Struct LaunchConfig 

pub struct LaunchConfig {
    pub grid_dim: (u32, u32, u32),
    pub block_dim: (u32, u32, u32),
    pub shared_mem_bytes: u32,
}

Kernel launch configuration (CUDA-style grid/block decomposition).

For CPU and Metal backends these values are informational only; the actual parallelism strategy is determined by the backend itself.

Fields

grid_dim: (u32, u32, u32)

Number of thread-blocks (x, y, z).

block_dim: (u32, u32, u32)

Threads per block (x, y, z).

shared_mem_bytes: u32

Dynamic shared memory per block in bytes.

Implementations

impl LaunchConfig

pub fn for_n_elements(n: usize) -> Self

Auto-compute a 1-D launch configuration for n elements.

Uses a block size of 256 threads and rounds the grid up to cover all elements. shared_mem_bytes is set to zero.

pub fn default_1d() -> Self

A sensible default 1-D config (1 block of 256 threads).

Trait Implementations

impl Clone for LaunchConfig

fn clone(&self) -> LaunchConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for LaunchConfig

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

Formats the value using the given formatter.

impl PartialEq for LaunchConfig

fn eq(&self, other: &LaunchConfig) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Copy for LaunchConfig

impl Eq for LaunchConfig

impl StructuralPartialEq for LaunchConfig