> -----Original Message-----
> From: Lorenzo Pieralisi <lorenzo.pieral...@arm.com>
> Sent: Monday, August 12, 2019 11:39 AM
> To: Haiyang Zhang <haiya...@microsoft.com>
> Cc: sas...@kernel.org; bhelg...@google.com; linux-
> hyp...@vger.kernel.org; linux-...@vger.kernel.org; KY Srinivasan
> <k...@microsoft.com>; Stephen Hemminger <sthem...@microsoft.com>;
> o...@aepfle.de; vkuznets <vkuzn...@redhat.com>; linux-
> ker...@vger.kernel.org
> Subject: Re: [PATCH v2] PCI: hv: Detect and fix Hyper-V PCI domain number
> collision
>
> On Tue, Aug 06, 2019 at 11:52:11PM +0000, Haiyang Zhang wrote:
> > Currently in Azure cloud, for passthrough devices including GPU, the
> > host sets the device instance ID's bytes 8 - 15 to a value derived from
> > the host HWID, which is the same on all devices in a VM. So, the device
> > instance ID's bytes 8 and 9 provided by the host are no longer unique.
> >
> > This can cause device passthrough to VMs to fail because the bytes 8 and
> > 9 is used as PCI domain number. So, as recommended by Azure host team,
> > we now use the bytes 4 and 5 which usually contain unique numbers as PCI
> > domain. The chance of collision is greatly reduced. In the rare cases of
> > collision, we will detect and find another number that is not in use.
>
> This is not clear at all. Why "finding another number" is fine with
> this patch while it is not with current kernel code ? Also does this
> have backward compatibility issues ?
The bytes 4, 5 have more uniqueness (info entropy) than bytes 8, 9, so we use
bytes 4, 5. On older hosts, bytes 4, 5 can also be used -- so it has no backward
compatibility issues.
> I do not understand if a collision is a problem or not from the
> log above.
Collision will cause the second device with the same domain number fails to
load.
I will include these info into the patch description.
>
> > Thanks to Michael Kelley <mikel...@microsoft.com> for proposing this
> idea.
>
> Add it as Suggested-by: tag.
I will add this line.
Thanks,
- Haiyang
