projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 1accb7e) | patch
Fix a crash in AVX2 when trying to broadcast a double into a 128-bit vector. There...
authorCraig Topper <craig.topper@gmail.com>
Tue, 10 Jan 2012 08:23:59 +0000 (08:23 +0000)
committerCraig Topper <craig.topper@gmail.com>
Tue, 10 Jan 2012 08:23:59 +0000 (08:23 +0000)
commita937633893432b4cdbd1a32602c566f0a006f2fd
treef499582fd517facaa9f4a17bed64bffed9aaea44tree | snapshot
parent1accb7ed98d823c291a4d5df172d0538451aba9ecommit | diff
Fix a crash in AVX2 when trying to broadcast a double into a 128-bit vector. There is no vbroadcastsd xmm, but we do need to support 64-bit integers broadcasted into xmm. Also factor the AVX check into the isVectorBroadcast function. This makes more sense since the AVX2 check was already inside.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147844 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/X86/X86ISelLowering.cpp diff | blob | history
test/CodeGen/X86/avx-vbroadcast.ll diff | blob | history
test/CodeGen/X86/avx2-vbroadcast.ll diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors