linux-kernel - [PATCH v4] sched/fair: do not scan twice in detach

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <20250721023939.19703-1-shijie@os.amperecomputing.com>
Date: Mon, 21 Jul 2025 10:39:39 +0800
From: Huang Shijie <shijie@...amperecomputing.com>
To: mingo@...hat.com,
	peterz@...radead.org,
	juri.lelli@...hat.com,
	vincent.guittot@...aro.org,
	vschneid@...hat.com
Cc: patches@...erecomputing.com,
	cl@...ux.com,
	Shubhang@...amperecomputing.com,
	dietmar.eggemann@....com,
	rostedt@...dmis.org,
	bsegall@...gle.com,
	mgorman@...e.de,
	linux-kernel@...r.kernel.org,
	Huang Shijie <shijie@...amperecomputing.com>
Subject: [PATCH v4] sched/fair: do not scan twice in detach_tasks()

detach_tasks() uses struct lb_env.loop_max as an env.src_rq->cfs_tasks
iteration count limit. It is however set without the source RQ lock held,
and besides detach_tasks() can be re-invoked after releasing and
re-acquiring the RQ lock per LBF_NEED_BREAK.

This means that env.loop_max and the actual length of env.src_rq->cfs_tasks
as observed within detach_tasks() can differ. This can cause some tasks to
be unnecessarily iterated over more than once, for instance:

  env.loop_max := 4
  detach_tasks()
    // Here env.src->cfs_tasks only contains two tasks which can't be
    // migrated anywhere, so they're put back in the list each time.
    env.src->cfs_tasks := [p1, p0]
    // The iteration goes:
    p0; cfs_tasks := [p0, p1]
    p1; cfs_tasks := [p1, p0]
    p0; cfs_tasks := [p0, p1]
    p1; cfs_tasks := [p1, p0]

    // IOW we iterate over each task twice

In the Specjbb test, the similar issues can be caught many times.
(Over 330,000 times in a 30-minites Specjbb test)

The patch introduces "first_back" to record the first task which
is put back to the task list. If we get a task which is equal to
first_back, we break the loop, and avoid to scan twice for it.

Reviewed-by: Valentin Schneider <vschneid@...hat.com>
Signed-off-by: Huang Shijie <shijie@...amperecomputing.com>
---
v3 --> v4:
    Changed the commit message suggested by Valentin Schneider.
    v3: https://lore.kernel.org/all/20250718063523.9232-1-shijie@os.amperecomputing.com/

v2 --> v3:
    Fix a typo in the commit message.
    v2: https://lore.kernel.org/all/20250718054709.8781-1-shijie@os.amperecomputing.com/

v1 --> v2:
    Add more comment from Valentin Schneider
    v1: https://lore.kernel.org/all/20250707083636.38380-1-shijie@os.amperecomputing.com/
---
 kernel/sched/fair.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 7e2963efe800..7cc9d50e3e11 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -9443,6 +9443,7 @@ static int detach_tasks(struct lb_env *env)
 {
 	struct list_head *tasks = &env->src_rq->cfs_tasks;
 	unsigned long util, load;
+	struct task_struct *first_back = NULL;
 	struct task_struct *p;
 	int detached = 0;
 
@@ -9481,6 +9482,12 @@ static int detach_tasks(struct lb_env *env)
 		}
 
 		p = list_last_entry(tasks, struct task_struct, se.group_node);
+		/*
+		 * We're back to an already visited task that couldn't be
+		 * detached, we've seen all there is to see.
+		 */
+		if (p == first_back)
+			break;
 
 		if (!can_migrate_task(p, env))
 			goto next;
@@ -9562,6 +9569,8 @@ static int detach_tasks(struct lb_env *env)
 			schedstat_inc(p->stats.nr_failed_migrations_hot);
 
 		list_move(&p->se.group_node, tasks);
+		if (!first_back)
+			first_back = p;
 	}
 
 	/*
-- 
2.40.1