Skip to content

Add perf triage for 2025-01-14 #2032

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.

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
Jan 15, 2025
Merged
Changes from all commits
Commits
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
143 changes: 143 additions & 0 deletions triage/2025-01-14.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,143 @@
# 2025-01-14 Triage Log

A quiet week with little change to the actual compiler performance. The biggest compiler regression was quickly recognized and reverted.

Triage done by **@rylev**.
Revision range: [0f1e965f..1ab85fbd](https://perf.rust-lang.org/?start=0f1e965fec3bc2f97b932e9dd8e85fca6d7faadc&end=1ab85fbd7474e8ce84d5283548f21472860de3e2&absolute=false&stat=instructions%3Au)

**Summary**:

| (instructions:u) | mean | range | count |
|:----------------------------------:|:------:|:---------------:|:-----:|
| Regressions ❌ <br /> (primary) | 0.4% | [0.1%, 1.8%] | 21 |
| Regressions ❌ <br /> (secondary) | 0.5% | [0.0%, 2.0%] | 35 |
| Improvements ✅ <br /> (primary) | -0.8% | [-2.7%, -0.3%] | 6 |
| Improvements ✅ <br /> (secondary) | -10.2% | [-27.8%, -0.1%] | 13 |
| All ❌✅ (primary) | 0.2% | [-2.7%, 1.8%] | 27 |


4 Regressions, 3 Improvements, 3 Mixed; 3 of them in rollups
44 artifact comparisons made in total

#### Regressions

Master bootstrap update [#135268](https://github.com/rust-lang/rust/pull/135268) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=ec12cd8436aab693d8847b3077c26fe2de4ed661&end=251206c27b619ccf3a08e2ac4c525dc343f08492&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:----:|:------------:|:-----:|
| Regressions ❌ <br /> (primary) | 0.2% | [0.2%, 0.2%] | 3 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | - | - | 0 |
| Improvements ✅ <br /> (secondary) | - | - | 0 |
| All ❌✅ (primary) | 0.2% | [0.2%, 0.2%] | 3 |
- Probably just noise, and if not, it's not really worth investigating.


mir_transform: implement `#[rustc_force_inline]` [#134082](https://github.com/rust-lang/rust/pull/134082) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=336209eef13882bd1e211b24779584cb7ef911eb&end=b1a7dfb91106018f47ed9dc9b27aee1977682868&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | 0.3% | [0.2%, 0.3%] | 10 |
| Regressions ❌ <br /> (secondary) | 0.3% | [0.2%, 0.3%] | 15 |
| Improvements ✅ <br /> (primary) | -0.1% | [-0.1%, -0.1%] | 1 |
| Improvements ✅ <br /> (secondary) | -0.3% | [-0.4%, -0.3%] | 2 |
| All ❌✅ (primary) | 0.2% | [-0.1%, 0.3%] | 11 |
- Looks like some regressions snuck back in since the last run - I'm unsure if something happened in a rebase to reintroduce them.
- The actual regressions are much smaller than the original perf run, and I don't think they're big enough to demand an investigation.


Add #[inline] to copy_from_slice [#135384](https://github.com/rust-lang/rust/pull/135384) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=627513a764a696b59bb2679cc1731a197a088a1d&end=48a426eca9df23b24b3559e545cf88dee61d4de9&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:----:|:------------:|:-----:|
| Regressions ❌ <br /> (primary) | 1.1% | [1.1%, 1.1%] | 1 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | - | - | 0 |
| Improvements ✅ <br /> (secondary) | - | - | 0 |
| All ❌✅ (primary) | 1.1% | [1.1%, 1.1%] | 1 |
- From the PR "Small chaotic effect is expected, small binary size improvements oddly line up with the effect I was going for. Seems unlikely they're connected though, I wouldn't be surprised to see a different list of affected scenarios after merge."


Rollup of 4 pull requests [#135438](https://github.com/rust-lang/rust/pull/135438) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=2ae9916816a448fcaab3b2da461de754eda0055a&end=1ab85fbd7474e8ce84d5283548f21472860de3e2&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | 0.4% | [0.4%, 0.4%] | 1 |
| Regressions ❌ <br /> (secondary) | 1.2% | [0.9%, 1.6%] | 6 |
| Improvements ✅ <br /> (primary) | - | - | 0 |
| Improvements ✅ <br /> (secondary) | -0.2% | [-0.2%, -0.2%] | 1 |
| All ❌✅ (primary) | 0.4% | [0.4%, 0.4%] | 1 |
- https://github.com/rust-lang/rust/pull/135426 is the offending PR. Working with the author on a solution.


