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, 65 insertions, 0 deletions

diff --git a/scripts/extract-vmlinux b/scripts/extract-vmlinux
new file mode 100755
index 000000000..266df9bc7
--- /dev/null
+++ b/scripts/extract-vmlinux
@@ -0,0 +1,65 @@
+#!/bin/sh
+# SPDX-License-Identifier: GPL-2.0-only
+# ----------------------------------------------------------------------
+# extract-vmlinux - Extract uncompressed vmlinux from a kernel image
+#
+# Inspired from extract-ikconfig
+# (c) 2009,2010 Dick Streefland <dick@streefland.net>
+#
+# (c) 2011      Corentin Chary <corentin.chary@gmail.com>
+#
+# ----------------------------------------------------------------------
+
+me=${0##*/}
+
+check_vmlinux()
+{
+	if file "$1" | grep -q 'Linux kernel.*boot executable' ||
+		readelf -h "$1" > /dev/null 2>&1
+	then
+		cat "$1"
+		echo "$me: Extracted vmlinux using '$2' from offset $3" >&2
+		exit 0
+	fi
+}
+
+try_decompress()
+{
+	# The obscure use of the "tr" filter is to work around older versions of
+	# "grep" that report the byte offset of the line instead of the pattern.
+
+	# Try to find the header ($1) and decompress from here
+	for	pos in `tr "$1\n$2" "\n$2=" < "$img" | grep -abo "^$2"`
+	do
+		pos=${pos%%:*}
+		tail -c+$pos "$img" | $3 > $tmp 2> /dev/null
+		check_vmlinux $tmp "$3" $pos
+	done
+}
+
+# Check invocation:
+img=$1
+if	[ $# -ne 1 -o ! -s "$img" ]
+then
+	echo "Usage: $me <kernel-image>" >&2
+	exit 2
+fi
+
+# Prepare temp files:
+tmp=$(mktemp /tmp/vmlinux-XXX)
+trap "rm -f $tmp" 0
+
+# That didn't work, so retry after decompression.
+try_decompress '\037\213\010' xy    gunzip
+try_decompress '\3757zXZ\000' abcde unxz
+try_decompress 'BZh'          xy    bunzip2
+try_decompress '\135\0\0\0'   xxx   unlzma
+try_decompress '\211\114\132' xy    'lzop -d'
+try_decompress '\002!L\030'   xxx   'lz4 -d'
+try_decompress '(\265/\375'   xxx   unzstd
+
+# Finally check for uncompressed images or objects:
+check_vmlinux "$img" cat 0
+
+# Bail out:
+echo "$me: Cannot find vmlinux." >&2