Re: [PATCH for-9.2 v12 00/11] hw/pci: SR-IOV related fixes and improvements

["Michael S. Tsirkin" <mst@redhat.com>]

On Mon, Aug 05, 2024 at 04:10:02PM +0900, Akihiko Odaki wrote:
> On 2024/08/05 15:39, Michael S. Tsirkin wrote:
> > On Sun, Aug 04, 2024 at 06:01:36PM +0900, Akihiko Odaki wrote:
> > > Supersedes: <20240714-rombar-v2-0-af1504ef55de@daynix.com>
> > > ("[PATCH v2 0/4] hw/pci: Convert rom_bar into OnOffAuto")
> > > 
> > > I submitted a RFC series[1] to add support for SR-IOV emulation to
> > > virtio-net-pci. During the development of the series, I fixed some
> > > trivial bugs and made improvements that I think are independently
> > > useful. This series extracts those fixes and improvements from the RFC
> > > series.
> > > 
> > > [1]: https://patchew.org/QEMU/20231210-sriov-v2-0-b959e8a6dfaf@daynix.com/
> > > 
> > > Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
> > > ---
> > > Changes in v12:
> > > - Changed to ignore invalid PCI_SRIOV_NUM_VF writes as done for
> > >    PCI_SRIOV_CTRL_VFE.
> > > - Updated the message for patch "hw/pci: Use -1 as the default value for
> > >    rombar". (Markus Armbruster)
> > > - Link to v11: https://lore.kernel.org/r/20240802-reuse-v11-0-fb83bb8c19fb@daynix.com
> > 
> > The igb issue is still with us, is it not?
> 
> Yes, if you are talking about the problem with s390x/libvirt. That is why it
> has for-9.2 tag.
> 
> It is actually not specific to igb but also affects nvme. The upcoming
> virtio-net-pci's SR-IOV will not be affected, but it is not present yet.

How about we fix it then? There's time enough for 9.2.
Also breaking any configurations should be documented in a commit log
of the change that causes the breakage. Pls do that in the future.

I understand the fact that you are trying to prevent errors
since it is hard to report them to guest cleanly.
But I guess we can pretend device does not respond to config cycles?


-- 
MST