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, 60 insertions, 0 deletions

diff --git a/tools/docs/parse-headers.py b/tools/docs/parse-headers.py
new file mode 100755
index 000000000..436acea4c
--- /dev/null
+++ b/tools/docs/parse-headers.py
@@ -0,0 +1,60 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2016, 2025 by Mauro Carvalho Chehab <mchehab@kernel.org>.
+# pylint: disable=C0103
+
+"""
+Convert a C header or source file ``FILE_IN``, into a ReStructured Text
+included via ..parsed-literal block with cross-references for the
+documentation files that describe the API. It accepts an optional
+``FILE_RULES`` file to describes what elements will be either ignored or
+be pointed to a non-default reference type/name.
+
+The output is written at ``FILE_OUT``.
+
+It is capable of identifying defines, functions, structs, typedefs,
+enums and enum symbols and create cross-references for all of them.
+It is also capable of distinguish #define used for specifying a Linux
+ioctl.
+
+The optional ``FILE_RULES`` contains a set of rules like:
+
+    ignore ioctl VIDIOC_ENUM_FMT
+    replace ioctl VIDIOC_DQBUF vidioc_qbuf
+    replace define V4L2_EVENT_MD_FL_HAVE_FRAME_SEQ :c:type:`v4l2_event_motion_det`
+"""
+
+import argparse, sys
+import os.path
+
+src_dir = os.path.dirname(os.path.realpath(__file__))
+sys.path.insert(0, os.path.join(src_dir, '../lib/python'))
+from kdoc.parse_data_structs import ParseDataStructs
+from kdoc.enrich_formatter import EnrichFormatter
+
+def main():
+    """Main function"""
+    parser = argparse.ArgumentParser(description=__doc__,
+                                     formatter_class=EnrichFormatter)
+
+    parser.add_argument("-d", "--debug", action="count", default=0,
+                        help="Increase debug level. Can be used multiple times")
+    parser.add_argument("-t", "--toc", action="store_true",
+                        help="instead of a literal block, outputs a TOC table at the RST file")
+
+    parser.add_argument("file_in", help="Input C file")
+    parser.add_argument("file_out", help="Output RST file")
+    parser.add_argument("file_rules", nargs="?",
+                        help="Exceptions file (optional)")
+
+    args = parser.parse_args()
+
+    parser = ParseDataStructs(debug=args.debug)
+    parser.parse_file(args.file_in, args.file_rules)
+
+    parser.debug_print()
+    parser.write_output(args.file_in, args.file_out, args.toc)
+
+
+if __name__ == "__main__":
+    main()