mirrors/dynarmic
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

emit_x64_vector_floating_point: Simplify FPVector{Min,Max}

Browse source
This commit is contained in:
MerryMage 2018-08-03 11:46:55 +01:00
parent 4a1ce797cb
commit 1ef388d1cd
1 changed files with 42 additions and 101 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

143
src/backend_x64/emit_x64_vector_floating_point.cpp
View file

@ -566,15 +566,15 @@ void EmitX64::EmitFPVectorGreaterEqual64(EmitContext& ctx, IR::Inst* inst) {
ctx.reg_alloc.DefineValue(inst, b); ctx.reg_alloc.DefineValue(inst, b);
} }
template<size_t fsize> template<size_t fsize, bool is_max>
static void EmitFPVectorMax(BlockOfCode& code, EmitContext& ctx, IR::Inst* inst) { static void EmitFPVectorMinMax(BlockOfCode& code, EmitContext& ctx, IR::Inst* inst) {
if (ctx.FPSCR_DN()) { if (ctx.FPSCR_DN()) {
auto args = ctx.reg_alloc.GetArgumentInfo(inst); auto args = ctx.reg_alloc.GetArgumentInfo(inst);
const Xbyak::Xmm result = ctx.reg_alloc.UseScratchXmm(args[0]); const Xbyak::Xmm result = ctx.reg_alloc.UseScratchXmm(args[0]);
const Xbyak::Xmm xmm_b = ctx.FPSCR_FTZ() ? ctx.reg_alloc.UseScratchXmm(args[1]) : ctx.reg_alloc.UseXmm(args[1]); const Xbyak::Xmm xmm_b = ctx.FPSCR_FTZ() ? ctx.reg_alloc.UseScratchXmm(args[1]) : ctx.reg_alloc.UseXmm(args[1]);
const Xbyak::Xmm mask = xmm0; const Xbyak::Xmm mask = xmm0;
const Xbyak::Xmm anded = ctx.reg_alloc.ScratchXmm(); const Xbyak::Xmm eq = ctx.reg_alloc.ScratchXmm();
const Xbyak::Xmm nan_mask = ctx.reg_alloc.ScratchXmm(); const Xbyak::Xmm nan_mask = ctx.reg_alloc.ScratchXmm();
DenormalsAreZero<fsize>(code, ctx, {result, xmm_b}, mask); DenormalsAreZero<fsize>(code, ctx, {result, xmm_b}, mask);
@ -582,22 +582,32 @@ static void EmitFPVectorMax(BlockOfCode& code, EmitContext& ctx, IR::Inst* inst)
if (code.DoesCpuSupport(Xbyak::util::Cpu::tAVX)) { if (code.DoesCpuSupport(Xbyak::util::Cpu::tAVX)) {
FCODE(vcmpeqp)(mask, result, xmm_b); FCODE(vcmpeqp)(mask, result, xmm_b);
FCODE(vcmpunordp)(nan_mask, result, xmm_b); FCODE(vcmpunordp)(nan_mask, result, xmm_b);
FCODE(vandp)(anded, result, xmm_b); if constexpr (is_max) {
FCODE(vmaxp)(result, result, xmm_b); FCODE(vandp)(eq, result, xmm_b);
FCODE(blendvp)(result, anded); FCODE(vmaxp)(result, result, xmm_b);
} else {
FCODE(vorp)(eq, result, xmm_b);
FCODE(vminp)(result, result, xmm_b);
}
FCODE(blendvp)(result, eq);
FCODE(vblendvp)(result, result, GetNaNVector<fsize>(code), nan_mask); FCODE(vblendvp)(result, result, GetNaNVector<fsize>(code), nan_mask);
} else { } else {
code.movaps(mask, result); code.movaps(mask, result);
code.movaps(anded, result); code.movaps(eq, result);
code.movaps(nan_mask, result); code.movaps(nan_mask, result);
FCODE(cmpneqp)(mask, xmm_b); FCODE(cmpneqp)(mask, xmm_b);
