Skip to content

Commit f9b3dec

Browse files
rylevrylev
committed
Add triage for 2023-01-03
1 parent 00440a4 commit f9b3dec

File tree

1 file changed

+121
-0
lines changed

1 file changed

+121
-0
lines changed

‎triage/2023-01-03.md

Lines changed: 121 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,121 @@
1+
# 2023-01-03 Triage Log
2+
3+
Fairly busy week with some massive performance improvements at the expense of some significant albeit smaller regressions. The main wins came in a long-standing PR from @cjgillot to enable encoding spans in metadata relative to their enclosing item. This causes more work in full compilation which causes some regressions up to 5% but can lead to very large wins in incremental compilation scenarios (up to ~70%). For example, the clap crate compiles 68% faster after a small 1 line change than it did previously.
4+
5+
Triage done by **@rylev**.
6+
Revision range: [b38a6d..b43596](https://perf.rust-lang.org/?start=b38a6d373cb254697411147c0e49cd2e84864258&end=b435960c4cfd3975651c7051be56d7f5d6c201ab&absolute=false&stat=instructions%3Au)
7+
8+
**Summary**:
9+
10+
| (instructions:u) | mean | range | count |
11+
|:----------------------------------:|:-----:|:---------------:|:-----:|
12+
| Regressions ❌ <br /> (primary) | 1.6% | [0.3%, 4.6%] | 97 |
13+
| Regressions ❌ <br /> (secondary) | 1.8% | [0.2%, 7.6%] | 60 |
14+
| Improvements ✅ <br /> (primary) | -9.7% | [-68.7%, -0.2%] | 53 |
15+
| Improvements ✅ <br /> (secondary) | -1.7% | [-15.3%, -0.1%] | 62 |
16+
| All ❌✅ (primary) | -2.4% | [-68.7%, 4.6%] | 150 |
17+
18+
1 Regressions, 1 Improvements, 4 Mixed; 1 of them in rollups
19+
47 artifact comparisons made in total
20+
21+
#### Regressions
22+
23+
Rollup of 8 pull requests [#106228](https://github.com/rust-lang/rust/pull/106228) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=270c94e484e19764a2832ef918c95224eb3f17c7&end=9709a438721d679572f0f7c389cd1fa60922fbef&stat=instructions:u)
24+
25+
| (instructions:u) | mean | range | count |
26+
|:----------------------------------:|:----:|:------------:|:-----:|
27+
| Regressions ❌ <br /> (primary) | 0.4% | [0.2%, 0.6%] | 7 |
28+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
29+
| Improvements ✅ <br /> (primary) | - | - | 0 |
30+
| Improvements ✅ <br /> (secondary) | - | - | 0 |
31+
| All ❌✅ (primary) | 0.4% | [0.2%, 0.6%] | 7 |
32+
- Kicked off a perf run on the only likely culprit. The regression is small enough that I don't think it's a big deal if we don't figure this out.
33+
34+
Allow trait method paths to satisfy const Fn bounds [#106210](https://github.com/rust-lang/rust/pull/106210) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=2c7536eaae59b44ff5259183ca98384129d69dd0&end=973a4db8d590d1ddfa4d07bd216aeffd28ba354e&stat=instructions:u)
35+
36+
| (instructions:u) | mean | range | count |
37+
|:----------------------------------:|:-----:|:--------------:|:-----:|
38+
| Regressions ❌ <br /> (primary) | 0.3% | [0.2%, 0.5%] | 16 |
39+
| Regressions ❌ <br /> (secondary) | 0.4% | [0.2%, 0.7%] | 16 |
40+
| Improvements ✅ <br /> (primary) | - | - | 0 |
41+
| Improvements ✅ <br /> (secondary) | -1.4% | [-1.4%, -1.4%] | 1 |
42+
| All ❌✅ (primary) | 0.3% | [0.2%, 0.5%] | 16 |
43+
- Massive amount of noise reversed in [#106268](https://github.com/rust-lang/rust/pull/106268). No idea where this noise comes from.
44+
45+
46+
#### Improvements
47+
48+
49+
fix comment for `TokenCursor::desugar` [#106268](https://github.com/rust-lang/rust/pull/106268) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=808be91da07c91e4d6006cca0ad78385785a741e&end=e5e5fcb0b758fcf7f149cc9206155dcfa18ec909&stat=instructions:u)
50+
51+
| (instructions:u) | mean | range | count |
52+
|:----------------------------------:|:-----:|:--------------:|:-----:|
53+
| Regressions ❌ <br /> (primary) | - | - | 0 |
54+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
55+
| Improvements ✅ <br /> (primary) | -0.3% | [-0.5%, -0.2%] | 17 |
56+
| Improvements ✅ <br /> (secondary) | -0.3% | [-0.7%, -0.2%] | 15 |
57+
| All ❌✅ (primary) | -0.3% | [-0.5%, -0.2%] | 17 |
58+
- Massive amount of noise coming in [#106210](https://github.com/rust-lang/rust/pull/106210) reversed here. No idea where this noise comes from.
59+
60+
61+
Abolish `QueryVTable` in favour of more assoc items on `QueryConfig` [#106307](https://github.com/rust-lang/rust/pull/106307) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=23b1cc197a3a6d0c1ab78a5c6e7cdad17681f1c9&end=d6f99e535a301a421dfee52a7c25bb4bdf420344&stat=instructions:u)
