diff --git a/triage/2024-10-29.md b/triage/2024-10-29.md new file mode 100644 index 000000000..54294ad27 --- /dev/null +++ b/triage/2024-10-29.md @@ -0,0 +1,231 @@ +# 2024-10-29 Triage Log + +This week saw a lot of activity both on the regressions and improvements side. There was one large +regression, which was immediately reverted. Overall, the week ended up being positive, thanks to +a rollup PR that caused a tiny improvement to almost all benchmarks. + +Triage done by **@kobzol**. +Revision range: [3e33bda0..c8a8c820](https://perf.rust-lang.org/?start=3e33bda0326586a6e1e34d0f5c060ca6d116e6a4&end=c8a8c82035439cb2404b8f24ca0bc18209d534ca&absolute=false&stat=instructions%3Au) + +**Summary**: + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:-----:|:--------------:|:-----:| +| Regressions ❌
(primary) | 0.7% | [0.2%, 2.7%] | 15 | +| Regressions ❌
(secondary) | 0.8% | [0.1%, 1.6%] | 22 | +| Improvements ✅
(primary) | -0.6% | [-1.5%, -0.2%] | 153 | +| Improvements ✅
(secondary) | -0.7% | [-1.9%, -0.1%] | 80 | +| All ❌✅ (primary) | -0.5% | [-1.5%, 2.7%] | 168 | + + +6 Regressions, 6 Improvements, 4 Mixed; 6 of them in rollups +58 artifact comparisons made in total + +#### Regressions + +Revise arm platform notes regarding soft float [#130987](https://github.com/rust-lang/rust/pull/130987) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=3ec4308f6cb4bba3140d736d9ebd53b234fa7df8&end=edbd9398dcc0e3b6ffe0dd0cd4578d37d1d36974&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:----:|:------------:|:-----:| +| Regressions ❌
(primary) | 2.9% | [2.9%, 2.9%] | 1 | +| Regressions ❌
(secondary) | - | - | 0 | +| Improvements ✅
(primary) | - | - | 0 | +| Improvements ✅
(secondary) | - | - | 0 | +| All ❌✅ (primary) | 2.9% | [2.9%, 2.9%] | 1 | + +- Noise, this PR only modified documentation. +- Marked as triaged. + +Dont consider predicates that may hold as impossible in `is_impossible_associated_item` [#131840](https://github.com/rust-lang/rust/pull/131840) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=4392847410ddd67f6734dd9845f9742ff9e85c83&end=814df6e50eaf89b90793e7d9618bb60f1f18377a&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:----:|:------------:|:-----:| +| Regressions ❌
(primary) | 0.6% | [0.2%, 2.3%] | 9 | +| Regressions ❌
(secondary) | 0.2% | [0.2%, 0.2%] | 1 | +| Improvements ✅
(primary) | - | - | 0 | +| Improvements ✅
(secondary) | - | - | 0 | +| All ❌✅ (primary) | 0.6% | [0.2%, 2.3%] | 9 | + +- The largest regression was just noise, the rest was small enough and was deemed to be acceptable. +- Marked as triaged. + +nightly feature tracking: get rid of the per-feature bool fields [#132027](https://github.com/rust-lang/rust/pull/132027) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=ffd978b7bf4ccdc74fe6c1b048d253eb3f3aa341&end=be01dabfefd2daa4574b974f571c7852085d60cb&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:----:|:------------:|:-----:| +| Regressions ❌
(primary) | 1.2% | [1.1%, 1.2%] | 4 | +| Regressions ❌
(secondary) | 0.5% | [0.1%, 1.1%] | 34 | +| Improvements ✅
(primary) | - | - | 0 | +| Improvements ✅
(secondary) | - | - | 0 | +| All ❌✅ (primary) | 1.2% | [1.1%, 1.2%] | 4 | + +- The small regressions were only on tiny benchmarks and stress tests, which was deemed to be acceptable. + +Rollup of 10 pull requests [#132094](https://github.com/rust-lang/rust/pull/132094) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=b8bb2968ce1e44d01520c9d59ee6299ed66df3f9&end=55b7f8e800a6a43657a8582450323d546297c950&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:----:|:------------:|:-----:| +| Regressions ❌
(primary) | 0.2% | [0.1%, 0.2%] | 4 | +| Regressions ❌
(secondary) | - | - | 0 | +| Improvements ✅
(primary) | - | - | 0 | +| Improvements ✅
(secondary) | - | - | 0 | +| All ❌✅ (primary) | 0.2% | [0.1%, 0.2%] | 4 | + +- The tiny regressions were reverted in a follow-up merge. +- Marked as triaged. + +Emit future-incompatibility lint when calling/declaring functions with vectors that require missing target feature [#127731](https://github.com/rust-lang/rust/pull/127731) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=45089ec19ebebec88bace6ec237244ff0eaa7ad3&end=6faf0bd3e561f1a0c81f3eafe0ce0e688385d70e&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:-----:|:--------------:|:-----:| +| Regressions ❌
