Re: [Xen-devel] Notes for upcoming PCI emulation call

[Roger Pau Monné <roger.pau@xxxxxxxxxx>]

On Wed, May 02, 2018 at 08:03:52PM +1000, Alexey G wrote:
> - Emulating (a simplest) upstream PCIe hierarchy for passed through PCIe
> devices. The issue was described in details here:
> http://lists.gnu.org/archive/html/qemu-devel/2018-03/msg03593.html
> 
> Latter problem must be resolved properly by introducing emulated PCIe
> Root Ports for passed through devices.

In order to solve it properly bridges are required. But as you point
out in your email, it can be worked around. Since there's already
quite a lot of stuff to discuss, and AFAICT bridge support is not
mandatory for an initial implementation I would suggest that the
bridge discussion is deferred until MCFG support for bus 0 is
implemented.

> While this approach is beneficial in many ways, it will affect
> compatibility with QEMU very, very badly. For example, NVDIMM support
> patches from Intel rely on QEMU ACPI facilities which can become
> completely inoperational due to removing emulated NB+SB and their
> corresponding subtypes and properties.

There are multiple issues with NVDIMM on Xen, and we are still
discussing how NVDIMM devices should be handled on Dom0, I think at
this stage ACPI is a quite minor issue compared to the other
challenges presented by NVDIMM.

And in any case in order for vNVDIMM to support PVH guests the QEMU
code is certainly not going to be used on Xen.

Thanks, Roger.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel