On Thu, Apr 22 2010, Jörn Engel wrote:
quoted text
> On Thu, 22 April 2010 12:39:53 +0200, Jens Axboe wrote:
> > 
> > Here's a series for fixing these. At this point they are totally
> > untested except that I did compile them. Note that your analysis
> > appeared correct for all cases but ocfs2, which does use get_sb_bdev()
> > and hence gets ->s_bdi assigned.
> > 
> > You can see them here, I'll post the series soon:
> > 
> > http://git.kernel.dk/?p=linux-2.6-block.git;a=shortlog;h=refs/heads/for-linus
> > 
> > The first patch is a helper addition, the rest are per-fs fixups.
> 
> Looks good at a cursory glance.  What's still missing is some sort of
> assertion.  You are a smart person and missed this problem, twice even.
> Even if you hadn't, a not so smart person can add a new filesystem and
> miss s_bdi, like I did.  We want some automatism to catch this.

I totally agree, we want some way to catch this problem in the future.
Really the check needs to be something ala:

        if (!sb->s_bdi && mnt_has_storage_backing() && rw_mount)
                yell_and_fail;

but I'm not sure how best to accomplish that. We can check for ->s_bdev
and mtd like you did, but that does not catch network file systems for
instance.

-- 
Jens Axboe

--
unsubscribe notice
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/