summaryrefslogtreecommitdiffstats
path: root/drivers/edac/imh_base.c
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2026-08-30 09:22:00 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2026-08-30 09:22:00 -0700
commit034dd340b08be1f2f0477ad16131d609f9dbd53c (patch)
tree2536f4b2d7893ccd916b5abdf3c454ad6edc03c2 /drivers/edac/imh_base.c
downloadlinux-stable-034dd340b08be1f2f0477ad16131d609f9dbd53c.tar.gz
linux-stable-034dd340b08be1f2f0477ad16131d609f9dbd53c.zip
Merge tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted
Pull tracing fixes from Steven Rostedt: - Fix error output of boot instance creation failure Currently if a boot instance creation fails, instead of printing out the name of the instance that failed, it prints "(null)". That is because it prints "cur_str" that had already been processed by strsep(). Print the saved name instead. While at it, print the error code of the failure. - Fix use-after-free for same named historgrams Histograms can be named so that they can be used in multiple events. But if the named histogram has a variable attached, the second event that uses the named histogram which duplicates it and needs to free the original after duplication leaves the old variable in place and still visible. If another histogram uses than variable, it will use the stale one which will try to reference the freed duplicate histogram and crash the kernel. Free the duplicate variables along with the duplicated histogram data. - Check return value of kthread_run() in event self test The events self tests uses a kthread for testing but does not check if it succeeded in creating a kthread. If the kthread creation were to fail, the code will still try to call kthread_stop() on the error returned. - Fix race between reading trace_pipe and updating subbuffer size If a user is reading the trace_pipe file at the same time they update the ring buffer sub-buffer size, can cause the trace_pipe read to read stale data. Add trace_access_lock() around updating the ring buffer sub-buffer size. - Fix eventfs_inode on failure path in creation of the events directory In the creation of the "events" directory, if after allocating the eventfs_inode a failure is detected, it calls cleanup_ei() which calls free_ei(). The free_ei() will test if eventfs_inode being freed has no children. It is a bug if it does. But on the failure case of the creation of the "events" directory, the children lists have not yet been initialized and the free will trigger a warning because list_empty() on an uninitialized list returns false. Move the initialization into init_ei() where it makes more sense and makes sure that a created eventfs_inode has its lists initialized upon creation. - Check return value of kthread_run() in ftrace direct sample code The sample code that shows how to use the ftrace direct calls does not test the return of kthread_run() to see if it succeeds. Return a failure if the kthread_run() doesn't succeed. - Clear user events state on fork in case of alloc failure On fork, the child gets a pointer to the parent's user events state. It makes a copy of it then updates the child's pointer to it. But if the allocation fails, the duplication function leaves the child with a pointer to its parent's descriptor. When the child cleans up its data, it will free the parent's descriptor while the parent is still using it. In the duplication function, set the child's user_event_mm to NULL before testing if the allocation succeeded, and when it exits it will not free the parent's descriptor. - Fix retry exhaustion in simple ring buffer reader swap simple_ring_buffer_swap_reader_page() starts with retry set to 8 and post-decrements it only after a failed link replacement. On the final attempt, a successful replacement leaves retry at zero, while a failed replacement leaves it at -1. But the check for success expects the retry value to be non-zero and exits with an error on zero. This is the opposite result. Fix it. - Fail nicely when the remote swap_reader_page() returns an error Currently, if the swap_reader_page() of a remote buffer fails, it triggers a WARN_ON_ONCE() and continues normally. Instead, have it exit with an error and a pr_warn() print instead of a full WARNING. * tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: ring-buffer: Stop remote reader update when page swap fails tracing: Fix retry exhaustion in simple ring buffer reader swap tracing/user_events: Clear copied tracing state before fork duplication samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-multi-modify samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-modify eventfs: Initialize ei->children and ei->list in init_ei() tracing: Fix use-after-free in trace_pipe read on sub-buffer order change tracing: Fix crash passing ERR_PTR to kthread_stop() tracing: Fix use-after-free with same-name named triggers tracing: Fix logged instance name on creation failure
Diffstat (limited to 'drivers/edac/imh_base.c')
-rw-r--r--drivers/edac/imh_base.c631
1 files changed, 631 insertions, 0 deletions
diff --git a/drivers/edac/imh_base.c b/drivers/edac/imh_base.c
new file mode 100644
index 000000000..6ca0df031
--- /dev/null
+++ b/drivers/edac/imh_base.c
@@ -0,0 +1,631 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Driver for Intel(R) servers with Integrated Memory/IO Hub-based memory controller.
