diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2025-10-07 14:34:14 +0900 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-10-07 14:34:14 +0900 |
| commit | ff394cd0f675cefeaf8438e43eda143a8587ae57 (patch) | |
| tree | 3106394b9fb9d2c9c712f4b6ac38fbc9f84dbc98 /llvm/examples/OptSubcommand/llvm-hello-sub.cpp | |
| parent | 3a3a524f54edbd59d36006614d626e25fbb5e89e (diff) | |
| download | llvm-ff394cd0f675cefeaf8438e43eda143a8587ae57.zip llvm-ff394cd0f675cefeaf8438e43eda143a8587ae57.tar.gz llvm-ff394cd0f675cefeaf8438e43eda143a8587ae57.tar.bz2 | |
llvm-reduce: Add new pass to inline call sites (#134223)
Added a primitive heuristic to avoid blowing up the code size
which could use more thought.
This helps cleanup some basic examples I've been looking at where
there is a worse result when just running a bug through the full
optimization pipeline vs. running just a single pass at the point
of failure.
Diffstat (limited to 'llvm/examples/OptSubcommand/llvm-hello-sub.cpp')
0 files changed, 0 insertions, 0 deletions