Re: [Mesa-dev] [PATCH 05/23] nir: Introduce new nir_intrinsic_load_per_vertex_input intrinsics.

2015-10-01 Thread Kenneth Graunke
On Wednesday, September 30, 2015 12:58:09 AM Kenneth Graunke wrote: > Discuss :) NAK on patches 3-5. It turns out they break if the outermost array index is non-constant. I've got two new patches that replace them. signature.asc Description: This is a digitally signed message part. ___

Re: [Mesa-dev] [PATCH 05/23] nir: Introduce new nir_intrinsic_load_per_vertex_input intrinsics.

2015-09-30 Thread Jason Ekstrand
On Wed, Sep 30, 2015 at 12:58 AM, Kenneth Graunke wrote: > Geometry and tessellation shaders process multiple vertices; their > inputs are arrays indexed by the vertex number. While GLSL makes > this look like a normal array, it can be very different behind the > scenes. > > On Intel hardware, al

[Mesa-dev] [PATCH 05/23] nir: Introduce new nir_intrinsic_load_per_vertex_input intrinsics.

2015-09-30 Thread Kenneth Graunke
Geometry and tessellation shaders process multiple vertices; their inputs are arrays indexed by the vertex number. While GLSL makes this look like a normal array, it can be very different behind the scenes. On Intel hardware, all inputs for a particular vertex are stored together - as if they wer