replace Criterion with Divan (#12000)

<!--
if this PR closes one or more issues, you can automatically link the PR
with
them by using one of the [*linking
keywords*](https://docs.github.com/en/issues/tracking-your-work-with-issues/linking-a-pull-request-to-an-issue#linking-a-pull-request-to-an-issue-using-a-keyword),
e.g.
- this PR should close #xxxx
- fixes #xxxx

you can also mention related issues, PRs or discussions!
-->

# Description
<!--
Thank you for improving Nushell. Please, check our [contributing
guide](../CONTRIBUTING.md) and talk to the core team before making major
changes.

Description of your pull request goes here. **Provide examples and/or
screenshots** if your changes affect the user experience.
--> 
This is a test of changing out the current criterion microbenchmark tool
to [Divan](https://nikolaivazquez.com/blog/divan/), a new and more
straightforward microbenchmark suit.
Itself states it is robust to noise, and even allow it to be used in CI
settings. It by default has no external dependencies and is very fast to
run, the sampling method allows it to be a lot faster compared to
criterion requiring less samples.
The output is also nicely displayed and easy to get a quick overview of
the performance.


![image](https://github.com/nushell/nushell/assets/17986183/587a1fb1-1da3-402c-b668-a27fde9a0657)


# User-Facing Changes
<!-- List of all changes that impact the user experience here. This
helps us keep track of breaking changes. -->

# Tests + Formatting
<!--
Don't forget to add tests that cover your changes.

Make sure you've run and fixed any issues with these commands:

- `cargo fmt --all -- --check` to check standard code formatting (`cargo
fmt --all` applies these changes)
- `cargo clippy --workspace -- -D warnings -D clippy::unwrap_used` to
check that you're using the standard code style
- `cargo test --workspace` to check that all tests pass (on Windows make
sure to [enable developer
mode](https://learn.microsoft.com/en-us/windows/apps/get-started/developer-mode-features-and-debugging))
- `cargo run -- -c "use std testing; testing run-tests --path
crates/nu-std"` to run the tests for the standard library

> **Note**
> from `nushell` you can also use the `toolkit` as follows
> ```bash
> use toolkit.nu # or use an `env_change` hook to activate it
automatically
> toolkit check pr
> ```
-->

# After Submitting
<!-- If your PR had any user-facing changes, update [the
documentation](https://github.com/nushell/nushell.github.io) after the
PR is merged, if necessary. This will help us keep the docs up to date.
-->
This commit is contained in:
Filip Andersson 2024-03-01 19:09:21 +01:00 committed by GitHub
parent 38a42905ae
commit 9ff92c6878
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
4 changed files with 187 additions and 273 deletions

