Inline world get (#2520)

# Objective

While looking at the code of `World`, I noticed two basic functions (`get` and `get_mut`) that are probably called a lot and with simple code that are not `inline`

## Solution

- Add benchmark to check impact
- Add `#[inline]`


```
group                                            this pr                                main
-----                                            ----                                   ----
world_entity/50000_entities                      1.00   115.9±11.90µs        ? ?/sec    1.71   198.5±29.54µs        ? ?/sec
world_get/50000_entities_SparseSet               1.00   409.9±46.96µs        ? ?/sec    1.18   483.5±36.41µs        ? ?/sec
world_get/50000_entities_Table                   1.00   391.3±29.83µs        ? ?/sec    1.16   455.6±57.85µs        ? ?/sec
world_query_for_each/50000_entities_SparseSet    1.02   121.3±18.36µs        ? ?/sec    1.00   119.4±13.88µs        ? ?/sec
world_query_for_each/50000_entities_Table        1.03     13.8±0.96µs        ? ?/sec    1.00     13.3±0.54µs        ? ?/sec
world_query_get/50000_entities_SparseSet         1.00   666.9±54.36µs        ? ?/sec    1.03   687.1±57.77µs        ? ?/sec
world_query_get/50000_entities_Table             1.01   584.4±55.12µs        ? ?/sec    1.00   576.3±36.13µs        ? ?/sec
world_query_iter/50000_entities_SparseSet        1.01   169.7±19.50µs        ? ?/sec    1.00   168.6±32.56µs        ? ?/sec
world_query_iter/50000_entities_Table            1.00     26.2±1.38µs        ? ?/sec    1.91     50.0±4.40µs        ? ?/sec
```

I didn't add benchmarks for the mutable path but I don't see how it could hurt to make it inline too...

benches/Cargo.toml

@@ -21,6 +21,11 @@ name = "commands"
 path = "benches/bevy_ecs/commands.rs"
 harness = false
 
+[[bench]]
+name = "world_get"
+path = "benches/bevy_ecs/world_get.rs"
+harness = false
+
 [[bench]]
 name = "iter"
 path = "benches/bevy_tasks/iter.rs"

