Re: [PATCH 0/3] scsi: scsi-mq: don't hold host_busy in IO path

[Laurence Oberman <loberman@redhat.com>]

On Fri, 2018-04-27 at 15:48 +0000, Bart Van Assche wrote:
> On Fri, 2018-04-27 at 09:39 -0600, Jens Axboe wrote:
> > blk_mq_tagset_busy_iter(&shost->tag_set, scsi_host_check_in_flight,
> > 			&in_flight);
> > return in_flight.cnt + atomic_read(&shost->host_busy);
> > 
> > The atomic read is basically free, once we get rid of the dirty of
> > that
> > variable on each IO.
> 
> Hello Jens,
> 
> What makes you think that " + atomic_read(&shost->host_busy)" is
> necessary?
> I am not aware of any code outside the SCSI core that modifies the
> host_busy
> member.
> 
> Thanks,
> 
> Bart.
> 
> 
> 

As part of testing latest upstream in MQ and non-MQ I intend to test
this patch series fully on actual hardware
F/C 8G to memory backed array LUNS and of course SRP/RDMA
I have started working on this and will report back.
Thanks
Laurence