chipid(dev->of_node, &config.chip_id); + /* We shouldn't have gotten this far if we can't parse the chip_id */ + if (WARN_ON(ret)) + return ret; + + dev->platform_data = &config; + priv->gpu_pdev = to_platform_device(dev); + + info = adreno_info(config.chip_id); + /* We shouldn't have gotten this far if we don't recognize the GPU: */ + if (WARN_ON(!info)) + return -ENXIO; + + config.info = info; + + DBG("Found GPU: %"ADRENO_CHIPID_FMT, ADRENO_CHIPID_ARGS(config.chip_id)); + + priv->is_a2xx = info->family < ADRENO_3XX; + priv->has_cached_coherent = + !!(info->quirks & ADRENO_QUIRK_HAS_CACHED_COHERENT); + + gpu = info->funcs->init(drm); + if (IS_ERR(gpu)) { + dev_warn(drm->dev, "failed to load adreno gpu\n"); + return PTR_ERR(gpu); + } + + ret = dev_pm_opp_of_find_icc_paths(dev, NULL); + if (ret) + return ret; + + return 0; +} + +static int adreno_system_suspend(struct device *dev); +static void adreno_unbind(struct device *dev, struct device *master, + void *data) +{ + struct msm_drm_private *priv = dev_get_drvdata(master); + struct msm_gpu *gpu = dev_to_gpu(dev); + + if (pm_runtime_enabled(dev)) + WARN_ON_ONCE(adreno_system_suspend(dev)); + gpu->funcs->destroy(gpu); + + priv->gpu_pdev = NULL; +} + +static const struct component_ops a3xx_ops = { + .bind = adreno_bind, + .unbind = adreno_unbind, +}; + +static int adreno_probe(struct platform_device *pdev) +{ + if (of_device_is_compatible(pdev->dev.of_node, "amd,imageon") || + msm_gpu_no_components()) + return msm_gpu_probe(pdev, &a3xx_ops); + + return component_add(&pdev->dev, &a3xx_ops); +} + +static void adreno_remove(struct platform_device *pdev) +{ + struct msm_drm_private *priv = platform_get_drvdata(pdev); + + if (priv->kms_init) + component_del(&pdev->dev, &a3xx_ops); + else + msm_gpu_remove(pdev, &a3xx_ops); +} + +static void adreno_shutdown(struct platform_device *pdev) +{ + WARN_ON_ONCE(adreno_system_suspend(&pdev->dev)); +} + +static const struct of_device_id dt_match[] = { + { .compatible = "qcom,adreno" }, + { .compatible = "qcom,adreno-3xx" }, + /* for compatibility with imx5 gpu: */ + { .compatible = "amd,imageon" }, + /* for backwards compat w/ downstream kgsl DT files: */ + { .compatible = "qcom,kgsl-3d0" }, + {} +}; +MODULE_DEVICE_TABLE(of, dt_match); + +static int adreno_runtime_resume(struct device *dev) +{ + struct msm_gpu *gpu = dev_to_gpu(dev); + int ret = gpu->funcs->pm_resume(gpu); + if (!ret) + ret = msm_perfcntr_resume(gpu); + return ret; +} + +static int adreno_runtime_suspend(struct device *dev) +{ + struct msm_gpu *gpu = dev_to_gpu(dev); + + /* + * We should be holding a runpm ref, which will prevent + * runtime suspend. In the system suspend path, we've + * already waited for active jobs to complete. + */ + WARN_ON_ONCE(gpu->active_submits); + + msm_perfcntr_suspend(gpu); + + return gpu->funcs->pm_suspend(gpu); +} + +static void suspend_scheduler(struct msm_gpu *gpu) +{ + int i; + + /* + * Shut down the scheduler before we force suspend, so that + * suspend isn't racing with scheduler kthread feeding us + * more work. + * + * Note, we just want to park the thread, and let any jobs + * that are already on the hw queue complete normally, as + * opposed to the drm_sched_stop() path used for handling + * faulting/timed-out jobs. We can't really cancel any jobs + * already on the hw queue without racing with the GPU. + */ + for (i = 0; i < gpu->nr_rings; i++) { + struct drm_gpu_scheduler *sched = &gpu->rb[i]->sched; + + drm_sched_wqueue_stop(sched); + } +} + +static void resume_scheduler(struct msm_gpu *gpu) +{ + int i; + + for (i = 0; i < gpu->nr_rings; i++) { + struct drm_gpu_scheduler *sched = &gpu->rb[i]->sched; + + drm_sched_wqueue_start(sched); + } +} + +static int adreno_system_suspend(struct device *dev) +{ + struct msm_gpu *gpu = dev_to_gpu(dev); + int remaining, ret; + + if (!gpu) + return 0; + + suspend_scheduler(gpu); + + remaining = wait_event_timeout(gpu->retire_event, + gpu->active_submits == 0, + msecs_to_jiffies(1000)); + if (remaining == 0) { + dev_err(dev, "Timeout waiting for GPU to suspend\n"); + ret = -EBUSY; + goto out; + } + + ret = pm_runtime_force_suspend(dev); +out: + if (ret) + resume_scheduler(gpu); + + return ret; +} + +static int adreno_system_resume(struct device *dev) +{ + struct msm_gpu *gpu = dev_to_gpu(dev); + + if (!gpu) + return 0; + + resume_scheduler(gpu); + return pm_runtime_force_resume(dev); +} + +static const struct dev_pm_ops adreno_pm_ops = { + SYSTEM_SLEEP_PM_OPS(adreno_system_suspend, adreno_system_resume) + RUNTIME_PM_OPS(adreno_runtime_suspend, adreno_runtime_resume, NULL) +}; + +static struct platform_driver adreno_driver = { + .probe = adreno_probe, + .remove = adreno_remove, + .shutdown = adreno_shutdown, + .driver = { + .name = "adreno", + .of_match_table = dt_match, + .pm = &adreno_pm_ops, + }, +}; + +void __init adreno_register(void) +{ + if (skip_gpu) + return; + + platform_driver_register(&adreno_driver); +} + +void __exit adreno_unregister(void) +{ + if (skip_gpu) + return; + + platform_driver_unregister(&adreno_driver); +} diff --git a/drivers/gpu/drm/msm/adreno/adreno_gen7_0_0_snapshot.h b/drivers/gpu/drm/msm/adreno/adreno_gen7_0_0_snapshot.h new file mode 100644 index 000000000..d513e03fe --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/adreno_gen7_0_0_snapshot.h @@ -0,0 +1,935 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2020-2021, The Linux Foundation. All rights reserved. + * Copyright (c) 2022-2023 Qualcomm Innovation Center, Inc. All rights reserved. + */ +#ifndef __ADRENO_GEN7_0_0_SNAPSHOT_H +#define __ADRENO_GEN7_0_0_SNAPSHOT_H + +#include "a6xx_gpu_state.h" + +static const u32 gen7_0_0_debugbus_blocks[] = { + A7XX_DBGBUS_CP_0_0, + A7XX_DBGBUS_CP_0_1, + A7XX_DBGBUS_RBBM, + A7XX_DBGBUS_HLSQ, + A7XX_DBGBUS_UCHE_0, + A7XX_DBGBUS_TESS_BR, + A7XX_DBGBUS_TESS_BV, + A7XX_DBGBUS_PC_BR, + A7XX_DBGBUS_PC_BV, + A7XX_DBGBUS_VFDP_BR, + A7XX_DBGBUS_VFDP_BV, + A7XX_DBGBUS_VPC_BR, + A7XX_DBGBUS_VPC_BV, + A7XX_DBGBUS_TSE_BR, + A7XX_DBGBUS_TSE_BV, + A7XX_DBGBUS_RAS_BR, + A7XX_DBGBUS_RAS_BV, + A7XX_DBGBUS_VSC, + A7XX_DBGBUS_COM_0, + A7XX_DBGBUS_LRZ_BR, + A7XX_DBGBUS_LRZ_BV, + A7XX_DBGBUS_UFC_0, + A7XX_DBGBUS_UFC_1, + A7XX_DBGBUS_GMU_GX, + A7XX_DBGBUS_DBGC, + A7XX_DBGBUS_GPC_BR, + A7XX_DBGBUS_GPC_BV, + A7XX_DBGBUS_LARC, + A7XX_DBGBUS_HLSQ_SPTP, + A7XX_DBGBUS_RB_0, + A7XX_DBGBUS_RB_1, + A7XX_DBGBUS_RB_2, + A7XX_DBGBUS_RB_3, + A7XX_DBGBUS_UCHE_WRAPPER, + A7XX_DBGBUS_CCU_0, + A7XX_DBGBUS_CCU_1, + A7XX_DBGBUS_CCU_2, + A7XX_DBGBUS_CCU_3, + A7XX_DBGBUS_VFD_BR_0, + A7XX_DBGBUS_VFD_BR_1, + A7XX_DBGBUS_VFD_BR_2, + A7XX_DBGBUS_VFD_BR_3, + A7XX_DBGBUS_VFD_BR_4, + A7XX_DBGBUS_VFD_BR_5, + A7XX_DBGBUS_VFD_BR_6, + A7XX_DBGBUS_VFD_BR_7, + A7XX_DBGBUS_VFD_BV_0, + A7XX_DBGBUS_VFD_BV_1, + A7XX_DBGBUS_VFD_BV_2, + A7XX_DBGBUS_VFD_BV_3, + A7XX_DBGBUS_USP_0, + A7XX_DBGBUS_USP_1, + A7XX_DBGBUS_USP_2, + A7XX_DBGBUS_USP_3, + A7XX_DBGBUS_TP_0, + A7XX_DBGBUS_TP_1, + A7XX_DBGBUS_TP_2, + A7XX_DBGBUS_TP_3, + A7XX_DBGBUS_TP_4, + A7XX_DBGBUS_TP_5, + A7XX_DBGBUS_TP_6, + A7XX_DBGBUS_TP_7, + A7XX_DBGBUS_USPTP_0, + A7XX_DBGBUS_USPTP_1, + A7XX_DBGBUS_USPTP_2, + A7XX_DBGBUS_USPTP_3, + A7XX_DBGBUS_USPTP_4, + A7XX_DBGBUS_USPTP_5, + A7XX_DBGBUS_USPTP_6, + A7XX_DBGBUS_USPTP_7, +}; + +static const struct gen7_shader_block gen7_0_0_shader_blocks[] = { + {A7XX_TP0_TMO_DATA, 0x200, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_TP0_SMO_DATA, 0x80, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_TP0_MIPMAP_BASE_DATA, 0x3c0, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_INST_DATA, 0x800, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_INST_DATA_1, 0x800, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_0_DATA, 0x800, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_1_DATA, 0x800, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_2_DATA, 0x800, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_3_DATA, 0x800, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_4_DATA, 0x800, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_5_DATA, 0x800, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_6_DATA, 0x800, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_7_DATA, 0x800, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_CB_RAM, 0x390, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_INST_TAG, 0x90, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_INST_DATA_2, 0x200, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_TMO_TAG, 0x80, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_SMO_TAG, 0x80, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_STATE_DATA, 0x40, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_HWAVE_RAM, 0x100, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_L0_INST_BUF, 0x50, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_8_DATA, 0x800, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_9_DATA, 0x800, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_10_DATA, 0x800, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_11_DATA, 0x800, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_12_DATA, 0x200, 4, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_HLSQ_CVS_BE_CTXT_BUF_RAM_TAG, 0x10, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CVS_BE_CTXT_BUF_RAM_TAG, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CPS_BE_CTXT_BUF_RAM_TAG, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CVS_BE_CTXT_BUF_RAM, 0x300, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CVS_BE_CTXT_BUF_RAM, 0x300, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CPS_BE_CTXT_BUF_RAM, 0x300, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CVS_RAM, 0x1c0, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CVS_RAM, 0x1c0, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CPS_RAM, 0x300, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CPS_RAM, 0x300, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CVS_RAM_TAG, 0x40, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CVS_RAM_TAG, 0x40, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CPS_RAM_TAG, 0x40, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CPS_RAM_TAG, 0x40, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_ICB_CVS_CB_BASE_TAG, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_ICB_CVS_CB_BASE_TAG, 0x10, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_ICB_CPS_CB_BASE_TAG, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_ICB_CPS_CB_BASE_TAG, 0x10, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CVS_MISC_RAM, 0x280, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CVS_MISC_RAM, 0x280, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CPS_MISC_RAM, 0x800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CPS_MISC_RAM, 0x800, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CPS_MISC_RAM_1, 0x200, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INST_RAM, 0x800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INST_RAM, 0x800, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INST_RAM, 0x800, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CVS_CONST_RAM, 0x800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CVS_CONST_RAM, 0x800, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CPS_CONST_RAM, 0x800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CPS_CONST_RAM, 0x800, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CVS_MISC_RAM_TAG, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CVS_MISC_RAM_TAG, 0x10, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CPS_MISC_RAM_TAG, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CPS_MISC_RAM_TAG, 0x10, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INST_RAM_TAG, 0x80, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INST_RAM_TAG, 0x80, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INST_RAM_TAG, 0x80, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CVS_CONST_RAM_TAG, 0x64, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CVS_CONST_RAM_TAG, 0x64, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CPS_CONST_RAM_TAG, 0x64, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CPS_CONST_RAM_TAG, 0x64, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INST_RAM_1, 0x800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_STPROC_META, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_BV_BE_META, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_BV_BE_META, 0x10, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_DATAPATH_META, 0x20, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_FRONTEND_META, 0x40, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_FRONTEND_META, 0x40, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_FRONTEND_META, 0x40, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INDIRECT_META, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_BACKEND_META, 0x40, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_BACKEND_META, 0x40, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_BACKEND_META, 0x40, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, +}; + +static const u32 gen7_0_0_pre_crashdumper_gpu_registers[] = { + 0x00210, 0x00210, 0x00212, 0x00213, 0x03c00, 0x03c0b, 0x03c40, 0x03c42, + 0x03c45, 0x03c47, 0x03c49, 0x03c4a, 0x03cc0, 0x03cd1, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_pre_crashdumper_gpu_registers), 8)); + +static const u32 gen7_0_0_post_crashdumper_registers[] = { + 0x00535, 0x00535, 0x0f400, 0x0f400, 0x0f800, 0x0f803, 0x0fc00, 0x0fc01, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_post_crashdumper_registers), 8)); + +static const u32 gen7_0_0_gpu_registers[] = { + 0x00000, 0x00000, 0x00002, 0x00002, 0x00011, 0x00012, 0x00016, 0x0001b, + 0x0001f, 0x00032, 0x00038, 0x0003c, 0x00042, 0x00042, 0x00044, 0x00044, + 0x00047, 0x00047, 0x00049, 0x0004a, 0x0004c, 0x0004c, 0x00050, 0x00050, + 0x00056, 0x00056, 0x00073, 0x00075, 0x000ad, 0x000ae, 0x000b0, 0x000b0, + 0x000b4, 0x000b4, 0x000b8, 0x000b8, 0x000bc, 0x000bc, 0x000c0, 0x000c0, + 0x000c4, 0x000c4, 0x000c8, 0x000c8, 0x000cc, 0x000cc, 0x000d0, 0x000d0, + 0x000d4, 0x000d4, 0x000d8, 0x000d8, 0x000dc, 0x000dc, 0x000e0, 0x000e0, + 0x000e4, 0x000e4, 0x000e8, 0x000e8, 0x000ec, 0x000ec, 0x000f0, 0x000f0, + 0x000f4, 0x000f4, 0x000f8, 0x000f8, 0x00100, 0x00100, 0x00104, 0x0010b, + 0x0010f, 0x0011d, 0x0012f, 0x0012f, 0x00200, 0x0020d, 0x00211, 0x00211, + 0x00215, 0x00243, 0x00260, 0x00268, 0x00272, 0x00274, 0x00281, 0x0028d, + 0x00300, 0x00401, 0x00410, 0x00451, 0x00460, 0x004a3, 0x004c0, 0x004d1, + 0x00500, 0x00500, 0x00507, 0x0050b, 0x0050f, 0x0050f, 0x00511, 0x00511, + 0x00533, 0x00534, 0x00536, 0x00536, 0x00540, 0x00555, 0x00564, 0x00567, + 0x00574, 0x00577, 0x005fb, 0x005ff, 0x00800, 0x00808, 0x00810, 0x00813, + 0x00820, 0x00821, 0x00823, 0x00827, 0x00830, 0x00834, 0x0083f, 0x00841, + 0x00843, 0x00847, 0x0084f, 0x00886, 0x008a0, 0x008ab, 0x008c0, 0x008c0, + 0x008c4, 0x008c5, 0x008d0, 0x008dd, 0x008e0, 0x008e6, 0x008f0, 0x008f3, + 0x00900, 0x00903, 0x00908, 0x00911, 0x00928, 0x0093e, 0x00942, 0x0094d, + 0x00980, 0x00984, 0x0098d, 0x0098f, 0x009b0, 0x009b4, 0x009c2, 0x009c9, + 0x009ce, 0x009d7, 0x009e0, 0x009e7, 0x00a00, 0x00a00, 0x00a02, 0x00a03, + 0x00a10, 0x00a4f, 0x00a61, 0x00a9f, 0x00ad0, 0x00adb, 0x00b00, 0x00b31, + 0x00b35, 0x00b3c, 0x00b40, 0x00b40, 0x00c00, 0x00c00, 0x00c02, 0x00c04, + 0x00c06, 0x00c06, 0x00c10, 0x00cd9, 0x00ce0, 0x00d0c, 0x00df0, 0x00df4, + 0x00e01, 0x00e02, 0x00e07, 0x00e0e, 0x00e10, 0x00e13, 0x00e17, 0x00e19, + 0x00e1b, 0x00e2b, 0x00e30, 0x00e32, 0x00e38, 0x00e3c, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_gpu_registers), 8)); + +static const u32 gen7_0_0_gmu_registers[] = { + 0x10001, 0x10001, 0x10003, 0x10003, 0x10401, 0x10401, 0x10403, 0x10403, + 0x10801, 0x10801, 0x10803, 0x10803, 0x10c01, 0x10c01, 0x10c03, 0x10c03, + 0x11001, 0x11001, 0x11003, 0x11003, 0x11401, 0x11401, 0x11403, 0x11403, + 0x11801, 0x11801, 0x11803, 0x11803, 0x11c01, 0x11c01, 0x11c03, 0x11c03, + 0x1f400, 0x1f40d, 0x1f40f, 0x1f411, 0x1f500, 0x1f500, 0x1f507, 0x1f507, + 0x1f509, 0x1f50b, 0x1f800, 0x1f804, 0x1f807, 0x1f808, 0x1f80b, 0x1f80c, + 0x1f80f, 0x1f80f, 0x1f811, 0x1f811, 0x1f813, 0x1f817, 0x1f819, 0x1f81c, + 0x1f824, 0x1f82a, 0x1f82d, 0x1f830, 0x1f840, 0x1f853, 0x1f860, 0x1f860, + 0x1f870, 0x1f879, 0x1f87f, 0x1f87f, 0x1f888, 0x1f889, 0x1f8a0, 0x1f8a2, + 0x1f8a4, 0x1f8af, 0x1f8c0, 0x1f8c1, 0x1f8c3, 0x1f8c4, 0x1f8d0, 0x1f8d0, + 0x1f8ec, 0x1f8ec, 0x1f8f0, 0x1f8f1, 0x1f910, 0x1f914, 0x1f920, 0x1f921, + 0x1f924, 0x1f925, 0x1f928, 0x1f929, 0x1f92c, 0x1f92d, 0x1f940, 0x1f940, + 0x1f942, 0x1f944, 0x1f948, 0x1f94a, 0x1f94f, 0x1f951, 0x1f958, 0x1f95a, + 0x1f95d, 0x1f95d, 0x1f962, 0x1f962, 0x1f964, 0x1f96b, 0x1f970, 0x1f979, + 0x1f980, 0x1f981, 0x1f984, 0x1f986, 0x1f992, 0x1f993, 0x1f996, 0x1f99e, + 0x1f9c0, 0x1f9c0, 0x1f9c5, 0x1f9d4, 0x1f9f0, 0x1f9f1, 0x1f9f8, 0x1f9fa, + 0x1fa00, 0x1fa03, 0x20000, 0x20005, 0x20008, 0x2000c, 0x20010, 0x20012, + 0x20018, 0x20018, 0x20020, 0x20023, 0x20030, 0x20031, 0x23801, 0x23801, + 0x23803, 0x23803, 0x23805, 0x23805, 0x23807, 0x23807, 0x23809, 0x23809, + 0x2380b, 0x2380b, 0x2380d, 0x2380d, 0x2380f, 0x2380f, 0x23811, 0x23811, + 0x23813, 0x23813, 0x23815, 0x23815, 0x23817, 0x23817, 0x23819, 0x23819, + 0x2381b, 0x2381b, 0x2381d, 0x2381d, 0x2381f, 0x23820, 0x23822, 0x23822, + 0x23824, 0x23824, 0x23826, 0x23826, 0x23828, 0x23828, 0x2382a, 0x2382a, + 0x2382c, 0x2382c, 0x2382e, 0x2382e, 0x23830, 0x23830, 0x23832, 0x23832, + 0x23834, 0x23834, 0x23836, 0x23836, 0x23838, 0x23838, 0x2383a, 0x2383a, + 0x2383c, 0x2383c, 0x2383e, 0x2383e, 0x23840, 0x23847, 0x23b00, 0x23b01, + 0x23b03, 0x23b03, 0x23b05, 0x23b0e, 0x23b10, 0x23b13, 0x23b15, 0x23b16, + 0x23b20, 0x23b20, 0x23b28, 0x23b28, 0x23b30, 0x23b30, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_gmu_registers), 8)); + +static const u32 gen7_0_0_gmugx_registers[] = { + 0x1a400, 0x1a41f, 0x1a440, 0x1a45f, 0x1a480, 0x1a49f, 0x1a4c0, 0x1a4df, + 0x1a500, 0x1a51f, 0x1a540, 0x1a55f, 0x1a580, 0x1a59f, 0x1a5c0, 0x1a5df, + 0x1a780, 0x1a781, 0x1a783, 0x1a785, 0x1a787, 0x1a789, 0x1a78b, 0x1a78d, + 0x1a78f, 0x1a791, 0x1a793, 0x1a795, 0x1a797, 0x1a799, 0x1a79b, 0x1a79b, + 0x1a7c0, 0x1a7c1, 0x1a7c4, 0x1a7c5, 0x1a7c8, 0x1a7c9, 0x1a7cc, 0x1a7cd, + 0x1a7d0, 0x1a7d1, 0x1a7d4, 0x1a7d5, 0x1a7d8, 0x1a7d9, 0x1a7fc, 0x1a7fd, + 0x1a800, 0x1a802, 0x1a804, 0x1a804, 0x1a816, 0x1a816, 0x1a81e, 0x1a81e, + 0x1a826, 0x1a826, 0x1a82e, 0x1a82e, 0x1a836, 0x1a836, 0x1a83e, 0x1a83e, + 0x1a846, 0x1a846, 0x1a860, 0x1a862, 0x1a864, 0x1a867, 0x1a870, 0x1a870, + 0x1a883, 0x1a884, 0x1a8c0, 0x1a8c2, 0x1a8c4, 0x1a8c7, 0x1a8d0, 0x1a8d3, + 0x1a900, 0x1a92b, 0x1a940, 0x1a940, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_gmugx_registers), 8)); + +static const u32 gen7_0_0_noncontext_pipe_br_registers[] = { + 0x00887, 0x0088c, 0x08600, 0x08600, 0x08602, 0x08602, 0x08610, 0x0861b, + 0x08620, 0x08620, 0x08630, 0x08630, 0x08637, 0x08639, 0x08640, 0x08640, + 0x09600, 0x09600, 0x09602, 0x09603, 0x0960a, 0x09616, 0x09624, 0x0963a, + 0x09640, 0x09640, 0x09e00, 0x09e00, 0x09e02, 0x09e07, 0x09e0a, 0x09e16, + 0x09e19, 0x09e19, 0x09e1c, 0x09e1c, 0x09e20, 0x09e25, 0x09e30, 0x09e31, + 0x09e40, 0x09e51, 0x09e64, 0x09e64, 0x09e70, 0x09e72, 0x09e78, 0x09e79, + 0x09e80, 0x09fff, 0x0a600, 0x0a600, 0x0a603, 0x0a603, 0x0a610, 0x0a61f, + 0x0a630, 0x0a631, 0x0a638, 0x0a638, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_noncontext_pipe_br_registers), 8)); + +static const u32 gen7_0_0_noncontext_pipe_bv_registers[] = { + 0x00887, 0x0088c, 0x08600, 0x08600, 0x08602, 0x08602, 0x08610, 0x0861b, + 0x08620, 0x08620, 0x08630, 0x08630, 0x08637, 0x08639, 0x08640, 0x08640, + 0x09600, 0x09600, 0x09602, 0x09603, 0x0960a, 0x09616, 0x09624, 0x0963a, + 0x09640, 0x09640, 0x09e00, 0x09e00, 0x09e02, 0x09e07, 0x09e0a, 0x09e16, + 0x09e19, 0x09e19, 0x09e1c, 0x09e1c, 0x09e20, 0x09e25, 0x09e30, 0x09e31, + 0x09e40, 0x09e51, 0x09e64, 0x09e64, 0x09e70, 0x09e72, 0x09e78, 0x09e79, + 0x09e80, 0x09fff, 0x0a600, 0x0a600, 0x0a603, 0x0a603, 0x0a610, 0x0a61f, + 0x0a630, 0x0a631, 0x0a638, 0x0a638, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_noncontext_pipe_bv_registers), 8)); + +static const u32 gen7_0_0_noncontext_pipe_lpac_registers[] = { + 0x00887, 0x0088c, 0x00f80, 0x00f80, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_noncontext_pipe_lpac_registers), 8)); + +static const u32 gen7_0_0_noncontext_rb_rac_pipe_br_registers[] = { + 0x08e10, 0x08e1c, 0x08e20, 0x08e25, 0x08e51, 0x08e5a, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_noncontext_rb_rac_pipe_br_registers), 8)); + +static const u32 gen7_0_0_noncontext_rb_rbp_pipe_br_registers[] = { + 0x08e01, 0x08e01, 0x08e04, 0x08e04, 0x08e06, 0x08e09, 0x08e0c, 0x08e0c, + 0x08e28, 0x08e28, 0x08e2c, 0x08e35, 0x08e3b, 0x08e3f, 0x08e50, 0x08e50, + 0x08e5b, 0x08e5d, 0x08e5f, 0x08e5f, 0x08e61, 0x08e61, 0x08e63, 0x08e65, + 0x08e68, 0x08e68, 0x08e70, 0x08e79, 0x08e80, 0x08e8f, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_noncontext_rb_rbp_pipe_br_registers), 8)); + +/* Block: GRAS Cluster: A7XX_CLUSTER_GRAS Pipeline: PIPE_BR */ +static const u32 gen7_0_0_gras_cluster_gras_pipe_br_registers[] = { + 0x08000, 0x08008, 0x08010, 0x08092, 0x08094, 0x08099, 0x0809b, 0x0809d, + 0x080a0, 0x080a7, 0x080af, 0x080f1, 0x080f4, 0x080f6, 0x080f8, 0x080fa, + 0x08100, 0x08107, 0x08109, 0x0810b, 0x08110, 0x08110, 0x08120, 0x0813f, + 0x08400, 0x08406, 0x0840a, 0x0840b, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_gras_cluster_gras_pipe_br_registers), 8)); + +/* Block: GRAS Cluster: A7XX_CLUSTER_GRAS Pipeline: PIPE_BV */ +static const u32 gen7_0_0_gras_cluster_gras_pipe_bv_registers[] = { + 0x08000, 0x08008, 0x08010, 0x08092, 0x08094, 0x08099, 0x0809b, 0x0809d, + 0x080a0, 0x080a7, 0x080af, 0x080f1, 0x080f4, 0x080f6, 0x080f8, 0x080fa, + 0x08100, 0x08107, 0x08109, 0x0810b, 0x08110, 0x08110, 0x08120, 0x0813f, + 0x08400, 0x08406, 0x0840a, 0x0840b, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_gras_cluster_gras_pipe_bv_registers), 8)); + +/* Block: PC Cluster: A7XX_CLUSTER_FE Pipeline: PIPE_BR */ +static const u32 gen7_0_0_pc_cluster_fe_pipe_br_registers[] = { + 0x09800, 0x09804, 0x09806, 0x0980a, 0x09810, 0x09811, 0x09884, 0x09886, + 0x09b00, 0x09b08, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_pc_cluster_fe_pipe_br_registers), 8)); + +/* Block: PC Cluster: A7XX_CLUSTER_FE Pipeline: PIPE_BV */ +static const u32 gen7_0_0_pc_cluster_fe_pipe_bv_registers[] = { + 0x09800, 0x09804, 0x09806, 0x0980a, 0x09810, 0x09811, 0x09884, 0x09886, + 0x09b00, 0x09b08, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_pc_cluster_fe_pipe_bv_registers), 8)); + +/* Block: RB_RAC Cluster: A7XX_CLUSTER_PS Pipeline: PIPE_BR */ +static const u32 gen7_0_0_rb_rac_cluster_ps_pipe_br_registers[] = { + 0x08802, 0x08802, 0x08804, 0x08806, 0x08809, 0x0880a, 0x0880e, 0x08811, + 0x08818, 0x0881e, 0x08821, 0x08821, 0x08823, 0x08826, 0x08829, 0x08829, + 0x0882b, 0x0882e, 0x08831, 0x08831, 0x08833, 0x08836, 0x08839, 0x08839, + 0x0883b, 0x0883e, 0x08841, 0x08841, 0x08843, 0x08846, 0x08849, 0x08849, + 0x0884b, 0x0884e, 0x08851, 0x08851, 0x08853, 0x08856, 0x08859, 0x08859, + 0x0885b, 0x0885e, 0x08860, 0x08864, 0x08870, 0x08870, 0x08873, 0x08876, + 0x08878, 0x08879, 0x08882, 0x08885, 0x08887, 0x08889, 0x08891, 0x08891, + 0x08898, 0x08898, 0x088c0, 0x088c1, 0x088e5, 0x088e5, 0x088f4, 0x088f5, + 0x08a00, 0x08a05, 0x08a10, 0x08a15, 0x08a20, 0x08a25, 0x08a30, 0x08a35, + 0x08c00, 0x08c01, 0x08c18, 0x08c1f, 0x08c26, 0x08c34, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_rb_rac_cluster_ps_pipe_br_registers), 8)); + +/* Block: RB_RBP Cluster: A7XX_CLUSTER_PS Pipeline: PIPE_BR */ +static const u32 gen7_0_0_rb_rbp_cluster_ps_pipe_br_registers[] = { + 0x08800, 0x08801, 0x08803, 0x08803, 0x0880b, 0x0880d, 0x08812, 0x08812, + 0x08820, 0x08820, 0x08822, 0x08822, 0x08827, 0x08828, 0x0882a, 0x0882a, + 0x0882f, 0x08830, 0x08832, 0x08832, 0x08837, 0x08838, 0x0883a, 0x0883a, + 0x0883f, 0x08840, 0x08842, 0x08842, 0x08847, 0x08848, 0x0884a, 0x0884a, + 0x0884f, 0x08850, 0x08852, 0x08852, 0x08857, 0x08858, 0x0885a, 0x0885a, + 0x0885f, 0x0885f, 0x08865, 0x08865, 0x08871, 0x08872, 0x08877, 0x08877, + 0x08880, 0x08881, 0x08886, 0x08886, 0x08890, 0x08890, 0x088d0, 0x088e4, + 0x088e8, 0x088ea, 0x088f0, 0x088f0, 0x08900, 0x0891a, 0x08927, 0x08928, + 0x08c17, 0x08c17, 0x08c20, 0x08c25, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_rb_rbp_cluster_ps_pipe_br_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_PS Pipeline: PIPE_BR Location: HLSQ_STATE */ +static const u32 gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_state_registers[] = { + 0x0a980, 0x0a980, 0x0a982, 0x0a984, 0x0a99e, 0x0a99e, 0x0a9a7, 0x0a9a7, + 0x0a9aa, 0x0a9aa, 0x0a9ae, 0x0a9b0, 0x0a9b3, 0x0a9b5, 0x0a9ba, 0x0a9ba, + 0x0a9bc, 0x0a9bc, 0x0a9c4, 0x0a9c4, 0x0a9cd, 0x0a9cd, 0x0a9e0, 0x0a9fc, + 0x0aa00, 0x0aa00, 0x0aa30, 0x0aa31, 0x0aa40, 0x0aabf, 0x0ab00, 0x0ab03, + 0x0ab05, 0x0ab05, 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_state_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_PS Pipeline: PIPE_LPAC Location: HLSQ_STATE */ +static const u32 gen7_0_0_sp_cluster_sp_ps_pipe_lpac_hlsq_state_registers[] = { + 0x0a9b0, 0x0a9b0, 0x0a9b3, 0x0a9b5, 0x0a9ba, 0x0a9ba, 0x0a9bc, 0x0a9bc, + 0x0a9c4, 0x0a9c4, 0x0a9cd, 0x0a9cd, 0x0a9e2, 0x0a9e3, 0x0a9e6, 0x0a9fc, + 0x0aa00, 0x0aa00, 0x0aa31, 0x0aa31, 0x0ab00, 0x0ab01, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_ps_pipe_lpac_hlsq_state_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_PS Pipeline: PIPE_BR Location: HLSQ_DP */ +static const u32 gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_dp_registers[] = { + 0x0a9b1, 0x0a9b1, 0x0a9c6, 0x0a9cb, 0x0a9d4, 0x0a9df, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_dp_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_PS Pipeline: PIPE_LPAC Location: HLSQ_DP */ +static const u32 gen7_0_0_sp_cluster_sp_ps_pipe_lpac_hlsq_dp_registers[] = { + 0x0a9b1, 0x0a9b1, 0x0a9d4, 0x0a9df, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_ps_pipe_lpac_hlsq_dp_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_PS Pipeline: PIPE_BR Location: SP_TOP */ +static const u32 gen7_0_0_sp_cluster_sp_ps_pipe_br_sp_top_registers[] = { + 0x0a980, 0x0a980, 0x0a982, 0x0a984, 0x0a99e, 0x0a9a2, 0x0a9a7, 0x0a9a8, + 0x0a9aa, 0x0a9aa, 0x0a9ae, 0x0a9ae, 0x0a9b0, 0x0a9b1, 0x0a9b3, 0x0a9b5, + 0x0a9ba, 0x0a9bc, 0x0a9e0, 0x0a9f9, 0x0aa00, 0x0aa00, 0x0ab00, 0x0ab00, + 0x0ab02, 0x0ab02, 0x0ab04, 0x0ab05, 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_ps_pipe_br_sp_top_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_PS Pipeline: PIPE_LPAC Location: SP_TOP */ +static const u32 gen7_0_0_sp_cluster_sp_ps_pipe_lpac_sp_top_registers[] = { + 0x0a9b0, 0x0a9b1, 0x0a9b3, 0x0a9b5, 0x0a9ba, 0x0a9bc, 0x0a9e2, 0x0a9e3, + 0x0a9e6, 0x0a9f9, 0x0aa00, 0x0aa00, 0x0ab00, 0x0ab00, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_ps_pipe_lpac_sp_top_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_PS Pipeline: PIPE_BR Location: uSPTP */ +static const u32 gen7_0_0_sp_cluster_sp_ps_pipe_br_usptp_registers[] = { + 0x0a980, 0x0a982, 0x0a985, 0x0a9a6, 0x0a9a8, 0x0a9a9, 0x0a9ab, 0x0a9ae, + 0x0a9b0, 0x0a9b3, 0x0a9b6, 0x0a9b9, 0x0a9bb, 0x0a9bf, 0x0a9c2, 0x0a9c3, + 0x0a9cd, 0x0a9cd, 0x0a9d0, 0x0a9d3, 0x0aa30, 0x0aa31, 0x0aa40, 0x0aabf, + 0x0ab00, 0x0ab05, 0x0ab21, 0x0ab22, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_ps_pipe_br_usptp_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_PS Pipeline: PIPE_LPAC Location: uSPTP */ +static const u32 gen7_0_0_sp_cluster_sp_ps_pipe_lpac_usptp_registers[] = { + 0x0a9b0, 0x0a9b3, 0x0a9b6, 0x0a9b9, 0x0a9bb, 0x0a9be, 0x0a9c2, 0x0a9c3, + 0x0a9cd, 0x0a9cd, 0x0a9d0, 0x0a9d3, 0x0aa31, 0x0aa31, 0x0ab00, 0x0ab01, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_ps_pipe_lpac_usptp_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_VS Pipeline: PIPE_BR Location: HLSQ_STATE */ +static const u32 gen7_0_0_sp_cluster_sp_vs_pipe_br_hlsq_state_registers[] = { + 0x0a800, 0x0a800, 0x0a81b, 0x0a81d, 0x0a822, 0x0a822, 0x0a824, 0x0a824, + 0x0a827, 0x0a82a, 0x0a830, 0x0a830, 0x0a833, 0x0a835, 0x0a83a, 0x0a83a, + 0x0a83c, 0x0a83c, 0x0a83f, 0x0a840, 0x0a85b, 0x0a85d, 0x0a862, 0x0a862, + 0x0a864, 0x0a864, 0x0a867, 0x0a867, 0x0a870, 0x0a870, 0x0a88c, 0x0a88e, + 0x0a893, 0x0a893, 0x0a895, 0x0a895, 0x0a898, 0x0a898, 0x0a89a, 0x0a89d, + 0x0a8a0, 0x0a8af, 0x0a8c0, 0x0a8c3, 0x0ab00, 0x0ab03, 0x0ab05, 0x0ab05, + 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_vs_pipe_br_hlsq_state_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_VS Pipeline: PIPE_BV Location: HLSQ_STATE */ +static const u32 gen7_0_0_sp_cluster_sp_vs_pipe_bv_hlsq_state_registers[] = { + 0x0a800, 0x0a800, 0x0a81b, 0x0a81d, 0x0a822, 0x0a822, 0x0a824, 0x0a824, + 0x0a827, 0x0a82a, 0x0a830, 0x0a830, 0x0a833, 0x0a835, 0x0a83a, 0x0a83a, + 0x0a83c, 0x0a83c, 0x0a83f, 0x0a840, 0x0a85b, 0x0a85d, 0x0a862, 0x0a862, + 0x0a864, 0x0a864, 0x0a867, 0x0a867, 0x0a870, 0x0a870, 0x0a88c, 0x0a88e, + 0x0a893, 0x0a893, 0x0a895, 0x0a895, 0x0a898, 0x0a898, 0x0a89a, 0x0a89d, + 0x0a8a0, 0x0a8af, 0x0a8c0, 0x0a8c3, 0x0ab00, 0x0ab02, 0x0ab0a, 0x0ab1b, + 0x0ab20, 0x0ab20, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_vs_pipe_bv_hlsq_state_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_VS Pipeline: PIPE_BR Location: SP_TOP */ +static const u32 gen7_0_0_sp_cluster_sp_vs_pipe_br_sp_top_registers[] = { + 0x0a800, 0x0a800, 0x0a81c, 0x0a81d, 0x0a822, 0x0a824, 0x0a830, 0x0a831, + 0x0a834, 0x0a835, 0x0a83a, 0x0a83c, 0x0a840, 0x0a840, 0x0a85c, 0x0a85d, + 0x0a862, 0x0a864, 0x0a870, 0x0a871, 0x0a88d, 0x0a88e, 0x0a893, 0x0a895, + 0x0a8a0, 0x0a8af, 0x0ab00, 0x0ab00, 0x0ab02, 0x0ab02, 0x0ab04, 0x0ab05, + 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_vs_pipe_br_sp_top_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_VS Pipeline: PIPE_BV Location: SP_TOP */ +static const u32 gen7_0_0_sp_cluster_sp_vs_pipe_bv_sp_top_registers[] = { + 0x0a800, 0x0a800, 0x0a81c, 0x0a81d, 0x0a822, 0x0a824, 0x0a830, 0x0a831, + 0x0a834, 0x0a835, 0x0a83a, 0x0a83c, 0x0a840, 0x0a840, 0x0a85c, 0x0a85d, + 0x0a862, 0x0a864, 0x0a870, 0x0a871, 0x0a88d, 0x0a88e, 0x0a893, 0x0a895, + 0x0a8a0, 0x0a8af, 0x0ab00, 0x0ab00, 0x0ab02, 0x0ab02, 0x0ab0a, 0x0ab1b, + 0x0ab20, 0x0ab20, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_vs_pipe_bv_sp_top_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_VS Pipeline: PIPE_BR Location: uSPTP */ +static const u32 gen7_0_0_sp_cluster_sp_vs_pipe_br_usptp_registers[] = { + 0x0a800, 0x0a81b, 0x0a81e, 0x0a821, 0x0a823, 0x0a827, 0x0a830, 0x0a833, + 0x0a836, 0x0a839, 0x0a83b, 0x0a85b, 0x0a85e, 0x0a861, 0x0a863, 0x0a867, + 0x0a870, 0x0a88c, 0x0a88f, 0x0a892, 0x0a894, 0x0a898, 0x0a8c0, 0x0a8c3, + 0x0ab00, 0x0ab05, 0x0ab21, 0x0ab22, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_vs_pipe_br_usptp_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_VS Pipeline: PIPE_BV Location: uSPTP */ +static const u32 gen7_0_0_sp_cluster_sp_vs_pipe_bv_usptp_registers[] = { + 0x0a800, 0x0a81b, 0x0a81e, 0x0a821, 0x0a823, 0x0a827, 0x0a830, 0x0a833, + 0x0a836, 0x0a839, 0x0a83b, 0x0a85b, 0x0a85e, 0x0a861, 0x0a863, 0x0a867, + 0x0a870, 0x0a88c, 0x0a88f, 0x0a892, 0x0a894, 0x0a898, 0x0a8c0, 0x0a8c3, + 0x0ab00, 0x0ab02, 0x0ab21, 0x0ab22, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_vs_pipe_bv_usptp_registers), 8)); + +/* Block: TPL1 Cluster: A7XX_CLUSTER_SP_PS Pipeline: PIPE_BR */ +static const u32 gen7_0_0_tpl1_cluster_sp_ps_pipe_br_registers[] = { + 0x0b180, 0x0b183, 0x0b190, 0x0b195, 0x0b2c0, 0x0b2d5, 0x0b300, 0x0b307, + 0x0b309, 0x0b309, 0x0b310, 0x0b310, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_tpl1_cluster_sp_ps_pipe_br_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_PS Pipeline: PIPE_BV Location: HLSQ_STATE */ +static const u32 gen7_0_0_sp_cluster_sp_ps_pipe_bv_hlsq_state_registers[] = { + 0x0ab00, 0x0ab02, 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_ps_pipe_bv_hlsq_state_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_PS Pipeline: PIPE_BV Location: SP_TOP */ +static const u32 gen7_0_0_sp_cluster_sp_ps_pipe_bv_sp_top_registers[] = { + 0x0ab00, 0x0ab00, 0x0ab02, 0x0ab02, 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_ps_pipe_bv_sp_top_registers), 8)); + +/* Block: SP Cluster: A7XX_CLUSTER_SP_PS Pipeline: PIPE_BV Location: uSPTP */ +static const u32 gen7_0_0_sp_cluster_sp_ps_pipe_bv_usptp_registers[] = { + 0x0ab00, 0x0ab02, 0x0ab21, 0x0ab22, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_cluster_sp_ps_pipe_bv_usptp_registers), 8)); + +/* Block: TPL1 Cluster: A7XX_CLUSTER_SP_PS Pipeline: PIPE_BV */ +static const u32 gen7_0_0_tpl1_cluster_sp_ps_pipe_bv_registers[] = { + 0x0b300, 0x0b307, 0x0b309, 0x0b309, 0x0b310, 0x0b310, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_tpl1_cluster_sp_ps_pipe_bv_registers), 8)); + +/* Block: TPL1 Cluster: A7XX_CLUSTER_SP_PS Pipeline: PIPE_LPAC */ +static const u32 gen7_0_0_tpl1_cluster_sp_ps_pipe_lpac_registers[] = { + 0x0b180, 0x0b181, 0x0b300, 0x0b301, 0x0b307, 0x0b307, 0x0b309, 0x0b309, + 0x0b310, 0x0b310, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_tpl1_cluster_sp_ps_pipe_lpac_registers), 8)); + +/* Block: TPL1 Cluster: A7XX_CLUSTER_SP_VS Pipeline: PIPE_BR */ +static const u32 gen7_0_0_tpl1_cluster_sp_vs_pipe_br_registers[] = { + 0x0b300, 0x0b307, 0x0b309, 0x0b309, 0x0b310, 0x0b310, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_tpl1_cluster_sp_vs_pipe_br_registers), 8)); + +/* Block: TPL1 Cluster: A7XX_CLUSTER_SP_VS Pipeline: PIPE_BV */ +static const u32 gen7_0_0_tpl1_cluster_sp_vs_pipe_bv_registers[] = { + 0x0b300, 0x0b307, 0x0b309, 0x0b309, 0x0b310, 0x0b310, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_tpl1_cluster_sp_vs_pipe_bv_registers), 8)); + +/* Block: VFD Cluster: A7XX_CLUSTER_FE Pipeline: PIPE_BR */ +static const u32 gen7_0_0_vfd_cluster_fe_pipe_br_registers[] = { + 0x0a000, 0x0a009, 0x0a00e, 0x0a0ef, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_vfd_cluster_fe_pipe_br_registers), 8)); + +/* Block: VFD Cluster: A7XX_CLUSTER_FE Pipeline: PIPE_BV */ +static const u32 gen7_0_0_vfd_cluster_fe_pipe_bv_registers[] = { + 0x0a000, 0x0a009, 0x0a00e, 0x0a0ef, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_vfd_cluster_fe_pipe_bv_registers), 8)); + +/* Block: VPC Cluster: A7XX_CLUSTER_FE Pipeline: PIPE_BR */ +static const u32 gen7_0_0_vpc_cluster_fe_pipe_br_registers[] = { + 0x09300, 0x09307, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_vpc_cluster_fe_pipe_br_registers), 8)); + +/* Block: VPC Cluster: A7XX_CLUSTER_FE Pipeline: PIPE_BV */ +static const u32 gen7_0_0_vpc_cluster_fe_pipe_bv_registers[] = { + 0x09300, 0x09307, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_vpc_cluster_fe_pipe_bv_registers), 8)); + +/* Block: VPC Cluster: A7XX_CLUSTER_PC_VS Pipeline: PIPE_BR */ +static const u32 gen7_0_0_vpc_cluster_pc_vs_pipe_br_registers[] = { + 0x09101, 0x0910c, 0x09300, 0x09307, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_vpc_cluster_pc_vs_pipe_br_registers), 8)); + +/* Block: VPC Cluster: A7XX_CLUSTER_PC_VS Pipeline: PIPE_BV */ +static const u32 gen7_0_0_vpc_cluster_pc_vs_pipe_bv_registers[] = { + 0x09101, 0x0910c, 0x09300, 0x09307, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_vpc_cluster_pc_vs_pipe_bv_registers), 8)); + +/* Block: VPC Cluster: A7XX_CLUSTER_VPC_PS Pipeline: PIPE_BR */ +static const u32 gen7_0_0_vpc_cluster_vpc_ps_pipe_br_registers[] = { + 0x09200, 0x0920f, 0x09212, 0x09216, 0x09218, 0x09236, 0x09300, 0x09307, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_vpc_cluster_vpc_ps_pipe_br_registers), 8)); + +/* Block: VPC Cluster: A7XX_CLUSTER_VPC_PS Pipeline: PIPE_BV */ +static const u32 gen7_0_0_vpc_cluster_vpc_ps_pipe_bv_registers[] = { + 0x09200, 0x0920f, 0x09212, 0x09216, 0x09218, 0x09236, 0x09300, 0x09307, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_vpc_cluster_vpc_ps_pipe_bv_registers), 8)); + +/* Block: SP Cluster: noncontext Pipeline: PIPE_BR Location: HLSQ_STATE */ +static const u32 gen7_0_0_sp_noncontext_pipe_br_hlsq_state_registers[] = { + 0x0ae52, 0x0ae52, 0x0ae60, 0x0ae67, 0x0ae69, 0x0ae73, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_noncontext_pipe_br_hlsq_state_registers), 8)); + +/* Block: SP Cluster: noncontext Pipeline: PIPE_BR Location: SP_TOP */ +static const u32 gen7_0_0_sp_noncontext_pipe_br_sp_top_registers[] = { + 0x0ae00, 0x0ae00, 0x0ae02, 0x0ae04, 0x0ae06, 0x0ae09, 0x0ae0c, 0x0ae0c, + 0x0ae0f, 0x0ae0f, 0x0ae28, 0x0ae2b, 0x0ae35, 0x0ae35, 0x0ae3a, 0x0ae3f, + 0x0ae50, 0x0ae52, 0x0ae80, 0x0aea3, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_noncontext_pipe_br_sp_top_registers), 8)); + +/* Block: SP Cluster: noncontext Pipeline: PIPE_BR Location: uSPTP */ +static const u32 gen7_0_0_sp_noncontext_pipe_br_usptp_registers[] = { + 0x0ae00, 0x0ae00, 0x0ae02, 0x0ae04, 0x0ae06, 0x0ae09, 0x0ae0c, 0x0ae0c, + 0x0ae0f, 0x0ae0f, 0x0ae30, 0x0ae32, 0x0ae35, 0x0ae35, 0x0ae3a, 0x0ae3b, + 0x0ae3e, 0x0ae3f, 0x0ae50, 0x0ae52, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_noncontext_pipe_br_usptp_registers), 8)); + +/* Block: SP Cluster: noncontext Pipeline: PIPE_LPAC Location: HLSQ_STATE */ +static const u32 gen7_0_0_sp_noncontext_pipe_lpac_hlsq_state_registers[] = { + 0x0af88, 0x0af8a, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_noncontext_pipe_lpac_hlsq_state_registers), 8)); + +/* Block: SP Cluster: noncontext Pipeline: PIPE_LPAC Location: SP_TOP */ +static const u32 gen7_0_0_sp_noncontext_pipe_lpac_sp_top_registers[] = { + 0x0af80, 0x0af84, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_noncontext_pipe_lpac_sp_top_registers), 8)); + +/* Block: SP Cluster: noncontext Pipeline: PIPE_LPAC Location: uSPTP */ +static const u32 gen7_0_0_sp_noncontext_pipe_lpac_usptp_registers[] = { + 0x0af80, 0x0af84, 0x0af90, 0x0af92, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_sp_noncontext_pipe_lpac_usptp_registers), 8)); + +/* Block: TPl1 Cluster: noncontext Pipeline: PIPE_NONE */ +static const u32 gen7_0_0_tpl1_noncontext_pipe_none_registers[] = { + 0x0b600, 0x0b600, 0x0b602, 0x0b602, 0x0b604, 0x0b604, 0x0b608, 0x0b60c, + 0x0b60f, 0x0b621, 0x0b630, 0x0b633, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_tpl1_noncontext_pipe_none_registers), 8)); + +/* Block: TPl1 Cluster: noncontext Pipeline: PIPE_BR */ +static const u32 gen7_0_0_tpl1_noncontext_pipe_br_registers[] = { + 0x0b600, 0x0b600, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_tpl1_noncontext_pipe_br_registers), 8)); + +/* Block: TPl1 Cluster: noncontext Pipeline: PIPE_LPAC */ +static const u32 gen7_0_0_tpl1_noncontext_pipe_lpac_registers[] = { + 0x0b780, 0x0b780, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_tpl1_noncontext_pipe_lpac_registers), 8)); + +static const struct gen7_sel_reg gen7_0_0_rb_rac_sel = { + .host_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_HOST, + .cd_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_CD, + .val = 0x0, +}; + +static const struct gen7_sel_reg gen7_0_0_rb_rbp_sel = { + .host_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_HOST, + .cd_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_CD, + .val = 0x9, +}; + +static const struct gen7_cluster_registers gen7_0_0_clusters[] = { + { A7XX_CLUSTER_NONE, PIPE_BR, STATE_NON_CONTEXT, + gen7_0_0_noncontext_pipe_br_registers, }, + { A7XX_CLUSTER_NONE, PIPE_BV, STATE_NON_CONTEXT, + gen7_0_0_noncontext_pipe_bv_registers, }, + { A7XX_CLUSTER_NONE, PIPE_LPAC, STATE_NON_CONTEXT, + gen7_0_0_noncontext_pipe_lpac_registers, }, + { A7XX_CLUSTER_NONE, PIPE_BR, STATE_NON_CONTEXT, + gen7_0_0_noncontext_rb_rac_pipe_br_registers, &gen7_0_0_rb_rac_sel, }, + { A7XX_CLUSTER_NONE, PIPE_BR, STATE_NON_CONTEXT, + gen7_0_0_noncontext_rb_rbp_pipe_br_registers, &gen7_0_0_rb_rbp_sel, }, + { A7XX_CLUSTER_GRAS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_gras_cluster_gras_pipe_br_registers, }, + { A7XX_CLUSTER_GRAS, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_0_0_gras_cluster_gras_pipe_bv_registers, }, + { A7XX_CLUSTER_GRAS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_gras_cluster_gras_pipe_br_registers, }, + { A7XX_CLUSTER_GRAS, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_0_0_gras_cluster_gras_pipe_bv_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_pc_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_0_0_pc_cluster_fe_pipe_bv_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_pc_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_0_0_pc_cluster_fe_pipe_bv_registers, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_rb_rac_cluster_ps_pipe_br_registers, &gen7_0_0_rb_rac_sel, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_rb_rac_cluster_ps_pipe_br_registers, &gen7_0_0_rb_rac_sel, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_rb_rbp_cluster_ps_pipe_br_registers, &gen7_0_0_rb_rbp_sel, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_rb_rbp_cluster_ps_pipe_br_registers, &gen7_0_0_rb_rbp_sel, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_vfd_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_0_0_vfd_cluster_fe_pipe_bv_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_vfd_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_0_0_vfd_cluster_fe_pipe_bv_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_fe_pipe_bv_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_fe_pipe_bv_registers, }, + { A7XX_CLUSTER_PC_VS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_pc_vs_pipe_br_registers, }, + { A7XX_CLUSTER_PC_VS, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_pc_vs_pipe_bv_registers, }, + { A7XX_CLUSTER_PC_VS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_pc_vs_pipe_br_registers, }, + { A7XX_CLUSTER_PC_VS, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_pc_vs_pipe_bv_registers, }, + { A7XX_CLUSTER_VPC_PS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_vpc_ps_pipe_br_registers, }, + { A7XX_CLUSTER_VPC_PS, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_vpc_ps_pipe_bv_registers, }, + { A7XX_CLUSTER_VPC_PS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_vpc_ps_pipe_br_registers, }, + { A7XX_CLUSTER_VPC_PS, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_vpc_ps_pipe_bv_registers, }, +}; + +static const struct gen7_sptp_cluster_registers gen7_0_0_sptp_clusters[] = { + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_BR, 0, A7XX_HLSQ_STATE, + gen7_0_0_sp_noncontext_pipe_br_hlsq_state_registers, 0xae00 }, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_BR, 0, A7XX_SP_TOP, + gen7_0_0_sp_noncontext_pipe_br_sp_top_registers, 0xae00 }, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_0_0_sp_noncontext_pipe_br_usptp_registers, 0xae00 }, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_LPAC, 0, A7XX_HLSQ_STATE, + gen7_0_0_sp_noncontext_pipe_lpac_hlsq_state_registers, 0xaf80 }, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_LPAC, 0, A7XX_SP_TOP, + gen7_0_0_sp_noncontext_pipe_lpac_sp_top_registers, 0xaf80 }, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_LPAC, 0, A7XX_USPTP, + gen7_0_0_sp_noncontext_pipe_lpac_usptp_registers, 0xaf80 }, + { A7XX_CLUSTER_NONE, A7XX_TP0_NCTX_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_0_0_tpl1_noncontext_pipe_br_registers, 0xb600 }, + { A7XX_CLUSTER_NONE, A7XX_TP0_NCTX_REG, PIPE_LPAC, 0, A7XX_USPTP, + gen7_0_0_tpl1_noncontext_pipe_lpac_registers, 0xb780 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_HLSQ_STATE, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_HLSQ_DP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_dp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_HLSQ_STATE, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_HLSQ_DP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_dp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_HLSQ_STATE, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_HLSQ_DP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_dp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_HLSQ_STATE, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_HLSQ_DP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_dp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_HLSQ_STATE, + gen7_0_0_sp_cluster_sp_ps_pipe_lpac_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_HLSQ_DP, + gen7_0_0_sp_cluster_sp_ps_pipe_lpac_hlsq_dp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_ps_pipe_lpac_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_ps_pipe_lpac_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_HLSQ_STATE, + gen7_0_0_sp_cluster_sp_vs_pipe_br_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BV, 0, A7XX_HLSQ_STATE, + gen7_0_0_sp_cluster_sp_vs_pipe_bv_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_vs_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BV, 0, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_vs_pipe_bv_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_vs_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BV, 0, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_vs_pipe_bv_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_HLSQ_STATE, + gen7_0_0_sp_cluster_sp_vs_pipe_br_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BV, 1, A7XX_HLSQ_STATE, + gen7_0_0_sp_cluster_sp_vs_pipe_bv_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_vs_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BV, 1, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_vs_pipe_bv_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_vs_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BV, 1, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_vs_pipe_bv_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_TP0_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_ps_pipe_br_registers, 0xb000 }, + { A7XX_CLUSTER_SP_PS, A7XX_TP0_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_ps_pipe_br_registers, 0xb000 }, + { A7XX_CLUSTER_SP_PS, A7XX_TP0_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_ps_pipe_br_registers, 0xb000 }, + { A7XX_CLUSTER_SP_PS, A7XX_TP0_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_ps_pipe_br_registers, 0xb000 }, + { A7XX_CLUSTER_SP_PS, A7XX_TP0_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_ps_pipe_lpac_registers, 0xb000 }, + { A7XX_CLUSTER_SP_VS, A7XX_TP0_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_vs_pipe_br_registers, 0xb000 }, + { A7XX_CLUSTER_SP_VS, A7XX_TP0_CTX0_3D_CVS_REG, PIPE_BV, 0, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_vs_pipe_bv_registers, 0xb000 }, + { A7XX_CLUSTER_SP_VS, A7XX_TP0_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_vs_pipe_br_registers, 0xb000 }, + { A7XX_CLUSTER_SP_VS, A7XX_TP0_CTX1_3D_CVS_REG, PIPE_BV, 1, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_vs_pipe_bv_registers, 0xb000 }, +}; + +static const u32 gen7_0_0_rscc_registers[] = { + 0x14000, 0x14036, 0x14040, 0x14042, 0x14080, 0x14084, 0x14089, 0x1408c, + 0x14091, 0x14094, 0x14099, 0x1409c, 0x140a1, 0x140a4, 0x140a9, 0x140ac, + 0x14100, 0x14102, 0x14114, 0x14119, 0x14124, 0x1412e, 0x14140, 0x14143, + 0x14180, 0x14197, 0x14340, 0x14342, 0x14344, 0x14347, 0x1434c, 0x14373, + 0x143ec, 0x143ef, 0x143f4, 0x1441b, 0x14494, 0x14497, 0x1449c, 0x144c3, + 0x1453c, 0x1453f, 0x14544, 0x1456b, 0x145e4, 0x145e7, 0x145ec, 0x14613, + 0x1468c, 0x1468f, 0x14694, 0x146bb, 0x14734, 0x14737, 0x1473c, 0x14763, + 0x147dc, 0x147df, 0x147e4, 0x1480b, 0x14884, 0x14887, 0x1488c, 0x148b3, + 0x1492c, 0x1492f, 0x14934, 0x1495b, 0x14f51, 0x14f54, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_rscc_registers), 8)); + +static const u32 gen7_0_0_cpr_registers[] = { + 0x26800, 0x26805, 0x26808, 0x2680c, 0x26814, 0x26814, 0x2681c, 0x2681c, + 0x26820, 0x26838, 0x26840, 0x26840, 0x26848, 0x26848, 0x26850, 0x26850, + 0x26880, 0x26898, 0x26980, 0x269b0, 0x269c0, 0x269c8, 0x269e0, 0x269ee, + 0x269fb, 0x269ff, 0x26a02, 0x26a07, 0x26a09, 0x26a0b, 0x26a10, 0x26b0f, + 0x27440, 0x27441, 0x27444, 0x27444, 0x27480, 0x274a2, 0x274ac, 0x274ac, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_cpr_registers), 8)); + +static const u32 gen7_0_0_gpucc_registers[] = { + 0x24000, 0x2400e, 0x24400, 0x2440e, 0x24800, 0x24805, 0x24c00, 0x24cff, + 0x25800, 0x25804, 0x25c00, 0x25c04, 0x26000, 0x26004, 0x26400, 0x26405, + 0x26414, 0x2641d, 0x2642a, 0x26430, 0x26432, 0x26432, 0x26441, 0x26455, + 0x26466, 0x26468, 0x26478, 0x2647a, 0x26489, 0x2648a, 0x2649c, 0x2649e, + 0x264a0, 0x264a3, 0x264b3, 0x264b5, 0x264c5, 0x264c7, 0x264d6, 0x264d8, + 0x264e8, 0x264e9, 0x264f9, 0x264fc, 0x2650b, 0x2650c, 0x2651c, 0x2651e, + 0x26540, 0x26570, 0x26600, 0x26616, 0x26620, 0x2662d, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_gpucc_registers), 8)); + +static const u32 gen7_0_0_cx_misc_registers[] = { + 0x27800, 0x27800, 0x27810, 0x27814, 0x27820, 0x27824, 0x27832, 0x27857, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_cx_misc_registers), 8)); + +static const u32 gen7_0_0_dpm_registers[] = { + 0x1aa00, 0x1aa06, 0x1aa09, 0x1aa0a, 0x1aa0c, 0x1aa0d, 0x1aa0f, 0x1aa12, + 0x1aa14, 0x1aa47, 0x1aa50, 0x1aa51, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_0_0_dpm_registers), 8)); + +static const struct gen7_reg_list gen7_0_0_reg_list[] = { + { gen7_0_0_gpu_registers, NULL }, + { gen7_0_0_cx_misc_registers, NULL }, + { gen7_0_0_dpm_registers, NULL }, + { NULL, NULL }, +}; + +static const u32 *gen7_0_0_external_core_regs[] = { + gen7_0_0_gpucc_registers, + gen7_0_0_cpr_registers, +}; +#endif /*_ADRENO_GEN7_0_0_SNAPSHOT_H */ diff --git a/drivers/gpu/drm/msm/adreno/adreno_gen7_17_0_snapshot.h b/drivers/gpu/drm/msm/adreno/adreno_gen7_17_0_snapshot.h new file mode 100644 index 000000000..00a4a0fc9 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/adreno_gen7_17_0_snapshot.h @@ -0,0 +1,428 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) Qualcomm Innovation Center, Inc. All rights reserved. + */ +#ifndef __ADRENO_GEN7_17_0_SNAPSHOT_H +#define __ADRENO_GEN7_17_0_SNAPSHOT_H + +#include "a6xx_gpu_state.h" + +/* + * Snapshot tables for Adreno A722 (Eliza). + * Cluster sub-arrays that are identical to A730 reference gen7_0_0_* + * symbols; adreno_gen7_0_0_snapshot.h is included first in the TU. + */ + +static const u32 gen7_17_0_rscc_registers[] = { + 0x14000, 0x14034, 0x14036, 0x14036, 0x14040, 0x14042, 0x14044, 0x14045, + 0x14047, 0x14047, 0x14080, 0x14084, 0x14089, 0x1408c, 0x14091, 0x14094, + 0x14099, 0x1409c, 0x140a1, 0x140a4, 0x140a9, 0x140ac, 0x140b1, 0x140b4, + 0x140b9, 0x140bc, 0x14100, 0x14104, 0x14114, 0x14119, 0x14124, 0x14132, + 0x14154, 0x1416b, 0x14340, 0x14341, 0x14344, 0x14344, 0x14346, 0x1437c, + 0x143f0, 0x143f8, 0x143fa, 0x143fe, 0x14400, 0x14404, 0x14406, 0x1440a, + 0x1440c, 0x14410, 0x14412, 0x14416, 0x14418, 0x1441c, 0x1441e, 0x14422, + 0x14424, 0x14424, 0x14498, 0x144a0, 0x144a2, 0x144a6, 0x144a8, 0x144ac, + 0x144ae, 0x144b2, 0x144b4, 0x144b8, 0x144ba, 0x144be, 0x144c0, 0x144c4, + 0x144c6, 0x144ca, 0x144cc, 0x144cc, 0x14540, 0x14548, 0x1454a, 0x1454e, + 0x14550, 0x14554, 0x14556, 0x1455a, 0x1455c, 0x14560, 0x14562, 0x14566, + 0x14568, 0x1456c, 0x1456e, 0x14572, 0x14574, 0x14574, 0x145e8, 0x145f0, + 0x145f2, 0x145f6, 0x145f8, 0x145fc, 0x145fe, 0x14602, 0x14604, 0x14608, + 0x1460a, 0x1460e, 0x14610, 0x14614, 0x14616, 0x1461a, 0x1461c, 0x1461c, + 0x14690, 0x14698, 0x1469a, 0x1469e, 0x146a0, 0x146a4, 0x146a6, 0x146aa, + 0x146ac, 0x146b0, 0x146b2, 0x146b6, 0x146b8, 0x146bc, 0x146be, 0x146c2, + 0x146c4, 0x146c4, 0x14738, 0x14740, 0x14742, 0x14746, 0x14748, 0x1474c, + 0x1474e, 0x14752, 0x14754, 0x14758, 0x1475a, 0x1475e, 0x14760, 0x14764, + 0x14766, 0x1476a, 0x1476c, 0x1476c, 0x147e0, 0x147e8, 0x147ea, 0x147ee, + 0x147f0, 0x147f4, 0x147f6, 0x147fa, 0x147fc, 0x14800, 0x14802, 0x14806, + 0x14808, 0x1480c, 0x1480e, 0x14812, 0x14814, 0x14814, 0x14888, 0x14890, + 0x14892, 0x14896, 0x14898, 0x1489c, 0x1489e, 0x148a2, 0x148a4, 0x148a8, + 0x148aa, 0x148ae, 0x148b0, 0x148b4, 0x148b6, 0x148ba, 0x148bc, 0x148bc, + 0x14930, 0x14938, 0x1493a, 0x1493e, 0x14940, 0x14944, 0x14946, 0x1494a, + 0x1494c, 0x14950, 0x14952, 0x14956, 0x14958, 0x1495c, 0x1495e, 0x14962, + 0x14964, 0x14964, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_rscc_registers), 8)); + +static const u32 gen7_17_0_cpr_registers[] = { + 0x26800, 0x26805, 0x26808, 0x2680c, 0x26814, 0x26814, 0x2681c, 0x2681c, + 0x26820, 0x26838, 0x26840, 0x26840, 0x26848, 0x26848, 0x26850, 0x26850, + 0x26880, 0x2688e, 0x26980, 0x269b0, 0x269c0, 0x269c2, 0x269c6, 0x269c8, + 0x269e0, 0x269ee, 0x269fb, 0x269ff, 0x26a02, 0x26a07, 0x26a09, 0x26a0b, + 0x26a10, 0x26b0f, 0x27440, 0x27441, 0x27444, 0x27444, 0x27480, 0x274a2, + 0x274ac, 0x274c4, 0x274c8, 0x274da, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_cpr_registers), 8)); + +static const u32 gen7_17_0_gpucc_registers[] = { + 0x24000, 0x2400f, 0x24400, 0x2440f, 0x24800, 0x24805, 0x24c00, 0x24cff, + 0x25400, 0x25404, 0x25800, 0x25804, 0x25c00, 0x25c04, 0x26000, 0x26004, + 0x26400, 0x26405, 0x26414, 0x2641d, 0x2642a, 0x2642c, 0x2642e, 0x26432, + 0x26434, 0x26434, 0x26443, 0x26457, 0x26459, 0x2645d, 0x2645f, 0x26464, + 0x26477, 0x26479, 0x26489, 0x2648b, 0x2649a, 0x2649b, 0x264ad, 0x264af, + 0x264b1, 0x264b5, 0x264d6, 0x264d8, 0x264e7, 0x264e9, 0x264f9, 0x264fa, + 0x2650a, 0x2650d, 0x2651f, 0x26520, 0x2652d, 0x2652f, 0x2653e, 0x2653e, + 0x26540, 0x2654e, 0x26554, 0x26573, 0x26576, 0x26576, 0x26593, 0x26593, + 0x26600, 0x26616, 0x26620, 0x2662d, 0x26630, 0x26631, 0x26635, 0x26635, + 0x26637, 0x26637, 0x2663a, 0x2663a, 0x26642, 0x26642, 0x26656, 0x26658, + 0x2665b, 0x2665d, 0x2665f, 0x26662, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_gpucc_registers), 8)); + +static const u32 *gen7_17_0_external_core_regs[] = { + gen7_17_0_gpucc_registers, + gen7_17_0_cpr_registers, +}; + +static const u32 gen7_17_0_debugbus_blocks[] = { + A7XX_DBGBUS_CP_0_0, + A7XX_DBGBUS_CP_0_1, + A7XX_DBGBUS_RBBM, + A7XX_DBGBUS_HLSQ, + A7XX_DBGBUS_UCHE_0, + A7XX_DBGBUS_TESS_BR, + A7XX_DBGBUS_PC_BR, + A7XX_DBGBUS_VFDP_BR, + A7XX_DBGBUS_VPC_BR, + A7XX_DBGBUS_TSE_BR, + A7XX_DBGBUS_RAS_BR, + A7XX_DBGBUS_VSC, + A7XX_DBGBUS_COM_0, + A7XX_DBGBUS_LRZ_BR, + A7XX_DBGBUS_UFC_0, + A7XX_DBGBUS_UFC_1, + A7XX_DBGBUS_GMU_GX, + A7XX_DBGBUS_DBGC, + A7XX_DBGBUS_GPC_BR, + A7XX_DBGBUS_LARC, + A7XX_DBGBUS_HLSQ_SPTP, + A7XX_DBGBUS_RB_0, + A7XX_DBGBUS_RB_1, + A7XX_DBGBUS_UCHE_WRAPPER, + A7XX_DBGBUS_CCU_0, + A7XX_DBGBUS_CCU_1, + A7XX_DBGBUS_VFD_BR_0, + A7XX_DBGBUS_VFD_BR_1, + A7XX_DBGBUS_VFD_BR_2, + A7XX_DBGBUS_VFD_BR_3, + A7XX_DBGBUS_USP_0, + A7XX_DBGBUS_USP_1, + A7XX_DBGBUS_TP_0, + A7XX_DBGBUS_TP_1, + A7XX_DBGBUS_TP_2, + A7XX_DBGBUS_TP_3, + A7XX_DBGBUS_USPTP_0, + A7XX_DBGBUS_USPTP_1, + A7XX_DBGBUS_USPTP_2, + A7XX_DBGBUS_USPTP_3, +}; + +static const struct gen7_sel_reg gen7_17_0_rb_rac_sel = { + .host_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_HOST, + .cd_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_CD, + .val = 0x0, +}; + +static const struct gen7_sel_reg gen7_17_0_rb_rbp_sel = { + .host_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_HOST, + .cd_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_CD, + .val = 0x9, +}; + +static const u32 gen7_17_0_post_crashdumper_registers[] = { + 0x00535, 0x00535, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_post_crashdumper_registers), 8)); + +static const u32 gen7_17_0_gpu_registers[] = { + 0x00000, 0x00000, 0x00002, 0x00002, 0x00011, 0x00012, 0x00016, 0x0001b, + 0x0001f, 0x00032, 0x00038, 0x0003c, 0x00042, 0x00042, 0x00044, 0x00044, + 0x00047, 0x00047, 0x00049, 0x0004a, 0x0004c, 0x0004c, 0x00050, 0x00050, + 0x00056, 0x00056, 0x00073, 0x00075, 0x000ad, 0x000ae, 0x000b0, 0x000b0, + 0x000b4, 0x000b4, 0x000b8, 0x000b8, 0x000bc, 0x000bc, 0x000c0, 0x000c0, + 0x000c4, 0x000c4, 0x000c8, 0x000c8, 0x000cc, 0x000cc, 0x000d0, 0x000d0, + 0x000d4, 0x000d4, 0x000d8, 0x000d8, 0x000dc, 0x000dc, 0x000e0, 0x000e0, + 0x000e4, 0x000e4, 0x000e8, 0x000e8, 0x000ec, 0x000ec, 0x000f0, 0x000f0, + 0x000f4, 0x000f4, 0x000f8, 0x000f8, 0x00100, 0x00100, 0x00104, 0x0010b, + 0x0010f, 0x0011d, 0x0012f, 0x0012f, 0x00200, 0x0020d, 0x00215, 0x00243, + 0x00260, 0x00268, 0x00272, 0x00274, 0x00286, 0x00286, 0x0028a, 0x0028a, + 0x0028c, 0x0028c, 0x00300, 0x00401, 0x00500, 0x00500, 0x00507, 0x0050b, + 0x0050f, 0x0050f, 0x00511, 0x00511, 0x00533, 0x00534, 0x00540, 0x00555, + 0x00564, 0x00567, 0x00800, 0x00808, 0x00810, 0x00813, 0x00820, 0x00821, + 0x00823, 0x00827, 0x00830, 0x00834, 0x00840, 0x00841, 0x00843, 0x00847, + 0x0084f, 0x00886, 0x008a0, 0x008ab, 0x008c0, 0x008c0, 0x008c4, 0x008c5, + 0x008d0, 0x008dd, 0x008f0, 0x008f3, 0x00900, 0x00903, 0x00908, 0x00911, + 0x00928, 0x0093e, 0x00942, 0x0094d, 0x00980, 0x00984, 0x0098d, 0x0098f, + 0x009b0, 0x009b4, 0x009c2, 0x009c9, 0x009ce, 0x009d7, 0x00a00, 0x00a00, + 0x00a02, 0x00a03, 0x00a10, 0x00a4f, 0x00a67, 0x00a6c, 0x00a9c, 0x00a9f, + 0x00c00, 0x00c00, 0x00c02, 0x00c04, 0x00c06, 0x00c06, 0x00c10, 0x00cd9, + 0x00ce0, 0x00d0c, 0x00df0, 0x00df4, 0x00e01, 0x00e02, 0x00e07, 0x00e0e, + 0x00e10, 0x00e12, 0x00e17, 0x00e17, 0x00e19, 0x00e19, 0x00e1b, 0x00e2b, + 0x00e30, 0x00e32, 0x00e38, 0x00e3c, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_gpu_registers), 8)); + +static const u32 gen7_17_0_dbgc_registers[] = { + 0x00600, 0x0061c, 0x0061e, 0x00634, 0x00640, 0x0065a, 0x00679, 0x0067a, + 0x00699, 0x00699, 0x0069b, 0x0069e, 0x18400, 0x1841c, 0x1841e, 0x18434, + 0x18440, 0x1845c, 0x18479, 0x1847c, 0x18580, 0x18581, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_dbgc_registers), 8)); + +static const u32 gen7_17_0_noncontext_pipe_br_registers[] = { + 0x00887, 0x0088c, 0x08600, 0x08600, 0x08602, 0x08602, 0x08610, 0x0861b, + 0x08620, 0x08620, 0x08630, 0x08630, 0x08637, 0x08639, 0x08640, 0x08640, + 0x09600, 0x09600, 0x09602, 0x09603, 0x0960a, 0x09616, 0x09624, 0x0963a, + 0x09640, 0x09640, 0x09e00, 0x09e00, 0x09e02, 0x09e07, 0x09e0a, 0x09e16, + 0x09e19, 0x09e19, 0x09e1c, 0x09e1c, 0x09e20, 0x09e25, 0x09e30, 0x09e31, + 0x09e40, 0x09e51, 0x09e64, 0x09e64, 0x09e70, 0x09e72, 0x09e78, 0x09e79, + 0x09e80, 0x09fff, 0x0a600, 0x0a600, 0x0a603, 0x0a603, 0x0a610, 0x0a61f, + 0x0a630, 0x0a631, 0x0a638, 0x0a638, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_noncontext_pipe_br_registers), 8)); + +static const u32 gen7_17_0_noncontext_rb_rac_pipe_br_registers[] = { + 0x08e10, 0x08e1c, 0x08e20, 0x08e25, 0x08e51, 0x08e54, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_noncontext_rb_rac_pipe_br_registers), 8)); + +static const u32 gen7_17_0_noncontext_rb_rbp_pipe_br_registers[] = { + 0x08e01, 0x08e01, 0x08e04, 0x08e04, 0x08e06, 0x08e09, 0x08e0c, 0x08e0c, + 0x08e28, 0x08e28, 0x08e2c, 0x08e35, 0x08e3b, 0x08e3f, 0x08e50, 0x08e50, + 0x08e5b, 0x08e5d, 0x08e5f, 0x08e5f, 0x08e61, 0x08e61, 0x08e63, 0x08e65, + 0x08e68, 0x08e68, 0x08e70, 0x08e79, 0x08e80, 0x08e8f, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_noncontext_rb_rbp_pipe_br_registers), 8)); + +static const u32 gen7_17_0_pc_cluster_fe_pipe_br_registers[] = { + 0x09800, 0x09804, 0x09806, 0x0980a, 0x09810, 0x09811, 0x09884, 0x09886, + 0x09970, 0x09972, 0x09b00, 0x09b08, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_pc_cluster_fe_pipe_br_registers), 8)); + +static const u32 gen7_17_0_sp_cluster_sp_ps_pipe_lpac_hlsq_state_registers[] = { + 0x0aa40, 0x0aabf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_sp_cluster_sp_ps_pipe_lpac_hlsq_state_registers), 8)); + +static const u32 gen7_17_0_sp_cluster_sp_ps_pipe_lpac_usptp_registers[] = { + 0x0aa40, 0x0aabf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_sp_cluster_sp_ps_pipe_lpac_usptp_registers), 8)); + +static const u32 gen7_17_0_non_context_tpl1_pipe_none_usptp_registers[] = { + 0x0b602, 0x0b602, 0x0b604, 0x0b604, 0x0b608, 0x0b60c, 0x0b60f, 0x0b621, + 0x0b630, 0x0b633, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_non_context_tpl1_pipe_none_usptp_registers), 8)); + +static const u32 gen7_17_0_non_context_tpl1_pipe_br_usptp_registers[] = { + 0x0b600, 0x0b600, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_non_context_tpl1_pipe_br_usptp_registers), 8)); + +static const u32 gen7_17_0_tpl1_cluster_sp_vs_pipe_br_usptp_registers[] = { + 0x0b300, 0x0b307, 0x0b309, 0x0b309, 0x0b310, 0x0b310, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_tpl1_cluster_sp_vs_pipe_br_usptp_registers), 8)); + +static const u32 gen7_17_0_tpl1_cluster_sp_ps_pipe_br_usptp_registers[] = { + 0x0b180, 0x0b183, 0x0b190, 0x0b195, 0x0b2c0, 0x0b2d5, 0x0b300, 0x0b307, + 0x0b309, 0x0b309, 0x0b310, 0x0b310, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_17_0_tpl1_cluster_sp_ps_pipe_br_usptp_registers), 8)); + +/* No BV pipe — gen7_0_0_* sub-arrays are shared from adreno_gen7_0_0_snapshot.h */ +static struct gen7_cluster_registers gen7_17_0_clusters[] = { + { A7XX_CLUSTER_NONE, PIPE_BR, STATE_NON_CONTEXT, + gen7_17_0_noncontext_pipe_br_registers, }, + { A7XX_CLUSTER_NONE, PIPE_BR, STATE_NON_CONTEXT, + gen7_17_0_noncontext_rb_rac_pipe_br_registers, &gen7_17_0_rb_rac_sel, }, + { A7XX_CLUSTER_NONE, PIPE_BR, STATE_NON_CONTEXT, + gen7_17_0_noncontext_rb_rbp_pipe_br_registers, &gen7_17_0_rb_rbp_sel, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_rb_rac_cluster_ps_pipe_br_registers, &gen7_17_0_rb_rac_sel, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_rb_rac_cluster_ps_pipe_br_registers, &gen7_17_0_rb_rac_sel, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_rb_rbp_cluster_ps_pipe_br_registers, &gen7_17_0_rb_rbp_sel, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_rb_rbp_cluster_ps_pipe_br_registers, &gen7_17_0_rb_rbp_sel, }, + { A7XX_CLUSTER_GRAS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_gras_cluster_gras_pipe_br_registers, }, + { A7XX_CLUSTER_GRAS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_gras_cluster_gras_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_17_0_pc_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_17_0_pc_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_vfd_cluster_fe_pipe_bv_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_vfd_cluster_fe_pipe_bv_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_PC_VS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_pc_vs_pipe_br_registers, }, + { A7XX_CLUSTER_PC_VS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_pc_vs_pipe_br_registers, }, + { A7XX_CLUSTER_VPC_PS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_vpc_ps_pipe_br_registers, }, + { A7XX_CLUSTER_VPC_PS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_vpc_ps_pipe_br_registers, }, +}; + +/* No BV pipe; 2 SPs, 2 USPTPs */ +static struct gen7_sptp_cluster_registers gen7_17_0_sptp_clusters[] = { + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_BR, 0, A7XX_HLSQ_STATE, + gen7_0_0_sp_noncontext_pipe_br_hlsq_state_registers, 0xae00 }, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_BR, 0, A7XX_SP_TOP, + gen7_0_0_sp_noncontext_pipe_br_sp_top_registers, 0xae00 }, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_0_0_sp_noncontext_pipe_br_usptp_registers, 0xae00 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_HLSQ_STATE, + gen7_0_0_sp_cluster_sp_vs_pipe_br_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_vs_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_vs_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_HLSQ_STATE, + gen7_0_0_sp_cluster_sp_vs_pipe_br_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_vs_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_vs_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_HLSQ_STATE, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_HLSQ_DP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_dp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_HLSQ_STATE, + gen7_17_0_sp_cluster_sp_ps_pipe_lpac_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_USPTP, + gen7_17_0_sp_cluster_sp_ps_pipe_lpac_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_HLSQ_STATE, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_HLSQ_DP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_dp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_HLSQ_DP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_dp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_HLSQ_DP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_dp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_SP_TOP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_USPTP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_NONE, A7XX_TP0_NCTX_REG, PIPE_NONE, 0, A7XX_USPTP, + gen7_17_0_non_context_tpl1_pipe_none_usptp_registers, 0xb600 }, + { A7XX_CLUSTER_NONE, A7XX_TP0_NCTX_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_17_0_non_context_tpl1_pipe_br_usptp_registers, 0xb600 }, + { A7XX_CLUSTER_SP_VS, A7XX_TP0_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_17_0_tpl1_cluster_sp_vs_pipe_br_usptp_registers, 0xb000 }, + { A7XX_CLUSTER_SP_VS, A7XX_TP0_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_17_0_tpl1_cluster_sp_vs_pipe_br_usptp_registers, 0xb000 }, + { A7XX_CLUSTER_SP_PS, A7XX_TP0_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_17_0_tpl1_cluster_sp_ps_pipe_br_usptp_registers, 0xb000 }, + { A7XX_CLUSTER_SP_PS, A7XX_TP0_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_17_0_tpl1_cluster_sp_ps_pipe_br_usptp_registers, 0xb000 }, + { A7XX_CLUSTER_SP_PS, A7XX_TP0_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_USPTP, + gen7_17_0_tpl1_cluster_sp_ps_pipe_br_usptp_registers, 0xb000 }, + { A7XX_CLUSTER_SP_PS, A7XX_TP0_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_USPTP, + gen7_17_0_tpl1_cluster_sp_ps_pipe_br_usptp_registers, 0xb000 }, +}; + +static struct gen7_shader_block gen7_17_0_shader_blocks[] = { + { A7XX_TP0_TMO_DATA, 0x0200, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_TP0_SMO_DATA, 0x0080, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_TP0_MIPMAP_BASE_DATA, 0x03c0, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_INST_DATA, 0x0800, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_INST_DATA_1, 0x0800, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_0_DATA, 0x0800, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_1_DATA, 0x0800, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_2_DATA, 0x0800, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_3_DATA, 0x0800, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_4_DATA, 0x0800, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_5_DATA, 0x0800, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_CB_RAM, 0x0390, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_INST_TAG, 0x0090, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_TMO_TAG, 0x0080, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_SMO_TAG, 0x0080, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_STATE_DATA, 0x0040, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_HWAVE_RAM, 0x0100, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_L0_INST_BUF, 0x0050, 2, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_HLSQ_CVS_BE_CTXT_BUF_RAM_TAG, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CPS_BE_CTXT_BUF_RAM_TAG, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CVS_BE_CTXT_BUF_RAM, 0x0300, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CPS_BE_CTXT_BUF_RAM, 0x0300, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CHUNK_CVS_RAM, 0x01c0, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CHUNK_CPS_RAM, 0x0300, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CHUNK_CVS_RAM_TAG, 0x0040, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CHUNK_CPS_RAM_TAG, 0x0040, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_ICB_CVS_CB_BASE_TAG, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_ICB_CPS_CB_BASE_TAG, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CVS_MISC_RAM, 0x0280, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CPS_MISC_RAM, 0x0800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CPS_MISC_RAM_1, 0x0200, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_INST_RAM, 0x0800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CVS_CONST_RAM, 0x0800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CPS_CONST_RAM, 0x0800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CVS_MISC_RAM_TAG, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CPS_MISC_RAM_TAG, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_INST_RAM_TAG, 0x0080, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CVS_CONST_RAM_TAG, 0x0064, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CPS_CONST_RAM_TAG, 0x0064, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_INST_RAM_1, 0x0800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_STPROC_META, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_BV_BE_META, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_DATAPATH_META, 0x0020, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_FRONTEND_META, 0x0040, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_INDIRECT_META, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_BACKEND_META, 0x0040, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, +}; + +static struct gen7_reg_list gen7_17_0_reg_list[] = { + { gen7_17_0_gpu_registers, NULL }, + { gen7_17_0_dbgc_registers, NULL }, + { NULL, NULL }, +}; + +static const struct a6xx_indexed_registers gen7_17_0_cp_indexed_reglist[] = { + { "CP_SQE_STAT", REG_A6XX_CP_SQE_STAT_ADDR, + REG_A6XX_CP_SQE_STAT_DATA, 0x40, NULL }, + { "CP_DRAW_STATE", REG_A6XX_CP_DRAW_STATE_ADDR, + REG_A6XX_CP_DRAW_STATE_DATA, 0x100, NULL }, + { "CP_SQE_UCODE_DBG", REG_A6XX_CP_SQE_UCODE_DBG_ADDR, + REG_A6XX_CP_SQE_UCODE_DBG_DATA, 0x8000, NULL }, + { "CP_ROQ_DBG", REG_A6XX_CP_ROQ_DBG_ADDR, + REG_A6XX_CP_ROQ_DBG_DATA, 0, a7xx_get_cp_roq_size }, +}; + +#endif /* __ADRENO_GEN7_17_0_SNAPSHOT_H */ diff --git a/drivers/gpu/drm/msm/adreno/adreno_gen7_2_0_snapshot.h b/drivers/gpu/drm/msm/adreno/adreno_gen7_2_0_snapshot.h new file mode 100644 index 000000000..7897622ea --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/adreno_gen7_2_0_snapshot.h @@ -0,0 +1,755 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2020-2021, The Linux Foundation. All rights reserved. + * Copyright (c) 2022-2023 Qualcomm Innovation Center, Inc. All rights reserved. + */ +#ifndef __ADRENO_GEN7_2_0_SNAPSHOT_H +#define __ADRENO_GEN7_2_0_SNAPSHOT_H + +#include "a6xx_gpu_state.h" + +static const u32 gen7_2_0_debugbus_blocks[] = { + A7XX_DBGBUS_CP_0_0, + A7XX_DBGBUS_CP_0_1, + A7XX_DBGBUS_RBBM, + A7XX_DBGBUS_HLSQ, + A7XX_DBGBUS_UCHE_0, + A7XX_DBGBUS_UCHE_1, + A7XX_DBGBUS_TESS_BR, + A7XX_DBGBUS_TESS_BV, + A7XX_DBGBUS_PC_BR, + A7XX_DBGBUS_PC_BV, + A7XX_DBGBUS_VFDP_BR, + A7XX_DBGBUS_VFDP_BV, + A7XX_DBGBUS_VPC_BR, + A7XX_DBGBUS_VPC_BV, + A7XX_DBGBUS_TSE_BR, + A7XX_DBGBUS_TSE_BV, + A7XX_DBGBUS_RAS_BR, + A7XX_DBGBUS_RAS_BV, + A7XX_DBGBUS_VSC, + A7XX_DBGBUS_COM_0, + A7XX_DBGBUS_LRZ_BR, + A7XX_DBGBUS_LRZ_BV, + A7XX_DBGBUS_UFC_0, + A7XX_DBGBUS_UFC_1, + A7XX_DBGBUS_GMU_GX, + A7XX_DBGBUS_DBGC, + A7XX_DBGBUS_GPC_BR, + A7XX_DBGBUS_GPC_BV, + A7XX_DBGBUS_LARC, + A7XX_DBGBUS_HLSQ_SPTP, + A7XX_DBGBUS_RB_0, + A7XX_DBGBUS_RB_1, + A7XX_DBGBUS_RB_2, + A7XX_DBGBUS_RB_3, + A7XX_DBGBUS_RB_4, + A7XX_DBGBUS_RB_5, + A7XX_DBGBUS_UCHE_WRAPPER, + A7XX_DBGBUS_CCU_0, + A7XX_DBGBUS_CCU_1, + A7XX_DBGBUS_CCU_2, + A7XX_DBGBUS_CCU_3, + A7XX_DBGBUS_CCU_4, + A7XX_DBGBUS_CCU_5, + A7XX_DBGBUS_VFD_BR_0, + A7XX_DBGBUS_VFD_BR_1, + A7XX_DBGBUS_VFD_BR_2, + A7XX_DBGBUS_VFD_BR_3, + A7XX_DBGBUS_VFD_BR_4, + A7XX_DBGBUS_VFD_BR_5, + A7XX_DBGBUS_VFD_BV_0, + A7XX_DBGBUS_VFD_BV_1, + A7XX_DBGBUS_USP_0, + A7XX_DBGBUS_USP_1, + A7XX_DBGBUS_USP_2, + A7XX_DBGBUS_USP_3, + A7XX_DBGBUS_USP_4, + A7XX_DBGBUS_USP_5, + A7XX_DBGBUS_TP_0, + A7XX_DBGBUS_TP_1, + A7XX_DBGBUS_TP_2, + A7XX_DBGBUS_TP_3, + A7XX_DBGBUS_TP_4, + A7XX_DBGBUS_TP_5, + A7XX_DBGBUS_TP_6, + A7XX_DBGBUS_TP_7, + A7XX_DBGBUS_TP_8, + A7XX_DBGBUS_TP_9, + A7XX_DBGBUS_TP_10, + A7XX_DBGBUS_TP_11, + A7XX_DBGBUS_USPTP_0, + A7XX_DBGBUS_USPTP_1, + A7XX_DBGBUS_USPTP_2, + A7XX_DBGBUS_USPTP_3, + A7XX_DBGBUS_USPTP_4, + A7XX_DBGBUS_USPTP_5, + A7XX_DBGBUS_USPTP_6, + A7XX_DBGBUS_USPTP_7, + A7XX_DBGBUS_USPTP_8, + A7XX_DBGBUS_USPTP_9, + A7XX_DBGBUS_USPTP_10, + A7XX_DBGBUS_USPTP_11, + A7XX_DBGBUS_CCHE_0, + A7XX_DBGBUS_CCHE_1, + A7XX_DBGBUS_CCHE_2, +}; + +static const struct gen7_shader_block gen7_2_0_shader_blocks[] = { + {A7XX_TP0_TMO_DATA, 0x200, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_TP0_SMO_DATA, 0x80, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_TP0_MIPMAP_BASE_DATA, 0x3c0, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_INST_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_INST_DATA_1, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_0_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_1_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_2_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_3_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_4_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_5_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_6_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_7_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_CB_RAM, 0x390, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_13_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_14_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_INST_TAG, 0xc0, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_INST_DATA_2, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_TMO_TAG, 0x80, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_SMO_TAG, 0x80, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_STATE_DATA, 0x40, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_HWAVE_RAM, 0x100, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_L0_INST_BUF, 0x50, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_8_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_9_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_10_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_11_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_SP_LB_12_DATA, 0x800, 6, 2, PIPE_BR, A7XX_USPTP}, + {A7XX_HLSQ_CVS_BE_CTXT_BUF_RAM_TAG, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CVS_BE_CTXT_BUF_RAM_TAG, 0x10, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CPS_BE_CTXT_BUF_RAM_TAG, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CVS_BE_CTXT_BUF_RAM, 0x300, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CVS_BE_CTXT_BUF_RAM, 0x300, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CPS_BE_CTXT_BUF_RAM, 0x300, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CVS_RAM, 0x1c0, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CVS_RAM, 0x1c0, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CPS_RAM, 0x300, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CPS_RAM, 0x180, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CVS_RAM_TAG, 0x40, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CVS_RAM_TAG, 0x40, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CPS_RAM_TAG, 0x40, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CHUNK_CPS_RAM_TAG, 0x40, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_ICB_CVS_CB_BASE_TAG, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_ICB_CVS_CB_BASE_TAG, 0x10, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_ICB_CPS_CB_BASE_TAG, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_ICB_CPS_CB_BASE_TAG, 0x10, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CVS_MISC_RAM, 0x280, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CVS_MISC_RAM, 0x280, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CPS_MISC_RAM, 0x800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CPS_MISC_RAM, 0x200, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CPS_MISC_RAM_1, 0x1c0, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INST_RAM, 0x800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INST_RAM, 0x800, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INST_RAM, 0x200, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CVS_MISC_RAM_TAG, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CVS_MISC_RAM_TAG, 0x10, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CPS_MISC_RAM_TAG, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_CPS_MISC_RAM_TAG, 0x10, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INST_RAM_TAG, 0x80, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INST_RAM_TAG, 0x80, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INST_RAM_TAG, 0x80, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CVS_CONST_RAM_TAG, 0x64, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CVS_CONST_RAM_TAG, 0x38, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CPS_CONST_RAM_TAG, 0x64, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CPS_CONST_RAM_TAG, 0x10, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CVS_CONST_RAM, 0x800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CVS_CONST_RAM, 0x800, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CPS_CONST_RAM, 0x800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_GFX_CPS_CONST_RAM, 0x800, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INST_RAM_1, 0x800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_STPROC_META, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_BV_BE_META, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_BV_BE_META, 0x10, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_DATAPATH_META, 0x20, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_FRONTEND_META, 0x80, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_FRONTEND_META, 0x80, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_FRONTEND_META, 0x80, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_INDIRECT_META, 0x10, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_BACKEND_META, 0x40, 1, 1, PIPE_BR, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_BACKEND_META, 0x40, 1, 1, PIPE_BV, A7XX_HLSQ_STATE}, + {A7XX_HLSQ_BACKEND_META, 0x40, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE}, +}; + +static const u32 gen7_2_0_gpu_registers[] = { + 0x00000, 0x00000, 0x00002, 0x00002, 0x00011, 0x00012, 0x00016, 0x0001b, + 0x0001f, 0x00032, 0x00038, 0x0003c, 0x00042, 0x00042, 0x00044, 0x00044, + 0x00047, 0x00047, 0x00049, 0x0004a, 0x0004c, 0x0004c, 0x00050, 0x00050, + 0x00056, 0x00056, 0x00073, 0x0007d, 0x000ad, 0x000ae, 0x000b0, 0x000b0, + 0x000b4, 0x000b4, 0x000b8, 0x000b8, 0x000bc, 0x000bc, 0x000c0, 0x000c0, + 0x000c4, 0x000c4, 0x000c8, 0x000c8, 0x000cc, 0x000cc, 0x000d0, 0x000d0, + 0x000d4, 0x000d4, 0x000d8, 0x000d8, 0x000dc, 0x000dc, 0x000e0, 0x000e0, + 0x000e4, 0x000e4, 0x000e8, 0x000e8, 0x000ec, 0x000ec, 0x000f0, 0x000f0, + 0x000f4, 0x000f4, 0x000f8, 0x000f8, 0x00100, 0x00100, 0x00104, 0x0010c, + 0x0010f, 0x0011d, 0x0012f, 0x0012f, 0x00200, 0x0020d, 0x00211, 0x00211, + 0x00215, 0x00253, 0x00260, 0x00270, 0x00272, 0x00274, 0x00281, 0x0028d, + 0x00300, 0x00401, 0x00410, 0x00451, 0x00460, 0x004a3, 0x004c0, 0x004d1, + 0x00500, 0x00500, 0x00507, 0x0050b, 0x0050f, 0x0050f, 0x00511, 0x00511, + 0x00533, 0x00536, 0x00540, 0x00555, 0x00564, 0x00567, 0x00574, 0x00577, + 0x00584, 0x0059b, 0x005fb, 0x005ff, 0x00800, 0x00808, 0x00810, 0x00813, + 0x00820, 0x00821, 0x00823, 0x00827, 0x00830, 0x00834, 0x0083f, 0x00841, + 0x00843, 0x00847, 0x0084f, 0x00886, 0x008a0, 0x008ab, 0x008c0, 0x008c0, + 0x008c4, 0x008c6, 0x008d0, 0x008dd, 0x008e0, 0x008e6, 0x008f0, 0x008f3, + 0x00900, 0x00903, 0x00908, 0x00911, 0x00928, 0x0093e, 0x00942, 0x0094d, + 0x00980, 0x00984, 0x0098d, 0x0098f, 0x009b0, 0x009b4, 0x009c2, 0x009c9, + 0x009ce, 0x009d7, 0x009e0, 0x009e7, 0x00a00, 0x00a00, 0x00a02, 0x00a03, + 0x00a10, 0x00a4f, 0x00a61, 0x00a9f, 0x00ad0, 0x00adb, 0x00b00, 0x00b31, + 0x00b35, 0x00b3c, 0x00b40, 0x00b40, 0x00c00, 0x00c00, 0x00c02, 0x00c04, + 0x00c06, 0x00c06, 0x00c10, 0x00cd9, 0x00ce0, 0x00d0c, 0x00df0, 0x00df4, + 0x00e01, 0x00e02, 0x00e07, 0x00e0e, 0x00e10, 0x00e13, 0x00e17, 0x00e19, + 0x00e1b, 0x00e2b, 0x00e30, 0x00e32, 0x00e38, 0x00e3c, 0x00e40, 0x00e4b, + 0x0ec00, 0x0ec01, 0x0ec05, 0x0ec05, 0x0ec07, 0x0ec07, 0x0ec0a, 0x0ec0a, + 0x0ec12, 0x0ec12, 0x0ec26, 0x0ec28, 0x0ec2b, 0x0ec2d, 0x0ec2f, 0x0ec2f, + 0x0ec40, 0x0ec41, 0x0ec45, 0x0ec45, 0x0ec47, 0x0ec47, 0x0ec4a, 0x0ec4a, + 0x0ec52, 0x0ec52, 0x0ec66, 0x0ec68, 0x0ec6b, 0x0ec6d, 0x0ec6f, 0x0ec6f, + 0x0ec80, 0x0ec81, 0x0ec85, 0x0ec85, 0x0ec87, 0x0ec87, 0x0ec8a, 0x0ec8a, + 0x0ec92, 0x0ec92, 0x0eca6, 0x0eca8, 0x0ecab, 0x0ecad, 0x0ecaf, 0x0ecaf, + 0x0ecc0, 0x0ecc1, 0x0ecc5, 0x0ecc5, 0x0ecc7, 0x0ecc7, 0x0ecca, 0x0ecca, + 0x0ecd2, 0x0ecd2, 0x0ece6, 0x0ece8, 0x0eceb, 0x0eced, 0x0ecef, 0x0ecef, + 0x0ed00, 0x0ed01, 0x0ed05, 0x0ed05, 0x0ed07, 0x0ed07, 0x0ed0a, 0x0ed0a, + 0x0ed12, 0x0ed12, 0x0ed26, 0x0ed28, 0x0ed2b, 0x0ed2d, 0x0ed2f, 0x0ed2f, + 0x0ed40, 0x0ed41, 0x0ed45, 0x0ed45, 0x0ed47, 0x0ed47, 0x0ed4a, 0x0ed4a, + 0x0ed52, 0x0ed52, 0x0ed66, 0x0ed68, 0x0ed6b, 0x0ed6d, 0x0ed6f, 0x0ed6f, + 0x0ed80, 0x0ed81, 0x0ed85, 0x0ed85, 0x0ed87, 0x0ed87, 0x0ed8a, 0x0ed8a, + 0x0ed92, 0x0ed92, 0x0eda6, 0x0eda8, 0x0edab, 0x0edad, 0x0edaf, 0x0edaf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_gpu_registers), 8)); + +static const u32 gen7_2_0_gmu_registers[] = { + 0x10001, 0x10001, 0x10003, 0x10003, 0x10401, 0x10401, 0x10403, 0x10403, + 0x10801, 0x10801, 0x10803, 0x10803, 0x10c01, 0x10c01, 0x10c03, 0x10c03, + 0x11001, 0x11001, 0x11003, 0x11003, 0x11401, 0x11401, 0x11403, 0x11403, + 0x11801, 0x11801, 0x11803, 0x11803, 0x11c01, 0x11c01, 0x11c03, 0x11c03, + 0x1a79b, 0x1a79b, 0x1a7ac, 0x1a7b9, 0x1a7dc, 0x1a7dd, 0x1a7e0, 0x1a7e1, + 0x1a803, 0x1a803, 0x1a805, 0x1a806, 0x1a84e, 0x1a84e, 0x1a856, 0x1a856, + 0x1f400, 0x1f40d, 0x1f40f, 0x1f411, 0x1f500, 0x1f500, 0x1f507, 0x1f507, + 0x1f509, 0x1f50b, 0x1f700, 0x1f701, 0x1f704, 0x1f706, 0x1f708, 0x1f709, + 0x1f70c, 0x1f70d, 0x1f710, 0x1f711, 0x1f713, 0x1f716, 0x1f720, 0x1f724, + 0x1f729, 0x1f729, 0x1f730, 0x1f747, 0x1f760, 0x1f761, 0x1f764, 0x1f76b, + 0x1f800, 0x1f804, 0x1f807, 0x1f808, 0x1f80b, 0x1f80c, 0x1f80f, 0x1f80f, + 0x1f811, 0x1f811, 0x1f813, 0x1f817, 0x1f819, 0x1f81c, 0x1f824, 0x1f82a, + 0x1f82d, 0x1f830, 0x1f840, 0x1f853, 0x1f860, 0x1f860, 0x1f862, 0x1f864, + 0x1f868, 0x1f868, 0x1f870, 0x1f879, 0x1f87f, 0x1f87f, 0x1f888, 0x1f889, + 0x1f8a0, 0x1f8a2, 0x1f890, 0x1f892, 0x1f894, 0x1f896, 0x1f8a4, 0x1f8af, + 0x1f8b8, 0x1f8b9, 0x1f8c0, 0x1f8c1, 0x1f8c3, 0x1f8c4, 0x1f8d0, 0x1f8d0, + 0x1f8ec, 0x1f8ec, 0x1f8f0, 0x1f8f1, 0x1f910, 0x1f917, 0x1f920, 0x1f921, + 0x1f924, 0x1f925, 0x1f928, 0x1f929, 0x1f92c, 0x1f92d, 0x1f940, 0x1f940, + 0x1f942, 0x1f944, 0x1f948, 0x1f94a, 0x1f94f, 0x1f951, 0x1f954, 0x1f955, + 0x1f958, 0x1f95a, 0x1f95d, 0x1f95d, 0x1f962, 0x1f96b, 0x1f970, 0x1f979, + 0x1f97c, 0x1f97c, 0x1f980, 0x1f981, 0x1f984, 0x1f986, 0x1f992, 0x1f993, + 0x1f996, 0x1f99e, 0x1f9c0, 0x1f9c0, 0x1f9c5, 0x1f9d4, 0x1f9f0, 0x1f9f1, + 0x1f9f8, 0x1f9fa, 0x1f9fc, 0x1f9fc, 0x1fa00, 0x1fa03, 0x20000, 0x20012, + 0x20018, 0x20018, 0x2001a, 0x2001a, 0x20020, 0x20024, 0x20030, 0x20031, + 0x20034, 0x20036, 0x23801, 0x23801, 0x23803, 0x23803, 0x23805, 0x23805, + 0x23807, 0x23807, 0x23809, 0x23809, 0x2380b, 0x2380b, 0x2380d, 0x2380d, + 0x2380f, 0x2380f, 0x23811, 0x23811, 0x23813, 0x23813, 0x23815, 0x23815, + 0x23817, 0x23817, 0x23819, 0x23819, 0x2381b, 0x2381b, 0x2381d, 0x2381d, + 0x2381f, 0x23820, 0x23822, 0x23822, 0x23824, 0x23824, 0x23826, 0x23826, + 0x23828, 0x23828, 0x2382a, 0x2382a, 0x2382c, 0x2382c, 0x2382e, 0x2382e, + 0x23830, 0x23830, 0x23832, 0x23832, 0x23834, 0x23834, 0x23836, 0x23836, + 0x23838, 0x23838, 0x2383a, 0x2383a, 0x2383c, 0x2383c, 0x2383e, 0x2383e, + 0x23840, 0x23847, 0x23b00, 0x23b01, 0x23b03, 0x23b03, 0x23b05, 0x23b0e, + 0x23b10, 0x23b13, 0x23b15, 0x23b16, 0x23b20, 0x23b20, 0x23b28, 0x23b28, + 0x23b30, 0x23b30, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_gmu_registers), 8)); + +static const u32 gen7_2_0_gmugx_registers[] = { + 0x1a400, 0x1a41f, 0x1a440, 0x1a45f, 0x1a480, 0x1a49f, 0x1a4c0, 0x1a4df, + 0x1a500, 0x1a51f, 0x1a540, 0x1a55f, 0x1a580, 0x1a59f, 0x1a5c0, 0x1a5df, + 0x1a600, 0x1a61f, 0x1a640, 0x1a65f, 0x1a780, 0x1a781, 0x1a783, 0x1a785, + 0x1a787, 0x1a789, 0x1a78b, 0x1a78d, 0x1a78f, 0x1a791, 0x1a793, 0x1a795, + 0x1a797, 0x1a799, 0x1a79c, 0x1a79d, 0x1a79f, 0x1a79f, 0x1a7a0, 0x1a7a1, + 0x1a7a3, 0x1a7a3, 0x1a7a8, 0x1a7ab, 0x1a7c0, 0x1a7c1, 0x1a7c4, 0x1a7c5, + 0x1a7c8, 0x1a7c9, 0x1a7cc, 0x1a7cd, 0x1a7d0, 0x1a7d1, 0x1a7d4, 0x1a7d5, + 0x1a7d8, 0x1a7d9, 0x1a7fc, 0x1a7fd, 0x1a800, 0x1a802, 0x1a804, 0x1a804, + 0x1a816, 0x1a816, 0x1a81e, 0x1a81e, 0x1a826, 0x1a826, 0x1a82e, 0x1a82e, + 0x1a836, 0x1a836, 0x1a83e, 0x1a83e, 0x1a846, 0x1a846, 0x1a860, 0x1a862, + 0x1a864, 0x1a867, 0x1a870, 0x1a870, 0x1a883, 0x1a884, 0x1a8c0, 0x1a8c2, + 0x1a8c4, 0x1a8c7, 0x1a8d0, 0x1a8d3, 0x1a900, 0x1a92b, 0x1a940, 0x1a940, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_gmugx_registers), 8)); + +static const u32 gen7_2_0_noncontext_pipe_br_registers[] = { + 0x00887, 0x0088c, 0x08600, 0x08600, 0x08602, 0x08602, 0x08610, 0x0861b, + 0x08620, 0x08620, 0x08630, 0x08630, 0x08637, 0x08639, 0x08640, 0x08640, + 0x09600, 0x09600, 0x09602, 0x09603, 0x0960a, 0x09616, 0x09624, 0x0963a, + 0x09640, 0x09640, 0x09e00, 0x09e00, 0x09e02, 0x09e07, 0x09e0a, 0x09e16, + 0x09e19, 0x09e19, 0x09e1c, 0x09e1c, 0x09e20, 0x09e25, 0x09e30, 0x09e31, + 0x09e40, 0x09e51, 0x09e64, 0x09e64, 0x09e70, 0x09e72, 0x09e78, 0x09e79, + 0x09e80, 0x09fff, 0x0a600, 0x0a600, 0x0a603, 0x0a603, 0x0a610, 0x0a61f, + 0x0a630, 0x0a631, 0x0a638, 0x0a63c, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_noncontext_pipe_br_registers), 8)); + +static const u32 gen7_2_0_noncontext_pipe_bv_registers[] = { + 0x00887, 0x0088c, 0x08600, 0x08600, 0x08602, 0x08602, 0x08610, 0x0861b, + 0x08620, 0x08620, 0x08630, 0x08630, 0x08637, 0x08639, 0x08640, 0x08640, + 0x09600, 0x09600, 0x09602, 0x09603, 0x0960a, 0x09616, 0x09624, 0x0963a, + 0x09640, 0x09640, 0x09e00, 0x09e00, 0x09e02, 0x09e07, 0x09e0a, 0x09e16, + 0x09e19, 0x09e19, 0x09e1c, 0x09e1c, 0x09e20, 0x09e25, 0x09e30, 0x09e31, + 0x09e40, 0x09e51, 0x09e64, 0x09e64, 0x09e70, 0x09e72, 0x09e78, 0x09e79, + 0x09e80, 0x09fff, 0x0a600, 0x0a600, 0x0a603, 0x0a603, 0x0a610, 0x0a61f, + 0x0a630, 0x0a631, 0x0a638, 0x0a63c, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_noncontext_pipe_bv_registers), 8)); + +static const u32 gen7_2_0_noncontext_rb_rac_pipe_br_registers[] = { + 0x08e10, 0x08e1c, 0x08e20, 0x08e25, 0x08e51, 0x08e5a, 0x08ea0, 0x08ea3, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_noncontext_rb_rac_pipe_br_registers), 8)); + +static const u32 gen7_2_0_noncontext_rb_rbp_pipe_br_registers[] = { + 0x08e01, 0x08e01, 0x08e04, 0x08e04, 0x08e06, 0x08e09, 0x08e0c, 0x08e0c, + 0x08e28, 0x08e28, 0x08e2c, 0x08e35, 0x08e3b, 0x08e40, 0x08e50, 0x08e50, + 0x08e5b, 0x08e5d, 0x08e5f, 0x08e5f, 0x08e61, 0x08e61, 0x08e63, 0x08e66, + 0x08e68, 0x08e69, 0x08e70, 0x08e79, 0x08e80, 0x08e8f, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_noncontext_rb_rbp_pipe_br_registers), 8)); + +static const u32 gen7_2_0_gras_cluster_gras_pipe_br_registers[] = { + 0x08000, 0x0800c, 0x08010, 0x08092, 0x08094, 0x08099, 0x0809b, 0x0809d, + 0x080a0, 0x080a7, 0x080af, 0x080f1, 0x080f4, 0x080f6, 0x080f8, 0x080fa, + 0x08100, 0x08107, 0x08109, 0x0810b, 0x08110, 0x08113, 0x08120, 0x0813f, + 0x08400, 0x08406, 0x0840a, 0x0840b, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_gras_cluster_gras_pipe_br_registers), 8)); + +static const u32 gen7_2_0_gras_cluster_gras_pipe_bv_registers[] = { + 0x08000, 0x0800c, 0x08010, 0x08092, 0x08094, 0x08099, 0x0809b, 0x0809d, + 0x080a0, 0x080a7, 0x080af, 0x080f1, 0x080f4, 0x080f6, 0x080f8, 0x080fa, + 0x08100, 0x08107, 0x08109, 0x0810b, 0x08110, 0x08113, 0x08120, 0x0813f, + 0x08400, 0x08406, 0x0840a, 0x0840b, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_gras_cluster_gras_pipe_bv_registers), 8)); + +static const u32 gen7_2_0_rb_rac_cluster_ps_pipe_br_registers[] = { + 0x08802, 0x08802, 0x08804, 0x08806, 0x08809, 0x0880a, 0x0880e, 0x08811, + 0x08818, 0x0881e, 0x08821, 0x08821, 0x08823, 0x08826, 0x08829, 0x08829, + 0x0882b, 0x0882e, 0x08831, 0x08831, 0x08833, 0x08836, 0x08839, 0x08839, + 0x0883b, 0x0883e, 0x08841, 0x08841, 0x08843, 0x08846, 0x08849, 0x08849, + 0x0884b, 0x0884e, 0x08851, 0x08851, 0x08853, 0x08856, 0x08859, 0x08859, + 0x0885b, 0x0885e, 0x08860, 0x08864, 0x08870, 0x08870, 0x08873, 0x08876, + 0x08878, 0x08879, 0x08882, 0x08885, 0x08887, 0x08889, 0x08891, 0x08891, + 0x08898, 0x08899, 0x088c0, 0x088c1, 0x088e5, 0x088e5, 0x088f4, 0x088f5, + 0x08a00, 0x08a05, 0x08a10, 0x08a15, 0x08a20, 0x08a25, 0x08a30, 0x08a35, + 0x08c00, 0x08c01, 0x08c18, 0x08c1f, 0x08c26, 0x08c34, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_rb_rac_cluster_ps_pipe_br_registers), 8)); + +static const u32 gen7_2_0_sp_cluster_sp_ps_pipe_br_hlsq_state_registers[] = { + 0x0a980, 0x0a984, 0x0a99e, 0x0a99e, 0x0a9a7, 0x0a9a7, 0x0a9aa, 0x0a9aa, + 0x0a9ae, 0x0a9b0, 0x0a9b2, 0x0a9b5, 0x0a9ba, 0x0a9ba, 0x0a9bc, 0x0a9bc, + 0x0a9c4, 0x0a9c4, 0x0a9cd, 0x0a9cd, 0x0a9e0, 0x0a9fc, 0x0aa00, 0x0aa00, + 0x0aa30, 0x0aa31, 0x0aa40, 0x0aabf, 0x0ab00, 0x0ab03, 0x0ab05, 0x0ab05, + 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_sp_cluster_sp_ps_pipe_br_hlsq_state_registers), 8)); + +static const u32 gen7_2_0_sp_cluster_sp_ps_pipe_br_sp_top_registers[] = { + 0x0a980, 0x0a980, 0x0a982, 0x0a984, 0x0a99e, 0x0a9a2, 0x0a9a7, 0x0a9a8, + 0x0a9aa, 0x0a9aa, 0x0a9ae, 0x0a9ae, 0x0a9b0, 0x0a9b1, 0x0a9b3, 0x0a9b5, + 0x0a9ba, 0x0a9bc, 0x0a9c5, 0x0a9c5, 0x0a9e0, 0x0a9f9, 0x0aa00, 0x0aa01, + 0x0ab00, 0x0ab00, 0x0ab02, 0x0ab02, 0x0ab04, 0x0ab05, 0x0ab0a, 0x0ab1b, + 0x0ab20, 0x0ab20, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_sp_cluster_sp_ps_pipe_br_sp_top_registers), 8)); + +static const u32 gen7_2_0_sp_cluster_sp_ps_pipe_br_usptp_registers[] = { + 0x0a980, 0x0a982, 0x0a985, 0x0a9a6, 0x0a9a8, 0x0a9a9, 0x0a9ab, 0x0a9ae, + 0x0a9b0, 0x0a9b3, 0x0a9b6, 0x0a9b9, 0x0a9bb, 0x0a9bf, 0x0a9c2, 0x0a9c3, + 0x0a9c5, 0x0a9c5, 0x0a9cd, 0x0a9cd, 0x0a9d0, 0x0a9d3, 0x0aa01, 0x0aa01, + 0x0aa30, 0x0aa31, 0x0aa40, 0x0aabf, 0x0ab00, 0x0ab05, 0x0ab21, 0x0ab22, + 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_sp_cluster_sp_ps_pipe_br_usptp_registers), 8)); + +static const u32 gen7_2_0_sp_cluster_sp_ps_pipe_lpac_hlsq_state_registers[] = { + 0x0a9b0, 0x0a9b0, 0x0a9b2, 0x0a9b5, 0x0a9ba, 0x0a9ba, 0x0a9bc, 0x0a9bc, + 0x0a9c4, 0x0a9c4, 0x0a9cd, 0x0a9cd, 0x0a9e2, 0x0a9e3, 0x0a9e6, 0x0a9fc, + 0x0aa00, 0x0aa00, 0x0aa31, 0x0aa31, 0x0ab00, 0x0ab01, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_sp_cluster_sp_ps_pipe_lpac_hlsq_state_registers), 8)); + +static const u32 gen7_2_0_sp_cluster_sp_ps_pipe_lpac_sp_top_registers[] = { + 0x0a9b0, 0x0a9b1, 0x0a9b3, 0x0a9b5, 0x0a9ba, 0x0a9bc, 0x0a9c5, 0x0a9c5, + 0x0a9e2, 0x0a9e3, 0x0a9e6, 0x0a9f9, 0x0aa00, 0x0aa00, 0x0ab00, 0x0ab00, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_sp_cluster_sp_ps_pipe_lpac_sp_top_registers), 8)); + +static const u32 gen7_2_0_sp_cluster_sp_ps_pipe_lpac_usptp_registers[] = { + 0x0a9b0, 0x0a9b3, 0x0a9b6, 0x0a9b9, 0x0a9bb, 0x0a9be, 0x0a9c2, 0x0a9c3, + 0x0a9c5, 0x0a9c5, 0x0a9cd, 0x0a9cd, 0x0a9d0, 0x0a9d3, 0x0aa31, 0x0aa31, + 0x0ab00, 0x0ab01, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_sp_cluster_sp_ps_pipe_lpac_usptp_registers), 8)); + +static const u32 gen7_2_0_sp_cluster_sp_vs_pipe_br_hlsq_state_registers[] = { + 0x0a800, 0x0a801, 0x0a81b, 0x0a81d, 0x0a822, 0x0a822, 0x0a824, 0x0a824, + 0x0a827, 0x0a82a, 0x0a830, 0x0a830, 0x0a832, 0x0a835, 0x0a83a, 0x0a83a, + 0x0a83c, 0x0a83c, 0x0a83f, 0x0a841, 0x0a85b, 0x0a85d, 0x0a862, 0x0a862, + 0x0a864, 0x0a864, 0x0a867, 0x0a867, 0x0a870, 0x0a870, 0x0a872, 0x0a872, + 0x0a88c, 0x0a88e, 0x0a893, 0x0a893, 0x0a895, 0x0a895, 0x0a898, 0x0a898, + 0x0a89a, 0x0a89d, 0x0a8a0, 0x0a8af, 0x0a8c0, 0x0a8c3, 0x0ab00, 0x0ab03, + 0x0ab05, 0x0ab05, 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_sp_cluster_sp_vs_pipe_br_hlsq_state_registers), 8)); + +static const u32 gen7_2_0_sp_cluster_sp_vs_pipe_br_sp_top_registers[] = { + 0x0a800, 0x0a800, 0x0a81c, 0x0a81d, 0x0a822, 0x0a824, 0x0a82d, 0x0a82d, + 0x0a82f, 0x0a831, 0x0a834, 0x0a835, 0x0a83a, 0x0a83c, 0x0a840, 0x0a840, + 0x0a85c, 0x0a85d, 0x0a862, 0x0a864, 0x0a868, 0x0a868, 0x0a870, 0x0a871, + 0x0a88d, 0x0a88e, 0x0a893, 0x0a895, 0x0a899, 0x0a899, 0x0a8a0, 0x0a8af, + 0x0ab00, 0x0ab00, 0x0ab02, 0x0ab02, 0x0ab04, 0x0ab05, 0x0ab0a, 0x0ab1b, + 0x0ab20, 0x0ab20, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_sp_cluster_sp_vs_pipe_br_sp_top_registers), 8)); + +static const u32 gen7_2_0_sp_cluster_sp_vs_pipe_br_usptp_registers[] = { + 0x0a800, 0x0a81b, 0x0a81e, 0x0a821, 0x0a823, 0x0a827, 0x0a82d, 0x0a82d, + 0x0a82f, 0x0a833, 0x0a836, 0x0a839, 0x0a83b, 0x0a85b, 0x0a85e, 0x0a861, + 0x0a863, 0x0a868, 0x0a870, 0x0a88c, 0x0a88f, 0x0a892, 0x0a894, 0x0a899, + 0x0a8c0, 0x0a8c3, 0x0ab00, 0x0ab05, 0x0ab21, 0x0ab22, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_sp_cluster_sp_vs_pipe_br_usptp_registers), 8)); + +static const u32 gen7_2_0_sp_cluster_sp_vs_pipe_bv_hlsq_state_registers[] = { + 0x0a800, 0x0a801, 0x0a81b, 0x0a81d, 0x0a822, 0x0a822, 0x0a824, 0x0a824, + 0x0a827, 0x0a82a, 0x0a830, 0x0a830, 0x0a832, 0x0a835, 0x0a83a, 0x0a83a, + 0x0a83c, 0x0a83c, 0x0a83f, 0x0a841, 0x0a85b, 0x0a85d, 0x0a862, 0x0a862, + 0x0a864, 0x0a864, 0x0a867, 0x0a867, 0x0a870, 0x0a870, 0x0a872, 0x0a872, + 0x0a88c, 0x0a88e, 0x0a893, 0x0a893, 0x0a895, 0x0a895, 0x0a898, 0x0a898, + 0x0a89a, 0x0a89d, 0x0a8a0, 0x0a8af, 0x0a8c0, 0x0a8c3, 0x0ab00, 0x0ab02, + 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_sp_cluster_sp_vs_pipe_bv_hlsq_state_registers), 8)); + +static const u32 gen7_2_0_sp_cluster_sp_vs_pipe_bv_sp_top_registers[] = { + 0x0a800, 0x0a800, 0x0a81c, 0x0a81d, 0x0a822, 0x0a824, 0x0a82d, 0x0a82d, + 0x0a82f, 0x0a831, 0x0a834, 0x0a835, 0x0a83a, 0x0a83c, 0x0a840, 0x0a840, + 0x0a85c, 0x0a85d, 0x0a862, 0x0a864, 0x0a868, 0x0a868, 0x0a870, 0x0a871, + 0x0a88d, 0x0a88e, 0x0a893, 0x0a895, 0x0a899, 0x0a899, 0x0a8a0, 0x0a8af, + 0x0ab00, 0x0ab00, 0x0ab02, 0x0ab02, 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_sp_cluster_sp_vs_pipe_bv_sp_top_registers), 8)); + +static const u32 gen7_2_0_sp_cluster_sp_vs_pipe_bv_usptp_registers[] = { + 0x0a800, 0x0a81b, 0x0a81e, 0x0a821, 0x0a823, 0x0a827, 0x0a82d, 0x0a82d, + 0x0a82f, 0x0a833, 0x0a836, 0x0a839, 0x0a83b, 0x0a85b, 0x0a85e, 0x0a861, + 0x0a863, 0x0a868, 0x0a870, 0x0a88c, 0x0a88f, 0x0a892, 0x0a894, 0x0a899, + 0x0a8c0, 0x0a8c3, 0x0ab00, 0x0ab02, 0x0ab21, 0x0ab22, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_sp_cluster_sp_vs_pipe_bv_usptp_registers), 8)); + +static const u32 gen7_2_0_sp_noncontext_pipe_lpac_hlsq_state_registers[] = { + 0x0af88, 0x0af8b, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_sp_noncontext_pipe_lpac_hlsq_state_registers), 8)); + +static const struct gen7_sel_reg gen7_2_0_rb_rac_sel = { + .host_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_HOST, + .cd_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_CD, + .val = 0x0, +}; + +static const struct gen7_sel_reg gen7_2_0_rb_rbp_sel = { + .host_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_HOST, + .cd_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_CD, + .val = 0x9, +}; + +static const struct gen7_cluster_registers gen7_2_0_clusters[] = { + { A7XX_CLUSTER_NONE, PIPE_BR, STATE_NON_CONTEXT, + gen7_2_0_noncontext_pipe_br_registers, }, + { A7XX_CLUSTER_NONE, PIPE_BV, STATE_NON_CONTEXT, + gen7_2_0_noncontext_pipe_bv_registers, }, + { A7XX_CLUSTER_NONE, PIPE_LPAC, STATE_NON_CONTEXT, + gen7_0_0_noncontext_pipe_lpac_registers, }, + { A7XX_CLUSTER_NONE, PIPE_BR, STATE_NON_CONTEXT, + gen7_2_0_noncontext_rb_rac_pipe_br_registers, &gen7_2_0_rb_rac_sel, }, + { A7XX_CLUSTER_NONE, PIPE_BR, STATE_NON_CONTEXT, + gen7_2_0_noncontext_rb_rbp_pipe_br_registers, &gen7_2_0_rb_rbp_sel, }, + { A7XX_CLUSTER_GRAS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_2_0_gras_cluster_gras_pipe_br_registers, }, + { A7XX_CLUSTER_GRAS, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_2_0_gras_cluster_gras_pipe_bv_registers, }, + { A7XX_CLUSTER_GRAS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_2_0_gras_cluster_gras_pipe_br_registers, }, + { A7XX_CLUSTER_GRAS, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_2_0_gras_cluster_gras_pipe_bv_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_pc_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_0_0_pc_cluster_fe_pipe_bv_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_pc_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_0_0_pc_cluster_fe_pipe_bv_registers, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_2_0_rb_rac_cluster_ps_pipe_br_registers, &gen7_2_0_rb_rac_sel, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_2_0_rb_rac_cluster_ps_pipe_br_registers, &gen7_2_0_rb_rac_sel, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_rb_rbp_cluster_ps_pipe_br_registers, &gen7_2_0_rb_rbp_sel, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_rb_rbp_cluster_ps_pipe_br_registers, &gen7_2_0_rb_rbp_sel, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_vfd_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_0_0_vfd_cluster_fe_pipe_bv_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_vfd_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_0_0_vfd_cluster_fe_pipe_bv_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_fe_pipe_bv_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_fe_pipe_br_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_fe_pipe_bv_registers, }, + { A7XX_CLUSTER_PC_VS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_pc_vs_pipe_br_registers, }, + { A7XX_CLUSTER_PC_VS, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_pc_vs_pipe_bv_registers, }, + { A7XX_CLUSTER_PC_VS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_pc_vs_pipe_br_registers, }, + { A7XX_CLUSTER_PC_VS, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_pc_vs_pipe_bv_registers, }, + { A7XX_CLUSTER_VPC_PS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_vpc_ps_pipe_br_registers, }, + { A7XX_CLUSTER_VPC_PS, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_0_0_vpc_cluster_vpc_ps_pipe_bv_registers, }, + { A7XX_CLUSTER_VPC_PS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_vpc_ps_pipe_br_registers, }, + { A7XX_CLUSTER_VPC_PS, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_0_0_vpc_cluster_vpc_ps_pipe_bv_registers, }, +}; + +static const struct gen7_sptp_cluster_registers gen7_2_0_sptp_clusters[] = { + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_BR, 0, A7XX_HLSQ_STATE, + gen7_0_0_sp_noncontext_pipe_br_hlsq_state_registers, 0xae00 }, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_BR, 0, A7XX_SP_TOP, + gen7_0_0_sp_noncontext_pipe_br_sp_top_registers, 0xae00 }, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_0_0_sp_noncontext_pipe_br_usptp_registers, 0xae00 }, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_LPAC, 0, A7XX_HLSQ_STATE, + gen7_2_0_sp_noncontext_pipe_lpac_hlsq_state_registers, 0xaf80 }, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_LPAC, 0, A7XX_SP_TOP, + gen7_0_0_sp_noncontext_pipe_lpac_sp_top_registers, 0xaf80 }, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_LPAC, 0, A7XX_USPTP, + gen7_0_0_sp_noncontext_pipe_lpac_usptp_registers, 0xaf80 }, + { A7XX_CLUSTER_NONE, A7XX_TP0_NCTX_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_0_0_tpl1_noncontext_pipe_br_registers, 0xb600 }, + { A7XX_CLUSTER_NONE, A7XX_TP0_NCTX_REG, PIPE_NONE, 0, A7XX_USPTP, + gen7_0_0_tpl1_noncontext_pipe_none_registers, 0xb600 }, + { A7XX_CLUSTER_NONE, A7XX_TP0_NCTX_REG, PIPE_LPAC, 0, A7XX_USPTP, + gen7_0_0_tpl1_noncontext_pipe_lpac_registers, 0xb780 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_HLSQ_STATE, + gen7_2_0_sp_cluster_sp_ps_pipe_br_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_HLSQ_DP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_dp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_SP_TOP, + gen7_2_0_sp_cluster_sp_ps_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_2_0_sp_cluster_sp_ps_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_HLSQ_STATE, + gen7_2_0_sp_cluster_sp_ps_pipe_br_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_HLSQ_DP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_dp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_SP_TOP, + gen7_2_0_sp_cluster_sp_ps_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_2_0_sp_cluster_sp_ps_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_HLSQ_DP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_dp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_HLSQ_DP, + gen7_0_0_sp_cluster_sp_ps_pipe_br_hlsq_dp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_SP_TOP, + gen7_2_0_sp_cluster_sp_ps_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_USPTP, + gen7_2_0_sp_cluster_sp_ps_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_SP_TOP, + gen7_2_0_sp_cluster_sp_ps_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_USPTP, + gen7_2_0_sp_cluster_sp_ps_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_HLSQ_STATE, + gen7_2_0_sp_cluster_sp_ps_pipe_lpac_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_HLSQ_DP, + gen7_0_0_sp_cluster_sp_ps_pipe_lpac_hlsq_dp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_SP_TOP, + gen7_2_0_sp_cluster_sp_ps_pipe_lpac_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_USPTP, + gen7_2_0_sp_cluster_sp_ps_pipe_lpac_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_HLSQ_STATE, + gen7_2_0_sp_cluster_sp_vs_pipe_br_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BV, 0, A7XX_HLSQ_STATE, + gen7_2_0_sp_cluster_sp_vs_pipe_bv_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_SP_TOP, + gen7_2_0_sp_cluster_sp_vs_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BV, 0, A7XX_SP_TOP, + gen7_2_0_sp_cluster_sp_vs_pipe_bv_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_2_0_sp_cluster_sp_vs_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BV, 0, A7XX_USPTP, + gen7_2_0_sp_cluster_sp_vs_pipe_bv_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_HLSQ_STATE, + gen7_2_0_sp_cluster_sp_vs_pipe_br_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BV, 1, A7XX_HLSQ_STATE, + gen7_2_0_sp_cluster_sp_vs_pipe_bv_hlsq_state_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_SP_TOP, + gen7_2_0_sp_cluster_sp_vs_pipe_br_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BV, 1, A7XX_SP_TOP, + gen7_2_0_sp_cluster_sp_vs_pipe_bv_sp_top_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_2_0_sp_cluster_sp_vs_pipe_br_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BV, 1, A7XX_USPTP, + gen7_2_0_sp_cluster_sp_vs_pipe_bv_usptp_registers, 0xa800 }, + { A7XX_CLUSTER_SP_PS, A7XX_TP0_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_ps_pipe_br_registers, 0xb000 }, + { A7XX_CLUSTER_SP_PS, A7XX_TP0_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_ps_pipe_br_registers, 0xb000 }, + { A7XX_CLUSTER_SP_PS, A7XX_TP0_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_ps_pipe_br_registers, 0xb000 }, + { A7XX_CLUSTER_SP_PS, A7XX_TP0_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_ps_pipe_br_registers, 0xb000 }, + { A7XX_CLUSTER_SP_PS, A7XX_TP0_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_ps_pipe_lpac_registers, 0xb000 }, + { A7XX_CLUSTER_SP_VS, A7XX_TP0_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_vs_pipe_br_registers, 0xb000 }, + { A7XX_CLUSTER_SP_VS, A7XX_TP0_CTX0_3D_CVS_REG, PIPE_BV, 0, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_vs_pipe_bv_registers, 0xb000 }, + { A7XX_CLUSTER_SP_VS, A7XX_TP0_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_vs_pipe_br_registers, 0xb000 }, + { A7XX_CLUSTER_SP_VS, A7XX_TP0_CTX1_3D_CVS_REG, PIPE_BV, 1, A7XX_USPTP, + gen7_0_0_tpl1_cluster_sp_vs_pipe_bv_registers, 0xb000 }, +}; + +static const u32 gen7_2_0_dbgc_registers[] = { + 0x005ff, 0x0061c, 0x0061e, 0x00634, 0x00640, 0x0065e, 0x00679, 0x0067e, + 0x00699, 0x00699, 0x0069b, 0x0069e, 0x006a0, 0x006a3, 0x006c0, 0x006c1, + 0x18400, 0x1841c, 0x1841e, 0x18434, 0x18440, 0x1845c, 0x18479, 0x1847c, + 0x18580, 0x18581, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_dbgc_registers), 8)); + +static const u32 gen7_2_0_rscc_registers[] = { + 0x14000, 0x14036, 0x14040, 0x14047, 0x14080, 0x14084, 0x14089, 0x1408c, + 0x14091, 0x14094, 0x14099, 0x1409c, 0x140a1, 0x140a4, 0x140a9, 0x140ac, + 0x14100, 0x14104, 0x14114, 0x14119, 0x14124, 0x14132, 0x14154, 0x1416b, + 0x14340, 0x14342, 0x14344, 0x1437c, 0x143f0, 0x143f8, 0x143fa, 0x143fe, + 0x14400, 0x14404, 0x14406, 0x1440a, 0x1440c, 0x14410, 0x14412, 0x14416, + 0x14418, 0x1441c, 0x1441e, 0x14422, 0x14424, 0x14424, 0x14498, 0x144a0, + 0x144a2, 0x144a6, 0x144a8, 0x144ac, 0x144ae, 0x144b2, 0x144b4, 0x144b8, + 0x144ba, 0x144be, 0x144c0, 0x144c4, 0x144c6, 0x144ca, 0x144cc, 0x144cc, + 0x14540, 0x14548, 0x1454a, 0x1454e, 0x14550, 0x14554, 0x14556, 0x1455a, + 0x1455c, 0x14560, 0x14562, 0x14566, 0x14568, 0x1456c, 0x1456e, 0x14572, + 0x14574, 0x14574, 0x145e8, 0x145f0, 0x145f2, 0x145f6, 0x145f8, 0x145fc, + 0x145fe, 0x14602, 0x14604, 0x14608, 0x1460a, 0x1460e, 0x14610, 0x14614, + 0x14616, 0x1461a, 0x1461c, 0x1461c, 0x14690, 0x14698, 0x1469a, 0x1469e, + 0x146a0, 0x146a4, 0x146a6, 0x146aa, 0x146ac, 0x146b0, 0x146b2, 0x146b6, + 0x146b8, 0x146bc, 0x146be, 0x146c2, 0x146c4, 0x146c4, 0x14738, 0x14740, + 0x14742, 0x14746, 0x14748, 0x1474c, 0x1474e, 0x14752, 0x14754, 0x14758, + 0x1475a, 0x1475e, 0x14760, 0x14764, 0x14766, 0x1476a, 0x1476c, 0x1476c, + 0x147e0, 0x147e8, 0x147ea, 0x147ee, 0x147f0, 0x147f4, 0x147f6, 0x147fa, + 0x147fc, 0x14800, 0x14802, 0x14806, 0x14808, 0x1480c, 0x1480e, 0x14812, + 0x14814, 0x14814, 0x14888, 0x14890, 0x14892, 0x14896, 0x14898, 0x1489c, + 0x1489e, 0x148a2, 0x148a4, 0x148a8, 0x148aa, 0x148ae, 0x148b0, 0x148b4, + 0x148b6, 0x148ba, 0x148bc, 0x148bc, 0x14930, 0x14938, 0x1493a, 0x1493e, + 0x14940, 0x14944, 0x14946, 0x1494a, 0x1494c, 0x14950, 0x14952, 0x14956, + 0x14958, 0x1495c, 0x1495e, 0x14962, 0x14964, 0x14964, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_rscc_registers), 8)); + +static const u32 gen7_2_0_cpr_registers[] = { + 0x26800, 0x26805, 0x26808, 0x2680c, 0x26814, 0x26814, 0x2681c, 0x2681c, + 0x26820, 0x26838, 0x26840, 0x26840, 0x26848, 0x26848, 0x26850, 0x26850, + 0x26880, 0x2689e, 0x26980, 0x269b0, 0x269c0, 0x269c8, 0x269e0, 0x269ee, + 0x269fb, 0x269ff, 0x26a02, 0x26a07, 0x26a09, 0x26a0b, 0x26a10, 0x26b0f, + 0x27440, 0x27441, 0x27444, 0x27444, 0x27480, 0x274a2, 0x274ac, 0x274ad, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_cpr_registers), 8)); + +static const u32 gen7_2_0_dpm_lkg_registers[] = { + 0x21c00, 0x21c00, 0x21c08, 0x21c09, 0x21c0e, 0x21c0f, 0x21c4f, 0x21c50, + 0x21c52, 0x21c52, 0x21c54, 0x21c56, 0x21c58, 0x21c5a, 0x21c5c, 0x21c60, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_dpm_lkg_registers), 8)); + +static const u32 gen7_2_0_gpucc_registers[] = { + 0x24000, 0x2400f, 0x24400, 0x2440f, 0x24800, 0x24805, 0x24c00, 0x24cff, + 0x25400, 0x25404, 0x25800, 0x25804, 0x25c00, 0x25c04, 0x26000, 0x26004, + 0x26400, 0x26405, 0x26414, 0x2641d, 0x2642a, 0x26430, 0x26432, 0x26433, + 0x26441, 0x2644b, 0x2644d, 0x26457, 0x26466, 0x26468, 0x26478, 0x2647a, + 0x26489, 0x2648a, 0x2649c, 0x2649e, 0x264a0, 0x264a4, 0x264c5, 0x264c7, + 0x264d6, 0x264d8, 0x264e8, 0x264e9, 0x264f9, 0x264fc, 0x2651c, 0x2651e, + 0x26540, 0x26576, 0x26600, 0x26616, 0x26620, 0x2662d, 0x26630, 0x26631, + 0x26635, 0x26635, 0x26637, 0x26637, 0x2663a, 0x2663a, 0x26642, 0x26642, + 0x26656, 0x26658, 0x2665b, 0x2665d, 0x2665f, 0x26662, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_gpucc_registers), 8)); + +static const u32 gen7_2_0_cx_misc_registers[] = { + 0x27800, 0x27800, 0x27810, 0x27814, 0x27820, 0x27824, 0x27832, 0x27857, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_cx_misc_registers), 8)); + +static const u32 gen7_2_0_dpm_registers[] = { + 0x1aa00, 0x1aa06, 0x1aa09, 0x1aa0a, 0x1aa0c, 0x1aa0d, 0x1aa0f, 0x1aa12, + 0x1aa14, 0x1aa47, 0x1aa50, 0x1aa51, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_2_0_dpm_registers), 8)); + +static const struct gen7_reg_list gen7_2_0_reg_list[] = { + { gen7_2_0_gpu_registers, NULL }, + { gen7_2_0_cx_misc_registers, NULL }, + { gen7_2_0_dpm_registers, NULL }, + { gen7_2_0_dbgc_registers, NULL }, + { NULL, NULL }, +}; + +static const u32 *gen7_2_0_external_core_regs[] = { + gen7_2_0_gpucc_registers, + gen7_2_0_cpr_registers, + gen7_2_0_dpm_lkg_registers, +}; +#endif /*_ADRENO_GEN7_2_0_SNAPSHOT_H */ diff --git a/drivers/gpu/drm/msm/adreno/adreno_gen7_9_0_snapshot.h b/drivers/gpu/drm/msm/adreno/adreno_gen7_9_0_snapshot.h new file mode 100644 index 000000000..20125d1aa --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/adreno_gen7_9_0_snapshot.h @@ -0,0 +1,1446 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2020-2021, The Linux Foundation. All rights reserved. + * Copyright (c) 2022-2023 Qualcomm Innovation Center, Inc. All rights reserved. + */ +#ifndef __ADRENO_GEN7_9_0_SNAPSHOT_H +#define __ADRENO_GEN7_9_0_SNAPSHOT_H + +#include "a6xx_gpu_state.h" + +static const u32 gen7_9_0_debugbus_blocks[] = { + A7XX_DBGBUS_CP_0_0, + A7XX_DBGBUS_CP_0_1, + A7XX_DBGBUS_RBBM, + A7XX_DBGBUS_HLSQ, + A7XX_DBGBUS_UCHE_0, + A7XX_DBGBUS_UCHE_1, + A7XX_DBGBUS_TESS_BR, + A7XX_DBGBUS_TESS_BV, + A7XX_DBGBUS_PC_BR, + A7XX_DBGBUS_PC_BV, + A7XX_DBGBUS_VFDP_BR, + A7XX_DBGBUS_VFDP_BV, + A7XX_DBGBUS_VPC_BR, + A7XX_DBGBUS_VPC_BV, + A7XX_DBGBUS_TSE_BR, + A7XX_DBGBUS_TSE_BV, + A7XX_DBGBUS_RAS_BR, + A7XX_DBGBUS_RAS_BV, + A7XX_DBGBUS_VSC, + A7XX_DBGBUS_COM_0, + A7XX_DBGBUS_LRZ_BR, + A7XX_DBGBUS_LRZ_BV, + A7XX_DBGBUS_UFC_0, + A7XX_DBGBUS_UFC_1, + A7XX_DBGBUS_GMU_GX, + A7XX_DBGBUS_DBGC, + A7XX_DBGBUS_GPC_BR, + A7XX_DBGBUS_GPC_BV, + A7XX_DBGBUS_LARC, + A7XX_DBGBUS_HLSQ_SPTP, + A7XX_DBGBUS_RB_0, + A7XX_DBGBUS_RB_1, + A7XX_DBGBUS_RB_2, + A7XX_DBGBUS_RB_3, + A7XX_DBGBUS_RB_4, + A7XX_DBGBUS_RB_5, + A7XX_DBGBUS_UCHE_WRAPPER, + A7XX_DBGBUS_CCU_0, + A7XX_DBGBUS_CCU_1, + A7XX_DBGBUS_CCU_2, + A7XX_DBGBUS_CCU_3, + A7XX_DBGBUS_CCU_4, + A7XX_DBGBUS_CCU_5, + A7XX_DBGBUS_VFD_BR_0, + A7XX_DBGBUS_VFD_BR_1, + A7XX_DBGBUS_VFD_BR_2, + A7XX_DBGBUS_VFD_BV_0, + A7XX_DBGBUS_VFD_BV_1, + A7XX_DBGBUS_VFD_BV_2, + A7XX_DBGBUS_USP_0, + A7XX_DBGBUS_USP_1, + A7XX_DBGBUS_USP_2, + A7XX_DBGBUS_USP_3, + A7XX_DBGBUS_USP_4, + A7XX_DBGBUS_USP_5, + A7XX_DBGBUS_TP_0, + A7XX_DBGBUS_TP_1, + A7XX_DBGBUS_TP_2, + A7XX_DBGBUS_TP_3, + A7XX_DBGBUS_TP_4, + A7XX_DBGBUS_TP_5, + A7XX_DBGBUS_TP_6, + A7XX_DBGBUS_TP_7, + A7XX_DBGBUS_TP_8, + A7XX_DBGBUS_TP_9, + A7XX_DBGBUS_TP_10, + A7XX_DBGBUS_TP_11, + A7XX_DBGBUS_USPTP_0, + A7XX_DBGBUS_USPTP_1, + A7XX_DBGBUS_USPTP_2, + A7XX_DBGBUS_USPTP_3, + A7XX_DBGBUS_USPTP_4, + A7XX_DBGBUS_USPTP_5, + A7XX_DBGBUS_USPTP_6, + A7XX_DBGBUS_USPTP_7, + A7XX_DBGBUS_USPTP_8, + A7XX_DBGBUS_USPTP_9, + A7XX_DBGBUS_USPTP_10, + A7XX_DBGBUS_USPTP_11, + A7XX_DBGBUS_CCHE_0, + A7XX_DBGBUS_CCHE_1, + A7XX_DBGBUS_CCHE_2, + A7XX_DBGBUS_VPC_DSTR_0, + A7XX_DBGBUS_VPC_DSTR_1, + A7XX_DBGBUS_VPC_DSTR_2, + A7XX_DBGBUS_HLSQ_DP_STR_0, + A7XX_DBGBUS_HLSQ_DP_STR_1, + A7XX_DBGBUS_HLSQ_DP_STR_2, + A7XX_DBGBUS_HLSQ_DP_STR_3, + A7XX_DBGBUS_HLSQ_DP_STR_4, + A7XX_DBGBUS_HLSQ_DP_STR_5, + A7XX_DBGBUS_UFC_DSTR_0, + A7XX_DBGBUS_UFC_DSTR_1, + A7XX_DBGBUS_UFC_DSTR_2, + A7XX_DBGBUS_CGC_SUBCORE, + A7XX_DBGBUS_CGC_CORE, +}; + +static const u32 gen7_9_0_gbif_debugbus_blocks[] = { + A7XX_DBGBUS_GBIF_GX, +}; + +static const u32 gen7_9_0_cx_debugbus_blocks[] = { + A7XX_DBGBUS_CX, + A7XX_DBGBUS_GMU_CX, + A7XX_DBGBUS_GBIF_CX, +}; + +static const struct gen7_shader_block gen7_9_0_shader_blocks[] = { + { A7XX_TP0_TMO_DATA, 0x0200, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_TP0_SMO_DATA, 0x0080, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_TP0_MIPMAP_BASE_DATA, 0x03C0, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_INST_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_INST_DATA_1, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_0_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_1_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_2_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_3_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_4_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_5_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_6_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_7_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_CB_RAM, 0x0390, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_13_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_14_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_INST_TAG, 0x00C0, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_INST_DATA_2, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_TMO_TAG, 0x0080, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_SMO_TAG, 0x0080, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_STATE_DATA, 0x0040, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_HWAVE_RAM, 0x0100, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_L0_INST_BUF, 0x0050, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_8_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_9_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_10_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_11_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_SP_LB_12_DATA, 0x0800, 6, 2, PIPE_BR, A7XX_USPTP }, + { A7XX_HLSQ_DATAPATH_DSTR_META, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_DATAPATH_DSTR_META, 0x0010, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_L2STC_TAG_RAM, 0x0200, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_L2STC_INFO_CMD, 0x0474, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CVS_BE_CTXT_BUF_RAM_TAG, 0x0080, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CVS_BE_CTXT_BUF_RAM_TAG, 0x0080, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CPS_BE_CTXT_BUF_RAM_TAG, 0x0080, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CVS_BE_CTXT_BUF_RAM, 0x0400, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CVS_BE_CTXT_BUF_RAM, 0x0400, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CPS_BE_CTXT_BUF_RAM, 0x0400, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CHUNK_CVS_RAM, 0x01C0, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CHUNK_CVS_RAM, 0x01C0, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CHUNK_CPS_RAM, 0x0300, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CHUNK_CPS_RAM, 0x0180, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CHUNK_CVS_RAM_TAG, 0x0040, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CHUNK_CVS_RAM_TAG, 0x0040, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CHUNK_CPS_RAM_TAG, 0x0040, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CHUNK_CPS_RAM_TAG, 0x0040, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_ICB_CVS_CB_BASE_TAG, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_ICB_CVS_CB_BASE_TAG, 0x0010, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_ICB_CPS_CB_BASE_TAG, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_ICB_CPS_CB_BASE_TAG, 0x0010, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CVS_MISC_RAM, 0x0540, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CVS_MISC_RAM, 0x0540, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CPS_MISC_RAM, 0x0640, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CPS_MISC_RAM, 0x00B0, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CPS_MISC_RAM_1, 0x0800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_INST_RAM, 0x0800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_INST_RAM, 0x0800, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_INST_RAM, 0x0200, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CVS_CONST_RAM, 0x0800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CVS_CONST_RAM, 0x0800, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CPS_CONST_RAM, 0x0800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CPS_CONST_RAM, 0x0800, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CVS_MISC_RAM_TAG, 0x0050, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CVS_MISC_RAM_TAG, 0x0050, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CPS_MISC_RAM_TAG, 0x0050, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_CPS_MISC_RAM_TAG, 0x0008, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_INST_RAM_TAG, 0x0014, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_INST_RAM_TAG, 0x0010, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_INST_RAM_TAG, 0x0004, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CVS_CONST_RAM_TAG, 0x0040, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CVS_CONST_RAM_TAG, 0x0040, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CPS_CONST_RAM_TAG, 0x0040, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_CPS_CONST_RAM_TAG, 0x0020, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_LOCAL_MISC_RAM, 0x03C0, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_LOCAL_MISC_RAM, 0x0280, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_LOCAL_MISC_RAM, 0x0050, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_LOCAL_MISC_RAM_TAG, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_GFX_LOCAL_MISC_RAM_TAG, 0x0008, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_INST_RAM_1, 0x0800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_STPROC_META, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_BV_BE_META, 0x0018, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_BV_BE_META, 0x0018, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_INST_RAM_2, 0x0800, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_DATAPATH_META, 0x0020, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_FRONTEND_META, 0x0080, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_FRONTEND_META, 0x0080, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_FRONTEND_META, 0x0080, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_INDIRECT_META, 0x0010, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_BACKEND_META, 0x0040, 1, 1, PIPE_BR, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_BACKEND_META, 0x0040, 1, 1, PIPE_BV, A7XX_HLSQ_STATE }, + { A7XX_HLSQ_BACKEND_META, 0x0040, 1, 1, PIPE_LPAC, A7XX_HLSQ_STATE }, +}; + +/* + * Block : ['PRE_CRASHDUMPER', 'GBIF'] + * pairs : 2 (Regs:5), 5 (Regs:38) + */ +static const u32 gen7_9_0_pre_crashdumper_gpu_registers[] = { + 0x00210, 0x00213, 0x00536, 0x00536, 0x03c00, 0x03c0b, 0x03c40, 0x03c42, + 0x03c45, 0x03c47, 0x03c49, 0x03c4a, 0x03cc0, 0x03cd1, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_pre_crashdumper_gpu_registers), 8)); + +/* + * Block : ['BROADCAST', 'CP', 'GRAS', 'GXCLKCTL'] + * Block : ['PC', 'RBBM', 'RDVM', 'UCHE'] + * Block : ['VFD', 'VPC', 'VSC'] + * Pipeline: PIPE_NONE + * pairs : 196 (Regs:1778) + */ +static const u32 gen7_9_0_gpu_registers[] = { + 0x00000, 0x00000, 0x00002, 0x00002, 0x00011, 0x00012, 0x00016, 0x0001b, + 0x0001f, 0x00032, 0x00038, 0x0003c, 0x00044, 0x00044, 0x00047, 0x00047, + 0x00049, 0x0004a, 0x0004c, 0x0004c, 0x00056, 0x00056, 0x00073, 0x0007d, + 0x00090, 0x000a8, 0x000ad, 0x000ad, 0x00117, 0x00117, 0x00120, 0x00122, + 0x00130, 0x0013f, 0x00142, 0x0015f, 0x00162, 0x00164, 0x00166, 0x00171, + 0x00173, 0x00174, 0x00176, 0x0017b, 0x0017e, 0x00180, 0x00183, 0x00192, + 0x00195, 0x00196, 0x00199, 0x0019a, 0x0019d, 0x001a2, 0x001aa, 0x001ae, + 0x001b9, 0x001b9, 0x001bb, 0x001bb, 0x001be, 0x001be, 0x001c1, 0x001c2, + 0x001c5, 0x001c5, 0x001c7, 0x001c7, 0x001c9, 0x001c9, 0x001cb, 0x001ce, + 0x001d1, 0x001df, 0x001e1, 0x001e3, 0x001e5, 0x001e5, 0x001e7, 0x001e9, + 0x00200, 0x0020d, 0x00215, 0x00253, 0x00260, 0x00260, 0x00264, 0x00270, + 0x00272, 0x00274, 0x00281, 0x00281, 0x00283, 0x00283, 0x00289, 0x0028d, + 0x00290, 0x002a2, 0x002c0, 0x002c1, 0x00300, 0x00401, 0x00410, 0x00451, + 0x00460, 0x004a3, 0x004c0, 0x004d1, 0x00500, 0x00500, 0x00507, 0x0050b, + 0x0050f, 0x0050f, 0x00511, 0x00511, 0x00533, 0x00535, 0x00540, 0x0055b, + 0x00564, 0x00567, 0x00574, 0x00577, 0x00584, 0x0059b, 0x005fb, 0x005ff, + 0x00800, 0x00808, 0x00810, 0x00813, 0x00820, 0x00821, 0x00823, 0x00827, + 0x00830, 0x00834, 0x0083f, 0x00841, 0x00843, 0x00847, 0x0084f, 0x00886, + 0x008a0, 0x008ab, 0x008c0, 0x008c0, 0x008c4, 0x008c4, 0x008c6, 0x008c6, + 0x008d0, 0x008dd, 0x008e0, 0x008e6, 0x008f0, 0x008f3, 0x00900, 0x00903, + 0x00908, 0x00911, 0x00928, 0x0093e, 0x00942, 0x0094d, 0x00980, 0x00984, + 0x0098d, 0x0098f, 0x009b0, 0x009b4, 0x009c2, 0x009c9, 0x009ce, 0x009d7, + 0x009e0, 0x009e7, 0x00a00, 0x00a00, 0x00a02, 0x00a03, 0x00a10, 0x00a4f, + 0x00a61, 0x00a9f, 0x00ad0, 0x00adb, 0x00b00, 0x00b31, 0x00b35, 0x00b3c, + 0x00b40, 0x00b40, 0x00b70, 0x00b73, 0x00b78, 0x00b79, 0x00b7c, 0x00b7d, + 0x00b80, 0x00b81, 0x00b84, 0x00b85, 0x00b88, 0x00b89, 0x00b8c, 0x00b8d, + 0x00b90, 0x00b93, 0x00b98, 0x00b99, 0x00b9c, 0x00b9d, 0x00ba0, 0x00ba1, + 0x00ba4, 0x00ba5, 0x00ba8, 0x00ba9, 0x00bac, 0x00bad, 0x00bb0, 0x00bb1, + 0x00bb4, 0x00bb5, 0x00bb8, 0x00bb9, 0x00bbc, 0x00bbd, 0x00bc0, 0x00bc1, + 0x00c00, 0x00c00, 0x00c02, 0x00c04, 0x00c06, 0x00c06, 0x00c10, 0x00cd9, + 0x00ce0, 0x00d0c, 0x00df0, 0x00df4, 0x00e01, 0x00e02, 0x00e07, 0x00e0e, + 0x00e10, 0x00e13, 0x00e17, 0x00e19, 0x00e1c, 0x00e2b, 0x00e30, 0x00e32, + 0x00e3a, 0x00e3d, 0x00e50, 0x00e5b, 0x02840, 0x0287f, 0x0ec00, 0x0ec01, + 0x0ec05, 0x0ec05, 0x0ec07, 0x0ec07, 0x0ec0a, 0x0ec0a, 0x0ec12, 0x0ec12, + 0x0ec26, 0x0ec28, 0x0ec2b, 0x0ec2d, 0x0ec2f, 0x0ec2f, 0x0ec40, 0x0ec41, + 0x0ec45, 0x0ec45, 0x0ec47, 0x0ec47, 0x0ec4a, 0x0ec4a, 0x0ec52, 0x0ec52, + 0x0ec66, 0x0ec68, 0x0ec6b, 0x0ec6d, 0x0ec6f, 0x0ec6f, 0x0ec80, 0x0ec81, + 0x0ec85, 0x0ec85, 0x0ec87, 0x0ec87, 0x0ec8a, 0x0ec8a, 0x0ec92, 0x0ec92, + 0x0eca6, 0x0eca8, 0x0ecab, 0x0ecad, 0x0ecaf, 0x0ecaf, 0x0ecc0, 0x0ecc1, + 0x0ecc5, 0x0ecc5, 0x0ecc7, 0x0ecc7, 0x0ecca, 0x0ecca, 0x0ecd2, 0x0ecd2, + 0x0ece6, 0x0ece8, 0x0eceb, 0x0eced, 0x0ecef, 0x0ecef, 0x0ed00, 0x0ed01, + 0x0ed05, 0x0ed05, 0x0ed07, 0x0ed07, 0x0ed0a, 0x0ed0a, 0x0ed12, 0x0ed12, + 0x0ed26, 0x0ed28, 0x0ed2b, 0x0ed2d, 0x0ed2f, 0x0ed2f, 0x0ed40, 0x0ed41, + 0x0ed45, 0x0ed45, 0x0ed47, 0x0ed47, 0x0ed4a, 0x0ed4a, 0x0ed52, 0x0ed52, + 0x0ed66, 0x0ed68, 0x0ed6b, 0x0ed6d, 0x0ed6f, 0x0ed6f, 0x0ed80, 0x0ed81, + 0x0ed85, 0x0ed85, 0x0ed87, 0x0ed87, 0x0ed8a, 0x0ed8a, 0x0ed92, 0x0ed92, + 0x0eda6, 0x0eda8, 0x0edab, 0x0edad, 0x0edaf, 0x0edaf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_gpu_registers), 8)); + +static const u32 gen7_9_0_gxclkctl_registers[] = { + 0x18800, 0x18800, 0x18808, 0x1880b, 0x18820, 0x18822, 0x18830, 0x18830, + 0x18834, 0x1883b, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_gxclkctl_registers), 8)); + +/* + * Block : ['GMUAO', 'GMUCX', 'GMUCX_RAM'] + * Pipeline: PIPE_NONE + * pairs : 134 (Regs:429) + */ +static const u32 gen7_9_0_gmu_registers[] = { + 0x10001, 0x10001, 0x10003, 0x10003, 0x10401, 0x10401, 0x10403, 0x10403, + 0x10801, 0x10801, 0x10803, 0x10803, 0x10c01, 0x10c01, 0x10c03, 0x10c03, + 0x11001, 0x11001, 0x11003, 0x11003, 0x11401, 0x11401, 0x11403, 0x11403, + 0x11801, 0x11801, 0x11803, 0x11803, 0x11c01, 0x11c01, 0x11c03, 0x11c03, + 0x1f400, 0x1f40b, 0x1f40f, 0x1f411, 0x1f500, 0x1f500, 0x1f507, 0x1f507, + 0x1f509, 0x1f50b, 0x1f700, 0x1f701, 0x1f704, 0x1f706, 0x1f708, 0x1f709, + 0x1f70c, 0x1f70d, 0x1f710, 0x1f711, 0x1f713, 0x1f716, 0x1f718, 0x1f71d, + 0x1f720, 0x1f724, 0x1f729, 0x1f729, 0x1f730, 0x1f747, 0x1f750, 0x1f756, + 0x1f758, 0x1f759, 0x1f75c, 0x1f75c, 0x1f760, 0x1f761, 0x1f764, 0x1f76b, + 0x1f770, 0x1f775, 0x1f780, 0x1f785, 0x1f790, 0x1f798, 0x1f7a0, 0x1f7a8, + 0x1f7b0, 0x1f7b3, 0x1f800, 0x1f804, 0x1f807, 0x1f808, 0x1f80b, 0x1f80c, + 0x1f80f, 0x1f80f, 0x1f811, 0x1f811, 0x1f813, 0x1f817, 0x1f819, 0x1f81c, + 0x1f824, 0x1f82a, 0x1f82d, 0x1f830, 0x1f840, 0x1f853, 0x1f860, 0x1f860, + 0x1f862, 0x1f866, 0x1f868, 0x1f869, 0x1f870, 0x1f879, 0x1f87f, 0x1f881, + 0x1f890, 0x1f896, 0x1f8a0, 0x1f8a2, 0x1f8a4, 0x1f8af, 0x1f8b8, 0x1f8b9, + 0x1f8c0, 0x1f8c1, 0x1f8c3, 0x1f8c4, 0x1f8d0, 0x1f8d0, 0x1f8ec, 0x1f8ec, + 0x1f8f0, 0x1f8f1, 0x1f910, 0x1f917, 0x1f920, 0x1f921, 0x1f924, 0x1f925, + 0x1f928, 0x1f929, 0x1f92c, 0x1f92d, 0x1f942, 0x1f944, 0x1f948, 0x1f94a, + 0x1f94f, 0x1f951, 0x1f954, 0x1f955, 0x1f95d, 0x1f95d, 0x1f962, 0x1f96b, + 0x1f970, 0x1f971, 0x1f973, 0x1f977, 0x1f97c, 0x1f97c, 0x1f980, 0x1f981, + 0x1f984, 0x1f986, 0x1f992, 0x1f993, 0x1f996, 0x1f99e, 0x1f9c5, 0x1f9d4, + 0x1f9f0, 0x1f9f1, 0x1f9f8, 0x1f9fa, 0x1f9fc, 0x1f9fc, 0x1fa00, 0x1fa03, + 0x20000, 0x20013, 0x20018, 0x2001a, 0x20020, 0x20021, 0x20024, 0x20025, + 0x2002a, 0x2002c, 0x20030, 0x20031, 0x20034, 0x20036, 0x23801, 0x23801, + 0x23803, 0x23803, 0x23805, 0x23805, 0x23807, 0x23807, 0x23809, 0x23809, + 0x2380b, 0x2380b, 0x2380d, 0x2380d, 0x2380f, 0x2380f, 0x23811, 0x23811, + 0x23813, 0x23813, 0x23815, 0x23815, 0x23817, 0x23817, 0x23819, 0x23819, + 0x2381b, 0x2381b, 0x2381d, 0x2381d, 0x2381f, 0x23820, 0x23822, 0x23822, + 0x23824, 0x23824, 0x23826, 0x23826, 0x23828, 0x23828, 0x2382a, 0x2382a, + 0x2382c, 0x2382c, 0x2382e, 0x2382e, 0x23830, 0x23830, 0x23832, 0x23832, + 0x23834, 0x23834, 0x23836, 0x23836, 0x23838, 0x23838, 0x2383a, 0x2383a, + 0x2383c, 0x2383c, 0x2383e, 0x2383e, 0x23840, 0x23847, 0x23b00, 0x23b01, + 0x23b03, 0x23b03, 0x23b05, 0x23b0e, 0x23b10, 0x23b13, 0x23b15, 0x23b16, + 0x23b28, 0x23b28, 0x23b30, 0x23b30, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_gmu_registers), 8)); + +/* + * Block : ['GMUGX'] + * Pipeline: PIPE_NONE + * pairs : 44 (Regs:454) + */ +static const u32 gen7_9_0_gmugx_registers[] = { + 0x1a400, 0x1a41f, 0x1a440, 0x1a45f, 0x1a480, 0x1a49f, 0x1a4c0, 0x1a4df, + 0x1a500, 0x1a51f, 0x1a540, 0x1a55f, 0x1a580, 0x1a59f, 0x1a600, 0x1a61f, + 0x1a640, 0x1a65f, 0x1a780, 0x1a781, 0x1a783, 0x1a785, 0x1a787, 0x1a789, + 0x1a78b, 0x1a78d, 0x1a78f, 0x1a791, 0x1a793, 0x1a795, 0x1a797, 0x1a799, + 0x1a79b, 0x1a79d, 0x1a79f, 0x1a7a1, 0x1a7a3, 0x1a7a3, 0x1a7a8, 0x1a7b9, + 0x1a7c0, 0x1a7c1, 0x1a7c4, 0x1a7c5, 0x1a7c8, 0x1a7c9, 0x1a7cc, 0x1a7cd, + 0x1a7d0, 0x1a7d1, 0x1a7d4, 0x1a7d5, 0x1a7d8, 0x1a7d9, 0x1a7dc, 0x1a7dd, + 0x1a7e0, 0x1a7e1, 0x1a7fc, 0x1a7fd, 0x1a800, 0x1a808, 0x1a816, 0x1a816, + 0x1a81e, 0x1a81e, 0x1a826, 0x1a826, 0x1a82e, 0x1a82e, 0x1a836, 0x1a836, + 0x1a83e, 0x1a83e, 0x1a846, 0x1a846, 0x1a84e, 0x1a84e, 0x1a856, 0x1a856, + 0x1a883, 0x1a884, 0x1a890, 0x1a8b3, 0x1a900, 0x1a92b, 0x1a940, 0x1a940, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_gmugx_registers), 8)); + +/* + * Block : ['CX_MISC'] + * Pipeline: PIPE_NONE + * pairs : 7 (Regs:56) + */ +static const u32 gen7_9_0_cx_misc_registers[] = { + 0x27800, 0x27800, 0x27810, 0x27814, 0x27820, 0x27824, 0x27828, 0x2782a, + 0x27832, 0x27857, 0x27880, 0x27881, 0x27c00, 0x27c01, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_cx_misc_registers), 8)); + +/* + * Block : ['DBGC'] + * Pipeline: PIPE_NONE + * pairs : 19 (Regs:155) + */ +static const u32 gen7_9_0_dbgc_registers[] = { + 0x00600, 0x0061c, 0x0061e, 0x00634, 0x00640, 0x00643, 0x0064e, 0x00652, + 0x00654, 0x0065e, 0x00699, 0x00699, 0x0069b, 0x0069e, 0x006c2, 0x006e4, + 0x006e6, 0x006e6, 0x006e9, 0x006e9, 0x006eb, 0x006eb, 0x006f1, 0x006f4, + 0x00700, 0x00707, 0x00718, 0x00718, 0x00720, 0x00729, 0x00740, 0x0074a, + 0x00758, 0x00758, 0x00760, 0x00762, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_dbgc_registers), 8)); + +/* + * Block : ['CX_DBGC'] + * Pipeline: PIPE_NONE + * pairs : 7 (Regs:75) + */ +static const u32 gen7_9_0_cx_dbgc_registers[] = { + 0x18400, 0x1841c, 0x1841e, 0x18434, 0x18440, 0x18443, 0x1844e, 0x18452, + 0x18454, 0x1845e, 0x18520, 0x18520, 0x18580, 0x18581, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_cx_dbgc_registers), 8)); + +/* + * Block : ['BROADCAST', 'CP', 'CX_DBGC', 'CX_MISC', 'DBGC', 'GBIF'] + * Block : ['GMUAO', 'GMUCX', 'GMUGX', 'GRAS', 'GXCLKCTL', 'PC'] + * Block : ['RBBM', 'RDVM', 'UCHE', 'VFD', 'VPC', 'VSC'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_NONE + * pairs : 29 (Regs:573) + */ +static const u32 gen7_9_0_non_context_pipe_br_registers[] = { + 0x00887, 0x0088c, 0x08600, 0x08602, 0x08610, 0x0861b, 0x08620, 0x08620, + 0x08630, 0x08630, 0x08637, 0x08639, 0x08640, 0x08640, 0x09600, 0x09603, + 0x0960a, 0x09616, 0x09624, 0x0963a, 0x09640, 0x09640, 0x09e00, 0x09e00, + 0x09e02, 0x09e07, 0x09e0a, 0x09e16, 0x09e18, 0x09e1a, 0x09e1c, 0x09e1c, + 0x09e20, 0x09e25, 0x09e30, 0x09e31, 0x09e40, 0x09e51, 0x09e64, 0x09e6c, + 0x09e70, 0x09e72, 0x09e78, 0x09e79, 0x09e80, 0x09fff, 0x0a600, 0x0a600, + 0x0a603, 0x0a603, 0x0a610, 0x0a61f, 0x0a630, 0x0a631, 0x0a638, 0x0a63c, + 0x0a640, 0x0a65f, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_pipe_br_registers), 8)); + +/* + * Block : ['BROADCAST', 'CP', 'CX_DBGC', 'CX_MISC', 'DBGC', 'GBIF'] + * Block : ['GMUAO', 'GMUCX', 'GMUGX', 'GRAS', 'GXCLKCTL', 'PC'] + * Block : ['RBBM', 'RDVM', 'UCHE', 'VFD', 'VPC', 'VSC'] + * Pipeline: PIPE_BV + * Cluster : A7XX_CLUSTER_NONE + * pairs : 29 (Regs:573) + */ +static const u32 gen7_9_0_non_context_pipe_bv_registers[] = { + 0x00887, 0x0088c, 0x08600, 0x08602, 0x08610, 0x0861b, 0x08620, 0x08620, + 0x08630, 0x08630, 0x08637, 0x08639, 0x08640, 0x08640, 0x09600, 0x09603, + 0x0960a, 0x09616, 0x09624, 0x0963a, 0x09640, 0x09640, 0x09e00, 0x09e00, + 0x09e02, 0x09e07, 0x09e0a, 0x09e16, 0x09e18, 0x09e1a, 0x09e1c, 0x09e1c, + 0x09e20, 0x09e25, 0x09e30, 0x09e31, 0x09e40, 0x09e51, 0x09e64, 0x09e6c, + 0x09e70, 0x09e72, 0x09e78, 0x09e79, 0x09e80, 0x09fff, 0x0a600, 0x0a600, + 0x0a603, 0x0a603, 0x0a610, 0x0a61f, 0x0a630, 0x0a631, 0x0a638, 0x0a63c, + 0x0a640, 0x0a65f, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_pipe_bv_registers), 8)); + +/* + * Block : ['BROADCAST', 'CP', 'CX_DBGC', 'CX_MISC', 'DBGC', 'GBIF'] + * Block : ['GMUAO', 'GMUCX', 'GMUGX', 'GRAS', 'GXCLKCTL', 'PC'] + * Block : ['RBBM', 'RDVM', 'UCHE', 'VFD', 'VPC', 'VSC'] + * Pipeline: PIPE_LPAC + * Cluster : A7XX_CLUSTER_NONE + * pairs : 2 (Regs:7) + */ +static const u32 gen7_9_0_non_context_pipe_lpac_registers[] = { + 0x00887, 0x0088c, 0x00f80, 0x00f80, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_pipe_lpac_registers), 8)); + +/* + * Block : ['RB'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_NONE + * pairs : 5 (Regs:37) + */ +static const u32 gen7_9_0_non_context_rb_pipe_br_rac_registers[] = { + 0x08e10, 0x08e1c, 0x08e20, 0x08e25, 0x08e51, 0x08e5a, 0x08e6a, 0x08e6d, + 0x08ea0, 0x08ea3, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_rb_pipe_br_rac_registers), 8)); + +/* + * Block : ['RB'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_NONE + * pairs : 15 (Regs:66) + */ +static const u32 gen7_9_0_non_context_rb_pipe_br_rbp_registers[] = { + 0x08e01, 0x08e01, 0x08e04, 0x08e04, 0x08e06, 0x08e09, 0x08e0c, 0x08e0c, + 0x08e28, 0x08e28, 0x08e2c, 0x08e35, 0x08e3b, 0x08e40, 0x08e50, 0x08e50, + 0x08e5b, 0x08e5d, 0x08e5f, 0x08e5f, 0x08e61, 0x08e61, 0x08e63, 0x08e66, + 0x08e68, 0x08e69, 0x08e70, 0x08e7d, 0x08e80, 0x08e8f, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_rb_pipe_br_rbp_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_NONE + * Location: A7XX_HLSQ_STATE + * pairs : 4 (Regs:28) + */ +static const u32 gen7_9_0_non_context_sp_pipe_br_hlsq_state_registers[] = { + 0x0ae52, 0x0ae52, 0x0ae60, 0x0ae67, 0x0ae69, 0x0ae75, 0x0aec0, 0x0aec5, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_sp_pipe_br_hlsq_state_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_NONE + * Location: A7XX_SP_TOP + * pairs : 10 (Regs:61) + */ +static const u32 gen7_9_0_non_context_sp_pipe_br_sp_top_registers[] = { + 0x0ae00, 0x0ae00, 0x0ae02, 0x0ae04, 0x0ae06, 0x0ae0a, 0x0ae0c, 0x0ae0c, + 0x0ae0f, 0x0ae0f, 0x0ae28, 0x0ae2b, 0x0ae35, 0x0ae35, 0x0ae3a, 0x0ae3f, + 0x0ae50, 0x0ae52, 0x0ae80, 0x0aea3, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_sp_pipe_br_sp_top_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_NONE + * Location: A7XX_USPTP + * pairs : 12 (Regs:62) + */ +static const u32 gen7_9_0_non_context_sp_pipe_br_usptp_registers[] = { + 0x0ae00, 0x0ae00, 0x0ae02, 0x0ae04, 0x0ae06, 0x0ae0a, 0x0ae0c, 0x0ae0c, + 0x0ae0f, 0x0ae0f, 0x0ae28, 0x0ae2b, 0x0ae30, 0x0ae32, 0x0ae35, 0x0ae35, + 0x0ae3a, 0x0ae3b, 0x0ae3e, 0x0ae3f, 0x0ae50, 0x0ae52, 0x0ae80, 0x0aea3, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_sp_pipe_br_usptp_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_NONE + * Location: A7XX_HLSQ_DP_STR + * pairs : 2 (Regs:5) + */ +static const u32 gen7_9_0_non_context_sp_pipe_br_hlsq_dp_str_registers[] = { + 0x0ae6b, 0x0ae6c, 0x0ae73, 0x0ae75, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_sp_pipe_br_hlsq_dp_str_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_LPAC + * Cluster : A7XX_CLUSTER_NONE + * Location: A7XX_HLSQ_STATE + * pairs : 1 (Regs:5) + */ +static const u32 gen7_9_0_non_context_sp_pipe_lpac_hlsq_state_registers[] = { + 0x0af88, 0x0af8c, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_sp_pipe_lpac_hlsq_state_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_LPAC + * Cluster : A7XX_CLUSTER_NONE + * Location: A7XX_SP_TOP + * pairs : 1 (Regs:6) + */ +static const u32 gen7_9_0_non_context_sp_pipe_lpac_sp_top_registers[] = { + 0x0af80, 0x0af85, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_sp_pipe_lpac_sp_top_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_LPAC + * Cluster : A7XX_CLUSTER_NONE + * Location: A7XX_USPTP + * pairs : 2 (Regs:9) + */ +static const u32 gen7_9_0_non_context_sp_pipe_lpac_usptp_registers[] = { + 0x0af80, 0x0af85, 0x0af90, 0x0af92, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_sp_pipe_lpac_usptp_registers), 8)); + +/* + * Block : ['TPL1'] + * Pipeline: PIPE_NONE + * Cluster : A7XX_CLUSTER_NONE + * Location: A7XX_USPTP + * pairs : 5 (Regs:29) + */ +static const u32 gen7_9_0_non_context_tpl1_pipe_none_usptp_registers[] = { + 0x0b602, 0x0b602, 0x0b604, 0x0b604, 0x0b608, 0x0b60c, 0x0b610, 0x0b621, + 0x0b630, 0x0b633, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_tpl1_pipe_none_usptp_registers), 8)); + +/* + * Block : ['TPL1'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_NONE + * Location: A7XX_USPTP + * pairs : 1 (Regs:1) + */ +static const u32 gen7_9_0_non_context_tpl1_pipe_br_usptp_registers[] = { + 0x0b600, 0x0b600, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_tpl1_pipe_br_usptp_registers), 8)); + +/* + * Block : ['TPL1'] + * Pipeline: PIPE_LPAC + * Cluster : A7XX_CLUSTER_NONE + * Location: A7XX_USPTP + * pairs : 1 (Regs:1) + */ +static const u32 gen7_9_0_non_context_tpl1_pipe_lpac_usptp_registers[] = { + 0x0b780, 0x0b780, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_non_context_tpl1_pipe_lpac_usptp_registers), 8)); + +/* + * Block : ['GRAS'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_GRAS + * pairs : 14 (Regs:293) + */ +static const u32 gen7_9_0_gras_pipe_br_cluster_gras_registers[] = { + 0x08000, 0x0800c, 0x08010, 0x08092, 0x08094, 0x08099, 0x0809b, 0x0809d, + 0x080a0, 0x080a7, 0x080af, 0x080f1, 0x080f4, 0x080f6, 0x080f8, 0x080fa, + 0x08100, 0x08107, 0x08109, 0x0810b, 0x08110, 0x08116, 0x08120, 0x0813f, + 0x08400, 0x08406, 0x0840a, 0x0840b, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_gras_pipe_br_cluster_gras_registers), 8)); + +/* + * Block : ['GRAS'] + * Pipeline: PIPE_BV + * Cluster : A7XX_CLUSTER_GRAS + * pairs : 14 (Regs:293) + */ +static const u32 gen7_9_0_gras_pipe_bv_cluster_gras_registers[] = { + 0x08000, 0x0800c, 0x08010, 0x08092, 0x08094, 0x08099, 0x0809b, 0x0809d, + 0x080a0, 0x080a7, 0x080af, 0x080f1, 0x080f4, 0x080f6, 0x080f8, 0x080fa, + 0x08100, 0x08107, 0x08109, 0x0810b, 0x08110, 0x08116, 0x08120, 0x0813f, + 0x08400, 0x08406, 0x0840a, 0x0840b, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_gras_pipe_bv_cluster_gras_registers), 8)); + +/* + * Block : ['PC'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_FE + * pairs : 6 (Regs:31) + */ +static const u32 gen7_9_0_pc_pipe_br_cluster_fe_registers[] = { + 0x09800, 0x09804, 0x09806, 0x0980a, 0x09810, 0x09811, 0x09884, 0x09886, + 0x09970, 0x09972, 0x09b00, 0x09b0c, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_pc_pipe_br_cluster_fe_registers), 8)); + +/* + * Block : ['PC'] + * Pipeline: PIPE_BV + * Cluster : A7XX_CLUSTER_FE + * pairs : 6 (Regs:31) + */ +static const u32 gen7_9_0_pc_pipe_bv_cluster_fe_registers[] = { + 0x09800, 0x09804, 0x09806, 0x0980a, 0x09810, 0x09811, 0x09884, 0x09886, + 0x09970, 0x09972, 0x09b00, 0x09b0c, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_pc_pipe_bv_cluster_fe_registers), 8)); + +/* + * Block : ['VFD'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_FE + * pairs : 2 (Regs:236) + */ +static const u32 gen7_9_0_vfd_pipe_br_cluster_fe_registers[] = { + 0x0a000, 0x0a009, 0x0a00e, 0x0a0ef, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_vfd_pipe_br_cluster_fe_registers), 8)); + +/* + * Block : ['VFD'] + * Pipeline: PIPE_BV + * Cluster : A7XX_CLUSTER_FE + * pairs : 2 (Regs:236) + */ +static const u32 gen7_9_0_vfd_pipe_bv_cluster_fe_registers[] = { + 0x0a000, 0x0a009, 0x0a00e, 0x0a0ef, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_vfd_pipe_bv_cluster_fe_registers), 8)); + +/* + * Block : ['VPC'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_FE + * pairs : 2 (Regs:18) + */ +static const u32 gen7_9_0_vpc_pipe_br_cluster_fe_registers[] = { + 0x09300, 0x0930a, 0x09311, 0x09317, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_vpc_pipe_br_cluster_fe_registers), 8)); + +/* + * Block : ['VPC'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_PC_VS + * pairs : 3 (Regs:30) + */ +static const u32 gen7_9_0_vpc_pipe_br_cluster_pc_vs_registers[] = { + 0x09101, 0x0910c, 0x09300, 0x0930a, 0x09311, 0x09317, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_vpc_pipe_br_cluster_pc_vs_registers), 8)); + +/* + * Block : ['VPC'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_VPC_PS + * pairs : 5 (Regs:76) + */ +static const u32 gen7_9_0_vpc_pipe_br_cluster_vpc_ps_registers[] = { + 0x09200, 0x0920f, 0x09212, 0x09216, 0x09218, 0x0923c, 0x09300, 0x0930a, + 0x09311, 0x09317, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_vpc_pipe_br_cluster_vpc_ps_registers), 8)); + +/* + * Block : ['VPC'] + * Pipeline: PIPE_BV + * Cluster : A7XX_CLUSTER_FE + * pairs : 2 (Regs:18) + */ +static const u32 gen7_9_0_vpc_pipe_bv_cluster_fe_registers[] = { + 0x09300, 0x0930a, 0x09311, 0x09317, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_vpc_pipe_bv_cluster_fe_registers), 8)); + +/* + * Block : ['VPC'] + * Pipeline: PIPE_BV + * Cluster : A7XX_CLUSTER_PC_VS + * pairs : 3 (Regs:30) + */ +static const u32 gen7_9_0_vpc_pipe_bv_cluster_pc_vs_registers[] = { + 0x09101, 0x0910c, 0x09300, 0x0930a, 0x09311, 0x09317, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_vpc_pipe_bv_cluster_pc_vs_registers), 8)); + +/* + * Block : ['VPC'] + * Pipeline: PIPE_BV + * Cluster : A7XX_CLUSTER_VPC_PS + * pairs : 5 (Regs:76) + */ +static const u32 gen7_9_0_vpc_pipe_bv_cluster_vpc_ps_registers[] = { + 0x09200, 0x0920f, 0x09212, 0x09216, 0x09218, 0x0923c, 0x09300, 0x0930a, + 0x09311, 0x09317, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_vpc_pipe_bv_cluster_vpc_ps_registers), 8)); + +/* + * Block : ['RB'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_PS + * pairs : 39 (Regs:133) + */ +static const u32 gen7_9_0_rb_pipe_br_cluster_ps_rac_registers[] = { + 0x08802, 0x08802, 0x08804, 0x08806, 0x08809, 0x0880a, 0x0880e, 0x08811, + 0x08818, 0x0881e, 0x08821, 0x08821, 0x08823, 0x08826, 0x08829, 0x08829, + 0x0882b, 0x0882e, 0x08831, 0x08831, 0x08833, 0x08836, 0x08839, 0x08839, + 0x0883b, 0x0883e, 0x08841, 0x08841, 0x08843, 0x08846, 0x08849, 0x08849, + 0x0884b, 0x0884e, 0x08851, 0x08851, 0x08853, 0x08856, 0x08859, 0x08859, + 0x0885b, 0x0885e, 0x08860, 0x08864, 0x08870, 0x08870, 0x08873, 0x08876, + 0x08878, 0x08879, 0x08882, 0x08885, 0x08887, 0x08889, 0x08891, 0x08891, + 0x08898, 0x08899, 0x088c0, 0x088c1, 0x088e5, 0x088e5, 0x088f4, 0x088f5, + 0x08a00, 0x08a05, 0x08a10, 0x08a15, 0x08a20, 0x08a25, 0x08a30, 0x08a35, + 0x08c00, 0x08c01, 0x08c18, 0x08c1f, 0x08c26, 0x08c34, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_rb_pipe_br_cluster_ps_rac_registers), 8)); + +/* + * Block : ['RB'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_PS + * pairs : 34 (Regs:100) + */ +static const u32 gen7_9_0_rb_pipe_br_cluster_ps_rbp_registers[] = { + 0x08800, 0x08801, 0x08803, 0x08803, 0x0880b, 0x0880d, 0x08812, 0x08812, + 0x08820, 0x08820, 0x08822, 0x08822, 0x08827, 0x08828, 0x0882a, 0x0882a, + 0x0882f, 0x08830, 0x08832, 0x08832, 0x08837, 0x08838, 0x0883a, 0x0883a, + 0x0883f, 0x08840, 0x08842, 0x08842, 0x08847, 0x08848, 0x0884a, 0x0884a, + 0x0884f, 0x08850, 0x08852, 0x08852, 0x08857, 0x08858, 0x0885a, 0x0885a, + 0x0885f, 0x0885f, 0x08865, 0x08865, 0x08871, 0x08872, 0x08877, 0x08877, + 0x08880, 0x08881, 0x08886, 0x08886, 0x08890, 0x08890, 0x088d0, 0x088e4, + 0x088e8, 0x088ea, 0x088f0, 0x088f0, 0x08900, 0x0891a, 0x08927, 0x08928, + 0x08c17, 0x08c17, 0x08c20, 0x08c25, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_rb_pipe_br_cluster_ps_rbp_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_SP_VS + * Location: A7XX_HLSQ_STATE + * pairs : 29 (Regs:215) + */ +static const u32 gen7_9_0_sp_pipe_br_cluster_sp_vs_hlsq_state_registers[] = { + 0x0a800, 0x0a801, 0x0a81b, 0x0a81d, 0x0a822, 0x0a822, 0x0a824, 0x0a824, + 0x0a827, 0x0a82a, 0x0a830, 0x0a830, 0x0a832, 0x0a835, 0x0a83a, 0x0a83a, + 0x0a83c, 0x0a83c, 0x0a83f, 0x0a841, 0x0a85b, 0x0a85d, 0x0a862, 0x0a862, + 0x0a864, 0x0a864, 0x0a867, 0x0a867, 0x0a870, 0x0a870, 0x0a872, 0x0a872, + 0x0a88c, 0x0a88e, 0x0a893, 0x0a893, 0x0a895, 0x0a895, 0x0a898, 0x0a898, + 0x0a89a, 0x0a89d, 0x0a8a0, 0x0a8af, 0x0a8c0, 0x0a8c3, 0x0a974, 0x0a977, + 0x0ab00, 0x0ab03, 0x0ab05, 0x0ab05, 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, + 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_br_cluster_sp_vs_hlsq_state_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_SP_VS + * Location: A7XX_SP_TOP + * pairs : 22 (Regs:73) + */ +static const u32 gen7_9_0_sp_pipe_br_cluster_sp_vs_sp_top_registers[] = { + 0x0a800, 0x0a800, 0x0a81c, 0x0a81d, 0x0a822, 0x0a824, 0x0a82d, 0x0a82d, + 0x0a82f, 0x0a831, 0x0a834, 0x0a835, 0x0a83a, 0x0a83c, 0x0a840, 0x0a840, + 0x0a85c, 0x0a85d, 0x0a862, 0x0a864, 0x0a868, 0x0a868, 0x0a870, 0x0a871, + 0x0a88d, 0x0a88e, 0x0a893, 0x0a895, 0x0a899, 0x0a899, 0x0a8a0, 0x0a8af, + 0x0a974, 0x0a977, 0x0ab00, 0x0ab00, 0x0ab02, 0x0ab02, 0x0ab04, 0x0ab05, + 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_br_cluster_sp_vs_sp_top_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_SP_VS + * Location: A7XX_USPTP + * pairs : 16 (Regs:269) + */ +static const u32 gen7_9_0_sp_pipe_br_cluster_sp_vs_usptp_registers[] = { + 0x0a800, 0x0a81b, 0x0a81e, 0x0a821, 0x0a823, 0x0a827, 0x0a82d, 0x0a82d, + 0x0a82f, 0x0a833, 0x0a836, 0x0a839, 0x0a83b, 0x0a85b, 0x0a85e, 0x0a861, + 0x0a863, 0x0a868, 0x0a870, 0x0a88c, 0x0a88f, 0x0a892, 0x0a894, 0x0a899, + 0x0a8c0, 0x0a8c3, 0x0ab00, 0x0ab05, 0x0ab21, 0x0ab22, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_br_cluster_sp_vs_usptp_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_SP_PS + * Location: A7XX_HLSQ_STATE + * pairs : 21 (Regs:334) + */ +static const u32 gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_state_registers[] = { + 0x0a980, 0x0a984, 0x0a99e, 0x0a99e, 0x0a9a7, 0x0a9a7, 0x0a9aa, 0x0a9aa, + 0x0a9ae, 0x0a9b0, 0x0a9b2, 0x0a9b5, 0x0a9ba, 0x0a9ba, 0x0a9bc, 0x0a9bc, + 0x0a9c4, 0x0a9c4, 0x0a9c6, 0x0a9c6, 0x0a9cd, 0x0a9cd, 0x0a9e0, 0x0a9fc, + 0x0aa00, 0x0aa00, 0x0aa30, 0x0aa31, 0x0aa40, 0x0aabf, 0x0aaf2, 0x0aaf3, + 0x0ab00, 0x0ab03, 0x0ab05, 0x0ab05, 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, + 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_state_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_SP_PS + * Location: A7XX_HLSQ_DP + * pairs : 3 (Regs:19) + */ +static const u32 gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_dp_registers[] = { + 0x0a9b1, 0x0a9b1, 0x0a9c6, 0x0a9cb, 0x0a9d4, 0x0a9df, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_dp_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_SP_PS + * Location: A7XX_SP_TOP + * pairs : 18 (Regs:77) + */ +static const u32 gen7_9_0_sp_pipe_br_cluster_sp_ps_sp_top_registers[] = { + 0x0a980, 0x0a980, 0x0a982, 0x0a984, 0x0a99e, 0x0a9a2, 0x0a9a7, 0x0a9a8, + 0x0a9aa, 0x0a9aa, 0x0a9ae, 0x0a9ae, 0x0a9b0, 0x0a9b1, 0x0a9b3, 0x0a9b5, + 0x0a9ba, 0x0a9bc, 0x0a9c5, 0x0a9c5, 0x0a9e0, 0x0a9f9, 0x0aa00, 0x0aa03, + 0x0aaf2, 0x0aaf3, 0x0ab00, 0x0ab00, 0x0ab02, 0x0ab02, 0x0ab04, 0x0ab05, + 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_br_cluster_sp_ps_sp_top_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_SP_PS + * Location: A7XX_USPTP + * pairs : 17 (Regs:333) + */ +static const u32 gen7_9_0_sp_pipe_br_cluster_sp_ps_usptp_registers[] = { + 0x0a980, 0x0a982, 0x0a985, 0x0a9a6, 0x0a9a8, 0x0a9a9, 0x0a9ab, 0x0a9ae, + 0x0a9b0, 0x0a9b3, 0x0a9b6, 0x0a9b9, 0x0a9bb, 0x0a9bf, 0x0a9c2, 0x0a9c3, + 0x0a9c5, 0x0a9c5, 0x0a9cd, 0x0a9cd, 0x0a9d0, 0x0a9d3, 0x0aa01, 0x0aa03, + 0x0aa30, 0x0aa31, 0x0aa40, 0x0aabf, 0x0ab00, 0x0ab05, 0x0ab21, 0x0ab22, + 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_br_cluster_sp_ps_usptp_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_SP_PS + * Location: A7XX_HLSQ_DP_STR + * pairs : 1 (Regs:6) + */ +static const u32 gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_dp_str_registers[] = { + 0x0a9c6, 0x0a9cb, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_dp_str_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BV + * Cluster : A7XX_CLUSTER_SP_VS + * Location: A7XX_HLSQ_STATE + * pairs : 28 (Regs:213) + */ +static const u32 gen7_9_0_sp_pipe_bv_cluster_sp_vs_hlsq_state_registers[] = { + 0x0a800, 0x0a801, 0x0a81b, 0x0a81d, 0x0a822, 0x0a822, 0x0a824, 0x0a824, + 0x0a827, 0x0a82a, 0x0a830, 0x0a830, 0x0a832, 0x0a835, 0x0a83a, 0x0a83a, + 0x0a83c, 0x0a83c, 0x0a83f, 0x0a841, 0x0a85b, 0x0a85d, 0x0a862, 0x0a862, + 0x0a864, 0x0a864, 0x0a867, 0x0a867, 0x0a870, 0x0a870, 0x0a872, 0x0a872, + 0x0a88c, 0x0a88e, 0x0a893, 0x0a893, 0x0a895, 0x0a895, 0x0a898, 0x0a898, + 0x0a89a, 0x0a89d, 0x0a8a0, 0x0a8af, 0x0a8c0, 0x0a8c3, 0x0a974, 0x0a977, + 0x0ab00, 0x0ab02, 0x0ab0a, 0x0ab1b, 0x0ab20, 0x0ab20, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_bv_cluster_sp_vs_hlsq_state_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BV + * Cluster : A7XX_CLUSTER_SP_VS + * Location: A7XX_SP_TOP + * pairs : 21 (Regs:71) + */ +static const u32 gen7_9_0_sp_pipe_bv_cluster_sp_vs_sp_top_registers[] = { + 0x0a800, 0x0a800, 0x0a81c, 0x0a81d, 0x0a822, 0x0a824, 0x0a82d, 0x0a82d, + 0x0a82f, 0x0a831, 0x0a834, 0x0a835, 0x0a83a, 0x0a83c, 0x0a840, 0x0a840, + 0x0a85c, 0x0a85d, 0x0a862, 0x0a864, 0x0a868, 0x0a868, 0x0a870, 0x0a871, + 0x0a88d, 0x0a88e, 0x0a893, 0x0a895, 0x0a899, 0x0a899, 0x0a8a0, 0x0a8af, + 0x0a974, 0x0a977, 0x0ab00, 0x0ab00, 0x0ab02, 0x0ab02, 0x0ab0a, 0x0ab1b, + 0x0ab20, 0x0ab20, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_bv_cluster_sp_vs_sp_top_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_BV + * Cluster : A7XX_CLUSTER_SP_VS + * Location: A7XX_USPTP + * pairs : 16 (Regs:266) + */ +static const u32 gen7_9_0_sp_pipe_bv_cluster_sp_vs_usptp_registers[] = { + 0x0a800, 0x0a81b, 0x0a81e, 0x0a821, 0x0a823, 0x0a827, 0x0a82d, 0x0a82d, + 0x0a82f, 0x0a833, 0x0a836, 0x0a839, 0x0a83b, 0x0a85b, 0x0a85e, 0x0a861, + 0x0a863, 0x0a868, 0x0a870, 0x0a88c, 0x0a88f, 0x0a892, 0x0a894, 0x0a899, + 0x0a8c0, 0x0a8c3, 0x0ab00, 0x0ab02, 0x0ab21, 0x0ab22, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_bv_cluster_sp_vs_usptp_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_LPAC + * Cluster : A7XX_CLUSTER_SP_PS + * Location: A7XX_HLSQ_STATE + * pairs : 14 (Regs:299) + */ +static const u32 gen7_9_0_sp_pipe_lpac_cluster_sp_ps_hlsq_state_registers[] = { + 0x0a9b0, 0x0a9b0, 0x0a9b2, 0x0a9b5, 0x0a9ba, 0x0a9ba, 0x0a9bc, 0x0a9bc, + 0x0a9c4, 0x0a9c4, 0x0a9cd, 0x0a9cd, 0x0a9e2, 0x0a9e3, 0x0a9e6, 0x0a9fc, + 0x0aa00, 0x0aa00, 0x0aa31, 0x0aa35, 0x0aa40, 0x0aabf, 0x0aaf3, 0x0aaf3, + 0x0ab00, 0x0ab01, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_lpac_cluster_sp_ps_hlsq_state_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_LPAC + * Cluster : A7XX_CLUSTER_SP_PS + * Location: A7XX_HLSQ_DP + * pairs : 2 (Regs:13) + */ +static const u32 gen7_9_0_sp_pipe_lpac_cluster_sp_ps_hlsq_dp_registers[] = { + 0x0a9b1, 0x0a9b1, 0x0a9d4, 0x0a9df, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_lpac_cluster_sp_ps_hlsq_dp_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_LPAC + * Cluster : A7XX_CLUSTER_SP_PS + * Location: A7XX_SP_TOP + * pairs : 9 (Regs:34) + */ +static const u32 gen7_9_0_sp_pipe_lpac_cluster_sp_ps_sp_top_registers[] = { + 0x0a9b0, 0x0a9b1, 0x0a9b3, 0x0a9b5, 0x0a9ba, 0x0a9bc, 0x0a9c5, 0x0a9c5, + 0x0a9e2, 0x0a9e3, 0x0a9e6, 0x0a9f9, 0x0aa00, 0x0aa00, 0x0aaf3, 0x0aaf3, + 0x0ab00, 0x0ab00, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_lpac_cluster_sp_ps_sp_top_registers), 8)); + +/* + * Block : ['SP'] + * Pipeline: PIPE_LPAC + * Cluster : A7XX_CLUSTER_SP_PS + * Location: A7XX_USPTP + * pairs : 11 (Regs:279) + */ +static const u32 gen7_9_0_sp_pipe_lpac_cluster_sp_ps_usptp_registers[] = { + 0x0a9b0, 0x0a9b3, 0x0a9b6, 0x0a9b9, 0x0a9bb, 0x0a9be, 0x0a9c2, 0x0a9c3, + 0x0a9c5, 0x0a9c5, 0x0a9cd, 0x0a9cd, 0x0a9d0, 0x0a9d3, 0x0aa31, 0x0aa31, + 0x0aa40, 0x0aabf, 0x0ab00, 0x0ab01, 0x0ab40, 0x0abbf, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_sp_pipe_lpac_cluster_sp_ps_usptp_registers), 8)); + +/* + * Block : ['TPL1'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_SP_VS + * Location: A7XX_USPTP + * pairs : 3 (Regs:10) + */ +static const u32 gen7_9_0_tpl1_pipe_br_cluster_sp_vs_usptp_registers[] = { + 0x0b300, 0x0b307, 0x0b309, 0x0b309, 0x0b310, 0x0b310, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_tpl1_pipe_br_cluster_sp_vs_usptp_registers), 8)); + +/* + * Block : ['TPL1'] + * Pipeline: PIPE_BR + * Cluster : A7XX_CLUSTER_SP_PS + * Location: A7XX_USPTP + * pairs : 6 (Regs:42) + */ +static const u32 gen7_9_0_tpl1_pipe_br_cluster_sp_ps_usptp_registers[] = { + 0x0b180, 0x0b183, 0x0b190, 0x0b195, 0x0b2c0, 0x0b2d5, 0x0b300, 0x0b307, + 0x0b309, 0x0b309, 0x0b310, 0x0b310, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_tpl1_pipe_br_cluster_sp_ps_usptp_registers), 8)); + +/* + * Block : ['TPL1'] + * Pipeline: PIPE_BV + * Cluster : A7XX_CLUSTER_SP_VS + * Location: A7XX_USPTP + * pairs : 3 (Regs:10) + */ +static const u32 gen7_9_0_tpl1_pipe_bv_cluster_sp_vs_usptp_registers[] = { + 0x0b300, 0x0b307, 0x0b309, 0x0b309, 0x0b310, 0x0b310, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_tpl1_pipe_bv_cluster_sp_vs_usptp_registers), 8)); + +/* + * Block : ['TPL1'] + * Pipeline: PIPE_LPAC + * Cluster : A7XX_CLUSTER_SP_PS + * Location: A7XX_USPTP + * pairs : 5 (Regs:7) + */ +static const u32 gen7_9_0_tpl1_pipe_lpac_cluster_sp_ps_usptp_registers[] = { + 0x0b180, 0x0b181, 0x0b300, 0x0b301, 0x0b307, 0x0b307, 0x0b309, 0x0b309, + 0x0b310, 0x0b310, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_tpl1_pipe_lpac_cluster_sp_ps_usptp_registers), 8)); + +static const struct gen7_sel_reg gen7_9_0_rb_rac_sel = { + .host_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_HOST, + .cd_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_CD, + .val = 0, +}; + +static const struct gen7_sel_reg gen7_9_0_rb_rbp_sel = { + .host_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_HOST, + .cd_reg = REG_A6XX_RB_SUB_BLOCK_SEL_CNTL_CD, + .val = 0x9, +}; + +static const struct gen7_cluster_registers gen7_9_0_clusters[] = { + { A7XX_CLUSTER_NONE, PIPE_BR, STATE_NON_CONTEXT, + gen7_9_0_non_context_pipe_br_registers, }, + { A7XX_CLUSTER_NONE, PIPE_BV, STATE_NON_CONTEXT, + gen7_9_0_non_context_pipe_bv_registers, }, + { A7XX_CLUSTER_NONE, PIPE_LPAC, STATE_NON_CONTEXT, + gen7_9_0_non_context_pipe_lpac_registers, }, + { A7XX_CLUSTER_NONE, PIPE_BR, STATE_NON_CONTEXT, + gen7_9_0_non_context_rb_pipe_br_rac_registers, &gen7_9_0_rb_rac_sel, }, + { A7XX_CLUSTER_NONE, PIPE_BR, STATE_NON_CONTEXT, + gen7_9_0_non_context_rb_pipe_br_rbp_registers, &gen7_9_0_rb_rbp_sel, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_9_0_rb_pipe_br_cluster_ps_rac_registers, &gen7_9_0_rb_rac_sel, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_9_0_rb_pipe_br_cluster_ps_rac_registers, &gen7_9_0_rb_rac_sel, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_9_0_rb_pipe_br_cluster_ps_rbp_registers, &gen7_9_0_rb_rbp_sel, }, + { A7XX_CLUSTER_PS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_9_0_rb_pipe_br_cluster_ps_rbp_registers, &gen7_9_0_rb_rbp_sel, }, + { A7XX_CLUSTER_GRAS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_9_0_gras_pipe_br_cluster_gras_registers, }, + { A7XX_CLUSTER_GRAS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_9_0_gras_pipe_br_cluster_gras_registers, }, + { A7XX_CLUSTER_GRAS, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_9_0_gras_pipe_bv_cluster_gras_registers, }, + { A7XX_CLUSTER_GRAS, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_9_0_gras_pipe_bv_cluster_gras_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_9_0_pc_pipe_br_cluster_fe_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_9_0_pc_pipe_br_cluster_fe_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_9_0_pc_pipe_bv_cluster_fe_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_9_0_pc_pipe_bv_cluster_fe_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_9_0_vfd_pipe_br_cluster_fe_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_9_0_vfd_pipe_br_cluster_fe_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_9_0_vfd_pipe_bv_cluster_fe_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_9_0_vfd_pipe_bv_cluster_fe_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_9_0_vpc_pipe_br_cluster_fe_registers, }, + { A7XX_CLUSTER_FE, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_9_0_vpc_pipe_br_cluster_fe_registers, }, + { A7XX_CLUSTER_PC_VS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_9_0_vpc_pipe_br_cluster_pc_vs_registers, }, + { A7XX_CLUSTER_PC_VS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_9_0_vpc_pipe_br_cluster_pc_vs_registers, }, + { A7XX_CLUSTER_VPC_PS, PIPE_BR, STATE_FORCE_CTXT_0, + gen7_9_0_vpc_pipe_br_cluster_vpc_ps_registers, }, + { A7XX_CLUSTER_VPC_PS, PIPE_BR, STATE_FORCE_CTXT_1, + gen7_9_0_vpc_pipe_br_cluster_vpc_ps_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_9_0_vpc_pipe_bv_cluster_fe_registers, }, + { A7XX_CLUSTER_FE, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_9_0_vpc_pipe_bv_cluster_fe_registers, }, + { A7XX_CLUSTER_PC_VS, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_9_0_vpc_pipe_bv_cluster_pc_vs_registers, }, + { A7XX_CLUSTER_PC_VS, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_9_0_vpc_pipe_bv_cluster_pc_vs_registers, }, + { A7XX_CLUSTER_VPC_PS, PIPE_BV, STATE_FORCE_CTXT_0, + gen7_9_0_vpc_pipe_bv_cluster_vpc_ps_registers, }, + { A7XX_CLUSTER_VPC_PS, PIPE_BV, STATE_FORCE_CTXT_1, + gen7_9_0_vpc_pipe_bv_cluster_vpc_ps_registers, }, +}; + +static const struct gen7_sptp_cluster_registers gen7_9_0_sptp_clusters[] = { + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_BR, 0, A7XX_HLSQ_STATE, + gen7_9_0_non_context_sp_pipe_br_hlsq_state_registers, 0xae00}, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_BR, 0, A7XX_SP_TOP, + gen7_9_0_non_context_sp_pipe_br_sp_top_registers, 0xae00}, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_9_0_non_context_sp_pipe_br_usptp_registers, 0xae00}, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_BR, 0, A7XX_HLSQ_DP_STR, + gen7_9_0_non_context_sp_pipe_br_hlsq_dp_str_registers, 0xae00}, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_LPAC, 0, A7XX_HLSQ_STATE, + gen7_9_0_non_context_sp_pipe_lpac_hlsq_state_registers, 0xaf80}, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_LPAC, 0, A7XX_SP_TOP, + gen7_9_0_non_context_sp_pipe_lpac_sp_top_registers, 0xaf80}, + { A7XX_CLUSTER_NONE, A7XX_SP_NCTX_REG, PIPE_LPAC, 0, A7XX_USPTP, + gen7_9_0_non_context_sp_pipe_lpac_usptp_registers, 0xaf80}, + { A7XX_CLUSTER_NONE, A7XX_TP0_NCTX_REG, PIPE_NONE, 0, A7XX_USPTP, + gen7_9_0_non_context_tpl1_pipe_none_usptp_registers, 0xb600}, + { A7XX_CLUSTER_NONE, A7XX_TP0_NCTX_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_9_0_non_context_tpl1_pipe_br_usptp_registers, 0xb600}, + { A7XX_CLUSTER_NONE, A7XX_TP0_NCTX_REG, PIPE_LPAC, 0, A7XX_USPTP, + gen7_9_0_non_context_tpl1_pipe_lpac_usptp_registers, 0xb780}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_HLSQ_STATE, + gen7_9_0_sp_pipe_br_cluster_sp_vs_hlsq_state_registers, 0xa800}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_SP_TOP, + gen7_9_0_sp_pipe_br_cluster_sp_vs_sp_top_registers, 0xa800}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_9_0_sp_pipe_br_cluster_sp_vs_usptp_registers, 0xa800}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BV, 0, A7XX_HLSQ_STATE, + gen7_9_0_sp_pipe_bv_cluster_sp_vs_hlsq_state_registers, 0xa800}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BV, 0, A7XX_SP_TOP, + gen7_9_0_sp_pipe_bv_cluster_sp_vs_sp_top_registers, 0xa800}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BV, 0, A7XX_USPTP, + gen7_9_0_sp_pipe_bv_cluster_sp_vs_usptp_registers, 0xa800}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_HLSQ_STATE, + gen7_9_0_sp_pipe_br_cluster_sp_vs_hlsq_state_registers, 0xa800}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_SP_TOP, + gen7_9_0_sp_pipe_br_cluster_sp_vs_sp_top_registers, 0xa800}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_9_0_sp_pipe_br_cluster_sp_vs_usptp_registers, 0xa800}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BV, 1, A7XX_HLSQ_STATE, + gen7_9_0_sp_pipe_bv_cluster_sp_vs_hlsq_state_registers, 0xa800}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BV, 1, A7XX_SP_TOP, + gen7_9_0_sp_pipe_bv_cluster_sp_vs_sp_top_registers, 0xa800}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BV, 1, A7XX_USPTP, + gen7_9_0_sp_pipe_bv_cluster_sp_vs_usptp_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_HLSQ_STATE, + gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_state_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_HLSQ_DP, + gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_dp_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_SP_TOP, + gen7_9_0_sp_pipe_br_cluster_sp_ps_sp_top_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_9_0_sp_pipe_br_cluster_sp_ps_usptp_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_HLSQ_DP_STR, + gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_dp_str_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_HLSQ_STATE, + gen7_9_0_sp_pipe_lpac_cluster_sp_ps_hlsq_state_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_HLSQ_DP, + gen7_9_0_sp_pipe_lpac_cluster_sp_ps_hlsq_dp_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_SP_TOP, + gen7_9_0_sp_pipe_lpac_cluster_sp_ps_sp_top_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_USPTP, + gen7_9_0_sp_pipe_lpac_cluster_sp_ps_usptp_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_HLSQ_STATE, + gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_state_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_HLSQ_DP, + gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_dp_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_SP_TOP, + gen7_9_0_sp_pipe_br_cluster_sp_ps_sp_top_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_9_0_sp_pipe_br_cluster_sp_ps_usptp_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_HLSQ_DP_STR, + gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_dp_str_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_HLSQ_DP, + gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_dp_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_SP_TOP, + gen7_9_0_sp_pipe_br_cluster_sp_ps_sp_top_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_USPTP, + gen7_9_0_sp_pipe_br_cluster_sp_ps_usptp_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_HLSQ_DP_STR, + gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_dp_str_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_HLSQ_DP, + gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_dp_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_SP_TOP, + gen7_9_0_sp_pipe_br_cluster_sp_ps_sp_top_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_USPTP, + gen7_9_0_sp_pipe_br_cluster_sp_ps_usptp_registers, 0xa800}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_HLSQ_DP_STR, + gen7_9_0_sp_pipe_br_cluster_sp_ps_hlsq_dp_str_registers, 0xa800}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_9_0_tpl1_pipe_br_cluster_sp_vs_usptp_registers, 0xb000}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX0_3D_CVS_REG, PIPE_BV, 0, A7XX_USPTP, + gen7_9_0_tpl1_pipe_bv_cluster_sp_vs_usptp_registers, 0xb000}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_9_0_tpl1_pipe_br_cluster_sp_vs_usptp_registers, 0xb000}, + { A7XX_CLUSTER_SP_VS, A7XX_SP_CTX1_3D_CVS_REG, PIPE_BV, 1, A7XX_USPTP, + gen7_9_0_tpl1_pipe_bv_cluster_sp_vs_usptp_registers, 0xb000}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_BR, 0, A7XX_USPTP, + gen7_9_0_tpl1_pipe_br_cluster_sp_ps_usptp_registers, 0xb000}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX0_3D_CPS_REG, PIPE_LPAC, 0, A7XX_USPTP, + gen7_9_0_tpl1_pipe_lpac_cluster_sp_ps_usptp_registers, 0xb000}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX1_3D_CPS_REG, PIPE_BR, 1, A7XX_USPTP, + gen7_9_0_tpl1_pipe_br_cluster_sp_ps_usptp_registers, 0xb000}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX2_3D_CPS_REG, PIPE_BR, 2, A7XX_USPTP, + gen7_9_0_tpl1_pipe_br_cluster_sp_ps_usptp_registers, 0xb000}, + { A7XX_CLUSTER_SP_PS, A7XX_SP_CTX3_3D_CPS_REG, PIPE_BR, 3, A7XX_USPTP, + gen7_9_0_tpl1_pipe_br_cluster_sp_ps_usptp_registers, 0xb000}, +}; + +static const struct a6xx_indexed_registers gen7_9_0_cp_indexed_reg_list[] = { + { "CP_SQE_STAT", REG_A6XX_CP_SQE_STAT_ADDR, + REG_A6XX_CP_SQE_STAT_DATA, 0x00040}, + { "CP_DRAW_STATE", REG_A6XX_CP_DRAW_STATE_ADDR, + REG_A6XX_CP_DRAW_STATE_DATA, 0x00200}, + { "CP_ROQ_DBG", REG_A6XX_CP_ROQ_DBG_ADDR, + REG_A6XX_CP_ROQ_DBG_DATA, 0x00800}, + { "CP_SQE_UCODE_DBG", REG_A6XX_CP_SQE_UCODE_DBG_ADDR, + REG_A6XX_CP_SQE_UCODE_DBG_DATA, 0x08000}, + { "CP_BV_DRAW_STATE", REG_A7XX_CP_BV_DRAW_STATE_ADDR, + REG_A7XX_CP_BV_DRAW_STATE_DATA, 0x00200}, + { "CP_BV_ROQ_DBG", REG_A7XX_CP_BV_ROQ_DBG_ADDR, + REG_A7XX_CP_BV_ROQ_DBG_DATA, 0x00800}, + { "CP_BV_SQE_UCODE_DBG", REG_A7XX_CP_BV_SQE_UCODE_DBG_ADDR, + REG_A7XX_CP_BV_SQE_UCODE_DBG_DATA, 0x08000}, + { "CP_BV_SQE_STAT", REG_A7XX_CP_BV_SQE_STAT_ADDR, + REG_A7XX_CP_BV_SQE_STAT_DATA, 0x00040}, + { "CP_RESOURCE_TABLE_DBG", REG_A7XX_CP_RESOURCE_TABLE_DBG_ADDR, + REG_A7XX_CP_RESOURCE_TABLE_DBG_DATA, 0x04100}, + { "CP_LPAC_DRAW_STATE", REG_A7XX_CP_LPAC_DRAW_STATE_ADDR, + REG_A7XX_CP_LPAC_DRAW_STATE_DATA, 0x00200}, + { "CP_LPAC_ROQ_DBG", REG_A7XX_CP_LPAC_ROQ_DBG_ADDR, + REG_A7XX_CP_LPAC_ROQ_DBG_DATA, 0x00200}, + { "CP_SQE_AC_UCODE_DBG", REG_A7XX_CP_SQE_AC_UCODE_DBG_ADDR, + REG_A7XX_CP_SQE_AC_UCODE_DBG_DATA, 0x08000}, + { "CP_SQE_AC_STAT", REG_A7XX_CP_SQE_AC_STAT_ADDR, + REG_A7XX_CP_SQE_AC_STAT_DATA, 0x00040}, + { "CP_LPAC_FIFO_DBG", REG_A7XX_CP_LPAC_FIFO_DBG_ADDR, + REG_A7XX_CP_LPAC_FIFO_DBG_DATA, 0x00040}, + { "CP_AQE_ROQ_0", REG_A7XX_CP_AQE_ROQ_DBG_ADDR_0, + REG_A7XX_CP_AQE_ROQ_DBG_DATA_0, 0x00100}, + { "CP_AQE_ROQ_1", REG_A7XX_CP_AQE_ROQ_DBG_ADDR_1, + REG_A7XX_CP_AQE_ROQ_DBG_DATA_1, 0x00100}, + { "CP_AQE_UCODE_DBG_0", REG_A7XX_CP_AQE_UCODE_DBG_ADDR_0, + REG_A7XX_CP_AQE_UCODE_DBG_DATA_0, 0x08000}, + { "CP_AQE_UCODE_DBG_1", REG_A7XX_CP_AQE_UCODE_DBG_ADDR_1, + REG_A7XX_CP_AQE_UCODE_DBG_DATA_1, 0x08000}, + { "CP_AQE_STAT_0", REG_A7XX_CP_AQE_STAT_ADDR_0, + REG_A7XX_CP_AQE_STAT_DATA_0, 0x00040}, + { "CP_AQE_STAT_1", REG_A7XX_CP_AQE_STAT_ADDR_1, + REG_A7XX_CP_AQE_STAT_DATA_1, 0x00040}, +}; + +static const struct gen7_reg_list gen7_9_0_reg_list[] = { + { gen7_9_0_gpu_registers, NULL}, + { gen7_9_0_cx_misc_registers, NULL}, + { gen7_9_0_cx_dbgc_registers, NULL}, + { gen7_9_0_dbgc_registers, NULL}, + { NULL, NULL}, +}; + +static const u32 gen7_9_0_cpr_registers[] = { + 0x26800, 0x26805, 0x26808, 0x2680d, 0x26814, 0x26815, 0x2681c, 0x2681c, + 0x26820, 0x26839, 0x26840, 0x26841, 0x26848, 0x26849, 0x26850, 0x26851, + 0x26880, 0x268a1, 0x26980, 0x269b0, 0x269c0, 0x269c8, 0x269e0, 0x269ee, + 0x269fb, 0x269ff, 0x26a02, 0x26a07, 0x26a09, 0x26a0b, 0x26a10, 0x26b0f, + 0x27440, 0x27441, 0x27444, 0x27444, 0x27480, 0x274a2, 0x274ac, 0x274c4, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_cpr_registers), 8)); + +static const u32 gen7_9_0_dpm_registers[] = { + 0x1aa00, 0x1aa06, 0x1aa09, 0x1aa0a, 0x1aa0c, 0x1aa0d, 0x1aa0f, 0x1aa12, + 0x1aa14, 0x1aa47, 0x1aa50, 0x1aa51, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_dpm_registers), 8)); + +static const u32 gen7_9_0_dpm_leakage_registers[] = { + 0x21c00, 0x21c00, 0x21c08, 0x21c09, 0x21c0e, 0x21c0f, 0x21c4f, 0x21c50, + 0x21c52, 0x21c52, 0x21c54, 0x21c56, 0x21c58, 0x21c5a, 0x21c5c, 0x21c60, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_dpm_leakage_registers), 8)); + +static const u32 gen7_9_0_gfx_gpu_acd_registers[] = { + 0x18c00, 0x18c16, 0x18c20, 0x18c2d, 0x18c30, 0x18c31, 0x18c35, 0x18c35, + 0x18c37, 0x18c37, 0x18c3a, 0x18c3a, 0x18c42, 0x18c42, 0x18c56, 0x18c58, + 0x18c5b, 0x18c5d, 0x18c5f, 0x18c62, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_gfx_gpu_acd_registers), 8)); + +static const u32 gen7_9_0_gpucc_registers[] = { + 0x24000, 0x2400f, 0x24400, 0x2440f, 0x24800, 0x24805, 0x24c00, 0x24cff, + 0x25400, 0x25404, 0x25800, 0x25804, 0x25c00, 0x25c04, 0x26000, 0x26004, + 0x26400, 0x26405, 0x26414, 0x2641d, 0x2642a, 0x26430, 0x26432, 0x26434, + 0x26441, 0x2644b, 0x2644d, 0x26463, 0x26466, 0x26468, 0x26478, 0x2647a, + 0x26489, 0x2648a, 0x2649c, 0x2649e, 0x264a0, 0x264a6, 0x264c5, 0x264c7, + 0x264d6, 0x264d8, 0x264e8, 0x264e9, 0x264f9, 0x264fc, 0x2650b, 0x2650b, + 0x2651c, 0x2651e, 0x26540, 0x2654e, 0x26554, 0x26573, 0x26576, 0x2657a, + UINT_MAX, UINT_MAX, + +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_gpucc_registers), 8)); + +static const u32 gen7_9_0_isense_registers[] = { + 0x22c3a, 0x22c3c, 0x22c41, 0x22c41, 0x22c46, 0x22c47, 0x22c4c, 0x22c4c, + 0x22c51, 0x22c51, 0x22c56, 0x22c56, 0x22c5b, 0x22c5b, 0x22c60, 0x22c60, + 0x22c65, 0x22c65, 0x22c6a, 0x22c70, 0x22c75, 0x22c75, 0x22c7a, 0x22c7a, + 0x22c7f, 0x22c7f, 0x22c84, 0x22c85, 0x22c8a, 0x22c8a, 0x22c8f, 0x22c8f, + 0x23000, 0x23009, 0x2300e, 0x2300e, 0x23013, 0x23013, 0x23018, 0x23018, + 0x2301d, 0x2301d, 0x23022, 0x23022, 0x23027, 0x23032, 0x23037, 0x23037, + 0x2303c, 0x2303c, 0x23041, 0x23041, 0x23046, 0x23046, 0x2304b, 0x2304b, + 0x23050, 0x23050, 0x23055, 0x23055, 0x2305a, 0x2305a, 0x2305f, 0x2305f, + 0x23064, 0x23064, 0x23069, 0x2306a, 0x2306f, 0x2306f, 0x23074, 0x23075, + 0x2307a, 0x2307e, 0x23083, 0x23083, 0x23088, 0x23088, 0x2308d, 0x2308d, + 0x23092, 0x23092, 0x230e2, 0x230e2, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_isense_registers), 8)); + +static const u32 gen7_9_0_rscc_registers[] = { + 0x14000, 0x14036, 0x14040, 0x14047, 0x14080, 0x14084, 0x14089, 0x1408c, + 0x14091, 0x14094, 0x14099, 0x1409c, 0x140a1, 0x140a4, 0x140a9, 0x140ac, + 0x14100, 0x14104, 0x14114, 0x14119, 0x14124, 0x14132, 0x14154, 0x1416b, + 0x14340, 0x14342, 0x14344, 0x1437c, 0x143f0, 0x143f8, 0x143fa, 0x143fe, + 0x14400, 0x14404, 0x14406, 0x1440a, 0x1440c, 0x14410, 0x14412, 0x14416, + 0x14418, 0x1441c, 0x1441e, 0x14422, 0x14424, 0x14424, 0x14498, 0x144a0, + 0x144a2, 0x144a6, 0x144a8, 0x144ac, 0x144ae, 0x144b2, 0x144b4, 0x144b8, + 0x144ba, 0x144be, 0x144c0, 0x144c4, 0x144c6, 0x144ca, 0x144cc, 0x144cc, + 0x14540, 0x14548, 0x1454a, 0x1454e, 0x14550, 0x14554, 0x14556, 0x1455a, + 0x1455c, 0x14560, 0x14562, 0x14566, 0x14568, 0x1456c, 0x1456e, 0x14572, + 0x14574, 0x14574, 0x145e8, 0x145f0, 0x145f2, 0x145f6, 0x145f8, 0x145fc, + 0x145fe, 0x14602, 0x14604, 0x14608, 0x1460a, 0x1460e, 0x14610, 0x14614, + 0x14616, 0x1461a, 0x1461c, 0x1461c, 0x14690, 0x14698, 0x1469a, 0x1469e, + 0x146a0, 0x146a4, 0x146a6, 0x146aa, 0x146ac, 0x146b0, 0x146b2, 0x146b6, + 0x146b8, 0x146bc, 0x146be, 0x146c2, 0x146c4, 0x146c4, 0x14738, 0x14740, + 0x14742, 0x14746, 0x14748, 0x1474c, 0x1474e, 0x14752, 0x14754, 0x14758, + 0x1475a, 0x1475e, 0x14760, 0x14764, 0x14766, 0x1476a, 0x1476c, 0x1476c, + 0x147e0, 0x147e8, 0x147ea, 0x147ee, 0x147f0, 0x147f4, 0x147f6, 0x147fa, + 0x147fc, 0x14800, 0x14802, 0x14806, 0x14808, 0x1480c, 0x1480e, 0x14812, + 0x14814, 0x14814, 0x14888, 0x14890, 0x14892, 0x14896, 0x14898, 0x1489c, + 0x1489e, 0x148a2, 0x148a4, 0x148a8, 0x148aa, 0x148ae, 0x148b0, 0x148b4, + 0x148b6, 0x148ba, 0x148bc, 0x148bc, 0x14930, 0x14938, 0x1493a, 0x1493e, + 0x14940, 0x14944, 0x14946, 0x1494a, 0x1494c, 0x14950, 0x14952, 0x14956, + 0x14958, 0x1495c, 0x1495e, 0x14962, 0x14964, 0x14964, + UINT_MAX, UINT_MAX, +}; +static_assert(IS_ALIGNED(sizeof(gen7_9_0_rscc_registers), 8)); + +static const u32 *gen7_9_0_external_core_regs[] = { + gen7_9_0_gpucc_registers, + gen7_9_0_gxclkctl_registers, + gen7_9_0_cpr_registers, + gen7_9_0_dpm_registers, + gen7_9_0_dpm_leakage_registers, + gen7_9_0_gfx_gpu_acd_registers, +}; +#endif /*_ADRENO_GEN7_9_0_SNAPSHOT_H */ diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c new file mode 100644 index 000000000..5832dc25d --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -0,0 +1,1270 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + * + * Copyright (c) 2014 The Linux Foundation. All rights reserved. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "adreno_gpu.h" +#include "a6xx_gpu.h" +#include "msm_gem.h" +#include "msm_mmu.h" + +static u64 address_space_size = 0; +MODULE_PARM_DESC(address_space_size, "Override for size of processes private GPU address space"); +module_param(address_space_size, ullong, 0600); + +static bool zap_available = true; + +static int zap_shader_load_mdt(struct msm_gpu *gpu, const char *fwname, + u32 pasid) +{ + struct device *dev = &gpu->pdev->dev; + const struct firmware *fw; + const char *signed_fwname = NULL; + struct device_node *np; + struct resource r; + phys_addr_t mem_phys; + ssize_t mem_size; + void *mem_region = NULL; + int ret; + + if (!IS_ENABLED(CONFIG_ARCH_QCOM)) { + zap_available = false; + return -EINVAL; + } + + np = of_get_available_child_by_name(dev->of_node, "zap-shader"); + if (!np) { + zap_available = false; + return -ENODEV; + } + + /* We need PAS to be able to load the firmware */ + if (!qcom_pas_is_available()) { + DRM_DEV_ERROR(dev, "PAS is not available\n"); + return -EPROBE_DEFER; + } + + ret = of_reserved_mem_region_to_resource(np, 0, &r); + if (ret) { + zap_available = false; + return ret; + } + mem_phys = r.start; + + /* + * Check for a firmware-name property. This is the new scheme + * to handle firmware that may be signed with device specific + * keys, allowing us to have a different zap fw path for different + * devices. + * + * If the firmware-name property is found, we bypass the + * adreno_request_fw() mechanism, because we don't need to handle + * the /lib/firmware/qcom/... vs /lib/firmware/... case. + * + * If the firmware-name property is not found, for backwards + * compatibility we fall back to the fwname from the gpulist + * table. + */ + of_property_read_string_index(np, "firmware-name", 0, &signed_fwname); + if (signed_fwname) { + fwname = signed_fwname; + ret = request_firmware_direct(&fw, fwname, gpu->dev->dev); + if (ret) + fw = ERR_PTR(ret); + } else if (fwname) { + /* Request the MDT file from the default location: */ + fw = adreno_request_fw(to_adreno_gpu(gpu), fwname); + } else { + /* + * For new targets, we require the firmware-name property, + * if a zap-shader is required, rather than falling back + * to a firmware name specified in gpulist. + * + * Because the firmware is signed with a (potentially) + * device specific key, having the name come from gpulist + * was a bad idea, and is only provided for backwards + * compatibility for older targets. + */ + return -ENOENT; + } + + if (IS_ERR(fw)) { + DRM_DEV_ERROR(dev, "Unable to load %s\n", fwname); + return PTR_ERR(fw); + } + + /* Figure out how much memory we need */ + mem_size = qcom_mdt_get_size(fw); + if (mem_size < 0) { + ret = mem_size; + goto out; + } + + if (mem_size > resource_size(&r)) { + DRM_DEV_ERROR(dev, + "memory region is too small to load the MDT\n"); + ret = -E2BIG; + goto out; + } + + /* Allocate memory for the firmware image */ + mem_region = memremap(mem_phys, mem_size, MEMREMAP_WC); + if (!mem_region) { + ret = -ENOMEM; + goto out; + } + + /* + * Load the rest of the MDT + * + * Note that we could be dealing with two different paths, since + * with upstream linux-firmware it would be in a qcom/ subdir.. + * adreno_request_fw() handles this, but qcom_mdt_load() does + * not. But since we've already gotten through adreno_request_fw() + * we know which of the two cases it is: + */ + if (signed_fwname || (to_adreno_gpu(gpu)->fwloc == FW_LOCATION_LEGACY)) { + ret = qcom_mdt_load(dev, fw, fwname, pasid, + mem_region, mem_phys, mem_size, NULL); + } else { + char *newname; + + newname = kasprintf(GFP_KERNEL, "qcom/%s", fwname); + + ret = qcom_mdt_load(dev, fw, newname, pasid, + mem_region, mem_phys, mem_size, NULL); + kfree(newname); + } + if (ret) + goto out; + + /* Send the image to the secure world */ + ret = qcom_pas_auth_and_reset(pasid); + + /* + * If the pas call returns -EOPNOTSUPP we assume that this target + * doesn't need/support the zap shader so quietly fail + */ + if (ret == -EOPNOTSUPP) + zap_available = false; + else if (ret) + DRM_DEV_ERROR(dev, "Unable to authorize the image\n"); + +out: + if (mem_region) + memunmap(mem_region); + + release_firmware(fw); + + return ret; +} + +int adreno_zap_shader_load(struct msm_gpu *gpu, u32 pasid) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + + /* Short cut if we determine the zap shader isn't available/needed */ + if (!zap_available) + return -ENODEV; + + return zap_shader_load_mdt(gpu, adreno_gpu->info->zapfw, pasid); +} + +struct drm_gpuvm * +adreno_create_vm(struct msm_gpu *gpu, + struct platform_device *pdev) +{ + return adreno_iommu_create_vm(gpu, pdev, 0); +} + +struct drm_gpuvm * +adreno_iommu_create_vm(struct msm_gpu *gpu, + struct platform_device *pdev, + unsigned long quirks) +{ + struct iommu_domain_geometry *geometry; + struct msm_mmu *mmu; + struct drm_gpuvm *vm; + u64 start, size; + + mmu = msm_iommu_gpu_new(&pdev->dev, gpu, quirks); + if (IS_ERR(mmu)) + return ERR_CAST(mmu); + + geometry = msm_iommu_get_geometry(mmu); + if (IS_ERR(geometry)) + return ERR_CAST(geometry); + + /* + * Use the aperture start or SZ_16M, whichever is greater. This will + * ensure that we align with the allocated pagetable range while still + * allowing room in the lower 32 bits for GMEM and whatnot + */ + start = max_t(u64, SZ_16M, geometry->aperture_start); + size = geometry->aperture_end - start + 1; + + vm = msm_gem_vm_create(gpu->dev, mmu, "gpu", start & GENMASK_ULL(48, 0), + size, true); + + if (IS_ERR(vm) && !IS_ERR(mmu)) + mmu->funcs->destroy(mmu); + + return vm; +} + +u64 adreno_private_vm_size(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct adreno_smmu_priv *adreno_smmu = dev_get_drvdata(&gpu->pdev->dev); + const struct io_pgtable_cfg *ttbr1_cfg; + + if (address_space_size) + return address_space_size; + + if (adreno_gpu->info->quirks & ADRENO_QUIRK_4GB_VA) + return SZ_4G; + + if (!adreno_smmu || !adreno_smmu->get_ttbr1_cfg) + return SZ_4G; + + ttbr1_cfg = adreno_smmu->get_ttbr1_cfg(adreno_smmu->cookie); + + /* + * Userspace VM is actually using TTBR0, but both are the same size, + * with b48 (sign bit) selecting which TTBRn to use. So if IAS is + * 48, the total (kernel+user) address space size is effectively + * 49 bits. But what userspace is control of is the lower 48. + */ + return BIT(ttbr1_cfg->ias) - ADRENO_VM_START; +} + +void adreno_check_and_reenable_stall(struct adreno_gpu *adreno_gpu) +{ + struct msm_gpu *gpu = &adreno_gpu->base; + struct msm_drm_private *priv = gpu->dev->dev_private; + unsigned long flags; + + /* + * Wait until the cooldown period has passed and we would actually + * collect a crashdump to re-enable stall-on-fault. + */ + spin_lock_irqsave(&priv->fault_stall_lock, flags); + if (!priv->stall_enabled && + ktime_after(ktime_get(), priv->stall_reenable_time) && + !READ_ONCE(gpu->crashstate)) { + struct msm_mmu *mmu = to_msm_vm(gpu->vm)->mmu; + + priv->stall_enabled = true; + + mmu->funcs->set_stall(mmu, true); + } + spin_unlock_irqrestore(&priv->fault_stall_lock, flags); +} + +#define ARM_SMMU_FSR_TF BIT(1) +#define ARM_SMMU_FSR_PF BIT(3) +#define ARM_SMMU_FSR_EF BIT(4) +#define ARM_SMMU_FSR_SS BIT(30) + +int adreno_fault_handler(struct msm_gpu *gpu, unsigned long iova, int flags, + struct adreno_smmu_fault_info *info, const char *block, + u32 scratch[4]) +{ + struct adreno_gpu *adreno_gpu = container_of(gpu, struct adreno_gpu, base); + struct msm_drm_private *priv = gpu->dev->dev_private; + struct msm_mmu *mmu = to_msm_vm(gpu->vm)->mmu; + const char *type = "UNKNOWN"; + bool do_devcoredump = info && (info->fsr & ARM_SMMU_FSR_SS) && + !READ_ONCE(gpu->crashstate); + unsigned long irq_flags; + + /* + * In case there is a subsequent storm of pagefaults, disable + * stall-on-fault for at least half a second. + */ + spin_lock_irqsave(&priv->fault_stall_lock, irq_flags); + if (priv->stall_enabled) { + priv->stall_enabled = false; + + mmu->funcs->set_stall(mmu, false); + } + + priv->stall_reenable_time = ktime_add_ms(ktime_get(), 500); + spin_unlock_irqrestore(&priv->fault_stall_lock, irq_flags); + + /* + * Print a default message if we couldn't get the data from the + * adreno-smmu-priv + */ + if (!info) { + pr_warn_ratelimited("*** gpu fault: iova=%.16lx flags=%d (%u,%u,%u,%u)\n", + iova, flags, + scratch[0], scratch[1], scratch[2], scratch[3]); + + return 0; + } + + if (info->fsr & ARM_SMMU_FSR_TF) + type = "TRANSLATION"; + else if (info->fsr & ARM_SMMU_FSR_PF) + type = "PERMISSION"; + else if (info->fsr & ARM_SMMU_FSR_EF) + type = "EXTERNAL"; + + pr_warn_ratelimited("*** gpu fault: ttbr0=%.16llx iova=%.16lx dir=%s type=%s source=%s (%u,%u,%u,%u)\n", + info->ttbr0, iova, + flags & IOMMU_FAULT_WRITE ? "WRITE" : "READ", + type, block, + scratch[0], scratch[1], scratch[2], scratch[3]); + + if (do_devcoredump) { + struct msm_gpu_fault_info fault_info = {}; + + /* Turn off the hangcheck timer to keep it from bothering us */ + timer_delete(&gpu->hangcheck_timer); + + /* Let any concurrent GMU transactions know that the MMU may be + * blocked for a while and they should wait on us. + */ + reinit_completion(&adreno_gpu->fault_coredump_done); + + fault_info.ttbr0 = info->ttbr0; + fault_info.iova = iova; + fault_info.flags = flags; + fault_info.type = type; + fault_info.block = block; + + msm_gpu_fault_crashstate_capture(gpu, &fault_info); + + complete_all(&adreno_gpu->fault_coredump_done); + } + + return 0; +} + +static bool +valid_per_process_vm(struct msm_gpu *gpu, struct drm_gpuvm *vm) +{ + return vm && (vm != gpu->vm); +} + +int adreno_get_param(struct msm_gpu *gpu, struct msm_context *ctx, + uint32_t param, uint64_t *value, uint32_t *len) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct drm_device *drm = gpu->dev; + /* Note ctx can be NULL when called from rd_open(): */ + struct drm_gpuvm *vm = ctx ? msm_context_vm(drm, ctx) : NULL; + + /* No pointer params yet */ + if (*len != 0) + return UERR(EINVAL, drm, "invalid len"); + + switch (param) { + case MSM_PARAM_GPU_ID: + *value = adreno_gpu->info->revn; + return 0; + case MSM_PARAM_GMEM_SIZE: + *value = adreno_gpu->info->gmem; + return 0; + case MSM_PARAM_GMEM_BASE: + if (adreno_gpu->info->family >= ADRENO_6XX_GEN3) + *value = 0; + else + *value = 0x100000; + return 0; + case MSM_PARAM_CHIP_ID: + *value = adreno_gpu->chip_id; + if (!adreno_gpu->info->revn) + *value |= ((uint64_t) adreno_gpu->speedbin) << 32; + return 0; + case MSM_PARAM_MAX_FREQ: + *value = adreno_gpu->base.fast_rate; + return 0; + case MSM_PARAM_TIMESTAMP: + if (adreno_gpu->funcs->get_timestamp) { + pm_runtime_get_sync(&gpu->pdev->dev); + *value = adreno_gpu->funcs->get_timestamp(gpu); + pm_runtime_put_autosuspend(&gpu->pdev->dev); + + return 0; + } + return -EINVAL; + case MSM_PARAM_PRIORITIES: + *value = gpu->nr_rings * NR_SCHED_PRIORITIES; + return 0; + case MSM_PARAM_PP_PGTABLE: + *value = 0; + return 0; + case MSM_PARAM_FAULTS: + if (vm) + *value = gpu->global_faults + to_msm_vm(vm)->faults; + else + *value = gpu->global_faults; + return 0; + case MSM_PARAM_SUSPENDS: + *value = gpu->suspend_count; + return 0; + case MSM_PARAM_VA_START: + if (!valid_per_process_vm(gpu, vm)) + return UERR(EINVAL, drm, "requires per-process pgtables"); + *value = vm->mm_start; + return 0; + case MSM_PARAM_VA_SIZE: + if (!valid_per_process_vm(gpu, vm)) + return UERR(EINVAL, drm, "requires per-process pgtables"); + *value = vm->mm_range; + return 0; + case MSM_PARAM_HIGHEST_BANK_BIT: + if (!adreno_gpu->ubwc_config) + return UERR(ENOENT, drm, "no UBWC on this platform"); + *value = adreno_gpu->ubwc_config->highest_bank_bit; + return 0; + case MSM_PARAM_RAYTRACING: + *value = adreno_gpu->has_ray_tracing; + return 0; + case MSM_PARAM_UBWC_SWIZZLE: + if (!adreno_gpu->ubwc_config) + return UERR(ENOENT, drm, "no UBWC on this platform"); + *value = qcom_ubwc_swizzle(adreno_gpu->ubwc_config); + return 0; + case MSM_PARAM_MACROTILE_MODE: + if (!adreno_gpu->ubwc_config) + return UERR(ENOENT, drm, "no UBWC on this platform"); + *value = qcom_ubwc_macrotile_mode(adreno_gpu->ubwc_config); + return 0; + case MSM_PARAM_UCHE_TRAP_BASE: + *value = adreno_gpu->uche_trap_base; + return 0; + case MSM_PARAM_HAS_PRR: + *value = adreno_smmu_has_prr(gpu); + return 0; + case MSM_PARAM_AQE: + *value = !!(adreno_gpu->funcs->aqe_is_enabled && + adreno_gpu->funcs->aqe_is_enabled(adreno_gpu)); + return 0; + default: + return UERR(EINVAL, drm, "%s: invalid param: %u", gpu->name, param); + } +} + +int adreno_set_param(struct msm_gpu *gpu, struct msm_context *ctx, + uint32_t param, uint64_t value, uint32_t len) +{ + struct drm_device *drm = gpu->dev; + + switch (param) { + case MSM_PARAM_COMM: + case MSM_PARAM_CMDLINE: + /* kstrdup_quotable_cmdline() limits to PAGE_SIZE, so + * that should be a reasonable upper bound + */ + if (len > PAGE_SIZE) + return UERR(EINVAL, drm, "invalid len"); + break; + default: + if (len != 0) + return UERR(EINVAL, drm, "invalid len"); + } + + switch (param) { + case MSM_PARAM_COMM: + case MSM_PARAM_CMDLINE: { + char *str, **paramp; + + str = memdup_user_nul(u64_to_user_ptr(value), len); + if (IS_ERR(str)) + return PTR_ERR(str); + + mutex_lock(&gpu->lock); + + if (param == MSM_PARAM_COMM) { + paramp = &ctx->comm; + } else { + paramp = &ctx->cmdline; + } + + kfree(*paramp); + *paramp = str; + + mutex_unlock(&gpu->lock); + + return 0; + } + case MSM_PARAM_SYSPROF: + if (!perfmon_capable()) + return UERR(EPERM, drm, "invalid permissions"); + return msm_context_set_sysprof(ctx, gpu, value); + case MSM_PARAM_EN_VM_BIND: { + guard(rwsem_read)(&ctx->ctxlock); + + /* We can only support VM_BIND with per-process pgtables: */ + if (!gpu->funcs->create_private_vm) + return UERR(EINVAL, drm, "requires per-process pgtables"); + + /* + * We can only swtich to VM_BIND mode if the VM has not yet + * been created: + */ + if (ctx->vm) + return UERR(EBUSY, drm, "VM already created"); + + ctx->userspace_managed_vm = value; + + return 0; + } + default: + return UERR(EINVAL, drm, "%s: invalid param: %u", gpu->name, param); + } +} + +const struct firmware * +adreno_request_fw(struct adreno_gpu *adreno_gpu, const char *fwname) +{ + struct drm_device *drm = adreno_gpu->base.dev; + const struct firmware *fw = NULL; + char *newname; + int ret; + + newname = kasprintf(GFP_KERNEL, "qcom/%s", fwname); + if (!newname) + return ERR_PTR(-ENOMEM); + + /* + * Try first to load from qcom/$fwfile using a direct load (to avoid + * a potential timeout waiting for usermode helper) + */ + if ((adreno_gpu->fwloc == FW_LOCATION_UNKNOWN) || + (adreno_gpu->fwloc == FW_LOCATION_NEW)) { + + ret = request_firmware_direct(&fw, newname, drm->dev); + if (!ret) { + DRM_DEV_INFO(drm->dev, "loaded %s from new location\n", + newname); + adreno_gpu->fwloc = FW_LOCATION_NEW; + goto out; + } else if (adreno_gpu->fwloc != FW_LOCATION_UNKNOWN) { + DRM_DEV_ERROR(drm->dev, "failed to load %s: %d\n", + newname, ret); + fw = ERR_PTR(ret); + goto out; + } + } + + /* + * Then try the legacy location without qcom/ prefix + */ + if ((adreno_gpu->fwloc == FW_LOCATION_UNKNOWN) || + (adreno_gpu->fwloc == FW_LOCATION_LEGACY)) { + + ret = request_firmware_direct(&fw, fwname, drm->dev); + if (!ret) { + DRM_DEV_INFO(drm->dev, "loaded %s from legacy location\n", + fwname); + adreno_gpu->fwloc = FW_LOCATION_LEGACY; + goto out; + } else if (adreno_gpu->fwloc != FW_LOCATION_UNKNOWN) { + DRM_DEV_ERROR(drm->dev, "failed to load %s: %d\n", + fwname, ret); + fw = ERR_PTR(ret); + goto out; + } + } + + /* + * Finally fall back to request_firmware() for cases where the + * usermode helper is needed (I think mainly android) + */ + if ((adreno_gpu->fwloc == FW_LOCATION_UNKNOWN) || + (adreno_gpu->fwloc == FW_LOCATION_HELPER)) { + + ret = request_firmware(&fw, newname, drm->dev); + if (!ret) { + DRM_DEV_INFO(drm->dev, "loaded %s with helper\n", + newname); + adreno_gpu->fwloc = FW_LOCATION_HELPER; + goto out; + } else if (adreno_gpu->fwloc != FW_LOCATION_UNKNOWN) { + DRM_DEV_ERROR(drm->dev, "failed to load %s: %d\n", + newname, ret); + fw = ERR_PTR(ret); + goto out; + } + } + + DRM_DEV_ERROR(drm->dev, "failed to load %s\n", fwname); + fw = ERR_PTR(-ENOENT); +out: + kfree(newname); + return fw; +} + +int adreno_load_fw(struct adreno_gpu *adreno_gpu) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(adreno_gpu->info->fw); i++) { + const struct firmware *fw; + + if (!adreno_gpu->info->fw[i]) + continue; + + /* Skip loading GMU firmware with GMU Wrapper */ + if (adreno_has_gmu_wrapper(adreno_gpu) && i == ADRENO_FW_GMU) + continue; + + /* Skip if the firmware has already been loaded */ + if (adreno_gpu->fw[i]) + continue; + + fw = adreno_request_fw(adreno_gpu, adreno_gpu->info->fw[i]); + if (IS_ERR(fw)) + return PTR_ERR(fw); + + adreno_gpu->fw[i] = fw; + } + + return 0; +} + +struct drm_gem_object *adreno_fw_create_bo(struct msm_gpu *gpu, + const struct firmware *fw, u64 *iova) +{ + struct drm_gem_object *bo; + void *ptr; + + ptr = msm_gem_kernel_new(gpu->dev, fw->size - 4, + MSM_BO_WC | MSM_BO_GPU_READONLY, gpu->vm, &bo, iova); + + if (IS_ERR(ptr)) + return ERR_CAST(ptr); + + memcpy(ptr, &fw->data[4], fw->size - 4); + + msm_gem_put_vaddr(bo); + + return bo; +} + +int adreno_hw_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + int ret; + + VERB("%s", gpu->name); + + if (adreno_gpu->info->family >= ADRENO_6XX_GEN1 && + qcom_scm_set_gpu_smmu_aperture_is_available()) { + /* We currently always use context bank 0, so hard code this */ + ret = qcom_scm_set_gpu_smmu_aperture(0); + if (ret) + DRM_DEV_ERROR(gpu->dev->dev, "unable to set SMMU aperture: %d\n", ret); + } + + for (int i = 0; i < gpu->nr_rings; i++) { + struct msm_ringbuffer *ring = gpu->rb[i]; + + if (!ring) + continue; + + ring->cur = ring->start; + ring->next = ring->start; + ring->memptrs->rptr = 0; + ring->memptrs->bv_fence = ring->fctx->completed_fence; + + /* Detect and clean up an impossible fence, ie. if GPU managed + * to scribble something invalid, we don't want that to confuse + * us into mistakingly believing that submits have completed. + */ + if (fence_before(ring->fctx->last_fence, ring->memptrs->fence)) { + ring->memptrs->fence = ring->fctx->last_fence; + } + } + + return 0; +} + +/* Use this helper to read rptr, since a430 doesn't update rptr in memory */ +static uint32_t get_rptr(struct adreno_gpu *adreno_gpu, + struct msm_ringbuffer *ring) +{ + struct msm_gpu *gpu = &adreno_gpu->base; + + return gpu->funcs->get_rptr(gpu, ring); +} + +struct msm_ringbuffer *adreno_active_ring(struct msm_gpu *gpu) +{ + return gpu->rb[0]; +} + +void adreno_recover(struct msm_gpu *gpu) +{ + struct drm_device *dev = gpu->dev; + int ret; + + msm_perfcntr_suspend(gpu); + gpu->funcs->pm_suspend(gpu); + gpu->funcs->pm_resume(gpu); + msm_perfcntr_resume(gpu); + + ret = msm_gpu_hw_init(gpu); + if (ret) { + DRM_DEV_ERROR(dev->dev, "gpu hw init failed: %d\n", ret); + /* hmm, oh well? */ + } +} + +void adreno_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring, u32 reg) +{ + uint32_t wptr; + + /* Copy the shadow to the actual register */ + ring->cur = ring->next; + + /* + * Mask wptr value that we calculate to fit in the HW range. This is + * to account for the possibility that the last command fit exactly into + * the ringbuffer and rb->next hasn't wrapped to zero yet + */ + wptr = get_wptr(ring); + + /* ensure writes to ringbuffer have hit system memory: */ + mb(); + + gpu_write(gpu, reg, wptr); +} + +bool adreno_idle(struct msm_gpu *gpu, struct msm_ringbuffer *ring) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + uint32_t wptr = get_wptr(ring); + + /* wait for CP to drain ringbuffer: */ + if (!spin_until(get_rptr(adreno_gpu, ring) == wptr)) + return true; + + /* TODO maybe we need to reset GPU here to recover from hang? */ + DRM_ERROR("%s: timeout waiting to drain ringbuffer %d rptr/wptr = %X/%X\n", + gpu->name, ring->id, get_rptr(adreno_gpu, ring), wptr); + + return false; +} + +int adreno_gpu_state_get(struct msm_gpu *gpu, struct msm_gpu_state *state) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + int i, count = 0; + + WARN_ON(!mutex_is_locked(&gpu->lock)); + + kref_init(&state->ref); + + ktime_get_real_ts64(&state->time); + + for (i = 0; i < gpu->nr_rings; i++) { + int size = 0, j; + + state->ring[i].fence = gpu->rb[i]->memptrs->fence; + state->ring[i].iova = gpu->rb[i]->iova; + state->ring[i].seqno = gpu->rb[i]->fctx->last_fence; + state->ring[i].rptr = get_rptr(adreno_gpu, gpu->rb[i]); + state->ring[i].wptr = get_wptr(gpu->rb[i]); + + /* Copy at least 'wptr' dwords of the data */ + size = state->ring[i].wptr; + + /* After wptr find the last non zero dword to save space */ + for (j = state->ring[i].wptr; j < MSM_GPU_RINGBUFFER_SZ >> 2; j++) + if (gpu->rb[i]->start[j]) + size = j + 1; + + if (size) { + state->ring[i].data = kvmemdup(gpu->rb[i]->start, size << 2, GFP_KERNEL); + if (state->ring[i].data) + state->ring[i].data_size = size << 2; + } + } + + /* Some targets prefer to collect their own registers */ + if (!adreno_gpu->registers) + return 0; + + /* Count the number of registers */ + for (i = 0; adreno_gpu->registers[i] != ~0; i += 2) + count += adreno_gpu->registers[i + 1] - + adreno_gpu->registers[i] + 1; + + state->registers = kcalloc(count * 2, sizeof(u32), GFP_KERNEL); + if (state->registers) { + int pos = 0; + + for (i = 0; adreno_gpu->registers[i] != ~0; i += 2) { + u32 start = adreno_gpu->registers[i]; + u32 end = adreno_gpu->registers[i + 1]; + u32 addr; + + for (addr = start; addr <= end; addr++) { + state->registers[pos++] = addr; + state->registers[pos++] = gpu_read(gpu, addr); + } + } + + state->nr_registers = count; + } + + return 0; +} + +void adreno_gpu_state_destroy(struct msm_gpu_state *state) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(state->ring); i++) + kvfree(state->ring[i].data); + + for (i = 0; state->bos && i < state->nr_bos; i++) + kvfree(state->bos[i].data); + + kfree(state->vm_logs); + kfree(state->bos); + kfree(state->comm); + kfree(state->cmd); + kfree(state->registers); +} + +static void adreno_gpu_state_kref_destroy(struct kref *kref) +{ + struct msm_gpu_state *state = container_of(kref, + struct msm_gpu_state, ref); + + adreno_gpu_state_destroy(state); + kfree(state); +} + +int adreno_gpu_state_put(struct msm_gpu_state *state) +{ + if (IS_ERR_OR_NULL(state)) + return 1; + + return kref_put(&state->ref, adreno_gpu_state_kref_destroy); +} + +#if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP) + +static char *adreno_gpu_ascii85_encode(u32 *src, size_t len) +{ + void *buf; + size_t buf_itr = 0, buffer_size; + char out[ASCII85_BUFSZ]; + long l; + int i; + + if (!src || !len) + return NULL; + + l = ascii85_encode_len(len); + + /* + * Ascii85 outputs either a 5 byte string or a 1 byte string. So we + * account for the worst case of 5 bytes per dword plus the 1 for '\0' + */ + buffer_size = (l * 5) + 1; + + buf = kvmalloc(buffer_size, GFP_KERNEL); + if (!buf) + return NULL; + + for (i = 0; i < l; i++) + buf_itr += scnprintf(buf + buf_itr, buffer_size - buf_itr, "%s", + ascii85_encode(src[i], out)); + + return buf; +} + +/* len is expected to be in bytes + * + * WARNING: *ptr should be allocated with kvmalloc or friends. It can be free'd + * with kvfree() and replaced with a newly kvmalloc'd buffer on the first call + * when the unencoded raw data is encoded + */ +void adreno_show_object(struct drm_printer *p, void **ptr, int len, + bool *encoded) +{ + if (!*ptr || !len) + return; + + if (!*encoded) { + long datalen, i; + u32 *buf = *ptr; + + /* + * Only dump the non-zero part of the buffer - rarely will + * any data completely fill the entire allocated size of + * the buffer. + */ + for (datalen = 0, i = 0; i < len >> 2; i++) + if (buf[i]) + datalen = ((i + 1) << 2); + + /* + * If we reach here, then the originally captured binary buffer + * will be replaced with the ascii85 encoded string + */ + *ptr = adreno_gpu_ascii85_encode(buf, datalen); + + kvfree(buf); + + *encoded = true; + } + + if (!*ptr) + return; + + drm_puts(p, " data: !!ascii85 |\n"); + drm_puts(p, " "); + + drm_puts(p, *ptr); + + drm_puts(p, "\n"); +} + +void adreno_show(struct msm_gpu *gpu, struct msm_gpu_state *state, + struct drm_printer *p) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + int i; + + if (IS_ERR_OR_NULL(state)) + return; + + drm_printf(p, "revision: %u (%"ADRENO_CHIPID_FMT")\n", + adreno_gpu->info->revn, + ADRENO_CHIPID_ARGS(adreno_gpu->chip_id)); + /* + * If this is state collected due to iova fault, so fault related info + * + * TTBR0 would not be zero, so this is a good way to distinguish + */ + if (state->fault_info.ttbr0) { + const struct msm_gpu_fault_info *info = &state->fault_info; + + drm_puts(p, "fault-info:\n"); + drm_printf(p, " - ttbr0=%.16llx\n", info->ttbr0); + drm_printf(p, " - iova=%.16lx\n", info->iova); + drm_printf(p, " - dir=%s\n", info->flags & IOMMU_FAULT_WRITE ? "WRITE" : "READ"); + drm_printf(p, " - type=%s\n", info->type); + drm_printf(p, " - source=%s\n", info->block); + + /* Information extracted from what we think are the current + * pgtables. Hopefully the TTBR0 matches what we've extracted + * from the SMMU registers in smmu_info! + */ + drm_puts(p, "pgtable-fault-info:\n"); + drm_printf(p, " - ttbr0: %.16llx\n", (u64)info->pgtbl_ttbr0); + drm_printf(p, " - asid: %d\n", info->asid); + drm_printf(p, " - ptes: %.16llx %.16llx %.16llx %.16llx\n", + info->ptes[0], info->ptes[1], info->ptes[2], info->ptes[3]); + } + + if (state->vm_logs) { + drm_puts(p, "vm-log:\n"); + for (i = 0; i < state->nr_vm_logs; i++) { + struct msm_gem_vm_log_entry *e = &state->vm_logs[i]; + drm_printf(p, " - %s:%d: 0x%016llx-0x%016llx\n", + e->op, e->queue_id, e->iova, + e->iova + e->range); + } + } + + drm_printf(p, "rbbm-status: 0x%08x\n", state->rbbm_status); + + drm_puts(p, "ringbuffer:\n"); + + for (i = 0; i < gpu->nr_rings; i++) { + drm_printf(p, " - id: %d\n", i); + drm_printf(p, " iova: 0x%016llx\n", state->ring[i].iova); + drm_printf(p, " last-fence: %u\n", state->ring[i].seqno); + drm_printf(p, " retired-fence: %u\n", state->ring[i].fence); + drm_printf(p, " rptr: %u\n", state->ring[i].rptr); + drm_printf(p, " wptr: %u\n", state->ring[i].wptr); + drm_printf(p, " size: %u\n", MSM_GPU_RINGBUFFER_SZ); + + adreno_show_object(p, &state->ring[i].data, + state->ring[i].data_size, &state->ring[i].encoded); + } + + if (state->bos) { + drm_puts(p, "bos:\n"); + + for (i = 0; i < state->nr_bos; i++) { + drm_printf(p, " - iova: 0x%016llx\n", + state->bos[i].iova); + drm_printf(p, " size: %zd\n", state->bos[i].size); + drm_printf(p, " flags: 0x%x\n", state->bos[i].flags); + drm_printf(p, " name: %-32s\n", state->bos[i].name); + + adreno_show_object(p, &state->bos[i].data, + state->bos[i].size, &state->bos[i].encoded); + } + } + + if (state->nr_registers) { + drm_puts(p, "registers:\n"); + + for (i = 0; i < state->nr_registers; i++) { + drm_printf(p, " - { offset: 0x%04x, value: 0x%08x }\n", + state->registers[i * 2] << 2, + state->registers[(i * 2) + 1]); + } + } +} +#endif + +/* Dump common gpu status and scratch registers on any hang, to make + * the hangcheck logs more useful. The scratch registers seem always + * safe to read when GPU has hung (unlike some other regs, depending + * on how the GPU hung), and they are useful to match up to cmdstream + * dumps when debugging hangs: + */ +void adreno_dump_info(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + int i; + + printk("revision: %u (%"ADRENO_CHIPID_FMT")\n", + adreno_gpu->info->revn, + ADRENO_CHIPID_ARGS(adreno_gpu->chip_id)); + + for (i = 0; i < gpu->nr_rings; i++) { + struct msm_ringbuffer *ring = gpu->rb[i]; + + printk("rb %d: fence: %d/%d\n", i, + ring->memptrs->fence, + ring->fctx->last_fence); + + printk("rptr: %d\n", get_rptr(adreno_gpu, ring)); + printk("rb wptr: %d\n", get_wptr(ring)); + } +} + +/* would be nice to not have to duplicate the _show() stuff with printk(): */ +void adreno_dump(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + int i; + + if (!adreno_gpu->registers) + return; + + /* dump these out in a form that can be parsed by demsm: */ + printk("IO:region %s 00000000 00020000\n", gpu->name); + for (i = 0; adreno_gpu->registers[i] != ~0; i += 2) { + uint32_t start = adreno_gpu->registers[i]; + uint32_t end = adreno_gpu->registers[i+1]; + uint32_t addr; + + for (addr = start; addr <= end; addr++) { + uint32_t val = gpu_read(gpu, addr); + printk("IO:R %08x %08x\n", addr<<2, val); + } + } +} + +static uint32_t ring_freewords(struct msm_ringbuffer *ring) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(ring->gpu); + uint32_t size = MSM_GPU_RINGBUFFER_SZ >> 2; + /* Use ring->next to calculate free size */ + uint32_t wptr = ring->next - ring->start; + uint32_t rptr = get_rptr(adreno_gpu, ring); + return (rptr + (size - 1) - wptr) % size; +} + +void adreno_wait_ring(struct msm_ringbuffer *ring, uint32_t ndwords) +{ + if (spin_until(ring_freewords(ring) >= ndwords)) + DRM_DEV_ERROR(ring->gpu->dev->dev, + "timeout waiting for space in ringbuffer %d\n", + ring->id); +} + +static int adreno_get_pwrlevels(struct device *dev, + struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + unsigned long freq = ULONG_MAX; + struct dev_pm_opp *opp; + int ret; + + gpu->fast_rate = 0; + + /* devm_pm_opp_of_add_table may error out but will still create an OPP table */ + ret = devm_pm_opp_of_add_table(dev); + if (ret == -ENODEV) { + /* Special cases for ancient hw with ancient DT bindings */ + if (adreno_is_a2xx(adreno_gpu)) { + dev_warn(dev, "Unable to find the OPP table. Falling back to 200 MHz.\n"); + dev_pm_opp_add(dev, 200000000, 0); + } else if (adreno_is_a320(adreno_gpu)) { + dev_warn(dev, "Unable to find the OPP table. Falling back to 450 MHz.\n"); + dev_pm_opp_add(dev, 450000000, 0); + } else { + DRM_DEV_ERROR(dev, "Unable to find the OPP table\n"); + return -ENODEV; + } + } else if (ret) { + DRM_DEV_ERROR(dev, "Unable to set the OPP table\n"); + return ret; + } + + /* Find the fastest defined rate */ + opp = dev_pm_opp_find_freq_floor(dev, &freq); + if (IS_ERR(opp)) + return PTR_ERR(opp); + + gpu->fast_rate = freq; + dev_pm_opp_put(opp); + + DBG("fast_rate=%u, slow_rate=27000000", gpu->fast_rate); + + return 0; +} + +int adreno_gpu_ocmem_init(struct device *dev, struct adreno_gpu *adreno_gpu, + struct adreno_ocmem *adreno_ocmem) +{ + struct ocmem_buf *ocmem_hdl; + struct ocmem *ocmem; + + ocmem = of_get_ocmem(dev); + if (IS_ERR(ocmem)) { + if (PTR_ERR(ocmem) == -ENODEV) { + /* + * Return success since either the ocmem property was + * not specified in device tree, or ocmem support is + * not compiled into the kernel. + */ + return 0; + } + + return PTR_ERR(ocmem); + } + + ocmem_hdl = ocmem_allocate(ocmem, OCMEM_GRAPHICS, adreno_gpu->info->gmem); + if (IS_ERR(ocmem_hdl)) + return PTR_ERR(ocmem_hdl); + + adreno_ocmem->ocmem = ocmem; + adreno_ocmem->base = ocmem_hdl->addr; + adreno_ocmem->hdl = ocmem_hdl; + + if (WARN_ON(ocmem_hdl->len != adreno_gpu->info->gmem)) + return -ENOMEM; + + return 0; +} + +void adreno_gpu_ocmem_cleanup(struct adreno_ocmem *adreno_ocmem) +{ + if (adreno_ocmem && adreno_ocmem->base) + ocmem_free(adreno_ocmem->ocmem, OCMEM_GRAPHICS, + adreno_ocmem->hdl); +} + +int adreno_read_speedbin(struct device *dev, u32 *speedbin) +{ + return nvmem_cell_read_variable_le_u32(dev, "speed_bin", speedbin); +} + +int adreno_gpu_init(struct drm_device *drm, struct platform_device *pdev, + struct adreno_gpu *adreno_gpu, + const struct adreno_gpu_funcs *funcs, int nr_rings) +{ + struct device *dev = &pdev->dev; + struct adreno_platform_config *config = dev->platform_data; + struct msm_gpu_config adreno_gpu_config = { 0 }; + struct msm_gpu *gpu = &adreno_gpu->base; + const char *gpu_name; + int ret; + + adreno_gpu->funcs = funcs; + adreno_gpu->info = config->info; + adreno_gpu->chip_id = config->chip_id; + + gpu->allow_relocs = config->info->family < ADRENO_6XX_GEN1; + gpu->pdev = pdev; + + /* Only handle the core clock when GMU is not in use (or is absent). */ + if (adreno_has_gmu_wrapper(adreno_gpu) || + adreno_has_rgmu(adreno_gpu) || + adreno_gpu->info->family < ADRENO_6XX_GEN1) { + /* + * This can only be done before devm_pm_opp_of_add_table(), or + * dev_pm_opp_set_config() will WARN_ON() + */ + if (IS_ERR(devm_clk_get(dev, "core"))) { + /* + * If "core" is absent, go for the legacy clock name. + * If we got this far in probing, it's a given one of + * them exists. + */ + devm_pm_opp_set_clkname(dev, "core_clk"); + } else + devm_pm_opp_set_clkname(dev, "core"); + } + + gpu_name = devm_kasprintf(dev, GFP_KERNEL, "%"ADRENO_CHIPID_FMT, + ADRENO_CHIPID_ARGS(config->chip_id)); + if (!gpu_name) + return -ENOMEM; + + adreno_gpu_config.ioname = "kgsl_3d0_reg_memory"; + + adreno_gpu_config.nr_rings = nr_rings; + + ret = adreno_get_pwrlevels(dev, gpu); + if (ret) + return ret; + + init_completion(&adreno_gpu->fault_coredump_done); + complete_all(&adreno_gpu->fault_coredump_done); + + pm_runtime_set_autosuspend_delay(dev, + adreno_gpu->info->inactive_period); + pm_runtime_use_autosuspend(dev); + + return msm_gpu_init(drm, pdev, &adreno_gpu->base, &funcs->base, + gpu_name, &adreno_gpu_config); +} + +void adreno_gpu_cleanup(struct adreno_gpu *adreno_gpu) +{ + struct msm_gpu *gpu = &adreno_gpu->base; + struct msm_drm_private *priv = gpu->dev ? gpu->dev->dev_private : NULL; + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(adreno_gpu->info->fw); i++) + release_firmware(adreno_gpu->fw[i]); + + pm_runtime_dont_use_autosuspend(&gpu->pdev->dev); + + if (priv && pm_runtime_enabled(&priv->gpu_pdev->dev)) + pm_runtime_disable(&priv->gpu_pdev->dev); + + msm_gpu_cleanup(&adreno_gpu->base); +} diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h new file mode 100644 index 000000000..114a40f79 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h @@ -0,0 +1,769 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + * + * Copyright (c) 2014,2017, 2019 The Linux Foundation. All rights reserved. + */ + +#ifndef __ADRENO_GPU_H__ +#define __ADRENO_GPU_H__ + +#include +#include + +#include + +#include "msm_gpu.h" + +#include "adreno_common.xml.h" +#include "adreno_pm4.xml.h" + +extern bool snapshot_debugbus; + +enum { + ADRENO_FW_PM4 = 0, + ADRENO_FW_SQE = 0, /* a6xx */ + ADRENO_FW_PFP = 1, + ADRENO_FW_GMU = 1, /* a6xx */ + ADRENO_FW_GPMU = 2, + ADRENO_FW_AQE = 3, + ADRENO_FW_MAX, +}; + +/** + * @enum adreno_family: identify generation and possibly sub-generation + * + * In some cases there are distinct sub-generations within a major revision + * so it helps to be able to group the GPU devices by generation and if + * necessary sub-generation. + */ +enum adreno_family { + ADRENO_2XX_GEN1, /* a20x */ + ADRENO_2XX_GEN2, /* a22x */ + ADRENO_3XX, + ADRENO_4XX, + ADRENO_5XX, + ADRENO_6XX_GEN1, /* a630 family */ + ADRENO_6XX_GEN2, /* a640 family */ + ADRENO_6XX_GEN3, /* a650 family */ + ADRENO_6XX_GEN4, /* a660 family */ + ADRENO_7XX_GEN1, /* a730 family */ + ADRENO_7XX_GEN2, /* a740 family */ + ADRENO_7XX_GEN3, /* a750 family */ + ADRENO_8XX_GEN1, /* a830 family */ + ADRENO_8XX_GEN2, /* a840 family */ +}; + +#define ADRENO_QUIRK_TWO_PASS_USE_WFI BIT(0) +#define ADRENO_QUIRK_FAULT_DETECT_MASK BIT(1) +#define ADRENO_QUIRK_LMLOADKILL_DISABLE BIT(2) +#define ADRENO_QUIRK_HAS_HW_APRIV BIT(3) +#define ADRENO_QUIRK_HAS_CACHED_COHERENT BIT(4) +#define ADRENO_QUIRK_PREEMPTION BIT(5) +#define ADRENO_QUIRK_4GB_VA BIT(6) +#define ADRENO_QUIRK_IFPC BIT(7) +#define ADRENO_QUIRK_SOFTFUSE BIT(8) + +/* Helper for formating the chip_id in the way that userspace tools like + * crashdec expect. + */ +#define ADRENO_CHIPID_FMT "08x" +#define ADRENO_CHIPID_ARGS(_c) (_c) + +struct adreno_gpu; + +struct adreno_gpu_funcs { + struct msm_gpu_funcs base; + struct msm_gpu *(*init)(struct drm_device *dev); + u64 (*get_timestamp)(struct msm_gpu *gpu); + void (*bus_halt)(struct adreno_gpu *adreno_gpu, bool gx_off); + int (*mmu_fault_handler)(void *arg, unsigned long iova, int flags, void *data); + bool (*gx_is_on)(struct adreno_gpu *adreno_gpu); + bool (*aqe_is_enabled)(struct adreno_gpu *adreno_gpu); +}; + +struct adreno_reglist { + u32 offset; + u32 value; +}; + +/* Reglist with pipe information */ +struct adreno_reglist_pipe { + u32 offset; + u32 value; + u32 pipe; +}; + +struct adreno_speedbin { + uint16_t fuse; + uint16_t speedbin; +}; + +struct a6xx_info; + +struct adreno_info { + const char *machine; + /** + * @chipids: Table of matching chip-ids + * + * Terminated with 0 sentinal + */ + uint32_t *chip_ids; + enum adreno_family family; + uint32_t revn; + const char *fw[ADRENO_FW_MAX]; + uint32_t gmem; + u64 quirks; + const struct adreno_gpu_funcs *funcs; + const char *zapfw; + u32 inactive_period; + union { + const struct a6xx_info *a6xx; + }; + /** + * @speedbins: Optional table of fuse to speedbin mappings + * + * Consists of pairs of fuse, index mappings, terminated with + * {SHRT_MAX, 0} sentinal. + */ + struct adreno_speedbin *speedbins; + u64 preempt_record_size; +}; + +#define ADRENO_CHIP_IDS(tbl...) (uint32_t[]) { tbl, 0 } + +struct adreno_gpulist { + const struct adreno_info *gpus; + unsigned gpus_count; +}; + +#define DECLARE_ADRENO_GPULIST(name) \ +const struct adreno_gpulist name ## _gpulist = { \ + name ## _gpus, ARRAY_SIZE(name ## _gpus) \ +} + +/* + * Helper to build a speedbin table, ie. the table: + * fuse | speedbin + * -----+--------- + * 0 | 0 + * 169 | 1 + * 174 | 2 + * + * would be declared as: + * + * .speedbins = ADRENO_SPEEDBINS( + * { 0, 0 }, + * { 169, 1 }, + * { 174, 2 }, + * ), + */ +#define ADRENO_SPEEDBINS(tbl...) (struct adreno_speedbin[]) { tbl {SHRT_MAX, 0} } + +struct adreno_protect { + const uint32_t *regs; + uint32_t count; + uint32_t count_max; +}; + +#define DECLARE_ADRENO_PROTECT(name, __count_max) \ +static const struct adreno_protect name = { \ + .regs = name ## _regs, \ + .count = ARRAY_SIZE(name ## _regs), \ + .count_max = __count_max, \ +}; + +struct adreno_reglist_list { + /** @reg: List of register **/ + const u32 *regs; + /** @count: Number of registers in the list **/ + u32 count; +}; + +#define DECLARE_ADRENO_REGLIST_LIST(name) \ +static const struct adreno_reglist_list name = { \ + .regs = name ## _regs, \ + .count = ARRAY_SIZE(name ## _regs), \ +}; + +struct adreno_reglist_pipe_list { + /** @reg: List of register **/ + const struct adreno_reglist_pipe *regs; + /** @count: Number of registers in the list **/ + u32 count; +}; + +#define DECLARE_ADRENO_REGLIST_PIPE_LIST(name) \ +static const struct adreno_reglist_pipe_list name = { \ + .regs = name ## _regs, \ + .count = ARRAY_SIZE(name ## _regs), \ +}; + +struct adreno_gpu { + struct msm_gpu base; + const struct adreno_info *info; + uint32_t chip_id; + uint16_t speedbin; + const struct adreno_gpu_funcs *funcs; + + struct completion fault_coredump_done; + + /* interesting register offsets to dump: */ + const unsigned int *registers; + + /* + * Are we loading fw from legacy path? Prior to addition + * of gpu firmware to linux-firmware, the fw files were + * placed in toplevel firmware directory, following qcom's + * android kernel. But linux-firmware preferred they be + * placed in a 'qcom' subdirectory. + * + * For backwards compatibility, we try first to load from + * the new path, using request_firmware_direct() to avoid + * any potential timeout waiting for usermode helper, then + * fall back to the old path (with direct load). And + * finally fall back to request_firmware() with the new + * path to allow the usermode helper. + */ + enum { + FW_LOCATION_UNKNOWN = 0, + FW_LOCATION_NEW, /* /lib/firmware/qcom/$fwfile */ + FW_LOCATION_LEGACY, /* /lib/firmware/$fwfile */ + FW_LOCATION_HELPER, + } fwloc; + + /* firmware: */ + const struct firmware *fw[ADRENO_FW_MAX]; + + const struct qcom_ubwc_cfg_data *ubwc_config; + + /* + * Register offsets are different between some GPUs. + * GPU specific offsets will be exported by GPU specific + * code (a3xx_gpu.c) and stored in this common location. + */ + const unsigned int *reg_offsets; + bool gmu_is_wrapper; + + bool has_ray_tracing; + + u64 uche_trap_base; +}; +#define to_adreno_gpu(x) container_of(x, struct adreno_gpu, base) + +struct adreno_ocmem { + struct ocmem *ocmem; + unsigned long base; + void *hdl; +}; + +/* platform config data (ie. from DT, or pdata) */ +struct adreno_platform_config { + uint32_t chip_id; + const struct adreno_info *info; +}; + +#define ADRENO_IDLE_TIMEOUT msecs_to_jiffies(1000) + +#define spin_until(X) ({ \ + int __ret = -ETIMEDOUT; \ + unsigned long __t = jiffies + ADRENO_IDLE_TIMEOUT; \ + do { \ + if (X) { \ + __ret = 0; \ + break; \ + } \ + } while (time_before(jiffies, __t)); \ + __ret; \ +}) + +static inline uint8_t adreno_patchid(const struct adreno_gpu *gpu) +{ + /* It is probably ok to assume legacy "adreno_rev" format + * for all a6xx devices, but probably best to limit this + * to older things. + */ + WARN_ON_ONCE(gpu->info->family >= ADRENO_6XX_GEN1); + return gpu->chip_id & 0xff; +} + +static inline bool adreno_is_revn(const struct adreno_gpu *gpu, uint32_t revn) +{ + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->revn == revn; +} + +static inline bool adreno_has_gmu_wrapper(const struct adreno_gpu *gpu) +{ + return gpu->gmu_is_wrapper; +} + +static inline bool adreno_is_a2xx(const struct adreno_gpu *gpu) +{ + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->family <= ADRENO_2XX_GEN2; +} + +static inline bool adreno_is_a20x(const struct adreno_gpu *gpu) +{ + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->family == ADRENO_2XX_GEN1; +} + +static inline bool adreno_is_a225(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 225); +} + +static inline bool adreno_is_a305(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 305); +} + +static inline bool adreno_is_a305b(const struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x03000512; +} + +static inline bool adreno_is_a306(const struct adreno_gpu *gpu) +{ + /* yes, 307, because a305c is 306 */ + return adreno_is_revn(gpu, 307); +} + +static inline bool adreno_is_a306a(const struct adreno_gpu *gpu) +{ + /* a306a (marketing name is a308) */ + return adreno_is_revn(gpu, 308); +} + +static inline bool adreno_is_a320(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 320); +} + +static inline bool adreno_is_a330(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 330); +} + +static inline bool adreno_is_a330v2(const struct adreno_gpu *gpu) +{ + return adreno_is_a330(gpu) && (adreno_patchid(gpu) > 0); +} + +static inline int adreno_is_a405(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 405); +} + +static inline int adreno_is_a420(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 420); +} + +static inline int adreno_is_a430(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 430); +} + +static inline int adreno_is_a505(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 505); +} + +static inline int adreno_is_a506(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 506); +} + +static inline int adreno_is_a508(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 508); +} + +static inline int adreno_is_a509(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 509); +} + +static inline int adreno_is_a510(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 510); +} + +static inline int adreno_is_a512(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 512); +} + +static inline int adreno_is_a530(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 530); +} + +static inline int adreno_is_a540(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 540); +} + +static inline int adreno_is_a610(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 610); +} + +static inline int adreno_is_a612(const struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x06010200; +} + +static inline bool adreno_has_rgmu(const struct adreno_gpu *gpu) +{ + return adreno_is_a612(gpu); +} + +static inline int adreno_is_a618(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 618); +} + +static inline int adreno_is_a619(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 619); +} + +static inline int adreno_is_a619_holi(const struct adreno_gpu *gpu) +{ + return adreno_is_a619(gpu) && adreno_has_gmu_wrapper(gpu); +} + +static inline int adreno_is_a621(const struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x06020100; +} + +static inline int adreno_is_a623(const struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x06020300; +} + +static inline int adreno_is_a630(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 630); +} + +static inline int adreno_is_a640(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 640); +} + +static inline int adreno_is_a650(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 650); +} + +static inline int adreno_is_7c3(const struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x06030500; +} + +static inline int adreno_is_a660(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 660); +} + +static inline int adreno_is_a680(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 680); +} + +static inline int adreno_is_a663(const struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x06060300; +} + +static inline int adreno_is_a690(const struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x06090000; +} + +static inline int adreno_is_a702(const struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x07000200; +} + +static inline int adreno_is_a610_family(const struct adreno_gpu *gpu) +{ + if (WARN_ON_ONCE(!gpu->info)) + return false; + return adreno_is_a610(gpu) || + adreno_is_a612(gpu) || + adreno_is_a702(gpu); +} + +/* TODO: 615/616 */ +static inline int adreno_is_a615_family(const struct adreno_gpu *gpu) +{ + return adreno_is_a618(gpu) || + adreno_is_a619(gpu); +} + +static inline int adreno_is_a630_family(const struct adreno_gpu *gpu) +{ + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->family == ADRENO_6XX_GEN1; +} + +static inline int adreno_is_a660_family(const struct adreno_gpu *gpu) +{ + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->family == ADRENO_6XX_GEN4; +} + +/* check for a650, a660, or any derivatives */ +static inline int adreno_is_a650_family(const struct adreno_gpu *gpu) +{ + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->family == ADRENO_6XX_GEN3 || + gpu->info->family == ADRENO_6XX_GEN4; +} + +static inline int adreno_is_a640_family(const struct adreno_gpu *gpu) +{ + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->family == ADRENO_6XX_GEN2; +} + +static inline int adreno_is_a730(struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x07030001; +} + +static inline int adreno_is_a740(struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x43050a01; +} + +static inline int adreno_is_a750(struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x43051401; +} + +static inline int adreno_is_x185(struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x43050c01; +} + +static inline int adreno_is_a722(struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x43020100; +} + +static inline int adreno_is_a740_family(struct adreno_gpu *gpu) +{ + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->family == ADRENO_7XX_GEN2 || + gpu->info->family == ADRENO_7XX_GEN3; +} + +static inline int adreno_is_a750_family(struct adreno_gpu *gpu) +{ + return gpu->info->family == ADRENO_7XX_GEN3; +} + +static inline int adreno_is_a7xx(struct adreno_gpu *gpu) +{ + /* Update with non-fake (i.e. non-A702) Gen 7 GPUs */ + return gpu->info->family == ADRENO_7XX_GEN1 || + adreno_is_a740_family(gpu); +} + +static inline int adreno_is_a8xx(struct adreno_gpu *gpu) +{ + return gpu->info->family >= ADRENO_8XX_GEN1; +} + +static inline int adreno_is_a810(struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x44010000; +} + +static inline int adreno_is_x285(struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x44070001; +} + +static inline int adreno_is_a840(struct adreno_gpu *gpu) +{ + return gpu->info->chip_ids[0] == 0x44050a01; +} + +/* Put vm_start above 32b to catch issues with not setting xyz_BASE_HI */ +#define ADRENO_VM_START 0x100000000ULL +u64 adreno_private_vm_size(struct msm_gpu *gpu); +int adreno_get_param(struct msm_gpu *gpu, struct msm_context *ctx, + uint32_t param, uint64_t *value, uint32_t *len); +int adreno_set_param(struct msm_gpu *gpu, struct msm_context *ctx, + uint32_t param, uint64_t value, uint32_t len); +const struct firmware *adreno_request_fw(struct adreno_gpu *adreno_gpu, + const char *fwname); +struct drm_gem_object *adreno_fw_create_bo(struct msm_gpu *gpu, + const struct firmware *fw, u64 *iova); +int adreno_hw_init(struct msm_gpu *gpu); +void adreno_recover(struct msm_gpu *gpu); +void adreno_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring, u32 reg); +bool adreno_idle(struct msm_gpu *gpu, struct msm_ringbuffer *ring); +#if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP) +void adreno_show(struct msm_gpu *gpu, struct msm_gpu_state *state, + struct drm_printer *p); +#endif +void adreno_dump_info(struct msm_gpu *gpu); +void adreno_dump(struct msm_gpu *gpu); +void adreno_wait_ring(struct msm_ringbuffer *ring, uint32_t ndwords); +struct msm_ringbuffer *adreno_active_ring(struct msm_gpu *gpu); + +int adreno_gpu_ocmem_init(struct device *dev, struct adreno_gpu *adreno_gpu, + struct adreno_ocmem *ocmem); +void adreno_gpu_ocmem_cleanup(struct adreno_ocmem *ocmem); + +int adreno_gpu_init(struct drm_device *drm, struct platform_device *pdev, + struct adreno_gpu *gpu, const struct adreno_gpu_funcs *funcs, + int nr_rings); +void adreno_gpu_cleanup(struct adreno_gpu *gpu); +int adreno_load_fw(struct adreno_gpu *adreno_gpu); + +void adreno_gpu_state_destroy(struct msm_gpu_state *state); + +int adreno_gpu_state_get(struct msm_gpu *gpu, struct msm_gpu_state *state); +int adreno_gpu_state_put(struct msm_gpu_state *state); +void adreno_show_object(struct drm_printer *p, void **ptr, int len, + bool *encoded); + +/* + * Common helper function to initialize the default address space for arm-smmu + * attached targets + */ +struct drm_gpuvm * +adreno_create_vm(struct msm_gpu *gpu, + struct platform_device *pdev); + +struct drm_gpuvm * +adreno_iommu_create_vm(struct msm_gpu *gpu, + struct platform_device *pdev, + unsigned long quirks); + +int adreno_fault_handler(struct msm_gpu *gpu, unsigned long iova, int flags, + struct adreno_smmu_fault_info *info, const char *block, + u32 scratch[4]); + +void adreno_check_and_reenable_stall(struct adreno_gpu *gpu); + +int adreno_read_speedbin(struct device *dev, u32 *speedbin); + +/* + * For a5xx and a6xx targets load the zap shader that is used to pull the GPU + * out of secure mode + */ +int adreno_zap_shader_load(struct msm_gpu *gpu, u32 pasid); + +/* ringbuffer helpers (the parts that are adreno specific) */ + +static inline void +OUT_PKT0(struct msm_ringbuffer *ring, uint16_t regindx, uint16_t cnt) +{ + adreno_wait_ring(ring, cnt+1); + OUT_RING(ring, CP_TYPE0_PKT | ((cnt-1) << 16) | (regindx & 0x7FFF)); +} + +/* no-op packet: */ +static inline void +OUT_PKT2(struct msm_ringbuffer *ring) +{ + adreno_wait_ring(ring, 1); + OUT_RING(ring, CP_TYPE2_PKT); +} + +static inline void +OUT_PKT3(struct msm_ringbuffer *ring, uint8_t opcode, uint16_t cnt) +{ + adreno_wait_ring(ring, cnt+1); + OUT_RING(ring, CP_TYPE3_PKT | ((cnt-1) << 16) | ((opcode & 0xFF) << 8)); +} + +static inline u32 PM4_PARITY(u32 val) +{ + return (0x9669 >> (0xF & (val ^ + (val >> 4) ^ (val >> 8) ^ (val >> 12) ^ + (val >> 16) ^ ((val) >> 20) ^ (val >> 24) ^ + (val >> 28)))) & 1; +} + +/* Maximum number of values that can be executed for one opcode */ +#define TYPE4_MAX_PAYLOAD 127 + +#define PKT4(_reg, _cnt) \ + (CP_TYPE4_PKT | ((_cnt) << 0) | (PM4_PARITY((_cnt)) << 7) | \ + (((_reg) & 0x3FFFF) << 8) | (PM4_PARITY((_reg)) << 27)) + +static inline void +OUT_PKT4(struct msm_ringbuffer *ring, uint16_t regindx, uint16_t cnt) +{ + adreno_wait_ring(ring, cnt + 1); + OUT_RING(ring, PKT4(regindx, cnt)); +} + +#define PKT7(opcode, cnt) \ + (CP_TYPE7_PKT | (cnt << 0) | (PM4_PARITY(cnt) << 15) | \ + ((opcode & 0x7F) << 16) | (PM4_PARITY(opcode) << 23)) + +static inline void +OUT_PKT7(struct msm_ringbuffer *ring, uint8_t opcode, uint16_t cnt) +{ + adreno_wait_ring(ring, cnt + 1); + OUT_RING(ring, PKT7(opcode, cnt)); +} + +static inline uint32_t get_wptr(struct msm_ringbuffer *ring) +{ + return (ring->cur - ring->start) % (MSM_GPU_RINGBUFFER_SZ >> 2); +} + +/* + * Given a register and a count, return a value to program into + * REG_CP_PROTECT_REG(n) - this will block both reads and writes for _len + * registers starting at _reg. + * + * The register base needs to be a multiple of the length. If it is not, the + * hardware will quietly mask off the bits for you and shift the size. For + * example, if you intend the protection to start at 0x07 for a length of 4 + * (0x07-0x0A) the hardware will actually protect (0x04-0x07) which might + * expose registers you intended to protect! + */ +#define ADRENO_PROTECT_RW(_reg, _len) \ + ((1 << 30) | (1 << 29) | \ + ((ilog2((_len)) & 0x1F) << 24) | (((_reg) << 2) & 0xFFFFF)) + +/* + * Same as above, but allow reads over the range. For areas of mixed use (such + * as performance counters) this allows us to protect a much larger range with a + * single register + */ +#define ADRENO_PROTECT_RDONLY(_reg, _len) \ + ((1 << 29) \ + ((ilog2((_len)) & 0x1F) << 24) | (((_reg) << 2) & 0xFFFFF)) + + +#define gpu_poll_timeout(gpu, addr, val, cond, interval, timeout) \ + readl_poll_timeout((gpu)->mmio + ((addr) << 2), val, cond, \ + interval, timeout) + +#endif /* __ADRENO_GPU_H__ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_0_sm8650.