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, 165 insertions, 0 deletions

diff --git a/include/trace/misc/fs.h b/include/trace/misc/fs.h
new file mode 100644
index 000000000..7ead1c61f
--- /dev/null
+++ b/include/trace/misc/fs.h
@@ -0,0 +1,165 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Display helpers for generic filesystem items
+ *
+ * Author: Chuck Lever <chuck.lever@oracle.com>
+ *
+ * Copyright (c) 2020, Oracle and/or its affiliates.
+ */
+
+#include <linux/fs.h>
+
+#define show_fs_dirent_type(x) \
+	__print_symbolic(x, \
+		{ DT_UNKNOWN,		"UNKNOWN" }, \
+		{ DT_FIFO,		"FIFO" }, \
+		{ DT_CHR,		"CHR" }, \
+		{ DT_DIR,		"DIR" }, \
+		{ DT_BLK,		"BLK" }, \
+		{ DT_REG,		"REG" }, \
+		{ DT_LNK,		"LNK" }, \
+		{ DT_SOCK,		"SOCK" }, \
+		{ DT_WHT,		"WHT" })
+
+#define show_fs_fcntl_open_flags(x) \
+	__print_flags(x, "|", \
+		{ O_WRONLY,		"O_WRONLY" }, \
+		{ O_RDWR,		"O_RDWR" }, \
+		{ O_CREAT,		"O_CREAT" }, \
+		{ O_EXCL,		"O_EXCL" }, \
+		{ O_NOCTTY,		"O_NOCTTY" }, \
+		{ O_TRUNC,		"O_TRUNC" }, \
+		{ O_APPEND,		"O_APPEND" }, \
+		{ O_NONBLOCK,		"O_NONBLOCK" }, \
+		{ O_DSYNC,		"O_DSYNC" }, \
+		{ O_DIRECT,		"O_DIRECT" }, \
+		{ O_LARGEFILE,		"O_LARGEFILE" }, \
+		{ O_DIRECTORY,		"O_DIRECTORY" }, \
+		{ O_NOFOLLOW,		"O_NOFOLLOW" }, \
+		{ O_NOATIME,		"O_NOATIME" }, \
+		{ O_CLOEXEC,		"O_CLOEXEC" })
+
+#define __fmode_flag(x)	{ (__force unsigned long)FMODE_##x, #x }
+#define show_fs_fmode_flags(x) \
+	__print_flags(x, "|", \
+		__fmode_flag(READ), \
+		__fmode_flag(WRITE), \
+		__fmode_flag(EXEC))
+
+#ifdef CONFIG_64BIT
+#define show_fs_fcntl_cmd(x) \
+	__print_symbolic(x, \
+		{ F_DUPFD,		"DUPFD" }, \
+		{ F_GETFD,		"GETFD" }, \
+		{ F_SETFD,		"SETFD" }, \
+		{ F_GETFL,		"GETFL" }, \
+		{ F_SETFL,		"SETFL" }, \
+		{ F_GETLK,		"GETLK" }, \
+		{ F_SETLK,		"SETLK" }, \
+		{ F_SETLKW,		"SETLKW" }, \
+		{ F_SETOWN,		"SETOWN" }, \
+		{ F_GETOWN,		"GETOWN" }, \
+		{ F_SETSIG,		"SETSIG" }, \
+		{ F_GETSIG,		"GETSIG" }, \
+		{ F_SETOWN_EX,		"SETOWN_EX" }, \
+		{ F_GETOWN_EX,		"GETOWN_EX" }, \
+		{ F_GETOWNER_UIDS,	"GETOWNER_UIDS" }, \
+		{ F_OFD_GETLK,		"OFD_GETLK" }, \
+		{ F_OFD_SETLK,		"OFD_SETLK" }, \
+		{ F_OFD_SETLKW,		"OFD_SETLKW" })
+#else /* CONFIG_64BIT */
+#define show_fs_fcntl_cmd(x) \
+	__print_symbolic(x, \
+		{ F_DUPFD,		"DUPFD" }, \
+		{ F_GETFD,		"GETFD" }, \
+		{ F_SETFD,		"SETFD" }, \
+		{ F_GETFL,		"GETFL" }, \
+		{ F_SETFL,		"SETFL" }, \
+		{ F_GETLK,		"GETLK" }, \
+		{ F_SETLK,		"SETLK" }, \
+		{ F_SETLKW,		"SETLKW" }, \
+		{ F_SETOWN,		"SETOWN" }, \
+		{ F_GETOWN,		"GETOWN" }, \
+		{ F_SETSIG,		"SETSIG" }, \
+		{ F_GETSIG,		"GETSIG" }, \
