Skip to content

Commit f38de3c

Browse files
Mark-SimulacrumMark-Simulacrum
authored
Add triage for this week (#1693)
1 parent 481d7ba commit f38de3c

File tree

1 file changed

+159
-0
lines changed

1 file changed

+159
-0
lines changed

‎triage/2023-08-08.md

Lines changed: 159 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,159 @@
1+
# 2023-08-08 Triage Log
2+
3+
Overall a very positive week, primarily due to an [upgrade to LLVM
4+
17](https://github.com/rust-lang/rust/pull/114048) and some changes to [lint
5+
execution](https://github.com/rust-lang/rust/pull/113734). Memory usage is down
6+
[4-7%](https://perf.rust-lang.org/?start=828bdc2c26f5c95773c4ecf72870919f16417b66&end=443c3161dd04f4c1b656a626f9079921bee9c326&absolute=false&stat=max-rss&kind=percentfromfirst)
7+
over the last week and wall times are down
8+
[3-5%](https://perf.rust-lang.org/?start=828bdc2c26f5c95773c4ecf72870919f16417b66&end=443c3161dd04f4c1b656a626f9079921bee9c326&absolute=false&stat=wall-time&kind=percentfromfirst).
9+
10+
Triage done by **@simulacrum**.
11+
Revision range: [828bdc2c26f5c95773c4ecf72870919f16417b66..443c3161dd04f4c1b656a626f9079921bee9c326](https://perf.rust-lang.org/?start=828bdc2c26f5c95773c4ecf72870919f16417b66&end=443c3161dd04f4c1b656a626f9079921bee9c326&absolute=false&stat=instructions%3Au)
12+
13+
**Summary**:
14+
15+
| (instructions:u) | mean | range | count |
16+
|:----------------------------------:|:-----:|:---------------:|:-----:|
17+
| Regressions ❌ <br /> (primary) | - | - | 0 |
18+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
19+
| Improvements ✅ <br /> (primary) | -3.1% | [-7.1%, -1.2%] | 253 |
20+
| Improvements ✅ <br /> (secondary) | -4.1% | [-28.4%, -0.3%] | 233 |
21+
| All ❌✅ (primary) | -3.1% | [-7.1%, -1.2%] | 253 |
22+
23+
24+
2 Regressions, 7 Improvements, 2 Mixed; 2 of them in rollups
25+
64 artifact comparisons made in total
26+
27+
#### Regressions
28+
29+
Rollup of 9 pull requests [#114481](https://github.com/rust-lang/rust/pull/114481) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=fe896efa974ea759b10b9754234ae61e9cab28b3&end=e4c14468460b22ac7a289a422fb7fee606e12a0b&stat=instructions:u)
30+
31+
| (instructions:u) | mean | range | count |
32+
|:----------------------------------:|:----:|:------------:|:-----:|
33+
| Regressions ❌ <br /> (primary) | 0.8% | [0.4%, 1.5%] | 80 |
34+
| Regressions ❌ <br /> (secondary) | 0.7% | [0.2%, 1.2%] | 25 |
35+
| Improvements ✅ <br /> (primary) | - | - | 0 |
36+
| Improvements ✅ <br /> (secondary) | - | - | 0 |
37+
| All ❌✅ (primary) | 0.8% | [0.4%, 1.5%] | 80 |
38+
39+
Unclear what is causing this regression. It's entirely limited to the
40+
incr-patched and incr-unchanged scenarios, and digging into one of the
41+
benchmarks (bitmaps-3.1.0 check incr-unchanged) seems to show a mix of noise
42+
and, in aggregate, significant improvements over the last several days. Not
43+
clear which PR in this rollup could have caused this change in behavior.
44+
45+
See [these comments](https://github.com/rust-lang/rust/pull/114481#issuecomment-1666438972) for some more details.
46+
47+
Re-enable atomic loads and stores for all RISC-V targets [#98333](https://github.com/rust-lang/rust/pull/98333) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=e173a8e6634d787a1529ff6ec5dc63273425ca34&end=90f0b24ad3e7fc0dc0e419c9da30d74629cd5736&stat=instructions:u)
48+
49+
| (instructions:u) | mean | range | count |
50+
|:----------------------------------:|:----:|:------------:|:-----:|
51+
| Regressions ❌ <br /> (primary) | 0.8% | [0.4%, 1.3%] | 18 |
52+
| Regressions ❌ <br /> (secondary) | 0.6% | [0.2%, 0.9%] | 12 |