198
Cargo.lock generated
View file

@ -98,12 +98,6 @@ dependencies = [
"libc",
]
[[package]]
name = "anes"
version = "0.1.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4b46cbb362ab8752921c97e041f5e366ee6297bd428a31275b9fcf1e380f7299"
[[package]]
name = "ansi-str"
version = "0.8.0"
@ -599,12 +593,6 @@ version = "0.3.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "df8670b8c7b9dae1793364eafadf7239c40d669904660c5960d74cfd80b46a53"
[[package]]
name = "cast"
version = "0.3.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "37b2a672a2cb129a2e41c10b1224bb368f9f37a2b16b612598138befd7b37eb5"
[[package]]
name = "castaway"
version = "0.2.2"
@ -723,33 +711,6 @@ dependencies = [
"stacker",
]
[[package]]
name = "ciborium"
version = "0.2.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "42e69ffd6f0917f5c029256a24d0161db17cea3997d185db0d35926308770f0e"
dependencies = [
"ciborium-io",
"ciborium-ll",
"serde",
]
[[package]]
name = "ciborium-io"
version = "0.2.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "05afea1e0a06c9be33d539b876f1ce3692f4afea2cb41f740e7743225ed1c757"
[[package]]
name = "ciborium-ll"
version = "0.2.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "57663b653d948a338bfb3eeba9bb2fd5fcfaecb9e199e87e1eda4d9e8b240fd9"
dependencies = [
"ciborium-io",
"half",
]
[[package]]
name = "clang-sys"
version = "1.7.0"
@ -840,6 +801,12 @@ dependencies = [
"static_assertions",
]
[[package]]
name = "condtype"
version = "1.3.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "baf0a07a401f374238ab8e2f11a104d2851bf9ce711ec69804834de8af45c7af"
[[package]]
name = "console"
version = "0.15.8"
@ -942,42 +909,6 @@ dependencies = [
"cfg-if",
]
[[package]]
name = "criterion"
version = "0.5.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f2b12d017a929603d80db1831cd3a24082f8137ce19c69e6447f54f5fc8d692f"
dependencies = [
"anes",
"cast",
"ciborium",
"clap",
"criterion-plot",
"is-terminal",
"itertools 0.10.5",
"num-traits",
"once_cell",
"oorandom",
"plotters",
"rayon",
"regex",
"serde",
"serde_derive",
"serde_json",
"tinytemplate",
"walkdir",
]
[[package]]
name = "criterion-plot"
version = "0.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6b50826342786a51a89e2da3a28f1c32b06e387201bc2d19791f622c673706b1"
dependencies = [
"cast",
"itertools 0.10.5",
]
[[package]]
name = "crossbeam-channel"
version = "0.5.11"
@ -1216,6 +1147,31 @@ dependencies = [
"winapi",
]
[[package]]
name = "divan"
version = "0.1.14"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a0d567df2c9c2870a43f3f2bd65aaeb18dbce1c18f217c3e564b4fbaeb3ee56c"
dependencies = [
"cfg-if",
"clap",
"condtype",
"divan-macros",
"libc",
"regex-lite",
]
[[package]]
name = "divan-macros"
version = "0.1.14"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "27540baf49be0d484d8f0130d7d8da3011c32a44d4fc873368154f1510e574a2"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.48",
]
[[package]]
name = "dlib"
version = "0.5.2"
@ -1751,16 +1707,6 @@ dependencies = [
"tracing",
]
[[package]]
name = "half"