+ * Copyright (c) 2025, Intel Corporation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/io.h>
+#include <asm/cpu_device_id.h>
+#include <asm/intel-family.h>
+#include <asm/mce.h>
+#include <asm/cpu.h>
+#include "edac_module.h"
+#include "skx_common.h"
+
+#define IMH_REVISION "v0.0.1"
+#define EDAC_MOD_STR "imh_edac"
+
+/* Debug macros */
+#define imh_printk(level, fmt, arg...) \
+ edac_printk(level, "imh", fmt, ##arg)
+
+/* Configuration Agent(Ubox) */
+#define MMIO_BASE_H(reg) (((u64)GET_BITFIELD(reg, 0, 29)) << 23)
+#define SOCKET_ID(reg) GET_BITFIELD(reg, 0, 3)
+
+/* PUNIT */
+#define DDR_IMC_BITMAP(reg) GET_BITFIELD(reg, 23, 30)
+
+/* Memory Controller */
+#define ECC_ENABLED(reg) GET_BITFIELD(reg, 2, 2)
+#define DIMM_POPULATED(reg) GET_BITFIELD(reg, 15, 15)
+
+/* System Cache Agent(SCA) */
+#define TOLM(reg) (((u64)GET_BITFIELD(reg, 16, 31)) << 16)
+#define TOHM(reg) (((u64)GET_BITFIELD(reg, 16, 51)) << 16)
+
+/* Home Agent (HA) */
+#define NMCACHING(reg) GET_BITFIELD(reg, 8, 8)
+
+/**
+ * struct local_reg - A register as described in the local package view.
+ *
+ * @pkg: (input) The package where the register is located.
+ * @pbase: (input) The IP MMIO base physical address in the local package view.
+ * @size: (input) The IP MMIO size.
+ * @offset: (input) The register offset from the IP MMIO base @pbase.
+ * @width: (input) The register width in byte.
+ * @vbase: (internal) The IP MMIO base virtual address.
+ * @val: (output) The register value.
+ */
+struct local_reg {
+ int pkg;
+ u64 pbase;
+ u32 size;
+ u32 offset;
+ u8 width;
+ void __iomem *vbase;
+ u64 val;
+};
+
+#define DEFINE_LOCAL_REG(name, cfg, package, north, ip_name, ip_idx, reg_name) \
+ struct local_reg name = { \
+ .pkg = package, \
+ .pbase = (north ? (cfg)->mmio_base_l_north : \
+ (cfg)->mmio_base_l_south) + \
+ (cfg)->ip_name##_base + \
+ (cfg)->ip_name##_size * (ip_idx), \
+ .size = (cfg)->ip_name##_size, \
+ .offset = (cfg)->ip_name##_reg_##reg_name##_offset, \
+ .width = (cfg)->ip_name##_reg_##reg_name##_width, \
+ }
+
+static struct res_config *res_cfg;
+static int retry_rd_err_log;
+
+#define REG_RRL_DEFINE(a0, a1, a2, a3, a4, a5, a6, b0, b1, b2, b3) \
+ { \
+ .set_num = 4, \
+ .reg_num = 7, \
+ .sources = {RRL_SRC_FRE_SCRUB, RRL_SRC_FRE_DEMAND, RRL_SRC_LRE_SCRUB, RRL_SRC_LRE_DEMAND}, \
+ .offsets = { \
+ {a0, a1, a2, a3, a4, a5, a6}, \
+ {a0 + 4, a1 + 4, a2 + 8, a3 + 4, a4 + 4, a5 + 8, a6 + 8}, \
+ {a0 + 8, a1 + 8, a2 + 16, a3 + 8, a4 + 8, a5 + 16, a6 + 16}, \
+ {a0 + 12, a1 + 12, a2 + 24, a3 + 12, a4 + 12, a5 + 24, a6 + 24}, \
+ }, \
+ .widths = {4, 4, 8, 4, 4, 8, 8}, \
+ .v_mask = BIT(0), \
+ .uc_mask = BIT(1), \
+ .over_mask = BIT(2), \
+ .en_mask = BIT(12), \
+ .en_patspr_mask = BIT(14), \
+ .noover_mask = BIT(15), \
+ .cecnt_num = 4, \
+ .cecnt_offsets = {b0, b1, b2, b3}, \
+ .cecnt_widths = {8, 8, 8, 8}, \
