Skip to content

Fix the behavior of NoCpuCulling when toggled at runtime. #23534

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
alice-i-cecile merged 1 commit into from
Mar 28, 2026
Merged

Fix the behavior of NoCpuCulling when toggled at runtime. #23534

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
238 changes: 144 additions & 94 deletions crates/bevy_pbr/src/render/mesh.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -773,6 +773,9 @@ bitflags::bitflags! {
/// The mesh had morph targets last frame and so they should be taken
/// into account for motion vector computation.
const HAS_PREVIOUS_MORPH = 1 << 4;
/// CPU culling has been disabled because the `NoCpuCulling` component
/// is present on the mesh instance.
const NO_CPU_CULLING = 1 << 5;
}
}

Expand Down Expand Up @@ -1091,15 +1094,17 @@ impl RenderMeshInstanceSharedFlat {
tag: Option<&MeshTag>,
not_shadow_caster: bool,
no_automatic_batching: bool,
no_cpu_culling: bool,
aabb: Option<&Aabb>,
) -> Self {
Self::for_cpu_building(
Self::new(
previous_transform,
mesh,
tag,
default(),
not_shadow_caster,
no_automatic_batching,
no_cpu_culling,
aabb,
)
}
Expand All @@ -1113,6 +1118,28 @@ impl RenderMeshInstanceSharedFlat {
not_shadow_caster: bool,
no_automatic_batching: bool,
aabb: Option<&Aabb>,
) -> Self {
Self::new(
previous_transform,
mesh,
tag,
material_bindings_index,
not_shadow_caster,
no_automatic_batching,
false,
aabb,
)
}

fn new(
previous_transform: Option<&PreviousGlobalTransform>,
mesh: &Mesh3d,
tag: Option<&MeshTag>,
material_bindings_index: MaterialBindingId,
not_shadow_caster: bool,
no_automatic_batching: bool,
no_cpu_culling: bool,
aabb: Option<&Aabb>,
) -> Self {
let mut mesh_instance_flags = RenderMeshInstanceFlags::empty();
mesh_instance_flags.set(RenderMeshInstanceFlags::SHADOW_CASTER, !not_shadow_caster);
Expand All @@ -1124,6 +1151,7 @@ impl RenderMeshInstanceSharedFlat {
RenderMeshInstanceFlags::HAS_PREVIOUS_TRANSFORM,
previous_transform.is_some(),
);
mesh_instance_flags.set(RenderMeshInstanceFlags::NO_CPU_CULLING, no_cpu_culling);

RenderMeshInstanceSharedFlat {
asset_id: mesh.id().into(),
Expand Down Expand Up @@ -1521,7 +1549,7 @@ impl RenderMeshInstanceGpuPrepared {
render_mesh_instances: &mut MainEntityHashMap<RenderMeshInstanceGpu>,
current_input_buffer: &mut InstanceInputUniformBuffer<MeshInputUniform>,
previous_input_buffer: &PreviousInstanceInputUniformBuffer<MeshInputUniform>,
) -> Option<u32> {
) -> u32 {
// Did the last frame contain this entity as well?
let current_uniform_index;
match render_mesh_instances.entry(entity) {
Expand Down Expand Up @@ -1574,7 +1602,7 @@ impl RenderMeshInstanceGpuPrepared {
}
}

