AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

aco: optimize v_and(a, v_subbrev_co(0, 0, vcc)) -> v_cndmask(0, a, vcc)

Browse Source 
fossils-db (Vega10):
Totals from 7786 (5.70% of 136546) affected shaders:
SGPRs: 517778 -> 518626 (+0.16%); split: -0.01%, +0.17%
VGPRs: 488252 -> 488084 (-0.03%); split: -0.04%, +0.01%
CodeSize: 42282068 -> 42250152 (-0.08%); split: -0.16%, +0.09%
MaxWaves: 35697 -> 35716 (+0.05%); split: +0.06%, -0.01%
Instrs: 8319309 -> 8304792 (-0.17%); split: -0.18%, +0.00%
Cycles: 88619440 -> 88489636 (-0.15%); split: -0.16%, +0.01%
VMEM: 2788278 -> 2780431 (-0.28%); split: +0.06%, -0.35%
SMEM: 570364 -> 569370 (-0.17%); split: +0.12%, -0.30%
VClause: 144906 -> 144908 (+0.00%); split: -0.05%, +0.05%
SClause: 302143 -> 302055 (-0.03%); split: -0.04%, +0.01%
Copies: 579124 -> 578779 (-0.06%); split: -0.14%, +0.08%
PreSGPRs: 327695 -> 328845 (+0.35%); split: -0.00%, +0.35%
PreVGPRs: 434280 -> 433954 (-0.08%)

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7438>
This commit is contained in:
Samuel Pitoiset
2020-11-03 18:50:32 +01:00
committed by Marge Bot
parent 2bbe01b186
commit bae5487659
2 changed files with 87 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/amd/compiler/tests/test_optimizer.cpp
+42
View File
@@ -80,3 +80,45 @@ BEGIN_TEST(optimize.neg)
finish_opt_test();
}
END_TEST
Temp create_subbrev_co(Operand op0, Operand op1, Operand op2)
{
return bld.vop2_e64(aco_opcode::v_subbrev_co_u32, bld.def(v1), bld.hint_vcc(bld.def(bld.lm)), op0, op1, op2);
}
BEGIN_TEST(optimize.cndmask)
for (unsigned i = GFX9; i <= GFX10; i++) {
//>> v1: %a, s1: %b, s2: %c, s2: %_:exec = p_startpgm
if (!setup_cs("v1 s1 s2", (chip_class)i))
continue;
Temp subbrev;
//! v1: %res0 = v_cndmask_b32 0, %a, %c
//! p_unit_test 0, %res0
subbrev = create_subbrev_co(Operand(0u), Operand(0u), Operand(inputs[2]));
writeout(0, bld.vop2(aco_opcode::v_and_b32, bld.def(v1), inputs[0], subbrev));
//! v1: %res1 = v_cndmask_b32 0, 42, %c
//! p_unit_test 1, %res1
subbrev = create_subbrev_co(Operand(0u), Operand(0u), Operand(inputs[2]));
writeout(1, bld.vop2(aco_opcode::v_and_b32, bld.def(v1), Operand(42u), subbrev));
//~gfx9! v1: %subbrev, s2: %_ = v_subbrev_co_u32 0, 0, %c
//~gfx9! v1: %res2 = v_and_b32 %b, %subbrev
//~gfx10! v1: %res2 = v_cndmask_b32 0, %b, %c
//! p_unit_test 2, %res2
subbrev = create_subbrev_co(Operand(0u), Operand(0u), Operand(inputs[2]));
writeout(2, bld.vop2(aco_opcode::v_and_b32, bld.def(v1), inputs[1], subbrev));
//! v1: %subbrev1, s2: %_ = v_subbrev_co_u32 0, 0, %c
//! v1: %xor = v_xor_b32 %a, %subbrev1
//! v1: %res3 = v_cndmask_b32 0, %xor, %c
//! p_unit_test 3, %res3
subbrev = create_subbrev_co(Operand(0u), Operand(0u), Operand(inputs[2]));
Temp xor_a = bld.vop2(aco_opcode::v_xor_b32, bld.def(v1), inputs[0], subbrev);
writeout(3, bld.vop2(aco_opcode::v_and_b32, bld.def(v1), xor_a, subbrev));
finish_opt_test();
}
END_TEST