mesa/src/amd/llvm
Samuel Pitoiset 9f005f1f85 radv: enable lowering of GS intrinsics for the LLVM backend
This replaces emit_vertex with:
   if (vertex_count < max_vertices) {
      emit_vertex_with_counter vertex_count ...
      vertex_count += 1
   }

Which is exactly what NIR->LLVM was doing but at NIR level. This
pass is already called by ACO.

pipeline-db changes on GFX10:
Totals from affected shaders:
SGPRS: 1952 -> 1912 (-2.05 %)
VGPRS: 2112 -> 2044 (-3.22 %)
Code Size: 189368 -> 185620 (-1.98 %) bytes
Max Waves: 494 -> 491 (-0.61 %)

No pipeline-db changes on other generations.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4182>
2020-04-08 08:24:05 +02:00
..
ac_llvm_build.c ac/nir: use llvm.amdgcn.rcp in ac_build_fdiv() 2020-03-27 08:05:43 +01:00
ac_llvm_build.h ac/nir: use llvm.amdgcn.rcp in ac_build_fdiv() 2020-03-27 08:05:43 +01:00
ac_llvm_cull.c ac/cull: don't read Position.Z if it's not needed for culling 2020-01-15 15:06:20 -05:00
ac_llvm_cull.h amd: Move all amd/common code that depends on LLVM to amd/llvm. 2019-10-08 00:44:08 +00:00
ac_llvm_helper.cpp radv: remove useless RADV_DEBUG=unsafemath debug option 2019-11-15 09:07:34 +01:00
ac_llvm_util.c ac: don't set old denormals flags with LLVM >= 11 2020-03-17 20:47:48 +00:00
ac_llvm_util.h radeonsi: remove AMD_DEBUG=sisched option 2020-03-06 11:35:12 +01:00
ac_nir_to_llvm.c radv: enable lowering of GS intrinsics for the LLVM backend 2020-04-08 08:24:05 +02:00
ac_nir_to_llvm.h radeonsi/nir: don't run si_nir_opts again if there is no change 2019-11-25 16:48:27 -05:00
ac_shader_abi.h radv: enable lowering of GS intrinsics for the LLVM backend 2020-04-08 08:24:05 +02:00
meson.build meson: inline inc_common 2020-03-28 21:36:54 +01:00