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, 73 insertions, 0 deletions

diff --git a/include/rv/kunit.h b/include/rv/kunit.h
new file mode 100644
index 000000000..31e0b93c4
--- /dev/null
+++ b/include/rv/kunit.h
@@ -0,0 +1,73 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2026-2029 Red Hat, Inc. Gabriele Monaco <gmonaco@redhat.com>
+ *
+ * Declaration of wrappers to allow mocking core functionality, like current,
+ * and other testing utilities.
+ * Necessary only when mocking may be needed. If the RV KUnit test is
+ * enabled, the wrappers incur an additional function call overhead.
+ */
+
+#ifndef _RV_KUNIT_H
+#define _RV_KUNIT_H
+
+#if IS_ENABLED(CONFIG_RV_MONITORS_KUNIT_TEST)
+
+#include <kunit/test.h>
+#include <kunit/test-bug.h>
+#include <linux/delay.h>
+
+int rv_set_testing(struct kunit_suite *suite);
+void rv_clear_testing(struct kunit_suite *suite);
+
+#define RV_KUNIT_MAX_MOCK_TASKS 8
+
+struct rv_kunit_ctx {
+	int reactions, expected;
+	int mock_task_count;
+	struct task_struct *mock_tasks[RV_KUNIT_MAX_MOCK_TASKS];
+};
+
+#define RV_KUNIT_EXPECT_REACTION(test, ctx)                             \
+	do {                                                            \
+		KUNIT_EXPECT_EQ(test, ctx->reactions, ++ctx->expected); \
+		if (ctx->reactions != ctx->expected)                    \
+			ctx->expected = ctx->reactions;                 \
+	} while (0)
+
+#define RV_KUNIT_EXPECT_NO_REACTION(test, ctx)                        \
+	do {                                                          \
+		KUNIT_EXPECT_EQ(test, ctx->reactions, ctx->expected); \
+		if (ctx->reactions != ctx->expected)                  \
+			ctx->expected = ctx->reactions;               \
+	} while (0)
+
+#define RV_KUNIT_EXPECT_REACTION_HERE(test, ctx)                             \
+	for (int __done = ({ RV_KUNIT_EXPECT_NO_REACTION(test, ctx); 0; });  \
+	     !__done;                                                        \
+	     __done = ({ RV_KUNIT_EXPECT_REACTION(test, ctx); 1; }))
+
+struct rv_kunit_mon {
+	struct rv_monitor *rv_this;
+	int (*monitor_init)(void);
+	void (*monitor_destroy)(void);
+	bool is_per_task;
+	int *task_slot;
+	void (*task_reset)(struct task_struct *task);
+};
+
+void prepare_test(struct kunit *test, const struct rv_kunit_mon *mon);
+void teardown_test(void *arg);
+struct task_struct *rv_kunit_alloc_mock_task(struct kunit *test);
+
+void rv_mock_current(struct task_struct *tsk);
+struct task_struct *rv_get_mock_current(void);
+
+#define rv_get_current() (unlikely(kunit_get_current_test()) ? rv_get_mock_current() : current)
+
+#else /* !CONFIG_RV_MONITORS_KUNIT_TEST */
+
+#define rv_get_current() current
+
+#endif /* CONFIG_RV_MONITORS_KUNIT_TEST */
+#endif /* _RV_KUNIT_H */