Inlining in the Glasgow Haskell Compiler:Empirical Investigation and Improvement

5 matt_d 1 7/22/2025, 10:05:54 PM era.ed.ac.uk โ†—

Comments (1)

internet_points ยท 9h ago
Wow. They really did the work, profiling and exploring by search, training GA/ML, ending up with what seems like a usable solution. Fingers crossed this is something that can end up in GHC (if the aptly named section 6.6.6 "Effect On Binary Size and Compilation Time" doesn't end up being too bad in practice)