News for hackers. Stuff that matters.
Re:Ask HN: Does a JIT typically use machine learning to boos... by PaulHoule on Thursday February 09, 2023 @10:25PM (#34735237)
Machine learning for compilers is a big research area, this query turns up documents on that

https://arxiv.org/search/?query=machine+learning+compiler+op...

and also documents on compilers for machine language programs.

An advanced optimizing compiler can work very hard to optimize the code, particularly with "profile guided optimization" or having the compiler generate a large number of candidate binaries and choosing the best. You'd better believe people treat this as a mathematical "optimization" problem and try out the most powerful tools.

The JIT, however, runs every time the program starts up for every user and if the JIT is not fast the delay becomes untolerable. Now the JIT does have the advantage that it can do profile-guided optimization based on tier0 profiling but whatever it does has to be fast so in that sense it must use simple algorithms. As that article points out the developers of a commercial runtime like .NET are very concerned about optimizing the optimization process.

There are no comments on this comment yet.