version = "2.3.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bc52e53916c08643f1b56ec082790d1e86a32e58dc5268f897f313fbae7b4872"
dependencies = [
"cfg-if",
"crunchy",
]
[[package]]
name = "halfbrown"
version = "0.2.4"
@ -2064,17 +2010,6 @@ dependencies = [
"once_cell",
]
[[package]]
name = "is-terminal"
version = "0.4.10"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0bad00257d07be169d870ab665980b06cdb366d792ad690bf2e76876dc503455"
dependencies = [
"hermit-abi",
"rustix",
"windows-sys 0.52.0",
]
[[package]]
name = "is-wsl"
version = "0.4.0"
@ -2106,15 +2041,6 @@ dependencies = [
"winapi",
]
[[package]]
name = "itertools"
version = "0.10.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b0fd2260e829bddf4cb6ea802289de2f86d6a7a690192fbe91b3f46e0f2c8473"
dependencies = [
"either",
]
[[package]]
name = "itertools"
version = "0.11.0"
@ -2791,9 +2717,9 @@ name = "nu"
version = "0.90.2"
dependencies = [
"assert_cmd",
"criterion",
"crossterm",
"ctrlc",
"divan",
"log",
"miette",
"mimalloc",
@ -3551,12 +3477,6 @@ version = "1.19.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3fdb12b2476b595f9358c5161aa467c2438859caa136dec86c26fdd2efe17b92"
[[package]]
name = "oorandom"
version = "11.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0ab1bc2a289d34bd04a330323ac98a1b4bc82c9d9fcb1e66b63caa84da26b575"
[[package]]
name = "open"
version = "5.0.1"
@ -3914,34 +3834,6 @@ dependencies = [
"array-init-cursor",
]
[[package]]
name = "plotters"
version = "0.3.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d2c224ba00d7cadd4d5c660deaf2098e5e80e07846537c51f9cfa4be50c1fd45"
dependencies = [
"num-traits",
"plotters-backend",
"plotters-svg",
"wasm-bindgen",
"web-sys",
]
[[package]]
name = "plotters-backend"
version = "0.3.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9e76628b4d3a7581389a35d5b6e2139607ad7c75b17aed325f210aa91f4a9609"
[[package]]
name = "plotters-svg"
version = "0.3.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "38f6d39893cca0701371e3c27294f09797214b86f1fb951b89ade8ec04e2abab"
dependencies = [
"plotters-backend",
]
[[package]]
name = "polars"
version = "0.37.0"
@ -4753,6 +4645,12 @@ dependencies = [
"regex-syntax",
]
[[package]]
name = "regex-lite"
version = "0.1.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "30b661b2f27137bdbc16f00eda72866a92bb28af1753ffbd56744fb6e2e9cd8e"
[[package]]
name = "regex-syntax"
version = "0.8.2"
@ -5808,16 +5706,6 @@ dependencies = [
"crunchy",
]
[[package]]
name = "tinytemplate"
version = "1.2.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "be4d6b5f19ff7664e8c98d03e2139cb510db9b0a60b55f8e8709b689d939b6bc"
dependencies = [
"serde",
"serde_json",
]
[[package]]
name = "tinyvec"
version = "1.6.0"
@ -6506,16 +6394,6 @@ dependencies = [
"pkg-config",
]
[[package]]
name = "web-sys"
version = "0.3.67"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "58cd2333b6e0be7a39605f0e255892fd7418a682d8da8fe042fe25128794d2ed"
dependencies = [
"js-sys",
"wasm-bindgen",
]
[[package]]
name = "which"
version = "6.0.0"

