==================================================Ascend ============================= test session starts ============================== platform linux -- Python 3.9.19, pytest-6.2.5, py-1.11.0, pluggy-1.5.0 rootdir: /home/jenkins/mindspore/testcases/testcases/tests/st/graph_kernel, configfile: ../../../../../../sault/virtual_test/virtualenv_003/sault/config/pytest.ini plugins: mock-3.14.0, hydra-core-1.3.2, forked-1.6.0, anyio-4.9.0, xdist-1.32.0 collected 1 item test_dvm_inplace.py TotalTime = 0.336976, [21] [bootstrap]: 0.00106948 [type_inference]: 0.274913 [event_method]: 0.00011777 [auto_monad]: 0.00034828 [graph_reusing]: 8.62e-06 [inline]: 2.85002e-06 [add_attr]: 0.0479529, [1] [add_attr_with_inline]: 0.0479368, [1] [Cycle 1]: 0.00015117, [2] [tag_attr]: 4.837e-05 [meta_addattr_fg_expand]: 1.015e-05 [parallel-infer-symbol]: 4.59998e-06 [pre_auto_parallel]: 8.171e-05 [insert-virtual-dataset]: 2.79001e-06 [parallel-infer-symbol-second]: 7.90023e-07 [dataset_repeat_opt]: 2.88e-06 [pipeline_split]: 2.02999e-06 [optimize]: 0.0111824, [53] [py_interpret_to_execute]: 4.713e-05 [rewriter_before_opt_a]: 0.00016578 [opt_a]: 0.00731619, [2] [Cycle 1]: 0.00557797, [45] [expand_dump_flag]: 3.63999e-06 [switch_simplify]: 0.00016263 [loop_unroll]: 5.567e-05 [a_1]: 0.00169578 [with_stream_mark]: 4.391e-05 [recompute_prepare]: 2.984e-05 [updatestate_depend_eliminate]: 8.362e-05 [updatestate_assign_eliminate]: 1.562e-05 [updatestate_loads_eliminate]: 1.251e-05 [parameter_eliminate]: 3.31001e-06 [a_2]: 0.00026672 [accelerated_algorithm]: 4.44e-05 [shard]: 2.23998e-06 [meta_shard_fg_expand]: 4.42998e-06 [shard_inline]: 1.72e-05 [merge_send_recv]: 3.588e-05 [auto_parallel]: 1.799e-05 [parallel]: 4.988e-05 [flash_sp]: 2.407e-05 [merge_comm]: 1.139e-05 [allreduce_fusion]: 8.64e-06 [matmul_add_comm_reduction]: 2.096e-05 [allreduce_slice_to_reducescatter]: 6.69999e-07 [virtual_shard_identity]: 2.245e-05 [virtual_dataset]: 1.677e-05 [get_grad_eliminate_]: 1.589e-05 [virtual_output]: 1.558e-05 [merge_forward]: 1.004e-05 [cell_reuse_recompute_pass]: 2.61e-06 [offload_activation]: 1.894e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.126e-05 [merge_recompute_call_nodes]: 1.74e-06 [before_grad]: 2.838e-05 [set_forward_comm_id_for_comm_node_pass]: 1.183e-05 [meta_fg_expand]: 7.71001e-06 [flash_sp_send_recv_attached]: 6.34001e-06 [receive_attached]: 1.219e-05 [after_resolve]: 2.784e-05 [a_after_grad]: 2.675e-05 [renormalize]: 0.00208629 [add_forward_monad_depend]: 1.179e-05 [auto_monad_grad]: 2.88998e-06 [auto_monad_eliminator]: 4.444e-05 [cse]: 7.818e-05 [a_3]: 0.00012873 [Cycle 2]: 0.00172089, [45] [expand_dump_flag]: 2.31e-06 [switch_simplify]: 1.956e-05 [loop_unroll]: 1.553e-05 [a_1]: 0.0004929 [with_stream_mark]: 2.735e-05 [recompute_prepare]: 1.818e-05 [updatestate_depend_eliminate]: 1.066e-05 [updatestate_assign_eliminate]: 1.049e-05 [updatestate_loads_eliminate]: 1.173e-05 [parameter_eliminate]: 2.72001e-06 [a_2]: 0.00023546 [accelerated_algorithm]: 2.487e-05 [shard]: 2.21e-06 [meta_shard_fg_expand]: 4.24002e-06 [shard_inline]: 1.679e-05 [merge_send_recv]: 1.682e-05 [auto_parallel]: 1.652e-05 [parallel]: 1.019e-05 [flash_sp]: 4.68001e-06 [merge_comm]: 1.031e-05 [allreduce_fusion]: 9.07999e-06 [matmul_add_comm_reduction]: 1.893e-05 [allreduce_slice_to_reducescatter]: 9.00007e-07 [virtual_shard_identity]: 2.027e-05 [virtual_dataset]: 1.617e-05 [get_grad_eliminate_]: 1.544e-05 [virtual_output]: 1.574e-05 [merge_forward]: 1.129e-05 [cell_reuse_recompute_pass]: 2.88e-06 [offload_activation]: 1.932e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.415e-05 [merge_recompute_call_nodes]: 2.06e-06 [before_grad]: 2.737e-05 [set_forward_comm_id_for_comm_node_pass]: 9.57999e-06 [meta_fg_expand]: 6.88e-06 [flash_sp_send_recv_attached]: 1.92001e-06 [receive_attached]: 2.47001e-06 [after_resolve]: 2.743e-05 [a_after_grad]: 2.67e-05 [renormalize]: 9.00181e-08 [add_forward_monad_depend]: 4.47e-06 [auto_monad_grad]: 2.34999e-06 [auto_monad_eliminator]: 3.278e-05 [cse]: 5.46e-05 [a_3]: 0.00014792 [py_interpret_to_execute_after_opt_a]: 3.191e-05 [slice_cell_reuse_recomputed_activation]: 2.76e-06 [rewriter_after_opt_a]: 0.000247 [convert_after_rewriter]: 2.23e-05 [order_py_execute_after_rewriter]: 1.125e-05 [mutable_eliminate]: 0.00077228 [opt_b]: 0.00054865, [1] [Cycle 1]: 0.00053758, [7] [b_1]: 0.00037183 [b_2]: 2.003e-05 [updatestate_depend_eliminate]: 1.821e-05 [updatestate_assign_eliminate]: 1.017e-05 [updatestate_loads_eliminate]: 1.216e-05 [renormalize]: 8.70001e-07 [cse]: 5.969e-05 [optimize_parallel_all_gather_comm]: 3.564e-05 [overlap_param_gather]: 1.92001e-06 [cconv]: 4.144e-05 [loop_unroll]: 0.0005773 [opt_after_cconv]: 0.00025421, [1] [Cycle 1]: 0.00024445, [7] [c_1]: 0.00010933 [parameter_eliminate]: 5.89e-06 [updatestate_depend_eliminate]: 1.534e-05 [updatestate_assign_eliminate]: 9.92001e-06 [updatestate_loads_eliminate]: 1.166e-05 [cse]: 5.153e-05 [renormalize]: 3.59985e-07 [remove_dup_value]: 2.813e-05 [tuple_transform]: 0.00017095, [1] [Cycle 1]: 0.00016529, [4] [d_1]: 0.0001221 [none_parameter_eliminate]: 2.04999e-06 [renormalize]: 1.50001e-07 [switch_simplify]: 1.805e-05 [partial_unused_args_eliminate]: 2.02999e-06 [add_recomputation]: 0.00012815 [cse_after_recomputation]: 5.526e-05, [1] [Cycle 1]: 4.885e-05, [1] [cse]: 4.049e-05 [environ_conv]: 3.733e-05 [swap_dp_allreduce_reducescatter]: 1.401e-05 [bias_add_comm_swap]: 3.48e-06 [label_micro_interleaved_index]: 8.40001e-06 [label_fine_grained_interleaved_index]: 2.93e-06 [merge_cast_opt]: 1.42e-06 [slice_recompute_activation]: 2.24001e-06 [micro_interleaved_order_control]: 3.23e-06 [assign_add_opt]: 1.95001e-06 [ForceFp32Comm]: 9.79984e-07 [remove_cast_before_assign_add]: 9.20001e-07 [full_micro_interleaved_order_control]: 2.35002e-06 [reorder_send_recv_between_fp_bp]: 2.98e-06 [comm_op_add_attrs]: 1.40999e-06 [add_comm_op_reuse_tag]: 1.29e-06 [interleave_split_concat_branches]: 1.16002e-06 [interleave_parallel_branches]: 1.07998e-06 [overlap_opt_shard_in_pipeline]: 2.634e-05 [overlap_opt_shard_grad_in_pipeline]: 1.93997e-06 [control_data_broadcast_order]: 3.438e-05 [grouped_pairwise_exchange_alltoall]: 1.67001e-06 [offloading_packed_experts]: 8.91002e-06 [overlap_recompute_and_grad_model_parallel]: 8.95001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.27e-06 [overlap_recompute_allgather_and_fa_grad]: 1.47999e-06 [overlap_recompute_comm]: 2.14e-06 [overlap_grad_ring_attention]: 8.33999e-06 [overlap_grad_flash_sp]: 6.108e-05 [begin_end_overlap_inline]: 7.89994e-07 [split_matmul_comm_elemetwise]: 2.26e-06 [split_layernorm_comm]: 2.06003e-06 [handle_group_info]: 1.15001e-06 [symbol_engine_optimizer]: 0.00018604, [1] [Cycle 1]: 0.0001793, [6] [build]: 3.009e-05 [elim_shapecalc]: 2.823e-05 [elim_not_effective]: 3.452e-05 [opt_reshape]: 1.669e-05 [fold_const_symbol]: 2.607e-05 [renormalize]: 4.00003e-07 [detach_backward]: 2.71999e-06 [pipeline_parallel_scheduler]: 1.72001e-06 [auto_monad_reorder]: 6.744e-05 [get_jit_bprop_graph]: 2.13002e-06 [rewriter_after_jit_bprop_graph]: 7.38999e-06 [opt_after_jit_grad]: 0.00085148 [validate]: 0.00012476 Sums bootstrap : 0.001069s : 0.37% type_inference : 0.274913s : 95.54% event_method : 0.000118s : 0.04% auto_monad : 0.000348s : 0.12% graph_reusing : 0.000009s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000048s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000010s : 0.00% parallel-infer-symbol : 0.000005s : 0.00% pre_auto_parallel : 0.000082s : 0.03% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000047s : 0.02% optimize.rewriter_before_opt_a : 0.000166s : 0.06% optimize.opt_a.expand_dump_flag : 0.000006s : 0.00% optimize.opt_a.switch_simplify : 0.000182s : 0.06% optimize.opt_a.loop_unroll : 0.000071s : 0.02% optimize.opt_a.a_1 : 0.002189s : 0.76% optimize.opt_a.with_stream_mark : 0.000071s : 0.02% optimize.opt_a.recompute_prepare : 0.000048s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000094s : 0.03% optimize.opt_a.updatestate_assign_eliminate : 0.000026s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000024s : 0.01% optimize.opt_a.parameter_eliminate : 0.000006s : 0.00% optimize.opt_a.a_2 : 0.000502s : 0.17% optimize.opt_a.accelerated_algorithm : 0.000069s : 0.02% optimize.opt_a.shard : 0.000004s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000009s : 0.00% optimize.opt_a.shard_inline : 0.000034s : 0.01% optimize.opt_a.merge_send_recv : 0.000053s : 0.02% optimize.opt_a.auto_parallel : 0.000035s : 0.01% optimize.opt_a.parallel : 0.000060s : 0.02% optimize.opt_a.flash_sp : 0.000029s : 0.01% optimize.opt_a.merge_comm : 0.000022s : 0.01% optimize.opt_a.allreduce_fusion : 0.000018s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000040s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000043s : 0.01% optimize.opt_a.virtual_dataset : 0.000033s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000031s : 0.01% optimize.opt_a.virtual_output : 0.000031s : 0.01% optimize.opt_a.merge_forward : 0.000021s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000038s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000065s : 0.02% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000056s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000021s : 0.01% optimize.opt_a.meta_fg_expand : 0.000015s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000008s : 0.00% optimize.opt_a.receive_attached : 0.000015s : 0.01% optimize.opt_a.after_resolve : 0.000055s : 0.02% optimize.opt_a.a_after_grad : 0.000053s : 0.02% optimize.opt_a.renormalize : 0.002086s : 0.73% optimize.opt_a.add_forward_monad_depend : 0.000016s : 0.01% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000077s : 0.03% optimize.opt_a.cse : 0.000133s : 0.05% optimize.opt_a.a_3 : 0.000277s : 0.10% optimize.py_interpret_to_execute_after_opt_a : 0.000032s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000247s : 0.09% optimize.convert_after_rewriter : 0.000022s : 0.01% optimize.order_py_execute_after_rewriter : 0.000011s : 0.00% optimize.mutable_eliminate : 0.000772s : 0.27% optimize.opt_b.b_1 : 0.000372s : 0.13% optimize.opt_b.b_2 : 0.000020s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000018s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000010s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000012s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000060s : 0.02% optimize.optimize_parallel_all_gather_comm : 0.000036s : 0.01% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000041s : 0.01% optimize.loop_unroll : 0.000577s : 0.20% optimize.opt_after_cconv.c_1 : 0.000109s : 0.04% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000015s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000010s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000012s : 0.00% optimize.opt_after_cconv.cse : 0.000052s : 0.02% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000028s : 0.01% optimize.tuple_transform.d_1 : 0.000122s : 0.04% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000018s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000128s : 0.04% optimize.cse_after_recomputation.cse : 0.000040s : 0.01% optimize.environ_conv : 0.000037s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000014s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000008s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000026s : 0.01% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000034s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000009s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000009s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000008s : 0.00% optimize.overlap_grad_flash_sp : 0.000061s : 0.02% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000030s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000028s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000035s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000017s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000026s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000067s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.00% opt_after_jit_grad : 0.000851s : 0.30% validate : 0.000125s : 0.04% Time group info: ------[substitution.] 0.000643 182 10.17% : 0.000065s : 10: substitution.arithmetic_simplify 1.74% : 0.000011s : 2: substitution.depend_value_elim 0.76% : 0.000005s : 10: substitution.elim_not_effective 0.62% : 0.000004s : 10: substitution.fold_const_symbol 1.87% : 0.000012s : 15: substitution.graph_param_transform 59.69% : 0.000384s : 9: substitution.inline 1.55% : 0.000010s : 20: substitution.j_node_and_user_rematch 3.58% : 0.000023s : 4: substitution.less_batch_normalization 1.02% : 0.000007s : 6: substitution.load_eliminater 2.23% : 0.000014s : 20: substitution.remove_not_recompute_node 1.57% : 0.000010s : 10: substitution.replace_old_param 1.85% : 0.000012s : 2: substitution.switch_simplify 6.35% : 0.000041s : 30: substitution.updatestate_pure_node_eliminater 6.98% : 0.000045s : 34: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.274821 2 99.41% : 0.273212s : 1: type_inference.infer 0.59% : 0.001609s : 1: type_inference.specialize ------[replace.] 