Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Mon, 6 Apr 2020 15:19:07 +0000
From:   Luis Chamberlain <mcgrof@kernel.org>
To:     Nicolai Stange <nstange@suse.de>
Cc:     Bart Van Assche <bvanassche@acm.org>,
        Eric Sandeen <sandeen@sandeen.net>, axboe@kernel.dk,
        viro@zeniv.linux.org.uk, gregkh@linuxfoundation.org,
        rostedt@goodmis.org, mingo@redhat.com, jack@suse.cz,
        ming.lei@redhat.com, mhocko@suse.com, linux-block@vger.kernel.org,
        linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
        Omar Sandoval <osandov@fb.com>,
        Hannes Reinecke <hare@suse.com>,
        Michal Hocko <mhocko@kernel.org>,
        syzbot+603294af2d01acfdd6da@syzkaller.appspotmail.com
Subject: Re: [RFC 2/3] blktrace: fix debugfs use after free
Message-ID: <20200406151907.GD11244@42.do-not-panic.com>
References: <20200402000002.7442-1-mcgrof@kernel.org>
 <20200402000002.7442-3-mcgrof@kernel.org>
 <3640b16b-abda-5160-301a-6a0ee67365b4@acm.org>
 <b827d03c-e097-06c3-02ab-00df42b5fc0e@sandeen.net>
 <75aa4cff-1b90-ebd4-17a4-c1cb6d390b30@acm.org>
 <87d08lj7l6.fsf@suse.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <87d08lj7l6.fsf@suse.de>
User-Agent: Mutt/1.10.1 (2018-07-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Apr 06, 2020 at 11:18:13AM +0200, Nicolai Stange wrote:
> Bart Van Assche <bvanassche@acm.org> writes:

> So I'd suggest to drop patch [3/3] from this series and modify this
> patch [2/3] here to move the blk_q_debugfs_unregister(q) invocation from
> __blk_release_queue() to blk_unregister_queue() instead.

I'll take a stab.

> > Additionally, I think the following changes fix that problem by using
> > q->debugfs_dir in the blktrace code instead of debugfs_lookup():
> 
> That would fix the UAF, but !queue_is_mq() queues wouldn't get a debugfs
> directory created for them by blktrace anymore?

It would, it would just be done early on init as well, and it would now be
shared with the queue_is_mq() case.

  Luis