FCODE(cmpordp)(nan_mask, xmm_b); FCODE(cmpordp)(nan_mask, xmm_b);
code.andps(anded, xmm_b); if constexpr (is_max) {
FCODE(maxp)(result, xmm_b); code.andps(eq, xmm_b);
FCODE(maxp)(result, xmm_b);
} else {
code.orps(eq, xmm_b);
FCODE(minp)(result, xmm_b);
}
code.andps(result, mask); code.andps(result, mask);
code.andnps(mask, anded); code.andnps(mask, eq);
code.orps(result, mask); code.orps(result, mask);
code.andps(result, nan_mask); code.andps(result, nan_mask);
@ -612,7 +622,7 @@ static void EmitFPVectorMax(BlockOfCode& code, EmitContext& ctx, IR::Inst* inst)
EmitThreeOpVectorOperation<fsize, DefaultIndexer>(code, ctx, inst, [&](const Xbyak::Xmm& result, Xbyak::Xmm xmm_b){ EmitThreeOpVectorOperation<fsize, DefaultIndexer>(code, ctx, inst, [&](const Xbyak::Xmm& result, Xbyak::Xmm xmm_b){
const Xbyak::Xmm mask = xmm0; const Xbyak::Xmm mask = xmm0;
const Xbyak::Xmm anded = ctx.reg_alloc.ScratchXmm(); const Xbyak::Xmm eq = ctx.reg_alloc.ScratchXmm();
if (ctx.FPSCR_FTZ()) { if (ctx.FPSCR_FTZ()) {
const Xbyak::Xmm prev_xmm_b = xmm_b; const Xbyak::Xmm prev_xmm_b = xmm_b;
@ -627,117 +637,48 @@ static void EmitFPVectorMax(BlockOfCode& code, EmitContext& ctx, IR::Inst* inst)
if (code.DoesCpuSupport(Xbyak::util::Cpu::tAVX)) { if (code.DoesCpuSupport(Xbyak::util::Cpu::tAVX)) {
FCODE(vcmpeqp)(mask, result, xmm_b); FCODE(vcmpeqp)(mask, result, xmm_b);
FCODE(vandp)(anded, result, xmm_b); if constexpr (is_max) {
FCODE(vmaxp)(result, result, xmm_b); FCODE(vandp)(eq, result, xmm_b);
FCODE(blendvp)(result, anded); FCODE(vmaxp)(result, result, xmm_b);
} else {
FCODE(vorp)(eq, result, xmm_b);
FCODE(vminp)(result, result, xmm_b);
}
FCODE(blendvp)(result, eq);
} else { } else {
code.movaps(mask, result); code.movaps(mask, result);
code.movaps(anded, result); code.movaps(eq, result);
FCODE(cmpneqp)(mask, xmm_b); FCODE(cmpneqp)(mask, xmm_b);
code.andps(anded, xmm_b); if constexpr (is_max) {
FCODE(maxp)(result, xmm_b); code.andps(eq, xmm_b);
FCODE(maxp)(result, xmm_b);
} else {
code.orps(eq, xmm_b);
FCODE(minp)(result, xmm_b);
}
code.andps(result, mask); code.andps(result, mask);
code.andnps(mask, anded); code.andnps(mask, eq);
code.orps(result, mask); code.orps(result, mask);
} }
}); });
} }
void EmitX64::EmitFPVectorMax32(EmitContext& ctx, IR::Inst* inst) { void EmitX64::EmitFPVectorMax32(EmitContext& ctx, IR::Inst* inst) {
EmitFPVectorMax<32>(code, ctx, inst); EmitFPVectorMinMax<32, true>(code, ctx, inst);
} }
void EmitX64::EmitFPVectorMax64(EmitContext& ctx, IR::Inst* inst) { void EmitX64::EmitFPVectorMax64(EmitContext& ctx, IR::Inst* inst) {
EmitFPVectorMax<64>(code, ctx, inst); EmitFPVectorMinMax<64, true>(code, ctx, inst);
}
template<size_t fsize>
static void EmitFPVectorMin(BlockOfCode& code, EmitContext& ctx, IR::Inst* inst) {
if (ctx.FPSCR_DN()) {
auto args = ctx.reg_alloc.GetArgumentInfo(inst);