0.000151 11 58.28% : 0.000088s : 9: replace.inline 41.72% : 0.000063s : 2: replace.switch_simplify ------[match.] 0.000388 11 97.33% : 0.000378s : 9: match.inline 2.67% : 0.000010s : 2: match.switch_simplify ------[predicate.] 0.000667 4397 1.05% : 0.000007s : 48: predicate.accumulaten_eliminater 0.84% : 0.000006s : 15: predicate.ad_related_special_op_eliminate 0.64% : 0.000004s : 30: predicate.addn_check_dump 1.27% : 0.000008s : 48: predicate.addn_zero_filter 1.01% : 0.000007s : 48: predicate.adjust_all_reduce_mul_add 2.48% : 0.000017s : 78: predicate.arithmetic_simplify 1.01% : 0.000007s : 48: predicate.cast_eliminate 0.63% : 0.000004s : 30: predicate.check_bprop_eliminate 0.60% : 0.000004s : 30: predicate.compare_switch_simplify 0.17% : 0.000001s : 15: predicate.const_output_eliminate 0.65% : 0.000004s : 30: predicate.depend_value_elim 1.06% : 0.000007s : 48: predicate.dict_get_item_const_eliminator 1.06% : 0.000007s : 48: predicate.dict_get_item_eliminator 1.00% : 0.000007s : 48: predicate.dict_set_item_eliminator 0.97% : 0.000006s : 30: predicate.dumpgradient_eliminate 0.20% : 0.000001s : 15: predicate.elim_not_effective 0.50% : 0.000003s : 15: predicate.elim_shapecalc_of_broadcastargs 1.39% : 0.000009s : 63: predicate.environ_add_const_eliminate 1.27% : 0.000008s : 63: predicate.environ_get_add_eliminate 1.29% : 0.000009s : 63: predicate.environ_get_depend_swap 1.95% : 0.000013s : 93: predicate.environ_get_eliminate 1.20% : 0.000008s : 63: predicate.environ_get_set_eliminate 1.16% : 0.000008s : 57: predicate.exchange_switch_depend_value 1.70% : 0.000011s : 57: predicate.float_depend_g_call 0.59% : 0.000004s : 30: predicate.float_environ_get_switch 0.91% : 0.000006s : 45: predicate.float_tuple_getitem_switch 0.18% : 0.000001s : 15: predicate.fold_const_symbol 0.70% : 0.000005s : 30: predicate.get_grad_eliminate 0.19% : 0.000001s : 15: predicate.graph_param_transform 0.57% : 0.000004s : 30: predicate.incorporate_call 0.55% : 0.000004s : 30: predicate.incorporate_call_switch 5.51% : 0.000037s : 195: predicate.inline 0.97% : 0.000006s : 30: predicate.inline_without_move 0.34% : 0.000002s : 30: predicate.j_node_and_user_rematch 0.83% : 0.000006s : 30: predicate.less_batch_normalization 1.67% : 0.000011s : 78: predicate.list_to_tuple_eliminator_ 2.53% : 0.000017s : 126: predicate.load_eliminater 0.92% : 0.000006s : 15: predicate.loop_unroll_after_grad 1.74% : 0.000012s : 83: predicate.loop_unroll_before_grad 1.81% : 0.000012s : 78: predicate.make_slice_get_slice_eliminator 0.67% : 0.000004s : 30: predicate.merge_addn 0.60% : 0.000004s : 30: predicate.micro_step_allgather_replace 0.62% : 0.000004s : 30: predicate.mini_step_allgather_replace 0.94% : 0.000006s : 48: predicate.minmaximum_grad 1.12% : 0.000007s : 15: predicate.mutable_eliminate 0.36% : 0.000002s : 15: predicate.opt_reshape 0.41% : 0.000003s : 15: predicate.parallel_virtual_node 1.72% : 0.000011s : 57: predicate.partial_defer_inline 1.41% : 0.000009s : 63: predicate.partial_eliminate 0.98% : 0.000006s : 48: predicate.print_const_string_wrapper 0.65% : 0.000004s : 30: predicate.reduce_all_const_elim 1.16% : 0.000008s : 48: predicate.reduce_eliminate 2.47% : 0.000016s : 126: predicate.redundant_stop_gradient_eliminater 0.40% : 0.000003s : 30: predicate.remove_not_recompute_node 1.07% : 0.000007s : 78: predicate.replace_applicator 0.44% : 0.000003s : 30: predicate.replace_old_param 0.26% : 0.000002s : 15: predicate.reset_defer_inline 1.10% : 0.000007s : 48: predicate.reshape_eliminate 0.69% : 0.000005s : 30: predicate.row_tensor_add_zeros_like 0.40% : 0.000003s : 15: predicate.row_tensor_eliminate 0.74% : 0.000005s : 30: predicate.same_eliminate 0.44% : 0.000003s : 36: predicate.set_cell_output_no_recompute 0.91% : 0.000006s : 30: predicate.shard_identity_eliminate 0.70% : 0.000005s : 30: predicate.special_op_eliminate 0.75% : 0.000005s : 30: predicate.specialize_transform 0.89% : 0.000006s : 30: predicate.split_environ_get_set_with_tuple_value 0.90% : 0.000006s : 30: predicate.stack_unstack_eliminate 0.36% : 0.000002s : 15: predicate.switch_call_monad_eliminater 1.24% : 0.000008s : 57: predicate.switch_defer_inline 1.87% : 0.000012s : 87: predicate.switch_layer_defer_inline 4.35% : 0.000029s : 189: predicate.switch_simplify 0.93% : 0.000006s : 48: predicate.tile_eliminate 0.97% : 0.000006s : 48: predicate.transpose_eliminate 1.74% : 0.000012s : 78: predicate.tuple_list_convert_item_index_to_positive 1.71% : 0.000011s : 78: predicate.tuple_list_get_item_const_eliminator 1.67% : 0.000011s : 78: predicate.tuple_list_get_item_depend_reorder 3.00% : 0.000020s : 108: predicate.tuple_list_get_item_eliminator 1.67% : 0.000011s : 78: predicate.tuple_list_get_set_item_eliminator 2.41% : 0.000016s : 108: predicate.tuple_list_set_item_eliminator 1.65% : 0.000011s : 78: predicate.tuple_to_list_eliminator_ 2.57% : 0.000017s : 126: predicate.updatestate_pure_node_eliminater 3.35% : 0.000022s : 156: predicate.updatestate_useless_node_eliminater 0.35% : 0.000002s : 15: predicate.value_based_eliminate 0.75% : 0.000005s : 30: predicate.virtual_dataset_eliminate 0.79% : 0.000005s : 30: predicate.virtual_output_eliminate 0.33% : 0.000002s : 15: predicate.virtual_view_grad_eliminate 0.41% : 0.000003s : 15: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000975 13 33.83% : 0.000330s : 2: func_graph_cloner_run.FuncGraphClonerGraph 66.17% : 0.000645s : 11: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.402393 192 0.00% : 0.000004s : 1: ForceFp32Comm 11.92% : 0.047959s : 1: add_attr 11.91% : 0.047941s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.03% : 0.000135s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.09% : 0.000361s : 1: auto_monad 0.02% : 0.000078s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.28% : 0.001121s : 1: bootstrap 0.01% : 0.000045s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000038s : 1: control_data_broadcast_order 0.01% : 0.000027s : 1: convert_after_rewriter 0.01% : 0.000059s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.01% : 0.000044s : 1: environ_conv 0.03% : 0.000128s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000013s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000012s : 1: label_micro_interleaved_index 0.15% : 0.000592s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.20% : 0.000789s : 1: mutable_eliminate 0.00% : 0.000012s : 1: offloading_packed_experts 0.01% : 0.000032s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000037s : 1: opt.transform.mutable_eliminate 0.90% : 0.003617s : 78: opt.transform.opt_a 0.03% : 0.000108s : 1: opt.transform.opt_after_cconv 0.02% : 0.000067s : 1: opt.transform.opt_after_jit_grad 0.09% : 0.000359s : 28: opt.transform.opt_b 0.03% : 0.000137s : 2: opt.transform.opt_trans_graph 0.02% : 0.000100s : 4: opt.transform.symbol_engine_opt 1.82% : 0.007320s : 1: opt_a 0.06% : 0.000258s : 1: opt_after_cconv 0.22% : 0.000868s : 1: opt_after_jit_grad 0.14% : 0.000553s : 1: opt_b 2.78% : 0.011188s : 1: optimize 0.01% : 0.000040s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000015s : 1: order_py_execute_after_rewriter 0.02% : 0.000068s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000013s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.01% : 0.000031s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000013s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.02% : 0.000087s : 1: pre_auto_parallel 0.01% : 0.000052s : 1: py_interpret_to_execute 0.01% : 0.000037s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.01% : 0.000032s : 1: remove_dup_value 0.27% : 0.001076s : 1: renormalize.infer 0.25% : 0.000991s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000011s : 1: rewriter_after_jit_bprop_graph 0.06% : 0.000258s : 1: rewriter_after_opt_a 0.04% : 0.000174s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000018s : 1: swap_dp_allreduce_reducescatter 0.05% : 0.000189s : 1: symbol_engine_optimizer 0.04% : 0.000174s : 1: tuple_transform 68.33% : 0.274940s : 1: type_inference TotalTime = 0.638517, [21] [bootstrap]: 0.00090105 [type_inference]: 0.399461 [event_method]: 0.00014581 [auto_monad]: 0.00034465 [graph_reusing]: 1.031e-05 [inline]: 3.53999e-06 [add_attr]: 0.016749, [1] [add_attr_with_inline]: 0.0167323, [1] [Cycle 1]: 0.00010952, [2] [tag_attr]: 4.835e-05 [meta_addattr_fg_expand]: 1.008e-05 [parallel-infer-symbol]: 4.05998e-06 [pre_auto_parallel]: 0.00014602 [insert-virtual-dataset]: 2.99001e-06 [parallel-infer-symbol-second]: 1.04e-06 [dataset_repeat_opt]: 3.44001e-06 [pipeline_split]: 1.82999e-06 [optimize]: 0.219809, [53] [py_interpret_to_execute]: 6.226e-05 [rewriter_before_opt_a]: 0.00016239 [opt_a]: 0.216153, [2] [Cycle 1]: 0.214518, [45] [expand_dump_flag]: 5.12e-06 [switch_simplify]: 0.00017015 [loop_unroll]: 5.765e-05 [a_1]: 0.210347 [with_stream_mark]: 4.233e-05 [recompute_prepare]: 2.909e-05 [updatestate_depend_eliminate]: 7.327e-05 [updatestate_assign_eliminate]: 1.175e-05 [updatestate_loads_eliminate]: 1.257e-05 [parameter_eliminate]: 2.76999e-06 [a_2]: 0.00025247 [accelerated_algorithm]: 3.983e-05 [shard]: 2.35002e-06 [meta_shard_fg_expand]: 5.89999e-06 [shard_inline]: 1.662e-05 [merge_send_recv]: 1.445e-05 [auto_parallel]: 2.663e-05 [parallel]: 0.00023605 [flash_sp]: 1.478e-05 [merge_comm]: 1.037e-05 [allreduce_fusion]: 8.62e-06 [matmul_add_comm_reduction]: 1.918e-05 [allreduce_slice_to_reducescatter]: 9.39996e-07 [virtual_shard_identity]: 2.332e-05 [virtual_dataset]: 1.676e-05 [get_grad_eliminate_]: 1.597e-05 [virtual_output]: 1.623e-05 [merge_forward]: 9.44998e-06 [cell_reuse_recompute_pass]: 1.30999e-06 [offload_activation]: 1.9e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.289e-05 [merge_recompute_call_nodes]: 1.86003e-06 [before_grad]: 2.824e-05 [set_forward_comm_id_for_comm_node_pass]: 9.47999e-06 [meta_fg_expand]: 8.74e-06 [flash_sp_send_recv_attached]: 5.04e-06 [receive_attached]: 2.44999e-06 [after_resolve]: 2.56e-05 [a_after_grad]: 2.703e-05 [renormalize]: 0.00220757 [add_forward_monad_depend]: 1.029e-05 [auto_monad_grad]: 2.74001e-06 [auto_monad_eliminator]: 3.414e-05 [cse]: 0.00015396 [a_3]: 0.00012971 [Cycle 2]: 0.00161777, [45] [expand_dump_flag]: 3.35998e-06 [switch_simplify]: 1.896e-05 [loop_unroll]: 1.641e-05 [a_1]: 0.00052518 [with_stream_mark]: 2.217e-05 [recompute_prepare]: 1.717e-05 [updatestate_depend_eliminate]: 9.51003e-06 [updatestate_assign_eliminate]: 1.059e-05 [updatestate_loads_eliminate]: 1.092e-05 [parameter_eliminate]: 2.17999e-06 [a_2]: 0.00023391 [accelerated_algorithm]: 2.134e-05 [shard]: 1.87999e-06 [meta_shard_fg_expand]: 3.82998e-06 [shard_inline]: 1.637e-05 [merge_send_recv]: 1.479e-05 [auto_parallel]: 1.498e-05 [parallel]: 9.02e-06 [flash_sp]: 3.71999e-06 [merge_comm]: 1.002e-05 [allreduce_fusion]: 1.1e-05 [matmul_add_comm_reduction]: 1.726e-05 [allreduce_slice_to_reducescatter]: 6.69999e-07 [virtual_shard_identity]: 1.724e-05 [virtual_dataset]: 1.537e-05 [get_grad_eliminate_]: 1.691e-05 [virtual_output]: 1.534e-05 [merge_forward]: 9.77999e-06 [cell_reuse_recompute_pass]: 2.29001e-06 [offload_activation]: 1.782e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.962e-05 [merge_recompute_call_nodes]: 1.30999e-06 [before_grad]: 2.657e-05 [set_forward_comm_id_for_comm_node_pass]: 8.94998e-06 [meta_fg_expand]: 7.7e-06 [flash_sp_send_recv_attached]: 2.04999e-06 [receive_attached]: 2.46998e-06 [after_resolve]: 2.468e-05 [a_after_grad]: 2.594e-05 [renormalize]: 7.00238e-08 [add_forward_monad_depend]: 1.91e-06 [auto_monad_grad]: 1.82999e-06 [auto_monad_eliminator]: 2.159e-05 [cse]: 4.526e-05 [a_3]: 0.0001122 [py_interpret_to_execute_after_opt_a]: 2.817e-05 [slice_cell_reuse_recomputed_activation]: 2.15002e-06 [rewriter_after_opt_a]: 0.00024062 [convert_after_rewriter]: 1.634e-05 [order_py_execute_after_rewriter]: 1.157e-05 [mutable_eliminate]: 0.00082565 [opt_b]: 0.0005413, [1] [Cycle 1]: 0.00053257, [7] [b_1]: 0.00037293 [b_2]: 1.894e-05 [updatestate_depend_eliminate]: 1.526e-05 [updatestate_assign_eliminate]: 1.001e-05 [updatestate_loads_eliminate]: 1.253e-05 [renormalize]: 8.80013e-07 [cse]: 6.237e-05 [optimize_parallel_all_gather_comm]: 3.314e-05 [overlap_param_gather]: 1.94e-06 [cconv]: 3.7e-05 [loop_unroll]: 0.00052992 [opt_after_cconv]: 0.00023747, [1] [Cycle 1]: 0.00023027, [7] [c_1]: 0.00011696 [parameter_eliminate]: 5.19e-06 [updatestate_depend_eliminate]: 1.08e-05 [updatestate_assign_eliminate]: 8.3e-06 [updatestate_loads_eliminate]: 8.99e-06 [cse]: 4.311e-05 [renormalize]: 9.79984e-07 [remove_dup_value]: 2.81e-05 [tuple_transform]: 0.0001632, [1] [Cycle 1]: 0.00015785, [4] [d_1]: 0.00012105 [none_parameter_eliminate]: 1.85001e-06 [renormalize]: 1.69995e-07 [switch_simplify]: 1.75e-05 [partial_unused_args_eliminate]: 1.92999e-06 [add_recomputation]: 0.00010962 [cse_after_recomputation]: 4.632e-05, [1] [Cycle 1]: 4.115e-05, [1] [cse]: 3.485e-05 [environ_conv]: 1.623e-05 [swap_dp_allreduce_reducescatter]: 1.277e-05 [bias_add_comm_swap]: 2.93e-06 [label_micro_interleaved_index]: 5.26002e-06 [label_fine_grained_interleaved_index]: 2.68e-06 [merge_cast_opt]: 1.60999e-06 [slice_recompute_activation]: 2.34999e-06 [micro_interleaved_order_control]: 2.06e-06 [assign_add_opt]: 1.55999e-06 [ForceFp32Comm]: 9.20001e-07 [remove_cast_before_assign_add]: 1.17e-06 [full_micro_interleaved_order_control]: 2.32999e-06 [reorder_send_recv_between_fp_bp]: 2.88998e-06 [comm_op_add_attrs]: 1.32e-06 [add_comm_op_reuse_tag]: 1.35001e-06 [interleave_split_concat_branches]: 1.15999e-06 [interleave_parallel_branches]: 1.08001e-06 [overlap_opt_shard_in_pipeline]: 1.24e-06 [overlap_opt_shard_grad_in_pipeline]: 1.72001e-06 [control_data_broadcast_order]: 2.75e-05 [grouped_pairwise_exchange_alltoall]: 1.71e-06 [offloading_packed_experts]: 8.22998e-06 [overlap_recompute_and_grad_model_parallel]: 8.60999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.40001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.52999e-06 [overlap_recompute_comm]: 2.48e-06 [overlap_grad_ring_attention]: 8.08999e-06 [overlap_grad_flash_sp]: 4.59e-05 [begin_end_overlap_inline]: 5.69999e-07 [split_matmul_comm_elemetwise]: 2.52001e-06 [split_layernorm_comm]: 1.77999e-06 [handle_group_info]: 1.21997e-06 [symbol_engine_optimizer]: 0.00015449, [1] [Cycle 1]: 0.00014843, [6] [build]: 1.582e-05 [elim_shapecalc]: 2.486e-05 [elim_not_effective]: 2.998e-05 [opt_reshape]: 1.924e-05 [fold_const_symbol]: 2.628e-05 [renormalize]: 2.30008e-07 [detach_backward]: 2.29001e-06 [pipeline_parallel_scheduler]: 1.74e-06 [auto_monad_reorder]: 5.256e-05 [get_jit_bprop_graph]: 2.26e-06 [rewriter_after_jit_bprop_graph]: 5.97001e-06 [opt_after_jit_grad]: 0.00054995 [validate]: 0.0001037 Sums bootstrap : 0.000901s : 0.15% type_inference : 0.399461s : 64.36% event_method : 0.000146s : 0.02% auto_monad : 0.000345s : 0.06% graph_reusing : 0.000010s : 0.00% inline : 0.000004s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000048s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000010s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000146s : 0.02% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000062s : 0.01% optimize.rewriter_before_opt_a : 0.000162s : 0.03% optimize.opt_a.expand_dump_flag : 0.000008s : 0.00% optimize.opt_a.switch_simplify : 0.000189s : 0.03% optimize.opt_a.loop_unroll : 0.000074s : 0.01% optimize.opt_a.a_1 : 0.210872s : 33.98% optimize.opt_a.with_stream_mark : 0.000064s : 0.01% optimize.opt_a.recompute_prepare : 0.000046s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000083s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000022s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000023s : 0.00% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000486s : 0.08% optimize.opt_a.accelerated_algorithm : 0.000061s : 0.01% optimize.opt_a.shard : 0.000004s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000010s : 0.00% optimize.opt_a.shard_inline : 0.000033s : 0.01% optimize.opt_a.merge_send_recv : 0.000029s : 0.00% optimize.opt_a.auto_parallel : 0.000042s : 0.01% optimize.opt_a.parallel : 0.000245s : 0.04% optimize.opt_a.flash_sp : 0.000018s : 0.00% optimize.opt_a.merge_comm : 0.000020s : 0.00% optimize.opt_a.allreduce_fusion : 0.000020s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000036s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000041s : 0.01% optimize.opt_a.virtual_dataset : 0.000032s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000033s : 0.01% optimize.opt_a.virtual_output : 0.000032s : 0.01% optimize.opt_a.merge_forward : 0.000019s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.00% optimize.opt_a.offload_activation : 0.000037s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000063s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000055s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000018s : 0.00% optimize.opt_a.meta_fg_expand : 0.000016s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000007s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000050s : 0.01% optimize.opt_a.a_after_grad : 0.000053s : 0.01% optimize.opt_a.renormalize : 0.002208s : 0.36% optimize.opt_a.add_forward_monad_depend : 0.000012s : 0.00% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000056s : 0.01% optimize.opt_a.cse : 0.000199s : 0.03% optimize.opt_a.a_3 : 0.000242s : 0.04% optimize.py_interpret_to_execute_after_opt_a : 0.000028s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000241s : 0.04% optimize.convert_after_rewriter : 0.000016s : 0.00% optimize.order_py_execute_after_rewriter : 0.000012s : 0.00% optimize.mutable_eliminate : 0.000826s : 0.13% optimize.opt_b.b_1 : 0.000373s : 0.06% optimize.opt_b.b_2 : 0.000019s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000015s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000010s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000013s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000062s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000033s : 0.01% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000037s : 0.01% optimize.loop_unroll : 0.000530s : 0.09% optimize.opt_after_cconv.c_1 : 0.000117s : 0.02% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000011s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000008s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000009s : 0.00% optimize.opt_after_cconv.cse : 0.000043s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000028s : 0.00% optimize.tuple_transform.d_1 : 0.000121s : 0.02% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000018s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000110s : 0.02% optimize.cse_after_recomputation.cse : 0.000035s : 0.01% optimize.environ_conv : 0.000016s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000013s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000028s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000008s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000009s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000008s : 0.00% optimize.overlap_grad_flash_sp : 0.000046s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000016s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000025s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000030s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000019s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000026s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000053s : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.000550s : 0.09% validate : 0.000104s : 0.02% Time group info: ------[substitution.] 0.000687 182 7.27% : 0.000050s : 10: substitution.arithmetic_simplify 1.47% : 0.000010s : 2: substitution.depend_value_elim 0.61% : 0.000004s : 10: substitution.elim_not_effective 0.55% : 0.000004s : 10: substitution.fold_const_symbol 1.83% : 0.000013s : 15: substitution.graph_param_transform 66.66% : 0.000458s : 9: substitution.inline 1.55% : 0.000011s : 20: substitution.j_node_and_user_rematch 3.10% : 0.000021s : 4: substitution.less_batch_normalization 0.96% : 0.000007s : 6: substitution.load_eliminater 2.15% : 0.000015s : 20: substitution.remove_not_recompute_node 1.42% : 0.000010s : 10: substitution.replace_old_param 1.75% : 0.000012s : 2: substitution.switch_simplify 5.01% : 0.000034s : 30: substitution.updatestate_pure_node_eliminater 5.66% : 0.000039s : 34: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.399295 2 99.43% : 0.397023s : 1: type_inference.infer 0.57% : 0.002271s : 1: type_inference.specialize ------[replace.] 0.000154 11 55.87% : 0.000086s : 9: replace.inline 44.13% : 0.000068s : 2: replace.switch_simplify ------[match.] 0.000461 11 97.72% : 0.000451s : 9: match.inline 2.28% : 0.000010s : 2: match.switch_simplify ------[predicate.] 0.000687 4397 1.03% : 0.000007s : 48: predicate.accumulaten_eliminater 0.61% : 0.000004s : 15: predicate.ad_related_special_op_eliminate 0.59% : 0.000004s : 30: predicate.addn_check_dump 1.08% : 0.000007s : 48: predicate.addn_zero_filter 0.97% : 0.000007s : 48: predicate.adjust_all_reduce_mul_add 2.28% : 0.000016s : 78: predicate.arithmetic_simplify 1.16% : 0.000008s : 48: predicate.cast_eliminate 0.71% : 0.000005s : 30: predicate.check_bprop_eliminate 0.64% : 0.000004s : 30: predicate.compare_switch_simplify 0.17% : 0.000001s : 15: predicate.const_output_eliminate 0.59% : 0.000004s : 30: predicate.depend_value_elim 1.19% : 0.000008s : 48: predicate.dict_get_item_const_eliminator 1.12% : 0.000008s : 48: predicate.dict_get_item_eliminator 0.98% : 0.000007s : 48: predicate.dict_set_item_eliminator 0.71% : 0.000005s : 30: predicate.dumpgradient_eliminate 0.19% : 0.000001s : 15: predicate.elim_not_effective 0.38% : 0.000003s : 15: predicate.elim_shapecalc_of_broadcastargs 1.44% : 0.000010s : 63: predicate.environ_add_const_eliminate 1.23% : 0.000008s : 63: predicate.environ_get_add_eliminate 2.17% : 0.000015s : 63: predicate.environ_get_depend_swap 1.85% : 0.000013s : 93: predicate.environ_get_eliminate 1.23% : 0.000008s : 63: predicate.environ_get_set_eliminate 1.20% : 0.000008s : 57: predicate.exchange_switch_depend_value 1.68% : 0.000012s : 57: predicate.float_depend_g_call 0.60% : 0.000004s : 30: predicate.float_environ_get_switch 0.88% : 0.000006s : 45: predicate.float_tuple_getitem_switch 0.18% : 0.000001s : 15: predicate.fold_const_symbol 0.66% : 0.000005s : 30: predicate.get_grad_eliminate 0.21% : 0.000001s : 15: predicate.graph_param_transform 0.57% : 0.000004s : 30: predicate.incorporate_call 0.53% : 0.000004s : 30: predicate.incorporate_call_switch 5.56% : 0.000038s : 195: predicate.inline 0.74% : 0.000005s : 30: predicate.inline_without_move 0.32% : 0.000002s : 30: predicate.j_node_and_user_rematch 0.89% : 0.000006s : 30: predicate.less_batch_normalization 1.75% : 0.000012s : 78: predicate.list_to_tuple_eliminator_ 2.59% : 0.000018s : 126: predicate.load_eliminater 0.70% : 0.000005s : 15: predicate.loop_unroll_after_grad 1.68% : 0.000012s : 83: predicate.loop_unroll_before_grad 1.66% : 0.000011s : 78: predicate.make_slice_get_slice_eliminator 0.68% : 0.000005s : 30: predicate.merge_addn 0.64% : 0.000004s : 30: predicate.micro_step_allgather_replace 0.64% : 0.000004s : 30: predicate.mini_step_allgather_replace 0.99% : 0.000007s : 48: predicate.minmaximum_grad 0.85% : 0.000006s : 15: predicate.mutable_eliminate 0.32% : 0.000002s : 15: predicate.opt_reshape 0.36% : 0.000002s : 15: predicate.parallel_virtual_node 2.02% : 0.000014s : 57: predicate.partial_defer_inline 1.37% : 0.000009s : 63: predicate.partial_eliminate 0.95% : 0.000007s : 48: predicate.print_const_string_wrapper 0.60% : 0.000004s : 30: predicate.reduce_all_const_elim 1.47% : 0.000010s : 48: predicate.reduce_eliminate 2.64% : 0.000018s : 126: predicate.redundant_stop_gradient_eliminater 0.41% : 0.000003s : 30: predicate.remove_not_recompute_node 1.08% : 0.000007s : 78: predicate.replace_applicator 0.39% : 0.000003s : 30: predicate.replace_old_param 0.27% : 0.000002s : 15: predicate.reset_defer_inline 1.05% : 0.000007s : 48: predicate.reshape_eliminate 0.70% : 0.000005s : 30: predicate.row_tensor_add_zeros_like 0.35% : 0.000002s : 15: predicate.row_tensor_eliminate 0.81% : 0.000006s : 30: predicate.same_eliminate 0.46% : 0.000003s : 36: predicate.set_cell_output_no_recompute 0.69% : 0.000005s : 30: predicate.shard_identity_eliminate 0.69% : 0.000005s : 30: predicate.special_op_eliminate 0.70% : 0.000005s : 30: predicate.specialize_transform 0.80% : 0.000005s : 30: predicate.split_environ_get_set_with_tuple_value 0.80% : 0.000006s : 30: predicate.stack_unstack_eliminate 0.33% : 0.000002s : 15: predicate.switch_call_monad_eliminater 1.23% : 0.000008s : 57: predicate.switch_defer_inline 1.93% : 0.000013s : 87: predicate.switch_layer_defer_inline 4.25% : 0.000029s : 189: predicate.switch_simplify 1.04% : 0.000007s : 48: predicate.tile_eliminate 0.96% : 0.000007s : 48: predicate.transpose_eliminate 1.63% : 0.000011s : 78: predicate.tuple_list_convert_item_index_to_positive 1.96% : 0.000013s : 78: predicate.tuple_list_get_item_const_eliminator 1.70% : 0.000012s : 78: predicate.tuple_list_get_item_depend_reorder 2.85% : 0.000020s : 108: predicate.tuple_list_get_item_eliminator 1.65% : 0.000011s : 78: predicate.tuple_list_get_set_item_eliminator 2.66% : 0.000018s : 108: predicate.tuple_list_set_item_eliminator 1.77% : 0.000012s : 78: predicate.tuple_to_list_eliminator_ 2.71% : 0.000019s : 126: predicate.updatestate_pure_node_eliminater 3.53% : 0.000024s : 156: predicate.updatestate_useless_node_eliminater 0.36% : 0.000002s : 15: predicate.value_based_eliminate 0.69% : 0.000005s : 30: predicate.virtual_dataset_eliminate 0.70% : 0.000005s : 30: predicate.virtual_output_eliminate 0.29% : 0.000002s : 15: predicate.virtual_view_grad_eliminate 0.33% : 0.000002s : 15: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.001198 13 37.37% : 0.000448s : 2: func_graph_cloner_run.FuncGraphClonerGraph 62.63% : 0.000750s : 11: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.090134 192 0.00% : 0.000004s : 1: ForceFp32Comm 1.54% : 0.016757s : 1: add_attr 1.54% : 0.016738s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000114s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.03% : 0.000358s : 1: auto_monad 0.01% : 0.000057s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.09% : 0.000935s : 1: bootstrap 0.00% : 0.000041s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000031s : 1: control_data_broadcast_order 