Migrate to core::hint::black_box() (#16980)

# Objective

Many of our benchmarks use
[`criterion::black_box()`](https://docs.rs/criterion/latest/criterion/fn.black_box.html),
which is used to prevent the compiler from optimizing away computation
that we're trying to time. This can be slow, though, because
`criterion::black_box()` forces a point read each time it is called
through
[`ptr::road_volatile()`](https://doc.rust-lang.org/stable/std/ptr/fn.read_volatile.html).

In Rust 1.66, the standard library introduced
[`core::hint::black_box()`](https://doc.rust-lang.org/nightly/std/hint/fn.black_box.html)
(and `std::hint::black_box()`). This is an intended replacement for
`criterion`'s version that uses compiler intrinsics instead of volatile
pointer reads, and thus has no runtime overhead. This increases
benchmark accuracy, which is always nice 👍

Note that benchmarks may _appear_ to improve in performance after this
change, but that's just because we are eliminating the pointer read
overhead.

## Solution

- Deny `criterion::black_box` in `clippy.toml`.
- Fix all imports.

## Testing

- `cargo clippy -p benches --benches`

benches/benches/bevy_ecs/change_detection.rs

@@ -1,3 +1,5 @@
+use core::hint::black_box;
+
 use bevy_ecs::{
     component::{Component, Mutable},
     entity::Entity,
@@ -5,7 +7,7 @@ use bevy_ecs::{
     query::QueryFilter,
     world::World,
 };
-use criterion::{black_box, criterion_group, Criterion};
+use criterion::{criterion_group, Criterion};
 use rand::{prelude::SliceRandom, SeedableRng};
 use rand_chacha::ChaCha8Rng;
 

benches/benches/bevy_ecs/empty_archetypes.rs

@@ -1,5 +1,7 @@
+use core::hint::black_box;
+
 use bevy_ecs::{component::Component, prelude::*, schedule::ExecutorKind, world::World};
-use criterion::{black_box, criterion_group, BenchmarkId, Criterion};
+use criterion::{criterion_group, BenchmarkId, Criterion};
 
 criterion_group!(benches, empty_archetypes);
 

benches/benches/bevy_ecs/entity_cloning.rs

@@ -1,10 +1,12 @@
+use core::hint::black_box;
+
 use bevy_ecs::bundle::Bundle;
 use bevy_ecs::reflect::AppTypeRegistry;
 use bevy_ecs::{component::Component, reflect::ReflectComponent, world::World};
 use bevy_hierarchy::{BuildChildren, CloneEntityHierarchyExt};
 use bevy_math::Mat4;
 use bevy_reflect::{GetTypeRegistration, Reflect};
-use criterion::{black_box, criterion_group, criterion_main, Bencher, Criterion};
+use criterion::{criterion_group, criterion_main, Bencher, Criterion};
 
 criterion_group!(benches, reflect_benches, clone_benches);
 criterion_main!(benches);

benches/benches/bevy_ecs/observers/propagation.rs

@@ -1,9 +1,11 @@
+use core::hint::black_box;
+
 use bevy_ecs::{
     component::Component, entity::Entity, event::Event, observer::Trigger, world::World,
 };
 use bevy_hierarchy::{BuildChildren, Parent};
 
-use criterion::{black_box, Criterion};
+use criterion::Criterion;
 use rand::SeedableRng;
 use rand::{seq::IteratorRandom, Rng};
 use rand_chacha::ChaCha8Rng;

benches/benches/bevy_ecs/observers/simple.rs

@@ -1,6 +1,8 @@
+use core::hint::black_box;
+
 use bevy_ecs::{entity::Entity, event::Event, observer::Trigger, world::World};
 
-use criterion::{black_box, Criterion};
+use criterion::Criterion;
 use rand::{prelude::SliceRandom, SeedableRng};
 use rand_chacha::ChaCha8Rng;
 fn deterministic_rand() -> ChaCha8Rng {

benches/benches/bevy_ecs/world/commands.rs

@@ -1,9 +1,11 @@
+use core::hint::black_box;
+
 use bevy_ecs::{
     component::Component,
     system::Commands,
     world::{Command, CommandQueue, World},
 };
-use criterion::{black_box, Criterion};
+use criterion::Criterion;
 
