Mojo function

copy_accum_to_gmem

copy_accum_to_gmem[c_type: DType, c_layout: Layout, c_smem_layout: Layout, c_desc_layout: Layout, num_accum_pipeline_stages: Int, /, *, repeat: Int, accum_type: DType, cta_group: Int, epilogue_dtype: DType, block_tile_shape: IndexList[3], mma_shape: IndexList[3], num_output_warps: UInt, c_swizzle: TensorMapSwizzle = TensorMapSwizzle.SWIZZLE_128B, elementwise_compute_lambda_fn: OptionalReg[elementwise_compute_lambda_type] = 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], mma_output_stage: UInt32, tmem_offset: UInt32, c_coord: Tuple[UInt32, UInt32], c_shape: Tuple[UInt32, UInt32])