Struct wgpu::RenderPass

source ·
pub struct RenderPass<'encoder> { /* private fields */ }
Expand description

In-progress recording of a render pass: a list of render commands in a CommandEncoder.

It can be created with CommandEncoder::begin_render_pass(), whose RenderPassDescriptor specifies the attachments (textures) that will be rendered to.

Most of the methods on RenderPass serve one of two purposes, identifiable by their names:

  • draw_*(): Drawing (that is, encoding a render command, which, when executed by the GPU, will rasterize something and execute shaders).
  • set_*(): Setting part of the render state for future drawing commands.

A render pass may contain any number of drawing commands, and before/between each command the render state may be updated however you wish; each drawing command will be executed using the render state that has been set when the draw_*() function is called.

Corresponds to WebGPU GPURenderPassEncoder.

Implementations§

source§

impl<'encoder> RenderPass<'encoder>

source

pub fn forget_lifetime(self) -> RenderPass<'static>

Drops the lifetime relationship to the parent command encoder, making usage of the encoder while this pass is recorded a run-time error instead.

Attention: As long as the render pass has not been ended, any mutating operation on the parent command encoder will cause a run-time error and invalidate it! By default, the lifetime constraint prevents this, but it can be useful to handle this at run time, such as when storing the pass and encoder in the same data structure.

This operation has no effect on pass recording. It’s a safe operation, since CommandEncoder is in a locked state as long as the pass is active regardless of the lifetime constraint or its absence.

source

pub fn set_bind_group<'a>( &mut self, index: u32, bind_group: impl Into<Option<&'a BindGroup>>, offsets: &[DynamicOffset] )

Sets the active bind group for a given bind group index. The bind group layout in the active pipeline when any draw_*() method is called must match the layout of this bind group.

If the bind group have dynamic offsets, provide them in binding order. These offsets have to be aligned to Limits::min_uniform_buffer_offset_alignment or Limits::min_storage_buffer_offset_alignment appropriately.

Subsequent draw calls’ shader executions will be able to access data in these bind groups.

source

pub fn set_pipeline(&mut self, pipeline: &RenderPipeline)

Sets the active render pipeline.

Subsequent draw calls will exhibit the behavior defined by pipeline.

source

pub fn set_blend_constant(&mut self, color: Color)

Sets the blend color as used by some of the blending modes.

Subsequent blending tests will test against this value. If this method has not been called, the blend constant defaults to Color::TRANSPARENT (all components zero).

source

pub fn set_index_buffer( &mut self, buffer_slice: BufferSlice<'_>, index_format: IndexFormat )

Sets the active index buffer.

Subsequent calls to draw_indexed on this RenderPass will use buffer as the source index buffer.

source

pub fn set_vertex_buffer(&mut self, slot: u32, buffer_slice: BufferSlice<'_>)

Assign a vertex buffer to a slot.

Subsequent calls to draw and draw_indexed on this RenderPass will use buffer as one of the source vertex buffers.

The slot refers to the index of the matching descriptor in VertexState::buffers.

source

pub fn set_scissor_rect(&mut self, x: u32, y: u32, width: u32, height: u32)

Sets the scissor rectangle used during the rasterization stage. After transformation into viewport coordinates.

Subsequent draw calls will discard any fragments which fall outside the scissor rectangle. If this method has not been called, the scissor rectangle defaults to the entire bounds of the render targets.

The function of the scissor rectangle resembles set_viewport(), but it does not affect the coordinate system, only which fragments are discarded.

source

pub fn set_viewport( &mut self, x: f32, y: f32, w: f32, h: f32, min_depth: f32, max_depth: f32 )

Sets the viewport used during the rasterization stage to linearly map from normalized device coordinates to viewport coordinates.

Subsequent draw calls will only draw within this region. If this method has not been called, the viewport defaults to the entire bounds of the render targets.

source

pub fn set_stencil_reference(&mut self, reference: u32)

Sets the stencil reference.

Subsequent stencil tests will test against this value. If this method has not been called, the stencil reference value defaults to 0.

source

pub fn insert_debug_marker(&mut self, label: &str)

Inserts debug marker.

source

pub fn push_debug_group(&mut self, label: &str)

Start record commands and group it into debug marker group.

source

pub fn pop_debug_group(&mut self)

Stops command recording and creates debug group.

source

pub fn draw(&mut self, vertices: Range<u32>, instances: Range<u32>)

Draws primitives from the active vertex buffer(s).

