|
| 1 | +# 2025-04-29 Triage Log |
| 2 | + |
| 3 | +Strange week with lots of noise peeking through the performance runs. The only really significant change was a performance improvement that comes from allowing out of order encoding of the dep graph. |
| 4 | + |
| 5 | +Triage done by **@rylev**. |
| 6 | +Revision range: [8f2819b0..25cdf1f6](https://perf.rust-lang.org/?start=8f2819b0e3428d0aee05fa60e91e0211c2aea053&end=25cdf1f67463c9365d8d83778c933ec7480e940b&absolute=false&stat=instructions%3Au) |
| 7 | + |
| 8 | +**Summary**: |
| 9 | + |
| 10 | +| (instructions:u) | mean | range | count | |
| 11 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 12 | +| Regressions ❌ <br /> (primary) | 0.5% | [0.1%, 3.0%] | 77 | |
| 13 | +| Regressions ❌ <br /> (secondary) | 0.6% | [0.1%, 2.4%] | 77 | |
| 14 | +| Improvements ✅ <br /> (primary) | -0.7% | [-1.3%, -0.2%] | 106 | |
| 15 | +| Improvements ✅ <br /> (secondary) | -0.7% | [-1.2%, -0.2%] | 29 | |
| 16 | +| All ❌✅ (primary) | -0.2% | [-1.3%, 3.0%] | 183 | |
| 17 | + |
| 18 | + |
| 19 | +4 Regressions, 2 Improvements, 4 Mixed; 2 of them in rollups |
| 20 | +38 artifact comparisons made in total |
| 21 | + |
| 22 | +#### Regressions |
| 23 | + |
| 24 | +Refactor git change detection in bootstrap [#138591](https://github.com/rust-lang/rust/pull/138591) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=1a5bf12f6586d724ed5ff40e58e06c0233560c0e&end=645d0ad2a4f145ae576e442ec5c73c0f8eed829b&stat=instructions:u) |
| 25 | + |
| 26 | +| (instructions:u) | mean | range | count | |
| 27 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 28 | +| Regressions ❌ <br /> (primary) | 0.2% | [0.1%, 0.3%] | 13 | |
| 29 | +| Regressions ❌ <br /> (secondary) | 0.3% | [0.2%, 0.6%] | 42 | |
| 30 | +| Improvements ✅ <br /> (primary) | -0.2% | [-0.2%, -0.2%] | 1 | |
| 31 | +| Improvements ✅ <br /> (secondary) | - | - | 0 | |
| 32 | +| All ❌✅ (primary) | 0.2% | [-0.2%, 0.3%] | 14 | |
| 33 | +- Looks like noise. |
| 34 | + |
| 35 | + |
| 36 | +Properly stall coroutine witnesses in new solver [#138845](https://github.com/rust-lang/rust/pull/138845) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=df35ff6c354f1f1fbf430b84e7dea37dfe997f34&end=fa58ce343ad498196d799a7381869e79938e952a&stat=instructions:u) |
| 37 | + |
| 38 | +| (instructions:u) | mean | range | count | |
| 39 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 40 | +| Regressions ❌ <br /> (primary) | 0.2% | [0.1%, 0.4%] | 25 | |
| 41 | +| Regressions ❌ <br /> (secondary) | 0.4% | [0.1%, 0.6%] | 48 | |
| 42 | +| Improvements ✅ <br /> (primary) | -0.2% | [-0.2%, -0.2%] | 1 | |
| 43 | +| Improvements ✅ <br /> (secondary) | - | - | 0 | |
| 44 | +| All ❌✅ (primary) | 0.2% | [-0.2%, 0.4%] | 26 | |
| 45 | +- The regressions are small enough that this isn't a huge concern, and the regressions themselves might be due to some underlying perf instability. |
| 46 | + |
| 47 | + |
| 48 | +Implement a lint for implicit autoref of raw pointer dereference - take 2 [#123239](https://github.com/rust-lang/rust/pull/123239) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=0134651fb81314870903e21b1bcbdd993d75b61a&end=a932eb36f8adf6c8cdfc450f063943da3112d621&stat=instructions:u) |
