Port 1.9 GC heuristics to v1.12.1+RAI #261
Closed
Conversation
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
* Port pool stats to 1.12 * Increment/decrement current_pg_count --------- Co-authored-by: K Pamnany <kpamnany@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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Outdated suggestions cannot be applied.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.
Summary
This PR ports the 1.9 GC heuristics code (also used by 1.10.2+RAI) to 1.12.1+RAI. For context: Julia 1.12 uses a variant of MemBalancer with a few guardrails added by this PR.
Correctness Assessment
I was able to build Julia with this patch. I also ran the GCBenchmarks to assess correctness and didn't observe any correctness bugs.
Performance Assessment
I ran the GCBenchmarks with this PR, and compared its performance with our branch of Julia 1.10 (1.10.2+RAI). It's not an apples-to-apples comparison, but it should let us check if GC performance hasn't regressed significantly between these two versions.
I summarized GC time for each benchmark in the table below. All benchmarks were run with a single mutator and GC thread:
(TODO).