HomePhabricator

[x86] allow narrowing of vector UINT_TO_FP

Authored by spatel on Feb 22 2019, 7:47 AM.

Description

[x86] allow narrowing of vector UINT_TO_FP

As discussed in:
D56864
D58197

Always use the narrow (128-bit) instruction when possible.
We already had the signed int version of this transform.

llvm-svn: 354675

Details

Committed
spatelFeb 22 2019, 7:47 AM
Parents
rG1baf7896cc2b: [x86] simplify code in combineExtractSubvector; NFC
Branches
Unknown
Tags
Unknown