| commit | 429670f0e8e25f6f6deb41a4e17a4ca0f91f1325 | [log] [tgz] |
|---|---|---|
| author | Chandler Carruth <chandlerc@gmail.com> | Wed Oct 01 02:25:54 2014 +0000 |
| committer | Chandler Carruth <chandlerc@gmail.com> | Wed Oct 01 02:25:54 2014 +0000 |
| tree | 8196e928b823132014dad2a6ba49e98fb1359165 | |
| parent | 771ac70aed5efb591a78d22b44a4589183e6c3f0 [diff] |
[x86] Hoist the zext-lowering up in the v4i32 lowering routine -- it is the same speed as pshufd but we can fold loads into the pmovzx instructions. This fixes some regressions that came up in the regression test suite for the new vector shuffle lowering. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218733 91177308-0d34-0410-b5e6-96231b3b80d8