When optimzing a mul by immediate into two, the resulting mul's should get a x86 specific node to avoid dag combiner from hacking on them further.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@68066 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Evan Cheng
2009-03-30 21:36:47 +00:00
parent e6f350d755
commit 73f24c9f0d
5 changed files with 12 additions and 5 deletions

View File

@@ -7176,6 +7176,7 @@ const char *X86TargetLowering::getTargetNodeName(unsigned Opcode) const {
case X86ISD::UMUL: return "X86ISD::UMUL";
case X86ISD::INC: return "X86ISD::INC";
case X86ISD::DEC: return "X86ISD::DEC";
case X86ISD::MUL_IMM: return "X86ISD::MUL_IMM";
}
}
@@ -8458,14 +8459,14 @@ static SDValue PerformMulCombine(SDNode *N, SelectionDAG &DAG,
NewMul = DAG.getNode(ISD::SHL, DL, VT, N->getOperand(0),
DAG.getConstant(Log2_64(MulAmt1), MVT::i8));
else
NewMul = DAG.getNode(ISD::MUL, DL, VT, N->getOperand(0),
NewMul = DAG.getNode(X86ISD::MUL_IMM, DL, VT, N->getOperand(0),
DAG.getConstant(MulAmt1, VT));
if (isPowerOf2_64(MulAmt2))
NewMul = DAG.getNode(ISD::SHL, DL, VT, NewMul,
DAG.getConstant(Log2_64(MulAmt2), MVT::i8));
else
NewMul = DAG.getNode(ISD::MUL, DL, VT, NewMul,
NewMul = DAG.getNode(X86ISD::MUL_IMM, DL, VT, NewMul,
DAG.getConstant(MulAmt2, VT));
// Do not add new nodes to DAG combiner worklist.