Struct MegakernelLaunchGeometry 

pub struct MegakernelLaunchGeometry {
    pub workgroup_size_x: u32,
    pub slot_count: u32,
    pub dispatch_grid: [u32; 3],
}

Host-side launch geometry for a finite megakernel dispatch.

Fields

workgroup_size_x: u32

Lanes per worker workgroup used to compile the program.

slot_count: u32

Ring slots allocated for the dispatch, padded to a full workgroup.

dispatch_grid: [u32; 3]

Grid submitted to the backend.

Implementations

impl MegakernelLaunchGeometry

pub fn from_item_count( item_count: usize, worker_count: u32, max_workgroup_size_x: u32, ) -> Result<Self, BackendError>

Build geometry for item_count host work items.

Errors

Returns BackendError when the host queue cannot be represented by the u32 ring protocol.

pub fn from_slots( slot_count: u32, worker_count: u32, max_workgroup_size_x: u32, ) -> Self

Build geometry for an already-sized ring.

pub const fn covering_worker_groups(&self) -> u32

Number of worker workgroups needed to cover every ring slot exactly once.

pub fn dispatch_config(&self, timeout: Option<Duration>) -> DispatchConfig

Build the backend dispatch config that matches this launch geometry.

Trait Implementations

impl Clone for MegakernelLaunchGeometry

fn clone(&self) -> MegakernelLaunchGeometry

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for MegakernelLaunchGeometry

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

Formats the value using the given formatter.

impl PartialEq for MegakernelLaunchGeometry

fn eq(&self, other: &MegakernelLaunchGeometry) -> 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 MegakernelLaunchGeometry

impl Eq for MegakernelLaunchGeometry

impl StructuralPartialEq for MegakernelLaunchGeometry