fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2025-12-20 18:10:11 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 2

aco/isel: emit exec copy for ballot(true)

Browse source 
Once copy propagated in the optimizer, this will allow
using nir_opt_uniform_subgroup without too many regressions.

Foz-DB Navi48:
Totals from 405 (0.41% of 97637) affected shaders:
Instrs: 3796716 -> 3796894 (+0.00%); split: -0.00%, +0.00%
CodeSize: 20116136 -> 20116652 (+0.00%); split: -0.00%, +0.00%
Latency: 18326661 -> 18327114 (+0.00%); split: -0.00%, +0.00%
InvThroughput: 3353206 -> 3353268 (+0.00%); split: -0.00%, +0.00%
Copies: 292307 -> 293830 (+0.52%)
SALU: 507523 -> 507738 (+0.04%)

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38830>
This commit is contained in:
Georg Lehmann 2025-11-29 11:12:13 +01:00 • committed by Marge Bot
parent 72e3071751
commit 911e1ce168
1 changed files with 2 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
src/amd/compiler/instruction_selection/aco_select_nir_intrinsics.cpp
View file

@ -4210,6 +4210,8 @@ visit_intrinsic(isel_context* ctx, nir_intrinsic_instr* instr)
Definition def = dst.size() == bld.lm.size() ? Definition(dst) : bld.def(bld.lm); Definition def = dst.size() == bld.lm.size() ? Definition(dst) : bld.def(bld.lm);
if (instr->intrinsic == nir_intrinsic_ballot_relaxed) if (instr->intrinsic == nir_intrinsic_ballot_relaxed)
src = bld.copy(def, src); src = bld.copy(def, src);
else if (nir_src_is_const(instr->src[0]) && nir_src_as_uint(instr->src[0]))
src = bld.copy(def, Operand(exec, bld.lm));
else else
src = bld.sop2(Builder::s_and, def, bld.def(s1, scc), src, Operand(exec, bld.lm)); src = bld.sop2(Builder::s_and, def, bld.def(s1, scc), src, Operand(exec, bld.lm));
if (dst.size() != bld.lm.size()) { if (dst.size() != bld.lm.size()) {