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, 77 insertions, 0 deletions

diff --git a/drivers/scsi/sr.h b/drivers/scsi/sr.h
new file mode 100644
index 000000000..2d92f9cb6
--- /dev/null
+++ b/drivers/scsi/sr.h
@@ -0,0 +1,77 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ *      sr.h by David Giller
+ *      CD-ROM disk driver header file
+ *      
+ *      adapted from:
+ *      sd.h Copyright (C) 1992 Drew Eckhardt 
+ *      SCSI disk driver header file by
+ *              Drew Eckhardt 
+ *
+ *      <drew@colorado.edu>
+ *
+ *       Modified by Eric Youngdale eric@andante.org to
+ *       add scatter-gather, multiple outstanding request, and other
+ *       enhancements.
+ */
+
+#ifndef _SR_H
+#define _SR_H
+
+#include <linux/mutex.h>
+
+#define MAX_RETRIES	3
+#define SR_TIMEOUT	(30 * HZ)
+
+struct scsi_device;
+
+/* The CDROM is fairly slow, so we need a little extra time */
+/* In fact, it is very slow if it has to spin up first */
+#define IOCTL_TIMEOUT 30*HZ
+
+
+typedef struct scsi_cd {
+	unsigned capacity;	/* size in blocks                       */
+	struct scsi_device *device;
+	unsigned int vendor;	/* vendor code, see sr_vendor.c         */
+	unsigned long ms_offset;	/* for reading multisession-CD's        */
+	unsigned use:1;		/* is this device still supportable     */
+	unsigned xa_flag:1;	/* CD has XA sectors ? */
+	unsigned readcd_known:1;	/* drive supports READ_CD (0xbe) */
+	unsigned readcd_cdda:1;	/* reading audio data using READ_CD */
+	unsigned media_present:1;	/* media is present */
+
+	/* GET_EVENT spurious event handling, blk layer guarantees exclusion */
+	int tur_mismatch;		/* nr of get_event TUR mismatches */
+	bool tur_changed:1;		/* changed according to TUR */
+	bool get_event_changed:1;	/* changed according to GET_EVENT */
+	bool ignore_get_event:1;	/* GET_EVENT is unreliable, use TUR */
+
+	struct cdrom_device_info cdi;
+	struct mutex lock;
+	struct gendisk *disk;
+} Scsi_CD;
+
+#define sr_printk(prefix, cd, fmt, a...) \
+	sdev_prefix_printk(prefix, (cd)->device, (cd)->cdi.name, fmt, ##a)
+
+int sr_do_ioctl(Scsi_CD *, struct packet_command *);
+
+int sr_lock_door(struct cdrom_device_info *, int);
+int sr_tray_move(struct cdrom_device_info *, int);
+int sr_drive_status(struct cdrom_device_info *, int);
+int sr_disk_status(struct cdrom_device_info *);
+int sr_get_last_session(struct cdrom_device_info *, struct cdrom_multisession *);
+int sr_get_mcn(struct cdrom_device_info *, struct cdrom_mcn *);
+int sr_reset(struct cdrom_device_info *);
+int sr_select_speed(struct cdrom_device_info *cdi, unsigned long speed);
+int sr_audio_ioctl(struct cdrom_device_info *, unsigned int, void *);
+
+int sr_is_xa(Scsi_CD *);
+
+/* sr_vendor.c */
+void sr_vendor_init(Scsi_CD *);
+int sr_cd_check(struct cdrom_device_info *);
+int sr_set_blocklength(Scsi_CD *, int blocklength);
+
+#endif