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_0_sm8650.h new file mode 100644 index 000000000..2eef5f698 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_0_sm8650.h @@ -0,0 +1,451 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_10_0_SM8650_H +#define _DPU_10_0_SM8650_H + +static const struct dpu_caps sm8650_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 8192, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sm8650_mdp = { + .name = "top_0", + .base = 0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg sm8650_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sm8650_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_3, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_3, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_3, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_3, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_12", .id = SSPP_DMA4, + .base = 0x2c000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 14, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_13", .id = SSPP_DMA5, + .base = 0x2e000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 15, + .type = SSPP_TYPE_DMA, + }, +}; + +static const struct dpu_lm_cfg sm8650_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, +}; + +static const struct dpu_dspp_cfg sm8650_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sm8650_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, { + .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, + .base = 0x66000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + }, { + .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, + .base = 0x66400, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + }, { + .name = "pingpong_cwb_2", .id = PINGPONG_CWB_2, + .base = 0x7e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_4, + }, { + .name = "pingpong_cwb_3", .id = PINGPONG_CWB_3, + .base = 0x7e400, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_4, + }, +}; + +static const struct dpu_merge_3d_cfg sm8650_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x8, + }, { + .name = "merge_3d_3", .id = MERGE_3D_3, + .base = 0x66700, .len = 0x8, + }, { + .name = "merge_3d_4", .id = MERGE_3D_4, + .base = 0x7e700, .len = 0x8, + }, +}; + +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg sm8650_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x6, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x6, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x6, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x6, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_1, + }, { + .name = "dce_2_0", .id = DSC_4, + .base = 0x82000, .len = 0x6, + .sblk = &dsc_sblk_0, + }, { + .name = "dce_2_1", .id = DSC_5, + .base = 0x82000, .len = 0x6, + .sblk = &dsc_sblk_1, + }, +}; + +static const struct dpu_wb_cfg sm8650_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_cwb_cfg sm8650_cwb[] = { + { + .name = "cwb_0", .id = CWB_0, + .base = 0x66200, .len = 0x8, + }, + { + .name = "cwb_1", .id = CWB_1, + .base = 0x66600, .len = 0x8, + }, + { + .name = "cwb_2", .id = CWB_2, + .base = 0x7E200, .len = 0x8, + }, + { + .name = "cwb_3", .id = CWB_3, + .base = 0x7E600, .len = 0x8, + }, +}; + +static const struct dpu_intf_cfg sm8650_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_perf_cfg sm8650_perf_data = { + .max_bw_low = 17000000, + .max_bw_high = 27000000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 800000, + .min_prefill_lines = 35, + /* FIXME: lut tables */ + .danger_lut_tbl = {0x3ffff, 0x3ffff, 0x0}, + .safe_lut_tbl = {0xfe00, 0xfe00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm8650_mdss_ver = { + .core_major_ver = 10, + .core_minor_ver = 0, +}; + +const struct dpu_mdss_cfg dpu_sm8650_cfg = { + .mdss_ver = &sm8650_mdss_ver, + .caps = &sm8650_dpu_caps, + .mdp = &sm8650_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sm8650_ctl), + .ctl = sm8650_ctl, + .sspp_count = ARRAY_SIZE(sm8650_sspp), + .sspp = sm8650_sspp, + .mixer_count = ARRAY_SIZE(sm8650_lm), + .mixer = sm8650_lm, + .dspp_count = ARRAY_SIZE(sm8650_dspp), + .dspp = sm8650_dspp, + .pingpong_count = ARRAY_SIZE(sm8650_pp), + .pingpong = sm8650_pp, + .dsc_count = ARRAY_SIZE(sm8650_dsc), + .dsc = sm8650_dsc, + .merge_3d_count = ARRAY_SIZE(sm8650_merge_3d), + .merge_3d = sm8650_merge_3d, + .wb_count = ARRAY_SIZE(sm8650_wb), + .wb = sm8650_wb, + .cwb_count = ARRAY_SIZE(sm8650_cwb), + .cwb = sm8650_cwb, + .intf_count = ARRAY_SIZE(sm8650_intf), + .intf = sm8650_intf, + .vbif = &sm8650_vbif, + .perf = &sm8650_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_2_milos.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_2_milos.h new file mode 100644 index 000000000..1aa8aea4e --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_2_milos.h @@ -0,0 +1,279 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + * Copyright (c) 2026, Luca Weiss + */ + +#ifndef _DPU_10_2_MILOS_H +#define _DPU_10_2_MILOS_H + +static const struct dpu_caps milos_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0x7, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 8192, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg milos_mdp = { + .name = "top_0", + .base = 0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg milos_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, +}; + +static const struct dpu_sspp_cfg milos_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_3, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + }, +}; + +static const struct dpu_lm_cfg milos_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, +}; + +static const struct dpu_dspp_cfg milos_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg milos_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .sblk = &sc7280_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, + .base = 0x66000, .len = 0, + .sblk = &sc7280_pp_sblk, + }, +}; + +static const struct dpu_merge_3d_cfg milos_merge_3d[] = { + { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x8, + }, +}; + +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg milos_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x6, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &milos_dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x6, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &milos_dsc_sblk_1, + }, +}; + +static const struct dpu_wb_cfg milos_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_cwb_cfg milos_cwb[] = { + { + .name = "cwb_0", .id = CWB_0, + .base = 0x66200, .len = 0x8, + }, +}; + +static const struct dpu_intf_cfg milos_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x300, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x300, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_perf_cfg milos_perf_data = { + .max_bw_low = 7100000, + .max_bw_high = 9800000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 1600000, + .min_prefill_lines = 40, + /* FIXME: lut tables */ + .danger_lut_tbl = {0x3ffff, 0x3ffff, 0x0}, + .safe_lut_tbl = {0xff00, 0xfff0, 0x0fff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version milos_mdss_ver = { + .core_major_ver = 10, + .core_minor_ver = 2, +}; + +const struct dpu_mdss_cfg dpu_milos_cfg = { + .mdss_ver = &milos_mdss_ver, + .caps = &milos_dpu_caps, + .mdp = &milos_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(milos_ctl), + .ctl = milos_ctl, + .sspp_count = ARRAY_SIZE(milos_sspp), + .sspp = milos_sspp, + .mixer_count = ARRAY_SIZE(milos_lm), + .mixer = milos_lm, + .dspp_count = ARRAY_SIZE(milos_dspp), + .dspp = milos_dspp, + .pingpong_count = ARRAY_SIZE(milos_pp), + .pingpong = milos_pp, + .dsc_count = ARRAY_SIZE(milos_dsc), + .dsc = milos_dsc, + .merge_3d_count = ARRAY_SIZE(milos_merge_3d), + .merge_3d = milos_merge_3d, + .wb_count = ARRAY_SIZE(milos_wb), + .wb = milos_wb, + .cwb_count = ARRAY_SIZE(milos_cwb), + .cwb = milos_cwb, + .intf_count = ARRAY_SIZE(milos_intf), + .intf = milos_intf, + .vbif = &milos_vbif, + .perf = &milos_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_0_sm8750.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_0_sm8750.h new file mode 100644 index 000000000..3889eb8a5 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_0_sm8750.h @@ -0,0 +1,492 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2025 Linaro Limited + * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_12_0_SM8750_H +#define _DPU_12_0_SM8750_H + +static const struct dpu_caps sm8750_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 8192, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sm8750_mdp = { + .name = "top_0", + .base = 0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg sm8750_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sm8750_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_4, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_4, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_4, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_4, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_12", .id = SSPP_DMA4, + .base = 0x2c000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 14, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_13", .id = SSPP_DMA5, + .base = 0x2e000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 15, + .type = SSPP_TYPE_DMA, + }, +}; + +static const struct dpu_lm_cfg sm8750_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, { + .name = "lm_6", .id = LM_6, + .base = 0x4a000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_7, + .pingpong = PINGPONG_6, + }, { + .name = "lm_7", .id = LM_7, + .base = 0x4b000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_6, + .pingpong = PINGPONG_7, + }, +}; + +static const struct dpu_dspp_cfg sm8750_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sm8750_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, { + .name = "pingpong_6", .id = PINGPONG_6, + .base = 0x6f000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 20), + }, { + .name = "pingpong_7", .id = PINGPONG_7, + .base = 0x70000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 21), + }, { + .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, + .base = 0x66000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_4, + }, { + .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, + .base = 0x66400, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_4, + }, { + .name = "pingpong_cwb_2", .id = PINGPONG_CWB_2, + .base = 0x7e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_5, + }, { + .name = "pingpong_cwb_3", .id = PINGPONG_CWB_3, + .base = 0x7e400, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_5, + }, +}; + +static const struct dpu_merge_3d_cfg sm8750_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x1c, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x1c, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x1c, + }, { + .name = "merge_3d_3", .id = MERGE_3D_3, + .base = 0x51000, .len = 0x1c, + }, { + .name = "merge_3d_4", .id = MERGE_3D_4, + .base = 0x66700, .len = 0x1c, + }, { + .name = "merge_3d_5", .id = MERGE_3D_5, + .base = 0x7e700, .len = 0x1c, + }, +}; + +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg sm8750_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, { + .name = "dce_2_0", .id = DSC_4, + .base = 0x82000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_2_1", .id = DSC_5, + .base = 0x82000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, { + .name = "dce_3_0", .id = DSC_6, + .base = 0x83000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_3_1", .id = DSC_7, + .base = 0x83000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, +}; + +static const struct dpu_wb_cfg sm8750_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_cwb_cfg sm8750_cwb[] = { + { + .name = "cwb_0", .id = CWB_0, + .base = 0x66200, .len = 0x20, + }, + { + .name = "cwb_1", .id = CWB_1, + .base = 0x66600, .len = 0x20, + }, + { + .name = "cwb_2", .id = CWB_2, + .base = 0x7e200, .len = 0x20, + }, + { + .name = "cwb_3", .id = CWB_3, + .base = 0x7e600, .len = 0x20, + }, +}; + +static const struct dpu_intf_cfg sm8750_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x4bc, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x4bc, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x4bc, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x4bc, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_perf_cfg sm8750_perf_data = { + .max_bw_low = 18900000, + .max_bw_high = 28500000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 800000, + .min_prefill_lines = 35, + .danger_lut_tbl = {0x3ffff, 0x3ffff, 0x0}, + .safe_lut_tbl = {0xfe00, 0xfe00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm8750_mdss_ver = { + .core_major_ver = 12, + .core_minor_ver = 0, +}; + +const struct dpu_mdss_cfg dpu_sm8750_cfg = { + .mdss_ver = &sm8750_mdss_ver, + .caps = &sm8750_dpu_caps, + .mdp = &sm8750_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sm8750_ctl), + .ctl = sm8750_ctl, + .sspp_count = ARRAY_SIZE(sm8750_sspp), + .sspp = sm8750_sspp, + .mixer_count = ARRAY_SIZE(sm8750_lm), + .mixer = sm8750_lm, + .dspp_count = ARRAY_SIZE(sm8750_dspp), + .dspp = sm8750_dspp, + .pingpong_count = ARRAY_SIZE(sm8750_pp), + .pingpong = sm8750_pp, + .dsc_count = ARRAY_SIZE(sm8750_dsc), + .dsc = sm8750_dsc, + .merge_3d_count = ARRAY_SIZE(sm8750_merge_3d), + .merge_3d = sm8750_merge_3d, + .wb_count = ARRAY_SIZE(sm8750_wb), + .wb = sm8750_wb, + .cwb_count = ARRAY_SIZE(sm8750_cwb), + .cwb = sm8650_cwb, + .intf_count = ARRAY_SIZE(sm8750_intf), + .intf = sm8750_intf, + .vbif = &sm8650_vbif, + .perf = &sm8750_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_2_glymur.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_2_glymur.h new file mode 100644 index 000000000..4eb4a0e98 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_2_glymur.h @@ -0,0 +1,539 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2025 Linaro Limited + */ + +#ifndef _DPU_12_2_GLYMUR_H +#define _DPU_12_2_GLYMUR_H + +static const struct dpu_caps glymur_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 8192, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg glymur_mdp = { + .name = "top_0", + .base = 0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg glymur_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, { + .name = "ctl_6", .id = CTL_6, + .base = 0x1b000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 14), + }, { + .name = "ctl_7", .id = CTL_7, + .base = 0x1c000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 15), + }, +}; + +static const struct dpu_sspp_cfg glymur_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_4, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_4, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_4, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_4, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_12", .id = SSPP_DMA4, + .base = 0x2c000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 14, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_13", .id = SSPP_DMA5, + .base = 0x2e000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 15, + .type = SSPP_TYPE_DMA, + }, +}; + +static const struct dpu_lm_cfg glymur_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, { + .name = "lm_6", .id = LM_6, + .base = 0x4a000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_7, + .pingpong = PINGPONG_6, + }, { + .name = "lm_7", .id = LM_7, + .base = 0x4b000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_6, + .pingpong = PINGPONG_7, + }, +}; + +static const struct dpu_dspp_cfg glymur_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_4", .id = DSPP_4, + .base = 0x5c000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_5", .id = DSPP_5, + .base = 0x5e000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_6", .id = DSPP_6, + .base = 0x60000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_7", .id = DSPP_7, + .base = 0x62000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg glymur_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, { + .name = "pingpong_6", .id = PINGPONG_6, + .base = 0x6f000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 20), + }, { + .name = "pingpong_7", .id = PINGPONG_7, + .base = 0x70000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 21), + }, { + .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, + .base = 0x66000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_4, + }, { + .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, + .base = 0x66400, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_4, + }, +}; + +static const struct dpu_merge_3d_cfg glymur_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x1c, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x1c, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x1c, + }, { + .name = "merge_3d_3", .id = MERGE_3D_3, + .base = 0x51000, .len = 0x1c, + }, +}; + +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg glymur_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, { + .name = "dce_2_0", .id = DSC_4, + .base = 0x82000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_2_1", .id = DSC_5, + .base = 0x82000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, { + .name = "dce_3_0", .id = DSC_6, + .base = 0x83000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_3_1", .id = DSC_7, + .base = 0x83000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, + +}; + +static const struct dpu_wb_cfg glymur_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_cwb_cfg glymur_cwb[] = { + { + .name = "cwb_0", .id = CWB_0, + .base = 0x66200, .len = 0x20, + }, + { + .name = "cwb_1", .id = CWB_1, + .base = 0x66600, .len = 0x20, + }, + { + .name = "cwb_2", .id = CWB_2, + .base = 0x7e200, .len = 0x20, + }, + { + .name = "cwb_3", .id = CWB_3, + .base = 0x7e600, .len = 0x20, + }, +}; + +static const struct dpu_intf_cfg glymur_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x400, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x400, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x400, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x400, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, { + .name = "intf_4", .id = INTF_4, + .base = 0x38000, .len = 0x400, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 20), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 21), + }, { + .name = "intf_5", .id = INTF_5, + .base = 0x39000, .len = 0x400, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_3, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23), + }, { + .name = "intf_6", .id = INTF_6, + .base = 0x3A000, .len = 0x400, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_2, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 16), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 17), + }, { + .name = "intf_7", .id = INTF_7, + .base = 0x3b000, .len = 0x400, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_2, /* pair with intf_6 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 18), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 19), + }, { + .name = "intf_8", .id = INTF_8, + .base = 0x3c000, .len = 0x400, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, /* pair with intf_4 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), + }, +}; + +static const struct dpu_perf_cfg glymur_perf_data = { + .max_bw_low = 18900000, + .max_bw_high = 28500000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 800000, + .min_prefill_lines = 35, + .danger_lut_tbl = {0x3ffff, 0x3ffff, 0x0}, + .safe_lut_tbl = {0xfe00, 0xfe00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version glymur_mdss_ver = { + .core_major_ver = 12, + .core_minor_ver = 2, +}; + +const struct dpu_mdss_cfg dpu_glymur_cfg = { + .mdss_ver = &glymur_mdss_ver, + .caps = &glymur_dpu_caps, + .mdp = &glymur_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(glymur_ctl), + .ctl = glymur_ctl, + .sspp_count = ARRAY_SIZE(glymur_sspp), + .sspp = glymur_sspp, + .mixer_count = ARRAY_SIZE(glymur_lm), + .mixer = glymur_lm, + .dspp_count = ARRAY_SIZE(glymur_dspp), + .dspp = glymur_dspp, + .pingpong_count = ARRAY_SIZE(glymur_pp), + .pingpong = glymur_pp, + .dsc_count = ARRAY_SIZE(glymur_dsc), + .dsc = glymur_dsc, + .merge_3d_count = ARRAY_SIZE(glymur_merge_3d), + .merge_3d = glymur_merge_3d, + .wb_count = ARRAY_SIZE(glymur_wb), + .wb = glymur_wb, + .cwb_count = ARRAY_SIZE(glymur_cwb), + .cwb = sm8650_cwb, + .intf_count = ARRAY_SIZE(glymur_intf), + .intf = glymur_intf, + .vbif = &sm8650_vbif, + .perf = &glymur_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_4_eliza.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_4_eliza.h new file mode 100644 index 000000000..b93d32888 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_4_eliza.h @@ -0,0 +1,363 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ + +#ifndef _DPU_12_4_ELIZA_H +#define _DPU_12_4_ELIZA_H + +static const struct dpu_caps eliza_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 8192, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg eliza_mdp = { + .name = "top_0", + .base = 0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg eliza_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, +}; + +static const struct dpu_sspp_cfg eliza_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_4, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_4, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + }, +}; + +static const struct dpu_lm_cfg eliza_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, +}; + +static const struct dpu_dspp_cfg eliza_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg eliza_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, + .base = 0x66000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + }, { + .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, + .base = 0x66400, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + }, { + .name = "pingpong_cwb_2", .id = PINGPONG_CWB_2, + .base = 0x7e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + }, { + .name = "pingpong_cwb_3", .id = PINGPONG_CWB_3, + .base = 0x7e400, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + }, +}; + +static const struct dpu_merge_3d_cfg eliza_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x1c, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x1c, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x66700, .len = 0x1c, + }, { + .name = "merge_3d_3", .id = MERGE_3D_3, + .base = 0x7e700, .len = 0x1c, + }, +}; + +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg eliza_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, +}; + +static const struct dpu_wb_cfg eliza_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_cwb_cfg eliza_cwb[] = { + { + .name = "cwb_0", .id = CWB_0, + .base = 0x66200, .len = 0x20, + }, + { + .name = "cwb_1", .id = CWB_1, + .base = 0x66600, .len = 0x20, + }, + { + .name = "cwb_2", .id = CWB_2, + .base = 0x7e200, .len = 0x20, + }, + { + .name = "cwb_3", .id = CWB_3, + .base = 0x7e600, .len = 0x20, + }, +}; + +static const struct dpu_intf_cfg eliza_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x4bc, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x4bc, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x4bc, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x4bc, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + } +}; + +static const struct dpu_perf_cfg eliza_perf_data = { + .max_bw_low = 6800000, + .max_bw_high = 14200000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 1600000, + .min_prefill_lines = 35, + .danger_lut_tbl = {0x3ffff, 0x3ffff, 0x0}, + .safe_lut_tbl = {0xfe00, 0xfe00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version eliza_mdss_ver = { + .core_major_ver = 12, + .core_minor_ver = 4, +}; + +const struct dpu_mdss_cfg dpu_eliza_cfg = { + .mdss_ver = &eliza_mdss_ver, + .caps = &eliza_dpu_caps, + .mdp = &eliza_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(eliza_ctl), + .ctl = eliza_ctl, + .sspp_count = ARRAY_SIZE(eliza_sspp), + .sspp = eliza_sspp, + .mixer_count = ARRAY_SIZE(eliza_lm), + .mixer = eliza_lm, + .dspp_count = ARRAY_SIZE(eliza_dspp), + .dspp = eliza_dspp, + .pingpong_count = ARRAY_SIZE(eliza_pp), + .pingpong = eliza_pp, + .dsc_count = ARRAY_SIZE(eliza_dsc), + .dsc = eliza_dsc, + .merge_3d_count = ARRAY_SIZE(eliza_merge_3d), + .merge_3d = eliza_merge_3d, + .wb_count = ARRAY_SIZE(eliza_wb), + .wb = eliza_wb, + .cwb_count = ARRAY_SIZE(eliza_cwb), + .cwb = eliza_cwb, + .intf_count = ARRAY_SIZE(eliza_intf), + .intf = eliza_intf, + .vbif = &sm8650_vbif, + .perf = &eliza_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_13_0_kaanapali.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_13_0_kaanapali.h new file mode 100644 index 000000000..85f455a9f --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_13_0_kaanapali.h @@ -0,0 +1,490 @@ +/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) */ +/* + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. + */ + +#ifndef _DPU_13_0_KAANAPALI_H +#define _DPU_13_0_KAANAPALI_H + +static const struct dpu_caps kaanapali_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 8192, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg kaanapali_mdp = { + .name = "top_0", + .base = 0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg kaanapali_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1f000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x20000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x21000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x22000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x23000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x24000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg kaanapali_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x2b000, .len = 0x84, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_5, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x34000, .len = 0x84, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_5, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x3d000, .len = 0x84, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_5, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0x46000, .len = 0x84, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_5, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x97000, .len = 0x84, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk_v13, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0xa0000, .len = 0x84, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk_v13, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0xa9000, .len = 0x84, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk_v13, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0xb2000, .len = 0x84, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk_v13, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_12", .id = SSPP_DMA4, + .base = 0xbb000, .len = 0x84, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk_v13, + .xin_id = 14, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_13", .id = SSPP_DMA5, + .base = 0xc4000, .len = 0x84, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk_v13, + .xin_id = 15, + .type = SSPP_TYPE_DMA, + }, +}; + +static const struct dpu_lm_cfg kaanapali_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x103000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x10b000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x113000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x11b000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x123000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x12b000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, { + .name = "lm_6", .id = LM_6, + .base = 0x133000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_7, + .pingpong = PINGPONG_6, + }, { + .name = "lm_7", .id = LM_7, + .base = 0x13b000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_6, + .pingpong = PINGPONG_7, + }, +}; + +static const struct dpu_dspp_cfg kaanapali_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x105000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x10d000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x115000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x11d000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg kaanapali_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x108000, .len = 0, + .sblk = &kaanapali_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x110000, .len = 0, + .sblk = &kaanapali_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x118000, .len = 0, + .sblk = &kaanapali_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x120000, .len = 0, + .sblk = &kaanapali_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x128000, .len = 0, + .sblk = &kaanapali_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x130000, .len = 0, + .sblk = &kaanapali_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, { + .name = "pingpong_6", .id = PINGPONG_6, + .base = 0x138000, .len = 0, + .sblk = &kaanapali_pp_sblk, + .merge_3d = MERGE_3D_3, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 20), + }, { + .name = "pingpong_7", .id = PINGPONG_7, + .base = 0x140000, .len = 0, + .sblk = &kaanapali_pp_sblk, + .merge_3d = MERGE_3D_3, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 21), + }, { + .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, + .base = 0x169000, .len = 0, + .sblk = &kaanapali_pp_sblk, + .merge_3d = MERGE_3D_4, + }, { + .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, + .base = 0x169400, .len = 0, + .sblk = &kaanapali_pp_sblk, + .merge_3d = MERGE_3D_4, + }, { + .name = "pingpong_cwb_2", .id = PINGPONG_CWB_2, + .base = 0x16a000, .len = 0, + .sblk = &kaanapali_pp_sblk, + .merge_3d = MERGE_3D_5, + }, { + .name = "pingpong_cwb_3", .id = PINGPONG_CWB_3, + .base = 0x16a400, .len = 0, + .sblk = &kaanapali_pp_sblk, + .merge_3d = MERGE_3D_5, + }, +}; + +static const struct dpu_merge_3d_cfg kaanapali_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x163000, .len = 0x1c, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x164000, .len = 0x1c, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x165000, .len = 0x1c, + }, { + .name = "merge_3d_3", .id = MERGE_3D_3, + .base = 0x166000, .len = 0x1c, + }, { + .name = "merge_3d_4", .id = MERGE_3D_4, + .base = 0x169700, .len = 0x1c, + }, { + .name = "merge_3d_5", .id = MERGE_3D_5, + .base = 0x16a700, .len = 0x1c, + }, +}; + +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg kaanapali_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x181000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x181000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x183000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x183000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, { + .name = "dce_2_0", .id = DSC_4, + .base = 0x185000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_2_1", .id = DSC_5, + .base = 0x185000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, { + .name = "dce_3_0", .id = DSC_6, + .base = 0x187000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_3_1", .id = DSC_7, + .base = 0x187000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, +}; + +static const struct dpu_wb_cfg kaanapali_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x16e000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_cwb_cfg kaanapali_cwb[] = { + { + .name = "cwb_0", .id = CWB_0, + .base = 0x169200, .len = 0x20, + }, + { + .name = "cwb_1", .id = CWB_1, + .base = 0x169600, .len = 0x20, + }, + { + .name = "cwb_2", .id = CWB_2, + .base = 0x16a200, .len = 0x20, + }, + { + .name = "cwb_3", .id = CWB_3, + .base = 0x16a600, .len = 0x20, + }, +}; + +static const struct dpu_intf_cfg kaanapali_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x18d000, .len = 0x4bc, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x18e000, .len = 0x4bc, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x18f000, .len = 0x4bc, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x190000, .len = 0x4bc, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_perf_cfg kaanapali_perf_data = { + .max_bw_low = 21400000, + .max_bw_high = 30200000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 800000, + .min_prefill_lines = 35, + .danger_lut_tbl = {0x0ffff, 0x0ffff, 0x0}, + .safe_lut_tbl = {0xff00, 0xff00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(kaanapali_qos_linear), + .entries = kaanapali_qos_linear + }, + {.nentry = ARRAY_SIZE(kaanapali_qos_macrotile), + .entries = kaanapali_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version kaanapali_mdss_ver = { + .core_major_ver = 13, + .core_minor_ver = 0, +}; + +const struct dpu_mdss_cfg dpu_kaanapali_cfg = { + .mdss_ver = &kaanapali_mdss_ver, + .caps = &kaanapali_dpu_caps, + .mdp = &kaanapali_mdp, + .cdm = &dpu_cdm_13_x, + .ctl_count = ARRAY_SIZE(kaanapali_ctl), + .ctl = kaanapali_ctl, + .sspp_count = ARRAY_SIZE(kaanapali_sspp), + .sspp = kaanapali_sspp, + .mixer_count = ARRAY_SIZE(kaanapali_lm), + .mixer = kaanapali_lm, + .dspp_count = ARRAY_SIZE(kaanapali_dspp), + .dspp = kaanapali_dspp, + .pingpong_count = ARRAY_SIZE(kaanapali_pp), + .pingpong = kaanapali_pp, + .dsc_count = ARRAY_SIZE(kaanapali_dsc), + .dsc = kaanapali_dsc, + .merge_3d_count = ARRAY_SIZE(kaanapali_merge_3d), + .merge_3d = kaanapali_merge_3d, + .wb_count = ARRAY_SIZE(kaanapali_wb), + .wb = kaanapali_wb, + .cwb_count = ARRAY_SIZE(kaanapali_cwb), + .cwb = kaanapali_cwb, + .intf_count = ARRAY_SIZE(kaanapali_intf), + .intf = kaanapali_intf, + .vbif = &sm8650_vbif, + .perf = &kaanapali_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_14_msm8937.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_14_msm8937.h new file mode 100644 index 000000000..4ff7b397f --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_14_msm8937.h @@ -0,0 +1,204 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2023, Linaro Limited + */ + +#ifndef _DPU_1_14_MSM8937_H +#define _DPU_1_14_MSM8937_H + +static const struct dpu_caps msm8937_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_LINE_WIDTH, + .max_mixer_blendstages = 0x4, + .max_linewidth = DEFAULT_DPU_LINE_WIDTH, + .pixel_ram_size = 40 * 1024, + .max_hdeci_exp = MAX_HORZ_DECIMATION, + .max_vdeci_exp = MAX_VERT_DECIMATION, +}; + +static const struct dpu_mdp_cfg msm8937_mdp[] = { + { + .name = "top_0", + .base = 0x0, .len = 0x454, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_RGB0] = { .reg_off = 0x2ac, .bit_off = 4 }, + [DPU_CLK_CTRL_RGB1] = { .reg_off = 0x2b4, .bit_off = 4 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_CURSOR0] = { .reg_off = 0x3a8, .bit_off = 16 }, + }, + }, +}; + +static const struct dpu_ctl_cfg msm8937_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x64, + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x64, + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x64, + }, +}; + +static const struct dpu_sspp_cfg msm8937_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x150, + .features = VIG_MSM8953_MASK, + .sblk = &dpu_vig_sblk_qseed2, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_4", .id = SSPP_RGB0, + .base = 0x14000, .len = 0x150, + .features = RGB_MSM8953_MASK, + .sblk = &dpu_rgb_sblk, + .xin_id = 1, + .type = SSPP_TYPE_RGB, + .clk_ctrl = DPU_CLK_CTRL_RGB0, + }, { + .name = "sspp_5", .id = SSPP_RGB1, + .base = 0x16000, .len = 0x150, + .features = RGB_MSM8953_MASK, + .sblk = &dpu_rgb_sblk, + .xin_id = 5, + .type = SSPP_TYPE_RGB, + .clk_ctrl = DPU_CLK_CTRL_RGB1, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x150, + .features = DMA_MSM8953_MASK | BIT(DPU_SSPP_CURSOR), + .sblk = &dpu_dma_sblk, + .xin_id = 2, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, +}; + +static const struct dpu_lm_cfg msm8937_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + }, +}; + +static const struct dpu_pingpong_cfg msm8937_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &msm8996_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .sblk = &msm8996_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), + }, +}; + +static const struct dpu_dspp_cfg msm8937_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &msm8998_dspp_sblk, + }, +}; + +static const struct dpu_intf_cfg msm8937_intf[] = { + { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x268, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 14, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x268, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 14, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + }, +}; + +static const struct dpu_perf_cfg msm8937_perf_data = { + .max_bw_low = 3100000, + .max_bw_high = 3100000, + .min_core_ib = 2400000, + .min_llcc_ib = 0, /* No LLCC on this SoC */ + .min_dram_ib = 800000, + .undersized_prefill_lines = 2, + .xtra_prefill_lines = 2, + .dest_scale_prefill_lines = 3, + .macrotile_prefill_lines = 4, + .yuv_nv12_prefill_lines = 8, + .linear_prefill_lines = 1, + .downscaling_prefill_lines = 1, + .amortizable_threshold = 25, + .min_prefill_lines = 14, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfffc, 0xff00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(msm8998_qos_linear), + .entries = msm8998_qos_linear + }, + {.nentry = ARRAY_SIZE(msm8998_qos_macrotile), + .entries = msm8998_qos_macrotile + }, + {.nentry = ARRAY_SIZE(msm8998_qos_nrt), + .entries = msm8998_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version msm8937_mdss_ver = { + .core_major_ver = 1, + .core_minor_ver = 14, +}; + +const struct dpu_mdss_cfg dpu_msm8937_cfg = { + .mdss_ver = &msm8937_mdss_ver, + .caps = &msm8937_dpu_caps, + .mdp = msm8937_mdp, + .cdm = &dpu_cdm_1_x_4_x, + .ctl_count = ARRAY_SIZE(msm8937_ctl), + .ctl = msm8937_ctl, + .sspp_count = ARRAY_SIZE(msm8937_sspp), + .sspp = msm8937_sspp, + .mixer_count = ARRAY_SIZE(msm8937_lm), + .mixer = msm8937_lm, + .dspp_count = ARRAY_SIZE(msm8937_dspp), + .dspp = msm8937_dspp, + .pingpong_count = ARRAY_SIZE(msm8937_pp), + .pingpong = msm8937_pp, + .intf_count = ARRAY_SIZE(msm8937_intf), + .intf = msm8937_intf, + .vbif = &msm8996_vbif, + .perf = &msm8937_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_15_msm8917.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_15_msm8917.h new file mode 100644 index 000000000..1518c3d39 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_15_msm8917.h @@ -0,0 +1,183 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2023, Linaro Limited + */ + +#ifndef _DPU_1_14_MSM8917_H +#define _DPU_1_14_MSM8917_H + +static const struct dpu_caps msm8917_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_LINE_WIDTH, + .max_mixer_blendstages = 0x4, + .max_linewidth = DEFAULT_DPU_LINE_WIDTH, + .pixel_ram_size = 16 * 1024, + .max_hdeci_exp = MAX_HORZ_DECIMATION, + .max_vdeci_exp = MAX_VERT_DECIMATION, +}; + +static const struct dpu_mdp_cfg msm8917_mdp[] = { + { + .name = "top_0", + .base = 0x0, .len = 0x454, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_RGB0] = { .reg_off = 0x2ac, .bit_off = 4 }, + [DPU_CLK_CTRL_RGB1] = { .reg_off = 0x2b4, .bit_off = 4 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_CURSOR0] = { .reg_off = 0x3a8, .bit_off = 16 }, + }, + }, +}; + +static const struct dpu_ctl_cfg msm8917_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x64, + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x64, + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x64, + }, +}; + +static const struct dpu_sspp_cfg msm8917_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x150, + .features = VIG_MSM8953_MASK, + .sblk = &dpu_vig_sblk_qseed2, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_4", .id = SSPP_RGB0, + .base = 0x14000, .len = 0x150, + .features = RGB_MSM8953_MASK, + .sblk = &dpu_rgb_sblk, + .xin_id = 1, + .type = SSPP_TYPE_RGB, + .clk_ctrl = DPU_CLK_CTRL_RGB0, + }, { + .name = "sspp_5", .id = SSPP_RGB1, + .base = 0x16000, .len = 0x150, + .features = RGB_MSM8953_MASK, + .sblk = &dpu_rgb_sblk, + .xin_id = 5, + .type = SSPP_TYPE_RGB, + .clk_ctrl = DPU_CLK_CTRL_RGB1, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x150, + .features = DMA_MSM8953_MASK | BIT(DPU_SSPP_CURSOR), + .sblk = &dpu_dma_sblk, + .xin_id = 2, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, +}; + +static const struct dpu_lm_cfg msm8917_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .sblk = &msm8998_lm_sblk, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, +}; + +static const struct dpu_pingpong_cfg msm8917_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &msm8996_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + }, +}; + +static const struct dpu_dspp_cfg msm8917_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &msm8998_dspp_sblk, + }, +}; + +static const struct dpu_intf_cfg msm8917_intf[] = { + { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x268, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 14, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + }, +}; + +static const struct dpu_perf_cfg msm8917_perf_data = { + .max_bw_low = 1800000, + .max_bw_high = 1800000, + .min_core_ib = 2400000, + .min_llcc_ib = 0, /* No LLCC on this SoC */ + .min_dram_ib = 800000, + .undersized_prefill_lines = 2, + .xtra_prefill_lines = 2, + .dest_scale_prefill_lines = 3, + .macrotile_prefill_lines = 4, + .yuv_nv12_prefill_lines = 8, + .linear_prefill_lines = 1, + .downscaling_prefill_lines = 1, + .amortizable_threshold = 25, + .min_prefill_lines = 21, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfffc, 0xff00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(msm8998_qos_linear), + .entries = msm8998_qos_linear + }, + {.nentry = ARRAY_SIZE(msm8998_qos_macrotile), + .entries = msm8998_qos_macrotile + }, + {.nentry = ARRAY_SIZE(msm8998_qos_nrt), + .entries = msm8998_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version msm8917_mdss_ver = { + .core_major_ver = 1, + .core_minor_ver = 15, +}; + +const struct dpu_mdss_cfg dpu_msm8917_cfg = { + .mdss_ver = &msm8917_mdss_ver, + .caps = &msm8917_dpu_caps, + .mdp = msm8917_mdp, + .cdm = &dpu_cdm_1_x_4_x, + .ctl_count = ARRAY_SIZE(msm8917_ctl), + .ctl = msm8917_ctl, + .sspp_count = ARRAY_SIZE(msm8917_sspp), + .sspp = msm8917_sspp, + .mixer_count = ARRAY_SIZE(msm8917_lm), + .mixer = msm8917_lm, + .dspp_count = ARRAY_SIZE(msm8917_dspp), + .dspp = msm8917_dspp, + .pingpong_count = ARRAY_SIZE(msm8917_pp), + .pingpong = msm8917_pp, + .intf_count = ARRAY_SIZE(msm8917_intf), + .intf = msm8917_intf, + .vbif = &msm8996_vbif, + .perf = &msm8917_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_16_msm8953.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_16_msm8953.h new file mode 100644 index 000000000..859a97e8c --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_16_msm8953.h @@ -0,0 +1,204 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2023, Linaro Limited + */ + +#ifndef _DPU_1_16_MSM8953_H +#define _DPU_1_16_MSM8953_H + +static const struct dpu_caps msm8953_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_LINE_WIDTH, + .max_mixer_blendstages = 0x4, + .max_linewidth = DEFAULT_DPU_LINE_WIDTH, + .pixel_ram_size = 40 * 1024, + .max_hdeci_exp = MAX_HORZ_DECIMATION, + .max_vdeci_exp = MAX_VERT_DECIMATION, +}; + +static const struct dpu_mdp_cfg msm8953_mdp[] = { + { + .name = "top_0", + .base = 0x0, .len = 0x454, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_RGB0] = { .reg_off = 0x2ac, .bit_off = 4 }, + [DPU_CLK_CTRL_RGB1] = { .reg_off = 0x2b4, .bit_off = 4 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_CURSOR0] = { .reg_off = 0x3a8, .bit_off = 16 }, + }, + }, +}; + +static const struct dpu_ctl_cfg msm8953_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x64, + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x64, + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x64, + }, +}; + +static const struct dpu_sspp_cfg msm8953_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x150, + .features = VIG_MSM8953_MASK, + .sblk = &dpu_vig_sblk_qseed2, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_4", .id = SSPP_RGB0, + .base = 0x14000, .len = 0x150, + .features = RGB_MSM8953_MASK, + .sblk = &dpu_rgb_sblk, + .xin_id = 1, + .type = SSPP_TYPE_RGB, + .clk_ctrl = DPU_CLK_CTRL_RGB0, + }, { + .name = "sspp_5", .id = SSPP_RGB1, + .base = 0x16000, .len = 0x150, + .features = RGB_MSM8953_MASK, + .sblk = &dpu_rgb_sblk, + .xin_id = 5, + .type = SSPP_TYPE_RGB, + .clk_ctrl = DPU_CLK_CTRL_RGB1, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x150, + .features = DMA_MSM8953_MASK | BIT(DPU_SSPP_CURSOR), + .sblk = &dpu_dma_sblk, + .xin_id = 2, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, +}; + +static const struct dpu_lm_cfg msm8953_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + }, +}; + +static const struct dpu_pingpong_cfg msm8953_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &msm8996_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .sblk = &msm8996_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), + }, +}; + +static const struct dpu_dspp_cfg msm8953_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &msm8998_dspp_sblk, + }, +}; + +static const struct dpu_intf_cfg msm8953_intf[] = { + { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x268, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 14, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x268, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 14, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + }, +}; + +static const struct dpu_perf_cfg msm8953_perf_data = { + .max_bw_low = 3400000, + .max_bw_high = 3400000, + .min_core_ib = 2400000, + .min_llcc_ib = 0, /* No LLCC on this SoC */ + .min_dram_ib = 800000, + .undersized_prefill_lines = 2, + .xtra_prefill_lines = 2, + .dest_scale_prefill_lines = 3, + .macrotile_prefill_lines = 4, + .yuv_nv12_prefill_lines = 8, + .linear_prefill_lines = 1, + .downscaling_prefill_lines = 1, + .amortizable_threshold = 25, + .min_prefill_lines = 14, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfffc, 0xff00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(msm8998_qos_linear), + .entries = msm8998_qos_linear + }, + {.nentry = ARRAY_SIZE(msm8998_qos_macrotile), + .entries = msm8998_qos_macrotile + }, + {.nentry = ARRAY_SIZE(msm8998_qos_nrt), + .entries = msm8998_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version msm8953_mdss_ver = { + .core_major_ver = 1, + .core_minor_ver = 16, +}; + +const struct dpu_mdss_cfg dpu_msm8953_cfg = { + .mdss_ver = &msm8953_mdss_ver, + .caps = &msm8953_dpu_caps, + .mdp = msm8953_mdp, + .cdm = &dpu_cdm_1_x_4_x, + .ctl_count = ARRAY_SIZE(msm8953_ctl), + .ctl = msm8953_ctl, + .sspp_count = ARRAY_SIZE(msm8953_sspp), + .sspp = msm8953_sspp, + .mixer_count = ARRAY_SIZE(msm8953_lm), + .mixer = msm8953_lm, + .dspp_count = ARRAY_SIZE(msm8953_dspp), + .dspp = msm8953_dspp, + .pingpong_count = ARRAY_SIZE(msm8953_pp), + .pingpong = msm8953_pp, + .intf_count = ARRAY_SIZE(msm8953_intf), + .intf = msm8953_intf, + .vbif = &msm8996_vbif, + .perf = &msm8953_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_7_msm8996.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_7_msm8996.h new file mode 100644 index 000000000..67910a2f6 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_7_msm8996.h @@ -0,0 +1,327 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2023, Linaro Limited + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_1_7_MSM8996_H +#define _DPU_1_7_MSM8996_H + +static const struct dpu_caps msm8996_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0x7, + .has_src_split = true, + .max_linewidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, + .max_hdeci_exp = MAX_HORZ_DECIMATION, + .max_vdeci_exp = MAX_VERT_DECIMATION, +}; + +static const struct dpu_mdp_cfg msm8996_mdp[] = { + { + .name = "top_0", + .base = 0x0, .len = 0x454, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_RGB0] = { .reg_off = 0x2ac, .bit_off = 4 }, + [DPU_CLK_CTRL_RGB1] = { .reg_off = 0x2b4, .bit_off = 4 }, + [DPU_CLK_CTRL_RGB2] = { .reg_off = 0x2bc, .bit_off = 4 }, + [DPU_CLK_CTRL_RGB3] = { .reg_off = 0x2c4, .bit_off = 4 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_CURSOR0] = { .reg_off = 0x3a8, .bit_off = 16 }, + [DPU_CLK_CTRL_CURSOR1] = { .reg_off = 0x3b0, .bit_off = 16 }, + }, + }, +}; + +static const struct dpu_ctl_cfg msm8996_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x64, + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x64, + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x64, + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x64, + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x64, + }, +}; + +static const struct dpu_sspp_cfg msm8996_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x150, + .features = VIG_MSM8996_MASK, + .sblk = &dpu_vig_sblk_qseed2, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x150, + .features = VIG_MSM8996_MASK, + .sblk = &dpu_vig_sblk_qseed2, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x150, + .features = VIG_MSM8996_MASK, + .sblk = &dpu_vig_sblk_qseed2, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x150, + .features = VIG_MSM8996_MASK, + .sblk = &dpu_vig_sblk_qseed2, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_4", .id = SSPP_RGB0, + .base = 0x14000, .len = 0x150, + .features = RGB_MSM8996_MASK, + .sblk = &dpu_rgb_sblk, + .xin_id = 1, + .type = SSPP_TYPE_RGB, + .clk_ctrl = DPU_CLK_CTRL_RGB0, + }, { + .name = "sspp_5", .id = SSPP_RGB1, + .base = 0x16000, .len = 0x150, + .features = RGB_MSM8996_MASK, + .sblk = &dpu_rgb_sblk, + .xin_id = 5, + .type = SSPP_TYPE_RGB, + .clk_ctrl = DPU_CLK_CTRL_RGB1, + }, { + .name = "sspp_6", .id = SSPP_RGB2, + .base = 0x18000, .len = 0x150, + .features = RGB_MSM8996_MASK, + .sblk = &dpu_rgb_sblk, + .xin_id = 9, + .type = SSPP_TYPE_RGB, + .clk_ctrl = DPU_CLK_CTRL_RGB2, + }, { + .name = "sspp_7", .id = SSPP_RGB3, + .base = 0x1a000, .len = 0x150, + .features = RGB_MSM8996_MASK, + .sblk = &dpu_rgb_sblk, + .xin_id = 13, + .type = SSPP_TYPE_RGB, + .clk_ctrl = DPU_CLK_CTRL_RGB3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x150, + .features = DMA_MSM8996_MASK, + .sblk = &dpu_dma_sblk, + .xin_id = 2, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x150, + .features = DMA_MSM8996_MASK, + .sblk = &dpu_dma_sblk, + .xin_id = 10, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, +}; + +static const struct dpu_lm_cfg msm8996_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_2, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, +}; + +static const struct dpu_pingpong_cfg msm8996_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &msm8996_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .sblk = &msm8996_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .sblk = &msm8996_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x71800, .len = 0xd4, + .sblk = &msm8996_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15), + }, +}; + +static const struct dpu_dsc_cfg msm8996_dsc[] = { + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + }, { + .name = "dsc_1", .id = DSC_1, + .base = 0x80400, .len = 0x140, + }, +}; + +static const struct dpu_dspp_cfg msm8996_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &msm8998_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &msm8998_dspp_sblk, + }, +}; + +static const struct dpu_intf_cfg msm8996_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x268, + .type = INTF_NONE, + .prog_fetch_lines_worst_case = 25, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x268, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 25, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x268, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 25, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x6b800, .len = 0x268, + .type = INTF_HDMI, + .prog_fetch_lines_worst_case = 25, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_perf_cfg msm8996_perf_data = { + .max_bw_low = 9600000, + .max_bw_high = 9600000, + .min_core_ib = 2400000, + .min_llcc_ib = 0, /* No LLCC on this SoC */ + .min_dram_ib = 800000, + .undersized_prefill_lines = 2, + .xtra_prefill_lines = 2, + .dest_scale_prefill_lines = 3, + .macrotile_prefill_lines = 4, + .yuv_nv12_prefill_lines = 8, + .linear_prefill_lines = 1, + .downscaling_prefill_lines = 1, + .amortizable_threshold = 25, + .min_prefill_lines = 21, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfffc, 0xff00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(msm8998_qos_linear), + .entries = msm8998_qos_linear + }, + {.nentry = ARRAY_SIZE(msm8998_qos_macrotile), + .entries = msm8998_qos_macrotile + }, + {.nentry = ARRAY_SIZE(msm8998_qos_nrt), + .entries = msm8998_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version msm8996_mdss_ver = { + .core_major_ver = 1, + .core_minor_ver = 7, +}; + +const struct dpu_mdss_cfg dpu_msm8996_cfg = { + .mdss_ver = &msm8996_mdss_ver, + .caps = &msm8996_dpu_caps, + .mdp = msm8996_mdp, + .cdm = &dpu_cdm_1_x_4_x, + .ctl_count = ARRAY_SIZE(msm8996_ctl), + .ctl = msm8996_ctl, + .sspp_count = ARRAY_SIZE(msm8996_sspp), + .sspp = msm8996_sspp, + .mixer_count = ARRAY_SIZE(msm8996_lm), + .mixer = msm8996_lm, + .dspp_count = ARRAY_SIZE(msm8996_dspp), + .dspp = msm8996_dspp, + .pingpong_count = ARRAY_SIZE(msm8996_pp), + .pingpong = msm8996_pp, + .dsc_count = ARRAY_SIZE(msm8996_dsc), + .dsc = msm8996_dsc, + .intf_count = ARRAY_SIZE(msm8996_intf), + .intf = msm8996_intf, + .vbif = &msm8996_vbif, + .perf = &msm8996_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h new file mode 100644 index 000000000..67c1463d3 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h @@ -0,0 +1,312 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_3_0_MSM8998_H +#define _DPU_3_0_MSM8998_H + +static const struct dpu_caps msm8998_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0x7, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, + .max_hdeci_exp = MAX_HORZ_DECIMATION, + .max_vdeci_exp = MAX_VERT_DECIMATION, +}; + +static const struct dpu_mdp_cfg msm8998_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x458, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 12 }, + [DPU_CLK_CTRL_CURSOR0] = { .reg_off = 0x3a8, .bit_off = 16 }, + [DPU_CLK_CTRL_CURSOR1] = { .reg_off = 0x3b0, .bit_off = 16 }, + }, +}; + +static const struct dpu_ctl_cfg msm8998_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x94, + .features = BIT(DPU_CTL_SPLIT_DISPLAY), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x94, + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x94, + .features = BIT(DPU_CTL_SPLIT_DISPLAY), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x94, + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x94, + }, +}; + +static const struct dpu_sspp_cfg msm8998_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1ac, + .features = VIG_MSM8998_MASK, + .sblk = &dpu_vig_sblk_qseed3_1_2, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1ac, + .features = VIG_MSM8998_MASK, + .sblk = &dpu_vig_sblk_qseed3_1_2, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x1ac, + .features = VIG_MSM8998_MASK, + .sblk = &dpu_vig_sblk_qseed3_1_2, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x1ac, + .features = VIG_MSM8998_MASK, + .sblk = &dpu_vig_sblk_qseed3_1_2, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1ac, + .features = DMA_MSM8998_MASK, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1ac, + .features = DMA_MSM8998_MASK, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1ac, + .features = DMA_CURSOR_MSM8998_MASK, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x1ac, + .features = DMA_CURSOR_MSM8998_MASK, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, +}; + +static const struct dpu_lm_cfg msm8998_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_2, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, +}; + +static const struct dpu_pingpong_cfg msm8998_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x71800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15), + }, +}; + +static const struct dpu_dsc_cfg msm8998_dsc[] = { + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + }, { + .name = "dsc_1", .id = DSC_1, + .base = 0x80400, .len = 0x140, + }, +}; + +static const struct dpu_dspp_cfg msm8998_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &msm8998_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &msm8998_dspp_sblk, + }, +}; + +static const struct dpu_intf_cfg msm8998_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 21, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x280, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 21, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x280, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 21, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x6b800, .len = 0x280, + .type = INTF_HDMI, + .prog_fetch_lines_worst_case = 21, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_perf_cfg msm8998_perf_data = { + .max_bw_low = 6700000, + .max_bw_high = 6700000, + .min_core_ib = 2400000, + .min_llcc_ib = 800000, + .min_dram_ib = 800000, + .undersized_prefill_lines = 2, + .xtra_prefill_lines = 2, + .dest_scale_prefill_lines = 3, + .macrotile_prefill_lines = 4, + .yuv_nv12_prefill_lines = 8, + .linear_prefill_lines = 1, + .downscaling_prefill_lines = 1, + .amortizable_threshold = 25, + .min_prefill_lines = 25, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfffc, 0xff00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(msm8998_qos_linear), + .entries = msm8998_qos_linear + }, + {.nentry = ARRAY_SIZE(msm8998_qos_macrotile), + .entries = msm8998_qos_macrotile + }, + {.nentry = ARRAY_SIZE(msm8998_qos_nrt), + .entries = msm8998_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 200, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version msm8998_mdss_ver = { + .core_major_ver = 3, + .core_minor_ver = 0, +}; + +const struct dpu_mdss_cfg dpu_msm8998_cfg = { + .mdss_ver = &msm8998_mdss_ver, + .caps = &msm8998_dpu_caps, + .mdp = &msm8998_mdp, + .cdm = &dpu_cdm_1_x_4_x, + .ctl_count = ARRAY_SIZE(msm8998_ctl), + .ctl = msm8998_ctl, + .sspp_count = ARRAY_SIZE(msm8998_sspp), + .sspp = msm8998_sspp, + .mixer_count = ARRAY_SIZE(msm8998_lm), + .mixer = msm8998_lm, + .dspp_count = ARRAY_SIZE(msm8998_dspp), + .dspp = msm8998_dspp, + .pingpong_count = ARRAY_SIZE(msm8998_pp), + .pingpong = msm8998_pp, + .dsc_count = ARRAY_SIZE(msm8998_dsc), + .dsc = msm8998_dsc, + .intf_count = ARRAY_SIZE(msm8998_intf), + .intf = msm8998_intf, + .vbif = &msm8998_vbif, + .perf = &msm8998_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_2_sdm660.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_2_sdm660.h new file mode 100644 index 000000000..843440298 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_2_sdm660.h @@ -0,0 +1,276 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2023. Linaro Inc. All rights reserved. + */ + +#ifndef _DPU_3_2_SDM660_H +#define _DPU_3_2_SDM660_H + +static const struct dpu_caps sdm660_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0x7, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, + .max_hdeci_exp = MAX_HORZ_DECIMATION, + .max_vdeci_exp = MAX_VERT_DECIMATION, +}; + +static const struct dpu_mdp_cfg sdm660_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x458, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_CURSOR0] = { .reg_off = 0x3a8, .bit_off = 16 }, + }, +}; + +static const struct dpu_ctl_cfg sdm660_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x94, + .features = BIT(DPU_CTL_SPLIT_DISPLAY), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x94, + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x94, + .features = BIT(DPU_CTL_SPLIT_DISPLAY), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x94, + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x94, + }, +}; + +static const struct dpu_sspp_cfg sdm660_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1ac, + .features = VIG_MSM8998_MASK, + .sblk = &dpu_vig_sblk_qseed3_1_2, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1ac, + .features = VIG_MSM8998_MASK, + .sblk = &dpu_vig_sblk_qseed3_1_2, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1ac, + .features = DMA_MSM8998_MASK, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1ac, + .features = DMA_MSM8998_MASK, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1ac, + .features = DMA_CURSOR_MSM8998_MASK, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, +}; + +static const struct dpu_lm_cfg sdm660_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_2, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, +}; + +static const struct dpu_pingpong_cfg sdm660_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x71800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15), + }, +}; + +static const struct dpu_dsc_cfg sdm660_dsc[] = { + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + }, { + .name = "dsc_1", .id = DSC_1, + .base = 0x80400, .len = 0x140, + }, +}; + +static const struct dpu_dspp_cfg sdm660_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &msm8998_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &msm8998_dspp_sblk, + }, +}; + +static const struct dpu_intf_cfg sdm660_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 21, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x280, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 21, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x280, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 21, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + }, +}; + +static const struct dpu_perf_cfg sdm660_perf_data = { + .max_bw_low = 6600000, + .max_bw_high = 6600000, + .min_core_ib = 3100000, + .min_llcc_ib = 800000, + .min_dram_ib = 800000, + .undersized_prefill_lines = 2, + .xtra_prefill_lines = 2, + .dest_scale_prefill_lines = 3, + .macrotile_prefill_lines = 4, + .yuv_nv12_prefill_lines = 8, + .linear_prefill_lines = 1, + .downscaling_prefill_lines = 1, + .amortizable_threshold = 25, + .min_prefill_lines = 25, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfffc, 0xff00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(msm8998_qos_linear), + .entries = msm8998_qos_linear + }, + {.nentry = ARRAY_SIZE(msm8998_qos_macrotile), + .entries = msm8998_qos_macrotile + }, + {.nentry = ARRAY_SIZE(msm8998_qos_nrt), + .entries = msm8998_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 200, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sdm660_mdss_ver = { + .core_major_ver = 3, + .core_minor_ver = 2, +}; + +const struct dpu_mdss_cfg dpu_sdm660_cfg = { + .mdss_ver = &sdm660_mdss_ver, + .caps = &sdm660_dpu_caps, + .mdp = &sdm660_mdp, + .cdm = &dpu_cdm_1_x_4_x, + .ctl_count = ARRAY_SIZE(sdm660_ctl), + .ctl = sdm660_ctl, + .sspp_count = ARRAY_SIZE(sdm660_sspp), + .sspp = sdm660_sspp, + .mixer_count = ARRAY_SIZE(sdm660_lm), + .mixer = sdm660_lm, + .dspp_count = ARRAY_SIZE(sdm660_dspp), + .dspp = sdm660_dspp, + .pingpong_count = ARRAY_SIZE(sdm660_pp), + .pingpong = sdm660_pp, + .dsc_count = ARRAY_SIZE(sdm660_dsc), + .dsc = sdm660_dsc, + .intf_count = ARRAY_SIZE(sdm660_intf), + .intf = sdm660_intf, + .vbif = &msm8998_vbif, + .perf = &sdm660_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_3_sdm630.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_3_sdm630.h new file mode 100644 index 000000000..ef5777aee --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_3_sdm630.h @@ -0,0 +1,214 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2023. Linaro Inc. All rights reserved. + */ + +#ifndef _DPU_3_3_SDM630_H +#define _DPU_3_3_SDM630_H + +static const struct dpu_caps sdm630_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_LINE_WIDTH, + .max_mixer_blendstages = 0x7, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = DEFAULT_DPU_LINE_WIDTH, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, + .max_hdeci_exp = MAX_HORZ_DECIMATION, + .max_vdeci_exp = MAX_VERT_DECIMATION, +}; + +static const struct dpu_mdp_cfg sdm630_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x458, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_CURSOR0] = { .reg_off = 0x3a8, .bit_off = 16 }, + }, +}; + +static const struct dpu_ctl_cfg sdm630_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x94, + .features = BIT(DPU_CTL_SPLIT_DISPLAY), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x94, + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x94, + .features = BIT(DPU_CTL_SPLIT_DISPLAY), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x94, + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x94, + }, +}; + +static const struct dpu_sspp_cfg sdm630_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1ac, + .features = VIG_MSM8998_MASK, + .sblk = &dpu_vig_sblk_qseed3_1_2, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1ac, + .features = DMA_MSM8998_MASK, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1ac, + .features = DMA_MSM8998_MASK, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1ac, + .features = DMA_CURSOR_MSM8998_MASK, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, +}; + +static const struct dpu_lm_cfg sdm630_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .pingpong = PINGPONG_2, + }, +}; + +static const struct dpu_pingpong_cfg sdm630_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14), + }, +}; + +static const struct dpu_dspp_cfg sdm630_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &msm8998_dspp_sblk, + }, +}; + +static const struct dpu_intf_cfg sdm630_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 21, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x280, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 21, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + }, +}; + +static const struct dpu_perf_cfg sdm630_perf_data = { + .max_bw_low = 4100000, + .max_bw_high = 4100000, + .min_core_ib = 3200000, + .min_llcc_ib = 800000, + .min_dram_ib = 800000, + .undersized_prefill_lines = 2, + .xtra_prefill_lines = 2, + .dest_scale_prefill_lines = 3, + .macrotile_prefill_lines = 4, + .yuv_nv12_prefill_lines = 8, + .linear_prefill_lines = 1, + .downscaling_prefill_lines = 1, + .amortizable_threshold = 25, + .min_prefill_lines = 25, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfffc, 0xff00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(msm8998_qos_linear), + .entries = msm8998_qos_linear + }, + {.nentry = ARRAY_SIZE(msm8998_qos_macrotile), + .entries = msm8998_qos_macrotile + }, + {.nentry = ARRAY_SIZE(msm8998_qos_nrt), + .entries = msm8998_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 200, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sdm630_mdss_ver = { + .core_major_ver = 3, + .core_minor_ver = 3, +}; + +const struct dpu_mdss_cfg dpu_sdm630_cfg = { + .mdss_ver = &sdm630_mdss_ver, + .caps = &sdm630_dpu_caps, + .mdp = &sdm630_mdp, + .cdm = &dpu_cdm_1_x_4_x, + .ctl_count = ARRAY_SIZE(sdm630_ctl), + .ctl = sdm630_ctl, + .sspp_count = ARRAY_SIZE(sdm630_sspp), + .sspp = sdm630_sspp, + .mixer_count = ARRAY_SIZE(sdm630_lm), + .mixer = sdm630_lm, + .dspp_count = ARRAY_SIZE(sdm630_dspp), + .dspp = sdm630_dspp, + .pingpong_count = ARRAY_SIZE(sdm630_pp), + .pingpong = sdm630_pp, + .intf_count = ARRAY_SIZE(sdm630_intf), + .intf = sdm630_intf, + .vbif = &msm8998_vbif, + .perf = &sdm630_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h new file mode 100644 index 000000000..b608d2bb6 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h @@ -0,0 +1,332 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_4_0_SDM845_H +#define _DPU_4_0_SDM845_H + +static const struct dpu_caps sdm845_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, + .max_hdeci_exp = MAX_HORZ_DECIMATION, + .max_vdeci_exp = MAX_VERT_DECIMATION, +}; + +static const struct dpu_mdp_cfg sdm845_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x45c, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + }, +}; + +static const struct dpu_ctl_cfg sdm845_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0xe4, + .features = BIT(DPU_CTL_SPLIT_DISPLAY), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0xe4, + .features = BIT(DPU_CTL_SPLIT_DISPLAY), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0xe4, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0xe4, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0xe4, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, +}; + +static const struct dpu_sspp_cfg sdm845_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1c8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_1_3, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1c8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_1_3, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x1c8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_1_3, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x1c8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_1_3, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1c8, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1c8, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1c8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x1c8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, +}; + +static const struct dpu_lm_cfg sdm845_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, +}; + +static const struct dpu_dspp_cfg sdm845_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sdm845_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x71800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15), + }, +}; + +static const struct dpu_dsc_cfg sdm845_dsc[] = { + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + }, { + .name = "dsc_1", .id = DSC_1, + .base = 0x80400, .len = 0x140, + }, { + .name = "dsc_2", .id = DSC_2, + .base = 0x80800, .len = 0x140, + }, { + .name = "dsc_3", .id = DSC_3, + .base = 0x80c00, .len = 0x140, + }, +}; + +static const struct dpu_intf_cfg sdm845_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x280, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x280, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x6b800, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_perf_cfg sdm845_perf_data = { + .max_bw_low = 6800000, + .max_bw_high = 6800000, + .min_core_ib = 2400000, + .min_llcc_ib = 800000, + .min_dram_ib = 800000, + .undersized_prefill_lines = 2, + .xtra_prefill_lines = 2, + .dest_scale_prefill_lines = 3, + .macrotile_prefill_lines = 4, + .yuv_nv12_prefill_lines = 8, + .linear_prefill_lines = 1, + .downscaling_prefill_lines = 1, + .amortizable_threshold = 25, + .min_prefill_lines = 24, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfff0, 0xf000, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sdm845_qos_linear), + .entries = sdm845_qos_linear + }, + {.nentry = ARRAY_SIZE(sdm845_qos_macrotile), + .entries = sdm845_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sdm845_qos_nrt), + .entries = sdm845_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sdm845_mdss_ver = { + .core_major_ver = 4, + .core_minor_ver = 0, +}; + +const struct dpu_mdss_cfg dpu_sdm845_cfg = { + .mdss_ver = &sdm845_mdss_ver, + .caps = &sdm845_dpu_caps, + .mdp = &sdm845_mdp, + .cdm = &dpu_cdm_1_x_4_x, + .ctl_count = ARRAY_SIZE(sdm845_ctl), + .ctl = sdm845_ctl, + .sspp_count = ARRAY_SIZE(sdm845_sspp), + .sspp = sdm845_sspp, + .mixer_count = ARRAY_SIZE(sdm845_lm), + .mixer = sdm845_lm, + .dspp_count = ARRAY_SIZE(sdm845_dspp), + .dspp = sdm845_dspp, + .pingpong_count = ARRAY_SIZE(sdm845_pp), + .pingpong = sdm845_pp, + .dsc_count = ARRAY_SIZE(sdm845_dsc), + .dsc = sdm845_dsc, + .intf_count = ARRAY_SIZE(sdm845_intf), + .intf = sdm845_intf, + .vbif = &sdm845_vbif, + .perf = &sdm845_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_1_sdm670.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_1_sdm670.h new file mode 100644 index 000000000..54b4a83ee --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_1_sdm670.h @@ -0,0 +1,151 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2023, Richard Acayan. All rights reserved. + */ + +#ifndef _DPU_4_1_SDM670_H +#define _DPU_4_1_SDM670_H + +static const struct dpu_mdp_cfg sdm670_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x45c, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + }, +}; + +static const struct dpu_sspp_cfg sdm670_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1c8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_1_3, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1c8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_1_3, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1c8, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1c8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1c8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, +}; + +static const struct dpu_lm_cfg sdm670_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_2, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, +}; + +static const struct dpu_dspp_cfg sdm670_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_dsc_cfg sdm670_dsc[] = { + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + }, { + .name = "dsc_1", .id = DSC_1, + .base = 0x80400, .len = 0x140, + }, +}; + +static const struct dpu_mdss_version sdm670_mdss_ver = { + .core_major_ver = 4, + .core_minor_ver = 1, +}; + +const struct dpu_mdss_cfg dpu_sdm670_cfg = { + .mdss_ver = &sdm670_mdss_ver, + .caps = &sdm845_dpu_caps, + .mdp = &sdm670_mdp, + .cdm = &dpu_cdm_1_x_4_x, + .ctl_count = ARRAY_SIZE(sdm845_ctl), + .ctl = sdm845_ctl, + .sspp_count = ARRAY_SIZE(sdm670_sspp), + .sspp = sdm670_sspp, + .mixer_count = ARRAY_SIZE(sdm670_lm), + .mixer = sdm670_lm, + .dspp_count = ARRAY_SIZE(sdm670_dspp), + .dspp = sdm670_dspp, + .pingpong_count = ARRAY_SIZE(sdm845_pp), + .pingpong = sdm845_pp, + .dsc_count = ARRAY_SIZE(sdm670_dsc), + .dsc = sdm670_dsc, + .intf_count = ARRAY_SIZE(sdm845_intf), + .intf = sdm845_intf, + .vbif = &sdm845_vbif, + .perf = &sdm845_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h new file mode 100644 index 000000000..9729a65d3 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h @@ -0,0 +1,387 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_5_0_SM8150_H +#define _DPU_5_0_SM8150_H + +static const struct dpu_caps sm8150_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 4096, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, + .max_hdeci_exp = MAX_HORZ_DECIMATION, + .max_vdeci_exp = MAX_VERT_DECIMATION, +}; + +static const struct dpu_mdp_cfg sm8150_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x45c, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x2bc, .bit_off = 16 }, + }, +}; + +static const struct dpu_ctl_cfg sm8150_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a00, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sm8150_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f0, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_1_4, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1f0, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_1_4, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x1f0, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_1_4, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x1f0, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_1_4, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f0, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f0, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f0, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x1f0, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, +}; + +static const struct dpu_lm_cfg sm8150_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, +}; + +static const struct dpu_dspp_cfg sm8150_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sm8150_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x71800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x72000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x72800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, +}; + +static const struct dpu_merge_3d_cfg sm8150_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x83000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x83100, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x83200, .len = 0x8, + }, +}; + +static const struct dpu_dsc_cfg sm8150_dsc[] = { + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + }, { + .name = "dsc_1", .id = DSC_1, + .base = 0x80400, .len = 0x140, + }, { + .name = "dsc_2", .id = DSC_2, + .base = 0x80800, .len = 0x140, + }, { + .name = "dsc_3", .id = DSC_3, + .base = 0x80c00, .len = 0x140, + }, +}; + +static const struct dpu_wb_cfg sm8150_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_intf_cfg sm8150_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2bc, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x2bc, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x6b800, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_perf_cfg sm8150_perf_data = { + .max_bw_low = 12800000, + .max_bw_high = 12800000, + .min_core_ib = 2400000, + .min_llcc_ib = 800000, + .min_dram_ib = 800000, + .min_prefill_lines = 24, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfff8, 0xf000, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sm8150_qos_linear), + .entries = sm8150_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm8150_mdss_ver = { + .core_major_ver = 5, + .core_minor_ver = 0, +}; + +const struct dpu_mdss_cfg dpu_sm8150_cfg = { + .mdss_ver = &sm8150_mdss_ver, + .caps = &sm8150_dpu_caps, + .mdp = &sm8150_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sm8150_ctl), + .ctl = sm8150_ctl, + .sspp_count = ARRAY_SIZE(sm8150_sspp), + .sspp = sm8150_sspp, + .mixer_count = ARRAY_SIZE(sm8150_lm), + .mixer = sm8150_lm, + .dspp_count = ARRAY_SIZE(sm8150_dspp), + .dspp = sm8150_dspp, + .dsc_count = ARRAY_SIZE(sm8150_dsc), + .dsc = sm8150_dsc, + .pingpong_count = ARRAY_SIZE(sm8150_pp), + .pingpong = sm8150_pp, + .merge_3d_count = ARRAY_SIZE(sm8150_merge_3d), + .merge_3d = sm8150_merge_3d, + .wb_count = ARRAY_SIZE(sm8150_wb), + .wb = sm8150_wb, + .intf_count = ARRAY_SIZE(sm8150_intf), + .intf = sm8150_intf, + .vbif = &sdm845_vbif, + .perf = &sm8150_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h new file mode 100644 index 000000000..fb18de029 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h @@ -0,0 +1,411 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_5_1_SC8180X_H +#define _DPU_5_1_SC8180X_H + +static const struct dpu_caps sc8180x_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 4096, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, + .max_hdeci_exp = MAX_HORZ_DECIMATION, + .max_vdeci_exp = MAX_VERT_DECIMATION, +}; + +static const struct dpu_mdp_cfg sc8180x_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x45c, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x2bc, .bit_off = 16 }, + }, +}; + +static const struct dpu_ctl_cfg sc8180x_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a00, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sc8180x_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f0, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_1_4, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1f0, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_1_4, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x1f0, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_1_4, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x1f0, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_1_4, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f0, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f0, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f0, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x1f0, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, +}; + +static const struct dpu_lm_cfg sc8180x_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, +}; + +static const struct dpu_dspp_cfg sc8180x_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sc8180x_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x71800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x72000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x72800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, +}; + +static const struct dpu_merge_3d_cfg sc8180x_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x83000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x83100, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x83200, .len = 0x8, + }, +}; + +static const struct dpu_dsc_cfg sc8180x_dsc[] = { + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + }, { + .name = "dsc_1", .id = DSC_1, + .base = 0x80400, .len = 0x140, + }, { + .name = "dsc_2", .id = DSC_2, + .base = 0x80800, .len = 0x140, + }, { + .name = "dsc_3", .id = DSC_3, + .base = 0x80c00, .len = 0x140, + }, { + .name = "dsc_4", .id = DSC_4, + .base = 0x81000, .len = 0x140, + }, { + .name = "dsc_5", .id = DSC_5, + .base = 0x81400, .len = 0x140, + }, +}; + +static const struct dpu_wb_cfg sc8180x_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_intf_cfg sc8180x_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2bc, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x2bc, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, + /* INTF_3 is for MST, wired to INTF_DP 0 and 1, use dummy index until this is supported */ + { + .name = "intf_3", .id = INTF_3, + .base = 0x6b800, .len = 0x280, + .type = INTF_DP, + .controller_id = 999, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, { + .name = "intf_4", .id = INTF_4, + .base = 0x6c000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 20), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 21), + }, { + .name = "intf_5", .id = INTF_5, + .base = 0x6c800, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_2, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23), + }, +}; + +static const struct dpu_perf_cfg sc8180x_perf_data = { + .max_bw_low = 9600000, + .max_bw_high = 9600000, + .min_core_ib = 2400000, + .min_llcc_ib = 800000, + .min_dram_ib = 800000, + .min_prefill_lines = 24, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfff0, 0xf000, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sc8180x_mdss_ver = { + .core_major_ver = 5, + .core_minor_ver = 1, +}; + +const struct dpu_mdss_cfg dpu_sc8180x_cfg = { + .mdss_ver = &sc8180x_mdss_ver, + .caps = &sc8180x_dpu_caps, + .mdp = &sc8180x_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sc8180x_ctl), + .ctl = sc8180x_ctl, + .sspp_count = ARRAY_SIZE(sc8180x_sspp), + .sspp = sc8180x_sspp, + .mixer_count = ARRAY_SIZE(sc8180x_lm), + .mixer = sc8180x_lm, + .dspp_count = ARRAY_SIZE(sc8180x_dspp), + .dspp = sc8180x_dspp, + .dsc_count = ARRAY_SIZE(sc8180x_dsc), + .dsc = sc8180x_dsc, + .pingpong_count = ARRAY_SIZE(sc8180x_pp), + .pingpong = sc8180x_pp, + .merge_3d_count = ARRAY_SIZE(sc8180x_merge_3d), + .merge_3d = sc8180x_merge_3d, + .wb_count = ARRAY_SIZE(sc8180x_wb), + .wb = sc8180x_wb, + .intf_count = ARRAY_SIZE(sc8180x_intf), + .intf = sc8180x_intf, + .vbif = &sdm845_vbif, + .perf = &sc8180x_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_2_sm7150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_2_sm7150.h new file mode 100644 index 000000000..92a614686 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_2_sm7150.h @@ -0,0 +1,315 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2018-2019, The Linux Foundation. All rights reserved. + * Copyright (c) 2024, Danila Tikhonov + */ + +#ifndef _DPU_5_2_SM7150_H +#define _DPU_5_2_SM7150_H + +static const struct dpu_caps sm7150_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 2880, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, + .max_hdeci_exp = MAX_HORZ_DECIMATION, + .max_vdeci_exp = MAX_VERT_DECIMATION, +}; + +static const struct dpu_mdp_cfg sm7150_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x45c, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x2bc, .bit_off = 16 }, + }, +}; + +static const struct dpu_ctl_cfg sm7150_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a00, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sm7150_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f0, + .features = VIG_SDM845_MASK_NO_SDMA, + .sblk = &dpu_vig_sblk_qseed3_2_4, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1f0, + .features = VIG_SDM845_MASK_NO_SDMA, + .sblk = &dpu_vig_sblk_qseed3_2_4, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f0, + .features = DMA_SDM845_MASK_NO_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f0, + .features = DMA_SDM845_MASK_NO_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f0, + .features = DMA_CURSOR_SDM845_MASK_NO_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, +}; + +static const struct dpu_lm_cfg sm7150_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, +}; + +static const struct dpu_dspp_cfg sm7150_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sm7150_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x71800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, +}; + +static const struct dpu_merge_3d_cfg sm7150_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x83000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x83100, .len = 0x8, + }, +}; + +static const struct dpu_dsc_cfg sm7150_dsc[] = { + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + }, { + .name = "dsc_1", .id = DSC_1, + .base = 0x80400, .len = 0x140, + }, +}; + +static const struct dpu_intf_cfg sm7150_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2bc, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x2bc, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x6b800, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_wb_cfg sm7150_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_perf_cfg sm7150_perf_data = { + .max_bw_low = 7100000, + .max_bw_high = 7100000, + .min_core_ib = 2400000, + .min_llcc_ib = 800000, + .min_dram_ib = 800000, + .min_prefill_lines = 24, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfff8, 0xf000, 0xffff}, + .qos_lut_tbl = { + { + .nentry = ARRAY_SIZE(sm8150_qos_linear), + .entries = sm8150_qos_linear + }, { + .nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, { + .nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm7150_mdss_ver = { + .core_major_ver = 5, + .core_minor_ver = 2, +}; + +const struct dpu_mdss_cfg dpu_sm7150_cfg = { + .mdss_ver = &sm7150_mdss_ver, + .caps = &sm7150_dpu_caps, + .mdp = &sm7150_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sm7150_ctl), + .ctl = sm7150_ctl, + .sspp_count = ARRAY_SIZE(sm7150_sspp), + .sspp = sm7150_sspp, + .mixer_count = ARRAY_SIZE(sm7150_lm), + .mixer = sm7150_lm, + .dspp_count = ARRAY_SIZE(sm7150_dspp), + .dspp = sm7150_dspp, + .pingpong_count = ARRAY_SIZE(sm7150_pp), + .pingpong = sm7150_pp, + .merge_3d_count = ARRAY_SIZE(sm7150_merge_3d), + .merge_3d = sm7150_merge_3d, + .dsc_count = ARRAY_SIZE(sm7150_dsc), + .dsc = sm7150_dsc, + .intf_count = ARRAY_SIZE(sm7150_intf), + .intf = sm7150_intf, + .wb_count = ARRAY_SIZE(sm7150_wb), + .wb = sm7150_wb, + .vbif = &sdm845_vbif, + .perf = &sm7150_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_3_sm6150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_3_sm6150.h new file mode 100644 index 000000000..87488d3ed --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_3_sm6150.h @@ -0,0 +1,252 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2024 Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#ifndef _DPU_5_3_SM6150_H +#define _DPU_5_3_SM6150_H + +static const struct dpu_caps sm6150_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0x9, + .has_dim_layer = true, + .has_idle_pc = true, + .max_linewidth = 2160, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, + .max_hdeci_exp = MAX_HORZ_DECIMATION, + .max_vdeci_exp = MAX_VERT_DECIMATION, +}; + +static const struct dpu_mdp_cfg sm6150_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x45c, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x2bc, .bit_off = 16 }, + }, +}; + +static const struct dpu_ctl_cfg sm6150_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a00, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sm6150_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f0, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_2_4, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f0, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f0, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f0, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x1f0, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, +}; + +static const struct dpu_lm_cfg sm6150_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .sblk = &sdm845_lm_sblk, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .sblk = &sdm845_lm_sblk, + .pingpong = PINGPONG_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .sblk = &sdm845_lm_sblk, + .pingpong = PINGPONG_2, + }, +}; + +static const struct dpu_dspp_cfg sm6150_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sm6150_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, +}; + +static const struct dpu_wb_cfg sm6150_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .maxlinewidth = 2160, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_intf_cfg sm6150_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x6b800, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_perf_cfg sm6150_perf_data = { + .max_bw_low = 4800000, + .max_bw_high = 4800000, + .min_core_ib = 2400000, + .min_llcc_ib = 0, + .min_dram_ib = 800000, + .min_prefill_lines = 24, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfff8, 0xf000, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sm8150_qos_linear), + .entries = sm8150_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm6150_mdss_ver = { + .core_major_ver = 5, + .core_minor_ver = 3, +}; + +const struct dpu_mdss_cfg dpu_sm6150_cfg = { + .mdss_ver = &sm6150_mdss_ver, + .caps = &sm6150_dpu_caps, + .mdp = &sm6150_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sm6150_ctl), + .ctl = sm6150_ctl, + .sspp_count = ARRAY_SIZE(sm6150_sspp), + .sspp = sm6150_sspp, + .mixer_count = ARRAY_SIZE(sm6150_lm), + .mixer = sm6150_lm, + .dspp_count = ARRAY_SIZE(sm6150_dspp), + .dspp = sm6150_dspp, + .pingpong_count = ARRAY_SIZE(sm6150_pp), + .pingpong = sm6150_pp, + .wb_count = ARRAY_SIZE(sm6150_wb), + .wb = sm6150_wb, + .intf_count = ARRAY_SIZE(sm6150_intf), + .intf = sm6150_intf, + .vbif = &sdm845_vbif, + .perf = &sm6150_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h new file mode 100644 index 000000000..64be51e30 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h @@ -0,0 +1,223 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2023 Marijn Suijten . All rights reserved. + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_5_4_SM6125_H +#define _DPU_5_4_SM6125_H + +static const struct dpu_caps sm6125_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0x6, + .has_dim_layer = true, + .has_idle_pc = true, + .max_linewidth = 2160, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, + .max_hdeci_exp = MAX_HORZ_DECIMATION, + .max_vdeci_exp = MAX_VERT_DECIMATION, +}; + +static const struct dpu_mdp_cfg sm6125_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x45c, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x2bc, .bit_off = 16 }, + }, +}; + +static const struct dpu_ctl_cfg sm6125_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a00, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sm6125_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f0, + .features = VIG_SDM845_MASK_NO_SDMA, + .sblk = &dpu_vig_sblk_qseed3_2_4, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f0, + .features = DMA_SDM845_MASK_NO_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f0, + .features = DMA_SDM845_MASK_NO_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, +}; + +static const struct dpu_lm_cfg sm6125_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .sblk = &sdm845_lm_sblk, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + .lm_pair = LM_1, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .sblk = &sdm845_lm_sblk, + .pingpong = PINGPONG_1, + .dspp = 0, + .lm_pair = LM_0, + }, +}; + +static const struct dpu_dspp_cfg sm6125_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sm6125_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .merge_3d = 0, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .merge_3d = 0, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, +}; + +static const struct dpu_wb_cfg sm6125_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .maxlinewidth = 2160, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_intf_cfg sm6125_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .type = INTF_DSI, + .controller_id = 0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, +}; + +static const struct dpu_perf_cfg sm6125_perf_data = { + .max_bw_low = 4100000, + .max_bw_high = 4100000, + .min_core_ib = 2400000, + .min_llcc_ib = 0, /* No LLCC on this SoC */ + .min_dram_ib = 800000, + .min_prefill_lines = 24, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfff8, 0xf000, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sm8150_qos_linear), + .entries = sm8150_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm6125_mdss_ver = { + .core_major_ver = 5, + .core_minor_ver = 4, +}; + +const struct dpu_mdss_cfg dpu_sm6125_cfg = { + .mdss_ver = &sm6125_mdss_ver, + .caps = &sm6125_dpu_caps, + .mdp = &sm6125_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sm6125_ctl), + .ctl = sm6125_ctl, + .sspp_count = ARRAY_SIZE(sm6125_sspp), + .sspp = sm6125_sspp, + .mixer_count = ARRAY_SIZE(sm6125_lm), + .mixer = sm6125_lm, + .dspp_count = ARRAY_SIZE(sm6125_dspp), + .dspp = sm6125_dspp, + .pingpong_count = ARRAY_SIZE(sm6125_pp), + .pingpong = sm6125_pp, + .wb_count = ARRAY_SIZE(sm6125_wb), + .wb = sm6125_wb, + .intf_count = ARRAY_SIZE(sm6125_intf), + .intf = sm6125_intf, + .vbif = &sdm845_vbif, + .perf = &sm6125_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h new file mode 100644 index 000000000..0e311d54e --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h @@ -0,0 +1,386 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_6_0_SM8250_H +#define _DPU_6_0_SM8250_H + +static const struct dpu_caps sm8250_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 4096, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sm8250_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x2bc, .bit_off = 16 }, + }, +}; + +static const struct dpu_ctl_cfg sm8250_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a00, .len = 0x1e0, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sm8250_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1f8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x1f8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x1f8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f8, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, +}; + +static const struct dpu_lm_cfg sm8250_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, +}; + +static const struct dpu_dspp_cfg sm8250_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sm8250_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x71800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x72000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x72800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, +}; + +static const struct dpu_merge_3d_cfg sm8250_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x83000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x83100, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x83200, .len = 0x8, + }, +}; + +static const struct dpu_dsc_cfg sm8250_dsc[] = { + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + }, { + .name = "dsc_1", .id = DSC_1, + .base = 0x80400, .len = 0x140, + }, { + .name = "dsc_2", .id = DSC_2, + .base = 0x80800, .len = 0x140, + }, { + .name = "dsc_3", .id = DSC_3, + .base = 0x80c00, .len = 0x140, + }, +}; + +static const struct dpu_intf_cfg sm8250_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x2c0, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x6b800, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_wb_cfg sm8250_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_perf_cfg sm8250_perf_data = { + .max_bw_low = 13700000, + .max_bw_high = 16600000, + .min_core_ib = 4800000, + .min_llcc_ib = 0, + .min_dram_ib = 800000, + .min_prefill_lines = 35, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfff0, 0xff00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm8250_mdss_ver = { + .core_major_ver = 6, + .core_minor_ver = 0, +}; + +const struct dpu_mdss_cfg dpu_sm8250_cfg = { + .mdss_ver = &sm8250_mdss_ver, + .caps = &sm8250_dpu_caps, + .mdp = &sm8250_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sm8250_ctl), + .ctl = sm8250_ctl, + .sspp_count = ARRAY_SIZE(sm8250_sspp), + .sspp = sm8250_sspp, + .mixer_count = ARRAY_SIZE(sm8250_lm), + .mixer = sm8250_lm, + .dspp_count = ARRAY_SIZE(sm8250_dspp), + .dspp = sm8250_dspp, + .dsc_count = ARRAY_SIZE(sm8250_dsc), + .dsc = sm8250_dsc, + .pingpong_count = ARRAY_SIZE(sm8250_pp), + .pingpong = sm8250_pp, + .merge_3d_count = ARRAY_SIZE(sm8250_merge_3d), + .merge_3d = sm8250_merge_3d, + .intf_count = ARRAY_SIZE(sm8250_intf), + .intf = sm8250_intf, + .vbif = &sdm845_vbif, + .wb_count = ARRAY_SIZE(sm8250_wb), + .wb = sm8250_wb, + .perf = &sm8250_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h new file mode 100644 index 000000000..d6f7ee24c --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h @@ -0,0 +1,217 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_6_2_SC7180_H +#define _DPU_6_2_SC7180_H + +static const struct dpu_caps sc7180_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0x9, + .has_dim_layer = true, + .has_idle_pc = true, + .max_linewidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sc7180_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x2bc, .bit_off = 16 }, + }, +}; + +static const struct dpu_ctl_cfg sc7180_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1dc, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1dc, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1dc, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, +}; + +static const struct dpu_sspp_cfg sc7180_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_SDM845_MASK_NO_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK_NO_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK_NO_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK_NO_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, +}; + +static const struct dpu_lm_cfg sc7180_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sc7180_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sc7180_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + }, +}; + +static const struct dpu_dspp_cfg sc7180_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sc7180_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, +}; + +static const struct dpu_intf_cfg sc7180_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, +}; + +static const struct dpu_wb_cfg sc7180_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_perf_cfg sc7180_perf_data = { + .max_bw_low = 6800000, + .max_bw_high = 6800000, + .min_core_ib = 2400000, + .min_llcc_ib = 800000, + .min_dram_ib = 1600000, + .min_prefill_lines = 24, + .danger_lut_tbl = {0xff, 0xffff, 0x0}, + .safe_lut_tbl = {0xfff0, 0xff00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sc7180_mdss_ver = { + .core_major_ver = 6, + .core_minor_ver = 2, +}; + +const struct dpu_mdss_cfg dpu_sc7180_cfg = { + .mdss_ver = &sc7180_mdss_ver, + .caps = &sc7180_dpu_caps, + .mdp = &sc7180_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sc7180_ctl), + .ctl = sc7180_ctl, + .sspp_count = ARRAY_SIZE(sc7180_sspp), + .sspp = sc7180_sspp, + .mixer_count = ARRAY_SIZE(sc7180_lm), + .mixer = sc7180_lm, + .dspp_count = ARRAY_SIZE(sc7180_dspp), + .dspp = sc7180_dspp, + .pingpong_count = ARRAY_SIZE(sc7180_pp), + .pingpong = sc7180_pp, + .intf_count = ARRAY_SIZE(sc7180_intf), + .intf = sc7180_intf, + .wb_count = ARRAY_SIZE(sc7180_wb), + .wb = sc7180_wb, + .vbif = &sdm845_vbif, + .perf = &sc7180_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h new file mode 100644 index 000000000..20a2e9ff5 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h @@ -0,0 +1,151 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_6_3_SM6115_H +#define _DPU_6_3_SM6115_H + +static const struct dpu_caps sm6115_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_LINE_WIDTH, + .max_mixer_blendstages = 0x4, + .has_dim_layer = true, + .has_idle_pc = true, + .max_linewidth = 2160, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sm6115_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + }, +}; + +static const struct dpu_ctl_cfg sm6115_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1dc, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, +}; + +static const struct dpu_sspp_cfg sm6115_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_SDM845_MASK_NO_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK_NO_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, +}; + +static const struct dpu_lm_cfg sm6115_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .sblk = &qcm2290_lm_sblk, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, +}; + +static const struct dpu_dspp_cfg sm6115_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sm6115_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, +}; + +static const struct dpu_intf_cfg sm6115_intf[] = { + { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, +}; + +static const struct dpu_perf_cfg sm6115_perf_data = { + .max_bw_low = 3100000, + .max_bw_high = 4000000, + .min_core_ib = 2400000, + .min_llcc_ib = 800000, + .min_dram_ib = 800000, + .min_prefill_lines = 24, + .danger_lut_tbl = {0xff, 0xffff, 0x0}, + .safe_lut_tbl = {0xfff0, 0xff00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm6115_mdss_ver = { + .core_major_ver = 6, + .core_minor_ver = 3, +}; + +const struct dpu_mdss_cfg dpu_sm6115_cfg = { + .mdss_ver = &sm6115_mdss_ver, + .caps = &sm6115_dpu_caps, + .mdp = &sm6115_mdp, + .ctl_count = ARRAY_SIZE(sm6115_ctl), + .ctl = sm6115_ctl, + .sspp_count = ARRAY_SIZE(sm6115_sspp), + .sspp = sm6115_sspp, + .mixer_count = ARRAY_SIZE(sm6115_lm), + .mixer = sm6115_lm, + .dspp_count = ARRAY_SIZE(sm6115_dspp), + .dspp = sm6115_dspp, + .pingpong_count = ARRAY_SIZE(sm6115_pp), + .pingpong = sm6115_pp, + .intf_count = ARRAY_SIZE(sm6115_intf), + .intf = sm6115_intf, + .vbif = &sdm845_vbif, + .perf = &sm6115_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h new file mode 100644 index 000000000..dd891703e --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h @@ -0,0 +1,235 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + * Copyright (c) 2023, Linaro Limited + */ + +#ifndef _DPU_6_4_SM6350_H +#define _DPU_6_4_SM6350_H + +static const struct dpu_caps sm6350_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0x7, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .max_linewidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sm6350_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x2bc, .bit_off = 16 }, + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg sm6350_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1dc, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1dc, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1dc, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x1dc, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, +}; + +static const struct dpu_sspp_cfg sm6350_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_SDM845_MASK_NO_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK_NO_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK_NO_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK_NO_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, +}; + +static const struct dpu_lm_cfg sm6350_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sc7180_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sc7180_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = 0, + }, +}; + +static const struct dpu_dspp_cfg sm6350_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static struct dpu_pingpong_cfg sm6350_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, +}; + +static const struct dpu_dsc_cfg sm6350_dsc[] = { + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + }, +}; + +static const struct dpu_wb_cfg sm6350_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .maxlinewidth = 1920, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_intf_cfg sm6350_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 35, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 35, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, +}; + +static const struct dpu_perf_cfg sm6350_perf_data = { + .max_bw_low = 4200000, + .max_bw_high = 5100000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 1600000, + .min_prefill_lines = 35, + /* TODO: confirm danger_lut_tbl */ + .danger_lut_tbl = {0xffff, 0xffff, 0x0}, + .safe_lut_tbl = {0xff00, 0xff00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sm6350_qos_linear_macrotile), + .entries = sm6350_qos_linear_macrotile + }, + {.nentry = ARRAY_SIZE(sm6350_qos_linear_macrotile), + .entries = sm6350_qos_linear_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm6350_mdss_ver = { + .core_major_ver = 6, + .core_minor_ver = 4, +}; + +const struct dpu_mdss_cfg dpu_sm6350_cfg = { + .mdss_ver = &sm6350_mdss_ver, + .caps = &sm6350_dpu_caps, + .mdp = &sm6350_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sm6350_ctl), + .ctl = sm6350_ctl, + .sspp_count = ARRAY_SIZE(sm6350_sspp), + .sspp = sm6350_sspp, + .mixer_count = ARRAY_SIZE(sm6350_lm), + .mixer = sm6350_lm, + .dspp_count = ARRAY_SIZE(sm6350_dspp), + .dspp = sm6350_dspp, + .dsc_count = ARRAY_SIZE(sm6350_dsc), + .dsc = sm6350_dsc, + .pingpong_count = ARRAY_SIZE(sm6350_pp), + .pingpong = sm6350_pp, + .wb_count = ARRAY_SIZE(sm6350_wb), + .wb = sm6350_wb, + .intf_count = ARRAY_SIZE(sm6350_intf), + .intf = sm6350_intf, + .vbif = &sdm845_vbif, + .perf = &sm6350_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h new file mode 100644 index 000000000..7ae7530aa --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h @@ -0,0 +1,144 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_6_5_QCM2290_H +#define _DPU_6_5_QCM2290_H + +static const struct dpu_caps qcm2290_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_LINE_WIDTH, + .max_mixer_blendstages = 0x4, + .has_dim_layer = true, + .has_idle_pc = true, + .max_linewidth = 2160, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg qcm2290_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + }, +}; + +static const struct dpu_ctl_cfg qcm2290_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1dc, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, +}; + +static const struct dpu_sspp_cfg qcm2290_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_QCM2290_MASK, + .sblk = &dpu_vig_sblk_noscale, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK_NO_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, +}; + +static const struct dpu_lm_cfg qcm2290_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .sblk = &qcm2290_lm_sblk, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, +}; + +static const struct dpu_dspp_cfg qcm2290_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg qcm2290_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, +}; + +static const struct dpu_intf_cfg qcm2290_intf[] = { + { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, +}; + +static const struct dpu_perf_cfg qcm2290_perf_data = { + .max_bw_low = 2700000, + .max_bw_high = 2700000, + .min_core_ib = 1300000, + .min_llcc_ib = 0, + .min_dram_ib = 1600000, + .min_prefill_lines = 24, + .danger_lut_tbl = {0xff, 0x0, 0x0}, + .safe_lut_tbl = {0xfff0, 0x0, 0x0}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(qcm2290_qos_linear), + .entries = qcm2290_qos_linear + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version qcm2290_mdss_ver = { + .core_major_ver = 6, + .core_minor_ver = 5, +}; + +const struct dpu_mdss_cfg dpu_qcm2290_cfg = { + .mdss_ver = &qcm2290_mdss_ver, + .caps = &qcm2290_dpu_caps, + .mdp = &qcm2290_mdp, + .ctl_count = ARRAY_SIZE(qcm2290_ctl), + .ctl = qcm2290_ctl, + .sspp_count = ARRAY_SIZE(qcm2290_sspp), + .sspp = qcm2290_sspp, + .mixer_count = ARRAY_SIZE(qcm2290_lm), + .mixer = qcm2290_lm, + .dspp_count = ARRAY_SIZE(qcm2290_dspp), + .dspp = qcm2290_dspp, + .pingpong_count = ARRAY_SIZE(qcm2290_pp), + .pingpong = qcm2290_pp, + .intf_count = ARRAY_SIZE(qcm2290_intf), + .intf = qcm2290_intf, + .vbif = &sdm845_vbif, + .perf = &qcm2290_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h new file mode 100644 index 000000000..fc7ceac85 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h @@ -0,0 +1,162 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + * Copyright (c) 2023, Linaro Limited + */ + +#ifndef _DPU_6_9_SM6375_H +#define _DPU_6_9_SM6375_H + +static const struct dpu_caps sm6375_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_LINE_WIDTH, + .max_mixer_blendstages = 0x4, + .has_dim_layer = true, + .has_idle_pc = true, + .max_linewidth = 2160, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sm6375_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + }, +}; + +static const struct dpu_ctl_cfg sm6375_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1dc, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, +}; + +static const struct dpu_sspp_cfg sm6375_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_SDM845_MASK_NO_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK_NO_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, +}; + +static const struct dpu_lm_cfg sm6375_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .sblk = &qcm2290_lm_sblk, + .lm_pair = 0, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, +}; + +static const struct dpu_dspp_cfg sm6375_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sm6375_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .sblk = &sdm845_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, +}; + +static const struct dpu_dsc_cfg sm6375_dsc[] = { + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + }, +}; + +static const struct dpu_intf_cfg sm6375_intf[] = { + { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, +}; + +static const struct dpu_perf_cfg sm6375_perf_data = { + .max_bw_low = 5200000, + .max_bw_high = 6200000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, /* No LLCC on this SoC */ + .min_dram_ib = 1600000, + .min_prefill_lines = 24, + /* TODO: confirm danger_lut_tbl */ + .danger_lut_tbl = {0xffff, 0xffff, 0x0}, + .safe_lut_tbl = {0xfe00, 0xfe00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sm6350_qos_linear_macrotile), + .entries = sm6350_qos_linear_macrotile + }, + {.nentry = ARRAY_SIZE(sm6350_qos_linear_macrotile), + .entries = sm6350_qos_linear_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm6375_mdss_ver = { + .core_major_ver = 6, + .core_minor_ver = 9, +}; + +const struct dpu_mdss_cfg dpu_sm6375_cfg = { + .mdss_ver = &sm6375_mdss_ver, + .caps = &sm6375_dpu_caps, + .mdp = &sm6375_mdp, + .ctl_count = ARRAY_SIZE(sm6375_ctl), + .ctl = sm6375_ctl, + .sspp_count = ARRAY_SIZE(sm6375_sspp), + .sspp = sm6375_sspp, + .mixer_count = ARRAY_SIZE(sm6375_lm), + .mixer = sm6375_lm, + .dspp_count = ARRAY_SIZE(sm6375_dspp), + .dspp = sm6375_dspp, + .dsc_count = ARRAY_SIZE(sm6375_dsc), + .dsc = sm6375_dsc, + .pingpong_count = ARRAY_SIZE(sm6375_pp), + .pingpong = sm6375_pp, + .intf_count = ARRAY_SIZE(sm6375_intf), + .intf = sm6375_intf, + .vbif = &sdm845_vbif, + .perf = &sm6375_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h new file mode 100644 index 000000000..d06c14f05 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h @@ -0,0 +1,398 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_7_0_SM8350_H +#define _DPU_7_0_SM8350_H + +static const struct dpu_caps sm8350_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 4096, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sm8350_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x2bc, .bit_off = 16 }, + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg sm8350_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x1e8, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x1e8, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x1e8, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x1e8, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x1e8, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x1e8, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sm8350_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1f8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x1f8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x1f8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f8, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, +}; + +static const struct dpu_lm_cfg sm8350_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, +}; + +static const struct dpu_dspp_cfg sm8350_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sm8350_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, +}; + +static const struct dpu_merge_3d_cfg sm8350_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x8, + }, +}; + +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg sm8350_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x4, + .sblk = &dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x4, + .sblk = &dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_1, + }, +}; + +static const struct dpu_wb_cfg sm8350_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_intf_cfg sm8350_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x2c4, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x2c4, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_perf_cfg sm8350_perf_data = { + .max_bw_low = 11800000, + .max_bw_high = 15500000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 800000, + .min_prefill_lines = 40, + /* FIXME: lut tables */ + .danger_lut_tbl = {0x3ffff, 0x3ffff, 0x0}, + .safe_lut_tbl = {0xfe00, 0xfe00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm8350_mdss_ver = { + .core_major_ver = 7, + .core_minor_ver = 0, +}; + +const struct dpu_mdss_cfg dpu_sm8350_cfg = { + .mdss_ver = &sm8350_mdss_ver, + .caps = &sm8350_dpu_caps, + .mdp = &sm8350_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sm8350_ctl), + .ctl = sm8350_ctl, + .sspp_count = ARRAY_SIZE(sm8350_sspp), + .sspp = sm8350_sspp, + .mixer_count = ARRAY_SIZE(sm8350_lm), + .mixer = sm8350_lm, + .dspp_count = ARRAY_SIZE(sm8350_dspp), + .dspp = sm8350_dspp, + .pingpong_count = ARRAY_SIZE(sm8350_pp), + .pingpong = sm8350_pp, + .dsc_count = ARRAY_SIZE(sm8350_dsc), + .dsc = sm8350_dsc, + .merge_3d_count = ARRAY_SIZE(sm8350_merge_3d), + .merge_3d = sm8350_merge_3d, + .wb_count = ARRAY_SIZE(sm8350_wb), + .wb = sm8350_wb, + .intf_count = ARRAY_SIZE(sm8350_intf), + .intf = sm8350_intf, + .vbif = &sdm845_vbif, + .perf = &sm8350_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h new file mode 100644 index 000000000..99b8a890f --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h @@ -0,0 +1,269 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_7_2_SC7280_H +#define _DPU_7_2_SC7280_H + +static const struct dpu_caps sc7280_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0x7, + .has_dim_layer = true, + .has_idle_pc = true, + .max_linewidth = 2400, + .has_3d_merge = true, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sc7280_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x2014, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x2bc, .bit_off = 16 }, + }, +}; + +static const struct dpu_ctl_cfg sc7280_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x1e8, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x1e8, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x1e8, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x1e8, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, +}; + +static const struct dpu_sspp_cfg sc7280_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_SC7280_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0_rot_v2, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, +}; + +static const struct dpu_lm_cfg sc7280_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sc7180_lm_sblk, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sc7180_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &sc7180_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, +}; + +static const struct dpu_dspp_cfg sc7280_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sc7280_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, +}; + +static const struct dpu_merge_3d_cfg sc7280_merge_3d[] = { + { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x8, + }, +}; + +/* NOTE: sc7280 only has one DSC hard slice encoder */ +static const struct dpu_dsc_cfg sc7280_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, +}; + +static const struct dpu_wb_cfg sc7280_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_intf_cfg sc7280_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x2c4, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_5", .id = INTF_5, + .base = 0x39000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23), + }, +}; + +static const struct dpu_perf_cfg sc7280_perf_data = { + .max_bw_low = 4700000, + .max_bw_high = 8800000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 1600000, + .min_prefill_lines = 24, + .danger_lut_tbl = {0xffff, 0xffff, 0x0}, + .safe_lut_tbl = {0xff00, 0xff00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sc7280_mdss_ver = { + .core_major_ver = 7, + .core_minor_ver = 2, +}; + +const struct dpu_mdss_cfg dpu_sc7280_cfg = { + .mdss_ver = &sc7280_mdss_ver, + .caps = &sc7280_dpu_caps, + .mdp = &sc7280_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sc7280_ctl), + .ctl = sc7280_ctl, + .sspp_count = ARRAY_SIZE(sc7280_sspp), + .sspp = sc7280_sspp, + .dspp_count = ARRAY_SIZE(sc7280_dspp), + .dspp = sc7280_dspp, + .mixer_count = ARRAY_SIZE(sc7280_lm), + .mixer = sc7280_lm, + .pingpong_count = ARRAY_SIZE(sc7280_pp), + .pingpong = sc7280_pp, + .merge_3d_count = ARRAY_SIZE(sc7280_merge_3d), + .merge_3d = sc7280_merge_3d, + .dsc_count = ARRAY_SIZE(sc7280_dsc), + .dsc = sc7280_dsc, + .wb_count = ARRAY_SIZE(sc7280_wb), + .wb = sc7280_wb, + .intf_count = ARRAY_SIZE(sc7280_intf), + .intf = sc7280_intf, + .vbif = &sdm845_vbif, + .perf = &sc7280_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h new file mode 100644 index 000000000..0ba907711 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h @@ -0,0 +1,427 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_8_0_SC8280XP_H +#define _DPU_8_0_SC8280XP_H + +static const struct dpu_caps sc8280xp_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 11, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 5120, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sc8280xp_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg sc8280xp_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sc8280xp_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x2ac, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x2ac, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x2ac, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x2ac, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_0, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x2ac, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x2ac, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x2ac, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x2ac, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, +}; + +static const struct dpu_lm_cfg sc8280xp_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, +}; + +static const struct dpu_dspp_cfg sc8280xp_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sc8280xp_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, +}; + +static const struct dpu_merge_3d_cfg sc8280xp_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x8, + }, +}; + +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg sc8280xp_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x4, + .sblk = &dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x4, + .sblk = &dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_1, + }, { + .name = "dce_2_0", .id = DSC_4, + .base = 0x82000, .len = 0x4, + .sblk = &dsc_sblk_0, + }, { + .name = "dce_2_1", .id = DSC_5, + .base = 0x82000, .len = 0x4, + .sblk = &dsc_sblk_1, + }, +}; + +static const struct dpu_intf_cfg sc8280xp_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, { + .name = "intf_4", .id = INTF_4, + .base = 0x38000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 20), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 21), + }, { + .name = "intf_5", .id = INTF_5, + .base = 0x39000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_3, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23), + }, { + .name = "intf_6", .id = INTF_6, + .base = 0x3a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_2, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 16), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 17), + }, { + .name = "intf_7", .id = INTF_7, + .base = 0x3b000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_2, /* pair with intf_6 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 18), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 19), + }, { + .name = "intf_8", .id = INTF_8, + .base = 0x3c000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, /* pair with intf_8 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), + }, +}; + +static const struct dpu_perf_cfg sc8280xp_perf_data = { + .max_bw_low = 13600000, + .max_bw_high = 18200000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 800000, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfe00, 0xfe00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc8180x_qos_linear), + .entries = sc8180x_qos_linear + }, + {.nentry = ARRAY_SIZE(sc8180x_qos_macrotile), + .entries = sc8180x_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sc8280xp_mdss_ver = { + .core_major_ver = 8, + .core_minor_ver = 0, +}; + +const struct dpu_mdss_cfg dpu_sc8280xp_cfg = { + .mdss_ver = &sc8280xp_mdss_ver, + .caps = &sc8280xp_dpu_caps, + .mdp = &sc8280xp_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sc8280xp_ctl), + .ctl = sc8280xp_ctl, + .sspp_count = ARRAY_SIZE(sc8280xp_sspp), + .sspp = sc8280xp_sspp, + .mixer_count = ARRAY_SIZE(sc8280xp_lm), + .mixer = sc8280xp_lm, + .dspp_count = ARRAY_SIZE(sc8280xp_dspp), + .dspp = sc8280xp_dspp, + .pingpong_count = ARRAY_SIZE(sc8280xp_pp), + .pingpong = sc8280xp_pp, + .dsc_count = ARRAY_SIZE(sc8280xp_dsc), + .dsc = sc8280xp_dsc, + .merge_3d_count = ARRAY_SIZE(sc8280xp_merge_3d), + .merge_3d = sc8280xp_merge_3d, + .intf_count = ARRAY_SIZE(sc8280xp_intf), + .intf = sc8280xp_intf, + .vbif = &sdm845_vbif, + .perf = &sc8280xp_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h new file mode 100644 index 000000000..170a709f9 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h @@ -0,0 +1,411 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_8_1_SM8450_H +#define _DPU_8_1_SM8450_H + +static const struct dpu_caps sm8450_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 5120, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sm8450_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x2bc, .bit_off = 16 }, + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg sm8450_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sm8450_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x32c, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_1, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x32c, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_1, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x32c, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_1, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x32c, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_1, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x32c, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x32c, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x32c, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x32c, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, +}; + +static const struct dpu_lm_cfg sm8450_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, +}; + +static const struct dpu_dspp_cfg sm8450_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sm8450_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, { + .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, + .base = 0x65800, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + }, { + .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, + .base = 0x65c00, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + }, +}; + +static const struct dpu_merge_3d_cfg sm8450_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x8, + }, { + .name = "merge_3d_3", .id = MERGE_3D_3, + .base = 0x65f00, .len = 0x8, + }, +}; + +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg sm8450_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x4, + .sblk = &dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x4, + .sblk = &dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_1, + }, +}; + +static const struct dpu_wb_cfg sm8450_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_intf_cfg sm8450_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_perf_cfg sm8450_perf_data = { + .max_bw_low = 13600000, + .max_bw_high = 18200000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 800000, + .min_prefill_lines = 35, + /* FIXME: lut tables */ + .danger_lut_tbl = {0x3ffff, 0x3ffff, 0x0}, + .safe_lut_tbl = {0xfe00, 0xfe00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm8450_mdss_ver = { + .core_major_ver = 8, + .core_minor_ver = 1, +}; + +const struct dpu_mdss_cfg dpu_sm8450_cfg = { + .mdss_ver = &sm8450_mdss_ver, + .caps = &sm8450_dpu_caps, + .mdp = &sm8450_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sm8450_ctl), + .ctl = sm8450_ctl, + .sspp_count = ARRAY_SIZE(sm8450_sspp), + .sspp = sm8450_sspp, + .mixer_count = ARRAY_SIZE(sm8450_lm), + .mixer = sm8450_lm, + .dspp_count = ARRAY_SIZE(sm8450_dspp), + .dspp = sm8450_dspp, + .pingpong_count = ARRAY_SIZE(sm8450_pp), + .pingpong = sm8450_pp, + .dsc_count = ARRAY_SIZE(sm8450_dsc), + .dsc = sm8450_dsc, + .merge_3d_count = ARRAY_SIZE(sm8450_merge_3d), + .merge_3d = sm8450_merge_3d, + .wb_count = ARRAY_SIZE(sm8450_wb), + .wb = sm8450_wb, + .intf_count = ARRAY_SIZE(sm8450_intf), + .intf = sm8450_intf, + .vbif = &sdm845_vbif, + .perf = &sm8450_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h new file mode 100644 index 000000000..9cc5e22e1 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h @@ -0,0 +1,450 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2024 Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#ifndef _DPU_8_4_SA8775P_H +#define _DPU_8_4_SA8775P_H + +static const struct dpu_caps sa8775p_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 5120, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sa8775p_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x2bc, .bit_off = 16 }, + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg sa8775p_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x204, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sa8775p_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x32c, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_1, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x32c, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_1, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x32c, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_1, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x32c, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_1, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x32c, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x32c, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x32c, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x32c, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, +}; + +static const struct dpu_lm_cfg sa8775p_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, +}; + +static const struct dpu_dspp_cfg sa8775p_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sa8775p_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, { + .name = "pingpong_6", .id = PINGPONG_CWB_0, + .base = 0x65800, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + }, { + .name = "pingpong_7", .id = PINGPONG_CWB_1, + .base = 0x65c00, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + }, +}; + +static const struct dpu_merge_3d_cfg sa8775p_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x8, + }, { + .name = "merge_3d_3", .id = MERGE_3D_3, + .base = 0x65f00, .len = 0x8, + }, +}; + +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg sa8775p_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x4, + .sblk = &dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x4, + .sblk = &dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_1, + }, { + .name = "dce_2_0", .id = DSC_4, + .base = 0x82000, .len = 0x4, + .sblk = &dsc_sblk_0, + }, { + .name = "dce_2_1", .id = DSC_5, + .base = 0x82000, .len = 0x4, + .sblk = &dsc_sblk_1, + }, +}; + +static const struct dpu_wb_cfg sa8775p_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_intf_cfg sa8775p_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, { + .name = "intf_4", .id = INTF_4, + .base = 0x38000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 20), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 21), + }, { + .name = "intf_6", .id = INTF_6, + .base = 0x3A000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 16), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 17), + }, { + .name = "intf_7", .id = INTF_7, + .base = 0x3b000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 18), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 19), + }, { + .name = "intf_8", .id = INTF_8, + .base = 0x3c000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, /* pair with intf_4 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), + }, +}; + +static const struct dpu_perf_cfg sa8775p_perf_data = { + .max_bw_low = 13600000, + .max_bw_high = 18200000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 800000, + .min_prefill_lines = 35, + /* FIXME: lut tables */ + .danger_lut_tbl = {0x3ffff, 0x3ffff, 0x0}, + .safe_lut_tbl = {0xfff0, 0xfff0, 0x1}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sm6350_qos_linear_macrotile), + .entries = sm6350_qos_linear_macrotile + }, + {.nentry = ARRAY_SIZE(sm6350_qos_linear_macrotile), + .entries = sm6350_qos_linear_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sa8775p_mdss_ver = { + .core_major_ver = 8, + .core_minor_ver = 4, +}; + +const struct dpu_mdss_cfg dpu_sa8775p_cfg = { + .mdss_ver = &sa8775p_mdss_ver, + .caps = &sa8775p_dpu_caps, + .mdp = &sa8775p_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sa8775p_ctl), + .ctl = sa8775p_ctl, + .sspp_count = ARRAY_SIZE(sa8775p_sspp), + .sspp = sa8775p_sspp, + .mixer_count = ARRAY_SIZE(sa8775p_lm), + .mixer = sa8775p_lm, + .dspp_count = ARRAY_SIZE(sa8775p_dspp), + .dspp = sa8775p_dspp, + .pingpong_count = ARRAY_SIZE(sa8775p_pp), + .pingpong = sa8775p_pp, + .dsc_count = ARRAY_SIZE(sa8775p_dsc), + .dsc = sa8775p_dsc, + .merge_3d_count = ARRAY_SIZE(sa8775p_merge_3d), + .merge_3d = sa8775p_merge_3d, + .wb_count = ARRAY_SIZE(sa8775p_wb), + .wb = sa8775p_wb, + .intf_count = ARRAY_SIZE(sa8775p_intf), + .intf = sa8775p_intf, + .vbif = &sdm845_vbif, + .perf = &sa8775p_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h new file mode 100644 index 000000000..b3d0f2218 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h @@ -0,0 +1,406 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_9_0_SM8550_H +#define _DPU_9_0_SM8550_H + +static const struct dpu_caps sm8550_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 5120, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sm8550_mdp = { + .name = "top_0", + .base = 0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg sm8550_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sm8550_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_2, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_2, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_2, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_2, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_12", .id = SSPP_DMA4, + .base = 0x2c000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 14, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_13", .id = SSPP_DMA5, + .base = 0x2e000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 15, + .type = SSPP_TYPE_DMA, + }, +}; + +static const struct dpu_lm_cfg sm8550_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, +}; + +static const struct dpu_dspp_cfg sm8550_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; +static const struct dpu_pingpong_cfg sm8550_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, { + .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, + .base = 0x66000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + }, { + .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, + .base = 0x66400, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + }, +}; + +static const struct dpu_merge_3d_cfg sm8550_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x8, + }, { + .name = "merge_3d_3", .id = MERGE_3D_3, + .base = 0x66700, .len = 0x8, + }, +}; + +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg sm8550_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x4, + .sblk = &dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x4, + .sblk = &dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_1, + }, +}; + +static const struct dpu_wb_cfg sm8550_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_intf_cfg sm8550_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_perf_cfg sm8550_perf_data = { + .max_bw_low = 13600000, + .max_bw_high = 18200000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 800000, + .min_prefill_lines = 35, + /* FIXME: lut tables */ + .danger_lut_tbl = {0x3ffff, 0x3ffff, 0x0}, + .safe_lut_tbl = {0xfe00, 0xfe00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm8550_mdss_ver = { + .core_major_ver = 9, + .core_minor_ver = 0, +}; + +const struct dpu_mdss_cfg dpu_sm8550_cfg = { + .mdss_ver = &sm8550_mdss_ver, + .caps = &sm8550_dpu_caps, + .mdp = &sm8550_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sm8550_ctl), + .ctl = sm8550_ctl, + .sspp_count = ARRAY_SIZE(sm8550_sspp), + .sspp = sm8550_sspp, + .mixer_count = ARRAY_SIZE(sm8550_lm), + .mixer = sm8550_lm, + .dspp_count = ARRAY_SIZE(sm8550_dspp), + .dspp = sm8550_dspp, + .pingpong_count = ARRAY_SIZE(sm8550_pp), + .pingpong = sm8550_pp, + .dsc_count = ARRAY_SIZE(sm8550_dsc), + .dsc = sm8550_dsc, + .merge_3d_count = ARRAY_SIZE(sm8550_merge_3d), + .merge_3d = sm8550_merge_3d, + .wb_count = ARRAY_SIZE(sm8550_wb), + .wb = sm8550_wb, + .intf_count = ARRAY_SIZE(sm8550_intf), + .intf = sm8550_intf, + .vbif = &sm8550_vbif, + .perf = &sm8550_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h new file mode 100644 index 000000000..7effe7120 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h @@ -0,0 +1,406 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_9_1_SAR2130P_H +#define _DPU_9_1_SAR2130P_H + +static const struct dpu_caps sar2130p_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 5120, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sar2130p_mdp = { + .name = "top_0", + .base = 0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg sar2130p_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sar2130p_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_2, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_2, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_2, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_2, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_12", .id = SSPP_DMA4, + .base = 0x2c000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 14, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_13", .id = SSPP_DMA5, + .base = 0x2e000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 15, + .type = SSPP_TYPE_DMA, + }, +}; + +static const struct dpu_lm_cfg sar2130p_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, +}; + +static const struct dpu_dspp_cfg sar2130p_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; +static const struct dpu_pingpong_cfg sar2130p_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, { + .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, + .base = 0x66000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + }, { + .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, + .base = 0x66400, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + }, +}; + +static const struct dpu_merge_3d_cfg sar2130p_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x8, + }, { + .name = "merge_3d_3", .id = MERGE_3D_3, + .base = 0x66700, .len = 0x8, + }, +}; + +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg sar2130p_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x4, + .sblk = &dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x4, + .sblk = &dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_1, + }, +}; + +static const struct dpu_wb_cfg sar2130p_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_intf_cfg sar2130p_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_perf_cfg sar2130p_perf_data = { + .max_bw_low = 13600000, + .max_bw_high = 18200000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 800000, + .min_prefill_lines = 35, + /* FIXME: lut tables */ + .danger_lut_tbl = {0x3ffff, 0x3ffff, 0x0}, + .safe_lut_tbl = {0xfe00, 0xfe00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 0, .wr_enable = 0}, + {.rd_enable = 0, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sar2130p_mdss_ver = { + .core_major_ver = 9, + .core_minor_ver = 1, +}; + +const struct dpu_mdss_cfg dpu_sar2130p_cfg = { + .mdss_ver = &sar2130p_mdss_ver, + .caps = &sar2130p_dpu_caps, + .mdp = &sar2130p_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sar2130p_ctl), + .ctl = sar2130p_ctl, + .sspp_count = ARRAY_SIZE(sar2130p_sspp), + .sspp = sar2130p_sspp, + .mixer_count = ARRAY_SIZE(sar2130p_lm), + .mixer = sar2130p_lm, + .dspp_count = ARRAY_SIZE(sar2130p_dspp), + .dspp = sar2130p_dspp, + .pingpong_count = ARRAY_SIZE(sar2130p_pp), + .pingpong = sar2130p_pp, + .dsc_count = ARRAY_SIZE(sar2130p_dsc), + .dsc = sar2130p_dsc, + .merge_3d_count = ARRAY_SIZE(sar2130p_merge_3d), + .merge_3d = sar2130p_merge_3d, + .wb_count = ARRAY_SIZE(sar2130p_wb), + .wb = sar2130p_wb, + .intf_count = ARRAY_SIZE(sar2130p_intf), + .intf = sar2130p_intf, + .vbif = &sm8550_vbif, + .perf = &sar2130p_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h new file mode 100644 index 000000000..7b3febf87 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h @@ -0,0 +1,446 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2023, Linaro Limited + */ + +#ifndef _DPU_9_2_X1E80100_H +#define _DPU_9_2_X1E80100_H + +static const struct dpu_caps x1e80100_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 5120, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg x1e80100_mdp = { + .name = "top_0", + .base = 0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg x1e80100_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x290, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg x1e80100_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_3, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_3, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_3, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_3, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_12", .id = SSPP_DMA4, + .base = 0x2c000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 14, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_13", .id = SSPP_DMA5, + .base = 0x2e000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 15, + .type = SSPP_TYPE_DMA, + }, +}; + +static const struct dpu_lm_cfg x1e80100_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, +}; + +static const struct dpu_dspp_cfg x1e80100_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg x1e80100_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, { + .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, + .base = 0x66000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + }, { + .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, + .base = 0x66400, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + }, +}; + +static const struct dpu_merge_3d_cfg x1e80100_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x8, + }, { + .name = "merge_3d_3", .id = MERGE_3D_3, + .base = 0x66700, .len = 0x8, + }, +}; + +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg x1e80100_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x4, + .sblk = &dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x4, + .sblk = &dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_1, + }, +}; + +static const struct dpu_wb_cfg x1e80100_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .xin_id = 6, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_intf_cfg x1e80100_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x300, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, { + .name = "intf_4", .id = INTF_4, + .base = 0x38000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 20), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 21), + }, { + .name = "intf_5", .id = INTF_5, + .base = 0x39000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_3, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23), + }, { + .name = "intf_6", .id = INTF_6, + .base = 0x3A000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_2, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 16), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 17), + }, { + .name = "intf_7", .id = INTF_7, + .base = 0x3b000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_2, /* pair with intf_6 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 18), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 19), + }, { + .name = "intf_8", .id = INTF_8, + .base = 0x3c000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, /* pair with intf_4 for DP MST */ + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), + }, +}; + +static const struct dpu_perf_cfg x1e80100_perf_data = { + .max_bw_low = 13600000, + .max_bw_high = 18200000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 800000, + .min_prefill_lines = 35, + /* FIXME: lut tables */ + .danger_lut_tbl = {0x3ffff, 0x3ffff, 0x0}, + .safe_lut_tbl = {0xfe00, 0xfe00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version x1e80100_mdss_ver = { + .core_major_ver = 9, + .core_minor_ver = 2, +}; + +const struct dpu_mdss_cfg dpu_x1e80100_cfg = { + .mdss_ver = &x1e80100_mdss_ver, + .caps = &x1e80100_dpu_caps, + .mdp = &x1e80100_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(x1e80100_ctl), + .ctl = x1e80100_ctl, + .sspp_count = ARRAY_SIZE(x1e80100_sspp), + .sspp = x1e80100_sspp, + .mixer_count = ARRAY_SIZE(x1e80100_lm), + .mixer = x1e80100_lm, + .dspp_count = ARRAY_SIZE(x1e80100_dspp), + .dspp = x1e80100_dspp, + .pingpong_count = ARRAY_SIZE(x1e80100_pp), + .pingpong = x1e80100_pp, + .dsc_count = ARRAY_SIZE(x1e80100_dsc), + .dsc = x1e80100_dsc, + .merge_3d_count = ARRAY_SIZE(x1e80100_merge_3d), + .merge_3d = x1e80100_merge_3d, + .wb_count = ARRAY_SIZE(x1e80100_wb), + .wb = x1e80100_wb, + .intf_count = ARRAY_SIZE(x1e80100_intf), + .intf = x1e80100_intf, + .vbif = &sm8550_vbif, + .perf = &x1e80100_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.h new file mode 100644 index 000000000..e7183cf05 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.h @@ -0,0 +1,34 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef __DPU_CORE_IRQ_H__ +#define __DPU_CORE_IRQ_H__ + +#include "dpu_kms.h" +#include "dpu_hw_interrupts.h" + +void dpu_core_irq_preinstall(struct msm_kms *kms); + +void dpu_core_irq_uninstall(struct msm_kms *kms); + +irqreturn_t dpu_core_irq(struct msm_kms *kms); + +u32 dpu_core_irq_read( + struct dpu_kms *dpu_kms, + unsigned int irq_idx); + +int dpu_core_irq_register_callback( + struct dpu_kms *dpu_kms, + unsigned int irq_idx, + void (*irq_cb)(void *arg), + void *irq_arg); + +int dpu_core_irq_unregister_callback( + struct dpu_kms *dpu_kms, + unsigned int irq_idx); + +void dpu_debugfs_core_irq_init(struct dpu_kms *dpu_kms, + struct dentry *parent); + +#endif /* __DPU_CORE_IRQ_H__ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c new file mode 100644 index 000000000..fea173e37 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c @@ -0,0 +1,519 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2016-2018, The Linux Foundation. All rights reserved. + */ + +#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ + +#include +#include +#include +#include +#include +#include +#include + +#include "dpu_kms.h" +#include "dpu_trace.h" +#include "dpu_crtc.h" +#include "dpu_core_perf.h" + +/** + * enum dpu_perf_mode - performance tuning mode + * @DPU_PERF_MODE_NORMAL: performance controlled by user mode client + * @DPU_PERF_MODE_MINIMUM: performance bounded by minimum setting + * @DPU_PERF_MODE_FIXED: performance bounded by fixed setting + * @DPU_PERF_MODE_MAX: maximum value, used for error checking + */ +enum dpu_perf_mode { + DPU_PERF_MODE_NORMAL, + DPU_PERF_MODE_MINIMUM, + DPU_PERF_MODE_FIXED, + DPU_PERF_MODE_MAX +}; + +/** + * dpu_core_perf_adjusted_mode_clk - Adjust given mode clock rate according to + * the perf clock factor. + * @mode_clk_rate: unadjusted mode clock rate + * @perf_cfg: performance configuration + */ +u64 dpu_core_perf_adjusted_mode_clk(u64 mode_clk_rate, + const struct dpu_perf_cfg *perf_cfg) +{ + u32 clk_factor; + + clk_factor = perf_cfg->clk_inefficiency_factor; + if (clk_factor) { + mode_clk_rate *= clk_factor; + do_div(mode_clk_rate, 100); + } + + return mode_clk_rate; +} + +/** + * _dpu_core_perf_calc_bw() - to calculate BW per crtc + * @perf_cfg: performance configuration + * @crtc: pointer to a crtc + * Return: returns aggregated BW for all planes in crtc. + */ +static u64 _dpu_core_perf_calc_bw(const struct dpu_perf_cfg *perf_cfg, + struct drm_crtc *crtc) +{ + struct drm_plane *plane; + struct dpu_plane_state *pstate; + u64 crtc_plane_bw = 0; + u32 bw_factor; + + drm_atomic_crtc_for_each_plane(plane, crtc) { + pstate = to_dpu_plane_state(plane->state); + if (!pstate) + continue; + + crtc_plane_bw += pstate->plane_fetch_bw; + } + + bw_factor = perf_cfg->bw_inefficiency_factor; + if (bw_factor) { + crtc_plane_bw *= bw_factor; + do_div(crtc_plane_bw, 100); + } + + return crtc_plane_bw; +} + +/** + * _dpu_core_perf_calc_clk() - to calculate clock per crtc + * @perf_cfg: performance configuration + * @crtc: pointer to a crtc + * @state: pointer to a crtc state + * Return: returns max clk for all planes in crtc. + */ +static u64 _dpu_core_perf_calc_clk(const struct dpu_perf_cfg *perf_cfg, + struct drm_crtc *crtc, struct drm_crtc_state *state) +{ + struct drm_plane *plane; + struct dpu_plane_state *pstate; + struct drm_display_mode *mode; + u64 mode_clk; + + mode = &state->adjusted_mode; + + mode_clk = (u64)mode->vtotal * mode->hdisplay * drm_mode_vrefresh(mode); + + drm_atomic_crtc_for_each_plane(plane, crtc) { + pstate = to_dpu_plane_state(plane->state); + if (!pstate) + continue; + + mode_clk = max(pstate->plane_clk, mode_clk); + } + + return dpu_core_perf_adjusted_mode_clk(mode_clk, perf_cfg); +} + +static struct dpu_kms *_dpu_crtc_get_kms(struct drm_crtc *crtc) +{ + struct msm_drm_private *priv; + priv = crtc->dev->dev_private; + return to_dpu_kms(priv->kms); +} + +static void _dpu_core_perf_calc_crtc(const struct dpu_core_perf *core_perf, + struct drm_crtc *crtc, + struct drm_crtc_state *state, + struct dpu_core_perf_params *perf) +{ + const struct dpu_perf_cfg *perf_cfg = core_perf->perf_cfg; + + if (!perf_cfg || !crtc || !state || !perf) { + DPU_ERROR("invalid parameters\n"); + return; + } + + perf->bw_ctl = _dpu_core_perf_calc_bw(perf_cfg, crtc); + perf->max_per_pipe_ib = perf_cfg->min_dram_ib; + perf->core_clk_rate = _dpu_core_perf_calc_clk(perf_cfg, crtc, state); + DRM_DEBUG_ATOMIC( + "crtc=%d clk_rate=%llu core_ib=%u core_ab=%u\n", + crtc->base.id, perf->core_clk_rate, + perf->max_per_pipe_ib, + (u32)DIV_ROUND_UP_ULL(perf->bw_ctl, 1000)); +} + +static void dpu_core_perf_aggregate(struct drm_device *ddev, + enum dpu_crtc_client_type curr_client_type, + struct dpu_core_perf_params *perf) +{ + struct dpu_crtc_state *dpu_cstate; + struct drm_crtc *tmp_crtc; + + drm_for_each_crtc(tmp_crtc, ddev) { + if (tmp_crtc->enabled && + curr_client_type == dpu_crtc_get_client_type(tmp_crtc)) { + dpu_cstate = to_dpu_crtc_state(tmp_crtc->state); + + perf->max_per_pipe_ib = max(perf->max_per_pipe_ib, + dpu_cstate->new_perf.max_per_pipe_ib); + + perf->bw_ctl += dpu_cstate->new_perf.bw_ctl; + + DRM_DEBUG_ATOMIC("crtc=%d bw=%llu\n", + tmp_crtc->base.id, + dpu_cstate->new_perf.bw_ctl); + } + } +} + +/** + * dpu_core_perf_crtc_check - validate performance of the given crtc state + * @crtc: Pointer to crtc + * @state: Pointer to new crtc state + * return: zero if success, or error code otherwise + */ +int dpu_core_perf_crtc_check(struct drm_crtc *crtc, + struct drm_crtc_state *state) +{ + u32 bw, threshold; + struct dpu_crtc_state *dpu_cstate; + struct dpu_kms *kms; + struct dpu_core_perf_params perf = { 0 }; + + if (!crtc || !state) { + DPU_ERROR("invalid crtc\n"); + return -EINVAL; + } + + kms = _dpu_crtc_get_kms(crtc); + + /* we only need bandwidth check on real-time clients (interfaces) */ + if (dpu_crtc_get_client_type(crtc) == NRT_CLIENT) + return 0; + + dpu_cstate = to_dpu_crtc_state(state); + + /* obtain new values */ + _dpu_core_perf_calc_crtc(&kms->perf, crtc, state, &dpu_cstate->new_perf); + + dpu_core_perf_aggregate(crtc->dev, dpu_crtc_get_client_type(crtc), &perf); + + /* convert bandwidth to kb */ + bw = DIV_ROUND_UP_ULL(perf.bw_ctl, 1000); + DRM_DEBUG_ATOMIC("calculated bandwidth=%uk\n", bw); + + threshold = kms->perf.perf_cfg->max_bw_high; + + DRM_DEBUG_ATOMIC("final threshold bw limit = %d\n", threshold); + + if (!threshold) { + DPU_ERROR("no bandwidth limits specified\n"); + return -E2BIG; + } else if (bw > threshold) { + DPU_ERROR("exceeds bandwidth: %ukb > %ukb\n", bw, + threshold); + return -E2BIG; + } + + return 0; +} + +static int _dpu_core_perf_crtc_update_bus(struct dpu_kms *kms, + struct drm_crtc *crtc) +{ + struct dpu_core_perf_params perf = { 0 }; + int i, ret = 0; + u32 avg_bw; + u32 peak_bw; + + if (!kms->num_paths) + return 0; + + if (kms->perf.perf_tune.mode == DPU_PERF_MODE_MINIMUM) { + avg_bw = 0; + peak_bw = 0; + } else if (kms->perf.perf_tune.mode == DPU_PERF_MODE_FIXED) { + avg_bw = kms->perf.fix_core_ab_vote; + peak_bw = kms->perf.fix_core_ib_vote; + } else { + dpu_core_perf_aggregate(crtc->dev, dpu_crtc_get_client_type(crtc), &perf); + + avg_bw = div_u64(perf.bw_ctl, 1000); /*Bps_to_icc*/ + peak_bw = perf.max_per_pipe_ib; + } + + avg_bw /= kms->num_paths; + + for (i = 0; i < kms->num_paths; i++) + icc_set_bw(kms->path[i], avg_bw, peak_bw); + + return ret; +} + +/** + * dpu_core_perf_crtc_release_bw() - request zero bandwidth + * @crtc: pointer to a crtc + * + * Function checks a state variable for the crtc, if all pending commit + * requests are done, meaning no more bandwidth is needed, release + * bandwidth request. + */ +void dpu_core_perf_crtc_release_bw(struct drm_crtc *crtc) +{ + struct dpu_crtc *dpu_crtc; + struct dpu_kms *kms; + + if (!crtc) { + DPU_ERROR("invalid crtc\n"); + return; + } + + kms = _dpu_crtc_get_kms(crtc); + dpu_crtc = to_dpu_crtc(crtc); + + if (atomic_dec_return(&kms->bandwidth_ref) > 0) + return; + + /* Release the bandwidth */ + if (kms->perf.enable_bw_release) { + trace_dpu_cmd_release_bw(crtc->base.id); + DRM_DEBUG_ATOMIC("Release BW crtc=%d\n", crtc->base.id); + dpu_crtc->cur_perf.bw_ctl = 0; + _dpu_core_perf_crtc_update_bus(kms, crtc); + } +} + +static u64 _dpu_core_perf_get_core_clk_rate(struct dpu_kms *kms) +{ + u64 clk_rate; + struct drm_crtc *crtc; + struct dpu_crtc_state *dpu_cstate; + + if (kms->perf.perf_tune.mode == DPU_PERF_MODE_FIXED) + return kms->perf.fix_core_clk_rate; + + if (kms->perf.perf_tune.mode == DPU_PERF_MODE_MINIMUM) + return kms->perf.max_core_clk_rate; + + clk_rate = 0; + drm_for_each_crtc(crtc, kms->dev) { + if (crtc->enabled) { + dpu_cstate = to_dpu_crtc_state(crtc->state); + clk_rate = max(dpu_cstate->new_perf.core_clk_rate, + clk_rate); + } + } + + return clk_rate; +} + +/** + * dpu_core_perf_crtc_update - update performance of the given crtc + * @crtc: Pointer to crtc + * @params_changed: true if crtc parameters are modified + * return: zero if success, or error code otherwise + */ +int dpu_core_perf_crtc_update(struct drm_crtc *crtc, + int params_changed) +{ + struct dpu_core_perf_params *new, *old; + bool update_bus = false, update_clk = false; + u64 clk_rate = 0; + struct dpu_crtc *dpu_crtc; + struct dpu_crtc_state *dpu_cstate; + struct dpu_kms *kms; + int ret; + + if (!crtc) { + DPU_ERROR("invalid crtc\n"); + return -EINVAL; + } + + kms = _dpu_crtc_get_kms(crtc); + + dpu_crtc = to_dpu_crtc(crtc); + dpu_cstate = to_dpu_crtc_state(crtc->state); + + DRM_DEBUG_ATOMIC("crtc:%d enabled:%d core_clk:%llu\n", + crtc->base.id, crtc->enabled, kms->perf.core_clk_rate); + + old = &dpu_crtc->cur_perf; + new = &dpu_cstate->new_perf; + + if (crtc->enabled) { + /* + * cases for bus bandwidth update. + * 1. new bandwidth vote - "ab or ib vote" is higher + * than current vote for update request. + * 2. new bandwidth vote - "ab or ib vote" is lower + * than current vote at end of commit or stop. + */ + if ((params_changed && ((new->bw_ctl > old->bw_ctl) || + (new->max_per_pipe_ib > old->max_per_pipe_ib))) || + (!params_changed && ((new->bw_ctl < old->bw_ctl) || + (new->max_per_pipe_ib < old->max_per_pipe_ib)))) { + DRM_DEBUG_ATOMIC("crtc=%d p=%d new_bw=%llu,old_bw=%llu\n", + crtc->base.id, params_changed, + new->bw_ctl, old->bw_ctl); + old->bw_ctl = new->bw_ctl; + old->max_per_pipe_ib = new->max_per_pipe_ib; + update_bus = true; + } + + if ((params_changed && new->core_clk_rate > old->core_clk_rate) || + (!params_changed && new->core_clk_rate < old->core_clk_rate)) { + old->core_clk_rate = new->core_clk_rate; + update_clk = true; + } + } else { + DRM_DEBUG_ATOMIC("crtc=%d disable\n", crtc->base.id); + memset(old, 0, sizeof(*old)); + update_bus = true; + update_clk = true; + } + + trace_dpu_perf_crtc_update(crtc->base.id, new->bw_ctl, + new->core_clk_rate, !crtc->enabled, update_bus, update_clk); + + if (update_bus) { + ret = _dpu_core_perf_crtc_update_bus(kms, crtc); + if (ret) { + DPU_ERROR("crtc-%d: failed to update bus bw vote\n", + crtc->base.id); + return ret; + } + } + + /* + * Update the clock after bandwidth vote to ensure + * bandwidth is available before clock rate is increased. + */ + if (update_clk) { + clk_rate = _dpu_core_perf_get_core_clk_rate(kms); + + DRM_DEBUG_ATOMIC("clk:%llu\n", clk_rate); + + trace_dpu_core_perf_update_clk(kms->dev, !crtc->enabled, clk_rate); + + /* If we're going offline, PM callbacks will disable the clocks instead */ + if (!clk_rate) + return 0; + + clk_rate = min(clk_rate, kms->perf.max_core_clk_rate); + ret = dev_pm_opp_set_rate(&kms->pdev->dev, clk_rate); + if (ret) { + DPU_ERROR("failed to set core clock rate %llu\n", clk_rate); + return ret; + } + + kms->perf.core_clk_rate = clk_rate; + DRM_DEBUG_ATOMIC("update clk rate = %lld HZ\n", clk_rate); + } + return 0; +} + +#ifdef CONFIG_DEBUG_FS + +static ssize_t _dpu_core_perf_mode_write(struct file *file, + const char __user *user_buf, size_t count, loff_t *ppos) +{ + struct dpu_core_perf *perf = file->private_data; + u32 perf_mode = 0; + int ret; + + ret = kstrtouint_from_user(user_buf, count, 0, &perf_mode); + if (ret) + return ret; + + if (perf_mode >= DPU_PERF_MODE_MAX) + return -EINVAL; + + if (perf_mode == DPU_PERF_MODE_FIXED) { + DRM_INFO("fix performance mode\n"); + } else if (perf_mode == DPU_PERF_MODE_MINIMUM) { + /* run the driver with max clk and BW vote */ + DRM_INFO("minimum performance mode\n"); + } else if (perf_mode == DPU_PERF_MODE_NORMAL) { + /* reset the perf tune params to 0 */ + DRM_INFO("normal performance mode\n"); + } + perf->perf_tune.mode = perf_mode; + + return count; +} + +static ssize_t _dpu_core_perf_mode_read(struct file *file, + char __user *buff, size_t count, loff_t *ppos) +{ + struct dpu_core_perf *perf = file->private_data; + int len; + char buf[128]; + + len = scnprintf(buf, sizeof(buf), + "mode %d\n", + perf->perf_tune.mode); + + return simple_read_from_buffer(buff, count, ppos, buf, len); +} + +static const struct file_operations dpu_core_perf_mode_fops = { + .open = simple_open, + .read = _dpu_core_perf_mode_read, + .write = _dpu_core_perf_mode_write, +}; + +/** + * dpu_core_perf_debugfs_init - initialize debugfs for core performance context + * @dpu_kms: Pointer to the dpu_kms struct + * @parent: Pointer to parent debugfs + */ +int dpu_core_perf_debugfs_init(struct dpu_kms *dpu_kms, struct dentry *parent) +{ + struct dpu_core_perf *perf = &dpu_kms->perf; + struct dentry *entry; + + entry = debugfs_create_dir("core_perf", parent); + + debugfs_create_u64("max_core_clk_rate", 0600, entry, + &perf->max_core_clk_rate); + debugfs_create_u64("core_clk_rate", 0600, entry, + &perf->core_clk_rate); + debugfs_create_u32("enable_bw_release", 0600, entry, + (u32 *)&perf->enable_bw_release); + debugfs_create_u32("low_core_ab", 0400, entry, + (u32 *)&perf->perf_cfg->max_bw_low); + debugfs_create_u32("max_core_ab", 0400, entry, + (u32 *)&perf->perf_cfg->max_bw_high); + debugfs_create_u32("min_core_ib", 0400, entry, + (u32 *)&perf->perf_cfg->min_core_ib); + debugfs_create_u32("min_llcc_ib", 0400, entry, + (u32 *)&perf->perf_cfg->min_llcc_ib); + debugfs_create_u32("min_dram_ib", 0400, entry, + (u32 *)&perf->perf_cfg->min_dram_ib); + debugfs_create_file("perf_mode", 0600, entry, + (u32 *)perf, &dpu_core_perf_mode_fops); + debugfs_create_u64("fix_core_clk_rate", 0600, entry, + &perf->fix_core_clk_rate); + debugfs_create_u32("fix_core_ib_vote", 0600, entry, + &perf->fix_core_ib_vote); + debugfs_create_u32("fix_core_ab_vote", 0600, entry, + &perf->fix_core_ab_vote); + + return 0; +} +#endif + +/** + * dpu_core_perf_init - initialize the given core performance context + * @perf: Pointer to core performance context + * @perf_cfg: Pointer to platform performance configuration + * @max_core_clk_rate: Maximum core clock rate + */ +int dpu_core_perf_init(struct dpu_core_perf *perf, + const struct dpu_perf_cfg *perf_cfg, + unsigned long max_core_clk_rate) +{ + perf->perf_cfg = perf_cfg; + perf->max_core_clk_rate = max_core_clk_rate; + + return 0; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h new file mode 100644 index 000000000..3740bc974 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h @@ -0,0 +1,76 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2016-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_CORE_PERF_H_ +#define _DPU_CORE_PERF_H_ + +#include +#include +#include +#include + +#include "dpu_hw_catalog.h" + +/** + * struct dpu_core_perf_params - definition of performance parameters + * @max_per_pipe_ib: maximum instantaneous bandwidth request + * @bw_ctl: arbitrated bandwidth request + * @core_clk_rate: core clock rate request + */ +struct dpu_core_perf_params { + u32 max_per_pipe_ib; + u64 bw_ctl; + u64 core_clk_rate; +}; + +/** + * struct dpu_core_perf_tune - definition of performance tuning control + * @mode: performance mode + */ +struct dpu_core_perf_tune { + u32 mode; +}; + +/** + * struct dpu_core_perf - definition of core performance context + * @perf_cfg: Platform-specific performance configuration + * @core_clk_rate: current core clock rate + * @max_core_clk_rate: maximum allowable core clock rate + * @perf_tune: debug control for performance tuning + * @enable_bw_release: debug control for bandwidth release + * @fix_core_clk_rate: fixed core clock request in Hz used in mode 2 + * @fix_core_ib_vote: fixed core ib vote in KBps used in mode 2 + * @fix_core_ab_vote: fixed core ab vote in KBps used in mode 2 + */ +struct dpu_core_perf { + const struct dpu_perf_cfg *perf_cfg; + u64 core_clk_rate; + u64 max_core_clk_rate; + struct dpu_core_perf_tune perf_tune; + u32 enable_bw_release; + u64 fix_core_clk_rate; + u32 fix_core_ib_vote; + u32 fix_core_ab_vote; +}; + +u64 dpu_core_perf_adjusted_mode_clk(u64 clk_rate, + const struct dpu_perf_cfg *perf_cfg); + +int dpu_core_perf_crtc_check(struct drm_crtc *crtc, + struct drm_crtc_state *state); + +int dpu_core_perf_crtc_update(struct drm_crtc *crtc, + int params_changed); + +void dpu_core_perf_crtc_release_bw(struct drm_crtc *crtc); + +int dpu_core_perf_init(struct dpu_core_perf *perf, + const struct dpu_perf_cfg *perf_cfg, + unsigned long max_core_clk_rate); + +struct dpu_kms; + +int dpu_core_perf_debugfs_init(struct dpu_kms *dpu_kms, struct dentry *parent); + +#endif /* _DPU_CORE_PERF_H_ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c new file mode 100644 index 000000000..42d0a529b --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c @@ -0,0 +1,1941 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2022-2023 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2014-2021 The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "dpu_kms.h" +#include "dpu_hw_lm.h" +#include "dpu_hw_ctl.h" +#include "dpu_hw_dspp.h" +#include "dpu_crtc.h" +#include "dpu_plane.h" +#include "dpu_encoder.h" +#include "dpu_vbif.h" +#include "dpu_core_perf.h" +#include "dpu_trace.h" + +/* layer mixer index on dpu_crtc */ +#define LEFT_MIXER 0 +#define RIGHT_MIXER 1 + +/* timeout in ms waiting for frame done */ +#define DPU_CRTC_FRAME_DONE_TIMEOUT_MS 60 + +#define CONVERT_S3_15(val) \ + (((((u64)val) & ~BIT_ULL(63)) >> 17) & GENMASK_ULL(17, 0)) + +static struct dpu_kms *_dpu_crtc_get_kms(struct drm_crtc *crtc) +{ + struct msm_drm_private *priv = crtc->dev->dev_private; + + return to_dpu_kms(priv->kms); +} + +static struct drm_encoder *get_encoder_from_crtc(struct drm_crtc *crtc) +{ + struct drm_device *dev = crtc->dev; + struct drm_encoder *encoder; + + drm_for_each_encoder(encoder, dev) + if (encoder->crtc == crtc) + return encoder; + + return NULL; +} + +static enum dpu_crtc_crc_source dpu_crtc_parse_crc_source(const char *src_name) +{ + if (!src_name || + !strcmp(src_name, "none")) + return DPU_CRTC_CRC_SOURCE_NONE; + if (!strcmp(src_name, "auto") || + !strcmp(src_name, "lm")) + return DPU_CRTC_CRC_SOURCE_LAYER_MIXER; + if (!strcmp(src_name, "encoder")) + return DPU_CRTC_CRC_SOURCE_ENCODER; + + return DPU_CRTC_CRC_SOURCE_INVALID; +} + +static int dpu_crtc_verify_crc_source(struct drm_crtc *crtc, + const char *src_name, size_t *values_cnt) +{ + enum dpu_crtc_crc_source source = dpu_crtc_parse_crc_source(src_name); + struct dpu_crtc_state *crtc_state = to_dpu_crtc_state(crtc->state); + + if (source < 0) { + DRM_DEBUG_DRIVER("Invalid source %s for CRTC%d\n", src_name, crtc->index); + return -EINVAL; + } + + if (source == DPU_CRTC_CRC_SOURCE_LAYER_MIXER) { + *values_cnt = crtc_state->num_mixers; + } else if (source == DPU_CRTC_CRC_SOURCE_ENCODER) { + struct drm_encoder *drm_enc; + + *values_cnt = 0; + + drm_for_each_encoder_mask(drm_enc, crtc->dev, crtc->state->encoder_mask) + *values_cnt += dpu_encoder_get_crc_values_cnt(drm_enc); + } + + return 0; +} + +static void dpu_crtc_setup_lm_misr(struct dpu_crtc_state *crtc_state) +{ + struct dpu_crtc_mixer *m; + int i; + + for (i = 0; i < crtc_state->num_mixers; ++i) { + m = &crtc_state->mixers[i]; + + if (!m->hw_lm || !m->hw_lm->ops.setup_misr) + continue; + + /* Calculate MISR over 1 frame */ + m->hw_lm->ops.setup_misr(m->hw_lm); + } +} + +static void dpu_crtc_setup_encoder_misr(struct drm_crtc *crtc) +{ + struct drm_encoder *drm_enc; + + drm_for_each_encoder_mask(drm_enc, crtc->dev, crtc->state->encoder_mask) + dpu_encoder_setup_misr(drm_enc); +} + +static int dpu_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name) +{ + enum dpu_crtc_crc_source source = dpu_crtc_parse_crc_source(src_name); + enum dpu_crtc_crc_source current_source; + struct dpu_crtc_state *crtc_state; + struct drm_device *drm_dev = crtc->dev; + + bool was_enabled; + bool enable = false; + int ret = 0; + + if (source < 0) { + DRM_DEBUG_DRIVER("Invalid CRC source %s for CRTC%d\n", src_name, crtc->index); + return -EINVAL; + } + + ret = drm_modeset_lock(&crtc->mutex, NULL); + + if (ret) + return ret; + + enable = (source != DPU_CRTC_CRC_SOURCE_NONE); + crtc_state = to_dpu_crtc_state(crtc->state); + + spin_lock_irq(&drm_dev->event_lock); + current_source = crtc_state->crc_source; + spin_unlock_irq(&drm_dev->event_lock); + + was_enabled = (current_source != DPU_CRTC_CRC_SOURCE_NONE); + + if (!was_enabled && enable) { + ret = drm_crtc_vblank_get(crtc); + + if (ret) + goto cleanup; + + } else if (was_enabled && !enable) { + drm_crtc_vblank_put(crtc); + } + + spin_lock_irq(&drm_dev->event_lock); + crtc_state->crc_source = source; + spin_unlock_irq(&drm_dev->event_lock); + + crtc_state->crc_frame_skip_count = 0; + + if (source == DPU_CRTC_CRC_SOURCE_LAYER_MIXER) + dpu_crtc_setup_lm_misr(crtc_state); + else if (source == DPU_CRTC_CRC_SOURCE_ENCODER) + dpu_crtc_setup_encoder_misr(crtc); + else + ret = -EINVAL; + +cleanup: + drm_modeset_unlock(&crtc->mutex); + + return ret; +} + +static u32 dpu_crtc_get_vblank_counter(struct drm_crtc *crtc) +{ + struct drm_encoder *encoder = get_encoder_from_crtc(crtc); + if (!encoder) { + DRM_ERROR("no encoder found for crtc %d\n", crtc->index); + return 0; + } + + return dpu_encoder_get_vsync_count(encoder); +} + +static int dpu_crtc_get_lm_crc(struct drm_crtc *crtc, + struct dpu_crtc_state *crtc_state) +{ + struct dpu_crtc_mixer *m; + u32 crcs[CRTC_DUAL_MIXERS]; + + int rc = 0; + int i; + + BUILD_BUG_ON(ARRAY_SIZE(crcs) != ARRAY_SIZE(crtc_state->mixers)); + + for (i = 0; i < crtc_state->num_mixers; ++i) { + + m = &crtc_state->mixers[i]; + + if (!m->hw_lm || !m->hw_lm->ops.collect_misr) + continue; + + rc = m->hw_lm->ops.collect_misr(m->hw_lm, &crcs[i]); + + if (rc) { + if (rc != -ENODATA) + DRM_DEBUG_DRIVER("MISR read failed\n"); + return rc; + } + } + + return drm_crtc_add_crc_entry(crtc, true, + drm_crtc_accurate_vblank_count(crtc), crcs); +} + +static int dpu_crtc_get_encoder_crc(struct drm_crtc *crtc) +{ + struct drm_encoder *drm_enc; + int rc, pos = 0; + u32 crcs[INTF_MAX]; + + drm_for_each_encoder_mask(drm_enc, crtc->dev, crtc->state->encoder_mask) { + rc = dpu_encoder_get_crc(drm_enc, crcs, pos); + if (rc < 0) { + if (rc != -ENODATA) + DRM_DEBUG_DRIVER("MISR read failed\n"); + + return rc; + } + + pos += rc; + } + + return drm_crtc_add_crc_entry(crtc, true, + drm_crtc_accurate_vblank_count(crtc), crcs); +} + +static int dpu_crtc_get_crc(struct drm_crtc *crtc) +{ + struct dpu_crtc_state *crtc_state = to_dpu_crtc_state(crtc->state); + + /* Skip first 2 frames in case of "uncooked" CRCs */ + if (crtc_state->crc_frame_skip_count < 2) { + crtc_state->crc_frame_skip_count++; + return 0; + } + + if (crtc_state->crc_source == DPU_CRTC_CRC_SOURCE_LAYER_MIXER) + return dpu_crtc_get_lm_crc(crtc, crtc_state); + else if (crtc_state->crc_source == DPU_CRTC_CRC_SOURCE_ENCODER) + return dpu_crtc_get_encoder_crc(crtc); + + return -EINVAL; +} + +static bool dpu_crtc_get_scanout_position(struct drm_crtc *crtc, + bool in_vblank_irq, + int *vpos, int *hpos, + ktime_t *stime, ktime_t *etime, + const struct drm_display_mode *mode) +{ + unsigned int pipe = crtc->index; + struct drm_encoder *encoder; + int line, vsw, vbp, vactive_start, vactive_end, vfp_end; + + encoder = get_encoder_from_crtc(crtc); + if (!encoder) { + DRM_ERROR("no encoder found for crtc %d\n", pipe); + return false; + } + + vsw = mode->crtc_vsync_end - mode->crtc_vsync_start; + vbp = mode->crtc_vtotal - mode->crtc_vsync_end; + + /* + * the line counter is 1 at the start of the VSYNC pulse and VTOTAL at + * the end of VFP. Translate the porch values relative to the line + * counter positions. + */ + + vactive_start = vsw + vbp + 1; + vactive_end = vactive_start + mode->crtc_vdisplay; + + /* last scan line before VSYNC */ + vfp_end = mode->crtc_vtotal; + + if (stime) + *stime = ktime_get(); + + line = dpu_encoder_get_linecount(encoder); + + if (line < vactive_start) + line -= vactive_start; + else if (line > vactive_end) + line = line - vfp_end - vactive_start; + else + line -= vactive_start; + + *vpos = line; + *hpos = 0; + + if (etime) + *etime = ktime_get(); + + return true; +} + +static void _dpu_crtc_setup_blend_cfg(struct dpu_crtc_mixer *mixer, + struct dpu_plane_state *pstate, + const struct msm_format *format, + const struct dpu_mdss_version *mdss_ver) +{ + struct dpu_hw_mixer *lm = mixer->hw_lm; + u32 blend_op; + u32 fg_alpha, bg_alpha; + + fg_alpha = pstate->base.alpha; + + /* default to opaque blending */ + if (pstate->base.pixel_blend_mode == DRM_MODE_BLEND_PIXEL_NONE || + !format->alpha_enable) { + blend_op = DPU_BLEND_FG_ALPHA_FG_CONST | + DPU_BLEND_BG_ALPHA_BG_CONST; + bg_alpha = DRM_BLEND_ALPHA_OPAQUE - fg_alpha; + } else if (pstate->base.pixel_blend_mode == DRM_MODE_BLEND_PREMULTI) { + blend_op = DPU_BLEND_FG_ALPHA_FG_CONST | + DPU_BLEND_BG_ALPHA_FG_PIXEL; + if (fg_alpha != DRM_BLEND_ALPHA_OPAQUE) { + bg_alpha = fg_alpha; + blend_op |= DPU_BLEND_BG_MOD_ALPHA | + DPU_BLEND_BG_INV_MOD_ALPHA; + } else { + bg_alpha = 0; + blend_op |= DPU_BLEND_BG_INV_ALPHA; + } + } else { + /* coverage blending */ + blend_op = DPU_BLEND_FG_ALPHA_FG_PIXEL | + DPU_BLEND_BG_ALPHA_FG_PIXEL; + if (fg_alpha != DRM_BLEND_ALPHA_OPAQUE) { + bg_alpha = fg_alpha; + blend_op |= DPU_BLEND_FG_MOD_ALPHA | + DPU_BLEND_FG_INV_MOD_ALPHA | + DPU_BLEND_BG_MOD_ALPHA | + DPU_BLEND_BG_INV_MOD_ALPHA; + } else { + bg_alpha = 0; + blend_op |= DPU_BLEND_BG_INV_ALPHA; + } + } + + lm->ops.setup_blend_config(lm, pstate->stage, + fg_alpha, bg_alpha, blend_op); + + DRM_DEBUG_ATOMIC("format:%p4cc, alpha_en:%u blend_op:0x%x\n", + &format->pixel_format, format->alpha_enable, blend_op); +} + +static void _dpu_crtc_program_lm_output_roi(struct drm_crtc *crtc) +{ + struct dpu_crtc_state *crtc_state; + int lm_idx; + + crtc_state = to_dpu_crtc_state(crtc->state); + + for (lm_idx = 0; lm_idx < crtc_state->num_mixers; lm_idx++) { + const struct drm_rect *lm_roi = &crtc_state->lm_bounds[lm_idx]; + struct dpu_hw_mixer *hw_lm = crtc_state->mixers[lm_idx].hw_lm; + struct dpu_hw_mixer_cfg cfg; + + if (!lm_roi || !drm_rect_visible(lm_roi)) + continue; + + cfg.out_width = drm_rect_width(lm_roi); + cfg.out_height = drm_rect_height(lm_roi); + cfg.right_mixer = lm_idx & 0x1; + cfg.flags = 0; + hw_lm->ops.setup_mixer_out(hw_lm, &cfg); + } +} + +static void _dpu_crtc_blend_setup_pipe(struct drm_crtc *crtc, + struct drm_plane *plane, + struct dpu_crtc_mixer *mixer, + u32 lms_in_pair, + enum dpu_stage stage, + const struct msm_format *format, + uint64_t modifier, + struct dpu_sw_pipe *pipe, + unsigned int stage_idx, + struct dpu_hw_stage_cfg *stage_cfg + ) +{ + u32 lm_idx; + enum dpu_sspp sspp_idx; + struct drm_plane_state *state; + + sspp_idx = pipe->sspp->idx; + + state = plane->state; + + trace_dpu_crtc_setup_mixer(DRMID(crtc), DRMID(plane), + state, to_dpu_plane_state(state), stage_idx, + format->pixel_format, pipe, + modifier); + + DRM_DEBUG_ATOMIC("crtc %d stage:%d - plane %d sspp %d fb %d multirect_idx %d\n", + crtc->base.id, + stage, + plane->base.id, + sspp_idx - SSPP_NONE, + state->fb ? state->fb->base.id : -1, + pipe->multirect_index); + + stage_cfg->stage[stage][stage_idx] = sspp_idx; + stage_cfg->multirect_index[stage][stage_idx] = pipe->multirect_index; + + /* blend config update */ + for (lm_idx = 0; lm_idx < lms_in_pair; lm_idx++) + mixer[lm_idx].lm_ctl->ops.update_pending_flush_sspp(mixer[lm_idx].lm_ctl, sspp_idx); +} + +static void _dpu_crtc_blend_setup_mixer(struct drm_crtc *crtc, + struct dpu_crtc *dpu_crtc, struct dpu_crtc_mixer *mixer, + struct dpu_hw_stage_cfg *stage_cfg) +{ + struct drm_plane *plane; + struct drm_framebuffer *fb; + struct drm_plane_state *state; + struct dpu_crtc_state *cstate = to_dpu_crtc_state(crtc->state); + struct dpu_plane_state *pstate = NULL; + const struct msm_format *format; + struct dpu_hw_ctl *ctl = mixer->lm_ctl; + u32 lm_idx, stage, i, pipe_idx, head_pipe_in_stage, lms_in_pair; + bool bg_alpha_enable = false; + DECLARE_BITMAP(active_fetch, SSPP_MAX); + DECLARE_BITMAP(active_pipes, SSPP_MAX); + + memset(active_fetch, 0, sizeof(active_fetch)); + memset(active_pipes, 0, sizeof(active_pipes)); + drm_atomic_crtc_for_each_plane(plane, crtc) { + state = plane->state; + if (!state) + continue; + + if (!state->visible) + continue; + + pstate = to_dpu_plane_state(state); + fb = state->fb; + + format = msm_framebuffer_format(pstate->base.fb); + + if (pstate->stage == DPU_STAGE_BASE && format->alpha_enable) + bg_alpha_enable = true; + + /* loop pipe per mixer pair with config in stage structure */ + for (stage = 0; stage < STAGES_PER_PLANE; stage++) { + head_pipe_in_stage = stage * PIPES_PER_STAGE; + for (i = 0; i < PIPES_PER_STAGE; i++) { + pipe_idx = i + head_pipe_in_stage; + if (!pstate->pipe[pipe_idx].sspp) + continue; + lms_in_pair = min(cstate->num_mixers - (stage * PIPES_PER_STAGE), + PIPES_PER_STAGE); + set_bit(pstate->pipe[pipe_idx].sspp->idx, active_fetch); + set_bit(pstate->pipe[pipe_idx].sspp->idx, active_pipes); + _dpu_crtc_blend_setup_pipe(crtc, plane, + &mixer[head_pipe_in_stage], + lms_in_pair, + pstate->stage, + format, fb ? fb->modifier : 0, + &pstate->pipe[pipe_idx], i, + &stage_cfg[stage]); + } + } + + /* blend config update */ + for (lm_idx = 0; lm_idx < cstate->num_mixers; lm_idx++) { + _dpu_crtc_setup_blend_cfg(mixer + lm_idx, pstate, format, + ctl->mdss_ver); + + if (bg_alpha_enable && !format->alpha_enable) + mixer[lm_idx].mixer_op_mode = 0; + else + mixer[lm_idx].mixer_op_mode |= + 1 << pstate->stage; + } + } + + if (ctl->ops.set_active_fetch_pipes) + ctl->ops.set_active_fetch_pipes(ctl, active_fetch); + + if (ctl->ops.set_active_pipes) + ctl->ops.set_active_pipes(ctl, active_pipes); + + _dpu_crtc_program_lm_output_roi(crtc); +} + +/** + * _dpu_crtc_blend_setup - configure crtc mixers + * @crtc: Pointer to drm crtc structure + */ +static void _dpu_crtc_blend_setup(struct drm_crtc *crtc) +{ + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); + struct dpu_crtc_state *cstate = to_dpu_crtc_state(crtc->state); + struct dpu_crtc_mixer *mixer = cstate->mixers; + struct dpu_hw_ctl *ctl; + struct dpu_hw_mixer *lm; + struct dpu_hw_stage_cfg stage_cfg[STAGES_PER_PLANE]; + DECLARE_BITMAP(active_lms, LM_MAX); + int i; + + DRM_DEBUG_ATOMIC("%s\n", dpu_crtc->name); + + for (i = 0; i < cstate->num_mixers; i++) { + mixer[i].mixer_op_mode = 0; + if (mixer[i].lm_ctl->ops.clear_all_blendstages) + mixer[i].lm_ctl->ops.clear_all_blendstages( + mixer[i].lm_ctl); + if (mixer[i].lm_ctl->ops.set_active_fetch_pipes) + mixer[i].lm_ctl->ops.set_active_fetch_pipes(mixer[i].lm_ctl, NULL); + if (mixer[i].lm_ctl->ops.set_active_pipes) + mixer[i].lm_ctl->ops.set_active_pipes(mixer[i].lm_ctl, NULL); + + if (mixer[i].hw_lm->ops.clear_all_blendstages) + mixer[i].hw_lm->ops.clear_all_blendstages(mixer[i].hw_lm); + } + + /* initialize stage cfg */ + memset(&stage_cfg, 0, sizeof(stage_cfg)); + memset(active_lms, 0, sizeof(active_lms)); + + _dpu_crtc_blend_setup_mixer(crtc, dpu_crtc, mixer, stage_cfg); + + for (i = 0; i < cstate->num_mixers; i++) { + ctl = mixer[i].lm_ctl; + lm = mixer[i].hw_lm; + + lm->ops.setup_alpha_out(lm, mixer[i].mixer_op_mode); + + /* stage config flush mask */ + ctl->ops.update_pending_flush_mixer(ctl, + mixer[i].hw_lm->idx); + + set_bit(lm->idx, active_lms); + if (ctl->ops.set_active_lms) + ctl->ops.set_active_lms(ctl, active_lms); + + DRM_DEBUG_ATOMIC("lm %d, op_mode 0x%X, ctl %d\n", + mixer[i].hw_lm->idx - LM_0, + mixer[i].mixer_op_mode, + ctl->idx - CTL_0); + + /* + * call dpu_hw_ctl_setup_blendstage() to blend layers per stage cfg. + * stage data is shared between PIPES_PER_STAGE pipes. + */ + if (ctl->ops.setup_blendstage) + ctl->ops.setup_blendstage(ctl, mixer[i].hw_lm->idx, + &stage_cfg[i / PIPES_PER_STAGE]); + + if (lm->ops.setup_blendstage) + lm->ops.setup_blendstage(lm, mixer[i].hw_lm->idx, + &stage_cfg[i / PIPES_PER_STAGE]); + } +} + +/** + * _dpu_crtc_complete_flip - signal pending page_flip events + * Any pending vblank events are added to the vblank_event_list + * so that the next vblank interrupt shall signal them. + * However PAGE_FLIP events are not handled through the vblank_event_list. + * This API signals any pending PAGE_FLIP events requested through + * DRM_IOCTL_MODE_PAGE_FLIP and are cached in the dpu_crtc->event. + * @crtc: Pointer to drm crtc structure + */ +static void _dpu_crtc_complete_flip(struct drm_crtc *crtc) +{ + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); + struct drm_device *dev = crtc->dev; + unsigned long flags; + + spin_lock_irqsave(&dev->event_lock, flags); + if (dpu_crtc->event) { + DRM_DEBUG_VBL("%s: send event: %p\n", dpu_crtc->name, + dpu_crtc->event); + trace_dpu_crtc_complete_flip(DRMID(crtc)); + drm_crtc_send_vblank_event(crtc, dpu_crtc->event); + dpu_crtc->event = NULL; + } + spin_unlock_irqrestore(&dev->event_lock, flags); +} + +/** + * dpu_crtc_get_intf_mode - get interface mode of the given crtc + * @crtc: Pointert to crtc + */ +enum dpu_intf_mode dpu_crtc_get_intf_mode(struct drm_crtc *crtc) +{ + struct drm_encoder *encoder; + + /* + * TODO: This function is called from dpu debugfs and as part of atomic + * check. When called from debugfs, the crtc->mutex must be held to + * read crtc->state. However reading crtc->state from atomic check isn't + * allowed (unless you have a good reason, a big comment, and a deep + * understanding of how the atomic/modeset locks work (<- and this is + * probably not possible)). So we'll keep the WARN_ON here for now, but + * really we need to figure out a better way to track our operating mode + */ + WARN_ON(!drm_modeset_is_locked(&crtc->mutex)); + + /* TODO: Returns the first INTF_MODE, could there be multiple values? */ + drm_for_each_encoder_mask(encoder, crtc->dev, crtc->state->encoder_mask) + return dpu_encoder_get_intf_mode(encoder); + + return INTF_MODE_NONE; +} + +/** + * dpu_crtc_vblank_callback - called on vblank irq, issues completion events + * @crtc: Pointer to drm crtc object + */ +void dpu_crtc_vblank_callback(struct drm_crtc *crtc) +{ + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); + + /* keep statistics on vblank callback - with auto reset via debugfs */ + if (ktime_compare(dpu_crtc->vblank_cb_time, ktime_set(0, 0)) == 0) + dpu_crtc->vblank_cb_time = ktime_get(); + else + dpu_crtc->vblank_cb_count++; + + dpu_crtc_get_crc(crtc); + + drm_crtc_handle_vblank(crtc); + trace_dpu_crtc_vblank_cb(DRMID(crtc)); +} + +static void dpu_crtc_frame_event_work(struct kthread_work *work) +{ + struct dpu_crtc_frame_event *fevent = container_of(work, + struct dpu_crtc_frame_event, work); + struct drm_crtc *crtc = fevent->crtc; + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); + unsigned long flags; + bool frame_done = false; + + DPU_ATRACE_BEGIN("crtc_frame_event"); + + DRM_DEBUG_ATOMIC("crtc%d event:%u ts:%lld\n", crtc->base.id, fevent->event, + ktime_to_ns(fevent->ts)); + + if (fevent->event & (DPU_ENCODER_FRAME_EVENT_DONE + | DPU_ENCODER_FRAME_EVENT_ERROR + | DPU_ENCODER_FRAME_EVENT_PANEL_DEAD)) { + + if (atomic_read(&dpu_crtc->frame_pending) < 1) { + /* ignore vblank when not pending */ + } else if (atomic_dec_return(&dpu_crtc->frame_pending) == 0) { + /* release bandwidth and other resources */ + trace_dpu_crtc_frame_event_done(DRMID(crtc), + fevent->event); + dpu_core_perf_crtc_release_bw(crtc); + } else { + trace_dpu_crtc_frame_event_more_pending(DRMID(crtc), + fevent->event); + } + + if (fevent->event & (DPU_ENCODER_FRAME_EVENT_DONE + | DPU_ENCODER_FRAME_EVENT_ERROR)) + frame_done = true; + } + + if (fevent->event & DPU_ENCODER_FRAME_EVENT_PANEL_DEAD) + DPU_ERROR("crtc%d ts:%lld received panel dead event\n", + crtc->base.id, ktime_to_ns(fevent->ts)); + + if (frame_done) + complete_all(&dpu_crtc->frame_done_comp); + + spin_lock_irqsave(&dpu_crtc->spin_lock, flags); + list_add_tail(&fevent->list, &dpu_crtc->frame_event_list); + spin_unlock_irqrestore(&dpu_crtc->spin_lock, flags); + DPU_ATRACE_END("crtc_frame_event"); +} + +/** + * dpu_crtc_frame_event_cb - crtc frame event callback API + * @crtc: Pointer to crtc + * @event: Event to process + * + * Encoder may call this for different events from different context - IRQ, + * user thread, commit_thread, etc. Each event should be carefully reviewed and + * should be processed in proper task context to avoid schedulin delay or + * properly manage the irq context's bottom half processing. + */ +void dpu_crtc_frame_event_cb(struct drm_crtc *crtc, u32 event) +{ + struct dpu_crtc *dpu_crtc; + struct msm_drm_private *priv; + struct dpu_crtc_frame_event *fevent; + unsigned long flags; + u32 crtc_id; + + /* Nothing to do on idle event */ + if (event & DPU_ENCODER_FRAME_EVENT_IDLE) + return; + + dpu_crtc = to_dpu_crtc(crtc); + priv = crtc->dev->dev_private; + crtc_id = drm_crtc_index(crtc); + + trace_dpu_crtc_frame_event_cb(DRMID(crtc), event); + + spin_lock_irqsave(&dpu_crtc->spin_lock, flags); + fevent = list_first_entry_or_null(&dpu_crtc->frame_event_list, + struct dpu_crtc_frame_event, list); + if (fevent) + list_del_init(&fevent->list); + spin_unlock_irqrestore(&dpu_crtc->spin_lock, flags); + + if (!fevent) { + DRM_ERROR_RATELIMITED("crtc%d event %d overflow\n", crtc->base.id, event); + return; + } + + fevent->event = event; + fevent->crtc = crtc; + fevent->ts = ktime_get(); + kthread_queue_work(priv->kms->event_thread[crtc_id].worker, &fevent->work); +} + +/** + * dpu_crtc_complete_commit - callback signalling completion of current commit + * @crtc: Pointer to drm crtc object + */ +void dpu_crtc_complete_commit(struct drm_crtc *crtc) +{ + trace_dpu_crtc_complete_commit(DRMID(crtc)); + dpu_core_perf_crtc_update(crtc, 0); + _dpu_crtc_complete_flip(crtc); +} + +static int _dpu_crtc_check_and_setup_lm_bounds(struct drm_crtc *crtc, + struct drm_crtc_state *state) +{ + struct dpu_crtc_state *cstate = to_dpu_crtc_state(state); + struct drm_display_mode *adj_mode = &state->adjusted_mode; + u32 crtc_split_width = adj_mode->hdisplay / cstate->num_mixers; + struct dpu_kms *dpu_kms = _dpu_crtc_get_kms(crtc); + int i; + + /* if we cannot merge 2 LMs (no 3d mux) better to fail earlier + * before even checking the width after the split + */ + if (!dpu_kms->catalog->caps->has_3d_merge && + adj_mode->hdisplay > dpu_kms->catalog->caps->max_mixer_width) + return -E2BIG; + + for (i = 0; i < cstate->num_mixers; i++) { + struct drm_rect *r = &cstate->lm_bounds[i]; + r->x1 = crtc_split_width * i; + r->y1 = 0; + r->x2 = r->x1 + crtc_split_width; + r->y2 = adj_mode->vdisplay; + + trace_dpu_crtc_setup_lm_bounds(DRMID(crtc), i, r); + + if (drm_rect_width(r) > dpu_kms->catalog->caps->max_mixer_width) + return -E2BIG; + } + + return 0; +} + +static void _dpu_crtc_get_pcc_coeff(struct drm_crtc_state *state, + struct dpu_hw_pcc_cfg *cfg) +{ + struct drm_color_ctm *ctm; + + memset(cfg, 0, sizeof(struct dpu_hw_pcc_cfg)); + + ctm = (struct drm_color_ctm *)state->ctm->data; + + if (!ctm) + return; + + cfg->r.r = CONVERT_S3_15(ctm->matrix[0]); + cfg->g.r = CONVERT_S3_15(ctm->matrix[1]); + cfg->b.r = CONVERT_S3_15(ctm->matrix[2]); + + cfg->r.g = CONVERT_S3_15(ctm->matrix[3]); + cfg->g.g = CONVERT_S3_15(ctm->matrix[4]); + cfg->b.g = CONVERT_S3_15(ctm->matrix[5]); + + cfg->r.b = CONVERT_S3_15(ctm->matrix[6]); + cfg->g.b = CONVERT_S3_15(ctm->matrix[7]); + cfg->b.b = CONVERT_S3_15(ctm->matrix[8]); +} + +static void _dpu_crtc_get_gc_lut(struct drm_crtc_state *state, + struct dpu_hw_gc_lut *gc_lut) +{ + struct drm_color_lut *lut; + int i; + u32 val_even, val_odd; + + lut = (struct drm_color_lut *)state->gamma_lut->data; + + if (!lut) + return; + + /* Pack 1024 10-bit entries in 512 32-bit registers */ + for (i = 0; i < PGC_TBL_LEN; i++) { + val_even = drm_color_lut_extract(lut[i * 2].green, 10); + val_odd = drm_color_lut_extract(lut[i * 2 + 1].green, 10); + gc_lut->c0[i] = val_even | (val_odd << 16); + val_even = drm_color_lut_extract(lut[i * 2].blue, 10); + val_odd = drm_color_lut_extract(lut[i * 2 + 1].blue, 10); + gc_lut->c1[i] = val_even | (val_odd << 16); + val_even = drm_color_lut_extract(lut[i * 2].red, 10); + val_odd = drm_color_lut_extract(lut[i * 2 + 1].red, 10); + gc_lut->c2[i] = val_even | (val_odd << 16); + } + + /* Disable 8-bit rounding mode */ + gc_lut->flags = 0; +} + +static void _dpu_crtc_setup_cp_blocks(struct drm_crtc *crtc) +{ + struct drm_crtc_state *state = crtc->state; + struct dpu_crtc_state *cstate = to_dpu_crtc_state(crtc->state); + struct dpu_crtc_mixer *mixer = cstate->mixers; + struct dpu_hw_pcc_cfg cfg; + struct dpu_hw_gc_lut *gc_lut; + struct dpu_hw_ctl *ctl; + struct dpu_hw_dspp *dspp; + int i; + + + if (!state->color_mgmt_changed && !drm_atomic_crtc_needs_modeset(state)) + return; + + for (i = 0; i < cstate->num_mixers; i++) { + ctl = mixer[i].lm_ctl; + dspp = mixer[i].hw_dspp; + + if (!dspp) + continue; + + if (dspp->ops.setup_pcc) { + if (!state->ctm) { + dspp->ops.setup_pcc(dspp, NULL); + } else { + _dpu_crtc_get_pcc_coeff(state, &cfg); + dspp->ops.setup_pcc(dspp, &cfg); + } + + /* stage config flush mask */ + ctl->ops.update_pending_flush_dspp(ctl, + mixer[i].hw_dspp->idx, DPU_DSPP_PCC); + } + + if (dspp->ops.setup_gc) { + if (!state->gamma_lut) { + dspp->ops.setup_gc(dspp, NULL); + } else { + gc_lut = kzalloc_obj(*gc_lut); + if (!gc_lut) + continue; + _dpu_crtc_get_gc_lut(state, gc_lut); + dspp->ops.setup_gc(dspp, gc_lut); + kfree(gc_lut); + } + + /* stage config flush mask */ + ctl->ops.update_pending_flush_dspp(ctl, + mixer[i].hw_dspp->idx, DPU_DSPP_GC); + } + } +} + +static void dpu_crtc_atomic_begin(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + struct dpu_crtc_state *cstate = to_dpu_crtc_state(crtc->state); + struct drm_encoder *encoder; + + if (!crtc->state->enable) { + DRM_DEBUG_ATOMIC("crtc%d -> enable %d, skip atomic_begin\n", + crtc->base.id, crtc->state->enable); + return; + } + + DRM_DEBUG_ATOMIC("crtc%d\n", crtc->base.id); + + _dpu_crtc_check_and_setup_lm_bounds(crtc, crtc->state); + + /* encoder will trigger pending mask now */ + drm_for_each_encoder_mask(encoder, crtc->dev, crtc->state->encoder_mask) + dpu_encoder_trigger_kickoff_pending(encoder); + + /* + * If no mixers have been allocated in dpu_crtc_atomic_check(), + * it means we are trying to flush a CRTC whose state is disabled: + * nothing else needs to be done. + */ + if (unlikely(!cstate->num_mixers)) + return; + + _dpu_crtc_blend_setup(crtc); + + _dpu_crtc_setup_cp_blocks(crtc); + + /* + * PP_DONE irq is only used by command mode for now. + * It is better to request pending before FLUSH and START trigger + * to make sure no pp_done irq missed. + * This is safe because no pp_done will happen before SW trigger + * in command mode. + */ +} + +static void dpu_crtc_atomic_flush(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + struct dpu_crtc *dpu_crtc; + struct drm_device *dev; + struct drm_plane *plane; + struct msm_drm_private *priv; + unsigned long flags; + struct dpu_crtc_state *cstate; + + if (!crtc->state->enable) { + DRM_DEBUG_ATOMIC("crtc%d -> enable %d, skip atomic_flush\n", + crtc->base.id, crtc->state->enable); + return; + } + + DRM_DEBUG_ATOMIC("crtc%d\n", crtc->base.id); + + dpu_crtc = to_dpu_crtc(crtc); + cstate = to_dpu_crtc_state(crtc->state); + dev = crtc->dev; + priv = dev->dev_private; + + if (crtc->index >= ARRAY_SIZE(priv->kms->event_thread)) { + DPU_ERROR("invalid crtc index[%d]\n", crtc->index); + return; + } + + WARN_ON(dpu_crtc->event); + spin_lock_irqsave(&dev->event_lock, flags); + dpu_crtc->event = crtc->state->event; + crtc->state->event = NULL; + spin_unlock_irqrestore(&dev->event_lock, flags); + + /* + * If no mixers has been allocated in dpu_crtc_atomic_check(), + * it means we are trying to flush a CRTC whose state is disabled: + * nothing else needs to be done. + */ + if (unlikely(!cstate->num_mixers)) + return; + + /* update performance setting before crtc kickoff */ + dpu_core_perf_crtc_update(crtc, 1); + + /* + * Final plane updates: Give each plane a chance to complete all + * required writes/flushing before crtc's "flush + * everything" call below. + */ + drm_atomic_crtc_for_each_plane(plane, crtc) { + if (dpu_crtc->smmu_state.transition_error) + dpu_plane_set_error(plane, true); + dpu_plane_flush(plane); + } + + /* Kickoff will be scheduled by outer layer */ +} + +/** + * dpu_crtc_destroy_state - state destroy hook + * @crtc: drm CRTC + * @state: CRTC state object to release + */ +static void dpu_crtc_destroy_state(struct drm_crtc *crtc, + struct drm_crtc_state *state) +{ + struct dpu_crtc_state *cstate = to_dpu_crtc_state(state); + + DRM_DEBUG_ATOMIC("crtc%d\n", crtc->base.id); + + __drm_atomic_helper_crtc_destroy_state(state); + + kfree(cstate); +} + +static int _dpu_crtc_wait_for_frame_done(struct drm_crtc *crtc) +{ + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); + int ret, rc = 0; + + if (!atomic_read(&dpu_crtc->frame_pending)) { + DRM_DEBUG_ATOMIC("no frames pending\n"); + return 0; + } + + DPU_ATRACE_BEGIN("frame done completion wait"); + ret = wait_for_completion_timeout(&dpu_crtc->frame_done_comp, + msecs_to_jiffies(DPU_CRTC_FRAME_DONE_TIMEOUT_MS)); + if (!ret) { + DRM_ERROR("frame done wait timed out, ret:%d\n", ret); + rc = -ETIMEDOUT; + } + DPU_ATRACE_END("frame done completion wait"); + + return rc; +} + +static int dpu_crtc_kickoff_clone_mode(struct drm_crtc *crtc) +{ + struct drm_encoder *encoder; + struct drm_encoder *rt_encoder = NULL, *wb_encoder = NULL; + struct dpu_kms *dpu_kms = _dpu_crtc_get_kms(crtc); + + /* Find encoder for real time display */ + drm_for_each_encoder_mask(encoder, crtc->dev, + crtc->state->encoder_mask) { + if (encoder->encoder_type == DRM_MODE_ENCODER_VIRTUAL) + wb_encoder = encoder; + else + rt_encoder = encoder; + } + + if (!rt_encoder || !wb_encoder) { + DRM_DEBUG_ATOMIC("real time or wb encoder not found\n"); + return -EINVAL; + } + + dpu_encoder_prepare_for_kickoff(wb_encoder); + dpu_encoder_prepare_for_kickoff(rt_encoder); + + dpu_vbif_clear_errors(dpu_kms); + + /* + * Kickoff real time encoder last as it's the encoder that + * will do the flush + */ + dpu_encoder_kickoff(wb_encoder); + dpu_encoder_kickoff(rt_encoder); + + /* Don't start frame done timers until the kickoffs have finished */ + dpu_encoder_start_frame_done_timer(wb_encoder); + dpu_encoder_start_frame_done_timer(rt_encoder); + + return 0; +} + +/** + * dpu_crtc_commit_kickoff - trigger kickoff of the commit for this crtc + * @crtc: Pointer to drm crtc object + */ +void dpu_crtc_commit_kickoff(struct drm_crtc *crtc) +{ + struct drm_encoder *encoder; + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); + struct dpu_kms *dpu_kms = _dpu_crtc_get_kms(crtc); + struct dpu_crtc_state *cstate = to_dpu_crtc_state(crtc->state); + + /* + * If no mixers has been allocated in dpu_crtc_atomic_check(), + * it means we are trying to start a CRTC whose state is disabled: + * nothing else needs to be done. + */ + if (unlikely(!cstate->num_mixers)) + return; + + DPU_ATRACE_BEGIN("crtc_commit"); + + drm_for_each_encoder_mask(encoder, crtc->dev, + crtc->state->encoder_mask) { + if (!dpu_encoder_is_valid_for_commit(encoder)) { + DRM_DEBUG_ATOMIC("invalid FB not kicking off crtc\n"); + goto end; + } + } + + if (drm_crtc_in_clone_mode(crtc->state)) { + if (dpu_crtc_kickoff_clone_mode(crtc)) + goto end; + } else { + /* + * Encoder will flush/start now, unless it has a tx pending. + * If so, it may delay and flush at an irq event (e.g. ppdone) + */ + drm_for_each_encoder_mask(encoder, crtc->dev, + crtc->state->encoder_mask) + dpu_encoder_prepare_for_kickoff(encoder); + + dpu_vbif_clear_errors(dpu_kms); + + drm_for_each_encoder_mask(encoder, crtc->dev, + crtc->state->encoder_mask) { + dpu_encoder_kickoff(encoder); + dpu_encoder_start_frame_done_timer(encoder); + } + } + + if (atomic_inc_return(&dpu_crtc->frame_pending) == 1) { + /* acquire bandwidth and other resources */ + DRM_DEBUG_ATOMIC("crtc%d first commit\n", crtc->base.id); + } else + DRM_DEBUG_ATOMIC("crtc%d commit\n", crtc->base.id); + + dpu_crtc->play_count++; + + reinit_completion(&dpu_crtc->frame_done_comp); + +end: + DPU_ATRACE_END("crtc_commit"); +} + +static void dpu_crtc_reset(struct drm_crtc *crtc) +{ + struct dpu_crtc_state *cstate = kzalloc_obj(*cstate); + + if (crtc->state) + dpu_crtc_destroy_state(crtc, crtc->state); + + if (cstate) + __drm_atomic_helper_crtc_reset(crtc, &cstate->base); + else + __drm_atomic_helper_crtc_reset(crtc, NULL); +} + +/** + * dpu_crtc_duplicate_state - state duplicate hook + * @crtc: Pointer to drm crtc structure + */ +static struct drm_crtc_state *dpu_crtc_duplicate_state(struct drm_crtc *crtc) +{ + struct dpu_crtc_state *cstate, *old_cstate = to_dpu_crtc_state(crtc->state); + + cstate = kmemdup(old_cstate, sizeof(*old_cstate), GFP_KERNEL); + if (!cstate) { + DPU_ERROR("failed to allocate state\n"); + return NULL; + } + + /* duplicate base helper */ + __drm_atomic_helper_crtc_duplicate_state(crtc, &cstate->base); + + return &cstate->base; +} + +static void dpu_crtc_atomic_print_state(struct drm_printer *p, + const struct drm_crtc_state *state) +{ + const struct dpu_crtc_state *cstate = to_dpu_crtc_state(state); + int i; + + for (i = 0; i < cstate->num_mixers; i++) { + drm_printf(p, "\tlm[%d]=%d\n", i, cstate->mixers[i].hw_lm->idx - LM_0); + drm_printf(p, "\tctl[%d]=%d\n", i, cstate->mixers[i].lm_ctl->idx - CTL_0); + if (cstate->mixers[i].hw_dspp) + drm_printf(p, "\tdspp[%d]=%d\n", i, cstate->mixers[i].hw_dspp->idx - DSPP_0); + } +} + +static void dpu_crtc_disable(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + struct drm_crtc_state *old_crtc_state = drm_atomic_get_old_crtc_state(state, + crtc); + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); + struct dpu_crtc_state *cstate = to_dpu_crtc_state(crtc->state); + struct drm_encoder *encoder; + unsigned long flags; + bool release_bandwidth = false; + + DRM_DEBUG_KMS("crtc%d\n", crtc->base.id); + + /* If disable is triggered while in self refresh mode, + * reset the encoder software state so that in enable + * it won't trigger a warn while assigning crtc. + */ + if (old_crtc_state->self_refresh_active) { + drm_for_each_encoder_mask(encoder, crtc->dev, + old_crtc_state->encoder_mask) { + dpu_encoder_assign_crtc(encoder, NULL); + } + return; + } + + /* Disable/save vblank irq handling */ + drm_crtc_vblank_off(crtc); + + drm_for_each_encoder_mask(encoder, crtc->dev, + old_crtc_state->encoder_mask) { + /* in video mode, we hold an extra bandwidth reference + * as we cannot drop bandwidth at frame-done if any + * crtc is being used in video mode. + */ + if (dpu_encoder_get_intf_mode(encoder) == INTF_MODE_VIDEO) + release_bandwidth = true; + + /* + * If disable is triggered during psr active(e.g: screen dim in PSR), + * we will need encoder->crtc connection to process the device sleep & + * preserve it during psr sequence. + */ + if (!crtc->state->self_refresh_active) + dpu_encoder_assign_crtc(encoder, NULL); + } + + /* wait for frame_event_done completion */ + if (_dpu_crtc_wait_for_frame_done(crtc)) + DPU_ERROR("crtc%d wait for frame done failed;frame_pending%d\n", + crtc->base.id, + atomic_read(&dpu_crtc->frame_pending)); + + trace_dpu_crtc_disable(DRMID(crtc), false, dpu_crtc); + dpu_crtc->enabled = false; + + if (atomic_read(&dpu_crtc->frame_pending)) { + trace_dpu_crtc_disable_frame_pending(DRMID(crtc), + atomic_read(&dpu_crtc->frame_pending)); + if (release_bandwidth) + dpu_core_perf_crtc_release_bw(crtc); + atomic_set(&dpu_crtc->frame_pending, 0); + } + + dpu_core_perf_crtc_update(crtc, 0); + + /* disable clk & bw control until clk & bw properties are set */ + cstate->bw_control = false; + cstate->bw_split_vote = false; + + if (crtc->state->event && !crtc->state->active) { + spin_lock_irqsave(&crtc->dev->event_lock, flags); + drm_crtc_send_vblank_event(crtc, crtc->state->event); + crtc->state->event = NULL; + spin_unlock_irqrestore(&crtc->dev->event_lock, flags); + } + + pm_runtime_put_sync(crtc->dev->dev); +} + +static void dpu_crtc_enable(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); + struct drm_encoder *encoder; + bool request_bandwidth = false; + struct drm_crtc_state *old_crtc_state; + + old_crtc_state = drm_atomic_get_old_crtc_state(state, crtc); + + pm_runtime_get_sync(crtc->dev->dev); + + DRM_DEBUG_KMS("crtc%d\n", crtc->base.id); + + drm_for_each_encoder_mask(encoder, crtc->dev, crtc->state->encoder_mask) { + /* in video mode, we hold an extra bandwidth reference + * as we cannot drop bandwidth at frame-done if any + * crtc is being used in video mode. + */ + if (dpu_encoder_get_intf_mode(encoder) == INTF_MODE_VIDEO) + request_bandwidth = true; + } + + if (request_bandwidth) + atomic_inc(&_dpu_crtc_get_kms(crtc)->bandwidth_ref); + + trace_dpu_crtc_enable(DRMID(crtc), true, dpu_crtc); + dpu_crtc->enabled = true; + + if (!old_crtc_state->self_refresh_active) { + drm_for_each_encoder_mask(encoder, crtc->dev, crtc->state->encoder_mask) + dpu_encoder_assign_crtc(encoder, crtc); + } + + /* Enable/restore vblank irq handling */ + drm_crtc_vblank_on(crtc); +} + +static bool dpu_crtc_needs_dirtyfb(struct drm_crtc_state *cstate) +{ + struct drm_crtc *crtc = cstate->crtc; + struct drm_encoder *encoder; + + if (cstate->self_refresh_active) + return true; + + drm_for_each_encoder_mask (encoder, crtc->dev, cstate->encoder_mask) { + if (dpu_encoder_get_intf_mode(encoder) == INTF_MODE_CMD) { + return true; + } + } + + return false; +} + +static int dpu_crtc_assign_planes(struct drm_crtc *crtc, struct drm_crtc_state *crtc_state) +{ + int total_planes = crtc->dev->mode_config.num_total_plane; + struct drm_atomic_commit *state = crtc_state->state; + struct dpu_global_state *global_state; + struct drm_plane_state **states; + struct drm_plane *plane; + int ret; + + global_state = dpu_kms_get_global_state(crtc_state->state); + if (IS_ERR(global_state)) + return PTR_ERR(global_state); + + if (!crtc_state->enable) + return 0; + + states = kzalloc_objs(*states, total_planes); + if (!states) + return -ENOMEM; + + drm_atomic_crtc_state_for_each_plane(plane, crtc_state) { + struct drm_plane_state *plane_state = + drm_atomic_get_plane_state(state, plane); + + if (IS_ERR(plane_state)) { + ret = PTR_ERR(plane_state); + goto done; + } + + states[plane_state->normalized_zpos] = plane_state; + } + + ret = dpu_assign_plane_resources(global_state, state, crtc, states, total_planes); + +done: + kfree(states); + return ret; +} + +static int dpu_crtc_reassign_planes(struct drm_crtc *crtc, struct drm_crtc_state *crtc_state) +{ + struct dpu_global_state *global_state; + + global_state = dpu_kms_get_global_state(crtc_state->state); + if (IS_ERR(global_state)) + return PTR_ERR(global_state); + + dpu_rm_release_all_sspp(global_state, crtc); + + return dpu_crtc_assign_planes(crtc, crtc_state); +} + +#define MAX_CHANNELS_PER_CRTC PIPES_PER_PLANE +#define MAX_HDISPLAY_SPLIT 1080 + +static struct msm_display_topology dpu_crtc_get_topology( + struct drm_crtc *crtc, + struct dpu_kms *dpu_kms, + struct drm_crtc_state *crtc_state) +{ + struct drm_display_mode *mode = &crtc_state->adjusted_mode; + struct msm_display_topology topology = {0}; + struct drm_encoder *drm_enc; + + drm_for_each_encoder_mask(drm_enc, crtc->dev, crtc_state->encoder_mask) + dpu_encoder_update_topology(drm_enc, &topology, crtc_state->state, + &crtc_state->adjusted_mode); + + topology.cwb_enabled = drm_crtc_in_clone_mode(crtc_state); + + /* + * Datapath topology selection + * + * Dual display + * 2 LM, 2 INTF ( Split display using 2 interfaces) + * + * Single display + * 1 LM, 1 INTF + * 2 LM, 1 INTF (stream merge to support high resolution interfaces) + * + * If DSC is enabled, use 2 LMs for 2:2:1 topology + * + * Add dspps to the reservation requirements if ctm or gamma_lut are requested + * + * Only hardcode num_lm to 2 for cases where num_intf == 2 and CWB is not + * enabled. This is because in cases where CWB is enabled, num_intf will + * count both the WB and real-time phys encoders. + * + * For non-DSC CWB usecases, have the num_lm be decided by the + * (mode->hdisplay > MAX_HDISPLAY_SPLIT) check. + */ + + if (topology.num_intf == 2 && !topology.cwb_enabled) + topology.num_lm = 2; + else if (topology.num_dsc == 2) + topology.num_lm = 2; + else if (dpu_kms->catalog->caps->has_3d_merge && + topology.num_dsc == 0) + topology.num_lm = (mode->hdisplay > MAX_HDISPLAY_SPLIT) ? 