linux-rt-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: minyard@acm.org
To: linux-rt-users@vger.kernel.org
Cc: minyard@acm.org, Corey Minyard <cminyard@mvista.com>
Subject: [PATCH] Fix a lockup in wait_for_completion() and friends
Date: Wed,  8 May 2019 15:27:59 -0500	[thread overview]
Message-ID: <20190508202759.13842-1-minyard@acm.org> (raw)

From: Corey Minyard <cminyard@mvista.com>

The function call do_wait_for_common() has a race condition that
can result in lockups waiting for completions.  Adding the thread
to (and removing the thread from) the wait queue for the completion
is done outside the do loop in that function.  However, if the thread
is woken up with swake_up_locked(), that function will delete the
entry from the wait queue.  If that happens and another thread sneaks
in and decrements the done count in the completion to zero, the
loop will go around again, but the thread will no longer be in the
wait queue, so there is no way to wake it up.

Fix it by adding/removing the thread to/from the wait queue inside
the do loop.

Fixes: a04ff6b4ec4ee7e ("completion: Use simple wait queues")
Signed-off-by: Corey Minyard <cminyard@mvista.com>
---
This looks like a fairly serious bug, I guess, but I've never seen a
report on it before.

I found it because I have an out-of-tree feature (hopefully in tree some
day) that takes a core dump of a running process without killing it.  It
makes extensive use of completions, and the test code is fairly brutal.
It didn't lock up on stock 4.19, but failed with the RT patches applied.

The funny thing is, I've never seen this test code fail before on earlier
releases, but it locks up pretty reliably on 4.19-rt.  It looks like this
bug goes back to at least the 4.4-rt kernel.  But we haven't received any
customer reports of failures.

The feature and test are in a public tree if someone wants to try to
reproduce this.  But hopefully this is pretty obvious with the explaination.

Also, you could put the DECLARE_SWAITQUEUE() outside the loop, I think,
but maybe it's cleaner or safer to declare it in the loop?  If someone
cares I can test it that way.

-corey

 kernel/sched/completion.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/kernel/sched/completion.c b/kernel/sched/completion.c
index 755a58084978..4cde33cf8b28 100644
--- a/kernel/sched/completion.c
+++ b/kernel/sched/completion.c
@@ -70,10 +70,10 @@ do_wait_for_common(struct completion *x,
 		   long (*action)(long), long timeout, int state)
 {
 	if (!x->done) {
-		DECLARE_SWAITQUEUE(wait);
-
-		__prepare_to_swait(&x->wait, &wait);
 		do {
+			DECLARE_SWAITQUEUE(wait);
+
+			__prepare_to_swait(&x->wait, &wait);
 			if (signal_pending_state(state, current)) {
 				timeout = -ERESTARTSYS;
 				break;
@@ -82,8 +82,8 @@ do_wait_for_common(struct completion *x,
 			raw_spin_unlock_irq(&x->wait.lock);
 			timeout = action(timeout);
 			raw_spin_lock_irq(&x->wait.lock);
+			__finish_swait(&x->wait, &wait);
 		} while (!x->done && timeout);
-		__finish_swait(&x->wait, &wait);
 		if (!x->done)
 			return timeout;
 	}
-- 
2.17.1


             reply	other threads:[~2019-05-08 20:28 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-05-08 20:27 minyard [this message]
2019-05-13 12:13 ` [PATCH] Fix a lockup in wait_for_completion() and friends Daniel Wagner
2019-05-13 12:19   ` Daniel Wagner

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=20190508202759.13842-1-minyard@acm.org \
    --to=minyard@acm.org \
    --cc=cminyard@mvista.com \
    --cc=linux-rt-users@vger.kernel.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).