Some(current_uniform_index)
current_uniform_index
}
}

Expand Down Expand Up @@ -1715,14 +1743,26 @@ impl RenderGpuCulledEntities {
///
/// The `render_layers` argument specifies the set of render layers that the
/// entity belongs to.
pub fn update(&mut self, new_entity: MainEntity, render_layers: RenderLayers) {
pub fn update(
&mut self,
new_entity: MainEntity,
render_layers: RenderLayers,
no_cpu_culling: bool,
) {
match self.entities.entry(new_entity) {
Entry::Occupied(mut occupied_entry) => {
occupied_entry.insert(render_layers);
if no_cpu_culling {
occupied_entry.insert(render_layers);
} else {
occupied_entry.remove();
self.removed.push(new_entity);
}
}
Entry::Vacant(vacant_entry) => {
vacant_entry.insert(render_layers);
self.added.push(new_entity);
if no_cpu_culling {
vacant_entry.insert(render_layers);
self.added.push(new_entity);
}
}
}
}
Expand Down Expand Up @@ -1899,14 +1939,14 @@ pub fn extract_meshes_for_gpu_building(
Changed<Aabb>,
Changed<Mesh3d>,
Changed<MeshTag>,
(
Or<(
Changed<NoFrustumCulling>,
Changed<NotShadowReceiver>,
Changed<TransmittedShadowReceiver>,
Changed<NotShadowCaster>,
Changed<NoAutomaticBatching>,
Changed<NoCpuCulling>,
),
)>,
Changed<VisibilityRange>,
Changed<SkinnedMesh>,
)>,
Expand Down Expand Up @@ -2122,6 +2162,7 @@ fn extract_mesh_for_gpu_building(
tag,
not_shadow_caster,
no_automatic_batching,
no_cpu_culling,
aabb,
);

Expand Down Expand Up @@ -2511,81 +2552,84 @@ pub fn collect_meshes_for_gpu_building(
let removed_tx = removed_tx.clone();
scope.spawn(async move {
let _span = info_span!("prepared_mesh_producer").entered();
changed_cpu_culling
.drain(..)
.chain(changed_gpu_culling.drain(..))
.for_each(
|(entity, mesh_instance_builder, mesh_culling_builder)| {
match mesh_instance_builder.prepare(
entity,
mesh_allocator,
mesh_material_ids,
render_material_bindings,
render_lightmaps,
skin_uniforms,
morph_indices,
frame_count,
) {
Some(mut prepared) => {
if let Some(render_mesh_instance) =
render_mesh_instances.get(&entity)
&& prepared.render_layers
== render_mesh_instance.render_layers
{
// We can take a fast path and
// write directly to shared
// memory, since the only fields
// that changed are POD fields.

prepared.shared.write_to_blob(
&render_mesh_instance.shared,
);
render_mesh_instance
.gpu_specific
.set_world_space_center(prepared.center);

let current_uniform_index =
render_mesh_instance
.gpu_specific
.current_uniform_index();

let previous_mesh_input_uniform =
current_input_buffer
.get_unchecked(current_uniform_index);
let previous_input_index =
previous_input_buffer
.push(previous_mesh_input_uniform);
prepared
.mesh_input_uniform
.previous_input_index =
previous_input_index;

current_input_buffer.set(
current_uniform_index,
prepared.mesh_input_uniform,
);
mesh_culling_data_buffer.set(
current_uniform_index,
mesh_culling_builder,
);
} else {
// This is the slow path. Send
// the mesh instance to the
// collection sink.
let data = (
entity,
prepared,
Some(mesh_culling_builder),
);
prepared_tx.send(data).ok();
}
}
None => {
reextract_tx.send(entity).ok();
}
}
},
);
for (entity, mesh_instance_builder, mesh_culling_builder) in
changed_cpu_culling
.drain(..)
.chain(changed_gpu_culling.drain(..))
{
let Some(mut prepared) = mesh_instance_builder.prepare(
entity,
mesh_allocator,
mesh_material_ids,
render_material_bindings,
render_lightmaps,
skin_uniforms,
morph_indices,
frame_count,
) else {
reextract_tx.send(entity).ok();
continue;
};

let Some(render_mesh_instance) = render_mesh_instances.get(&entity)
else {
// We must take the slow path because we
// haven't seen the mesh instance yet. Send
// the mesh instance to the collection sink.
let _ = prepared_tx.send((
entity,
prepared,
Some(mesh_culling_builder),
));
continue;
};

if prepared.render_layers != render_mesh_instance.render_layers
|| prepared
.shared
.flags
.contains(RenderMeshInstanceFlags::NO_CPU_CULLING)
!= render_mesh_instance
.shared
.flags()
.contains(RenderMeshInstanceFlags::NO_CPU_CULLING)
{
// We must take the slow path because the
// instance either changed render layers or
// CPU/GPU culling mode. Send the mesh
// instance to the collection sink.
let _ = prepared_tx.send((
entity,
prepared,
Some(mesh_culling_builder),
));
continue;
}

