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, 112 insertions, 0 deletions

diff --git a/drivers/mfd/rohm-pwrbutton.c b/drivers/mfd/rohm-pwrbutton.c
new file mode 100644
index 000000000..96a6d1f01
--- /dev/null
+++ b/drivers/mfd/rohm-pwrbutton.c
@@ -0,0 +1,112 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Shared helper for ROHM PMIC power button registration
+ *
+ * Copyright 2018, 2019 ROHM Semiconductors
+ * Copyright 2026 Google LLC
+ */
+
+#include <linux/device/devres.h>
+#include <linux/gfp_types.h>
+#include <linux/input.h>
+#include <linux/mfd/core.h>
+#include <linux/module.h>
+#include <linux/property.h>
+#include <linux/slab.h>
+
+#include "rohm-pwrbutton.h"
+
+#define GPIO_KEYS  0	/* Node corresponding to gpio-keys device itself */
+#define PWRON_KEY  1	/* Node describing power button in gpio-keys */
+
+static int rohm_pwrbutton_register_swnodes(const struct software_node *nodes)
+{
+	const struct software_node * const node_group[] = {
+		&nodes[GPIO_KEYS], &nodes[PWRON_KEY], NULL
+	};
+
+	return software_node_register_node_group(node_group);
+}
+
+static void rohm_pwrbutton_unregister_swnodes(void *data)
+{
+	const struct software_node *nodes = data;
+	const struct software_node * const node_group[] = {
+		&nodes[GPIO_KEYS], &nodes[PWRON_KEY], NULL
+	};
+
+	software_node_unregister_node_group(node_group);
+}
+
+int rohm_register_pwrbutton(struct device *dev, int irq, const char *name,
+			    bool wakeup, struct irq_domain *irq_domain)
+{
+	const struct resource res[] = {
+		DEFINE_RES_IRQ_NAMED(irq, name),
+	};
+	struct mfd_cell gpio_keys_cell = {
+		.name = "gpio-keys",
+		.resources = res,
+		.num_resources = ARRAY_SIZE(res),
+	};
+	struct property_entry *parent_props;
+	struct property_entry *child_props;
+	struct software_node *nodes;
+	int n_props;
+	int ret;
+
+	if (irq <= 0)
+		return -EINVAL;
+
+	nodes = devm_kcalloc(dev, 2, sizeof(*nodes), GFP_KERNEL);
+	if (!nodes)
+		return -ENOMEM;
+
+	nodes[GPIO_KEYS].name = devm_kasprintf(dev, GFP_KERNEL, "%s-power-key", dev_name(dev));
+	if (!nodes[GPIO_KEYS].name)
+		return -ENOMEM;
+
+	parent_props = devm_kcalloc(dev, 2, sizeof(*parent_props), GFP_KERNEL);
+	if (!parent_props)
+		return -ENOMEM;
+
+	parent_props[0] = PROPERTY_ENTRY_STRING("label", name);
+	nodes[GPIO_KEYS].properties = parent_props;
+
+	n_props = 2; /* linux,code and terminator */
+	if (wakeup)
+		n_props++;
+
+	child_props = devm_kcalloc(dev, n_props, sizeof(*child_props), GFP_KERNEL);
+	if (!child_props)
+		return -ENOMEM;
+
+	child_props[0] = PROPERTY_ENTRY_U32("linux,code", KEY_POWER);
+	if (wakeup)
+		child_props[1] = PROPERTY_ENTRY_BOOL("wakeup-source");
+
+	nodes[PWRON_KEY].parent = &nodes[GPIO_KEYS];
+	nodes[PWRON_KEY].properties = child_props;
+
+	ret = rohm_pwrbutton_register_swnodes(nodes);
+	if (ret)
+		return ret;
+
+	ret = devm_add_action_or_reset(dev, rohm_pwrbutton_unregister_swnodes, nodes);
+	if (ret)
+		return ret;
+
+	gpio_keys_cell.swnode = &nodes[GPIO_KEYS];
+
+	ret = devm_mfd_add_devices(dev, PLATFORM_DEVID_AUTO, &gpio_keys_cell, 1,
+				   NULL, 0, irq_domain);
+	if (ret)
+		return dev_err_probe(dev, ret, "Failed to register power-button");
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(rohm_register_pwrbutton);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Dmitry Torokhov <dmitry.torokhov@gmail.com>");
+MODULE_DESCRIPTION("Shared helper for ROHM PMIC power button registration");