PyPy hands-on
summary
PyPy presents a fast Python interpreter plus a language-development framework; benchmarks claim up to 100x speedups over CPython on some workloads and typical cases often run several times faster (example shown with a 10 million-iteration loop). It uses a tracing JIT that records hot loops, emits machine code with guards and bailouts, and targets x86 and ARM. The hands-on covers running apps under PyPy, using the jitviewer log file to inspect traces and assembly, and turning a toy language interpreter into a JIT-friendly, high-performance implementation.
AI-written summary · about
chapters (8)
- 0:00Welcome to PyPy
- 1:10What PyPy is and performance
- 2:56Training setup
- 4:22Run buggy Python on PyPy
- 6:12How PyPy's JIT works
- 9:46JIT demo with simple program
- 13:02Inspect JIT code with tools
- 20:12Hands-on prep
related talks
57mPython and PyPy performance (not) for dummies
22mPyPy and the future of the Python ecosystem
42mThe state of PyPy
48mPython Performance: Past, Present and Future
58mPython performance Past, Present, Future
17mInside a PyPy JIT: is that even legal?
39mCPython, Grumpy, PyPy - When, How, Why
26mWhy are Python and Ruby slow?