Skip to content

eckertliam/inline-ml

Repository files navigation

Inline-ML

Machine Learning-Guided Inlining for LLVM

Inline-ML is a research prototype that applies machine learning to LLVM IR inlining decisions. Originally created for a capstone project, it now serves as a standalone CLI tool that explores data-driven compiler optimization. The system learns from real-world C codebases to predict when inlining should occur, using interpretable models like XGBoost.

About

Predicting function inlining using LLVM IR + XGBoost

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published