From: "Michael S. Tsirkin" <mst@redhat.com>
To: Parav Pandit <parav@nvidia.com>
Cc: "virtio-dev@lists.oasis-open.org"
<virtio-dev@lists.oasis-open.org>,
"cohuck@redhat.com" <cohuck@redhat.com>,
"virtio-comment@lists.oasis-open.org"
<virtio-comment@lists.oasis-open.org>,
Shahaf Shuler <shahafs@nvidia.com>
Subject: [virtio-comment] Re: [PATCH 00/11] Introduce transitional mmr pci device
Date: Mon, 3 Apr 2023 11:23:05 -0400 [thread overview]
Message-ID: <20230403111735-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <PH0PR12MB5481EC65708410304D9C1EA3DC929@PH0PR12MB5481.namprd12.prod.outlook.com>
On Mon, Apr 03, 2023 at 03:16:53PM +0000, Parav Pandit wrote:
>
>
> > From: Michael S. Tsirkin <mst@redhat.com>
> > Sent: Monday, April 3, 2023 11:07 AM
>
> > > > OTOH it is presumably required for scalability anyway, no?
> > > No.
> > > Most new generation SIOV and SR-IOV devices operate without any para-
> > virtualization.
> >
> > Don't see the connection to PV. You need an emulation layer in the host if you
> > want to run legacy guests. Looks like it could do transport vq just as well.
> >
> Transport vq for legacy MMR purpose seems fine with its latency and DMA overheads.
> Your question was about "scalability".
> After your latest response, I am unclear what "scalability" means.
> Do you mean saving the register space in the PCI device?
yes that's how you used scalability in the past.
> If yes, than, no for legacy guests for scalability it is not required, because the legacy register is subset of 1.x.
Weird. what does guest being legacy have to do with a wish to save
registers on the host hardware? You don't have so many legacy guests as
modern guests? Why?
>
> > > > And presumably it can all be done in firmware ...
> > > > Is there actual hardware that can't implement transport vq but is
> > > > going to implement the mmr spec?
> > > >
> > > Nvidia and Marvell DPUs implement MMR spec.
> >
> > Hmm implement it in what sense exactly?
> >
> Do not follow the question.
> The proposed series will be implemented as PCI SR-IOV devices using MMR spec.
>
> > > Transport VQ has very high latency and DMA overheads for 2 to 4 bytes
> > read/write.
> >
> > How many of these 2 byte accesses trigger from a typical guest?
> >
> Mostly during the VM boot time. 20 to 40 registers read write access.
That is not a lot! How long does a DMA operation take then?
> > > And before discussing "why not that approach", lets finish reviewing "this
> > approach" first.
> >
> > That's a weird way to put it. We don't want so many ways to do legacy if we can
> > help it.
> Sure, so lets finish the review of current proposal details.
> At the moment
> a. I don't see any visible gain of transport VQ other than device reset part I explained.
For example, we do not need a new range of device IDs and existing
drivers can bind on the host.
> b. it can be a way with high latency, DMA overheads on the virtqueue for read/writes for small access.
numbers?