const Xbyak::Xmm result = ctx.reg_alloc.UseScratchXmm(args[0]);
const Xbyak::Xmm xmm_b = ctx.FPSCR_FTZ() ? ctx.reg_alloc.UseScratchXmm(args[1]) : ctx.reg_alloc.UseXmm(args[1]);
const Xbyak::Xmm mask = xmm0;
const Xbyak::Xmm ored = ctx.reg_alloc.ScratchXmm();
const Xbyak::Xmm nan_mask = ctx.reg_alloc.ScratchXmm();
DenormalsAreZero<fsize>(code, ctx, {result, xmm_b}, mask);
if (code.DoesCpuSupport(Xbyak::util::Cpu::tAVX)) {
FCODE(vcmpeqp)(mask, result, xmm_b);
FCODE(vcmpunordp)(nan_mask, result, xmm_b);
FCODE(vorp)(ored, result, xmm_b);
FCODE(vminp)(result, result, xmm_b);
FCODE(blendvp)(result, ored);
FCODE(vblendvp)(result, result, GetNaNVector<fsize>(code), nan_mask);
} else {
code.movaps(mask, result);
code.movaps(ored, result);
code.movaps(nan_mask, result);
FCODE(cmpneqp)(mask, xmm_b);
FCODE(cmpordp)(nan_mask, xmm_b);
code.orps(ored, xmm_b);
FCODE(minp)(result, xmm_b);
code.andps(result, mask);
code.andnps(mask, ored);
code.orps(result, mask);
code.andps(result, nan_mask);
code.andnps(nan_mask, GetNaNVector<fsize>(code));
code.orps(result, nan_mask);
}
ctx.reg_alloc.DefineValue(inst, result);
return;
}
EmitThreeOpVectorOperation<fsize, DefaultIndexer>(code, ctx, inst, [&](const Xbyak::Xmm& result, Xbyak::Xmm xmm_b){
const Xbyak::Xmm mask = xmm0;
const Xbyak::Xmm ored = ctx.reg_alloc.ScratchXmm();
if (ctx.FPSCR_FTZ()) {
const Xbyak::Xmm prev_xmm_b = xmm_b;
xmm_b = ctx.reg_alloc.ScratchXmm();
code.movaps(xmm_b, prev_xmm_b);
DenormalsAreZero<fsize>(code, ctx, {result, xmm_b}, mask);
}
// What we are doing here is handling the case when the inputs are differently signed zeros.
// x86-64 treats differently signed zeros as equal while ARM does not.
// Thus if we OR together things that x86-64 thinks are equal we'll get the negative zero.
if (code.DoesCpuSupport(Xbyak::util::Cpu::tAVX)) {
FCODE(vcmpeqp)(mask, result, xmm_b);
FCODE(vorp)(ored, result, xmm_b);
FCODE(vminp)(result, result, xmm_b);
FCODE(blendvp)(result, ored);
} else {
code.movaps(mask, result);
code.movaps(ored, result);
FCODE(cmpneqp)(mask, xmm_b);
code.orps(ored, xmm_b);
FCODE(minp)(result, xmm_b);
code.andps(result, mask);
code.andnps(mask, ored);
code.orps(result, mask);
}
});
} }
void EmitX64::EmitFPVectorMin32(EmitContext& ctx, IR::Inst* inst) { void EmitX64::EmitFPVectorMin32(EmitContext& ctx, IR::Inst* inst) {
EmitFPVectorMin<32>(code, ctx, inst); EmitFPVectorMinMax<32, false>(code, ctx, inst);
} }
void EmitX64::EmitFPVectorMin64(EmitContext& ctx, IR::Inst* inst) { void EmitX64::EmitFPVectorMin64(EmitContext& ctx, IR::Inst* inst) {
EmitFPVectorMin<64>(code, ctx, inst); EmitFPVectorMinMax<64, false>(code, ctx, inst);
} }
void EmitX64::EmitFPVectorMul32(EmitContext& ctx, IR::Inst* inst) { void EmitX64::EmitFPVectorMul32(EmitContext& ctx, IR::Inst* inst) {