Re: [PATCH 2.6.19 5/5] fs: freeze_bdev with semaphore not mutex

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Wednesday, 8 November 2006 00:18, Eric Sandeen wrote:
> Rafael J. Wysocki wrote:
> > On Tuesday, 7 November 2006 23:45, Eric Sandeen wrote:
> >> Andrew Morton wrote:
> >>
> >>>> --- linux-2.6.19-rc4.orig/fs/buffer.c	2006-11-07 17:06:20.000000000 +0000
> >>>> +++ linux-2.6.19-rc4/fs/buffer.c	2006-11-07 17:26:04.000000000 +0000
> >>>> @@ -188,7 +188,9 @@ struct super_block *freeze_bdev(struct b
> >>>>  {
> >>>>  	struct super_block *sb;
> >>>>  
> >>>> -	mutex_lock(&bdev->bd_mount_mutex);
> >>>> +	if (down_trylock(&bdev->bd_mount_sem))
> >>>> +		return -EBUSY;
> >>>> +
> >>> This is a functional change which isn't described in the changelog.  What's
> >>> happening here?
> >> Only allow one bdev-freezer in at a time, rather than queueing them up?
> > 
> > But freeze_bdev() is supposed to return the result of get_super(bdev)
> > _unconditionally_.  Moreover, in its current form freeze_bdev() _cannot_
> > _fail_, so I don't see how this change doesn't break any existing code.
> 
> Well, it could return NULL.  Is that a failure?

It will only fail if get_super(bdev) returns NULL, but if you call
freeze_bdev(sb->s_bdev), then it can't do that.

> But, nobody is checking for an outright error, certainly.  Especially
> when the error hasn't been ERR_PTR'd.  :)  So I agree, that's not so good.
> 
> But, how is a stampede of fs-freezers -supposed- to work?  I could
> imagine something like a freezer count, and the filesystem is only
> unfrozen after everyone has thawed?  Or should only one freezer be
> active at a time... which is what we have now I guess.

I think it shouldn't be possible to freeze an fs more than once.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/