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, 166 insertions, 0 deletions

diff --git a/include/trace/events/lock.h b/include/trace/events/lock.h
new file mode 100644
index 000000000..b1d5b18c4
--- /dev/null
+++ b/include/trace/events/lock.h
@@ -0,0 +1,166 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM lock
+
+#if !defined(_TRACE_LOCK_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_LOCK_H
+
+#include <linux/tracepoint.h>
+
+/* flags for lock:contention_begin */
+#define LCB_F_SPIN	(1U << 0)
+#define LCB_F_READ	(1U << 1)
+#define LCB_F_WRITE	(1U << 2)
+#define LCB_F_RT	(1U << 3)
+#define LCB_F_PERCPU	(1U << 4)
+#define LCB_F_MUTEX	(1U << 5)
+
+
+#ifdef CONFIG_LOCKDEP
+
+#include <linux/lockdep.h>
+
+TRACE_EVENT(lock_acquire,
+
+	TP_PROTO(struct lockdep_map *lock, unsigned int subclass,
+		int trylock, int read, int check,
+		struct lockdep_map *next_lock, unsigned long ip),
+
+	TP_ARGS(lock, subclass, trylock, read, check, next_lock, ip),
+
+	TP_STRUCT__entry(
+		__field(unsigned int, flags)
+		__string(name, lock->name)
+		__field(void *, lockdep_addr)
+	),
+
+	TP_fast_assign(
+		__entry->flags = (trylock ? 1 : 0) | (read ? 2 : 0);
+		__assign_str(name);
+		__entry->lockdep_addr = lock;
+	),
+
+	TP_printk("%p %s%s%s", __entry->lockdep_addr,
+		  (__entry->flags & 1) ? "try " : "",
+		  (__entry->flags & 2) ? "read " : "",
+		  __get_str(name))
+);
+
+DECLARE_EVENT_CLASS(lock,
+
+	TP_PROTO(struct lockdep_map *lock, unsigned long ip),
+
+	TP_ARGS(lock, ip),
+
+	TP_STRUCT__entry(
+		__string(	name, 	lock->name	)
+		__field(	void *, lockdep_addr	)
+	),
+
+	TP_fast_assign(
+		__assign_str(name);
+		__entry->lockdep_addr = lock;
+	),
+
+	TP_printk("%p %s",  __entry->lockdep_addr, __get_str(name))
+);
+
+DEFINE_EVENT(lock, lock_release,
+
+	TP_PROTO(struct lockdep_map *lock, unsigned long ip),
+
+	TP_ARGS(lock, ip)
+);
+
+#ifdef CONFIG_LOCK_STAT
+
+DEFINE_EVENT(lock, lock_contended,
+
+	TP_PROTO(struct lockdep_map *lock, unsigned long ip),
+
+	TP_ARGS(lock, ip)
+);
+
+DEFINE_EVENT(lock, lock_acquired,
+
+	TP_PROTO(struct lockdep_map *lock, unsigned long ip),
+
+	TP_ARGS(lock, ip)
+);
+
+#endif /* CONFIG_LOCK_STAT */
+#endif /* CONFIG_LOCKDEP */
+
+TRACE_EVENT(contention_begin,
+
+	TP_PROTO(void *lock, unsigned int flags),
+
+	TP_ARGS(lock, flags),
+
+	TP_STRUCT__entry(
+		__field(void *, lock_addr)
+		__field(unsigned int, flags)
+	),
+
+	TP_fast_assign(
+		__entry->lock_addr = lock;
+		__entry->flags = flags;
+	),
+
+	TP_printk("%p (flags=%s)", __entry->lock_addr,
+		  __print_flags(__entry->flags, "|",
+				{ LCB_F_SPIN,		"SPIN" },
+				{ LCB_F_READ,		"READ" },
+				{ LCB_F_WRITE,		"WRITE" },
+				{ LCB_F_RT,		"RT" },
+				{ LCB_F_PERCPU,		"PERCPU" },
+				{ LCB_F_MUTEX,		"MUTEX" }
+			  ))
+);
+
+TRACE_EVENT(contention_end,
+
+	TP_PROTO(void *lock, int ret),
+
+	TP_ARGS(lock, ret),
+
+	TP_STRUCT__entry(
+		__field(void *, lock_addr)
+		__field(int, ret)
+	),
+
+	TP_fast_assign(
+		__entry->lock_addr = lock;
+		__entry->ret = ret;
+	),
+
+	TP_printk("%p (ret=%d)", __entry->lock_addr, __entry->ret)
+);
+
+/* kernel/locking/mutex.c */
+int arch_contended_release_trace_reg(void);
+void arch_contended_release_trace_unreg(void);
+
+TRACE_EVENT_FN(contended_release,
+
+	TP_PROTO(void *lock),
+
+	TP_ARGS(lock),
+
+	TP_STRUCT__entry(
+		__field(void *, lock_addr)
+	),
+
+	TP_fast_assign(
+		__entry->lock_addr = lock;
+	),
+
+	TP_printk("%p", __entry->lock_addr),
+
+	arch_contended_release_trace_reg, arch_contended_release_trace_unreg
+);
+
+#endif /* _TRACE_LOCK_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>