#### Improvements

Rollup of 6 pull requests [#135260](https://github.com/rust-lang/rust/pull/135260) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=a580b5c379b4fca50dfe5afc0fc0ce00921e4e00&end=e26ff2f9086fc449b963df578f8641c31846abe6&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | -0.3% | [-0.4%, -0.3%] | 4 |
| Improvements ✅ <br /> (secondary) | -0.5% | [-0.5%, -0.5%] | 1 |
| All ❌✅ (primary) | -0.3% | [-0.4%, -0.3%] | 4 |


Remove special-casing for argument patterns in MIR typeck (attempt to fix perf regression of #133858) [#135273](https://github.com/rust-lang/rust/pull/135273) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=67951d946a158bc70949150ca06265e912752096&end=b44e14f762fd4062faeba97f8f2bc470298ec1ac&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:---------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | -3.0% | [-17.4%, -0.2%] | 16 |
| Improvements ✅ <br /> (secondary) | - | - | 0 |
| All ❌✅ (primary) | -3.0% | [-17.4%, -0.2%] | 16 |


Add an InstSimplify for repetitive array expressions [#135274](https://github.com/rust-lang/rust/pull/135274) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=a2d7c8144f3f237612b37420fe2b6b08408d5b1d&end=7e4077d06fc073442c567d58885b47ed2c5121b8&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:------:|:---------------:|:-----:|
| Regressions ❌ <br /> (primary) | - | - | 0 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | - | - | 0 |
| Improvements ✅ <br /> (secondary) | -14.7% | [-27.9%, -1.3%] | 9 |
| All ❌✅ (primary) | - | - | 0 |


#### Mixed

`best_blame_constraint`: Blame better constraints when the region graph has cycles from invariance or `'static` [#133858](https://github.com/rust-lang/rust/pull/133858) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=9c87288a7d2f03625a813df6d3bfe43c09ad4f5a&end=6afee111c2faf86ba884ea748967130abad37b52&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | 3.2% | [0.1%, 21.0%] | 17 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | - | - | 0 |
| Improvements ✅ <br /> (secondary) | -0.3% | [-0.3%, -0.3%] | 1 |
| All ❌✅ (primary) | 3.2% | [0.1%, 21.0%] | 17 |
- Partially reverted in https://github.com/rust-lang/rust/pull/135273


[mir-opt] GVN some more transmute cases [#133324](https://github.com/rust-lang/rust/pull/133324) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=65d7296fc9868af121adb5e07358a3815a80a7a1&end=b6b8361bce8561fb8786ad33ca1abfdf4bc487b6&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | 0.7% | [0.1%, 1.4%] | 3 |
| Regressions ❌ <br /> (secondary) | - | - | 0 |
| Improvements ✅ <br /> (primary) | -0.6% | [-3.0%, -0.2%] | 11 |
| Improvements ✅ <br /> (secondary) | -0.3% | [-0.5%, -0.2%] | 7 |
| All ❌✅ (primary) | -0.3% | [-3.0%, 1.4%] | 14 |
- [Original perf run was deemed acceptable](https://github.com/rust-lang/rust/pull/133324#issuecomment-2520872151) and the final perf run wasn't as bad.


Rollup of 6 pull requests [#135357](https://github.com/rust-lang/rust/pull/135357) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=7e4077d06fc073442c567d58885b47ed2c5121b8&end=ce55b2052db2c9d6deb2dbc5cb2f56d163515a99&stat=instructions:u)

| (instructions:u) | mean | range | count |
|:----------------------------------:|:-----:|:--------------:|:-----:|
| Regressions ❌ <br /> (primary) | 0.4% | [0.3%, 0.6%] | 4 |
| Regressions ❌ <br /> (secondary) | 0.9% | [0.2%, 2.0%] | 7 |
| Improvements ✅ <br /> (primary) | -0.5% | [-0.5%, -0.5%] | 1 |
| Improvements ✅ <br /> (secondary) | - | - | 0 |
| All ❌✅ (primary) | 0.2% | [-0.5%, 0.6%] | 5 |

Loading