Skip to content

Add triage for 2024-04-22 #2074

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Apr 22, 2025
Merged

Add triage for 2024-04-22 #2074

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
106 changes: 106 additions & 0 deletions triage/2024-04-22.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,106 @@
# 2025-04-22 Triage Log

Mostly positive week. Most of the improvements come from a revert of a regression from a few weeks ago, but we also get nice wins from re-using Sized fast-path, coming from Sized hierarchy implementation work.
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hm.. maybe I should mention removing Nonterminal, too, that's been a long time coming.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You could, but I doubt that anyone outside from a few compiler people know that that means :)


Triage done by **@panstromek**.
Revision range: [15f58c46..8f2819b0](https://perf.rust-lang.org/?start=15f58c46da79399961a09db0c650a2f90f442e6b&end=8f2819b0e3428d0aee05fa60e91e0211c2aea053&absolute=false&stat=instructions%3Au)

**Summary**:

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:---------------:|:-----:|
| Regressions ❌ <br /> (primary) | 1.3% | [0.4%, 2.1%] | 7 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | -1.0% | [-12.9%, -0.1%] | 144 |
| Improvements ✅ <br /> (secondary) | -2.2% | [-12.3%, -0.2%] | 111 |
| All ❌✅ (primary) | -0.9% | [-12.9%, 2.1%] | 151 |


0 Regressions, 4 Improvements, 2 Mixed; 3 of them in rollups
35 artifact comparisons made in total

#### Regressions



#### Improvements

re-use `Sized` fast-path [#139577](https://github.com/rust-lang/rust/pull/139577) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=990039ec53d5bffe0ec77391e00f0e5be05924e8&end=2da29dbe8fe23df1c7c4ab1d8740ca3c32b15526&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:---------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | -0.6% | [-1.2%, -0.3%] | 55 |
| Improvements ✅ <br /> (secondary) | -2.1% | [-12.2%, -0.4%] | 44 |
| All ❌✅ (primary) | -0.6% | [-1.2%, -0.3%] | 55 |


Rollup of 17 pull requests [#139845](https://github.com/rust-lang/rust/pull/139845) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=58c2dd9a54a325f4ce96f70332ceb07a3b58f0e5&end=f433fa46b0fd27d35219357ad75f54d294081bc4&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | - | - | 0 |
| Improvements ✅ <br /> (secondary) | -1.1% | [-1.7%, -0.2%] | 9 |
| All ❌✅ (primary) | - | - | 0 |

Unexpected improvement on `many-assoc-items` stress test. **@Zalathar** identified the cause as [#139669] (https://github.com/rust-lang/rust/pull/139669).


Revert "Deduplicate template parameter creation" [#139878](https://github.com/rust-lang/rust/pull/139878) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=38c560ae681d5c0d3fd615eaedc537a282fb1086&end=cacb9eed381ed19ba936fc019d63d9b9e694007e&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:---------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | -5.8% | [-12.9%, -0.6%] | 11 |
| Improvements ✅ <br /> (secondary) | -2.9% | [-8.7%, -0.3%] | 34 |
| All ❌✅ (primary) | -5.8% | [-12.9%, -0.6%] | 11 |

Reverts previously triaged regression from roughly a month ago.

Rollup of 8 pull requests [#140043](https://github.com/rust-lang/rust/pull/140043) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=90fd16eb5be9255006c95e8af12a0d43854dc1a9&end=49e5e4e3a5610c240a717cb99003a5d5d3356679&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | -0.6% | [-1.2%, -0.3%] | 19 |
| Improvements ✅ <br /> (secondary) | - | - | 0 |
| All ❌✅ (primary) | -0.6% | [-1.2%, -0.3%] | 19 |

All improvements are on bitmaps (both 3.1 and updated 3.2.1 versions). Looks a bit unexpected. Could be a type system fix in [#139762](https://github.com/rust-lang/rust/pull/139762), which is the only change that seems relevant.

#### Mixed

Remove `Nonterminal` and `TokenKind::Interpolated` [#124141](https://github.com/rust-lang/rust/pull/124141) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=15f58c46da79399961a09db0c650a2f90f442e6b&end=f836ae4e663b6e8938096b8559e094d18361be55&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | 1.0% | [0.3%, 2.1%] | 11 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | -0.5% | [-1.4%, -0.2%] | 48 |
| Improvements ✅ <br /> (secondary) | -1.0% | [-2.4%, -0.3%] | 36 |
| All ❌✅ (primary) | -0.2% | [-1.4%, 2.1%] | 59 |

Few real regressions on hyper, but mostly on small incremental runs. Outweighed by a lot of improvements, already marked as triaged by **@mark-simulacrum**.

Rollup of 8 pull requests [#139992](https://github.com/rust-lang/rust/pull/139992) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=a15cce2690e8fab72422515c9dc02c6fbc506733&end=1f76d219c906f0112bb1872f33aa977164c53fa6&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | 0.5% | [0.5%, 0.5%] | 1 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | -0.5% | [-0.6%, -0.5%] | 2 |
| Improvements ✅ <br /> (secondary) | -0.5% | [-0.5%, -0.5%] | 1 |
| All ❌✅ (primary) | -0.2% | [-0.6%, 0.5%] | 3 |

Improvements outweigh regressions.

Somewhat small regression in debug build of Cargo looks real - the benchmark is a bit noisy and it partially recovered in next PR but it hasn't fully recovered since then.

This is also a large bootstrap win, which seems a bit unexpected. Most of the change is in the frontend.

Almost no PR in this rollup touches runtime code, so I suspect this is either a libc update or the drop handling fix from [#139977](https://github.com/rust-lang/rust/pull/139977), because adds code in a generic wrapper around intrinsic.
Loading