diff options
| author | Filipe Cabecinhas <me@filcab.net> | 2014-04-24 00:38:14 +0000 |
|---|---|---|
| committer | Filipe Cabecinhas <me@filcab.net> | 2014-04-24 00:38:14 +0000 |
| commit | 1a80595a2b803949608e11d522d715a54c3a11bc (patch) | |
| tree | 2e5d3bfdd9a02c8c616c944bd97b6697c1b13789 /llvm/lib/Transforms/Hello | |
| parent | 60728177fb0f443b5e07a3f65827c510011c0ffe (diff) | |
| download | bcm5719-llvm-1a80595a2b803949608e11d522d715a54c3a11bc.tar.gz bcm5719-llvm-1a80595a2b803949608e11d522d715a54c3a11bc.zip | |
Optimize some special cases for SSE4a insertqi
Summary:
Since the upper 64 bits of the destination register are undefined when
performing this operation, we can substitute it and let the optimizer
figure out that only a copy is needed.
Also added range merging, if an instruction copies a range that can be
merged with a previous copied range.
Added test cases for both optimizations.
Reviewers: grosbach, nadav
CC: llvm-commits
Differential Revision: http://reviews.llvm.org/D3357
llvm-svn: 207055
Diffstat (limited to 'llvm/lib/Transforms/Hello')
0 files changed, 0 insertions, 0 deletions
