[GlobalISel][AMDGPU] Lower G_SMULH/G_UMULH
Reviewed By: arsenm, foad
Differential Revision: https://reviews.llvm.org/D85653
diff --git a/llvm/lib/CodeGen/GlobalISel/LegalizerHelper.cpp b/llvm/lib/CodeGen/GlobalISel/LegalizerHelper.cpp
index 196dbf2..0004223 100644
--- a/llvm/lib/CodeGen/GlobalISel/LegalizerHelper.cpp
+++ b/llvm/lib/CodeGen/GlobalISel/LegalizerHelper.cpp
@@ -2837,6 +2837,9 @@
case TargetOpcode::G_SADDO:
case TargetOpcode::G_SSUBO:
return lowerSADDO_SSUBO(MI);
+ case TargetOpcode::G_UMULH:
+ case TargetOpcode::G_SMULH:
+ return lowerSMULH_UMULH(MI);
case TargetOpcode::G_SMULO:
case TargetOpcode::G_UMULO: {
// Generate G_UMULH/G_SMULH to check for overflow and a normal G_MUL for the
@@ -6144,3 +6147,25 @@
MI.eraseFromParent();
return Legalized;
}
+
+LegalizerHelper::LegalizeResult
+LegalizerHelper::lowerSMULH_UMULH(MachineInstr &MI) {
+ bool IsSigned = MI.getOpcode() == TargetOpcode::G_SMULH;
+ unsigned ExtOp = IsSigned ? TargetOpcode::G_SEXT : TargetOpcode::G_ZEXT;
+ Register Result = MI.getOperand(0).getReg();
+ LLT OrigTy = MRI.getType(Result);
+ auto SizeInBits = OrigTy.getScalarSizeInBits();
+ LLT WideTy = OrigTy.changeElementSize(SizeInBits * 2);
+
+ auto LHS = MIRBuilder.buildInstr(ExtOp, {WideTy}, {MI.getOperand(1)});
+ auto RHS = MIRBuilder.buildInstr(ExtOp, {WideTy}, {MI.getOperand(2)});
+ auto Mul = MIRBuilder.buildMul(WideTy, LHS, RHS);
+ unsigned ShiftOp = IsSigned ? TargetOpcode::G_ASHR : TargetOpcode::G_LSHR;
+
+ auto ShiftAmt = MIRBuilder.buildConstant(WideTy, SizeInBits);
+ auto Shifted = MIRBuilder.buildInstr(ShiftOp, {WideTy}, {Mul, ShiftAmt});
+ MIRBuilder.buildTrunc(Result, Shifted);
+
+ MI.eraseFromParent();
+ return Legalized;
+}