Re: 5.13 i915/PAT regression on Brasswell, adding nopat to the kernel commandline worksaround this

[Hans de Goede]

Hi,

On 5/12/21 1:57 PM, Christoph Hellwig wrote:
> On Wed, May 12, 2021 at 01:15:03PM +0200, Peter Zijlstra wrote:
>> IIRC it's because of 74ffa5a3e685 ("mm: add remap_pfn_range_notrack"),
>> which added a sanity check to make sure expectations were met. It turns
>> out they were not.
>>
>> The bug is not new, the warning is. AFAIK the i915 team is aware, but
>> other than that I've not followed.
> 
> 
> The actual culprit is b12d691ea5e0 ("i915: fix remap_io_sg to verify the
> pgprot"), but otherwise agreed.  Someone the i915 maintainers all seem
> to be on vacation as the previous report did not manage to trigger any
> kind of reply.

I can confirm that reverting that commit restores i915 functionality with
5.13-rc1 on the Braswell machine on which I have been testing this.

Regards,

Hans