Skip to content

Commit

Permalink
Improve docs.
Browse files Browse the repository at this point in the history 
* More cross-linking.
* Remove `extern crate` stuff from doc tests / snippets.
  • Loading branch information
@waywardmonkeys
waywardmonkeys committed Jul 27, 2024
1 parent 6d28458 commit 19eddb8
Show file tree
Hide file tree
Showing 5 changed files with 57 additions and 70 deletions.
74 changes: 32 additions & 42 deletions src/bencher.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -21,16 +21,22 @@ use crate::async_executor::AsyncExecutor;
/// way to time a routine and each has advantages and disadvantages.
///
/// * If you want to do the iteration and measurement yourself (eg. passing the iteration count
/// to a separate process), use `iter_custom`.
/// to a separate process), use [`iter_custom`].
/// * If your routine requires no per-iteration setup and returns a value with an expensive `drop`
/// method, use `iter_with_large_drop`.
/// * If your routine requires some per-iteration setup that shouldn't be timed, use `iter_batched`
/// or `iter_batched_ref`. See [`BatchSize`] for a discussion of batch sizes.
/// method, use [`iter_with_large_drop`].
/// * If your routine requires some per-iteration setup that shouldn't be timed, use [`iter_batched`]
/// or [`iter_batched_ref`]. See [`BatchSize`] for a discussion of batch sizes.
/// If the setup value implements `Drop` and you don't want to include the `drop` time in the
/// measurement, use `iter_batched_ref`, otherwise use `iter_batched`. These methods are also
/// measurement, use [`iter_batched_ref`], otherwise use [`iter_batched`]. These methods are also
/// suitable for benchmarking routines which return a value with an expensive `drop` method,
/// but are more complex than `iter_with_large_drop`.
/// * Otherwise, use `iter`.
/// but are more complex than [`iter_with_large_drop`].
/// * Otherwise, use [`iter`].
///
/// [`iter`]: Bencher::iter
/// [`iter_custom`]: Bencher::iter_custom
/// [`iter_with_large_drop`]: Bencher::iter_with_large_drop
/// [`iter_batched`]: Bencher::iter_batched
/// [`iter_batched_ref`]: Bencher::iter_batched_ref
pub struct Bencher<'a, M: Measurement = WallTime> {
pub(crate) iterated: bool, // Have we iterated this benchmark?
pub(crate) iters: u64, // Number of times to iterate this benchmark
Expand All @@ -56,9 +62,7 @@ impl<'a, M: Measurement> Bencher<'a, M> {
/// # Example
///
/// ```rust
/// #[macro_use] extern crate criterion;
///
/// use criterion::*;
/// use criterion::{criterion_group, criterion_main, Criterion};
///
/// // The function to benchmark
/// fn foo() {
Expand Down Expand Up @@ -97,12 +101,11 @@ impl<'a, M: Measurement> Bencher<'a, M> {
/// and coordinate with multiple threads).
///
/// # Timing model
/// Custom, the timing model is whatever is returned as the Duration from `routine`.
/// Custom, the timing model is whatever is returned as the [`Duration`] from `routine`.
///
/// # Example
/// ```rust
/// #[macro_use] extern crate criterion;
/// use criterion::*;
/// use criterion::{criterion_group, criterion_main, Criterion};
/// use std::time::Instant;
///
/// fn foo() {