0.00% : 0.000021s : 1: convert_after_rewriter 0.00% : 0.000049s : 1: cse_after_recomputation 0.00% : 0.000007s : 1: dataset_repeat_opt 0.00% : 0.000005s : 1: detach_backward 0.00% : 0.000020s : 1: environ_conv 0.01% : 0.000159s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000015s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000008s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.05% : 0.000540s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.08% : 0.000840s : 1: mutable_eliminate 0.00% : 0.000011s : 1: offloading_packed_experts 0.00% : 0.000027s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000033s : 1: opt.transform.mutable_eliminate 19.47% : 0.212254s : 78: opt.transform.opt_a 0.01% : 0.000115s : 1: opt.transform.opt_after_cconv 0.01% : 0.000056s : 1: opt.transform.opt_after_jit_grad 0.03% : 0.000359s : 28: opt.transform.opt_b 0.01% : 0.000136s : 2: opt.transform.opt_trans_graph 0.01% : 0.000096s : 4: opt.transform.symbol_engine_opt 19.83% : 0.216157s : 1: opt_a 0.02% : 0.000241s : 1: opt_after_cconv 0.05% : 0.000560s : 1: opt_after_jit_grad 0.05% : 0.000545s : 1: opt_b 20.16% : 0.219815s : 1: optimize 0.00% : 0.000037s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000015s : 1: order_py_execute_after_rewriter 0.00% : 0.000050s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000012s : 1: overlap_grad_ring_attention 0.00% : 0.000004s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000012s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000006s : 1: pipeline_split 0.01% : 0.000153s : 1: pre_auto_parallel 0.01% : 0.000067s : 1: py_interpret_to_execute 0.00% : 0.000033s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000032s : 1: remove_dup_value 0.11% : 0.001251s : 1: renormalize.infer 0.09% : 0.000943s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.02% : 0.000248s : 1: rewriter_after_opt_a 0.02% : 0.000168s : 1: rewriter_before_opt_a 0.00% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000016s : 1: swap_dp_allreduce_reducescatter 0.01% : 0.000157s : 1: symbol_engine_optimizer 0.02% : 0.000166s : 1: tuple_transform 36.65% : 0.399497s : 1: type_inference [WARNING] CORE(166799,ffff961eff30,python3.9):2026-01-29-17:41:37.591.504 [mindspore/core/utils/convert_utils.cc:923] GetConversionType] For operators with side effects, there is an implicit type conversion between Float16 and BFloat16, which may result in loss of precision. It is recommended to use Float32. TotalTime = 0.15563, [21] [bootstrap]: 0.00054374 [type_inference]: 0.138686 [event_method]: 2.397e-05 [auto_monad]: 0.00031186 [graph_reusing]: 7.13e-06 [inline]: 3.04999e-06 [add_attr]: 0.00427204, [1] [add_attr_with_inline]: 0.00425958, [1] [Cycle 1]: 8.786e-05, [2] [tag_attr]: 3.439e-05 [meta_addattr_fg_expand]: 7.74002e-06 [parallel-infer-symbol]: 4.1e-06 [pre_auto_parallel]: 5.17e-05 [insert-virtual-dataset]: 2.89001e-06 [parallel-infer-symbol-second]: 7.60017e-07 [dataset_repeat_opt]: 2.44001e-06 [pipeline_split]: 2.17999e-06 [optimize]: 0.010592, [53] [py_interpret_to_execute]: 3.951e-05 [rewriter_before_opt_a]: 0.00012006 [opt_a]: 0.00662266, [2] [Cycle 1]: 0.00473991, [45] [expand_dump_flag]: 3.83999e-06 [switch_simplify]: 5.724e-05 [loop_unroll]: 4.298e-05 [a_1]: 0.00124456 [with_stream_mark]: 3.192e-05 [recompute_prepare]: 2.742e-05 [updatestate_depend_eliminate]: 8e-05 [updatestate_assign_eliminate]: 1.554e-05 [updatestate_loads_eliminate]: 1.536e-05 [parameter_eliminate]: 2.40002e-06 [a_2]: 0.00028204 [accelerated_algorithm]: 4.326e-05 [shard]: 2.63e-06 [meta_shard_fg_expand]: 4.33001e-06 [shard_inline]: 1.92e-05 [merge_send_recv]: 1.702e-05 [auto_parallel]: 1.581e-05 [parallel]: 2.841e-05 [flash_sp]: 1.182e-05 [merge_comm]: 1.262e-05 [allreduce_fusion]: 1.158e-05 [matmul_add_comm_reduction]: 2.053e-05 [allreduce_slice_to_reducescatter]: 8.90024e-07 [virtual_shard_identity]: 2.113e-05 [virtual_dataset]: 1.829e-05 [get_grad_eliminate_]: 1.87e-05 [virtual_output]: 1.862e-05 [merge_forward]: 1.216e-05 [cell_reuse_recompute_pass]: 1.15999e-06 [offload_activation]: 2.117e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.733e-05 [merge_recompute_call_nodes]: 1.55999e-06 [before_grad]: 3.371e-05 [set_forward_comm_id_for_comm_node_pass]: 1.235e-05 [meta_fg_expand]: 8.79e-06 [flash_sp_send_recv_attached]: 6.01e-06 [receive_attached]: 2.98e-06 [after_resolve]: 2.617e-05 [a_after_grad]: 3.098e-05 [renormalize]: 0.00186674 [add_forward_monad_depend]: 1.021e-05 [auto_monad_grad]: 2.98e-06 [auto_monad_eliminator]: 4.366e-05 [cse]: 0.00011864 [a_3]: 0.00014749 [Cycle 2]: 0.001869, [45] [expand_dump_flag]: 2.91999e-06 [switch_simplify]: 2.144e-05 [loop_unroll]: 1.828e-05 [a_1]: 0.0005924 [with_stream_mark]: 2.606e-05 [recompute_prepare]: 1.898e-05 [updatestate_depend_eliminate]: 1.201e-05 [updatestate_assign_eliminate]: 1.465e-05 [updatestate_loads_eliminate]: 1.477e-05 [parameter_eliminate]: 2.78e-06 [a_2]: 0.00027428 [accelerated_algorithm]: 2.502e-05 [shard]: 2.46e-06 [meta_shard_fg_expand]: 5.01997e-06 [shard_inline]: 1.869e-05 [merge_send_recv]: 1.798e-05 [auto_parallel]: 1.802e-05 [parallel]: 1.061e-05 [flash_sp]: 4.53001e-06 [merge_comm]: 1.262e-05 [allreduce_fusion]: 1.121e-05 [matmul_add_comm_reduction]: 1.921e-05 [allreduce_slice_to_reducescatter]: 6.40022e-07 [virtual_shard_identity]: 2.208e-05 [virtual_dataset]: 1.815e-05 [get_grad_eliminate_]: 1.8e-05 [virtual_output]: 1.798e-05 [merge_forward]: 1.158e-05 [cell_reuse_recompute_pass]: 2.61e-06 [offload_activation]: 2.336e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.702e-05 [merge_recompute_call_nodes]: 1.77001e-06 [before_grad]: 3.336e-05 [set_forward_comm_id_for_comm_node_pass]: 1.291e-05 [meta_fg_expand]: 9.05999e-06 [flash_sp_send_recv_attached]: 1.81998e-06 [receive_attached]: 2.34999e-06 [after_resolve]: 2.704e-05 [a_after_grad]: 3.04e-05 [renormalize]: 2.69996e-07 [add_forward_monad_depend]: 3.04999e-06 [auto_monad_grad]: 2.43e-06 [auto_monad_eliminator]: 3.113e-05 [cse]: 5.986e-05 [a_3]: 0.00012672 [py_interpret_to_execute_after_opt_a]: 3.283e-05 [slice_cell_reuse_recomputed_activation]: 2.91999e-06 [rewriter_after_opt_a]: 0.00031716 [convert_after_rewriter]: 2.065e-05 [order_py_execute_after_rewriter]: 1.283e-05 [mutable_eliminate]: 0.00084048 [opt_b]: 0.00063626, [1] [Cycle 1]: 0.00062743, [7] [b_1]: 0.00044903 [b_2]: 2.208e-05 [updatestate_depend_eliminate]: 1.795e-05 [updatestate_assign_eliminate]: 1.465e-05 [updatestate_loads_eliminate]: 1.573e-05 [renormalize]: 5.69999e-07 [cse]: 6.673e-05 [optimize_parallel_all_gather_comm]: 3.944e-05 [overlap_param_gather]: 2.07001e-06 [cconv]: 4.267e-05 [loop_unroll]: 0.00052087 [opt_after_cconv]: 0.00026945, [1] [Cycle 1]: 0.00026244, [7] [c_1]: 0.00012776 [parameter_eliminate]: 5.49e-06 [updatestate_depend_eliminate]: 1.497e-05 [updatestate_assign_eliminate]: 1.266e-05 [updatestate_loads_eliminate]: 1.321e-05 [cse]: 5.193e-05 [renormalize]: 3.59985e-07 [remove_dup_value]: 6.776e-05 [tuple_transform]: 0.00018864, [1] [Cycle 1]: 0.00018323, [4] [d_1]: 0.00014218 [none_parameter_eliminate]: 2.68e-06 [renormalize]: 1.50001e-07 [switch_simplify]: 1.93e-05 [partial_unused_args_eliminate]: 2.35002e-06 [add_recomputation]: 0.00013695 [cse_after_recomputation]: 5.36e-05, [1] [Cycle 1]: 4.813e-05, [1] [cse]: 4.249e-05 [environ_conv]: 1.913e-05 [swap_dp_allreduce_reducescatter]: 1.617e-05 [bias_add_comm_swap]: 2.74999e-06 [label_micro_interleaved_index]: 6.57002e-06 [label_fine_grained_interleaved_index]: 3.02002e-06 [merge_cast_opt]: 1.44e-06 [slice_recompute_activation]: 2.36998e-06 [micro_interleaved_order_control]: 2.69999e-06 [assign_add_opt]: 1.69e-06 [ForceFp32Comm]: 8.90024e-07 [remove_cast_before_assign_add]: 1.07998e-06 [full_micro_interleaved_order_control]: 2.36e-06 [reorder_send_recv_between_fp_bp]: 2.61e-06 [comm_op_add_attrs]: 1.65001e-06 [add_comm_op_reuse_tag]: 1.05001e-06 [interleave_split_concat_branches]: 1.17999e-06 [interleave_parallel_branches]: 1.07e-06 [overlap_opt_shard_in_pipeline]: 1.44e-06 [overlap_opt_shard_grad_in_pipeline]: 2.22999e-06 [control_data_broadcast_order]: 3.397e-05 [grouped_pairwise_exchange_alltoall]: 1.77001e-06 [offloading_packed_experts]: 9.84999e-06 [overlap_recompute_and_grad_model_parallel]: 9.42001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.17e-06 [overlap_recompute_allgather_and_fa_grad]: 1.54e-06 [overlap_recompute_comm]: 2.79999e-06 [overlap_grad_ring_attention]: 9.05999e-06 [overlap_grad_flash_sp]: 5.201e-05 [begin_end_overlap_inline]: 5.69999e-07 [split_matmul_comm_elemetwise]: 2.59999e-06 [split_layernorm_comm]: 1.89e-06 [handle_group_info]: 1.12999e-06 [symbol_engine_optimizer]: 0.00016864, [1] [Cycle 1]: 0.00016323, [6] [build]: 1.406e-05 [elim_shapecalc]: 2.574e-05 [elim_not_effective]: 3.669e-05 [opt_reshape]: 2.218e-05 [fold_const_symbol]: 3.32e-05 [renormalize]: 2.00002e-07 [detach_backward]: 2.48998e-06 [pipeline_parallel_scheduler]: 1.54e-06 [auto_monad_reorder]: 7.86e-05 [get_jit_bprop_graph]: 2.97002e-06 [rewriter_after_jit_bprop_graph]: 6.89001e-06 [opt_after_jit_grad]: 0.00072799 [validate]: 0.00010149 Sums bootstrap : 0.000544s : 0.36% type_inference : 0.138686s : 92.31% event_method : 0.000024s : 0.02% auto_monad : 0.000312s : 0.21% graph_reusing : 0.000007s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000034s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000008s : 0.01% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000052s : 0.03% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000040s : 0.03% optimize.rewriter_before_opt_a : 0.000120s : 0.08% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000079s : 0.05% optimize.opt_a.loop_unroll : 0.000061s : 0.04% optimize.opt_a.a_1 : 0.001837s : 1.22% optimize.opt_a.with_stream_mark : 0.000058s : 0.04% optimize.opt_a.recompute_prepare : 0.000046s : 0.03% optimize.opt_a.updatestate_depend_eliminate : 0.000092s : 0.06% optimize.opt_a.updatestate_assign_eliminate : 0.000030s : 0.02% optimize.opt_a.updatestate_loads_eliminate : 0.000030s : 0.02% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000556s : 0.37% optimize.opt_a.accelerated_algorithm : 0.000068s : 0.05% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000009s : 0.01% optimize.opt_a.shard_inline : 0.000038s : 0.03% optimize.opt_a.merge_send_recv : 0.000035s : 0.02% optimize.opt_a.auto_parallel : 0.000034s : 0.02% optimize.opt_a.parallel : 0.000039s : 0.03% optimize.opt_a.flash_sp : 0.000016s : 0.01% optimize.opt_a.merge_comm : 0.000025s : 0.02% optimize.opt_a.allreduce_fusion : 0.000023s : 0.02% optimize.opt_a.matmul_add_comm_reduction : 0.000040s : 0.03% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000043s : 0.03% optimize.opt_a.virtual_dataset : 0.000036s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000037s : 0.02% optimize.opt_a.virtual_output : 0.000037s : 0.02% optimize.opt_a.merge_forward : 0.000024s : 0.02% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.00% optimize.opt_a.offload_activation : 0.000045s : 0.03% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000074s : 0.05% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000067s : 0.04% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000025s : 0.02% optimize.opt_a.meta_fg_expand : 0.000018s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000008s : 0.01% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000053s : 0.04% optimize.opt_a.a_after_grad : 0.000061s : 0.04% optimize.opt_a.renormalize : 0.001867s : 1.24% optimize.opt_a.add_forward_monad_depend : 0.000013s : 0.01% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000075s : 0.05% optimize.opt_a.cse : 0.000178s : 0.12% optimize.opt_a.a_3 : 0.000274s : 0.18% optimize.py_interpret_to_execute_after_opt_a : 0.000033s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000317s : 0.21% optimize.convert_after_rewriter : 0.000021s : 0.01% optimize.order_py_execute_after_rewriter : 0.000013s : 0.01% optimize.mutable_eliminate : 0.000840s : 0.56% optimize.opt_b.b_1 : 0.000449s : 0.30% optimize.opt_b.b_2 : 0.000022s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000018s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000015s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000016s : 0.01% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000067s : 0.04% optimize.optimize_parallel_all_gather_comm : 0.000039s : 0.03% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000043s : 0.03% optimize.loop_unroll : 0.000521s : 0.35% optimize.opt_after_cconv.c_1 : 0.000128s : 0.09% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000015s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000013s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000013s : 0.01% optimize.opt_after_cconv.cse : 0.000052s : 0.03% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000068s : 0.05% optimize.tuple_transform.d_1 : 0.000142s : 0.09% optimize.tuple_transform.none_parameter_eliminate : 0.000003s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000019s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000137s : 0.09% optimize.cse_after_recomputation.cse : 0.000042s : 0.03% optimize.environ_conv : 0.000019s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000016s : 0.01% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000007s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000002s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000034s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000010s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000009s : 0.01% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000009s : 0.01% optimize.overlap_grad_flash_sp : 0.000052s : 0.03% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000014s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000026s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000037s : 0.02% optimize.symbol_engine_optimizer.opt_reshape : 0.000022s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000033s : 0.02% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000079s : 0.05% get_jit_bprop_graph : 0.000003s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.00% opt_after_jit_grad : 0.000728s : 0.48% validate : 0.000101s : 0.07% Time group info: ------[substitution.] 