Enum poc_kokkos_rs::routines::parameters::RangePolicy

pub enum RangePolicy<const N: usize> {
    RangePolicy(Range<usize>),
    MDRangePolicy([Range<usize>; N]),
    TeamPolicy {
        league_size: usize,
        team_size: usize,
        vector_size: usize,
    },
    PerTeam,
    PerThread,
    TeamThreadRange,
    TeamThreadMDRange,
    TeamVectorRange,
    TeamVectorMDRange,
    ThreadVectorRange,
    ThreadVectorMDRange,
}

Range Policy enum.

This holds information related to the looping structure adopted by the routine.

Variants

RangePolicy(Range<usize>)

1D iteration range.

MDRangePolicy([Range<usize>; N])

N-dimensional iteration range.

TeamPolicy

Fields

league_size: usize

Number of team.

team_size: usize

Number of threads per team.

vector_size: usize

Number of vector

Team-based iteration policy.

PerTeam

Policy used to ensure each team execute the body once and only once.

PerThread

Policy used to ensure each thread execute the body once and only once.

TeamThreadRange

Medium-level depth. Can host further nests using vectors.

TeamThreadMDRange

Medium-level depth. Can host further nests using vectors.

TeamVectorRange

Medium-level depth. Cannot host further nests.

TeamVectorMDRange

Medium-level depth. Cannot host further nests.

ThreadVectorRange

Inner-level depth. Cannot host further nests.

ThreadVectorMDRange

Inner-level depth. Cannot host further nests.

Trait Implementations

impl<const N: usize> Clone for RangePolicy<N>

fn clone(&self) -> RangePolicy<N>

Returns a copy of the value.

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

Performs copy-assignment from source.

impl<const N: usize> Debug for RangePolicy<N>

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

Formats the value using the given formatter.