diff options
| author | zhongyunde <zhongyunde@huawei.com> | 2023-06-07 21:50:54 +0800 |
|---|---|---|
| committer | zhongyunde <zhongyunde@huawei.com> | 2023-06-07 22:08:29 +0800 |
| commit | df19d872277fcb28344e3f1e35cb5d960d506bc4 (patch) | |
| tree | 4e51edd103b0a6347f93d041fc4576bb3c36bd11 /clang/lib/Frontend/CompilerInvocation.cpp | |
| parent | e60b30d5e3878e7d91f8872ec4c4dca00d4a2dfc (diff) | |
| download | llvm-df19d872277fcb28344e3f1e35cb5d960d506bc4.zip llvm-df19d872277fcb28344e3f1e35cb5d960d506bc4.tar.gz llvm-df19d872277fcb28344e3f1e35cb5d960d506bc4.tar.bz2 | |
[LV] Add option to tune the cost model, NFC
For Neon, the default nonconst stride cost is conservative,
and it is a local variable, which is not convenience to
to tune the loop vectorize.
So I try to use a option, which is similar to SVEGatherOverhead brought in D115143.
Fix https://github.com/llvm/llvm-project/issues/63082.
Reviewed By: dmgreen, fhahn
Differential Revision: https://reviews.llvm.org/D152253
Diffstat (limited to 'clang/lib/Frontend/CompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions