1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-25 22:12:57 +02:00
llvm-mirror/test/Transforms/LoopVectorize/PowerPC
Michael Kuperstein 247caf8b05 [InstCombine] scalarizePHI should not assume the code it sees has been CSE'd
scalarizePHI only looked for phis that have exactly two uses - the "latch"
use, and an extract. Unfortunately, we can not assume all equivalent extracts
are CSE'd, since InstCombine itself may create an extract which is a duplicate
of an existing one. This extends it to handle several distinct extracts from
the same index.

This should fix at least some of the  performance regressions from PR27988.

Differential Revision: http://reviews.llvm.org/D20983

llvm-svn: 271961
2016-06-06 23:38:33 +00:00
..
agg-interleave-a2.ll
large-loop-rdx.ll
lit.local.cfg
small-loop-rdx.ll
stride-vectorization.ll
vectorize-only-for-real.ll [LoopVectorize] Don't vectorize loops when everything will be scalarized 2016-03-30 19:37:08 +00:00
vsx-tsvc-s173.ll [InstCombine] scalarizePHI should not assume the code it sees has been CSE'd 2016-06-06 23:38:33 +00:00