On Tue, 13 Sep 2016 09:08:04 +0200
Maxime Coquelin <address@hidden> wrote:
On 09/12/2016 09:58 PM, Michael S. Tsirkin wrote:
On Mon, Sep 12, 2016 at 08:22:50PM +0200, Maxime Coquelin wrote:
On 09/12/2016 10:51 AM, Cornelia Huck wrote:
On Sat, 10 Sep 2016 10:23:37 +0200
Maxime Coquelin <address@hidden> wrote:
Currently, devices are plugged before features are negotiated.
If the backend doesn't support VIRTIO_F_VERSION_1, the transport
need to rewind some settings.
This is the case for CCW, for which a post_plugged callback had
been introduced, where max_rev field is just updated if
VIRTIO_F_VERSION_1 is not supported by the backend.
For PCI, implementing the post_plugged would be much more
s/the//
complicated, so it needs to know whether the backend supports
VIRTIO_F_VERSION_1 at plug time.
Currently, nothing is done for PCI. Modern capabilitities get
exposed to the guest even if VIRTIO_F_VERSION_1 is not supported
by the backend, which confuses the guest.
This patch proposes to replace existing post_plugged solution
Nit: The patch does not propose anything, it just does it :)
Michael,
Should I send a v2 fixing the above comments, or you can handle them
when applying the patch?
Thanks,
Maxime
It's easier if you post v2 including all acks.
Ok, v2 is ready, waiting to collect some acks.
In the meanwhile, I've verified that everything works as expected as
well with an old host kernel (3.18) and your patch for ccw. Therefore,
Tested-by: Cornelia Huck <address@hidden> [ccw]
Reviewed-by: Cornelia Huck <address@hidden>