53+
| Improvements ✅ <br /> (primary) | - | - | 0 |
54+
| Improvements ✅ <br /> (secondary) | - | - | 0 |
55+
| All ❌✅ (primary) | 0.8% | [0.4%, 1.3%] | 18 |
56+
57+
This PR has a very similar regression profile to the previous rollup in this list (#114481).
58+
59+
#### Improvements
60+
61+
Remove `constness` from `TraitPredicate` [#114202](https://github.com/rust-lang/rust/pull/114202) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=7637653b9fdd2b2f2dc3b378891eb5cf6aea9f5d&end=20747af8940cf683970e76ec9a433305940c05f3&stat=instructions:u)
62+
63+
| (instructions:u) | mean | range | count |
64+
|:----------------------------------:|:-----:|:--------------:|:-----:|
65+
| Regressions ❌ <br /> (primary) | - | - | 0 |
66+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
67+
| Improvements ✅ <br /> (primary) | -0.7% | [-0.9%, -0.4%] | 16 |
68+
| Improvements ✅ <br /> (secondary) | -0.6% | [-1.2%, -0.3%] | 13 |
69+
| All ❌✅ (primary) | -0.7% | [-0.9%, -0.4%] | 16 |
70+
71+
cg_llvm: stop identifying ADTs in LLVM IR [#114350](https://github.com/rust-lang/rust/pull/114350) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=098c1db7e55e7229a520a2ec8e460e1e300e109b&end=73dc6f03a2a8c0e8b8b25fb97d7f4805995ce0b9&stat=instructions:u)
72+
73+
| (instructions:u) | mean | range | count |
74+
|:----------------------------------:|:-----:|:--------------:|:-----:|
75+
| Regressions ❌ <br /> (primary) | - | - | 0 |
76+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
77+
| Improvements ✅ <br /> (primary) | -0.8% | [-1.2%, -0.6%] | 5 |
78+
| Improvements ✅ <br /> (secondary) | -1.1% | [-1.5%, -0.6%] | 6 |
79+
| All ❌✅ (primary) | -0.8% | [-1.2%, -0.6%] | 5 |
80+
81+
Convert builtin "global" late lints to run per module [#113734](https://github.com/rust-lang/rust/pull/113734) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=fca59ab5f0e7df7d816bed77a32abc0045ebe80b&end=67626b8e89ff57269f1d67cb9bc037088c058b42&stat=instructions:u)
82+
83+
| (instructions:u) | mean | range | count |
84+
|:----------------------------------:|:-----:|:---------------:|:-----:|
85+
| Regressions ❌ <br /> (primary) | - | - | 0 |
86+
| Regressions ❌ <br /> (secondary) | 0.5% | [0.4%, 0.8%] | 3 |
87+
| Improvements ✅ <br /> (primary) | -1.7% | [-3.4%, -0.2%] | 88 |
88+
| Improvements ✅ <br /> (secondary) | -4.2% | [-26.5%, -0.2%] | 85 |
89+
| All ❌✅ (primary) | -1.7% | [-3.4%, -0.2%] | 88 |
90+
91+
parent_module_from_def_id does not need to be a query. [#114516](https://github.com/rust-lang/rust/pull/114516) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=e59540968e0c6fc3c37d92a639b83fc8fcee5e81&end=5973bfbd38fcefb63090a588733931e933498dfd&stat=instructions:u)
92+
93+
| (instructions:u) | mean | range | count |
94+
|:----------------------------------:|:-----:|:--------------:|:-----:|
95+
| Regressions ❌ <br /> (primary) | 0.3% | [0.2%, 0.4%] | 2 |
96+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
97+
| Improvements ✅ <br /> (primary) | -1.4% | [-2.0%, -1.1%] | 17 |
98+
| Improvements ✅ <br /> (secondary) | -1.2% | [-1.5%, -0.3%] | 11 |
99+
| All ❌✅ (primary) | -1.2% | [-2.0%, 0.4%] | 19 |
100+
101+
Steal MIR for CTFE when possible. [#114502](https://github.com/rust-lang/rust/pull/114502) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=85fbb571497d13cfb828de9b0d3e78656b9203c1&end=f3623871cfa0763c95ebd6ceafaa6dc2e44ca68f&stat=instructions:u)
102+
103+
| (instructions:u) | mean | range | count |
104+
|:----------------------------------:|:-----:|:--------------:|:-----:|
105+
| Regressions ❌ <br /> (primary) | - | - | 0 |
106+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
107+
| Improvements ✅ <br /> (primary) | -0.6% | [-0.6%, -0.6%] | 1 |
108+
| Improvements ✅ <br /> (secondary) | -1.5% | [-2.6%, -0.5%] | 19 |
109+
| All ❌✅ (primary) | -0.6% | [-0.6%, -0.6%] | 1 |
110+
111+
Update to LLVM 17 [#114048](https://github.com/rust-lang/rust/pull/114048) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=03a119b0b0e310d22d94399b24ed030056050f13&end=443c3161dd04f4c1b656a626f9079921bee9c326&stat=instructions:u)
112+
113+
| (instructions:u) | mean | range | count |
114+
|:----------------------------------:|:-----:|:---------------:|:-----:|
115+
| Regressions ❌ <br /> (primary) | - | - | 0 |
116+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
117+
| Improvements ✅ <br /> (primary) | -2.2% | [-5.6%, -0.8%] | 173 |
118+
| Improvements ✅ <br /> (secondary) | -2.2% | [-12.9%, -0.5%] | 199 |
119+
| All ❌✅ (primary) | -2.2% | [-5.6%, -0.8%] | 173 |
120+
121+
This represents a large win across multiple benchmarks. Unfortunately it also
122+
regresses artifact sizes for libLLVM.so by 50MB. Investigation into why is
123+
[ongoing](https://github.com/rust-lang/rust/pull/114048#issuecomment-1669252031).
124+
125+
Do not run ConstProp on mir_for_ctfe. [#114459](https://github.com/rust-lang/rust/pull/114459) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=67626b8e89ff57269f1d67cb9bc037088c058b42&end=1cabb8ed2301faf193132e71ca5b10f82913e48a&stat=instructions:u)
126+
127+
| (instructions:u) | mean | range | count |
128+
|:----------------------------------:|:-----:|:--------------:|:-----:|
129+
| Regressions ❌ <br /> (primary) | 0.2% | [0.2%, 0.2%] | 1 |
130+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
131+
| Improvements ✅ <br /> (primary) | -0.8% | [-1.2%, -0.5%] | 8 |
132+
| Improvements ✅ <br /> (secondary) | -1.7% | [-3.0%, -1.1%] | 7 |
133+
| All ❌✅ (primary) | -0.7% | [-1.2%, 0.2%] | 9 |
134+
135+
The one regression is actually within noise floor for the relevant benchmark.
136+
137+
#### Mixed
138+
139+
Optimize `Iterator` implementation for `&mut impl Iterator + Sized` [#111200](https://github.com/rust-lang/rust/pull/111200) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=9565b68b741fd0466b3af075139e819e7d73ee38&end=eb088b8b9d98f1af1b0e61bbdcd8686e1b0db7b6&stat=instructions:u)
140+
141+
| (instructions:u) | mean | range | count |
142+
|:----------------------------------:|:-----:|:--------------:|:-----:|
143+
| Regressions ❌ <br /> (primary) | 0.4% | [0.3%, 0.5%] | 6 |
144+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
145+
| Improvements ✅ <br /> (primary) | - | - | 0 |
146+
| Improvements ✅ <br /> (secondary) | -0.7% | [-0.9%, -0.5%] | 2 |
147+
| All ❌✅ (primary) | 0.4% | [0.3%, 0.5%] | 6 |
148+
149+
Rollup of 7 pull requests [#114604](https://github.com/rust-lang/rust/pull/114604) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=443c3161dd04f4c1b656a626f9079921bee9c326&end=8e7fd551311d424e4e63fa45906a2a928fce96a7&stat=instructions:u)
150+
151+
| (instructions:u) | mean | range | count |
152+
|:----------------------------------:|:-----:|:--------------:|:-----:|
153+
| Regressions ❌ <br /> (primary) | 1.4% | [0.4%, 2.0%] | 3 |
154+
| Regressions ❌ <br /> (secondary) | 1.2% | [1.1%, 1.3%] | 6 |
155+
| Improvements ✅ <br /> (primary) | -0.7% | [-0.9%, -0.4%] | 2 |
156+
| Improvements ✅ <br /> (secondary) | -1.0% | [-1.5%, -0.5%] | 2 |
157+
| All ❌✅ (primary) | 0.5% | [-0.9%, 2.0%] | 5 |
158+
159+
Investigation into cause is [ongoing](https://github.com/rust-lang/rust/pull/114604#issuecomment-1669350724).

0 commit comments

Comments
 (0)