2 : 1; + else + topology.num_lm = 1; + + if (crtc_state->ctm || crtc_state->gamma_lut) + topology.num_dspp = topology.num_lm; + + return topology; +} + +static int dpu_crtc_assign_resources(struct drm_crtc *crtc, + struct drm_crtc_state *crtc_state) +{ + struct dpu_hw_blk *hw_ctl[MAX_CHANNELS_PER_CRTC]; + struct dpu_hw_blk *hw_lm[MAX_CHANNELS_PER_CRTC]; + struct dpu_hw_blk *hw_dspp[MAX_CHANNELS_PER_CRTC]; + int i, num_lm, num_ctl, num_dspp; + struct dpu_kms *dpu_kms = _dpu_crtc_get_kms(crtc); + struct dpu_global_state *global_state; + struct dpu_crtc_state *cstate; + struct msm_display_topology topology; + int ret; + + /* + * Release and Allocate resources on every modeset + */ + global_state = dpu_kms_get_global_state(crtc_state->state); + if (IS_ERR(global_state)) + return PTR_ERR(global_state); + + dpu_rm_release(global_state, crtc); + + if (!crtc_state->enable) + return 0; + + topology = dpu_crtc_get_topology(crtc, dpu_kms, crtc_state); + ret = dpu_rm_reserve(&dpu_kms->rm, global_state, + crtc_state->crtc, &topology); + if (ret) + return ret; + + cstate = to_dpu_crtc_state(crtc_state); + + num_ctl = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + crtc_state->crtc, + DPU_HW_BLK_CTL, hw_ctl, + ARRAY_SIZE(hw_ctl)); + num_lm = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + crtc_state->crtc, + DPU_HW_BLK_LM, hw_lm, + ARRAY_SIZE(hw_lm)); + num_dspp = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + crtc_state->crtc, + DPU_HW_BLK_DSPP, hw_dspp, + ARRAY_SIZE(hw_dspp)); + + for (i = 0; i < num_lm; i++) { + int ctl_idx = (i < num_ctl) ? i : (num_ctl-1); + + cstate->mixers[i].hw_lm = to_dpu_hw_mixer(hw_lm[i]); + cstate->mixers[i].lm_ctl = to_dpu_hw_ctl(hw_ctl[ctl_idx]); + if (i < num_dspp) + cstate->mixers[i].hw_dspp = to_dpu_hw_dspp(hw_dspp[i]); + } + + cstate->num_mixers = num_lm; + + return 0; +} + +/** + * dpu_crtc_check_mode_changed: check if full modeset is required + * @old_crtc_state: Previous CRTC state + * @new_crtc_state: Corresponding CRTC state to be checked + * + * Check if the changes in the object properties demand full mode set. + */ +int dpu_crtc_check_mode_changed(struct drm_crtc_state *old_crtc_state, + struct drm_crtc_state *new_crtc_state) +{ + struct drm_encoder *drm_enc; + struct drm_crtc *crtc = new_crtc_state->crtc; + bool clone_mode_enabled = drm_crtc_in_clone_mode(old_crtc_state); + bool clone_mode_requested = drm_crtc_in_clone_mode(new_crtc_state); + + DRM_DEBUG_ATOMIC("%d\n", crtc->base.id); + + /* there might be cases where encoder needs a modeset too */ + drm_for_each_encoder_mask(drm_enc, crtc->dev, new_crtc_state->encoder_mask) { + if (dpu_encoder_needs_modeset(drm_enc, new_crtc_state->state)) + new_crtc_state->mode_changed = true; + } + + if ((clone_mode_requested && !clone_mode_enabled) || + (!clone_mode_requested && clone_mode_enabled)) + new_crtc_state->mode_changed = true; + + return 0; +} + +static int dpu_crtc_atomic_check(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + struct drm_crtc_state *crtc_state = drm_atomic_get_new_crtc_state(state, + crtc); + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); + struct dpu_crtc_state *cstate = to_dpu_crtc_state(crtc_state); + + const struct drm_plane_state *pstate; + struct drm_plane *plane; + + int rc = 0; + + bool needs_dirtyfb = dpu_crtc_needs_dirtyfb(crtc_state); + + /* don't reallocate resources if only ACTIVE has beeen changed */ + if (crtc_state->mode_changed || crtc_state->connectors_changed || + crtc_state->color_mgmt_changed) { + rc = dpu_crtc_assign_resources(crtc, crtc_state); + if (rc < 0) + return rc; + } + + if (crtc_state->planes_changed || crtc_state->zpos_changed) { + if (dpu_use_virtual_planes) + rc = dpu_crtc_reassign_planes(crtc, crtc_state); + else + rc = dpu_crtc_assign_planes(crtc, crtc_state); + if (rc < 0) + return rc; + } + + if (!crtc_state->enable || !drm_atomic_crtc_effectively_active(crtc_state)) { + DRM_DEBUG_ATOMIC("crtc%d -> enable %d, active %d, skip atomic_check\n", + crtc->base.id, crtc_state->enable, + crtc_state->active); + memset(&cstate->new_perf, 0, sizeof(cstate->new_perf)); + return 0; + } + + DRM_DEBUG_ATOMIC("%s: check\n", dpu_crtc->name); + + if (cstate->num_mixers) { + rc = _dpu_crtc_check_and_setup_lm_bounds(crtc, crtc_state); + if (rc) + return rc; + } + + /* FIXME: move this to dpu_plane_atomic_check? */ + drm_atomic_crtc_state_for_each_plane_state(plane, pstate, crtc_state) { + struct dpu_plane_state *dpu_pstate = to_dpu_plane_state(pstate); + + if (IS_ERR_OR_NULL(pstate)) { + rc = PTR_ERR(pstate); + DPU_ERROR("%s: failed to get plane%d state, %d\n", + dpu_crtc->name, plane->base.id, rc); + return rc; + } + + if (!pstate->visible) + continue; + + dpu_pstate->needs_dirtyfb = needs_dirtyfb; + } + + atomic_inc(&_dpu_crtc_get_kms(crtc)->bandwidth_ref); + + rc = dpu_core_perf_crtc_check(crtc, crtc_state); + if (rc) { + DPU_ERROR("crtc%d failed performance check %d\n", + crtc->base.id, rc); + return rc; + } + + return 0; +} + +static enum drm_mode_status dpu_crtc_mode_valid(struct drm_crtc *crtc, + const struct drm_display_mode *mode) +{ + struct dpu_kms *dpu_kms = _dpu_crtc_get_kms(crtc); + u64 adjusted_mode_clk; + + /* if there is no 3d_mux block we cannot merge LMs so we cannot + * split the large layer into 2 LMs, filter out such modes + */ + if (!dpu_kms->catalog->caps->has_3d_merge && + mode->hdisplay > dpu_kms->catalog->caps->max_mixer_width) + return MODE_BAD_HVALUE; + + adjusted_mode_clk = dpu_core_perf_adjusted_mode_clk(mode->clock, + dpu_kms->perf.perf_cfg); + + if (dpu_kms->catalog->caps->has_3d_merge) + adjusted_mode_clk /= 2; + + /* + * The given mode, adjusted for the perf clock factor, should not exceed + * the max core clock rate + */ + if (dpu_kms->perf.max_core_clk_rate < adjusted_mode_clk * 1000) + return MODE_CLOCK_HIGH; + + /* + * max crtc width is equal to the max mixer width * 2 and max height is 4K + */ + return drm_mode_validate_size(mode, + 2 * dpu_kms->catalog->caps->max_mixer_width, + 4096); +} + +/** + * dpu_crtc_vblank - enable or disable vblanks for this crtc + * @crtc: Pointer to drm crtc object + * @en: true to enable vblanks, false to disable + */ +int dpu_crtc_vblank(struct drm_crtc *crtc, bool en) +{ + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); + struct drm_encoder *enc; + + trace_dpu_crtc_vblank(DRMID(&dpu_crtc->base), en, dpu_crtc); + + /* + * Normally we would iterate through encoder_mask in crtc state to find + * attached encoders. In this case, we might be disabling vblank _after_ + * encoder_mask has been cleared. + * + * Instead, we "assign" a crtc to the encoder in enable and clear it in + * disable (which is also after encoder_mask is cleared). So instead of + * using encoder mask, we'll ask the encoder to toggle itself iff it's + * currently assigned to our crtc. + * + * Note also that this function cannot be called while crtc is disabled + * since we use drm_crtc_vblank_on/off. So we don't need to worry + * about the assigned crtcs being inconsistent with the current state + * (which means no need to worry about modeset locks). + */ + list_for_each_entry(enc, &crtc->dev->mode_config.encoder_list, head) { + trace_dpu_crtc_vblank_enable(DRMID(crtc), DRMID(enc), en, + dpu_crtc); + + dpu_encoder_toggle_vblank_for_crtc(enc, crtc, en); + } + + return 0; +} + +/** + * dpu_crtc_get_num_lm - Get mixer number in this CRTC pipeline + * @state: Pointer to drm crtc state object + */ +unsigned int dpu_crtc_get_num_lm(const struct drm_crtc_state *state) +{ + struct dpu_crtc_state *cstate = to_dpu_crtc_state(state); + + return cstate->num_mixers; +} + +#ifdef CONFIG_DEBUG_FS +static int _dpu_debugfs_status_show(struct seq_file *s, void *data) +{ + struct dpu_crtc *dpu_crtc; + struct dpu_plane_state *pstate = NULL; + struct dpu_crtc_mixer *m; + + struct drm_crtc *crtc; + struct drm_plane *plane; + struct drm_display_mode *mode; + struct drm_framebuffer *fb; + struct drm_plane_state *state; + struct dpu_crtc_state *cstate; + + int i, out_width; + + dpu_crtc = s->private; + crtc = &dpu_crtc->base; + + drm_modeset_lock_all(crtc->dev); + cstate = to_dpu_crtc_state(crtc->state); + + mode = &crtc->state->adjusted_mode; + out_width = mode->hdisplay / cstate->num_mixers; + + seq_printf(s, "crtc:%d width:%d height:%d\n", crtc->base.id, + mode->hdisplay, mode->vdisplay); + + seq_puts(s, "\n"); + + for (i = 0; i < cstate->num_mixers; ++i) { + m = &cstate->mixers[i]; + seq_printf(s, "\tmixer:%d ctl:%d width:%d height:%d\n", + m->hw_lm->idx - LM_0, m->lm_ctl->idx - CTL_0, + out_width, mode->vdisplay); + } + + seq_puts(s, "\n"); + + drm_atomic_crtc_for_each_plane(plane, crtc) { + pstate = to_dpu_plane_state(plane->state); + state = plane->state; + + if (!pstate || !state) + continue; + + seq_printf(s, "\tplane:%u stage:%d\n", plane->base.id, + pstate->stage); + + if (plane->state->fb) { + fb = plane->state->fb; + + seq_printf(s, "\tfb:%d image format:%4.4s wxh:%ux%u ", + fb->base.id, (char *) &fb->format->format, + fb->width, fb->height); + for (i = 0; i < ARRAY_SIZE(fb->format->cpp); ++i) + seq_printf(s, "cpp[%d]:%u ", + i, fb->format->cpp[i]); + seq_puts(s, "\n\t"); + + seq_printf(s, "modifier:%8llu ", fb->modifier); + seq_puts(s, "\n"); + + seq_puts(s, "\t"); + for (i = 0; i < ARRAY_SIZE(fb->pitches); i++) + seq_printf(s, "pitches[%d]:%8u ", i, + fb->pitches[i]); + seq_puts(s, "\n"); + + seq_puts(s, "\t"); + for (i = 0; i < ARRAY_SIZE(fb->offsets); i++) + seq_printf(s, "offsets[%d]:%8u ", i, + fb->offsets[i]); + seq_puts(s, "\n"); + } + + seq_printf(s, "\tsrc_x:%4d src_y:%4d src_w:%4d src_h:%4d\n", + state->src_x, state->src_y, state->src_w, state->src_h); + + seq_printf(s, "\tdst x:%4d dst_y:%4d dst_w:%4d dst_h:%4d\n", + state->crtc_x, state->crtc_y, state->crtc_w, + state->crtc_h); + + for (i = 0; i < PIPES_PER_PLANE; i++) { + if (!pstate->pipe[i].sspp) + continue; + seq_printf(s, "\tsspp[%d]:%s\n", + i, pstate->pipe[i].sspp->cap->name); + seq_printf(s, "\tmultirect[%d]: mode: %d index: %d\n", + i, pstate->pipe[i].multirect_mode, + pstate->pipe[i].multirect_index); + } + + seq_puts(s, "\n"); + } + if (dpu_crtc->vblank_cb_count) { + ktime_t diff = ktime_sub(ktime_get(), dpu_crtc->vblank_cb_time); + s64 diff_ms = ktime_to_ms(diff); + s64 fps = diff_ms ? div_s64( + dpu_crtc->vblank_cb_count * 1000, diff_ms) : 0; + + seq_printf(s, + "vblank fps:%lld count:%u total:%llums total_framecount:%llu\n", + fps, dpu_crtc->vblank_cb_count, + ktime_to_ms(diff), dpu_crtc->play_count); + + /* reset time & count for next measurement */ + dpu_crtc->vblank_cb_count = 0; + dpu_crtc->vblank_cb_time = ktime_set(0, 0); + } + + drm_modeset_unlock_all(crtc->dev); + + return 0; +} + +DEFINE_SHOW_ATTRIBUTE(_dpu_debugfs_status); + +static int dpu_crtc_debugfs_state_show(struct seq_file *s, void *v) +{ + struct drm_crtc *crtc = s->private; + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); + + seq_printf(s, "client type: %d\n", dpu_crtc_get_client_type(crtc)); + seq_printf(s, "intf_mode: %d\n", dpu_crtc_get_intf_mode(crtc)); + seq_printf(s, "core_clk_rate: %llu\n", + dpu_crtc->cur_perf.core_clk_rate); + seq_printf(s, "bw_ctl: %uk\n", + (u32)DIV_ROUND_UP_ULL(dpu_crtc->cur_perf.bw_ctl, 1000)); + seq_printf(s, "max_per_pipe_ib: %u\n", + dpu_crtc->cur_perf.max_per_pipe_ib); + + return 0; +} +DEFINE_SHOW_ATTRIBUTE(dpu_crtc_debugfs_state); + +static int _dpu_crtc_init_debugfs(struct drm_crtc *crtc) +{ + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); + + debugfs_create_file("status", 0400, + crtc->debugfs_entry, + dpu_crtc, &_dpu_debugfs_status_fops); + debugfs_create_file("state", 0600, + crtc->debugfs_entry, + &dpu_crtc->base, + &dpu_crtc_debugfs_state_fops); + + return 0; +} +#else +static int _dpu_crtc_init_debugfs(struct drm_crtc *crtc) +{ + return 0; +} +#endif /* CONFIG_DEBUG_FS */ + +static int dpu_crtc_late_register(struct drm_crtc *crtc) +{ + return _dpu_crtc_init_debugfs(crtc); +} + +static const struct drm_crtc_funcs dpu_crtc_funcs = { + .set_config = drm_atomic_helper_set_config, + .page_flip = drm_atomic_helper_page_flip, + .reset = dpu_crtc_reset, + .atomic_duplicate_state = dpu_crtc_duplicate_state, + .atomic_destroy_state = dpu_crtc_destroy_state, + .atomic_print_state = dpu_crtc_atomic_print_state, + .late_register = dpu_crtc_late_register, + .verify_crc_source = dpu_crtc_verify_crc_source, + .set_crc_source = dpu_crtc_set_crc_source, + .enable_vblank = msm_crtc_enable_vblank, + .disable_vblank = msm_crtc_disable_vblank, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, + .get_vblank_counter = dpu_crtc_get_vblank_counter, +}; + +static const struct drm_crtc_helper_funcs dpu_crtc_helper_funcs = { + .atomic_disable = dpu_crtc_disable, + .atomic_enable = dpu_crtc_enable, + .atomic_check = dpu_crtc_atomic_check, + .atomic_begin = dpu_crtc_atomic_begin, + .atomic_flush = dpu_crtc_atomic_flush, + .mode_valid = dpu_crtc_mode_valid, + .get_scanout_position = dpu_crtc_get_scanout_position, +}; + +/** + * dpu_crtc_init - create a new crtc object + * @dev: dpu device + * @plane: base plane + * @cursor: cursor plane + * @return: new crtc object or error + * + * initialize CRTC + */ +struct drm_crtc *dpu_crtc_init(struct drm_device *dev, struct drm_plane *plane, + struct drm_plane *cursor) +{ + struct msm_drm_private *priv = dev->dev_private; + struct dpu_kms *dpu_kms = to_dpu_kms(priv->kms); + struct drm_crtc *crtc = NULL; + struct dpu_crtc *dpu_crtc; + int i, ret; + + dpu_crtc = drmm_crtc_alloc_with_planes(dev, struct dpu_crtc, base, + plane, cursor, + &dpu_crtc_funcs, + NULL); + + if (IS_ERR(dpu_crtc)) + return ERR_CAST(dpu_crtc); + + crtc = &dpu_crtc->base; + crtc->dev = dev; + + spin_lock_init(&dpu_crtc->spin_lock); + atomic_set(&dpu_crtc->frame_pending, 0); + + init_completion(&dpu_crtc->frame_done_comp); + + INIT_LIST_HEAD(&dpu_crtc->frame_event_list); + + for (i = 0; i < ARRAY_SIZE(dpu_crtc->frame_events); i++) { + INIT_LIST_HEAD(&dpu_crtc->frame_events[i].list); + list_add(&dpu_crtc->frame_events[i].list, + &dpu_crtc->frame_event_list); + kthread_init_work(&dpu_crtc->frame_events[i].work, + dpu_crtc_frame_event_work); + } + + drm_crtc_helper_add(crtc, &dpu_crtc_helper_funcs); + + if (dpu_kms->catalog->dspp_count) { + const struct dpu_dspp_cfg *dspp = &dpu_kms->catalog->dspp[0]; + + if (dspp->sblk->gc.base) { + drm_mode_crtc_set_gamma_size(crtc, DPU_GAMMA_LUT_SIZE); + drm_crtc_enable_color_mgmt(crtc, 0, true, DPU_GAMMA_LUT_SIZE); + } else { + drm_crtc_enable_color_mgmt(crtc, 0, true, 0); + } + } + + /* save user friendly CRTC name for later */ + snprintf(dpu_crtc->name, DPU_CRTC_NAME_SIZE, "crtc%u", crtc->base.id); + + /* initialize event handling */ + spin_lock_init(&dpu_crtc->event_lock); + + ret = drm_self_refresh_helper_init(crtc); + if (ret) { + DPU_ERROR("Failed to initialize %s with self-refresh helpers %d\n", + crtc->name, ret); + return ERR_PTR(ret); + } + + DRM_DEBUG_KMS("%s: successfully initialized crtc\n", dpu_crtc->name); + return crtc; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.h new file mode 100644 index 000000000..6eaba5696 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.h @@ -0,0 +1,272 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2021 The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#ifndef _DPU_CRTC_H_ +#define _DPU_CRTC_H_ + +#include +#include +#include "dpu_kms.h" +#include "dpu_core_perf.h" + +#define DPU_CRTC_NAME_SIZE 12 + +/* define the maximum number of in-flight frame events */ +#define DPU_CRTC_FRAME_EVENT_SIZE 4 + +/** + * enum dpu_crtc_client_type: crtc client type + * @RT_CLIENT: RealTime client like video/cmd mode display + * voting through apps rsc + * @NRT_CLIENT: Non-RealTime client like WB display + * voting through apps rsc + */ +enum dpu_crtc_client_type { + RT_CLIENT, + NRT_CLIENT, +}; + +/** + * enum dpu_crtc_smmu_state: smmu state + * @ATTACHED: all the context banks are attached. + * @DETACHED: all the context banks are detached. + * @ATTACH_ALL_REQ: transient state of attaching context banks. + * @DETACH_ALL_REQ: transient state of detaching context banks. + */ +enum dpu_crtc_smmu_state { + ATTACHED = 0, + DETACHED, + ATTACH_ALL_REQ, + DETACH_ALL_REQ, +}; + +/** + * enum dpu_crtc_smmu_state_transition_type: state transition type + * @NONE: no pending state transitions + * @PRE_COMMIT: state transitions should be done before processing the commit + * @POST_COMMIT: state transitions to be done after processing the commit. + */ +enum dpu_crtc_smmu_state_transition_type { + NONE, + PRE_COMMIT, + POST_COMMIT +}; + +/** + * struct dpu_crtc_smmu_state_data: stores the smmu state and transition type + * @state: current state of smmu context banks + * @transition_type: transition request type + * @transition_error: whether there is error while transitioning the state + */ +struct dpu_crtc_smmu_state_data { + uint32_t state; + uint32_t transition_type; + uint32_t transition_error; +}; + +/** + * enum dpu_crtc_crc_source: CRC source + * @DPU_CRTC_CRC_SOURCE_NONE: no source set + * @DPU_CRTC_CRC_SOURCE_LAYER_MIXER: CRC in layer mixer + * @DPU_CRTC_CRC_SOURCE_ENCODER: CRC in encoder + * @DPU_CRTC_CRC_SOURCE_INVALID: Invalid source + */ +enum dpu_crtc_crc_source { + DPU_CRTC_CRC_SOURCE_NONE = 0, + DPU_CRTC_CRC_SOURCE_LAYER_MIXER, + DPU_CRTC_CRC_SOURCE_ENCODER, + DPU_CRTC_CRC_SOURCE_MAX, + DPU_CRTC_CRC_SOURCE_INVALID = -1 +}; + +/** + * struct dpu_crtc_mixer: stores the map for each virtual pipeline in the CRTC + * @hw_lm: LM HW Driver context + * @lm_ctl: CTL Path HW driver context + * @lm_dspp: DSPP HW driver context + * @mixer_op_mode: mixer blending operation mode + * @flush_mask: mixer flush mask for ctl, mixer and pipe + */ +struct dpu_crtc_mixer { + struct dpu_hw_mixer *hw_lm; + struct dpu_hw_ctl *lm_ctl; + struct dpu_hw_dspp *hw_dspp; + u32 mixer_op_mode; +}; + +/** + * struct dpu_crtc_frame_event: stores crtc frame event for crtc processing + * @work: base work structure + * @crtc: Pointer to crtc handling this event + * @list: event list + * @ts: timestamp at queue entry + * @event: event identifier + */ +struct dpu_crtc_frame_event { + struct kthread_work work; + struct drm_crtc *crtc; + struct list_head list; + ktime_t ts; + u32 event; +}; + +/* + * Maximum number of free event structures to cache + */ +#define DPU_CRTC_MAX_EVENT_COUNT 16 + +/** + * struct dpu_crtc - virtualized CRTC data structure + * @base : Base drm crtc structure + * @name : ASCII description of this crtc + * @event : Pointer to last received drm vblank event. If there is a + * pending vblank event, this will be non-null. + * @vsync_count : Running count of received vsync events + * @drm_requested_vblank : Whether vblanks have been enabled in the encoder + * @property_info : Opaque structure for generic property support + * @property_defaults : Array of default values for generic property support + * @vblank_cb_count : count of vblank callback since last reset + * @play_count : frame count between crtc enable and disable + * @vblank_cb_time : ktime at vblank count reset + * @enabled : whether the DPU CRTC is currently enabled. updated in the + * commit-thread, not state-swap time which is earlier, so + * safe to make decisions on during VBLANK on/off work + * @feature_list : list of color processing features supported on a crtc + * @active_list : list of color processing features are active + * @dirty_list : list of color processing features are dirty + * @ad_dirty: list containing ad properties that are dirty + * @ad_active: list containing ad properties that are active + * @frame_pending : Whether or not an update is pending + * @frame_events : static allocation of in-flight frame events + * @frame_event_list : available frame event list + * @spin_lock : spin lock for frame event, transaction status, etc... + * @frame_done_comp : for frame_event_done synchronization + * @event_thread : Pointer to event handler thread + * @event_worker : Event worker queue + * @event_lock : Spinlock around event handling code + * @phandle: Pointer to power handler + * @cur_perf : current performance committed to clock/bandwidth driver + * @crc_source : CRC source + */ +struct dpu_crtc { + struct drm_crtc base; + char name[DPU_CRTC_NAME_SIZE]; + + struct drm_pending_vblank_event *event; + u32 vsync_count; + + u32 vblank_cb_count; + u64 play_count; + ktime_t vblank_cb_time; + bool enabled; + + struct list_head feature_list; + struct list_head active_list; + struct list_head dirty_list; + struct list_head ad_dirty; + struct list_head ad_active; + + atomic_t frame_pending; + struct dpu_crtc_frame_event frame_events[DPU_CRTC_FRAME_EVENT_SIZE]; + struct list_head frame_event_list; + spinlock_t spin_lock; + struct completion frame_done_comp; + + /* for handling internal event thread */ + spinlock_t event_lock; + + struct dpu_core_perf_params cur_perf; + + struct dpu_crtc_smmu_state_data smmu_state; +}; + +#define to_dpu_crtc(x) container_of(x, struct dpu_crtc, base) + +/** + * struct dpu_crtc_state - dpu container for atomic crtc state + * @base: Base drm crtc state structure + * @bw_control : true if bw/clk controlled by core bw/clk properties + * @bw_split_vote : true if bw controlled by llcc/dram bw properties + * @lm_bounds : LM boundaries based on current mode full resolution, no ROI. + * Origin top left of CRTC. + * @property_state: Local storage for msm_prop properties + * @property_values: Current crtc property values + * @input_fence_timeout_ns : Cached input fence timeout, in ns + * @new_perf: new performance state being requested + * @num_mixers : Number of mixers in use + * @mixers : List of active mixers + * @num_ctls : Number of ctl paths in use + * @hw_ctls : List of active ctl paths + * @crc_source : CRC source + * @crc_frame_skip_count: Number of frames skipped before getting CRC + */ +struct dpu_crtc_state { + struct drm_crtc_state base; + + bool bw_control; + bool bw_split_vote; + struct drm_rect lm_bounds[CRTC_DUAL_MIXERS]; + + uint64_t input_fence_timeout_ns; + + struct dpu_core_perf_params new_perf; + + /* HW Resources reserved for the crtc */ + u32 num_mixers; + struct dpu_crtc_mixer mixers[CRTC_DUAL_MIXERS]; + + u32 num_ctls; + struct dpu_hw_ctl *hw_ctls[CRTC_DUAL_MIXERS]; + + enum dpu_crtc_crc_source crc_source; + int crc_frame_skip_count; +}; + +#define to_dpu_crtc_state(x) \ + container_of(x, struct dpu_crtc_state, base) + +/** + * dpu_crtc_frame_pending - return the number of pending frames + * @crtc: Pointer to drm crtc object + */ +static inline int dpu_crtc_frame_pending(struct drm_crtc *crtc) +{ + return crtc ? atomic_read(&to_dpu_crtc(crtc)->frame_pending) : -EINVAL; +} + +int dpu_crtc_check_mode_changed(struct drm_crtc_state *old_crtc_state, + struct drm_crtc_state *new_crtc_state); + +int dpu_crtc_vblank(struct drm_crtc *crtc, bool en); + +void dpu_crtc_vblank_callback(struct drm_crtc *crtc); + +void dpu_crtc_commit_kickoff(struct drm_crtc *crtc); + +void dpu_crtc_complete_commit(struct drm_crtc *crtc); + +struct drm_crtc *dpu_crtc_init(struct drm_device *dev, struct drm_plane *plane, + struct drm_plane *cursor); + +enum dpu_intf_mode dpu_crtc_get_intf_mode(struct drm_crtc *crtc); + +/** + * dpu_crtc_get_client_type - check the crtc type- rt, nrt etc. + * @crtc: Pointer to crtc + */ +static inline enum dpu_crtc_client_type dpu_crtc_get_client_type( + struct drm_crtc *crtc) +{ + return crtc && crtc->state ? RT_CLIENT : NRT_CLIENT; +} + +void dpu_crtc_frame_event_cb(struct drm_crtc *crtc, u32 event); + +unsigned int dpu_crtc_get_num_lm(const struct drm_crtc_state *state); + +#endif /* _DPU_CRTC_H_ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c new file mode 100644 index 000000000..1f20695f8 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c @@ -0,0 +1,2946 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013 Red Hat + * Copyright (c) 2014-2018, 2020-2021 The Linux Foundation. All rights reserved. + * Copyright (c) 2022-2024 Qualcomm Innovation Center, Inc. All rights reserved. + * + * Author: Rob Clark + */ + +#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ +#include +#include +#include + +#include +#include +#include +#include +#include + +#include "msm_drv.h" +#include "dpu_kms.h" +#include "dpu_hwio.h" +#include "dpu_hw_catalog.h" +#include "dpu_hw_intf.h" +#include "dpu_hw_ctl.h" +#include "dpu_hw_cwb.h" +#include "dpu_hw_dspp.h" +#include "dpu_hw_dsc.h" +#include "dpu_hw_merge3d.h" +#include "dpu_hw_cdm.h" +#include "dpu_formats.h" +#include "dpu_encoder_phys.h" +#include "dpu_crtc.h" +#include "dpu_trace.h" +#include "dpu_core_irq.h" +#include "disp/msm_disp_snapshot.h" + +#define DPU_DEBUG_ENC(e, fmt, ...) DRM_DEBUG_ATOMIC("enc%d " fmt,\ + (e) ? (e)->base.base.id : -1, ##__VA_ARGS__) + +#define DPU_ERROR_ENC(e, fmt, ...) DPU_ERROR("enc%d " fmt,\ + (e) ? (e)->base.base.id : -1, ##__VA_ARGS__) + +#define DPU_ERROR_ENC_RATELIMITED(e, fmt, ...) DPU_ERROR_RATELIMITED("enc%d " fmt,\ + (e) ? (e)->base.base.id : -1, ##__VA_ARGS__) + +/* + * Two to anticipate panels that can do cmd/vid dynamic switching + * plan is to create all possible physical encoder types, and switch between + * them at runtime + */ +#define NUM_PHYS_ENCODER_TYPES 2 + +#define MAX_PHYS_ENCODERS_PER_VIRTUAL \ + (MAX_H_TILES_PER_DISPLAY * NUM_PHYS_ENCODER_TYPES) + +#define MAX_CHANNELS_PER_ENC 2 +#define MAX_CWB_PER_ENC 2 + +#define IDLE_SHORT_TIMEOUT 1 + +/* timeout in frames waiting for frame done */ +#define DPU_ENCODER_FRAME_DONE_TIMEOUT_FRAMES 5 + +/** + * enum dpu_enc_rc_events - events for resource control state machine + * @DPU_ENC_RC_EVENT_KICKOFF: + * This event happens at NORMAL priority. + * Event that signals the start of the transfer. When this event is + * received, enable MDP/DSI core clocks. Regardless of the previous + * state, the resource should be in ON state at the end of this event. + * @DPU_ENC_RC_EVENT_FRAME_DONE: + * This event happens at INTERRUPT level. + * Event signals the end of the data transfer after the PP FRAME_DONE + * event. At the end of this event, a delayed work is scheduled to go to + * IDLE_PC state after IDLE_TIMEOUT time. + * @DPU_ENC_RC_EVENT_PRE_STOP: + * This event happens at NORMAL priority. + * This event, when received during the ON state, leave the RC STATE + * in the PRE_OFF state. It should be followed by the STOP event as + * part of encoder disable. + * If received during IDLE or OFF states, it will do nothing. + * @DPU_ENC_RC_EVENT_STOP: + * This event happens at NORMAL priority. + * When this event is received, disable all the MDP/DSI core clocks, and + * disable IRQs. It should be called from the PRE_OFF or IDLE states. + * IDLE is expected when IDLE_PC has run, and PRE_OFF did nothing. + * PRE_OFF is expected when PRE_STOP was executed during the ON state. + * Resource state should be in OFF at the end of the event. + * @DPU_ENC_RC_EVENT_ENTER_IDLE: + * This event happens at NORMAL priority from a work item. + * Event signals that there were no frame updates for IDLE_TIMEOUT time. + * This would disable MDP/DSI core clocks and change the resource state + * to IDLE. + */ +enum dpu_enc_rc_events { + DPU_ENC_RC_EVENT_KICKOFF = 1, + DPU_ENC_RC_EVENT_FRAME_DONE, + DPU_ENC_RC_EVENT_PRE_STOP, + DPU_ENC_RC_EVENT_STOP, + DPU_ENC_RC_EVENT_ENTER_IDLE +}; + +/* + * enum dpu_enc_rc_states - states that the resource control maintains + * @DPU_ENC_RC_STATE_OFF: Resource is in OFF state + * @DPU_ENC_RC_STATE_PRE_OFF: Resource is transitioning to OFF state + * @DPU_ENC_RC_STATE_ON: Resource is in ON state + * @DPU_ENC_RC_STATE_MODESET: Resource is in modeset state + * @DPU_ENC_RC_STATE_IDLE: Resource is in IDLE state + */ +enum dpu_enc_rc_states { + DPU_ENC_RC_STATE_OFF, + DPU_ENC_RC_STATE_PRE_OFF, + DPU_ENC_RC_STATE_ON, + DPU_ENC_RC_STATE_IDLE +}; + +/** + * struct dpu_encoder_virt - virtual encoder. Container of one or more physical + * encoders. Virtual encoder manages one "logical" display. Physical + * encoders manage one intf block, tied to a specific panel/sub-panel. + * Virtual encoder defers as much as possible to the physical encoders. + * Virtual encoder registers itself with the DRM Framework as the encoder. + * @base: drm_encoder base class for registration with DRM + * @enc_spinlock: Virtual-Encoder-Wide Spin Lock for IRQ purposes + * @enabled: True if the encoder is active, protected by enc_lock + * @commit_done_timedout: True if there has been a timeout on commit after + * enabling the encoder. + * @num_phys_encs: Actual number of physical encoders contained. + * @phys_encs: Container of physical encoders managed. + * @cur_master: Pointer to the current master in this mode. Optimization + * Only valid after enable. Cleared as disable. + * @cur_slave: As above but for the slave encoder. + * @hw_pp: Handle to the pingpong blocks used for the display. No. + * pingpong blocks can be different than num_phys_encs. + * @hw_cwb: Handle to the CWB muxes used for concurrent writeback + * display. Number of CWB muxes can be different than + * num_phys_encs. + * @hw_dsc: Handle to the DSC blocks used for the display. + * @dsc_mask: Bitmask of used DSC blocks. + * @cwb_mask: Bitmask of used CWB muxes + * @intfs_swapped: Whether or not the phys_enc interfaces have been swapped + * for partial update right-only cases, such as pingpong + * split where virtual pingpong does not generate IRQs + * @crtc: Pointer to the currently assigned crtc. Normally you + * would use crtc->state->encoder_mask to determine the + * link between encoder/crtc. However in this case we need + * to track crtc in the disable() hook which is called + * _after_ encoder_mask is cleared. + * @connector: If a mode is set, cached pointer to the active connector + * @enc_lock: Lock around physical encoder + * create/destroy/enable/disable + * @frame_busy_mask: Bitmask tracking which phys_enc we are still + * busy processing current command. + * Bit0 = phys_encs[0] etc. + * @frame_done_timeout_ms: frame done timeout in ms + * @frame_done_timeout_cnt: atomic counter tracking the number of frame + * done timeouts + * @frame_done_timer: watchdog timer for frame done event + * @disp_info: local copy of msm_display_info struct + * @idle_pc_supported: indicate if idle power collaps is supported + * @rc_lock: resource control mutex lock to protect + * virt encoder over various state changes + * @rc_state: resource controller state + * @delayed_off_work: delayed worker to schedule disabling of + * clks and resources after IDLE_TIMEOUT time. + * @topology: topology of the display + * @idle_timeout: idle timeout duration in milliseconds + * @wide_bus_en: wide bus is enabled on this interface + * @dsc: drm_dsc_config pointer, for DSC-enabled encoders + */ +struct dpu_encoder_virt { + struct drm_encoder base; + spinlock_t enc_spinlock; + + bool enabled; + bool commit_done_timedout; + + unsigned int num_phys_encs; + struct dpu_encoder_phys *phys_encs[MAX_PHYS_ENCODERS_PER_VIRTUAL]; + struct dpu_encoder_phys *cur_master; + struct dpu_encoder_phys *cur_slave; + struct dpu_hw_pingpong *hw_pp[MAX_CHANNELS_PER_ENC]; + struct dpu_hw_cwb *hw_cwb[MAX_CWB_PER_ENC]; + struct dpu_hw_dsc *hw_dsc[MAX_CHANNELS_PER_ENC]; + + unsigned int dsc_mask; + unsigned int cwb_mask; + + bool intfs_swapped; + + struct drm_crtc *crtc; + struct drm_connector *connector; + + struct mutex enc_lock; + DECLARE_BITMAP(frame_busy_mask, MAX_PHYS_ENCODERS_PER_VIRTUAL); + + atomic_t frame_done_timeout_ms; + atomic_t frame_done_timeout_cnt; + struct timer_list frame_done_timer; + + struct msm_display_info disp_info; + + bool idle_pc_supported; + struct mutex rc_lock; + enum dpu_enc_rc_states rc_state; + struct delayed_work delayed_off_work; + struct msm_display_topology topology; + + u32 idle_timeout; + + bool wide_bus_en; + + /* DSC configuration */ + struct drm_dsc_config *dsc; +}; + +#define to_dpu_encoder_virt(x) container_of(x, struct dpu_encoder_virt, base) + +static u32 dither_matrix[DITHER_MATRIX_SZ] = { + 15, 7, 13, 5, 3, 11, 1, 9, 12, 4, 14, 6, 0, 8, 2, 10 +}; + +/** + * dpu_encoder_get_drm_fmt - return DRM fourcc format + * @phys_enc: Pointer to physical encoder structure + */ +u32 dpu_encoder_get_drm_fmt(struct dpu_encoder_phys *phys_enc) +{ + struct drm_encoder *drm_enc; + struct dpu_encoder_virt *dpu_enc; + struct drm_display_info *info; + struct drm_display_mode *mode; + + drm_enc = phys_enc->parent; + dpu_enc = to_dpu_encoder_virt(drm_enc); + info = &dpu_enc->connector->display_info; + mode = &phys_enc->cached_mode; + + if (drm_mode_is_420_only(info, mode)) + return DRM_FORMAT_YUV420; + + return DRM_FORMAT_RGB888; +} + +/** + * dpu_encoder_needs_periph_flush - return true if physical encoder requires + * peripheral flush + * @phys_enc: Pointer to physical encoder structure + */ +bool dpu_encoder_needs_periph_flush(struct dpu_encoder_phys *phys_enc) +{ + struct drm_encoder *drm_enc; + struct dpu_encoder_virt *dpu_enc; + struct msm_display_info *disp_info; + struct msm_drm_private *priv; + struct drm_display_mode *mode; + + drm_enc = phys_enc->parent; + dpu_enc = to_dpu_encoder_virt(drm_enc); + disp_info = &dpu_enc->disp_info; + priv = drm_enc->dev->dev_private; + mode = &phys_enc->cached_mode; + + return phys_enc->hw_intf->cap->type == INTF_DP && + msm_dp_needs_periph_flush(priv->kms->dp[disp_info->h_tile_instance[0]], mode); +} + +/** + * dpu_encoder_is_widebus_enabled - return bool value if widebus is enabled + * @drm_enc: Pointer to previously created drm encoder structure + */ +bool dpu_encoder_is_widebus_enabled(const struct drm_encoder *drm_enc) +{ + const struct dpu_encoder_virt *dpu_enc; + struct msm_drm_private *priv = drm_enc->dev->dev_private; + const struct msm_display_info *disp_info; + int index; + + dpu_enc = to_dpu_encoder_virt(drm_enc); + disp_info = &dpu_enc->disp_info; + index = disp_info->h_tile_instance[0]; + + if (disp_info->intf_type == INTF_DP) + return msm_dp_wide_bus_available(priv->kms->dp[index]); + else if (disp_info->intf_type == INTF_DSI) + return msm_dsi_wide_bus_enabled(priv->kms->dsi[index]); + + return false; +} + +/** + * dpu_encoder_is_dsc_enabled - indicate whether dsc is enabled + * for the encoder. + * @drm_enc: Pointer to previously created drm encoder structure + */ +bool dpu_encoder_is_dsc_enabled(const struct drm_encoder *drm_enc) +{ + const struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); + + return dpu_enc->dsc ? true : false; +} + +/** + * dpu_encoder_get_crc_values_cnt - get number of physical encoders contained + * in virtual encoder that can collect CRC values + * @drm_enc: Pointer to previously created drm encoder structure + * Returns: Number of physical encoders for given drm encoder + */ +int dpu_encoder_get_crc_values_cnt(const struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc; + int i, num_intf = 0; + + dpu_enc = to_dpu_encoder_virt(drm_enc); + + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + + if (phys->hw_intf && phys->hw_intf->ops.setup_misr + && phys->hw_intf->ops.collect_misr) + num_intf++; + } + + return num_intf; +} + +/** + * dpu_encoder_setup_misr - enable misr calculations + * @drm_enc: Pointer to previously created drm encoder structure + */ +void dpu_encoder_setup_misr(const struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc; + + int i; + + dpu_enc = to_dpu_encoder_virt(drm_enc); + + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + + if (!phys->hw_intf || !phys->hw_intf->ops.setup_misr) + continue; + + phys->hw_intf->ops.setup_misr(phys->hw_intf); + } +} + +/** + * dpu_encoder_get_crc - get the crc value from interface blocks + * @drm_enc: Pointer to previously created drm encoder structure + * @crcs: array to fill with CRC data + * @pos: offset into the @crcs array + * Returns: 0 on success, error otherwise + */ +int dpu_encoder_get_crc(const struct drm_encoder *drm_enc, u32 *crcs, int pos) +{ + struct dpu_encoder_virt *dpu_enc; + + int i, rc = 0, entries_added = 0; + + if (!drm_enc->crtc) { + DRM_ERROR("no crtc found for encoder %d\n", drm_enc->index); + return -EINVAL; + } + + dpu_enc = to_dpu_encoder_virt(drm_enc); + + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + + if (!phys->hw_intf || !phys->hw_intf->ops.collect_misr) + continue; + + rc = phys->hw_intf->ops.collect_misr(phys->hw_intf, &crcs[pos + entries_added]); + if (rc) + return rc; + entries_added++; + } + + return entries_added; +} + +static void _dpu_encoder_setup_dither(struct dpu_hw_pingpong *hw_pp, unsigned bpc) +{ + struct dpu_hw_dither_cfg dither_cfg = { 0 }; + + if (!hw_pp->ops.setup_dither) + return; + + switch (bpc) { + case 6: + dither_cfg.c0_bitdepth = 6; + dither_cfg.c1_bitdepth = 6; + dither_cfg.c2_bitdepth = 6; + dither_cfg.c3_bitdepth = 6; + dither_cfg.temporal_en = 0; + break; + default: + hw_pp->ops.setup_dither(hw_pp, NULL); + return; + } + + memcpy(&dither_cfg.matrix, dither_matrix, + sizeof(u32) * DITHER_MATRIX_SZ); + + hw_pp->ops.setup_dither(hw_pp, &dither_cfg); +} + +static char *dpu_encoder_helper_get_intf_type(enum dpu_intf_mode intf_mode) +{ + switch (intf_mode) { + case INTF_MODE_VIDEO: + return "INTF_MODE_VIDEO"; + case INTF_MODE_CMD: + return "INTF_MODE_CMD"; + case INTF_MODE_WB_BLOCK: + return "INTF_MODE_WB_BLOCK"; + case INTF_MODE_WB_LINE: + return "INTF_MODE_WB_LINE"; + default: + return "INTF_MODE_UNKNOWN"; + } +} + +/** + * dpu_encoder_helper_report_irq_timeout - utility to report error that irq has + * timed out, including reporting frame error event to crtc and debug dump + * @phys_enc: Pointer to physical encoder structure + * @intr_idx: Failing interrupt index + */ +void dpu_encoder_helper_report_irq_timeout(struct dpu_encoder_phys *phys_enc, + enum dpu_intr_idx intr_idx) +{ + DRM_ERROR("irq timeout id=%u, intf_mode=%s intf=%d wb=%d, pp=%d, intr=%d\n", + DRMID(phys_enc->parent), + dpu_encoder_helper_get_intf_type(phys_enc->intf_mode), + phys_enc->hw_intf ? phys_enc->hw_intf->idx - INTF_0 : -1, + phys_enc->hw_wb ? phys_enc->hw_wb->idx - WB_0 : -1, + phys_enc->hw_pp->idx - PINGPONG_0, intr_idx); + + dpu_encoder_frame_done_callback(phys_enc->parent, phys_enc, + DPU_ENCODER_FRAME_EVENT_ERROR); +} + +static int dpu_encoder_helper_wait_event_timeout(int32_t drm_id, + u32 irq_idx, struct dpu_encoder_wait_info *info); + +/** + * dpu_encoder_helper_wait_for_irq - utility to wait on an irq. + * note: will call dpu_encoder_helper_wait_for_irq on timeout + * @phys_enc: Pointer to physical encoder structure + * @irq_idx: IRQ index + * @func: IRQ callback to be called in case of timeout + * @wait_info: wait info struct + * @return: 0 or -ERROR + */ +int dpu_encoder_helper_wait_for_irq(struct dpu_encoder_phys *phys_enc, + unsigned int irq_idx, + void (*func)(void *arg), + struct dpu_encoder_wait_info *wait_info) +{ + u32 irq_status; + int ret; + + if (!wait_info) { + DPU_ERROR("invalid params\n"); + return -EINVAL; + } + /* note: do master / slave checking outside */ + + /* return EWOULDBLOCK since we know the wait isn't necessary */ + if (phys_enc->enable_state == DPU_ENC_DISABLED) { + DRM_ERROR("encoder is disabled id=%u, callback=%ps, IRQ=[%d, %d]\n", + DRMID(phys_enc->parent), func, + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx)); + return -EWOULDBLOCK; + } + + if (irq_idx == 0) { + DRM_DEBUG_KMS("skip irq wait id=%u, callback=%ps\n", + DRMID(phys_enc->parent), func); + return 0; + } + + DRM_DEBUG_KMS("id=%u, callback=%ps, IRQ=[%d, %d], pp=%d, pending_cnt=%d\n", + DRMID(phys_enc->parent), func, + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx), phys_enc->hw_pp->idx - PINGPONG_0, + atomic_read(wait_info->atomic_cnt)); + + ret = dpu_encoder_helper_wait_event_timeout( + DRMID(phys_enc->parent), + irq_idx, + wait_info); + + if (ret <= 0) { + irq_status = dpu_core_irq_read(phys_enc->dpu_kms, irq_idx); + if (irq_status) { + unsigned long flags; + + DRM_DEBUG_KMS("IRQ=[%d, %d] not triggered id=%u, callback=%ps, pp=%d, atomic_cnt=%d\n", + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx), + DRMID(phys_enc->parent), func, + phys_enc->hw_pp->idx - PINGPONG_0, + atomic_read(wait_info->atomic_cnt)); + local_irq_save(flags); + func(phys_enc); + local_irq_restore(flags); + ret = 0; + } else { + ret = -ETIMEDOUT; + DRM_DEBUG_KMS("IRQ=[%d, %d] timeout id=%u, callback=%ps, pp=%d, atomic_cnt=%d\n", + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx), + DRMID(phys_enc->parent), func, + phys_enc->hw_pp->idx - PINGPONG_0, + atomic_read(wait_info->atomic_cnt)); + } + } else { + ret = 0; + trace_dpu_enc_irq_wait_success(DRMID(phys_enc->parent), + func, DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx), + phys_enc->hw_pp->idx - PINGPONG_0, + atomic_read(wait_info->atomic_cnt)); + } + + return ret; +} + +/** + * dpu_encoder_get_vsync_count - get vsync count for the encoder. + * @drm_enc: Pointer to previously created drm encoder structure + */ +int dpu_encoder_get_vsync_count(struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); + struct dpu_encoder_phys *phys = dpu_enc ? dpu_enc->cur_master : NULL; + return phys ? atomic_read(&phys->vsync_cnt) : 0; +} + +/** + * dpu_encoder_get_linecount - get interface line count for the encoder. + * @drm_enc: Pointer to previously created drm encoder structure + */ +int dpu_encoder_get_linecount(struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc; + struct dpu_encoder_phys *phys; + int linecount = 0; + + dpu_enc = to_dpu_encoder_virt(drm_enc); + phys = dpu_enc ? dpu_enc->cur_master : NULL; + + if (phys && phys->ops.get_line_count) + linecount = phys->ops.get_line_count(phys); + + return linecount; +} + +/** + * dpu_encoder_helper_split_config - split display configuration helper function + * This helper function may be used by physical encoders to configure + * the split display related registers. + * @phys_enc: Pointer to physical encoder structure + * @interface: enum dpu_intf setting + */ +void dpu_encoder_helper_split_config( + struct dpu_encoder_phys *phys_enc, + enum dpu_intf interface) +{ + struct dpu_encoder_virt *dpu_enc; + struct split_pipe_cfg cfg = { 0 }; + struct dpu_hw_mdp *hw_mdptop; + struct msm_display_info *disp_info; + + if (!phys_enc->hw_mdptop || !phys_enc->parent) { + DPU_ERROR("invalid arg(s), encoder %d\n", phys_enc != NULL); + return; + } + + dpu_enc = to_dpu_encoder_virt(phys_enc->parent); + hw_mdptop = phys_enc->hw_mdptop; + disp_info = &dpu_enc->disp_info; + + if (disp_info->intf_type != INTF_DSI) + return; + + /** + * disable split modes since encoder will be operating in as the only + * encoder, either for the entire use case in the case of, for example, + * single DSI, or for this frame in the case of left/right only partial + * update. + */ + if (phys_enc->split_role == ENC_ROLE_SOLO) { + if (hw_mdptop->ops.setup_split_pipe) + hw_mdptop->ops.setup_split_pipe(hw_mdptop, &cfg); + return; + } + + cfg.en = true; + cfg.mode = phys_enc->intf_mode; + cfg.intf = interface; + + if (cfg.en && phys_enc->ops.needs_single_flush && + phys_enc->ops.needs_single_flush(phys_enc)) + cfg.split_flush_en = true; + + if (phys_enc->split_role == ENC_ROLE_MASTER) { + DPU_DEBUG_ENC(dpu_enc, "enable %d\n", cfg.en); + + if (hw_mdptop->ops.setup_split_pipe) + hw_mdptop->ops.setup_split_pipe(hw_mdptop, &cfg); + } +} + +/** + * dpu_encoder_use_dsc_merge - returns true if the encoder uses DSC merge topology. + * @drm_enc: Pointer to previously created drm encoder structure + */ +bool dpu_encoder_use_dsc_merge(struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); + int i, intf_count = 0, num_dsc = 0; + + for (i = 0; i < MAX_PHYS_ENCODERS_PER_VIRTUAL; i++) + if (dpu_enc->phys_encs[i]) + intf_count++; + + for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) + if (dpu_enc->hw_dsc[i]) + num_dsc++; + + return (num_dsc > 0) && (num_dsc > intf_count); +} + +/** + * dpu_encoder_get_dsc_config - get DSC config for the DPU encoder + * This helper function is used by physical encoder to get DSC config + * used for this encoder. + * @drm_enc: Pointer to encoder structure + */ +struct drm_dsc_config *dpu_encoder_get_dsc_config(struct drm_encoder *drm_enc) +{ + struct msm_drm_private *priv = drm_enc->dev->dev_private; + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); + int index = dpu_enc->disp_info.h_tile_instance[0]; + + if (dpu_enc->disp_info.intf_type == INTF_DSI) + return msm_dsi_get_dsc_config(priv->kms->dsi[index]); + + return NULL; +} + +void dpu_encoder_update_topology(struct drm_encoder *drm_enc, + struct msm_display_topology *topology, + struct drm_atomic_commit *state, + const struct drm_display_mode *adj_mode) +{ + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); + struct msm_drm_private *priv = dpu_enc->base.dev->dev_private; + struct msm_display_info *disp_info = &dpu_enc->disp_info; + struct dpu_kms *dpu_kms = to_dpu_kms(priv->kms); + struct drm_connector *connector; + struct drm_connector_state *conn_state; + struct drm_framebuffer *fb; + struct drm_dsc_config *dsc; + + int i; + + for (i = 0; i < MAX_PHYS_ENCODERS_PER_VIRTUAL; i++) + if (dpu_enc->phys_encs[i]) + topology->num_intf++; + + dsc = dpu_encoder_get_dsc_config(drm_enc); + + /* We only support 2 DSC mode (with 2 LM and 1 INTF) */ + if (dsc) { + /* + * Use 2 DSC encoders, 2 layer mixers and 1 or 2 interfaces + * when Display Stream Compression (DSC) is enabled, + * and when enough DSC blocks are available. + * This is power-optimal and can drive up to (including) 4k + * screens. + */ + WARN(topology->num_intf > 2, + "DSC topology cannot support more than 2 interfaces\n"); + if (topology->num_intf >= 2 || dpu_kms->catalog->dsc_count >= 2) + topology->num_dsc = 2; + else + topology->num_dsc = 1; + } + + connector = drm_atomic_get_new_connector_for_encoder(state, drm_enc); + if (!connector) + return; + conn_state = drm_atomic_get_new_connector_state(state, connector); + if (!conn_state) + return; + + /* + * Use CDM only for writeback or DP at the moment as other interfaces cannot handle it. + * If writeback itself cannot handle cdm for some reason it will fail in its atomic_check() + * earlier. + */ + if (disp_info->intf_type == INTF_WB && conn_state->writeback_job) { + fb = conn_state->writeback_job->fb; + + if (fb && MSM_FORMAT_IS_YUV(msm_framebuffer_format(fb))) + topology->num_cdm++; + } else if (disp_info->intf_type == INTF_DP) { + if (drm_mode_is_420_only(&connector->display_info, adj_mode)) + topology->num_cdm++; + } +} + +bool dpu_encoder_needs_modeset(struct drm_encoder *drm_enc, struct drm_atomic_commit *state) +{ + struct drm_connector *connector; + struct drm_connector_state *conn_state; + struct drm_framebuffer *fb; + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); + + if (!drm_enc || !state) + return false; + + connector = drm_atomic_get_new_connector_for_encoder(state, drm_enc); + if (!connector) + return false; + + conn_state = drm_atomic_get_new_connector_state(state, connector); + if (!conn_state) + return false; + + /** + * These checks are duplicated from dpu_encoder_update_topology() since + * CRTC and encoder don't hold topology information + */ + if (dpu_enc->disp_info.intf_type == INTF_WB && conn_state->writeback_job) { + fb = conn_state->writeback_job->fb; + if (fb && MSM_FORMAT_IS_YUV(msm_framebuffer_format(fb))) { + if (!dpu_enc->cur_master->hw_cdm) + return true; + } else { + if (dpu_enc->cur_master->hw_cdm) + return true; + } + } + + return false; +} + +static void _dpu_encoder_update_vsync_source(struct dpu_encoder_virt *dpu_enc, + struct msm_display_info *disp_info) +{ + struct dpu_vsync_source_cfg vsync_cfg = { 0 }; + struct msm_drm_private *priv; + struct dpu_kms *dpu_kms; + struct dpu_hw_mdp *hw_mdptop; + struct drm_encoder *drm_enc; + struct dpu_encoder_phys *phys_enc; + int i; + + if (!dpu_enc || !disp_info) { + DPU_ERROR("invalid param dpu_enc:%d or disp_info:%d\n", + dpu_enc != NULL, disp_info != NULL); + return; + } else if (dpu_enc->num_phys_encs > ARRAY_SIZE(dpu_enc->hw_pp)) { + DPU_ERROR("invalid num phys enc %d/%d\n", + dpu_enc->num_phys_encs, + (int) ARRAY_SIZE(dpu_enc->hw_pp)); + return; + } + + drm_enc = &dpu_enc->base; + /* this pointers are checked in virt_enable_helper */ + priv = drm_enc->dev->dev_private; + + dpu_kms = to_dpu_kms(priv->kms); + hw_mdptop = dpu_kms->hw_mdp; + if (!hw_mdptop) { + DPU_ERROR("invalid mdptop\n"); + return; + } + + vsync_cfg.vsync_source = disp_info->vsync_source; + vsync_cfg.frame_rate = drm_mode_vrefresh(&dpu_enc->base.crtc->state->adjusted_mode); + + if (hw_mdptop->ops.setup_vsync_source) { + for (i = 0; i < dpu_enc->num_phys_encs; i++) + vsync_cfg.ppnumber[i] = dpu_enc->hw_pp[i]->idx; + + vsync_cfg.pp_count = dpu_enc->num_phys_encs; + + hw_mdptop->ops.setup_vsync_source(hw_mdptop, &vsync_cfg); + } + + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + phys_enc = dpu_enc->phys_encs[i]; + + if (phys_enc->has_intf_te && phys_enc->hw_intf->ops.vsync_sel) + phys_enc->hw_intf->ops.vsync_sel(phys_enc->hw_intf, + &vsync_cfg); + } +} + +static void _dpu_encoder_irq_enable(struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc; + int i; + + if (!drm_enc) { + DPU_ERROR("invalid encoder\n"); + return; + } + + dpu_enc = to_dpu_encoder_virt(drm_enc); + + DPU_DEBUG_ENC(dpu_enc, "\n"); + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + + phys->ops.irq_enable(phys); + } +} + +static void _dpu_encoder_irq_disable(struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc; + int i; + + if (!drm_enc) { + DPU_ERROR("invalid encoder\n"); + return; + } + + dpu_enc = to_dpu_encoder_virt(drm_enc); + + DPU_DEBUG_ENC(dpu_enc, "\n"); + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + + phys->ops.irq_disable(phys); + } +} + +static void _dpu_encoder_resource_enable(struct drm_encoder *drm_enc) +{ + struct msm_drm_private *priv; + struct dpu_kms *dpu_kms; + struct dpu_encoder_virt *dpu_enc; + + dpu_enc = to_dpu_encoder_virt(drm_enc); + priv = drm_enc->dev->dev_private; + dpu_kms = to_dpu_kms(priv->kms); + + trace_dpu_enc_rc_enable(DRMID(drm_enc)); + + if (!dpu_enc->cur_master) { + DPU_ERROR("encoder master not set\n"); + return; + } + + /* enable DPU core clks */ + pm_runtime_get_sync(&dpu_kms->pdev->dev); + + /* enable all the irq */ + _dpu_encoder_irq_enable(drm_enc); +} + +static void _dpu_encoder_resource_disable(struct drm_encoder *drm_enc) +{ + struct msm_drm_private *priv; + struct dpu_kms *dpu_kms; + struct dpu_encoder_virt *dpu_enc; + + dpu_enc = to_dpu_encoder_virt(drm_enc); + priv = drm_enc->dev->dev_private; + dpu_kms = to_dpu_kms(priv->kms); + + trace_dpu_enc_rc_disable(DRMID(drm_enc)); + + if (!dpu_enc->cur_master) { + DPU_ERROR("encoder master not set\n"); + return; + } + + /* disable all the irq */ + _dpu_encoder_irq_disable(drm_enc); + + /* disable DPU core clks */ + pm_runtime_put_sync(&dpu_kms->pdev->dev); +} + +static int dpu_encoder_resource_control(struct drm_encoder *drm_enc, + u32 sw_event) +{ + struct dpu_encoder_virt *dpu_enc; + struct msm_drm_private *priv; + bool is_vid_mode = false; + + if (!drm_enc || !drm_enc->dev || !drm_enc->crtc) { + DPU_ERROR("invalid parameters\n"); + return -EINVAL; + } + dpu_enc = to_dpu_encoder_virt(drm_enc); + priv = drm_enc->dev->dev_private; + is_vid_mode = !dpu_enc->disp_info.is_cmd_mode; + + /* + * when idle_pc is not supported, process only KICKOFF, STOP and MODESET + * events and return early for other events (ie wb display). + */ + if (!dpu_enc->idle_pc_supported && + (sw_event != DPU_ENC_RC_EVENT_KICKOFF && + sw_event != DPU_ENC_RC_EVENT_STOP && + sw_event != DPU_ENC_RC_EVENT_PRE_STOP)) + return 0; + + trace_dpu_enc_rc(DRMID(drm_enc), sw_event, dpu_enc->idle_pc_supported, + dpu_enc->rc_state, "begin"); + + switch (sw_event) { + case DPU_ENC_RC_EVENT_KICKOFF: + /* cancel delayed off work, if any */ + if (cancel_delayed_work_sync(&dpu_enc->delayed_off_work)) + DPU_DEBUG_ENC(dpu_enc, "sw_event:%d, work cancelled\n", + sw_event); + + mutex_lock(&dpu_enc->rc_lock); + + /* return if the resource control is already in ON state */ + if (dpu_enc->rc_state == DPU_ENC_RC_STATE_ON) { + DRM_DEBUG_ATOMIC("id;%u, sw_event:%d, rc in ON state\n", + DRMID(drm_enc), sw_event); + mutex_unlock(&dpu_enc->rc_lock); + return 0; + } else if (dpu_enc->rc_state != DPU_ENC_RC_STATE_OFF && + dpu_enc->rc_state != DPU_ENC_RC_STATE_IDLE) { + DRM_DEBUG_ATOMIC("id;%u, sw_event:%d, rc in state %d\n", + DRMID(drm_enc), sw_event, + dpu_enc->rc_state); + mutex_unlock(&dpu_enc->rc_lock); + return -EINVAL; + } + + if (is_vid_mode && dpu_enc->rc_state == DPU_ENC_RC_STATE_IDLE) + _dpu_encoder_irq_enable(drm_enc); + else + _dpu_encoder_resource_enable(drm_enc); + + dpu_enc->rc_state = DPU_ENC_RC_STATE_ON; + + trace_dpu_enc_rc(DRMID(drm_enc), sw_event, + dpu_enc->idle_pc_supported, dpu_enc->rc_state, + "kickoff"); + + mutex_unlock(&dpu_enc->rc_lock); + break; + + case DPU_ENC_RC_EVENT_FRAME_DONE: + /* + * mutex lock is not used as this event happens at interrupt + * context. And locking is not required as, the other events + * like KICKOFF and STOP does a wait-for-idle before executing + * the resource_control + */ + if (dpu_enc->rc_state != DPU_ENC_RC_STATE_ON) { + DRM_DEBUG_KMS("id:%d, sw_event:%d,rc:%d-unexpected\n", + DRMID(drm_enc), sw_event, + dpu_enc->rc_state); + return -EINVAL; + } + + /* + * schedule off work item only when there are no + * frames pending + */ + if (dpu_crtc_frame_pending(drm_enc->crtc) > 1) { + DRM_DEBUG_KMS("id:%d skip schedule work\n", + DRMID(drm_enc)); + return 0; + } + + queue_delayed_work(priv->kms->wq, &dpu_enc->delayed_off_work, + msecs_to_jiffies(dpu_enc->idle_timeout)); + + trace_dpu_enc_rc(DRMID(drm_enc), sw_event, + dpu_enc->idle_pc_supported, dpu_enc->rc_state, + "frame done"); + break; + + case DPU_ENC_RC_EVENT_PRE_STOP: + /* cancel delayed off work, if any */ + if (cancel_delayed_work_sync(&dpu_enc->delayed_off_work)) + DPU_DEBUG_ENC(dpu_enc, "sw_event:%d, work cancelled\n", + sw_event); + + mutex_lock(&dpu_enc->rc_lock); + + if (is_vid_mode && + dpu_enc->rc_state == DPU_ENC_RC_STATE_IDLE) { + _dpu_encoder_irq_enable(drm_enc); + } + /* skip if is already OFF or IDLE, resources are off already */ + else if (dpu_enc->rc_state == DPU_ENC_RC_STATE_OFF || + dpu_enc->rc_state == DPU_ENC_RC_STATE_IDLE) { + DRM_DEBUG_KMS("id:%u, sw_event:%d, rc in %d state\n", + DRMID(drm_enc), sw_event, + dpu_enc->rc_state); + mutex_unlock(&dpu_enc->rc_lock); + return 0; + } + + dpu_enc->rc_state = DPU_ENC_RC_STATE_PRE_OFF; + + trace_dpu_enc_rc(DRMID(drm_enc), sw_event, + dpu_enc->idle_pc_supported, dpu_enc->rc_state, + "pre stop"); + + mutex_unlock(&dpu_enc->rc_lock); + break; + + case DPU_ENC_RC_EVENT_STOP: + mutex_lock(&dpu_enc->rc_lock); + + /* return if the resource control is already in OFF state */ + if (dpu_enc->rc_state == DPU_ENC_RC_STATE_OFF) { + DRM_DEBUG_KMS("id: %u, sw_event:%d, rc in OFF state\n", + DRMID(drm_enc), sw_event); + mutex_unlock(&dpu_enc->rc_lock); + return 0; + } else if (dpu_enc->rc_state == DPU_ENC_RC_STATE_ON) { + DRM_ERROR("id: %u, sw_event:%d, rc in state %d\n", + DRMID(drm_enc), sw_event, dpu_enc->rc_state); + mutex_unlock(&dpu_enc->rc_lock); + return -EINVAL; + } + + /** + * expect to arrive here only if in either idle state or pre-off + * and in IDLE state the resources are already disabled + */ + if (dpu_enc->rc_state == DPU_ENC_RC_STATE_PRE_OFF) + _dpu_encoder_resource_disable(drm_enc); + + dpu_enc->rc_state = DPU_ENC_RC_STATE_OFF; + + trace_dpu_enc_rc(DRMID(drm_enc), sw_event, + dpu_enc->idle_pc_supported, dpu_enc->rc_state, + "stop"); + + mutex_unlock(&dpu_enc->rc_lock); + break; + + case DPU_ENC_RC_EVENT_ENTER_IDLE: + mutex_lock(&dpu_enc->rc_lock); + + if (dpu_enc->rc_state != DPU_ENC_RC_STATE_ON) { + DRM_ERROR("id: %u, sw_event:%d, rc:%d !ON state\n", + DRMID(drm_enc), sw_event, dpu_enc->rc_state); + mutex_unlock(&dpu_enc->rc_lock); + return 0; + } + + /* + * if we are in ON but a frame was just kicked off, + * ignore the IDLE event, it's probably a stale timer event + */ + if (dpu_enc->frame_busy_mask[0]) { + DRM_ERROR("id:%u, sw_event:%d, rc:%d frame pending\n", + DRMID(drm_enc), sw_event, dpu_enc->rc_state); + mutex_unlock(&dpu_enc->rc_lock); + return 0; + } + + if (is_vid_mode) + _dpu_encoder_irq_disable(drm_enc); + else + _dpu_encoder_resource_disable(drm_enc); + + dpu_enc->rc_state = DPU_ENC_RC_STATE_IDLE; + + trace_dpu_enc_rc(DRMID(drm_enc), sw_event, + dpu_enc->idle_pc_supported, dpu_enc->rc_state, + "idle"); + + mutex_unlock(&dpu_enc->rc_lock); + break; + + default: + DRM_ERROR("id:%u, unexpected sw_event: %d\n", DRMID(drm_enc), + sw_event); + trace_dpu_enc_rc(DRMID(drm_enc), sw_event, + dpu_enc->idle_pc_supported, dpu_enc->rc_state, + "error"); + break; + } + + trace_dpu_enc_rc(DRMID(drm_enc), sw_event, + dpu_enc->idle_pc_supported, dpu_enc->rc_state, + "end"); + return 0; +} + +/** + * dpu_encoder_prepare_wb_job - prepare writeback job for the encoder. + * @drm_enc: Pointer to previously created drm encoder structure + * @job: Pointer to the current drm writeback job + */ +void dpu_encoder_prepare_wb_job(struct drm_encoder *drm_enc, + struct drm_writeback_job *job) +{ + struct dpu_encoder_virt *dpu_enc; + int i; + + dpu_enc = to_dpu_encoder_virt(drm_enc); + + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + + if (phys->ops.prepare_wb_job) + phys->ops.prepare_wb_job(phys, job); + + } +} + +/** + * dpu_encoder_cleanup_wb_job - cleanup writeback job for the encoder. + * @drm_enc: Pointer to previously created drm encoder structure + * @job: Pointer to the current drm writeback job + */ +void dpu_encoder_cleanup_wb_job(struct drm_encoder *drm_enc, + struct drm_writeback_job *job) +{ + struct dpu_encoder_virt *dpu_enc; + int i; + + dpu_enc = to_dpu_encoder_virt(drm_enc); + + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + + if (phys->ops.cleanup_wb_job) + phys->ops.cleanup_wb_job(phys, job); + + } +} + +static void dpu_encoder_virt_atomic_mode_set(struct drm_encoder *drm_enc, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state) +{ + struct dpu_encoder_virt *dpu_enc; + struct msm_drm_private *priv; + struct dpu_kms *dpu_kms; + struct dpu_global_state *global_state; + struct dpu_hw_blk *hw_pp[MAX_CHANNELS_PER_ENC]; + struct dpu_hw_blk *hw_ctl[MAX_CHANNELS_PER_ENC]; + struct dpu_hw_blk *hw_dsc[MAX_CHANNELS_PER_ENC]; + struct dpu_hw_blk *hw_cwb[MAX_CHANNELS_PER_ENC]; + int num_ctl, num_pp, num_dsc, num_pp_per_intf; + int num_cwb = 0; + bool is_cwb_encoder; + unsigned int dsc_mask = 0; + unsigned int cwb_mask = 0; + int i; + + if (!drm_enc) { + DPU_ERROR("invalid encoder\n"); + return; + } + + dpu_enc = to_dpu_encoder_virt(drm_enc); + DPU_DEBUG_ENC(dpu_enc, "\n"); + + priv = drm_enc->dev->dev_private; + dpu_kms = to_dpu_kms(priv->kms); + is_cwb_encoder = drm_crtc_in_clone_mode(crtc_state) && + dpu_enc->disp_info.intf_type == INTF_WB; + + global_state = dpu_kms_get_existing_global_state(dpu_kms); + if (IS_ERR_OR_NULL(global_state)) { + DPU_ERROR("Failed to get global state"); + return; + } + + trace_dpu_enc_mode_set(DRMID(drm_enc)); + + /* Query resource that have been reserved in atomic check step. */ + if (is_cwb_encoder) { + num_pp = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + drm_enc->crtc, + DPU_HW_BLK_DCWB_PINGPONG, + hw_pp, ARRAY_SIZE(hw_pp)); + num_cwb = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + drm_enc->crtc, + DPU_HW_BLK_CWB, + hw_cwb, ARRAY_SIZE(hw_cwb)); + } else { + num_pp = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + drm_enc->crtc, + DPU_HW_BLK_PINGPONG, hw_pp, + ARRAY_SIZE(hw_pp)); + } + + for (i = 0; i < num_cwb; i++) { + dpu_enc->hw_cwb[i] = to_dpu_hw_cwb(hw_cwb[i]); + cwb_mask |= BIT(dpu_enc->hw_cwb[i]->idx - CWB_0); + } + + dpu_enc->cwb_mask = cwb_mask; + + num_ctl = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + drm_enc->crtc, DPU_HW_BLK_CTL, hw_ctl, ARRAY_SIZE(hw_ctl)); + + for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) + dpu_enc->hw_pp[i] = i < num_pp ? to_dpu_hw_pingpong(hw_pp[i]) + : NULL; + + num_dsc = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + drm_enc->crtc, DPU_HW_BLK_DSC, + hw_dsc, ARRAY_SIZE(hw_dsc)); + for (i = 0; i < num_dsc; i++) { + dpu_enc->hw_dsc[i] = to_dpu_hw_dsc(hw_dsc[i]); + dsc_mask |= BIT(dpu_enc->hw_dsc[i]->idx - DSC_0); + } + + dpu_enc->dsc_mask = dsc_mask; + + if ((dpu_enc->disp_info.intf_type == INTF_WB && conn_state->writeback_job) || + dpu_enc->disp_info.intf_type == INTF_DP) { + struct dpu_hw_blk *hw_cdm = NULL; + + dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + drm_enc->crtc, DPU_HW_BLK_CDM, + &hw_cdm, 1); + dpu_enc->cur_master->hw_cdm = hw_cdm ? to_dpu_hw_cdm(hw_cdm) : NULL; + } + + /* + * There may be 4 PP and 2 INTF for quad pipe case, so INTF is not + * mapped to PP 1:1. Let's calculate the stride with pipe/INTF + */ + num_pp_per_intf = num_pp / dpu_enc->num_phys_encs; + + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + + phys->hw_pp = dpu_enc->hw_pp[num_pp_per_intf * i]; + if (!phys->hw_pp) { + DPU_ERROR_ENC(dpu_enc, + "no pp block assigned at idx: %d\n", i); + return; + } + + /* Use first (and only) CTL if active CTLs are supported */ + if (num_ctl == 1) + phys->hw_ctl = to_dpu_hw_ctl(hw_ctl[0]); + else + phys->hw_ctl = i < num_ctl ? to_dpu_hw_ctl(hw_ctl[i]) : NULL; + if (!phys->hw_ctl) { + DPU_ERROR_ENC(dpu_enc, + "no ctl block assigned at idx: %d\n", i); + return; + } + + phys->cached_mode = crtc_state->adjusted_mode; + if (phys->ops.atomic_mode_set) + phys->ops.atomic_mode_set(phys, crtc_state, conn_state); + } +} + +static void _dpu_encoder_virt_enable_helper(struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc = NULL; + int i; + + if (!drm_enc || !drm_enc->dev) { + DPU_ERROR("invalid parameters\n"); + return; + } + + dpu_enc = to_dpu_encoder_virt(drm_enc); + if (!dpu_enc || !dpu_enc->cur_master) { + DPU_ERROR("invalid dpu encoder/master\n"); + return; + } + + + if (dpu_enc->disp_info.intf_type == INTF_DP && + dpu_enc->cur_master->hw_mdptop && + dpu_enc->cur_master->hw_mdptop->ops.intf_audio_select) + dpu_enc->cur_master->hw_mdptop->ops.intf_audio_select( + dpu_enc->cur_master->hw_mdptop); + + if (dpu_enc->disp_info.is_cmd_mode) + _dpu_encoder_update_vsync_source(dpu_enc, &dpu_enc->disp_info); + + if (dpu_enc->disp_info.intf_type == INTF_DSI && + !WARN_ON(dpu_enc->num_phys_encs == 0)) { + unsigned bpc = dpu_enc->connector->display_info.bpc; + for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) { + if (!dpu_enc->hw_pp[i]) + continue; + _dpu_encoder_setup_dither(dpu_enc->hw_pp[i], bpc); + } + } +} + +/** + * dpu_encoder_virt_runtime_resume - pm runtime resume the encoder configs + * @drm_enc: encoder pointer + */ +void dpu_encoder_virt_runtime_resume(struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); + + mutex_lock(&dpu_enc->enc_lock); + + if (!dpu_enc->enabled) + goto out; + + if (dpu_enc->cur_slave && dpu_enc->cur_slave->ops.restore) + dpu_enc->cur_slave->ops.restore(dpu_enc->cur_slave); + if (dpu_enc->cur_master && dpu_enc->cur_master->ops.restore) + dpu_enc->cur_master->ops.restore(dpu_enc->cur_master); + + _dpu_encoder_virt_enable_helper(drm_enc); + +out: + mutex_unlock(&dpu_enc->enc_lock); +} + +static void dpu_encoder_virt_atomic_enable(struct drm_encoder *drm_enc, + struct drm_atomic_commit *state) +{ + struct dpu_encoder_virt *dpu_enc = NULL; + int ret = 0; + struct drm_display_mode *cur_mode = NULL; + + dpu_enc = to_dpu_encoder_virt(drm_enc); + dpu_enc->dsc = dpu_encoder_get_dsc_config(drm_enc); + + atomic_set(&dpu_enc->frame_done_timeout_cnt, 0); + + mutex_lock(&dpu_enc->enc_lock); + + dpu_enc->commit_done_timedout = false; + + dpu_enc->connector = drm_atomic_get_new_connector_for_encoder(state, drm_enc); + + cur_mode = &dpu_enc->base.crtc->state->adjusted_mode; + + dpu_enc->wide_bus_en = dpu_encoder_is_widebus_enabled(drm_enc); + + trace_dpu_enc_enable(DRMID(drm_enc), cur_mode->hdisplay, + cur_mode->vdisplay); + + /* always enable slave encoder before master */ + if (dpu_enc->cur_slave && dpu_enc->cur_slave->ops.enable) + dpu_enc->cur_slave->ops.enable(dpu_enc->cur_slave); + + if (dpu_enc->cur_master && dpu_enc->cur_master->ops.enable) + dpu_enc->cur_master->ops.enable(dpu_enc->cur_master); + + ret = dpu_encoder_resource_control(drm_enc, DPU_ENC_RC_EVENT_KICKOFF); + if (ret) { + DPU_ERROR_ENC(dpu_enc, "dpu resource control failed: %d\n", + ret); + goto out; + } + + _dpu_encoder_virt_enable_helper(drm_enc); + + dpu_enc->enabled = true; + +out: + mutex_unlock(&dpu_enc->enc_lock); +} + +static void dpu_encoder_virt_atomic_disable(struct drm_encoder *drm_enc, + struct drm_atomic_commit *state) +{ + struct dpu_encoder_virt *dpu_enc = NULL; + struct drm_crtc *crtc; + struct drm_crtc_state *old_state = NULL; + int i = 0; + + dpu_enc = to_dpu_encoder_virt(drm_enc); + DPU_DEBUG_ENC(dpu_enc, "\n"); + + crtc = drm_atomic_get_old_crtc_for_encoder(state, drm_enc); + if (crtc) + old_state = drm_atomic_get_old_crtc_state(state, crtc); + + /* + * The encoder is already disabled if self refresh mode was set earlier, + * in the old_state for the corresponding crtc. + */ + if (old_state && old_state->self_refresh_active) + return; + + mutex_lock(&dpu_enc->enc_lock); + dpu_enc->enabled = false; + + trace_dpu_enc_disable(DRMID(drm_enc)); + + /* wait for idle */ + dpu_encoder_wait_for_tx_complete(drm_enc); + + dpu_encoder_resource_control(drm_enc, DPU_ENC_RC_EVENT_PRE_STOP); + + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + + if (phys->ops.disable) + phys->ops.disable(phys); + } + + + /* after phys waits for frame-done, should be no more frames pending */ + if (atomic_xchg(&dpu_enc->frame_done_timeout_ms, 0)) { + DPU_ERROR("enc%d timeout pending\n", drm_enc->base.id); + timer_delete_sync(&dpu_enc->frame_done_timer); + } + + dpu_encoder_resource_control(drm_enc, DPU_ENC_RC_EVENT_STOP); + + dpu_enc->connector = NULL; + + DPU_DEBUG_ENC(dpu_enc, "encoder disabled\n"); + + mutex_unlock(&dpu_enc->enc_lock); +} + +static struct dpu_hw_intf *dpu_encoder_get_intf(const struct dpu_mdss_cfg *catalog, + struct dpu_rm *dpu_rm, + enum dpu_intf_type type, u32 controller_id) +{ + int i = 0; + + if (type == INTF_WB) + return NULL; + + for (i = 0; i < catalog->intf_count; i++) { + if (catalog->intf[i].type == type + && catalog->intf[i].controller_id == controller_id) { + return dpu_rm_get_intf(dpu_rm, catalog->intf[i].id); + } + } + + return NULL; +} + +/** + * dpu_encoder_vblank_callback - Notify virtual encoder of vblank IRQ reception + * @drm_enc: Pointer to drm encoder structure + * @phy_enc: Pointer to physical encoder + * Note: This is called from IRQ handler context. + */ +void dpu_encoder_vblank_callback(struct drm_encoder *drm_enc, + struct dpu_encoder_phys *phy_enc) +{ + struct dpu_encoder_virt *dpu_enc = NULL; + unsigned long lock_flags; + + if (!drm_enc || !phy_enc) + return; + + DPU_ATRACE_BEGIN("encoder_vblank_callback"); + dpu_enc = to_dpu_encoder_virt(drm_enc); + + atomic_inc(&phy_enc->vsync_cnt); + + spin_lock_irqsave(&dpu_enc->enc_spinlock, lock_flags); + if (dpu_enc->crtc) + dpu_crtc_vblank_callback(dpu_enc->crtc); + spin_unlock_irqrestore(&dpu_enc->enc_spinlock, lock_flags); + + DPU_ATRACE_END("encoder_vblank_callback"); +} + +/** + * dpu_encoder_underrun_callback - Notify virtual encoder of underrun IRQ reception + * @drm_enc: Pointer to drm encoder structure + * @phy_enc: Pointer to physical encoder + * Note: This is called from IRQ handler context. + */ +void dpu_encoder_underrun_callback(struct drm_encoder *drm_enc, + struct dpu_encoder_phys *phy_enc) +{ + if (!phy_enc) + return; + + DPU_ATRACE_BEGIN("encoder_underrun_callback"); + atomic_inc(&phy_enc->underrun_cnt); + + /* trigger dump only on the first underrun */ + if (atomic_read(&phy_enc->underrun_cnt) == 1) + msm_disp_snapshot_state(drm_enc->dev); + + trace_dpu_enc_underrun_cb(DRMID(drm_enc), + atomic_read(&phy_enc->underrun_cnt)); + DPU_ATRACE_END("encoder_underrun_callback"); +} + +/** + * dpu_encoder_assign_crtc - Link the encoder to the crtc it's assigned to + * @drm_enc: encoder pointer + * @crtc: crtc pointer + */ +void dpu_encoder_assign_crtc(struct drm_encoder *drm_enc, struct drm_crtc *crtc) +{ + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); + unsigned long lock_flags; + + spin_lock_irqsave(&dpu_enc->enc_spinlock, lock_flags); + /* crtc should always be cleared before re-assigning */ + WARN_ON(crtc && dpu_enc->crtc); + dpu_enc->crtc = crtc; + spin_unlock_irqrestore(&dpu_enc->enc_spinlock, lock_flags); +} + +/** + * dpu_encoder_toggle_vblank_for_crtc - Toggles vblank interrupts on or off if + * the encoder is assigned to the given crtc + * @drm_enc: encoder pointer + * @crtc: crtc pointer + * @enable: true if vblank should be enabled + */ +void dpu_encoder_toggle_vblank_for_crtc(struct drm_encoder *drm_enc, + struct drm_crtc *crtc, bool enable) +{ + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); + unsigned long lock_flags; + int i; + + trace_dpu_enc_vblank_cb(DRMID(drm_enc), enable); + + spin_lock_irqsave(&dpu_enc->enc_spinlock, lock_flags); + if (dpu_enc->crtc != crtc) { + spin_unlock_irqrestore(&dpu_enc->enc_spinlock, lock_flags); + return; + } + spin_unlock_irqrestore(&dpu_enc->enc_spinlock, lock_flags); + + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + + if (phys->ops.control_vblank_irq) + phys->ops.control_vblank_irq(phys, enable); + } +} + +/** + * dpu_encoder_frame_done_callback - Notify virtual encoder that this phys + * encoder completes last request frame + * @drm_enc: Pointer to drm encoder structure + * @ready_phys: Pointer to physical encoder + * @event: Event to process + */ +void dpu_encoder_frame_done_callback( + struct drm_encoder *drm_enc, + struct dpu_encoder_phys *ready_phys, u32 event) +{ + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); + unsigned int i; + + if (event & (DPU_ENCODER_FRAME_EVENT_DONE + | DPU_ENCODER_FRAME_EVENT_ERROR + | DPU_ENCODER_FRAME_EVENT_PANEL_DEAD)) { + + if (!dpu_enc->frame_busy_mask[0]) { + /** + * suppress frame_done without waiter, + * likely autorefresh + */ + trace_dpu_enc_frame_done_cb_not_busy(DRMID(drm_enc), event, + dpu_encoder_helper_get_intf_type(ready_phys->intf_mode), + ready_phys->hw_intf ? ready_phys->hw_intf->idx : -1, + ready_phys->hw_wb ? ready_phys->hw_wb->idx : -1); + return; + } + + /* One of the physical encoders has become idle */ + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + if (dpu_enc->phys_encs[i] == ready_phys) { + trace_dpu_enc_frame_done_cb(DRMID(drm_enc), i, + dpu_enc->frame_busy_mask[0]); + clear_bit(i, dpu_enc->frame_busy_mask); + } + } + + if (!dpu_enc->frame_busy_mask[0]) { + atomic_set(&dpu_enc->frame_done_timeout_ms, 0); + timer_delete(&dpu_enc->frame_done_timer); + + dpu_encoder_resource_control(drm_enc, + DPU_ENC_RC_EVENT_FRAME_DONE); + + if (dpu_enc->crtc) + dpu_crtc_frame_event_cb(dpu_enc->crtc, event); + } + } else { + if (dpu_enc->crtc) + dpu_crtc_frame_event_cb(dpu_enc->crtc, event); + } +} + +static void dpu_encoder_off_work(struct work_struct *work) +{ + struct dpu_encoder_virt *dpu_enc = container_of(work, + struct dpu_encoder_virt, delayed_off_work.work); + + dpu_encoder_resource_control(&dpu_enc->base, + DPU_ENC_RC_EVENT_ENTER_IDLE); + + dpu_encoder_frame_done_callback(&dpu_enc->base, NULL, + DPU_ENCODER_FRAME_EVENT_IDLE); +} + +/** + * _dpu_encoder_trigger_flush - trigger flush for a physical encoder + * @drm_enc: Pointer to drm encoder structure + * @phys: Pointer to physical encoder structure + * @extra_flush_bits: Additional bit mask to include in flush trigger + */ +static void _dpu_encoder_trigger_flush(struct drm_encoder *drm_enc, + struct dpu_encoder_phys *phys, uint32_t extra_flush_bits) +{ + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); + struct dpu_hw_ctl *ctl; + int pending_kickoff_cnt; + u32 ret = UINT_MAX; + + if (!phys->hw_pp) { + DPU_ERROR("invalid pingpong hw\n"); + return; + } + + ctl = phys->hw_ctl; + if (!ctl->ops.trigger_flush) { + DPU_ERROR("missing trigger cb\n"); + return; + } + + pending_kickoff_cnt = dpu_encoder_phys_inc_pending(phys); + + /* Return early if encoder is writeback and in clone mode */ + if (drm_enc->encoder_type == DRM_MODE_ENCODER_VIRTUAL && + dpu_enc->cwb_mask) { + DPU_DEBUG("encoder %d skip flush for concurrent writeback encoder\n", + DRMID(drm_enc)); + return; + } + + + if (extra_flush_bits && ctl->ops.update_pending_flush) + ctl->ops.update_pending_flush(ctl, extra_flush_bits); + + ctl->ops.trigger_flush(ctl); + + if (ctl->ops.get_pending_flush) + ret = ctl->ops.get_pending_flush(ctl); + + trace_dpu_enc_trigger_flush(DRMID(drm_enc), + dpu_encoder_helper_get_intf_type(phys->intf_mode), + phys->hw_intf ? phys->hw_intf->idx : -1, + phys->hw_wb ? phys->hw_wb->idx : -1, + pending_kickoff_cnt, ctl->idx, + extra_flush_bits, ret); +} + +/** + * _dpu_encoder_trigger_start - trigger start for a physical encoder + * @phys: Pointer to physical encoder structure + */ +static void _dpu_encoder_trigger_start(struct dpu_encoder_phys *phys) +{ + struct dpu_encoder_virt *dpu_enc; + + if (!phys) { + DPU_ERROR("invalid argument(s)\n"); + return; + } + + if (!phys->hw_pp) { + DPU_ERROR("invalid pingpong hw\n"); + return; + } + + dpu_enc = to_dpu_encoder_virt(phys->parent); + + if (phys->parent->encoder_type == DRM_MODE_ENCODER_VIRTUAL && + dpu_enc->cwb_mask) { + DPU_DEBUG("encoder %d CWB enabled, skipping\n", DRMID(phys->parent)); + return; + } + + if (phys->ops.trigger_start && phys->enable_state != DPU_ENC_DISABLED) + phys->ops.trigger_start(phys); +} + +/** + * dpu_encoder_helper_trigger_start - control start helper function + * This helper function may be optionally specified by physical + * encoders if they require ctl_start triggering. + * @phys_enc: Pointer to physical encoder structure + */ +void dpu_encoder_helper_trigger_start(struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_ctl *ctl; + + ctl = phys_enc->hw_ctl; + if (ctl->ops.trigger_start) { + ctl->ops.trigger_start(ctl); + trace_dpu_enc_trigger_start(DRMID(phys_enc->parent), ctl->idx); + } +} + +static int dpu_encoder_helper_wait_event_timeout( + int32_t drm_id, + unsigned int irq_idx, + struct dpu_encoder_wait_info *info) +{ + int rc = 0; + s64 expected_time = ktime_to_ms(ktime_get()) + info->timeout_ms; + s64 jiffies = msecs_to_jiffies(info->timeout_ms); + s64 time; + + do { + rc = wait_event_timeout(*(info->wq), + atomic_read(info->atomic_cnt) == 0, jiffies); + time = ktime_to_ms(ktime_get()); + + trace_dpu_enc_wait_event_timeout(drm_id, + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx), + rc, time, + expected_time, + atomic_read(info->atomic_cnt)); + /* If we timed out, counter is valid and time is less, wait again */ + } while (atomic_read(info->atomic_cnt) && (rc == 0) && + (time < expected_time)); + + return rc; +} + +static void dpu_encoder_helper_hw_reset(struct dpu_encoder_phys *phys_enc) +{ + struct dpu_encoder_virt *dpu_enc; + struct dpu_hw_ctl *ctl; + int rc; + struct drm_encoder *drm_enc; + + dpu_enc = to_dpu_encoder_virt(phys_enc->parent); + ctl = phys_enc->hw_ctl; + drm_enc = phys_enc->parent; + + if (!ctl->ops.reset) + return; + + DRM_DEBUG_KMS("id:%u ctl %d reset\n", DRMID(drm_enc), + ctl->idx); + + rc = ctl->ops.reset(ctl); + if (rc) { + DPU_ERROR_ENC(dpu_enc, "ctl %d reset failure\n", ctl->idx); + msm_disp_snapshot_state(drm_enc->dev); + } + + phys_enc->enable_state = DPU_ENC_ENABLED; +} + +/** + * _dpu_encoder_kickoff_phys - handle physical encoder kickoff + * Iterate through the physical encoders and perform consolidated flush + * and/or control start triggering as needed. This is done in the virtual + * encoder rather than the individual physical ones in order to handle + * use cases that require visibility into multiple physical encoders at + * a time. + * @dpu_enc: Pointer to virtual encoder structure + */ +static void _dpu_encoder_kickoff_phys(struct dpu_encoder_virt *dpu_enc) +{ + struct dpu_hw_ctl *ctl; + uint32_t i, pending_flush; + unsigned long lock_flags; + + pending_flush = 0x0; + + /* update pending counts and trigger kickoff ctl flush atomically */ + spin_lock_irqsave(&dpu_enc->enc_spinlock, lock_flags); + + /* don't perform flush/start operations for slave encoders */ + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + + if (phys->enable_state == DPU_ENC_DISABLED) + continue; + + ctl = phys->hw_ctl; + + /* + * This is cleared in frame_done worker, which isn't invoked + * for async commits. So don't set this for async, since it'll + * roll over to the next commit. + */ + if (phys->split_role != ENC_ROLE_SLAVE) + set_bit(i, dpu_enc->frame_busy_mask); + + if (!phys->ops.needs_single_flush || + !phys->ops.needs_single_flush(phys)) + _dpu_encoder_trigger_flush(&dpu_enc->base, phys, 0x0); + else if (ctl->ops.get_pending_flush) + pending_flush |= ctl->ops.get_pending_flush(ctl); + } + + /* for split flush, combine pending flush masks and send to master */ + if (pending_flush && dpu_enc->cur_master) { + _dpu_encoder_trigger_flush( + &dpu_enc->base, + dpu_enc->cur_master, + pending_flush); + } + + _dpu_encoder_trigger_start(dpu_enc->cur_master); + + spin_unlock_irqrestore(&dpu_enc->enc_spinlock, lock_flags); +} + +/** + * dpu_encoder_trigger_kickoff_pending - Clear the flush bits from previous + * kickoff and trigger the ctl prepare progress for command mode display. + * @drm_enc: encoder pointer + */ +void dpu_encoder_trigger_kickoff_pending(struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc; + struct dpu_encoder_phys *phys; + unsigned int i; + struct dpu_hw_ctl *ctl; + struct msm_display_info *disp_info; + + if (!drm_enc) { + DPU_ERROR("invalid encoder\n"); + return; + } + dpu_enc = to_dpu_encoder_virt(drm_enc); + disp_info = &dpu_enc->disp_info; + + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + phys = dpu_enc->phys_encs[i]; + + ctl = phys->hw_ctl; + ctl->ops.clear_pending_flush(ctl); + + /* update only for command mode primary ctl */ + if ((phys == dpu_enc->cur_master) && + disp_info->is_cmd_mode + && ctl->ops.trigger_pending) + ctl->ops.trigger_pending(ctl); + } +} + +static u32 _dpu_encoder_calculate_linetime(struct dpu_encoder_virt *dpu_enc, + struct drm_display_mode *mode) +{ + u64 pclk_rate; + u32 pclk_period; + u32 line_time; + + /* + * For linetime calculation, only operate on master encoder. + */ + if (!dpu_enc->cur_master) + return 0; + + if (!dpu_enc->cur_master->ops.get_line_count) { + DPU_ERROR("get_line_count function not defined\n"); + return 0; + } + + pclk_rate = mode->clock; /* pixel clock in kHz */ + if (pclk_rate == 0) { + DPU_ERROR("pclk is 0, cannot calculate line time\n"); + return 0; + } + + pclk_period = DIV_ROUND_UP_ULL(1000000000ull, pclk_rate); + if (pclk_period == 0) { + DPU_ERROR("pclk period is 0\n"); + return 0; + } + + /* + * Line time calculation based on Pixel clock and HTOTAL. + * Final unit is in ns. + */ + line_time = (pclk_period * mode->htotal) / 1000; + if (line_time == 0) { + DPU_ERROR("line time calculation is 0\n"); + return 0; + } + + DPU_DEBUG_ENC(dpu_enc, + "clk_rate=%lldkHz, clk_period=%d, linetime=%dns\n", + pclk_rate, pclk_period, line_time); + + return line_time; +} + +/** + * dpu_encoder_vsync_time - get the time of the next vsync + * @drm_enc: encoder pointer + * @wakeup_time: pointer to ktime_t to write the vsync time to + */ +int dpu_encoder_vsync_time(struct drm_encoder *drm_enc, ktime_t *wakeup_time) +{ + struct drm_display_mode *mode; + struct dpu_encoder_virt *dpu_enc; + u32 cur_line; + u32 line_time; + u32 vtotal, time_to_vsync; + ktime_t cur_time; + + dpu_enc = to_dpu_encoder_virt(drm_enc); + + if (!drm_enc->crtc || !drm_enc->crtc->state) { + DPU_ERROR("crtc/crtc state object is NULL\n"); + return -EINVAL; + } + mode = &drm_enc->crtc->state->adjusted_mode; + + line_time = _dpu_encoder_calculate_linetime(dpu_enc, mode); + if (!line_time) + return -EINVAL; + + cur_line = dpu_enc->cur_master->ops.get_line_count(dpu_enc->cur_master); + + vtotal = mode->vtotal; + if (cur_line >= vtotal) + time_to_vsync = line_time * vtotal; + else + time_to_vsync = line_time * (vtotal - cur_line); + + if (time_to_vsync == 0) { + DPU_ERROR("time to vsync should not be zero, vtotal=%d\n", + vtotal); + return -EINVAL; + } + + cur_time = ktime_get(); + *wakeup_time = ktime_add_ns(cur_time, time_to_vsync); + + DPU_DEBUG_ENC(dpu_enc, + "cur_line=%u vtotal=%u time_to_vsync=%u, cur_time=%lld, wakeup_time=%lld\n", + cur_line, vtotal, time_to_vsync, + ktime_to_ms(cur_time), + ktime_to_ms(*wakeup_time)); + return 0; +} + +static u32 +dpu_encoder_dsc_initial_line_calc(struct drm_dsc_config *dsc, + u32 enc_ip_width) +{ + int ssm_delay, total_pixels, soft_slice_per_enc; + + soft_slice_per_enc = enc_ip_width / dsc->slice_width; + + /* + * minimum number of initial line pixels is a sum of: + * 1. sub-stream multiplexer delay (83 groups for 8bpc, + * 91 for 10 bpc) * 3 + * 2. for two soft slice cases, add extra sub-stream multiplexer * 3 + * 3. the initial xmit delay + * 4. total pipeline delay through the "lock step" of encoder (47) + * 5. 6 additional pixels as the output of the rate buffer is + * 48 bits wide + */ + ssm_delay = ((dsc->bits_per_component < 10) ? 84 : 92); + total_pixels = ssm_delay * 3 + dsc->initial_xmit_delay + 47; + if (soft_slice_per_enc > 1) + total_pixels += (ssm_delay * 3); + return DIV_ROUND_UP(total_pixels, dsc->slice_width); +} + +static void dpu_encoder_dsc_pipe_cfg(struct dpu_hw_ctl *ctl, + struct dpu_hw_dsc *hw_dsc, + struct dpu_hw_pingpong *hw_pp, + struct drm_dsc_config *dsc, + u32 common_mode, + u32 initial_lines) +{ + if (hw_dsc->ops.dsc_config) + hw_dsc->ops.dsc_config(hw_dsc, dsc, common_mode, initial_lines); + + if (hw_dsc->ops.dsc_config_thresh) + hw_dsc->ops.dsc_config_thresh(hw_dsc, dsc); + + if (hw_pp->ops.setup_dsc) + hw_pp->ops.setup_dsc(hw_pp); + + if (hw_dsc->ops.dsc_bind_pingpong_blk) + hw_dsc->ops.dsc_bind_pingpong_blk(hw_dsc, hw_pp->idx); + + if (hw_pp->ops.enable_dsc) + hw_pp->ops.enable_dsc(hw_pp); + + if (ctl->ops.update_pending_flush_dsc) + ctl->ops.update_pending_flush_dsc(ctl, hw_dsc->idx); +} + +static void dpu_encoder_prep_dsc(struct dpu_encoder_virt *dpu_enc, + struct drm_dsc_config *dsc) +{ + struct dpu_encoder_phys *enc_master = dpu_enc->cur_master; + struct dpu_hw_ctl *ctl = enc_master->hw_ctl; + struct dpu_hw_dsc *hw_dsc[MAX_CHANNELS_PER_ENC]; + struct dpu_hw_pingpong *hw_pp[MAX_CHANNELS_PER_ENC]; + int this_frame_slices; + int intf_ip_w, enc_ip_w; + int dsc_common_mode; + int pic_width; + u32 initial_lines; + int num_dsc = 0; + int i; + + for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) { + hw_pp[i] = dpu_enc->hw_pp[i]; + hw_dsc[i] = dpu_enc->hw_dsc[i]; + + if (!hw_pp[i] || !hw_dsc[i]) + break; + + num_dsc++; + } + + pic_width = dsc->pic_width; + + dsc_common_mode = 0; + if (num_dsc > 1) + dsc_common_mode |= DSC_MODE_SPLIT_PANEL; + if (dpu_encoder_use_dsc_merge(enc_master->parent)) + dsc_common_mode |= DSC_MODE_MULTIPLEX; + if (enc_master->intf_mode == INTF_MODE_VIDEO) + dsc_common_mode |= DSC_MODE_VIDEO; + + this_frame_slices = pic_width / dsc->slice_width; + intf_ip_w = this_frame_slices * dsc->slice_width; + + enc_ip_w = intf_ip_w / num_dsc; + initial_lines = dpu_encoder_dsc_initial_line_calc(dsc, enc_ip_w); + + for (i = 0; i < num_dsc; i++) + dpu_encoder_dsc_pipe_cfg(ctl, hw_dsc[i], hw_pp[i], + dsc, dsc_common_mode, initial_lines); +} + +/** + * dpu_encoder_prepare_for_kickoff - schedule double buffer flip of the ctl + * path (i.e. ctl flush and start) at next appropriate time. + * Immediately: if no previous commit is outstanding. + * Delayed: Block until next trigger can be issued. + * @drm_enc: encoder pointer + */ +void dpu_encoder_prepare_for_kickoff(struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc; + struct dpu_encoder_phys *phys; + bool needs_hw_reset = false; + unsigned int i; + + dpu_enc = to_dpu_encoder_virt(drm_enc); + + trace_dpu_enc_prepare_kickoff(DRMID(drm_enc)); + + /* prepare for next kickoff, may include waiting on previous kickoff */ + DPU_ATRACE_BEGIN("enc_prepare_for_kickoff"); + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + phys = dpu_enc->phys_encs[i]; + if (phys->ops.prepare_for_kickoff) + phys->ops.prepare_for_kickoff(phys); + if (phys->enable_state == DPU_ENC_ERR_NEEDS_HW_RESET) + needs_hw_reset = true; + } + DPU_ATRACE_END("enc_prepare_for_kickoff"); + + dpu_encoder_resource_control(drm_enc, DPU_ENC_RC_EVENT_KICKOFF); + + /* if any phys needs reset, reset all phys, in-order */ + if (needs_hw_reset) { + trace_dpu_enc_prepare_kickoff_reset(DRMID(drm_enc)); + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + dpu_encoder_helper_hw_reset(dpu_enc->phys_encs[i]); + } + } + + if (dpu_enc->dsc) + dpu_encoder_prep_dsc(dpu_enc, dpu_enc->dsc); +} + +/** + * dpu_encoder_is_valid_for_commit - check if encode has valid parameters for commit. + * @drm_enc: Pointer to drm encoder structure + */ +bool dpu_encoder_is_valid_for_commit(struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc; + unsigned int i; + struct dpu_encoder_phys *phys; + + dpu_enc = to_dpu_encoder_virt(drm_enc); + + if (drm_enc->encoder_type == DRM_MODE_ENCODER_VIRTUAL) { + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + phys = dpu_enc->phys_encs[i]; + if (phys->ops.is_valid_for_commit && !phys->ops.is_valid_for_commit(phys)) { + DPU_DEBUG("invalid FB not kicking off\n"); + return false; + } + } + } + + return true; +} + +/** + * dpu_encoder_start_frame_done_timer - Start the encoder frame done timer + * @drm_enc: Pointer to drm encoder structure + */ +void dpu_encoder_start_frame_done_timer(struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc; + unsigned long timeout_ms; + + dpu_enc = to_dpu_encoder_virt(drm_enc); + timeout_ms = DPU_ENCODER_FRAME_DONE_TIMEOUT_FRAMES * 1000 / + drm_mode_vrefresh(&drm_enc->crtc->state->adjusted_mode); + + atomic_set(&dpu_enc->frame_done_timeout_ms, timeout_ms); + mod_timer(&dpu_enc->frame_done_timer, + jiffies + msecs_to_jiffies(timeout_ms)); + +} + +/** + * dpu_encoder_kickoff - trigger a double buffer flip of the ctl path + * (i.e. ctl flush and start) immediately. + * @drm_enc: encoder pointer + */ +void dpu_encoder_kickoff(struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc; + struct dpu_encoder_phys *phys; + unsigned int i; + + DPU_ATRACE_BEGIN("encoder_kickoff"); + dpu_enc = to_dpu_encoder_virt(drm_enc); + + trace_dpu_enc_kickoff(DRMID(drm_enc)); + + /* All phys encs are ready to go, trigger the kickoff */ + _dpu_encoder_kickoff_phys(dpu_enc); + + /* allow phys encs to handle any post-kickoff business */ + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + phys = dpu_enc->phys_encs[i]; + if (phys->ops.handle_post_kickoff) + phys->ops.handle_post_kickoff(phys); + } + + DPU_ATRACE_END("encoder_kickoff"); +} + +static void dpu_encoder_helper_reset_mixers(struct dpu_encoder_phys *phys_enc) +{ + int i, num_lm; + struct dpu_global_state *global_state; + struct dpu_hw_blk *hw_lm[2]; + struct dpu_hw_mixer *hw_mixer[2]; + struct dpu_hw_ctl *ctl = phys_enc->hw_ctl; + + /* reset all mixers for this encoder */ + if (ctl->ops.clear_all_blendstages) + ctl->ops.clear_all_blendstages(ctl); + + global_state = dpu_kms_get_existing_global_state(phys_enc->dpu_kms); + + num_lm = dpu_rm_get_assigned_resources(&phys_enc->dpu_kms->rm, global_state, + phys_enc->parent->crtc, DPU_HW_BLK_LM, hw_lm, ARRAY_SIZE(hw_lm)); + + for (i = 0; i < num_lm; i++) { + hw_mixer[i] = to_dpu_hw_mixer(hw_lm[i]); + if (ctl->ops.update_pending_flush_mixer) + ctl->ops.update_pending_flush_mixer(ctl, hw_mixer[i]->idx); + + /* clear all blendstages */ + if (ctl->ops.setup_blendstage) + ctl->ops.setup_blendstage(ctl, hw_mixer[i]->idx, NULL); + + if (hw_mixer[i]->ops.clear_all_blendstages) + hw_mixer[i]->ops.clear_all_blendstages(hw_mixer[i]); + + if (ctl->ops.set_active_lms) + ctl->ops.set_active_lms(ctl, NULL); + + if (ctl->ops.set_active_fetch_pipes) + ctl->ops.set_active_fetch_pipes(ctl, NULL); + + if (ctl->ops.set_active_pipes) + ctl->ops.set_active_pipes(ctl, NULL); + } +} + +static void dpu_encoder_dsc_pipe_clr(struct dpu_hw_ctl *ctl, + struct dpu_hw_dsc *hw_dsc, + struct dpu_hw_pingpong *hw_pp) +{ + if (hw_dsc->ops.dsc_disable) + hw_dsc->ops.dsc_disable(hw_dsc); + + if (hw_pp->ops.disable_dsc) + hw_pp->ops.disable_dsc(hw_pp); + + if (hw_dsc->ops.dsc_bind_pingpong_blk) + hw_dsc->ops.dsc_bind_pingpong_blk(hw_dsc, PINGPONG_NONE); + + if (ctl->ops.update_pending_flush_dsc) + ctl->ops.update_pending_flush_dsc(ctl, hw_dsc->idx); +} + +static void dpu_encoder_unprep_dsc(struct dpu_encoder_virt *dpu_enc) +{ + /* coding only for 2LM, 2enc, 1 dsc config */ + struct dpu_encoder_phys *enc_master = dpu_enc->cur_master; + struct dpu_hw_ctl *ctl = enc_master->hw_ctl; + struct dpu_hw_dsc *hw_dsc[MAX_CHANNELS_PER_ENC]; + struct dpu_hw_pingpong *hw_pp[MAX_CHANNELS_PER_ENC]; + int i; + + for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) { + hw_pp[i] = dpu_enc->hw_pp[i]; + hw_dsc[i] = dpu_enc->hw_dsc[i]; + + if (hw_pp[i] && hw_dsc[i]) + dpu_encoder_dsc_pipe_clr(ctl, hw_dsc[i], hw_pp[i]); + } +} + +/** + * dpu_encoder_helper_phys_cleanup - helper to cleanup dpu pipeline + * @phys_enc: Pointer to physical encoder structure + */ +void dpu_encoder_helper_phys_cleanup(struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_ctl *ctl = phys_enc->hw_ctl; + struct dpu_hw_intf_cfg intf_cfg = { 0 }; + int i; + struct dpu_encoder_virt *dpu_enc; + + dpu_enc = to_dpu_encoder_virt(phys_enc->parent); + + ctl->ops.reset(ctl); + + dpu_encoder_helper_reset_mixers(phys_enc); + + /* + * TODO: move the once-only operation like CTL flush/trigger + * into dpu_encoder_virt_disable() and all operations which need + * to be done per phys encoder into the phys_disable() op. + */ + if (phys_enc->hw_wb) { + /* disable the PP block */ + if (phys_enc->hw_wb->ops.bind_pingpong_blk) + phys_enc->hw_wb->ops.bind_pingpong_blk(phys_enc->hw_wb, PINGPONG_NONE); + + /* mark WB flush as pending */ + if (ctl->ops.update_pending_flush_wb) + ctl->ops.update_pending_flush_wb(ctl, phys_enc->hw_wb->idx); + } else { + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + if (dpu_enc->phys_encs[i] && phys_enc->hw_intf->ops.bind_pingpong_blk) + phys_enc->hw_intf->ops.bind_pingpong_blk( + dpu_enc->phys_encs[i]->hw_intf, + PINGPONG_NONE); + + /* mark INTF flush as pending */ + if (ctl->ops.update_pending_flush_intf) + ctl->ops.update_pending_flush_intf(ctl, + dpu_enc->phys_encs[i]->hw_intf->idx); + } + } + + if (phys_enc->hw_pp && phys_enc->hw_pp->ops.setup_dither) + phys_enc->hw_pp->ops.setup_dither(phys_enc->hw_pp, NULL); + + if (dpu_enc->cwb_mask) + dpu_encoder_helper_phys_setup_cwb(phys_enc, false); + + /* reset the merge 3D HW block */ + if (phys_enc->hw_pp && phys_enc->hw_pp->merge_3d) { + phys_enc->hw_pp->merge_3d->ops.setup_3d_mode(phys_enc->hw_pp->merge_3d, + BLEND_3D_NONE); + if (ctl->ops.update_pending_flush_merge_3d) + ctl->ops.update_pending_flush_merge_3d(ctl, + phys_enc->hw_pp->merge_3d->idx); + } + + if (phys_enc->hw_cdm) { + if (phys_enc->hw_cdm->ops.bind_pingpong_blk && phys_enc->hw_pp) + phys_enc->hw_cdm->ops.bind_pingpong_blk(phys_enc->hw_cdm, + PINGPONG_NONE); + if (ctl->ops.update_pending_flush_cdm) + ctl->ops.update_pending_flush_cdm(ctl, + phys_enc->hw_cdm->idx); + } + + if (dpu_enc->dsc) { + dpu_encoder_unprep_dsc(dpu_enc); + dpu_enc->dsc = NULL; + } + + intf_cfg.stream_sel = 0; /* Don't care value for video mode */ + intf_cfg.mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc); + intf_cfg.dsc = dpu_encoder_helper_get_dsc(phys_enc); + intf_cfg.cwb = dpu_enc->cwb_mask; + + if (phys_enc->hw_intf) + intf_cfg.intf = phys_enc->hw_intf->idx; + if (phys_enc->hw_wb) + intf_cfg.wb = phys_enc->hw_wb->idx; + + if (phys_enc->hw_pp && phys_enc->hw_pp->merge_3d) + intf_cfg.merge_3d = phys_enc->hw_pp->merge_3d->idx; + + if (ctl->ops.reset_intf_cfg) + ctl->ops.reset_intf_cfg(ctl, &intf_cfg); + + ctl->ops.trigger_flush(ctl); + ctl->ops.trigger_start(ctl); + ctl->ops.clear_pending_flush(ctl); +} + +void dpu_encoder_helper_phys_setup_cwb(struct dpu_encoder_phys *phys_enc, + bool enable) +{ + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(phys_enc->parent); + struct dpu_hw_cwb *hw_cwb; + struct dpu_hw_ctl *hw_ctl; + struct dpu_hw_cwb_setup_cfg cwb_cfg; + + struct dpu_kms *dpu_kms; + struct dpu_global_state *global_state; + struct dpu_hw_blk *rt_pp_list[MAX_CHANNELS_PER_ENC]; + int num_pp; + + if (!phys_enc->hw_wb) + return; + + hw_ctl = phys_enc->hw_ctl; + + if (!phys_enc->hw_ctl) { + DPU_DEBUG("[wb:%d] no ctl assigned\n", + phys_enc->hw_wb->idx - WB_0); + return; + } + + dpu_kms = phys_enc->dpu_kms; + global_state = dpu_kms_get_existing_global_state(dpu_kms); + num_pp = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + phys_enc->parent->crtc, + DPU_HW_BLK_PINGPONG, rt_pp_list, + ARRAY_SIZE(rt_pp_list)); + + if (num_pp == 0 || num_pp > MAX_CHANNELS_PER_ENC) { + DPU_DEBUG_ENC(dpu_enc, "invalid num_pp %d\n", num_pp); + return; + } + + /* + * The CWB mux supports using LM or DSPP as tap points. For now, + * always use LM tap point + */ + cwb_cfg.input = INPUT_MODE_LM_OUT; + + for (int i = 0; i < MAX_CWB_PER_ENC; i++) { + hw_cwb = dpu_enc->hw_cwb[i]; + if (!hw_cwb) + continue; + + if (enable) { + struct dpu_hw_pingpong *hw_pp = + to_dpu_hw_pingpong(rt_pp_list[i]); + cwb_cfg.pp_idx = hw_pp->idx; + } else { + cwb_cfg.pp_idx = PINGPONG_NONE; + } + + hw_cwb->ops.config_cwb(hw_cwb, &cwb_cfg); + + if (hw_ctl->ops.update_pending_flush_cwb) + hw_ctl->ops.update_pending_flush_cwb(hw_ctl, hw_cwb->idx); + } +} + +/** + * dpu_encoder_helper_phys_setup_cdm - setup chroma down sampling block + * @phys_enc: Pointer to physical encoder + * @dpu_fmt: Pinter to the format description + * @output_type: HDMI/WB + */ +void dpu_encoder_helper_phys_setup_cdm(struct dpu_encoder_phys *phys_enc, + const struct msm_format *dpu_fmt, + u32 output_type) +{ + struct dpu_hw_cdm *hw_cdm; + struct dpu_hw_cdm_cfg *cdm_cfg; + struct dpu_hw_pingpong *hw_pp; + int ret; + + if (!phys_enc) + return; + + cdm_cfg = &phys_enc->cdm_cfg; + hw_pp = phys_enc->hw_pp; + hw_cdm = phys_enc->hw_cdm; + + if (!hw_cdm) + return; + + if (!MSM_FORMAT_IS_YUV(dpu_fmt)) { + DPU_DEBUG("[enc:%d] cdm_disable fmt:%p4cc\n", DRMID(phys_enc->parent), + &dpu_fmt->pixel_format); + if (hw_cdm->ops.bind_pingpong_blk) + hw_cdm->ops.bind_pingpong_blk(hw_cdm, PINGPONG_NONE); + + return; + } + + memset(cdm_cfg, 0, sizeof(struct dpu_hw_cdm_cfg)); + + cdm_cfg->output_width = phys_enc->cached_mode.hdisplay; + cdm_cfg->output_height = phys_enc->cached_mode.vdisplay; + cdm_cfg->output_fmt = dpu_fmt; + cdm_cfg->output_type = output_type; + cdm_cfg->output_bit_depth = MSM_FORMAT_IS_DX(dpu_fmt) ? + CDM_CDWN_OUTPUT_10BIT : CDM_CDWN_OUTPUT_8BIT; + cdm_cfg->csc_cfg = &dpu_csc10_rgb2yuv_601l; + + /* enable 10 bit logic */ + switch (cdm_cfg->output_fmt->chroma_sample) { + case CHROMA_FULL: + cdm_cfg->h_cdwn_type = CDM_CDWN_DISABLE; + cdm_cfg->v_cdwn_type = CDM_CDWN_DISABLE; + break; + case CHROMA_H2V1: + cdm_cfg->h_cdwn_type = CDM_CDWN_COSITE; + cdm_cfg->v_cdwn_type = CDM_CDWN_DISABLE; + break; + case CHROMA_420: + cdm_cfg->h_cdwn_type = CDM_CDWN_COSITE; + cdm_cfg->v_cdwn_type = CDM_CDWN_OFFSITE; + break; + case CHROMA_H1V2: + default: + DPU_ERROR("[enc:%d] unsupported chroma sampling type\n", + DRMID(phys_enc->parent)); + cdm_cfg->h_cdwn_type = CDM_CDWN_DISABLE; + cdm_cfg->v_cdwn_type = CDM_CDWN_DISABLE; + break; + } + + DPU_DEBUG("[enc:%d] cdm_enable:%d,%d,%p4cc,%d,%d,%d,%d]\n", + DRMID(phys_enc->parent), cdm_cfg->output_width, + cdm_cfg->output_height, &cdm_cfg->output_fmt->pixel_format, + cdm_cfg->output_type, cdm_cfg->output_bit_depth, + cdm_cfg->h_cdwn_type, cdm_cfg->v_cdwn_type); + + if (hw_cdm->ops.enable) { + cdm_cfg->pp_id = hw_pp->idx; + ret = hw_cdm->ops.enable(hw_cdm, cdm_cfg); + if (ret < 0) { + DPU_ERROR("[enc:%d] failed to enable CDM; ret:%d\n", + DRMID(phys_enc->parent), ret); + return; + } + } +} + +#ifdef CONFIG_DEBUG_FS +static int _dpu_encoder_status_show(struct seq_file *s, void *data) +{ + struct drm_encoder *drm_enc = s->private; + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); + int i; + + mutex_lock(&dpu_enc->enc_lock); + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + + seq_printf(s, "intf:%d wb:%d vsync:%8d underrun:%8d frame_done_cnt:%d", + phys->hw_intf ? phys->hw_intf->idx - INTF_0 : -1, + phys->hw_wb ? phys->hw_wb->idx - WB_0 : -1, + atomic_read(&phys->vsync_cnt), + atomic_read(&phys->underrun_cnt), + atomic_read(&dpu_enc->frame_done_timeout_cnt)); + + seq_printf(s, "mode: %s\n", dpu_encoder_helper_get_intf_type(phys->intf_mode)); + } + mutex_unlock(&dpu_enc->enc_lock); + + return 0; +} + +DEFINE_SHOW_ATTRIBUTE(_dpu_encoder_status); + +static void dpu_encoder_debugfs_init(struct drm_encoder *drm_enc, struct dentry *root) +{ + /* don't error check these */ + debugfs_create_file("status", 0600, + root, drm_enc, &_dpu_encoder_status_fops); +} +#else +#define dpu_encoder_debugfs_init NULL +#endif + +static int dpu_encoder_virt_add_phys_encs( + struct drm_device *dev, + struct msm_display_info *disp_info, + struct dpu_encoder_virt *dpu_enc, + struct dpu_enc_phys_init_params *params) +{ + struct dpu_encoder_phys *enc = NULL; + + DPU_DEBUG_ENC(dpu_enc, "\n"); + + /* + * We may create up to NUM_PHYS_ENCODER_TYPES physical encoder types + * in this function, check up-front. + */ + if (dpu_enc->num_phys_encs + NUM_PHYS_ENCODER_TYPES >= + ARRAY_SIZE(dpu_enc->phys_encs)) { + DPU_ERROR_ENC(dpu_enc, "too many physical encoders %d\n", + dpu_enc->num_phys_encs); + return -EINVAL; + } + + + if (disp_info->intf_type == INTF_WB) { + enc = dpu_encoder_phys_wb_init(dev, params); + + if (IS_ERR(enc)) { + DPU_ERROR_ENC(dpu_enc, "failed to init wb enc: %ld\n", + PTR_ERR(enc)); + return PTR_ERR(enc); + } + + dpu_enc->phys_encs[dpu_enc->num_phys_encs] = enc; + ++dpu_enc->num_phys_encs; + } else if (disp_info->is_cmd_mode) { + enc = dpu_encoder_phys_cmd_init(dev, params); + + if (IS_ERR(enc)) { + DPU_ERROR_ENC(dpu_enc, "failed to init cmd enc: %ld\n", + PTR_ERR(enc)); + return PTR_ERR(enc); + } + + dpu_enc->phys_encs[dpu_enc->num_phys_encs] = enc; + ++dpu_enc->num_phys_encs; + } else { + enc = dpu_encoder_phys_vid_init(dev, params); + + if (IS_ERR(enc)) { + DPU_ERROR_ENC(dpu_enc, "failed to init vid enc: %ld\n", + PTR_ERR(enc)); + return PTR_ERR(enc); + } + + dpu_enc->phys_encs[dpu_enc->num_phys_encs] = enc; + ++dpu_enc->num_phys_encs; + } + + if (params->split_role == ENC_ROLE_SLAVE) + dpu_enc->cur_slave = enc; + else + dpu_enc->cur_master = enc; + + return 0; +} + +/** + * dpu_encoder_get_clones - Calculate the possible_clones for DPU encoder + * @drm_enc: DRM encoder pointer + * Returns: possible_clones mask + */ +uint32_t dpu_encoder_get_clones(struct drm_encoder *drm_enc) +{ + struct drm_encoder *curr; + int type = drm_enc->encoder_type; + uint32_t clone_mask = drm_encoder_mask(drm_enc); + + /* + * Set writeback as possible clones of real-time DSI encoders and vice + * versa + * + * Writeback encoders can't be clones of each other and DSI + * encoders can't be clones of each other. + * + * TODO: Add DP encoders as valid possible clones for writeback encoders + * (and vice versa) once concurrent writeback has been validated for DP + */ + drm_for_each_encoder(curr, drm_enc->dev) { + if ((type == DRM_MODE_ENCODER_VIRTUAL && + curr->encoder_type == DRM_MODE_ENCODER_DSI) || + (type == DRM_MODE_ENCODER_DSI && + curr->encoder_type == DRM_MODE_ENCODER_VIRTUAL)) + clone_mask |= drm_encoder_mask(curr); + } + + return clone_mask; +} + +static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc, + struct dpu_kms *dpu_kms, + struct msm_display_info *disp_info) +{ + int ret = 0; + int i = 0; + struct dpu_enc_phys_init_params phys_params; + + if (!dpu_enc) { + DPU_ERROR("invalid arg(s), enc %d\n", dpu_enc != NULL); + return -EINVAL; + } + + dpu_enc->cur_master = NULL; + + memset(&phys_params, 0, sizeof(phys_params)); + phys_params.dpu_kms = dpu_kms; + phys_params.parent = &dpu_enc->base; + phys_params.enc_spinlock = &dpu_enc->enc_spinlock; + + WARN_ON(disp_info->num_of_h_tiles < 1); + + DPU_DEBUG("dsi_info->num_of_h_tiles %d\n", disp_info->num_of_h_tiles); + + if (disp_info->intf_type != INTF_WB) + dpu_enc->idle_pc_supported = + dpu_kms->catalog->caps->has_idle_pc; + + mutex_lock(&dpu_enc->enc_lock); + for (i = 0; i < disp_info->num_of_h_tiles && !ret; i++) { + /* + * Left-most tile is at index 0, content is controller id + * h_tile_instance_ids[2] = {0, 1}; DSI0 = left, DSI1 = right + * h_tile_instance_ids[2] = {1, 0}; DSI1 = left, DSI0 = right + */ + u32 controller_id = disp_info->h_tile_instance[i]; + + if (disp_info->num_of_h_tiles > 1) { + if (i == 0) + phys_params.split_role = ENC_ROLE_MASTER; + else + phys_params.split_role = ENC_ROLE_SLAVE; + } else { + phys_params.split_role = ENC_ROLE_SOLO; + } + + DPU_DEBUG("h_tile_instance %d = %d, split_role %d\n", + i, controller_id, phys_params.split_role); + + phys_params.hw_intf = dpu_encoder_get_intf(dpu_kms->catalog, &dpu_kms->rm, + disp_info->intf_type, + controller_id); + + if (disp_info->intf_type == INTF_WB && controller_id < WB_MAX) + phys_params.hw_wb = dpu_rm_get_wb(&dpu_kms->rm, controller_id); + + if (!phys_params.hw_intf && !phys_params.hw_wb) { + DPU_ERROR_ENC(dpu_enc, "no intf or wb block assigned at idx: %d\n", i); + ret = -EINVAL; + break; + } + + if (phys_params.hw_intf && phys_params.hw_wb) { + DPU_ERROR_ENC(dpu_enc, + "invalid phys both intf and wb block at idx: %d\n", i); + ret = -EINVAL; + break; + } + + ret = dpu_encoder_virt_add_phys_encs(dpu_kms->dev, disp_info, + dpu_enc, &phys_params); + if (ret) { + DPU_ERROR_ENC(dpu_enc, "failed to add phys encs\n"); + break; + } + } + + mutex_unlock(&dpu_enc->enc_lock); + + return ret; +} + +static void dpu_encoder_frame_done_timeout(struct timer_list *t) +{ + struct dpu_encoder_virt *dpu_enc = timer_container_of(dpu_enc, t, + frame_done_timer); + struct drm_encoder *drm_enc = &dpu_enc->base; + u32 event; + + if (!drm_enc->dev) { + DPU_ERROR("invalid parameters\n"); + return; + } + + if (!dpu_enc->frame_busy_mask[0] || !dpu_enc->crtc) { + DRM_DEBUG_KMS("id:%u invalid timeout frame_busy_mask=%lu\n", + DRMID(drm_enc), dpu_enc->frame_busy_mask[0]); + return; + } else if (!atomic_xchg(&dpu_enc->frame_done_timeout_ms, 0)) { + DRM_DEBUG_KMS("id:%u invalid timeout\n", DRMID(drm_enc)); + return; + } + + DPU_ERROR_ENC_RATELIMITED(dpu_enc, "frame done timeout\n"); + + if (atomic_inc_return(&dpu_enc->frame_done_timeout_cnt) == 1) + msm_disp_snapshot_state(drm_enc->dev); + + event = DPU_ENCODER_FRAME_EVENT_ERROR; + trace_dpu_enc_frame_done_timeout(DRMID(drm_enc), event); + dpu_crtc_frame_event_cb(dpu_enc->crtc, event); +} + +static const struct drm_encoder_helper_funcs dpu_encoder_helper_funcs = { + .atomic_mode_set = dpu_encoder_virt_atomic_mode_set, + .atomic_disable = dpu_encoder_virt_atomic_disable, + .atomic_enable = dpu_encoder_virt_atomic_enable, +}; + +static const struct drm_encoder_funcs dpu_encoder_funcs = { + .debugfs_init = dpu_encoder_debugfs_init, +}; + +/** + * dpu_encoder_init - initialize virtual encoder object + * @dev: Pointer to drm device structure + * @drm_enc_mode: corresponding DRM_MODE_ENCODER_* constant + * @disp_info: Pointer to display information structure + * Returns: Pointer to newly created drm encoder + */ +struct drm_encoder *dpu_encoder_init(struct drm_device *dev, + int drm_enc_mode, + struct msm_display_info *disp_info) +{ + struct msm_drm_private *priv = dev->dev_private; + struct dpu_kms *dpu_kms = to_dpu_kms(priv->kms); + struct dpu_encoder_virt *dpu_enc; + int ret; + + dpu_enc = drmm_encoder_alloc(dev, struct dpu_encoder_virt, base, + &dpu_encoder_funcs, drm_enc_mode, NULL); + if (IS_ERR(dpu_enc)) + return ERR_CAST(dpu_enc); + + drm_encoder_helper_add(&dpu_enc->base, &dpu_encoder_helper_funcs); + + spin_lock_init(&dpu_enc->enc_spinlock); + dpu_enc->enabled = false; + mutex_init(&dpu_enc->enc_lock); + mutex_init(&dpu_enc->rc_lock); + + ret = dpu_encoder_setup_display(dpu_enc, dpu_kms, disp_info); + if (ret) { + DPU_ERROR("failed to setup encoder\n"); + return ERR_PTR(-ENOMEM); + } + + atomic_set(&dpu_enc->frame_done_timeout_ms, 0); + atomic_set(&dpu_enc->frame_done_timeout_cnt, 0); + timer_setup(&dpu_enc->frame_done_timer, + dpu_encoder_frame_done_timeout, 0); + + INIT_DELAYED_WORK(&dpu_enc->delayed_off_work, + dpu_encoder_off_work); + dpu_enc->idle_timeout = IDLE_TIMEOUT; + + memcpy(&dpu_enc->disp_info, disp_info, sizeof(*disp_info)); + + DPU_DEBUG_ENC(dpu_enc, "created\n"); + + return &dpu_enc->base; +} + +/** + * dpu_encoder_wait_for_commit_done() - Wait for encoder to flush pending state + * @drm_enc: encoder pointer + * + * Wait for hardware to have flushed the current pending changes to hardware at + * a vblank or CTL_START. Physical encoders will map this differently depending + * on the type: vid mode -> vsync_irq, cmd mode -> CTL_START. + * + * Return: 0 on success, -EWOULDBLOCK if already signaled, error otherwise + */ +int dpu_encoder_wait_for_commit_done(struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc = NULL; + int i, ret = 0; + + if (!drm_enc) { + DPU_ERROR("invalid encoder\n"); + return -EINVAL; + } + dpu_enc = to_dpu_encoder_virt(drm_enc); + DPU_DEBUG_ENC(dpu_enc, "\n"); + + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + + if (phys->ops.wait_for_commit_done) { + DPU_ATRACE_BEGIN("wait_for_commit_done"); + ret = phys->ops.wait_for_commit_done(phys); + DPU_ATRACE_END("wait_for_commit_done"); + if (ret == -ETIMEDOUT && !dpu_enc->commit_done_timedout) { + dpu_enc->commit_done_timedout = true; + msm_disp_snapshot_state(drm_enc->dev); + } + if (ret) + return ret; + } + } + + return ret; +} + +/** + * dpu_encoder_wait_for_tx_complete() - Wait for encoder to transfer pixels to panel + * @drm_enc: encoder pointer + * + * Wait for the hardware to transfer all the pixels to the panel. Physical + * encoders will map this differently depending on the type: vid mode -> vsync_irq, + * cmd mode -> pp_done. + * + * Return: 0 on success, -EWOULDBLOCK if already signaled, error otherwise + */ +int dpu_encoder_wait_for_tx_complete(struct drm_encoder *drm_enc) +{ + struct dpu_encoder_virt *dpu_enc = NULL; + int i, ret = 0; + + if (!drm_enc) { + DPU_ERROR("invalid encoder\n"); + return -EINVAL; + } + dpu_enc = to_dpu_encoder_virt(drm_enc); + DPU_DEBUG_ENC(dpu_enc, "\n"); + + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + + if (phys->ops.wait_for_tx_complete) { + DPU_ATRACE_BEGIN("wait_for_tx_complete"); + ret = phys->ops.wait_for_tx_complete(phys); + DPU_ATRACE_END("wait_for_tx_complete"); + if (ret) + return ret; + } + } + + return ret; +} + +/** + * dpu_encoder_get_intf_mode - get interface mode of the given encoder + * @encoder: Pointer to drm encoder object + */ +enum dpu_intf_mode dpu_encoder_get_intf_mode(struct drm_encoder *encoder) +{ + struct dpu_encoder_virt *dpu_enc = NULL; + + if (!encoder) { + DPU_ERROR("invalid encoder\n"); + return INTF_MODE_NONE; + } + dpu_enc = to_dpu_encoder_virt(encoder); + + if (dpu_enc->cur_master) + return dpu_enc->cur_master->intf_mode; + + if (dpu_enc->num_phys_encs) + return dpu_enc->phys_encs[0]->intf_mode; + + return INTF_MODE_NONE; +} + +/** + * dpu_encoder_helper_get_cwb_mask - get CWB blocks mask for the DPU encoder + * @phys_enc: Pointer to physical encoder structure + */ +unsigned int dpu_encoder_helper_get_cwb_mask(struct dpu_encoder_phys *phys_enc) +{ + struct drm_encoder *encoder = phys_enc->parent; + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(encoder); + + return dpu_enc->cwb_mask; +} + +/** + * dpu_encoder_helper_get_dsc - get DSC blocks mask for the DPU encoder + * This helper function is used by physical encoder to get DSC blocks mask + * used for this encoder. + * @phys_enc: Pointer to physical encoder structure + */ +unsigned int dpu_encoder_helper_get_dsc(struct dpu_encoder_phys *phys_enc) +{ + struct drm_encoder *encoder = phys_enc->parent; + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(encoder); + + return dpu_enc->dsc_mask; +} + +void dpu_encoder_phys_init(struct dpu_encoder_phys *phys_enc, + struct dpu_enc_phys_init_params *p) +{ + phys_enc->hw_mdptop = p->dpu_kms->hw_mdp; + phys_enc->hw_intf = p->hw_intf; + phys_enc->hw_wb = p->hw_wb; + phys_enc->parent = p->parent; + phys_enc->dpu_kms = p->dpu_kms; + phys_enc->split_role = p->split_role; + phys_enc->enc_spinlock = p->enc_spinlock; + phys_enc->enable_state = DPU_ENC_DISABLED; + + atomic_set(&phys_enc->pending_kickoff_cnt, 0); + atomic_set(&phys_enc->pending_ctlstart_cnt, 0); + + atomic_set(&phys_enc->vsync_cnt, 0); + atomic_set(&phys_enc->underrun_cnt, 0); + + init_waitqueue_head(&phys_enc->pending_kickoff_wq); +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h new file mode 100644 index 000000000..4942097e7 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h @@ -0,0 +1,101 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022-2024 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#ifndef __DPU_ENCODER_H__ +#define __DPU_ENCODER_H__ + +#include +#include "dpu_hw_mdss.h" + +#define DPU_ENCODER_FRAME_EVENT_DONE BIT(0) +#define DPU_ENCODER_FRAME_EVENT_ERROR BIT(1) +#define DPU_ENCODER_FRAME_EVENT_PANEL_DEAD BIT(2) +#define DPU_ENCODER_FRAME_EVENT_IDLE BIT(3) + +#define IDLE_TIMEOUT (66 - 16/2) + +#define MAX_H_TILES_PER_DISPLAY 2 + +/** + * struct msm_display_info - defines display properties + * @intf_type: INTF_ type + * @num_of_h_tiles: Number of horizontal tiles in case of split interface + * @h_tile_instance: Controller instance used per tile. Number of elements is + * based on num_of_h_tiles + * @is_cmd_mode Boolean to indicate if the CMD mode is requested + * @vsync_source: Source of the TE signal for DSI CMD devices + */ +struct msm_display_info { + enum dpu_intf_type intf_type; + uint32_t num_of_h_tiles; + uint32_t h_tile_instance[MAX_H_TILES_PER_DISPLAY]; + bool is_cmd_mode; + enum dpu_vsync_source vsync_source; +}; + +void dpu_encoder_assign_crtc(struct drm_encoder *encoder, + struct drm_crtc *crtc); + +void dpu_encoder_toggle_vblank_for_crtc(struct drm_encoder *encoder, + struct drm_crtc *crtc, bool enable); + +void dpu_encoder_prepare_for_kickoff(struct drm_encoder *encoder); + +void dpu_encoder_trigger_kickoff_pending(struct drm_encoder *encoder); + +void dpu_encoder_kickoff(struct drm_encoder *encoder); + +int dpu_encoder_vsync_time(struct drm_encoder *drm_enc, ktime_t *wakeup_time); + +int dpu_encoder_wait_for_commit_done(struct drm_encoder *drm_encoder); + +int dpu_encoder_wait_for_tx_complete(struct drm_encoder *drm_encoder); + +enum dpu_intf_mode dpu_encoder_get_intf_mode(struct drm_encoder *encoder); + +void dpu_encoder_virt_runtime_resume(struct drm_encoder *encoder); + +uint32_t dpu_encoder_get_clones(struct drm_encoder *drm_enc); + +struct drm_encoder *dpu_encoder_init(struct drm_device *dev, + int drm_enc_mode, + struct msm_display_info *disp_info); + +int dpu_encoder_get_linecount(struct drm_encoder *drm_enc); + +int dpu_encoder_get_vsync_count(struct drm_encoder *drm_enc); + +bool dpu_encoder_is_widebus_enabled(const struct drm_encoder *drm_enc); + +bool dpu_encoder_is_dsc_enabled(const struct drm_encoder *drm_enc); + +int dpu_encoder_get_crc_values_cnt(const struct drm_encoder *drm_enc); + +void dpu_encoder_setup_misr(const struct drm_encoder *drm_encoder); + +int dpu_encoder_get_crc(const struct drm_encoder *drm_enc, u32 *crcs, int pos); + +bool dpu_encoder_use_dsc_merge(struct drm_encoder *drm_enc); + +void dpu_encoder_update_topology(struct drm_encoder *drm_enc, + struct msm_display_topology *topology, + struct drm_atomic_commit *state, + const struct drm_display_mode *adj_mode); + +bool dpu_encoder_needs_modeset(struct drm_encoder *drm_enc, struct drm_atomic_commit *state); + +void dpu_encoder_prepare_wb_job(struct drm_encoder *drm_enc, + struct drm_writeback_job *job); + +void dpu_encoder_cleanup_wb_job(struct drm_encoder *drm_enc, + struct drm_writeback_job *job); + +bool dpu_encoder_is_valid_for_commit(struct drm_encoder *drm_enc); + +void dpu_encoder_start_frame_done_timer(struct drm_encoder *drm_enc); +#endif /* __DPU_ENCODER_H__ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h new file mode 100644 index 000000000..61b22d949 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h @@ -0,0 +1,356 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022-2024 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018 The Linux Foundation. All rights reserved. + */ + +#ifndef __DPU_ENCODER_PHYS_H__ +#define __DPU_ENCODER_PHYS_H__ + +#include +#include + +#include "dpu_kms.h" +#include "dpu_hw_intf.h" +#include "dpu_hw_wb.h" +#include "dpu_hw_pingpong.h" +#include "dpu_hw_cdm.h" +#include "dpu_hw_ctl.h" +#include "dpu_hw_top.h" +#include "dpu_hw_util.h" +#include "dpu_encoder.h" +#include "dpu_crtc.h" + +#define DPU_ENCODER_NAME_MAX 16 + +/* wait for at most 2 vsync for lowest refresh rate (24hz) */ +#define KICKOFF_TIMEOUT_MS 84 +#define KICKOFF_TIMEOUT_JIFFIES msecs_to_jiffies(KICKOFF_TIMEOUT_MS) + +/** + * enum dpu_enc_split_role - Role this physical encoder will play in a + * split-panel configuration, where one panel is master, and others slaves. + * Masters have extra responsibilities, like managing the VBLANK IRQ. + * @ENC_ROLE_SOLO: This is the one and only panel. This encoder is master. + * @ENC_ROLE_MASTER: This encoder is the master of a split panel config. + * @ENC_ROLE_SLAVE: This encoder is not the master of a split panel config. + */ +enum dpu_enc_split_role { + ENC_ROLE_SOLO, + ENC_ROLE_MASTER, + ENC_ROLE_SLAVE, +}; + +/** + * enum dpu_enc_enable_state - current enabled state of the physical encoder + * @DPU_ENC_DISABLING: Encoder transitioning to disable state + * Events bounding transition are encoder type specific + * @DPU_ENC_DISABLED: Encoder is disabled + * @DPU_ENC_ENABLING: Encoder transitioning to enabled + * Events bounding transition are encoder type specific + * @DPU_ENC_ENABLED: Encoder is enabled + * @DPU_ENC_ERR_NEEDS_HW_RESET: Encoder is enabled, but requires a hw_reset + * to recover from a previous error + */ +enum dpu_enc_enable_state { + DPU_ENC_DISABLING, + DPU_ENC_DISABLED, + DPU_ENC_ENABLING, + DPU_ENC_ENABLED, + DPU_ENC_ERR_NEEDS_HW_RESET +}; + +struct dpu_encoder_phys; + +/** + * struct dpu_encoder_phys_ops - Interface the physical encoders provide to + * the containing virtual encoder. + * @prepare_commit: MSM Atomic Call, start of atomic commit sequence + * @is_master: Whether this phys_enc is the current master + * encoder. Can be switched at enable time. Based + * on split_role and current mode (CMD/VID). + * @atomic_mode_set: DRM Call. Set a DRM mode. + * This likely caches the mode, for use at enable. + * @enable: DRM Call. Enable a DRM mode. + * @disable: DRM Call. Disable mode. + * @control_vblank_irq Register/Deregister for VBLANK IRQ + * @wait_for_commit_done: Wait for hardware to have flushed the + * current pending frames to hardware + * @wait_for_tx_complete: Wait for hardware to transfer the pixels + * to the panel + * @wait_for_vblank: Wait for VBLANK, for sub-driver internal use + * @prepare_for_kickoff: Do any work necessary prior to a kickoff + * For CMD encoder, may wait for previous tx done + * @handle_post_kickoff: Do any work necessary post-kickoff work + * @trigger_start: Process start event on physical encoder + * @needs_single_flush: Whether encoder slaves need to be flushed + * @irq_enable: Handler to enable all the encoder IRQs + * @irq_disable: Handler to disable all the encoder IRQs + * @prepare_idle_pc: phys encoder can update the vsync_enable status + * on idle power collapse prepare + * @restore: Restore all the encoder configs. + * @get_line_count: Obtain current vertical line count + */ + +struct dpu_encoder_phys_ops { + void (*prepare_commit)(struct dpu_encoder_phys *encoder); + bool (*is_master)(struct dpu_encoder_phys *encoder); + void (*atomic_mode_set)(struct dpu_encoder_phys *encoder, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state); + void (*enable)(struct dpu_encoder_phys *encoder); + void (*disable)(struct dpu_encoder_phys *encoder); + int (*control_vblank_irq)(struct dpu_encoder_phys *enc, bool enable); + int (*wait_for_commit_done)(struct dpu_encoder_phys *phys_enc); + int (*wait_for_tx_complete)(struct dpu_encoder_phys *phys_enc); + void (*prepare_for_kickoff)(struct dpu_encoder_phys *phys_enc); + void (*handle_post_kickoff)(struct dpu_encoder_phys *phys_enc); + void (*trigger_start)(struct dpu_encoder_phys *phys_enc); + bool (*needs_single_flush)(struct dpu_encoder_phys *phys_enc); + void (*irq_enable)(struct dpu_encoder_phys *phys); + void (*irq_disable)(struct dpu_encoder_phys *phys); + void (*prepare_idle_pc)(struct dpu_encoder_phys *phys_enc); + void (*restore)(struct dpu_encoder_phys *phys); + int (*get_line_count)(struct dpu_encoder_phys *phys); + int (*get_frame_count)(struct dpu_encoder_phys *phys); + void (*prepare_wb_job)(struct dpu_encoder_phys *phys_enc, + struct drm_writeback_job *job); + void (*cleanup_wb_job)(struct dpu_encoder_phys *phys_enc, + struct drm_writeback_job *job); + bool (*is_valid_for_commit)(struct dpu_encoder_phys *phys_enc); +}; + +/** + * enum dpu_intr_idx - dpu encoder interrupt index + * @INTR_IDX_VSYNC: Vsync interrupt for video mode panel + * @INTR_IDX_PINGPONG: Pingpong done interrupt for cmd mode panel + * @INTR_IDX_UNDERRUN: Underrun interrupt for video and cmd mode panel + * @INTR_IDX_RDPTR: Readpointer done interrupt for cmd mode panel + * @INTR_IDX_WB_DONE: Writeback done interrupt for virtual connector + */ +enum dpu_intr_idx { + INTR_IDX_VSYNC, + INTR_IDX_PINGPONG, + INTR_IDX_UNDERRUN, + INTR_IDX_CTL_START, + INTR_IDX_RDPTR, + INTR_IDX_WB_DONE, + INTR_IDX_MAX, +}; + +/** + * struct dpu_encoder_phys - physical encoder that drives a single INTF block + * tied to a specific panel / sub-panel. Abstract type, sub-classed by + * phys_vid or phys_cmd for video mode or command mode encs respectively. + * @parent: Pointer to the containing virtual encoder + * @ops: Operations exposed to the virtual encoder + * @parent_ops: Callbacks exposed by the parent to the phys_enc + * @hw_mdptop: Hardware interface to the top registers + * @hw_ctl: Hardware interface to the ctl registers + * @hw_pp: Hardware interface to the ping pong registers + * @hw_intf: Hardware interface to the intf registers + * @hw_wb: Hardware interface to the wb registers + * @hw_cdm: Hardware interface to the CDM registers + * @dpu_kms: Pointer to the dpu_kms top level + * @cdm_cfg: CDM block config needed to store WB/DP block's CDM configuration + * @cached_mode: DRM mode cached at mode_set time, acted on in enable + * @vblank_ctl_lock: Vblank ctl mutex lock to protect vblank_refcount + * @enabled: Whether the encoder has enabled and running a mode + * @split_role: Role to play in a split-panel configuration + * @intf_mode: Interface mode + * @enc_spinlock: Virtual-Encoder-Wide Spin Lock for IRQ purposes + * @enable_state: Enable state tracking + * @vblank_refcount: Reference count of vblank request + * @vsync_cnt: Vsync count for the physical encoder + * @underrun_cnt: Underrun count for the physical encoder + * @pending_kickoff_cnt: Atomic counter tracking the number of kickoffs + * vs. the number of done/vblank irqs. Should hover + * between 0-2 Incremented when a new kickoff is + * scheduled. Decremented in irq handler + * @pending_ctlstart_cnt: Atomic counter tracking the number of ctl start + * pending. + * @pending_kickoff_wq: Wait queue for blocking until kickoff completes + * @irq: IRQ indices + * @has_intf_te: Interface TE configuration support + */ +struct dpu_encoder_phys { + struct drm_encoder *parent; + struct dpu_encoder_phys_ops ops; + struct dpu_hw_mdp *hw_mdptop; + struct dpu_hw_ctl *hw_ctl; + struct dpu_hw_pingpong *hw_pp; + struct dpu_hw_intf *hw_intf; + struct dpu_hw_wb *hw_wb; + struct dpu_hw_cdm *hw_cdm; + struct dpu_kms *dpu_kms; + struct dpu_hw_cdm_cfg cdm_cfg; + struct drm_display_mode cached_mode; + struct mutex vblank_ctl_lock; + enum dpu_enc_split_role split_role; + enum dpu_intf_mode intf_mode; + spinlock_t *enc_spinlock; + enum dpu_enc_enable_state enable_state; + int vblank_refcount; + atomic_t vsync_cnt; + atomic_t underrun_cnt; + atomic_t pending_ctlstart_cnt; + atomic_t pending_kickoff_cnt; + wait_queue_head_t pending_kickoff_wq; + unsigned int irq[INTR_IDX_MAX]; + bool has_intf_te; +}; + +static inline int dpu_encoder_phys_inc_pending(struct dpu_encoder_phys *phys) +{ + atomic_inc_return(&phys->pending_ctlstart_cnt); + return atomic_inc_return(&phys->pending_kickoff_cnt); +} + +/** + * struct dpu_encoder_phys_wb - sub-class of dpu_encoder_phys to handle command + * mode specific operations + * @base: Baseclass physical encoder structure + * @wbirq_refcount: Reference count of writeback interrupt + * @wb_done_timeout_cnt: number of wb done irq timeout errors + * @wb_cfg: writeback block config to store fb related details + * @wb_conn: backpointer to writeback connector + * @wb_job: backpointer to current writeback job + * @dest: dpu buffer layout for current writeback output buffer + */ +struct dpu_encoder_phys_wb { + struct dpu_encoder_phys base; + atomic_t wbirq_refcount; + int wb_done_timeout_cnt; + struct dpu_hw_wb_cfg wb_cfg; + struct drm_writeback_connector *wb_conn; + struct drm_writeback_job *wb_job; + struct dpu_hw_fmt_layout dest; +}; + +/** + * struct dpu_encoder_phys_cmd - sub-class of dpu_encoder_phys to handle command + * mode specific operations + * @base: Baseclass physical encoder structure + * @intf_idx: Intf Block index used by this phys encoder + * @stream_sel: Stream selection for multi-stream interfaces + * @serialize_wait4pp: serialize wait4pp feature waits for pp_done interrupt + * after ctl_start instead of before next frame kickoff + * @pp_timeout_report_cnt: number of pingpong done irq timeout errors + * @pending_vblank_cnt: Atomic counter tracking pending wait for VBLANK + * @pending_vblank_wq: Wait queue for blocking until VBLANK received + */ +struct dpu_encoder_phys_cmd { + struct dpu_encoder_phys base; + int stream_sel; + bool serialize_wait4pp; + int pp_timeout_report_cnt; + atomic_t pending_vblank_cnt; + wait_queue_head_t pending_vblank_wq; +}; + +/** + * struct dpu_enc_phys_init_params - initialization parameters for phys encs + * @dpu_kms: Pointer to the dpu_kms top level + * @parent: Pointer to the containing virtual encoder + * @parent_ops: Callbacks exposed by the parent to the phys_enc + * @split_role: Role to play in a split-panel configuration + * @hw_intf: Hardware interface to the intf registers + * @hw_wb: Hardware interface to the wb registers + * @enc_spinlock: Virtual-Encoder-Wide Spin Lock for IRQ purposes + */ +struct dpu_enc_phys_init_params { + struct dpu_kms *dpu_kms; + struct drm_encoder *parent; + enum dpu_enc_split_role split_role; + struct dpu_hw_intf *hw_intf; + struct dpu_hw_wb *hw_wb; + spinlock_t *enc_spinlock; +}; + +/** + * dpu_encoder_wait_info - container for passing arguments to irq wait functions + * @wq: wait queue structure + * @atomic_cnt: wait until atomic_cnt equals zero + * @timeout_ms: timeout value in milliseconds + */ +struct dpu_encoder_wait_info { + wait_queue_head_t *wq; + atomic_t *atomic_cnt; + s64 timeout_ms; +}; + +struct dpu_encoder_phys *dpu_encoder_phys_vid_init(struct drm_device *dev, + struct dpu_enc_phys_init_params *p); + +struct dpu_encoder_phys *dpu_encoder_phys_cmd_init(struct drm_device *dev, + struct dpu_enc_phys_init_params *p); + +struct dpu_encoder_phys *dpu_encoder_phys_wb_init(struct drm_device *dev, + struct dpu_enc_phys_init_params *p); + +void dpu_encoder_helper_trigger_start(struct dpu_encoder_phys *phys_enc); + +static inline enum dpu_3d_blend_mode dpu_encoder_helper_get_3d_blend_mode( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_crtc_state *dpu_cstate; + + if (!phys_enc || phys_enc->enable_state == DPU_ENC_DISABLING) + return BLEND_3D_NONE; + + dpu_cstate = to_dpu_crtc_state(phys_enc->parent->crtc->state); + + /* Use merge_3d unless DSC MERGE topology is used */ + if (phys_enc->split_role == ENC_ROLE_SOLO && + dpu_cstate->num_mixers == CRTC_DUAL_MIXERS && + !dpu_encoder_use_dsc_merge(phys_enc->parent)) + return BLEND_3D_H_ROW_INT; + + return BLEND_3D_NONE; +} + +unsigned int dpu_encoder_helper_get_cwb_mask(struct dpu_encoder_phys *phys_enc); + +unsigned int dpu_encoder_helper_get_dsc(struct dpu_encoder_phys *phys_enc); + +struct drm_dsc_config *dpu_encoder_get_dsc_config(struct drm_encoder *drm_enc); + +u32 dpu_encoder_get_drm_fmt(struct dpu_encoder_phys *phys_enc); + +bool dpu_encoder_needs_periph_flush(struct dpu_encoder_phys *phys_enc); + +void dpu_encoder_helper_split_config( + struct dpu_encoder_phys *phys_enc, + enum dpu_intf interface); + +void dpu_encoder_helper_report_irq_timeout(struct dpu_encoder_phys *phys_enc, + enum dpu_intr_idx intr_idx); + +int dpu_encoder_helper_wait_for_irq(struct dpu_encoder_phys *phys_enc, + unsigned int irq, + void (*func)(void *arg), + struct dpu_encoder_wait_info *wait_info); + +void dpu_encoder_helper_phys_cleanup(struct dpu_encoder_phys *phys_enc); + +void dpu_encoder_helper_phys_setup_cwb(struct dpu_encoder_phys *phys_enc, + bool enable); + +void dpu_encoder_helper_phys_setup_cdm(struct dpu_encoder_phys *phys_enc, + const struct msm_format *dpu_fmt, + u32 output_type); + +void dpu_encoder_vblank_callback(struct drm_encoder *drm_enc, + struct dpu_encoder_phys *phy_enc); + +void dpu_encoder_underrun_callback(struct drm_encoder *drm_enc, + struct dpu_encoder_phys *phy_enc); + +void dpu_encoder_frame_done_callback( + struct drm_encoder *drm_enc, + struct dpu_encoder_phys *ready_phys, u32 event); + +void dpu_encoder_phys_init(struct dpu_encoder_phys *phys, + struct dpu_enc_phys_init_params *p); + +#endif /* __dpu_encoder_phys_H__ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c new file mode 100644 index 000000000..45079ee59 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c @@ -0,0 +1,784 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2015-2018, 2020-2021 The Linux Foundation. All rights reserved. + */ + +#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ +#include +#include +#include "dpu_encoder_phys.h" +#include "dpu_hw_interrupts.h" +#include "dpu_hw_pingpong.h" +#include "dpu_core_irq.h" +#include "dpu_formats.h" +#include "dpu_trace.h" +#include "disp/msm_disp_snapshot.h" + +#include + +#define DPU_DEBUG_CMDENC(e, fmt, ...) DPU_DEBUG("enc%d intf%d " fmt, \ + (e) && (e)->base.parent ? \ + (e)->base.parent->base.id : -1, \ + (e) ? (e)->base.hw_intf->idx - INTF_0 : -1, ##__VA_ARGS__) + +#define DPU_ERROR_CMDENC(e, fmt, ...) DPU_ERROR("enc%d intf%d " fmt, \ + (e) && (e)->base.parent ? \ + (e)->base.parent->base.id : -1, \ + (e) ? (e)->base.hw_intf->idx - INTF_0 : -1, ##__VA_ARGS__) + +#define to_dpu_encoder_phys_cmd(x) \ + container_of(x, struct dpu_encoder_phys_cmd, base) + +#define PP_TIMEOUT_MAX_TRIALS 10 + +/* + * Tearcheck sync start and continue thresholds are empirically found + * based on common panels In the future, may want to allow panels to override + * these default values + */ +#define DEFAULT_TEARCHECK_SYNC_THRESH_START 4 +#define DEFAULT_TEARCHECK_SYNC_THRESH_CONTINUE 4 + +static void dpu_encoder_phys_cmd_enable_te(struct dpu_encoder_phys *phys_enc); + +static bool dpu_encoder_phys_cmd_is_master(struct dpu_encoder_phys *phys_enc) +{ + return (phys_enc->split_role != ENC_ROLE_SLAVE); +} + +static void _dpu_encoder_phys_cmd_update_intf_cfg( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_encoder_phys_cmd *cmd_enc = + to_dpu_encoder_phys_cmd(phys_enc); + struct dpu_hw_ctl *ctl; + struct dpu_hw_intf_cfg intf_cfg = { 0 }; + struct dpu_hw_intf_cmd_mode_cfg cmd_mode_cfg = {}; + + ctl = phys_enc->hw_ctl; + if (!ctl->ops.setup_intf_cfg) + return; + + intf_cfg.intf = phys_enc->hw_intf->idx; + if (phys_enc->split_role == ENC_ROLE_MASTER) + intf_cfg.intf_master = phys_enc->hw_intf->idx; + intf_cfg.intf_mode_sel = DPU_CTL_MODE_SEL_CMD; + intf_cfg.stream_sel = cmd_enc->stream_sel; + intf_cfg.mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc); + intf_cfg.dsc = dpu_encoder_helper_get_dsc(phys_enc); + ctl->ops.setup_intf_cfg(ctl, &intf_cfg); + + /* setup which pp blk will connect to this intf */ + if (phys_enc->dpu_kms->catalog->mdss_ver->core_major_ver >= 5 && + phys_enc->hw_intf->ops.bind_pingpong_blk) + phys_enc->hw_intf->ops.bind_pingpong_blk( + phys_enc->hw_intf, + phys_enc->hw_pp->idx); + + if (intf_cfg.dsc != 0) + cmd_mode_cfg.data_compress = true; + + cmd_mode_cfg.wide_bus_en = dpu_encoder_is_widebus_enabled(phys_enc->parent); + + if (phys_enc->hw_intf->ops.program_intf_cmd_cfg) + phys_enc->hw_intf->ops.program_intf_cmd_cfg(phys_enc->hw_intf, &cmd_mode_cfg); +} + +static void dpu_encoder_phys_cmd_pp_tx_done_irq(void *arg) +{ + struct dpu_encoder_phys *phys_enc = arg; + unsigned long lock_flags; + int new_cnt; + u32 event = DPU_ENCODER_FRAME_EVENT_DONE; + + if (!phys_enc->hw_pp) + return; + + DPU_ATRACE_BEGIN("pp_done_irq"); + /* notify all synchronous clients first, then asynchronous clients */ + dpu_encoder_frame_done_callback(phys_enc->parent, phys_enc, event); + + spin_lock_irqsave(phys_enc->enc_spinlock, lock_flags); + new_cnt = atomic_add_unless(&phys_enc->pending_kickoff_cnt, -1, 0); + spin_unlock_irqrestore(phys_enc->enc_spinlock, lock_flags); + + trace_dpu_enc_phys_cmd_pp_tx_done(DRMID(phys_enc->parent), + phys_enc->hw_pp->idx - PINGPONG_0, + new_cnt, event); + + /* Signal any waiting atomic commit thread */ + wake_up_all(&phys_enc->pending_kickoff_wq); + DPU_ATRACE_END("pp_done_irq"); +} + +static void dpu_encoder_phys_cmd_te_rd_ptr_irq(void *arg) +{ + struct dpu_encoder_phys *phys_enc = arg; + struct dpu_encoder_phys_cmd *cmd_enc; + + DPU_ATRACE_BEGIN("rd_ptr_irq"); + cmd_enc = to_dpu_encoder_phys_cmd(phys_enc); + + dpu_encoder_vblank_callback(phys_enc->parent, phys_enc); + + atomic_add_unless(&cmd_enc->pending_vblank_cnt, -1, 0); + wake_up_all(&cmd_enc->pending_vblank_wq); + DPU_ATRACE_END("rd_ptr_irq"); +} + +static void dpu_encoder_phys_cmd_ctl_start_irq(void *arg) +{ + struct dpu_encoder_phys *phys_enc = arg; + + DPU_ATRACE_BEGIN("ctl_start_irq"); + + atomic_add_unless(&phys_enc->pending_ctlstart_cnt, -1, 0); + + /* Signal any waiting ctl start interrupt */ + wake_up_all(&phys_enc->pending_kickoff_wq); + DPU_ATRACE_END("ctl_start_irq"); +} + +static void dpu_encoder_phys_cmd_underrun_irq(void *arg) +{ + struct dpu_encoder_phys *phys_enc = arg; + + dpu_encoder_underrun_callback(phys_enc->parent, phys_enc); +} + +static void dpu_encoder_phys_cmd_atomic_mode_set( + struct dpu_encoder_phys *phys_enc, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state) +{ + phys_enc->irq[INTR_IDX_CTL_START] = phys_enc->hw_ctl->caps->intr_start; + + phys_enc->irq[INTR_IDX_PINGPONG] = phys_enc->hw_pp->caps->intr_done; + + if (phys_enc->has_intf_te) + phys_enc->irq[INTR_IDX_RDPTR] = phys_enc->hw_intf->cap->intr_tear_rd_ptr; + else + phys_enc->irq[INTR_IDX_RDPTR] = phys_enc->hw_pp->caps->intr_rdptr; + + phys_enc->irq[INTR_IDX_UNDERRUN] = phys_enc->hw_intf->cap->intr_underrun; +} + +static int _dpu_encoder_phys_cmd_handle_ppdone_timeout( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_encoder_phys_cmd *cmd_enc = + to_dpu_encoder_phys_cmd(phys_enc); + u32 frame_event = DPU_ENCODER_FRAME_EVENT_ERROR; + bool do_log = false; + struct drm_encoder *drm_enc; + + if (!phys_enc->hw_pp) + return -EINVAL; + + drm_enc = phys_enc->parent; + + cmd_enc->pp_timeout_report_cnt++; + if (cmd_enc->pp_timeout_report_cnt == PP_TIMEOUT_MAX_TRIALS) { + frame_event |= DPU_ENCODER_FRAME_EVENT_PANEL_DEAD; + do_log = true; + } else if (cmd_enc->pp_timeout_report_cnt == 1) { + do_log = true; + } + + trace_dpu_enc_phys_cmd_pdone_timeout(DRMID(drm_enc), + phys_enc->hw_pp->idx - PINGPONG_0, + cmd_enc->pp_timeout_report_cnt, + atomic_read(&phys_enc->pending_kickoff_cnt), + frame_event); + + /* to avoid flooding, only log first time, and "dead" time */ + if (do_log) { + DRM_ERROR("id:%d pp:%d kickoff timeout %d cnt %d koff_cnt %d\n", + DRMID(drm_enc), + phys_enc->hw_pp->idx - PINGPONG_0, + phys_enc->hw_ctl->idx - CTL_0, + cmd_enc->pp_timeout_report_cnt, + atomic_read(&phys_enc->pending_kickoff_cnt)); + msm_disp_snapshot_state(drm_enc->dev); + dpu_core_irq_unregister_callback(phys_enc->dpu_kms, + phys_enc->irq[INTR_IDX_RDPTR]); + } + + atomic_add_unless(&phys_enc->pending_kickoff_cnt, -1, 0); + + /* request a ctl reset before the next kickoff */ + phys_enc->enable_state = DPU_ENC_ERR_NEEDS_HW_RESET; + + dpu_encoder_frame_done_callback(phys_enc->parent, phys_enc, frame_event); + + return -ETIMEDOUT; +} + +static int _dpu_encoder_phys_cmd_wait_for_idle( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_encoder_phys_cmd *cmd_enc = + to_dpu_encoder_phys_cmd(phys_enc); + struct dpu_encoder_wait_info wait_info; + int ret; + + wait_info.wq = &phys_enc->pending_kickoff_wq; + wait_info.atomic_cnt = &phys_enc->pending_kickoff_cnt; + wait_info.timeout_ms = KICKOFF_TIMEOUT_MS; + + ret = dpu_encoder_helper_wait_for_irq(phys_enc, + phys_enc->irq[INTR_IDX_PINGPONG], + dpu_encoder_phys_cmd_pp_tx_done_irq, + &wait_info); + if (ret == -ETIMEDOUT) + _dpu_encoder_phys_cmd_handle_ppdone_timeout(phys_enc); + else if (!ret) + cmd_enc->pp_timeout_report_cnt = 0; + + return ret; +} + +static int dpu_encoder_phys_cmd_control_vblank_irq( + struct dpu_encoder_phys *phys_enc, + bool enable) +{ + int ret = 0; + int refcount; + + if (!phys_enc->hw_pp) { + DPU_ERROR("invalid encoder\n"); + return -EINVAL; + } + + mutex_lock(&phys_enc->vblank_ctl_lock); + refcount = phys_enc->vblank_refcount; + + /* Slave encoders don't report vblank */ + if (!dpu_encoder_phys_cmd_is_master(phys_enc)) + goto end; + + /* IRQ not yet initialized */ + if (!phys_enc->irq[INTR_IDX_RDPTR]) { + ret = -EINVAL; + goto end; + } + + /* protect against negative */ + if (!enable && refcount == 0) { + ret = -EINVAL; + goto end; + } + + DRM_DEBUG_KMS("id:%u pp:%d enable=%s/%d\n", DRMID(phys_enc->parent), + phys_enc->hw_pp->idx - PINGPONG_0, + str_true_false(enable), refcount); + + if (enable) { + if (phys_enc->vblank_refcount == 0) + ret = dpu_core_irq_register_callback(phys_enc->dpu_kms, + phys_enc->irq[INTR_IDX_RDPTR], + dpu_encoder_phys_cmd_te_rd_ptr_irq, + phys_enc); + if (!ret) + phys_enc->vblank_refcount++; + } else if (!enable) { + if (phys_enc->vblank_refcount == 1) + ret = dpu_core_irq_unregister_callback(phys_enc->dpu_kms, + phys_enc->irq[INTR_IDX_RDPTR]); + if (!ret) + phys_enc->vblank_refcount--; + } + +end: + mutex_unlock(&phys_enc->vblank_ctl_lock); + if (ret) { + DRM_ERROR("vblank irq err id:%u pp:%d ret:%d, enable %s/%d\n", + DRMID(phys_enc->parent), + phys_enc->hw_pp->idx - PINGPONG_0, ret, + str_true_false(enable), refcount); + } + + return ret; +} + +static void dpu_encoder_phys_cmd_irq_enable(struct dpu_encoder_phys *phys_enc) +{ + trace_dpu_enc_phys_cmd_irq_enable(DRMID(phys_enc->parent), + phys_enc->hw_pp->idx - PINGPONG_0, + phys_enc->vblank_refcount); + + dpu_core_irq_register_callback(phys_enc->dpu_kms, + phys_enc->irq[INTR_IDX_PINGPONG], + dpu_encoder_phys_cmd_pp_tx_done_irq, + phys_enc); + dpu_core_irq_register_callback(phys_enc->dpu_kms, + phys_enc->irq[INTR_IDX_UNDERRUN], + dpu_encoder_phys_cmd_underrun_irq, + phys_enc); + dpu_encoder_phys_cmd_control_vblank_irq(phys_enc, true); + + if (dpu_encoder_phys_cmd_is_master(phys_enc) && phys_enc->irq[INTR_IDX_CTL_START]) + dpu_core_irq_register_callback(phys_enc->dpu_kms, + phys_enc->irq[INTR_IDX_CTL_START], + dpu_encoder_phys_cmd_ctl_start_irq, + phys_enc); +} + +static void dpu_encoder_phys_cmd_irq_disable(struct dpu_encoder_phys *phys_enc) +{ + trace_dpu_enc_phys_cmd_irq_disable(DRMID(phys_enc->parent), + phys_enc->hw_pp->idx - PINGPONG_0, + phys_enc->vblank_refcount); + + if (dpu_encoder_phys_cmd_is_master(phys_enc) && phys_enc->irq[INTR_IDX_CTL_START]) + dpu_core_irq_unregister_callback(phys_enc->dpu_kms, + phys_enc->irq[INTR_IDX_CTL_START]); + + dpu_core_irq_unregister_callback(phys_enc->dpu_kms, phys_enc->irq[INTR_IDX_UNDERRUN]); + dpu_encoder_phys_cmd_control_vblank_irq(phys_enc, false); + dpu_core_irq_unregister_callback(phys_enc->dpu_kms, phys_enc->irq[INTR_IDX_PINGPONG]); +} + +static void dpu_encoder_phys_cmd_tearcheck_config( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_encoder_phys_cmd *cmd_enc = + to_dpu_encoder_phys_cmd(phys_enc); + struct dpu_hw_tear_check tc_cfg = { 0 }; + struct drm_display_mode *mode; + bool tc_enable = true; + unsigned long vsync_hz; + struct dpu_kms *dpu_kms; + + /* + * TODO: if/when resource allocation is refactored, move this to a + * place where the driver can actually return an error. + */ + if (!phys_enc->has_intf_te && + (!phys_enc->hw_pp || + !phys_enc->hw_pp->ops.enable_tearcheck)) { + DPU_DEBUG_CMDENC(cmd_enc, "tearcheck not supported\n"); + return; + } + + DPU_DEBUG_CMDENC(cmd_enc, "intf %d pp %d\n", + phys_enc->hw_intf ? phys_enc->hw_intf->idx - INTF_0 : -1, + phys_enc->hw_pp ? phys_enc->hw_pp->idx - PINGPONG_0 : -1); + + mode = &phys_enc->cached_mode; + + dpu_kms = phys_enc->dpu_kms; + + /* + * TE default: dsi byte clock calculated base on 70 fps; + * around 14 ms to complete a kickoff cycle if te disabled; + * vclk_line base on 60 fps; write is faster than read; + * init == start == rdptr; + * + * vsync_count is ratio of MDP VSYNC clock frequency to LCD panel + * frequency divided by the no. of rows (lines) in the LCDpanel. + */ + vsync_hz = dpu_kms_get_clk_rate(dpu_kms, "vsync"); + if (!vsync_hz) { + DPU_DEBUG_CMDENC(cmd_enc, "invalid - no vsync clock\n"); + return; + } + + tc_cfg.vsync_count = vsync_hz / + (mode->vtotal * drm_mode_vrefresh(mode)); + + /* + * Set the sync_cfg_height to twice vtotal so that if we lose a + * TE event coming from the display TE pin we won't stall immediately + */ + tc_cfg.hw_vsync_mode = 1; + tc_cfg.sync_cfg_height = mode->vtotal * 2; + tc_cfg.vsync_init_val = mode->vdisplay; + tc_cfg.sync_threshold_start = DEFAULT_TEARCHECK_SYNC_THRESH_START; + tc_cfg.sync_threshold_continue = DEFAULT_TEARCHECK_SYNC_THRESH_CONTINUE; + tc_cfg.start_pos = mode->vdisplay; + tc_cfg.rd_ptr_irq = mode->vdisplay + 1; + + DPU_DEBUG_CMDENC(cmd_enc, + "tc vsync_clk_speed_hz %lu vtotal %u vrefresh %u\n", + vsync_hz, mode->vtotal, drm_mode_vrefresh(mode)); + DPU_DEBUG_CMDENC(cmd_enc, + "tc enable %u start_pos %u rd_ptr_irq %u\n", + tc_enable, tc_cfg.start_pos, tc_cfg.rd_ptr_irq); + DPU_DEBUG_CMDENC(cmd_enc, + "tc hw_vsync_mode %u vsync_count %u vsync_init_val %u\n", + tc_cfg.hw_vsync_mode, tc_cfg.vsync_count, + tc_cfg.vsync_init_val); + DPU_DEBUG_CMDENC(cmd_enc, + "tc cfgheight %u thresh_start %u thresh_cont %u\n", + tc_cfg.sync_cfg_height, tc_cfg.sync_threshold_start, + tc_cfg.sync_threshold_continue); + + if (phys_enc->has_intf_te) + phys_enc->hw_intf->ops.enable_tearcheck(phys_enc->hw_intf, &tc_cfg); + else + phys_enc->hw_pp->ops.enable_tearcheck(phys_enc->hw_pp, &tc_cfg); +} + +static void _dpu_encoder_phys_cmd_pingpong_config( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_encoder_phys_cmd *cmd_enc = + to_dpu_encoder_phys_cmd(phys_enc); + + if (!phys_enc->hw_pp || !phys_enc->hw_ctl->ops.setup_intf_cfg) { + DPU_ERROR("invalid arg(s), enc %d\n", phys_enc != NULL); + return; + } + + DPU_DEBUG_CMDENC(cmd_enc, "pp %d, enabling mode:\n", + phys_enc->hw_pp->idx - PINGPONG_0); + drm_mode_debug_printmodeline(&phys_enc->cached_mode); + + _dpu_encoder_phys_cmd_update_intf_cfg(phys_enc); + dpu_encoder_phys_cmd_tearcheck_config(phys_enc); +} + +static bool dpu_encoder_phys_cmd_needs_single_flush( + struct dpu_encoder_phys *phys_enc) +{ + /** + * we do separate flush for each CTL and let + * CTL_START synchronize them + */ + return false; +} + +static void dpu_encoder_phys_cmd_enable_helper( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_ctl *ctl; + + if (!phys_enc->hw_pp) { + DPU_ERROR("invalid arg(s), encoder %d\n", phys_enc != NULL); + return; + } + + dpu_encoder_helper_split_config(phys_enc, phys_enc->hw_intf->idx); + + _dpu_encoder_phys_cmd_pingpong_config(phys_enc); + + ctl = phys_enc->hw_ctl; + ctl->ops.update_pending_flush_intf(ctl, phys_enc->hw_intf->idx); +} + +static void dpu_encoder_phys_cmd_enable(struct dpu_encoder_phys *phys_enc) +{ + struct dpu_encoder_phys_cmd *cmd_enc = + to_dpu_encoder_phys_cmd(phys_enc); + + if (!phys_enc->hw_pp) { + DPU_ERROR("invalid phys encoder\n"); + return; + } + + DPU_DEBUG_CMDENC(cmd_enc, "pp %d\n", phys_enc->hw_pp->idx - PINGPONG_0); + + if (phys_enc->enable_state == DPU_ENC_ENABLED) { + DPU_ERROR("already enabled\n"); + return; + } + + dpu_encoder_phys_cmd_enable_helper(phys_enc); + phys_enc->enable_state = DPU_ENC_ENABLED; +} + +static void _dpu_encoder_phys_cmd_connect_te( + struct dpu_encoder_phys *phys_enc, bool enable) +{ + if (phys_enc->has_intf_te) { + if (!phys_enc->hw_intf || !phys_enc->hw_intf->ops.connect_external_te) + return; + + trace_dpu_enc_phys_cmd_connect_te(DRMID(phys_enc->parent), enable); + phys_enc->hw_intf->ops.connect_external_te(phys_enc->hw_intf, enable); + } else { + if (!phys_enc->hw_pp || !phys_enc->hw_pp->ops.connect_external_te) + return; + + trace_dpu_enc_phys_cmd_connect_te(DRMID(phys_enc->parent), enable); + phys_enc->hw_pp->ops.connect_external_te(phys_enc->hw_pp, enable); + } +} + +static void dpu_encoder_phys_cmd_prepare_idle_pc( + struct dpu_encoder_phys *phys_enc) +{ + _dpu_encoder_phys_cmd_connect_te(phys_enc, false); +} + +static int dpu_encoder_phys_cmd_get_line_count( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_pingpong *hw_pp; + struct dpu_hw_intf *hw_intf; + + if (!dpu_encoder_phys_cmd_is_master(phys_enc)) + return -EINVAL; + + if (phys_enc->has_intf_te) { + hw_intf = phys_enc->hw_intf; + if (!hw_intf || !hw_intf->ops.get_line_count) + return -EINVAL; + return hw_intf->ops.get_line_count(hw_intf); + } + + hw_pp = phys_enc->hw_pp; + if (!hw_pp || !hw_pp->ops.get_line_count) + return -EINVAL; + return hw_pp->ops.get_line_count(hw_pp); +} + +static void dpu_encoder_phys_cmd_disable(struct dpu_encoder_phys *phys_enc) +{ + struct dpu_encoder_phys_cmd *cmd_enc = + to_dpu_encoder_phys_cmd(phys_enc); + struct dpu_hw_ctl *ctl; + + if (phys_enc->enable_state == DPU_ENC_DISABLED) { + DPU_ERROR_CMDENC(cmd_enc, "already disabled\n"); + return; + } + + if (phys_enc->has_intf_te) { + DRM_DEBUG_KMS("id:%u intf:%d state:%d\n", DRMID(phys_enc->parent), + phys_enc->hw_intf->idx - INTF_0, + phys_enc->enable_state); + + if (phys_enc->hw_intf->ops.disable_tearcheck) + phys_enc->hw_intf->ops.disable_tearcheck(phys_enc->hw_intf); + } else { + if (!phys_enc->hw_pp) { + DPU_ERROR("invalid encoder\n"); + return; + } + + DRM_DEBUG_KMS("id:%u pp:%d state:%d\n", DRMID(phys_enc->parent), + phys_enc->hw_pp->idx - PINGPONG_0, + phys_enc->enable_state); + + if (phys_enc->hw_pp->ops.disable_tearcheck) + phys_enc->hw_pp->ops.disable_tearcheck(phys_enc->hw_pp); + } + + if (phys_enc->hw_intf->ops.bind_pingpong_blk) { + phys_enc->hw_intf->ops.bind_pingpong_blk( + phys_enc->hw_intf, + PINGPONG_NONE); + + ctl = phys_enc->hw_ctl; + ctl->ops.update_pending_flush_intf(ctl, phys_enc->hw_intf->idx); + } + + phys_enc->enable_state = DPU_ENC_DISABLED; +} + +static void dpu_encoder_phys_cmd_prepare_for_kickoff( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_encoder_phys_cmd *cmd_enc = + to_dpu_encoder_phys_cmd(phys_enc); + int ret; + + if (!phys_enc->hw_pp) { + DPU_ERROR("invalid encoder\n"); + return; + } + DRM_DEBUG_KMS("id:%u pp:%d pending_cnt:%d\n", DRMID(phys_enc->parent), + phys_enc->hw_pp->idx - PINGPONG_0, + atomic_read(&phys_enc->pending_kickoff_cnt)); + + /* + * Mark kickoff request as outstanding. If there are more than one, + * outstanding, then we have to wait for the previous one to complete + */ + ret = _dpu_encoder_phys_cmd_wait_for_idle(phys_enc); + if (ret) { + /* force pending_kickoff_cnt 0 to discard failed kickoff */ + atomic_set(&phys_enc->pending_kickoff_cnt, 0); + DRM_ERROR("failed wait_for_idle: id:%u ret:%d pp:%d\n", + DRMID(phys_enc->parent), ret, + phys_enc->hw_pp->idx - PINGPONG_0); + } + + dpu_encoder_phys_cmd_enable_te(phys_enc); + + DPU_DEBUG_CMDENC(cmd_enc, "pp:%d pending_cnt %d\n", + phys_enc->hw_pp->idx - PINGPONG_0, + atomic_read(&phys_enc->pending_kickoff_cnt)); +} + +static void dpu_encoder_phys_cmd_enable_te(struct dpu_encoder_phys *phys_enc) +{ + if (!phys_enc) + return; + if (!dpu_encoder_phys_cmd_is_master(phys_enc)) + return; + + if (phys_enc->has_intf_te) { + if (!phys_enc->hw_intf->ops.disable_autorefresh) + return; + + phys_enc->hw_intf->ops.disable_autorefresh( + phys_enc->hw_intf, + DRMID(phys_enc->parent), + phys_enc->cached_mode.vdisplay); + } else { + if (!phys_enc->hw_pp || + !phys_enc->hw_pp->ops.disable_autorefresh) + return; + + phys_enc->hw_pp->ops.disable_autorefresh( + phys_enc->hw_pp, + DRMID(phys_enc->parent), + phys_enc->cached_mode.vdisplay); + } +} + +static int _dpu_encoder_phys_cmd_wait_for_ctl_start( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_encoder_phys_cmd *cmd_enc = + to_dpu_encoder_phys_cmd(phys_enc); + struct dpu_encoder_wait_info wait_info; + int ret; + + wait_info.wq = &phys_enc->pending_kickoff_wq; + wait_info.atomic_cnt = &phys_enc->pending_ctlstart_cnt; + wait_info.timeout_ms = KICKOFF_TIMEOUT_MS; + + ret = dpu_encoder_helper_wait_for_irq(phys_enc, + phys_enc->irq[INTR_IDX_CTL_START], + dpu_encoder_phys_cmd_ctl_start_irq, + &wait_info); + if (ret == -ETIMEDOUT) { + DPU_ERROR_CMDENC(cmd_enc, "ctl start interrupt wait failed\n"); + ret = -EINVAL; + } else if (!ret) + ret = 0; + + return ret; +} + +static int dpu_encoder_phys_cmd_wait_for_tx_complete( + struct dpu_encoder_phys *phys_enc) +{ + int rc; + + rc = _dpu_encoder_phys_cmd_wait_for_idle(phys_enc); + if (rc) { + DRM_ERROR("failed wait_for_idle: id:%u ret:%d intf:%d\n", + DRMID(phys_enc->parent), rc, + phys_enc->hw_intf->idx - INTF_0); + } + + return rc; +} + +static int dpu_encoder_phys_cmd_wait_for_commit_done( + struct dpu_encoder_phys *phys_enc) +{ + /* only required for master controller */ + if (!dpu_encoder_phys_cmd_is_master(phys_enc)) + return 0; + + if (phys_enc->irq[INTR_IDX_CTL_START] && + !phys_enc->hw_ctl->ops.is_started(phys_enc->hw_ctl)) + return _dpu_encoder_phys_cmd_wait_for_ctl_start(phys_enc); + + return dpu_encoder_phys_cmd_wait_for_tx_complete(phys_enc); +} + +static void dpu_encoder_phys_cmd_handle_post_kickoff( + struct dpu_encoder_phys *phys_enc) +{ + /** + * re-enable external TE, either for the first time after enabling + * or if disabled for Autorefresh + */ + _dpu_encoder_phys_cmd_connect_te(phys_enc, true); +} + +static void dpu_encoder_phys_cmd_trigger_start( + struct dpu_encoder_phys *phys_enc) +{ + dpu_encoder_helper_trigger_start(phys_enc); +} + +static void dpu_encoder_phys_cmd_init_ops( + struct dpu_encoder_phys_ops *ops) +{ + ops->is_master = dpu_encoder_phys_cmd_is_master; + ops->atomic_mode_set = dpu_encoder_phys_cmd_atomic_mode_set; + ops->enable = dpu_encoder_phys_cmd_enable; + ops->disable = dpu_encoder_phys_cmd_disable; + ops->control_vblank_irq = dpu_encoder_phys_cmd_control_vblank_irq; + ops->wait_for_commit_done = dpu_encoder_phys_cmd_wait_for_commit_done; + ops->prepare_for_kickoff = dpu_encoder_phys_cmd_prepare_for_kickoff; + ops->wait_for_tx_complete = dpu_encoder_phys_cmd_wait_for_tx_complete; + ops->trigger_start = dpu_encoder_phys_cmd_trigger_start; + ops->needs_single_flush = dpu_encoder_phys_cmd_needs_single_flush; + ops->irq_enable = dpu_encoder_phys_cmd_irq_enable; + ops->irq_disable = dpu_encoder_phys_cmd_irq_disable; + ops->restore = dpu_encoder_phys_cmd_enable_helper; + ops->prepare_idle_pc = dpu_encoder_phys_cmd_prepare_idle_pc; + ops->handle_post_kickoff = dpu_encoder_phys_cmd_handle_post_kickoff; + ops->get_line_count = dpu_encoder_phys_cmd_get_line_count; +} + +/** + * dpu_encoder_phys_cmd_init - Construct a new command mode physical encoder + * @dev: Corresponding device for devres management + * @p: Pointer to init params structure + * Return: Error code or newly allocated encoder + */ +struct dpu_encoder_phys *dpu_encoder_phys_cmd_init(struct drm_device *dev, + struct dpu_enc_phys_init_params *p) +{ + struct dpu_encoder_phys *phys_enc = NULL; + struct dpu_encoder_phys_cmd *cmd_enc = NULL; + + DPU_DEBUG("intf\n"); + + cmd_enc = drmm_kzalloc(dev, sizeof(*cmd_enc), GFP_KERNEL); + if (!cmd_enc) { + DPU_ERROR("failed to allocate\n"); + return ERR_PTR(-ENOMEM); + } + phys_enc = &cmd_enc->base; + + dpu_encoder_phys_init(phys_enc, p); + + mutex_init(&phys_enc->vblank_ctl_lock); + phys_enc->vblank_refcount = 0; + + dpu_encoder_phys_cmd_init_ops(&phys_enc->ops); + phys_enc->intf_mode = INTF_MODE_CMD; + cmd_enc->stream_sel = 0; + + if (!phys_enc->hw_intf) { + DPU_ERROR_CMDENC(cmd_enc, "no INTF provided\n"); + return ERR_PTR(-EINVAL); + } + + /* DPU before 5.0 use PINGPONG for TE handling */ + if (phys_enc->dpu_kms->catalog->mdss_ver->core_major_ver >= 5) + phys_enc->has_intf_te = true; + + if (phys_enc->has_intf_te && !phys_enc->hw_intf->ops.enable_tearcheck) { + DPU_ERROR_CMDENC(cmd_enc, "tearcheck not supported\n"); + return ERR_PTR(-EINVAL); + } + + atomic_set(&cmd_enc->pending_vblank_cnt, 0); + init_waitqueue_head(&cmd_enc->pending_vblank_wq); + + DPU_DEBUG_CMDENC(cmd_enc, "created\n"); + + return phys_enc; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c new file mode 100644 index 000000000..ba810f26e --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c @@ -0,0 +1,791 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2015-2018, 2020-2021 The Linux Foundation. All rights reserved. + */ + +#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ +#include "dpu_encoder_phys.h" +#include "dpu_hw_interrupts.h" +#include "dpu_hw_merge3d.h" +#include "dpu_core_irq.h" +#include "dpu_formats.h" +#include "dpu_trace.h" +#include "disp/msm_disp_snapshot.h" +#include "msm_dsc_helper.h" + +#include +#include + +#define DPU_DEBUG_VIDENC(e, fmt, ...) DPU_DEBUG("enc%d intf%d " fmt, \ + (e) && (e)->parent ? \ + (e)->parent->base.id : -1, \ + (e) && (e)->hw_intf ? \ + (e)->hw_intf->idx - INTF_0 : -1, ##__VA_ARGS__) + +#define DPU_ERROR_VIDENC(e, fmt, ...) DPU_ERROR("enc%d intf%d " fmt, \ + (e) && (e)->parent ? \ + (e)->parent->base.id : -1, \ + (e) && (e)->hw_intf ? \ + (e)->hw_intf->idx - INTF_0 : -1, ##__VA_ARGS__) + +#define to_dpu_encoder_phys_vid(x) \ + container_of(x, struct dpu_encoder_phys_vid, base) + +static bool dpu_encoder_phys_vid_is_master( + struct dpu_encoder_phys *phys_enc) +{ + bool ret = false; + + if (phys_enc->split_role != ENC_ROLE_SLAVE) + ret = true; + + return ret; +} + +static void drm_mode_to_intf_timing_params( + const struct dpu_encoder_phys *phys_enc, + const struct drm_display_mode *mode, + struct dpu_hw_intf_timing_params *timing) +{ + memset(timing, 0, sizeof(*timing)); + + if ((mode->htotal < mode->hsync_end) + || (mode->hsync_start < mode->hdisplay) + || (mode->vtotal < mode->vsync_end) + || (mode->vsync_start < mode->vdisplay) + || (mode->hsync_end < mode->hsync_start) + || (mode->vsync_end < mode->vsync_start)) { + DPU_ERROR( + "invalid params - hstart:%d,hend:%d,htot:%d,hdisplay:%d\n", + mode->hsync_start, mode->hsync_end, + mode->htotal, mode->hdisplay); + DPU_ERROR("vstart:%d,vend:%d,vtot:%d,vdisplay:%d\n", + mode->vsync_start, mode->vsync_end, + mode->vtotal, mode->vdisplay); + return; + } + + /* + * https://www.kernel.org/doc/htmldocs/drm/ch02s05.html + * Active Region Front Porch Sync Back Porch + * <-----------------><------------><-----><-----------> + * <- [hv]display ---> + * <--------- [hv]sync_start ------> + * <----------------- [hv]sync_end -------> + * <---------------------------- [hv]total -------------> + */ + timing->width = mode->hdisplay; /* active width */ + timing->height = mode->vdisplay; /* active height */ + timing->xres = timing->width; + timing->yres = timing->height; + timing->h_back_porch = mode->htotal - mode->hsync_end; + timing->h_front_porch = mode->hsync_start - mode->hdisplay; + timing->v_back_porch = mode->vtotal - mode->vsync_end; + timing->v_front_porch = mode->vsync_start - mode->vdisplay; + timing->hsync_pulse_width = mode->hsync_end - mode->hsync_start; + timing->vsync_pulse_width = mode->vsync_end - mode->vsync_start; + timing->hsync_polarity = (mode->flags & DRM_MODE_FLAG_NHSYNC) ? 1 : 0; + timing->vsync_polarity = (mode->flags & DRM_MODE_FLAG_NVSYNC) ? 1 : 0; + timing->border_clr = 0; + timing->underflow_clr = 0xff; + timing->hsync_skew = mode->hskew; + + /* DSI controller cannot handle active-low sync signals. */ + if (phys_enc->hw_intf->cap->type == INTF_DSI) { + timing->hsync_polarity = 0; + timing->vsync_polarity = 0; + } + + timing->wide_bus_en = dpu_encoder_is_widebus_enabled(phys_enc->parent); + timing->compression_en = dpu_encoder_is_dsc_enabled(phys_enc->parent); + + /* + * For DP/EDP, Shift timings to align it to bottom right. + * wide_bus_en is set for everything excluding SDM845 & + * porch changes cause DisplayPort failure and HDMI tearing. + */ + if (phys_enc->hw_intf->cap->type == INTF_DP && timing->wide_bus_en) { + timing->h_back_porch += timing->h_front_porch; + timing->h_front_porch = 0; + timing->v_back_porch += timing->v_front_porch; + timing->v_front_porch = 0; + } + + /* + * for DP, divide the horizonal parameters by 2 when + * widebus is enabled + */ + if (phys_enc->hw_intf->cap->type == INTF_DP && timing->wide_bus_en) { + timing->width = timing->width >> 1; + timing->xres = timing->xres >> 1; + timing->h_back_porch = timing->h_back_porch >> 1; + timing->h_front_porch = timing->h_front_porch >> 1; + timing->hsync_pulse_width = timing->hsync_pulse_width >> 1; + } + + /* + * for DSI, if compression is enabled, then divide the horizonal active + * timing parameters by compression ratio. bits of 3 components(R/G/B) + * is compressed into bits of 1 pixel. + */ + if (phys_enc->hw_intf->cap->type != INTF_DP && timing->compression_en) { + struct drm_dsc_config *dsc = + dpu_encoder_get_dsc_config(phys_enc->parent); + /* + * TODO: replace drm_dsc_get_bpp_int with logic to handle + * fractional part if there is fraction + */ + timing->width = timing->width * drm_dsc_get_bpp_int(dsc) / + (dsc->bits_per_component * 3); + timing->xres = timing->width; + timing->dce_bytes_per_line = msm_dsc_get_bytes_per_line(dsc); + } +} + +static u32 get_horizontal_total(const struct dpu_hw_intf_timing_params *timing) +{ + u32 active = timing->xres; + u32 inactive = + timing->h_back_porch + timing->h_front_porch + + timing->hsync_pulse_width; + return active + inactive; +} + +static u32 get_vertical_total(const struct dpu_hw_intf_timing_params *timing) +{ + u32 active = timing->yres; + u32 inactive = + timing->v_back_porch + timing->v_front_porch + + timing->vsync_pulse_width; + return active + inactive; +} + +/* + * programmable_fetch_get_num_lines: + * Number of fetch lines in vertical front porch + * @timing: Pointer to the intf timing information for the requested mode + * + * Returns the number of fetch lines in vertical front porch at which mdp + * can start fetching the next frame. + * + * Number of needed prefetch lines is anything that cannot be absorbed in the + * start of frame time (back porch + vsync pulse width). + * + * Some panels have very large VFP, however we only need a total number of + * lines based on the chip worst case latencies. + */ +static u32 programmable_fetch_get_num_lines( + struct dpu_encoder_phys *phys_enc, + const struct dpu_hw_intf_timing_params *timing) +{ + u32 worst_case_needed_lines = + phys_enc->hw_intf->cap->prog_fetch_lines_worst_case; + u32 start_of_frame_lines = + timing->v_back_porch + timing->vsync_pulse_width; + u32 needed_vfp_lines = worst_case_needed_lines - start_of_frame_lines; + u32 actual_vfp_lines = 0; + + /* Fetch must be outside active lines, otherwise undefined. */ + if (start_of_frame_lines >= worst_case_needed_lines) { + DPU_DEBUG_VIDENC(phys_enc, + "prog fetch is not needed, large vbp+vsw\n"); + actual_vfp_lines = 0; + } else if (timing->v_front_porch < needed_vfp_lines) { + /* Warn fetch needed, but not enough porch in panel config */ + pr_warn_once + ("low vbp+vfp may lead to perf issues in some cases\n"); + DPU_DEBUG_VIDENC(phys_enc, + "less vfp than fetch req, using entire vfp\n"); + actual_vfp_lines = timing->v_front_porch; + } else { + DPU_DEBUG_VIDENC(phys_enc, "room in vfp for needed prefetch\n"); + actual_vfp_lines = needed_vfp_lines; + } + + DPU_DEBUG_VIDENC(phys_enc, + "v_front_porch %u v_back_porch %u vsync_pulse_width %u\n", + timing->v_front_porch, timing->v_back_porch, + timing->vsync_pulse_width); + DPU_DEBUG_VIDENC(phys_enc, + "wc_lines %u needed_vfp_lines %u actual_vfp_lines %u\n", + worst_case_needed_lines, needed_vfp_lines, actual_vfp_lines); + + return actual_vfp_lines; +} + +/* + * programmable_fetch_config: Programs HW to prefetch lines by offsetting + * the start of fetch into the vertical front porch for cases where the + * vsync pulse width and vertical back porch time is insufficient + * + * Gets # of lines to pre-fetch, then calculate VSYNC counter value. + * HW layer requires VSYNC counter of first pixel of tgt VFP line. + * + * @timing: Pointer to the intf timing information for the requested mode + */ +static void programmable_fetch_config(struct dpu_encoder_phys *phys_enc, + const struct dpu_hw_intf_timing_params *timing) +{ + struct dpu_hw_intf_prog_fetch f = { 0 }; + u32 vfp_fetch_lines = 0; + u32 horiz_total = 0; + u32 vert_total = 0; + u32 vfp_fetch_start_vsync_counter = 0; + unsigned long lock_flags; + + if (WARN_ON_ONCE(!phys_enc->hw_intf->ops.setup_prg_fetch)) + return; + + vfp_fetch_lines = programmable_fetch_get_num_lines(phys_enc, timing); + if (vfp_fetch_lines) { + vert_total = get_vertical_total(timing); + horiz_total = get_horizontal_total(timing); + vfp_fetch_start_vsync_counter = + (vert_total - vfp_fetch_lines) * horiz_total + 1; + f.enable = 1; + f.fetch_start = vfp_fetch_start_vsync_counter; + } + + DPU_DEBUG_VIDENC(phys_enc, + "vfp_fetch_lines %u vfp_fetch_start_vsync_counter %u\n", + vfp_fetch_lines, vfp_fetch_start_vsync_counter); + + spin_lock_irqsave(phys_enc->enc_spinlock, lock_flags); + phys_enc->hw_intf->ops.setup_prg_fetch(phys_enc->hw_intf, &f); + spin_unlock_irqrestore(phys_enc->enc_spinlock, lock_flags); +} + +static void dpu_encoder_phys_vid_setup_timing_engine( + struct dpu_encoder_phys *phys_enc) +{ + struct drm_display_mode mode; + struct dpu_hw_intf_timing_params timing_params = { 0 }; + const struct msm_format *fmt = NULL; + u32 fmt_fourcc; + unsigned long lock_flags; + struct dpu_hw_intf_cfg intf_cfg = { 0 }; + + drm_mode_init(&mode, &phys_enc->cached_mode); + + if (!phys_enc->hw_ctl->ops.setup_intf_cfg) { + DPU_ERROR("invalid encoder %d\n", phys_enc != NULL); + return; + } + + if (!phys_enc->hw_intf->ops.setup_timing_gen) { + DPU_ERROR("timing engine setup is not supported\n"); + return; + } + + DPU_DEBUG_VIDENC(phys_enc, "enabling mode:\n"); + drm_mode_debug_printmodeline(&mode); + + fmt_fourcc = dpu_encoder_get_drm_fmt(phys_enc); + + if (phys_enc->split_role != ENC_ROLE_SOLO || fmt_fourcc == DRM_FORMAT_YUV420) { + mode.hdisplay >>= 1; + mode.htotal >>= 1; + mode.hsync_start >>= 1; + mode.hsync_end >>= 1; + mode.hskew >>= 1; + + DPU_DEBUG_VIDENC(phys_enc, + "split_role %d, halve horizontal %d %d %d %d %d\n", + phys_enc->split_role, + mode.hdisplay, mode.htotal, + mode.hsync_start, mode.hsync_end, + mode.hskew); + } + + drm_mode_to_intf_timing_params(phys_enc, &mode, &timing_params); + + fmt = mdp_get_format(&phys_enc->dpu_kms->base, fmt_fourcc, 0); + DPU_DEBUG_VIDENC(phys_enc, "fmt_fourcc 0x%X\n", fmt_fourcc); + + if (phys_enc->hw_cdm) + intf_cfg.cdm = phys_enc->hw_cdm->idx; + intf_cfg.intf = phys_enc->hw_intf->idx; + if (phys_enc->split_role == ENC_ROLE_MASTER) + intf_cfg.intf_master = phys_enc->hw_intf->idx; + intf_cfg.intf_mode_sel = DPU_CTL_MODE_SEL_VID; + intf_cfg.stream_sel = 0; /* Don't care value for video mode */ + intf_cfg.mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc); + intf_cfg.dsc = dpu_encoder_helper_get_dsc(phys_enc); + if (intf_cfg.mode_3d && phys_enc->hw_pp->merge_3d) + intf_cfg.merge_3d = phys_enc->hw_pp->merge_3d->idx; + + spin_lock_irqsave(phys_enc->enc_spinlock, lock_flags); + phys_enc->hw_intf->ops.setup_timing_gen(phys_enc->hw_intf, + &timing_params, fmt); + phys_enc->hw_ctl->ops.setup_intf_cfg(phys_enc->hw_ctl, &intf_cfg); + + /* setup which pp blk will connect to this intf */ + if (phys_enc->hw_intf->ops.bind_pingpong_blk) + phys_enc->hw_intf->ops.bind_pingpong_blk( + phys_enc->hw_intf, + phys_enc->hw_pp->idx); + + if (phys_enc->hw_pp->merge_3d) + phys_enc->hw_pp->merge_3d->ops.setup_3d_mode(phys_enc->hw_pp->merge_3d, intf_cfg.mode_3d); + + spin_unlock_irqrestore(phys_enc->enc_spinlock, lock_flags); + + programmable_fetch_config(phys_enc, &timing_params); +} + +static void dpu_encoder_phys_vid_vblank_irq(void *arg) +{ + struct dpu_encoder_phys *phys_enc = arg; + struct dpu_hw_ctl *hw_ctl; + unsigned long lock_flags; + u32 flush_register = 0; + + hw_ctl = phys_enc->hw_ctl; + + DPU_ATRACE_BEGIN("vblank_irq"); + + dpu_encoder_vblank_callback(phys_enc->parent, phys_enc); + + atomic_read(&phys_enc->pending_kickoff_cnt); + + /* + * only decrement the pending flush count if we've actually flushed + * hardware. due to sw irq latency, vblank may have already happened + * so we need to double-check with hw that it accepted the flush bits + */ + spin_lock_irqsave(phys_enc->enc_spinlock, lock_flags); + if (hw_ctl->ops.get_flush_register) + flush_register = hw_ctl->ops.get_flush_register(hw_ctl); + + if (!(flush_register & hw_ctl->ops.get_pending_flush(hw_ctl))) + atomic_add_unless(&phys_enc->pending_kickoff_cnt, -1, 0); + spin_unlock_irqrestore(phys_enc->enc_spinlock, lock_flags); + + /* Signal any waiting atomic commit thread */ + wake_up_all(&phys_enc->pending_kickoff_wq); + + dpu_encoder_frame_done_callback(phys_enc->parent, phys_enc, + DPU_ENCODER_FRAME_EVENT_DONE); + + DPU_ATRACE_END("vblank_irq"); +} + +static void dpu_encoder_phys_vid_underrun_irq(void *arg) +{ + struct dpu_encoder_phys *phys_enc = arg; + + dpu_encoder_underrun_callback(phys_enc->parent, phys_enc); +} + +static bool dpu_encoder_phys_vid_needs_single_flush( + struct dpu_encoder_phys *phys_enc) +{ + return !(phys_enc->dpu_kms->catalog->mdss_ver->core_major_ver >= 5) && + phys_enc->split_role != ENC_ROLE_SOLO; +} + +static void dpu_encoder_phys_vid_atomic_mode_set( + struct dpu_encoder_phys *phys_enc, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state) +{ + phys_enc->irq[INTR_IDX_VSYNC] = phys_enc->hw_intf->cap->intr_vsync; + + phys_enc->irq[INTR_IDX_UNDERRUN] = phys_enc->hw_intf->cap->intr_underrun; +} + +static int dpu_encoder_phys_vid_control_vblank_irq( + struct dpu_encoder_phys *phys_enc, + bool enable) +{ + int ret = 0; + int refcount; + + mutex_lock(&phys_enc->vblank_ctl_lock); + refcount = phys_enc->vblank_refcount; + + /* Slave encoders don't report vblank */ + if (!dpu_encoder_phys_vid_is_master(phys_enc)) + goto end; + + /* protect against negative */ + if (!enable && refcount == 0) { + ret = -EINVAL; + goto end; + } + + DRM_DEBUG_VBL("id:%u enable=%d/%d\n", DRMID(phys_enc->parent), enable, + refcount); + + if (enable) { + if (phys_enc->vblank_refcount == 0) + ret = dpu_core_irq_register_callback(phys_enc->dpu_kms, + phys_enc->irq[INTR_IDX_VSYNC], + dpu_encoder_phys_vid_vblank_irq, + phys_enc); + if (!ret) + phys_enc->vblank_refcount++; + } else if (!enable) { + if (phys_enc->vblank_refcount == 1) + ret = dpu_core_irq_unregister_callback(phys_enc->dpu_kms, + phys_enc->irq[INTR_IDX_VSYNC]); + if (!ret) + phys_enc->vblank_refcount--; + } + +end: + mutex_unlock(&phys_enc->vblank_ctl_lock); + if (ret) { + DRM_ERROR("failed: id:%u intf:%d ret:%d enable:%d refcnt:%d\n", + DRMID(phys_enc->parent), + phys_enc->hw_intf->idx - INTF_0, ret, enable, + refcount); + } + return ret; +} + +static void dpu_encoder_phys_vid_enable(struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_ctl *ctl; + const struct msm_format *fmt; + u32 fmt_fourcc; + u32 mode_3d; + + ctl = phys_enc->hw_ctl; + fmt_fourcc = dpu_encoder_get_drm_fmt(phys_enc); + fmt = mdp_get_format(&phys_enc->dpu_kms->base, fmt_fourcc, 0); + mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc); + + DPU_DEBUG_VIDENC(phys_enc, "\n"); + + if (WARN_ON(!phys_enc->hw_intf->ops.enable_timing)) + return; + + dpu_encoder_helper_split_config(phys_enc, phys_enc->hw_intf->idx); + + dpu_encoder_helper_phys_setup_cdm(phys_enc, fmt, CDM_CDWN_OUTPUT_HDMI); + + dpu_encoder_phys_vid_setup_timing_engine(phys_enc); + + /* + * For single flush cases (dual-ctl or pp-split), skip setting the + * flush bit for the slave intf, since both intfs use same ctl + * and HW will only flush the master. + */ + if (dpu_encoder_phys_vid_needs_single_flush(phys_enc) && + !dpu_encoder_phys_vid_is_master(phys_enc)) + goto skip_flush; + + ctl->ops.update_pending_flush_intf(ctl, phys_enc->hw_intf->idx); + if (mode_3d && ctl->ops.update_pending_flush_merge_3d && + phys_enc->hw_pp->merge_3d) + ctl->ops.update_pending_flush_merge_3d(ctl, phys_enc->hw_pp->merge_3d->idx); + + if (ctl->ops.update_pending_flush_cdm && phys_enc->hw_cdm) + ctl->ops.update_pending_flush_cdm(ctl, phys_enc->hw_cdm->idx); + + /* + * Peripheral flush must be updated whenever flushing SDP packets is needed. + * SDP packets are required for any YUV format (YUV420, YUV422, YUV444). + */ + if (ctl->ops.update_pending_flush_periph && dpu_encoder_needs_periph_flush(phys_enc)) + ctl->ops.update_pending_flush_periph(ctl, phys_enc->hw_intf->idx); + +skip_flush: + DPU_DEBUG_VIDENC(phys_enc, + "update pending flush ctl %d intf %d\n", + ctl->idx - CTL_0, phys_enc->hw_intf->idx); + + atomic_set(&phys_enc->underrun_cnt, 0); + + /* ctl_flush & timing engine enable will be triggered by framework */ + if (phys_enc->enable_state == DPU_ENC_DISABLED) + phys_enc->enable_state = DPU_ENC_ENABLING; +} + +static int dpu_encoder_phys_vid_wait_for_tx_complete( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_encoder_wait_info wait_info; + int ret; + + wait_info.wq = &phys_enc->pending_kickoff_wq; + wait_info.atomic_cnt = &phys_enc->pending_kickoff_cnt; + wait_info.timeout_ms = KICKOFF_TIMEOUT_MS; + + if (!dpu_encoder_phys_vid_is_master(phys_enc)) { + return 0; + } + + /* Wait for kickoff to complete */ + ret = dpu_encoder_helper_wait_for_irq(phys_enc, + phys_enc->irq[INTR_IDX_VSYNC], + dpu_encoder_phys_vid_vblank_irq, + &wait_info); + + if (ret == -ETIMEDOUT) { + dpu_encoder_helper_report_irq_timeout(phys_enc, INTR_IDX_VSYNC); + } + + return ret; +} + +static int dpu_encoder_phys_vid_wait_for_commit_done( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_ctl *hw_ctl = phys_enc->hw_ctl; + int ret; + + if (!hw_ctl) + return 0; + + ret = wait_event_timeout(phys_enc->pending_kickoff_wq, + (hw_ctl->ops.get_flush_register(hw_ctl) == 0), + msecs_to_jiffies(50)); + if (ret <= 0) { + DPU_ERROR("vblank timeout: %x\n", hw_ctl->ops.get_flush_register(hw_ctl)); + return -ETIMEDOUT; + } + + return 0; +} + +static void dpu_encoder_phys_vid_prepare_for_kickoff( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_ctl *ctl; + int rc; + struct drm_encoder *drm_enc; + + drm_enc = phys_enc->parent; + + ctl = phys_enc->hw_ctl; + if (!ctl->ops.wait_reset_status) + return; + + /* + * hw supports hardware initiated ctl reset, so before we kickoff a new + * frame, need to check and wait for hw initiated ctl reset completion + */ + rc = ctl->ops.wait_reset_status(ctl); + if (rc) { + DPU_ERROR_VIDENC(phys_enc, "ctl %d reset failure: %d\n", + ctl->idx, rc); + msm_disp_snapshot_state(drm_enc->dev); + dpu_core_irq_unregister_callback(phys_enc->dpu_kms, + phys_enc->irq[INTR_IDX_VSYNC]); + } +} + +static void dpu_encoder_phys_vid_disable(struct dpu_encoder_phys *phys_enc) +{ + unsigned long lock_flags; + int ret; + struct dpu_hw_intf_status intf_status = {0}; + + if (!phys_enc->parent || !phys_enc->parent->dev) { + DPU_ERROR("invalid encoder/device\n"); + return; + } + + if (!phys_enc->hw_intf) { + DPU_ERROR("invalid hw_intf %d hw_ctl %d\n", + phys_enc->hw_intf != NULL, phys_enc->hw_ctl != NULL); + return; + } + + if (WARN_ON(!phys_enc->hw_intf->ops.enable_timing)) + return; + + if (phys_enc->enable_state == DPU_ENC_DISABLED) { + DPU_ERROR("already disabled\n"); + return; + } + + spin_lock_irqsave(phys_enc->enc_spinlock, lock_flags); + phys_enc->hw_intf->ops.enable_timing(phys_enc->hw_intf, 0); + if (dpu_encoder_phys_vid_is_master(phys_enc)) + dpu_encoder_phys_inc_pending(phys_enc); + spin_unlock_irqrestore(phys_enc->enc_spinlock, lock_flags); + + /* + * Wait for a vsync so we know the ENABLE=0 latched before + * the (connector) source of the vsync's gets disabled, + * otherwise we end up in a funny state if we re-enable + * before the disable latches, which results that some of + * the settings changes for the new modeset (like new + * scanout buffer) don't latch properly.. + */ + if (dpu_encoder_phys_vid_is_master(phys_enc)) { + ret = dpu_encoder_phys_vid_wait_for_tx_complete(phys_enc); + if (ret) { + atomic_set(&phys_enc->pending_kickoff_cnt, 0); + DRM_ERROR("wait disable failed: id:%u intf:%d ret:%d\n", + DRMID(phys_enc->parent), + phys_enc->hw_intf->idx - INTF_0, ret); + } + } + + if (phys_enc->hw_intf && phys_enc->hw_intf->ops.get_status) + phys_enc->hw_intf->ops.get_status(phys_enc->hw_intf, &intf_status); + + /* + * Wait for a vsync if timing en status is on after timing engine + * is disabled. + */ + if (intf_status.is_en && dpu_encoder_phys_vid_is_master(phys_enc)) { + spin_lock_irqsave(phys_enc->enc_spinlock, lock_flags); + dpu_encoder_phys_inc_pending(phys_enc); + spin_unlock_irqrestore(phys_enc->enc_spinlock, lock_flags); + ret = dpu_encoder_phys_vid_wait_for_tx_complete(phys_enc); + if (ret) { + atomic_set(&phys_enc->pending_kickoff_cnt, 0); + DRM_ERROR("wait disable failed: id:%u intf:%d ret:%d\n", + DRMID(phys_enc->parent), + phys_enc->hw_intf->idx - INTF_0, ret); + } + } + + dpu_encoder_helper_phys_cleanup(phys_enc); + phys_enc->enable_state = DPU_ENC_DISABLED; +} + +static void dpu_encoder_phys_vid_handle_post_kickoff( + struct dpu_encoder_phys *phys_enc) +{ + unsigned long lock_flags; + + /* + * Video mode must flush CTL before enabling timing engine + * Video encoders need to turn on their interfaces now + */ + if (phys_enc->enable_state == DPU_ENC_ENABLING) { + trace_dpu_enc_phys_vid_post_kickoff(DRMID(phys_enc->parent), + phys_enc->hw_intf->idx - INTF_0); + spin_lock_irqsave(phys_enc->enc_spinlock, lock_flags); + phys_enc->hw_intf->ops.enable_timing(phys_enc->hw_intf, 1); + spin_unlock_irqrestore(phys_enc->enc_spinlock, lock_flags); + phys_enc->enable_state = DPU_ENC_ENABLED; + } +} + +static void dpu_encoder_phys_vid_irq_enable(struct dpu_encoder_phys *phys_enc) +{ + int ret; + + trace_dpu_enc_phys_vid_irq_enable(DRMID(phys_enc->parent), + phys_enc->hw_intf->idx - INTF_0, + phys_enc->vblank_refcount); + + ret = dpu_encoder_phys_vid_control_vblank_irq(phys_enc, true); + if (WARN_ON(ret)) + return; + + dpu_core_irq_register_callback(phys_enc->dpu_kms, + phys_enc->irq[INTR_IDX_UNDERRUN], + dpu_encoder_phys_vid_underrun_irq, + phys_enc); +} + +static void dpu_encoder_phys_vid_irq_disable(struct dpu_encoder_phys *phys_enc) +{ + trace_dpu_enc_phys_vid_irq_disable(DRMID(phys_enc->parent), + phys_enc->hw_intf->idx - INTF_0, + phys_enc->vblank_refcount); + + dpu_encoder_phys_vid_control_vblank_irq(phys_enc, false); + dpu_core_irq_unregister_callback(phys_enc->dpu_kms, + phys_enc->irq[INTR_IDX_UNDERRUN]); +} + +static int dpu_encoder_phys_vid_get_line_count( + struct dpu_encoder_phys *phys_enc) +{ + if (!dpu_encoder_phys_vid_is_master(phys_enc)) + return -EINVAL; + + if (!phys_enc->hw_intf || !phys_enc->hw_intf->ops.get_line_count) + return -EINVAL; + + return phys_enc->hw_intf->ops.get_line_count(phys_enc->hw_intf); +} + +static int dpu_encoder_phys_vid_get_frame_count( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_intf_status s = {0}; + u32 fetch_start = 0; + struct drm_display_mode mode; + + drm_mode_init(&mode, &phys_enc->cached_mode); + + if (!dpu_encoder_phys_vid_is_master(phys_enc)) + return -EINVAL; + + if (!phys_enc->hw_intf || !phys_enc->hw_intf->ops.get_status) + return -EINVAL; + + phys_enc->hw_intf->ops.get_status(phys_enc->hw_intf, &s); + + if (s.is_prog_fetch_en && s.is_en) { + fetch_start = mode.vtotal - (mode.vsync_start - mode.vdisplay); + if ((s.line_count > fetch_start) && + (s.line_count <= mode.vtotal)) + return s.frame_count + 1; + } + + return s.frame_count; +} + +static void dpu_encoder_phys_vid_init_ops(struct dpu_encoder_phys_ops *ops) +{ + ops->is_master = dpu_encoder_phys_vid_is_master; + ops->atomic_mode_set = dpu_encoder_phys_vid_atomic_mode_set; + ops->enable = dpu_encoder_phys_vid_enable; + ops->disable = dpu_encoder_phys_vid_disable; + ops->control_vblank_irq = dpu_encoder_phys_vid_control_vblank_irq; + ops->wait_for_commit_done = dpu_encoder_phys_vid_wait_for_commit_done; + ops->wait_for_tx_complete = dpu_encoder_phys_vid_wait_for_tx_complete; + ops->irq_enable = dpu_encoder_phys_vid_irq_enable; + ops->irq_disable = dpu_encoder_phys_vid_irq_disable; + ops->prepare_for_kickoff = dpu_encoder_phys_vid_prepare_for_kickoff; + ops->handle_post_kickoff = dpu_encoder_phys_vid_handle_post_kickoff; + ops->needs_single_flush = dpu_encoder_phys_vid_needs_single_flush; + ops->get_line_count = dpu_encoder_phys_vid_get_line_count; + ops->get_frame_count = dpu_encoder_phys_vid_get_frame_count; +} + +/** + * dpu_encoder_phys_vid_init - Construct a new video mode physical encoder + * @dev: Corresponding device for devres management + * @p: Pointer to init params structure + * Return: Error code or newly allocated encoder + */ +struct dpu_encoder_phys *dpu_encoder_phys_vid_init(struct drm_device *dev, + struct dpu_enc_phys_init_params *p) +{ + struct dpu_encoder_phys *phys_enc = NULL; + + if (!p) { + DPU_ERROR("failed to create encoder due to invalid parameter\n"); + return ERR_PTR(-EINVAL); + } + + phys_enc = drmm_kzalloc(dev, sizeof(*phys_enc), GFP_KERNEL); + if (!phys_enc) { + DPU_ERROR("failed to create encoder due to memory allocation error\n"); + return ERR_PTR(-ENOMEM); + } + + DPU_DEBUG_VIDENC(phys_enc, "\n"); + + dpu_encoder_phys_init(phys_enc, p); + mutex_init(&phys_enc->vblank_ctl_lock); + phys_enc->vblank_refcount = 0; + + dpu_encoder_phys_vid_init_ops(&phys_enc->ops); + phys_enc->intf_mode = INTF_MODE_VIDEO; + + DPU_DEBUG_VIDENC(phys_enc, "created intf idx:%d\n", p->hw_intf->idx); + + return phys_enc; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c new file mode 100644 index 000000000..22433bfbe --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c @@ -0,0 +1,694 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2022-2024 Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ + +#include + +#include +#include + +#include "dpu_encoder_phys.h" +#include "dpu_formats.h" +#include "dpu_hw_top.h" +#include "dpu_hw_wb.h" +#include "dpu_hw_lm.h" +#include "dpu_hw_merge3d.h" +#include "dpu_hw_interrupts.h" +#include "dpu_core_irq.h" +#include "dpu_vbif.h" +#include "dpu_crtc.h" +#include "disp/msm_disp_snapshot.h" + +#define to_dpu_encoder_phys_wb(x) \ + container_of(x, struct dpu_encoder_phys_wb, base) + +/** + * dpu_encoder_phys_wb_is_master - report wb always as master encoder + * @phys_enc: Pointer to physical encoder + */ +static bool dpu_encoder_phys_wb_is_master(struct dpu_encoder_phys *phys_enc) +{ + /* there is only one physical enc for dpu_writeback */ + return true; +} + +static bool _dpu_encoder_phys_wb_clk_force_ctrl(struct dpu_hw_wb *wb, + struct dpu_hw_mdp *mdp, + bool enable, bool *forced_on) +{ + if (wb->ops.setup_clk_force_ctrl) { + *forced_on = wb->ops.setup_clk_force_ctrl(wb, enable); + return true; + } + + if (mdp->ops.setup_clk_force_ctrl) { + *forced_on = mdp->ops.setup_clk_force_ctrl(mdp, wb->caps->clk_ctrl, enable); + return true; + } + + return false; +} + +/** + * dpu_encoder_phys_wb_set_ot_limit - set OT limit for writeback interface + * @phys_enc: Pointer to physical encoder + */ +static void dpu_encoder_phys_wb_set_ot_limit( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_wb *hw_wb = phys_enc->hw_wb; + struct dpu_vbif_set_ot_params ot_params; + bool forced_on = false; + + memset(&ot_params, 0, sizeof(ot_params)); + ot_params.xin_id = hw_wb->caps->xin_id; + ot_params.num = hw_wb->idx - WB_0; + ot_params.width = phys_enc->cached_mode.hdisplay; + ot_params.height = phys_enc->cached_mode.vdisplay; + ot_params.is_wfd = !dpu_encoder_helper_get_cwb_mask(phys_enc); + ot_params.frame_rate = drm_mode_vrefresh(&phys_enc->cached_mode); + /* XXX: WB on MSM8996 should use VBIF_NRT */ + ot_params.rd = false; + + if (!_dpu_encoder_phys_wb_clk_force_ctrl(hw_wb, phys_enc->dpu_kms->hw_mdp, + true, &forced_on)) + return; + + dpu_vbif_set_ot_limit(phys_enc->dpu_kms, &ot_params); + + if (forced_on) + _dpu_encoder_phys_wb_clk_force_ctrl(hw_wb, phys_enc->dpu_kms->hw_mdp, + false, &forced_on); +} + +/** + * dpu_encoder_phys_wb_set_qos_remap - set QoS remapper for writeback + * @phys_enc: Pointer to physical encoder + */ +static void dpu_encoder_phys_wb_set_qos_remap( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_wb *hw_wb; + struct dpu_vbif_set_qos_params qos_params; + bool forced_on = false; + + if (!phys_enc || !phys_enc->parent || !phys_enc->parent->crtc) { + DPU_ERROR("invalid arguments\n"); + return; + } + + if (!phys_enc->hw_wb || !phys_enc->hw_wb->caps) { + DPU_ERROR("invalid writeback hardware\n"); + return; + } + + hw_wb = phys_enc->hw_wb; + + memset(&qos_params, 0, sizeof(qos_params)); + /* XXX: WB on MSM8996 should use VBIF_NRT */ + qos_params.xin_id = hw_wb->caps->xin_id; + qos_params.num = hw_wb->idx - WB_0; + qos_params.is_rt = dpu_encoder_helper_get_cwb_mask(phys_enc); + + DPU_DEBUG("[qos_remap] wb:%d xin:%d is_rt:%d\n", + qos_params.num, + qos_params.xin_id, qos_params.is_rt); + + if (!_dpu_encoder_phys_wb_clk_force_ctrl(hw_wb, phys_enc->dpu_kms->hw_mdp, + true, &forced_on)) + return; + + dpu_vbif_set_qos_remap(phys_enc->dpu_kms, &qos_params); + + if (forced_on) + _dpu_encoder_phys_wb_clk_force_ctrl(hw_wb, phys_enc->dpu_kms->hw_mdp, + false, &forced_on); +} + +/** + * dpu_encoder_phys_wb_set_qos - set QoS/danger/safe LUTs for writeback + * @phys_enc: Pointer to physical encoder + */ +static void dpu_encoder_phys_wb_set_qos(struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_wb *hw_wb; + struct dpu_hw_qos_cfg qos_cfg; + const struct dpu_mdss_cfg *catalog; + const struct dpu_qos_lut_tbl *qos_lut_tb; + + if (!phys_enc || !phys_enc->dpu_kms || !phys_enc->dpu_kms->catalog) { + DPU_ERROR("invalid parameter(s)\n"); + return; + } + + catalog = phys_enc->dpu_kms->catalog; + + hw_wb = phys_enc->hw_wb; + + memset(&qos_cfg, 0, sizeof(struct dpu_hw_qos_cfg)); + qos_cfg.danger_safe_en = true; + qos_cfg.danger_lut = + catalog->perf->danger_lut_tbl[DPU_QOS_LUT_USAGE_NRT]; + + qos_cfg.safe_lut = catalog->perf->safe_lut_tbl[DPU_QOS_LUT_USAGE_NRT]; + + qos_lut_tb = &catalog->perf->qos_lut_tbl[DPU_QOS_LUT_USAGE_NRT]; + qos_cfg.creq_lut = _dpu_hw_get_qos_lut(qos_lut_tb, 0); + + if (hw_wb->ops.setup_qos_lut) + hw_wb->ops.setup_qos_lut(hw_wb, &qos_cfg); +} + +/** + * dpu_encoder_phys_wb_setup_fb - setup output framebuffer + * @phys_enc: Pointer to physical encoder + * @format: Format of the framebuffer + */ +static void dpu_encoder_phys_wb_setup_fb(struct dpu_encoder_phys *phys_enc, + const struct msm_format *format) +{ + struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc); + struct dpu_hw_wb *hw_wb; + struct dpu_hw_wb_cfg *wb_cfg; + u32 cdp_usage; + + if (!phys_enc || !phys_enc->dpu_kms || !phys_enc->dpu_kms->catalog) { + DPU_ERROR("invalid encoder\n"); + return; + } + + hw_wb = phys_enc->hw_wb; + wb_cfg = &wb_enc->wb_cfg; + if (dpu_encoder_helper_get_cwb_mask(phys_enc)) + cdp_usage = DPU_PERF_CDP_USAGE_RT; + else + cdp_usage = DPU_PERF_CDP_USAGE_NRT; + + wb_cfg->intf_mode = phys_enc->intf_mode; + wb_cfg->roi.x1 = 0; + wb_cfg->roi.x2 = phys_enc->cached_mode.hdisplay; + wb_cfg->roi.y1 = 0; + wb_cfg->roi.y2 = phys_enc->cached_mode.vdisplay; + + if (hw_wb->ops.setup_roi) + hw_wb->ops.setup_roi(hw_wb, wb_cfg); + + if (hw_wb->ops.setup_outformat) + hw_wb->ops.setup_outformat(hw_wb, wb_cfg, format); + + if (hw_wb->ops.setup_cdp) { + const struct dpu_perf_cfg *perf = phys_enc->dpu_kms->catalog->perf; + + hw_wb->ops.setup_cdp(hw_wb, format, + perf->cdp_cfg[cdp_usage].wr_enable); + } + + if (hw_wb->ops.setup_outaddress) + hw_wb->ops.setup_outaddress(hw_wb, wb_cfg); +} + +/** + * dpu_encoder_phys_wb_setup_ctl - setup wb pipeline for ctl path + * @phys_enc:Pointer to physical encoder + */ +static void dpu_encoder_phys_wb_setup_ctl(struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_wb *hw_wb; + struct dpu_hw_cdm *hw_cdm; + + if (!phys_enc) { + DPU_ERROR("invalid encoder\n"); + return; + } + + hw_wb = phys_enc->hw_wb; + hw_cdm = phys_enc->hw_cdm; + + if (phys_enc->dpu_kms->catalog->mdss_ver->core_major_ver >= 5 && + (phys_enc->hw_ctl && + phys_enc->hw_ctl->ops.setup_intf_cfg)) { + struct dpu_hw_intf_cfg intf_cfg = {0}; + struct dpu_hw_pingpong *hw_pp = phys_enc->hw_pp; + enum dpu_3d_blend_mode mode_3d; + + mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc); + + intf_cfg.intf = DPU_NONE; + intf_cfg.wb = hw_wb->idx; + intf_cfg.cwb = dpu_encoder_helper_get_cwb_mask(phys_enc); + + if (mode_3d && hw_pp && hw_pp->merge_3d) + intf_cfg.merge_3d = hw_pp->merge_3d->idx; + + if (hw_cdm) + intf_cfg.cdm = hw_cdm->idx; + + if (hw_pp && hw_pp->merge_3d && hw_pp->merge_3d->ops.setup_3d_mode) + hw_pp->merge_3d->ops.setup_3d_mode(hw_pp->merge_3d, mode_3d); + + /* setup which pp blk will connect to this wb */ + if (hw_pp && hw_wb->ops.bind_pingpong_blk) + hw_wb->ops.bind_pingpong_blk(hw_wb, hw_pp->idx); + + phys_enc->hw_ctl->ops.setup_intf_cfg(phys_enc->hw_ctl, &intf_cfg); + } else if (phys_enc->hw_ctl && phys_enc->hw_ctl->ops.setup_intf_cfg) { + struct dpu_hw_intf_cfg intf_cfg = {0}; + + intf_cfg.intf = DPU_NONE; + intf_cfg.wb = hw_wb->idx; + intf_cfg.mode_3d = + dpu_encoder_helper_get_3d_blend_mode(phys_enc); + phys_enc->hw_ctl->ops.setup_intf_cfg(phys_enc->hw_ctl, &intf_cfg); + } +} + +/** + * _dpu_encoder_phys_wb_update_flush - flush hardware update + * @phys_enc: Pointer to physical encoder + */ +static void _dpu_encoder_phys_wb_update_flush(struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_wb *hw_wb; + struct dpu_hw_ctl *hw_ctl; + struct dpu_hw_pingpong *hw_pp; + struct dpu_hw_cdm *hw_cdm; + u32 pending_flush = 0; + u32 mode_3d; + + if (!phys_enc) + return; + + hw_wb = phys_enc->hw_wb; + hw_pp = phys_enc->hw_pp; + hw_ctl = phys_enc->hw_ctl; + hw_cdm = phys_enc->hw_cdm; + mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc); + + DPU_DEBUG("[wb:%d]\n", hw_wb->idx - WB_0); + + if (!hw_ctl) { + DPU_DEBUG("[wb:%d] no ctl assigned\n", hw_wb->idx - WB_0); + return; + } + + if (hw_ctl->ops.update_pending_flush_wb) + hw_ctl->ops.update_pending_flush_wb(hw_ctl, hw_wb->idx); + + if (mode_3d && hw_ctl->ops.update_pending_flush_merge_3d && + hw_pp && hw_pp->merge_3d) + hw_ctl->ops.update_pending_flush_merge_3d(hw_ctl, + hw_pp->merge_3d->idx); + + if (hw_cdm && hw_ctl->ops.update_pending_flush_cdm) + hw_ctl->ops.update_pending_flush_cdm(hw_ctl, hw_cdm->idx); + + if (hw_ctl->ops.get_pending_flush) + pending_flush = hw_ctl->ops.get_pending_flush(hw_ctl); + + DPU_DEBUG("Pending flush mask for CTL_%d is 0x%x, WB %d\n", + hw_ctl->idx - CTL_0, pending_flush, + hw_wb->idx - WB_0); +} + +/** + * dpu_encoder_phys_wb_setup - setup writeback encoder + * @phys_enc: Pointer to physical encoder + */ +static void dpu_encoder_phys_wb_setup( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_wb *hw_wb = phys_enc->hw_wb; + struct drm_display_mode mode = phys_enc->cached_mode; + struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc); + const struct msm_format *format; + + format = msm_framebuffer_format(wb_enc->wb_job->fb); + + DPU_DEBUG("[mode_set:%d, \"%s\",%d,%d]\n", + hw_wb->idx - WB_0, mode.name, + mode.hdisplay, mode.vdisplay); + + dpu_encoder_phys_wb_set_ot_limit(phys_enc); + + dpu_encoder_phys_wb_set_qos_remap(phys_enc); + + dpu_encoder_phys_wb_set_qos(phys_enc); + + dpu_encoder_phys_wb_setup_fb(phys_enc, format); + + dpu_encoder_helper_phys_setup_cdm(phys_enc, format, CDM_CDWN_OUTPUT_WB); + + dpu_encoder_helper_phys_setup_cwb(phys_enc, true); + + dpu_encoder_phys_wb_setup_ctl(phys_enc); +} + +/** + * dpu_encoder_phys_wb_done_irq - writeback interrupt handler + * @arg: Pointer to writeback encoder + */ +static void dpu_encoder_phys_wb_done_irq(void *arg) +{ + struct dpu_encoder_phys *phys_enc = arg; + struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc); + + struct dpu_hw_wb *hw_wb = phys_enc->hw_wb; + unsigned long lock_flags; + u32 event = DPU_ENCODER_FRAME_EVENT_DONE; + + DPU_DEBUG("[wb:%d]\n", hw_wb->idx - WB_0); + + dpu_encoder_frame_done_callback(phys_enc->parent, phys_enc, event); + + dpu_encoder_vblank_callback(phys_enc->parent, phys_enc); + + spin_lock_irqsave(phys_enc->enc_spinlock, lock_flags); + atomic_add_unless(&phys_enc->pending_kickoff_cnt, -1, 0); + spin_unlock_irqrestore(phys_enc->enc_spinlock, lock_flags); + + if (wb_enc->wb_conn) + drm_writeback_signal_completion(wb_enc->wb_conn, 0); + + /* Signal any waiting atomic commit thread */ + wake_up_all(&phys_enc->pending_kickoff_wq); +} + +/** + * dpu_encoder_phys_wb_irq_enable - irq control of WB + * @phys: Pointer to physical encoder + */ +static void dpu_encoder_phys_wb_irq_enable(struct dpu_encoder_phys *phys) +{ + + struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys); + + if (atomic_inc_return(&wb_enc->wbirq_refcount) == 1) + dpu_core_irq_register_callback(phys->dpu_kms, + phys->irq[INTR_IDX_WB_DONE], + dpu_encoder_phys_wb_done_irq, + phys); +} + +/** + * dpu_encoder_phys_wb_irq_disable - irq control of WB + * @phys: Pointer to physical encoder + */ +static void dpu_encoder_phys_wb_irq_disable(struct dpu_encoder_phys *phys) +{ + + struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys); + + if (atomic_dec_return(&wb_enc->wbirq_refcount) == 0) + dpu_core_irq_unregister_callback(phys->dpu_kms, phys->irq[INTR_IDX_WB_DONE]); +} + +static void dpu_encoder_phys_wb_atomic_mode_set( + struct dpu_encoder_phys *phys_enc, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state) +{ + + phys_enc->irq[INTR_IDX_WB_DONE] = phys_enc->hw_wb->caps->intr_wb_done; +} + +static void _dpu_encoder_phys_wb_handle_wbdone_timeout( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc); + u32 frame_event = DPU_ENCODER_FRAME_EVENT_ERROR; + + wb_enc->wb_done_timeout_cnt++; + + if (wb_enc->wb_done_timeout_cnt == 1) + msm_disp_snapshot_state(phys_enc->parent->dev); + + atomic_add_unless(&phys_enc->pending_kickoff_cnt, -1, 0); + + /* request a ctl reset before the next kickoff */ + phys_enc->enable_state = DPU_ENC_ERR_NEEDS_HW_RESET; + + if (wb_enc->wb_conn) + drm_writeback_signal_completion(wb_enc->wb_conn, 0); + + dpu_encoder_frame_done_callback(phys_enc->parent, phys_enc, frame_event); +} + +/** + * dpu_encoder_phys_wb_wait_for_commit_done - wait until request is committed + * @phys_enc: Pointer to physical encoder + */ +static int dpu_encoder_phys_wb_wait_for_commit_done( + struct dpu_encoder_phys *phys_enc) +{ + int ret; + struct dpu_encoder_wait_info wait_info; + struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc); + + wait_info.wq = &phys_enc->pending_kickoff_wq; + wait_info.atomic_cnt = &phys_enc->pending_kickoff_cnt; + wait_info.timeout_ms = KICKOFF_TIMEOUT_MS; + + ret = dpu_encoder_helper_wait_for_irq(phys_enc, + phys_enc->irq[INTR_IDX_WB_DONE], + dpu_encoder_phys_wb_done_irq, &wait_info); + if (ret == -ETIMEDOUT) + _dpu_encoder_phys_wb_handle_wbdone_timeout(phys_enc); + else if (!ret) + wb_enc->wb_done_timeout_cnt = 0; + + return ret; +} + +/** + * dpu_encoder_phys_wb_prepare_for_kickoff - pre-kickoff processing + * @phys_enc: Pointer to physical encoder + * Returns: Zero on success + */ +static void dpu_encoder_phys_wb_prepare_for_kickoff( + struct dpu_encoder_phys *phys_enc) +{ + struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc); + struct drm_connector *drm_conn; + struct drm_connector_state *state; + + DPU_DEBUG("[wb:%d]\n", phys_enc->hw_wb->idx - WB_0); + + if (!wb_enc->wb_conn || !wb_enc->wb_job) { + DPU_ERROR("invalid wb_conn or wb_job\n"); + return; + } + + drm_conn = &wb_enc->wb_conn->base; + state = drm_conn->state; + + if (wb_enc->wb_conn && wb_enc->wb_job) + drm_writeback_queue_job(wb_enc->wb_conn, state); + + dpu_encoder_phys_wb_setup(phys_enc); + + _dpu_encoder_phys_wb_update_flush(phys_enc); +} + +/** + * dpu_encoder_phys_wb_needs_single_flush - trigger flush processing + * @phys_enc: Pointer to physical encoder + */ +static bool dpu_encoder_phys_wb_needs_single_flush(struct dpu_encoder_phys *phys_enc) +{ + DPU_DEBUG("[wb:%d]\n", phys_enc->hw_wb->idx - WB_0); + return false; +} + +/** + * dpu_encoder_phys_wb_handle_post_kickoff - post-kickoff processing + * @phys_enc: Pointer to physical encoder + */ +static void dpu_encoder_phys_wb_handle_post_kickoff( + struct dpu_encoder_phys *phys_enc) +{ + DPU_DEBUG("[wb:%d]\n", phys_enc->hw_wb->idx - WB_0); + +} + +/** + * dpu_encoder_phys_wb_enable - enable writeback encoder + * @phys_enc: Pointer to physical encoder + */ +static void dpu_encoder_phys_wb_enable(struct dpu_encoder_phys *phys_enc) +{ + DPU_DEBUG("[wb:%d]\n", phys_enc->hw_wb->idx - WB_0); + phys_enc->enable_state = DPU_ENC_ENABLED; +} +/** + * dpu_encoder_phys_wb_disable - disable writeback encoder + * @phys_enc: Pointer to physical encoder + */ +static void dpu_encoder_phys_wb_disable(struct dpu_encoder_phys *phys_enc) +{ + struct dpu_hw_wb *hw_wb = phys_enc->hw_wb; + + DPU_DEBUG("[wb:%d]\n", hw_wb->idx - WB_0); + + if (phys_enc->enable_state == DPU_ENC_DISABLED) { + DPU_ERROR("encoder is already disabled\n"); + return; + } + + /* reset h/w before final flush */ + phys_enc->hw_ctl->ops.clear_pending_flush(phys_enc->hw_ctl); + + /* + * New CTL reset sequence from 5.0 MDP onwards. + * If has_3d_merge_reset is not set, legacy reset + * sequence is executed. + * + * Legacy reset sequence has not been implemented yet. + * Any target earlier than SM8150 will need it and when + * WB support is added to those targets will need to add + * the legacy teardown sequence as well. + */ + if (phys_enc->dpu_kms->catalog->mdss_ver->core_major_ver >= 5) + dpu_encoder_helper_phys_cleanup(phys_enc); + + phys_enc->enable_state = DPU_ENC_DISABLED; +} + +static void dpu_encoder_phys_wb_prepare_wb_job(struct dpu_encoder_phys *phys_enc, + struct drm_writeback_job *job) +{ + const struct msm_format *format; + struct dpu_hw_wb_cfg *wb_cfg; + int ret; + struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc); + + if (!job->fb) + return; + + wb_enc->wb_job = job; + wb_enc->wb_conn = job->connector; + + wb_cfg = &wb_enc->wb_cfg; + + memset(wb_cfg, 0, sizeof(struct dpu_hw_wb_cfg)); + + ret = msm_framebuffer_prepare(job->fb, false); + if (ret) { + DPU_ERROR("prep fb failed, %d\n", ret); + return; + } + + format = msm_framebuffer_format(job->fb); + + ret = dpu_format_populate_plane_sizes(job->fb, &wb_cfg->dest); + if (ret) { + DPU_DEBUG("failed to populate plane sizes%d\n", ret); + return; + } + + dpu_format_populate_addrs(job->fb, &wb_cfg->dest); + + wb_cfg->dest.width = job->fb->width; + wb_cfg->dest.height = job->fb->height; + wb_cfg->dest.num_planes = format->num_planes; + + if ((format->fetch_type == MDP_PLANE_PLANAR) && + (format->element[0] == C1_B_Cb)) + swap(wb_cfg->dest.plane_addr[1], wb_cfg->dest.plane_addr[2]); + + DPU_DEBUG("[fb_offset:%8.8x,%8.8x,%8.8x,%8.8x]\n", + wb_cfg->dest.plane_addr[0], wb_cfg->dest.plane_addr[1], + wb_cfg->dest.plane_addr[2], wb_cfg->dest.plane_addr[3]); + + DPU_DEBUG("[fb_stride:%8.8x,%8.8x,%8.8x,%8.8x]\n", + wb_cfg->dest.plane_pitch[0], wb_cfg->dest.plane_pitch[1], + wb_cfg->dest.plane_pitch[2], wb_cfg->dest.plane_pitch[3]); +} + +static void dpu_encoder_phys_wb_cleanup_wb_job(struct dpu_encoder_phys *phys_enc, + struct drm_writeback_job *job) +{ + struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc); + + if (!job->fb) + return; + + msm_framebuffer_cleanup(job->fb, false); + wb_enc->wb_job = NULL; + wb_enc->wb_conn = NULL; +} + +static bool dpu_encoder_phys_wb_is_valid_for_commit(struct dpu_encoder_phys *phys_enc) +{ + struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc); + + if (wb_enc->wb_job) + return true; + else + return false; +} + +/** + * dpu_encoder_phys_wb_init_ops - initialize writeback operations + * @ops: Pointer to encoder operation table + */ +static void dpu_encoder_phys_wb_init_ops(struct dpu_encoder_phys_ops *ops) +{ + ops->is_master = dpu_encoder_phys_wb_is_master; + ops->atomic_mode_set = dpu_encoder_phys_wb_atomic_mode_set; + ops->enable = dpu_encoder_phys_wb_enable; + ops->disable = dpu_encoder_phys_wb_disable; + ops->wait_for_commit_done = dpu_encoder_phys_wb_wait_for_commit_done; + ops->prepare_for_kickoff = dpu_encoder_phys_wb_prepare_for_kickoff; + ops->handle_post_kickoff = dpu_encoder_phys_wb_handle_post_kickoff; + ops->needs_single_flush = dpu_encoder_phys_wb_needs_single_flush; + ops->trigger_start = dpu_encoder_helper_trigger_start; + ops->prepare_wb_job = dpu_encoder_phys_wb_prepare_wb_job; + ops->cleanup_wb_job = dpu_encoder_phys_wb_cleanup_wb_job; + ops->irq_enable = dpu_encoder_phys_wb_irq_enable; + ops->irq_disable = dpu_encoder_phys_wb_irq_disable; + ops->is_valid_for_commit = dpu_encoder_phys_wb_is_valid_for_commit; + +} + +/** + * dpu_encoder_phys_wb_init - initialize writeback encoder + * @dev: Corresponding device for devres management + * @p: Pointer to init info structure with initialization params + */ +struct dpu_encoder_phys *dpu_encoder_phys_wb_init(struct drm_device *dev, + struct dpu_enc_phys_init_params *p) +{ + struct dpu_encoder_phys *phys_enc = NULL; + struct dpu_encoder_phys_wb *wb_enc = NULL; + + DPU_DEBUG("\n"); + + if (!p || !p->parent) { + DPU_ERROR("invalid params\n"); + return ERR_PTR(-EINVAL); + } + + wb_enc = drmm_kzalloc(dev, sizeof(*wb_enc), GFP_KERNEL); + if (!wb_enc) { + DPU_ERROR("failed to allocate wb phys_enc enc\n"); + return ERR_PTR(-ENOMEM); + } + + phys_enc = &wb_enc->base; + + dpu_encoder_phys_init(phys_enc, p); + + dpu_encoder_phys_wb_init_ops(&phys_enc->ops); + phys_enc->intf_mode = INTF_MODE_WB_LINE; + + atomic_set(&wb_enc->wbirq_refcount, 0); + + wb_enc->wb_done_timeout_cnt = 0; + + DPU_DEBUG("Created dpu_encoder_phys for wb %d\n", phys_enc->hw_wb->idx); + + return phys_enc; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c new file mode 100644 index 000000000..590922c4f --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c @@ -0,0 +1,364 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ + +#include +#include + +#include "dpu_kms.h" +#include "dpu_formats.h" + +#define DPU_UBWC_PLANE_SIZE_ALIGNMENT 4096 + +/* + * struct dpu_media_color_map - maps drm format to media format + * @format: DRM base pixel format + * @color: Media API color related to DRM format + */ +struct dpu_media_color_map { + uint32_t format; + uint32_t color; +}; + +/* _dpu_get_v_h_subsample_rate - Get subsample rates for all formats we support + * Note: Not using the drm_format_*_subsampling since we have formats + */ +static void _dpu_get_v_h_subsample_rate( + enum mdp_chroma_samp_type chroma_sample, + uint32_t *v_sample, + uint32_t *h_sample) +{ + if (!v_sample || !h_sample) + return; + + switch (chroma_sample) { + case CHROMA_H2V1: + *v_sample = 1; + *h_sample = 2; + break; + case CHROMA_H1V2: + *v_sample = 2; + *h_sample = 1; + break; + case CHROMA_420: + *v_sample = 2; + *h_sample = 2; + break; + default: + *v_sample = 1; + *h_sample = 1; + break; + } +} + +static int _dpu_format_populate_plane_sizes_ubwc( + const struct msm_format *fmt, + struct drm_framebuffer *fb, + struct dpu_hw_fmt_layout *layout) +{ + bool meta = MSM_FORMAT_IS_UBWC(fmt); + + if (MSM_FORMAT_IS_YUV(fmt)) { + unsigned int stride, y_sclines, uv_sclines; + unsigned int y_tile_width, y_tile_height; + unsigned int y_meta_stride, y_meta_scanlines; + unsigned int uv_meta_stride, uv_meta_scanlines; + + if (MSM_FORMAT_IS_DX(fmt)) { + if (fmt->flags & MSM_FORMAT_FLAG_UNPACK_TIGHT) { + /* can't use round_up() here because 192 is NPoT */ + stride = roundup(fb->width, 192); + stride = round_up(stride * 4 / 3, 256); + y_tile_width = 48; + } else { + stride = round_up(fb->width * 2, 256); + y_tile_width = 32; + } + + y_sclines = round_up(fb->height, 16); + uv_sclines = round_up((fb->height+1)>>1, 16); + y_tile_height = 4; + } else { + stride = round_up(fb->width, 128); + y_tile_width = 32; + + y_sclines = round_up(fb->height, 32); + uv_sclines = round_up((fb->height+1)>>1, 32); + y_tile_height = 8; + } + + layout->plane_pitch[0] = stride; + layout->plane_size[0] = round_up(layout->plane_pitch[0] * + y_sclines, DPU_UBWC_PLANE_SIZE_ALIGNMENT); + + layout->plane_pitch[1] = stride; + layout->plane_size[1] = round_up(layout->plane_pitch[1] * + uv_sclines, DPU_UBWC_PLANE_SIZE_ALIGNMENT); + + if (!meta) + return 0; + + y_meta_stride = DIV_ROUND_UP(fb->width, y_tile_width); + layout->plane_pitch[2] = round_up(y_meta_stride, 64); + + y_meta_scanlines = DIV_ROUND_UP(fb->height, y_tile_height); + y_meta_scanlines = round_up(y_meta_scanlines, 16); + layout->plane_size[2] = round_up(layout->plane_pitch[2] * + y_meta_scanlines, DPU_UBWC_PLANE_SIZE_ALIGNMENT); + + uv_meta_stride = DIV_ROUND_UP((fb->width+1)>>1, y_tile_width / 2); + layout->plane_pitch[3] = round_up(uv_meta_stride, 64); + + uv_meta_scanlines = DIV_ROUND_UP((fb->height+1)>>1, y_tile_height); + uv_meta_scanlines = round_up(uv_meta_scanlines, 16); + layout->plane_size[3] = round_up(layout->plane_pitch[3] * + uv_meta_scanlines, DPU_UBWC_PLANE_SIZE_ALIGNMENT); + } else { + unsigned int rgb_scanlines, rgb_meta_scanlines, rgb_meta_stride; + + layout->plane_pitch[0] = round_up(fb->width * fmt->bpp, 256); + rgb_scanlines = round_up(fb->height, 16); + layout->plane_size[0] = round_up(layout->plane_pitch[0] * + rgb_scanlines, DPU_UBWC_PLANE_SIZE_ALIGNMENT); + + if (!meta) + return 0; + + /* uAPI leaves plane[1] empty and plane[2] as meta */ + layout->num_planes += 1; + + rgb_meta_stride = DIV_ROUND_UP(fb->width, 16); + layout->plane_pitch[2] = round_up(rgb_meta_stride, 64); + + rgb_meta_scanlines = DIV_ROUND_UP(fb->height, 4); + rgb_meta_scanlines = round_up(rgb_meta_scanlines, 16); + + layout->plane_size[2] = round_up(layout->plane_pitch[2] * + rgb_meta_scanlines, DPU_UBWC_PLANE_SIZE_ALIGNMENT); + } + + return 0; +} + +static int _dpu_format_populate_plane_sizes_linear( + const struct msm_format *fmt, + struct drm_framebuffer *fb, + struct dpu_hw_fmt_layout *layout) +{ + int i; + + /* Due to memset above, only need to set planes of interest */ + if (fmt->fetch_type == MDP_PLANE_INTERLEAVED) { + layout->plane_size[0] = fb->width * fb->height * fmt->bpp; + layout->plane_pitch[0] = fb->width * fmt->bpp; + } else { + uint32_t v_subsample, h_subsample; + uint32_t chroma_samp; + uint32_t bpp = 1; + + chroma_samp = fmt->chroma_sample; + _dpu_get_v_h_subsample_rate(chroma_samp, &v_subsample, + &h_subsample); + + if (fb->width % h_subsample || fb->height % v_subsample) { + DRM_ERROR("mismatch in subsample vs dimensions\n"); + return -EINVAL; + } + + if ((fmt->pixel_format == DRM_FORMAT_NV12) && + (MSM_FORMAT_IS_DX(fmt))) + bpp = 2; + layout->plane_pitch[0] = fb->width * bpp; + layout->plane_pitch[1] = layout->plane_pitch[0] / h_subsample; + layout->plane_size[0] = layout->plane_pitch[0] * fb->height; + layout->plane_size[1] = layout->plane_pitch[1] * + (fb->height / v_subsample); + + if (fmt->fetch_type == MDP_PLANE_PSEUDO_PLANAR) { + layout->plane_size[1] *= 2; + layout->plane_pitch[1] *= 2; + } else { + /* planar */ + layout->plane_size[2] = layout->plane_size[1]; + layout->plane_pitch[2] = layout->plane_pitch[1]; + } + } + + /* + * linear format: allow user allocated pitches if they are greater than + * the requirement. + * ubwc format: pitch values are computed uniformly across + * all the components based on ubwc specifications. + */ + for (i = 0; i < layout->num_planes && i < DPU_MAX_PLANES; ++i) { + if (layout->plane_pitch[i] <= fb->pitches[i]) { + layout->plane_pitch[i] = fb->pitches[i]; + } else { + DRM_DEBUG("plane %u expected pitch %u, fb %u\n", + i, layout->plane_pitch[i], fb->pitches[i]); + return -EINVAL; + } + } + + return 0; +} + +/** + * dpu_format_populate_plane_sizes - populate non-address part of the layout based on + * fb, and format found in the fb + * @fb: framebuffer pointer + * @layout: format layout structure to populate + * + * Return: error code on failure or 0 if new addresses were populated + */ +int dpu_format_populate_plane_sizes( + struct drm_framebuffer *fb, + struct dpu_hw_fmt_layout *layout) +{ + const struct msm_format *fmt; + int ret, i; + + if (!layout || !fb) { + DRM_ERROR("invalid pointer\n"); + return -EINVAL; + } + + if (fb->width > DPU_MAX_IMG_WIDTH || + fb->height > DPU_MAX_IMG_HEIGHT) { + DRM_ERROR("image dimensions outside max range\n"); + return -ERANGE; + } + + fmt = msm_framebuffer_format(fb); + + memset(layout, 0, sizeof(struct dpu_hw_fmt_layout)); + layout->width = fb->width; + layout->height = fb->height; + layout->num_planes = fmt->num_planes; + + if (MSM_FORMAT_IS_UBWC(fmt) || MSM_FORMAT_IS_TILE(fmt)) + ret = _dpu_format_populate_plane_sizes_ubwc(fmt, fb, layout); + else + ret = _dpu_format_populate_plane_sizes_linear(fmt, fb, layout); + + if (ret) + return ret; + + for (i = 0; i < DPU_MAX_PLANES; i++) + layout->total_size += layout->plane_size[i]; + + return 0; +} + +static void _dpu_format_populate_addrs_ubwc(struct drm_framebuffer *fb, + struct dpu_hw_fmt_layout *layout) +{ + const struct msm_format *fmt; + uint32_t base_addr = 0; + bool meta; + + base_addr = msm_framebuffer_iova(fb, 0); + + fmt = msm_framebuffer_format(fb); + meta = MSM_FORMAT_IS_UBWC(fmt); + + /* Per-format logic for verifying active planes */ + if (MSM_FORMAT_IS_YUV(fmt)) { + /************************************************/ + /* UBWC ** */ + /* buffer ** DPU PLANE */ + /* format ** */ + /************************************************/ + /* ------------------- ** -------------------- */ + /* | Y meta | ** | Y bitstream | */ + /* | data | ** | plane | */ + /* ------------------- ** -------------------- */ + /* | Y bitstream | ** | CbCr bitstream | */ + /* | data | ** | plane | */ + /* ------------------- ** -------------------- */ + /* | Cbcr metadata | ** | Y meta | */ + /* | data | ** | plane | */ + /* ------------------- ** -------------------- */ + /* | CbCr bitstream | ** | CbCr meta | */ + /* | data | ** | plane | */ + /* ------------------- ** -------------------- */ + /************************************************/ + + /* configure Y bitstream plane */ + layout->plane_addr[0] = base_addr + layout->plane_size[2]; + + /* configure CbCr bitstream plane */ + layout->plane_addr[1] = base_addr + layout->plane_size[0] + + layout->plane_size[2] + layout->plane_size[3]; + + if (!meta) + return; + + /* configure Y metadata plane */ + layout->plane_addr[2] = base_addr; + + /* configure CbCr metadata plane */ + layout->plane_addr[3] = base_addr + layout->plane_size[0] + + layout->plane_size[2]; + + } else { + /************************************************/ + /* UBWC ** */ + /* buffer ** DPU PLANE */ + /* format ** */ + /************************************************/ + /* ------------------- ** -------------------- */ + /* | RGB meta | ** | RGB bitstream | */ + /* | data | ** | plane | */ + /* ------------------- ** -------------------- */ + /* | RGB bitstream | ** | NONE | */ + /* | data | ** | | */ + /* ------------------- ** -------------------- */ + /* ** | RGB meta | */ + /* ** | plane | */ + /* ** -------------------- */ + /************************************************/ + + layout->plane_addr[0] = base_addr + layout->plane_size[2]; + layout->plane_addr[1] = 0; + + if (!meta) + return; + + layout->plane_addr[2] = base_addr; + layout->plane_addr[3] = 0; + } +} + +static void _dpu_format_populate_addrs_linear(struct drm_framebuffer *fb, + struct dpu_hw_fmt_layout *layout) +{ + unsigned int i; + + /* Populate addresses for simple formats here */ + for (i = 0; i < layout->num_planes; ++i) + layout->plane_addr[i] = msm_framebuffer_iova(fb, i); + } + +/** + * dpu_format_populate_addrs - populate buffer addresses based on + * mmu, fb, and format found in the fb + * @fb: framebuffer pointer + * @layout: format layout structure to populate + */ +void dpu_format_populate_addrs(struct drm_framebuffer *fb, + struct dpu_hw_fmt_layout *layout) +{ + const struct msm_format *fmt; + + fmt = msm_framebuffer_format(fb); + + /* Populate the addresses given the fb */ + if (MSM_FORMAT_IS_UBWC(fmt) || + MSM_FORMAT_IS_TILE(fmt)) + _dpu_format_populate_addrs_ubwc(fb, layout); + else + _dpu_format_populate_addrs_linear(fb, layout); +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h new file mode 100644 index 000000000..dc03f522e --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h @@ -0,0 +1,41 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_FORMATS_H +#define _DPU_FORMATS_H + +#include +#include "msm_gem.h" +#include "dpu_hw_mdss.h" + +/** + * dpu_find_format - validate if the pixel format is supported + * @format: dpu format + * @supported_formats: supported formats by dpu HW + * @num_formatss: total number of formats + * + * Return: false if not valid format, true on success + */ +static inline bool dpu_find_format(u32 format, const u32 *supported_formats, + size_t num_formats) +{ + int i; + + for (i = 0; i < num_formats; i++) { + /* check for valid formats supported */ + if (format == supported_formats[i]) + return true; + } + + return false; +} + +void dpu_format_populate_addrs(struct drm_framebuffer *fb, + struct dpu_hw_fmt_layout *layout); + +int dpu_format_populate_plane_sizes( + struct drm_framebuffer *fb, + struct dpu_hw_fmt_layout *layout); + +#endif /*_DPU_FORMATS_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c new file mode 100644 index 000000000..9a993cdfa --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c @@ -0,0 +1,801 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + * Copyright (c) 2022-2023, Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ +#include +#include +#include +#include "dpu_hw_mdss.h" +#include "dpu_hw_interrupts.h" +#include "dpu_hw_catalog.h" +#include "dpu_kms.h" + +#define VIG_BASE_MASK \ + (BIT(DPU_SSPP_QOS) |\ + BIT(DPU_SSPP_CDP) |\ + BIT(DPU_SSPP_TS_PREFILL) | BIT(DPU_SSPP_EXCL_RECT)) + +#define VIG_MASK \ + (VIG_BASE_MASK | \ + BIT(DPU_SSPP_CSC_10BIT)) + +#define VIG_MSM8953_MASK \ + (BIT(DPU_SSPP_QOS) |\ + BIT(DPU_SSPP_SCALER_QSEED2) |\ + BIT(DPU_SSPP_CSC)) + +#define VIG_MSM8996_MASK \ + (BIT(DPU_SSPP_QOS) | BIT(DPU_SSPP_CDP) |\ + BIT(DPU_SSPP_TS_PREFILL) | BIT(DPU_SSPP_SCALER_QSEED2) |\ + BIT(DPU_SSPP_CSC)) + +#define VIG_MSM8998_MASK \ + (VIG_MASK | BIT(DPU_SSPP_SCALER_QSEED3_COMPATIBLE)) + +#define VIG_SDM845_MASK_NO_SDMA \ + (VIG_MASK | BIT(DPU_SSPP_SCALER_QSEED3_COMPATIBLE)) + +#define VIG_SDM845_MASK_SDMA \ + (VIG_SDM845_MASK_NO_SDMA | BIT(DPU_SSPP_SMART_DMA_V2)) + +#define VIG_QCM2290_MASK (VIG_BASE_MASK) + +#define DMA_MSM8953_MASK \ + (BIT(DPU_SSPP_QOS)) + +#define DMA_MSM8996_MASK \ + (BIT(DPU_SSPP_QOS) | BIT(DPU_SSPP_TS_PREFILL) | BIT(DPU_SSPP_CDP)) + +#define DMA_MSM8998_MASK \ + (BIT(DPU_SSPP_QOS) |\ + BIT(DPU_SSPP_TS_PREFILL) | BIT(DPU_SSPP_TS_PREFILL_REC1) |\ + BIT(DPU_SSPP_CDP) | BIT(DPU_SSPP_EXCL_RECT)) + +#define VIG_SC7280_MASK \ + (VIG_SDM845_MASK_NO_SDMA | BIT(DPU_SSPP_INLINE_ROTATION)) + +#define VIG_SC7280_MASK_SDMA \ + (VIG_SC7280_MASK | BIT(DPU_SSPP_SMART_DMA_V2)) + +#define DMA_SDM845_MASK_NO_SDMA \ + (BIT(DPU_SSPP_QOS) | \ + BIT(DPU_SSPP_TS_PREFILL) | BIT(DPU_SSPP_TS_PREFILL_REC1) |\ + BIT(DPU_SSPP_CDP) | BIT(DPU_SSPP_EXCL_RECT)) + +#define DMA_CURSOR_SDM845_MASK_NO_SDMA \ + (DMA_SDM845_MASK_NO_SDMA | BIT(DPU_SSPP_CURSOR)) + +#define DMA_SDM845_MASK_SDMA \ + (DMA_SDM845_MASK_NO_SDMA | BIT(DPU_SSPP_SMART_DMA_V2)) + +#define DMA_CURSOR_SDM845_MASK_SDMA \ + (DMA_CURSOR_SDM845_MASK_NO_SDMA | BIT(DPU_SSPP_SMART_DMA_V2)) + +#define DMA_CURSOR_MSM8996_MASK \ + (DMA_MSM8996_MASK | BIT(DPU_SSPP_CURSOR)) + +#define DMA_CURSOR_MSM8998_MASK \ + (DMA_MSM8998_MASK | BIT(DPU_SSPP_CURSOR)) + +#define RGB_MSM8953_MASK \ + (BIT(DPU_SSPP_QOS)) + +#define RGB_MSM8996_MASK \ + (BIT(DPU_SSPP_QOS) | BIT(DPU_SSPP_CDP) |\ + BIT(DPU_SSPP_TS_PREFILL) | BIT(DPU_SSPP_SCALER_RGB)) + +#define MIXER_MSM8998_MASK \ + (BIT(DPU_MIXER_SOURCESPLIT)) + +#define WB_SDM845_MASK (BIT(DPU_WB_LINE_MODE) | \ + BIT(DPU_WB_UBWC) | \ + BIT(DPU_WB_YUV_CONFIG) | \ + BIT(DPU_WB_PIPE_ALPHA) | \ + BIT(DPU_WB_XY_ROI_OFFSET) | \ + BIT(DPU_WB_QOS) | \ + BIT(DPU_WB_QOS_8LVL) | \ + BIT(DPU_WB_CDP)) + +#define DEFAULT_PIXEL_RAM_SIZE (50 * 1024) +#define DEFAULT_DPU_LINE_WIDTH 2048 +#define DEFAULT_DPU_OUTPUT_LINE_WIDTH 2560 + +#define MAX_HORZ_DECIMATION 4 +#define MAX_VERT_DECIMATION 4 + +#define STRCAT(X, Y) (X Y) + +static const uint32_t plane_formats[] = { + DRM_FORMAT_ARGB8888, + DRM_FORMAT_ABGR8888, + DRM_FORMAT_RGBA8888, + DRM_FORMAT_BGRA8888, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_RGBX8888, + DRM_FORMAT_BGRX8888, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_ARGB2101010, + DRM_FORMAT_XRGB2101010, + DRM_FORMAT_RGB888, + DRM_FORMAT_BGR888, + DRM_FORMAT_RGB565, + DRM_FORMAT_BGR565, + DRM_FORMAT_ARGB1555, + DRM_FORMAT_ABGR1555, + DRM_FORMAT_RGBA5551, + DRM_FORMAT_BGRA5551, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_XBGR1555, + DRM_FORMAT_RGBX5551, + DRM_FORMAT_BGRX5551, + DRM_FORMAT_ARGB4444, + DRM_FORMAT_ABGR4444, + DRM_FORMAT_RGBA4444, + DRM_FORMAT_BGRA4444, + DRM_FORMAT_XRGB4444, + DRM_FORMAT_XBGR4444, + DRM_FORMAT_RGBX4444, + DRM_FORMAT_BGRX4444, +}; + +static const uint32_t plane_formats_yuv[] = { + DRM_FORMAT_ARGB8888, + DRM_FORMAT_ABGR8888, + DRM_FORMAT_RGBA8888, + DRM_FORMAT_BGRX8888, + DRM_FORMAT_BGRA8888, + DRM_FORMAT_ARGB2101010, + DRM_FORMAT_XRGB2101010, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_RGBX8888, + DRM_FORMAT_RGB888, + DRM_FORMAT_BGR888, + DRM_FORMAT_RGB565, + DRM_FORMAT_BGR565, + DRM_FORMAT_ARGB1555, + DRM_FORMAT_ABGR1555, + DRM_FORMAT_RGBA5551, + DRM_FORMAT_BGRA5551, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_XBGR1555, + DRM_FORMAT_RGBX5551, + DRM_FORMAT_BGRX5551, + DRM_FORMAT_ARGB4444, + DRM_FORMAT_ABGR4444, + DRM_FORMAT_RGBA4444, + DRM_FORMAT_BGRA4444, + DRM_FORMAT_XRGB4444, + DRM_FORMAT_XBGR4444, + DRM_FORMAT_RGBX4444, + DRM_FORMAT_BGRX4444, + + DRM_FORMAT_P010, + DRM_FORMAT_NV12, + DRM_FORMAT_NV21, + DRM_FORMAT_NV16, + DRM_FORMAT_NV61, + DRM_FORMAT_VYUY, + DRM_FORMAT_UYVY, + DRM_FORMAT_YUYV, + DRM_FORMAT_YVYU, + DRM_FORMAT_YUV420, + DRM_FORMAT_YVU420, +}; + +static const u32 rotation_v2_formats[] = { + DRM_FORMAT_NV12, + /* TODO add formats after validation */ +}; + +static const u32 wb2_formats_rgb_yuv[] = { + DRM_FORMAT_RGB565, + DRM_FORMAT_BGR565, + DRM_FORMAT_RGB888, + DRM_FORMAT_ARGB8888, + DRM_FORMAT_RGBA8888, + DRM_FORMAT_ABGR8888, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_RGBX8888, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_ARGB1555, + DRM_FORMAT_RGBA5551, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_RGBX5551, + DRM_FORMAT_ARGB4444, + DRM_FORMAT_RGBA4444, + DRM_FORMAT_RGBX4444, + DRM_FORMAT_XRGB4444, + DRM_FORMAT_BGR888, + DRM_FORMAT_BGRA8888, + DRM_FORMAT_BGRX8888, + DRM_FORMAT_ABGR1555, + DRM_FORMAT_BGRA5551, + DRM_FORMAT_XBGR1555, + DRM_FORMAT_BGRX5551, + DRM_FORMAT_ABGR4444, + DRM_FORMAT_BGRA4444, + DRM_FORMAT_BGRX4444, + DRM_FORMAT_XBGR4444, + DRM_FORMAT_NV12, +}; + +/************************************************************* + * SSPP sub blocks config + *************************************************************/ + +#define SSPP_SCALER_VER(maj, min) (((maj) << 16) | (min)) + +/* SSPP common configuration */ +#define _VIG_SBLK(scaler_ver) \ + { \ + .scaler_blk = {.name = "scaler", \ + .version = scaler_ver, \ + .base = 0xa00, .len = 0xa0,}, \ + .csc_blk = {.name = "csc", \ + .base = 0x1a00, .len = 0x100,}, \ + .format_list = plane_formats_yuv, \ + .num_formats = ARRAY_SIZE(plane_formats_yuv), \ + .rotation_cfg = NULL, \ + } + +/* kaanapali SSPP common configuration */ +#define _VIG_SBLK_REC0_REC1(scaler_ver) \ + { \ + .sspp_rec0_blk = {.name = "sspp_rec0", \ + .base = 0x1000, .len = 0x180,}, \ + .csc_blk = {.name = "csc", \ + .base = 0x1800, .len = 0x100,}, \ + .scaler_blk = {.name = "scaler", \ + .version = scaler_ver, \ + .base = 0x2000, .len = 0xec,}, \ + .sspp_rec1_blk = {.name = "sspp_rec1", \ + .base = 0x3000, .len = 0x180,}, \ + .format_list = plane_formats_yuv, \ + .num_formats = ARRAY_SIZE(plane_formats_yuv), \ + .rotation_cfg = NULL, \ + } + +#define _VIG_SBLK_ROT(scaler_ver, rot_cfg) \ + { \ + .scaler_blk = {.name = "scaler", \ + .version = scaler_ver, \ + .base = 0xa00, .len = 0xa0,}, \ + .csc_blk = {.name = "csc", \ + .base = 0x1a00, .len = 0x100,}, \ + .format_list = plane_formats_yuv, \ + .num_formats = ARRAY_SIZE(plane_formats_yuv), \ + .rotation_cfg = rot_cfg, \ + } + +#define _VIG_SBLK_NOSCALE() \ + { \ + .format_list = plane_formats, \ + .num_formats = ARRAY_SIZE(plane_formats), \ + } + +/* qseed2 is not supported, so disabled scaling */ +#define _VIG_SBLK_QSEED2() \ + { \ + .scaler_blk = {.name = "scaler", \ + /* no version for qseed2 */ \ + .base = 0x200, .len = 0xa0,}, \ + .csc_blk = {.name = "csc", \ + .base = 0x320, .len = 0x100,}, \ + .format_list = plane_formats, \ + .num_formats = ARRAY_SIZE(plane_formats), \ + .rotation_cfg = NULL, \ + } + +#define _RGB_SBLK() \ + { \ + .scaler_blk = {.name = "scaler", \ + .base = 0x200, .len = 0x28,}, \ + .format_list = plane_formats, \ + .num_formats = ARRAY_SIZE(plane_formats), \ + } + +#define _DMA_SBLK() \ + { \ + .format_list = plane_formats, \ + .num_formats = ARRAY_SIZE(plane_formats), \ + } + +#define _DMA_SBLK_V13() \ + { \ + .sspp_rec0_blk = {.name = "sspp_rec0", \ + .base = 0x1000, .len = 0x180,}, \ + .sspp_rec1_blk = {.name = "sspp_rec1", \ + .base = 0x3000, .len = 0x180,}, \ + .format_list = plane_formats, \ + .num_formats = ARRAY_SIZE(plane_formats), \ + } + +static const struct dpu_rotation_cfg dpu_rot_sc7280_cfg_v2 = { + .rot_maxheight = 1088, + .rot_num_formats = ARRAY_SIZE(rotation_v2_formats), + .rot_format_list = rotation_v2_formats, +}; + +static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed2 = + _VIG_SBLK_QSEED2(); + +static const struct dpu_sspp_sub_blks dpu_vig_sblk_noscale = + _VIG_SBLK_NOSCALE(); + +static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed3_1_2 = + _VIG_SBLK(SSPP_SCALER_VER(1, 2)); + +static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed3_1_3 = + _VIG_SBLK(SSPP_SCALER_VER(1, 3)); + +static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed3_1_4 = + _VIG_SBLK(SSPP_SCALER_VER(1, 4)); + +static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed3_2_4 = + _VIG_SBLK(SSPP_SCALER_VER(2, 4)); + +static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed3_3_0 = + _VIG_SBLK(SSPP_SCALER_VER(3, 0)); + +static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed3_3_0_rot_v2 = + _VIG_SBLK_ROT(SSPP_SCALER_VER(3, 0), + &dpu_rot_sc7280_cfg_v2); + +static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed3_3_1 = + _VIG_SBLK(SSPP_SCALER_VER(3, 1)); + +static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed3_3_2 = + _VIG_SBLK(SSPP_SCALER_VER(3, 2)); + +static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed3_3_3 = + _VIG_SBLK(SSPP_SCALER_VER(3, 3)); + +static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed3_3_4 = + _VIG_SBLK(SSPP_SCALER_VER(3, 4)); + +static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed3_3_5 = + _VIG_SBLK_REC0_REC1(SSPP_SCALER_VER(3, 5)); + +static const struct dpu_sspp_sub_blks dpu_rgb_sblk = _RGB_SBLK(); + +static const struct dpu_sspp_sub_blks dpu_dma_sblk = _DMA_SBLK(); + +static const struct dpu_sspp_sub_blks dpu_dma_sblk_v13 = _DMA_SBLK_V13(); + +/************************************************************* + * MIXER sub blocks config + *************************************************************/ + +static const struct dpu_lm_sub_blks msm8998_lm_sblk = { + .maxblendstages = 7, /* excluding base layer */ + .blendstage_base = { /* offsets relative to mixer base */ + 0x20, 0x50, 0x80, 0xb0, 0x230, + 0x260, 0x290 + }, +}; + +static const struct dpu_lm_sub_blks sdm845_lm_sblk = { + .maxblendstages = 11, /* excluding base layer */ + .blendstage_base = { /* offsets relative to mixer base */ + 0x20, 0x38, 0x50, 0x68, 0x80, 0x98, + 0xb0, 0xc8, 0xe0, 0xf8, 0x110 + }, +}; + +static const struct dpu_lm_sub_blks sc7180_lm_sblk = { + .maxblendstages = 7, /* excluding base layer */ + .blendstage_base = { /* offsets relative to mixer base */ + 0x20, 0x38, 0x50, 0x68, 0x80, 0x98, 0xb0 + }, +}; + +static const struct dpu_lm_sub_blks sm8750_lm_sblk = { + .maxblendstages = 11, /* excluding base layer */ + .blendstage_base = { /* offsets relative to mixer base */ + /* 0x40 + n*0x30 */ + 0x40, 0x70, 0xa0, 0xd0, 0x100, 0x130, 0x160, 0x190, 0x1c0, + 0x1f0, 0x220 + }, +}; + +static const struct dpu_lm_sub_blks qcm2290_lm_sblk = { + .maxblendstages = 4, /* excluding base layer */ + .blendstage_base = { /* offsets relative to mixer base */ + 0x20, 0x38, 0x50, 0x68 + }, +}; + +/************************************************************* + * DSPP sub blocks config + *************************************************************/ +static const struct dpu_dspp_sub_blks msm8998_dspp_sblk = { + .pcc = {.name = "pcc", .base = 0x1700, + .len = 0x90, .version = 0x10007}, + .gc = {.name = "gc", .base = 0x17c0, + .len = 0x40, .version = 0x10007}, +}; + +static const struct dpu_dspp_sub_blks sdm845_dspp_sblk = { + .pcc = {.name = "pcc", .base = 0x1700, + .len = 0x90, .version = 0x40000}, + .gc = {.name = "gc", .base = 0x17c0, + .len = 0x40, .version = 0x10008}, +}; + +static const struct dpu_dspp_sub_blks sm8750_dspp_sblk = { + .pcc = {.name = "pcc", .base = 0x1700, + .len = 0x90, .version = 0x60000}, +}; + +/************************************************************* + * PINGPONG sub blocks config + *************************************************************/ + +static const struct dpu_pingpong_sub_blks msm8996_pp_sblk = { + /* No dither block */ +}; + +static const struct dpu_pingpong_sub_blks sdm845_pp_sblk = { + .dither = {.name = "dither", .base = 0x30e0, + .len = 0x20, .version = 0x10000}, +}; + +static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = { + .dither = {.name = "dither", .base = 0xe0, + .len = 0x20, .version = 0x20000}, +}; + +static const struct dpu_pingpong_sub_blks kaanapali_pp_sblk = { + .dither = {.name = "dither", .base = 0xc0, + .len = 0x40, .version = 0x30000}, +}; + +/************************************************************* + * DSC sub blocks config + *************************************************************/ +static const struct dpu_dsc_sub_blks dsc_sblk_0 = { + .enc = {.name = "enc", .base = 0x100, .len = 0x9c}, + .ctl = {.name = "ctl", .base = 0xF00, .len = 0x10}, +}; + +static const struct dpu_dsc_sub_blks dsc_sblk_1 = { + .enc = {.name = "enc", .base = 0x200, .len = 0x9c}, + .ctl = {.name = "ctl", .base = 0xF80, .len = 0x10}, +}; + +static const struct dpu_dsc_sub_blks milos_dsc_sblk_0 = { + .enc = {.name = "enc", .base = 0x100, .len = 0x100}, + .ctl = {.name = "ctl", .base = 0xF00, .len = 0x80}, +}; + +static const struct dpu_dsc_sub_blks milos_dsc_sblk_1 = { + .enc = {.name = "enc", .base = 0x200, .len = 0x100}, + .ctl = {.name = "ctl", .base = 0xF80, .len = 0x80}, +}; + +static const struct dpu_dsc_sub_blks sm8750_dsc_sblk_0 = { + .enc = {.name = "enc", .base = 0x100, .len = 0x100}, + .ctl = {.name = "ctl", .base = 0xF00, .len = 0x24}, +}; + +static const struct dpu_dsc_sub_blks sm8750_dsc_sblk_1 = { + .enc = {.name = "enc", .base = 0x200, .len = 0x100}, + .ctl = {.name = "ctl", .base = 0xF80, .len = 0x24}, +}; + +/************************************************************* + * CDM block config + *************************************************************/ +static const struct dpu_cdm_cfg dpu_cdm_1_x_4_x = { + .name = "cdm_0", + .id = CDM_0, + .len = 0x224, + .base = 0x79200, +}; + +static const struct dpu_cdm_cfg dpu_cdm_5_x = { + .name = "cdm_0", + .id = CDM_0, + .len = 0x228, + .base = 0x79200, +}; + +static const struct dpu_cdm_cfg dpu_cdm_13_x = { + .name = "cdm_0", + .id = CDM_0, + .len = 0x240, + .base = 0x19e000, +}; + +/************************************************************* + * VBIF sub blocks config + *************************************************************/ +/* VBIF QOS remap */ +static const u32 msm8998_rt_pri_lvl[] = {1, 2, 2, 2}; +static const u32 msm8998_nrt_pri_lvl[] = {1, 1, 1, 1}; +static const u32 sdm845_rt_pri_lvl[] = {3, 3, 4, 4, 5, 5, 6, 6}; +static const u32 sdm845_nrt_pri_lvl[] = {3, 3, 3, 3, 3, 3, 3, 3}; +static const u32 sm8650_rt_pri_lvl[] = {4, 4, 5, 5, 5, 5, 5, 6}; + +static const struct dpu_vbif_dynamic_ot_cfg msm8998_ot_rdwr_cfg[] = { + { + .pps = 1920 * 1080 * 30, + .ot_limit = 2, + }, + { + .pps = 1920 * 1080 * 60, + .ot_limit = 4, + }, + { + .pps = 3840 * 2160 * 30, + .ot_limit = 16, + }, +}; + +static const struct dpu_vbif_cfg milos_vbif = { + .len = 0x1074, + .features = BIT(DPU_VBIF_QOS_REMAP), + .xin_halt_timeout = 0x4000, + .qos_rp_remap_size = 0x40, + .qos_rt_tbl = { + .npriority_lvl = ARRAY_SIZE(sdm845_rt_pri_lvl), + .priority_lvl = sdm845_rt_pri_lvl, + }, + .qos_nrt_tbl = { + .npriority_lvl = ARRAY_SIZE(sdm845_nrt_pri_lvl), + .priority_lvl = sdm845_nrt_pri_lvl, + }, + .memtype_count = 16, + .memtype = {3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3}, +}; + +static const struct dpu_vbif_cfg msm8996_vbif = { + .len = 0x1040, + .default_ot_rd_limit = 32, + .default_ot_wr_limit = 16, + .features = BIT(DPU_VBIF_QOS_REMAP) | BIT(DPU_VBIF_QOS_OTLIM), + .xin_halt_timeout = 0x4000, + .qos_rp_remap_size = 0x20, + .dynamic_ot_rd_tbl = { + .count = ARRAY_SIZE(msm8998_ot_rdwr_cfg), + .cfg = msm8998_ot_rdwr_cfg, + }, + .dynamic_ot_wr_tbl = { + .count = ARRAY_SIZE(msm8998_ot_rdwr_cfg), + .cfg = msm8998_ot_rdwr_cfg, + }, + .qos_rt_tbl = { + .npriority_lvl = ARRAY_SIZE(msm8998_rt_pri_lvl), + .priority_lvl = msm8998_rt_pri_lvl, + }, + .qos_nrt_tbl = { + .npriority_lvl = ARRAY_SIZE(msm8998_nrt_pri_lvl), + .priority_lvl = msm8998_nrt_pri_lvl, + }, +}; + +static const struct dpu_vbif_cfg msm8998_vbif = { + .len = 0x1040, + .default_ot_rd_limit = 32, + .default_ot_wr_limit = 32, + .features = BIT(DPU_VBIF_QOS_REMAP) | BIT(DPU_VBIF_QOS_OTLIM), + .xin_halt_timeout = 0x4000, + .qos_rp_remap_size = 0x20, + .dynamic_ot_rd_tbl = { + .count = ARRAY_SIZE(msm8998_ot_rdwr_cfg), + .cfg = msm8998_ot_rdwr_cfg, + }, + .dynamic_ot_wr_tbl = { + .count = ARRAY_SIZE(msm8998_ot_rdwr_cfg), + .cfg = msm8998_ot_rdwr_cfg, + }, + .qos_rt_tbl = { + .npriority_lvl = ARRAY_SIZE(msm8998_rt_pri_lvl), + .priority_lvl = msm8998_rt_pri_lvl, + }, + .qos_nrt_tbl = { + .npriority_lvl = ARRAY_SIZE(msm8998_nrt_pri_lvl), + .priority_lvl = msm8998_nrt_pri_lvl, + }, + .memtype_count = 14, + .memtype = {2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2}, +}; + +static const struct dpu_vbif_cfg sdm845_vbif = { + .len = 0x1040, + .features = BIT(DPU_VBIF_QOS_REMAP), + .xin_halt_timeout = 0x4000, + .qos_rp_remap_size = 0x40, + .qos_rt_tbl = { + .npriority_lvl = ARRAY_SIZE(sdm845_rt_pri_lvl), + .priority_lvl = sdm845_rt_pri_lvl, + }, + .qos_nrt_tbl = { + .npriority_lvl = ARRAY_SIZE(sdm845_nrt_pri_lvl), + .priority_lvl = sdm845_nrt_pri_lvl, + }, + .memtype_count = 14, + .memtype = {3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3}, +}; + +static const struct dpu_vbif_cfg sm8550_vbif = { + .len = 0x1040, + .features = BIT(DPU_VBIF_QOS_REMAP), + .xin_halt_timeout = 0x4000, + .qos_rp_remap_size = 0x40, + .qos_rt_tbl = { + .npriority_lvl = ARRAY_SIZE(sdm845_rt_pri_lvl), + .priority_lvl = sdm845_rt_pri_lvl, + }, + .qos_nrt_tbl = { + .npriority_lvl = ARRAY_SIZE(sdm845_nrt_pri_lvl), + .priority_lvl = sdm845_nrt_pri_lvl, + }, + .memtype_count = 16, + .memtype = {3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3}, +}; + +static const struct dpu_vbif_cfg sm8650_vbif = { + .len = 0x1074, + .features = BIT(DPU_VBIF_QOS_REMAP), + .xin_halt_timeout = 0x4000, + .qos_rp_remap_size = 0x40, + .qos_rt_tbl = { + .npriority_lvl = ARRAY_SIZE(sm8650_rt_pri_lvl), + .priority_lvl = sm8650_rt_pri_lvl, + }, + .qos_nrt_tbl = { + .npriority_lvl = ARRAY_SIZE(sdm845_nrt_pri_lvl), + .priority_lvl = sdm845_nrt_pri_lvl, + }, + .memtype_count = 16, + .memtype = {3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3}, +}; + +/************************************************************* + * PERF data config + *************************************************************/ + +/* SSPP QOS LUTs */ +static const struct dpu_qos_lut_entry msm8998_qos_linear[] = { + {.fl = 4, .lut = 0x1b}, + {.fl = 5, .lut = 0x5b}, + {.fl = 6, .lut = 0x15b}, + {.fl = 7, .lut = 0x55b}, + {.fl = 8, .lut = 0x155b}, + {.fl = 9, .lut = 0x555b}, + {.fl = 10, .lut = 0x1555b}, + {.fl = 11, .lut = 0x5555b}, + {.fl = 12, .lut = 0x15555b}, + {.fl = 0, .lut = 0x55555b} +}; + +static const struct dpu_qos_lut_entry sdm845_qos_linear[] = { + {.fl = 4, .lut = 0x357}, + {.fl = 5, .lut = 0x3357}, + {.fl = 6, .lut = 0x23357}, + {.fl = 7, .lut = 0x223357}, + {.fl = 8, .lut = 0x2223357}, + {.fl = 9, .lut = 0x22223357}, + {.fl = 10, .lut = 0x222223357}, + {.fl = 11, .lut = 0x2222223357}, + {.fl = 12, .lut = 0x22222223357}, + {.fl = 13, .lut = 0x222222223357}, + {.fl = 14, .lut = 0x1222222223357}, + {.fl = 0, .lut = 0x11222222223357} +}; + +static const struct dpu_qos_lut_entry msm8998_qos_macrotile[] = { + {.fl = 10, .lut = 0x1aaff}, + {.fl = 11, .lut = 0x5aaff}, + {.fl = 12, .lut = 0x15aaff}, + {.fl = 0, .lut = 0x55aaff}, +}; + +static const struct dpu_qos_lut_entry sc7180_qos_linear[] = { + {.fl = 0, .lut = 0x0011222222335777}, +}; + +static const struct dpu_qos_lut_entry kaanapali_qos_linear[] = { + {.fl = 0, .lut = 0x0011223344556666}, +}; + +static const struct dpu_qos_lut_entry sm6350_qos_linear_macrotile[] = { + {.fl = 0, .lut = 0x0011223445566777 }, +}; + +static const struct dpu_qos_lut_entry sm8150_qos_linear[] = { + {.fl = 0, .lut = 0x0011222222223357 }, +}; + +static const struct dpu_qos_lut_entry sc8180x_qos_linear[] = { + {.fl = 4, .lut = 0x0000000000000357 }, +}; + +static const struct dpu_qos_lut_entry qcm2290_qos_linear[] = { + {.fl = 0, .lut = 0x0011222222335777}, +}; + +static const struct dpu_qos_lut_entry sdm845_qos_macrotile[] = { + {.fl = 10, .lut = 0x344556677}, + {.fl = 11, .lut = 0x3344556677}, + {.fl = 12, .lut = 0x23344556677}, + {.fl = 13, .lut = 0x223344556677}, + {.fl = 14, .lut = 0x1223344556677}, + {.fl = 0, .lut = 0x112233344556677}, +}; + +static const struct dpu_qos_lut_entry sc7180_qos_macrotile[] = { + {.fl = 0, .lut = 0x0011223344556677}, +}; + +static const struct dpu_qos_lut_entry kaanapali_qos_macrotile[] = { + {.fl = 0, .lut = 0x0011223344556666}, +}; + +static const struct dpu_qos_lut_entry sc8180x_qos_macrotile[] = { + {.fl = 10, .lut = 0x0000000344556677}, +}; + +static const struct dpu_qos_lut_entry msm8998_qos_nrt[] = { + {.fl = 0, .lut = 0x0}, +}; + +static const struct dpu_qos_lut_entry sdm845_qos_nrt[] = { + {.fl = 0, .lut = 0x0}, +}; + +static const struct dpu_qos_lut_entry sc7180_qos_nrt[] = { + {.fl = 0, .lut = 0x0}, +}; + +/************************************************************* + * Hardware catalog + *************************************************************/ + +#include "catalog/dpu_1_7_msm8996.h" +#include "catalog/dpu_1_14_msm8937.h" +#include "catalog/dpu_1_15_msm8917.h" +#include "catalog/dpu_1_16_msm8953.h" + +#include "catalog/dpu_3_0_msm8998.h" +#include "catalog/dpu_3_2_sdm660.h" +#include "catalog/dpu_3_3_sdm630.h" + +#include "catalog/dpu_4_0_sdm845.h" +#include "catalog/dpu_4_1_sdm670.h" + +#include "catalog/dpu_5_0_sm8150.h" +#include "catalog/dpu_5_1_sc8180x.h" +#include "catalog/dpu_5_2_sm7150.h" +#include "catalog/dpu_5_3_sm6150.h" +#include "catalog/dpu_5_4_sm6125.h" + +#include "catalog/dpu_6_0_sm8250.h" +#include "catalog/dpu_6_2_sc7180.h" +#include "catalog/dpu_6_3_sm6115.h" +#include "catalog/dpu_6_4_sm6350.h" +#include "catalog/dpu_6_5_qcm2290.h" +#include "catalog/dpu_6_9_sm6375.h" + +#include "catalog/dpu_7_0_sm8350.h" +#include "catalog/dpu_7_2_sc7280.h" + +#include "catalog/dpu_8_0_sc8280xp.h" +#include "catalog/dpu_8_1_sm8450.h" +#include "catalog/dpu_8_4_sa8775p.h" + +#include "catalog/dpu_9_0_sm8550.h" +#include "catalog/dpu_9_1_sar2130p.h" +#include "catalog/dpu_9_2_x1e80100.h" + +#include "catalog/dpu_10_0_sm8650.h" +#include "catalog/dpu_10_2_milos.h" + +#include "catalog/dpu_12_0_sm8750.h" +#include "catalog/dpu_12_2_glymur.h" +#include "catalog/dpu_12_4_eliza.h" +#include "catalog/dpu_13_0_kaanapali.h" diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h new file mode 100644 index 000000000..f45faf873 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h @@ -0,0 +1,801 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022-2023, Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_HW_CATALOG_H +#define _DPU_HW_CATALOG_H + +#include +#include +#include +#include + +/** + * Max hardware block count: For ex: max 12 SSPP pipes or + * 5 ctl paths. In all cases, it can have max 12 hardware blocks + * based on current design + */ +#define MAX_BLOCKS 12 + +#define DPU_HW_BLK_NAME_LEN 16 + +#define DPU_MAX_IMG_WIDTH 0x3fff +#define DPU_MAX_IMG_HEIGHT 0x3fff + +#define CRTC_DUAL_MIXERS 2 + +#define MAX_XIN_COUNT 16 + +/** + * SSPP sub-blocks/features + * @DPU_SSPP_SCALER_QSEED2, QSEED2 algorithm support + * @DPU_SSPP_SCALER_QSEED3_COMPATIBLE, QSEED3-compatible alogorithm support (includes QSEED3, QSEED3LITE and QSEED4) + * @DPU_SSPP_SCALER_RGB, RGB Scaler, supported by RGB pipes + * @DPU_SSPP_CSC, Support of Color space converion + * @DPU_SSPP_CSC_10BIT, Support of 10-bit Color space conversion + * @DPU_SSPP_CURSOR, SSPP can be used as a cursor layer + * @DPU_SSPP_QOS, SSPP support QoS control, danger/safe/creq + * @DPU_SSPP_EXCL_RECT, SSPP supports exclusion rect + * @DPU_SSPP_SMART_DMA_V1, SmartDMA 1.0 support + * @DPU_SSPP_SMART_DMA_V2, SmartDMA 2.0 support + * @DPU_SSPP_TS_PREFILL Supports prefill with traffic shaper + * @DPU_SSPP_TS_PREFILL_REC1 Supports prefill with traffic shaper multirec + * @DPU_SSPP_CDP Supports client driven prefetch + * @DPU_SSPP_INLINE_ROTATION Support inline rotation + * @DPU_SSPP_MAX maximum value + */ +enum { + DPU_SSPP_SCALER_QSEED2 = 0x1, + DPU_SSPP_SCALER_QSEED3_COMPATIBLE, + DPU_SSPP_SCALER_RGB, + DPU_SSPP_CSC, + DPU_SSPP_CSC_10BIT, + DPU_SSPP_CURSOR, + DPU_SSPP_QOS, + DPU_SSPP_EXCL_RECT, + DPU_SSPP_SMART_DMA_V1, + DPU_SSPP_SMART_DMA_V2, + DPU_SSPP_TS_PREFILL, + DPU_SSPP_TS_PREFILL_REC1, + DPU_SSPP_CDP, + DPU_SSPP_INLINE_ROTATION, + DPU_SSPP_MAX +}; + +/* + * MIXER sub-blocks/features + * @DPU_MIXER_SOURCESPLIT Layer mixer supports source-split configuration + * @DPU_MIXER_MAX maximum value + */ +enum { + DPU_MIXER_SOURCESPLIT = 0x1, + DPU_MIXER_MAX, +}; + +/** + * DSPP sub-blocks + * @DPU_DSPP_PCC Panel color correction block + * @DPU_DSPP_GC Gamma correction block + */ +enum { + DPU_DSPP_PCC = 0x1, + DPU_DSPP_GC, + DPU_DSPP_MAX +}; + +/** + * CTL sub-blocks + * @DPU_CTL_SPLIT_DISPLAY: CTL supports video mode split display + * @DPU_CTL_MAX + */ +enum { + DPU_CTL_SPLIT_DISPLAY = 0x1, + DPU_CTL_MAX +}; + +/** + * WB sub-blocks and features + * @DPU_WB_LINE_MODE Writeback module supports line/linear mode + * @DPU_WB_BLOCK_MODE Writeback module supports block mode read + * @DPU_WB_CHROMA_DOWN, Writeback chroma down block, + * @DPU_WB_DOWNSCALE, Writeback integer downscaler, + * @DPU_WB_DITHER, Dither block + * @DPU_WB_TRAFFIC_SHAPER, Writeback traffic shaper bloc + * @DPU_WB_UBWC, Writeback Universal bandwidth compression + * @DPU_WB_YUV_CONFIG Writeback supports output of YUV colorspace + * @DPU_WB_PIPE_ALPHA Writeback supports pipe alpha + * @DPU_WB_XY_ROI_OFFSET Writeback supports x/y-offset of out ROI in + * the destination image + * @DPU_WB_QOS, Writeback supports QoS control, danger/safe/creq + * @DPU_WB_QOS_8LVL, Writeback supports 8-level QoS control + * @DPU_WB_CDP Writeback supports client driven prefetch + * @DPU_WB_CROP CWB supports cropping + * @DPU_WB_MAX maximum value + */ +enum { + DPU_WB_LINE_MODE = 0x1, + DPU_WB_BLOCK_MODE, + DPU_WB_UBWC, + DPU_WB_YUV_CONFIG, + DPU_WB_PIPE_ALPHA, + DPU_WB_XY_ROI_OFFSET, + DPU_WB_QOS, + DPU_WB_QOS_8LVL, + DPU_WB_CDP, + DPU_WB_CROP, + DPU_WB_MAX +}; + +/** + * VBIF sub-blocks and features + * @DPU_VBIF_QOS_OTLIM VBIF supports OT Limit + * @DPU_VBIF_QOS_REMAP VBIF supports QoS priority remap + * @DPU_VBIF_MAX maximum value + */ +enum { + DPU_VBIF_QOS_OTLIM = 0x1, + DPU_VBIF_QOS_REMAP, + DPU_VBIF_MAX +}; + +/** + * DSC sub-blocks/features + * @DPU_DSC_NATIVE_42x_EN Supports NATIVE_422_EN and NATIVE_420_EN encoding + * @DPU_DSC_MAX + */ +enum { + DPU_DSC_NATIVE_42x_EN = 0x1, + DPU_DSC_MAX +}; + +/** + * MACRO DPU_HW_BLK_INFO - information of HW blocks inside DPU + * @name: string name for debug purposes + * @id: enum identifying this block + * @base: register base offset to mdss + * @len: length of hardware block + */ +#define DPU_HW_BLK_INFO \ + char name[DPU_HW_BLK_NAME_LEN]; \ + u32 id; \ + u32 base; \ + u32 len + +/** + * struct dpu_scaler_blk: Scaler information + * @name: string name for debug purposes + * @base: offset of this sub-block relative to the block offset + * @len: register block length of this sub-block + * @version: qseed block revision, on QSEED3+ platforms this is the value of + * scaler_blk.base + QSEED3_HW_VERSION registers. + */ +struct dpu_scaler_blk { + char name[DPU_HW_BLK_NAME_LEN]; + u32 base; + u32 len; + u32 version; +}; + +struct dpu_csc_blk { + char name[DPU_HW_BLK_NAME_LEN]; + u32 base; + u32 len; +}; + +/** + * struct dpu_pp_blk : Pixel processing sub-blk information + * @name: string name for debug purposes + * @base: offset of this sub-block relative to the block offset + * @len: register block length of this sub-block + * @version: HW Algorithm version + */ +struct dpu_pp_blk { + char name[DPU_HW_BLK_NAME_LEN]; + u32 base; + u32 len; + u32 version; +}; + +/** + * struct dpu_dsc_blk - DSC Encoder sub-blk information + * @name: string name for debug purposes + * @base: offset of this sub-block relative to the block offset + * @len: register block length of this sub-block + */ +struct dpu_dsc_blk { + char name[DPU_HW_BLK_NAME_LEN]; + u32 base; + u32 len; +}; + +/** + * struct dpu_sspp_v13_rec_blk - SSPP REC sub-blk information + * @name: string name for debug purposes + * @base: offset of this sub-block relative to the block offset + * @len: register block length of this sub-block + */ +struct dpu_sspp_v13_rec_blk { + char name[DPU_HW_BLK_NAME_LEN]; + u32 base; + u32 len; +}; + +/** + * enum dpu_qos_lut_usage - define QoS LUT use cases + */ +enum dpu_qos_lut_usage { + DPU_QOS_LUT_USAGE_LINEAR, + DPU_QOS_LUT_USAGE_MACROTILE, + DPU_QOS_LUT_USAGE_NRT, + DPU_QOS_LUT_USAGE_MAX, +}; + +/** + * struct dpu_qos_lut_entry - define QoS LUT table entry + * @fl: fill level, or zero on last entry to indicate default lut + * @lut: lut to use if equal to or less than fill level + */ +struct dpu_qos_lut_entry { + u32 fl; + u64 lut; +}; + +/** + * struct dpu_qos_lut_tbl - define QoS LUT table + * @nentry: number of entry in this table + * @entries: Pointer to table entries + */ +struct dpu_qos_lut_tbl { + u32 nentry; + const struct dpu_qos_lut_entry *entries; +}; + +/** + * struct dpu_rotation_cfg - define inline rotation config + * @rot_maxheight: max pre rotated height allowed for rotation + * @rot_num_formats: number of elements in @rot_format_list + * @rot_format_list: list of supported rotator formats + */ +struct dpu_rotation_cfg { + u32 rot_maxheight; + size_t rot_num_formats; + const u32 *rot_format_list; +}; + +/** + * struct dpu_caps - define DPU capabilities + * @max_mixer_width max layer mixer line width support. + * @max_mixer_blendstages max layer mixer blend stages or + * supported z order + * @has_src_split source split feature status + * @has_dim_layer dim layer feature status + * @has_idle_pc indicate if idle power collapse feature is supported + * @has_3d_merge indicate if 3D merge is supported + * @max_linewidth max linewidth for sspp + * @pixel_ram_size size of latency hiding and de-tiling buffer in bytes + * @max_hdeci_exp max horizontal decimation supported (max is 2^value) + * @max_vdeci_exp max vertical decimation supported (max is 2^value) + */ +struct dpu_caps { + u32 max_mixer_width; + u32 max_mixer_blendstages; + bool has_src_split; + bool has_dim_layer; + bool has_idle_pc; + bool has_3d_merge; + /* SSPP limits */ + u32 max_linewidth; + u32 pixel_ram_size; + u32 max_hdeci_exp; + u32 max_vdeci_exp; +}; + +/** + * struct dpu_sspp_sub_blks : SSPP sub-blocks + * common: Pointer to common configurations shared by sub blocks + * @max_per_pipe_bw: maximum allowable bandwidth of this pipe in kBps + * @qseed_ver: qseed version + * @scaler_blk: + * @csc_blk: + * @format_list: Pointer to list of supported formats + * @num_formats: Number of supported formats + * @dpu_rotation_cfg: inline rotation configuration + */ +struct dpu_sspp_sub_blks { + u32 max_per_pipe_bw; + u32 qseed_ver; + struct dpu_scaler_blk scaler_blk; + struct dpu_pp_blk csc_blk; + struct dpu_sspp_v13_rec_blk sspp_rec0_blk; + struct dpu_sspp_v13_rec_blk sspp_rec1_blk; + + const u32 *format_list; + u32 num_formats; + const struct dpu_rotation_cfg *rotation_cfg; +}; + +/** + * struct dpu_lm_sub_blks: information of mixer block + * @maxwidth: Max pixel width supported by this mixer + * @maxblendstages: Max number of blend-stages supported + * @blendstage_base: Blend-stage register base offset + */ +struct dpu_lm_sub_blks { + u32 maxblendstages; + u32 blendstage_base[MAX_BLOCKS]; +}; + +/** + * struct dpu_dspp_sub_blks: Information of DSPP block + * @pcc: pixel color correction block + * @gc: gamma correction block + */ +struct dpu_dspp_sub_blks { + struct dpu_pp_blk pcc; + struct dpu_pp_blk gc; +}; + +struct dpu_pingpong_sub_blks { + struct dpu_pp_blk dither; +}; + +/** + * struct dpu_dsc_sub_blks - DSC sub-blks + * @enc: DSC encoder sub-block + * @ctl: DSC controller sub-block + */ +struct dpu_dsc_sub_blks { + struct dpu_dsc_blk enc; + struct dpu_dsc_blk ctl; +}; + +/** + * dpu_clk_ctrl_type - Defines top level clock control signals + */ +enum dpu_clk_ctrl_type { + DPU_CLK_CTRL_NONE, + DPU_CLK_CTRL_VIG0, + DPU_CLK_CTRL_VIG1, + DPU_CLK_CTRL_VIG2, + DPU_CLK_CTRL_VIG3, + DPU_CLK_CTRL_VIG4, + DPU_CLK_CTRL_RGB0, + DPU_CLK_CTRL_RGB1, + DPU_CLK_CTRL_RGB2, + DPU_CLK_CTRL_RGB3, + DPU_CLK_CTRL_DMA0, + DPU_CLK_CTRL_DMA1, + DPU_CLK_CTRL_DMA2, + DPU_CLK_CTRL_DMA3, + DPU_CLK_CTRL_DMA4, + DPU_CLK_CTRL_DMA5, + DPU_CLK_CTRL_CURSOR0, + DPU_CLK_CTRL_CURSOR1, + DPU_CLK_CTRL_INLINE_ROT0_SSPP, + DPU_CLK_CTRL_REG_DMA, + DPU_CLK_CTRL_WB2, + DPU_CLK_CTRL_MAX, +}; + +/* struct dpu_clk_ctrl_reg : Clock control register + * @reg_off: register offset + * @bit_off: bit offset + */ +struct dpu_clk_ctrl_reg { + u32 reg_off; + u32 bit_off; +}; + +/* struct dpu_mdp_cfg : MDP TOP-BLK instance info + * @id: index identifying this block + * @base: register base offset to mdss + * @clk_ctrls clock control register definition + */ +struct dpu_mdp_cfg { + DPU_HW_BLK_INFO; + struct dpu_clk_ctrl_reg clk_ctrls[DPU_CLK_CTRL_MAX]; +}; + +/* struct dpu_ctl_cfg : MDP CTL instance info + * @id: index identifying this block + * @base: register base offset to mdss + * @features bit mask identifying sub-blocks/features + * @intr_start: interrupt index for CTL_START + */ +struct dpu_ctl_cfg { + DPU_HW_BLK_INFO; + unsigned long features; + unsigned int intr_start; +}; + +/** + * struct dpu_sspp_cfg - information of source pipes + * @id: index identifying this block + * @base register offset of this block + * @features bit mask identifying sub-blocks/features + * @sblk: SSPP sub-blocks information + * @xin_id: bus client identifier + * @clk_ctrl clock control identifier + * @type sspp type identifier + */ +struct dpu_sspp_cfg { + DPU_HW_BLK_INFO; + unsigned long features; + const struct dpu_sspp_sub_blks *sblk; + u32 xin_id; + enum dpu_clk_ctrl_type clk_ctrl; + u32 type; +}; + +/** + * struct dpu_lm_cfg - information of layer mixer blocks + * @id: index identifying this block + * @base register offset of this block + * @features bit mask identifying sub-blocks/features + * @sblk: LM Sub-blocks information + * @pingpong: ID of connected PingPong, PINGPONG_NONE if unsupported + * @lm_pair: ID of LM that can be controlled by same CTL + */ +struct dpu_lm_cfg { + DPU_HW_BLK_INFO; + unsigned long features; + const struct dpu_lm_sub_blks *sblk; + u32 pingpong; + u32 dspp; + unsigned long lm_pair; +}; + +/** + * struct dpu_dspp_cfg - information of DSPP blocks + * @id enum identifying this block + * @base register offset of this block + * supported by this block + * @sblk sub-blocks information + */ +struct dpu_dspp_cfg { + DPU_HW_BLK_INFO; + const struct dpu_dspp_sub_blks *sblk; +}; + +/** + * struct dpu_pingpong_cfg - information of PING-PONG blocks + * @id enum identifying this block + * @base register offset of this block + * @intr_done: index for PINGPONG done interrupt + * @intr_rdptr: index for PINGPONG readpointer done interrupt + * @sblk sub-blocks information + */ +struct dpu_pingpong_cfg { + DPU_HW_BLK_INFO; + u32 merge_3d; + unsigned int intr_done; + unsigned int intr_rdptr; + const struct dpu_pingpong_sub_blks *sblk; +}; + +/** + * struct dpu_merge_3d_cfg - information of DSPP blocks + * @id enum identifying this block + * @base register offset of this block + * @sblk sub-blocks information + */ +struct dpu_merge_3d_cfg { + DPU_HW_BLK_INFO; + const struct dpu_merge_3d_sub_blks *sblk; +}; + +/** + * struct dpu_dsc_cfg - information of DSC blocks + * @id enum identifying this block + * @base register offset of this block + * @len: length of hardware block + * @features bit mask identifying sub-blocks/features + * @sblk: sub-blocks information + */ +struct dpu_dsc_cfg { + DPU_HW_BLK_INFO; + unsigned long features; + const struct dpu_dsc_sub_blks *sblk; +}; + +/** + * struct dpu_intf_cfg - information of timing engine blocks + * @id enum identifying this block + * @base register offset of this block + * @type: Interface type(DSI, DP, HDMI) + * @controller_id: Controller Instance ID in case of multiple of intf type + * @prog_fetch_lines_worst_case Worst case latency num lines needed to prefetch + * @intr_underrun: index for INTF underrun interrupt + * @intr_vsync: index for INTF VSYNC interrupt + * @intr_tear_rd_ptr: Index for INTF TEAR_RD_PTR interrupt + */ +struct dpu_intf_cfg { + DPU_HW_BLK_INFO; + u32 type; /* interface type*/ + u32 controller_id; + u32 prog_fetch_lines_worst_case; + unsigned int intr_underrun; + unsigned int intr_vsync; + unsigned int intr_tear_rd_ptr; +}; + +/** + * struct dpu_wb_cfg - information of writeback blocks + * @DPU_HW_BLK_INFO: refer to the description above for DPU_HW_BLK_INFO + * @maxlinewidth: max line width supported by writeback block + * @xin_id: bus client identifier + * @intr_wb_done: interrupt index for WB_DONE + * @format_list: list of formats supported by this writeback block + * @num_formats: number of formats supported by this writeback block + * @clk_ctrl: clock control identifier + */ +struct dpu_wb_cfg { + DPU_HW_BLK_INFO; + unsigned long features; + u32 maxlinewidth; + u32 xin_id; + unsigned int intr_wb_done; + const u32 *format_list; + u32 num_formats; + enum dpu_clk_ctrl_type clk_ctrl; +}; + +/* + * struct dpu_cwb_cfg : MDP CWB mux instance info + * @id: enum identifying this block + * @base: register base offset to mdss + * @features bit mask identifying sub-blocks/features + */ +struct dpu_cwb_cfg { + DPU_HW_BLK_INFO; +}; + +/** + * struct dpu_vbif_dynamic_ot_cfg - dynamic OT setting + * @pps pixel per seconds + * @ot_limit OT limit to use up to specified pixel per second + */ +struct dpu_vbif_dynamic_ot_cfg { + u64 pps; + u32 ot_limit; +}; + +/** + * struct dpu_vbif_dynamic_ot_tbl - dynamic OT setting table + * @count length of cfg + * @cfg pointer to array of configuration settings with + * ascending requirements + */ +struct dpu_vbif_dynamic_ot_tbl { + u32 count; + const struct dpu_vbif_dynamic_ot_cfg *cfg; +}; + +/** + * struct dpu_vbif_qos_tbl - QoS priority table + * @npriority_lvl num of priority level + * @priority_lvl pointer to array of priority level in ascending order + */ +struct dpu_vbif_qos_tbl { + u32 npriority_lvl; + const u32 *priority_lvl; +}; + +/** + * struct dpu_vbif_cfg - information of VBIF blocks + * @len: length of hardware block + * @features bit mask identifying sub-blocks/features + * @ot_rd_limit default OT read limit + * @ot_wr_limit default OT write limit + * @xin_halt_timeout maximum time (in usec) for xin to halt + * @qos_rp_remap_size size of VBIF_XINL_QOS_RP_REMAP register space + * @dynamic_ot_rd_tbl dynamic OT read configuration table + * @dynamic_ot_wr_tbl dynamic OT write configuration table + * @qos_rt_tbl real-time QoS priority table + * @qos_nrt_tbl non-real-time QoS priority table + * @memtype_count number of defined memtypes + * @memtype array of xin memtype definitions + */ +struct dpu_vbif_cfg { + u32 len; + unsigned long features; + u32 default_ot_rd_limit; + u32 default_ot_wr_limit; + u32 xin_halt_timeout; + u32 qos_rp_remap_size; + struct dpu_vbif_dynamic_ot_tbl dynamic_ot_rd_tbl; + struct dpu_vbif_dynamic_ot_tbl dynamic_ot_wr_tbl; + struct dpu_vbif_qos_tbl qos_rt_tbl; + struct dpu_vbif_qos_tbl qos_nrt_tbl; + u32 memtype_count; + u32 memtype[MAX_XIN_COUNT]; +}; + +/** + * struct dpu_cdm_cfg - information of chroma down blocks + * @name string name for debug purposes + * @id enum identifying this block + * @base register offset of this block + */ +struct dpu_cdm_cfg { + DPU_HW_BLK_INFO; +}; + +/** + * Define CDP use cases + * @DPU_PERF_CDP_UDAGE_RT: real-time use cases + * @DPU_PERF_CDP_USAGE_NRT: non real-time use cases such as WFD + */ +enum { + DPU_PERF_CDP_USAGE_RT, + DPU_PERF_CDP_USAGE_NRT, + DPU_PERF_CDP_USAGE_MAX +}; + +/** + * struct dpu_perf_cdp_cfg - define CDP use case configuration + * @rd_enable: true if read pipe CDP is enabled + * @wr_enable: true if write pipe CDP is enabled + */ +struct dpu_perf_cdp_cfg { + bool rd_enable; + bool wr_enable; +}; + +/** + * struct dpu_mdss_version - DPU's major and minor versions + * @core_major_ver: DPU core's major version + * @core_minor_ver: DPU core's minor version + */ +struct dpu_mdss_version { + u8 core_major_ver; + u8 core_minor_ver; +}; + +/** + * struct dpu_perf_cfg - performance control settings + * @max_bw_low low threshold of maximum bandwidth (kbps) + * @max_bw_high high threshold of maximum bandwidth (kbps) + * @min_core_ib minimum bandwidth for core (kbps) + * @min_core_ib minimum mnoc ib vote in kbps + * @min_llcc_ib minimum llcc ib vote in kbps + * @min_dram_ib minimum dram ib vote in kbps + * @undersized_prefill_lines undersized prefill in lines + * @xtra_prefill_lines extra prefill latency in lines + * @dest_scale_prefill_lines destination scaler latency in lines + * @macrotile_perfill_lines macrotile latency in lines + * @yuv_nv12_prefill_lines yuv_nv12 latency in lines + * @linear_prefill_lines linear latency in lines + * @downscaling_prefill_lines downscaling latency in lines + * @amortizable_theshold minimum y position for traffic shaping prefill + * @min_prefill_lines minimum pipeline latency in lines + * @clk_inefficiency_factor DPU src clock inefficiency factor + * @bw_inefficiency_factor DPU axi bus bw inefficiency factor + * @safe_lut_tbl: LUT tables for safe signals + * @danger_lut_tbl: LUT tables for danger signals + * @qos_lut_tbl: LUT tables for QoS signals + * @cdp_cfg cdp use case configurations + */ +struct dpu_perf_cfg { + u32 max_bw_low; + u32 max_bw_high; + u32 min_core_ib; + u32 min_llcc_ib; + u32 min_dram_ib; + u32 undersized_prefill_lines; + u32 xtra_prefill_lines; + u32 dest_scale_prefill_lines; + u32 macrotile_prefill_lines; + u32 yuv_nv12_prefill_lines; + u32 linear_prefill_lines; + u32 downscaling_prefill_lines; + u32 amortizable_threshold; + u32 min_prefill_lines; + u32 clk_inefficiency_factor; + u32 bw_inefficiency_factor; + u32 safe_lut_tbl[DPU_QOS_LUT_USAGE_MAX]; + u32 danger_lut_tbl[DPU_QOS_LUT_USAGE_MAX]; + struct dpu_qos_lut_tbl qos_lut_tbl[DPU_QOS_LUT_USAGE_MAX]; + struct dpu_perf_cdp_cfg cdp_cfg[DPU_PERF_CDP_USAGE_MAX]; +}; + +/** + * struct dpu_mdss_cfg - information of MDSS HW + * This is the main catalog data structure representing + * this HW version. Contains dpu's major and minor versions, + * number of instances, register offsets, capabilities of the + * all MDSS HW sub-blocks. + * + * @dma_formats Supported formats for dma pipe + * @cursor_formats Supported formats for cursor pipe + * @vig_formats Supported formats for vig pipe + */ +struct dpu_mdss_cfg { + const struct dpu_mdss_version *mdss_ver; + + const struct dpu_caps *caps; + + const struct dpu_mdp_cfg *mdp; + + u32 ctl_count; + const struct dpu_ctl_cfg *ctl; + + u32 sspp_count; + const struct dpu_sspp_cfg *sspp; + + u32 mixer_count; + const struct dpu_lm_cfg *mixer; + + u32 pingpong_count; + const struct dpu_pingpong_cfg *pingpong; + + u32 merge_3d_count; + const struct dpu_merge_3d_cfg *merge_3d; + + u32 dsc_count; + const struct dpu_dsc_cfg *dsc; + + u32 intf_count; + const struct dpu_intf_cfg *intf; + + const struct dpu_vbif_cfg *vbif; + + u32 wb_count; + const struct dpu_wb_cfg *wb; + + const struct dpu_cdm_cfg *cdm; + + u32 ad_count; + + u32 dspp_count; + const struct dpu_dspp_cfg *dspp; + + u32 cwb_count; + const struct dpu_cwb_cfg *cwb; + + /* Add additional block data structures here */ + + const struct dpu_perf_cfg *perf; + const struct dpu_format_extended *dma_formats; + const struct dpu_format_extended *cursor_formats; + const struct dpu_format_extended *vig_formats; +}; + +extern const struct dpu_mdss_cfg dpu_eliza_cfg; +extern const struct dpu_mdss_cfg dpu_glymur_cfg; +extern const struct dpu_mdss_cfg dpu_kaanapali_cfg; +extern const struct dpu_mdss_cfg dpu_milos_cfg; +extern const struct dpu_mdss_cfg dpu_msm8917_cfg; +extern const struct dpu_mdss_cfg dpu_msm8937_cfg; +extern const struct dpu_mdss_cfg dpu_msm8953_cfg; +extern const struct dpu_mdss_cfg dpu_msm8996_cfg; +extern const struct dpu_mdss_cfg dpu_msm8998_cfg; +extern const struct dpu_mdss_cfg dpu_sar2130p_cfg; +extern const struct dpu_mdss_cfg dpu_sdm630_cfg; +extern const struct dpu_mdss_cfg dpu_sdm660_cfg; +extern const struct dpu_mdss_cfg dpu_sdm845_cfg; +extern const struct dpu_mdss_cfg dpu_sdm670_cfg; +extern const struct dpu_mdss_cfg dpu_sm8150_cfg; +extern const struct dpu_mdss_cfg dpu_sc8180x_cfg; +extern const struct dpu_mdss_cfg dpu_sm7150_cfg; +extern const struct dpu_mdss_cfg dpu_sm8250_cfg; +extern const struct dpu_mdss_cfg dpu_sc7180_cfg; +extern const struct dpu_mdss_cfg dpu_sm6115_cfg; +extern const struct dpu_mdss_cfg dpu_sm6125_cfg; +extern const struct dpu_mdss_cfg dpu_sm6150_cfg; +extern const struct dpu_mdss_cfg dpu_sm6350_cfg; +extern const struct dpu_mdss_cfg dpu_qcm2290_cfg; +extern const struct dpu_mdss_cfg dpu_sm6375_cfg; +extern const struct dpu_mdss_cfg dpu_sm8350_cfg; +extern const struct dpu_mdss_cfg dpu_sc7280_cfg; +extern const struct dpu_mdss_cfg dpu_sc8280xp_cfg; +extern const struct dpu_mdss_cfg dpu_sm8450_cfg; +extern const struct dpu_mdss_cfg dpu_sa8775p_cfg; +extern const struct dpu_mdss_cfg dpu_sm8550_cfg; +extern const struct dpu_mdss_cfg dpu_sm8650_cfg; +extern const struct dpu_mdss_cfg dpu_sm8750_cfg; +extern const struct dpu_mdss_cfg dpu_x1e80100_cfg; + +#endif /* _DPU_HW_CATALOG_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cdm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cdm.c new file mode 100644 index 000000000..3f88c3641 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cdm.c @@ -0,0 +1,257 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2023, The Linux Foundation. All rights reserved. + */ + +#include + +#include + +#include "dpu_hw_mdss.h" +#include "dpu_hw_util.h" +#include "dpu_hw_catalog.h" +#include "dpu_hw_cdm.h" +#include "dpu_kms.h" + +#define CDM_CSC_10_OPMODE 0x000 +#define CDM_CSC_10_BASE 0x004 + +#define CDM_CDWN2_OP_MODE 0x100 +#define CDM_CDWN2_CLAMP_OUT 0x104 +#define CDM_CDWN2_PARAMS_3D_0 0x108 +#define CDM_CDWN2_PARAMS_3D_1 0x10C +#define CDM_CDWN2_COEFF_COSITE_H_0 0x110 +#define CDM_CDWN2_COEFF_COSITE_H_1 0x114 +#define CDM_CDWN2_COEFF_COSITE_H_2 0x118 +#define CDM_CDWN2_COEFF_OFFSITE_H_0 0x11C +#define CDM_CDWN2_COEFF_OFFSITE_H_1 0x120 +#define CDM_CDWN2_COEFF_OFFSITE_H_2 0x124 +#define CDM_CDWN2_COEFF_COSITE_V 0x128 +#define CDM_CDWN2_COEFF_OFFSITE_V 0x12C +#define CDM_CDWN2_OUT_SIZE 0x130 + +#define CDM_HDMI_PACK_OP_MODE 0x200 +#define CDM_CSC_10_MATRIX_COEFF_0 0x004 + +#define CDM_MUX 0x224 + +/* CDM CDWN2 sub-block bit definitions */ +#define CDM_CDWN2_OP_MODE_EN BIT(0) +#define CDM_CDWN2_OP_MODE_ENABLE_H BIT(1) +#define CDM_CDWN2_OP_MODE_ENABLE_V BIT(2) +#define CDM_CDWN2_OP_MODE_BITS_OUT_8BIT BIT(7) +#define CDM_CDWN2_V_PIXEL_METHOD_MASK GENMASK(6, 5) +#define CDM_CDWN2_H_PIXEL_METHOD_MASK GENMASK(4, 3) + +/* CDM CSC10 sub-block bit definitions */ +#define CDM_CSC10_OP_MODE_EN BIT(0) +#define CDM_CSC10_OP_MODE_SRC_FMT_YUV BIT(1) +#define CDM_CSC10_OP_MODE_DST_FMT_YUV BIT(2) + +/* CDM HDMI pack sub-block bit definitions */ +#define CDM_HDMI_PACK_OP_MODE_EN BIT(0) + +/* + * Horizontal coefficients for cosite chroma downscale + * s13 representation of coefficients + */ +static u32 cosite_h_coeff[] = {0x00000016, 0x000001cc, 0x0100009e}; + +/* + * Horizontal coefficients for offsite chroma downscale + */ +static u32 offsite_h_coeff[] = {0x000b0005, 0x01db01eb, 0x00e40046}; + +/* + * Vertical coefficients for cosite chroma downscale + */ +static u32 cosite_v_coeff[] = {0x00080004}; +/* + * Vertical coefficients for offsite chroma downscale + */ +static u32 offsite_v_coeff[] = {0x00060002}; + +static int dpu_hw_cdm_setup_cdwn(struct dpu_hw_cdm *ctx, struct dpu_hw_cdm_cfg *cfg) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + u32 opmode; + u32 out_size; + + switch (cfg->h_cdwn_type) { + case CDM_CDWN_DISABLE: + opmode = 0; + break; + case CDM_CDWN_PIXEL_DROP: + opmode = CDM_CDWN2_OP_MODE_ENABLE_H | + FIELD_PREP(CDM_CDWN2_H_PIXEL_METHOD_MASK, + CDM_CDWN2_METHOD_PIXEL_DROP); + break; + case CDM_CDWN_AVG: + opmode = CDM_CDWN2_OP_MODE_ENABLE_H | + FIELD_PREP(CDM_CDWN2_H_PIXEL_METHOD_MASK, + CDM_CDWN2_METHOD_AVG); + break; + case CDM_CDWN_COSITE: + opmode = CDM_CDWN2_OP_MODE_ENABLE_H | + FIELD_PREP(CDM_CDWN2_H_PIXEL_METHOD_MASK, + CDM_CDWN2_METHOD_COSITE); + DPU_REG_WRITE(c, CDM_CDWN2_COEFF_COSITE_H_0, + cosite_h_coeff[0]); + DPU_REG_WRITE(c, CDM_CDWN2_COEFF_COSITE_H_1, + cosite_h_coeff[1]); + DPU_REG_WRITE(c, CDM_CDWN2_COEFF_COSITE_H_2, + cosite_h_coeff[2]); + break; + case CDM_CDWN_OFFSITE: + opmode = CDM_CDWN2_OP_MODE_ENABLE_H | + FIELD_PREP(CDM_CDWN2_H_PIXEL_METHOD_MASK, CDM_CDWN2_METHOD_OFFSITE); + DPU_REG_WRITE(c, CDM_CDWN2_COEFF_OFFSITE_H_0, + offsite_h_coeff[0]); + DPU_REG_WRITE(c, CDM_CDWN2_COEFF_OFFSITE_H_1, + offsite_h_coeff[1]); + DPU_REG_WRITE(c, CDM_CDWN2_COEFF_OFFSITE_H_2, + offsite_h_coeff[2]); + break; + default: + DPU_ERROR("%s invalid horz down sampling type\n", __func__); + return -EINVAL; + } + + switch (cfg->v_cdwn_type) { + case CDM_CDWN_DISABLE: + /* if its only Horizontal downsample, we dont need to do anything here */ + break; + case CDM_CDWN_PIXEL_DROP: + opmode |= CDM_CDWN2_OP_MODE_ENABLE_V | + FIELD_PREP(CDM_CDWN2_V_PIXEL_METHOD_MASK, + CDM_CDWN2_METHOD_PIXEL_DROP); + break; + case CDM_CDWN_AVG: + opmode |= CDM_CDWN2_OP_MODE_ENABLE_V | + FIELD_PREP(CDM_CDWN2_V_PIXEL_METHOD_MASK, + CDM_CDWN2_METHOD_AVG); + break; + case CDM_CDWN_COSITE: + opmode |= CDM_CDWN2_OP_MODE_ENABLE_V | + FIELD_PREP(CDM_CDWN2_V_PIXEL_METHOD_MASK, + CDM_CDWN2_METHOD_COSITE); + DPU_REG_WRITE(c, + CDM_CDWN2_COEFF_COSITE_V, + cosite_v_coeff[0]); + break; + case CDM_CDWN_OFFSITE: + opmode |= CDM_CDWN2_OP_MODE_ENABLE_V | + FIELD_PREP(CDM_CDWN2_V_PIXEL_METHOD_MASK, + CDM_CDWN2_METHOD_OFFSITE); + DPU_REG_WRITE(c, + CDM_CDWN2_COEFF_OFFSITE_V, + offsite_v_coeff[0]); + break; + default: + return -EINVAL; + } + + if (cfg->output_bit_depth != CDM_CDWN_OUTPUT_10BIT) + opmode |= CDM_CDWN2_OP_MODE_BITS_OUT_8BIT; + + if (cfg->v_cdwn_type || cfg->h_cdwn_type) + opmode |= CDM_CDWN2_OP_MODE_EN; /* EN CDWN module */ + else + opmode &= ~CDM_CDWN2_OP_MODE_EN; + + out_size = (cfg->output_width & 0xFFFF) | ((cfg->output_height & 0xFFFF) << 16); + DPU_REG_WRITE(c, CDM_CDWN2_OUT_SIZE, out_size); + DPU_REG_WRITE(c, CDM_CDWN2_OP_MODE, opmode); + DPU_REG_WRITE(c, CDM_CDWN2_CLAMP_OUT, ((0x3FF << 16) | 0x0)); + + return 0; +} + +static int dpu_hw_cdm_enable(struct dpu_hw_cdm *ctx, struct dpu_hw_cdm_cfg *cdm) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + const struct msm_format *fmt; + u32 opmode = 0; + u32 csc = 0; + + if (!ctx || !cdm) + return -EINVAL; + + fmt = cdm->output_fmt; + + if (!MSM_FORMAT_IS_YUV(fmt)) + return -EINVAL; + + dpu_hw_csc_setup(&ctx->hw, CDM_CSC_10_MATRIX_COEFF_0, cdm->csc_cfg, true); + dpu_hw_cdm_setup_cdwn(ctx, cdm); + + if (cdm->output_type == CDM_CDWN_OUTPUT_HDMI) { + if (fmt->chroma_sample == CHROMA_H1V2) + return -EINVAL; /*unsupported format */ + opmode = CDM_HDMI_PACK_OP_MODE_EN; + opmode |= (fmt->chroma_sample << 1); + } + + csc |= CDM_CSC10_OP_MODE_DST_FMT_YUV; + csc &= ~CDM_CSC10_OP_MODE_SRC_FMT_YUV; + csc |= CDM_CSC10_OP_MODE_EN; + + if (ctx && ctx->ops.bind_pingpong_blk) + ctx->ops.bind_pingpong_blk(ctx, cdm->pp_id); + + DPU_REG_WRITE(c, CDM_CSC_10_OPMODE, csc); + DPU_REG_WRITE(c, CDM_HDMI_PACK_OP_MODE, opmode); + return 0; +} + +static void dpu_hw_cdm_bind_pingpong_blk(struct dpu_hw_cdm *ctx, const enum dpu_pingpong pp) +{ + struct dpu_hw_blk_reg_map *c; + int mux_cfg; + + c = &ctx->hw; + + mux_cfg = DPU_REG_READ(c, CDM_MUX); + mux_cfg &= ~0xf; + + if (pp >= PINGPONG_CWB_0) + mux_cfg |= 0xd; + else if (pp) + mux_cfg |= (pp - PINGPONG_0) & 0x7; + else + mux_cfg |= 0xf; + + DPU_REG_WRITE(c, CDM_MUX, mux_cfg); +} + +/** + * dpu_hw_cdm_init - initializes the cdm hw driver object. + * should be called once before accessing every cdm. + * @dev: DRM device handle + * @cfg: CDM catalog entry for which driver object is required + * @addr : mapped register io address of MDSS + * @mdss_rev: mdss hw core revision + */ +struct dpu_hw_cdm *dpu_hw_cdm_init(struct drm_device *dev, + const struct dpu_cdm_cfg *cfg, void __iomem *addr, + const struct dpu_mdss_version *mdss_rev) +{ + struct dpu_hw_cdm *c; + + c = drmm_kzalloc(dev, sizeof(*c), GFP_KERNEL); + if (!c) + return ERR_PTR(-ENOMEM); + + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_CDM; + + /* Assign ops */ + c->idx = cfg->id; + c->caps = cfg; + + c->ops.enable = dpu_hw_cdm_enable; + if (mdss_rev->core_major_ver >= 5) + c->ops.bind_pingpong_blk = dpu_hw_cdm_bind_pingpong_blk; + + return c; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cdm.