Re: [virtio-comment] Re: [PATCH] virtio-net: define feature of per-packet RSS hash delivery

[Yuri Benditovich <yuri.benditovich@daynix.com>]

OK, I will send the next version of the patch.
I'm just waiting for merge of previous RSS patch to avoid heavy rebase.

On Mon, Jan 20, 2020 at 6:14 PM Michael S. Tsirkin <mst@redhat.com> wrote:
>
> On Fri, Dec 13, 2019 at 03:38:11AM -0500, Michael S. Tsirkin wrote:
> > On Fri, Dec 13, 2019 at 10:21:53AM +0200, Yuri Benditovich wrote:
> > > On Fri, Dec 13, 2019 at 9:16 AM Michael S. Tsirkin <mst@redhat.com> wrote:
> > > >
> > > > On Fri, Dec 13, 2019 at 06:34:39AM +0200, Yuri Benditovich wrote:
> > > > >
> > > > > So there are 2 possible profiles of devices that 'do not have RSS' but
> > > > > deliver hash:
> > > > > 1. The device is in general can do RSS, but has only one pair of
> > > > > queues and does only hash calculation
> > > > > 2. The device has multiple queues but is configured to do auto
> > > > > steering and we want also hash from it.
> > > > >
> > > > > My impression is that case 2 is rather academic one (at least looking forward).
> > > > > Do you agree?
> > > >
> > > > I am not sure I agree. auto mq is much easier to configure than RSS
> > > > so I image it's useful even if one wants hash.
> > >
> > > With RSS and hash delivery we're mainly targeting virtio-net in hardware.
> > > Auto mq is much more complicated, correct?
> >
> > Hardware implementing it exists, it's not too hard, just some tables
> > to maintain.
>
> So what is the resolution here? To me it seems easier to just allow
> this than worry whether it's useful. And devices that do not
> want to support some combinations, don't have to, right?
>
> > > >
> > > > --
> > > > MST
> > > >
> >
> >
> > This publicly archived list offers a means to provide input to the
> > OASIS Virtual I/O Device (VIRTIO) TC.
> >
> > In order to verify user consent to the Feedback License terms and
> > to minimize spam in the list archive, subscription is required
> > before posting.
> >
> > Subscribe: virtio-comment-subscribe@lists.oasis-open.org
> > Unsubscribe: virtio-comment-unsubscribe@lists.oasis-open.org
> > List help: virtio-comment-help@lists.oasis-open.org
> > List archive: https://lists.oasis-open.org/archives/virtio-comment/
> > Feedback License: https://www.oasis-open.org/who/ipr/feedback_license.pdf
> > List Guidelines: https://www.oasis-open.org/policies-guidelines/mailing-lists
> > Committee: https://www.oasis-open.org/committees/virtio/
> > Join OASIS: https://www.oasis-open.org/join/
>
>
> This publicly archived list offers a means to provide input to the
> OASIS Virtual I/O Device (VIRTIO) TC.
>
> In order to verify user consent to the Feedback License terms and
> to minimize spam in the list archive, subscription is required
> before posting.
>
> Subscribe: virtio-comment-subscribe@lists.oasis-open.org
> Unsubscribe: virtio-comment-unsubscribe@lists.oasis-open.org
> List help: virtio-comment-help@lists.oasis-open.org
> List archive: https://lists.oasis-open.org/archives/virtio-comment/
> Feedback License: https://www.oasis-open.org/who/ipr/feedback_license.pdf
> List Guidelines: https://www.oasis-open.org/policies-guidelines/mailing-lists
> Committee: https://www.oasis-open.org/committees/virtio/
> Join OASIS: https://www.oasis-open.org/join/
>