Re: [PATCH v2] PCI: Add ATS-disable quirk for AMD Stoney GPUs

[Bjorn Helgaas]

On Thu, Apr 20, 2017 at 02:11:42PM +0200, Joerg Roedel wrote:
> On Sat, Apr 08, 2017 at 09:41:07AM +0200, Lukas Wunner wrote:
> > On Fri, Apr 07, 2017 at 04:32:18PM +0200, Joerg Roedel wrote:
> > > From: Joerg Roedel <jroedel@xxxxxxx>
> > > 
> > > ATS is broken on this hardware and causes IOMMU stalls and
> > > system failure. Disable ATS on these devices to make them
> > > usable again with IOMMU enabled.
> > 
> > AMD Stoney Ridge is an x86 CPU + GPU combo and this quirk pertains
> > to the GPU, right?
> > 
> > In that case the quirk should go to arch/x86.  Paul Menzel (+cc)
> > has just complained on linux-pci@ that final fixups are taking half
> > a second, and I think that could be reduced if more efforts were
> > spent to move arch-specific quirks out of the catch-all in
> > drivers/pci/quirks.c.
> 
> The affected hardware here might be x86-only, but ATS is not. If a
> broken ATS-capable plug-in card appears, we need this in generic code
> anyway.

It could go in either arch/x86/pci/fixup.c or drivers/pci/quirks.c.

It's not clear to me exactly what the hardware defect is or where it
is.  If it's in the CPU or in a GPU that can only be found on x86, I
think arch/x86/pci/fixup.c is the appropriate place.

If it's in a GPU that could be found on other arches,
drivers/pci/quirks.c would be the appropriate place.

I don't personally think the possibility of a plugin card with broken
ATS is a real reason to put this quirk in drivers/pci/quirks.c.  It's
a trivial patch and easy to copy or move later if we need to.

Bjorn