projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 29630ff) | patch
Make this kind of lowering to be supported by 256-bit instructions:
authorBruno Cardoso Lopes <bruno.cardoso@gmail.com>
Tue, 2 Aug 2011 16:06:18 +0000 (16:06 +0000)
committerBruno Cardoso Lopes <bruno.cardoso@gmail.com>
Tue, 2 Aug 2011 16:06:18 +0000 (16:06 +0000)
commitac5f13fe3f4b79fa188d407dc97de9c9f8fdf368
tree940d415b63f59f5c0bdc2e6816e40b26a22d05fctree | snapshot
parent29630ff214085365d11f1a58251ba07326e21398commit | diff
Make this kind of lowering to be supported by 256-bit instructions:
  shuffle (scalar_to_vector (load (ptr + 4))), undef, <0, 0, 0, 0>
To:
  shuffle (vload ptr)), undef, <1, 1, 1, 1>
Fix PR10494

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136691 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/X86/X86ISelLowering.cpp diff | blob | history
test/CodeGen/X86/avx-256-splat.ll diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors