(This is the documentation for SDL3, which is under heavy development and the API is changing! SDL2 is the current stable version!)

SDL_BeginGPUComputePass

Begins a compute pass on a command buffer.

Header File

Syntax

SDL_GPUComputePass* SDL_BeginGPUComputePass(
    SDL_GPUCommandBuffer *command_buffer,
    const SDL_GPUStorageTextureWriteOnlyBinding *storage_texture_bindings,
    Uint32 num_storage_texture_bindings,
    const SDL_GPUStorageBufferWriteOnlyBinding *storage_buffer_bindings,
    Uint32 num_storage_buffer_bindings);

Function Parameters

SDL_GPUCommandBuffer *	command_buffer	a command buffer.
const SDL_GPUStorageTextureWriteOnlyBinding *	storage_texture_bindings	an array of writeable storage texture binding structs.
Uint32	num_storage_texture_bindings	the number of storage textures to bind from the array.
const SDL_GPUStorageBufferWriteOnlyBinding *	storage_buffer_bindings	an array of writeable storage buffer binding structs.
Uint32	num_storage_buffer_bindings	the number of storage buffers to bind from the array.

Return Value

(SDL_GPUComputePass *) Returns a compute pass handle.

Remarks

A compute pass is defined by a set of texture subresources and buffers that will be written to by compute pipelines. These textures and buffers must have been created with the COMPUTE_STORAGE_WRITE bit. All operations related to compute pipelines must take place inside of a compute pass. You must not begin another compute pass, or a render pass or copy pass before ending the compute pass.

A VERY IMPORTANT NOTE Textures and buffers bound as write-only MUST NOT be read from during the compute pass. Doing so will result in undefined behavior. If your compute work requires reading the output from a previous dispatch, you MUST end the current compute pass and begin a new one before you can safely access the data.

Version

This function is available since SDL 3.0.0.