+}
+
+static struct reg_rrl dmr_reg_rrl_ddr_subch0 = REG_RRL_DEFINE(
+ 0x2dc0, 0x2dd0, 0x2de0, 0x2e00, 0x2e10, 0x2f70, 0x0200,
+ 0x2c10, 0x2c18, 0x2c20, 0x2c28);
+static struct reg_rrl dmr_reg_rrl_ddr_subch1 = REG_RRL_DEFINE(
+ 0x6dc0, 0x6dd0, 0x6de0, 0x6e00, 0x6e10, 0x6f70, 0x4200,
+ 0x6c10, 0x6c18, 0x6c20, 0x6c28);
+
+static void __read_local_reg(void *reg)
+{
+ struct local_reg *r = (struct local_reg *)reg;
+
+ r->val = skx_readx(r->vbase + r->offset, r->width);
+}
+
+/* Read a local-view register. */
+static bool read_local_reg(struct local_reg *reg)
+{
+ int cpu;
+
+ /* Get the target CPU in the package @reg->pkg. */
+ for_each_online_cpu(cpu) {
+ if (reg->pkg == topology_physical_package_id(cpu))
+ break;
+ }
+
+ if (cpu >= nr_cpu_ids)
+ return false;
+
+ reg->vbase = ioremap(reg->pbase, reg->size);
+ if (!reg->vbase) {
+ imh_printk(KERN_ERR, "Failed to ioremap 0x%llx\n", reg->pbase);
+ return false;
+ }
+
+ /* Get the target CPU to read the register. */
+ smp_call_function_single(cpu, __read_local_reg, reg, 1);
+ iounmap(reg->vbase);
+
+ return true;
+}
+
+/* Get the bitmap of memory controller instances in package @pkg. */
+static u32 get_imc_bitmap(struct res_config *cfg, int pkg, bool north)
+{
+ DEFINE_LOCAL_REG(reg, cfg, pkg, north, pcu, 0, capid3);
+
+ if (!read_local_reg(&reg))
+ return 0;
+
+ edac_dbg(2, "Pkg%d %s mc instances bitmap 0x%llx (reg 0x%llx)\n",
+ pkg, north ? "north" : "south",
+ DDR_IMC_BITMAP(reg.val), reg.val);
+
+ return DDR_IMC_BITMAP(reg.val);
+}
+
+static void imc_release(struct device *dev)
+{
+ edac_dbg(2, "imc device %s released\n", dev_name(dev));
+ kfree(dev);
+}
+
+static int __get_ddr_munits(struct res_config *cfg, struct skx_dev *d,
+ bool north, int lmc)
+{
+ unsigned long size = cfg->ddr_chan_mmio_sz * cfg->ddr_chan_num;
+ unsigned long bitmap = get_imc_bitmap(cfg, d->pkg, north);
+ void __iomem *mbase;
+ struct device *dev;
+ int i, rc, pmc;
+ u64 base;
+
+ for_each_set_bit(i, &bitmap, sizeof(bitmap) * 8) {
+ base = north ? d->mmio_base_h_north : d->mmio_base_h_south;
+ base += cfg->ddr_imc_base + size * i;
+
+ edac_dbg(2, "Pkg%d mc%d mmio base 0x%llx size 0x%lx\n",
+ d->pkg, lmc, base, size);
+
+ /* Set up the imc MMIO. */
+ mbase = ioremap(base, size);
+ if (!mbase) {
+ imh_printk(KERN_ERR, "Failed to ioremap 0x%llx\n", base);
+ return -ENOMEM;
+ }
+
+ d->imc[lmc].mbase = mbase;
+ d->imc[lmc].lmc = lmc;
+
+ /* Create the imc device instance. */
+ dev = kzalloc_obj(*dev);
+ if (!dev)
+ return -ENOMEM;
+
+ dev->release = imc_release;
+ device_initialize(dev);
+ rc = dev_set_name(dev, "0x%llx", base);
+ if (rc) {
+ imh_printk(KERN_ERR, "Failed to set dev name\n");
+ put_device(dev);
+ return rc;
+ }
+
+ d->imc[lmc].dev = dev;
+
+ /* Set up the imc index mapping. */
+ pmc = north ? i : 8 + i;
+ skx_set_mc_mapping(d, pmc, lmc);
+
+ lmc++;
+ }
+
+ return lmc;