Expand Down Expand Up @@ -148,8 +151,9 @@ impl<'a, M: Measurement> Bencher<'a, M> {
/// Times a `routine` by collecting its output on each iteration. This avoids timing the
/// destructor of the value returned by `routine`.
///
/// WARNING: This requires `O(iters * mem::size_of::<O>())` of memory, and `iters` is not under the
/// control of the caller. If this causes out-of-memory errors, use `iter_batched` instead.
/// WARNING: This requires `O(iters * mem::size_of::<O>())` of memory, and `iters` is not
/// under the control of the caller. If this causes out-of-memory errors, use
/// [`iter_batched`](Self::iter_batched) instead.
///
/// # Timing model
///
Expand All @@ -160,9 +164,7 @@ impl<'a, M: Measurement> Bencher<'a, M> {
/// # Example
///
/// ```rust
/// #[macro_use] extern crate criterion;
///
/// use criterion::*;
/// use criterion::{criterion_group, criterion_main, Criterion};
///
/// fn create_vector() -> Vec<u64> {
/// # vec![]
Expand Down Expand Up @@ -203,9 +205,7 @@ impl<'a, M: Measurement> Bencher<'a, M> {
/// # Example
///
/// ```rust
/// #[macro_use] extern crate criterion;
///
/// use criterion::*;
/// use criterion::{criterion_group, criterion_main, BatchSize, Criterion};
///
/// fn create_scrambled_data() -> Vec<u64> {
/// # vec![]
Expand Down Expand Up @@ -293,9 +293,7 @@ impl<'a, M: Measurement> Bencher<'a, M> {
/// # Example
///
/// ```rust
/// #[macro_use] extern crate criterion;
///
/// use criterion::*;
/// use criterion::{criterion_group, criterion_main, BatchSize, Criterion};
///
/// fn create_scrambled_data() -> Vec<u64> {
/// # vec![]
Expand Down Expand Up @@ -408,9 +406,7 @@ impl<'a, 'b, A: AsyncExecutor, M: Measurement> AsyncBencher<'a, 'b, A, M> {
/// # Example
///
/// ```rust
/// #[macro_use] extern crate criterion;
///
/// use criterion::*;
/// use criterion::{criterion_group, criterion_main, Criterion};
/// use criterion::async_executor::FuturesExecutor;
///
/// // The function to benchmark
Expand Down Expand Up @@ -454,12 +450,11 @@ impl<'a, 'b, A: AsyncExecutor, M: Measurement> AsyncBencher<'a, 'b, A, M> {
/// and coordinate with multiple threads).
///
/// # Timing model
/// Custom, the timing model is whatever is returned as the Duration from `routine`.
/// Custom, the timing model is whatever is returned as the [`Duration`] from `routine`.
///
/// # Example
/// ```rust
/// #[macro_use] extern crate criterion;
/// use criterion::*;
/// use criterion::{criterion_group, criterion_main, Criterion};
/// use criterion::async_executor::FuturesExecutor;
/// use std::time::Instant;
///
Expand Down Expand Up @@ -513,8 +508,9 @@ impl<'a, 'b, A: AsyncExecutor, M: Measurement> AsyncBencher<'a, 'b, A, M> {
/// Times a `routine` by collecting its output on each iteration. This avoids timing the
/// destructor of the value returned by `routine`.
///
/// WARNING: This requires `O(iters * mem::size_of::<O>())` of memory, and `iters` is not under the
/// control of the caller. If this causes out-of-memory errors, use `iter_batched` instead.
/// WARNING: This requires `O(iters * mem::size_of::<O>())` of memory, and `iters`
/// is not under the control of the caller. If this causes out-of-memory errors, use
