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, 72 insertions, 0 deletions

diff --git a/net/ipv4/fib_notifier.c b/net/ipv4/fib_notifier.c
new file mode 100644
index 000000000..b1551c265
--- /dev/null
+++ b/net/ipv4/fib_notifier.c
@@ -0,0 +1,72 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/rtnetlink.h>
+#include <linux/notifier.h>
+#include <linux/socket.h>
+#include <linux/kernel.h>
+#include <linux/export.h>
+#include <net/net_namespace.h>
+#include <net/fib_notifier.h>
+#include <net/ip_fib.h>
+
+int call_fib4_notifier(struct notifier_block *nb,
+		       enum fib_event_type event_type,
+		       struct fib_notifier_info *info)
+{
+	info->family = AF_INET;
+	return call_fib_notifier(nb, event_type, info);
+}
+
+int call_fib4_notifiers(struct net *net, enum fib_event_type event_type,
+			struct fib_notifier_info *info)
+{
+	ASSERT_RTNL();
+
+	info->family = AF_INET;
+	/* Paired with READ_ONCE() in fib4_seq_read() */
+	WRITE_ONCE(net->ipv4.fib_seq, net->ipv4.fib_seq + 1);
+	return call_fib_notifiers(net, event_type, info);
+}
+
+static unsigned int fib4_seq_read(const struct net *net)
+{
+	/* Paired with WRITE_ONCE() in call_fib4_notifiers() */
+	return READ_ONCE(net->ipv4.fib_seq) + fib4_rules_seq_read(net);
+}
+
+static int fib4_dump(struct net *net, struct notifier_block *nb,
+		     struct netlink_ext_ack *extack)
+{
+	int err;
+
+	err = fib4_rules_dump(net, nb, extack);
+	if (err)
+		return err;
+
+	return fib_notify(net, nb, extack);
+}
+
+static const struct fib_notifier_ops fib4_notifier_ops_template = {
+	.family		= AF_INET,
+	.fib_seq_read	= fib4_seq_read,
+	.fib_dump	= fib4_dump,
+	.owner		= THIS_MODULE,
+};
+
+int __net_init fib4_notifier_init(struct net *net)
+{
+	struct fib_notifier_ops *ops;
+
+	net->ipv4.fib_seq = 0;
+
+	ops = fib_notifier_ops_register(&fib4_notifier_ops_template, net);
+	if (IS_ERR(ops))
+		return PTR_ERR(ops);
+	net->ipv4.notifier_ops = ops;
+
+	return 0;
+}
+
+void __net_exit fib4_notifier_exit(struct net *net)
+{
+	fib_notifier_ops_unregister(net->ipv4.notifier_ops);
+}