[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Thu, 11 Nov 2021 08:02:27 +0000
From: "Wang, Wei W" <wei.w.wang@...el.com>
To: Stefan Hajnoczi <stefanha@...hat.com>
CC: "sgarzare@...hat.com" <sgarzare@...hat.com>,
"davem@...emloft.net" <davem@...emloft.net>,
"kuba@...nel.org" <kuba@...nel.org>,
"mst@...hat.com" <mst@...hat.com>,
Paolo Bonzini <pbonzini@...hat.com>,
"kys@...rosoft.com" <kys@...rosoft.com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"virtualization@...ts.linux-foundation.org"
<virtualization@...ts.linux-foundation.org>,
"Yamahata, Isaku" <isaku.yamahata@...el.com>,
"Nakajima, Jun" <jun.nakajima@...el.com>,
"Kleen, Andi" <andi.kleen@...el.com>,
Andra Paraschiv <andraprs@...zon.com>,
"Sergio Lopez Pascual" <slp@...hat.com>
Subject: RE: [RFC] hypercall-vsock: add a new vsock transport
> From: Stefan Hajnoczi <stefanha@...hat.com>
On Wednesday, November 10, 2021 5:35 PM, Stefan Hajnoczi wrote:
> AF_VSOCK is designed to allow multiple transports, so why not. There is a cost
> to developing and maintaining a vsock transport though.
Yes. The effort could be reduced via simplifying the design as much as possible:
e.g. no ring operations - guest just sends a packet each time for the host to read.
(this transport isn't targeting for high performance)
>
> I think Amazon Nitro enclaves use virtio-vsock and I've CCed Andra in case she
> has thoughts on the pros/cons and how to minimize the trusted computing
> base.
Thanks for adding more related person to the discussion loop.
>
> If simplicity is the top priority then VIRTIO's MMIO transport without indirect
> descriptors and using the packed virtqueue layout reduces the size of the
> implementation:
> https://docs.oasis-open.org/virtio/virtio/v1.1/cs01/virtio-v1.1-cs01.html#x1-1
> 440002
I listed some considerations for virtio-mmio in the response to Michael.
Please have a check if any different thoughts.
Thanks,
Wei
Powered by blists - more mailing lists