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, 228 insertions, 0 deletions

diff --git a/include/trace/events/objagg.h b/include/trace/events/objagg.h
new file mode 100644
index 000000000..fcec0fc9e
--- /dev/null
+++ b/include/trace/events/objagg.h
@@ -0,0 +1,228 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 */
+/* Copyright (c) 2018 Mellanox Technologies. All rights reserved */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM objagg
+
+#if !defined(__TRACE_OBJAGG_H) || defined(TRACE_HEADER_MULTI_READ)
+#define __TRACE_OBJAGG_H
+
+#include <linux/tracepoint.h>
+
+struct objagg;
+struct objagg_obj;
+
+TRACE_EVENT(objagg_create,
+	TP_PROTO(const struct objagg *objagg),
+
+	TP_ARGS(objagg),
+
+	TP_STRUCT__entry(
+		__field(const void *, objagg)
+	),
+
+	TP_fast_assign(
+		__entry->objagg = objagg;
+	),
+
+	TP_printk("objagg %p", __entry->objagg)
+);
+
+TRACE_EVENT(objagg_destroy,
+	TP_PROTO(const struct objagg *objagg),
+
+	TP_ARGS(objagg),
+
+	TP_STRUCT__entry(
+		__field(const void *, objagg)
+	),
+
+	TP_fast_assign(
+		__entry->objagg = objagg;
+	),
+
+	TP_printk("objagg %p", __entry->objagg)
+);
+
+TRACE_EVENT(objagg_obj_create,
+	TP_PROTO(const struct objagg *objagg,
+		 const struct objagg_obj *obj),
+
+	TP_ARGS(objagg, obj),
+
+	TP_STRUCT__entry(
+		__field(const void *, objagg)
+		__field(const void *, obj)
+	),
+
+	TP_fast_assign(
+		__entry->objagg = objagg;
+		__entry->obj = obj;
+	),
+
+	TP_printk("objagg %p, obj %p", __entry->objagg, __entry->obj)
+);
+
+TRACE_EVENT(objagg_obj_destroy,
+	TP_PROTO(const struct objagg *objagg,
+		 const struct objagg_obj *obj),
+
+	TP_ARGS(objagg, obj),
+
+	TP_STRUCT__entry(
+		__field(const void *, objagg)
+		__field(const void *, obj)
+	),
+
+	TP_fast_assign(
+		__entry->objagg = objagg;
+		__entry->obj = obj;
+	),
+
+	TP_printk("objagg %p, obj %p", __entry->objagg, __entry->obj)
+);
+
+TRACE_EVENT(objagg_obj_get,
+	TP_PROTO(const struct objagg *objagg,
+		 const struct objagg_obj *obj,
+		 unsigned int refcount),
+
+	TP_ARGS(objagg, obj, refcount),
+
+	TP_STRUCT__entry(
+		__field(const void *, objagg)
+		__field(const void *, obj)
+		__field(unsigned int, refcount)
+	),
+
+	TP_fast_assign(
+		__entry->objagg = objagg;
+		__entry->obj = obj;
+		__entry->refcount = refcount;
+	),
+
+	TP_printk("objagg %p, obj %p, refcount %u",
+		  __entry->objagg, __entry->obj, __entry->refcount)
+);
+
+TRACE_EVENT(objagg_obj_put,
+	TP_PROTO(const struct objagg *objagg,
+		 const struct objagg_obj *obj,
+		 unsigned int refcount),
+
+	TP_ARGS(objagg, obj, refcount),
+
+	TP_STRUCT__entry(
+		__field(const void *, objagg)
+		__field(const void *, obj)
+		__field(unsigned int, refcount)
+	),
+
+	TP_fast_assign(
+		__entry->objagg = objagg;
+		__entry->obj = obj;
+		__entry->refcount = refcount;
+	),
+
+	TP_printk("objagg %p, obj %p, refcount %u",
+		  __entry->objagg, __entry->obj, __entry->refcount)
+);
+
+TRACE_EVENT(objagg_obj_parent_assign,
+	TP_PROTO(const struct objagg *objagg,
+		 const struct objagg_obj *obj,
+		 const struct objagg_obj *parent,
+		 unsigned int parent_refcount),
+
+	TP_ARGS(objagg, obj, parent, parent_refcount),
+
+	TP_STRUCT__entry(
+		__field(const void *, objagg)
+		__field(const void *, obj)
+		__field(const void *, parent)
+		__field(unsigned int, parent_refcount)
+	),
+
+	TP_fast_assign(
+		__entry->objagg = objagg;
+		__entry->obj = obj;
+		__entry->parent = parent;
+		__entry->parent_refcount = parent_refcount;
+	),
+
+	TP_printk("objagg %p, obj %p, parent %p, parent_refcount %u",
+		  __entry->objagg, __entry->obj,
+		  __entry->parent, __entry->parent_refcount)
+);
+
+TRACE_EVENT(objagg_obj_parent_unassign,
+	TP_PROTO(const struct objagg *objagg,
+		 const struct objagg_obj *obj,
+		 const struct objagg_obj *parent,
+		 unsigned int parent_refcount),
+
+	TP_ARGS(objagg, obj, parent, parent_refcount),
+
+	TP_STRUCT__entry(
+		__field(const void *, objagg)
+		__field(const void *, obj)
+		__field(const void *, parent)
+		__field(unsigned int, parent_refcount)
+	),
+
+	TP_fast_assign(
+		__entry->objagg = objagg;
+		__entry->obj = obj;
+		__entry->parent = parent;
+		__entry->parent_refcount = parent_refcount;
+	),
+
+	TP_printk("objagg %p, obj %p, parent %p, parent_refcount %u",
+		  __entry->objagg, __entry->obj,
+		  __entry->parent, __entry->parent_refcount)
+);
+
+TRACE_EVENT(objagg_obj_root_create,
+	TP_PROTO(const struct objagg *objagg,
+		 const struct objagg_obj *obj),
+
+	TP_ARGS(objagg, obj),
+
+	TP_STRUCT__entry(
+		__field(const void *, objagg)
+		__field(const void *, obj)
+	),
+
+	TP_fast_assign(
+		__entry->objagg = objagg;
+		__entry->obj = obj;
+	),
+
+	TP_printk("objagg %p, obj %p",
+		  __entry->objagg, __entry->obj)
+);
+
+TRACE_EVENT(objagg_obj_root_destroy,
+	TP_PROTO(const struct objagg *objagg,
+		 const struct objagg_obj *obj),
+
+	TP_ARGS(objagg, obj),
+
+	TP_STRUCT__entry(
+		__field(const void *, objagg)
+		__field(const void *, obj)
+	),
+
+	TP_fast_assign(
+		__entry->objagg = objagg;
+		__entry->obj = obj;
+	),
+
+	TP_printk("objagg %p, obj %p",
+		  __entry->objagg, __entry->obj)
+);
+
+#endif /* __TRACE_OBJAGG_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>