diff options
| author | Alexandros Lamprineas <alexandros.lamprineas@arm.com> | 2018-12-06 16:11:58 +0000 |
|---|---|---|
| committer | Alexandros Lamprineas <alexandros.lamprineas@arm.com> | 2018-12-06 16:11:58 +0000 |
| commit | e4c91f5c4c52cd36b8238d29b933f2866ea0d735 (patch) | |
| tree | f7260782da76417bfab8b49bba4cacd5d3533d4e /llvm/lib/DebugInfo/PDB/Native/ModuleDebugStream.cpp | |
| parent | 64ad0ad5ed92fc9881075cffebce418d05f61288 (diff) | |
| download | llvm-e4c91f5c4c52cd36b8238d29b933f2866ea0d735.zip llvm-e4c91f5c4c52cd36b8238d29b933f2866ea0d735.tar.gz llvm-e4c91f5c4c52cd36b8238d29b933f2866ea0d735.tar.bz2 | |
[GVN] Don't perform scalar PRE on GEPs
Partial Redundancy Elimination of GEPs prevents CodeGenPrepare from
sinking the addressing mode computation of memory instructions back
to its uses. The problem comes from the insertion of PHIs, which
confuse CGP and make it bail.
I've autogenerated the check lines of an existing test and added a
store instruction to demonstrate the motivation behind this change.
The store is now using the gep instead of a phi.
Differential Revision: https://reviews.llvm.org/D55009
llvm-svn: 348496
Diffstat (limited to 'llvm/lib/DebugInfo/PDB/Native/ModuleDebugStream.cpp')
0 files changed, 0 insertions, 0 deletions