View file

@ -102,7 +102,7 @@ nix = { version = "0.27", default-features = false, features = [
[dev-dependencies]
nu-test-support = { path = "./crates/nu-test-support", version = "0.90.2" }
assert_cmd = "2.0"
criterion = "0.5"
divan = "0.1.14"
pretty_assertions = "1.4"
rstest = { version = "0.18", default-features = false }
serial_test = "3.0"
@ -182,7 +182,6 @@ bench = false
reedline = { git = "https://github.com/nushell/reedline", branch = "main" }
# nu-ansi-term = {git = "https://github.com/nushell/nu-ansi-term.git", branch = "main"}
# Criterion benchmarking setup
# Run all benchmarks with `cargo bench`
# Run individual benchmarks like `cargo bench -- <regex>` e.g. `cargo bench -- parse`
[[bench]]

View file

@ -1,6 +1,6 @@
# Criterion benchmarks
# Divan benchmarks
These are benchmarks using [Criterion](https://github.com/bheisler/criterion.rs), a microbenchmarking tool for Rust.
These are benchmarks using [Divan](https://github.com/nvzqz/divan), a microbenchmarking tool for Rust.
Run all benchmarks with `cargo bench`

View file

@ -1,4 +1,3 @@
use criterion::{criterion_group, criterion_main, BatchSize, Criterion};
use nu_cli::eval_source;
use nu_parser::parse;
use nu_plugin::{Encoder, EncodingType, PluginCallResponse, PluginOutput};
@ -8,6 +7,11 @@ use nu_protocol::{
use nu_utils::{get_default_config, get_default_env};
use std::path::{Path, PathBuf};
fn main() {
// Run registered benchmarks.
divan::main();
}
fn load_bench_commands() -> EngineState {
nu_command::add_shell_command_context(nu_cmd_lang::create_default_context())
}
@ -31,41 +35,7 @@ fn get_home_path(engine_state: &EngineState) -> PathBuf {
.unwrap_or_default()
}
// FIXME: All benchmarks live in this 1 file to speed up build times when benchmarking.
// When the *_benchmarks functions were in different files, `cargo bench` would build
// an executable for every single one - incredibly slowly. Would be nice to figure out
// a way to split things up again.
fn parser_benchmarks(c: &mut Criterion) {
let mut engine_state = load_bench_commands();
let home_path = get_home_path(&engine_state);
// parsing config.nu breaks without PWD set, so set a valid path
engine_state.add_env_var(
"PWD".into(),
Value::string(home_path.to_string_lossy(), Span::test_data()),
);
let default_env = get_default_env().as_bytes();
c.bench_function("parse_default_env_file", |b| {
b.iter_batched(
|| nu_protocol::engine::StateWorkingSet::new(&engine_state),
|mut working_set| parse(&mut working_set, None, default_env, false),
BatchSize::SmallInput,
)
});
let default_config = get_default_config().as_bytes();
c.bench_function("parse_default_config_file", |b| {
b.iter_batched(
|| nu_protocol::engine::StateWorkingSet::new(&engine_state),
|mut working_set| parse(&mut working_set, None, default_config, false),
BatchSize::SmallInput,
)
});
}
fn eval_benchmarks(c: &mut Criterion) {
fn setup_engine() -> EngineState {
let mut engine_state = load_bench_commands();
let home_path = get_home_path(&engine_state);
@ -79,33 +49,78 @@ fn eval_benchmarks(c: &mut Criterion) {
.expect("Failed to create nushell constant.");
engine_state.set_variable_const_val(NU_VARIABLE_ID, nu_const);
c.bench_function("eval default_env.nu", |b| {
b.iter(|| {
let mut stack = nu_protocol::engine::Stack::new();
eval_source(
&mut engine_state,
&mut stack,
get_default_env().as_bytes(),
"default_env.nu",
PipelineData::empty(),
false,
)
})
});
engine_state
}
c.bench_function("eval default_config.nu", |b| {
b.iter(|| {
let mut stack = nu_protocol::engine::Stack::new();
// FIXME: All benchmarks live in this 1 file to speed up build times when benchmarking.
// When the *_benchmarks functions were in different files, `cargo bench` would build
// an executable for every single one - incredibly slowly. Would be nice to figure out
// a way to split things up again.
#[divan::bench_group()]
mod parser_benchmarks {
use super::*;
#[divan::bench()]
fn parse_default_config_file(bencher: divan::Bencher) {
let engine_state = setup_engine();
let default_env = get_default_config().as_bytes();
bencher
.with_inputs(|| nu_protocol::engine::StateWorkingSet::new(&engine_state))
.bench_refs(|mut working_set| parse(&mut working_set, None, default_env, false))
}
#[divan::bench()]
fn parse_default_env_file(bencher: divan::Bencher) {
let engine_state = setup_engine();
let default_env = get_default_env().as_bytes();
bencher
.with_inputs(|| nu_protocol::engine::StateWorkingSet::new(&engine_state))
