sched/cache: Keep nr_pref_llc_running in the runnable domain, to fix LLC mis-scheduling buggrafted

alb_break_llc() decides whether to break LLC preference during active
load balance. It does so by testing that every runnable fair task on the
source rq prefers its LLC:

	env->src_rq->nr_pref_llc_running == env->src_rq->cfs.h_nr_runnable

But the two counters cover different sets. nr_pref_llc_running is updated
in account_llc_enqueue()/account_llc_dequeue(), next to cfs_rq->nr_queued,
so it follows queued tasks. h_nr_runnable is updated in set_delayed()/
clear_delayed() and drops delay-dequeued tasks.

So under DELAY_DEQUEUE, a preferring task that goes to sleep stays counted
in nr_pref_llc_running while h_nr_runnable falls. The equality then breaks,
alb_break_llc() returns false, and active balance is free to pull a task
off its preferred LLC. Active balance only moves runnable tasks, and this
is the only LLC check it consults: once the stopper runs, LBF_ACTIVE_LB
skips the per-task test in can_migrate_task(). The runnable set is the one
we want.

Fix it on the counter side. A task should be counted in
nr_pref_llc_running exactly while it is both queued on its preferred LLC
(pref_llc_queued) and runnable (!sched_delayed). Define that membership
once in task_pref_llc_runnable(), and adjust the counter only through
pref_llc_running_inc()/pref_llc_running_dec() from the four sites that
change either input: account_llc_enqueue(), account_llc_dequeue(),
set_delayed() and clear_delayed(). Gating every update on the same
predicate keeps the delay, wake and dequeue paths from double-counting
or underflowing; see the comments at those sites for the ordering.

nr_llc_running and sd->llc_counts are not touched and stay on queued
semantics.

Fixes: 714059f79ff0 ("sched/cache: Handle moving single tasks to/from their preferred LLC")
Closes: https://lore.kernel.org/lkml/20260827135000.735138-1-zhanxusheng@xiaomi.com/
Reported-by: Zhan Xusheng <zhanxusheng@xiaomi.com>
Suggested-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Kayra Cizmeci <kayracizmeci@gmail.com>
Cc: <stable@kernel.org> # v7.2.x
Link: https://patch.msgid.link/06af61afedac32e6477f57feb4d658f6c411c3af.1790035273.git.tim.c.chen@linux.intel.com