/// [`iter_batched`](Self::iter_batched) instead.
///
/// # Timing model
///
Expand All @@ -525,9 +521,7 @@ impl<'a, 'b, A: AsyncExecutor, M: Measurement> AsyncBencher<'a, 'b, A, M> {
/// # Example
///
/// ```rust
/// #[macro_use] extern crate criterion;
///
/// use criterion::*;
/// use criterion::{criterion_group, criterion_main, Criterion};
/// use criterion::async_executor::FuturesExecutor;
///
/// async fn create_vector() -> Vec<u64> {
Expand Down Expand Up @@ -580,9 +574,7 @@ impl<'a, 'b, A: AsyncExecutor, M: Measurement> AsyncBencher<'a, 'b, A, M> {
/// # Example
///
/// ```rust
/// #[macro_use] extern crate criterion;
///
/// use criterion::*;
/// use criterion::{criterion_group, criterion_main, BatchSize, Criterion};
/// use criterion::async_executor::FuturesExecutor;
///
/// fn create_scrambled_data() -> Vec<u64> {
Expand Down Expand Up @@ -678,9 +670,7 @@ impl<'a, 'b, A: AsyncExecutor, M: Measurement> AsyncBencher<'a, 'b, A, M> {
/// # Example
///
/// ```rust
/// #[macro_use] extern crate criterion;
///
/// use criterion::*;
/// use criterion::{criterion_group, criterion_main, BatchSize, Criterion};
/// use criterion::async_executor::FuturesExecutor;
///
/// fn create_scrambled_data() -> Vec<u64> {
Expand All @@ -698,7 +688,7 @@ impl<'a, 'b, A: AsyncExecutor, M: Measurement> AsyncBencher<'a, 'b, A, M> {
///
/// c.bench_function("with_setup", move |b| {
/// // This will avoid timing the clone call.
/// b.iter_batched(|| data.clone(), |mut data| async move { sort(&mut data).await }, BatchSize::SmallInput)
/// b.iter_batched_ref(|| data.clone(), |mut data| async move { sort(&mut data).await }, BatchSize::SmallInput)
/// });
/// }
///
Expand Down
3 changes: 1 addition & 2 deletions src/benchmark_group.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -16,8 +16,7 @@ use std::time::Duration;
/// # Examples:
///
/// ```no_run
/// #[macro_use] extern crate criterion;
/// use self::criterion::*;
/// use criterion::{criterion_group, criterion_main, BenchmarkId, Criterion, Throughput};
/// use std::time::Duration;
///
/// fn bench_simple(c: &mut Criterion) {
Expand Down
32 changes: 18 additions & 14 deletions src/lib.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -1154,8 +1154,7 @@ https://bheisler.github.io/criterion.rs/book/faq.html
/// # Examples:
///
/// ```rust
/// #[macro_use] extern crate criterion;
/// use self::criterion::*;
/// use criterion::{criterion_group, criterion_main, Criterion};
///
/// fn bench_simple(c: &mut Criterion) {
/// let mut group = c.benchmark_group("My Group");
Expand Down Expand Up @@ -1187,13 +1186,13 @@ impl<M> Criterion<M>
where
M: Measurement + 'static,
{
/// Benchmarks a function. For comparing multiple functions, see `benchmark_group`.
/// Benchmarks a function. For comparing multiple functions, see
/// [`benchmark_group`](Self::benchmark_group).
///
/// # Example
///
/// ```rust
/// #[macro_use] extern crate criterion;
/// use self::criterion::*;
/// use criterion::{criterion_group, criterion_main, Criterion};
///
/// fn bench(c: &mut Criterion) {
/// // Setup (construct data, allocate memory, etc)
Expand All @@ -1218,13 +1217,12 @@ where
}