0.000464 203 7.41% : 0.000034s : 4: substitution.cast_eliminate 2.30% : 0.000011s : 2: substitution.depend_value_elim 1.26% : 0.000006s : 14: substitution.elim_not_effective 1.05% : 0.000005s : 14: substitution.fold_const_symbol 3.22% : 0.000015s : 17: substitution.graph_param_transform 52.40% : 0.000243s : 4: substitution.inline 2.64% : 0.000012s : 28: substitution.j_node_and_user_rematch 4.94% : 0.000023s : 2: substitution.less_batch_normalization 2.06% : 0.000010s : 12: substitution.load_eliminater 3.78% : 0.000018s : 28: substitution.remove_not_recompute_node 1.79% : 0.000008s : 6: substitution.replace_old_param 8.12% : 0.000038s : 32: substitution.updatestate_pure_node_eliminater 9.01% : 0.000042s : 40: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.138602 2 99.25% : 0.137557s : 1: type_inference.infer 0.75% : 0.001045s : 1: type_inference.specialize ------[replace.] 0.000046 4 100.00% : 0.000046s : 4: replace.inline ------[match.] 0.000240 4 100.00% : 0.000240s : 4: match.inline ------[predicate.] 0.000644 4350 0.93% : 0.000006s : 43: predicate.accumulaten_eliminater 0.97% : 0.000006s : 17: predicate.ad_related_special_op_eliminate 0.72% : 0.000005s : 34: predicate.addn_check_dump 1.04% : 0.000007s : 43: predicate.addn_zero_filter 0.93% : 0.000006s : 43: predicate.adjust_all_reduce_mul_add 1.95% : 0.000013s : 77: predicate.arithmetic_simplify 0.91% : 0.000006s : 43: predicate.cast_eliminate 0.79% : 0.000005s : 34: predicate.check_bprop_eliminate 0.71% : 0.000005s : 34: predicate.compare_switch_simplify 0.19% : 0.000001s : 17: predicate.const_output_eliminate 0.73% : 0.000005s : 34: predicate.depend_value_elim 0.97% : 0.000006s : 43: predicate.dict_get_item_const_eliminator 1.00% : 0.000006s : 43: predicate.dict_get_item_eliminator 0.94% : 0.000006s : 43: predicate.dict_set_item_eliminator 0.91% : 0.000006s : 34: predicate.dumpgradient_eliminate 0.25% : 0.000002s : 17: predicate.elim_not_effective 0.43% : 0.000003s : 17: predicate.elim_shapecalc_of_broadcastargs 1.31% : 0.000008s : 60: predicate.environ_add_const_eliminate 1.21% : 0.000008s : 60: predicate.environ_get_add_eliminate 1.26% : 0.000008s : 60: predicate.environ_get_depend_swap 2.00% : 0.000013s : 94: predicate.environ_get_eliminate 1.25% : 0.000008s : 60: predicate.environ_get_set_eliminate 1.00% : 0.000006s : 47: predicate.exchange_switch_depend_value 1.42% : 0.000009s : 47: predicate.float_depend_g_call 0.72% : 0.000005s : 34: predicate.float_environ_get_switch 1.11% : 0.000007s : 51: predicate.float_tuple_getitem_switch 0.20% : 0.000001s : 17: predicate.fold_const_symbol 0.82% : 0.000005s : 34: predicate.get_grad_eliminate 0.29% : 0.000002s : 17: predicate.graph_param_transform 0.71% : 0.000005s : 34: predicate.incorporate_call 0.68% : 0.000004s : 34: predicate.incorporate_call_switch 5.52% : 0.000036s : 192: predicate.inline 0.97% : 0.000006s : 34: predicate.inline_without_move 0.36% : 0.000002s : 34: predicate.j_node_and_user_rematch 1.10% : 0.000007s : 34: predicate.less_batch_normalization 1.82% : 0.000012s : 77: predicate.list_to_tuple_eliminator_ 2.54% : 0.000016s : 120: predicate.load_eliminater 0.77% : 0.000005s : 17: predicate.loop_unroll_after_grad 1.38% : 0.000009s : 64: predicate.loop_unroll_before_grad 1.78% : 0.000011s : 77: predicate.make_slice_get_slice_eliminator 0.76% : 0.000005s : 34: predicate.merge_addn 0.71% : 0.000005s : 34: predicate.micro_step_allgather_replace 0.80% : 0.000005s : 34: predicate.mini_step_allgather_replace 0.88% : 0.000006s : 43: predicate.minmaximum_grad 0.87% : 0.000006s : 17: predicate.mutable_eliminate 0.42% : 0.000003s : 17: predicate.opt_reshape 0.42% : 0.000003s : 17: predicate.parallel_virtual_node 1.31% : 0.000008s : 47: predicate.partial_defer_inline 1.42% : 0.000009s : 60: predicate.partial_eliminate 0.89% : 0.000006s : 43: predicate.print_const_string_wrapper 0.74% : 0.000005s : 34: predicate.reduce_all_const_elim 1.13% : 0.000007s : 43: predicate.reduce_eliminate 2.48% : 0.000016s : 120: predicate.redundant_stop_gradient_eliminater 0.41% : 0.000003s : 34: predicate.remove_not_recompute_node 1.17% : 0.000008s : 77: predicate.replace_applicator 0.51% : 0.000003s : 34: predicate.replace_old_param 0.29% : 0.000002s : 17: predicate.reset_defer_inline 0.98% : 0.000006s : 43: predicate.reshape_eliminate 0.76% : 0.000005s : 34: predicate.row_tensor_add_zeros_like 0.46% : 0.000003s : 17: predicate.row_tensor_eliminate 0.81% : 0.000005s : 34: predicate.same_eliminate 0.51% : 0.000003s : 40: predicate.set_cell_output_no_recompute 0.89% : 0.000006s : 34: predicate.shard_identity_eliminate 0.87% : 0.000006s : 34: predicate.special_op_eliminate 0.84% : 0.000005s : 34: predicate.specialize_transform 1.00% : 0.000006s : 34: predicate.split_environ_get_set_with_tuple_value 0.85% : 0.000005s : 34: predicate.stack_unstack_eliminate 0.45% : 0.000003s : 17: predicate.switch_call_monad_eliminater 1.08% : 0.000007s : 47: predicate.switch_defer_inline 1.78% : 0.000011s : 81: predicate.switch_layer_defer_inline 3.88% : 0.000025s : 162: predicate.switch_simplify 0.88% : 0.000006s : 43: predicate.tile_eliminate 0.90% : 0.000006s : 43: predicate.transpose_eliminate 1.81% : 0.000012s : 77: predicate.tuple_list_convert_item_index_to_positive 1.77% : 0.000011s : 77: predicate.tuple_list_get_item_const_eliminator 1.74% : 0.000011s : 77: predicate.tuple_list_get_item_depend_reorder 2.82% : 0.000018s : 111: predicate.tuple_list_get_item_eliminator 1.89% : 0.000012s : 77: predicate.tuple_list_get_set_item_eliminator 2.59% : 0.000017s : 111: predicate.tuple_list_set_item_eliminator 1.66% : 0.000011s : 77: predicate.tuple_to_list_eliminator_ 2.64% : 0.000017s : 120: predicate.updatestate_pure_node_eliminater 3.57% : 0.000023s : 154: predicate.updatestate_useless_node_eliminater 0.40% : 0.000003s : 17: predicate.value_based_eliminate 0.79% : 0.000005s : 34: predicate.virtual_dataset_eliminate 0.79% : 0.000005s : 34: predicate.virtual_output_eliminate 0.36% : 0.000002s : 17: predicate.virtual_view_grad_eliminate 0.44% : 0.000003s : 17: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000679 7 4.09% : 0.000028s : 1: func_graph_cloner_run.FuncGraphClonerGraph 95.91% : 0.000651s : 6: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.176379 192 0.00% : 0.000004s : 1: ForceFp32Comm 2.43% : 0.004279s : 1: add_attr 2.42% : 0.004265s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.08% : 0.000142s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.18% : 0.000321s : 1: auto_monad 0.05% : 0.000083s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.33% : 0.000579s : 1: bootstrap 0.03% : 0.000046s : 1: cconv 0.00% : 0.000005s : 1: comm_op_add_attrs 0.02% : 0.000037s : 1: control_data_broadcast_order 0.01% : 0.000026s : 1: convert_after_rewriter 0.03% : 0.000057s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.01% : 0.000023s : 1: environ_conv 0.02% : 0.000033s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.01% : 0.000011s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.01% : 0.000010s : 1: label_micro_interleaved_index 0.30% : 0.000531s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.48% : 0.000855s : 1: mutable_eliminate 0.01% : 0.000013s : 1: offloading_packed_experts 0.02% : 0.000029s : 1: opt.transform.loop_unroll_optimizer 0.02% : 0.000035s : 1: opt.transform.mutable_eliminate 1.85% : 0.003264s : 78: opt.transform.opt_a 0.07% : 0.000126s : 1: opt.transform.opt_after_cconv 0.04% : 0.000070s : 1: opt.transform.opt_after_jit_grad 0.25% : 0.000437s : 28: opt.transform.opt_b 0.09% : 0.000159s : 2: opt.transform.opt_trans_graph 0.06% : 0.000114s : 4: opt.transform.symbol_engine_opt 3.76% : 0.006626s : 1: opt_a 0.15% : 0.000273s : 1: opt_after_cconv 0.42% : 0.000741s : 1: opt_after_jit_grad 0.36% : 0.000641s : 1: opt_b 6.01% : 0.010598s : 1: optimize 0.02% : 0.000044s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000016s : 1: order_py_execute_after_rewriter 0.03% : 0.000056s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000013s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000012s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.03% : 0.000056s : 1: pre_auto_parallel 0.03% : 0.000044s : 1: py_interpret_to_execute 0.02% : 0.000038s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.04% : 0.000072s : 1: remove_dup_value 0.55% : 0.000978s : 1: renormalize.infer 0.50% : 0.000876s : 1: renormalize.specialize 0.00% : 0.000005s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.19% : 0.000327s : 1: rewriter_after_opt_a 0.07% : 0.000125s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000020s : 1: swap_dp_allreduce_reducescatter 0.10% : 0.000172s : 1: symbol_engine_optimizer 0.11% : 0.000192s : 1: tuple_transform 78.65% : 0.138720s : 1: type_inference . TotalTime = 0.216384, [21] [bootstrap]: 0.00062805 [type_inference]: 0.0120093 [event_method]: 2.062e-05 [auto_monad]: 0.00030864 [graph_reusing]: 6.71999e-06 [inline]: 3.77998e-06 [add_attr]: 0.00410911, [1] [add_attr_with_inline]: 0.00409631, [1] [Cycle 1]: 9.291e-05, [2] [tag_attr]: 3.678e-05 [meta_addattr_fg_expand]: 8.32998e-06 [parallel-infer-symbol]: 3.94002e-06 [pre_auto_parallel]: 5.163e-05 [insert-virtual-dataset]: 3.2e-06 [parallel-infer-symbol-second]: 8.70001e-07 [dataset_repeat_opt]: 2.32001e-06 [pipeline_split]: 1.92001e-06 [optimize]: 0.197851, [53] [py_interpret_to_execute]: 4.225e-05 [rewriter_before_opt_a]: 0.00011685 [opt_a]: 0.193433, [2] [Cycle 1]: 0.191348, [45] [expand_dump_flag]: 3.53999e-06 [switch_simplify]: 5.731e-05 [loop_unroll]: 4.282e-05 [a_1]: 0.187669 [with_stream_mark]: 6.471e-05 [recompute_prepare]: 3.636e-05 [updatestate_depend_eliminate]: 0.00012521 [updatestate_assign_eliminate]: 1.644e-05 [updatestate_loads_eliminate]: 1.691e-05 [parameter_eliminate]: 3.6e-06 [a_2]: 0.00028857 [accelerated_algorithm]: 4.4e-05 [shard]: 3.26001e-06 [meta_shard_fg_expand]: 8.28999e-06 [shard_inline]: 2.033e-05 [merge_send_recv]: 1.933e-05 [auto_parallel]: 5.712e-05 [parallel]: 2.925e-05 [flash_sp]: 1.579e-05 [merge_comm]: 1.215e-05 [allreduce_fusion]: 1.188e-05 [matmul_add_comm_reduction]: 2.199e-05 [allreduce_slice_to_reducescatter]: 9.60019e-07 [virtual_shard_identity]: 2.26e-05 [virtual_dataset]: 1.928e-05 [get_grad_eliminate_]: 1.835e-05 [virtual_output]: 1.918e-05 [merge_forward]: 1.123e-05 [cell_reuse_recompute_pass]: 2.76999e-06 [offload_activation]: 2.23e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.136e-05 [merge_recompute_call_nodes]: 1.80001e-06 [before_grad]: 3.503e-05 [set_forward_comm_id_for_comm_node_pass]: 1.263e-05 [meta_fg_expand]: 9.65002e-06 [flash_sp_send_recv_attached]: 6.46e-06 [receive_attached]: 2.61e-06 [after_resolve]: 2.842e-05 [a_after_grad]: 3.306e-05 [renormalize]: 0.00179761 [add_forward_monad_depend]: 1.186e-05 [auto_monad_grad]: 3.44001e-06 [auto_monad_eliminator]: 4.436e-05 [cse]: 0.00012343 [a_3]: 0.00015077 [Cycle 2]: 0.00206809, [45] [expand_dump_flag]: 2.67001e-06 [switch_simplify]: 2.254e-05 [loop_unroll]: 1.975e-05 [a_1]: 0.00060836 [with_stream_mark]: 3.446e-05 [recompute_prepare]: 2.736e-05 [updatestate_depend_eliminate]: 1.292e-05 [updatestate_assign_eliminate]: 1.625e-05 [updatestate_loads_eliminate]: 1.602e-05 [parameter_eliminate]: 2.99999e-06 [a_2]: 0.00027976 [accelerated_algorithm]: 2.892e-05 [shard]: 2.65002e-06 [meta_shard_fg_expand]: 6.31e-06 [shard_inline]: 1.959e-05 [merge_send_recv]: 3.46e-05 [auto_parallel]: 2.02e-05 [parallel]: 1.36e-05 [flash_sp]: 5.62001e-06 [merge_comm]: 1.403e-05 [allreduce_fusion]: 1.286e-05 [matmul_add_comm_reduction]: 2.445e-05 [allreduce_slice_to_reducescatter]: 9.39996e-07 [virtual_shard_identity]: 2.68e-05 [virtual_dataset]: 2.001e-05 [get_grad_eliminate_]: 1.826e-05 [virtual_output]: 2.084e-05 [merge_forward]: 1.299e-05 [cell_reuse_recompute_pass]: 3.32002e-06 [offload_activation]: 2.47e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.678e-05 [merge_recompute_call_nodes]: 1.84e-06 [before_grad]: 3.883e-05 [set_forward_comm_id_for_comm_node_pass]: 1.407e-05 [meta_fg_expand]: 9.22001e-06 [flash_sp_send_recv_attached]: 2.26e-06 [receive_attached]: 2.88e-06 [after_resolve]: 