Nuances of a Tracing JIT Compiler
summary
Describes practical experience using Lua and LuaJIT for business logic, explains why the team forked LuaJIT due to architectural limits, and discusses details of how a tracing JIT compiler works and was adapted in their custom fork for production use.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 7:37JIT compiler issues
- 13:17JIT compiler analysis
- 17:10Improving analysis tools
- 20:45Tracing optimization
- 25:00Table indexing problem
- 31:07Loop and side-trace issues
- 32:56Summary
- 35:33Q&A
related talks
48mHow we work on the stability of our Lua implementation
40mJoy of Meta-Tracing Just-in-Time Compilation: More Than Just a VM Generator
01h 04mJust-in-Time Compilation: The Next Big Thing?
30mAn Incremental Approach to JIT Construction
19mReusing Just-in-Time Compiled Code
46mIR JIT Framework: the base for the next generation of JIT in PHP
44mJIT Me Baby One More Time
45mTwo-level Just-in-Time Compilation with One Interpreter and One Engine