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, 76 insertions, 0 deletions

diff --git a/Documentation/ABI/testing/sysfs-class-wakeup b/Documentation/ABI/testing/sysfs-class-wakeup
new file mode 100644
index 000000000..754aab8b6
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-class-wakeup
@@ -0,0 +1,76 @@
+What:		/sys/class/wakeup/
+Date:		June 2019
+Contact:	Tri Vo <trong@android.com>
+Description:
+		The /sys/class/wakeup/ directory contains pointers to all
+		wakeup sources in the kernel at that moment in time.
+
+What:		/sys/class/wakeup/.../name
+Date:		June 2019
+Contact:	Tri Vo <trong@android.com>
+Description:
+		This file contains the name of the wakeup source.
+
+What:		/sys/class/wakeup/.../active_count
+Date:		June 2019
+Contact:	Tri Vo <trong@android.com>
+Description:
+		This file contains the number of times the wakeup source was
+		activated.
+
+What:		/sys/class/wakeup/.../event_count
+Date:		June 2019
+Contact:	Tri Vo <trong@android.com>
+Description:
+		This file contains the number of signaled wakeup events
+		associated with the wakeup source.
+
+What:		/sys/class/wakeup/.../wakeup_count
+Date:		June 2019
+Contact:	Tri Vo <trong@android.com>
+Description:
+		This file contains the number of times the wakeup source might
+		abort suspend.
+
+What:		/sys/class/wakeup/.../expire_count
+Date:		June 2019
+Contact:	Tri Vo <trong@android.com>
+Description:
+		This file contains the number of times the wakeup source's
+		timeout has expired.
+
+What:		/sys/class/wakeup/.../active_time_ms
+Date:		June 2019
+Contact:	Tri Vo <trong@android.com>
+Description:
+		This file contains the amount of time the wakeup source has
+		been continuously active, in milliseconds.  If the wakeup
+		source is not active, this file contains '0'.
+
+What:		/sys/class/wakeup/.../total_time_ms
+Date:		June 2019
+Contact:	Tri Vo <trong@android.com>
+Description:
+		This file contains the total amount of time this wakeup source
+		has been active, in milliseconds.
+
+What:		/sys/class/wakeup/.../max_time_ms
+Date:		June 2019
+Contact:	Tri Vo <trong@android.com>
+Description:
+		This file contains the maximum amount of time this wakeup
+		source has been continuously active, in milliseconds.
+
+What:		/sys/class/wakeup/.../last_change_ms
+Date:		June 2019
+Contact:	Tri Vo <trong@android.com>
+Description:
+		This file contains the monotonic clock time when the wakeup
+		source was touched last time, in milliseconds.
+
+What:		/sys/class/wakeup/.../prevent_suspend_time_ms
+Date:		June 2019
+Contact:	Tri Vo <trong@android.com>
+Description:
+		The file contains the total amount of time this wakeup source
+		has been preventing autosleep, in milliseconds.