diff options
| -rw-r--r-- | llvm/include/llvm/Analysis/LoopAccessAnalysis.h | 3 | ||||
| -rw-r--r-- | llvm/lib/Analysis/LoopAccessAnalysis.cpp | 10 |
2 files changed, 11 insertions, 2 deletions
diff --git a/llvm/include/llvm/Analysis/LoopAccessAnalysis.h b/llvm/include/llvm/Analysis/LoopAccessAnalysis.h index 717076a233b..795f4cd746d 100644 --- a/llvm/include/llvm/Analysis/LoopAccessAnalysis.h +++ b/llvm/include/llvm/Analysis/LoopAccessAnalysis.h @@ -321,6 +321,9 @@ private: /// \brief Check whether the data dependence could prevent store-load /// forwarding. + /// + /// \return false if we shouldn't vectorize at all or avoid larger + /// vectorization factors by limiting MaxSafeDepDistBytes. bool couldPreventStoreLoadForward(unsigned Distance, unsigned TypeByteSize); }; diff --git a/llvm/lib/Analysis/LoopAccessAnalysis.cpp b/llvm/lib/Analysis/LoopAccessAnalysis.cpp index eff8b2a613c..0b41da5bbe1 100644 --- a/llvm/lib/Analysis/LoopAccessAnalysis.cpp +++ b/llvm/lib/Analysis/LoopAccessAnalysis.cpp @@ -1081,14 +1081,20 @@ bool MemoryDepChecker::couldPreventStoreLoadForward(unsigned Distance, // hence on your typical architecture store-load forwarding does not take // place. Vectorizing in such cases does not make sense. // Store-load forwarding distance. - const unsigned NumCyclesForStoreLoadThroughMemory = 8 * TypeByteSize; + + // After this many iterations store-to-load forwarding conflicts should not + // cause any slowdowns. + const unsigned NumItersForStoreLoadThroughMemory = 8 * TypeByteSize; // Maximum vector factor. unsigned MaxVFWithoutSLForwardIssues = std::min( VectorizerParams::MaxVectorWidth * TypeByteSize, MaxSafeDepDistBytes); + // Compute the smallest VF at which the store and load would be misaligned. for (unsigned VF = 2 * TypeByteSize; VF <= MaxVFWithoutSLForwardIssues; VF *= 2) { - if (Distance % VF && Distance / VF < NumCyclesForStoreLoadThroughMemory) { + // If the number of vector iteration between the store and the load are + // small we could incur conflicts. + if (Distance % VF && Distance / VF < NumItersForStoreLoadThroughMemory) { MaxVFWithoutSLForwardIssues = (VF >>= 1); break; } |
