diff options
| author | Rhys Perry <pendingchaos02@gmail.com> | 2022-10-26 21:11:31 +0100 |
|---|---|---|
| committer | Marge Bot <emma+marge@anholt.net> | 2022-11-01 12:42:43 +0000 |
| commit | 6113ee650a272dc737a200e276de474e083a2fdf (patch) | |
| tree | 1e2a309793c6662db91a431f2a659c7ad47a7039 /src/amd/compiler/aco_register_allocation.cpp | |
| parent | 16d2c7ad557b46104f91365ab3405f0a3ed7e36d (diff) | |
| download | mesa-6113ee650a272dc737a200e276de474e083a2fdf.tar.gz | |
aco/gfx11: fix FS input loads in quad-divergent control flow
This is not ideal and it would be great to somehow make it better some
day.
fossil-db (gfx1100):
Totals from 5208 (3.86% of 135032) affected shaders:
MaxWaves: 127058 -> 126962 (-0.08%); split: +0.01%, -0.09%
Instrs: 3983440 -> 4072736 (+2.24%); split: -0.00%, +2.24%
CodeSize: 21872468 -> 22230852 (+1.64%); split: -0.00%, +1.64%
VGPRs: 206688 -> 206984 (+0.14%); split: -0.05%, +0.20%
Latency: 37447383 -> 37491197 (+0.12%); split: -0.05%, +0.17%
InvThroughput: 6421955 -> 6422348 (+0.01%); split: -0.03%, +0.03%
VClause: 71579 -> 71545 (-0.05%); split: -0.09%, +0.04%
SClause: 148289 -> 147146 (-0.77%); split: -0.84%, +0.07%
Copies: 259011 -> 258084 (-0.36%); split: -0.61%, +0.25%
Branches: 101366 -> 101314 (-0.05%); split: -0.10%, +0.05%
PreSGPRs: 223482 -> 223460 (-0.01%); split: -0.21%, +0.20%
PreVGPRs: 184448 -> 184744 (+0.16%)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19370>
Diffstat (limited to 'src/amd/compiler/aco_register_allocation.cpp')
| -rw-r--r-- | src/amd/compiler/aco_register_allocation.cpp | 4 |
1 files changed, 3 insertions, 1 deletions
diff --git a/src/amd/compiler/aco_register_allocation.cpp b/src/amd/compiler/aco_register_allocation.cpp index 9cbc20810aa..0a3401ead49 100644 --- a/src/amd/compiler/aco_register_allocation.cpp +++ b/src/amd/compiler/aco_register_allocation.cpp @@ -607,7 +607,9 @@ get_subdword_definition_info(Program* program, const aco_ptr<Instruction>& instr amd_gfx_level gfx_level = program->gfx_level; if (instr->isPseudo()) { - if (gfx_level >= GFX8) + if (instr->opcode == aco_opcode::p_interp_gfx11) + return std::make_pair(4u, 4u); + else if (gfx_level >= GFX8) return std::make_pair(rc.bytes() % 2 == 0 ? 2 : 1, rc.bytes()); else return std::make_pair(4, rc.size() * 4u); |