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, 88 insertions, 0 deletions

diff --git a/tools/perf/util/maps.h b/tools/perf/util/maps.h
new file mode 100644
index 000000000..4ec9b7453
--- /dev/null
+++ b/tools/perf/util/maps.h
@@ -0,0 +1,88 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __PERF_MAPS_H
+#define __PERF_MAPS_H
+
+#include <linux/refcount.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include <linux/types.h>
+
+struct ref_reloc_sym;
+struct machine;
+struct map;
+struct maps;
+
+#define KMAP_NAME_LEN 256
+
+struct kmap {
+	struct ref_reloc_sym *ref_reloc_sym;
+	struct maps	     *kmaps;
+	char		     name[KMAP_NAME_LEN];
+};
+
+struct maps *maps__new(struct machine *machine);
+bool maps__empty(struct maps *maps);
+int maps__copy_from(struct maps *maps, struct maps *parent);
+
+struct maps *maps__get(struct maps *maps);
+void maps__put(struct maps *maps);
+
+static inline void __maps__zput(struct maps **map)
+{
+	maps__put(*map);
+	*map = NULL;
+}
+
+#define maps__zput(map) __maps__zput(&map)
+
+bool maps__equal(struct maps *a, struct maps *b);
+
+/* Iterate over map calling cb for each entry. */
+int maps__for_each_map(struct maps *maps, int (*cb)(struct map *map, void *data), void *data);
+/* Iterate over map removing an entry if cb returns true. */
+void maps__remove_maps(struct maps *maps, bool (*cb)(struct map *map, void *data), void *data);
+
+struct machine *maps__machine(const struct maps *maps);
+unsigned int maps__nr_maps(const struct maps *maps); /* Test only. */
+refcount_t *maps__refcnt(struct maps *maps); /* Test only. */
+
+#ifdef HAVE_LIBUNWIND_SUPPORT
+void *maps__addr_space(const struct maps *maps);
+void maps__set_addr_space(struct maps *maps, void *addr_space);
+uint16_t maps__e_machine(const struct maps *maps);
+void maps__set_e_machine(struct maps *maps, uint16_t e_machine);
+#endif
+#ifdef HAVE_LIBDW_SUPPORT
+void *maps__libdw_addr_space_dwfl(const struct maps *maps);
+void maps__set_libdw_addr_space_dwfl(struct maps *maps, void *dwfl);
+#endif
+
+size_t maps__fprintf(struct maps *maps, FILE *fp);
+
+int maps__load_maps(struct maps *maps);
+int maps__insert(struct maps *maps, struct map *map);
+void maps__remove(struct maps *maps, struct map *map);
+int maps__mutate_mapping(struct maps *maps, struct map *map,
+			 int (*mutate_cb)(struct map *map, void *data), void *data);
+
+struct map *maps__find(struct maps *maps, u64 addr);
+struct symbol *maps__find_symbol(struct maps *maps, u64 addr, struct map **mapp);
+struct symbol *maps__find_symbol_by_name(struct maps *maps, const char *name, struct map **mapp);
+
+struct addr_map_symbol;
+
+int maps__find_ams(struct maps *maps, struct addr_map_symbol *ams);
+
+int maps__fixup_overlap_and_insert(struct maps *maps, struct map *new);
+
+struct map *maps__find_by_name(struct maps *maps, const char *name);
+
+struct map *maps__find_next_entry(struct maps *maps, struct map *map);
+
+int maps__merge_in(struct maps *kmaps, struct map *new_map);
+
+void maps__fixup_end(struct maps *maps);
+
+void maps__load_first(struct maps *maps);
+
+#endif // __PERF_MAPS_H