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, 126 insertions, 0 deletions

diff --git a/drivers/input/misc/tps6594-pwrbutton.c b/drivers/input/misc/tps6594-pwrbutton.c
new file mode 100644
index 000000000..cd039b386
--- /dev/null
+++ b/drivers/input/misc/tps6594-pwrbutton.c
@@ -0,0 +1,126 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * power button driver for TI TPS6594 PMICs
+ *
+ * Copyright (C) 2025 Critical Link LLC - https://www.criticallink.com/
+ */
+#include <linux/init.h>
+#include <linux/input.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/mfd/tps6594.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+#include <linux/slab.h>
+
+struct tps6594_pwrbutton {
+	struct device *dev;
+	struct input_dev *idev;
+	char phys[32];
+};
+
+static irqreturn_t tps6594_pb_push_irq(int irq, void *_pwr)
+{
+	struct tps6594_pwrbutton *pwr = _pwr;
+
+	input_report_key(pwr->idev, KEY_POWER, 1);
+	pm_wakeup_event(pwr->dev, 0);
+	input_sync(pwr->idev);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t tps6594_pb_release_irq(int irq, void *_pwr)
+{
+	struct tps6594_pwrbutton *pwr = _pwr;
+
+	input_report_key(pwr->idev, KEY_POWER, 0);
+	input_sync(pwr->idev);
+
+	return IRQ_HANDLED;
+}
+
+static int tps6594_pb_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct tps6594_pwrbutton *pwr;
+	struct input_dev *idev;
+	int error;
+	int push_irq;
+	int release_irq;
+
+	pwr = devm_kzalloc(dev, sizeof(*pwr), GFP_KERNEL);
+	if (!pwr)
+		return -ENOMEM;
+
+	idev = devm_input_allocate_device(dev);
+	if (!idev)
+		return -ENOMEM;
+
+	idev->name = pdev->name;
+	snprintf(pwr->phys, sizeof(pwr->phys), "%s/input0",
+		 pdev->name);
+	idev->phys = pwr->phys;
+	idev->id.bustype = BUS_I2C;
+
+	input_set_capability(idev, EV_KEY, KEY_POWER);
+
+	pwr->dev = dev;
+	pwr->idev = idev;
+	device_init_wakeup(dev, true);
+
+	push_irq = platform_get_irq(pdev, 0);
+	if (push_irq < 0)
+		return -EINVAL;
+
+	release_irq = platform_get_irq(pdev, 1);
+	if (release_irq < 0)
+		return -EINVAL;
+
+	error = devm_request_threaded_irq(dev, push_irq, NULL,
+					  tps6594_pb_push_irq,
+					  IRQF_ONESHOT,
+					  pdev->resource[0].name, pwr);
+	if (error) {
+		dev_err(dev, "failed to request push IRQ #%d: %d\n", push_irq,
+			error);
+		return error;
+	}
+
+	error = devm_request_threaded_irq(dev, release_irq, NULL,
+					  tps6594_pb_release_irq,
+					  IRQF_ONESHOT,
+					  pdev->resource[1].name, pwr);
+	if (error) {
+		dev_err(dev, "failed to request release IRQ #%d: %d\n",
+			release_irq, error);
+		return error;
+	}
+
+	error = input_register_device(idev);
+	if (error) {
+		dev_err(dev, "Can't register power button: %d\n", error);
+		return error;
+	}
+
+	return 0;
+}
+
+static const struct platform_device_id tps6594_pwrbtn_id_table[] = {
+	{ "tps6594-pwrbutton", },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(platform, tps6594_pwrbtn_id_table);
+
+static struct platform_driver tps6594_pb_driver = {
+	.probe = tps6594_pb_probe,
+	.driver = {
+		.name = "tps6594_pwrbutton",
+	},
+	.id_table = tps6594_pwrbtn_id_table,
+};
+module_platform_driver(tps6594_pb_driver);
+
+MODULE_DESCRIPTION("TPS6594 Power Button");
+MODULE_LICENSE("GPL");