--
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/
next prev parent reply other threads:[~2023-04-03 15:23 UTC|newest]
Thread overview: 199+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-30 22:58 [virtio-comment] [PATCH 00/11] Introduce transitional mmr pci device Parav Pandit
2023-03-30 22:58 ` [virtio-comment] [PATCH 01/11] transport-pci: Use lowecase alphabets Parav Pandit
2023-03-30 22:58 ` [virtio-comment] [PATCH 02/11] transport-pci: Move transitional device id to legacy section Parav Pandit
2023-03-31 6:43 ` [virtio-comment] " Michael S. Tsirkin
2023-03-31 21:24 ` [virtio-comment] " Parav Pandit
2023-04-02 7:54 ` [virtio-comment] " Michael S. Tsirkin
2023-04-03 14:42 ` [virtio-comment] " Parav Pandit
2023-04-03 14:50 ` [virtio-comment] " Michael S. Tsirkin
2023-04-03 14:58 ` [virtio-comment] " Parav Pandit
2023-04-03 15:14 ` [virtio-comment] " Michael S. Tsirkin
2023-03-30 22:58 ` [virtio-comment] [PATCH 03/11] transport-pci: Split notes of PCI Device Layout Parav Pandit
2023-03-30 22:58 ` [virtio-comment] [PATCH 04/11] transport-pci: Rename and move legacy PCI Device layout section Parav Pandit
2023-03-30 22:58 ` [virtio-comment] [PATCH 05/11] introduction: Add missing helping verb Parav Pandit
2023-03-30 22:58 ` [virtio-comment] [PATCH 06/11] introduction: Introduce transitional MMR interface Parav Pandit
2023-04-07 9:17 ` [virtio-comment] " Michael S. Tsirkin
2023-03-30 22:58 ` [virtio-comment] [PATCH 07/11] transport-pci: Introduce transitional MMR device id Parav Pandit
2023-04-04 7:28 ` [virtio-comment] " Michael S. Tsirkin
2023-04-04 16:08 ` Parav Pandit
2023-04-07 12:03 ` Michael S. Tsirkin
2023-04-07 15:18 ` Parav Pandit
2023-04-07 15:51 ` Michael S. Tsirkin
2023-04-09 3:15 ` Parav Pandit
2023-04-10 10:18 ` Michael S. Tsirkin
2023-04-10 14:34 ` Parav Pandit
2023-04-10 19:58 ` Michael S. Tsirkin
2023-04-10 20:16 ` Parav Pandit
2023-04-07 8:37 ` Michael S. Tsirkin
2023-03-30 22:58 ` [virtio-comment] [PATCH 08/11] transport-pci: Introduce virtio extended capability Parav Pandit
2023-04-04 7:35 ` [virtio-comment] " Michael S. Tsirkin
2023-04-04 7:54 ` Cornelia Huck
2023-04-04 12:43 ` Michael S. Tsirkin
2023-04-04 13:19 ` Cornelia Huck
2023-04-04 14:37 ` Michael S. Tsirkin
2023-04-10 16:21 ` Parav Pandit
2023-04-10 19:49 ` Michael S. Tsirkin
2023-04-10 19:57 ` [virtio-comment] " Parav Pandit
2023-04-10 20:02 ` [virtio-comment] " Michael S. Tsirkin
2023-04-11 8:39 ` Cornelia Huck
2023-04-04 21:18 ` [virtio-comment] " Parav Pandit
2023-04-05 5:10 ` [virtio-comment] " Michael S. Tsirkin
2023-04-05 13:16 ` [virtio-comment] " Parav Pandit
2023-04-07 8:15 ` [virtio-comment] " Michael S. Tsirkin
2023-04-10 1:36 ` [virtio-comment] Re: [virtio-dev] " Jason Wang
2023-04-10 6:24 ` Michael S. Tsirkin
2023-04-10 7:16 ` Jason Wang
2023-04-10 10:04 ` Michael S. Tsirkin
2023-04-11 2:19 ` Jason Wang
2023-04-11 7:00 ` Michael S. Tsirkin
2023-04-11 9:07 ` Jason Wang
2023-04-11 10:43 ` Michael S. Tsirkin
2023-04-11 13:59 ` [virtio-comment] " Parav Pandit
2023-04-11 14:11 ` [virtio-comment] " Michael S. Tsirkin
2023-04-11 13:47 ` Parav Pandit
