|
| 1 | +# 2026-02-17 Triage Log |
| 2 | + |
| 3 | +Several pull requests introduced (usually very small) regressions across the board this week. On the |
| 4 | +other hand, [#151380](https://github.com/rust-lang/rust/pull/151380) provided a nice performance win in the inference engine. |
| 5 | +I would also like to bring attention to [#152375](https://github.com/rust-lang/rust/pull/152375), |
| 6 | +which improved the parallel frontend. It is not shown in this report, because we don't yet have |
| 7 | +many benchmarks for the parallel frontend, but this PR seemingly improved the `check` (wall-time) |
| 8 | +performance with multiple frontend threads on several real-world crates by 5-10%! |
| 9 | + |
| 10 | +Triage done by **@kobzol**. |
| 11 | +Revision range: [39219ceb..3c9faa0d](https://perf.rust-lang.org/?start=39219ceb97d1b37dda72517daa9ebe8364ffe186&end=3c9faa0d037b9eecda4a440cc482ff7f960fb8a5&absolute=false&stat=instructions%3Au) |
| 12 | + |
| 13 | +**Summary**: |
| 14 | + |
| 15 | +| (instructions:u) | mean | range | count | |
| 16 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 17 | +| Regressions ❌ <br /> (primary) | 0.7% | [0.2%, 3.1%] | 96 | |
| 18 | +| Regressions ❌ <br /> (secondary) | 1.1% | [0.0%, 5.7%] | 62 | |
| 19 | +| Improvements ✅ <br /> (primary) | -0.4% | [-0.9%, -0.2%] | 8 | |
| 20 | +| Improvements ✅ <br /> (secondary) | -2.6% | [-7.0%, -0.0%] | 45 | |
| 21 | +| All ❌✅ (primary) | 0.6% | [-0.9%, 3.1%] | 104 | |
| 22 | + |
| 23 | + |
| 24 | +2 Regressions, 0 Improvements, 9 Mixed; 4 of them in rollups |
| 25 | +36 artifact comparisons made in total |
| 26 | + |
| 27 | +#### Regressions |
| 28 | + |
| 29 | +Rollup of 17 pull requests [#152562](https://github.com/rust-lang/rust/pull/152562) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=47611e16044c68ef27bac31c35fda2ba1dc20b73&end=d7daac06d87e1252d10eaa44960164faac46beff&stat=instructions:u) |
| 30 | + |
| 31 | +| (instructions:u) | mean | range | count | |
| 32 | +|:----------------------------------:|:----:|:------------:|:-----:| |
| 33 | +| Regressions ❌ <br /> (primary) | - | - | 0 | |
| 34 | +| Regressions ❌ <br /> (secondary) | 0.3% | [0.2%, 0.5%] | 11 | |
| 35 | +| Improvements ✅ <br /> (primary) | - | - | 0 | |
| 36 | +| Improvements ✅ <br /> (secondary) | - | - | 0 | |
| 37 | +| All ❌✅ (primary) | - | - | 0 | |
| 38 | + |
| 39 | +- Noise that got reverted in a follow-up PR. |
| 40 | +- Already marked as triaged. |
| 41 | + |
| 42 | +replace box_new with lower-level intrinsics [#148190](https://github.com/rust-lang/rust/pull/148190) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=71e00273c0921e1bc850ae8cc4161fbb44cfa848&end=3c9faa0d037b9eecda4a440cc482ff7f960fb8a5&stat=instructions:u) |
| 43 | + |
| 44 | +| (instructions:u) | mean | range | count | |
| 45 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 46 | +| Regressions ❌ <br /> (primary) | 0.8% | [0.5%, 1.9%] | 8 | |
| 47 | +| Regressions ❌ <br /> (secondary) | 1.8% | [0.4%, 5.5%] | 24 | |
| 48 | +| Improvements ✅ <br /> (primary) | -0.8% | [-0.8%, -0.8%] | 1 | |
| 49 | +| Improvements ✅ <br /> (secondary) | -0.3% | [-0.3%, -0.3%] | 1 | |
| 50 | +| All ❌✅ (primary) | 0.6% | [-0.8%, 1.9%] | 9 | |
| 51 | + |
| 52 | +- Small performance hit across the board caused by refactoring how Box works internally. |
| 53 | +- Was deemed acceptable pre-merge. |
| 54 | +- Marked as triaged. |
| 55 | + |
| 56 | +#### Mixed |
| 57 | + |
| 58 | +Rollup of 12 pull requests [#152399](https://github.com/rust-lang/rust/pull/152399) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=71dc761bfe791895c5997dda654dca1dad817413&end=18d13b5332916ffca8eadb9106d54b5b434e9978&stat=instructions:u) |
| 59 | + |
| 60 | +| (instructions:u) | mean | range | count | |
| 61 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 62 | +| Regressions ❌ <br /> (primary) | 0.2% | [0.2%, 0.2%] | 3 | |
