Skip to content

Commit a8651d5

Browse files
Mark-SimulacrumMark-Simulacrum
authored
Merge pull request #1885 from Mark-Simulacrum/triage
Add triage for this week
2 parents 39bfe5c + e3160fc commit a8651d5

File tree

1 file changed

+129
-0
lines changed

1 file changed

+129
-0
lines changed

‎triage/2024-04-01.md

+129
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,129 @@
1+
# 2024-04-01 Triage Log
2+
3+
A pretty quiet week, with most changes (dropped from the report below) being
4+
due to continuing bimodality in the performance data. No particularly notable
5+
changes landed.
6+
7+
Triage done by **@simulacrum**.
8+
Revision range: [73476d49904751f8d90ce904e16dfbc278083d2c..3d5528c287860b918e178a34f04ff903325571b3](https://perf.rust-lang.org/?start=73476d49904751f8d90ce904e16dfbc278083d2c&end=3d5528c287860b918e178a34f04ff903325571b3&absolute=false&stat=instructions%3Au)
9+
10+
**Summary**:
11+
12+
| (instructions:u) | mean | range | count |
13+
|:----------------------------------:|:-----:|:--------------:|:-----:|
14+
| Regressions ❌ <br /> (primary) | 0.5% | [0.2%, 1.1%] | 15 |
15+
| Regressions ❌ <br /> (secondary) | 0.5% | [0.2%, 1.8%] | 30 |
16+
| Improvements ✅ <br /> (primary) | -0.5% | [-1.6%, -0.2%] | 29 |
17+
| Improvements ✅ <br /> (secondary) | -0.7% | [-1.5%, -0.2%] | 11 |
18+
| All ❌✅ (primary) | -0.2% | [-1.6%, 1.1%] | 44 |
19+
20+
21+
1 Regressions, 2 Improvements, 5 Mixed; 0 of them in rollups
22+
61 artifact comparisons made in total
23+
24+
#### Regressions
25+
26+
Use FxIndexMap instead FxHashMap to stabilize iteration order in EffectiveVisibilities [#121851](https://github.com/rust-lang/rust/pull/121851) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=a8cfc83801301c2b4f0fd030192e268eeb15d473&end=bf71daedc29e7a240261acd1516378047e311a6f&stat=instructions:u)
27+
28+
| (instructions:u) | mean | range | count |
29+
|:----------------------------------:|:----:|:------------:|:-----:|
30+
| Regressions ❌ <br /> (primary) | 0.3% | [0.2%, 0.4%] | 13 |
31+
| Regressions ❌ <br /> (secondary) | 0.4% | [0.3%, 0.4%] | 3 |
32+
| Improvements ✅ <br /> (primary) | - | - | 0 |
33+
| Improvements ✅ <br /> (secondary) | - | - | 0 |
34+
| All ❌✅ (primary) | 0.3% | [0.2%, 0.4%] | 13 |
35+
36+
Justified by avoiding bugs in incremental compilation. See [comment](https://github.com/rust-lang/rust/pull/121851#issuecomment-1976192758) for details.
37+
38+
#### Improvements
39+
40+
Avoid some unnecessary query invocations. [#121387](https://github.com/rust-lang/rust/pull/121387) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=b13a71a2e77f4625d1a2b8a5b9488414686ebca9&end=519d892f9523fe40cc11ec07323ffc2792614742&stat=instructions:u)
41+
42+
| (instructions:u) | mean | range | count |
43+
|:----------------------------------:|:-----:|:--------------:|:-----:|
44+
| Regressions ❌ <br /> (primary) | - | - | 0 |
45+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
46+
| Improvements ✅ <br /> (primary) | -1.0% | [-1.7%, -0.3%] | 2 |
47+
| Improvements ✅ <br /> (secondary) | - | - | 0 |
48+
| All ❌✅ (primary) | -1.0% | [-1.7%, -0.3%] | 2 |
49+
50+
51+
Simplify proc macro bridge state [#122939](https://github.com/rust-lang/rust/pull/122939) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=519d892f9523fe40cc11ec07323ffc2792614742&end=536606bc5df7e6045c6dbedc2f852ba3d38581f1&stat=instructions:u)
52+
53+
| (instructions:u) | mean | range | count |
54+
|:----------------------------------:|:-----:|:--------------:|:-----:|
55+
| Regressions ❌ <br /> (primary) | - | - | 0 |
56+
| Regressions ❌ <br /> (secondary) | 0.3% | [0.3%, 0.3%] | 1 |
57+
| Improvements ✅ <br /> (primary) | -0.3% | [-0.4%, -0.3%] | 2 |
58+
| Improvements ✅ <br /> (secondary) | -1.2% | [-1.9%, -0.4%] | 6 |
59+
| All ❌✅ (primary) | -0.3% | [-0.4%, -0.3%] | 2 |
60+
61+
#### Mixed
62+
63+
Less generic code for Vec allocations [#122396](https://github.com/rust-lang/rust/pull/122396) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=0157da41ee9a1f6fb1a20e00ea62ab4122e2feaa&end=c9f8f3438a8134a413aa5d4903e0196e44e37bbc&stat=instructions:u)
64+
65+
| (instructions:u) | mean | range | count |
66+
|:----------------------------------:|:-----:|:--------------:|:-----:|