+}
+
+static bool get_ddr_munits(struct res_config *cfg, struct skx_dev *d)
+{
+ int lmc = __get_ddr_munits(cfg, d, true, 0);
+
+ if (lmc < 0)
+ return false;
+
+ lmc = __get_ddr_munits(cfg, d, false, lmc);
+ if (lmc <= 0)
+ return false;
+
+ return true;
+}
+
+static bool get_socket_id(struct res_config *cfg, struct skx_dev *d)
+{
+ DEFINE_LOCAL_REG(reg, cfg, d->pkg, true, ubox, 0, socket_id);
+ u8 src_id;
+ int i;
+
+ if (!read_local_reg(&reg))
+ return false;
+
+ src_id = SOCKET_ID(reg.val);
+ edac_dbg(2, "socket id 0x%x (reg 0x%llx)\n", src_id, reg.val);
+
+ for (i = 0; i < cfg->ddr_imc_num; i++)
+ d->imc[i].src_id = src_id;
+
+ return true;
+}
+
+/* Get TOLM (Top Of Low Memory) and TOHM (Top Of High Memory) parameters. */
+static bool imh_get_tolm_tohm(struct res_config *cfg, u64 *tolm, u64 *tohm)
+{
+ DEFINE_LOCAL_REG(reg, cfg, 0, true, sca, 0, tolm);
+
+ if (!read_local_reg(&reg))
+ return false;
+
+ *tolm = TOLM(reg.val);
+ edac_dbg(2, "tolm 0x%llx (reg 0x%llx)\n", *tolm, reg.val);
+
+ DEFINE_LOCAL_REG(reg2, cfg, 0, true, sca, 0, tohm);
+
+ if (!read_local_reg(&reg2))
+ return false;
+
+ *tohm = TOHM(reg2.val);
+ edac_dbg(2, "tohm 0x%llx (reg 0x%llx)\n", *tohm, reg2.val);
+
+ return true;
+}
+
+/* Get the system-view MMIO_BASE_H for {north,south}-IMH. */
+static int imh_get_all_mmio_base_h(struct res_config *cfg, struct list_head *edac_list)
+{
+ int i, n = topology_max_packages(), imc_num = cfg->ddr_imc_num + cfg->hbm_imc_num;
+ struct skx_dev *d;
+
+ for (i = 0; i < n; i++) {
+ d = kzalloc_flex(*d, imc, imc_num);
+ if (!d)
+ return -ENOMEM;
+
+ DEFINE_LOCAL_REG(reg, cfg, i, true, ubox, 0, mmio_base);
+
+ /* Get MMIO_BASE_H for the north-IMH. */
+ if (!read_local_reg(&reg) || !reg.val) {
+ kfree(d);
+ imh_printk(KERN_ERR, "Pkg%d has no north mmio_base_h\n", i);
+ return -ENODEV;
+ }
+
+ d->mmio_base_h_north = MMIO_BASE_H(reg.val);
+ edac_dbg(2, "Pkg%d north mmio_base_h 0x%llx (reg 0x%llx)\n",
+ i, d->mmio_base_h_north, reg.val);
+
+ /* Get MMIO_BASE_H for the south-IMH (optional). */
+ DEFINE_LOCAL_REG(reg2, cfg, i, false, ubox, 0, mmio_base);
+
+ if (read_local_reg(&reg2)) {
+ d->mmio_base_h_south = MMIO_BASE_H(reg2.val);
+ edac_dbg(2, "Pkg%d south mmio_base_h 0x%llx (reg 0x%llx)\n",
+ i, d->mmio_base_h_south, reg2.val);
+ }
+
+ d->pkg = i;
+ d->num_imc = imc_num;
+ skx_init_mc_mapping(d);
+ list_add_tail(&d->list, edac_list);
+ }
+
+ return 0;
+}
+
+/* Get the number of per-package memory controllers. */
+static int imh_get_imc_num(struct res_config *cfg)
+{
+ int imc_num = hweight32(get_imc_bitmap(cfg, 0, true)) +
+ hweight32(get_imc_bitmap(cfg, 0, false));
+
+ if (!imc_num) {
+ imh_printk(KERN_ERR, "Invalid mc number\n");
+ return -ENODEV;
+ }
+
+ if (cfg->ddr_imc_num != imc_num) {
+ /*
+ * Update the configuration data to reflect the number of
+ * present DDR memory controllers.
