| commit | 05901d80badd51fae6e17eb4fd8d53d6e5a7bdce | [log] [tgz] |
|---|---|---|
| author | Chandler Carruth <chandlerc@gmail.com> | Thu Sep 25 10:21:15 2014 +0000 |
| committer | Chandler Carruth <chandlerc@gmail.com> | Thu Sep 25 10:21:15 2014 +0000 |
| tree | dddef560d7e06c1e2906e1acf9ecc3234340ba65 | |
| parent | f220c5387bde32ec9cdd7beb244bd7c133901e15 [diff] |
[x86] Teach the new vector shuffle lowering a fancier way to lower 256-bit vectors with lane-crossing. Rather than immediately decomposing to 128-bit vectors, try flipping the 256-bit vector lanes, shuffling them and blending them together. This reduces our worst case shuffle by a pretty significant margin across the board. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218446 91177308-0d34-0410-b5e6-96231b3b80d8