The active vertex buffer(s) can be set with RenderPass::set_vertex_buffer. Does not use an Index Buffer. If you need this see RenderPass::draw_indexed

Panics if vertices Range is outside of the range of the vertices range of any set vertex buffer.

vertices: The range of vertices to draw. instances: Range of Instances to draw. Use 0..1 if instance buffers are not used. E.g.of how its used internally

for instance_id in instance_range {
    for vertex_id in vertex_range {
        let vertex = vertex[vertex_id];
        vertex_shader(vertex, vertex_id, instance_id);
    }
}

This drawing command uses the current render state, as set by preceding set_*() methods. It is not affected by changes to the state that are performed after it is called.

source

pub fn draw_indexed( &mut self, indices: Range<u32>, base_vertex: i32, instances: Range<u32> )

Draws indexed primitives using the active index buffer and the active vertex buffers.

The active index buffer can be set with RenderPass::set_index_buffer The active vertex buffers can be set with RenderPass::set_vertex_buffer.

Panics if indices Range is outside of the range of the indices range of any set index buffer.

indices: The range of indices to draw. base_vertex: value added to each index value before indexing into the vertex buffers. instances: Range of Instances to draw. Use 0..1 if instance buffers are not used. E.g.of how its used internally

for instance_id in instance_range {
    for index_index in index_range {
        let vertex_id = index_buffer[index_index];
        let adjusted_vertex_id = vertex_id + base_vertex;
        let vertex = vertex[adjusted_vertex_id];
        vertex_shader(vertex, adjusted_vertex_id, instance_id);
    }
}

This drawing command uses the current render state, as set by preceding set_*() methods. It is not affected by changes to the state that are performed after it is called.

source

pub fn draw_indirect( &mut self, indirect_buffer: &Buffer, indirect_offset: BufferAddress )

Draws primitives from the active vertex buffer(s) based on the contents of the indirect_buffer.

This is like calling RenderPass::draw but the contents of the call are specified in the indirect_buffer. The structure expected in indirect_buffer must conform to DrawIndirectArgs.

Indirect drawing has some caveats depending on the features available. We are not currently able to validate these and issue an error.

See details on the individual flags for more information.

source

pub fn draw_indexed_indirect( &mut self, indirect_buffer: &Buffer, indirect_offset: BufferAddress )

Draws indexed primitives using the active index buffer and the active vertex buffers, based on the contents of the indirect_buffer.

This is like calling RenderPass::draw_indexed but the contents of the call are specified in the indirect_buffer. The structure expected in indirect_buffer must conform to DrawIndexedIndirectArgs.

Indirect drawing has some caveats depending on the features available. We are not currently able to validate these and issue an error.

See details on the individual flags for more information.

source

pub fn execute_bundles<'a, I: IntoIterator<Item = &'a RenderBundle>>( &mut self, render_bundles: I )

Execute a render bundle, which is a set of pre-recorded commands that can be run together.

Commands in the bundle do not inherit this render pass’s current render state, and after the bundle has executed, the state is cleared (reset to defaults, not the previous state).

source§

impl<'encoder> RenderPass<'encoder>

Features::MULTI_DRAW_INDIRECT must be enabled on the device in order to call these functions.

source

pub fn multi_draw_indirect( &mut self, indirect_buffer: &Buffer, indirect_offset: BufferAddress, count: u32 )

Dispatches multiple draw calls from the active vertex buffer(s) based on the contents of the indirect_buffer. count draw calls are issued.

The active vertex buffers can be set with RenderPass::set_vertex_buffer.

The structure expected in indirect_buffer must conform to DrawIndirectArgs. These draw structures are expected to be tightly packed.

This drawing command uses the current render state, as set by preceding set_*() methods. It is not affected by changes to the state that are performed after it is called.

source

pub fn multi_draw_indexed_indirect( &mut self, indirect_buffer: &Buffer, indirect_offset: BufferAddress, count: u32 )

Dispatches multiple draw calls from the active index buffer and the active vertex buffers, based on the contents of the indirect_buffer. count draw calls are issued.

The active index buffer can be set with RenderPass::set_index_buffer, while the active vertex buffers can be set with RenderPass::set_vertex_buffer.

The structure expected in indirect_buffer must conform to DrawIndexedIndirectArgs. These draw structures are expected to be tightly packed.

This drawing command uses the current render state, as set by preceding set_*() methods. It is not affected by changes to the state that are performed after it is called.

source§

impl<'encoder> RenderPass<'encoder>

Features::MULTI_DRAW_INDIRECT_COUNT must be enabled on the device in order to call these functions.

source

pub fn multi_draw_indirect_count( &mut self, indirect_buffer: &Buffer, indirect_offset: BufferAddress, count_buffer: &Buffer, count_offset: BufferAddress, max_count: u32 )

Dispatches multiple draw calls from the active vertex buffer(s) based on the contents of the indirect_buffer. The count buffer is read to determine how many draws to issue.

