On Wed, Aug 25, 2021 at 09:48:19AM -0600, Alex Williamson wrote:
> On Wed, 25 Aug 2021 14:43:03 +0200
> Christoph Hellwig <hch@xxxxxx> wrote:
>
> > On Tue, Aug 24, 2021 at 02:25:08PM -0600, Alex Williamson wrote:
> > > I think this turns into the patch below on top of Yishai's
> > > vfio-pci-core series. Please verify. If you'd like something
> > > different, please post an update. Thanks,
> >
> > The change looks fine to me. Does that mean you want me to rebase
> > on top of the above series?
>
> I wish the answer here was more obvious, but we're still waiting for
> PCI and scripts/mod/ acks for that series. I note however that while
> the functionality of other driver's having a userspace driver discovery
> mechanism hinges on patches 9 & 10 of Yishai's series, those patches
> can also be cleanly moved to the end of the series, or a follow-on
> series if necessary, and current vfio-pci binding continues to work.
>
> Should we go ahead and get final reviews for {1-8,11-13} of Yishai's
> series to get them in my next branch so we have a consistent base to
> work from? Thanks,
I see Bjorn has just now ack'd the PCI parts so I think it is also
fine to go ahead. A subsystem is ultimately responsible to define the
modalias format under its bus, the file2alias changes are supposed to
track it.
Also I'm looking at past changes to file2alias and I don't really see
many acks..
Thanks,
Jason