| 63 | +| Regressions ❌ <br /> (secondary) | 0.8% | [0.1%, 1.4%] | 26 | |
| 64 | +| Improvements ✅ <br /> (primary) | - | - | 0 | |
| 65 | +| Improvements ✅ <br /> (secondary) | -0.2% | [-0.3%, -0.1%] | 8 | |
| 66 | +| All ❌✅ (primary) | 0.2% | [0.2%, 0.2%] | 3 | |
| 67 | + |
| 68 | +- The largest regressions seem to be caused by [#152327](https://github.com/rust-lang/rust/pull/152327). |
| 69 | +- That PR fixed some compiler bugs, so the regression on a secondary benchmarks seems justifiable. |
| 70 | +- Marked as triaged. |
| 71 | + |
| 72 | +Update mgca to use `type const` syntax instead of the `#[type_const]` attribute. [#152324](https://github.com/rust-lang/rust/pull/152324) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=18d13b5332916ffca8eadb9106d54b5b434e9978&end=381e9ef09effbed5f1378a24dc6e461fea35a0e1&stat=instructions:u) |
| 73 | + |
| 74 | +| (instructions:u) | mean | range | count | |
| 75 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 76 | +| Regressions ❌ <br /> (primary) | 0.4% | [0.1%, 0.7%] | 18 | |
| 77 | +| Regressions ❌ <br /> (secondary) | 0.5% | [0.0%, 1.7%] | 13 | |
| 78 | +| Improvements ✅ <br /> (primary) | - | - | 0 | |
| 79 | +| Improvements ✅ <br /> (secondary) | -0.2% | [-0.3%, -0.0%] | 4 | |
| 80 | +| All ❌✅ (primary) | 0.4% | [0.1%, 0.7%] | 18 | |
| 81 | + |
| 82 | +- It is unclear whether the regressions were expected or not. |
| 83 | +- Asked the reviewer to follow up. |
| 84 | +- Not marking as triaged yet. |
| 85 | + |
| 86 | +Update to LLVM 22.1.0-rc3 [#152428](https://github.com/rust-lang/rust/pull/152428) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=7dc2e92b83be02dc07f87be7e94266d6e48e5ca5&end=7057231bd78d6c7893f905ea1832365d4c5efe17&stat=instructions:u) |
| 87 | + |
| 88 | +| (instructions:u) | mean | range | count | |
| 89 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 90 | +| Regressions ❌ <br /> (primary) | 0.2% | [0.2%, 0.2%] | 1 | |
| 91 | +| Regressions ❌ <br /> (secondary) | 0.1% | [0.0%, 0.2%] | 4 | |
| 92 | +| Improvements ✅ <br /> (primary) | - | - | 0 | |
| 93 | +| Improvements ✅ <br /> (secondary) | -0.3% | [-0.5%, -0.1%] | 10 | |
| 94 | +| All ❌✅ (primary) | 0.2% | [0.2%, 0.2%] | 1 | |
| 95 | + |
| 96 | +- A few tiny changes that got over the noise threshold, but overall a wash. |
| 97 | +- Marked as triaged. |
| 98 | + |
| 99 | +Rollup of 7 pull requests [#152506](https://github.com/rust-lang/rust/pull/152506) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=5fdff787e6ef7cd4458f81e69cbb4d2a39439a98&end=605f49b27444a738ea4032cb77e3bdc4eb811bab&stat=instructions:u) |
| 100 | + |
| 101 | +| (instructions:u) | mean | range | count | |
| 102 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 103 | +| Regressions ❌ <br /> (primary) | 0.2% | [0.2%, 0.2%] | 1 | |
| 104 | +| Regressions ❌ <br /> (secondary) | 0.2% | [0.1%, 0.2%] | 3 | |
| 105 | +| Improvements ✅ <br /> (primary) | - | - | 0 | |
| 106 | +| Improvements ✅ <br /> (secondary) | -0.1% | [-0.2%, -0.1%] | 2 | |
| 107 | +| All ❌✅ (primary) | 0.2% | [0.2%, 0.2%] | 1 | |
| 108 | + |
| 109 | +- Very small regressions across the board, but I don't think it's worth it to dig deeper. |
| 110 | +- Marked as triaged. |
| 111 | + |
| 112 | +Start using pattern types in libcore (NonZero and friends) [#148537](https://github.com/rust-lang/rust/pull/148537) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=7ad4e69ad585d8ff214f7b42d01f1959eda08f40&end=bb8b30a5fce59f5d3d17a8f010601c59f0f19d79&stat=instructions:u) |
| 113 | + |
| 114 | +| (instructions:u) | mean | range | count | |
| 115 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 116 | +| Regressions ❌ <br /> (primary) | 1.9% | [0.1%, 3.8%] | 12 | |
| 117 | +| Regressions ❌ <br /> (secondary) | 2.2% | [0.1%, 3.6%] | 4 | |
| 118 | +| Improvements ✅ <br /> (primary) | - | - | 0 | |
| 119 | +| Improvements ✅ <br /> (secondary) | -0.1% | [-0.1%, -0.1%] | 2 | |
| 120 | +| All ❌✅ (primary) | 1.9% | [0.1%, 3.8%] | 12 | |