benches/benches/bevy_ecs/world_get.rs

@@ -0,0 +1,161 @@
+use bevy::ecs::{
+    component::{ComponentDescriptor, StorageType},
+    entity::Entity,
+    world::World,
+};
+use criterion::{black_box, criterion_group, criterion_main, Criterion};
+
+criterion_group!(
+    benches,
+    world_entity,
+    world_get,
+    world_query_get,
+    world_query_iter,
+    world_query_for_each,
+);
+criterion_main!(benches);
+
+struct A(f32);
+
+const RANGE: std::ops::Range<u32> = 5..6;
+
+fn setup(entity_count: u32, storage: StorageType) -> World {
+    let mut world = World::default();
+    world
+        .register_component(ComponentDescriptor::new::<A>(storage))
+        .unwrap();
+    world.spawn_batch((0..entity_count).map(|_| (A(0.0),)));
+    world
+}
+
+fn world_entity(criterion: &mut Criterion) {
+    let mut group = criterion.benchmark_group("world_entity");
+    group.warm_up_time(std::time::Duration::from_millis(500));
+    group.measurement_time(std::time::Duration::from_secs(4));
+
+    for entity_count in RANGE.map(|i| i * 10_000) {
+        group.bench_function(format!("{}_entities", entity_count), |bencher| {
+            let world = setup(entity_count, StorageType::Table);
+
+            bencher.iter(|| {
+                for i in 0..entity_count {
+                    let entity = Entity::new(i);
+                    black_box(world.entity(entity));
+                }
+            });
+        });
+    }
+
+    group.finish();
+}
+
+fn world_get(criterion: &mut Criterion) {
+    let mut group = criterion.benchmark_group("world_get");
+    group.warm_up_time(std::time::Duration::from_millis(500));
+    group.measurement_time(std::time::Duration::from_secs(4));
+
+    for entity_count in RANGE.map(|i| i * 10_000) {
+        for storage in [StorageType::Table, StorageType::SparseSet] {
+            group.bench_function(
+                format!("{}_entities_{:?}", entity_count, storage),
+                |bencher| {
+                    let world = setup(entity_count, storage);
+
+                    bencher.iter(|| {
+                        for i in 0..entity_count {
+                            let entity = Entity::new(i);
+                            assert!(world.get::<A>(entity).is_some());
+                        }
+                    });
+                },
+            );
+        }
+    }
+
+    group.finish();
+}
+
+fn world_query_get(criterion: &mut Criterion) {
+    let mut group = criterion.benchmark_group("world_query_get");
+    group.warm_up_time(std::time::Duration::from_millis(500));
+    group.measurement_time(std::time::Duration::from_secs(4));
+
+    for entity_count in RANGE.map(|i| i * 10_000) {
+        for storage in [StorageType::Table, StorageType::SparseSet] {
+            group.bench_function(
+                format!("{}_entities_{:?}", entity_count, storage),
+                |bencher| {
+                    let mut world = setup(entity_count, storage);
+                    let mut query = world.query::<&A>();
+
+                    bencher.iter(|| {
+                        for i in 0..entity_count {
+                            let entity = Entity::new(i);
+                            assert!(query.get(&world, entity).is_ok());
+                        }
+                    });
+                },
+            );
+        }
+    }
+
+    group.finish();
+}
+
+fn world_query_iter(criterion: &mut Criterion) {
+    let mut group = criterion.benchmark_group("world_query_iter");
+    group.warm_up_time(std::time::Duration::from_millis(500));
+    group.measurement_time(std::time::Duration::from_secs(4));
+
+    for entity_count in RANGE.map(|i| i * 10_000) {
+        for storage in [StorageType::Table, StorageType::SparseSet] {
+            group.bench_function(
+                format!("{}_entities_{:?}", entity_count, storage),
+                |bencher| {
+                    let mut world = setup(entity_count, storage);
+                    let mut query = world.query::<&A>();
+
+                    bencher.iter(|| {
+                        let mut count = 0;
+                        for comp in query.iter(&world) {
+                            black_box(comp);
+                            count += 1;
+                        }
+                        assert_eq!(black_box(count), entity_count);
+                    });
+                },
+            );
+        }
+    }
+
+    group.finish();
+}
+
+fn world_query_for_each(criterion: &mut Criterion) {
+    let mut group = criterion.benchmark_group("world_query_for_each");
+    group.warm_up_time(std::time::Duration::from_millis(500));
+    group.measurement_time(std::time::Duration::from_secs(4));
+
+    for entity_count in RANGE.map(|i| i * 10_000) {
+        for storage in [StorageType::Table, StorageType::SparseSet] {
+            group.bench_function(
+                format!("{}_entities_{:?}", entity_count, storage),
+                |bencher| {
+                    let mut world = setup(entity_count, storage);
+                    let mut query = world.query::<&A>();
+
+                    bencher.iter(|| {
+                        let mut count = 0;
+                        query.for_each(&world, |comp| {
+                            black_box(comp);
+                            count += 1;
+                        });
+                        assert_eq!(black_box(count), entity_count);
+                    });
+                },
+            );
+        }
+    }
+
+    group.finish();
+}

crates/bevy_ecs/src/entity/mod.rs

@@ -344,6 +344,7 @@ impl Entities {
     /// Access the location storage of an entity.
     ///
     /// Must not be called on pending entities.
+    #[inline]
     pub fn get_mut(&mut self, entity: Entity) -> Option<&mut EntityLocation> {
         let meta = &mut self.meta[entity.id as usize];
         if meta.generation == entity.generation {
@@ -354,6 +355,7 @@ impl Entities {
     }
 
     /// Returns `Ok(Location { archetype: 0, index: undefined })` for pending entities.
+    #[inline]
     pub fn get(&self, entity: Entity) -> Option<EntityLocation> {
         if (entity.id as usize) < self.meta.len() {
             let meta = &self.meta[entity.id as usize];