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, 76 insertions, 0 deletions

diff --git a/tools/perf/util/block-range.h b/tools/perf/util/block-range.h
new file mode 100644
index 000000000..ec0fb534b
--- /dev/null
+++ b/tools/perf/util/block-range.h
@@ -0,0 +1,76 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __PERF_BLOCK_RANGE_H
+#define __PERF_BLOCK_RANGE_H
+
+#include <stdbool.h>
+#include <linux/rbtree.h>
+#include <linux/types.h>
+
+struct symbol;
+
+/*
+ * struct block_range - non-overlapping parts of basic blocks
+ * @node:	treenode
+ * @start:	inclusive start of range
+ * @end:	inclusive end of range
+ * @is_target:	@start is a jump target
+ * @is_branch:	@end is a branch instruction
+ * @coverage:	number of blocks that cover this range
+ * @taken:	number of times the branch is taken (requires @is_branch)
+ * @pred:	number of times the taken branch was predicted
+ */
+struct block_range {
+	struct rb_node node;
+
+	struct symbol *sym;
+
+	u64 start;
+	u64 end;
+
+	int is_target, is_branch;
+
+	u64 coverage;
+	u64 entry;
+	u64 taken;
+	u64 pred;
+};
+
+static inline struct block_range *block_range__next(struct block_range *br)
+{
+	struct rb_node *n = rb_next(&br->node);
+	if (!n)
+		return NULL;
+	return rb_entry(n, struct block_range, node);
+}
+
+struct block_range_iter {
+	struct block_range *start;
+	struct block_range *end;
+};
+
+static inline struct block_range *block_range_iter(struct block_range_iter *iter)
+{
+	return iter->start;
+}
+
+static inline bool block_range_iter__next(struct block_range_iter *iter)
+{
+	if (iter->start == iter->end)
+		return false;
+
+	iter->start = block_range__next(iter->start);
+	return true;
+}
+
+static inline bool block_range_iter__valid(struct block_range_iter *iter)
+{
+	if (!iter->start || !iter->end)
+		return false;
+	return true;
+}
+
+extern struct block_range *block_range__find(u64 addr);
+extern struct block_range_iter block_range__create(u64 start, u64 end);
+extern double block_range__coverage(struct block_range *br);
+
+#endif /* __PERF_BLOCK_RANGE_H */