diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2017-05-17 20:00:00 +0000 |
|---|---|---|
| committer | Matt Arsenault <Matthew.Arsenault@amd.com> | 2017-05-17 20:00:00 +0000 |
| commit | 786eeea23eef657c95a3b94c8903478ad38462da (patch) | |
| tree | 1b77ee721f5f0f2f66bf0047efb8104c97601512 /llvm/test/CodeGen/Hexagon/frame-offset-overflow.ll | |
| parent | 91b46c99be2f612c482b99e1196f5265b0fb8b9f (diff) | |
| download | bcm5719-llvm-786eeea23eef657c95a3b94c8903478ad38462da.tar.gz bcm5719-llvm-786eeea23eef657c95a3b94c8903478ad38462da.zip | |
AMDGPU: Try to use op_sel when selecting packed instructions
Avoids instructions to pack a vector when the source is really
a scalar being broadcast.
Also be smarter and look for per-component fneg.
Doesn't yet handle scalar from upper half of register
or other swizzles.
llvm-svn: 303291
Diffstat (limited to 'llvm/test/CodeGen/Hexagon/frame-offset-overflow.ll')
0 files changed, 0 insertions, 0 deletions