1 files changed, 151 insertions, 0 deletions

diff --git a/include/sound/sof/stream.h b/include/sound/sof/stream.h
new file mode 100644
index 000000000..1bb25487d
--- /dev/null
+++ b/include/sound/sof/stream.h
@@ -0,0 +1,151 @@
+/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause) */
+/*
+ * This file is provided under a dual BSD/GPLv2 license.  When using or
+ * redistributing this file, you may do so under either license.
+ *
+ * Copyright(c) 2018 Intel Corporation
+ */
+
+#ifndef __INCLUDE_SOUND_SOF_STREAM_H__
+#define __INCLUDE_SOUND_SOF_STREAM_H__
+
+#include <sound/sof/header.h>
+
+/*
+ * Stream configuration.
+ */
+
+#define SOF_IPC_MAX_CHANNELS			8
+
+/* common sample rates for use in masks */
+#define SOF_RATE_8000		(1 <<  0) /**< 8000Hz  */
+#define SOF_RATE_11025		(1 <<  1) /**< 11025Hz */
+#define SOF_RATE_12000		(1 <<  2) /**< 12000Hz */
+#define SOF_RATE_16000		(1 <<  3) /**< 16000Hz */
+#define SOF_RATE_22050		(1 <<  4) /**< 22050Hz */
+#define SOF_RATE_24000		(1 <<  5) /**< 24000Hz */
+#define SOF_RATE_32000		(1 <<  6) /**< 32000Hz */
+#define SOF_RATE_44100		(1 <<  7) /**< 44100Hz */
+#define SOF_RATE_48000		(1 <<  8) /**< 48000Hz */
+#define SOF_RATE_64000		(1 <<  9) /**< 64000Hz */
+#define SOF_RATE_88200		(1 << 10) /**< 88200Hz */
+#define SOF_RATE_96000		(1 << 11) /**< 96000Hz */
+#define SOF_RATE_176400		(1 << 12) /**< 176400Hz */
+#define SOF_RATE_192000		(1 << 13) /**< 192000Hz */
+
+/* continuous and non-standard rates for flexibility */
+#define SOF_RATE_CONTINUOUS	(1 << 30)  /**< range */
+#define SOF_RATE_KNOT		(1 << 31)  /**< non-continuous */
+
+/* generic PCM flags for runtime settings */
+#define SOF_PCM_FLAG_XRUN_STOP	(1 << 0) /**< Stop on any XRUN */
+
+/* stream PCM frame format */
+enum sof_ipc_frame {
+	SOF_IPC_FRAME_S16_LE = 0,
+	SOF_IPC_FRAME_S24_4LE,
+	SOF_IPC_FRAME_S32_LE,
+	SOF_IPC_FRAME_FLOAT,
+	/* other formats here */
+};
+
+/* stream buffer format */
+enum sof_ipc_buffer_format {
+	SOF_IPC_BUFFER_INTERLEAVED,
+	SOF_IPC_BUFFER_NONINTERLEAVED,
+	/* other formats here */
+};
+
+/* stream direction */
+enum sof_ipc_stream_direction {
+	SOF_IPC_STREAM_PLAYBACK = 0,
+	SOF_IPC_STREAM_CAPTURE,
+};
+
+/* stream ring info */
+struct sof_ipc_host_buffer {
+	struct sof_ipc_hdr hdr;
+	uint32_t phy_addr;
+	uint32_t pages;
+	uint32_t size;
+	uint32_t reserved[3];
+} __packed;
+
+struct sof_ipc_stream_params {
+	struct sof_ipc_hdr hdr;
+	struct sof_ipc_host_buffer buffer;
+	uint32_t direction;	/**< enum sof_ipc_stream_direction */
+	uint32_t frame_fmt;	/**< enum sof_ipc_frame */
+	uint32_t buffer_fmt;	/**< enum sof_ipc_buffer_format */
+	uint32_t rate;
+	uint16_t stream_tag;
+	uint16_t channels;
+	uint16_t sample_valid_bytes;
+	uint16_t sample_container_bytes;
+
+	uint32_t host_period_bytes;
+	uint16_t no_stream_position; /**< 1 means don't send stream position */
+	uint8_t cont_update_posn; /**< 1 means continuous update stream position */
+	uint8_t reserved0;
+	int16_t ext_data_length; /**< 0, means no extended data */
+	uint8_t reserved[2];
+	uint16_t chmap[SOF_IPC_MAX_CHANNELS];	/**< channel map - SOF_CHMAP_ */
+	uint8_t ext_data[]; /**< extended data */
+} __packed;
+
+/* PCM params info - SOF_IPC_STREAM_PCM_PARAMS */
+struct sof_ipc_pcm_params {
+	struct sof_ipc_cmd_hdr hdr;
+	uint32_t comp_id;
+	uint32_t flags;		/**< generic PCM flags - SOF_PCM_FLAG_ */
+	uint32_t reserved[2];
+	struct sof_ipc_stream_params params;
+}  __packed;
+
+/* PCM params info reply - SOF_IPC_STREAM_PCM_PARAMS_REPLY */
+struct sof_ipc_pcm_params_reply {
+	struct sof_ipc_reply rhdr;
+	uint32_t comp_id;
+	uint32_t posn_offset;
+} __packed;
+
+/* free stream - SOF_IPC_STREAM_PCM_PARAMS */
+struct sof_ipc_stream {
+	struct sof_ipc_cmd_hdr hdr;
+	uint32_t comp_id;
+} __packed;
+
+/* flags indicating which time stamps are in sync with each other */
+#define	SOF_TIME_HOST_SYNC	(1 << 0)
+#define	SOF_TIME_DAI_SYNC	(1 << 1)
+#define	SOF_TIME_WALL_SYNC	(1 << 2)
+#define	SOF_TIME_STAMP_SYNC	(1 << 3)
+
+/* flags indicating which time stamps are valid */
+#define	SOF_TIME_HOST_VALID	(1 << 8)
+#define	SOF_TIME_DAI_VALID	(1 << 9)
+#define	SOF_TIME_WALL_VALID	(1 << 10)
+#define	SOF_TIME_STAMP_VALID	(1 << 11)
+
+/* flags indicating time stamps are 64bit else 3use low 32bit */
+#define	SOF_TIME_HOST_64	(1 << 16)
+#define	SOF_TIME_DAI_64		(1 << 17)
+#define	SOF_TIME_WALL_64	(1 << 18)
+#define	SOF_TIME_STAMP_64	(1 << 19)
+
+struct sof_ipc_stream_posn {
+	struct sof_ipc_reply rhdr;
+	uint32_t comp_id;	/**< host component ID */
+	uint32_t flags;		/**< SOF_TIME_ */
+	uint32_t wallclock_hz;	/**< frequency of wallclock in Hz */
+	uint32_t timestamp_ns;	/**< resolution of timestamp in ns */
+	uint64_t host_posn;	/**< host DMA position in bytes */
+	uint64_t dai_posn;	/**< DAI DMA position in bytes */
+	uint64_t comp_posn;	/**< comp position in bytes */
+	uint64_t wallclock;	/**< audio wall clock */
+	uint64_t timestamp;	/**< system time stamp */
+	uint32_t xrun_comp_id;	/**< comp ID of XRUN component */
+	int32_t xrun_size;	/**< XRUN size in bytes */
+}  __packed;
+
+#endif