[Mesa-stable] Request for update of commits proposed for stable branch

[Marek Olšák]

Sounds good to me.

Marek

On Tue, Jul 30, 2013 at 12:51 AM, Deucher, Alexander
<Alexander.Deucher at amd.com> wrote:
>> -----Original Message-----
>> From: Carl Worth [mailto:cworth at cworth.org]
>> Sent: Monday, July 29, 2013 6:29 PM
>> To: Deucher, Alexander
>> Cc: Stellard, Thomas; Marek Olšák; mesa-stable at lists.freedesktop.org
>> Subject: RE: Request for update of commits proposed for stable branch
>>
>> "Deucher, Alexander" <Alexander.Deucher at amd.com> writes:
>> > I'll take a look this week.  Thanks!
>>
>> Any update here, Alex?
>
> Sorry, you released the last update before I had had a chance to look at it and I figured it was too late at that point.
>
>>
>> >> commit 49c1fc7044eaaa5c2dca05ff4a709be8e3636871
>> >> Author: Alex Deucher <alexander.deucher at amd.com>
>> >> Date:   Tue Mar 19 18:11:20 2013 -0400
>> >>
>> >>     r600g: don't emit SQ_DYN_GPR_RESOURCE_LIMIT_1 on cayman
>>
>> I looked a bit at this one, and I couldn't find any mention of
>> SQ_DYN_GPR_RESOURCE_LIMIT_1 in the function of interest (on the 9.1
>> branch). So I'm guessing there's nothing to be done here, (we can't
>> remove code that doesn't exist).
>
> Yeah, that one can be dropped.
>
>>
>> >> commit 4539f8e20af286d1f521eb016c89c6d9af0b801c
>> >> Author: Alex Deucher <alexander.deucher at amd.com>
>> >> Date:   Fri May 3 09:56:31 2013 -0400
>> >>
>> >>     r600g: don't emit surface_sync after FLUSH_AND_INV_EVENT
>>
>> This one is harder for me to trivially reject.
>>
>> There's code in the 9.1 branch that's similar-ish to the code being
>> modified by the patch, but it's clearly different, so it's not obvious
>> to me what should be done.
>>
>> I'm going to skip both of these commits for the imminent 9.1.6 release,
>> and unless you advocate specifically, I won't look at these again for
>> future 9.1 releases.
>
> You can skip this one as well.  The flush code has changed too much in the master branch so we'd need to pull a number of changes back.  With 9.2 imminent, we can just leave the old flush code in 9.1 unless Tom or Marek have any objections.
>
> Alex
>
>