aio-libs / multidict / 37203551251

86%

push

github

Move md_pop_item() out of line (#1671)

<!-- Thank you for your contribution! -->

## What do these changes do?

Marks `md_pop_item()` as `NOINLINE`, so the body of `popitem()` stays
out of its slot `multidict_popitem()`. The method is rarely called;
keeping it out of line keeps the source simple and the slot small, at
the cost of one call.

Measured with `benchmarks/callgrind_driver.py --include-multidict-only`
on 3.14.7 and 3.14.7t: only the `popitem` row moved, by the call
overhead. GIL: MultiDict 485.4 to 494.4 Ir (+1.9%), CIMultiDict 1089.9
to 1098.9 (+0.8%). FT: MultiDict 540.0 to 549.8 (+1.8%), CIMultiDict
1159.2 to 1166.0 (+0.6%). Every other row was unchanged.

## Are there changes in behavior for the user?

No.

## Is it a substantial burden for the maintainers to support this?

No.

## Related issue number

N/A

## Checklist

- [x] I think the code is well written
- [ ] Unit tests for the changes exist: N/A, no behaviour change;
covered by the existing `popitem()` tests
- [ ] Documentation reflects the changes: N/A
- [ ] If you provide code modification, please add yourself to
`CONTRIBUTORS.txt`: N/A
- [x] Add a new news fragment into the `CHANGES/` folder
- [x] `make doc-spelling` passes and any new technical words are added
to `docs/spelling_wordlist.txt`

<details>
<summary>Agent run details (optional, for reviewers)</summary>

Tests: C extension 3.14.7 4077 passed, 3.14.7t 4062 passed, 3.10.17 4081
passed; pure Python (`MULTIDICT_NO_EXTENSIONS=1 --no-c-extensions`)
3.14.7 1604 passed.
Inlining: `tools/check_inlining.py` ok on 3.14.7 (16 rules) and 3.14.7t
(18 rules); `htkeysiter_init` and `_md_unlink_at` stay inlined in
`md_pop_item()`.
Spelling: `make doc-spelling` ok.
</details>

Drafted with Claude Code (Claude Opus 5.5); reviewed by @asvetlov.

792 of 1584 branches covered (50.0%)

Branch coverage included in aggregate %.

9315 of 10167 relevant lines covered (91.62%)

1.83 hits per line