Skip to content

Commit d729dd7

Browse files
Mark-SimulacrumMark-Simulacrum
authored
Merge pull request #2336 from Mark-Simulacrum/triage
Add perf triage for this week
2 parents 3cf7ac5 + e755985 commit d729dd7

File tree

1 file changed

+127
-0
lines changed

1 file changed

+127
-0
lines changed

‎triage/2025/2025-12-02.md‎

Lines changed: 127 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,127 @@
1+
# 2025-12-03 Triage Log
2+
3+
A fairly quiet week overall, despite a slightly higher than usual amount of merged PRs.
4+
5+
Triage done by **@simulacrum**.
6+
Revision range: [b64df9d1..eca9d93f](https://perf.rust-lang.org/?start=b64df9d1012f2482b54a4d959548cf8fc67e820c&end=eca9d93f9057f9a48ff691bd65e7daf2f94c1b67&absolute=false&stat=instructions%3Au)
7+
8+
**Summary**:
9+
10+
| (instructions:u) | mean | range | count |
11+
|:----------------------------------:|:-----:|:--------------:|:-----:|
12+
| Regressions ❌ <br /> (primary) | 0.5% | [0.2%, 1.0%] | 152 |
13+
| Regressions ❌ <br /> (secondary) | 0.7% | [0.1%, 2.4%] | 98 |
14+
| Improvements ✅ <br /> (primary) | -0.3% | [-1.3%, -0.1%] | 23 |
15+
| Improvements ✅ <br /> (secondary) | -0.3% | [-1.1%, -0.1%] | 53 |
16+
| All ❌✅ (primary) | 0.4% | [-1.3%, 1.0%] | 175 |
17+
18+
3 Regressions, 1 Improvement, 4 Mixed; 3 of them in rollups
19+
43 artifact comparisons made in total
20+
21+
#### Regressions
22+
23+
Rollup of 12 pull requests [#149351](https://github.com/rust-lang/rust/pull/149351) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=c797096598b80ba4a40397ad7e91584fcd3df5f1&end=99ca3fc4ec35e11d1f034143e8d8f0b79ef4c1d7&stat=instructions:u)
24+
25+
| (instructions:u) | mean | range | count |
26+
|:----------------------------------:|:----:|:------------:|:-----:|
27+
| Regressions ❌ <br /> (primary) | 0.1% | [0.1%, 0.1%] | 1 |
28+
| Regressions ❌ <br /> (secondary) | 0.8% | [0.3%, 1.3%] | 8 |
29+
| Improvements ✅ <br /> (primary) | - | - | 0 |
30+
| Improvements ✅ <br /> (secondary) | - | - | 0 |
31+
| All ❌✅ (primary) | 0.1% | [0.1%, 0.1%] | 1 |
32+
33+
This regressed due to [#147936](https://github.com/rust-lang/rust/pull/147936#issuecomment-3583927293)
34+
which added intrinsics for offload on the large-workspace benchmark. It's not
35+
clear why this had such a large effect though.
36+
37+
Rollup of 6 pull requests [#149387](https://github.com/rust-lang/rust/pull/149387) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=cf8a95590a1b768b7711f2631d5b5e3ead464de7&end=f392ed53ca3884c96430f2a283ffffa836b204f5&stat=instructions:u)
38+
39+
| (instructions:u) | mean | range | count |
40+
|:----------------------------------:|:----:|:------------:|:-----:|
41+
| Regressions ❌ <br /> (primary) | - | - | 0 |
42+
| Regressions ❌ <br /> (secondary) | 1.7% | [0.1%, 2.6%] | 4 |
43+
| Improvements ✅ <br /> (primary) | - | - | 0 |
44+
| Improvements ✅ <br /> (secondary) | - | - | 0 |
45+
| All ❌✅ (primary) | - | - | 0 |
46+
47+
This regressed some next-solver benchmarks. Root cause not investigated but surface level inspection suggests this is a bug fix so probably warranted, marked as triaged.
48+
49+
Rollup of 3 pull requests [#149462](https://github.com/rust-lang/rust/pull/149462) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=8a3a6bdb68b4d4c9ed922840808b02015741331e&end=3ff30e7eafc1da7104c3960187d17939172428ed&stat=instructions:u)
50+
51+
| (instructions:u) | mean | range | count |
52+
|:----------------------------------:|:-----:|:--------------:|:-----:|
53+
| Regressions ❌ <br /> (primary) | 0.5% | [0.1%, 1.0%] | 160 |
54+
| Regressions ❌ <br /> (secondary) | 0.6% | [0.1%, 2.0%] | 82 |
55+
| Improvements ✅ <br /> (primary) | - | - | 0 |
56+
| Improvements ✅ <br /> (secondary) | -0.5% | [-0.5%, -0.5%] | 1 |
57+
| All ❌✅ (primary) | 0.5% | [0.1%, 1.0%] | 160 |
58+
59+
This is due to ongoing work to support open APIs, see [comment](https://github.com/rust-lang/rust/pull/149462#issuecomment-3592715205) with analysis.
60+
61+
#### Improvements
62+
63+
Remove context-dependent `!` fallback [#148871](https://github.com/rust-lang/rust/pull/148871) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=88bd39beb342e2ae7ded1eb7d58697416686d679&end=d645a4c9c563b80048ce5f32845e754a67f11efa&stat=instructions:u)