The indirect buffer must be long enough to account for max_count draws, however only count draws will be read. If count is greater than max_count, max_count will be used.

The active vertex buffers can be set with RenderPass::set_vertex_buffer.

The structure expected in indirect_buffer must conform to DrawIndirectArgs. These draw structures are expected to be tightly packed.

The structure expected in count_buffer is the following:

#[repr(C)]
struct DrawIndirectCount {
    count: u32, // Number of draw calls to issue.
}

This drawing command uses the current render state, as set by preceding set_*() methods. It is not affected by changes to the state that are performed after it is called.

source

pub fn multi_draw_indexed_indirect_count( &mut self, indirect_buffer: &Buffer, indirect_offset: BufferAddress, count_buffer: &Buffer, count_offset: BufferAddress, max_count: u32 )

Dispatches multiple draw calls from the active index buffer and the active vertex buffers, based on the contents of the indirect_buffer. The count buffer is read to determine how many draws to issue.

The indirect buffer must be long enough to account for max_count draws, however only count draws will be read. If count is greater than max_count, max_count will be used.

The active index buffer can be set with RenderPass::set_index_buffer, while the active vertex buffers can be set with RenderPass::set_vertex_buffer.

The structure expected in indirect_buffer must conform to DrawIndexedIndirectArgs.

These draw structures are expected to be tightly packed.

The structure expected in count_buffer is the following:

#[repr(C)]
struct DrawIndexedIndirectCount {
    count: u32, // Number of draw calls to issue.
}

This drawing command uses the current render state, as set by preceding set_*() methods. It is not affected by changes to the state that are performed after it is called.

source§

impl<'encoder> RenderPass<'encoder>

Features::PUSH_CONSTANTS must be enabled on the device in order to call these functions.

source

pub fn set_push_constants( &mut self, stages: ShaderStages, offset: u32, data: &[u8] )

Set push constant data for subsequent draw calls.

Write the bytes in data at offset offset within push constant storage, all of which are accessible by all the pipeline stages in stages, and no others. Both offset and the length of data must be multiples of PUSH_CONSTANT_ALIGNMENT, which is always 4.

For example, if offset is 4 and data is eight bytes long, this call will write data to bytes 4..12 of push constant storage.

Stage matching

Every byte in the affected range of push constant storage must be accessible to exactly the same set of pipeline stages, which must match stages. If there are two bytes of storage that are accessible by different sets of pipeline stages - say, one is accessible by fragment shaders, and the other is accessible by both fragment shaders and vertex shaders - then no single set_push_constants call may affect both of them; to write both, you must make multiple calls, each with the appropriate stages value.

Which pipeline stages may access a given byte is determined by the pipeline’s PushConstant global variable and (if it is a struct) its members’ offsets.

For example, suppose you have twelve bytes of push constant storage, where bytes 0..8 are accessed by the vertex shader, and bytes 4..12 are accessed by the fragment shader. This means there are three byte ranges each accessed by a different set of stages:

  • Bytes 0..4 are accessed only by the fragment shader.

  • Bytes 4..8 are accessed by both the fragment shader and the vertex shader.

  • Bytes 8..12 are accessed only by the vertex shader.

To write all twelve bytes requires three set_push_constants calls, one for each range, each passing the matching stages mask.

source§

impl<'encoder> RenderPass<'encoder>

Features::TIMESTAMP_QUERY_INSIDE_PASSES must be enabled on the device in order to call these functions.

source

pub fn write_timestamp(&mut self, query_set: &QuerySet, query_index: u32)

Issue a timestamp command at this point in the queue. The timestamp will be written to the specified query set, at the specified index.

Must be multiplied by Queue::get_timestamp_period to get the value in nanoseconds. Absolute values have no meaning, but timestamps can be subtracted to get the time it takes for a string of operations to complete.

source§

impl<'encoder> RenderPass<'encoder>

source

pub fn begin_occlusion_query(&mut self, query_index: u32)

Start a occlusion query on this render pass. It can be ended with end_occlusion_query. Occlusion queries may not be nested.

source

pub fn end_occlusion_query(&mut self)

End the occlusion query on this render pass. It can be started with begin_occlusion_query. Occlusion queries may not be nested.

source§

impl<'encoder> RenderPass<'encoder>

Features::PIPELINE_STATISTICS_QUERY must be enabled on the device in order to call these functions.

source

pub fn begin_pipeline_statistics_query( &mut self, query_set: &QuerySet, query_index: u32 )

Start a pipeline statistics query on this render pass. It can be ended with end_pipeline_statistics_query. Pipeline statistics queries may not be nested.

source

pub fn end_pipeline_statistics_query(&mut self)

End the pipeline statistics query on this render pass. It can be started with begin_pipeline_statistics_query. Pipeline statistics queries may not be nested.

