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, 122 insertions, 0 deletions

diff --git a/drivers/mfd/rt5120.c b/drivers/mfd/rt5120.c
new file mode 100644
index 000000000..a229eb292
--- /dev/null
+++ b/drivers/mfd/rt5120.c
@@ -0,0 +1,122 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2022 Richtek Technology Corp.
+ * Author: ChiYuan Huang <cy_huang@richtek.com>
+ */
+
+#include <linux/i2c.h>
+#include <linux/kernel.h>
+#include <linux/mfd/core.h>
+#include <linux/module.h>
+#include <linux/regmap.h>
+
+#define RT5120_REG_INTENABLE	0x1D
+#define RT5120_REG_INTSTAT	0x1E
+#define RT5120_REG_FZCMODE	0x44
+
+#define RT5120_INT_HOTDIE	0
+#define RT5120_INT_PWRKEY_REL	5
+#define RT5120_INT_PWRKEY_PRESS	6
+
+static const struct regmap_range rt5120_rd_yes_ranges[] = {
+	regmap_reg_range(0x03, 0x13),
+	regmap_reg_range(0x1c, 0x20),
+	regmap_reg_range(0x44, 0x44),
+};
+
+static const struct regmap_range rt5120_wr_yes_ranges[] = {
+	regmap_reg_range(0x06, 0x13),
+	regmap_reg_range(0x1c, 0x20),
+	regmap_reg_range(0x44, 0x44),
+};
+
+static const struct regmap_access_table rt5120_rd_table = {
+	.yes_ranges = rt5120_rd_yes_ranges,
+	.n_yes_ranges = ARRAY_SIZE(rt5120_rd_yes_ranges),
+};
+
+static const struct regmap_access_table rt5120_wr_table = {
+	.yes_ranges = rt5120_wr_yes_ranges,
+	.n_yes_ranges = ARRAY_SIZE(rt5120_wr_yes_ranges),
+};
+
+static const struct regmap_config rt5120_regmap_config = {
+	.reg_bits = 8,
+	.val_bits = 8,
+	.max_register = RT5120_REG_FZCMODE,
+
+	.wr_table = &rt5120_wr_table,
+	.rd_table = &rt5120_rd_table,
+};
+
+static const struct regmap_irq rt5120_irqs[] = {
+	REGMAP_IRQ_REG_LINE(RT5120_INT_HOTDIE, 8),
+	REGMAP_IRQ_REG_LINE(RT5120_INT_PWRKEY_REL, 8),
+	REGMAP_IRQ_REG_LINE(RT5120_INT_PWRKEY_PRESS, 8),
+};
+
+static const struct regmap_irq_chip rt5120_irq_chip = {
+	.name = "rt5120-pmic",
+	.status_base = RT5120_REG_INTSTAT,
+	.unmask_base = RT5120_REG_INTENABLE,
+	.ack_base = RT5120_REG_INTSTAT,
+	.use_ack = true,
+	.num_regs = 1,
+	.irqs = rt5120_irqs,
+	.num_irqs = ARRAY_SIZE(rt5120_irqs),
+};
+
+static const struct resource rt5120_regulator_resources[] = {
+	DEFINE_RES_IRQ(RT5120_INT_HOTDIE),
+};
+
+static const struct resource rt5120_pwrkey_resources[] = {
+	DEFINE_RES_IRQ_NAMED(RT5120_INT_PWRKEY_PRESS, "pwrkey-press"),
+	DEFINE_RES_IRQ_NAMED(RT5120_INT_PWRKEY_REL, "pwrkey-release"),
+};
+
+static const struct mfd_cell rt5120_devs[] = {
+	MFD_CELL_RES("rt5120-regulator", rt5120_regulator_resources),
+	MFD_CELL_OF("rt5120-pwrkey", rt5120_pwrkey_resources, NULL, 0, 0, "richtek,rt5120-pwrkey"),
+};
+
+static int rt5120_probe(struct i2c_client *i2c)
+{
+	struct device *dev = &i2c->dev;
+	struct regmap *regmap;
+	struct regmap_irq_chip_data *irq_data;
+	int ret;
+
+	regmap = devm_regmap_init_i2c(i2c, &rt5120_regmap_config);
+	if (IS_ERR(regmap))
+		return dev_err_probe(dev, PTR_ERR(regmap),
+				     "Failed to init regmap\n");
+
+	ret = devm_regmap_add_irq_chip(dev, regmap, i2c->irq, IRQF_ONESHOT, 0,
+				       &rt5120_irq_chip, &irq_data);
+	if (ret)
+		return dev_err_probe(dev, ret, "Failed to add IRQ chip\n");
+
+	return devm_mfd_add_devices(dev, PLATFORM_DEVID_AUTO, rt5120_devs,
+				    ARRAY_SIZE(rt5120_devs), NULL, 0,
+				    regmap_irq_get_domain(irq_data));
+}
+
+static const struct of_device_id rt5120_device_match_table[] = {
+	{ .compatible = "richtek,rt5120" },
+	{}
+};
+MODULE_DEVICE_TABLE(of, rt5120_device_match_table);
+
+static struct i2c_driver rt5120_driver = {
+	.driver = {
+		.name = "rt5120",
+		.of_match_table = rt5120_device_match_table,
+	},
+	.probe = rt5120_probe,
+};
+module_i2c_driver(rt5120_driver);
+
+MODULE_AUTHOR("ChiYuan Huang <cy_huang@richtek.com>");
+MODULE_DESCRIPTION("Richtek RT5120 I2C driver");
+MODULE_LICENSE("GPL v2");