Code Quality - inlining #792
Merged
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
This commit adds some inlining requirements on lambda functions used on the fast path for the allocations path. This improves the g++ code generation by preventing the stack allocation of the lambda which is then not optimised away.
Contributor
There was a problem hiding this comment.
Pull Request Overview
This PR adds SNMALLOC_FAST_PATH_LAMBDA to various lambdas to encourage inlining on critical paths and comments out fork-prevention logic to remove a performance bottleneck during remote allocator posting.
- Annotated several lambda definitions with the
SNMALLOC_FAST_PATH_LAMBDAattribute. - Disabled the
PreventForkguard infreelist_queue.hand expanded the comment explaining why. - Updated comments to describe alternative post-fork handling without enabling the fork-prevention code.
Reviewed Changes
Copilot reviewed 3 out of 3 changed files in this pull request and generated 3 comments.
| File | Description |
|---|---|
| src/snmalloc/mem/remotecache.h | Added SNMALLOC_FAST_PATH_LAMBDA to a forwarding lambda for inlining. |
| src/snmalloc/mem/freelist_queue.h | Commented out PreventFork usage and expanded explanatory comments. |
| src/snmalloc/mem/corealloc.h | Annotated multiple lambdas with SNMALLOC_FAST_PATH_LAMBDA for inlining. |
src/snmalloc/mem/freelist_queue.h
Outdated
| // // with the following code, but it is not currently enabled as it | ||
| // // has negative performance impact. | ||
| // // An alternative would be to reset the queue on the child postfork | ||
| // // handler to ensure that the queue has not be blackholed. |
There was a problem hiding this comment.
Fix grammar: change 'has not be blackholed' to 'has not been blackholed'.
Suggested change
| // // handler to ensure that the queue has not be blackholed. | |
| // // handler to ensure that the queue has not been blackholed. |
SchrodingerZhu
approved these changes
Jul 9, 2025
Member
Author
|
This was running on an Azure Linux 3 instance with clang++ (18.1.8) and g++ (13.2.0) |
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.
This PR adds some inlining attributes to lambdas on the fast path. Some versions of g++ and clang++ were not inlining as much as expected, and leading to poor codegen. This PR fixes that.
Additionally, it disables the fork prevention during remote allocator posting. This was visible in some traces as causing a bottleneck.