6502/llvm-6502
1
0
Fork 0
mirror of https://github.com/c64scene-ar/llvm-6502.git synced 2025-02-19 06:31:18 +00:00
Code Issues Projects Releases Wiki Activity

Teach DAG combiner to constant fold fneg of a BUILD_VECTOR of constants.

Browse Source 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163483 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Craig Topper 2012-09-09 22:58:45 +00:00
parent 236bcf1fcd
commit 956342b210
2 changed files with 27 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
lib/CodeGen/SelectionDAG/DAGCombiner.cpp
View File

@ -413,7 +413,7 @@ static char isNegatibleForFree(SDValue Op, bool LegalOperations,
!TLI.isOperationLegalOrCustom(ISD::FSUB, Op.getValueType()))
return 0;
// fold (fsub (fadd A, B)) -> (fsub (fneg A), B)
// fold (fneg (fadd A, B)) -> (fsub (fneg A), B)
if (char V = isNegatibleForFree(Op.getOperand(0), LegalOperations, TLI,
Options, Depth + 1))
return V;
@ -6409,6 +6409,30 @@ SDValue DAGCombiner::visitFNEG(SDNode *N) {
SDValue N0 = N->getOperand(0);
EVT VT = N->getValueType(0);
if (VT.isVector() && !LegalOperations) {
// If operand is a BUILD_VECTOR node, see if we can constant fold it.
if (N0.getOpcode() == ISD::BUILD_VECTOR) {
SmallVector<SDValue, 8> Ops;
for (unsigned i = 0, e = N0.getNumOperands(); i != e; ++i) {
SDValue Op = N0.getOperand(i);
if (Op.getOpcode() != ISD::UNDEF &&
Op.getOpcode() != ISD::ConstantFP)
break;
EVT EltVT = Op.getValueType();
SDValue FoldOp = DAG.getNode(ISD::FNEG, N0.getDebugLoc(), EltVT, Op);
if (FoldOp.getOpcode() != ISD::UNDEF &&
FoldOp.getOpcode() != ISD::ConstantFP)
break;
Ops.push_back(FoldOp);
AddToWorkList(FoldOp.getNode());
}
if (Ops.size() == N0.getNumOperands())
return DAG.getNode(ISD::BUILD_VECTOR, N->getDebugLoc(),
VT, &Ops[0], Ops.size());
}
}
if (isNegatibleForFree(N0, LegalOperations, DAG.getTargetLoweringInfo(),
&DAG.getTarget().Options))
return GetNegatedExpression(N0, DAG, LegalOperations);

4
test/CodeGen/X86/avx-basic.ll
View File

@ -109,8 +109,8 @@ allocas:
; rdar://10566486
; CHECK: fneg
; CHECK: vxorps
define <16 x float> @fneg(<16 x float> addrspace(1)* nocapture %out) nounwind {
%1 = fsub <16 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>
define <16 x float> @fneg(<16 x float> %a) nounwind {
%1 = fsub <16 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %a
ret <16 x float> %1
}