Re: [Mesa-dev] [PATCH 30/31] nir/algebraic: Add some optimizations for D3D-style booleans

2018-10-23 Thread Jason Ekstrand
On Tue, Oct 23, 2018 at 1:38 AM Connor Abbott wrote: > On Tue, Oct 23, 2018 at 12:16 AM Jason Ekstrand > wrote: > > > > D3D booleans use a 32-bit 0/-1 representation. Because this previously > > matched NIR exactly, we didn't have to really optimize for it. Now that > > we have 1-bit

Re: [Mesa-dev] [PATCH 30/31] nir/algebraic: Add some optimizations for D3D-style booleans

2018-10-23 Thread Connor Abbott
On Tue, Oct 23, 2018 at 12:16 AM Jason Ekstrand wrote: > > D3D booleans use a 32-bit 0/-1 representation. Because this previously > matched NIR exactly, we didn't have to really optimize for it. Now that > we have 1-bit booleans, we need some specific optimizations to chew > through the

Re: [Mesa-dev] [PATCH 30/31] nir/algebraic: Add some optimizations for D3D-style booleans

2018-10-22 Thread Bas Nieuwenhuizen
On Tue, Oct 23, 2018 at 12:16 AM Jason Ekstrand wrote: > > D3D booleans use a 32-bit 0/-1 representation. Because this previously > matched NIR exactly, we didn't have to really optimize for it. Now that > we have 1-bit booleans, we need some specific optimizations to chew > through the

[Mesa-dev] [PATCH 30/31] nir/algebraic: Add some optimizations for D3D-style booleans

2018-10-22 Thread Jason Ekstrand
D3D booleans use a 32-bit 0/-1 representation. Because this previously matched NIR exactly, we didn't have to really optimize for it. Now that we have 1-bit booleans, we need some specific optimizations to chew through the D3D12-style booleans. --- src/compiler/nir/nir_opt_algebraic.py | 13