Re: [PATCH RFC] virtio-pci: new config layout: using memory BAR

[Rusty Russell <rusty@rustcorp.com.au>]

"Michael S. Tsirkin" <mst@redhat.com> writes:
>> By my count, net still has 7 feature bits left, so I don't think the
>> feature bits are likely to be a limitation in the next 6 months?
>
> Yes but you wanted a generic transport feature bit
> for flexible SG layout.
> Are you happy with VIRTIO_NET_F_ANY_HEADER_SG then?

If we need it within 6 months, I'd rather do that: this feature would
then be assumed for 1.0, and reserved.  We may do that for other
features, too (the committee will have to review).

>> MMIO is a bigger problem.  Linux guests are happy with it: does it break
>> the Windows drivers?
>> 
>> Thanks,
>> Rusty.
>
> You mean make BAR0 an MMIO BAR?
> Yes, it would break current windows guests.
> Further, as long as we use same address to notify all queues,
> we would also need to decode the instruction on x86 and that's
> measureably slower than PIO.
> We could go back to discussing hypercall use for notifications,
> but that has its own set of issues...

We might have something ready to deploy in 3 months, but realistically,
I'd rather have it ready and tested outside the main git tree(s) and
push it once the standard is committed.

Cheers,
Rusty.