+ */
+ cfg->ddr_imc_num = imc_num;
+ edac_dbg(2, "Set ddr mc number %d\n", imc_num);
+ }
+
+ return 0;
+}
+
+/* Get all memory controllers' parameters. */
+static int imh_get_munits(struct res_config *cfg, struct list_head *edac_list)
+{
+ struct skx_imc *imc;
+ struct skx_dev *d;
+ u8 mc = 0;
+ int i;
+
+ list_for_each_entry(d, edac_list, list) {
+ if (!get_ddr_munits(cfg, d)) {
+ imh_printk(KERN_ERR, "No mc found\n");
+ return -ENODEV;
+ }
+
+ if (!get_socket_id(cfg, d)) {
+ imh_printk(KERN_ERR, "Failed to get socket id\n");
+ return -ENODEV;
+ }
+
+ for (i = 0; i < cfg->ddr_imc_num; i++) {
+ imc = &d->imc[i];
+ if (!imc->mbase)
+ continue;
+
+ imc->chan_mmio_sz = cfg->ddr_chan_mmio_sz;
+ imc->num_channels = cfg->ddr_chan_num;
+ imc->num_dimms = cfg->ddr_dimm_num;
+ imc->mc = mc++;
+ }
+ }
+
+ return 0;
+}
+
+static bool check_2lm_enabled(struct res_config *cfg, struct skx_dev *d, int ha_idx)
+{
+ DEFINE_LOCAL_REG(reg, cfg, d->pkg, true, ha, ha_idx, mode);
+
+ if (!read_local_reg(&reg))
+ return false;
+
+ if (!NMCACHING(reg.val))
+ return false;
+
+ edac_dbg(2, "2-level memory configuration (reg 0x%llx, ha idx %d)\n", reg.val, ha_idx);
+ return true;
+}
+
+/* Check whether the system has a 2-level memory configuration. */
+static bool imh_2lm_enabled(struct res_config *cfg, struct list_head *head)
+{
+ struct skx_dev *d;
+ int i;
+
+ list_for_each_entry(d, head, list) {
+ for (i = 0; i < cfg->ddr_imc_num; i++)
+ if (check_2lm_enabled(cfg, d, i))
+ return true;
+ }
+
+ return false;
+}
+
+static u32 read_imc_mcmtr(struct res_config *cfg, struct skx_imc *imc, int chan)
+{
+ return (u32)skx_read_imc_reg(imc, chan, cfg->ddr_reg_mcmtr_offset, cfg->ddr_reg_mcmtr_width);
+}
+
+static u32 read_imc_dimmmtr(struct res_config *cfg, struct skx_imc *imc, int chan, int dimm)
+{
+ return (u32)skx_read_imc_reg(imc, chan, cfg->ddr_reg_dimmmtr_offset +
+ cfg->ddr_reg_dimmmtr_width * dimm,
+ cfg->ddr_reg_dimmmtr_width);
+}
+
+static bool ecc_enabled(u32 mcmtr)
+{
+ return (bool)ECC_ENABLED(mcmtr);
+}
+
+static bool dimm_populated(u32 dimmmtr)
+{
+ return (bool)DIMM_POPULATED(dimmmtr);
+}
+
+/* Get each DIMM's configurations of the memory controller @mci. */
+static int imh_get_dimm_config(struct mem_ctl_info *mci, struct res_config *cfg)
+{
+ struct skx_pvt *pvt = mci->pvt_info;
+ struct skx_imc *imc = pvt->imc;
+ struct dimm_info *dimm;
+ u32 mcmtr, dimmmtr;
