diff options
| author | Roman Lebedev <lebedev.ri@gmail.com> | 2018-07-09 19:06:42 +0000 |
|---|---|---|
| committer | Roman Lebedev <lebedev.ri@gmail.com> | 2018-07-09 19:06:42 +0000 |
| commit | 5ccae1750b26871670015755f2d2f1acd4860cb7 (patch) | |
| tree | 269c2d892dcba4907b3862e286146bf4d96ed295 /llvm/lib/Target/AMDGPU/AMDGPUAlwaysInlinePass.cpp | |
| parent | 22a0c8dbc8bc008928b319d44ad296738940fe72 (diff) | |
| download | bcm5719-llvm-5ccae1750b26871670015755f2d2f1acd4860cb7.tar.gz bcm5719-llvm-5ccae1750b26871670015755f2d2f1acd4860cb7.zip | |
[X86][TLI] DAGCombine: Unfold variable bit-clearing mask to two shifts.
Summary:
This adds a reverse transform for the instcombine canonicalizations
that were added in D47980, D47981.
As discussed later, that was worse at least for the code size,
and potentially for the performance, too.
https://rise4fun.com/Alive/Zmpl
Reviewers: craig.topper, RKSimon, spatel
Reviewed By: spatel
Subscribers: reames, llvm-commits
Differential Revision: https://reviews.llvm.org/D48768
llvm-svn: 336585
Diffstat (limited to 'llvm/lib/Target/AMDGPU/AMDGPUAlwaysInlinePass.cpp')
0 files changed, 0 insertions, 0 deletions