/// Benchmarks a function with an input. For comparing multiple functions or multiple inputs,
/// see `benchmark_group`.
/// see [`benchmark_group`](Self::benchmark_group).
///
/// # Example
///
/// ```rust
/// #[macro_use] extern crate criterion;
/// use self::criterion::*;
/// use criterion::{criterion_group, criterion_main, BenchmarkId, Criterion};
///
/// fn bench(c: &mut Criterion) {
/// // Setup (construct data, allocate memory, etc)
Expand Down Expand Up @@ -1285,7 +1283,7 @@ pub enum Throughput {
Elements(u64),
}

/// Axis scaling type
/// Axis scaling type. Specified via [`PlotConfiguration::summary_scale`].
#[derive(Debug, Default, Clone, Copy)]
pub enum AxisScale {
/// Axes scale linearly
Expand Down Expand Up @@ -1318,17 +1316,21 @@ pub struct PlotConfiguration {

impl PlotConfiguration {
#[must_use]
/// Set the axis scale (linear or logarithmic) for the summary plots. Typically, you would
/// set this to logarithmic if benchmarking over a range of inputs which scale exponentially.
/// Defaults to linear.
/// Set the axis scale ([linear] or [logarithmic]) for the summary plots.
///
/// Typically, you would set this to logarithmic if benchmarking over a
/// range of inputs which scale exponentially. Defaults to [`AxisScale::Linear`].
///
/// [linear]: AxisScale::Linear
/// [logarithmic]: AxisScale::Logarithmic
pub fn summary_scale(mut self, new_scale: AxisScale) -> PlotConfiguration {
self.summary_scale = new_scale;
self
}
}

/// This enum allows the user to control how Criterion.rs chooses the iteration count when sampling.
/// The default is Auto, which will choose a method automatically based on the iteration time during
/// The default is `Auto`, which will choose a method automatically based on the iteration time during
/// the warm-up phase.
#[derive(Debug, Default, Clone, Copy)]
pub enum SamplingMode {
Expand All @@ -1343,10 +1345,11 @@ pub enum SamplingMode {

/// Keep the iteration count the same for all samples. This is not recommended, as it affects
/// the statistics that Criterion.rs can compute. However, it requires fewer iterations than
/// the Linear method and therefore is more suitable for very long-running benchmarks where
/// the `Linear` method and therefore is more suitable for very long-running benchmarks where
/// benchmark execution time is more of a problem and statistical precision is less important.
Flat,
}

impl SamplingMode {
pub(crate) fn choose_sampling_mode(
&self,
Expand Down Expand Up @@ -1380,6 +1383,7 @@ pub(crate) enum ActualSamplingMode {
Linear,
Flat,
}

impl ActualSamplingMode {
pub(crate) fn iteration_counts(
&self,
Expand Down
14 changes: 4 additions & 10 deletions src/macros.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -18,9 +18,7 @@
/// Complete form:
///
/// ```
/// # #[macro_use]
/// # extern crate criterion;
/// # use criterion::Criterion;
/// # use criterion::{criterion_group, Criterion};
/// # fn bench_method1(c: &mut Criterion) {
/// # }
/// #
Expand All @@ -44,9 +42,7 @@
/// Compact Form:
///
/// ```
/// # #[macro_use]
/// # extern crate criterion;
/// # use criterion::Criterion;
/// # use criterion::{criterion_group, Criterion};
/// # fn bench_method1(c: &mut Criterion) {
/// # }
/// #
Expand Down Expand Up @@ -100,10 +96,8 @@ macro_rules! criterion_group {
///
/// Since we've disabled the default benchmark harness, we need to add our own:
///
/// ```ignore
/// #[macro_use]
/// extern crate criterion;
/// use criterion::Criterion;
/// ```no_run
/// use criterion::{criterion_group, criterion_main, Criterion};
/// fn bench_method1(c: &mut Criterion) {
/// }
///
Expand Down
4 changes: 2 additions & 2 deletions src/measurement.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -9,8 +9,8 @@ use std::time::{Duration, Instant};
/// Trait providing functions to format measured values to string so that they can be displayed on
/// the command line or in the reports. The functions of this trait take measured values in f64
/// form; implementors can assume that the values are of the same scale as those produced by the
/// associated [MeasuredValue](trait.MeasuredValue.html) (eg. if your measurement produces values in
/// nanoseconds, the values passed to the formatter will be in nanoseconds).
/// associated [`Measuremeant`] (eg. if your measurement produces values in nanoseconds, the
/// values passed to the formatter will be in nanoseconds).
///
/// Implementors are encouraged to format the values in a way that is intuitive for humans and
/// uses the SI prefix system. For example, the format used by [`WallTime`] can display the value
Expand Down

0 comments on commit 19eddb8

Please sign in to comment.