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cdm.h new file mode 100644 index 000000000..75e6dae0f --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cdm.h @@ -0,0 +1,134 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2023, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_HW_CDM_H +#define _DPU_HW_CDM_H + +#include "dpu_hw_mdss.h" +#include "dpu_hw_top.h" + +struct dpu_hw_cdm; + +/** + * struct dpu_hw_cdm_cfg : current configuration of CDM block + * + * @output_width: output ROI width of CDM block + * @output_height: output ROI height of CDM block + * @output_bit_depth: output bit-depth of CDM block + * @h_cdwn_type: downsample type used for horizontal pixels + * @v_cdwn_type: downsample type used for vertical pixels + * @output_fmt: handle to msm_format of CDM block + * @csc_cfg: handle to CSC matrix programmed for CDM block + * @output_type: interface to which CDM is paired (HDMI/WB) + * @pp_id: ping-pong block to which CDM is bound to + */ +struct dpu_hw_cdm_cfg { + u32 output_width; + u32 output_height; + u32 output_bit_depth; + u32 h_cdwn_type; + u32 v_cdwn_type; + const struct msm_format *output_fmt; + const struct dpu_csc_cfg *csc_cfg; + u32 output_type; + int pp_id; +}; + +/* + * These values are used indicate which type of downsample is used + * in the horizontal/vertical direction for the CDM block. + */ +enum dpu_hw_cdwn_type { + CDM_CDWN_DISABLE, + CDM_CDWN_PIXEL_DROP, + CDM_CDWN_AVG, + CDM_CDWN_COSITE, + CDM_CDWN_OFFSITE, +}; + +/* + * CDM block can be paired with WB or HDMI block. These values match + * the input with which the CDM block is paired. + */ +enum dpu_hw_cdwn_output_type { + CDM_CDWN_OUTPUT_HDMI, + CDM_CDWN_OUTPUT_WB, +}; + +/* + * CDM block can give an 8-bit or 10-bit output. These values + * are used to indicate the output bit depth of CDM block + */ +enum dpu_hw_cdwn_output_bit_depth { + CDM_CDWN_OUTPUT_8BIT, + CDM_CDWN_OUTPUT_10BIT, +}; + +/* + * CDM block can downsample using different methods. These values + * are used to indicate the downsample method which can be used + * either in the horizontal or vertical direction. + */ +enum dpu_hw_cdwn_op_mode_method_h_v { + CDM_CDWN2_METHOD_PIXEL_DROP, + CDM_CDWN2_METHOD_AVG, + CDM_CDWN2_METHOD_COSITE, + CDM_CDWN2_METHOD_OFFSITE +}; + +/** + * struct dpu_hw_cdm_ops : Interface to the chroma down Hw driver functions + * Assumption is these functions will be called after + * clocks are enabled + * @enable: Enables the output to interface and programs the + * output packer + * @bind_pingpong_blk: enable/disable the connection with pingpong which + * will feed pixels to this cdm + */ +struct dpu_hw_cdm_ops { + /** + * @enable: Enable the CDM module + * @cdm Pointer to chroma down context + */ + int (*enable)(struct dpu_hw_cdm *cdm, struct dpu_hw_cdm_cfg *cfg); + + /** + * @bind_pingpong_blk: Enable/disable the connection with pingpong + * @cdm Pointer to chroma down context + * @pp pingpong block id. + */ + void (*bind_pingpong_blk)(struct dpu_hw_cdm *cdm, const enum dpu_pingpong pp); +}; + +/** + * struct dpu_hw_cdm - cdm description + * @base: Hardware block base structure + * @hw: Block hardware details + * @idx: CDM index + * @caps: Pointer to cdm_cfg + * @ops: handle to operations possible for this CDM + */ +struct dpu_hw_cdm { + struct dpu_hw_blk base; + struct dpu_hw_blk_reg_map hw; + + /* chroma down */ + const struct dpu_cdm_cfg *caps; + enum dpu_cdm idx; + + /* ops */ + struct dpu_hw_cdm_ops ops; +}; + +struct dpu_hw_cdm *dpu_hw_cdm_init(struct drm_device *dev, + const struct dpu_cdm_cfg *cdm, void __iomem *addr, + const struct dpu_mdss_version *mdss_rev); + +static inline struct dpu_hw_cdm *to_dpu_hw_cdm(struct dpu_hw_blk *hw) +{ + return container_of(hw, struct dpu_hw_cdm, base); +} + +#endif /*_DPU_HW_CDM_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c new file mode 100644 index 000000000..e8729fe1c --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c @@ -0,0 +1,874 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + * Copyright (c) 2022-2024 Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#include + +#include + +#include "dpu_hwio.h" +#include "dpu_hw_ctl.h" +#include "dpu_kms.h" +#include "dpu_trace.h" + +#define CTL_LAYER(lm) \ + (((lm) == LM_5) ? (0x024) : (((lm) - LM_0) * 0x004)) +#define CTL_LAYER_EXT(lm) \ + (0x40 + (((lm) - LM_0) * 0x004)) +#define CTL_LAYER_EXT2(lm) \ + (0x70 + (((lm) - LM_0) * 0x004)) +#define CTL_LAYER_EXT3(lm) \ + (0xA0 + (((lm) - LM_0) * 0x004)) +#define CTL_LAYER_EXT4(lm) \ + (0xB8 + (((lm) - LM_0) * 0x004)) +#define CTL_TOP 0x014 +#define CTL_FLUSH 0x018 +#define CTL_START 0x01C +#define CTL_PREPARE 0x0d0 +#define CTL_SW_RESET 0x030 +#define CTL_LAYER_EXTN_OFFSET 0x40 +#define CTL_MERGE_3D_ACTIVE 0x0E4 +#define CTL_DSC_ACTIVE 0x0E8 +#define CTL_WB_ACTIVE 0x0EC +#define CTL_CWB_ACTIVE 0x0F0 +#define CTL_INTF_ACTIVE 0x0F4 +#define CTL_CDM_ACTIVE 0x0F8 +#define CTL_FETCH_PIPE_ACTIVE 0x0FC +#define CTL_MERGE_3D_FLUSH 0x100 +#define CTL_DSC_FLUSH 0x104 +#define CTL_WB_FLUSH 0x108 +#define CTL_CWB_FLUSH 0x10C +#define CTL_INTF_FLUSH 0x110 +#define CTL_CDM_FLUSH 0x114 +#define CTL_PERIPH_FLUSH 0x128 +#define CTL_PIPE_ACTIVE 0x12c +#define CTL_LAYER_ACTIVE 0x130 +#define CTL_INTF_MASTER 0x134 +#define CTL_DSPP_n_FLUSH(n) ((0x13C) + ((n) * 4)) + +#define CTL_MIXER_BORDER_OUT BIT(24) +#define CTL_FLUSH_MASK_CTL BIT(17) + +#define DPU_REG_RESET_TIMEOUT_US 2000 +#define MERGE_3D_IDX 23 +#define DSC_IDX 22 +#define CDM_IDX 26 +#define PERIPH_IDX 30 +#define INTF_IDX 31 +#define WB_IDX 16 +#define CWB_IDX 28 +#define DSPP_IDX 29 /* From DPU hw rev 7.x.x */ +#define CTL_INVALID_BIT 0xffff +#define CTL_DEFAULT_GROUP_ID 0xf + +static const u32 fetch_tbl[SSPP_MAX] = {CTL_INVALID_BIT, 16, 17, 18, 19, + CTL_INVALID_BIT, CTL_INVALID_BIT, CTL_INVALID_BIT, CTL_INVALID_BIT, 0, + 1, 2, 3, 4, 5}; + +static const u32 lm_tbl[LM_MAX] = {CTL_INVALID_BIT, 0, 1, 2, 3, 4, 5, 6, 7}; + +static int _mixer_stages(const struct dpu_lm_cfg *mixer, int count, + enum dpu_lm lm) +{ + int i; + int stages = -EINVAL; + + for (i = 0; i < count; i++) { + if (lm == mixer[i].id) { + stages = mixer[i].sblk->maxblendstages; + break; + } + } + + return stages; +} + +static inline u32 dpu_hw_ctl_get_flush_register(struct dpu_hw_ctl *ctx) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + + return DPU_REG_READ(c, CTL_FLUSH); +} + +static inline void dpu_hw_ctl_trigger_start(struct dpu_hw_ctl *ctx) +{ + trace_dpu_hw_ctl_trigger_start(ctx->pending_flush_mask, + dpu_hw_ctl_get_flush_register(ctx)); + DPU_REG_WRITE(&ctx->hw, CTL_START, 0x1); +} + +static inline bool dpu_hw_ctl_is_started(struct dpu_hw_ctl *ctx) +{ + return !!(DPU_REG_READ(&ctx->hw, CTL_START) & BIT(0)); +} + +static inline void dpu_hw_ctl_trigger_pending(struct dpu_hw_ctl *ctx) +{ + trace_dpu_hw_ctl_trigger_prepare(ctx->pending_flush_mask, + dpu_hw_ctl_get_flush_register(ctx)); + DPU_REG_WRITE(&ctx->hw, CTL_PREPARE, 0x1); +} + +static inline void dpu_hw_ctl_clear_pending_flush(struct dpu_hw_ctl *ctx) +{ + trace_dpu_hw_ctl_clear_pending_flush(ctx->pending_flush_mask, + dpu_hw_ctl_get_flush_register(ctx)); + ctx->pending_flush_mask = 0x0; + ctx->pending_intf_flush_mask = 0; + ctx->pending_wb_flush_mask = 0; + ctx->pending_cwb_flush_mask = 0; + ctx->pending_periph_flush_mask = 0; + ctx->pending_merge_3d_flush_mask = 0; + ctx->pending_dsc_flush_mask = 0; + ctx->pending_cdm_flush_mask = 0; + + memset(ctx->pending_dspp_flush_mask, 0, + sizeof(ctx->pending_dspp_flush_mask)); +} + +static inline void dpu_hw_ctl_update_pending_flush(struct dpu_hw_ctl *ctx, + u32 flushbits) +{ + trace_dpu_hw_ctl_update_pending_flush(flushbits, + ctx->pending_flush_mask); + ctx->pending_flush_mask |= flushbits; +} + +static u32 dpu_hw_ctl_get_pending_flush(struct dpu_hw_ctl *ctx) +{ + return ctx->pending_flush_mask; +} + +static inline void dpu_hw_ctl_trigger_flush_v1(struct dpu_hw_ctl *ctx) +{ + int dspp; + + if (ctx->pending_flush_mask & BIT(MERGE_3D_IDX)) + DPU_REG_WRITE(&ctx->hw, CTL_MERGE_3D_FLUSH, + ctx->pending_merge_3d_flush_mask); + if (ctx->pending_flush_mask & BIT(INTF_IDX)) + DPU_REG_WRITE(&ctx->hw, CTL_INTF_FLUSH, + ctx->pending_intf_flush_mask); + if (ctx->pending_flush_mask & BIT(WB_IDX)) + DPU_REG_WRITE(&ctx->hw, CTL_WB_FLUSH, + ctx->pending_wb_flush_mask); + if (ctx->pending_flush_mask & BIT(CWB_IDX)) + DPU_REG_WRITE(&ctx->hw, CTL_CWB_FLUSH, + ctx->pending_cwb_flush_mask); + + if (ctx->pending_flush_mask & BIT(DSPP_IDX)) + for (dspp = DSPP_0; dspp < DSPP_MAX; dspp++) { + if (ctx->pending_dspp_flush_mask[dspp - DSPP_0]) + DPU_REG_WRITE(&ctx->hw, + CTL_DSPP_n_FLUSH(dspp - DSPP_0), + ctx->pending_dspp_flush_mask[dspp - DSPP_0]); + } + + if (ctx->pending_flush_mask & BIT(PERIPH_IDX)) + DPU_REG_WRITE(&ctx->hw, CTL_PERIPH_FLUSH, + ctx->pending_periph_flush_mask); + + if (ctx->pending_flush_mask & BIT(DSC_IDX)) + DPU_REG_WRITE(&ctx->hw, CTL_DSC_FLUSH, + ctx->pending_dsc_flush_mask); + + if (ctx->pending_flush_mask & BIT(CDM_IDX)) + DPU_REG_WRITE(&ctx->hw, CTL_CDM_FLUSH, + ctx->pending_cdm_flush_mask); + + DPU_REG_WRITE(&ctx->hw, CTL_FLUSH, ctx->pending_flush_mask); +} + +static inline void dpu_hw_ctl_trigger_flush(struct dpu_hw_ctl *ctx) +{ + trace_dpu_hw_ctl_trigger_pending_flush(ctx->pending_flush_mask, + dpu_hw_ctl_get_flush_register(ctx)); + DPU_REG_WRITE(&ctx->hw, CTL_FLUSH, ctx->pending_flush_mask); +} + +static void dpu_hw_ctl_update_pending_flush_sspp(struct dpu_hw_ctl *ctx, + enum dpu_sspp sspp) +{ + switch (sspp) { + case SSPP_VIG0: + ctx->pending_flush_mask |= BIT(0); + break; + case SSPP_VIG1: + ctx->pending_flush_mask |= BIT(1); + break; + case SSPP_VIG2: + ctx->pending_flush_mask |= BIT(2); + break; + case SSPP_VIG3: + ctx->pending_flush_mask |= BIT(18); + break; + case SSPP_RGB0: + ctx->pending_flush_mask |= BIT(3); + break; + case SSPP_RGB1: + ctx->pending_flush_mask |= BIT(4); + break; + case SSPP_RGB2: + ctx->pending_flush_mask |= BIT(5); + break; + case SSPP_RGB3: + ctx->pending_flush_mask |= BIT(19); + break; + case SSPP_DMA0: + ctx->pending_flush_mask |= BIT(11); + break; + case SSPP_DMA1: + ctx->pending_flush_mask |= BIT(12); + break; + case SSPP_DMA2: + ctx->pending_flush_mask |= BIT(24); + break; + case SSPP_DMA3: + ctx->pending_flush_mask |= BIT(25); + break; + case SSPP_DMA4: + ctx->pending_flush_mask |= BIT(13); + break; + case SSPP_DMA5: + ctx->pending_flush_mask |= BIT(14); + break; + case SSPP_CURSOR0: + ctx->pending_flush_mask |= BIT(22); + break; + case SSPP_CURSOR1: + ctx->pending_flush_mask |= BIT(23); + break; + default: + break; + } +} + +static void dpu_hw_ctl_update_pending_flush_mixer(struct dpu_hw_ctl *ctx, + enum dpu_lm lm) +{ + switch (lm) { + case LM_0: + ctx->pending_flush_mask |= BIT(6); + break; + case LM_1: + ctx->pending_flush_mask |= BIT(7); + break; + case LM_2: + ctx->pending_flush_mask |= BIT(8); + break; + case LM_3: + ctx->pending_flush_mask |= BIT(9); + break; + case LM_4: + ctx->pending_flush_mask |= BIT(10); + break; + case LM_5: + ctx->pending_flush_mask |= BIT(20); + break; + case LM_6: + ctx->pending_flush_mask |= BIT(21); + break; + case LM_7: + ctx->pending_flush_mask |= BIT(27); + break; + default: + break; + } + + ctx->pending_flush_mask |= CTL_FLUSH_MASK_CTL; +} + +static void dpu_hw_ctl_update_pending_flush_intf(struct dpu_hw_ctl *ctx, + enum dpu_intf intf) +{ + switch (intf) { + case INTF_0: + ctx->pending_flush_mask |= BIT(31); + break; + case INTF_1: + ctx->pending_flush_mask |= BIT(30); + break; + case INTF_2: + ctx->pending_flush_mask |= BIT(29); + break; + case INTF_3: + ctx->pending_flush_mask |= BIT(28); + break; + default: + break; + } +} + +static void dpu_hw_ctl_update_pending_flush_wb(struct dpu_hw_ctl *ctx, + enum dpu_wb wb) +{ + switch (wb) { + case WB_0: + case WB_1: + case WB_2: + ctx->pending_flush_mask |= BIT(WB_IDX); + break; + default: + break; + } +} + +static void dpu_hw_ctl_update_pending_flush_cdm(struct dpu_hw_ctl *ctx, enum dpu_cdm cdm_num) +{ + /* update pending flush only if CDM_0 is flushed */ + if (cdm_num == CDM_0) + ctx->pending_flush_mask |= BIT(CDM_IDX); +} + +static void dpu_hw_ctl_update_pending_flush_wb_v1(struct dpu_hw_ctl *ctx, + enum dpu_wb wb) +{ + ctx->pending_wb_flush_mask |= BIT(wb - WB_0); + ctx->pending_flush_mask |= BIT(WB_IDX); +} + +static void dpu_hw_ctl_update_pending_flush_cwb_v1(struct dpu_hw_ctl *ctx, + enum dpu_cwb cwb) +{ + ctx->pending_cwb_flush_mask |= BIT(cwb - CWB_0); + ctx->pending_flush_mask |= BIT(CWB_IDX); +} + +static void dpu_hw_ctl_update_pending_flush_intf_v1(struct dpu_hw_ctl *ctx, + enum dpu_intf intf) +{ + ctx->pending_intf_flush_mask |= BIT(intf - INTF_0); + ctx->pending_flush_mask |= BIT(INTF_IDX); +} + +static void dpu_hw_ctl_update_pending_flush_periph_v1(struct dpu_hw_ctl *ctx, + enum dpu_intf intf) +{ + ctx->pending_periph_flush_mask |= BIT(intf - INTF_0); + ctx->pending_flush_mask |= BIT(PERIPH_IDX); +} + +static void dpu_hw_ctl_update_pending_flush_merge_3d_v1(struct dpu_hw_ctl *ctx, + enum dpu_merge_3d merge_3d) +{ + ctx->pending_merge_3d_flush_mask |= BIT(merge_3d - MERGE_3D_0); + ctx->pending_flush_mask |= BIT(MERGE_3D_IDX); +} + +static void dpu_hw_ctl_update_pending_flush_dsc_v1(struct dpu_hw_ctl *ctx, + enum dpu_dsc dsc_num) +{ + ctx->pending_dsc_flush_mask |= BIT(dsc_num - DSC_0); + ctx->pending_flush_mask |= BIT(DSC_IDX); +} + +static void dpu_hw_ctl_update_pending_flush_cdm_v1(struct dpu_hw_ctl *ctx, enum dpu_cdm cdm_num) +{ + ctx->pending_cdm_flush_mask |= BIT(cdm_num - CDM_0); + ctx->pending_flush_mask |= BIT(CDM_IDX); +} + +static void dpu_hw_ctl_update_pending_flush_dspp(struct dpu_hw_ctl *ctx, + enum dpu_dspp dspp, u32 dspp_sub_blk) +{ + switch (dspp) { + case DSPP_0: + ctx->pending_flush_mask |= BIT(13); + break; + case DSPP_1: + ctx->pending_flush_mask |= BIT(14); + break; + case DSPP_2: + ctx->pending_flush_mask |= BIT(15); + break; + case DSPP_3: + ctx->pending_flush_mask |= BIT(21); + break; + default: + break; + } +} + +static void dpu_hw_ctl_update_pending_flush_dspp_sub_blocks( + struct dpu_hw_ctl *ctx, enum dpu_dspp dspp, u32 dspp_sub_blk) +{ + if (dspp >= DSPP_MAX) + return; + + switch (dspp_sub_blk) { + case DPU_DSPP_PCC: + ctx->pending_dspp_flush_mask[dspp - DSPP_0] |= BIT(4); + break; + case DPU_DSPP_GC: + ctx->pending_dspp_flush_mask[dspp - DSPP_0] |= BIT(5); + break; + default: + return; + } + + ctx->pending_flush_mask |= BIT(DSPP_IDX); +} + +static u32 dpu_hw_ctl_poll_reset_status(struct dpu_hw_ctl *ctx, u32 timeout_us) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + ktime_t timeout; + u32 status; + + timeout = ktime_add_us(ktime_get(), timeout_us); + + /* + * it takes around 30us to have mdp finish resetting its ctl path + * poll every 50us so that reset should be completed at 1st poll + */ + do { + status = DPU_REG_READ(c, CTL_SW_RESET); + status &= 0x1; + if (status) + usleep_range(20, 50); + } while (status && ktime_compare_safe(ktime_get(), timeout) < 0); + + return status; +} + +static int dpu_hw_ctl_reset_control(struct dpu_hw_ctl *ctx) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + + pr_debug("issuing hw ctl reset for ctl:%d\n", ctx->idx); + DPU_REG_WRITE(c, CTL_SW_RESET, 0x1); + if (dpu_hw_ctl_poll_reset_status(ctx, DPU_REG_RESET_TIMEOUT_US)) + return -EINVAL; + + return 0; +} + +static int dpu_hw_ctl_wait_reset_status(struct dpu_hw_ctl *ctx) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + u32 status; + + status = DPU_REG_READ(c, CTL_SW_RESET); + status &= 0x01; + if (!status) + return 0; + + pr_debug("hw ctl reset is set for ctl:%d\n", ctx->idx); + if (dpu_hw_ctl_poll_reset_status(ctx, DPU_REG_RESET_TIMEOUT_US)) { + pr_err("hw recovery is not complete for ctl:%d\n", ctx->idx); + return -EINVAL; + } + + return 0; +} + +static void dpu_hw_ctl_clear_all_blendstages(struct dpu_hw_ctl *ctx) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + int i; + + for (i = 0; i < ctx->mixer_count; i++) { + enum dpu_lm mixer_id = ctx->mixer_hw_caps[i].id; + + DPU_REG_WRITE(c, CTL_LAYER(mixer_id), 0); + DPU_REG_WRITE(c, CTL_LAYER_EXT(mixer_id), 0); + DPU_REG_WRITE(c, CTL_LAYER_EXT2(mixer_id), 0); + DPU_REG_WRITE(c, CTL_LAYER_EXT3(mixer_id), 0); + } + + DPU_REG_WRITE(c, CTL_FETCH_PIPE_ACTIVE, 0); +} + +struct ctl_blend_config { + int idx, shift, ext_shift; +}; + +static const struct ctl_blend_config ctl_blend_config[][2] = { + [SSPP_NONE] = { { -1 }, { -1 } }, + [SSPP_MAX] = { { -1 }, { -1 } }, + [SSPP_VIG0] = { { 0, 0, 0 }, { 3, 0 } }, + [SSPP_VIG1] = { { 0, 3, 2 }, { 3, 4 } }, + [SSPP_VIG2] = { { 0, 6, 4 }, { 3, 8 } }, + [SSPP_VIG3] = { { 0, 26, 6 }, { 3, 12 } }, + [SSPP_RGB0] = { { 0, 9, 8 }, { -1 } }, + [SSPP_RGB1] = { { 0, 12, 10 }, { -1 } }, + [SSPP_RGB2] = { { 0, 15, 12 }, { -1 } }, + [SSPP_RGB3] = { { 0, 29, 14 }, { -1 } }, + [SSPP_DMA0] = { { 0, 18, 16 }, { 2, 8 } }, + [SSPP_DMA1] = { { 0, 21, 18 }, { 2, 12 } }, + [SSPP_DMA2] = { { 2, 0 }, { 2, 16 } }, + [SSPP_DMA3] = { { 2, 4 }, { 2, 20 } }, + [SSPP_DMA4] = { { 4, 0 }, { 4, 8 } }, + [SSPP_DMA5] = { { 4, 4 }, { 4, 12 } }, + [SSPP_CURSOR0] = { { 1, 20 }, { -1 } }, + [SSPP_CURSOR1] = { { 1, 26 }, { -1 } }, +}; + +static void dpu_hw_ctl_setup_blendstage(struct dpu_hw_ctl *ctx, + enum dpu_lm lm, struct dpu_hw_stage_cfg *stage_cfg) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + u32 mix, ext, mix_ext; + u32 mixercfg[5] = { 0 }; + int i, j; + int stages; + int pipes_per_stage; + + stages = _mixer_stages(ctx->mixer_hw_caps, ctx->mixer_count, lm); + if (stages < 0) + return; + + if (test_bit(DPU_MIXER_SOURCESPLIT, + &ctx->mixer_hw_caps->features)) + pipes_per_stage = PIPES_PER_STAGE; + else + pipes_per_stage = 1; + + mixercfg[0] = CTL_MIXER_BORDER_OUT; /* always set BORDER_OUT */ + + if (!stage_cfg) + goto exit; + + for (i = 0; i <= stages; i++) { + /* overflow to ext register if 'i + 1 > 7' */ + mix = (i + 1) & 0x7; + ext = i >= 7; + mix_ext = (i + 1) & 0xf; + + for (j = 0 ; j < pipes_per_stage; j++) { + enum dpu_sspp_multirect_index rect_index = + stage_cfg->multirect_index[i][j]; + enum dpu_sspp pipe = stage_cfg->stage[i][j]; + const struct ctl_blend_config *cfg = + &ctl_blend_config[pipe][rect_index == DPU_SSPP_RECT_1]; + + /* + * CTL_LAYER has 3-bit field (and extra bits in EXT register), + * all EXT registers has 4-bit fields. + */ + if (cfg->idx == -1) { + continue; + } else if (cfg->idx == 0) { + mixercfg[0] |= mix << cfg->shift; + mixercfg[1] |= ext << cfg->ext_shift; + } else { + mixercfg[cfg->idx] |= mix_ext << cfg->shift; + } + } + } + +exit: + DPU_REG_WRITE(c, CTL_LAYER(lm), mixercfg[0]); + DPU_REG_WRITE(c, CTL_LAYER_EXT(lm), mixercfg[1]); + DPU_REG_WRITE(c, CTL_LAYER_EXT2(lm), mixercfg[2]); + DPU_REG_WRITE(c, CTL_LAYER_EXT3(lm), mixercfg[3]); + if (ctx->mdss_ver->core_major_ver >= 9) + DPU_REG_WRITE(c, CTL_LAYER_EXT4(lm), mixercfg[4]); +} + + +static void dpu_hw_ctl_intf_cfg_v1(struct dpu_hw_ctl *ctx, + struct dpu_hw_intf_cfg *cfg) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + u32 intf_active = 0; + u32 dsc_active = 0; + u32 wb_active = 0; + u32 cwb_active = 0; + u32 mode_sel = 0; + u32 merge_3d_active = 0; + + /* CTL_TOP[31:28] carries group_id to collate CTL paths + * per VM. Explicitly disable it until VM support is + * added in SW. Power on reset value is not disable. + */ + if (ctx->mdss_ver->core_major_ver >= 7) + mode_sel = CTL_DEFAULT_GROUP_ID << 28; + + if (cfg->intf_mode_sel == DPU_CTL_MODE_SEL_CMD) + mode_sel |= BIT(17); + + intf_active = DPU_REG_READ(c, CTL_INTF_ACTIVE); + wb_active = DPU_REG_READ(c, CTL_WB_ACTIVE); + cwb_active = DPU_REG_READ(c, CTL_CWB_ACTIVE); + dsc_active = DPU_REG_READ(c, CTL_DSC_ACTIVE); + merge_3d_active = DPU_REG_READ(c, CTL_MERGE_3D_ACTIVE); + + if (cfg->intf) + intf_active |= BIT(cfg->intf - INTF_0); + + if (cfg->wb) + wb_active |= BIT(cfg->wb - WB_0); + + if (cfg->cwb) + cwb_active |= cfg->cwb; + + if (cfg->dsc) + dsc_active |= cfg->dsc; + + if (cfg->merge_3d) + merge_3d_active |= BIT(cfg->merge_3d - MERGE_3D_0); + + DPU_REG_WRITE(c, CTL_TOP, mode_sel); + DPU_REG_WRITE(c, CTL_INTF_ACTIVE, intf_active); + DPU_REG_WRITE(c, CTL_WB_ACTIVE, wb_active); + DPU_REG_WRITE(c, CTL_CWB_ACTIVE, cwb_active); + DPU_REG_WRITE(c, CTL_DSC_ACTIVE, dsc_active); + DPU_REG_WRITE(c, CTL_MERGE_3D_ACTIVE, merge_3d_active); + + if (cfg->intf_master) + DPU_REG_WRITE(c, CTL_INTF_MASTER, BIT(cfg->intf_master - INTF_0)); + + if (cfg->cdm) + DPU_REG_WRITE(c, CTL_CDM_ACTIVE, cfg->cdm); +} + +static void dpu_hw_ctl_intf_cfg(struct dpu_hw_ctl *ctx, + struct dpu_hw_intf_cfg *cfg) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + u32 intf_cfg = 0; + + intf_cfg |= (cfg->intf & 0xF) << 4; + + if (cfg->mode_3d) { + intf_cfg |= BIT(19); + intf_cfg |= (cfg->mode_3d - 0x1) << 20; + } + + if (cfg->wb) + intf_cfg |= (cfg->wb & 0x3) + 2; + + switch (cfg->intf_mode_sel) { + case DPU_CTL_MODE_SEL_VID: + intf_cfg &= ~BIT(17); + intf_cfg &= ~(0x3 << 15); + break; + case DPU_CTL_MODE_SEL_CMD: + intf_cfg |= BIT(17); + intf_cfg |= ((cfg->stream_sel & 0x3) << 15); + break; + default: + pr_err("unknown interface type %d\n", cfg->intf_mode_sel); + return; + } + + DPU_REG_WRITE(c, CTL_TOP, intf_cfg); +} + +static void dpu_hw_ctl_reset_intf_cfg_v1(struct dpu_hw_ctl *ctx, + struct dpu_hw_intf_cfg *cfg) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + u32 intf_active = 0; + u32 intf_master = 0; + u32 wb_active = 0; + u32 cwb_active = 0; + u32 merge3d_active = 0; + u32 dsc_active; + u32 cdm_active; + + /* + * This API resets each portion of the CTL path namely, + * clearing the sspps staged on the lm, merge_3d block, + * interfaces , writeback etc to ensure clean teardown of the pipeline. + * This will be used for writeback to begin with to have a + * proper teardown of the writeback session but upon further + * validation, this can be extended to all interfaces. + */ + if (cfg->merge_3d) { + merge3d_active = DPU_REG_READ(c, CTL_MERGE_3D_ACTIVE); + merge3d_active &= ~BIT(cfg->merge_3d - MERGE_3D_0); + DPU_REG_WRITE(c, CTL_MERGE_3D_ACTIVE, + merge3d_active); + } + + if (ctx->ops.clear_all_blendstages) + ctx->ops.clear_all_blendstages(ctx); + + if (ctx->ops.set_active_lms) + ctx->ops.set_active_lms(ctx, NULL); + + if (ctx->ops.set_active_fetch_pipes) + ctx->ops.set_active_fetch_pipes(ctx, NULL); + + if (ctx->ops.set_active_pipes) + ctx->ops.set_active_pipes(ctx, NULL); + + if (cfg->intf) { + intf_active = DPU_REG_READ(c, CTL_INTF_ACTIVE); + intf_active &= ~BIT(cfg->intf - INTF_0); + DPU_REG_WRITE(c, CTL_INTF_ACTIVE, intf_active); + + intf_master = DPU_REG_READ(c, CTL_INTF_MASTER); + + /* Unset this intf as master, if it is the current master */ + if (intf_master == BIT(cfg->intf - INTF_0)) { + DPU_DEBUG_DRIVER("Unsetting INTF_%d master\n", cfg->intf - INTF_0); + DPU_REG_WRITE(c, CTL_INTF_MASTER, 0); + } + } + + if (cfg->cwb) { + cwb_active = DPU_REG_READ(c, CTL_CWB_ACTIVE); + cwb_active &= ~cfg->cwb; + DPU_REG_WRITE(c, CTL_CWB_ACTIVE, cwb_active); + } + + if (cfg->wb) { + wb_active = DPU_REG_READ(c, CTL_WB_ACTIVE); + wb_active &= ~BIT(cfg->wb - WB_0); + DPU_REG_WRITE(c, CTL_WB_ACTIVE, wb_active); + } + + if (cfg->dsc) { + dsc_active = DPU_REG_READ(c, CTL_DSC_ACTIVE); + dsc_active &= ~cfg->dsc; + DPU_REG_WRITE(c, CTL_DSC_ACTIVE, dsc_active); + } + + if (cfg->cdm) { + cdm_active = DPU_REG_READ(c, CTL_CDM_ACTIVE); + cdm_active &= ~cfg->cdm; + DPU_REG_WRITE(c, CTL_CDM_ACTIVE, cdm_active); + } +} + +static void dpu_hw_ctl_set_active_fetch_pipes(struct dpu_hw_ctl *ctx, + unsigned long *fetch_active) +{ + int i; + u32 val = 0; + + if (fetch_active) { + for (i = 0; i < SSPP_MAX; i++) { + if (test_bit(i, fetch_active) && + fetch_tbl[i] != CTL_INVALID_BIT) + val |= BIT(fetch_tbl[i]); + } + } + + DPU_REG_WRITE(&ctx->hw, CTL_FETCH_PIPE_ACTIVE, val); +} + +static void dpu_hw_ctl_set_active_pipes(struct dpu_hw_ctl *ctx, + unsigned long *active_pipes) +{ + int i; + u32 val = 0; + + if (active_pipes) { + for (i = 0; i < SSPP_MAX; i++) { + if (test_bit(i, active_pipes) && + fetch_tbl[i] != CTL_INVALID_BIT) + val |= BIT(fetch_tbl[i]); + } + } + + DPU_REG_WRITE(&ctx->hw, CTL_PIPE_ACTIVE, val); +} + +static void dpu_hw_ctl_set_active_lms(struct dpu_hw_ctl *ctx, + unsigned long *active_lms) +{ + int i; + u32 val = 0; + + if (active_lms) { + for (i = LM_0; i < LM_MAX; i++) { + if (test_bit(i, active_lms) && + lm_tbl[i] != CTL_INVALID_BIT) + val |= BIT(lm_tbl[i]); + } + } + + DPU_REG_WRITE(&ctx->hw, CTL_LAYER_ACTIVE, val); +} + +/** + * dpu_hw_ctl_init() - Initializes the ctl_path hw driver object. + * Should be called before accessing any ctl_path register. + * @dev: Corresponding device for devres management + * @cfg: ctl_path catalog entry for which driver object is required + * @addr: mapped register io address of MDP + * @mdss_ver: dpu core's major and minor versions + * @mixer_count: Number of mixers in @mixer + * @mixer: Pointer to an array of Layer Mixers defined in the catalog + */ +struct dpu_hw_ctl *dpu_hw_ctl_init(struct drm_device *dev, + const struct dpu_ctl_cfg *cfg, + void __iomem *addr, + const struct dpu_mdss_version *mdss_ver, + u32 mixer_count, + const struct dpu_lm_cfg *mixer) +{ + struct dpu_hw_ctl *c; + + c = drmm_kzalloc(dev, sizeof(*c), GFP_KERNEL); + if (!c) + return ERR_PTR(-ENOMEM); + + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_CTL; + + c->caps = cfg; + c->mdss_ver = mdss_ver; + + if (mdss_ver->core_major_ver >= 5) { + c->ops.trigger_flush = dpu_hw_ctl_trigger_flush_v1; + c->ops.setup_intf_cfg = dpu_hw_ctl_intf_cfg_v1; + c->ops.reset_intf_cfg = dpu_hw_ctl_reset_intf_cfg_v1; + c->ops.update_pending_flush_intf = + dpu_hw_ctl_update_pending_flush_intf_v1; + + c->ops.update_pending_flush_periph = + dpu_hw_ctl_update_pending_flush_periph_v1; + + c->ops.update_pending_flush_merge_3d = + dpu_hw_ctl_update_pending_flush_merge_3d_v1; + c->ops.update_pending_flush_wb = dpu_hw_ctl_update_pending_flush_wb_v1; + c->ops.update_pending_flush_cwb = dpu_hw_ctl_update_pending_flush_cwb_v1; + c->ops.update_pending_flush_dsc = + dpu_hw_ctl_update_pending_flush_dsc_v1; + c->ops.update_pending_flush_cdm = dpu_hw_ctl_update_pending_flush_cdm_v1; + } else { + c->ops.trigger_flush = dpu_hw_ctl_trigger_flush; + c->ops.setup_intf_cfg = dpu_hw_ctl_intf_cfg; + c->ops.update_pending_flush_intf = + dpu_hw_ctl_update_pending_flush_intf; + c->ops.update_pending_flush_wb = dpu_hw_ctl_update_pending_flush_wb; + c->ops.update_pending_flush_cdm = dpu_hw_ctl_update_pending_flush_cdm; + } + c->ops.clear_pending_flush = dpu_hw_ctl_clear_pending_flush; + c->ops.update_pending_flush = dpu_hw_ctl_update_pending_flush; + c->ops.get_pending_flush = dpu_hw_ctl_get_pending_flush; + c->ops.get_flush_register = dpu_hw_ctl_get_flush_register; + c->ops.trigger_start = dpu_hw_ctl_trigger_start; + c->ops.is_started = dpu_hw_ctl_is_started; + c->ops.trigger_pending = dpu_hw_ctl_trigger_pending; + c->ops.reset = dpu_hw_ctl_reset_control; + c->ops.wait_reset_status = dpu_hw_ctl_wait_reset_status; + if (mdss_ver->core_major_ver < 12) { + c->ops.clear_all_blendstages = dpu_hw_ctl_clear_all_blendstages; + c->ops.setup_blendstage = dpu_hw_ctl_setup_blendstage; + } else { + c->ops.set_active_pipes = dpu_hw_ctl_set_active_pipes; + c->ops.set_active_lms = dpu_hw_ctl_set_active_lms; + } + c->ops.update_pending_flush_sspp = dpu_hw_ctl_update_pending_flush_sspp; + c->ops.update_pending_flush_mixer = dpu_hw_ctl_update_pending_flush_mixer; + if (mdss_ver->core_major_ver >= 7) + c->ops.update_pending_flush_dspp = dpu_hw_ctl_update_pending_flush_dspp_sub_blocks; + else + c->ops.update_pending_flush_dspp = dpu_hw_ctl_update_pending_flush_dspp; + + if (mdss_ver->core_major_ver >= 7) + c->ops.set_active_fetch_pipes = dpu_hw_ctl_set_active_fetch_pipes; + + c->idx = cfg->id; + c->mixer_count = mixer_count; + c->mixer_hw_caps = mixer; + + return c; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h new file mode 100644 index 000000000..e535bf013 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h @@ -0,0 +1,361 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + * Copyright (c) 2022-2024 Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#ifndef _DPU_HW_CTL_H +#define _DPU_HW_CTL_H + +#include "dpu_hw_mdss.h" +#include "dpu_hw_util.h" +#include "dpu_hw_catalog.h" +#include "dpu_hw_sspp.h" + +/** + * enum dpu_ctl_mode_sel: Interface mode selection + * @DPU_CTL_MODE_SEL_VID: Video mode interface + * @DPU_CTL_MODE_SEL_CMD: Command mode interface + */ +enum dpu_ctl_mode_sel { + DPU_CTL_MODE_SEL_VID = 0, + DPU_CTL_MODE_SEL_CMD +}; + +struct dpu_hw_ctl; +/** + * struct dpu_hw_stage_cfg - blending stage cfg + * @stage : SSPP_ID at each stage + * @multirect_index: index of the rectangle of SSPP. + */ +struct dpu_hw_stage_cfg { + enum dpu_sspp stage[DPU_STAGE_MAX][PIPES_PER_STAGE]; + enum dpu_sspp_multirect_index multirect_index + [DPU_STAGE_MAX][PIPES_PER_STAGE]; +}; + +/** + * struct dpu_hw_intf_cfg :Describes how the DPU writes data to output interface + * @intf : Interface id + * @intf_master: Master interface id in the dual pipe topology + * @wb: Writeback mode + * @mode_3d: 3d mux configuration + * @merge_3d: 3d merge block used + * @intf_mode_sel: Interface mode, cmd / vid + * @cdm: CDM block used + * @stream_sel: Stream selection for multi-stream interfaces + * @dsc: DSC BIT masks used + * @cwb: CWB BIT masks used + */ +struct dpu_hw_intf_cfg { + enum dpu_intf intf; + enum dpu_intf intf_master; + enum dpu_wb wb; + enum dpu_3d_blend_mode mode_3d; + enum dpu_merge_3d merge_3d; + enum dpu_ctl_mode_sel intf_mode_sel; + enum dpu_cdm cdm; + int stream_sel; + unsigned int cwb; + unsigned int dsc; +}; + +/** + * struct dpu_hw_ctl_ops - Interface to the wb Hw driver functions + * Assumption is these functions will be called after clocks are enabled + */ +struct dpu_hw_ctl_ops { + /** + * @trigger_start: kickoff hw operation for Sw controlled interfaces + * DSI cmd mode and WB interface are SW controlled + * @ctx : ctl path ctx pointer + */ + void (*trigger_start)(struct dpu_hw_ctl *ctx); + + /** + * @is_started: check if the ctl is started + * @ctx : ctl path ctx pointer + * @Return: true if started, false if stopped + */ + bool (*is_started)(struct dpu_hw_ctl *ctx); + + /** + * @trigger_pending: kickoff prepare is in progress hw operation for sw + * controlled interfaces: DSI cmd mode and WB interface + * are SW controlled + * @ctx : ctl path ctx pointer + */ + void (*trigger_pending)(struct dpu_hw_ctl *ctx); + + /** + * @clear_pending_flush: Clear the value of the cached pending_flush_mask + * No effect on hardware. + * Required to be implemented. + * @ctx : ctl path ctx pointer + */ + void (*clear_pending_flush)(struct dpu_hw_ctl *ctx); + + /** + * @get_pending_flush: Query the value of the cached pending_flush_mask + * No effect on hardware + * @ctx : ctl path ctx pointer + */ + u32 (*get_pending_flush)(struct dpu_hw_ctl *ctx); + + /** + * @update_pending_flush: OR in the given flushbits to the cached + * pending_flush_mask. + * No effect on hardware + * @ctx : ctl path ctx pointer + * @flushbits : module flushmask + */ + void (*update_pending_flush)(struct dpu_hw_ctl *ctx, + u32 flushbits); + + /** + * @update_pending_flush_wb: OR in the given flushbits to the + * cached pending_(wb_)flush_mask. + * No effect on hardware + * @ctx : ctl path ctx pointer + * @blk : writeback block index + */ + void (*update_pending_flush_wb)(struct dpu_hw_ctl *ctx, + enum dpu_wb blk); + + /** + * @update_pending_flush_cwb: OR in the given flushbits to the + * cached pending_(cwb_)flush_mask. + * No effect on hardware + * @ctx : ctl path ctx pointer + * @blk : concurrent writeback block index + */ + void (*update_pending_flush_cwb)(struct dpu_hw_ctl *ctx, + enum dpu_cwb blk); + + /** + * @update_pending_flush_intf: OR in the given flushbits to the + * cached pending_(intf_)flush_mask. + * No effect on hardware + * @ctx : ctl path ctx pointer + * @blk : interface block index + */ + void (*update_pending_flush_intf)(struct dpu_hw_ctl *ctx, + enum dpu_intf blk); + + /** + * @update_pending_flush_periph: OR in the given flushbits to the + * cached pending_(periph_)flush_mask. + * No effect on hardware + * @ctx : ctl path ctx pointer + * @blk : interface block index + */ + void (*update_pending_flush_periph)(struct dpu_hw_ctl *ctx, + enum dpu_intf blk); + + /** + * @update_pending_flush_merge_3d: OR in the given flushbits to the + * cached pending_(merge_3d_)flush_mask. + * No effect on hardware + * @ctx : ctl path ctx pointer + * @blk : interface block index + */ + void (*update_pending_flush_merge_3d)(struct dpu_hw_ctl *ctx, + enum dpu_merge_3d blk); + + /** + * @update_pending_flush_sspp: OR in the given flushbits to the + * cached pending_flush_mask. + * No effect on hardware + * @ctx : ctl path ctx pointer + * @blk : SSPP block index + */ + void (*update_pending_flush_sspp)(struct dpu_hw_ctl *ctx, + enum dpu_sspp blk); + + /** + * @update_pending_flush_mixer: OR in the given flushbits to the + * cached pending_flush_mask. + * No effect on hardware + * @ctx : ctl path ctx pointer + * @blk : LM block index + */ + void (*update_pending_flush_mixer)(struct dpu_hw_ctl *ctx, + enum dpu_lm blk); + + /** + * @update_pending_flush_dspp: OR in the given flushbits to the + * cached pending_flush_mask. + * No effect on hardware + * @ctx : ctl path ctx pointer + * @blk : DSPP block index + * @dspp_sub_blk : DSPP sub-block index + */ + void (*update_pending_flush_dspp)(struct dpu_hw_ctl *ctx, + enum dpu_dspp blk, u32 dspp_sub_blk); + + /** + * @update_pending_flush_dsc: OR in the given flushbits to the + * cached pending_(dsc_)flush_mask. + * No effect on hardware + * @ctx: ctl path ctx pointer + * @blk: interface block index + */ + void (*update_pending_flush_dsc)(struct dpu_hw_ctl *ctx, + enum dpu_dsc blk); + + /** + * @update_pending_flush_cdm: OR in the given flushbits to the + * cached pending_(cdm_)flush_mask. + * No effect on hardware + * @ctx: ctl path ctx pointer + * @cdm_num: idx of cdm to be flushed + */ + void (*update_pending_flush_cdm)(struct dpu_hw_ctl *ctx, enum dpu_cdm cdm_num); + + /** + * @trigger_flush: Write the value of the pending_flush_mask to hardware + * @ctx : ctl path ctx pointer + */ + void (*trigger_flush)(struct dpu_hw_ctl *ctx); + + /** + * @get_flush_register: Read the value of the flush register + * @ctx : ctl path ctx pointer + * @Return: value of the ctl flush register. + */ + u32 (*get_flush_register)(struct dpu_hw_ctl *ctx); + + /** + * @setup_intf_cfg: Setup ctl_path interface config + * @ctx + * @cfg : interface config structure pointer + */ + void (*setup_intf_cfg)(struct dpu_hw_ctl *ctx, + struct dpu_hw_intf_cfg *cfg); + + /** + * @reset_intf_cfg: reset ctl_path interface config + * @ctx : ctl path ctx pointer + * @cfg : interface config structure pointer + */ + void (*reset_intf_cfg)(struct dpu_hw_ctl *ctx, + struct dpu_hw_intf_cfg *cfg); + + /** + * @reset: reset function for this ctl type + */ + int (*reset)(struct dpu_hw_ctl *c); + + /** + * @wait_reset_status: checks ctl reset status + * @ctx : ctl path ctx pointer + * + * This function checks the ctl reset status bit. + * If the reset bit is set, it keeps polling the status till the hw + * reset is complete. + * Returns: 0 on success or -error if reset incomplete within interval + */ + int (*wait_reset_status)(struct dpu_hw_ctl *ctx); + + /** + * @clear_all_blendstages: Set all blend stages to disabled + * @ctx : ctl path ctx pointer + */ + void (*clear_all_blendstages)(struct dpu_hw_ctl *ctx); + + /** + * @setup_blendstage: Configure layer mixer to pipe configuration + * @ctx : ctl path ctx pointer + * @lm : layer mixer enumeration + * @cfg : blend stage configuration + */ + void (*setup_blendstage)(struct dpu_hw_ctl *ctx, + enum dpu_lm lm, struct dpu_hw_stage_cfg *cfg); + + /** + * @set_active_fetch_pipes: Set active pipes attached to this CTL + * @ctx: ctl path ctx pointer + * @active_pipes: bitmap of enum dpu_sspp + */ + void (*set_active_fetch_pipes)(struct dpu_hw_ctl *ctx, + unsigned long *fetch_active); + + /** + * @set_active_pipes: Set active pipes attached to this CTL + * @ctx: ctl path ctx pointer + * @active_pipes: bitmap of enum dpu_sspp + */ + void (*set_active_pipes)(struct dpu_hw_ctl *ctx, + unsigned long *active_pipes); + + /** + * @set_active_lms: Set active layer mixers attached to this CTL + * @ctx: ctl path ctx pointer + * @active_lms: bitmap of enum dpu_lm + */ + void (*set_active_lms)(struct dpu_hw_ctl *ctx, + unsigned long *active_lms); +}; + +/** + * struct dpu_hw_ctl : CTL PATH driver object + * @base: hardware block base structure + * @hw: block register map object + * @idx: control path index + * @caps: control path capabilities + * @mixer_count: number of mixers + * @mixer_hw_caps: mixer hardware capabilities + * @pending_flush_mask: storage for pending ctl_flush managed via ops + * @pending_intf_flush_mask: pending INTF flush + * @pending_wb_flush_mask: pending WB flush + * @pending_cwb_flush_mask: pending CWB flush + * @pending_periph_flush_mask: pending PERIPH flush + * @pending_merge_3d_flush_mask: pending MERGE 3D flush + * @pending_dspp_flush_mask: pending DSPP flush + * @pending_dsc_flush_mask: pending DSC flush + * @pending_cdm_flush_mask: pending CDM flush + * @mdss_ver: MDSS revision information + * @ops: operation list + */ +struct dpu_hw_ctl { + struct dpu_hw_blk base; + struct dpu_hw_blk_reg_map hw; + + /* ctl path */ + int idx; + const struct dpu_ctl_cfg *caps; + int mixer_count; + const struct dpu_lm_cfg *mixer_hw_caps; + u32 pending_flush_mask; + u32 pending_intf_flush_mask; + u32 pending_wb_flush_mask; + u32 pending_cwb_flush_mask; + u32 pending_periph_flush_mask; + u32 pending_merge_3d_flush_mask; + u32 pending_dspp_flush_mask[DSPP_MAX - DSPP_0]; + u32 pending_dsc_flush_mask; + u32 pending_cdm_flush_mask; + + const struct dpu_mdss_version *mdss_ver; + + /* ops */ + struct dpu_hw_ctl_ops ops; +}; + +/** + * to_dpu_hw_ctl - convert base object dpu_hw_base to container + * @hw: Pointer to base hardware block + * return: Pointer to hardware block container + */ +static inline struct dpu_hw_ctl *to_dpu_hw_ctl(struct dpu_hw_blk *hw) +{ + return container_of(hw, struct dpu_hw_ctl, base); +} + +struct dpu_hw_ctl *dpu_hw_ctl_init(struct drm_device *dev, + const struct dpu_ctl_cfg *cfg, + void __iomem *addr, + const struct dpu_mdss_version *mdss_ver, + u32 mixer_count, + const struct dpu_lm_cfg *mixer); + +#endif /*_DPU_HW_CTL_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cwb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cwb.c new file mode 100644 index 000000000..ae785f4ff --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cwb.c @@ -0,0 +1,75 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2024 Qualcomm Innovation Center, Inc. All rights reserved + */ + +#include +#include "dpu_hw_cwb.h" + +#include + +#define CWB_MUX 0x000 +#define CWB_MODE 0x004 + +/* CWB mux block bit definitions */ +#define CWB_MUX_MASK GENMASK(3, 0) +#define CWB_MODE_MASK GENMASK(2, 0) + +static void dpu_hw_cwb_config(struct dpu_hw_cwb *ctx, + struct dpu_hw_cwb_setup_cfg *cwb_cfg) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + int cwb_mux_cfg = 0xF; + enum dpu_pingpong pp; + enum cwb_mode_input input; + + if (!cwb_cfg) + return; + + input = cwb_cfg->input; + pp = cwb_cfg->pp_idx; + + if (input >= INPUT_MODE_MAX) + return; + + /* + * The CWB_MUX register takes the pingpong index for the real-time + * display + */ + if ((pp != PINGPONG_NONE) && (pp < PINGPONG_MAX)) + cwb_mux_cfg = FIELD_PREP(CWB_MUX_MASK, pp - PINGPONG_0); + + input = FIELD_PREP(CWB_MODE_MASK, input); + + DPU_REG_WRITE(c, CWB_MUX, cwb_mux_cfg); + DPU_REG_WRITE(c, CWB_MODE, input); +} + +/** + * dpu_hw_cwb_init() - Initializes the writeback hw driver object with cwb. + * @dev: Corresponding device for devres management + * @cfg: wb_path catalog entry for which driver object is required + * @addr: mapped register io address of MDP + * Return: Error code or allocated dpu_hw_wb context + */ +struct dpu_hw_cwb *dpu_hw_cwb_init(struct drm_device *dev, + const struct dpu_cwb_cfg *cfg, + void __iomem *addr) +{ + struct dpu_hw_cwb *c; + + if (!addr) + return ERR_PTR(-EINVAL); + + c = drmm_kzalloc(dev, sizeof(*c), GFP_KERNEL); + if (!c) + return ERR_PTR(-ENOMEM); + + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_CWB; + + c->idx = cfg->id; + c->ops.config_cwb = dpu_hw_cwb_config; + + return c; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cwb.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cwb.h new file mode 100644 index 000000000..ed7bfcee7 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_cwb.h @@ -0,0 +1,69 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2024 Qualcomm Innovation Center, Inc. All rights reserved + */ + +#ifndef _DPU_HW_CWB_H +#define _DPU_HW_CWB_H + +#include "dpu_hw_util.h" + +struct dpu_hw_cwb; + +enum cwb_mode_input { + INPUT_MODE_LM_OUT, + INPUT_MODE_DSPP_OUT, + INPUT_MODE_MAX +}; + +/** + * struct dpu_hw_cwb_setup_cfg : Describes configuration for CWB mux + * @pp_idx: Index of the real-time pinpong that the CWB mux will + * feed the CWB mux + * @input: Input tap point + */ +struct dpu_hw_cwb_setup_cfg { + enum dpu_pingpong pp_idx; + enum cwb_mode_input input; +}; + +/** + * struct dpu_hw_cwb_ops : Interface to the cwb hw driver functions + * @config_cwb: configure CWB mux + */ +struct dpu_hw_cwb_ops { + void (*config_cwb)(struct dpu_hw_cwb *ctx, + struct dpu_hw_cwb_setup_cfg *cwb_cfg); +}; + +/** + * struct dpu_hw_cwb : CWB mux driver object + * @base: Hardware block base structure + * @hw: Block hardware details + * @idx: CWB index + * @ops: handle to operations possible for this CWB + */ +struct dpu_hw_cwb { + struct dpu_hw_blk base; + struct dpu_hw_blk_reg_map hw; + + enum dpu_cwb idx; + + struct dpu_hw_cwb_ops ops; +}; + +/** + * to_dpu_hw_cwb - convert base object dpu_hw_base to container + * @hw: Pointer to base hardware block + * return: Pointer to hardware block container + */ +static inline struct dpu_hw_cwb *to_dpu_hw_cwb(struct dpu_hw_blk *hw) +{ + return container_of(hw, struct dpu_hw_cwb, base); +} + +struct dpu_hw_cwb *dpu_hw_cwb_init(struct drm_device *dev, + const struct dpu_cwb_cfg *cfg, + void __iomem *addr); + +#endif /*_DPU_HW_CWB_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c new file mode 100644 index 000000000..3a149caa7 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c @@ -0,0 +1,216 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2020-2022, Linaro Limited + */ + +#include + +#include + +#include "dpu_kms.h" +#include "dpu_hw_catalog.h" +#include "dpu_hwio.h" +#include "dpu_hw_mdss.h" +#include "dpu_hw_dsc.h" + +#define DSC_COMMON_MODE 0x000 +#define DSC_ENC 0x004 +#define DSC_PICTURE 0x008 +#define DSC_SLICE 0x00C +#define DSC_CHUNK_SIZE 0x010 +#define DSC_DELAY 0x014 +#define DSC_SCALE_INITIAL 0x018 +#define DSC_SCALE_DEC_INTERVAL 0x01C +#define DSC_SCALE_INC_INTERVAL 0x020 +#define DSC_FIRST_LINE_BPG_OFFSET 0x024 +#define DSC_BPG_OFFSET 0x028 +#define DSC_DSC_OFFSET 0x02C +#define DSC_FLATNESS 0x030 +#define DSC_RC_MODEL_SIZE 0x034 +#define DSC_RC 0x038 +#define DSC_RC_BUF_THRESH 0x03C +#define DSC_RANGE_MIN_QP 0x074 +#define DSC_RANGE_MAX_QP 0x0B0 +#define DSC_RANGE_BPG_OFFSET 0x0EC + +#define DSC_CTL(m) (0x1800 - 0x3FC * (m - DSC_0)) + +static void dpu_hw_dsc_disable(struct dpu_hw_dsc *dsc) +{ + struct dpu_hw_blk_reg_map *c = &dsc->hw; + + DPU_REG_WRITE(c, DSC_COMMON_MODE, 0); +} + +static void dpu_hw_dsc_config(struct dpu_hw_dsc *hw_dsc, + struct drm_dsc_config *dsc, + u32 mode, + u32 initial_lines) +{ + struct dpu_hw_blk_reg_map *c = &hw_dsc->hw; + u32 data; + u32 slice_last_group_size; + u32 det_thresh_flatness; + bool is_cmd_mode = !(mode & DSC_MODE_VIDEO); + bool input_10_bits = dsc->bits_per_component == 10; + + DPU_REG_WRITE(c, DSC_COMMON_MODE, mode); + + if (is_cmd_mode) + initial_lines += 1; + + slice_last_group_size = (dsc->slice_width + 2) % 3; + + data = (initial_lines << 20); + data |= (slice_last_group_size << 18); + /* bpp is 6.4 format, 4 LSBs bits are for fractional part */ + data |= (dsc->bits_per_pixel << 8); + data |= (dsc->block_pred_enable << 7); + data |= (dsc->line_buf_depth << 3); + data |= (dsc->simple_422 << 2); + data |= (dsc->convert_rgb << 1); + data |= input_10_bits; + + DPU_REG_WRITE(c, DSC_ENC, data); + + data = dsc->pic_width << 16; + data |= dsc->pic_height; + DPU_REG_WRITE(c, DSC_PICTURE, data); + + data = dsc->slice_width << 16; + data |= dsc->slice_height; + DPU_REG_WRITE(c, DSC_SLICE, data); + + data = dsc->slice_chunk_size << 16; + DPU_REG_WRITE(c, DSC_CHUNK_SIZE, data); + + data = dsc->initial_dec_delay << 16; + data |= dsc->initial_xmit_delay; + DPU_REG_WRITE(c, DSC_DELAY, data); + + data = dsc->initial_scale_value; + DPU_REG_WRITE(c, DSC_SCALE_INITIAL, data); + + data = dsc->scale_decrement_interval; + DPU_REG_WRITE(c, DSC_SCALE_DEC_INTERVAL, data); + + data = dsc->scale_increment_interval; + DPU_REG_WRITE(c, DSC_SCALE_INC_INTERVAL, data); + + data = dsc->first_line_bpg_offset; + DPU_REG_WRITE(c, DSC_FIRST_LINE_BPG_OFFSET, data); + + data = dsc->nfl_bpg_offset << 16; + data |= dsc->slice_bpg_offset; + DPU_REG_WRITE(c, DSC_BPG_OFFSET, data); + + data = dsc->initial_offset << 16; + data |= dsc->final_offset; + DPU_REG_WRITE(c, DSC_DSC_OFFSET, data); + + det_thresh_flatness = drm_dsc_flatness_det_thresh(dsc); + data = det_thresh_flatness << 10; + data |= dsc->flatness_max_qp << 5; + data |= dsc->flatness_min_qp; + DPU_REG_WRITE(c, DSC_FLATNESS, data); + + data = dsc->rc_model_size; + DPU_REG_WRITE(c, DSC_RC_MODEL_SIZE, data); + + data = dsc->rc_tgt_offset_low << 18; + data |= dsc->rc_tgt_offset_high << 14; + data |= dsc->rc_quant_incr_limit1 << 9; + data |= dsc->rc_quant_incr_limit0 << 4; + data |= dsc->rc_edge_factor; + DPU_REG_WRITE(c, DSC_RC, data); +} + +static void dpu_hw_dsc_config_thresh(struct dpu_hw_dsc *hw_dsc, + struct drm_dsc_config *dsc) +{ + struct drm_dsc_rc_range_parameters *rc = dsc->rc_range_params; + struct dpu_hw_blk_reg_map *c = &hw_dsc->hw; + u32 off; + int i; + + off = DSC_RC_BUF_THRESH; + for (i = 0; i < DSC_NUM_BUF_RANGES - 1 ; i++) { + DPU_REG_WRITE(c, off, dsc->rc_buf_thresh[i]); + off += 4; + } + + off = DSC_RANGE_MIN_QP; + for (i = 0; i < DSC_NUM_BUF_RANGES; i++) { + DPU_REG_WRITE(c, off, rc[i].range_min_qp); + off += 4; + } + + off = DSC_RANGE_MAX_QP; + for (i = 0; i < 15; i++) { + DPU_REG_WRITE(c, off, rc[i].range_max_qp); + off += 4; + } + + off = DSC_RANGE_BPG_OFFSET; + for (i = 0; i < 15; i++) { + DPU_REG_WRITE(c, off, rc[i].range_bpg_offset); + off += 4; + } +} + +static void dpu_hw_dsc_bind_pingpong_blk( + struct dpu_hw_dsc *hw_dsc, + const enum dpu_pingpong pp) +{ + struct dpu_hw_blk_reg_map *c = &hw_dsc->hw; + int mux_cfg = 0xF; + u32 dsc_ctl_offset; + + dsc_ctl_offset = DSC_CTL(hw_dsc->idx); + + if (pp) + mux_cfg = (pp - PINGPONG_0) & 0x7; + + if (pp) + DRM_DEBUG_KMS("Binding dsc:%d to pp:%d\n", + hw_dsc->idx - DSC_0, pp - PINGPONG_0); + else + DRM_DEBUG_KMS("Unbinding dsc:%d from any pp\n", + hw_dsc->idx - DSC_0); + + DPU_REG_WRITE(c, dsc_ctl_offset, mux_cfg); +} + +/** + * dpu_hw_dsc_init() - Initializes the DSC hw driver object. + * @dev: Corresponding device for devres management + * @cfg: DSC catalog entry for which driver object is required + * @addr: Mapped register io address of MDP + * @mdss_ver: dpu core's major and minor versions + * Return: Error code or allocated dpu_hw_dsc context + */ +struct dpu_hw_dsc *dpu_hw_dsc_init(struct drm_device *dev, + const struct dpu_dsc_cfg *cfg, + void __iomem *addr, + const struct dpu_mdss_version *mdss_ver) +{ + struct dpu_hw_dsc *c; + + c = drmm_kzalloc(dev, sizeof(*c), GFP_KERNEL); + if (!c) + return ERR_PTR(-ENOMEM); + + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_DSC; + + c->idx = cfg->id; + c->caps = cfg; + + c->ops.dsc_disable = dpu_hw_dsc_disable; + c->ops.dsc_config = dpu_hw_dsc_config; + c->ops.dsc_config_thresh = dpu_hw_dsc_config_thresh; + if (mdss_ver->core_major_ver >= 5) + c->ops.dsc_bind_pingpong_blk = dpu_hw_dsc_bind_pingpong_blk; + + return c; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h new file mode 100644 index 000000000..39d93b9df --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h @@ -0,0 +1,83 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2020-2022, Linaro Limited + * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved + */ + +#ifndef _DPU_HW_DSC_H +#define _DPU_HW_DSC_H + +#include + +#define DSC_MODE_SPLIT_PANEL BIT(0) +#define DSC_MODE_MULTIPLEX BIT(1) +#define DSC_MODE_VIDEO BIT(2) + +struct dpu_hw_dsc; + +/** + * struct dpu_hw_dsc_ops - interface to the dsc hardware driver functions + * Assumption is these functions will be called after clocks are enabled + */ +struct dpu_hw_dsc_ops { + /** + * @dsc_disable: disable dsc + * @hw_dsc: Pointer to dsc context + */ + void (*dsc_disable)(struct dpu_hw_dsc *hw_dsc); + + /** + * @dsc_config: configures dsc encoder + * @hw_dsc: Pointer to dsc context + * @dsc: panel dsc parameters + * @mode: dsc topology mode to be set + * @initial_lines: amount of initial lines to be used + */ + void (*dsc_config)(struct dpu_hw_dsc *hw_dsc, + struct drm_dsc_config *dsc, + u32 mode, + u32 initial_lines); + + /** + * @dsc_config_thresh: programs panel thresholds + * @hw_dsc: Pointer to dsc context + * @dsc: panel dsc parameters + */ + void (*dsc_config_thresh)(struct dpu_hw_dsc *hw_dsc, + struct drm_dsc_config *dsc); + + /** + * @dsc_bind_pingpong_blk: binds pixel output from a DSC block + * to a pingpong block + */ + void (*dsc_bind_pingpong_blk)(struct dpu_hw_dsc *hw_dsc, + enum dpu_pingpong pp); +}; + +struct dpu_hw_dsc { + struct dpu_hw_blk base; + struct dpu_hw_blk_reg_map hw; + + /* dsc */ + enum dpu_dsc idx; + const struct dpu_dsc_cfg *caps; + + /* ops */ + struct dpu_hw_dsc_ops ops; +}; + +struct dpu_hw_dsc *dpu_hw_dsc_init(struct drm_device *dev, + const struct dpu_dsc_cfg *cfg, + void __iomem *addr, + const struct dpu_mdss_version *mdss_ver); + +struct dpu_hw_dsc *dpu_hw_dsc_init_1_2(struct drm_device *dev, + const struct dpu_dsc_cfg *cfg, + void __iomem *addr); + +static inline struct dpu_hw_dsc *to_dpu_hw_dsc(struct dpu_hw_blk *hw) +{ + return container_of(hw, struct dpu_hw_dsc, base); +} + +#endif /* _DPU_HW_DSC_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c new file mode 100644 index 000000000..b3395e9c3 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c @@ -0,0 +1,396 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2020-2021, The Linux Foundation. All rights reserved. + * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved + */ + +#include + +#include + +#include "dpu_kms.h" +#include "dpu_hw_catalog.h" +#include "dpu_hwio.h" +#include "dpu_hw_mdss.h" +#include "dpu_hw_dsc.h" + +#define DSC_CMN_MAIN_CNF 0x00 + +/* DPU_DSC_ENC register offsets */ +#define ENC_DF_CTRL 0x00 +#define ENC_GENERAL_STATUS 0x04 +#define ENC_HSLICE_STATUS 0x08 +#define ENC_OUT_STATUS 0x0C +#define ENC_INT_STAT 0x10 +#define ENC_INT_CLR 0x14 +#define ENC_INT_MASK 0x18 +#define DSC_MAIN_CONF 0x30 +#define DSC_PICTURE_SIZE 0x34 +#define DSC_SLICE_SIZE 0x38 +#define DSC_MISC_SIZE 0x3C +#define DSC_HRD_DELAYS 0x40 +#define DSC_RC_SCALE 0x44 +#define DSC_RC_SCALE_INC_DEC 0x48 +#define DSC_RC_OFFSETS_1 0x4C +#define DSC_RC_OFFSETS_2 0x50 +#define DSC_RC_OFFSETS_3 0x54 +#define DSC_RC_OFFSETS_4 0x58 +#define DSC_FLATNESS_QP 0x5C +#define DSC_RC_MODEL_SIZE 0x60 +#define DSC_RC_CONFIG 0x64 +#define DSC_RC_BUF_THRESH_0 0x68 +#define DSC_RC_BUF_THRESH_1 0x6C +#define DSC_RC_BUF_THRESH_2 0x70 +#define DSC_RC_BUF_THRESH_3 0x74 +#define DSC_RC_MIN_QP_0 0x78 +#define DSC_RC_MIN_QP_1 0x7C +#define DSC_RC_MIN_QP_2 0x80 +#define DSC_RC_MAX_QP_0 0x84 +#define DSC_RC_MAX_QP_1 0x88 +#define DSC_RC_MAX_QP_2 0x8C +#define DSC_RC_RANGE_BPG_OFFSETS_0 0x90 +#define DSC_RC_RANGE_BPG_OFFSETS_1 0x94 +#define DSC_RC_RANGE_BPG_OFFSETS_2 0x98 + +/* DPU_DSC_CTL register offsets */ +#define DSC_CTL 0x00 +#define DSC_CFG 0x04 +#define DSC_DATA_IN_SWAP 0x08 +#define DSC_CLK_CTRL 0x0C + +static int _dsc_calc_output_buf_max_addr(struct dpu_hw_dsc *hw_dsc, int num_softslice) +{ + int max_addr = 2400 / num_softslice; + + if (hw_dsc->caps->features & BIT(DPU_DSC_NATIVE_42x_EN)) + max_addr /= 2; + + return max_addr - 1; +}; + +static void dpu_hw_dsc_disable_1_2(struct dpu_hw_dsc *hw_dsc) +{ + struct dpu_hw_blk_reg_map *hw; + const struct dpu_dsc_sub_blks *sblk; + + if (!hw_dsc) + return; + + hw = &hw_dsc->hw; + sblk = hw_dsc->caps->sblk; + DPU_REG_WRITE(hw, sblk->ctl.base + DSC_CFG, 0); + + DPU_REG_WRITE(hw, sblk->enc.base + ENC_DF_CTRL, 0); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_MAIN_CONF, 0); +} + +static void dpu_hw_dsc_config_1_2(struct dpu_hw_dsc *hw_dsc, + struct drm_dsc_config *dsc, + u32 mode, + u32 initial_lines) +{ + struct dpu_hw_blk_reg_map *hw; + const struct dpu_dsc_sub_blks *sblk; + u32 data = 0; + u32 det_thresh_flatness; + u32 num_active_slice_per_enc; + u32 bpp; + + if (!hw_dsc || !dsc) + return; + + hw = &hw_dsc->hw; + + sblk = hw_dsc->caps->sblk; + + if (mode & DSC_MODE_SPLIT_PANEL) + data |= BIT(0); + + if (mode & DSC_MODE_MULTIPLEX) + data |= BIT(1); + + num_active_slice_per_enc = dsc->slice_count; + if (mode & DSC_MODE_MULTIPLEX) + num_active_slice_per_enc = dsc->slice_count / 2; + + data |= (num_active_slice_per_enc & 0x3) << 7; + + DPU_REG_WRITE(hw, DSC_CMN_MAIN_CNF, data); + + data = (initial_lines & 0xff); + + if (mode & DSC_MODE_VIDEO) + data |= BIT(9); + + data |= (_dsc_calc_output_buf_max_addr(hw_dsc, num_active_slice_per_enc) << 18); + + DPU_REG_WRITE(hw, sblk->enc.base + ENC_DF_CTRL, data); + + data = (dsc->dsc_version_minor & 0xf) << 28; + if (dsc->dsc_version_minor == 0x2) { + if (dsc->native_422) + data |= BIT(22); + if (dsc->native_420) + data |= BIT(21); + } + + bpp = dsc->bits_per_pixel; + /* as per hw requirement bpp should be programmed + * twice the actual value in case of 420 or 422 encoding + */ + if (dsc->native_422 || dsc->native_420) + bpp = 2 * bpp; + + data |= bpp << 10; + + if (dsc->block_pred_enable) + data |= BIT(20); + + if (dsc->convert_rgb) + data |= BIT(4); + + data |= (dsc->line_buf_depth & 0xf) << 6; + data |= dsc->bits_per_component & 0xf; + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_MAIN_CONF, data); + + data = (dsc->pic_width & 0xffff) | + ((dsc->pic_height & 0xffff) << 16); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_PICTURE_SIZE, data); + + data = (dsc->slice_width & 0xffff) | + ((dsc->slice_height & 0xffff) << 16); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_SLICE_SIZE, data); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_MISC_SIZE, + (dsc->slice_chunk_size) & 0xffff); + + data = (dsc->initial_xmit_delay & 0xffff) | + ((dsc->initial_dec_delay & 0xffff) << 16); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_HRD_DELAYS, data); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_SCALE, + dsc->initial_scale_value & 0x3f); + + data = (dsc->scale_increment_interval & 0xffff) | + ((dsc->scale_decrement_interval & 0x7ff) << 16); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_SCALE_INC_DEC, data); + + data = (dsc->first_line_bpg_offset & 0x1f) | + ((dsc->second_line_bpg_offset & 0x1f) << 5); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_OFFSETS_1, data); + + data = (dsc->nfl_bpg_offset & 0xffff) | + ((dsc->slice_bpg_offset & 0xffff) << 16); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_OFFSETS_2, data); + + data = (dsc->initial_offset & 0xffff) | + ((dsc->final_offset & 0xffff) << 16); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_OFFSETS_3, data); + + data = (dsc->nsl_bpg_offset & 0xffff) | + ((dsc->second_line_offset_adj & 0xffff) << 16); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_OFFSETS_4, data); + + det_thresh_flatness = drm_dsc_flatness_det_thresh(dsc); + data = (dsc->flatness_min_qp & 0x1f) | + ((dsc->flatness_max_qp & 0x1f) << 5) | + ((det_thresh_flatness & 0xff) << 10); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_FLATNESS_QP, data); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MODEL_SIZE, + (dsc->rc_model_size) & 0xffff); + + data = dsc->rc_edge_factor & 0xf; + data |= (dsc->rc_quant_incr_limit0 & 0x1f) << 8; + data |= (dsc->rc_quant_incr_limit1 & 0x1f) << 13; + data |= (dsc->rc_tgt_offset_high & 0xf) << 20; + data |= (dsc->rc_tgt_offset_low & 0xf) << 24; + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_CONFIG, data); + + /* program the dsc wrapper */ + data = BIT(0); /* encoder enable */ + if (dsc->native_422) + data |= BIT(8); + else if (dsc->native_420) + data |= BIT(9); + if (!dsc->convert_rgb) + data |= BIT(10); + if (dsc->bits_per_component == 8) + data |= BIT(11); + if (mode & DSC_MODE_SPLIT_PANEL) + data |= BIT(12); + if (mode & DSC_MODE_MULTIPLEX) + data |= BIT(13); + if (!(mode & DSC_MODE_VIDEO)) + data |= BIT(17); + + DPU_REG_WRITE(hw, sblk->ctl.base + DSC_CFG, data); +} + +static void dpu_hw_dsc_config_thresh_1_2(struct dpu_hw_dsc *hw_dsc, + struct drm_dsc_config *dsc) +{ + struct dpu_hw_blk_reg_map *hw; + const struct dpu_dsc_sub_blks *sblk; + struct drm_dsc_rc_range_parameters *rc; + + if (!hw_dsc || !dsc) + return; + + hw = &hw_dsc->hw; + + sblk = hw_dsc->caps->sblk; + + rc = dsc->rc_range_params; + + /* + * With BUF_THRESH -- 14 in total + * each register contains 4 thresh values with the last register + * containing only 2 thresh values + */ + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_BUF_THRESH_0, + (dsc->rc_buf_thresh[0] << 0) | + (dsc->rc_buf_thresh[1] << 8) | + (dsc->rc_buf_thresh[2] << 16) | + (dsc->rc_buf_thresh[3] << 24)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_BUF_THRESH_1, + (dsc->rc_buf_thresh[4] << 0) | + (dsc->rc_buf_thresh[5] << 8) | + (dsc->rc_buf_thresh[6] << 16) | + (dsc->rc_buf_thresh[7] << 24)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_BUF_THRESH_2, + (dsc->rc_buf_thresh[8] << 0) | + (dsc->rc_buf_thresh[9] << 8) | + (dsc->rc_buf_thresh[10] << 16) | + (dsc->rc_buf_thresh[11] << 24)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_BUF_THRESH_3, + (dsc->rc_buf_thresh[12] << 0) | + (dsc->rc_buf_thresh[13] << 8)); + + /* + * with min/max_QP -- 5 bits + * each register contains 5 min_qp or max_qp for total of 15 + * + * With BPG_OFFSET -- 6 bits + * each register contains 5 BPG_offset for total of 15 + */ + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MIN_QP_0, + (rc[0].range_min_qp << 0) | + (rc[1].range_min_qp << 5) | + (rc[2].range_min_qp << 10) | + (rc[3].range_min_qp << 15) | + (rc[4].range_min_qp << 20)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MAX_QP_0, + (rc[0].range_max_qp << 0) | + (rc[1].range_max_qp << 5) | + (rc[2].range_max_qp << 10) | + (rc[3].range_max_qp << 15) | + (rc[4].range_max_qp << 20)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_RANGE_BPG_OFFSETS_0, + (rc[0].range_bpg_offset << 0) | + (rc[1].range_bpg_offset << 6) | + (rc[2].range_bpg_offset << 12) | + (rc[3].range_bpg_offset << 18) | + (rc[4].range_bpg_offset << 24)); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MIN_QP_1, + (rc[5].range_min_qp << 0) | + (rc[6].range_min_qp << 5) | + (rc[7].range_min_qp << 10) | + (rc[8].range_min_qp << 15) | + (rc[9].range_min_qp << 20)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MAX_QP_1, + (rc[5].range_max_qp << 0) | + (rc[6].range_max_qp << 5) | + (rc[7].range_max_qp << 10) | + (rc[8].range_max_qp << 15) | + (rc[9].range_max_qp << 20)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_RANGE_BPG_OFFSETS_1, + (rc[5].range_bpg_offset << 0) | + (rc[6].range_bpg_offset << 6) | + (rc[7].range_bpg_offset << 12) | + (rc[8].range_bpg_offset << 18) | + (rc[9].range_bpg_offset << 24)); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MIN_QP_2, + (rc[10].range_min_qp << 0) | + (rc[11].range_min_qp << 5) | + (rc[12].range_min_qp << 10) | + (rc[13].range_min_qp << 15) | + (rc[14].range_min_qp << 20)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MAX_QP_2, + (rc[10].range_max_qp << 0) | + (rc[11].range_max_qp << 5) | + (rc[12].range_max_qp << 10) | + (rc[13].range_max_qp << 15) | + (rc[14].range_max_qp << 20)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_RANGE_BPG_OFFSETS_2, + (rc[10].range_bpg_offset << 0) | + (rc[11].range_bpg_offset << 6) | + (rc[12].range_bpg_offset << 12) | + (rc[13].range_bpg_offset << 18) | + (rc[14].range_bpg_offset << 24)); +} + +static void dpu_hw_dsc_bind_pingpong_blk_1_2(struct dpu_hw_dsc *hw_dsc, + const enum dpu_pingpong pp) +{ + struct dpu_hw_blk_reg_map *hw; + const struct dpu_dsc_sub_blks *sblk; + int mux_cfg = 0xf; /* Disabled */ + + hw = &hw_dsc->hw; + + sblk = hw_dsc->caps->sblk; + + if (pp) + mux_cfg = (pp - PINGPONG_0) & 0x7; + + DPU_REG_WRITE(hw, sblk->ctl.base + DSC_CTL, mux_cfg); +} + +static void _setup_dcs_ops_1_2(struct dpu_hw_dsc_ops *ops) +{ + ops->dsc_disable = dpu_hw_dsc_disable_1_2; + ops->dsc_config = dpu_hw_dsc_config_1_2; + ops->dsc_config_thresh = dpu_hw_dsc_config_thresh_1_2; + ops->dsc_bind_pingpong_blk = dpu_hw_dsc_bind_pingpong_blk_1_2; +} + +/** + * dpu_hw_dsc_init_1_2() - initializes the v1.2 DSC hw driver object + * @dev: Corresponding device for devres management + * @cfg: DSC catalog entry for which driver object is required + * @addr: Mapped register io address of MDP + * Returns: Error code or allocated dpu_hw_dsc context + */ +struct dpu_hw_dsc *dpu_hw_dsc_init_1_2(struct drm_device *dev, + const struct dpu_dsc_cfg *cfg, + void __iomem *addr) +{ + struct dpu_hw_dsc *c; + + c = drmm_kzalloc(dev, sizeof(*c), GFP_KERNEL); + if (!c) + return ERR_PTR(-ENOMEM); + + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_DSC; + + c->idx = cfg->id; + c->caps = cfg; + _setup_dcs_ops_1_2(&c->ops); + + return c; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c new file mode 100644 index 000000000..23dcbe1ce --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c @@ -0,0 +1,151 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#include + +#include "dpu_hwio.h" +#include "dpu_hw_catalog.h" +#include "dpu_hw_lm.h" +#include "dpu_hw_dspp.h" +#include "dpu_kms.h" + + +/* DSPP_PCC */ +#define PCC_EN BIT(0) +#define PCC_DIS 0 +#define PCC_RED_R_OFF 0x10 +#define PCC_RED_G_OFF 0x1C +#define PCC_RED_B_OFF 0x28 +#define PCC_GREEN_R_OFF 0x14 +#define PCC_GREEN_G_OFF 0x20 +#define PCC_GREEN_B_OFF 0x2C +#define PCC_BLUE_R_OFF 0x18 +#define PCC_BLUE_G_OFF 0x24 +#define PCC_BLUE_B_OFF 0x30 + +/* DSPP_GC */ +#define GC_EN BIT(0) +#define GC_DIS 0 +#define GC_8B_ROUND_EN BIT(1) +#define GC_LUT_SWAP_OFF 0x1c +#define GC_C0_OFF 0x4 +#define GC_C1_OFF 0xc +#define GC_C2_OFF 0x14 +#define GC_C0_INDEX_OFF 0x8 +#define GC_C1_INDEX_OFF 0x10 +#define GC_C2_INDEX_OFF 0x18 + +static void dpu_setup_dspp_pcc(struct dpu_hw_dspp *ctx, + struct dpu_hw_pcc_cfg *cfg) +{ + + u32 base; + + if (!ctx) { + DRM_ERROR("invalid ctx %p\n", ctx); + return; + } + + base = ctx->cap->sblk->pcc.base; + + if (!base) { + DRM_ERROR("invalid ctx %p pcc base 0x%x\n", ctx, base); + return; + } + + if (!cfg) { + DRM_DEBUG_DRIVER("disable pcc feature\n"); + DPU_REG_WRITE(&ctx->hw, base, PCC_DIS); + return; + } + + DPU_REG_WRITE(&ctx->hw, base + PCC_RED_R_OFF, cfg->r.r); + DPU_REG_WRITE(&ctx->hw, base + PCC_RED_G_OFF, cfg->r.g); + DPU_REG_WRITE(&ctx->hw, base + PCC_RED_B_OFF, cfg->r.b); + + DPU_REG_WRITE(&ctx->hw, base + PCC_GREEN_R_OFF, cfg->g.r); + DPU_REG_WRITE(&ctx->hw, base + PCC_GREEN_G_OFF, cfg->g.g); + DPU_REG_WRITE(&ctx->hw, base + PCC_GREEN_B_OFF, cfg->g.b); + + DPU_REG_WRITE(&ctx->hw, base + PCC_BLUE_R_OFF, cfg->b.r); + DPU_REG_WRITE(&ctx->hw, base + PCC_BLUE_G_OFF, cfg->b.g); + DPU_REG_WRITE(&ctx->hw, base + PCC_BLUE_B_OFF, cfg->b.b); + + DPU_REG_WRITE(&ctx->hw, base, PCC_EN); +} + +static void dpu_setup_dspp_gc(struct dpu_hw_dspp *ctx, + struct dpu_hw_gc_lut *gc_lut) +{ + int i = 0; + u32 base, reg; + + if (!ctx) { + DRM_ERROR("invalid ctx\n"); + return; + } + + base = ctx->cap->sblk->gc.base; + + if (!base) { + DRM_ERROR("invalid ctx %p gc base\n", ctx); + return; + } + + if (!gc_lut) { + DRM_DEBUG_DRIVER("disable gc feature\n"); + DPU_REG_WRITE(&ctx->hw, base, GC_DIS); + return; + } + + DPU_REG_WRITE(&ctx->hw, base + GC_C0_INDEX_OFF, 0); + DPU_REG_WRITE(&ctx->hw, base + GC_C1_INDEX_OFF, 0); + DPU_REG_WRITE(&ctx->hw, base + GC_C2_INDEX_OFF, 0); + + for (i = 0; i < PGC_TBL_LEN; i++) { + DPU_REG_WRITE(&ctx->hw, base + GC_C0_OFF, gc_lut->c0[i]); + DPU_REG_WRITE(&ctx->hw, base + GC_C1_OFF, gc_lut->c1[i]); + DPU_REG_WRITE(&ctx->hw, base + GC_C2_OFF, gc_lut->c2[i]); + } + + DPU_REG_WRITE(&ctx->hw, base + GC_LUT_SWAP_OFF, BIT(0)); + + reg = GC_EN | ((gc_lut->flags & PGC_8B_ROUND) ? GC_8B_ROUND_EN : 0); + DPU_REG_WRITE(&ctx->hw, base, reg); +} + +/** + * dpu_hw_dspp_init() - Initializes the DSPP hw driver object. + * should be called once before accessing every DSPP. + * @dev: Corresponding device for devres management + * @cfg: DSPP catalog entry for which driver object is required + * @addr: Mapped register io address of MDP + * Return: pointer to structure or ERR_PTR + */ +struct dpu_hw_dspp *dpu_hw_dspp_init(struct drm_device *dev, + const struct dpu_dspp_cfg *cfg, + void __iomem *addr) +{ + struct dpu_hw_dspp *c; + + if (!addr) + return ERR_PTR(-EINVAL); + + c = drmm_kzalloc(dev, sizeof(*c), GFP_KERNEL); + if (!c) + return ERR_PTR(-ENOMEM); + + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_DSPP; + + /* Assign ops */ + c->idx = cfg->id; + c->cap = cfg; + if (c->cap->sblk->pcc.base) + c->ops.setup_pcc = dpu_setup_dspp_pcc; + if (c->cap->sblk->gc.base) + c->ops.setup_gc = dpu_setup_dspp_gc; + + return c; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.h new file mode 100644 index 000000000..b47b77880 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.h @@ -0,0 +1,112 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_HW_DSPP_H +#define _DPU_HW_DSPP_H + +struct dpu_hw_dspp; + +/** + * struct dpu_hw_pcc_coeff - PCC coefficient structure for each color + * component. + * @r: red coefficient. + * @g: green coefficient. + * @b: blue coefficient. + */ + +struct dpu_hw_pcc_coeff { + __u32 r; + __u32 g; + __u32 b; +}; + +/** + * struct dpu_hw_pcc_cfg - pcc feature structure + * @r: red coefficients. + * @g: green coefficients. + * @b: blue coefficients. + */ +struct dpu_hw_pcc_cfg { + struct dpu_hw_pcc_coeff r; + struct dpu_hw_pcc_coeff g; + struct dpu_hw_pcc_coeff b; +}; + +#define DPU_GAMMA_LUT_SIZE 1024 +#define PGC_TBL_LEN 512 +#define PGC_8B_ROUND BIT(0) + +/** + * struct dpu_hw_gc_lut - gc lut feature structure + * @flags: flags for the feature values can be: + * - PGC_8B_ROUND + * @c0: color0 component lut + * @c1: color1 component lut + * @c2: color2 component lut + */ +struct dpu_hw_gc_lut { + __u64 flags; + __u32 c0[PGC_TBL_LEN]; + __u32 c1[PGC_TBL_LEN]; + __u32 c2[PGC_TBL_LEN]; +}; + +/** + * struct dpu_hw_dspp_ops - interface to the dspp hardware driver functions + * Caller must call the init function to get the dspp context for each dspp + * Assumption is these functions will be called after clocks are enabled + */ +struct dpu_hw_dspp_ops { + /** + * @setup_pcc: setup_pcc - setup dspp pcc + * @ctx: Pointer to dspp context + * @cfg: Pointer to configuration + */ + void (*setup_pcc)(struct dpu_hw_dspp *ctx, struct dpu_hw_pcc_cfg *cfg); + + /** + * setup_gc - setup dspp gc + * @ctx: Pointer to dspp context + * @gc_lut: Pointer to lut content + */ + void (*setup_gc)(struct dpu_hw_dspp *ctx, struct dpu_hw_gc_lut *gc_lut); + +}; + +/** + * struct dpu_hw_dspp - dspp description + * @base: Hardware block base structure + * @hw: Block hardware details + * @idx: DSPP index + * @cap: Pointer to layer_cfg + * @ops: Pointer to operations possible for this DSPP + */ +struct dpu_hw_dspp { + struct dpu_hw_blk base; + struct dpu_hw_blk_reg_map hw; + + /* dspp */ + int idx; + const struct dpu_dspp_cfg *cap; + + /* Ops */ + struct dpu_hw_dspp_ops ops; +}; + +/** + * to_dpu_hw_dspp - convert base object dpu_hw_base to container + * @hw: Pointer to base hardware block + * return: Pointer to hardware block container + */ +static inline struct dpu_hw_dspp *to_dpu_hw_dspp(struct dpu_hw_blk *hw) +{ + return container_of(hw, struct dpu_hw_dspp, base); +} + +struct dpu_hw_dspp *dpu_hw_dspp_init(struct drm_device *dev, + const struct dpu_dspp_cfg *cfg, + void __iomem *addr); + +#endif /*_DPU_HW_DSPP_H */ + diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c new file mode 100644 index 000000000..5b7cd5241 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c @@ -0,0 +1,815 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2016-2018, The Linux Foundation. All rights reserved. + */ + +#include +#include +#include + +#include + +#include "dpu_core_irq.h" +#include "dpu_kms.h" +#include "dpu_hw_interrupts.h" +#include "dpu_hw_util.h" +#include "dpu_hw_mdss.h" +#include "dpu_trace.h" + +/* + * Register offsets in MDSS register file for the interrupt registers + * w.r.t. the MDP base + */ +#define MDP_INTF_OFF(intf) (0x6A000 + 0x800 * (intf)) +#define MDP_INTF_INTR_EN(intf) (MDP_INTF_OFF(intf) + 0x1c0) +#define MDP_INTF_INTR_STATUS(intf) (MDP_INTF_OFF(intf) + 0x1c4) +#define MDP_INTF_INTR_CLEAR(intf) (MDP_INTF_OFF(intf) + 0x1c8) +#define MDP_INTF_TEAR_OFF(intf) (0x6D700 + 0x100 * (intf)) +#define MDP_INTF_INTR_TEAR_EN(intf) (MDP_INTF_TEAR_OFF(intf) + 0x000) +#define MDP_INTF_INTR_TEAR_STATUS(intf) (MDP_INTF_TEAR_OFF(intf) + 0x004) +#define MDP_INTF_INTR_TEAR_CLEAR(intf) (MDP_INTF_TEAR_OFF(intf) + 0x008) +#define MDP_AD4_OFF(ad4) (0x7C000 + 0x1000 * (ad4)) +#define MDP_AD4_INTR_EN_OFF(ad4) (MDP_AD4_OFF(ad4) + 0x41c) +#define MDP_AD4_INTR_CLEAR_OFF(ad4) (MDP_AD4_OFF(ad4) + 0x424) +#define MDP_AD4_INTR_STATUS_OFF(ad4) (MDP_AD4_OFF(ad4) + 0x420) +#define MDP_INTF_REV_7xxx_OFF(intf) (0x34000 + 0x1000 * (intf)) +#define MDP_INTF_REV_7xxx_INTR_EN(intf) (MDP_INTF_REV_7xxx_OFF(intf) + 0x1c0) +#define MDP_INTF_REV_7xxx_INTR_STATUS(intf) (MDP_INTF_REV_7xxx_OFF(intf) + 0x1c4) +#define MDP_INTF_REV_7xxx_INTR_CLEAR(intf) (MDP_INTF_REV_7xxx_OFF(intf) + 0x1c8) +#define MDP_INTF_REV_7xxx_TEAR_OFF(intf) (0x34800 + 0x1000 * (intf)) +#define MDP_INTF_REV_7xxx_INTR_TEAR_EN(intf) (MDP_INTF_REV_7xxx_TEAR_OFF(intf) + 0x000) +#define MDP_INTF_REV_7xxx_INTR_TEAR_STATUS(intf) (MDP_INTF_REV_7xxx_TEAR_OFF(intf) + 0x004) +#define MDP_INTF_REV_7xxx_INTR_TEAR_CLEAR(intf) (MDP_INTF_REV_7xxx_TEAR_OFF(intf) + 0x008) + +#define MDP_INTF_REV_13xx_OFF(intf) (0x18d000 + 0x1000 * (intf)) +#define MDP_INTF_REV_13xx_INTR_EN(intf) (MDP_INTF_REV_13xx_OFF(intf) + 0x1c0) +#define MDP_INTF_REV_13xx_INTR_STATUS(intf) (MDP_INTF_REV_13xx_OFF(intf) + 0x1c4) +#define MDP_INTF_REV_13xx_INTR_CLEAR(intf) (MDP_INTF_REV_13xx_OFF(intf) + 0x1c8) +#define MDP_INTF_REV_13xx_TEAR_OFF(intf) (0x18d800 + 0x1000 * (intf)) +#define MDP_INTF_REV_13xx_INTR_TEAR_EN(intf) (MDP_INTF_REV_13xx_TEAR_OFF(intf) + 0x000) +#define MDP_INTF_REV_13xx_INTR_TEAR_STATUS(intf) (MDP_INTF_REV_13xx_TEAR_OFF(intf) + 0x004) +#define MDP_INTF_REV_13xx_INTR_TEAR_CLEAR(intf) (MDP_INTF_REV_13xx_TEAR_OFF(intf) + 0x008) + +/** + * struct dpu_intr_reg - array of DPU register sets + * @clr_off: offset to CLEAR reg + * @en_off: offset to ENABLE reg + * @status_off: offset to STATUS reg + */ +struct dpu_intr_reg { + u32 clr_off; + u32 en_off; + u32 status_off; +}; + +/* + * dpu_intr_set_legacy - List of DPU interrupt registers for DPU <= 6.x + */ +static const struct dpu_intr_reg dpu_intr_set_legacy[] = { + [MDP_SSPP_TOP0_INTR] = { + INTR_CLEAR, + INTR_EN, + INTR_STATUS + }, + [MDP_SSPP_TOP0_INTR2] = { + INTR2_CLEAR, + INTR2_EN, + INTR2_STATUS + }, + [MDP_SSPP_TOP0_HIST_INTR] = { + HIST_INTR_CLEAR, + HIST_INTR_EN, + HIST_INTR_STATUS + }, + [MDP_INTF0_INTR] = { + MDP_INTF_INTR_CLEAR(0), + MDP_INTF_INTR_EN(0), + MDP_INTF_INTR_STATUS(0) + }, + [MDP_INTF1_INTR] = { + MDP_INTF_INTR_CLEAR(1), + MDP_INTF_INTR_EN(1), + MDP_INTF_INTR_STATUS(1) + }, + [MDP_INTF2_INTR] = { + MDP_INTF_INTR_CLEAR(2), + MDP_INTF_INTR_EN(2), + MDP_INTF_INTR_STATUS(2) + }, + [MDP_INTF3_INTR] = { + MDP_INTF_INTR_CLEAR(3), + MDP_INTF_INTR_EN(3), + MDP_INTF_INTR_STATUS(3) + }, + [MDP_INTF4_INTR] = { + MDP_INTF_INTR_CLEAR(4), + MDP_INTF_INTR_EN(4), + MDP_INTF_INTR_STATUS(4) + }, + [MDP_INTF5_INTR] = { + MDP_INTF_INTR_CLEAR(5), + MDP_INTF_INTR_EN(5), + MDP_INTF_INTR_STATUS(5) + }, + [MDP_INTF1_TEAR_INTR] = { + MDP_INTF_INTR_TEAR_CLEAR(1), + MDP_INTF_INTR_TEAR_EN(1), + MDP_INTF_INTR_TEAR_STATUS(1) + }, + [MDP_INTF2_TEAR_INTR] = { + MDP_INTF_INTR_TEAR_CLEAR(2), + MDP_INTF_INTR_TEAR_EN(2), + MDP_INTF_INTR_TEAR_STATUS(2) + }, + [MDP_AD4_0_INTR] = { + MDP_AD4_INTR_CLEAR_OFF(0), + MDP_AD4_INTR_EN_OFF(0), + MDP_AD4_INTR_STATUS_OFF(0), + }, + [MDP_AD4_1_INTR] = { + MDP_AD4_INTR_CLEAR_OFF(1), + MDP_AD4_INTR_EN_OFF(1), + MDP_AD4_INTR_STATUS_OFF(1), + }, +}; + +/* + * dpu_intr_set_7xxx - List of DPU interrupt registers for DPU >= 7.0 + */ +static const struct dpu_intr_reg dpu_intr_set_7xxx[] = { + [MDP_SSPP_TOP0_INTR] = { + INTR_CLEAR, + INTR_EN, + INTR_STATUS + }, + [MDP_SSPP_TOP0_INTR2] = { + INTR2_CLEAR, + INTR2_EN, + INTR2_STATUS + }, + [MDP_SSPP_TOP0_HIST_INTR] = { + HIST_INTR_CLEAR, + HIST_INTR_EN, + HIST_INTR_STATUS + }, + [MDP_INTF0_INTR] = { + MDP_INTF_REV_7xxx_INTR_CLEAR(0), + MDP_INTF_REV_7xxx_INTR_EN(0), + MDP_INTF_REV_7xxx_INTR_STATUS(0) + }, + [MDP_INTF1_INTR] = { + MDP_INTF_REV_7xxx_INTR_CLEAR(1), + MDP_INTF_REV_7xxx_INTR_EN(1), + MDP_INTF_REV_7xxx_INTR_STATUS(1) + }, + [MDP_INTF1_TEAR_INTR] = { + MDP_INTF_REV_7xxx_INTR_TEAR_CLEAR(1), + MDP_INTF_REV_7xxx_INTR_TEAR_EN(1), + MDP_INTF_REV_7xxx_INTR_TEAR_STATUS(1) + }, + [MDP_INTF2_INTR] = { + MDP_INTF_REV_7xxx_INTR_CLEAR(2), + MDP_INTF_REV_7xxx_INTR_EN(2), + MDP_INTF_REV_7xxx_INTR_STATUS(2) + }, + [MDP_INTF2_TEAR_INTR] = { + MDP_INTF_REV_7xxx_INTR_TEAR_CLEAR(2), + MDP_INTF_REV_7xxx_INTR_TEAR_EN(2), + MDP_INTF_REV_7xxx_INTR_TEAR_STATUS(2) + }, + [MDP_INTF3_INTR] = { + MDP_INTF_REV_7xxx_INTR_CLEAR(3), + MDP_INTF_REV_7xxx_INTR_EN(3), + MDP_INTF_REV_7xxx_INTR_STATUS(3) + }, + [MDP_INTF4_INTR] = { + MDP_INTF_REV_7xxx_INTR_CLEAR(4), + MDP_INTF_REV_7xxx_INTR_EN(4), + MDP_INTF_REV_7xxx_INTR_STATUS(4) + }, + [MDP_INTF5_INTR] = { + MDP_INTF_REV_7xxx_INTR_CLEAR(5), + MDP_INTF_REV_7xxx_INTR_EN(5), + MDP_INTF_REV_7xxx_INTR_STATUS(5) + }, + [MDP_INTF6_INTR] = { + MDP_INTF_REV_7xxx_INTR_CLEAR(6), + MDP_INTF_REV_7xxx_INTR_EN(6), + MDP_INTF_REV_7xxx_INTR_STATUS(6) + }, + [MDP_INTF7_INTR] = { + MDP_INTF_REV_7xxx_INTR_CLEAR(7), + MDP_INTF_REV_7xxx_INTR_EN(7), + MDP_INTF_REV_7xxx_INTR_STATUS(7) + }, + [MDP_INTF8_INTR] = { + MDP_INTF_REV_7xxx_INTR_CLEAR(8), + MDP_INTF_REV_7xxx_INTR_EN(8), + MDP_INTF_REV_7xxx_INTR_STATUS(8) + }, +}; + +/* + * dpu_intr_set_13xx - List of DPU interrupt registers for DPU >= 13.0 + */ +static const struct dpu_intr_reg dpu_intr_set_13xx[] = { + [MDP_SSPP_TOP0_INTR] = { + INTR_CLEAR, + INTR_EN, + INTR_STATUS + }, + [MDP_SSPP_TOP0_INTR2] = { + INTR2_CLEAR, + INTR2_EN, + INTR2_STATUS + }, + [MDP_SSPP_TOP0_HIST_INTR] = { + HIST_INTR_CLEAR, + HIST_INTR_EN, + HIST_INTR_STATUS + }, + [MDP_INTF0_INTR] = { + MDP_INTF_REV_13xx_INTR_CLEAR(0), + MDP_INTF_REV_13xx_INTR_EN(0), + MDP_INTF_REV_13xx_INTR_STATUS(0) + }, + [MDP_INTF1_INTR] = { + MDP_INTF_REV_13xx_INTR_CLEAR(1), + MDP_INTF_REV_13xx_INTR_EN(1), + MDP_INTF_REV_13xx_INTR_STATUS(1) + }, + [MDP_INTF1_TEAR_INTR] = { + MDP_INTF_REV_13xx_INTR_TEAR_CLEAR(1), + MDP_INTF_REV_13xx_INTR_TEAR_EN(1), + MDP_INTF_REV_13xx_INTR_TEAR_STATUS(1) + }, + [MDP_INTF2_INTR] = { + MDP_INTF_REV_13xx_INTR_CLEAR(2), + MDP_INTF_REV_13xx_INTR_EN(2), + MDP_INTF_REV_13xx_INTR_STATUS(2) + }, + [MDP_INTF2_TEAR_INTR] = { + MDP_INTF_REV_13xx_INTR_TEAR_CLEAR(2), + MDP_INTF_REV_13xx_INTR_TEAR_EN(2), + MDP_INTF_REV_13xx_INTR_TEAR_STATUS(2) + }, + [MDP_INTF3_INTR] = { + MDP_INTF_REV_13xx_INTR_CLEAR(3), + MDP_INTF_REV_13xx_INTR_EN(3), + MDP_INTF_REV_13xx_INTR_STATUS(3) + }, + [MDP_INTF4_INTR] = { + MDP_INTF_REV_13xx_INTR_CLEAR(4), + MDP_INTF_REV_13xx_INTR_EN(4), + MDP_INTF_REV_13xx_INTR_STATUS(4) + }, + [MDP_INTF5_INTR] = { + MDP_INTF_REV_13xx_INTR_CLEAR(5), + MDP_INTF_REV_13xx_INTR_EN(5), + MDP_INTF_REV_13xx_INTR_STATUS(5) + }, + [MDP_INTF6_INTR] = { + MDP_INTF_REV_13xx_INTR_CLEAR(6), + MDP_INTF_REV_13xx_INTR_EN(6), + MDP_INTF_REV_13xx_INTR_STATUS(6) + }, + [MDP_INTF7_INTR] = { + MDP_INTF_REV_13xx_INTR_CLEAR(7), + MDP_INTF_REV_13xx_INTR_EN(7), + MDP_INTF_REV_13xx_INTR_STATUS(7) + }, + [MDP_INTF8_INTR] = { + MDP_INTF_REV_13xx_INTR_CLEAR(8), + MDP_INTF_REV_13xx_INTR_EN(8), + MDP_INTF_REV_13xx_INTR_STATUS(8) + }, +}; + +#define DPU_IRQ_MASK(irq_idx) (BIT(DPU_IRQ_BIT(irq_idx))) + +static inline bool dpu_core_irq_is_valid(unsigned int irq_idx) +{ + return irq_idx && irq_idx <= DPU_NUM_IRQS; +} + +static inline struct dpu_hw_intr_entry *dpu_core_irq_get_entry(struct dpu_hw_intr *intr, + unsigned int irq_idx) +{ + return &intr->irq_tbl[irq_idx - 1]; +} + +/** + * dpu_core_irq_callback_handler - dispatch core interrupts + * @dpu_kms: Pointer to DPU's KMS structure + * @irq_idx: interrupt index + */ +static void dpu_core_irq_callback_handler(struct dpu_kms *dpu_kms, unsigned int irq_idx) +{ + struct dpu_hw_intr_entry *irq_entry = dpu_core_irq_get_entry(dpu_kms->hw_intr, irq_idx); + + VERB("IRQ=[%d, %d]\n", DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx)); + + if (!irq_entry->cb) { + DRM_ERROR("no registered cb, IRQ=[%d, %d]\n", + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx)); + return; + } + + atomic_inc(&irq_entry->count); + + /* + * Perform registered function callback + */ + irq_entry->cb(irq_entry->arg); +} + +/** + * dpu_core_irq - core IRQ handler + * @kms: MSM KMS handle + * @return: interrupt handling status + */ +irqreturn_t dpu_core_irq(struct msm_kms *kms) +{ + struct dpu_kms *dpu_kms = to_dpu_kms(kms); + struct dpu_hw_intr *intr = dpu_kms->hw_intr; + int reg_idx; + unsigned int irq_idx; + u32 irq_status; + u32 enable_mask; + int bit; + unsigned long irq_flags; + + if (!intr) + return IRQ_NONE; + + spin_lock_irqsave(&intr->irq_lock, irq_flags); + for (reg_idx = 0; reg_idx < MDP_INTR_MAX; reg_idx++) { + if (!test_bit(reg_idx, &intr->irq_mask)) + continue; + + /* Read interrupt status */ + irq_status = DPU_REG_READ(&intr->hw, intr->intr_set[reg_idx].status_off); + + /* Read enable mask */ + enable_mask = DPU_REG_READ(&intr->hw, intr->intr_set[reg_idx].en_off); + + /* and clear the interrupt */ + if (irq_status) + DPU_REG_WRITE(&intr->hw, intr->intr_set[reg_idx].clr_off, + irq_status); + + /* Finally update IRQ status based on enable mask */ + irq_status &= enable_mask; + + if (!irq_status) + continue; + + /* + * Search through matching intr status. + */ + while ((bit = ffs(irq_status)) != 0) { + irq_idx = DPU_IRQ_IDX(reg_idx, bit - 1); + + dpu_core_irq_callback_handler(dpu_kms, irq_idx); + + /* + * When callback finish, clear the irq_status + * with the matching mask. Once irq_status + * is all cleared, the search can be stopped. + */ + irq_status &= ~BIT(bit - 1); + } + } + + /* ensure register writes go through */ + wmb(); + + spin_unlock_irqrestore(&intr->irq_lock, irq_flags); + + return IRQ_HANDLED; +} + +static int dpu_hw_intr_enable_irq_locked(struct dpu_hw_intr *intr, + unsigned int irq_idx) +{ + int reg_idx; + const struct dpu_intr_reg *reg; + const char *dbgstr = NULL; + uint32_t cache_irq_mask; + + if (!intr) + return -EINVAL; + + if (!dpu_core_irq_is_valid(irq_idx)) { + pr_err("invalid IRQ=[%d, %d]\n", + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx)); + return -EINVAL; + } + + /* + * The cache_irq_mask and hardware RMW operations needs to be done + * under irq_lock and it's the caller's responsibility to ensure that's + * held. + */ + assert_spin_locked(&intr->irq_lock); + + reg_idx = DPU_IRQ_REG(irq_idx); + reg = &intr->intr_set[reg_idx]; + + /* Is this interrupt register supported on the platform */ + if (WARN_ON(!reg->en_off)) + return -EINVAL; + + cache_irq_mask = intr->cache_irq_mask[reg_idx]; + if (cache_irq_mask & DPU_IRQ_MASK(irq_idx)) { + dbgstr = "already "; + } else { + dbgstr = ""; + + cache_irq_mask |= DPU_IRQ_MASK(irq_idx); + /* Cleaning any pending interrupt */ + DPU_REG_WRITE(&intr->hw, reg->clr_off, DPU_IRQ_MASK(irq_idx)); + /* Enabling interrupts with the new mask */ + DPU_REG_WRITE(&intr->hw, reg->en_off, cache_irq_mask); + + /* ensure register write goes through */ + wmb(); + + intr->cache_irq_mask[reg_idx] = cache_irq_mask; + } + + pr_debug("DPU IRQ=[%d, %d] %senabled: MASK:0x%.8lx, CACHE-MASK:0x%.8x\n", + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx), dbgstr, + DPU_IRQ_MASK(irq_idx), cache_irq_mask); + + return 0; +} + +static int dpu_hw_intr_disable_irq_locked(struct dpu_hw_intr *intr, + unsigned int irq_idx) +{ + int reg_idx; + const struct dpu_intr_reg *reg; + const char *dbgstr = NULL; + uint32_t cache_irq_mask; + + if (!intr) + return -EINVAL; + + if (!dpu_core_irq_is_valid(irq_idx)) { + pr_err("invalid IRQ=[%d, %d]\n", + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx)); + return -EINVAL; + } + + /* + * The cache_irq_mask and hardware RMW operations needs to be done + * under irq_lock and it's the caller's responsibility to ensure that's + * held. + */ + assert_spin_locked(&intr->irq_lock); + + reg_idx = DPU_IRQ_REG(irq_idx); + reg = &intr->intr_set[reg_idx]; + + cache_irq_mask = intr->cache_irq_mask[reg_idx]; + if ((cache_irq_mask & DPU_IRQ_MASK(irq_idx)) == 0) { + dbgstr = "already "; + } else { + dbgstr = ""; + + cache_irq_mask &= ~DPU_IRQ_MASK(irq_idx); + /* Disable interrupts based on the new mask */ + DPU_REG_WRITE(&intr->hw, reg->en_off, cache_irq_mask); + /* Cleaning any pending interrupt */ + DPU_REG_WRITE(&intr->hw, reg->clr_off, DPU_IRQ_MASK(irq_idx)); + + /* ensure register write goes through */ + wmb(); + + intr->cache_irq_mask[reg_idx] = cache_irq_mask; + } + + pr_debug("DPU IRQ=[%d, %d] %sdisabled: MASK:0x%.8lx, CACHE-MASK:0x%.8x\n", + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx), dbgstr, + DPU_IRQ_MASK(irq_idx), cache_irq_mask); + + return 0; +} + +static void dpu_clear_irqs(struct dpu_kms *dpu_kms) +{ + struct dpu_hw_intr *intr = dpu_kms->hw_intr; + int i; + + if (!intr) + return; + + for (i = 0; i < MDP_INTR_MAX; i++) { + if (test_bit(i, &intr->irq_mask)) + DPU_REG_WRITE(&intr->hw, + intr->intr_set[i].clr_off, 0xffffffff); + } + + /* ensure register writes go through */ + wmb(); +} + +static void dpu_disable_all_irqs(struct dpu_kms *dpu_kms) +{ + struct dpu_hw_intr *intr = dpu_kms->hw_intr; + int i; + + if (!intr) + return; + + for (i = 0; i < MDP_INTR_MAX; i++) { + if (test_bit(i, &intr->irq_mask)) + DPU_REG_WRITE(&intr->hw, + intr->intr_set[i].en_off, 0x00000000); + } + + /* ensure register writes go through */ + wmb(); +} + +/** + * dpu_core_irq_read - IRQ helper function for reading IRQ status + * @dpu_kms: DPU handle + * @irq_idx: irq index + * @return: non-zero if irq detected; otherwise no irq detected + */ +u32 dpu_core_irq_read(struct dpu_kms *dpu_kms, + unsigned int irq_idx) +{ + struct dpu_hw_intr *intr = dpu_kms->hw_intr; + int reg_idx; + unsigned long irq_flags; + u32 intr_status; + + if (!intr) + return 0; + + if (!dpu_core_irq_is_valid(irq_idx)) { + pr_err("invalid IRQ=[%d, %d]\n", DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx)); + return 0; + } + + spin_lock_irqsave(&intr->irq_lock, irq_flags); + + reg_idx = DPU_IRQ_REG(irq_idx); + intr_status = DPU_REG_READ(&intr->hw, + intr->intr_set[reg_idx].status_off) & + DPU_IRQ_MASK(irq_idx); + if (intr_status) + DPU_REG_WRITE(&intr->hw, intr->intr_set[reg_idx].clr_off, + intr_status); + + /* ensure register writes go through */ + wmb(); + + spin_unlock_irqrestore(&intr->irq_lock, irq_flags); + + return intr_status; +} + +/** + * dpu_hw_intr_init(): Initializes the interrupts hw object + * @dev: Corresponding device for devres management + * @addr: mapped register io address of MDP + * @m: pointer to MDSS catalog data + */ +struct dpu_hw_intr *dpu_hw_intr_init(struct drm_device *dev, + void __iomem *addr, + const struct dpu_mdss_cfg *m) +{ + struct dpu_hw_intr *intr; + unsigned int i; + + if (!addr || !m) + return ERR_PTR(-EINVAL); + + intr = drmm_kzalloc(dev, sizeof(*intr), GFP_KERNEL); + if (!intr) + return ERR_PTR(-ENOMEM); + + if (m->mdss_ver->core_major_ver >= 13) + intr->intr_set = dpu_intr_set_13xx; + else if (m->mdss_ver->core_major_ver >= 7) + intr->intr_set = dpu_intr_set_7xxx; + else + intr->intr_set = dpu_intr_set_legacy; + + intr->hw.blk_addr = addr + m->mdp[0].base; + + intr->irq_mask = BIT(MDP_SSPP_TOP0_INTR) | + BIT(MDP_SSPP_TOP0_INTR2) | + BIT(MDP_SSPP_TOP0_HIST_INTR); + for (i = 0; i < m->intf_count; i++) { + const struct dpu_intf_cfg *intf = &m->intf[i]; + + if (intf->type == INTF_NONE) + continue; + + intr->irq_mask |= BIT(MDP_INTFn_INTR(intf->id)); + + if (intf->intr_tear_rd_ptr) + intr->irq_mask |= BIT(DPU_IRQ_REG(intf->intr_tear_rd_ptr)); + } + + spin_lock_init(&intr->irq_lock); + + return intr; +} + +/** + * dpu_core_irq_register_callback - For registering callback function on IRQ + * interrupt + * @dpu_kms: DPU handle + * @irq_idx: irq index + * @irq_cb: IRQ callback function. + * @irq_arg: IRQ callback argument. + * @return: 0 for success registering callback, otherwise failure + * + * This function supports registration of multiple callbacks for each interrupt. + */ +int dpu_core_irq_register_callback(struct dpu_kms *dpu_kms, + unsigned int irq_idx, + void (*irq_cb)(void *arg), + void *irq_arg) +{ + struct dpu_hw_intr_entry *irq_entry; + unsigned long irq_flags; + int ret; + + if (!irq_cb) { + DPU_ERROR("IRQ=[%d, %d] NULL callback\n", + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx)); + return -EINVAL; + } + + if (!dpu_core_irq_is_valid(irq_idx)) { + DPU_ERROR("invalid IRQ=[%d, %d] irq_cb:%ps\n", + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx), irq_cb); + return -EINVAL; + } + + VERB("[%pS] IRQ=[%d, %d]\n", __builtin_return_address(0), + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx)); + + spin_lock_irqsave(&dpu_kms->hw_intr->irq_lock, irq_flags); + + irq_entry = dpu_core_irq_get_entry(dpu_kms->hw_intr, irq_idx); + if (unlikely(WARN_ON(irq_entry->cb))) { + spin_unlock_irqrestore(&dpu_kms->hw_intr->irq_lock, irq_flags); + + return -EBUSY; + } + + trace_dpu_core_irq_register_callback(DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx), irq_cb); + irq_entry->arg = irq_arg; + irq_entry->cb = irq_cb; + + ret = dpu_hw_intr_enable_irq_locked( + dpu_kms->hw_intr, + irq_idx); + if (ret) + DPU_ERROR("Failed/ to enable IRQ=[%d, %d]\n", + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx)); + spin_unlock_irqrestore(&dpu_kms->hw_intr->irq_lock, irq_flags); + + trace_dpu_irq_register_success(DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx)); + + return 0; +} + +/** + * dpu_core_irq_unregister_callback - For unregistering callback function on IRQ + * interrupt + * @dpu_kms: DPU handle + * @irq_idx: irq index + * @return: 0 for success registering callback, otherwise failure + * + * This function supports registration of multiple callbacks for each interrupt. + */ +int dpu_core_irq_unregister_callback(struct dpu_kms *dpu_kms, + unsigned int irq_idx) +{ + struct dpu_hw_intr_entry *irq_entry; + unsigned long irq_flags; + int ret; + + if (!dpu_core_irq_is_valid(irq_idx)) { + DPU_ERROR("invalid IRQ=[%d, %d]\n", + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx)); + return -EINVAL; + } + + VERB("[%pS] IRQ=[%d, %d]\n", __builtin_return_address(0), + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx)); + + spin_lock_irqsave(&dpu_kms->hw_intr->irq_lock, irq_flags); + trace_dpu_core_irq_unregister_callback(DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx)); + + ret = dpu_hw_intr_disable_irq_locked(dpu_kms->hw_intr, irq_idx); + if (ret) + DPU_ERROR("Failed to disable IRQ=[%d, %d]: %d\n", + DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx), ret); + + irq_entry = dpu_core_irq_get_entry(dpu_kms->hw_intr, irq_idx); + irq_entry->cb = NULL; + irq_entry->arg = NULL; + + spin_unlock_irqrestore(&dpu_kms->hw_intr->irq_lock, irq_flags); + + trace_dpu_irq_unregister_success(DPU_IRQ_REG(irq_idx), DPU_IRQ_BIT(irq_idx)); + + return 0; +} + +#ifdef CONFIG_DEBUG_FS +static int dpu_debugfs_core_irq_show(struct seq_file *s, void *v) +{ + struct dpu_kms *dpu_kms = s->private; + struct dpu_hw_intr_entry *irq_entry; + unsigned long irq_flags; + int i, irq_count; + void *cb; + + for (i = 1; i <= DPU_NUM_IRQS; i++) { + spin_lock_irqsave(&dpu_kms->hw_intr->irq_lock, irq_flags); + irq_entry = dpu_core_irq_get_entry(dpu_kms->hw_intr, i); + irq_count = atomic_read(&irq_entry->count); + cb = irq_entry->cb; + spin_unlock_irqrestore(&dpu_kms->hw_intr->irq_lock, irq_flags); + + if (irq_count || cb) + seq_printf(s, "IRQ=[%d, %d] count:%d cb:%ps\n", + DPU_IRQ_REG(i), DPU_IRQ_BIT(i), irq_count, cb); + } + + return 0; +} + +DEFINE_SHOW_ATTRIBUTE(dpu_debugfs_core_irq); + +/** + * dpu_debugfs_core_irq_init - register core irq debugfs + * @dpu_kms: pointer to kms + * @parent: debugfs directory root + */ +void dpu_debugfs_core_irq_init(struct dpu_kms *dpu_kms, + struct dentry *parent) +{ + debugfs_create_file("core_irq", 0600, parent, dpu_kms, + &dpu_debugfs_core_irq_fops); +} +#endif + +/** + * dpu_core_irq_preinstall - perform pre-installation of core IRQ handler + * @kms: MSM KMS handle + * @return: none + */ +void dpu_core_irq_preinstall(struct msm_kms *kms) +{ + struct dpu_kms *dpu_kms = to_dpu_kms(kms); + struct dpu_hw_intr_entry *irq_entry; + int i; + + pm_runtime_get_sync(&dpu_kms->pdev->dev); + dpu_clear_irqs(dpu_kms); + dpu_disable_all_irqs(dpu_kms); + pm_runtime_put_sync(&dpu_kms->pdev->dev); + + for (i = 1; i <= DPU_NUM_IRQS; i++) { + irq_entry = dpu_core_irq_get_entry(dpu_kms->hw_intr, i); + atomic_set(&irq_entry->count, 0); + } +} + +/** + * dpu_core_irq_uninstall - uninstall core IRQ handler + * @kms: MSM KMS handle + * @return: none + */ +void dpu_core_irq_uninstall(struct msm_kms *kms) +{ + struct dpu_kms *dpu_kms = to_dpu_kms(kms); + struct dpu_hw_intr_entry *irq_entry; + int i; + + if (!dpu_kms->hw_intr) + return; + + pm_runtime_get_sync(&dpu_kms->pdev->dev); + for (i = 1; i <= DPU_NUM_IRQS; i++) { + irq_entry = dpu_core_irq_get_entry(dpu_kms->hw_intr, i); + if (irq_entry->cb) + DPU_ERROR("IRQ=[%d, %d] still enabled/registered\n", + DPU_IRQ_REG(i), DPU_IRQ_BIT(i)); + } + + dpu_clear_irqs(dpu_kms); + dpu_disable_all_irqs(dpu_kms); + pm_runtime_put_sync(&dpu_kms->pdev->dev); +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h new file mode 100644 index 000000000..142358a10 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h @@ -0,0 +1,75 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2016-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_HW_INTERRUPTS_H +#define _DPU_HW_INTERRUPTS_H + +#include + +#include "dpu_hwio.h" +#include "dpu_hw_catalog.h" +#include "dpu_hw_util.h" +#include "dpu_hw_mdss.h" + +/* When making changes be sure to sync with dpu_intr_set */ +enum dpu_hw_intr_reg { + MDP_SSPP_TOP0_INTR, + MDP_SSPP_TOP0_INTR2, + MDP_SSPP_TOP0_HIST_INTR, + /* All MDP_INTFn_INTR should come sequentially */ + MDP_INTF0_INTR, + MDP_INTF1_INTR, + MDP_INTF2_INTR, + MDP_INTF3_INTR, + MDP_INTF4_INTR, + MDP_INTF5_INTR, + MDP_INTF6_INTR, + MDP_INTF7_INTR, + MDP_INTF8_INTR, + MDP_INTF1_TEAR_INTR, + MDP_INTF2_TEAR_INTR, + MDP_AD4_0_INTR, + MDP_AD4_1_INTR, + MDP_INTR_MAX, +}; + +#define MDP_INTFn_INTR(intf) (MDP_INTF0_INTR + (intf - INTF_0)) + +#define DPU_IRQ_IDX(reg_idx, offset) (1 + reg_idx * 32 + offset) +#define DPU_IRQ_REG(irq_idx) ((irq_idx - 1) / 32) +#define DPU_IRQ_BIT(irq_idx) ((irq_idx - 1) % 32) + +#define DPU_NUM_IRQS (MDP_INTR_MAX * 32) + +struct dpu_hw_intr_entry { + void (*cb)(void *arg); + void *arg; + atomic_t count; +}; + +/** + * struct dpu_hw_intr: hw interrupts handling data structure + * @hw: virtual address mapping + * @ops: function pointer mapping for IRQ handling + * @cache_irq_mask: array of IRQ enable masks reg storage created during init + * @save_irq_status: array of IRQ status reg storage created during init + * @irq_lock: spinlock for accessing IRQ resources + * @irq_cb_tbl: array of IRQ callbacks + */ +struct dpu_hw_intr { + struct dpu_hw_blk_reg_map hw; + u32 cache_irq_mask[MDP_INTR_MAX]; + u32 *save_irq_status; + spinlock_t irq_lock; + unsigned long irq_mask; + const struct dpu_intr_reg *intr_set; + + struct dpu_hw_intr_entry irq_tbl[DPU_NUM_IRQS]; +}; + +struct dpu_hw_intr *dpu_hw_intr_init(struct drm_device *dev, + void __iomem *addr, + const struct dpu_mdss_cfg *m); + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c new file mode 100644 index 000000000..ac82b69ae --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c @@ -0,0 +1,672 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2022-2023 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#include "dpu_hwio.h" +#include "dpu_hw_catalog.h" +#include "dpu_hw_intf.h" +#include "dpu_kms.h" +#include "dpu_trace.h" + +#include + +#include + +#define INTF_TIMING_ENGINE_EN 0x000 +#define INTF_CONFIG 0x004 +#define INTF_HSYNC_CTL 0x008 +#define INTF_VSYNC_PERIOD_F0 0x00C +#define INTF_VSYNC_PERIOD_F1 0x010 +#define INTF_VSYNC_PULSE_WIDTH_F0 0x014 +#define INTF_VSYNC_PULSE_WIDTH_F1 0x018 +#define INTF_DISPLAY_V_START_F0 0x01C +#define INTF_DISPLAY_V_START_F1 0x020 +#define INTF_DISPLAY_V_END_F0 0x024 +#define INTF_DISPLAY_V_END_F1 0x028 +#define INTF_ACTIVE_V_START_F0 0x02C +#define INTF_ACTIVE_V_START_F1 0x030 +#define INTF_ACTIVE_V_END_F0 0x034 +#define INTF_ACTIVE_V_END_F1 0x038 +#define INTF_DISPLAY_HCTL 0x03C +#define INTF_ACTIVE_HCTL 0x040 +#define INTF_BORDER_COLOR 0x044 +#define INTF_UNDERFLOW_COLOR 0x048 +#define INTF_HSYNC_SKEW 0x04C +#define INTF_POLARITY_CTL 0x050 +#define INTF_TEST_CTL 0x054 +#define INTF_TP_COLOR0 0x058 +#define INTF_TP_COLOR1 0x05C +#define INTF_CONFIG2 0x060 +#define INTF_DISPLAY_DATA_HCTL 0x064 +#define INTF_ACTIVE_DATA_HCTL 0x068 + +#define INTF_DSI_CMD_MODE_TRIGGER_EN 0x084 +#define INTF_PANEL_FORMAT 0x090 + +#define INTF_FRAME_LINE_COUNT_EN 0x0A8 +#define INTF_FRAME_COUNT 0x0AC +#define INTF_LINE_COUNT 0x0B0 + +#define INTF_DEFLICKER_CONFIG 0x0F0 +#define INTF_DEFLICKER_STRNG_COEFF 0x0F4 +#define INTF_DEFLICKER_WEAK_COEFF 0x0F8 + +#define INTF_TPG_ENABLE 0x100 +#define INTF_TPG_MAIN_CONTROL 0x104 +#define INTF_TPG_VIDEO_CONFIG 0x108 +#define INTF_TPG_COMPONENT_LIMITS 0x10C +#define INTF_TPG_RECTANGLE 0x110 +#define INTF_TPG_INITIAL_VALUE 0x114 +#define INTF_TPG_BLK_WHITE_PATTERN_FRAMES 0x118 +#define INTF_TPG_RGB_MAPPING 0x11C +#define INTF_PROG_FETCH_START 0x170 +#define INTF_PROG_ROT_START 0x174 + +#define INTF_MISR_CTRL 0x180 +#define INTF_MISR_SIGNATURE 0x184 + +#define INTF_WD_TIMER_0_CTL 0x230 +#define INTF_WD_TIMER_0_CTL2 0x234 +#define INTF_WD_TIMER_0_LOAD_VALUE 0x238 + +#define INTF_MUX 0x25C +#define INTF_STATUS 0x26C +#define INTF_AVR_CONTROL 0x270 +#define INTF_AVR_MODE 0x274 +#define INTF_AVR_TRIGGER 0x278 +#define INTF_AVR_VTOTAL 0x27C +#define INTF_TEAR_MDP_VSYNC_SEL 0x280 +#define INTF_TEAR_TEAR_CHECK_EN 0x284 +#define INTF_TEAR_SYNC_CONFIG_VSYNC 0x288 +#define INTF_TEAR_SYNC_CONFIG_HEIGHT 0x28C +#define INTF_TEAR_SYNC_WRCOUNT 0x290 +#define INTF_TEAR_VSYNC_INIT_VAL 0x294 +#define INTF_TEAR_INT_COUNT_VAL 0x298 +#define INTF_TEAR_SYNC_THRESH 0x29C +#define INTF_TEAR_START_POS 0x2A0 +#define INTF_TEAR_RD_PTR_IRQ 0x2A4 +#define INTF_TEAR_WR_PTR_IRQ 0x2A8 +#define INTF_TEAR_OUT_LINE_COUNT 0x2AC +#define INTF_TEAR_LINE_COUNT 0x2B0 +#define INTF_TEAR_AUTOREFRESH_CONFIG 0x2B4 + +#define INTF_CFG_ACTIVE_H_EN BIT(29) +#define INTF_CFG_ACTIVE_V_EN BIT(30) + +#define INTF_CFG2_DATABUS_WIDEN BIT(0) +#define INTF_CFG2_DATA_HCTL_EN BIT(4) +#define INTF_CFG2_DCE_DATA_COMPRESS BIT(12) + + +static void dpu_hw_intf_setup_timing_engine(struct dpu_hw_intf *intf, + const struct dpu_hw_intf_timing_params *p, + const struct msm_format *fmt) +{ + struct dpu_hw_blk_reg_map *c = &intf->hw; + u32 hsync_period, vsync_period; + u32 display_v_start, display_v_end; + u32 hsync_start_x, hsync_end_x; + u32 hsync_data_start_x, hsync_data_end_x; + u32 active_h_start, active_h_end; + u32 active_v_start, active_v_end; + u32 active_hctl, display_hctl, hsync_ctl; + u32 polarity_ctl, den_polarity; + u32 panel_format; + u32 intf_cfg, intf_cfg2 = 0; + u32 display_data_hctl = 0, active_data_hctl = 0; + u32 data_width; + bool dp_intf = false; + + /* read interface_cfg */ + intf_cfg = DPU_REG_READ(c, INTF_CONFIG); + + if (intf->cap->type == INTF_DP) + dp_intf = true; + + hsync_period = p->hsync_pulse_width + p->h_back_porch + p->width + + p->h_front_porch; + vsync_period = p->vsync_pulse_width + p->v_back_porch + p->height + + p->v_front_porch; + + display_v_start = ((p->vsync_pulse_width + p->v_back_porch) * + hsync_period) + p->hsync_skew; + display_v_end = ((vsync_period - p->v_front_porch) * hsync_period) + + p->hsync_skew - 1; + + hsync_start_x = p->h_back_porch + p->hsync_pulse_width; + hsync_end_x = hsync_period - p->h_front_porch - 1; + + if (p->width != p->xres) { /* border fill added */ + active_h_start = hsync_start_x; + active_h_end = active_h_start + p->xres - 1; + } else { + active_h_start = 0; + active_h_end = 0; + } + + if (p->height != p->yres) { /* border fill added */ + active_v_start = display_v_start; + active_v_end = active_v_start + (p->yres * hsync_period) - 1; + } else { + active_v_start = 0; + active_v_end = 0; + } + + if (active_h_end) { + active_hctl = (active_h_end << 16) | active_h_start; + intf_cfg |= INTF_CFG_ACTIVE_H_EN; + } else { + active_hctl = 0; + } + + if (active_v_end) + intf_cfg |= INTF_CFG_ACTIVE_V_EN; + + hsync_ctl = (hsync_period << 16) | p->hsync_pulse_width; + display_hctl = (hsync_end_x << 16) | hsync_start_x; + + if (p->wide_bus_en) + intf_cfg2 |= INTF_CFG2_DATABUS_WIDEN; + + data_width = p->width; + + /* + * If widebus is disabled: + * For uncompressed stream, the data is valid for the entire active + * window period. + * For compressed stream, data is valid for a shorter time period + * inside the active window depending on the compression ratio. + * + * If widebus is enabled: + * For uncompressed stream, data is valid for only half the active + * window, since the data rate is doubled in this mode. + * For compressed stream, data validity window needs to be adjusted for + * compression ratio and then further halved. + * + * For the compression mode in DP case, the p->width is already + * adjusted in drm_mode_to_intf_timing_params(). + */ + if (p->compression_en && !dp_intf) { + if (p->wide_bus_en) + data_width = DIV_ROUND_UP(p->dce_bytes_per_line, 6); + else + data_width = DIV_ROUND_UP(p->dce_bytes_per_line, 3); + } else if (p->wide_bus_en && !dp_intf) { + data_width = p->width >> 1; + } + + /* TODO: handle DSC+DP case, we only handle DSC+DSI case so far */ + if (p->compression_en && !dp_intf && + intf->mdss_ver->core_major_ver >= 7) + intf_cfg2 |= INTF_CFG2_DCE_DATA_COMPRESS; + + hsync_data_start_x = hsync_start_x; + hsync_data_end_x = hsync_start_x + data_width - 1; + + display_data_hctl = (hsync_data_end_x << 16) | hsync_data_start_x; + + if (dp_intf) { + /* DP timing adjustment */ + display_v_start += p->hsync_pulse_width + p->h_back_porch; + display_v_end -= p->h_front_porch; + + active_h_start = hsync_start_x; + active_h_end = active_h_start + p->xres - 1; + active_v_start = display_v_start; + active_v_end = active_v_start + (p->yres * hsync_period) - 1; + + active_hctl = (active_h_end << 16) | active_h_start; + display_hctl = active_hctl; + + intf_cfg |= INTF_CFG_ACTIVE_H_EN | INTF_CFG_ACTIVE_V_EN; + } + + den_polarity = 0; + polarity_ctl = (den_polarity << 2) | /* DEN Polarity */ + (p->vsync_polarity << 1) | /* VSYNC Polarity */ + (p->hsync_polarity << 0); /* HSYNC Polarity */ + + if (!MSM_FORMAT_IS_YUV(fmt)) + panel_format = (fmt->bpc_g_y | + (fmt->bpc_b_cb << 2) | + (fmt->bpc_r_cr << 4) | + (0x21 << 8)); + else + /* Interface treats all the pixel data in RGB888 format */ + panel_format = (BPC8 | + (BPC8 << 2) | + (BPC8 << 4) | + (0x21 << 8)); + + DPU_REG_WRITE(c, INTF_HSYNC_CTL, hsync_ctl); + DPU_REG_WRITE(c, INTF_VSYNC_PERIOD_F0, vsync_period * hsync_period); + DPU_REG_WRITE(c, INTF_VSYNC_PULSE_WIDTH_F0, + p->vsync_pulse_width * hsync_period); + DPU_REG_WRITE(c, INTF_DISPLAY_HCTL, display_hctl); + DPU_REG_WRITE(c, INTF_DISPLAY_V_START_F0, display_v_start); + DPU_REG_WRITE(c, INTF_DISPLAY_V_END_F0, display_v_end); + DPU_REG_WRITE(c, INTF_ACTIVE_HCTL, active_hctl); + DPU_REG_WRITE(c, INTF_ACTIVE_V_START_F0, active_v_start); + DPU_REG_WRITE(c, INTF_ACTIVE_V_END_F0, active_v_end); + DPU_REG_WRITE(c, INTF_BORDER_COLOR, p->border_clr); + DPU_REG_WRITE(c, INTF_UNDERFLOW_COLOR, p->underflow_clr); + DPU_REG_WRITE(c, INTF_HSYNC_SKEW, p->hsync_skew); + DPU_REG_WRITE(c, INTF_POLARITY_CTL, polarity_ctl); + DPU_REG_WRITE(c, INTF_FRAME_LINE_COUNT_EN, 0x3); + DPU_REG_WRITE(c, INTF_CONFIG, intf_cfg); + DPU_REG_WRITE(c, INTF_PANEL_FORMAT, panel_format); + if (intf->mdss_ver->core_major_ver >= 5) { + /* + * DATA_HCTL_EN controls data timing which can be different from + * video timing. It is recommended to enable it for all cases, except + * if compression is enabled in 1 pixel per clock mode + */ + if (!(p->compression_en && !p->wide_bus_en)) + intf_cfg2 |= INTF_CFG2_DATA_HCTL_EN; + + DPU_REG_WRITE(c, INTF_CONFIG2, intf_cfg2); + DPU_REG_WRITE(c, INTF_DISPLAY_DATA_HCTL, display_data_hctl); + DPU_REG_WRITE(c, INTF_ACTIVE_DATA_HCTL, active_data_hctl); + } +} + +static void dpu_hw_intf_enable_timing_engine( + struct dpu_hw_intf *intf, + u8 enable) +{ + struct dpu_hw_blk_reg_map *c = &intf->hw; + /* Note: Display interface select is handled in top block hw layer */ + DPU_REG_WRITE(c, INTF_TIMING_ENGINE_EN, enable != 0); +} + +static void dpu_hw_intf_setup_prg_fetch( + struct dpu_hw_intf *intf, + const struct dpu_hw_intf_prog_fetch *fetch) +{ + struct dpu_hw_blk_reg_map *c = &intf->hw; + int fetch_enable; + + /* + * Fetch should always be outside the active lines. If the fetching + * is programmed within active region, hardware behavior is unknown. + */ + + fetch_enable = DPU_REG_READ(c, INTF_CONFIG); + if (fetch->enable) { + fetch_enable |= BIT(31); + DPU_REG_WRITE(c, INTF_PROG_FETCH_START, + fetch->fetch_start); + } else { + fetch_enable &= ~BIT(31); + } + + DPU_REG_WRITE(c, INTF_CONFIG, fetch_enable); +} + +static void dpu_hw_intf_bind_pingpong_blk( + struct dpu_hw_intf *intf, + const enum dpu_pingpong pp) +{ + struct dpu_hw_blk_reg_map *c = &intf->hw; + u32 mux_cfg; + + mux_cfg = DPU_REG_READ(c, INTF_MUX); + mux_cfg &= ~0xf; + + if (pp) + mux_cfg |= (pp - PINGPONG_0) & 0x7; + else + mux_cfg |= 0xf; + + DPU_REG_WRITE(c, INTF_MUX, mux_cfg); +} + +static void dpu_hw_intf_get_status( + struct dpu_hw_intf *intf, + struct dpu_hw_intf_status *s) +{ + struct dpu_hw_blk_reg_map *c = &intf->hw; + + if (intf->mdss_ver->core_major_ver >= 5) + s->is_en = DPU_REG_READ(c, INTF_STATUS) & BIT(0); + else + s->is_en = DPU_REG_READ(c, INTF_TIMING_ENGINE_EN); + + s->is_prog_fetch_en = !!(DPU_REG_READ(c, INTF_CONFIG) & BIT(31)); + if (s->is_en) { + s->frame_count = DPU_REG_READ(c, INTF_FRAME_COUNT); + s->line_count = DPU_REG_READ(c, INTF_LINE_COUNT); + } else { + s->line_count = 0; + s->frame_count = 0; + } +} + +static u32 dpu_hw_intf_get_line_count(struct dpu_hw_intf *intf) +{ + struct dpu_hw_blk_reg_map *c; + + if (!intf) + return 0; + + c = &intf->hw; + + return DPU_REG_READ(c, INTF_LINE_COUNT); +} + +static void dpu_hw_intf_setup_misr(struct dpu_hw_intf *intf) +{ + dpu_hw_setup_misr(&intf->hw, INTF_MISR_CTRL, 0x1); +} + +static int dpu_hw_intf_collect_misr(struct dpu_hw_intf *intf, u32 *misr_value) +{ + return dpu_hw_collect_misr(&intf->hw, INTF_MISR_CTRL, INTF_MISR_SIGNATURE, misr_value); +} + +static int dpu_hw_intf_enable_te(struct dpu_hw_intf *intf, + struct dpu_hw_tear_check *te) +{ + struct dpu_hw_blk_reg_map *c; + int cfg; + + if (!intf) + return -EINVAL; + + c = &intf->hw; + + cfg = BIT(19); /* VSYNC_COUNTER_EN */ + if (te->hw_vsync_mode) + cfg |= BIT(20); + + cfg |= te->vsync_count; + + DPU_REG_WRITE(c, INTF_TEAR_SYNC_CONFIG_VSYNC, cfg); + DPU_REG_WRITE(c, INTF_TEAR_SYNC_CONFIG_HEIGHT, te->sync_cfg_height); + DPU_REG_WRITE(c, INTF_TEAR_VSYNC_INIT_VAL, te->vsync_init_val); + DPU_REG_WRITE(c, INTF_TEAR_RD_PTR_IRQ, te->rd_ptr_irq); + DPU_REG_WRITE(c, INTF_TEAR_START_POS, te->start_pos); + DPU_REG_WRITE(c, INTF_TEAR_SYNC_THRESH, + ((te->sync_threshold_continue << 16) | + te->sync_threshold_start)); + DPU_REG_WRITE(c, INTF_TEAR_SYNC_WRCOUNT, + (te->start_pos + te->sync_threshold_start + 1)); + + DPU_REG_WRITE(c, INTF_TEAR_TEAR_CHECK_EN, 1); + + return 0; +} + +static void dpu_hw_intf_setup_autorefresh_config(struct dpu_hw_intf *intf, + u32 frame_count, bool enable) +{ + struct dpu_hw_blk_reg_map *c; + u32 refresh_cfg; + + c = &intf->hw; + refresh_cfg = DPU_REG_READ(c, INTF_TEAR_AUTOREFRESH_CONFIG); + if (enable) + refresh_cfg = BIT(31) | frame_count; + else + refresh_cfg &= ~BIT(31); + + DPU_REG_WRITE(c, INTF_TEAR_AUTOREFRESH_CONFIG, refresh_cfg); +} + +/* + * dpu_hw_intf_get_autorefresh_config - Get autorefresh config from HW + * @intf: DPU intf structure + * @frame_count: Used to return the current frame count from hw + * + * Returns: True if autorefresh enabled, false if disabled. + */ +static bool dpu_hw_intf_get_autorefresh_config(struct dpu_hw_intf *intf, + u32 *frame_count) +{ + u32 val = DPU_REG_READ(&intf->hw, INTF_TEAR_AUTOREFRESH_CONFIG); + + if (frame_count != NULL) + *frame_count = val & 0xffff; + return !!((val & BIT(31)) >> 31); +} + +static int dpu_hw_intf_disable_te(struct dpu_hw_intf *intf) +{ + struct dpu_hw_blk_reg_map *c; + + if (!intf) + return -EINVAL; + + c = &intf->hw; + DPU_REG_WRITE(c, INTF_TEAR_TEAR_CHECK_EN, 0); + return 0; +} + +static int dpu_hw_intf_connect_external_te(struct dpu_hw_intf *intf, + bool enable_external_te) +{ + struct dpu_hw_blk_reg_map *c = &intf->hw; + u32 cfg; + int orig; + + if (!intf) + return -EINVAL; + + c = &intf->hw; + cfg = DPU_REG_READ(c, INTF_TEAR_SYNC_CONFIG_VSYNC); + orig = (bool)(cfg & BIT(20)); + if (enable_external_te) + cfg |= BIT(20); + else + cfg &= ~BIT(20); + DPU_REG_WRITE(c, INTF_TEAR_SYNC_CONFIG_VSYNC, cfg); + trace_dpu_intf_connect_ext_te(intf->idx - INTF_0, cfg); + + return orig; +} + +static int dpu_hw_intf_get_vsync_info(struct dpu_hw_intf *intf, + struct dpu_hw_pp_vsync_info *info) +{ + struct dpu_hw_blk_reg_map *c = &intf->hw; + u32 val; + + if (!intf || !info) + return -EINVAL; + + c = &intf->hw; + + val = DPU_REG_READ(c, INTF_TEAR_VSYNC_INIT_VAL); + info->rd_ptr_init_val = val & 0xffff; + + val = DPU_REG_READ(c, INTF_TEAR_INT_COUNT_VAL); + info->rd_ptr_frame_count = (val & 0xffff0000) >> 16; + info->rd_ptr_line_count = val & 0xffff; + + val = DPU_REG_READ(c, INTF_TEAR_LINE_COUNT); + info->wr_ptr_line_count = val & 0xffff; + + val = DPU_REG_READ(c, INTF_FRAME_COUNT); + info->intf_frame_count = val; + + return 0; +} + +static void dpu_hw_intf_vsync_sel(struct dpu_hw_intf *intf, + struct dpu_vsync_source_cfg *cfg) +{ + struct dpu_hw_blk_reg_map *c; + + if (!intf) + return; + + c = &intf->hw; + + DPU_REG_WRITE(c, INTF_TEAR_MDP_VSYNC_SEL, (cfg->vsync_source & 0xf)); +} + +static void dpu_hw_intf_vsync_sel_v8(struct dpu_hw_intf *intf, + struct dpu_vsync_source_cfg *cfg) +{ + struct dpu_hw_blk_reg_map *c; + + if (!intf) + return; + + c = &intf->hw; + + if (cfg->vsync_source >= DPU_VSYNC_SOURCE_WD_TIMER_4 && + cfg->vsync_source <= DPU_VSYNC_SOURCE_WD_TIMER_1) { + pr_warn_once("DPU 8.x supports only GPIOs and timer0 as TE sources\n"); + return; + } + + if (cfg->vsync_source == DPU_VSYNC_SOURCE_WD_TIMER_0) { + u32 reg; + + DPU_REG_WRITE(c, INTF_WD_TIMER_0_LOAD_VALUE, + CALCULATE_WD_LOAD_VALUE(cfg->frame_rate)); + + DPU_REG_WRITE(c, INTF_WD_TIMER_0_CTL, BIT(0)); /* clear timer */ + + reg = BIT(8); /* enable heartbeat timer */ + reg |= BIT(0); /* enable WD timer */ + reg |= BIT(1); /* select default 16 clock ticks */ + DPU_REG_WRITE(c, INTF_WD_TIMER_0_CTL2, reg); + + /* make sure that timers are enabled/disabled for vsync state */ + wmb(); + } + + dpu_hw_intf_vsync_sel(intf, cfg); +} + +static void dpu_hw_intf_disable_autorefresh(struct dpu_hw_intf *intf, + uint32_t encoder_id, u16 vdisplay) +{ + struct dpu_hw_pp_vsync_info info; + int trial = 0; + + /* If autorefresh is already disabled, we have nothing to do */ + if (!dpu_hw_intf_get_autorefresh_config(intf, NULL)) + return; + + /* + * If autorefresh is enabled, disable it and make sure it is safe to + * proceed with current frame commit/push. Sequence followed is, + * 1. Disable TE + * 2. Disable autorefresh config + * 4. Poll for frame transfer ongoing to be false + * 5. Enable TE back + */ + + dpu_hw_intf_connect_external_te(intf, false); + dpu_hw_intf_setup_autorefresh_config(intf, 0, false); + + do { + udelay(DPU_ENC_MAX_POLL_TIMEOUT_US); + if ((trial * DPU_ENC_MAX_POLL_TIMEOUT_US) + > (KICKOFF_TIMEOUT_MS * USEC_PER_MSEC)) { + DPU_ERROR("enc%d intf%d disable autorefresh failed\n", + encoder_id, intf->idx - INTF_0); + break; + } + + trial++; + + dpu_hw_intf_get_vsync_info(intf, &info); + } while (info.wr_ptr_line_count > 0 && + info.wr_ptr_line_count < vdisplay); + + dpu_hw_intf_connect_external_te(intf, true); + + DPU_DEBUG("enc%d intf%d disabled autorefresh\n", + encoder_id, intf->idx - INTF_0); + +} + +static void dpu_hw_intf_program_intf_cmd_cfg(struct dpu_hw_intf *intf, + struct dpu_hw_intf_cmd_mode_cfg *cmd_mode_cfg) +{ + u32 intf_cfg2 = DPU_REG_READ(&intf->hw, INTF_CONFIG2); + + if (cmd_mode_cfg->data_compress) + intf_cfg2 |= INTF_CFG2_DCE_DATA_COMPRESS; + + if (cmd_mode_cfg->wide_bus_en) + intf_cfg2 |= INTF_CFG2_DATABUS_WIDEN; + + DPU_REG_WRITE(&intf->hw, INTF_CONFIG2, intf_cfg2); +} + +/** + * dpu_hw_intf_init() - Initializes the INTF driver for the passed + * interface catalog entry. + * @dev: Corresponding device for devres management + * @cfg: interface catalog entry for which driver object is required + * @addr: mapped register io address of MDP + * @mdss_rev: dpu core's major and minor versions + */ +struct dpu_hw_intf *dpu_hw_intf_init(struct drm_device *dev, + const struct dpu_intf_cfg *cfg, + void __iomem *addr, + const struct dpu_mdss_version *mdss_rev) +{ + struct dpu_hw_intf *c; + + if (cfg->type == INTF_NONE) { + DPU_DEBUG("Skip intf %d with type NONE\n", cfg->id - INTF_0); + return NULL; + } + + c = drmm_kzalloc(dev, sizeof(*c), GFP_KERNEL); + if (!c) + return ERR_PTR(-ENOMEM); + + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_INTF; + + /* + * Assign ops + */ + c->idx = cfg->id; + c->cap = cfg; + + c->mdss_ver = mdss_rev; + + c->ops.setup_timing_gen = dpu_hw_intf_setup_timing_engine; + c->ops.setup_prg_fetch = dpu_hw_intf_setup_prg_fetch; + c->ops.get_status = dpu_hw_intf_get_status; + c->ops.enable_timing = dpu_hw_intf_enable_timing_engine; + c->ops.get_line_count = dpu_hw_intf_get_line_count; + c->ops.setup_misr = dpu_hw_intf_setup_misr; + c->ops.collect_misr = dpu_hw_intf_collect_misr; + + if (mdss_rev->core_major_ver >= 5) + c->ops.bind_pingpong_blk = dpu_hw_intf_bind_pingpong_blk; + + /* INTF TE is only for DSI interfaces */ + if (mdss_rev->core_major_ver >= 5 && cfg->type == INTF_DSI) { + WARN_ON(!cfg->intr_tear_rd_ptr); + + c->ops.enable_tearcheck = dpu_hw_intf_enable_te; + c->ops.disable_tearcheck = dpu_hw_intf_disable_te; + c->ops.connect_external_te = dpu_hw_intf_connect_external_te; + if (mdss_rev->core_major_ver >= 8) + c->ops.vsync_sel = dpu_hw_intf_vsync_sel_v8; + else + c->ops.vsync_sel = dpu_hw_intf_vsync_sel; + c->ops.disable_autorefresh = dpu_hw_intf_disable_autorefresh; + } + + /* Technically, INTF_CONFIG2 is present for DPU 5.0+, but + * we can configure it for DPU 7.0+ since the wide bus and DSC flags + * would not be set for DPU < 7.0 anyways + */ + if (mdss_rev->core_major_ver >= 7) + c->ops.program_intf_cmd_cfg = dpu_hw_intf_program_intf_cmd_cfg; + + return c; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h new file mode 100644 index 000000000..badd26305 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h @@ -0,0 +1,135 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022-2023 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_HW_INTF_H +#define _DPU_HW_INTF_H + +#include "dpu_hw_catalog.h" +#include "dpu_hw_mdss.h" +#include "dpu_hw_util.h" + +struct dpu_hw_intf; +struct dpu_vsync_source_cfg; + +/* intf timing settings */ +struct dpu_hw_intf_timing_params { + u32 width; /* active width */ + u32 height; /* active height */ + u32 xres; /* Display panel width */ + u32 yres; /* Display panel height */ + + u32 h_back_porch; + u32 h_front_porch; + u32 v_back_porch; + u32 v_front_porch; + u32 hsync_pulse_width; + u32 vsync_pulse_width; + u32 hsync_polarity; + u32 vsync_polarity; + u32 border_clr; + u32 underflow_clr; + u32 hsync_skew; + + bool wide_bus_en; + bool compression_en; + u32 dce_bytes_per_line; +}; + +struct dpu_hw_intf_prog_fetch { + u8 enable; + /* vsync counter for the front porch pixel line */ + u32 fetch_start; +}; + +struct dpu_hw_intf_status { + u8 is_en; /* interface timing engine is enabled or not */ + u8 is_prog_fetch_en; /* interface prog fetch counter is enabled or not */ + u32 frame_count; /* frame count since timing engine enabled */ + u32 line_count; /* current line count including blanking */ +}; + +struct dpu_hw_intf_cmd_mode_cfg { + u8 data_compress; /* enable data compress between dpu and dsi */ + u8 wide_bus_en; /* enable databus widen mode */ +}; + +/** + * struct dpu_hw_intf_ops : Interface to the interface Hw driver functions + * Assumption is these functions will be called after clocks are enabled + * @setup_timing_gen : programs the timing engine + * @setup_prg_fetch : enables/disables the programmable fetch logic + * @enable_timing: enable/disable timing engine + * @get_status: returns if timing engine is enabled or not + * @get_line_count: reads current vertical line counter + * @bind_pingpong_blk: enable/disable the connection with pingpong which will + * feed pixels to this interface + * @setup_misr: enable/disable MISR + * @collect_misr: read MISR signature + * @enable_tearcheck: Enables vsync generation and sets up init value of read + * pointer and programs the tear check configuration + * @disable_tearcheck: Disables tearcheck block + * @connect_external_te: Read, modify, write to either set or clear listening to external TE + * Returns 1 if TE was originally connected, 0 if not, or -ERROR + * @vsync_sel: Select vsync signal for tear-effect configuration + * @disable_autorefresh: Disable autorefresh if enabled + * @program_intf_cmd_cfg: Program the DPU to interface datapath for command mode + */ +struct dpu_hw_intf_ops { + void (*setup_timing_gen)(struct dpu_hw_intf *intf, + const struct dpu_hw_intf_timing_params *p, + const struct msm_format *fmt); + + void (*setup_prg_fetch)(struct dpu_hw_intf *intf, + const struct dpu_hw_intf_prog_fetch *fetch); + + void (*enable_timing)(struct dpu_hw_intf *intf, + u8 enable); + + void (*get_status)(struct dpu_hw_intf *intf, + struct dpu_hw_intf_status *status); + + u32 (*get_line_count)(struct dpu_hw_intf *intf); + + void (*bind_pingpong_blk)(struct dpu_hw_intf *intf, + const enum dpu_pingpong pp); + void (*setup_misr)(struct dpu_hw_intf *intf); + int (*collect_misr)(struct dpu_hw_intf *intf, u32 *misr_value); + + // Tearcheck on INTF since DPU 5.0.0 + + int (*enable_tearcheck)(struct dpu_hw_intf *intf, struct dpu_hw_tear_check *cfg); + + int (*disable_tearcheck)(struct dpu_hw_intf *intf); + + int (*connect_external_te)(struct dpu_hw_intf *intf, bool enable_external_te); + + void (*vsync_sel)(struct dpu_hw_intf *intf, struct dpu_vsync_source_cfg *cfg); + + void (*disable_autorefresh)(struct dpu_hw_intf *intf, uint32_t encoder_id, u16 vdisplay); + + void (*program_intf_cmd_cfg)(struct dpu_hw_intf *intf, + struct dpu_hw_intf_cmd_mode_cfg *cmd_mode_cfg); +}; + +struct dpu_hw_intf { + struct dpu_hw_blk_reg_map hw; + + /* intf */ + enum dpu_intf idx; + const struct dpu_intf_cfg *cap; + + const struct dpu_mdss_version *mdss_ver; + + /* ops */ + struct dpu_hw_intf_ops ops; +}; + +struct dpu_hw_intf *dpu_hw_intf_init(struct drm_device *dev, + const struct dpu_intf_cfg *cfg, + void __iomem *addr, + const struct dpu_mdss_version *mdss_rev); + +#endif /*_DPU_HW_INTF_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c new file mode 100644 index 000000000..b7779726b --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c @@ -0,0 +1,397 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2022-2023 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2021, The Linux Foundation. All rights reserved. + */ + +#include + +#include "dpu_kms.h" +#include "dpu_hw_catalog.h" +#include "dpu_hwio.h" +#include "dpu_hw_lm.h" +#include "dpu_hw_mdss.h" + +#define LM_OP_MODE 0x00 +#define LM_OUT_SIZE 0x04 +#define LM_BORDER_COLOR_0 0x08 +#define LM_BORDER_COLOR_1 0x010 + +/* These register are offset to mixer base + stage base */ +#define LM_BLEND0_OP 0x00 + +/* = v12 DPU */ +#define LM_BG_SRC_SEL_V12 0x14 +#define LM_BG_SRC_SEL_V12_RESET_VALUE 0x0000c0c0 +#define LM_BORDER_COLOR_0_V12 0x1c +#define LM_BORDER_COLOR_1_V12 0x20 + +/* >= v12 DPU with offset to mixer base + stage base */ +#define LM_BLEND0_FG_SRC_SEL_V12 0x04 +#define LM_BLEND0_CONST_ALPHA_V12 0x08 +#define LM_FG_COLOR_FILL_COLOR_0_V12 0x0c +#define LM_FG_COLOR_FILL_COLOR_1_V12 0x10 +#define LM_FG_COLOR_FILL_SIZE_V12 0x14 +#define LM_FG_COLOR_FILL_XY_V12 0x18 + +#define LM_BLEND0_FG_ALPHA 0x04 +#define LM_BLEND0_BG_ALPHA 0x08 + +#define LM_MISR_CTRL 0x310 +#define LM_MISR_SIGNATURE 0x314 + + +/** + * _stage_offset(): returns the relative offset of the blend registers + * for the stage to be setup + * @ctx: mixer ctx contains the mixer to be programmed + * @stage: stage index to setup + */ +static inline int _stage_offset(struct dpu_hw_mixer *ctx, enum dpu_stage stage) +{ + const struct dpu_lm_sub_blks *sblk = ctx->cap->sblk; + if (stage != DPU_STAGE_BASE && stage <= sblk->maxblendstages) + return sblk->blendstage_base[stage - DPU_STAGE_0]; + + return -EINVAL; +} + +static void dpu_hw_lm_setup_out(struct dpu_hw_mixer *ctx, + struct dpu_hw_mixer_cfg *mixer) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + u32 outsize; + u32 op_mode; + + op_mode = DPU_REG_READ(c, LM_OP_MODE); + + outsize = mixer->out_height << 16 | mixer->out_width; + DPU_REG_WRITE(c, LM_OUT_SIZE, outsize); + + /* SPLIT_LEFT_RIGHT */ + if (mixer->right_mixer) + op_mode |= BIT(31); + else + op_mode &= ~BIT(31); + DPU_REG_WRITE(c, LM_OP_MODE, op_mode); +} + +static void dpu_hw_lm_setup_border_color(struct dpu_hw_mixer *ctx, + struct dpu_mdss_color *color, + u8 border_en) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + + if (border_en) { + DPU_REG_WRITE(c, LM_BORDER_COLOR_0, + (color->color_0 & 0xFFF) | + ((color->color_1 & 0xFFF) << 0x10)); + DPU_REG_WRITE(c, LM_BORDER_COLOR_1, + (color->color_2 & 0xFFF) | + ((color->color_3 & 0xFFF) << 0x10)); + } +} + +static void dpu_hw_lm_setup_border_color_v12(struct dpu_hw_mixer *ctx, + struct dpu_mdss_color *color, + u8 border_en) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + + if (border_en) { + DPU_REG_WRITE(c, LM_BORDER_COLOR_0_V12, + (color->color_0 & 0x3ff) | + ((color->color_1 & 0x3ff) << 16)); + DPU_REG_WRITE(c, LM_BORDER_COLOR_1_V12, + (color->color_2 & 0x3ff) | + ((color->color_3 & 0x3ff) << 16)); + } +} + +static void dpu_hw_lm_setup_misr(struct dpu_hw_mixer *ctx) +{ + dpu_hw_setup_misr(&ctx->hw, LM_MISR_CTRL, 0x0); +} + +static int dpu_hw_lm_collect_misr(struct dpu_hw_mixer *ctx, u32 *misr_value) +{ + return dpu_hw_collect_misr(&ctx->hw, LM_MISR_CTRL, LM_MISR_SIGNATURE, misr_value); +} + +static void dpu_hw_lm_setup_blend_config_combined_alpha(struct dpu_hw_mixer *ctx, + u32 stage, + u16 fg_alpha, u16 bg_alpha, + u32 blend_op) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + int stage_off; + u32 const_alpha; + + if (stage == DPU_STAGE_BASE) + return; + + stage_off = _stage_offset(ctx, stage); + if (WARN_ON(stage_off < 0)) + return; + + const_alpha = (bg_alpha >> 8) | ((fg_alpha >> 8) << 16); + DPU_REG_WRITE(c, LM_BLEND0_CONST_ALPHA + stage_off, const_alpha); + DPU_REG_WRITE(c, LM_BLEND0_OP + stage_off, blend_op); +} + +static void +dpu_hw_lm_setup_blend_config_combined_alpha_v12(struct dpu_hw_mixer *ctx, + u32 stage, + u16 fg_alpha, u16 bg_alpha, + u32 blend_op) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + int stage_off; + u32 const_alpha; + + if (stage == DPU_STAGE_BASE) + return; + + stage_off = _stage_offset(ctx, stage); + if (WARN_ON(stage_off < 0)) + return; + + const_alpha = (bg_alpha >> 6) | ((fg_alpha >> 6) << 16); + DPU_REG_WRITE(c, LM_BLEND0_CONST_ALPHA_V12 + stage_off, const_alpha); + DPU_REG_WRITE(c, LM_BLEND0_OP + stage_off, blend_op); +} + +static void dpu_hw_lm_setup_blend_config(struct dpu_hw_mixer *ctx, + u32 stage, + u16 fg_alpha, u16 bg_alpha, + u32 blend_op) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + int stage_off; + + if (stage == DPU_STAGE_BASE) + return; + + stage_off = _stage_offset(ctx, stage); + if (WARN_ON(stage_off < 0)) + return; + + DPU_REG_WRITE(c, LM_BLEND0_FG_ALPHA + stage_off, fg_alpha >> 8); + DPU_REG_WRITE(c, LM_BLEND0_BG_ALPHA + stage_off, bg_alpha >> 8); + DPU_REG_WRITE(c, LM_BLEND0_OP + stage_off, blend_op); +} + +static void dpu_hw_lm_setup_color3(struct dpu_hw_mixer *ctx, + uint32_t mixer_op_mode) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + int op_mode; + + /* read the existing op_mode configuration */ + op_mode = DPU_REG_READ(c, LM_OP_MODE); + + op_mode = (op_mode & (BIT(31) | BIT(30))) | mixer_op_mode; + + DPU_REG_WRITE(c, LM_OP_MODE, op_mode); +} + +static void dpu_hw_lm_setup_color3_v12(struct dpu_hw_mixer *ctx, + uint32_t mixer_op_mode) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + int op_mode, stages, stage_off, i; + + stages = ctx->cap->sblk->maxblendstages; + if (stages <= 0) + return; + + for (i = DPU_STAGE_0; i <= stages; i++) { + stage_off = _stage_offset(ctx, i); + if (WARN_ON(stage_off < 0)) + return; + + /* set color_out3 bit in blend0_op when enabled in mixer_op_mode */ + op_mode = DPU_REG_READ(c, LM_BLEND0_OP + stage_off); + if (mixer_op_mode & BIT(i)) + op_mode |= BIT(30); + else + op_mode &= ~BIT(30); + + DPU_REG_WRITE(c, LM_BLEND0_OP + stage_off, op_mode); + } +} + +static int _set_staged_sspp(u32 stage, struct dpu_hw_stage_cfg *stage_cfg, + int pipes_per_stage, u32 *value) +{ + int i; + u32 pipe_type = 0, pipe_id = 0, rec_id = 0; + u32 src_sel[PIPES_PER_STAGE]; + + *value = LM_BG_SRC_SEL_V12_RESET_VALUE; + if (!stage_cfg || !pipes_per_stage) + return 0; + + for (i = 0; i < pipes_per_stage; i++) { + enum dpu_sspp pipe = stage_cfg->stage[stage][i]; + enum dpu_sspp_multirect_index rect_index = stage_cfg->multirect_index[stage][i]; + + src_sel[i] = LM_BG_SRC_SEL_V12_RESET_VALUE; + + if (!pipe) + continue; + + /* translate pipe data to SWI pipe_type, pipe_id */ + if (pipe >= SSPP_DMA0 && pipe <= SSPP_DMA5) { + pipe_type = 0; + pipe_id = pipe - SSPP_DMA0; + } else if (pipe >= SSPP_VIG0 && pipe <= SSPP_VIG3) { + pipe_type = 1; + pipe_id = pipe - SSPP_VIG0; + } else { + DPU_ERROR("invalid rec-%d pipe:%d\n", i, pipe); + return -EINVAL; + } + + /* translate rec data to SWI rec_id */ + if (rect_index == DPU_SSPP_RECT_SOLO || rect_index == DPU_SSPP_RECT_0) { + rec_id = 0; + } else if (rect_index == DPU_SSPP_RECT_1) { + rec_id = 1; + } else { + DPU_ERROR("invalid rec-%d rect_index:%d\n", i, rect_index); + rec_id = 0; + } + + /* calculate SWI value for rec-0 and rec-1 and store it temporary buffer */ + src_sel[i] = (((pipe_type & 0x3) << 6) | ((rec_id & 0x3) << 4) | (pipe_id & 0xf)); + } + + /* calculate final SWI register value for rec-0 and rec-1 */ + *value = 0; + for (i = 0; i < pipes_per_stage; i++) + *value |= src_sel[i] << (i * 8); + + return 0; +} + +static int dpu_hw_lm_setup_blendstage(struct dpu_hw_mixer *ctx, enum dpu_lm lm, + struct dpu_hw_stage_cfg *stage_cfg) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + int i, ret, stages, stage_off, pipes_per_stage; + u32 value; + + stages = ctx->cap->sblk->maxblendstages; + if (stages <= 0) + return -EINVAL; + + if (test_bit(DPU_MIXER_SOURCESPLIT, &ctx->cap->features)) + pipes_per_stage = PIPES_PER_STAGE; + else + pipes_per_stage = 1; + + /* + * When stage configuration is empty, we can enable the + * border color by setting the corresponding LAYER_ACTIVE bit + * and un-staging all the pipes from the layer mixer. + */ + if (!stage_cfg) + DPU_REG_WRITE(c, LM_BG_SRC_SEL_V12, LM_BG_SRC_SEL_V12_RESET_VALUE); + + for (i = DPU_STAGE_0; i <= stages; i++) { + stage_off = _stage_offset(ctx, i); + if (stage_off < 0) + return stage_off; + + ret = _set_staged_sspp(i, stage_cfg, pipes_per_stage, &value); + if (ret) + return ret; + + DPU_REG_WRITE(c, LM_BLEND0_FG_SRC_SEL_V12 + stage_off, value); + } + + return 0; +} + +static int dpu_hw_lm_clear_all_blendstages(struct dpu_hw_mixer *ctx) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + int i, stages, stage_off; + + stages = ctx->cap->sblk->maxblendstages; + if (stages <= 0) + return -EINVAL; + + DPU_REG_WRITE(c, LM_BG_SRC_SEL_V12, LM_BG_SRC_SEL_V12_RESET_VALUE); + + for (i = DPU_STAGE_0; i <= stages; i++) { + stage_off = _stage_offset(ctx, i); + if (stage_off < 0) + return stage_off; + + DPU_REG_WRITE(c, LM_BLEND0_FG_SRC_SEL_V12 + stage_off, + LM_BG_SRC_SEL_V12_RESET_VALUE); + } + + return 0; +} + +/** + * dpu_hw_lm_init() - Initializes the mixer hw driver object. + * should be called once before accessing every mixer. + * @dev: Corresponding device for devres management + * @cfg: mixer catalog entry for which driver object is required + * @addr: mapped register io address of MDP + * @mdss_ver: DPU core's major and minor versions + */ +struct dpu_hw_mixer *dpu_hw_lm_init(struct drm_device *dev, + const struct dpu_lm_cfg *cfg, + void __iomem *addr, + const struct dpu_mdss_version *mdss_ver) +{ + struct dpu_hw_mixer *c; + + if (cfg->pingpong == PINGPONG_NONE) { + DPU_DEBUG("skip mixer %d without pingpong\n", cfg->id); + return NULL; + } + + c = drmm_kzalloc(dev, sizeof(*c), GFP_KERNEL); + if (!c) + return ERR_PTR(-ENOMEM); + + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_LM; + + /* Assign ops */ + c->idx = cfg->id; + c->cap = cfg; + c->ops.setup_mixer_out = dpu_hw_lm_setup_out; + if (mdss_ver->core_major_ver >= 12) + c->ops.setup_blend_config = dpu_hw_lm_setup_blend_config_combined_alpha_v12; + else if (mdss_ver->core_major_ver >= 4) + c->ops.setup_blend_config = dpu_hw_lm_setup_blend_config_combined_alpha; + else + c->ops.setup_blend_config = dpu_hw_lm_setup_blend_config; + if (mdss_ver->core_major_ver < 12) { + c->ops.setup_alpha_out = dpu_hw_lm_setup_color3; + c->ops.setup_border_color = dpu_hw_lm_setup_border_color; + } else { + c->ops.setup_alpha_out = dpu_hw_lm_setup_color3_v12; + c->ops.setup_blendstage = dpu_hw_lm_setup_blendstage; + c->ops.clear_all_blendstages = dpu_hw_lm_clear_all_blendstages; + c->ops.setup_border_color = dpu_hw_lm_setup_border_color_v12; + } + c->ops.setup_misr = dpu_hw_lm_setup_misr; + c->ops.collect_misr = dpu_hw_lm_collect_misr; + + return c; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h new file mode 100644 index 000000000..380ca673f --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h @@ -0,0 +1,118 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022-2023 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2021, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_HW_LM_H +#define _DPU_HW_LM_H + +#include "dpu_hw_mdss.h" +#include "dpu_hw_util.h" + +struct dpu_hw_mixer; +struct dpu_hw_stage_cfg; + +struct dpu_hw_mixer_cfg { + u32 out_width; + u32 out_height; + bool right_mixer; + int flags; +}; + +struct dpu_hw_color3_cfg { + u8 keep_fg[DPU_STAGE_MAX]; +}; + +/** + * struct dpu_hw_lm_ops : Interface to the mixer Hw driver functions + * Assumption is these functions will be called after clocks are enabled + */ +struct dpu_hw_lm_ops { + /** + * @setup_mixer_out: Sets up mixer output width and height + * and border color if enabled + */ + void (*setup_mixer_out)(struct dpu_hw_mixer *ctx, + struct dpu_hw_mixer_cfg *cfg); + + /** + * @setup_blend_config: Alpha blending configuration + * for the specified stage + */ + void (*setup_blend_config)(struct dpu_hw_mixer *ctx, uint32_t stage, + u16 fg_alpha, u16 bg_alpha, uint32_t blend_op); + + /** + * @setup_alpha_out: Alpha color component selection from either fg or bg + */ + void (*setup_alpha_out)(struct dpu_hw_mixer *ctx, uint32_t mixer_op); + + /** + * @clear_all_blendstages: Clear layer mixer to pipe configuration + * @ctx : mixer ctx pointer + * Returns: 0 on success or -error + */ + int (*clear_all_blendstages)(struct dpu_hw_mixer *ctx); + + /** + * @setup_blendstage: Configure layer mixer to pipe configuration + * @ctx : mixer ctx pointer + * @lm : layer mixer enumeration + * @stage_cfg : blend stage configuration + * Returns: 0 on success or -error + */ + int (*setup_blendstage)(struct dpu_hw_mixer *ctx, enum dpu_lm lm, + struct dpu_hw_stage_cfg *stage_cfg); + + /** + * @setup_border_color : enable/disable border color + */ + void (*setup_border_color)(struct dpu_hw_mixer *ctx, + struct dpu_mdss_color *color, + u8 border_en); + + /** + * @setup_misr: Enable/disable MISR + */ + void (*setup_misr)(struct dpu_hw_mixer *ctx); + + /** + * @collect_misr: Read MISR signature + */ + int (*collect_misr)(struct dpu_hw_mixer *ctx, u32 *misr_value); +}; + +struct dpu_hw_mixer { + struct dpu_hw_blk base; + struct dpu_hw_blk_reg_map hw; + + /* lm */ + enum dpu_lm idx; + const struct dpu_lm_cfg *cap; + const struct dpu_mdp_cfg *mdp; + const struct dpu_ctl_cfg *ctl; + + /* ops */ + struct dpu_hw_lm_ops ops; + + /* store mixer info specific to display */ + struct dpu_hw_mixer_cfg cfg; +}; + +/** + * to_dpu_hw_mixer - convert base object dpu_hw_base to container + * @hw: Pointer to base hardware block + * return: Pointer to hardware block container + */ +static inline struct dpu_hw_mixer *to_dpu_hw_mixer(struct dpu_hw_blk *hw) +{ + return container_of(hw, struct dpu_hw_mixer, base); +} + +struct dpu_hw_mixer *dpu_hw_lm_init(struct drm_device *dev, + const struct dpu_lm_cfg *cfg, + void __iomem *addr, + const struct dpu_mdss_version *mdss_ver); + +#endif /*_DPU_HW_LM_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h new file mode 100644 index 000000000..0e65bf5dd --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h @@ -0,0 +1,415 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2024 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_HW_MDSS_H +#define _DPU_HW_MDSS_H + +#include +#include + +#include "msm_drv.h" + +#include "disp/mdp_format.h" + +#define DPU_DBG_NAME "dpu" + +#define DPU_NONE 0 + +#ifndef DPU_CSC_MATRIX_COEFF_SIZE +#define DPU_CSC_MATRIX_COEFF_SIZE 9 +#endif + +#ifndef DPU_CSC_CLAMP_SIZE +#define DPU_CSC_CLAMP_SIZE 6 +#endif + +#ifndef DPU_CSC_BIAS_SIZE +#define DPU_CSC_BIAS_SIZE 3 +#endif + +#ifndef DPU_MAX_PLANES +#define DPU_MAX_PLANES 4 +#endif + +#define STAGES_PER_PLANE 1 +#define PIPES_PER_STAGE 2 +#define PIPES_PER_PLANE (PIPES_PER_STAGE * STAGES_PER_PLANE) +#ifndef DPU_MAX_DE_CURVES +#define DPU_MAX_DE_CURVES 3 +#endif + +#define DPU_BLEND_FG_ALPHA_FG_CONST (0 << 0) +#define DPU_BLEND_FG_ALPHA_BG_CONST (1 << 0) +#define DPU_BLEND_FG_ALPHA_FG_PIXEL (2 << 0) +#define DPU_BLEND_FG_ALPHA_BG_PIXEL (3 << 0) +#define DPU_BLEND_FG_INV_ALPHA (1 << 2) +#define DPU_BLEND_FG_MOD_ALPHA (1 << 3) +#define DPU_BLEND_FG_INV_MOD_ALPHA (1 << 4) +#define DPU_BLEND_FG_TRANSP_EN (1 << 5) +#define DPU_BLEND_BG_ALPHA_FG_CONST (0 << 8) +#define DPU_BLEND_BG_ALPHA_BG_CONST (1 << 8) +#define DPU_BLEND_BG_ALPHA_FG_PIXEL (2 << 8) +#define DPU_BLEND_BG_ALPHA_BG_PIXEL (3 << 8) +#define DPU_BLEND_BG_INV_ALPHA (1 << 10) +#define DPU_BLEND_BG_MOD_ALPHA (1 << 11) +#define DPU_BLEND_BG_INV_MOD_ALPHA (1 << 12) +#define DPU_BLEND_BG_TRANSP_EN (1 << 13) + +enum dpu_vsync_source { + DPU_VSYNC_SOURCE_GPIO_0, + DPU_VSYNC_SOURCE_GPIO_1, + DPU_VSYNC_SOURCE_GPIO_2, + DPU_VSYNC_SOURCE_INTF_0 = 3, + DPU_VSYNC_SOURCE_INTF_1, + DPU_VSYNC_SOURCE_INTF_2, + DPU_VSYNC_SOURCE_INTF_3, + DPU_VSYNC_SOURCE_WD_TIMER_4 = 11, + DPU_VSYNC_SOURCE_WD_TIMER_3, + DPU_VSYNC_SOURCE_WD_TIMER_2, + DPU_VSYNC_SOURCE_WD_TIMER_1, + DPU_VSYNC_SOURCE_WD_TIMER_0, +}; + +enum dpu_hw_blk_type { + DPU_HW_BLK_TOP = 0, + DPU_HW_BLK_SSPP, + DPU_HW_BLK_LM, + DPU_HW_BLK_CTL, + DPU_HW_BLK_PINGPONG, + DPU_HW_BLK_DCWB_PINGPONG, + DPU_HW_BLK_INTF, + DPU_HW_BLK_WB, + DPU_HW_BLK_DSPP, + DPU_HW_BLK_MERGE_3D, + DPU_HW_BLK_DSC, + DPU_HW_BLK_CDM, + DPU_HW_BLK_CWB, + DPU_HW_BLK_MAX, +}; + +enum dpu_sspp { + SSPP_NONE, + SSPP_VIG0, + SSPP_VIG1, + SSPP_VIG2, + SSPP_VIG3, + SSPP_RGB0, + SSPP_RGB1, + SSPP_RGB2, + SSPP_RGB3, + SSPP_DMA0, + SSPP_DMA1, + SSPP_DMA2, + SSPP_DMA3, + SSPP_DMA4, + SSPP_DMA5, + SSPP_CURSOR0, + SSPP_CURSOR1, + SSPP_MAX +}; + +enum dpu_sspp_type { + SSPP_TYPE_VIG, + SSPP_TYPE_RGB, + SSPP_TYPE_DMA, + SSPP_TYPE_CURSOR, + SSPP_TYPE_MAX +}; + +enum dpu_lm { + LM_0 = 1, + LM_1, + LM_2, + LM_3, + LM_4, + LM_5, + LM_6, + LM_7, + LM_MAX +}; + +enum dpu_stage { + DPU_STAGE_BASE = 0, + DPU_STAGE_0, + DPU_STAGE_1, + DPU_STAGE_2, + DPU_STAGE_3, + DPU_STAGE_4, + DPU_STAGE_5, + DPU_STAGE_6, + DPU_STAGE_7, + DPU_STAGE_8, + DPU_STAGE_9, + DPU_STAGE_10, + DPU_STAGE_MAX +}; +enum dpu_dspp { + DSPP_0 = 1, + DSPP_1, + DSPP_2, + DSPP_3, + DSPP_4, + DSPP_5, + DSPP_6, + DSPP_7, + DSPP_MAX +}; + +enum dpu_ctl { + CTL_0 = 1, + CTL_1, + CTL_2, + CTL_3, + CTL_4, + CTL_5, + CTL_6, + CTL_7, + CTL_MAX +}; + +enum dpu_dsc { + DSC_NONE = 0, + DSC_0, + DSC_1, + DSC_2, + DSC_3, + DSC_4, + DSC_5, + DSC_6, + DSC_7, + DSC_MAX +}; + +enum dpu_cdm { + CDM_0 = 1, + CDM_MAX +}; + +enum dpu_pingpong { + PINGPONG_NONE, + PINGPONG_0, + PINGPONG_1, + PINGPONG_2, + PINGPONG_3, + PINGPONG_4, + PINGPONG_5, + PINGPONG_6, + PINGPONG_7, + PINGPONG_CWB_0, + PINGPONG_CWB_1, + PINGPONG_CWB_2, + PINGPONG_CWB_3, + PINGPONG_S0, + PINGPONG_MAX +}; + +enum dpu_merge_3d { + MERGE_3D_0 = 1, + MERGE_3D_1, + MERGE_3D_2, + MERGE_3D_3, + MERGE_3D_4, + MERGE_3D_5, + MERGE_3D_MAX +}; + +enum dpu_intf { + INTF_0 = 1, + INTF_1, + INTF_2, + INTF_3, + INTF_4, + INTF_5, + INTF_6, + INTF_7, + INTF_8, + INTF_MAX +}; + +/* + * Historically these values correspond to the values written to the + * DISP_INTF_SEL register, which had to programmed manually. On newer MDP + * generations this register is NOP, but we keep the values for historical + * reasons. + */ +enum dpu_intf_type { + INTF_NONE = 0x0, + INTF_DSI = 0x1, + INTF_HDMI = 0x3, + INTF_LCDC = 0x5, + /* old eDP found on 8x74 and 8x84 */ + INTF_EDP = 0x9, + /* both DP and eDP, handled by the new DP driver */ + INTF_DP = 0xa, + + /* virtual interfaces */ + INTF_WB = 0x100, +}; + +enum dpu_intf_mode { + INTF_MODE_NONE = 0, + INTF_MODE_CMD, + INTF_MODE_VIDEO, + INTF_MODE_WB_BLOCK, + INTF_MODE_WB_LINE, + INTF_MODE_MAX +}; + +enum dpu_wb { + WB_0 = 1, + WB_1, + WB_2, + WB_3, + WB_MAX +}; + +enum dpu_cwb { + CWB_0 = 0x1, + CWB_1, + CWB_2, + CWB_3, + CWB_MAX +}; + +enum dpu_wd_timer { + WD_TIMER_0 = 0x1, + WD_TIMER_1, + WD_TIMER_2, + WD_TIMER_3, + WD_TIMER_4, + WD_TIMER_5, + WD_TIMER_MAX +}; + +/** + * enum dpu_3d_blend_mode + * Desribes how the 3d data is blended + * @BLEND_3D_NONE : 3d blending not enabled + * @BLEND_3D_FRAME_INT : Frame interleaving + * @BLEND_3D_H_ROW_INT : Horizontal row interleaving + * @BLEND_3D_V_ROW_INT : vertical row interleaving + * @BLEND_3D_COL_INT : column interleaving + * @BLEND_3D_MAX : + */ +enum dpu_3d_blend_mode { + BLEND_3D_NONE = 0, + BLEND_3D_FRAME_INT, + BLEND_3D_H_ROW_INT, + BLEND_3D_V_ROW_INT, + BLEND_3D_COL_INT, + BLEND_3D_MAX +}; + +/** + * struct dpu_hw_fmt_layout - format information of the source pixel data + * @num_planes: number of planes (including meta data planes) + * @width: image width + * @height: image height + * @total_size: total size in bytes + * @plane_addr: address of each plane + * @plane_size: length of each plane + * @plane_pitch: pitch of each plane + */ +struct dpu_hw_fmt_layout { + uint32_t num_planes; + uint32_t width; + uint32_t height; + uint32_t total_size; + uint32_t plane_addr[DPU_MAX_PLANES]; + uint32_t plane_size[DPU_MAX_PLANES]; + uint32_t plane_pitch[DPU_MAX_PLANES]; +}; + +struct dpu_csc_cfg { + /* matrix coefficients in S15.16 format */ + uint32_t csc_mv[DPU_CSC_MATRIX_COEFF_SIZE]; + uint32_t csc_pre_bv[DPU_CSC_BIAS_SIZE]; + uint32_t csc_post_bv[DPU_CSC_BIAS_SIZE]; + uint32_t csc_pre_lv[DPU_CSC_CLAMP_SIZE]; + uint32_t csc_post_lv[DPU_CSC_CLAMP_SIZE]; +}; + +/** + * struct dpu_mdss_color - mdss color description + * color 0 : green + * color 1 : blue + * color 2 : red + * color 3 : alpha + */ +struct dpu_mdss_color { + u32 color_0; + u32 color_1; + u32 color_2; + u32 color_3; +}; + +/* + * Define bit masks for h/w logging. + */ +#define DPU_DBG_MASK_NONE (1 << 0) +#define DPU_DBG_MASK_INTF (1 << 1) +#define DPU_DBG_MASK_LM (1 << 2) +#define DPU_DBG_MASK_CTL (1 << 3) +#define DPU_DBG_MASK_PINGPONG (1 << 4) +#define DPU_DBG_MASK_SSPP (1 << 5) +#define DPU_DBG_MASK_WB (1 << 6) +#define DPU_DBG_MASK_TOP (1 << 7) +#define DPU_DBG_MASK_VBIF (1 << 8) +#define DPU_DBG_MASK_ROT (1 << 9) +#define DPU_DBG_MASK_DSPP (1 << 10) +#define DPU_DBG_MASK_DSC (1 << 11) +#define DPU_DBG_MASK_CDM (1 << 12) +#define DPU_DBG_MASK_CWB (1 << 13) + +/** + * struct dpu_hw_tear_check - Struct contains parameters to configure + * tear-effect module. This structure is used to configure tear-check + * logic present either in ping-pong or in interface module. + * @vsync_count: Ratio of MDP VSYNC clk freq(Hz) to refresh rate divided + * by no of lines + * @sync_cfg_height: Total vertical lines (display height - 1) + * @vsync_init_val: Init value to which the read pointer gets loaded at + * vsync edge + * @sync_threshold_start: Read pointer threshold start ROI for write operation + * @sync_threshold_continue: The minimum number of lines the write pointer + * needs to be above the read pointer + * @start_pos: The position from which the start_threshold value is added + * @rd_ptr_irq: The read pointer line at which interrupt has to be generated + * @hw_vsync_mode: Sync with external frame sync input + */ +struct dpu_hw_tear_check { + /* + * This is ratio of MDP VSYNC clk freq(Hz) to + * refresh rate divided by no of lines + */ + u32 vsync_count; + u32 sync_cfg_height; + u32 vsync_init_val; + u32 sync_threshold_start; + u32 sync_threshold_continue; + u32 start_pos; + u32 rd_ptr_irq; + u8 hw_vsync_mode; +}; + +/** + * struct dpu_hw_pp_vsync_info - Struct contains parameters to configure + * read and write pointers for command mode panels + * @rd_ptr_init_val: Value of rd pointer at vsync edge + * @rd_ptr_frame_count: Num frames sent since enabling interface + * @rd_ptr_line_count: Current line on panel (rd ptr) + * @wr_ptr_line_count: Current line within pp fifo (wr ptr) + * @intf_frame_count: Frames read from intf + */ +struct dpu_hw_pp_vsync_info { + u32 rd_ptr_init_val; + u32 rd_ptr_frame_count; + u32 rd_ptr_line_count; + u32 wr_ptr_line_count; + u32 intf_frame_count; +}; + +#endif /* _DPU_HW_MDSS_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c new file mode 100644 index 000000000..83b1dbecd --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c @@ -0,0 +1,67 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#include + +#include + +#include "dpu_hw_mdss.h" +#include "dpu_hwio.h" +#include "dpu_hw_catalog.h" +#include "dpu_hw_merge3d.h" +#include "dpu_kms.h" +#include "dpu_trace.h" + +#define MERGE_3D_MUX 0x000 +#define MERGE_3D_MODE 0x004 + +static void dpu_hw_merge_3d_setup_3d_mode(struct dpu_hw_merge_3d *merge_3d, + enum dpu_3d_blend_mode mode_3d) +{ + struct dpu_hw_blk_reg_map *c; + u32 data; + + + c = &merge_3d->hw; + if (mode_3d == BLEND_3D_NONE) { + DPU_REG_WRITE(c, MERGE_3D_MODE, 0); + DPU_REG_WRITE(c, MERGE_3D_MUX, 0); + } else { + data = BIT(0) | ((mode_3d - 1) << 1); + DPU_REG_WRITE(c, MERGE_3D_MODE, data); + } +} + +static void _setup_merge_3d_ops(struct dpu_hw_merge_3d *c) +{ + c->ops.setup_3d_mode = dpu_hw_merge_3d_setup_3d_mode; +}; + +/** + * dpu_hw_merge_3d_init() - Initializes the merge_3d driver for the passed + * merge3d catalog entry. + * @dev: Corresponding device for devres management + * @cfg: Pingpong catalog entry for which driver object is required + * @addr: Mapped register io address of MDP + * Return: Error code or allocated dpu_hw_merge_3d context + */ +struct dpu_hw_merge_3d *dpu_hw_merge_3d_init(struct drm_device *dev, + const struct dpu_merge_3d_cfg *cfg, + void __iomem *addr) +{ + struct dpu_hw_merge_3d *c; + + c = drmm_kzalloc(dev, sizeof(*c), GFP_KERNEL); + if (!c) + return ERR_PTR(-ENOMEM); + + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_PINGPONG; + + c->idx = cfg->id; + c->caps = cfg; + _setup_merge_3d_ops(c); + + return c; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.h new file mode 100644 index 000000000..b57f88187 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.h @@ -0,0 +1,51 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_HW_MERGE3D_H +#define _DPU_HW_MERGE3D_H + +#include "dpu_hw_catalog.h" +#include "dpu_hw_mdss.h" +#include "dpu_hw_util.h" + +struct dpu_hw_merge_3d; + +/** + * struct dpu_hw_merge_3d_ops : Interface to the merge_3d Hw driver functions + * Assumption is these functions will be called after clocks are enabled + * @setup_3d_mode : enable 3D merge + */ +struct dpu_hw_merge_3d_ops { + void (*setup_3d_mode)(struct dpu_hw_merge_3d *merge_3d, + enum dpu_3d_blend_mode mode_3d); + +}; + +struct dpu_hw_merge_3d { + struct dpu_hw_blk base; + struct dpu_hw_blk_reg_map hw; + + /* merge_3d */ + enum dpu_merge_3d idx; + const struct dpu_merge_3d_cfg *caps; + + /* ops */ + struct dpu_hw_merge_3d_ops ops; +}; + +/** + * to_dpu_hw_merge_3d - convert base object dpu_hw_base to container + * @hw: Pointer to base hardware block + * return: Pointer to hardware block container + */ +static inline struct dpu_hw_merge_3d *to_dpu_hw_merge_3d(struct dpu_hw_blk *hw) +{ + return container_of(hw, struct dpu_hw_merge_3d, base); +} + +struct dpu_hw_merge_3d *dpu_hw_merge_3d_init(struct drm_device *dev, + const struct dpu_merge_3d_cfg *cfg, + void __iomem *addr); + +#endif /*_DPU_HW_MERGE3D_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c new file mode 100644 index 000000000..138071be5 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c @@ -0,0 +1,332 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#include + +#include + +#include "dpu_hw_mdss.h" +#include "dpu_hwio.h" +#include "dpu_hw_catalog.h" +#include "dpu_hw_pingpong.h" +#include "dpu_kms.h" +#include "dpu_trace.h" + +#define PP_TEAR_CHECK_EN 0x000 +#define PP_SYNC_CONFIG_VSYNC 0x004 +#define PP_SYNC_CONFIG_HEIGHT 0x008 +#define PP_SYNC_WRCOUNT 0x00C +#define PP_VSYNC_INIT_VAL 0x010 +#define PP_INT_COUNT_VAL 0x014 +#define PP_SYNC_THRESH 0x018 +#define PP_START_POS 0x01C +#define PP_RD_PTR_IRQ 0x020 +#define PP_WR_PTR_IRQ 0x024 +#define PP_OUT_LINE_COUNT 0x028 +#define PP_LINE_COUNT 0x02C +#define PP_AUTOREFRESH_CONFIG 0x030 + +#define PP_FBC_MODE 0x034 +#define PP_FBC_BUDGET_CTL 0x038 +#define PP_FBC_LOSSY_MODE 0x03C +#define PP_DSC_MODE 0x0a0 +#define PP_DCE_DATA_IN_SWAP 0x0ac +#define PP_DCE_DATA_OUT_SWAP 0x0c8 + +#define PP_DITHER_EN 0x000 +#define PP_DITHER_BITDEPTH 0x004 +#define PP_DITHER_MATRIX 0x008 + +#define DITHER_DEPTH_MAP_INDEX 9 + +static u32 dither_depth_map[DITHER_DEPTH_MAP_INDEX] = { + 0, 0, 0, 0, 0, 0, 0, 1, 2 +}; + +static void dpu_hw_pp_setup_dither(struct dpu_hw_pingpong *pp, + struct dpu_hw_dither_cfg *cfg) +{ + struct dpu_hw_blk_reg_map *c; + u32 i, base, data = 0; + + c = &pp->hw; + base = pp->caps->sblk->dither.base; + if (!cfg) { + DPU_REG_WRITE(c, base + PP_DITHER_EN, 0); + return; + } + + data = dither_depth_map[cfg->c0_bitdepth] & REG_MASK(2); + data |= (dither_depth_map[cfg->c1_bitdepth] & REG_MASK(2)) << 2; + data |= (dither_depth_map[cfg->c2_bitdepth] & REG_MASK(2)) << 4; + data |= (dither_depth_map[cfg->c3_bitdepth] & REG_MASK(2)) << 6; + data |= (cfg->temporal_en) ? (1 << 8) : 0; + + DPU_REG_WRITE(c, base + PP_DITHER_BITDEPTH, data); + + for (i = 0; i < DITHER_MATRIX_SZ - 3; i += 4) { + data = (cfg->matrix[i] & REG_MASK(4)) | + ((cfg->matrix[i + 1] & REG_MASK(4)) << 4) | + ((cfg->matrix[i + 2] & REG_MASK(4)) << 8) | + ((cfg->matrix[i + 3] & REG_MASK(4)) << 12); + DPU_REG_WRITE(c, base + PP_DITHER_MATRIX + i, data); + } + DPU_REG_WRITE(c, base + PP_DITHER_EN, 1); +} + +static int dpu_hw_pp_enable_te(struct dpu_hw_pingpong *pp, + struct dpu_hw_tear_check *te) +{ + struct dpu_hw_blk_reg_map *c; + int cfg; + + if (!pp || !te) + return -EINVAL; + c = &pp->hw; + + cfg = BIT(19); /*VSYNC_COUNTER_EN */ + if (te->hw_vsync_mode) + cfg |= BIT(20); + + cfg |= te->vsync_count; + + DPU_REG_WRITE(c, PP_SYNC_CONFIG_VSYNC, cfg); + DPU_REG_WRITE(c, PP_SYNC_CONFIG_HEIGHT, te->sync_cfg_height); + DPU_REG_WRITE(c, PP_VSYNC_INIT_VAL, te->vsync_init_val); + DPU_REG_WRITE(c, PP_RD_PTR_IRQ, te->rd_ptr_irq); + DPU_REG_WRITE(c, PP_START_POS, te->start_pos); + DPU_REG_WRITE(c, PP_SYNC_THRESH, + ((te->sync_threshold_continue << 16) | + te->sync_threshold_start)); + DPU_REG_WRITE(c, PP_SYNC_WRCOUNT, + (te->start_pos + te->sync_threshold_start + 1)); + + DPU_REG_WRITE(c, PP_TEAR_CHECK_EN, 1); + + return 0; +} + +static void dpu_hw_pp_setup_autorefresh_config(struct dpu_hw_pingpong *pp, + u32 frame_count, bool enable) +{ + DPU_REG_WRITE(&pp->hw, PP_AUTOREFRESH_CONFIG, + enable ? (BIT(31) | frame_count) : 0); +} + +/* + * dpu_hw_pp_get_autorefresh_config - Get autorefresh config from HW + * @pp: DPU pingpong structure + * @frame_count: Used to return the current frame count from hw + * + * Returns: True if autorefresh enabled, false if disabled. + */ +static bool dpu_hw_pp_get_autorefresh_config(struct dpu_hw_pingpong *pp, + u32 *frame_count) +{ + u32 val = DPU_REG_READ(&pp->hw, PP_AUTOREFRESH_CONFIG); + if (frame_count != NULL) + *frame_count = val & 0xffff; + return !!((val & BIT(31)) >> 31); +} + +static int dpu_hw_pp_disable_te(struct dpu_hw_pingpong *pp) +{ + struct dpu_hw_blk_reg_map *c; + + if (!pp) + return -EINVAL; + c = &pp->hw; + + DPU_REG_WRITE(c, PP_TEAR_CHECK_EN, 0); + return 0; +} + +static int dpu_hw_pp_connect_external_te(struct dpu_hw_pingpong *pp, + bool enable_external_te) +{ + struct dpu_hw_blk_reg_map *c = &pp->hw; + u32 cfg; + int orig; + + if (!pp) + return -EINVAL; + + c = &pp->hw; + cfg = DPU_REG_READ(c, PP_SYNC_CONFIG_VSYNC); + orig = (bool)(cfg & BIT(20)); + if (enable_external_te) + cfg |= BIT(20); + else + cfg &= ~BIT(20); + DPU_REG_WRITE(c, PP_SYNC_CONFIG_VSYNC, cfg); + trace_dpu_pp_connect_ext_te(pp->idx - PINGPONG_0, cfg); + + return orig; +} + +static int dpu_hw_pp_get_vsync_info(struct dpu_hw_pingpong *pp, + struct dpu_hw_pp_vsync_info *info) +{ + struct dpu_hw_blk_reg_map *c; + u32 val; + + if (!pp || !info) + return -EINVAL; + c = &pp->hw; + + val = DPU_REG_READ(c, PP_VSYNC_INIT_VAL); + info->rd_ptr_init_val = val & 0xffff; + + val = DPU_REG_READ(c, PP_INT_COUNT_VAL); + info->rd_ptr_frame_count = (val & 0xffff0000) >> 16; + info->rd_ptr_line_count = val & 0xffff; + + val = DPU_REG_READ(c, PP_LINE_COUNT); + info->wr_ptr_line_count = val & 0xffff; + + return 0; +} + +static u32 dpu_hw_pp_get_line_count(struct dpu_hw_pingpong *pp) +{ + struct dpu_hw_blk_reg_map *c = &pp->hw; + u32 height, init; + u32 line = 0xFFFF; + + if (!pp) + return 0; + c = &pp->hw; + + init = DPU_REG_READ(c, PP_VSYNC_INIT_VAL) & 0xFFFF; + height = DPU_REG_READ(c, PP_SYNC_CONFIG_HEIGHT) & 0xFFFF; + + if (height < init) + return line; + + line = DPU_REG_READ(c, PP_INT_COUNT_VAL) & 0xFFFF; + + if (line < init) + line += (0xFFFF - init); + else + line -= init; + + return line; +} + +static void dpu_hw_pp_disable_autorefresh(struct dpu_hw_pingpong *pp, + uint32_t encoder_id, u16 vdisplay) +{ + struct dpu_hw_pp_vsync_info info; + int trial = 0; + + /* If autorefresh is already disabled, we have nothing to do */ + if (!dpu_hw_pp_get_autorefresh_config(pp, NULL)) + return; + + /* + * If autorefresh is enabled, disable it and make sure it is safe to + * proceed with current frame commit/push. Sequence followed is, + * 1. Disable TE + * 2. Disable autorefresh config + * 4. Poll for frame transfer ongoing to be false + * 5. Enable TE back + */ + + dpu_hw_pp_connect_external_te(pp, false); + dpu_hw_pp_setup_autorefresh_config(pp, 0, false); + + do { + udelay(DPU_ENC_MAX_POLL_TIMEOUT_US); + if ((trial * DPU_ENC_MAX_POLL_TIMEOUT_US) + > (KICKOFF_TIMEOUT_MS * USEC_PER_MSEC)) { + DPU_ERROR("enc%d pp%d disable autorefresh failed\n", + encoder_id, pp->idx - PINGPONG_0); + break; + } + + trial++; + + dpu_hw_pp_get_vsync_info(pp, &info); + } while (info.wr_ptr_line_count > 0 && + info.wr_ptr_line_count < vdisplay); + + dpu_hw_pp_connect_external_te(pp, true); + + DPU_DEBUG("enc%d pp%d disabled autorefresh\n", + encoder_id, pp->idx - PINGPONG_0); +} + +static int dpu_hw_pp_dsc_enable(struct dpu_hw_pingpong *pp) +{ + struct dpu_hw_blk_reg_map *c = &pp->hw; + + DPU_REG_WRITE(c, PP_DSC_MODE, 1); + return 0; +} + +static void dpu_hw_pp_dsc_disable(struct dpu_hw_pingpong *pp) +{ + struct dpu_hw_blk_reg_map *c = &pp->hw; + + DPU_REG_WRITE(c, PP_DSC_MODE, 0); +} + +static int dpu_hw_pp_setup_dsc(struct dpu_hw_pingpong *pp) +{ + struct dpu_hw_blk_reg_map *pp_c = &pp->hw; + int data; + + data = DPU_REG_READ(pp_c, PP_DCE_DATA_OUT_SWAP); + data |= BIT(18); /* endian flip */ + DPU_REG_WRITE(pp_c, PP_DCE_DATA_OUT_SWAP, data); + return 0; +} + +/** + * dpu_hw_pingpong_init() - initializes the pingpong driver for the passed + * pingpong catalog entry. + * @dev: Corresponding device for devres management + * @cfg: Pingpong catalog entry for which driver object is required + * @addr: Mapped register io address of MDP + * @mdss_rev: dpu core's major and minor versions + * Return: Error code or allocated dpu_hw_pingpong context + */ +struct dpu_hw_pingpong *dpu_hw_pingpong_init(struct drm_device *dev, + const struct dpu_pingpong_cfg *cfg, + void __iomem *addr, + const struct dpu_mdss_version *mdss_rev) +{ + struct dpu_hw_pingpong *c; + + c = drmm_kzalloc(dev, sizeof(*c), GFP_KERNEL); + if (!c) + return ERR_PTR(-ENOMEM); + + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_PINGPONG; + + c->idx = cfg->id; + c->caps = cfg; + + if (mdss_rev->core_major_ver < 5) { + WARN_ON(!cfg->intr_rdptr); + + c->ops.enable_tearcheck = dpu_hw_pp_enable_te; + c->ops.disable_tearcheck = dpu_hw_pp_disable_te; + c->ops.connect_external_te = dpu_hw_pp_connect_external_te; + c->ops.get_line_count = dpu_hw_pp_get_line_count; + c->ops.disable_autorefresh = dpu_hw_pp_disable_autorefresh; + } + + if (mdss_rev->core_major_ver < 7) { + c->ops.setup_dsc = dpu_hw_pp_setup_dsc; + c->ops.enable_dsc = dpu_hw_pp_dsc_enable; + c->ops.disable_dsc = dpu_hw_pp_dsc_disable; + } + + if (mdss_rev->core_major_ver >= 3) + c->ops.setup_dither = dpu_hw_pp_setup_dither; + + return c; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h new file mode 100644 index 000000000..effd012d8 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h @@ -0,0 +1,126 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_HW_PINGPONG_H +#define _DPU_HW_PINGPONG_H + +#include "dpu_hw_catalog.h" +#include "dpu_hw_mdss.h" +#include "dpu_hw_util.h" + +#define DITHER_MATRIX_SZ 16 + +struct dpu_hw_pingpong; + +/** + * struct dpu_hw_dither_cfg - dither feature structure + * @flags: for customizing operations + * @temporal_en: temperal dither enable + * @c0_bitdepth: c0 component bit depth + * @c1_bitdepth: c1 component bit depth + * @c2_bitdepth: c2 component bit depth + * @c3_bitdepth: c2 component bit depth + * @matrix: dither strength matrix + */ +struct dpu_hw_dither_cfg { + u64 flags; + u32 temporal_en; + u32 c0_bitdepth; + u32 c1_bitdepth; + u32 c2_bitdepth; + u32 c3_bitdepth; + u32 matrix[DITHER_MATRIX_SZ]; +}; + +/** + * struct dpu_hw_pingpong_ops : Interface to the pingpong Hw driver functions + * Assumption is these functions will be called after clocks are enabled + * @enable_tearcheck: program and enable tear check block + * @disable_tearcheck: disable able tear check block + * @setup_dither : function to program the dither hw block + * @get_line_count: obtain current vertical line counter + */ +struct dpu_hw_pingpong_ops { + /** + * @enable_tearcheck: enables vysnc generation and sets up init value of + * read pointer and programs the tear check cofiguration + */ + int (*enable_tearcheck)(struct dpu_hw_pingpong *pp, + struct dpu_hw_tear_check *cfg); + + /** + * @disable_tearcheck: disables tear check block + */ + int (*disable_tearcheck)(struct dpu_hw_pingpong *pp); + + /** + * @connect_external_te: read, modify, write to either set or clear + * listening to external TE + * @Return: 1 if TE was originally connected, 0 if not, or -ERROR + */ + int (*connect_external_te)(struct dpu_hw_pingpong *pp, + bool enable_external_te); + + /** + * @get_line_count: Obtain current vertical line counter + */ + u32 (*get_line_count)(struct dpu_hw_pingpong *pp); + + /** + * @disable_autorefresh: Disable autorefresh if enabled + */ + void (*disable_autorefresh)(struct dpu_hw_pingpong *pp, uint32_t encoder_id, u16 vdisplay); + + /** + * @setup_dither: Setup dither matix for pingpong block + */ + void (*setup_dither)(struct dpu_hw_pingpong *pp, + struct dpu_hw_dither_cfg *cfg); + /** + * @enable_dsc: Enable DSC + */ + int (*enable_dsc)(struct dpu_hw_pingpong *pp); + + /** + * @disable_dsc: Disable DSC + */ + void (*disable_dsc)(struct dpu_hw_pingpong *pp); + + /** + * @setup_dsc: Setup DSC + */ + int (*setup_dsc)(struct dpu_hw_pingpong *pp); +}; + +struct dpu_hw_merge_3d; + +struct dpu_hw_pingpong { + struct dpu_hw_blk base; + struct dpu_hw_blk_reg_map hw; + + /* pingpong */ + enum dpu_pingpong idx; + const struct dpu_pingpong_cfg *caps; + struct dpu_hw_merge_3d *merge_3d; + + /* ops */ + struct dpu_hw_pingpong_ops ops; +}; + +/** + * to_dpu_hw_pingpong - convert base object dpu_hw_base to container + * @hw: Pointer to base hardware block + * return: Pointer to hardware block container + */ +static inline struct dpu_hw_pingpong *to_dpu_hw_pingpong(struct dpu_hw_blk *hw) +{ + return container_of(hw, struct dpu_hw_pingpong, base); +} + +struct dpu_hw_pingpong *dpu_hw_pingpong_init(struct drm_device *dev, + const struct dpu_pingpong_cfg *cfg, + void __iomem *addr, + const struct dpu_mdss_version *mdss_rev); + +#endif /*_DPU_HW_PINGPONG_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c new file mode 100644 index 000000000..b5e50fc19 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c @@ -0,0 +1,757 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#include + +#include "dpu_hwio.h" +#include "dpu_hw_catalog.h" +#include "dpu_hw_lm.h" +#include "dpu_hw_sspp.h" +#include "dpu_kms.h" + +#include +#include + +#include + +#define DPU_FETCH_CONFIG_RESET_VALUE 0x00000087 + +/* SSPP registers */ +#define SSPP_SRC_SIZE 0x00 +#define SSPP_SRC_XY 0x08 +#define SSPP_OUT_SIZE 0x0c +#define SSPP_OUT_XY 0x10 +#define SSPP_SRC0_ADDR 0x14 +#define SSPP_SRC1_ADDR 0x18 +#define SSPP_SRC2_ADDR 0x1C +#define SSPP_SRC3_ADDR 0x20 +#define SSPP_SRC_YSTRIDE0 0x24 +#define SSPP_SRC_YSTRIDE1 0x28 +#define SSPP_SRC_FORMAT 0x30 +#define SSPP_SRC_UNPACK_PATTERN 0x34 +#define SSPP_SRC_OP_MODE 0x38 +#define SSPP_SRC_CONSTANT_COLOR 0x3c +#define SSPP_EXCL_REC_CTL 0x40 +#define SSPP_UBWC_STATIC_CTRL 0x44 +#define SSPP_FETCH_CONFIG 0x48 +#define SSPP_DANGER_LUT 0x60 +#define SSPP_SAFE_LUT 0x64 +#define SSPP_CREQ_LUT 0x68 +#define SSPP_QOS_CTRL 0x6C +#define SSPP_SRC_ADDR_SW_STATUS 0x70 +#define SSPP_CREQ_LUT_0 0x74 +#define SSPP_CREQ_LUT_1 0x78 +#define SSPP_DECIMATION_CONFIG 0xB4 +#define SSPP_SW_PIX_EXT_C0_LR 0x100 +#define SSPP_SW_PIX_EXT_C0_TB 0x104 +#define SSPP_SW_PIX_EXT_C0_REQ_PIXELS 0x108 +#define SSPP_SW_PIX_EXT_C1C2_LR 0x110 +#define SSPP_SW_PIX_EXT_C1C2_TB 0x114 +#define SSPP_SW_PIX_EXT_C1C2_REQ_PIXELS 0x118 +#define SSPP_SW_PIX_EXT_C3_LR 0x120 +#define SSPP_SW_PIX_EXT_C3_TB 0x124 +#define SSPP_SW_PIX_EXT_C3_REQ_PIXELS 0x128 +#define SSPP_TRAFFIC_SHAPER 0x130 +#define SSPP_CDP_CNTL 0x134 +#define SSPP_UBWC_ERROR_STATUS 0x138 +#define SSPP_CDP_CNTL_REC1 0x13c +#define SSPP_TRAFFIC_SHAPER_PREFILL 0x150 +#define SSPP_TRAFFIC_SHAPER_REC1_PREFILL 0x154 +#define SSPP_TRAFFIC_SHAPER_REC1 0x158 +#define SSPP_OUT_SIZE_REC1 0x160 +#define SSPP_OUT_XY_REC1 0x164 +#define SSPP_SRC_XY_REC1 0x168 +#define SSPP_SRC_SIZE_REC1 0x16C +#define SSPP_MULTIRECT_OPMODE 0x170 +#define SSPP_SRC_FORMAT_REC1 0x174 +#define SSPP_SRC_UNPACK_PATTERN_REC1 0x178 +#define SSPP_SRC_OP_MODE_REC1 0x17C +#define SSPP_SRC_CONSTANT_COLOR_REC1 0x180 +#define SSPP_EXCL_REC_SIZE_REC1 0x184 +#define SSPP_EXCL_REC_XY_REC1 0x188 +#define SSPP_EXCL_REC_SIZE 0x1B4 +#define SSPP_EXCL_REC_XY 0x1B8 +#define SSPP_UBWC_STATIC_CTRL_REC1 0x1c0 +#define SSPP_UBWC_ERROR_STATUS_REC1 0x1c8 +#define SSPP_CLK_CTRL 0x330 + +/* SSPP_SRC_OP_MODE & OP_MODE_REC1 */ +#define MDSS_MDP_OP_DEINTERLACE BIT(22) +#define MDSS_MDP_OP_DEINTERLACE_ODD BIT(23) +#define MDSS_MDP_OP_IGC_ROM_1 BIT(18) +#define MDSS_MDP_OP_IGC_ROM_0 BIT(17) +#define MDSS_MDP_OP_IGC_EN BIT(16) +#define MDSS_MDP_OP_FLIP_UD BIT(14) +#define MDSS_MDP_OP_FLIP_LR BIT(13) +#define MDSS_MDP_OP_BWC_EN BIT(0) +#define MDSS_MDP_OP_PE_OVERRIDE BIT(31) +#define MDSS_MDP_OP_BWC_LOSSLESS (0 << 1) +#define MDSS_MDP_OP_BWC_Q_HIGH (1 << 1) +#define MDSS_MDP_OP_BWC_Q_MED (2 << 1) + +/* SSPP_QOS_CTRL */ +#define SSPP_QOS_CTRL_VBLANK_EN BIT(16) +#define SSPP_QOS_CTRL_DANGER_SAFE_EN BIT(0) +#define SSPP_QOS_CTRL_DANGER_VBLANK_MASK 0x3 +#define SSPP_QOS_CTRL_DANGER_VBLANK_OFF 4 +#define SSPP_QOS_CTRL_CREQ_VBLANK_MASK 0x3 +#define SSPP_QOS_CTRL_CREQ_VBLANK_OFF 20 + +/* DPU_SSPP_SCALER_QSEED2 */ +#define SSPP_VIG_OP_MODE 0x0 +#define SCALE_CONFIG 0x04 +#define COMP0_3_PHASE_STEP_X 0x10 +#define COMP0_3_PHASE_STEP_Y 0x14 +#define COMP1_2_PHASE_STEP_X 0x18 +#define COMP1_2_PHASE_STEP_Y 0x1c +#define COMP0_3_INIT_PHASE_X 0x20 +#define COMP0_3_INIT_PHASE_Y 0x24 +#define COMP1_2_INIT_PHASE_X 0x28 +#define COMP1_2_INIT_PHASE_Y 0x2C +#define VIG_0_QSEED2_SHARP 0x30 + +/* SSPP_TRAFFIC_SHAPER and _REC1 */ +#define SSPP_TRAFFIC_SHAPER_BPC_MAX 0xFF + +/* + * Definitions for ViG op modes + */ +#define VIG_OP_CSC_DST_DATAFMT BIT(19) +#define VIG_OP_CSC_SRC_DATAFMT BIT(18) +#define VIG_OP_CSC_EN BIT(17) +#define VIG_OP_MEM_PROT_CONT BIT(15) +#define VIG_OP_MEM_PROT_VAL BIT(14) +#define VIG_OP_MEM_PROT_SAT BIT(13) +#define VIG_OP_MEM_PROT_HUE BIT(12) +#define VIG_OP_HIST BIT(8) +#define VIG_OP_SKY_COL BIT(7) +#define VIG_OP_FOIL BIT(6) +#define VIG_OP_SKIN_COL BIT(5) +#define VIG_OP_PA_EN BIT(4) +#define VIG_OP_PA_SAT_ZERO_EXP BIT(2) +#define VIG_OP_MEM_PROT_BLEND BIT(1) + +/* + * Definitions for CSC 10 op modes + */ +#define SSPP_VIG_CSC_10_OP_MODE 0x0 +#define VIG_CSC_10_SRC_DATAFMT BIT(1) +#define VIG_CSC_10_EN BIT(0) +#define CSC_10BIT_OFFSET 4 + +/* traffic shaper clock in Hz */ +#define TS_CLK 19200000 + + +static void dpu_hw_sspp_setup_multirect(struct dpu_sw_pipe *pipe) +{ + struct dpu_hw_sspp *ctx = pipe->sspp; + + if (!ctx) + return; + + dpu_hw_setup_multirect_impl(pipe, ctx, SSPP_MULTIRECT_OPMODE); +} + +void dpu_hw_setup_multirect_impl(struct dpu_sw_pipe *pipe, + struct dpu_hw_sspp *ctx, u32 op_mode_off) +{ + u32 mode_mask; + + if (pipe->multirect_index == DPU_SSPP_RECT_SOLO) { + /** + * if rect index is RECT_SOLO, we cannot expect a + * virtual plane sharing the same SSPP id. So we go + * and disable multirect + */ + mode_mask = 0; + } else { + mode_mask = DPU_REG_READ(&ctx->hw, op_mode_off); + mode_mask |= pipe->multirect_index; + if (pipe->multirect_mode == DPU_SSPP_MULTIRECT_TIME_MX) + mode_mask |= BIT(2); + else + mode_mask &= ~BIT(2); + } + + DPU_REG_WRITE(&ctx->hw, op_mode_off, mode_mask); +} + +void dpu_hw_sspp_setup_opmode(struct dpu_hw_sspp *ctx, + u32 mask, u8 en) +{ + const struct dpu_sspp_sub_blks *sblk = ctx->cap->sblk; + u32 opmode; + + if (!test_bit(DPU_SSPP_SCALER_QSEED2, &ctx->cap->features) || + !test_bit(DPU_SSPP_CSC, &ctx->cap->features)) + return; + + opmode = DPU_REG_READ(&ctx->hw, sblk->scaler_blk.base + SSPP_VIG_OP_MODE); + + if (en) + opmode |= mask; + else + opmode &= ~mask; + + DPU_REG_WRITE(&ctx->hw, sblk->scaler_blk.base + SSPP_VIG_OP_MODE, opmode); +} + +void dpu_hw_sspp_setup_csc10_opmode(struct dpu_hw_sspp *ctx, + u32 mask, u8 en) +{ + const struct dpu_sspp_sub_blks *sblk = ctx->cap->sblk; + u32 opmode; + + opmode = DPU_REG_READ(&ctx->hw, sblk->csc_blk.base + SSPP_VIG_CSC_10_OP_MODE); + if (en) + opmode |= mask; + else + opmode &= ~mask; + + DPU_REG_WRITE(&ctx->hw, sblk->csc_blk.base + SSPP_VIG_CSC_10_OP_MODE, opmode); +} + +/* + * Setup source pixel format, flip, + */ +static void dpu_hw_sspp_setup_format(struct dpu_sw_pipe *pipe, + const struct msm_format *fmt, u32 flags) +{ + struct dpu_hw_sspp *ctx = pipe->sspp; + u32 op_mode_off, unpack_pat_off, format_off, ubwc_ctrl_off, ubwc_error_off; + + if (!ctx || !fmt) + return; + + if (pipe->multirect_index == DPU_SSPP_RECT_SOLO || + pipe->multirect_index == DPU_SSPP_RECT_0) { + op_mode_off = SSPP_SRC_OP_MODE; + unpack_pat_off = SSPP_SRC_UNPACK_PATTERN; + format_off = SSPP_SRC_FORMAT; + ubwc_ctrl_off = SSPP_UBWC_STATIC_CTRL; + ubwc_error_off = SSPP_UBWC_ERROR_STATUS; + } else { + op_mode_off = SSPP_SRC_OP_MODE_REC1; + unpack_pat_off = SSPP_SRC_UNPACK_PATTERN_REC1; + format_off = SSPP_SRC_FORMAT_REC1; + + /* reg wasn't present before DPU 8.0 */ + if (ctx->mdss_ver->core_major_ver >= 8) { + ubwc_ctrl_off = SSPP_UBWC_STATIC_CTRL_REC1; + ubwc_error_off = SSPP_UBWC_ERROR_STATUS_REC1; + } else { + ubwc_ctrl_off = SSPP_UBWC_STATIC_CTRL; + ubwc_error_off = SSPP_UBWC_ERROR_STATUS; + } + } + + if (fmt->fetch_mode != MDP_FETCH_LINEAR) { + u32 hbb = ctx->ubwc->highest_bank_bit - 13; + + DPU_REG_WRITE(&ctx->hw, SSPP_FETCH_CONFIG, + DPU_FETCH_CONFIG_RESET_VALUE | + hbb << 18); + } + + dpu_hw_setup_format_impl(pipe, fmt, flags, ctx, op_mode_off, + unpack_pat_off, format_off, + ubwc_ctrl_off, ubwc_error_off); +} + +void dpu_hw_setup_format_impl(struct dpu_sw_pipe *pipe, const struct msm_format *fmt, + u32 flags, struct dpu_hw_sspp *ctx, u32 op_mode_off, + u32 unpack_pat_off, u32 format_off, u32 ubwc_ctrl_off, + u32 ubwc_error_off) +{ + struct dpu_hw_blk_reg_map *c; + u32 chroma_samp, unpack, src_format; + u32 opmode; + u32 fast_clear; + + c = &ctx->hw; + opmode = DPU_REG_READ(c, op_mode_off); + opmode &= ~(MDSS_MDP_OP_FLIP_LR | MDSS_MDP_OP_FLIP_UD | + MDSS_MDP_OP_BWC_EN | MDSS_MDP_OP_PE_OVERRIDE); + + if (flags & DPU_SSPP_FLIP_LR) + opmode |= MDSS_MDP_OP_FLIP_LR; + if (flags & DPU_SSPP_FLIP_UD) + opmode |= MDSS_MDP_OP_FLIP_UD; + + chroma_samp = fmt->chroma_sample; + if (flags & DPU_SSPP_SOURCE_ROTATED_90) { + if (chroma_samp == CHROMA_H2V1) + chroma_samp = CHROMA_H1V2; + else if (chroma_samp == CHROMA_H1V2) + chroma_samp = CHROMA_H2V1; + } + + src_format = (chroma_samp << 23) | (fmt->fetch_type << 19) | + (fmt->bpc_a << 6) | (fmt->bpc_r_cr << 4) | + (fmt->bpc_b_cb << 2) | (fmt->bpc_g_y << 0); + + if (flags & DPU_SSPP_ROT_90) + src_format |= BIT(11); /* ROT90 */ + + if (fmt->alpha_enable && fmt->fetch_type == MDP_PLANE_INTERLEAVED) + src_format |= BIT(8); /* SRCC3_EN */ + + if (flags & DPU_SSPP_SOLID_FILL) + src_format |= BIT(22); + + unpack = (fmt->element[3] << 24) | (fmt->element[2] << 16) | + (fmt->element[1] << 8) | (fmt->element[0] << 0); + src_format |= ((fmt->unpack_count - 1) << 12) | + ((fmt->flags & MSM_FORMAT_FLAG_UNPACK_TIGHT ? 1 : 0) << 17) | + ((fmt->flags & MSM_FORMAT_FLAG_UNPACK_ALIGN_MSB ? 1 : 0) << 18) | + ((fmt->bpp - 1) << 9); + + if (fmt->fetch_mode != MDP_FETCH_LINEAR) { + u32 hbb = ctx->ubwc->highest_bank_bit - 13; + u32 ctrl_val = 0; + + if (MSM_FORMAT_IS_UBWC(fmt)) + opmode |= MDSS_MDP_OP_BWC_EN; + src_format |= (fmt->fetch_mode & 3) << 30; /*FRAME_FORMAT */ + + if (ctx->ubwc->ubwc_enc_version > UBWC_6_0) { + DRM_WARN_ONCE("Unsupported UBWC version %x\n", ctx->ubwc->ubwc_enc_version); + } else if (ctx->ubwc->ubwc_enc_version >= UBWC_5_0) { + if (!MSM_FORMAT_IS_YUV(fmt)) { + ctrl_val = BIT(30); + if (!MSM_FORMAT_IS_DX(fmt)) /* and not FP16, but it's unsupported */ + ctrl_val |= BIT(31); + } + /* SDE also sets bits for lossy formats, but we don't support them yet */ + } else if (ctx->ubwc->ubwc_enc_version >= UBWC_4_0) { + ctrl_val = MSM_FORMAT_IS_YUV(fmt) ? 0 : BIT(30); + } else if (ctx->ubwc->ubwc_enc_version >= UBWC_3_0) { + ctrl_val = BIT(30) | qcom_ubwc_swizzle(ctx->ubwc) | (hbb << 4); + } else if (ctx->ubwc->ubwc_enc_version >= UBWC_2_0) { + fast_clear = fmt->alpha_enable ? BIT(31) : 0; + ctrl_val = fast_clear | qcom_ubwc_swizzle(ctx->ubwc) | (hbb << 4); + } else if (ctx->ubwc->ubwc_enc_version >= UBWC_1_0) { + fast_clear = fmt->alpha_enable ? BIT(31) : 0; + ctrl_val = fast_clear | + (qcom_ubwc_swizzle(ctx->ubwc) & UBWC_SWIZZLE_ENABLE_LVL1) | + BIT(8) | (hbb << 4); + } else { + DRM_WARN_ONCE("Unsupported UBWC version %x\n", ctx->ubwc->ubwc_enc_version); + } + + DPU_REG_WRITE(c, ubwc_ctrl_off, ctrl_val); + } + + opmode |= MDSS_MDP_OP_PE_OVERRIDE; + + /* if this is YUV pixel format, enable CSC */ + if (MSM_FORMAT_IS_YUV(fmt)) + src_format |= BIT(15); + + if (MSM_FORMAT_IS_DX(fmt)) + src_format |= BIT(14); + + /* update scaler opmode, if appropriate */ + if (test_bit(DPU_SSPP_CSC, &ctx->cap->features)) + dpu_hw_sspp_setup_opmode(ctx, VIG_OP_CSC_EN | VIG_OP_CSC_SRC_DATAFMT, + MSM_FORMAT_IS_YUV(fmt)); + else if (test_bit(DPU_SSPP_CSC_10BIT, &ctx->cap->features)) + dpu_hw_sspp_setup_csc10_opmode(ctx, + VIG_CSC_10_EN | VIG_CSC_10_SRC_DATAFMT, + MSM_FORMAT_IS_YUV(fmt)); + + DPU_REG_WRITE(c, format_off, src_format); + DPU_REG_WRITE(c, unpack_pat_off, unpack); + DPU_REG_WRITE(c, op_mode_off, opmode); + + /* clear previous UBWC error */ + DPU_REG_WRITE(c, ubwc_error_off, BIT(31)); +} + +static void dpu_hw_sspp_setup_pe_config(struct dpu_hw_sspp *ctx, + struct dpu_hw_pixel_ext *pe_ext) +{ + struct dpu_hw_blk_reg_map *c; + u8 color; + u32 lr_pe[4], tb_pe[4], tot_req_pixels[4]; + const u32 bytemask = 0xff; + const u32 shortmask = 0xffff; + + if (!ctx || !pe_ext) + return; + + c = &ctx->hw; + + /* program SW pixel extension override for all pipes*/ + for (color = 0; color < DPU_MAX_PLANES; color++) { + /* color 2 has the same set of registers as color 1 */ + if (color == 2) + continue; + + lr_pe[color] = ((pe_ext->right_ftch[color] & bytemask) << 24)| + ((pe_ext->right_rpt[color] & bytemask) << 16)| + ((pe_ext->left_ftch[color] & bytemask) << 8)| + (pe_ext->left_rpt[color] & bytemask); + + tb_pe[color] = ((pe_ext->btm_ftch[color] & bytemask) << 24)| + ((pe_ext->btm_rpt[color] & bytemask) << 16)| + ((pe_ext->top_ftch[color] & bytemask) << 8)| + (pe_ext->top_rpt[color] & bytemask); + + tot_req_pixels[color] = (((pe_ext->roi_h[color] + + pe_ext->num_ext_pxls_top[color] + + pe_ext->num_ext_pxls_btm[color]) & shortmask) << 16) | + ((pe_ext->roi_w[color] + + pe_ext->num_ext_pxls_left[color] + + pe_ext->num_ext_pxls_right[color]) & shortmask); + } + + /* color 0 */ + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_LR, lr_pe[0]); + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_TB, tb_pe[0]); + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_REQ_PIXELS, + tot_req_pixels[0]); + + /* color 1 and color 2 */ + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_LR, lr_pe[1]); + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_TB, tb_pe[1]); + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_REQ_PIXELS, + tot_req_pixels[1]); + + /* color 3 */ + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_LR, lr_pe[3]); + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_TB, lr_pe[3]); + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_REQ_PIXELS, + tot_req_pixels[3]); +} + +void dpu_hw_sspp_setup_scaler3(struct dpu_hw_sspp *ctx, + struct dpu_hw_scaler3_cfg *scaler3_cfg, + const struct msm_format *format) +{ + if (!ctx || !scaler3_cfg) + return; + + dpu_hw_setup_scaler3(&ctx->hw, scaler3_cfg, + ctx->cap->sblk->scaler_blk.base, + ctx->cap->sblk->scaler_blk.version, + format); +} + +/* + * dpu_hw_sspp_setup_rects() + */ +static void dpu_hw_sspp_setup_rects(struct dpu_sw_pipe *pipe, + struct dpu_sw_pipe_cfg *cfg) +{ + struct dpu_hw_sspp *ctx = pipe->sspp; + u32 src_size_off, src_xy_off, out_size_off, out_xy_off; + + if (!ctx || !cfg) + return; + + if (pipe->multirect_index == DPU_SSPP_RECT_SOLO || + pipe->multirect_index == DPU_SSPP_RECT_0) { + src_size_off = SSPP_SRC_SIZE; + src_xy_off = SSPP_SRC_XY; + out_size_off = SSPP_OUT_SIZE; + out_xy_off = SSPP_OUT_XY; + } else { + src_size_off = SSPP_SRC_SIZE_REC1; + src_xy_off = SSPP_SRC_XY_REC1; + out_size_off = SSPP_OUT_SIZE_REC1; + out_xy_off = SSPP_OUT_XY_REC1; + } + + dpu_hw_setup_rects_impl(pipe, cfg, ctx, src_size_off, + src_xy_off, out_size_off, out_xy_off); +} + +static void dpu_hw_sspp_setup_sourceaddress(struct dpu_sw_pipe *pipe, + struct dpu_hw_fmt_layout *layout) +{ + struct dpu_hw_sspp *ctx = pipe->sspp; + u32 ystride0, ystride1; + int i; + + if (!ctx) + return; + + if (pipe->multirect_index == DPU_SSPP_RECT_SOLO) { + for (i = 0; i < ARRAY_SIZE(layout->plane_addr); i++) + DPU_REG_WRITE(&ctx->hw, SSPP_SRC0_ADDR + i * 0x4, + layout->plane_addr[i]); + } else if (pipe->multirect_index == DPU_SSPP_RECT_0) { + DPU_REG_WRITE(&ctx->hw, SSPP_SRC0_ADDR, + layout->plane_addr[0]); + DPU_REG_WRITE(&ctx->hw, SSPP_SRC2_ADDR, + layout->plane_addr[2]); + } else { + DPU_REG_WRITE(&ctx->hw, SSPP_SRC1_ADDR, + layout->plane_addr[0]); + DPU_REG_WRITE(&ctx->hw, SSPP_SRC3_ADDR, + layout->plane_addr[2]); + } + + if (pipe->multirect_index == DPU_SSPP_RECT_SOLO) { + ystride0 = (layout->plane_pitch[0]) | + (layout->plane_pitch[1] << 16); + ystride1 = (layout->plane_pitch[2]) | + (layout->plane_pitch[3] << 16); + } else { + ystride0 = DPU_REG_READ(&ctx->hw, SSPP_SRC_YSTRIDE0); + ystride1 = DPU_REG_READ(&ctx->hw, SSPP_SRC_YSTRIDE1); + + if (pipe->multirect_index == DPU_SSPP_RECT_0) { + ystride0 = (ystride0 & 0xFFFF0000) | + (layout->plane_pitch[0] & 0x0000FFFF); + ystride1 = (ystride1 & 0xFFFF0000)| + (layout->plane_pitch[2] & 0x0000FFFF); + } else { + ystride0 = (ystride0 & 0x0000FFFF) | + ((layout->plane_pitch[0] << 16) & + 0xFFFF0000); + ystride1 = (ystride1 & 0x0000FFFF) | + ((layout->plane_pitch[2] << 16) & + 0xFFFF0000); + } + } + + DPU_REG_WRITE(&ctx->hw, SSPP_SRC_YSTRIDE0, ystride0); + DPU_REG_WRITE(&ctx->hw, SSPP_SRC_YSTRIDE1, ystride1); +} + +void dpu_hw_sspp_setup_csc(struct dpu_hw_sspp *ctx, + const struct dpu_csc_cfg *data) +{ + u32 offset; + bool csc10 = false; + + if (!ctx || !data) + return; + + offset = ctx->cap->sblk->csc_blk.base; + + if (test_bit(DPU_SSPP_CSC_10BIT, &ctx->cap->features)) { + offset += CSC_10BIT_OFFSET; + csc10 = true; + } + + dpu_hw_csc_setup(&ctx->hw, offset, data, csc10); +} + +static void dpu_hw_sspp_setup_solidfill(struct dpu_sw_pipe *pipe, u32 color) +{ + struct dpu_hw_sspp *ctx = pipe->sspp; + u32 const_clr_off; + + if (!ctx) + return; + + if (pipe->multirect_index == DPU_SSPP_RECT_SOLO || + pipe->multirect_index == DPU_SSPP_RECT_0) + const_clr_off = SSPP_SRC_CONSTANT_COLOR; + else + const_clr_off = SSPP_SRC_CONSTANT_COLOR_REC1; + + dpu_hw_setup_solidfill_impl(pipe, color, ctx, const_clr_off); +} + +void dpu_hw_setup_solidfill_impl(struct dpu_sw_pipe *pipe, + u32 color, struct dpu_hw_sspp *ctx, + u32 const_clr_off) +{ + struct dpu_hw_fmt_layout cfg; + + /* cleanup source addresses */ + memset(&cfg, 0, sizeof(cfg)); + ctx->ops.setup_sourceaddress(pipe, &cfg); + + DPU_REG_WRITE(&ctx->hw, const_clr_off, color); +} + +static void dpu_hw_sspp_setup_qos_lut(struct dpu_hw_sspp *ctx, + struct dpu_hw_qos_cfg *cfg) +{ + if (!ctx || !cfg) + return; + + _dpu_hw_setup_qos_lut(&ctx->hw, SSPP_DANGER_LUT, + ctx->mdss_ver->core_major_ver >= 4, + cfg); +} + +void dpu_hw_sspp_setup_qos_ctrl_impl(struct dpu_hw_sspp *ctx, + bool danger_safe_en, u32 ctrl_off) +{ + DPU_REG_WRITE(&ctx->hw, ctrl_off, + danger_safe_en ? SSPP_QOS_CTRL_DANGER_SAFE_EN : 0); +} + +static void dpu_hw_sspp_setup_qos_ctrl(struct dpu_hw_sspp *ctx, + bool danger_safe_en) +{ + if (!ctx) + return; + + dpu_hw_sspp_setup_qos_ctrl_impl(ctx, danger_safe_en, SSPP_QOS_CTRL); +} + +static void dpu_hw_sspp_setup_cdp(struct dpu_sw_pipe *pipe, + const struct msm_format *fmt, + bool enable) +{ + struct dpu_hw_sspp *ctx = pipe->sspp; + u32 cdp_cntl_offset = 0; + + if (!ctx) + return; + + if (pipe->multirect_index == DPU_SSPP_RECT_SOLO || + pipe->multirect_index == DPU_SSPP_RECT_0) + cdp_cntl_offset = SSPP_CDP_CNTL; + else + cdp_cntl_offset = SSPP_CDP_CNTL_REC1; + + dpu_setup_cdp(&ctx->hw, cdp_cntl_offset, fmt, enable); +} + +static bool dpu_hw_sspp_setup_clk_force_ctrl(struct dpu_hw_sspp *ctx, bool enable) +{ + static const struct dpu_clk_ctrl_reg sspp_clk_ctrl = { + .reg_off = SSPP_CLK_CTRL, + .bit_off = 0 + }; + + return dpu_hw_clk_force_ctrl(&ctx->hw, &sspp_clk_ctrl, enable); +} + +static void _setup_layer_ops(struct dpu_hw_sspp *c, + unsigned long features, const struct dpu_mdss_version *mdss_rev) +{ + c->ops.setup_format = dpu_hw_sspp_setup_format; + c->ops.setup_rects = dpu_hw_sspp_setup_rects; + c->ops.setup_sourceaddress = dpu_hw_sspp_setup_sourceaddress; + c->ops.setup_solidfill = dpu_hw_sspp_setup_solidfill; + c->ops.setup_pe = dpu_hw_sspp_setup_pe_config; + + if (test_bit(DPU_SSPP_QOS, &features)) { + c->ops.setup_qos_lut = dpu_hw_sspp_setup_qos_lut; + c->ops.setup_qos_ctrl = dpu_hw_sspp_setup_qos_ctrl; + } + + if (test_bit(DPU_SSPP_CSC, &features) || + test_bit(DPU_SSPP_CSC_10BIT, &features)) + c->ops.setup_csc = dpu_hw_sspp_setup_csc; + + if (test_bit(DPU_SSPP_SMART_DMA_V1, &c->cap->features) || + test_bit(DPU_SSPP_SMART_DMA_V2, &c->cap->features)) + c->ops.setup_multirect = dpu_hw_sspp_setup_multirect; + + if (test_bit(DPU_SSPP_SCALER_QSEED3_COMPATIBLE, &features)) + c->ops.setup_scaler = dpu_hw_sspp_setup_scaler3; + + if (test_bit(DPU_SSPP_CDP, &features)) + c->ops.setup_cdp = dpu_hw_sspp_setup_cdp; + + if (mdss_rev->core_major_ver >= 9) + c->ops.setup_clk_force_ctrl = dpu_hw_sspp_setup_clk_force_ctrl; +} + +#ifdef CONFIG_DEBUG_FS +int _dpu_hw_sspp_init_debugfs(struct dpu_hw_sspp *hw_pipe, struct dpu_kms *kms, + struct dentry *entry) +{ + const struct dpu_sspp_cfg *cfg = hw_pipe->cap; + const struct dpu_sspp_sub_blks *sblk = cfg->sblk; + struct dentry *debugfs_root; + char sspp_name[32]; + + snprintf(sspp_name, sizeof(sspp_name), "%d", hw_pipe->idx); + + /* create overall sub-directory for the pipe */ + debugfs_root = + debugfs_create_dir(sspp_name, entry); + + /* don't error check these */ + debugfs_create_xul("features", 0600, + debugfs_root, (unsigned long *)&hw_pipe->cap->features); + + /* add register dump support */ + dpu_debugfs_create_regset32("src_blk", 0400, + debugfs_root, + cfg->base, + cfg->len, + kms); + + if (sblk->scaler_blk.len) + dpu_debugfs_create_regset32("scaler_blk", 0400, + debugfs_root, + sblk->scaler_blk.base + cfg->base, + sblk->scaler_blk.len, + kms); + + if (cfg->features & BIT(DPU_SSPP_CSC) || + cfg->features & BIT(DPU_SSPP_CSC_10BIT)) + dpu_debugfs_create_regset32("csc_blk", 0400, + debugfs_root, + sblk->csc_blk.base + cfg->base, + sblk->csc_blk.len, + kms); + + debugfs_create_u32("xin_id", + 0400, + debugfs_root, + (u32 *) &cfg->xin_id); + debugfs_create_u32("clk_ctrl", + 0400, + debugfs_root, + (u32 *) &cfg->clk_ctrl); + + return 0; +} +#endif + +/** + * dpu_hw_sspp_init() - Initializes the sspp hw driver object. + * Should be called once before accessing every pipe. + * @dev: Corresponding device for devres management + * @cfg: Pipe catalog entry for which driver object is required + * @addr: Mapped register io address of MDP + * @mdss_data: UBWC / MDSS configuration data + * @mdss_rev: dpu core's major and minor versions + */ +struct dpu_hw_sspp *dpu_hw_sspp_init(struct drm_device *dev, + const struct dpu_sspp_cfg *cfg, + void __iomem *addr, + const struct qcom_ubwc_cfg_data *mdss_data, + const struct dpu_mdss_version *mdss_rev) +{ + struct dpu_hw_sspp *hw_pipe; + + if (!addr) + return ERR_PTR(-EINVAL); + + hw_pipe = drmm_kzalloc(dev, sizeof(*hw_pipe), GFP_KERNEL); + if (!hw_pipe) + return ERR_PTR(-ENOMEM); + + hw_pipe->hw.blk_addr = addr + cfg->base; + hw_pipe->hw.log_mask = DPU_DBG_MASK_SSPP; + + /* Assign ops */ + hw_pipe->ubwc = mdss_data; + hw_pipe->idx = cfg->id; + hw_pipe->cap = cfg; + + hw_pipe->mdss_ver = mdss_rev; + + if (mdss_rev->core_major_ver >= 13) + dpu_hw_sspp_init_v13(hw_pipe, hw_pipe->cap->features, mdss_rev); + else + _setup_layer_ops(hw_pipe, hw_pipe->cap->features, mdss_rev); + + return hw_pipe; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h new file mode 100644 index 000000000..69d68cc91 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h @@ -0,0 +1,392 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_HW_SSPP_H +#define _DPU_HW_SSPP_H + +#include "dpu_hw_catalog.h" +#include "dpu_hw_mdss.h" +#include "dpu_hw_util.h" +#include "dpu_formats.h" + +struct dpu_hw_sspp; + +#define DPU_SSPP_MAX_PITCH_SIZE 0xffff + +/* + * Flags + */ +#define DPU_SSPP_FLIP_LR BIT(0) +#define DPU_SSPP_FLIP_UD BIT(1) +#define DPU_SSPP_SOURCE_ROTATED_90 BIT(2) +#define DPU_SSPP_ROT_90 BIT(3) +#define DPU_SSPP_SOLID_FILL BIT(4) + +/* + * Component indices + */ +enum { + DPU_SSPP_COMP_0, + DPU_SSPP_COMP_1_2, + DPU_SSPP_COMP_2, + DPU_SSPP_COMP_3, + + DPU_SSPP_COMP_MAX +}; + +/** + * enum dpu_sspp_multirect_index - multirect mode + * @DPU_SSPP_RECT_SOLO: multirect disabled + * @DPU_SSPP_RECT_0: rect0 of a multirect pipe + * @DPU_SSPP_RECT_1: rect1 of a multirect pipe + * + * Note: HW supports multirect with either RECT0 or + * RECT1. Considering no benefit of such configs over + * SOLO mode and to keep the plane management simple, + * we dont support single rect multirect configs. + */ +enum dpu_sspp_multirect_index { + DPU_SSPP_RECT_SOLO = 0, + DPU_SSPP_RECT_0, + DPU_SSPP_RECT_1, +}; + +enum dpu_sspp_multirect_mode { + DPU_SSPP_MULTIRECT_NONE = 0, + DPU_SSPP_MULTIRECT_PARALLEL, + DPU_SSPP_MULTIRECT_TIME_MX, +}; + +enum { + DPU_FRAME_LINEAR, + DPU_FRAME_TILE_A4X, + DPU_FRAME_TILE_A5X, +}; + +enum dpu_hw_filter { + DPU_SCALE_FILTER_NEAREST = 0, + DPU_SCALE_FILTER_BIL, + DPU_SCALE_FILTER_PCMN, + DPU_SCALE_FILTER_CA, + DPU_SCALE_FILTER_MAX +}; + +enum dpu_hw_filter_alpa { + DPU_SCALE_ALPHA_PIXEL_REP, + DPU_SCALE_ALPHA_BIL +}; + +enum dpu_hw_filter_yuv { + DPU_SCALE_2D_4X4, + DPU_SCALE_2D_CIR, + DPU_SCALE_1D_SEP, + DPU_SCALE_BIL +}; + +struct dpu_hw_sharp_cfg { + u32 strength; + u32 edge_thr; + u32 smooth_thr; + u32 noise_thr; +}; + +struct dpu_hw_pixel_ext { + /* scaling factors are enabled for this input layer */ + uint8_t enable_pxl_ext; + + int init_phase_x[DPU_MAX_PLANES]; + int phase_step_x[DPU_MAX_PLANES]; + int init_phase_y[DPU_MAX_PLANES]; + int phase_step_y[DPU_MAX_PLANES]; + + /* + * Number of pixels extension in left, right, top and bottom direction + * for all color components. This pixel value for each color component + * should be sum of fetch + repeat pixels. + */ + int num_ext_pxls_left[DPU_MAX_PLANES]; + int num_ext_pxls_right[DPU_MAX_PLANES]; + int num_ext_pxls_top[DPU_MAX_PLANES]; + int num_ext_pxls_btm[DPU_MAX_PLANES]; + + /* + * Number of pixels needs to be overfetched in left, right, top and + * bottom directions from source image for scaling. + */ + int left_ftch[DPU_MAX_PLANES]; + int right_ftch[DPU_MAX_PLANES]; + int top_ftch[DPU_MAX_PLANES]; + int btm_ftch[DPU_MAX_PLANES]; + + /* + * Number of pixels needs to be repeated in left, right, top and + * bottom directions for scaling. + */ + int left_rpt[DPU_MAX_PLANES]; + int right_rpt[DPU_MAX_PLANES]; + int top_rpt[DPU_MAX_PLANES]; + int btm_rpt[DPU_MAX_PLANES]; + + uint32_t roi_w[DPU_MAX_PLANES]; + uint32_t roi_h[DPU_MAX_PLANES]; + + /* + * Filter type to be used for scaling in horizontal and vertical + * directions + */ + enum dpu_hw_filter horz_filter[DPU_MAX_PLANES]; + enum dpu_hw_filter vert_filter[DPU_MAX_PLANES]; + +}; + +/** + * struct dpu_sw_pipe_cfg : software pipe configuration + * @src_rect: src ROI, caller takes into account the different operations + * such as decimation, flip etc to program this field + * @dst_rect: destination ROI. + * @rotation: simplified drm rotation hint + */ +struct dpu_sw_pipe_cfg { + struct drm_rect src_rect; + struct drm_rect dst_rect; + unsigned int rotation; +}; + +/** + * struct dpu_hw_pipe_ts_cfg - traffic shaper configuration + * @size: size to prefill in bytes, or zero to disable + * @time: time to prefill in usec, or zero to disable + */ +struct dpu_hw_pipe_ts_cfg { + u64 size; + u64 time; +}; + +/** + * struct dpu_sw_pipe - software pipe description + * @sspp: backing SSPP pipe + * @multirect_index: index of the rectangle of SSPP + * @multirect_mode: parallel or time multiplex multirect mode + */ +struct dpu_sw_pipe { + struct dpu_hw_sspp *sspp; + enum dpu_sspp_multirect_index multirect_index; + enum dpu_sspp_multirect_mode multirect_mode; +}; + +/** + * struct dpu_hw_sspp_ops - interface to the SSPP Hw driver functions + * Caller must call the init function to get the pipe context for each pipe + * Assumption is these functions will be called after clocks are enabled + */ +struct dpu_hw_sspp_ops { + /** + * @setup_format: setup pixel format cropping rectangle, flip + * @pipe: Pointer to software pipe context + * @cfg: Pointer to pipe config structure + * @flags: Extra flags for format config + */ + void (*setup_format)(struct dpu_sw_pipe *pipe, + const struct msm_format *fmt, u32 flags); + + /** + * @setup_rects: setup pipe ROI rectangles + * @pipe: Pointer to software pipe context + * @cfg: Pointer to pipe config structure + */ + void (*setup_rects)(struct dpu_sw_pipe *pipe, + struct dpu_sw_pipe_cfg *cfg); + + /** + * @setup_pe: setup pipe pixel extension + * @ctx: Pointer to pipe context + * @pe_ext: Pointer to pixel ext settings + */ + void (*setup_pe)(struct dpu_hw_sspp *ctx, + struct dpu_hw_pixel_ext *pe_ext); + + /** + * @setup_sourceaddress: setup pipe source addresses + * @pipe: Pointer to software pipe context + * @layout: format layout information for programming buffer to hardware + */ + void (*setup_sourceaddress)(struct dpu_sw_pipe *ctx, + struct dpu_hw_fmt_layout *layout); + + /** + * @setup_csc: setup color space coversion + * @ctx: Pointer to pipe context + * @data: Pointer to config structure + */ + void (*setup_csc)(struct dpu_hw_sspp *ctx, const struct dpu_csc_cfg *data); + + /** + * @setup_solidfill: enable/disable colorfill + * @pipe: Pointer to software pipe context + * @const_color: Fill color value + * @flags: Pipe flags + */ + void (*setup_solidfill)(struct dpu_sw_pipe *pipe, u32 color); + + /** + * @setup_multirect: setup multirect configuration + * @pipe: Pointer to software pipe context + */ + + void (*setup_multirect)(struct dpu_sw_pipe *pipe); + + /** + * @setup_sharpening: setup sharpening + * @ctx: Pointer to pipe context + * @cfg: Pointer to config structure + */ + void (*setup_sharpening)(struct dpu_hw_sspp *ctx, + struct dpu_hw_sharp_cfg *cfg); + + /** + * @setup_qos_lut: setup QoS LUTs + * @ctx: Pointer to pipe context + * @cfg: LUT configuration + */ + void (*setup_qos_lut)(struct dpu_hw_sspp *ctx, + struct dpu_hw_qos_cfg *cfg); + + /** + * @setup_qos_ctrl: setup QoS control + * @ctx: Pointer to pipe context + * @danger_safe_en: flags controlling enabling of danger/safe QoS/LUT + */ + void (*setup_qos_ctrl)(struct dpu_hw_sspp *ctx, + bool danger_safe_en); + + /** + * @setup_clk_force_ctrl: setup clock force control + * @ctx: Pointer to pipe context + * @enable: enable clock force if true + */ + bool (*setup_clk_force_ctrl)(struct dpu_hw_sspp *ctx, + bool enable); + + /** + * @setup_histogram: setup histograms + * @ctx: Pointer to pipe context + * @cfg: Pointer to histogram configuration + */ + void (*setup_histogram)(struct dpu_hw_sspp *ctx, + void *cfg); + + /** + * @setup_scaler: setup scaler + * @scaler3_cfg: Pointer to scaler configuration + * @format: pixel format parameters + */ + void (*setup_scaler)(struct dpu_hw_sspp *ctx, + struct dpu_hw_scaler3_cfg *scaler3_cfg, + const struct msm_format *format); + + /** + * @setup_cdp: setup client driven prefetch + * @pipe: Pointer to software pipe context + * @fmt: format used by the sw pipe + * @enable: whether the CDP should be enabled for this pipe + */ + void (*setup_cdp)(struct dpu_sw_pipe *pipe, + const struct msm_format *fmt, + bool enable); +}; + +/** + * struct dpu_hw_sspp - pipe description + * @base: hardware block base structure + * @hw: block hardware details + * @ubwc: UBWC configuration data + * @idx: pipe index + * @cap: pointer to layer_cfg + * @mdss_ver: MDSS version info to use for feature checks + * @ops: pointer to operations possible for this pipe + */ +struct dpu_hw_sspp { + struct dpu_hw_blk base; + struct dpu_hw_blk_reg_map hw; + const struct qcom_ubwc_cfg_data *ubwc; + + /* Pipe */ + enum dpu_sspp idx; + const struct dpu_sspp_cfg *cap; + + const struct dpu_mdss_version *mdss_ver; + + /* Ops */ + struct dpu_hw_sspp_ops ops; +}; + +struct dpu_kms; + +struct dpu_hw_sspp *dpu_hw_sspp_init(struct drm_device *dev, + const struct dpu_sspp_cfg *cfg, + void __iomem *addr, + const struct qcom_ubwc_cfg_data *mdss_data, + const struct dpu_mdss_version *mdss_rev); + +int _dpu_hw_sspp_init_debugfs(struct dpu_hw_sspp *hw_pipe, struct dpu_kms *kms, + struct dentry *entry); + +void dpu_hw_sspp_setup_opmode(struct dpu_hw_sspp *ctx, + u32 mask, u8 en); + +void dpu_hw_sspp_setup_csc10_opmode(struct dpu_hw_sspp *ctx, + u32 mask, u8 en); + +void dpu_hw_sspp_setup_scaler3(struct dpu_hw_sspp *ctx, + struct dpu_hw_scaler3_cfg *scaler3_cfg, + const struct msm_format *format); + +void dpu_hw_sspp_setup_csc(struct dpu_hw_sspp *ctx, + const struct dpu_csc_cfg *data); + +void dpu_hw_setup_multirect_impl(struct dpu_sw_pipe *pipe, + struct dpu_hw_sspp *ctx, + u32 op_mode_off); + +void dpu_hw_setup_format_impl(struct dpu_sw_pipe *pipe, const struct msm_format *fmt, + u32 flags, struct dpu_hw_sspp *ctx, + u32 op_mode_off, u32 unpack_pat_off, u32 format_off, + u32 ubwc_ctrl_off, u32 ubwc_err_off); + +static inline void dpu_hw_setup_rects_impl(struct dpu_sw_pipe *pipe, struct dpu_sw_pipe_cfg *cfg, + struct dpu_hw_sspp *ctx, u32 src_size_off, + u32 src_xy_off, u32 out_size_off, u32 out_xy_off) +{ + struct dpu_hw_blk_reg_map *c; + u32 src_size, src_xy, dst_size, dst_xy; + + c = &ctx->hw; + + /* src and dest rect programming */ + src_xy = (cfg->src_rect.y1 << 16) | cfg->src_rect.x1; + src_size = (drm_rect_height(&cfg->src_rect) << 16) | + drm_rect_width(&cfg->src_rect); + dst_xy = (cfg->dst_rect.y1 << 16) | cfg->dst_rect.x1; + dst_size = (drm_rect_height(&cfg->dst_rect) << 16) | + drm_rect_width(&cfg->dst_rect); + + /* rectangle register programming */ + DPU_REG_WRITE(c, src_size_off, src_size); + DPU_REG_WRITE(c, src_xy_off, src_xy); + DPU_REG_WRITE(c, out_size_off, dst_size); + DPU_REG_WRITE(c, out_xy_off, dst_xy); +} + +void dpu_hw_setup_solidfill_impl(struct dpu_sw_pipe *pipe, + u32 color, struct dpu_hw_sspp *ctx, u32 const_clr_off); + +void dpu_hw_sspp_setup_qos_ctrl_impl(struct dpu_hw_sspp *ctx, + bool danger_safe_en, u32 ctrl_off); + +void dpu_hw_sspp_init_v13(struct dpu_hw_sspp *c, + unsigned long features, + const struct dpu_mdss_version *mdss_rev); + +#endif /*_DPU_HW_SSPP_H */ + diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp_v13.