 #[derive(Component)]
 struct A;

benches/benches/bevy_ecs/world/world_get.rs

@@ -1,3 +1,5 @@
+use core::hint::black_box;
+
 use bevy_ecs::{
     bundle::Bundle,
     component::Component,
@@ -5,7 +7,7 @@ use bevy_ecs::{
     system::{Query, SystemState},
     world::World,
 };
-use criterion::{black_box, Criterion};
+use criterion::Criterion;
 use rand::{prelude::SliceRandom, SeedableRng};
 use rand_chacha::ChaCha8Rng;
 

benches/benches/bevy_math/bezier.rs

@@ -1,4 +1,6 @@
-use criterion::{black_box, criterion_group, Criterion};
+use core::hint::black_box;
+
+use criterion::{criterion_group, Criterion};
 
 use bevy_math::prelude::*;
 

benches/benches/bevy_picking/ray_mesh_intersection.rs

@@ -1,6 +1,8 @@
+use core::hint::black_box;
+
 use bevy_math::{Dir3, Mat4, Ray3d, Vec3};
 use bevy_picking::mesh_picking::ray_cast;
-use criterion::{black_box, criterion_group, Criterion};
+use criterion::{criterion_group, Criterion};
 
 fn ptoxznorm(p: u32, size: u32) -> (f32, f32) {
     let ij = (p / (size), p % (size));

benches/benches/bevy_reflect/list.rs

@@ -1,10 +1,10 @@
-use core::{iter, time::Duration};
+use core::{hint::black_box, iter, time::Duration};
 
 use benches::bench;
 use bevy_reflect::{DynamicList, List};
 use criterion::{
-    black_box, criterion_group, measurement::Measurement, AxisScale, BatchSize, BenchmarkGroup,
-    BenchmarkId, Criterion, PlotConfiguration, Throughput,
+    criterion_group, measurement::Measurement, AxisScale, BatchSize, BenchmarkGroup, BenchmarkId,
+    Criterion, PlotConfiguration, Throughput,
 };
 
 criterion_group!(

benches/benches/bevy_reflect/map.rs

@@ -1,11 +1,11 @@
-use core::{fmt::Write, iter, time::Duration};
+use core::{fmt::Write, hint::black_box, iter, time::Duration};
 
 use benches::bench;
 use bevy_reflect::{DynamicMap, Map};
 use bevy_utils::HashMap;
 use criterion::{
-    black_box, criterion_group, measurement::Measurement, AxisScale, BatchSize, BenchmarkGroup,
-    BenchmarkId, Criterion, PlotConfiguration, Throughput,
+    criterion_group, measurement::Measurement, AxisScale, BatchSize, BenchmarkGroup, BenchmarkId,
+    Criterion, PlotConfiguration, Throughput,
 };
 
 criterion_group!(

benches/benches/bevy_reflect/path.rs

@@ -1,8 +1,8 @@
-use core::{fmt::Write, str, time::Duration};
+use core::{fmt::Write, hint::black_box, str, time::Duration};
 
 use benches::bench;
 use bevy_reflect::ParsedPath;
-use criterion::{black_box, criterion_group, BatchSize, BenchmarkId, Criterion, Throughput};
+use criterion::{criterion_group, BatchSize, BenchmarkId, Criterion, Throughput};
 use rand::{distributions::Uniform, Rng, SeedableRng};
 use rand_chacha::ChaCha8Rng;
 

benches/benches/bevy_reflect/struct.rs

@@ -1,10 +1,10 @@
-use core::time::Duration;
+use core::{hint::black_box, time::Duration};
 
 use benches::bench;
 use bevy_reflect::{DynamicStruct, GetField, PartialReflect, Reflect, Struct};
 use criterion::{
-    black_box, criterion_group, measurement::Measurement, AxisScale, BatchSize, BenchmarkGroup,
-    BenchmarkId, Criterion, PlotConfiguration, Throughput,
+    criterion_group, measurement::Measurement, AxisScale, BatchSize, BenchmarkGroup, BenchmarkId,
+    Criterion, PlotConfiguration, Throughput,
 };
 
 criterion_group!(

benches/benches/bevy_render/render_layers.rs

@@ -1,4 +1,6 @@
-use criterion::{black_box, criterion_group, Criterion};
+use core::hint::black_box;
+
+use criterion::{criterion_group, Criterion};
 
 use bevy_render::view::RenderLayers;
 

benches/benches/bevy_render/torus.rs

@@ -1,4 +1,6 @@
-use criterion::{black_box, criterion_group, Criterion};
+use core::hint::black_box;
+
+use criterion::{criterion_group, Criterion};
 
 use bevy_render::mesh::TorusMeshBuilder;
 

benches/benches/bevy_tasks/iter.rs

@@ -1,5 +1,7 @@
+use core::hint::black_box;
+
 use bevy_tasks::{ParallelIterator, TaskPoolBuilder};
-use criterion::{black_box, criterion_group, BenchmarkId, Criterion};
+use criterion::{criterion_group, BenchmarkId, Criterion};
 
 struct ParChunks<'a, T>(core::slice::Chunks<'a, T>);
 impl<'a, T> ParallelIterator<core::slice::Iter<'a, T>> for ParChunks<'a, T>

clippy.toml

@@ -41,4 +41,5 @@ disallowed-methods = [
   { path = "f32::asinh", reason = "use bevy_math::ops::asinh instead for libm determinism" },
   { path = "f32::acosh", reason = "use bevy_math::ops::acosh instead for libm determinism" },
   { path = "f32::atanh", reason = "use bevy_math::ops::atanh instead for libm determinism" },
+  { path = "criterion::black_box", reason = "use core::hint::black_box instead" },
 ]