Mojo function

naive_blockwise_scaled_fp8_matmul

naive_blockwise_scaled_fp8_matmul[c_type: DType, a_type: DType, b_type: DType, scales_type: DType, c_shape: DimList, a_shape: DimList, b_shape: DimList, a_scale_shape: DimList, b_scale_shape: DimList, //, *, BLOCK_DIM: Int = 16, transpose_b: Bool = False, elementwise_lambda_fn: OptionalReg[fn[DType, Int, Int](IndexList[2], SIMD[$0, $1]) capturing -> None] = OptionalReg[fn[DType, Int, Int](IndexList[2], SIMD[$0, $1]) capturing -> None]({:i1 0, 1}), s_type: DType = get_accum_type[::DType,::DType]()](c_device: NDBuffer[c_type, 2, origin, c_shape], a_device: NDBuffer[a_type, 2, origin, a_shape], b_device: NDBuffer[b_type, 2, origin, b_shape], a_scales_device: NDBuffer[scales_type, 2, origin, a_scale_shape], b_scales_device: NDBuffer[scales_type, 2, origin, b_scale_shape], ctx: DeviceContext)