#14672: Make likelyhood of branches/conditions available throughout the compiler. -------------------------------------+------------------------------------- Reporter: AndreasK | Owner: (none) Type: task | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.2 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Phab:D4316 Wiki Page: | Phab:D4324 Phab:D4327 -------------------------------------+------------------------------------- Comment (by AndreasK): Replying to [comment:8 svenpanne]:
Just a general remark: [https://phabricator.haskell.org/D4327 D4327] talks about annotations as a source of likelihood values, which is fine, but having actual data from previous runs, i.e. using profile-guided optimization, would probably have much more potential.
I was talking about annotations primarily because they seem like the low hanging fruit. As far as the native codegen is concerned it makes no difference if the data comes from user suggestions or actual measurements. Having good data on all code paths would certainly be better then just an estimate for some paths the user annotated. Once we have a way to pass/use the information throughout the passes adding PGO is the next logical step and should be possible without touching anything past core. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/14672#comment:10 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler