diff options
| author | Chad Rosier <mcrosier@apple.com> | 2012-03-09 02:00:48 +0000 |
|---|---|---|
| committer | Chad Rosier <mcrosier@apple.com> | 2012-03-09 02:00:48 +0000 |
| commit | abd6674166d99a8da98269051fbd869d9d8c3ba1 (patch) | |
| tree | 4b68f0ac12b9221bc2931e98600677d9a545ca4a /lib/Analysis/InlineCost.cpp | |
| parent | d04ec0c855176ebddd459c044bdd24f49938fae4 (diff) | |
Fix a regression from r147481.
Original commit message from r147481:
DAGCombine for transforming 128->256 casts into a vmovaps, rather
then a vxorps + vinsertf128 pair if the original vector came from a load.
Fix:
Unaligned loads need to generate a vmovups.
rdar://10974078
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@152366 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'lib/Analysis/InlineCost.cpp')
0 files changed, 0 insertions, 0 deletions