Enum ComputeDesc 

pub enum ComputeDesc {
    F32,
    F64,
    MinF32,
    MinF64,
    MinF16,
    MinBf16,
    Tf32,
    C32F,
}

Selects the cuTENSOR mixed-precision compute path. MinF32 is the natural pairing for f16/bf16 inputs that accumulate in f32.

Variants

F32

CUTENSOR_R_32F — full fp32, no min-precision.

F64

CUTENSOR_R_64F — full fp64.

MinF32

CUTENSOR_R_MIN_32F — fp32 accumulation with min-precision kernels. Default for f32 / f16 / bf16 inputs.

MinF64

CUTENSOR_R_MIN_64F — fp64 accumulation, min-precision.

MinF16

CUTENSOR_R_MIN_16F — pure half-precision compute.

MinBf16

CUTENSOR_R_MIN_16BF — pure bf16 compute.

Tf32

CUTENSOR_R_MIN_TF32 — TF32 accumulation (Ampere+).

C32F

CUTENSOR_C_32F — complex fp32 inputs/compute.

Implementations

impl ComputeDesc

pub fn tag(self) -> &'static str

Trait Implementations

impl Clone for ComputeDesc

fn clone(&self) -> ComputeDesc

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ComputeDesc

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

Formats the value using the given formatter.

impl Hash for ComputeDesc

fn hash<__H: Hasher>(&self, state: &mut __H)

Feeds this value into the given Hasher.

fn hash_slice<H>(data: &[Self], state: &mut H)

where H: Hasher, Self: Sized,

Feeds a slice of this type into the given Hasher.

impl PartialEq for ComputeDesc

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

impl Eq for ComputeDesc

impl StructuralPartialEq for ComputeDesc