Mojo function

multi_stage_store_C

multi_stage_store_C[c_type: DType, c_smem_layout: Layout, c_layout: Layout, c_desc_layout: Layout, num_accum_pipeline_stages: Int, /, *, input_type: DType, accum_type: DType, block_tile_shape: IndexList[3], mma_shape: IndexList[3], stage_stride_cols: Int, c_swizzle: TensorMapSwizzle = TensorMapSwizzle.SWIZZLE_128B, cta_group: Int = 1, num_output_warps: Int = 4, max_tmem_cols: Int = 512, elementwise_compute_lambda_fn: OptionalReg[fn[dtype: DType, width: Int, *, alignment: Int = 1](IndexList[2], SIMD[dtype, width]) capturing -> SIMD[dtype, width]] = None, register_based_epilogue: Bool = True, transpose_c: Bool = False](c_iter: LayoutTensorIter[c_type, c_smem_layout, MutAnyOrigin, address_space=AddressSpace.SHARED, alignment=128], c_tma_op: TMATensorTile[c_type, c_layout, c_desc_layout], mma_output_pipeline: ProducerConsumerPipeline[num_accum_pipeline_stages], tmem_addr: UInt32, work_tile_coord: Tuple[UInt32, UInt32, UInt32], elect_one_warp: Bool, M: UInt32, N: UInt32)