62+
63+
| (instructions:u) | mean | range | count |
64+
|:----------------------------------:|:-----:|:--------------:|:-----:|
65+
| Regressions ❌ <br /> (primary) | - | - | 0 |
66+
| Regressions ❌ <br /> (secondary) | 0.6% | [0.4%, 1.0%] | 4 |
67+
| Improvements ✅ <br /> (primary) | -0.5% | [-1.0%, -0.2%] | 63 |
68+
| Improvements ✅ <br /> (secondary) | -0.6% | [-1.3%, -0.4%] | 60 |
69+
| All ❌✅ (primary) | -0.5% | [-1.0%, -0.2%] | 63 |
70+
71+
72+
#### Mixed
73+
74+
Catch panics/unwinding in destruction of TLS values [#105426](https://github.com/rust-lang/rust/pull/105426) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=f6cc345be403d481b4518e151218118c2a9eb4bb&end=7c991868c60a4afc1ee6334b912ea96061a2c98d&stat=instructions:u)
75+
76+
| (instructions:u) | mean | range | count |
77+
|:----------------------------------:|:-----:|:---------------:|:-----:|
78+
| Regressions ❌ <br /> (primary) | 0.7% | [0.3%, 1.9%] | 11 |
79+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
80+
| Improvements ✅ <br /> (primary) | -0.3% | [-0.6%, -0.2%] | 41 |
81+
| Improvements ✅ <br /> (secondary) | -1.8% | [-14.6%, -0.2%] | 31 |
82+
| All ❌✅ (primary) | -0.1% | [-0.6%, 1.9%] | 52 |
83+
- Very much unsure where the performance changes are coming from. Some show LLVM regressions but not all. LLVM also does not explain the improvements.
84+
- Ask [in the PR](https://github.com/rust-lang/rust/pull/105426#issuecomment-1370051574) for thoughts.
85+
86+
87+
Use some more `const_eval_select` in pointer methods for compile times [#106275](https://github.com/rust-lang/rust/pull/106275) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=726bbfc8f0d6eb431dfa620c1c8fc3848020cd3c&end=dd01a163c1cff9461ddd85e7b3c2c0109657658f&stat=instructions:u)
88+
89+
| (instructions:u) | mean | range | count |
90+
|:----------------------------------:|:-----:|:--------------:|:-----:|
91+
| Regressions ❌ <br /> (primary) | 0.4% | [0.4%, 0.4%] | 1 |
92+
| Regressions ❌ <br /> (secondary) | 0.7% | [0.6%, 0.9%] | 2 |
93+
| Improvements ✅ <br /> (primary) | -0.5% | [-0.9%, -0.2%] | 28 |
94+
| Improvements ✅ <br /> (secondary) | -0.8% | [-0.8%, -0.7%] | 3 |
95+
| All ❌✅ (primary) | -0.4% | [-0.9%, 0.4%] | 29 |
96+
- Author investigated why this might have caused a regression in `regex` and found nothing. The gains much outweigh the regressions so we should be fine keeping it.
97+
98+
99+
Encode spans relative to the enclosing item -- enable on nightly [#84762](https://github.com/rust-lang/rust/pull/84762) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=f89003eda8917ff99f8ee3fb5c812310a58c014b&end=fb9dfa8ceffb985105be3176d7ed2f99515ea377&stat=instructions:u)
100+
101+
| (instructions:u) | mean | range | count |
102+
|:----------------------------------:|:------:|:---------------:|:-----:|
103+
| Regressions ❌ <br /> (primary) | 1.7% | [0.2%, 5.0%] | 108 |
104+
| Regressions ❌ <br /> (secondary) | 1.8% | [0.2%, 7.6%] | 90 |
105+
| Improvements ✅ <br /> (primary) | -20.0% | [-68.9%, -0.6%] | 25 |
106+
| Improvements ✅ <br /> (secondary) | -2.7% | [-4.8%, -1.5%] | 11 |
107+
| All ❌✅ (primary) | -2.4% | [-68.9%, 5.0%] | 133 |
108+
- There was an [FCP](https://github.com/rust-lang/rust/pull/84762#issuecomment-1327518372) to decide whether the trade off of some regressions for some huge incremental performance wins was worth it.
109+
- This FCP passed and so the PR was merged.
110+
111+
112+
Reenable limited top-down MIR inlining [#106364](https://github.com/rust-lang/rust/pull/106364) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=d6f99e535a301a421dfee52a7c25bb4bdf420344&end=67d16171d7eebc08057a46aeed887252e7cc286b&stat=instructions:u)
113+
114+
| (instructions:u) | mean | range | count |
115+
|:----------------------------------:|:-----:|:--------------:|:-----:|
116+
| Regressions ❌ <br /> (primary) | 0.5% | [0.3%, 1.1%] | 4 |
117+
| Regressions ❌ <br /> (secondary) | 0.6% | [0.5%, 0.6%] | 4 |
118+
| Improvements ✅ <br /> (primary) | -0.6% | [-1.2%, -0.2%] | 9 |
119+
| Improvements ✅ <br /> (secondary) | -0.9% | [-2.4%, -0.2%] | 18 |
120+
| All ❌✅ (primary) | -0.2% | [-1.2%, 1.1%] | 13 |
121+
- As pointed out [here](https://github.com/rust-lang/rust/pull/106364#issuecomment-1369111291) this was effectively a performance wash with regressions and improvements canceling themselves out.

0 commit comments

Comments
 (0)