From: "Michael S. Tsirkin" <mst@redhat.com>
To: Jason Wang <jasowang@redhat.com>
Cc: Parav Pandit <parav@nvidia.com>,
Max Gurtovoy <mgurtovoy@nvidia.com>,
"virtio-comment@lists.oasis-open.org"
<virtio-comment@lists.oasis-open.org>,
"cohuck@redhat.com" <cohuck@redhat.com>,
"virtio-dev@lists.oasis-open.org"
<virtio-dev@lists.oasis-open.org>, Oren Duer <oren@nvidia.com>,
Shahaf Shuler <shahafs@nvidia.com>,
"aadam@redhat.com" <aadam@redhat.com>,
"virtio@lists.oasis-open.org" <virtio@lists.oasis-open.org>
Subject: Re: [virtio-comment] RE: [PATCH v5 2/7] Introduce admin command set
Date: Tue, 28 Jun 2022 10:24:19 -0400 [thread overview]
Message-ID: <20220628095247-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <CACGkMEuF8Eshy8vKfpX_-P66wXGiBowRTwurUMUq9K0FxbfKBA@mail.gmail.com>
On Thu, Jun 23, 2022 at 11:34:42AM +0800, Jason Wang wrote:
> On Thu, Jun 23, 2022 at 10:57 AM Parav Pandit <parav@nvidia.com> wrote:
> >
> >
> > > From: Jason Wang <jasowang@redhat.com>
> > > Sent: Wednesday, June 22, 2022 10:42 PM
> >
> > [..]
> > > > Are you suggesting only feature bits through register-based transport or all
> > > commands via register-based transport?
> > >
> > > For feature bits, if we don't have admin virtqueue as transport,
> > Right.
> > Management tasks (features) done via AQ are negotiated via the AQ (regardless of device type being management dev or managed dev).
> > It is not replacing usual feature bits negotiation that has strong relationships with device initialization sequence.
>
> Just to clarify,
>
> E.g if the admin virtqueue is implemented in PF. I suggest using the
> existing PCI transport based feature negotiation since:
>
> 1) we have selector, so we don't need any new registered and we don't
> need worry about the scalability
I think this is a point that needs addressing.
To be more specific. The concern raised was that feature bits are memory
mapped. Accessing them on PCI requires an MMIO write. According to the
PCI spec it is not legal for a device to defer accepting a write until
another write is accepted (since that leads to deadlocks).
Thus any info programmed using MMIO writes has to reside in
on-card memory and can not be offloaded to system RAM.
After thinking about these issues I have an idea:
At the moment VQs are already never programmed before FEATURES_OK,
features are never programmed after FEATURES_OK.
This implies that device can actually store features in
VQ state registers temporarily until FEATURES_OK.
This is more than 24 bytes of memory per VQ, with at least
one data VQ and one admin VQ, this will be sufficient for a long time.
Thus, all that we need to do is prohibit programming VQs
before FEATURES_OK more strongly.
I can work on that if the idea is acceptable to others.
--
MST
next prev parent reply other threads:[~2022-06-28 14:24 UTC|newest]
Thread overview: 103+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-04-26 22:58 [PATCH v5 0/7] Introduce device group and device management Max Gurtovoy
2022-04-26 22:58 ` [virtio-comment] [PATCH v5 1/7] Introduce device group Max Gurtovoy
2022-05-15 15:25 ` Michael S. Tsirkin
2022-05-18 13:14 ` [virtio-comment] " Max Gurtovoy
2022-05-18 13:32 ` Cornelia Huck
2022-06-01 13:43 ` Max Gurtovoy
2022-06-02 2:21 ` Jason Wang
2022-06-02 6:59 ` Michael S. Tsirkin
2022-06-27 21:52 ` Max Gurtovoy
2022-06-28 18:54 ` Michael S. Tsirkin
2022-07-06 11:25 ` Max Gurtovoy
2022-07-06 11:42 ` Michael S. Tsirkin
2022-07-06 12:01 ` Max Gurtovoy
2022-07-06 12:23 ` Michael S. Tsirkin
2022-07-06 15:18 ` Max Gurtovoy
2022-04-26 22:58 ` [PATCH v5 2/7] Introduce admin command set Max Gurtovoy
2022-05-15 15:23 ` Michael S. Tsirkin
2022-05-16 21:08 ` [virtio-comment] " Parav Pandit
2022-05-17 10:08 ` [virtio-dev] " Cornelia Huck
2022-05-18 13:42 ` Max Gurtovoy
2022-05-17 11:48 ` Michael S. Tsirkin