+		{ F_GETLK64,		"GETLK64" }, \
+		{ F_SETLK64,		"SETLK64" }, \
+		{ F_SETLKW64,		"SETLKW64" }, \
+		{ F_SETOWN_EX,		"SETOWN_EX" }, \
+		{ F_GETOWN_EX,		"GETOWN_EX" }, \
+		{ F_GETOWNER_UIDS,	"GETOWNER_UIDS" }, \
+		{ F_OFD_GETLK,		"OFD_GETLK" }, \
+		{ F_OFD_SETLK,		"OFD_SETLK" }, \
+		{ F_OFD_SETLKW,		"OFD_SETLKW" })
+#endif /* CONFIG_64BIT */
+
+#define show_fs_fcntl_lock_type(x) \
+	__print_symbolic(x, \
+		{ F_RDLCK,		"RDLCK" }, \
+		{ F_WRLCK,		"WRLCK" }, \
+		{ F_UNLCK,		"UNLCK" })
+
+#define show_fs_lookup_flags(flags) \
+	__print_flags(flags, "|", \
+		{ LOOKUP_FOLLOW,	"FOLLOW" }, \
+		{ LOOKUP_DIRECTORY,	"DIRECTORY" }, \
+		{ LOOKUP_AUTOMOUNT,	"AUTOMOUNT" }, \
+		{ LOOKUP_EMPTY,		"EMPTY" }, \
+		{ LOOKUP_DOWN,		"DOWN" }, \
+		{ LOOKUP_MOUNTPOINT,	"MOUNTPOINT" }, \
+		{ LOOKUP_REVAL,		"REVAL" }, \
+		{ LOOKUP_RCU,		"RCU" }, \
+		{ LOOKUP_OPEN,		"OPEN" }, \
+		{ LOOKUP_CREATE,	"CREATE" }, \
+		{ LOOKUP_EXCL,		"EXCL" }, \
+		{ LOOKUP_RENAME_TARGET,	"RENAME_TARGET" }, \
+		{ LOOKUP_PARENT,	"PARENT" }, \
+		{ LOOKUP_NO_SYMLINKS,	"NO_SYMLINKS" }, \
+		{ LOOKUP_NO_MAGICLINKS,	"NO_MAGICLINKS" }, \
+		{ LOOKUP_NO_XDEV,	"NO_XDEV" }, \
+		{ LOOKUP_BENEATH,	"BENEATH" }, \
+		{ LOOKUP_IN_ROOT,	"IN_ROOT" }, \
+		{ LOOKUP_CACHED,	"CACHED" })
+
+#define show_ia_valid_flags(flags)			\
+	__print_flags(flags, "|",			\
+		{ ATTR_MODE,		"MODE" },	\
+		{ ATTR_UID,		"UID" },	\
+		{ ATTR_GID,		"GID" },	\
+		{ ATTR_SIZE,		"SIZE" },	\
+		{ ATTR_ATIME,		"ATIME" },	\
+		{ ATTR_MTIME,		"MTIME" },	\
+		{ ATTR_CTIME,		"CTIME" },	\
+		{ ATTR_ATIME_SET,	"ATIME_SET" },	\
+		{ ATTR_MTIME_SET,	"MTIME_SET" },	\
+		{ ATTR_FORCE,		"FORCE" },	\
+		{ ATTR_KILL_SUID,	"KILL_SUID" },	\
+		{ ATTR_KILL_SGID,	"KILL_SGID" },	\
+		{ ATTR_FILE,		"FILE" },	\
+		{ ATTR_KILL_PRIV,	"KILL_PRIV" },	\
+		{ ATTR_OPEN,		"OPEN" },	\
+		{ ATTR_TIMES_SET,	"TIMES_SET" },	\
+		{ ATTR_TOUCH,		"TOUCH"},	\
+		{ ATTR_DELEG,		"DELEG"})
+
+#define show_statx_mask(flags)					\
+	__print_flags(flags, "|",				\
+		{ STATX_TYPE,		"TYPE" },		\
+		{ STATX_MODE,		"MODE" },		\
+		{ STATX_NLINK,		"NLINK" },		\
+		{ STATX_UID,		"UID" },		\
+		{ STATX_GID,		"GID" },		\
+		{ STATX_ATIME,		"ATIME" },		\
+		{ STATX_MTIME,		"MTIME" },		\
+		{ STATX_CTIME,		"CTIME" },		\
+		{ STATX_INO,		"INO" },		\
+		{ STATX_SIZE,		"SIZE" },		\
+		{ STATX_BLOCKS,		"BLOCKS" },		\
+		{ STATX_BASIC_STATS,	"BASIC_STATS" },	\
+		{ STATX_BTIME,		"BTIME" },		\
+		{ STATX_MNT_ID,		"MNT_ID" },		\
+		{ STATX_DIOALIGN,	"DIOALIGN" },		\
+		{ STATX_MNT_ID_UNIQUE,	"MNT_ID_UNIQUE" },	\
+		{ STATX_SUBVOL,		"SUBVOL" },		\
+		{ STATX_WRITE_ATOMIC,	"WRITE_ATOMIC" },	\
+		{ STATX_DIO_READ_ALIGN,	"DIO_READ_ALIGN" })