fix custom call-site performance regression#5814
Merged
Conversation
builtin check done when the entry is set for the expected target
Member
Author
|
and here's some raw numbers :
JRuby 9.2.7.0 master this branch things are much more interesting, when Fixnum gets re-opened (fairly common in gems and the whole point of doing #4736 in the first place) :
JRuby 9.2.7.0 this branch (re-opening has no effect) |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
restores interpreter 'fast' Fixnum/Float operations performance, which degraded after doing the
isBuiltinchecks (#4736)there's a bit of refactoring needed for the CachingCallSite API used
we're adding a
CacheEntry setCache(CacheEntry entry, IRubyObject self)and checks such as
isBuiltin(RubyClass)change to ->isBuitlin(IRubyObject self)... this is the way JRuby intends to use these APIs anyway, so no harm done there
the gist of better performance is adding logic in custom sites when
setCachehappens,do the builtin and target check there so that on the hot-path, taking the direct specific (Fixnum/Float) branch becomes 2
instanceofchecks (edit: plus the obvious type "token" check)