Re: [virtio-dev] Re: [RFC] Virtio RDMA

[Jason Wang <jasowang@redhat.com>]

On Wed, Feb 16, 2022 at 4:01 PM Junji Wei <weijunji@bytedance.com> wrote:
>
>
> > On Feb 16, 2022, at 3:26 PM, Leon Romanovsky <leon@kernel.org> wrote:
> >
> > On Wed, Feb 16, 2022 at 03:13:11PM +0800, Junji Wei wrote:
> >>
> >
> > <...>
> >
> >>>> We can't. So do you mean we can implement virtio-rdma only for IB in the future?
> >>>
> >>> It's probably virtio-IB but we need to listen to others.
> >>
> >> Agreed, one problem is that there might be some duplicated works.
> >
> > Once it will be needed, the code can be refactored. IB and RoCE are
> > different from user perspective, so combining them into one virtio-rdma
> > module doesn't give too much advantage.
>
> Yes, code would be fine. But there maybe some duplicated contents in spec.

We can carefully design the interfaces in order to eliminate the
duplication in the spec.

Thanks

>
> >
> >>
> >>>
> >>>>
> >>>>>> And currently virtio-rdma doesn't have a strong dependency on
> >>>>>> virtio-net (except for gid and ah stuffs). Is it OK to mix them up?
> >>>>>
> >>>>> There are a bunch of hardware vendors that ship a converged Ethernet
> >>>>> adapter. It simplifies the management and deployment.
> >>>>
> >>>> Virtio-rdma is not depend on virtio-net, we can bind it to another ethernet device
> >>>> via mac address in the future. And is it too mass to mix up two different device
> >>>> in one spec?
> >>>
> >>> So either should be fine, we just need to figure out which one is
> >>> better. What I meant is to extend the virtio-net to be capable of
> >>> converged ethernet.
> >>
> >> Got it. One question is whether there will be some cases that user want
> >> to use virtio-rdma binding to other types of ethernet device such as
> >> passthroughed net device. In this case, we donât need a virtio-net
> >> device actually.
> >
> > What is the use case for this virtio-rdma? Especially in context of RXE.
>
> Hmm... yes, we didnât find one. In passthrough case we can use RXE directly.
>
> Thanks.
>