On Thu, Aug 25, 2022 at 11:50 PM Suwan Kim <suwan.kim027@xxxxxxxxx> wrote:
>
> On Thu, Aug 25, 2022 at 2:32 AM Stefan Hajnoczi <stefanha@xxxxxxxxxx> wrote:
> >
> > On Wed, Aug 24, 2022 at 10:16:10PM +0900, Kim Suwan wrote:
> > > Hi Stefan,
> > >
> > > On Wed, Aug 24, 2022 at 5:56 AM Stefan Hajnoczi <stefanha@xxxxxxxxxx> wrote:
> > > >
> > > > On Tue, Aug 23, 2022 at 11:50:05PM +0900, Suwan Kim wrote:
> > > > > @@ -409,6 +409,8 @@ static bool virtblk_add_req_batch(struct virtio_blk_vq *vq,
> > > > > virtblk_unmap_data(req, vbr);
> > > > > virtblk_cleanup_cmd(req);
> > > > > rq_list_add(requeue_list, req);
> > > > > + } else {
> > > > > + blk_mq_start_request(req);
> > > > > }
> > > >
> > > > The device may see new requests as soon as virtblk_add_req() is called
> > > > above. Therefore the device may complete the request before
> > > > blk_mq_start_request() is called.
> > > >
> > > > rq->io_start_time_ns = ktime_get_ns() will be after the request was
> > > > actually submitted.
> > > >
> > > > I think blk_mq_start_request() needs to be called before
> > > > virtblk_add_req().
> > > >
> > >
> > > But if blk_mq_start_request() is called before virtblk_add_req()
> > > and virtblk_add_req() fails, it can trigger WARN_ON_ONCE() at
> > > virtio_queue_rq().
> > >
> > > With regard to the race condition between virtblk_add_req() and
> > > completion, I think that the race condition can not happen because
> > > virtblk_add_req() holds vq lock with irq saving and completion side
> > > (virtblk_done, virtblk_poll) need to acquire the vq lock also.
> > > Moreover, virtblk_done() is irq context so I think it can not be
> > > executed until virtblk_add_req() releases the lock.
> >
> > I agree there is no race condition regarding the ordering of
> > blk_mq_start_request() and request completion. The spinlock prevents
> > that and I wasn't concerned about that part.
> >
> > The issue is that the timestamp will be garbage. If we capture the
> > timestamp during/after the request is executing, then the collected
> > statistics will be wrong.
> >
> > Can you look for another solution that doesn't break the timestamp?
> >
> > FWIW I see that the rq->state state machine allows returning to the idle
> > state once the request has been started: __blk_mq_requeue_request().
>
> I considered blk_mq_requeue_request() to handle error cases but
> I didn't use it because I think it can make the error path request
> processing slower than requeuing an error request to plug list again.
>
> But there doesn't seem to be any other option that doesn't break
> the timestamp.
>
> As you said, I will use __blk_mq_requeue_request() and send
> new patch soon.
>
> To Alexandre,
>
> I will share new diff soon. Could you please test one more time?
Absolutely! Thanks for looking into this.
