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, 111 insertions, 0 deletions

diff --git a/drivers/reset/reset-berlin.c b/drivers/reset/reset-berlin.c
new file mode 100644
index 000000000..578fe8670
--- /dev/null
+++ b/drivers/reset/reset-berlin.c
@@ -0,0 +1,111 @@
+/*
+ * Copyright (C) 2014 Marvell Technology Group Ltd.
+ *
+ * Marvell Berlin reset driver
+ *
+ * Antoine Tenart <antoine.tenart@free-electrons.com>
+ * Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/delay.h>
+#include <linux/io.h>
+#include <linux/mfd/syscon.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+#include <linux/reset-controller.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+
+#define BERLIN_MAX_RESETS	32
+
+#define to_berlin_reset_priv(p)		\
+	container_of((p), struct berlin_reset_priv, rcdev)
+
+struct berlin_reset_priv {
+	struct regmap			*regmap;
+	struct reset_controller_dev	rcdev;
+};
+
+static int berlin_reset_reset(struct reset_controller_dev *rcdev,
+			      unsigned long id)
+{
+	struct berlin_reset_priv *priv = to_berlin_reset_priv(rcdev);
+	int offset = id >> 8;
+	int mask = BIT(id & 0x1f);
+
+	regmap_write(priv->regmap, offset, mask);
+
+	/* let the reset be effective */
+	udelay(10);
+
+	return 0;
+}
+
+static const struct reset_control_ops berlin_reset_ops = {
+	.reset	= berlin_reset_reset,
+};
+
+static int berlin_reset_xlate(struct reset_controller_dev *rcdev,
+			      const struct of_phandle_args *reset_spec)
+{
+	unsigned int offset, bit;
+
+	offset = reset_spec->args[0];
+	bit = reset_spec->args[1];
+
+	if (bit >= BERLIN_MAX_RESETS)
+		return -EINVAL;
+
+	return (offset << 8) | bit;
+}
+
+static int berlin2_reset_probe(struct platform_device *pdev)
+{
+	struct device_node *parent_np;
+	struct berlin_reset_priv *priv;
+
+	priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	parent_np = of_get_parent(pdev->dev.of_node);
+	priv->regmap = syscon_node_to_regmap(parent_np);
+	of_node_put(parent_np);
+	if (IS_ERR(priv->regmap))
+		return PTR_ERR(priv->regmap);
+
+	priv->rcdev.owner = THIS_MODULE;
+	priv->rcdev.ops = &berlin_reset_ops;
+	priv->rcdev.of_node = pdev->dev.of_node;
+	priv->rcdev.of_reset_n_cells = 2;
+	priv->rcdev.of_xlate = berlin_reset_xlate;
+
+	return reset_controller_register(&priv->rcdev);
+}
+
+static const struct of_device_id berlin_reset_dt_match[] = {
+	{ .compatible = "marvell,berlin2-reset" },
+	{ },
+};
+MODULE_DEVICE_TABLE(of, berlin_reset_dt_match);
+
+static struct platform_driver berlin_reset_driver = {
+	.probe	= berlin2_reset_probe,
+	.driver	= {
+		.name = "berlin2-reset",
+		.of_match_table = berlin_reset_dt_match,
+	},
+};
+module_platform_driver(berlin_reset_driver);
+
+MODULE_AUTHOR("Antoine Tenart <antoine.tenart@free-electrons.com>");
+MODULE_AUTHOR("Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com>");
+MODULE_DESCRIPTION("Synaptics Berlin reset controller");
+MODULE_LICENSE("GPL");