Skip to content

Commit dc7deb4

Browse files
rylevrylev
authored
Merge pull request #1808 from rylev/perf-triage-2024-01-30
2 parents 07a4e12 + 54a18b2 commit dc7deb4

File tree

1 file changed

+121
-0
lines changed

1 file changed

+121
-0
lines changed

‎triage/2024-01-30.md

Lines changed: 121 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,121 @@
1+
# 2024-01-30 Triage Log
2+
3+
This was a very quiet week with only one PR having any real impact on overall compiler performance. The removal of the internal `StructuralEq` trait saw a roughly 0.4% improvement on average across nearly 50 real-world benchmarks.
4+
5+
Triage done by **@rylev**.
6+
Revision range: [d6b151fc7..5c9c3c7](https://perf.rust-lang.org/?start=d6b151fc77e213bf637db0f12c1965ace3ffe255&end=5c9c3c7871d603ba13d38372830eca0c9013e575&absolute=false&stat=instructions%3Au)
7+
8+
**Summary**:
9+
10+
| (instructions:u) | mean | range | count |
11+
|:----------------------------------:|:-----:|:--------------:|:-----:|
12+
| Regressions ❌ <br /> (primary) | 0.5% | [0.3%, 0.7%] | 5 |
13+
| Regressions ❌ <br /> (secondary) | 0.5% | [0.2%, 1.4%] | 10 |
14+
| Improvements ✅ <br /> (primary) | -0.5% | [-1.5%, -0.2%] | 48 |
15+
| Improvements ✅ <br /> (secondary) | -2.3% | [-7.7%, -0.4%] | 36 |
16+
| All ❌✅ (primary) | -0.4% | [-1.5%, 0.7%] | 53 |
17+
18+
19+
0 Regressions, 4 Improvements, 4 Mixed; 3 of them in rollups
20+
37 artifact comparisons made in total
21+
22+
#### Regressions
23+
24+
25+
26+
#### Improvements
27+
28+
Only use dense bitsets in dataflow analyses [#116152](https://github.com/rust-lang/rust/pull/116152) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=8b94152af68a0ed6d6af0b5ba57491e40481008e&end=0e4243538b9119654c22dce688f8a63c81864de9&stat=instructions:u)
29+
30+
| (instructions:u) | mean | range | count |
31+
|:----------------------------------:|:-----:|:--------------:|:-----:|
32+
| Regressions ❌ <br /> (primary) | - | - | 0 |
33+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
34+
| Improvements ✅ <br /> (primary) | -0.4% | [-0.5%, -0.2%] | 5 |
35+
| Improvements ✅ <br /> (secondary) | -0.8% | [-1.1%, -0.4%] | 2 |
36+
| All ❌✅ (primary) | -0.4% | [-0.5%, -0.2%] | 5 |
37+
38+
39+
const-eval interning: get rid of type-driven traversal [#119044](https://github.com/rust-lang/rust/pull/119044) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=0e4243538b9119654c22dce688f8a63c81864de9&end=6265a95b3735a51cca1d48b8bbc3c8a6a035e15b&stat=instructions:u)
40+
41+
| (instructions:u) | mean | range | count |
42+
|:----------------------------------:|:-----:|:--------------:|:-----:|
43+
| Regressions ❌ <br /> (primary) | - | - | 0 |
44+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
45+
| Improvements ✅ <br /> (primary) | -0.9% | [-1.1%, -0.8%] | 3 |
46+
| Improvements ✅ <br /> (secondary) | -3.8% | [-8.0%, -0.6%] | 14 |
47+
| All ❌✅ (primary) | -0.9% | [-1.1%, -0.8%] | 3 |
48+
49+
50+
Remove all ConstPropNonsense [#119627](https://github.com/rust-lang/rust/pull/119627) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=7ffc697ce10f19447c0ce338428ae4b9bc0c041c&end=68411c955458ffbc98649d6dd057c4be3b187f38&stat=instructions:u)
51+
52+
| (instructions:u) | mean | range | count |
53+
|:----------------------------------:|:-----:|:--------------:|:-----:|
54+
| Regressions ❌ <br /> (primary) | - | - | 0 |
55+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
56+
| Improvements ✅ <br /> (primary) | -0.4% | [-0.6%, -0.2%] | 11 |
57+
| Improvements ✅ <br /> (secondary) | -0.5% | [-0.5%, -0.5%] | 3 |
58+
| All ❌✅ (primary) | -0.4% | [-0.6%, -0.2%] | 11 |
59+
60+
61+
remove StructuralEq trait [#116167](https://github.com/rust-lang/rust/pull/116167) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=0c1fb2a1e65a57073dc62dd502319a00cf753371&end=dd2559e08e1530806740931037d6bb83ef956161&stat=instructions:u)
62+
63+
