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/net/ieee802154/sysfs.c b/net/ieee802154/sysfs.c
new file mode 100644
index 000000000..6708160eb
--- /dev/null
+++ b/net/ieee802154/sysfs.c
@@ -0,0 +1,111 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *
+ * Authors:
+ * Alexander Aring <aar@pengutronix.de>
+ *
+ * Based on: net/wireless/sysfs.c
+ */
+
+#include <linux/device.h>
+#include <linux/rtnetlink.h>
+
+#include <net/cfg802154.h>
+
+#include "core.h"
+#include "sysfs.h"
+#include "rdev-ops.h"
+
+static inline struct cfg802154_registered_device *
+dev_to_rdev(struct device *dev)
+{
+	return container_of(dev, struct cfg802154_registered_device,
+			    wpan_phy.dev);
+}
+
+#define SHOW_FMT(name, fmt, member)					\
+static ssize_t name ## _show(struct device *dev,			\
+			     struct device_attribute *attr,		\
+			     char *buf)					\
+{									\
+	return sprintf(buf, fmt "\n", dev_to_rdev(dev)->member);	\
+}									\
+static DEVICE_ATTR_RO(name)
+
+SHOW_FMT(index, "%d", wpan_phy_idx);
+
+static ssize_t name_show(struct device *dev,
+			 struct device_attribute *attr,
+			 char *buf)
+{
+	struct wpan_phy *wpan_phy = &dev_to_rdev(dev)->wpan_phy;
+
+	return sprintf(buf, "%s\n", dev_name(&wpan_phy->dev));
+}
+static DEVICE_ATTR_RO(name);
+
+static void wpan_phy_release(struct device *dev)
+{
+	struct cfg802154_registered_device *rdev = dev_to_rdev(dev);
+
+	cfg802154_dev_free(rdev);
+}
+
+static struct attribute *pmib_attrs[] = {
+	&dev_attr_index.attr,
+	&dev_attr_name.attr,
+	NULL,
+};
+ATTRIBUTE_GROUPS(pmib);
+
+#ifdef CONFIG_PM_SLEEP
+static int wpan_phy_suspend(struct device *dev)
+{
+	struct cfg802154_registered_device *rdev = dev_to_rdev(dev);
+	int ret = 0;
+
+	if (rdev->ops->suspend) {
+		rtnl_lock();
+		ret = rdev_suspend(rdev);
+		rtnl_unlock();
+	}
+
+	return ret;
+}
+
+static int wpan_phy_resume(struct device *dev)
+{
+	struct cfg802154_registered_device *rdev = dev_to_rdev(dev);
+	int ret = 0;
+
+	if (rdev->ops->resume) {
+		rtnl_lock();
+		ret = rdev_resume(rdev);
+		rtnl_unlock();
+	}
+
+	return ret;
+}
+
+static SIMPLE_DEV_PM_OPS(wpan_phy_pm_ops, wpan_phy_suspend, wpan_phy_resume);
+#define WPAN_PHY_PM_OPS (&wpan_phy_pm_ops)
+#else
+#define WPAN_PHY_PM_OPS NULL
+#endif
+
+const struct class wpan_phy_class = {
+	.name = "ieee802154",
+	.dev_release = wpan_phy_release,
+	.dev_groups = pmib_groups,
+	.pm = WPAN_PHY_PM_OPS,
+};
+
+int wpan_phy_sysfs_init(void)
+{
+	return class_register(&wpan_phy_class);
+}
+
+void wpan_phy_sysfs_exit(void)
+{
+	class_unregister(&wpan_phy_class);
+}