Struct ComputeShaderSimulator 

pub struct ComputeShaderSimulator { /* private fields */ }

High-level entry point for simulated GPU compute.

Manages a default group size and provides factory methods for creating typed ShaderKernel instances.

Implementations

impl ComputeShaderSimulator

pub fn new(default_group_size: usize) -> Self

Create a simulator with the given default work-group size.

Sizes of 0 are normalised to 64 (a common GPU default).

pub fn default_group_size(&self) -> usize

The default number of threads per work group.

pub fn create_kernel<T: Send + Sync + 'static>( &self, name: impl Into<String>, f: impl Fn(&ThreadGroupContext, &mut T) + Send + Sync + 'static, ) -> ShaderKernel<T>

Create a kernel using the simulator’s default group size.

pub fn create_kernel_with_group_size<T: Send + Sync + 'static>( &self, name: impl Into<String>, group_size: usize, f: impl Fn(&ThreadGroupContext, &mut T) + Send + Sync + 'static, ) -> ShaderKernel<T>

Create a kernel with a custom group size, ignoring the simulator default.

pub fn dispatch<T: Send + Sync>( &self, kernel: &ShaderKernel<T>, data: &mut [T], work_groups: usize, )

Dispatch kernel over data using work_groups work groups.

pub fn dispatch_with_barrier<T: Send + Sync + Clone>( &self, kernel: &ShaderKernel<T>, data: &mut [T], work_groups: usize, )

Dispatch kernel and wait for all threads to complete.

Rayon’s par_iter_mut already joins all threads before returning, so this is semantically equivalent to dispatch. The method exists to model GPU barriers explicitly in calling code.

Trait Implementations

impl Clone for ComputeShaderSimulator

fn clone(&self) -> ComputeShaderSimulator

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ComputeShaderSimulator

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

Formats the value using the given formatter.