67+
| Regressions ❌ <br /> (primary) | 0.6% | [0.3%, 0.8%] | 2 |
68+
| Regressions ❌ <br /> (secondary) | 0.3% | [0.3%, 0.3%] | 1 |
69+
| Improvements ✅ <br /> (primary) | -0.5% | [-1.4%, -0.1%] | 12 |
70+
| Improvements ✅ <br /> (secondary) | -4.4% | [-4.4%, -4.4%] | 1 |
71+
| All ❌✅ (primary) | -0.3% | [-1.4%, 0.8%] | 14 |
72+
73+
Improvements mostly outweigh regressions, including the improvements in binary size.
74+
75+
Rollup of 8 pull requests [#123147](https://github.com/rust-lang/rust/pull/123147) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=9d709549481f77a8ab3111e7da2156d8d8181cdb&end=d779a7a25f67fced5f8fea232ef407c5b228a22f&stat=instructions:u)
76+
77+
| (instructions:u) | mean | range | count |
78+
|:----------------------------------:|:-----:|:--------------:|:-----:|
79+
| Regressions ❌ <br /> (primary) | 1.0% | [0.8%, 1.2%] | 2 |
80+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
81+
| Improvements ✅ <br /> (primary) | -0.4% | [-0.7%, -0.3%] | 3 |
82+
| Improvements ✅ <br /> (secondary) | - | - | 0 |
83+
| All ❌✅ (primary) | 0.2% | [-0.7%, 1.2%] | 5 |
84+
85+
Regressions appear to be potentially genuine, but also not particularly material, limited to just one benchmark.
86+
87+
Codegen const panic messages as function calls [#122671](https://github.com/rust-lang/rust/pull/122671) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=d74804636fa57e80d1e213fa9d2d65b27216b515&end=db2f9759f43167755d4eebb0a1358df9766a505e&stat=instructions:u)
88+
89+
| (instructions:u) | mean | range | count |
90+
|:----------------------------------:|:-----:|:--------------:|:-----:|
91+
| Regressions ❌ <br /> (primary) | 0.4% | [0.3%, 0.4%] | 2 |
92+
| Regressions ❌ <br /> (secondary) | 0.3% | [0.2%, 0.4%] | 7 |
93+
| Improvements ✅ <br /> (primary) | -0.7% | [-0.9%, -0.6%] | 3 |
94+
| Improvements ✅ <br /> (secondary) | - | - | 0 |
95+
| All ❌✅ (primary) | -0.3% | [-0.9%, 0.4%] | 5 |
96+
97+
This is primarily a win in binary sizes, particularly for code built with
98+
overflow checks which add many more callsites that are made cheaper with this
99+
optimization.
100+
101+
Eliminate `UbChecks` for non-standard libraries [#122975](https://github.com/rust-lang/rust/pull/122975) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=db2f9759f43167755d4eebb0a1358df9766a505e&end=760e567af5398a0d8c512f904e551e1f38e00d79&stat=instructions:u)
102+
103+
| (instructions:u) | mean | range | count |
104+
|:----------------------------------:|:-----:|:--------------:|:-----:|
105+
| Regressions ❌ <br /> (primary) | 0.9% | [0.3%, 1.7%] | 7 |
106+
| Regressions ❌ <br /> (secondary) | - | - | 0 |
107+
| Improvements ✅ <br /> (primary) | -0.5% | [-0.8%, -0.4%] | 5 |
108+
| Improvements ✅ <br /> (secondary) | -0.2% | [-0.3%, -0.1%] | 2 |
109+
| All ❌✅ (primary) | 0.3% | [-0.8%, 1.7%] | 12 |
110+
111+
Mixed results look fairly real. This change affects codegen of downstream
112+
libraries (i.e., most of the perf suite) so it's expected to lead to mostly
113+
improvements. It's not clear precisely where the regressions are coming from
114+
(maybe LLVM is able to optimize more?). The largest regression seems to resolve
115+
itself in #122976 (see next mixed PR in this report) but it's not clear that
116+
there is any direct relationship between the two.
117+
118+
Remove len argument from RawVec::reserve_for_push [#122976](https://github.com/rust-lang/rust/pull/122976) [(Comparison Link)](https://perf.rust-lang.org/compare.html?start=a3cfa031fa7726a957d73e6cad5744eb9706f56d&end=877d36b1928b5a4f7d193517b48290ecbe404d71&stat=instructions:u)
119+
120+
| (instructions:u) | mean | range | count |
121+
|:----------------------------------:|:-----:|:--------------:|:-----:|
122+
| Regressions ❌ <br /> (primary) | - | - | 0 |
123+
| Regressions ❌ <br /> (secondary) | 2.8% | [2.8%, 2.8%] | 1 |
124+
| Improvements ✅ <br /> (primary) | -1.7% | [-1.7%, -1.7%] | 1 |
125+
| Improvements ✅ <br /> (secondary) | -0.4% | [-0.4%, -0.3%] | 5 |
126+
| All ❌✅ (primary) | -1.7% | [-1.7%, -1.7%] | 1 |
127+
128+
Appears to primarily resolve regressions introduced by #122975 (previous mixed
129+
PR in this report). Potentially just noise as such.

0 commit comments

Comments
 (0)