Use match_shared in the main eval loop #1698
Merged
Conversation
This file contains 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
Radvendii
reviewed
Oct 25, 2023
There was a problem hiding this comment.
There are a bunch of places where you've removed clone() calls because now we're taking ownership, but there are other places where the clone() calls have stuck around, and I'm not sure if that's because something different is happening there or because you missed them.
I only marked a couple of them, but saw a bunch more.
|
Good catch, I thought I had them all, but did an additional search and fix and got a few more. Now there shouldn't be remaining undue clones in |
yannham
force-pushed
the
optimize/eval-matche-shared
branch
from
020b712 to
39bad0d
Compare
Radvendii
approved these changes
Oct 25, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
A long standing todo was to use
match_sharedinside the main eval loop, such that some operations (typically array and record closurization) don't have to unduly copy data if they can take ownership on a 1-countedRc. We were just waiting for nice formatting which has landed in #1677.It doesn't seem to have a huge effect on #1622 (the effect is expected to be more salient on either small programs or programs with huge arrays and records), in one direction or another, and the benchmarks are way too volatile right now to make any sense of them (it's an important, but a separate one). If anything, this PR improves the documentation and the signature of various evaluation functions by returning closures instead of tuples.