dsg/mesa
2
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

aco: skip fract for sin/cos on gfx6-8 if the src is already in range

Browse source 
Foz-DB Polaris10:
Totals from 1301 (1.86% of 69950) affected shaders:
Instrs: 1447217 -> 1445610 (-0.11%); split: -0.11%, +0.00%
CodeSize: 7775988 -> 7769588 (-0.08%); split: -0.08%, +0.00%
SGPRs: 101712 -> 101776 (+0.06%)
SpillSGPRs: 931 -> 927 (-0.43%)
Latency: 16119433 -> 16115293 (-0.03%); split: -0.03%, +0.01%
InvThroughput: 9605952 -> 9577042 (-0.30%); split: -0.31%, +0.01%
VClause: 24591 -> 24593 (+0.01%); split: -0.01%, +0.02%
SClause: 29656 -> 29655 (-0.00%)
Copies: 133968 -> 134001 (+0.02%); split: -0.01%, +0.03%
VALU: 1157855 -> 1156235 (-0.14%)
SALU: 124626 -> 124639 (+0.01%); split: -0.00%, +0.01%

Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/40545>
This commit is contained in:
Georg Lehmann 2026-03-20 17:58:23 +01:00 • committed by Marge Bot
parent 57e2b272d5
commit 559a35dcb3
1 changed files with 6 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

8
src/amd/compiler/instruction_selection/aco_select_nir_alu.cpp
View file

@ -2529,9 +2529,13 @@ visit_alu_instr(isel_context* ctx, nir_alu_instr* instr)
}
Temp src = get_alu_src(ctx, instr->src[0]);
/* before GFX9, v_sin and v_cos had a valid input domain of [-256, +256] */
if (ctx->options->gfx_level < GFX9)
src = bld.vop1(fract, bld.def(rc), src);
if (ctx->options->gfx_level < GFX9) {
fp_class_mask fp_class = nir_analyze_fp_class(&ctx->fp_class_ht, instr->src[0].src.ssa);
if (fp_class & (FP_CLASS_ANY_INF | FP_CLASS_LT_NEG_ONE | FP_CLASS_GT_POS_ONE))
src = bld.vop1(fract, bld.def(rc), src);
}
if (dst.regClass() == rc) {
bld.vop1(opcode, Definition(dst), src);