Struct TosaInt8MatmulCase 

pub struct TosaInt8MatmulCase {
    pub name: &'static str,
    pub artifact: &'static [u8],
    pub inputs: &'static [PackedTensor],
    pub zero_points: [i8; 2],
    pub outputs: &'static [Int32Tensor],
}

A stable TOSA INT8 matrix multiplication with exact INT32 accumulation.

Fields

name: &'static str

Diagnostic case name.

artifact: &'static [u8]

TOSA 1.0 FlatBuffer payload.

inputs: &'static [PackedTensor]

Signed INT8 block inputs in declared slot order.

zero_points: [i8; 2]

Compile-time zero points for the left and right operands.

outputs: &'static [Int32Tensor]

Exact INT32 block outputs in declared slot order.

Implementations

impl TosaInt8MatmulCase

pub fn output_matches(&self, output: usize, actual: &[i32]) -> bool

Compare one backend output with the exact INT32 oracle.

Trait Implementations

impl Clone for TosaInt8MatmulCase

fn clone(&self) -> TosaInt8MatmulCase

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for TosaInt8MatmulCase

impl Debug for TosaInt8MatmulCase

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

Formats the value using the given formatter.

impl Eq for TosaInt8MatmulCase

impl PartialEq for TosaInt8MatmulCase

fn eq(&self, other: &TosaInt8MatmulCase) -> 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 StructuralPartialEq for TosaInt8MatmulCase