2023-04-11 14:02 ` Michael S. Tsirkin
2023-04-11 14:07 ` Parav Pandit
2023-04-11 14:10 ` Michael S. Tsirkin
2023-04-11 14:30 ` Parav Pandit
2023-04-10 17:54 ` Parav Pandit
2023-04-10 17:58 ` Parav Pandit
2023-04-11 3:28 ` Jason Wang
2023-04-11 19:01 ` [virtio-comment] " Parav Pandit
2023-04-11 21:25 ` [virtio-comment] " Michael S. Tsirkin
2023-04-12 0:40 ` [virtio-comment] " Parav Pandit
2023-04-12 2:56 ` [virtio-comment] " Michael S. Tsirkin
2023-04-12 4:07 ` Jason Wang
2023-04-12 4:20 ` Michael S. Tsirkin
2023-04-12 4:53 ` Jason Wang
2023-04-12 5:25 ` Michael S. Tsirkin
2023-04-12 5:37 ` Jason Wang
2023-04-13 17:03 ` Michael S. Tsirkin
2023-04-12 4:04 ` Jason Wang
2023-04-12 4:13 ` [virtio-comment] " Parav Pandit
2023-04-12 4:20 ` [virtio-comment] " Michael S. Tsirkin
2023-04-12 4:55 ` Jason Wang
2023-05-19 6:10 ` [virtio-comment] " Michael S. Tsirkin
2023-05-19 21:02 ` [virtio-comment] " Parav Pandit
2023-05-21 5:57 ` [virtio-comment] " Michael S. Tsirkin
2023-05-21 13:24 ` [virtio-comment] " Parav Pandit
2023-05-21 14:34 ` [virtio-comment] " Michael S. Tsirkin
2023-03-30 22:58 ` [virtio-comment] [PATCH 09/11] transport-pci: Describe PCI MMR dev config registers Parav Pandit
2023-04-07 8:55 ` [virtio-comment] " Michael S. Tsirkin
2023-04-10 1:33 ` Jason Wang
2023-04-10 6:14 ` Michael S. Tsirkin
2023-04-10 6:20 ` Jason Wang
2023-04-10 6:39 ` Michael S. Tsirkin
2023-04-10 7:20 ` Jason Wang
2023-04-10 10:06 ` Michael S. Tsirkin
2023-04-11 2:13 ` Jason Wang
2023-04-11 7:04 ` Michael S. Tsirkin
2023-04-11 9:01 ` Jason Wang
[not found] ` <CALBs2cXURMEzCGnULicXbsBfwnKE5cZOz=M-_hhFCXZ=Lqb9Nw@mail.gmail.com>
2023-04-11 10:39 ` Michael S. Tsirkin
2023-04-11 10:42 ` Michael S. Tsirkin
2023-04-12 3:58 ` Jason Wang
2023-04-12 4:15 ` Michael S. Tsirkin
2023-04-12 4:51 ` Jason Wang
2023-04-12 5:01 ` Parav Pandit
2023-04-12 5:14 ` Jason Wang
2023-04-12 5:30 ` Parav Pandit
2023-04-12 5:38 ` Jason Wang
2023-04-12 5:55 ` Parav Pandit
2023-04-12 6:15 ` Jason Wang
2023-04-12 14:23 ` Parav Pandit
2023-04-13 1:48 ` Jason Wang
2023-04-13 3:31 ` Parav Pandit
2023-04-13 5:14 ` Jason Wang
2023-04-13 17:19 ` Michael S. Tsirkin
2023-04-13 19:39 ` Parav Pandit
2023-04-14 3:09 ` Jason Wang
2023-04-14 3:18 ` Parav Pandit
2023-04-14 3:37 ` Jason Wang
2023-04-14 3:51 ` Parav Pandit
2023-04-14 7:05 ` Michael S. Tsirkin
2023-04-17 3:22 ` Jason Wang
2023-04-17 17:23 ` Parav Pandit
2023-04-17 20:26 ` Michael S. Tsirkin
2023-04-17 20:28 ` Parav Pandit
2023-04-18 0:36 ` Jason Wang
2023-04-18 1:30 ` Parav Pandit
2023-04-18 11:58 ` Michael S. Tsirkin
2023-04-18 12:09 ` Parav Pandit
2023-04-18 12:30 ` Michael S. Tsirkin
2023-04-18 12:36 ` Parav Pandit
2023-04-18 1:01 ` Jason Wang
2023-04-18 1:48 ` Parav Pandit
2023-04-13 17:24 ` Parav Pandit
2023-04-13 21:02 ` Michael S. Tsirkin
2023-04-13 21:08 ` Parav Pandit
2023-04-14 2:36 ` Jason Wang
2023-04-14 2:43 ` Parav Pandit
2023-04-14 6:57 ` Michael S. Tsirkin
2023-04-16 13:41 ` Parav Pandit
2023-04-16 20:44 ` Michael S. Tsirkin
2023-04-17 16:59 ` Parav Pandit
2023-04-18 1:09 ` Jason Wang
2023-04-18 1:37 ` Parav Pandit