| 49 | + |
| 50 | +| (instructions:u) | mean | range | count | |
| 51 | +|:----------------------------------:|:----:|:------------:|:-----:| |
| 52 | +| Regressions ❌ <br /> (primary) | 1.8% | [0.3%, 3.6%] | 6 | |
| 53 | +| Regressions ❌ <br /> (secondary) | - | - | 0 | |
| 54 | +| Improvements ✅ <br /> (primary) | - | - | 0 | |
| 55 | +| Improvements ✅ <br /> (secondary) | - | - | 0 | |
| 56 | +| All ❌✅ (primary) | 1.8% | [0.3%, 3.6%] | 6 | |
| 57 | +- Perf regression reversed in https://github.com/rust-lang/rust/pull/140406 |
| 58 | + |
| 59 | + |
| 60 | +Async drop codegen [#123948](https://github.com/rust-lang/rust/pull/123948) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=a932eb36f8adf6c8cdfc450f063943da3112d621&end=7d65abfe80f9eee93296d1ce08f845c9bf7039f8&stat=instructions:u) |
| 61 | + |
| 62 | +| (instructions:u) | mean | range | count | |
| 63 | +|:----------------------------------:|:----:|:------------:|:-----:| |
| 64 | +| Regressions ❌ <br /> (primary) | 0.4% | [0.2%, 0.6%] | 12 | |
| 65 | +| Regressions ❌ <br /> (secondary) | 1.0% | [0.2%, 2.4%] | 28 | |
| 66 | +| Improvements ✅ <br /> (primary) | - | - | 0 | |
| 67 | +| Improvements ✅ <br /> (secondary) | - | - | 0 | |
| 68 | +| All ❌✅ (primary) | 0.4% | [0.2%, 0.6%] | 12 | |
| 69 | +- Ask the author and reviewer whether this is worth pursuing more after their initial perf investigation. |
| 70 | + |
| 71 | + |
| 72 | +#### Improvements |
| 73 | + |
| 74 | +Remove `token::{Open,Close}Delim` [#139897](https://github.com/rust-lang/rust/pull/139897) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=d6c1e454aa8af5e7e59fbf5c4e7d3128d2f99582&end=fae7785b60ea7fe1ad293352c057a5b7be73d245&stat=instructions:u) |
| 75 | + |
| 76 | +| (instructions:u) | mean | range | count | |
| 77 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 78 | +| Regressions ❌ <br /> (primary) | - | - | 0 | |
| 79 | +| Regressions ❌ <br /> (secondary) | - | - | 0 | |
| 80 | +| Improvements ✅ <br /> (primary) | -0.4% | [-0.6%, -0.2%] | 14 | |
| 81 | +| Improvements ✅ <br /> (secondary) | - | - | 0 | |
| 82 | +| All ❌✅ (primary) | -0.4% | [-0.6%, -0.2%] | 14 | |
| 83 | + |
| 84 | + |
| 85 | +Rollup of 8 pull requests [#140165](https://github.com/rust-lang/rust/pull/140165) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=6bc57c6bf7d0024ad9ea5a2c112f3fc9c383c8a4&end=1a5bf12f6586d724ed5ff40e58e06c0233560c0e&stat=instructions:u) |
| 86 | + |
| 87 | +| (instructions:u) | mean | range | count | |
| 88 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 89 | +| Regressions ❌ <br /> (primary) | 0.2% | [0.2%, 0.2%] | 1 | |
| 90 | +| Regressions ❌ <br /> (secondary) | - | - | 0 | |
| 91 | +| Improvements ✅ <br /> (primary) | -0.2% | [-0.3%, -0.1%] | 17 | |
| 92 | +| Improvements ✅ <br /> (secondary) | -0.3% | [-0.6%, -0.2%] | 48 | |
| 93 | +| All ❌✅ (primary) | -0.2% | [-0.3%, 0.2%] | 18 | |
| 94 | + |
| 95 | + |
| 96 | +#### Mixed |
| 97 | + |
| 98 | +Clippy subtree update [#139983](https://github.com/rust-lang/rust/pull/139983) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=553600e0f5f5a7d492de6d95ccb2f057005f5651&end=df35ff6c354f1f1fbf430b84e7dea37dfe997f34&stat=instructions:u) |
| 99 | + |
| 100 | +| (instructions:u) | mean | range | count | |
| 101 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 102 | +| Regressions ❌ <br /> (primary) | 0.2% | [0.2%, 0.2%] | 1 | |
| 103 | +| Regressions ❌ <br /> (secondary) | - | - | 0 | |
| 104 | +| Improvements ✅ <br /> (primary) | -0.2% | [-0.3%, -0.1%] | 7 | |
| 105 | +| Improvements ✅ <br /> (secondary) | -0.3% | [-0.6%, -0.2%] | 36 | |
| 106 | +| All ❌✅ (primary) | -0.1% | [-0.3%, 0.2%] | 8 | |
| 107 | +- The small regression is fine in light of the larger and more numerous improvements. |
| 108 | + |
| 109 | + |
| 110 | +Rollup of 8 pull requests [#140256](https://github.com/rust-lang/rust/pull/140256) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=3c877f6a477380ed61155d3bf816df09c9e05b9e&end=d7ea436a02d5de4033fcf7fd4eb8ed965d0f574c&stat=instructions:u) |
| 111 | + |
| 112 | +| (instructions:u) | mean | range | count | |
| 113 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 114 | +| Regressions ❌ <br /> (primary) | 0.2% | [0.2%, 0.2%] | 1 | |
| 115 | +| Regressions ❌ <br /> (secondary) | - | - | 0 | |
| 116 | +| Improvements ✅ <br /> (primary) | - | - | 0 | |
| 117 | +| Improvements ✅ <br /> (secondary) | -0.3% | [-0.6%, -0.2%] | 23 | |
| 118 | +| All ❌✅ (primary) | 0.2% | [0.2%, 0.2%] | 1 | |
| 119 | +- The single regression looks to be noise. |
| 120 | + |
| 121 | + |
| 122 | +set subsections_via_symbols for ld64 helper sections [#139752](https://github.com/rust-lang/rust/pull/139752) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=d7ea436a02d5de4033fcf7fd4eb8ed965d0f574c&end=847e3ee6b0e614937eee4e6d8f61094411eadcc0&stat=instructions:u) |
| 123 | + |
| 124 | +| (instructions:u) | mean | range | count | |
| 125 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 126 | +| Regressions ❌ <br /> (primary) | 0.3% | [0.2%, 0.3%] | 2 | |
| 127 | +| Regressions ❌ <br /> (secondary) | 0.3% | [0.2%, 0.6%] | 30 | |
| 128 | +| Improvements ✅ <br /> (primary) | -0.2% | [-0.2%, -0.2%] | 1 | |
| 129 | +| Improvements ✅ <br /> (secondary) | - | - | 0 | |
| 130 | +| All ❌✅ (primary) | 0.1% | [-0.2%, 0.3%] | 3 | |
| 131 | +- Not sure where the regressions are coming from since this should only impact macOS. |
| 132 | + |
| 133 | + |
| 134 | +Allow out of order dep graph node encoding [#139756](https://github.com/rust-lang/rust/pull/139756) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=8f43b85954d2f3d8fc00a7504c603e5ca9eb0695&end=e3e432d4d65a55e6db167598e96db2bcb163e316&stat=instructions:u) |
| 135 | + |
| 136 | +| (instructions:u) | mean | range | count | |
| 137 | +|:----------------------------------:|:-----:|:--------------:|:-----:| |
| 138 | +| Regressions ❌ <br /> (primary) | 0.3% | [0.3%, 0.3%] | 7 | |
| 139 | +| Regressions ❌ <br /> (secondary) | 0.4% | [0.2%, 0.6%] | 15 | |
| 140 | +| Improvements ✅ <br /> (primary) | -0.7% | [-1.3%, -0.3%] | 115 | |
| 141 | +| Improvements ✅ <br /> (secondary) | -0.8% | [-1.4%, -0.2%] | 27 | |
| 142 | +| All ❌✅ (primary) | -0.7% | [-1.3%, 0.3%] | 122 | |
| 143 | +- Perf improvements massively outweigh the regressions. |
| 144 | + |
0 commit comments