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp_v13.c new file mode 100644 index 000000000..f8f96ad97 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp_v13.c @@ -0,0 +1,323 @@ +// SPDX-License-Identifier: BSD-3-Clause +/* + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. + */ + +#include +#include + +#include "dpu_hw_sspp.h" + +/* >= v13 DPU */ +/* CMN Registers -> Source Surface Processing Pipe Common SSPP registers */ +/* Name Offset */ +#define SSPP_CMN_CLK_CTRL 0x0 +#define SSPP_CMN_CLK_STATUS 0x4 +#define SSPP_CMN_MULTI_REC_OP_MODE 0x10 +#define SSPP_CMN_ADDR_CONFIG 0x14 +#define SSPP_CMN_CAC_CTRL 0x20 +#define SSPP_CMN_SYS_CACHE_MODE 0x24 +#define SSPP_CMN_QOS_CTRL 0x28 + +#define SSPP_CMN_FILL_LEVEL_SCALE 0x3c +#define SSPP_CMN_FILL_LEVELS 0x40 +#define SSPP_CMN_STATUS 0x44 +#define SSPP_CMN_FETCH_DMA_RD_OTS 0x48 +#define SSPP_CMN_FETCH_DTB_WR_PLANE0 0x4c +#define SSPP_CMN_FETCH_DTB_WR_PLANE1 0x50 +#define SSPP_CMN_FETCH_DTB_WR_PLANE2 0x54 +#define SSPP_CMN_DTB_UNPACK_RD_PLANE0 0x58 +#define SSPP_CMN_DTB_UNPACK_RD_PLANE1 0x5c +#define SSPP_CMN_DTB_UNPACK_RD_PLANE2 0x60 +#define SSPP_CMN_UNPACK_LINE_COUNT 0x64 +#define SSPP_CMN_TPG_CONTROL 0x68 +#define SSPP_CMN_TPG_CONFIG 0x6c +#define SSPP_CMN_TPG_COMPONENT_LIMITS 0x70 +#define SSPP_CMN_TPG_RECTANGLE 0x74 +#define SSPP_CMN_TPG_BLACK_WHITE_PATTERN_FRAMES 0x78 +#define SSPP_CMN_TPG_RGB_MAPPING 0x7c +#define SSPP_CMN_TPG_PATTERN_GEN_INIT_VAL 0x80 + +/*RECRegisterset*/ +/*Name Offset*/ +#define SSPP_REC_SRC_FORMAT 0x0 +#define SSPP_REC_SRC_UNPACK_PATTERN 0x4 +#define SSPP_REC_SRC_OP_MODE 0x8 +#define SSPP_REC_SRC_CONSTANT_COLOR 0xc +#define SSPP_REC_SRC_IMG_SIZE 0x10 +#define SSPP_REC_SRC_SIZE 0x14 +#define SSPP_REC_SRC_XY 0x18 +#define SSPP_REC_OUT_SIZE 0x1c +#define SSPP_REC_OUT_XY 0x20 +#define SSPP_REC_SW_PIX_EXT_LR 0x24 +#define SSPP_REC_SW_PIX_EXT_TB 0x28 +#define SSPP_REC_SRC_SIZE_ODX 0x30 +#define SSPP_REC_SRC_XY_ODX 0x34 +#define SSPP_REC_OUT_SIZE_ODX 0x38 +#define SSPP_REC_OUT_XY_ODX 0x3c +#define SSPP_REC_SW_PIX_EXT_LR_ODX 0x40 +#define SSPP_REC_SW_PIX_EXT_TB_ODX 0x44 +#define SSPP_REC_PRE_DOWN_SCALE 0x48 +#define SSPP_REC_SRC0_ADDR 0x4c +#define SSPP_REC_SRC1_ADDR 0x50 +#define SSPP_REC_SRC2_ADDR 0x54 +#define SSPP_REC_SRC3_ADDR 0x58 +#define SSPP_REC_SRC_YSTRIDE0 0x5c +#define SSPP_REC_SRC_YSTRIDE1 0x60 +#define SSPP_REC_CURRENT_SRC0_ADDR 0x64 +#define SSPP_REC_CURRENT_SRC1_ADDR 0x68 +#define SSPP_REC_CURRENT_SRC2_ADDR 0x6c +#define SSPP_REC_CURRENT_SRC3_ADDR 0x70 +#define SSPP_REC_SRC_ADDR_SW_STATUS 0x74 +#define SSPP_REC_CDP_CNTL 0x78 +#define SSPP_REC_TRAFFIC_SHAPER 0x7c +#define SSPP_REC_TRAFFIC_SHAPER_PREFILL 0x80 +#define SSPP_REC_PD_MEM_ALLOC 0x84 +#define SSPP_REC_QOS_CLAMP 0x88 +#define SSPP_REC_UIDLE_CTRL_VALUE 0x8c +#define SSPP_REC_UBWC_STATIC_CTRL 0x90 +#define SSPP_REC_UBWC_STATIC_CTRL_OVERRIDE 0x94 +#define SSPP_REC_UBWC_STATS_ROI 0x98 +#define SSPP_REC_UBWC_STATS_WORST_TILE_ROW_BW_ROI0 0x9c +#define SSPP_REC_UBWC_STATS_TOTAL_BW_ROI0 0xa0 +#define SSPP_REC_UBWC_STATS_WORST_TILE_ROW_BW_ROI1 0xa4 +#define SSPP_REC_UBWC_STATS_TOTAL_BW_ROI1 0xa8 +#define SSPP_REC_UBWC_STATS_WORST_TILE_ROW_BW_ROI2 0xac +#define SSPP_REC_UBWC_STATS_TOTAL_BW_ROI2 0xb0 +#define SSPP_REC_EXCL_REC_CTRL 0xb4 +#define SSPP_REC_EXCL_REC_SIZE 0xb8 +#define SSPP_REC_EXCL_REC_XY 0xbc +#define SSPP_REC_LINE_INSERTION_CTRL 0xc0 +#define SSPP_REC_LINE_INSERTION_OUT_SIZE 0xc4 +#define SSPP_REC_FETCH_PIPE_ACTIVE 0xc8 +#define SSPP_REC_META_ERROR_STATUS 0xcc +#define SSPP_REC_UBWC_ERROR_STATUS 0xd0 +#define SSPP_REC_FLUSH_CTRL 0xd4 +#define SSPP_REC_INTR_EN 0xd8 +#define SSPP_REC_INTR_STATUS 0xdc +#define SSPP_REC_INTR_CLEAR 0xe0 +#define SSPP_REC_HSYNC_STATUS 0xe4 +#define SSPP_REC_FP16_CONFIG 0x150 +#define SSPP_REC_FP16_CSC_MATRIX_COEFF_R_0 0x154 +#define SSPP_REC_FP16_CSC_MATRIX_COEFF_R_1 0x158 +#define SSPP_REC_FP16_CSC_MATRIX_COEFF_G_0 0x15c +#define SSPP_REC_FP16_CSC_MATRIX_COEFF_G_1 0x160 +#define SSPP_REC_FP16_CSC_MATRIX_COEFF_B_0 0x164 +#define SSPP_REC_FP16_CSC_MATRIX_COEFF_B_1 0x168 +#define SSPP_REC_FP16_CSC_PRE_CLAMP_R 0x16c +#define SSPP_REC_FP16_CSC_PRE_CLAMP_G 0x170 +#define SSPP_REC_FP16_CSC_PRE_CLAMP_B 0x174 +#define SSPP_REC_FP16_CSC_POST_CLAMP 0x178 + +static inline u32 dpu_hw_sspp_calculate_rect_off(enum dpu_sspp_multirect_index rect_index, + struct dpu_hw_sspp *ctx) +{ + return (rect_index == DPU_SSPP_RECT_SOLO || rect_index == DPU_SSPP_RECT_0) ? + ctx->cap->sblk->sspp_rec0_blk.base : ctx->cap->sblk->sspp_rec1_blk.base; +} + +static void dpu_hw_sspp_setup_multirect_v13(struct dpu_sw_pipe *pipe) +{ + struct dpu_hw_sspp *ctx = pipe->sspp; + + if (!ctx) + return; + + dpu_hw_setup_multirect_impl(pipe, ctx, SSPP_CMN_MULTI_REC_OP_MODE); +} + +static void dpu_hw_sspp_setup_format_v13(struct dpu_sw_pipe *pipe, + const struct msm_format *fmt, u32 flags) +{ + struct dpu_hw_sspp *ctx = pipe->sspp; + u32 op_mode_off, unpack_pat_off, format_off; + u32 ubwc_ctrl_off, ubwc_err_off; + u32 offset; + + if (!ctx || !fmt) + return; + + offset = dpu_hw_sspp_calculate_rect_off(pipe->multirect_index, ctx); + + op_mode_off = offset + SSPP_REC_SRC_OP_MODE; + unpack_pat_off = offset + SSPP_REC_SRC_UNPACK_PATTERN; + format_off = offset + SSPP_REC_SRC_FORMAT; + ubwc_ctrl_off = offset + SSPP_REC_UBWC_STATIC_CTRL; + ubwc_err_off = offset + SSPP_REC_UBWC_ERROR_STATUS; + + dpu_hw_setup_format_impl(pipe, fmt, flags, ctx, op_mode_off, + unpack_pat_off, format_off, ubwc_ctrl_off, ubwc_err_off); +} + +static void dpu_hw_sspp_setup_pe_config_v13(struct dpu_hw_sspp *ctx, + struct dpu_hw_pixel_ext *pe_ext) +{ + struct dpu_hw_blk_reg_map *c; + u8 color; + u32 lr_pe[4], tb_pe[4]; + const u32 bytemask = 0xff; + u32 offset; + + if (!ctx || !pe_ext) + return; + + offset = ctx->cap->sblk->sspp_rec0_blk.base; + + c = &ctx->hw; + /* program SW pixel extension override for all pipes*/ + for (color = 0; color < DPU_MAX_PLANES; color++) { + /* color 2 has the same set of registers as color 1 */ + if (color == 2) + continue; + + lr_pe[color] = ((pe_ext->right_ftch[color] & bytemask) << 24) | + ((pe_ext->right_rpt[color] & bytemask) << 16) | + ((pe_ext->left_ftch[color] & bytemask) << 8) | + (pe_ext->left_rpt[color] & bytemask); + + tb_pe[color] = ((pe_ext->btm_ftch[color] & bytemask) << 24) | + ((pe_ext->btm_rpt[color] & bytemask) << 16) | + ((pe_ext->top_ftch[color] & bytemask) << 8) | + (pe_ext->top_rpt[color] & bytemask); + } + + /* color 0 */ + DPU_REG_WRITE(c, SSPP_REC_SW_PIX_EXT_LR + offset, lr_pe[0]); + DPU_REG_WRITE(c, SSPP_REC_SW_PIX_EXT_TB + offset, tb_pe[0]); + + /* color 1 and color 2 */ + DPU_REG_WRITE(c, SSPP_REC_SW_PIX_EXT_LR_ODX + offset, lr_pe[1]); + DPU_REG_WRITE(c, SSPP_REC_SW_PIX_EXT_TB_ODX + offset, tb_pe[1]); +} + +static void dpu_hw_sspp_setup_rects_v13(struct dpu_sw_pipe *pipe, + struct dpu_sw_pipe_cfg *cfg) +{ + struct dpu_hw_sspp *ctx = pipe->sspp; + u32 src_size_off, src_xy_off, out_size_off, out_xy_off; + u32 offset; + + if (!ctx || !cfg) + return; + + offset = dpu_hw_sspp_calculate_rect_off(pipe->multirect_index, ctx); + + src_size_off = offset + SSPP_REC_SRC_SIZE; + src_xy_off = offset + SSPP_REC_SRC_XY; + out_size_off = offset + SSPP_REC_OUT_SIZE; + out_xy_off = offset + SSPP_REC_OUT_XY; + + dpu_hw_setup_rects_impl(pipe, cfg, ctx, src_size_off, + src_xy_off, out_size_off, out_xy_off); +} + +static void dpu_hw_sspp_setup_sourceaddress_v13(struct dpu_sw_pipe *pipe, + struct dpu_hw_fmt_layout *layout) +{ + struct dpu_hw_sspp *ctx = pipe->sspp; + int i; + u32 offset, ystride0, ystride1; + + if (!ctx) + return; + + offset = dpu_hw_sspp_calculate_rect_off(pipe->multirect_index, ctx); + + for (i = 0; i < ARRAY_SIZE(layout->plane_addr); i++) + DPU_REG_WRITE(&ctx->hw, offset + SSPP_REC_SRC0_ADDR + i * 0x4, + layout->plane_addr[i]); + + ystride0 = (layout->plane_pitch[0]) | (layout->plane_pitch[2] << 16); + ystride1 = (layout->plane_pitch[1]) | (layout->plane_pitch[3] << 16); + + DPU_REG_WRITE(&ctx->hw, offset + SSPP_REC_SRC_YSTRIDE0, ystride0); + DPU_REG_WRITE(&ctx->hw, offset + SSPP_REC_SRC_YSTRIDE1, ystride1); +} + +static void dpu_hw_sspp_setup_solidfill_v13(struct dpu_sw_pipe *pipe, u32 color) +{ + struct dpu_hw_sspp *ctx = pipe->sspp; + u32 const_clr_off; + u32 offset; + + if (!ctx) + return; + + offset = dpu_hw_sspp_calculate_rect_off(pipe->multirect_index, ctx); + const_clr_off = offset + SSPP_REC_SRC_CONSTANT_COLOR; + + dpu_hw_setup_solidfill_impl(pipe, color, ctx, const_clr_off); +} + +static void dpu_hw_sspp_setup_qos_lut_v13(struct dpu_hw_sspp *ctx, + struct dpu_hw_qos_cfg *cfg) +{ + if (!ctx || !cfg) + return; + + dpu_hw_setup_qos_lut_v13(&ctx->hw, cfg); +} + +static void dpu_hw_sspp_setup_qos_ctrl_v13(struct dpu_hw_sspp *ctx, + bool danger_safe_en) +{ + if (!ctx) + return; + + dpu_hw_sspp_setup_qos_ctrl_impl(ctx, danger_safe_en, SSPP_CMN_QOS_CTRL); +} + +static void dpu_hw_sspp_setup_cdp_v13(struct dpu_sw_pipe *pipe, + const struct msm_format *fmt, + bool enable) +{ + struct dpu_hw_sspp *ctx = pipe->sspp; + u32 offset = 0; + + if (!ctx) + return; + + offset = dpu_hw_sspp_calculate_rect_off(pipe->multirect_index, ctx); + dpu_setup_cdp(&ctx->hw, offset + SSPP_REC_CDP_CNTL, fmt, enable); +} + +static bool dpu_hw_sspp_setup_clk_force_ctrl_v13(struct dpu_hw_sspp *ctx, bool enable) +{ + static const struct dpu_clk_ctrl_reg sspp_clk_ctrl = { + .reg_off = SSPP_CMN_CLK_CTRL, + .bit_off = 0 + }; + + return dpu_hw_clk_force_ctrl(&ctx->hw, &sspp_clk_ctrl, enable); +} + +void dpu_hw_sspp_init_v13(struct dpu_hw_sspp *c, + unsigned long features, const struct dpu_mdss_version *mdss_rev) +{ + c->ops.setup_format = dpu_hw_sspp_setup_format_v13; + c->ops.setup_rects = dpu_hw_sspp_setup_rects_v13; + c->ops.setup_sourceaddress = dpu_hw_sspp_setup_sourceaddress_v13; + c->ops.setup_solidfill = dpu_hw_sspp_setup_solidfill_v13; + c->ops.setup_pe = dpu_hw_sspp_setup_pe_config_v13; + + if (test_bit(DPU_SSPP_QOS, &features)) { + c->ops.setup_qos_lut = dpu_hw_sspp_setup_qos_lut_v13; + c->ops.setup_qos_ctrl = dpu_hw_sspp_setup_qos_ctrl_v13; + } + + if (test_bit(DPU_SSPP_CSC, &features) || + test_bit(DPU_SSPP_CSC_10BIT, &features)) + c->ops.setup_csc = dpu_hw_sspp_setup_csc; + + if (test_bit(DPU_SSPP_SMART_DMA_V1, &c->cap->features) || + test_bit(DPU_SSPP_SMART_DMA_V2, &c->cap->features)) + c->ops.setup_multirect = dpu_hw_sspp_setup_multirect_v13; + + if (test_bit(DPU_SSPP_SCALER_QSEED3_COMPATIBLE, &features)) + c->ops.setup_scaler = dpu_hw_sspp_setup_scaler3; + + if (test_bit(DPU_SSPP_CDP, &features)) + c->ops.setup_cdp = dpu_hw_sspp_setup_cdp_v13; + + c->ops.setup_clk_force_ctrl = dpu_hw_sspp_setup_clk_force_ctrl_v13; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c new file mode 100644 index 000000000..1ebd75d4f --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c @@ -0,0 +1,312 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#include + +#include + +#include "dpu_hwio.h" +#include "dpu_hw_catalog.h" +#include "dpu_hw_top.h" +#include "dpu_kms.h" + +#define FLD_SPLIT_DISPLAY_CMD BIT(1) +#define FLD_SMART_PANEL_FREE_RUN BIT(2) +#define FLD_INTF_1_SW_TRG_MUX BIT(4) +#define FLD_INTF_2_SW_TRG_MUX BIT(8) +#define FLD_TE_LINE_INTER_WATERLEVEL_MASK 0xFFFF + +#define TRAFFIC_SHAPER_EN BIT(31) +#define TRAFFIC_SHAPER_RD_CLIENT(num) (0x030 + (num * 4)) +#define TRAFFIC_SHAPER_WR_CLIENT(num) (0x060 + (num * 4)) +#define TRAFFIC_SHAPER_FIXPOINT_FACTOR 4 + +static void dpu_hw_setup_split_pipe(struct dpu_hw_mdp *mdp, + struct split_pipe_cfg *cfg) +{ + struct dpu_hw_blk_reg_map *c; + u32 upper_pipe = 0; + u32 lower_pipe = 0; + + if (!mdp || !cfg) + return; + + c = &mdp->hw; + + if (cfg->en) { + if (cfg->mode == INTF_MODE_CMD) { + lower_pipe = FLD_SPLIT_DISPLAY_CMD; + /* interface controlling sw trigger */ + if (cfg->intf == INTF_2) + lower_pipe |= FLD_INTF_1_SW_TRG_MUX; + else + lower_pipe |= FLD_INTF_2_SW_TRG_MUX; + upper_pipe = lower_pipe; + } else { + if (cfg->intf == INTF_2) { + lower_pipe = FLD_INTF_1_SW_TRG_MUX; + upper_pipe = FLD_INTF_2_SW_TRG_MUX; + } else { + lower_pipe = FLD_INTF_2_SW_TRG_MUX; + upper_pipe = FLD_INTF_1_SW_TRG_MUX; + } + } + } + + DPU_REG_WRITE(c, SSPP_SPARE, cfg->split_flush_en ? 0x1 : 0x0); + DPU_REG_WRITE(c, SPLIT_DISPLAY_LOWER_PIPE_CTRL, lower_pipe); + DPU_REG_WRITE(c, SPLIT_DISPLAY_UPPER_PIPE_CTRL, upper_pipe); + DPU_REG_WRITE(c, SPLIT_DISPLAY_EN, cfg->en & 0x1); +} + +static bool dpu_hw_setup_clk_force_ctrl(struct dpu_hw_mdp *mdp, + enum dpu_clk_ctrl_type clk_ctrl, bool enable) +{ + if (!mdp) + return false; + + if (clk_ctrl <= DPU_CLK_CTRL_NONE || clk_ctrl >= DPU_CLK_CTRL_MAX) + return false; + + return dpu_hw_clk_force_ctrl(&mdp->hw, &mdp->caps->clk_ctrls[clk_ctrl], enable); +} + + +static void dpu_hw_get_danger_status(struct dpu_hw_mdp *mdp, + struct dpu_danger_safe_status *status) +{ + struct dpu_hw_blk_reg_map *c; + u32 value; + + if (!mdp || !status) + return; + + c = &mdp->hw; + + value = DPU_REG_READ(c, DANGER_STATUS); + status->mdp = (value >> 0) & 0x3; + status->sspp[SSPP_VIG0] = (value >> 4) & 0x3; + status->sspp[SSPP_VIG1] = (value >> 6) & 0x3; + status->sspp[SSPP_VIG2] = (value >> 8) & 0x3; + status->sspp[SSPP_VIG3] = (value >> 10) & 0x3; + status->sspp[SSPP_RGB0] = (value >> 12) & 0x3; + status->sspp[SSPP_RGB1] = (value >> 14) & 0x3; + status->sspp[SSPP_RGB2] = (value >> 16) & 0x3; + status->sspp[SSPP_RGB3] = (value >> 18) & 0x3; + status->sspp[SSPP_DMA0] = (value >> 20) & 0x3; + status->sspp[SSPP_DMA1] = (value >> 22) & 0x3; + status->sspp[SSPP_DMA2] = (value >> 28) & 0x3; + status->sspp[SSPP_DMA3] = (value >> 30) & 0x3; + status->sspp[SSPP_CURSOR0] = (value >> 24) & 0x3; + status->sspp[SSPP_CURSOR1] = (value >> 26) & 0x3; +} + +static void dpu_hw_setup_wd_timer(struct dpu_hw_mdp *mdp, + struct dpu_vsync_source_cfg *cfg) +{ + struct dpu_hw_blk_reg_map *c; + u32 reg, wd_load_value, wd_ctl, wd_ctl2; + + if (!mdp || !cfg) + return; + + c = &mdp->hw; + + if (cfg->vsync_source >= DPU_VSYNC_SOURCE_WD_TIMER_4 && + cfg->vsync_source <= DPU_VSYNC_SOURCE_WD_TIMER_0) { + switch (cfg->vsync_source) { + case DPU_VSYNC_SOURCE_WD_TIMER_4: + wd_load_value = MDP_WD_TIMER_4_LOAD_VALUE; + wd_ctl = MDP_WD_TIMER_4_CTL; + wd_ctl2 = MDP_WD_TIMER_4_CTL2; + break; + case DPU_VSYNC_SOURCE_WD_TIMER_3: + wd_load_value = MDP_WD_TIMER_3_LOAD_VALUE; + wd_ctl = MDP_WD_TIMER_3_CTL; + wd_ctl2 = MDP_WD_TIMER_3_CTL2; + break; + case DPU_VSYNC_SOURCE_WD_TIMER_2: + wd_load_value = MDP_WD_TIMER_2_LOAD_VALUE; + wd_ctl = MDP_WD_TIMER_2_CTL; + wd_ctl2 = MDP_WD_TIMER_2_CTL2; + break; + case DPU_VSYNC_SOURCE_WD_TIMER_1: + wd_load_value = MDP_WD_TIMER_1_LOAD_VALUE; + wd_ctl = MDP_WD_TIMER_1_CTL; + wd_ctl2 = MDP_WD_TIMER_1_CTL2; + break; + case DPU_VSYNC_SOURCE_WD_TIMER_0: + default: + wd_load_value = MDP_WD_TIMER_0_LOAD_VALUE; + wd_ctl = MDP_WD_TIMER_0_CTL; + wd_ctl2 = MDP_WD_TIMER_0_CTL2; + break; + } + + DPU_REG_WRITE(c, wd_load_value, + CALCULATE_WD_LOAD_VALUE(cfg->frame_rate)); + + DPU_REG_WRITE(c, wd_ctl, BIT(0)); /* clear timer */ + reg = DPU_REG_READ(c, wd_ctl2); + reg |= BIT(8); /* enable heartbeat timer */ + reg |= BIT(0); /* enable WD timer */ + DPU_REG_WRITE(c, wd_ctl2, reg); + + /* make sure that timers are enabled/disabled for vsync state */ + wmb(); + } +} + +static void dpu_hw_setup_vsync_sel(struct dpu_hw_mdp *mdp, + struct dpu_vsync_source_cfg *cfg) +{ + struct dpu_hw_blk_reg_map *c; + u32 reg, i; + static const u32 pp_offset[PINGPONG_MAX] = {0xC, 0x8, 0x4, 0x13, 0x18}; + + if (!mdp || !cfg || (cfg->pp_count > ARRAY_SIZE(cfg->ppnumber))) + return; + + c = &mdp->hw; + + reg = DPU_REG_READ(c, MDP_VSYNC_SEL); + for (i = 0; i < cfg->pp_count; i++) { + int pp_idx = cfg->ppnumber[i] - PINGPONG_0; + + if (pp_idx >= ARRAY_SIZE(pp_offset)) + continue; + + reg &= ~(0xf << pp_offset[pp_idx]); + reg |= (cfg->vsync_source & 0xf) << pp_offset[pp_idx]; + } + DPU_REG_WRITE(c, MDP_VSYNC_SEL, reg); + + dpu_hw_setup_wd_timer(mdp, cfg); +} + +static void dpu_hw_get_safe_status(struct dpu_hw_mdp *mdp, + struct dpu_danger_safe_status *status) +{ + struct dpu_hw_blk_reg_map *c; + u32 value; + + if (!mdp || !status) + return; + + c = &mdp->hw; + + value = DPU_REG_READ(c, SAFE_STATUS); + status->mdp = (value >> 0) & 0x1; + status->sspp[SSPP_VIG0] = (value >> 4) & 0x1; + status->sspp[SSPP_VIG1] = (value >> 6) & 0x1; + status->sspp[SSPP_VIG2] = (value >> 8) & 0x1; + status->sspp[SSPP_VIG3] = (value >> 10) & 0x1; + status->sspp[SSPP_RGB0] = (value >> 12) & 0x1; + status->sspp[SSPP_RGB1] = (value >> 14) & 0x1; + status->sspp[SSPP_RGB2] = (value >> 16) & 0x1; + status->sspp[SSPP_RGB3] = (value >> 18) & 0x1; + status->sspp[SSPP_DMA0] = (value >> 20) & 0x1; + status->sspp[SSPP_DMA1] = (value >> 22) & 0x1; + status->sspp[SSPP_DMA2] = (value >> 28) & 0x1; + status->sspp[SSPP_DMA3] = (value >> 30) & 0x1; + status->sspp[SSPP_CURSOR0] = (value >> 24) & 0x1; + status->sspp[SSPP_CURSOR1] = (value >> 26) & 0x1; +} + +static void dpu_hw_intf_audio_select(struct dpu_hw_mdp *mdp) +{ + struct dpu_hw_blk_reg_map *c; + + if (!mdp) + return; + + c = &mdp->hw; + + DPU_REG_WRITE(c, HDMI_DP_CORE_SELECT, 0x1); +} + +static void dpu_hw_dp_phy_intf_sel(struct dpu_hw_mdp *mdp, + enum dpu_dp_phy_sel phys[2]) +{ + struct dpu_hw_blk_reg_map *c = &mdp->hw; + unsigned int intf; + u32 sel = 0; + + sel |= FIELD_PREP(MDP_DP_PHY_INTF_SEL_INTF0, phys[0]); + sel |= FIELD_PREP(MDP_DP_PHY_INTF_SEL_INTF1, phys[1]); + + for (intf = 0; intf < 2; intf++) { + switch (phys[intf]) { + case DPU_DP_PHY_0: + sel |= FIELD_PREP(MDP_DP_PHY_INTF_SEL_PHY0, intf + 1); + break; + case DPU_DP_PHY_1: + sel |= FIELD_PREP(MDP_DP_PHY_INTF_SEL_PHY1, intf + 1); + break; + case DPU_DP_PHY_2: + sel |= FIELD_PREP(MDP_DP_PHY_INTF_SEL_PHY2, intf + 1); + break; + default: + /* ignore */ + break; + } + } + + DPU_REG_WRITE(c, MDP_DP_PHY_INTF_SEL, sel); +} + +static void _setup_mdp_ops(struct dpu_hw_mdp_ops *ops, + const struct dpu_mdss_version *mdss_rev) +{ + ops->setup_split_pipe = dpu_hw_setup_split_pipe; + ops->setup_clk_force_ctrl = dpu_hw_setup_clk_force_ctrl; + ops->get_danger_status = dpu_hw_get_danger_status; + + if (mdss_rev->core_major_ver < 5) + ops->setup_vsync_source = dpu_hw_setup_vsync_sel; + else if (mdss_rev->core_major_ver < 8) + ops->setup_vsync_source = dpu_hw_setup_wd_timer; + + ops->get_safe_status = dpu_hw_get_safe_status; + + if (mdss_rev->core_major_ver >= 5) + ops->dp_phy_intf_sel = dpu_hw_dp_phy_intf_sel; + + if (mdss_rev->core_major_ver == 4 || + mdss_rev->core_major_ver == 5) + ops->intf_audio_select = dpu_hw_intf_audio_select; +} + +/** + * dpu_hw_mdptop_init - initializes the top driver for the passed config + * @dev: Corresponding device for devres management + * @cfg: MDP TOP configuration from catalog + * @addr: Mapped register io address of MDP + * @mdss_rev: dpu core's major and minor versions + */ +struct dpu_hw_mdp *dpu_hw_mdptop_init(struct drm_device *dev, + const struct dpu_mdp_cfg *cfg, + void __iomem *addr, + const struct dpu_mdss_version *mdss_rev) +{ + struct dpu_hw_mdp *mdp; + + if (!addr) + return ERR_PTR(-EINVAL); + + mdp = drmm_kzalloc(dev, sizeof(*mdp), GFP_KERNEL); + if (!mdp) + return ERR_PTR(-ENOMEM); + + mdp->hw.blk_addr = addr + cfg->base; + mdp->hw.log_mask = DPU_DBG_MASK_TOP; + + /* + * Assign ops + */ + mdp->caps = cfg; + _setup_mdp_ops(&mdp->ops, mdss_rev); + + return mdp; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.h new file mode 100644 index 000000000..80279d87c --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.h @@ -0,0 +1,164 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_HW_TOP_H +#define _DPU_HW_TOP_H + +#include "dpu_hw_catalog.h" +#include "dpu_hw_mdss.h" +#include "dpu_hw_util.h" + +struct dpu_hw_mdp; + +/** + * struct traffic_shaper_cfg: traffic shaper configuration + * @en : enable/disable traffic shaper + * @rd_client : true if read client; false if write client + * @client_id : client identifier + * @bpc_denom : denominator of byte per clk + * @bpc_numer : numerator of byte per clk + */ +struct traffic_shaper_cfg { + bool en; + bool rd_client; + u32 client_id; + u32 bpc_denom; + u64 bpc_numer; +}; + +/** + * struct split_pipe_cfg - pipe configuration for dual display panels + * @en : Enable/disable dual pipe configuration + * @mode : Panel interface mode + * @intf : Interface id for main control path + * @split_flush_en: Allows both the paths to be flushed when master path is + * flushed + */ +struct split_pipe_cfg { + bool en; + enum dpu_intf_mode mode; + enum dpu_intf intf; + bool split_flush_en; +}; + +/** + * struct dpu_danger_safe_status: danger and safe status signals + * @mdp: top level status + * @sspp: source pipe status + */ +struct dpu_danger_safe_status { + u8 mdp; + u8 sspp[SSPP_MAX]; +}; + +/** + * struct dpu_vsync_source_cfg - configure vsync source and configure the + * watchdog timers if required. + * @pp_count: number of ping pongs active + * @frame_rate: Display frame rate + * @ppnumber: ping pong index array + * @vsync_source: vsync source selection + */ +struct dpu_vsync_source_cfg { + u32 pp_count; + u32 frame_rate; + u32 ppnumber[PINGPONG_MAX]; + enum dpu_vsync_source vsync_source; +}; + +enum dpu_dp_phy_sel { + DPU_DP_PHY_NONE, + DPU_DP_PHY_0, + DPU_DP_PHY_1, + DPU_DP_PHY_2, +}; + +/** + * struct dpu_hw_mdp_ops - interface to the MDP TOP Hw driver functions + * Assumption is these functions will be called after clocks are enabled. + */ +struct dpu_hw_mdp_ops { + /** + * @setup_split_pipe : Programs the pipe control registers. + * Registers are not double buffered, this + * function should be called before timing control enable + * @mdp : mdp top context driver + * @cfg : upper and lower part of pipe configuration + */ + void (*setup_split_pipe)(struct dpu_hw_mdp *mdp, + struct split_pipe_cfg *p); + + /** + * @setup_traffic_shaper : programs traffic shaper control. + * @mdp : mdp top context driver + * @cfg : traffic shaper configuration + */ + void (*setup_traffic_shaper)(struct dpu_hw_mdp *mdp, + struct traffic_shaper_cfg *cfg); + + /** + * @setup_clk_force_ctrl: set clock force control + * @mdp: mdp top context driver + * @clk_ctrl: clock to be controlled + * @enable: force on enable + * @return: if the clock is forced-on by this function + */ + bool (*setup_clk_force_ctrl)(struct dpu_hw_mdp *mdp, + enum dpu_clk_ctrl_type clk_ctrl, bool enable); + + /** + * @get_danger_status: get danger status + * @mdp: mdp top context driver + * @status: Pointer to danger safe status + */ + void (*get_danger_status)(struct dpu_hw_mdp *mdp, + struct dpu_danger_safe_status *status); + + /** + * @setup_vsync_source: setup vsync source configuration details + * @mdp: mdp top context driver + * @cfg: vsync source selection configuration + */ + void (*setup_vsync_source)(struct dpu_hw_mdp *mdp, + struct dpu_vsync_source_cfg *cfg); + + /** + * @get_safe_status: get safe status + * @mdp: mdp top context driver + * @status: Pointer to danger safe status + */ + void (*get_safe_status)(struct dpu_hw_mdp *mdp, + struct dpu_danger_safe_status *status); + + /** + * @dp_phy_intf_sel: configure intf to phy mapping + * @mdp: mdp top context driver + * @phys: list of phys the DP interfaces should be connected to. 0 disables the INTF. + */ + void (*dp_phy_intf_sel)(struct dpu_hw_mdp *mdp, enum dpu_dp_phy_sel phys[2]); + + /** + * @intf_audio_select: select the external interface for audio + * @mdp: mdp top context driver + */ + void (*intf_audio_select)(struct dpu_hw_mdp *mdp); +}; + +struct dpu_hw_mdp { + struct dpu_hw_blk base; + struct dpu_hw_blk_reg_map hw; + + /* top */ + const struct dpu_mdp_cfg *caps; + + /* ops */ + struct dpu_hw_mdp_ops ops; +}; + +struct dpu_hw_mdp *dpu_hw_mdptop_init(struct drm_device *dev, + const struct dpu_mdp_cfg *cfg, + void __iomem *addr, + const struct dpu_mdss_version *mdss_rev); + +#endif /*_DPU_HW_TOP_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c new file mode 100644 index 000000000..c7a5f4a0d --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c @@ -0,0 +1,621 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2022-2023 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ +#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ + +#include "msm_drv.h" +#include "dpu_kms.h" +#include "dpu_hw_mdss.h" +#include "dpu_hw_util.h" + +/* using a file static variables for debugfs access */ +static u32 dpu_hw_util_log_mask = DPU_DBG_MASK_NONE; + +/* DPU_SCALER_QSEED3 */ +#define QSEED3_HW_VERSION 0x00 +#define QSEED3_OP_MODE 0x04 +#define QSEED3_RGB2Y_COEFF 0x08 +#define QSEED3_PHASE_INIT 0x0C +#define QSEED3_PHASE_STEP_Y_H 0x10 +#define QSEED3_PHASE_STEP_Y_V 0x14 +#define QSEED3_PHASE_STEP_UV_H 0x18 +#define QSEED3_PHASE_STEP_UV_V 0x1C +#define QSEED3_PRELOAD 0x20 +#define QSEED3_DE_SHARPEN 0x24 +#define QSEED3_DE_SHARPEN_CTL 0x28 +#define QSEED3_DE_SHAPE_CTL 0x2C +#define QSEED3_DE_THRESHOLD 0x30 +#define QSEED3_DE_ADJUST_DATA_0 0x34 +#define QSEED3_DE_ADJUST_DATA_1 0x38 +#define QSEED3_DE_ADJUST_DATA_2 0x3C +#define QSEED3_SRC_SIZE_Y_RGB_A 0x40 +#define QSEED3_SRC_SIZE_UV 0x44 +#define QSEED3_DST_SIZE 0x48 +#define QSEED3_COEF_LUT_CTRL 0x4C +#define QSEED3_COEF_LUT_SWAP_BIT 0 +#define QSEED3_COEF_LUT_DIR_BIT 1 +#define QSEED3_COEF_LUT_Y_CIR_BIT 2 +#define QSEED3_COEF_LUT_UV_CIR_BIT 3 +#define QSEED3_COEF_LUT_Y_SEP_BIT 4 +#define QSEED3_COEF_LUT_UV_SEP_BIT 5 +#define QSEED3_BUFFER_CTRL 0x50 +#define QSEED3_CLK_CTRL0 0x54 +#define QSEED3_CLK_CTRL1 0x58 +#define QSEED3_CLK_STATUS 0x5C +#define QSEED3_PHASE_INIT_Y_H 0x90 +#define QSEED3_PHASE_INIT_Y_V 0x94 +#define QSEED3_PHASE_INIT_UV_H 0x98 +#define QSEED3_PHASE_INIT_UV_V 0x9C +#define QSEED3_COEF_LUT 0x100 +#define QSEED3_FILTERS 5 +#define QSEED3_LUT_REGIONS 4 +#define QSEED3_CIRCULAR_LUTS 9 +#define QSEED3_SEPARABLE_LUTS 10 +#define QSEED3_LUT_SIZE 60 +#define QSEED3_ENABLE 2 +#define QSEED3_DIR_LUT_SIZE (200 * sizeof(u32)) +#define QSEED3_CIR_LUT_SIZE \ + (QSEED3_LUT_SIZE * QSEED3_CIRCULAR_LUTS * sizeof(u32)) +#define QSEED3_SEP_LUT_SIZE \ + (QSEED3_LUT_SIZE * QSEED3_SEPARABLE_LUTS * sizeof(u32)) + +/* DPU_SCALER_QSEED3LITE */ +#define QSEED3LITE_COEF_LUT_Y_SEP_BIT 4 +#define QSEED3LITE_COEF_LUT_UV_SEP_BIT 5 +#define QSEED3LITE_COEF_LUT_CTRL 0x4C +#define QSEED3LITE_COEF_LUT_SWAP_BIT 0 +#define QSEED3LITE_DIR_FILTER_WEIGHT 0x60 +#define QSEED3LITE_FILTERS 2 +#define QSEED3LITE_SEPARABLE_LUTS 10 +#define QSEED3LITE_LUT_SIZE 33 +#define QSEED3LITE_SEP_LUT_SIZE \ + (QSEED3LITE_LUT_SIZE * QSEED3LITE_SEPARABLE_LUTS * sizeof(u32)) + +/* QOS_LUT */ +#define QOS_DANGER_LUT 0x00 +#define QOS_SAFE_LUT 0x04 +#define QOS_CREQ_LUT 0x08 +#define QOS_QOS_CTRL 0x0C +#define QOS_CREQ_LUT_0 0x14 +#define QOS_CREQ_LUT_1 0x18 + +/* CMN_QOS_LUT */ +#define SSPP_CMN_QOS_CTRL 0x28 +#define SSPP_CMN_DANGER_LUT 0x2c +#define SSPP_CMN_SAFE_LUT 0x30 +#define SSPP_CMN_CREQ_LUT_0 0x34 +#define SSPP_CMN_CREQ_LUT_1 0x38 + +/* QOS_QOS_CTRL */ +#define QOS_QOS_CTRL_DANGER_SAFE_EN BIT(0) +#define QOS_QOS_CTRL_DANGER_VBLANK_MASK GENMASK(5, 4) +#define QOS_QOS_CTRL_VBLANK_EN BIT(16) +#define QOS_QOS_CTRL_CREQ_VBLANK_MASK GENMASK(21, 20) + +void dpu_reg_write(struct dpu_hw_blk_reg_map *c, + u32 reg_off, + u32 val, + const char *name) +{ + /* don't need to mutex protect this */ + if (c->log_mask & dpu_hw_util_log_mask) + DPU_DEBUG_DRIVER("[%s:0x%X] <= 0x%X\n", + name, reg_off, val); + writel_relaxed(val, c->blk_addr + reg_off); +} + +int dpu_reg_read(struct dpu_hw_blk_reg_map *c, u32 reg_off) +{ + return readl_relaxed(c->blk_addr + reg_off); +} + +u32 *dpu_hw_util_get_log_mask_ptr(void) +{ + return &dpu_hw_util_log_mask; +} + +static void _dpu_hw_setup_scaler3_lut(struct dpu_hw_blk_reg_map *c, + struct dpu_hw_scaler3_cfg *scaler3_cfg, u32 offset) +{ + int i, j, filter; + int config_lut = 0x0; + unsigned long lut_flags; + u32 lut_addr, lut_offset, lut_len; + u32 *lut[QSEED3_FILTERS] = {NULL, NULL, NULL, NULL, NULL}; + static const uint32_t off_tbl[QSEED3_FILTERS][QSEED3_LUT_REGIONS][2] = { + {{18, 0x000}, {12, 0x120}, {12, 0x1E0}, {8, 0x2A0} }, + {{6, 0x320}, {3, 0x3E0}, {3, 0x440}, {3, 0x4A0} }, + {{6, 0x500}, {3, 0x5c0}, {3, 0x620}, {3, 0x680} }, + {{6, 0x380}, {3, 0x410}, {3, 0x470}, {3, 0x4d0} }, + {{6, 0x560}, {3, 0x5f0}, {3, 0x650}, {3, 0x6b0} }, + }; + + lut_flags = (unsigned long) scaler3_cfg->lut_flag; + if (test_bit(QSEED3_COEF_LUT_DIR_BIT, &lut_flags) && + (scaler3_cfg->dir_len == QSEED3_DIR_LUT_SIZE)) { + lut[0] = scaler3_cfg->dir_lut; + config_lut = 1; + } + if (test_bit(QSEED3_COEF_LUT_Y_CIR_BIT, &lut_flags) && + (scaler3_cfg->y_rgb_cir_lut_idx < QSEED3_CIRCULAR_LUTS) && + (scaler3_cfg->cir_len == QSEED3_CIR_LUT_SIZE)) { + lut[1] = scaler3_cfg->cir_lut + + scaler3_cfg->y_rgb_cir_lut_idx * QSEED3_LUT_SIZE; + config_lut = 1; + } + if (test_bit(QSEED3_COEF_LUT_UV_CIR_BIT, &lut_flags) && + (scaler3_cfg->uv_cir_lut_idx < QSEED3_CIRCULAR_LUTS) && + (scaler3_cfg->cir_len == QSEED3_CIR_LUT_SIZE)) { + lut[2] = scaler3_cfg->cir_lut + + scaler3_cfg->uv_cir_lut_idx * QSEED3_LUT_SIZE; + config_lut = 1; + } + if (test_bit(QSEED3_COEF_LUT_Y_SEP_BIT, &lut_flags) && + (scaler3_cfg->y_rgb_sep_lut_idx < QSEED3_SEPARABLE_LUTS) && + (scaler3_cfg->sep_len == QSEED3_SEP_LUT_SIZE)) { + lut[3] = scaler3_cfg->sep_lut + + scaler3_cfg->y_rgb_sep_lut_idx * QSEED3_LUT_SIZE; + config_lut = 1; + } + if (test_bit(QSEED3_COEF_LUT_UV_SEP_BIT, &lut_flags) && + (scaler3_cfg->uv_sep_lut_idx < QSEED3_SEPARABLE_LUTS) && + (scaler3_cfg->sep_len == QSEED3_SEP_LUT_SIZE)) { + lut[4] = scaler3_cfg->sep_lut + + scaler3_cfg->uv_sep_lut_idx * QSEED3_LUT_SIZE; + config_lut = 1; + } + + if (config_lut) { + for (filter = 0; filter < QSEED3_FILTERS; filter++) { + if (!lut[filter]) + continue; + lut_offset = 0; + for (i = 0; i < QSEED3_LUT_REGIONS; i++) { + lut_addr = QSEED3_COEF_LUT + offset + + off_tbl[filter][i][1]; + lut_len = off_tbl[filter][i][0] << 2; + for (j = 0; j < lut_len; j++) { + DPU_REG_WRITE(c, + lut_addr, + (lut[filter])[lut_offset++]); + lut_addr += 4; + } + } + } + } + + if (test_bit(QSEED3_COEF_LUT_SWAP_BIT, &lut_flags)) + DPU_REG_WRITE(c, QSEED3_COEF_LUT_CTRL + offset, BIT(0)); + +} + +static void _dpu_hw_setup_scaler3lite_lut(struct dpu_hw_blk_reg_map *c, + struct dpu_hw_scaler3_cfg *scaler3_cfg, u32 offset) +{ + int j, filter; + int config_lut = 0x0; + unsigned long lut_flags; + u32 lut_addr, lut_offset; + u32 *lut[QSEED3LITE_FILTERS] = {NULL, NULL}; + static const uint32_t off_tbl[QSEED3_FILTERS] = { 0x000, 0x200 }; + + DPU_REG_WRITE(c, QSEED3LITE_DIR_FILTER_WEIGHT + offset, scaler3_cfg->dir_weight); + + if (!scaler3_cfg->sep_lut) + return; + + lut_flags = (unsigned long) scaler3_cfg->lut_flag; + if (test_bit(QSEED3_COEF_LUT_Y_SEP_BIT, &lut_flags) && + (scaler3_cfg->y_rgb_sep_lut_idx < QSEED3LITE_SEPARABLE_LUTS) && + (scaler3_cfg->sep_len == QSEED3LITE_SEP_LUT_SIZE)) { + lut[0] = scaler3_cfg->sep_lut + + scaler3_cfg->y_rgb_sep_lut_idx * QSEED3LITE_LUT_SIZE; + config_lut = 1; + } + if (test_bit(QSEED3_COEF_LUT_UV_SEP_BIT, &lut_flags) && + (scaler3_cfg->uv_sep_lut_idx < QSEED3LITE_SEPARABLE_LUTS) && + (scaler3_cfg->sep_len == QSEED3LITE_SEP_LUT_SIZE)) { + lut[1] = scaler3_cfg->sep_lut + + scaler3_cfg->uv_sep_lut_idx * QSEED3LITE_LUT_SIZE; + config_lut = 1; + } + + if (config_lut) { + for (filter = 0; filter < QSEED3LITE_FILTERS; filter++) { + if (!lut[filter]) + continue; + lut_offset = 0; + lut_addr = QSEED3_COEF_LUT + offset + off_tbl[filter]; + for (j = 0; j < QSEED3LITE_LUT_SIZE; j++) { + DPU_REG_WRITE(c, + lut_addr, + (lut[filter])[lut_offset++]); + lut_addr += 4; + } + } + } + + if (test_bit(QSEED3_COEF_LUT_SWAP_BIT, &lut_flags)) + DPU_REG_WRITE(c, QSEED3_COEF_LUT_CTRL + offset, BIT(0)); + +} + +static void _dpu_hw_setup_scaler3_de(struct dpu_hw_blk_reg_map *c, + struct dpu_hw_scaler3_de_cfg *de_cfg, u32 offset) +{ + u32 sharp_lvl, sharp_ctl, shape_ctl, de_thr; + u32 adjust_a, adjust_b, adjust_c; + + if (!de_cfg->enable) + return; + + sharp_lvl = (de_cfg->sharpen_level1 & 0x1FF) | + ((de_cfg->sharpen_level2 & 0x1FF) << 16); + + sharp_ctl = ((de_cfg->limit & 0xF) << 9) | + ((de_cfg->prec_shift & 0x7) << 13) | + ((de_cfg->clip & 0x7) << 16); + + shape_ctl = (de_cfg->thr_quiet & 0xFF) | + ((de_cfg->thr_dieout & 0x3FF) << 16); + + de_thr = (de_cfg->thr_low & 0x3FF) | + ((de_cfg->thr_high & 0x3FF) << 16); + + adjust_a = (de_cfg->adjust_a[0] & 0x3FF) | + ((de_cfg->adjust_a[1] & 0x3FF) << 10) | + ((de_cfg->adjust_a[2] & 0x3FF) << 20); + + adjust_b = (de_cfg->adjust_b[0] & 0x3FF) | + ((de_cfg->adjust_b[1] & 0x3FF) << 10) | + ((de_cfg->adjust_b[2] & 0x3FF) << 20); + + adjust_c = (de_cfg->adjust_c[0] & 0x3FF) | + ((de_cfg->adjust_c[1] & 0x3FF) << 10) | + ((de_cfg->adjust_c[2] & 0x3FF) << 20); + + DPU_REG_WRITE(c, QSEED3_DE_SHARPEN + offset, sharp_lvl); + DPU_REG_WRITE(c, QSEED3_DE_SHARPEN_CTL + offset, sharp_ctl); + DPU_REG_WRITE(c, QSEED3_DE_SHAPE_CTL + offset, shape_ctl); + DPU_REG_WRITE(c, QSEED3_DE_THRESHOLD + offset, de_thr); + DPU_REG_WRITE(c, QSEED3_DE_ADJUST_DATA_0 + offset, adjust_a); + DPU_REG_WRITE(c, QSEED3_DE_ADJUST_DATA_1 + offset, adjust_b); + DPU_REG_WRITE(c, QSEED3_DE_ADJUST_DATA_2 + offset, adjust_c); + +} + +void dpu_hw_setup_scaler3(struct dpu_hw_blk_reg_map *c, + struct dpu_hw_scaler3_cfg *scaler3_cfg, + u32 scaler_offset, u32 scaler_version, + const struct msm_format *format) +{ + u32 op_mode = 0; + u32 phase_init, preload, src_y_rgb, src_uv, dst; + + if (!scaler3_cfg->enable) + goto end; + + op_mode |= BIT(0); + op_mode |= (scaler3_cfg->y_rgb_filter_cfg & 0x3) << 16; + + if (format && MSM_FORMAT_IS_YUV(format)) { + op_mode |= BIT(12); + op_mode |= (scaler3_cfg->uv_filter_cfg & 0x3) << 24; + } + + op_mode |= (scaler3_cfg->blend_cfg & 1) << 31; + op_mode |= (scaler3_cfg->dir_en) ? BIT(4) : 0; + + preload = + ((scaler3_cfg->preload_x[0] & 0x7F) << 0) | + ((scaler3_cfg->preload_y[0] & 0x7F) << 8) | + ((scaler3_cfg->preload_x[1] & 0x7F) << 16) | + ((scaler3_cfg->preload_y[1] & 0x7F) << 24); + + src_y_rgb = (scaler3_cfg->src_width[0] & 0x1FFFF) | + ((scaler3_cfg->src_height[0] & 0x1FFFF) << 16); + + src_uv = (scaler3_cfg->src_width[1] & 0x1FFFF) | + ((scaler3_cfg->src_height[1] & 0x1FFFF) << 16); + + dst = (scaler3_cfg->dst_width & 0x1FFFF) | + ((scaler3_cfg->dst_height & 0x1FFFF) << 16); + + if (scaler3_cfg->de.enable) { + _dpu_hw_setup_scaler3_de(c, &scaler3_cfg->de, scaler_offset); + op_mode |= BIT(8); + } + + if (scaler3_cfg->lut_flag) { + if (scaler_version < 0x2004) + _dpu_hw_setup_scaler3_lut(c, scaler3_cfg, scaler_offset); + else + _dpu_hw_setup_scaler3lite_lut(c, scaler3_cfg, scaler_offset); + } + + if (scaler_version == 0x1002) { + phase_init = + ((scaler3_cfg->init_phase_x[0] & 0x3F) << 0) | + ((scaler3_cfg->init_phase_y[0] & 0x3F) << 8) | + ((scaler3_cfg->init_phase_x[1] & 0x3F) << 16) | + ((scaler3_cfg->init_phase_y[1] & 0x3F) << 24); + DPU_REG_WRITE(c, QSEED3_PHASE_INIT + scaler_offset, phase_init); + } else { + DPU_REG_WRITE(c, QSEED3_PHASE_INIT_Y_H + scaler_offset, + scaler3_cfg->init_phase_x[0] & 0x1FFFFF); + DPU_REG_WRITE(c, QSEED3_PHASE_INIT_Y_V + scaler_offset, + scaler3_cfg->init_phase_y[0] & 0x1FFFFF); + DPU_REG_WRITE(c, QSEED3_PHASE_INIT_UV_H + scaler_offset, + scaler3_cfg->init_phase_x[1] & 0x1FFFFF); + DPU_REG_WRITE(c, QSEED3_PHASE_INIT_UV_V + scaler_offset, + scaler3_cfg->init_phase_y[1] & 0x1FFFFF); + } + + DPU_REG_WRITE(c, QSEED3_PHASE_STEP_Y_H + scaler_offset, + scaler3_cfg->phase_step_x[0] & 0xFFFFFF); + + DPU_REG_WRITE(c, QSEED3_PHASE_STEP_Y_V + scaler_offset, + scaler3_cfg->phase_step_y[0] & 0xFFFFFF); + + DPU_REG_WRITE(c, QSEED3_PHASE_STEP_UV_H + scaler_offset, + scaler3_cfg->phase_step_x[1] & 0xFFFFFF); + + DPU_REG_WRITE(c, QSEED3_PHASE_STEP_UV_V + scaler_offset, + scaler3_cfg->phase_step_y[1] & 0xFFFFFF); + + DPU_REG_WRITE(c, QSEED3_PRELOAD + scaler_offset, preload); + + DPU_REG_WRITE(c, QSEED3_SRC_SIZE_Y_RGB_A + scaler_offset, src_y_rgb); + + DPU_REG_WRITE(c, QSEED3_SRC_SIZE_UV + scaler_offset, src_uv); + + DPU_REG_WRITE(c, QSEED3_DST_SIZE + scaler_offset, dst); + +end: + if (format && !MSM_FORMAT_IS_DX(format)) + op_mode |= BIT(14); + + if (format && format->alpha_enable) { + op_mode |= BIT(10); + if (scaler_version == 0x1002) + op_mode |= (scaler3_cfg->alpha_filter_cfg & 0x1) << 30; + else + op_mode |= (scaler3_cfg->alpha_filter_cfg & 0x3) << 29; + } + + DPU_REG_WRITE(c, QSEED3_OP_MODE + scaler_offset, op_mode); +} + +void dpu_hw_csc_setup(struct dpu_hw_blk_reg_map *c, + u32 csc_reg_off, + const struct dpu_csc_cfg *data, bool csc10) +{ + static const u32 matrix_shift = 7; + u32 clamp_shift = csc10 ? 16 : 8; + u32 val; + + /* matrix coeff - convert S15.16 to S4.9 */ + val = ((data->csc_mv[0] >> matrix_shift) & 0x1FFF) | + (((data->csc_mv[1] >> matrix_shift) & 0x1FFF) << 16); + DPU_REG_WRITE(c, csc_reg_off, val); + val = ((data->csc_mv[2] >> matrix_shift) & 0x1FFF) | + (((data->csc_mv[3] >> matrix_shift) & 0x1FFF) << 16); + DPU_REG_WRITE(c, csc_reg_off + 0x4, val); + val = ((data->csc_mv[4] >> matrix_shift) & 0x1FFF) | + (((data->csc_mv[5] >> matrix_shift) & 0x1FFF) << 16); + DPU_REG_WRITE(c, csc_reg_off + 0x8, val); + val = ((data->csc_mv[6] >> matrix_shift) & 0x1FFF) | + (((data->csc_mv[7] >> matrix_shift) & 0x1FFF) << 16); + DPU_REG_WRITE(c, csc_reg_off + 0xc, val); + val = (data->csc_mv[8] >> matrix_shift) & 0x1FFF; + DPU_REG_WRITE(c, csc_reg_off + 0x10, val); + + /* Pre clamp */ + val = (data->csc_pre_lv[0] << clamp_shift) | data->csc_pre_lv[1]; + DPU_REG_WRITE(c, csc_reg_off + 0x14, val); + val = (data->csc_pre_lv[2] << clamp_shift) | data->csc_pre_lv[3]; + DPU_REG_WRITE(c, csc_reg_off + 0x18, val); + val = (data->csc_pre_lv[4] << clamp_shift) | data->csc_pre_lv[5]; + DPU_REG_WRITE(c, csc_reg_off + 0x1c, val); + + /* Post clamp */ + val = (data->csc_post_lv[0] << clamp_shift) | data->csc_post_lv[1]; + DPU_REG_WRITE(c, csc_reg_off + 0x20, val); + val = (data->csc_post_lv[2] << clamp_shift) | data->csc_post_lv[3]; + DPU_REG_WRITE(c, csc_reg_off + 0x24, val); + val = (data->csc_post_lv[4] << clamp_shift) | data->csc_post_lv[5]; + DPU_REG_WRITE(c, csc_reg_off + 0x28, val); + + /* Pre-Bias */ + DPU_REG_WRITE(c, csc_reg_off + 0x2c, data->csc_pre_bv[0]); + DPU_REG_WRITE(c, csc_reg_off + 0x30, data->csc_pre_bv[1]); + DPU_REG_WRITE(c, csc_reg_off + 0x34, data->csc_pre_bv[2]); + + /* Post-Bias */ + DPU_REG_WRITE(c, csc_reg_off + 0x38, data->csc_post_bv[0]); + DPU_REG_WRITE(c, csc_reg_off + 0x3c, data->csc_post_bv[1]); + DPU_REG_WRITE(c, csc_reg_off + 0x40, data->csc_post_bv[2]); +} + +/** + * _dpu_hw_get_qos_lut - get LUT mapping based on fill level + * @tbl: Pointer to LUT table + * @total_fl: fill level + * Return: LUT setting corresponding to the fill level + */ +u64 _dpu_hw_get_qos_lut(const struct dpu_qos_lut_tbl *tbl, + u32 total_fl) +{ + int i; + + if (!tbl || !tbl->nentry || !tbl->entries) + return 0; + + for (i = 0; i < tbl->nentry; i++) + if (total_fl <= tbl->entries[i].fl) + return tbl->entries[i].lut; + + /* if last fl is zero, use as default */ + if (!tbl->entries[i-1].fl) + return tbl->entries[i-1].lut; + + return 0; +} + +void _dpu_hw_setup_qos_lut(struct dpu_hw_blk_reg_map *c, u32 offset, + bool qos_8lvl, + const struct dpu_hw_qos_cfg *cfg) +{ + DPU_REG_WRITE(c, offset + QOS_DANGER_LUT, cfg->danger_lut); + DPU_REG_WRITE(c, offset + QOS_SAFE_LUT, cfg->safe_lut); + + if (qos_8lvl) { + DPU_REG_WRITE(c, offset + QOS_CREQ_LUT_0, cfg->creq_lut); + DPU_REG_WRITE(c, offset + QOS_CREQ_LUT_1, cfg->creq_lut >> 32); + } else { + DPU_REG_WRITE(c, offset + QOS_CREQ_LUT, cfg->creq_lut); + } + + DPU_REG_WRITE(c, offset + QOS_QOS_CTRL, + cfg->danger_safe_en ? QOS_QOS_CTRL_DANGER_SAFE_EN : 0); +} + +void dpu_hw_setup_qos_lut_v13(struct dpu_hw_blk_reg_map *c, + const struct dpu_hw_qos_cfg *cfg) +{ + DPU_REG_WRITE(c, SSPP_CMN_DANGER_LUT, cfg->danger_lut); + DPU_REG_WRITE(c, SSPP_CMN_SAFE_LUT, cfg->safe_lut); + DPU_REG_WRITE(c, SSPP_CMN_CREQ_LUT_0, cfg->creq_lut); + DPU_REG_WRITE(c, SSPP_CMN_CREQ_LUT_1, cfg->creq_lut >> 32); + DPU_REG_WRITE(c, SSPP_CMN_QOS_CTRL, + cfg->danger_safe_en ? QOS_QOS_CTRL_DANGER_SAFE_EN : 0); +} + +/* + * note: Aside from encoders, input_sel should be set to 0x0 by default + */ +void dpu_hw_setup_misr(struct dpu_hw_blk_reg_map *c, + u32 misr_ctrl_offset, u8 input_sel) +{ + u32 config = 0; + + DPU_REG_WRITE(c, misr_ctrl_offset, MISR_CTRL_STATUS_CLEAR); + + /* Clear old MISR value (in case it's read before a new value is calculated)*/ + wmb(); + + config = MISR_FRAME_COUNT | MISR_CTRL_ENABLE | MISR_CTRL_FREE_RUN_MASK | + ((input_sel & 0xF) << 24); + DPU_REG_WRITE(c, misr_ctrl_offset, config); +} + +int dpu_hw_collect_misr(struct dpu_hw_blk_reg_map *c, + u32 misr_ctrl_offset, + u32 misr_signature_offset, + u32 *misr_value) +{ + u32 ctrl = 0; + + if (!misr_value) + return -EINVAL; + + ctrl = DPU_REG_READ(c, misr_ctrl_offset); + + if (!(ctrl & MISR_CTRL_ENABLE)) + return -ENODATA; + + if (!(ctrl & MISR_CTRL_STATUS)) + return -EINVAL; + + *misr_value = DPU_REG_READ(c, misr_signature_offset); + + return 0; +} + +#define CDP_ENABLE BIT(0) +#define CDP_UBWC_META_ENABLE BIT(1) +#define CDP_TILE_AMORTIZE_ENABLE BIT(2) +#define CDP_PRELOAD_AHEAD_64 BIT(3) + +void dpu_setup_cdp(struct dpu_hw_blk_reg_map *c, u32 offset, + const struct msm_format *fmt, bool enable) +{ + u32 cdp_cntl = CDP_PRELOAD_AHEAD_64; + + if (enable) + cdp_cntl |= CDP_ENABLE; + if (MSM_FORMAT_IS_UBWC(fmt)) + cdp_cntl |= CDP_UBWC_META_ENABLE; + if (MSM_FORMAT_IS_UBWC(fmt) || + MSM_FORMAT_IS_TILE(fmt)) + cdp_cntl |= CDP_TILE_AMORTIZE_ENABLE; + + DPU_REG_WRITE(c, offset, cdp_cntl); +} + +bool dpu_hw_clk_force_ctrl(struct dpu_hw_blk_reg_map *c, + const struct dpu_clk_ctrl_reg *clk_ctrl_reg, + bool enable) +{ + u32 reg_val, new_val; + bool clk_forced_on; + + reg_val = DPU_REG_READ(c, clk_ctrl_reg->reg_off); + + if (enable) + new_val = reg_val | BIT(clk_ctrl_reg->bit_off); + else + new_val = reg_val & ~BIT(clk_ctrl_reg->bit_off); + + DPU_REG_WRITE(c, clk_ctrl_reg->reg_off, new_val); + + clk_forced_on = !(reg_val & BIT(clk_ctrl_reg->bit_off)); + + return clk_forced_on; +} + +#define TO_S15D16(_x_)((_x_) << 7) + +const struct dpu_csc_cfg dpu_csc_YUV2RGB_601L = { + { + /* S15.16 format */ + 0x00012A00, 0x00000000, 0x00019880, + 0x00012A00, 0xFFFF9B80, 0xFFFF3000, + 0x00012A00, 0x00020480, 0x00000000, + }, + /* signed bias */ + { 0xfff0, 0xff80, 0xff80,}, + { 0x0, 0x0, 0x0,}, + /* unsigned clamp */ + { 0x10, 0xeb, 0x10, 0xf0, 0x10, 0xf0,}, + { 0x00, 0xff, 0x00, 0xff, 0x00, 0xff,}, +}; + +const struct dpu_csc_cfg dpu_csc10_YUV2RGB_601L = { + { + /* S15.16 format */ + 0x00012A00, 0x00000000, 0x00019880, + 0x00012A00, 0xFFFF9B80, 0xFFFF3000, + 0x00012A00, 0x00020480, 0x00000000, + }, + /* signed bias */ + { 0xffc0, 0xfe00, 0xfe00,}, + { 0x0, 0x0, 0x0,}, + /* unsigned clamp */ + { 0x40, 0x3ac, 0x40, 0x3c0, 0x40, 0x3c0,}, + { 0x00, 0x3ff, 0x00, 0x3ff, 0x00, 0x3ff,}, +}; + +const struct dpu_csc_cfg dpu_csc10_rgb2yuv_601l = { + { + TO_S15D16(0x0083), TO_S15D16(0x0102), TO_S15D16(0x0032), + TO_S15D16(0x1fb5), TO_S15D16(0x1f6c), TO_S15D16(0x00e1), + TO_S15D16(0x00e1), TO_S15D16(0x1f45), TO_S15D16(0x1fdc) + }, + { 0x00, 0x00, 0x00 }, + { 0x0040, 0x0200, 0x0200 }, + { 0x000, 0x3ff, 0x000, 0x3ff, 0x000, 0x3ff }, + { 0x040, 0x3ac, 0x040, 0x3c0, 0x040, 0x3c0 }, +}; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h new file mode 100644 index 000000000..628befc65 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h @@ -0,0 +1,385 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022-2023 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2021, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_HW_UTIL_H +#define _DPU_HW_UTIL_H + +#include +#include +#include "dpu_hw_mdss.h" +#include "dpu_hw_catalog.h" + +#define REG_MASK(n) ((BIT(n)) - 1) +#define MISR_FRAME_COUNT 0x1 +#define MISR_CTRL_ENABLE BIT(8) +#define MISR_CTRL_STATUS BIT(9) +#define MISR_CTRL_STATUS_CLEAR BIT(10) +#define MISR_CTRL_FREE_RUN_MASK BIT(31) + +#define TO_S15D16(_x_)((_x_) << 7) + +#define MDP_TICK_COUNT 16 +#define XO_CLK_RATE 19200 +#define MS_TICKS_IN_SEC 1000 + +#define CALCULATE_WD_LOAD_VALUE(fps) \ + ((uint32_t)((MS_TICKS_IN_SEC * XO_CLK_RATE)/(MDP_TICK_COUNT * fps))) + +extern const struct dpu_csc_cfg dpu_csc_YUV2RGB_601L; +extern const struct dpu_csc_cfg dpu_csc10_YUV2RGB_601L; +extern const struct dpu_csc_cfg dpu_csc10_rgb2yuv_601l; + +/* + * This is the common struct maintained by each sub block + * for mapping the register offsets in this block to the + * absoulute IO address + * @blk_addr: hw block register mapped address + * @log_mask: log mask for this block + */ +struct dpu_hw_blk_reg_map { + void __iomem *blk_addr; + u32 log_mask; +}; + +/** + * struct dpu_hw_blk - opaque hardware block object + */ +struct dpu_hw_blk { + /* opaque */ +}; + +/** + * struct dpu_hw_scaler3_de_cfg : QSEEDv3 detail enhancer configuration + * @enable: detail enhancer enable/disable + * @sharpen_level1: sharpening strength for noise + * @sharpen_level2: sharpening strength for signal + * @ clip: clip shift + * @ limit: limit value + * @ thr_quiet: quiet threshold + * @ thr_dieout: dieout threshold + * @ thr_high: low threshold + * @ thr_high: high threshold + * @ prec_shift: precision shift + * @ adjust_a: A-coefficients for mapping curve + * @ adjust_b: B-coefficients for mapping curve + * @ adjust_c: C-coefficients for mapping curve + */ +struct dpu_hw_scaler3_de_cfg { + u32 enable; + int16_t sharpen_level1; + int16_t sharpen_level2; + uint16_t clip; + uint16_t limit; + uint16_t thr_quiet; + uint16_t thr_dieout; + uint16_t thr_low; + uint16_t thr_high; + uint16_t prec_shift; + int16_t adjust_a[DPU_MAX_DE_CURVES]; + int16_t adjust_b[DPU_MAX_DE_CURVES]; + int16_t adjust_c[DPU_MAX_DE_CURVES]; +}; + + +/** + * struct dpu_hw_scaler3_cfg : QSEEDv3 configuration + * @enable: scaler enable + * @dir_en: direction detection block enable + * @ init_phase_x: horizontal initial phase + * @ phase_step_x: horizontal phase step + * @ init_phase_y: vertical initial phase + * @ phase_step_y: vertical phase step + * @ preload_x: horizontal preload value + * @ preload_y: vertical preload value + * @ src_width: source width + * @ src_height: source height + * @ dst_width: destination width + * @ dst_height: destination height + * @ y_rgb_filter_cfg: y/rgb plane filter configuration + * @ uv_filter_cfg: uv plane filter configuration + * @ alpha_filter_cfg: alpha filter configuration + * @ blend_cfg: blend coefficients configuration + * @ lut_flag: scaler LUT update flags + * 0x1 swap LUT bank + * 0x2 update 2D filter LUT + * 0x4 update y circular filter LUT + * 0x8 update uv circular filter LUT + * 0x10 update y separable filter LUT + * 0x20 update uv separable filter LUT + * @ dir_lut_idx: 2D filter LUT index + * @ y_rgb_cir_lut_idx: y circular filter LUT index + * @ uv_cir_lut_idx: uv circular filter LUT index + * @ y_rgb_sep_lut_idx: y circular filter LUT index + * @ uv_sep_lut_idx: uv separable filter LUT index + * @ dir_lut: pointer to 2D LUT + * @ cir_lut: pointer to circular filter LUT + * @ sep_lut: pointer to separable filter LUT + * @ de: detail enhancer configuration + * @ dir_weight: Directional weight + */ +struct dpu_hw_scaler3_cfg { + u32 enable; + u32 dir_en; + int32_t init_phase_x[DPU_MAX_PLANES]; + int32_t phase_step_x[DPU_MAX_PLANES]; + int32_t init_phase_y[DPU_MAX_PLANES]; + int32_t phase_step_y[DPU_MAX_PLANES]; + + u32 preload_x[DPU_MAX_PLANES]; + u32 preload_y[DPU_MAX_PLANES]; + u32 src_width[DPU_MAX_PLANES]; + u32 src_height[DPU_MAX_PLANES]; + + u32 dst_width; + u32 dst_height; + + u32 y_rgb_filter_cfg; + u32 uv_filter_cfg; + u32 alpha_filter_cfg; + u32 blend_cfg; + + u32 lut_flag; + u32 dir_lut_idx; + + u32 y_rgb_cir_lut_idx; + u32 uv_cir_lut_idx; + u32 y_rgb_sep_lut_idx; + u32 uv_sep_lut_idx; + u32 *dir_lut; + size_t dir_len; + u32 *cir_lut; + size_t cir_len; + u32 *sep_lut; + size_t sep_len; + + /* + * Detail enhancer settings + */ + struct dpu_hw_scaler3_de_cfg de; + + u32 dir_weight; +}; + +/** + * struct dpu_drm_pix_ext_v1 - version 1 of pixel ext structure + * @num_ext_pxls_lr: Number of total horizontal pixels + * @num_ext_pxls_tb: Number of total vertical lines + * @left_ftch: Number of extra pixels to overfetch from left + * @right_ftch: Number of extra pixels to overfetch from right + * @top_ftch: Number of extra lines to overfetch from top + * @btm_ftch: Number of extra lines to overfetch from bottom + * @left_rpt: Number of extra pixels to repeat from left + * @right_rpt: Number of extra pixels to repeat from right + * @top_rpt: Number of extra lines to repeat from top + * @btm_rpt: Number of extra lines to repeat from bottom + */ +struct dpu_drm_pix_ext_v1 { + /* + * Number of pixels ext in left, right, top and bottom direction + * for all color components. + */ + int32_t num_ext_pxls_lr[DPU_MAX_PLANES]; + int32_t num_ext_pxls_tb[DPU_MAX_PLANES]; + + /* + * Number of pixels needs to be overfetched in left, right, top + * and bottom directions from source image for scaling. + */ + int32_t left_ftch[DPU_MAX_PLANES]; + int32_t right_ftch[DPU_MAX_PLANES]; + int32_t top_ftch[DPU_MAX_PLANES]; + int32_t btm_ftch[DPU_MAX_PLANES]; + /* + * Number of pixels needs to be repeated in left, right, top and + * bottom directions for scaling. + */ + int32_t left_rpt[DPU_MAX_PLANES]; + int32_t right_rpt[DPU_MAX_PLANES]; + int32_t top_rpt[DPU_MAX_PLANES]; + int32_t btm_rpt[DPU_MAX_PLANES]; + +}; + +/** + * struct dpu_drm_de_v1 - version 1 of detail enhancer structure + * @enable: Enables/disables detail enhancer + * @sharpen_level1: Sharpening strength for noise + * @sharpen_level2: Sharpening strength for context + * @clip: Clip coefficient + * @limit: Detail enhancer limit factor + * @thr_quiet: Quite zone threshold + * @thr_dieout: Die-out zone threshold + * @thr_low: Linear zone left threshold + * @thr_high: Linear zone right threshold + * @prec_shift: Detail enhancer precision + * @adjust_a: Mapping curves A coefficients + * @adjust_b: Mapping curves B coefficients + * @adjust_c: Mapping curves C coefficients + */ +struct dpu_drm_de_v1 { + uint32_t enable; + int16_t sharpen_level1; + int16_t sharpen_level2; + uint16_t clip; + uint16_t limit; + uint16_t thr_quiet; + uint16_t thr_dieout; + uint16_t thr_low; + uint16_t thr_high; + uint16_t prec_shift; + int16_t adjust_a[DPU_MAX_DE_CURVES]; + int16_t adjust_b[DPU_MAX_DE_CURVES]; + int16_t adjust_c[DPU_MAX_DE_CURVES]; +}; + +/** + * struct dpu_drm_scaler_v2 - version 2 of struct dpu_drm_scaler + * @enable: Scaler enable + * @dir_en: Detail enhancer enable + * @pe: Pixel extension settings + * @horz_decimate: Horizontal decimation factor + * @vert_decimate: Vertical decimation factor + * @init_phase_x: Initial scaler phase values for x + * @phase_step_x: Phase step values for x + * @init_phase_y: Initial scaler phase values for y + * @phase_step_y: Phase step values for y + * @preload_x: Horizontal preload value + * @preload_y: Vertical preload value + * @src_width: Source width + * @src_height: Source height + * @dst_width: Destination width + * @dst_height: Destination height + * @y_rgb_filter_cfg: Y/RGB plane filter configuration + * @uv_filter_cfg: UV plane filter configuration + * @alpha_filter_cfg: Alpha filter configuration + * @blend_cfg: Selection of blend coefficients + * @lut_flag: LUT configuration flags + * @dir_lut_idx: 2d 4x4 LUT index + * @y_rgb_cir_lut_idx: Y/RGB circular LUT index + * @uv_cir_lut_idx: UV circular LUT index + * @y_rgb_sep_lut_idx: Y/RGB separable LUT index + * @uv_sep_lut_idx: UV separable LUT index + * @de: Detail enhancer settings + */ +struct dpu_drm_scaler_v2 { + /* + * General definitions + */ + uint32_t enable; + uint32_t dir_en; + + /* + * Pix ext settings + */ + struct dpu_drm_pix_ext_v1 pe; + + /* + * Decimation settings + */ + uint32_t horz_decimate; + uint32_t vert_decimate; + + /* + * Phase settings + */ + int32_t init_phase_x[DPU_MAX_PLANES]; + int32_t phase_step_x[DPU_MAX_PLANES]; + int32_t init_phase_y[DPU_MAX_PLANES]; + int32_t phase_step_y[DPU_MAX_PLANES]; + + uint32_t preload_x[DPU_MAX_PLANES]; + uint32_t preload_y[DPU_MAX_PLANES]; + uint32_t src_width[DPU_MAX_PLANES]; + uint32_t src_height[DPU_MAX_PLANES]; + + uint32_t dst_width; + uint32_t dst_height; + + uint32_t y_rgb_filter_cfg; + uint32_t uv_filter_cfg; + uint32_t alpha_filter_cfg; + uint32_t blend_cfg; + + uint32_t lut_flag; + uint32_t dir_lut_idx; + + /* for Y(RGB) and UV planes*/ + uint32_t y_rgb_cir_lut_idx; + uint32_t uv_cir_lut_idx; + uint32_t y_rgb_sep_lut_idx; + uint32_t uv_sep_lut_idx; + + /* + * Detail enhancer settings + */ + struct dpu_drm_de_v1 de; +}; + +/** + * struct dpu_hw_qos_cfg: pipe QoS configuration + * @danger_lut: LUT for generate danger level based on fill level + * @safe_lut: LUT for generate safe level based on fill level + * @creq_lut: LUT for generate creq level based on fill level + * @creq_vblank: creq value generated to vbif during vertical blanking + * @danger_vblank: danger value generated during vertical blanking + * @vblank_en: enable creq_vblank and danger_vblank during vblank + * @danger_safe_en: enable danger safe generation + */ +struct dpu_hw_qos_cfg { + u32 danger_lut; + u32 safe_lut; + u64 creq_lut; + bool danger_safe_en; +}; + +u32 *dpu_hw_util_get_log_mask_ptr(void); + +void dpu_reg_write(struct dpu_hw_blk_reg_map *c, + u32 reg_off, + u32 val, + const char *name); +int dpu_reg_read(struct dpu_hw_blk_reg_map *c, u32 reg_off); + +#define DPU_REG_WRITE(c, off, val) dpu_reg_write(c, off, val, #off) +#define DPU_REG_READ(c, off) dpu_reg_read(c, off) + +void *dpu_hw_util_get_dir(void); + +void dpu_hw_setup_scaler3(struct dpu_hw_blk_reg_map *c, + struct dpu_hw_scaler3_cfg *scaler3_cfg, + u32 scaler_offset, u32 scaler_version, + const struct msm_format *format); + +void dpu_hw_csc_setup(struct dpu_hw_blk_reg_map *c, + u32 csc_reg_off, + const struct dpu_csc_cfg *data, bool csc10); + +void dpu_setup_cdp(struct dpu_hw_blk_reg_map *c, u32 offset, + const struct msm_format *fmt, bool enable); + +u64 _dpu_hw_get_qos_lut(const struct dpu_qos_lut_tbl *tbl, + u32 total_fl); + +void _dpu_hw_setup_qos_lut(struct dpu_hw_blk_reg_map *c, u32 offset, + bool qos_8lvl, + const struct dpu_hw_qos_cfg *cfg); + +void dpu_hw_setup_qos_lut_v13(struct dpu_hw_blk_reg_map *c, + const struct dpu_hw_qos_cfg *cfg); + +void dpu_hw_setup_misr(struct dpu_hw_blk_reg_map *c, + u32 misr_ctrl_offset, u8 input_sel); + +int dpu_hw_collect_misr(struct dpu_hw_blk_reg_map *c, + u32 misr_ctrl_offset, + u32 misr_signature_offset, + u32 *misr_value); + +bool dpu_hw_clk_force_ctrl(struct dpu_hw_blk_reg_map *c, + const struct dpu_clk_ctrl_reg *clk_ctrl_reg, + bool enable); + +#endif /* _DPU_HW_UTIL_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c new file mode 100644 index 000000000..112df3f31 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c @@ -0,0 +1,245 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#include + +#include "dpu_hwio.h" +#include "dpu_hw_catalog.h" +#include "dpu_hw_vbif.h" + +#define VBIF_VERSION 0x0000 +#define VBIF_CLK_FORCE_CTRL0 0x0008 +#define VBIF_CLK_FORCE_CTRL1 0x000C +#define VBIF_QOS_REMAP_00 0x0020 +#define VBIF_QOS_REMAP_01 0x0024 +#define VBIF_QOS_REMAP_10 0x0028 +#define VBIF_QOS_REMAP_11 0x002C +#define VBIF_WRITE_GATHER_EN 0x00AC +#define VBIF_IN_RD_LIM_CONF0 0x00B0 +#define VBIF_IN_RD_LIM_CONF1 0x00B4 +#define VBIF_IN_RD_LIM_CONF2 0x00B8 +#define VBIF_IN_WR_LIM_CONF0 0x00C0 +#define VBIF_IN_WR_LIM_CONF1 0x00C4 +#define VBIF_IN_WR_LIM_CONF2 0x00C8 +#define VBIF_OUT_RD_LIM_CONF0 0x00D0 +#define VBIF_OUT_WR_LIM_CONF0 0x00D4 +#define VBIF_OUT_AXI_AMEMTYPE_CONF0 0x0160 +#define VBIF_OUT_AXI_AMEMTYPE_CONF1 0x0164 +#define VBIF_XIN_PND_ERR 0x0190 +#define VBIF_XIN_SRC_ERR 0x0194 +#define VBIF_XIN_CLR_ERR 0x019C +#define VBIF_XIN_HALT_CTRL0 0x0200 +#define VBIF_XIN_HALT_CTRL1 0x0204 +#define VBIF_XINL_QOS_RP_REMAP_000 0x0550 +#define VBIF_XINL_QOS_LVL_REMAP_000(vbif) (VBIF_XINL_QOS_RP_REMAP_000 + (vbif)->cap->qos_rp_remap_size) + +static void dpu_hw_clear_errors(struct dpu_hw_vbif *vbif, + u32 *pnd_errors, u32 *src_errors) +{ + struct dpu_hw_blk_reg_map *c; + u32 pnd, src; + + if (!vbif) + return; + c = &vbif->hw; + pnd = DPU_REG_READ(c, VBIF_XIN_PND_ERR); + src = DPU_REG_READ(c, VBIF_XIN_SRC_ERR); + + if (pnd_errors) + *pnd_errors = pnd; + if (src_errors) + *src_errors = src; + + DPU_REG_WRITE(c, VBIF_XIN_CLR_ERR, pnd | src); +} + +static void dpu_hw_set_mem_type(struct dpu_hw_vbif *vbif, + u32 xin_id, u32 value) +{ + struct dpu_hw_blk_reg_map *c; + u32 reg_off; + u32 bit_off; + u32 reg_val; + + /* + * Assume 4 bits per bit field, 8 fields per 32-bit register so + * 16 bit fields maximum across two registers + */ + if (!vbif || xin_id >= MAX_XIN_COUNT || xin_id >= 16) + return; + + c = &vbif->hw; + + if (xin_id >= 8) { + xin_id -= 8; + reg_off = VBIF_OUT_AXI_AMEMTYPE_CONF1; + } else { + reg_off = VBIF_OUT_AXI_AMEMTYPE_CONF0; + } + bit_off = (xin_id & 0x7) * 4; + reg_val = DPU_REG_READ(c, reg_off); + reg_val &= ~(0x7 << bit_off); + reg_val |= (value & 0x7) << bit_off; + DPU_REG_WRITE(c, reg_off, reg_val); +} + +static void dpu_hw_set_limit_conf(struct dpu_hw_vbif *vbif, + u32 xin_id, bool rd, u32 limit) +{ + struct dpu_hw_blk_reg_map *c = &vbif->hw; + u32 reg_val; + u32 reg_off; + u32 bit_off; + + if (rd) + reg_off = VBIF_IN_RD_LIM_CONF0; + else + reg_off = VBIF_IN_WR_LIM_CONF0; + + reg_off += (xin_id / 4) * 4; + bit_off = (xin_id % 4) * 8; + reg_val = DPU_REG_READ(c, reg_off); + reg_val &= ~(0xFF << bit_off); + reg_val |= (limit) << bit_off; + DPU_REG_WRITE(c, reg_off, reg_val); +} + +static u32 dpu_hw_get_limit_conf(struct dpu_hw_vbif *vbif, + u32 xin_id, bool rd) +{ + struct dpu_hw_blk_reg_map *c = &vbif->hw; + u32 reg_val; + u32 reg_off; + u32 bit_off; + u32 limit; + + if (rd) + reg_off = VBIF_IN_RD_LIM_CONF0; + else + reg_off = VBIF_IN_WR_LIM_CONF0; + + reg_off += (xin_id / 4) * 4; + bit_off = (xin_id % 4) * 8; + reg_val = DPU_REG_READ(c, reg_off); + limit = (reg_val >> bit_off) & 0xFF; + + return limit; +} + +static void dpu_hw_set_halt_ctrl(struct dpu_hw_vbif *vbif, + u32 xin_id, bool enable) +{ + struct dpu_hw_blk_reg_map *c = &vbif->hw; + u32 reg_val; + + reg_val = DPU_REG_READ(c, VBIF_XIN_HALT_CTRL0); + + if (enable) + reg_val |= BIT(xin_id); + else + reg_val &= ~BIT(xin_id); + + DPU_REG_WRITE(c, VBIF_XIN_HALT_CTRL0, reg_val); +} + +static bool dpu_hw_get_halt_ctrl(struct dpu_hw_vbif *vbif, + u32 xin_id) +{ + struct dpu_hw_blk_reg_map *c = &vbif->hw; + u32 reg_val; + + reg_val = DPU_REG_READ(c, VBIF_XIN_HALT_CTRL1); + + return (reg_val & BIT(xin_id)) ? true : false; +} + +static void dpu_hw_set_qos_remap(struct dpu_hw_vbif *vbif, + u32 xin_id, u32 level, u32 remap_level) +{ + struct dpu_hw_blk_reg_map *c; + u32 reg_lvl, reg_val, reg_val_lvl, mask, reg_high, reg_shift; + + if (!vbif) + return; + + c = &vbif->hw; + + reg_lvl = VBIF_XINL_QOS_LVL_REMAP_000(vbif); + reg_high = ((xin_id & 0x8) >> 3) * 4 + (level * 8); + reg_shift = (xin_id & 0x7) * 4; + + reg_val = DPU_REG_READ(c, VBIF_XINL_QOS_RP_REMAP_000 + reg_high); + reg_val_lvl = DPU_REG_READ(c, reg_lvl + reg_high); + + mask = 0x7 << reg_shift; + + reg_val &= ~mask; + reg_val |= (remap_level << reg_shift) & mask; + + reg_val_lvl &= ~mask; + reg_val_lvl |= (remap_level << reg_shift) & mask; + + DPU_REG_WRITE(c, VBIF_XINL_QOS_RP_REMAP_000 + reg_high, reg_val); + DPU_REG_WRITE(c, reg_lvl + reg_high, reg_val_lvl); +} + +static void dpu_hw_set_write_gather_en(struct dpu_hw_vbif *vbif, u32 xin_id) +{ + struct dpu_hw_blk_reg_map *c; + u32 reg_val; + + if (!vbif || xin_id >= MAX_XIN_COUNT) + return; + + c = &vbif->hw; + + reg_val = DPU_REG_READ(c, VBIF_WRITE_GATHER_EN); + reg_val |= BIT(xin_id); + DPU_REG_WRITE(c, VBIF_WRITE_GATHER_EN, reg_val); +} + +static void _setup_vbif_ops(struct dpu_hw_vbif_ops *ops, + unsigned long cap) +{ + ops->set_limit_conf = dpu_hw_set_limit_conf; + ops->get_limit_conf = dpu_hw_get_limit_conf; + ops->set_halt_ctrl = dpu_hw_set_halt_ctrl; + ops->get_halt_ctrl = dpu_hw_get_halt_ctrl; + if (test_bit(DPU_VBIF_QOS_REMAP, &cap)) + ops->set_qos_remap = dpu_hw_set_qos_remap; + ops->set_mem_type = dpu_hw_set_mem_type; + ops->clear_errors = dpu_hw_clear_errors; + ops->set_write_gather_en = dpu_hw_set_write_gather_en; +} + +/** + * dpu_hw_vbif_init() - Initializes the VBIF driver for the passed + * VBIF catalog entry. + * @dev: Corresponding device for devres management + * @cfg: VBIF catalog entry for which driver object is required + * @addr: Mapped register io address of MDSS + */ +struct dpu_hw_vbif *dpu_hw_vbif_init(struct drm_device *dev, + const struct dpu_vbif_cfg *cfg, + void __iomem *addr) +{ + struct dpu_hw_vbif *c; + + c = drmm_kzalloc(dev, sizeof(*c), GFP_KERNEL); + if (!c) + return ERR_PTR(-ENOMEM); + + c->hw.blk_addr = addr; + c->hw.log_mask = DPU_DBG_MASK_VBIF; + + /* + * Assign ops + */ + c->cap = cfg; + _setup_vbif_ops(&c->ops, c->cap->features); + + /* no need to register sub-range in dpu dbg, dump entire vbif io base */ + + return c; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.h new file mode 100644 index 000000000..96ec4e35e --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.h @@ -0,0 +1,111 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_HW_VBIF_H +#define _DPU_HW_VBIF_H + +#include "dpu_hw_catalog.h" +#include "dpu_hw_mdss.h" +#include "dpu_hw_util.h" + +struct dpu_hw_vbif; + +/** + * struct dpu_hw_vbif_ops : Interface to the VBIF hardware driver functions + * Assumption is these functions will be called after clocks are enabled + */ +struct dpu_hw_vbif_ops { + /** + * @set_limit_conf: set transaction limit config + * @vbif: vbif context driver + * @xin_id: client interface identifier + * @rd: true for read limit; false for write limit + * @limit: outstanding transaction limit + */ + void (*set_limit_conf)(struct dpu_hw_vbif *vbif, + u32 xin_id, bool rd, u32 limit); + + /** + * @get_limit_conf: get transaction limit config + * @vbif: vbif context driver + * @xin_id: client interface identifier + * @rd: true for read limit; false for write limit + * @return: outstanding transaction limit + */ + u32 (*get_limit_conf)(struct dpu_hw_vbif *vbif, + u32 xin_id, bool rd); + + /** + * @set_halt_ctrl: set halt control + * @vbif: vbif context driver + * @xin_id: client interface identifier + * @enable: halt control enable + */ + void (*set_halt_ctrl)(struct dpu_hw_vbif *vbif, + u32 xin_id, bool enable); + + /** + * @get_halt_ctrl: get halt control + * @vbif: vbif context driver + * @xin_id: client interface identifier + * @return: halt control enable + */ + bool (*get_halt_ctrl)(struct dpu_hw_vbif *vbif, + u32 xin_id); + + /** + * @set_qos_remap: set QoS priority remap + * @vbif: vbif context driver + * @xin_id: client interface identifier + * @level: priority level + * @remap_level: remapped level + */ + void (*set_qos_remap)(struct dpu_hw_vbif *vbif, + u32 xin_id, u32 level, u32 remap_level); + + /** + * @set_mem_type: set memory type + * @vbif: vbif context driver + * @xin_id: client interface identifier + * @value: memory type value + */ + void (*set_mem_type)(struct dpu_hw_vbif *vbif, + u32 xin_id, u32 value); + + /** + * @clear_errors: clear any vbif errors + * This function clears any detected pending/source errors + * on the VBIF interface, and optionally returns the detected + * error mask(s). + * @vbif: vbif context driver + * @pnd_errors: pointer to pending error reporting variable + * @src_errors: pointer to source error reporting variable + */ + void (*clear_errors)(struct dpu_hw_vbif *vbif, + u32 *pnd_errors, u32 *src_errors); + + /** + * @set_write_gather_en: set write_gather enable + * @vbif: vbif context driver + * @xin_id: client interface identifier + */ + void (*set_write_gather_en)(struct dpu_hw_vbif *vbif, u32 xin_id); +}; + +struct dpu_hw_vbif { + /* base */ + struct dpu_hw_blk_reg_map hw; + + /* vbif */ + const struct dpu_vbif_cfg *cap; + + /* ops */ + struct dpu_hw_vbif_ops ops; +}; + +struct dpu_hw_vbif *dpu_hw_vbif_init(struct drm_device *dev, + const struct dpu_vbif_cfg *cfg, + void __iomem *addr); + +#endif /*_DPU_HW_VBIF_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c new file mode 100644 index 000000000..4da4bd6a9 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c @@ -0,0 +1,262 @@ +// SPDX-License-Identifier: GPL-2.0-only + /* + * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved + */ + +#include + +#include "dpu_hw_mdss.h" +#include "dpu_hwio.h" +#include "dpu_hw_catalog.h" +#include "dpu_hw_wb.h" +#include "dpu_formats.h" +#include "dpu_kms.h" + +#define WB_DST_FORMAT 0x000 +#define WB_DST_OP_MODE 0x004 +#define WB_DST_PACK_PATTERN 0x008 +#define WB_DST0_ADDR 0x00C +#define WB_DST1_ADDR 0x010 +#define WB_DST2_ADDR 0x014 +#define WB_DST3_ADDR 0x018 +#define WB_DST_YSTRIDE0 0x01C +#define WB_DST_YSTRIDE1 0x020 +#define WB_DST_YSTRIDE1 0x020 +#define WB_DST_DITHER_BITDEPTH 0x024 +#define WB_DST_MATRIX_ROW0 0x030 +#define WB_DST_MATRIX_ROW1 0x034 +#define WB_DST_MATRIX_ROW2 0x038 +#define WB_DST_MATRIX_ROW3 0x03C +#define WB_DST_WRITE_CONFIG 0x048 +#define WB_ROTATION_DNSCALER 0x050 +#define WB_ROTATOR_PIPE_DOWNSCALER 0x054 +#define WB_N16_INIT_PHASE_X_C03 0x060 +#define WB_N16_INIT_PHASE_X_C12 0x064 +#define WB_N16_INIT_PHASE_Y_C03 0x068 +#define WB_N16_INIT_PHASE_Y_C12 0x06C +#define WB_OUT_SIZE 0x074 +#define WB_ALPHA_X_VALUE 0x078 +#define WB_DANGER_LUT 0x084 +#define WB_SAFE_LUT 0x088 +#define WB_QOS_CTRL 0x090 +#define WB_CREQ_LUT_0 0x098 +#define WB_CREQ_LUT_1 0x09C +#define WB_UBWC_STATIC_CTRL 0x144 +#define WB_MUX 0x150 +#define WB_CROP_CTRL 0x154 +#define WB_CROP_OFFSET 0x158 +#define WB_CLK_CTRL 0x178 +#define WB_CSC_BASE 0x260 +#define WB_DST_ADDR_SW_STATUS 0x2B0 +#define WB_CDP_CNTL 0x2B4 +#define WB_OUT_IMAGE_SIZE 0x2C0 +#define WB_OUT_XY 0x2C4 + +static void dpu_hw_wb_setup_outaddress(struct dpu_hw_wb *ctx, + struct dpu_hw_wb_cfg *data) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + + DPU_REG_WRITE(c, WB_DST0_ADDR, data->dest.plane_addr[0]); + DPU_REG_WRITE(c, WB_DST1_ADDR, data->dest.plane_addr[1]); + DPU_REG_WRITE(c, WB_DST2_ADDR, data->dest.plane_addr[2]); + DPU_REG_WRITE(c, WB_DST3_ADDR, data->dest.plane_addr[3]); +} + +static void dpu_hw_wb_setup_format(struct dpu_hw_wb *ctx, + struct dpu_hw_wb_cfg *data, + const struct msm_format *fmt) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + u32 dst_format, pattern, ystride0, ystride1, outsize, chroma_samp; + u32 write_config = 0; + u32 opmode = 0; + u32 dst_addr_sw = 0; + + chroma_samp = fmt->chroma_sample; + + dst_format = (chroma_samp << 23) | + (fmt->fetch_type << 19) | + (fmt->bpc_a << 6) | + (fmt->bpc_r_cr << 4) | + (fmt->bpc_b_cb << 2) | + (fmt->bpc_g_y << 0); + + if (fmt->bpc_a || fmt->alpha_enable) { + dst_format |= BIT(8); /* DSTC3_EN */ + if (!fmt->alpha_enable || + !(ctx->caps->features & BIT(DPU_WB_PIPE_ALPHA))) + dst_format |= BIT(14); /* DST_ALPHA_X */ + } + + if (MSM_FORMAT_IS_YUV(fmt)) + dst_format |= BIT(15); + + pattern = (fmt->element[3] << 24) | + (fmt->element[2] << 16) | + (fmt->element[1] << 8) | + (fmt->element[0] << 0); + + dst_format |= ((fmt->flags & MSM_FORMAT_FLAG_UNPACK_ALIGN_MSB ? 1 : 0) << 18) | + ((fmt->flags & MSM_FORMAT_FLAG_UNPACK_TIGHT ? 1 : 0) << 17) | + ((fmt->unpack_count - 1) << 12) | + ((fmt->bpp - 1) << 9); + + ystride0 = data->dest.plane_pitch[0] | + (data->dest.plane_pitch[1] << 16); + ystride1 = data->dest.plane_pitch[2] | + (data->dest.plane_pitch[3] << 16); + + if (drm_rect_height(&data->roi) && drm_rect_width(&data->roi)) + outsize = (drm_rect_height(&data->roi) << 16) | drm_rect_width(&data->roi); + else + outsize = (data->dest.height << 16) | data->dest.width; + + DPU_REG_WRITE(c, WB_ALPHA_X_VALUE, 0xFF); + DPU_REG_WRITE(c, WB_DST_FORMAT, dst_format); + DPU_REG_WRITE(c, WB_DST_OP_MODE, opmode); + DPU_REG_WRITE(c, WB_DST_PACK_PATTERN, pattern); + DPU_REG_WRITE(c, WB_DST_YSTRIDE0, ystride0); + DPU_REG_WRITE(c, WB_DST_YSTRIDE1, ystride1); + DPU_REG_WRITE(c, WB_OUT_SIZE, outsize); + DPU_REG_WRITE(c, WB_DST_WRITE_CONFIG, write_config); + DPU_REG_WRITE(c, WB_DST_ADDR_SW_STATUS, dst_addr_sw); +} + +static void dpu_hw_wb_roi(struct dpu_hw_wb *ctx, struct dpu_hw_wb_cfg *wb) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + u32 image_size, out_size, out_xy; + + image_size = (wb->dest.height << 16) | wb->dest.width; + out_xy = 0; + out_size = (drm_rect_height(&wb->roi) << 16) | drm_rect_width(&wb->roi); + + DPU_REG_WRITE(c, WB_OUT_IMAGE_SIZE, image_size); + DPU_REG_WRITE(c, WB_OUT_XY, out_xy); + DPU_REG_WRITE(c, WB_OUT_SIZE, out_size); +} + +static void dpu_hw_wb_setup_qos_lut(struct dpu_hw_wb *ctx, + struct dpu_hw_qos_cfg *cfg) +{ + if (!ctx || !cfg) + return; + + _dpu_hw_setup_qos_lut(&ctx->hw, WB_DANGER_LUT, + test_bit(DPU_WB_QOS_8LVL, &ctx->caps->features), + cfg); +} + +static void dpu_hw_wb_setup_qos_lut_v13(struct dpu_hw_wb *ctx, + struct dpu_hw_qos_cfg *cfg) +{ + if (!ctx || !cfg) + return; + + dpu_hw_setup_qos_lut_v13(&ctx->hw, cfg); +} + +static void dpu_hw_wb_setup_cdp(struct dpu_hw_wb *ctx, + const struct msm_format *fmt, + bool enable) +{ + if (!ctx) + return; + + dpu_setup_cdp(&ctx->hw, WB_CDP_CNTL, fmt, enable); +} + +static void dpu_hw_wb_bind_pingpong_blk( + struct dpu_hw_wb *ctx, + const enum dpu_pingpong pp) +{ + struct dpu_hw_blk_reg_map *c; + int mux_cfg; + + if (!ctx) + return; + + c = &ctx->hw; + + mux_cfg = DPU_REG_READ(c, WB_MUX); + mux_cfg &= ~0xf; + + if (pp >= PINGPONG_CWB_0) + mux_cfg |= (pp < PINGPONG_CWB_2) ? 0xd : 0xb; + else if (pp) + mux_cfg |= (pp - PINGPONG_0) & 0x7; + else + mux_cfg |= 0xf; + + DPU_REG_WRITE(c, WB_MUX, mux_cfg); +} + +static bool dpu_hw_wb_setup_clk_force_ctrl(struct dpu_hw_wb *ctx, bool enable) +{ + static const struct dpu_clk_ctrl_reg wb_clk_ctrl = { + .reg_off = WB_CLK_CTRL, + .bit_off = 0 + }; + + return dpu_hw_clk_force_ctrl(&ctx->hw, &wb_clk_ctrl, enable); +} + +static void _setup_wb_ops(struct dpu_hw_wb_ops *ops, + unsigned long features, const struct dpu_mdss_version *mdss_rev) +{ + ops->setup_outaddress = dpu_hw_wb_setup_outaddress; + ops->setup_outformat = dpu_hw_wb_setup_format; + + if (test_bit(DPU_WB_XY_ROI_OFFSET, &features)) + ops->setup_roi = dpu_hw_wb_roi; + + if (test_bit(DPU_WB_QOS, &features)) { + if (mdss_rev->core_major_ver >= 13) + ops->setup_qos_lut = dpu_hw_wb_setup_qos_lut_v13; + else + ops->setup_qos_lut = dpu_hw_wb_setup_qos_lut; + } + + if (test_bit(DPU_WB_CDP, &features)) + ops->setup_cdp = dpu_hw_wb_setup_cdp; + + if (mdss_rev->core_major_ver >= 5) + ops->bind_pingpong_blk = dpu_hw_wb_bind_pingpong_blk; + + if (mdss_rev->core_major_ver >= 9) + ops->setup_clk_force_ctrl = dpu_hw_wb_setup_clk_force_ctrl; +} + +/** + * dpu_hw_wb_init() - Initializes the writeback hw driver object. + * @dev: Corresponding device for devres management + * @cfg: wb_path catalog entry for which driver object is required + * @addr: mapped register io address of MDP + * @mdss_rev: dpu core's major and minor versions + * Return: Error code or allocated dpu_hw_wb context + */ +struct dpu_hw_wb *dpu_hw_wb_init(struct drm_device *dev, + const struct dpu_wb_cfg *cfg, + void __iomem *addr, + const struct dpu_mdss_version *mdss_rev) +{ + struct dpu_hw_wb *c; + + if (!addr) + return ERR_PTR(-EINVAL); + + c = drmm_kzalloc(dev, sizeof(*c), GFP_KERNEL); + if (!c) + return ERR_PTR(-ENOMEM); + + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_WB; + + /* Assign ops */ + c->idx = cfg->id; + c->caps = cfg; + _setup_wb_ops(&c->ops, c->caps->features, mdss_rev); + + return c; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h new file mode 100644 index 000000000..cfdbb5bb2 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h @@ -0,0 +1,83 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved + */ + +#ifndef _DPU_HW_WB_H +#define _DPU_HW_WB_H + +#include "dpu_hw_catalog.h" +#include "dpu_hw_mdss.h" +#include "dpu_hw_top.h" +#include "dpu_hw_util.h" +#include "dpu_hw_pingpong.h" + +struct dpu_hw_wb; + +struct dpu_hw_wb_cfg { + struct dpu_hw_fmt_layout dest; + enum dpu_intf_mode intf_mode; + struct drm_rect roi; + struct drm_rect crop; +}; + +/** + * struct dpu_hw_wb_ops : Interface to the wb hw driver functions + * Assumption is these functions will be called after clocks are enabled + * @setup_outaddress: setup output address from the writeback job + * @setup_outformat: setup output format of writeback block from writeback job + * @setup_roi: setup ROI (Region of Interest) parameters + * @setup_qos_lut: setup qos LUT for writeback block based on input + * @setup_cdp: setup chroma down prefetch block for writeback block + * @setup_clk_force_ctrl: setup clock force control + * @bind_pingpong_blk: enable/disable the connection with ping-pong block + */ +struct dpu_hw_wb_ops { + void (*setup_outaddress)(struct dpu_hw_wb *ctx, + struct dpu_hw_wb_cfg *wb); + + void (*setup_outformat)(struct dpu_hw_wb *ctx, + struct dpu_hw_wb_cfg *wb, + const struct msm_format *fmt); + + void (*setup_roi)(struct dpu_hw_wb *ctx, + struct dpu_hw_wb_cfg *wb); + + void (*setup_qos_lut)(struct dpu_hw_wb *ctx, + struct dpu_hw_qos_cfg *cfg); + + void (*setup_cdp)(struct dpu_hw_wb *ctx, + const struct msm_format *fmt, + bool enable); + + bool (*setup_clk_force_ctrl)(struct dpu_hw_wb *ctx, + bool enable); + + void (*bind_pingpong_blk)(struct dpu_hw_wb *ctx, + const enum dpu_pingpong pp); +}; + +/** + * struct dpu_hw_wb : WB driver object + * @hw: block hardware details + * @idx: hardware index number within type + * @caps: hardware capabilities + * @ops: function pointers + */ +struct dpu_hw_wb { + struct dpu_hw_blk_reg_map hw; + + /* wb path */ + int idx; + const struct dpu_wb_cfg *caps; + + /* ops */ + struct dpu_hw_wb_ops ops; +}; + +struct dpu_hw_wb *dpu_hw_wb_init(struct drm_device *dev, + const struct dpu_wb_cfg *cfg, + void __iomem *addr, + const struct dpu_mdss_version *mdss_rev); + +#endif /*_DPU_HW_WB_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hwio.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hwio.h new file mode 100644 index 000000000..054fe097e --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hwio.h @@ -0,0 +1,73 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_HWIO_H +#define _DPU_HWIO_H + +#include "dpu_hw_util.h" + +/** + * MDP TOP block Register and bit fields and defines + */ +#define DISP_INTF_SEL 0x004 +#define INTR_EN 0x010 +#define INTR_STATUS 0x014 +#define INTR_CLEAR 0x018 +#define INTR2_EN 0x008 +#define INTR2_STATUS 0x00c +#define SSPP_SPARE 0x028 +#define INTR2_CLEAR 0x02c +#define HIST_INTR_EN 0x01c +#define HIST_INTR_STATUS 0x020 +#define HIST_INTR_CLEAR 0x024 +#define SPLIT_DISPLAY_EN 0x2F4 +#define SPLIT_DISPLAY_UPPER_PIPE_CTRL 0x2F8 +#define DSPP_IGC_COLOR0_RAM_LUTN 0x300 +#define DSPP_IGC_COLOR1_RAM_LUTN 0x304 +#define DSPP_IGC_COLOR2_RAM_LUTN 0x308 +#define DANGER_STATUS 0x360 +#define SAFE_STATUS 0x364 +#define HW_EVENTS_CTL 0x37C +#define MDP_WD_TIMER_0_CTL 0x380 +#define MDP_WD_TIMER_0_CTL2 0x384 +#define MDP_WD_TIMER_0_LOAD_VALUE 0x388 +#define MDP_WD_TIMER_1_CTL 0x390 +#define MDP_WD_TIMER_1_CTL2 0x394 +#define MDP_WD_TIMER_1_LOAD_VALUE 0x398 +#define CLK_CTRL3 0x3A8 +#define CLK_STATUS3 0x3AC +#define CLK_CTRL4 0x3B0 +#define CLK_STATUS4 0x3B4 +#define CLK_CTRL5 0x3B8 +#define CLK_STATUS5 0x3BC +#define CLK_CTRL7 0x3D0 +#define CLK_STATUS7 0x3D4 +#define SPLIT_DISPLAY_LOWER_PIPE_CTRL 0x3F0 +#define SPLIT_DISPLAY_TE_LINE_INTERVAL 0x3F4 +#define INTF_SW_RESET_MASK 0x3FC +#define HDMI_DP_CORE_SELECT 0x408 +#define MDP_OUT_CTL_0 0x410 +#define MDP_VSYNC_SEL 0x414 +#define MDP_WD_TIMER_2_CTL 0x420 +#define MDP_WD_TIMER_2_CTL2 0x424 +#define MDP_WD_TIMER_2_LOAD_VALUE 0x428 +#define MDP_WD_TIMER_3_CTL 0x430 +#define MDP_WD_TIMER_3_CTL2 0x434 +#define MDP_WD_TIMER_3_LOAD_VALUE 0x438 +#define MDP_WD_TIMER_4_CTL 0x440 +#define MDP_WD_TIMER_4_CTL2 0x444 +#define MDP_WD_TIMER_4_LOAD_VALUE 0x448 +#define DCE_SEL 0x450 + +#define MDP_DP_PHY_INTF_SEL 0x460 +#define MDP_DP_PHY_INTF_SEL_INTF0 GENMASK(2, 0) +#define MDP_DP_PHY_INTF_SEL_INTF1 GENMASK(5, 3) +#define MDP_DP_PHY_INTF_SEL_PHY0 GENMASK(8, 6) +#define MDP_DP_PHY_INTF_SEL_PHY1 GENMASK(11, 9) +#define MDP_DP_PHY_INTF_SEL_PHY2 GENMASK(14, 12) + +#define MDP_PERIPH_TOP0 MDP_WD_TIMER_0_CTL +#define MDP_PERIPH_TOP0_END CLK_CTRL3 + +#endif /*_DPU_HWIO_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c new file mode 100644 index 000000000..da3556eb6 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c @@ -0,0 +1,1540 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013 Red Hat + * Copyright (c) 2014-2018, The Linux Foundation. All rights reserved. + * Copyright (c) 2022-2024 Qualcomm Innovation Center, Inc. All rights reserved. + * + * Author: Rob Clark + */ + +#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +#include + +#include "msm_drv.h" +#include "msm_mmu.h" +#include "msm_gem.h" +#include "disp/msm_disp_snapshot.h" + +#include "dpu_core_irq.h" +#include "dpu_crtc.h" +#include "dpu_encoder.h" +#include "dpu_formats.h" +#include "dpu_hw_vbif.h" +#include "dpu_kms.h" +#include "dpu_plane.h" +#include "dpu_vbif.h" +#include "dpu_writeback.h" + +#define CREATE_TRACE_POINTS +#include "dpu_trace.h" + +/* + * To enable overall DRM driver logging + * # echo 0x2 > /sys/module/drm/parameters/debug + * + * To enable DRM driver h/w logging + * # echo > /sys/kernel/debug/dri/0/debug/hw_log_mask + * + * See dpu_hw_mdss.h for h/w logging mask definitions (search for DPU_DBG_MASK_) + */ +#define DPU_DEBUGFS_DIR "msm_dpu" +#define DPU_DEBUGFS_HWMASKNAME "hw_log_mask" + +bool dpu_use_virtual_planes = true; +module_param(dpu_use_virtual_planes, bool, 0); + +static int dpu_kms_hw_init(struct msm_kms *kms); +static void _dpu_kms_mmu_destroy(struct dpu_kms *dpu_kms); + +#ifdef CONFIG_DEBUG_FS +static int _dpu_danger_signal_status(struct seq_file *s, + bool danger_status) +{ + struct dpu_danger_safe_status status; + struct dpu_kms *kms = s->private; + int i; + + if (!kms->hw_mdp) { + DPU_ERROR("invalid arg(s)\n"); + return 0; + } + + memset(&status, 0, sizeof(struct dpu_danger_safe_status)); + + pm_runtime_get_sync(&kms->pdev->dev); + if (danger_status) { + seq_puts(s, "\nDanger signal status:\n"); + if (kms->hw_mdp->ops.get_danger_status) + kms->hw_mdp->ops.get_danger_status(kms->hw_mdp, + &status); + } else { + seq_puts(s, "\nSafe signal status:\n"); + if (kms->hw_mdp->ops.get_safe_status) + kms->hw_mdp->ops.get_safe_status(kms->hw_mdp, + &status); + } + pm_runtime_put_sync(&kms->pdev->dev); + + seq_printf(s, "MDP : 0x%x\n", status.mdp); + + for (i = SSPP_VIG0; i < SSPP_MAX; i++) + seq_printf(s, "SSPP%d : 0x%x \n", i - SSPP_VIG0, + status.sspp[i]); + seq_puts(s, "\n"); + + return 0; +} + +static int dpu_debugfs_danger_stats_show(struct seq_file *s, void *v) +{ + return _dpu_danger_signal_status(s, true); +} +DEFINE_SHOW_ATTRIBUTE(dpu_debugfs_danger_stats); + +static int dpu_debugfs_safe_stats_show(struct seq_file *s, void *v) +{ + return _dpu_danger_signal_status(s, false); +} +DEFINE_SHOW_ATTRIBUTE(dpu_debugfs_safe_stats); + +static ssize_t _dpu_plane_danger_read(struct file *file, + char __user *buff, size_t count, loff_t *ppos) +{ + struct dpu_kms *kms = file->private_data; + int len; + char buf[40]; + + len = scnprintf(buf, sizeof(buf), "%d\n", !kms->has_danger_ctrl); + + return simple_read_from_buffer(buff, count, ppos, buf, len); +} + +static void _dpu_plane_set_danger_state(struct dpu_kms *kms, bool enable) +{ + struct drm_plane *plane; + + drm_for_each_plane(plane, kms->dev) { + if (plane->fb && plane->state) { + dpu_plane_danger_signal_ctrl(plane, enable); + DPU_DEBUG("plane:%d img:%dx%d ", + plane->base.id, plane->fb->width, + plane->fb->height); + DPU_DEBUG("src[%d,%d,%d,%d] dst[%d,%d,%d,%d]\n", + plane->state->src_x >> 16, + plane->state->src_y >> 16, + plane->state->src_w >> 16, + plane->state->src_h >> 16, + plane->state->crtc_x, plane->state->crtc_y, + plane->state->crtc_w, plane->state->crtc_h); + } else { + DPU_DEBUG("Inactive plane:%d\n", plane->base.id); + } + } +} + +static ssize_t _dpu_plane_danger_write(struct file *file, + const char __user *user_buf, size_t count, loff_t *ppos) +{ + struct dpu_kms *kms = file->private_data; + int disable_panic; + int ret; + + ret = kstrtouint_from_user(user_buf, count, 0, &disable_panic); + if (ret) + return ret; + + if (disable_panic) { + /* Disable panic signal for all active pipes */ + DPU_DEBUG("Disabling danger:\n"); + _dpu_plane_set_danger_state(kms, false); + kms->has_danger_ctrl = false; + } else { + /* Enable panic signal for all active pipes */ + DPU_DEBUG("Enabling danger:\n"); + kms->has_danger_ctrl = true; + _dpu_plane_set_danger_state(kms, true); + } + + return count; +} + +static const struct file_operations dpu_plane_danger_enable = { + .open = simple_open, + .read = _dpu_plane_danger_read, + .write = _dpu_plane_danger_write, +}; + +static void dpu_debugfs_danger_init(struct dpu_kms *dpu_kms, + struct dentry *parent) +{ + struct dentry *entry = debugfs_create_dir("danger", parent); + + debugfs_create_file("danger_status", 0600, entry, + dpu_kms, &dpu_debugfs_danger_stats_fops); + debugfs_create_file("safe_status", 0600, entry, + dpu_kms, &dpu_debugfs_safe_stats_fops); + debugfs_create_file("disable_danger", 0600, entry, + dpu_kms, &dpu_plane_danger_enable); + +} + +/* + * Companion structure for dpu_debugfs_create_regset32. + */ +struct dpu_debugfs_regset32 { + uint32_t offset; + uint32_t blk_len; + struct dpu_kms *dpu_kms; +}; + +static int dpu_regset32_show(struct seq_file *s, void *data) +{ + struct dpu_debugfs_regset32 *regset = s->private; + struct dpu_kms *dpu_kms = regset->dpu_kms; + void __iomem *base; + uint32_t i, addr; + + if (!dpu_kms->mmio) + return 0; + + base = dpu_kms->mmio + regset->offset; + + /* insert padding spaces, if needed */ + if (regset->offset & 0xF) { + seq_printf(s, "[%x]", regset->offset & ~0xF); + for (i = 0; i < (regset->offset & 0xF); i += 4) + seq_puts(s, " "); + } + + pm_runtime_get_sync(&dpu_kms->pdev->dev); + + /* main register output */ + for (i = 0; i < regset->blk_len; i += 4) { + addr = regset->offset + i; + if ((addr & 0xF) == 0x0) + seq_printf(s, i ? "\n[%x]" : "[%x]", addr); + seq_printf(s, " %08x", readl_relaxed(base + i)); + } + seq_puts(s, "\n"); + pm_runtime_put_sync(&dpu_kms->pdev->dev); + + return 0; +} +DEFINE_SHOW_ATTRIBUTE(dpu_regset32); + +/** + * dpu_debugfs_create_regset32 - Create register read back file for debugfs + * + * This function is almost identical to the standard debugfs_create_regset32() + * function, with the main difference being that a list of register + * names/offsets do not need to be provided. The 'read' function simply outputs + * sequential register values over a specified range. + * + * @name: File name within debugfs + * @mode: File mode within debugfs + * @parent: Parent directory entry within debugfs, can be NULL + * @offset: sub-block offset + * @length: sub-block length, in bytes + * @dpu_kms: pointer to dpu kms structure + */ +void dpu_debugfs_create_regset32(const char *name, umode_t mode, + void *parent, + uint32_t offset, uint32_t length, struct dpu_kms *dpu_kms) +{ + struct dpu_debugfs_regset32 *regset; + + if (WARN_ON(!name || !dpu_kms || !length)) + return; + + regset = devm_kzalloc(&dpu_kms->pdev->dev, sizeof(*regset), GFP_KERNEL); + if (!regset) + return; + + /* make sure offset is a multiple of 4 */ + regset->offset = round_down(offset, 4); + regset->blk_len = length; + regset->dpu_kms = dpu_kms; + + debugfs_create_file(name, mode, parent, regset, &dpu_regset32_fops); +} + +static void dpu_debugfs_sspp_init(struct dpu_kms *dpu_kms, struct dentry *debugfs_root) +{ + struct dentry *entry = debugfs_create_dir("sspp", debugfs_root); + int i; + + if (IS_ERR(entry)) + return; + + for (i = SSPP_NONE; i < SSPP_MAX; i++) { + struct dpu_hw_sspp *hw = dpu_rm_get_sspp(&dpu_kms->rm, i); + + if (!hw) + continue; + + _dpu_hw_sspp_init_debugfs(hw, dpu_kms, entry); + } +} + +static int dpu_kms_debugfs_init(struct msm_kms *kms, struct drm_minor *minor) +{ + struct dpu_kms *dpu_kms = to_dpu_kms(kms); + void *p = dpu_hw_util_get_log_mask_ptr(); + struct dentry *entry; + + if (!p) + return -EINVAL; + + /* Only create a set of debugfs for the primary node, ignore render nodes */ + if (minor->type != DRM_MINOR_PRIMARY) + return 0; + + entry = debugfs_create_dir("debug", minor->debugfs_root); + + debugfs_create_x32(DPU_DEBUGFS_HWMASKNAME, 0600, entry, p); + + dpu_debugfs_danger_init(dpu_kms, entry); + dpu_debugfs_vbif_init(dpu_kms, entry); + dpu_debugfs_core_irq_init(dpu_kms, entry); + dpu_debugfs_sspp_init(dpu_kms, entry); + + return dpu_core_perf_debugfs_init(dpu_kms, entry); +} +#endif + +/* Global/shared object state funcs */ + +/* + * This is a helper that returns the private state currently in operation. + * Note that this would return the "old_state" if called in the atomic check + * path, and the "new_state" after the atomic swap has been done. + */ +struct dpu_global_state * +dpu_kms_get_existing_global_state(struct dpu_kms *dpu_kms) +{ + return to_dpu_global_state(dpu_kms->global_state.state); +} + +/* + * This acquires the modeset lock set aside for global state, creates + * a new duplicated private object state. + */ +struct dpu_global_state *dpu_kms_get_global_state(struct drm_atomic_commit *s) +{ + struct msm_drm_private *priv = s->dev->dev_private; + struct dpu_kms *dpu_kms = to_dpu_kms(priv->kms); + struct drm_private_state *priv_state; + + priv_state = drm_atomic_get_private_obj_state(s, + &dpu_kms->global_state); + if (IS_ERR(priv_state)) + return ERR_CAST(priv_state); + + return to_dpu_global_state(priv_state); +} + +static struct drm_private_state * +dpu_kms_global_duplicate_state(struct drm_private_obj *obj) +{ + struct dpu_global_state *state; + + state = kmemdup(obj->state, sizeof(*state), GFP_KERNEL); + if (!state) + return NULL; + + __drm_atomic_helper_private_obj_duplicate_state(obj, &state->base); + + return &state->base; +} + +static void dpu_kms_global_destroy_state(struct drm_private_obj *obj, + struct drm_private_state *state) +{ + struct dpu_global_state *dpu_state = to_dpu_global_state(state); + + kfree(dpu_state); +} + +static struct drm_private_state * +dpu_kms_global_create_state(struct drm_private_obj *obj) +{ + struct drm_device *dev = obj->dev; + struct msm_drm_private *priv = dev->dev_private; + struct dpu_kms *dpu_kms = to_dpu_kms(priv->kms); + struct dpu_global_state *dpu_state; + + dpu_state = kzalloc_obj(*dpu_state); + if (!dpu_state) + return ERR_PTR(-ENOMEM); + + __drm_atomic_helper_private_obj_create_state(obj, &dpu_state->base); + dpu_state->rm = &dpu_kms->rm; + + return &dpu_state->base; +} + +static void dpu_kms_global_print_state(struct drm_printer *p, + const struct drm_private_state *state) +{ + const struct dpu_global_state *global_state = to_dpu_global_state(state); + + dpu_rm_print_state(p, global_state); +} + +static const struct drm_private_state_funcs dpu_kms_global_state_funcs = { + .atomic_create_state = dpu_kms_global_create_state, + .atomic_duplicate_state = dpu_kms_global_duplicate_state, + .atomic_destroy_state = dpu_kms_global_destroy_state, + .atomic_print_state = dpu_kms_global_print_state, +}; + +static void dpu_kms_global_obj_fini(struct dpu_kms *dpu_kms) +{ + drm_atomic_private_obj_fini(&dpu_kms->global_state); +} + +static int dpu_kms_parse_data_bus_icc_path(struct dpu_kms *dpu_kms) +{ + struct icc_path *path0; + struct icc_path *path1; + struct device *dpu_dev = &dpu_kms->pdev->dev; + + path0 = msm_icc_get(dpu_dev, "mdp0-mem"); + path1 = msm_icc_get(dpu_dev, "mdp1-mem"); + + if (IS_ERR_OR_NULL(path0)) + return PTR_ERR_OR_ZERO(path0); + + dpu_kms->path[0] = path0; + dpu_kms->num_paths = 1; + + if (!IS_ERR_OR_NULL(path1)) { + dpu_kms->path[1] = path1; + dpu_kms->num_paths++; + } + return 0; +} + +static int dpu_kms_enable_vblank(struct msm_kms *kms, struct drm_crtc *crtc) +{ + return dpu_crtc_vblank(crtc, true); +} + +static void dpu_kms_disable_vblank(struct msm_kms *kms, struct drm_crtc *crtc) +{ + dpu_crtc_vblank(crtc, false); +} + +static void dpu_kms_enable_commit(struct msm_kms *kms) +{ + struct dpu_kms *dpu_kms = to_dpu_kms(kms); + pm_runtime_get_sync(&dpu_kms->pdev->dev); +} + +static void dpu_kms_disable_commit(struct msm_kms *kms) +{ + struct dpu_kms *dpu_kms = to_dpu_kms(kms); + pm_runtime_put_sync(&dpu_kms->pdev->dev); +} + +static int dpu_kms_check_mode_changed(struct msm_kms *kms, struct drm_atomic_commit *state) +{ + struct drm_crtc_state *new_crtc_state; + struct drm_crtc_state *old_crtc_state; + struct drm_crtc *crtc; + int i; + + for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) + dpu_crtc_check_mode_changed(old_crtc_state, new_crtc_state); + + return 0; +} + +static void dpu_kms_flush_commit(struct msm_kms *kms, unsigned crtc_mask) +{ + struct dpu_kms *dpu_kms = to_dpu_kms(kms); + struct drm_crtc *crtc; + + for_each_crtc_mask(dpu_kms->dev, crtc, crtc_mask) { + if (!crtc->state->active) + continue; + + trace_dpu_kms_commit(DRMID(crtc)); + dpu_crtc_commit_kickoff(crtc); + } +} + +static void dpu_kms_complete_commit(struct msm_kms *kms, unsigned crtc_mask) +{ + struct dpu_kms *dpu_kms = to_dpu_kms(kms); + struct drm_crtc *crtc; + + DPU_ATRACE_BEGIN("kms_complete_commit"); + + for_each_crtc_mask(dpu_kms->dev, crtc, crtc_mask) + dpu_crtc_complete_commit(crtc); + + DPU_ATRACE_END("kms_complete_commit"); +} + +static void dpu_kms_wait_for_commit_done(struct msm_kms *kms, + struct drm_crtc *crtc) +{ + struct drm_encoder *encoder; + struct drm_device *dev; + int ret; + + if (!kms || !crtc || !crtc->state) { + DPU_ERROR("invalid params\n"); + return; + } + + dev = crtc->dev; + + if (!crtc->state->enable) { + DPU_DEBUG("[crtc:%d] not enable\n", crtc->base.id); + return; + } + + if (!drm_atomic_crtc_effectively_active(crtc->state)) { + DPU_DEBUG("[crtc:%d] not active\n", crtc->base.id); + return; + } + + list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { + if (encoder->crtc != crtc) + continue; + /* + * Wait for post-flush if necessary to delay before + * plane_cleanup. For example, wait for vsync in case of video + * mode panels. This may be a no-op for command mode panels. + */ + trace_dpu_kms_wait_for_commit_done(DRMID(crtc)); + ret = dpu_encoder_wait_for_commit_done(encoder); + if (ret && ret != -EWOULDBLOCK) { + DPU_ERROR("wait for commit done returned %d\n", ret); + break; + } + } +} + +static void dpu_kms_wait_flush(struct msm_kms *kms, unsigned crtc_mask) +{ + struct dpu_kms *dpu_kms = to_dpu_kms(kms); + struct drm_crtc *crtc; + + for_each_crtc_mask(dpu_kms->dev, crtc, crtc_mask) + dpu_kms_wait_for_commit_done(kms, crtc); +} + +static const char *dpu_vsync_sources[] = { + [DPU_VSYNC_SOURCE_GPIO_0] = "mdp_vsync_p", + [DPU_VSYNC_SOURCE_GPIO_1] = "mdp_vsync_s", + [DPU_VSYNC_SOURCE_GPIO_2] = "mdp_vsync_e", + [DPU_VSYNC_SOURCE_INTF_0] = "mdp_intf0", + [DPU_VSYNC_SOURCE_INTF_1] = "mdp_intf1", + [DPU_VSYNC_SOURCE_INTF_2] = "mdp_intf2", + [DPU_VSYNC_SOURCE_INTF_3] = "mdp_intf3", + [DPU_VSYNC_SOURCE_WD_TIMER_0] = "timer0", + [DPU_VSYNC_SOURCE_WD_TIMER_1] = "timer1", + [DPU_VSYNC_SOURCE_WD_TIMER_2] = "timer2", + [DPU_VSYNC_SOURCE_WD_TIMER_3] = "timer3", + [DPU_VSYNC_SOURCE_WD_TIMER_4] = "timer4", +}; + +static int dpu_kms_dsi_set_te_source(struct msm_display_info *info, + struct msm_dsi *dsi) +{ + const char *te_source = msm_dsi_get_te_source(dsi); + int i; + + if (!te_source) { + info->vsync_source = DPU_VSYNC_SOURCE_GPIO_0; + return 0; + } + + /* we can not use match_string since dpu_vsync_sources is a sparse array */ + for (i = 0; i < ARRAY_SIZE(dpu_vsync_sources); i++) { + if (dpu_vsync_sources[i] && + !strcmp(dpu_vsync_sources[i], te_source)) { + info->vsync_source = i; + return 0; + } + } + + return -EINVAL; +} + +static int _dpu_kms_initialize_dsi(struct drm_device *dev, + struct msm_drm_private *priv, + struct dpu_kms *dpu_kms) +{ + struct drm_encoder *encoder = NULL; + struct msm_display_info info; + int i, rc = 0; + + if (!(priv->kms->dsi[0] || priv->kms->dsi[1])) + return rc; + + /* + * We support following confiurations: + * - Single DSI host (dsi0 or dsi1) + * - Two independent DSI hosts + * - Bonded DSI0 and DSI1 hosts + * + * TODO: Support swapping DSI0 and DSI1 in the bonded setup. + */ + for (i = 0; i < ARRAY_SIZE(priv->kms->dsi); i++) { + int other = (i + 1) % 2; + + if (!priv->kms->dsi[i]) + continue; + + if (msm_dsi_is_bonded_dsi(priv->kms->dsi[i]) && + !msm_dsi_is_master_dsi(priv->kms->dsi[i])) + continue; + + memset(&info, 0, sizeof(info)); + info.intf_type = INTF_DSI; + + info.h_tile_instance[info.num_of_h_tiles++] = i; + if (msm_dsi_is_bonded_dsi(priv->kms->dsi[i])) + info.h_tile_instance[info.num_of_h_tiles++] = other; + + info.is_cmd_mode = msm_dsi_is_cmd_mode(priv->kms->dsi[i]); + + rc = dpu_kms_dsi_set_te_source(&info, priv->kms->dsi[i]); + if (rc) { + DPU_ERROR("failed to identify TE source for dsi display\n"); + return rc; + } + + encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_DSI, &info); + if (IS_ERR(encoder)) { + DPU_ERROR("encoder init failed for dsi display\n"); + return PTR_ERR(encoder); + } + + rc = msm_dsi_modeset_init(priv->kms->dsi[i], dev, encoder); + if (rc) { + DPU_ERROR("modeset_init failed for dsi[%d], rc = %d\n", + i, rc); + break; + } + + if (msm_dsi_is_bonded_dsi(priv->kms->dsi[i]) && priv->kms->dsi[other]) { + rc = msm_dsi_modeset_init(priv->kms->dsi[other], dev, encoder); + if (rc) { + DPU_ERROR("modeset_init failed for dsi[%d], rc = %d\n", + other, rc); + break; + } + } + } + + return rc; +} + +static int _dpu_kms_initialize_displayport(struct drm_device *dev, + struct msm_drm_private *priv, + struct dpu_kms *dpu_kms) +{ + struct drm_encoder *encoder = NULL; + struct msm_display_info info; + bool yuv_supported; + int rc; + int i; + + for (i = 0; i < ARRAY_SIZE(priv->kms->dp); i++) { + if (!priv->kms->dp[i]) + continue; + + memset(&info, 0, sizeof(info)); + info.num_of_h_tiles = 1; + info.h_tile_instance[0] = i; + info.intf_type = INTF_DP; + + encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_TMDS, &info); + if (IS_ERR(encoder)) { + DPU_ERROR("encoder init failed for dsi display\n"); + return PTR_ERR(encoder); + } + + yuv_supported = !!dpu_kms->catalog->cdm; + rc = msm_dp_modeset_init(priv->kms->dp[i], dev, encoder, yuv_supported); + if (rc) { + DPU_ERROR("modeset_init failed for DP, rc = %d\n", rc); + return rc; + } + } + + return 0; +} + +static int _dpu_kms_initialize_hdmi(struct drm_device *dev, + struct msm_drm_private *priv, + struct dpu_kms *dpu_kms) +{ + struct drm_encoder *encoder = NULL; + struct msm_display_info info; + int rc; + + if (!priv->kms->hdmi) + return 0; + + memset(&info, 0, sizeof(info)); + info.num_of_h_tiles = 1; + info.h_tile_instance[0] = 0; + info.intf_type = INTF_HDMI; + + encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_TMDS, &info); + if (IS_ERR(encoder)) { + DPU_ERROR("encoder init failed for HDMI display\n"); + return PTR_ERR(encoder); + } + + rc = msm_hdmi_modeset_init(priv->kms->hdmi, dev, encoder); + if (rc) { + DPU_ERROR("modeset_init failed for DP, rc = %d\n", rc); + return rc; + } + + return 0; +} + +static int _dpu_kms_initialize_writeback(struct drm_device *dev, + struct msm_drm_private *priv, struct dpu_kms *dpu_kms, + const u32 *wb_formats, int n_formats) +{ + struct drm_encoder *encoder = NULL; + struct msm_display_info info; + const enum dpu_wb wb_idx = WB_2; + u32 maxlinewidth; + int rc; + + memset(&info, 0, sizeof(info)); + + info.num_of_h_tiles = 1; + /* use only WB idx 2 instance for DPU */ + info.h_tile_instance[0] = wb_idx; + info.intf_type = INTF_WB; + + maxlinewidth = dpu_rm_get_wb(&dpu_kms->rm, info.h_tile_instance[0])->caps->maxlinewidth; + + encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_VIRTUAL, &info); + if (IS_ERR(encoder)) { + DPU_ERROR("encoder init failed for dsi display\n"); + return PTR_ERR(encoder); + } + + rc = dpu_writeback_init(dev, encoder, wb_formats, n_formats, maxlinewidth); + if (rc) { + DPU_ERROR("dpu_writeback_init, rc = %d\n", rc); + return rc; + } + + return 0; +} + +/** + * _dpu_kms_setup_displays - create encoders, bridges and connectors + * for underlying displays + * @dev: Pointer to drm device structure + * @priv: Pointer to private drm device data + * @dpu_kms: Pointer to dpu kms structure + * Returns: Zero on success + */ +static int _dpu_kms_setup_displays(struct drm_device *dev, + struct msm_drm_private *priv, + struct dpu_kms *dpu_kms) +{ + int rc = 0; + int i; + + rc = _dpu_kms_initialize_dsi(dev, priv, dpu_kms); + if (rc) { + DPU_ERROR("initialize_dsi failed, rc = %d\n", rc); + return rc; + } + + rc = _dpu_kms_initialize_displayport(dev, priv, dpu_kms); + if (rc) { + DPU_ERROR("initialize_DP failed, rc = %d\n", rc); + return rc; + } + + rc = _dpu_kms_initialize_hdmi(dev, priv, dpu_kms); + if (rc) { + DPU_ERROR("initialize HDMI failed, rc = %d\n", rc); + return rc; + } + + /* Since WB isn't a driver check the catalog before initializing */ + if (dpu_kms->catalog->wb_count) { + for (i = 0; i < dpu_kms->catalog->wb_count; i++) { + if (dpu_kms->catalog->wb[i].id == WB_2) { + rc = _dpu_kms_initialize_writeback(dev, priv, dpu_kms, + dpu_kms->catalog->wb[i].format_list, + dpu_kms->catalog->wb[i].num_formats); + if (rc) { + DPU_ERROR("initialize_WB failed, rc = %d\n", rc); + return rc; + } + } + } + } + + return rc; +} + +#define MAX_PLANES 20 +static int _dpu_kms_drm_obj_init(struct dpu_kms *dpu_kms) +{ + struct drm_device *dev; + struct drm_plane *primary_planes[MAX_PLANES], *plane; + struct drm_plane *cursor_planes[MAX_PLANES] = { NULL }; + struct drm_crtc *crtc; + struct drm_encoder *encoder; + unsigned int num_encoders; + + struct msm_drm_private *priv; + const struct dpu_mdss_cfg *catalog; + + int primary_planes_idx = 0, cursor_planes_idx = 0, i, ret; + int max_crtc_count; + dev = dpu_kms->dev; + priv = dev->dev_private; + catalog = dpu_kms->catalog; + + /* + * Create encoder and query display drivers to create + * bridges and connectors + */ + ret = _dpu_kms_setup_displays(dev, priv, dpu_kms); + if (ret) + return ret; + + num_encoders = 0; + drm_for_each_encoder(encoder, dev) { + num_encoders++; + if (catalog->cwb_count > 0) + encoder->possible_clones = dpu_encoder_get_clones(encoder); + } + + max_crtc_count = min(catalog->mixer_count, num_encoders); + + /* Create the planes, keeping track of one primary/cursor per crtc */ + for (i = 0; i < catalog->sspp_count; i++) { + enum drm_plane_type type; + + if ((catalog->sspp[i].features & BIT(DPU_SSPP_CURSOR)) + && cursor_planes_idx < max_crtc_count) + type = DRM_PLANE_TYPE_CURSOR; + else if (primary_planes_idx < max_crtc_count) + type = DRM_PLANE_TYPE_PRIMARY; + else + type = DRM_PLANE_TYPE_OVERLAY; + + DPU_DEBUG("Create plane type %d with features %lx (cur %lx)\n", + type, catalog->sspp[i].features, + catalog->sspp[i].features & BIT(DPU_SSPP_CURSOR)); + + if (dpu_use_virtual_planes) + plane = dpu_plane_init_virtual(dev, type, (1UL << max_crtc_count) - 1); + else + plane = dpu_plane_init(dev, catalog->sspp[i].id, type, + (1UL << max_crtc_count) - 1); + if (IS_ERR(plane)) { + DPU_ERROR("dpu_plane_init failed\n"); + ret = PTR_ERR(plane); + return ret; + } + + if (type == DRM_PLANE_TYPE_CURSOR) + cursor_planes[cursor_planes_idx++] = plane; + else if (type == DRM_PLANE_TYPE_PRIMARY) + primary_planes[primary_planes_idx++] = plane; + } + + max_crtc_count = min(max_crtc_count, primary_planes_idx); + + /* Create one CRTC per encoder */ + for (i = 0; i < max_crtc_count; i++) { + crtc = dpu_crtc_init(dev, primary_planes[i], cursor_planes[i]); + if (IS_ERR(crtc)) { + ret = PTR_ERR(crtc); + return ret; + } + } + + /* All CRTCs are compatible with all encoders */ + drm_for_each_encoder(encoder, dev) + encoder->possible_crtcs = (1 << dev->mode_config.num_crtc) - 1; + + return 0; +} + +static void _dpu_kms_hw_destroy(struct dpu_kms *dpu_kms) +{ + dpu_kms->hw_intr = NULL; + + /* safe to call these more than once during shutdown */ + _dpu_kms_mmu_destroy(dpu_kms); + + dpu_kms->hw_vbif = NULL; + + dpu_kms_global_obj_fini(dpu_kms); + + dpu_kms->catalog = NULL; + + dpu_kms->hw_mdp = NULL; +} + +static void dpu_kms_destroy(struct msm_kms *kms) +{ + struct dpu_kms *dpu_kms; + + if (!kms) { + DPU_ERROR("invalid kms\n"); + return; + } + + dpu_kms = to_dpu_kms(kms); + + _dpu_kms_hw_destroy(dpu_kms); + + msm_kms_destroy(&dpu_kms->base); + + if (dpu_kms->rpm_enabled) + pm_runtime_disable(&dpu_kms->pdev->dev); +} + +static int dpu_irq_postinstall(struct msm_kms *kms) +{ + struct msm_drm_private *priv; + struct dpu_kms *dpu_kms = to_dpu_kms(kms); + + if (!dpu_kms || !dpu_kms->dev) + return -EINVAL; + + priv = dpu_kms->dev->dev_private; + if (!priv) + return -EINVAL; + + return 0; +} + +static void dpu_kms_mdp_snapshot(struct msm_disp_state *disp_state, struct msm_kms *kms) +{ + int i; + struct dpu_kms *dpu_kms; + const struct dpu_mdss_cfg *cat; + void __iomem *base; + + dpu_kms = to_dpu_kms(kms); + + cat = dpu_kms->catalog; + + pm_runtime_get_sync(&dpu_kms->pdev->dev); + + /* dump CTL sub-blocks HW regs info */ + for (i = 0; i < cat->ctl_count; i++) + msm_disp_snapshot_add_block(disp_state, cat->ctl[i].len, + dpu_kms->mmio + cat->ctl[i].base, "%s", + cat->ctl[i].name); + + /* dump DSPP sub-blocks HW regs info */ + for (i = 0; i < cat->dspp_count; i++) { + base = dpu_kms->mmio + cat->dspp[i].base; + msm_disp_snapshot_add_block(disp_state, cat->dspp[i].len, base, + "%s", cat->dspp[i].name); + + if (cat->dspp[i].sblk && cat->dspp[i].sblk->pcc.len > 0) + msm_disp_snapshot_add_block(disp_state, cat->dspp[i].sblk->pcc.len, + base + cat->dspp[i].sblk->pcc.base, "%s_%s", + cat->dspp[i].name, + cat->dspp[i].sblk->pcc.name); + } + + /* dump INTF sub-blocks HW regs info */ + for (i = 0; i < cat->intf_count; i++) + msm_disp_snapshot_add_block(disp_state, cat->intf[i].len, + dpu_kms->mmio + cat->intf[i].base, "%s", + cat->intf[i].name); + + /* dump PP sub-blocks HW regs info */ + for (i = 0; i < cat->pingpong_count; i++) { + base = dpu_kms->mmio + cat->pingpong[i].base; + msm_disp_snapshot_add_block(disp_state, cat->pingpong[i].len, base, + "%s", cat->pingpong[i].name); + + /* TE2 sub-block has length of 0, so will not print it */ + + if (cat->pingpong[i].sblk && cat->pingpong[i].sblk->dither.len > 0) + msm_disp_snapshot_add_block(disp_state, cat->pingpong[i].sblk->dither.len, + base + cat->pingpong[i].sblk->dither.base, + "%s_%s", cat->pingpong[i].name, + cat->pingpong[i].sblk->dither.name); + } + + /* dump SSPP sub-blocks HW regs info */ + for (i = 0; i < cat->sspp_count; i++) { + base = dpu_kms->mmio + cat->sspp[i].base; + msm_disp_snapshot_add_block(disp_state, cat->sspp[i].len, base, + "%s", cat->sspp[i].name); + + if (cat->sspp[i].sblk && cat->sspp[i].sblk->scaler_blk.len > 0) + msm_disp_snapshot_add_block(disp_state, cat->sspp[i].sblk->scaler_blk.len, + base + cat->sspp[i].sblk->scaler_blk.base, + "%s_%s", cat->sspp[i].name, + cat->sspp[i].sblk->scaler_blk.name); + + if (cat->sspp[i].sblk && cat->sspp[i].sblk->csc_blk.len > 0) + msm_disp_snapshot_add_block(disp_state, cat->sspp[i].sblk->csc_blk.len, + base + cat->sspp[i].sblk->csc_blk.base, + "%s_%s", cat->sspp[i].name, + cat->sspp[i].sblk->csc_blk.name); + } + + /* dump LM sub-blocks HW regs info */ + for (i = 0; i < cat->mixer_count; i++) + msm_disp_snapshot_add_block(disp_state, cat->mixer[i].len, + dpu_kms->mmio + cat->mixer[i].base, + "%s", cat->mixer[i].name); + + /* dump WB sub-blocks HW regs info */ + for (i = 0; i < cat->wb_count; i++) + msm_disp_snapshot_add_block(disp_state, cat->wb[i].len, + dpu_kms->mmio + cat->wb[i].base, "%s", + cat->wb[i].name); + + if (dpu_kms->catalog->mdss_ver->core_major_ver >= 8) { + msm_disp_snapshot_add_block(disp_state, MDP_PERIPH_TOP0, + dpu_kms->mmio + cat->mdp[0].base, "top"); + msm_disp_snapshot_add_block(disp_state, cat->mdp[0].len - MDP_PERIPH_TOP0_END, + dpu_kms->mmio + cat->mdp[0].base + MDP_PERIPH_TOP0_END, "top_2"); + } else { + msm_disp_snapshot_add_block(disp_state, cat->mdp[0].len, + dpu_kms->mmio + cat->mdp[0].base, "top"); + } + + /* dump CWB sub-blocks HW regs info */ + for (i = 0; i < cat->cwb_count; i++) + msm_disp_snapshot_add_block(disp_state, cat->cwb[i].len, + dpu_kms->mmio + cat->cwb[i].base, cat->cwb[i].name); + + /* dump DSC sub-blocks HW regs info */ + for (i = 0; i < cat->dsc_count; i++) { + base = dpu_kms->mmio + cat->dsc[i].base; + msm_disp_snapshot_add_block(disp_state, cat->dsc[i].len, base, + "%s", cat->dsc[i].name); + + if (cat->mdss_ver->core_major_ver >= 7) { + struct dpu_dsc_blk enc = cat->dsc[i].sblk->enc; + struct dpu_dsc_blk ctl = cat->dsc[i].sblk->ctl; + + msm_disp_snapshot_add_block(disp_state, enc.len, base + enc.base, "%s_%s", + cat->dsc[i].name, enc.name); + msm_disp_snapshot_add_block(disp_state, ctl.len, base + ctl.base, "%s_%s", + cat->dsc[i].name, ctl.name); + } + } + + if (cat->cdm) + msm_disp_snapshot_add_block(disp_state, cat->cdm->len, + dpu_kms->mmio + cat->cdm->base, + "%s", cat->cdm->name); + + const struct dpu_vbif_cfg *vbif = dpu_kms->catalog->vbif; + + msm_disp_snapshot_add_block(disp_state, vbif->len, + dpu_kms->vbif, + "vbif"); + + pm_runtime_put_sync(&dpu_kms->pdev->dev); +} + +static const struct msm_kms_funcs kms_funcs = { + .hw_init = dpu_kms_hw_init, + .irq_preinstall = dpu_core_irq_preinstall, + .irq_postinstall = dpu_irq_postinstall, + .irq_uninstall = dpu_core_irq_uninstall, + .irq = dpu_core_irq, + .enable_commit = dpu_kms_enable_commit, + .disable_commit = dpu_kms_disable_commit, + .check_mode_changed = dpu_kms_check_mode_changed, + .flush_commit = dpu_kms_flush_commit, + .wait_flush = dpu_kms_wait_flush, + .complete_commit = dpu_kms_complete_commit, + .enable_vblank = dpu_kms_enable_vblank, + .disable_vblank = dpu_kms_disable_vblank, + .destroy = dpu_kms_destroy, + .snapshot = dpu_kms_mdp_snapshot, +#ifdef CONFIG_DEBUG_FS + .debugfs_init = dpu_kms_debugfs_init, +#endif +}; + +static void _dpu_kms_mmu_destroy(struct dpu_kms *dpu_kms) +{ + struct msm_mmu *mmu; + + if (!dpu_kms->base.vm) + return; + + mmu = to_msm_vm(dpu_kms->base.vm)->mmu; + + mmu->funcs->detach(mmu); + drm_gpuvm_put(dpu_kms->base.vm); + + dpu_kms->base.vm = NULL; +} + +static int _dpu_kms_mmu_init(struct dpu_kms *dpu_kms) +{ + struct drm_gpuvm *vm; + + vm = msm_kms_init_vm(dpu_kms->dev, dpu_kms->dev->dev->parent); + if (IS_ERR(vm)) + return PTR_ERR(vm); + + dpu_kms->base.vm = vm; + + return 0; +} + +/** + * dpu_kms_get_clk_rate() - get the clock rate + * @dpu_kms: pointer to dpu_kms structure + * @clock_name: clock name to get the rate + * + * Return: current clock rate + */ +unsigned long dpu_kms_get_clk_rate(struct dpu_kms *dpu_kms, char *clock_name) +{ + struct clk *clk; + + clk = msm_clk_bulk_get_clock(dpu_kms->clocks, dpu_kms->num_clocks, clock_name); + if (!clk) + return 0; + + return clk_get_rate(clk); +} + +#define DPU_PERF_DEFAULT_MAX_CORE_CLK_RATE 412500000 + +static int dpu_kms_hw_init(struct msm_kms *kms) +{ + struct dpu_kms *dpu_kms; + struct drm_device *dev; + int rc = -EINVAL; + unsigned long max_core_clk_rate; + u32 core_rev; + + if (!kms) { + DPU_ERROR("invalid kms\n"); + return rc; + } + + dpu_kms = to_dpu_kms(kms); + dev = dpu_kms->dev; + + dev->mode_config.cursor_width = 512; + dev->mode_config.cursor_height = 512; + + drm_atomic_private_obj_init(dpu_kms->dev, &dpu_kms->global_state, + &dpu_kms_global_state_funcs); + + atomic_set(&dpu_kms->bandwidth_ref, 0); + + rc = pm_runtime_resume_and_get(&dpu_kms->pdev->dev); + if (rc < 0) + goto error; + + core_rev = readl_relaxed(dpu_kms->mmio + 0x0); + + pr_info("dpu hardware revision:0x%x\n", core_rev); + + dpu_kms->catalog = of_device_get_match_data(dev->dev); + if (!dpu_kms->catalog) { + DPU_ERROR("device config not known!\n"); + rc = -EINVAL; + goto err_pm_put; + } + + /* + * Now we need to read the HW catalog and initialize resources such as + * clocks, regulators, GDSC/MMAGIC, ioremap the register ranges etc + */ + rc = _dpu_kms_mmu_init(dpu_kms); + if (rc) { + DPU_ERROR("dpu_kms_mmu_init failed: %d\n", rc); + goto err_pm_put; + } + + dpu_kms->mdss = qcom_ubwc_config_get_data(); + if (IS_ERR(dpu_kms->mdss)) { + rc = PTR_ERR(dpu_kms->mdss); + DPU_ERROR("failed to get UBWC config data: %d\n", rc); + goto err_pm_put; + } + + if (!dpu_kms->mdss) { + rc = -EINVAL; + DPU_ERROR("NULL MDSS data\n"); + goto err_pm_put; + } + + rc = dpu_rm_init(dev, &dpu_kms->rm, dpu_kms->catalog, dpu_kms->mdss, dpu_kms->mmio); + if (rc) { + DPU_ERROR("rm init failed: %d\n", rc); + goto err_pm_put; + } + + dpu_kms->hw_mdp = dpu_hw_mdptop_init(dev, + dpu_kms->catalog->mdp, + dpu_kms->mmio, + dpu_kms->catalog->mdss_ver); + if (IS_ERR(dpu_kms->hw_mdp)) { + rc = PTR_ERR(dpu_kms->hw_mdp); + DPU_ERROR("failed to get hw_mdp: %d\n", rc); + dpu_kms->hw_mdp = NULL; + goto err_pm_put; + } + + struct dpu_hw_vbif *hw; + const struct dpu_vbif_cfg *vbif = dpu_kms->catalog->vbif; + + hw = dpu_hw_vbif_init(dev, vbif, dpu_kms->vbif); + if (IS_ERR(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed to init vbif: %d\n", rc); + goto err_pm_put; + } + + dpu_kms->hw_vbif = hw; + + /* TODO: use the same max_freq as in dpu_kms_hw_init */ + max_core_clk_rate = dpu_kms_get_clk_rate(dpu_kms, "core"); + if (!max_core_clk_rate) { + DPU_DEBUG("max core clk rate not determined, using default\n"); + max_core_clk_rate = DPU_PERF_DEFAULT_MAX_CORE_CLK_RATE; + } + + rc = dpu_core_perf_init(&dpu_kms->perf, dpu_kms->catalog->perf, max_core_clk_rate); + if (rc) { + DPU_ERROR("failed to init perf %d\n", rc); + goto err_pm_put; + } + + /* + * We need to program DP <-> PHY relationship only for SC8180X since it + * has fewer DP controllers than DP PHYs. + * If any other platform requires the same kind of programming, or if + * the INTF <->DP relationship isn't static anymore, this needs to be + * configured through the DT. + */ + if (of_device_is_compatible(dpu_kms->pdev->dev.of_node, "qcom,sc8180x-dpu")) + dpu_kms->hw_mdp->ops.dp_phy_intf_sel(dpu_kms->hw_mdp, (unsigned int[]){ 1, 2, }); + + dpu_kms->hw_intr = dpu_hw_intr_init(dev, dpu_kms->mmio, dpu_kms->catalog); + if (IS_ERR(dpu_kms->hw_intr)) { + rc = PTR_ERR(dpu_kms->hw_intr); + DPU_ERROR("hw_intr init failed: %d\n", rc); + dpu_kms->hw_intr = NULL; + goto err_pm_put; + } + + dev->mode_config.min_width = 0; + dev->mode_config.min_height = 0; + + dev->mode_config.max_width = DPU_MAX_IMG_WIDTH; + dev->mode_config.max_height = DPU_MAX_IMG_HEIGHT; + + dev->max_vblank_count = 0xffffffff; + /* Disable vblank irqs aggressively for power-saving */ + dev->vblank_disable_immediate = true; + + /* + * _dpu_kms_drm_obj_init should create the DRM related objects + * i.e. CRTCs, planes, encoders, connectors and so forth + */ + rc = _dpu_kms_drm_obj_init(dpu_kms); + if (rc) { + DPU_ERROR("modeset init failed: %d\n", rc); + goto err_pm_put; + } + + dpu_vbif_init_memtypes(dpu_kms); + + pm_runtime_put_sync(&dpu_kms->pdev->dev); + + return 0; + +err_pm_put: + pm_runtime_put_sync(&dpu_kms->pdev->dev); +error: + _dpu_kms_hw_destroy(dpu_kms); + + return rc; +} + +static int dpu_kms_init(struct drm_device *ddev) +{ + struct msm_drm_private *priv = ddev->dev_private; + struct device *dev = ddev->dev; + struct platform_device *pdev = to_platform_device(dev); + struct dpu_kms *dpu_kms = to_dpu_kms(priv->kms); + struct dev_pm_opp *opp; + int ret = 0; + unsigned long max_freq = ULONG_MAX; + + opp = dev_pm_opp_find_freq_floor(dev, &max_freq); + if (!IS_ERR(opp)) + dev_pm_opp_put(opp); + + dev_pm_opp_set_rate(dev, max_freq); + + ret = msm_kms_init(&dpu_kms->base, &kms_funcs); + if (ret) { + DPU_ERROR("failed to init kms, ret=%d\n", ret); + return ret; + } + dpu_kms->dev = ddev; + + pm_runtime_enable(&pdev->dev); + dpu_kms->rpm_enabled = true; + + return 0; +} + +static int dpu_kms_mmap_mdp5(struct dpu_kms *dpu_kms) +{ + struct platform_device *pdev = dpu_kms->pdev; + struct platform_device *mdss_dev; + int ret; + + if (!dev_is_platform(dpu_kms->pdev->dev.parent)) + return -EINVAL; + + mdss_dev = to_platform_device(dpu_kms->pdev->dev.parent); + + dpu_kms->mmio = msm_ioremap(pdev, "mdp_phys"); + if (IS_ERR(dpu_kms->mmio)) { + ret = PTR_ERR(dpu_kms->mmio); + DPU_ERROR("mdp register memory map failed: %d\n", ret); + dpu_kms->mmio = NULL; + return ret; + } + DRM_DEBUG("mapped dpu address space @%p\n", dpu_kms->mmio); + + dpu_kms->vbif = msm_ioremap_mdss(mdss_dev, dpu_kms->pdev, "vbif_phys"); + if (IS_ERR(dpu_kms->vbif)) { + ret = PTR_ERR(dpu_kms->vbif); + DPU_ERROR("vbif register memory map failed: %d\n", ret); + dpu_kms->vbif = NULL; + return ret; + } + + return 0; +} + +static int dpu_kms_mmap_dpu(struct dpu_kms *dpu_kms) +{ + struct platform_device *pdev = dpu_kms->pdev; + int ret; + + dpu_kms->mmio = msm_ioremap(pdev, "mdp"); + if (IS_ERR(dpu_kms->mmio)) { + ret = PTR_ERR(dpu_kms->mmio); + DPU_ERROR("mdp register memory map failed: %d\n", ret); + dpu_kms->mmio = NULL; + return ret; + } + DRM_DEBUG("mapped dpu address space @%p\n", dpu_kms->mmio); + + dpu_kms->vbif = msm_ioremap(pdev, "vbif"); + if (IS_ERR(dpu_kms->vbif)) { + ret = PTR_ERR(dpu_kms->vbif); + DPU_ERROR("vbif register memory map failed: %d\n", ret); + dpu_kms->vbif = NULL; + return ret; + } + + return 0; +} + +static int dpu_dev_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct dpu_kms *dpu_kms; + int irq; + int ret = 0; + + if (!msm_disp_drv_should_bind(&pdev->dev, true)) + return -ENODEV; + + dpu_kms = devm_kzalloc(dev, sizeof(*dpu_kms), GFP_KERNEL); + if (!dpu_kms) + return -ENOMEM; + + dpu_kms->pdev = pdev; + + ret = devm_pm_opp_set_clkname(dev, "core"); + if (ret) + return ret; + /* OPP table is optional */ + ret = devm_pm_opp_of_add_table(dev); + if (ret && ret != -ENODEV) + return dev_err_probe(dev, ret, "invalid OPP table in device tree\n"); + + ret = devm_clk_bulk_get_all(&pdev->dev, &dpu_kms->clocks); + if (ret < 0) + return dev_err_probe(dev, ret, "failed to parse clocks\n"); + + dpu_kms->num_clocks = ret; + + irq = platform_get_irq(pdev, 0); + if (irq < 0) + return dev_err_probe(dev, irq, "failed to get irq\n"); + + dpu_kms->base.irq = irq; + + if (of_device_is_compatible(dpu_kms->pdev->dev.of_node, "qcom,mdp5")) + ret = dpu_kms_mmap_mdp5(dpu_kms); + else + ret = dpu_kms_mmap_dpu(dpu_kms); + if (ret) + return ret; + + ret = dpu_kms_parse_data_bus_icc_path(dpu_kms); + if (ret) + return ret; + + return msm_drv_probe(&pdev->dev, dpu_kms_init, &dpu_kms->base); +} + +static void dpu_dev_remove(struct platform_device *pdev) +{ + component_master_del(&pdev->dev, &msm_drm_ops); +} + +static int __maybe_unused dpu_runtime_suspend(struct device *dev) +{ + int i; + struct platform_device *pdev = to_platform_device(dev); + struct msm_drm_private *priv = platform_get_drvdata(pdev); + struct dpu_kms *dpu_kms = to_dpu_kms(priv->kms); + + clk_bulk_disable_unprepare(dpu_kms->num_clocks, dpu_kms->clocks); + + for (i = 0; i < dpu_kms->num_paths; i++) + icc_set_bw(dpu_kms->path[i], 0, 0); + + return 0; +} + +static int __maybe_unused dpu_runtime_resume(struct device *dev) +{ + int rc = -1; + struct platform_device *pdev = to_platform_device(dev); + struct msm_drm_private *priv = platform_get_drvdata(pdev); + struct dpu_kms *dpu_kms = to_dpu_kms(priv->kms); + struct drm_encoder *encoder; + struct drm_device *ddev; + + ddev = dpu_kms->dev; + + rc = clk_bulk_prepare_enable(dpu_kms->num_clocks, dpu_kms->clocks); + if (rc) { + DPU_ERROR("clock enable failed rc:%d\n", rc); + return rc; + } + + dpu_vbif_init_memtypes(dpu_kms); + + drm_for_each_encoder(encoder, ddev) + dpu_encoder_virt_runtime_resume(encoder); + + return rc; +} + +static const struct dev_pm_ops dpu_pm_ops = { + SET_RUNTIME_PM_OPS(dpu_runtime_suspend, dpu_runtime_resume, NULL) + SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, + pm_runtime_force_resume) + .prepare = msm_kms_pm_prepare, + .complete = msm_kms_pm_complete, +}; + +static const struct of_device_id dpu_dt_match[] = { + { .compatible = "qcom,eliza-dpu", .data = &dpu_eliza_cfg, }, + { .compatible = "qcom,glymur-dpu", .data = &dpu_glymur_cfg, }, + { .compatible = "qcom,kaanapali-dpu", .data = &dpu_kaanapali_cfg, }, + { .compatible = "qcom,milos-dpu", .data = &dpu_milos_cfg, }, + { .compatible = "qcom,msm8917-mdp5", .data = &dpu_msm8917_cfg, }, + { .compatible = "qcom,msm8937-mdp5", .data = &dpu_msm8937_cfg, }, + { .compatible = "qcom,msm8953-mdp5", .data = &dpu_msm8953_cfg, }, + { .compatible = "qcom,msm8996-mdp5", .data = &dpu_msm8996_cfg, }, + { .compatible = "qcom,msm8998-dpu", .data = &dpu_msm8998_cfg, }, + { .compatible = "qcom,qcm2290-dpu", .data = &dpu_qcm2290_cfg, }, + { .compatible = "qcom,sa8775p-dpu", .data = &dpu_sa8775p_cfg, }, + { .compatible = "qcom,sar2130p-dpu", .data = &dpu_sar2130p_cfg, }, + { .compatible = "qcom,sdm630-mdp5", .data = &dpu_sdm630_cfg, }, + { .compatible = "qcom,sdm660-mdp5", .data = &dpu_sdm660_cfg, }, + { .compatible = "qcom,sdm670-dpu", .data = &dpu_sdm670_cfg, }, + { .compatible = "qcom,sdm845-dpu", .data = &dpu_sdm845_cfg, }, + { .compatible = "qcom,sc7180-dpu", .data = &dpu_sc7180_cfg, }, + { .compatible = "qcom,sc7280-dpu", .data = &dpu_sc7280_cfg, }, + { .compatible = "qcom,sc8180x-dpu", .data = &dpu_sc8180x_cfg, }, + { .compatible = "qcom,sc8280xp-dpu", .data = &dpu_sc8280xp_cfg, }, + { .compatible = "qcom,sm6115-dpu", .data = &dpu_sm6115_cfg, }, + { .compatible = "qcom,sm6125-dpu", .data = &dpu_sm6125_cfg, }, + { .compatible = "qcom,sm6150-dpu", .data = &dpu_sm6150_cfg, }, + { .compatible = "qcom,sm6350-dpu", .data = &dpu_sm6350_cfg, }, + { .compatible = "qcom,sm6375-dpu", .data = &dpu_sm6375_cfg, }, + { .compatible = "qcom,sm7150-dpu", .data = &dpu_sm7150_cfg, }, + { .compatible = "qcom,sm8150-dpu", .data = &dpu_sm8150_cfg, }, + { .compatible = "qcom,sm8250-dpu", .data = &dpu_sm8250_cfg, }, + { .compatible = "qcom,sm8350-dpu", .data = &dpu_sm8350_cfg, }, + { .compatible = "qcom,sm8450-dpu", .data = &dpu_sm8450_cfg, }, + { .compatible = "qcom,sm8550-dpu", .data = &dpu_sm8550_cfg, }, + { .compatible = "qcom,sm8650-dpu", .data = &dpu_sm8650_cfg, }, + { .compatible = "qcom,sm8750-dpu", .data = &dpu_sm8750_cfg, }, + { .compatible = "qcom,x1e80100-dpu", .data = &dpu_x1e80100_cfg, }, + {} +}; +MODULE_DEVICE_TABLE(of, dpu_dt_match); + +static struct platform_driver dpu_driver = { + .probe = dpu_dev_probe, + .remove = dpu_dev_remove, + .shutdown = msm_kms_shutdown, + .driver = { + .name = "msm_dpu", + .of_match_table = dpu_dt_match, + .pm = &dpu_pm_ops, + }, +}; + +void __init msm_dpu_register(void) +{ + platform_driver_register(&dpu_driver); +} + +void __exit msm_dpu_unregister(void) +{ + platform_driver_unregister(&dpu_driver); +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h new file mode 100644 index 000000000..e39831a39 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h @@ -0,0 +1,172 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#ifndef __DPU_KMS_H__ +#define __DPU_KMS_H__ + +#include + +#include + +#include "msm_drv.h" +#include "msm_kms.h" +#include "msm_mmu.h" +#include "msm_gem.h" +#include "dpu_hw_catalog.h" +#include "dpu_hw_ctl.h" +#include "dpu_hw_lm.h" +#include "dpu_hw_interrupts.h" +#include "dpu_hw_top.h" +#include "dpu_rm.h" +#include "dpu_core_perf.h" + +#define DRMID(x) ((x) ? (x)->base.id : -1) + +/** + * DPU_DEBUG - macro for kms/plane/crtc/encoder/connector logs + * @fmt: Pointer to format string + */ +#define DPU_DEBUG(fmt, ...) \ + DRM_DEBUG_DRIVER(fmt, ##__VA_ARGS__) + +/** + * DPU_DEBUG_DRIVER - macro for hardware driver logging + * @fmt: Pointer to format string + */ +#define DPU_DEBUG_DRIVER(fmt, ...) \ + DRM_DEBUG_DRIVER(fmt, ##__VA_ARGS__) + +#define DPU_ERROR(fmt, ...) pr_err("[dpu error]" fmt, ##__VA_ARGS__) +#define DPU_ERROR_RATELIMITED(fmt, ...) pr_err_ratelimited("[dpu error]" fmt, ##__VA_ARGS__) + +/** + * ktime_compare_safe - compare two ktime structures + * This macro is similar to the standard ktime_compare() function, but + * attempts to also handle ktime overflows. + * @A: First ktime value + * @B: Second ktime value + * Returns: -1 if A < B, 0 if A == B, 1 if A > B + */ +#define ktime_compare_safe(A, B) \ + ktime_compare(ktime_sub((A), (B)), ktime_set(0, 0)) + +extern bool dpu_use_virtual_planes; + +struct dpu_kms { + struct msm_kms base; + struct drm_device *dev; + const struct dpu_mdss_cfg *catalog; + const struct qcom_ubwc_cfg_data *mdss; + + /* io/register spaces: */ + void __iomem *mmio, *vbif; + + struct regulator *vdd; + struct regulator *mmagic; + struct regulator *venus; + + struct dpu_hw_intr *hw_intr; + + struct dpu_core_perf perf; + + /* + * Global private object state, Do not access directly, use + * dpu_kms_global_get_state() + */ + struct drm_private_obj global_state; + + struct dpu_rm rm; + + struct dpu_hw_vbif *hw_vbif; + struct dpu_hw_mdp *hw_mdp; + + bool has_danger_ctrl; + + struct platform_device *pdev; + bool rpm_enabled; + + struct clk_bulk_data *clocks; + size_t num_clocks; + + /* reference count bandwidth requests, so we know when we can + * release bandwidth. Each atomic update increments, and frame- + * done event decrements. Additionally, for video mode, the + * reference is incremented when crtc is enabled, and decremented + * when disabled. + */ + atomic_t bandwidth_ref; + struct icc_path *path[2]; + u32 num_paths; +}; + +struct vsync_info { + u32 frame_count; + u32 line_count; +}; + +#define DPU_ENC_WR_PTR_START_TIMEOUT_US 20000 + +#define DPU_ENC_MAX_POLL_TIMEOUT_US 2000 + +#define to_dpu_kms(x) container_of(x, struct dpu_kms, base) + +#define to_dpu_global_state(x) container_of(x, struct dpu_global_state, base) + +/* Global private object state for tracking resources that are shared across + * multiple kms objects (planes/crtcs/etc). + */ +struct dpu_global_state { + struct drm_private_state base; + + struct dpu_rm *rm; + + uint32_t pingpong_to_crtc_id[PINGPONG_MAX - PINGPONG_0]; + uint32_t mixer_to_crtc_id[LM_MAX - LM_0]; + uint32_t ctl_to_crtc_id[CTL_MAX - CTL_0]; + uint32_t dspp_to_crtc_id[DSPP_MAX - DSPP_0]; + uint32_t dsc_to_crtc_id[DSC_MAX - DSC_0]; + uint32_t cdm_to_crtc_id; + + uint32_t sspp_to_crtc_id[SSPP_MAX - SSPP_NONE]; + uint32_t cwb_to_crtc_id[CWB_MAX - CWB_0]; +}; + +struct dpu_global_state + *dpu_kms_get_existing_global_state(struct dpu_kms *dpu_kms); +struct dpu_global_state + *__must_check dpu_kms_get_global_state(struct drm_atomic_commit *s); + +/** + * Debugfs functions - extra helper functions for debugfs support + * + * Main debugfs documentation is located at, + * + * Documentation/filesystems/debugfs.rst + * + * @dpu_debugfs_create_regset32: Create 32-bit register dump file + */ + +void dpu_debugfs_create_regset32(const char *name, umode_t mode, + void *parent, + uint32_t offset, uint32_t length, struct dpu_kms *dpu_kms); + +/** + * DPU info management functions + * These functions/definitions allow for building up a 'dpu_info' structure + * containing one or more "key=value\n" entries. + */ +#define DPU_KMS_INFO_MAX_SIZE 4096 + +/** + * Vblank enable/disable functions + */ +int dpu_enable_vblank(struct msm_kms *kms, struct drm_crtc *crtc); +void dpu_disable_vblank(struct msm_kms *kms, struct drm_crtc *crtc); + +unsigned long dpu_kms_get_clk_rate(struct dpu_kms *dpu_kms, char *clock_name); + +#endif /* __dpu_kms_H__ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c new file mode 100644 index 000000000..7b92082d3 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c @@ -0,0 +1,1963 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2014-2018 The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ + +#include +#include + +#include +#include +#include +#include +#include +#include + +#include + +#include "msm_drv.h" +#include "dpu_kms.h" +#include "dpu_hw_sspp.h" +#include "dpu_hw_util.h" +#include "dpu_trace.h" +#include "dpu_crtc.h" +#include "dpu_vbif.h" +#include "dpu_plane.h" + +#define DPU_DEBUG_PLANE(pl, fmt, ...) DRM_DEBUG_ATOMIC("plane%d " fmt,\ + (pl) ? (pl)->base.base.id : -1, ##__VA_ARGS__) + +#define DPU_ERROR_PLANE(pl, fmt, ...) DPU_ERROR("plane%d " fmt,\ + (pl) ? (pl)->base.base.id : -1, ##__VA_ARGS__) + +#define DECIMATED_DIMENSION(dim, deci) (((dim) + ((1 << (deci)) - 1)) >> (deci)) +#define PHASE_STEP_SHIFT 21 +#define PHASE_STEP_UNIT_SCALE ((int) (1 << PHASE_STEP_SHIFT)) +#define PHASE_RESIDUAL 15 + +#define SHARP_STRENGTH_DEFAULT 32 +#define SHARP_EDGE_THR_DEFAULT 112 +#define SHARP_SMOOTH_THR_DEFAULT 8 +#define SHARP_NOISE_THR_DEFAULT 2 + +#define DPU_PLANE_COLOR_FILL_FLAG BIT(31) +#define DPU_ZPOS_MAX 255 + +/* + * Default Preload Values + */ +#define DPU_QSEED3_DEFAULT_PRELOAD_H 0x4 +#define DPU_QSEED3_DEFAULT_PRELOAD_V 0x3 +#define DPU_QSEED4_DEFAULT_PRELOAD_V 0x2 +#define DPU_QSEED4_DEFAULT_PRELOAD_H 0x4 + +#define DEFAULT_REFRESH_RATE 60 + +static const uint32_t qcom_compressed_supported_formats[] = { + DRM_FORMAT_ABGR8888, + DRM_FORMAT_ARGB8888, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_ARGB2101010, + DRM_FORMAT_XRGB2101010, + DRM_FORMAT_BGR565, + + DRM_FORMAT_NV12, + DRM_FORMAT_P010, +}; + +/* + * struct dpu_plane - local dpu plane structure + * @vm: address space pointer + * @csc_ptr: Points to dpu_csc_cfg structure to use for current + * @catalog: Points to dpu catalog structure + * @revalidate: force revalidation of all the plane properties + */ +struct dpu_plane { + struct drm_plane base; + + enum dpu_sspp pipe; + + uint32_t color_fill; + bool is_error; + bool is_rt_pipe; + const struct dpu_mdss_cfg *catalog; +}; + +static const uint64_t supported_format_modifiers[] = { + DRM_FORMAT_MOD_QCOM_COMPRESSED, + DRM_FORMAT_MOD_LINEAR, + DRM_FORMAT_MOD_INVALID +}; + +#define to_dpu_plane(x) container_of(x, struct dpu_plane, base) + +static struct dpu_kms *_dpu_plane_get_kms(struct drm_plane *plane) +{ + struct msm_drm_private *priv = plane->dev->dev_private; + + return to_dpu_kms(priv->kms); +} + +/** + * _dpu_plane_calc_bw - calculate bandwidth required for a plane + * @catalog: Points to dpu catalog structure + * @fmt: Pointer to source buffer format + * @mode: Pointer to drm display mode + * @pipe_cfg: Pointer to pipe configuration + * Result: Updates calculated bandwidth in the plane state. + * BW Equation: src_w * src_h * bpp * fps * (v_total / v_dest) + * Prefill BW Equation: line src bytes * line_time + */ +static u64 _dpu_plane_calc_bw(const struct dpu_mdss_cfg *catalog, + const struct msm_format *fmt, + const struct drm_display_mode *mode, + struct dpu_sw_pipe_cfg *pipe_cfg) +{ + int src_width, src_height, dst_height, fps; + u64 plane_pixel_rate, plane_bit_rate; + u64 plane_prefill_bw; + u64 plane_bw; + u32 hw_latency_lines; + u64 scale_factor; + int vbp, vpw, vfp; + + src_width = drm_rect_width(&pipe_cfg->src_rect); + src_height = drm_rect_height(&pipe_cfg->src_rect); + dst_height = drm_rect_height(&pipe_cfg->dst_rect); + fps = drm_mode_vrefresh(mode); + vbp = mode->vtotal - mode->vsync_end; + vpw = mode->vsync_end - mode->vsync_start; + vfp = mode->vsync_start - mode->vdisplay; + hw_latency_lines = catalog->perf->min_prefill_lines; + scale_factor = src_height > dst_height ? + mult_frac(src_height, 1, dst_height) : 1; + + plane_pixel_rate = src_width * mode->vtotal * fps; + plane_bit_rate = plane_pixel_rate * fmt->bpp; + + plane_bw = plane_bit_rate * scale_factor; + + plane_prefill_bw = plane_bw * hw_latency_lines; + + if ((vbp+vpw) > hw_latency_lines) + do_div(plane_prefill_bw, (vbp+vpw)); + else if ((vbp+vpw+vfp) < hw_latency_lines) + do_div(plane_prefill_bw, (vbp+vpw+vfp)); + else + do_div(plane_prefill_bw, hw_latency_lines); + + + return max(plane_bw, plane_prefill_bw); +} + +/** + * _dpu_plane_calc_clk - calculate clock required for a plane + * @mode: Pointer to drm display mode + * @pipe_cfg: Pointer to pipe configuration + * Result: Updates calculated clock in the plane state. + * Clock equation: dst_w * v_total * fps * (src_h / dst_h) + */ +static u64 _dpu_plane_calc_clk(const struct drm_display_mode *mode, + struct dpu_sw_pipe_cfg *pipe_cfg) +{ + int dst_width, src_height, dst_height, fps; + u64 plane_clk; + + src_height = drm_rect_height(&pipe_cfg->src_rect); + dst_width = drm_rect_width(&pipe_cfg->dst_rect); + dst_height = drm_rect_height(&pipe_cfg->dst_rect); + fps = drm_mode_vrefresh(mode); + + plane_clk = + dst_width * mode->vtotal * fps; + + if (src_height > dst_height) { + plane_clk *= src_height; + do_div(plane_clk, dst_height); + } + + return plane_clk; +} + +/** + * _dpu_plane_calc_fill_level - calculate fill level of the given source format + * @plane: Pointer to drm plane + * @pipe: Pointer to software pipe + * @lut_usage: LUT usecase + * @fmt: Pointer to source buffer format + * @src_width: width of source buffer + * Return: fill level corresponding to the source buffer/format or 0 if error + */ +static int _dpu_plane_calc_fill_level(struct drm_plane *plane, + struct dpu_sw_pipe *pipe, + enum dpu_qos_lut_usage lut_usage, + const struct msm_format *fmt, u32 src_width) +{ + struct dpu_plane *pdpu; + u32 fixed_buff_size; + u32 total_fl; + + if (!fmt || !pipe || !src_width || !fmt->bpp) { + DPU_ERROR("invalid arguments\n"); + return 0; + } + + if (lut_usage == DPU_QOS_LUT_USAGE_NRT) + return 0; + + pdpu = to_dpu_plane(plane); + fixed_buff_size = pdpu->catalog->caps->pixel_ram_size; + + /* FIXME: in multirect case account for the src_width of all the planes */ + + if (fmt->fetch_type == MDP_PLANE_PSEUDO_PLANAR) { + if (fmt->chroma_sample == CHROMA_420) { + /* NV12 */ + total_fl = (fixed_buff_size / 2) / + ((src_width + 32) * fmt->bpp); + } else { + /* non NV12 */ + total_fl = (fixed_buff_size / 2) * 2 / + ((src_width + 32) * fmt->bpp); + } + } else { + if (pipe->multirect_mode == DPU_SSPP_MULTIRECT_PARALLEL) { + total_fl = (fixed_buff_size / 2) * 2 / + ((src_width + 32) * fmt->bpp); + } else { + total_fl = (fixed_buff_size) * 2 / + ((src_width + 32) * fmt->bpp); + } + } + + DPU_DEBUG_PLANE(pdpu, "pnum:%d fmt: %p4cc w:%u fl:%u\n", + pipe->sspp->idx - SSPP_VIG0, + &fmt->pixel_format, + src_width, total_fl); + + return total_fl; +} + +/** + * _dpu_plane_set_qos_lut - set QoS LUT of the given plane + * @plane: Pointer to drm plane + * @pipe: Pointer to software pipe + * @fmt: Pointer to source buffer format + * @pipe_cfg: Pointer to pipe configuration + */ +static void _dpu_plane_set_qos_lut(struct drm_plane *plane, + struct dpu_sw_pipe *pipe, + const struct msm_format *fmt, struct dpu_sw_pipe_cfg *pipe_cfg) +{ + struct dpu_plane *pdpu = to_dpu_plane(plane); + struct dpu_hw_qos_cfg cfg; + u32 total_fl, lut_usage; + + if (!pdpu->is_rt_pipe) { + lut_usage = DPU_QOS_LUT_USAGE_NRT; + } else { + if (fmt && MSM_FORMAT_IS_LINEAR(fmt)) + lut_usage = DPU_QOS_LUT_USAGE_LINEAR; + else + lut_usage = DPU_QOS_LUT_USAGE_MACROTILE; + } + + total_fl = _dpu_plane_calc_fill_level(plane, pipe, lut_usage, fmt, + drm_rect_width(&pipe_cfg->src_rect)); + + cfg.creq_lut = _dpu_hw_get_qos_lut(&pdpu->catalog->perf->qos_lut_tbl[lut_usage], total_fl); + cfg.danger_lut = pdpu->catalog->perf->danger_lut_tbl[lut_usage]; + cfg.safe_lut = pdpu->catalog->perf->safe_lut_tbl[lut_usage]; + + if (pipe->sspp->idx != SSPP_CURSOR0 && + pipe->sspp->idx != SSPP_CURSOR1 && + pdpu->is_rt_pipe) + cfg.danger_safe_en = true; + + DPU_DEBUG_PLANE(pdpu, "pnum:%d ds:%d is_rt:%d\n", + pdpu->pipe - SSPP_VIG0, + cfg.danger_safe_en, + pdpu->is_rt_pipe); + + trace_dpu_perf_set_qos_luts(pipe->sspp->idx - SSPP_VIG0, + (fmt) ? fmt->pixel_format : 0, + pdpu->is_rt_pipe, total_fl, cfg.creq_lut, lut_usage); + + DPU_DEBUG_PLANE(pdpu, "pnum:%d fmt: %p4cc rt:%d fl:%u lut:0x%llx\n", + pdpu->pipe - SSPP_VIG0, + fmt ? &fmt->pixel_format : NULL, + pdpu->is_rt_pipe, total_fl, cfg.creq_lut); + + trace_dpu_perf_set_danger_luts(pdpu->pipe - SSPP_VIG0, + (fmt) ? fmt->pixel_format : 0, + (fmt) ? fmt->fetch_mode : 0, + cfg.danger_lut, + cfg.safe_lut); + + DPU_DEBUG_PLANE(pdpu, "pnum:%d fmt: %p4cc mode:%d luts[0x%x, 0x%x]\n", + pdpu->pipe - SSPP_VIG0, + fmt ? &fmt->pixel_format : NULL, + fmt ? fmt->fetch_mode : -1, + cfg.danger_lut, + cfg.safe_lut); + + pipe->sspp->ops.setup_qos_lut(pipe->sspp, &cfg); +} + +/** + * _dpu_plane_set_qos_ctrl - set QoS control of the given plane + * @plane: Pointer to drm plane + * @pipe: Pointer to software pipe + * @enable: true to enable QoS control + */ +static void _dpu_plane_set_qos_ctrl(struct drm_plane *plane, + struct dpu_sw_pipe *pipe, + bool enable) +{ + struct dpu_plane *pdpu = to_dpu_plane(plane); + + if (!pdpu->is_rt_pipe) + enable = false; + + DPU_DEBUG_PLANE(pdpu, "pnum:%d ds:%d is_rt:%d\n", + pdpu->pipe - SSPP_VIG0, + enable, + pdpu->is_rt_pipe); + + pipe->sspp->ops.setup_qos_ctrl(pipe->sspp, + enable); +} + +static bool _dpu_plane_sspp_clk_force_ctrl(struct dpu_hw_sspp *sspp, + struct dpu_hw_mdp *mdp, + bool enable, bool *forced_on) +{ + if (sspp->ops.setup_clk_force_ctrl) { + *forced_on = sspp->ops.setup_clk_force_ctrl(sspp, enable); + return true; + } + + if (mdp->ops.setup_clk_force_ctrl) { + *forced_on = mdp->ops.setup_clk_force_ctrl(mdp, sspp->cap->clk_ctrl, enable); + return true; + } + + return false; +} + +/** + * _dpu_plane_set_ot_limit - set OT limit for the given plane + * @plane: Pointer to drm plane + * @pipe: Pointer to software pipe + * @pipe_cfg: Pointer to pipe configuration + * @frame_rate: CRTC's frame rate + */ +static void _dpu_plane_set_ot_limit(struct drm_plane *plane, + struct dpu_sw_pipe *pipe, + struct dpu_sw_pipe_cfg *pipe_cfg, + int frame_rate) +{ + struct dpu_plane *pdpu = to_dpu_plane(plane); + struct dpu_vbif_set_ot_params ot_params; + struct dpu_kms *dpu_kms = _dpu_plane_get_kms(plane); + bool forced_on = false; + + memset(&ot_params, 0, sizeof(ot_params)); + ot_params.xin_id = pipe->sspp->cap->xin_id; + ot_params.num = pipe->sspp->idx - SSPP_NONE; + ot_params.width = drm_rect_width(&pipe_cfg->src_rect); + ot_params.height = drm_rect_height(&pipe_cfg->src_rect); + ot_params.is_wfd = !pdpu->is_rt_pipe; + ot_params.frame_rate = frame_rate; + ot_params.rd = true; + + if (!_dpu_plane_sspp_clk_force_ctrl(pipe->sspp, dpu_kms->hw_mdp, + true, &forced_on)) + return; + + dpu_vbif_set_ot_limit(dpu_kms, &ot_params); + + if (forced_on) + _dpu_plane_sspp_clk_force_ctrl(pipe->sspp, dpu_kms->hw_mdp, + false, &forced_on); +} + +/** + * _dpu_plane_set_qos_remap - set vbif QoS for the given plane + * @plane: Pointer to drm plane + * @pipe: Pointer to software pipe + */ +static void _dpu_plane_set_qos_remap(struct drm_plane *plane, + struct dpu_sw_pipe *pipe) +{ + struct dpu_plane *pdpu = to_dpu_plane(plane); + struct dpu_vbif_set_qos_params qos_params; + struct dpu_kms *dpu_kms = _dpu_plane_get_kms(plane); + bool forced_on = false; + + memset(&qos_params, 0, sizeof(qos_params)); + qos_params.xin_id = pipe->sspp->cap->xin_id; + qos_params.num = pipe->sspp->idx - SSPP_VIG0; + qos_params.is_rt = pdpu->is_rt_pipe; + + DPU_DEBUG_PLANE(pdpu, "pipe:%d xin:%d rt:%d\n", + qos_params.num, + qos_params.xin_id, qos_params.is_rt); + + if (!_dpu_plane_sspp_clk_force_ctrl(pipe->sspp, dpu_kms->hw_mdp, + true, &forced_on)) + return; + + dpu_vbif_set_qos_remap(dpu_kms, &qos_params); + + if (forced_on) + _dpu_plane_sspp_clk_force_ctrl(pipe->sspp, dpu_kms->hw_mdp, + false, &forced_on); +} + +static void _dpu_plane_setup_scaler3(struct dpu_hw_sspp *pipe_hw, + uint32_t src_w, uint32_t src_h, uint32_t dst_w, uint32_t dst_h, + struct dpu_hw_scaler3_cfg *scale_cfg, + const struct msm_format *fmt, + uint32_t chroma_subsmpl_h, uint32_t chroma_subsmpl_v, + unsigned int rotation) +{ + uint32_t i; + bool inline_rotation = rotation & DRM_MODE_ROTATE_90; + + /* + * For inline rotation cases, scaler config is post-rotation, + * so swap the dimensions here. However, pixel extension will + * need pre-rotation settings. + */ + if (inline_rotation) + swap(src_w, src_h); + + scale_cfg->phase_step_x[DPU_SSPP_COMP_0] = + mult_frac((1 << PHASE_STEP_SHIFT), src_w, dst_w); + scale_cfg->phase_step_y[DPU_SSPP_COMP_0] = + mult_frac((1 << PHASE_STEP_SHIFT), src_h, dst_h); + + + scale_cfg->phase_step_y[DPU_SSPP_COMP_1_2] = + scale_cfg->phase_step_y[DPU_SSPP_COMP_0] / chroma_subsmpl_v; + scale_cfg->phase_step_x[DPU_SSPP_COMP_1_2] = + scale_cfg->phase_step_x[DPU_SSPP_COMP_0] / chroma_subsmpl_h; + + scale_cfg->phase_step_x[DPU_SSPP_COMP_2] = + scale_cfg->phase_step_x[DPU_SSPP_COMP_1_2]; + scale_cfg->phase_step_y[DPU_SSPP_COMP_2] = + scale_cfg->phase_step_y[DPU_SSPP_COMP_1_2]; + + scale_cfg->phase_step_x[DPU_SSPP_COMP_3] = + scale_cfg->phase_step_x[DPU_SSPP_COMP_0]; + scale_cfg->phase_step_y[DPU_SSPP_COMP_3] = + scale_cfg->phase_step_y[DPU_SSPP_COMP_0]; + + for (i = 0; i < DPU_MAX_PLANES; i++) { + scale_cfg->src_width[i] = src_w; + scale_cfg->src_height[i] = src_h; + if (i == DPU_SSPP_COMP_1_2 || i == DPU_SSPP_COMP_2) { + scale_cfg->src_width[i] /= chroma_subsmpl_h; + scale_cfg->src_height[i] /= chroma_subsmpl_v; + } + + if (pipe_hw->cap->sblk->scaler_blk.version >= 0x3000) { + scale_cfg->preload_x[i] = DPU_QSEED4_DEFAULT_PRELOAD_H; + scale_cfg->preload_y[i] = DPU_QSEED4_DEFAULT_PRELOAD_V; + } else { + scale_cfg->preload_x[i] = DPU_QSEED3_DEFAULT_PRELOAD_H; + scale_cfg->preload_y[i] = DPU_QSEED3_DEFAULT_PRELOAD_V; + } + } + if (!(MSM_FORMAT_IS_YUV(fmt)) && (src_h == dst_h) + && (src_w == dst_w)) + return; + + scale_cfg->dst_width = dst_w; + scale_cfg->dst_height = dst_h; + scale_cfg->y_rgb_filter_cfg = DPU_SCALE_BIL; + scale_cfg->uv_filter_cfg = DPU_SCALE_BIL; + scale_cfg->alpha_filter_cfg = DPU_SCALE_ALPHA_BIL; + scale_cfg->lut_flag = 0; + scale_cfg->blend_cfg = 1; + scale_cfg->enable = 1; +} + +static void _dpu_plane_setup_pixel_ext(struct dpu_hw_scaler3_cfg *scale_cfg, + struct dpu_hw_pixel_ext *pixel_ext, + uint32_t src_w, uint32_t src_h, + uint32_t chroma_subsmpl_h, uint32_t chroma_subsmpl_v) +{ + int i; + + for (i = 0; i < DPU_MAX_PLANES; i++) { + uint32_t w = src_w, h = src_h; + + if (i == DPU_SSPP_COMP_1_2 || i == DPU_SSPP_COMP_2) { + w /= chroma_subsmpl_h; + h /= chroma_subsmpl_v; + } + + pixel_ext->num_ext_pxls_top[i] = h; + pixel_ext->num_ext_pxls_left[i] = w; + } +} + +static const struct dpu_csc_cfg *_dpu_plane_get_csc(struct dpu_sw_pipe *pipe, + const struct msm_format *fmt) +{ + const struct dpu_csc_cfg *csc_ptr; + + if (!MSM_FORMAT_IS_YUV(fmt)) + return NULL; + + if (BIT(DPU_SSPP_CSC_10BIT) & pipe->sspp->cap->features) + csc_ptr = &dpu_csc10_YUV2RGB_601L; + else + csc_ptr = &dpu_csc_YUV2RGB_601L; + + return csc_ptr; +} + +static void _dpu_plane_setup_scaler(struct dpu_sw_pipe *pipe, + const struct msm_format *fmt, bool color_fill, + struct dpu_sw_pipe_cfg *pipe_cfg) +{ + struct dpu_hw_sspp *pipe_hw = pipe->sspp; + const struct drm_format_info *info = drm_format_info(fmt->pixel_format); + struct dpu_hw_scaler3_cfg scaler3_cfg; + struct dpu_hw_pixel_ext pixel_ext; + u32 src_width = drm_rect_width(&pipe_cfg->src_rect); + u32 src_height = drm_rect_height(&pipe_cfg->src_rect); + u32 dst_width = drm_rect_width(&pipe_cfg->dst_rect); + u32 dst_height = drm_rect_height(&pipe_cfg->dst_rect); + + memset(&scaler3_cfg, 0, sizeof(scaler3_cfg)); + memset(&pixel_ext, 0, sizeof(pixel_ext)); + + /* don't chroma subsample if decimating */ + /* update scaler. calculate default config for QSEED3 */ + _dpu_plane_setup_scaler3(pipe_hw, + src_width, + src_height, + dst_width, + dst_height, + &scaler3_cfg, fmt, + info->hsub, info->vsub, + pipe_cfg->rotation); + + /* configure pixel extension based on scalar config */ + _dpu_plane_setup_pixel_ext(&scaler3_cfg, &pixel_ext, + src_width, src_height, info->hsub, info->vsub); + + if (pipe_hw->ops.setup_pe) + pipe_hw->ops.setup_pe(pipe_hw, + &pixel_ext); + + /** + * when programmed in multirect mode, scalar block will be + * bypassed. Still we need to update alpha and bitwidth + * ONLY for RECT0 + */ + if (pipe_hw->ops.setup_scaler && + pipe->multirect_index != DPU_SSPP_RECT_1) + pipe_hw->ops.setup_scaler(pipe_hw, + &scaler3_cfg, + fmt); +} + +static void _dpu_plane_color_fill_pipe(struct dpu_plane_state *pstate, + struct dpu_sw_pipe *pipe, + struct drm_rect *dst_rect, + u32 fill_color, + const struct msm_format *fmt) +{ + struct dpu_sw_pipe_cfg pipe_cfg; + + /* update sspp */ + if (!pipe->sspp->ops.setup_solidfill) + return; + + pipe->sspp->ops.setup_solidfill(pipe, fill_color); + + /* override scaler/decimation if solid fill */ + pipe_cfg.dst_rect = *dst_rect; + + pipe_cfg.src_rect.x1 = 0; + pipe_cfg.src_rect.y1 = 0; + pipe_cfg.src_rect.x2 = + drm_rect_width(&pipe_cfg.dst_rect); + pipe_cfg.src_rect.y2 = + drm_rect_height(&pipe_cfg.dst_rect); + + if (pipe->sspp->ops.setup_format) + pipe->sspp->ops.setup_format(pipe, fmt, DPU_SSPP_SOLID_FILL); + + if (pipe->sspp->ops.setup_rects) + pipe->sspp->ops.setup_rects(pipe, &pipe_cfg); + + _dpu_plane_setup_scaler(pipe, fmt, true, &pipe_cfg); +} + +/** + * _dpu_plane_color_fill - enables color fill on plane + * @pdpu: Pointer to DPU plane object + * @color: RGB fill color value, [23..16] Blue, [15..8] Green, [7..0] Red + * @alpha: 8-bit fill alpha value, 255 selects 100% alpha + */ +static void _dpu_plane_color_fill(struct dpu_plane *pdpu, + uint32_t color, uint32_t alpha) +{ + const struct msm_format *fmt; + const struct drm_plane *plane = &pdpu->base; + struct msm_drm_private *priv = plane->dev->dev_private; + struct dpu_plane_state *pstate = to_dpu_plane_state(plane->state); + u32 fill_color = (color & 0xFFFFFF) | ((alpha & 0xFF) << 24); + int i; + + DPU_DEBUG_PLANE(pdpu, "\n"); + + /* + * select fill format to match user property expectation, + * h/w only supports RGB variants + */ + fmt = mdp_get_format(priv->kms, DRM_FORMAT_ABGR8888, 0); + /* should not happen ever */ + if (!fmt) + return; + + /* update sspp */ + for (i = 0; i < PIPES_PER_PLANE; i++) { + if (!pstate->pipe[i].sspp) + continue; + _dpu_plane_color_fill_pipe(pstate, &pstate->pipe[i], + &pstate->pipe_cfg[i].dst_rect, + fill_color, fmt); + } +} + +static int dpu_plane_prepare_fb(struct drm_plane *plane, + struct drm_plane_state *new_state) +{ + struct drm_framebuffer *fb = new_state->fb; + struct dpu_plane *pdpu = to_dpu_plane(plane); + struct dpu_plane_state *pstate = to_dpu_plane_state(new_state); + int ret; + + if (!new_state->fb) + return 0; + + DPU_DEBUG_PLANE(pdpu, "FB[%u]\n", fb->base.id); + + /* + * TODO: Need to sort out the msm_framebuffer_prepare() call below so + * we can use msm_atomic_prepare_fb() instead of doing the + * implicit fence and fb prepare by hand here. + */ + drm_gem_plane_helper_prepare_fb(plane, new_state); + + ret = msm_framebuffer_prepare(new_state->fb, pstate->needs_dirtyfb); + if (ret) { + DPU_ERROR("failed to prepare framebuffer\n"); + return ret; + } + + return 0; +} + +static void dpu_plane_cleanup_fb(struct drm_plane *plane, + struct drm_plane_state *old_state) +{ + struct dpu_plane *pdpu = to_dpu_plane(plane); + struct dpu_plane_state *old_pstate; + + if (!old_state || !old_state->fb) + return; + + old_pstate = to_dpu_plane_state(old_state); + + DPU_DEBUG_PLANE(pdpu, "FB[%u]\n", old_state->fb->base.id); + + msm_framebuffer_cleanup(old_state->fb, old_pstate->needs_dirtyfb); +} + +static int dpu_plane_check_inline_rotation(struct dpu_plane *pdpu, + struct dpu_sw_pipe *pipe, + struct drm_rect src, + const struct msm_format *fmt) +{ + const struct dpu_sspp_sub_blks *sblk = pipe->sspp->cap->sblk; + size_t num_formats; + const u32 *supported_formats; + + if (!test_bit(DPU_SSPP_INLINE_ROTATION, &pipe->sspp->cap->features)) + return -EINVAL; + + if (!sblk->rotation_cfg) { + DPU_ERROR("invalid rotation cfg\n"); + return -EINVAL; + } + + if (drm_rect_width(&src) > sblk->rotation_cfg->rot_maxheight) { + DPU_DEBUG_PLANE(pdpu, "invalid height for inline rot:%d max:%d\n", + src.y2, sblk->rotation_cfg->rot_maxheight); + return -EINVAL; + } + + supported_formats = sblk->rotation_cfg->rot_format_list; + num_formats = sblk->rotation_cfg->rot_num_formats; + + if (!MSM_FORMAT_IS_UBWC(fmt) || + !dpu_find_format(fmt->pixel_format, supported_formats, num_formats)) + return -EINVAL; + + return 0; +} + +static int dpu_plane_atomic_check_pipe(struct dpu_plane *pdpu, + struct dpu_sw_pipe *pipe, + struct dpu_sw_pipe_cfg *pipe_cfg, + const struct drm_display_mode *mode, + struct drm_plane_state *new_plane_state) +{ + uint32_t min_src_size; + struct dpu_kms *kms = _dpu_plane_get_kms(&pdpu->base); + int ret; + const struct msm_format *fmt; + uint32_t supported_rotations; + const struct dpu_sspp_cfg *pipe_hw_caps; + const struct dpu_sspp_sub_blks *sblk; + + pipe_hw_caps = pipe->sspp->cap; + sblk = pipe->sspp->cap->sblk; + + /* + * We already have verified scaling against platform limitations. + * Now check if the SSPP supports scaling at all. + */ + if (!(sblk->scaler_blk.len && pipe->sspp->ops.setup_scaler) && + ((drm_rect_width(&new_plane_state->src) >> 16 != + drm_rect_width(&new_plane_state->dst)) || + (drm_rect_height(&new_plane_state->src) >> 16 != + drm_rect_height(&new_plane_state->dst)))) + return -ERANGE; + + fmt = msm_framebuffer_format(new_plane_state->fb); + + supported_rotations = DRM_MODE_REFLECT_MASK | DRM_MODE_ROTATE_0; + + if (pipe_hw_caps->features & BIT(DPU_SSPP_INLINE_ROTATION)) + supported_rotations |= DRM_MODE_ROTATE_90; + + pipe_cfg->rotation = drm_rotation_simplify(new_plane_state->rotation, + supported_rotations); + + min_src_size = MSM_FORMAT_IS_YUV(fmt) ? 2 : 1; + + if (MSM_FORMAT_IS_YUV(fmt) && + !pipe->sspp->cap->sblk->csc_blk.len) { + DPU_DEBUG_PLANE(pdpu, + "plane doesn't have csc for yuv\n"); + return -EINVAL; + } + + /* check src bounds */ + if (drm_rect_width(&pipe_cfg->src_rect) < min_src_size || + drm_rect_height(&pipe_cfg->src_rect) < min_src_size) { + DPU_DEBUG_PLANE(pdpu, "invalid source " DRM_RECT_FMT "\n", + DRM_RECT_ARG(&pipe_cfg->src_rect)); + return -E2BIG; + } + + /* valid yuv image */ + if (MSM_FORMAT_IS_YUV(fmt) && + (pipe_cfg->src_rect.x1 & 0x1 || + pipe_cfg->src_rect.y1 & 0x1 || + drm_rect_width(&pipe_cfg->src_rect) & 0x1 || + drm_rect_height(&pipe_cfg->src_rect) & 0x1)) { + DPU_DEBUG_PLANE(pdpu, "invalid yuv source " DRM_RECT_FMT "\n", + DRM_RECT_ARG(&pipe_cfg->src_rect)); + return -EINVAL; + } + + /* min dst support */ + if (drm_rect_width(&pipe_cfg->dst_rect) < 0x1 || + drm_rect_height(&pipe_cfg->dst_rect) < 0x1) { + DPU_DEBUG_PLANE(pdpu, "invalid dest rect " DRM_RECT_FMT "\n", + DRM_RECT_ARG(&pipe_cfg->dst_rect)); + return -EINVAL; + } + + if (pipe_cfg->rotation & DRM_MODE_ROTATE_90) { + ret = dpu_plane_check_inline_rotation(pdpu, pipe, pipe_cfg->src_rect, fmt); + if (ret) + return ret; + } + + /* max clk check */ + if (_dpu_plane_calc_clk(mode, pipe_cfg) > kms->perf.max_core_clk_rate) { + DPU_DEBUG_PLANE(pdpu, "plane exceeds max mdp core clk limits\n"); + return -E2BIG; + } + + return 0; +} + +#define MAX_UPSCALE_RATIO 20 +#define MAX_DOWNSCALE_RATIO 4 + +static int dpu_plane_atomic_check_nosspp(struct drm_plane *plane, + struct drm_plane_state *new_plane_state, + const struct drm_crtc_state *crtc_state) +{ + int i, ret = 0, min_scale, max_scale; + struct dpu_plane *pdpu = to_dpu_plane(plane); + struct dpu_plane_state *pstate = to_dpu_plane_state(new_plane_state); + struct drm_rect fb_rect = { 0 }; + + min_scale = FRAC_16_16(1, MAX_UPSCALE_RATIO); + max_scale = MAX_DOWNSCALE_RATIO << 16; + + ret = drm_atomic_helper_check_plane_state(new_plane_state, crtc_state, + min_scale, + max_scale, + true, true); + if (ret) { + DPU_DEBUG_PLANE(pdpu, "Check plane state failed (%d)\n", ret); + return ret; + } + if (!new_plane_state->visible) + return 0; + + pstate->stage = DPU_STAGE_0 + pstate->base.normalized_zpos; + if (pstate->stage >= pdpu->catalog->caps->max_mixer_blendstages) { + DPU_ERROR("> %d plane stages assigned\n", + pdpu->catalog->caps->max_mixer_blendstages - DPU_STAGE_0); + return -EINVAL; + } + + fb_rect.x2 = new_plane_state->fb->width; + fb_rect.y2 = new_plane_state->fb->height; + + /* Ensure fb size is supported */ + if (drm_rect_width(&fb_rect) > DPU_MAX_IMG_WIDTH || + drm_rect_height(&fb_rect) > DPU_MAX_IMG_HEIGHT) { + DPU_DEBUG_PLANE(pdpu, "invalid framebuffer " DRM_RECT_FMT "\n", + DRM_RECT_ARG(&fb_rect)); + return -E2BIG; + } + + ret = dpu_format_populate_plane_sizes(new_plane_state->fb, &pstate->layout); + if (ret) { + DPU_ERROR_PLANE(pdpu, "failed to get format plane sizes, %d\n", ret); + return ret; + } + + for (i = 0; i < pstate->layout.num_planes; i++) + if (pstate->layout.plane_pitch[i] > DPU_SSPP_MAX_PITCH_SIZE) + return -E2BIG; + + pstate->needs_qos_remap = drm_atomic_crtc_needs_modeset(crtc_state); + + return 0; +} + +static int dpu_plane_split(struct drm_plane *plane, + struct drm_plane_state *new_plane_state, + const struct drm_crtc_state *crtc_state) +{ + struct dpu_plane *pdpu = to_dpu_plane(plane); + struct dpu_kms *kms = _dpu_plane_get_kms(&pdpu->base); + u64 max_mdp_clk_rate = kms->perf.max_core_clk_rate; + struct dpu_plane_state *pstate = to_dpu_plane_state(new_plane_state); + struct dpu_sw_pipe_cfg *pipe_cfg; + struct dpu_sw_pipe_cfg *r_pipe_cfg; + const struct drm_display_mode *mode = &crtc_state->adjusted_mode; + uint32_t max_linewidth; + u32 num_lm; + int stage_id, num_stages; + + max_linewidth = pdpu->catalog->caps->max_linewidth; + + /* In non-virtual plane case, one mixer pair is always needed. */ + num_lm = dpu_crtc_get_num_lm(crtc_state); + if (dpu_use_virtual_planes) + num_stages = (num_lm + 1) / 2; + else + num_stages = 1; + + /* + * For wide plane that exceeds SSPP rectangle constrain, it needed to + * be split and mapped to 2 rectangles with 1 config for 2:2:1. + * For 2 interfaces cases, such as dual DSI, 2:2:2 topology is needed. + * If the width or clock exceeds hardware limitation in every half of + * screen, 4:4:2 topology is needed and virtual plane feature should + * be enabled to map plane to more than 1 SSPP. 2 stage configs are + * needed to serve 2 mixer pairs in this 4:4:2 case. So both left/right + * half of plane splitting, and splitting within the half of screen is + * needed in quad-pipe case. Check dest rectangle left/right clipping + * and iterate mixer configs for this plane first, then check wide + * rectangle splitting in every half next. + */ + for (stage_id = 0; stage_id < num_stages; stage_id++) { + struct drm_rect mixer_rect = { + .x1 = stage_id * mode->hdisplay / num_stages, + .y1 = 0, + .x2 = (stage_id + 1) * mode->hdisplay / num_stages, + .y2 = mode->vdisplay + }; + int cfg_idx = stage_id * PIPES_PER_STAGE; + + pipe_cfg = &pstate->pipe_cfg[cfg_idx]; + r_pipe_cfg = &pstate->pipe_cfg[cfg_idx + 1]; + + drm_rect_fp_to_int(&pipe_cfg->src_rect, &new_plane_state->src); + + drm_rect_rotate(&pipe_cfg->src_rect, + new_plane_state->fb->width, new_plane_state->fb->height, + new_plane_state->rotation); + + pipe_cfg->dst_rect = new_plane_state->dst; + + DPU_DEBUG_PLANE(pdpu, "checking src " DRM_RECT_FMT + " vs clip window " DRM_RECT_FMT "\n", + DRM_RECT_ARG(&pipe_cfg->src_rect), + DRM_RECT_ARG(&mixer_rect)); + + /* + * If this plane does not fall into mixer rect, check next + * mixer rect. + */ + if (!drm_rect_clip_scaled(&pipe_cfg->src_rect, + &pipe_cfg->dst_rect, + &mixer_rect)) { + memset(pipe_cfg, 0, 2 * sizeof(struct dpu_sw_pipe_cfg)); + + continue; + } + + pipe_cfg->dst_rect.x1 -= mixer_rect.x1; + pipe_cfg->dst_rect.x2 -= mixer_rect.x1; + + DPU_DEBUG_PLANE(pdpu, "Got clip src:" DRM_RECT_FMT " dst: " DRM_RECT_FMT "\n", + DRM_RECT_ARG(&pipe_cfg->src_rect), DRM_RECT_ARG(&pipe_cfg->dst_rect)); + + /* Split wide rect into 2 rect */ + if ((drm_rect_width(&pipe_cfg->src_rect) > max_linewidth) || + _dpu_plane_calc_clk(mode, pipe_cfg) > max_mdp_clk_rate) { + + if (drm_rect_width(&pipe_cfg->src_rect) > 2 * max_linewidth) { + DPU_DEBUG_PLANE(pdpu, "invalid src " DRM_RECT_FMT " line:%u\n", + DRM_RECT_ARG(&pipe_cfg->src_rect), max_linewidth); + return -E2BIG; + } + + memcpy(r_pipe_cfg, pipe_cfg, sizeof(struct dpu_sw_pipe_cfg)); + pipe_cfg->src_rect.x2 = (pipe_cfg->src_rect.x1 + pipe_cfg->src_rect.x2) >> 1; + pipe_cfg->dst_rect.x2 = (pipe_cfg->dst_rect.x1 + pipe_cfg->dst_rect.x2) >> 1; + r_pipe_cfg->src_rect.x1 = pipe_cfg->src_rect.x2; + r_pipe_cfg->dst_rect.x1 = pipe_cfg->dst_rect.x2; + DPU_DEBUG_PLANE(pdpu, "Split wide plane into:" + DRM_RECT_FMT " and " DRM_RECT_FMT "\n", + DRM_RECT_ARG(&pipe_cfg->src_rect), + DRM_RECT_ARG(&r_pipe_cfg->src_rect)); + } else { + memset(r_pipe_cfg, 0, sizeof(struct dpu_sw_pipe_cfg)); + } + + drm_rect_rotate_inv(&pipe_cfg->src_rect, + new_plane_state->fb->width, + new_plane_state->fb->height, + new_plane_state->rotation); + + if (drm_rect_width(&r_pipe_cfg->src_rect) != 0) + drm_rect_rotate_inv(&r_pipe_cfg->src_rect, + new_plane_state->fb->width, + new_plane_state->fb->height, + new_plane_state->rotation); + } + + return 0; +} + +static int dpu_plane_is_multirect_capable(struct dpu_hw_sspp *sspp, + struct dpu_sw_pipe_cfg *pipe_cfg, + const struct msm_format *fmt) +{ + if (drm_rect_width(&pipe_cfg->src_rect) != drm_rect_width(&pipe_cfg->dst_rect) || + drm_rect_height(&pipe_cfg->src_rect) != drm_rect_height(&pipe_cfg->dst_rect)) + return false; + + if (pipe_cfg->rotation & DRM_MODE_ROTATE_90) + return false; + + if (MSM_FORMAT_IS_YUV(fmt)) + return false; + + if (!sspp) + return true; + + if (!test_bit(DPU_SSPP_SMART_DMA_V1, &sspp->cap->features) && + !test_bit(DPU_SSPP_SMART_DMA_V2, &sspp->cap->features)) + return false; + + return true; +} + +static int dpu_plane_is_parallel_capable(struct dpu_sw_pipe_cfg *pipe_cfg, + const struct msm_format *fmt, + uint32_t max_linewidth) +{ + if (MSM_FORMAT_IS_UBWC(fmt) && + drm_rect_width(&pipe_cfg->src_rect) > max_linewidth / 2) + return false; + + return true; +} + +static int dpu_plane_is_multirect_parallel_capable(struct dpu_hw_sspp *sspp, + struct dpu_sw_pipe_cfg *pipe_cfg, + const struct msm_format *fmt, + uint32_t max_linewidth) +{ + return dpu_plane_is_multirect_capable(sspp, pipe_cfg, fmt) && + dpu_plane_is_parallel_capable(pipe_cfg, fmt, max_linewidth); +} + +static bool dpu_plane_get_single_pipe_in_stage(struct dpu_plane_state *pstate, + struct dpu_sw_pipe **single_pipe, + struct dpu_sw_pipe_cfg **single_pipe_cfg, + int stage_index) +{ + int pipe_idx; + + pipe_idx = stage_index * PIPES_PER_STAGE; + if (drm_rect_width(&pstate->pipe_cfg[pipe_idx].src_rect) != 0 && + drm_rect_width(&pstate->pipe_cfg[pipe_idx + 1].src_rect) == 0) { + *single_pipe = &pstate->pipe[pipe_idx]; + *single_pipe_cfg = &pstate->pipe_cfg[pipe_idx]; + return true; + } + + return false; +} + +static int dpu_plane_atomic_check_sspp(struct drm_plane *plane, + struct drm_atomic_commit *state, + const struct drm_crtc_state *crtc_state) +{ + struct drm_plane_state *new_plane_state = + drm_atomic_get_new_plane_state(state, plane); + struct dpu_plane *pdpu = to_dpu_plane(plane); + struct dpu_plane_state *pstate = to_dpu_plane_state(new_plane_state); + + struct dpu_sw_pipe *pipe; + struct dpu_sw_pipe_cfg *pipe_cfg; + int ret = 0, i; + + for (i = 0; i < PIPES_PER_PLANE; i++) { + pipe = &pstate->pipe[i]; + pipe_cfg = &pstate->pipe_cfg[i]; + if (!drm_rect_width(&pipe_cfg->src_rect)) + continue; + DPU_DEBUG_PLANE(pdpu, "pipe %d is in use, validate it\n", i); + ret = dpu_plane_atomic_check_pipe(pdpu, pipe, pipe_cfg, + &crtc_state->adjusted_mode, + new_plane_state); + if (ret) + return ret; + } + + return 0; +} + +static bool dpu_plane_try_multirect_parallel(struct dpu_sw_pipe *pipe, struct dpu_sw_pipe_cfg *pipe_cfg, + struct dpu_sw_pipe *r_pipe, struct dpu_sw_pipe_cfg *r_pipe_cfg, + struct dpu_hw_sspp *sspp, const struct msm_format *fmt, + uint32_t max_linewidth) +{ + r_pipe->sspp = NULL; + + pipe->multirect_index = DPU_SSPP_RECT_SOLO; + pipe->multirect_mode = DPU_SSPP_MULTIRECT_NONE; + + r_pipe->multirect_index = DPU_SSPP_RECT_SOLO; + r_pipe->multirect_mode = DPU_SSPP_MULTIRECT_NONE; + + if (drm_rect_width(&r_pipe_cfg->src_rect) != 0) { + if (!dpu_plane_is_multirect_parallel_capable(pipe->sspp, pipe_cfg, fmt, max_linewidth) || + !dpu_plane_is_multirect_parallel_capable(pipe->sspp, r_pipe_cfg, fmt, max_linewidth)) + return false; + + r_pipe->sspp = pipe->sspp; + + pipe->multirect_index = DPU_SSPP_RECT_0; + pipe->multirect_mode = DPU_SSPP_MULTIRECT_PARALLEL; + + r_pipe->multirect_index = DPU_SSPP_RECT_1; + r_pipe->multirect_mode = DPU_SSPP_MULTIRECT_PARALLEL; + } + + return true; +} + +static int dpu_plane_try_multirect_shared(struct dpu_plane_state *pstate, + struct dpu_plane_state *prev_adjacent_pstate, + const struct msm_format *fmt, + uint32_t max_linewidth, int stage_index) +{ + struct dpu_sw_pipe *pipe, *prev_pipe; + struct dpu_sw_pipe_cfg *pipe_cfg, *prev_pipe_cfg; + const struct msm_format *prev_fmt = msm_framebuffer_format(prev_adjacent_pstate->base.fb); + u16 max_tile_height = 1; + + if (!dpu_plane_get_single_pipe_in_stage(pstate, &pipe, + &pipe_cfg, stage_index)) + return false; + + if (!dpu_plane_get_single_pipe_in_stage(prev_adjacent_pstate, + &prev_pipe, &prev_pipe_cfg, + stage_index) || + prev_pipe->multirect_mode != DPU_SSPP_MULTIRECT_NONE) + return false; + + /* Do not validate SSPP of current plane when it is not ready */ + if (!dpu_plane_is_multirect_capable(pipe->sspp, pipe_cfg, fmt) || + !dpu_plane_is_multirect_capable(prev_pipe->sspp, prev_pipe_cfg, prev_fmt)) + return false; + + if (MSM_FORMAT_IS_UBWC(fmt)) + max_tile_height = max(max_tile_height, fmt->tile_height); + + if (MSM_FORMAT_IS_UBWC(prev_fmt)) + max_tile_height = max(max_tile_height, prev_fmt->tile_height); + + if (dpu_plane_is_parallel_capable(pipe_cfg, fmt, max_linewidth) && + dpu_plane_is_parallel_capable(prev_pipe_cfg, prev_fmt, max_linewidth) && + (pipe_cfg->dst_rect.x1 >= prev_pipe_cfg->dst_rect.x2 || + prev_pipe_cfg->dst_rect.x1 >= pipe_cfg->dst_rect.x2)) { + pipe->sspp = prev_pipe->sspp; + + pipe->multirect_index = DPU_SSPP_RECT_1; + pipe->multirect_mode = DPU_SSPP_MULTIRECT_PARALLEL; + + prev_pipe->multirect_index = DPU_SSPP_RECT_0; + prev_pipe->multirect_mode = DPU_SSPP_MULTIRECT_PARALLEL; + + return true; + } + + if (pipe_cfg->dst_rect.y1 >= prev_pipe_cfg->dst_rect.y2 + 2 * max_tile_height || + prev_pipe_cfg->dst_rect.y1 >= pipe_cfg->dst_rect.y2 + 2 * max_tile_height) { + pipe->sspp = prev_pipe->sspp; + + pipe->multirect_index = DPU_SSPP_RECT_1; + pipe->multirect_mode = DPU_SSPP_MULTIRECT_TIME_MX; + + prev_pipe->multirect_index = DPU_SSPP_RECT_0; + prev_pipe->multirect_mode = DPU_SSPP_MULTIRECT_TIME_MX; + + return true; + } + + return false; +} + +static int dpu_plane_atomic_check(struct drm_plane *plane, + struct drm_atomic_commit *state) +{ + struct drm_plane_state *plane_state = + drm_atomic_get_plane_state(state, plane); + struct drm_plane_state *old_plane_state = + drm_atomic_get_old_plane_state(state, plane); + int ret = 0; + struct drm_crtc_state *crtc_state = NULL; + + if (IS_ERR(plane_state)) + return PTR_ERR(plane_state); + + if (plane_state->crtc) + crtc_state = drm_atomic_get_new_crtc_state(state, + plane_state->crtc); + + ret = dpu_plane_atomic_check_nosspp(plane, plane_state, crtc_state); + if (ret) + return ret; + + if (!plane_state->visible) + return 0; + + /* + * Force resource reallocation if the format of FB or src/dst have + * changed. We might need to allocate different SSPP or SSPPs for this + * plane than the one used previously. + */ + if (!old_plane_state || !old_plane_state->fb || + old_plane_state->src_w != plane_state->src_w || + old_plane_state->src_h != plane_state->src_h || + old_plane_state->crtc_w != plane_state->crtc_w || + old_plane_state->crtc_h != plane_state->crtc_h || + msm_framebuffer_format(old_plane_state->fb) != + msm_framebuffer_format(plane_state->fb)) + crtc_state->planes_changed = true; + return 0; +} + +static int dpu_plane_assign_resource_in_stage(struct dpu_sw_pipe *pipe, + struct dpu_sw_pipe_cfg *pipe_cfg, + struct drm_plane_state *plane_state, + struct dpu_global_state *global_state, + struct drm_crtc *crtc, + struct dpu_rm_sspp_requirements *reqs) +{ + struct drm_plane *plane = plane_state->plane; + struct dpu_kms *dpu_kms = _dpu_plane_get_kms(plane); + struct dpu_sw_pipe *r_pipe = pipe + 1; + struct dpu_sw_pipe_cfg *r_pipe_cfg = pipe_cfg + 1; + + if (drm_rect_width(&pipe_cfg->src_rect) == 0) + return 0; + + pipe->sspp = dpu_rm_reserve_sspp(&dpu_kms->rm, global_state, crtc, reqs); + if (!pipe->sspp) + return -ENODEV; + pipe->multirect_index = DPU_SSPP_RECT_SOLO; + pipe->multirect_mode = DPU_SSPP_MULTIRECT_NONE; + + if (drm_rect_width(&r_pipe_cfg->src_rect) == 0) + return 0; + + if (dpu_plane_try_multirect_parallel(pipe, pipe_cfg, r_pipe, r_pipe_cfg, + pipe->sspp, + msm_framebuffer_format(plane_state->fb), + dpu_kms->catalog->caps->max_linewidth)) + return 0; + + r_pipe->sspp = dpu_rm_reserve_sspp(&dpu_kms->rm, global_state, crtc, reqs); + if (!r_pipe->sspp) + return -ENODEV; + r_pipe->multirect_index = DPU_SSPP_RECT_SOLO; + r_pipe->multirect_mode = DPU_SSPP_MULTIRECT_NONE; + + return 0; +} + +static int dpu_plane_virtual_assign_resources(struct drm_crtc *crtc, + struct dpu_global_state *global_state, + struct drm_atomic_commit *state, + struct drm_plane_state *plane_state, + const struct drm_crtc_state *crtc_state, + struct drm_plane_state **prev_adjacent_plane_state) +{ + struct drm_plane *plane = plane_state->plane; + struct dpu_kms *dpu_kms = _dpu_plane_get_kms(plane); + struct dpu_rm_sspp_requirements reqs; + struct dpu_plane_state *pstate, *prev_adjacent_pstate[STAGES_PER_PLANE]; + struct dpu_sw_pipe *pipe; + struct dpu_sw_pipe_cfg *pipe_cfg; + const struct msm_format *fmt; + int i, ret; + + pstate = to_dpu_plane_state(plane_state); + for (i = 0; i < STAGES_PER_PLANE; i++) + prev_adjacent_pstate[i] = prev_adjacent_plane_state[i] ? + to_dpu_plane_state(prev_adjacent_plane_state[i]) : NULL; + + for (i = 0; i < PIPES_PER_PLANE; i++) + pstate->pipe[i].sspp = NULL; + + if (!plane_state->fb) + return -EINVAL; + + ret = dpu_plane_split(plane, plane_state, crtc_state); + if (ret) + return ret; + + fmt = msm_framebuffer_format(plane_state->fb); + reqs.yuv = MSM_FORMAT_IS_YUV(fmt); + reqs.scale = (plane_state->src_w >> 16 != plane_state->crtc_w) || + (plane_state->src_h >> 16 != plane_state->crtc_h); + + reqs.rot90 = drm_rotation_90_or_270(plane_state->rotation); + + for (i = 0; i < STAGES_PER_PLANE; i++) { + if (prev_adjacent_pstate[i] && + dpu_plane_try_multirect_shared(pstate, prev_adjacent_pstate[i], fmt, + dpu_kms->catalog->caps->max_linewidth, + i)) + continue; + + if (dpu_plane_get_single_pipe_in_stage(pstate, &pipe, &pipe_cfg, i)) + prev_adjacent_plane_state[i] = plane_state; + + pipe = &pstate->pipe[i * PIPES_PER_STAGE]; + pipe_cfg = &pstate->pipe_cfg[i * PIPES_PER_STAGE]; + ret = dpu_plane_assign_resource_in_stage(pipe, pipe_cfg, + plane_state, + global_state, + crtc, &reqs); + if (ret) + return ret; + } + + return dpu_plane_atomic_check_sspp(plane, state, crtc_state); +} + +static int dpu_plane_assign_resources(struct drm_crtc *crtc, + struct dpu_global_state *global_state, + struct drm_atomic_commit *state, + struct drm_plane_state *plane_state, + const struct drm_crtc_state *crtc_state) +{ + struct drm_plane *plane = plane_state->plane; + struct dpu_kms *dpu_kms = _dpu_plane_get_kms(plane); + struct dpu_plane_state *pstate = to_dpu_plane_state(plane_state); + struct dpu_sw_pipe *pipe = &pstate->pipe[0]; + struct dpu_sw_pipe *r_pipe = &pstate->pipe[1]; + struct dpu_sw_pipe_cfg *pipe_cfg = &pstate->pipe_cfg[0]; + struct dpu_sw_pipe_cfg *r_pipe_cfg = &pstate->pipe_cfg[1]; + struct dpu_plane *pdpu = to_dpu_plane(plane); + int ret; + + pipe->sspp = dpu_rm_get_sspp(&dpu_kms->rm, pdpu->pipe); + if (!pipe->sspp) + return -EINVAL; + + ret = dpu_plane_split(plane, plane_state, crtc_state); + if (ret) + return ret; + + if (!dpu_plane_try_multirect_parallel(pipe, pipe_cfg, r_pipe, r_pipe_cfg, + pipe->sspp, + msm_framebuffer_format(plane_state->fb), + dpu_kms->catalog->caps->max_linewidth)) { + DPU_DEBUG_PLANE(pdpu, "invalid " DRM_RECT_FMT " /" DRM_RECT_FMT + " max_line:%u, can't use split source\n", + DRM_RECT_ARG(&pipe_cfg->src_rect), + DRM_RECT_ARG(&r_pipe_cfg->src_rect), + dpu_kms->catalog->caps->max_linewidth); + return -E2BIG; + } + + return dpu_plane_atomic_check_sspp(plane, state, crtc_state); +} + +int dpu_assign_plane_resources(struct dpu_global_state *global_state, + struct drm_atomic_commit *state, + struct drm_crtc *crtc, + struct drm_plane_state **states, + unsigned int num_planes) +{ + struct drm_plane_state *prev_adjacent_plane_state[STAGES_PER_PLANE] = { NULL }; + const struct drm_crtc_state *crtc_state = NULL; + unsigned int i; + int ret; + + for (i = 0; i < num_planes; i++) { + struct drm_plane_state *plane_state = states[i]; + + if (!plane_state || + !plane_state->visible) + continue; + + if (plane_state->crtc) + crtc_state = drm_atomic_get_new_crtc_state(state, + plane_state->crtc); + + if (!dpu_use_virtual_planes) + ret = dpu_plane_assign_resources(crtc, global_state, + state, plane_state, + crtc_state); + else + ret = dpu_plane_virtual_assign_resources(crtc, global_state, + state, plane_state, + crtc_state, + prev_adjacent_plane_state); + if (ret) + return ret; + } + + return 0; +} + +static void dpu_plane_flush_csc(struct dpu_plane *pdpu, struct dpu_sw_pipe *pipe) +{ + const struct msm_format *format = + msm_framebuffer_format(pdpu->base.state->fb); + const struct dpu_csc_cfg *csc_ptr; + + if (!pipe->sspp || !pipe->sspp->ops.setup_csc) + return; + + csc_ptr = _dpu_plane_get_csc(pipe, format); + if (!csc_ptr) + return; + + DPU_DEBUG_PLANE(pdpu, "using 0x%X 0x%X 0x%X...\n", + csc_ptr->csc_mv[0], + csc_ptr->csc_mv[1], + csc_ptr->csc_mv[2]); + + pipe->sspp->ops.setup_csc(pipe->sspp, csc_ptr); + +} + +/** + * dpu_plane_flush - final plane operations before commit flush + * @plane: Pointer to drm plane structure + */ +void dpu_plane_flush(struct drm_plane *plane) +{ + struct dpu_plane *pdpu; + struct dpu_plane_state *pstate; + int i; + + if (!plane || !plane->state) { + DPU_ERROR("invalid plane\n"); + return; + } + + pdpu = to_dpu_plane(plane); + pstate = to_dpu_plane_state(plane->state); + + /* + * These updates have to be done immediately before the plane flush + * timing, and may not be moved to the atomic_update/mode_set functions. + */ + if (pdpu->is_error) + /* force white frame with 100% alpha pipe output on error */ + _dpu_plane_color_fill(pdpu, 0xFFFFFF, 0xFF); + else if (pdpu->color_fill & DPU_PLANE_COLOR_FILL_FLAG) + /* force 100% alpha */ + _dpu_plane_color_fill(pdpu, pdpu->color_fill, 0xFF); + else { + for (i = 0; i < PIPES_PER_PLANE; i++) + dpu_plane_flush_csc(pdpu, &pstate->pipe[i]); + } + + /* flag h/w flush complete */ + if (plane->state) + pstate->pending = false; +} + +/** + * dpu_plane_set_error: enable/disable error condition + * @plane: pointer to drm_plane structure + * @error: error value to set + */ +void dpu_plane_set_error(struct drm_plane *plane, bool error) +{ + struct dpu_plane *pdpu; + + if (!plane) + return; + + pdpu = to_dpu_plane(plane); + pdpu->is_error = error; +} + +static void dpu_plane_sspp_update_pipe(struct drm_plane *plane, + struct dpu_sw_pipe *pipe, + struct dpu_sw_pipe_cfg *pipe_cfg, + const struct msm_format *fmt, + int frame_rate, + struct dpu_hw_fmt_layout *layout) +{ + uint32_t src_flags; + struct dpu_plane *pdpu = to_dpu_plane(plane); + struct drm_plane_state *state = plane->state; + struct dpu_plane_state *pstate = to_dpu_plane_state(state); + + if (layout && pipe->sspp->ops.setup_sourceaddress) { + trace_dpu_plane_set_scanout(pipe, layout); + pipe->sspp->ops.setup_sourceaddress(pipe, layout); + } + + /* override for color fill */ + if (pdpu->color_fill & DPU_PLANE_COLOR_FILL_FLAG) { + _dpu_plane_set_qos_ctrl(plane, pipe, false); + + /* skip remaining processing on color fill */ + return; + } + + if (pipe->sspp->ops.setup_rects) { + pipe->sspp->ops.setup_rects(pipe, + pipe_cfg); + } + + _dpu_plane_setup_scaler(pipe, fmt, false, pipe_cfg); + + if (pipe->sspp->ops.setup_multirect) + pipe->sspp->ops.setup_multirect( + pipe); + + if (pipe->sspp->ops.setup_format) { + unsigned int rotation = pipe_cfg->rotation; + + src_flags = 0x0; + + if (rotation & DRM_MODE_REFLECT_X) + src_flags |= DPU_SSPP_FLIP_LR; + + if (rotation & DRM_MODE_REFLECT_Y) + src_flags |= DPU_SSPP_FLIP_UD; + + if (rotation & DRM_MODE_ROTATE_90) + src_flags |= DPU_SSPP_ROT_90; + + /* update format */ + pipe->sspp->ops.setup_format(pipe, fmt, src_flags); + + if (pipe->sspp->ops.setup_cdp) { + const struct dpu_perf_cfg *perf = pdpu->catalog->perf; + + pipe->sspp->ops.setup_cdp(pipe, fmt, + perf->cdp_cfg[DPU_PERF_CDP_USAGE_RT].rd_enable); + } + } + + _dpu_plane_set_qos_lut(plane, pipe, fmt, pipe_cfg); + + if (pipe->sspp->idx != SSPP_CURSOR0 && + pipe->sspp->idx != SSPP_CURSOR1) + _dpu_plane_set_ot_limit(plane, pipe, pipe_cfg, frame_rate); + + if (pstate->needs_qos_remap) + _dpu_plane_set_qos_remap(plane, pipe); +} + +static void dpu_plane_sspp_atomic_update(struct drm_plane *plane, + struct drm_plane_state *new_state) +{ + struct dpu_plane *pdpu = to_dpu_plane(plane); + struct drm_plane_state *state = plane->state; + struct dpu_plane_state *pstate = to_dpu_plane_state(state); + struct drm_crtc *crtc = state->crtc; + struct drm_framebuffer *fb = state->fb; + bool is_rt_pipe; + const struct msm_format *fmt = + msm_framebuffer_format(fb); + int i; + + pstate->pending = true; + + is_rt_pipe = (dpu_crtc_get_client_type(crtc) != NRT_CLIENT); + pstate->needs_qos_remap |= (is_rt_pipe != pdpu->is_rt_pipe); + pdpu->is_rt_pipe = is_rt_pipe; + + dpu_format_populate_addrs(new_state->fb, &pstate->layout); + + DPU_DEBUG_PLANE(pdpu, "FB[%u] " DRM_RECT_FP_FMT "->crtc%u " DRM_RECT_FMT + ", %p4cc ubwc %d\n", fb->base.id, DRM_RECT_FP_ARG(&state->src), + crtc->base.id, DRM_RECT_ARG(&state->dst), + &fmt->pixel_format, MSM_FORMAT_IS_UBWC(fmt)); + + for (i = 0; i < PIPES_PER_PLANE; i++) { + if (!drm_rect_width(&pstate->pipe_cfg[i].src_rect)) + continue; + dpu_plane_sspp_update_pipe(plane, &pstate->pipe[i], + &pstate->pipe_cfg[i], fmt, + drm_mode_vrefresh(&crtc->mode), + &pstate->layout); + } + + if (pstate->needs_qos_remap) + pstate->needs_qos_remap = false; + + pstate->plane_fetch_bw = 0; + pstate->plane_clk = 0; + for (i = 0; i < PIPES_PER_PLANE; i++) { + if (!drm_rect_width(&pstate->pipe_cfg[i].src_rect)) + continue; + pstate->plane_fetch_bw += _dpu_plane_calc_bw(pdpu->catalog, fmt, + &crtc->mode, &pstate->pipe_cfg[i]); + + pstate->plane_clk = max(pstate->plane_clk, + _dpu_plane_calc_clk(&crtc->mode, + &pstate->pipe_cfg[i])); + } +} + +static void _dpu_plane_atomic_disable(struct drm_plane *plane) +{ + struct drm_plane_state *state = plane->state; + struct dpu_plane_state *pstate = to_dpu_plane_state(state); + struct dpu_sw_pipe *pipe; + int i; + + for (i = 0; i < PIPES_PER_PLANE; i += 1) { + pipe = &pstate->pipe[i]; + if (!pipe->sspp) + continue; + + trace_dpu_plane_disable(DRMID(plane), false, + pstate->pipe[i].multirect_mode); + + if (i % PIPES_PER_STAGE == 0) + continue; + + /* + * clear multirect for the right pipe so that the SSPP + * can be further reused in the solo mode + */ + pipe->multirect_index = DPU_SSPP_RECT_SOLO; + pipe->multirect_mode = DPU_SSPP_MULTIRECT_NONE; + if (pipe->sspp->ops.setup_multirect) + pipe->sspp->ops.setup_multirect(pipe); + } + + pstate->pending = true; +} + +static void dpu_plane_atomic_update(struct drm_plane *plane, + struct drm_atomic_commit *state) +{ + struct dpu_plane *pdpu = to_dpu_plane(plane); + struct drm_plane_state *new_state = drm_atomic_get_new_plane_state(state, + plane); + + pdpu->is_error = false; + + DPU_DEBUG_PLANE(pdpu, "\n"); + + if (!new_state->visible) { + _dpu_plane_atomic_disable(plane); + } else { + dpu_plane_sspp_atomic_update(plane, new_state); + } +} + +static void dpu_plane_destroy_state(struct drm_plane *plane, + struct drm_plane_state *state) +{ + __drm_atomic_helper_plane_destroy_state(state); + kfree(to_dpu_plane_state(state)); +} + +static struct drm_plane_state * +dpu_plane_duplicate_state(struct drm_plane *plane) +{ + struct dpu_plane *pdpu; + struct dpu_plane_state *pstate; + struct dpu_plane_state *old_state; + + if (!plane) { + DPU_ERROR("invalid plane\n"); + return NULL; + } else if (!plane->state) { + DPU_ERROR("invalid plane state\n"); + return NULL; + } + + old_state = to_dpu_plane_state(plane->state); + pdpu = to_dpu_plane(plane); + pstate = kmemdup(old_state, sizeof(*old_state), GFP_KERNEL); + if (!pstate) { + DPU_ERROR_PLANE(pdpu, "failed to allocate state\n"); + return NULL; + } + + DPU_DEBUG_PLANE(pdpu, "\n"); + + pstate->pending = false; + + __drm_atomic_helper_plane_duplicate_state(plane, &pstate->base); + + return &pstate->base; +} + +static const char * const multirect_mode_name[] = { + [DPU_SSPP_MULTIRECT_NONE] = "none", + [DPU_SSPP_MULTIRECT_PARALLEL] = "parallel", + [DPU_SSPP_MULTIRECT_TIME_MX] = "time_mx", +}; + +static const char * const multirect_index_name[] = { + [DPU_SSPP_RECT_SOLO] = "solo", + [DPU_SSPP_RECT_0] = "rect_0", + [DPU_SSPP_RECT_1] = "rect_1", +}; + +static const char *dpu_get_multirect_mode(enum dpu_sspp_multirect_mode mode) +{ + if (WARN_ON(mode >= ARRAY_SIZE(multirect_mode_name))) + return "unknown"; + + return multirect_mode_name[mode]; +} + +static const char *dpu_get_multirect_index(enum dpu_sspp_multirect_index index) +{ + if (WARN_ON(index >= ARRAY_SIZE(multirect_index_name))) + return "unknown"; + + return multirect_index_name[index]; +} + +static void dpu_plane_atomic_print_state(struct drm_printer *p, + const struct drm_plane_state *state) +{ + const struct dpu_plane_state *pstate = to_dpu_plane_state(state); + const struct dpu_sw_pipe *pipe; + const struct dpu_sw_pipe_cfg *pipe_cfg; + int i; + + drm_printf(p, "\tstage=%d\n", pstate->stage); + + for (i = 0; i < PIPES_PER_PLANE; i++) { + pipe = &pstate->pipe[i]; + if (!pipe->sspp) + continue; + pipe_cfg = &pstate->pipe_cfg[i]; + drm_printf(p, "\tsspp[%d]=%s\n", i, pipe->sspp->cap->name); + drm_printf(p, "\tmultirect_mode[%d]=%s\n", i, + dpu_get_multirect_mode(pipe->multirect_mode)); + drm_printf(p, "\tmultirect_index[%d]=%s\n", i, + dpu_get_multirect_index(pipe->multirect_index)); + drm_printf(p, "\tsrc[%d]=" DRM_RECT_FMT "\n", i, + DRM_RECT_ARG(&pipe_cfg->src_rect)); + drm_printf(p, "\tdst[%d]=" DRM_RECT_FMT "\n", i, + DRM_RECT_ARG(&pipe_cfg->dst_rect)); + } +} + +static void dpu_plane_reset(struct drm_plane *plane) +{ + struct dpu_plane *pdpu; + struct dpu_plane_state *pstate; + + if (!plane) { + DPU_ERROR("invalid plane\n"); + return; + } + + pdpu = to_dpu_plane(plane); + DPU_DEBUG_PLANE(pdpu, "\n"); + + /* remove previous state, if present */ + if (plane->state) { + dpu_plane_destroy_state(plane, plane->state); + plane->state = NULL; + } + + pstate = kzalloc_obj(*pstate); + if (!pstate) { + DPU_ERROR_PLANE(pdpu, "failed to allocate state\n"); + return; + } + + __drm_atomic_helper_plane_reset(plane, &pstate->base); +} + +#ifdef CONFIG_DEBUG_FS +void dpu_plane_danger_signal_ctrl(struct drm_plane *plane, bool enable) +{ + struct dpu_plane *pdpu = to_dpu_plane(plane); + struct dpu_plane_state *pstate = to_dpu_plane_state(plane->state); + struct dpu_kms *dpu_kms = _dpu_plane_get_kms(plane); + int i; + + if (!pdpu->is_rt_pipe) + return; + + pm_runtime_get_sync(&dpu_kms->pdev->dev); + for (i = 0; i < PIPES_PER_PLANE; i++) { + if (!pstate->pipe[i].sspp) + continue; + _dpu_plane_set_qos_ctrl(plane, &pstate->pipe[i], enable); + } + pm_runtime_put_sync(&dpu_kms->pdev->dev); +} +#endif + +static bool dpu_plane_format_mod_supported(struct drm_plane *plane, + uint32_t format, uint64_t modifier) +{ + struct dpu_kms *dpu_kms = _dpu_plane_get_kms(plane); + bool has_no_ubwc = (dpu_kms->mdss->ubwc_enc_version == 0); + + if (modifier == DRM_FORMAT_MOD_LINEAR) + return true; + + if (modifier == DRM_FORMAT_MOD_QCOM_COMPRESSED && !has_no_ubwc) + return dpu_find_format(format, qcom_compressed_supported_formats, + ARRAY_SIZE(qcom_compressed_supported_formats)); + + return false; +} + +static const struct drm_plane_funcs dpu_plane_funcs = { + .update_plane = drm_atomic_helper_update_plane, + .disable_plane = drm_atomic_helper_disable_plane, + .reset = dpu_plane_reset, + .atomic_duplicate_state = dpu_plane_duplicate_state, + .atomic_destroy_state = dpu_plane_destroy_state, + .atomic_print_state = dpu_plane_atomic_print_state, + .format_mod_supported = dpu_plane_format_mod_supported, +}; + +static const struct drm_plane_helper_funcs dpu_plane_helper_funcs = { + .prepare_fb = dpu_plane_prepare_fb, + .cleanup_fb = dpu_plane_cleanup_fb, + .atomic_check = dpu_plane_atomic_check, + .atomic_update = dpu_plane_atomic_update, +}; + +static const struct drm_plane_helper_funcs dpu_plane_virtual_helper_funcs = { + .prepare_fb = dpu_plane_prepare_fb, + .cleanup_fb = dpu_plane_cleanup_fb, + .atomic_check = dpu_plane_atomic_check, + .atomic_update = dpu_plane_atomic_update, +}; + +/* initialize plane */ +static struct drm_plane *dpu_plane_init_common(struct drm_device *dev, + enum drm_plane_type type, + unsigned long possible_crtcs, + bool inline_rotation, + const uint32_t *format_list, + uint32_t num_formats, + enum dpu_sspp pipe) +{ + struct drm_plane *plane = NULL; + struct dpu_plane *pdpu; + struct msm_drm_private *priv = dev->dev_private; + struct dpu_kms *kms = to_dpu_kms(priv->kms); + uint32_t supported_rotations; + int ret; + + pdpu = drmm_universal_plane_alloc(dev, struct dpu_plane, base, + 0xff, &dpu_plane_funcs, + format_list, num_formats, + supported_format_modifiers, type, NULL); + if (IS_ERR(pdpu)) + return ERR_CAST(pdpu); + + /* cache local stuff for later */ + plane = &pdpu->base; + pdpu->pipe = pipe; + + pdpu->catalog = kms->catalog; + + ret = drm_plane_create_zpos_property(plane, 0, 0, DPU_ZPOS_MAX); + if (ret) + DPU_ERROR("failed to install zpos property, rc = %d\n", ret); + + drm_plane_create_alpha_property(plane); + drm_plane_create_blend_mode_property(plane, + BIT(DRM_MODE_BLEND_PIXEL_NONE) | + BIT(DRM_MODE_BLEND_PREMULTI) | + BIT(DRM_MODE_BLEND_COVERAGE)); + + supported_rotations = DRM_MODE_REFLECT_MASK | DRM_MODE_ROTATE_0 | DRM_MODE_ROTATE_180; + + if (inline_rotation) + supported_rotations |= DRM_MODE_ROTATE_MASK; + + drm_plane_create_rotation_property(plane, + DRM_MODE_ROTATE_0, supported_rotations); + + drm_plane_enable_fb_damage_clips(plane); + + DPU_DEBUG("%s created for pipe:%u id:%u\n", plane->name, + pipe, plane->base.id); + return plane; +} + +/** + * dpu_plane_init - create new dpu plane for the given pipe + * @dev: Pointer to DRM device + * @pipe: dpu hardware pipe identifier + * @type: Plane type - PRIMARY/OVERLAY/CURSOR + * @possible_crtcs: bitmask of crtc that can be attached to the given pipe + * + * Initialize the plane. + */ +struct drm_plane *dpu_plane_init(struct drm_device *dev, + uint32_t pipe, enum drm_plane_type type, + unsigned long possible_crtcs) +{ + struct drm_plane *plane = NULL; + struct msm_drm_private *priv = dev->dev_private; + struct dpu_kms *kms = to_dpu_kms(priv->kms); + struct dpu_hw_sspp *pipe_hw; + + /* initialize underlying h/w driver */ + pipe_hw = dpu_rm_get_sspp(&kms->rm, pipe); + if (!pipe_hw || !pipe_hw->cap || !pipe_hw->cap->sblk) { + DPU_ERROR("[%u]SSPP is invalid\n", pipe); + return ERR_PTR(-EINVAL); + } + + + plane = dpu_plane_init_common(dev, type, possible_crtcs, + pipe_hw->cap->features & BIT(DPU_SSPP_INLINE_ROTATION), + pipe_hw->cap->sblk->format_list, + pipe_hw->cap->sblk->num_formats, + pipe); + if (IS_ERR(plane)) + return plane; + + drm_plane_helper_add(plane, &dpu_plane_helper_funcs); + + DPU_DEBUG("%s created for pipe:%u id:%u\n", plane->name, + pipe, plane->base.id); + + return plane; +} + +/** + * dpu_plane_init_virtual - create new virtualized DPU plane + * @dev: Pointer to DRM device + * @type: Plane type - PRIMARY/OVERLAY/CURSOR + * @possible_crtcs: bitmask of crtc that can be attached to the given pipe + * + * Initialize the virtual plane with no backing SSPP / pipe. + */ +struct drm_plane *dpu_plane_init_virtual(struct drm_device *dev, + enum drm_plane_type type, + unsigned long possible_crtcs) +{ + struct drm_plane *plane = NULL; + struct msm_drm_private *priv = dev->dev_private; + struct dpu_kms *kms = to_dpu_kms(priv->kms); + bool has_inline_rotation = false; + const u32 *format_list = NULL; + u32 num_formats = 0; + int i; + + /* Determine the largest configuration that we can implement */ + for (i = 0; i < kms->catalog->sspp_count; i++) { + const struct dpu_sspp_cfg *cfg = &kms->catalog->sspp[i]; + + if (test_bit(DPU_SSPP_INLINE_ROTATION, &cfg->features)) + has_inline_rotation = true; + + if (!format_list || + cfg->sblk->csc_blk.len) { + format_list = cfg->sblk->format_list; + num_formats = cfg->sblk->num_formats; + } + } + + plane = dpu_plane_init_common(dev, type, possible_crtcs, + has_inline_rotation, + format_list, + num_formats, + SSPP_NONE); + if (IS_ERR(plane)) + return plane; + + drm_plane_helper_add(plane, &dpu_plane_virtual_helper_funcs); + + DPU_DEBUG("%s created virtual id:%u\n", plane->name, plane->base.id); + + return plane; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h new file mode 100644 index 000000000..d2e7c8c25 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h @@ -0,0 +1,78 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#ifndef _DPU_PLANE_H_ +#define _DPU_PLANE_H_ + +#include + +#include "dpu_kms.h" +#include "dpu_hw_mdss.h" +#include "dpu_hw_sspp.h" + +/** + * struct dpu_plane_state: Define dpu extension of drm plane state object + * @base: base drm plane state object + * @pipe: software pipe description array + * @pipe_cfg: software pipe configuration array + * @stage: assigned by crtc blender + * @needs_qos_remap: qos remap settings need to be updated + * @multirect_index: index of the rectangle of SSPP + * @multirect_mode: parallel or time multiplex multirect mode + * @pending: whether the current update is still pending + * @plane_fetch_bw: calculated BW per plane + * @plane_clk: calculated clk per plane + * @needs_dirtyfb: whether attached CRTC needs pixel data explicitly flushed + * @layout: framebuffer memory layout + */ +struct dpu_plane_state { + struct drm_plane_state base; + struct dpu_sw_pipe pipe[PIPES_PER_PLANE]; + struct dpu_sw_pipe_cfg pipe_cfg[PIPES_PER_PLANE]; + enum dpu_stage stage; + bool needs_qos_remap; + bool pending; + + u64 plane_fetch_bw; + u64 plane_clk; + + bool needs_dirtyfb; + + struct dpu_hw_fmt_layout layout; +}; + +#define to_dpu_plane_state(x) \ + container_of(x, struct dpu_plane_state, base) + +void dpu_plane_flush(struct drm_plane *plane); + +void dpu_plane_set_error(struct drm_plane *plane, bool error); + +struct drm_plane *dpu_plane_init(struct drm_device *dev, + uint32_t pipe, enum drm_plane_type type, + unsigned long possible_crtcs); + +struct drm_plane *dpu_plane_init_virtual(struct drm_device *dev, + enum drm_plane_type type, + unsigned long possible_crtcs); + +int dpu_plane_color_fill(struct drm_plane *plane, + uint32_t color, uint32_t alpha); + +#ifdef CONFIG_DEBUG_FS +void dpu_plane_danger_signal_ctrl(struct drm_plane *plane, bool enable); +#else +static inline void dpu_plane_danger_signal_ctrl(struct drm_plane *plane, bool enable) {} +#endif + +int dpu_assign_plane_resources(struct dpu_global_state *global_state, + struct drm_atomic_commit *state, + struct drm_crtc *crtc, + struct drm_plane_state **states, + unsigned int num_planes); + +#endif /* _DPU_PLANE_H_ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c new file mode 100644 index 000000000..7e77d88f8 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c @@ -0,0 +1,1053 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2016-2018, The Linux Foundation. All rights reserved. + * Copyright (c) 2023-2024 Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#define pr_fmt(fmt) "[drm:%s] " fmt, __func__ +#include "dpu_kms.h" +#include "dpu_hw_lm.h" +#include "dpu_hw_ctl.h" +#include "dpu_hw_cdm.h" +#include "dpu_hw_cwb.h" +#include "dpu_hw_pingpong.h" +#include "dpu_hw_sspp.h" +#include "dpu_hw_intf.h" +#include "dpu_hw_wb.h" +#include "dpu_hw_dspp.h" +#include "dpu_hw_merge3d.h" +#include "dpu_hw_dsc.h" +#include "dpu_encoder.h" +#include "dpu_trace.h" + + +static inline bool reserved_by_other(uint32_t *res_map, int idx, + uint32_t crtc_id) +{ + return res_map[idx] && res_map[idx] != crtc_id; +} + +/** + * dpu_rm_init - Read hardware catalog and create reservation tracking objects + * for all HW blocks. + * @dev: Corresponding device for devres management + * @rm: DPU Resource Manager handle + * @cat: Pointer to hardware catalog + * @mdss_data: Pointer to MDSS / UBWC configuration + * @mmio: mapped register io address of MDP + * @return: 0 on Success otherwise -ERROR + */ +int dpu_rm_init(struct drm_device *dev, + struct dpu_rm *rm, + const struct dpu_mdss_cfg *cat, + const struct qcom_ubwc_cfg_data *mdss_data, + void __iomem *mmio) +{ + int rc, i; + + if (!rm || !cat || !mmio) { + DPU_ERROR("invalid kms\n"); + return -EINVAL; + } + + /* Clear, setup lists */ + memset(rm, 0, sizeof(*rm)); + + rm->has_legacy_ctls = (cat->mdss_ver->core_major_ver < 5); + + /* Interrogate HW catalog and create tracking items for hw blocks */ + for (i = 0; i < cat->mixer_count; i++) { + struct dpu_hw_mixer *hw; + const struct dpu_lm_cfg *lm = &cat->mixer[i]; + + hw = dpu_hw_lm_init(dev, lm, mmio, cat->mdss_ver); + if (IS_ERR(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed lm object creation: err %d\n", rc); + goto fail; + } + rm->mixer_blks[lm->id - LM_0] = &hw->base; + } + + for (i = 0; i < cat->merge_3d_count; i++) { + struct dpu_hw_merge_3d *hw; + const struct dpu_merge_3d_cfg *merge_3d = &cat->merge_3d[i]; + + hw = dpu_hw_merge_3d_init(dev, merge_3d, mmio); + if (IS_ERR(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed merge_3d object creation: err %d\n", + rc); + goto fail; + } + rm->merge_3d_blks[merge_3d->id - MERGE_3D_0] = &hw->base; + } + + for (i = 0; i < cat->pingpong_count; i++) { + struct dpu_hw_pingpong *hw; + const struct dpu_pingpong_cfg *pp = &cat->pingpong[i]; + + hw = dpu_hw_pingpong_init(dev, pp, mmio, cat->mdss_ver); + if (IS_ERR(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed pingpong object creation: err %d\n", + rc); + goto fail; + } + if (pp->merge_3d && pp->merge_3d < MERGE_3D_MAX) + hw->merge_3d = to_dpu_hw_merge_3d(rm->merge_3d_blks[pp->merge_3d - MERGE_3D_0]); + rm->pingpong_blks[pp->id - PINGPONG_0] = &hw->base; + } + + for (i = 0; i < cat->intf_count; i++) { + struct dpu_hw_intf *hw; + const struct dpu_intf_cfg *intf = &cat->intf[i]; + + hw = dpu_hw_intf_init(dev, intf, mmio, cat->mdss_ver); + if (IS_ERR(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed intf object creation: err %d\n", rc); + goto fail; + } + rm->hw_intf[intf->id - INTF_0] = hw; + } + + for (i = 0; i < cat->wb_count; i++) { + struct dpu_hw_wb *hw; + const struct dpu_wb_cfg *wb = &cat->wb[i]; + + hw = dpu_hw_wb_init(dev, wb, mmio, cat->mdss_ver); + if (IS_ERR(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed wb object creation: err %d\n", rc); + goto fail; + } + rm->hw_wb[wb->id - WB_0] = hw; + } + + for (i = 0; i < cat->cwb_count; i++) { + struct dpu_hw_cwb *hw; + const struct dpu_cwb_cfg *cwb = &cat->cwb[i]; + + hw = dpu_hw_cwb_init(dev, cwb, mmio); + if (IS_ERR(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed cwb object creation: err %d\n", rc); + goto fail; + } + rm->cwb_blks[cwb->id - CWB_0] = &hw->base; + } + + for (i = 0; i < cat->ctl_count; i++) { + struct dpu_hw_ctl *hw; + const struct dpu_ctl_cfg *ctl = &cat->ctl[i]; + + hw = dpu_hw_ctl_init(dev, ctl, mmio, cat->mdss_ver, cat->mixer_count, cat->mixer); + if (IS_ERR(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed ctl object creation: err %d\n", rc); + goto fail; + } + rm->ctl_blks[ctl->id - CTL_0] = &hw->base; + } + + for (i = 0; i < cat->dspp_count; i++) { + struct dpu_hw_dspp *hw; + const struct dpu_dspp_cfg *dspp = &cat->dspp[i]; + + hw = dpu_hw_dspp_init(dev, dspp, mmio); + if (IS_ERR(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed dspp object creation: err %d\n", rc); + goto fail; + } + rm->dspp_blks[dspp->id - DSPP_0] = &hw->base; + } + + for (i = 0; i < cat->dsc_count; i++) { + struct dpu_hw_dsc *hw; + const struct dpu_dsc_cfg *dsc = &cat->dsc[i]; + + if (cat->mdss_ver->core_major_ver >= 7) + hw = dpu_hw_dsc_init_1_2(dev, dsc, mmio); + else + hw = dpu_hw_dsc_init(dev, dsc, mmio, cat->mdss_ver); + + if (IS_ERR(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed dsc object creation: err %d\n", rc); + goto fail; + } + rm->dsc_blks[dsc->id - DSC_0] = &hw->base; + } + + for (i = 0; i < cat->sspp_count; i++) { + struct dpu_hw_sspp *hw; + const struct dpu_sspp_cfg *sspp = &cat->sspp[i]; + + hw = dpu_hw_sspp_init(dev, sspp, mmio, mdss_data, cat->mdss_ver); + if (IS_ERR(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed sspp object creation: err %d\n", rc); + goto fail; + } + rm->hw_sspp[sspp->id - SSPP_NONE] = hw; + } + + if (cat->cdm) { + struct dpu_hw_cdm *hw; + + hw = dpu_hw_cdm_init(dev, cat->cdm, mmio, cat->mdss_ver); + if (IS_ERR(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed cdm object creation: err %d\n", rc); + goto fail; + } + rm->cdm_blk = &hw->base; + } + + return 0; + +fail: + return rc ? rc : -EFAULT; +} + +static bool _dpu_rm_needs_split_display(const struct msm_display_topology *top) +{ + return top->num_intf > 1; +} + +/** + * _dpu_rm_get_lm_peer - get the id of a mixer which is a peer of the primary + * @rm: dpu resource manager handle + * @primary_idx: index of primary mixer in rm->mixer_blks[] + * + * Returns: lm peer mixed id on success or %-EINVAL on error + */ +static int _dpu_rm_get_lm_peer(struct dpu_rm *rm, int primary_idx) +{ + const struct dpu_lm_cfg *prim_lm_cfg; + + prim_lm_cfg = to_dpu_hw_mixer(rm->mixer_blks[primary_idx])->cap; + + if (prim_lm_cfg->lm_pair >= LM_0 && prim_lm_cfg->lm_pair < LM_MAX) + return prim_lm_cfg->lm_pair - LM_0; + return -EINVAL; +} + +static int _dpu_rm_reserve_cwb_mux_and_pingpongs(struct dpu_rm *rm, + struct dpu_global_state *global_state, + uint32_t crtc_id, + struct msm_display_topology *topology) +{ + int num_cwb_mux = topology->num_lm, cwb_mux_count = 0; + int cwb_pp_start_idx = PINGPONG_CWB_0 - PINGPONG_0; + int cwb_pp_idx[MAX_BLOCKS]; + int cwb_mux_idx[MAX_BLOCKS]; + + /* + * Reserve additional dedicated CWB PINGPONG blocks and muxes for each + * mixer + * + * TODO: add support reserving resources for platforms with no + * PINGPONG_CWB + */ + for (int i = 0; i < ARRAY_SIZE(rm->mixer_blks) && + cwb_mux_count < num_cwb_mux; i++) { + for (int j = 0; j < ARRAY_SIZE(rm->cwb_blks); j++) { + /* + * Odd LMs must be assigned to odd CWB muxes and even + * LMs with even CWB muxes. + * + * Since the RM HW block array index is based on the HW + * block ids, we can also use the array index to enforce + * the odd/even rule. See dpu_rm_init() for more + * information + */ + if (reserved_by_other(global_state->cwb_to_crtc_id, j, crtc_id) || + i % 2 != j % 2) + continue; + + cwb_mux_idx[cwb_mux_count] = j; + cwb_pp_idx[cwb_mux_count] = j + cwb_pp_start_idx; + cwb_mux_count++; + break; + } + } + + if (cwb_mux_count != num_cwb_mux) { + DPU_ERROR("Unable to reserve all CWB PINGPONGs\n"); + return -ENAVAIL; + } + + for (int i = 0; i < cwb_mux_count; i++) { + global_state->pingpong_to_crtc_id[cwb_pp_idx[i]] = crtc_id; + global_state->cwb_to_crtc_id[cwb_mux_idx[i]] = crtc_id; + } + + return 0; +} + +/** + * _dpu_rm_check_lm_and_get_connected_blks - check if proposed layer mixer meets + * proposed use case requirements, incl. hardwired dependent blocks like + * pingpong + * @rm: dpu resource manager handle + * @global_state: resources shared across multiple kms objects + * @crtc_id: crtc id requesting for allocation + * @lm_idx: index of proposed layer mixer in rm->mixer_blks[], function checks + * if lm, and all other hardwired blocks connected to the lm (pp) is + * available and appropriate + * @pp_idx: output parameter, index of pingpong block attached to the layer + * mixer in rm->pingpong_blks[]. + * @dspp_idx: output parameter, index of dspp block attached to the layer + * mixer in rm->dspp_blks[]. + * @topology: selected topology for the display + * Return: true if lm matches all requirements, false otherwise + */ +static bool _dpu_rm_check_lm_and_get_connected_blks(struct dpu_rm *rm, + struct dpu_global_state *global_state, + uint32_t crtc_id, int lm_idx, int *pp_idx, int *dspp_idx, + struct msm_display_topology *topology) +{ + const struct dpu_lm_cfg *lm_cfg; + int idx; + + /* Already reserved? */ + if (reserved_by_other(global_state->mixer_to_crtc_id, lm_idx, crtc_id)) { + DPU_DEBUG("LM_%d already reserved\n", lm_idx); + return false; + } + + lm_cfg = to_dpu_hw_mixer(rm->mixer_blks[lm_idx])->cap; + idx = lm_cfg->pingpong - PINGPONG_0; + if (idx < 0 || idx >= ARRAY_SIZE(rm->pingpong_blks) || !rm->pingpong_blks[idx]) { + DPU_ERROR("LM_%d, invalid PP_%d\n", lm_idx, idx); + return false; + } + + if (reserved_by_other(global_state->pingpong_to_crtc_id, idx, crtc_id)) { + DPU_DEBUG("LM_%d PP_%d already reserved\n", lm_idx, idx); + return false; + } + *pp_idx = idx; + + if (!topology->num_dspp) + return true; + + idx = lm_cfg->dspp - DSPP_0; + if (idx < 0 || idx >= ARRAY_SIZE(rm->dspp_blks) || !rm->dspp_blks[idx]) { + DPU_ERROR("LM_%d, invalid DSPP_%d\n", lm_idx, idx); + return false; + } + + if (reserved_by_other(global_state->dspp_to_crtc_id, idx, crtc_id)) { + DPU_DEBUG("LM_%d DSPP_%d already reserved\n", lm_idx, idx); + return false; + } + *dspp_idx = idx; + + return true; +} + +static int _dpu_rm_reserve_lms(struct dpu_rm *rm, + struct dpu_global_state *global_state, + uint32_t crtc_id, + struct msm_display_topology *topology) + +{ + int lm_idx[MAX_BLOCKS]; + int pp_idx[MAX_BLOCKS]; + int dspp_idx[MAX_BLOCKS] = {0}; + int i, lm_count = 0; + + if (!topology->num_lm) { + DPU_ERROR("zero LMs in topology\n"); + return -EINVAL; + } + + /* Find a primary mixer */ + for (i = 0; i < ARRAY_SIZE(rm->mixer_blks) && + lm_count < topology->num_lm; i++) { + if (!rm->mixer_blks[i]) + continue; + + /* + * Reset lm_count to an even index. This will drop the previous + * primary mixer if failed to find its peer. + */ + lm_count &= ~1; + lm_idx[lm_count] = i; + + if (!_dpu_rm_check_lm_and_get_connected_blks(rm, global_state, + crtc_id, i, &pp_idx[lm_count], + &dspp_idx[lm_count], topology)) { + continue; + } + + ++lm_count; + + /* Valid primary mixer found, find matching peers */ + if (lm_count < topology->num_lm) { + int j = _dpu_rm_get_lm_peer(rm, i); + + /* ignore the peer if there is an error or if the peer was already processed */ + if (j < 0 || j < i) + continue; + + if (!rm->mixer_blks[j]) + continue; + + if (!_dpu_rm_check_lm_and_get_connected_blks(rm, + global_state, crtc_id, j, + &pp_idx[lm_count], &dspp_idx[lm_count], + topology)) { + continue; + } + + lm_idx[lm_count] = j; + ++lm_count; + } + } + + if (lm_count != topology->num_lm) { + DPU_DEBUG("unable to find appropriate mixers\n"); + return -ENAVAIL; + } + + for (i = 0; i < lm_count; i++) { + global_state->mixer_to_crtc_id[lm_idx[i]] = crtc_id; + global_state->pingpong_to_crtc_id[pp_idx[i]] = crtc_id; + global_state->dspp_to_crtc_id[dspp_idx[i]] = + topology->num_dspp ? crtc_id : 0; + + trace_dpu_rm_reserve_lms(lm_idx[i] + LM_0, crtc_id, + pp_idx[i] + PINGPONG_0); + } + + return 0; +} + +static int _dpu_rm_reserve_ctls( + struct dpu_rm *rm, + struct dpu_global_state *global_state, + uint32_t crtc_id, + const struct msm_display_topology *top) +{ + int ctl_idx[MAX_BLOCKS]; + int i = 0, j, num_ctls; + bool needs_split_display; + + if (rm->has_legacy_ctls) { + /* + * TODO: check if there is a need for special handling if + * DPU < 5.0 get CWB support. + */ + num_ctls = top->num_intf; + + needs_split_display = _dpu_rm_needs_split_display(top); + } else { + /* use single CTL */ + num_ctls = 1; + needs_split_display = false; + } + + for (j = 0; j < ARRAY_SIZE(rm->ctl_blks); j++) { + const struct dpu_hw_ctl *ctl; + unsigned long features; + bool has_split_display; + + if (!rm->ctl_blks[j]) + continue; + if (reserved_by_other(global_state->ctl_to_crtc_id, j, crtc_id)) + continue; + + ctl = to_dpu_hw_ctl(rm->ctl_blks[j]); + features = ctl->caps->features; + has_split_display = BIT(DPU_CTL_SPLIT_DISPLAY) & features; + + DPU_DEBUG("CTL_%d caps 0x%lX\n", j, features); + + if (needs_split_display != has_split_display) + continue; + + ctl_idx[i] = j; + DPU_DEBUG("CTL_%d match\n", j); + + if (++i == num_ctls) + break; + + } + + if (i != num_ctls) + return -ENAVAIL; + + for (i = 0; i < ARRAY_SIZE(ctl_idx) && i < num_ctls; i++) { + global_state->ctl_to_crtc_id[ctl_idx[i]] = crtc_id; + trace_dpu_rm_reserve_ctls(i + CTL_0, crtc_id); + } + + return 0; +} + +static int _dpu_rm_pingpong_next_index(struct dpu_global_state *global_state, + int start, + uint32_t crtc_id) +{ + int i; + + for (i = start; i < (PINGPONG_MAX - PINGPONG_0); i++) { + if (global_state->pingpong_to_crtc_id[i] == crtc_id) + return i; + } + + return -ENAVAIL; +} + +static int _dpu_rm_pingpong_dsc_check(int dsc_idx, int pp_idx) +{ + /* + * DSC with even index must be used with the PINGPONG with even index + * DSC with odd index must be used with the PINGPONG with odd index + */ + if ((dsc_idx & 0x01) != (pp_idx & 0x01)) + return -ENAVAIL; + + return 0; +} + +static int _dpu_rm_dsc_alloc(struct dpu_rm *rm, + struct dpu_global_state *global_state, + uint32_t crtc_id, + const struct msm_display_topology *top) +{ + int num_dsc = 0; + int pp_idx = 0; + int dsc_idx; + int ret; + + for (dsc_idx = 0; dsc_idx < ARRAY_SIZE(rm->dsc_blks) && + num_dsc < top->num_dsc; dsc_idx++) { + if (!rm->dsc_blks[dsc_idx]) + continue; + + if (reserved_by_other(global_state->dsc_to_crtc_id, dsc_idx, crtc_id)) + continue; + + pp_idx = _dpu_rm_pingpong_next_index(global_state, pp_idx, crtc_id); + if (pp_idx < 0) + return -ENAVAIL; + + ret = _dpu_rm_pingpong_dsc_check(dsc_idx, pp_idx); + if (ret) + return -ENAVAIL; + + global_state->dsc_to_crtc_id[dsc_idx] = crtc_id; + num_dsc++; + pp_idx++; + } + + if (num_dsc < top->num_dsc) { + DPU_ERROR("DSC allocation failed num_dsc=%d required=%d\n", + num_dsc, top->num_dsc); + return -ENAVAIL; + } + + return 0; +} + +static int _dpu_rm_dsc_alloc_pair(struct dpu_rm *rm, + struct dpu_global_state *global_state, + uint32_t crtc_id, + const struct msm_display_topology *top) +{ + int num_dsc = 0; + int dsc_idx, pp_idx = 0; + int ret; + + /* only start from even dsc index */ + for (dsc_idx = 0; dsc_idx < ARRAY_SIZE(rm->dsc_blks) && + num_dsc < top->num_dsc; dsc_idx += 2) { + if (!rm->dsc_blks[dsc_idx] || + !rm->dsc_blks[dsc_idx + 1]) + continue; + + /* consective dsc index to be paired */ + if (reserved_by_other(global_state->dsc_to_crtc_id, dsc_idx, crtc_id) || + reserved_by_other(global_state->dsc_to_crtc_id, dsc_idx + 1, crtc_id)) + continue; + + pp_idx = _dpu_rm_pingpong_next_index(global_state, pp_idx, crtc_id); + if (pp_idx < 0) + return -ENAVAIL; + + ret = _dpu_rm_pingpong_dsc_check(dsc_idx, pp_idx); + if (ret) { + pp_idx = 0; + continue; + } + + pp_idx = _dpu_rm_pingpong_next_index(global_state, pp_idx + 1, crtc_id); + if (pp_idx < 0) + return -ENAVAIL; + + ret = _dpu_rm_pingpong_dsc_check(dsc_idx + 1, pp_idx); + if (ret) { + pp_idx = 0; + continue; + } + + global_state->dsc_to_crtc_id[dsc_idx] = crtc_id; + global_state->dsc_to_crtc_id[dsc_idx + 1] = crtc_id; + num_dsc += 2; + pp_idx++; /* start for next pair */ + } + + if (num_dsc < top->num_dsc) { + DPU_ERROR("DSC allocation failed num_dsc=%d required=%d\n", + num_dsc, top->num_dsc); + return -ENAVAIL; + } + + return 0; +} + +static int _dpu_rm_reserve_dsc(struct dpu_rm *rm, + struct dpu_global_state *global_state, + uint32_t crtc_id, + const struct msm_display_topology *top) +{ + if (!top->num_dsc || !top->num_intf) + return 0; + + /* + * Facts: + * 1) no pingpong split (two layer mixers shared one pingpong) + * 2) DSC pair starts from even index, such as index(0,1), (2,3), etc + * 3) even PINGPONG connects to even DSC + * 4) odd PINGPONG connects to odd DSC + * 5) pair: encoder +--> pp_idx_0 --> dsc_idx_0 + * +--> pp_idx_1 --> dsc_idx_1 + */ + + /* num_dsc should be either 1, 2 or 4 */ + if (top->num_dsc > top->num_intf) /* merge mode */ + return _dpu_rm_dsc_alloc_pair(rm, global_state, crtc_id, top); + else + return _dpu_rm_dsc_alloc(rm, global_state, crtc_id, top); + + return 0; +} + +static int _dpu_rm_reserve_cdm(struct dpu_rm *rm, + struct dpu_global_state *global_state, + uint32_t crtc_id, + int num_cdm) +{ + /* try allocating only one CDM block */ + if (!rm->cdm_blk) { + DPU_ERROR("CDM block does not exist\n"); + return -EIO; + } + + if (num_cdm > 1) { + DPU_ERROR("More than 1 INTF requesting CDM\n"); + return -EINVAL; + } + + if (global_state->cdm_to_crtc_id) { + DPU_ERROR("CDM_0 is already allocated\n"); + return -EIO; + } + + global_state->cdm_to_crtc_id = crtc_id; + + return 0; +} + +static int _dpu_rm_make_reservation( + struct dpu_rm *rm, + struct dpu_global_state *global_state, + uint32_t crtc_id, + struct msm_display_topology *topology) +{ + int ret; + + ret = _dpu_rm_reserve_lms(rm, global_state, crtc_id, topology); + if (ret) { + DPU_ERROR("unable to find appropriate mixers\n"); + return ret; + } + + if (topology->cwb_enabled) { + ret = _dpu_rm_reserve_cwb_mux_and_pingpongs(rm, global_state, + crtc_id, topology); + if (ret) + return ret; + } + + ret = _dpu_rm_reserve_ctls(rm, global_state, crtc_id, + topology); + if (ret) { + DPU_ERROR("unable to find appropriate CTL\n"); + return ret; + } + + ret = _dpu_rm_reserve_dsc(rm, global_state, crtc_id, topology); + if (ret) + return ret; + + if (topology->num_cdm > 0) { + ret = _dpu_rm_reserve_cdm(rm, global_state, crtc_id, topology->num_cdm); + if (ret) { + DPU_ERROR("unable to find CDM blk\n"); + return ret; + } + } + + return ret; +} + +static void _dpu_rm_clear_mapping(uint32_t *res_mapping, int cnt, + uint32_t crtc_id) +{ + int i; + + for (i = 0; i < cnt; i++) { + if (res_mapping[i] == crtc_id) + res_mapping[i] = 0; + } +} + +/** + * dpu_rm_release - Given the encoder for the display chain, release any + * HW blocks previously reserved for that use case. + * @global_state: resources shared across multiple kms objects + * @crtc: DRM CRTC handle + * @return: 0 on Success otherwise -ERROR + */ +void dpu_rm_release(struct dpu_global_state *global_state, + struct drm_crtc *crtc) +{ + uint32_t crtc_id = crtc->base.id; + + _dpu_rm_clear_mapping(global_state->pingpong_to_crtc_id, + ARRAY_SIZE(global_state->pingpong_to_crtc_id), crtc_id); + _dpu_rm_clear_mapping(global_state->mixer_to_crtc_id, + ARRAY_SIZE(global_state->mixer_to_crtc_id), crtc_id); + _dpu_rm_clear_mapping(global_state->ctl_to_crtc_id, + ARRAY_SIZE(global_state->ctl_to_crtc_id), crtc_id); + _dpu_rm_clear_mapping(global_state->dsc_to_crtc_id, + ARRAY_SIZE(global_state->dsc_to_crtc_id), crtc_id); + _dpu_rm_clear_mapping(global_state->dspp_to_crtc_id, + ARRAY_SIZE(global_state->dspp_to_crtc_id), crtc_id); + _dpu_rm_clear_mapping(&global_state->cdm_to_crtc_id, 1, crtc_id); + _dpu_rm_clear_mapping(global_state->cwb_to_crtc_id, + ARRAY_SIZE(global_state->cwb_to_crtc_id), crtc_id); +} + +/** + * dpu_rm_reserve - Given a CRTC->Encoder->Connector display chain, analyze + * the use connections and user requirements, specified through related + * topology control properties, and reserve hardware blocks to that + * display chain. + * HW blocks can then be accessed through dpu_rm_get_* functions. + * HW Reservations should be released via dpu_rm_release_hw. + * @rm: DPU Resource Manager handle + * @global_state: resources shared across multiple kms objects + * @crtc: DRM CRTC handle + * @topology: Pointer to topology info for the display + * @return: 0 on Success otherwise -ERROR + */ +int dpu_rm_reserve( + struct dpu_rm *rm, + struct dpu_global_state *global_state, + struct drm_crtc *crtc, + struct msm_display_topology *topology) +{ + int ret; + + if (IS_ERR(global_state)) { + DPU_ERROR("failed to global state\n"); + return PTR_ERR(global_state); + } + + DRM_DEBUG_KMS("reserving hw for crtc %d\n", crtc->base.id); + + DRM_DEBUG_KMS("num_lm: %d num_dsc: %d num_intf: %d\n", + topology->num_lm, topology->num_dsc, + topology->num_intf); + + ret = _dpu_rm_make_reservation(rm, global_state, crtc->base.id, topology); + if (ret) + DPU_ERROR("failed to reserve hw resources: %d\n", ret); + + return ret; +} + +static struct dpu_hw_sspp *dpu_rm_try_sspp(struct dpu_rm *rm, + struct dpu_global_state *global_state, + struct drm_crtc *crtc, + struct dpu_rm_sspp_requirements *reqs, + unsigned int type) +{ + uint32_t crtc_id = crtc->base.id; + struct dpu_hw_sspp *hw_sspp; + int i; + + for (i = 0; i < ARRAY_SIZE(rm->hw_sspp); i++) { + if (!rm->hw_sspp[i]) + continue; + + if (global_state->sspp_to_crtc_id[i]) + continue; + + hw_sspp = rm->hw_sspp[i]; + + if (hw_sspp->cap->type != type) + continue; + + if (reqs->scale && !hw_sspp->cap->sblk->scaler_blk.len) + continue; + + // TODO: QSEED2 and RGB scalers are not yet supported + if (reqs->scale && !hw_sspp->ops.setup_scaler) + continue; + + if (reqs->yuv && !hw_sspp->cap->sblk->csc_blk.len) + continue; + + if (reqs->rot90 && !