2023-04-14 6:58 ` Michael S. Tsirkin
2023-04-14 3:08 ` Jason Wang
2023-04-14 3:13 ` Parav Pandit
2023-04-14 3:18 ` Jason Wang
2023-04-14 3:22 ` Parav Pandit
2023-04-14 3:29 ` Jason Wang
2023-04-11 13:57 ` Parav Pandit
2023-04-12 4:33 ` Michael S. Tsirkin
2023-03-30 22:58 ` [virtio-comment] [PATCH 10/11] transport-pci: Use driver notification PCI capability Parav Pandit
2023-04-12 4:31 ` [virtio-comment] " Michael S. Tsirkin
2023-04-12 4:37 ` [virtio-comment] " Parav Pandit
2023-04-12 4:43 ` [virtio-comment] " Michael S. Tsirkin
2023-04-12 4:48 ` [virtio-comment] " Parav Pandit
2023-04-12 5:02 ` [virtio-comment] " Michael S. Tsirkin
2023-04-12 5:06 ` Parav Pandit
2023-04-12 5:17 ` Michael S. Tsirkin
2023-04-12 5:24 ` Parav Pandit
2023-04-12 5:27 ` Michael S. Tsirkin
2023-03-30 22:58 ` [virtio-comment] [PATCH 11/11] conformance: Add transitional MMR interface conformance Parav Pandit
2023-03-31 7:03 ` [virtio-comment] Re: [PATCH 00/11] Introduce transitional mmr pci device Michael S. Tsirkin
2023-03-31 21:43 ` Parav Pandit
2023-04-03 14:53 ` Michael S. Tsirkin
2023-04-03 14:57 ` [virtio-comment] " Parav Pandit
2023-04-03 15:06 ` [virtio-comment] " Michael S. Tsirkin
2023-04-03 15:16 ` [virtio-comment] " Parav Pandit
2023-04-03 15:23 ` Michael S. Tsirkin [this message]
2023-04-03 15:34 ` [virtio-comment] " Michael S. Tsirkin
2023-04-03 15:47 ` [virtio-comment] " Parav Pandit
2023-04-03 17:28 ` [virtio-comment] " Michael S. Tsirkin
2023-04-03 17:35 ` Parav Pandit
2023-04-03 17:39 ` Michael S. Tsirkin
2023-04-03 15:36 ` Parav Pandit
2023-04-03 17:16 ` Michael S. Tsirkin
2023-04-03 17:29 ` Parav Pandit
2023-04-03 18:02 ` Michael S. Tsirkin
2023-04-03 20:25 ` Parav Pandit
2023-04-03 21:04 ` Michael S. Tsirkin
2023-04-03 22:00 ` Parav Pandit
2023-04-07 9:35 ` Michael S. Tsirkin
2023-04-10 1:52 ` Jason Wang
2023-04-03 14:45 ` [virtio-comment] " Stefan Hajnoczi
2023-04-03 14:53 ` Parav Pandit
2023-04-03 17:48 ` Michael S. Tsirkin
2023-04-03 19:11 ` Stefan Hajnoczi
2023-04-03 20:03 ` Michael S. Tsirkin
2023-04-03 19:48 ` Parav Pandit
2023-04-03 20:02 ` Michael S. Tsirkin
2023-04-03 20:42 ` Parav Pandit
2023-04-03 21:14 ` Michael S. Tsirkin
2023-04-03 22:08 ` Parav Pandit
2023-04-03 19:10 ` Stefan Hajnoczi
2023-04-03 20:27 ` Parav Pandit
2023-04-04 14:30 ` Stefan Hajnoczi
2023-04-12 4:48 ` [virtio-comment] " Michael S. Tsirkin
2023-04-12 4:52 ` [virtio-comment] " Parav Pandit
2023-04-12 5:12 ` [virtio-comment] " Michael S. Tsirkin
2023-04-12 5:15 ` Parav Pandit
2023-04-12 5:23 ` Michael S. Tsirkin
2023-04-12 5:39 ` Parav Pandit
2023-04-12 6:02 ` Parav Pandit
2023-04-12 5:10 ` [virtio-comment] Re: [virtio-dev] " Halil Pasic
2023-04-25 2:42 ` [virtio-comment] " Parav Pandit
2023-05-02 7:17 ` David Edmondson
2023-05-02 13:54 ` Parav Pandit
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230403111735-mutt-send-email-mst@kernel.org \
--to=mst@redhat.com \
--cc=cohuck@redhat.com \
--cc=parav@nvidia.com \
--cc=shahafs@nvidia.com \
--cc=virtio-comment@lists.oasis-open.org \
--cc=virtio-dev@lists.oasis-open.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).