(primary) | 5.0% | [0.3%, 16.9%] | 77 | +| Regressions ❌
(secondary) | 4.7% | [0.1%, 29.4%] | 30 | +| Improvements ✅
(primary) | -0.1% | [-0.1%, -0.1%] | 1 | +| Improvements ✅
(secondary) | - | - | 0 | +| All ❌✅ (primary) | 4.9% | [-0.1%, 16.9%] | 78 | + +- This was a large regression that was immediately reverted in https://github.com/rust-lang/rust/pull/132152. +- Investigation into how to land this change without such large regressions is happening in https://github.com/rust-lang/rust/pull/132173 +- Marked as triaged. + +bump stdarch [#132145](https://github.com/rust-lang/rust/pull/132145) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=9f57edf2e261ade3c926d69459162057f15164d6&end=3f1be1ec7ec3d8e80beb381ee82164a0aa3ca777&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:----:|:------------:|:-----:| +| Regressions ❌
(primary) | 0.9% | [0.9%, 0.9%] | 1 | +| Regressions ❌
(secondary) | - | - | 0 | +| Improvements ✅
(primary) | - | - | 0 | +| Improvements ✅
(secondary) | - | - | 0 | +| All ❌✅ (primary) | 0.9% | [0.9%, 0.9%] | 1 | + +- One tiny regression caused by a dependency upgrade, no need to investigate further. +- Marked as triaged. + +#### Improvements + +(ci) Update macOS Xcode to 15 [#131570](https://github.com/rust-lang/rust/pull/131570) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=31e102c509192189be841954f38962eb3cd9e2e2&end=4392847410ddd67f6734dd9845f9742ff9e85c83&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:-----:|:--------------:|:-----:| +| Regressions ❌
(primary) | - | - | 0 | +| Regressions ❌
(secondary) | - | - | 0 | +| Improvements ✅
(primary) | -3.6% | [-3.6%, -3.6%] | 1 | +| Improvements ✅
(secondary) | - | - | 0 | +| All ❌✅ (primary) | -3.6% | [-3.6%, -3.6%] | 1 | + + +better default capacity for str::replace [#131929](https://github.com/rust-lang/rust/pull/131929) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=8bf64f106ac0cd1f2b9c6fd27965dd2dc4b862dd&end=b13176595d8a9fd27334747080764a53a18f4c24&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:-----:|:--------------:|:-----:| +| Regressions ❌
(primary) | - | - | 0 | +| Regressions ❌
(secondary) | - | - | 0 | +| Improvements ✅
(primary) | -0.3% | [-0.4%, -0.2%] | 3 | +| Improvements ✅
(secondary) | - | - | 0 | +| All ❌✅ (primary) | -0.3% | [-0.4%, -0.2%] | 3 | + + +rustdoc: hash assets at rustdoc build time [#131951](https://github.com/rust-lang/rust/pull/131951) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=55b7f8e800a6a43657a8582450323d546297c950&end=8aca4bab080b2c81065645fc070acca7a060f8a3&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:-----:|:--------------:|:-----:| +| Regressions ❌
(primary) | - | - | 0 | +| Regressions ❌
(secondary) | - | - | 0 | +| Improvements ✅
(primary) | -0.2% | [-0.3%, -0.1%] | 6 | +| Improvements ✅
(secondary) | -0.3% | [-0.3%, -0.1%] | 17 | +| All ❌✅ (primary) | -0.2% | [-0.3%, -0.1%] | 6 | + + +Revert #127731 "Emit error when calling/declaring functions with unavailable …" [#132152](https://github.com/rust-lang/rust/pull/132152) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=a06b7cbe21967a86050fa92dab843c8afda1c28e&end=ae4c6b66402cdbda8daa626216efd94e2dbe94b2&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:-----:|:---------------:|:-----:| +| Regressions ❌
(primary) | - | - | 0 | +| Regressions ❌
(secondary) | 0.3% | [0.3%, 0.3%] | 1 | +| Improvements ✅
(primary) | -4.7% | [-14.4%, -0.3%] | 75 | +| Improvements ✅
(secondary) | -4.5% | [-22.7%, -0.2%] | 27 | +| All ❌✅ (primary) | -4.7% | [-14.4%, -0.3%] | 75 | + +- Revert of the previously mentioned [#127731](https://github.com/rust-lang/rust/pull/127731). + +Rollup of 3 pull requests [#132171](https://github.com/rust-lang/rust/pull/132171) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=ae4c6b66402cdbda8daa626216efd94e2dbe94b2&end=80d0d927d5069b67cc08c0c65b48e7b6e0cdeeb5&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:-----:|:--------------:|:-----:| +| Regressions ❌
(primary) | - | - | 0 | +| Regressions ❌
(secondary) | - | - | 0 | +| Improvements ✅
