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, 99 insertions, 0 deletions

diff --git a/drivers/nvmem/u-boot-env.c b/drivers/nvmem/u-boot-env.c
new file mode 100644
index 000000000..467b28891
--- /dev/null
+++ b/drivers/nvmem/u-boot-env.c
@@ -0,0 +1,99 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2022 Rafał Miłecki <rafal@milecki.pl>
+ */
+
+#include <linux/module.h>
+#include <linux/mtd/mtd.h>
+#include <linux/nvmem-provider.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+
+#include "layouts/u-boot-env.h"
+
+struct u_boot_env {
+	struct device *dev;
+	struct nvmem_device *nvmem;
+	enum u_boot_env_format format;
+
+	struct mtd_info *mtd;
+};
+
+static int u_boot_env_read(void *context, unsigned int offset, void *val,
+			   size_t bytes)
+{
+	struct u_boot_env *priv = context;
+	struct device *dev = priv->dev;
+	size_t bytes_read;
+	int err;
+
+	err = mtd_read(priv->mtd, offset, bytes, &bytes_read, val);
+	if (err && !mtd_is_bitflip(err)) {
+		dev_err(dev, "Failed to read from mtd: %d\n", err);
+		return err;
+	}
+
+	if (bytes_read != bytes) {
+		dev_err(dev, "Failed to read %zu bytes\n", bytes);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static int u_boot_env_probe(struct platform_device *pdev)
+{
+	struct nvmem_config config = {
+		.name = "u-boot-env",
+		.reg_read = u_boot_env_read,
+	};
+	struct device *dev = &pdev->dev;
+	struct device_node *np = dev->of_node;
+	struct u_boot_env *priv;
+
+	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+	priv->dev = dev;
+
+	priv->format = (uintptr_t)of_device_get_match_data(dev);
+
+	priv->mtd = of_get_mtd_device_by_node(np);
+	if (IS_ERR(priv->mtd)) {
+		dev_err_probe(dev, PTR_ERR(priv->mtd), "Failed to get %pOF MTD\n", np);
+		return PTR_ERR(priv->mtd);
+	}
+
+	config.dev = dev;
+	config.priv = priv;
+	config.size = priv->mtd->size;
+
+	priv->nvmem = devm_nvmem_register(dev, &config);
+	if (IS_ERR(priv->nvmem))
+		return PTR_ERR(priv->nvmem);
+
+	return u_boot_env_parse(dev, priv->nvmem, priv->format);
+}
+
+static const struct of_device_id u_boot_env_of_match_table[] = {
+	{ .compatible = "u-boot,env", .data = (void *)U_BOOT_FORMAT_SINGLE, },
+	{ .compatible = "u-boot,env-redundant-bool", .data = (void *)U_BOOT_FORMAT_REDUNDANT, },
+	{ .compatible = "u-boot,env-redundant-count", .data = (void *)U_BOOT_FORMAT_REDUNDANT, },
+	{ .compatible = "brcm,env", .data = (void *)U_BOOT_FORMAT_BROADCOM, },
+	{},
+};
+
+static struct platform_driver u_boot_env_driver = {
+	.probe = u_boot_env_probe,
+	.driver = {
+		.name = "u_boot_env",
+		.of_match_table = u_boot_env_of_match_table,
+	},
+};
+module_platform_driver(u_boot_env_driver);
+
+MODULE_AUTHOR("Rafał Miłecki");
+MODULE_DESCRIPTION("U-Boot environment variables support module");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(of, u_boot_env_of_match_table);