Re: [Mesa-dev] [PATCH] ac/nir: Only clamp shadow reference on radeonsi.

2017-11-02 Thread Nicolai Hähnle
On 23.10.2017 03:32, Bas Nieuwenhuizen wrote: Vulkan CTS does not expect the value to be clamped (at least for D32), and it makes a differences even though depth is in [0,1], due to strict inequalities. I couldn't find anything in the Vulkan spec about this, but the test seemed to be copied from

Re: [Mesa-dev] [PATCH] ac/nir: Only clamp shadow reference on radeonsi.

2017-10-22 Thread Dave Airlie
On 23 October 2017 at 11:32, Bas Nieuwenhuizen wrote: > Vulkan CTS does not expect the value to be clamped (at least for D32), > and it makes a differences even though depth is in [0,1], due > to strict inequalities. > > I couldn't find anything in the Vulkan spec about this, but the test > seemed

[Mesa-dev] [PATCH] ac/nir: Only clamp shadow reference on radeonsi.

2017-10-22 Thread Bas Nieuwenhuizen
Vulkan CTS does not expect the value to be clamped (at least for D32), and it makes a differences even though depth is in [0,1], due to strict inequalities. I couldn't find anything in the Vulkan spec about this, but the test seemed to be copied from GL tests and the GL spec only specifies clampin