|
11 | 11 | ## [Unreleased] |
12 | 12 |
|
13 | 13 |
|
| 14 | + |
| 15 | + |
| 16 | + |
| 17 | + |
| 18 | +<br> |
| 19 | + |
| 20 | +## [0.6.0] - 2025-08-29 |
14 | 21 | ### Added |
15 | 22 | - Scheduled perf workflow: `.github/workflows/perf.yml` |
16 | 23 | - Runs twice weekly (Sun/Wed 02:00 UTC) and via manual dispatch |
|
28 | 35 | - CI perf workflow restricted to `main` and protected with concurrency guard |
29 | 36 | - Replaced `#[inline(always)]` with `#[inline]` for hot-path methods in `src/histogram.rs` |
30 | 37 | - Resolves `clippy::inline_always` under `-D warnings` while preserving optimizer freedom |
| 38 | +- `docs/API.md`: moved the Histogram section under “Types → Stats → Histogram” and added a runnable example with feature-gating notes |
| 39 | +- `docs/features/README.md`: clarified `--no-default-features` usage comment and guidance to opt back in with `-F benchmark` / `-F metrics` |
| 40 | +- `docs/BENCHMARK.md` and `docs/METRICS.md`: added consistent feature-gating notes near the top |
| 41 | +- `docs/features/BENCHMARK.md` and `docs/features/METRICS.md`: added consistent feature-gating notes; fixed header label in Metrics page |
| 42 | + |
| 43 | +### Documentation |
| 44 | +- Doc sweep for clarity and consistency on feature gates across Benchmark and Metrics docs |
| 45 | +- Emphasized opt-in perf gates: `perf-tests` with `PERF_TESTS=1` |
| 46 | +- Clarified and emphasized that histogram percentile inputs are clamped to [0.0, 1.0] across `README.md` and `docs/API.md`. Out-of-range inputs map to min/max. |
| 47 | + |
| 48 | +### Fixed |
| 49 | +- Clippy lints in benches: |
| 50 | + - `benches/timers.rs`: use assign-op `+=` pattern |
| 51 | + - `benches/histogram_hot.rs`: removed unnecessary casts |
| 52 | +- Added explicit ignore reasons for perf-gated tests/benches in `src/histogram.rs` |
| 53 | +- Percentile interpolation overflow in `src/histogram.rs` when handling large bucket widths/counts: switched to u128 intermediates to prevent overflow. |
| 54 | +- Ensured percentile 1.0 returns the true maximum in `Histogram::percentiles()` (explicit post-pass to set p==1.0 to max), aligning with `percentile(1.0)` behavior. |
| 55 | + |
| 56 | +### Maintenance |
| 57 | +- CI locally: `cargo fmt --all -- --check`, `cargo clippy --all-targets --all-features`, `cargo test --all-features`, `cargo doc --no-deps` |
| 58 | + |
| 59 | + |
31 | 60 |
|
32 | 61 |
|
33 | 62 | <br> |
@@ -202,7 +231,9 @@ Initial pre-dev release for backup. |
202 | 231 |
|
203 | 232 |
|
204 | 233 |
|
205 | | -[Unreleased]: https://github.com/jamesgober/rust-benchmark/compare/v0.5.8...HEAD |
| 234 | +[Unreleased]: https://github.com/jamesgober/rust-benchmark/compare/v0.6.0...HEAD |
| 235 | +[0.8.0]: https://github.com/jamesgober/rust-benchmark/compare/v0.7.0...v0.8.0 |
| 236 | +[0.7.0]: https://github.com/jamesgober/rust-benchmark/compare/v0.6.0...v0.7.0 |
206 | 237 | [0.6.0]: https://github.com/jamesgober/rust-benchmark/compare/v0.5.8...v0.6.0 |
207 | 238 | [0.5.8]: https://github.com/jamesgober/rust-benchmark/compare/v0.5.7...v0.5.8 |
208 | 239 | [0.5.7]: https://github.com/jamesgober/rust-benchmark/compare/v0.5.0...v0.5.7 |
|
0 commit comments