RE: making changes to agp code?

["Jan Beulich" <jbeulich@novell.com>]

>>> "Langsdorf, Mark" <mark.langsdorf@amd.com> 29.03.07 00:37 >>>
>> >> On a second look I believe the implementation is broken even 
>> >> on native, as long as !CONFIG_FLATMEM, since there's an
>> >> assumption that an invalid PFN cannot be followed by a valid
>> >> one. For that reason, I think the code needs to be changed to
>> >> call e820_any_mapped() (just like aperture.c does). I have a
>> >> tentative patch to do that, but don't have a working box with
>> >> an 8151.
>> >
>> >I do.  You can send it to me for testing.
>> 
>> Attached - depending on what tree you want to apply it on you
>> may have to tweak it a little.
>
>I applied it to xen-unstable with some tweaking (my version
>doesn't seem to have an i386 e820-xen.c ??) and to 2.6.20.

Yes, i386's e820.c is a 2.6.20 addition, which I followed in our Xen
port. But I assume the patch is not likely to be applied to -unstable
anyway, because of it touching a few files not in the sparse tree
(unless Keir feels otherwise, or if your iommu stuff will have a strict
dependency on this). The primary intention is to post the non-Xen
parts to kernel.org and pick up when it got merged.

>System booted correctly and ran fine.
>
>Acked-by: Mark Langsdorf <mark.langsdorf@amd.com>

Thanks!

Jan