diff options
| author | Arnold Schwaighofer <aschwaighofer@apple.com> | 2013-04-18 17:22:34 +0000 |
|---|---|---|
| committer | Arnold Schwaighofer <aschwaighofer@apple.com> | 2013-04-18 17:22:34 +0000 |
| commit | a3fb330d05e85107d01ecf133355d0c6a88196fd (patch) | |
| tree | bb72c5b84952d2c0a48de9bb27b8637979329384 /tools/llvm-objdump/llvm-objdump.cpp | |
| parent | bff177676c32b88e19b8230cf048b5d7bdc7d657 (diff) | |
LoopVectorizer: Recognize min/max reductions
A min/max operation is represented by a select(cmp(lt/le/gt/ge, X, Y), X, Y)
sequence in LLVM. If we see such a sequence we can treat it just as any other
commutative binary instruction and reduce it.
This appears to help bzip2 by about 1.5% on an imac12,2.
radar://12960601
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179773 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'tools/llvm-objdump/llvm-objdump.cpp')
0 files changed, 0 insertions, 0 deletions