.bench_refs(|mut working_set| parse(&mut working_set, None, default_env, false))
}
}
#[divan::bench_group()]
mod eval_benchmarks {
use super::*;
#[divan::bench()]
fn eval_default_env(bencher: divan::Bencher) {
let default_env = get_default_env().as_bytes();
let fname = "default_env.nu";
bencher
.with_inputs(|| (setup_engine(), nu_protocol::engine::Stack::new()))
.bench_values(|(mut engine_state, mut stack)| {
eval_source(
&mut engine_state,
&mut stack,
get_default_config().as_bytes(),
"default_config.nu",
default_env,
fname,
PipelineData::empty(),
false,
)
})
});
}
#[divan::bench()]
fn eval_default_config(bencher: divan::Bencher) {
let default_env = get_default_config().as_bytes();
let fname = "default_config.nu";
bencher
.with_inputs(|| (setup_engine(), nu_protocol::engine::Stack::new()))
.bench_values(|(mut engine_state, mut stack)| {
eval_source(
&mut engine_state,
&mut stack,
default_env,
fname,
PipelineData::empty(),
false,
)
})
}
}
// generate a new table data with `row_cnt` rows, `col_cnt` columns.
@ -119,54 +134,76 @@ fn encoding_test_data(row_cnt: usize, col_cnt: usize) -> Value {
Value::list(vec![record; row_cnt], Span::test_data())
}
fn encoding_benchmarks(c: &mut Criterion) {
let mut group = c.benchmark_group("Encoding");
let test_cnt_pairs = [(100, 5), (10000, 15)];
for (row_cnt, col_cnt) in test_cnt_pairs.into_iter() {
for fmt in ["json", "msgpack"] {
group.bench_function(&format!("{fmt} encode {row_cnt} * {col_cnt}"), |b| {
let mut res = vec![];
#[divan::bench_group()]
mod encoding_benchmarks {
use super::*;
#[divan::bench(args = [(100, 5), (10000, 15)])]
fn json_encode(bencher: divan::Bencher, (row_cnt, col_cnt): (usize, usize)) {
let test_data = PluginOutput::CallResponse(
0,
PluginCallResponse::value(encoding_test_data(row_cnt, col_cnt)),
);
let encoder = EncodingType::try_from_bytes(fmt.as_bytes()).unwrap();
b.iter(|| encoder.encode(&test_data, &mut res))
});
let encoder = EncodingType::try_from_bytes(b"json").unwrap();
bencher
.with_inputs(|| (vec![]))
.bench_values(|mut res| encoder.encode(&test_data, &mut res))
}
#[divan::bench(args = [(100, 5), (10000, 15)])]
fn msgpack_encode(bencher: divan::Bencher, (row_cnt, col_cnt): (usize, usize)) {
let test_data = PluginOutput::CallResponse(
0,
PluginCallResponse::value(encoding_test_data(row_cnt, col_cnt)),
);
let encoder = EncodingType::try_from_bytes(b"msgpack").unwrap();
bencher
.with_inputs(|| (vec![]))
.bench_values(|mut res| encoder.encode(&test_data, &mut res))
}
group.finish();
}
fn decoding_benchmarks(c: &mut Criterion) {
let mut group = c.benchmark_group("Decoding");
let test_cnt_pairs = [(100, 5), (10000, 15)];
for (row_cnt, col_cnt) in test_cnt_pairs.into_iter() {
for fmt in ["json", "msgpack"] {
group.bench_function(&format!("{fmt} decode for {row_cnt} * {col_cnt}"), |b| {
let mut res = vec![];
#[divan::bench_group()]
mod decoding_benchmarks {
use super::*;
#[divan::bench(args = [(100, 5), (10000, 15)])]
fn json_decode(bencher: divan::Bencher, (row_cnt, col_cnt): (usize, usize)) {
let test_data = PluginOutput::CallResponse(
0,
PluginCallResponse::value(encoding_test_data(row_cnt, col_cnt)),
);
let encoder = EncodingType::try_from_bytes(fmt.as_bytes()).unwrap();
let encoder = EncodingType::try_from_bytes(b"json").unwrap();
let mut res = vec![];
encoder.encode(&test_data, &mut res).unwrap();
let mut binary_data = std::io::Cursor::new(res);
b.iter(|| -> Result<Option<PluginOutput>, _> {
bencher
.with_inputs(|| {
let mut binary_data = std::io::Cursor::new(res.clone());
binary_data.set_position(0);
binary_data
})
.bench_values(|mut binary_data| -> Result<Option<PluginOutput>, _> {
encoder.decode(&mut binary_data)
})
});
}
}
group.finish();
}
criterion_group!(
benches,
parser_benchmarks,
eval_benchmarks,
encoding_benchmarks,
decoding_benchmarks
);
criterion_main!(benches);
#[divan::bench(args = [(100, 5), (10000, 15)])]
fn msgpack_decode(bencher: divan::Bencher, (row_cnt, col_cnt): (usize, usize)) {
let test_data = PluginOutput::CallResponse(
0,
PluginCallResponse::value(encoding_test_data(row_cnt, col_cnt)),
);
let encoder = EncodingType::try_from_bytes(b"msgpack").unwrap();
let mut res = vec![];
encoder.encode(&test_data, &mut res).unwrap();
bencher
.with_inputs(|| {
let mut binary_data = std::io::Cursor::new(res.clone());
binary_data.set_position(0);
binary_data
})
.bench_values(|mut binary_data| -> Result<Option<PluginOutput>, _> {
encoder.decode(&mut binary_data)
})
}
}