diff options
| author | Kyle Butt <kyle+llvm@iteratee.net> | 2016-01-08 01:55:13 +0000 |
|---|---|---|
| committer | Kyle Butt <kyle+llvm@iteratee.net> | 2016-01-08 01:55:13 +0000 |
| commit | a02ce98bd4d406b0aa3b440dfcacfbd23a8b95fa (patch) | |
| tree | bfd38fb3c9ee8197ac271bdc70b971031acff97c /llvm/lib/Analysis | |
| parent | d72a458d28b180b7d1c224df2e00653975ac4359 (diff) | |
| download | bcm5719-llvm-a02ce98bd4d406b0aa3b440dfcacfbd23a8b95fa.tar.gz bcm5719-llvm-a02ce98bd4d406b0aa3b440dfcacfbd23a8b95fa.zip | |
[Vectorization] Actually return from error case in isStridedPtr
The early return seems to be missed. This causes a radical and wrong loop
optimization on powerpc. It isn't reproducible on x86_64, because
"UseInterleaved" is false.
Patch by Tim Shen.
llvm-svn: 257134
Diffstat (limited to 'llvm/lib/Analysis')
| -rw-r--r-- | llvm/lib/Analysis/LoopAccessAnalysis.cpp | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/llvm/lib/Analysis/LoopAccessAnalysis.cpp b/llvm/lib/Analysis/LoopAccessAnalysis.cpp index d7896ade354..8bcdcb86201 100644 --- a/llvm/lib/Analysis/LoopAccessAnalysis.cpp +++ b/llvm/lib/Analysis/LoopAccessAnalysis.cpp @@ -845,6 +845,7 @@ int llvm::isStridedPtr(PredicatedScalarEvolution &PSE, Value *Ptr, if (Lp != AR->getLoop()) { DEBUG(dbgs() << "LAA: Bad stride - Not striding over innermost loop " << *Ptr << " SCEV: " << *PtrScev << "\n"); + return 0; } // The address calculation must not wrap. Otherwise, a dependence could be |