+ int i, j, ndimms;
+
+ for (i = 0; i < imc->num_channels; i++) {
+ if (!imc->mbase)
+ continue;
+
+ mcmtr = read_imc_mcmtr(cfg, imc, i);
+
+ for (ndimms = 0, j = 0; j < imc->num_dimms; j++) {
+ dimmmtr = read_imc_dimmmtr(cfg, imc, i, j);
+ edac_dbg(1, "mcmtr 0x%x dimmmtr 0x%x (mc%d ch%d dimm%d)\n",
+ mcmtr, dimmmtr, imc->mc, i, j);
+
+ if (!dimm_populated(dimmmtr))
+ continue;
+
+ dimm = edac_get_dimm(mci, i, j, 0);
+ ndimms += skx_get_dimm_info(dimmmtr, 0, 0, dimm,
+ imc, i, j, cfg);
+ }
+
+ if (ndimms && !ecc_enabled(mcmtr)) {
+ imh_printk(KERN_ERR, "ECC is disabled on mc%d ch%d\n",
+ imc->mc, i);
+ return -ENODEV;
+ }
+ }
+
+ return 0;
+}
+
+/* Register all memory controllers to the EDAC core. */
+static int imh_register_mci(struct res_config *cfg, struct list_head *edac_list)
+{
+ struct skx_imc *imc;
+ struct skx_dev *d;
+ int i, rc;
+
+ list_for_each_entry(d, edac_list, list) {
+ for (i = 0; i < cfg->ddr_imc_num; i++) {
+ imc = &d->imc[i];
+ if (!imc->mbase)
+ continue;
+
+ rc = skx_register_mci(imc, imc->dev,
+ dev_name(imc->dev),
+ "Intel IMH-based Socket",
+ EDAC_MOD_STR,
+ imh_get_dimm_config, cfg);
+ if (rc)
+ return rc;
+ }
+ }
+
+ return 0;
+}
+
+static struct res_config dmr_cfg = {
+ .type = DMR,
+ .support_ddr5 = true,
+ .mmio_base_l_north = 0xf6800000,
+ .mmio_base_l_south = 0xf6000000,
+ .ddr_chan_num = 1,
+ .ddr_dimm_num = 2,
+ .ddr_imc_base = 0x39b000,
+ .ddr_chan_mmio_sz = 0x8000,
+ .ddr_reg_mcmtr_offset = 0x360,
+ .ddr_reg_mcmtr_width = 4,
+ .ddr_reg_dimmmtr_offset = 0x370,
+ .ddr_reg_dimmmtr_width = 4,
+ .ubox_base = 0x0,
+ .ubox_size = 0x2000,
+ .ubox_reg_mmio_base_offset = 0x580,
+ .ubox_reg_mmio_base_width = 4,
+ .ubox_reg_socket_id_offset = 0x1080,
+ .ubox_reg_socket_id_width = 4,
+ .pcu_base = 0x3000,
+ .pcu_size = 0x10000,
+ .pcu_reg_capid3_offset = 0x290,
+ .pcu_reg_capid3_width = 4,
+ .sca_base = 0x24c000,
+ .sca_size = 0x2500,
+ .sca_reg_tolm_offset = 0x2100,
+ .sca_reg_tolm_width = 8,
+ .sca_reg_tohm_offset = 0x2108,
+ .sca_reg_tohm_width = 8,
+ .ha_base = 0x3eb000,
+ .ha_size = 0x1000,
+ .ha_reg_mode_offset = 0x4a0,
+ .ha_reg_mode_width = 4,
+ .reg_rrl_ddr[0] = &dmr_reg_rrl_ddr_subch0,
+ .reg_rrl_ddr[1] = &dmr_reg_rrl_ddr_subch1,
+};
+
+static const struct x86_cpu_id imh_cpuids[] = {
+ X86_MATCH_VFM(INTEL_DIAMONDRAPIDS_X, &dmr_cfg),
+ {}
+};
+MODULE_DEVICE_TABLE(x86cpu, imh_cpuids);
+
