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, 70 insertions, 0 deletions

diff --git a/io_uring/alloc_cache.h b/io_uring/alloc_cache.h
new file mode 100644
index 000000000..962b6e2d0
--- /dev/null
+++ b/io_uring/alloc_cache.h
@@ -0,0 +1,70 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef IOU_ALLOC_CACHE_H
+#define IOU_ALLOC_CACHE_H
+
+#include <linux/io_uring_types.h>
+#include <linux/kasan.h>
+
+/*
+ * Don't allow the cache to grow beyond this size.
+ */
+#define IO_ALLOC_CACHE_MAX	128
+
+void io_alloc_cache_free(struct io_alloc_cache *cache,
+			 void (*free)(const void *));
+bool io_alloc_cache_init(struct io_alloc_cache *cache,
+			 unsigned max_nr, unsigned int size,
+			 unsigned int init_bytes);
+
+void *io_cache_alloc_new(struct io_alloc_cache *cache, gfp_t gfp);
+
+static inline bool io_alloc_cache_put(struct io_alloc_cache *cache,
+				      void *entry)
+{
+	if (cache->nr_cached < cache->max_cached) {
+		if (!kasan_mempool_poison_object(entry))
+			return false;
+		cache->entries[cache->nr_cached++] = entry;
+		return true;
+	}
+	return false;
+}
+
+static inline void *io_alloc_cache_get(struct io_alloc_cache *cache)
+{
+	if (cache->nr_cached) {
+		void *entry = cache->entries[--cache->nr_cached];
+
+		/*
+		 * If KASAN is enabled, always clear the initial bytes that
+		 * must be zeroed post alloc, in case any of them overlap
+		 * with KASAN storage.
+		 */
+#if defined(CONFIG_KASAN)
+		kasan_mempool_unpoison_object(entry, cache->elem_size);
+		if (cache->init_clear)
+			memset(entry, 0, cache->init_clear);
+#endif
+		return entry;
+	}
+
+	return NULL;
+}
+
+static inline void *io_cache_alloc(struct io_alloc_cache *cache, gfp_t gfp)
+{
+	void *obj;
+
+	obj = io_alloc_cache_get(cache);
+	if (obj)
+		return obj;
+	return io_cache_alloc_new(cache, gfp);
+}
+
+static inline void io_cache_free(struct io_alloc_cache *cache, void *obj)
+{
+	if (!io_alloc_cache_put(cache, obj))
+		kvfree(obj);
+}
+
+#endif