All of lore.kernel.org
 help / color / mirror / Atom feed
From: Kirill Korotaev <dev@sw.ru>
To: Andrew Morton <akpm@osdl.org>, Con Kolivas <kernel@kolivas.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	devel@openvz.org, Ingo Molnar <mingo@elte.hu>
Subject: [PATCH] sched: CPU hotplug race vs. set_cpus_allowed()
Date: Mon, 26 Jun 2006 11:58:49 +0400	[thread overview]
Message-ID: <449F93B9.9070305@sw.ru> (raw)

[-- Attachment #1: Type: text/plain, Size: 519 bytes --]

There is a race between set_cpus_allowed()
and move_task_off_dead_cpu().
__migrate_task() doesn't report any err code, so
task can be left on its runqueue if its cpus_allowed mask
changed so that dest_cpu is not longer a possible target.
Also, chaning cpus_allowed mask requires rq->lock being held.

Changes from original post according to to Con Colivas notes:
- added comment
- cleanup code a bit

Signed-Off-By: Kirill Korotaev <dev@openvz.org>
Acked-By: Ingo Molnar <mingo@elte.hu>

Kirill
P.S. against 2.6.17-mm1

[-- Attachment #2: diff-migrate-task --]
[-- Type: text/plain, Size: 1989 bytes --]

--- linux-2.6.17-mm1s.orig/kernel/sched.c	2006-06-21 18:53:17.000000000 +0400
+++ linux-2.6.17-mm1.dev/kernel/sched.c	2006-06-26 11:54:19.000000000 +0400
@@ -4857,13 +4857,16 @@ EXPORT_SYMBOL_GPL(set_cpus_allowed);
  *
  * So we race with normal scheduler movements, but that's OK, as long
  * as the task is no longer on this CPU.
+ *
+ * Returns non-zero if task was successfully migrated.
  */
-static void __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu)
+static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu)
 {
 	runqueue_t *rq_dest, *rq_src;
+	int ret = 0;
 
 	if (unlikely(cpu_is_offline(dest_cpu)))
-		return;
+		return ret;
 
 	rq_src = cpu_rq(src_cpu);
 	rq_dest = cpu_rq(dest_cpu);
@@ -4891,9 +4894,10 @@ static void __migrate_task(struct task_s
 		if (TASK_PREEMPTS_CURR(p, rq_dest))
 			resched_task(rq_dest->curr);
 	}
-
+	ret = 1;
 out:
 	double_rq_unlock(rq_src, rq_dest);
+	return ret;
 }
 
 /*
@@ -4964,10 +4968,13 @@ int sigstop_on_cpu_lost;
 /* Figure out where task on dead CPU should go, use force if neccessary. */
 static void move_task_off_dead_cpu(int dead_cpu, struct task_struct *tsk)
 {
+	runqueue_t *rq;
+	unsigned long flags;
 	int dest_cpu;
 	cpumask_t mask;
 	int force = 0;
 
+restart:
 	/* On same node? */
 	mask = node_to_cpumask(cpu_to_node(dead_cpu));
 	cpus_and(mask, mask, tsk->cpus_allowed);
@@ -4979,8 +4986,10 @@ static void move_task_off_dead_cpu(int d
 
 	/* No more Mr. Nice Guy. */
 	if (dest_cpu == NR_CPUS) {
+		rq = task_rq_lock(tsk, &flags);
 		cpus_setall(tsk->cpus_allowed);
 		dest_cpu = any_online_cpu(tsk->cpus_allowed);
+		task_rq_unlock(rq, &flags);
 
 		/*
 		 * Don't tell them about moving exiting tasks or
@@ -5000,7 +5009,8 @@ static void move_task_off_dead_cpu(int d
 		if (tsk->mm && sigstop_on_cpu_lost)
 			force = 1;
 	}
-	__migrate_task(tsk, dead_cpu, dest_cpu);
+	if (!__migrate_task(tsk, dead_cpu, dest_cpu))
+		goto restart;
 
 	if (force)
 		force_sig_specific(SIGSTOP, tsk);

             reply	other threads:[~2006-06-26  7:59 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-06-26  7:58 Kirill Korotaev [this message]
  -- strict thread matches above, loose matches on Subject: below --
2006-06-23  8:16 [PATCH] sched: CPU hotplug race vs. set_cpus_allowed() Kirill Korotaev
2006-06-23  8:18 ` Ingo Molnar
2006-06-23 10:53 ` Con Kolivas
2006-06-24  5:25 ` Andrew Morton

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=449F93B9.9070305@sw.ru \
    --to=dev@sw.ru \
    --cc=akpm@osdl.org \
    --cc=devel@openvz.org \
    --cc=kernel@kolivas.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.