Struct SingleLane 

pub struct SingleLane<T: GpuValue, const LANE: u8> { /* private fields */ }

A value that exists ONLY in a specific lane.

Models the result of a reduction — only one lane has the answer. Prevents the common bug of reading a reduction result from all lanes (undefined behavior in CUDA).

Implementations

impl<T: GpuValue, const LANE: u8> SingleLane<T, LANE>

pub fn new(value: T) -> Self

pub fn get(self) -> T

Read the value. Only valid in the owning lane.

pub fn broadcast(self) -> Uniform<T>

Broadcast to all lanes — the ONLY safe way to share with other lanes.

Trait Implementations

impl<T: Clone + GpuValue, const LANE: u8> Clone for SingleLane<T, LANE>

fn clone(&self) -> SingleLane<T, LANE>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<T: Debug + GpuValue, const LANE: u8> Debug for SingleLane<T, LANE>

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

Formats the value using the given formatter.

impl<T: PartialEq + GpuValue, const LANE: u8> PartialEq for SingleLane<T, LANE>

fn eq(&self, other: &SingleLane<T, LANE>) -> 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<T: Copy + GpuValue, const LANE: u8> Copy for SingleLane<T, LANE>

impl<T: GpuValue, const LANE: u8> StructuralPartialEq for SingleLane<T, LANE>