Improve adaptive allocator thread local performance (#15741)#16107
Merged
Conversation
Motivation: Adaptive allocator perform costly atomic operations in the thread local path, which reduce its performance Modification: Reduce the amount of atomic operations in the thread local allocation's fast path Result: Fixes netty#15571 These are the different variations I want to test: - [x] Uses unguarded `Recycler`s - [x] Implements "compressed" local free list (LIFO) - [x] Use a mpsc q for the reuse chunk q in the thread-local case **NO VISIBLE IMPROVEMENTS** - [x] Guards `nextInLine`'s `getAndSet` with a null check via volatile `get` first, since size classed chunks rarely end up into `nextInLine` (i.e. which is mostly `null`) **NO VISIBLE IMPROVEMENTS** - [x] Implements a var handle based `MpscIntQueue` (done at 1c4e1e4) **NO VISIBLE IMPROVEMENTS** - [x] Remove the live/raw ref cnt as mentioned at netty#15736 (comment) - [ ] Remove the ref count for size classed chunks (see 8953bbe and 8cb1bf0) - [ ] Use the "pinned" Recycler instead of the `FastThreadLocal`-based one (cherry picked from commit accd981)
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Motivation:
Adaptive allocator perform costly atomic operations in the thread local path, which reduce its performance
Modification:
Reduce the amount of atomic operations in the thread local allocation's fast path
Result:
Fixes #15571
These are the different variations I want to test:
RecyclersnextInLine'sgetAndSetwith a null check via volatilegetfirst, since size classed chunks rarely end up intonextInLine(i.e. which is mostlynull)NO VISIBLE IMPROVEMENTS
MpscIntQueue(done at 1c4e1e4)NO VISIBLE IMPROVEMENTS
8cb1bf0)
FastThreadLocal-based one(cherry picked from commit accd981)