| commit | 273949e71bf6db3a81d8f2c2dff9cede8b73ea32 | [log] [tgz] |
|---|---|---|
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | Mon Jul 02 15:14:07 2018 +0000 |
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | Mon Jul 02 15:14:07 2018 +0000 |
| tree | d0b28f1aa64c21610bd47ecc8bb24ff39fe90852 | |
| parent | 3e5a75b28dfc5452ee1d74a5873e859356529be0 [diff] |
[X86][SSE] Blend any v8i16/v4i32 shift with 2 shift unique values We were only doing this for basic blends, despite shuffle lowering now being good enough to handle more complex blends. This means that the two v8i16 splat shifts are performed in parallel instead of serially as the general shift case. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@336113 91177308-0d34-0410-b5e6-96231b3b80d8