On 14.05.2024 16:33, Stewart Hildebrand wrote:
> From: Oleksandr Andrushchenko <[email protected]>
> 
> Assign SBDF to the PCI devices being passed through with bus 0.
> The resulting topology is where PCIe devices reside on the bus 0 of the
> root complex itself (embedded endpoints).
> This implementation is limited to 32 devices which are allowed on
> a single PCI bus.
> 
> Please note, that at the moment only function 0 of a multifunction
> device can be passed through.
> 
> Signed-off-by: Oleksandr Andrushchenko <[email protected]>
> Signed-off-by: Volodymyr Babchuk <[email protected]>
> Signed-off-by: Stewart Hildebrand <[email protected]>

I'm largely okay, so
Acked-by: Jan Beulich <[email protected]>

Still my wariness remains towards the multi-function bit (it being set is
at least misleading when the device then has no further functions), and I
also remain not fully convinced by the earlier reply towards phantom
functions. While those indeed don't have any config space, is their setting
up in the IOMMU tables actually handled correctly for this virtual SBDF
case? It may well be (as the virtual coordinates _should_ indeed not matter
at all), but that's not entirely obvious.

Jan

Reply via email to