| commit | aeef43e1bb2002691342816e0e50f090ae8edbe6 | [log] [tgz] |
|---|---|---|
| author | Craig Topper <craig.topper@intel.com> | Fri Nov 02 22:48:02 2018 +0000 |
| committer | Craig Topper <craig.topper@intel.com> | Fri Nov 02 22:48:02 2018 +0000 |
| tree | c7f8f3f3ee397fc6ba780aae8d3b65e415055573 | |
| parent | ef5bf36c741d327d2139bb3754c5cb7ec6fdbd4d [diff] |
[X86] In LowerEXTEND_VECTOR_INREG, emit a vector shuffle instead of directly using X86ISD::UNPCKL The majority of the changes are because the rest of shuffle lowering/combining prefers to replace the undef input with the other operand. Using UNPCKL directly seemed to avoid this and just grabbed a randomish register for the undef which can create false dependencies. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346050 91177308-0d34-0410-b5e6-96231b3b80d8