(hw_sspp->cap->features & DPU_SSPP_INLINE_ROTATION)) + continue; + + global_state->sspp_to_crtc_id[i] = crtc_id; + + return rm->hw_sspp[i]; + } + + return NULL; +} + +/** + * dpu_rm_reserve_sspp - Reserve the required SSPP for the provided CRTC + * @rm: DPU Resource Manager handle + * @global_state: private global state + * @crtc: DRM CRTC handle + * @reqs: SSPP required features + */ +struct dpu_hw_sspp *dpu_rm_reserve_sspp(struct dpu_rm *rm, + struct dpu_global_state *global_state, + struct drm_crtc *crtc, + struct dpu_rm_sspp_requirements *reqs) +{ + struct dpu_hw_sspp *hw_sspp = NULL; + + if (!reqs->scale && !reqs->yuv) + hw_sspp = dpu_rm_try_sspp(rm, global_state, crtc, reqs, SSPP_TYPE_DMA); + if (!hw_sspp && !reqs->yuv) + hw_sspp = dpu_rm_try_sspp(rm, global_state, crtc, reqs, SSPP_TYPE_RGB); + if (!hw_sspp) + hw_sspp = dpu_rm_try_sspp(rm, global_state, crtc, reqs, SSPP_TYPE_VIG); + + return hw_sspp; +} + +/** + * dpu_rm_release_all_sspp - Given the CRTC, release all SSPP + * blocks previously reserved for that use case. + * @global_state: resources shared across multiple kms objects + * @crtc: DRM CRTC handle + */ +void dpu_rm_release_all_sspp(struct dpu_global_state *global_state, + struct drm_crtc *crtc) +{ + uint32_t crtc_id = crtc->base.id; + + _dpu_rm_clear_mapping(global_state->sspp_to_crtc_id, + ARRAY_SIZE(global_state->sspp_to_crtc_id), crtc_id); +} + +static char *dpu_hw_blk_type_name[] = { + [DPU_HW_BLK_TOP] = "TOP", + [DPU_HW_BLK_SSPP] = "SSPP", + [DPU_HW_BLK_LM] = "LM", + [DPU_HW_BLK_CTL] = "CTL", + [DPU_HW_BLK_PINGPONG] = "pingpong", + [DPU_HW_BLK_INTF] = "INTF", + [DPU_HW_BLK_WB] = "WB", + [DPU_HW_BLK_DSPP] = "DSPP", + [DPU_HW_BLK_MERGE_3D] = "merge_3d", + [DPU_HW_BLK_DSC] = "DSC", + [DPU_HW_BLK_CDM] = "CDM", + [DPU_HW_BLK_MAX] = "unknown", +}; + +/** + * dpu_rm_get_assigned_resources - Get hw resources of the given type that are + * assigned to this encoder + * @rm: DPU Resource Manager handle + * @global_state: resources shared across multiple kms objects + * @crtc: DRM CRTC handle + * @type: resource type to return data for + * @blks: pointer to the array to be filled by HW resources + * @blks_size: size of the @blks array + */ +int dpu_rm_get_assigned_resources(struct dpu_rm *rm, + struct dpu_global_state *global_state, struct drm_crtc *crtc, + enum dpu_hw_blk_type type, struct dpu_hw_blk **blks, int blks_size) +{ + uint32_t crtc_id = crtc->base.id; + struct dpu_hw_blk **hw_blks; + uint32_t *hw_to_crtc_id; + int i, num_blks, max_blks; + + switch (type) { + case DPU_HW_BLK_PINGPONG: + case DPU_HW_BLK_DCWB_PINGPONG: + hw_blks = rm->pingpong_blks; + hw_to_crtc_id = global_state->pingpong_to_crtc_id; + max_blks = ARRAY_SIZE(rm->pingpong_blks); + break; + case DPU_HW_BLK_LM: + hw_blks = rm->mixer_blks; + hw_to_crtc_id = global_state->mixer_to_crtc_id; + max_blks = ARRAY_SIZE(rm->mixer_blks); + break; + case DPU_HW_BLK_CTL: + hw_blks = rm->ctl_blks; + hw_to_crtc_id = global_state->ctl_to_crtc_id; + max_blks = ARRAY_SIZE(rm->ctl_blks); + break; + case DPU_HW_BLK_DSPP: + hw_blks = rm->dspp_blks; + hw_to_crtc_id = global_state->dspp_to_crtc_id; + max_blks = ARRAY_SIZE(rm->dspp_blks); + break; + case DPU_HW_BLK_DSC: + hw_blks = rm->dsc_blks; + hw_to_crtc_id = global_state->dsc_to_crtc_id; + max_blks = ARRAY_SIZE(rm->dsc_blks); + break; + case DPU_HW_BLK_CDM: + hw_blks = &rm->cdm_blk; + hw_to_crtc_id = &global_state->cdm_to_crtc_id; + max_blks = 1; + break; + case DPU_HW_BLK_CWB: + hw_blks = rm->cwb_blks; + hw_to_crtc_id = global_state->cwb_to_crtc_id; + max_blks = ARRAY_SIZE(rm->cwb_blks); + break; + default: + DPU_ERROR("blk type %d not managed by rm\n", type); + return 0; + } + + num_blks = 0; + for (i = 0; i < max_blks; i++) { + if (hw_to_crtc_id[i] != crtc_id) + continue; + + if (type == DPU_HW_BLK_PINGPONG) { + struct dpu_hw_pingpong *pp = to_dpu_hw_pingpong(hw_blks[i]); + + if (pp->idx >= PINGPONG_CWB_0) + continue; + } + + if (type == DPU_HW_BLK_DCWB_PINGPONG) { + struct dpu_hw_pingpong *pp = to_dpu_hw_pingpong(hw_blks[i]); + + if (pp->idx < PINGPONG_CWB_0) + continue; + } + + if (num_blks == blks_size) { + DPU_ERROR("More than %d %s assigned to crtc %d\n", + blks_size, dpu_hw_blk_type_name[type], crtc_id); + break; + } + if (!hw_blks[i]) { + DPU_ERROR("%s unavailable to assign to crtc %d\n", + dpu_hw_blk_type_name[type], crtc_id); + break; + } + blks[num_blks++] = hw_blks[i]; + } + + return num_blks; +} + +static void dpu_rm_print_state_helper(struct drm_printer *p, + struct dpu_hw_blk *blk, + uint32_t mapping) +{ + if (!blk) + drm_puts(p, "- "); + else if (!mapping) + drm_puts(p, "# "); + else + drm_printf(p, "%d ", mapping); +} + + +/** + * dpu_rm_print_state - output the RM private state + * @p: DRM printer + * @global_state: global state + */ +void dpu_rm_print_state(struct drm_printer *p, + const struct dpu_global_state *global_state) +{ + const struct dpu_rm *rm = global_state->rm; + int i; + + drm_puts(p, "resource mapping:\n"); + drm_puts(p, "\tpingpong="); + for (i = 0; i < ARRAY_SIZE(global_state->pingpong_to_crtc_id); i++) + dpu_rm_print_state_helper(p, rm->pingpong_blks[i], + global_state->pingpong_to_crtc_id[i]); + drm_puts(p, "\n"); + + drm_puts(p, "\tmixer="); + for (i = 0; i < ARRAY_SIZE(global_state->mixer_to_crtc_id); i++) + dpu_rm_print_state_helper(p, rm->mixer_blks[i], + global_state->mixer_to_crtc_id[i]); + drm_puts(p, "\n"); + + drm_puts(p, "\tctl="); + for (i = 0; i < ARRAY_SIZE(global_state->ctl_to_crtc_id); i++) + dpu_rm_print_state_helper(p, rm->ctl_blks[i], + global_state->ctl_to_crtc_id[i]); + drm_puts(p, "\n"); + + drm_puts(p, "\tdspp="); + for (i = 0; i < ARRAY_SIZE(global_state->dspp_to_crtc_id); i++) + dpu_rm_print_state_helper(p, rm->dspp_blks[i], + global_state->dspp_to_crtc_id[i]); + drm_puts(p, "\n"); + + drm_puts(p, "\tdsc="); + for (i = 0; i < ARRAY_SIZE(global_state->dsc_to_crtc_id); i++) + dpu_rm_print_state_helper(p, rm->dsc_blks[i], + global_state->dsc_to_crtc_id[i]); + drm_puts(p, "\n"); + + drm_puts(p, "\tcdm="); + dpu_rm_print_state_helper(p, rm->cdm_blk, + global_state->cdm_to_crtc_id); + drm_puts(p, "\n"); + + drm_puts(p, "\tsspp="); + /* skip SSPP_NONE and start from the next index */ + for (i = SSPP_NONE + 1; i < ARRAY_SIZE(global_state->sspp_to_crtc_id); i++) + dpu_rm_print_state_helper(p, rm->hw_sspp[i] ? &rm->hw_sspp[i]->base : NULL, + global_state->sspp_to_crtc_id[i]); + drm_puts(p, "\n"); + + drm_puts(p, "\tcwb="); + for (i = 0; i < ARRAY_SIZE(global_state->cwb_to_crtc_id); i++) + dpu_rm_print_state_helper(p, rm->cwb_blks[i], + global_state->cwb_to_crtc_id[i]); + drm_puts(p, "\n"); +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h new file mode 100644 index 000000000..ccd64404f --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h @@ -0,0 +1,129 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2016-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef __DPU_RM_H__ +#define __DPU_RM_H__ + +#include + +#include "msm_kms.h" +#include "dpu_hw_top.h" + +struct dpu_global_state; + +/** + * struct dpu_rm - DPU dynamic hardware resource manager + * @pingpong_blks: array of pingpong hardware resources + * @mixer_blks: array of layer mixer hardware resources + * @ctl_blks: array of ctl hardware resources + * @hw_intf: array of intf hardware resources + * @hw_wb: array of wb hardware resources + * @hw_cwb: array of cwb hardware resources + * @dspp_blks: array of dspp hardware resources + * @hw_sspp: array of sspp hardware resources + * @cdm_blk: cdm hardware resource + * @has_legacy_ctls: DPU uses pre-ACTIVE CTL blocks. + */ +struct dpu_rm { + struct dpu_hw_blk *pingpong_blks[PINGPONG_MAX - PINGPONG_0]; + struct dpu_hw_blk *mixer_blks[LM_MAX - LM_0]; + struct dpu_hw_blk *ctl_blks[CTL_MAX - CTL_0]; + struct dpu_hw_intf *hw_intf[INTF_MAX - INTF_0]; + struct dpu_hw_wb *hw_wb[WB_MAX - WB_0]; + struct dpu_hw_blk *cwb_blks[CWB_MAX - CWB_0]; + struct dpu_hw_blk *dspp_blks[DSPP_MAX - DSPP_0]; + struct dpu_hw_blk *merge_3d_blks[MERGE_3D_MAX - MERGE_3D_0]; + struct dpu_hw_blk *dsc_blks[DSC_MAX - DSC_0]; + struct dpu_hw_sspp *hw_sspp[SSPP_MAX - SSPP_NONE]; + struct dpu_hw_blk *cdm_blk; + bool has_legacy_ctls; +}; + +struct dpu_rm_sspp_requirements { + bool yuv; + bool scale; + bool rot90; +}; + +/** + * struct msm_display_topology - defines a display topology pipeline + * @num_lm: number of layer mixers used + * @num_intf: number of interfaces the panel is mounted on + * @num_dspp: number of dspp blocks used + * @num_dsc: number of Display Stream Compression (DSC) blocks used + * @num_cdm: indicates how many outputs are requesting cdm block for + * this display topology + * @cwb_enabled: indicates whether CWB is enabled for this display topology + */ +struct msm_display_topology { + u32 num_lm; + u32 num_intf; + u32 num_dspp; + u32 num_dsc; + int num_cdm; + bool cwb_enabled; +}; + +int dpu_rm_init(struct drm_device *dev, + struct dpu_rm *rm, + const struct dpu_mdss_cfg *cat, + const struct qcom_ubwc_cfg_data *mdss_data, + void __iomem *mmio); + +int dpu_rm_reserve(struct dpu_rm *rm, + struct dpu_global_state *global_state, + struct drm_crtc *crtc, + struct msm_display_topology *topology); + +void dpu_rm_release(struct dpu_global_state *global_state, + struct drm_crtc *crtc); + +struct dpu_hw_sspp *dpu_rm_reserve_sspp(struct dpu_rm *rm, + struct dpu_global_state *global_state, + struct drm_crtc *crtc, + struct dpu_rm_sspp_requirements *reqs); + +void dpu_rm_release_all_sspp(struct dpu_global_state *global_state, + struct drm_crtc *crtc); + +int dpu_rm_get_assigned_resources(struct dpu_rm *rm, + struct dpu_global_state *global_state, struct drm_crtc *crtc, + enum dpu_hw_blk_type type, struct dpu_hw_blk **blks, int blks_size); + +void dpu_rm_print_state(struct drm_printer *p, + const struct dpu_global_state *global_state); + +/** + * dpu_rm_get_intf - Return a struct dpu_hw_intf instance given it's index. + * @rm: DPU Resource Manager handle + * @intf_idx: INTF's index + */ +static inline struct dpu_hw_intf *dpu_rm_get_intf(struct dpu_rm *rm, enum dpu_intf intf_idx) +{ + return rm->hw_intf[intf_idx - INTF_0]; +} + +/** + * dpu_rm_get_wb - Return a struct dpu_hw_wb instance given it's index. + * @rm: DPU Resource Manager handle + * @wb_idx: WB index + */ +static inline struct dpu_hw_wb *dpu_rm_get_wb(struct dpu_rm *rm, enum dpu_wb wb_idx) +{ + return rm->hw_wb[wb_idx - WB_0]; +} + +/** + * dpu_rm_get_sspp - Return a struct dpu_hw_sspp instance given it's index. + * @rm: DPU Resource Manager handle + * @sspp_idx: SSPP index + */ +static inline struct dpu_hw_sspp *dpu_rm_get_sspp(struct dpu_rm *rm, enum dpu_sspp sspp_idx) +{ + return rm->hw_sspp[sspp_idx - SSPP_NONE]; +} + +#endif /* __DPU_RM_H__ */ + diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h new file mode 100644 index 000000000..805d11749 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h @@ -0,0 +1,1007 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2014-2018, The Linux Foundation. All rights reserved. + */ + +#if !defined(_DPU_TRACE_H_) || defined(TRACE_HEADER_MULTI_READ) +#define _DPU_TRACE_H_ + +#include +#include +#include + +#include +#include "dpu_crtc.h" +#include "dpu_encoder_phys.h" +#include "dpu_hw_mdss.h" +#include "dpu_hw_vbif.h" +#include "dpu_plane.h" + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM dpu +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE dpu_trace + +TRACE_EVENT(dpu_perf_set_qos_luts, + TP_PROTO(u32 pnum, u32 fmt, bool rt, u32 fl, + u32 lut, u32 lut_usage), + TP_ARGS(pnum, fmt, rt, fl, lut, lut_usage), + TP_STRUCT__entry( + __field(u32, pnum) + __field(u32, fmt) + __field(bool, rt) + __field(u32, fl) + __field(u64, lut) + __field(u32, lut_usage) + ), + TP_fast_assign( + __entry->pnum = pnum; + __entry->fmt = fmt; + __entry->rt = rt; + __entry->fl = fl; + __entry->lut = lut; + __entry->lut_usage = lut_usage; + ), + TP_printk("pnum=%d fmt=%x rt=%d fl=%d lut=0x%llx lut_usage=%d", + __entry->pnum, __entry->fmt, + __entry->rt, __entry->fl, + __entry->lut, __entry->lut_usage) +); + +TRACE_EVENT(dpu_perf_set_danger_luts, + TP_PROTO(u32 pnum, u32 fmt, u32 mode, u32 danger_lut, + u32 safe_lut), + TP_ARGS(pnum, fmt, mode, danger_lut, safe_lut), + TP_STRUCT__entry( + __field(u32, pnum) + __field(u32, fmt) + __field(u32, mode) + __field(u32, danger_lut) + __field(u32, safe_lut) + ), + TP_fast_assign( + __entry->pnum = pnum; + __entry->fmt = fmt; + __entry->mode = mode; + __entry->danger_lut = danger_lut; + __entry->safe_lut = safe_lut; + ), + TP_printk("pnum=%d fmt=%x mode=%d luts[0x%x, 0x%x]", + __entry->pnum, __entry->fmt, + __entry->mode, __entry->danger_lut, + __entry->safe_lut) +); + +TRACE_EVENT(dpu_perf_set_ot, + TP_PROTO(u32 pnum, u32 xin_id, u32 rd_lim), + TP_ARGS(pnum, xin_id, rd_lim), + TP_STRUCT__entry( + __field(u32, pnum) + __field(u32, xin_id) + __field(u32, rd_lim) + ), + TP_fast_assign( + __entry->pnum = pnum; + __entry->xin_id = xin_id; + __entry->rd_lim = rd_lim; + ), + TP_printk("pnum:%d xin_id:%d ot:%d", + __entry->pnum, __entry->xin_id, __entry->rd_lim) +) + +TRACE_EVENT(dpu_cmd_release_bw, + TP_PROTO(u32 crtc_id), + TP_ARGS(crtc_id), + TP_STRUCT__entry( + __field(u32, crtc_id) + ), + TP_fast_assign( + __entry->crtc_id = crtc_id; + ), + TP_printk("crtc:%d", __entry->crtc_id) +); + +TRACE_EVENT(tracing_mark_write, + TP_PROTO(int pid, const char *name, bool trace_begin), + TP_ARGS(pid, name, trace_begin), + TP_STRUCT__entry( + __field(int, pid) + __string(trace_name, name) + __field(bool, trace_begin) + ), + TP_fast_assign( + __entry->pid = pid; + __assign_str(trace_name); + __entry->trace_begin = trace_begin; + ), + TP_printk("%s|%d|%s", __entry->trace_begin ? "B" : "E", + __entry->pid, __get_str(trace_name)) +) + +TRACE_EVENT(dpu_trace_counter, + TP_PROTO(int pid, char *name, int value), + TP_ARGS(pid, name, value), + TP_STRUCT__entry( + __field(int, pid) + __string(counter_name, name) + __field(int, value) + ), + TP_fast_assign( + __entry->pid = current->tgid; + __assign_str(counter_name); + __entry->value = value; + ), + TP_printk("%d|%s|%d", __entry->pid, + __get_str(counter_name), __entry->value) +) + +TRACE_EVENT(dpu_perf_crtc_update, + TP_PROTO(u32 crtc, u64 bw_ctl, u32 core_clk_rate, + bool stop_req, bool update_bus, bool update_clk), + TP_ARGS(crtc, bw_ctl, core_clk_rate, stop_req, update_bus, update_clk), + TP_STRUCT__entry( + __field(u32, crtc) + __field(u64, bw_ctl) + __field(u32, core_clk_rate) + __field(bool, stop_req) + __field(u32, update_bus) + __field(u32, update_clk) + ), + TP_fast_assign( + __entry->crtc = crtc; + __entry->bw_ctl = bw_ctl; + __entry->core_clk_rate = core_clk_rate; + __entry->stop_req = stop_req; + __entry->update_bus = update_bus; + __entry->update_clk = update_clk; + ), + TP_printk( + "crtc=%d bw_ctl=%llu clk_rate=%u stop_req=%d u_bus=%d u_clk=%d", + __entry->crtc, + __entry->bw_ctl, + __entry->core_clk_rate, + __entry->stop_req, + __entry->update_bus, + __entry->update_clk) +); + +DECLARE_EVENT_CLASS(dpu_irq_template, + TP_PROTO(unsigned int irq_reg, unsigned int irq_bit), + TP_ARGS(irq_reg, irq_bit), + TP_STRUCT__entry( + __field( unsigned int, irq_reg ) + __field( unsigned int, irq_bit ) + ), + TP_fast_assign( + __entry->irq_reg = irq_reg; + __entry->irq_bit = irq_bit; + ), + TP_printk("IRQ=[%d, %d]", __entry->irq_reg, __entry->irq_bit) +); +DEFINE_EVENT(dpu_irq_template, dpu_irq_register_success, + TP_PROTO(unsigned int irq_reg, unsigned int irq_bit), + TP_ARGS(irq_reg, irq_bit) +); +DEFINE_EVENT(dpu_irq_template, dpu_irq_unregister_success, + TP_PROTO(unsigned int irq_reg, unsigned int irq_bit), + TP_ARGS(irq_reg, irq_bit) +); + +TRACE_EVENT(dpu_enc_irq_wait_success, + TP_PROTO(uint32_t drm_id, void *func, + unsigned int irq_reg, unsigned int irq_bit, enum dpu_pingpong pp_idx, int atomic_cnt), + TP_ARGS(drm_id, func, irq_reg, irq_bit, pp_idx, atomic_cnt), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( void *, func ) + __field( unsigned int, irq_reg ) + __field( unsigned int, irq_bit ) + __field( enum dpu_pingpong, pp_idx ) + __field( int, atomic_cnt ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->func = func; + __entry->irq_reg = irq_reg; + __entry->irq_bit = irq_bit; + __entry->pp_idx = pp_idx; + __entry->atomic_cnt = atomic_cnt; + ), + TP_printk("id=%u, callback=%ps, IRQ=[%d, %d], pp=%d, atomic_cnt=%d", + __entry->drm_id, __entry->func, + __entry->irq_reg, __entry->irq_bit, __entry->pp_idx, __entry->atomic_cnt) +); + +DECLARE_EVENT_CLASS(dpu_drm_obj_template, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + ), + TP_printk("id=%u", __entry->drm_id) +); +DEFINE_EVENT(dpu_drm_obj_template, dpu_enc_atomic_check, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); +DEFINE_EVENT(dpu_drm_obj_template, dpu_enc_mode_set, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); +DEFINE_EVENT(dpu_drm_obj_template, dpu_enc_disable, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); +DEFINE_EVENT(dpu_drm_obj_template, dpu_enc_kickoff, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); +DEFINE_EVENT(dpu_drm_obj_template, dpu_enc_prepare_kickoff, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); +DEFINE_EVENT(dpu_drm_obj_template, dpu_enc_prepare_kickoff_reset, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); +DEFINE_EVENT(dpu_drm_obj_template, dpu_crtc_complete_flip, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); +DEFINE_EVENT(dpu_drm_obj_template, dpu_crtc_vblank_cb, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); +DEFINE_EVENT(dpu_drm_obj_template, dpu_crtc_complete_commit, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); +DEFINE_EVENT(dpu_drm_obj_template, dpu_kms_commit, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); +DEFINE_EVENT(dpu_drm_obj_template, dpu_kms_wait_for_commit_done, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); +DEFINE_EVENT(dpu_drm_obj_template, dpu_crtc_runtime_resume, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); +DEFINE_EVENT(dpu_drm_obj_template, dpu_enc_rc_enable, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); +DEFINE_EVENT(dpu_drm_obj_template, dpu_enc_rc_disable, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); + +TRACE_EVENT(dpu_enc_enable, + TP_PROTO(uint32_t drm_id, int hdisplay, int vdisplay), + TP_ARGS(drm_id, hdisplay, vdisplay), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( int, hdisplay ) + __field( int, vdisplay ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->hdisplay = hdisplay; + __entry->vdisplay = vdisplay; + ), + TP_printk("id=%u, mode=%dx%d", + __entry->drm_id, __entry->hdisplay, __entry->vdisplay) +); + +DECLARE_EVENT_CLASS(dpu_enc_keyval_template, + TP_PROTO(uint32_t drm_id, int val), + TP_ARGS(drm_id, val), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( int, val ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->val = val; + ), + TP_printk("id=%u, val=%d", __entry->drm_id, __entry->val) +); +DEFINE_EVENT(dpu_enc_keyval_template, dpu_enc_underrun_cb, + TP_PROTO(uint32_t drm_id, int count), + TP_ARGS(drm_id, count) +); +DEFINE_EVENT(dpu_enc_keyval_template, dpu_enc_trigger_start, + TP_PROTO(uint32_t drm_id, int ctl_idx), + TP_ARGS(drm_id, ctl_idx) +); + +TRACE_EVENT(dpu_enc_atomic_check_flags, + TP_PROTO(uint32_t drm_id, unsigned int flags), + TP_ARGS(drm_id, flags), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( unsigned int, flags ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->flags = flags; + ), + TP_printk("id=%u, flags=%u", + __entry->drm_id, __entry->flags) +); + +DECLARE_EVENT_CLASS(dpu_enc_id_enable_template, + TP_PROTO(uint32_t drm_id, bool enable), + TP_ARGS(drm_id, enable), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( bool, enable ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->enable = enable; + ), + TP_printk("id=%u, enable=%s", + __entry->drm_id, __entry->enable ? "true" : "false") +); +DEFINE_EVENT(dpu_enc_id_enable_template, dpu_enc_vblank_cb, + TP_PROTO(uint32_t drm_id, bool enable), + TP_ARGS(drm_id, enable) +); +DEFINE_EVENT(dpu_enc_id_enable_template, dpu_enc_phys_cmd_connect_te, + TP_PROTO(uint32_t drm_id, bool enable), + TP_ARGS(drm_id, enable) +); + +TRACE_EVENT(dpu_enc_rc, + TP_PROTO(uint32_t drm_id, u32 sw_event, bool idle_pc_supported, + int rc_state, const char *stage), + TP_ARGS(drm_id, sw_event, idle_pc_supported, rc_state, stage), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( u32, sw_event ) + __field( bool, idle_pc_supported ) + __field( int, rc_state ) + __string( stage_str, stage ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->sw_event = sw_event; + __entry->idle_pc_supported = idle_pc_supported; + __entry->rc_state = rc_state; + __assign_str(stage_str); + ), + TP_printk("%s: id:%u, sw_event:%d, idle_pc_supported:%s, rc_state:%d", + __get_str(stage_str), __entry->drm_id, __entry->sw_event, + __entry->idle_pc_supported ? "true" : "false", + __entry->rc_state) +); + +TRACE_EVENT(dpu_enc_frame_done_cb_not_busy, + TP_PROTO(uint32_t drm_id, u32 event, char *intf_mode, enum dpu_intf intf_idx, + enum dpu_wb wb_idx), + TP_ARGS(drm_id, event, intf_mode, intf_idx, wb_idx), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( u32, event ) + __string( intf_mode_str, intf_mode ) + __field( enum dpu_intf, intf_idx ) + __field( enum dpu_wb, wb_idx ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->event = event; + __assign_str(intf_mode_str); + __entry->intf_idx = intf_idx; + __entry->wb_idx = wb_idx; + ), + TP_printk("id=%u, event=%u, intf_mode=%s intf=%d wb=%d", __entry->drm_id, + __entry->event, __get_str(intf_mode_str), + __entry->intf_idx, __entry->wb_idx) +); + +TRACE_EVENT(dpu_enc_frame_done_cb, + TP_PROTO(uint32_t drm_id, unsigned int idx, + unsigned long frame_busy_mask), + TP_ARGS(drm_id, idx, frame_busy_mask), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( unsigned int, idx ) + __field( unsigned long, frame_busy_mask ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->idx = idx; + __entry->frame_busy_mask = frame_busy_mask; + ), + TP_printk("id=%u, idx=%u, frame_busy_mask=%lx", __entry->drm_id, + __entry->idx, __entry->frame_busy_mask) +); + +TRACE_EVENT(dpu_enc_trigger_flush, + TP_PROTO(uint32_t drm_id, char *intf_mode, enum dpu_intf intf_idx, enum dpu_wb wb_idx, + int pending_kickoff_cnt, int ctl_idx, u32 extra_flush_bits, + u32 pending_flush_ret), + TP_ARGS(drm_id, intf_mode, intf_idx, wb_idx, pending_kickoff_cnt, ctl_idx, + extra_flush_bits, pending_flush_ret), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __string( intf_mode_str, intf_mode ) + __field( enum dpu_intf, intf_idx ) + __field( enum dpu_wb, wb_idx ) + __field( int, pending_kickoff_cnt ) + __field( int, ctl_idx ) + __field( u32, extra_flush_bits ) + __field( u32, pending_flush_ret ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __assign_str(intf_mode_str); + __entry->intf_idx = intf_idx; + __entry->wb_idx = wb_idx; + __entry->pending_kickoff_cnt = pending_kickoff_cnt; + __entry->ctl_idx = ctl_idx; + __entry->extra_flush_bits = extra_flush_bits; + __entry->pending_flush_ret = pending_flush_ret; + ), + TP_printk("id=%u, intf_mode=%s, intf_idx=%d, wb_idx=%d, pending_kickoff_cnt=%d ctl_idx=%d " + "extra_flush_bits=0x%x pending_flush_ret=0x%x", + __entry->drm_id, __get_str(intf_mode_str), __entry->intf_idx, __entry->wb_idx, + __entry->pending_kickoff_cnt, __entry->ctl_idx, + __entry->extra_flush_bits, __entry->pending_flush_ret) +); + +DECLARE_EVENT_CLASS(dpu_id_event_template, + TP_PROTO(uint32_t drm_id, u32 event), + TP_ARGS(drm_id, event), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( u32, event ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->event = event; + ), + TP_printk("id=%u, event=%u", __entry->drm_id, __entry->event) +); +DEFINE_EVENT(dpu_id_event_template, dpu_enc_frame_done_timeout, + TP_PROTO(uint32_t drm_id, u32 event), + TP_ARGS(drm_id, event) +); +DEFINE_EVENT(dpu_id_event_template, dpu_crtc_frame_event_cb, + TP_PROTO(uint32_t drm_id, u32 event), + TP_ARGS(drm_id, event) +); +DEFINE_EVENT(dpu_id_event_template, dpu_crtc_frame_event_done, + TP_PROTO(uint32_t drm_id, u32 event), + TP_ARGS(drm_id, event) +); +DEFINE_EVENT(dpu_id_event_template, dpu_crtc_frame_event_more_pending, + TP_PROTO(uint32_t drm_id, u32 event), + TP_ARGS(drm_id, event) +); + +TRACE_EVENT(dpu_enc_wait_event_timeout, + TP_PROTO(uint32_t drm_id, unsigned int irq_reg, unsigned int irq_bit, int rc, s64 time, + s64 expected_time, int atomic_cnt), + TP_ARGS(drm_id, irq_reg, irq_bit, rc, time, expected_time, atomic_cnt), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( unsigned int, irq_reg ) + __field( unsigned int, irq_bit ) + __field( int, rc ) + __field( s64, time ) + __field( s64, expected_time ) + __field( int, atomic_cnt ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->irq_reg = irq_reg; + __entry->irq_bit = irq_bit; + __entry->rc = rc; + __entry->time = time; + __entry->expected_time = expected_time; + __entry->atomic_cnt = atomic_cnt; + ), + TP_printk("id=%u, IRQ=[%d, %d], rc=%d, time=%lld, expected=%lld cnt=%d", + __entry->drm_id, __entry->irq_reg, __entry->irq_bit, __entry->rc, __entry->time, + __entry->expected_time, __entry->atomic_cnt) +); + +TRACE_EVENT(dpu_enc_phys_cmd_irq_enable, + TP_PROTO(uint32_t drm_id, enum dpu_pingpong pp, + int refcnt), + TP_ARGS(drm_id, pp, refcnt), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( enum dpu_pingpong, pp ) + __field( int, refcnt ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->pp = pp; + __entry->refcnt = refcnt; + ), + TP_printk("id=%u, pp=%d, refcnt=%d", __entry->drm_id, + __entry->pp, + __entry->refcnt) +); + +TRACE_EVENT(dpu_enc_phys_cmd_irq_disable, + TP_PROTO(uint32_t drm_id, enum dpu_pingpong pp, + int refcnt), + TP_ARGS(drm_id, pp, refcnt), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( enum dpu_pingpong, pp ) + __field( int, refcnt ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->pp = pp; + __entry->refcnt = refcnt; + ), + TP_printk("id=%u, pp=%d, refcnt=%d", __entry->drm_id, + __entry->pp, + __entry->refcnt) +); + +TRACE_EVENT(dpu_enc_phys_cmd_pp_tx_done, + TP_PROTO(uint32_t drm_id, enum dpu_pingpong pp, int new_count, + u32 event), + TP_ARGS(drm_id, pp, new_count, event), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( enum dpu_pingpong, pp ) + __field( int, new_count ) + __field( u32, event ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->pp = pp; + __entry->new_count = new_count; + __entry->event = event; + ), + TP_printk("id=%u, pp=%d, new_count=%d, event=%u", __entry->drm_id, + __entry->pp, __entry->new_count, __entry->event) +); + +TRACE_EVENT(dpu_enc_phys_cmd_pdone_timeout, + TP_PROTO(uint32_t drm_id, enum dpu_pingpong pp, int timeout_count, + int kickoff_count, u32 event), + TP_ARGS(drm_id, pp, timeout_count, kickoff_count, event), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( enum dpu_pingpong, pp ) + __field( int, timeout_count ) + __field( int, kickoff_count ) + __field( u32, event ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->pp = pp; + __entry->timeout_count = timeout_count; + __entry->kickoff_count = kickoff_count; + __entry->event = event; + ), + TP_printk("id=%u, pp=%d, timeout_count=%d, kickoff_count=%d, event=%u", + __entry->drm_id, __entry->pp, __entry->timeout_count, + __entry->kickoff_count, __entry->event) +); + +TRACE_EVENT(dpu_enc_phys_vid_post_kickoff, + TP_PROTO(uint32_t drm_id, enum dpu_intf intf_idx), + TP_ARGS(drm_id, intf_idx), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( enum dpu_intf, intf_idx ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->intf_idx = intf_idx; + ), + TP_printk("id=%u, intf_idx=%d", __entry->drm_id, __entry->intf_idx) +); + +TRACE_EVENT(dpu_enc_phys_vid_irq_enable, + TP_PROTO(uint32_t drm_id, enum dpu_intf intf_idx, + int refcnt), + TP_ARGS(drm_id, intf_idx, refcnt), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( enum dpu_intf, intf_idx ) + __field( int, refcnt ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->intf_idx = intf_idx; + __entry->refcnt = refcnt; + ), + TP_printk("id=%u, intf_idx=%d refcnt=%d", __entry->drm_id, + __entry->intf_idx, + __entry->drm_id) +); + +TRACE_EVENT(dpu_enc_phys_vid_irq_disable, + TP_PROTO(uint32_t drm_id, enum dpu_intf intf_idx, + int refcnt), + TP_ARGS(drm_id, intf_idx, refcnt), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( enum dpu_intf, intf_idx ) + __field( int, refcnt ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->intf_idx = intf_idx; + __entry->refcnt = refcnt; + ), + TP_printk("id=%u, intf_idx=%d refcnt=%d", __entry->drm_id, + __entry->intf_idx, + __entry->drm_id) +); + +TRACE_EVENT(dpu_crtc_setup_mixer, + TP_PROTO(uint32_t crtc_id, uint32_t plane_id, + struct drm_plane_state *state, struct dpu_plane_state *pstate, + uint32_t stage_idx, uint32_t pixel_format, + struct dpu_sw_pipe *pipe, uint64_t modifier), + TP_ARGS(crtc_id, plane_id, state, pstate, stage_idx, + pixel_format, pipe, modifier), + TP_STRUCT__entry( + __field( uint32_t, crtc_id ) + __field( uint32_t, plane_id ) + __field( uint32_t, fb_id ) + __field_struct( struct drm_rect, src_rect ) + __field_struct( struct drm_rect, dst_rect ) + __field( uint32_t, stage_idx ) + __field( enum dpu_stage, stage ) + __field( enum dpu_sspp, sspp ) + __field( uint32_t, multirect_idx ) + __field( uint32_t, multirect_mode ) + __field( uint32_t, pixel_format ) + __field( uint64_t, modifier ) + ), + TP_fast_assign( + __entry->crtc_id = crtc_id; + __entry->plane_id = plane_id; + __entry->fb_id = state ? state->fb->base.id : 0; + __entry->src_rect = drm_plane_state_src(state); + __entry->dst_rect = drm_plane_state_dest(state); + __entry->stage_idx = stage_idx; + __entry->stage = pstate->stage; + __entry->sspp = pipe->sspp->idx; + __entry->multirect_idx = pipe->multirect_index; + __entry->multirect_mode = pipe->multirect_mode; + __entry->pixel_format = pixel_format; + __entry->modifier = modifier; + ), + TP_printk("crtc_id:%u plane_id:%u fb_id:%u src:" DRM_RECT_FP_FMT + " dst:" DRM_RECT_FMT " stage_idx:%u stage:%d, sspp:%d " + "multirect_index:%d multirect_mode:%u pix_format:%u " + "modifier:%llu", + __entry->crtc_id, __entry->plane_id, __entry->fb_id, + DRM_RECT_FP_ARG(&__entry->src_rect), + DRM_RECT_ARG(&__entry->dst_rect), + __entry->stage_idx, __entry->stage, __entry->sspp, + __entry->multirect_idx, __entry->multirect_mode, + __entry->pixel_format, __entry->modifier) +); + +TRACE_EVENT(dpu_crtc_setup_lm_bounds, + TP_PROTO(uint32_t drm_id, int mixer, struct drm_rect *bounds), + TP_ARGS(drm_id, mixer, bounds), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( int, mixer ) + __field_struct( struct drm_rect, bounds ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->mixer = mixer; + __entry->bounds = *bounds; + ), + TP_printk("id:%u mixer:%d bounds:" DRM_RECT_FMT, __entry->drm_id, + __entry->mixer, DRM_RECT_ARG(&__entry->bounds)) +); + +TRACE_EVENT(dpu_crtc_vblank_enable, + TP_PROTO(uint32_t drm_id, uint32_t enc_id, bool enable, + struct dpu_crtc *crtc), + TP_ARGS(drm_id, enc_id, enable, crtc), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( uint32_t, enc_id ) + __field( bool, enable ) + __field( bool, enabled ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->enc_id = enc_id; + __entry->enable = enable; + __entry->enabled = crtc->enabled; + ), + TP_printk("id:%u encoder:%u enable:%s state{enabled:%s}", + __entry->drm_id, __entry->enc_id, + __entry->enable ? "true" : "false", + __entry->enabled ? "true" : "false") +); + +DECLARE_EVENT_CLASS(dpu_crtc_enable_template, + TP_PROTO(uint32_t drm_id, bool enable, struct dpu_crtc *crtc), + TP_ARGS(drm_id, enable, crtc), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( bool, enable ) + __field( bool, enabled ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->enable = enable; + __entry->enabled = crtc->enabled; + ), + TP_printk("id:%u enable:%s state{enabled:%s}", + __entry->drm_id, __entry->enable ? "true" : "false", + __entry->enabled ? "true" : "false") +); +DEFINE_EVENT(dpu_crtc_enable_template, dpu_crtc_enable, + TP_PROTO(uint32_t drm_id, bool enable, struct dpu_crtc *crtc), + TP_ARGS(drm_id, enable, crtc) +); +DEFINE_EVENT(dpu_crtc_enable_template, dpu_crtc_disable, + TP_PROTO(uint32_t drm_id, bool enable, struct dpu_crtc *crtc), + TP_ARGS(drm_id, enable, crtc) +); +DEFINE_EVENT(dpu_crtc_enable_template, dpu_crtc_vblank, + TP_PROTO(uint32_t drm_id, bool enable, struct dpu_crtc *crtc), + TP_ARGS(drm_id, enable, crtc) +); + +TRACE_EVENT(dpu_crtc_disable_frame_pending, + TP_PROTO(uint32_t drm_id, int frame_pending), + TP_ARGS(drm_id, frame_pending), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( int, frame_pending ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->frame_pending = frame_pending; + ), + TP_printk("id:%u frame_pending:%d", __entry->drm_id, + __entry->frame_pending) +); + +TRACE_EVENT(dpu_plane_set_scanout, + TP_PROTO(struct dpu_sw_pipe *pipe, struct dpu_hw_fmt_layout *layout), + TP_ARGS(pipe, layout), + TP_STRUCT__entry( + __field( enum dpu_sspp, index ) + __field_struct( struct dpu_hw_fmt_layout, layout ) + __field( enum dpu_sspp_multirect_index, multirect_index) + ), + TP_fast_assign( + __entry->index = pipe->sspp->idx; + __entry->layout = *layout; + __entry->multirect_index = pipe->multirect_index; + ), + TP_printk("index:%d layout:{%ux%u @ [%u/%u, %u/%u, %u/%u, %u/%u]} " + "multirect_index:%d", __entry->index, __entry->layout.width, + __entry->layout.height, __entry->layout.plane_addr[0], + __entry->layout.plane_size[0], + __entry->layout.plane_addr[1], + __entry->layout.plane_size[1], + __entry->layout.plane_addr[2], + __entry->layout.plane_size[2], + __entry->layout.plane_addr[3], + __entry->layout.plane_size[3], __entry->multirect_index) +); + +TRACE_EVENT(dpu_plane_disable, + TP_PROTO(uint32_t drm_id, bool is_virtual, uint32_t multirect_mode), + TP_ARGS(drm_id, is_virtual, multirect_mode), + TP_STRUCT__entry( + __field( uint32_t, drm_id ) + __field( bool, is_virtual ) + __field( uint32_t, multirect_mode ) + ), + TP_fast_assign( + __entry->drm_id = drm_id; + __entry->is_virtual = is_virtual; + __entry->multirect_mode = multirect_mode; + ), + TP_printk("id:%u is_virtual:%s multirect_mode:%u", __entry->drm_id, + __entry->is_virtual ? "true" : "false", + __entry->multirect_mode) +); + +DECLARE_EVENT_CLASS(dpu_rm_iter_template, + TP_PROTO(uint32_t id, uint32_t enc_id), + TP_ARGS(id, enc_id), + TP_STRUCT__entry( + __field( uint32_t, id ) + __field( uint32_t, enc_id ) + ), + TP_fast_assign( + __entry->id = id; + __entry->enc_id = enc_id; + ), + TP_printk("id:%d enc_id:%u", __entry->id, __entry->enc_id) +); +DEFINE_EVENT(dpu_rm_iter_template, dpu_rm_reserve_intf, + TP_PROTO(uint32_t id, uint32_t enc_id), + TP_ARGS(id, enc_id) +); +DEFINE_EVENT(dpu_rm_iter_template, dpu_rm_reserve_ctls, + TP_PROTO(uint32_t id, uint32_t enc_id), + TP_ARGS(id, enc_id) +); + +TRACE_EVENT(dpu_rm_reserve_lms, + TP_PROTO(uint32_t id, uint32_t enc_id, uint32_t pp_id), + TP_ARGS(id, enc_id, pp_id), + TP_STRUCT__entry( + __field( uint32_t, id ) + __field( uint32_t, enc_id ) + __field( uint32_t, pp_id ) + ), + TP_fast_assign( + __entry->id = id; + __entry->enc_id = enc_id; + __entry->pp_id = pp_id; + ), + TP_printk("id:%d enc_id:%u pp_id:%u", __entry->id, + __entry->enc_id, __entry->pp_id) +); + +TRACE_EVENT(dpu_vbif_wait_xin_halt_fail, + TP_PROTO(u32 xin_id), + TP_ARGS(xin_id), + TP_STRUCT__entry( + __field( u32, xin_id ) + ), + TP_fast_assign( + __entry->xin_id = xin_id; + ), + TP_printk("xin_id:%u", __entry->xin_id) +); + +TRACE_EVENT(dpu_pp_connect_ext_te, + TP_PROTO(enum dpu_pingpong pp, u32 cfg), + TP_ARGS(pp, cfg), + TP_STRUCT__entry( + __field( enum dpu_pingpong, pp ) + __field( u32, cfg ) + ), + TP_fast_assign( + __entry->pp = pp; + __entry->cfg = cfg; + ), + TP_printk("pp:%d cfg:%u", __entry->pp, __entry->cfg) +); + +TRACE_EVENT(dpu_intf_connect_ext_te, + TP_PROTO(enum dpu_intf intf, u32 cfg), + TP_ARGS(intf, cfg), + TP_STRUCT__entry( + __field( enum dpu_intf, intf ) + __field( u32, cfg ) + ), + TP_fast_assign( + __entry->intf = intf; + __entry->cfg = cfg; + ), + TP_printk("intf:%d cfg:%u", __entry->intf, __entry->cfg) +); + +TRACE_EVENT(dpu_core_irq_register_callback, + TP_PROTO(unsigned int irq_reg, unsigned int irq_bit, void *callback), + TP_ARGS(irq_reg, irq_bit, callback), + TP_STRUCT__entry( + __field( unsigned int, irq_reg ) + __field( unsigned int, irq_bit ) + __field( void *, callback) + ), + TP_fast_assign( + __entry->irq_reg = irq_reg; + __entry->irq_bit = irq_bit; + __entry->callback = callback; + ), + TP_printk("IRQ=[%d, %d] callback:%ps", __entry->irq_reg, __entry->irq_bit, + __entry->callback) +); + +TRACE_EVENT(dpu_core_irq_unregister_callback, + TP_PROTO(unsigned int irq_reg, unsigned int irq_bit), + TP_ARGS(irq_reg, irq_bit), + TP_STRUCT__entry( + __field( unsigned int, irq_reg ) + __field( unsigned int, irq_bit ) + ), + TP_fast_assign( + __entry->irq_reg = irq_reg; + __entry->irq_bit = irq_bit; + ), + TP_printk("IRQ=[%d, %d]", __entry->irq_reg, __entry->irq_bit) +); + +TRACE_EVENT(dpu_core_perf_update_clk, + TP_PROTO(struct drm_device *dev, bool stop_req, u64 clk_rate), + TP_ARGS(dev, stop_req, clk_rate), + TP_STRUCT__entry( + __string( dev_name, dev->unique ) + __field( bool, stop_req ) + __field( u64, clk_rate ) + ), + TP_fast_assign( + __assign_str(dev_name); + __entry->stop_req = stop_req; + __entry->clk_rate = clk_rate; + ), + TP_printk("dev:%s stop_req:%s clk_rate:%llu", __get_str(dev_name), + __entry->stop_req ? "true" : "false", __entry->clk_rate) +); + +TRACE_EVENT(dpu_hw_ctl_update_pending_flush, + TP_PROTO(u32 new_bits, u32 pending_mask), + TP_ARGS(new_bits, pending_mask), + TP_STRUCT__entry( + __field( u32, new_bits ) + __field( u32, pending_mask ) + ), + TP_fast_assign( + __entry->new_bits = new_bits; + __entry->pending_mask = pending_mask; + ), + TP_printk("new=%x existing=%x", __entry->new_bits, + __entry->pending_mask) +); + +DECLARE_EVENT_CLASS(dpu_hw_ctl_pending_flush_template, + TP_PROTO(u32 pending_mask, u32 ctl_flush), + TP_ARGS(pending_mask, ctl_flush), + TP_STRUCT__entry( + __field( u32, pending_mask ) + __field( u32, ctl_flush ) + ), + TP_fast_assign( + __entry->pending_mask = pending_mask; + __entry->ctl_flush = ctl_flush; + ), + TP_printk("pending_mask=%x CTL_FLUSH=%x", __entry->pending_mask, + __entry->ctl_flush) +); +DEFINE_EVENT(dpu_hw_ctl_pending_flush_template, dpu_hw_ctl_clear_pending_flush, + TP_PROTO(u32 pending_mask, u32 ctl_flush), + TP_ARGS(pending_mask, ctl_flush) +); +DEFINE_EVENT(dpu_hw_ctl_pending_flush_template, + dpu_hw_ctl_trigger_pending_flush, + TP_PROTO(u32 pending_mask, u32 ctl_flush), + TP_ARGS(pending_mask, ctl_flush) +); +DEFINE_EVENT(dpu_hw_ctl_pending_flush_template, dpu_hw_ctl_trigger_prepare, + TP_PROTO(u32 pending_mask, u32 ctl_flush), + TP_ARGS(pending_mask, ctl_flush) +); +DEFINE_EVENT(dpu_hw_ctl_pending_flush_template, dpu_hw_ctl_trigger_start, + TP_PROTO(u32 pending_mask, u32 ctl_flush), + TP_ARGS(pending_mask, ctl_flush) +); + +#define DPU_ATRACE_END(name) trace_tracing_mark_write(current->tgid, name, 0) +#define DPU_ATRACE_BEGIN(name) trace_tracing_mark_write(current->tgid, name, 1) +#define DPU_ATRACE_FUNC() DPU_ATRACE_BEGIN(__func__) + +#define DPU_ATRACE_INT(name, value) \ + trace_dpu_trace_counter(current->tgid, name, value) + +#endif /* _DPU_TRACE_H_ */ + +/* This part must be outside protection */ +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#include diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c new file mode 100644 index 000000000..0c6fa9bb0 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c @@ -0,0 +1,314 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. + */ + +#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ + +#include +#include + +#include "dpu_vbif.h" +#include "dpu_hw_vbif.h" +#include "dpu_trace.h" + +/** + * _dpu_vbif_wait_for_xin_halt - wait for the xin to halt + * @vbif: Pointer to hardware vbif driver + * @xin_id: Client interface identifier + * @return: 0 if success; error code otherwise + */ +static int _dpu_vbif_wait_for_xin_halt(struct dpu_hw_vbif *vbif, u32 xin_id) +{ + ktime_t timeout; + bool status; + int rc; + + if (!vbif || !vbif->cap || !vbif->ops.get_halt_ctrl) { + DPU_ERROR("invalid arguments vbif %d\n", vbif != NULL); + return -EINVAL; + } + + timeout = ktime_add_us(ktime_get(), vbif->cap->xin_halt_timeout); + for (;;) { + status = vbif->ops.get_halt_ctrl(vbif, xin_id); + if (status) + break; + if (ktime_compare_safe(ktime_get(), timeout) > 0) { + status = vbif->ops.get_halt_ctrl(vbif, xin_id); + break; + } + usleep_range(501, 1000); + } + + if (!status) { + rc = -ETIMEDOUT; + DPU_ERROR("VBIF client %d not halting. TIMEDOUT.\n", xin_id); + } else { + rc = 0; + DRM_DEBUG_ATOMIC("VBIF client %d is halted\n", xin_id); + } + + return rc; +} + +/** + * _dpu_vbif_apply_dynamic_ot_limit - determine OT based on usecase parameters + * @vbif: Pointer to hardware vbif driver + * @ot_lim: Pointer to OT limit to be modified + * @params: Pointer to usecase parameters + */ +static void _dpu_vbif_apply_dynamic_ot_limit(struct dpu_hw_vbif *vbif, + u32 *ot_lim, struct dpu_vbif_set_ot_params *params) +{ + u64 pps; + const struct dpu_vbif_dynamic_ot_tbl *tbl; + u32 i; + + if (!vbif || !(vbif->cap->features & BIT(DPU_VBIF_QOS_OTLIM))) + return; + + /* Dynamic OT setting done only for WFD */ + if (!params->is_wfd) + return; + + pps = params->frame_rate; + pps *= params->width; + pps *= params->height; + + tbl = params->rd ? &vbif->cap->dynamic_ot_rd_tbl : + &vbif->cap->dynamic_ot_wr_tbl; + + for (i = 0; i < tbl->count; i++) { + if (pps <= tbl->cfg[i].pps) { + *ot_lim = tbl->cfg[i].ot_limit; + break; + } + } + + DRM_DEBUG_ATOMIC("VBIF xin:%d w:%d h:%d fps:%d pps:%llu ot:%u\n", + params->xin_id, + params->width, params->height, params->frame_rate, + pps, *ot_lim); +} + +/** + * _dpu_vbif_get_ot_limit - get OT based on usecase & configuration parameters + * @vbif: Pointer to hardware vbif driver + * @params: Pointer to usecase parameters + * @return: OT limit + */ +static u32 _dpu_vbif_get_ot_limit(struct dpu_hw_vbif *vbif, + struct dpu_vbif_set_ot_params *params) +{ + u32 ot_lim = 0; + u32 val; + + if (!vbif || !vbif->cap) { + DPU_ERROR("invalid arguments vbif %d\n", vbif != NULL); + return -EINVAL; + } + + if (vbif->cap->default_ot_wr_limit && !params->rd) + ot_lim = vbif->cap->default_ot_wr_limit; + else if (vbif->cap->default_ot_rd_limit && params->rd) + ot_lim = vbif->cap->default_ot_rd_limit; + + /* + * If default ot is not set from dt/catalog, + * then do not configure it. + */ + if (ot_lim == 0) + goto exit; + + /* Modify the limits if the target and the use case requires it */ + _dpu_vbif_apply_dynamic_ot_limit(vbif, &ot_lim, params); + + if (vbif && vbif->ops.get_limit_conf) { + val = vbif->ops.get_limit_conf(vbif, + params->xin_id, params->rd); + if (val == ot_lim) + ot_lim = 0; + } + +exit: + DRM_DEBUG_ATOMIC("VBIF xin:%d ot_lim:%d\n", params->xin_id, ot_lim); + return ot_lim; +} + +/** + * dpu_vbif_set_ot_limit - set OT based on usecase & configuration parameters + * @dpu_kms: DPU handler + * @params: Pointer to usecase parameters + * + * Note this function would block waiting for bus halt. + */ +void dpu_vbif_set_ot_limit(struct dpu_kms *dpu_kms, + struct dpu_vbif_set_ot_params *params) +{ + struct dpu_hw_vbif *vbif; + u32 ot_lim; + int ret; + + vbif = dpu_kms->hw_vbif; + if (!vbif) { + DRM_DEBUG_ATOMIC("invalid arguments vbif %d\n", vbif != NULL); + return; + } + + if (!vbif->ops.set_limit_conf || !vbif->ops.set_halt_ctrl) + return; + + /* set write_gather_en for all write clients */ + if (vbif->ops.set_write_gather_en && !params->rd) + vbif->ops.set_write_gather_en(vbif, params->xin_id); + + ot_lim = _dpu_vbif_get_ot_limit(vbif, params) & 0xFF; + + if (ot_lim == 0) + return; + + trace_dpu_perf_set_ot(params->num, params->xin_id, ot_lim); + + vbif->ops.set_limit_conf(vbif, params->xin_id, params->rd, ot_lim); + + vbif->ops.set_halt_ctrl(vbif, params->xin_id, true); + + ret = _dpu_vbif_wait_for_xin_halt(vbif, params->xin_id); + if (ret) + trace_dpu_vbif_wait_xin_halt_fail(params->xin_id); + + vbif->ops.set_halt_ctrl(vbif, params->xin_id, false); +} + +/** + * dpu_vbif_set_qos_remap - set QoS priority level remap + * @dpu_kms: DPU handler + * @params: Pointer to QoS configuration parameters + */ +void dpu_vbif_set_qos_remap(struct dpu_kms *dpu_kms, + struct dpu_vbif_set_qos_params *params) +{ + struct dpu_hw_vbif *vbif; + const struct dpu_vbif_qos_tbl *qos_tbl; + int i; + + if (!params) { + DPU_ERROR("invalid arguments\n"); + return; + } + + vbif = dpu_kms->hw_vbif; + + if (!vbif || !vbif->cap) { + DPU_ERROR("invalid vbif\n"); + return; + } + + if (!vbif->ops.set_qos_remap) { + DRM_DEBUG_ATOMIC("qos remap not supported\n"); + return; + } + + qos_tbl = params->is_rt ? &vbif->cap->qos_rt_tbl : + &vbif->cap->qos_nrt_tbl; + + if (!qos_tbl->npriority_lvl || !qos_tbl->priority_lvl) { + DRM_DEBUG_ATOMIC("qos tbl not defined\n"); + return; + } + + for (i = 0; i < qos_tbl->npriority_lvl; i++) { + DRM_DEBUG_ATOMIC("VBIF xin:%d lvl:%d/%d\n", + params->xin_id, i, + qos_tbl->priority_lvl[i]); + vbif->ops.set_qos_remap(vbif, params->xin_id, i, + qos_tbl->priority_lvl[i]); + } +} + +/** + * dpu_vbif_clear_errors - clear any vbif errors + * @dpu_kms: DPU handler + */ +void dpu_vbif_clear_errors(struct dpu_kms *dpu_kms) +{ + struct dpu_hw_vbif *vbif; + u32 pnd, src; + + vbif = dpu_kms->hw_vbif; + if (vbif && vbif->ops.clear_errors) { + vbif->ops.clear_errors(vbif, &pnd, &src); + if (pnd || src) { + DRM_DEBUG_KMS("VBIF: pnd 0x%X, src 0x%X\n", pnd, src); + } + } +} + +/** + * dpu_vbif_init_memtypes - initialize xin memory types for vbif + * @dpu_kms: DPU handler + */ +void dpu_vbif_init_memtypes(struct dpu_kms *dpu_kms) +{ + struct dpu_hw_vbif *vbif; + int j; + + vbif = dpu_kms->hw_vbif; + if (vbif && vbif->cap && vbif->ops.set_mem_type) { + for (j = 0; j < vbif->cap->memtype_count; j++) + vbif->ops.set_mem_type(vbif, j, vbif->cap->memtype[j]); + } +} + +#ifdef CONFIG_DEBUG_FS + +void dpu_debugfs_vbif_init(struct dpu_kms *dpu_kms, struct dentry *debugfs_root) +{ + const struct dpu_vbif_cfg *vbif = dpu_kms->catalog->vbif; + char vbif_name[32]; + struct dentry *debugfs_vbif; + int j; + + debugfs_vbif = debugfs_create_dir("vbif", debugfs_root); + + debugfs_create_u32("features", 0600, debugfs_vbif, + (u32 *)&vbif->features); + + debugfs_create_u32("xin_halt_timeout", 0400, debugfs_vbif, + (u32 *)&vbif->xin_halt_timeout); + + debugfs_create_u32("default_rd_ot_limit", 0400, debugfs_vbif, + (u32 *)&vbif->default_ot_rd_limit); + + debugfs_create_u32("default_wr_ot_limit", 0400, debugfs_vbif, + (u32 *)&vbif->default_ot_wr_limit); + + for (j = 0; j < vbif->dynamic_ot_rd_tbl.count; j++) { + const struct dpu_vbif_dynamic_ot_cfg *cfg = + &vbif->dynamic_ot_rd_tbl.cfg[j]; + + snprintf(vbif_name, sizeof(vbif_name), + "dynamic_ot_rd_%d_pps", j); + debugfs_create_u64(vbif_name, 0400, debugfs_vbif, + (u64 *)&cfg->pps); + snprintf(vbif_name, sizeof(vbif_name), + "dynamic_ot_rd_%d_ot_limit", j); + debugfs_create_u32(vbif_name, 0400, debugfs_vbif, + (u32 *)&cfg->ot_limit); + } + + for (j = 0; j < vbif->dynamic_ot_wr_tbl.count; j++) { + const struct dpu_vbif_dynamic_ot_cfg *cfg = + &vbif->dynamic_ot_wr_tbl.cfg[j]; + + snprintf(vbif_name, sizeof(vbif_name), + "dynamic_ot_wr_%d_pps", j); + debugfs_create_u64(vbif_name, 0400, debugfs_vbif, + (u64 *)&cfg->pps); + snprintf(vbif_name, sizeof(vbif_name), + "dynamic_ot_wr_%d_ot_limit", j); + debugfs_create_u32(vbif_name, 0400, debugfs_vbif, + (u32 *)&cfg->ot_limit); + } +} +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.h new file mode 100644 index 000000000..f47a89cb3 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.h @@ -0,0 +1,49 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2016-2018, The Linux Foundation. All rights reserved. + */ + +#ifndef __DPU_VBIF_H__ +#define __DPU_VBIF_H__ + +#include "dpu_kms.h" + +struct dpu_vbif_set_ot_params { + u32 xin_id; + u32 num; + u32 width; + u32 height; + u32 frame_rate; + bool rd; + bool is_wfd; +}; + +struct dpu_vbif_set_memtype_params { + u32 xin_id; + bool is_cacheable; +}; + +/** + * struct dpu_vbif_set_qos_params - QoS remapper parameter + * @xin_id: client interface identifier + * @num: pipe identifier (debug only) + * @is_rt: true if pipe is used in real-time use case + */ +struct dpu_vbif_set_qos_params { + u32 xin_id; + u32 num; + bool is_rt; +}; + +void dpu_vbif_set_ot_limit(struct dpu_kms *dpu_kms, + struct dpu_vbif_set_ot_params *params); + +void dpu_vbif_set_qos_remap(struct dpu_kms *dpu_kms, + struct dpu_vbif_set_qos_params *params); + +void dpu_vbif_clear_errors(struct dpu_kms *dpu_kms); + +void dpu_vbif_init_memtypes(struct dpu_kms *dpu_kms); + +void dpu_debugfs_vbif_init(struct dpu_kms *dpu_kms, struct dentry *debugfs_root); + +#endif /* __DPU_VBIF_H__ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.c new file mode 100644 index 000000000..e7b09013a --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.c @@ -0,0 +1,145 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#include +#include +#include + +#include "dpu_writeback.h" + +static int dpu_wb_conn_get_modes(struct drm_connector *connector) +{ + struct drm_device *dev = connector->dev; + struct msm_drm_private *priv = dev->dev_private; + struct dpu_kms *dpu_kms = to_dpu_kms(priv->kms); + + /* + * We should ideally be limiting the modes only to the maxlinewidth but + * on some chipsets this will allow even 4k modes to be added which will + * fail the per SSPP bandwidth checks. So, till we have dual-SSPP support + * and source split support added lets limit the modes based on max_mixer_width + * as 4K modes can then be supported. + */ + return drm_add_modes_noedid(connector, dpu_kms->catalog->caps->max_mixer_width, + dev->mode_config.max_height); +} + +static int dpu_wb_conn_atomic_check(struct drm_connector *connector, + struct drm_atomic_commit *state) +{ + struct drm_writeback_connector *wb_conn = drm_connector_to_writeback(connector); + struct dpu_wb_connector *dpu_wb_conn = to_dpu_wb_conn(wb_conn); + struct drm_connector_state *conn_state = + drm_atomic_get_new_connector_state(state, connector); + struct drm_crtc *crtc; + struct drm_crtc_state *crtc_state; + const struct drm_display_mode *mode; + struct drm_framebuffer *fb; + + DPU_DEBUG("[atomic_check:%d]\n", connector->base.id); + + if (!conn_state || !conn_state->connector) { + DPU_ERROR("invalid connector state\n"); + return -EINVAL; + } + + crtc = conn_state->crtc; + if (!crtc) + return 0; + + if (!conn_state->writeback_job || !conn_state->writeback_job->fb) + return 0; + + crtc_state = drm_atomic_get_crtc_state(state, crtc); + if (IS_ERR(crtc_state)) + return PTR_ERR(crtc_state); + + mode = &crtc_state->mode; + + fb = conn_state->writeback_job->fb; + + DPU_DEBUG("[fb_id:%u][fb:%u,%u][mode:\"%s\":%ux%u]\n", fb->base.id, fb->width, fb->height, + mode->name, mode->hdisplay, mode->vdisplay); + + if (fb->width != mode->hdisplay) { + DPU_ERROR("invalid fb w=%d, mode w=%d\n", fb->width, mode->hdisplay); + return -EINVAL; + } else if (fb->height != mode->vdisplay) { + DPU_ERROR("invalid fb h=%d, mode h=%d\n", fb->height, mode->vdisplay); + return -EINVAL; + } else if (fb->width > dpu_wb_conn->maxlinewidth) { + DPU_ERROR("invalid fb w=%d, maxlinewidth=%u\n", + fb->width, dpu_wb_conn->maxlinewidth); + return -EINVAL; + } else if (fb->modifier != DRM_FORMAT_MOD_LINEAR) { + DPU_ERROR("unsupported fb modifier:%#llx\n", fb->modifier); + return -EINVAL; + } + + return drm_atomic_helper_check_wb_connector_state(conn_state->connector, conn_state->state); +} + +static const struct drm_connector_funcs dpu_wb_conn_funcs = { + .reset = drm_atomic_helper_connector_reset, + .fill_modes = drm_helper_probe_single_connector_modes, + .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, +}; + +static int dpu_wb_conn_prepare_job(struct drm_writeback_connector *connector, + struct drm_writeback_job *job) +{ + + struct dpu_wb_connector *dpu_wb_conn = to_dpu_wb_conn(connector); + + if (!job->fb) + return 0; + + dpu_encoder_prepare_wb_job(dpu_wb_conn->wb_enc, job); + + return 0; +} + +static void dpu_wb_conn_cleanup_job(struct drm_writeback_connector *connector, + struct drm_writeback_job *job) +{ + struct dpu_wb_connector *dpu_wb_conn = to_dpu_wb_conn(connector); + + if (!job->fb) + return; + + dpu_encoder_cleanup_wb_job(dpu_wb_conn->wb_enc, job); +} + +static const struct drm_connector_helper_funcs dpu_wb_conn_helper_funcs = { + .get_modes = dpu_wb_conn_get_modes, + .atomic_check = dpu_wb_conn_atomic_check, + .prepare_writeback_job = dpu_wb_conn_prepare_job, + .cleanup_writeback_job = dpu_wb_conn_cleanup_job, +}; + +int dpu_writeback_init(struct drm_device *dev, struct drm_encoder *enc, + const u32 *format_list, u32 num_formats, u32 maxlinewidth) +{ + struct dpu_wb_connector *dpu_wb_conn; + int rc = 0; + + dpu_wb_conn = drmm_kzalloc(dev, sizeof(*dpu_wb_conn), GFP_KERNEL); + if (!dpu_wb_conn) + return -ENOMEM; + + dpu_wb_conn->maxlinewidth = maxlinewidth; + + drm_connector_helper_add(&dpu_wb_conn->base.base, &dpu_wb_conn_helper_funcs); + + rc = drmm_writeback_connector_init(dev, &dpu_wb_conn->base, + &dpu_wb_conn_funcs, enc, + format_list, num_formats); + + if (!rc) + dpu_wb_conn->wb_enc = enc; + + return rc; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.h new file mode 100644 index 000000000..4b11cca80 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.h @@ -0,0 +1,32 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#ifndef _DPU_WRITEBACK_H +#define _DPU_WRITEBACK_H + +#include +#include +#include +#include + +#include "msm_drv.h" +#include "dpu_kms.h" +#include "dpu_encoder_phys.h" + +struct dpu_wb_connector { + struct drm_writeback_connector base; + struct drm_encoder *wb_enc; + u32 maxlinewidth; +}; + +static inline struct dpu_wb_connector *to_dpu_wb_conn(struct drm_writeback_connector *conn) +{ + return container_of(conn, struct dpu_wb_connector, base); +} + +int dpu_writeback_init(struct drm_device *dev, struct drm_encoder *enc, + const u32 *format_list, u32 num_formats, u32 maxlinewidth); + +#endif /*_DPU_WRITEBACK_H */ diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c new file mode 100644 index 000000000..57dfce584 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c @@ -0,0 +1,665 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#include +#include +#include +#include +#include +#include + +#include "mdp4_kms.h" +#include "msm_gem.h" + +struct mdp4_crtc { + struct drm_crtc base; + char name[8]; + int ovlp; + enum mdp4_dma dma; + bool enabled; + + /* which mixer/encoder we route output to: */ + int mixer; + + struct { + spinlock_t lock; + bool stale; + uint32_t width, height; + uint32_t x, y; + + /* next cursor to scan-out: */ + uint32_t next_iova; + struct drm_gem_object *next_bo; + + /* current cursor being scanned out: */ + struct drm_gem_object *scanout_bo; + } cursor; + + + /* if there is a pending flip, these will be non-null: */ + struct drm_pending_vblank_event *event; + + /* Bits have been flushed at the last commit, + * used to decide if a vsync has happened since last commit. + */ + u32 flushed_mask; + +#define PENDING_CURSOR 0x1 +#define PENDING_FLIP 0x2 + atomic_t pending; + + /* for unref'ing cursor bo's after scanout completes: */ + struct drm_flip_work unref_cursor_work; + + struct mdp_irq vblank; + struct mdp_irq err; +}; +#define to_mdp4_crtc(x) container_of(x, struct mdp4_crtc, base) + +static struct mdp4_kms *get_kms(struct drm_crtc *crtc) +{ + struct msm_drm_private *priv = crtc->dev->dev_private; + return to_mdp4_kms(to_mdp_kms(priv->kms)); +} + +static void request_pending(struct drm_crtc *crtc, uint32_t pending) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + + atomic_or(pending, &mdp4_crtc->pending); + mdp_irq_register(&get_kms(crtc)->base, &mdp4_crtc->vblank); +} + +static void crtc_flush(struct drm_crtc *crtc) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct mdp4_kms *mdp4_kms = get_kms(crtc); + struct drm_plane *plane; + uint32_t flush = 0; + + drm_atomic_crtc_for_each_plane(plane, crtc) { + enum mdp4_pipe pipe_id = mdp4_plane_pipe(plane); + flush |= pipe2flush(pipe_id); + } + + flush |= ovlp2flush(mdp4_crtc->ovlp); + + DBG("%s: flush=%08x", mdp4_crtc->name, flush); + + mdp4_crtc->flushed_mask = flush; + + mdp4_write(mdp4_kms, REG_MDP4_OVERLAY_FLUSH, flush); +} + +/* if file!=NULL, this is preclose potential cancel-flip path */ +static void complete_flip(struct drm_crtc *crtc, struct drm_file *file) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct drm_device *dev = crtc->dev; + struct drm_pending_vblank_event *event; + unsigned long flags; + + spin_lock_irqsave(&dev->event_lock, flags); + event = mdp4_crtc->event; + if (event) { + mdp4_crtc->event = NULL; + DBG("%s: send event: %p", mdp4_crtc->name, event); + drm_crtc_send_vblank_event(crtc, event); + } + spin_unlock_irqrestore(&dev->event_lock, flags); +} + +static void unref_cursor_worker(struct drm_flip_work *work, void *val) +{ + struct mdp4_crtc *mdp4_crtc = + container_of(work, struct mdp4_crtc, unref_cursor_work); + struct mdp4_kms *mdp4_kms = get_kms(&mdp4_crtc->base); + struct msm_kms *kms = &mdp4_kms->base.base; + + msm_gem_unpin_iova(val, kms->vm); + drm_gem_object_put(val); +} + +/* statically (for now) map planes to mixer stage (z-order): */ +static const int idxs[] = { + [VG1] = 1, + [VG2] = 2, + [RGB1] = 0, + [RGB2] = 0, + [RGB3] = 0, + [VG3] = 3, + [VG4] = 4, + +}; + +/* setup mixer config, for which we need to consider all crtc's and + * the planes attached to them + * + * TODO may possibly need some extra locking here + */ +static void setup_mixer(struct mdp4_kms *mdp4_kms) +{ + struct drm_mode_config *config = &mdp4_kms->dev->mode_config; + struct drm_crtc *crtc; + uint32_t mixer_cfg = 0; + static const enum mdp_mixer_stage_id stages[] = { + STAGE_BASE, STAGE0, STAGE1, STAGE2, STAGE3, + }; + + list_for_each_entry(crtc, &config->crtc_list, head) { + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct drm_plane *plane; + + drm_atomic_crtc_for_each_plane(plane, crtc) { + enum mdp4_pipe pipe_id = mdp4_plane_pipe(plane); + int idx = idxs[pipe_id]; + mixer_cfg = mixercfg(mixer_cfg, mdp4_crtc->mixer, + pipe_id, stages[idx]); + } + } + + mdp4_write(mdp4_kms, REG_MDP4_LAYERMIXER_IN_CFG, mixer_cfg); +} + +static void blend_setup(struct drm_crtc *crtc) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct mdp4_kms *mdp4_kms = get_kms(crtc); + struct drm_plane *plane; + int i, ovlp = mdp4_crtc->ovlp; + bool alpha[4]= { false, false, false, false }; + + mdp4_write(mdp4_kms, REG_MDP4_OVLP_TRANSP_LOW0(ovlp), 0); + mdp4_write(mdp4_kms, REG_MDP4_OVLP_TRANSP_LOW1(ovlp), 0); + mdp4_write(mdp4_kms, REG_MDP4_OVLP_TRANSP_HIGH0(ovlp), 0); + mdp4_write(mdp4_kms, REG_MDP4_OVLP_TRANSP_HIGH1(ovlp), 0); + + drm_atomic_crtc_for_each_plane(plane, crtc) { + enum mdp4_pipe pipe_id = mdp4_plane_pipe(plane); + int idx = idxs[pipe_id]; + if (idx > 0) { + const struct msm_format *format = + msm_framebuffer_format(plane->state->fb); + alpha[idx-1] = format->alpha_enable; + } + } + + for (i = 0; i < 4; i++) { + uint32_t op; + + if (alpha[i]) { + op = MDP4_OVLP_STAGE_OP_FG_ALPHA(FG_PIXEL) | + MDP4_OVLP_STAGE_OP_BG_ALPHA(FG_PIXEL) | + MDP4_OVLP_STAGE_OP_BG_INV_ALPHA; + } else { + op = MDP4_OVLP_STAGE_OP_FG_ALPHA(FG_CONST) | + MDP4_OVLP_STAGE_OP_BG_ALPHA(BG_CONST); + } + + mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_FG_ALPHA(ovlp, i), 0xff); + mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_BG_ALPHA(ovlp, i), 0x00); + mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_OP(ovlp, i), op); + mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_CO3(ovlp, i), 1); + mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_TRANSP_LOW0(ovlp, i), 0); + mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_TRANSP_LOW1(ovlp, i), 0); + mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_TRANSP_HIGH0(ovlp, i), 0); + mdp4_write(mdp4_kms, REG_MDP4_OVLP_STAGE_TRANSP_HIGH1(ovlp, i), 0); + } + + setup_mixer(mdp4_kms); +} + +static void mdp4_crtc_mode_set_nofb(struct drm_crtc *crtc) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct mdp4_kms *mdp4_kms = get_kms(crtc); + enum mdp4_dma dma = mdp4_crtc->dma; + int ovlp = mdp4_crtc->ovlp; + struct drm_display_mode *mode; + + if (WARN_ON(!crtc->state)) + return; + + mode = &crtc->state->adjusted_mode; + + DBG("%s: set mode: " DRM_MODE_FMT, + mdp4_crtc->name, DRM_MODE_ARG(mode)); + + mdp4_write(mdp4_kms, REG_MDP4_DMA_SRC_SIZE(dma), + MDP4_DMA_SRC_SIZE_WIDTH(mode->hdisplay) | + MDP4_DMA_SRC_SIZE_HEIGHT(mode->vdisplay)); + + /* take data from pipe: */ + mdp4_write(mdp4_kms, REG_MDP4_DMA_SRC_BASE(dma), 0); + mdp4_write(mdp4_kms, REG_MDP4_DMA_SRC_STRIDE(dma), 0); + mdp4_write(mdp4_kms, REG_MDP4_DMA_DST_SIZE(dma), + MDP4_DMA_DST_SIZE_WIDTH(0) | + MDP4_DMA_DST_SIZE_HEIGHT(0)); + + mdp4_write(mdp4_kms, REG_MDP4_OVLP_BASE(ovlp), 0); + mdp4_write(mdp4_kms, REG_MDP4_OVLP_SIZE(ovlp), + MDP4_OVLP_SIZE_WIDTH(mode->hdisplay) | + MDP4_OVLP_SIZE_HEIGHT(mode->vdisplay)); + mdp4_write(mdp4_kms, REG_MDP4_OVLP_STRIDE(ovlp), 0); + + mdp4_write(mdp4_kms, REG_MDP4_OVLP_CFG(ovlp), 1); + + if (dma == DMA_E) { + mdp4_write(mdp4_kms, REG_MDP4_DMA_E_QUANT(0), 0x00ff0000); + mdp4_write(mdp4_kms, REG_MDP4_DMA_E_QUANT(1), 0x00ff0000); + mdp4_write(mdp4_kms, REG_MDP4_DMA_E_QUANT(2), 0x00ff0000); + } +} + +static void mdp4_crtc_atomic_disable(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct mdp4_kms *mdp4_kms = get_kms(crtc); + unsigned long flags; + + DBG("%s", mdp4_crtc->name); + + if (WARN_ON(!mdp4_crtc->enabled)) + return; + + /* Disable/save vblank irq handling before power is disabled */ + drm_crtc_vblank_off(crtc); + + mdp_irq_unregister(&mdp4_kms->base, &mdp4_crtc->err); + mdp4_disable(mdp4_kms); + + if (crtc->state->event && !crtc->state->active) { + WARN_ON(mdp4_crtc->event); + spin_lock_irqsave(&mdp4_kms->dev->event_lock, flags); + drm_crtc_send_vblank_event(crtc, crtc->state->event); + crtc->state->event = NULL; + spin_unlock_irqrestore(&mdp4_kms->dev->event_lock, flags); + } + + mdp4_crtc->enabled = false; +} + +static void mdp4_crtc_atomic_enable(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct mdp4_kms *mdp4_kms = get_kms(crtc); + + DBG("%s", mdp4_crtc->name); + + if (WARN_ON(mdp4_crtc->enabled)) + return; + + mdp4_enable(mdp4_kms); + + /* Restore vblank irq handling after power is enabled */ + drm_crtc_vblank_on(crtc); + + mdp_irq_register(&mdp4_kms->base, &mdp4_crtc->err); + + crtc_flush(crtc); + + mdp4_crtc->enabled = true; +} + +static int mdp4_crtc_atomic_check(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + DBG("%s: check", mdp4_crtc->name); + // TODO anything else to check? + return 0; +} + +static void mdp4_crtc_atomic_begin(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + DBG("%s: begin", mdp4_crtc->name); +} + +static void mdp4_crtc_atomic_flush(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct drm_device *dev = crtc->dev; + unsigned long flags; + + DBG("%s: event: %p", mdp4_crtc->name, crtc->state->event); + + WARN_ON(mdp4_crtc->event); + + spin_lock_irqsave(&dev->event_lock, flags); + mdp4_crtc->event = crtc->state->event; + crtc->state->event = NULL; + spin_unlock_irqrestore(&dev->event_lock, flags); + + blend_setup(crtc); + crtc_flush(crtc); + request_pending(crtc, PENDING_FLIP); +} + +#define CURSOR_WIDTH 64 +#define CURSOR_HEIGHT 64 + +/* called from IRQ to update cursor related registers (if needed). The + * cursor registers, other than x/y position, appear not to be double + * buffered, and changing them other than from vblank seems to trigger + * underflow. + */ +static void update_cursor(struct drm_crtc *crtc) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct mdp4_kms *mdp4_kms = get_kms(crtc); + struct msm_kms *kms = &mdp4_kms->base.base; + enum mdp4_dma dma = mdp4_crtc->dma; + unsigned long flags; + + spin_lock_irqsave(&mdp4_crtc->cursor.lock, flags); + if (mdp4_crtc->cursor.stale) { + struct drm_gem_object *next_bo = mdp4_crtc->cursor.next_bo; + struct drm_gem_object *prev_bo = mdp4_crtc->cursor.scanout_bo; + uint64_t iova = mdp4_crtc->cursor.next_iova; + + if (next_bo) { + /* take a obj ref + iova ref when we start scanning out: */ + drm_gem_object_get(next_bo); + msm_gem_get_and_pin_iova(next_bo, kms->vm, &iova); + + /* enable cursor: */ + mdp4_write(mdp4_kms, REG_MDP4_DMA_CURSOR_SIZE(dma), + MDP4_DMA_CURSOR_SIZE_WIDTH(mdp4_crtc->cursor.width) | + MDP4_DMA_CURSOR_SIZE_HEIGHT(mdp4_crtc->cursor.height)); + mdp4_write(mdp4_kms, REG_MDP4_DMA_CURSOR_BASE(dma), iova); + mdp4_write(mdp4_kms, REG_MDP4_DMA_CURSOR_BLEND_CONFIG(dma), + MDP4_DMA_CURSOR_BLEND_CONFIG_FORMAT(CURSOR_ARGB) | + MDP4_DMA_CURSOR_BLEND_CONFIG_CURSOR_EN); + } else { + /* disable cursor: */ + mdp4_write(mdp4_kms, REG_MDP4_DMA_CURSOR_BASE(dma), + mdp4_kms->blank_cursor_iova); + } + + /* and drop the iova ref + obj rev when done scanning out: */ + if (prev_bo) + drm_flip_work_queue(&mdp4_crtc->unref_cursor_work, prev_bo); + + mdp4_crtc->cursor.scanout_bo = next_bo; + mdp4_crtc->cursor.stale = false; + } + + mdp4_write(mdp4_kms, REG_MDP4_DMA_CURSOR_POS(dma), + MDP4_DMA_CURSOR_POS_X(mdp4_crtc->cursor.x) | + MDP4_DMA_CURSOR_POS_Y(mdp4_crtc->cursor.y)); + + spin_unlock_irqrestore(&mdp4_crtc->cursor.lock, flags); +} + +static int mdp4_crtc_cursor_set(struct drm_crtc *crtc, + struct drm_file *file_priv, uint32_t handle, + uint32_t width, uint32_t height) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct mdp4_kms *mdp4_kms = get_kms(crtc); + struct msm_kms *kms = &mdp4_kms->base.base; + struct drm_device *dev = crtc->dev; + struct drm_gem_object *cursor_bo, *old_bo; + unsigned long flags; + uint64_t iova; + int ret; + + if ((width > CURSOR_WIDTH) || (height > CURSOR_HEIGHT)) { + DRM_DEV_ERROR(dev->dev, "bad cursor size: %dx%d\n", width, height); + return -EINVAL; + } + + if (handle) { + cursor_bo = drm_gem_object_lookup(file_priv, handle); + if (!cursor_bo) + return -ENOENT; + } else { + cursor_bo = NULL; + } + + if (cursor_bo) { + ret = msm_gem_get_and_pin_iova(cursor_bo, kms->vm, &iova); + if (ret) + goto fail; + } else { + iova = 0; + } + + spin_lock_irqsave(&mdp4_crtc->cursor.lock, flags); + old_bo = mdp4_crtc->cursor.next_bo; + mdp4_crtc->cursor.next_bo = cursor_bo; + mdp4_crtc->cursor.next_iova = iova; + mdp4_crtc->cursor.width = width; + mdp4_crtc->cursor.height = height; + mdp4_crtc->cursor.stale = true; + spin_unlock_irqrestore(&mdp4_crtc->cursor.lock, flags); + + if (old_bo) { + /* drop our previous reference: */ + drm_flip_work_queue(&mdp4_crtc->unref_cursor_work, old_bo); + } + + request_pending(crtc, PENDING_CURSOR); + + return 0; + +fail: + drm_gem_object_put(cursor_bo); + return ret; +} + +static int mdp4_crtc_cursor_move(struct drm_crtc *crtc, int x, int y) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + unsigned long flags; + + spin_lock_irqsave(&mdp4_crtc->cursor.lock, flags); + mdp4_crtc->cursor.x = x; + mdp4_crtc->cursor.y = y; + spin_unlock_irqrestore(&mdp4_crtc->cursor.lock, flags); + + crtc_flush(crtc); + request_pending(crtc, PENDING_CURSOR); + + return 0; +} + +static const struct drm_crtc_funcs mdp4_crtc_funcs = { + .set_config = drm_atomic_helper_set_config, + .page_flip = drm_atomic_helper_page_flip, + .cursor_set = mdp4_crtc_cursor_set, + .cursor_move = mdp4_crtc_cursor_move, + .reset = drm_atomic_helper_crtc_reset, + .atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_crtc_destroy_state, + .enable_vblank = msm_crtc_enable_vblank, + .disable_vblank = msm_crtc_disable_vblank, +}; + +static const struct drm_crtc_helper_funcs mdp4_crtc_helper_funcs = { + .mode_set_nofb = mdp4_crtc_mode_set_nofb, + .atomic_check = mdp4_crtc_atomic_check, + .atomic_begin = mdp4_crtc_atomic_begin, + .atomic_flush = mdp4_crtc_atomic_flush, + .atomic_enable = mdp4_crtc_atomic_enable, + .atomic_disable = mdp4_crtc_atomic_disable, +}; + +static void mdp4_crtc_vblank_irq(struct mdp_irq *irq, uint32_t irqstatus) +{ + struct mdp4_crtc *mdp4_crtc = container_of(irq, struct mdp4_crtc, vblank); + struct drm_crtc *crtc = &mdp4_crtc->base; + struct msm_drm_private *priv = crtc->dev->dev_private; + unsigned pending; + + mdp_irq_unregister(&get_kms(crtc)->base, &mdp4_crtc->vblank); + + pending = atomic_xchg(&mdp4_crtc->pending, 0); + + if (pending & PENDING_FLIP) { + complete_flip(crtc, NULL); + } + + if (pending & PENDING_CURSOR) { + update_cursor(crtc); + drm_flip_work_commit(&mdp4_crtc->unref_cursor_work, priv->kms->wq); + } +} + +static void mdp4_crtc_err_irq(struct mdp_irq *irq, uint32_t irqstatus) +{ + struct mdp4_crtc *mdp4_crtc = container_of(irq, struct mdp4_crtc, err); + struct drm_crtc *crtc = &mdp4_crtc->base; + DBG("%s: error: %08x", mdp4_crtc->name, irqstatus); + crtc_flush(crtc); +} + +static void mdp4_crtc_wait_for_flush_done(struct drm_crtc *crtc) +{ + struct drm_device *dev = crtc->dev; + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct mdp4_kms *mdp4_kms = get_kms(crtc); + wait_queue_head_t *queue = drm_crtc_vblank_waitqueue(crtc); + int ret; + + ret = drm_crtc_vblank_get(crtc); + if (ret) + return; + + ret = wait_event_timeout(*queue, + !(mdp4_read(mdp4_kms, REG_MDP4_OVERLAY_FLUSH) & + mdp4_crtc->flushed_mask), + msecs_to_jiffies(50)); + if (ret <= 0) + dev_warn(dev->dev, "vblank time out, crtc=%s\n", mdp4_crtc->base.name); + + mdp4_crtc->flushed_mask = 0; + + drm_crtc_vblank_put(crtc); +} + +uint32_t mdp4_crtc_vblank(struct drm_crtc *crtc) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + return mdp4_crtc->vblank.irqmask; +} + +/* set dma config, ie. the format the encoder wants. */ +void mdp4_crtc_set_config(struct drm_crtc *crtc, uint32_t config) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct mdp4_kms *mdp4_kms = get_kms(crtc); + + mdp4_write(mdp4_kms, REG_MDP4_DMA_CONFIG(mdp4_crtc->dma), config); +} + +/* set interface for routing crtc->encoder: */ +void mdp4_crtc_set_intf(struct drm_crtc *crtc, enum mdp4_intf intf, int mixer) +{ + struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); + struct mdp4_kms *mdp4_kms = get_kms(crtc); + uint32_t intf_sel; + + intf_sel = mdp4_read(mdp4_kms, REG_MDP4_DISP_INTF_SEL); + + switch (mdp4_crtc->dma) { + case DMA_P: + intf_sel &= ~MDP4_DISP_INTF_SEL_PRIM__MASK; + intf_sel |= MDP4_DISP_INTF_SEL_PRIM(intf); + break; + case DMA_S: + intf_sel &= ~MDP4_DISP_INTF_SEL_SEC__MASK; + intf_sel |= MDP4_DISP_INTF_SEL_SEC(intf); + break; + case DMA_E: + intf_sel &= ~MDP4_DISP_INTF_SEL_EXT__MASK; + intf_sel |= MDP4_DISP_INTF_SEL_EXT(intf); + break; + } + + if (intf == INTF_DSI_VIDEO) { + intf_sel &= ~MDP4_DISP_INTF_SEL_DSI_CMD; + intf_sel |= MDP4_DISP_INTF_SEL_DSI_VIDEO; + } else if (intf == INTF_DSI_CMD) { + intf_sel &= ~MDP4_DISP_INTF_SEL_DSI_VIDEO; + intf_sel |= MDP4_DISP_INTF_SEL_DSI_CMD; + } + + mdp4_crtc->mixer = mixer; + + blend_setup(crtc); + + DBG("%s: intf_sel=%08x", mdp4_crtc->name, intf_sel); + + mdp4_write(mdp4_kms, REG_MDP4_DISP_INTF_SEL, intf_sel); +} + +void mdp4_crtc_wait_for_commit_done(struct drm_crtc *crtc) +{ + /* wait_for_flush_done is the only case for now. + * Later we will have command mode CRTC to wait for + * other event. + */ + mdp4_crtc_wait_for_flush_done(crtc); +} + +static const char *dma_names[] = { + "DMA_P", "DMA_S", "DMA_E", +}; + +static void mdp4_crtc_flip_cleanup(struct drm_device *dev, void *ptr) +{ + struct mdp4_crtc *mdp4_crtc = ptr; + + drm_flip_work_cleanup(&mdp4_crtc->unref_cursor_work); +} + +/* initialize crtc */ +struct drm_crtc *mdp4_crtc_init(struct drm_device *dev, + struct drm_plane *plane, int ovlp_id, + enum mdp4_dma dma_id) +{ + struct drm_crtc *crtc = NULL; + struct mdp4_crtc *mdp4_crtc; + int ret; + + mdp4_crtc = drmm_crtc_alloc_with_planes(dev, struct mdp4_crtc, base, + plane, NULL, + &mdp4_crtc_funcs, NULL); + if (IS_ERR(mdp4_crtc)) + return ERR_CAST(mdp4_crtc); + + crtc = &mdp4_crtc->base; + + mdp4_crtc->ovlp = ovlp_id; + mdp4_crtc->dma = dma_id; + + mdp4_crtc->vblank.irqmask = dma2irq(mdp4_crtc->dma); + mdp4_crtc->vblank.irq = mdp4_crtc_vblank_irq; + + mdp4_crtc->err.irqmask = dma2err(mdp4_crtc->dma); + mdp4_crtc->err.irq = mdp4_crtc_err_irq; + + snprintf(mdp4_crtc->name, sizeof(mdp4_crtc->name), "%s:%d", + dma_names[dma_id], ovlp_id); + + spin_lock_init(&mdp4_crtc->cursor.lock); + + drm_flip_work_init(&mdp4_crtc->unref_cursor_work, + "unref cursor", unref_cursor_worker); + ret = drmm_add_action_or_reset(dev, mdp4_crtc_flip_cleanup, mdp4_crtc); + if (ret) + return ERR_PTR(ret); + + drm_crtc_helper_add(crtc, &mdp4_crtc_helper_funcs); + + return crtc; +} diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_dsi_encoder.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_dsi_encoder.c new file mode 100644 index 000000000..74dafe710 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_dsi_encoder.c @@ -0,0 +1,153 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2015, The Linux Foundation. All rights reserved. + * Copyright (c) 2014, Inforce Computing. All rights reserved. + * + * Author: Vinay Simha + */ + +#include +#include + +#include "mdp4_kms.h" + +#ifdef CONFIG_DRM_MSM_DSI + +struct mdp4_dsi_encoder { + struct drm_encoder base; + struct drm_panel *panel; + bool enabled; +}; +#define to_mdp4_dsi_encoder(x) container_of(x, struct mdp4_dsi_encoder, base) + +static struct mdp4_kms *get_kms(struct drm_encoder *encoder) +{ + struct msm_drm_private *priv = encoder->dev->dev_private; + return to_mdp4_kms(to_mdp_kms(priv->kms)); +} + +static void mdp4_dsi_encoder_mode_set(struct drm_encoder *encoder, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + struct mdp4_kms *mdp4_kms = get_kms(encoder); + uint32_t dsi_hsync_skew, vsync_period, vsync_len, ctrl_pol; + uint32_t display_v_start, display_v_end; + uint32_t hsync_start_x, hsync_end_x; + + mode = adjusted_mode; + + DBG("set mode: " DRM_MODE_FMT, DRM_MODE_ARG(mode)); + + ctrl_pol = 0; + if (mode->flags & DRM_MODE_FLAG_NHSYNC) + ctrl_pol |= MDP4_DSI_CTRL_POLARITY_HSYNC_LOW; + if (mode->flags & DRM_MODE_FLAG_NVSYNC) + ctrl_pol |= MDP4_DSI_CTRL_POLARITY_VSYNC_LOW; + /* probably need to get DATA_EN polarity from panel.. */ + + dsi_hsync_skew = 0; /* get this from panel? */ + + hsync_start_x = (mode->htotal - mode->hsync_start); + hsync_end_x = mode->htotal - (mode->hsync_start - mode->hdisplay) - 1; + + vsync_period = mode->vtotal * mode->htotal; + vsync_len = (mode->vsync_end - mode->vsync_start) * mode->htotal; + display_v_start = (mode->vtotal - mode->vsync_start) * mode->htotal + dsi_hsync_skew; + display_v_end = vsync_period - ((mode->vsync_start - mode->vdisplay) * mode->htotal) + dsi_hsync_skew - 1; + + mdp4_write(mdp4_kms, REG_MDP4_DSI_HSYNC_CTRL, + MDP4_DSI_HSYNC_CTRL_PULSEW(mode->hsync_end - mode->hsync_start) | + MDP4_DSI_HSYNC_CTRL_PERIOD(mode->htotal)); + mdp4_write(mdp4_kms, REG_MDP4_DSI_VSYNC_PERIOD, vsync_period); + mdp4_write(mdp4_kms, REG_MDP4_DSI_VSYNC_LEN, vsync_len); + mdp4_write(mdp4_kms, REG_MDP4_DSI_DISPLAY_HCTRL, + MDP4_DSI_DISPLAY_HCTRL_START(hsync_start_x) | + MDP4_DSI_DISPLAY_HCTRL_END(hsync_end_x)); + mdp4_write(mdp4_kms, REG_MDP4_DSI_DISPLAY_VSTART, display_v_start); + mdp4_write(mdp4_kms, REG_MDP4_DSI_DISPLAY_VEND, display_v_end); + + mdp4_write(mdp4_kms, REG_MDP4_DSI_CTRL_POLARITY, ctrl_pol); + mdp4_write(mdp4_kms, REG_MDP4_DSI_UNDERFLOW_CLR, + MDP4_DSI_UNDERFLOW_CLR_ENABLE_RECOVERY | + MDP4_DSI_UNDERFLOW_CLR_COLOR(0xff)); + mdp4_write(mdp4_kms, REG_MDP4_DSI_ACTIVE_HCTL, + MDP4_DSI_ACTIVE_HCTL_START(0) | + MDP4_DSI_ACTIVE_HCTL_END(0)); + mdp4_write(mdp4_kms, REG_MDP4_DSI_HSYNC_SKEW, dsi_hsync_skew); + mdp4_write(mdp4_kms, REG_MDP4_DSI_BORDER_CLR, 0); + mdp4_write(mdp4_kms, REG_MDP4_DSI_ACTIVE_VSTART, 0); + mdp4_write(mdp4_kms, REG_MDP4_DSI_ACTIVE_VEND, 0); +} + +static void mdp4_dsi_encoder_disable(struct drm_encoder *encoder) +{ + struct mdp4_dsi_encoder *mdp4_dsi_encoder = to_mdp4_dsi_encoder(encoder); + struct mdp4_kms *mdp4_kms = get_kms(encoder); + + if (!mdp4_dsi_encoder->enabled) + return; + + mdp4_write(mdp4_kms, REG_MDP4_DSI_ENABLE, 0); + + /* + * Wait for a vsync so we know the ENABLE=0 latched before + * the (connector) source of the vsync's gets disabled, + * otherwise we end up in a funny state if we re-enable + * before the disable latches, which results that some of + * the settings changes for the new modeset (like new + * scanout buffer) don't latch properly.. + */ + mdp_irq_wait(&mdp4_kms->base, MDP4_IRQ_PRIMARY_VSYNC); + + mdp4_dsi_encoder->enabled = false; +} + +static void mdp4_dsi_encoder_enable(struct drm_encoder *encoder) +{ + struct mdp4_dsi_encoder *mdp4_dsi_encoder = to_mdp4_dsi_encoder(encoder); + struct mdp4_kms *mdp4_kms = get_kms(encoder); + + if (mdp4_dsi_encoder->enabled) + return; + + mdp4_crtc_set_config(encoder->crtc, + MDP4_DMA_CONFIG_PACK_ALIGN_MSB | + MDP4_DMA_CONFIG_DEFLKR_EN | + MDP4_DMA_CONFIG_DITHER_EN | + MDP4_DMA_CONFIG_R_BPC(BPC8) | + MDP4_DMA_CONFIG_G_BPC(BPC8) | + MDP4_DMA_CONFIG_B_BPC(BPC8) | + MDP4_DMA_CONFIG_PACK(0x21)); + + mdp4_crtc_set_intf(encoder->crtc, INTF_DSI_VIDEO, 0); + + mdp4_write(mdp4_kms, REG_MDP4_DSI_ENABLE, 1); + + mdp4_dsi_encoder->enabled = true; +} + +static const struct drm_encoder_helper_funcs mdp4_dsi_encoder_helper_funcs = { + .mode_set = mdp4_dsi_encoder_mode_set, + .disable = mdp4_dsi_encoder_disable, + .enable = mdp4_dsi_encoder_enable, +}; + +/* initialize encoder */ +struct drm_encoder *mdp4_dsi_encoder_init(struct drm_device *dev) +{ + struct drm_encoder *encoder; + struct mdp4_dsi_encoder *mdp4_dsi_encoder; + + mdp4_dsi_encoder = drmm_encoder_alloc(dev, struct mdp4_dsi_encoder, base, + NULL, DRM_MODE_ENCODER_DSI, NULL); + if (IS_ERR(mdp4_dsi_encoder)) + return ERR_CAST(mdp4_dsi_encoder); + + encoder = &mdp4_dsi_encoder->base; + + drm_encoder_helper_add(encoder, &mdp4_dsi_encoder_helper_funcs); + + return encoder; +} +#endif /* CONFIG_DRM_MSM_DSI */ diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c new file mode 100644 index 000000000..3b70764b4 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c @@ -0,0 +1,190 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#include +#include + +#include "mdp4_kms.h" + +struct mdp4_dtv_encoder { + struct drm_encoder base; + struct clk *hdmi_clk; + struct clk *mdp_clk; + unsigned long int pixclock; + bool enabled; + uint32_t bsc; +}; +#define to_mdp4_dtv_encoder(x) container_of(x, struct mdp4_dtv_encoder, base) + +static struct mdp4_kms *get_kms(struct drm_encoder *encoder) +{ + struct msm_drm_private *priv = encoder->dev->dev_private; + return to_mdp4_kms(to_mdp_kms(priv->kms)); +} + +static void mdp4_dtv_encoder_mode_set(struct drm_encoder *encoder, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + struct mdp4_dtv_encoder *mdp4_dtv_encoder = to_mdp4_dtv_encoder(encoder); + struct mdp4_kms *mdp4_kms = get_kms(encoder); + uint32_t dtv_hsync_skew, vsync_period, vsync_len, ctrl_pol; + uint32_t display_v_start, display_v_end; + uint32_t hsync_start_x, hsync_end_x; + + mode = adjusted_mode; + + DBG("set mode: " DRM_MODE_FMT, DRM_MODE_ARG(mode)); + + mdp4_dtv_encoder->pixclock = mode->clock * 1000; + + DBG("pixclock=%lu", mdp4_dtv_encoder->pixclock); + + ctrl_pol = 0; + if (mode->flags & DRM_MODE_FLAG_NHSYNC) + ctrl_pol |= MDP4_DTV_CTRL_POLARITY_HSYNC_LOW; + if (mode->flags & DRM_MODE_FLAG_NVSYNC) + ctrl_pol |= MDP4_DTV_CTRL_POLARITY_VSYNC_LOW; + /* probably need to get DATA_EN polarity from panel.. */ + + dtv_hsync_skew = 0; /* get this from panel? */ + + hsync_start_x = (mode->htotal - mode->hsync_start); + hsync_end_x = mode->htotal - (mode->hsync_start - mode->hdisplay) - 1; + + vsync_period = mode->vtotal * mode->htotal; + vsync_len = (mode->vsync_end - mode->vsync_start) * mode->htotal; + display_v_start = (mode->vtotal - mode->vsync_start) * mode->htotal + dtv_hsync_skew; + display_v_end = vsync_period - ((mode->vsync_start - mode->vdisplay) * mode->htotal) + dtv_hsync_skew - 1; + + mdp4_write(mdp4_kms, REG_MDP4_DTV_HSYNC_CTRL, + MDP4_DTV_HSYNC_CTRL_PULSEW(mode->hsync_end - mode->hsync_start) | + MDP4_DTV_HSYNC_CTRL_PERIOD(mode->htotal)); + mdp4_write(mdp4_kms, REG_MDP4_DTV_VSYNC_PERIOD, vsync_period); + mdp4_write(mdp4_kms, REG_MDP4_DTV_VSYNC_LEN, vsync_len); + mdp4_write(mdp4_kms, REG_MDP4_DTV_DISPLAY_HCTRL, + MDP4_DTV_DISPLAY_HCTRL_START(hsync_start_x) | + MDP4_DTV_DISPLAY_HCTRL_END(hsync_end_x)); + mdp4_write(mdp4_kms, REG_MDP4_DTV_DISPLAY_VSTART, display_v_start); + mdp4_write(mdp4_kms, REG_MDP4_DTV_DISPLAY_VEND, display_v_end); + mdp4_write(mdp4_kms, REG_MDP4_DTV_BORDER_CLR, 0); + mdp4_write(mdp4_kms, REG_MDP4_DTV_UNDERFLOW_CLR, + MDP4_DTV_UNDERFLOW_CLR_ENABLE_RECOVERY | + MDP4_DTV_UNDERFLOW_CLR_COLOR(0xff)); + mdp4_write(mdp4_kms, REG_MDP4_DTV_HSYNC_SKEW, dtv_hsync_skew); + mdp4_write(mdp4_kms, REG_MDP4_DTV_CTRL_POLARITY, ctrl_pol); + mdp4_write(mdp4_kms, REG_MDP4_DTV_ACTIVE_HCTL, + MDP4_DTV_ACTIVE_HCTL_START(0) | + MDP4_DTV_ACTIVE_HCTL_END(0)); + mdp4_write(mdp4_kms, REG_MDP4_DTV_ACTIVE_VSTART, 0); + mdp4_write(mdp4_kms, REG_MDP4_DTV_ACTIVE_VEND, 0); +} + +static void mdp4_dtv_encoder_disable(struct drm_encoder *encoder) +{ + struct mdp4_dtv_encoder *mdp4_dtv_encoder = to_mdp4_dtv_encoder(encoder); + struct mdp4_kms *mdp4_kms = get_kms(encoder); + + if (WARN_ON(!mdp4_dtv_encoder->enabled)) + return; + + mdp4_write(mdp4_kms, REG_MDP4_DTV_ENABLE, 0); + + /* + * Wait for a vsync so we know the ENABLE=0 latched before + * the (connector) source of the vsync's gets disabled, + * otherwise we end up in a funny state if we re-enable + * before the disable latches, which results that some of + * the settings changes for the new modeset (like new + * scanout buffer) don't latch properly.. + */ + mdp_irq_wait(&mdp4_kms->base, MDP4_IRQ_EXTERNAL_VSYNC); + + clk_disable_unprepare(mdp4_dtv_encoder->hdmi_clk); + clk_disable_unprepare(mdp4_dtv_encoder->mdp_clk); + + mdp4_dtv_encoder->enabled = false; +} + +static void mdp4_dtv_encoder_enable(struct drm_encoder *encoder) +{ + struct drm_device *dev = encoder->dev; + struct mdp4_dtv_encoder *mdp4_dtv_encoder = to_mdp4_dtv_encoder(encoder); + struct mdp4_kms *mdp4_kms = get_kms(encoder); + unsigned long pc = mdp4_dtv_encoder->pixclock; + int ret; + + if (WARN_ON(mdp4_dtv_encoder->enabled)) + return; + + mdp4_crtc_set_config(encoder->crtc, + MDP4_DMA_CONFIG_R_BPC(BPC8) | + MDP4_DMA_CONFIG_G_BPC(BPC8) | + MDP4_DMA_CONFIG_B_BPC(BPC8) | + MDP4_DMA_CONFIG_PACK(0x21)); + mdp4_crtc_set_intf(encoder->crtc, INTF_LCDC_DTV, 1); + + DBG("setting mdp_clk=%lu", pc); + + ret = clk_set_rate(mdp4_dtv_encoder->mdp_clk, pc); + if (ret) + DRM_DEV_ERROR(dev->dev, "failed to set mdp_clk to %lu: %d\n", + pc, ret); + + ret = clk_prepare_enable(mdp4_dtv_encoder->mdp_clk); + if (ret) + DRM_DEV_ERROR(dev->dev, "failed to enabled mdp_clk: %d\n", ret); + + ret = clk_prepare_enable(mdp4_dtv_encoder->hdmi_clk); + if (ret) + DRM_DEV_ERROR(dev->dev, "failed to enable hdmi_clk: %d\n", ret); + + mdp4_write(mdp4_kms, REG_MDP4_DTV_ENABLE, 1); + + mdp4_dtv_encoder->enabled = true; +} + +static const struct drm_encoder_helper_funcs mdp4_dtv_encoder_helper_funcs = { + .mode_set = mdp4_dtv_encoder_mode_set, + .enable = mdp4_dtv_encoder_enable, + .disable = mdp4_dtv_encoder_disable, +}; + +long mdp4_dtv_round_pixclk(struct drm_encoder *encoder, unsigned long rate) +{ + struct mdp4_dtv_encoder *mdp4_dtv_encoder = to_mdp4_dtv_encoder(encoder); + return clk_round_rate(mdp4_dtv_encoder->mdp_clk, rate); +} + +/* initialize encoder */ +struct drm_encoder *mdp4_dtv_encoder_init(struct drm_device *dev) +{ + struct drm_encoder *encoder; + struct mdp4_dtv_encoder *mdp4_dtv_encoder; + + mdp4_dtv_encoder = drmm_encoder_alloc(dev, struct mdp4_dtv_encoder, base, + NULL, DRM_MODE_ENCODER_TMDS, NULL); + if (IS_ERR(mdp4_dtv_encoder)) + return ERR_CAST(mdp4_dtv_encoder); + + encoder = &mdp4_dtv_encoder->base; + + drm_encoder_helper_add(encoder, &mdp4_dtv_encoder_helper_funcs); + + mdp4_dtv_encoder->hdmi_clk = devm_clk_get(dev->dev, "hdmi_clk"); + if (IS_ERR(mdp4_dtv_encoder->hdmi_clk)) { + DRM_DEV_ERROR(dev->dev, "failed to get hdmi_clk\n"); + return ERR_CAST(mdp4_dtv_encoder->hdmi_clk); + } + + mdp4_dtv_encoder->mdp_clk = devm_clk_get(dev->dev, "tv_clk"); + if (IS_ERR(mdp4_dtv_encoder->mdp_clk)) { + DRM_DEV_ERROR(dev->dev, "failed to get tv_clk\n"); + return ERR_CAST(mdp4_dtv_encoder->mdp_clk); + } + + return encoder; +} diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_irq.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_irq.c new file mode 100644 index 000000000..ddcdd5e87 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_irq.c @@ -0,0 +1,110 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#include +#include + +#include "msm_drv.h" +#include "mdp4_kms.h" + +void mdp4_set_irqmask(struct mdp_kms *mdp_kms, uint32_t irqmask, + uint32_t old_irqmask) +{ + mdp4_write(to_mdp4_kms(mdp_kms), REG_MDP4_INTR_CLEAR, + irqmask ^ (irqmask & old_irqmask)); + mdp4_write(to_mdp4_kms(mdp_kms), REG_MDP4_INTR_ENABLE, irqmask); +} + +static void mdp4_irq_error_handler(struct mdp_irq *irq, uint32_t irqstatus) +{ + struct mdp4_kms *mdp4_kms = container_of(irq, struct mdp4_kms, error_handler); + static DEFINE_RATELIMIT_STATE(rs, 5*HZ, 1); + extern bool dumpstate; + + DRM_ERROR_RATELIMITED("errors: %08x\n", irqstatus); + + if (dumpstate && __ratelimit(&rs)) { + struct drm_printer p = drm_info_printer(mdp4_kms->dev->dev); + drm_state_dump(mdp4_kms->dev, &p); + } +} + +void mdp4_irq_preinstall(struct msm_kms *kms) +{ + struct mdp4_kms *mdp4_kms = to_mdp4_kms(to_mdp_kms(kms)); + mdp4_enable(mdp4_kms); + mdp4_write(mdp4_kms, REG_MDP4_INTR_CLEAR, 0xffffffff); + mdp4_write(mdp4_kms, REG_MDP4_INTR_ENABLE, 0x00000000); + mdp4_disable(mdp4_kms); +} + +int mdp4_irq_postinstall(struct msm_kms *kms) +{ + struct mdp_kms *mdp_kms = to_mdp_kms(kms); + struct mdp4_kms *mdp4_kms = to_mdp4_kms(mdp_kms); + struct mdp_irq *error_handler = &mdp4_kms->error_handler; + + error_handler->irq = mdp4_irq_error_handler; + error_handler->irqmask = MDP4_IRQ_PRIMARY_INTF_UDERRUN | + MDP4_IRQ_EXTERNAL_INTF_UDERRUN; + + mdp_irq_register(mdp_kms, error_handler); + + return 0; +} + +void mdp4_irq_uninstall(struct msm_kms *kms) +{ + struct mdp4_kms *mdp4_kms = to_mdp4_kms(to_mdp_kms(kms)); + mdp4_enable(mdp4_kms); + mdp4_write(mdp4_kms, REG_MDP4_INTR_ENABLE, 0x00000000); + mdp4_disable(mdp4_kms); +} + +irqreturn_t mdp4_irq(struct msm_kms *kms) +{ + struct mdp_kms *mdp_kms = to_mdp_kms(kms); + struct mdp4_kms *mdp4_kms = to_mdp4_kms(mdp_kms); + struct drm_device *dev = mdp4_kms->dev; + struct drm_crtc *crtc; + uint32_t status, enable; + + enable = mdp4_read(mdp4_kms, REG_MDP4_INTR_ENABLE); + status = mdp4_read(mdp4_kms, REG_MDP4_INTR_STATUS) & enable; + mdp4_write(mdp4_kms, REG_MDP4_INTR_CLEAR, status); + + VERB("status=%08x", status); + + mdp_dispatch_irqs(mdp_kms, status); + + drm_for_each_crtc(crtc, dev) + if (status & mdp4_crtc_vblank(crtc)) + drm_crtc_handle_vblank(crtc); + + return IRQ_HANDLED; +} + +int mdp4_enable_vblank(struct msm_kms *kms, struct drm_crtc *crtc) +{ + struct mdp4_kms *mdp4_kms = to_mdp4_kms(to_mdp_kms(kms)); + + mdp4_enable(mdp4_kms); + mdp_update_vblank_mask(to_mdp_kms(kms), + mdp4_crtc_vblank(crtc), true); + mdp4_disable(mdp4_kms); + + return 0; +} + +void mdp4_disable_vblank(struct msm_kms *kms, struct drm_crtc *crtc) +{ + struct mdp4_kms *mdp4_kms = to_mdp4_kms(to_mdp_kms(kms)); + + mdp4_enable(mdp4_kms); + mdp_update_vblank_mask(to_mdp_kms(kms), + mdp4_crtc_vblank(crtc), false); + mdp4_disable(mdp4_kms); +} diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c new file mode 100644 index 000000000..9b1d1982e --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c @@ -0,0 +1,575 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#include + +#include +#include +#include + +#include "msm_drv.h" +#include "msm_gem.h" +#include "msm_mmu.h" +#include "mdp4_kms.h" + +static int mdp4_hw_init(struct msm_kms *kms) +{ + struct mdp4_kms *mdp4_kms = to_mdp4_kms(to_mdp_kms(kms)); + struct drm_device *dev = mdp4_kms->dev; + u32 dmap_cfg, vg_cfg; + unsigned long clk; + + pm_runtime_get_sync(dev->dev); + + if (mdp4_kms->rev > 1) { + mdp4_write(mdp4_kms, REG_MDP4_CS_CONTROLLER0, 0x0707ffff); + mdp4_write(mdp4_kms, REG_MDP4_CS_CONTROLLER1, 0x03073f3f); + } + + mdp4_write(mdp4_kms, REG_MDP4_PORTMAP_MODE, 0x3); + + /* max read pending cmd config, 3 pending requests: */ + mdp4_write(mdp4_kms, REG_MDP4_READ_CNFG, 0x02222); + + clk = clk_get_rate(mdp4_kms->clk); + + if ((mdp4_kms->rev >= 1) || (clk >= 90000000)) { + dmap_cfg = 0x47; /* 16 bytes-burst x 8 req */ + vg_cfg = 0x47; /* 16 bytes-burs x 8 req */ + } else { + dmap_cfg = 0x27; /* 8 bytes-burst x 8 req */ + vg_cfg = 0x43; /* 16 bytes-burst x 4 req */ + } + + DBG("fetch config: dmap=%02x, vg=%02x", dmap_cfg, vg_cfg); + + mdp4_write(mdp4_kms, REG_MDP4_DMA_FETCH_CONFIG(DMA_P), dmap_cfg); + mdp4_write(mdp4_kms, REG_MDP4_DMA_FETCH_CONFIG(DMA_E), dmap_cfg); + + mdp4_write(mdp4_kms, REG_MDP4_PIPE_FETCH_CONFIG(VG1), vg_cfg); + mdp4_write(mdp4_kms, REG_MDP4_PIPE_FETCH_CONFIG(VG2), vg_cfg); + mdp4_write(mdp4_kms, REG_MDP4_PIPE_FETCH_CONFIG(RGB1), vg_cfg); + mdp4_write(mdp4_kms, REG_MDP4_PIPE_FETCH_CONFIG(RGB2), vg_cfg); + + if (mdp4_kms->rev >= 2) + mdp4_write(mdp4_kms, REG_MDP4_LAYERMIXER_IN_CFG_UPDATE_METHOD, 1); + mdp4_write(mdp4_kms, REG_MDP4_LAYERMIXER_IN_CFG, 0); + + /* disable CSC matrix / YUV by default: */ + mdp4_write(mdp4_kms, REG_MDP4_PIPE_OP_MODE(VG1), 0); + mdp4_write(mdp4_kms, REG_MDP4_PIPE_OP_MODE(VG2), 0); + mdp4_write(mdp4_kms, REG_MDP4_DMA_P_OP_MODE, 0); + mdp4_write(mdp4_kms, REG_MDP4_DMA_S_OP_MODE, 0); + mdp4_write(mdp4_kms, REG_MDP4_OVLP_CSC_CONFIG(1), 0); + mdp4_write(mdp4_kms, REG_MDP4_OVLP_CSC_CONFIG(2), 0); + + if (mdp4_kms->rev > 1) + mdp4_write(mdp4_kms, REG_MDP4_RESET_STATUS, 1); + + pm_runtime_put_sync(dev->dev); + + return 0; +} + +static void mdp4_enable_commit(struct msm_kms *kms) +{ + struct mdp4_kms *mdp4_kms = to_mdp4_kms(to_mdp_kms(kms)); + mdp4_enable(mdp4_kms); +} + +static void mdp4_disable_commit(struct msm_kms *kms) +{ + struct mdp4_kms *mdp4_kms = to_mdp4_kms(to_mdp_kms(kms)); + mdp4_disable(mdp4_kms); +} + +static void mdp4_flush_commit(struct msm_kms *kms, unsigned crtc_mask) +{ + /* TODO */ +} + +static void mdp4_wait_flush(struct msm_kms *kms, unsigned crtc_mask) +{ + struct mdp4_kms *mdp4_kms = to_mdp4_kms(to_mdp_kms(kms)); + struct drm_crtc *crtc; + + for_each_crtc_mask(mdp4_kms->dev, crtc, crtc_mask) + mdp4_crtc_wait_for_commit_done(crtc); +} + +static void mdp4_complete_commit(struct msm_kms *kms, unsigned crtc_mask) +{ +} + +static long mdp4_round_pixclk(struct msm_kms *kms, unsigned long rate, + struct drm_encoder *encoder) +{ + /* if we had >1 encoder, we'd need something more clever: */ + switch (encoder->encoder_type) { + case DRM_MODE_ENCODER_TMDS: + return mdp4_dtv_round_pixclk(encoder, rate); + case DRM_MODE_ENCODER_LVDS: + case DRM_MODE_ENCODER_DSI: + default: + return rate; + } +} + +static void mdp4_destroy(struct msm_kms *kms) +{ + struct mdp4_kms *mdp4_kms = to_mdp4_kms(to_mdp_kms(kms)); + struct device *dev = mdp4_kms->dev->dev; + + if (mdp4_kms->blank_cursor_iova) + msm_gem_unpin_iova(mdp4_kms->blank_cursor_bo, kms->vm); + drm_gem_object_put(mdp4_kms->blank_cursor_bo); + + if (kms->vm) { + struct msm_mmu *mmu = to_msm_vm(kms->vm)->mmu; + + mmu->funcs->detach(mmu); + drm_gpuvm_put(kms->vm); + } + + if (mdp4_kms->rpm_enabled) + pm_runtime_disable(dev); + + mdp_kms_destroy(&mdp4_kms->base); +} + +static const struct mdp_kms_funcs kms_funcs = { + .base = { + .hw_init = mdp4_hw_init, + .irq_preinstall = mdp4_irq_preinstall, + .irq_postinstall = mdp4_irq_postinstall, + .irq_uninstall = mdp4_irq_uninstall, + .irq = mdp4_irq, + .enable_vblank = mdp4_enable_vblank, + .disable_vblank = mdp4_disable_vblank, + .enable_commit = mdp4_enable_commit, + .disable_commit = mdp4_disable_commit, + .flush_commit = mdp4_flush_commit, + .wait_flush = mdp4_wait_flush, + .complete_commit = mdp4_complete_commit, + .round_pixclk = mdp4_round_pixclk, + .destroy = mdp4_destroy, + }, + .set_irqmask = mdp4_set_irqmask, +}; + +int mdp4_disable(struct mdp4_kms *mdp4_kms) +{ + DBG(""); + + clk_disable_unprepare(mdp4_kms->clk); + clk_disable_unprepare(mdp4_kms->pclk); + clk_disable_unprepare(mdp4_kms->lut_clk); + clk_disable_unprepare(mdp4_kms->axi_clk); + + return 0; +} + +int mdp4_enable(struct mdp4_kms *mdp4_kms) +{ + DBG(""); + + clk_prepare_enable(mdp4_kms->clk); + clk_prepare_enable(mdp4_kms->pclk); + clk_prepare_enable(mdp4_kms->lut_clk); + clk_prepare_enable(mdp4_kms->axi_clk); + + return 0; +} + + +static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms, + int intf_type) +{ + struct drm_device *dev = mdp4_kms->dev; + struct msm_drm_private *priv = dev->dev_private; + struct drm_encoder *encoder; + struct drm_connector *connector; + struct drm_bridge *next_bridge; + int dsi_id; + int ret; + + switch (intf_type) { + case DRM_MODE_ENCODER_LVDS: + /* + * bail out early if there is no panel node (no need to + * initialize LCDC encoder and LVDS connector) + */ + next_bridge = devm_drm_of_get_bridge(dev->dev, dev->dev->of_node, 0, 0); + if (IS_ERR(next_bridge)) { + ret = PTR_ERR(next_bridge); + if (ret == -ENODEV) + return 0; + return ret; + } + + encoder = mdp4_lcdc_encoder_init(dev); + if (IS_ERR(encoder)) { + DRM_DEV_ERROR(dev->dev, "failed to construct LCDC encoder\n"); + return PTR_ERR(encoder); + } + + /* LCDC can be hooked to DMA_P (TODO: Add DMA_S later?) */ + encoder->possible_crtcs = 1 << DMA_P; + + ret = drm_bridge_attach(encoder, next_bridge, NULL, DRM_BRIDGE_ATTACH_NO_CONNECTOR); + if (ret) { + DRM_DEV_ERROR(dev->dev, "failed to attach LVDS panel/bridge: %d\n", ret); + + return ret; + } + + connector = drm_bridge_connector_init(dev, encoder); + if (IS_ERR(connector)) { + DRM_DEV_ERROR(dev->dev, "failed to initialize LVDS connector\n"); + return PTR_ERR(connector); + } + + break; + case DRM_MODE_ENCODER_TMDS: + encoder = mdp4_dtv_encoder_init(dev); + if (IS_ERR(encoder)) { + DRM_DEV_ERROR(dev->dev, "failed to construct DTV encoder\n"); + return PTR_ERR(encoder); + } + + /* DTV can be hooked to DMA_E: */ + encoder->possible_crtcs = 1 << 1; + + if (priv->kms->hdmi) { + /* Construct bridge/connector for HDMI: */ + ret = msm_hdmi_modeset_init(priv->kms->hdmi, dev, encoder); + if (ret) { + DRM_DEV_ERROR(dev->dev, "failed to initialize HDMI: %d\n", ret); + return ret; + } + } + + break; + case DRM_MODE_ENCODER_DSI: + /* only DSI1 supported for now */ + dsi_id = 0; + + if (!priv->kms->dsi[dsi_id]) + break; + + encoder = mdp4_dsi_encoder_init(dev); + if (IS_ERR(encoder)) { + ret = PTR_ERR(encoder); + DRM_DEV_ERROR(dev->dev, + "failed to construct DSI encoder: %d\n", ret); + return ret; + } + + /* TODO: Add DMA_S later? */ + encoder->possible_crtcs = 1 << DMA_P; + + ret = msm_dsi_modeset_init(priv->kms->dsi[dsi_id], dev, encoder); + if (ret) { + DRM_DEV_ERROR(dev->dev, "failed to initialize DSI: %d\n", + ret); + return ret; + } + + break; + default: + DRM_DEV_ERROR(dev->dev, "Invalid or unsupported interface\n"); + return -EINVAL; + } + + return 0; +} + +static int modeset_init(struct mdp4_kms *mdp4_kms) +{ + struct drm_device *dev = mdp4_kms->dev; + struct drm_plane *plane; + struct drm_crtc *crtc; + int i, ret; + static const enum mdp4_pipe rgb_planes[] = { + RGB1, RGB2, + }; + static const enum mdp4_pipe vg_planes[] = { + VG1, VG2, + }; + static const enum mdp4_dma mdp4_crtcs[] = { + DMA_P, DMA_E, + }; + static const char * const mdp4_crtc_names[] = { + "DMA_P", "DMA_E", + }; + static const int mdp4_intfs[] = { + DRM_MODE_ENCODER_LVDS, + DRM_MODE_ENCODER_DSI, + DRM_MODE_ENCODER_TMDS, + }; + + /* construct non-private planes: */ + for (i = 0; i < ARRAY_SIZE(vg_planes); i++) { + plane = mdp4_plane_init(dev, vg_planes[i], false); + if (IS_ERR(plane)) { + DRM_DEV_ERROR(dev->dev, + "failed to construct plane for VG%d\n", i + 1); + ret = PTR_ERR(plane); + goto fail; + } + } + + for (i = 0; i < ARRAY_SIZE(mdp4_crtcs); i++) { + plane = mdp4_plane_init(dev, rgb_planes[i], true); + if (IS_ERR(plane)) { + DRM_DEV_ERROR(dev->dev, + "failed to construct plane for RGB%d\n", i + 1); + ret = PTR_ERR(plane); + goto fail; + } + + crtc = mdp4_crtc_init(dev, plane, i, + mdp4_crtcs[i]); + if (IS_ERR(crtc)) { + DRM_DEV_ERROR(dev->dev, "failed to construct crtc for %s\n", + mdp4_crtc_names[i]); + ret = PTR_ERR(crtc); + goto fail; + } + } + + /* + * we currently set up two relatively fixed paths: + * + * LCDC/LVDS path: RGB1 -> DMA_P -> LCDC -> LVDS + * or + * DSI path: RGB1 -> DMA_P -> DSI1 -> DSI Panel + * + * DTV/HDMI path: RGB2 -> DMA_E -> DTV -> HDMI + */ + + for (i = 0; i < ARRAY_SIZE(mdp4_intfs); i++) { + ret = mdp4_modeset_init_intf(mdp4_kms, mdp4_intfs[i]); + if (ret) { + DRM_DEV_ERROR(dev->dev, "failed to initialize intf: %d, %d\n", + i, ret); + goto fail; + } + } + + return 0; + +fail: + return ret; +} + +static void read_mdp_hw_revision(struct mdp4_kms *mdp4_kms, + u32 *major, u32 *minor) +{ + struct drm_device *dev = mdp4_kms->dev; + u32 version; + + mdp4_enable(mdp4_kms); + version = mdp4_read(mdp4_kms, REG_MDP4_VERSION); + mdp4_disable(mdp4_kms); + + *major = FIELD(version, MDP4_VERSION_MAJOR); + *minor = FIELD(version, MDP4_VERSION_MINOR); + + DRM_DEV_INFO(dev->dev, "MDP4 version v%d.%d", *major, *minor); +} + +static int mdp4_kms_init(struct drm_device *dev) +{ + struct msm_drm_private *priv = dev->dev_private; + struct mdp4_kms *mdp4_kms = to_mdp4_kms(to_mdp_kms(priv->kms)); + struct msm_kms *kms = NULL; + struct drm_gpuvm *vm; + int ret; + u32 major, minor; + unsigned long max_clk; + + /* TODO: Chips that aren't apq8064 have a 200 Mhz max_clk */ + max_clk = 266667000; + + ret = mdp_kms_init(&mdp4_kms->base, &kms_funcs); + if (ret) { + DRM_DEV_ERROR(dev->dev, "failed to init kms\n"); + return ret; + } + + kms = priv->kms; + + mdp4_kms->dev = dev; + + if (mdp4_kms->vdd) { + ret = regulator_enable(mdp4_kms->vdd); + if (ret) { + DRM_DEV_ERROR(dev->dev, "failed to enable regulator vdd: %d\n", ret); + return ret; + } + } + + clk_set_rate(mdp4_kms->clk, max_clk); + + read_mdp_hw_revision(mdp4_kms, &major, &minor); + + if (major != 4) { + DRM_DEV_ERROR(dev->dev, "unexpected MDP version: v%d.%d\n", + major, minor); + ret = -ENXIO; + return ret; + } + + mdp4_kms->rev = minor; + + if (mdp4_kms->rev >= 2) { + if (!mdp4_kms->lut_clk) { + DRM_DEV_ERROR(dev->dev, "failed to get lut_clk\n"); + ret = -ENODEV; + return ret; + } + clk_set_rate(mdp4_kms->lut_clk, max_clk); + } + + pm_runtime_enable(dev->dev); + mdp4_kms->rpm_enabled = true; + + /* make sure things are off before attaching iommu (bootloader could + * have left things on, in which case we'll start getting faults if + * we don't disable): + */ + mdp4_enable(mdp4_kms); + mdp4_write(mdp4_kms, REG_MDP4_DTV_ENABLE, 0); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_ENABLE, 0); + mdp4_write(mdp4_kms, REG_MDP4_DSI_ENABLE, 0); + mdp4_disable(mdp4_kms); + mdelay(16); + + vm = msm_kms_init_vm(mdp4_kms->dev, NULL); + if (IS_ERR(vm)) { + ret = PTR_ERR(vm); + return ret; + } + + kms->vm = vm; + + ret = modeset_init(mdp4_kms); + if (ret) { + DRM_DEV_ERROR(dev->dev, "modeset_init failed: %d\n", ret); + return ret; + } + + mdp4_kms->blank_cursor_bo = msm_gem_new(dev, SZ_16K, MSM_BO_WC | MSM_BO_SCANOUT, NULL); + if (IS_ERR(mdp4_kms->blank_cursor_bo)) { + ret = PTR_ERR(mdp4_kms->blank_cursor_bo); + DRM_DEV_ERROR(dev->dev, "could not allocate blank-cursor bo: %d\n", ret); + mdp4_kms->blank_cursor_bo = NULL; + return ret; + } + + ret = msm_gem_get_and_pin_iova(mdp4_kms->blank_cursor_bo, kms->vm, + &mdp4_kms->blank_cursor_iova); + if (ret) { + DRM_DEV_ERROR(dev->dev, "could not pin blank-cursor bo: %d\n", ret); + return ret; + } + + dev->mode_config.min_width = 0; + dev->mode_config.min_height = 0; + dev->mode_config.max_width = 2048; + dev->mode_config.max_height = 2048; + + return 0; +} + +static const struct dev_pm_ops mdp4_pm_ops = { + .prepare = msm_kms_pm_prepare, + .complete = msm_kms_pm_complete, +}; + +static int mdp4_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct mdp4_kms *mdp4_kms; + int irq; + + mdp4_kms = devm_kzalloc(dev, sizeof(*mdp4_kms), GFP_KERNEL); + if (!mdp4_kms) + return -ENOMEM; + + mdp4_kms->mmio = msm_ioremap(pdev, NULL); + if (IS_ERR(mdp4_kms->mmio)) + return PTR_ERR(mdp4_kms->mmio); + + irq = platform_get_irq(pdev, 0); + if (irq < 0) + return dev_err_probe(dev, irq, "failed to get irq\n"); + + mdp4_kms->base.base.irq = irq; + + /* NOTE: driver for this regulator still missing upstream.. use + * _get_exclusive() and ignore the error if it does not exist + * (and hope that the bootloader left it on for us) + */ + mdp4_kms->vdd = devm_regulator_get_exclusive(&pdev->dev, "vdd"); + if (IS_ERR(mdp4_kms->vdd)) + mdp4_kms->vdd = NULL; + + mdp4_kms->clk = devm_clk_get(&pdev->dev, "core_clk"); + if (IS_ERR(mdp4_kms->clk)) + return dev_err_probe(dev, PTR_ERR(mdp4_kms->clk), "failed to get core_clk\n"); + + mdp4_kms->pclk = devm_clk_get(&pdev->dev, "iface_clk"); + if (IS_ERR(mdp4_kms->pclk)) + mdp4_kms->pclk = NULL; + + mdp4_kms->axi_clk = devm_clk_get(&pdev->dev, "bus_clk"); + if (IS_ERR(mdp4_kms->axi_clk)) + return dev_err_probe(dev, PTR_ERR(mdp4_kms->axi_clk), "failed to get axi_clk\n"); + + /* + * This is required for revn >= 2. Handle errors here and let the kms + * init bail out if the clock is not provided. + */ + mdp4_kms->lut_clk = devm_clk_get_optional(&pdev->dev, "lut_clk"); + if (IS_ERR(mdp4_kms->lut_clk)) + return dev_err_probe(dev, PTR_ERR(mdp4_kms->lut_clk), "failed to get lut_clk\n"); + + return msm_drv_probe(&pdev->dev, mdp4_kms_init, &mdp4_kms->base.base); +} + +static void mdp4_remove(struct platform_device *pdev) +{ + component_master_del(&pdev->dev, &msm_drm_ops); +} + +static const struct of_device_id mdp4_dt_match[] = { + { .compatible = "qcom,mdp4" }, + { /* sentinel */ } +}; +MODULE_DEVICE_TABLE(of, mdp4_dt_match); + +static struct platform_driver mdp4_platform_driver = { + .probe = mdp4_probe, + .remove = mdp4_remove, + .shutdown = msm_kms_shutdown, + .driver = { + .name = "mdp4", + .of_match_table = mdp4_dt_match, + .pm = &mdp4_pm_ops, + }, +}; + +void __init msm_mdp4_register(void) +{ + platform_driver_register(&mdp4_platform_driver); +} + +void __exit msm_mdp4_unregister(void) +{ + platform_driver_unregister(&mdp4_platform_driver); +} diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h new file mode 100644 index 000000000..06458d4ee --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h @@ -0,0 +1,207 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#ifndef __MDP4_KMS_H__ +#define __MDP4_KMS_H__ + +#include + +#include "msm_drv.h" +#include "msm_kms.h" +#include "disp/mdp_kms.h" +#include "mdp4.xml.h" + +struct device_node; + +struct mdp4_kms { + struct mdp_kms base; + + struct drm_device *dev; + + int rev; + + void __iomem *mmio; + + struct regulator *vdd; + + struct clk *clk; + struct clk *pclk; + struct clk *lut_clk; + struct clk *axi_clk; + + struct mdp_irq error_handler; + + bool rpm_enabled; + + /* empty/blank cursor bo to use when cursor is "disabled" */ + struct drm_gem_object *blank_cursor_bo; + uint64_t blank_cursor_iova; +}; +#define to_mdp4_kms(x) container_of(x, struct mdp4_kms, base) + +static inline void mdp4_write(struct mdp4_kms *mdp4_kms, u32 reg, u32 data) +{ + writel(data, mdp4_kms->mmio + reg); +} + +static inline u32 mdp4_read(struct mdp4_kms *mdp4_kms, u32 reg) +{ + return readl(mdp4_kms->mmio + reg); +} + +static inline uint32_t pipe2flush(enum mdp4_pipe pipe) +{ + switch (pipe) { + case VG1: return MDP4_OVERLAY_FLUSH_VG1; + case VG2: return MDP4_OVERLAY_FLUSH_VG2; + case RGB1: return MDP4_OVERLAY_FLUSH_RGB1; + case RGB2: return MDP4_OVERLAY_FLUSH_RGB2; + default: return 0; + } +} + +static inline uint32_t ovlp2flush(int ovlp) +{ + switch (ovlp) { + case 0: return MDP4_OVERLAY_FLUSH_OVLP0; + case 1: return MDP4_OVERLAY_FLUSH_OVLP1; + default: return 0; + } +} + +static inline uint32_t dma2irq(enum mdp4_dma dma) +{ + switch (dma) { + case DMA_P: return MDP4_IRQ_DMA_P_DONE; + case DMA_S: return MDP4_IRQ_DMA_S_DONE; + case DMA_E: return MDP4_IRQ_DMA_E_DONE; + default: return 0; + } +} + +static inline uint32_t dma2err(enum mdp4_dma dma) +{ + switch (dma) { + case DMA_P: return MDP4_IRQ_PRIMARY_INTF_UDERRUN; + case DMA_S: return 0; // ??? + case DMA_E: return MDP4_IRQ_EXTERNAL_INTF_UDERRUN; + default: return 0; + } +} + +static inline uint32_t mixercfg(uint32_t mixer_cfg, int mixer, + enum mdp4_pipe pipe, enum mdp_mixer_stage_id stage) +{ + switch (pipe) { + case VG1: + mixer_cfg &= ~(MDP4_LAYERMIXER_IN_CFG_PIPE0__MASK | + MDP4_LAYERMIXER_IN_CFG_PIPE0_MIXER1); + mixer_cfg |= MDP4_LAYERMIXER_IN_CFG_PIPE0(stage) | + COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE0_MIXER1); + break; + case VG2: + mixer_cfg &= ~(MDP4_LAYERMIXER_IN_CFG_PIPE1__MASK | + MDP4_LAYERMIXER_IN_CFG_PIPE1_MIXER1); + mixer_cfg |= MDP4_LAYERMIXER_IN_CFG_PIPE1(stage) | + COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE1_MIXER1); + break; + case RGB1: + mixer_cfg &= ~(MDP4_LAYERMIXER_IN_CFG_PIPE2__MASK | + MDP4_LAYERMIXER_IN_CFG_PIPE2_MIXER1); + mixer_cfg |= MDP4_LAYERMIXER_IN_CFG_PIPE2(stage) | + COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE2_MIXER1); + break; + case RGB2: + mixer_cfg &= ~(MDP4_LAYERMIXER_IN_CFG_PIPE3__MASK | + MDP4_LAYERMIXER_IN_CFG_PIPE3_MIXER1); + mixer_cfg |= MDP4_LAYERMIXER_IN_CFG_PIPE3(stage) | + COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE3_MIXER1); + break; + case RGB3: + mixer_cfg &= ~(MDP4_LAYERMIXER_IN_CFG_PIPE4__MASK | + MDP4_LAYERMIXER_IN_CFG_PIPE4_MIXER1); + mixer_cfg |= MDP4_LAYERMIXER_IN_CFG_PIPE4(stage) | + COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE4_MIXER1); + break; + case VG3: + mixer_cfg &= ~(MDP4_LAYERMIXER_IN_CFG_PIPE5__MASK | + MDP4_LAYERMIXER_IN_CFG_PIPE5_MIXER1); + mixer_cfg |= MDP4_LAYERMIXER_IN_CFG_PIPE5(stage) | + COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE5_MIXER1); + break; + case VG4: + mixer_cfg &= ~(MDP4_LAYERMIXER_IN_CFG_PIPE6__MASK | + MDP4_LAYERMIXER_IN_CFG_PIPE6_MIXER1); + mixer_cfg |= MDP4_LAYERMIXER_IN_CFG_PIPE6(stage) | + COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE6_MIXER1); + break; + default: + WARN(1, "invalid pipe"); + break; + } + + return mixer_cfg; +} + +int mdp4_disable(struct mdp4_kms *mdp4_kms); +int mdp4_enable(struct mdp4_kms *mdp4_kms); + +void mdp4_set_irqmask(struct mdp_kms *mdp_kms, uint32_t irqmask, + uint32_t old_irqmask); +void mdp4_irq_preinstall(struct msm_kms *kms); +int mdp4_irq_postinstall(struct msm_kms *kms); +void mdp4_irq_uninstall(struct msm_kms *kms); +irqreturn_t mdp4_irq(struct msm_kms *kms); +int mdp4_enable_vblank(struct msm_kms *kms, struct drm_crtc *crtc); +void mdp4_disable_vblank(struct msm_kms *kms, struct drm_crtc *crtc); + +static inline uint32_t mdp4_pipe_caps(enum mdp4_pipe pipe) +{ + switch (pipe) { + case VG1: + case VG2: + case VG3: + case VG4: + return MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | MDP_PIPE_CAP_CSC; + case RGB1: + case RGB2: + case RGB3: + return MDP_PIPE_CAP_SCALE; + default: + return 0; + } +} + +enum mdp4_pipe mdp4_plane_pipe(struct drm_plane *plane); +struct drm_plane *mdp4_plane_init(struct drm_device *dev, + enum mdp4_pipe pipe_id, bool private_plane); + +uint32_t mdp4_crtc_vblank(struct drm_crtc *crtc); +void mdp4_crtc_set_config(struct drm_crtc *crtc, uint32_t config); +void mdp4_crtc_set_intf(struct drm_crtc *crtc, enum mdp4_intf intf, int mixer); +void mdp4_crtc_wait_for_commit_done(struct drm_crtc *crtc); +struct drm_crtc *mdp4_crtc_init(struct drm_device *dev, + struct drm_plane *plane, int ovlp_id, + enum mdp4_dma dma_id); + +long mdp4_dtv_round_pixclk(struct drm_encoder *encoder, unsigned long rate); +struct drm_encoder *mdp4_dtv_encoder_init(struct drm_device *dev); + +struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev); + +#ifdef CONFIG_DRM_MSM_DSI +struct drm_encoder *mdp4_dsi_encoder_init(struct drm_device *dev); +#else +static inline struct drm_encoder *mdp4_dsi_encoder_init(struct drm_device *dev) +{ + return ERR_PTR(-ENODEV); +} +#endif + +struct clk *mdp4_get_lcdc_clock(struct drm_device *dev); + +#endif /* __MDP4_KMS_H__ */ diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c new file mode 100644 index 000000000..105187305 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c @@ -0,0 +1,396 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2014 Red Hat + * Author: Rob Clark + * Author: Vinay Simha + */ + +#include + +#include +#include + +#include "mdp4_kms.h" + +struct mdp4_lcdc_encoder { + struct drm_encoder base; + struct drm_panel *panel; + struct clk *lcdc_clk; + unsigned long int pixclock; + struct regulator_bulk_data regs[3]; + bool enabled; + uint32_t bsc; +}; +#define to_mdp4_lcdc_encoder(x) container_of(x, struct mdp4_lcdc_encoder, base) + +static struct mdp4_kms *get_kms(struct drm_encoder *encoder) +{ + struct msm_drm_private *priv = encoder->dev->dev_private; + return to_mdp4_kms(to_mdp_kms(priv->kms)); +} + +/* this should probably be a helper: */ +static struct drm_connector *get_connector(struct drm_encoder *encoder) +{ + struct drm_device *dev = encoder->dev; + struct drm_connector *connector; + + list_for_each_entry(connector, &dev->mode_config.connector_list, head) + if (connector->encoder == encoder) + return connector; + + return NULL; +} + +static void setup_phy(struct drm_encoder *encoder) +{ + struct drm_device *dev = encoder->dev; + struct drm_connector *connector = get_connector(encoder); + struct mdp4_kms *mdp4_kms = get_kms(encoder); + uint32_t lvds_intf = 0, lvds_phy_cfg0 = 0; + int bpp, nchan, swap; + + if (!connector) + return; + + bpp = 3 * connector->display_info.bpc; + + if (!bpp) + bpp = 18; + + /* TODO, these should come from panel somehow: */ + nchan = 1; + swap = 0; + + switch (bpp) { + case 24: + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_MUX_CTL_3_TO_0(0), + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT0(0x08) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT1(0x05) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT2(0x04) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT3(0x03)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_MUX_CTL_6_TO_4(0), + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT4(0x02) | + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT5(0x01) | + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT6(0x00)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_MUX_CTL_3_TO_0(1), + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT0(0x11) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT1(0x10) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT2(0x0d) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT3(0x0c)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_MUX_CTL_6_TO_4(1), + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT4(0x0b) | + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT5(0x0a) | + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT6(0x09)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_MUX_CTL_3_TO_0(2), + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT0(0x1a) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT1(0x19) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT2(0x18) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT3(0x15)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_MUX_CTL_6_TO_4(2), + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT4(0x14) | + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT5(0x13) | + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT6(0x12)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_MUX_CTL_3_TO_0(3), + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT0(0x1b) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT1(0x17) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT2(0x16) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT3(0x0f)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_MUX_CTL_6_TO_4(3), + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT4(0x0e) | + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT5(0x07) | + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT6(0x06)); + if (nchan == 2) { + lvds_intf |= MDP4_LCDC_LVDS_INTF_CTL_CH2_DATA_LANE3_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH2_DATA_LANE2_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH2_DATA_LANE1_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH2_DATA_LANE0_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH1_DATA_LANE3_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH1_DATA_LANE2_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH1_DATA_LANE1_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH1_DATA_LANE0_EN; + } else { + lvds_intf |= MDP4_LCDC_LVDS_INTF_CTL_CH1_DATA_LANE3_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH1_DATA_LANE2_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH1_DATA_LANE1_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH1_DATA_LANE0_EN; + } + break; + + case 18: + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_MUX_CTL_3_TO_0(0), + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT0(0x0a) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT1(0x07) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT2(0x06) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT3(0x05)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_MUX_CTL_6_TO_4(0), + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT4(0x04) | + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT5(0x03) | + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT6(0x02)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_MUX_CTL_3_TO_0(1), + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT0(0x13) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT1(0x12) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT2(0x0f) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT3(0x0e)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_MUX_CTL_6_TO_4(1), + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT4(0x0d) | + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT5(0x0c) | + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT6(0x0b)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_MUX_CTL_3_TO_0(2), + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT0(0x1a) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT1(0x19) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT2(0x18) | + MDP4_LCDC_LVDS_MUX_CTL_3_TO_0_BIT3(0x17)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_MUX_CTL_6_TO_4(2), + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT4(0x16) | + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT5(0x15) | + MDP4_LCDC_LVDS_MUX_CTL_6_TO_4_BIT6(0x14)); + if (nchan == 2) { + lvds_intf |= MDP4_LCDC_LVDS_INTF_CTL_CH2_DATA_LANE2_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH2_DATA_LANE1_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH2_DATA_LANE0_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH1_DATA_LANE2_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH1_DATA_LANE1_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH1_DATA_LANE0_EN; + } else { + lvds_intf |= MDP4_LCDC_LVDS_INTF_CTL_CH1_DATA_LANE2_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH1_DATA_LANE1_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH1_DATA_LANE0_EN; + } + lvds_intf |= MDP4_LCDC_LVDS_INTF_CTL_RGB_OUT; + break; + + default: + DRM_DEV_ERROR(dev->dev, "unknown bpp: %d\n", bpp); + return; + } + + switch (nchan) { + case 1: + lvds_phy_cfg0 = MDP4_LVDS_PHY_CFG0_CHANNEL0; + lvds_intf |= MDP4_LCDC_LVDS_INTF_CTL_CH1_CLK_LANE_EN | + MDP4_LCDC_LVDS_INTF_CTL_MODE_SEL; + break; + case 2: + lvds_phy_cfg0 = MDP4_LVDS_PHY_CFG0_CHANNEL0 | + MDP4_LVDS_PHY_CFG0_CHANNEL1; + lvds_intf |= MDP4_LCDC_LVDS_INTF_CTL_CH2_CLK_LANE_EN | + MDP4_LCDC_LVDS_INTF_CTL_CH1_CLK_LANE_EN; + break; + default: + DRM_DEV_ERROR(dev->dev, "unknown # of channels: %d\n", nchan); + return; + } + + if (swap) + lvds_intf |= MDP4_LCDC_LVDS_INTF_CTL_CH_SWAP; + + lvds_intf |= MDP4_LCDC_LVDS_INTF_CTL_ENABLE; + + mdp4_write(mdp4_kms, REG_MDP4_LVDS_PHY_CFG0, lvds_phy_cfg0); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_INTF_CTL, lvds_intf); + mdp4_write(mdp4_kms, REG_MDP4_LVDS_PHY_CFG2, 0x30); + + mb(); + udelay(1); + lvds_phy_cfg0 |= MDP4_LVDS_PHY_CFG0_SERIALIZATION_ENBLE; + mdp4_write(mdp4_kms, REG_MDP4_LVDS_PHY_CFG0, lvds_phy_cfg0); +} + +static void mdp4_lcdc_encoder_mode_set(struct drm_encoder *encoder, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + struct mdp4_lcdc_encoder *mdp4_lcdc_encoder = + to_mdp4_lcdc_encoder(encoder); + struct mdp4_kms *mdp4_kms = get_kms(encoder); + uint32_t lcdc_hsync_skew, vsync_period, vsync_len, ctrl_pol; + uint32_t display_v_start, display_v_end; + uint32_t hsync_start_x, hsync_end_x; + + mode = adjusted_mode; + + DBG("set mode: " DRM_MODE_FMT, DRM_MODE_ARG(mode)); + + mdp4_lcdc_encoder->pixclock = mode->clock * 1000; + + DBG("pixclock=%lu", mdp4_lcdc_encoder->pixclock); + + ctrl_pol = 0; + if (mode->flags & DRM_MODE_FLAG_NHSYNC) + ctrl_pol |= MDP4_LCDC_CTRL_POLARITY_HSYNC_LOW; + if (mode->flags & DRM_MODE_FLAG_NVSYNC) + ctrl_pol |= MDP4_LCDC_CTRL_POLARITY_VSYNC_LOW; + /* probably need to get DATA_EN polarity from panel.. */ + + lcdc_hsync_skew = 0; /* get this from panel? */ + + hsync_start_x = (mode->htotal - mode->hsync_start); + hsync_end_x = mode->htotal - (mode->hsync_start - mode->hdisplay) - 1; + + vsync_period = mode->vtotal * mode->htotal; + vsync_len = (mode->vsync_end - mode->vsync_start) * mode->htotal; + display_v_start = (mode->vtotal - mode->vsync_start) * mode->htotal + lcdc_hsync_skew; + display_v_end = vsync_period - ((mode->vsync_start - mode->vdisplay) * mode->htotal) + lcdc_hsync_skew - 1; + + mdp4_write(mdp4_kms, REG_MDP4_LCDC_HSYNC_CTRL, + MDP4_LCDC_HSYNC_CTRL_PULSEW(mode->hsync_end - mode->hsync_start) | + MDP4_LCDC_HSYNC_CTRL_PERIOD(mode->htotal)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_VSYNC_PERIOD, vsync_period); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_VSYNC_LEN, vsync_len); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_DISPLAY_HCTRL, + MDP4_LCDC_DISPLAY_HCTRL_START(hsync_start_x) | + MDP4_LCDC_DISPLAY_HCTRL_END(hsync_end_x)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_DISPLAY_VSTART, display_v_start); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_DISPLAY_VEND, display_v_end); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_BORDER_CLR, 0); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_UNDERFLOW_CLR, + MDP4_LCDC_UNDERFLOW_CLR_ENABLE_RECOVERY | + MDP4_LCDC_UNDERFLOW_CLR_COLOR(0xff)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_HSYNC_SKEW, lcdc_hsync_skew); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_CTRL_POLARITY, ctrl_pol); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_ACTIVE_HCTL, + MDP4_LCDC_ACTIVE_HCTL_START(0) | + MDP4_LCDC_ACTIVE_HCTL_END(0)); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_ACTIVE_VSTART, 0); + mdp4_write(mdp4_kms, REG_MDP4_LCDC_ACTIVE_VEND, 0); +} + +static void mdp4_lcdc_encoder_disable(struct drm_encoder *encoder) +{ + struct mdp4_lcdc_encoder *mdp4_lcdc_encoder = + to_mdp4_lcdc_encoder(encoder); + struct mdp4_kms *mdp4_kms = get_kms(encoder); + + if (WARN_ON(!mdp4_lcdc_encoder->enabled)) + return; + + mdp4_write(mdp4_kms, REG_MDP4_LCDC_ENABLE, 0); + + /* + * Wait for a vsync so we know the ENABLE=0 latched before + * the (connector) source of the vsync's gets disabled, + * otherwise we end up in a funny state if we re-enable + * before the disable latches, which results that some of + * the settings changes for the new modeset (like new + * scanout buffer) don't latch properly.. + */ + mdp_irq_wait(&mdp4_kms->base, MDP4_IRQ_PRIMARY_VSYNC); + + clk_disable_unprepare(mdp4_lcdc_encoder->lcdc_clk); + + regulator_bulk_disable(ARRAY_SIZE(mdp4_lcdc_encoder->regs), + mdp4_lcdc_encoder->regs); + + mdp4_lcdc_encoder->enabled = false; +} + +static void mdp4_lcdc_encoder_enable(struct drm_encoder *encoder) +{ + struct drm_device *dev = encoder->dev; + struct mdp4_lcdc_encoder *mdp4_lcdc_encoder = + to_mdp4_lcdc_encoder(encoder); + unsigned long pc = mdp4_lcdc_encoder->pixclock; + struct mdp4_kms *mdp4_kms = get_kms(encoder); + uint32_t config; + int ret; + + if (WARN_ON(mdp4_lcdc_encoder->enabled)) + return; + + /* TODO: hard-coded for 18bpp: */ + config = + MDP4_DMA_CONFIG_R_BPC(BPC6) | + MDP4_DMA_CONFIG_G_BPC(BPC6) | + MDP4_DMA_CONFIG_B_BPC(BPC6) | + MDP4_DMA_CONFIG_PACK(0x21) | + MDP4_DMA_CONFIG_DEFLKR_EN | + MDP4_DMA_CONFIG_DITHER_EN; + + if (!of_property_read_bool(dev->dev->of_node, "qcom,lcdc-align-lsb")) + config |= MDP4_DMA_CONFIG_PACK_ALIGN_MSB; + + mdp4_crtc_set_config(encoder->crtc, config); + mdp4_crtc_set_intf(encoder->crtc, INTF_LCDC_DTV, 0); + + ret = regulator_bulk_enable(ARRAY_SIZE(mdp4_lcdc_encoder->regs), + mdp4_lcdc_encoder->regs); + if (ret) + DRM_DEV_ERROR(dev->dev, "failed to enable regulators: %d\n", ret); + + DBG("setting lcdc_clk=%lu", pc); + ret = clk_set_rate(mdp4_lcdc_encoder->lcdc_clk, pc); + if (ret) + DRM_DEV_ERROR(dev->dev, "failed to configure lcdc_clk: %d\n", ret); + ret = clk_prepare_enable(mdp4_lcdc_encoder->lcdc_clk); + if (ret) + DRM_DEV_ERROR(dev->dev, "failed to enable lcdc_clk: %d\n", ret); + + setup_phy(encoder); + + mdp4_write(mdp4_kms, REG_MDP4_LCDC_ENABLE, 1); + + mdp4_lcdc_encoder->enabled = true; +} + +static enum drm_mode_status +mdp4_lcdc_encoder_mode_valid(struct drm_encoder *encoder, + const struct drm_display_mode *mode) +{ + struct mdp4_lcdc_encoder *mdp4_lcdc_encoder = + to_mdp4_lcdc_encoder(encoder); + long actual, requested; + + requested = 1000 * mode->clock; + actual = clk_round_rate(mdp4_lcdc_encoder->lcdc_clk, requested); + + DBG("requested=%ld, actual=%ld", requested, actual); + + if (actual != requested) + return MODE_CLOCK_RANGE; + + return MODE_OK; +} + +static const struct drm_encoder_helper_funcs mdp4_lcdc_encoder_helper_funcs = { + .mode_set = mdp4_lcdc_encoder_mode_set, + .disable = mdp4_lcdc_encoder_disable, + .enable = mdp4_lcdc_encoder_enable, + .mode_valid = mdp4_lcdc_encoder_mode_valid, +}; + +/* initialize encoder */ +struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev) +{ + struct drm_encoder *encoder; + struct mdp4_lcdc_encoder *mdp4_lcdc_encoder; + int ret; + + mdp4_lcdc_encoder = drmm_encoder_alloc(dev, struct mdp4_lcdc_encoder, base, + NULL, DRM_MODE_ENCODER_LVDS, NULL); + if (IS_ERR(mdp4_lcdc_encoder)) + return ERR_CAST(mdp4_lcdc_encoder); + + encoder = &mdp4_lcdc_encoder->base; + + drm_encoder_helper_add(encoder, &mdp4_lcdc_encoder_helper_funcs); + + mdp4_lcdc_encoder->lcdc_clk = mdp4_get_lcdc_clock(dev); + if (IS_ERR(mdp4_lcdc_encoder->lcdc_clk)) { + DRM_DEV_ERROR(dev->dev, "failed to get lvds_clk\n"); + return ERR_CAST(mdp4_lcdc_encoder->lcdc_clk); + } + + /* TODO: different regulators in other cases? */ + mdp4_lcdc_encoder->regs[0].supply = "lvds-vccs-3p3v"; + mdp4_lcdc_encoder->regs[1].supply = "lvds-pll-vdda"; + mdp4_lcdc_encoder->regs[2].supply = "lvds-vdda"; + + ret = devm_regulator_bulk_get(dev->dev, + ARRAY_SIZE(mdp4_lcdc_encoder->regs), + mdp4_lcdc_encoder->regs); + if (ret) + return ERR_PTR(ret); + + return encoder; +} diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c new file mode 100644 index 000000000..04c49bf3d --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c @@ -0,0 +1,183 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2014 Red Hat + * Author: Rob Clark + */ + +#include +#include + +#include "mdp4_kms.h" + +struct mdp4_lvds_pll { + struct clk_hw pll_hw; + struct drm_device *dev; + unsigned long pixclk; +}; +#define to_mdp4_lvds_pll(x) container_of(x, struct mdp4_lvds_pll, pll_hw) + +static struct mdp4_kms *get_kms(struct mdp4_lvds_pll *lvds_pll) +{ + struct msm_drm_private *priv = lvds_pll->dev->dev_private; + return to_mdp4_kms(to_mdp_kms(priv->kms)); +} + +struct pll_rate { + unsigned long rate; + struct { + uint32_t val; + uint32_t reg; + } conf[32]; +}; + +/* NOTE: keep sorted highest freq to lowest: */ +static const struct pll_rate freqtbl[] = { + { 72000000, { + { 0x8f, REG_MDP4_LVDS_PHY_PLL_CTRL_1 }, + { 0x30, REG_MDP4_LVDS_PHY_PLL_CTRL_2 }, + { 0xc6, REG_MDP4_LVDS_PHY_PLL_CTRL_3 }, + { 0x10, REG_MDP4_LVDS_PHY_PLL_CTRL_5 }, + { 0x07, REG_MDP4_LVDS_PHY_PLL_CTRL_6 }, + { 0x62, REG_MDP4_LVDS_PHY_PLL_CTRL_7 }, + { 0x41, REG_MDP4_LVDS_PHY_PLL_CTRL_8 }, + { 0x0d, REG_MDP4_LVDS_PHY_PLL_CTRL_9 }, + { 0, 0 } } + }, +}; + +static const struct pll_rate *find_rate(unsigned long rate) +{ + int i; + for (i = 1; i < ARRAY_SIZE(freqtbl); i++) + if (rate > freqtbl[i].rate) + return &freqtbl[i-1]; + return &freqtbl[i-1]; +} + +static int mdp4_lvds_pll_enable(struct clk_hw *hw) +{ + struct mdp4_lvds_pll *lvds_pll = to_mdp4_lvds_pll(hw); + struct mdp4_kms *mdp4_kms = get_kms(lvds_pll); + const struct pll_rate *pll_rate = find_rate(lvds_pll->pixclk); + int i; + + DBG("pixclk=%lu (%lu)", lvds_pll->pixclk, pll_rate->rate); + + if (WARN_ON(!pll_rate)) + return -EINVAL; + + mdp4_write(mdp4_kms, REG_MDP4_LCDC_LVDS_PHY_RESET, 0x33); + + for (i = 0; pll_rate->conf[i].reg; i++) + mdp4_write(mdp4_kms, pll_rate->conf[i].reg, pll_rate->conf[i].val); + + mdp4_write(mdp4_kms, REG_MDP4_LVDS_PHY_PLL_CTRL_0, 0x01); + + /* Wait until LVDS PLL is locked and ready */ + while (!mdp4_read(mdp4_kms, REG_MDP4_LVDS_PHY_PLL_LOCKED)) + cpu_relax(); + + return 0; +} + +static void mdp4_lvds_pll_disable(struct clk_hw *hw) +{ + struct mdp4_lvds_pll *lvds_pll = to_mdp4_lvds_pll(hw); + struct mdp4_kms *mdp4_kms = get_kms(lvds_pll); + + DBG(""); + + mdp4_write(mdp4_kms, REG_MDP4_LVDS_PHY_CFG0, 0x0); + mdp4_write(mdp4_kms, REG_MDP4_LVDS_PHY_PLL_CTRL_0, 0x0); +} + +static unsigned long mdp4_lvds_pll_recalc_rate(struct clk_hw *hw, + unsigned long parent_rate) +{ + struct mdp4_lvds_pll *lvds_pll = to_mdp4_lvds_pll(hw); + return lvds_pll->pixclk; +} + +static int mdp4_lvds_pll_determine_rate(struct clk_hw *hw, + struct clk_rate_request *req) +{ + const struct pll_rate *pll_rate = find_rate(req->rate); + + req->rate = pll_rate->rate; + + return 0; +} + +static int mdp4_lvds_pll_set_rate(struct clk_hw *hw, unsigned long rate, + unsigned long parent_rate) +{ + struct mdp4_lvds_pll *lvds_pll = to_mdp4_lvds_pll(hw); + lvds_pll->pixclk = rate; + return 0; +} + + +static const struct clk_ops mdp4_lvds_pll_ops = { + .enable = mdp4_lvds_pll_enable, + .disable = mdp4_lvds_pll_disable, + .recalc_rate = mdp4_lvds_pll_recalc_rate, + .determine_rate = mdp4_lvds_pll_determine_rate, + .set_rate = mdp4_lvds_pll_set_rate, +}; + +static const struct clk_parent_data mdp4_lvds_pll_parents[] = { + { .fw_name = "pxo", .name = "pxo", }, +}; + +static struct clk_init_data pll_init = { + .name = "mdp4_lvds_pll", + .ops = &mdp4_lvds_pll_ops, + .parent_data = mdp4_lvds_pll_parents, + .num_parents = ARRAY_SIZE(mdp4_lvds_pll_parents), +}; + +static struct clk_hw *mdp4_lvds_pll_init(struct drm_device *dev) +{ + struct mdp4_lvds_pll *lvds_pll; + int ret; + + lvds_pll = devm_kzalloc(dev->dev, sizeof(*lvds_pll), GFP_KERNEL); + if (!lvds_pll) + return ERR_PTR(-ENOMEM); + + lvds_pll->dev = dev; + + lvds_pll->pll_hw.init = &pll_init; + ret = devm_clk_hw_register(dev->dev, &lvds_pll->pll_hw); + if (ret) + return ERR_PTR(ret); + + ret = devm_of_clk_add_hw_provider(dev->dev, of_clk_hw_simple_get, &lvds_pll->pll_hw); + if (ret) + return ERR_PTR(ret); + + return &lvds_pll->pll_hw; +} + +struct clk *mdp4_get_lcdc_clock(struct drm_device *dev) +{ + struct clk_hw *hw; + struct clk *clk; + + + /* TODO: do we need different pll in other cases? */ + hw = mdp4_lvds_pll_init(dev); + if (IS_ERR(hw)) { + DRM_DEV_ERROR(dev->dev, "failed to register LVDS PLL\n"); + return ERR_CAST(hw); + } + + clk = devm_clk_get(dev->dev, "lcdc_clk"); + if (clk == ERR_PTR(-ENOENT)) { + drm_warn(dev, "can't get LCDC clock, using PLL directly\n"); + + return devm_clk_hw_get_clk(dev->dev, hw, "lcdc_clk"); + } + + return clk; +} diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c new file mode 100644 index 000000000..9459f70ce --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c @@ -0,0 +1,436 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#include +#include +#include +#include +#include + +#include "mdp4_kms.h" + +#define DOWN_SCALE_MAX 8 +#define UP_SCALE_MAX 8 + +struct mdp4_plane { + struct drm_plane base; + const char *name; + + enum mdp4_pipe pipe; +}; +#define to_mdp4_plane(x) container_of(x, struct mdp4_plane, base) + +/* MDP format helper functions */ +static inline +enum mdp4_frame_format mdp4_get_frame_format(struct drm_framebuffer *fb) +{ + bool is_tile = false; + + if (fb->modifier == DRM_FORMAT_MOD_SAMSUNG_64_32_TILE) + is_tile = true; + + if (fb->format->format == DRM_FORMAT_NV12 && is_tile) + return FRAME_TILE_YCBCR_420; + + return FRAME_LINEAR; +} + +static void mdp4_plane_set_scanout(struct drm_plane *plane, + struct drm_framebuffer *fb); +static int mdp4_plane_mode_set(struct drm_plane *plane, + struct drm_crtc *crtc, struct drm_framebuffer *fb, + int crtc_x, int crtc_y, + unsigned int crtc_w, unsigned int crtc_h, + uint32_t src_x, uint32_t src_y, + uint32_t src_w, uint32_t src_h); + +static struct mdp4_kms *get_kms(struct drm_plane *plane) +{ + struct msm_drm_private *priv = plane->dev->dev_private; + return to_mdp4_kms(to_mdp_kms(priv->kms)); +} + +/* helper to install properties which are common to planes and crtcs */ +static void mdp4_plane_install_properties(struct drm_plane *plane, + struct drm_mode_object *obj) +{ + // XXX +} + +static int mdp4_plane_set_property(struct drm_plane *plane, + struct drm_property *property, uint64_t val) +{ + // XXX + return -EINVAL; +} + +static const struct drm_plane_funcs mdp4_plane_funcs = { + .update_plane = drm_atomic_helper_update_plane, + .disable_plane = drm_atomic_helper_disable_plane, + .set_property = mdp4_plane_set_property, + .reset = drm_atomic_helper_plane_reset, + .atomic_duplicate_state = drm_atomic_helper_plane_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_plane_destroy_state, +}; + +static int mdp4_plane_prepare_fb(struct drm_plane *plane, + struct drm_plane_state *new_state) +{ + if (!new_state->fb) + return 0; + + drm_gem_plane_helper_prepare_fb(plane, new_state); + + return msm_framebuffer_prepare(new_state->fb, false); +} + +static void mdp4_plane_cleanup_fb(struct drm_plane *plane, + struct drm_plane_state *old_state) +{ + struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane); + struct drm_framebuffer *fb = old_state->fb; + + if (!fb) + return; + + DBG("%s: cleanup: FB[%u]", mdp4_plane->name, fb->base.id); + msm_framebuffer_cleanup(fb, false); +} + + +static int mdp4_plane_atomic_check(struct drm_plane *plane, + struct drm_atomic_commit *state) +{ + return 0; +} + +static void mdp4_plane_atomic_update(struct drm_plane *plane, + struct drm_atomic_commit *state) +{ + struct drm_plane_state *new_state = drm_atomic_get_new_plane_state(state, + plane); + int ret; + + ret = mdp4_plane_mode_set(plane, + new_state->crtc, new_state->fb, + new_state->crtc_x, new_state->crtc_y, + new_state->crtc_w, new_state->crtc_h, + new_state->src_x, new_state->src_y, + new_state->src_w, new_state->src_h); + /* atomic_check should have ensured that this doesn't fail */ + WARN_ON(ret < 0); +} + +static const struct drm_plane_helper_funcs mdp4_plane_helper_funcs = { + .prepare_fb = mdp4_plane_prepare_fb, + .cleanup_fb = mdp4_plane_cleanup_fb, + .atomic_check = mdp4_plane_atomic_check, + .atomic_update = mdp4_plane_atomic_update, +}; + +static void mdp4_plane_set_scanout(struct drm_plane *plane, + struct drm_framebuffer *fb) +{ + struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane); + struct mdp4_kms *mdp4_kms = get_kms(plane); + enum mdp4_pipe pipe = mdp4_plane->pipe; + + mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRC_STRIDE_A(pipe), + MDP4_PIPE_SRC_STRIDE_A_P0(fb->pitches[0]) | + MDP4_PIPE_SRC_STRIDE_A_P1(fb->pitches[1])); + + mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRC_STRIDE_B(pipe), + MDP4_PIPE_SRC_STRIDE_B_P2(fb->pitches[2]) | + MDP4_PIPE_SRC_STRIDE_B_P3(fb->pitches[3])); + + mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRCP0_BASE(pipe), + msm_framebuffer_iova(fb, 0)); + mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRCP1_BASE(pipe), + msm_framebuffer_iova(fb, 1)); + mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRCP2_BASE(pipe), + msm_framebuffer_iova(fb, 2)); + mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRCP3_BASE(pipe), + msm_framebuffer_iova(fb, 3)); +} + +static void mdp4_write_csc_config(struct mdp4_kms *mdp4_kms, + enum mdp4_pipe pipe, struct csc_cfg *csc) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(csc->matrix); i++) { + mdp4_write(mdp4_kms, REG_MDP4_PIPE_CSC_MV(pipe, i), + csc->matrix[i]); + } + + for (i = 0; i < ARRAY_SIZE(csc->post_bias) ; i++) { + mdp4_write(mdp4_kms, REG_MDP4_PIPE_CSC_PRE_BV(pipe, i), + csc->pre_bias[i]); + + mdp4_write(mdp4_kms, REG_MDP4_PIPE_CSC_POST_BV(pipe, i), + csc->post_bias[i]); + } + + for (i = 0; i < ARRAY_SIZE(csc->post_clamp) ; i++) { + mdp4_write(mdp4_kms, REG_MDP4_PIPE_CSC_PRE_LV(pipe, i), + csc->pre_clamp[i]); + + mdp4_write(mdp4_kms, REG_MDP4_PIPE_CSC_POST_LV(pipe, i), + csc->post_clamp[i]); + } +} + +#define MDP4_VG_PHASE_STEP_DEFAULT 0x20000000 + +static int mdp4_plane_mode_set(struct drm_plane *plane, + struct drm_crtc *crtc, struct drm_framebuffer *fb, + int crtc_x, int crtc_y, + unsigned int crtc_w, unsigned int crtc_h, + uint32_t src_x, uint32_t src_y, + uint32_t src_w, uint32_t src_h) +{ + struct drm_device *dev = plane->dev; + struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane); + struct mdp4_kms *mdp4_kms = get_kms(plane); + enum mdp4_pipe pipe = mdp4_plane->pipe; + const struct msm_format *format; + uint32_t op_mode = 0; + uint32_t phasex_step = MDP4_VG_PHASE_STEP_DEFAULT; + uint32_t phasey_step = MDP4_VG_PHASE_STEP_DEFAULT; + enum mdp4_frame_format frame_type; + + if (!(crtc && fb)) { + DBG("%s: disabled!", mdp4_plane->name); + return 0; + } + + frame_type = mdp4_get_frame_format(fb); + + /* src values are in Q16 fixed point, convert to integer: */ + src_x = src_x >> 16; + src_y = src_y >> 16; + src_w = src_w >> 16; + src_h = src_h >> 16; + + DBG("%s: FB[%u] %u,%u,%u,%u -> CRTC[%u] %d,%d,%u,%u", mdp4_plane->name, + fb->base.id, src_x, src_y, src_w, src_h, + crtc->base.id, crtc_x, crtc_y, crtc_w, crtc_h); + + format = msm_framebuffer_format(fb); + + if (src_w > (crtc_w * DOWN_SCALE_MAX)) { + DRM_DEV_ERROR(dev->dev, "Width down scaling exceeds limits!\n"); + return -ERANGE; + } + + if (src_h > (crtc_h * DOWN_SCALE_MAX)) { + DRM_DEV_ERROR(dev->dev, "Height down scaling exceeds limits!\n"); + return -ERANGE; + } + + if (crtc_w > (src_w * UP_SCALE_MAX)) { + DRM_DEV_ERROR(dev->dev, "Width up scaling exceeds limits!\n"); + return -ERANGE; + } + + if (crtc_h > (src_h * UP_SCALE_MAX)) { + DRM_DEV_ERROR(dev->dev, "Height up scaling exceeds limits!\n"); + return -ERANGE; + } + + if (src_w != crtc_w) { + uint32_t sel_unit = SCALE_FIR; + op_mode |= MDP4_PIPE_OP_MODE_SCALEX_EN; + + if (MSM_FORMAT_IS_YUV(format)) { + if (crtc_w > src_w) + sel_unit = SCALE_PIXEL_RPT; + else if (crtc_w <= (src_w / 4)) + sel_unit = SCALE_MN_PHASE; + + op_mode |= MDP4_PIPE_OP_MODE_SCALEX_UNIT_SEL(sel_unit); + phasex_step = mult_frac(MDP4_VG_PHASE_STEP_DEFAULT, + src_w, crtc_w); + } + } + + if (src_h != crtc_h) { + uint32_t sel_unit = SCALE_FIR; + op_mode |= MDP4_PIPE_OP_MODE_SCALEY_EN; + + if (MSM_FORMAT_IS_YUV(format)) { + + if (crtc_h > src_h) + sel_unit = SCALE_PIXEL_RPT; + else if (crtc_h <= (src_h / 4)) + sel_unit = SCALE_MN_PHASE; + + op_mode |= MDP4_PIPE_OP_MODE_SCALEY_UNIT_SEL(sel_unit); + phasey_step = mult_frac(MDP4_VG_PHASE_STEP_DEFAULT, + src_h, crtc_h); + } + } + + mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRC_SIZE(pipe), + MDP4_PIPE_SRC_SIZE_WIDTH(src_w) | + MDP4_PIPE_SRC_SIZE_HEIGHT(src_h)); + + mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRC_XY(pipe), + MDP4_PIPE_SRC_XY_X(src_x) | + MDP4_PIPE_SRC_XY_Y(src_y)); + + mdp4_write(mdp4_kms, REG_MDP4_PIPE_DST_SIZE(pipe), + MDP4_PIPE_DST_SIZE_WIDTH(crtc_w) | + MDP4_PIPE_DST_SIZE_HEIGHT(crtc_h)); + + mdp4_write(mdp4_kms, REG_MDP4_PIPE_DST_XY(pipe), + MDP4_PIPE_DST_XY_X(crtc_x) | + MDP4_PIPE_DST_XY_Y(crtc_y)); + + mdp4_plane_set_scanout(plane, fb); + + mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRC_FORMAT(pipe), + MDP4_PIPE_SRC_FORMAT_A_BPC(format->bpc_a) | + MDP4_PIPE_SRC_FORMAT_R_BPC(format->bpc_r_cr) | + MDP4_PIPE_SRC_FORMAT_G_BPC(format->bpc_g_y) | + MDP4_PIPE_SRC_FORMAT_B_BPC(format->bpc_b_cb) | + COND(format->alpha_enable, MDP4_PIPE_SRC_FORMAT_ALPHA_ENABLE) | + MDP4_PIPE_SRC_FORMAT_CPP(format->bpp - 1) | + MDP4_PIPE_SRC_FORMAT_UNPACK_COUNT(format->unpack_count - 1) | + MDP4_PIPE_SRC_FORMAT_FETCH_PLANES(format->fetch_type) | + MDP4_PIPE_SRC_FORMAT_CHROMA_SAMP(format->chroma_sample) | + MDP4_PIPE_SRC_FORMAT_FRAME_FORMAT(frame_type) | + COND(format->flags & MSM_FORMAT_FLAG_UNPACK_TIGHT, + MDP4_PIPE_SRC_FORMAT_UNPACK_TIGHT)); + + mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRC_UNPACK(pipe), + MDP4_PIPE_SRC_UNPACK_ELEM0(format->element[0]) | + MDP4_PIPE_SRC_UNPACK_ELEM1(format->element[1]) | + MDP4_PIPE_SRC_UNPACK_ELEM2(format->element[2]) | + MDP4_PIPE_SRC_UNPACK_ELEM3(format->element[3])); + + if (MSM_FORMAT_IS_YUV(format)) { + struct csc_cfg *csc = mdp_get_default_csc_cfg(CSC_YUV2RGB); + + op_mode |= MDP4_PIPE_OP_MODE_SRC_YCBCR; + op_mode |= MDP4_PIPE_OP_MODE_CSC_EN; + mdp4_write_csc_config(mdp4_kms, pipe, csc); + } + + mdp4_write(mdp4_kms, REG_MDP4_PIPE_OP_MODE(pipe), op_mode); + mdp4_write(mdp4_kms, REG_MDP4_PIPE_PHASEX_STEP(pipe), phasex_step); + mdp4_write(mdp4_kms, REG_MDP4_PIPE_PHASEY_STEP(pipe), phasey_step); + + if (frame_type != FRAME_LINEAR) + mdp4_write(mdp4_kms, REG_MDP4_PIPE_SSTILE_FRAME_SIZE(pipe), + MDP4_PIPE_SSTILE_FRAME_SIZE_WIDTH(src_w) | + MDP4_PIPE_SSTILE_FRAME_SIZE_HEIGHT(src_h)); + + return 0; +} + +static const char *pipe_names[] = { + "VG1", "VG2", + "RGB1", "RGB2", "RGB3", + "VG3", "VG4", +}; + +enum mdp4_pipe mdp4_plane_pipe(struct drm_plane *plane) +{ + struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane); + return mdp4_plane->pipe; +} + +static const uint64_t supported_format_modifiers[] = { + DRM_FORMAT_MOD_SAMSUNG_64_32_TILE, + DRM_FORMAT_MOD_LINEAR, + DRM_FORMAT_MOD_INVALID +}; + +static const uint32_t mdp4_rgb_formats[] = { + DRM_FORMAT_ARGB8888, + DRM_FORMAT_ABGR8888, + DRM_FORMAT_RGBA8888, + DRM_FORMAT_BGRA8888, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_RGBX8888, + DRM_FORMAT_BGRX8888, + DRM_FORMAT_RGB888, + DRM_FORMAT_BGR888, + DRM_FORMAT_RGB565, + DRM_FORMAT_BGR565, +}; + +static const uint32_t mdp4_rgb_yuv_formats[] = { + DRM_FORMAT_ARGB8888, + DRM_FORMAT_ABGR8888, + DRM_FORMAT_RGBA8888, + DRM_FORMAT_BGRA8888, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_RGBX8888, + DRM_FORMAT_BGRX8888, + DRM_FORMAT_RGB888, + DRM_FORMAT_BGR888, + DRM_FORMAT_RGB565, + DRM_FORMAT_BGR565, + + DRM_FORMAT_NV12, + DRM_FORMAT_NV21, + DRM_FORMAT_NV16, + DRM_FORMAT_NV61, + DRM_FORMAT_VYUY, + DRM_FORMAT_UYVY, + DRM_FORMAT_YUYV, + DRM_FORMAT_YVYU, + DRM_FORMAT_YUV420, + DRM_FORMAT_YVU420, +}; + +/* initialize plane */ +struct drm_plane *mdp4_plane_init(struct drm_device *dev, + enum mdp4_pipe pipe_id, bool private_plane) +{ + struct drm_plane *plane = NULL; + struct mdp4_plane *mdp4_plane; + enum drm_plane_type type; + uint32_t pipe_caps; + const uint32_t *formats; + size_t nformats; + + type = private_plane ? DRM_PLANE_TYPE_PRIMARY : DRM_PLANE_TYPE_OVERLAY; + + pipe_caps = mdp4_pipe_caps(pipe_id); + if (pipe_supports_yuv(pipe_caps)) { + formats = mdp4_rgb_yuv_formats; + nformats = ARRAY_SIZE(mdp4_rgb_yuv_formats); + } else { + formats = mdp4_rgb_formats; + nformats = ARRAY_SIZE(mdp4_rgb_formats); + } + + mdp4_plane = drmm_universal_plane_alloc(dev, struct mdp4_plane, base, + 0xff, &mdp4_plane_funcs, + formats, nformats, + supported_format_modifiers, + type, NULL); + if (IS_ERR(mdp4_plane)) + return ERR_CAST(mdp4_plane); + + plane = &mdp4_plane->base; + + mdp4_plane->pipe = pipe_id; + mdp4_plane->name = pipe_names[pipe_id]; + + drm_plane_helper_add(plane, &mdp4_plane_helper_funcs); + + mdp4_plane_install_properties(plane, &plane->base); + + drm_plane_enable_fb_damage_clips(plane); + + return plane; +} diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c new file mode 100644 index 000000000..7c91fc191 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c @@ -0,0 +1,1084 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2014-2015 The Linux Foundation. All rights reserved. + */ + +#include "mdp5_kms.h" +#include "mdp5_cfg.h" + +struct mdp5_cfg_handler { + int revision; + struct mdp5_cfg config; +}; + +/* mdp5_cfg must be exposed (used in mdp5.xml.h) */ +const struct mdp5_cfg_hw *mdp5_cfg = NULL; + +static const struct mdp5_cfg_hw msm8x26_config = { + .name = "msm8x26", + .mdp = { + .count = 1, + .caps = MDP_CAP_SMP | + 0, + }, + .smp = { + .mmb_count = 7, + .mmb_size = 4096, + .clients = { + [SSPP_VIG0] = 1, + [SSPP_DMA0] = 4, + [SSPP_RGB0] = 7, + }, + }, + .ctl = { + .count = 2, + .base = { 0x00500, 0x00600 }, + .flush_hw_mask = 0x0003ffff, + }, + .pipe_vig = { + .count = 1, + .base = { 0x01100 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | + MDP_PIPE_CAP_CSC | + 0, + }, + .pipe_rgb = { + .count = 1, + .base = { 0x01d00 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | + 0, + }, + .pipe_dma = { + .count = 1, + .base = { 0x02900 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + 0, + }, + .lm = { + .count = 2, + .base = { 0x03100, 0x03d00 }, + .instances = { + { .id = 0, .pp = 0, .dspp = 0, + .caps = MDP_LM_CAP_DISPLAY, }, + { .id = 1, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB }, + }, + .nb_stages = 2, + .max_width = 2048, + .max_height = 0xFFFF, + }, + .dspp = { + .count = 1, + .base = { 0x04500 }, + }, + .pp = { + .count = 1, + .base = { 0x21a00 }, + }, + .intf = { + .base = { 0x00000, 0x21200 }, + .connect = { + [0] = INTF_DISABLED, + [1] = INTF_DSI, + }, + }, + .perf = { + .ab_inefficiency = 100, + .ib_inefficiency = 200, + .clk_inefficiency = 125 + }, + .max_clk = 200000000, +}; + +static const struct mdp5_cfg_hw msm8x74_config = { + .name = "msm8x74", + .mdp = { + .count = 1, + .caps = MDP_CAP_SMP | + 0, + }, + .smp = { + .mmb_count = 22, + .mmb_size = 4096, + .clients = { + [SSPP_VIG0] = 1, [SSPP_VIG1] = 4, [SSPP_VIG2] = 7, + [SSPP_DMA0] = 10, [SSPP_DMA1] = 13, + [SSPP_RGB0] = 16, [SSPP_RGB1] = 17, [SSPP_RGB2] = 18, + }, + }, + .ctl = { + .count = 5, + .base = { 0x00500, 0x00600, 0x00700, 0x00800, 0x00900 }, + .flush_hw_mask = 0x0003ffff, + }, + .pipe_vig = { + .count = 3, + .base = { 0x01100, 0x01500, 0x01900 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | MDP_PIPE_CAP_CSC | + MDP_PIPE_CAP_DECIMATION, + }, + .pipe_rgb = { + .count = 3, + .base = { 0x01d00, 0x02100, 0x02500 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | MDP_PIPE_CAP_DECIMATION, + }, + .pipe_dma = { + .count = 2, + .base = { 0x02900, 0x02d00 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP, + }, + .lm = { + .count = 5, + .base = { 0x03100, 0x03500, 0x03900, 0x03d00, 0x04100 }, + .instances = { + { .id = 0, .pp = 0, .dspp = 0, + .caps = MDP_LM_CAP_DISPLAY, }, + { .id = 1, .pp = 1, .dspp = 1, + .caps = MDP_LM_CAP_DISPLAY, }, + { .id = 2, .pp = 2, .dspp = 2, + .caps = MDP_LM_CAP_DISPLAY, }, + { .id = 3, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB, }, + { .id = 4, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB, }, + }, + .nb_stages = 5, + .max_width = 2048, + .max_height = 0xFFFF, + }, + .dspp = { + .count = 3, + .base = { 0x04500, 0x04900, 0x04d00 }, + }, + .ad = { + .count = 2, + .base = { 0x13000, 0x13200 }, + }, + .pp = { + .count = 3, + .base = { 0x12c00, 0x12d00, 0x12e00 }, + }, + .intf = { + .base = { 0x12400, 0x12600, 0x12800, 0x12a00 }, + .connect = { + [0] = INTF_eDP, + [1] = INTF_DSI, + [2] = INTF_DSI, + [3] = INTF_HDMI, + }, + }, + .perf = { + .ab_inefficiency = 200, + .ib_inefficiency = 120, + .clk_inefficiency = 125 + }, + .max_clk = 320000000, +}; + +static const struct mdp5_cfg_hw apq8084_config = { + .name = "apq8084", + .mdp = { + .count = 1, + .caps = MDP_CAP_SMP | + MDP_CAP_SRC_SPLIT | + 0, + }, + .smp = { + .mmb_count = 44, + .mmb_size = 8192, + .clients = { + [SSPP_VIG0] = 1, [SSPP_VIG1] = 4, + [SSPP_VIG2] = 7, [SSPP_VIG3] = 19, + [SSPP_DMA0] = 10, [SSPP_DMA1] = 13, + [SSPP_RGB0] = 16, [SSPP_RGB1] = 17, + [SSPP_RGB2] = 18, [SSPP_RGB3] = 22, + }, + .reserved_state[0] = GENMASK(7, 0), /* first 8 MMBs */ + .reserved = { + /* Two SMP blocks are statically tied to RGB pipes: */ + [16] = 2, [17] = 2, [18] = 2, [22] = 2, + }, + }, + .ctl = { + .count = 5, + .base = { 0x00500, 0x00600, 0x00700, 0x00800, 0x00900 }, + .flush_hw_mask = 0x003fffff, + }, + .pipe_vig = { + .count = 4, + .base = { 0x01100, 0x01500, 0x01900, 0x01d00 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | MDP_PIPE_CAP_CSC | + MDP_PIPE_CAP_DECIMATION, + }, + .pipe_rgb = { + .count = 4, + .base = { 0x02100, 0x02500, 0x02900, 0x02d00 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | MDP_PIPE_CAP_DECIMATION, + }, + .pipe_dma = { + .count = 2, + .base = { 0x03100, 0x03500 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP, + }, + .lm = { + .count = 6, + .base = { 0x03900, 0x03d00, 0x04100, 0x04500, 0x04900, 0x04d00 }, + .instances = { + { .id = 0, .pp = 0, .dspp = 0, + .caps = MDP_LM_CAP_DISPLAY | + MDP_LM_CAP_PAIR, }, + { .id = 1, .pp = 1, .dspp = 1, + .caps = MDP_LM_CAP_DISPLAY, }, + { .id = 2, .pp = 2, .dspp = 2, + .caps = MDP_LM_CAP_DISPLAY | + MDP_LM_CAP_PAIR, }, + { .id = 3, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB, }, + { .id = 4, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB, }, + { .id = 5, .pp = 3, .dspp = 3, + .caps = MDP_LM_CAP_DISPLAY, }, + }, + .nb_stages = 5, + .max_width = 2048, + .max_height = 0xFFFF, + }, + .dspp = { + .count = 4, + .base = { 0x05100, 0x05500, 0x05900, 0x05d00 }, + + }, + .ad = { + .count = 3, + .base = { 0x13400, 0x13600, 0x13800 }, + }, + .pp = { + .count = 4, + .base = { 0x12e00, 0x12f00, 0x13000, 0x13100 }, + }, + .intf = { + .base = { 0x12400, 0x12600, 0x12800, 0x12a00, 0x12c00 }, + .connect = { + [0] = INTF_eDP, + [1] = INTF_DSI, + [2] = INTF_DSI, + [3] = INTF_HDMI, + }, + }, + .perf = { + .ab_inefficiency = 200, + .ib_inefficiency = 120, + .clk_inefficiency = 105 + }, + .max_clk = 320000000, +}; + +static const struct mdp5_cfg_hw msm8x16_config = { + .name = "msm8x16", + .mdp = { + .count = 1, + .base = { 0x0 }, + .caps = MDP_CAP_SMP | + 0, + }, + .smp = { + .mmb_count = 8, + .mmb_size = 8192, + .clients = { + [SSPP_VIG0] = 1, [SSPP_DMA0] = 4, + [SSPP_RGB0] = 7, [SSPP_RGB1] = 8, + }, + }, + .ctl = { + .count = 5, + .base = { 0x01000, 0x01200, 0x01400, 0x01600, 0x01800 }, + .flush_hw_mask = 0x4003ffff, + }, + .pipe_vig = { + .count = 1, + .base = { 0x04000 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | MDP_PIPE_CAP_CSC | + MDP_PIPE_CAP_DECIMATION, + }, + .pipe_rgb = { + .count = 2, + .base = { 0x14000, 0x16000 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_DECIMATION, + }, + .pipe_dma = { + .count = 1, + .base = { 0x24000 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP, + }, + .lm = { + .count = 2, /* LM0 and LM3 */ + .base = { 0x44000, 0x47000 }, + .instances = { + { .id = 0, .pp = 0, .dspp = 0, + .caps = MDP_LM_CAP_DISPLAY, }, + { .id = 3, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB }, + }, + .nb_stages = 8, + .max_width = 2048, + .max_height = 0xFFFF, + }, + .dspp = { + .count = 1, + .base = { 0x54000 }, + + }, + .intf = { + .base = { 0x00000, 0x6a800 }, + .connect = { + [0] = INTF_DISABLED, + [1] = INTF_DSI, + }, + }, + .perf = { + .ab_inefficiency = 100, + .ib_inefficiency = 200, + .clk_inefficiency = 105 + }, + .max_clk = 320000000, +}; + +static const struct mdp5_cfg_hw msm8x36_config = { + .name = "msm8x36", + .mdp = { + .count = 1, + .base = { 0x0 }, + .caps = MDP_CAP_SMP | + 0, + }, + .smp = { + .mmb_count = 8, + .mmb_size = 10240, + .clients = { + [SSPP_VIG0] = 1, [SSPP_DMA0] = 4, + [SSPP_RGB0] = 7, [SSPP_RGB1] = 8, + }, + }, + .ctl = { + .count = 3, + .base = { 0x01000, 0x01200, 0x01400 }, + .flush_hw_mask = 0x4003ffff, + }, + .pipe_vig = { + .count = 1, + .base = { 0x04000 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | MDP_PIPE_CAP_CSC | + MDP_PIPE_CAP_DECIMATION, + }, + .pipe_rgb = { + .count = 2, + .base = { 0x14000, 0x16000 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_DECIMATION, + }, + .pipe_dma = { + .count = 1, + .base = { 0x24000 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP, + }, + .lm = { + .count = 2, + .base = { 0x44000, 0x47000 }, + .instances = { + { .id = 0, .pp = 0, .dspp = 0, + .caps = MDP_LM_CAP_DISPLAY, }, + { .id = 1, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB, }, + }, + .nb_stages = 8, + .max_width = 2560, + .max_height = 0xFFFF, + }, + .pp = { + .count = 1, + .base = { 0x70000 }, + }, + .ad = { + .count = 1, + .base = { 0x78000 }, + }, + .dspp = { + .count = 1, + .base = { 0x54000 }, + }, + .intf = { + .base = { 0x00000, 0x6a800, 0x6b000 }, + .connect = { + [0] = INTF_DISABLED, + [1] = INTF_DSI, + [2] = INTF_DSI, + }, + }, + .perf = { + .ab_inefficiency = 100, + .ib_inefficiency = 200, + .clk_inefficiency = 105 + }, + .max_clk = 366670000, +}; + +static const struct mdp5_cfg_hw msm8x94_config = { + .name = "msm8x94", + .mdp = { + .count = 1, + .caps = MDP_CAP_SMP | + MDP_CAP_SRC_SPLIT | + 0, + }, + .smp = { + .mmb_count = 44, + .mmb_size = 8192, + .clients = { + [SSPP_VIG0] = 1, [SSPP_VIG1] = 4, + [SSPP_VIG2] = 7, [SSPP_VIG3] = 19, + [SSPP_DMA0] = 10, [SSPP_DMA1] = 13, + [SSPP_RGB0] = 16, [SSPP_RGB1] = 17, + [SSPP_RGB2] = 18, [SSPP_RGB3] = 22, + }, + .reserved_state[0] = GENMASK(23, 0), /* first 24 MMBs */ + .reserved = { + [1] = 1, [4] = 1, [7] = 1, [19] = 1, + [16] = 5, [17] = 5, [18] = 5, [22] = 5, + }, + }, + .ctl = { + .count = 5, + .base = { 0x01000, 0x01200, 0x01400, 0x01600, 0x01800 }, + .flush_hw_mask = 0xf0ffffff, + }, + .pipe_vig = { + .count = 4, + .base = { 0x04000, 0x06000, 0x08000, 0x0a000 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | MDP_PIPE_CAP_CSC | + MDP_PIPE_CAP_DECIMATION, + }, + .pipe_rgb = { + .count = 4, + .base = { 0x14000, 0x16000, 0x18000, 0x1a000 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | MDP_PIPE_CAP_DECIMATION, + }, + .pipe_dma = { + .count = 2, + .base = { 0x24000, 0x26000 }, + .caps = MDP_PIPE_CAP_HFLIP | MDP_PIPE_CAP_VFLIP, + }, + .lm = { + .count = 6, + .base = { 0x44000, 0x45000, 0x46000, 0x47000, 0x48000, 0x49000 }, + .instances = { + { .id = 0, .pp = 0, .dspp = 0, + .caps = MDP_LM_CAP_DISPLAY | + MDP_LM_CAP_PAIR, }, + { .id = 1, .pp = 1, .dspp = 1, + .caps = MDP_LM_CAP_DISPLAY, }, + { .id = 2, .pp = 2, .dspp = 2, + .caps = MDP_LM_CAP_DISPLAY | + MDP_LM_CAP_PAIR, }, + { .id = 3, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB, }, + { .id = 4, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB, }, + { .id = 5, .pp = 3, .dspp = 3, + .caps = MDP_LM_CAP_DISPLAY, }, + }, + .nb_stages = 8, + .max_width = 2048, + .max_height = 0xFFFF, + }, + .dspp = { + .count = 4, + .base = { 0x54000, 0x56000, 0x58000, 0x5a000 }, + + }, + .ad = { + .count = 3, + .base = { 0x78000, 0x78800, 0x79000 }, + }, + .pp = { + .count = 4, + .base = { 0x70000, 0x70800, 0x71000, 0x71800 }, + }, + .intf = { + .base = { 0x6a000, 0x6a800, 0x6b000, 0x6b800, 0x6c000 }, + .connect = { + [0] = INTF_DISABLED, + [1] = INTF_DSI, + [2] = INTF_DSI, + [3] = INTF_HDMI, + }, + }, + .perf = { + .ab_inefficiency = 100, + .ib_inefficiency = 100, + .clk_inefficiency = 105 + }, + .max_clk = 400000000, +}; + +static const struct mdp5_cfg_hw msm8x96_config = { + .name = "msm8x96", + .mdp = { + .count = 1, + .caps = MDP_CAP_DSC | + MDP_CAP_CDM | + MDP_CAP_SRC_SPLIT | + 0, + }, + .ctl = { + .count = 5, + .base = { 0x01000, 0x01200, 0x01400, 0x01600, 0x01800 }, + .flush_hw_mask = 0xf4ffffff, + }, + .pipe_vig = { + .count = 4, + .base = { 0x04000, 0x06000, 0x08000, 0x0a000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | + MDP_PIPE_CAP_CSC | + MDP_PIPE_CAP_DECIMATION | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_rgb = { + .count = 4, + .base = { 0x14000, 0x16000, 0x18000, 0x1a000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | + MDP_PIPE_CAP_DECIMATION | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_dma = { + .count = 2, + .base = { 0x24000, 0x26000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_cursor = { + .count = 2, + .base = { 0x34000, 0x36000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SW_PIX_EXT | + MDP_PIPE_CAP_CURSOR | + 0, + }, + + .lm = { + .count = 6, + .base = { 0x44000, 0x45000, 0x46000, 0x47000, 0x48000, 0x49000 }, + .instances = { + { .id = 0, .pp = 0, .dspp = 0, + .caps = MDP_LM_CAP_DISPLAY | + MDP_LM_CAP_PAIR, }, + { .id = 1, .pp = 1, .dspp = 1, + .caps = MDP_LM_CAP_DISPLAY, }, + { .id = 2, .pp = 2, .dspp = -1, + .caps = MDP_LM_CAP_DISPLAY | + MDP_LM_CAP_PAIR, }, + { .id = 3, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB, }, + { .id = 4, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB, }, + { .id = 5, .pp = 3, .dspp = -1, + .caps = MDP_LM_CAP_DISPLAY, }, + }, + .nb_stages = 8, + .max_width = 2560, + .max_height = 0xFFFF, + }, + .dspp = { + .count = 2, + .base = { 0x54000, 0x56000 }, + }, + .ad = { + .count = 3, + .base = { 0x78000, 0x78800, 0x79000 }, + }, + .pp = { + .count = 4, + .base = { 0x70000, 0x70800, 0x71000, 0x71800 }, + }, + .cdm = { + .count = 1, + .base = { 0x79200 }, + }, + .dsc = { + .count = 2, + .base = { 0x80000, 0x80400 }, + }, + .intf = { + .base = { 0x6a000, 0x6a800, 0x6b000, 0x6b800, 0x6c000 }, + .connect = { + [0] = INTF_DISABLED, + [1] = INTF_DSI, + [2] = INTF_DSI, + [3] = INTF_HDMI, + }, + }, + .perf = { + .ab_inefficiency = 100, + .ib_inefficiency = 200, + .clk_inefficiency = 105 + }, + .max_clk = 412500000, +}; + +static const struct mdp5_cfg_hw msm8x76_config = { + .name = "msm8x76", + .mdp = { + .count = 1, + .caps = MDP_CAP_SMP | + MDP_CAP_DSC | + MDP_CAP_SRC_SPLIT | + 0, + }, + .ctl = { + .count = 3, + .base = { 0x01000, 0x01200, 0x01400 }, + .flush_hw_mask = 0xffffffff, + }, + .smp = { + .mmb_count = 10, + .mmb_size = 10240, + .clients = { + [SSPP_VIG0] = 1, [SSPP_VIG1] = 9, + [SSPP_DMA0] = 4, + [SSPP_RGB0] = 7, [SSPP_RGB1] = 8, + }, + }, + .pipe_vig = { + .count = 2, + .base = { 0x04000, 0x06000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | + MDP_PIPE_CAP_CSC | + MDP_PIPE_CAP_DECIMATION | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_rgb = { + .count = 2, + .base = { 0x14000, 0x16000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_DECIMATION | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_dma = { + .count = 1, + .base = { 0x24000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_cursor = { + .count = 1, + .base = { 0x440DC }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SW_PIX_EXT | + MDP_PIPE_CAP_CURSOR | + 0, + }, + .lm = { + .count = 2, + .base = { 0x44000, 0x45000 }, + .instances = { + { .id = 0, .pp = 0, .dspp = 0, + .caps = MDP_LM_CAP_DISPLAY, }, + { .id = 1, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB }, + }, + .nb_stages = 8, + .max_width = 2560, + .max_height = 0xFFFF, + }, + .dspp = { + .count = 1, + .base = { 0x54000 }, + + }, + .pp = { + .count = 3, + .base = { 0x70000, 0x70800, 0x72000 }, + }, + .dsc = { + .count = 2, + .base = { 0x80000, 0x80400 }, + }, + .intf = { + .base = { 0x6a000, 0x6a800, 0x6b000 }, + .connect = { + [0] = INTF_DISABLED, + [1] = INTF_DSI, + [2] = INTF_DSI, + }, + }, + .max_clk = 360000000, +}; + +static const struct mdp5_cfg_hw msm8x53_config = { + .name = "msm8x53", + .mdp = { + .count = 1, + .caps = MDP_CAP_CDM, + }, + .ctl = { + .count = 3, + .base = { 0x01000, 0x01200, 0x01400 }, + .flush_hw_mask = 0xffffffff, + }, + .pipe_vig = { + .count = 1, + .base = { 0x04000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | + MDP_PIPE_CAP_CSC | + MDP_PIPE_CAP_DECIMATION | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_rgb = { + .count = 2, + .base = { 0x14000, 0x16000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_DECIMATION | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_dma = { + .count = 1, + .base = { 0x24000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_cursor = { + .count = 1, + .base = { 0x34000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SW_PIX_EXT | + MDP_PIPE_CAP_CURSOR | + 0, + }, + + .lm = { + .count = 3, + .base = { 0x44000, 0x45000 }, + .instances = { + { .id = 0, .pp = 0, .dspp = 0, + .caps = MDP_LM_CAP_DISPLAY | + MDP_LM_CAP_PAIR }, + { .id = 1, .pp = 1, .dspp = -1, + .caps = MDP_LM_CAP_DISPLAY }, + }, + .nb_stages = 5, + .max_width = 2048, + .max_height = 0xFFFF, + }, + .dspp = { + .count = 1, + .base = { 0x54000 }, + + }, + .pp = { + .count = 2, + .base = { 0x70000, 0x70800 }, + }, + .cdm = { + .count = 1, + .base = { 0x79200 }, + }, + .intf = { + .base = { 0x6a000, 0x6a800, 0x6b000 }, + .connect = { + [0] = INTF_DISABLED, + [1] = INTF_DSI, + [2] = INTF_DSI, + }, + }, + .perf = { + .ab_inefficiency = 100, + .ib_inefficiency = 200, + .clk_inefficiency = 105 + }, + .max_clk = 400000000, +}; + +static const struct mdp5_cfg_hw msm8917_config = { + .name = "msm8917", + .mdp = { + .count = 1, + .caps = MDP_CAP_CDM, + }, + .ctl = { + .count = 3, + .base = { 0x01000, 0x01200, 0x01400 }, + .flush_hw_mask = 0xffffffff, + }, + .pipe_vig = { + .count = 1, + .base = { 0x04000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | + MDP_PIPE_CAP_CSC | + MDP_PIPE_CAP_DECIMATION | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_rgb = { + .count = 2, + .base = { 0x14000, 0x16000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_DECIMATION | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_dma = { + .count = 1, + .base = { 0x24000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_cursor = { + .count = 1, + .base = { 0x34000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SW_PIX_EXT | + MDP_PIPE_CAP_CURSOR | + 0, + }, + + .lm = { + .count = 2, + .base = { 0x44000, 0x45000 }, + .instances = { + { .id = 0, .pp = 0, .dspp = 0, + .caps = MDP_LM_CAP_DISPLAY, }, + { .id = 1, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB }, + }, + .nb_stages = 8, + .max_width = 2048, + .max_height = 0xFFFF, + }, + .dspp = { + .count = 1, + .base = { 0x54000 }, + + }, + .pp = { + .count = 1, + .base = { 0x70000 }, + }, + .cdm = { + .count = 1, + .base = { 0x79200 }, + }, + .intf = { + .base = { 0x6a000, 0x6a800 }, + .connect = { + [0] = INTF_DISABLED, + [1] = INTF_DSI, + }, + }, + .max_clk = 320000000, +}; + +static const struct mdp5_cfg_hw msm8937_config = { + .name = "msm8937", + .mdp = { + .count = 1, + .caps = MDP_CAP_CDM, + }, + .ctl = { + .count = 3, + .base = { 0x01000, 0x01200, 0x01400 }, + .flush_hw_mask = 0xffffffff, + }, + .pipe_vig = { + .count = 1, + .base = { 0x04000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | + MDP_PIPE_CAP_CSC | + MDP_PIPE_CAP_DECIMATION | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_rgb = { + .count = 2, + .base = { 0x14000, 0x16000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_DECIMATION | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_dma = { + .count = 1, + .base = { 0x24000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_cursor = { + .count = 1, + .base = { 0x34000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SW_PIX_EXT | + MDP_PIPE_CAP_CURSOR | + 0, + }, + + .lm = { + .count = 2, + .base = { 0x44000, 0x45000 }, + .instances = { + { .id = 0, .pp = 0, .dspp = 0, + .caps = MDP_LM_CAP_DISPLAY | + MDP_LM_CAP_PAIR }, + { .id = 1, .pp = 1, .dspp = -1, + .caps = MDP_LM_CAP_DISPLAY }, + }, + .nb_stages = 5, + .max_width = 2048, + .max_height = 0xFFFF, + }, + .dspp = { + .count = 1, + .base = { 0x54000 }, + + }, + .pp = { + .count = 2, + .base = { 0x70000, 0x70800 }, + }, + .cdm = { + .count = 1, + .base = { 0x79200 }, + }, + .intf = { + .base = { 0x00000, 0x6a800, 0x6b000 }, + .connect = { + [0] = INTF_DISABLED, + [1] = INTF_DSI, + [2] = INTF_DSI, + }, + }, + .max_clk = 320000000, +}; + +static const struct mdp5_cfg_handler cfg_handlers_v1[] = { + { .revision = 1, .config = { .hw = &msm8x26_config } }, + { .revision = 2, .config = { .hw = &msm8x74_config } }, + { .revision = 3, .config = { .hw = &apq8084_config } }, + { .revision = 6, .config = { .hw = &msm8x16_config } }, + { .revision = 8, .config = { .hw = &msm8x36_config } }, + { .revision = 9, .config = { .hw = &msm8x94_config } }, + { .revision = 7, .config = { .hw = &msm8x96_config } }, + { .revision = 11, .config = { .hw = &msm8x76_config } }, + { .revision = 14, .config = { .hw = &msm8937_config } }, + { .revision = 15, .config = { .hw = &msm8917_config } }, + { .revision = 16, .config = { .hw = &msm8x53_config } }, +}; + +const struct mdp5_cfg_hw *mdp5_cfg_get_hw_config(struct mdp5_cfg_handler *cfg_handler) +{ + return cfg_handler->config.hw; +} + +struct mdp5_cfg *mdp5_cfg_get_config(struct mdp5_cfg_handler *cfg_handler) +{ + return &cfg_handler->config; +} + +int mdp5_cfg_get_hw_rev(struct mdp5_cfg_handler *cfg_handler) +{ + return cfg_handler->revision; +} + +struct mdp5_cfg_handler *mdp5_cfg_init(struct mdp5_kms *mdp5_kms, + uint32_t major, uint32_t minor) +{ + struct drm_device *dev = mdp5_kms->dev; + struct mdp5_cfg_handler *cfg_handler; + const struct mdp5_cfg_handler *cfg_handlers; + int i, num_handlers; + + cfg_handler = devm_kzalloc(dev->dev, sizeof(*cfg_handler), GFP_KERNEL); + if (unlikely(!cfg_handler)) { + return ERR_PTR(-ENOMEM); + } + + switch (major) { + case 1: + cfg_handlers = cfg_handlers_v1; + num_handlers = ARRAY_SIZE(cfg_handlers_v1); + break; + default: + DRM_DEV_ERROR(dev->dev, "unexpected MDP major version: v%d.%d\n", + major, minor); + return ERR_PTR(-ENXIO); + } + + /* only after mdp5_cfg global pointer's init can we access the hw */ + for (i = 0; i < num_handlers; i++) { + if (cfg_handlers[i].revision != minor) + continue; + mdp5_cfg = cfg_handlers[i].config.hw; + + break; + } + if (unlikely(!mdp5_cfg)) { + DRM_DEV_ERROR(dev->dev, "unexpected MDP minor revision: v%d.%d\n", + major, minor); + return ERR_PTR(-ENXIO); + } + + cfg_handler->revision = minor; + cfg_handler->config.hw = mdp5_cfg; + + DBG("MDP5: %s hw config selected", mdp5_cfg->name); + + return cfg_handler; +} diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.h b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.h new file mode 100644 index 000000000..4b988e69f --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.h @@ -0,0 +1,136 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2014 The Linux Foundation. All rights reserved. + */ + +#ifndef __MDP5_CFG_H__ +#define __MDP5_CFG_H__ + +#include "msm_drv.h" + +/* + * mdp5_cfg + * + * This module configures the dynamic offsets used by mdp5.xml.h + * (initialized in mdp5_cfg.c) + */ +extern const struct mdp5_cfg_hw *mdp5_cfg; + +#define MAX_CTL 8 +#define MAX_BASES 8 +#define MAX_SMP_BLOCKS 44 +#define MAX_CLIENTS 32 + +typedef DECLARE_BITMAP(mdp5_smp_state_t, MAX_SMP_BLOCKS); + +#define MDP5_SUB_BLOCK_DEFINITION \ + unsigned int count; \ + uint32_t base[MAX_BASES] + +struct mdp5_sub_block { + MDP5_SUB_BLOCK_DEFINITION; +}; + +struct mdp5_lm_instance { + int id; + int pp; + int dspp; + uint32_t caps; +}; + +struct mdp5_lm_block { + MDP5_SUB_BLOCK_DEFINITION; + struct mdp5_lm_instance instances[MAX_BASES]; + uint32_t nb_stages; /* number of stages per blender */ + uint32_t max_width; /* Maximum output resolution */ + uint32_t max_height; +}; + +struct mdp5_pipe_block { + MDP5_SUB_BLOCK_DEFINITION; + uint32_t caps; /* pipe capabilities */ +}; + +struct mdp5_ctl_block { + MDP5_SUB_BLOCK_DEFINITION; + uint32_t flush_hw_mask; /* FLUSH register's hardware mask */ +}; + +struct mdp5_smp_block { + int mmb_count; /* number of SMP MMBs */ + int mmb_size; /* MMB: size in bytes */ + uint32_t clients[MAX_CLIENTS]; /* SMP port allocation /pipe */ + mdp5_smp_state_t reserved_state;/* SMP MMBs statically allocated */ + uint8_t reserved[MAX_CLIENTS]; /* # of MMBs allocated per client */ +}; + +struct mdp5_mdp_block { + MDP5_SUB_BLOCK_DEFINITION; + uint32_t caps; /* MDP capabilities: MDP_CAP_xxx bits */ +}; + +struct mdp5_wb_instance { + int id; + int lm; +}; + +struct mdp5_wb_block { + MDP5_SUB_BLOCK_DEFINITION; + struct mdp5_wb_instance instances[MAX_BASES]; +}; + +#define MDP5_INTF_NUM_MAX 5 + +struct mdp5_intf_block { + uint32_t base[MAX_BASES]; + u32 connect[MDP5_INTF_NUM_MAX]; /* array of enum mdp5_intf_type */ +}; + +struct mdp5_perf_block { + u32 ab_inefficiency; + u32 ib_inefficiency; + u32 clk_inefficiency; +}; + +struct mdp5_cfg_hw { + char *name; + + struct mdp5_mdp_block mdp; + struct mdp5_smp_block smp; + struct mdp5_ctl_block ctl; + struct mdp5_pipe_block pipe_vig; + struct mdp5_pipe_block pipe_rgb; + struct mdp5_pipe_block pipe_dma; + struct mdp5_pipe_block pipe_cursor; + struct mdp5_lm_block lm; + struct mdp5_sub_block dspp; + struct mdp5_sub_block ad; + struct mdp5_sub_block pp; + struct mdp5_sub_block dsc; + struct mdp5_sub_block cdm; + struct mdp5_wb_block wb; + struct mdp5_intf_block intf; + struct mdp5_perf_block perf; + + uint32_t max_clk; +}; + +struct mdp5_cfg { + const struct mdp5_cfg_hw *hw; +}; + +struct mdp5_kms; +struct mdp5_cfg_handler; + +const struct mdp5_cfg_hw *mdp5_cfg_get_hw_config(struct mdp5_cfg_handler *cfg_hnd); +struct mdp5_cfg *mdp5_cfg_get_config(struct mdp5_cfg_handler *cfg_hnd); +int mdp5_cfg_get_hw_rev(struct mdp5_cfg_handler *cfg_hnd); + +#define mdp5_cfg_intf_is_virtual(intf_type) ({ \ + typeof(intf_type) __val = (intf_type); \ + (__val) >= INTF_VIRTUAL ? true : false; }) + +struct mdp5_cfg_handler *mdp5_cfg_init(struct mdp5_kms *mdp5_kms, + uint32_t major, uint32_t minor); + +#endif /* __MDP5_CFG_H__ */ diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c new file mode 100644 index 000000000..e5662412d --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c @@ -0,0 +1,161 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2015, The Linux Foundation. All rights reserved. + */ + +#include +#include + +#include "mdp5_kms.h" + +#ifdef CONFIG_DRM_MSM_DSI + +static struct mdp5_kms *get_kms(struct drm_encoder *encoder) +{ + struct msm_drm_private *priv = encoder->dev->dev_private; + return to_mdp5_kms(to_mdp_kms(priv->kms)); +} + +#define VSYNC_CLK_RATE 19200000 +static int pingpong_tearcheck_setup(struct drm_encoder *encoder, + struct drm_display_mode *mode) +{ + struct mdp5_kms *mdp5_kms = get_kms(encoder); + struct device *dev = encoder->dev->dev; + u32 total_lines, vclks_line, cfg; + long vsync_clk_speed; + struct mdp5_hw_mixer *mixer = mdp5_crtc_get_mixer(encoder->crtc); + int pp_id = mixer->pp; + + if (IS_ERR_OR_NULL(mdp5_kms->vsync_clk)) { + DRM_DEV_ERROR(dev, "vsync_clk is not initialized\n"); + return -EINVAL; + } + + total_lines = mode->vtotal * drm_mode_vrefresh(mode); + if (!total_lines) { + DRM_DEV_ERROR(dev, "%s: vtotal(%d) or vrefresh(%d) is 0\n", + __func__, mode->vtotal, drm_mode_vrefresh(mode)); + return -EINVAL; + } + + vsync_clk_speed = clk_round_rate(mdp5_kms->vsync_clk, VSYNC_CLK_RATE); + if (vsync_clk_speed <= 0) { + DRM_DEV_ERROR(dev, "vsync_clk round rate failed %ld\n", + vsync_clk_speed); + return -EINVAL; + } + vclks_line = vsync_clk_speed / total_lines; + + cfg = MDP5_PP_SYNC_CONFIG_VSYNC_COUNTER_EN + | MDP5_PP_SYNC_CONFIG_VSYNC_IN_EN; + cfg |= MDP5_PP_SYNC_CONFIG_VSYNC_COUNT(vclks_line); + + /* + * Tearcheck emits a blanking signal every vclks_line * vtotal * 2 ticks on + * the vsync_clk equating to roughly half the desired panel refresh rate. + * This is only necessary as stability fallback if interrupts from the + * panel arrive too late or not at all, but is currently used by default + * because these panel interrupts are not wired up yet. + */ + mdp5_write(mdp5_kms, REG_MDP5_PP_SYNC_CONFIG_VSYNC(pp_id), cfg); + mdp5_write(mdp5_kms, + REG_MDP5_PP_SYNC_CONFIG_HEIGHT(pp_id), (2 * mode->vtotal)); + + mdp5_write(mdp5_kms, + REG_MDP5_PP_VSYNC_INIT_VAL(pp_id), mode->vdisplay); + mdp5_write(mdp5_kms, REG_MDP5_PP_RD_PTR_IRQ(pp_id), mode->vdisplay + 1); + mdp5_write(mdp5_kms, REG_MDP5_PP_START_POS(pp_id), mode->vdisplay); + mdp5_write(mdp5_kms, REG_MDP5_PP_SYNC_THRESH(pp_id), + MDP5_PP_SYNC_THRESH_START(4) | + MDP5_PP_SYNC_THRESH_CONTINUE(4)); + mdp5_write(mdp5_kms, REG_MDP5_PP_AUTOREFRESH_CONFIG(pp_id), 0x0); + + return 0; +} + +static int pingpong_tearcheck_enable(struct drm_encoder *encoder) +{ + struct mdp5_kms *mdp5_kms = get_kms(encoder); + struct mdp5_hw_mixer *mixer = mdp5_crtc_get_mixer(encoder->crtc); + int pp_id = mixer->pp; + int ret; + + ret = clk_set_rate(mdp5_kms->vsync_clk, + clk_round_rate(mdp5_kms->vsync_clk, VSYNC_CLK_RATE)); + if (ret) { + DRM_DEV_ERROR(encoder->dev->dev, + "vsync_clk clk_set_rate failed, %d\n", ret); + return ret; + } + ret = clk_prepare_enable(mdp5_kms->vsync_clk); + if (ret) { + DRM_DEV_ERROR(encoder->dev->dev, + "vsync_clk clk_prepare_enable failed, %d\n", ret); + return ret; + } + + mdp5_write(mdp5_kms, REG_MDP5_PP_TEAR_CHECK_EN(pp_id), 1); + + return 0; +} + +static void pingpong_tearcheck_disable(struct drm_encoder *encoder) +{ + struct mdp5_kms *mdp5_kms = get_kms(encoder); + struct mdp5_hw_mixer *mixer = mdp5_crtc_get_mixer(encoder->crtc); + int pp_id = mixer->pp; + + mdp5_write(mdp5_kms, REG_MDP5_PP_TEAR_CHECK_EN(pp_id), 0); + clk_disable_unprepare(mdp5_kms->vsync_clk); +} + +void mdp5_cmd_encoder_mode_set(struct drm_encoder *encoder, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + mode = adjusted_mode; + + DBG("set mode: " DRM_MODE_FMT, DRM_MODE_ARG(mode)); + pingpong_tearcheck_setup(encoder, mode); + mdp5_crtc_set_pipeline(encoder->crtc); +} + +void mdp5_cmd_encoder_disable(struct drm_encoder *encoder) +{ + struct mdp5_encoder *mdp5_cmd_enc = to_mdp5_encoder(encoder); + struct mdp5_ctl *ctl = mdp5_cmd_enc->ctl; + struct mdp5_interface *intf = mdp5_cmd_enc->intf; + struct mdp5_pipeline *pipeline = mdp5_crtc_get_pipeline(encoder->crtc); + + if (WARN_ON(!mdp5_cmd_enc->enabled)) + return; + + pingpong_tearcheck_disable(encoder); + + mdp5_ctl_set_encoder_state(ctl, pipeline, false); + mdp5_ctl_commit(ctl, pipeline, mdp_ctl_flush_mask_encoder(intf), true); + + mdp5_cmd_enc->enabled = false; +} + +void mdp5_cmd_encoder_enable(struct drm_encoder *encoder) +{ + struct mdp5_encoder *mdp5_cmd_enc = to_mdp5_encoder(encoder); + struct mdp5_ctl *ctl = mdp5_cmd_enc->ctl; + struct mdp5_interface *intf = mdp5_cmd_enc->intf; + struct mdp5_pipeline *pipeline = mdp5_crtc_get_pipeline(encoder->crtc); + + if (WARN_ON(mdp5_cmd_enc->enabled)) + return; + + if (pingpong_tearcheck_enable(encoder)) + return; + + mdp5_ctl_commit(ctl, pipeline, mdp_ctl_flush_mask_encoder(intf), true); + + mdp5_ctl_set_encoder_state(ctl, pipeline, true); + + mdp5_cmd_enc->enabled = true; +} +#endif /* CONFIG_DRM_MSM_DSI */ diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c new file mode 100644 index 000000000..4c4a897fc --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c @@ -0,0 +1,1359 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2014-2015 The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "mdp5_kms.h" +#include "msm_gem.h" + +#define CURSOR_WIDTH 64 +#define CURSOR_HEIGHT 64 + +struct mdp5_crtc { + struct drm_crtc base; + int id; + bool enabled; + + spinlock_t lm_lock; /* protect REG_MDP5_LM_* registers */ + + /* if there is a pending flip, these will be non-null: */ + struct drm_pending_vblank_event *event; + + /* Bits have been flushed at the last commit, + * used to decide if a vsync has happened since last commit. + */ + u32 flushed_mask; + +#define PENDING_CURSOR 0x1 +#define PENDING_FLIP 0x2 + atomic_t pending; + + /* for unref'ing cursor bo's after scanout completes: */ + struct drm_flip_work unref_cursor_work; + + struct mdp_irq vblank; + struct mdp_irq err; + struct mdp_irq pp_done; + + struct completion pp_completion; + + bool lm_cursor_enabled; + + struct { + /* protect REG_MDP5_LM_CURSOR* registers and cursor scanout_bo*/ + spinlock_t lock; + + /* current cursor being scanned out: */ + struct drm_gem_object *scanout_bo; + uint64_t iova; + uint32_t width, height; + int x, y; + } cursor; +}; +#define to_mdp5_crtc(x) container_of(x, struct mdp5_crtc, base) + +static void mdp5_crtc_restore_cursor(struct drm_crtc *crtc); + +static struct mdp5_kms *get_kms(struct drm_crtc *crtc) +{ + struct msm_drm_private *priv = crtc->dev->dev_private; + return to_mdp5_kms(to_mdp_kms(priv->kms)); +} + +static void request_pending(struct drm_crtc *crtc, uint32_t pending) +{ + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + + atomic_or(pending, &mdp5_crtc->pending); + mdp_irq_register(&get_kms(crtc)->base, &mdp5_crtc->vblank); +} + +static void request_pp_done_pending(struct drm_crtc *crtc) +{ + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + reinit_completion(&mdp5_crtc->pp_completion); +} + +static u32 crtc_flush(struct drm_crtc *crtc, u32 flush_mask) +{ + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + struct mdp5_ctl *ctl = mdp5_cstate->ctl; + struct mdp5_pipeline *pipeline = &mdp5_cstate->pipeline; + bool start = !mdp5_cstate->defer_start; + + mdp5_cstate->defer_start = false; + + DBG("%s: flush=%08x", crtc->name, flush_mask); + + return mdp5_ctl_commit(ctl, pipeline, flush_mask, start); +} + +/* + * flush updates, to make sure hw is updated to new scanout fb, + * so that we can safely queue unref to current fb (ie. next + * vblank we know hw is done w/ previous scanout_fb). + */ +static u32 crtc_flush_all(struct drm_crtc *crtc) +{ + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + struct mdp5_hw_mixer *mixer, *r_mixer; + struct drm_plane *plane; + uint32_t flush_mask = 0; + + /* this should not happen: */ + if (WARN_ON(!mdp5_cstate->ctl)) + return 0; + + drm_atomic_crtc_for_each_plane(plane, crtc) { + if (!plane->state->visible) + continue; + flush_mask |= mdp5_plane_get_flush(plane); + } + + mixer = mdp5_cstate->pipeline.mixer; + flush_mask |= mdp_ctl_flush_mask_lm(mixer->lm); + + r_mixer = mdp5_cstate->pipeline.r_mixer; + if (r_mixer) + flush_mask |= mdp_ctl_flush_mask_lm(r_mixer->lm); + + return crtc_flush(crtc, flush_mask); +} + +/* if file!=NULL, this is preclose potential cancel-flip path */ +static void complete_flip(struct drm_crtc *crtc, struct drm_file *file) +{ + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + struct mdp5_pipeline *pipeline = &mdp5_cstate->pipeline; + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + struct mdp5_ctl *ctl = mdp5_cstate->ctl; + struct drm_device *dev = crtc->dev; + struct drm_pending_vblank_event *event; + unsigned long flags; + + spin_lock_irqsave(&dev->event_lock, flags); + event = mdp5_crtc->event; + if (event) { + mdp5_crtc->event = NULL; + DBG("%s: send event: %p", crtc->name, event); + drm_crtc_send_vblank_event(crtc, event); + } + spin_unlock_irqrestore(&dev->event_lock, flags); + + if (ctl && !crtc->state->enable) { + /* set STAGE_UNUSED for all layers */ + mdp5_ctl_blend(ctl, pipeline, NULL, NULL, 0, 0); + /* XXX: What to do here? */ + /* mdp5_crtc->ctl = NULL; */ + } +} + +static void unref_cursor_worker(struct drm_flip_work *work, void *val) +{ + struct mdp5_crtc *mdp5_crtc = + container_of(work, struct mdp5_crtc, unref_cursor_work); + struct mdp5_kms *mdp5_kms = get_kms(&mdp5_crtc->base); + struct msm_kms *kms = &mdp5_kms->base.base; + + msm_gem_unpin_iova(val, kms->vm); + drm_gem_object_put(val); +} + +static void mdp5_crtc_flip_cleanup(struct drm_device *dev, void *ptr) +{ + struct mdp5_crtc *mdp5_crtc = ptr; + + drm_flip_work_cleanup(&mdp5_crtc->unref_cursor_work); +} + +static inline u32 mdp5_lm_use_fg_alpha_mask(enum mdp_mixer_stage_id stage) +{ + switch (stage) { + case STAGE0: return MDP5_LM_BLEND_COLOR_OUT_STAGE0_FG_ALPHA; + case STAGE1: return MDP5_LM_BLEND_COLOR_OUT_STAGE1_FG_ALPHA; + case STAGE2: return MDP5_LM_BLEND_COLOR_OUT_STAGE2_FG_ALPHA; + case STAGE3: return MDP5_LM_BLEND_COLOR_OUT_STAGE3_FG_ALPHA; + case STAGE4: return MDP5_LM_BLEND_COLOR_OUT_STAGE4_FG_ALPHA; + case STAGE5: return MDP5_LM_BLEND_COLOR_OUT_STAGE5_FG_ALPHA; + case STAGE6: return MDP5_LM_BLEND_COLOR_OUT_STAGE6_FG_ALPHA; + default: + return 0; + } +} + +/* + * left/right pipe offsets for the stage array used in blend_setup() + */ +#define PIPE_LEFT 0 +#define PIPE_RIGHT 1 + +/* + * blend_setup() - blend all the planes of a CRTC + * + * If no base layer is available, border will be enabled as the base layer. + * Otherwise all layers will be blended based on their stage calculated + * in mdp5_crtc_atomic_check. + */ +static void blend_setup(struct drm_crtc *crtc) +{ + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + struct mdp5_pipeline *pipeline = &mdp5_cstate->pipeline; + struct mdp5_kms *mdp5_kms = get_kms(crtc); + struct drm_plane *plane; + struct mdp5_plane_state *pstate, *pstates[STAGE_MAX + 1] = {NULL}; + const struct msm_format *format; + struct mdp5_hw_mixer *mixer = pipeline->mixer; + uint32_t lm = mixer->lm; + struct mdp5_hw_mixer *r_mixer = pipeline->r_mixer; + uint32_t r_lm = r_mixer ? r_mixer->lm : 0; + struct mdp5_ctl *ctl = mdp5_cstate->ctl; + uint32_t blend_op, fg_alpha, bg_alpha, ctl_blend_flags = 0; + unsigned long flags; + enum mdp5_pipe stage[STAGE_MAX + 1][MAX_PIPE_STAGE] = { { SSPP_NONE } }; + enum mdp5_pipe r_stage[STAGE_MAX + 1][MAX_PIPE_STAGE] = { { SSPP_NONE } }; + int i, plane_cnt = 0; + bool bg_alpha_enabled = false; + u32 mixer_op_mode = 0; + u32 val; +#define blender(stage) ((stage) - STAGE0) + + spin_lock_irqsave(&mdp5_crtc->lm_lock, flags); + + /* ctl could be released already when we are shutting down: */ + /* XXX: Can this happen now? */ + if (!ctl) + goto out; + + /* Collect all plane information */ + drm_atomic_crtc_for_each_plane(plane, crtc) { + enum mdp5_pipe right_pipe; + + if (!plane->state->visible) + continue; + + pstate = to_mdp5_plane_state(plane->state); + pstates[pstate->stage] = pstate; + stage[pstate->stage][PIPE_LEFT] = mdp5_plane_pipe(plane); + /* + * if we have a right mixer, stage the same pipe as we + * have on the left mixer + */ + if (r_mixer) + r_stage[pstate->stage][PIPE_LEFT] = + mdp5_plane_pipe(plane); + /* + * if we have a right pipe (i.e, the plane comprises of 2 + * hwpipes, then stage the right pipe on the right side of both + * the layer mixers + */ + right_pipe = mdp5_plane_right_pipe(plane); + if (right_pipe) { + stage[pstate->stage][PIPE_RIGHT] = right_pipe; + r_stage[pstate->stage][PIPE_RIGHT] = right_pipe; + } + + plane_cnt++; + } + + if (!pstates[STAGE_BASE]) { + ctl_blend_flags |= MDP5_CTL_BLEND_OP_FLAG_BORDER_OUT; + DBG("Border Color is enabled"); + } else if (plane_cnt) { + format = msm_framebuffer_format(pstates[STAGE_BASE]->base.fb); + + if (format->alpha_enable) + bg_alpha_enabled = true; + } + + /* The reset for blending */ + for (i = STAGE0; i <= STAGE_MAX; i++) { + if (!pstates[i]) + continue; + + format = msm_framebuffer_format(pstates[i]->base.fb); + plane = pstates[i]->base.plane; + blend_op = MDP5_LM_BLEND_OP_MODE_FG_ALPHA(FG_CONST) | + MDP5_LM_BLEND_OP_MODE_BG_ALPHA(BG_CONST); + fg_alpha = pstates[i]->base.alpha >> 8; + bg_alpha = 0xFF - fg_alpha; + + if (!format->alpha_enable && bg_alpha_enabled) + mixer_op_mode = 0; + else + mixer_op_mode |= mdp5_lm_use_fg_alpha_mask(i); + + DBG("Stage %d fg_alpha %x bg_alpha %x", i, fg_alpha, bg_alpha); + + if (format->alpha_enable && + pstates[i]->base.pixel_blend_mode == DRM_MODE_BLEND_PREMULTI) { + blend_op = MDP5_LM_BLEND_OP_MODE_FG_ALPHA(FG_CONST) | + MDP5_LM_BLEND_OP_MODE_BG_ALPHA(FG_PIXEL); + if (fg_alpha != 0xff) { + bg_alpha = fg_alpha; + blend_op |= + MDP5_LM_BLEND_OP_MODE_BG_MOD_ALPHA | + MDP5_LM_BLEND_OP_MODE_BG_INV_MOD_ALPHA; + } else { + blend_op |= MDP5_LM_BLEND_OP_MODE_BG_INV_ALPHA; + } + } else if (format->alpha_enable && + pstates[i]->base.pixel_blend_mode == DRM_MODE_BLEND_COVERAGE) { + blend_op = MDP5_LM_BLEND_OP_MODE_FG_ALPHA(FG_PIXEL) | + MDP5_LM_BLEND_OP_MODE_BG_ALPHA(FG_PIXEL); + if (fg_alpha != 0xff) { + bg_alpha = fg_alpha; + blend_op |= + MDP5_LM_BLEND_OP_MODE_FG_MOD_ALPHA | + MDP5_LM_BLEND_OP_MODE_FG_INV_MOD_ALPHA | + MDP5_LM_BLEND_OP_MODE_BG_MOD_ALPHA | + MDP5_LM_BLEND_OP_MODE_BG_INV_MOD_ALPHA; + } else { + blend_op |= MDP5_LM_BLEND_OP_MODE_BG_INV_ALPHA; + } + } + + mdp5_write(mdp5_kms, REG_MDP5_LM_BLEND_OP_MODE(lm, + blender(i)), blend_op); + mdp5_write(mdp5_kms, REG_MDP5_LM_BLEND_FG_ALPHA(lm, + blender(i)), fg_alpha); + mdp5_write(mdp5_kms, REG_MDP5_LM_BLEND_BG_ALPHA(lm, + blender(i)), bg_alpha); + if (r_mixer) { + mdp5_write(mdp5_kms, REG_MDP5_LM_BLEND_OP_MODE(r_lm, + blender(i)), blend_op); + mdp5_write(mdp5_kms, REG_MDP5_LM_BLEND_FG_ALPHA(r_lm, + blender(i)), fg_alpha); + mdp5_write(mdp5_kms, REG_MDP5_LM_BLEND_BG_ALPHA(r_lm, + blender(i)), bg_alpha); + } + } + + val = mdp5_read(mdp5_kms, REG_MDP5_LM_BLEND_COLOR_OUT(lm)); + mdp5_write(mdp5_kms, REG_MDP5_LM_BLEND_COLOR_OUT(lm), + val | mixer_op_mode); + if (r_mixer) { + val = mdp5_read(mdp5_kms, REG_MDP5_LM_BLEND_COLOR_OUT(r_lm)); + mdp5_write(mdp5_kms, REG_MDP5_LM_BLEND_COLOR_OUT(r_lm), + val | mixer_op_mode); + } + + mdp5_ctl_blend(ctl, pipeline, stage, r_stage, plane_cnt, + ctl_blend_flags); +out: + spin_unlock_irqrestore(&mdp5_crtc->lm_lock, flags); +} + +static void mdp5_crtc_mode_set_nofb(struct drm_crtc *crtc) +{ + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + struct mdp5_kms *mdp5_kms = get_kms(crtc); + struct mdp5_hw_mixer *mixer = mdp5_cstate->pipeline.mixer; + struct mdp5_hw_mixer *r_mixer = mdp5_cstate->pipeline.r_mixer; + uint32_t lm = mixer->lm; + u32 mixer_width, val; + unsigned long flags; + struct drm_display_mode *mode; + + if (WARN_ON(!crtc->state)) + return; + + mode = &crtc->state->adjusted_mode; + + DBG("%s: set mode: " DRM_MODE_FMT, crtc->name, DRM_MODE_ARG(mode)); + + mixer_width = mode->hdisplay; + if (r_mixer) + mixer_width /= 2; + + spin_lock_irqsave(&mdp5_crtc->lm_lock, flags); + mdp5_write(mdp5_kms, REG_MDP5_LM_OUT_SIZE(lm), + MDP5_LM_OUT_SIZE_WIDTH(mixer_width) | + MDP5_LM_OUT_SIZE_HEIGHT(mode->vdisplay)); + + /* Assign mixer to LEFT side in source split mode */ + val = mdp5_read(mdp5_kms, REG_MDP5_LM_BLEND_COLOR_OUT(lm)); + val &= ~MDP5_LM_BLEND_COLOR_OUT_SPLIT_LEFT_RIGHT; + mdp5_write(mdp5_kms, REG_MDP5_LM_BLEND_COLOR_OUT(lm), val); + + if (r_mixer) { + u32 r_lm = r_mixer->lm; + + mdp5_write(mdp5_kms, REG_MDP5_LM_OUT_SIZE(r_lm), + MDP5_LM_OUT_SIZE_WIDTH(mixer_width) | + MDP5_LM_OUT_SIZE_HEIGHT(mode->vdisplay)); + + /* Assign mixer to RIGHT side in source split mode */ + val = mdp5_read(mdp5_kms, REG_MDP5_LM_BLEND_COLOR_OUT(r_lm)); + val |= MDP5_LM_BLEND_COLOR_OUT_SPLIT_LEFT_RIGHT; + mdp5_write(mdp5_kms, REG_MDP5_LM_BLEND_COLOR_OUT(r_lm), val); + } + + spin_unlock_irqrestore(&mdp5_crtc->lm_lock, flags); +} + +static struct drm_encoder *get_encoder_from_crtc(struct drm_crtc *crtc) +{ + struct drm_device *dev = crtc->dev; + struct drm_encoder *encoder; + + drm_for_each_encoder(encoder, dev) + if (encoder->crtc == crtc) + return encoder; + + return NULL; +} + +static bool mdp5_crtc_get_scanout_position(struct drm_crtc *crtc, + bool in_vblank_irq, + int *vpos, int *hpos, + ktime_t *stime, ktime_t *etime, + const struct drm_display_mode *mode) +{ + unsigned int pipe = crtc->index; + struct drm_encoder *encoder; + int line, vsw, vbp, vactive_start, vactive_end, vfp_end; + + + encoder = get_encoder_from_crtc(crtc); + if (!encoder) { + DRM_ERROR("no encoder found for crtc %d\n", pipe); + return false; + } + + vsw = mode->crtc_vsync_end - mode->crtc_vsync_start; + vbp = mode->crtc_vtotal - mode->crtc_vsync_end; + + /* + * the line counter is 1 at the start of the VSYNC pulse and VTOTAL at + * the end of VFP. Translate the porch values relative to the line + * counter positions. + */ + + vactive_start = vsw + vbp + 1; + + vactive_end = vactive_start + mode->crtc_vdisplay; + + /* last scan line before VSYNC */ + vfp_end = mode->crtc_vtotal; + + if (stime) + *stime = ktime_get(); + + line = mdp5_encoder_get_linecount(encoder); + + if (line < vactive_start) + line -= vactive_start; + else if (line > vactive_end) + line = line - vfp_end - vactive_start; + else + line -= vactive_start; + + *vpos = line; + *hpos = 0; + + if (etime) + *etime = ktime_get(); + + return true; +} + +static u32 mdp5_crtc_get_vblank_counter(struct drm_crtc *crtc) +{ + struct drm_encoder *encoder; + + encoder = get_encoder_from_crtc(crtc); + if (!encoder) + return 0; + + return mdp5_encoder_get_framecount(encoder); +} + +static void mdp5_crtc_atomic_disable(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + struct mdp5_kms *mdp5_kms = get_kms(crtc); + struct device *dev = &mdp5_kms->pdev->dev; + unsigned long flags; + + DBG("%s", crtc->name); + + if (WARN_ON(!mdp5_crtc->enabled)) + return; + + /* Disable/save vblank irq handling before power is disabled */ + drm_crtc_vblank_off(crtc); + + if (mdp5_cstate->cmd_mode) + mdp_irq_unregister(&mdp5_kms->base, &mdp5_crtc->pp_done); + + mdp_irq_unregister(&mdp5_kms->base, &mdp5_crtc->err); + pm_runtime_put_sync(dev); + + if (crtc->state->event && !crtc->state->active) { + WARN_ON(mdp5_crtc->event); + spin_lock_irqsave(&mdp5_kms->dev->event_lock, flags); + drm_crtc_send_vblank_event(crtc, crtc->state->event); + crtc->state->event = NULL; + spin_unlock_irqrestore(&mdp5_kms->dev->event_lock, flags); + } + + mdp5_crtc->enabled = false; +} + +static void mdp5_crtc_vblank_on(struct drm_crtc *crtc) +{ + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + struct mdp5_interface *intf = mdp5_cstate->pipeline.intf; + u32 count; + + count = intf->mode == MDP5_INTF_DSI_MODE_COMMAND ? 0 : 0xffffffff; + drm_crtc_set_max_vblank_count(crtc, count); + + drm_crtc_vblank_on(crtc); +} + +static void mdp5_crtc_atomic_enable(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + struct mdp5_kms *mdp5_kms = get_kms(crtc); + struct device *dev = &mdp5_kms->pdev->dev; + + DBG("%s", crtc->name); + + if (WARN_ON(mdp5_crtc->enabled)) + return; + + pm_runtime_get_sync(dev); + + if (mdp5_crtc->lm_cursor_enabled) { + /* + * Restore LM cursor state, as it might have been lost + * with suspend: + */ + if (mdp5_crtc->cursor.iova) { + unsigned long flags; + + spin_lock_irqsave(&mdp5_crtc->cursor.lock, flags); + mdp5_crtc_restore_cursor(crtc); + spin_unlock_irqrestore(&mdp5_crtc->cursor.lock, flags); + + mdp5_ctl_set_cursor(mdp5_cstate->ctl, + &mdp5_cstate->pipeline, 0, true); + } else { + mdp5_ctl_set_cursor(mdp5_cstate->ctl, + &mdp5_cstate->pipeline, 0, false); + } + } + + /* Restore vblank irq handling after power is enabled */ + mdp5_crtc_vblank_on(crtc); + + mdp5_crtc_mode_set_nofb(crtc); + + mdp_irq_register(&mdp5_kms->base, &mdp5_crtc->err); + + if (mdp5_cstate->cmd_mode) + mdp_irq_register(&mdp5_kms->base, &mdp5_crtc->pp_done); + + mdp5_crtc->enabled = true; +} + +static int mdp5_crtc_setup_pipeline(struct drm_crtc *crtc, + struct drm_crtc_state *new_crtc_state, + bool need_right_mixer) +{ + struct mdp5_crtc_state *mdp5_cstate = + to_mdp5_crtc_state(new_crtc_state); + struct mdp5_pipeline *pipeline = &mdp5_cstate->pipeline; + struct mdp5_interface *intf; + bool new_mixer = false; + + new_mixer = !pipeline->mixer; + + if ((need_right_mixer && !pipeline->r_mixer) || + (!need_right_mixer && pipeline->r_mixer)) + new_mixer = true; + + if (new_mixer) { + struct mdp5_hw_mixer *old_mixer = pipeline->mixer; + struct mdp5_hw_mixer *old_r_mixer = pipeline->r_mixer; + u32 caps; + int ret; + + caps = MDP_LM_CAP_DISPLAY; + if (need_right_mixer) + caps |= MDP_LM_CAP_PAIR; + + ret = mdp5_mixer_assign(new_crtc_state->state, crtc, caps, + &pipeline->mixer, need_right_mixer ? + &pipeline->r_mixer : NULL); + if (ret) + return ret; + + ret = mdp5_mixer_release(new_crtc_state->state, old_mixer); + if (ret) + return ret; + + if (old_r_mixer) { + ret = mdp5_mixer_release(new_crtc_state->state, old_r_mixer); + if (ret) + return ret; + + if (!need_right_mixer) + pipeline->r_mixer = NULL; + } + } + + /* + * these should have been already set up in the encoder's atomic + * check (called by drm_atomic_helper_check_modeset) + */ + intf = pipeline->intf; + + mdp5_cstate->err_irqmask = intf2err(intf->num); + mdp5_cstate->vblank_irqmask = intf2vblank(pipeline->mixer, intf); + + if ((intf->type == INTF_DSI) && + (intf->mode == MDP5_INTF_DSI_MODE_COMMAND)) { + mdp5_cstate->pp_done_irqmask = lm2ppdone(pipeline->mixer); + mdp5_cstate->cmd_mode = true; + } else { + mdp5_cstate->pp_done_irqmask = 0; + mdp5_cstate->cmd_mode = false; + } + + return 0; +} + +struct plane_state { + struct drm_plane *plane; + struct mdp5_plane_state *state; +}; + +static int pstate_cmp(const void *a, const void *b) +{ + struct plane_state *pa = (struct plane_state *)a; + struct plane_state *pb = (struct plane_state *)b; + return pa->state->base.normalized_zpos - pb->state->base.normalized_zpos; +} + +/* is there a helper for this? */ +static bool is_fullscreen(struct drm_crtc_state *cstate, + struct drm_plane_state *pstate) +{ + return (pstate->crtc_x <= 0) && (pstate->crtc_y <= 0) && + ((pstate->crtc_x + pstate->crtc_w) >= cstate->mode.hdisplay) && + ((pstate->crtc_y + pstate->crtc_h) >= cstate->mode.vdisplay); +} + +static enum mdp_mixer_stage_id get_start_stage(struct drm_crtc *crtc, + struct drm_crtc_state *new_crtc_state, + struct drm_plane_state *bpstate) +{ + struct mdp5_crtc_state *mdp5_cstate = + to_mdp5_crtc_state(new_crtc_state); + + /* + * if we're in source split mode, it's mandatory to have + * border out on the base stage + */ + if (mdp5_cstate->pipeline.r_mixer) + return STAGE0; + + /* if the bottom-most layer is not fullscreen, we need to use + * it for solid-color: + */ + if (!is_fullscreen(new_crtc_state, bpstate)) + return STAGE0; + + return STAGE_BASE; +} + +static int mdp5_crtc_atomic_check(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + struct drm_crtc_state *crtc_state = drm_atomic_get_new_crtc_state(state, + crtc); + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc_state); + struct mdp5_interface *intf = mdp5_cstate->pipeline.intf; + struct mdp5_kms *mdp5_kms = get_kms(crtc); + struct drm_plane *plane; + struct drm_device *dev = crtc->dev; + struct plane_state pstates[STAGE_MAX + 1]; + const struct mdp5_cfg_hw *hw_cfg; + const struct drm_plane_state *pstate; + const struct drm_display_mode *mode = &crtc_state->adjusted_mode; + bool cursor_plane = false; + bool need_right_mixer = false; + int cnt = 0, i; + int ret; + enum mdp_mixer_stage_id start; + + DBG("%s: check", crtc->name); + + drm_atomic_crtc_state_for_each_plane_state(plane, pstate, crtc_state) { + struct mdp5_plane_state *mdp5_pstate = + to_mdp5_plane_state(pstate); + + if (!pstate->visible) + continue; + + pstates[cnt].plane = plane; + pstates[cnt].state = to_mdp5_plane_state(pstate); + + mdp5_pstate->needs_dirtyfb = + intf->mode == MDP5_INTF_DSI_MODE_COMMAND; + + /* + * if any plane on this crtc uses 2 hwpipes, then we need + * the crtc to have a right hwmixer. + */ + if (pstates[cnt].state->r_hwpipe) + need_right_mixer = true; + cnt++; + + if (plane->type == DRM_PLANE_TYPE_CURSOR) + cursor_plane = true; + } + + /* bail out early if there aren't any planes */ + if (!cnt) + return 0; + + hw_cfg = mdp5_cfg_get_hw_config(mdp5_kms->cfg); + + /* + * we need a right hwmixer if the mode's width is greater than a single + * LM's max width + */ + if (mode->hdisplay > hw_cfg->lm.max_width) + need_right_mixer = true; + + ret = mdp5_crtc_setup_pipeline(crtc, crtc_state, need_right_mixer); + if (ret) { + DRM_DEV_ERROR(dev->dev, "couldn't assign mixers %d\n", ret); + return ret; + } + + /* assign a stage based on sorted zpos property */ + sort(pstates, cnt, sizeof(pstates[0]), pstate_cmp, NULL); + + /* trigger a warning if cursor isn't the highest zorder */ + WARN_ON(cursor_plane && + (pstates[cnt - 1].plane->type != DRM_PLANE_TYPE_CURSOR)); + + start = get_start_stage(crtc, crtc_state, &pstates[0].state->base); + + /* verify that there are not too many planes attached to crtc + * and that we don't have conflicting mixer stages: + */ + if ((cnt + start - 1) >= hw_cfg->lm.nb_stages) { + DRM_DEV_ERROR(dev->dev, "too many planes! cnt=%d, start stage=%d\n", + cnt, start); + return -EINVAL; + } + + for (i = 0; i < cnt; i++) { + if (cursor_plane && (i == (cnt - 1))) + pstates[i].state->stage = hw_cfg->lm.nb_stages; + else + pstates[i].state->stage = start + i; + DBG("%s: assign pipe %s on stage=%d", crtc->name, + pstates[i].plane->name, + pstates[i].state->stage); + } + + return 0; +} + +static void mdp5_crtc_atomic_begin(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + DBG("%s: begin", crtc->name); +} + +static void mdp5_crtc_atomic_flush(struct drm_crtc *crtc, + struct drm_atomic_commit *state) +{ + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + struct drm_device *dev = crtc->dev; + unsigned long flags; + + DBG("%s: event: %p", crtc->name, crtc->state->event); + + WARN_ON(mdp5_crtc->event); + + spin_lock_irqsave(&dev->event_lock, flags); + mdp5_crtc->event = crtc->state->event; + crtc->state->event = NULL; + spin_unlock_irqrestore(&dev->event_lock, flags); + + /* + * If no CTL has been allocated in mdp5_crtc_atomic_check(), + * it means we are trying to flush a CRTC whose state is disabled: + * nothing else needs to be done. + */ + /* XXX: Can this happen now ? */ + if (unlikely(!mdp5_cstate->ctl)) + return; + + blend_setup(crtc); + + /* PP_DONE irq is only used by command mode for now. + * It is better to request pending before FLUSH and START trigger + * to make sure no pp_done irq missed. + * This is safe because no pp_done will happen before SW trigger + * in command mode. + */ + if (mdp5_cstate->cmd_mode) + request_pp_done_pending(crtc); + + mdp5_crtc->flushed_mask = crtc_flush_all(crtc); + + /* XXX are we leaking out state here? */ + mdp5_crtc->vblank.irqmask = mdp5_cstate->vblank_irqmask; + mdp5_crtc->err.irqmask = mdp5_cstate->err_irqmask; + mdp5_crtc->pp_done.irqmask = mdp5_cstate->pp_done_irqmask; + + request_pending(crtc, PENDING_FLIP); +} + +static void get_roi(struct drm_crtc *crtc, uint32_t *roi_w, uint32_t *roi_h) +{ + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + uint32_t xres = crtc->mode.hdisplay; + uint32_t yres = crtc->mode.vdisplay; + + /* + * Cursor Region Of Interest (ROI) is a plane read from cursor + * buffer to render. The ROI region is determined by the visibility of + * the cursor point. In the default Cursor image the cursor point will + * be at the top left of the cursor image. + * + * Without rotation: + * If the cursor point reaches the right (xres - x < cursor.width) or + * bottom (yres - y < cursor.height) boundary of the screen, then ROI + * width and ROI height need to be evaluated to crop the cursor image + * accordingly. + * (xres-x) will be new cursor width when x > (xres - cursor.width) + * (yres-y) will be new cursor height when y > (yres - cursor.height) + * + * With rotation: + * We get negative x and/or y coordinates. + * (cursor.width - abs(x)) will be new cursor width when x < 0 + * (cursor.height - abs(y)) will be new cursor width when y < 0 + */ + if (mdp5_crtc->cursor.x >= 0) + *roi_w = min(mdp5_crtc->cursor.width, xres - + mdp5_crtc->cursor.x); + else + *roi_w = mdp5_crtc->cursor.width - abs(mdp5_crtc->cursor.x); + if (mdp5_crtc->cursor.y >= 0) + *roi_h = min(mdp5_crtc->cursor.height, yres - + mdp5_crtc->cursor.y); + else + *roi_h = mdp5_crtc->cursor.height - abs(mdp5_crtc->cursor.y); +} + +static void mdp5_crtc_restore_cursor(struct drm_crtc *crtc) +{ + const struct drm_format_info *info = drm_format_info(DRM_FORMAT_ARGB8888); + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + struct mdp5_kms *mdp5_kms = get_kms(crtc); + const enum mdp5_cursor_alpha cur_alpha = CURSOR_ALPHA_PER_PIXEL; + uint32_t blendcfg, stride; + uint32_t x, y, src_x, src_y, width, height; + uint32_t roi_w, roi_h; + int lm; + + assert_spin_locked(&mdp5_crtc->cursor.lock); + + lm = mdp5_cstate->pipeline.mixer->lm; + + x = mdp5_crtc->cursor.x; + y = mdp5_crtc->cursor.y; + width = mdp5_crtc->cursor.width; + height = mdp5_crtc->cursor.height; + + stride = width * info->cpp[0]; + + get_roi(crtc, &roi_w, &roi_h); + + /* If cusror buffer overlaps due to rotation on the + * upper or left screen border the pixel offset inside + * the cursor buffer of the ROI is the positive overlap + * distance. + */ + if (mdp5_crtc->cursor.x < 0) { + src_x = abs(mdp5_crtc->cursor.x); + x = 0; + } else { + src_x = 0; + } + if (mdp5_crtc->cursor.y < 0) { + src_y = abs(mdp5_crtc->cursor.y); + y = 0; + } else { + src_y = 0; + } + DBG("%s: x=%d, y=%d roi_w=%d roi_h=%d src_x=%d src_y=%d", + crtc->name, x, y, roi_w, roi_h, src_x, src_y); + + mdp5_write(mdp5_kms, REG_MDP5_LM_CURSOR_STRIDE(lm), stride); + mdp5_write(mdp5_kms, REG_MDP5_LM_CURSOR_FORMAT(lm), + MDP5_LM_CURSOR_FORMAT_FORMAT(CURSOR_FMT_ARGB8888)); + mdp5_write(mdp5_kms, REG_MDP5_LM_CURSOR_IMG_SIZE(lm), + MDP5_LM_CURSOR_IMG_SIZE_SRC_H(height) | + MDP5_LM_CURSOR_IMG_SIZE_SRC_W(width)); + mdp5_write(mdp5_kms, REG_MDP5_LM_CURSOR_SIZE(lm), + MDP5_LM_CURSOR_SIZE_ROI_H(roi_h) | + MDP5_LM_CURSOR_SIZE_ROI_W(roi_w)); + mdp5_write(mdp5_kms, REG_MDP5_LM_CURSOR_START_XY(lm), + MDP5_LM_CURSOR_START_XY_Y_START(y) | + MDP5_LM_CURSOR_START_XY_X_START(x)); + mdp5_write(mdp5_kms, REG_MDP5_LM_CURSOR_XY(lm), + MDP5_LM_CURSOR_XY_SRC_Y(src_y) | + MDP5_LM_CURSOR_XY_SRC_X(src_x)); + mdp5_write(mdp5_kms, REG_MDP5_LM_CURSOR_BASE_ADDR(lm), + mdp5_crtc->cursor.iova); + + blendcfg = MDP5_LM_CURSOR_BLEND_CONFIG_BLEND_EN; + blendcfg |= MDP5_LM_CURSOR_BLEND_CONFIG_BLEND_ALPHA_SEL(cur_alpha); + mdp5_write(mdp5_kms, REG_MDP5_LM_CURSOR_BLEND_CONFIG(lm), blendcfg); +} + +static int mdp5_crtc_cursor_set(struct drm_crtc *crtc, + struct drm_file *file, uint32_t handle, + uint32_t width, uint32_t height) +{ + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + struct mdp5_pipeline *pipeline = &mdp5_cstate->pipeline; + struct drm_device *dev = crtc->dev; + struct mdp5_kms *mdp5_kms = get_kms(crtc); + struct platform_device *pdev = mdp5_kms->pdev; + struct msm_kms *kms = &mdp5_kms->base.base; + struct drm_gem_object *cursor_bo, *old_bo = NULL; + struct mdp5_ctl *ctl; + int ret; + uint32_t flush_mask = mdp_ctl_flush_mask_cursor(0); + bool cursor_enable = true; + unsigned long flags; + + if (!mdp5_crtc->lm_cursor_enabled) { + dev_warn(dev->dev, + "cursor_set is deprecated with cursor planes\n"); + return -EINVAL; + } + + if ((width > CURSOR_WIDTH) || (height > CURSOR_HEIGHT)) { + DRM_DEV_ERROR(dev->dev, "bad cursor size: %dx%d\n", width, height); + return -EINVAL; + } + + ctl = mdp5_cstate->ctl; + if (!ctl) + return -EINVAL; + + /* don't support LM cursors when we have source split enabled */ + if (mdp5_cstate->pipeline.r_mixer) + return -EINVAL; + + if (!handle) { + DBG("Cursor off"); + cursor_enable = false; + mdp5_crtc->cursor.iova = 0; + pm_runtime_get_sync(&pdev->dev); + goto set_cursor; + } + + cursor_bo = drm_gem_object_lookup(file, handle); + if (!cursor_bo) + return -ENOENT; + + ret = msm_gem_get_and_pin_iova(cursor_bo, kms->vm, + &mdp5_crtc->cursor.iova); + if (ret) { + drm_gem_object_put(cursor_bo); + return -EINVAL; + } + + pm_runtime_get_sync(&pdev->dev); + + spin_lock_irqsave(&mdp5_crtc->cursor.lock, flags); + old_bo = mdp5_crtc->cursor.scanout_bo; + + mdp5_crtc->cursor.scanout_bo = cursor_bo; + mdp5_crtc->cursor.width = width; + mdp5_crtc->cursor.height = height; + + mdp5_crtc_restore_cursor(crtc); + + spin_unlock_irqrestore(&mdp5_crtc->cursor.lock, flags); + +set_cursor: + ret = mdp5_ctl_set_cursor(ctl, pipeline, 0, cursor_enable); + if (ret) { + DRM_DEV_ERROR(dev->dev, "failed to %sable cursor: %d\n", + cursor_enable ? "en" : "dis", ret); + goto end; + } + + crtc_flush(crtc, flush_mask); + +end: + pm_runtime_put_sync(&pdev->dev); + if (old_bo) { + drm_flip_work_queue(&mdp5_crtc->unref_cursor_work, old_bo); + /* enable vblank to complete cursor work: */ + request_pending(crtc, PENDING_CURSOR); + } + return ret; +} + +static int mdp5_crtc_cursor_move(struct drm_crtc *crtc, int x, int y) +{ + struct mdp5_kms *mdp5_kms = get_kms(crtc); + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + uint32_t flush_mask = mdp_ctl_flush_mask_cursor(0); + struct drm_device *dev = crtc->dev; + uint32_t roi_w; + uint32_t roi_h; + unsigned long flags; + + if (!mdp5_crtc->lm_cursor_enabled) { + dev_warn(dev->dev, + "cursor_move is deprecated with cursor planes\n"); + return -EINVAL; + } + + /* don't support LM cursors when we have source split enabled */ + if (mdp5_cstate->pipeline.r_mixer) + return -EINVAL; + + /* In case the CRTC is disabled, just drop the cursor update */ + if (unlikely(!crtc->state->enable)) + return 0; + + /* accept negative x/y coordinates up to maximum cursor overlap */ + mdp5_crtc->cursor.x = x = max(x, -(int)mdp5_crtc->cursor.width); + mdp5_crtc->cursor.y = y = max(y, -(int)mdp5_crtc->cursor.height); + + get_roi(crtc, &roi_w, &roi_h); + + pm_runtime_get_sync(&mdp5_kms->pdev->dev); + + spin_lock_irqsave(&mdp5_crtc->cursor.lock, flags); + mdp5_crtc_restore_cursor(crtc); + spin_unlock_irqrestore(&mdp5_crtc->cursor.lock, flags); + + crtc_flush(crtc, flush_mask); + + pm_runtime_put_sync(&mdp5_kms->pdev->dev); + + return 0; +} + +static void +mdp5_crtc_atomic_print_state(struct drm_printer *p, + const struct drm_crtc_state *state) +{ + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(state); + struct mdp5_pipeline *pipeline = &mdp5_cstate->pipeline; + struct mdp5_kms *mdp5_kms = get_kms(state->crtc); + + if (WARN_ON(!pipeline)) + return; + + if (mdp5_cstate->ctl) + drm_printf(p, "\tctl=%d\n", mdp5_ctl_get_ctl_id(mdp5_cstate->ctl)); + + drm_printf(p, "\thwmixer=%s\n", pipeline->mixer ? + pipeline->mixer->name : "(null)"); + + if (mdp5_kms->caps & MDP_CAP_SRC_SPLIT) + drm_printf(p, "\tright hwmixer=%s\n", pipeline->r_mixer ? + pipeline->r_mixer->name : "(null)"); + + drm_printf(p, "\tcmd_mode=%d\n", mdp5_cstate->cmd_mode); +} + +static struct drm_crtc_state * +mdp5_crtc_duplicate_state(struct drm_crtc *crtc) +{ + struct mdp5_crtc_state *mdp5_cstate; + + if (WARN_ON(!crtc->state)) + return NULL; + + mdp5_cstate = kmemdup(to_mdp5_crtc_state(crtc->state), + sizeof(*mdp5_cstate), GFP_KERNEL); + if (!mdp5_cstate) + return NULL; + + __drm_atomic_helper_crtc_duplicate_state(crtc, &mdp5_cstate->base); + + return &mdp5_cstate->base; +} + +static void mdp5_crtc_destroy_state(struct drm_crtc *crtc, struct drm_crtc_state *state) +{ + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(state); + + __drm_atomic_helper_crtc_destroy_state(state); + + kfree(mdp5_cstate); +} + +static void mdp5_crtc_reset(struct drm_crtc *crtc) +{ + struct mdp5_crtc_state *mdp5_cstate = kzalloc_obj(*mdp5_cstate); + + if (crtc->state) + mdp5_crtc_destroy_state(crtc, crtc->state); + + if (mdp5_cstate) + __drm_atomic_helper_crtc_reset(crtc, &mdp5_cstate->base); + else + __drm_atomic_helper_crtc_reset(crtc, NULL); +} + +static const struct drm_crtc_funcs mdp5_crtc_no_lm_cursor_funcs = { + .set_config = drm_atomic_helper_set_config, + .page_flip = drm_atomic_helper_page_flip, + .reset = mdp5_crtc_reset, + .atomic_duplicate_state = mdp5_crtc_duplicate_state, + .atomic_destroy_state = mdp5_crtc_destroy_state, + .atomic_print_state = mdp5_crtc_atomic_print_state, + .get_vblank_counter = mdp5_crtc_get_vblank_counter, + .enable_vblank = msm_crtc_enable_vblank, + .disable_vblank = msm_crtc_disable_vblank, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, +}; + +static const struct drm_crtc_funcs mdp5_crtc_funcs = { + .set_config = drm_atomic_helper_set_config, + .page_flip = drm_atomic_helper_page_flip, + .reset = mdp5_crtc_reset, + .atomic_duplicate_state = mdp5_crtc_duplicate_state, + .atomic_destroy_state = mdp5_crtc_destroy_state, + .cursor_set = mdp5_crtc_cursor_set, + .cursor_move = mdp5_crtc_cursor_move, + .atomic_print_state = mdp5_crtc_atomic_print_state, + .get_vblank_counter = mdp5_crtc_get_vblank_counter, + .enable_vblank = msm_crtc_enable_vblank, + .disable_vblank = msm_crtc_disable_vblank, + .get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp, +}; + +static const struct drm_crtc_helper_funcs mdp5_crtc_helper_funcs = { + .mode_set_nofb = mdp5_crtc_mode_set_nofb, + .atomic_check = mdp5_crtc_atomic_check, + .atomic_begin = mdp5_crtc_atomic_begin, + .atomic_flush = mdp5_crtc_atomic_flush, + .atomic_enable = mdp5_crtc_atomic_enable, + .atomic_disable = mdp5_crtc_atomic_disable, + .get_scanout_position = mdp5_crtc_get_scanout_position, +}; + +static void mdp5_crtc_vblank_irq(struct mdp_irq *irq, uint32_t irqstatus) +{ + struct mdp5_crtc *mdp5_crtc = container_of(irq, struct mdp5_crtc, vblank); + struct drm_crtc *crtc = &mdp5_crtc->base; + struct msm_drm_private *priv = crtc->dev->dev_private; + unsigned pending; + + mdp_irq_unregister(&get_kms(crtc)->base, &mdp5_crtc->vblank); + + pending = atomic_xchg(&mdp5_crtc->pending, 0); + + if (pending & PENDING_FLIP) { + complete_flip(crtc, NULL); + } + + if (pending & PENDING_CURSOR) + drm_flip_work_commit(&mdp5_crtc->unref_cursor_work, priv->kms->wq); +} + +static void mdp5_crtc_err_irq(struct mdp_irq *irq, uint32_t irqstatus) +{ + struct mdp5_crtc *mdp5_crtc = container_of(irq, struct mdp5_crtc, err); + + DBG("%s: error: %08x", mdp5_crtc->base.name, irqstatus); +} + +static void mdp5_crtc_pp_done_irq(struct mdp_irq *irq, uint32_t irqstatus) +{ + struct mdp5_crtc *mdp5_crtc = container_of(irq, struct mdp5_crtc, + pp_done); + + complete_all(&mdp5_crtc->pp_completion); +} + +static void mdp5_crtc_wait_for_pp_done(struct drm_crtc *crtc) +{ + struct drm_device *dev = crtc->dev; + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + int ret; + + ret = wait_for_completion_timeout(&mdp5_crtc->pp_completion, + msecs_to_jiffies(50)); + if (ret == 0) + dev_warn_ratelimited(dev->dev, "pp done time out, lm=%d\n", + mdp5_cstate->pipeline.mixer->lm); +} + +static void mdp5_crtc_wait_for_flush_done(struct drm_crtc *crtc) +{ + struct drm_device *dev = crtc->dev; + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + struct mdp5_ctl *ctl = mdp5_cstate->ctl; + wait_queue_head_t *queue = drm_crtc_vblank_waitqueue(crtc); + int ret; + + /* Should not call this function if crtc is disabled. */ + if (!ctl) + return; + + ret = drm_crtc_vblank_get(crtc); + if (ret) + return; + + ret = wait_event_timeout(*queue, + ((mdp5_ctl_get_commit_status(ctl) & + mdp5_crtc->flushed_mask) == 0), + msecs_to_jiffies(50)); + if (ret <= 0) + dev_warn(dev->dev, "vblank time out, crtc=%d\n", mdp5_crtc->id); + + mdp5_crtc->flushed_mask = 0; + + drm_crtc_vblank_put(crtc); +} + +uint32_t mdp5_crtc_vblank(struct drm_crtc *crtc) +{ + struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc); + return mdp5_crtc->vblank.irqmask; +} + +void mdp5_crtc_set_pipeline(struct drm_crtc *crtc) +{ + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + struct mdp5_kms *mdp5_kms = get_kms(crtc); + + /* should this be done elsewhere ? */ + mdp_irq_update(&mdp5_kms->base); + + mdp5_ctl_set_pipeline(mdp5_cstate->ctl, &mdp5_cstate->pipeline); +} + +struct mdp5_ctl *mdp5_crtc_get_ctl(struct drm_crtc *crtc) +{ + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + + return mdp5_cstate->ctl; +} + +struct mdp5_hw_mixer *mdp5_crtc_get_mixer(struct drm_crtc *crtc) +{ + struct mdp5_crtc_state *mdp5_cstate; + + if (WARN_ON(!crtc)) + return ERR_PTR(-EINVAL); + + mdp5_cstate = to_mdp5_crtc_state(crtc->state); + + return WARN_ON(!mdp5_cstate->pipeline.mixer) ? + ERR_PTR(-EINVAL) : mdp5_cstate->pipeline.mixer; +} + +struct mdp5_pipeline *mdp5_crtc_get_pipeline(struct drm_crtc *crtc) +{ + struct mdp5_crtc_state *mdp5_cstate; + + if (WARN_ON(!crtc)) + return ERR_PTR(-EINVAL); + + mdp5_cstate = to_mdp5_crtc_state(crtc->state); + + return &mdp5_cstate->pipeline; +} + +void mdp5_crtc_wait_for_commit_done(struct drm_crtc *crtc) +{ + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc->state); + + if (mdp5_cstate->cmd_mode) + mdp5_crtc_wait_for_pp_done(crtc); + else + mdp5_crtc_wait_for_flush_done(crtc); +} + +/* initialize crtc */ +struct drm_crtc *mdp5_crtc_init(struct drm_device *dev, + struct drm_plane *plane, + struct drm_plane *cursor_plane, int id) +{ + struct drm_crtc *crtc = NULL; + struct mdp5_crtc *mdp5_crtc; + int ret; + + mdp5_crtc = drmm_crtc_alloc_with_planes(dev, struct mdp5_crtc, base, + plane, cursor_plane, + cursor_plane ? + &mdp5_crtc_no_lm_cursor_funcs : + &mdp5_crtc_funcs, + NULL); + if (IS_ERR(mdp5_crtc)) + return ERR_CAST(mdp5_crtc); + + crtc = &mdp5_crtc->base; + + mdp5_crtc->id = id; + + spin_lock_init(&mdp5_crtc->lm_lock); + spin_lock_init(&mdp5_crtc->cursor.lock); + init_completion(&mdp5_crtc->pp_completion); + + mdp5_crtc->vblank.irq = mdp5_crtc_vblank_irq; + mdp5_crtc->err.irq = mdp5_crtc_err_irq; + mdp5_crtc->pp_done.irq = mdp5_crtc_pp_done_irq; + + mdp5_crtc->lm_cursor_enabled = cursor_plane ? false : true; + + drm_flip_work_init(&mdp5_crtc->unref_cursor_work, + "unref cursor", unref_cursor_worker); + ret = drmm_add_action_or_reset(dev, mdp5_crtc_flip_cleanup, mdp5_crtc); + if (ret) + return ERR_PTR(ret); + + drm_crtc_helper_add(crtc, &mdp5_crtc_helper_funcs); + + return crtc; +} diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.c new file mode 100644 index 000000000..1eca14061 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.c @@ -0,0 +1,662 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2014-2015 The Linux Foundation. All rights reserved. + */ + +#include +#include "mdp5_kms.h" +#include "mdp5_ctl.h" + +/* + * CTL - MDP Control Pool Manager + * + * Controls are shared between all display interfaces. + * + * They are intended to be used for data path configuration. + * The top level register programming describes the complete data path for + * a specific data path ID - REG_MDP5_CTL_*(, ...) + * + * Hardware capabilities determine the number of concurrent data paths + */ + +#define CTL_STAT_BUSY 0x1 +#define CTL_STAT_BOOKED 0x2 + +struct mdp5_ctl { + struct mdp5_ctl_manager *ctlm; + + u32 id; + + /* CTL status bitmask */ + u32 status; + + bool encoder_enabled; + + /* pending flush_mask bits */ + u32 flush_mask; + + /* REG_MDP5_CTL_*() registers access info + lock: */ + spinlock_t hw_lock; + u32 reg_offset; + + /* when do CTL registers need to be flushed? (mask of trigger bits) */ + u32 pending_ctl_trigger; + + bool cursor_on; +}; + +struct mdp5_ctl_manager { + struct drm_device *dev; + + /* number of CTL / Layer Mixers in this hw config: */ + u32 nlm; + u32 nctl; + + /* to filter out non-present bits in the current hardware config */ + u32 flush_hw_mask; + + /* pool of CTLs + lock to protect resource allocation (ctls[i].busy) */ + spinlock_t pool_lock; + struct mdp5_ctl ctls[MAX_CTL]; +}; + +static inline +struct mdp5_kms *get_kms(struct mdp5_ctl_manager *ctl_mgr) +{ + struct msm_drm_private *priv = ctl_mgr->dev->dev_private; + + return to_mdp5_kms(to_mdp_kms(priv->kms)); +} + +static inline +void ctl_write(struct mdp5_ctl *ctl, u32 reg, u32 data) +{ + struct mdp5_kms *mdp5_kms = get_kms(ctl->ctlm); + + (void)ctl->reg_offset; /* TODO use this instead of mdp5_write */ + mdp5_write(mdp5_kms, reg, data); +} + +static inline +u32 ctl_read(struct mdp5_ctl *ctl, u32 reg) +{ + struct mdp5_kms *mdp5_kms = get_kms(ctl->ctlm); + + (void)ctl->reg_offset; /* TODO use this instead of mdp5_write */ + return mdp5_read(mdp5_kms, reg); +} + +static void set_display_intf(struct mdp5_kms *mdp5_kms, + struct mdp5_interface *intf) +{ + unsigned long flags; + u32 intf_sel; + + spin_lock_irqsave(&mdp5_kms->resource_lock, flags); + intf_sel = mdp5_read(mdp5_kms, REG_MDP5_DISP_INTF_SEL); + + switch (intf->num) { + case 0: + intf_sel &= ~MDP5_DISP_INTF_SEL_INTF0__MASK; + intf_sel |= MDP5_DISP_INTF_SEL_INTF0(intf->type); + break; + case 1: + intf_sel &= ~MDP5_DISP_INTF_SEL_INTF1__MASK; + intf_sel |= MDP5_DISP_INTF_SEL_INTF1(intf->type); + break; + case 2: + intf_sel &= ~MDP5_DISP_INTF_SEL_INTF2__MASK; + intf_sel |= MDP5_DISP_INTF_SEL_INTF2(intf->type); + break; + case 3: + intf_sel &= ~MDP5_DISP_INTF_SEL_INTF3__MASK; + intf_sel |= MDP5_DISP_INTF_SEL_INTF3(intf->type); + break; + default: + BUG(); + break; + } + + mdp5_write(mdp5_kms, REG_MDP5_DISP_INTF_SEL, intf_sel); + spin_unlock_irqrestore(&mdp5_kms->resource_lock, flags); +} + +static void set_ctl_op(struct mdp5_ctl *ctl, struct mdp5_pipeline *pipeline) +{ + unsigned long flags; + struct mdp5_interface *intf = pipeline->intf; + u32 ctl_op = 0; + + if (!mdp5_cfg_intf_is_virtual(intf->type)) + ctl_op |= MDP5_CTL_OP_INTF_NUM(INTF0 + intf->num); + + switch (intf->type) { + case INTF_DSI: + if (intf->mode == MDP5_INTF_DSI_MODE_COMMAND) + ctl_op |= MDP5_CTL_OP_CMD_MODE; + break; + + case INTF_WB: + if (intf->mode == MDP5_INTF_WB_MODE_LINE) + ctl_op |= MDP5_CTL_OP_MODE(MODE_WB_2_LINE); + break; + + default: + break; + } + + if (pipeline->r_mixer) + ctl_op |= MDP5_CTL_OP_PACK_3D_ENABLE | + MDP5_CTL_OP_PACK_3D(1); + + spin_lock_irqsave(&ctl->hw_lock, flags); + ctl_write(ctl, REG_MDP5_CTL_OP(ctl->id), ctl_op); + spin_unlock_irqrestore(&ctl->hw_lock, flags); +} + +int mdp5_ctl_set_pipeline(struct mdp5_ctl *ctl, struct mdp5_pipeline *pipeline) +{ + struct mdp5_kms *mdp5_kms = get_kms(ctl->ctlm); + struct mdp5_interface *intf = pipeline->intf; + + /* Virtual interfaces need not set a display intf (e.g.: Writeback) */ + if (!mdp5_cfg_intf_is_virtual(intf->type)) + set_display_intf(mdp5_kms, intf); + + set_ctl_op(ctl, pipeline); + + return 0; +} + +static bool start_signal_needed(struct mdp5_ctl *ctl, + struct mdp5_pipeline *pipeline) +{ + struct mdp5_interface *intf = pipeline->intf; + + if (!ctl->encoder_enabled) + return false; + + switch (intf->type) { + case INTF_WB: + return true; + case INTF_DSI: + return intf->mode == MDP5_INTF_DSI_MODE_COMMAND; + default: + return false; + } +} + +/* + * send_start_signal() - Overlay Processor Start Signal + * + * For a given control operation (display pipeline), a START signal needs to be + * executed in order to kick off operation and activate all layers. + * e.g.: DSI command mode, Writeback + */ +static void send_start_signal(struct mdp5_ctl *ctl) +{ + unsigned long flags; + + spin_lock_irqsave(&ctl->hw_lock, flags); + ctl_write(ctl, REG_MDP5_CTL_START(ctl->id), 1); + spin_unlock_irqrestore(&ctl->hw_lock, flags); +} + +/** + * mdp5_ctl_set_encoder_state() - set the encoder state + * + * @ctl: the CTL instance + * @pipeline: the encoder's INTF + MIXER configuration + * @enabled: true, when encoder is ready for data streaming; false, otherwise. + * + * Note: + * This encoder state is needed to trigger START signal (data path kickoff). + */ +int mdp5_ctl_set_encoder_state(struct mdp5_ctl *ctl, + struct mdp5_pipeline *pipeline, + bool enabled) +{ + struct mdp5_interface *intf = pipeline->intf; + + if (WARN_ON(!ctl)) + return -EINVAL; + + ctl->encoder_enabled = enabled; + DBG("intf_%d: %s", intf->num, str_on_off(enabled)); + + if (start_signal_needed(ctl, pipeline)) { + send_start_signal(ctl); + } + + return 0; +} + +/* + * Note: + * CTL registers need to be flushed after calling this function + * (call mdp5_ctl_commit() with mdp_ctl_flush_mask_ctl() mask) + */ +int mdp5_ctl_set_cursor(struct mdp5_ctl *ctl, struct mdp5_pipeline *pipeline, + int cursor_id, bool enable) +{ + struct mdp5_ctl_manager *ctl_mgr = ctl->ctlm; + unsigned long flags; + u32 blend_cfg; + struct mdp5_hw_mixer *mixer = pipeline->mixer; + + if (WARN_ON(!mixer)) { + DRM_DEV_ERROR(ctl_mgr->dev->dev, "CTL %d cannot find LM", + ctl->id); + return -EINVAL; + } + + if (pipeline->r_mixer) { + DRM_DEV_ERROR(ctl_mgr->dev->dev, "unsupported configuration"); + return -EINVAL; + } + + spin_lock_irqsave(&ctl->hw_lock, flags); + + blend_cfg = ctl_read(ctl, REG_MDP5_CTL_LAYER_REG(ctl->id, mixer->lm)); + + if (enable) + blend_cfg |= MDP5_CTL_LAYER_REG_CURSOR_OUT; + else + blend_cfg &= ~MDP5_CTL_LAYER_REG_CURSOR_OUT; + + ctl_write(ctl, REG_MDP5_CTL_LAYER_REG(ctl->id, mixer->lm), blend_cfg); + ctl->cursor_on = enable; + + spin_unlock_irqrestore(&ctl->hw_lock, flags); + + ctl->pending_ctl_trigger = mdp_ctl_flush_mask_cursor(cursor_id); + + return 0; +} + +static u32 mdp_ctl_blend_mask(enum mdp5_pipe pipe, + enum mdp_mixer_stage_id stage) +{ + switch (pipe) { + case SSPP_VIG0: return MDP5_CTL_LAYER_REG_VIG0(stage); + case SSPP_VIG1: return MDP5_CTL_LAYER_REG_VIG1(stage); + case SSPP_VIG2: return MDP5_CTL_LAYER_REG_VIG2(stage); + case SSPP_RGB0: return MDP5_CTL_LAYER_REG_RGB0(stage); + case SSPP_RGB1: return MDP5_CTL_LAYER_REG_RGB1(stage); + case SSPP_RGB2: return MDP5_CTL_LAYER_REG_RGB2(stage); + case SSPP_DMA0: return MDP5_CTL_LAYER_REG_DMA0(stage); + case SSPP_DMA1: return MDP5_CTL_LAYER_REG_DMA1(stage); + case SSPP_VIG3: return MDP5_CTL_LAYER_REG_VIG3(stage); + case SSPP_RGB3: return MDP5_CTL_LAYER_REG_RGB3(stage); + case SSPP_CURSOR0: + case SSPP_CURSOR1: + default: return 0; + } +} + +static u32 mdp_ctl_blend_ext_mask(enum mdp5_pipe pipe, + enum mdp_mixer_stage_id stage) +{ + if (stage < STAGE6 && (pipe != SSPP_CURSOR0 && pipe != SSPP_CURSOR1)) + return 0; + + switch (pipe) { + case SSPP_VIG0: return MDP5_CTL_LAYER_EXT_REG_VIG0_BIT3; + case SSPP_VIG1: return MDP5_CTL_LAYER_EXT_REG_VIG1_BIT3; + case SSPP_VIG2: return MDP5_CTL_LAYER_EXT_REG_VIG2_BIT3; + case SSPP_RGB0: return MDP5_CTL_LAYER_EXT_REG_RGB0_BIT3; + case SSPP_RGB1: return MDP5_CTL_LAYER_EXT_REG_RGB1_BIT3; + case SSPP_RGB2: return MDP5_CTL_LAYER_EXT_REG_RGB2_BIT3; + case SSPP_DMA0: return MDP5_CTL_LAYER_EXT_REG_DMA0_BIT3; + case SSPP_DMA1: return MDP5_CTL_LAYER_EXT_REG_DMA1_BIT3; + case SSPP_VIG3: return MDP5_CTL_LAYER_EXT_REG_VIG3_BIT3; + case SSPP_RGB3: return MDP5_CTL_LAYER_EXT_REG_RGB3_BIT3; + case SSPP_CURSOR0: return MDP5_CTL_LAYER_EXT_REG_CURSOR0(stage); + case SSPP_CURSOR1: return MDP5_CTL_LAYER_EXT_REG_CURSOR1(stage); + default: return 0; + } +} + +static void mdp5_ctl_reset_blend_regs(struct mdp5_ctl *ctl) +{ + unsigned long flags; + struct mdp5_ctl_manager *ctl_mgr = ctl->ctlm; + int i; + + spin_lock_irqsave(&ctl->hw_lock, flags); + + for (i = 0; i < ctl_mgr->nlm; i++) { + ctl_write(ctl, REG_MDP5_CTL_LAYER_REG(ctl->id, i), 0x0); + ctl_write(ctl, REG_MDP5_CTL_LAYER_EXT_REG(ctl->id, i), 0x0); + } + + spin_unlock_irqrestore(&ctl->hw_lock, flags); +} + +#define PIPE_LEFT 0 +#define PIPE_RIGHT 1 +int mdp5_ctl_blend(struct mdp5_ctl *ctl, struct mdp5_pipeline *pipeline, + enum mdp5_pipe stage[][MAX_PIPE_STAGE], + enum mdp5_pipe r_stage[][MAX_PIPE_STAGE], + u32 stage_cnt, u32 ctl_blend_op_flags) +{ + struct mdp5_hw_mixer *mixer = pipeline->mixer; + struct mdp5_hw_mixer *r_mixer = pipeline->r_mixer; + unsigned long flags; + u32 blend_cfg = 0, blend_ext_cfg = 0; + u32 r_blend_cfg = 0, r_blend_ext_cfg = 0; + int i, start_stage; + + mdp5_ctl_reset_blend_regs(ctl); + + if (ctl_blend_op_flags & MDP5_CTL_BLEND_OP_FLAG_BORDER_OUT) { + start_stage = STAGE0; + blend_cfg |= MDP5_CTL_LAYER_REG_BORDER_COLOR; + if (r_mixer) + r_blend_cfg |= MDP5_CTL_LAYER_REG_BORDER_COLOR; + } else { + start_stage = STAGE_BASE; + } + + for (i = start_stage; stage_cnt && i <= STAGE_MAX; i++) { + blend_cfg |= + mdp_ctl_blend_mask(stage[i][PIPE_LEFT], i) | + mdp_ctl_blend_mask(stage[i][PIPE_RIGHT], i); + blend_ext_cfg |= + mdp_ctl_blend_ext_mask(stage[i][PIPE_LEFT], i) | + mdp_ctl_blend_ext_mask(stage[i][PIPE_RIGHT], i); + if (r_mixer) { + r_blend_cfg |= + mdp_ctl_blend_mask(r_stage[i][PIPE_LEFT], i) | + mdp_ctl_blend_mask(r_stage[i][PIPE_RIGHT], i); + r_blend_ext_cfg |= + mdp_ctl_blend_ext_mask(r_stage[i][PIPE_LEFT], i) | + mdp_ctl_blend_ext_mask(r_stage[i][PIPE_RIGHT], i); + } + } + + spin_lock_irqsave(&ctl->hw_lock, flags); + if (ctl->cursor_on) + blend_cfg |= MDP5_CTL_LAYER_REG_CURSOR_OUT; + + ctl_write(ctl, REG_MDP5_CTL_LAYER_REG(ctl->id, mixer->lm), blend_cfg); + ctl_write(ctl, REG_MDP5_CTL_LAYER_EXT_REG(ctl->id, mixer->lm), + blend_ext_cfg); + if (r_mixer) { + ctl_write(ctl, REG_MDP5_CTL_LAYER_REG(ctl->id, r_mixer->lm), + r_blend_cfg); + ctl_write(ctl, REG_MDP5_CTL_LAYER_EXT_REG(ctl->id, r_mixer->lm), + r_blend_ext_cfg); + } + spin_unlock_irqrestore(&ctl->hw_lock, flags); + + ctl->pending_ctl_trigger = mdp_ctl_flush_mask_lm(mixer->lm); + if (r_mixer) + ctl->pending_ctl_trigger |= mdp_ctl_flush_mask_lm(r_mixer->lm); + + DBG("lm%d: blend config = 0x%08x. ext_cfg = 0x%08x", mixer->lm, + blend_cfg, blend_ext_cfg); + if (r_mixer) + DBG("lm%d: blend config = 0x%08x. ext_cfg = 0x%08x", + r_mixer->lm, r_blend_cfg, r_blend_ext_cfg); + + return 0; +} + +u32 mdp_ctl_flush_mask_encoder(struct mdp5_interface *intf) +{ + if (intf->type == INTF_WB) + return MDP5_CTL_FLUSH_WB; + + switch (intf->num) { + case 0: return MDP5_CTL_FLUSH_TIMING_0; + case 1: return MDP5_CTL_FLUSH_TIMING_1; + case 2: return MDP5_CTL_FLUSH_TIMING_2; + case 3: return MDP5_CTL_FLUSH_TIMING_3; + default: return 0; + } +} + +u32 mdp_ctl_flush_mask_cursor(int cursor_id) +{ + switch (cursor_id) { + case 0: return MDP5_CTL_FLUSH_CURSOR_0; + case 1: return MDP5_CTL_FLUSH_CURSOR_1; + default: return 0; + } +} + +u32 mdp_ctl_flush_mask_pipe(enum mdp5_pipe pipe) +{ + switch (pipe) { + case SSPP_VIG0: return MDP5_CTL_FLUSH_VIG0; + case SSPP_VIG1: return MDP5_CTL_FLUSH_VIG1; + case SSPP_VIG2: return MDP5_CTL_FLUSH_VIG2; + case SSPP_RGB0: return MDP5_CTL_FLUSH_RGB0; + case SSPP_RGB1: return MDP5_CTL_FLUSH_RGB1; + case SSPP_RGB2: return MDP5_CTL_FLUSH_RGB2; + case SSPP_DMA0: return MDP5_CTL_FLUSH_DMA0; + case SSPP_DMA1: return MDP5_CTL_FLUSH_DMA1; + case SSPP_VIG3: return MDP5_CTL_FLUSH_VIG3; + case SSPP_RGB3: return MDP5_CTL_FLUSH_RGB3; + case SSPP_CURSOR0: return MDP5_CTL_FLUSH_CURSOR_0; + case SSPP_CURSOR1: return MDP5_CTL_FLUSH_CURSOR_1; + default: return 0; + } +} + +u32 mdp_ctl_flush_mask_lm(int lm) +{ + switch (lm) { + case 0: return MDP5_CTL_FLUSH_LM0; + case 1: return MDP5_CTL_FLUSH_LM1; + case 2: return MDP5_CTL_FLUSH_LM2; + case 3: return MDP5_CTL_FLUSH_LM3; + case 4: return MDP5_CTL_FLUSH_LM4; + case 5: return MDP5_CTL_FLUSH_LM5; + default: return 0; + } +} + +static u32 fix_sw_flush(struct mdp5_ctl *ctl, struct mdp5_pipeline *pipeline, + u32 flush_mask) +{ + struct mdp5_ctl_manager *ctl_mgr = ctl->ctlm; + u32 sw_mask = 0; +#define BIT_NEEDS_SW_FIX(bit) \ + (!(ctl_mgr->flush_hw_mask & bit) && (flush_mask & bit)) + + /* for some targets, cursor bit is the same as LM bit */ + if (BIT_NEEDS_SW_FIX(MDP5_CTL_FLUSH_CURSOR_0)) + sw_mask |= mdp_ctl_flush_mask_lm(pipeline->mixer->lm); + + return sw_mask; +} + +/** + * mdp5_ctl_commit() - Register Flush + * + * @ctl: the CTL instance + * @pipeline: the encoder's INTF + MIXER configuration + * @flush_mask: bitmask of display controller hw blocks to flush + * @start: if true, immediately update flush registers and set START + * bit, otherwise accumulate flush_mask bits until we are + * ready to START + * + * The flush register is used to indicate several registers are all + * programmed, and are safe to update to the back copy of the double + * buffered registers. + * + * Some registers FLUSH bits are shared when the hardware does not have + * dedicated bits for them; handling these is the job of fix_sw_flush(). + * + * CTL registers need to be flushed in some circumstances; if that is the + * case, some trigger bits will be present in both flush mask and + * ctl->pending_ctl_trigger. + * + * Return H/W flushed bit mask. + */ +u32 mdp5_ctl_commit(struct mdp5_ctl *ctl, + struct mdp5_pipeline *pipeline, + u32 flush_mask, bool start) +{ + struct mdp5_ctl_manager *ctl_mgr = ctl->ctlm; + unsigned long flags; + u32 flush_id = ctl->id; + u32 curr_ctl_flush_mask; + + VERB("flush_mask=%x, trigger=%x", flush_mask, ctl->pending_ctl_trigger); + + if (ctl->pending_ctl_trigger & flush_mask) { + flush_mask |= MDP5_CTL_FLUSH_CTL; + ctl->pending_ctl_trigger = 0; + } + + flush_mask |= fix_sw_flush(ctl, pipeline, flush_mask); + + flush_mask &= ctl_mgr->flush_hw_mask; + + curr_ctl_flush_mask = flush_mask; + + if (!start) { + ctl->flush_mask |= flush_mask; + return curr_ctl_flush_mask; + } else { + flush_mask |= ctl->flush_mask; + ctl->flush_mask = 0; + } + + if (flush_mask) { + spin_lock_irqsave(&ctl->hw_lock, flags); + ctl_write(ctl, REG_MDP5_CTL_FLUSH(flush_id), flush_mask); + spin_unlock_irqrestore(&ctl->hw_lock, flags); + } + + if (start_signal_needed(ctl, pipeline)) { + send_start_signal(ctl); + } + + return curr_ctl_flush_mask; +} + +u32 mdp5_ctl_get_commit_status(struct mdp5_ctl *ctl) +{ + return ctl_read(ctl, REG_MDP5_CTL_FLUSH(ctl->id)); +} + +int mdp5_ctl_get_ctl_id(struct mdp5_ctl *ctl) +{ + return WARN_ON(!ctl) ? -EINVAL : ctl->id; +} + +/* + * mdp5_ctl_request() - CTL allocation + * + * Try to return booked CTL for @intf_num is 1 or 2, unbooked for other INTFs. + * If no CTL is available in preferred category, allocate from the other one. + * + * @return fail if no CTL is available. + */ +struct mdp5_ctl *mdp5_ctlm_request(struct mdp5_ctl_manager *ctl_mgr, + int intf_num) +{ + struct mdp5_ctl *ctl = NULL; + const u32 checkm = CTL_STAT_BUSY | CTL_STAT_BOOKED; + u32 match = ((intf_num == 1) || (intf_num == 2)) ? CTL_STAT_BOOKED : 0; + unsigned long flags; + int c; + + spin_lock_irqsave(&ctl_mgr->pool_lock, flags); + + /* search the preferred */ + for (c = 0; c < ctl_mgr->nctl; c++) + if ((ctl_mgr->ctls[c].status & checkm) == match) + goto found; + + dev_warn(ctl_mgr->dev->dev, + "fall back to the other CTL category for INTF %d!\n", intf_num); + + match ^= CTL_STAT_BOOKED; + for (c = 0; c < ctl_mgr->nctl; c++) + if ((ctl_mgr->ctls[c].status & checkm) == match) + goto found; + + DRM_DEV_ERROR(ctl_mgr->dev->dev, "No more CTL available!"); + goto unlock; + +found: + ctl = &ctl_mgr->ctls[c]; + ctl->status |= CTL_STAT_BUSY; + ctl->pending_ctl_trigger = 0; + DBG("CTL %d allocated", ctl->id); + +unlock: + spin_unlock_irqrestore(&ctl_mgr->pool_lock, flags); + return ctl; +} + +void mdp5_ctlm_hw_reset(struct mdp5_ctl_manager *ctl_mgr) +{ + unsigned long flags; + int c; + + for (c = 0; c < ctl_mgr->nctl; c++) { + struct mdp5_ctl *ctl = &ctl_mgr->ctls[c]; + + spin_lock_irqsave(&ctl->hw_lock, flags); + ctl_write(ctl, REG_MDP5_CTL_OP(ctl->id), 0); + spin_unlock_irqrestore(&ctl->hw_lock, flags); + } +} + +struct mdp5_ctl_manager *mdp5_ctlm_init(struct drm_device *dev, + void __iomem *mmio_base, struct mdp5_cfg_handler *cfg_hnd) +{ + struct mdp5_ctl_manager *ctl_mgr; + const struct mdp5_cfg_hw *hw_cfg = mdp5_cfg_get_hw_config(cfg_hnd); + const struct mdp5_ctl_block *ctl_cfg = &hw_cfg->ctl; + unsigned long flags; + int c, ret; + + ctl_mgr = devm_kzalloc(dev->dev, sizeof(*ctl_mgr), GFP_KERNEL); + if (!ctl_mgr) { + DRM_DEV_ERROR(dev->dev, "failed to allocate CTL manager\n"); + return ERR_PTR(-ENOMEM); + } + + if (WARN_ON(ctl_cfg->count > MAX_CTL)) { + DRM_DEV_ERROR(dev->dev, "Increase static pool size to at least %d\n", + ctl_cfg->count); + return ERR_PTR(-ENOSPC); + } + + /* initialize the CTL manager: */ + ctl_mgr->dev = dev; + ctl_mgr->nlm = hw_cfg->lm.count; + ctl_mgr->nctl = ctl_cfg->count; + ctl_mgr->flush_hw_mask = ctl_cfg->flush_hw_mask; + spin_lock_init(&ctl_mgr->pool_lock); + + /* initialize each CTL of the pool: */ + spin_lock_irqsave(&ctl_mgr->pool_lock, flags); + for (c = 0; c < ctl_mgr->nctl; c++) { + struct mdp5_ctl *ctl = &ctl_mgr->ctls[c]; + + if (WARN_ON(!ctl_cfg->base[c])) { + DRM_DEV_ERROR(dev->dev, "CTL_%d: base is null!