| (instructions:u) | mean | range | count |
64+
|:----------------------------------:|:-----:|:--------------:|:-----:|
65+
| Regressions ❌ <br /> (primary) | 0.3% | [0.3%, 0.3%] | 1 |
66+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
67+
| Improvements ✅ <br /> (primary) | -0.4% | [-0.6%, -0.2%] | 48 |
68+
| Improvements ✅ <br /> (secondary) | -1.3% | [-3.0%, -0.3%] | 23 |
69+
| All ❌✅ (primary) | -0.4% | [-0.6%, 0.3%] | 49 |
70+
71+
72+
#### Mixed
73+
74+
Rollup of 10 pull requests [#120335](https://github.com/rust-lang/rust/pull/120335) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=d93feccb35183fa66fee77e7a2c9d4bf4d01695c&end=5bd5d214effd494f4bafb29b3a7a2f6c2070ca5c&stat=instructions:u)
75+
76+
| (instructions:u) | mean | range | count |
77+
|:----------------------------------:|:-----:|:--------------:|:-----:|
78+
| Regressions ❌ <br /> (primary) | 0.4% | [0.3%, 0.8%] | 9 |
79+
| Regressions ❌ <br /> (secondary) | 0.7% | [0.6%, 0.9%] | 2 |
80+
| Improvements ✅ <br /> (primary) | - | - | 0 |
81+
| Improvements ✅ <br /> (secondary) | -0.7% | [-0.7%, -0.7%] | 1 |
82+
| All ❌✅ (primary) | 0.4% | [0.3%, 0.8%] | 9 |
83+
- Small regressions in doc benchmarks
84+
- Most likely caused by [addings `AsyncFn` family of traits](https://github.com/rust-lang/rust/pull/119305).
85+
- Author has been pinged, but it's probably acceptable cost given that we're documenting more.
86+
87+
88+
Rollup of 9 pull requests [#120375](https://github.com/rust-lang/rust/pull/120375) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=cdd4ff8d8179197ba25708e0d85a4fc30f0a4971&end=e7bbe8ce933123a8e327c79e38bcb790595e6a65&stat=instructions:u)
89+
90+
| (instructions:u) | mean | range | count |
91+
|:----------------------------------:|:-----:|:--------------:|:-----:|
92+
| Regressions ❌ <br /> (primary) | - | - | 0 |
93+
| Regressions ❌ <br /> (secondary) | 0.3% | [0.3%, 0.4%] | 3 |
94+
| Improvements ✅ <br /> (primary) | -0.2% | [-0.2%, -0.2%] | 6 |
95+
| Improvements ✅ <br /> (secondary) | - | - | 0 |
96+
| All ❌✅ (primary) | -0.2% | [-0.2%, -0.2%] | 6 |
97+
- Small number of improvements outweigh the small number of regressions and so it's not really worth investigating further.
98+
99+
100+
Rollup of 12 pull requests [#120401](https://github.com/rust-lang/rust/pull/120401) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=c073f56a416b541fc66e46fb5265dad870da9229&end=b362939be16f9324dd9e6e36e22b606020068d75&stat=instructions:u)
101+
102+
| (instructions:u) | mean | range | count |
103+
|:----------------------------------:|:-----:|:--------------:|:-----:|
104+
| Regressions ❌ <br /> (primary) | - | - | 0 |
105+
| Regressions ❌ <br /> (secondary) | 0.6% | [0.5%, 1.1%] | 7 |
106+
| Improvements ✅ <br /> (primary) | - | - | 0 |
107+
| Improvements ✅ <br /> (secondary) | -3.1% | [-3.1%, -3.1%] | 1 |
108+
| All ❌✅ (primary) | - | - | 0 |
109+
- Very small regressions in match stress test is not worth investigating
110+
111+
112+
Merge into larger interval set [#120024](https://github.com/rust-lang/rust/pull/120024) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=6b4f1c5e782c72a047a23e922decd33e7d462345&end=635124704849eeead4e3a7bb6e663c5351571d93&stat=instructions:u)
113+
114+
| (instructions:u) | mean | range | count |
115+
|:----------------------------------:|:-----:|:--------------:|:-----:|
116+
| Regressions ❌ <br /> (primary) | 0.3% | [0.2%, 0.4%] | 14 |
117+
| Regressions ❌ <br /> (secondary) | 0.2% | [0.2%, 0.2%] | 1 |
118+
| Improvements ✅ <br /> (primary) | - | - | 0 |
119+
| Improvements ✅ <br /> (secondary) | -0.3% | [-0.3%, -0.2%] | 2 |
120+
| All ❌✅ (primary) | 0.3% | [0.2%, 0.4%] | 14 |
121+
- The reviewer decided that the performance hit here was worth it given that we are now protecting against truly extreme cases that the perf benchmark suite doesn't test against.

0 commit comments

Comments
 (0)