maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

AMDGPU/GlobalISel: Invert parameter for div/rem lowering function

This commit is contained in:
Matt Arsenault 2020-06-24 11:12:02 -04:00
parent 6a3cffce3e
commit b5c4e6c148
1 changed files with 13 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
View File

@ -2537,7 +2537,7 @@ void AMDGPULegalizerInfo::legalizeUDIV_UREM32Impl(MachineIRBuilder &B,
Register DstReg,
Register Num,
Register Den,
bool IsRem) const {
bool IsDiv) const {
const LLT S1 = LLT::scalar(1);
const LLT S32 = LLT::scalar(32);
@ -2604,7 +2604,9 @@ void AMDGPULegalizerInfo::legalizeUDIV_UREM32Impl(MachineIRBuilder &B,
auto Div = B.buildSelect(S32, Tmp1, Quotient_A_One, Quotient);
// Div = (Remainder_GE_Zero ? Div : Quotient_S_One)
if (IsRem) {
if (IsDiv) {
B.buildSelect(DstReg, Remainder_GE_Zero, Div, Quotient_S_One);
} else {
Div = B.buildSelect(S32, Remainder_GE_Zero, Div, Quotient_S_One);
// Calculate Rem result:
@ -2618,19 +2620,17 @@ void AMDGPULegalizerInfo::legalizeUDIV_UREM32Impl(MachineIRBuilder &B,
// Rem = (Remainder_GE_Zero ? Rem : Remainder_A_Den)
B.buildSelect(DstReg, Remainder_GE_Zero, Rem, Remainder_A_Den);
} else {
B.buildSelect(DstReg, Remainder_GE_Zero, Div, Quotient_S_One);
}
}
bool AMDGPULegalizerInfo::legalizeUDIV_UREM32(MachineInstr &MI,
MachineRegisterInfo &MRI,
MachineIRBuilder &B) const {
const bool IsRem = MI.getOpcode() == AMDGPU::G_UREM;
const bool IsDiv = MI.getOpcode() == AMDGPU::G_UDIV;
Register DstReg = MI.getOperand(0).getReg();
Register Num = MI.getOperand(1).getReg();
Register Den = MI.getOperand(2).getReg();
legalizeUDIV_UREM32Impl(B, DstReg, Num, Den, IsRem);
legalizeUDIV_UREM32Impl(B, DstReg, Num, Den, IsDiv);
MI.eraseFromParent();
return true;
}
@ -2811,7 +2811,7 @@ bool AMDGPULegalizerInfo::legalizeSDIV_SREM32(MachineInstr &MI,
MachineIRBuilder &B) const {
const LLT S32 = LLT::scalar(32);
const bool IsRem = MI.getOpcode() == AMDGPU::G_SREM;
const bool IsDiv = MI.getOpcode() == AMDGPU::G_SDIV;
Register DstReg = MI.getOperand(0).getReg();
Register LHS = MI.getOperand(1).getReg();
Register RHS = MI.getOperand(2).getReg();
@ -2827,16 +2827,16 @@ bool AMDGPULegalizerInfo::legalizeSDIV_SREM32(MachineInstr &MI,
RHS = B.buildXor(S32, RHS, RHSign).getReg(0);
Register UDivRem = MRI.createGenericVirtualRegister(S32);
legalizeUDIV_UREM32Impl(B, UDivRem, LHS, RHS, IsRem);
legalizeUDIV_UREM32Impl(B, UDivRem, LHS, RHS, IsDiv);
if (IsRem) {
auto RSign = LHSign; // Remainder sign is the same as LHS
UDivRem = B.buildXor(S32, UDivRem, RSign).getReg(0);
B.buildSub(DstReg, UDivRem, RSign);
} else {
if (IsDiv) {
auto DSign = B.buildXor(S32, LHSign, RHSign);
UDivRem = B.buildXor(S32, UDivRem, DSign).getReg(0);
B.buildSub(DstReg, UDivRem, DSign);
} else {
auto RSign = LHSign; // Remainder sign is the same as LHS
UDivRem = B.buildXor(S32, UDivRem, RSign).getReg(0);
B.buildSub(DstReg, UDivRem, RSign);
}
MI.eraseFromParent();