64+
65+
| (instructions:u) | mean | range | count |
66+
|:----------------------------------:|:-----:|:--------------:|:-----:|
67+
| Regressions ❌ <br /> (primary) | - | - | 0 |
68+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
69+
| Improvements ✅ <br /> (primary) | -0.3% | [-0.4%, -0.1%] | 38 |
70+
| Improvements ✅ <br /> (secondary) | -0.3% | [-0.5%, -0.1%] | 40 |
71+
| All ❌✅ (primary) | -0.3% | [-0.4%, -0.1%] | 38 |
72+
73+
74+
#### Mixed
75+
76+
v0 mangling for std on nightly [#149148](https://github.com/rust-lang/rust/pull/149148) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=cdb4236e654a49c3035269588fe22dfafc0cfa3a&end=7934bbdf84a6b9a30297caf4f4f38286dedf876a&stat=instructions:u)
77+
78+
| (instructions:u) | mean | range | count |
79+
|:----------------------------------:|:-----:|:--------------:|:-----:|
80+
| Regressions ❌ <br /> (primary) | 0.8% | [0.6%, 0.8%] | 5 |
81+
| Regressions ❌ <br /> (secondary) | 0.5% | [0.0%, 0.9%] | 23 |
82+
| Improvements ✅ <br /> (primary) | - | - | 0 |
83+
| Improvements ✅ <br /> (secondary) | -0.3% | [-1.1%, -0.1%] | 5 |
84+
| All ❌✅ (primary) | 0.8% | [0.6%, 0.8%] | 5 |
85+
86+
Expected regressions due to binary size increases, see [comment](https://github.com/rust-lang/rust/pull/149148#issuecomment-3576262278).
87+
88+
Fix issue with callsite inline attribute not being applied sometimes. [#147404](https://github.com/rust-lang/rust/pull/147404) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=cc3eee7fbe17ea4b7238531cb97e1b7b8bd6afce&end=1eb0657f78777f0b4d6bcc49c126d5d35212cae5&stat=instructions:u)
89+
90+
| (instructions:u) | mean | range | count |
91+
|:----------------------------------:|:-----:|:--------------:|:-----:|
92+
| Regressions ❌ <br /> (primary) | 0.4% | [0.3%, 0.4%] | 2 |
93+
| Regressions ❌ <br /> (secondary) | 0.2% | [0.1%, 0.3%] | 5 |
94+
| Improvements ✅ <br /> (primary) | -0.1% | [-0.1%, -0.1%] | 2 |
95+
| Improvements ✅ <br /> (secondary) | -0.2% | [-0.4%, -0.1%] | 22 |
96+
| All ❌✅ (primary) | 0.1% | [-0.1%, 0.4%] | 4 |
97+
98+
Mixed results. Arguably a correctness fix, so marking as triaged. Regressions
99+
are due to doing a bit more work in codegen.
100+
101+
Compute jump threading opportunities in a single pass [#142821](https://github.com/rust-lang/rust/pull/142821) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=1d60f9e070c1039b263e0f035c0f03dfcc610d0f&end=4ad239f4156aa4e7df5ac9eb90ff0ab3d0089d1c&stat=instructions:u)
102+
103+
| (instructions:u) | mean | range | count |
104+
|:----------------------------------:|:-----:|:--------------:|:-----:|
105+
| Regressions ❌ <br /> (primary) | 0.4% | [0.1%, 0.8%] | 18 |
106+
| Regressions ❌ <br /> (secondary) | 0.2% | [0.1%, 1.0%] | 19 |
107+
| Improvements ✅ <br /> (primary) | -0.7% | [-1.4%, -0.3%] | 8 |
108+
| Improvements ✅ <br /> (secondary) | -0.3% | [-0.8%, -0.1%] | 27 |
109+
| All ❌✅ (primary) | 0.1% | [-1.4%, 0.8%] | 26 |
110+
111+
Mixed results as expected from MIR opt changes. Generally only a slight
112+
regression and the PR has other benefits (simplification of code and what looks
113+
like binary size wins).
114+
115+
Move even more early buffered lints to dyn lint diagnostics [#147634](https://github.com/rust-lang/rust/pull/147634) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=63b1db05801271e400954e41b8600a3cf1482363&end=47cd7120d9b4e1b64eb27c87522a07888197fae8&stat=instructions:u)
116+
117+
| (instructions:u) | mean | range | count |
118+
|:----------------------------------:|:-----:|:--------------:|:-----:|
119+
| Regressions ❌ <br /> (primary) | - | - | 0 |
120+
| Regressions ❌ <br /> (secondary) | 0.2% | [0.2%, 0.2%] | 1 |
121+
| Improvements ✅ <br /> (primary) | - | - | 0 |
122+
| Improvements ✅ <br /> (secondary) | -0.5% | [-0.9%, -0.1%] | 13 |
123+
| All ❌✅ (primary) | - | - | 0 |
124+
125+
Only a single secondary run regressed (ucd) so while it doesn't look spurious I
126+
don't think it merits investigation, especially coupled with some improvements.
127+
Marking as triaged.

0 commit comments

Comments
 (0)