[Mesa-dev] [PATCH] radv: remove sisched hack for talos

Timothy Arceri tarceri at itsqueeze.com
Fri Mar 15 22:47:47 UTC 2019


On 16/3/19 6:53 am, Dieter Nützel wrote:
> Am 15.03.2019 15:20, schrieb Samuel Pitoiset:
>> Results of my benchmarks are:
>>
>> 3 runs at 1080p:
>>
>> GFX8: -1%
>>
>> GFX9: -1.12%
>>
>> 3 runs at 4k:
>>
>> GFX8: -2%
>>
>> GFX9: -1.85%
>>
>> I'm actually not sure if we want to remove it...
> 
> Yes, my hint is we should wait until Marek is back from vacation.
> Running all the time with AMD_DEBUB (R600_DEBUG)=nir,sisched 'cause it's
> worth it...

Can you point to specific games that are helped?

> 
> Dieter
> 
>> On 3/15/19 11:25 AM, Timothy Arceri wrote:
>>> This was added in 8a7d4092d260 but no longer seems to have any
>>> impact on performance.
>>> ---
>>>   src/amd/vulkan/radv_device.c | 10 +---------
>>>   1 file changed, 1 insertion(+), 9 deletions(-)
>>>
>>> diff --git a/src/amd/vulkan/radv_device.c b/src/amd/vulkan/radv_device.c
>>> index 9570c15af02..56421dbc74b 100644
>>> --- a/src/amd/vulkan/radv_device.c
>>> +++ b/src/amd/vulkan/radv_device.c
>>> @@ -499,15 +499,7 @@ radv_handle_per_app_options(struct radv_instance 
>>> *instance,
>>>       if (!name)
>>>           return;
>>>   -    if (!strcmp(name, "Talos - Linux - 32bit") ||
>>> -        !strcmp(name, "Talos - Linux - 64bit")) {
>>> -        if (!(instance->debug_flags & RADV_DEBUG_NO_SISCHED)) {
>>> -            /* Force enable LLVM sisched for Talos because it looks
>>> -             * safe and it gives few more FPS.
>>> -             */
>>> -            instance->perftest_flags |= RADV_PERFTEST_SISCHED;
>>> -        }
>>> -    } else if (!strcmp(name, "DOOM_VFR")) {
>>> +    if (!strcmp(name, "DOOM_VFR")) {
>>>           /* Work around a Doom VFR game bug */
>>>           instance->debug_flags |= RADV_DEBUG_NO_DYNAMIC_BOUNDS;
>>>       }
>> _______________________________________________
>> mesa-dev mailing list
>> mesa-dev at lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/mesa-dev


More information about the mesa-dev mailing list