Linear Algebra

    /**
     * @brief GEMM for 1 byte types.
     *
     * @tparam AlphaType for the scalar alpha
     * @tparam OCMShapeA Type for the matrix A (In OCM)
     * @tparam OCMShapeB Type for the matrix B (In OCM)
     * @tparam BetaType Type for the scalar beta
     * @tparam OCMShapeC Type for the matrix C (In OCM)
     * @param alpha the scalar alpha.
     * @param matA the first matrix, A.
     * @param matB the second matrix, B.
     * @param beta the scalar beta.
     * @param matC matrix C.
     * @param matOut output matrix.
     *
     */
    template <typename AlphaType,
              typename OCMShapeA,
              typename OCMShapeB,
              typename BetaType,
              typename OCMShapeC,
              isOcmTensor<OCMShapeA> = 0,
              isOcmTensor<OCMShapeB> = 0,
              isOcmTensor<OCMShapeC> = 0,
              std::enable_if_t<std::is_integral<typename OCMShapeB::elemType>::value &&
                                 sizeof(typename OCMShapeB::elemType) == 1,
                               int>  = 0>
    INLINE void gemm(
      AlphaType alpha, OCMShapeA& matA, OCMShapeB& matB, BetaType beta, OCMShapeC& matC, OCMShapeC& matOut) {