Re: [RFC PATCH] scsi: fix oops in scsi_uninit_cmd()

[Jason Yan]

On 2019/3/14 7:51, Bart Van Assche wrote:
> On Thu, 2019-02-21 at 16:53 +0800, Jason Yan wrote:
> > On 2019/2/20 23:18, Christoph Hellwig wrote:
> > > [fullquote removed, please follow proper mail etiquette]
> > >
> > > On Tue, Feb 19, 2019 at 08:56:28AM -0800, Bart Van Assche wrote:
> > > > regression in the SCSI sd driver due to the switch from the legacy block
> > > > layer to scsi-mq. The above patch introduces two atomic operations in the
> > > > hot path and hence would introduce a performance regression. I think this
> > > > can be avoided by making sure that sd_uninit_command() gets called before
> > > > the request tag is freed. What changes would be required to make the block
> > > > layer core call sd_uninit_command() before the request tag is freed? Would
> > > > introducing prep_rq_fn and unprep_rq_fn callbacks in struct blk_mq_ops and
> > > > making sure that the SCSI core sets these callback function pointers
> > > > appropriately be sufficient? Would such a change allow to simplify the NVMe
> > > > initiator driver? Are there any alternatives to this approach that are more
> > > > elegant?
> > >
> > > Additional indirect calls in the I/O fast path is something I'd rather
> > > avoid.  But I don't fully understand the problem yet - where do
> > > we release a disk reference from blk_update_request?
> >
> > When userspace close the fd after blk_update_request() and before
> > scsi_mq_uninit_cmd(), a disk reference will be released. It is not the
> > blk_update_request() directly released it.
> >
> > close
> >      ->sd_release
> >         ->scsi_disk_put
> >           ->scsi_disk_release
> >             ->disk->private_data = NULL;
> >
> > The userspace can close the fd because blk_update_request() returned the
> > last IO , the userspace application does not have to stuck on read() or
> > write(). The window is very small, but it can be reproduce every day
> > in our testcases. So I'm very curious why. One possible explanation is
> > that we enabled kernel preempt(CONFIG_PREEMPT).
> >
> > And why can't we move that release to __blk_mq_end_request?
>
> Hi Jason,
>
> What is the current status of this issue?

Hi Bart,

I did not find any other approach that will not affect the hot path. I 
don't know if you guys have other suggestions?

> Thanks,
>
> Bart.
>
> .