Inline assembly is crucial for many performance-critical applications and low-level programming. However, its compiler support in LLVM and Clang is complicated and mostly undocumented. In this talk, we’re going to give an overview on how inline assembly is processed in Clang / LLVM and show concrete examples of how to implement target-specific inline assembly handling logic. We hope that this can provide a clear guideline for developers to add inline assembly support in future backends and targets.