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

emit_x64_floating_point: Deduplicate EmitFPMulAdd implementation

Browse source
This commit is contained in:
MerryMage 2018-07-30 20:16:45 +01:00
parent e038fe72df
commit 2019d32743
1 changed files with 19 additions and 28 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

47
src/backend_x64/emit_x64_floating_point.cpp
View file

@ -14,6 +14,7 @@
#include "common/common_types.h" #include "common/common_types.h"
#include "common/fp/fpcr.h" #include "common/fp/fpcr.h"
#include "common/fp/fpsr.h" #include "common/fp/fpsr.h"
#include "common/fp/info.h"
#include "common/fp/op.h" #include "common/fp/op.h"
#include "common/fp/rounding_mode.h" #include "common/fp/rounding_mode.h"
#include "common/fp/util.h" #include "common/fp/util.h"
@ -648,8 +649,22 @@ void EmitX64::EmitFPMul64(EmitContext& ctx, IR::Inst* inst) {
FPThreeOp<64>(code, ctx, inst, &Xbyak::CodeGenerator::mulsd); FPThreeOp<64>(code, ctx, inst, &Xbyak::CodeGenerator::mulsd);
} }
template<typename FPT> template<size_t fsize>
static void EmitFPMulAddFallback(BlockOfCode& code, EmitContext& ctx, IR::Inst* inst) { static void EmitFPMulAdd(BlockOfCode& code, EmitContext& ctx, IR::Inst* inst) {
using FPT = mp::unsigned_integer_of_size<fsize>;
if (code.DoesCpuSupport(Xbyak::util::Cpu::tFMA)) {
FPFourOp<fsize>(code, ctx, inst, [&](Xbyak::Xmm result, Xbyak::Xmm operand2, Xbyak::Xmm operand3) {
FCODE(vfmadd231s)(result, operand2, operand3);
}, [](FPT a, FPT b, FPT c) -> FPT {
if (FP::IsQNaN(a) && ((FP::IsInf(b) && FP::IsZero(c)) || (FP::IsZero(b) && FP::IsInf(c)))) {
return FP::FPInfo<FPT>::DefaultNaN();
}
return *FP::ProcessNaNs(a, b, c);
});
return;
}
auto args = ctx.reg_alloc.GetArgumentInfo(inst); auto args = ctx.reg_alloc.GetArgumentInfo(inst);
ctx.reg_alloc.HostCall(inst, args[0], args[1], args[2]); ctx.reg_alloc.HostCall(inst, args[0], args[1], args[2]);
code.mov(code.ABI_PARAM4.cvt32(), ctx.FPCR()); code.mov(code.ABI_PARAM4.cvt32(), ctx.FPCR());
@ -666,35 +681,11 @@ static void EmitFPMulAddFallback(BlockOfCode& code, EmitContext& ctx, IR::Inst*
} }
void EmitX64::EmitFPMulAdd32(EmitContext& ctx, IR::Inst* inst) { void EmitX64::EmitFPMulAdd32(EmitContext& ctx, IR::Inst* inst) {
if (code.DoesCpuSupport(Xbyak::util::Cpu::tFMA)) { EmitFPMulAdd<32>(code, ctx, inst);
FPFourOp<32>(code, ctx, inst, [&](Xbyak::Xmm result, Xbyak::Xmm operand2, Xbyak::Xmm operand3) {
code.vfmadd231ss(result, operand2, operand3);
}, [](u32 a, u32 b, u32 c) -> u32 {
if (FP::IsQNaN(a) && ((FP::IsInf(b) && FP::IsZero(c)) || (FP::IsZero(b) && FP::IsInf(c)))) {
return f32_nan;
}
return *FP::ProcessNaNs(a, b, c);
});
return;
}
EmitFPMulAddFallback<u32>(code, ctx, inst);
} }
void EmitX64::EmitFPMulAdd64(EmitContext& ctx, IR::Inst* inst) { void EmitX64::EmitFPMulAdd64(EmitContext& ctx, IR::Inst* inst) {
if (code.DoesCpuSupport(Xbyak::util::Cpu::tFMA)) { EmitFPMulAdd<64>(code, ctx, inst);
FPFourOp<64>(code, ctx, inst, [&](Xbyak::Xmm result, Xbyak::Xmm operand2, Xbyak::Xmm operand3) {
code.vfmadd231sd(result, operand2, operand3);
}, [](u64 a, u64 b, u64 c) -> u64 {
if (FP::IsQNaN(a) && ((FP::IsInf(b) && FP::IsZero(c)) || (FP::IsZero(b) && FP::IsInf(c)))) {
return f64_nan;
}
return *FP::ProcessNaNs(a, b, c);
});
return;
}
EmitFPMulAddFallback<u64>(code, ctx, inst);
} }
template<typename FPT> template<typename FPT>