| 121 | + |
| 122 | +- The regression was deemed [acceptable](https://github.com/rust-lang/rust/pull/148537#issuecomment-3883425825) pre-merge, and the post-merge results correspond to the previous perf. runs. |
| 123 | +- Marked as triaged. |
| 124 | + |
| 125 | +Make operational semantics of pattern matching independent of crate and module [#150681](https://github.com/rust-lang/rust/pull/150681) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=5d04477ea8118ba54d9ce402848dda19a41d61eb&end=f8463896a9b36a04899c013bd8825a7fd29dd7a4&stat=instructions:u) |
| 126 | + |
| 127 | +| (instructions:u) | mean | range | count | |
| 128 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 129 | +| Regressions ❌ <br /> (primary) | 0.6% | [0.1%, 1.1%] | 7 | |
| 130 | +| Regressions ❌ <br /> (secondary) | 0.2% | [0.1%, 0.2%] | 12 | |
| 131 | +| Improvements ✅ <br /> (primary) | -2.6% | [-2.6%, -2.6%] | 1 | |
| 132 | +| Improvements ✅ <br /> (secondary) | -2.3% | [-2.3%, -2.3%] | 1 | |
| 133 | +| All ❌✅ (primary) | 0.2% | [-2.6%, 1.1%] | 8 | |
| 134 | + |
| 135 | +- A few small regressions across the board, mainly in `doc` builds, the rest of the profiles are a wash. Given the change made here, I suppose the perf. regression is expected. |
| 136 | +- Marked as triaged. |
| 137 | + |
| 138 | +Rollup of 7 pull requests [#152632](https://github.com/rust-lang/rust/pull/152632) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=a33907a7a5381473eec8bcfa0c56e05a856a911c&end=7bee525095c0872e87c038c412c781b9bbb3f5dc&stat=instructions:u) |
| 139 | + |
| 140 | +| (instructions:u) | mean | range | count | |
| 141 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 142 | +| Regressions ❌ <br /> (primary) | 0.4% | [0.1%, 0.8%] | 6 | |
| 143 | +| Regressions ❌ <br /> (secondary) | 0.4% | [0.1%, 0.8%] | 3 | |
| 144 | +| Improvements ✅ <br /> (primary) | -0.5% | [-1.0%, -0.2%] | 3 | |
| 145 | +| Improvements ✅ <br /> (secondary) | -1.0% | [-2.0%, -0.5%] | 5 | |
| 146 | +| All ❌✅ (primary) | 0.1% | [-1.0%, 0.8%] | 9 | |
| 147 | + |
| 148 | +- Performance-wise it's a wash. |
| 149 | +- Marked as triaged. |
| 150 | + |
| 151 | +Shallow resolve ty and const vars to their root vars [#151380](https://github.com/rust-lang/rust/pull/151380) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=7bee525095c0872e87c038c412c781b9bbb3f5dc&end=75b9d89c68a4a92758349272aa4e35a8eb7a6708&stat=instructions:u) |
| 152 | + |
| 153 | +| (instructions:u) | mean | range | count | |
| 154 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 155 | +| Regressions ❌ <br /> (primary) | - | - | 0 | |
| 156 | +| Regressions ❌ <br /> (secondary) | 0.3% | [0.2%, 0.4%] | 4 | |
| 157 | +| Improvements ✅ <br /> (primary) | -0.3% | [-0.4%, -0.1%] | 14 | |
| 158 | +| Improvements ✅ <br /> (secondary) | -2.7% | [-7.1%, -0.1%] | 50 | |
| 159 | +| All ❌✅ (primary) | -0.3% | [-0.4%, -0.1%] | 14 | |
| 160 | + |
| 161 | +- Many more wins than losses. |
| 162 | +- Marked as triaged. |
| 163 | + |
| 164 | +Big query system cleanups [#152636](https://github.com/rust-lang/rust/pull/152636) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=139651428df86cf88443295542c12ea617cbb587&end=fef627b1ebdc7369ddf8a4031a5d25733ac1fb34&stat=instructions:u) |
| 165 | + |
| 166 | +| (instructions:u) | mean | range | count | |
| 167 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 168 | +| Regressions ❌ <br /> (primary) | 0.3% | [0.2%, 0.4%] | 10 | |
| 169 | +| Regressions ❌ <br /> (secondary) | 0.2% | [0.1%, 0.3%] | 9 | |
| 170 | +| Improvements ✅ <br /> (primary) | - | - | 0 | |
| 171 | +| Improvements ✅ <br /> (secondary) | -0.4% | [-0.4%, -0.3%] | 4 | |
| 172 | +| All ❌✅ (primary) | 0.3% | [0.2%, 0.4%] | 10 | |
| 173 | + |
| 174 | +- The small performance regressions were investigated and deemed to be caused mostly by inlining perturbations. |
| 175 | +- Already marked as triaged. |
0 commit comments