(primary) | -0.3% | [-0.3%, -0.2%] | 2 | +| Improvements ✅
(secondary) | -0.3% | [-0.4%, -0.2%] | 16 | +| All ❌✅ (primary) | -0.3% | [-0.3%, -0.2%] | 2 | + +- The small wins probably come from [#132168](https://github.com/rust-lang/rust/pull/132168). + +Rollup of 9 pull requests [#132277](https://github.com/rust-lang/rust/pull/132277) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=a9d17627d241645a54c1134a20f1596127fedb60&end=2df8dbb1b37168c59eca2884502a1b79892858a9&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:-----:|:--------------:|:-----:| +| Regressions ❌
(primary) | - | - | 0 | +| Regressions ❌
(secondary) | - | - | 0 | +| Improvements ✅
(primary) | -0.5% | [-1.7%, -0.1%] | 186 | +| Improvements ✅
(secondary) | -0.6% | [-1.7%, -0.1%] | 92 | +| All ❌✅ (primary) | -0.5% | [-1.7%, -0.1%] | 186 | + +- The large amount of small wins might have been caused by [#130259](https://github.com/rust-lang/rust/pull/130259). + +#### Mixed + +Rollup of 5 pull requests [#132053](https://github.com/rust-lang/rust/pull/132053) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=e1f306899514ea80abc1d1c9f6a57762afb304a3&end=9abfcb49002b82f46ad66ceb0b26cee8be2dc658&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:-----:|:--------------:|:-----:| +| Regressions ❌
(primary) | 1.0% | [0.3%, 1.8%] | 2 | +| Regressions ❌
(secondary) | - | - | 0 | +| Improvements ✅
(primary) | - | - | 0 | +| Improvements ✅
(secondary) | -0.5% | [-0.5%, -0.5%] | 1 | +| All ❌✅ (primary) | 1.0% | [0.3%, 1.8%] | 2 | + +- Large number of small-ish regressions that were just under the statistical threshold. +- Might have been caused by LLVM doing more work because of [#132031](https://github.com/rust-lang/rust/pull/132031). +- Not marking as triaged yet, investigation ongoing. + +Represent trait constness as a distinct predicate [#131985](https://github.com/rust-lang/rust/pull/131985) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=f61306d47bc98af8bb9d15f1adf6086785590a8c&end=1d4a7670d4f37bfbae2d89ec3ec07cd40dbc5a5d&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:-----:|:--------------:|:-----:| +| Regressions ❌
(primary) | 0.5% | [0.1%, 1.2%] | 16 | +| Regressions ❌
(secondary) | 0.9% | [0.1%, 1.5%] | 22 | +| Improvements ✅
(primary) | -0.3% | [-0.5%, -0.1%] | 44 | +| Improvements ✅
(secondary) | -0.3% | [-0.6%, -0.1%] | 13 | +| All ❌✅ (primary) | -0.1% | [-0.5%, 1.2%] | 60 | + +- This was a large internal refactoring that had mostly positive or neutral effects on performance, + although it did cause regressions in `doc` builds on several benchmarks. +- It was deemed acceptable during review. +- Marked as triaged. + +Rollup of 4 pull requests [#132116](https://github.com/rust-lang/rust/pull/132116) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=1d4a7670d4f37bfbae2d89ec3ec07cd40dbc5a5d&end=a93c1718c80b9f100056c8eec3fc37fbd6424134&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:-----:|:--------------:|:-----:| +| Regressions ❌
(primary) | 0.3% | [0.1%, 1.8%] | 49 | +| Regressions ❌
(secondary) | 0.8% | [0.1%, 1.3%] | 6 | +| Improvements ✅
(primary) | -0.3% | [-0.3%, -0.3%] | 1 | +| Improvements ✅
(secondary) | -0.7% | [-1.8%, -0.1%] | 3 | +| All ❌✅ (primary) | 0.3% | [-0.3%, 1.8%] | 50 | + +- The regression was caused by [#131983](https://github.com/rust-lang/rust/pull/131983), which + stabilized a new behavior change that will appear in Edition 2024. +- Not marking as triaged yet, there might be a way to claw some of the regressions back. + +Rollup of 12 pull requests [#132317](https://github.com/rust-lang/rust/pull/132317) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=c8a8c82035439cb2404b8f24ca0bc18209d534ca&end=2dece5bb62f234f5622a08289c5a3d1555cd7843&stat=instructions:u) + +| (instructions:u) | mean | range | count | +|:----------------------------------:|:-----:|:--------------:|:-----:| +| Regressions ❌
(primary) | 0.2% | [0.2%, 0.4%] | 4 | +| Regressions ❌
(secondary) | 0.2% | [0.2%, 0.2%] | 1 | +| Improvements ✅
(primary) | -0.2% | [-0.4%, -0.1%] | 6 | +| Improvements ✅
(secondary) | -0.3% | [-0.3%, -0.3%] | 1 | +| All ❌✅ (primary) | -0.0% | [-0.4%, 0.4%] | 10 | + +- Performance results were a wash. +- Marked as triaged.