Trait Implementations§

source§

impl<'encoder> Debug for RenderPass<'encoder>

source§

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter. Read more
source§

impl<'a> RenderEncoder<'a> for RenderPass<'a>

source§

fn set_bind_group( &mut self, index: u32, bind_group: Option<&'a BindGroup>, offsets: &[DynamicOffset] )

Sets the active bind group for a given bind group index. The bind group layout in the active pipeline when any draw() function is called must match the layout of this bind group. Read more
source§

fn set_pipeline(&mut self, pipeline: &'a RenderPipeline)

Sets the active render pipeline. Read more
source§

fn set_index_buffer( &mut self, buffer_slice: BufferSlice<'a>, index_format: IndexFormat )

Sets the active index buffer. Read more
source§

fn set_vertex_buffer(&mut self, slot: u32, buffer_slice: BufferSlice<'a>)

Assign a vertex buffer to a slot. Read more
source§

fn draw(&mut self, vertices: Range<u32>, instances: Range<u32>)

Draws primitives from the active vertex buffer(s). Read more
source§

fn draw_indexed( &mut self, indices: Range<u32>, base_vertex: i32, instances: Range<u32> )

Draws indexed primitives using the active index buffer and the active vertex buffers. Read more
source§

fn draw_indirect( &mut self, indirect_buffer: &'a Buffer, indirect_offset: BufferAddress )

Draws primitives from the active vertex buffer(s) based on the contents of the indirect_buffer. Read more
source§

fn draw_indexed_indirect( &mut self, indirect_buffer: &'a Buffer, indirect_offset: BufferAddress )

Draws indexed primitives using the active index buffer and the active vertex buffers, based on the contents of the indirect_buffer. Read more
source§

fn set_push_constants(&mut self, stages: ShaderStages, offset: u32, data: &[u8])

wgt::Features::PUSH_CONSTANTS must be enabled on the device in order to call this function. Read more

Auto Trait Implementations§

§

impl<'encoder> !RefUnwindSafe for RenderPass<'encoder>

§

impl<'encoder> Send for RenderPass<'encoder>

§

impl<'encoder> Sync for RenderPass<'encoder>

§

impl<'encoder> Unpin for RenderPass<'encoder>

§

impl<'encoder> !UnwindSafe for RenderPass<'encoder>

Blanket Implementations§

source§

impl<T> Any for T
where T: 'static + ?Sized,

source§

fn type_id(&self) -> TypeId

Gets the TypeId of self. Read more
source§

impl<T> Borrow<T> for T
where T: ?Sized,

source§

fn borrow(&self) -> &T

Immutably borrows from an owned value. Read more
source§

impl<T> BorrowMut<T> for T
where T: ?Sized,

source§

fn borrow_mut(&mut self) -> &mut T

Mutably borrows from an owned value. Read more
§

impl<T> Downcast<T> for T

§

fn downcast(&self) -> &T

source§

impl<T> From<T> for T

source§

fn from(t: T) -> T

Returns the argument unchanged.

source§

impl<T, U> Into<U> for T
where U: From<T>,

source§

fn into(self) -> U

Calls U::from(self).

That is, this conversion is whatever the implementation of From<T> for U chooses to do.

source§

impl<T, U> TryFrom<U> for T
where U: Into<T>,

§

type Error = Infallible

The type returned in the event of a conversion error.
source§

fn try_from(value: U) -> Result<T, <T as TryFrom<U>>::Error>

Performs the conversion.
source§

impl<T, U> TryInto<U> for T
where U: TryFrom<T>,

§

type Error = <U as TryFrom<T>>::Error

The type returned in the event of a conversion error.
source§

fn try_into(self) -> Result<U, <U as TryFrom<T>>::Error>

Performs the conversion.
§

impl<T> Upcast<T> for T

§

fn upcast(&self) -> Option<&T>

§

impl<T> WasmNotSend for T
where T: Send,

§

impl<T> WasmNotSendSync for T

§

impl<T> WasmNotSync for T
where T: Sync,