From 0d6526f82c3cdefcca47f73f5fc08dc6f335eac6 Mon Sep 17 00:00:00 2001 From: Tim Chen Date: Mon, 21 Sep 2026 17:37:22 -0700 Subject: sched/cache: Keep nr_pref_llc_running in the runnable domain, to fix LLC mis-scheduling bug 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 Suggested-by: Chen Yu Signed-off-by: Tim Chen Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Reviewed-by: Kayra Cizmeci Cc: # v7.2.x Link: https://patch.msgid.link/06af61afedac32e6477f57feb4d658f6c411c3af.1790035273.git.tim.c.chen@linux.intel.com --- tools/bootconfig/scripts/xbc.sh | 56 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 56 insertions(+) create mode 100644 tools/bootconfig/scripts/xbc.sh (limited to 'tools/bootconfig/scripts/xbc.sh') diff --git a/tools/bootconfig/scripts/xbc.sh b/tools/bootconfig/scripts/xbc.sh new file mode 100644 index 000000000..1f0ebf50d --- /dev/null +++ b/tools/bootconfig/scripts/xbc.sh @@ -0,0 +1,56 @@ +#!/bin/sh +# SPDX-License-Identifier: GPL-2.0-only + +# bootconfig utility functions + +XBC_TMPFILE= +XBC_BASEDIR=`dirname $0` +BOOTCONFIG=${BOOTCONFIG:=$XBC_BASEDIR/../bootconfig} +if [ ! -x "$BOOTCONFIG" ]; then + BOOTCONFIG=`which bootconfig` + if [ -z "$BOOTCONFIG" ]; then + echo "Erorr: bootconfig command is not found" 1>&2 + exit 1 + fi +fi + +xbc_cleanup() { + if [ "$XBC_TMPFILE" ]; then + rm -f "$XBC_TMPFILE" + fi +} + +xbc_init() { # bootconfig-file + xbc_cleanup + XBC_TMPFILE=`mktemp bconf-XXXX` + trap xbc_cleanup EXIT TERM + + $BOOTCONFIG -l $1 > $XBC_TMPFILE || exit 1 +} + +nr_args() { # args + echo $# +} + +xbc_get_val() { # key [maxnum] + if [ "$2" ]; then + MAXOPT="-L $2" + fi + grep "^$1 =" $XBC_TMPFILE | cut -d= -f2- | \ + sed -e 's/", /" /g' -e "s/',/' /g" | \ + xargs $MAXOPT -n 1 echo +} + +xbc_has_key() { # key + grep -q "^$1 =" $XBC_TMPFILE +} + +xbc_has_branch() { # prefix-key + grep -q "^$1" $XBC_TMPFILE +} + +xbc_subkeys() { # prefix-key depth [subkey-pattern] + __keys=`echo $1 | sed "s/\./ /g"` + __s=`nr_args $__keys` + grep "^$1$3" $XBC_TMPFILE | cut -d= -f1| cut -d. -f$((__s + 1))-$((__s + $2)) | uniq +} -- cgit v1.3.1