Swap allocated_size and total_size to prevent cache misses on fast path. #13834
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
![copybara-service[bot]](https://avatars.githubusercontent.com/in/44061?s=60&v=4){kind=small}
Note: I don't know how to simulate cache miss reliably in a microbenchmark, so the change does not show any significant diff on the added BMs. Motivation: 1. It's more aligned with class design: total_size is the capacity of Rep, so it makes more sense to store it there. 2. On the fast path, we check allocated_size, and with this change we do not to dereference Rep to read it (slow path is unchanged, as it needs both capacity and allocated size). 3. fast path is inlined and as a result we reduce code bloat. Also added removed redundant check with MaybeExtend in some methods: these methods already know that call to extend is necessary, so no need to go through `Reserve` route. nit: deduped code by reusing AddAllocatedForParse in other Add functions. PiperOrigin-RevId: 562501058
|
Auto-closing Copybara pull request |
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Outdated suggestions cannot be applied.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.
Swap allocated_size and total_size to prevent cache misses on fast path.
Note: I don't know how to simulate cache miss reliably in a microbenchmark, so the change does not show any significant diff on the added BMs.
Motivation:
Also added removed redundant check with MaybeExtend in some methods: these methods already know that call to extend is necessary, so no need to go through
Reserveroute.nit: deduped code by reusing AddAllocatedForParse in other Add functions.