Re: [PATCH 2/4] x86/amd_nb: add support for newer PCI topologies

[Srinivas Pandruvada]

On Wed, 2018-11-07 at 15:30 -0800, Srinivas Pandruvada wrote:
> [...]
> 
> > Sure, you can't *force* OEMs to supply a given ACPI device, but you
> > can certainly say "if you want this functionality, supply INT3401
> > devices."  That's what you do with PNP0A03 (PCI host bridges), for
> > example.  If an OEM doesn't supply PNP0A03 devices, the system can
> > boot just fine as long as you don't need PCI.
> > 
> > This model of using the PCI IDs forces OS vendors to release
> > updates
> > for every new platform.  I guess you must have considered that and
> > decided whatever benefit you're getting was worth the cost.
> 
> Not worth cost. This is a pain. Every release we end up adding a
> single
> line change to many drivers adding a PCI device id. 
> Since there is no unique class_mask for PCI device for these devices,
> we need to add device_id for each generation even if there is no
> change.
> Instead if we have some feature to say don't enumerate for PCI device
> id < X and a black list, it will save lot of work.
This still needs some work on our internal PCI device allocation scheme
, where we can reserve a block of ids for a PCI device for same
functionality from generation to generation.

Thanks,
Srinivas