Cool stuff. I've never looked at PCI IOMMU traces before.
I don't see anything obvious when looking at those traces though, so we're probably looking at some type of chipset bug here.
I guess we're down to trial and error then? I e, try random changes to stuff that win7 does but we don't (or the other way around), and see if it helps?
E g, try don't enable unsol before probing (win 7 does not do that), or don't write 0xff to STATETS, etc...and see if anything of that makes STATETS detect the codec.
Cool stuff. I've never looked at PCI IOMMU traces before.
I don't see anything obvious when looking at those traces though, so we're probably looking at some type of chipset bug here.
I guess we're down to trial and error then? I e, try random changes to stuff that win7 does but we don't (or the other way around), and see if it helps?
E g, try don't enable unsol before probing (win 7 does not do that), or don't write 0xff to STATETS, etc...and see if anything of that makes STATETS detect the codec.