Re: [Qemu-devel] Headsup: windows virtio networking does not work on current git

[Anthony Liguori <aliguori@us.ibm.com>]

Michael Tokarev <mjt@tls.msk.ru> writes:

> 02.02.2013 00:36, Anthony Liguori wrote:
>> Michael Tokarev <mjt@tls.msk.ru> writes:
>>
>>> 02.02.2013 00:18, Michael Tokarev wrote:
>>>> Just a heads-up for now, no real diagnostics or anything like that.
>>>>
>>>> Current git master (a9c87c586ba9ee290792a98dc126b2861b7f8b03), when booted
>>>> a windows guest, results in no virtio-net inside.  Neither winXP nor Win7,
>>>> neither older nor latest (22 Jan 2013) virtio-net drivers works.
>>>>
>>>> Windows displays a yellow exclamation mark near the virtio-net device and
>>>> says it can't start the device (Code 10).
>>>>
>>>> Linux guests work fine, quick test anyway.
>>>>
>>>> Cc'ing Jason since his virtio-net changes was last.  But I repeat: no
>>>> diagnostics as of yet, no bisection.
>>>
>>> Bisection was easy, since win works fine right before the multiqueue
>>> virtio-net series.  This is the first bad commit:
>>
>> Adding Vadim and Michael.
>>
>> If you use -M pc-1.3 or explicitly disable multiqueue, does the driver work?
>
> Neither one of these nor both makes any visible difference.
> Neither does -M pc-1.1 (just in case).

Hrm, then it's very likely not a driver problem.  Thanks.

Regards,

Anthony Liguori

>
>> Regards,
>>
>> Anthony Liguori
>>
>>>
>>> commit fed699f9ca6ae8a0fb62803334cf46fa64d1eb91
>>> Author: Jason Wang <jasowang@redhat.com>
>>> Date:   Wed Jan 30 19:12:39 2013 +0800
>>>
>>>       virtio-net: multiqueue support
>>>
>>>       This patch implements both userspace and vhost support for multiple queue
>>>       virtio-net (VIRTIO_NET_F_MQ). This is done by introducing an array of
>>>       VirtIONetQueue to VirtIONet.
>>>
>>>       Signed-off-by: Jason Wang <jasowang@redhat.com>
>>>       Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
>>>
>>> After this commit, win guest (winXP and win7) shows yellow
>>> exclamation sign and is unable to start the device with
>>> code 10.
>>>
>>> FWIW.  I'm not sure it is a good idea to make a release with
>>> such a breakage, even rc0.
>>>
>>> Thanks,
>>>
>>> /mjt