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, 79 insertions, 0 deletions

diff --git a/scripts/livepatch/fix-patch-lines b/scripts/livepatch/fix-patch-lines
new file mode 100755
index 000000000..fa7d4f659
--- /dev/null
+++ b/scripts/livepatch/fix-patch-lines
@@ -0,0 +1,79 @@
+#!/usr/bin/awk -f
+# SPDX-License-Identifier: GPL-2.0
+#
+# Use #line directives to preserve original __LINE__ numbers across patches to
+# avoid unwanted compilation changes.
+
+BEGIN {
+	in_hunk = 0
+	skip    = 0
+}
+
+/^--- / {
+	skip = $2 !~ /\.(c|h)$/
+	print
+	next
+}
+
+/^@@/ {
+	if (skip) {
+		print
+		next
+	}
+
+	in_hunk = 1
+
+	# @@ -1,3 +1,4 @@:
+	#   1: line number in old file
+	#   3: how many lines the hunk covers in old file
+	#   1: line number in new file
+	#   4: how many lines the hunk covers in new file
+
+	match($0, /^@@ -([0-9]+)(,([0-9]+))? \+([0-9]+)(,([0-9]+))? @@/, m)
+
+	# Set 'cur' to the old file's line number at the start of the hunk.  It
+	# gets incremented for every context line and every line removal, so
+	# that it always represents the old file's current line number.
+	cur = m[1]
+
+	# last = last line number of current hunk
+	last = cur + (m[3] ? m[3] : 1) - 1
+
+	need_line_directive = 0
+
+	print
+	next
+}
+
+{
+	if (skip || !in_hunk || $0 ~ /^\\ No newline at end of file/) {
+		print
+		next
+	}
+
+	# change line
+	if ($0 ~ /^[+-]/) {
+		# inject #line after this group of changes
+		need_line_directive = 1
+
+		if ($0 ~ /^-/)
+			cur++
+
+		print
+		next
+	}
+
+	# If this is the first context line after a group of changes, inject
+	# the #line directive to force the compiler to correct the line
+	# numbering to match the original file.
+	if (need_line_directive) {
+		print "+#line " cur
+		need_line_directive = 0
+	}
+
+	if (cur == last)
+		in_hunk = 0
+
+	cur++
+	print
+}