projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: b7e230d) | patch
AVX: Fix a bug in getTargetVShiftNode. The shift amount has to be a 128bit vector...
authorNadav Rotem <nadav.rotem@intel.com>
Sat, 14 Jul 2012 22:26:05 +0000 (22:26 +0000)
committerNadav Rotem <nadav.rotem@intel.com>
Sat, 14 Jul 2012 22:26:05 +0000 (22:26 +0000)
commit65f489fd7d876c3e624938cd46d2475c7f365a8a
treed95f9beeb8ab1aac258e5b2bb82714fada0ad911tree | snapshot
parentb7e230d999deec5c90ee51a7c2bbc6ee5be0a914commit | diff
AVX: Fix a bug in getTargetVShiftNode. The shift amount has to be a 128bit vector with the same element type as the input vector.
This is needed because of the patterns we have for the VP[SLL/SRA/SRL][W/D/Q] instructions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@160222 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/X86/X86ISelLowering.cpp diff | blob | history
test/CodeGen/X86/2012-07-15-vshl.ll [new file with mode: 0644] blob
C/C++ LLVM-based compilers that forbids OOTA behaviors