2022-05-18 14:09 ` Max Gurtovoy
2022-05-18 14:42 ` [virtio] " Cornelia Huck
2022-05-18 14:48 ` Max Gurtovoy
2022-05-31 20:39 ` Parav Pandit
2022-06-20 9:23 ` Michael S. Tsirkin
2022-06-20 9:49 ` Michael S. Tsirkin
2022-06-20 9:59 ` Michael S. Tsirkin
2022-06-20 11:06 ` Parav Pandit
2022-06-20 16:46 ` Michael S. Tsirkin
2022-06-20 16:54 ` Max Gurtovoy
2022-06-20 17:04 ` Michael S. Tsirkin
2022-06-20 17:19 ` Parav Pandit
2022-06-20 20:53 ` Michael S. Tsirkin
2022-06-20 23:54 ` Parav Pandit
2022-06-20 17:16 ` Parav Pandit
2022-06-23 1:26 ` Jason Wang
2022-06-23 2:07 ` Parav Pandit
2022-06-23 2:41 ` Jason Wang
2022-06-23 2:57 ` Parav Pandit
2022-06-23 3:34 ` Jason Wang
2022-06-28 14:24 ` Michael S. Tsirkin [this message]
2022-06-29 8:43 ` Jason Wang
2022-06-29 9:02 ` Michael S. Tsirkin
2022-06-30 1:53 ` Jason Wang
2022-05-18 13:39 ` [virtio-comment] " Max Gurtovoy
2022-05-18 13:50 ` [virtio] " Cornelia Huck
2022-05-18 14:16 ` Max Gurtovoy
2022-06-20 22:26 ` Michael S. Tsirkin
2022-06-20 21:08 ` Michael S. Tsirkin
2022-04-26 22:58 ` [PATCH v5 3/7] Introduce new destination type for admin commands Max Gurtovoy
2022-05-15 15:01 ` Michael S. Tsirkin
2022-05-18 14:27 ` [virtio-comment] " Max Gurtovoy
2022-05-15 15:09 ` Michael S. Tsirkin
2022-05-16 21:21 ` Parav Pandit
2022-05-16 23:33 ` Michael S. Tsirkin
2022-05-18 14:36 ` Max Gurtovoy
2022-05-18 14:34 ` Max Gurtovoy
2022-05-18 23:55 ` Michael S. Tsirkin
2022-04-26 22:58 ` [PATCH v5 4/7] Introduce virtio admin virtqueue Max Gurtovoy
2022-05-15 14:59 ` Michael S. Tsirkin
2022-05-18 14:37 ` Max Gurtovoy
2022-05-18 23:56 ` Michael S. Tsirkin
2022-04-26 22:58 ` [PATCH v5 5/7] Add miscellaneous configuration structure for PCI Max Gurtovoy
2022-05-15 14:49 ` Michael S. Tsirkin
2022-06-01 14:46 ` Max Gurtovoy
2022-05-15 14:57 ` Michael S. Tsirkin
2022-05-17 10:12 ` [virtio] " Cornelia Huck
2022-05-18 14:42 ` Max Gurtovoy
2022-05-18 23:58 ` Michael S. Tsirkin
2022-04-26 22:58 ` [PATCH v5 6/7] Introduce MGMT admin commands Max Gurtovoy
2022-05-15 14:37 ` Michael S. Tsirkin
2022-05-16 21:47 ` Parav Pandit
2022-05-17 12:31 ` [virtio-comment] " Michael S. Tsirkin
2022-05-18 15:14 ` Max Gurtovoy
2022-05-17 2:28 ` Jason Wang
2022-05-18 15:27 ` Max Gurtovoy
2022-05-18 16:41 ` Michael S. Tsirkin
2022-05-18 23:10 ` Max Gurtovoy
2022-05-18 15:03 ` Max Gurtovoy
2022-06-20 9:45 ` Michael S. Tsirkin
2022-04-26 22:58 ` [PATCH v5 7/7] RFC: add initial support for configuring feature bits Max Gurtovoy
2022-05-15 14:38 ` Michael S. Tsirkin
2022-05-18 15:31 ` Max Gurtovoy
2022-05-18 16:34 ` Michael S. Tsirkin
2022-05-18 23:18 ` Max Gurtovoy
2022-05-15 15:27 ` [PATCH v5 0/7] Introduce device group and device management Michael S. Tsirkin
2022-05-18 15:32 ` Max Gurtovoy
2022-07-05 13:56 ` Michael S. Tsirkin
2022-07-05 15:11 ` [virtio-comment] " Parav Pandit
2022-07-06 2:54 ` [virtio-dev] " Jason Wang
2022-07-06 10:10 ` Michael S. Tsirkin
2022-07-06 10:46 ` [virtio-comment] " Parav Pandit
2022-07-06 11:00 ` Michael S. Tsirkin
2022-07-06 20:45 ` Parav Pandit
2022-07-24 21:09 ` Michael S. Tsirkin
2022-07-24 21:25 ` [virtio-comment] " Parav Pandit
2022-07-24 23:41 ` Michael S. Tsirkin
2022-07-25 2:53 ` [virtio-comment] " Parav Pandit
2022-07-25 7:44 ` Michael S. Tsirkin
2022-07-30 13:21 ` [virtio-comment] " Parav Pandit
2022-07-31 15:38 ` Michael S. Tsirkin
2022-08-02 17:40 ` 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=20220628095247-mutt-send-email-mst@kernel.org \
--to=mst@redhat.com \
--cc=aadam@redhat.com \
--cc=cohuck@redhat.com \
--cc=jasowang@redhat.com \
--cc=mgurtovoy@nvidia.com \
--cc=oren@nvidia.com \
--cc=parav@nvidia.com \
--cc=shahafs@nvidia.com \
--cc=virtio-comment@lists.oasis-open.org \
--cc=virtio-dev@lists.oasis-open.org \
--cc=virtio@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.