Perf/dlmalloc guest allocator - #869
Open
jotabulacios wants to merge 16 commits into
Open
Conversation
Collaborator
|
/bench |
Benchmark — real block (
|
| Metric | main | PR | Δ |
|---|---|---|---|
| Peak heap | 52060 MB | 52560 MB | +500 MB (+1.0%) ⚪ |
| Prove time | 196.587s | 154.176s | -42.411s (-21.6%) 🟢 |
🎉 Improvement on the real block — prove time down 21.6%.
Prove-time spread 2.7% (154.176s / 157.986s / 153.831s)
Commit: b525057 · Baseline: cached · Runner: self-hosted bench
Collaborator
|
/bench |
jotabulacios
marked this pull request as ready for review
July 29, 2026 13:51
…nto perf/dlmalloc-guest-allocator
Contributor
|
/bench |
Contributor
|
/bench |
Contributor
|
/bench-gpu |
GPU Benchmark (ABBA) —
|
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The guest allocator moves from embedded-alloc's TLSF heap to dlmalloc, with TLSF kept one
flag away (
tlsf-alloc) for fallback. dlmalloc sits on a bump provider that hands itpage-aligned chunks of the guest heap, so it still reuses freed memory — no OOM — while
each alloc and free costs fewer guest instructions than TLSF.
Isolated A/B on the same tree, only the feature differs: guest cycles drop 11.2% on a
10-transfer ethrex block (1,702,365 → 1,511,225) and 13.1% on a 20-transfer one
(3,915,738 → 3,404,649). In the guest profile the allocator goes from 24.4% of cycles to
9.5%.