Re: Regression in next with ext4 oops

From: Tony Lindgren <tony@atomide.com>
To: Kalle Valo <kvalo@codeaurora.org>
Cc: Al Viro <viro@ZenIV.linux.org.uk>,
	"Theodore Ts'o" <tytso@mit.edu>, Jan Kara <jack@suse.cz>,
	Eric Biggers <ebiggers@google.com>,
	Andreas Dilger <adilger@dilger.ca>,
	Jaegeuk Kim <jaegeuk@kernel.org>,
	linux-ext4@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-omap@vger.kernel.org
Subject: Re: Regression in next with ext4 oops
Date: Tue, 4 Oct 2016 13:34:04 -0700	[thread overview]
Message-ID: <20161004203404.cxx767awxo7ti4v7@atomide.com> (raw)
In-Reply-To: <87fuobq4v1.fsf@kamboji.qca.qualcomm.com>

* Kalle Valo <kvalo@codeaurora.org> [161004 12:42]:
> Tony Lindgren <tony@atomide.com> writes:
> 
> > * Tony Lindgren <tony@atomide.com> [161004 12:17]:
> >> Hi,
> >> 
> >> * Al Viro <viro@ZenIV.linux.org.uk> [161004 08:00]:
> >> > On Tue, Oct 04, 2016 at 10:02:31AM -0400, Theodore Ts'o wrote:
> >> > > On Tue, Oct 04, 2016 at 11:00:41AM +0200, Jan Kara wrote:
> >> > > > Never seen this but I suspect it is a fallout from Al's directory locking
> >> > > > changes. In particular ext4_htree_fill_tree() builds rb-tree of found
> >> > > > directory entries in file->private_data (and generally modifies the
> >> > > > structure stored there) but after Al's changes we don't have exclusive
> >> > > > access to struct file if I'm right so if two processes end up calling
> >> > > > getdents() for the same 'struct file' we are doomed.
> >> > > 
> >> > > I haven't seen it either, and I've been doing a lot of testing on the
> >> > > ext4 test branch.  So I'm guessing Tony has the only reliable repro
> >> > > for the problem at the moment.  That being said, it shouldn't be that
> >> > > hard to create a test case for this and add it to xfstests.
> >> > > 
> >> > > I'm pretty sure Jan is right about this, though, but it would be great
> >> > > to a get a quick confirmation from Tony if at all possible.
> >> > 
> >> > Jan is wrong - we do have per-struct-file serialization for getdents()
> >> > et.al.  It might be a race between getdents() on *different* struct
> >> > file for the same directory, but ->private_data is not a problem.
> >> 
> >> OK found the guilty person after git bisect and that's me.
> >> 
> >> Git bisect points to commit d776fc86b82f ("wlcore: sdio: Populate config
> >> firmware data"), so adding Kalle to Cc.
> >> 
> >> Looks like update-initramfs does rmmod of wlcore_sdio and that triggers
> >> some issue with the wlcore driver or with SDIO/MMC. Or maybe it's a memory
> >> corruption issue. I don't know yet exactly what's going on here yet but
> >> I plan to find out after some lunch.
> >
> > And the patch below seems to fix the issue as the driver is now
> > using devm_kzalloc. Will do some more testing and then will post
> > a proper patch. The same issue might be there for SPI glue also.
> 
> This was already posted and you even acked it :)
> 
> wlcore: sdio: drop kfree for memory allocated with devm_kzalloc
> 
> https://patchwork.kernel.org/patch/9353985/

Heh well now we know :) Can you please apply it as it fixes a memory
corruption issue?

The SPI glue does not have this same issue.

Regards,

Tony