6502/llvm-6502
1
0
Fork 0
mirror of https://github.com/c64scene-ar/llvm-6502.git synced 2025-11-01 00:17:01 +00:00
Code Issues Projects Releases Wiki Activity

Fixed a bug in printing "cmp" pseudo ops.

Browse Source 
> This IR code
> %res = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a1, i8 14)
> fails with assertion:
>
> llc: X86ATTInstPrinter.cpp:62: void llvm::X86ATTInstPrinter::printSSECC(const llvm::MCInst*, unsigned int, llvm::raw_ostream&): Assertion `0 && "Invalid ssecc argument!"' failed.
> 0  llc             0x0000000001355803
> 1  llc             0x0000000001355dc9
> 2  libpthread.so.0 0x00007f79a30575d0
> 3  libc.so.6       0x00007f79a23a1945 gsignal + 53
> 4  libc.so.6       0x00007f79a23a2f21 abort + 385
> 5  libc.so.6       0x00007f79a239a810 __assert_fail + 240
> 6  llc             0x00000000011858d5 llvm::X86ATTInstPrinter::printSSECC(llvm::MCInst const*, unsigned int, llvm::raw_ostream&) + 119

I added the full testing for all possible pseudo-ops of cmp.
I extended X86AsmPrinter.cpp and X86IntelInstPrinter.cpp.

You'l also see lines alignments (unrelated to this fix) in X86IselLowering.cpp from my previous check-in.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@150068 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Elena Demikhovsky
2012-02-08 08:37:26 +00:00
parent bc0e4bf754
commit f602040c49
5 changed files with 183 additions and 41 deletions
Download Patch File Download Diff File Expand all files Collapse all files

35
lib/Target/X86/X86ISelLowering.cpp
View File

@@ -14597,41 +14597,42 @@ static SDValue PerformSExtCombine(SDNode *N, SelectionDAG &DAG,
if (!DCI.isBeforeLegalizeOps())
return SDValue();
if (!Subtarget->hasAVX()) return SDValue();
if (!Subtarget->hasAVX())
return SDValue();
// Optimize vectors in AVX mode
// Sign extend v8i16 to v8i32 and
// v4i32 to v4i64
//
// Divide input vector into two parts
// for v4i32 the shuffle mask will be { 0, 1, -1, -1} {2, 3, -1, -1}
// use vpmovsx instruction to extend v4i32 -> v2i64; v8i16 -> v4i32
// concat the vectors to original VT
// Optimize vectors in AVX mode
// Sign extend v8i16 to v8i32 and
// v4i32 to v4i64
//
// Divide input vector into two parts
// for v4i32 the shuffle mask will be { 0, 1, -1, -1} {2, 3, -1, -1}
// use vpmovsx instruction to extend v4i32 -> v2i64; v8i16 -> v4i32
// concat the vectors to original VT
EVT VT = N->getValueType(0);
SDValue Op = N->getOperand(0);
EVT OpVT = Op.getValueType();
DebugLoc dl = N->getDebugLoc();
if (((VT == MVT::v4i64) && (OpVT == MVT::v4i32)) ||
((VT == MVT::v8i32) && (OpVT == MVT::v8i16))) {
if ((VT == MVT::v4i64 && OpVT == MVT::v4i32) ||
(VT == MVT::v8i32 && OpVT == MVT::v8i16)) {
unsigned NumElems = OpVT.getVectorNumElements();
SmallVector<int,8> ShufMask1(NumElems, -1);
for (unsigned i=0; i< NumElems/2; i++) ShufMask1[i] = i;
for (unsigned i = 0; i < NumElems/2; i++) ShufMask1[i] = i;
SDValue OpLo = DAG.getVectorShuffle(OpVT, dl, Op, DAG.getUNDEF(OpVT),
ShufMask1.data());
ShufMask1.data());
SmallVector<int,8> ShufMask2(NumElems, -1);
for (unsigned i=0; i< NumElems/2; i++) ShufMask2[i] = i+NumElems/2;
for (unsigned i = 0; i < NumElems/2; i++) ShufMask2[i] = i + NumElems/2;
SDValue OpHi = DAG.getVectorShuffle(OpVT, dl, Op, DAG.getUNDEF(OpVT),
ShufMask2.data());
ShufMask2.data());
EVT HalfVT = EVT::getVectorVT(*DAG.getContext(), VT.getScalarType(),
VT.getVectorNumElements()/2);
VT.getVectorNumElements()/2);
OpLo = DAG.getNode(X86ISD::VSEXT_MOVL, dl, HalfVT, OpLo);
OpHi = DAG.getNode(X86ISD::VSEXT_MOVL, dl, HalfVT, OpHi);