diff options
| author | Slava Zakharin <szakharin@nvidia.com> | 2022-10-03 16:54:12 -0700 |
|---|---|---|
| committer | Slava Zakharin <szakharin@nvidia.com> | 2022-10-04 13:28:03 -0700 |
| commit | 42fead6834ef3a96c765ea545c2d2bac951f7e98 (patch) | |
| tree | ca157a7710f73a953ae13a4879c55c575160b864 /llvm/lib/Transforms/Utils/LoopVersioning.cpp | |
| parent | 4769976c49be468d7629d513080e6959a25adcfe (diff) | |
| download | llvm-42fead6834ef3a96c765ea545c2d2bac951f7e98.zip llvm-42fead6834ef3a96c765ea545c2d2bac951f7e98.tar.gz llvm-42fead6834ef3a96c765ea545c2d2bac951f7e98.tar.bz2 | |
[flang][tco] Engineering option for running only CodeGen passes.
This option allows running only CodeGen passes and then translating
FIR to LLVM IR. I am using it to fetch optimized FIR after bbc,
hand-modify it and then produce LLVM IR that can be fed to clang.
Diffstat (limited to 'llvm/lib/Transforms/Utils/LoopVersioning.cpp')
0 files changed, 0 insertions, 0 deletions