From: Jason Wang <[email protected]>
To: Stefan Hajnoczi <[email protected]>
Cc: Christoph Hellwig <[email protected]>,
Jens Axboe <[email protected]>,
Ferry Meng <[email protected]>,
"Michael S . Tsirkin" <[email protected]>,
[email protected], [email protected],
[email protected], [email protected],
Joseph Qi <[email protected]>,
Jeffle Xu <[email protected]>
Subject: Re: [PATCH 0/3][RFC] virtio-blk: add io_uring passthrough support for virtio-blk
Date: Tue, 17 Dec 2024 10:12:28 +0800 [thread overview]
Message-ID: <CACGkMEtPdhrXnTYgF4eCC7x7fbh53hgOJ9TytYmZR=z+nFexTQ@mail.gmail.com> (raw)
In-Reply-To: <CAJSP0QXU_uNqL-9LmLRkDdPPSdUAGdesQ2DFuCMHnjyEuREvXQ@mail.gmail.com>
On Tue, Dec 17, 2024 at 12:14 AM Stefan Hajnoczi <[email protected]> wrote:
>
> On Mon, 16 Dec 2024 at 10:54, Christoph Hellwig <[email protected]> wrote:
> >
> > Hacking passthrough into virtio_blk seems like not very good layering.
> > If you have a use case where you want to use the core kernel virtio code
> > but not the protocol drivers we'll probably need a virtqueue passthrough
> > option of some kind.
>
> I think people are finding that submitting I/O via uring_cmd is faster
> than traditional io_uring. The use case isn't really passthrough, it's
> bypass :).
>
> That's why I asked Jens to weigh in on whether there is a generic
> block layer solution here. If uring_cmd is faster then maybe a generic
> uring_cmd I/O interface can be defined without tying applications to
> device-specific commands. Or maybe the traditional io_uring code path
> can be optimized so that bypass is no longer attractive.
>
> The virtio-level virtqueue passthrough idea is interesting for use
> cases that mix passthrough applications with non-passthrough
> applications. VFIO isn't enough because it prevents sharing and
> excludes non-passthrough applications. Something similar to VDPA
> might be able to pass through just a subset of virtqueues that
> userspace could access via the vhost_vdpa driver.
I thought it could be reused as a mixing approach like this. The vDPA
driver might just do a shadow virtqueue so in fact we just replace
io_uring here with the virtqueue. Or if we think vDPA is heavyweight,
vhost-blk could be another way.
> This approach
> doesn't scale if many applications are running at the same time
> because the number of virtqueues is finite and often the same as the
> number of CPUs.
>
> Stefan
>
Thanks
next prev parent reply other threads:[~2024-12-17 2:12 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-12-03 12:14 [PATCH 0/3][RFC] virtio-blk: add io_uring passthrough support for virtio-blk Ferry Meng
2024-12-03 12:14 ` [PATCH 1/3] virtio-blk: add virtio-blk chardev support Ferry Meng
2024-12-03 12:14 ` [PATCH 2/3] virtio-blk: add uring_cmd support for I/O passthru on chardev Ferry Meng
2024-12-04 15:19 ` kernel test robot
2024-12-03 12:14 ` [PATCH 3/3] virtio-blk: add uring_cmd iopoll support Ferry Meng
2024-12-04 21:47 ` [PATCH 0/3][RFC] virtio-blk: add io_uring passthrough support for virtio-blk Stefan Hajnoczi
2024-12-05 9:51 ` [Resend]Re: " Ferry Meng
2024-12-16 2:01 ` Ferry Meng
2024-12-16 7:38 ` Jason Wang
2024-12-16 12:07 ` Ferry Meng
2024-12-17 2:08 ` Jason Wang
2024-12-17 6:04 ` Ferry Meng
2024-12-16 15:54 ` Christoph Hellwig
2024-12-16 16:13 ` Stefan Hajnoczi
2024-12-17 2:12 ` Jason Wang [this message]
2024-12-17 6:08 ` Jingbo Xu
2024-12-17 17:54 ` Jens Axboe
2024-12-17 21:00 ` Stefan Hajnoczi
2024-12-17 21:07 ` Jens Axboe
2024-12-18 3:35 ` Ferry Meng
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='CACGkMEtPdhrXnTYgF4eCC7x7fbh53hgOJ9TytYmZR=z+nFexTQ@mail.gmail.com' \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
/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