qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Pankaj Gupta <pagupta@redhat.com>
To: Stefan Hajnoczi <stefanha@redhat.com>
Cc: mst@redhat.com, amit@kernel.org, qemu-devel@nongnu.org,
	slopezpa@redhat.com
Subject: Re: [Qemu-devel] [PATCH 2/2] virtio: Set status early in VirtIODevice parent object
Date: Wed, 27 Jun 2018 07:06:45 -0400 (EDT)	[thread overview]
Message-ID: <2067006616.46310203.1530097605117.JavaMail.zimbra@redhat.com> (raw)
In-Reply-To: <20180627102142.GB5955@stefanha-x1.localdomain>


> 
> On Wed, Jun 27, 2018 at 01:52:48PM +0530, Pankaj Gupta wrote:
> >  To process pending requests for driver status 'VIRTIO_CONFIG_S_DRIVER_OK',
> >  virtio-rng 'set_status' calls 'is_guest_ready' function. This checks if
> >  virtqueue is ready and status is set to 'VIRTIO_CONFIG_S_DRIVER_OK'.
> > 
> >  This call is made before new status is updated in VirtIODevice parent
> >  object
> >  and calling function uses old status value.
> > 
> > Signed-off-by: Pankaj Gupta <pagupta@redhat.com>
> > ---
> >  hw/virtio/virtio.c | 3 ++-
> >  1 file changed, 2 insertions(+), 1 deletion(-)
> > 
> > diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
> > index d4e4d98b59..37dc59a686 100644
> > --- a/hw/virtio/virtio.c
> > +++ b/hw/virtio/virtio.c
> > @@ -1154,10 +1154,11 @@ int virtio_set_status(VirtIODevice *vdev, uint8_t
> > val)
> >              }
> >          }
> >      }
> > +    vdev->status = val;
> > +
> >      if (k->set_status) {
> >          k->set_status(vdev, val);
> >      }
> > -    vdev->status = val;
> 
> This breaks existing ->set_status() callbacks that rely on vdev->status
> containing the old value.  Have you audited them?

I did not audit them all. I thought that's the way it should be base object
first get updated. But you are right, it can break things where old status is used.

> 
> For example, virtio_net_set_status -> virtio_net_vnet_endian_status uses
> vdev->status!
> 
> It may be easier to modify virtio-rng.c so that the old status value
> isn't used.

Just updating status only for virtio_rng_set_status looks okay? This will avoid
multiple copies of status as different functions are using it.

I will send a V2.

+static void virtio_rng_set_status(VirtIODevice *vdev, uint8_t status)
+{
+    VirtIORNG *vrng = VIRTIO_RNG(vdev);
+
+    if (!vdev->vm_running) {
+        return;
+    }
+    vdev->status = status;
+
+    /* Something changed, try to process buffers */
+    virtio_rng_process(vrng);
+}
+


Thanks,
Pankaj

> 
> Stefan
> 

      reply	other threads:[~2018-06-27 11:06 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-06-27  8:22 [Qemu-devel] [PATCH 0/2] virtio-rng: fix old guest kernel hang Pankaj Gupta
2018-06-27  8:22 ` [Qemu-devel] [PATCH 1/2] virtio-rng: process pending requests when driver is ready Pankaj Gupta
2018-06-27  8:22 ` [Qemu-devel] [PATCH 2/2] virtio: Set status early in VirtIODevice parent object Pankaj Gupta
2018-06-27 10:21   ` Stefan Hajnoczi
2018-06-27 11:06     ` Pankaj Gupta [this message]

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=2067006616.46310203.1530097605117.JavaMail.zimbra@redhat.com \
    --to=pagupta@redhat.com \
    --cc=amit@kernel.org \
    --cc=mst@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=slopezpa@redhat.com \
    --cc=stefanha@redhat.com \
    /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).