diff options
| author | Ricardo Jesus <rjj@nvidia.com> | 2024-07-26 10:12:25 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-07-26 10:12:25 +0100 |
| commit | 25da8e5a9798a1c25526e4dc8db571fc25cca187 (patch) | |
| tree | 2b8b149c2a88d3bd3c543adfb95aa83459e70cd0 /flang/lib/Frontend/CompilerInvocation.cpp | |
| parent | 7a3bc44c89b675b5fda7e90cf59e74d54778917f (diff) | |
| download | llvm-25da8e5a9798a1c25526e4dc8db571fc25cca187.zip llvm-25da8e5a9798a1c25526e4dc8db571fc25cca187.tar.gz llvm-25da8e5a9798a1c25526e4dc8db571fc25cca187.tar.bz2 | |
Reapply "[LICM] Fold associative binary ops to promote code hoisting (#81608)" (#100377)
This reapplies a more strict version of
https://github.com/llvm/llvm-project/commit/f2ccf80136a01ca69f766becafb329db6c54c0c8.
Perform the transformation
"(LV op C1) op C2" ==> "LV op (C1 op C2)"
where op is an associative binary op, LV is a loop variant, and C1 and
C2 are loop invariants, and hoist (C1 op C2) into the preheader.
For now this fold is restricted to ADDs.
Diffstat (limited to 'flang/lib/Frontend/CompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions