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, 66 insertions, 0 deletions

diff --git a/include/net/libeth/cache.h b/include/net/libeth/cache.h
new file mode 100644
index 000000000..bdb0c043c
--- /dev/null
+++ b/include/net/libeth/cache.h
@@ -0,0 +1,66 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright (C) 2024 Intel Corporation */
+
+#ifndef __LIBETH_CACHE_H
+#define __LIBETH_CACHE_H
+
+#include <linux/cache.h>
+
+/**
+ * libeth_cacheline_group_assert - make sure cacheline group size is expected
+ * @type: type of the structure containing the group
+ * @grp: group name inside the struct
+ * @sz: expected group size
+ */
+#if defined(CONFIG_64BIT) && SMP_CACHE_BYTES == 64
+#define libeth_cacheline_group_assert(type, grp, sz)			      \
+	static_assert(offsetof(type, __cacheline_group_end__##grp) -	      \
+		      offsetofend(type, __cacheline_group_begin__##grp) ==    \
+		      (sz))
+#define __libeth_cacheline_struct_assert(type, sz)			      \
+	static_assert(sizeof(type) == (sz))
+#else /* !CONFIG_64BIT || SMP_CACHE_BYTES != 64 */
+#define libeth_cacheline_group_assert(type, grp, sz)			      \
+	static_assert(offsetof(type, __cacheline_group_end__##grp) -	      \
+		      offsetofend(type, __cacheline_group_begin__##grp) <=    \
+		      (sz))
+#define __libeth_cacheline_struct_assert(type, sz)			      \
+	static_assert(sizeof(type) <= (sz))
+#endif /* !CONFIG_64BIT || SMP_CACHE_BYTES != 64 */
+
+#define __libeth_cls1(sz1)	SMP_CACHE_ALIGN(sz1)
+#define __libeth_cls2(sz1, sz2)	(SMP_CACHE_ALIGN(sz1) + SMP_CACHE_ALIGN(sz2))
+#define __libeth_cls3(sz1, sz2, sz3)					      \
+	(SMP_CACHE_ALIGN(sz1) + SMP_CACHE_ALIGN(sz2) + SMP_CACHE_ALIGN(sz3))
+#define __libeth_cls(...)						      \
+	CONCATENATE(__libeth_cls, COUNT_ARGS(__VA_ARGS__))(__VA_ARGS__)
+
+/**
+ * libeth_cacheline_struct_assert - make sure CL-based struct size is expected
+ * @type: type of the struct
+ * @...: from 1 to 3 CL group sizes (read-mostly, read-write, cold)
+ *
+ * When a struct contains several CL groups, it's difficult to predict its size
+ * on different architectures. The macro instead takes sizes of all of the
+ * groups the structure contains and generates the final struct size.
+ */
+#define libeth_cacheline_struct_assert(type, ...)			      \
+	__libeth_cacheline_struct_assert(type, __libeth_cls(__VA_ARGS__));    \
+	static_assert(__alignof(type) >= SMP_CACHE_BYTES)
+
+/**
+ * libeth_cacheline_set_assert - make sure CL-based struct layout is expected
+ * @type: type of the struct
+ * @ro: expected size of the read-mostly group
+ * @rw: expected size of the read-write group
+ * @c: expected size of the cold group
+ *
+ * Check that each group size is expected and then do final struct size check.
+ */
+#define libeth_cacheline_set_assert(type, ro, rw, c)			      \
+	libeth_cacheline_group_assert(type, read_mostly, ro);		      \
+	libeth_cacheline_group_assert(type, read_write, rw);		      \
+	libeth_cacheline_group_assert(type, cold, c);			      \
+	libeth_cacheline_struct_assert(type, ro, rw, c)
+
+#endif /* __LIBETH_CACHE_H */