[Mesa-dev] [PATCH 0/2] i965/hsw: Set MOCS for surfaces

[Matt Turner]

On Sun, May 5, 2013 at 8:53 PM, Chad Versace
<chad.versace at linux.intel.com> wrote:
> On 05/04/2013 03:20 PM, Matt Turner wrote:
>>
>> On Sat, May 4, 2013 at 12:14 AM, Matt Turner <mattst88 at gmail.com> wrote:
>>>
>>> On Fri, May 3, 2013 at 10:42 PM, Chad Versace
>>> <chad.versace at linux.intel.com> wrote:
>>>>
>>>> +32% on GLBenchmark 2.5 Egypt Offscreen DXT1. Tested on Harris Beach GT3
>>>> with
>>>> Android.
>>>>
>>>> Please please please, someone run Piglit on this series.  I currently
>>>> have no
>>>> Haswell hardware that is stable on non-Android Linux, and so can't run
>>>> Piglit
>>>> myself.  I don't want to commit this series until it's been regression
>>>> tested.
>>>
>>>
>>> I just tried on a GT2 Haswell and I have hangs in the baseline piglit
>>> run. I could try on the Haswell on my desk at the office tomorrow if
>>> needed. Let me know if you'd like me to try.
>>
>>
>> No luck with that one either. Same result. :(
>
>
> Thanks for running Piglit. And double thanks for taking a weekend office
> trip.
>
> Boo hangs, boo. Did you determine which test hangs? I'll borrow a Haswell
> tomorrow and take a look at Piglit.

No, it didn't seem to be any particular test. It would typically
happen 10 or so tests into a piglit run, but didn't seem to correspond
to anything specifically. I also noticed that loading gdm causes a GPU
reset, so I had to start gnome-terminal with an xinitrc.

I'll investigate more tomorrow, but I wonder if a recent change is to
blame. I don't remember having any trouble like this before.