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, 62 insertions, 0 deletions

diff --git a/net/openvswitch/meter.h b/net/openvswitch/meter.h
new file mode 100644
index 000000000..8bbf983cd
--- /dev/null
+++ b/net/openvswitch/meter.h
@@ -0,0 +1,62 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2017 Nicira, Inc.
+ */
+
+#ifndef METER_H
+#define METER_H 1
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/netlink.h>
+#include <linux/openvswitch.h>
+#include <linux/skbuff.h>
+#include <linux/bits.h>
+
+#include "flow.h"
+struct datapath;
+
+#define DP_MAX_BANDS		1
+#define DP_METER_ARRAY_SIZE_MIN	BIT_ULL(10)
+#define DP_METER_NUM_MAX	(200000UL)
+
+struct dp_meter_band {
+	u32 type;
+	u32 rate;
+	u32 burst_size;
+	u64 bucket; /* 1/1000 packets, or in bits */
+	struct ovs_flow_stats stats;
+};
+
+struct dp_meter {
+	spinlock_t lock;    /* Per meter lock */
+	struct rcu_head rcu;
+	u32 id;
+	u16 kbps:1, keep_stats:1;
+	u16 n_bands;
+	u32 max_delta_t;
+	u64 used;
+	struct ovs_flow_stats stats;
+	struct dp_meter_band bands[] __counted_by(n_bands);
+};
+
+struct dp_meter_instance {
+	struct rcu_head rcu;
+	u32 n_meters;
+	struct dp_meter __rcu *dp_meters[] __counted_by(n_meters);
+};
+
+struct dp_meter_table {
+	struct dp_meter_instance __rcu *ti;
+	u32 count;
+	u32 max_meters_allowed;
+};
+
+extern struct genl_family dp_meter_genl_family;
+int ovs_meters_init(struct datapath *dp);
+void ovs_meters_exit(struct datapath *dp);
+bool ovs_meter_execute(struct datapath *dp, struct sk_buff *skb,
+		       struct sw_flow_key *key, u32 meter_id);
+
+#endif /* meter.h */