Inside CPython 3.11's new specializing, adaptive interpreter
summary
Covers CPython 3.11’s new specializing, adaptive interpreter that speeds Python by about 25% on average. Explains how the interpreter observes running code and applies specialization/adaptation techniques to optimize execution, outlines challenges of optimizing a dynamic language, and shows tools to observe the interpreter and planned follow-up improvements for Python 3.12 and later.
AI-written summary · about
chapters (9)
- 0:00Start
- 3:32Quickening
- 4:25Superinstructions
- 4:56Adaptive Instructions
- 6:43LOAD_ATTR_INSTANCE_VALUE
- 13:45Inline Caches
- 15:31Source Code
- 15:51Specialist
- 17:03CPython 3.12
related talks
01h 03mA tour of CPython's runtime
24mHow two undergrads from the other side of the planet are speeding up...
25mHow two teams are working together to make Python better for all
29mWhat they don't tell you about building a JIT compiler for CPython
32mWriting performant code for modern Python interpreters
38mDynamic Code Instrumentation with Hacked Interpreters
42mOptimizing Code Performance for Python Internals
49mThe Evolution of Python: Progress or Dead End