| commit | 3e38505d0613c6d284201665f018135a643b5f8d | [log] [tgz] |
|---|---|---|
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | Tue Feb 19 17:23:55 2019 +0000 |
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | Tue Feb 19 17:23:55 2019 +0000 |
| tree | 5bf3e8400c8f375110a4a19e6d1b172348eaf9f2 | |
| parent | be849ca61f09d3c048ca39146df12f144d68264a [diff] |
[X86][AVX2] Hide VPBLENDD instructions behind AVX2 predicate This was the cause of the regression in D57888 - the commuted load pattern wasn't hidden by the predicate so once we enabled v4i32 blends on SSE41+ targets then isel was incorrectly matched against AVX2+ instructions. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@354358 91177308-0d34-0410-b5e6-96231b3b80d8