Trait ReductionOp 

pub trait ReductionOp<T>:
    Sealed
    + Copy
    + 'static
where
    T: NumericElement,
{
    // Required methods
    unsafe fn accumulate<Arch>(
        acc: <Arch as SimdKernel<T>>::Vector,
        v: <Arch as SimdKernel<T>>::Vector,
    ) -> <Arch as SimdKernel<T>>::Vector
       where Arch: SimdKernel<T>;
    unsafe fn finalize<Arch>(acc: <Arch as SimdKernel<T>>::Vector) -> T
       where Arch: SimdKernel<T>;
    fn identity_scalar() -> T;
    fn scalar_combine(a: T, b: T) -> T;

    // Provided methods
    unsafe fn fma_pair_accumulate<Arch>(
        acc: <Arch as SimdKernel<T>>::Vector,
        a: <Arch as SimdKernel<T>>::Vector,
        b: <Arch as SimdKernel<T>>::Vector,
    ) -> <Arch as SimdKernel<T>>::Vector
       where Arch: SimdKernel<T> { ... }
    fn scalar_accumulate(acc: T, elem: T) -> T { ... }
    unsafe fn identity_vector<Arch>() -> <Arch as SimdKernel<T>>::Vector
       where Arch: SimdKernel<T> { ... }
    unsafe fn transform_vector<Arch>(
        v: <Arch as SimdKernel<T>>::Vector,
    ) -> <Arch as SimdKernel<T>>::Vector
       where Arch: SimdKernel<T> { ... }
    unsafe fn combine_vectors<Arch>(
        a: <Arch as SimdKernel<T>>::Vector,
        b: <Arch as SimdKernel<T>>::Vector,
    ) -> <Arch as SimdKernel<T>>::Vector
       where Arch: SimdKernel<T> { ... }
}

Sealed ZST trait for SIMD horizontal reduction strategies.

Implementors define how a vector accumulator is updated (accumulate) and how the final scalar result is extracted (finalize). Both methods are #[inline(always)] and carry no branching — DCE eliminates unused strategies entirely.

Identity Element

identity_scalar() returns the reduction identity (0 for Sum, T::MAX_VALUE for Min, T::MIN_VALUE for Max). It is used for empty-slice fast paths and for combining the scalar tail with the SIMD result via scalar_combine.

Required Methods

unsafe fn accumulate<Arch>( acc: <Arch as SimdKernel<T>>::Vector, v: <Arch as SimdKernel<T>>::Vector, ) -> <Arch as SimdKernel<T>>::Vector

where Arch: SimdKernel<T>,

Merge a new data vector v into accumulator acc.

Safety

Processor must support the target feature of Arch.

unsafe fn finalize<Arch>(acc: <Arch as SimdKernel<T>>::Vector) -> T

where Arch: SimdKernel<T>,

Reduce the final accumulator to a scalar.

Safety

Processor must support the target feature of Arch.

fn identity_scalar() -> T

The identity element for this reduction as a scalar.

Default: must be overridden. Sum returns T::ZERO, Min returns T::MAX_VALUE, Max returns T::MIN_VALUE.

fn scalar_combine(a: T, b: T) -> T

Combine two scalar partial results using this reduction.

For Sum: addition. For Min: min_scalar. For Max: max_scalar.

Provided Methods

unsafe fn fma_pair_accumulate<Arch>( acc: <Arch as SimdKernel<T>>::Vector, a: <Arch as SimdKernel<T>>::Vector, b: <Arch as SimdKernel<T>>::Vector, ) -> <Arch as SimdKernel<T>>::Vector

where Arch: SimdKernel<T>,

FMA-aware pairwise accumulation: acc = fuse(acc, a, b) where fuse may use a single fused multiply-add instruction rather than a separate mul + accumulate.

Default: Self::accumulate(acc, Arch::mul(a, b)) — a correct two-instruction fallback. Override this for Dot and similar operations that can exploit Arch::fmadd.

Safety

Processor must support the target feature of Arch.

fn scalar_accumulate(acc: T, elem: T) -> T

Accumulate a single scalar element elem into a scalar accumulator acc.

Default: delegates to Self::scalar_combine(acc, elem). Override this for reductions whose SIMD accumulate applies a per-element transform (e.g. SquaredSum applies elem * elem before adding). The scalar tail path uses this method instead of scalar_combine to maintain correctness for slices shorter than Arch::LANE_COUNT.

unsafe fn identity_vector<Arch>() -> <Arch as SimdKernel<T>>::Vector

where Arch: SimdKernel<T>,

Splat the identity element into a vector register.

Default: Arch::splat(Self::identity_scalar()). Backends may override.

Safety

Processor must support the target feature of Arch.

unsafe fn transform_vector<Arch>( v: <Arch as SimdKernel<T>>::Vector, ) -> <Arch as SimdKernel<T>>::Vector

where Arch: SimdKernel<T>,

Per-element lane transform applied before combining (identity by default).

Reductions with a per-element transform (AbsSum applies abs) override this so the reduce loop can seed unrolled accumulators with transform_vector(load(...)) instead of raw loads.

Safety

Processor must support the target feature of Arch.

unsafe fn combine_vectors<Arch>( a: <Arch as SimdKernel<T>>::Vector, b: <Arch as SimdKernel<T>>::Vector, ) -> <Arch as SimdKernel<T>>::Vector

where Arch: SimdKernel<T>,

Merge two partial accumulators WITHOUT the per-element transform.

accumulate is combine_vectors(acc, transform_vector(v)); the reduce loop’s cross-accumulator merge must use this method, because the partials are already transformed. Default delegates to accumulate, which is correct exactly when transform_vector is the identity — transform-bearing ops must override both.

Safety

Processor must support the target feature of Arch.

Dyn Compatibility

This trait is not dyn compatible.

In older versions of Rust, dyn compatibility was called "object safety".

Implementors

impl<T> ReductionOp<T> for AbsMax

where T: NumericElement,

impl<T> ReductionOp<T> for AbsSum

where T: NumericElement,

impl<T> ReductionOp<T> for Dot

where T: NumericElement,

impl<T> ReductionOp<T> for Max

where T: NumericElement,

impl<T> ReductionOp<T> for Min

where T: NumericElement,

impl<T> ReductionOp<T> for Product

where T: NumericElement,

impl<T> ReductionOp<T> for Sum

where T: NumericElement,