aboutsummaryrefslogtreecommitdiff
path: root/lib/Transforms/Vectorize/BBVectorize.cpp
diff options
context:
space:
mode:
authorHal Finkel <hfinkel@anl.gov>2012-04-13 17:15:33 +0000
committerHal Finkel <hfinkel@anl.gov>2012-04-13 17:15:33 +0000
commit064551e94c9763c7e74121b88f9cbccc0969946a (patch)
treee7318aec8e99f055e9fad22b1a6db83c6eba7f7e /lib/Transforms/Vectorize/BBVectorize.cpp
parente92077f11e2ca85d98fe41cf3aba2cb813e57c9a (diff)
By default, use Early-CSE instead of GVN for vectorization cleanup.
As has been suggested by Duncan and others, Early-CSE and GVN should do similar redundancy elimination, but Early-CSE is much less expensive. Most of my autovectorization benchmarks show a performance regresion, but all of these are < 0.1%, and so I think that it is still worth using the less expensive pass. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154673 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'lib/Transforms/Vectorize/BBVectorize.cpp')
0 files changed, 0 insertions, 0 deletions