// If we got here, we can take a fast path and
// write directly to shared memory, since the
// only fields that changed are POD fields.

prepared.shared.write_to_blob(&render_mesh_instance.shared);
render_mesh_instance
.gpu_specific
.set_world_space_center(prepared.center);

let current_uniform_index =
render_mesh_instance.gpu_specific.current_uniform_index();

let previous_mesh_input_uniform =
current_input_buffer.get_unchecked(current_uniform_index);
let previous_input_index =
previous_input_buffer.push(previous_mesh_input_uniform);
prepared.mesh_input_uniform.previous_input_index =
previous_input_index;

current_input_buffer
.set(current_uniform_index, prepared.mesh_input_uniform);
mesh_culling_data_buffer
.set(current_uniform_index, mesh_culling_builder);
}

for entity in removed.drain(..) {
removed_tx.send(entity).unwrap();
Expand All @@ -2606,25 +2650,31 @@ pub fn collect_meshes_for_gpu_building(

while let Ok(batch) = prepared_rx.recv() {
let (entity, prepared, mesh_culling_builder) = batch;
let Some(instance_data_index) = prepared.update(
let instance_data_index = prepared.update(
entity,
&mut *render_mesh_instances,
current_input_buffer,
previous_input_buffer,
) else {
continue;
};
);
if let Some(mesh_culling_data) = mesh_culling_builder {
mesh_culling_data.update(&mut mesh_culling_data_buffer, instance_data_index);
}
// If the instance is already visible, just update the layers.
// Otherwise, mark it as newly-added.
let render_layers = render_mesh_instances
.get(&entity)
.and_then(|render_mesh_instance| render_mesh_instance.render_layers.as_ref())
.cloned()
.unwrap_or_else(RenderLayers::default);
render_gpu_culled_entities.update(entity, render_layers);
let (render_layers, no_cpu_culling) = match render_mesh_instances.get(&entity) {
None => (RenderLayers::default(), false),
Some(render_mesh_instance) => (
render_mesh_instance
.render_layers
.clone()
.unwrap_or_default(),
render_mesh_instance
.shared
.flags()
.contains(RenderMeshInstanceFlags::NO_CPU_CULLING),
),
};
render_gpu_culled_entities.update(entity, render_layers, no_cpu_culling);
}
while let Ok(batch) = removed_rx.recv() {
let entity = batch;
Expand Down
23 changes: 23 additions & 0 deletions examples/large_scenes/bevy_city/src/settings.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,5 @@
use bevy::{
camera::visibility::NoCpuCulling,
camera_controller::free_camera::FreeCameraState,
feathers::{
self,
Expand All @@ -21,6 +22,7 @@ pub struct Settings {
pub shadow_maps_enabled: bool,
pub contact_shadows_enabled: bool,
pub wireframe_enabled: bool,
pub cpu_culling: bool,
}

impl Default for Settings {
Expand All @@ -30,6 +32,7 @@ impl Default for Settings {
shadow_maps_enabled: true,
contact_shadows_enabled: true,
wireframe_enabled: false,
cpu_culling: true,
}
}
}
Expand Down Expand Up @@ -122,6 +125,26 @@ pub fn setup_settings_ui(mut commands: Commands) {
}
)
),
(
checkbox(Checked, Spawn((Text::new("CPU culling"), ThemedText))),
observe(checkbox_self_update),
observe(
|change: On<ValueChange<bool>>,
mut settings: ResMut<Settings>,
mut commands: Commands,
meshes: Query<Entity, With<Mesh3d>>| {
settings.cpu_culling = change.value;

for entity in meshes.iter() {
if settings.cpu_culling {
commands.entity(entity).remove::<NoCpuCulling>();
} else {
commands.entity(entity).insert(NoCpuCulling);
}
}
}
)
),
(
button(
ButtonProps::default(),
Expand Down