Re: [PATCH] block: revert pushing the final release of request_queue to a workqueue.

[Ming Lei]

On Mon, Feb 10, 2020 at 10:13:22AM +0800, yukuai (C) wrote:
> On 2020/2/10 9:00, Bart Van Assche wrote:
> > Have you already noticed patch "[PATCH] blktrace: Protect q->blk_trace
> > with RCU"? If not, have you already tried to verify whether that patch
> > fixes the use-after-free detected by syzbot?
> 
> I just tested and confirmed the patch didn't fix the problem.

Right, the two are for fixing different issue.

> 
> By the way, I think Ming is right about "So release not by wq just
> reduces the chance, instead of fixing it completely.", and "move
> blk_mq_debugfs_unregister() into blk_unregister_queue()" is a good
> choice. However, blk_trace_shutdown() and blk_mq_exit_queue() also
> remove some files or dirs, and they may need to move to
> blk_unregister_queue().

Right.

Fortunately, we hold sysfs_dir_lock in blk_unregister_queue(), so
the issue can be fixed by check & remove with holding the same lock
in blk_trace_free().


Thanks, 
Ming