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, 58 insertions, 0 deletions

diff --git a/drivers/nvmem/layouts/fixed-layout.c b/drivers/nvmem/layouts/fixed-layout.c
new file mode 100644
index 000000000..635d448b3
--- /dev/null
+++ b/drivers/nvmem/layouts/fixed-layout.c
@@ -0,0 +1,58 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright 2026 Bootlin
+ *
+ * Authors: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
+ */
+
+#include <linux/nvmem-provider.h>
+#include <linux/of.h>
+
+#include "../internals.h"
+
+static int fixed_layout_add_cells(struct nvmem_layout *layout)
+{
+	struct device_node *np;
+	int ret;
+
+	np = of_nvmem_layout_get_container(layout->nvmem);
+	if (!np)
+		return -ENOENT;
+
+	ret = nvmem_add_cells_from_dt(layout->nvmem, np);
+	of_node_put(np);
+
+	return ret;
+}
+
+static int fixed_layout_probe(struct nvmem_layout *layout)
+{
+	layout->add_cells = fixed_layout_add_cells;
+
+	return nvmem_layout_register(layout);
+}
+
+static void fixed_layout_remove(struct nvmem_layout *layout)
+{
+	nvmem_layout_unregister(layout);
+}
+
+static const struct of_device_id fixed_layout_of_match_table[] = {
+	{ .compatible = "fixed-layout", },
+	{},
+};
+
+static struct nvmem_layout_driver fixed_layout_layout = {
+	.driver = {
+		.name = "fixed-layout",
+		.of_match_table = fixed_layout_of_match_table,
+	},
+	.probe = fixed_layout_probe,
+	.remove = fixed_layout_remove,
+};
+module_nvmem_layout_driver(fixed_layout_layout);
+
+MODULE_AUTHOR("Mathieu Dubois-Briand");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(of, fixed_layout_of_match_table);
+MODULE_DESCRIPTION("NVMEM fixed-layout driver");