Mojo function

blackwell_tma_umma_warp_specialized_blockwise_fp8_kernel

blackwell_tma_umma_warp_specialized_blockwise_fp8_kernel[a_type: DType, b_type: DType, c_type: DType, a_rank: Int, a_tile_shape: IndexList[a_rank], a_desc_shape: IndexList[a_rank], b_rank: Int, b_tile_shape: IndexList[b_rank], b_desc_shape: IndexList[b_rank], c_rank: Int, c_tile_shape: IndexList[c_rank], c_desc_shape: IndexList[c_rank], a_scales_rank: Int, a_scales_tile_shape: IndexList[a_scales_rank], a_scales_desc_shape: IndexList[a_scales_rank], a_scales_type: DType, b_scales_type: DType, b_scales_layout: TensorLayout, transpose_b: Bool, config: MatmulConfig[a_type, b_type, c_type, transpose_b], num_pipeline_stages: Int, cluster_shape: StaticTuple[Int32, 3]](a_tma_op: TMATensorTile[a_type, a_rank, a_tile_shape, a_desc_shape], b_tma_op: TMATensorTile[b_type, b_rank, b_tile_shape, b_desc_shape], c_tma_op: TMATensorTile[c_type, c_rank, c_tile_shape, c_desc_shape], a_scales_tma_op: TMATensorTile[a_scales_type, a_scales_rank, a_scales_tile_shape, a_scales_desc_shape], cluster_dim: StaticTuple[Int32, 3], num_iters: Int, b_scales: TileTensor[b_scales_type, b_scales_layout, ImmutAnyOrigin], problem_shape: StaticTuple[Int32, 3])