\n", c); + ret = -EINVAL; + spin_unlock_irqrestore(&ctl_mgr->pool_lock, flags); + return ERR_PTR(ret); + } + ctl->ctlm = ctl_mgr; + ctl->id = c; + ctl->reg_offset = ctl_cfg->base[c]; + ctl->status = 0; + spin_lock_init(&ctl->hw_lock); + } + + spin_unlock_irqrestore(&ctl_mgr->pool_lock, flags); + DBG("Pool of %d CTLs created.", ctl_mgr->nctl); + + return ctl_mgr; +} diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.h b/drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.h new file mode 100644 index 000000000..0c45f7874 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.h @@ -0,0 +1,76 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2014 The Linux Foundation. All rights reserved. + */ + +#ifndef __MDP5_CTL_H__ +#define __MDP5_CTL_H__ + +#include "msm_drv.h" + +/* + * CTL Manager prototypes: + * mdp5_ctlm_init() returns a ctlm (CTL Manager) handler, + * which is then used to call the other mdp5_ctlm_*(ctlm, ...) functions. + */ +struct mdp5_ctl_manager; +struct mdp5_ctl_manager *mdp5_ctlm_init(struct drm_device *dev, + void __iomem *mmio_base, struct mdp5_cfg_handler *cfg_hnd); +void mdp5_ctlm_hw_reset(struct mdp5_ctl_manager *ctlm); + +/* + * CTL prototypes: + * mdp5_ctl_request(ctlm, ...) returns a ctl (CTL resource) handler, + * which is then used to call the other mdp5_ctl_*(ctl, ...) functions. + */ +struct mdp5_ctl *mdp5_ctlm_request(struct mdp5_ctl_manager *ctlm, int intf_num); + +int mdp5_ctl_get_ctl_id(struct mdp5_ctl *ctl); + +struct mdp5_interface; +struct mdp5_pipeline; +int mdp5_ctl_set_pipeline(struct mdp5_ctl *ctl, struct mdp5_pipeline *p); +int mdp5_ctl_set_encoder_state(struct mdp5_ctl *ctl, struct mdp5_pipeline *p, + bool enabled); + +int mdp5_ctl_set_cursor(struct mdp5_ctl *ctl, struct mdp5_pipeline *pipeline, + int cursor_id, bool enable); + +#define MAX_PIPE_STAGE 2 + +/* + * mdp5_ctl_blend() - Blend multiple layers on a Layer Mixer (LM) + * + * @stage: array to contain the pipe num for each stage + * @stage_cnt: valid stage number in stage array + * @ctl_blend_op_flags: blender operation mode flags + * + * Note: + * CTL registers need to be flushed after calling this function + * (call mdp5_ctl_commit() with mdp_ctl_flush_mask_ctl() mask) + */ +#define MDP5_CTL_BLEND_OP_FLAG_BORDER_OUT BIT(0) +int mdp5_ctl_blend(struct mdp5_ctl *ctl, struct mdp5_pipeline *pipeline, + enum mdp5_pipe stage[][MAX_PIPE_STAGE], + enum mdp5_pipe r_stage[][MAX_PIPE_STAGE], + u32 stage_cnt, u32 ctl_blend_op_flags); + +/** + * mdp_ctl_flush_mask...() - Register FLUSH masks + * + * These masks are used to specify which block(s) need to be flushed + * through @flush_mask parameter in mdp5_ctl_commit(.., flush_mask). + */ +u32 mdp_ctl_flush_mask_lm(int lm); +u32 mdp_ctl_flush_mask_pipe(enum mdp5_pipe pipe); +u32 mdp_ctl_flush_mask_cursor(int cursor_id); +u32 mdp_ctl_flush_mask_encoder(struct mdp5_interface *intf); + +/* @flush_mask: see CTL flush masks definitions below */ +u32 mdp5_ctl_commit(struct mdp5_ctl *ctl, struct mdp5_pipeline *pipeline, + u32 flush_mask, bool start); +u32 mdp5_ctl_get_commit_status(struct mdp5_ctl *ctl); + + + +#endif /* __MDP5_CTL_H__ */ diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_encoder.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_encoder.c new file mode 100644 index 000000000..eaba3b2d7 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_encoder.c @@ -0,0 +1,307 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2014, The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#include +#include + +#include "mdp5_kms.h" + +static struct mdp5_kms *get_kms(struct drm_encoder *encoder) +{ + struct msm_drm_private *priv = encoder->dev->dev_private; + return to_mdp5_kms(to_mdp_kms(priv->kms)); +} + +static void mdp5_vid_encoder_mode_set(struct drm_encoder *encoder, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder); + struct mdp5_kms *mdp5_kms = get_kms(encoder); + struct drm_device *dev = encoder->dev; + struct drm_connector *connector; + int intf = mdp5_encoder->intf->num; + uint32_t dtv_hsync_skew, vsync_period, vsync_len, ctrl_pol; + uint32_t display_v_start, display_v_end; + uint32_t hsync_start_x, hsync_end_x; + uint32_t format = 0x2100; + unsigned long flags; + + mode = adjusted_mode; + + DBG("set mode: " DRM_MODE_FMT, DRM_MODE_ARG(mode)); + + ctrl_pol = 0; + + /* DSI controller cannot handle active-low sync signals. */ + if (mdp5_encoder->intf->type != INTF_DSI) { + if (mode->flags & DRM_MODE_FLAG_NHSYNC) + ctrl_pol |= MDP5_INTF_POLARITY_CTL_HSYNC_LOW; + if (mode->flags & DRM_MODE_FLAG_NVSYNC) + ctrl_pol |= MDP5_INTF_POLARITY_CTL_VSYNC_LOW; + } + /* probably need to get DATA_EN polarity from panel.. */ + + dtv_hsync_skew = 0; /* get this from panel? */ + + /* Get color format from panel, default is 8bpc */ + list_for_each_entry(connector, &dev->mode_config.connector_list, head) { + if (connector->encoder == encoder) { + switch (connector->display_info.bpc) { + case 4: + format |= 0; + break; + case 5: + format |= 0x15; + break; + case 6: + format |= 0x2A; + break; + case 8: + default: + format |= 0x3F; + break; + } + break; + } + } + + hsync_start_x = (mode->htotal - mode->hsync_start); + hsync_end_x = mode->htotal - (mode->hsync_start - mode->hdisplay) - 1; + + vsync_period = mode->vtotal * mode->htotal; + vsync_len = (mode->vsync_end - mode->vsync_start) * mode->htotal; + display_v_start = (mode->vtotal - mode->vsync_start) * mode->htotal + dtv_hsync_skew; + display_v_end = vsync_period - ((mode->vsync_start - mode->vdisplay) * mode->htotal) + dtv_hsync_skew - 1; + + /* + * For edp only: + * DISPLAY_V_START = (VBP * HCYCLE) + HBP + * DISPLAY_V_END = (VBP + VACTIVE) * HCYCLE - 1 - HFP + */ + if (mdp5_encoder->intf->type == INTF_eDP) { + display_v_start += mode->htotal - mode->hsync_start; + display_v_end -= mode->hsync_start - mode->hdisplay; + } + + spin_lock_irqsave(&mdp5_encoder->intf_lock, flags); + + mdp5_write(mdp5_kms, REG_MDP5_INTF_HSYNC_CTL(intf), + MDP5_INTF_HSYNC_CTL_PULSEW(mode->hsync_end - mode->hsync_start) | + MDP5_INTF_HSYNC_CTL_PERIOD(mode->htotal)); + mdp5_write(mdp5_kms, REG_MDP5_INTF_VSYNC_PERIOD_F0(intf), vsync_period); + mdp5_write(mdp5_kms, REG_MDP5_INTF_VSYNC_LEN_F0(intf), vsync_len); + mdp5_write(mdp5_kms, REG_MDP5_INTF_DISPLAY_HCTL(intf), + MDP5_INTF_DISPLAY_HCTL_START(hsync_start_x) | + MDP5_INTF_DISPLAY_HCTL_END(hsync_end_x)); + mdp5_write(mdp5_kms, REG_MDP5_INTF_DISPLAY_VSTART_F0(intf), display_v_start); + mdp5_write(mdp5_kms, REG_MDP5_INTF_DISPLAY_VEND_F0(intf), display_v_end); + mdp5_write(mdp5_kms, REG_MDP5_INTF_BORDER_COLOR(intf), 0); + mdp5_write(mdp5_kms, REG_MDP5_INTF_UNDERFLOW_COLOR(intf), 0xff); + mdp5_write(mdp5_kms, REG_MDP5_INTF_HSYNC_SKEW(intf), dtv_hsync_skew); + mdp5_write(mdp5_kms, REG_MDP5_INTF_POLARITY_CTL(intf), ctrl_pol); + mdp5_write(mdp5_kms, REG_MDP5_INTF_ACTIVE_HCTL(intf), + MDP5_INTF_ACTIVE_HCTL_START(0) | + MDP5_INTF_ACTIVE_HCTL_END(0)); + mdp5_write(mdp5_kms, REG_MDP5_INTF_ACTIVE_VSTART_F0(intf), 0); + mdp5_write(mdp5_kms, REG_MDP5_INTF_ACTIVE_VEND_F0(intf), 0); + mdp5_write(mdp5_kms, REG_MDP5_INTF_PANEL_FORMAT(intf), format); + mdp5_write(mdp5_kms, REG_MDP5_INTF_FRAME_LINE_COUNT_EN(intf), 0x3); /* frame+line? */ + + spin_unlock_irqrestore(&mdp5_encoder->intf_lock, flags); + + mdp5_crtc_set_pipeline(encoder->crtc); +} + +static void mdp5_vid_encoder_disable(struct drm_encoder *encoder) +{ + struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder); + struct mdp5_kms *mdp5_kms = get_kms(encoder); + struct mdp5_ctl *ctl = mdp5_encoder->ctl; + struct mdp5_pipeline *pipeline = mdp5_crtc_get_pipeline(encoder->crtc); + struct mdp5_hw_mixer *mixer = mdp5_crtc_get_mixer(encoder->crtc); + struct mdp5_interface *intf = mdp5_encoder->intf; + int intfn = mdp5_encoder->intf->num; + unsigned long flags; + + if (WARN_ON(!mdp5_encoder->enabled)) + return; + + mdp5_ctl_set_encoder_state(ctl, pipeline, false); + + spin_lock_irqsave(&mdp5_encoder->intf_lock, flags); + mdp5_write(mdp5_kms, REG_MDP5_INTF_TIMING_ENGINE_EN(intfn), 0); + spin_unlock_irqrestore(&mdp5_encoder->intf_lock, flags); + mdp5_ctl_commit(ctl, pipeline, mdp_ctl_flush_mask_encoder(intf), true); + + /* + * Wait for a vsync so we know the ENABLE=0 latched before + * the (connector) source of the vsync's gets disabled, + * otherwise we end up in a funny state if we re-enable + * before the disable latches, which results that some of + * the settings changes for the new modeset (like new + * scanout buffer) don't latch properly.. + */ + mdp_irq_wait(&mdp5_kms->base, intf2vblank(mixer, intf)); + + mdp5_encoder->enabled = false; +} + +static void mdp5_vid_encoder_enable(struct drm_encoder *encoder) +{ + struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder); + struct mdp5_kms *mdp5_kms = get_kms(encoder); + struct mdp5_ctl *ctl = mdp5_encoder->ctl; + struct mdp5_interface *intf = mdp5_encoder->intf; + struct mdp5_pipeline *pipeline = mdp5_crtc_get_pipeline(encoder->crtc); + int intfn = intf->num; + unsigned long flags; + + if (WARN_ON(mdp5_encoder->enabled)) + return; + + spin_lock_irqsave(&mdp5_encoder->intf_lock, flags); + mdp5_write(mdp5_kms, REG_MDP5_INTF_TIMING_ENGINE_EN(intfn), 1); + spin_unlock_irqrestore(&mdp5_encoder->intf_lock, flags); + mdp5_ctl_commit(ctl, pipeline, mdp_ctl_flush_mask_encoder(intf), true); + + mdp5_ctl_set_encoder_state(ctl, pipeline, true); + + mdp5_encoder->enabled = true; +} + +static void mdp5_encoder_mode_set(struct drm_encoder *encoder, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder); + struct mdp5_interface *intf = mdp5_encoder->intf; + + if (intf->mode == MDP5_INTF_DSI_MODE_COMMAND) + mdp5_cmd_encoder_mode_set(encoder, mode, adjusted_mode); + else + mdp5_vid_encoder_mode_set(encoder, mode, adjusted_mode); +} + +static void mdp5_encoder_disable(struct drm_encoder *encoder) +{ + struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder); + struct mdp5_interface *intf = mdp5_encoder->intf; + + if (intf->mode == MDP5_INTF_DSI_MODE_COMMAND) + mdp5_cmd_encoder_disable(encoder); + else + mdp5_vid_encoder_disable(encoder); +} + +static void mdp5_encoder_enable(struct drm_encoder *encoder) +{ + struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder); + struct mdp5_interface *intf = mdp5_encoder->intf; + /* this isn't right I think */ + struct drm_crtc_state *cstate = encoder->crtc->state; + + mdp5_encoder_mode_set(encoder, &cstate->mode, &cstate->adjusted_mode); + + if (intf->mode == MDP5_INTF_DSI_MODE_COMMAND) + mdp5_cmd_encoder_enable(encoder); + else + mdp5_vid_encoder_enable(encoder); +} + +static int mdp5_encoder_atomic_check(struct drm_encoder *encoder, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state) +{ + struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder); + struct mdp5_crtc_state *mdp5_cstate = to_mdp5_crtc_state(crtc_state); + struct mdp5_interface *intf = mdp5_encoder->intf; + struct mdp5_ctl *ctl = mdp5_encoder->ctl; + + mdp5_cstate->ctl = ctl; + mdp5_cstate->pipeline.intf = intf; + + /* + * This is a bit awkward, but we want to flush the CTL and hit the + * START bit at most once for an atomic update. In the non-full- + * modeset case, this is done from crtc->atomic_flush(), but that + * is too early in the case of full modeset, in which case we + * defer to encoder->enable(). But we need to *know* whether + * encoder->enable() will be called to do this: + */ + if (drm_atomic_crtc_needs_modeset(crtc_state)) + mdp5_cstate->defer_start = true; + + return 0; +} + +static const struct drm_encoder_helper_funcs mdp5_encoder_helper_funcs = { + .disable = mdp5_encoder_disable, + .enable = mdp5_encoder_enable, + .atomic_check = mdp5_encoder_atomic_check, +}; + +int mdp5_encoder_get_linecount(struct drm_encoder *encoder) +{ + struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder); + struct mdp5_kms *mdp5_kms = get_kms(encoder); + int intf = mdp5_encoder->intf->num; + + return mdp5_read(mdp5_kms, REG_MDP5_INTF_LINE_COUNT(intf)); +} + +u32 mdp5_encoder_get_framecount(struct drm_encoder *encoder) +{ + struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder); + struct mdp5_kms *mdp5_kms = get_kms(encoder); + int intf = mdp5_encoder->intf->num; + + return mdp5_read(mdp5_kms, REG_MDP5_INTF_FRAME_COUNT(intf)); +} + +void mdp5_encoder_set_intf_mode(struct drm_encoder *encoder, bool cmd_mode) +{ + struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder); + struct mdp5_interface *intf = mdp5_encoder->intf; + + /* TODO: Expand this to set writeback modes too */ + if (cmd_mode) { + WARN_ON(intf->type != INTF_DSI); + intf->mode = MDP5_INTF_DSI_MODE_COMMAND; + } else { + if (intf->type == INTF_DSI) + intf->mode = MDP5_INTF_DSI_MODE_VIDEO; + else + intf->mode = MDP5_INTF_MODE_NONE; + } +} + +/* initialize encoder */ +struct drm_encoder *mdp5_encoder_init(struct drm_device *dev, + struct mdp5_interface *intf, + struct mdp5_ctl *ctl) +{ + struct drm_encoder *encoder = NULL; + struct mdp5_encoder *mdp5_encoder; + int enc_type = (intf->type == INTF_DSI) ? + DRM_MODE_ENCODER_DSI : DRM_MODE_ENCODER_TMDS; + + mdp5_encoder = drmm_encoder_alloc(dev, struct mdp5_encoder, base, + NULL, enc_type, NULL); + if (IS_ERR(mdp5_encoder)) + return ERR_CAST(mdp5_encoder); + + encoder = &mdp5_encoder->base; + mdp5_encoder->ctl = ctl; + mdp5_encoder->intf = intf; + + spin_lock_init(&mdp5_encoder->intf_lock); + + drm_encoder_helper_add(encoder, &mdp5_encoder_helper_funcs); + + return encoder; +} diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_irq.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_irq.c new file mode 100644 index 000000000..b40ed3a84 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_irq.c @@ -0,0 +1,123 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#include + +#include +#include + +#include "msm_drv.h" +#include "mdp5_kms.h" + +void mdp5_set_irqmask(struct mdp_kms *mdp_kms, uint32_t irqmask, + uint32_t old_irqmask) +{ + mdp5_write(to_mdp5_kms(mdp_kms), REG_MDP5_INTR_CLEAR, + irqmask ^ (irqmask & old_irqmask)); + mdp5_write(to_mdp5_kms(mdp_kms), REG_MDP5_INTR_EN, irqmask); +} + +static void mdp5_irq_error_handler(struct mdp_irq *irq, uint32_t irqstatus) +{ + struct mdp5_kms *mdp5_kms = container_of(irq, struct mdp5_kms, error_handler); + static DEFINE_RATELIMIT_STATE(rs, 5*HZ, 1); + extern bool dumpstate; + + DRM_ERROR_RATELIMITED("errors: %08x\n", irqstatus); + + if (dumpstate && __ratelimit(&rs)) { + struct drm_printer p = drm_info_printer(mdp5_kms->dev->dev); + drm_state_dump(mdp5_kms->dev, &p); + } +} + +void mdp5_irq_preinstall(struct msm_kms *kms) +{ + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); + struct device *dev = &mdp5_kms->pdev->dev; + + pm_runtime_get_sync(dev); + mdp5_write(mdp5_kms, REG_MDP5_INTR_CLEAR, 0xffffffff); + mdp5_write(mdp5_kms, REG_MDP5_INTR_EN, 0x00000000); + pm_runtime_put_sync(dev); +} + +int mdp5_irq_postinstall(struct msm_kms *kms) +{ + struct mdp_kms *mdp_kms = to_mdp_kms(kms); + struct mdp5_kms *mdp5_kms = to_mdp5_kms(mdp_kms); + struct device *dev = &mdp5_kms->pdev->dev; + struct mdp_irq *error_handler = &mdp5_kms->error_handler; + + error_handler->irq = mdp5_irq_error_handler; + error_handler->irqmask = MDP5_IRQ_INTF0_UNDER_RUN | + MDP5_IRQ_INTF1_UNDER_RUN | + MDP5_IRQ_INTF2_UNDER_RUN | + MDP5_IRQ_INTF3_UNDER_RUN; + + pm_runtime_get_sync(dev); + mdp_irq_register(mdp_kms, error_handler); + pm_runtime_put_sync(dev); + + return 0; +} + +void mdp5_irq_uninstall(struct msm_kms *kms) +{ + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); + struct device *dev = &mdp5_kms->pdev->dev; + + pm_runtime_get_sync(dev); + mdp5_write(mdp5_kms, REG_MDP5_INTR_EN, 0x00000000); + pm_runtime_put_sync(dev); +} + +irqreturn_t mdp5_irq(struct msm_kms *kms) +{ + struct mdp_kms *mdp_kms = to_mdp_kms(kms); + struct mdp5_kms *mdp5_kms = to_mdp5_kms(mdp_kms); + struct drm_device *dev = mdp5_kms->dev; + struct drm_crtc *crtc; + uint32_t status, enable; + + enable = mdp5_read(mdp5_kms, REG_MDP5_INTR_EN); + status = mdp5_read(mdp5_kms, REG_MDP5_INTR_STATUS) & enable; + mdp5_write(mdp5_kms, REG_MDP5_INTR_CLEAR, status); + + VERB("status=%08x", status); + + mdp_dispatch_irqs(mdp_kms, status); + + drm_for_each_crtc(crtc, dev) + if (status & mdp5_crtc_vblank(crtc)) + drm_crtc_handle_vblank(crtc); + + return IRQ_HANDLED; +} + +int mdp5_enable_vblank(struct msm_kms *kms, struct drm_crtc *crtc) +{ + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); + struct device *dev = &mdp5_kms->pdev->dev; + + pm_runtime_get_sync(dev); + mdp_update_vblank_mask(to_mdp_kms(kms), + mdp5_crtc_vblank(crtc), true); + pm_runtime_put_sync(dev); + + return 0; +} + +void mdp5_disable_vblank(struct msm_kms *kms, struct drm_crtc *crtc) +{ + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); + struct device *dev = &mdp5_kms->pdev->dev; + + pm_runtime_get_sync(dev); + mdp_update_vblank_mask(to_mdp_kms(kms), + mdp5_crtc_vblank(crtc), false); + pm_runtime_put_sync(dev); +} diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c new file mode 100644 index 000000000..3934cd060 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c @@ -0,0 +1,927 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2014, The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#include +#include +#include + +#include +#include +#include +#include + +#include "msm_drv.h" +#include "msm_gem.h" +#include "msm_mmu.h" +#include "mdp5_kms.h" + +static int mdp5_hw_init(struct msm_kms *kms) +{ + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); + struct device *dev = &mdp5_kms->pdev->dev; + unsigned long flags; + + pm_runtime_get_sync(dev); + + /* Magic unknown register writes: + * + * W VBIF:0x004 00000001 (mdss_mdp.c:839) + * W MDP5:0x2e0 0xe9 (mdss_mdp.c:839) + * W MDP5:0x2e4 0x55 (mdss_mdp.c:839) + * W MDP5:0x3ac 0xc0000ccc (mdss_mdp.c:839) + * W MDP5:0x3b4 0xc0000ccc (mdss_mdp.c:839) + * W MDP5:0x3bc 0xcccccc (mdss_mdp.c:839) + * W MDP5:0x4a8 0xcccc0c0 (mdss_mdp.c:839) + * W MDP5:0x4b0 0xccccc0c0 (mdss_mdp.c:839) + * W MDP5:0x4b8 0xccccc000 (mdss_mdp.c:839) + * + * Downstream fbdev driver gets these register offsets/values + * from DT.. not really sure what these registers are or if + * different values for different boards/SoC's, etc. I guess + * they are the golden registers. + * + * Not setting these does not seem to cause any problem. But + * we may be getting lucky with the bootloader initializing + * them for us. OTOH, if we can always count on the bootloader + * setting the golden registers, then perhaps we don't need to + * care. + */ + + spin_lock_irqsave(&mdp5_kms->resource_lock, flags); + mdp5_write(mdp5_kms, REG_MDP5_DISP_INTF_SEL, 0); + spin_unlock_irqrestore(&mdp5_kms->resource_lock, flags); + + mdp5_ctlm_hw_reset(mdp5_kms->ctlm); + + pm_runtime_put_sync(dev); + + return 0; +} + +/* Global/shared object state funcs */ + +/* + * This is a helper that returns the private state currently in operation. + * Note that this would return the "old_state" if called in the atomic check + * path, and the "new_state" after the atomic swap has been done. + */ +struct mdp5_global_state * +mdp5_get_existing_global_state(struct mdp5_kms *mdp5_kms) +{ + return to_mdp5_global_state(mdp5_kms->glob_state.state); +} + +/* + * This acquires the modeset lock set aside for global state, creates + * a new duplicated private object state. + */ +struct mdp5_global_state *mdp5_get_global_state(struct drm_atomic_commit *s) +{ + struct msm_drm_private *priv = s->dev->dev_private; + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(priv->kms)); + struct drm_private_state *priv_state; + + priv_state = drm_atomic_get_private_obj_state(s, &mdp5_kms->glob_state); + if (IS_ERR(priv_state)) + return ERR_CAST(priv_state); + + return to_mdp5_global_state(priv_state); +} + +static struct drm_private_state * +mdp5_global_duplicate_state(struct drm_private_obj *obj) +{ + struct mdp5_global_state *state; + + state = kmemdup(obj->state, sizeof(*state), GFP_KERNEL); + if (!state) + return NULL; + + __drm_atomic_helper_private_obj_duplicate_state(obj, &state->base); + + return &state->base; +} + +static void mdp5_global_destroy_state(struct drm_private_obj *obj, + struct drm_private_state *state) +{ + struct mdp5_global_state *mdp5_state = to_mdp5_global_state(state); + + kfree(mdp5_state); +} + +static struct drm_private_state * +mdp5_global_create_state(struct drm_private_obj *obj) +{ + struct drm_device *dev = obj->dev; + struct msm_drm_private *priv = dev->dev_private; + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(priv->kms)); + struct mdp5_global_state *mdp5_state; + + mdp5_state = kzalloc_obj(*mdp5_state); + if (!mdp5_state) + return ERR_PTR(-ENOMEM); + + __drm_atomic_helper_private_obj_create_state(obj, &mdp5_state->base); + mdp5_state->mdp5_kms = mdp5_kms; + + return &mdp5_state->base; +} + +static void mdp5_global_print_state(struct drm_printer *p, + const struct drm_private_state *state) +{ + struct mdp5_global_state *mdp5_state = to_mdp5_global_state(state); + + if (mdp5_state->mdp5_kms->smp) + mdp5_smp_dump(mdp5_state->mdp5_kms->smp, p, mdp5_state); +} + +static const struct drm_private_state_funcs mdp5_global_state_funcs = { + .atomic_create_state = mdp5_global_create_state, + .atomic_duplicate_state = mdp5_global_duplicate_state, + .atomic_destroy_state = mdp5_global_destroy_state, + .atomic_print_state = mdp5_global_print_state, +}; + +static void mdp5_enable_commit(struct msm_kms *kms) +{ + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); + pm_runtime_get_sync(&mdp5_kms->pdev->dev); +} + +static void mdp5_disable_commit(struct msm_kms *kms) +{ + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); + pm_runtime_put_sync(&mdp5_kms->pdev->dev); +} + +static void mdp5_prepare_commit(struct msm_kms *kms, struct drm_atomic_commit *state) +{ + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); + struct mdp5_global_state *global_state; + + global_state = mdp5_get_existing_global_state(mdp5_kms); + + if (mdp5_kms->smp) + mdp5_smp_prepare_commit(mdp5_kms->smp, &global_state->smp); +} + +static void mdp5_flush_commit(struct msm_kms *kms, unsigned crtc_mask) +{ + /* TODO */ +} + +static void mdp5_wait_flush(struct msm_kms *kms, unsigned crtc_mask) +{ + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); + struct drm_crtc *crtc; + + for_each_crtc_mask(mdp5_kms->dev, crtc, crtc_mask) + mdp5_crtc_wait_for_commit_done(crtc); +} + +static void mdp5_complete_commit(struct msm_kms *kms, unsigned crtc_mask) +{ + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); + struct mdp5_global_state *global_state; + + global_state = mdp5_get_existing_global_state(mdp5_kms); + + if (mdp5_kms->smp) + mdp5_smp_complete_commit(mdp5_kms->smp, &global_state->smp); +} + +static void mdp5_destroy(struct mdp5_kms *mdp5_kms); + +static void mdp5_kms_destroy(struct msm_kms *kms) +{ + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); + + if (kms->vm) { + struct msm_mmu *mmu = to_msm_vm(kms->vm)->mmu; + + mmu->funcs->detach(mmu); + drm_gpuvm_put(kms->vm); + } + + mdp_kms_destroy(&mdp5_kms->base); + mdp5_destroy(mdp5_kms); +} + +static const struct mdp_kms_funcs kms_funcs = { + .base = { + .hw_init = mdp5_hw_init, + .irq_preinstall = mdp5_irq_preinstall, + .irq_postinstall = mdp5_irq_postinstall, + .irq_uninstall = mdp5_irq_uninstall, + .irq = mdp5_irq, + .enable_vblank = mdp5_enable_vblank, + .disable_vblank = mdp5_disable_vblank, + .flush_commit = mdp5_flush_commit, + .enable_commit = mdp5_enable_commit, + .disable_commit = mdp5_disable_commit, + .prepare_commit = mdp5_prepare_commit, + .wait_flush = mdp5_wait_flush, + .complete_commit = mdp5_complete_commit, + .destroy = mdp5_kms_destroy, + }, + .set_irqmask = mdp5_set_irqmask, +}; + +static int mdp5_disable(struct mdp5_kms *mdp5_kms) +{ + DBG(""); + + mdp5_kms->enable_count--; + WARN_ON(mdp5_kms->enable_count < 0); + + clk_disable_unprepare(mdp5_kms->tbu_rt_clk); + clk_disable_unprepare(mdp5_kms->tbu_clk); + clk_disable_unprepare(mdp5_kms->ahb_clk); + clk_disable_unprepare(mdp5_kms->axi_clk); + clk_disable_unprepare(mdp5_kms->core_clk); + clk_disable_unprepare(mdp5_kms->lut_clk); + + return 0; +} + +static int mdp5_enable(struct mdp5_kms *mdp5_kms) +{ + DBG(""); + + mdp5_kms->enable_count++; + + clk_prepare_enable(mdp5_kms->ahb_clk); + clk_prepare_enable(mdp5_kms->axi_clk); + clk_prepare_enable(mdp5_kms->core_clk); + clk_prepare_enable(mdp5_kms->lut_clk); + clk_prepare_enable(mdp5_kms->tbu_clk); + clk_prepare_enable(mdp5_kms->tbu_rt_clk); + + return 0; +} + +static struct drm_encoder *construct_encoder(struct mdp5_kms *mdp5_kms, + struct mdp5_interface *intf, + struct mdp5_ctl *ctl) +{ + struct drm_device *dev = mdp5_kms->dev; + struct drm_encoder *encoder; + + encoder = mdp5_encoder_init(dev, intf, ctl); + if (IS_ERR(encoder)) { + DRM_DEV_ERROR(dev->dev, "failed to construct encoder\n"); + return encoder; + } + + return encoder; +} + +static int get_dsi_id_from_intf(const struct mdp5_cfg_hw *hw_cfg, int intf_num) +{ + const enum mdp5_intf_type *intfs = hw_cfg->intf.connect; + const int intf_cnt = ARRAY_SIZE(hw_cfg->intf.connect); + int id = 0, i; + + for (i = 0; i < intf_cnt; i++) { + if (intfs[i] == INTF_DSI) { + if (intf_num == i) + return id; + + id++; + } + } + + return -EINVAL; +} + +static int modeset_init_intf(struct mdp5_kms *mdp5_kms, + struct mdp5_interface *intf) +{ + struct drm_device *dev = mdp5_kms->dev; + struct msm_drm_private *priv = dev->dev_private; + struct mdp5_ctl_manager *ctlm = mdp5_kms->ctlm; + struct mdp5_ctl *ctl; + struct drm_encoder *encoder; + int ret = 0; + + switch (intf->type) { + case INTF_eDP: + DRM_DEV_INFO(dev->dev, "Skipping eDP interface %d\n", intf->num); + break; + case INTF_HDMI: + if (!priv->kms->hdmi) + break; + + ctl = mdp5_ctlm_request(ctlm, intf->num); + if (!ctl) { + ret = -EINVAL; + break; + } + + encoder = construct_encoder(mdp5_kms, intf, ctl); + if (IS_ERR(encoder)) { + ret = PTR_ERR(encoder); + break; + } + + ret = msm_hdmi_modeset_init(priv->kms->hdmi, dev, encoder); + break; + case INTF_DSI: + { + const struct mdp5_cfg_hw *hw_cfg = + mdp5_cfg_get_hw_config(mdp5_kms->cfg); + int dsi_id = get_dsi_id_from_intf(hw_cfg, intf->num); + + if ((dsi_id >= ARRAY_SIZE(priv->kms->dsi)) || (dsi_id < 0)) { + DRM_DEV_ERROR(dev->dev, "failed to find dsi from intf %d\n", + intf->num); + ret = -EINVAL; + break; + } + + if (!priv->kms->dsi[dsi_id]) + break; + + ctl = mdp5_ctlm_request(ctlm, intf->num); + if (!ctl) { + ret = -EINVAL; + break; + } + + encoder = construct_encoder(mdp5_kms, intf, ctl); + if (IS_ERR(encoder)) { + ret = PTR_ERR(encoder); + break; + } + + ret = msm_dsi_modeset_init(priv->kms->dsi[dsi_id], dev, encoder); + if (!ret) + mdp5_encoder_set_intf_mode(encoder, + msm_dsi_is_cmd_mode(priv->kms->dsi[dsi_id])); + + break; + } + default: + DRM_DEV_ERROR(dev->dev, "unknown intf: %d\n", intf->type); + ret = -EINVAL; + break; + } + + return ret; +} + +static int modeset_init(struct mdp5_kms *mdp5_kms) +{ + struct drm_device *dev = mdp5_kms->dev; + unsigned int num_crtcs; + int i, ret, pi = 0, ci = 0; + struct drm_plane *primary[MAX_BASES] = { NULL }; + struct drm_plane *cursor[MAX_BASES] = { NULL }; + struct drm_encoder *encoder; + unsigned int num_encoders; + + /* + * Construct encoders and modeset initialize connector devices + * for each external display interface. + */ + for (i = 0; i < mdp5_kms->num_intfs; i++) { + ret = modeset_init_intf(mdp5_kms, mdp5_kms->intfs[i]); + if (ret) + goto fail; + } + + num_encoders = 0; + drm_for_each_encoder(encoder, dev) + num_encoders++; + + /* + * We should ideally have less number of encoders (set up by parsing + * the MDP5 interfaces) than the number of layer mixers present in HW, + * but let's be safe here anyway + */ + num_crtcs = min(num_encoders, mdp5_kms->num_hwmixers); + + /* + * Construct planes equaling the number of hw pipes, and CRTCs for the + * N encoders set up by the driver. The first N planes become primary + * planes for the CRTCs, with the remainder as overlay planes: + */ + for (i = 0; i < mdp5_kms->num_hwpipes; i++) { + struct mdp5_hw_pipe *hwpipe = mdp5_kms->hwpipes[i]; + struct drm_plane *plane; + enum drm_plane_type type; + + if (i < num_crtcs) + type = DRM_PLANE_TYPE_PRIMARY; + else if (hwpipe->caps & MDP_PIPE_CAP_CURSOR) + type = DRM_PLANE_TYPE_CURSOR; + else + type = DRM_PLANE_TYPE_OVERLAY; + + plane = mdp5_plane_init(dev, type); + if (IS_ERR(plane)) { + ret = PTR_ERR(plane); + DRM_DEV_ERROR(dev->dev, "failed to construct plane %d (%d)\n", i, ret); + goto fail; + } + + if (type == DRM_PLANE_TYPE_PRIMARY) + primary[pi++] = plane; + if (type == DRM_PLANE_TYPE_CURSOR) + cursor[ci++] = plane; + } + + for (i = 0; i < num_crtcs; i++) { + struct drm_crtc *crtc; + + crtc = mdp5_crtc_init(dev, primary[i], cursor[i], i); + if (IS_ERR(crtc)) { + ret = PTR_ERR(crtc); + DRM_DEV_ERROR(dev->dev, "failed to construct crtc %d (%d)\n", i, ret); + goto fail; + } + } + + /* + * Now that we know the number of crtcs we've created, set the possible + * crtcs for the encoders + */ + drm_for_each_encoder(encoder, dev) + encoder->possible_crtcs = (1 << dev->mode_config.num_crtc) - 1; + + return 0; + +fail: + return ret; +} + +static void read_mdp_hw_revision(struct mdp5_kms *mdp5_kms, + u32 *major, u32 *minor) +{ + struct device *dev = &mdp5_kms->pdev->dev; + u32 version; + + pm_runtime_get_sync(dev); + version = mdp5_read(mdp5_kms, REG_MDP5_HW_VERSION); + pm_runtime_put_sync(dev); + + *major = FIELD(version, MDP5_HW_VERSION_MAJOR); + *minor = FIELD(version, MDP5_HW_VERSION_MINOR); + + DRM_DEV_INFO(dev, "MDP5 version v%d.%d", *major, *minor); +} + +static int get_clk(struct platform_device *pdev, struct clk **clkp, + const char *name, bool mandatory) +{ + struct device *dev = &pdev->dev; + struct clk *clk = msm_clk_get(pdev, name); + if (IS_ERR(clk) && mandatory) { + DRM_DEV_ERROR(dev, "failed to get %s (%ld)\n", name, PTR_ERR(clk)); + return PTR_ERR(clk); + } + if (IS_ERR(clk)) + DBG("skipping %s", name); + else + *clkp = clk; + + return 0; +} + +static int mdp5_init(struct platform_device *pdev, struct drm_device *dev); + +static int mdp5_kms_init(struct drm_device *dev) +{ + struct msm_drm_private *priv = dev->dev_private; + struct platform_device *pdev; + struct mdp5_kms *mdp5_kms; + struct mdp5_cfg *config; + struct msm_kms *kms = priv->kms; + struct drm_gpuvm *vm; + int i, ret; + + ret = mdp5_init(to_platform_device(dev->dev), dev); + if (ret) + return ret; + + mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); + + pdev = mdp5_kms->pdev; + + ret = mdp_kms_init(&mdp5_kms->base, &kms_funcs); + if (ret) { + DRM_DEV_ERROR(&pdev->dev, "failed to init kms\n"); + return ret; + } + + config = mdp5_cfg_get_config(mdp5_kms->cfg); + + /* make sure things are off before attaching iommu (bootloader could + * have left things on, in which case we'll start getting faults if + * we don't disable): + */ + pm_runtime_get_sync(&pdev->dev); + for (i = 0; i < MDP5_INTF_NUM_MAX; i++) { + if (mdp5_cfg_intf_is_virtual(config->hw->intf.connect[i]) || + !config->hw->intf.base[i]) + continue; + mdp5_write(mdp5_kms, REG_MDP5_INTF_TIMING_ENGINE_EN(i), 0); + + mdp5_write(mdp5_kms, REG_MDP5_INTF_FRAME_LINE_COUNT_EN(i), 0x3); + } + mdelay(16); + + vm = msm_kms_init_vm(mdp5_kms->dev, pdev->dev.parent); + if (IS_ERR(vm)) { + ret = PTR_ERR(vm); + return ret; + } + + kms->vm = vm; + + pm_runtime_put_sync(&pdev->dev); + + ret = modeset_init(mdp5_kms); + if (ret) { + DRM_DEV_ERROR(&pdev->dev, "modeset_init failed: %d\n", ret); + return ret; + } + + dev->mode_config.min_width = 0; + dev->mode_config.min_height = 0; + dev->mode_config.max_width = 0xffff; + dev->mode_config.max_height = 0xffff; + + dev->max_vblank_count = 0; /* max_vblank_count is set on each CRTC */ + dev->vblank_disable_immediate = true; + + return 0; +} + +static void mdp5_destroy(struct mdp5_kms *mdp5_kms) +{ + if (mdp5_kms->rpm_enabled) + pm_runtime_disable(&mdp5_kms->pdev->dev); + + drm_atomic_private_obj_fini(&mdp5_kms->glob_state); +} + +static int construct_pipes(struct mdp5_kms *mdp5_kms, int cnt, + const enum mdp5_pipe *pipes, const uint32_t *offsets, + uint32_t caps) +{ + struct drm_device *dev = mdp5_kms->dev; + int i, ret; + + for (i = 0; i < cnt; i++) { + struct mdp5_hw_pipe *hwpipe; + + hwpipe = mdp5_pipe_init(dev, pipes[i], offsets[i], caps); + if (IS_ERR(hwpipe)) { + ret = PTR_ERR(hwpipe); + DRM_DEV_ERROR(dev->dev, "failed to construct pipe for %s (%d)\n", + pipe2name(pipes[i]), ret); + return ret; + } + hwpipe->idx = mdp5_kms->num_hwpipes; + mdp5_kms->hwpipes[mdp5_kms->num_hwpipes++] = hwpipe; + } + + return 0; +} + +static int hwpipe_init(struct mdp5_kms *mdp5_kms) +{ + static const enum mdp5_pipe rgb_planes[] = { + SSPP_RGB0, SSPP_RGB1, SSPP_RGB2, SSPP_RGB3, + }; + static const enum mdp5_pipe vig_planes[] = { + SSPP_VIG0, SSPP_VIG1, SSPP_VIG2, SSPP_VIG3, + }; + static const enum mdp5_pipe dma_planes[] = { + SSPP_DMA0, SSPP_DMA1, + }; + static const enum mdp5_pipe cursor_planes[] = { + SSPP_CURSOR0, SSPP_CURSOR1, + }; + const struct mdp5_cfg_hw *hw_cfg; + int ret; + + hw_cfg = mdp5_cfg_get_hw_config(mdp5_kms->cfg); + + /* Construct RGB pipes: */ + ret = construct_pipes(mdp5_kms, hw_cfg->pipe_rgb.count, rgb_planes, + hw_cfg->pipe_rgb.base, hw_cfg->pipe_rgb.caps); + if (ret) + return ret; + + /* Construct video (VIG) pipes: */ + ret = construct_pipes(mdp5_kms, hw_cfg->pipe_vig.count, vig_planes, + hw_cfg->pipe_vig.base, hw_cfg->pipe_vig.caps); + if (ret) + return ret; + + /* Construct DMA pipes: */ + ret = construct_pipes(mdp5_kms, hw_cfg->pipe_dma.count, dma_planes, + hw_cfg->pipe_dma.base, hw_cfg->pipe_dma.caps); + if (ret) + return ret; + + /* Construct cursor pipes: */ + ret = construct_pipes(mdp5_kms, hw_cfg->pipe_cursor.count, + cursor_planes, hw_cfg->pipe_cursor.base, + hw_cfg->pipe_cursor.caps); + if (ret) + return ret; + + return 0; +} + +static int hwmixer_init(struct mdp5_kms *mdp5_kms) +{ + struct drm_device *dev = mdp5_kms->dev; + const struct mdp5_cfg_hw *hw_cfg; + int i, ret; + + hw_cfg = mdp5_cfg_get_hw_config(mdp5_kms->cfg); + + for (i = 0; i < hw_cfg->lm.count; i++) { + struct mdp5_hw_mixer *mixer; + + mixer = mdp5_mixer_init(dev, &hw_cfg->lm.instances[i]); + if (IS_ERR(mixer)) { + ret = PTR_ERR(mixer); + DRM_DEV_ERROR(dev->dev, "failed to construct LM%d (%d)\n", + i, ret); + return ret; + } + + mixer->idx = mdp5_kms->num_hwmixers; + mdp5_kms->hwmixers[mdp5_kms->num_hwmixers++] = mixer; + } + + return 0; +} + +static int interface_init(struct mdp5_kms *mdp5_kms) +{ + struct drm_device *dev = mdp5_kms->dev; + const struct mdp5_cfg_hw *hw_cfg; + const enum mdp5_intf_type *intf_types; + int i; + + hw_cfg = mdp5_cfg_get_hw_config(mdp5_kms->cfg); + intf_types = hw_cfg->intf.connect; + + for (i = 0; i < ARRAY_SIZE(hw_cfg->intf.connect); i++) { + struct mdp5_interface *intf; + + if (intf_types[i] == INTF_DISABLED) + continue; + + intf = devm_kzalloc(dev->dev, sizeof(*intf), GFP_KERNEL); + if (!intf) { + DRM_DEV_ERROR(dev->dev, "failed to construct INTF%d\n", i); + return -ENOMEM; + } + + intf->num = i; + intf->type = intf_types[i]; + intf->mode = MDP5_INTF_MODE_NONE; + intf->idx = mdp5_kms->num_intfs; + mdp5_kms->intfs[mdp5_kms->num_intfs++] = intf; + } + + return 0; +} + +static int mdp5_init(struct platform_device *pdev, struct drm_device *dev) +{ + struct msm_drm_private *priv = dev->dev_private; + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(priv->kms)); + struct mdp5_cfg *config; + u32 major, minor; + int ret; + + mdp5_kms->dev = dev; + + drm_atomic_private_obj_init(mdp5_kms->dev, &mdp5_kms->glob_state, + &mdp5_global_state_funcs); + + /* we need to set a default rate before enabling. Set a safe + * rate first, then figure out hw revision, and then set a + * more optimal rate: + */ + clk_set_rate(mdp5_kms->core_clk, 200000000); + + pm_runtime_enable(&pdev->dev); + mdp5_kms->rpm_enabled = true; + + read_mdp_hw_revision(mdp5_kms, &major, &minor); + + mdp5_kms->cfg = mdp5_cfg_init(mdp5_kms, major, minor); + if (IS_ERR(mdp5_kms->cfg)) { + ret = PTR_ERR(mdp5_kms->cfg); + mdp5_kms->cfg = NULL; + goto fail; + } + + config = mdp5_cfg_get_config(mdp5_kms->cfg); + mdp5_kms->caps = config->hw->mdp.caps; + + /* TODO: compute core clock rate at runtime */ + clk_set_rate(mdp5_kms->core_clk, config->hw->max_clk); + + /* + * Some chipsets have a Shared Memory Pool (SMP), while others + * have dedicated latency buffering per source pipe instead; + * this section initializes the SMP: + */ + if (mdp5_kms->caps & MDP_CAP_SMP) { + mdp5_kms->smp = mdp5_smp_init(mdp5_kms, &config->hw->smp); + if (IS_ERR(mdp5_kms->smp)) { + ret = PTR_ERR(mdp5_kms->smp); + mdp5_kms->smp = NULL; + goto fail; + } + } + + mdp5_kms->ctlm = mdp5_ctlm_init(dev, mdp5_kms->mmio, mdp5_kms->cfg); + if (IS_ERR(mdp5_kms->ctlm)) { + ret = PTR_ERR(mdp5_kms->ctlm); + mdp5_kms->ctlm = NULL; + goto fail; + } + + ret = hwpipe_init(mdp5_kms); + if (ret) + goto fail; + + ret = hwmixer_init(mdp5_kms); + if (ret) + goto fail; + + ret = interface_init(mdp5_kms); + if (ret) + goto fail; + + return 0; +fail: + mdp5_destroy(mdp5_kms); + return ret; +} + +static int mdp5_setup_interconnect(struct platform_device *pdev) +{ + struct icc_path *path0 = msm_icc_get(&pdev->dev, "mdp0-mem"); + struct icc_path *path1 = msm_icc_get(&pdev->dev, "mdp1-mem"); + struct icc_path *path_rot = msm_icc_get(&pdev->dev, "rotator-mem"); + + if (IS_ERR(path0)) + return PTR_ERR(path0); + + if (!path0) { + /* no interconnect support is not necessarily a fatal + * condition, the platform may simply not have an + * interconnect driver yet. But warn about it in case + * bootloader didn't setup bus clocks high enough for + * scanout. + */ + dev_warn(&pdev->dev, "No interconnect support may cause display underflows!\n"); + return 0; + } + + icc_set_bw(path0, 0, MBps_to_icc(6400)); + + if (!IS_ERR_OR_NULL(path1)) + icc_set_bw(path1, 0, MBps_to_icc(6400)); + if (!IS_ERR_OR_NULL(path_rot)) + icc_set_bw(path_rot, 0, MBps_to_icc(6400)); + + return 0; +} + +static int mdp5_dev_probe(struct platform_device *pdev) +{ + struct mdp5_kms *mdp5_kms; + int ret, irq; + + DBG(""); + + if (!msm_disp_drv_should_bind(&pdev->dev, false)) + return -ENODEV; + + mdp5_kms = devm_kzalloc(&pdev->dev, sizeof(*mdp5_kms), GFP_KERNEL); + if (!mdp5_kms) + return -ENOMEM; + + ret = mdp5_setup_interconnect(pdev); + if (ret) + return ret; + + mdp5_kms->pdev = pdev; + + spin_lock_init(&mdp5_kms->resource_lock); + + mdp5_kms->mmio = msm_ioremap(pdev, "mdp_phys"); + if (IS_ERR(mdp5_kms->mmio)) + return PTR_ERR(mdp5_kms->mmio); + + /* mandatory clocks: */ + ret = get_clk(pdev, &mdp5_kms->axi_clk, "bus", true); + if (ret) + return ret; + ret = get_clk(pdev, &mdp5_kms->ahb_clk, "iface", true); + if (ret) + return ret; + ret = get_clk(pdev, &mdp5_kms->core_clk, "core", true); + if (ret) + return ret; + ret = get_clk(pdev, &mdp5_kms->vsync_clk, "vsync", true); + if (ret) + return ret; + + /* optional clocks: */ + get_clk(pdev, &mdp5_kms->lut_clk, "lut", false); + get_clk(pdev, &mdp5_kms->tbu_clk, "tbu", false); + get_clk(pdev, &mdp5_kms->tbu_rt_clk, "tbu_rt", false); + + irq = platform_get_irq(pdev, 0); + if (irq < 0) + return dev_err_probe(&pdev->dev, irq, "failed to get irq\n"); + + mdp5_kms->base.base.irq = irq; + + return msm_drv_probe(&pdev->dev, mdp5_kms_init, &mdp5_kms->base.base); +} + +static void mdp5_dev_remove(struct platform_device *pdev) +{ + DBG(""); + component_master_del(&pdev->dev, &msm_drm_ops); +} + +static __maybe_unused int mdp5_runtime_suspend(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct msm_drm_private *priv = platform_get_drvdata(pdev); + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(priv->kms)); + + DBG(""); + + return mdp5_disable(mdp5_kms); +} + +static __maybe_unused int mdp5_runtime_resume(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct msm_drm_private *priv = platform_get_drvdata(pdev); + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(priv->kms)); + + DBG(""); + + return mdp5_enable(mdp5_kms); +} + +static const struct dev_pm_ops mdp5_pm_ops = { + SET_RUNTIME_PM_OPS(mdp5_runtime_suspend, mdp5_runtime_resume, NULL) + .prepare = msm_kms_pm_prepare, + .complete = msm_kms_pm_complete, +}; + +static const struct of_device_id mdp5_dt_match[] = { + { .compatible = "qcom,mdp5", }, + /* to support downstream DT files */ + { .compatible = "qcom,mdss_mdp", }, + {} +}; +MODULE_DEVICE_TABLE(of, mdp5_dt_match); + +static struct platform_driver mdp5_driver = { + .probe = mdp5_dev_probe, + .remove = mdp5_dev_remove, + .shutdown = msm_kms_shutdown, + .driver = { + .name = "msm_mdp", + .of_match_table = mdp5_dt_match, + .pm = &mdp5_pm_ops, + }, +}; + +void __init msm_mdp_register(void) +{ + DBG(""); + platform_driver_register(&mdp5_driver); +} + +void __exit msm_mdp_unregister(void) +{ + DBG(""); + platform_driver_unregister(&mdp5_driver); +} diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.h b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.h new file mode 100644 index 000000000..7bf2172fc --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.h @@ -0,0 +1,317 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#ifndef __MDP5_KMS_H__ +#define __MDP5_KMS_H__ + +#include "msm_drv.h" +#include "msm_kms.h" +#include "disp/mdp_kms.h" +#include "mdp5_cfg.h" /* must be included before mdp5.xml.h */ +#include "mdp5.xml.h" +#include "mdp5_pipe.h" +#include "mdp5_mixer.h" +#include "mdp5_ctl.h" +#include "mdp5_smp.h" + +struct mdp5_kms { + struct mdp_kms base; + + struct drm_device *dev; + + struct platform_device *pdev; + + unsigned num_hwpipes; + struct mdp5_hw_pipe *hwpipes[SSPP_MAX]; + + unsigned num_hwmixers; + struct mdp5_hw_mixer *hwmixers[8]; + + unsigned num_intfs; + struct mdp5_interface *intfs[5]; + + struct mdp5_cfg_handler *cfg; + uint32_t caps; /* MDP capabilities (MDP_CAP_XXX bits) */ + + /* + * Global private object state, Do not access directly, use + * mdp5_global_get_state() + */ + struct drm_private_obj glob_state; + + struct mdp5_smp *smp; + struct mdp5_ctl_manager *ctlm; + + /* io/register spaces: */ + void __iomem *mmio; + + struct clk *axi_clk; + struct clk *ahb_clk; + struct clk *core_clk; + struct clk *lut_clk; + struct clk *tbu_clk; + struct clk *tbu_rt_clk; + struct clk *vsync_clk; + + /* + * lock to protect access to global resources: ie., following register: + * - REG_MDP5_DISP_INTF_SEL + */ + spinlock_t resource_lock; + + bool rpm_enabled; + + struct mdp_irq error_handler; + + int enable_count; +}; +#define to_mdp5_kms(x) container_of(x, struct mdp5_kms, base) + +/* Global private object state for tracking resources that are shared across + * multiple kms objects (planes/crtcs/etc). + */ +#define to_mdp5_global_state(x) container_of(x, struct mdp5_global_state, base) +struct mdp5_global_state { + struct drm_private_state base; + + struct drm_atomic_commit *state; + struct mdp5_kms *mdp5_kms; + + struct mdp5_hw_pipe_state hwpipe; + struct mdp5_hw_mixer_state hwmixer; + struct mdp5_smp_state smp; +}; + +struct mdp5_global_state * mdp5_get_existing_global_state(struct mdp5_kms *mdp5_kms); +struct mdp5_global_state *__must_check mdp5_get_global_state(struct drm_atomic_commit *s); + +/* Atomic plane state. Subclasses the base drm_plane_state in order to + * track assigned hwpipe and hw specific state. + */ +struct mdp5_plane_state { + struct drm_plane_state base; + + struct mdp5_hw_pipe *hwpipe; + struct mdp5_hw_pipe *r_hwpipe; /* right hwpipe */ + + /* assigned by crtc blender */ + enum mdp_mixer_stage_id stage; + + /* whether attached CRTC needs pixel data explicitly flushed to + * display (ex. DSI command mode display) + */ + bool needs_dirtyfb; +}; +#define to_mdp5_plane_state(x) \ + container_of(x, struct mdp5_plane_state, base) + +struct mdp5_pipeline { + struct mdp5_interface *intf; + struct mdp5_hw_mixer *mixer; + struct mdp5_hw_mixer *r_mixer; /* right mixer */ +}; + +struct mdp5_crtc_state { + struct drm_crtc_state base; + + struct mdp5_ctl *ctl; + struct mdp5_pipeline pipeline; + + /* these are derivatives of intf/mixer state in mdp5_pipeline */ + u32 vblank_irqmask; + u32 err_irqmask; + u32 pp_done_irqmask; + + bool cmd_mode; + + /* should we not write CTL[n].START register on flush? If the + * encoder has changed this is set to true, since encoder->enable() + * is called after crtc state is committed, but we only want to + * write the CTL[n].START register once. This lets us defer + * writing CTL[n].START until encoder->enable() + */ + bool defer_start; +}; +#define to_mdp5_crtc_state(x) \ + container_of(x, struct mdp5_crtc_state, base) + +enum mdp5_intf_mode { + MDP5_INTF_MODE_NONE = 0, + + /* Modes used for DSI interface (INTF_DSI type): */ + MDP5_INTF_DSI_MODE_VIDEO, + MDP5_INTF_DSI_MODE_COMMAND, + + /* Modes used for WB interface (INTF_WB type): */ + MDP5_INTF_WB_MODE_BLOCK, + MDP5_INTF_WB_MODE_LINE, +}; + +struct mdp5_interface { + int idx; + int num; /* display interface number */ + enum mdp5_intf_type type; + enum mdp5_intf_mode mode; +}; + +struct mdp5_encoder { + struct drm_encoder base; + spinlock_t intf_lock; /* protect REG_MDP5_INTF_* registers */ + bool enabled; + uint32_t bsc; + + struct mdp5_interface *intf; + struct mdp5_ctl *ctl; +}; +#define to_mdp5_encoder(x) container_of(x, struct mdp5_encoder, base) + +static inline void mdp5_write(struct mdp5_kms *mdp5_kms, u32 reg, u32 data) +{ + WARN_ON(mdp5_kms->enable_count <= 0); + writel(data, mdp5_kms->mmio + reg); +} + +static inline u32 mdp5_read(struct mdp5_kms *mdp5_kms, u32 reg) +{ + WARN_ON(mdp5_kms->enable_count <= 0); + return readl(mdp5_kms->mmio + reg); +} + +static inline const char *stage2name(enum mdp_mixer_stage_id stage) +{ + static const char *names[] = { +#define NAME(n) [n] = #n + NAME(STAGE_UNUSED), NAME(STAGE_BASE), + NAME(STAGE0), NAME(STAGE1), NAME(STAGE2), + NAME(STAGE3), NAME(STAGE4), NAME(STAGE6), +#undef NAME + }; + return names[stage]; +} + +static inline const char *pipe2name(enum mdp5_pipe pipe) +{ + static const char *names[] = { +#define NAME(n) [SSPP_ ## n] = #n + NAME(VIG0), NAME(VIG1), NAME(VIG2), + NAME(RGB0), NAME(RGB1), NAME(RGB2), + NAME(DMA0), NAME(DMA1), + NAME(VIG3), NAME(RGB3), + NAME(CURSOR0), NAME(CURSOR1), +#undef NAME + }; + return names[pipe]; +} + +static inline int pipe2nclients(enum mdp5_pipe pipe) +{ + switch (pipe) { + case SSPP_RGB0: + case SSPP_RGB1: + case SSPP_RGB2: + case SSPP_RGB3: + return 1; + default: + return 3; + } +} + +static inline uint32_t intf2err(int intf_num) +{ + switch (intf_num) { + case 0: return MDP5_IRQ_INTF0_UNDER_RUN; + case 1: return MDP5_IRQ_INTF1_UNDER_RUN; + case 2: return MDP5_IRQ_INTF2_UNDER_RUN; + case 3: return MDP5_IRQ_INTF3_UNDER_RUN; + default: return 0; + } +} + +static inline uint32_t intf2vblank(struct mdp5_hw_mixer *mixer, + struct mdp5_interface *intf) +{ + /* + * In case of DSI Command Mode, the Ping Pong's read pointer IRQ + * acts as a Vblank signal. The Ping Pong buffer used is bound to + * layer mixer. + */ + + if ((intf->type == INTF_DSI) && + (intf->mode == MDP5_INTF_DSI_MODE_COMMAND)) + return MDP5_IRQ_PING_PONG_0_RD_PTR << mixer->pp; + + if (intf->type == INTF_WB) + return MDP5_IRQ_WB_2_DONE; + + switch (intf->num) { + case 0: return MDP5_IRQ_INTF0_VSYNC; + case 1: return MDP5_IRQ_INTF1_VSYNC; + case 2: return MDP5_IRQ_INTF2_VSYNC; + case 3: return MDP5_IRQ_INTF3_VSYNC; + default: return 0; + } +} + +static inline uint32_t lm2ppdone(struct mdp5_hw_mixer *mixer) +{ + return MDP5_IRQ_PING_PONG_0_DONE << mixer->pp; +} + +void mdp5_set_irqmask(struct mdp_kms *mdp_kms, uint32_t irqmask, + uint32_t old_irqmask); +void mdp5_irq_preinstall(struct msm_kms *kms); +int mdp5_irq_postinstall(struct msm_kms *kms); +void mdp5_irq_uninstall(struct msm_kms *kms); +irqreturn_t mdp5_irq(struct msm_kms *kms); +int mdp5_enable_vblank(struct msm_kms *kms, struct drm_crtc *crtc); +void mdp5_disable_vblank(struct msm_kms *kms, struct drm_crtc *crtc); +int mdp5_irq_domain_init(struct mdp5_kms *mdp5_kms); +void mdp5_irq_domain_fini(struct mdp5_kms *mdp5_kms); + +uint32_t mdp5_plane_get_flush(struct drm_plane *plane); +enum mdp5_pipe mdp5_plane_pipe(struct drm_plane *plane); +enum mdp5_pipe mdp5_plane_right_pipe(struct drm_plane *plane); +struct drm_plane *mdp5_plane_init(struct drm_device *dev, + enum drm_plane_type type); + +struct mdp5_ctl *mdp5_crtc_get_ctl(struct drm_crtc *crtc); +uint32_t mdp5_crtc_vblank(struct drm_crtc *crtc); + +struct mdp5_hw_mixer *mdp5_crtc_get_mixer(struct drm_crtc *crtc); +struct mdp5_pipeline *mdp5_crtc_get_pipeline(struct drm_crtc *crtc); +void mdp5_crtc_set_pipeline(struct drm_crtc *crtc); +void mdp5_crtc_wait_for_commit_done(struct drm_crtc *crtc); +struct drm_crtc *mdp5_crtc_init(struct drm_device *dev, + struct drm_plane *plane, + struct drm_plane *cursor_plane, int id); + +struct drm_encoder *mdp5_encoder_init(struct drm_device *dev, + struct mdp5_interface *intf, struct mdp5_ctl *ctl); +void mdp5_encoder_set_intf_mode(struct drm_encoder *encoder, bool cmd_mode); +int mdp5_encoder_get_linecount(struct drm_encoder *encoder); +u32 mdp5_encoder_get_framecount(struct drm_encoder *encoder); + +#ifdef CONFIG_DRM_MSM_DSI +void mdp5_cmd_encoder_mode_set(struct drm_encoder *encoder, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode); +void mdp5_cmd_encoder_disable(struct drm_encoder *encoder); +void mdp5_cmd_encoder_enable(struct drm_encoder *encoder); +#else +static inline void mdp5_cmd_encoder_mode_set(struct drm_encoder *encoder, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ +} +static inline void mdp5_cmd_encoder_disable(struct drm_encoder *encoder) +{ +} +static inline void mdp5_cmd_encoder_enable(struct drm_encoder *encoder) +{ +} +#endif + +#endif /* __MDP5_KMS_H__ */ diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_mixer.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_mixer.c new file mode 100644 index 000000000..253b29629 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_mixer.c @@ -0,0 +1,164 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2017 The Linux Foundation. All rights reserved. + */ + +#include "mdp5_kms.h" + +/* + * As of now, there are only 2 combinations possible for source split: + * + * Left | Right + * -----|------ + * LM0 | LM1 + * LM2 | LM5 + * + */ +static int lm_right_pair[] = { 1, -1, 5, -1, -1, -1 }; + +static int get_right_pair_idx(struct mdp5_kms *mdp5_kms, int lm) +{ + int i; + int pair_lm; + + pair_lm = lm_right_pair[lm]; + if (pair_lm < 0) + return -EINVAL; + + for (i = 0; i < mdp5_kms->num_hwmixers; i++) { + struct mdp5_hw_mixer *mixer = mdp5_kms->hwmixers[i]; + + if (mixer->lm == pair_lm) + return mixer->idx; + } + + return -1; +} + +int mdp5_mixer_assign(struct drm_atomic_commit *s, struct drm_crtc *crtc, + uint32_t caps, struct mdp5_hw_mixer **mixer, + struct mdp5_hw_mixer **r_mixer) +{ + struct msm_drm_private *priv = s->dev->dev_private; + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(priv->kms)); + struct mdp5_global_state *global_state = mdp5_get_global_state(s); + struct mdp5_hw_mixer_state *new_state; + int i; + + if (IS_ERR(global_state)) + return PTR_ERR(global_state); + + new_state = &global_state->hwmixer; + + for (i = 0; i < mdp5_kms->num_hwmixers; i++) { + struct mdp5_hw_mixer *cur = mdp5_kms->hwmixers[i]; + + /* + * skip if already in-use by a different CRTC. If there is a + * mixer already assigned to this CRTC, it means this call is + * a request to get an additional right mixer. Assume that the + * existing mixer is the 'left' one, and try to see if we can + * get its corresponding 'right' pair. + */ + if (new_state->hwmixer_to_crtc[cur->idx] && + new_state->hwmixer_to_crtc[cur->idx] != crtc) + continue; + + /* skip if doesn't support some required caps: */ + if (caps & ~cur->caps) + continue; + + if (r_mixer) { + int pair_idx; + + pair_idx = get_right_pair_idx(mdp5_kms, cur->lm); + if (pair_idx < 0) + return -EINVAL; + + if (new_state->hwmixer_to_crtc[pair_idx]) + continue; + + *r_mixer = mdp5_kms->hwmixers[pair_idx]; + } + + /* + * prefer a pair-able LM over an unpairable one. We can + * switch the CRTC from Normal mode to Source Split mode + * without requiring a full modeset if we had already + * assigned this CRTC a pair-able LM. + * + * TODO: There will be assignment sequences which would + * result in the CRTC requiring a full modeset, even + * if we have the LM resources to prevent it. For a platform + * with a few displays, we don't run out of pair-able LMs + * so easily. For now, ignore the possibility of requiring + * a full modeset. + */ + if (!(*mixer) || cur->caps & MDP_LM_CAP_PAIR) + *mixer = cur; + } + + if (!(*mixer)) + return -ENOMEM; + + if (r_mixer && !(*r_mixer)) + return -ENOMEM; + + DBG("assigning Layer Mixer %d to crtc %s", (*mixer)->lm, crtc->name); + + new_state->hwmixer_to_crtc[(*mixer)->idx] = crtc; + if (r_mixer) { + DBG("assigning Right Layer Mixer %d to crtc %s", (*r_mixer)->lm, + crtc->name); + new_state->hwmixer_to_crtc[(*r_mixer)->idx] = crtc; + } + + return 0; +} + +int mdp5_mixer_release(struct drm_atomic_commit *s, struct mdp5_hw_mixer *mixer) +{ + struct mdp5_global_state *global_state = mdp5_get_global_state(s); + struct mdp5_hw_mixer_state *new_state; + + if (!mixer) + return 0; + + if (IS_ERR(global_state)) + return PTR_ERR(global_state); + + new_state = &global_state->hwmixer; + + if (WARN_ON(!new_state->hwmixer_to_crtc[mixer->idx])) + return -EINVAL; + + DBG("%s: release from crtc %s", mixer->name, + new_state->hwmixer_to_crtc[mixer->idx]->name); + + new_state->hwmixer_to_crtc[mixer->idx] = NULL; + + return 0; +} + +static const char * const mixer_names[] = { + "LM0", "LM1", "LM2", "LM3", "LM4", "LM5", +}; + +struct mdp5_hw_mixer *mdp5_mixer_init(struct drm_device *dev, + const struct mdp5_lm_instance *lm) +{ + struct mdp5_hw_mixer *mixer; + + mixer = devm_kzalloc(dev->dev, sizeof(*mixer), GFP_KERNEL); + if (!mixer) + return ERR_PTR(-ENOMEM); + + mixer->name = mixer_names[lm->id]; + mixer->lm = lm->id; + mixer->caps = lm->caps; + mixer->pp = lm->pp; + mixer->dspp = lm->dspp; + mixer->flush_mask = mdp_ctl_flush_mask_lm(lm->id); + + return mixer; +} diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_mixer.h b/drivers/gpu/drm/msm/disp/mdp5/mdp5_mixer.h new file mode 100644 index 000000000..46bca461b --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_mixer.h @@ -0,0 +1,36 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2017 The Linux Foundation. All rights reserved. + */ + +#ifndef __MDP5_LM_H__ +#define __MDP5_LM_H__ + +/* represents a hw Layer Mixer, one (or more) is dynamically assigned to a crtc */ +struct mdp5_hw_mixer { + int idx; + + const char *name; + + int lm; /* the LM instance # */ + uint32_t caps; + int pp; + int dspp; + + uint32_t flush_mask; /* used to commit LM registers */ +}; + +/* global atomic state of assignment between CRTCs and Layer Mixers: */ +struct mdp5_hw_mixer_state { + struct drm_crtc *hwmixer_to_crtc[8]; +}; + +struct mdp5_hw_mixer *mdp5_mixer_init(struct drm_device *dev, + const struct mdp5_lm_instance *lm); +int mdp5_mixer_assign(struct drm_atomic_commit *s, struct drm_crtc *crtc, + uint32_t caps, struct mdp5_hw_mixer **mixer, + struct mdp5_hw_mixer **r_mixer); +int mdp5_mixer_release(struct drm_atomic_commit *s, + struct mdp5_hw_mixer *mixer); + +#endif /* __MDP5_LM_H__ */ diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_pipe.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_pipe.c new file mode 100644 index 000000000..606e18973 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_pipe.c @@ -0,0 +1,171 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2016 Red Hat + * Author: Rob Clark + */ + +#include "mdp5_kms.h" + +int mdp5_pipe_assign(struct drm_atomic_commit *s, struct drm_plane *plane, + uint32_t caps, uint32_t blkcfg, + struct mdp5_hw_pipe **hwpipe, + struct mdp5_hw_pipe **r_hwpipe) +{ + struct msm_drm_private *priv = s->dev->dev_private; + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(priv->kms)); + struct mdp5_global_state *new_global_state, *old_global_state; + struct mdp5_hw_pipe_state *old_state, *new_state; + int i, j; + + new_global_state = mdp5_get_global_state(s); + if (IS_ERR(new_global_state)) + return PTR_ERR(new_global_state); + + /* grab old_state after mdp5_get_global_state(), since now we hold lock: */ + old_global_state = mdp5_get_existing_global_state(mdp5_kms); + + old_state = &old_global_state->hwpipe; + new_state = &new_global_state->hwpipe; + + for (i = 0; i < mdp5_kms->num_hwpipes; i++) { + struct mdp5_hw_pipe *cur = mdp5_kms->hwpipes[i]; + + /* skip if already in-use.. check both new and old state, + * since we cannot immediately re-use a pipe that is + * released in the current update in some cases: + * (1) mdp5 can have SMP (non-double-buffered) + * (2) hw pipe previously assigned to different CRTC + * (vblanks might not be aligned) + */ + if (new_state->hwpipe_to_plane[cur->idx] || + old_state->hwpipe_to_plane[cur->idx]) + continue; + + /* skip if doesn't support some required caps: */ + if (caps & ~cur->caps) + continue; + + /* + * don't assign a cursor pipe to a plane that isn't going to + * be used as a cursor + */ + if (cur->caps & MDP_PIPE_CAP_CURSOR && + plane->type != DRM_PLANE_TYPE_CURSOR) + continue; + + /* possible candidate, take the one with the + * fewest unneeded caps bits set: + */ + if (!(*hwpipe) || (hweight_long(cur->caps & ~caps) < + hweight_long((*hwpipe)->caps & ~caps))) { + bool r_found = false; + + if (r_hwpipe) { + for (j = i + 1; j < mdp5_kms->num_hwpipes; + j++) { + struct mdp5_hw_pipe *r_cur = + mdp5_kms->hwpipes[j]; + + /* reject different types of hwpipes */ + if (r_cur->caps != cur->caps) + continue; + + /* respect priority, eg. VIG0 > VIG1 */ + if (cur->pipe > r_cur->pipe) + continue; + + *r_hwpipe = r_cur; + r_found = true; + break; + } + } + + if (!r_hwpipe || r_found) + *hwpipe = cur; + } + } + + if (!(*hwpipe)) + return -ENOMEM; + + if (r_hwpipe && !(*r_hwpipe)) + return -ENOMEM; + + if (mdp5_kms->smp) { + int ret; + + /* We don't support SMP and 2 hwpipes/plane together */ + WARN_ON(r_hwpipe); + + DBG("%s: alloc SMP blocks", (*hwpipe)->name); + ret = mdp5_smp_assign(mdp5_kms->smp, &new_global_state->smp, + (*hwpipe)->pipe, blkcfg); + if (ret) + return -ENOMEM; + + (*hwpipe)->blkcfg = blkcfg; + } + + DBG("%s: assign to plane %s for caps %x", + (*hwpipe)->name, plane->name, caps); + new_state->hwpipe_to_plane[(*hwpipe)->idx] = plane; + + if (r_hwpipe) { + DBG("%s: assign to right of plane %s for caps %x", + (*r_hwpipe)->name, plane->name, caps); + new_state->hwpipe_to_plane[(*r_hwpipe)->idx] = plane; + } + + return 0; +} + +int mdp5_pipe_release(struct drm_atomic_commit *s, struct mdp5_hw_pipe *hwpipe) +{ + struct msm_drm_private *priv = s->dev->dev_private; + struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(priv->kms)); + struct mdp5_global_state *state; + struct mdp5_hw_pipe_state *new_state; + + if (!hwpipe) + return 0; + + state = mdp5_get_global_state(s); + if (IS_ERR(state)) + return PTR_ERR(state); + + new_state = &state->hwpipe; + + if (WARN_ON(!new_state->hwpipe_to_plane[hwpipe->idx])) + return -EINVAL; + + DBG("%s: release from plane %s", hwpipe->name, + new_state->hwpipe_to_plane[hwpipe->idx]->name); + + if (mdp5_kms->smp) { + DBG("%s: free SMP blocks", hwpipe->name); + mdp5_smp_release(mdp5_kms->smp, &state->smp, hwpipe->pipe); + } + + new_state->hwpipe_to_plane[hwpipe->idx] = NULL; + + return 0; +} + +struct mdp5_hw_pipe *mdp5_pipe_init(struct drm_device *dev, + enum mdp5_pipe pipe, + uint32_t reg_offset, uint32_t caps) +{ + struct mdp5_hw_pipe *hwpipe; + + hwpipe = devm_kzalloc(dev->dev, sizeof(*hwpipe), GFP_KERNEL); + if (!hwpipe) + return ERR_PTR(-ENOMEM); + + hwpipe->name = pipe2name(pipe); + hwpipe->pipe = pipe; + hwpipe->reg_offset = reg_offset; + hwpipe->caps = caps; + hwpipe->flush_mask = mdp_ctl_flush_mask_pipe(pipe); + + return hwpipe; +} diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_pipe.h b/drivers/gpu/drm/msm/disp/mdp5/mdp5_pipe.h new file mode 100644 index 000000000..37485d2fd --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_pipe.h @@ -0,0 +1,46 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2016 Red Hat + * Author: Rob Clark + */ + +#ifndef __MDP5_PIPE_H__ +#define __MDP5_PIPE_H__ + +/* TODO: Add SSPP_MAX in mdp5.xml.h */ +#define SSPP_MAX (SSPP_CURSOR1 + 1) + +/* represents a hw pipe, which is dynamically assigned to a plane */ +struct mdp5_hw_pipe { + int idx; + + const char *name; + enum mdp5_pipe pipe; + + uint32_t reg_offset; + uint32_t caps; + + uint32_t flush_mask; /* used to commit pipe registers */ + + /* number of smp blocks per plane, ie: + * nblks_y | (nblks_u << 8) | (nblks_v << 16) + */ + uint32_t blkcfg; +}; + +/* global atomic state of assignment between pipes and planes: */ +struct mdp5_hw_pipe_state { + struct drm_plane *hwpipe_to_plane[SSPP_MAX]; +}; + +int mdp5_pipe_assign(struct drm_atomic_commit *s, struct drm_plane *plane, + uint32_t caps, uint32_t blkcfg, + struct mdp5_hw_pipe **hwpipe, + struct mdp5_hw_pipe **r_hwpipe); +int mdp5_pipe_release(struct drm_atomic_commit *s, struct mdp5_hw_pipe *hwpipe); + +struct mdp5_hw_pipe *mdp5_pipe_init(struct drm_device *dev, + enum mdp5_pipe pipe, + uint32_t reg_offset, uint32_t caps); + +#endif /* __MDP5_PIPE_H__ */ diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c new file mode 100644 index 000000000..841f444a8 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c @@ -0,0 +1,1040 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2014-2015 The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#include +#include +#include +#include +#include +#include +#include + +#include "mdp5_kms.h" + +struct mdp5_plane { + struct drm_plane base; +}; +#define to_mdp5_plane(x) container_of(x, struct mdp5_plane, base) + +static int mdp5_plane_mode_set(struct drm_plane *plane, + struct drm_crtc *crtc, struct drm_framebuffer *fb, + struct drm_rect *src, struct drm_rect *dest); + +static struct mdp5_kms *get_kms(struct drm_plane *plane) +{ + struct msm_drm_private *priv = plane->dev->dev_private; + return to_mdp5_kms(to_mdp_kms(priv->kms)); +} + +static bool plane_enabled(struct drm_plane_state *state) +{ + return state->visible; +} + +/* helper to install properties which are common to planes and crtcs */ +static void mdp5_plane_install_properties(struct drm_plane *plane, + struct drm_mode_object *obj) +{ + unsigned int zpos; + + drm_plane_create_rotation_property(plane, + DRM_MODE_ROTATE_0, + DRM_MODE_ROTATE_0 | + DRM_MODE_ROTATE_180 | + DRM_MODE_REFLECT_X | + DRM_MODE_REFLECT_Y); + drm_plane_create_alpha_property(plane); + drm_plane_create_blend_mode_property(plane, + BIT(DRM_MODE_BLEND_PIXEL_NONE) | + BIT(DRM_MODE_BLEND_PREMULTI) | + BIT(DRM_MODE_BLEND_COVERAGE)); + + if (plane->type == DRM_PLANE_TYPE_PRIMARY) + zpos = STAGE_BASE; + else + zpos = STAGE0 + drm_plane_index(plane); + drm_plane_create_zpos_property(plane, zpos, 1, 255); +} + +static void +mdp5_plane_atomic_print_state(struct drm_printer *p, + const struct drm_plane_state *state) +{ + struct mdp5_plane_state *pstate = to_mdp5_plane_state(state); + struct mdp5_kms *mdp5_kms = get_kms(state->plane); + + drm_printf(p, "\thwpipe=%s\n", pstate->hwpipe ? + pstate->hwpipe->name : "(null)"); + if (mdp5_kms->caps & MDP_CAP_SRC_SPLIT) + drm_printf(p, "\tright-hwpipe=%s\n", + pstate->r_hwpipe ? pstate->r_hwpipe->name : + "(null)"); + drm_printf(p, "\tblend_mode=%u\n", pstate->base.pixel_blend_mode); + drm_printf(p, "\tzpos=%u\n", pstate->base.zpos); + drm_printf(p, "\tnormalized_zpos=%u\n", pstate->base.normalized_zpos); + drm_printf(p, "\talpha=%u\n", pstate->base.alpha); + drm_printf(p, "\tstage=%s\n", stage2name(pstate->stage)); +} + +static void mdp5_plane_reset(struct drm_plane *plane) +{ + struct mdp5_plane_state *mdp5_state; + + if (plane->state) + __drm_atomic_helper_plane_destroy_state(plane->state); + + kfree(to_mdp5_plane_state(plane->state)); + plane->state = NULL; + mdp5_state = kzalloc_obj(*mdp5_state); + if (!mdp5_state) + return; + __drm_atomic_helper_plane_reset(plane, &mdp5_state->base); +} + +static struct drm_plane_state * +mdp5_plane_duplicate_state(struct drm_plane *plane) +{ + struct mdp5_plane_state *mdp5_state; + + if (WARN_ON(!plane->state)) + return NULL; + + mdp5_state = kmemdup(to_mdp5_plane_state(plane->state), + sizeof(*mdp5_state), GFP_KERNEL); + if (!mdp5_state) + return NULL; + + __drm_atomic_helper_plane_duplicate_state(plane, &mdp5_state->base); + + return &mdp5_state->base; +} + +static void mdp5_plane_destroy_state(struct drm_plane *plane, + struct drm_plane_state *state) +{ + struct mdp5_plane_state *pstate = to_mdp5_plane_state(state); + + __drm_atomic_helper_plane_destroy_state(state); + + kfree(pstate); +} + +static const struct drm_plane_funcs mdp5_plane_funcs = { + .update_plane = drm_atomic_helper_update_plane, + .disable_plane = drm_atomic_helper_disable_plane, + .reset = mdp5_plane_reset, + .atomic_duplicate_state = mdp5_plane_duplicate_state, + .atomic_destroy_state = mdp5_plane_destroy_state, + .atomic_print_state = mdp5_plane_atomic_print_state, +}; + +static int mdp5_plane_prepare_fb(struct drm_plane *plane, + struct drm_plane_state *new_state) +{ + bool needs_dirtyfb = to_mdp5_plane_state(new_state)->needs_dirtyfb; + + if (!new_state->fb) + return 0; + + drm_gem_plane_helper_prepare_fb(plane, new_state); + + return msm_framebuffer_prepare(new_state->fb, needs_dirtyfb); +} + +static void mdp5_plane_cleanup_fb(struct drm_plane *plane, + struct drm_plane_state *old_state) +{ + struct drm_framebuffer *fb = old_state->fb; + bool needed_dirtyfb = to_mdp5_plane_state(old_state)->needs_dirtyfb; + + if (!fb) + return; + + DBG("%s: cleanup: FB[%u]", plane->name, fb->base.id); + msm_framebuffer_cleanup(fb, needed_dirtyfb); +} + +static int mdp5_plane_atomic_check_with_state(struct drm_crtc_state *crtc_state, + struct drm_plane_state *state) +{ + struct mdp5_plane_state *mdp5_state = to_mdp5_plane_state(state); + struct drm_plane *plane = state->plane; + struct drm_plane_state *old_state = plane->state; + struct mdp5_cfg *config = mdp5_cfg_get_config(get_kms(plane)->cfg); + bool new_hwpipe = false; + bool need_right_hwpipe = false; + uint32_t max_width, max_height; + bool out_of_bounds = false; + uint32_t caps = 0; + int min_scale, max_scale; + int ret; + + DBG("%s: check (%d -> %d)", plane->name, + plane_enabled(old_state), plane_enabled(state)); + + max_width = config->hw->lm.max_width << 16; + max_height = config->hw->lm.max_height << 16; + + /* Make sure source dimensions are within bounds. */ + if (state->src_h > max_height) + out_of_bounds = true; + + if (state->src_w > max_width) { + /* If source split is supported, we can go up to 2x + * the max LM width, but we'd need to stage another + * hwpipe to the right LM. So, the drm_plane would + * consist of 2 hwpipes. + */ + if (config->hw->mdp.caps & MDP_CAP_SRC_SPLIT && + (state->src_w <= 2 * max_width)) + need_right_hwpipe = true; + else + out_of_bounds = true; + } + + if (out_of_bounds) { + struct drm_rect src = drm_plane_state_src(state); + DBG("Invalid source size "DRM_RECT_FP_FMT, + DRM_RECT_FP_ARG(&src)); + return -ERANGE; + } + + min_scale = FRAC_16_16(1, 8); + max_scale = FRAC_16_16(8, 1); + + ret = drm_atomic_helper_check_plane_state(state, crtc_state, + min_scale, max_scale, + true, true); + if (ret) + return ret; + + if (plane_enabled(state)) { + unsigned int rotation; + const struct msm_format *format; + struct mdp5_kms *mdp5_kms = get_kms(plane); + uint32_t blkcfg = 0; + + format = msm_framebuffer_format(state->fb); + if (MSM_FORMAT_IS_YUV(format)) + caps |= MDP_PIPE_CAP_SCALE | MDP_PIPE_CAP_CSC; + + if (((state->src_w >> 16) != state->crtc_w) || + ((state->src_h >> 16) != state->crtc_h)) + caps |= MDP_PIPE_CAP_SCALE; + + rotation = drm_rotation_simplify(state->rotation, + DRM_MODE_ROTATE_0 | + DRM_MODE_REFLECT_X | + DRM_MODE_REFLECT_Y); + + if (rotation & DRM_MODE_REFLECT_X) + caps |= MDP_PIPE_CAP_HFLIP; + + if (rotation & DRM_MODE_REFLECT_Y) + caps |= MDP_PIPE_CAP_VFLIP; + + if (plane->type == DRM_PLANE_TYPE_CURSOR) + caps |= MDP_PIPE_CAP_CURSOR; + + /* (re)allocate hw pipe if we don't have one or caps-mismatch: */ + if (!mdp5_state->hwpipe || (caps & ~mdp5_state->hwpipe->caps)) + new_hwpipe = true; + + /* + * (re)allocte hw pipe if we're either requesting for 2 hw pipes + * or we're switching from 2 hw pipes to 1 hw pipe because the + * new src_w can be supported by 1 hw pipe itself. + */ + if ((need_right_hwpipe && !mdp5_state->r_hwpipe) || + (!need_right_hwpipe && mdp5_state->r_hwpipe)) + new_hwpipe = true; + + if (mdp5_kms->smp) { + const struct msm_format *format = + msm_framebuffer_format(state->fb); + + blkcfg = mdp5_smp_calculate(mdp5_kms->smp, format, + state->src_w >> 16, false); + + if (mdp5_state->hwpipe && (mdp5_state->hwpipe->blkcfg != blkcfg)) + new_hwpipe = true; + } + + /* (re)assign hwpipe if needed, otherwise keep old one: */ + if (new_hwpipe) { + /* TODO maybe we want to re-assign hwpipe sometimes + * in cases when we no-longer need some caps to make + * it available for other planes? + */ + struct mdp5_hw_pipe *old_hwpipe = mdp5_state->hwpipe; + struct mdp5_hw_pipe *old_right_hwpipe = + mdp5_state->r_hwpipe; + struct mdp5_hw_pipe *new_hwpipe = NULL; + struct mdp5_hw_pipe *new_right_hwpipe = NULL; + + ret = mdp5_pipe_assign(state->state, plane, caps, + blkcfg, &new_hwpipe, + need_right_hwpipe ? + &new_right_hwpipe : NULL); + if (ret) { + DBG("%s: failed to assign hwpipe(s)!", + plane->name); + return ret; + } + + mdp5_state->hwpipe = new_hwpipe; + if (need_right_hwpipe) + mdp5_state->r_hwpipe = new_right_hwpipe; + else + /* + * set it to NULL so that the driver knows we + * don't have a right hwpipe when committing a + * new state + */ + mdp5_state->r_hwpipe = NULL; + + + ret = mdp5_pipe_release(state->state, old_hwpipe); + if (ret) + return ret; + + ret = mdp5_pipe_release(state->state, old_right_hwpipe); + if (ret) + return ret; + + } + } else { + ret = mdp5_pipe_release(state->state, mdp5_state->hwpipe); + if (ret) + return ret; + + ret = mdp5_pipe_release(state->state, mdp5_state->r_hwpipe); + if (ret) + return ret; + + mdp5_state->hwpipe = mdp5_state->r_hwpipe = NULL; + } + + return 0; +} + +static int mdp5_plane_atomic_check(struct drm_plane *plane, + struct drm_atomic_commit *state) +{ + struct drm_plane_state *old_plane_state = drm_atomic_get_old_plane_state(state, + plane); + struct drm_plane_state *new_plane_state = drm_atomic_get_new_plane_state(state, + plane); + struct drm_crtc *crtc; + struct drm_crtc_state *crtc_state; + + crtc = new_plane_state->crtc ? new_plane_state->crtc : old_plane_state->crtc; + if (!crtc) + return 0; + + crtc_state = drm_atomic_get_new_crtc_state(state, crtc); + if (WARN_ON(!crtc_state)) + return -EINVAL; + + return mdp5_plane_atomic_check_with_state(crtc_state, new_plane_state); +} + +static void mdp5_plane_atomic_update(struct drm_plane *plane, + struct drm_atomic_commit *state) +{ + struct drm_plane_state *new_state = drm_atomic_get_new_plane_state(state, + plane); + + DBG("%s: update", plane->name); + + if (plane_enabled(new_state)) { + int ret; + + ret = mdp5_plane_mode_set(plane, + new_state->crtc, new_state->fb, + &new_state->src, &new_state->dst); + /* atomic_check should have ensured that this doesn't fail */ + WARN_ON(ret < 0); + } +} + +static int mdp5_plane_atomic_async_check(struct drm_plane *plane, + struct drm_atomic_commit *state, bool flip) +{ + struct drm_plane_state *new_plane_state = drm_atomic_get_new_plane_state(state, + plane); + struct mdp5_plane_state *mdp5_state = to_mdp5_plane_state(new_plane_state); + struct drm_crtc_state *crtc_state; + int min_scale, max_scale; + int ret; + + crtc_state = drm_atomic_get_new_crtc_state(state, + new_plane_state->crtc); + if (WARN_ON(!crtc_state)) + return -EINVAL; + + if (!crtc_state->active) + return -EINVAL; + + /* don't use fast path if we don't have a hwpipe allocated yet */ + if (!mdp5_state->hwpipe) + return -EINVAL; + + /* only allow changing of position(crtc x/y or src x/y) in fast path */ + if (plane->state->crtc != new_plane_state->crtc || + plane->state->src_w != new_plane_state->src_w || + plane->state->src_h != new_plane_state->src_h || + plane->state->crtc_w != new_plane_state->crtc_w || + plane->state->crtc_h != new_plane_state->crtc_h || + !plane->state->fb || + plane->state->fb != new_plane_state->fb) + return -EINVAL; + + min_scale = FRAC_16_16(1, 8); + max_scale = FRAC_16_16(8, 1); + + ret = drm_atomic_helper_check_plane_state(new_plane_state, crtc_state, + min_scale, max_scale, + true, true); + if (ret) + return ret; + + /* + * if the visibility of the plane changes (i.e, if the cursor is + * clipped out completely, we can't take the async path because + * we need to stage/unstage the plane from the Layer Mixer(s). We + * also assign/unassign the hwpipe(s) tied to the plane. We avoid + * taking the fast path for both these reasons. + */ + if (new_plane_state->visible != plane->state->visible) + return -EINVAL; + + return 0; +} + +static void mdp5_plane_atomic_async_update(struct drm_plane *plane, + struct drm_atomic_commit *state) +{ + struct drm_plane_state *new_state = drm_atomic_get_new_plane_state(state, + plane); + struct drm_framebuffer *old_fb = plane->state->fb; + + plane->state->src_x = new_state->src_x; + plane->state->src_y = new_state->src_y; + plane->state->crtc_x = new_state->crtc_x; + plane->state->crtc_y = new_state->crtc_y; + + if (plane_enabled(new_state)) { + struct mdp5_ctl *ctl; + struct mdp5_pipeline *pipeline = + mdp5_crtc_get_pipeline(new_state->crtc); + int ret; + + ret = mdp5_plane_mode_set(plane, new_state->crtc, new_state->fb, + &new_state->src, &new_state->dst); + WARN_ON(ret < 0); + + ctl = mdp5_crtc_get_ctl(new_state->crtc); + + mdp5_ctl_commit(ctl, pipeline, mdp5_plane_get_flush(plane), true); + } + + *to_mdp5_plane_state(plane->state) = + *to_mdp5_plane_state(new_state); + + new_state->fb = old_fb; +} + +static const struct drm_plane_helper_funcs mdp5_plane_helper_funcs = { + .prepare_fb = mdp5_plane_prepare_fb, + .cleanup_fb = mdp5_plane_cleanup_fb, + .atomic_check = mdp5_plane_atomic_check, + .atomic_update = mdp5_plane_atomic_update, + .atomic_async_check = mdp5_plane_atomic_async_check, + .atomic_async_update = mdp5_plane_atomic_async_update, +}; + +static void set_scanout_locked(struct mdp5_kms *mdp5_kms, + enum mdp5_pipe pipe, + struct drm_framebuffer *fb) +{ + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC_STRIDE_A(pipe), + MDP5_PIPE_SRC_STRIDE_A_P0(fb->pitches[0]) | + MDP5_PIPE_SRC_STRIDE_A_P1(fb->pitches[1])); + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC_STRIDE_B(pipe), + MDP5_PIPE_SRC_STRIDE_B_P2(fb->pitches[2]) | + MDP5_PIPE_SRC_STRIDE_B_P3(fb->pitches[3])); + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC0_ADDR(pipe), + msm_framebuffer_iova(fb, 0)); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC1_ADDR(pipe), + msm_framebuffer_iova(fb, 1)); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC2_ADDR(pipe), + msm_framebuffer_iova(fb, 2)); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC3_ADDR(pipe), + msm_framebuffer_iova(fb, 3)); +} + +/* Note: mdp5_plane->pipe_lock must be locked */ +static void csc_disable(struct mdp5_kms *mdp5_kms, enum mdp5_pipe pipe) +{ + uint32_t value = mdp5_read(mdp5_kms, REG_MDP5_PIPE_OP_MODE(pipe)) & + ~MDP5_PIPE_OP_MODE_CSC_1_EN; + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_OP_MODE(pipe), value); +} + +/* Note: mdp5_plane->pipe_lock must be locked */ +static void csc_enable(struct mdp5_kms *mdp5_kms, enum mdp5_pipe pipe, + struct csc_cfg *csc) +{ + uint32_t i, mode = 0; /* RGB, no CSC */ + uint32_t *matrix; + + if (unlikely(!csc)) + return; + + if ((csc->type == CSC_YUV2RGB) || (CSC_YUV2YUV == csc->type)) + mode |= MDP5_PIPE_OP_MODE_CSC_SRC_DATA_FORMAT(DATA_FORMAT_YUV); + if ((csc->type == CSC_RGB2YUV) || (CSC_YUV2YUV == csc->type)) + mode |= MDP5_PIPE_OP_MODE_CSC_DST_DATA_FORMAT(DATA_FORMAT_YUV); + mode |= MDP5_PIPE_OP_MODE_CSC_1_EN; + mdp5_write(mdp5_kms, REG_MDP5_PIPE_OP_MODE(pipe), mode); + + matrix = csc->matrix; + mdp5_write(mdp5_kms, REG_MDP5_PIPE_CSC_1_MATRIX_COEFF_0(pipe), + MDP5_PIPE_CSC_1_MATRIX_COEFF_0_COEFF_11(matrix[0]) | + MDP5_PIPE_CSC_1_MATRIX_COEFF_0_COEFF_12(matrix[1])); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_CSC_1_MATRIX_COEFF_1(pipe), + MDP5_PIPE_CSC_1_MATRIX_COEFF_1_COEFF_13(matrix[2]) | + MDP5_PIPE_CSC_1_MATRIX_COEFF_1_COEFF_21(matrix[3])); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_CSC_1_MATRIX_COEFF_2(pipe), + MDP5_PIPE_CSC_1_MATRIX_COEFF_2_COEFF_22(matrix[4]) | + MDP5_PIPE_CSC_1_MATRIX_COEFF_2_COEFF_23(matrix[5])); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_CSC_1_MATRIX_COEFF_3(pipe), + MDP5_PIPE_CSC_1_MATRIX_COEFF_3_COEFF_31(matrix[6]) | + MDP5_PIPE_CSC_1_MATRIX_COEFF_3_COEFF_32(matrix[7])); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_CSC_1_MATRIX_COEFF_4(pipe), + MDP5_PIPE_CSC_1_MATRIX_COEFF_4_COEFF_33(matrix[8])); + + for (i = 0; i < ARRAY_SIZE(csc->pre_bias); i++) { + uint32_t *pre_clamp = csc->pre_clamp; + uint32_t *post_clamp = csc->post_clamp; + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_CSC_1_PRE_CLAMP(pipe, i), + MDP5_PIPE_CSC_1_PRE_CLAMP_REG_HIGH(pre_clamp[2*i+1]) | + MDP5_PIPE_CSC_1_PRE_CLAMP_REG_LOW(pre_clamp[2*i])); + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_CSC_1_POST_CLAMP(pipe, i), + MDP5_PIPE_CSC_1_POST_CLAMP_REG_HIGH(post_clamp[2*i+1]) | + MDP5_PIPE_CSC_1_POST_CLAMP_REG_LOW(post_clamp[2*i])); + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_CSC_1_PRE_BIAS(pipe, i), + MDP5_PIPE_CSC_1_PRE_BIAS_REG_VALUE(csc->pre_bias[i])); + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_CSC_1_POST_BIAS(pipe, i), + MDP5_PIPE_CSC_1_POST_BIAS_REG_VALUE(csc->post_bias[i])); + } +} + +#define PHASE_STEP_SHIFT 21 +#define DOWN_SCALE_RATIO_MAX 32 /* 2^(26-21) */ + +static int calc_phase_step(uint32_t src, uint32_t dst, uint32_t *out_phase) +{ + uint32_t unit; + + if (src == 0 || dst == 0) + return -EINVAL; + + /* + * PHASE_STEP_X/Y is coded on 26 bits (25:0), + * where 2^21 represents the unity "1" in fixed-point hardware design. + * This leaves 5 bits for the integer part (downscale case): + * -> maximum downscale ratio = 0b1_1111 = 31 + */ + if (src > (dst * DOWN_SCALE_RATIO_MAX)) + return -EOVERFLOW; + + unit = 1 << PHASE_STEP_SHIFT; + *out_phase = mult_frac(unit, src, dst); + + return 0; +} + +static int calc_scalex_steps(struct drm_plane *plane, + uint32_t pixel_format, uint32_t src, uint32_t dest, + uint32_t phasex_steps[COMP_MAX]) +{ + const struct drm_format_info *info = drm_format_info(pixel_format); + struct mdp5_kms *mdp5_kms = get_kms(plane); + struct device *dev = mdp5_kms->dev->dev; + uint32_t phasex_step; + int ret; + + ret = calc_phase_step(src, dest, &phasex_step); + if (ret) { + DRM_DEV_ERROR(dev, "X scaling (%d->%d) failed: %d\n", src, dest, ret); + return ret; + } + + phasex_steps[COMP_0] = phasex_step; + phasex_steps[COMP_3] = phasex_step; + phasex_steps[COMP_1_2] = phasex_step / info->hsub; + + return 0; +} + +static int calc_scaley_steps(struct drm_plane *plane, + uint32_t pixel_format, uint32_t src, uint32_t dest, + uint32_t phasey_steps[COMP_MAX]) +{ + const struct drm_format_info *info = drm_format_info(pixel_format); + struct mdp5_kms *mdp5_kms = get_kms(plane); + struct device *dev = mdp5_kms->dev->dev; + uint32_t phasey_step; + int ret; + + ret = calc_phase_step(src, dest, &phasey_step); + if (ret) { + DRM_DEV_ERROR(dev, "Y scaling (%d->%d) failed: %d\n", src, dest, ret); + return ret; + } + + phasey_steps[COMP_0] = phasey_step; + phasey_steps[COMP_3] = phasey_step; + phasey_steps[COMP_1_2] = phasey_step / info->vsub; + + return 0; +} + +static uint32_t get_scale_config(const struct msm_format *format, + uint32_t src, uint32_t dst, bool horz) +{ + const struct drm_format_info *info = drm_format_info(format->pixel_format); + bool yuv = MSM_FORMAT_IS_YUV(format); + bool scaling = yuv ? true : (src != dst); + uint32_t sub; + uint32_t ya_filter, uv_filter; + + if (!scaling) + return 0; + + if (yuv) { + sub = horz ? info->hsub : info->vsub; + uv_filter = ((src / sub) <= dst) ? + SCALE_FILTER_BIL : SCALE_FILTER_PCMN; + } + ya_filter = (src <= dst) ? SCALE_FILTER_BIL : SCALE_FILTER_PCMN; + + if (horz) + return MDP5_PIPE_SCALE_CONFIG_SCALEX_EN | + MDP5_PIPE_SCALE_CONFIG_SCALEX_FILTER_COMP_0(ya_filter) | + MDP5_PIPE_SCALE_CONFIG_SCALEX_FILTER_COMP_3(ya_filter) | + COND(yuv, MDP5_PIPE_SCALE_CONFIG_SCALEX_FILTER_COMP_1_2(uv_filter)); + else + return MDP5_PIPE_SCALE_CONFIG_SCALEY_EN | + MDP5_PIPE_SCALE_CONFIG_SCALEY_FILTER_COMP_0(ya_filter) | + MDP5_PIPE_SCALE_CONFIG_SCALEY_FILTER_COMP_3(ya_filter) | + COND(yuv, MDP5_PIPE_SCALE_CONFIG_SCALEY_FILTER_COMP_1_2(uv_filter)); +} + +static void calc_pixel_ext(const struct msm_format *format, + uint32_t src, uint32_t dst, uint32_t phase_step[2], + int pix_ext_edge1[COMP_MAX], int pix_ext_edge2[COMP_MAX], + bool horz) +{ + bool scaling = MSM_FORMAT_IS_YUV(format) ? true : (src != dst); + int i; + + /* + * Note: + * We assume here that: + * 1. PCMN filter is used for downscale + * 2. bilinear filter is used for upscale + * 3. we are in a single pipe configuration + */ + + for (i = 0; i < COMP_MAX; i++) { + pix_ext_edge1[i] = 0; + pix_ext_edge2[i] = scaling ? 1 : 0; + } +} + +static void mdp5_write_pixel_ext(struct mdp5_kms *mdp5_kms, enum mdp5_pipe pipe, + const struct msm_format *format, + uint32_t src_w, int pe_left[COMP_MAX], int pe_right[COMP_MAX], + uint32_t src_h, int pe_top[COMP_MAX], int pe_bottom[COMP_MAX]) +{ + const struct drm_format_info *info = drm_format_info(format->pixel_format); + uint32_t lr, tb, req; + int i; + + for (i = 0; i < COMP_MAX; i++) { + uint32_t roi_w = src_w; + uint32_t roi_h = src_h; + + if (MSM_FORMAT_IS_YUV(format) && i == COMP_1_2) { + roi_w /= info->hsub; + roi_h /= info->vsub; + } + + lr = (pe_left[i] >= 0) ? + MDP5_PIPE_SW_PIX_EXT_LR_LEFT_RPT(pe_left[i]) : + MDP5_PIPE_SW_PIX_EXT_LR_LEFT_OVF(pe_left[i]); + + lr |= (pe_right[i] >= 0) ? + MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_RPT(pe_right[i]) : + MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_OVF(pe_right[i]); + + tb = (pe_top[i] >= 0) ? + MDP5_PIPE_SW_PIX_EXT_TB_TOP_RPT(pe_top[i]) : + MDP5_PIPE_SW_PIX_EXT_TB_TOP_OVF(pe_top[i]); + + tb |= (pe_bottom[i] >= 0) ? + MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_RPT(pe_bottom[i]) : + MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_OVF(pe_bottom[i]); + + req = MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_LEFT_RIGHT(roi_w + + pe_left[i] + pe_right[i]); + + req |= MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_TOP_BOTTOM(roi_h + + pe_top[i] + pe_bottom[i]); + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SW_PIX_EXT_LR(pipe, i), lr); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SW_PIX_EXT_TB(pipe, i), tb); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS(pipe, i), req); + + DBG("comp-%d (L/R): rpt=%d/%d, ovf=%d/%d, req=%d", i, + FIELD(lr, MDP5_PIPE_SW_PIX_EXT_LR_LEFT_RPT), + FIELD(lr, MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_RPT), + FIELD(lr, MDP5_PIPE_SW_PIX_EXT_LR_LEFT_OVF), + FIELD(lr, MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_OVF), + FIELD(req, MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_LEFT_RIGHT)); + + DBG("comp-%d (T/B): rpt=%d/%d, ovf=%d/%d, req=%d", i, + FIELD(tb, MDP5_PIPE_SW_PIX_EXT_TB_TOP_RPT), + FIELD(tb, MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_RPT), + FIELD(tb, MDP5_PIPE_SW_PIX_EXT_TB_TOP_OVF), + FIELD(tb, MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_OVF), + FIELD(req, MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_TOP_BOTTOM)); + } +} + +struct pixel_ext { + int left[COMP_MAX]; + int right[COMP_MAX]; + int top[COMP_MAX]; + int bottom[COMP_MAX]; +}; + +struct phase_step { + u32 x[COMP_MAX]; + u32 y[COMP_MAX]; +}; + +static void mdp5_hwpipe_mode_set(struct mdp5_kms *mdp5_kms, + struct mdp5_hw_pipe *hwpipe, + struct drm_framebuffer *fb, + struct phase_step *step, + struct pixel_ext *pe, + u32 scale_config, u32 hdecm, u32 vdecm, + bool hflip, bool vflip, + int crtc_x, int crtc_y, + unsigned int crtc_w, unsigned int crtc_h, + u32 src_img_w, u32 src_img_h, + u32 src_x, u32 src_y, + u32 src_w, u32 src_h) +{ + enum mdp5_pipe pipe = hwpipe->pipe; + bool has_pe = hwpipe->caps & MDP_PIPE_CAP_SW_PIX_EXT; + const struct msm_format *format = + msm_framebuffer_format(fb); + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC_IMG_SIZE(pipe), + MDP5_PIPE_SRC_IMG_SIZE_WIDTH(src_img_w) | + MDP5_PIPE_SRC_IMG_SIZE_HEIGHT(src_img_h)); + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC_SIZE(pipe), + MDP5_PIPE_SRC_SIZE_WIDTH(src_w) | + MDP5_PIPE_SRC_SIZE_HEIGHT(src_h)); + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC_XY(pipe), + MDP5_PIPE_SRC_XY_X(src_x) | + MDP5_PIPE_SRC_XY_Y(src_y)); + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_OUT_SIZE(pipe), + MDP5_PIPE_OUT_SIZE_WIDTH(crtc_w) | + MDP5_PIPE_OUT_SIZE_HEIGHT(crtc_h)); + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_OUT_XY(pipe), + MDP5_PIPE_OUT_XY_X(crtc_x) | + MDP5_PIPE_OUT_XY_Y(crtc_y)); + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC_FORMAT(pipe), + MDP5_PIPE_SRC_FORMAT_A_BPC(format->bpc_a) | + MDP5_PIPE_SRC_FORMAT_R_BPC(format->bpc_r_cr) | + MDP5_PIPE_SRC_FORMAT_G_BPC(format->bpc_g_y) | + MDP5_PIPE_SRC_FORMAT_B_BPC(format->bpc_b_cb) | + COND(format->alpha_enable, MDP5_PIPE_SRC_FORMAT_ALPHA_ENABLE) | + MDP5_PIPE_SRC_FORMAT_CPP(format->bpp - 1) | + MDP5_PIPE_SRC_FORMAT_UNPACK_COUNT(format->unpack_count - 1) | + COND(format->flags & MSM_FORMAT_FLAG_UNPACK_TIGHT, + MDP5_PIPE_SRC_FORMAT_UNPACK_TIGHT) | + MDP5_PIPE_SRC_FORMAT_FETCH_TYPE(format->fetch_type) | + MDP5_PIPE_SRC_FORMAT_CHROMA_SAMP(format->chroma_sample)); + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC_UNPACK(pipe), + MDP5_PIPE_SRC_UNPACK_ELEM0(format->element[0]) | + MDP5_PIPE_SRC_UNPACK_ELEM1(format->element[1]) | + MDP5_PIPE_SRC_UNPACK_ELEM2(format->element[2]) | + MDP5_PIPE_SRC_UNPACK_ELEM3(format->element[3])); + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC_OP_MODE(pipe), + (hflip ? MDP5_PIPE_SRC_OP_MODE_FLIP_LR : 0) | + (vflip ? MDP5_PIPE_SRC_OP_MODE_FLIP_UD : 0) | + COND(has_pe, MDP5_PIPE_SRC_OP_MODE_SW_PIX_EXT_OVERRIDE) | + MDP5_PIPE_SRC_OP_MODE_BWC(BWC_LOSSLESS)); + + /* not using secure mode: */ + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC_ADDR_SW_STATUS(pipe), 0); + + if (hwpipe->caps & MDP_PIPE_CAP_SW_PIX_EXT) + mdp5_write_pixel_ext(mdp5_kms, pipe, format, + src_w, pe->left, pe->right, + src_h, pe->top, pe->bottom); + + if (hwpipe->caps & MDP_PIPE_CAP_SCALE) { + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SCALE_PHASE_STEP_X(pipe), + step->x[COMP_0]); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SCALE_PHASE_STEP_Y(pipe), + step->y[COMP_0]); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SCALE_CR_PHASE_STEP_X(pipe), + step->x[COMP_1_2]); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SCALE_CR_PHASE_STEP_Y(pipe), + step->y[COMP_1_2]); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_DECIMATION(pipe), + MDP5_PIPE_DECIMATION_VERT(vdecm) | + MDP5_PIPE_DECIMATION_HORZ(hdecm)); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_SCALE_CONFIG(pipe), + scale_config); + } + + if (hwpipe->caps & MDP_PIPE_CAP_CSC) { + if (MSM_FORMAT_IS_YUV(format)) + csc_enable(mdp5_kms, pipe, + mdp_get_default_csc_cfg(CSC_YUV2RGB)); + else + csc_disable(mdp5_kms, pipe); + } + + set_scanout_locked(mdp5_kms, pipe, fb); +} + +static int mdp5_plane_mode_set(struct drm_plane *plane, + struct drm_crtc *crtc, struct drm_framebuffer *fb, + struct drm_rect *src, struct drm_rect *dest) +{ + struct drm_plane_state *pstate = plane->state; + struct mdp5_hw_pipe *hwpipe = to_mdp5_plane_state(pstate)->hwpipe; + struct mdp5_kms *mdp5_kms = get_kms(plane); + enum mdp5_pipe pipe = hwpipe->pipe; + struct mdp5_hw_pipe *right_hwpipe; + const struct msm_format *format; + uint32_t nplanes, config = 0; + struct phase_step step = { { 0 } }; + struct pixel_ext pe = { { 0 } }; + uint32_t hdecm = 0, vdecm = 0; + uint32_t pix_format; + unsigned int rotation; + bool vflip, hflip; + int crtc_x, crtc_y; + unsigned int crtc_w, crtc_h; + uint32_t src_x, src_y; + uint32_t src_w, src_h; + uint32_t src_img_w, src_img_h; + int ret; + + nplanes = fb->format->num_planes; + + /* bad formats should already be rejected: */ + if (WARN_ON(nplanes > pipe2nclients(pipe))) + return -EINVAL; + + format = msm_framebuffer_format(fb); + pix_format = format->pixel_format; + + src_x = src->x1; + src_y = src->y1; + src_w = drm_rect_width(src); + src_h = drm_rect_height(src); + + crtc_x = dest->x1; + crtc_y = dest->y1; + crtc_w = drm_rect_width(dest); + crtc_h = drm_rect_height(dest); + + /* src values are in Q16 fixed point, convert to integer: */ + src_x = src_x >> 16; + src_y = src_y >> 16; + src_w = src_w >> 16; + src_h = src_h >> 16; + + src_img_w = min(fb->width, src_w); + src_img_h = min(fb->height, src_h); + + DBG("%s: FB[%u] %u,%u,%u,%u -> CRTC[%u] %d,%d,%u,%u", plane->name, + fb->base.id, src_x, src_y, src_w, src_h, + crtc->base.id, crtc_x, crtc_y, crtc_w, crtc_h); + + right_hwpipe = to_mdp5_plane_state(pstate)->r_hwpipe; + if (right_hwpipe) { + /* + * if the plane comprises of 2 hw pipes, assume that the width + * is split equally across them. The only parameters that varies + * between the 2 pipes are src_x and crtc_x + */ + crtc_w /= 2; + src_w /= 2; + src_img_w /= 2; + } + + ret = calc_scalex_steps(plane, pix_format, src_w, crtc_w, step.x); + if (ret) + return ret; + + ret = calc_scaley_steps(plane, pix_format, src_h, crtc_h, step.y); + if (ret) + return ret; + + if (hwpipe->caps & MDP_PIPE_CAP_SW_PIX_EXT) { + calc_pixel_ext(format, src_w, crtc_w, step.x, + pe.left, pe.right, true); + calc_pixel_ext(format, src_h, crtc_h, step.y, + pe.top, pe.bottom, false); + } + + /* TODO calc hdecm, vdecm */ + + /* SCALE is used to both scale and up-sample chroma components */ + config |= get_scale_config(format, src_w, crtc_w, true); + config |= get_scale_config(format, src_h, crtc_h, false); + DBG("scale config = %x", config); + + rotation = drm_rotation_simplify(pstate->rotation, + DRM_MODE_ROTATE_0 | + DRM_MODE_REFLECT_X | + DRM_MODE_REFLECT_Y); + hflip = !!(rotation & DRM_MODE_REFLECT_X); + vflip = !!(rotation & DRM_MODE_REFLECT_Y); + + mdp5_hwpipe_mode_set(mdp5_kms, hwpipe, fb, &step, &pe, + config, hdecm, vdecm, hflip, vflip, + crtc_x, crtc_y, crtc_w, crtc_h, + src_img_w, src_img_h, + src_x, src_y, src_w, src_h); + if (right_hwpipe) + mdp5_hwpipe_mode_set(mdp5_kms, right_hwpipe, fb, &step, &pe, + config, hdecm, vdecm, hflip, vflip, + crtc_x + crtc_w, crtc_y, crtc_w, crtc_h, + src_img_w, src_img_h, + src_x + src_w, src_y, src_w, src_h); + + return ret; +} + +/* + * Use this func and the one below only after the atomic state has been + * successfully swapped + */ +enum mdp5_pipe mdp5_plane_pipe(struct drm_plane *plane) +{ + struct mdp5_plane_state *pstate = to_mdp5_plane_state(plane->state); + + if (WARN_ON(!pstate->hwpipe)) + return SSPP_NONE; + + return pstate->hwpipe->pipe; +} + +enum mdp5_pipe mdp5_plane_right_pipe(struct drm_plane *plane) +{ + struct mdp5_plane_state *pstate = to_mdp5_plane_state(plane->state); + + if (!pstate->r_hwpipe) + return SSPP_NONE; + + return pstate->r_hwpipe->pipe; +} + +uint32_t mdp5_plane_get_flush(struct drm_plane *plane) +{ + struct mdp5_plane_state *pstate = to_mdp5_plane_state(plane->state); + u32 mask; + + if (WARN_ON(!pstate->hwpipe)) + return 0; + + mask = pstate->hwpipe->flush_mask; + + if (pstate->r_hwpipe) + mask |= pstate->r_hwpipe->flush_mask; + + return mask; +} + +static const uint32_t mdp5_plane_formats[] = { + DRM_FORMAT_ARGB8888, + DRM_FORMAT_ABGR8888, + DRM_FORMAT_RGBA8888, + DRM_FORMAT_BGRA8888, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_RGBX8888, + DRM_FORMAT_BGRX8888, + DRM_FORMAT_RGB888, + DRM_FORMAT_BGR888, + DRM_FORMAT_RGB565, + DRM_FORMAT_BGR565, + + DRM_FORMAT_NV12, + DRM_FORMAT_NV21, + DRM_FORMAT_NV16, + DRM_FORMAT_NV61, + DRM_FORMAT_VYUY, + DRM_FORMAT_UYVY, + DRM_FORMAT_YUYV, + DRM_FORMAT_YVYU, + DRM_FORMAT_YUV420, + DRM_FORMAT_YVU420, +}; + +/* initialize plane */ +struct drm_plane *mdp5_plane_init(struct drm_device *dev, + enum drm_plane_type type) +{ + struct drm_plane *plane = NULL; + struct mdp5_plane *mdp5_plane; + + mdp5_plane = drmm_universal_plane_alloc(dev, struct mdp5_plane, base, + 0xff, &mdp5_plane_funcs, + mdp5_plane_formats, ARRAY_SIZE(mdp5_plane_formats), + NULL, type, NULL); + if (IS_ERR(mdp5_plane)) + return ERR_CAST(mdp5_plane); + + plane = &mdp5_plane->base; + + drm_plane_helper_add(plane, &mdp5_plane_helper_funcs); + + mdp5_plane_install_properties(plane, &plane->base); + + drm_plane_enable_fb_damage_clips(plane); + + return plane; +} diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.c new file mode 100644 index 000000000..890d2f315 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.c @@ -0,0 +1,383 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2014, The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#include +#include + +#include "mdp5_kms.h" +#include "mdp5_smp.h" + + +struct mdp5_smp { + struct drm_device *dev; + + uint8_t reserved[MAX_CLIENTS]; /* fixed MMBs allocation per client */ + + int blk_cnt; + int blk_size; + + /* register cache */ + u32 alloc_w[22]; + u32 alloc_r[22]; + u32 pipe_reqprio_fifo_wm0[SSPP_MAX]; + u32 pipe_reqprio_fifo_wm1[SSPP_MAX]; + u32 pipe_reqprio_fifo_wm2[SSPP_MAX]; +}; + +static inline +struct mdp5_kms *get_kms(struct mdp5_smp *smp) +{ + struct msm_drm_private *priv = smp->dev->dev_private; + + return to_mdp5_kms(to_mdp_kms(priv->kms)); +} + +static inline u32 pipe2client(enum mdp5_pipe pipe, int plane) +{ +#define CID_UNUSED 0 + + if (WARN_ON(plane >= pipe2nclients(pipe))) + return CID_UNUSED; + + /* + * Note on SMP clients: + * For ViG pipes, fetch Y/Cr/Cb-components clients are always + * consecutive, and in that order. + * + * e.g.: + * if mdp5_cfg->smp.clients[SSPP_VIG0] = N, + * Y plane's client ID is N + * Cr plane's client ID is N + 1 + * Cb plane's client ID is N + 2 + */ + + return mdp5_cfg->smp.clients[pipe] + plane; +} + +/* allocate blocks for the specified request: */ +static int smp_request_block(struct mdp5_smp *smp, + struct mdp5_smp_state *state, + u32 cid, int nblks) +{ + void *cs = state->client_state[cid]; + int i, avail, cnt = smp->blk_cnt; + uint8_t reserved; + + /* we shouldn't be requesting blocks for an in-use client: */ + WARN_ON(!bitmap_empty(cs, cnt)); + + reserved = smp->reserved[cid]; + + if (reserved) { + nblks = max(0, nblks - reserved); + DBG("%d MMBs allocated (%d reserved)", nblks, reserved); + } + + avail = cnt - bitmap_weight(state->state, cnt); + if (nblks > avail) { + DRM_DEV_ERROR(smp->dev->dev, "out of blks (req=%d > avail=%d)\n", + nblks, avail); + return -ENOSPC; + } + + for (i = 0; i < nblks; i++) { + int blk = find_first_zero_bit(state->state, cnt); + set_bit(blk, cs); + set_bit(blk, state->state); + } + + return 0; +} + +static void set_fifo_thresholds(struct mdp5_smp *smp, + enum mdp5_pipe pipe, int nblks) +{ + u32 smp_entries_per_blk = smp->blk_size / (128 / BITS_PER_BYTE); + u32 val; + + /* 1/4 of SMP pool that is being fetched */ + val = (nblks * smp_entries_per_blk) / 4; + + smp->pipe_reqprio_fifo_wm0[pipe] = val * 1; + smp->pipe_reqprio_fifo_wm1[pipe] = val * 2; + smp->pipe_reqprio_fifo_wm2[pipe] = val * 3; +} + +/* + * NOTE: looks like if horizontal decimation is used (if we supported that) + * then the width used to calculate SMP block requirements is the post- + * decimated width. Ie. SMP buffering sits downstream of decimation (which + * presumably happens during the dma from scanout buffer). + */ +uint32_t mdp5_smp_calculate(struct mdp5_smp *smp, + const struct msm_format *format, + u32 width, bool hdecim) +{ + const struct drm_format_info *info = drm_format_info(format->pixel_format); + int i, hsub, nplanes, nlines; + uint32_t blkcfg = 0; + + nplanes = info->num_planes; + hsub = info->hsub; + + /* different if BWC (compressed framebuffer?) enabled: */ + nlines = 2; + + /* Newer MDPs have split/packing logic, which fetches sub-sampled + * U and V components (splits them from Y if necessary) and packs + * them together, writes to SMP using a single client. + */ + if (format->chroma_sample > CHROMA_FULL) { + nplanes = 2; + + /* if decimation is enabled, HW decimates less on the + * sub sampled chroma components + */ + if (hdecim && (hsub > 1)) + hsub = 1; + } + + for (i = 0; i < nplanes; i++) { + int n, fetch_stride, cpp; + + cpp = info->cpp[i]; + fetch_stride = width * cpp / (i ? hsub : 1); + + n = DIV_ROUND_UP(fetch_stride * nlines, smp->blk_size); + + blkcfg |= (n << (8 * i)); + } + + return blkcfg; +} + +int mdp5_smp_assign(struct mdp5_smp *smp, struct mdp5_smp_state *state, + enum mdp5_pipe pipe, uint32_t blkcfg) +{ + struct mdp5_kms *mdp5_kms = get_kms(smp); + struct drm_device *dev = mdp5_kms->dev; + int i, ret; + + for (i = 0; i < pipe2nclients(pipe); i++) { + u32 cid = pipe2client(pipe, i); + int n = blkcfg & 0xff; + + if (!n) + continue; + + DBG("%s[%d]: request %d SMP blocks", pipe2name(pipe), i, n); + ret = smp_request_block(smp, state, cid, n); + if (ret) { + DRM_DEV_ERROR(dev->dev, "Cannot allocate %d SMP blocks: %d\n", + n, ret); + return ret; + } + + blkcfg >>= 8; + } + + state->assigned |= (1 << pipe); + + return 0; +} + +/* Release SMP blocks for all clients of the pipe */ +void mdp5_smp_release(struct mdp5_smp *smp, struct mdp5_smp_state *state, + enum mdp5_pipe pipe) +{ + int i; + int cnt = smp->blk_cnt; + + for (i = 0; i < pipe2nclients(pipe); i++) { + u32 cid = pipe2client(pipe, i); + void *cs = state->client_state[cid]; + + /* update global state: */ + bitmap_andnot(state->state, state->state, cs, cnt); + + /* clear client's state */ + bitmap_zero(cs, cnt); + } + + state->released |= (1 << pipe); +} + +/* NOTE: SMP_ALLOC_* regs are *not* double buffered, so release has to + * happen after scanout completes. + */ +static unsigned update_smp_state(struct mdp5_smp *smp, + u32 cid, mdp5_smp_state_t *assigned) +{ + int cnt = smp->blk_cnt; + unsigned nblks = 0; + u32 blk, val; + + for_each_set_bit(blk, *assigned, cnt) { + int idx = blk / 3; + int fld = blk % 3; + + val = smp->alloc_w[idx]; + + switch (fld) { + case 0: + val &= ~MDP5_SMP_ALLOC_W_REG_CLIENT0__MASK; + val |= MDP5_SMP_ALLOC_W_REG_CLIENT0(cid); + break; + case 1: + val &= ~MDP5_SMP_ALLOC_W_REG_CLIENT1__MASK; + val |= MDP5_SMP_ALLOC_W_REG_CLIENT1(cid); + break; + case 2: + val &= ~MDP5_SMP_ALLOC_W_REG_CLIENT2__MASK; + val |= MDP5_SMP_ALLOC_W_REG_CLIENT2(cid); + break; + } + + smp->alloc_w[idx] = val; + smp->alloc_r[idx] = val; + + nblks++; + } + + return nblks; +} + +static void write_smp_alloc_regs(struct mdp5_smp *smp) +{ + struct mdp5_kms *mdp5_kms = get_kms(smp); + int i, num_regs; + + num_regs = smp->blk_cnt / 3 + 1; + + for (i = 0; i < num_regs; i++) { + mdp5_write(mdp5_kms, REG_MDP5_SMP_ALLOC_W_REG(i), + smp->alloc_w[i]); + mdp5_write(mdp5_kms, REG_MDP5_SMP_ALLOC_R_REG(i), + smp->alloc_r[i]); + } +} + +static void write_smp_fifo_regs(struct mdp5_smp *smp) +{ + struct mdp5_kms *mdp5_kms = get_kms(smp); + int i; + + for (i = 0; i < mdp5_kms->num_hwpipes; i++) { + struct mdp5_hw_pipe *hwpipe = mdp5_kms->hwpipes[i]; + enum mdp5_pipe pipe = hwpipe->pipe; + + mdp5_write(mdp5_kms, REG_MDP5_PIPE_REQPRIO_FIFO_WM_0(pipe), + smp->pipe_reqprio_fifo_wm0[pipe]); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_REQPRIO_FIFO_WM_1(pipe), + smp->pipe_reqprio_fifo_wm1[pipe]); + mdp5_write(mdp5_kms, REG_MDP5_PIPE_REQPRIO_FIFO_WM_2(pipe), + smp->pipe_reqprio_fifo_wm2[pipe]); + } +} + +void mdp5_smp_prepare_commit(struct mdp5_smp *smp, struct mdp5_smp_state *state) +{ + enum mdp5_pipe pipe; + + for_each_set_bit(pipe, &state->assigned, sizeof(state->assigned) * 8) { + unsigned i, nblks = 0; + + for (i = 0; i < pipe2nclients(pipe); i++) { + u32 cid = pipe2client(pipe, i); + void *cs = state->client_state[cid]; + + nblks += update_smp_state(smp, cid, cs); + + DBG("assign %s:%u, %u blks", + pipe2name(pipe), i, nblks); + } + + set_fifo_thresholds(smp, pipe, nblks); + } + + write_smp_alloc_regs(smp); + write_smp_fifo_regs(smp); + + state->assigned = 0; +} + +void mdp5_smp_complete_commit(struct mdp5_smp *smp, struct mdp5_smp_state *state) +{ + enum mdp5_pipe pipe; + + for_each_set_bit(pipe, &state->released, sizeof(state->released) * 8) { + DBG("release %s", pipe2name(pipe)); + set_fifo_thresholds(smp, pipe, 0); + } + + write_smp_fifo_regs(smp); + + state->released = 0; +} + +void mdp5_smp_dump(struct mdp5_smp *smp, struct drm_printer *p, + struct mdp5_global_state *global_state) +{ + struct mdp5_kms *mdp5_kms = get_kms(smp); + struct mdp5_hw_pipe_state *hwpstate; + struct mdp5_smp_state *state; + int total = 0, i, j; + + drm_printf(p, "name\tinuse\tplane\n"); + drm_printf(p, "----\t-----\t-----\n"); + + /* grab these *after* we hold the state_lock */ + hwpstate = &global_state->hwpipe; + state = &global_state->smp; + + for (i = 0; i < mdp5_kms->num_hwpipes; i++) { + struct mdp5_hw_pipe *hwpipe = mdp5_kms->hwpipes[i]; + struct drm_plane *plane = hwpstate->hwpipe_to_plane[hwpipe->idx]; + enum mdp5_pipe pipe = hwpipe->pipe; + for (j = 0; j < pipe2nclients(pipe); j++) { + u32 cid = pipe2client(pipe, j); + void *cs = state->client_state[cid]; + int inuse = bitmap_weight(cs, smp->blk_cnt); + + drm_printf(p, "%s:%d\t%d\t%s\n", + pipe2name(pipe), j, inuse, + plane ? plane->name : "(null)"); + + total += inuse; + } + } + + drm_printf(p, "TOTAL:\t%d\t(of %d)\n", total, smp->blk_cnt); + drm_printf(p, "AVAIL:\t%d\n", smp->blk_cnt - + bitmap_weight(state->state, smp->blk_cnt)); +} + + +struct mdp5_smp *mdp5_smp_init(struct mdp5_kms *mdp5_kms, const struct mdp5_smp_block *cfg) +{ + struct drm_device *dev = mdp5_kms->dev; + struct mdp5_smp_state *state; + struct mdp5_global_state *global_state; + struct mdp5_smp *smp; + + smp = devm_kzalloc(dev->dev, sizeof(*smp), GFP_KERNEL); + if (unlikely(!smp)) + return ERR_PTR(-ENOMEM); + + smp->dev = mdp5_kms->dev; + smp->blk_cnt = cfg->mmb_count; + smp->blk_size = cfg->mmb_size; + + global_state = mdp5_get_existing_global_state(mdp5_kms); + state = &global_state->smp; + + /* statically tied MMBs cannot be re-allocated: */ + bitmap_copy(state->state, cfg->reserved_state, smp->blk_cnt); + memcpy(smp->reserved, cfg->reserved, sizeof(smp->reserved)); + + return smp; +} diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.h b/drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.h new file mode 100644 index 000000000..1be983238 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.h @@ -0,0 +1,88 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2014, The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#ifndef __MDP5_SMP_H__ +#define __MDP5_SMP_H__ + +#include + +#include "msm_drv.h" + +/* + * SMP - Shared Memory Pool: + * + * SMP blocks are shared between all the clients, where each plane in + * a scanout buffer is a SMP client. Ie. scanout of 3 plane I420 on + * pipe VIG0 => 3 clients: VIG0_Y, VIG0_CB, VIG0_CR. + * + * Based on the size of the attached scanout buffer, a certain # of + * blocks must be allocated to that client out of the shared pool. + * + * In some hw, some blocks are statically allocated for certain pipes + * and CANNOT be re-allocated (eg: MMB0 and MMB1 both tied to RGB0). + * + * + * Atomic SMP State: + * + * On atomic updates that modify SMP configuration, the state is cloned + * (copied) and modified. For test-only, or in cases where atomic + * update fails (or if we hit ww_mutex deadlock/backoff condition) the + * new state is simply thrown away. + * + * Because the SMP registers are not double buffered, updates are a + * two step process: + * + * 1) in _prepare_commit() we configure things (via read-modify-write) + * for the newly assigned pipes, so we don't take away blocks + * assigned to pipes that are still scanning out + * 2) in _complete_commit(), after vblank/etc, we clear things for the + * released clients, since at that point old pipes are no longer + * scanning out. + */ +struct mdp5_smp_state { + /* global state of what blocks are in use: */ + mdp5_smp_state_t state; + + /* per client state of what blocks they are using: */ + mdp5_smp_state_t client_state[MAX_CLIENTS]; + + /* assigned pipes (hw updated at _prepare_commit()): */ + unsigned long assigned; + + /* released pipes (hw updated at _complete_commit()): */ + unsigned long released; +}; + +struct mdp5_kms; +struct mdp5_smp; + +/* + * SMP module prototypes: + * mdp5_smp_init() returns a SMP @handler, + * which is then used to call the other mdp5_smp_*(handler, ...) functions. + */ + +struct mdp5_smp *mdp5_smp_init(struct mdp5_kms *mdp5_kms, + const struct mdp5_smp_block *cfg); + +struct mdp5_global_state; +void mdp5_smp_dump(struct mdp5_smp *smp, struct drm_printer *p, + struct mdp5_global_state *global_state); + +uint32_t mdp5_smp_calculate(struct mdp5_smp *smp, + const struct msm_format *format, + u32 width, bool hdecim); + +int mdp5_smp_assign(struct mdp5_smp *smp, struct mdp5_smp_state *state, + enum mdp5_pipe pipe, uint32_t blkcfg); +void mdp5_smp_release(struct mdp5_smp *smp, struct mdp5_smp_state *state, + enum mdp5_pipe pipe); + +void mdp5_smp_prepare_commit(struct mdp5_smp *smp, struct mdp5_smp_state *state); +void mdp5_smp_complete_commit(struct mdp5_smp *smp, struct mdp5_smp_state *state); + +#endif /* __MDP5_SMP_H__ */ diff --git a/drivers/gpu/drm/msm/disp/mdp_format.c b/drivers/gpu/drm/msm/disp/mdp_format.c new file mode 100644 index 000000000..33da569ea --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp_format.c @@ -0,0 +1,659 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2014 The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#include +#include + +#include "msm_drv.h" +#include "mdp_kms.h" + +static struct csc_cfg csc_convert[CSC_MAX] = { + [CSC_RGB2RGB] = { + .type = CSC_RGB2RGB, + .matrix = { + 0x0200, 0x0000, 0x0000, + 0x0000, 0x0200, 0x0000, + 0x0000, 0x0000, 0x0200 + }, + .pre_bias = { 0x0, 0x0, 0x0 }, + .post_bias = { 0x0, 0x0, 0x0 }, + .pre_clamp = { 0x0, 0xff, 0x0, 0xff, 0x0, 0xff }, + .post_clamp = { 0x0, 0xff, 0x0, 0xff, 0x0, 0xff }, + }, + [CSC_YUV2RGB] = { + .type = CSC_YUV2RGB, + .matrix = { + 0x0254, 0x0000, 0x0331, + 0x0254, 0xff37, 0xfe60, + 0x0254, 0x0409, 0x0000 + }, + .pre_bias = { 0xfff0, 0xff80, 0xff80 }, + .post_bias = { 0x00, 0x00, 0x00 }, + .pre_clamp = { 0x00, 0xff, 0x00, 0xff, 0x00, 0xff }, + .post_clamp = { 0x00, 0xff, 0x00, 0xff, 0x00, 0xff }, + }, + [CSC_RGB2YUV] = { + .type = CSC_RGB2YUV, + .matrix = { + 0x0083, 0x0102, 0x0032, + 0x1fb5, 0x1f6c, 0x00e1, + 0x00e1, 0x1f45, 0x1fdc + }, + .pre_bias = { 0x00, 0x00, 0x00 }, + .post_bias = { 0x10, 0x80, 0x80 }, + .pre_clamp = { 0x00, 0xff, 0x00, 0xff, 0x00, 0xff }, + .post_clamp = { 0x10, 0xeb, 0x10, 0xf0, 0x10, 0xf0 }, + }, + [CSC_YUV2YUV] = { + .type = CSC_YUV2YUV, + .matrix = { + 0x0200, 0x0000, 0x0000, + 0x0000, 0x0200, 0x0000, + 0x0000, 0x0000, 0x0200 + }, + .pre_bias = { 0x00, 0x00, 0x00 }, + .post_bias = { 0x00, 0x00, 0x00 }, + .pre_clamp = { 0x00, 0xff, 0x00, 0xff, 0x00, 0xff }, + .post_clamp = { 0x00, 0xff, 0x00, 0xff, 0x00, 0xff }, + }, +}; + +#define MDP_TILE_HEIGHT_DEFAULT 1 +#define MDP_TILE_HEIGHT_UBWC 4 +#define MDP_TILE_HEIGHT_NV12 8 + +#define INTERLEAVED_RGB_FMT(fmt, bp, r, g, b, e0, e1, e2) \ +{ \ + .pixel_format = DRM_FORMAT_ ## fmt, \ + .fetch_type = MDP_PLANE_INTERLEAVED, \ + .alpha_enable = false, \ + .element = { (e0), (e1), (e2), 0 }, \ + .bpc_g_y = g, \ + .bpc_b_cb = b, \ + .bpc_r_cr = r, \ + .bpc_a = 0, \ + .chroma_sample = CHROMA_FULL, \ + .unpack_count = 3, \ + .bpp = bp, \ + .fetch_mode = MDP_FETCH_LINEAR, \ + .flags = MSM_FORMAT_FLAG_UNPACK_TIGHT, \ + .num_planes = 1, \ + .tile_height = MDP_TILE_HEIGHT_DEFAULT \ +} + +#define INTERLEAVED_RGBA_FMT(fmt, bp, a, r, g, b, e0, e1, e2, e3) \ +{ \ + .pixel_format = DRM_FORMAT_ ## fmt, \ + .fetch_type = MDP_PLANE_INTERLEAVED, \ + .alpha_enable = true, \ + .element = { (e0), (e1), (e2), (e3) }, \ + .bpc_g_y = g, \ + .bpc_b_cb = b, \ + .bpc_r_cr = r, \ + .bpc_a = a, \ + .chroma_sample = CHROMA_FULL, \ + .unpack_count = 4, \ + .bpp = bp, \ + .fetch_mode = MDP_FETCH_LINEAR, \ + .flags = MSM_FORMAT_FLAG_UNPACK_TIGHT, \ + .num_planes = 1, \ + .tile_height = MDP_TILE_HEIGHT_DEFAULT \ +} + +#define INTERLEAVED_RGBX_FMT(fmt, bp, a, r, g, b, e0, e1, e2, e3) \ +{ \ + .pixel_format = DRM_FORMAT_ ## fmt, \ + .fetch_type = MDP_PLANE_INTERLEAVED, \ + .alpha_enable = false, \ + .element = { (e0), (e1), (e2), (e3) }, \ + .bpc_g_y = g, \ + .bpc_b_cb = b, \ + .bpc_r_cr = r, \ + .bpc_a = a, \ + .chroma_sample = CHROMA_FULL, \ + .unpack_count = 4, \ + .bpp = bp, \ + .fetch_mode = MDP_FETCH_LINEAR, \ + .flags = MSM_FORMAT_FLAG_UNPACK_TIGHT, \ + .num_planes = 1, \ + .tile_height = MDP_TILE_HEIGHT_DEFAULT \ +} + +#define INTERLEAVED_RGBA_DX_FMT(fmt, bp, a, r, g, b, e0, e1, e2, e3) \ +{ \ + .pixel_format = DRM_FORMAT_ ## fmt, \ + .fetch_type = MDP_PLANE_INTERLEAVED, \ + .alpha_enable = true, \ + .element = { (e0), (e1), (e2), (e3) }, \ + .bpc_g_y = g, \ + .bpc_b_cb = b, \ + .bpc_r_cr = r, \ + .bpc_a = a, \ + .chroma_sample = CHROMA_FULL, \ + .unpack_count = 4, \ + .bpp = bp, \ + .fetch_mode = MDP_FETCH_LINEAR, \ + .flags = MSM_FORMAT_FLAG_UNPACK_TIGHT | \ + MSM_FORMAT_FLAG_DX, \ + .num_planes = 1, \ + .tile_height = MDP_TILE_HEIGHT_DEFAULT \ +} + +#define INTERLEAVED_RGBX_DX_FMT(fmt, bp, a, r, g, b, e0, e1, e2, e3) \ +{ \ + .pixel_format = DRM_FORMAT_ ## fmt, \ + .fetch_type = MDP_PLANE_INTERLEAVED, \ + .alpha_enable = false, \ + .element = { (e0), (e1), (e2), (e3) }, \ + .bpc_g_y = g, \ + .bpc_b_cb = b, \ + .bpc_r_cr = r, \ + .bpc_a = a, \ + .chroma_sample = CHROMA_FULL, \ + .unpack_count = 4, \ + .bpp = bp, \ + .fetch_mode = MDP_FETCH_LINEAR, \ + .flags = MSM_FORMAT_FLAG_UNPACK_TIGHT | \ + MSM_FORMAT_FLAG_DX, \ + .num_planes = 1, \ + .tile_height = MDP_TILE_HEIGHT_DEFAULT \ +} + +#define INTERLEAVED_RGB_FMT_TILED(fmt, bp, r, g, b, e0, e1, e2) \ +{ \ + .pixel_format = DRM_FORMAT_ ## fmt, \ + .fetch_type = MDP_PLANE_INTERLEAVED, \ + .alpha_enable = false, \ + .element = { (e0), (e1), (e2), 0 }, \ + .bpc_g_y = g, \ + .bpc_b_cb = b, \ + .bpc_r_cr = r, \ + .bpc_a = 0, \ + .chroma_sample = CHROMA_FULL, \ + .unpack_count = 3, \ + .bpp = bp, \ + .fetch_mode = MDP_FETCH_UBWC, \ + .flags = MSM_FORMAT_FLAG_UNPACK_TIGHT | \ + MSM_FORMAT_FLAG_COMPRESSED, \ + .num_planes = 2, \ + .tile_height = MDP_TILE_HEIGHT_UBWC, \ +} + +#define INTERLEAVED_RGBA_FMT_TILED(fmt, bp, a, r, g, b, e0, e1, e2, e3) \ +{ \ + .pixel_format = DRM_FORMAT_ ## fmt, \ + .fetch_type = MDP_PLANE_INTERLEAVED, \ + .alpha_enable = true, \ + .element = { (e0), (e1), (e2), (e3) }, \ + .bpc_g_y = g, \ + .bpc_b_cb = b, \ + .bpc_r_cr = r, \ + .bpc_a = a, \ + .chroma_sample = CHROMA_FULL, \ + .unpack_count = 4, \ + .bpp = bp, \ + .fetch_mode = MDP_FETCH_UBWC, \ + .flags = MSM_FORMAT_FLAG_UNPACK_TIGHT | \ + MSM_FORMAT_FLAG_COMPRESSED, \ + .num_planes = 2, \ + .tile_height = MDP_TILE_HEIGHT_UBWC, \ +} + +#define INTERLEAVED_RGBX_FMT_TILED(fmt, bp, a, r, g, b, e0, e1, e2, e3) \ +{ \ + .pixel_format = DRM_FORMAT_ ## fmt, \ + .fetch_type = MDP_PLANE_INTERLEAVED, \ + .alpha_enable = false, \ + .element = { (e0), (e1), (e2), (e3) }, \ + .bpc_g_y = g, \ + .bpc_b_cb = b, \ + .bpc_r_cr = r, \ + .bpc_a = a, \ + .chroma_sample = CHROMA_FULL, \ + .unpack_count = 4, \ + .bpp = bp, \ + .fetch_mode = MDP_FETCH_UBWC, \ + .flags = MSM_FORMAT_FLAG_UNPACK_TIGHT | \ + MSM_FORMAT_FLAG_COMPRESSED, \ + .num_planes = 2, \ + .tile_height = MDP_TILE_HEIGHT_UBWC, \ +} + +#define INTERLEAVED_RGBA_DX_FMT_TILED(fmt, bp, a, r, g, b, e0, e1, e2, e3) \ +{ \ + .pixel_format = DRM_FORMAT_ ## fmt, \ + .fetch_type = MDP_PLANE_INTERLEAVED, \ + .alpha_enable = true, \ + .element = { (e0), (e1), (e2), (e3) }, \ + .bpc_g_y = g, \ + .bpc_b_cb = b, \ + .bpc_r_cr = r, \ + .bpc_a = a, \ + .chroma_sample = CHROMA_FULL, \ + .unpack_count = 4, \ + .bpp = bp, \ + .fetch_mode = MDP_FETCH_UBWC, \ + .flags = MSM_FORMAT_FLAG_UNPACK_TIGHT | \ + MSM_FORMAT_FLAG_DX | \ + MSM_FORMAT_FLAG_COMPRESSED, \ + .num_planes = 2, \ + .tile_height = MDP_TILE_HEIGHT_UBWC, \ +} + +#define INTERLEAVED_YUV_FMT(fmt, bp, r, g, b, e0, e1, e2, e3, chroma) \ +{ \ + .pixel_format = DRM_FORMAT_ ## fmt, \ + .fetch_type = MDP_PLANE_INTERLEAVED, \ + .alpha_enable = false, \ + .element = { (e0), (e1), (e2), (e3)}, \ + .bpc_g_y = g, \ + .bpc_b_cb = b, \ + .bpc_r_cr = r, \ + .bpc_a = 0, \ + .chroma_sample = chroma, \ + .unpack_count = 4, \ + .bpp = bp, \ + .fetch_mode = MDP_FETCH_LINEAR, \ + .flags = MSM_FORMAT_FLAG_UNPACK_TIGHT | \ + MSM_FORMAT_FLAG_YUV, \ + .num_planes = 1, \ + .tile_height = MDP_TILE_HEIGHT_DEFAULT \ +} + +#define PSEUDO_YUV_FMT(fmt, r, g, b, e0, e1, chroma) \ +{ \ + .pixel_format = DRM_FORMAT_ ## fmt, \ + .fetch_type = MDP_PLANE_PSEUDO_PLANAR, \ + .alpha_enable = 0, \ + .element = { (e0), (e1), 0, 0 }, \ + .bpc_g_y = g, \ + .bpc_b_cb = b, \ + .bpc_r_cr = r, \ + .bpc_a = 0, \ + .chroma_sample = chroma, \ + .unpack_count = 2, \ + .bpp = 2, \ + .fetch_mode = MDP_FETCH_LINEAR, \ + .flags = MSM_FORMAT_FLAG_UNPACK_TIGHT | \ + MSM_FORMAT_FLAG_YUV, \ + .num_planes = 2, \ + .tile_height = MDP_TILE_HEIGHT_DEFAULT \ +} + +#define PSEUDO_YUV_FMT_TILED(fmt, r, g, b, e0, e1, chroma, flg, th) \ +{ \ + .pixel_format = DRM_FORMAT_ ## fmt, \ + .fetch_type = MDP_PLANE_PSEUDO_PLANAR, \ + .alpha_enable = 0, \ + .element = { (e0), (e1), 0, 0 }, \ + .bpc_g_y = g, \ + .bpc_b_cb = b, \ + .bpc_r_cr = r, \ + .bpc_a = 0, \ + .chroma_sample = chroma, \ + .unpack_count = 2, \ + .bpp = 2, \ + .fetch_mode = MDP_FETCH_UBWC, \ + .flags = MSM_FORMAT_FLAG_UNPACK_TIGHT | \ + MSM_FORMAT_FLAG_YUV | \ + MSM_FORMAT_FLAG_COMPRESSED | flg, \ + .num_planes = 4, \ + .tile_height = th \ +} + +#define PSEUDO_YUV_FMT_LOOSE(fmt, r, g, b, e0, e1, chroma) \ +{ \ + .pixel_format = DRM_FORMAT_ ## fmt, \ + .fetch_type = MDP_PLANE_PSEUDO_PLANAR, \ + .alpha_enable = 0, \ + .element = { (e0), (e1), 0, 0 }, \ + .bpc_g_y = g, \ + .bpc_b_cb = b, \ + .bpc_r_cr = r, \ + .bpc_a = 0, \ + .chroma_sample = chroma, \ + .unpack_count = 2, \ + .bpp = 2, \ + .fetch_mode = MDP_FETCH_LINEAR, \ + .flags = MSM_FORMAT_FLAG_UNPACK_ALIGN_MSB | \ + MSM_FORMAT_FLAG_DX | \ + MSM_FORMAT_FLAG_YUV, \ + .num_planes = 2, \ + .tile_height = MDP_TILE_HEIGHT_DEFAULT \ +} + +#define PLANAR_YUV_FMT(fmt, bp, r, g, b, e0, e1, e2, chroma) \ +{ \ + .pixel_format = DRM_FORMAT_ ## fmt, \ + .fetch_type = MDP_PLANE_PLANAR, \ + .alpha_enable = false, \ + .element = { (e0), (e1), (e2), 0 }, \ + .bpc_g_y = g, \ + .bpc_b_cb = b, \ + .bpc_r_cr = r, \ + .bpc_a = 0, \ + .chroma_sample = chroma, \ + .unpack_count = 1, \ + .bpp = bp, \ + .fetch_mode = MDP_FETCH_LINEAR, \ + .flags = MSM_FORMAT_FLAG_UNPACK_TIGHT | \ + MSM_FORMAT_FLAG_YUV, \ + .num_planes = 3, \ + .tile_height = MDP_TILE_HEIGHT_DEFAULT \ +} + +static const struct msm_format mdp_formats[] = { + INTERLEAVED_RGBA_FMT(ARGB8888, 4, + BPC8A, BPC8, BPC8, BPC8, + C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA), + + INTERLEAVED_RGBA_FMT(ABGR8888, 4, + BPC8A, BPC8, BPC8, BPC8, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + INTERLEAVED_RGBX_FMT(XBGR8888, 4, + BPC8A, BPC8, BPC8, BPC8, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + INTERLEAVED_RGBA_FMT(RGBA8888, 4, + BPC8A, BPC8, BPC8, BPC8, + C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr), + + INTERLEAVED_RGBA_FMT(BGRA8888, 4, + BPC8A, BPC8, BPC8, BPC8, + C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb), + + INTERLEAVED_RGBX_FMT(BGRX8888, 4, + BPC8A, BPC8, BPC8, BPC8, + C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb), + + INTERLEAVED_RGBX_FMT(XRGB8888, 4, + BPC8A, BPC8, BPC8, BPC8, + C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA), + + INTERLEAVED_RGBX_FMT(RGBX8888, 4, + BPC8A, BPC8, BPC8, BPC8, + C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr), + + INTERLEAVED_RGB_FMT(RGB888, 3, + BPC8, BPC8, BPC8, + C1_B_Cb, C0_G_Y, C2_R_Cr), + + INTERLEAVED_RGB_FMT(BGR888, 3, + BPC8, BPC8, BPC8, + C2_R_Cr, C0_G_Y, C1_B_Cb), + + INTERLEAVED_RGB_FMT(RGB565, 2, + BPC5, BPC6, BPC5, + C1_B_Cb, C0_G_Y, C2_R_Cr), + + INTERLEAVED_RGB_FMT(BGR565, 2, + BPC5, BPC6, BPC5, + C2_R_Cr, C0_G_Y, C1_B_Cb), + + INTERLEAVED_RGBA_FMT(ARGB1555, 2, + BPC1A, BPC5, BPC5, BPC5, + C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA), + + INTERLEAVED_RGBA_FMT(ABGR1555, 2, + BPC1A, BPC5, BPC5, BPC5, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + INTERLEAVED_RGBA_FMT(RGBA5551, 2, + BPC1A, BPC5, BPC5, BPC5, + C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr), + + INTERLEAVED_RGBA_FMT(BGRA5551, 2, + BPC1A, BPC5, BPC5, BPC5, + C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb), + + INTERLEAVED_RGBX_FMT(XRGB1555, 2, + BPC1A, BPC5, BPC5, BPC5, + C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA), + + INTERLEAVED_RGBX_FMT(XBGR1555, 2, + BPC1A, BPC5, BPC5, BPC5, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + INTERLEAVED_RGBX_FMT(RGBX5551, 2, + BPC1A, BPC5, BPC5, BPC5, + C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr), + + INTERLEAVED_RGBX_FMT(BGRX5551, 2, + BPC1A, BPC5, BPC5, BPC5, + C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb), + + INTERLEAVED_RGBA_FMT(ARGB4444, 2, + BPC4A, BPC4, BPC4, BPC4, + C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA), + + INTERLEAVED_RGBA_FMT(ABGR4444, 2, + BPC4A, BPC4, BPC4, BPC4, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + INTERLEAVED_RGBA_FMT(RGBA4444, 2, + BPC4A, BPC4, BPC4, BPC4, + C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr), + + INTERLEAVED_RGBA_FMT(BGRA4444, 2, + BPC4A, BPC4, BPC4, BPC4, + C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb), + + INTERLEAVED_RGBX_FMT(XRGB4444, 2, + BPC4A, BPC4, BPC4, BPC4, + C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA), + + INTERLEAVED_RGBX_FMT(XBGR4444, 2, + BPC4A, BPC4, BPC4, BPC4, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + INTERLEAVED_RGBX_FMT(RGBX4444, 2, + BPC4A, BPC4, BPC4, BPC4, + C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr), + + INTERLEAVED_RGBX_FMT(BGRX4444, 2, + BPC4A, BPC4, BPC4, BPC4, + C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb), + + INTERLEAVED_RGBA_DX_FMT(BGRA1010102, 4, + BPC8A, BPC8, BPC8, BPC8, + C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb), + + INTERLEAVED_RGBA_DX_FMT(RGBA1010102, 4, + BPC8A, BPC8, BPC8, BPC8, + C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr), + + INTERLEAVED_RGBA_DX_FMT(ABGR2101010, 4, + BPC8A, BPC8, BPC8, BPC8, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + INTERLEAVED_RGBA_DX_FMT(ARGB2101010, 4, + BPC8A, BPC8, BPC8, BPC8, + C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA), + + INTERLEAVED_RGBX_DX_FMT(XRGB2101010, 4, + BPC8A, BPC8, BPC8, BPC8, + C1_B_Cb, C0_G_Y, C2_R_Cr, C3_ALPHA), + + INTERLEAVED_RGBX_DX_FMT(BGRX1010102, 4, + BPC8A, BPC8, BPC8, BPC8, + C3_ALPHA, C2_R_Cr, C0_G_Y, C1_B_Cb), + + INTERLEAVED_RGBX_DX_FMT(XBGR2101010, 4, + BPC8A, BPC8, BPC8, BPC8, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + INTERLEAVED_RGBX_DX_FMT(RGBX1010102, 4, + BPC8A, BPC8, BPC8, BPC8, + C3_ALPHA, C1_B_Cb, C0_G_Y, C2_R_Cr), + + /* --- RGB formats above / YUV formats below this line --- */ + + /* 2 plane YUV */ + PSEUDO_YUV_FMT(NV12, + BPC8, BPC8, BPC8, + C1_B_Cb, C2_R_Cr, + CHROMA_420), + + PSEUDO_YUV_FMT(NV21, + BPC8, BPC8, BPC8, + C2_R_Cr, C1_B_Cb, + CHROMA_420), + + PSEUDO_YUV_FMT(NV16, + BPC8, BPC8, BPC8, + C1_B_Cb, C2_R_Cr, + CHROMA_H2V1), + + PSEUDO_YUV_FMT(NV61, + BPC8, BPC8, BPC8, + C2_R_Cr, C1_B_Cb, + CHROMA_H2V1), + + PSEUDO_YUV_FMT_LOOSE(P010, + BPC8, BPC8, BPC8, + C1_B_Cb, C2_R_Cr, + CHROMA_420), + + /* 1 plane YUV */ + INTERLEAVED_YUV_FMT(VYUY, 2, + BPC8, BPC8, BPC8, + C2_R_Cr, C0_G_Y, C1_B_Cb, C0_G_Y, + CHROMA_H2V1), + + INTERLEAVED_YUV_FMT(UYVY, 2, + BPC8, BPC8, BPC8, + C1_B_Cb, C0_G_Y, C2_R_Cr, C0_G_Y, + CHROMA_H2V1), + + INTERLEAVED_YUV_FMT(YUYV, 2, + BPC8, BPC8, BPC8, + C0_G_Y, C1_B_Cb, C0_G_Y, C2_R_Cr, + CHROMA_H2V1), + + INTERLEAVED_YUV_FMT(YVYU, 2, + BPC8, BPC8, BPC8, + C0_G_Y, C2_R_Cr, C0_G_Y, C1_B_Cb, + CHROMA_H2V1), + + /* 3 plane YUV */ + PLANAR_YUV_FMT(YUV420, 1, + BPC8, BPC8, BPC8, + C2_R_Cr, C1_B_Cb, C0_G_Y, + CHROMA_420), + + PLANAR_YUV_FMT(YVU420, 1, + BPC8, BPC8, BPC8, + C1_B_Cb, C2_R_Cr, C0_G_Y, + CHROMA_420), +}; + +/* + * UBWC formats table: + * This table holds the UBWC formats supported. + * If a compression ratio needs to be used for this or any other format, + * the data will be passed by user-space. + */ +static const struct msm_format mdp_formats_ubwc[] = { + INTERLEAVED_RGB_FMT_TILED(BGR565, 2, + BPC5, BPC6, BPC5, + C2_R_Cr, C0_G_Y, C1_B_Cb), + + INTERLEAVED_RGBA_FMT_TILED(ABGR8888, 4, + BPC8A, BPC8, BPC8, BPC8, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + /* ARGB8888 and ABGR8888 purposely have the same color + * ordering. The hardware only supports ABGR8888 UBWC + * natively. + */ + INTERLEAVED_RGBA_FMT_TILED(ARGB8888, 4, + BPC8A, BPC8, BPC8, BPC8, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + INTERLEAVED_RGBX_FMT_TILED(XBGR8888, 4, + BPC8A, BPC8, BPC8, BPC8, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + INTERLEAVED_RGBX_FMT_TILED(XRGB8888, 4, + BPC8A, BPC8, BPC8, BPC8, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + INTERLEAVED_RGBA_DX_FMT_TILED(ABGR2101010, 4, + BPC8A, BPC8, BPC8, BPC8, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + INTERLEAVED_RGBA_DX_FMT_TILED(XBGR2101010, 4, + BPC8A, BPC8, BPC8, BPC8, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + INTERLEAVED_RGBA_DX_FMT_TILED(XRGB2101010, 4, + BPC8A, BPC8, BPC8, BPC8, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + /* XRGB2101010 and ARGB2101010 purposely have the same color + * ordering. The hardware only supports ARGB2101010 UBWC + * natively. + */ + INTERLEAVED_RGBA_DX_FMT_TILED(ARGB2101010, 4, + BPC8A, BPC8, BPC8, BPC8, + C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA), + + PSEUDO_YUV_FMT_TILED(NV12, + BPC8, BPC8, BPC8, + C1_B_Cb, C2_R_Cr, + CHROMA_420, 0, + MDP_TILE_HEIGHT_NV12), + + PSEUDO_YUV_FMT_TILED(P010, + BPC8, BPC8, BPC8, + C1_B_Cb, C2_R_Cr, + CHROMA_420, MSM_FORMAT_FLAG_DX, + MDP_TILE_HEIGHT_UBWC), +}; + +const struct msm_format *mdp_get_format(struct msm_kms *kms, uint32_t format, + uint64_t modifier) +{ + const struct msm_format *map = NULL; + ssize_t map_size; + int i; + + switch (modifier) { + case 0: + map = mdp_formats; + map_size = ARRAY_SIZE(mdp_formats); + break; + case DRM_FORMAT_MOD_QCOM_COMPRESSED: + map = mdp_formats_ubwc; + map_size = ARRAY_SIZE(mdp_formats_ubwc); + break; + default: + drm_err(kms->dev, "unsupported format modifier %llX\n", modifier); + return NULL; + } + + for (i = 0; i < map_size; i++) { + const struct msm_format *f = &map[i]; + + if (f->pixel_format == format) + return f; + } + + drm_err(kms->dev, "unsupported fmt: %p4cc modifier 0x%llX\n", + &format, modifier); + + return NULL; +} + +struct csc_cfg *mdp_get_default_csc_cfg(enum csc_type type) +{ + if (WARN_ON(type >= CSC_MAX)) + return NULL; + + return &csc_convert[type]; +} diff --git a/drivers/gpu/drm/msm/disp/mdp_format.h b/drivers/gpu/drm/msm/disp/mdp_format.h new file mode 100644 index 000000000..915954bf5 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp_format.h @@ -0,0 +1,81 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2016-2018, The Linux Foundation. All rights reserved. + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#ifndef __MSM_FORMAT_H__ +#define __MSM_FORMAT_H__ + +#include "mdp_common.xml.h" + +enum msm_format_flags { + MSM_FORMAT_FLAG_YUV_BIT, + MSM_FORMAT_FLAG_DX_BIT, + MSM_FORMAT_FLAG_COMPRESSED_BIT, + MSM_FORMAT_FLAG_UNPACK_TIGHT_BIT, + MSM_FORMAT_FLAG_UNPACK_ALIGN_MSB_BIT, +}; + +#define MSM_FORMAT_FLAG_YUV BIT(MSM_FORMAT_FLAG_YUV_BIT) +#define MSM_FORMAT_FLAG_DX BIT(MSM_FORMAT_FLAG_DX_BIT) +#define MSM_FORMAT_FLAG_COMPRESSED BIT(MSM_FORMAT_FLAG_COMPRESSED_BIT) +#define MSM_FORMAT_FLAG_UNPACK_TIGHT BIT(MSM_FORMAT_FLAG_UNPACK_TIGHT_BIT) +#define MSM_FORMAT_FLAG_UNPACK_ALIGN_MSB BIT(MSM_FORMAT_FLAG_UNPACK_ALIGN_MSB_BIT) + +/* + * DPU HW,Component order color map + */ +enum { + C0_G_Y = 0, + C1_B_Cb = 1, + C2_R_Cr = 2, + C3_ALPHA = 3 +}; + +/** + * struct msm_format: defines the format configuration + * @pixel_format: format fourcc + * @bpc_g_y: element bit widths: BPC for G or Y + * @bpc_b_cb: element bit widths: BPC for B or Cb + * @bpc_r_cr: element bit widths: BPC for R or Cr + * @bpc_a: element bit widths: BPC for the alpha channel + * @element: element color ordering + * @fetch_type: how the color components are packed in pixel format + * @chroma_sample: chroma sub-samplng type + * @alpha_enable: whether the format has an alpha channel + * @unpack_count: number of the components to unpack + * @bpp: bytes per pixel + * @flags: usage bit flags + * @num_planes: number of planes (including meta data planes) + * @fetch_mode: linear, tiled, or ubwc hw fetch behavior + * @tile_height: format tile height + */ +struct msm_format { + uint32_t pixel_format; + enum mdp_bpc bpc_g_y, bpc_b_cb, bpc_r_cr; + enum mdp_bpc_alpha bpc_a; + u8 element[4]; + enum mdp_fetch_type fetch_type; + enum mdp_chroma_samp_type chroma_sample; + bool alpha_enable; + u8 unpack_count; + u8 bpp; + unsigned long flags; + u8 num_planes; + enum mdp_fetch_mode fetch_mode; + u16 tile_height; +}; + +#define MSM_FORMAT_IS_YUV(X) ((X)->flags & MSM_FORMAT_FLAG_YUV) +#define MSM_FORMAT_IS_DX(X) ((X)->flags & MSM_FORMAT_FLAG_DX) +#define MSM_FORMAT_IS_LINEAR(X) ((X)->fetch_mode == MDP_FETCH_LINEAR) +#define MSM_FORMAT_IS_TILE(X) \ + (((X)->fetch_mode == MDP_FETCH_UBWC) && \ + !((X)->flags & MSM_FORMAT_FLAG_COMPRESSED)) +#define MSM_FORMAT_IS_UBWC(X) \ + (((X)->fetch_mode == MDP_FETCH_UBWC) && \ + ((X)->flags & MSM_FORMAT_FLAG_COMPRESSED)) + +#endif diff --git a/drivers/gpu/drm/msm/disp/mdp_kms.c b/drivers/gpu/drm/msm/disp/mdp_kms.c new file mode 100644 index 000000000..3c35ccfc7 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp_kms.c @@ -0,0 +1,138 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + + +#include "msm_drv.h" +#include "mdp_kms.h" + + +struct mdp_irq_wait { + struct mdp_irq irq; + int count; +}; + +static DECLARE_WAIT_QUEUE_HEAD(wait_event); + +static DEFINE_SPINLOCK(list_lock); + +static void update_irq(struct mdp_kms *mdp_kms) +{ + struct mdp_irq *irq; + uint32_t irqmask = mdp_kms->vblank_mask; + + assert_spin_locked(&list_lock); + + list_for_each_entry(irq, &mdp_kms->irq_list, node) + irqmask |= irq->irqmask; + + mdp_kms->funcs->set_irqmask(mdp_kms, irqmask, mdp_kms->cur_irq_mask); + mdp_kms->cur_irq_mask = irqmask; +} + +/* if an mdp_irq's irqmask has changed, such as when mdp5 crtc<->encoder + * link changes, this must be called to figure out the new global irqmask + */ +void mdp_irq_update(struct mdp_kms *mdp_kms) +{ + unsigned long flags; + spin_lock_irqsave(&list_lock, flags); + update_irq(mdp_kms); + spin_unlock_irqrestore(&list_lock, flags); +} + +void mdp_dispatch_irqs(struct mdp_kms *mdp_kms, uint32_t status) +{ + struct mdp_irq *handler, *n; + unsigned long flags; + + spin_lock_irqsave(&list_lock, flags); + mdp_kms->in_irq = true; + list_for_each_entry_safe(handler, n, &mdp_kms->irq_list, node) { + if (handler->irqmask & status) { + spin_unlock_irqrestore(&list_lock, flags); + handler->irq(handler, handler->irqmask & status); + spin_lock_irqsave(&list_lock, flags); + } + } + mdp_kms->in_irq = false; + update_irq(mdp_kms); + spin_unlock_irqrestore(&list_lock, flags); + +} + +void mdp_update_vblank_mask(struct mdp_kms *mdp_kms, uint32_t mask, bool enable) +{ + unsigned long flags; + + spin_lock_irqsave(&list_lock, flags); + if (enable) + mdp_kms->vblank_mask |= mask; + else + mdp_kms->vblank_mask &= ~mask; + update_irq(mdp_kms); + spin_unlock_irqrestore(&list_lock, flags); +} + +static void wait_irq(struct mdp_irq *irq, uint32_t irqstatus) +{ + struct mdp_irq_wait *wait = + container_of(irq, struct mdp_irq_wait, irq); + wait->count--; + wake_up_all(&wait_event); +} + +void mdp_irq_wait(struct mdp_kms *mdp_kms, uint32_t irqmask) +{ + struct mdp_irq_wait wait = { + .irq = { + .irq = wait_irq, + .irqmask = irqmask, + }, + .count = 1, + }; + mdp_irq_register(mdp_kms, &wait.irq); + wait_event_timeout(wait_event, (wait.count <= 0), + msecs_to_jiffies(100)); + mdp_irq_unregister(mdp_kms, &wait.irq); +} + +void mdp_irq_register(struct mdp_kms *mdp_kms, struct mdp_irq *irq) +{ + unsigned long flags; + bool needs_update = false; + + spin_lock_irqsave(&list_lock, flags); + + if (!irq->registered) { + irq->registered = true; + list_add(&irq->node, &mdp_kms->irq_list); + needs_update = !mdp_kms->in_irq; + } + + spin_unlock_irqrestore(&list_lock, flags); + + if (needs_update) + mdp_irq_update(mdp_kms); +} + +void mdp_irq_unregister(struct mdp_kms *mdp_kms, struct mdp_irq *irq) +{ + unsigned long flags; + bool needs_update = false; + + spin_lock_irqsave(&list_lock, flags); + + if (irq->registered) { + irq->registered = false; + list_del(&irq->node); + needs_update = !mdp_kms->in_irq; + } + + spin_unlock_irqrestore(&list_lock, flags); + + if (needs_update) + mdp_irq_update(mdp_kms); +} diff --git a/drivers/gpu/drm/msm/disp/mdp_kms.h b/drivers/gpu/drm/msm/disp/mdp_kms.h new file mode 100644 index 000000000..068fbeac6 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/mdp_kms.h @@ -0,0 +1,126 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2013 Red Hat + * Author: Rob Clark + */ + +#ifndef __MDP_KMS_H__ +#define __MDP_KMS_H__ + +#include +#include +#include + +#include "mdp_format.h" +#include "msm_drv.h" +#include "msm_kms.h" +#include "mdp_common.xml.h" + +struct mdp_kms; + +struct mdp_kms_funcs { + struct msm_kms_funcs base; + void (*set_irqmask)(struct mdp_kms *mdp_kms, uint32_t irqmask, + uint32_t old_irqmask); +}; + +struct mdp_kms { + struct msm_kms base; + + const struct mdp_kms_funcs *funcs; + + /* irq handling: */ + bool in_irq; + struct list_head irq_list; /* list of mdp4_irq */ + uint32_t vblank_mask; /* irq bits set for userspace vblank */ + uint32_t cur_irq_mask; /* current irq mask */ +}; +#define to_mdp_kms(x) container_of(x, struct mdp_kms, base) + +static inline int mdp_kms_init(struct mdp_kms *mdp_kms, + const struct mdp_kms_funcs *funcs) +{ + mdp_kms->funcs = funcs; + INIT_LIST_HEAD(&mdp_kms->irq_list); + return msm_kms_init(&mdp_kms->base, &funcs->base); +} + +static inline void mdp_kms_destroy(struct mdp_kms *mdp_kms) +{ + msm_kms_destroy(&mdp_kms->base); +} + +/* + * irq helpers: + */ + +/* For transiently registering for different MDP irqs that various parts + * of the KMS code need during setup/configuration. These are not + * necessarily the same as what drm_vblank_get/put() are requesting, and + * the hysteresis in drm_vblank_put() is not necessarily desirable for + * internal housekeeping related irq usage. + */ +struct mdp_irq { + struct list_head node; + uint32_t irqmask; + bool registered; + void (*irq)(struct mdp_irq *irq, uint32_t irqstatus); +}; + +void mdp_dispatch_irqs(struct mdp_kms *mdp_kms, uint32_t status); +void mdp_update_vblank_mask(struct mdp_kms *mdp_kms, uint32_t mask, bool enable); +void mdp_irq_wait(struct mdp_kms *mdp_kms, uint32_t irqmask); +void mdp_irq_register(struct mdp_kms *mdp_kms, struct mdp_irq *irq); +void mdp_irq_unregister(struct mdp_kms *mdp_kms, struct mdp_irq *irq); +void mdp_irq_update(struct mdp_kms *mdp_kms); + +/* + * pixel format helpers: + */ + +/* MDP capabilities */ +#define MDP_CAP_SMP BIT(0) /* Shared Memory Pool */ +#define MDP_CAP_DSC BIT(1) /* VESA Display Stream Compression */ +#define MDP_CAP_CDM BIT(2) /* Chroma Down Module (HDMI 2.0 YUV) */ +#define MDP_CAP_SRC_SPLIT BIT(3) /* Source Split of SSPPs */ + +/* MDP pipe capabilities */ +#define MDP_PIPE_CAP_HFLIP BIT(0) +#define MDP_PIPE_CAP_VFLIP BIT(1) +#define MDP_PIPE_CAP_SCALE BIT(2) +#define MDP_PIPE_CAP_CSC BIT(3) +#define MDP_PIPE_CAP_DECIMATION BIT(4) +#define MDP_PIPE_CAP_SW_PIX_EXT BIT(5) +#define MDP_PIPE_CAP_CURSOR BIT(6) + +/* MDP layer mixer caps */ +#define MDP_LM_CAP_DISPLAY BIT(0) +#define MDP_LM_CAP_WB BIT(1) +#define MDP_LM_CAP_PAIR BIT(2) + +static inline bool pipe_supports_yuv(uint32_t pipe_caps) +{ + return (pipe_caps & MDP_PIPE_CAP_SCALE) && + (pipe_caps & MDP_PIPE_CAP_CSC); +} + +enum csc_type { + CSC_RGB2RGB = 0, + CSC_YUV2RGB, + CSC_RGB2YUV, + CSC_YUV2YUV, + CSC_MAX +}; + +struct csc_cfg { + enum csc_type type; + uint32_t matrix[9]; + uint32_t pre_bias[3]; + uint32_t post_bias[3]; + uint32_t pre_clamp[6]; + uint32_t post_clamp[6]; +}; + +struct csc_cfg *mdp_get_default_csc_cfg(enum csc_type); + +#endif /* __MDP_KMS_H__ */ diff --git a/drivers/gpu/drm/msm/disp/msm_disp_snapshot.c b/drivers/gpu/drm/msm/disp/msm_disp_snapshot.c new file mode 100644 index 000000000..d99771684 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/msm_disp_snapshot.c @@ -0,0 +1,138 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2020-2021, The Linux Foundation. All rights reserved. + */ + +#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ + +#include "msm_disp_snapshot.h" + +static ssize_t __maybe_unused disp_devcoredump_read(char *buffer, loff_t offset, + size_t count, void *data, size_t datalen) +{ + struct drm_print_iterator iter; + struct drm_printer p; + struct msm_disp_state *disp_state; + + disp_state = data; + + iter.data = buffer; + iter.offset = 0; + iter.start = offset; + iter.remain = count; + + p = drm_coredump_printer(&iter); + + msm_disp_state_print(disp_state, &p); + + return count - iter.remain; +} + +struct msm_disp_state * +msm_disp_snapshot_state_sync(struct msm_kms *kms) +{ + struct drm_device *drm_dev = kms->dev; + struct msm_disp_state *disp_state; + + WARN_ON(!mutex_is_locked(&kms->dump_mutex)); + + disp_state = kzalloc_obj(struct msm_disp_state); + if (!disp_state) + return ERR_PTR(-ENOMEM); + + disp_state->dev = drm_dev->dev; + disp_state->drm_dev = drm_dev; + + INIT_LIST_HEAD(&disp_state->blocks); + + msm_disp_snapshot_capture_state(disp_state); + + return disp_state; +} + +static void _msm_disp_snapshot_work(struct kthread_work *work) +{ + struct msm_kms *kms = container_of(work, struct msm_kms, dump_work); + struct msm_disp_state *disp_state; + struct drm_printer p; + + /* Serialize dumping here */ + mutex_lock(&kms->dump_mutex); + disp_state = msm_disp_snapshot_state_sync(kms); + mutex_unlock(&kms->dump_mutex); + + if (IS_ERR(disp_state)) + return; + + if (MSM_DISP_SNAPSHOT_DUMP_IN_CONSOLE) { + p = drm_info_printer(disp_state->drm_dev->dev); + msm_disp_state_print(disp_state, &p); + } + + /* + * If COREDUMP is disabled, the stub will call the free function. + * If there is a codedump pending for the device, the dev_coredumpm() + * will also free new coredump state. + */ + dev_coredumpm(disp_state->dev, THIS_MODULE, disp_state, 0, GFP_KERNEL, + disp_devcoredump_read, msm_disp_state_free); +} + +void msm_disp_snapshot_state(struct drm_device *drm_dev) +{ + struct msm_drm_private *priv; + struct msm_kms *kms; + + if (!drm_dev) { + DRM_ERROR("invalid params\n"); + return; + } + + priv = drm_dev->dev_private; + kms = priv->kms; + + kthread_queue_work(kms->dump_worker, &kms->dump_work); +} + +int msm_disp_snapshot_init(struct drm_device *drm_dev) +{ + struct msm_drm_private *priv; + struct msm_kms *kms; + + if (!drm_dev) { + DRM_ERROR("invalid params\n"); + return -EINVAL; + } + + priv = drm_dev->dev_private; + kms = priv->kms; + + mutex_init(&kms->dump_mutex); + + kms->dump_worker = kthread_run_worker(0, "%s", "disp_snapshot"); + if (IS_ERR(kms->dump_worker)) + DRM_ERROR("failed to create disp state task\n"); + + kthread_init_work(&kms->dump_work, _msm_disp_snapshot_work); + + return 0; +} + +void msm_disp_snapshot_destroy(struct drm_device *drm_dev) +{ + struct msm_kms *kms; + struct msm_drm_private *priv; + + if (!drm_dev) { + DRM_ERROR("invalid params\n"); + return; + } + + priv = drm_dev->dev_private; + kms = priv->kms; + + if (kms->dump_worker) + kthread_destroy_worker(kms->dump_worker); + + mutex_destroy(&kms->dump_mutex); +} diff --git a/drivers/gpu/drm/msm/disp/msm_disp_snapshot.h b/drivers/gpu/drm/msm/disp/msm_disp_snapshot.h new file mode 100644 index 000000000..38bde9e60 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/msm_disp_snapshot.h @@ -0,0 +1,145 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2020-2021, The Linux Foundation. All rights reserved. + */ + +#ifndef MSM_DISP_SNAPSHOT_H_ +#define MSM_DISP_SNAPSHOT_H_ + +#include +#include +#include "../../../drm_crtc_internal.h" +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "msm_kms.h" + +#define MSM_DISP_SNAPSHOT_MAX_BLKS 10 + +/* debug option to print the registers in logs */ +#define MSM_DISP_SNAPSHOT_DUMP_IN_CONSOLE 0 + +/* print debug ranges in groups of 4 u32s */ +#define REG_DUMP_ALIGN 16 + +/** + * struct msm_disp_state - structure to store current dpu state + * @dev: device pointer + * @drm_dev: drm device pointer + * @blocks: list head for hardware state blocks + * @atomic_state: atomic state duplicated at the time of the error + * @time: timestamp at which the coredump was captured + */ +struct msm_disp_state { + struct device *dev; + struct drm_device *drm_dev; + + struct list_head blocks; + + struct drm_atomic_commit *atomic_state; + + struct timespec64 time; +}; + +/** + * struct msm_disp_state_block - structure to store each hardware block state + * @name: name of the block + * @node: handle to the linked list head + * @size: size of the register space of this hardware block + * @state: array holding the register dump of this hardware block + * @base_addr: starting address of this hardware block's register space + */ +struct msm_disp_state_block { + char name[SZ_128]; + struct list_head node; + unsigned int size; + u32 *state; + void __iomem *base_addr; +}; + +/** + * msm_disp_snapshot_init - initialize display snapshot + * @drm_dev: drm device handle + * + * Returns: 0 or -ERROR + */ +int msm_disp_snapshot_init(struct drm_device *drm_dev); + +/** + * msm_disp_snapshot_destroy - destroy the display snapshot + * @drm_dev: drm device handle + * + * Returns: none + */ +void msm_disp_snapshot_destroy(struct drm_device *drm_dev); + +/** + * msm_disp_snapshot_state_sync - synchronously snapshot display state + * @kms: the kms object + * + * Returns: state or error + * + * Context: + * Must be called with &kms->dump_mutex held + */ +struct msm_disp_state *msm_disp_snapshot_state_sync(struct msm_kms *kms); + +/** + * msm_disp_snapshot_state - trigger to dump the display snapshot + * @drm_dev: handle to drm device + * + * Returns: none + */ +void msm_disp_snapshot_state(struct drm_device *drm_dev); + +/** + * msm_disp_state_print - print out the current dpu state + * @disp_state: handle to drm device + * @p: handle to drm printer + * + * Returns: none + */ +void msm_disp_state_print(struct msm_disp_state *disp_state, struct drm_printer *p); + +/** + * msm_disp_snapshot_capture_state - utility to capture atomic state and hw registers + * @disp_state: handle to msm_disp_state struct + * + * Returns: none + */ +void msm_disp_snapshot_capture_state(struct msm_disp_state *disp_state); + +/** + * msm_disp_state_free - free the memory after the coredump has been read + * @data: handle to struct msm_disp_state + * + * Returns: none + */ +void msm_disp_state_free(void *data); + +/** + * msm_disp_snapshot_add_block - add a hardware block with its register dump + * @disp_state: handle to struct msm_disp_state + * @len: size of the register space of the hardware block + * @base_addr: starting address of the register space of the hardware block + * @fmt: format in which the block names need to be printed + * + * Returns: none + */ +__printf(4, 5) +void msm_disp_snapshot_add_block(struct msm_disp_state *disp_state, u32 len, + void __iomem *base_addr, const char *fmt, ...); + +#endif /* MSM_DISP_SNAPSHOT_H_ */ diff --git a/drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c b/drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c new file mode 100644 index 000000000..850246fc8 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c @@ -0,0 +1,205 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2020-2021, The Linux Foundation. All rights reserved. + */ + +#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ + +#include + +#include "msm_disp_snapshot.h" + +static void msm_disp_state_dump_regs(u32 **reg, u32 len, void __iomem *base_addr) +{ + u32 len_padded; + u32 num_rows; + u32 x0, x4, x8, xc; + void __iomem *addr; + u32 *dump_addr = NULL; + void __iomem *end_addr; + int i; + + len_padded = round_up(len, REG_DUMP_ALIGN); + num_rows = DIV_ROUND_UP(len, REG_DUMP_ALIGN); + + addr = base_addr; + end_addr = base_addr + len; + + *reg = kvzalloc(len_padded, GFP_KERNEL); + if (!*reg) + return; + + dump_addr = *reg; + for (i = 0; i < num_rows; i++) { + x0 = (addr < end_addr) ? readl_relaxed(addr + 0x0) : 0; + x4 = (addr + 0x4 < end_addr) ? readl_relaxed(addr + 0x4) : 0; + x8 = (addr + 0x8 < end_addr) ? readl_relaxed(addr + 0x8) : 0; + xc = (addr + 0xc < end_addr) ? readl_relaxed(addr + 0xc) : 0; + + dump_addr[i * 4] = x0; + dump_addr[i * 4 + 1] = x4; + dump_addr[i * 4 + 2] = x8; + dump_addr[i * 4 + 3] = xc; + + addr += REG_DUMP_ALIGN; + } +} + +static void msm_disp_state_print_regs(const u32 *dump_addr, u32 len, + void __iomem *base_addr, struct drm_printer *p) +{ + void __iomem *addr, *end_addr; + int i; + u32 num_rows; + + if (!dump_addr) { + drm_printf(p, "Registers not stored\n"); + return; + } + + addr = base_addr; + end_addr = base_addr + len; + num_rows = len / REG_DUMP_ALIGN; + + for (i = 0; i < num_rows; i++) { + drm_printf(p, "0x%lx : %08x %08x %08x %08x\n", + (unsigned long)(addr - base_addr), + dump_addr[i * 4], dump_addr[i * 4 + 1], + dump_addr[i * 4 + 2], dump_addr[i * 4 + 3]); + addr += REG_DUMP_ALIGN; + } + + if (addr != end_addr) { + drm_printf(p, "0x%lx : %08x", + (unsigned long)(addr - base_addr), + dump_addr[i * 4]); + if (addr + 0x4 < end_addr) + drm_printf(p, " %08x", dump_addr[i * 4 + 1]); + if (addr + 0x8 < end_addr) + drm_printf(p, " %08x", dump_addr[i * 4 + 2]); + drm_printf(p, "\n"); + } +} + +void msm_disp_state_print(struct msm_disp_state *state, struct drm_printer *p) +{ + struct msm_disp_state_block *block, *tmp; + + if (!p) { + DRM_ERROR("invalid drm printer\n"); + return; + } + + drm_printf(p, "---\n"); + drm_printf(p, "kernel: %s\n", init_utsname()->release); + drm_printf(p, "module: " KBUILD_MODNAME "\n"); + drm_printf(p, "dpu devcoredump\n"); + drm_printf(p, "time: %ptSp\n", &state->time); + + list_for_each_entry_safe(block, tmp, &state->blocks, node) { + drm_printf(p, "====================%s================\n", block->name); + msm_disp_state_print_regs(block->state, block->size, block->base_addr, p); + } + + drm_printf(p, "===================dpu drm state================\n"); + + if (state->atomic_state) + drm_atomic_print_new_state(state->atomic_state, p); +} + +static void msm_disp_capture_atomic_state(struct msm_disp_state *disp_state) +{ + struct drm_device *ddev; + struct drm_modeset_acquire_ctx ctx; + + ktime_get_real_ts64(&disp_state->time); + + ddev = disp_state->drm_dev; + + drm_modeset_acquire_init(&ctx, 0); + + while (drm_modeset_lock_all_ctx(ddev, &ctx) != 0) + drm_modeset_backoff(&ctx); + + disp_state->atomic_state = drm_atomic_helper_duplicate_state(ddev, + &ctx); + drm_modeset_drop_locks(&ctx); + drm_modeset_acquire_fini(&ctx); +} + +void msm_disp_snapshot_capture_state(struct msm_disp_state *disp_state) +{ + struct msm_drm_private *priv; + struct drm_device *drm_dev; + struct msm_kms *kms; + int i; + + drm_dev = disp_state->drm_dev; + priv = drm_dev->dev_private; + kms = priv->kms; + + for (i = 0; i < ARRAY_SIZE(kms->dp); i++) { + if (!kms->dp[i]) + continue; + + msm_dp_snapshot(disp_state, kms->dp[i]); + } + + for (i = 0; i < ARRAY_SIZE(kms->dsi); i++) { + if (!kms->dsi[i]) + continue; + + msm_dsi_snapshot(disp_state, kms->dsi[i]); + } + + if (kms->funcs->snapshot) + kms->funcs->snapshot(disp_state, kms); + + msm_disp_capture_atomic_state(disp_state); +} + +void msm_disp_state_free(void *data) +{ + struct msm_disp_state *disp_state = data; + struct msm_disp_state_block *block, *tmp; + + if (disp_state->atomic_state) { + drm_atomic_commit_put(disp_state->atomic_state); + disp_state->atomic_state = NULL; + } + + list_for_each_entry_safe(block, tmp, &disp_state->blocks, node) { + list_del(&block->node); + kvfree(block->state); + kfree(block); + } + + kfree(disp_state); +} + +void msm_disp_snapshot_add_block(struct msm_disp_state *disp_state, u32 len, + void __iomem *base_addr, const char *fmt, ...) +{ + struct msm_disp_state_block *new_blk; + struct va_format vaf; + va_list va; + + new_blk = kzalloc_obj(struct msm_disp_state_block); + if (!new_blk) + return; + + va_start(va, fmt); + + vaf.fmt = fmt; + vaf.va = &va; + snprintf(new_blk->name, sizeof(new_blk->name), "%pV", &vaf); + + va_end(va); + + INIT_LIST_HEAD(&new_blk->node); + new_blk->size = len; + new_blk->base_addr = base_addr; + + msm_disp_state_dump_regs(&new_blk->state, new_blk->size, base_addr); + list_add_tail(&new_blk->node, &disp_state->blocks); +} diff --git a/drivers/gpu/drm/msm/dp/dp_audio.c b/drivers/gpu/drm/msm/dp/dp_audio.c new file mode 100644 index 000000000..df222943a --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_audio.c @@ -0,0 +1,377 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2016-2020, The Linux Foundation. All rights reserved. + */ + + +#define pr_fmt(fmt) "[drm-dp] %s: " fmt, __func__ + +#include + +#include +#include + +#include "dp_audio.h" +#include "dp_drm.h" +#include "dp_panel.h" +#include "dp_reg.h" +#include "dp_display.h" +#include "dp_utils.h" + +struct msm_dp_audio_private { + struct platform_device *pdev; + struct drm_device *drm_dev; + void __iomem *link_base; + + u32 channels; + + struct msm_dp_audio msm_dp_audio; +}; + +static inline u32 msm_dp_read_link(struct msm_dp_audio_private *audio, u32 offset) +{ + return readl_relaxed(audio->link_base + offset); +} + +static inline void msm_dp_write_link(struct msm_dp_audio_private *audio, + u32 offset, u32 data) +{ + /* + * To make sure link reg writes happens before any other operation, + * this function uses writel() instread of writel_relaxed() + */ + writel(data, audio->link_base + offset); +} + +static void msm_dp_audio_stream_sdp(struct msm_dp_audio_private *audio) +{ + struct dp_sdp_header sdp_hdr = { + .HB0 = 0x00, + .HB1 = 0x02, + .HB2 = 0x00, + .HB3 = audio->channels - 1, + }; + u32 header[2]; + + msm_dp_utils_pack_sdp_header(&sdp_hdr, header); + + msm_dp_write_link(audio, MMSS_DP_AUDIO_STREAM_0, header[0]); + msm_dp_write_link(audio, MMSS_DP_AUDIO_STREAM_1, header[1]); +} + +static void msm_dp_audio_timestamp_sdp(struct msm_dp_audio_private *audio) +{ + struct dp_sdp_header sdp_hdr = { + .HB0 = 0x00, + .HB1 = 0x01, + .HB2 = 0x17, + .HB3 = 0x0 | (0x11 << 2), + }; + u32 header[2]; + + msm_dp_utils_pack_sdp_header(&sdp_hdr, header); + + msm_dp_write_link(audio, MMSS_DP_AUDIO_TIMESTAMP_0, header[0]); + msm_dp_write_link(audio, MMSS_DP_AUDIO_TIMESTAMP_1, header[1]); +} + +static void msm_dp_audio_infoframe_sdp(struct msm_dp_audio_private *audio) +{ + struct dp_sdp_header sdp_hdr = { + .HB0 = 0x00, + .HB1 = 0x84, + .HB2 = 0x1b, + .HB3 = 0x0 | (0x11 << 2), + }; + u32 header[2]; + + msm_dp_utils_pack_sdp_header(&sdp_hdr, header); + + msm_dp_write_link(audio, MMSS_DP_AUDIO_INFOFRAME_0, header[0]); + msm_dp_write_link(audio, MMSS_DP_AUDIO_INFOFRAME_1, header[1]); +} + +static void msm_dp_audio_copy_management_sdp(struct msm_dp_audio_private *audio) +{ + struct dp_sdp_header sdp_hdr = { + .HB0 = 0x00, + .HB1 = 0x05, + .HB2 = 0x0f, + .HB3 = 0x00, + }; + u32 header[2]; + + msm_dp_utils_pack_sdp_header(&sdp_hdr, header); + + msm_dp_write_link(audio, MMSS_DP_AUDIO_COPYMANAGEMENT_0, header[0]); + msm_dp_write_link(audio, MMSS_DP_AUDIO_COPYMANAGEMENT_1, header[1]); +} + +static void msm_dp_audio_isrc_sdp(struct msm_dp_audio_private *audio) +{ + struct dp_sdp_header sdp_hdr = { + .HB0 = 0x00, + .HB1 = 0x06, + .HB2 = 0x0f, + .HB3 = 0x00, + }; + u32 header[2]; + u32 reg; + + /* XXX: is it necessary to preserve this field? */ + reg = msm_dp_read_link(audio, MMSS_DP_AUDIO_ISRC_1); + sdp_hdr.HB3 = FIELD_GET(HEADER_3_MASK, reg); + + msm_dp_utils_pack_sdp_header(&sdp_hdr, header); + + msm_dp_write_link(audio, MMSS_DP_AUDIO_ISRC_0, header[0]); + msm_dp_write_link(audio, MMSS_DP_AUDIO_ISRC_1, header[1]); +} + +static void msm_dp_audio_config_sdp(struct msm_dp_audio_private *audio) +{ + u32 sdp_cfg, sdp_cfg2; + + sdp_cfg = msm_dp_read_link(audio, MMSS_DP_SDP_CFG); + /* AUDIO_TIMESTAMP_SDP_EN */ + sdp_cfg |= BIT(1); + /* AUDIO_STREAM_SDP_EN */ + sdp_cfg |= BIT(2); + /* AUDIO_COPY_MANAGEMENT_SDP_EN */ + sdp_cfg |= BIT(5); + /* AUDIO_ISRC_SDP_EN */ + sdp_cfg |= BIT(6); + /* AUDIO_INFOFRAME_SDP_EN */ + sdp_cfg |= BIT(20); + + drm_dbg_dp(audio->drm_dev, "sdp_cfg = 0x%x\n", sdp_cfg); + + msm_dp_write_link(audio, MMSS_DP_SDP_CFG, sdp_cfg); + + sdp_cfg2 = msm_dp_read_link(audio, MMSS_DP_SDP_CFG2); + /* IFRM_REGSRC -> Do not use reg values */ + sdp_cfg2 &= ~BIT(0); + /* AUDIO_STREAM_HB3_REGSRC-> Do not use reg values */ + sdp_cfg2 &= ~BIT(1); + + drm_dbg_dp(audio->drm_dev, "sdp_cfg2 = 0x%x\n", sdp_cfg2); + + msm_dp_write_link(audio, MMSS_DP_SDP_CFG2, sdp_cfg2); +} + +static void msm_dp_audio_setup_sdp(struct msm_dp_audio_private *audio) +{ + msm_dp_audio_config_sdp(audio); + + msm_dp_audio_stream_sdp(audio); + msm_dp_audio_timestamp_sdp(audio); + msm_dp_audio_infoframe_sdp(audio); + msm_dp_audio_copy_management_sdp(audio); + msm_dp_audio_isrc_sdp(audio); +} + +static void msm_dp_audio_setup_acr(struct msm_dp_audio_private *audio) +{ + u32 select, acr_ctrl; + + switch (audio->msm_dp_audio.bw_code) { + case DP_LINK_BW_1_62: + select = 0; + break; + case DP_LINK_BW_2_7: + select = 1; + break; + case DP_LINK_BW_5_4: + select = 2; + break; + case DP_LINK_BW_8_1: + select = 3; + break; + default: + drm_dbg_dp(audio->drm_dev, "Unknown link rate\n"); + select = 0; + break; + } + + acr_ctrl = select << 4 | BIT(31) | BIT(8) | BIT(14); + + drm_dbg_dp(audio->drm_dev, "select: %#x, acr_ctrl: %#x\n", + select, acr_ctrl); + + msm_dp_write_link(audio, MMSS_DP_AUDIO_ACR_CTRL, acr_ctrl); +} + +static void msm_dp_audio_safe_to_exit_level(struct msm_dp_audio_private *audio) +{ + u32 safe_to_exit_level, mainlink_levels; + + switch (audio->msm_dp_audio.lane_count) { + case 1: + safe_to_exit_level = 14; + break; + case 2: + safe_to_exit_level = 8; + break; + case 4: + safe_to_exit_level = 5; + break; + default: + safe_to_exit_level = 14; + drm_dbg_dp(audio->drm_dev, + "setting the default safe_to_exit_level = %u\n", + safe_to_exit_level); + break; + } + + mainlink_levels = msm_dp_read_link(audio, REG_DP_MAINLINK_LEVELS); + mainlink_levels &= 0xFE0; + mainlink_levels |= safe_to_exit_level; + + drm_dbg_dp(audio->drm_dev, + "mainlink_level = 0x%x, safe_to_exit_level = 0x%x\n", + mainlink_levels, safe_to_exit_level); + + msm_dp_write_link(audio, REG_DP_MAINLINK_LEVELS, mainlink_levels); +} + +static void msm_dp_audio_enable(struct msm_dp_audio_private *audio, bool enable) +{ + u32 audio_ctrl; + + audio_ctrl = msm_dp_read_link(audio, MMSS_DP_AUDIO_CFG); + + if (enable) + audio_ctrl |= BIT(0); + else + audio_ctrl &= ~BIT(0); + + drm_dbg_dp(audio->drm_dev, "dp_audio_cfg = 0x%x\n", audio_ctrl); + + msm_dp_write_link(audio, MMSS_DP_AUDIO_CFG, audio_ctrl); + /* make sure audio engine is disabled */ + wmb(); +} + +static struct msm_dp_audio_private *msm_dp_audio_get_data(struct msm_dp *msm_dp_display) +{ + struct msm_dp_audio *msm_dp_audio; + + msm_dp_audio = msm_dp_display->msm_dp_audio; + if (!msm_dp_audio) { + DRM_ERROR("invalid msm_dp_audio data\n"); + return ERR_PTR(-EINVAL); + } + + return container_of(msm_dp_audio, struct msm_dp_audio_private, msm_dp_audio); +} + +int msm_dp_audio_prepare(struct drm_bridge *bridge, + struct drm_connector *connector, + struct hdmi_codec_daifmt *daifmt, + struct hdmi_codec_params *params) +{ + int rc = 0; + struct msm_dp_audio_private *audio; + struct msm_dp *msm_dp_display; + + msm_dp_display = to_dp_bridge(bridge)->msm_dp_display; + + /* + * there could be cases where sound card can be opened even + * before OR even when DP is not connected . This can cause + * unclocked access as the audio subsystem relies on the DP + * driver to maintain the correct state of clocks. To protect + * such cases check for connection status and bail out if not + * connected. + */ + if (!msm_dp_display->power_on) + goto end; + + audio = msm_dp_audio_get_data(msm_dp_display); + if (IS_ERR(audio)) { + rc = PTR_ERR(audio); + goto end; + } + + audio->channels = params->channels; + + msm_dp_audio_setup_sdp(audio); + msm_dp_audio_setup_acr(audio); + msm_dp_audio_safe_to_exit_level(audio); + msm_dp_audio_enable(audio, true); + msm_dp_display_signal_audio_start(msm_dp_display); + msm_dp_display->audio_enabled = true; + +end: + return rc; +} + +void msm_dp_audio_shutdown(struct drm_bridge *bridge, + struct drm_connector *connecter) +{ + struct msm_dp_audio_private *audio; + struct msm_dp *msm_dp_display; + + msm_dp_display = to_dp_bridge(bridge)->msm_dp_display; + audio = msm_dp_audio_get_data(msm_dp_display); + if (IS_ERR(audio)) { + DRM_ERROR("failed to get audio data\n"); + return; + } + + /* + * if audio was not enabled there is no need + * to execute the shutdown and we can bail out early. + * This also makes sure that we dont cause an unclocked + * access when audio subsystem calls this without DP being + * connected. is_connected cannot be used here as its set + * to false earlier than this call + */ + if (!msm_dp_display->audio_enabled) + return; + + msm_dp_audio_enable(audio, false); + /* signal the dp display to safely shutdown clocks */ + msm_dp_display_signal_audio_complete(msm_dp_display); +} + +struct msm_dp_audio *msm_dp_audio_get(struct platform_device *pdev, + void __iomem *link_base) +{ + int rc = 0; + struct msm_dp_audio_private *audio; + struct msm_dp_audio *msm_dp_audio; + + if (!pdev) { + DRM_ERROR("invalid input\n"); + rc = -EINVAL; + goto error; + } + + audio = devm_kzalloc(&pdev->dev, sizeof(*audio), GFP_KERNEL); + if (!audio) { + rc = -ENOMEM; + goto error; + } + + audio->pdev = pdev; + audio->link_base = link_base; + + msm_dp_audio = &audio->msm_dp_audio; + + return msm_dp_audio; +error: + return ERR_PTR(rc); +} + +void msm_dp_audio_put(struct msm_dp_audio *msm_dp_audio) +{ + struct msm_dp_audio_private *audio; + + if (!msm_dp_audio) + return; + + audio = container_of(msm_dp_audio, struct msm_dp_audio_private, msm_dp_audio); + + devm_kfree(&audio->pdev->dev, audio); +} diff --git a/drivers/gpu/drm/msm/dp/dp_audio.h b/drivers/gpu/drm/msm/dp/dp_audio.h new file mode 100644 index 000000000..ce2342856 --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_audio.h @@ -0,0 +1,57 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved. + */ + +#ifndef _DP_AUDIO_H_ +#define _DP_AUDIO_H_ + +#include + +#include + +struct drm_bridge; + +/** + * struct msm_dp_audio + * @lane_count: number of lanes configured in current session + * @bw_code: link rate's bandwidth code for current session + */ +struct msm_dp_audio { + u32 lane_count; + u32 bw_code; +}; + +/** + * msm_dp_audio_get() + * + * Creates and instance of dp audio. + * + * @pdev: caller's platform device instance. + * @link_base: pointer to the msm_dp_link resource. + * + * Returns the error code in case of failure, otherwize + * an instance of newly created msm_dp_module. + */ +struct msm_dp_audio *msm_dp_audio_get(struct platform_device *pdev, + void __iomem *link_base); + +/** + * msm_dp_audio_put() + * + * Cleans the msm_dp_audio instance. + * + * @msm_dp_audio: an instance of msm_dp_audio. + */ +void msm_dp_audio_put(struct msm_dp_audio *msm_dp_audio); + +int msm_dp_audio_prepare(struct drm_bridge *bridge, + struct drm_connector *connector, + struct hdmi_codec_daifmt *daifmt, + struct hdmi_codec_params *params); +void msm_dp_audio_shutdown(struct drm_bridge *bridge, + struct drm_connector *connector); + +#endif /* _DP_AUDIO_H_ */ + + diff --git a/drivers/gpu/drm/msm/dp/dp_aux.c b/drivers/gpu/drm/msm/dp/dp_aux.c new file mode 100644 index 000000000..3825a2fb4 --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_aux.c @@ -0,0 +1,729 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved. + */ + +#include +#include +#include +#include + +#include "dp_reg.h" +#include "dp_aux.h" + +enum msm_dp_aux_err { + DP_AUX_ERR_NONE, + DP_AUX_ERR_ADDR, + DP_AUX_ERR_TOUT, + DP_AUX_ERR_NACK, + DP_AUX_ERR_DEFER, + DP_AUX_ERR_NACK_DEFER, + DP_AUX_ERR_PHY, +}; + +struct msm_dp_aux_private { + struct device *dev; + void __iomem *aux_base; + + struct phy *phy; + + struct mutex mutex; + struct completion comp; + + enum msm_dp_aux_err aux_error_num; + u32 retry_cnt; + bool cmd_busy; + bool native; + bool read; + bool no_send_addr; + bool no_send_stop; + bool initted; + bool is_edp; + bool enable_xfers; + u32 offset; + u32 segment; + + struct drm_dp_aux msm_dp_aux; +}; + +static inline u32 msm_dp_read_aux(struct msm_dp_aux_private *aux, u32 offset) +{ + return readl_relaxed(aux->aux_base + offset); +} + +static inline void msm_dp_write_aux(struct msm_dp_aux_private *aux, + u32 offset, u32 data) +{ + /* + * To make sure aux reg writes happens before any other operation, + * this function uses writel() instread of writel_relaxed() + */ + writel(data, aux->aux_base + offset); +} + +static void msm_dp_aux_clear_hw_interrupts(struct msm_dp_aux_private *aux) +{ + msm_dp_read_aux(aux, REG_DP_PHY_AUX_INTERRUPT_STATUS); + msm_dp_write_aux(aux, REG_DP_PHY_AUX_INTERRUPT_CLEAR, 0x1f); + msm_dp_write_aux(aux, REG_DP_PHY_AUX_INTERRUPT_CLEAR, 0x9f); + msm_dp_write_aux(aux, REG_DP_PHY_AUX_INTERRUPT_CLEAR, 0); +} + +/* + * NOTE: resetting AUX controller will also clear any pending HPD related interrupts + */ +static void msm_dp_aux_reset(struct msm_dp_aux_private *aux) +{ + u32 aux_ctrl; + + aux_ctrl = msm_dp_read_aux(aux, REG_DP_AUX_CTRL); + + aux_ctrl |= DP_AUX_CTRL_RESET; + msm_dp_write_aux(aux, REG_DP_AUX_CTRL, aux_ctrl); + usleep_range(1000, 1100); /* h/w recommended delay */ + + aux_ctrl &= ~DP_AUX_CTRL_RESET; + msm_dp_write_aux(aux, REG_DP_AUX_CTRL, aux_ctrl); +} + +static void msm_dp_aux_enable(struct msm_dp_aux_private *aux) +{ + u32 aux_ctrl; + + aux_ctrl = msm_dp_read_aux(aux, REG_DP_AUX_CTRL); + + msm_dp_write_aux(aux, REG_DP_TIMEOUT_COUNT, 0xffff); + msm_dp_write_aux(aux, REG_DP_AUX_LIMITS, 0xffff); + + aux_ctrl |= DP_AUX_CTRL_ENABLE; + msm_dp_write_aux(aux, REG_DP_AUX_CTRL, aux_ctrl); +} + +static void msm_dp_aux_disable(struct msm_dp_aux_private *aux) +{ + u32 aux_ctrl; + + aux_ctrl = msm_dp_read_aux(aux, REG_DP_AUX_CTRL); + aux_ctrl &= ~DP_AUX_CTRL_ENABLE; + msm_dp_write_aux(aux, REG_DP_AUX_CTRL, aux_ctrl); +} + +static int msm_dp_aux_wait_for_hpd_connect_state(struct msm_dp_aux_private *aux, + unsigned long wait_us) +{ + u32 state; + + /* poll for hpd connected status every 2ms and timeout after wait_us */ + return readl_poll_timeout(aux->aux_base + + REG_DP_DP_HPD_INT_STATUS, + state, state & DP_DP_HPD_STATE_STATUS_CONNECTED, + min(wait_us, 2000), wait_us); +} + +#define MAX_AUX_RETRIES 5 + +static ssize_t msm_dp_aux_write(struct msm_dp_aux_private *aux, + struct drm_dp_aux_msg *msg) +{ + u8 data[4]; + u32 reg; + ssize_t len; + u8 *msgdata = msg->buffer; + int const AUX_CMD_FIFO_LEN = 128; + int i = 0; + + if (aux->read) + len = 0; + else + len = msg->size; + + /* + * cmd fifo only has depth of 144 bytes + * limit buf length to 128 bytes here + */ + if (len > AUX_CMD_FIFO_LEN - 4) { + DRM_ERROR("buf size greater than allowed size of 128 bytes\n"); + return -EINVAL; + } + + /* Pack cmd and write to HW */ + data[0] = (msg->address >> 16) & 0xf; /* addr[19:16] */ + if (aux->read) + data[0] |= BIT(4); /* R/W */ + + data[1] = msg->address >> 8; /* addr[15:8] */ + data[2] = msg->address; /* addr[7:0] */ + data[3] = msg->size - 1; /* len[7:0] */ + + for (i = 0; i < len + 4; i++) { + reg = (i < 4) ? data[i] : msgdata[i - 4]; + reg <<= DP_AUX_DATA_OFFSET; + reg &= DP_AUX_DATA_MASK; + reg |= DP_AUX_DATA_WRITE; + /* index = 0, write */ + if (i == 0) + reg |= DP_AUX_DATA_INDEX_WRITE; + msm_dp_write_aux(aux, REG_DP_AUX_DATA, reg); + } + + msm_dp_write_aux(aux, REG_DP_AUX_TRANS_CTRL, 0); + msm_dp_aux_clear_hw_interrupts(aux); + + reg = 0; /* Transaction number == 1 */ + if (!aux->native) { /* i2c */ + reg |= DP_AUX_TRANS_CTRL_I2C; + + if (aux->no_send_addr) + reg |= DP_AUX_TRANS_CTRL_NO_SEND_ADDR; + + if (aux->no_send_stop) + reg |= DP_AUX_TRANS_CTRL_NO_SEND_STOP; + } + + reg |= DP_AUX_TRANS_CTRL_GO; + msm_dp_write_aux(aux, REG_DP_AUX_TRANS_CTRL, reg); + + return len; +} + +static ssize_t msm_dp_aux_cmd_fifo_tx(struct msm_dp_aux_private *aux, + struct drm_dp_aux_msg *msg) +{ + ssize_t ret; + unsigned long time_left; + + reinit_completion(&aux->comp); + + ret = msm_dp_aux_write(aux, msg); + if (ret < 0) + return ret; + + time_left = wait_for_completion_timeout(&aux->comp, + msecs_to_jiffies(250)); + if (!time_left) + return -ETIMEDOUT; + + return ret; +} + +static ssize_t msm_dp_aux_cmd_fifo_rx(struct msm_dp_aux_private *aux, + struct drm_dp_aux_msg *msg) +{ + u32 data; + u8 *dp; + u32 i, actual_i; + u32 len = msg->size; + + data = msm_dp_read_aux(aux, REG_DP_AUX_TRANS_CTRL); + data &= ~DP_AUX_TRANS_CTRL_GO; + msm_dp_write_aux(aux, REG_DP_AUX_TRANS_CTRL, data); + + data = DP_AUX_DATA_INDEX_WRITE; /* INDEX_WRITE */ + data |= DP_AUX_DATA_READ; /* read */ + + msm_dp_write_aux(aux, REG_DP_AUX_DATA, data); + + dp = msg->buffer; + + /* discard first byte */ + data = msm_dp_read_aux(aux, REG_DP_AUX_DATA); + + for (i = 0; i < len; i++) { + data = msm_dp_read_aux(aux, REG_DP_AUX_DATA); + *dp++ = (u8)((data >> DP_AUX_DATA_OFFSET) & 0xff); + + actual_i = (data >> DP_AUX_DATA_INDEX_OFFSET) & 0xFF; + if (i != actual_i) + break; + } + + return i; +} + +static void msm_dp_aux_update_offset_and_segment(struct msm_dp_aux_private *aux, + struct drm_dp_aux_msg *input_msg) +{ + u32 edid_address = 0x50; + u32 segment_address = 0x30; + bool i2c_read = input_msg->request & + (DP_AUX_I2C_READ & DP_AUX_NATIVE_READ); + u8 *data; + + if (aux->native || i2c_read || ((input_msg->address != edid_address) && + (input_msg->address != segment_address))) + return; + + + data = input_msg->buffer; + if (input_msg->address == segment_address) + aux->segment = *data; + else + aux->offset = *data; +} + +/** + * msm_dp_aux_transfer_helper() - helper function for EDID read transactions + * + * @aux: DP AUX private structure + * @input_msg: input message from DRM upstream APIs + * @send_seg: send the segment to sink + * + * return: void + * + * This helper function is used to fix EDID reads for non-compliant + * sinks that do not handle the i2c middle-of-transaction flag correctly. + */ +static void msm_dp_aux_transfer_helper(struct msm_dp_aux_private *aux, + struct drm_dp_aux_msg *input_msg, + bool send_seg) +{ + struct drm_dp_aux_msg helper_msg; + u32 message_size = 0x10; + u32 segment_address = 0x30; + u32 const edid_block_length = 0x80; + bool i2c_mot = input_msg->request & DP_AUX_I2C_MOT; + bool i2c_read = input_msg->request & + (DP_AUX_I2C_READ & DP_AUX_NATIVE_READ); + + if (!i2c_mot || !i2c_read || (input_msg->size == 0)) + return; + + /* + * Sending the segment value and EDID offset will be performed + * from the DRM upstream EDID driver for each block. Avoid + * duplicate AUX transactions related to this while reading the + * first 16 bytes of each block. + */ + if (!(aux->offset % edid_block_length) || !send_seg) + goto end; + + aux->read = false; + aux->cmd_busy = true; + aux->no_send_addr = true; + aux->no_send_stop = true; + + /* + * Send the segment address for every i2c read in which the + * middle-of-tranaction flag is set. This is required to support EDID + * reads of more than 2 blocks as the segment address is reset to 0 + * since we are overriding the middle-of-transaction flag for read + * transactions. + */ + + if (aux->segment) { + memset(&helper_msg, 0, sizeof(helper_msg)); + helper_msg.address = segment_address; + helper_msg.buffer = &aux->segment; + helper_msg.size = 1; + msm_dp_aux_cmd_fifo_tx(aux, &helper_msg); + } + + /* + * Send the offset address for every i2c read in which the + * middle-of-transaction flag is set. This will ensure that the sink + * will update its read pointer and return the correct portion of the + * EDID buffer in the subsequent i2c read trasntion triggered in the + * native AUX transfer function. + */ + memset(&helper_msg, 0, sizeof(helper_msg)); + helper_msg.address = input_msg->address; + helper_msg.buffer = &aux->offset; + helper_msg.size = 1; + msm_dp_aux_cmd_fifo_tx(aux, &helper_msg); + +end: + aux->offset += message_size; + if (aux->offset == 0x80 || aux->offset == 0x100) + aux->segment = 0x0; /* reset segment at end of block */ +} + +/* + * This function does the real job to process an AUX transaction. + * It will call aux_reset() function to reset the AUX channel, + * if the waiting is timeout. + */ +static ssize_t msm_dp_aux_transfer(struct drm_dp_aux *msm_dp_aux, + struct drm_dp_aux_msg *msg) +{ + ssize_t ret; + int const aux_cmd_native_max = 16; + int const aux_cmd_i2c_max = 128; + struct msm_dp_aux_private *aux; + + aux = container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + + aux->native = msg->request & (DP_AUX_NATIVE_WRITE & DP_AUX_NATIVE_READ); + + /* Ignore address only message */ + if (msg->size == 0 || !msg->buffer) { + msg->reply = aux->native ? + DP_AUX_NATIVE_REPLY_ACK : DP_AUX_I2C_REPLY_ACK; + return msg->size; + } + + /* msg sanity check */ + if ((aux->native && msg->size > aux_cmd_native_max) || + msg->size > aux_cmd_i2c_max) { + DRM_ERROR("%s: invalid msg: size(%zu), request(%x)\n", + __func__, msg->size, msg->request); + return -EINVAL; + } + + ret = pm_runtime_resume_and_get(msm_dp_aux->dev); + if (ret) + return ret; + + mutex_lock(&aux->mutex); + if (!aux->initted) { + ret = -EIO; + goto exit; + } + + /* + * If we're using DP and an external display isn't connected then the + * transfer won't succeed. Return right away. If we don't do this we + * can end up with long timeouts if someone tries to access the DP AUX + * character device when no DP device is connected. + */ + if (!aux->is_edp && !aux->enable_xfers) { + ret = -ENXIO; + goto exit; + } + + msm_dp_aux_update_offset_and_segment(aux, msg); + msm_dp_aux_transfer_helper(aux, msg, true); + + aux->read = msg->request & (DP_AUX_I2C_READ & DP_AUX_NATIVE_READ); + aux->cmd_busy = true; + + if (aux->read) { + aux->no_send_addr = true; + aux->no_send_stop = false; + } else { + aux->no_send_addr = true; + aux->no_send_stop = true; + } + + ret = msm_dp_aux_cmd_fifo_tx(aux, msg); + if (ret < 0) { + if (aux->native) { + aux->retry_cnt++; + if (!(aux->retry_cnt % MAX_AUX_RETRIES)) + phy_calibrate(aux->phy); + } + /* reset aux if link is in connected state */ + if (msm_dp_aux_is_link_connected(msm_dp_aux)) + msm_dp_aux_reset(aux); + } else { + aux->retry_cnt = 0; + switch (aux->aux_error_num) { + case DP_AUX_ERR_NONE: + if (aux->read) + ret = msm_dp_aux_cmd_fifo_rx(aux, msg); + msg->reply = aux->native ? DP_AUX_NATIVE_REPLY_ACK : DP_AUX_I2C_REPLY_ACK; + break; + case DP_AUX_ERR_DEFER: + msg->reply = aux->native ? DP_AUX_NATIVE_REPLY_DEFER : DP_AUX_I2C_REPLY_DEFER; + break; + case DP_AUX_ERR_PHY: + case DP_AUX_ERR_ADDR: + case DP_AUX_ERR_NACK: + case DP_AUX_ERR_NACK_DEFER: + msg->reply = aux->native ? DP_AUX_NATIVE_REPLY_NACK : DP_AUX_I2C_REPLY_NACK; + break; + case DP_AUX_ERR_TOUT: + ret = -ETIMEDOUT; + break; + } + } + + aux->cmd_busy = false; + +exit: + mutex_unlock(&aux->mutex); + pm_runtime_put_sync(msm_dp_aux->dev); + + return ret; +} + +irqreturn_t msm_dp_aux_isr(struct drm_dp_aux *msm_dp_aux, u32 isr) +{ + struct msm_dp_aux_private *aux; + + if (!msm_dp_aux) { + DRM_ERROR("invalid input\n"); + return IRQ_NONE; + } + + aux = container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + + if (!aux->cmd_busy) { + DRM_ERROR("Unexpected DP AUX IRQ %#010x when not busy\n", isr); + return IRQ_NONE; + } + + /* + * The logic below assumes only one error bit is set (other than "done" + * which can apparently be set at the same time as some of the other + * bits). Warn if more than one get set so we know we need to improve + * the logic. + */ + if (hweight32(isr & ~DP_INTR_AUX_XFER_DONE) > 1) + DRM_WARN("Some DP AUX interrupts unhandled: %#010x\n", isr); + + if (isr & DP_INTR_AUX_ERROR) { + aux->aux_error_num = DP_AUX_ERR_PHY; + msm_dp_aux_clear_hw_interrupts(aux); + } else if (isr & DP_INTR_NACK_DEFER) { + aux->aux_error_num = DP_AUX_ERR_NACK_DEFER; + } else if (isr & DP_INTR_WRONG_ADDR) { + aux->aux_error_num = DP_AUX_ERR_ADDR; + } else if (isr & DP_INTR_TIMEOUT) { + aux->aux_error_num = DP_AUX_ERR_TOUT; + } else if (!aux->native && (isr & DP_INTR_I2C_NACK)) { + aux->aux_error_num = DP_AUX_ERR_NACK; + } else if (!aux->native && (isr & DP_INTR_I2C_DEFER)) { + if (isr & DP_INTR_AUX_XFER_DONE) + aux->aux_error_num = DP_AUX_ERR_NACK; + else + aux->aux_error_num = DP_AUX_ERR_DEFER; + } else if (isr & DP_INTR_AUX_XFER_DONE) { + aux->aux_error_num = DP_AUX_ERR_NONE; + } else { + DRM_WARN("Unexpected interrupt: %#010x\n", isr); + return IRQ_NONE; + } + + complete(&aux->comp); + + return IRQ_HANDLED; +} + +void msm_dp_aux_enable_xfers(struct drm_dp_aux *msm_dp_aux, bool enabled) +{ + struct msm_dp_aux_private *aux; + + aux = container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + aux->enable_xfers = enabled; +} + +void msm_dp_aux_reconfig(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux; + + aux = container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + + phy_calibrate(aux->phy); + msm_dp_aux_reset(aux); +} + +void msm_dp_aux_init(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux; + + if (!msm_dp_aux) { + DRM_ERROR("invalid input\n"); + return; + } + + aux = container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + + mutex_lock(&aux->mutex); + + msm_dp_aux_enable(aux); + aux->retry_cnt = 0; + aux->initted = true; + + mutex_unlock(&aux->mutex); +} + +void msm_dp_aux_deinit(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux; + + aux = container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + + mutex_lock(&aux->mutex); + + aux->initted = false; + msm_dp_aux_disable(aux); + + mutex_unlock(&aux->mutex); +} + +int msm_dp_aux_register(struct drm_dp_aux *msm_dp_aux) +{ + int ret; + + if (!msm_dp_aux) { + DRM_ERROR("invalid input\n"); + return -EINVAL; + } + + ret = drm_dp_aux_register(msm_dp_aux); + if (ret) { + DRM_ERROR("%s: failed to register drm aux: %d\n", __func__, + ret); + return ret; + } + + return 0; +} + +void msm_dp_aux_unregister(struct drm_dp_aux *msm_dp_aux) +{ + drm_dp_aux_unregister(msm_dp_aux); +} + +static int msm_dp_wait_hpd_asserted(struct drm_dp_aux *msm_dp_aux, + unsigned long wait_us) +{ + int ret; + struct msm_dp_aux_private *aux; + + aux = container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + + ret = pm_runtime_resume_and_get(aux->dev); + if (ret) + return ret; + + ret = msm_dp_aux_wait_for_hpd_connect_state(aux, wait_us); + pm_runtime_put_sync(aux->dev); + + return ret; +} + +void msm_dp_aux_hpd_enable(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux = + container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + u32 reg; + + /* Configure REFTIMER and enable it */ + reg = msm_dp_read_aux(aux, REG_DP_DP_HPD_REFTIMER); + reg |= DP_DP_HPD_REFTIMER_ENABLE; + msm_dp_write_aux(aux, REG_DP_DP_HPD_REFTIMER, reg); + + /* Enable HPD */ + msm_dp_write_aux(aux, REG_DP_DP_HPD_CTRL, DP_DP_HPD_CTRL_HPD_EN); +} + +void msm_dp_aux_hpd_disable(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux = + container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + u32 reg; + + reg = msm_dp_read_aux(aux, REG_DP_DP_HPD_REFTIMER); + reg &= ~DP_DP_HPD_REFTIMER_ENABLE; + msm_dp_write_aux(aux, REG_DP_DP_HPD_REFTIMER, reg); + + msm_dp_write_aux(aux, REG_DP_DP_HPD_CTRL, 0); +} + +void msm_dp_aux_hpd_intr_enable(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux = + container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + u32 reg; + + reg = msm_dp_read_aux(aux, REG_DP_DP_HPD_INT_MASK); + reg |= DP_DP_HPD_INT_MASK; + msm_dp_write_aux(aux, REG_DP_DP_HPD_INT_MASK, + reg & DP_DP_HPD_INT_MASK); +} + +void msm_dp_aux_hpd_intr_disable(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux = + container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + u32 reg; + + reg = msm_dp_read_aux(aux, REG_DP_DP_HPD_INT_MASK); + reg &= ~DP_DP_HPD_INT_MASK; + msm_dp_write_aux(aux, REG_DP_DP_HPD_INT_MASK, + reg & DP_DP_HPD_INT_MASK); +} + +u32 msm_dp_aux_get_hpd_intr_status(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux = + container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + int isr, mask; + + isr = msm_dp_read_aux(aux, REG_DP_DP_HPD_INT_STATUS); + msm_dp_write_aux(aux, REG_DP_DP_HPD_INT_ACK, + (isr & DP_DP_HPD_INT_MASK)); + mask = msm_dp_read_aux(aux, REG_DP_DP_HPD_INT_MASK); + + /* + * We only want to return interrupts that are unmasked to the caller. + * However, the interrupt status field also contains other + * informational bits about the HPD state status, so we only mask + * out the part of the register that tells us about which interrupts + * are pending. + */ + return isr & (mask | ~DP_DP_HPD_INT_MASK); +} + +u32 msm_dp_aux_is_link_connected(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux = + container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + u32 status; + + status = msm_dp_read_aux(aux, REG_DP_DP_HPD_INT_STATUS); + status >>= DP_DP_HPD_STATE_STATUS_BITS_SHIFT; + status &= DP_DP_HPD_STATE_STATUS_BITS_MASK; + + return status; +} + +struct drm_dp_aux *msm_dp_aux_get(struct device *dev, + struct phy *phy, + bool is_edp, + void __iomem *aux_base) +{ + struct msm_dp_aux_private *aux; + + aux = devm_kzalloc(dev, sizeof(*aux), GFP_KERNEL); + if (!aux) + return ERR_PTR(-ENOMEM); + + init_completion(&aux->comp); + aux->cmd_busy = false; + aux->is_edp = is_edp; + mutex_init(&aux->mutex); + + aux->dev = dev; + aux->phy = phy; + aux->retry_cnt = 0; + aux->aux_base = aux_base; + + /* + * Use the drm_dp_aux_init() to use the aux adapter + * before registering AUX with the DRM device so that + * msm eDP panel can be detected by generic_dep_panel_probe(). + */ + aux->msm_dp_aux.name = "dpu_dp_aux"; + aux->msm_dp_aux.dev = dev; + aux->msm_dp_aux.transfer = msm_dp_aux_transfer; + aux->msm_dp_aux.wait_hpd_asserted = msm_dp_wait_hpd_asserted; + drm_dp_aux_init(&aux->msm_dp_aux); + + return &aux->msm_dp_aux; +} + +void msm_dp_aux_put(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux; + + if (!msm_dp_aux) + return; + + aux = container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + + mutex_destroy(&aux->mutex); + + devm_kfree(aux->dev, aux); +} diff --git a/drivers/gpu/drm/msm/dp/dp_aux.h b/drivers/gpu/drm/msm/dp/dp_aux.h new file mode 100644 index 000000000..4be02e8b4 --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_aux.h @@ -0,0 +1,33 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved. + */ + +#ifndef _DP_AUX_H_ +#define _DP_AUX_H_ + +#include + +int msm_dp_aux_register(struct drm_dp_aux *msm_dp_aux); +void msm_dp_aux_unregister(struct drm_dp_aux *msm_dp_aux); +irqreturn_t msm_dp_aux_isr(struct drm_dp_aux *msm_dp_aux, u32 isr); +void msm_dp_aux_enable_xfers(struct drm_dp_aux *msm_dp_aux, bool enabled); +void msm_dp_aux_init(struct drm_dp_aux *msm_dp_aux); +void msm_dp_aux_deinit(struct drm_dp_aux *msm_dp_aux); +void msm_dp_aux_reconfig(struct drm_dp_aux *msm_dp_aux); + +void msm_dp_aux_hpd_enable(struct drm_dp_aux *msm_dp_aux); +void msm_dp_aux_hpd_disable(struct drm_dp_aux *msm_dp_aux); +void msm_dp_aux_hpd_intr_enable(struct drm_dp_aux *msm_dp_aux); +void msm_dp_aux_hpd_intr_disable(struct drm_dp_aux *msm_dp_aux); +u32 msm_dp_aux_get_hpd_intr_status(struct drm_dp_aux *msm_dp_aux); +u32 msm_dp_aux_is_link_connected(struct drm_dp_aux *msm_dp_aux); + +struct phy; +struct drm_dp_aux *msm_dp_aux_get(struct device *dev, + struct phy *phy, + bool is_edp, + void __iomem *aux_base); +void msm_dp_aux_put(struct drm_dp_aux *aux); + +#endif /*__DP_AUX_H_*/ diff --git a/drivers/gpu/drm/msm/dp/dp_ctrl.c b/drivers/gpu/drm/msm/dp/dp_ctrl.c new file mode 100644 index 000000000..59070f399 --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_ctrl.c @@ -0,0 +1,2813 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved. + */ + +#define pr_fmt(fmt) "[drm-dp] %s: " fmt, __func__ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +#include "dp_reg.h" +#include "dp_ctrl.h" +#include "dp_link.h" + +#define POLLING_SLEEP_US 1000 +#define POLLING_TIMEOUT_US 10000 + +#define DP_KHZ_TO_HZ 1000 +#define IDLE_PATTERN_COMPLETION_TIMEOUT_JIFFIES (30 * HZ / 1000) /* 30 ms */ +#define PSR_OPERATION_COMPLETION_TIMEOUT_JIFFIES (300 * HZ / 1000) /* 300 ms */ +#define WAIT_FOR_VIDEO_READY_TIMEOUT_JIFFIES (HZ / 2) + +#define DP_INTERRUPT_STATUS_ACK_SHIFT 1 +#define DP_INTERRUPT_STATUS_MASK_SHIFT 2 + +#define DP_INTERRUPT_STATUS1 \ + (DP_INTR_AUX_XFER_DONE| \ + DP_INTR_WRONG_ADDR | DP_INTR_TIMEOUT | \ + DP_INTR_NACK_DEFER | DP_INTR_WRONG_DATA_CNT | \ + DP_INTR_I2C_NACK | DP_INTR_I2C_DEFER | \ + DP_INTR_PLL_UNLOCKED | DP_INTR_AUX_ERROR) + +#define DP_INTERRUPT_STATUS1_ACK \ + (DP_INTERRUPT_STATUS1 << DP_INTERRUPT_STATUS_ACK_SHIFT) +#define DP_INTERRUPT_STATUS1_MASK \ + (DP_INTERRUPT_STATUS1 << DP_INTERRUPT_STATUS_MASK_SHIFT) + +#define DP_INTERRUPT_STATUS2 \ + (DP_INTR_READY_FOR_VIDEO | DP_INTR_IDLE_PATTERN_SENT | \ + DP_INTR_FRAME_END | DP_INTR_CRC_UPDATED) + +#define DP_INTERRUPT_STATUS2_ACK \ + (DP_INTERRUPT_STATUS2 << DP_INTERRUPT_STATUS_ACK_SHIFT) +#define DP_INTERRUPT_STATUS2_MASK \ + (DP_INTERRUPT_STATUS2 << DP_INTERRUPT_STATUS_MASK_SHIFT) + +#define DP_INTERRUPT_STATUS4 \ + (PSR_UPDATE_INT | PSR_CAPTURE_INT | PSR_EXIT_INT | \ + PSR_UPDATE_ERROR_INT | PSR_WAKE_ERROR_INT) + +#define DP_INTERRUPT_MASK4 \ + (PSR_UPDATE_MASK | PSR_CAPTURE_MASK | PSR_EXIT_MASK | \ + PSR_UPDATE_ERROR_MASK | PSR_WAKE_ERROR_MASK) + +#define DP_CTRL_INTR_READY_FOR_VIDEO BIT(0) +#define DP_CTRL_INTR_IDLE_PATTERN_SENT BIT(3) + +#define MR_LINK_TRAINING1 0x8 +#define MR_LINK_SYMBOL_ERM 0x80 +#define MR_LINK_PRBS7 0x100 +#define MR_LINK_CUSTOM80 0x200 +#define MR_LINK_TRAINING4 0x40 + +enum { + DP_TRAINING_NONE, + DP_TRAINING_1, + DP_TRAINING_2, +}; + +struct msm_dp_tu_calc_input { + u64 lclk; /* 162, 270, 540 and 810 */ + u64 pclk_khz; /* in KHz */ + u64 hactive; /* active h-width */ + u64 hporch; /* bp + fp + pulse */ + int nlanes; /* no.of.lanes */ + int bpp; /* bits */ + int pixel_enc; /* 444, 420, 422 */ + int dsc_en; /* dsc on/off */ + int async_en; /* async mode */ + int fec_en; /* fec */ + int compress_ratio; /* 2:1 = 200, 3:1 = 300, 3.75:1 = 375 */ + int num_of_dsc_slices; /* number of slices per line */ +}; + +struct msm_dp_vc_tu_mapping_table { + u32 vic; + u8 lanes; + u8 lrate; /* DP_LINK_RATE -> 162(6), 270(10), 540(20), 810 (30) */ + u8 bpp; + u8 valid_boundary_link; + u16 delay_start_link; + bool boundary_moderation_en; + u8 valid_lower_boundary_link; + u8 upper_boundary_count; + u8 lower_boundary_count; + u8 tu_size_minus1; +}; + +struct msm_dp_ctrl_private { + struct msm_dp_ctrl msm_dp_ctrl; + struct drm_device *drm_dev; + struct device *dev; + struct drm_dp_aux *aux; + struct msm_dp_link *link; + void __iomem *ahb_base; + void __iomem *link_base; + + struct phy *phy; + + unsigned int num_core_clks; + struct clk_bulk_data *core_clks; + + unsigned int num_link_clks; + struct clk_bulk_data *link_clks; + + struct clk *pixel_clk; + + union phy_configure_opts phy_opts; + + struct completion idle_comp; + struct completion psr_op_comp; + struct completion video_comp; + + u32 hw_revision; + + bool core_clks_on; + bool link_clks_on; + bool stream_clks_on; +}; + +static inline u32 msm_dp_read_ahb(const struct msm_dp_ctrl_private *ctrl, u32 offset) +{ + return readl_relaxed(ctrl->ahb_base + offset); +} + +static inline void msm_dp_write_ahb(struct msm_dp_ctrl_private *ctrl, + u32 offset, u32 data) +{ + /* + * To make sure phy reg writes happens before any other operation, + * this function uses writel() instread of writel_relaxed() + */ + writel(data, ctrl->ahb_base + offset); +} + +static inline u32 msm_dp_read_link(struct msm_dp_ctrl_private *ctrl, u32 offset) +{ + return readl_relaxed(ctrl->link_base + offset); +} + +static inline void msm_dp_write_link(struct msm_dp_ctrl_private *ctrl, + u32 offset, u32 data) +{ + /* + * To make sure link reg writes happens before any other operation, + * this function uses writel() instread of writel_relaxed() + */ + writel(data, ctrl->link_base + offset); +} + +static int msm_dp_aux_link_configure(struct drm_dp_aux *aux, + struct msm_dp_link_info *link) +{ + u8 lane_count, bw_code; + int err; + + lane_count = link->num_lanes; + + if (link->capabilities & DP_LINK_CAP_ENHANCED_FRAMING) + lane_count |= DP_LANE_COUNT_ENHANCED_FRAME_EN; + + err = drm_dp_dpcd_writeb(aux, DP_LANE_COUNT_SET, lane_count); + if (err < 0) + return err; + + if (link->use_rate_set) { + DRM_DEBUG_DP("using LINK_RATE_SET: 0x%02x", link->rate_set); + err = drm_dp_dpcd_writeb(aux, DP_LINK_RATE_SET, link->rate_set); + } else { + bw_code = drm_dp_link_rate_to_bw_code(link->rate); + DRM_DEBUG_DP("using LINK_BW_SET: 0x%02x", bw_code); + err = drm_dp_dpcd_writeb(aux, DP_LINK_BW_SET, bw_code); + } + + return err; +} + +/* + * NOTE: resetting DP controller will also clear any pending HPD related interrupts + */ +void msm_dp_ctrl_reset(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel) +{ + struct msm_dp_ctrl_private *ctrl = + container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + u32 sw_reset; + + sw_reset = msm_dp_read_ahb(ctrl, REG_DP_SW_RESET); + + sw_reset |= DP_SW_RESET; + msm_dp_write_ahb(ctrl, REG_DP_SW_RESET, sw_reset); + usleep_range(1000, 1100); /* h/w recommended delay */ + + sw_reset &= ~DP_SW_RESET; + msm_dp_write_ahb(ctrl, REG_DP_SW_RESET, sw_reset); + + if (!ctrl->hw_revision) + ctrl->hw_revision = msm_dp_read_ahb(ctrl, REG_DP_HW_VERSION); + panel->hw_revision = ctrl->hw_revision; +} + +static u32 msm_dp_ctrl_get_aux_interrupt(struct msm_dp_ctrl_private *ctrl) +{ + u32 intr, intr_ack; + + intr = msm_dp_read_ahb(ctrl, REG_DP_INTR_STATUS); + intr &= ~DP_INTERRUPT_STATUS1_MASK; + intr_ack = (intr & DP_INTERRUPT_STATUS1) + << DP_INTERRUPT_STATUS_ACK_SHIFT; + msm_dp_write_ahb(ctrl, REG_DP_INTR_STATUS, + intr_ack | DP_INTERRUPT_STATUS1_MASK); + + return intr; + +} + +static u32 msm_dp_ctrl_get_interrupt(struct msm_dp_ctrl_private *ctrl) +{ + u32 intr, intr_ack; + + intr = msm_dp_read_ahb(ctrl, REG_DP_INTR_STATUS2); + intr &= ~DP_INTERRUPT_STATUS2_MASK; + intr_ack = (intr & DP_INTERRUPT_STATUS2) + << DP_INTERRUPT_STATUS_ACK_SHIFT; + msm_dp_write_ahb(ctrl, REG_DP_INTR_STATUS2, + intr_ack | DP_INTERRUPT_STATUS2_MASK); + + return intr; +} + +void msm_dp_ctrl_enable_irq(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl = + container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + + msm_dp_write_ahb(ctrl, REG_DP_INTR_STATUS, + DP_INTERRUPT_STATUS1_MASK); + msm_dp_write_ahb(ctrl, REG_DP_INTR_STATUS2, + DP_INTERRUPT_STATUS2_MASK); +} + +void msm_dp_ctrl_disable_irq(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl = + container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + + msm_dp_write_ahb(ctrl, REG_DP_INTR_STATUS, 0x00); + msm_dp_write_ahb(ctrl, REG_DP_INTR_STATUS2, 0x00); +} + +static u32 msm_dp_ctrl_get_psr_interrupt(struct msm_dp_ctrl_private *ctrl) +{ + u32 intr, intr_ack; + + intr = msm_dp_read_ahb(ctrl, REG_DP_INTR_STATUS4); + intr_ack = (intr & DP_INTERRUPT_STATUS4) + << DP_INTERRUPT_STATUS_ACK_SHIFT; + msm_dp_write_ahb(ctrl, REG_DP_INTR_STATUS4, intr_ack); + + return intr; +} + +static void msm_dp_ctrl_config_psr_interrupt(struct msm_dp_ctrl_private *ctrl) +{ + msm_dp_write_ahb(ctrl, REG_DP_INTR_MASK4, DP_INTERRUPT_MASK4); +} + +static void msm_dp_ctrl_psr_mainlink_enable(struct msm_dp_ctrl_private *ctrl) +{ + u32 val; + + val = msm_dp_read_link(ctrl, REG_DP_MAINLINK_CTRL); + val |= DP_MAINLINK_CTRL_ENABLE; + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, val); +} + +static void msm_dp_ctrl_psr_mainlink_disable(struct msm_dp_ctrl_private *ctrl) +{ + u32 val; + + val = msm_dp_read_link(ctrl, REG_DP_MAINLINK_CTRL); + val &= ~DP_MAINLINK_CTRL_ENABLE; + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, val); +} + +static void msm_dp_ctrl_mainlink_enable(struct msm_dp_ctrl_private *ctrl) +{ + u32 mainlink_ctrl; + + drm_dbg_dp(ctrl->drm_dev, "enable\n"); + + mainlink_ctrl = msm_dp_read_link(ctrl, REG_DP_MAINLINK_CTRL); + + mainlink_ctrl &= ~(DP_MAINLINK_CTRL_RESET | + DP_MAINLINK_CTRL_ENABLE); + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, mainlink_ctrl); + + mainlink_ctrl |= DP_MAINLINK_CTRL_RESET; + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, mainlink_ctrl); + + mainlink_ctrl &= ~DP_MAINLINK_CTRL_RESET; + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, mainlink_ctrl); + + mainlink_ctrl |= (DP_MAINLINK_CTRL_ENABLE | + DP_MAINLINK_FB_BOUNDARY_SEL); + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, mainlink_ctrl); +} + +static void msm_dp_ctrl_mainlink_disable(struct msm_dp_ctrl_private *ctrl) +{ + u32 mainlink_ctrl; + + drm_dbg_dp(ctrl->drm_dev, "disable\n"); + + mainlink_ctrl = msm_dp_read_link(ctrl, REG_DP_MAINLINK_CTRL); + mainlink_ctrl &= ~DP_MAINLINK_CTRL_ENABLE; + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, mainlink_ctrl); +} + +static void msm_dp_setup_peripheral_flush(struct msm_dp_ctrl_private *ctrl) +{ + u32 mainlink_ctrl; + + mainlink_ctrl = msm_dp_read_link(ctrl, REG_DP_MAINLINK_CTRL); + + if (ctrl->hw_revision >= DP_HW_VERSION_1_2) + mainlink_ctrl |= DP_MAINLINK_FLUSH_MODE_SDE_PERIPH_UPDATE; + else + mainlink_ctrl |= DP_MAINLINK_FLUSH_MODE_UPDATE_SDP; + + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, mainlink_ctrl); +} + +static bool msm_dp_ctrl_mainlink_ready(struct msm_dp_ctrl_private *ctrl) +{ + u32 data; + int ret; + + /* Poll for mainlink ready status */ + ret = readl_poll_timeout(ctrl->link_base + REG_DP_MAINLINK_READY, + data, data & DP_MAINLINK_READY_FOR_VIDEO, + POLLING_SLEEP_US, POLLING_TIMEOUT_US); + if (ret < 0) { + DRM_ERROR("mainlink not ready\n"); + return false; + } + + return true; +} + +void msm_dp_ctrl_push_idle(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + + reinit_completion(&ctrl->idle_comp); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, DP_STATE_CTRL_PUSH_IDLE); + + if (!wait_for_completion_timeout(&ctrl->idle_comp, + IDLE_PATTERN_COMPLETION_TIMEOUT_JIFFIES)) + pr_warn("PUSH_IDLE pattern timedout\n"); + + drm_dbg_dp(ctrl->drm_dev, "mainlink off\n"); +} + +static void msm_dp_ctrl_config_ctrl_streams(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *msm_dp_panel) +{ + u32 config = 0, tbd; + + /* + * RMW: called from atomic_enable(), serialized by the DRM atomic framework. + * Clear stream-specific fields before OR-ing to avoid bit accumulation. + */ + config = msm_dp_read_link(ctrl, REG_DP_CONFIGURATION_CTRL); + config &= ~(DP_CONFIGURATION_CTRL_RGB_YUV_MASK | + DP_CONFIGURATION_CTRL_BPC_MASK | + DP_CONFIGURATION_CTRL_SEND_VSC); + + if (msm_dp_panel->msm_dp_mode.out_fmt_is_yuv_420) + config |= DP_CONFIGURATION_CTRL_RGB_YUV; /* YUV420 */ + + tbd = msm_dp_link_get_test_bits_depth(ctrl->link, + msm_dp_panel->msm_dp_mode.bpp); + + config |= tbd << DP_CONFIGURATION_CTRL_BPC_SHIFT; + + if (msm_dp_panel->psr_cap.version) + config |= DP_CONFIGURATION_CTRL_SEND_VSC; + + drm_dbg_dp(ctrl->drm_dev, "stream DP_CONFIGURATION_CTRL=0x%x\n", config); + + msm_dp_write_link(ctrl, REG_DP_CONFIGURATION_CTRL, config); +} + +static void msm_dp_ctrl_config_ctrl_link(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel) +{ + u32 config = 0; + const u8 *dpcd = panel->dpcd; + + /* Default-> LSCLK DIV: 1/4 LCLK */ + config |= (2 << DP_CONFIGURATION_CTRL_LSCLK_DIV_SHIFT); + + /* Scrambler reset enable */ + if (drm_dp_alternate_scrambler_reset_cap(dpcd)) + config |= DP_CONFIGURATION_CTRL_ASSR; + + /* Num of Lanes */ + config |= ((ctrl->link->link_params.num_lanes - 1) + << DP_CONFIGURATION_CTRL_NUM_OF_LANES_SHIFT); + + if (drm_dp_enhanced_frame_cap(dpcd)) + config |= DP_CONFIGURATION_CTRL_ENHANCED_FRAMING; + + config |= DP_CONFIGURATION_CTRL_P_INTERLACED; /* progressive video */ + + /* sync clock & static Mvid */ + config |= DP_CONFIGURATION_CTRL_STATIC_DYNAMIC_CN; + config |= DP_CONFIGURATION_CTRL_SYNC_ASYNC_CLK; + + drm_dbg_dp(ctrl->drm_dev, "link DP_CONFIGURATION_CTRL=0x%x\n", config); + + msm_dp_write_link(ctrl, REG_DP_CONFIGURATION_CTRL, config); +} + +static void msm_dp_ctrl_lane_mapping(struct msm_dp_ctrl_private *ctrl) +{ + u32 *lane_map = ctrl->link->lane_map; + u32 ln_mapping; + + ln_mapping = lane_map[0] << LANE0_MAPPING_SHIFT; + ln_mapping |= lane_map[1] << LANE1_MAPPING_SHIFT; + ln_mapping |= lane_map[2] << LANE2_MAPPING_SHIFT; + ln_mapping |= lane_map[3] << LANE3_MAPPING_SHIFT; + + msm_dp_write_link(ctrl, REG_DP_LOGICAL2PHYSICAL_LANE_MAPPING, + ln_mapping); +} + +static void msm_dp_ctrl_config_misc1_misc0(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *msm_dp_panel) +{ + u32 colorimetry_cfg, test_bits_depth, misc_val; + + test_bits_depth = msm_dp_link_get_test_bits_depth(ctrl->link, + msm_dp_panel->msm_dp_mode.bpp); + colorimetry_cfg = msm_dp_link_get_colorimetry_config(ctrl->link); + + misc_val = msm_dp_read_link(ctrl, REG_DP_MISC1_MISC0); + + /* clear bpp bits */ + misc_val &= ~(0x07 << DP_MISC0_TEST_BITS_DEPTH_SHIFT); + misc_val |= colorimetry_cfg << DP_MISC0_COLORIMETRY_CFG_SHIFT; + misc_val |= test_bits_depth << DP_MISC0_TEST_BITS_DEPTH_SHIFT; + /* Configure clock to synchronous mode */ + misc_val |= DP_MISC0_SYNCHRONOUS_CLK; + + drm_dbg_dp(ctrl->drm_dev, "misc settings = 0x%x\n", misc_val); + msm_dp_write_link(ctrl, REG_DP_MISC1_MISC0, misc_val); +} + +static void msm_dp_ctrl_configure_source_params(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel) +{ + msm_dp_ctrl_config_ctrl_streams(ctrl, panel); + + msm_dp_ctrl_config_misc1_misc0(ctrl, panel); + + msm_dp_panel_timing_cfg(panel, ctrl->msm_dp_ctrl.wide_bus_en); +} + +/* + * The structure and few functions present below are IP/Hardware + * specific implementation. Most of the implementation will not + * have coding comments + */ +struct tu_algo_data { + s64 lclk_fp; + s64 pclk_fp; + s64 lwidth; + s64 lwidth_fp; + s64 hbp_relative_to_pclk; + s64 hbp_relative_to_pclk_fp; + int nlanes; + int bpp; + int pixelEnc; + int dsc_en; + int async_en; + int bpc; + + uint delay_start_link_extra_pixclk; + int extra_buffer_margin; + s64 ratio_fp; + s64 original_ratio_fp; + + s64 err_fp; + s64 n_err_fp; + s64 n_n_err_fp; + int tu_size; + int tu_size_desired; + int tu_size_minus1; + + int valid_boundary_link; + s64 resulting_valid_fp; + s64 total_valid_fp; + s64 effective_valid_fp; + s64 effective_valid_recorded_fp; + int n_tus; + int n_tus_per_lane; + int paired_tus; + int remainder_tus; + int remainder_tus_upper; + int remainder_tus_lower; + int extra_bytes; + int filler_size; + int delay_start_link; + + int extra_pclk_cycles; + int extra_pclk_cycles_in_link_clk; + s64 ratio_by_tu_fp; + s64 average_valid2_fp; + int new_valid_boundary_link; + int remainder_symbols_exist; + int n_symbols; + s64 n_remainder_symbols_per_lane_fp; + s64 last_partial_tu_fp; + s64 TU_ratio_err_fp; + + int n_tus_incl_last_incomplete_tu; + int extra_pclk_cycles_tmp; + int extra_pclk_cycles_in_link_clk_tmp; + int extra_required_bytes_new_tmp; + int filler_size_tmp; + int lower_filler_size_tmp; + int delay_start_link_tmp; + + bool boundary_moderation_en; + int boundary_mod_lower_err; + int upper_boundary_count; + int lower_boundary_count; + int i_upper_boundary_count; + int i_lower_boundary_count; + int valid_lower_boundary_link; + int even_distribution_BF; + int even_distribution_legacy; + int even_distribution; + int min_hblank_violated; + s64 delay_start_time_fp; + s64 hbp_time_fp; + s64 hactive_time_fp; + s64 diff_abs_fp; + + s64 ratio; +}; + +static int _tu_param_compare(s64 a, s64 b) +{ + u32 a_sign; + u32 b_sign; + s64 a_temp, b_temp, minus_1; + + if (a == b) + return 0; + + minus_1 = drm_fixp_from_fraction(-1, 1); + + a_sign = (a >> 32) & 0x80000000 ? 1 : 0; + + b_sign = (b >> 32) & 0x80000000 ? 1 : 0; + + if (a_sign > b_sign) + return 2; + else if (b_sign > a_sign) + return 1; + + if (!a_sign && !b_sign) { /* positive */ + if (a > b) + return 1; + else + return 2; + } else { /* negative */ + a_temp = drm_fixp_mul(a, minus_1); + b_temp = drm_fixp_mul(b, minus_1); + + if (a_temp > b_temp) + return 2; + else + return 1; + } +} + +static void msm_dp_panel_update_tu_timings(struct msm_dp_tu_calc_input *in, + struct tu_algo_data *tu) +{ + int nlanes = in->nlanes; + int dsc_num_slices = in->num_of_dsc_slices; + int dsc_num_bytes = 0; + int numerator; + s64 pclk_dsc_fp; + s64 dwidth_dsc_fp; + s64 hbp_dsc_fp; + + int tot_num_eoc_symbols = 0; + int tot_num_hor_bytes = 0; + int tot_num_dummy_bytes = 0; + int dwidth_dsc_bytes = 0; + int eoc_bytes = 0; + + s64 temp1_fp, temp2_fp, temp3_fp; + + tu->lclk_fp = drm_fixp_from_fraction(in->lclk, 1); + tu->pclk_fp = drm_fixp_from_fraction(in->pclk_khz, 1000); + tu->lwidth = in->hactive; + tu->hbp_relative_to_pclk = in->hporch; + tu->nlanes = in->nlanes; + tu->bpp = in->bpp; + tu->pixelEnc = in->pixel_enc; + tu->dsc_en = in->dsc_en; + tu->async_en = in->async_en; + tu->lwidth_fp = drm_fixp_from_fraction(in->hactive, 1); + tu->hbp_relative_to_pclk_fp = drm_fixp_from_fraction(in->hporch, 1); + + if (tu->pixelEnc == 420) { + temp1_fp = drm_fixp_from_fraction(2, 1); + tu->pclk_fp = drm_fixp_div(tu->pclk_fp, temp1_fp); + tu->lwidth_fp = drm_fixp_div(tu->lwidth_fp, temp1_fp); + tu->hbp_relative_to_pclk_fp = + drm_fixp_div(tu->hbp_relative_to_pclk_fp, 2); + } + + if (tu->pixelEnc == 422) { + switch (tu->bpp) { + case 24: + tu->bpp = 16; + tu->bpc = 8; + break; + case 30: + tu->bpp = 20; + tu->bpc = 10; + break; + default: + tu->bpp = 16; + tu->bpc = 8; + break; + } + } else { + tu->bpc = tu->bpp/3; + } + + if (!in->dsc_en) + goto fec_check; + + temp1_fp = drm_fixp_from_fraction(in->compress_ratio, 100); + temp2_fp = drm_fixp_from_fraction(in->bpp, 1); + temp3_fp = drm_fixp_div(temp2_fp, temp1_fp); + temp2_fp = drm_fixp_mul(tu->lwidth_fp, temp3_fp); + + temp1_fp = drm_fixp_from_fraction(8, 1); + temp3_fp = drm_fixp_div(temp2_fp, temp1_fp); + + numerator = drm_fixp2int(temp3_fp); + + dsc_num_bytes = numerator / dsc_num_slices; + eoc_bytes = dsc_num_bytes % nlanes; + tot_num_eoc_symbols = nlanes * dsc_num_slices; + tot_num_hor_bytes = dsc_num_bytes * dsc_num_slices; + tot_num_dummy_bytes = (nlanes - eoc_bytes) * dsc_num_slices; + + if (dsc_num_bytes == 0) + pr_info("incorrect no of bytes per slice=%d\n", dsc_num_bytes); + + dwidth_dsc_bytes = (tot_num_hor_bytes + + tot_num_eoc_symbols + + (eoc_bytes == 0 ? 0 : tot_num_dummy_bytes)); + + dwidth_dsc_fp = drm_fixp_from_fraction(dwidth_dsc_bytes, 3); + + temp2_fp = drm_fixp_mul(tu->pclk_fp, dwidth_dsc_fp); + temp1_fp = drm_fixp_div(temp2_fp, tu->lwidth_fp); + pclk_dsc_fp = temp1_fp; + + temp1_fp = drm_fixp_div(pclk_dsc_fp, tu->pclk_fp); + temp2_fp = drm_fixp_mul(tu->hbp_relative_to_pclk_fp, temp1_fp); + hbp_dsc_fp = temp2_fp; + + /* output */ + tu->pclk_fp = pclk_dsc_fp; + tu->lwidth_fp = dwidth_dsc_fp; + tu->hbp_relative_to_pclk_fp = hbp_dsc_fp; + +fec_check: + if (in->fec_en) { + temp1_fp = drm_fixp_from_fraction(976, 1000); /* 0.976 */ + tu->lclk_fp = drm_fixp_mul(tu->lclk_fp, temp1_fp); + } +} + +static void _tu_valid_boundary_calc(struct tu_algo_data *tu) +{ + s64 temp1_fp, temp2_fp, temp, temp1, temp2; + int compare_result_1, compare_result_2, compare_result_3; + + temp1_fp = drm_fixp_from_fraction(tu->tu_size, 1); + temp2_fp = drm_fixp_mul(tu->ratio_fp, temp1_fp); + + tu->new_valid_boundary_link = drm_fixp2int_ceil(temp2_fp); + + temp = (tu->i_upper_boundary_count * + tu->new_valid_boundary_link + + tu->i_lower_boundary_count * + (tu->new_valid_boundary_link-1)); + tu->average_valid2_fp = drm_fixp_from_fraction(temp, + (tu->i_upper_boundary_count + + tu->i_lower_boundary_count)); + + temp1_fp = drm_fixp_from_fraction(tu->bpp, 8); + temp2_fp = tu->lwidth_fp; + temp1_fp = drm_fixp_mul(temp2_fp, temp1_fp); + temp2_fp = drm_fixp_div(temp1_fp, tu->average_valid2_fp); + tu->n_tus = drm_fixp2int(temp2_fp); + if ((temp2_fp & 0xFFFFFFFF) > 0xFFFFF000) + tu->n_tus += 1; + + temp1_fp = drm_fixp_from_fraction(tu->n_tus, 1); + temp2_fp = drm_fixp_mul(temp1_fp, tu->average_valid2_fp); + temp1_fp = drm_fixp_from_fraction(tu->n_symbols, 1); + temp2_fp = temp1_fp - temp2_fp; + temp1_fp = drm_fixp_from_fraction(tu->nlanes, 1); + temp2_fp = drm_fixp_div(temp2_fp, temp1_fp); + tu->n_remainder_symbols_per_lane_fp = temp2_fp; + + temp1_fp = drm_fixp_from_fraction(tu->tu_size, 1); + tu->last_partial_tu_fp = + drm_fixp_div(tu->n_remainder_symbols_per_lane_fp, + temp1_fp); + + if (tu->n_remainder_symbols_per_lane_fp != 0) + tu->remainder_symbols_exist = 1; + else + tu->remainder_symbols_exist = 0; + + temp1_fp = drm_fixp_from_fraction(tu->n_tus, tu->nlanes); + tu->n_tus_per_lane = drm_fixp2int(temp1_fp); + + tu->paired_tus = (int)((tu->n_tus_per_lane) / + (tu->i_upper_boundary_count + + tu->i_lower_boundary_count)); + + tu->remainder_tus = tu->n_tus_per_lane - tu->paired_tus * + (tu->i_upper_boundary_count + + tu->i_lower_boundary_count); + + if ((tu->remainder_tus - tu->i_upper_boundary_count) > 0) { + tu->remainder_tus_upper = tu->i_upper_boundary_count; + tu->remainder_tus_lower = tu->remainder_tus - + tu->i_upper_boundary_count; + } else { + tu->remainder_tus_upper = tu->remainder_tus; + tu->remainder_tus_lower = 0; + } + + temp = tu->paired_tus * (tu->i_upper_boundary_count * + tu->new_valid_boundary_link + + tu->i_lower_boundary_count * + (tu->new_valid_boundary_link - 1)) + + (tu->remainder_tus_upper * + tu->new_valid_boundary_link) + + (tu->remainder_tus_lower * + (tu->new_valid_boundary_link - 1)); + tu->total_valid_fp = drm_fixp_from_fraction(temp, 1); + + if (tu->remainder_symbols_exist) { + temp1_fp = tu->total_valid_fp + + tu->n_remainder_symbols_per_lane_fp; + temp2_fp = drm_fixp_from_fraction(tu->n_tus_per_lane, 1); + temp2_fp = temp2_fp + tu->last_partial_tu_fp; + temp1_fp = drm_fixp_div(temp1_fp, temp2_fp); + } else { + temp2_fp = drm_fixp_from_fraction(tu->n_tus_per_lane, 1); + temp1_fp = drm_fixp_div(tu->total_valid_fp, temp2_fp); + } + tu->effective_valid_fp = temp1_fp; + + temp1_fp = drm_fixp_from_fraction(tu->tu_size, 1); + temp2_fp = drm_fixp_mul(tu->ratio_fp, temp1_fp); + tu->n_n_err_fp = tu->effective_valid_fp - temp2_fp; + + temp1_fp = drm_fixp_from_fraction(tu->tu_size, 1); + temp2_fp = drm_fixp_mul(tu->ratio_fp, temp1_fp); + tu->n_err_fp = tu->average_valid2_fp - temp2_fp; + + tu->even_distribution = tu->n_tus % tu->nlanes == 0 ? 1 : 0; + + temp1_fp = drm_fixp_from_fraction(tu->bpp, 8); + temp2_fp = tu->lwidth_fp; + temp1_fp = drm_fixp_mul(temp2_fp, temp1_fp); + temp2_fp = drm_fixp_div(temp1_fp, tu->average_valid2_fp); + + if (temp2_fp) + tu->n_tus_incl_last_incomplete_tu = drm_fixp2int_ceil(temp2_fp); + else + tu->n_tus_incl_last_incomplete_tu = 0; + + temp1 = 0; + temp1_fp = drm_fixp_from_fraction(tu->tu_size, 1); + temp2_fp = drm_fixp_mul(tu->original_ratio_fp, temp1_fp); + temp1_fp = tu->average_valid2_fp - temp2_fp; + temp2_fp = drm_fixp_from_fraction(tu->n_tus_incl_last_incomplete_tu, 1); + temp1_fp = drm_fixp_mul(temp2_fp, temp1_fp); + + if (temp1_fp) + temp1 = drm_fixp2int_ceil(temp1_fp); + + temp = tu->i_upper_boundary_count * tu->nlanes; + temp1_fp = drm_fixp_from_fraction(tu->tu_size, 1); + temp2_fp = drm_fixp_mul(tu->original_ratio_fp, temp1_fp); + temp1_fp = drm_fixp_from_fraction(tu->new_valid_boundary_link, 1); + temp2_fp = temp1_fp - temp2_fp; + temp1_fp = drm_fixp_from_fraction(temp, 1); + temp2_fp = drm_fixp_mul(temp1_fp, temp2_fp); + + if (temp2_fp) + temp2 = drm_fixp2int_ceil(temp2_fp); + else + temp2 = 0; + tu->extra_required_bytes_new_tmp = (int)(temp1 + temp2); + + temp1_fp = drm_fixp_from_fraction(8, tu->bpp); + temp2_fp = drm_fixp_from_fraction( + tu->extra_required_bytes_new_tmp, 1); + temp1_fp = drm_fixp_mul(temp2_fp, temp1_fp); + + if (temp1_fp) + tu->extra_pclk_cycles_tmp = drm_fixp2int_ceil(temp1_fp); + else + tu->extra_pclk_cycles_tmp = 0; + + temp1_fp = drm_fixp_from_fraction(tu->extra_pclk_cycles_tmp, 1); + temp2_fp = drm_fixp_div(tu->lclk_fp, tu->pclk_fp); + temp1_fp = drm_fixp_mul(temp1_fp, temp2_fp); + + if (temp1_fp) + tu->extra_pclk_cycles_in_link_clk_tmp = + drm_fixp2int_ceil(temp1_fp); + else + tu->extra_pclk_cycles_in_link_clk_tmp = 0; + + tu->filler_size_tmp = tu->tu_size - tu->new_valid_boundary_link; + + tu->lower_filler_size_tmp = tu->filler_size_tmp + 1; + + tu->delay_start_link_tmp = tu->extra_pclk_cycles_in_link_clk_tmp + + tu->lower_filler_size_tmp + + tu->extra_buffer_margin; + + temp1_fp = drm_fixp_from_fraction(tu->delay_start_link_tmp, 1); + tu->delay_start_time_fp = drm_fixp_div(temp1_fp, tu->lclk_fp); + + compare_result_1 = _tu_param_compare(tu->n_n_err_fp, tu->diff_abs_fp); + if (compare_result_1 == 2) + compare_result_1 = 1; + else + compare_result_1 = 0; + + compare_result_2 = _tu_param_compare(tu->n_n_err_fp, tu->err_fp); + if (compare_result_2 == 2) + compare_result_2 = 1; + else + compare_result_2 = 0; + + compare_result_3 = _tu_param_compare(tu->hbp_time_fp, + tu->delay_start_time_fp); + if (compare_result_3 == 2) + compare_result_3 = 0; + else + compare_result_3 = 1; + + if (((tu->even_distribution == 1) || + ((tu->even_distribution_BF == 0) && + (tu->even_distribution_legacy == 0))) && + tu->n_err_fp >= 0 && tu->n_n_err_fp >= 0 && + compare_result_2 && + (compare_result_1 || (tu->min_hblank_violated == 1)) && + (tu->new_valid_boundary_link - 1) > 0 && + compare_result_3 && + (tu->delay_start_link_tmp <= 1023)) { + tu->upper_boundary_count = tu->i_upper_boundary_count; + tu->lower_boundary_count = tu->i_lower_boundary_count; + tu->err_fp = tu->n_n_err_fp; + tu->boundary_moderation_en = true; + tu->tu_size_desired = tu->tu_size; + tu->valid_boundary_link = tu->new_valid_boundary_link; + tu->effective_valid_recorded_fp = tu->effective_valid_fp; + tu->even_distribution_BF = 1; + tu->delay_start_link = tu->delay_start_link_tmp; + } else if (tu->boundary_mod_lower_err == 0) { + compare_result_1 = _tu_param_compare(tu->n_n_err_fp, + tu->diff_abs_fp); + if (compare_result_1 == 2) + tu->boundary_mod_lower_err = 1; + } +} + +static void _dp_ctrl_calc_tu(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_tu_calc_input *in, + struct msm_dp_vc_tu_mapping_table *tu_table) +{ + struct tu_algo_data *tu; + int compare_result_1, compare_result_2; + u64 temp = 0; + s64 temp_fp = 0, temp1_fp = 0, temp2_fp = 0; + + s64 LCLK_FAST_SKEW_fp = drm_fixp_from_fraction(6, 10000); /* 0.0006 */ + s64 const_p49_fp = drm_fixp_from_fraction(49, 100); /* 0.49 */ + s64 const_p56_fp = drm_fixp_from_fraction(56, 100); /* 0.56 */ + s64 RATIO_SCALE_fp = drm_fixp_from_fraction(1001, 1000); + + u8 DP_BRUTE_FORCE = 1; + s64 BRUTE_FORCE_THRESHOLD_fp = drm_fixp_from_fraction(1, 10); /* 0.1 */ + uint EXTRA_PIXCLK_CYCLE_DELAY = 4; + uint HBLANK_MARGIN = 4; + + tu = kzalloc_obj(*tu); + if (!tu) + return; + + msm_dp_panel_update_tu_timings(in, tu); + + tu->err_fp = drm_fixp_from_fraction(1000, 1); /* 1000 */ + + temp1_fp = drm_fixp_from_fraction(4, 1); + temp2_fp = drm_fixp_mul(temp1_fp, tu->lclk_fp); + temp_fp = drm_fixp_div(temp2_fp, tu->pclk_fp); + tu->extra_buffer_margin = drm_fixp2int_ceil(temp_fp); + + temp1_fp = drm_fixp_from_fraction(tu->bpp, 8); + temp2_fp = drm_fixp_mul(tu->pclk_fp, temp1_fp); + temp1_fp = drm_fixp_from_fraction(tu->nlanes, 1); + temp2_fp = drm_fixp_div(temp2_fp, temp1_fp); + tu->ratio_fp = drm_fixp_div(temp2_fp, tu->lclk_fp); + + tu->original_ratio_fp = tu->ratio_fp; + tu->boundary_moderation_en = false; + tu->upper_boundary_count = 0; + tu->lower_boundary_count = 0; + tu->i_upper_boundary_count = 0; + tu->i_lower_boundary_count = 0; + tu->valid_lower_boundary_link = 0; + tu->even_distribution_BF = 0; + tu->even_distribution_legacy = 0; + tu->even_distribution = 0; + tu->delay_start_time_fp = 0; + + tu->err_fp = drm_fixp_from_fraction(1000, 1); + tu->n_err_fp = 0; + tu->n_n_err_fp = 0; + + tu->ratio = drm_fixp2int(tu->ratio_fp); + temp1_fp = drm_fixp_from_fraction(tu->nlanes, 1); + div64_u64_rem(tu->lwidth_fp, temp1_fp, &temp2_fp); + if (temp2_fp != 0 && + !tu->ratio && tu->dsc_en == 0) { + tu->ratio_fp = drm_fixp_mul(tu->ratio_fp, RATIO_SCALE_fp); + tu->ratio = drm_fixp2int(tu->ratio_fp); + if (tu->ratio) + tu->ratio_fp = drm_fixp_from_fraction(1, 1); + } + + if (tu->ratio > 1) + tu->ratio = 1; + + if (tu->ratio == 1) + goto tu_size_calc; + + compare_result_1 = _tu_param_compare(tu->ratio_fp, const_p49_fp); + if (!compare_result_1 || compare_result_1 == 1) + compare_result_1 = 1; + else + compare_result_1 = 0; + + compare_result_2 = _tu_param_compare(tu->ratio_fp, const_p56_fp); + if (!compare_result_2 || compare_result_2 == 2) + compare_result_2 = 1; + else + compare_result_2 = 0; + + if (tu->dsc_en && compare_result_1 && compare_result_2) { + HBLANK_MARGIN += 4; + drm_dbg_dp(ctrl->drm_dev, + "increase HBLANK_MARGIN to %d\n", HBLANK_MARGIN); + } + +tu_size_calc: + for (tu->tu_size = 32; tu->tu_size <= 64; tu->tu_size++) { + temp1_fp = drm_fixp_from_fraction(tu->tu_size, 1); + temp2_fp = drm_fixp_mul(tu->ratio_fp, temp1_fp); + temp = drm_fixp2int_ceil(temp2_fp); + temp1_fp = drm_fixp_from_fraction(temp, 1); + tu->n_err_fp = temp1_fp - temp2_fp; + + if (tu->n_err_fp < tu->err_fp) { + tu->err_fp = tu->n_err_fp; + tu->tu_size_desired = tu->tu_size; + } + } + + tu->tu_size_minus1 = tu->tu_size_desired - 1; + + temp1_fp = drm_fixp_from_fraction(tu->tu_size_desired, 1); + temp2_fp = drm_fixp_mul(tu->ratio_fp, temp1_fp); + tu->valid_boundary_link = drm_fixp2int_ceil(temp2_fp); + + temp1_fp = drm_fixp_from_fraction(tu->bpp, 8); + temp2_fp = tu->lwidth_fp; + temp2_fp = drm_fixp_mul(temp2_fp, temp1_fp); + + temp1_fp = drm_fixp_from_fraction(tu->valid_boundary_link, 1); + temp2_fp = drm_fixp_div(temp2_fp, temp1_fp); + tu->n_tus = drm_fixp2int(temp2_fp); + if ((temp2_fp & 0xFFFFFFFF) > 0xFFFFF000) + tu->n_tus += 1; + + tu->even_distribution_legacy = tu->n_tus % tu->nlanes == 0 ? 1 : 0; + + drm_dbg_dp(ctrl->drm_dev, + "n_sym = %d, num_of_tus = %d\n", + tu->valid_boundary_link, tu->n_tus); + + temp1_fp = drm_fixp_from_fraction(tu->tu_size_desired, 1); + temp2_fp = drm_fixp_mul(tu->original_ratio_fp, temp1_fp); + temp1_fp = drm_fixp_from_fraction(tu->valid_boundary_link, 1); + temp2_fp = temp1_fp - temp2_fp; + temp1_fp = drm_fixp_from_fraction(tu->n_tus + 1, 1); + temp2_fp = drm_fixp_mul(temp1_fp, temp2_fp); + + temp = drm_fixp2int(temp2_fp); + if (temp && temp2_fp) + tu->extra_bytes = drm_fixp2int_ceil(temp2_fp); + else + tu->extra_bytes = 0; + + temp1_fp = drm_fixp_from_fraction(tu->extra_bytes, 1); + temp2_fp = drm_fixp_from_fraction(8, tu->bpp); + temp1_fp = drm_fixp_mul(temp1_fp, temp2_fp); + + if (temp && temp1_fp) + tu->extra_pclk_cycles = drm_fixp2int_ceil(temp1_fp); + else + tu->extra_pclk_cycles = drm_fixp2int(temp1_fp); + + temp1_fp = drm_fixp_div(tu->lclk_fp, tu->pclk_fp); + temp2_fp = drm_fixp_from_fraction(tu->extra_pclk_cycles, 1); + temp1_fp = drm_fixp_mul(temp2_fp, temp1_fp); + + if (temp1_fp) + tu->extra_pclk_cycles_in_link_clk = drm_fixp2int_ceil(temp1_fp); + else + tu->extra_pclk_cycles_in_link_clk = drm_fixp2int(temp1_fp); + + tu->filler_size = tu->tu_size_desired - tu->valid_boundary_link; + + temp1_fp = drm_fixp_from_fraction(tu->tu_size_desired, 1); + tu->ratio_by_tu_fp = drm_fixp_mul(tu->ratio_fp, temp1_fp); + + tu->delay_start_link = tu->extra_pclk_cycles_in_link_clk + + tu->filler_size + tu->extra_buffer_margin; + + tu->resulting_valid_fp = + drm_fixp_from_fraction(tu->valid_boundary_link, 1); + + temp1_fp = drm_fixp_from_fraction(tu->tu_size_desired, 1); + temp2_fp = drm_fixp_div(tu->resulting_valid_fp, temp1_fp); + tu->TU_ratio_err_fp = temp2_fp - tu->original_ratio_fp; + + temp1_fp = drm_fixp_from_fraction(HBLANK_MARGIN, 1); + temp1_fp = tu->hbp_relative_to_pclk_fp - temp1_fp; + tu->hbp_time_fp = drm_fixp_div(temp1_fp, tu->pclk_fp); + + temp1_fp = drm_fixp_from_fraction(tu->delay_start_link, 1); + tu->delay_start_time_fp = drm_fixp_div(temp1_fp, tu->lclk_fp); + + compare_result_1 = _tu_param_compare(tu->hbp_time_fp, + tu->delay_start_time_fp); + if (compare_result_1 == 2) /* if (hbp_time_fp < delay_start_time_fp) */ + tu->min_hblank_violated = 1; + + tu->hactive_time_fp = drm_fixp_div(tu->lwidth_fp, tu->pclk_fp); + + compare_result_2 = _tu_param_compare(tu->hactive_time_fp, + tu->delay_start_time_fp); + if (compare_result_2 == 2) + tu->min_hblank_violated = 1; + + tu->delay_start_time_fp = 0; + + /* brute force */ + + tu->delay_start_link_extra_pixclk = EXTRA_PIXCLK_CYCLE_DELAY; + tu->diff_abs_fp = tu->resulting_valid_fp - tu->ratio_by_tu_fp; + + temp = drm_fixp2int(tu->diff_abs_fp); + if (!temp && tu->diff_abs_fp <= 0xffff) + tu->diff_abs_fp = 0; + + /* if(diff_abs < 0) diff_abs *= -1 */ + if (tu->diff_abs_fp < 0) + tu->diff_abs_fp = drm_fixp_mul(tu->diff_abs_fp, -1); + + tu->boundary_mod_lower_err = 0; + if ((tu->diff_abs_fp != 0 && + ((tu->diff_abs_fp > BRUTE_FORCE_THRESHOLD_fp) || + (tu->even_distribution_legacy == 0) || + (DP_BRUTE_FORCE == 1))) || + (tu->min_hblank_violated == 1)) { + do { + tu->err_fp = drm_fixp_from_fraction(1000, 1); + + temp1_fp = drm_fixp_div(tu->lclk_fp, tu->pclk_fp); + temp2_fp = drm_fixp_from_fraction( + tu->delay_start_link_extra_pixclk, 1); + temp1_fp = drm_fixp_mul(temp2_fp, temp1_fp); + + if (temp1_fp) + tu->extra_buffer_margin = + drm_fixp2int_ceil(temp1_fp); + else + tu->extra_buffer_margin = 0; + + temp1_fp = drm_fixp_from_fraction(tu->bpp, 8); + temp1_fp = drm_fixp_mul(tu->lwidth_fp, temp1_fp); + + if (temp1_fp) + tu->n_symbols = drm_fixp2int_ceil(temp1_fp); + else + tu->n_symbols = 0; + + for (tu->tu_size = 32; tu->tu_size <= 64; tu->tu_size++) { + for (tu->i_upper_boundary_count = 1; + tu->i_upper_boundary_count <= 15; + tu->i_upper_boundary_count++) { + for (tu->i_lower_boundary_count = 1; + tu->i_lower_boundary_count <= 15; + tu->i_lower_boundary_count++) { + _tu_valid_boundary_calc(tu); + } + } + } + tu->delay_start_link_extra_pixclk--; + } while (tu->boundary_moderation_en != true && + tu->boundary_mod_lower_err == 1 && + tu->delay_start_link_extra_pixclk != 0); + + if (tu->boundary_moderation_en == true) { + temp1_fp = drm_fixp_from_fraction( + (tu->upper_boundary_count * + tu->valid_boundary_link + + tu->lower_boundary_count * + (tu->valid_boundary_link - 1)), 1); + temp2_fp = drm_fixp_from_fraction( + (tu->upper_boundary_count + + tu->lower_boundary_count), 1); + tu->resulting_valid_fp = + drm_fixp_div(temp1_fp, temp2_fp); + + temp1_fp = drm_fixp_from_fraction( + tu->tu_size_desired, 1); + tu->ratio_by_tu_fp = + drm_fixp_mul(tu->original_ratio_fp, temp1_fp); + + tu->valid_lower_boundary_link = + tu->valid_boundary_link - 1; + + temp1_fp = drm_fixp_from_fraction(tu->bpp, 8); + temp1_fp = drm_fixp_mul(tu->lwidth_fp, temp1_fp); + temp2_fp = drm_fixp_div(temp1_fp, + tu->resulting_valid_fp); + tu->n_tus = drm_fixp2int(temp2_fp); + + tu->tu_size_minus1 = tu->tu_size_desired - 1; + tu->even_distribution_BF = 1; + + temp1_fp = + drm_fixp_from_fraction(tu->tu_size_desired, 1); + temp2_fp = + drm_fixp_div(tu->resulting_valid_fp, temp1_fp); + tu->TU_ratio_err_fp = temp2_fp - tu->original_ratio_fp; + } + } + + temp2_fp = drm_fixp_mul(LCLK_FAST_SKEW_fp, tu->lwidth_fp); + + if (temp2_fp) + temp = drm_fixp2int_ceil(temp2_fp); + else + temp = 0; + + temp1_fp = drm_fixp_from_fraction(tu->nlanes, 1); + temp2_fp = drm_fixp_mul(tu->original_ratio_fp, temp1_fp); + temp1_fp = drm_fixp_from_fraction(tu->bpp, 8); + temp2_fp = drm_fixp_div(temp1_fp, temp2_fp); + temp1_fp = drm_fixp_from_fraction(temp, 1); + temp2_fp = drm_fixp_mul(temp1_fp, temp2_fp); + temp = drm_fixp2int(temp2_fp); + + if (tu->async_en) + tu->delay_start_link += (int)temp; + + temp1_fp = drm_fixp_from_fraction(tu->delay_start_link, 1); + tu->delay_start_time_fp = drm_fixp_div(temp1_fp, tu->lclk_fp); + + /* OUTPUTS */ + tu_table->valid_boundary_link = tu->valid_boundary_link; + tu_table->delay_start_link = tu->delay_start_link; + tu_table->boundary_moderation_en = tu->boundary_moderation_en; + tu_table->valid_lower_boundary_link = tu->valid_lower_boundary_link; + tu_table->upper_boundary_count = tu->upper_boundary_count; + tu_table->lower_boundary_count = tu->lower_boundary_count; + tu_table->tu_size_minus1 = tu->tu_size_minus1; + + drm_dbg_dp(ctrl->drm_dev, "TU: valid_boundary_link: %d\n", + tu_table->valid_boundary_link); + drm_dbg_dp(ctrl->drm_dev, "TU: delay_start_link: %d\n", + tu_table->delay_start_link); + drm_dbg_dp(ctrl->drm_dev, "TU: boundary_moderation_en: %d\n", + tu_table->boundary_moderation_en); + drm_dbg_dp(ctrl->drm_dev, "TU: valid_lower_boundary_link: %d\n", + tu_table->valid_lower_boundary_link); + drm_dbg_dp(ctrl->drm_dev, "TU: upper_boundary_count: %d\n", + tu_table->upper_boundary_count); + drm_dbg_dp(ctrl->drm_dev, "TU: lower_boundary_count: %d\n", + tu_table->lower_boundary_count); + drm_dbg_dp(ctrl->drm_dev, "TU: tu_size_minus1: %d\n", + tu_table->tu_size_minus1); + + kfree(tu); +} + +static void msm_dp_ctrl_calc_tu_parameters(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel, + struct msm_dp_vc_tu_mapping_table *tu_table) +{ + struct msm_dp_tu_calc_input in; + const struct drm_display_mode *drm_mode; + + drm_mode = &panel->msm_dp_mode.drm_mode; + + in.lclk = ctrl->link->link_params.rate / 1000; + in.pclk_khz = drm_mode->clock; + in.hactive = drm_mode->hdisplay; + in.hporch = drm_mode->htotal - drm_mode->hdisplay; + in.nlanes = ctrl->link->link_params.num_lanes; + in.bpp = panel->msm_dp_mode.bpp; + in.pixel_enc = panel->msm_dp_mode.out_fmt_is_yuv_420 ? 