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

aco: optimize v_add_u32(v_mul_lo_u16) -> v_mad_u32_u16

Browse Source 
fossils-db (Vega10):
Totals from 779 (0.56% of 139517) affected shaders:
CodeSize: 1187928 -> 1187508 (-0.04%); split: -0.04%, +0.00%
Instrs: 247353 -> 244608 (-1.11%); split: -1.11%, +0.00%
Cycles: 1127472 -> 1116420 (-0.98%); split: -0.98%, +0.00%
VMEM: 139720 -> 138297 (-1.02%); split: +0.00%, -1.02%
SMEM: 51069 -> 50735 (-0.65%); split: +0.04%, -0.69%
Copies: 11548 -> 11547 (-0.01%); split: -0.03%, +0.03%

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7425>
This commit is contained in:
Samuel Pitoiset
2020-11-02 15:34:25 +01:00
committed by Marge Bot
parent 20e48551ac
commit db9d13b4ff
2 changed files with 32 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/amd/compiler/tests/test_optimizer.cpp
+24
View File
@@ -195,6 +195,30 @@ BEGIN_TEST(optimize.mad_u32_u16)
//! p_unit_test 5, %res5
writeout(5, create_mad_u32_u16(Operand(42u), Operand(inputs[0]), Operand(0u), false));
//~gfx9! v1: %mul6 = v_mul_lo_u16 %a, %b
//~gfx9! v1: %res6 = v_add_u32 %mul6, %b
//~gfx10! v1: %mul6 = v_mul_lo_u16_e64 %a, %b
//~gfx10! v1: %res6 = v_add_u32 %mul6, %b
//! p_unit_test 6, %res6
Temp mul;
if (i >= GFX10) {
mul = bld.vop3(aco_opcode::v_mul_lo_u16_e64, bld.def(v1), inputs[0], inputs[1]);
} else {
mul = bld.vop2(aco_opcode::v_mul_lo_u16, bld.def(v1), inputs[0], inputs[1]);
}
writeout(6, bld.vadd32(bld.def(v1), mul, inputs[1]));
//~gfx9! v1: %res7 = v_mad_u32_u16 %a, %b, %b
//~gfx10! v1: (nuw)%mul7 = v_mul_lo_u16_e64 %a, %b
//~gfx10! v1: %res7 = v_add_u32 %mul7, %b
//! p_unit_test 7, %res7
if (i >= GFX10) {
mul = bld.nuw().vop3(aco_opcode::v_mul_lo_u16_e64, bld.def(v1), inputs[0], inputs[1]);
} else {
mul = bld.nuw().vop2(aco_opcode::v_mul_lo_u16, bld.def(v1), inputs[0], inputs[1]);
}
writeout(7, bld.vadd32(bld.def(v1), mul, inputs[1]));
finish_opt_test();
}
END_TEST