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, 99 insertions, 0 deletions

diff --git a/include/trace/events/task.h b/include/trace/events/task.h
new file mode 100644
index 000000000..b9a129eb5
--- /dev/null
+++ b/include/trace/events/task.h
@@ -0,0 +1,99 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM task
+
+#if !defined(_TRACE_TASK_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_TASK_H
+#include <linux/tracepoint.h>
+
+TRACE_EVENT(task_newtask,
+
+	TP_PROTO(struct task_struct *task, u64 clone_flags),
+
+	TP_ARGS(task, clone_flags),
+
+	TP_STRUCT__entry(
+		__field(	pid_t,	pid)
+		__array(	char,	comm, TASK_COMM_LEN)
+		__field(	u64,    clone_flags)
+		__field(	short,	oom_score_adj)
+	),
+
+	TP_fast_assign(
+		__entry->pid = task->pid;
+		memcpy(__entry->comm, task->comm, TASK_COMM_LEN);
+		__entry->clone_flags = clone_flags;
+		__entry->oom_score_adj = task->signal->oom_score_adj;
+	),
+
+	TP_printk("pid=%d comm=%s clone_flags=%llx oom_score_adj=%hd",
+		__entry->pid, __entry->comm,
+		__entry->clone_flags, __entry->oom_score_adj)
+);
+
+TRACE_EVENT(task_rename,
+
+	TP_PROTO(struct task_struct *task, const char *comm),
+
+	TP_ARGS(task, comm),
+
+	TP_STRUCT__entry(
+		__field(	pid_t,	pid)
+		__array(	char, oldcomm,  TASK_COMM_LEN)
+		__array(	char, newcomm,  TASK_COMM_LEN)
+		__field(	short,	oom_score_adj)
+	),
+
+	TP_fast_assign(
+		__entry->pid = task->pid;
+		memcpy(entry->oldcomm, task->comm, TASK_COMM_LEN);
+		strscpy(entry->newcomm, comm, TASK_COMM_LEN);
+		__entry->oom_score_adj = task->signal->oom_score_adj;
+	),
+
+	TP_printk("pid=%d oldcomm=%s newcomm=%s oom_score_adj=%hd",
+		__entry->pid, __entry->oldcomm,
+		__entry->newcomm, __entry->oom_score_adj)
+);
+
+/**
+ * task_prctl_unknown - called on unknown prctl() option
+ * @option:	option passed
+ * @arg2:	arg2 passed
+ * @arg3:	arg3 passed
+ * @arg4:	arg4 passed
+ * @arg5:	arg5 passed
+ *
+ * Called on an unknown prctl() option.
+ */
+TRACE_EVENT(task_prctl_unknown,
+
+	TP_PROTO(int option, unsigned long arg2, unsigned long arg3,
+		 unsigned long arg4, unsigned long arg5),
+
+	TP_ARGS(option, arg2, arg3, arg4, arg5),
+
+	TP_STRUCT__entry(
+		__field(	int,		option)
+		__field(	unsigned long,	arg2)
+		__field(	unsigned long,	arg3)
+		__field(	unsigned long,	arg4)
+		__field(	unsigned long,	arg5)
+	),
+
+	TP_fast_assign(
+		__entry->option = option;
+		__entry->arg2 = arg2;
+		__entry->arg3 = arg3;
+		__entry->arg4 = arg4;
+		__entry->arg5 = arg5;
+	),
+
+	TP_printk("option=%d arg2=%ld arg3=%ld arg4=%ld arg5=%ld",
+		  __entry->option, __entry->arg2, __entry->arg3, __entry->arg4, __entry->arg5)
+);
+
+#endif
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>