On 18.04.2019 14:26, David Marchand wrote:
> On Wed, Apr 17, 2019 at 4:27 PM Kevin Traynor <ktray...@redhat.com
> <mailto:ktray...@redhat.com>> wrote:
>
> On 16/04/2019 10:45, David Marchand wrote:
>
> > Note: this patch requires a fix for the vhost library submitted here:
> > http://patchwork.dpdk.org/patch/52680/
> >
> > Without it, this change will do nothing but have openvswitch complain
> > that the vhost device is unknown:
> >
> > dpdk|INFO|VHOST_CONFIG: vhost peer closed
> > dpdk|ERR|VHOST_CONFIG: (0) device not found.
> >
> > dpdk|INFO|VHOST_CONFIG: vhost peer closed
> > dpdk|ERR|VHOST_CONFIG: (1) device not found.
> >
> > Signed-off-by: David Marchand <david.march...@redhat.com
> <mailto:david.march...@redhat.com>>
>
> I think this probably shouldn't be merged until OVS is using a version
> of DPDK with the linked patch because it is introducing errors in the
> logs which can be alarming for a user. The DPDK fix should be part of
> DPDK 18.11.2.
>
>
> It sounds sane yes.
> Now, 18.11.2 is not ready yet (no pressure Kevin ;-)) as we just got 18.11.1.
>
> Ilya, if we go that way, what do you think of considering the first patch for
> merge now and the others two fixes for when 18.11.2 is ready ?
I think, we can go with first two patches now and merge the last one when
the 18.11.2 ready. Second patch should not make any harm as all modern
drivers has F_MQ support and we'll not waste much time on polling disabled
queues. Anyway, polling of disabled queues better than guest-controllable
reconfigurations. We also need to think about backporting the second patch.
> --
> David Marchand
_______________________________________________
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
