[3.5.y.z,extended,stable] Patch "livelock avoidance in sget()" has been added to staging queue

Message ID 1375609769-11479-1-git-send-email-luis.henriques@canonical.com
State New
Headers show

Commit Message

Luis Henriques Aug. 4, 2013, 9:49 a.m.
This is a note to let you know that I have just added a patch titled

    livelock avoidance in sget()

to the linux-3.5.y-queue branch of the 3.5.y.z extended stable tree 
which can be found at:


If you, or anyone else, feels it should not be added to this tree, please 
reply to this email.

For more information about the 3.5.y.z tree, see



From 91de9481108f8a3795d7b760bee5e1f36af5c92b Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 20 Jul 2013 03:13:55 +0400
Subject: [PATCH] livelock avoidance in sget()

commit acfec9a5a892f98461f52ed5770de99a3e571ae2 upstream.

Eric Sandeen has found a nasty livelock in sget() - take a mount(2) about
to fail.  The superblock is on ->fs_supers, ->s_umount is held exclusive,
->s_active is 1.  Along comes two more processes, trying to mount the same
thing; sget() in each is picking that superblock, bumping ->s_count and
trying to grab ->s_umount.  ->s_active is 3 now.  Original mount(2)
finally gets to deactivate_locked_super() on failure; ->s_active is 2,
superblock is still ->fs_supers because shutdown will *not* happen until
->s_active hits 0.  ->s_umount is dropped and now we have two processes
chasing each other:
s_active = 2, A acquired ->s_umount, B blocked
A sees that the damn thing is stillborn, does deactivate_locked_super()
s_active = 1, A drops ->s_umount, B gets it
A restarts the search and finds the same superblock.  And bumps it ->s_active.
s_active = 2, B holds ->s_umount, A blocked on trying to get it
... and we are in the earlier situation with A and B switched places.

The root cause, of course, is that ->s_active should not grow until we'd
got MS_BORN.  Then failing ->mount() will have deactivate_locked_super()
shut the damn thing down.  Fortunately, it's easy to do - the key point
is that grab_super() is called only for superblocks currently on ->fs_supers,
so it can bump ->s_count and grab ->s_umount first, then check MS_BORN and
bump ->s_active; we must never increment ->s_count for superblocks past
->kill_sb(), but grab_super() is never called for those.

The bug is pretty old; we would've caught it by now, if not for accidental
exclusion between sget() for block filesystems; the things like cgroup or
e.g. mtd-based filesystems don't have anything of that sort, so they get
bitten.  The right way to deal with that is obviously to fix sget()...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Luis Henriques <luis.henriques@canonical.com>
 fs/super.c | 25 ++++++++++---------------
 1 file changed, 10 insertions(+), 15 deletions(-)



diff --git a/fs/super.c b/fs/super.c
index cf00177..3c520a5 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -298,19 +298,19 @@  EXPORT_SYMBOL(deactivate_super);
  *	and want to turn it into a full-blown active reference.  grab_super()
  *	is called with sb_lock held and drops it.  Returns 1 in case of
  *	success, 0 if we had failed (superblock contents was already dead or
- *	dying when grab_super() had been called).
+ *	dying when grab_super() had been called).  Note that this is only
+ *	called for superblocks not in rundown mode (== ones still on ->fs_supers
+ *	of their type), so increment of ->s_count is OK here.
 static int grab_super(struct super_block *s) __releases(sb_lock)
-	if (atomic_inc_not_zero(&s->s_active)) {
-		spin_unlock(&sb_lock);
-		return 1;
-	}
-	/* it's going away */
-	/* wait for it to die */
+	if ((s->s_flags & MS_BORN) && atomic_inc_not_zero(&s->s_active)) {
+		put_super(s);
+		return 1;
+	}
 	return 0;
@@ -440,11 +440,6 @@  retry:
 				s = NULL;
-			down_write(&old->s_umount);
-			if (unlikely(!(old->s_flags & MS_BORN))) {
-				deactivate_locked_super(old);
-				goto retry;
-			}
 			return old;
@@ -677,10 +672,10 @@  restart:
 		if (hlist_unhashed(&sb->s_instances))
 		if (sb->s_bdev == bdev) {
-			if (grab_super(sb)) /* drops sb_lock */
-				return sb;
-			else
+			if (!grab_super(sb))
 				goto restart;
+			up_write(&sb->s_umount);
+			return sb;