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, 93 insertions, 0 deletions

diff --git a/drivers/scsi/fcoe/fcoe.h b/drivers/scsi/fcoe/fcoe.h
new file mode 100644
index 000000000..520c53512
--- /dev/null
+++ b/drivers/scsi/fcoe/fcoe.h
@@ -0,0 +1,93 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright(c) 2009 Intel Corporation. All rights reserved.
+ *
+ * Maintained at www.Open-FCoE.org
+ */
+
+#ifndef _FCOE_H_
+#define _FCOE_H_
+
+#include <linux/skbuff.h>
+#include <linux/kthread.h>
+
+#define FCOE_MAX_QUEUE_DEPTH	256
+#define FCOE_MIN_QUEUE_DEPTH	32
+
+#define FCOE_WORD_TO_BYTE	4
+
+#define FCOE_VERSION	"0.1"
+#define FCOE_NAME	"fcoe"
+#define FCOE_VENDOR	"Open-FCoE.org"
+
+#define FCOE_MAX_LUN		0xFFFF
+#define FCOE_MAX_FCP_TARGET	256
+
+#define FCOE_MAX_OUTSTANDING_COMMANDS	1024
+
+#define FCOE_MIN_XID		0x0000	/* the min xid supported by fcoe_sw */
+#define FCOE_MAX_XID		0x0FFF	/* the max xid supported by fcoe_sw */
+
+extern unsigned int fcoe_debug_logging;
+
+#define FCOE_LOGGING	    0x01 /* General logging, not categorized */
+#define FCOE_NETDEV_LOGGING 0x02 /* Netdevice logging */
+
+#define FCOE_CHECK_LOGGING(LEVEL, CMD)					\
+do {                                                            	\
+	if (unlikely(fcoe_debug_logging & LEVEL))			\
+		do {							\
+			CMD;						\
+		} while (0);						\
+} while (0)
+
+#define FCOE_DBG(fmt, args...)						\
+	FCOE_CHECK_LOGGING(FCOE_LOGGING,				\
+			   pr_info("fcoe: " fmt, ##args);)
+
+#define FCOE_NETDEV_DBG(netdev, fmt, args...)			\
+	FCOE_CHECK_LOGGING(FCOE_NETDEV_LOGGING,			\
+			   pr_info("fcoe: %s: " fmt,		\
+				   netdev->name, ##args);)
+
+/**
+ * struct fcoe_interface - A FCoE interface
+ * @list:	      Handle for a list of FCoE interfaces
+ * @netdev:	      The associated net device
+ * @fcoe_packet_type: FCoE packet type
+ * @fip_packet_type:  FIP packet type
+ * @oem:	      The offload exchange manager for all local port
+ *		      instances associated with this port
+ * @removed:	      Indicates fcoe interface removed from net device
+ * @priority:	      Priority for the FCoE packet (DCB)
+ * This structure is 1:1 with a net device.
+ */
+struct fcoe_interface {
+	struct list_head   list;
+	struct net_device  *netdev;
+	struct net_device  *realdev;
+	struct packet_type fcoe_packet_type;
+	struct packet_type fip_packet_type;
+	struct packet_type fip_vlan_packet_type;
+	struct fc_exch_mgr *oem;
+	u8	removed;
+	u8	priority;
+};
+
+#define fcoe_to_ctlr(x)						\
+	(struct fcoe_ctlr *)(((struct fcoe_ctlr *)(x)) - 1)
+
+#define fcoe_from_ctlr(x)			\
+	((struct fcoe_interface *)((x) + 1))
+
+/**
+ * fcoe_netdev() - Return the net device associated with a local port
+ * @lport: The local port to get the net device from
+ */
+static inline struct net_device *fcoe_netdev(const struct fc_lport *lport)
+{
+	return ((struct fcoe_interface *)
+			((struct fcoe_port *)lport_priv(lport))->priv)->netdev;
+}
+
+#endif /* _FCOE_H_ */