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

aco: allow overflow for some SMEM instructions

Browse Source 
fossil-db (Navi):
Totals from 10184 (7.49% of 135946) affected shaders:
CodeSize: 83419748 -> 82430824 (-1.19%); split: -1.19%, +0.01%
Instrs: 16054612 -> 15908523 (-0.91%); split: -0.93%, +0.02%
VMEM: 1608018 -> 1581829 (-1.63%); split: +0.20%, -1.83%
SMEM: 577031 -> 563492 (-2.35%); split: +0.10%, -2.45%
VClause: 242643 -> 242512 (-0.05%); split: -0.06%, +0.00%
SClause: 640966 -> 569897 (-11.09%)

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/2720>
This commit is contained in:
Rhys Perry
2020-07-17 15:01:41 +01:00
committed by Marge Bot
parent d169f09e37
commit 3a4847179b
4 changed files with 9 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/amd/compiler/aco_optimizer.cpp
+2 -1
View File
@@ -1001,13 +1001,14 @@ void label_instruction(opt_ctx &ctx, Block& block, aco_ptr<Instruction>& instr)
SMEM_instruction *smem = static_cast<SMEM_instruction *>(instr.get());
Temp base;
uint32_t offset;
bool prevent_overflow = smem->operands[0].size() > 2 || smem->prevent_overflow;
if (i == 1 && info.is_constant_or_literal(32) &&
((ctx.program->chip_class == GFX6 && info.val <= 0x3FF) ||
(ctx.program->chip_class == GFX7 && info.val <= 0xFFFFFFFF) ||
(ctx.program->chip_class >= GFX8 && info.val <= 0xFFFFF))) {
instr->operands[i] = Operand(info.val);
continue;
} else if (i == 1 && parse_base_offset(ctx, instr.get(), i, &base, &offset, true) && base.regClass() == s1 && offset <= 0xFFFFF && ctx.program->chip_class >= GFX9) {
} else if (i == 1 && parse_base_offset(ctx, instr.get(), i, &base, &offset, prevent_overflow) && base.regClass() == s1 && offset <= 0xFFFFF && ctx.program->chip_class >= GFX9) {
bool soe = smem->operands.size() >= (!smem->definitions.empty() ? 3 : 4);
if (soe &&
(!ctx.info[smem->operands.back().tempId()].is_constant_or_literal(32) ||