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, 144 insertions, 0 deletions

diff --git a/include/sound/sof/info.h b/include/sound/sof/info.h
new file mode 100644
index 000000000..08400fbe5
--- /dev/null
+++ b/include/sound/sof/info.h
@@ -0,0 +1,144 @@
+/* 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_INFO_H__
+#define __INCLUDE_SOUND_SOF_INFO_H__
+
+#include <sound/sof/header.h>
+#include <sound/sof/stream.h>
+
+/*
+ * Firmware boot and version
+ */
+
+#define SOF_IPC_MAX_ELEMS	16
+
+/*
+ * Firmware boot info flag bits (64-bit)
+ */
+#define SOF_IPC_INFO_BUILD		BIT(0)
+#define SOF_IPC_INFO_LOCKS		BIT(1)
+#define SOF_IPC_INFO_LOCKSV		BIT(2)
+#define SOF_IPC_INFO_GDB		BIT(3)
+#define SOF_IPC_INFO_D3_PERSISTENT	BIT(4)
+
+/* extended data types that can be appended onto end of sof_ipc_fw_ready */
+enum sof_ipc_ext_data {
+	SOF_IPC_EXT_UNUSED		= 0,
+	SOF_IPC_EXT_WINDOW		= 1,
+	SOF_IPC_EXT_CC_INFO		= 2,
+	SOF_IPC_EXT_PROBE_INFO		= 3,
+	SOF_IPC_EXT_USER_ABI_INFO	= 4,
+};
+
+/* Build u32 number in format MMmmmppp */
+#define SOF_FW_VER(MAJOR, MINOR, PATCH) ((uint32_t)( \
+	((MAJOR) << 24) | ((MINOR) << 12) | (PATCH)))
+
+/* FW version - SOF_IPC_GLB_VERSION */
+struct sof_ipc_fw_version {
+	struct sof_ipc_hdr hdr;
+	uint16_t major;
+	uint16_t minor;
+	uint16_t micro;
+	uint16_t build;
+	uint8_t date[12];
+	uint8_t time[10];
+	uint8_t tag[6];
+	uint32_t abi_version;
+	/* used to check FW and ldc file compatibility, reproducible value */
+	uint32_t src_hash;
+
+	/* reserved for future use */
+	uint32_t reserved[3];
+} __packed;
+
+/* FW ready Message - sent by firmware when boot has completed */
+struct sof_ipc_fw_ready {
+	struct sof_ipc_cmd_hdr hdr;
+	uint32_t dspbox_offset;	 /* dsp initiated IPC mailbox */
+	uint32_t hostbox_offset; /* host initiated IPC mailbox */
+	uint32_t dspbox_size;
+	uint32_t hostbox_size;
+	struct sof_ipc_fw_version version;
+
+	/* Miscellaneous flags */
+	uint64_t flags;
+
+	/* reserved for future use */
+	uint32_t reserved[4];
+} __packed;
+
+/*
+ * Extended Firmware data. All optional, depends on platform/arch.
+ */
+enum sof_ipc_region {
+	SOF_IPC_REGION_DOWNBOX	= 0,
+	SOF_IPC_REGION_UPBOX,
+	SOF_IPC_REGION_TRACE,
+	SOF_IPC_REGION_DEBUG,
+	SOF_IPC_REGION_STREAM,
+	SOF_IPC_REGION_REGS,
+	SOF_IPC_REGION_EXCEPTION,
+};
+
+struct sof_ipc_ext_data_hdr {
+	struct sof_ipc_cmd_hdr hdr;
+	uint32_t type;		/**< SOF_IPC_EXT_ */
+} __packed;
+
+struct sof_ipc_window_elem {
+	struct sof_ipc_hdr hdr;
+	uint32_t type;		/**< SOF_IPC_REGION_ */
+	uint32_t id;		/**< platform specific - used to map to host memory */
+	uint32_t flags;		/**< R, W, RW, etc - to define */
+	uint32_t size;		/**< size of region in bytes */
+	/* offset in window region as windows can be partitioned */
+	uint32_t offset;
+} __packed;
+
+/* extended data memory windows for IPC, trace and debug */
+struct sof_ipc_window {
+	struct sof_ipc_ext_data_hdr ext_hdr;
+	uint32_t num_windows;
+	struct sof_ipc_window_elem window[SOF_IPC_MAX_ELEMS];
+}  __packed;
+
+struct sof_ipc_cc_version {
+	struct sof_ipc_ext_data_hdr ext_hdr;
+	uint32_t major;
+	uint32_t minor;
+	uint32_t micro;
+
+	/* reserved for future use */
+	uint32_t reserved[4];
+
+	uint8_t name[16]; /* null terminated compiler name */
+	uint8_t optim[4]; /* null terminated compiler -O flag value */
+	uint8_t desc[32]; /* null terminated compiler description */
+} __packed;
+
+/* extended data: Probe setup */
+struct sof_ipc_probe_support {
+	struct sof_ipc_ext_data_hdr ext_hdr;
+
+	uint32_t probe_points_max;
+	uint32_t injection_dmas_max;
+
+	/* reserved for future use */
+	uint32_t reserved[2];
+} __packed;
+
+/* extended data: user abi version(s) */
+struct sof_ipc_user_abi_version {
+	struct sof_ipc_ext_data_hdr ext_hdr;
+
+	uint32_t abi_dbg_version;
+}  __packed;
+
+#endif