News for hackers. Stuff that matters.
Re:PyPy 2.0 Released by MichailP on Thursday May 09, 2013 @04:40PM (#5682811)
For example does PyPy always warm up, or can it store "warmed up" version of some function (where type of objects is inferred)? I know that this is not in accordance with highly dynamic nature of Python, but not all functions are highly dynamic.
Re:PyPy 2.0 Released by rcthompson on Thursday May 09, 2013 @07:46PM (#5683676)
My understanding (from a previous HN discussion about javascript) is that since the output from a "just-in-time compiler" is actually machine code generated on the fly, it includes direct references to memory locations that are only valid for the lifetime of the process. So the output of a JIT is simply not in a format that can be saved and reloaded later.
Re:PyPy 2.0 Released by nickik on Friday May 10, 2013 @10:30AM (#5686375)
Actually its possible to have a JIT like that but it is uncommon.
Re:PyPy 2.0 Released by cglace on Thursday May 09, 2013 @04:46PM (#5682847)
Do you mean can you save the state of the JIT for future use?
Re:PyPy 2.0 Released by rubinelli on Thursday May 09, 2013 @05:08PM (#5682955)
Yeah. Can you? Considering CPython already generates .pyc and checks timestamps, keeping some kind of JIT cache around wouldn't be a stretch.
Re:PyPy 2.0 Released by kingkilr on Thursday May 09, 2013 @05:39PM (#5683138)
No, the state of the JIT cannot be saved between executions.
Re:PyPy 2.0 Released by gcr on Thursday May 09, 2013 @09:36PM (#5684043)
I can't speak for python, but I do know that the impact of translating code for Racket's JIT (provided by GNU lightning, not LLVM) is negligable (ie. takes less time than the time you gain by JITting