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, 63 insertions, 0 deletions

diff --git a/include/media/media-dev-allocator.h b/include/media/media-dev-allocator.h
new file mode 100644
index 000000000..2ab54d426
--- /dev/null
+++ b/include/media/media-dev-allocator.h
@@ -0,0 +1,63 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/*
+ * media-dev-allocator.h - Media Controller Device Allocator API
+ *
+ * Copyright (c) 2019 Shuah Khan <shuah@kernel.org>
+ *
+ * Credits: Suggested by Laurent Pinchart <laurent.pinchart@ideasonboard.com>
+ */
+
+/*
+ * This file adds a global ref-counted Media Controller Device Instance API.
+ * A system wide global media device list is managed and each media device
+ * includes a kref count. The last put on the media device releases the media
+ * device instance.
+ */
+
+#ifndef _MEDIA_DEV_ALLOCATOR_H
+#define _MEDIA_DEV_ALLOCATOR_H
+
+struct usb_device;
+
+#if defined(CONFIG_MEDIA_CONTROLLER) && IS_ENABLED(CONFIG_USB)
+/**
+ * media_device_usb_allocate() - Allocate and return struct &media device
+ *
+ * @udev:		struct &usb_device pointer
+ * @module_name:	should be filled with %KBUILD_MODNAME
+ * @owner:		struct module pointer %THIS_MODULE for the driver.
+ *			%THIS_MODULE is null for a built-in driver.
+ *			It is safe even when %THIS_MODULE is null.
+ *
+ * This interface should be called to allocate a Media Device when multiple
+ * drivers share usb_device and the media device. This interface allocates
+ * &media_device structure and calls media_device_usb_init() to initialize
+ * it.
+ *
+ */
+struct media_device *media_device_usb_allocate(struct usb_device *udev,
+					       const char *module_name,
+					       struct module *owner);
+/**
+ * media_device_delete() - Release media device. Calls kref_put().
+ *
+ * @mdev:		struct &media_device pointer
+ * @module_name:	should be filled with %KBUILD_MODNAME
+ * @owner:		struct module pointer %THIS_MODULE for the driver.
+ *			%THIS_MODULE is null for a built-in driver.
+ *			It is safe even when %THIS_MODULE is null.
+ *
+ * This interface should be called to put Media Device Instance kref.
+ */
+void media_device_delete(struct media_device *mdev, const char *module_name,
+			 struct module *owner);
+#else
+static inline struct media_device *media_device_usb_allocate(
+			struct usb_device *udev, const char *module_name,
+			struct module *owner)
+			{ return NULL; }
+static inline void media_device_delete(
+			struct media_device *mdev, const char *module_name,
+			struct module *owner) { }
+#endif /* CONFIG_MEDIA_CONTROLLER && CONFIG_USB */
+#endif /* _MEDIA_DEV_ALLOCATOR_H */