projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 88d51ce) | patch
[SLP] Enable -slp-vectorize-hor by default.
authorCharlie Turner <charlie.turner@arm.com>
Wed, 11 Nov 2015 15:03:46 +0000 (15:03 +0000)
committerCharlie Turner <charlie.turner@arm.com>
Wed, 11 Nov 2015 15:03:46 +0000 (15:03 +0000)
commitca0bc2bcbc2891b286af3f8040194ea1d9798e38
tree6d99bb2af4b2662e9c4910c4df5e585137e89267tree | snapshot
parent88d51ce445ecad5825506e9e88ea0717d68e55d8commit | diff
[SLP] Enable -slp-vectorize-hor by default.

Measurements primarily on AArch64 have shown this feature does not
significantly effect compile-time. The are no significant perf changes in LNT,
but for AArch64 at least, there are wins in third party benchmarks.

As discussed on llvm-dev, we're going to try turning this on by default and see
how other targets react to the change.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@252733 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Transforms/Vectorize/SLPVectorizer.cpp diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors