Record recent benchmark results and add AGENTS.md - #19
Merged
Merged
Conversation
The two measured wins from late in the Jackson 3 rewrite (write +8.8% from one wrap helper, read allocation 2056 to 1912 B/op from not boxing integers), and a back-to-back comparison of the rewrite against main after the map-key and UUID changes.
Commands, and the rules this repository works by: round-trip safety first, no Jackson 2 compatibility goal, failure output is the caller's to discard, failing-first tests, long back-to-back benchmarks with a reused mapper, results in jmh/results, and Copilot reviews.
There was a problem hiding this comment.
Copilot review overview
🟢 Approval recommended
Only minor benchmark reproducibility metadata nits remain; no blocking issues were identified.
Review effort: Lite
Findings: 2
Open (2)
What changed in this PR
Documents recent benchmark results and adds repository guidance for coding agents.
Changes:
- Adds three benchmark result records.
- Adds
AGENTS.mdwith build, testing, benchmarking, and design guidance.
| File | Description |
|---|---|
jmh/results/2026-09-27-pr9-vs-main.md |
Compares PR #9 performance with current main. |
jmh/results/2026-09-23-integer-without-boxing.md |
Records allocation benchmark results. |
jmh/results/2026-09-22-write-wrap-helper.md |
Records write benchmark results. |
AGENTS.md |
Documents repository development practices. |
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Documentation only.
Benchmark results
Three files in
jmh/results/, the first entries since2026-09-21-single-pass-short-strings.md:2026-09-22-write-wrap-helper.md: writePojoMsgpack +8.8% from moving the try/catch ofwriteName,openContainerandcloseContainerinto onepack()helper (f8695a4).2026-09-23-integer-without-boxing.md: readPojoMsgpack allocation 2056 to 1912 B/op from keeping integers in primitive fields (1037af9).2026-09-27-pr9-vs-main.md: PR Rewrite for Jackson 3 with a native MessagePack encoder and decoder #9 against current main after Record how a string token reads as binary #10, Remove MessagePackKeySerializer #16 and Write a UUID value as its string #17, run back to back. POJO read and write are on par; the one string benchmark outside the error bars moved together with the JSON control, so it is recorded as inconclusive.For the two older runs the exact JMH options were not recorded, and the files say so.
AGENTS.md
Guidance for coding agents: the build, test, soak and benchmark commands, and the rules this repository has settled on (round-trip safety first, no Jackson 2 compatibility goal, failure output is the caller's to discard, failing-first tests, long back-to-back benchmarks with a reused mapper, results in
jmh/results/, Copilot reviews). It points todocs/DESIGN.mdfor the design.