Skip to content

Add perf triage for this week #2425

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
lqd merged 1 commit into from
Apr 6, 2026
+172 −0
Merged

Add perf triage for this week #2425

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
172 changes: 172 additions & 0 deletions triage/2026/2026-04-05.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,172 @@
# 2026-04-05 Triage Log

A shorter week than normal (probably due to later perf triage last week).
Overall fairly small changes scattered across various PRs, though the net
effect was slightly positive (-0.5% avg change). All changed ended up either
mixed or improvements this week.

Triage done by **@simulacrum**.
Revision range: [cf7da0b7..e73c56ab](https://perf.rust-lang.org/?start=cf7da0b7277cad05b79f91b60c290aa08a17a6f0&end=e73c56abd0baf3dbaafbdc3ce6072a416aade867&absolute=false&stat=instructions%3Au)

**Summary**:

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | 0.4% | [0.4%, 0.4%] | 2 |
| Regressions ❌ <br /> (secondary) | 2.7% | [0.2%, 7.9%] | 47 |
| Improvements ✅ <br /> (primary) | -0.5% | [-1.6%, -0.1%] | 164 |
| Improvements ✅ <br /> (secondary) | -0.5% | [-1.9%, -0.1%] | 115 |
| All ❌✅ (primary) | -0.5% | [-1.6%, 0.4%] | 166 |

0 Regressions, 3 Improvements, 8 Mixed; 5 of them in rollups
26 artifact comparisons made in total

#### Regressions

No regressions this week.

#### Improvements

Rollup of 4 pull requests [#154638](https://github.com/rust-lang/rust/pull/154638) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=0e95a0f4c677002a5d4ac5bc59d97885e6f51f71&end=48cc71ee88cd0f11217eced958b9930970da998b&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | -0.2% | [-0.3%, -0.2%] | 3 |
| Improvements ✅ <br /> (secondary) | - | - | 0 |
| All ❌✅ (primary) | -0.2% | [-0.3%, -0.2%] | 3 |


Rollup of 5 pull requests [#154659](https://github.com/rust-lang/rust/pull/154659) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=12ab1cf1fd3dc925a846b0eb8a4b39393140fdd3&end=3ebe60c8c30d1175019dcbdb957441cd84aa1220&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | -0.7% | [-1.1%, -0.2%] | 10 |
| Improvements ✅ <br /> (secondary) | - | - | 0 |
| All ❌✅ (primary) | -0.7% | [-1.1%, -0.2%] | 10 |


Streamline `CachingSourceMapView` [#154614](https://github.com/rust-lang/rust/pull/154614) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=009a6c1e8b559de780be565937d83e8f4a36d58d&end=8931f23941cf9ec0093f86a81337873b288344c9&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | -0.4% | [-0.7%, -0.1%] | 135 |
| Improvements ✅ <br /> (secondary) | -0.4% | [-1.4%, -0.1%] | 95 |
| All ❌✅ (primary) | -0.4% | [-0.7%, -0.1%] | 135 |


#### Mixed

Properly generalize unevaluated consts [#154053](https://github.com/rust-lang/rust/pull/154053) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=08cd08fbef8c6663b052d8d9e4930d5696cbb8a7&end=37cfa179be19cec2236736521bed3ea3c4a1ec97&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | 0.2% | [0.2%, 0.2%] | 3 |
| Improvements ✅ <br /> (primary) | -0.3% | [-0.3%, -0.3%] | 1 |
| Improvements ✅ <br /> (secondary) | -0.1% | [-0.2%, -0.0%] | 7 |
| All ❌✅ (primary) | -0.3% | [-0.3%, -0.3%] | 1 |

Regressions are limited to new solver, and fairly small. Most likely no need to
investigate deeper, especially as this also fixes a few bugs.

Rollup of 4 pull requests [#154668](https://github.com/rust-lang/rust/pull/154668) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=3ebe60c8c30d1175019dcbdb957441cd84aa1220&end=2bd7a97871a74d4333bd3edb6564136167ac604b&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | 0.2% | [0.1%, 0.4%] | 4 |
| Regressions ❌ <br /> (secondary) | 0.2% | [0.0%, 0.3%] | 14 |
| Improvements ✅ <br /> (primary) | - | - | 0 |
| Improvements ✅ <br /> (secondary) | -0.4% | [-1.1%, -0.0%] | 5 |
| All ❌✅ (primary) | 0.2% | [0.1%, 0.4%] | 4 |

Most likely caused by extending integers with new methods in
[#154356](https://github.com/rust-lang/rust/pull/154356). Enqueued a build to
confirm that theory, see [comment](https://github.com/rust-lang/rust/pull/154356#issuecomment-4191902607).

Don't use disk-cache for query `def_kind` [#154292](https://github.com/rust-lang/rust/pull/154292) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=2bd7a97871a74d4333bd3edb6564136167ac604b&end=7e46c5f6fb87f8cf4353e058479cef15d1d952b4&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | 0.3% | [0.0%, 1.1%] | 5 |
| Improvements ✅ <br /> (primary) | -0.3% | [-0.4%, -0.1%] | 62 |
| Improvements ✅ <br /> (secondary) | -0.4% | [-0.7%, -0.1%] | 28 |
| All ❌✅ (primary) | -0.3% | [-0.4%, -0.1%] | 62 |

Not entirely clear but the instruction count regressions look like they may be
within noise threshold in practice. Improvements also clearly outweigh the
small regressions, so marking as triaged.

Revert #151380 and #153869 [#154308](https://github.com/rust-lang/rust/pull/154308) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=7e46c5f6fb87f8cf4353e058479cef15d1d952b4&end=009a6c1e8b559de780be565937d83e8f4a36d58d&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | 0.3% | [0.1%, 0.5%] | 16 |
| Regressions ❌ <br /> (secondary) | 3.6% | [0.0%, 8.6%] | 36 |
| Improvements ✅ <br /> (primary) | - | - | 0 |
| Improvements ✅ <br /> (secondary) | -0.3% | [-1.1%, -0.0%] | 16 |
| All ❌✅ (primary) | 0.3% | [0.1%, 0.5%] | 16 |

This reverts two improvements that landed in a previous week, so regressions are expected. This is ultimately a correctness fix.

Revert "Fix: On wasm targets, call panic_in_cleanup if panic occurs in cleanup #151771" [#154700](https://github.com/rust-lang/rust/pull/154700) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=8931f23941cf9ec0093f86a81337873b288344c9&end=e6b64a2f4c696b840f8a384ec28690eed6a5d267&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | 0.2% | [0.0%, 0.3%] | 7 |
| Improvements ✅ <br /> (primary) | - | - | 0 |
| Improvements ✅ <br /> (secondary) | -0.1% | [-0.1%, -0.0%] | 2 |
| All ❌✅ (primary) | - | - | 0 |

> revert, tiny changes, could be noise, the original PR didn't have these as
> positive results, marking as triaged as there's nothing much to do or
> investigate (until it relands and shows perf changes)

(see https://github.com/rust-lang/rust/pull/154700#issuecomment-4176873246)

Rollup of 20 pull requests [#154727](https://github.com/rust-lang/rust/pull/154727) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=e6b64a2f4c696b840f8a384ec28690eed6a5d267&end=55e86c996809902e8bbad512cfb4d2c18be446d9&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | 0.1% | [0.1%, 0.1%] | 1 |
| Improvements ✅ <br /> (primary) | -0.3% | [-0.3%, -0.1%] | 5 |
| Improvements ✅ <br /> (secondary) | - | - | 0 |
| All ❌✅ (primary) | -0.3% | [-0.3%, -0.1%] | 5 |

Probably noise (at least the regression), either way not worth investigating further.

Port lint attributes to attribute parser [#152369](https://github.com/rust-lang/rust/pull/152369) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=f908263775fa86408d0d529dec82baa9ae350529&end=2972b5e59f1c5529b6ba770437812fd83ab4ebd4&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | 0.2% | [0.2%, 0.2%] | 1 |
| Regressions ❌ <br /> (secondary) | 0.3% | [0.2%, 0.5%] | 7 |
| Improvements ✅ <br /> (primary) | -0.3% | [-0.6%, -0.2%] | 20 |
| Improvements ✅ <br /> (secondary) | -0.4% | [-0.6%, -0.0%] | 37 |
| All ❌✅ (primary) | -0.3% | [-0.6%, 0.2%] | 21 |

Perf wins significantly outweigh the regressions.

Rollup of 5 pull requests [#154802](https://github.com/rust-lang/rust/pull/154802) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=981cf69479ded5e2de0cf9e16111c19d65be0369&end=0312a55fe4200208170b94bf287ca3cc7ea499ae&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | 0.2% | [0.0%, 0.4%] | 10 |
| Improvements ✅ <br /> (primary) | - | - | 0 |
| Improvements ✅ <br /> (secondary) | -0.6% | [-1.6%, -0.0%] | 3 |
| All ❌✅ (primary) | - | - | 0 |

Regression almost certainly due to
<https://github.com/rust-lang/rust/pull/154376#issuecomment-4130882120>, that
perf run matches pretty closely what we see here. I don't think further
investigation is warranted, the affected benchmark intentionally stresses the
lint system (unused-warnings).
Loading