From: Stefan Hajnoczi <stefanha@redhat.com>
To: Hanna Czenczek <hreitz@redhat.com>
Cc: qemu-block@nongnu.org, qemu-devel@nongnu.org,
Fiona Ebner <f.ebner@proxmox.com>,
Paolo Bonzini <pbonzini@redhat.com>,
Kevin Wolf <kwolf@redhat.com>,
"Michael S . Tsirkin" <mst@redhat.com>,
Fam Zheng <fam@euphon.net>
Subject: Re: [PATCH 2/2] virtio: Keep notifications disabled during drain
Date: Thu, 25 Jan 2024 16:32:06 -0500 [thread overview]
Message-ID: <20240125213206.GB58542@fedora> (raw)
In-Reply-To: <cb2e1577-9079-49ea-90e4-71d57b78290f@redhat.com>
[-- Attachment #1: Type: text/plain, Size: 1944 bytes --]
On Thu, Jan 25, 2024 at 07:32:12PM +0100, Hanna Czenczek wrote:
> On 25.01.24 19:18, Hanna Czenczek wrote:
> > On 25.01.24 19:03, Stefan Hajnoczi wrote:
> > > On Wed, Jan 24, 2024 at 06:38:30PM +0100, Hanna Czenczek wrote:
>
> [...]
>
> > > > @@ -3563,6 +3574,13 @@ void
> > > > virtio_queue_aio_attach_host_notifier(VirtQueue *vq, AioContext
> > > > *ctx)
> > > > aio_set_event_notifier_poll(ctx, &vq->host_notifier,
> > > > virtio_queue_host_notifier_aio_poll_begin,
> > > > virtio_queue_host_notifier_aio_poll_end);
> > > > +
> > > > + /*
> > > > + * We will have ignored notifications about new requests
> > > > from the guest
> > > > + * during the drain, so "kick" the virt queue to process
> > > > those requests
> > > > + * now.
> > > > + */
> > > > + virtio_queue_notify(vq->vdev, vq->queue_index);
> > > event_notifier_set(&vq->host_notifier) is easier to understand because
> > > it doesn't contain a non-host_notifier code path that we must not take.
> > >
> > > Is there a reason why you used virtio_queue_notify() instead?
> >
> > Not a good one anyway!
> >
> > virtio_queue_notify() is just what seemed obvious to me (i.e. to notify
> > the virtqueue). Before removal of the AioContext lock, calling
> > handle_output seemed safe. But, yes, there was the discussion on the
> > RFC that it really isn’t. I didn’t consider that means we must rely on
> > virtio_queue_notify() calling event_notifier_set(), so we may as well
> > call it explicitly here.
> >
> > I’ll fix it, thanks for pointing it out!
>
> (I think together with this change, I’ll also remove the
> event_notifier_set() call from virtio_blk_data_plane_start(). It’d
> obviously be a duplicate, and removing it shows why
> virtio_queue_aio_attach_host_notifier() should always kick the queue.)
Yes, it can be removed from start().
Stefan
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
next prev parent reply other threads:[~2024-01-29 13:21 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-24 17:38 [PATCH 0/2] virtio: Keep notifications disabled during drain Hanna Czenczek
2024-01-24 17:38 ` [PATCH 1/2] virtio-scsi: Attach event vq notifier with no_poll Hanna Czenczek
2024-01-24 22:00 ` Stefan Hajnoczi
2024-01-25 9:43 ` Fiona Ebner
2024-01-24 17:38 ` [PATCH 2/2] virtio: Keep notifications disabled during drain Hanna Czenczek
2024-01-25 9:43 ` Fiona Ebner
2024-01-25 18:03 ` Stefan Hajnoczi
2024-01-25 18:18 ` Hanna Czenczek
2024-01-25 18:32 ` Hanna Czenczek
2024-01-25 21:32 ` Stefan Hajnoczi [this message]
2024-01-25 18:05 ` [PATCH 0/2] " Stefan Hajnoczi
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=20240125213206.GB58542@fedora \
--to=stefanha@redhat.com \
--cc=f.ebner@proxmox.com \
--cc=fam@euphon.net \
--cc=hreitz@redhat.com \
--cc=kwolf@redhat.com \
--cc=mst@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-block@nongnu.org \
--cc=qemu-devel@nongnu.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).