+static struct notifier_block imh_mce_dec = {
+ .notifier_call = skx_mce_check_error,
+ .priority = MCE_PRIO_EDAC,
+};
+
+static int __init imh_init(void)
+{
+ const struct x86_cpu_id *id;
+ struct list_head *edac_list;
+ struct res_config *cfg;
+ const char *owner;
+ u64 tolm, tohm;
+ int rc;
+
+ edac_dbg(2, "\n");
+
+ if (ghes_get_devices())
+ return -EBUSY;
+
+ owner = edac_get_owner();
+ if (owner && strncmp(owner, EDAC_MOD_STR, sizeof(EDAC_MOD_STR)))
+ return -EBUSY;
+
+ if (cpu_feature_enabled(X86_FEATURE_HYPERVISOR))
+ return -ENODEV;
+
+ id = x86_match_cpu(imh_cpuids);
+ if (!id)
+ return -ENODEV;
+ cfg = (struct res_config *)id->driver_data;
+ skx_set_res_cfg(cfg);
+ res_cfg = cfg;
+
+ if (!imh_get_tolm_tohm(cfg, &tolm, &tohm))
+ return -ENODEV;
+
+ skx_set_hi_lo(tolm, tohm);
+
+ rc = imh_get_imc_num(cfg);
+ if (rc < 0)
+ goto fail;
+
+ edac_list = skx_get_edac_list();
+
+ rc = imh_get_all_mmio_base_h(cfg, edac_list);
+ if (rc)
+ goto fail;
+
+ rc = imh_get_munits(cfg, edac_list);
+ if (rc)
+ goto fail;
+
+ skx_set_mem_cfg(imh_2lm_enabled(cfg, edac_list));
+
+ rc = imh_register_mci(cfg, edac_list);
+ if (rc)
+ goto fail;
+
+ rc = skx_adxl_get();
+ if (rc)
+ goto fail;
+
+ opstate_init();
+ mce_register_decode_chain(&imh_mce_dec);
+ skx_setup_debug("imh_test");
+
+ cfg->rrl_ctrl_mode = retry_rd_err_log;
+ if (retry_rd_err_log && cfg->reg_rrl_ddr[0]) {
+ skx_set_show_rrl(skx_show_rrl);
+ if (retry_rd_err_log == RRL_CTRL_LINUX)
+ skx_enable_rrl(true);
+ }
+
+ imh_printk(KERN_INFO, "%s\n", IMH_REVISION);
+
+ return 0;
+fail:
+ skx_remove();
+ return rc;
+}
+
+static void __exit imh_exit(void)
+{
+ edac_dbg(2, "\n");
+
+ if (retry_rd_err_log && res_cfg->reg_rrl_ddr[0]) {
+ if (retry_rd_err_log == RRL_CTRL_LINUX)
+ skx_enable_rrl(false);
+ skx_set_show_rrl(NULL);
+ }
+
+ skx_teardown_debug();
+ mce_unregister_decode_chain(&imh_mce_dec);
+ skx_adxl_put();
+ skx_remove();
+}
+
+module_init(imh_init);
+module_exit(imh_exit);
+
+module_param(retry_rd_err_log, int, 0444);
+MODULE_PARM_DESC(retry_rd_err_log, "retry_rd_err_log: 0=off(default), 1=bios(Linux doesn't reset any control bits, but just reports values.), 2=linux(Linux tries to take control and resets mode bits, clear valid/UC bits after reading.)");
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Qiuxu Zhuo");
+MODULE_DESCRIPTION("MC Driver for Intel servers using IMH-based memory controller");