420 : 444; + in.dsc_en = 0; + in.async_en = 0; + in.fec_en = 0; + in.num_of_dsc_slices = 0; + in.compress_ratio = 100; + + _dp_ctrl_calc_tu(ctrl, &in, tu_table); +} + +static void msm_dp_ctrl_setup_tr_unit(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel) +{ + u32 msm_dp_tu = 0x0; + u32 valid_boundary = 0x0; + u32 valid_boundary2 = 0x0; + struct msm_dp_vc_tu_mapping_table tu_calc_table; + + msm_dp_ctrl_calc_tu_parameters(ctrl, panel, &tu_calc_table); + + msm_dp_tu |= tu_calc_table.tu_size_minus1; + valid_boundary |= tu_calc_table.valid_boundary_link; + valid_boundary |= (tu_calc_table.delay_start_link << 16); + + valid_boundary2 |= (tu_calc_table.valid_lower_boundary_link << 1); + valid_boundary2 |= (tu_calc_table.upper_boundary_count << 16); + valid_boundary2 |= (tu_calc_table.lower_boundary_count << 20); + + if (tu_calc_table.boundary_moderation_en) + valid_boundary2 |= BIT(0); + + pr_debug("dp_tu=0x%x, valid_boundary=0x%x, valid_boundary2=0x%x\n", + msm_dp_tu, valid_boundary, valid_boundary2); + + msm_dp_write_link(ctrl, REG_DP_VALID_BOUNDARY, valid_boundary); + msm_dp_write_link(ctrl, REG_DP_TU, msm_dp_tu); + msm_dp_write_link(ctrl, REG_DP_VALID_BOUNDARY_2, valid_boundary2); +} + +static int msm_dp_ctrl_wait4video_ready(struct msm_dp_ctrl_private *ctrl) +{ + int ret = 0; + + if (!wait_for_completion_timeout(&ctrl->video_comp, + WAIT_FOR_VIDEO_READY_TIMEOUT_JIFFIES)) { + DRM_ERROR("wait4video timedout\n"); + ret = -ETIMEDOUT; + } + return ret; +} + +static int msm_dp_ctrl_set_vx_px(struct msm_dp_ctrl_private *ctrl, + u8 v_level, u8 p_level) +{ + union phy_configure_opts *phy_opts = &ctrl->phy_opts; + + /* TODO: Update for all lanes instead of just first one */ + phy_opts->dp.voltage[0] = v_level; + phy_opts->dp.pre[0] = p_level; + phy_opts->dp.set_voltages = 1; + phy_configure(ctrl->phy, phy_opts); + phy_opts->dp.set_voltages = 0; + + return 0; +} + +static int msm_dp_ctrl_update_phy_vx_px(struct msm_dp_ctrl_private *ctrl, + enum drm_dp_phy dp_phy) +{ + struct msm_dp_link *link = ctrl->link; + int lane, lane_cnt, reg; + int ret = 0; + u8 buf[4]; + u32 max_level_reached = 0; + u32 voltage_swing_level = link->phy_params.v_level; + u32 pre_emphasis_level = link->phy_params.p_level; + + drm_dbg_dp(ctrl->drm_dev, + "voltage level: %d emphasis level: %d\n", + voltage_swing_level, pre_emphasis_level); + ret = msm_dp_ctrl_set_vx_px(ctrl, + voltage_swing_level, pre_emphasis_level); + + if (ret) + return ret; + + if (voltage_swing_level >= DP_TRAIN_LEVEL_MAX) { + drm_dbg_dp(ctrl->drm_dev, + "max. voltage swing level reached %d\n", + voltage_swing_level); + max_level_reached |= DP_TRAIN_MAX_SWING_REACHED; + } + + if (pre_emphasis_level >= DP_TRAIN_LEVEL_MAX) { + drm_dbg_dp(ctrl->drm_dev, + "max. pre-emphasis level reached %d\n", + pre_emphasis_level); + max_level_reached |= DP_TRAIN_MAX_PRE_EMPHASIS_REACHED; + } + + pre_emphasis_level <<= DP_TRAIN_PRE_EMPHASIS_SHIFT; + + lane_cnt = ctrl->link->link_params.num_lanes; + for (lane = 0; lane < lane_cnt; lane++) + buf[lane] = voltage_swing_level | pre_emphasis_level + | max_level_reached; + + drm_dbg_dp(ctrl->drm_dev, "sink: p|v=0x%x\n", + voltage_swing_level | pre_emphasis_level); + + if (dp_phy == DP_PHY_DPRX) + reg = DP_TRAINING_LANE0_SET; + else + reg = DP_TRAINING_LANE0_SET_PHY_REPEATER(dp_phy); + + ret = drm_dp_dpcd_write(ctrl->aux, reg, buf, lane_cnt); + if (ret == lane_cnt) + ret = 0; + + return ret; +} + +static bool msm_dp_ctrl_train_pattern_set(struct msm_dp_ctrl_private *ctrl, + u8 pattern, enum drm_dp_phy dp_phy) +{ + u8 buf; + int reg; + int ret = 0; + + drm_dbg_dp(ctrl->drm_dev, "sink: pattern=%x\n", pattern); + + buf = pattern; + + if (pattern && pattern != DP_TRAINING_PATTERN_4) + buf |= DP_LINK_SCRAMBLING_DISABLE; + + if (dp_phy == DP_PHY_DPRX) + reg = DP_TRAINING_PATTERN_SET; + else + reg = DP_TRAINING_PATTERN_SET_PHY_REPEATER(dp_phy); + + ret = drm_dp_dpcd_writeb(ctrl->aux, reg, buf); + return ret == 1; +} + +static int msm_dp_ctrl_set_pattern_state_bit(struct msm_dp_ctrl_private *ctrl, + u32 state_bit) +{ + int bit, ret; + u32 data; + + bit = BIT(state_bit - 1); + drm_dbg_dp(ctrl->drm_dev, "hw: bit=%d train=%d\n", bit, state_bit); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, bit); + + bit = BIT(state_bit - 1) << DP_MAINLINK_READY_LINK_TRAINING_SHIFT; + + /* Poll for mainlink ready status */ + ret = readx_poll_timeout(readl, ctrl->link_base + REG_DP_MAINLINK_READY, + data, data & bit, + POLLING_SLEEP_US, POLLING_TIMEOUT_US); + if (ret < 0) { + DRM_ERROR("set state_bit for link_train=%d failed\n", state_bit); + return ret; + } + + return 0; +} + +static int msm_dp_ctrl_link_train_1(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel, + int *training_step, enum drm_dp_phy dp_phy) +{ + int delay_us; + int tries, old_v_level, ret = 0; + u8 link_status[DP_LINK_STATUS_SIZE]; + int const maximum_retries = 4; + + delay_us = drm_dp_read_clock_recovery_delay(ctrl->aux, + panel->dpcd, dp_phy, false); + + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, 0); + + *training_step = DP_TRAINING_1; + + ret = msm_dp_ctrl_set_pattern_state_bit(ctrl, 1); + if (ret) + return ret; + msm_dp_ctrl_train_pattern_set(ctrl, DP_TRAINING_PATTERN_1 | + DP_LINK_SCRAMBLING_DISABLE, dp_phy); + + msm_dp_link_reset_phy_params_vx_px(ctrl->link); + ret = msm_dp_ctrl_update_phy_vx_px(ctrl, dp_phy); + if (ret) + return ret; + + tries = 0; + old_v_level = ctrl->link->phy_params.v_level; + for (tries = 0; tries < maximum_retries; tries++) { + fsleep(delay_us); + + ret = drm_dp_dpcd_read_phy_link_status(ctrl->aux, dp_phy, link_status); + if (ret) + return ret; + + if (drm_dp_clock_recovery_ok(link_status, + ctrl->link->link_params.num_lanes)) { + return 0; + } + + if (ctrl->link->phy_params.v_level >= + DP_TRAIN_LEVEL_MAX) { + DRM_ERROR_RATELIMITED("max v_level reached\n"); + return -EAGAIN; + } + + if (old_v_level != ctrl->link->phy_params.v_level) { + tries = 0; + old_v_level = ctrl->link->phy_params.v_level; + } + + msm_dp_link_adjust_levels(ctrl->link, link_status); + ret = msm_dp_ctrl_update_phy_vx_px(ctrl, dp_phy); + if (ret) + return ret; + } + + DRM_ERROR("max tries reached\n"); + return -ETIMEDOUT; +} + +static int msm_dp_ctrl_link_rate_down_shift(struct msm_dp_ctrl_private *ctrl) +{ + int ret = 0; + struct msm_dp_link_info *link_params = &ctrl->link->link_params; + + if (link_params->rate_set) { + --link_params->rate_set; + link_params->rate = link_params->supported_rates[link_params->rate_set]; + } else { + switch (link_params->rate) { + case 810000: + link_params->rate = 540000; + break; + case 540000: + link_params->rate = 270000; + break; + case 270000: + link_params->rate = 162000; + break; + case 162000: + default: + ret = -EINVAL; + break; + } + } + + if (!ret) { + drm_dbg_dp(ctrl->drm_dev, "new rate=0x%x\n", + link_params->rate); + } + + return ret; +} + +static int msm_dp_ctrl_link_lane_down_shift(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel) +{ + + if (ctrl->link->link_params.num_lanes == 1) + return -1; + + ctrl->link->link_params.num_lanes /= 2; + ctrl->link->link_params.rate = panel->link_info.rate; + + ctrl->link->phy_params.p_level = 0; + ctrl->link->phy_params.v_level = 0; + + return 0; +} + +static void msm_dp_ctrl_clear_training_pattern(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel, + enum drm_dp_phy dp_phy) +{ + int delay_us; + + msm_dp_ctrl_train_pattern_set(ctrl, DP_TRAINING_PATTERN_DISABLE, dp_phy); + + delay_us = drm_dp_read_channel_eq_delay(ctrl->aux, + panel->dpcd, dp_phy, false); + fsleep(delay_us); +} + +static int msm_dp_ctrl_link_train_2(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel, + int *training_step, enum drm_dp_phy dp_phy) +{ + int delay_us; + int tries = 0, ret = 0; + u8 pattern; + u32 state_ctrl_bit; + int const maximum_retries = 5; + u8 link_status[DP_LINK_STATUS_SIZE]; + + delay_us = drm_dp_read_channel_eq_delay(ctrl->aux, + panel->dpcd, dp_phy, false); + + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, 0); + + *training_step = DP_TRAINING_2; + + if (drm_dp_tps4_supported(panel->dpcd)) { + pattern = DP_TRAINING_PATTERN_4; + state_ctrl_bit = 4; + } else if (drm_dp_tps3_supported(panel->dpcd)) { + pattern = DP_TRAINING_PATTERN_3; + state_ctrl_bit = 3; + } else { + pattern = DP_TRAINING_PATTERN_2; + state_ctrl_bit = 2; + } + + ret = msm_dp_ctrl_set_pattern_state_bit(ctrl, state_ctrl_bit); + if (ret) + return ret; + + msm_dp_ctrl_train_pattern_set(ctrl, pattern, dp_phy); + + for (tries = 0; tries <= maximum_retries; tries++) { + fsleep(delay_us); + + ret = drm_dp_dpcd_read_phy_link_status(ctrl->aux, dp_phy, link_status); + if (ret) + return ret; + + if (drm_dp_channel_eq_ok(link_status, + ctrl->link->link_params.num_lanes)) { + return 0; + } + + msm_dp_link_adjust_levels(ctrl->link, link_status); + ret = msm_dp_ctrl_update_phy_vx_px(ctrl, dp_phy); + if (ret) + return ret; + + } + + return -ETIMEDOUT; +} + +static int msm_dp_ctrl_link_train_1_2(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel, + int *training_step, enum drm_dp_phy dp_phy) +{ + int ret; + + ret = msm_dp_ctrl_link_train_1(ctrl, panel, training_step, dp_phy); + if (ret) { + DRM_ERROR("link training #1 on phy %d failed. ret=%d\n", dp_phy, ret); + return ret; + } + drm_dbg_dp(ctrl->drm_dev, "link training #1 on phy %d successful\n", dp_phy); + + ret = msm_dp_ctrl_link_train_2(ctrl, panel, training_step, dp_phy); + if (ret) { + DRM_ERROR("link training #2 on phy %d failed. ret=%d\n", dp_phy, ret); + return ret; + } + drm_dbg_dp(ctrl->drm_dev, "link training #2 on phy %d successful\n", dp_phy); + + return 0; +} + +static int msm_dp_ctrl_link_train(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel, + int *training_step) +{ + int i; + int ret = 0; + const u8 *dpcd = panel->dpcd; + u8 encoding[] = { 0, DP_SET_ANSI_8B10B }; + u8 assr; + struct msm_dp_link_info link_info = {0}; + + msm_dp_ctrl_config_ctrl_link(ctrl, panel); + msm_dp_ctrl_config_ctrl_streams(ctrl, panel); + + link_info.num_lanes = ctrl->link->link_params.num_lanes; + link_info.rate = ctrl->link->link_params.rate; + link_info.capabilities = DP_LINK_CAP_ENHANCED_FRAMING; + + msm_dp_aux_link_configure(ctrl->aux, &link_info); + + if (drm_dp_max_downspread(dpcd)) + encoding[0] |= DP_SPREAD_AMP_0_5; + + /* config DOWNSPREAD_CTRL and MAIN_LINK_CHANNEL_CODING_SET */ + drm_dp_dpcd_write(ctrl->aux, DP_DOWNSPREAD_CTRL, encoding, 2); + + if (drm_dp_alternate_scrambler_reset_cap(dpcd)) { + assr = DP_ALTERNATE_SCRAMBLER_RESET_ENABLE; + drm_dp_dpcd_write(ctrl->aux, DP_EDP_CONFIGURATION_SET, + &assr, 1); + } + + for (i = ctrl->link->lttpr_count - 1; i >= 0; i--) { + enum drm_dp_phy dp_phy = DP_PHY_LTTPR(i); + + ret = msm_dp_ctrl_link_train_1_2(ctrl, panel, training_step, dp_phy); + msm_dp_ctrl_clear_training_pattern(ctrl, panel, dp_phy); + + if (ret) + break; + } + + if (ret) { + DRM_ERROR("link training of LTTPR(s) failed. ret=%d\n", ret); + goto end; + } + + ret = msm_dp_ctrl_link_train_1_2(ctrl, panel, training_step, DP_PHY_DPRX); + if (ret) { + DRM_ERROR("link training on sink failed. ret=%d\n", ret); + goto end; + } + +end: + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, 0); + + return ret; +} + +static int msm_dp_ctrl_setup_main_link(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel, + int *training_step) +{ + int ret = 0; + + msm_dp_ctrl_mainlink_enable(ctrl); + + if (ctrl->link->sink_request & DP_TEST_LINK_PHY_TEST_PATTERN) + return ret; + + /* + * As part of previous calls, DP controller state might have + * transitioned to PUSH_IDLE. In order to start transmitting + * a link training pattern, we have to first do soft reset. + */ + + ret = msm_dp_ctrl_link_train(ctrl, panel, training_step); + + return ret; +} + +int msm_dp_ctrl_core_clk_enable(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl; + int ret = 0; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + + if (ctrl->core_clks_on) { + drm_dbg_dp(ctrl->drm_dev, "core clks already enabled\n"); + return 0; + } + + ret = clk_bulk_prepare_enable(ctrl->num_core_clks, ctrl->core_clks); + if (ret) + return ret; + + ctrl->core_clks_on = true; + + drm_dbg_dp(ctrl->drm_dev, "enable core clocks \n"); + drm_dbg_dp(ctrl->drm_dev, "stream_clks:%s link_clks:%s core_clks:%s\n", + str_on_off(ctrl->stream_clks_on), + str_on_off(ctrl->link_clks_on), + str_on_off(ctrl->core_clks_on)); + + return 0; +} + +void msm_dp_ctrl_core_clk_disable(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + + clk_bulk_disable_unprepare(ctrl->num_core_clks, ctrl->core_clks); + + ctrl->core_clks_on = false; + + drm_dbg_dp(ctrl->drm_dev, "disable core clocks \n"); + drm_dbg_dp(ctrl->drm_dev, "stream_clks:%s link_clks:%s core_clks:%s\n", + str_on_off(ctrl->stream_clks_on), + str_on_off(ctrl->link_clks_on), + str_on_off(ctrl->core_clks_on)); +} + +static int msm_dp_ctrl_link_clk_enable(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl; + int ret = 0; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + + if (ctrl->link_clks_on) { + drm_dbg_dp(ctrl->drm_dev, "links clks already enabled\n"); + return 0; + } + + if (!ctrl->core_clks_on) { + drm_dbg_dp(ctrl->drm_dev, "Enable core clks before link clks\n"); + + msm_dp_ctrl_core_clk_enable(msm_dp_ctrl); + } + + ret = clk_bulk_prepare_enable(ctrl->num_link_clks, ctrl->link_clks); + if (ret) + return ret; + + ctrl->link_clks_on = true; + + drm_dbg_dp(ctrl->drm_dev, "enable link clocks\n"); + drm_dbg_dp(ctrl->drm_dev, "stream_clks:%s link_clks:%s core_clks:%s\n", + str_on_off(ctrl->stream_clks_on), + str_on_off(ctrl->link_clks_on), + str_on_off(ctrl->core_clks_on)); + + return 0; +} + +static void msm_dp_ctrl_link_clk_disable(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + + clk_bulk_disable_unprepare(ctrl->num_link_clks, ctrl->link_clks); + + ctrl->link_clks_on = false; + + drm_dbg_dp(ctrl->drm_dev, "disabled link clocks\n"); + drm_dbg_dp(ctrl->drm_dev, "stream_clks:%s link_clks:%s core_clks:%s\n", + str_on_off(ctrl->stream_clks_on), + str_on_off(ctrl->link_clks_on), + str_on_off(ctrl->core_clks_on)); +} + +static int msm_dp_ctrl_enable_mainlink_clocks(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel) +{ + int ret = 0; + struct phy *phy = ctrl->phy; + const u8 *dpcd = panel->dpcd; + + ctrl->phy_opts.dp.lanes = ctrl->link->link_params.num_lanes; + ctrl->phy_opts.dp.link_rate = ctrl->link->link_params.rate / 100; + ctrl->phy_opts.dp.ssc = drm_dp_max_downspread(dpcd); + + phy_configure(phy, &ctrl->phy_opts); + phy_power_on(phy); + + dev_pm_opp_set_rate(ctrl->dev, ctrl->link->link_params.rate * 1000); + ret = msm_dp_ctrl_link_clk_enable(&ctrl->msm_dp_ctrl); + if (ret) + DRM_ERROR("Unable to start link clocks. ret=%d\n", ret); + + drm_dbg_dp(ctrl->drm_dev, "link rate=%d\n", ctrl->link->link_params.rate); + + return ret; +} + +static void msm_dp_ctrl_enable_sdp(struct msm_dp_ctrl_private *ctrl) +{ + /* trigger sdp */ + msm_dp_write_link(ctrl, MMSS_DP_SDP_CFG3, UPDATE_SDP); + msm_dp_write_link(ctrl, MMSS_DP_SDP_CFG3, 0x0); +} + +static void msm_dp_ctrl_psr_enter(struct msm_dp_ctrl_private *ctrl) +{ + u32 cmd; + + cmd = msm_dp_read_link(ctrl, REG_PSR_CMD); + + cmd &= ~(PSR_ENTER | PSR_EXIT); + cmd |= PSR_ENTER; + + msm_dp_ctrl_enable_sdp(ctrl); + msm_dp_write_link(ctrl, REG_PSR_CMD, cmd); +} + +static void msm_dp_ctrl_psr_exit(struct msm_dp_ctrl_private *ctrl) +{ + u32 cmd; + + cmd = msm_dp_read_link(ctrl, REG_PSR_CMD); + + cmd &= ~(PSR_ENTER | PSR_EXIT); + cmd |= PSR_EXIT; + + msm_dp_ctrl_enable_sdp(ctrl); + msm_dp_write_link(ctrl, REG_PSR_CMD, cmd); +} + +void msm_dp_ctrl_config_psr(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel) +{ + struct msm_dp_ctrl_private *ctrl = container_of(msm_dp_ctrl, + struct msm_dp_ctrl_private, msm_dp_ctrl); + u32 cfg; + + if (!panel->psr_cap.version) + return; + + /* enable PSR1 function */ + cfg = msm_dp_read_link(ctrl, REG_PSR_CONFIG); + cfg |= PSR1_SUPPORTED; + msm_dp_write_link(ctrl, REG_PSR_CONFIG, cfg); + + msm_dp_ctrl_config_psr_interrupt(ctrl); + msm_dp_ctrl_enable_sdp(ctrl); + + cfg = DP_PSR_ENABLE; + drm_dp_dpcd_write(ctrl->aux, DP_PSR_EN_CFG, &cfg, 1); +} + +void msm_dp_ctrl_set_psr(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel, bool enter) +{ + struct msm_dp_ctrl_private *ctrl = container_of(msm_dp_ctrl, + struct msm_dp_ctrl_private, msm_dp_ctrl); + + if (!panel->psr_cap.version) + return; + + /* + * When entering PSR, + * 1. Send PSR enter SDP and wait for the PSR_UPDATE_INT + * 2. Turn off video + * 3. Disable the mainlink + * + * When exiting PSR, + * 1. Enable the mainlink + * 2. Send the PSR exit SDP + */ + if (enter) { + reinit_completion(&ctrl->psr_op_comp); + msm_dp_ctrl_psr_enter(ctrl); + + if (!wait_for_completion_timeout(&ctrl->psr_op_comp, + PSR_OPERATION_COMPLETION_TIMEOUT_JIFFIES)) { + DRM_ERROR("PSR_ENTRY timedout\n"); + msm_dp_ctrl_psr_exit(ctrl); + return; + } + + msm_dp_ctrl_push_idle(msm_dp_ctrl); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, 0); + + msm_dp_ctrl_psr_mainlink_disable(ctrl); + } else { + msm_dp_ctrl_psr_mainlink_enable(ctrl); + + msm_dp_ctrl_psr_exit(ctrl); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, DP_STATE_CTRL_SEND_VIDEO); + msm_dp_ctrl_wait4video_ready(ctrl); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, 0); + } +} + +static void msm_dp_ctrl_phy_reset(struct msm_dp_ctrl_private *ctrl) +{ + msm_dp_write_ahb(ctrl, REG_DP_PHY_CTRL, + DP_PHY_CTRL_SW_RESET | DP_PHY_CTRL_SW_RESET_PLL); + usleep_range(1000, 1100); /* h/w recommended delay */ + msm_dp_write_ahb(ctrl, REG_DP_PHY_CTRL, 0x0); +} + +void msm_dp_ctrl_phy_init(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl; + struct phy *phy; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + phy = ctrl->phy; + + msm_dp_ctrl_phy_reset(ctrl); + phy_init(phy); +} + +void msm_dp_ctrl_phy_exit(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl; + struct phy *phy; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + phy = ctrl->phy; + + msm_dp_ctrl_phy_reset(ctrl); + phy_exit(phy); +} + +static int msm_dp_ctrl_reinitialize_mainlink(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel) +{ + struct phy *phy = ctrl->phy; + int ret = 0; + + msm_dp_ctrl_mainlink_disable(ctrl); + ctrl->phy_opts.dp.lanes = ctrl->link->link_params.num_lanes; + phy_configure(phy, &ctrl->phy_opts); + + /* + * Disable and re-enable the mainlink clock since the + * link clock might have been adjusted as part of the + * link maintenance. + */ + msm_dp_ctrl_link_clk_disable(&ctrl->msm_dp_ctrl); + + phy_power_off(phy); + /* hw recommended delay before re-enabling clocks */ + msleep(20); + + ret = msm_dp_ctrl_enable_mainlink_clocks(ctrl, panel); + if (ret) { + DRM_ERROR("Failed to enable mainlink clks. ret=%d\n", ret); + return ret; + } + + return ret; +} + +static int msm_dp_ctrl_deinitialize_mainlink(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel) +{ + struct phy *phy; + + phy = ctrl->phy; + + msm_dp_ctrl_mainlink_disable(ctrl); + + msm_dp_ctrl_reset(&ctrl->msm_dp_ctrl, panel); + + msm_dp_ctrl_link_clk_disable(&ctrl->msm_dp_ctrl); + + phy_power_off(phy); + + /* aux channel down, reinit phy */ + phy_exit(phy); + phy_init(phy); + + return 0; +} + +static int msm_dp_ctrl_link_maintenance(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel) +{ + int ret = 0; + int training_step = DP_TRAINING_NONE; + + msm_dp_ctrl_push_idle(&ctrl->msm_dp_ctrl); + + ctrl->link->phy_params.p_level = 0; + ctrl->link->phy_params.v_level = 0; + + ret = msm_dp_ctrl_setup_main_link(ctrl, panel, &training_step); + if (ret) + goto end; + + msm_dp_ctrl_clear_training_pattern(ctrl, panel, DP_PHY_DPRX); + + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, DP_STATE_CTRL_SEND_VIDEO); + + ret = msm_dp_ctrl_wait4video_ready(ctrl); +end: + return ret; +} + +#define SCRAMBLER_RESET_COUNT_VALUE 0xFC + +static void msm_dp_ctrl_send_phy_pattern(struct msm_dp_ctrl_private *ctrl, + u32 pattern) +{ + u32 value = 0x0; + + /* Make sure to clear the current pattern before starting a new one */ + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, 0x0); + + drm_dbg_dp(ctrl->drm_dev, "pattern: %#x\n", pattern); + switch (pattern) { + case DP_PHY_TEST_PATTERN_D10_2: + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, + DP_STATE_CTRL_LINK_TRAINING_PATTERN1); + break; + + case DP_PHY_TEST_PATTERN_ERROR_COUNT: + value &= ~(1 << 16); + msm_dp_write_link(ctrl, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET, + value); + value |= SCRAMBLER_RESET_COUNT_VALUE; + msm_dp_write_link(ctrl, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET, + value); + msm_dp_write_link(ctrl, REG_DP_MAINLINK_LEVELS, + DP_MAINLINK_SAFE_TO_EXIT_LEVEL_2); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, + DP_STATE_CTRL_LINK_SYMBOL_ERR_MEASURE); + break; + + case DP_PHY_TEST_PATTERN_PRBS7: + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, + DP_STATE_CTRL_LINK_PRBS7); + break; + + case DP_PHY_TEST_PATTERN_80BIT_CUSTOM: + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, + DP_STATE_CTRL_LINK_TEST_CUSTOM_PATTERN); + /* 00111110000011111000001111100000 */ + msm_dp_write_link(ctrl, REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG0, + 0x3E0F83E0); + /* 00001111100000111110000011111000 */ + msm_dp_write_link(ctrl, REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG1, + 0x0F83E0F8); + /* 1111100000111110 */ + msm_dp_write_link(ctrl, REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG2, + 0x0000F83E); + break; + + case DP_PHY_TEST_PATTERN_CP2520: + value = msm_dp_read_link(ctrl, REG_DP_MAINLINK_CTRL); + value &= ~DP_MAINLINK_CTRL_SW_BYPASS_SCRAMBLER; + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, value); + + value = DP_HBR2_ERM_PATTERN; + msm_dp_write_link(ctrl, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET, + value); + value |= SCRAMBLER_RESET_COUNT_VALUE; + msm_dp_write_link(ctrl, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET, + value); + msm_dp_write_link(ctrl, REG_DP_MAINLINK_LEVELS, + DP_MAINLINK_SAFE_TO_EXIT_LEVEL_2); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, + DP_STATE_CTRL_LINK_SYMBOL_ERR_MEASURE); + value = msm_dp_read_link(ctrl, REG_DP_MAINLINK_CTRL); + value |= DP_MAINLINK_CTRL_ENABLE; + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, value); + break; + + case DP_PHY_TEST_PATTERN_SEL_MASK: + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, + DP_MAINLINK_CTRL_ENABLE); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, + DP_STATE_CTRL_LINK_TRAINING_PATTERN4); + break; + + default: + drm_dbg_dp(ctrl->drm_dev, + "No valid test pattern requested: %#x\n", pattern); + break; + } +} + +static bool msm_dp_ctrl_send_phy_test_pattern(struct msm_dp_ctrl_private *ctrl) +{ + bool success = false; + u32 pattern_sent = 0x0; + u32 pattern_requested = ctrl->link->phy_params.phy_test_pattern_sel; + + drm_dbg_dp(ctrl->drm_dev, "request: 0x%x\n", pattern_requested); + + if (msm_dp_ctrl_set_vx_px(ctrl, + ctrl->link->phy_params.v_level, + ctrl->link->phy_params.p_level)) { + DRM_ERROR("Failed to set v/p levels\n"); + return false; + } + msm_dp_ctrl_send_phy_pattern(ctrl, pattern_requested); + msm_dp_ctrl_update_phy_vx_px(ctrl, DP_PHY_DPRX); + msm_dp_link_send_test_response(ctrl->link); + + pattern_sent = msm_dp_read_link(ctrl, REG_DP_MAINLINK_READY); + + switch (pattern_sent) { + case MR_LINK_TRAINING1: + success = (pattern_requested == + DP_PHY_TEST_PATTERN_D10_2); + break; + case MR_LINK_SYMBOL_ERM: + success = ((pattern_requested == + DP_PHY_TEST_PATTERN_ERROR_COUNT) || + (pattern_requested == + DP_PHY_TEST_PATTERN_CP2520)); + break; + case MR_LINK_PRBS7: + success = (pattern_requested == + DP_PHY_TEST_PATTERN_PRBS7); + break; + case MR_LINK_CUSTOM80: + success = (pattern_requested == + DP_PHY_TEST_PATTERN_80BIT_CUSTOM); + break; + case MR_LINK_TRAINING4: + success = (pattern_requested == + DP_PHY_TEST_PATTERN_SEL_MASK); + break; + default: + success = false; + } + + drm_dbg_dp(ctrl->drm_dev, "%s: test->0x%x\n", + success ? "success" : "failed", pattern_requested); + return success; +} + +static int msm_dp_ctrl_on_pixel_clk(struct msm_dp_ctrl_private *ctrl, unsigned long pixel_rate) +{ + int ret; + + ret = clk_set_rate(ctrl->pixel_clk, pixel_rate * 1000); + if (ret) { + DRM_ERROR("Failed to set pixel clock rate. ret=%d\n", ret); + return ret; + } + + if (WARN_ON_ONCE(ctrl->stream_clks_on)) + return 0; + + ret = clk_prepare_enable(ctrl->pixel_clk); + if (ret) { + DRM_ERROR("Failed to start pixel clocks. ret=%d\n", ret); + return ret; + } + ctrl->stream_clks_on = true; + + return ret; +} + +void msm_dp_ctrl_off_pixel_clk(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + + if (ctrl->stream_clks_on) { + clk_disable_unprepare(ctrl->pixel_clk); + ctrl->stream_clks_on = false; + } +} + +static int msm_dp_ctrl_process_phy_test_request(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel) +{ + int ret; + unsigned long pixel_rate; + + if (!ctrl->link->phy_params.phy_test_pattern_sel) { + drm_dbg_dp(ctrl->drm_dev, + "no test pattern selected by sink\n"); + return 0; + } + + /* + * The global reset will need DP link related clocks to be + * running. Add the global reset just before disabling the + * link clocks and core clocks. + */ + msm_dp_ctrl_off_pixel_clk(&ctrl->msm_dp_ctrl); + msm_dp_ctrl_off_link(&ctrl->msm_dp_ctrl, panel); + + ret = msm_dp_ctrl_on_link(&ctrl->msm_dp_ctrl, panel); + if (ret) { + DRM_ERROR("failed to enable DP link controller\n"); + return ret; + } + + pixel_rate = panel->msm_dp_mode.drm_mode.clock; + ret = msm_dp_ctrl_on_pixel_clk(ctrl, pixel_rate); + if (ret) + return ret; + + msm_dp_ctrl_send_phy_test_pattern(ctrl); + + return 0; +} + +void msm_dp_ctrl_handle_sink_request(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel) +{ + struct msm_dp_ctrl_private *ctrl; + u32 sink_request = 0x0; + + if (!msm_dp_ctrl) { + DRM_ERROR("invalid input\n"); + return; + } + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + sink_request = ctrl->link->sink_request; + + if (sink_request & DP_TEST_LINK_PHY_TEST_PATTERN) { + drm_dbg_dp(ctrl->drm_dev, "PHY_TEST_PATTERN request\n"); + if (msm_dp_ctrl_process_phy_test_request(ctrl, panel)) { + DRM_ERROR("process phy_test_req failed\n"); + return; + } + } + + if (sink_request & DP_LINK_STATUS_UPDATED) { + if (msm_dp_ctrl_link_maintenance(ctrl, panel)) { + DRM_ERROR("LM failed: TEST_LINK_TRAINING\n"); + return; + } + } + + if (sink_request & DP_TEST_LINK_TRAINING) { + msm_dp_link_send_test_response(ctrl->link); + if (msm_dp_ctrl_link_maintenance(ctrl, panel)) { + DRM_ERROR("LM failed: TEST_LINK_TRAINING\n"); + return; + } + } +} + +static bool msm_dp_ctrl_clock_recovery_any_ok( + const u8 link_status[DP_LINK_STATUS_SIZE], + int lane_count) +{ + int reduced_cnt; + + if (lane_count <= 1) + return false; + + /* + * only interested in the lane number after reduced + * lane_count = 4, then only interested in 2 lanes + * lane_count = 2, then only interested in 1 lane + */ + reduced_cnt = lane_count >> 1; + + return drm_dp_clock_recovery_ok(link_status, reduced_cnt); +} + +static bool msm_dp_ctrl_channel_eq_ok(struct msm_dp_ctrl_private *ctrl) +{ + u8 link_status[DP_LINK_STATUS_SIZE]; + int num_lanes = ctrl->link->link_params.num_lanes; + + drm_dp_dpcd_read_link_status(ctrl->aux, link_status); + + return drm_dp_channel_eq_ok(link_status, num_lanes); +} + +int msm_dp_ctrl_on_link(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel) +{ + int rc = 0; + struct msm_dp_ctrl_private *ctrl; + u32 rate; + int link_train_max_retries = 5; + u32 const phy_cts_pixel_clk_khz = 148500; + u8 link_status[DP_LINK_STATUS_SIZE]; + unsigned int training_step; + unsigned long pixel_rate; + + if (!msm_dp_ctrl) + return -EINVAL; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + + rate = panel->link_info.rate; + pixel_rate = panel->msm_dp_mode.drm_mode.clock; + + msm_dp_ctrl_core_clk_enable(&ctrl->msm_dp_ctrl); + + if (ctrl->link->sink_request & DP_TEST_LINK_PHY_TEST_PATTERN) { + drm_dbg_dp(ctrl->drm_dev, + "using phy test link parameters\n"); + if (!pixel_rate) + pixel_rate = phy_cts_pixel_clk_khz; + } else { + ctrl->link->link_params.rate = rate; + ctrl->link->link_params.num_lanes = + panel->link_info.num_lanes; + if (panel->msm_dp_mode.out_fmt_is_yuv_420) + pixel_rate >>= 1; + } + + drm_dbg_dp(ctrl->drm_dev, "rate=%d, num_lanes=%d, pixel_rate=%lu\n", + ctrl->link->link_params.rate, ctrl->link->link_params.num_lanes, + pixel_rate); + + rc = msm_dp_ctrl_enable_mainlink_clocks(ctrl, panel); + if (rc) + return rc; + + while (--link_train_max_retries) { + training_step = DP_TRAINING_NONE; + rc = msm_dp_ctrl_setup_main_link(ctrl, panel, &training_step); + if (rc == 0) { + /* training completed successfully */ + break; + } else if (training_step == DP_TRAINING_1) { + /* link train_1 failed */ + if (!msm_dp_aux_is_link_connected(ctrl->aux)) + break; + + drm_dp_dpcd_read_link_status(ctrl->aux, link_status); + + rc = msm_dp_ctrl_link_rate_down_shift(ctrl); + if (rc < 0) { /* already in RBR = 1.6G */ + if (msm_dp_ctrl_clock_recovery_any_ok(link_status, + ctrl->link->link_params.num_lanes)) { + /* + * some lanes are ready, + * reduce lane number + */ + rc = msm_dp_ctrl_link_lane_down_shift(ctrl, panel); + if (rc < 0) { /* lane == 1 already */ + /* end with failure */ + break; + } + } else { + /* end with failure */ + break; /* lane == 1 already */ + } + } + } else if (training_step == DP_TRAINING_2) { + /* link train_2 failed */ + if (!msm_dp_aux_is_link_connected(ctrl->aux)) + break; + + drm_dp_dpcd_read_link_status(ctrl->aux, link_status); + + if (!drm_dp_clock_recovery_ok(link_status, + ctrl->link->link_params.num_lanes)) + rc = msm_dp_ctrl_link_rate_down_shift(ctrl); + else + rc = msm_dp_ctrl_link_lane_down_shift(ctrl, panel); + + if (rc < 0) { + /* end with failure */ + break; /* lane == 1 already */ + } + + /* stop link training before start re training */ + msm_dp_ctrl_clear_training_pattern(ctrl, panel, DP_PHY_DPRX); + } + + rc = msm_dp_ctrl_reinitialize_mainlink(ctrl, panel); + if (rc) { + DRM_ERROR("Failed to reinitialize mainlink. rc=%d\n", rc); + break; + } + } + + if (ctrl->link->sink_request & DP_TEST_LINK_PHY_TEST_PATTERN) + return rc; + + if (rc == 0) { /* link train successfully */ + /* + * do not stop train pattern here + * stop link training at on_stream + * to pass compliance test + */ + } else { + /* + * link training failed + * end txing train pattern here + */ + msm_dp_ctrl_clear_training_pattern(ctrl, panel, DP_PHY_DPRX); + + msm_dp_ctrl_deinitialize_mainlink(ctrl, panel); + rc = -ECONNRESET; + } + + return rc; +} + +static int msm_dp_ctrl_link_retrain(struct msm_dp_ctrl_private *ctrl, + struct msm_dp_panel *panel) +{ + int training_step = DP_TRAINING_NONE; + + return msm_dp_ctrl_setup_main_link(ctrl, panel, &training_step); +} + +static void msm_dp_ctrl_config_msa(struct msm_dp_ctrl_private *ctrl, + u32 rate, u32 stream_rate_khz, + bool is_ycbcr_420) +{ + u32 pixel_m, pixel_n; + u32 mvid, nvid, pixel_div, dispcc_input_rate; + u32 const nvid_fixed = DP_LINK_CONSTANT_N_VALUE; + u32 const link_rate_hbr2 = 540000; + u32 const link_rate_hbr3 = 810000; + unsigned long den, num; + + switch (rate) { + case link_rate_hbr3: + pixel_div = 6; + break; + case link_rate_hbr2: + pixel_div = 4; + break; + case 162000: + case 270000: + pixel_div = 2; + break; + default: + /* + * This cannot be reached but the compiler is not able to know + * that statically so return early to avoid a possibly invalid + * division. + */ + DRM_ERROR("Invalid pixel mux divider\n"); + return; + } + + dispcc_input_rate = (rate * 10) / pixel_div; + + rational_best_approximation(dispcc_input_rate, stream_rate_khz, + (unsigned long)(1 << 16) - 1, + (unsigned long)(1 << 16) - 1, &den, &num); + + den = ~(den - num); + den = den & 0xFFFF; + pixel_m = num; + pixel_n = den; + + mvid = (pixel_m & 0xFFFF) * 5; + nvid = (0xFFFF & (~pixel_n)) + (pixel_m & 0xFFFF); + + if (nvid < nvid_fixed) { + u32 temp; + + temp = (nvid_fixed / nvid) * nvid; + mvid = (nvid_fixed / nvid) * mvid; + nvid = temp; + } + + if (is_ycbcr_420) + mvid /= 2; + + if (link_rate_hbr2 == rate) + nvid *= 2; + + if (link_rate_hbr3 == rate) + nvid *= 3; + + drm_dbg_dp(ctrl->drm_dev, "mvid=0x%x, nvid=0x%x\n", mvid, nvid); + msm_dp_write_link(ctrl, REG_DP_SOFTWARE_MVID, mvid); + msm_dp_write_link(ctrl, REG_DP_SOFTWARE_NVID, nvid); +} + +int msm_dp_ctrl_prepare_stream_on(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel, + bool force_link_train) +{ + int ret = 0; + struct msm_dp_ctrl_private *ctrl; + + if (!msm_dp_ctrl) + return -EINVAL; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + + drm_dbg_dp(ctrl->drm_dev, "rate=%d, num_lanes=%d\n", + ctrl->link->link_params.rate, + ctrl->link->link_params.num_lanes); + + drm_dbg_dp(ctrl->drm_dev, + "core_clk_on=%d link_clk_on=%d stream_clk_on=%d\n", + ctrl->core_clks_on, ctrl->link_clks_on, ctrl->stream_clks_on); + + if (!ctrl->link_clks_on) { /* link clk is off */ + ret = msm_dp_ctrl_enable_mainlink_clocks(ctrl, panel); + if (ret) { + DRM_ERROR("Failed to start link clocks. ret=%d\n", ret); + return ret; + } + } + + if (force_link_train || !msm_dp_ctrl_channel_eq_ok(ctrl)) + msm_dp_ctrl_link_retrain(ctrl, panel); + + /* stop txing train pattern to end link training */ + msm_dp_ctrl_clear_training_pattern(ctrl, panel, DP_PHY_DPRX); + + return ret; +} + +int msm_dp_ctrl_on_stream(struct msm_dp_ctrl *msm_dp_ctrl, struct msm_dp_panel *panel) +{ + int ret = 0; + bool mainlink_ready = false; + struct msm_dp_ctrl_private *ctrl; + unsigned long pixel_rate; + unsigned long pixel_rate_orig; + + if (!msm_dp_ctrl) + return -EINVAL; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + + pixel_rate_orig = panel->msm_dp_mode.drm_mode.clock; + pixel_rate = pixel_rate_orig; + + if (msm_dp_ctrl->wide_bus_en || panel->msm_dp_mode.out_fmt_is_yuv_420) + pixel_rate >>= 1; + + drm_dbg_dp(ctrl->drm_dev, "pixel_rate=%lu\n", pixel_rate); + + ret = msm_dp_ctrl_on_pixel_clk(ctrl, pixel_rate); + if (ret) + return ret; + + /* + * Set up transfer unit values and set controller state to send + * video. + */ + reinit_completion(&ctrl->video_comp); + + msm_dp_ctrl_lane_mapping(ctrl); + msm_dp_setup_peripheral_flush(ctrl); + msm_dp_ctrl_config_ctrl_link(ctrl, panel); + + msm_dp_ctrl_configure_source_params(ctrl, panel); + + msm_dp_ctrl_config_msa(ctrl, + ctrl->link->link_params.rate, + pixel_rate_orig, + panel->msm_dp_mode.out_fmt_is_yuv_420); + + msm_dp_panel_clear_dsc_dto(panel); + + msm_dp_ctrl_setup_tr_unit(ctrl, panel); + + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, DP_STATE_CTRL_SEND_VIDEO); + + ret = msm_dp_ctrl_wait4video_ready(ctrl); + if (ret) + return ret; + + mainlink_ready = msm_dp_ctrl_mainlink_ready(ctrl); + drm_dbg_dp(ctrl->drm_dev, + "mainlink %s\n", mainlink_ready ? "READY" : "NOT READY"); + + return ret; +} + +void msm_dp_ctrl_reinit_phy(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl; + struct phy *phy; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + phy = ctrl->phy; + + /* aux channel down, reinit phy */ + phy_exit(phy); + phy_init(phy); +} + +void msm_dp_ctrl_off_link(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel) +{ + struct msm_dp_ctrl_private *ctrl; + struct phy *phy; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + phy = ctrl->phy; + + msm_dp_panel_disable_vsc_sdp(panel); + + msm_dp_ctrl_mainlink_disable(ctrl); + + msm_dp_ctrl_reset(&ctrl->msm_dp_ctrl, panel); + + msm_dp_ctrl_link_clk_disable(&ctrl->msm_dp_ctrl); + + phy_power_off(phy); +} + +irqreturn_t msm_dp_ctrl_isr(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel) +{ + struct msm_dp_ctrl_private *ctrl; + u32 isr; + irqreturn_t ret = IRQ_NONE; + + if (!msm_dp_ctrl) + return IRQ_NONE; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + + if (panel->psr_cap.version) { + isr = msm_dp_ctrl_get_psr_interrupt(ctrl); + + if (isr) + complete(&ctrl->psr_op_comp); + + if (isr & PSR_EXIT_INT) + drm_dbg_dp(ctrl->drm_dev, "PSR exit done\n"); + + if (isr & PSR_UPDATE_INT) + drm_dbg_dp(ctrl->drm_dev, "PSR frame update done\n"); + + if (isr & PSR_CAPTURE_INT) + drm_dbg_dp(ctrl->drm_dev, "PSR frame capture done\n"); + } + + isr = msm_dp_ctrl_get_interrupt(ctrl); + + if (isr & DP_CTRL_INTR_READY_FOR_VIDEO) { + drm_dbg_dp(ctrl->drm_dev, "dp_video_ready\n"); + complete(&ctrl->video_comp); + ret = IRQ_HANDLED; + } + + if (isr & DP_CTRL_INTR_IDLE_PATTERN_SENT) { + drm_dbg_dp(ctrl->drm_dev, "idle_patterns_sent\n"); + complete(&ctrl->idle_comp); + ret = IRQ_HANDLED; + } + + /* DP aux isr */ + isr = msm_dp_ctrl_get_aux_interrupt(ctrl); + if (isr) + ret |= msm_dp_aux_isr(ctrl->aux, isr); + + return ret; +} + +static const char *core_clks[] = { + "core_iface", + "core_aux", +}; + +static const char *ctrl_clks[] = { + "ctrl_link", + "ctrl_link_iface", +}; + +static int msm_dp_ctrl_clk_init(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl; + struct device *dev; + int i, rc; + + ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + dev = ctrl->dev; + + ctrl->num_core_clks = ARRAY_SIZE(core_clks); + ctrl->core_clks = devm_kcalloc(dev, ctrl->num_core_clks, sizeof(*ctrl->core_clks), GFP_KERNEL); + if (!ctrl->core_clks) + return -ENOMEM; + + for (i = 0; i < ctrl->num_core_clks; i++) + ctrl->core_clks[i].id = core_clks[i]; + + rc = devm_clk_bulk_get(dev, ctrl->num_core_clks, ctrl->core_clks); + if (rc) + return rc; + + ctrl->num_link_clks = ARRAY_SIZE(ctrl_clks); + ctrl->link_clks = devm_kcalloc(dev, ctrl->num_link_clks, sizeof(*ctrl->link_clks), GFP_KERNEL); + if (!ctrl->link_clks) + return -ENOMEM; + + for (i = 0; i < ctrl->num_link_clks; i++) + ctrl->link_clks[i].id = ctrl_clks[i]; + + rc = devm_clk_bulk_get(dev, ctrl->num_link_clks, ctrl->link_clks); + if (rc) + return rc; + + ctrl->pixel_clk = devm_clk_get(dev, "stream_pixel"); + if (IS_ERR(ctrl->pixel_clk)) + return PTR_ERR(ctrl->pixel_clk); + + return 0; +} + +struct msm_dp_ctrl *msm_dp_ctrl_get(struct device *dev, struct msm_dp_link *link, + struct drm_dp_aux *aux, + struct phy *phy, + void __iomem *ahb_base, + void __iomem *link_base) +{ + struct msm_dp_ctrl_private *ctrl; + int ret; + + if (!dev || !aux || !link) { + DRM_ERROR("invalid input\n"); + return ERR_PTR(-EINVAL); + } + + ctrl = devm_kzalloc(dev, sizeof(*ctrl), GFP_KERNEL); + if (!ctrl) { + DRM_ERROR("Mem allocation failure\n"); + return ERR_PTR(-ENOMEM); + } + + ret = devm_pm_opp_set_clkname(dev, "ctrl_link"); + if (ret) { + dev_err(dev, "invalid DP OPP table in device tree\n"); + /* caller do PTR_ERR(opp_table) */ + return (struct msm_dp_ctrl *)ERR_PTR(ret); + } + + /* OPP table is optional */ + ret = devm_pm_opp_of_add_table(dev); + if (ret) + dev_err(dev, "failed to add DP OPP table\n"); + + init_completion(&ctrl->idle_comp); + init_completion(&ctrl->psr_op_comp); + init_completion(&ctrl->video_comp); + + /* in parameters */ + ctrl->aux = aux; + ctrl->link = link; + ctrl->dev = dev; + ctrl->phy = phy; + ctrl->ahb_base = ahb_base; + ctrl->link_base = link_base; + + ret = msm_dp_ctrl_clk_init(&ctrl->msm_dp_ctrl); + if (ret) { + dev_err(dev, "failed to init clocks\n"); + return ERR_PTR(ret); + } + + return &ctrl->msm_dp_ctrl; +} diff --git a/drivers/gpu/drm/msm/dp/dp_ctrl.h b/drivers/gpu/drm/msm/dp/dp_ctrl.h new file mode 100644 index 000000000..5902cf7e7 --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_ctrl.h @@ -0,0 +1,58 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved. + */ + +#ifndef _DP_CTRL_H_ +#define _DP_CTRL_H_ + +#include "dp_aux.h" +#include "dp_panel.h" +#include "dp_link.h" + +struct msm_dp_ctrl { + bool wide_bus_en; +}; + +struct phy; + +int msm_dp_ctrl_on_link(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel); +int msm_dp_ctrl_on_stream(struct msm_dp_ctrl *msm_dp_ctrl, struct msm_dp_panel *panel); +int msm_dp_ctrl_prepare_stream_on(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel, + bool force_link_train); +void msm_dp_ctrl_off_link(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel); +void msm_dp_ctrl_off_pixel_clk(struct msm_dp_ctrl *msm_dp_ctrl); +void msm_dp_ctrl_push_idle(struct msm_dp_ctrl *msm_dp_ctrl); +irqreturn_t msm_dp_ctrl_isr(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel); +void msm_dp_ctrl_handle_sink_request(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel); +struct msm_dp_ctrl *msm_dp_ctrl_get(struct device *dev, + struct msm_dp_link *link, + struct drm_dp_aux *aux, + struct phy *phy, + void __iomem *ahb_base, + void __iomem *link_base); + +void msm_dp_ctrl_reset(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel); +void msm_dp_ctrl_phy_init(struct msm_dp_ctrl *msm_dp_ctrl); +void msm_dp_ctrl_phy_exit(struct msm_dp_ctrl *msm_dp_ctrl); +void msm_dp_ctrl_irq_phy_exit(struct msm_dp_ctrl *msm_dp_ctrl); + +void msm_dp_ctrl_set_psr(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel, bool enable); +void msm_dp_ctrl_config_psr(struct msm_dp_ctrl *msm_dp_ctrl, + struct msm_dp_panel *panel); + +int msm_dp_ctrl_core_clk_enable(struct msm_dp_ctrl *msm_dp_ctrl); +void msm_dp_ctrl_core_clk_disable(struct msm_dp_ctrl *msm_dp_ctrl); + +void msm_dp_ctrl_enable_irq(struct msm_dp_ctrl *msm_dp_ctrl); +void msm_dp_ctrl_disable_irq(struct msm_dp_ctrl *msm_dp_ctrl); + +void msm_dp_ctrl_reinit_phy(struct msm_dp_ctrl *msm_dp_ctrl); +#endif /* _DP_CTRL_H_ */ diff --git a/drivers/gpu/drm/msm/dp/dp_debug.c b/drivers/gpu/drm/msm/dp/dp_debug.c new file mode 100644 index 000000000..cf3838fcd --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_debug.c @@ -0,0 +1,240 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved. + */ + +#define pr_fmt(fmt)"[drm-dp] %s: " fmt, __func__ + +#ifdef CONFIG_DEBUG_FS + +#include +#include +#include + +#include "dp_aux.h" +#include "dp_ctrl.h" +#include "dp_debug.h" +#include "dp_display.h" + +#define DEBUG_NAME "msm_dp" + +struct msm_dp_debug_private { + struct msm_dp_link *link; + struct msm_dp_panel *panel; + struct drm_connector *connector; +}; + +static int msm_dp_debug_show(struct seq_file *seq, void *p) +{ + struct msm_dp_debug_private *debug = seq->private; + u64 lclk = 0; + u32 link_params_rate; + const struct drm_display_mode *drm_mode; + + if (!debug) + return -ENODEV; + + drm_mode = &debug->panel->msm_dp_mode.drm_mode; + + seq_printf(seq, "\tname = %s\n", DEBUG_NAME); + seq_printf(seq, "\tdrm_dp_link\n\t\trate = %u\n", + debug->panel->link_info.rate); + seq_printf(seq, "\t\tnum_lanes = %u\n", + debug->panel->link_info.num_lanes); + seq_printf(seq, "\t\tcapabilities = %lu\n", + debug->panel->link_info.capabilities); + seq_printf(seq, "\tdp_panel_info:\n\t\tactive = %dx%d\n", + drm_mode->hdisplay, + drm_mode->vdisplay); + seq_printf(seq, "\t\tback_porch = %dx%d\n", + drm_mode->htotal - drm_mode->hsync_end, + drm_mode->vtotal - drm_mode->vsync_end); + seq_printf(seq, "\t\tfront_porch = %dx%d\n", + drm_mode->hsync_start - drm_mode->hdisplay, + drm_mode->vsync_start - drm_mode->vdisplay); + seq_printf(seq, "\t\tsync_width = %dx%d\n", + drm_mode->hsync_end - drm_mode->hsync_start, + drm_mode->vsync_end - drm_mode->vsync_start); + seq_printf(seq, "\t\tactive_low = %dx%d\n", + debug->panel->msm_dp_mode.h_active_low, + debug->panel->msm_dp_mode.v_active_low); + seq_printf(seq, "\t\th_skew = %d\n", + drm_mode->hskew); + seq_printf(seq, "\t\trefresh rate = %d\n", + drm_mode_vrefresh(drm_mode)); + seq_printf(seq, "\t\tpixel clock khz = %d\n", + drm_mode->clock); + seq_printf(seq, "\t\tbpp = %d\n", + debug->panel->msm_dp_mode.bpp); + + /* Link Information */ + seq_printf(seq, "\tdp_link:\n\t\ttest_requested = %d\n", + debug->link->sink_request); + seq_printf(seq, "\t\tnum_lanes = %d\n", + debug->link->link_params.num_lanes); + link_params_rate = debug->link->link_params.rate; + seq_printf(seq, "\t\tbw_code = %d\n", + drm_dp_link_rate_to_bw_code(link_params_rate)); + lclk = debug->link->link_params.rate * 1000; + seq_printf(seq, "\t\tlclk = %lld\n", lclk); + seq_printf(seq, "\t\tv_level = %d\n", + debug->link->phy_params.v_level); + seq_printf(seq, "\t\tp_level = %d\n", + debug->link->phy_params.p_level); + + return 0; +} +DEFINE_SHOW_ATTRIBUTE(msm_dp_debug); + +static int msm_dp_test_data_show(struct seq_file *m, void *data) +{ + const struct msm_dp_debug_private *debug = m->private; + const struct drm_connector *connector = debug->connector; + u32 bpc; + + if (connector->status == connector_status_connected) { + bpc = debug->link->test_video.test_bit_depth; + seq_printf(m, "hdisplay: %d\n", + debug->link->test_video.test_h_width); + seq_printf(m, "vdisplay: %d\n", + debug->link->test_video.test_v_height); + seq_printf(m, "bpc: %u\n", + msm_dp_link_bit_depth_to_bpp(bpc) / 3); + } else { + seq_puts(m, "0"); + } + + return 0; +} +DEFINE_SHOW_ATTRIBUTE(msm_dp_test_data); + +static int msm_dp_test_type_show(struct seq_file *m, void *data) +{ + const struct msm_dp_debug_private *debug = m->private; + const struct drm_connector *connector = debug->connector; + + if (connector->status == connector_status_connected) + seq_printf(m, "%02x", DP_TEST_LINK_VIDEO_PATTERN); + else + seq_puts(m, "0"); + + return 0; +} +DEFINE_SHOW_ATTRIBUTE(msm_dp_test_type); + +static ssize_t msm_dp_test_active_write(struct file *file, + const char __user *ubuf, + size_t len, loff_t *offp) +{ + char *input_buffer; + int status = 0; + const struct msm_dp_debug_private *debug; + const struct drm_connector *connector; + int val = 0; + + debug = ((struct seq_file *)file->private_data)->private; + connector = debug->connector; + + if (len == 0) + return 0; + + input_buffer = memdup_user_nul(ubuf, len); + if (IS_ERR(input_buffer)) + return PTR_ERR(input_buffer); + + DRM_DEBUG_DRIVER("Copied %d bytes from user\n", (unsigned int)len); + + if (connector->status == connector_status_connected) { + status = kstrtoint(input_buffer, 10, &val); + if (status < 0) { + kfree(input_buffer); + return status; + } + DRM_DEBUG_DRIVER("Got %d for test active\n", val); + /* To prevent erroneous activation of the compliance + * testing code, only accept an actual value of 1 here + */ + if (val == 1) + debug->panel->video_test = true; + else + debug->panel->video_test = false; + } + kfree(input_buffer); + + *offp += len; + return len; +} + +static int msm_dp_test_active_show(struct seq_file *m, void *data) +{ + struct msm_dp_debug_private *debug = m->private; + struct drm_connector *connector = debug->connector; + + if (connector->status == connector_status_connected) { + if (debug->panel->video_test) + seq_puts(m, "1"); + else + seq_puts(m, "0"); + } else { + seq_puts(m, "0"); + } + + return 0; +} + +static int msm_dp_test_active_open(struct inode *inode, + struct file *file) +{ + return single_open(file, msm_dp_test_active_show, + inode->i_private); +} + +static const struct file_operations test_active_fops = { + .owner = THIS_MODULE, + .open = msm_dp_test_active_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = msm_dp_test_active_write +}; + +int msm_dp_debug_init(struct device *dev, struct msm_dp_panel *panel, + struct msm_dp_link *link, + struct drm_connector *connector, + struct dentry *root, bool is_edp) +{ + struct msm_dp_debug_private *debug; + + if (!dev || !panel || !link) { + DRM_ERROR("invalid input\n"); + return -EINVAL; + } + + debug = devm_kzalloc(dev, sizeof(*debug), GFP_KERNEL); + if (!debug) + return -ENOMEM; + + debug->link = link; + debug->panel = panel; + + debugfs_create_file("dp_debug", 0444, root, + debug, &msm_dp_debug_fops); + + if (!is_edp) { + debugfs_create_file("dp_test_active", 0444, + root, + debug, &test_active_fops); + + debugfs_create_file("dp_test_data", 0444, + root, + debug, &msm_dp_test_data_fops); + + debugfs_create_file("dp_test_type", 0444, + root, + debug, &msm_dp_test_type_fops); + } + + return 0; +} + +#endif diff --git a/drivers/gpu/drm/msm/dp/dp_debug.h b/drivers/gpu/drm/msm/dp/dp_debug.h new file mode 100644 index 000000000..a90083fec --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_debug.h @@ -0,0 +1,48 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved. + */ + +#ifndef _DP_DEBUG_H_ +#define _DP_DEBUG_H_ + +#include "dp_panel.h" +#include "dp_link.h" + +#if defined(CONFIG_DEBUG_FS) + +/** + * msm_dp_debug_init() - configure and get the DisplayPlot debug module data + * + * @dev: device instance of the caller + * @panel: instance of panel module + * @link: instance of link module + * @connector: double pointer to display connector + * @root: connector's debugfs root + * @is_edp: set for eDP connectors / panels + * return: pointer to allocated debug module data + * + * This function sets up the debug module and provides a way + * for debugfs input to be communicated with existing modules + */ +int msm_dp_debug_init(struct device *dev, struct msm_dp_panel *panel, + struct msm_dp_link *link, + struct drm_connector *connector, + struct dentry *root, + bool is_edp); + +#else + +static inline +int msm_dp_debug_init(struct device *dev, struct msm_dp_panel *panel, + struct msm_dp_link *link, + struct drm_connector *connector, + struct dentry *root, + bool is_edp) +{ + return -EINVAL; +} + +#endif /* defined(CONFIG_DEBUG_FS) */ + +#endif /* _DP_DEBUG_H_ */ diff --git a/drivers/gpu/drm/msm/dp/dp_display.c b/drivers/gpu/drm/msm/dp/dp_display.c new file mode 100644 index 000000000..4dcbd9b99 --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_display.c @@ -0,0 +1,1574 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "msm_drv.h" +#include "msm_kms.h" +#include "dp_ctrl.h" +#include "dp_aux.h" +#include "dp_reg.h" +#include "dp_link.h" +#include "dp_panel.h" +#include "dp_display.h" +#include "dp_drm.h" +#include "dp_audio.h" +#include "dp_debug.h" + +static bool psr_enabled = false; +module_param(psr_enabled, bool, 0); +MODULE_PARM_DESC(psr_enabled, "enable PSR for eDP and DP displays"); + +#define HPD_STRING_SIZE 30 + +enum { + ISR_DISCONNECTED, + ISR_CONNECT_PENDING, + ISR_CONNECTED, + ISR_HPD_IO_GLITCH_COUNT, + ISR_IRQ_HPD_PULSE_COUNT, + ISR_HPD_REPLUG_COUNT, +}; + +struct msm_dp_display_private { + int irq; + + unsigned int id; + + /* state variables */ + bool core_initialized; + bool phy_initialized; + bool audio_supported; + + struct mutex plugged_lock; + bool plugged; + + struct drm_device *drm_dev; + + struct drm_dp_aux *aux; + struct msm_dp_link *link; + struct msm_dp_panel *panel; + struct msm_dp_ctrl *ctrl; + + struct msm_dp msm_dp_display; + + /* wait for audio signaling */ + struct completion audio_comp; + + /* HPD IRQ handling */ + spinlock_t irq_thread_lock; + u32 hpd_isr_status; + + bool wide_bus_supported; + + struct msm_dp_audio *audio; + + void __iomem *ahb_base; + size_t ahb_len; + + void __iomem *aux_base; + size_t aux_len; + + void __iomem *link_base; + size_t link_len; + + void __iomem *p0_base; + size_t p0_len; +}; + +struct msm_dp_desc { + phys_addr_t io_start; + unsigned int id; + bool wide_bus_supported; +}; + +static const struct msm_dp_desc msm_dp_desc_glymur[] = { + { .io_start = 0x0af54000, .id = MSM_DP_CONTROLLER_0, .wide_bus_supported = true }, + { .io_start = 0x0af5c000, .id = MSM_DP_CONTROLLER_1, .wide_bus_supported = true }, + { .io_start = 0x0af64000, .id = MSM_DP_CONTROLLER_2, .wide_bus_supported = true }, + { .io_start = 0x0af6c000, .id = MSM_DP_CONTROLLER_3, .wide_bus_supported = true }, + {} +}; + +static const struct msm_dp_desc msm_dp_desc_sa8775p[] = { + { .io_start = 0x0af54000, .id = MSM_DP_CONTROLLER_0, .wide_bus_supported = true }, + { .io_start = 0x0af5c000, .id = MSM_DP_CONTROLLER_1, .wide_bus_supported = true }, + { .io_start = 0x22154000, .id = MSM_DP_CONTROLLER_0, .wide_bus_supported = true }, + { .io_start = 0x2215c000, .id = MSM_DP_CONTROLLER_1, .wide_bus_supported = true }, + {} +}; + +static const struct msm_dp_desc msm_dp_desc_sdm845[] = { + { .io_start = 0x0ae90000, .id = MSM_DP_CONTROLLER_0 }, + {} +}; + +static const struct msm_dp_desc msm_dp_desc_sc7180[] = { + { .io_start = 0x0ae90000, .id = MSM_DP_CONTROLLER_0, .wide_bus_supported = true }, + {} +}; + +static const struct msm_dp_desc msm_dp_desc_sc7280[] = { + { .io_start = 0x0ae90000, .id = MSM_DP_CONTROLLER_0, .wide_bus_supported = true }, + { .io_start = 0x0aea0000, .id = MSM_DP_CONTROLLER_1, .wide_bus_supported = true }, + {} +}; + +static const struct msm_dp_desc msm_dp_desc_sc8180x[] = { + { .io_start = 0x0ae90000, .id = MSM_DP_CONTROLLER_0, .wide_bus_supported = true }, + { .io_start = 0x0ae98000, .id = MSM_DP_CONTROLLER_1, .wide_bus_supported = true }, + { .io_start = 0x0ae9a000, .id = MSM_DP_CONTROLLER_2, .wide_bus_supported = true }, + {} +}; + +static const struct msm_dp_desc msm_dp_desc_sc8280xp[] = { + { .io_start = 0x0ae90000, .id = MSM_DP_CONTROLLER_0, .wide_bus_supported = true }, + { .io_start = 0x0ae98000, .id = MSM_DP_CONTROLLER_1, .wide_bus_supported = true }, + { .io_start = 0x0ae9a000, .id = MSM_DP_CONTROLLER_2, .wide_bus_supported = true }, + { .io_start = 0x0aea0000, .id = MSM_DP_CONTROLLER_3, .wide_bus_supported = true }, + { .io_start = 0x22090000, .id = MSM_DP_CONTROLLER_0, .wide_bus_supported = true }, + { .io_start = 0x22098000, .id = MSM_DP_CONTROLLER_1, .wide_bus_supported = true }, + { .io_start = 0x2209a000, .id = MSM_DP_CONTROLLER_2, .wide_bus_supported = true }, + { .io_start = 0x220a0000, .id = MSM_DP_CONTROLLER_3, .wide_bus_supported = true }, + {} +}; + +static const struct msm_dp_desc msm_dp_desc_sm8650[] = { + { .io_start = 0x0af54000, .id = MSM_DP_CONTROLLER_0, .wide_bus_supported = true }, + {} +}; + +static const struct msm_dp_desc msm_dp_desc_x1e80100[] = { + { .io_start = 0x0ae90000, .id = MSM_DP_CONTROLLER_0, .wide_bus_supported = true }, + { .io_start = 0x0ae98000, .id = MSM_DP_CONTROLLER_1, .wide_bus_supported = true }, + { .io_start = 0x0ae9a000, .id = MSM_DP_CONTROLLER_2, .wide_bus_supported = true }, + { .io_start = 0x0aea0000, .id = MSM_DP_CONTROLLER_3, .wide_bus_supported = true }, + {} +}; + +static const struct of_device_id msm_dp_dt_match[] = { + { .compatible = "qcom,glymur-dp", .data = &msm_dp_desc_glymur }, + { .compatible = "qcom,sa8775p-dp", .data = &msm_dp_desc_sa8775p }, + { .compatible = "qcom,sc7180-dp", .data = &msm_dp_desc_sc7180 }, + { .compatible = "qcom,sc7280-dp", .data = &msm_dp_desc_sc7280 }, + { .compatible = "qcom,sc7280-edp", .data = &msm_dp_desc_sc7280 }, + { .compatible = "qcom,sc8180x-dp", .data = &msm_dp_desc_sc8180x }, + { .compatible = "qcom,sc8180x-edp", .data = &msm_dp_desc_sc8180x }, + { .compatible = "qcom,sc8280xp-dp", .data = &msm_dp_desc_sc8280xp }, + { .compatible = "qcom,sc8280xp-edp", .data = &msm_dp_desc_sc8280xp }, + { .compatible = "qcom,sdm845-dp", .data = &msm_dp_desc_sdm845 }, + { .compatible = "qcom,sm8350-dp", .data = &msm_dp_desc_sc7180 }, + { .compatible = "qcom,sm8650-dp", .data = &msm_dp_desc_sm8650 }, + { .compatible = "qcom,x1e80100-dp", .data = &msm_dp_desc_x1e80100 }, + {} +}; +MODULE_DEVICE_TABLE(of, msm_dp_dt_match); + +static struct msm_dp_display_private *dev_get_dp_display_private(struct device *dev) +{ + struct msm_dp *dp = dev_get_drvdata(dev); + + return container_of(dp, struct msm_dp_display_private, msm_dp_display); +} + +void msm_dp_display_signal_audio_start(struct msm_dp *msm_dp_display) +{ + struct msm_dp_display_private *dp; + + dp = container_of(msm_dp_display, struct msm_dp_display_private, msm_dp_display); + + reinit_completion(&dp->audio_comp); +} + +void msm_dp_display_signal_audio_complete(struct msm_dp *msm_dp_display) +{ + struct msm_dp_display_private *dp; + + dp = container_of(msm_dp_display, struct msm_dp_display_private, msm_dp_display); + + complete_all(&dp->audio_comp); +} + +static int msm_dp_display_bind(struct device *dev, struct device *master, + void *data) +{ + int rc = 0; + struct msm_dp_display_private *dp = dev_get_dp_display_private(dev); + struct msm_drm_private *priv = dev_get_drvdata(master); + struct drm_device *drm = priv->dev; + + dp->msm_dp_display.drm_dev = drm; + priv->kms->dp[dp->id] = &dp->msm_dp_display; + + dp->drm_dev = drm; + dp->aux->drm_dev = drm; + rc = msm_dp_aux_register(dp->aux); + if (rc) { + DRM_ERROR("DRM DP AUX register failed\n"); + goto end; + } + + return 0; +end: + return rc; +} + +static void msm_dp_display_unbind(struct device *dev, struct device *master, + void *data) +{ + struct msm_dp_display_private *dp = dev_get_dp_display_private(dev); + struct msm_drm_private *priv = dev_get_drvdata(master); + + of_dp_aux_depopulate_bus(dp->aux); + + msm_dp_aux_unregister(dp->aux); + dp->drm_dev = NULL; + dp->aux->drm_dev = NULL; + priv->kms->dp[dp->id] = NULL; +} + +static const struct component_ops msm_dp_display_comp_ops = { + .bind = msm_dp_display_bind, + .unbind = msm_dp_display_unbind, +}; + +static int msm_dp_display_lttpr_init(struct msm_dp_display_private *dp, u8 *dpcd) +{ + int rc, lttpr_count; + + if (drm_dp_read_lttpr_common_caps(dp->aux, dpcd, dp->link->lttpr_common_caps)) + return 0; + + lttpr_count = drm_dp_lttpr_count(dp->link->lttpr_common_caps); + rc = drm_dp_lttpr_init(dp->aux, lttpr_count); + if (rc) { + DRM_ERROR("failed to set LTTPRs transparency mode, rc=%d\n", rc); + return 0; + } + + return lttpr_count; +} + +static int msm_dp_display_process_hpd_high(struct msm_dp_display_private *dp) +{ + struct drm_connector *connector = dp->msm_dp_display.connector; + const struct drm_display_info *info = &connector->display_info; + int rc = 0; + u8 dpcd[DP_RECEIVER_CAP_SIZE]; + const struct drm_edid *drm_edid = NULL; + + rc = drm_dp_read_dpcd_caps(dp->aux, dpcd); + if (rc) + goto end; + + dp->link->lttpr_count = msm_dp_display_lttpr_init(dp, dpcd); + + rc = msm_dp_panel_read_link_caps(dp->panel, connector); + if (rc) + goto end; + + drm_edid = drm_edid_read_ddc(connector, &dp->aux->ddc); + drm_edid_connector_update(connector, drm_edid); + + if (!drm_edid) { + DRM_ERROR("panel edid read failed\n"); + /* check edid read fail is due to unplug */ + if (!msm_dp_aux_is_link_connected(dp->aux)) + return -ETIMEDOUT; + } + + msm_dp_link_process_request(dp->link); + + if (!dp->msm_dp_display.is_edp) + drm_dp_set_subconnector_property(connector, + connector_status_connected, + dp->panel->dpcd, + dp->panel->downstream_ports); + + dp->msm_dp_display.psr_supported = dp->panel->psr_cap.version && psr_enabled; + + dp->audio_supported = info->has_audio; + msm_dp_panel_handle_sink_request(dp->panel, drm_edid); + + /* + * set sink to normal operation mode -- D0 + * before dpcd read + */ + msm_dp_link_psm_config(dp->link, &dp->panel->link_info, false); + + msm_dp_link_reset_phy_params_vx_px(dp->link); + +end: + drm_edid_free(drm_edid); + return rc; +} + +/** + * msm_dp_display_host_phy_init() - start up DP PHY + * @dp: main display data structure + * + * Prepare DP PHY for the AUX transactions to succeed. + * + * Returns: true if this call has initliazed the PHY and false if the PHY has + * already been setup beforehand. + */ +static bool msm_dp_display_host_phy_init(struct msm_dp_display_private *dp) +{ + drm_dbg_dp(dp->drm_dev, "type=%d core_init=%d phy_init=%d\n", + dp->msm_dp_display.connector_type, dp->core_initialized, + dp->phy_initialized); + + if (!dp->phy_initialized) { + msm_dp_ctrl_phy_init(dp->ctrl); + dp->phy_initialized = true; + return true; + } + + return false; +} + +static void msm_dp_display_host_phy_exit(struct msm_dp_display_private *dp) +{ + drm_dbg_dp(dp->drm_dev, "type=%d core_init=%d phy_init=%d\n", + dp->msm_dp_display.connector_type, dp->core_initialized, + dp->phy_initialized); + + if (dp->phy_initialized) { + msm_dp_ctrl_phy_exit(dp->ctrl); + dp->phy_initialized = false; + } +} + +static void msm_dp_display_host_init(struct msm_dp_display_private *dp) +{ + drm_dbg_dp(dp->drm_dev, "type=%d core_init=%d phy_init=%d\n", + dp->msm_dp_display.connector_type, dp->core_initialized, + dp->phy_initialized); + + msm_dp_ctrl_core_clk_enable(dp->ctrl); + msm_dp_ctrl_reset(dp->ctrl, dp->panel); + msm_dp_ctrl_enable_irq(dp->ctrl); + msm_dp_aux_init(dp->aux); + dp->core_initialized = true; +} + +static void msm_dp_display_host_deinit(struct msm_dp_display_private *dp) +{ + drm_dbg_dp(dp->drm_dev, "type=%d core_init=%d phy_init=%d\n", + dp->msm_dp_display.connector_type, dp->core_initialized, + dp->phy_initialized); + + msm_dp_ctrl_reset(dp->ctrl, dp->panel); + msm_dp_ctrl_disable_irq(dp->ctrl); + msm_dp_aux_deinit(dp->aux); + msm_dp_ctrl_core_clk_disable(dp->ctrl); + dp->core_initialized = false; +} + +static void msm_dp_display_handle_video_request(struct msm_dp_display_private *dp) +{ + if (dp->link->sink_request & DP_TEST_LINK_VIDEO_PATTERN) { + dp->panel->video_test = true; + msm_dp_link_send_test_response(dp->link); + } +} + +static int msm_dp_display_handle_irq_hpd(struct msm_dp_display_private *dp) +{ + u32 sink_request = dp->link->sink_request; + + drm_dbg_dp(dp->drm_dev, "%d\n", sink_request); + + msm_dp_ctrl_handle_sink_request(dp->ctrl, dp->panel); + + if (sink_request & DP_TEST_LINK_VIDEO_PATTERN) + msm_dp_display_handle_video_request(dp); + + return 0; +} + +static int msm_dp_hpd_plug_handle(struct msm_dp_display_private *dp) +{ + int ret; + struct platform_device *pdev = dp->msm_dp_display.pdev; + + drm_dbg_dp(dp->drm_dev, "Before, type=%d sink_count=%d\n", + dp->msm_dp_display.connector_type, + dp->link->sink_count); + + guard(mutex)(&dp->plugged_lock); + + ret = pm_runtime_resume_and_get(&pdev->dev); + if (ret) { + DRM_ERROR("failed to pm_runtime_resume\n"); + return ret; + } + + msm_dp_aux_enable_xfers(dp->aux, true); + + msm_dp_display_host_phy_init(dp); + + ret = msm_dp_display_process_hpd_high(dp); + + drm_dbg_dp(dp->drm_dev, "After, type=%d sink_count=%d\n", + dp->msm_dp_display.connector_type, + dp->link->sink_count); + + dp->plugged = true; + + return ret; +}; + +static void msm_dp_display_handle_plugged_change(struct msm_dp *msm_dp_display, + bool plugged) +{ + struct msm_dp_display_private *dp; + + dp = container_of(msm_dp_display, + struct msm_dp_display_private, msm_dp_display); + + /* notify audio subsystem only if sink supports audio */ + if (dp->audio_supported) + drm_connector_hdmi_audio_plugged_notify(msm_dp_display->connector, + plugged); +} + +static int msm_dp_hpd_unplug_handle(struct msm_dp_display_private *dp) +{ + struct platform_device *pdev = dp->msm_dp_display.pdev; + + dp->panel->video_test = false; + + msm_dp_aux_enable_xfers(dp->aux, false); + + drm_dbg_dp(dp->drm_dev, "Before, type=%d sink_count=%d\n", + dp->msm_dp_display.connector_type, + dp->link->sink_count); + + guard(mutex)(&dp->plugged_lock); + if (!dp->plugged) + return 0; + + /* Don't forget modes for eDP */ + if (!dp->msm_dp_display.is_edp) + drm_edid_connector_update(dp->msm_dp_display.connector, NULL); + + /* triggered by irq_hdp with sink_count = 0 */ + if (dp->link->sink_count == 0) + msm_dp_display_host_phy_exit(dp); + + /* + * We don't need separate work for disconnect as + * connect/attention interrupts are disabled + */ + if (!dp->msm_dp_display.is_edp) + drm_dp_set_subconnector_property(dp->msm_dp_display.connector, + connector_status_disconnected, + dp->panel->dpcd, + dp->panel->downstream_ports); + + /* signal the disconnect event early to ensure proper teardown */ + msm_dp_display_handle_plugged_change(&dp->msm_dp_display, false); + + drm_dbg_dp(dp->drm_dev, "After, type=%d, sink_count=%d\n", + dp->msm_dp_display.connector_type, + dp->link->sink_count); + + if (dp->plugged) { + pm_runtime_put_sync(&pdev->dev); + dp->plugged = false; + } + + return 0; +} + +static int msm_dp_irq_hpd_handle(struct msm_dp_display_private *dp) +{ + u32 sink_request; + int rc = 0; + + /* irq_hpd can happen at either connected or disconnected state */ + drm_dbg_dp(dp->drm_dev, "Before, type=%d, sink_count=%d\n", + dp->msm_dp_display.connector_type, + dp->link->sink_count); + + /* check for any test request issued by sink */ + rc = msm_dp_link_process_request(dp->link); + if (!rc) { + sink_request = dp->link->sink_request; + drm_dbg_dp(dp->drm_dev, "sink_request=%d\n", sink_request); + if (sink_request & DS_PORT_STATUS_CHANGED) + rc = msm_dp_display_process_hpd_high(dp); + else + rc = msm_dp_display_handle_irq_hpd(dp); + } + + drm_dbg_dp(dp->drm_dev, "After, type=%d, sink_count=%d\n", + dp->msm_dp_display.connector_type, + dp->link->sink_count); + + return rc; +} + +static void msm_dp_display_deinit_sub_modules(struct msm_dp_display_private *dp) +{ + msm_dp_audio_put(dp->audio); + msm_dp_aux_put(dp->aux); +} + +static int msm_dp_init_sub_modules(struct msm_dp_display_private *dp) +{ + int rc = 0; + struct device *dev = &dp->msm_dp_display.pdev->dev; + struct phy *phy; + + phy = devm_phy_get(dev, "dp"); + if (IS_ERR(phy)) + return PTR_ERR(phy); + + rc = phy_set_mode_ext(phy, PHY_MODE_DP, + dp->msm_dp_display.is_edp ? PHY_SUBMODE_EDP : PHY_SUBMODE_DP); + if (rc) { + DRM_ERROR("failed to set phy submode, rc = %d\n", rc); + goto error; + } + + dp->aux = msm_dp_aux_get(dev, phy, dp->msm_dp_display.is_edp, dp->aux_base); + if (IS_ERR(dp->aux)) { + rc = PTR_ERR(dp->aux); + DRM_ERROR("failed to initialize aux, rc = %d\n", rc); + dp->aux = NULL; + goto error; + } + + dp->link = msm_dp_link_get(dev, dp->aux); + if (IS_ERR(dp->link)) { + rc = PTR_ERR(dp->link); + DRM_ERROR("failed to initialize link, rc = %d\n", rc); + dp->link = NULL; + goto error_link; + } + + dp->panel = msm_dp_panel_get(dev, dp->aux, dp->link, dp->link_base, dp->p0_base); + if (IS_ERR(dp->panel)) { + rc = PTR_ERR(dp->panel); + DRM_ERROR("failed to initialize panel, rc = %d\n", rc); + dp->panel = NULL; + goto error_link; + } + + dp->ctrl = msm_dp_ctrl_get(dev, dp->link, dp->aux, + phy, dp->ahb_base, dp->link_base); + if (IS_ERR(dp->ctrl)) { + rc = PTR_ERR(dp->ctrl); + DRM_ERROR("failed to initialize ctrl, rc = %d\n", rc); + dp->ctrl = NULL; + goto error_link; + } + + dp->audio = msm_dp_audio_get(dp->msm_dp_display.pdev, dp->link_base); + if (IS_ERR(dp->audio)) { + rc = PTR_ERR(dp->audio); + pr_err("failed to initialize audio, rc = %d\n", rc); + dp->audio = NULL; + goto error_link; + } + + return rc; + +error_link: + msm_dp_aux_put(dp->aux); +error: + return rc; +} + +static int msm_dp_display_set_mode(struct msm_dp *msm_dp_display, + const struct drm_display_mode *adjusted_mode, + struct msm_dp_panel *msm_dp_panel) +{ + struct msm_dp_display_private *dp; + u32 bpp; + + dp = container_of(msm_dp_display, struct msm_dp_display_private, msm_dp_display); + + if (msm_dp_display_check_video_test(msm_dp_display)) + bpp = msm_dp_display_get_test_bpp(msm_dp_display); + else + bpp = msm_dp_panel->connector->display_info.bpc * 3; + + msm_dp_panel_init_panel_info(msm_dp_panel, adjusted_mode, bpp ? bpp : 24); + + /* populate wide_bus_support to different layers */ + dp->ctrl->wide_bus_en = + msm_dp_panel->msm_dp_mode.out_fmt_is_yuv_420 ? false : dp->wide_bus_supported; + return 0; +} + +static int msm_dp_display_prepare_link(struct msm_dp_display_private *dp) +{ + struct msm_dp *msm_dp_display = &dp->msm_dp_display; + int rc = 0; + bool force_link_train = false; + + drm_dbg_dp(dp->drm_dev, "sink_count=%d\n", dp->link->sink_count); + + if (msm_dp_display->is_edp) + msm_dp_hpd_plug_handle(dp); + + rc = pm_runtime_resume_and_get(&msm_dp_display->pdev->dev); + if (rc) { + DRM_ERROR("failed to pm_runtime_resume\n"); + return rc; + } + + if (dp->link->sink_count == 0) + return -ENOTCONN; + + if (!msm_dp_display->power_on) { + msm_dp_display_host_phy_init(dp); + force_link_train = true; + } + + rc = msm_dp_ctrl_on_link(dp->ctrl, dp->panel); + if (rc) { + DRM_ERROR("Failed link training (rc=%d)\n", rc); + // TODO: schedule drm_connector_set_link_status_property() + return rc; + } + + return msm_dp_ctrl_prepare_stream_on(dp->ctrl, dp->panel, force_link_train); +} + +static int msm_dp_display_enable(struct msm_dp_display_private *dp, + struct msm_dp_panel *msm_dp_panel) +{ + int rc = 0; + struct msm_dp *msm_dp_display = &dp->msm_dp_display; + + drm_dbg_dp(dp->drm_dev, "sink_count=%d\n", dp->link->sink_count); + if (msm_dp_display->power_on) { + drm_dbg_dp(dp->drm_dev, "Link already setup, return\n"); + return 0; + } + + rc = msm_dp_ctrl_on_stream(dp->ctrl, msm_dp_panel); + if (!rc) + msm_dp_display->power_on = true; + + return rc; +} + +static int msm_dp_display_post_enable(struct msm_dp *msm_dp_display) +{ + struct msm_dp_display_private *dp; + u32 rate; + + dp = container_of(msm_dp_display, struct msm_dp_display_private, msm_dp_display); + + rate = dp->link->link_params.rate; + + if (dp->audio_supported) { + dp->audio->bw_code = drm_dp_link_rate_to_bw_code(rate); + dp->audio->lane_count = dp->link->link_params.num_lanes; + } + + /* signal the connect event late to synchronize video and display */ + msm_dp_display_handle_plugged_change(msm_dp_display, true); + + if (msm_dp_display->psr_supported) + msm_dp_ctrl_config_psr(dp->ctrl, dp->panel); + + return 0; +} + +static void msm_dp_display_audio_notify_disable(struct msm_dp_display_private *dp) +{ + struct msm_dp *msm_dp_display = &dp->msm_dp_display; + + /* wait only if audio was enabled */ + if (msm_dp_display->audio_enabled) { + /* signal the disconnect event */ + msm_dp_display_handle_plugged_change(msm_dp_display, false); + if (!wait_for_completion_timeout(&dp->audio_comp, + HZ * 5)) + DRM_ERROR("audio comp timeout\n"); + } + + msm_dp_display->audio_enabled = false; +} + +static int msm_dp_display_disable(struct msm_dp_display_private *dp, + struct msm_dp_panel *msm_dp_panel) +{ + struct msm_dp *msm_dp_display = &dp->msm_dp_display; + + if (!msm_dp_display->power_on) + return 0; + + msm_dp_panel_disable_vsc_sdp(msm_dp_panel); + + msm_dp_ctrl_off_pixel_clk(dp->ctrl); + + /* dongle is still connected but sinks are disconnected */ + if (dp->link->sink_count == 0) + msm_dp_link_psm_config(dp->link, &msm_dp_panel->link_info, true); + + msm_dp_ctrl_off_link(dp->ctrl, msm_dp_panel); + + if (dp->link->sink_count == 0) + /* re-init the PHY so that we can listen to Dongle disconnect */ + msm_dp_ctrl_reinit_phy(dp->ctrl); + else + msm_dp_display_host_phy_exit(dp); + + msm_dp_display->power_on = false; + + drm_dbg_dp(dp->drm_dev, "sink count: %d\n", dp->link->sink_count); + return 0; +} + +/** + * msm_dp_display_mode_valid - callback to determine if specified mode is valid + * @dp: Pointer to dp display structure + * @info: display info + * @mode: Pointer to drm mode structure + * Returns: Validity status for specified mode + */ +enum drm_mode_status msm_dp_display_mode_valid(struct msm_dp *dp, + const struct drm_display_info *info, + const struct drm_display_mode *mode) +{ + const u32 num_components = 3, default_bpp = 24; + struct msm_dp_display_private *msm_dp_display; + struct msm_dp_link_info *link_info; + u32 mode_rate_khz = 0, supported_rate_khz = 0, mode_bpp = 0; + int mode_pclk_khz = mode->clock; + int link_pclk_khz; + bool is_yuv_420; + + if (!dp || !mode_pclk_khz || !dp->connector) { + DRM_ERROR("invalid params\n"); + return -EINVAL; + } + + msm_dp_display = container_of(dp, struct msm_dp_display_private, msm_dp_display); + link_info = &msm_dp_display->panel->link_info; + + is_yuv_420 = drm_mode_is_420_only(&dp->connector->display_info, mode); + + /* + * YUV 420 is carried over DP by signalling the colorimetry through a + * VSC SDP, so a 420-only mode cannot be driven without VSC SDP support. + */ + if (is_yuv_420 && !msm_dp_display->panel->vsc_sdp_supported) + return MODE_NO_420; + + link_pclk_khz = is_yuv_420 ? mode_pclk_khz / 2 : mode_pclk_khz; + + if (is_yuv_420 || msm_dp_display->wide_bus_supported) + mode_pclk_khz /= 2; + + if (mode_pclk_khz > DP_MAX_PIXEL_CLK_KHZ) + return MODE_CLOCK_HIGH; + + mode_bpp = dp->connector->display_info.bpc * num_components; + if (!mode_bpp) + mode_bpp = default_bpp; + + mode_bpp = msm_dp_panel_get_mode_bpp(msm_dp_display->panel, + mode_bpp, link_pclk_khz); + + mode_rate_khz = link_pclk_khz * mode_bpp; + supported_rate_khz = link_info->num_lanes * link_info->rate * 8; + + if (mode_rate_khz > supported_rate_khz) + return MODE_BAD; + + return MODE_OK; +} + +int msm_dp_display_get_modes(struct msm_dp *dp) +{ + struct msm_dp_display_private *msm_dp_display; + + if (!dp) { + DRM_ERROR("invalid params\n"); + return 0; + } + + msm_dp_display = container_of(dp, struct msm_dp_display_private, msm_dp_display); + + return drm_edid_connector_add_modes(msm_dp_display->panel->connector); +} + +bool msm_dp_display_check_video_test(struct msm_dp *dp) +{ + struct msm_dp_display_private *msm_dp_display; + + msm_dp_display = container_of(dp, struct msm_dp_display_private, msm_dp_display); + + return msm_dp_display->panel->video_test; +} + +int msm_dp_display_get_test_bpp(struct msm_dp *dp) +{ + struct msm_dp_display_private *msm_dp_display; + + if (!dp) { + DRM_ERROR("invalid params\n"); + return 0; + } + + msm_dp_display = container_of(dp, struct msm_dp_display_private, msm_dp_display); + + return msm_dp_link_bit_depth_to_bpp( + msm_dp_display->link->test_video.test_bit_depth); +} + +void msm_dp_snapshot(struct msm_disp_state *disp_state, struct msm_dp *dp) +{ + struct msm_dp_display_private *msm_dp_display; + + msm_dp_display = container_of(dp, struct msm_dp_display_private, msm_dp_display); + + /* + * if we are reading registers we need the link clocks to be on + * however till DP cable is connected this will not happen as we + * do not know the resolution to power up with. Hence check the + * power_on status before dumping DP registers to avoid crash due + * to unclocked access + */ + if (!dp->power_on) + return; + + msm_disp_snapshot_add_block(disp_state, msm_dp_display->ahb_len, + msm_dp_display->ahb_base, "dp_ahb"); + msm_disp_snapshot_add_block(disp_state, msm_dp_display->aux_len, + msm_dp_display->aux_base, "dp_aux"); + msm_disp_snapshot_add_block(disp_state, msm_dp_display->link_len, + msm_dp_display->link_base, "dp_link"); + msm_disp_snapshot_add_block(disp_state, msm_dp_display->p0_len, + msm_dp_display->p0_base, "dp_p0"); +} + +void msm_dp_display_set_psr(struct msm_dp *msm_dp_display, bool enter) +{ + struct msm_dp_display_private *dp; + + if (!msm_dp_display) { + DRM_ERROR("invalid params\n"); + return; + } + + dp = container_of(msm_dp_display, struct msm_dp_display_private, msm_dp_display); + msm_dp_ctrl_set_psr(dp->ctrl, dp->panel, enter); +} + +/** + * msm_dp_bridge_detect - callback to determine if connector is connected + * + * @bridge: Pointer to drm bridge structure + * @connector: Pointer to drm connector structure + * + * Returns: where there is a display connected to the DPTX (returning + * disconnected for branch devices without DP Sinks being connected). + */ +enum drm_connector_status msm_dp_bridge_detect(struct drm_bridge *bridge, + struct drm_connector *connector) +{ + struct msm_dp_bridge *msm_dp_bridge = to_dp_bridge(bridge); + struct msm_dp *dp = msm_dp_bridge->msm_dp_display; + int status = connector_status_disconnected; + struct msm_dp_display_private *priv; + u8 dpcd[DP_RECEIVER_CAP_SIZE]; + struct drm_dp_desc desc; + bool phy_deinit; + int ret; + + dp = to_dp_bridge(bridge)->msm_dp_display; + + priv = container_of(dp, struct msm_dp_display_private, msm_dp_display); + + guard(mutex)(&priv->plugged_lock); + ret = pm_runtime_resume_and_get(&dp->pdev->dev); + if (ret) { + DRM_ERROR("failed to pm_runtime_resume\n"); + return status; + } + + phy_deinit = msm_dp_display_host_phy_init(priv); + + msm_dp_aux_enable_xfers(priv->aux, true); + + ret = msm_dp_aux_is_link_connected(priv->aux); + DRM_DEBUG_DP("aux link status: %x\n", ret); + if (!priv->plugged && !ret) { + DRM_DEBUG_DP("aux not connected\n"); + priv->plugged = false; + goto end; + } + + ret = drm_dp_read_dpcd_caps(priv->aux, dpcd); + if (ret) { + DRM_DEBUG_DP("failed to read caps\n"); + priv->plugged = false; + goto end; + } + + ret = drm_dp_read_desc(priv->aux, &desc, drm_dp_is_branch(dpcd)); + if (ret) { + DRM_DEBUG_DP("failed to read desc\n"); + priv->plugged = false; + goto end; + } + + status = connector_status_connected; + priv->plugged = true; + + if (drm_dp_read_sink_count_cap(connector, dpcd, &desc)) { + int sink_count = drm_dp_read_sink_count(priv->aux); + + drm_dbg_dp(dp->drm_dev, "sink_count = %d\n", sink_count); + + if (sink_count <= 0) + status = connector_status_disconnected; + } + +end: + /* + * If we detected the DPRX, leave the controller on so that it doesn't + * lose the state. + */ + if (!priv->plugged) { + if (phy_deinit) { + msm_dp_aux_enable_xfers(priv->aux, false); + msm_dp_display_host_phy_exit(priv); + } + + pm_runtime_put_sync(&dp->pdev->dev); + } + + return status; +} + +static irqreturn_t msm_dp_display_irq_handler(int irq, void *dev_id) +{ + struct msm_dp_display_private *dp = dev_id; + u32 hpd_isr_status; + unsigned long flags; + irqreturn_t ret = IRQ_HANDLED; + + hpd_isr_status = msm_dp_aux_get_hpd_intr_status(dp->aux); + + if (hpd_isr_status & 0x0F) { + drm_dbg_dp(dp->drm_dev, "type=%d isr=0x%x\n", + dp->msm_dp_display.connector_type, hpd_isr_status); + + spin_lock_irqsave(&dp->irq_thread_lock, flags); + dp->hpd_isr_status |= hpd_isr_status; + ret = IRQ_WAKE_THREAD; + spin_unlock_irqrestore(&dp->irq_thread_lock, flags); + } + + /* DP controller isr */ + ret |= msm_dp_ctrl_isr(dp->ctrl, dp->panel); + + return ret; +} + +static irqreturn_t msm_dp_display_irq_thread(int irq, void *dev_id) +{ + struct msm_dp_display_private *dp = dev_id; + irqreturn_t ret = IRQ_NONE; + unsigned long flags; + u32 hpd_isr_status; + + spin_lock_irqsave(&dp->irq_thread_lock, flags); + hpd_isr_status = dp->hpd_isr_status; + dp->hpd_isr_status = 0; + spin_unlock_irqrestore(&dp->irq_thread_lock, flags); + + if (hpd_isr_status & DP_DP_HPD_UNPLUG_INT_MASK) + drm_bridge_hpd_notify(dp->msm_dp_display.bridge, + connector_status_disconnected); + + if (hpd_isr_status & DP_DP_HPD_PLUG_INT_MASK) + drm_bridge_hpd_notify(dp->msm_dp_display.bridge, + connector_status_connected); + + /* Send HPD as connected and distinguish it in the notifier */ + if (hpd_isr_status & DP_DP_IRQ_HPD_INT_MASK) + drm_bridge_hpd_notify(dp->msm_dp_display.bridge, + connector_status_connected); + + ret = IRQ_HANDLED; + + return ret; +} + +static int msm_dp_display_request_irq(struct msm_dp_display_private *dp) +{ + int rc = 0; + struct platform_device *pdev = dp->msm_dp_display.pdev; + + dp->irq = platform_get_irq(pdev, 0); + if (dp->irq < 0) { + DRM_ERROR("failed to get irq\n"); + return dp->irq; + } + + spin_lock_init(&dp->irq_thread_lock); + irq_set_status_flags(dp->irq, IRQ_NOAUTOEN); + rc = devm_request_threaded_irq(&pdev->dev, dp->irq, + msm_dp_display_irq_handler, + msm_dp_display_irq_thread, + IRQ_TYPE_LEVEL_HIGH, + "dp_display_isr", dp); + + if (rc < 0) { + DRM_ERROR("failed to request IRQ%u: %d\n", + dp->irq, rc); + return rc; + } + + return 0; +} + +static const struct msm_dp_desc *msm_dp_display_get_desc(struct platform_device *pdev) +{ + const struct msm_dp_desc *descs = of_device_get_match_data(&pdev->dev); + struct resource *res; + int i; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!res) + return NULL; + + for (i = 0; i < descs[i].io_start; i++) { + if (descs[i].io_start == res->start) + return &descs[i]; + } + + dev_err(&pdev->dev, "unknown displayport instance\n"); + return NULL; +} + +static int msm_dp_display_probe_tail(struct device *dev) +{ + struct msm_dp *dp = dev_get_drvdata(dev); + int ret; + + /* + * External bridges are mandatory for eDP interfaces: one has to + * provide at least an eDP panel (which gets wrapped into panel-bridge). + * + * For DisplayPort interfaces external bridges are optional, so + * silently ignore an error if one is not present (-ENODEV). + */ + dp->next_bridge = devm_drm_of_get_bridge(&dp->pdev->dev, dp->pdev->dev.of_node, 1, 0); + if (IS_ERR(dp->next_bridge)) { + ret = PTR_ERR(dp->next_bridge); + dp->next_bridge = NULL; + if (dp->is_edp || ret != -ENODEV) + return ret; + } + + ret = component_add(dev, &msm_dp_display_comp_ops); + if (ret) + DRM_ERROR("component add failed, rc=%d\n", ret); + + return ret; +} + +static int msm_dp_auxbus_done_probe(struct drm_dp_aux *aux) +{ + return msm_dp_display_probe_tail(aux->dev); +} + +static int msm_dp_display_get_connector_type(struct platform_device *pdev, + const struct msm_dp_desc *desc) +{ + struct device_node *node = pdev->dev.of_node; + struct device_node *aux_bus = of_get_child_by_name(node, "aux-bus"); + struct device_node *panel = of_get_child_by_name(aux_bus, "panel"); + int connector_type; + + if (panel) + connector_type = DRM_MODE_CONNECTOR_eDP; + else + connector_type = DRM_MODE_SUBCONNECTOR_DisplayPort; + + of_node_put(panel); + of_node_put(aux_bus); + + return connector_type; +} + +static void __iomem *msm_dp_ioremap(struct platform_device *pdev, int idx, size_t *len) +{ + struct resource *res; + void __iomem *base; + + base = devm_platform_get_and_ioremap_resource(pdev, idx, &res); + if (!IS_ERR(base)) + *len = resource_size(res); + + return base; +} + +#define DP_DEFAULT_AHB_OFFSET 0x0000 +#define DP_DEFAULT_AHB_SIZE 0x0200 +#define DP_DEFAULT_AUX_OFFSET 0x0200 +#define DP_DEFAULT_AUX_SIZE 0x0200 +#define DP_DEFAULT_LINK_OFFSET 0x0400 +#define DP_DEFAULT_LINK_SIZE 0x0C00 +#define DP_DEFAULT_P0_OFFSET 0x1000 +#define DP_DEFAULT_P0_SIZE 0x0400 + +static int msm_dp_display_get_io(struct msm_dp_display_private *display) +{ + struct platform_device *pdev = display->msm_dp_display.pdev; + + display->ahb_base = msm_dp_ioremap(pdev, 0, &display->ahb_len); + if (IS_ERR(display->ahb_base)) + return PTR_ERR(display->ahb_base); + + display->aux_base = msm_dp_ioremap(pdev, 1, &display->aux_len); + if (IS_ERR(display->aux_base)) { + if (display->aux_base != ERR_PTR(-EINVAL)) { + DRM_ERROR("unable to remap aux region: %pe\n", display->aux_base); + return PTR_ERR(display->aux_base); + } + + /* + * The initial binding had a single reg, but in order to + * support variation in the sub-region sizes this was split. + * msm_dp_ioremap() will fail with -EINVAL here if only a single + * reg is specified, so fill in the sub-region offsets and + * lengths based on this single region. + */ + if (display->ahb_len < DP_DEFAULT_P0_OFFSET + DP_DEFAULT_P0_SIZE) { + DRM_ERROR("legacy memory region not large enough\n"); + return -EINVAL; + } + + display->ahb_len = DP_DEFAULT_AHB_SIZE; + display->aux_base = display->ahb_base + DP_DEFAULT_AUX_OFFSET; + display->aux_len = DP_DEFAULT_AUX_SIZE; + display->link_base = display->ahb_base + DP_DEFAULT_LINK_OFFSET; + display->link_len = DP_DEFAULT_LINK_SIZE; + display->p0_base = display->ahb_base + DP_DEFAULT_P0_OFFSET; + display->p0_len = DP_DEFAULT_P0_SIZE; + + return 0; + } + + display->link_base = msm_dp_ioremap(pdev, 2, &display->link_len); + if (IS_ERR(display->link_base)) { + DRM_ERROR("unable to remap link region: %pe\n", display->link_base); + return PTR_ERR(display->link_base); + } + + display->p0_base = msm_dp_ioremap(pdev, 3, &display->p0_len); + if (IS_ERR(display->p0_base)) { + DRM_ERROR("unable to remap p0 region: %pe\n", display->p0_base); + return PTR_ERR(display->p0_base); + } + + return 0; +} + +static int msm_dp_display_probe(struct platform_device *pdev) +{ + int rc = 0; + struct msm_dp_display_private *dp; + const struct msm_dp_desc *desc; + + if (!pdev || !pdev->dev.of_node) { + DRM_ERROR("pdev not found\n"); + return -ENODEV; + } + + dp = devm_kzalloc(&pdev->dev, sizeof(*dp), GFP_KERNEL); + if (!dp) + return -ENOMEM; + + desc = msm_dp_display_get_desc(pdev); + if (!desc) + return -EINVAL; + + dp->msm_dp_display.pdev = pdev; + dp->id = desc->id; + dp->msm_dp_display.connector_type = msm_dp_display_get_connector_type(pdev, desc); + dp->wide_bus_supported = desc->wide_bus_supported; + dp->msm_dp_display.is_edp = + (dp->msm_dp_display.connector_type == DRM_MODE_CONNECTOR_eDP); + dp->hpd_isr_status = 0; + + mutex_init(&dp->plugged_lock); + + rc = msm_dp_display_get_io(dp); + if (rc) + return rc; + + rc = msm_dp_init_sub_modules(dp); + if (rc) { + DRM_ERROR("init sub module failed\n"); + return -EPROBE_DEFER; + } + + /* Store DP audio handle inside DP display */ + dp->msm_dp_display.msm_dp_audio = dp->audio; + + init_completion(&dp->audio_comp); + + platform_set_drvdata(pdev, &dp->msm_dp_display); + + rc = devm_pm_runtime_enable(&pdev->dev); + if (rc) + goto err; + + rc = msm_dp_display_request_irq(dp); + if (rc) + goto err; + + if (dp->msm_dp_display.is_edp) { + rc = devm_of_dp_aux_populate_bus(dp->aux, msm_dp_auxbus_done_probe); + if (rc) { + DRM_ERROR("eDP auxbus population failed, rc=%d\n", rc); + goto err; + } + } else { + rc = msm_dp_display_probe_tail(&pdev->dev); + if (rc) + goto err; + } + + return rc; + +err: + msm_dp_display_deinit_sub_modules(dp); + return rc; +} + +static void msm_dp_display_remove(struct platform_device *pdev) +{ + struct msm_dp_display_private *dp = dev_get_dp_display_private(&pdev->dev); + + component_del(&pdev->dev, &msm_dp_display_comp_ops); + msm_dp_display_deinit_sub_modules(dp); + platform_set_drvdata(pdev, NULL); +} + +static int msm_dp_pm_runtime_suspend(struct device *dev) +{ + struct msm_dp_display_private *dp = dev_get_dp_display_private(dev); + + disable_irq(dp->irq); + + if (dp->msm_dp_display.is_edp) { + msm_dp_display_host_phy_exit(dp); + msm_dp_aux_hpd_disable(dp->aux); + } + msm_dp_display_host_deinit(dp); + + return 0; +} + +static int msm_dp_pm_runtime_resume(struct device *dev) +{ + struct msm_dp_display_private *dp = dev_get_dp_display_private(dev); + + /* + * for eDP, host cotroller, HPD block and PHY are enabled here + * but with HPD irq disabled + * + * for DP, only host controller is enabled here. + * HPD block is enabled at msm_dp_bridge_hpd_enable() + * PHY will be enabled at plugin handler later + */ + msm_dp_display_host_init(dp); + if (dp->msm_dp_display.is_edp) { + msm_dp_aux_hpd_enable(dp->aux); + msm_dp_display_host_phy_init(dp); + } + + enable_irq(dp->irq); + return 0; +} + +static const struct dev_pm_ops msm_dp_pm_ops = { + SET_RUNTIME_PM_OPS(msm_dp_pm_runtime_suspend, msm_dp_pm_runtime_resume, NULL) + SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, + pm_runtime_force_resume) +}; + +static struct platform_driver msm_dp_display_driver = { + .probe = msm_dp_display_probe, + .remove = msm_dp_display_remove, + .driver = { + .name = "msm-dp-display", + .of_match_table = msm_dp_dt_match, + .suppress_bind_attrs = true, + .pm = &msm_dp_pm_ops, + }, +}; + +int __init msm_dp_register(void) +{ + int ret; + + ret = platform_driver_register(&msm_dp_display_driver); + if (ret) + DRM_ERROR("Dp display driver register failed"); + + return ret; +} + +void __exit msm_dp_unregister(void) +{ + platform_driver_unregister(&msm_dp_display_driver); +} + +bool msm_dp_needs_periph_flush(const struct msm_dp *msm_dp_display, + const struct drm_display_mode *mode) +{ + return drm_mode_is_420_only(&msm_dp_display->connector->display_info, mode); +} + +bool msm_dp_wide_bus_available(const struct msm_dp *msm_dp_display) +{ + struct msm_dp_display_private *dp; + + dp = container_of(msm_dp_display, struct msm_dp_display_private, msm_dp_display); + + if (dp->panel->msm_dp_mode.out_fmt_is_yuv_420) + return false; + + return dp->wide_bus_supported; +} + +void msm_dp_display_debugfs_init(struct msm_dp *msm_dp_display, struct dentry *root, bool is_edp) +{ + struct msm_dp_display_private *dp; + struct device *dev; + int rc; + + dp = container_of(msm_dp_display, struct msm_dp_display_private, msm_dp_display); + dev = &dp->msm_dp_display.pdev->dev; + + rc = msm_dp_debug_init(dev, dp->panel, dp->link, dp->msm_dp_display.connector, root, is_edp); + if (rc) + DRM_ERROR("failed to initialize debug, rc = %d\n", rc); +} + +int msm_dp_modeset_init(struct msm_dp *msm_dp_display, struct drm_device *dev, + struct drm_encoder *encoder, bool yuv_supported) +{ + struct msm_dp_display_private *msm_dp_priv; + int ret; + + msm_dp_display->drm_dev = dev; + + msm_dp_priv = container_of(msm_dp_display, struct msm_dp_display_private, msm_dp_display); + + ret = msm_dp_bridge_init(msm_dp_display, dev, encoder, yuv_supported); + if (ret) { + DRM_DEV_ERROR(dev->dev, + "failed to create dp bridge: %d\n", ret); + return ret; + } + + msm_dp_display->connector = msm_dp_drm_connector_init(msm_dp_display, encoder); + if (IS_ERR(msm_dp_display->connector)) { + ret = PTR_ERR(msm_dp_display->connector); + DRM_DEV_ERROR(dev->dev, + "failed to create dp connector: %d\n", ret); + msm_dp_display->connector = NULL; + return ret; + } + + msm_dp_priv->panel->connector = msm_dp_display->connector; + + return 0; +} + +void msm_dp_display_atomic_pre_enable(struct msm_dp *msm_dp_display, + struct drm_atomic_commit *state) +{ + struct msm_dp_display_private *dp; + struct drm_crtc *crtc; + struct drm_crtc_state *crtc_state; + + dp = container_of(msm_dp_display, struct msm_dp_display_private, msm_dp_display); + + crtc = drm_atomic_get_new_crtc_for_encoder(state, msm_dp_display->bridge->encoder); + if (!crtc) + return; + crtc_state = drm_atomic_get_new_crtc_state(state, crtc); + + /* + * The DPU encoder's .atomic_enable() reads the mode's YUV420 / wide bus + * state and runs before the bridge's .atomic_enable(), so the mode must + * be programmed here, in .atomic_pre_enable(). + */ + msm_dp_display_set_mode(msm_dp_display, &crtc_state->adjusted_mode, dp->panel); +} + +void msm_dp_display_atomic_enable(struct msm_dp *msm_dp_display, + struct drm_atomic_commit *state) +{ + int rc = 0; + struct msm_dp_display_private *dp; + + dp = container_of(msm_dp_display, struct msm_dp_display_private, msm_dp_display); + + rc = msm_dp_display_prepare_link(dp); + if (rc) { + DRM_ERROR("DP display prepare failed, rc=%d\n", rc); + return; + } + + rc = msm_dp_display_enable(dp, dp->panel); + if (rc) + DRM_ERROR("DP display enable failed, rc=%d\n", rc); + + rc = msm_dp_display_post_enable(msm_dp_display); + if (rc) { + DRM_ERROR("DP display post enable failed, rc=%d\n", rc); + msm_dp_display_disable(dp, dp->panel); + } + + drm_dbg_dp(msm_dp_display->drm_dev, "type=%d Done\n", msm_dp_display->connector_type); +} + +void msm_dp_display_atomic_disable(struct msm_dp *dp) +{ + struct msm_dp_display_private *msm_dp_display; + + msm_dp_display = container_of(dp, struct msm_dp_display_private, msm_dp_display); + + /* + * If .atomic_enable() bailed out - link training failure is the common + * case - the mainlink was never brought up and ->power_on stayed false. + * Driving the PUSH_IDLE pattern into a controller that was never + * enabled times out, and .atomic_post_disable() then drops the + * controller's runtime-PM reference without tearing the PHY back down, + * because msm_dp_display_disable() returns early on !power_on. On + * glymur (Snapdragon X2 Elite) that combination is answered by a + * TrustZone-level SOCCP/ADSP force-stop and a silent SoC reset. + * There is nothing to push idle, so leave it alone. + */ + if (!dp->power_on) + return; + + msm_dp_ctrl_push_idle(msm_dp_display->ctrl); +} + +static void msm_dp_display_unprepare(struct msm_dp_display_private *dp) +{ + struct msm_dp *msm_dp_display = &dp->msm_dp_display; + + pm_runtime_put_sync(&msm_dp_display->pdev->dev); + + drm_dbg_dp(dp->drm_dev, "type=%d Done\n", msm_dp_display->connector_type); +} + +void msm_dp_display_atomic_post_disable(struct msm_dp *dp) +{ + struct msm_dp_display_private *msm_dp_display; + + msm_dp_display = container_of(dp, struct msm_dp_display_private, msm_dp_display); + + if (dp->is_edp) + msm_dp_hpd_unplug_handle(msm_dp_display); + + msm_dp_display_audio_notify_disable(msm_dp_display); + + msm_dp_display_disable(msm_dp_display, msm_dp_display->panel); + + msm_dp_display_unprepare(msm_dp_display); +} + +void msm_dp_bridge_hpd_enable(struct drm_bridge *bridge) +{ + struct msm_dp_bridge *msm_dp_bridge = to_dp_bridge(bridge); + struct msm_dp *msm_dp_display = msm_dp_bridge->msm_dp_display; + struct msm_dp_display_private *dp = container_of(msm_dp_display, struct msm_dp_display_private, msm_dp_display); + + /* + * this is for external DP with hpd irq enabled case, + * step-1: msm_dp_pm_runtime_resume() enable dp host only + * step-2: enable hdp block and have hpd irq enabled here + * step-3: waiting for plugin irq while phy is not initialized + * step-4: DP PHY is initialized at plugin handler before link training + * + */ + if (pm_runtime_resume_and_get(&msm_dp_display->pdev->dev)) { + DRM_ERROR("failed to resume power\n"); + return; + } + + msm_dp_aux_hpd_enable(dp->aux); + msm_dp_aux_hpd_intr_enable(dp->aux); +} + +void msm_dp_bridge_hpd_disable(struct drm_bridge *bridge) +{ + struct msm_dp_bridge *msm_dp_bridge = to_dp_bridge(bridge); + struct msm_dp *msm_dp_display = msm_dp_bridge->msm_dp_display; + struct msm_dp_display_private *dp = container_of(msm_dp_display, struct msm_dp_display_private, msm_dp_display); + + msm_dp_aux_hpd_intr_disable(dp->aux); + msm_dp_aux_hpd_disable(dp->aux); + + pm_runtime_put_sync(&msm_dp_display->pdev->dev); +} + +void msm_dp_bridge_hpd_notify(struct drm_bridge *bridge, + struct drm_connector *connector, + enum drm_connector_status status) +{ + struct msm_dp_bridge *msm_dp_bridge = to_dp_bridge(bridge); + struct msm_dp *msm_dp_display = msm_dp_bridge->msm_dp_display; + struct msm_dp_display_private *dp = container_of(msm_dp_display, struct msm_dp_display_private, msm_dp_display); + u32 hpd_link_status = 0; + + if (pm_runtime_resume_and_get(&msm_dp_display->pdev->dev)) { + DRM_ERROR("failed to pm_runtime_resume\n"); + return; + } + + hpd_link_status = msm_dp_aux_is_link_connected(dp->aux); + + drm_dbg_dp(dp->drm_dev, "type=%d link hpd_link_status=0x%x, status=%d\n", + msm_dp_display->connector_type, hpd_link_status, status); + + if (status == connector_status_connected) { + if (hpd_link_status == ISR_HPD_REPLUG_COUNT) { + msm_dp_hpd_unplug_handle(dp); + msm_dp_hpd_plug_handle(dp); + } else if (hpd_link_status == ISR_IRQ_HPD_PULSE_COUNT) { + msm_dp_irq_hpd_handle(dp); + } else { + msm_dp_hpd_plug_handle(dp); + } + } else { + msm_dp_hpd_unplug_handle(dp); + } + + pm_runtime_put_sync(&msm_dp_display->pdev->dev); +} diff --git a/drivers/gpu/drm/msm/dp/dp_display.h b/drivers/gpu/drm/msm/dp/dp_display.h new file mode 100644 index 000000000..83117e761 --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_display.h @@ -0,0 +1,46 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved. + */ + +#ifndef _DP_DISPLAY_H_ +#define _DP_DISPLAY_H_ + +#include "dp_panel.h" +#include "disp/msm_disp_snapshot.h" + +#define DP_MAX_PIXEL_CLK_KHZ 675000 + +struct msm_dp { + struct drm_device *drm_dev; + struct platform_device *pdev; + struct drm_connector *connector; + struct drm_bridge *next_bridge; + struct drm_bridge *bridge; + bool audio_enabled; + bool power_on; + unsigned int connector_type; + bool is_edp; + + struct msm_dp_audio *msm_dp_audio; + bool psr_supported; +}; + +int msm_dp_display_get_modes(struct msm_dp *msm_dp_display); +bool msm_dp_display_check_video_test(struct msm_dp *msm_dp_display); +int msm_dp_display_get_test_bpp(struct msm_dp *msm_dp_display); +void msm_dp_display_signal_audio_start(struct msm_dp *msm_dp_display); +void msm_dp_display_signal_audio_complete(struct msm_dp *msm_dp_display); +void msm_dp_display_set_psr(struct msm_dp *dp, bool enter); +void msm_dp_display_debugfs_init(struct msm_dp *msm_dp_display, struct dentry *dentry, bool is_edp); +void msm_dp_display_atomic_post_disable(struct msm_dp *dp_display); +void msm_dp_display_atomic_disable(struct msm_dp *dp_display); +void msm_dp_display_atomic_pre_enable(struct msm_dp *dp_display, + struct drm_atomic_commit *state); +void msm_dp_display_atomic_enable(struct msm_dp *dp_display, + struct drm_atomic_commit *state); +enum drm_mode_status msm_dp_display_mode_valid(struct msm_dp *dp, + const struct drm_display_info *info, + const struct drm_display_mode *mode); + +#endif /* _DP_DISPLAY_H_ */ diff --git a/drivers/gpu/drm/msm/dp/dp_drm.c b/drivers/gpu/drm/msm/dp/dp_drm.c new file mode 100644 index 000000000..c1e4147bf --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_drm.c @@ -0,0 +1,375 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved. + */ + +#include +#include +#include +#include +#include +#include + +#include "msm_drv.h" +#include "msm_kms.h" +#include "dp_audio.h" +#include "dp_drm.h" + +/** + * msm_dp_bridge_get_modes - callback to add drm modes via drm_mode_probed_add() + * @bridge: Poiner to drm bridge + * @connector: Pointer to drm connector structure + * Returns: Number of modes added + */ +static int msm_dp_bridge_get_modes(struct drm_bridge *bridge, struct drm_connector *connector) +{ + int rc = 0; + struct msm_dp *dp; + + if (!connector) + return 0; + + dp = to_dp_bridge(bridge)->msm_dp_display; + + /* pluggable case assumes EDID is read when HPD */ + rc = msm_dp_display_get_modes(dp); + if (rc <= 0) { + DRM_ERROR("failed to get DP sink modes, rc=%d\n", rc); + return rc; + } else { + drm_dbg_dp(connector->dev, "No sink connected\n"); + } + return rc; +} + +static void msm_dp_bridge_debugfs_init(struct drm_bridge *bridge, struct dentry *root) +{ + struct msm_dp *dp = to_dp_bridge(bridge)->msm_dp_display; + + msm_dp_display_debugfs_init(dp, root, false); +} + +static void msm_dp_bridge_atomic_pre_enable(struct drm_bridge *drm_bridge, + struct drm_atomic_commit *state) +{ + struct msm_dp_bridge *dp_bridge = to_dp_bridge(drm_bridge); + struct msm_dp *dp = dp_bridge->msm_dp_display; + + msm_dp_display_atomic_pre_enable(dp, state); +} + +static void msm_dp_bridge_atomic_enable(struct drm_bridge *drm_bridge, + struct drm_atomic_commit *state) +{ + struct msm_dp_bridge *dp_bridge = to_dp_bridge(drm_bridge); + struct msm_dp *dp = dp_bridge->msm_dp_display; + + msm_dp_display_atomic_enable(dp, state); +} + +static void msm_dp_bridge_atomic_disable(struct drm_bridge *drm_bridge, + struct drm_atomic_commit *state) +{ + struct msm_dp_bridge *dp_bridge = to_dp_bridge(drm_bridge); + struct msm_dp *dp = dp_bridge->msm_dp_display; + + msm_dp_display_atomic_disable(dp); +} + +static void msm_dp_bridge_atomic_post_disable(struct drm_bridge *drm_bridge, + struct drm_atomic_commit *state) +{ + struct msm_dp_bridge *dp_bridge = to_dp_bridge(drm_bridge); + struct msm_dp *dp = dp_bridge->msm_dp_display; + + msm_dp_display_atomic_post_disable(dp); +} + +static enum drm_mode_status msm_dp_bridge_mode_valid(struct drm_bridge *drm_bridge, + const struct drm_display_info *info, + const struct drm_display_mode *mode) +{ + struct msm_dp_bridge *dp_bridge = to_dp_bridge(drm_bridge); + struct msm_dp *dp = dp_bridge->msm_dp_display; + + return msm_dp_display_mode_valid(dp, info, mode); +} + +static const struct drm_bridge_funcs msm_dp_bridge_ops = { + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_create_state = drm_atomic_helper_bridge_create_state, + .atomic_pre_enable = msm_dp_bridge_atomic_pre_enable, + .atomic_enable = msm_dp_bridge_atomic_enable, + .atomic_disable = msm_dp_bridge_atomic_disable, + .atomic_post_disable = msm_dp_bridge_atomic_post_disable, + .mode_valid = msm_dp_bridge_mode_valid, + .get_modes = msm_dp_bridge_get_modes, + .detect = msm_dp_bridge_detect, + .hpd_enable = msm_dp_bridge_hpd_enable, + .hpd_disable = msm_dp_bridge_hpd_disable, + .hpd_notify = msm_dp_bridge_hpd_notify, + .debugfs_init = msm_dp_bridge_debugfs_init, + + .dp_audio_prepare = msm_dp_audio_prepare, + .dp_audio_shutdown = msm_dp_audio_shutdown, +}; + +static int msm_edp_bridge_atomic_check(struct drm_bridge *drm_bridge, + struct drm_bridge_state *bridge_state, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state) +{ + struct msm_dp *dp = to_dp_bridge(drm_bridge)->msm_dp_display; + + if (WARN_ON(!conn_state)) + return -ENODEV; + + conn_state->self_refresh_aware = dp->psr_supported; + + if (!conn_state->crtc || !crtc_state) + return 0; + + if (crtc_state->self_refresh_active && !dp->psr_supported) + return -EINVAL; + + return 0; +} + +static void msm_edp_bridge_atomic_enable(struct drm_bridge *drm_bridge, + struct drm_atomic_commit *state) +{ + struct drm_crtc *crtc; + struct drm_crtc_state *old_crtc_state; + struct msm_dp_bridge *msm_dp_bridge = to_dp_bridge(drm_bridge); + struct msm_dp *dp = msm_dp_bridge->msm_dp_display; + + /* + * Check the old state of the crtc to determine if the panel + * was put into psr state previously by the msm_edp_bridge_atomic_disable. + * If the panel is in psr, just exit psr state and skip the full + * bridge enable sequence. + */ + crtc = drm_atomic_get_new_crtc_for_encoder(state, + drm_bridge->encoder); + if (!crtc) + return; + + old_crtc_state = drm_atomic_get_old_crtc_state(state, crtc); + + if (old_crtc_state && old_crtc_state->self_refresh_active) { + msm_dp_display_set_psr(dp, false); + return; + } + + msm_dp_display_atomic_enable(dp, state); +} + +static void msm_edp_bridge_atomic_disable(struct drm_bridge *drm_bridge, + struct drm_atomic_commit *atomic_state) +{ + struct drm_crtc *crtc; + struct drm_crtc_state *new_crtc_state = NULL, *old_crtc_state = NULL; + struct msm_dp_bridge *msm_dp_bridge = to_dp_bridge(drm_bridge); + struct msm_dp *dp = msm_dp_bridge->msm_dp_display; + + crtc = drm_atomic_get_old_crtc_for_encoder(atomic_state, + drm_bridge->encoder); + if (!crtc) + goto out; + + new_crtc_state = drm_atomic_get_new_crtc_state(atomic_state, crtc); + if (!new_crtc_state) + goto out; + + old_crtc_state = drm_atomic_get_old_crtc_state(atomic_state, crtc); + if (!old_crtc_state) + goto out; + + /* + * Set self refresh mode if current crtc state is active. + * + * If old crtc state is active, then this is a display disable + * call while the sink is in psr state. So, exit psr here. + * The eDP controller will be disabled in the + * msm_edp_bridge_atomic_post_disable function. + * + * We observed sink is stuck in self refresh if psr exit is skipped + * when display disable occurs while the sink is in psr state. + */ + if (new_crtc_state->self_refresh_active) { + msm_dp_display_set_psr(dp, true); + return; + } else if (old_crtc_state->self_refresh_active) { + msm_dp_display_set_psr(dp, false); + return; + } + +out: + msm_dp_bridge_atomic_disable(drm_bridge, atomic_state); +} + +static void msm_edp_bridge_atomic_post_disable(struct drm_bridge *drm_bridge, + struct drm_atomic_commit *atomic_state) +{ + struct drm_crtc *crtc; + struct drm_crtc_state *new_crtc_state = NULL; + + crtc = drm_atomic_get_old_crtc_for_encoder(atomic_state, + drm_bridge->encoder); + if (!crtc) + return; + + new_crtc_state = drm_atomic_get_new_crtc_state(atomic_state, crtc); + if (!new_crtc_state) + return; + + /* + * Self refresh mode is already set in msm_edp_bridge_atomic_disable. + */ + if (new_crtc_state->self_refresh_active) + return; + + msm_dp_bridge_atomic_post_disable(drm_bridge, atomic_state); +} + +/** + * msm_edp_bridge_mode_valid - callback to determine if specified mode is valid + * @bridge: Pointer to drm bridge structure + * @info: display info + * @mode: Pointer to drm mode structure + * Returns: Validity status for specified mode + */ +static enum drm_mode_status msm_edp_bridge_mode_valid(struct drm_bridge *bridge, + const struct drm_display_info *info, + const struct drm_display_mode *mode) +{ + struct msm_dp *dp; + int mode_pclk_khz = mode->clock; + + dp = to_dp_bridge(bridge)->msm_dp_display; + + if (!dp || !mode_pclk_khz || !dp->connector) { + DRM_ERROR("invalid params\n"); + return -EINVAL; + } + + if (msm_dp_wide_bus_available(dp)) + mode_pclk_khz /= 2; + + if (mode_pclk_khz > DP_MAX_PIXEL_CLK_KHZ) + return MODE_CLOCK_HIGH; + + /* + * The eDP controller currently does not have a reliable way of + * enabling panel power to read sink capabilities. So, we rely + * on the panel driver to populate only supported modes for now. + */ + return MODE_OK; +} + +static void msm_edp_bridge_debugfs_init(struct drm_bridge *bridge, struct dentry *root) +{ + struct msm_dp *dp = to_dp_bridge(bridge)->msm_dp_display; + + msm_dp_display_debugfs_init(dp, root, true); +} + +static const struct drm_bridge_funcs msm_edp_bridge_ops = { + .atomic_pre_enable = msm_dp_bridge_atomic_pre_enable, + .atomic_enable = msm_edp_bridge_atomic_enable, + .atomic_disable = msm_edp_bridge_atomic_disable, + .atomic_post_disable = msm_edp_bridge_atomic_post_disable, + .mode_valid = msm_edp_bridge_mode_valid, + .atomic_create_state = drm_atomic_helper_bridge_create_state, + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_check = msm_edp_bridge_atomic_check, + .debugfs_init = msm_edp_bridge_debugfs_init, +}; + +int msm_dp_bridge_init(struct msm_dp *msm_dp_display, struct drm_device *dev, + struct drm_encoder *encoder, bool yuv_supported) +{ + int rc; + struct msm_dp_bridge *msm_dp_bridge; + struct drm_bridge *bridge; + + msm_dp_bridge = devm_drm_bridge_alloc(dev->dev, struct msm_dp_bridge, bridge, + msm_dp_display->is_edp ? &msm_edp_bridge_ops : + &msm_dp_bridge_ops); + if (IS_ERR(msm_dp_bridge)) + return PTR_ERR(msm_dp_bridge); + + msm_dp_bridge->msm_dp_display = msm_dp_display; + + bridge = &msm_dp_bridge->bridge; + bridge->type = msm_dp_display->connector_type; + bridge->ycbcr_420_allowed = yuv_supported; + + /* + * Many ops only make sense for DP. Why? + * - Detect/HPD are used by DRM to know if a display is _physically_ + * there, not whether the display is powered on / finished initting. + * On eDP we assume the display is always there because you can't + * know until power is applied. If we don't implement the ops DRM will + * assume our display is always there. + * - Currently eDP mode reading is driven by the panel driver. This + * allows the panel driver to properly power itself on to read the + * modes. + */ + if (!msm_dp_display->is_edp) { + bridge->ops = + DRM_BRIDGE_OP_DP_AUDIO | + DRM_BRIDGE_OP_DETECT | + DRM_BRIDGE_OP_HPD | + DRM_BRIDGE_OP_MODES; + bridge->hdmi_audio_dev = &msm_dp_display->pdev->dev; + bridge->hdmi_audio_max_i2s_playback_channels = 8; + bridge->hdmi_audio_dai_port = -1; + } + + rc = devm_drm_bridge_add(dev->dev, bridge); + if (rc) { + DRM_ERROR("failed to add bridge, rc=%d\n", rc); + + return rc; + } + + rc = drm_bridge_attach(encoder, bridge, NULL, DRM_BRIDGE_ATTACH_NO_CONNECTOR); + if (rc) { + DRM_ERROR("failed to attach bridge, rc=%d\n", rc); + + return rc; + } + + if (msm_dp_display->next_bridge) { + rc = drm_bridge_attach(encoder, + msm_dp_display->next_bridge, bridge, + DRM_BRIDGE_ATTACH_NO_CONNECTOR); + if (rc < 0) { + DRM_ERROR("failed to attach panel bridge: %d\n", rc); + return rc; + } + } + + msm_dp_display->bridge = bridge; + + return 0; +} + +/* connector initialization */ +struct drm_connector *msm_dp_drm_connector_init(struct msm_dp *msm_dp_display, + struct drm_encoder *encoder) +{ + struct drm_connector *connector = NULL; + + connector = drm_bridge_connector_init(msm_dp_display->drm_dev, encoder); + if (IS_ERR(connector)) + return connector; + + if (!msm_dp_display->is_edp) + drm_connector_attach_dp_subconnector_property(connector); + + return connector; +} diff --git a/drivers/gpu/drm/msm/dp/dp_drm.h b/drivers/gpu/drm/msm/dp/dp_drm.h new file mode 100644 index 000000000..da412c788 --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_drm.h @@ -0,0 +1,36 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved. + */ + +#ifndef _DP_DRM_H_ +#define _DP_DRM_H_ + +#include +#include + +#include "msm_drv.h" +#include "dp_display.h" + +struct msm_dp_bridge { + struct drm_bridge bridge; + struct msm_dp *msm_dp_display; +}; + +#define to_dp_bridge(x) container_of((x), struct msm_dp_bridge, bridge) + +struct drm_connector *msm_dp_drm_connector_init(struct msm_dp *msm_dp_display, + struct drm_encoder *encoder); +int msm_dp_bridge_init(struct msm_dp *msm_dp_display, struct drm_device *dev, + struct drm_encoder *encoder, + bool yuv_supported); + +enum drm_connector_status msm_dp_bridge_detect(struct drm_bridge *bridge, + struct drm_connector *connector); +void msm_dp_bridge_hpd_enable(struct drm_bridge *bridge); +void msm_dp_bridge_hpd_disable(struct drm_bridge *bridge); +void msm_dp_bridge_hpd_notify(struct drm_bridge *bridge, + struct drm_connector *connector, + enum drm_connector_status status); + +#endif /* _DP_DRM_H_ */ diff --git a/drivers/gpu/drm/msm/dp/dp_link.c b/drivers/gpu/drm/msm/dp/dp_link.c new file mode 100644 index 000000000..34a91e194 --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_link.c @@ -0,0 +1,1350 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved. + */ + +#define pr_fmt(fmt) "[drm-dp] %s: " fmt, __func__ + +#include +#include +#include + +#include "dp_reg.h" +#include "dp_link.h" +#include "dp_panel.h" + +#define DP_LINK_RATE_HBR2 540000 /* kbytes */ +#define DP_TEST_REQUEST_MASK 0x7F + +enum audio_sample_rate { + AUDIO_SAMPLE_RATE_32_KHZ = 0x00, + AUDIO_SAMPLE_RATE_44_1_KHZ = 0x01, + AUDIO_SAMPLE_RATE_48_KHZ = 0x02, + AUDIO_SAMPLE_RATE_88_2_KHZ = 0x03, + AUDIO_SAMPLE_RATE_96_KHZ = 0x04, + AUDIO_SAMPLE_RATE_176_4_KHZ = 0x05, + AUDIO_SAMPLE_RATE_192_KHZ = 0x06, +}; + +enum audio_pattern_type { + AUDIO_TEST_PATTERN_OPERATOR_DEFINED = 0x00, + AUDIO_TEST_PATTERN_SAWTOOTH = 0x01, +}; + +struct msm_dp_link_request { + u32 test_requested; + u32 test_link_rate; + u32 test_lane_count; +}; + +struct msm_dp_link_private { + u32 prev_sink_count; + struct drm_device *drm_dev; + struct drm_dp_aux *aux; + struct msm_dp_link msm_dp_link; + + struct msm_dp_link_request request; + struct mutex psm_mutex; + u8 link_status[DP_LINK_STATUS_SIZE]; +}; + +static int msm_dp_aux_link_power_up(struct drm_dp_aux *aux, + struct msm_dp_link_info *link) +{ + u8 value; + ssize_t len; + int i; + + if (link->revision < 0x11) + return 0; + + len = drm_dp_dpcd_readb(aux, DP_SET_POWER, &value); + if (len < 0) + return len; + + value &= ~DP_SET_POWER_MASK; + value |= DP_SET_POWER_D0; + + /* retry for 1ms to give the sink time to wake up */ + for (i = 0; i < 3; i++) { + len = drm_dp_dpcd_writeb(aux, DP_SET_POWER, value); + usleep_range(1000, 2000); + if (len == 1) + break; + } + + return 0; +} + +static int msm_dp_aux_link_power_down(struct drm_dp_aux *aux, + struct msm_dp_link_info *link) +{ + u8 value; + int err; + + if (link->revision < 0x11) + return 0; + + err = drm_dp_dpcd_readb(aux, DP_SET_POWER, &value); + if (err < 0) + return err; + + value &= ~DP_SET_POWER_MASK; + value |= DP_SET_POWER_D3; + + err = drm_dp_dpcd_writeb(aux, DP_SET_POWER, value); + if (err < 0) + return err; + + return 0; +} + +static int msm_dp_link_get_period(struct msm_dp_link_private *link, int const addr) +{ + int ret = 0; + u8 data; + u32 const max_audio_period = 0xA; + + /* TEST_AUDIO_PERIOD_CH_XX */ + if (drm_dp_dpcd_readb(link->aux, addr, &data) < 0) { + DRM_ERROR("failed to read test_audio_period (0x%x)\n", addr); + ret = -EINVAL; + goto exit; + } + + /* Period - Bits 3:0 */ + data = data & 0xF; + if ((int)data > max_audio_period) { + DRM_ERROR("invalid test_audio_period_ch_1 = 0x%x\n", data); + ret = -EINVAL; + goto exit; + } + + ret = data; +exit: + return ret; +} + +static int msm_dp_link_parse_audio_channel_period(struct msm_dp_link_private *link) +{ + int ret = 0; + struct msm_dp_link_test_audio *req = &link->msm_dp_link.test_audio; + + ret = msm_dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH1); + if (ret == -EINVAL) + goto exit; + + req->test_audio_period_ch_1 = ret; + drm_dbg_dp(link->drm_dev, "test_audio_period_ch_1 = 0x%x\n", ret); + + ret = msm_dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH2); + if (ret == -EINVAL) + goto exit; + + req->test_audio_period_ch_2 = ret; + drm_dbg_dp(link->drm_dev, "test_audio_period_ch_2 = 0x%x\n", ret); + + /* TEST_AUDIO_PERIOD_CH_3 (Byte 0x275) */ + ret = msm_dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH3); + if (ret == -EINVAL) + goto exit; + + req->test_audio_period_ch_3 = ret; + drm_dbg_dp(link->drm_dev, "test_audio_period_ch_3 = 0x%x\n", ret); + + ret = msm_dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH4); + if (ret == -EINVAL) + goto exit; + + req->test_audio_period_ch_4 = ret; + drm_dbg_dp(link->drm_dev, "test_audio_period_ch_4 = 0x%x\n", ret); + + ret = msm_dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH5); + if (ret == -EINVAL) + goto exit; + + req->test_audio_period_ch_5 = ret; + drm_dbg_dp(link->drm_dev, "test_audio_period_ch_5 = 0x%x\n", ret); + + ret = msm_dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH6); + if (ret == -EINVAL) + goto exit; + + req->test_audio_period_ch_6 = ret; + drm_dbg_dp(link->drm_dev, "test_audio_period_ch_6 = 0x%x\n", ret); + + ret = msm_dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH7); + if (ret == -EINVAL) + goto exit; + + req->test_audio_period_ch_7 = ret; + drm_dbg_dp(link->drm_dev, "test_audio_period_ch_7 = 0x%x\n", ret); + + ret = msm_dp_link_get_period(link, DP_TEST_AUDIO_PERIOD_CH8); + if (ret == -EINVAL) + goto exit; + + req->test_audio_period_ch_8 = ret; + drm_dbg_dp(link->drm_dev, "test_audio_period_ch_8 = 0x%x\n", ret); +exit: + return ret; +} + +static int msm_dp_link_parse_audio_pattern_type(struct msm_dp_link_private *link) +{ + int ret = 0; + u8 data; + ssize_t rlen; + int const max_audio_pattern_type = 0x1; + + rlen = drm_dp_dpcd_readb(link->aux, + DP_TEST_AUDIO_PATTERN_TYPE, &data); + if (rlen < 0) { + DRM_ERROR("failed to read link audio mode. rlen=%zd\n", rlen); + return rlen; + } + + /* Audio Pattern Type - Bits 7:0 */ + if ((int)data > max_audio_pattern_type) { + DRM_ERROR("invalid audio pattern type = 0x%x\n", data); + ret = -EINVAL; + goto exit; + } + + link->msm_dp_link.test_audio.test_audio_pattern_type = data; + drm_dbg_dp(link->drm_dev, "audio pattern type = 0x%x\n", data); +exit: + return ret; +} + +static int msm_dp_link_parse_audio_mode(struct msm_dp_link_private *link) +{ + int ret = 0; + u8 data; + ssize_t rlen; + int const max_audio_sampling_rate = 0x6; + int const max_audio_channel_count = 0x8; + int sampling_rate = 0x0; + int channel_count = 0x0; + + rlen = drm_dp_dpcd_readb(link->aux, DP_TEST_AUDIO_MODE, &data); + if (rlen < 0) { + DRM_ERROR("failed to read link audio mode. rlen=%zd\n", rlen); + return rlen; + } + + /* Sampling Rate - Bits 3:0 */ + sampling_rate = data & 0xF; + if (sampling_rate > max_audio_sampling_rate) { + DRM_ERROR("sampling rate (0x%x) greater than max (0x%x)\n", + sampling_rate, max_audio_sampling_rate); + ret = -EINVAL; + goto exit; + } + + /* Channel Count - Bits 7:4 */ + channel_count = ((data & 0xF0) >> 4) + 1; + if (channel_count > max_audio_channel_count) { + DRM_ERROR("channel_count (0x%x) greater than max (0x%x)\n", + channel_count, max_audio_channel_count); + ret = -EINVAL; + goto exit; + } + + link->msm_dp_link.test_audio.test_audio_sampling_rate = sampling_rate; + link->msm_dp_link.test_audio.test_audio_channel_count = channel_count; + drm_dbg_dp(link->drm_dev, + "sampling_rate = 0x%x, channel_count = 0x%x\n", + sampling_rate, channel_count); +exit: + return ret; +} + +static int msm_dp_link_parse_audio_pattern_params(struct msm_dp_link_private *link) +{ + int ret = 0; + + ret = msm_dp_link_parse_audio_mode(link); + if (ret) + goto exit; + + ret = msm_dp_link_parse_audio_pattern_type(link); + if (ret) + goto exit; + + ret = msm_dp_link_parse_audio_channel_period(link); + +exit: + return ret; +} + +static bool msm_dp_link_is_video_pattern_valid(u32 pattern) +{ + switch (pattern) { + case DP_NO_TEST_PATTERN: + case DP_COLOR_RAMP: + case DP_BLACK_AND_WHITE_VERTICAL_LINES: + case DP_COLOR_SQUARE: + return true; + default: + return false; + } +} + +/** + * msm_dp_link_is_bit_depth_valid() - validates the bit depth requested + * @tbd: bit depth requested by the sink + * + * Returns true if the requested bit depth is supported. + */ +static bool msm_dp_link_is_bit_depth_valid(u32 tbd) +{ + /* DP_TEST_VIDEO_PATTERN_NONE is treated as invalid */ + switch (tbd) { + case DP_TEST_BIT_DEPTH_6: + case DP_TEST_BIT_DEPTH_8: + case DP_TEST_BIT_DEPTH_10: + return true; + default: + return false; + } +} + +static int msm_dp_link_parse_timing_params1(struct msm_dp_link_private *link, + int addr, int len, u32 *val) +{ + u8 bp[2]; + int rlen; + + if (len != 2) + return -EINVAL; + + /* Read the requested video link pattern (Byte 0x221). */ + rlen = drm_dp_dpcd_read(link->aux, addr, bp, len); + if (rlen < len) { + DRM_ERROR("failed to read 0x%x\n", addr); + return -EINVAL; + } + + *val = bp[1] | (bp[0] << 8); + + return 0; +} + +static int msm_dp_link_parse_timing_params2(struct msm_dp_link_private *link, + int addr, int len, + u32 *val1, u32 *val2) +{ + u8 bp[2]; + int rlen; + + if (len != 2) + return -EINVAL; + + /* Read the requested video link pattern (Byte 0x221). */ + rlen = drm_dp_dpcd_read(link->aux, addr, bp, len); + if (rlen < len) { + DRM_ERROR("failed to read 0x%x\n", addr); + return -EINVAL; + } + + *val1 = (bp[0] & BIT(7)) >> 7; + *val2 = bp[1] | ((bp[0] & 0x7F) << 8); + + return 0; +} + +static int msm_dp_link_parse_timing_params3(struct msm_dp_link_private *link, + int addr, u32 *val) +{ + u8 bp; + u32 len = 1; + int rlen; + + rlen = drm_dp_dpcd_read(link->aux, addr, &bp, len); + if (rlen < 1) { + DRM_ERROR("failed to read 0x%x\n", addr); + return -EINVAL; + } + *val = bp; + + return 0; +} + +/** + * msm_dp_link_parse_video_pattern_params() - parses video pattern parameters from DPCD + * @link: Display Port Driver data + * + * Returns 0 if it successfully parses the video link pattern and the link + * bit depth requested by the sink and, and if the values parsed are valid. + */ +static int msm_dp_link_parse_video_pattern_params(struct msm_dp_link_private *link) +{ + int ret = 0; + ssize_t rlen; + u8 bp; + + rlen = drm_dp_dpcd_readb(link->aux, DP_TEST_PATTERN, &bp); + if (rlen < 0) { + DRM_ERROR("failed to read link video pattern. rlen=%zd\n", + rlen); + return rlen; + } + + if (!msm_dp_link_is_video_pattern_valid(bp)) { + DRM_ERROR("invalid link video pattern = 0x%x\n", bp); + ret = -EINVAL; + return ret; + } + + link->msm_dp_link.test_video.test_video_pattern = bp; + + /* Read the requested color bit depth and dynamic range (Byte 0x232) */ + rlen = drm_dp_dpcd_readb(link->aux, DP_TEST_MISC0, &bp); + if (rlen < 0) { + DRM_ERROR("failed to read link bit depth. rlen=%zd\n", rlen); + return rlen; + } + + /* Dynamic Range */ + link->msm_dp_link.test_video.test_dyn_range = + (bp & DP_TEST_DYNAMIC_RANGE_CEA); + + /* Color bit depth */ + bp &= DP_TEST_BIT_DEPTH_MASK; + if (!msm_dp_link_is_bit_depth_valid(bp)) { + DRM_ERROR("invalid link bit depth = 0x%x\n", bp); + ret = -EINVAL; + return ret; + } + + link->msm_dp_link.test_video.test_bit_depth = bp; + + /* resolution timing params */ + ret = msm_dp_link_parse_timing_params1(link, DP_TEST_H_TOTAL_HI, 2, + &link->msm_dp_link.test_video.test_h_total); + if (ret) { + DRM_ERROR("failed to parse test_htotal(DP_TEST_H_TOTAL_HI)\n"); + return ret; + } + + ret = msm_dp_link_parse_timing_params1(link, DP_TEST_V_TOTAL_HI, 2, + &link->msm_dp_link.test_video.test_v_total); + if (ret) { + DRM_ERROR("failed to parse test_v_total(DP_TEST_V_TOTAL_HI)\n"); + return ret; + } + + ret = msm_dp_link_parse_timing_params1(link, DP_TEST_H_START_HI, 2, + &link->msm_dp_link.test_video.test_h_start); + if (ret) { + DRM_ERROR("failed to parse test_h_start(DP_TEST_H_START_HI)\n"); + return ret; + } + + ret = msm_dp_link_parse_timing_params1(link, DP_TEST_V_START_HI, 2, + &link->msm_dp_link.test_video.test_v_start); + if (ret) { + DRM_ERROR("failed to parse test_v_start(DP_TEST_V_START_HI)\n"); + return ret; + } + + ret = msm_dp_link_parse_timing_params2(link, DP_TEST_HSYNC_HI, 2, + &link->msm_dp_link.test_video.test_hsync_pol, + &link->msm_dp_link.test_video.test_hsync_width); + if (ret) { + DRM_ERROR("failed to parse (DP_TEST_HSYNC_HI)\n"); + return ret; + } + + ret = msm_dp_link_parse_timing_params2(link, DP_TEST_VSYNC_HI, 2, + &link->msm_dp_link.test_video.test_vsync_pol, + &link->msm_dp_link.test_video.test_vsync_width); + if (ret) { + DRM_ERROR("failed to parse (DP_TEST_VSYNC_HI)\n"); + return ret; + } + + ret = msm_dp_link_parse_timing_params1(link, DP_TEST_H_WIDTH_HI, 2, + &link->msm_dp_link.test_video.test_h_width); + if (ret) { + DRM_ERROR("failed to parse test_h_width(DP_TEST_H_WIDTH_HI)\n"); + return ret; + } + + ret = msm_dp_link_parse_timing_params1(link, DP_TEST_V_HEIGHT_HI, 2, + &link->msm_dp_link.test_video.test_v_height); + if (ret) { + DRM_ERROR("failed to parse test_v_height\n"); + return ret; + } + + ret = msm_dp_link_parse_timing_params3(link, DP_TEST_MISC1, + &link->msm_dp_link.test_video.test_rr_d); + link->msm_dp_link.test_video.test_rr_d &= DP_TEST_REFRESH_DENOMINATOR; + if (ret) { + DRM_ERROR("failed to parse test_rr_d (DP_TEST_MISC1)\n"); + return ret; + } + + ret = msm_dp_link_parse_timing_params3(link, DP_TEST_REFRESH_RATE_NUMERATOR, + &link->msm_dp_link.test_video.test_rr_n); + if (ret) { + DRM_ERROR("failed to parse test_rr_n\n"); + return ret; + } + + drm_dbg_dp(link->drm_dev, + "link video pattern = 0x%x\n" + "link dynamic range = 0x%x\n" + "link bit depth = 0x%x\n" + "TEST_H_TOTAL = %d, TEST_V_TOTAL = %d\n" + "TEST_H_START = %d, TEST_V_START = %d\n" + "TEST_HSYNC_POL = %d\n" + "TEST_HSYNC_WIDTH = %d\n" + "TEST_VSYNC_POL = %d\n" + "TEST_VSYNC_WIDTH = %d\n" + "TEST_H_WIDTH = %d\n" + "TEST_V_HEIGHT = %d\n" + "TEST_REFRESH_DENOMINATOR = %d\n" + "TEST_REFRESH_NUMERATOR = %d\n", + link->msm_dp_link.test_video.test_video_pattern, + link->msm_dp_link.test_video.test_dyn_range, + link->msm_dp_link.test_video.test_bit_depth, + link->msm_dp_link.test_video.test_h_total, + link->msm_dp_link.test_video.test_v_total, + link->msm_dp_link.test_video.test_h_start, + link->msm_dp_link.test_video.test_v_start, + link->msm_dp_link.test_video.test_hsync_pol, + link->msm_dp_link.test_video.test_hsync_width, + link->msm_dp_link.test_video.test_vsync_pol, + link->msm_dp_link.test_video.test_vsync_width, + link->msm_dp_link.test_video.test_h_width, + link->msm_dp_link.test_video.test_v_height, + link->msm_dp_link.test_video.test_rr_d, + link->msm_dp_link.test_video.test_rr_n); + + return ret; +} + +/** + * msm_dp_link_parse_link_training_params() - parses link training parameters from + * DPCD + * @link: Display Port Driver data + * + * Returns 0 if it successfully parses the link rate (Byte 0x219) and lane + * count (Byte 0x220), and if these values parse are valid. + */ +static int msm_dp_link_parse_link_training_params(struct msm_dp_link_private *link) +{ + u8 bp; + ssize_t rlen; + + rlen = drm_dp_dpcd_readb(link->aux, DP_TEST_LINK_RATE, &bp); + if (rlen < 0) { + DRM_ERROR("failed to read link rate. rlen=%zd\n", rlen); + return rlen; + } + + if (!is_link_rate_valid(bp)) { + DRM_ERROR("invalid link rate = 0x%x\n", bp); + return -EINVAL; + } + + link->request.test_link_rate = bp; + drm_dbg_dp(link->drm_dev, "link rate = 0x%x\n", + link->request.test_link_rate); + + rlen = drm_dp_dpcd_readb(link->aux, DP_TEST_LANE_COUNT, &bp); + if (rlen < 0) { + DRM_ERROR("failed to read lane count. rlen=%zd\n", rlen); + return rlen; + } + bp &= DP_MAX_LANE_COUNT_MASK; + + if (!is_lane_count_valid(bp)) { + DRM_ERROR("invalid lane count = 0x%x\n", bp); + return -EINVAL; + } + + link->request.test_lane_count = bp; + drm_dbg_dp(link->drm_dev, "lane count = 0x%x\n", + link->request.test_lane_count); + return 0; +} + +/** + * msm_dp_link_parse_phy_test_params() - parses the phy link parameters + * @link: Display Port Driver data + * + * Parses the DPCD (Byte 0x248) for the DP PHY link pattern that is being + * requested. + */ +static int msm_dp_link_parse_phy_test_params(struct msm_dp_link_private *link) +{ + u8 data; + ssize_t rlen; + + rlen = drm_dp_dpcd_readb(link->aux, DP_PHY_TEST_PATTERN, + &data); + if (rlen < 0) { + DRM_ERROR("failed to read phy link pattern. rlen=%zd\n", rlen); + return rlen; + } + + link->msm_dp_link.phy_params.phy_test_pattern_sel = data & 0x07; + + drm_dbg_dp(link->drm_dev, "phy_test_pattern_sel = 0x%x\n", data); + + switch (data) { + case DP_PHY_TEST_PATTERN_SEL_MASK: + case DP_PHY_TEST_PATTERN_NONE: + case DP_PHY_TEST_PATTERN_D10_2: + case DP_PHY_TEST_PATTERN_ERROR_COUNT: + case DP_PHY_TEST_PATTERN_PRBS7: + case DP_PHY_TEST_PATTERN_80BIT_CUSTOM: + case DP_PHY_TEST_PATTERN_CP2520: + return 0; + default: + return -EINVAL; + } +} + +/** + * msm_dp_link_is_video_audio_test_requested() - checks for audio/video link request + * @link: link requested by the sink + * + * Returns true if the requested link is a permitted audio/video link. + */ +static bool msm_dp_link_is_video_audio_test_requested(u32 link) +{ + u8 video_audio_test = (DP_TEST_LINK_VIDEO_PATTERN | + DP_TEST_LINK_AUDIO_PATTERN | + DP_TEST_LINK_AUDIO_DISABLED_VIDEO); + + return ((link & video_audio_test) && + !(link & ~video_audio_test)); +} + +/** + * msm_dp_link_parse_request() - parses link request parameters from sink + * @link: Display Port Driver data + * + * Parses the DPCD to check if an automated link is requested (Byte 0x201), + * and what type of link automation is being requested (Byte 0x218). + */ +static int msm_dp_link_parse_request(struct msm_dp_link_private *link) +{ + int ret = 0; + u8 data; + ssize_t rlen; + + /** + * Read the device service IRQ vector (Byte 0x201) to determine + * whether an automated link has been requested by the sink. + */ + rlen = drm_dp_dpcd_readb(link->aux, + DP_DEVICE_SERVICE_IRQ_VECTOR, &data); + if (rlen < 0) { + DRM_ERROR("aux read failed. rlen=%zd\n", rlen); + return rlen; + } + + drm_dbg_dp(link->drm_dev, "device service irq vector = 0x%x\n", data); + + if (!(data & DP_AUTOMATED_TEST_REQUEST)) { + drm_dbg_dp(link->drm_dev, "no test requested\n"); + return 0; + } + + /** + * Read the link request byte (Byte 0x218) to determine what type + * of automated link has been requested by the sink. + */ + rlen = drm_dp_dpcd_readb(link->aux, DP_TEST_REQUEST, &data); + if (rlen < 0) { + DRM_ERROR("aux read failed. rlen=%zd\n", rlen); + return rlen; + } + + if (!data || (data == DP_TEST_LINK_FAUX_PATTERN)) { + drm_dbg_dp(link->drm_dev, "link 0x%x not supported\n", data); + goto end; + } + + drm_dbg_dp(link->drm_dev, "Test:(0x%x) requested\n", data); + link->request.test_requested = data; + if (link->request.test_requested == DP_TEST_LINK_PHY_TEST_PATTERN) { + ret = msm_dp_link_parse_phy_test_params(link); + if (ret) + goto end; + ret = msm_dp_link_parse_link_training_params(link); + if (ret) + goto end; + } + + if (link->request.test_requested == DP_TEST_LINK_TRAINING) { + ret = msm_dp_link_parse_link_training_params(link); + if (ret) + goto end; + } + + if (msm_dp_link_is_video_audio_test_requested( + link->request.test_requested)) { + ret = msm_dp_link_parse_video_pattern_params(link); + if (ret) + goto end; + + ret = msm_dp_link_parse_audio_pattern_params(link); + } +end: + /* + * Send a DP_TEST_ACK if all link parameters are valid, otherwise send + * a DP_TEST_NAK. + */ + if (ret) { + link->msm_dp_link.test_response = DP_TEST_NAK; + } else { + if (link->request.test_requested != DP_TEST_LINK_EDID_READ) + link->msm_dp_link.test_response = DP_TEST_ACK; + else + link->msm_dp_link.test_response = + DP_TEST_EDID_CHECKSUM_WRITE; + } + + return ret; +} + +static int msm_dp_link_parse_sink_status_field(struct msm_dp_link_private *link) +{ + int ret; + + link->prev_sink_count = link->msm_dp_link.sink_count; + ret = drm_dp_read_sink_count(link->aux); + if (ret < 0) { + DRM_ERROR("DP parse sink count failed\n"); + return ret; + } + link->msm_dp_link.sink_count = ret; + + ret = drm_dp_dpcd_read_link_status(link->aux, + link->link_status); + if (ret < 0) { + DRM_ERROR("DP link status read failed\n"); + return ret; + } + + return msm_dp_link_parse_request(link); +} + +/** + * msm_dp_link_process_link_training_request() - processes new training requests + * @link: Display Port link data + * + * This function will handle new link training requests that are initiated by + * the sink. In particular, it will update the requested lane count and link + * rate, and then trigger the link retraining procedure. + * + * The function will return 0 if a link training request has been processed, + * otherwise it will return -EINVAL. + */ +static int msm_dp_link_process_link_training_request(struct msm_dp_link_private *link) +{ + if (link->request.test_requested != DP_TEST_LINK_TRAINING) + return -EINVAL; + + drm_dbg_dp(link->drm_dev, + "Test:0x%x link rate = 0x%x, lane count = 0x%x\n", + DP_TEST_LINK_TRAINING, + link->request.test_link_rate, + link->request.test_lane_count); + + link->msm_dp_link.link_params.num_lanes = link->request.test_lane_count; + link->msm_dp_link.link_params.rate = + drm_dp_bw_code_to_link_rate(link->request.test_link_rate); + + return 0; +} + +bool msm_dp_link_send_test_response(struct msm_dp_link *msm_dp_link) +{ + struct msm_dp_link_private *link = NULL; + int ret = 0; + + if (!msm_dp_link) { + DRM_ERROR("invalid input\n"); + return false; + } + + link = container_of(msm_dp_link, struct msm_dp_link_private, msm_dp_link); + + ret = drm_dp_dpcd_writeb(link->aux, DP_TEST_RESPONSE, + msm_dp_link->test_response); + + return ret == 1; +} + +int msm_dp_link_psm_config(struct msm_dp_link *msm_dp_link, + struct msm_dp_link_info *link_info, bool enable) +{ + struct msm_dp_link_private *link = NULL; + int ret = 0; + + if (!msm_dp_link) { + DRM_ERROR("invalid params\n"); + return -EINVAL; + } + + link = container_of(msm_dp_link, struct msm_dp_link_private, msm_dp_link); + + mutex_lock(&link->psm_mutex); + if (enable) + ret = msm_dp_aux_link_power_down(link->aux, link_info); + else + ret = msm_dp_aux_link_power_up(link->aux, link_info); + + if (ret) + DRM_ERROR("Failed to %s low power mode\n", enable ? + "enter" : "exit"); + + mutex_unlock(&link->psm_mutex); + return ret; +} + +bool msm_dp_link_send_edid_checksum(struct msm_dp_link *msm_dp_link, u8 checksum) +{ + struct msm_dp_link_private *link = NULL; + int ret = 0; + + if (!msm_dp_link) { + DRM_ERROR("invalid input\n"); + return false; + } + + link = container_of(msm_dp_link, struct msm_dp_link_private, msm_dp_link); + + ret = drm_dp_dpcd_writeb(link->aux, DP_TEST_EDID_CHECKSUM, + checksum); + return ret == 1; +} + +static void msm_dp_link_parse_vx_px(struct msm_dp_link_private *link) +{ + drm_dbg_dp(link->drm_dev, "vx: 0=%d, 1=%d, 2=%d, 3=%d\n", + drm_dp_get_adjust_request_voltage(link->link_status, 0), + drm_dp_get_adjust_request_voltage(link->link_status, 1), + drm_dp_get_adjust_request_voltage(link->link_status, 2), + drm_dp_get_adjust_request_voltage(link->link_status, 3)); + + drm_dbg_dp(link->drm_dev, "px: 0=%d, 1=%d, 2=%d, 3=%d\n", + drm_dp_get_adjust_request_pre_emphasis(link->link_status, 0), + drm_dp_get_adjust_request_pre_emphasis(link->link_status, 1), + drm_dp_get_adjust_request_pre_emphasis(link->link_status, 2), + drm_dp_get_adjust_request_pre_emphasis(link->link_status, 3)); + + /** + * Update the voltage and pre-emphasis levels as per DPCD request + * vector. + */ + drm_dbg_dp(link->drm_dev, + "Current: v_level = 0x%x, p_level = 0x%x\n", + link->msm_dp_link.phy_params.v_level, + link->msm_dp_link.phy_params.p_level); + link->msm_dp_link.phy_params.v_level = + drm_dp_get_adjust_request_voltage(link->link_status, 0); + link->msm_dp_link.phy_params.p_level = + drm_dp_get_adjust_request_pre_emphasis(link->link_status, 0); + + link->msm_dp_link.phy_params.p_level >>= DP_TRAIN_PRE_EMPHASIS_SHIFT; + + drm_dbg_dp(link->drm_dev, + "Requested: v_level = 0x%x, p_level = 0x%x\n", + link->msm_dp_link.phy_params.v_level, + link->msm_dp_link.phy_params.p_level); +} + +/** + * msm_dp_link_process_phy_test_pattern_request() - process new phy link requests + * @link: Display Port Driver data + * + * This function will handle new phy link pattern requests that are initiated + * by the sink. The function will return 0 if a phy link pattern has been + * processed, otherwise it will return -EINVAL. + */ +static int msm_dp_link_process_phy_test_pattern_request( + struct msm_dp_link_private *link) +{ + if (!(link->request.test_requested & DP_TEST_LINK_PHY_TEST_PATTERN)) { + drm_dbg_dp(link->drm_dev, "no phy test\n"); + return -EINVAL; + } + + if (!is_link_rate_valid(link->request.test_link_rate) || + !is_lane_count_valid(link->request.test_lane_count)) { + DRM_ERROR("Invalid: link rate = 0x%x,lane count = 0x%x\n", + link->request.test_link_rate, + link->request.test_lane_count); + return -EINVAL; + } + + drm_dbg_dp(link->drm_dev, + "Current: rate = 0x%x, lane count = 0x%x\n", + link->msm_dp_link.link_params.rate, + link->msm_dp_link.link_params.num_lanes); + + drm_dbg_dp(link->drm_dev, + "Requested: rate = 0x%x, lane count = 0x%x\n", + link->request.test_link_rate, + link->request.test_lane_count); + + link->msm_dp_link.link_params.num_lanes = link->request.test_lane_count; + link->msm_dp_link.link_params.rate = + drm_dp_bw_code_to_link_rate(link->request.test_link_rate); + + msm_dp_link_parse_vx_px(link); + + return 0; +} + +static bool msm_dp_link_read_psr_error_status(struct msm_dp_link_private *link) +{ + u8 status; + + drm_dp_dpcd_read(link->aux, DP_PSR_ERROR_STATUS, &status, 1); + + if (status & DP_PSR_LINK_CRC_ERROR) + DRM_ERROR("PSR LINK CRC ERROR\n"); + else if (status & DP_PSR_RFB_STORAGE_ERROR) + DRM_ERROR("PSR RFB STORAGE ERROR\n"); + else if (status & DP_PSR_VSC_SDP_UNCORRECTABLE_ERROR) + DRM_ERROR("PSR VSC SDP UNCORRECTABLE ERROR\n"); + else + return false; + + return true; +} + +static bool msm_dp_link_psr_capability_changed(struct msm_dp_link_private *link) +{ + u8 status; + + drm_dp_dpcd_read(link->aux, DP_PSR_ESI, &status, 1); + + if (status & DP_PSR_CAPS_CHANGE) { + drm_dbg_dp(link->drm_dev, "PSR Capability Change\n"); + return true; + } + + return false; +} + +static u8 get_link_status(const u8 link_status[DP_LINK_STATUS_SIZE], int r) +{ + return link_status[r - DP_LANE0_1_STATUS]; +} + +/** + * msm_dp_link_process_link_status_update() - processes link status updates + * @link: Display Port link module data + * + * This function will check for changes in the link status, e.g. clock + * recovery done on all lanes, and trigger link training if there is a + * failure/error on the link. + * + * The function will return 0 if the a link status update has been processed, + * otherwise it will return -EINVAL. + */ +static int msm_dp_link_process_link_status_update(struct msm_dp_link_private *link) +{ + bool channel_eq_done = drm_dp_channel_eq_ok(link->link_status, + link->msm_dp_link.link_params.num_lanes); + + bool clock_recovery_done = drm_dp_clock_recovery_ok(link->link_status, + link->msm_dp_link.link_params.num_lanes); + + drm_dbg_dp(link->drm_dev, + "channel_eq_done = %d, clock_recovery_done = %d\n", + channel_eq_done, clock_recovery_done); + + if (channel_eq_done && clock_recovery_done) + return -EINVAL; + + return 0; +} + +/** + * msm_dp_link_process_ds_port_status_change() - process port status changes + * @link: Display Port Driver data + * + * This function will handle downstream port updates that are initiated by + * the sink. If the downstream port status has changed, the EDID is read via + * AUX. + * + * The function will return 0 if a downstream port update has been + * processed, otherwise it will return -EINVAL. + */ +static int msm_dp_link_process_ds_port_status_change(struct msm_dp_link_private *link) +{ + if (get_link_status(link->link_status, DP_LANE_ALIGN_STATUS_UPDATED) & + DP_DOWNSTREAM_PORT_STATUS_CHANGED) + goto reset; + + if (link->prev_sink_count == link->msm_dp_link.sink_count) + return -EINVAL; + +reset: + /* reset prev_sink_count */ + link->prev_sink_count = link->msm_dp_link.sink_count; + + return 0; +} + +static bool msm_dp_link_is_video_pattern_requested(struct msm_dp_link_private *link) +{ + return (link->request.test_requested & DP_TEST_LINK_VIDEO_PATTERN) + && !(link->request.test_requested & + DP_TEST_LINK_AUDIO_DISABLED_VIDEO); +} + +static bool msm_dp_link_is_audio_pattern_requested(struct msm_dp_link_private *link) +{ + return (link->request.test_requested & DP_TEST_LINK_AUDIO_PATTERN); +} + +static void msm_dp_link_reset_data(struct msm_dp_link_private *link) +{ + link->request = (const struct msm_dp_link_request){ 0 }; + link->msm_dp_link.test_video = (const struct msm_dp_link_test_video){ 0 }; + link->msm_dp_link.test_video.test_bit_depth = DP_TEST_BIT_DEPTH_UNKNOWN; + link->msm_dp_link.test_audio = (const struct msm_dp_link_test_audio){ 0 }; + link->msm_dp_link.phy_params.phy_test_pattern_sel = 0; + link->msm_dp_link.sink_request = 0; + link->msm_dp_link.test_response = 0; +} + +/** + * msm_dp_link_process_request() - handle HPD IRQ transition to HIGH + * @msm_dp_link: pointer to link module data + * + * This function will handle the HPD IRQ state transitions from LOW to HIGH + * (including cases when there are back to back HPD IRQ HIGH) indicating + * the start of a new link training request or sink status update. + */ +int msm_dp_link_process_request(struct msm_dp_link *msm_dp_link) +{ + int ret = 0; + struct msm_dp_link_private *link; + + if (!msm_dp_link) { + DRM_ERROR("invalid input\n"); + return -EINVAL; + } + + link = container_of(msm_dp_link, struct msm_dp_link_private, msm_dp_link); + + msm_dp_link_reset_data(link); + + ret = msm_dp_link_parse_sink_status_field(link); + if (ret) + return ret; + + if (link->request.test_requested == DP_TEST_LINK_EDID_READ) { + msm_dp_link->sink_request |= DP_TEST_LINK_EDID_READ; + } else if (!msm_dp_link_process_ds_port_status_change(link)) { + msm_dp_link->sink_request |= DS_PORT_STATUS_CHANGED; + } else if (!msm_dp_link_process_link_training_request(link)) { + msm_dp_link->sink_request |= DP_TEST_LINK_TRAINING; + } else if (!msm_dp_link_process_phy_test_pattern_request(link)) { + msm_dp_link->sink_request |= DP_TEST_LINK_PHY_TEST_PATTERN; + } else if (msm_dp_link_read_psr_error_status(link)) { + DRM_ERROR("PSR IRQ_HPD received\n"); + } else if (msm_dp_link_psr_capability_changed(link)) { + drm_dbg_dp(link->drm_dev, "PSR Capability changed\n"); + } else { + ret = msm_dp_link_process_link_status_update(link); + if (!ret) { + msm_dp_link->sink_request |= DP_LINK_STATUS_UPDATED; + } else { + if (msm_dp_link_is_video_pattern_requested(link)) { + ret = 0; + msm_dp_link->sink_request |= DP_TEST_LINK_VIDEO_PATTERN; + } + if (msm_dp_link_is_audio_pattern_requested(link)) { + msm_dp_link->sink_request |= DP_TEST_LINK_AUDIO_PATTERN; + ret = -EINVAL; + } + } + } + + drm_dbg_dp(link->drm_dev, "sink request=%#x\n", + msm_dp_link->sink_request); + return ret; +} + +int msm_dp_link_get_colorimetry_config(struct msm_dp_link *msm_dp_link) +{ + u32 cc = DP_MISC0_COLORIMERY_CFG_LEGACY_RGB; + struct msm_dp_link_private *link; + + if (!msm_dp_link) { + DRM_ERROR("invalid input\n"); + return -EINVAL; + } + + link = container_of(msm_dp_link, struct msm_dp_link_private, msm_dp_link); + + /* + * Unless a video pattern CTS test is ongoing, use RGB_VESA + * Only RGB_VESA and RGB_CEA supported for now + */ + if (msm_dp_link_is_video_pattern_requested(link)) { + if (link->msm_dp_link.test_video.test_dyn_range & + DP_TEST_DYNAMIC_RANGE_CEA) + cc = DP_MISC0_COLORIMERY_CFG_CEA_RGB; + } + + return cc; +} + +int msm_dp_link_adjust_levels(struct msm_dp_link *msm_dp_link, u8 *link_status) +{ + int i; + u8 max_p_level; + int v_max = 0, p_max = 0; + struct msm_dp_link_private *link; + + if (!msm_dp_link) { + DRM_ERROR("invalid input\n"); + return -EINVAL; + } + + link = container_of(msm_dp_link, struct msm_dp_link_private, msm_dp_link); + + /* use the max level across lanes */ + for (i = 0; i < msm_dp_link->link_params.num_lanes; i++) { + u8 data_v = drm_dp_get_adjust_request_voltage(link_status, i); + u8 data_p = drm_dp_get_adjust_request_pre_emphasis(link_status, + i); + drm_dbg_dp(link->drm_dev, + "lane=%d req_vol_swing=%d req_pre_emphasis=%d\n", + i, data_v, data_p); + if (v_max < data_v) + v_max = data_v; + if (p_max < data_p) + p_max = data_p; + } + + msm_dp_link->phy_params.v_level = v_max >> DP_TRAIN_VOLTAGE_SWING_SHIFT; + msm_dp_link->phy_params.p_level = p_max >> DP_TRAIN_PRE_EMPHASIS_SHIFT; + + /** + * Adjust the voltage swing and pre-emphasis level combination to within + * the allowable range. + */ + if (msm_dp_link->phy_params.v_level > DP_TRAIN_LEVEL_MAX) { + drm_dbg_dp(link->drm_dev, + "Requested vSwingLevel=%d, change to %d\n", + msm_dp_link->phy_params.v_level, + DP_TRAIN_LEVEL_MAX); + msm_dp_link->phy_params.v_level = DP_TRAIN_LEVEL_MAX; + } + + if (msm_dp_link->phy_params.p_level > DP_TRAIN_LEVEL_MAX) { + drm_dbg_dp(link->drm_dev, + "Requested preEmphasisLevel=%d, change to %d\n", + msm_dp_link->phy_params.p_level, + DP_TRAIN_LEVEL_MAX); + msm_dp_link->phy_params.p_level = DP_TRAIN_LEVEL_MAX; + } + + max_p_level = DP_TRAIN_LEVEL_MAX - msm_dp_link->phy_params.v_level; + if (msm_dp_link->phy_params.p_level > max_p_level) { + drm_dbg_dp(link->drm_dev, + "Requested preEmphasisLevel=%d, change to %d\n", + msm_dp_link->phy_params.p_level, + max_p_level); + msm_dp_link->phy_params.p_level = max_p_level; + } + + drm_dbg_dp(link->drm_dev, "adjusted: v_level=%d, p_level=%d\n", + msm_dp_link->phy_params.v_level, msm_dp_link->phy_params.p_level); + + return 0; +} + +void msm_dp_link_reset_phy_params_vx_px(struct msm_dp_link *msm_dp_link) +{ + msm_dp_link->phy_params.v_level = 0; + msm_dp_link->phy_params.p_level = 0; +} + +u32 msm_dp_link_get_test_bits_depth(struct msm_dp_link *msm_dp_link, u32 bpp) +{ + u32 tbd; + struct msm_dp_link_private *link; + + link = container_of(msm_dp_link, struct msm_dp_link_private, msm_dp_link); + + /* + * Few simplistic rules and assumptions made here: + * 1. Test bit depth is bit depth per color component + * 2. Assume 3 color components + */ + switch (bpp) { + case 18: + tbd = DP_TEST_BIT_DEPTH_6; + break; + case 24: + tbd = DP_TEST_BIT_DEPTH_8; + break; + case 30: + tbd = DP_TEST_BIT_DEPTH_10; + break; + default: + drm_dbg_dp(link->drm_dev, "bpp=%d not supported, use bpc=8\n", + bpp); + tbd = DP_TEST_BIT_DEPTH_8; + break; + } + + tbd = (tbd >> DP_TEST_BIT_DEPTH_SHIFT); + + return tbd; +} + +static u32 msm_dp_link_link_frequencies(struct device_node *of_node) +{ + struct device_node *endpoint; + u64 frequency = 0; + int cnt; + + endpoint = of_graph_get_endpoint_by_regs(of_node, 1, 0); /* port@1 */ + if (!endpoint) + return 0; + + cnt = of_property_count_u64_elems(endpoint, "link-frequencies"); + + if (cnt > 0) + of_property_read_u64_index(endpoint, "link-frequencies", + cnt - 1, &frequency); + of_node_put(endpoint); + + do_div(frequency, + 10 * /* from symbol rate to link rate */ + 1000); /* kbytes */ + + return frequency; +} + +/* + * Always populate msm_dp_link->lane_map with 4 lanes. + * - Use DTS "data-lanes" if present; otherwise fall back to default mapping. + * - For partial definitions, fill remaining entries with unused lanes in + * ascending order. + */ +static int msm_dp_link_lane_map(struct device *dev, struct msm_dp_link *msm_dp_link) +{ + struct device_node *of_node = dev->of_node; + struct device_node *endpoint; + int cnt = msm_dp_link->max_dp_lanes; + u32 tmp[DP_MAX_NUM_DP_LANES]; + u32 map[DP_MAX_NUM_DP_LANES] = {0, 1, 2, 3}; /* default 1:1 mapping */ + bool used[DP_MAX_NUM_DP_LANES] = {false}; + int i, j = 0, ret = -EINVAL; + + endpoint = of_graph_get_endpoint_by_regs(of_node, 1, -1); + if (endpoint) { + ret = of_property_read_u32_array(endpoint, "data-lanes", tmp, cnt); + if (ret) + dev_dbg(dev, "endpoint data-lanes read failed (ret=%d)\n", ret); + } + + if (ret) { + ret = of_property_read_u32_array(of_node, "data-lanes", tmp, cnt); + if (ret) { + dev_info(dev, "data-lanes not defined, set to default\n"); + goto out; + } + } + + for (i = 0; i < cnt; i++) { + if (tmp[i] >= DP_MAX_NUM_DP_LANES) { + dev_err(dev, "data-lanes[%d]=%u out of range\n", i, tmp[i]); + return -EINVAL; + } + used[tmp[i]] = true; + map[i] = tmp[i]; + } + + /* Fill the remaining entries with unused physical lanes (ascending) */ + for (i = cnt; i < DP_MAX_NUM_DP_LANES && j < DP_MAX_NUM_DP_LANES; j++) { + if (!used[j]) + map[i++] = j; + } + +out: + if (endpoint) + of_node_put(endpoint); + + dev_dbg(dev, "data-lanes count %d <%d %d %d %d>\n", cnt, map[0], map[1], map[2], map[3]); + memcpy(msm_dp_link->lane_map, map, sizeof(map)); + return 0; +} + +static int msm_dp_link_parse_dt(struct device *dev, struct msm_dp_link *msm_dp_link) +{ + struct device_node *of_node = dev->of_node; + int cnt; + + /* + * data-lanes is the property of msm_dp_out endpoint + */ + cnt = drm_of_get_data_lanes_count_ep(of_node, 1, 0, 1, DP_MAX_NUM_DP_LANES); + if (cnt < 0) { + /* legacy code, data-lanes is the property of mdss_dp node */ + cnt = drm_of_get_data_lanes_count(of_node, 1, DP_MAX_NUM_DP_LANES); + } + + if (cnt > 0) + msm_dp_link->max_dp_lanes = cnt; + else + msm_dp_link->max_dp_lanes = DP_MAX_NUM_DP_LANES; /* 4 lanes */ + + if (msm_dp_link_lane_map(dev, msm_dp_link)) { + dev_err(dev, "failed to parse data-lanes\n"); + return -EINVAL; + } + + msm_dp_link->max_dp_link_rate = msm_dp_link_link_frequencies(of_node); + if (!msm_dp_link->max_dp_link_rate) + msm_dp_link->max_dp_link_rate = DP_LINK_RATE_HBR2; + + return 0; +} + +struct msm_dp_link *msm_dp_link_get(struct device *dev, struct drm_dp_aux *aux) +{ + struct msm_dp_link_private *link; + struct msm_dp_link *msm_dp_link; + int ret; + + if (!dev || !aux) { + DRM_ERROR("invalid input\n"); + return ERR_PTR(-EINVAL); + } + + link = devm_kzalloc(dev, sizeof(*link), GFP_KERNEL); + if (!link) + return ERR_PTR(-ENOMEM); + + link->aux = aux; + + mutex_init(&link->psm_mutex); + msm_dp_link = &link->msm_dp_link; + + ret = msm_dp_link_parse_dt(dev, msm_dp_link); + if (ret) + return ERR_PTR(ret); + + return msm_dp_link; +} diff --git a/drivers/gpu/drm/msm/dp/dp_link.h b/drivers/gpu/drm/msm/dp/dp_link.h new file mode 100644 index 000000000..76125e9c0 --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_link.h @@ -0,0 +1,133 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved. + */ + +#ifndef _DP_LINK_H_ +#define _DP_LINK_H_ + +#include "dp_aux.h" +#include + +#define DS_PORT_STATUS_CHANGED 0x200 +#define DP_TEST_BIT_DEPTH_UNKNOWN 0xFFFFFFFF +#define DP_LINK_CAP_ENHANCED_FRAMING (1 << 0) +#define DP_MAX_NUM_DP_LANES 4 + +struct msm_dp_link_info { + unsigned char revision; + unsigned int rate; + unsigned int supported_rates[DP_MAX_SUPPORTED_RATES]; + unsigned int rate_set; + bool use_rate_set; + unsigned int num_lanes; + unsigned long capabilities; +}; + +#define DP_TRAIN_LEVEL_MAX 3 + +struct msm_dp_link_test_video { + u32 test_video_pattern; + u32 test_bit_depth; + u32 test_dyn_range; + u32 test_h_total; + u32 test_v_total; + u32 test_h_start; + u32 test_v_start; + u32 test_hsync_pol; + u32 test_hsync_width; + u32 test_vsync_pol; + u32 test_vsync_width; + u32 test_h_width; + u32 test_v_height; + u32 test_rr_d; + u32 test_rr_n; +}; + +struct msm_dp_link_test_audio { + u32 test_audio_sampling_rate; + u32 test_audio_channel_count; + u32 test_audio_pattern_type; + u32 test_audio_period_ch_1; + u32 test_audio_period_ch_2; + u32 test_audio_period_ch_3; + u32 test_audio_period_ch_4; + u32 test_audio_period_ch_5; + u32 test_audio_period_ch_6; + u32 test_audio_period_ch_7; + u32 test_audio_period_ch_8; +}; + +struct msm_dp_link_phy_params { + u32 phy_test_pattern_sel; + u8 v_level; + u8 p_level; +}; + +struct msm_dp_link { + u8 lttpr_common_caps[DP_LTTPR_COMMON_CAP_SIZE]; + int lttpr_count; + + u32 sink_request; + u32 test_response; + + u8 sink_count; + struct msm_dp_link_test_video test_video; + struct msm_dp_link_test_audio test_audio; + struct msm_dp_link_phy_params phy_params; + struct msm_dp_link_info link_params; + + u32 lane_map[DP_MAX_NUM_DP_LANES]; + u32 max_dp_lanes; + u32 max_dp_link_rate; +}; + +/** + * msm_dp_link_bit_depth_to_bpp() - convert test bit depth to bpp + * @tbd: test bit depth + * + * Returns: the bits per pixel (bpp) to be used corresponding to the + * bit depth value. This function assumes that bit depth has + * already been validated. + */ +static inline u32 msm_dp_link_bit_depth_to_bpp(u32 tbd) +{ + /* + * Few simplistic rules and assumptions made here: + * 1. Bit depth is per color component + * 2. If bit depth is unknown return 0 + * 3. Assume 3 color components + */ + switch (tbd) { + case DP_TEST_BIT_DEPTH_6: + return 18; + case DP_TEST_BIT_DEPTH_8: + return 24; + case DP_TEST_BIT_DEPTH_10: + return 30; + case DP_TEST_BIT_DEPTH_UNKNOWN: + default: + return 0; + } +} + +void msm_dp_link_reset_phy_params_vx_px(struct msm_dp_link *msm_dp_link); +u32 msm_dp_link_get_test_bits_depth(struct msm_dp_link *msm_dp_link, u32 bpp); +int msm_dp_link_process_request(struct msm_dp_link *msm_dp_link); +int msm_dp_link_get_colorimetry_config(struct msm_dp_link *msm_dp_link); +int msm_dp_link_adjust_levels(struct msm_dp_link *msm_dp_link, u8 *link_status); +bool msm_dp_link_send_test_response(struct msm_dp_link *msm_dp_link); +int msm_dp_link_psm_config(struct msm_dp_link *msm_dp_link, + struct msm_dp_link_info *link_info, bool enable); +bool msm_dp_link_send_edid_checksum(struct msm_dp_link *msm_dp_link, u8 checksum); + +/** + * msm_dp_link_get() - get the functionalities of dp test module + * @dev: kernel device structure + * @aux: DisplayPort AUX channel + * + * return: a pointer to msm_dp_link struct + */ +struct msm_dp_link *msm_dp_link_get(struct device *dev, struct drm_dp_aux *aux); + +#endif /* _DP_LINK_H_ */ diff --git a/drivers/gpu/drm/msm/dp/dp_panel.c b/drivers/gpu/drm/msm/dp/dp_panel.c new file mode 100644 index 000000000..e76dad0f6 --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_panel.c @@ -0,0 +1,729 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved. + */ + +#include "dp_panel.h" +#include "dp_reg.h" +#include "dp_utils.h" + +#include +#include +#include +#include + +#include +#include +#include + +#define DP_INTF_CONFIG_DATABUS_WIDEN BIT(4) + +struct msm_dp_panel_private { + struct device *dev; + struct drm_device *drm_dev; + struct msm_dp_panel msm_dp_panel; + struct drm_dp_aux *aux; + struct msm_dp_link *link; + void __iomem *link_base; + void __iomem *p0_base; + bool panel_on; +}; + +static inline u32 msm_dp_read_link(struct msm_dp_panel_private *panel, u32 offset) +{ + return readl_relaxed(panel->link_base + offset); +} + +static inline void msm_dp_write_link(struct msm_dp_panel_private *panel, + u32 offset, u32 data) +{ + /* + * To make sure link reg writes happens before any other operation, + * this function uses writel() instread of writel_relaxed() + */ + writel(data, panel->link_base + offset); +} + +static inline void msm_dp_write_p0(struct msm_dp_panel_private *panel, + u32 offset, u32 data) +{ + /* + * To make sure interface reg writes happens before any other operation, + * this function uses writel() instread of writel_relaxed() + */ + writel(data, panel->p0_base + offset); +} + +static inline u32 msm_dp_read_p0(struct msm_dp_panel_private *panel, + u32 offset) +{ + /* + * To make sure interface reg writes happens before any other operation, + * this function uses writel() instread of writel_relaxed() + */ + return readl_relaxed(panel->p0_base + offset); +} + +static void msm_dp_panel_read_psr_cap(struct msm_dp_panel_private *panel) +{ + ssize_t rlen; + struct msm_dp_panel *msm_dp_panel; + + msm_dp_panel = &panel->msm_dp_panel; + + /* edp sink */ + if (msm_dp_panel->dpcd[DP_EDP_CONFIGURATION_CAP]) { + rlen = drm_dp_dpcd_read(panel->aux, DP_PSR_SUPPORT, + &msm_dp_panel->psr_cap, sizeof(msm_dp_panel->psr_cap)); + if (rlen == sizeof(msm_dp_panel->psr_cap)) { + drm_dbg_dp(panel->drm_dev, + "psr version: 0x%x, psr_cap: 0x%x\n", + msm_dp_panel->psr_cap.version, + msm_dp_panel->psr_cap.capabilities); + } else + DRM_ERROR("failed to read psr info, rlen=%zd\n", rlen); + } +} + +static int msm_dp_panel_read_dpcd(struct msm_dp_panel *msm_dp_panel) +{ + int rc, max_lttpr_lanes, max_lttpr_rate; + struct msm_dp_panel_private *panel; + struct msm_dp_link_info *link_info; + struct msm_dp_link *link; + u8 *dpcd, major, minor; + + panel = container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + dpcd = msm_dp_panel->dpcd; + rc = drm_dp_read_dpcd_caps(panel->aux, dpcd); + if (rc) + return rc; + + msm_dp_panel->vsc_sdp_supported = drm_dp_vsc_sdp_supported(panel->aux, dpcd); + link_info = &msm_dp_panel->link_info; + link_info->revision = dpcd[DP_DPCD_REV]; + major = (link_info->revision >> 4) & 0x0f; + minor = link_info->revision & 0x0f; + + link = panel->link; + drm_dbg_dp(panel->drm_dev, "max_lanes=%d max_link_rate=%d\n", + link->max_dp_lanes, link->max_dp_link_rate); + + max_lttpr_lanes = drm_dp_lttpr_max_lane_count(link->lttpr_common_caps); + max_lttpr_rate = drm_dp_lttpr_max_link_rate(link->lttpr_common_caps); + + /* eDP sink */ + if (msm_dp_panel->dpcd[DP_EDP_CONFIGURATION_CAP]) { + u8 edp_rev; + + rc = drm_dp_dpcd_read_byte(panel->aux, DP_EDP_DPCD_REV, &edp_rev); + if (rc) + return rc; + + drm_dbg_dp(panel->drm_dev, "edp_rev=0x%x\n", edp_rev); + + /* For eDP v1.4+, parse the SUPPORTED_LINK_RATES table */ + if (edp_rev >= DP_EDP_14) { + __le16 rates[DP_MAX_SUPPORTED_RATES]; + u8 bw_set; + int i; + + rc = drm_dp_dpcd_read_data(panel->aux, DP_SUPPORTED_LINK_RATES, + rates, sizeof(rates)); + if (rc) + return rc; + + rc = drm_dp_dpcd_read_byte(panel->aux, DP_LINK_BW_SET, &bw_set); + if (rc) + return rc; + + /* Find index of max supported link rate that does not exceed dtsi limits */ + for (i = 0; i < ARRAY_SIZE(rates); i++) { + /* + * The value from the DPCD multiplied by 200 gives + * the link rate in kHz. Divide by 10 to convert to + * symbol rate, accounting for 8b/10b encoding. + */ + u32 rate = (le16_to_cpu(rates[i]) * 200) / 10; + + if (!rate) + break; + + drm_dbg_dp(panel->drm_dev, + "SUPPORTED_LINK_RATES[%d]: %d\n", i, rate); + + /* + * Limit link rate from link-frequencies of endpoint + * property of dtsi + */ + if (rate > link->max_dp_link_rate) + break; + + /* Limit link rate from LTTPR capabilities, if any */ + if (max_lttpr_rate && rate > max_lttpr_rate) + break; + + link_info->rate = rate; + link_info->supported_rates[i] = rate; + link_info->rate_set = i; + } + + /* Only use LINK_RATE_SET if LINK_BW_SET hasn't already been written to */ + if (!bw_set && link_info->rate) + link_info->use_rate_set = true; + } + } + + /* Fall back on MAX_LINK_RATE/LINK_BW_SET (DP, eDP <= v1.3) */ + if (!link_info->rate) { + link_info->rate = drm_dp_max_link_rate(dpcd); + + /* Limit link rate from link-frequencies of endpoint property of dtsi */ + if (link_info->rate > link->max_dp_link_rate) + link_info->rate = link->max_dp_link_rate; + + /* Limit link rate from LTTPR capabilities, if any */ + if (max_lttpr_rate && max_lttpr_rate < link_info->rate) + link_info->rate = max_lttpr_rate; + } + + link_info->num_lanes = drm_dp_max_lane_count(dpcd); + + /* Limit data lanes from data-lanes of endpoint property of dtsi */ + if (link_info->num_lanes > link->max_dp_lanes) + link_info->num_lanes = link->max_dp_lanes; + + /* Limit data lanes from LTTPR capabilities, if any */ + if (max_lttpr_lanes && max_lttpr_lanes < link_info->num_lanes) + link_info->num_lanes = max_lttpr_lanes; + + drm_dbg_dp(panel->drm_dev, "version: %d.%d\n", major, minor); + drm_dbg_dp(panel->drm_dev, "link_rate=%d\n", link_info->rate); + drm_dbg_dp(panel->drm_dev, "link_rate_set=%d\n", link_info->rate_set); + drm_dbg_dp(panel->drm_dev, "use_rate_set=%d\n", link_info->use_rate_set); + drm_dbg_dp(panel->drm_dev, "lane_count=%d\n", link_info->num_lanes); + + if (drm_dp_enhanced_frame_cap(dpcd)) + link_info->capabilities |= DP_LINK_CAP_ENHANCED_FRAMING; + + msm_dp_panel_read_psr_cap(panel); + + return rc; +} + +static u32 msm_dp_panel_get_supported_bpp(struct msm_dp_panel *msm_dp_panel, + u32 mode_edid_bpp, u32 mode_pclk_khz) +{ + const struct msm_dp_link_info *link_info; + const u32 max_supported_bpp = 30, min_supported_bpp = 18; + u32 bpp, data_rate_khz; + + bpp = min(mode_edid_bpp, max_supported_bpp); + + link_info = &msm_dp_panel->link_info; + data_rate_khz = link_info->num_lanes * link_info->rate * 8; + + do { + if (mode_pclk_khz * bpp <= data_rate_khz) + return bpp; + bpp -= 6; + } while (bpp > min_supported_bpp); + + return min_supported_bpp; +} + +int msm_dp_panel_read_link_caps(struct msm_dp_panel *msm_dp_panel, + struct drm_connector *connector) +{ + int rc, bw_code; + int count; + struct msm_dp_panel_private *panel; + + if (!msm_dp_panel || !connector) { + DRM_ERROR("invalid input\n"); + return -EINVAL; + } + + panel = container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + + rc = msm_dp_panel_read_dpcd(msm_dp_panel); + if (rc) { + DRM_ERROR("read dpcd failed %d\n", rc); + return rc; + } + + bw_code = drm_dp_link_rate_to_bw_code(msm_dp_panel->link_info.rate); + if (!is_link_rate_valid(bw_code) || + !is_lane_count_valid(msm_dp_panel->link_info.num_lanes) || + (bw_code > msm_dp_panel->max_bw_code)) { + DRM_ERROR("Illegal link rate=%d lane=%d\n", msm_dp_panel->link_info.rate, + msm_dp_panel->link_info.num_lanes); + return -EINVAL; + } + + if (drm_dp_is_branch(msm_dp_panel->dpcd)) { + count = drm_dp_read_sink_count(panel->aux); + if (!count) { + panel->link->sink_count = 0; + return -ENOTCONN; + } + } + + rc = drm_dp_read_downstream_info(panel->aux, msm_dp_panel->dpcd, + msm_dp_panel->downstream_ports); + return rc; +} + +u32 msm_dp_panel_get_mode_bpp(struct msm_dp_panel *msm_dp_panel, + u32 mode_edid_bpp, u32 mode_pclk_khz) +{ + struct msm_dp_panel_private *panel; + u32 bpp; + + if (!msm_dp_panel || !mode_edid_bpp || !mode_pclk_khz) { + DRM_ERROR("invalid input\n"); + return 0; + } + + panel = container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + + if (msm_dp_panel->video_test) + bpp = msm_dp_link_bit_depth_to_bpp( + panel->link->test_video.test_bit_depth); + else + bpp = msm_dp_panel_get_supported_bpp(msm_dp_panel, mode_edid_bpp, + mode_pclk_khz); + + return bpp; +} + +static u8 msm_dp_panel_get_edid_checksum(const struct edid *edid) +{ + edid += edid->extensions; + + return edid->checksum; +} + +void msm_dp_panel_handle_sink_request(struct msm_dp_panel *msm_dp_panel, + const struct drm_edid *drm_edid) +{ + struct msm_dp_panel_private *panel; + + if (!msm_dp_panel) { + DRM_ERROR("invalid input\n"); + return; + } + + panel = container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + + if (panel->link->sink_request & DP_TEST_LINK_EDID_READ) { + /* FIXME: get rid of drm_edid_raw() */ + const struct edid *edid = drm_edid_raw(drm_edid); + u8 checksum; + + if (edid) + checksum = msm_dp_panel_get_edid_checksum(edid); + else + checksum = msm_dp_panel->connector->real_edid_checksum; + + msm_dp_link_send_edid_checksum(panel->link, checksum); + msm_dp_link_send_test_response(panel->link); + } +} + +static void msm_dp_panel_tpg_enable(struct msm_dp_panel *msm_dp_panel, + struct drm_display_mode *drm_mode) +{ + struct msm_dp_panel_private *panel = + container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + u32 hsync_period, vsync_period; + u32 display_v_start, display_v_end; + u32 hsync_start_x, hsync_end_x; + u32 v_sync_width; + u32 hsync_ctl; + u32 display_hctl; + + /* TPG config parameters*/ + hsync_period = drm_mode->htotal; + vsync_period = drm_mode->vtotal; + + display_v_start = ((drm_mode->vtotal - drm_mode->vsync_start) * + hsync_period); + display_v_end = ((vsync_period - (drm_mode->vsync_start - + drm_mode->vdisplay)) + * hsync_period) - 1; + + display_v_start += drm_mode->htotal - drm_mode->hsync_start; + display_v_end -= (drm_mode->hsync_start - drm_mode->hdisplay); + + hsync_start_x = drm_mode->htotal - drm_mode->hsync_start; + hsync_end_x = hsync_period - (drm_mode->hsync_start - + drm_mode->hdisplay) - 1; + + v_sync_width = drm_mode->vsync_end - drm_mode->vsync_start; + + hsync_ctl = (hsync_period << 16) | + (drm_mode->hsync_end - drm_mode->hsync_start); + display_hctl = (hsync_end_x << 16) | hsync_start_x; + + + msm_dp_write_p0(panel, MMSS_DP_INTF_HSYNC_CTL, hsync_ctl); + msm_dp_write_p0(panel, MMSS_DP_INTF_VSYNC_PERIOD_F0, vsync_period * + hsync_period); + msm_dp_write_p0(panel, MMSS_DP_INTF_VSYNC_PULSE_WIDTH_F0, v_sync_width * + hsync_period); + msm_dp_write_p0(panel, MMSS_DP_INTF_VSYNC_PERIOD_F1, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_VSYNC_PULSE_WIDTH_F1, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_DISPLAY_HCTL, display_hctl); + msm_dp_write_p0(panel, MMSS_DP_INTF_ACTIVE_HCTL, 0); + msm_dp_write_p0(panel, MMSS_INTF_DISPLAY_V_START_F0, display_v_start); + msm_dp_write_p0(panel, MMSS_DP_INTF_DISPLAY_V_END_F0, display_v_end); + msm_dp_write_p0(panel, MMSS_INTF_DISPLAY_V_START_F1, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_DISPLAY_V_END_F1, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_ACTIVE_V_START_F0, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_ACTIVE_V_END_F0, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_ACTIVE_V_START_F1, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_ACTIVE_V_END_F1, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_POLARITY_CTL, 0); + + msm_dp_write_p0(panel, MMSS_DP_TPG_MAIN_CONTROL, + DP_TPG_CHECKERED_RECT_PATTERN); + msm_dp_write_p0(panel, MMSS_DP_TPG_VIDEO_CONFIG, + DP_TPG_VIDEO_CONFIG_BPP_8BIT | + DP_TPG_VIDEO_CONFIG_RGB); + msm_dp_write_p0(panel, MMSS_DP_BIST_ENABLE, + DP_BIST_ENABLE_DPBIST_EN); + msm_dp_write_p0(panel, MMSS_DP_TIMING_ENGINE_EN, + DP_TIMING_ENGINE_EN_EN); + drm_dbg_dp(panel->drm_dev, "%s: enabled tpg\n", __func__); +} + +static void msm_dp_panel_tpg_disable(struct msm_dp_panel *msm_dp_panel) +{ + struct msm_dp_panel_private *panel = + container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + + msm_dp_write_p0(panel, MMSS_DP_TPG_MAIN_CONTROL, 0x0); + msm_dp_write_p0(panel, MMSS_DP_BIST_ENABLE, 0x0); + msm_dp_write_p0(panel, MMSS_DP_TIMING_ENGINE_EN, 0x0); +} + +void msm_dp_panel_tpg_config(struct msm_dp_panel *msm_dp_panel, bool enable) +{ + struct msm_dp_panel_private *panel; + + if (!msm_dp_panel) { + DRM_ERROR("invalid input\n"); + return; + } + + panel = container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + + if (!panel->panel_on) { + drm_dbg_dp(panel->drm_dev, + "DP panel not enabled, handle TPG on next on\n"); + return; + } + + if (!enable) { + msm_dp_panel_tpg_disable(msm_dp_panel); + return; + } + + drm_dbg_dp(panel->drm_dev, "calling panel's tpg_enable\n"); + msm_dp_panel_tpg_enable(msm_dp_panel, &panel->msm_dp_panel.msm_dp_mode.drm_mode); +} + +void msm_dp_panel_clear_dsc_dto(struct msm_dp_panel *msm_dp_panel) +{ + struct msm_dp_panel_private *panel = + container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + + msm_dp_write_p0(panel, MMSS_DP_DSC_DTO, 0x0); +} + +static void msm_dp_panel_send_vsc_sdp(struct msm_dp_panel_private *panel, struct dp_sdp *vsc_sdp) +{ + u32 header[2]; + u32 val; + int i; + + msm_dp_utils_pack_sdp_header(&vsc_sdp->sdp_header, header); + + msm_dp_write_link(panel, MMSS_DP_GENERIC0_0, header[0]); + msm_dp_write_link(panel, MMSS_DP_GENERIC0_1, header[1]); + + for (i = 0; i < sizeof(vsc_sdp->db); i += 4) { + val = ((vsc_sdp->db[i]) | (vsc_sdp->db[i + 1] << 8) | (vsc_sdp->db[i + 2] << 16) | + (vsc_sdp->db[i + 3] << 24)); + msm_dp_write_link(panel, MMSS_DP_GENERIC0_2 + i, val); + } +} + +static void msm_dp_panel_update_sdp(struct msm_dp_panel_private *panel) +{ + u32 hw_revision = panel->msm_dp_panel.hw_revision; + + if (hw_revision >= DP_HW_VERSION_1_0 && + hw_revision < DP_HW_VERSION_1_2) { + msm_dp_write_link(panel, MMSS_DP_SDP_CFG3, UPDATE_SDP); + msm_dp_write_link(panel, MMSS_DP_SDP_CFG3, 0x0); + } +} + +void msm_dp_panel_enable_vsc_sdp(struct msm_dp_panel *msm_dp_panel, struct dp_sdp *vsc_sdp) +{ + struct msm_dp_panel_private *panel = + container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + u32 cfg, cfg2, misc; + + cfg = msm_dp_read_link(panel, MMSS_DP_SDP_CFG); + cfg2 = msm_dp_read_link(panel, MMSS_DP_SDP_CFG2); + misc = msm_dp_read_link(panel, REG_DP_MISC1_MISC0); + + cfg |= GEN0_SDP_EN; + msm_dp_write_link(panel, MMSS_DP_SDP_CFG, cfg); + + cfg2 |= GENERIC0_SDPSIZE_VALID; + msm_dp_write_link(panel, MMSS_DP_SDP_CFG2, cfg2); + + msm_dp_panel_send_vsc_sdp(panel, vsc_sdp); + + /* indicates presence of VSC (BIT(6) of MISC1) */ + misc |= DP_MISC1_VSC_SDP; + + drm_dbg_dp(panel->drm_dev, "vsc sdp enable=1\n"); + + pr_debug("misc settings = 0x%x\n", misc); + msm_dp_write_link(panel, REG_DP_MISC1_MISC0, misc); + + msm_dp_panel_update_sdp(panel); +} + +void msm_dp_panel_disable_vsc_sdp(struct msm_dp_panel *msm_dp_panel) +{ + struct msm_dp_panel_private *panel = + container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + u32 cfg, cfg2, misc; + + cfg = msm_dp_read_link(panel, MMSS_DP_SDP_CFG); + cfg2 = msm_dp_read_link(panel, MMSS_DP_SDP_CFG2); + misc = msm_dp_read_link(panel, REG_DP_MISC1_MISC0); + + cfg &= ~GEN0_SDP_EN; + msm_dp_write_link(panel, MMSS_DP_SDP_CFG, cfg); + + cfg2 &= ~GENERIC0_SDPSIZE_VALID; + msm_dp_write_link(panel, MMSS_DP_SDP_CFG2, cfg2); + + /* switch back to MSA */ + misc &= ~DP_MISC1_VSC_SDP; + + drm_dbg_dp(panel->drm_dev, "vsc sdp enable=0\n"); + + pr_debug("misc settings = 0x%x\n", misc); + msm_dp_write_link(panel, REG_DP_MISC1_MISC0, misc); + + msm_dp_panel_update_sdp(panel); +} + +static int msm_dp_panel_setup_vsc_sdp_yuv_420(struct msm_dp_panel *msm_dp_panel) +{ + struct msm_dp_display_mode *msm_dp_mode; + struct drm_dp_vsc_sdp vsc_sdp_data; + struct dp_sdp vsc_sdp; + ssize_t len; + + if (!msm_dp_panel) { + DRM_ERROR("invalid input\n"); + return -EINVAL; + } + + msm_dp_mode = &msm_dp_panel->msm_dp_mode; + + memset(&vsc_sdp_data, 0, sizeof(vsc_sdp_data)); + + /* VSC SDP header as per table 2-118 of DP 1.4 specification */ + vsc_sdp_data.sdp_type = DP_SDP_VSC; + vsc_sdp_data.revision = 0x05; + vsc_sdp_data.length = 0x13; + + /* VSC SDP Payload for DB16 */ + vsc_sdp_data.pixelformat = DP_PIXELFORMAT_YUV420; + vsc_sdp_data.colorimetry = DP_COLORIMETRY_DEFAULT; + + /* VSC SDP Payload for DB17 */ + vsc_sdp_data.bpc = msm_dp_mode->bpp / 3; + vsc_sdp_data.dynamic_range = DP_DYNAMIC_RANGE_CTA; + + /* VSC SDP Payload for DB18 */ + vsc_sdp_data.content_type = DP_CONTENT_TYPE_GRAPHICS; + + len = drm_dp_vsc_sdp_pack(&vsc_sdp_data, &vsc_sdp); + if (len < 0) { + DRM_ERROR("unable to pack vsc sdp\n"); + return len; + } + + msm_dp_panel_enable_vsc_sdp(msm_dp_panel, &vsc_sdp); + + return 0; +} + +int msm_dp_panel_timing_cfg(struct msm_dp_panel *msm_dp_panel, bool wide_bus_en) +{ + u32 data, total_ver, total_hor; + struct msm_dp_panel_private *panel; + struct drm_display_mode *drm_mode; + u32 width_blanking; + u32 sync_start; + u32 msm_dp_active; + u32 total; + u32 reg; + + panel = container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + drm_mode = &panel->msm_dp_panel.msm_dp_mode.drm_mode; + + drm_dbg_dp(panel->drm_dev, "width=%d hporch= %d %d %d\n", + drm_mode->hdisplay, drm_mode->htotal - drm_mode->hsync_end, + drm_mode->hsync_start - drm_mode->hdisplay, + drm_mode->hsync_end - drm_mode->hsync_start); + + drm_dbg_dp(panel->drm_dev, "height=%d vporch= %d %d %d\n", + drm_mode->vdisplay, drm_mode->vtotal - drm_mode->vsync_end, + drm_mode->vsync_start - drm_mode->vdisplay, + drm_mode->vsync_end - drm_mode->vsync_start); + + total_hor = drm_mode->htotal; + + total_ver = drm_mode->vtotal; + + data = total_ver; + data <<= 16; + data |= total_hor; + + total = data; + + data = (drm_mode->vtotal - drm_mode->vsync_start); + data <<= 16; + data |= (drm_mode->htotal - drm_mode->hsync_start); + + sync_start = data; + + data = drm_mode->vsync_end - drm_mode->vsync_start; + data <<= 16; + data |= (panel->msm_dp_panel.msm_dp_mode.v_active_low << 31); + data |= drm_mode->hsync_end - drm_mode->hsync_start; + data |= (panel->msm_dp_panel.msm_dp_mode.h_active_low << 15); + + width_blanking = data; + + data = drm_mode->vdisplay; + data <<= 16; + data |= drm_mode->hdisplay; + + msm_dp_active = data; + + msm_dp_write_link(panel, REG_DP_TOTAL_HOR_VER, total); + msm_dp_write_link(panel, REG_DP_START_HOR_VER_FROM_SYNC, sync_start); + msm_dp_write_link(panel, REG_DP_HSYNC_VSYNC_WIDTH_POLARITY, width_blanking); + msm_dp_write_link(panel, REG_DP_ACTIVE_HOR_VER, msm_dp_active); + + reg = msm_dp_read_p0(panel, MMSS_DP_INTF_CONFIG); + if (wide_bus_en) + reg |= DP_INTF_CONFIG_DATABUS_WIDEN; + else + reg &= ~DP_INTF_CONFIG_DATABUS_WIDEN; + + drm_dbg_dp(panel->drm_dev, "wide_bus_en=%d reg=%#x\n", wide_bus_en, reg); + + msm_dp_write_p0(panel, MMSS_DP_INTF_CONFIG, reg); + + if (msm_dp_panel->msm_dp_mode.out_fmt_is_yuv_420) + msm_dp_panel_setup_vsc_sdp_yuv_420(msm_dp_panel); + + panel->panel_on = true; + + return 0; +} + +int msm_dp_panel_init_panel_info(struct msm_dp_panel *msm_dp_panel, + const struct drm_display_mode *adjusted_mode, + u32 bpp) +{ + struct drm_display_mode *drm_mode; + struct msm_dp_panel_private *panel; + + panel = container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + + drm_mode_copy(&msm_dp_panel->msm_dp_mode.drm_mode, adjusted_mode); + msm_dp_panel->msm_dp_mode.bpp = bpp; + msm_dp_panel->msm_dp_mode.v_active_low = + !!(adjusted_mode->flags & DRM_MODE_FLAG_NVSYNC); + msm_dp_panel->msm_dp_mode.h_active_low = + !!(adjusted_mode->flags & DRM_MODE_FLAG_NHSYNC); + msm_dp_panel->msm_dp_mode.out_fmt_is_yuv_420 = + drm_mode_is_420_only(&msm_dp_panel->connector->display_info, adjusted_mode) && + msm_dp_panel->vsc_sdp_supported; + + drm_mode = &msm_dp_panel->msm_dp_mode.drm_mode; + + /* + * print resolution info as this is a result + * of user initiated action of cable connection + */ + drm_dbg_dp(panel->drm_dev, "SET NEW RESOLUTION:\n"); + drm_dbg_dp(panel->drm_dev, "%dx%d@%dfps\n", + drm_mode->hdisplay, drm_mode->vdisplay, drm_mode_vrefresh(drm_mode)); + drm_dbg_dp(panel->drm_dev, + "h_porches(back|front|width) = (%d|%d|%d)\n", + drm_mode->htotal - drm_mode->hsync_end, + drm_mode->hsync_start - drm_mode->hdisplay, + drm_mode->hsync_end - drm_mode->hsync_start); + drm_dbg_dp(panel->drm_dev, + "v_porches(back|front|width) = (%d|%d|%d)\n", + drm_mode->vtotal - drm_mode->vsync_end, + drm_mode->vsync_start - drm_mode->vdisplay, + drm_mode->vsync_end - drm_mode->vsync_start); + drm_dbg_dp(panel->drm_dev, "pixel clock (KHz)=(%d)\n", + drm_mode->clock); + drm_dbg_dp(panel->drm_dev, "bpp = %d\n", msm_dp_panel->msm_dp_mode.bpp); + + msm_dp_panel->msm_dp_mode.bpp = msm_dp_panel_get_mode_bpp(msm_dp_panel, msm_dp_panel->msm_dp_mode.bpp, + msm_dp_panel->msm_dp_mode.drm_mode.clock); + + drm_dbg_dp(panel->drm_dev, "updated bpp = %d\n", + msm_dp_panel->msm_dp_mode.bpp); + + return 0; +} + +struct msm_dp_panel *msm_dp_panel_get(struct device *dev, struct drm_dp_aux *aux, + struct msm_dp_link *link, + void __iomem *link_base, + void __iomem *p0_base) +{ + struct msm_dp_panel_private *panel; + struct msm_dp_panel *msm_dp_panel; + + if (!dev || !aux || !link) { + DRM_ERROR("invalid input\n"); + return ERR_PTR(-EINVAL); + } + + panel = devm_kzalloc(dev, sizeof(*panel), GFP_KERNEL); + if (!panel) + return ERR_PTR(-ENOMEM); + + panel->dev = dev; + panel->aux = aux; + panel->link = link; + panel->link_base = link_base; + panel->p0_base = p0_base; + + msm_dp_panel = &panel->msm_dp_panel; + msm_dp_panel->max_bw_code = DP_LINK_BW_8_1; + + return msm_dp_panel; +} + diff --git a/drivers/gpu/drm/msm/dp/dp_panel.h b/drivers/gpu/drm/msm/dp/dp_panel.h new file mode 100644 index 000000000..4519ac374 --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_panel.h @@ -0,0 +1,95 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved. + */ + +#ifndef _DP_PANEL_H_ +#define _DP_PANEL_H_ + +#include +#include + +#include "dp_aux.h" +#include "dp_link.h" + +struct edid; + +struct msm_dp_display_mode { + struct drm_display_mode drm_mode; + u32 bpp; + u32 h_active_low; + u32 v_active_low; + bool out_fmt_is_yuv_420; +}; + +struct msm_dp_panel_psr { + u8 version; + u8 capabilities; +}; + +struct msm_dp_panel { + /* dpcd raw data */ + u8 dpcd[DP_RECEIVER_CAP_SIZE]; + u8 downstream_ports[DP_MAX_DOWNSTREAM_PORTS]; + + struct msm_dp_link_info link_info; + struct drm_connector *connector; + struct msm_dp_display_mode msm_dp_mode; + struct msm_dp_panel_psr psr_cap; + bool video_test; + bool vsc_sdp_supported; + u32 hw_revision; + + u32 max_bw_code; +}; + +int msm_dp_panel_init_panel_info(struct msm_dp_panel *msm_dp_panel, + const struct drm_display_mode *adjusted_mode, + u32 bpp); +int msm_dp_panel_deinit(struct msm_dp_panel *msm_dp_panel); +int msm_dp_panel_timing_cfg(struct msm_dp_panel *msm_dp_panel, bool wide_bus_en); +int msm_dp_panel_read_link_caps(struct msm_dp_panel *msm_dp_panel, + struct drm_connector *connector); +u32 msm_dp_panel_get_mode_bpp(struct msm_dp_panel *msm_dp_panel, u32 mode_max_bpp, + u32 mode_pclk_khz); +void msm_dp_panel_handle_sink_request(struct msm_dp_panel *msm_dp_panel, + const struct drm_edid *drm_edid); +void msm_dp_panel_tpg_config(struct msm_dp_panel *msm_dp_panel, bool enable); + +void msm_dp_panel_clear_dsc_dto(struct msm_dp_panel *msm_dp_panel); + +void msm_dp_panel_enable_vsc_sdp(struct msm_dp_panel *msm_dp_panel, struct dp_sdp *vsc_sdp); +void msm_dp_panel_disable_vsc_sdp(struct msm_dp_panel *msm_dp_panel); + +/** + * is_link_rate_valid() - validates the link rate + * @bw_code: link rate requested by the sink + * + * Returns: true if the requested link rate is supported. + */ +static inline bool is_link_rate_valid(u32 bw_code) +{ + return (bw_code == DP_LINK_BW_1_62 || + bw_code == DP_LINK_BW_2_7 || + bw_code == DP_LINK_BW_5_4 || + bw_code == DP_LINK_BW_8_1); +} + +/** + * is_lane_count_valid() - validates the lane count + * @lane_count: lane count requested by the sink + * + * Returns: true if the requested lane count is supported. + */ +static inline bool is_lane_count_valid(u32 lane_count) +{ + return (lane_count == 1 || + lane_count == 2 || + lane_count == 4); +} + +struct msm_dp_panel *msm_dp_panel_get(struct device *dev, struct drm_dp_aux *aux, + struct msm_dp_link *link, + void __iomem *link_base, + void __iomem *p0_base); +#endif /* _DP_PANEL_H_ */ diff --git a/drivers/gpu/drm/msm/dp/dp_reg.h b/drivers/gpu/drm/msm/dp/dp_reg.h new file mode 100644 index 000000000..dda4b642c --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_reg.h @@ -0,0 +1,368 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved. + */ + +#ifndef _DP_REG_H_ +#define _DP_REG_H_ + +#include +#include + +/* DP_TX Registers */ +#define REG_DP_HW_VERSION (0x00000000) +#define DP_HW_VERSION_1_0 0x10000000 +#define DP_HW_VERSION_1_2 0x10020000 + +#define REG_DP_SW_RESET (0x00000010) +#define DP_SW_RESET (0x00000001) + +#define REG_DP_PHY_CTRL (0x00000014) +#define DP_PHY_CTRL_SW_RESET_PLL (0x00000001) +#define DP_PHY_CTRL_SW_RESET (0x00000004) + +#define REG_DP_CLK_CTRL (0x00000018) +#define REG_DP_CLK_ACTIVE (0x0000001C) + +#define REG_DP_INTR_STATUS (0x00000020) +#define DP_INTR_HPD BIT(0) +#define DP_INTR_AUX_XFER_DONE BIT(3) +#define DP_INTR_WRONG_ADDR BIT(6) +#define DP_INTR_TIMEOUT BIT(9) +#define DP_INTR_NACK_DEFER BIT(12) +#define DP_INTR_WRONG_DATA_CNT BIT(15) +#define DP_INTR_I2C_NACK BIT(18) +#define DP_INTR_I2C_DEFER BIT(21) +#define DP_INTR_PLL_UNLOCKED BIT(24) +#define DP_INTR_AUX_ERROR BIT(27) + +#define REG_DP_INTR_STATUS2 (0x00000024) +#define DP_INTR_READY_FOR_VIDEO BIT(0) +#define DP_INTR_IDLE_PATTERN_SENT BIT(3) +#define DP_INTR_FRAME_END BIT(6) +#define DP_INTR_CRC_UPDATED BIT(9) + +#define REG_DP_INTR_STATUS3 (0x00000028) + +#define REG_DP_INTR_STATUS4 (0x0000002C) +#define PSR_UPDATE_INT (0x00000001) +#define PSR_CAPTURE_INT (0x00000004) +#define PSR_EXIT_INT (0x00000010) +#define PSR_UPDATE_ERROR_INT (0x00000040) +#define PSR_WAKE_ERROR_INT (0x00000100) + +#define REG_DP_INTR_MASK4 (0x00000030) +#define PSR_UPDATE_MASK (0x00000001) +#define PSR_CAPTURE_MASK (0x00000002) +#define PSR_EXIT_MASK (0x00000004) +#define PSR_UPDATE_ERROR_MASK (0x00000008) +#define PSR_WAKE_ERROR_MASK (0x00000010) + +#define REG_DP_DP_HPD_CTRL (0x00000000) +#define DP_DP_HPD_CTRL_HPD_EN (0x00000001) + +#define REG_DP_DP_HPD_INT_STATUS (0x00000004) + +#define REG_DP_DP_HPD_INT_ACK (0x00000008) +#define DP_DP_HPD_PLUG_INT_ACK (0x00000001) +#define DP_DP_IRQ_HPD_INT_ACK (0x00000002) +#define DP_DP_HPD_REPLUG_INT_ACK (0x00000004) +#define DP_DP_HPD_UNPLUG_INT_ACK (0x00000008) +#define DP_DP_HPD_STATE_STATUS_BITS_MASK (0x00000007) +#define DP_DP_HPD_STATE_STATUS_BITS_SHIFT (0x1D) + +#define REG_DP_DP_HPD_INT_MASK (0x0000000C) +#define DP_DP_HPD_PLUG_INT_MASK (0x00000001) +#define DP_DP_IRQ_HPD_INT_MASK (0x00000002) +#define DP_DP_HPD_REPLUG_INT_MASK (0x00000004) +#define DP_DP_HPD_UNPLUG_INT_MASK (0x00000008) +#define DP_DP_HPD_INT_MASK (DP_DP_HPD_PLUG_INT_MASK | \ + DP_DP_IRQ_HPD_INT_MASK | \ + DP_DP_HPD_REPLUG_INT_MASK | \ + DP_DP_HPD_UNPLUG_INT_MASK) +#define DP_DP_HPD_STATE_STATUS_CONNECTED (0x40000000) +#define DP_DP_HPD_STATE_STATUS_PENDING (0x20000000) +#define DP_DP_HPD_STATE_STATUS_DISCONNECTED (0x00000000) +#define DP_DP_HPD_STATE_STATUS_MASK (0xE0000000) + +#define REG_DP_DP_HPD_REFTIMER (0x00000018) +#define DP_DP_HPD_REFTIMER_ENABLE (1 << 16) + +#define REG_DP_DP_HPD_EVENT_TIME_0 (0x0000001C) +#define REG_DP_DP_HPD_EVENT_TIME_1 (0x00000020) +#define DP_DP_HPD_EVENT_TIME_0_VAL (0x3E800FA) +#define DP_DP_HPD_EVENT_TIME_1_VAL (0x1F407D0) + +#define REG_DP_AUX_CTRL (0x00000030) +#define DP_AUX_CTRL_ENABLE (0x00000001) +#define DP_AUX_CTRL_RESET (0x00000002) + +#define REG_DP_AUX_DATA (0x00000034) +#define DP_AUX_DATA_READ (0x00000001) +#define DP_AUX_DATA_WRITE (0x00000000) +#define DP_AUX_DATA_OFFSET (0x00000008) +#define DP_AUX_DATA_INDEX_OFFSET (0x00000010) +#define DP_AUX_DATA_MASK (0x0000ff00) +#define DP_AUX_DATA_INDEX_WRITE (0x80000000) + +#define REG_DP_AUX_TRANS_CTRL (0x00000038) +#define DP_AUX_TRANS_CTRL_I2C (0x00000100) +#define DP_AUX_TRANS_CTRL_GO (0x00000200) +#define DP_AUX_TRANS_CTRL_NO_SEND_ADDR (0x00000400) +#define DP_AUX_TRANS_CTRL_NO_SEND_STOP (0x00000800) + +#define REG_DP_TIMEOUT_COUNT (0x0000003C) +#define REG_DP_AUX_LIMITS (0x00000040) +#define REG_DP_AUX_STATUS (0x00000044) + +#define DP_DPCD_CP_IRQ (0x201) +#define DP_DPCD_RXSTATUS (0x69493) + +#define DP_INTERRUPT_TRANS_NUM (0x000000A0) + +#define REG_DP_MAINLINK_CTRL (0x00000000) +#define DP_MAINLINK_CTRL_ENABLE (0x00000001) +#define DP_MAINLINK_CTRL_RESET (0x00000002) +#define DP_MAINLINK_CTRL_SW_BYPASS_SCRAMBLER (0x00000010) +#define DP_MAINLINK_CTRL_FLUSH_MODE_MASK GENMASK(24, 23) +#define DP_MAINLINK_FLUSH_MODE_UPDATE_SDP FIELD_PREP(DP_MAINLINK_CTRL_FLUSH_MODE_MASK, 1) +#define DP_MAINLINK_FLUSH_MODE_SDE_PERIPH_UPDATE FIELD_PREP(DP_MAINLINK_CTRL_FLUSH_MODE_MASK, 3) +#define DP_MAINLINK_FB_BOUNDARY_SEL (0x02000000) + +#define REG_DP_STATE_CTRL (0x00000004) +#define DP_STATE_CTRL_LINK_TRAINING_PATTERN1 (0x00000001) +#define DP_STATE_CTRL_LINK_TRAINING_PATTERN2 (0x00000002) +#define DP_STATE_CTRL_LINK_TRAINING_PATTERN3 (0x00000004) +#define DP_STATE_CTRL_LINK_TRAINING_PATTERN4 (0x00000008) +#define DP_STATE_CTRL_LINK_SYMBOL_ERR_MEASURE (0x00000010) +#define DP_STATE_CTRL_LINK_PRBS7 (0x00000020) +#define DP_STATE_CTRL_LINK_TEST_CUSTOM_PATTERN (0x00000040) +#define DP_STATE_CTRL_SEND_VIDEO (0x00000080) +#define DP_STATE_CTRL_PUSH_IDLE (0x00000100) + +#define REG_DP_CONFIGURATION_CTRL (0x00000008) +#define DP_CONFIGURATION_CTRL_SYNC_ASYNC_CLK (0x00000001) +#define DP_CONFIGURATION_CTRL_STATIC_DYNAMIC_CN (0x00000002) +#define DP_CONFIGURATION_CTRL_P_INTERLACED (0x00000004) +#define DP_CONFIGURATION_CTRL_INTERLACED_BTF (0x00000008) +#define DP_CONFIGURATION_CTRL_NUM_OF_LANES (0x00000010) +#define DP_CONFIGURATION_CTRL_ENHANCED_FRAMING (0x00000040) +#define DP_CONFIGURATION_CTRL_SEND_VSC (0x00000080) +#define DP_CONFIGURATION_CTRL_BPC (0x00000100) +#define DP_CONFIGURATION_CTRL_BPC_MASK GENMASK(9, 8) +#define DP_CONFIGURATION_CTRL_ASSR (0x00000400) +#define DP_CONFIGURATION_CTRL_RGB_YUV (0x00000800) +#define DP_CONFIGURATION_CTRL_RGB_YUV_MASK GENMASK(12, 11) +#define DP_CONFIGURATION_CTRL_LSCLK_DIV (0x00002000) +#define DP_CONFIGURATION_CTRL_NUM_OF_LANES_SHIFT (0x04) +#define DP_CONFIGURATION_CTRL_BPC_SHIFT (0x08) +#define DP_CONFIGURATION_CTRL_LSCLK_DIV_SHIFT (0x0D) + +#define REG_DP_SOFTWARE_MVID (0x00000010) +#define REG_DP_SOFTWARE_NVID (0x00000018) +#define REG_DP_TOTAL_HOR_VER (0x0000001C) +#define REG_DP_START_HOR_VER_FROM_SYNC (0x00000020) +#define REG_DP_HSYNC_VSYNC_WIDTH_POLARITY (0x00000024) +#define REG_DP_ACTIVE_HOR_VER (0x00000028) + +#define REG_DP_MISC1_MISC0 (0x0000002C) +#define DP_MISC0_SYNCHRONOUS_CLK (0x00000001) +#define DP_MISC0_COLORIMETRY_CFG_SHIFT (0x00000001) +#define DP_MISC0_TEST_BITS_DEPTH_SHIFT (0x00000005) +#define DP_MISC1_VSC_SDP (0x00004000) + +#define DP_MISC0_COLORIMERY_CFG_LEGACY_RGB (0) +#define DP_MISC0_COLORIMERY_CFG_CEA_RGB (0x04) + +#define REG_DP_VALID_BOUNDARY (0x00000030) +#define REG_DP_VALID_BOUNDARY_2 (0x00000034) + +#define REG_DP_LOGICAL2PHYSICAL_LANE_MAPPING (0x00000038) +#define LANE0_MAPPING_SHIFT (0x00000000) +#define LANE1_MAPPING_SHIFT (0x00000002) +#define LANE2_MAPPING_SHIFT (0x00000004) +#define LANE3_MAPPING_SHIFT (0x00000006) + +#define REG_DP_MAINLINK_READY (0x00000040) +#define DP_MAINLINK_READY_FOR_VIDEO (0x00000001) +#define DP_MAINLINK_READY_LINK_TRAINING_SHIFT (0x00000003) + +#define REG_DP_MAINLINK_LEVELS (0x00000044) +#define DP_MAINLINK_SAFE_TO_EXIT_LEVEL_2 (0x00000002) + + +#define REG_DP_TU (0x0000004C) + +#define REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET (0x00000054) +#define DP_HBR2_ERM_PATTERN (0x00010000) + +#define REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG0 (0x000000C0) +#define REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG1 (0x000000C4) +#define REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG2 (0x000000C8) + +#define MMSS_DP_MISC1_MISC0 (0x0000002C) +#define MMSS_DP_AUDIO_TIMING_GEN (0x00000080) +#define MMSS_DP_AUDIO_TIMING_RBR_32 (0x00000084) +#define MMSS_DP_AUDIO_TIMING_HBR_32 (0x00000088) +#define MMSS_DP_AUDIO_TIMING_RBR_44 (0x0000008C) +#define MMSS_DP_AUDIO_TIMING_HBR_44 (0x00000090) +#define MMSS_DP_AUDIO_TIMING_RBR_48 (0x00000094) +#define MMSS_DP_AUDIO_TIMING_HBR_48 (0x00000098) + +#define REG_PSR_CONFIG (0x00000100) +#define DISABLE_PSR (0x00000000) +#define PSR1_SUPPORTED (0x00000001) +#define PSR2_WITHOUT_FRAMESYNC (0x00000002) +#define PSR2_WITH_FRAMESYNC (0x00000003) + +#define REG_PSR_CMD (0x00000110) +#define PSR_ENTER (0x00000001) +#define PSR_EXIT (0x00000002) + +#define MMSS_DP_PSR_CRC_RG (0x00000154) +#define MMSS_DP_PSR_CRC_B (0x00000158) + +#define REG_DP_COMPRESSION_MODE_CTRL (0x00000180) + +#define MMSS_DP_AUDIO_CFG (0x00000200) +#define MMSS_DP_AUDIO_STATUS (0x00000204) +#define MMSS_DP_AUDIO_PKT_CTRL (0x00000208) +#define MMSS_DP_AUDIO_PKT_CTRL2 (0x0000020C) +#define MMSS_DP_AUDIO_ACR_CTRL (0x00000210) +#define MMSS_DP_AUDIO_CTRL_RESET (0x00000214) + +#define MMSS_DP_SDP_CFG (0x00000228) +#define GEN0_SDP_EN (0x00020000) +#define MMSS_DP_SDP_CFG2 (0x0000022C) +#define MMSS_DP_AUDIO_TIMESTAMP_0 (0x00000230) +#define MMSS_DP_AUDIO_TIMESTAMP_1 (0x00000234) +#define GENERIC0_SDPSIZE_VALID (0x00010000) + +#define MMSS_DP_AUDIO_STREAM_0 (0x00000240) +#define MMSS_DP_AUDIO_STREAM_1 (0x00000244) + +#define MMSS_DP_SDP_CFG3 (0x0000024c) +#define UPDATE_SDP (0x00000001) + +#define MMSS_DP_EXTENSION_0 (0x00000250) +#define MMSS_DP_EXTENSION_1 (0x00000254) +#define MMSS_DP_EXTENSION_2 (0x00000258) +#define MMSS_DP_EXTENSION_3 (0x0000025C) +#define MMSS_DP_EXTENSION_4 (0x00000260) +#define MMSS_DP_EXTENSION_5 (0x00000264) +#define MMSS_DP_EXTENSION_6 (0x00000268) +#define MMSS_DP_EXTENSION_7 (0x0000026C) +#define MMSS_DP_EXTENSION_8 (0x00000270) +#define MMSS_DP_EXTENSION_9 (0x00000274) +#define MMSS_DP_AUDIO_COPYMANAGEMENT_0 (0x00000278) +#define MMSS_DP_AUDIO_COPYMANAGEMENT_1 (0x0000027C) +#define MMSS_DP_AUDIO_COPYMANAGEMENT_2 (0x00000280) +#define MMSS_DP_AUDIO_COPYMANAGEMENT_3 (0x00000284) +#define MMSS_DP_AUDIO_COPYMANAGEMENT_4 (0x00000288) +#define MMSS_DP_AUDIO_COPYMANAGEMENT_5 (0x0000028C) +#define MMSS_DP_AUDIO_ISRC_0 (0x00000290) +#define MMSS_DP_AUDIO_ISRC_1 (0x00000294) +#define MMSS_DP_AUDIO_ISRC_2 (0x00000298) +#define MMSS_DP_AUDIO_ISRC_3 (0x0000029C) +#define MMSS_DP_AUDIO_ISRC_4 (0x000002A0) +#define MMSS_DP_AUDIO_ISRC_5 (0x000002A4) +#define MMSS_DP_AUDIO_INFOFRAME_0 (0x000002A8) +#define MMSS_DP_AUDIO_INFOFRAME_1 (0x000002AC) +#define MMSS_DP_AUDIO_INFOFRAME_2 (0x000002B0) + +#define MMSS_DP_GENERIC0_0 (0x00000300) +#define MMSS_DP_GENERIC0_1 (0x00000304) +#define MMSS_DP_GENERIC0_2 (0x00000308) +#define MMSS_DP_GENERIC0_3 (0x0000030C) +#define MMSS_DP_GENERIC0_4 (0x00000310) +#define MMSS_DP_GENERIC0_5 (0x00000314) +#define MMSS_DP_GENERIC0_6 (0x00000318) +#define MMSS_DP_GENERIC0_7 (0x0000031C) +#define MMSS_DP_GENERIC0_8 (0x00000320) +#define MMSS_DP_GENERIC0_9 (0x00000324) +#define MMSS_DP_GENERIC1_0 (0x00000328) +#define MMSS_DP_GENERIC1_1 (0x0000032C) +#define MMSS_DP_GENERIC1_2 (0x00000330) +#define MMSS_DP_GENERIC1_3 (0x00000334) +#define MMSS_DP_GENERIC1_4 (0x00000338) +#define MMSS_DP_GENERIC1_5 (0x0000033C) +#define MMSS_DP_GENERIC1_6 (0x00000340) +#define MMSS_DP_GENERIC1_7 (0x00000344) +#define MMSS_DP_GENERIC1_8 (0x00000348) +#define MMSS_DP_GENERIC1_9 (0x0000034C) + +#define MMSS_DP_VSCEXT_0 (0x000002D0) +#define MMSS_DP_VSCEXT_1 (0x000002D4) +#define MMSS_DP_VSCEXT_2 (0x000002D8) +#define MMSS_DP_VSCEXT_3 (0x000002DC) +#define MMSS_DP_VSCEXT_4 (0x000002E0) +#define MMSS_DP_VSCEXT_5 (0x000002E4) +#define MMSS_DP_VSCEXT_6 (0x000002E8) +#define MMSS_DP_VSCEXT_7 (0x000002EC) +#define MMSS_DP_VSCEXT_8 (0x000002F0) +#define MMSS_DP_VSCEXT_9 (0x000002F4) + +#define MMSS_DP_BIST_ENABLE (0x00000000) +#define DP_BIST_ENABLE_DPBIST_EN (0x00000001) + +#define MMSS_DP_TIMING_ENGINE_EN (0x00000010) +#define DP_TIMING_ENGINE_EN_EN (0x00000001) + +#define MMSS_DP_INTF_CONFIG (0x00000014) +#define MMSS_DP_INTF_HSYNC_CTL (0x00000018) +#define MMSS_DP_INTF_VSYNC_PERIOD_F0 (0x0000001C) +#define MMSS_DP_INTF_VSYNC_PERIOD_F1 (0x00000020) +#define MMSS_DP_INTF_VSYNC_PULSE_WIDTH_F0 (0x00000024) +#define MMSS_DP_INTF_VSYNC_PULSE_WIDTH_F1 (0x00000028) +#define MMSS_INTF_DISPLAY_V_START_F0 (0x0000002C) +#define MMSS_INTF_DISPLAY_V_START_F1 (0x00000030) +#define MMSS_DP_INTF_DISPLAY_V_END_F0 (0x00000034) +#define MMSS_DP_INTF_DISPLAY_V_END_F1 (0x00000038) +#define MMSS_DP_INTF_ACTIVE_V_START_F0 (0x0000003C) +#define MMSS_DP_INTF_ACTIVE_V_START_F1 (0x00000040) +#define MMSS_DP_INTF_ACTIVE_V_END_F0 (0x00000044) +#define MMSS_DP_INTF_ACTIVE_V_END_F1 (0x00000048) +#define MMSS_DP_INTF_DISPLAY_HCTL (0x0000004C) +#define MMSS_DP_INTF_ACTIVE_HCTL (0x00000050) +#define MMSS_DP_INTF_POLARITY_CTL (0x00000058) + +#define MMSS_DP_TPG_MAIN_CONTROL (0x00000060) +#define MMSS_DP_DSC_DTO (0x0000007C) +#define DP_TPG_CHECKERED_RECT_PATTERN (0x00000100) + +#define MMSS_DP_TPG_VIDEO_CONFIG (0x00000064) +#define DP_TPG_VIDEO_CONFIG_BPP_8BIT (0x00000001) +#define DP_TPG_VIDEO_CONFIG_RGB (0x00000004) + +#define MMSS_DP_ASYNC_FIFO_CONFIG (0x00000088) + +#define REG_DP_PHY_AUX_INTERRUPT_CLEAR (0x0000004C) +#define REG_DP_PHY_AUX_BIST_CFG (0x00000050) +#define REG_DP_PHY_AUX_INTERRUPT_STATUS (0x000000BC) + +/* DP HDCP 1.3 registers */ +#define DP_HDCP_CTRL (0x0A0) +#define DP_HDCP_STATUS (0x0A4) +#define DP_HDCP_SW_UPPER_AKSV (0x098) +#define DP_HDCP_SW_LOWER_AKSV (0x09C) +#define DP_HDCP_ENTROPY_CTRL0 (0x350) +#define DP_HDCP_ENTROPY_CTRL1 (0x35C) +#define DP_HDCP_SHA_STATUS (0x0C8) +#define DP_HDCP_RCVPORT_DATA2_0 (0x0B0) +#define DP_HDCP_RCVPORT_DATA3 (0x0A4) +#define DP_HDCP_RCVPORT_DATA4 (0x0A8) +#define DP_HDCP_RCVPORT_DATA5 (0x0C0) +#define DP_HDCP_RCVPORT_DATA6 (0x0C4) + +#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_SHA_CTRL (0x024) +#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_SHA_DATA (0x028) +#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA0 (0x004) +#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA1 (0x008) +#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA7 (0x00C) +#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA8 (0x010) +#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA9 (0x014) +#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA10 (0x018) +#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA11 (0x01C) +#define HDCP_SEC_DP_TZ_HV_HLOS_HDCP_RCVPORT_DATA12 (0x020) + +#endif /* _DP_REG_H_ */ diff --git a/drivers/gpu/drm/msm/dp/dp_utils.c b/drivers/gpu/drm/msm/dp/dp_utils.c new file mode 100644 index 000000000..4a5ebb0c3 --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_utils.c @@ -0,0 +1,88 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2024, The Linux Foundation. All rights reserved. + */ + +#include + +#include "dp_utils.h" + +#define DP_SDP_HEADER_SIZE 8 + +u8 msm_dp_utils_get_g0_value(u8 data) +{ + u8 c[4]; + u8 g[4]; + u8 ret_data = 0; + u8 i; + + for (i = 0; i < 4; i++) + c[i] = (data >> i) & 0x01; + + g[0] = c[3]; + g[1] = c[0] ^ c[3]; + g[2] = c[1]; + g[3] = c[2]; + + for (i = 0; i < 4; i++) + ret_data = ((g[i] & 0x01) << i) | ret_data; + + return ret_data; +} + +u8 msm_dp_utils_get_g1_value(u8 data) +{ + u8 c[4]; + u8 g[4]; + u8 ret_data = 0; + u8 i; + + for (i = 0; i < 4; i++) + c[i] = (data >> i) & 0x01; + + g[0] = c[0] ^ c[3]; + g[1] = c[0] ^ c[1] ^ c[3]; + g[2] = c[1] ^ c[2]; + g[3] = c[2] ^ c[3]; + + for (i = 0; i < 4; i++) + ret_data = ((g[i] & 0x01) << i) | ret_data; + + return ret_data; +} + +u8 msm_dp_utils_calculate_parity(u32 data) +{ + u8 x0 = 0; + u8 x1 = 0; + u8 ci = 0; + u8 iData = 0; + u8 i = 0; + u8 parity_byte; + u8 num_byte = (data & 0xFF00) > 0 ? 8 : 2; + + for (i = 0; i < num_byte; i++) { + iData = (data >> i * 4) & 0xF; + + ci = iData ^ x1; + x1 = x0 ^ msm_dp_utils_get_g1_value(ci); + x0 = msm_dp_utils_get_g0_value(ci); + } + + parity_byte = x1 | (x0 << 4); + + return parity_byte; +} + +void msm_dp_utils_pack_sdp_header(struct dp_sdp_header *sdp_header, u32 header_buff[2]) +{ + header_buff[0] = FIELD_PREP(HEADER_0_MASK, sdp_header->HB0) | + FIELD_PREP(PARITY_0_MASK, msm_dp_utils_calculate_parity(sdp_header->HB0)) | + FIELD_PREP(HEADER_1_MASK, sdp_header->HB1) | + FIELD_PREP(PARITY_1_MASK, msm_dp_utils_calculate_parity(sdp_header->HB1)); + + header_buff[1] = FIELD_PREP(HEADER_2_MASK, sdp_header->HB2) | + FIELD_PREP(PARITY_2_MASK, msm_dp_utils_calculate_parity(sdp_header->HB2)) | + FIELD_PREP(HEADER_3_MASK, sdp_header->HB3) | + FIELD_PREP(PARITY_3_MASK, msm_dp_utils_calculate_parity(sdp_header->HB3)); +} diff --git a/drivers/gpu/drm/msm/dp/dp_utils.h b/drivers/gpu/drm/msm/dp/dp_utils.h new file mode 100644 index 000000000..2e4f98a86 --- /dev/null +++ b/drivers/gpu/drm/msm/dp/dp_utils.h @@ -0,0 +1,36 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2024, The Linux Foundation. All rights reserved. + */ + +#ifndef _DP_UTILS_H_ +#define _DP_UTILS_H_ + +#include +#include +#include + +#define HEADER_BYTE_0_BIT 0 +#define PARITY_BYTE_0_BIT 8 +#define HEADER_BYTE_1_BIT 16 +#define PARITY_BYTE_1_BIT 24 +#define HEADER_BYTE_2_BIT 0 +#define PARITY_BYTE_2_BIT 8 +#define HEADER_BYTE_3_BIT 16 +#define PARITY_BYTE_3_BIT 24 + +#define HEADER_0_MASK GENMASK(7, 0) +#define PARITY_0_MASK GENMASK(15, 8) +#define HEADER_1_MASK GENMASK(23, 16) +#define PARITY_1_MASK GENMASK(31, 24) +#define HEADER_2_MASK GENMASK(7, 0) +#define PARITY_2_MASK GENMASK(15, 8) +#define HEADER_3_MASK GENMASK(23, 16) +#define PARITY_3_MASK GENMASK(31, 24) + +u8 msm_dp_utils_get_g0_value(u8 data); +u8 msm_dp_utils_get_g1_value(u8 data); +u8 msm_dp_utils_calculate_parity(u32 data); +void msm_dp_utils_pack_sdp_header(struct dp_sdp_header *sdp_header, u32 header_buff[2]); + +#endif /* _DP_UTILS_H_ */ diff --git a/drivers/gpu/drm/msm/dsi/dsi.c b/drivers/gpu/drm/msm/dsi/dsi.c new file mode 100644 index 000000000..3c9f01ed6 --- /dev/null +++ b/drivers/gpu/drm/msm/dsi/dsi.c @@ -0,0 +1,270 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2015, The Linux Foundation. All rights reserved. + */ + +#include "dsi.h" + +bool msm_dsi_is_cmd_mode(struct msm_dsi *msm_dsi) +{ + unsigned long host_flags = msm_dsi_host_get_mode_flags(msm_dsi->host); + + return !(host_flags & MIPI_DSI_MODE_VIDEO); +} + +struct drm_dsc_config *msm_dsi_get_dsc_config(struct msm_dsi *msm_dsi) +{ + return msm_dsi_host_get_dsc_config(msm_dsi->host); +} + +bool msm_dsi_wide_bus_enabled(struct msm_dsi *msm_dsi) +{ + return msm_dsi_host_is_wide_bus_enabled(msm_dsi->host); +} + +static int dsi_get_phy(struct msm_dsi *msm_dsi) +{ + struct platform_device *pdev = msm_dsi->pdev; + struct platform_device *phy_pdev; + struct device_node *phy_node; + + phy_node = of_parse_phandle(pdev->dev.of_node, "phys", 0); + if (!phy_node) { + DRM_DEV_ERROR(&pdev->dev, "cannot find phy device\n"); + return -ENXIO; + } + + phy_pdev = of_find_device_by_node(phy_node); + if (phy_pdev) { + msm_dsi->phy = platform_get_drvdata(phy_pdev); + msm_dsi->phy_dev = &phy_pdev->dev; + } + + of_node_put(phy_node); + + if (!phy_pdev) { + DRM_DEV_ERROR(&pdev->dev, "%s: phy driver is not ready\n", __func__); + return -EPROBE_DEFER; + } + if (!msm_dsi->phy) { + put_device(&phy_pdev->dev); + DRM_DEV_ERROR(&pdev->dev, "%s: phy driver is not ready\n", __func__); + return -EPROBE_DEFER; + } + + return 0; +} + +static void dsi_destroy(struct msm_dsi *msm_dsi) +{ + if (!msm_dsi) + return; + + msm_dsi_manager_unregister(msm_dsi); + + if (msm_dsi->phy_dev) { + put_device(msm_dsi->phy_dev); + msm_dsi->phy = NULL; + msm_dsi->phy_dev = NULL; + } + + if (msm_dsi->host) { + msm_dsi_host_destroy(msm_dsi->host); + msm_dsi->host = NULL; + } + + platform_set_drvdata(msm_dsi->pdev, NULL); +} + +static struct msm_dsi *dsi_init(struct platform_device *pdev) +{ + struct msm_dsi *msm_dsi; + int ret; + + if (!pdev) + return ERR_PTR(-ENXIO); + + msm_dsi = devm_kzalloc(&pdev->dev, sizeof(*msm_dsi), GFP_KERNEL); + if (!msm_dsi) + return ERR_PTR(-ENOMEM); + DBG("dsi probed=%p", msm_dsi); + + msm_dsi->id = -1; + msm_dsi->pdev = pdev; + platform_set_drvdata(pdev, msm_dsi); + + /* Init dsi host */ + ret = msm_dsi_host_init(msm_dsi); + if (ret) + goto destroy_dsi; + + /* GET dsi PHY */ + ret = dsi_get_phy(msm_dsi); + if (ret) + goto destroy_dsi; + + /* Register to dsi manager */ + ret = msm_dsi_manager_register(msm_dsi); + if (ret) + goto destroy_dsi; + + return msm_dsi; + +destroy_dsi: + dsi_destroy(msm_dsi); + return ERR_PTR(ret); +} + +static int dsi_bind(struct device *dev, struct device *master, void *data) +{ + struct msm_drm_private *priv = dev_get_drvdata(master); + struct msm_dsi *msm_dsi = dev_get_drvdata(dev); + + /* + * Next bridge doesn't exist for the secondary DSI host in a bonded + * pair. + */ + if (!msm_dsi_is_bonded_dsi(msm_dsi) || + msm_dsi_is_master_dsi(msm_dsi)) { + struct drm_bridge *ext_bridge; + + ext_bridge = devm_drm_of_get_bridge(&msm_dsi->pdev->dev, + msm_dsi->pdev->dev.of_node, 1, 0); + if (IS_ERR(ext_bridge)) + return PTR_ERR(ext_bridge); + + msm_dsi->next_bridge = ext_bridge; + } + + priv->kms->dsi[msm_dsi->id] = msm_dsi; + + return 0; +} + +static void dsi_unbind(struct device *dev, struct device *master, + void *data) +{ + struct msm_drm_private *priv = dev_get_drvdata(master); + struct msm_dsi *msm_dsi = dev_get_drvdata(dev); + + msm_dsi_tx_buf_free(msm_dsi->host); + priv->kms->dsi[msm_dsi->id] = NULL; +} + +static const struct component_ops dsi_ops = { + .bind = dsi_bind, + .unbind = dsi_unbind, +}; + +int dsi_dev_attach(struct platform_device *pdev) +{ + return component_add(&pdev->dev, &dsi_ops); +} + +void dsi_dev_detach(struct platform_device *pdev) +{ + component_del(&pdev->dev, &dsi_ops); +} + +static int dsi_dev_probe(struct platform_device *pdev) +{ + struct msm_dsi *msm_dsi; + + DBG(""); + msm_dsi = dsi_init(pdev); + if (IS_ERR(msm_dsi)) { + /* Don't fail the bind if the dsi port is not connected */ + if (PTR_ERR(msm_dsi) == -ENODEV) + return 0; + else + return PTR_ERR(msm_dsi); + } + + return 0; +} + +static void dsi_dev_remove(struct platform_device *pdev) +{ + struct msm_dsi *msm_dsi = platform_get_drvdata(pdev); + + DBG(""); + dsi_destroy(msm_dsi); +} + +static const struct of_device_id dt_match[] = { + { .compatible = "qcom,mdss-dsi-ctrl" }, + + /* Deprecated, don't use */ + { .compatible = "qcom,dsi-ctrl-6g-qcm2290" }, + {} +}; +MODULE_DEVICE_TABLE(of, dt_match); + +static const struct dev_pm_ops dsi_pm_ops = { + SET_RUNTIME_PM_OPS(msm_dsi_runtime_suspend, msm_dsi_runtime_resume, NULL) + SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, + pm_runtime_force_resume) +}; + +static struct platform_driver dsi_driver = { + .probe = dsi_dev_probe, + .remove = dsi_dev_remove, + .driver = { + .name = "msm_dsi", + .of_match_table = dt_match, + .pm = &dsi_pm_ops, + }, +}; + +void __init msm_dsi_register(void) +{ + DBG(""); + msm_dsi_phy_driver_register(); + platform_driver_register(&dsi_driver); +} + +void __exit msm_dsi_unregister(void) +{ + DBG(""); + msm_dsi_phy_driver_unregister(); + platform_driver_unregister(&dsi_driver); +} + +int msm_dsi_modeset_init(struct msm_dsi *msm_dsi, struct drm_device *dev, + struct drm_encoder *encoder) +{ + int ret; + + msm_dsi->dev = dev; + + ret = msm_dsi_host_modeset_init(msm_dsi->host, dev); + if (ret) { + DRM_DEV_ERROR(dev->dev, "failed to modeset init host: %d\n", ret); + return ret; + } + + if (msm_dsi_is_bonded_dsi(msm_dsi) && + !msm_dsi_is_master_dsi(msm_dsi)) { + /* + * Do not return an eror here, + * Just skip creating encoder/connector for the slave-DSI. + */ + return 0; + } + + ret = msm_dsi_manager_connector_init(msm_dsi, encoder); + if (ret) { + DRM_DEV_ERROR(dev->dev, + "failed to create dsi connector: %d\n", ret); + return ret; + } + + return 0; +} + +void msm_dsi_snapshot(struct msm_disp_state *disp_state, struct msm_dsi *msm_dsi) +{ + msm_dsi_host_snapshot(disp_state, msm_dsi->host); + msm_dsi_phy_snapshot(disp_state, msm_dsi->phy); +} + diff --git a/drivers/gpu/drm/msm/dsi/dsi.h b/drivers/gpu/drm/msm/dsi/dsi.h new file mode 100644 index 000000000..93c028a12 --- /dev/null +++ b/drivers/gpu/drm/msm/dsi/dsi.h @@ -0,0 +1,155 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2015, The Linux Foundation. All rights reserved. + */ + +#ifndef __DSI_CONNECTOR_H__ +#define __DSI_CONNECTOR_H__ + +#include +#include + +#include +#include +#include + +#include "msm_drv.h" +#include "disp/msm_disp_snapshot.h" + +#define DSI_0 0 +#define DSI_1 1 +#define DSI_MAX 2 + +struct msm_dsi_phy_shared_timings; +struct msm_dsi_phy_clk_request; + +enum msm_dsi_phy_usecase { + MSM_DSI_PHY_STANDALONE, + MSM_DSI_PHY_MASTER, + MSM_DSI_PHY_SLAVE, +}; + +#define DSI_BUS_CLK_MAX 4 + +struct msm_dsi { + struct drm_device *dev; + struct platform_device *pdev; + + struct mipi_dsi_host *host; + struct msm_dsi_phy *phy; + const char *te_source; + + struct drm_bridge *next_bridge; + + struct device *phy_dev; + bool phy_enabled; + + int id; +}; + +/* dsi manager */ +int msm_dsi_manager_connector_init(struct msm_dsi *msm_dsi, + struct drm_encoder *encoder); +int msm_dsi_manager_cmd_xfer(int id, const struct mipi_dsi_msg *msg); +bool msm_dsi_manager_cmd_xfer_trigger(int id, u32 dma_base, u32 len); +int msm_dsi_manager_register(struct msm_dsi *msm_dsi); +void msm_dsi_manager_unregister(struct msm_dsi *msm_dsi); +void msm_dsi_manager_tpg_enable(void); + +/* dsi host */ +struct msm_dsi_host; +int msm_dsi_host_xfer_prepare(struct mipi_dsi_host *host, + const struct mipi_dsi_msg *msg); +void msm_dsi_host_xfer_restore(struct mipi_dsi_host *host, + const struct mipi_dsi_msg *msg); +int msm_dsi_host_cmd_tx(struct mipi_dsi_host *host, + const struct mipi_dsi_msg *msg); +int msm_dsi_host_cmd_rx(struct mipi_dsi_host *host, + const struct mipi_dsi_msg *msg); +void msm_dsi_host_cmd_xfer_commit(struct mipi_dsi_host *host, + u32 dma_base, u32 len); +int msm_dsi_host_enable(struct mipi_dsi_host *host); +int msm_dsi_host_disable(struct mipi_dsi_host *host); +void msm_dsi_host_enable_irq(struct mipi_dsi_host *host); +void msm_dsi_host_disable_irq(struct mipi_dsi_host *host); +int msm_dsi_host_power_on(struct mipi_dsi_host *host, + struct msm_dsi_phy_shared_timings *phy_shared_timings, + bool is_bonded_dsi, struct msm_dsi_phy *phy); +int msm_dsi_host_power_off(struct mipi_dsi_host *host); +int msm_dsi_host_set_display_mode(struct mipi_dsi_host *host, + const struct drm_display_mode *mode); +enum drm_mode_status msm_dsi_host_check_dsc(struct mipi_dsi_host *host, + const struct drm_display_mode *mode); +unsigned long msm_dsi_host_get_mode_flags(struct mipi_dsi_host *host); +int msm_dsi_host_register(struct mipi_dsi_host *host); +void msm_dsi_host_unregister(struct mipi_dsi_host *host); +void msm_dsi_host_set_phy_mode(struct mipi_dsi_host *host, + struct msm_dsi_phy *src_phy); +int msm_dsi_host_set_src_pll(struct mipi_dsi_host *host, + struct msm_dsi_phy *src_phy); +void msm_dsi_host_reset_phy(struct mipi_dsi_host *host); +void msm_dsi_host_get_phy_clk_req(struct mipi_dsi_host *host, + struct msm_dsi_phy_clk_request *clk_req, + bool is_bonded_dsi); +void msm_dsi_host_destroy(struct mipi_dsi_host *host); +int msm_dsi_host_modeset_init(struct mipi_dsi_host *host, + struct drm_device *dev); +int msm_dsi_host_init(struct msm_dsi *msm_dsi); +int msm_dsi_runtime_suspend(struct device *dev); +int msm_dsi_runtime_resume(struct device *dev); +int dsi_link_clk_set_rate_6g(struct msm_dsi_host *msm_host); +int dsi_link_clk_set_rate_6g_v2_9(struct msm_dsi_host *msm_host); +int dsi_link_clk_set_rate_v2(struct msm_dsi_host *msm_host); +int dsi_link_clk_enable_6g(struct msm_dsi_host *msm_host); +int dsi_link_clk_enable_v2(struct msm_dsi_host *msm_host); +void dsi_link_clk_disable_6g(struct msm_dsi_host *msm_host); +void dsi_link_clk_disable_v2(struct msm_dsi_host *msm_host); +unsigned long dsi_byte_clk_get_rate(struct mipi_dsi_host *host, bool is_bonded_dsi, + const struct drm_display_mode *mode); +int dsi_tx_buf_alloc_6g(struct msm_dsi_host *msm_host, int size); +int dsi_tx_buf_alloc_v2(struct msm_dsi_host *msm_host, int size); +void *dsi_tx_buf_get_6g(struct msm_dsi_host *msm_host); +void *dsi_tx_buf_get_v2(struct msm_dsi_host *msm_host); +void dsi_tx_buf_put_6g(struct msm_dsi_host *msm_host); +void msm_dsi_tx_buf_free(struct mipi_dsi_host *mipi_host); +int dsi_dma_base_get_6g(struct msm_dsi_host *msm_host, uint64_t *iova); +int dsi_dma_base_get_v2(struct msm_dsi_host *msm_host, uint64_t *iova); +int dsi_clk_init_v2(struct msm_dsi_host *msm_host); +int dsi_clk_init_6g_v2(struct msm_dsi_host *msm_host); +int dsi_clk_init_6g_v2_9(struct msm_dsi_host *msm_host); +int dsi_calc_clk_rate_v2(struct msm_dsi_host *msm_host, bool is_bonded_dsi); +int dsi_calc_clk_rate_6g(struct msm_dsi_host *msm_host, bool is_bonded_dsi); +void msm_dsi_host_snapshot(struct msm_disp_state *disp_state, struct mipi_dsi_host *host); +void msm_dsi_host_test_pattern_en(struct mipi_dsi_host *host); +struct drm_dsc_config *msm_dsi_host_get_dsc_config(struct mipi_dsi_host *host); +bool msm_dsi_host_is_wide_bus_enabled(struct mipi_dsi_host *host); + +/* dsi phy */ +struct msm_dsi_phy; +struct msm_dsi_phy_shared_timings { + u32 clk_post; + u32 clk_pre; + bool clk_pre_inc_by_2; + bool byte_intf_clk_div_2; +}; + +struct msm_dsi_phy_clk_request { + unsigned long bitclk_rate; + unsigned long escclk_rate; +}; + +void msm_dsi_phy_driver_register(void); +void msm_dsi_phy_driver_unregister(void); +int msm_dsi_phy_enable(struct msm_dsi_phy *phy, + struct msm_dsi_phy_clk_request *clk_req, + struct msm_dsi_phy_shared_timings *shared_timings); +void msm_dsi_phy_disable(struct msm_dsi_phy *phy); +void msm_dsi_phy_set_usecase(struct msm_dsi_phy *phy, + enum msm_dsi_phy_usecase uc); +void msm_dsi_phy_pll_save_state(struct msm_dsi_phy *phy); +int msm_dsi_phy_pll_restore_state(struct msm_dsi_phy *phy); +void msm_dsi_phy_snapshot(struct msm_disp_state *disp_state, struct msm_dsi_phy *phy); +bool msm_dsi_phy_set_continuous_clock(struct msm_dsi_phy *phy, bool enable); + +#endif /* __DSI_CONNECTOR_H__ */ + diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.c b/drivers/gpu/drm/msm/dsi/dsi_cfg.c new file mode 100644 index 000000000..da3fe6824 --- /dev/null +++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.c @@ -0,0 +1,364 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2015, The Linux Foundation. All rights reserved. + */ + +#include "dsi_cfg.h" + +static const char * const dsi_v2_bus_clk_names[] = { + "core_mmss", "iface", "bus", +}; + +static const struct regulator_bulk_data apq8064_dsi_regulators[] = { + { .supply = "vdda", .init_load_uA = 100000 }, /* 1.2 V */ + { .supply = "avdd", .init_load_uA = 10000 }, /* 3.0 V */ + { .supply = "vddio", .init_load_uA = 100000 }, /* 1.8 V */ +}; + +static const struct msm_dsi_config apq8064_dsi_cfg = { + .io_offset = 0, + .regulator_data = apq8064_dsi_regulators, + .num_regulators = ARRAY_SIZE(apq8064_dsi_regulators), + .bus_clk_names = dsi_v2_bus_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_v2_bus_clk_names), + .io_start = { + { 0x4700000, 0x5800000 }, + }, +}; + +static const char * const dsi_6g_bus_clk_names[] = { + "mdp_core", "iface", "bus", "core_mmss", +}; + +static const struct regulator_bulk_data msm8974_apq8084_regulators[] = { + { .supply = "vdd", .init_load_uA = 150000 }, /* 3.0 V */ + { .supply = "vdda", .init_load_uA = 100000 }, /* 1.2 V */ + { .supply = "vddio", .init_load_uA = 100000 }, /* 1.8 V */ +}; + +static const struct msm_dsi_config msm8974_apq8084_dsi_cfg = { + .io_offset = DSI_6G_REG_SHIFT, + .regulator_data = msm8974_apq8084_regulators, + .num_regulators = ARRAY_SIZE(msm8974_apq8084_regulators), + .bus_clk_names = dsi_6g_bus_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_6g_bus_clk_names), + .io_start = { + { 0xfd922800, 0xfd922b00 }, + }, +}; + +static const char * const dsi_v1_3_1_clk_names[] = { + "mdp_core", "iface", "bus", +}; + +static const struct regulator_bulk_data dsi_v1_3_1_regulators[] = { + { .supply = "vdda", .init_load_uA = 100000 }, /* 1.2 V */ + { .supply = "vddio", .init_load_uA = 100000 }, /* 1.8 V */ +}; + +static const struct msm_dsi_config msm8916_dsi_cfg = { + .io_offset = DSI_6G_REG_SHIFT, + .regulator_data = dsi_v1_3_1_regulators, + .num_regulators = ARRAY_SIZE(dsi_v1_3_1_regulators), + .bus_clk_names = dsi_v1_3_1_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_v1_3_1_clk_names), + .io_start = { + { 0x1a98000 }, + }, +}; + +static const struct msm_dsi_config msm8976_dsi_cfg = { + .io_offset = DSI_6G_REG_SHIFT, + .regulator_data = dsi_v1_3_1_regulators, + .num_regulators = ARRAY_SIZE(dsi_v1_3_1_regulators), + .bus_clk_names = dsi_v1_3_1_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_v1_3_1_clk_names), + .io_start = { + { 0x1a94000, 0x1a96000 }, + }, +}; + +static const struct regulator_bulk_data msm8994_dsi_regulators[] = { + { .supply = "vdda", .init_load_uA = 100000 }, /* 1.25 V */ + { .supply = "vddio", .init_load_uA = 100000 }, /* 1.8 V */ + { .supply = "vcca", .init_load_uA = 10000 }, /* 1.0 V */ + { .supply = "vdd", .init_load_uA = 100000 }, /* 1.8 V */ + { .supply = "lab_reg", .init_load_uA = -1 }, + { .supply = "ibb_reg", .init_load_uA = -1 }, +}; + +static const struct msm_dsi_config msm8994_dsi_cfg = { + .io_offset = DSI_6G_REG_SHIFT, + .regulator_data = msm8994_dsi_regulators, + .num_regulators = ARRAY_SIZE(msm8994_dsi_regulators), + .bus_clk_names = dsi_6g_bus_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_6g_bus_clk_names), + .io_start = { + { 0xfd998000, 0xfd9a0000 }, + }, +}; + +static const struct regulator_bulk_data msm8996_dsi_regulators[] = { + { .supply = "vdda", .init_load_uA = 18160 }, /* 1.25 V */ + { .supply = "vcca", .init_load_uA = 17000 }, /* 0.925 V */ + { .supply = "vddio", .init_load_uA = 100000 }, /* 1.8 V */ +}; + +static const struct msm_dsi_config msm8996_dsi_cfg = { + .io_offset = DSI_6G_REG_SHIFT, + .regulator_data = msm8996_dsi_regulators, + .num_regulators = ARRAY_SIZE(msm8996_dsi_regulators), + .bus_clk_names = dsi_6g_bus_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_6g_bus_clk_names), + .io_start = { + { 0x994000, 0x996000 }, + }, +}; + +static const char * const dsi_msm8998_bus_clk_names[] = { + "iface", "bus", "core", +}; + +static const struct regulator_bulk_data msm8998_dsi_regulators[] = { + { .supply = "vdd", .init_load_uA = 367000 }, /* 0.9 V */ + { .supply = "vdda", .init_load_uA = 62800 }, /* 1.2 V */ +}; + +static const struct msm_dsi_config msm8998_dsi_cfg = { + .io_offset = DSI_6G_REG_SHIFT, + .regulator_data = msm8998_dsi_regulators, + .num_regulators = ARRAY_SIZE(msm8998_dsi_regulators), + .bus_clk_names = dsi_msm8998_bus_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_msm8998_bus_clk_names), + .io_start = { + { 0xc994000, 0xc996000 }, + }, +}; + +static const char * const dsi_sdm660_bus_clk_names[] = { + "iface", "bus", "core", "core_mmss", +}; + +static const struct regulator_bulk_data sdm660_dsi_regulators[] = { + { .supply = "vdda", .init_load_uA = 12560 }, /* 1.2 V */ +}; + +static const struct msm_dsi_config sdm660_dsi_cfg = { + .io_offset = DSI_6G_REG_SHIFT, + .regulator_data = sdm660_dsi_regulators, + .num_regulators = ARRAY_SIZE(sdm660_dsi_regulators), + .bus_clk_names = dsi_sdm660_bus_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_sdm660_bus_clk_names), + .io_start = { + { 0xc994000, 0xc996000 }, + }, +}; + +static const char * const dsi_v2_4_clk_names[] = { + "iface", "bus", +}; + +static const struct regulator_bulk_data dsi_v2_4_regulators[] = { + { .supply = "vdda", .init_load_uA = 21800 }, /* 1.2 V */ + { .supply = "refgen" }, +}; + +static const struct msm_dsi_config sdm845_dsi_cfg = { + .io_offset = DSI_6G_REG_SHIFT, + .regulator_data = dsi_v2_4_regulators, + .num_regulators = ARRAY_SIZE(dsi_v2_4_regulators), + .bus_clk_names = dsi_v2_4_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_v2_4_clk_names), + .io_start = { + { 0xae94000, 0xae96000 }, /* SDM845 / SDM670 */ + { 0x5e94000 }, /* QCM2290 / SM6115 / SM6125 / SM6375 */ + }, +}; + +static const struct regulator_bulk_data sm8550_dsi_regulators[] = { + { .supply = "vdda", .init_load_uA = 16800 }, /* 1.2 V */ +}; + +static const struct msm_dsi_config sm8550_dsi_cfg = { + .io_offset = DSI_6G_REG_SHIFT, + .regulator_data = sm8550_dsi_regulators, + .num_regulators = ARRAY_SIZE(sm8550_dsi_regulators), + .bus_clk_names = dsi_v2_4_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_v2_4_clk_names), + .io_start = { + { 0xae94000, 0xae96000 }, + }, +}; + +static const struct regulator_bulk_data sm8650_dsi_regulators[] = { + { .supply = "vdda", .init_load_uA = 16600 }, /* 1.2 V */ +}; + +static const struct msm_dsi_config sm8650_dsi_cfg = { + .io_offset = DSI_6G_REG_SHIFT, + .regulator_data = sm8650_dsi_regulators, + .num_regulators = ARRAY_SIZE(sm8650_dsi_regulators), + .bus_clk_names = dsi_v2_4_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_v2_4_clk_names), + .io_start = { + { 0xae94000, 0xae96000 }, + }, +}; + +static const struct msm_dsi_config kaanapali_dsi_cfg = { + .io_offset = DSI_6G_REG_SHIFT, + .regulator_data = sm8650_dsi_regulators, + .num_regulators = ARRAY_SIZE(sm8650_dsi_regulators), + .bus_clk_names = dsi_v2_4_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_v2_4_clk_names), + .io_start = { + { 0x9ac0000, 0x9ac3000 }, + }, +}; + +static const struct regulator_bulk_data sc7280_dsi_regulators[] = { + { .supply = "vdda", .init_load_uA = 8350 }, /* 1.2 V */ + { .supply = "refgen" }, +}; + +static const struct msm_dsi_config sc7280_dsi_cfg = { + .io_offset = DSI_6G_REG_SHIFT, + .regulator_data = sc7280_dsi_regulators, + .num_regulators = ARRAY_SIZE(sc7280_dsi_regulators), + .bus_clk_names = dsi_v2_4_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_v2_4_clk_names), + .io_start = { + { 0xae94000, 0xae96000 }, + }, +}; + +static const struct regulator_bulk_data sa8775p_dsi_regulators[] = { + { .supply = "vdda", .init_load_uA = 8300 }, /* 1.2 V */ + { .supply = "refgen" }, +}; + +static const struct msm_dsi_config sa8775p_dsi_cfg = { + .io_offset = DSI_6G_REG_SHIFT, + .regulator_data = sa8775p_dsi_regulators, + .num_regulators = ARRAY_SIZE(sa8775p_dsi_regulators), + .bus_clk_names = dsi_v2_4_clk_names, + .num_bus_clks = ARRAY_SIZE(dsi_v2_4_clk_names), + .io_start = { + { 0xae94000, 0xae96000 }, + }, +}; + +static const struct msm_dsi_host_cfg_ops msm_dsi_v2_host_ops = { + .link_clk_set_rate = dsi_link_clk_set_rate_v2, + .link_clk_enable = dsi_link_clk_enable_v2, + .link_clk_disable = dsi_link_clk_disable_v2, + .clk_init_ver = dsi_clk_init_v2, + .tx_buf_alloc = dsi_tx_buf_alloc_v2, + .tx_buf_get = dsi_tx_buf_get_v2, + .tx_buf_put = NULL, + .dma_base_get = dsi_dma_base_get_v2, + .calc_clk_rate = dsi_calc_clk_rate_v2, +}; + +static const struct msm_dsi_host_cfg_ops msm_dsi_6g_host_ops = { + .link_clk_set_rate = dsi_link_clk_set_rate_6g, + .link_clk_enable = dsi_link_clk_enable_6g, + .link_clk_disable = dsi_link_clk_disable_6g, + .clk_init_ver = NULL, + .tx_buf_alloc = dsi_tx_buf_alloc_6g, + .tx_buf_get = dsi_tx_buf_get_6g, + .tx_buf_put = dsi_tx_buf_put_6g, + .dma_base_get = dsi_dma_base_get_6g, + .calc_clk_rate = dsi_calc_clk_rate_6g, +}; + +static const struct msm_dsi_host_cfg_ops msm_dsi_6g_v2_host_ops = { + .link_clk_set_rate = dsi_link_clk_set_rate_6g, + .link_clk_enable = dsi_link_clk_enable_6g, + .link_clk_disable = dsi_link_clk_disable_6g, + .clk_init_ver = dsi_clk_init_6g_v2, + .tx_buf_alloc = dsi_tx_buf_alloc_6g, + .tx_buf_get = dsi_tx_buf_get_6g, + .tx_buf_put = dsi_tx_buf_put_6g, + .dma_base_get = dsi_dma_base_get_6g, + .calc_clk_rate = dsi_calc_clk_rate_6g, +}; + +static const struct msm_dsi_host_cfg_ops msm_dsi_6g_v2_9_host_ops = { + .link_clk_set_rate = dsi_link_clk_set_rate_6g_v2_9, + .link_clk_enable = dsi_link_clk_enable_6g, + .link_clk_disable = dsi_link_clk_disable_6g, + .clk_init_ver = dsi_clk_init_6g_v2_9, + .tx_buf_alloc = dsi_tx_buf_alloc_6g, + .tx_buf_get = dsi_tx_buf_get_6g, + .tx_buf_put = dsi_tx_buf_put_6g, + .dma_base_get = dsi_dma_base_get_6g, + .calc_clk_rate = dsi_calc_clk_rate_6g, +}; + +static const struct msm_dsi_cfg_handler dsi_cfg_handlers[] = { + {MSM_DSI_VER_MAJOR_V2, MSM_DSI_V2_VER_MINOR_8064, + &apq8064_dsi_cfg, &msm_dsi_v2_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_0, + &msm8974_apq8084_dsi_cfg, &msm_dsi_6g_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_0_2, + &msm8974_apq8084_dsi_cfg, &msm_dsi_6g_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_1, + &msm8974_apq8084_dsi_cfg, &msm_dsi_6g_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_1_1, + &msm8974_apq8084_dsi_cfg, &msm_dsi_6g_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_2, + &msm8974_apq8084_dsi_cfg, &msm_dsi_6g_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_3, + &msm8994_dsi_cfg, &msm_dsi_6g_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_3_1, + &msm8916_dsi_cfg, &msm_dsi_6g_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_4_1, + &msm8996_dsi_cfg, &msm_dsi_6g_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_4_2, + &msm8976_dsi_cfg, &msm_dsi_6g_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_0_0, + &msm8998_dsi_cfg, &msm_dsi_6g_v2_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_1_0, + &sdm660_dsi_cfg, &msm_dsi_6g_v2_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_2_1, + &sdm845_dsi_cfg, &msm_dsi_6g_v2_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_3_0, + &sdm845_dsi_cfg, &msm_dsi_6g_v2_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_3_1, + &sdm845_dsi_cfg, &msm_dsi_6g_v2_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_4_0, + &sdm845_dsi_cfg, &msm_dsi_6g_v2_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_4_1, + &sdm845_dsi_cfg, &msm_dsi_6g_v2_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_5_0, + &sc7280_dsi_cfg, &msm_dsi_6g_v2_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_5_1, + &sa8775p_dsi_cfg, &msm_dsi_6g_v2_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_6_0, + &sdm845_dsi_cfg, &msm_dsi_6g_v2_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_7_0, + &sm8550_dsi_cfg, &msm_dsi_6g_v2_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_8_0, + &sm8650_dsi_cfg, &msm_dsi_6g_v2_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_9_0, + &sm8650_dsi_cfg, &msm_dsi_6g_v2_9_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_10_0, + &kaanapali_dsi_cfg, &msm_dsi_6g_v2_9_host_ops}, +}; + +const struct msm_dsi_cfg_handler *msm_dsi_cfg_get(u32 major, u32 minor) +{ + const struct msm_dsi_cfg_handler *cfg_hnd = NULL; + int i; + + for (i = ARRAY_SIZE(dsi_cfg_handlers) - 1; i >= 0; i--) { + if ((dsi_cfg_handlers[i].major == major) && + (dsi_cfg_handlers[i].minor == minor)) { + cfg_hnd = &dsi_cfg_handlers[i]; + break; + } + } + + return cfg_hnd; +} diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.h b/drivers/gpu/drm/msm/dsi/dsi_cfg.h new file mode 100644 index 000000000..ccf066796 --- /dev/null +++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.h @@ -0,0 +1,75 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2015, The Linux Foundation. All rights reserved. + */ + +#ifndef __MSM_DSI_CFG_H__ +#define __MSM_DSI_CFG_H__ + +#include "dsi.h" + +#define MSM_DSI_VER_MAJOR_V2 0x02 +#define MSM_DSI_VER_MAJOR_6G 0x03 +#define MSM_DSI_6G_VER_MINOR_V1_0 0x10000000 +#define MSM_DSI_6G_VER_MINOR_V1_0_2 0x10000002 +#define MSM_DSI_6G_VER_MINOR_V1_1 0x10010000 +#define MSM_DSI_6G_VER_MINOR_V1_1_1 0x10010001 +#define MSM_DSI_6G_VER_MINOR_V1_2 0x10020000 +#define MSM_DSI_6G_VER_MINOR_V1_3 0x10030000 +#define MSM_DSI_6G_VER_MINOR_V1_3_1 0x10030001 +#define MSM_DSI_6G_VER_MINOR_V1_4_1 0x10040001 +#define MSM_DSI_6G_VER_MINOR_V1_4_2 0x10040002 +#define MSM_DSI_6G_VER_MINOR_V2_0_0 0x20000000 +#define MSM_DSI_6G_VER_MINOR_V2_1_0 0x20010000 +#define MSM_DSI_6G_VER_MINOR_V2_2_1 0x20020001 +#define MSM_DSI_6G_VER_MINOR_V2_3_0 0x20030000 +#define MSM_DSI_6G_VER_MINOR_V2_3_1 0x20030001 +#define MSM_DSI_6G_VER_MINOR_V2_4_0 0x20040000 +#define MSM_DSI_6G_VER_MINOR_V2_4_1 0x20040001 +#define MSM_DSI_6G_VER_MINOR_V2_5_0 0x20050000 +#define MSM_DSI_6G_VER_MINOR_V2_5_1 0x20050001 +#define MSM_DSI_6G_VER_MINOR_V2_6_0 0x20060000 +#define MSM_DSI_6G_VER_MINOR_V2_7_0 0x20070000 +#define MSM_DSI_6G_VER_MINOR_V2_8_0 0x20080000 +#define MSM_DSI_6G_VER_MINOR_V2_9_0 0x20090000 +#define MSM_DSI_6G_VER_MINOR_V2_10_0 0x200a0000 + +#define MSM_DSI_V2_VER_MINOR_8064 0x0 + +#define DSI_6G_REG_SHIFT 4 + +/* Maximum number of configurations matched against the same hw revision */ +#define VARIANTS_MAX 2 + +struct msm_dsi_config { + u32 io_offset; + const struct regulator_bulk_data *regulator_data; + int num_regulators; + const char * const *bus_clk_names; + const int num_bus_clks; + const resource_size_t io_start[VARIANTS_MAX][DSI_MAX]; +}; + +struct msm_dsi_host_cfg_ops { + int (*link_clk_set_rate)(struct msm_dsi_host *msm_host); + int (*link_clk_enable)(struct msm_dsi_host *msm_host); + void (*link_clk_disable)(struct msm_dsi_host *msm_host); + int (*clk_init_ver)(struct msm_dsi_host *msm_host); + int (*tx_buf_alloc)(struct msm_dsi_host *msm_host, int size); + void* (*tx_buf_get)(struct msm_dsi_host *msm_host); + void (*tx_buf_put)(struct msm_dsi_host *msm_host); + int (*dma_base_get)(struct msm_dsi_host *msm_host, uint64_t *iova); + int (*calc_clk_rate)(struct msm_dsi_host *msm_host, bool is_bonded_dsi); +}; + +struct msm_dsi_cfg_handler { + u32 major; + u32 minor; + const struct msm_dsi_config *cfg; + const struct msm_dsi_host_cfg_ops *ops; +}; + +const struct msm_dsi_cfg_handler *msm_dsi_cfg_get(u32 major, u32 minor); + +#endif /* __MSM_DSI_CFG_H__ */ + diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c new file mode 100644 index 000000000..b292dfd26 --- /dev/null +++ b/drivers/gpu/drm/msm/dsi/dsi_host.c @@ -0,0 +1,2711 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2015, The Linux Foundation. All rights reserved. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include