2.915e-05 [a_after_grad]: 3.295e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 5.92999e-06 [auto_monad_grad]: 2.61e-06 [auto_monad_eliminator]: 3.778e-05 [cse]: 7.513e-05 [a_3]: 0.00013379 [py_interpret_to_execute_after_opt_a]: 3.473e-05 [slice_cell_reuse_recomputed_activation]: 2.22001e-06 [rewriter_after_opt_a]: 0.00024911 [convert_after_rewriter]: 2.374e-05 [order_py_execute_after_rewriter]: 1.276e-05 [mutable_eliminate]: 0.00090349 [opt_b]: 0.00068291, [1] [Cycle 1]: 0.00067109, [7] [b_1]: 0.00045414 [b_2]: 2.35e-05 [updatestate_depend_eliminate]: 2.138e-05 [updatestate_assign_eliminate]: 1.477e-05 [updatestate_loads_eliminate]: 1.708e-05 [renormalize]: 1.60001e-06 [cse]: 8.646e-05 [optimize_parallel_all_gather_comm]: 4.477e-05 [overlap_param_gather]: 2.27999e-06 [cconv]: 4.972e-05 [loop_unroll]: 0.00069453 [opt_after_cconv]: 0.00030218, [1] [Cycle 1]: 0.00029332, [7] [c_1]: 0.0001385 [parameter_eliminate]: 6.52001e-06 [updatestate_depend_eliminate]: 1.895e-05 [updatestate_assign_eliminate]: 1.287e-05 [updatestate_loads_eliminate]: 1.386e-05 [cse]: 6.189e-05 [renormalize]: 1.19998e-06 [remove_dup_value]: 0.00010856 [tuple_transform]: 0.00019931, [1] [Cycle 1]: 0.00019349, [4] [d_1]: 0.0001479 [none_parameter_eliminate]: 2.41998e-06 [renormalize]: 1.80007e-07 [switch_simplify]: 2.04e-05 [partial_unused_args_eliminate]: 2.11e-06 [add_recomputation]: 0.00014937 [cse_after_recomputation]: 6.255e-05, [1] [Cycle 1]: 5.705e-05, [1] [cse]: 4.9e-05 [environ_conv]: 2.279e-05 [swap_dp_allreduce_reducescatter]: 1.788e-05 [bias_add_comm_swap]: 4.08999e-06 [label_micro_interleaved_index]: 9.07001e-06 [label_fine_grained_interleaved_index]: 3.60003e-06 [merge_cast_opt]: 1.46998e-06 [slice_recompute_activation]: 2.26e-06 [micro_interleaved_order_control]: 3.10002e-06 [assign_add_opt]: 1.66e-06 [ForceFp32Comm]: 1.02e-06 [remove_cast_before_assign_add]: 1.27999e-06 [full_micro_interleaved_order_control]: 2.88e-06 [reorder_send_recv_between_fp_bp]: 2.98e-06 [comm_op_add_attrs]: 1.16002e-06 [add_comm_op_reuse_tag]: 1.07e-06 [interleave_split_concat_branches]: 1.27999e-06 [interleave_parallel_branches]: 1.10001e-06 [overlap_opt_shard_in_pipeline]: 2.06e-06 [overlap_opt_shard_grad_in_pipeline]: 2.23998e-06 [control_data_broadcast_order]: 3.979e-05 [grouped_pairwise_exchange_alltoall]: 1.93997e-06 [offloading_packed_experts]: 9.86e-06 [overlap_recompute_and_grad_model_parallel]: 1.037e-05 [overlap_grad_matmul_and_grad_allreduce]: 1.57001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.49e-06 [overlap_recompute_comm]: 2.88e-06 [overlap_grad_ring_attention]: 9.81e-06 [overlap_grad_flash_sp]: 5.403e-05 [begin_end_overlap_inline]: 4.89992e-07 [split_matmul_comm_elemetwise]: 2.98998e-06 [split_layernorm_comm]: 2.21998e-06 [handle_group_info]: 1.04e-06 [symbol_engine_optimizer]: 0.00020756, [1] [Cycle 1]: 0.00020008, [6] [build]: 2.289e-05 [elim_shapecalc]: 3.598e-05 [elim_not_effective]: 4.145e-05 [opt_reshape]: 2.346e-05 [fold_const_symbol]: 3.363e-05 [renormalize]: 1.80007e-07 [detach_backward]: 2.91999e-06 [pipeline_parallel_scheduler]: 1.87001e-06 [auto_monad_reorder]: 9.3e-05 [get_jit_bprop_graph]: 2.88e-06 [rewriter_after_jit_bprop_graph]: 8.35001e-06 [opt_after_jit_grad]: 0.00092514 [validate]: 0.00012097 Sums bootstrap : 0.000628s : 0.30% type_inference : 0.012009s : 5.69% event_method : 0.000021s : 0.01% auto_monad : 0.000309s : 0.15% graph_reusing : 0.000007s : 0.00% inline : 0.000004s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000037s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000008s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000052s : 0.02% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000042s : 0.02% optimize.rewriter_before_opt_a : 0.000117s : 0.06% optimize.opt_a.expand_dump_flag : 0.000006s : 0.00% optimize.opt_a.switch_simplify : 0.000080s : 0.04% optimize.opt_a.loop_unroll : 0.000063s : 0.03% optimize.opt_a.a_1 : 0.188278s : 89.26% optimize.opt_a.with_stream_mark : 0.000099s : 0.05% optimize.opt_a.recompute_prepare : 0.000064s : 0.03% optimize.opt_a.updatestate_depend_eliminate : 0.000138s : 0.07% optimize.opt_a.updatestate_assign_eliminate : 0.000033s : 0.02% optimize.opt_a.updatestate_loads_eliminate : 0.000033s : 0.02% optimize.opt_a.parameter_eliminate : 0.000007s : 0.00% optimize.opt_a.a_2 : 0.000568s : 0.27% optimize.opt_a.accelerated_algorithm : 0.000073s : 0.03% optimize.opt_a.shard : 0.000006s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000015s : 0.01% optimize.opt_a.shard_inline : 0.000040s : 0.02% optimize.opt_a.merge_send_recv : 0.000054s : 0.03% optimize.opt_a.auto_parallel : 0.000077s : 0.04% optimize.opt_a.parallel : 0.000043s : 0.02% optimize.opt_a.flash_sp : 0.000021s : 0.01% optimize.opt_a.merge_comm : 0.000026s : 0.01% optimize.opt_a.allreduce_fusion : 0.000025s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000046s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000049s : 0.02% optimize.opt_a.virtual_dataset : 0.000039s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000037s : 0.02% optimize.opt_a.virtual_output : 0.000040s : 0.02% optimize.opt_a.merge_forward : 0.000024s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000006s : 0.00% optimize.opt_a.offload_activation : 0.000047s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000088s : 0.04% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000074s : 0.04% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000027s : 0.01% optimize.opt_a.meta_fg_expand : 0.000019s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000009s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000058s : 0.03% optimize.opt_a.a_after_grad : 0.000066s : 0.03% optimize.opt_a.renormalize : 0.001798s : 0.85% optimize.opt_a.add_forward_monad_depend : 0.000018s : 0.01% optimize.opt_a.auto_monad_grad : 0.000006s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000082s : 0.04% optimize.opt_a.cse : 0.000199s : 0.09% optimize.opt_a.a_3 : 0.000285s : 0.13% optimize.py_interpret_to_execute_after_opt_a : 0.000035s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000249s : 0.12% optimize.convert_after_rewriter : 0.000024s : 0.01% optimize.order_py_execute_after_rewriter : 0.000013s : 0.01% optimize.mutable_eliminate : 0.000903s : 0.43% optimize.opt_b.b_1 : 0.000454s : 0.22% optimize.opt_b.b_2 : 0.000023s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000021s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000015s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000017s : 0.01% optimize.opt_b.renormalize : 0.000002s : 0.00% optimize.opt_b.cse : 0.000086s : 0.04% optimize.optimize_parallel_all_gather_comm : 0.000045s : 0.02% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000050s : 0.02% optimize.loop_unroll : 0.000695s : 0.33% optimize.opt_after_cconv.c_1 : 0.000139s : 0.07% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000019s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000013s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000014s : 0.01% optimize.opt_after_cconv.cse : 0.000062s : 0.03% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000109s : 0.05% optimize.tuple_transform.d_1 : 0.000148s : 0.07% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000020s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000149s : 0.07% optimize.cse_after_recomputation.cse : 0.000049s : 0.02% optimize.environ_conv : 0.000023s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000018s : 0.01% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000009s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000004s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000040s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000010s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000010s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000010s : 0.00% optimize.overlap_grad_flash_sp : 0.000054s : 0.03% optimize.begin_end_overlap_inline : 0.000000s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000023s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000036s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000041s : 0.02% optimize.symbol_engine_optimizer.opt_reshape : 0.000023s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000034s : 0.02% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000093s : 0.04% get_jit_bprop_graph : 0.000003s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.000925s : 0.44% validate : 0.000121s : 0.06% Time group info: ------[substitution.] 0.186802 203 99.73% : 0.186296s : 4: substitution.cast_eliminate 0.01% : 0.000012s : 2: substitution.depend_value_elim 0.00% : 0.000006s : 14: substitution.elim_not_effective 0.00% : 0.000005s : 14: substitution.fold_const_symbol 0.01% : 0.000015s : 17: substitution.graph_param_transform 0.15% : 0.000289s : 4: substitution.inline 0.01% : 0.000014s : 28: substitution.j_node_and_user_rematch 0.01% : 0.000024s : 2: substitution.less_batch_normalization 0.01% : 0.000011s : 12: substitution.load_eliminater 0.01% : 0.000020s : 28: substitution.remove_not_recompute_node 0.01% : 0.000010s : 6: substitution.replace_old_param 0.03% : 0.000051s : 32: substitution.updatestate_pure_node_eliminater 0.03% : 0.000049s : 40: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.011928 2 92.66% : 0.011052s : 1: type_inference.infer 7.34% : 0.000876s : 1: type_inference.specialize ------[replace.] 0.000047 4 100.00% : 0.000047s : 4: replace.inline ------[match.] 0.000286 4 100.00% : 0.000286s : 4: match.inline ------[predicate.] 0.000677 4350 0.92% : 0.000006s : 43: predicate.accumulaten_eliminater 1.00% : 0.000007s : 17: predicate.ad_related_special_op_eliminate 0.70% : 0.000005s : 34: predicate.addn_check_dump 0.91% : 0.000006s : 43: predicate.addn_zero_filter 0.89% : 0.000006s : 43: predicate.adjust_all_reduce_mul_add 2.18% : 0.000015s : 77: predicate.arithmetic_simplify 0.91% : 0.000006s : 43: predicate.cast_eliminate 0.72% : 0.000005s : 34: predicate.check_bprop_eliminate 0.69% : 0.000005s : 34: predicate.compare_switch_simplify 0.18% : 0.000001s : 17: predicate.const_output_eliminate 0.72% : 0.000005s : 34: predicate.depend_value_elim 0.93% : 0.000006s : 43: predicate.dict_get_item_const_eliminator 0.99% : 0.000007s : 43: predicate.dict_get_item_eliminator 0.91% : 0.000006s : 43: predicate.dict_set_item_eliminator 1.09% : 0.000007s : 34: predicate.dumpgradient_eliminate 0.33% : 0.000002s : 17: predicate.elim_not_effective 0.55% : 0.000004s : 17: predicate.elim_shapecalc_of_broadcastargs 1.25% : 0.000008s : 60: predicate.environ_add_const_eliminate 1.19% : 0.000008s : 60: predicate.environ_get_add_eliminate 1.18% : 0.000008s : 60: predicate.environ_get_depend_swap 1.94% : 0.000013s : 94: predicate.environ_get_eliminate 1.18% : 0.000008s : 60: predicate.environ_get_set_eliminate 0.93% : 0.000006s : 47: predicate.exchange_switch_depend_value 1.44% : 0.000010s : 47: predicate.float_depend_g_call 0.71% : 0.000005s : 34: predicate.float_environ_get_switch 1.05% : 0.000007s : 51: predicate.float_tuple_getitem_switch 0.19% : 0.000001s : 17: predicate.fold_const_symbol 0.75% : 0.000005s : 34: predicate.get_grad_eliminate 0.24% : 0.000002s : 17: predicate.graph_param_transform 0.73% : 0.000005s : 34: predicate.incorporate_call 0.66% : 0.000004s : 34: predicate.incorporate_call_switch 5.84% : 0.000040s : 192: predicate.inline 1.04% : 0.000007s : 34: predicate.inline_without_move 0.37% : 0.000002s : 34: predicate.j_node_and_user_rematch 1.22% : 0.000008s : 34: predicate.less_batch_normalization 1.72% : 0.000012s : 77: predicate.list_to_tuple_eliminator_ 2.48% : 0.000017s : 120: predicate.load_eliminater 1.00% : 0.000007s : 17: predicate.loop_unroll_after_grad 1.38% : 0.000009s : 64: predicate.loop_unroll_before_grad 1.72% : 0.000012s : 77: predicate.make_slice_get_slice_eliminator 0.70% : 0.000005s : 34: predicate.merge_addn 0.75% : 0.000005s : 34: predicate.micro_step_allgather_replace 0.71% : 0.000005s : 34: predicate.mini_step_allgather_replace 0.91% : 0.000006s : 43: predicate.minmaximum_grad 1.15% : 0.000008s : 17: predicate.mutable_eliminate 0.42% : 0.000003s : 17: predicate.opt_reshape 0.39% : 0.000003s : 17: predicate.parallel_virtual_node 1.38% : 0.000009s : 47: predicate.partial_defer_inline 1.35% : 0.000009s : 60: predicate.partial_eliminate 0.88% : 0.000006s : 43: predicate.print_const_string_wrapper 0.73% : 0.000005s : 34: predicate.reduce_all_const_elim 1.30% : 0.000009s : 43: predicate.reduce_eliminate 2.44% : 0.000017s : 120: predicate.redundant_stop_gradient_eliminater 0.47% : 0.000003s : 34: predicate.remove_not_recompute_node 1.09% : 0.000007s : 77: predicate.replace_applicator 0.49% : 0.000003s : 34: predicate.replace_old_param 0.27% : 0.000002s : 17: predicate.reset_defer_inline 1.15% : 0.000008s : 43: predicate.reshape_eliminate 0.75% : 0.000005s : 34: predicate.row_tensor_add_zeros_like 0.49% : 0.000003s : 17: predicate.row_tensor_eliminate 0.91% : 0.000006s : 34: predicate.same_eliminate 0.51% : 0.000003s : 40: predicate.set_cell_output_no_recompute 0.91% : 0.000006s : 34: predicate.shard_identity_eliminate 0.77% : 0.000005s : 34: predicate.special_op_eliminate 0.81% : 0.000005s : 34: predicate.specialize_transform 1.06% : 0.000007s : 34: predicate.split_environ_get_set_with_tuple_value 0.89% : 0.000006s : 34: predicate.stack_unstack_eliminate 0.42% : 0.000003s : 17: predicate.switch_call_monad_eliminater 1.02% : 0.000007s : 47: predicate.switch_defer_inline 1.71% : 0.000012s : 81: predicate.switch_layer_defer_inline 3.77% : 0.000026s : 162: predicate.switch_simplify 0.94% : 0.000006s : 43: predicate.tile_eliminate 0.94% : 0.000006s : 43: predicate.transpose_eliminate 1.82% : 0.000012s : 77: predicate.tuple_list_convert_item_index_to_positive 1.69% : 0.000011s : 77: predicate.tuple_list_get_item_const_eliminator 1.58% : 0.000011s : 77: predicate.tuple_list_get_item_depend_reorder 2.73% : 0.000018s : 111: predicate.tuple_list_get_item_eliminator 1.68% : 0.000011s : 77: predicate.tuple_list_get_set_item_eliminator 2.61% : 0.000018s : 111: predicate.tuple_list_set_item_eliminator 1.57% : 0.000011s : 77: predicate.tuple_to_list_eliminator_ 2.76% : 0.000019s : 120: predicate.updatestate_pure_node_eliminater 3.50% : 0.000024s : 154: predicate.updatestate_useless_node_eliminater 0.37% : 0.000003s : 17: predicate.value_based_eliminate 0.76% : 0.000005s : 34: predicate.virtual_dataset_eliminate 0.76% : 0.000005s : 34: predicate.virtual_output_eliminate 0.37% : 0.000003s : 17: predicate.virtual_view_grad_eliminate 0.44% : 0.000003s : 17: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000477 7 6.66% : 0.000032s : 1: func_graph_cloner_run.FuncGraphClonerGraph 93.34% : 0.000445s : 6: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.610702 192 0.00% : 0.000004s : 1: ForceFp32Comm 0.67% : 0.004116s : 1: add_attr 0.67% : 0.004101s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.03% : 0.000159s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.05% : 0.000320s : 1: auto_monad 0.02% : 0.000104s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.11% : 0.000661s : 1: bootstrap 0.01% : 0.000056s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000046s : 1: control_data_broadcast_order 0.00% : 0.000029s : 1: convert_after_rewriter 0.01% : 0.000066s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000027s : 1: environ_conv 0.00% : 0.000028s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000007s : 1: get_jit_bprop_graph 0.00% : 0.000012s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000008s : 1: inline 0.00% : 0.000008s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000012s : 1: label_micro_interleaved_index 0.12% : 0.000714s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.15% : 0.000923s : 1: mutable_eliminate 0.00% : 0.000013s : 1: offloading_packed_experts 0.01% : 0.000039s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000044s : 1: opt.transform.mutable_eliminate 31.07% : 0.189770s : 78: opt.transform.opt_a 0.02% : 0.000137s : 1: opt.transform.opt_after_cconv 0.01% : 0.000085s : 1: opt.transform.opt_after_jit_grad 0.07% : 0.000441s : 28: opt.transform.opt_b 0.03% : 0.000165s : 2: opt.transform.opt_trans_graph 0.02% : 0.000129s : 4: opt.transform.symbol_engine_opt 31.67% : 0.193437s : 1: opt_a 0.05% : 0.000307s : 1: opt_after_cconv 0.16% : 0.000952s : 1: opt_after_jit_grad 0.11% : 0.000687s : 1: opt_b 32.40% : 0.197857s : 1: optimize 0.01% : 0.000051s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000016s : 1: order_py_execute_after_rewriter 0.01% : 0.000060s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000014s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000014s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000007s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000056s : 1: pre_auto_parallel 0.01% : 0.000047s : 1: py_interpret_to_execute 0.01% : 0.000041s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.02% : 0.000115s : 1: remove_dup_value 0.17% : 0.001010s : 1: renormalize.infer 0.13% : 0.000772s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000011s : 1: rewriter_after_jit_bprop_graph 0.04% : 0.000264s : 1: rewriter_after_opt_a 0.02% : 0.000122s : 1: rewriter_before_opt_a 0.00% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000021s : 1: swap_dp_allreduce_reducescatter 0.03% : 0.000211s : 1: symbol_engine_optimizer 0.03% : 0.000203s : 1: tuple_transform 1.97% : 0.012042s : 1: type_inference [WARNING] PIPELINE(166799,ffff961eff30,python3.9):2026-01-29-17:41:49.475.860 [mindspore/ccsrc/frontend/jit/ps/validator.cc:244] CheckAssignReturnValue] Deprecated: the return value of Assign/AssignAdd/AssignSub operator will be removed in subsequent releases. You can modify the code from: a = P.Assign()(param, value) b = a * 2 to: P.Assign()(param, value) b = param * 2 Please check your code:In file /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/auto_generate/gen_ops_def.py:9741, 11~37 return assign_op(variable, value) ^~~~~~~~~~~~~~~~~~~~~~~~~~ TotalTime = 0.320677, [21] [bootstrap]: 0.116494 [type_inference]: 0.0151883 [event_method]: 1.785e-05 [auto_monad]: 0.00024849 [graph_reusing]: 6.36e-06 [inline]: 3.01001e-06 [add_attr]: 0.176248, [1] [add_attr_with_inline]: 0.176231, [1] [Cycle 1]: 0.00010076, [2] [tag_attr]: 3.632e-05 [meta_addattr_fg_expand]: 9.19e-06 [parallel-infer-symbol]: 4.65001e-06 [pre_auto_parallel]: 4.929e-05 [insert-virtual-dataset]: 2.61e-06 [parallel-infer-symbol-second]: 8.50006e-07 [dataset_repeat_opt]: 2.24999e-06 [pipeline_split]: 1.61998e-06 [optimize]: 0.0100948, [53] [py_interpret_to_execute]: 5.045e-05 [rewriter_before_opt_a]: 0.00010788 [opt_a]: 0.00598031, [2] [Cycle 1]: 0.00430955, [45] [expand_dump_flag]: 3.76999e-06 [switch_simplify]: 4.881e-05 [loop_unroll]: 3.406e-05 [a_1]: 0.00126369 [with_stream_mark]: 3.499e-05 [recompute_prepare]: 3.049e-05 [updatestate_depend_eliminate]: 0.00015222 [updatestate_assign_eliminate]: 5.994e-05 [updatestate_loads_eliminate]: 1.725e-05 [parameter_eliminate]: 2.41998e-06 [a_2]: 0.00025096 [accelerated_algorithm]: 4.687e-05 [shard]: 2.83e-06 [meta_shard_fg_expand]: 4.55001e-06 [shard_inline]: 1.6e-05 [merge_send_recv]: 1.535e-05 [auto_parallel]: 1.486e-05 [parallel]: 2.692e-05 [flash_sp]: 1.488e-05 [merge_comm]: 9.52001e-06 [allreduce_fusion]: 8.47e-06 [matmul_add_comm_reduction]: 1.798e-05 [allreduce_slice_to_reducescatter]: 1.45001e-06 [virtual_shard_identity]: 1.814e-05 [virtual_dataset]: 1.502e-05 [get_grad_eliminate_]: 1.53e-05 [virtual_output]: 1.614e-05 [merge_forward]: 9.02e-06 [cell_reuse_recompute_pass]: 1.82001e-06 [offload_activation]: 1.965e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.209e-05 [merge_recompute_call_nodes]: 2.36998e-06 [before_grad]: 2.868e-05 [set_forward_comm_id_for_comm_node_pass]: 9.61998e-06 [meta_fg_expand]: 7.55998e-06 [flash_sp_send_recv_attached]: 5.56e-06 [receive_attached]: 2.07999e-06 [after_resolve]: 2.386e-05 [a_after_grad]: 2.517e-05 [renormalize]: 0.00134989 [add_forward_monad_depend]: 1.078e-05 [auto_monad_grad]: 3.19001e-06 [auto_monad_eliminator]: 3.588e-05 [cse]: 0.00013303 [a_3]: 0.00013228 [Cycle 2]: 0.00165438, [45] [expand_dump_flag]: 3.38e-06 [switch_simplify]: 1.9e-05 [loop_unroll]: 1.513e-05 [a_1]: 0.00046713 [with_stream_mark]: 2.91e-05 [recompute_prepare]: 1.736e-05 [updatestate_depend_eliminate]: 9.61e-06 [updatestate_assign_eliminate]: 1.093e-05 [updatestate_loads_eliminate]: 1.52e-05 [parameter_eliminate]: 2.50002e-06 [a_2]: 0.0002262 [accelerated_algorithm]: 2.371e-05 [shard]: 2.54001e-06 [meta_shard_fg_expand]: 4.47e-06 [shard_inline]: 1.665e-05 [merge_send_recv]: 1.623e-05 [auto_parallel]: 1.645e-05 [parallel]: 1.099e-05 [flash_sp]: 4.52e-06 [merge_comm]: 9.45001e-06 [allreduce_fusion]: 8.55001e-06 [matmul_add_comm_reduction]: 1.809e-05 [allreduce_slice_to_reducescatter]: 8.60018e-07 [virtual_shard_identity]: 1.771e-05 [virtual_dataset]: 1.523e-05 [get_grad_eliminate_]: 1.469e-05 [virtual_output]: 1.544e-05 [merge_forward]: 9.82001e-06 [cell_reuse_recompute_pass]: 2.94001e-06 [offload_activation]: 1.954e-05 [cell_reuse_handle_not_recompute_node_pass]: 5.598e-05 [merge_recompute_call_nodes]: 1.59e-06 [before_grad]: 2.941e-05 [set_forward_comm_id_for_comm_node_pass]: 1.055e-05 [meta_fg_expand]: 6.61e-06 [flash_sp_send_recv_attached]: 2.32001e-06 [receive_attached]: 2.14e-06 [after_resolve]: 2.58e-05 [a_after_grad]: 2.365e-05 [renormalize]: 3.99887e-08 [add_forward_monad_depend]: 4.72e-06 [auto_monad_grad]: 2.79999e-06 [auto_monad_eliminator]: 3.036e-05 [cse]: 6.075e-05 [a_3]: 0.00010665 [py_interpret_to_execute_after_opt_a]: 2.9e-05 [slice_cell_reuse_recomputed_activation]: 2.86e-06 [rewriter_after_opt_a]: 0.00021514 [convert_after_rewriter]: 1.856e-05 [order_py_execute_after_rewriter]: 1.12e-05 [mutable_eliminate]: 0.00090681 [opt_b]: 0.00059836, [1] [Cycle 1]: 0.00058629, [7] [b_1]: 0.00040347 [b_2]: 1.964e-05 [updatestate_depend_eliminate]: 1.771e-05 [updatestate_assign_eliminate]: 1.063e-05 [updatestate_loads_eliminate]: 1.534e-05 [renormalize]: 1.17999e-06 [cse]: 6.808e-05 [optimize_parallel_all_gather_comm]: 3.844e-05 [overlap_param_gather]: 2.08002e-06 [cconv]: 4.487e-05 [loop_unroll]: 0.00067334 [opt_after_cconv]: 0.00024994, [1] [Cycle 1]: 0.00024278, [7] [c_1]: 0.00011593 [parameter_eliminate]: 6.65002e-06 [updatestate_depend_eliminate]: 1.33e-05 [updatestate_assign_eliminate]: 9.31e-06 [updatestate_loads_eliminate]: 1.206e-05 [cse]: 4.731e-05 [renormalize]: 9.09989e-07 [remove_dup_value]: 5.62e-05 [tuple_transform]: 0.00015584, [1] [Cycle 1]: 0.00014997, [4] [d_1]: 0.00011248 [none_parameter_eliminate]: 2.12999e-06 [renormalize]: 1.90019e-07 [switch_simplify]: 1.629e-05 [partial_unused_args_eliminate]: 2.50002e-06 [add_recomputation]: 0.00011208 [cse_after_recomputation]: 0.00014799, [1] [Cycle 1]: 0.00014139, [1] [cse]: 0.00012909 [environ_conv]: 1.755e-05 [swap_dp_allreduce_reducescatter]: 3.949e-05 [bias_add_comm_swap]: 3.35e-06 [label_micro_interleaved_index]: 8.89e-06 [label_fine_grained_interleaved_index]: 3.23e-06 [merge_cast_opt]: 1.42999e-06 [slice_recompute_activation]: 2.53e-06 [micro_interleaved_order_control]: 2.62001e-06 [assign_add_opt]: 1.60999e-06 [ForceFp32Comm]: 9.29984e-07 [remove_cast_before_assign_add]: 9.70002e-07 [full_micro_interleaved_order_control]: 2.44001e-06 [reorder_send_recv_between_fp_bp]: 3.14001e-06 [comm_op_add_attrs]: 1.17999e-06 [add_comm_op_reuse_tag]: 1.04e-06 [interleave_split_concat_branches]: 1.17e-06 [interleave_parallel_branches]: 1.40999e-06 [overlap_opt_shard_in_pipeline]: 1.57001e-06 [overlap_opt_shard_grad_in_pipeline]: 2.19001e-06 [control_data_broadcast_order]: 3.077e-05 [grouped_pairwise_exchange_alltoall]: 1.80001e-06 [offloading_packed_experts]: 8.42e-06 [overlap_recompute_and_grad_model_parallel]: 8.65999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.25001e-06 [overlap_recompute_allgather_and_fa_grad]: 2.46998e-06 [overlap_recompute_comm]: 2.37999e-06 [overlap_grad_ring_attention]: 7.87e-06 [overlap_grad_flash_sp]: 4.44e-05 [begin_end_overlap_inline]: 5.00004e-07 [split_matmul_comm_elemetwise]: 2.40002e-06 [split_layernorm_comm]: 1.99e-06 [handle_group_info]: 1.10999e-06 [symbol_engine_optimizer]: 0.00017814, [1] [Cycle 1]: 0.00017062, [6] [build]: 1.797e-05 [elim_shapecalc]: 3.156e-05 [elim_not_effective]: 3.401e-05 [opt_reshape]: 1.934e-05 [fold_const_symbol]: 2.587e-05 [renormalize]: 1.59984e-07 [detach_backward]: 3.30998e-06 [pipeline_parallel_scheduler]: 1.69e-06 [auto_monad_reorder]: 5.893e-05 [get_jit_bprop_graph]: 2.46998e-06 [rewriter_after_jit_bprop_graph]: 9.03002e-06 [opt_after_jit_grad]: 0.0007845 [validate]: 0.00123277 Sums bootstrap : 0.116494s : 81.38% type_inference : 0.015188s : 10.61% event_method : 0.000018s : 0.01% auto_monad : 0.000248s : 0.17% graph_reusing : 0.000006s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000036s : 0.03% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000009s : 0.01% parallel-infer-symbol : 0.000005s : 0.00% pre_auto_parallel : 0.000049s : 0.03% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000050s : 0.04% optimize.rewriter_before_opt_a : 0.000108s : 0.08% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000068s : 0.05% optimize.opt_a.loop_unroll : 0.000049s : 0.03% optimize.opt_a.a_1 : 0.001731s : 1.21% optimize.opt_a.with_stream_mark : 0.000064s : 0.04% optimize.opt_a.recompute_prepare : 0.000048s : 0.03% optimize.opt_a.updatestate_depend_eliminate : 0.000162s : 0.11% optimize.opt_a.updatestate_assign_eliminate : 0.000071s : 0.05% optimize.opt_a.updatestate_loads_eliminate : 0.000032s : 0.02% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000477s : 0.33% optimize.opt_a.accelerated_algorithm : 0.000071s : 0.05% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000009s : 0.01% optimize.opt_a.shard_inline : 0.000033s : 0.02% optimize.opt_a.merge_send_recv : 0.000032s : 0.02% optimize.opt_a.auto_parallel : 0.000031s : 0.02% optimize.opt_a.parallel : 0.000038s : 0.03% optimize.opt_a.flash_sp : 0.000019s : 0.01% optimize.opt_a.merge_comm : 0.000019s : 0.01% optimize.opt_a.allreduce_fusion : 0.000017s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000036s : 0.03% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000036s : 0.03% optimize.opt_a.virtual_dataset : 0.000030s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000030s : 0.02% optimize.opt_a.virtual_output : 0.000032s : 0.02% optimize.opt_a.merge_forward : 0.000019s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000039s : 0.03% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000088s : 0.06% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000058s : 0.04% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000020s : 0.01% optimize.opt_a.meta_fg_expand : 0.000014s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000008s : 0.01% optimize.opt_a.receive_attached : 0.000004s : 0.00% optimize.opt_a.after_resolve : 0.000050s : 0.03% optimize.opt_a.a_after_grad : 0.000049s : 0.03% optimize.opt_a.renormalize : 0.001350s : 0.94% optimize.opt_a.add_forward_monad_depend : 0.000016s : 0.01% optimize.opt_a.auto_monad_grad : 0.000006s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000066s : 0.05% optimize.opt_a.cse : 0.000194s : 0.14% optimize.opt_a.a_3 : 0.000239s : 0.17% optimize.py_interpret_to_execute_after_opt_a : 0.000029s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000215s : 0.15% optimize.convert_after_rewriter : 0.000019s : 0.01% optimize.order_py_execute_after_rewriter : 0.000011s : 0.01% optimize.mutable_eliminate : 0.000907s : 0.63% optimize.opt_b.b_1 : 0.000403s : 0.28% optimize.opt_b.b_2 : 0.000020s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000018s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000011s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000015s : 0.01% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000068s : 0.05% optimize.optimize_parallel_all_gather_comm : 0.000038s : 0.03% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000045s : 0.03% optimize.loop_unroll : 0.000673s : 0.47% optimize.opt_after_cconv.c_1 : 0.000116s : 0.08% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000013s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000009s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000012s : 0.01% optimize.opt_after_cconv.cse : 0.000047s : 0.03% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000056s : 0.04% optimize.tuple_transform.d_1 : 0.000112s : 0.08% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000016s : 0.01% optimize.partial_unused_args_eliminate : 0.000003s : 0.00% optimize.add_recomputation : 0.000112s : 0.08% optimize.cse_after_recomputation.cse : 0.000129s : 0.09% optimize.environ_conv : 0.000018s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000039s : 0.03% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000009s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000031s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000008s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000009s : 0.01% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000008s : 0.01% optimize.overlap_grad_flash_sp : 0.000044s : 0.03% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000018s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000032s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000034s : 0.02% optimize.symbol_engine_optimizer.opt_reshape : 0.000019s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000026s : 0.02% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000059s : 0.04% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000009s : 0.01% opt_after_jit_grad : 0.000784s : 0.55% validate : 0.001233s : 0.86% Time group info: ------[substitution.] 0.000582 143 2.04% : 0.000012s : 2: substitution.depend_value_elim 0.81% : 0.000005s : 10: substitution.elim_not_effective 0.63% : 0.000004s : 10: substitution.fold_const_symbol 2.03% : 0.000012s : 13: substitution.graph_param_transform 59.46% : 0.000346s : 4: substitution.inline 1.89% : 0.000011s : 20: substitution.j_node_and_user_rematch 4.37% : 0.000025s : 4: substitution.less_batch_normalization 1.27% : 0.000007s : 6: substitution.load_eliminater 2.71% : 0.000016s : 20: substitution.remove_not_recompute_node 1.48% : 0.000009s : 6: substitution.replace_old_param 5.90% : 0.000034s : 22: substitution.updatestate_pure_node_eliminater 17.41% : 0.000101s : 26: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.015114 2 94.52% : 0.014285s : 1: type_inference.infer 5.48% : 0.000829s : 1: type_inference.specialize ------[replace.] 0.000069 4 100.00% : 0.000069s : 4: replace.inline ------[match.] 0.000343 4 100.00% : 0.000343s : 4: match.inline ------[predicate.] 0.000538 3432 0.95% : 0.000005s : 35: predicate.accumulaten_eliminater 0.88% : 0.000005s : 13: predicate.ad_related_special_op_eliminate 0.65% : 0.000004s : 26: predicate.addn_check_dump 1.10% : 0.000006s : 35: predicate.addn_zero_filter 0.98% : 0.000005s : 35: predicate.adjust_all_reduce_mul_add 2.33% : 0.000013s : 61: predicate.arithmetic_simplify 0.96% : 0.000005s : 35: predicate.cast_eliminate 0.69% : 0.000004s : 26: predicate.check_bprop_eliminate 0.65% : 0.000004s : 26: predicate.compare_switch_simplify 0.19% : 0.000001s : 13: predicate.const_output_eliminate 0.71% : 0.000004s : 26: predicate.depend_value_elim 1.00% : 0.000005s : 35: predicate.dict_get_item_const_eliminator 1.22% : 0.000007s : 35: predicate.dict_get_item_eliminator 0.93% : 0.000005s : 35: predicate.dict_set_item_eliminator 0.97% : 0.000005s : 26: predicate.dumpgradient_eliminate 0.20% : 0.000001s : 13: predicate.elim_not_effective 0.47% : 0.000003s : 13: predicate.elim_shapecalc_of_broadcastargs 1.36% : 0.000007s : 48: predicate.environ_add_const_eliminate 1.28% : 0.000007s : 48: predicate.environ_get_add_eliminate 1.25% : 0.000007s : 48: predicate.environ_get_depend_swap 2.01% : 0.000011s : 74: predicate.environ_get_eliminate 1.26% : 0.000007s : 48: predicate.environ_get_set_eliminate 1.02% : 0.000005s : 39: predicate.exchange_switch_depend_value 1.50% : 0.000008s : 39: predicate.float_depend_g_call 0.76% : 0.000004s : 26: predicate.float_environ_get_switch 0.98% : 0.000005s : 39: predicate.float_tuple_getitem_switch 0.18% : 0.000001s : 13: predicate.fold_const_symbol 0.77% : 0.000004s : 26: predicate.get_grad_eliminate 0.22% : 0.000001s : 13: predicate.graph_param_transform 0.69% : 0.000004s : 26: predicate.incorporate_call 0.63% : 0.000003s : 26: predicate.incorporate_call_switch 5.42% : 0.000029s : 152: predicate.inline 0.79% : 0.000004s : 26: predicate.inline_without_move 0.34% : 0.000002s : 26: predicate.j_node_and_user_rematch 1.20% : 0.000006s : 26: predicate.less_batch_normalization 1.78% : 0.000010s : 61: predicate.list_to_tuple_eliminator_ 2.59% : 0.000014s : 96: predicate.load_eliminater 0.83% : 0.000004s : 13: predicate.loop_unroll_after_grad 1.39% : 0.000007s : 49: predicate.loop_unroll_before_grad 1.73% : 0.000009s : 61: predicate.make_slice_get_slice_eliminator 0.69% : 0.000004s : 26: predicate.merge_addn 0.76% : 0.000004s : 26: predicate.micro_step_allgather_replace 0.72% : 0.000004s : 26: predicate.mini_step_allgather_replace 0.87% : 0.000005s : 35: predicate.minmaximum_grad 1.31% : 0.000007s : 13: predicate.mutable_eliminate 0.36% : 0.000002s : 13: predicate.opt_reshape 0.40% : 0.000002s : 13: predicate.parallel_virtual_node 1.45% : 0.000008s : 39: predicate.partial_defer_inline 1.36% : 0.000007s : 48: predicate.partial_eliminate 0.94% : 0.000005s : 35: predicate.print_const_string_wrapper 0.77% : 0.000004s : 26: predicate.reduce_all_const_elim 1.13% : 0.000006s : 35: predicate.reduce_eliminate 2.48% : 0.000013s : 96: predicate.redundant_stop_gradient_eliminater 0.52% : 0.000003s : 26: predicate.remove_not_recompute_node 1.15% : 0.000006s : 61: predicate.replace_applicator 0.46% : 0.000002s : 26: predicate.replace_old_param 0.28% : 0.000002s : 13: predicate.reset_defer_inline 0.97% : 0.000005s : 35: predicate.reshape_eliminate 0.72% : 0.000004s : 26: predicate.row_tensor_add_zeros_like 0.41% : 0.000002s : 13: predicate.row_tensor_eliminate 0.92% : 0.000005s : 26: predicate.same_eliminate 0.47% : 0.000003s : 32: predicate.set_cell_output_no_recompute 0.90% : 0.000005s : 26: predicate.shard_identity_eliminate 0.87% : 0.000005s : 26: predicate.special_op_eliminate 0.85% : 0.000005s : 26: predicate.specialize_transform 0.87% : 0.000005s : 26: predicate.split_environ_get_set_with_tuple_value 0.78% : 0.000004s : 26: predicate.stack_unstack_eliminate 0.39% : 0.000002s : 13: predicate.switch_call_monad_eliminater 1.07% : 0.000006s : 39: predicate.switch_defer_inline 1.86% : 0.000010s : 65: predicate.switch_layer_defer_inline 3.60% : 0.000019s : 127: predicate.switch_simplify 1.01% : 0.000005s : 35: predicate.tile_eliminate 0.91% : 0.000005s : 35: predicate.transpose_eliminate 1.67% : 0.000009s : 61: predicate.tuple_list_convert_item_index_to_positive 1.70% : 0.000009s : 61: predicate.tuple_list_get_item_const_eliminator 1.94% : 0.000010s : 61: predicate.tuple_list_get_item_depend_reorder 2.93% : 0.000016s : 87: predicate.tuple_list_get_item_eliminator 1.72% : 0.000009s : 61: predicate.tuple_list_get_set_item_eliminator 2.45% : 0.000013s : 87: predicate.tuple_list_set_item_eliminator 1.57% : 0.000008s : 61: predicate.tuple_to_list_eliminator_ 2.64% : 0.000014s : 96: predicate.updatestate_pure_node_eliminater 3.41% : 0.000018s : 122: predicate.updatestate_useless_node_eliminater 0.42% : 0.000002s : 13: predicate.value_based_eliminate 0.81% : 0.000004s : 26: predicate.virtual_dataset_eliminate 0.78% : 0.000004s : 26: predicate.virtual_output_eliminate 0.32% : 0.000002s : 13: predicate.virtual_view_grad_eliminate 0.52% : 0.000003s : 13: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000508 7 5.68% : 0.000029s : 1: func_graph_cloner_run.FuncGraphClonerGraph 17.54% : 0.000089s : 1: func_graph_cloner_run.FuncGraphClonerNode 76.78% : 0.000390s : 5: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.510821 192 0.00% : 0.000004s : 1: ForceFp32Comm 34.50% : 0.176257s : 1: add_attr 34.50% : 0.176236s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.02% : 0.000117s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.05% : 0.000258s : 1: auto_monad 0.01% : 0.000066s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 22.81% : 0.116531s : 1: bootstrap 0.01% : 0.000049s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000034s : 1: control_data_broadcast_order 0.00% : 0.000023s : 1: convert_after_rewriter 0.03% : 0.000152s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000022s : 1: environ_conv 0.01% : 0.000026s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000010s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000012s : 1: label_micro_interleaved_index 0.13% : 0.000686s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.18% : 0.000925s : 1: mutable_eliminate 0.00% : 0.000011s : 1: offloading_packed_experts 0.01% : 0.000030s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000040s : 1: opt.transform.mutable_eliminate 0.58% : 0.002964s : 78: opt.transform.opt_a 0.02% : 0.000113s : 1: opt.transform.opt_after_cconv 0.01% : 0.000063s : 1: opt.transform.opt_after_jit_grad 0.08% : 0.000384s : 28: opt.transform.opt_b 0.02% : 0.000126s : 2: opt.transform.opt_trans_graph 0.02% : 0.000104s : 4: opt.transform.symbol_engine_opt 1.17% : 0.005985s : 1: opt_a 0.05% : 0.000254s : 1: opt_after_cconv 0.16% : 0.000803s : 1: opt_after_jit_grad 0.12% : 0.000603s : 1: opt_b 1.98% : 0.010104s : 1: optimize 0.01% : 0.000042s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000014s : 1: order_py_execute_after_rewriter 0.01% : 0.000051s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000012s : 1: overlap_grad_ring_attention 0.00% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000025s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000012s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000054s : 1: pre_auto_parallel 0.01% : 0.000055s : 1: py_interpret_to_execute 0.01% : 0.000035s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.01% : 0.000062s : 1: remove_dup_value 0.14% : 0.000713s : 1: renormalize.infer 0.12% : 0.000624s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.04% : 0.000227s : 1: rewriter_after_opt_a 0.02% : 0.000112s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000045s : 1: swap_dp_allreduce_reducescatter 0.04% : 0.000181s : 1: symbol_engine_optimizer 0.03% : 0.000159s : 1: tuple_transform 2.98% : 0.015215s : 1: type_inference [WARNING] PIPELINE(166799,ffff961eff30,python3.9):2026-01-29-17:41:53.575.517 [mindspore/ccsrc/frontend/jit/ps/validator.cc:244] CheckAssignReturnValue] Deprecated: the return value of Assign/AssignAdd/AssignSub operator will be removed in subsequent releases. You can modify the code from: a = P.Assign()(param, value) b = a * 2 to: P.Assign()(param, value) b = param * 2 Please check your code:In file /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/auto_generate/gen_ops_def.py:9741, 11~37 return assign_op(variable, value) ^~~~~~~~~~~~~~~~~~~~~~~~~~ TotalTime = 0.34776, [21] [bootstrap]: 0.00060092 [type_inference]: 0.200051 [event_method]: 1.652e-05 [auto_monad]: 0.00022581 [graph_reusing]: 6.64001e-06 [inline]: 3.25e-06 [add_attr]: 0.0047643, [1] [add_attr_with_inline]: 0.00475168, [1] [Cycle 1]: 7.679e-05, [2] [tag_attr]: 2.51e-05 [meta_addattr_fg_expand]: 6.64999e-06 [parallel-infer-symbol]: 3.63e-06 [pre_auto_parallel]: 4.767e-05 [insert-virtual-dataset]: 2.91e-06 [parallel-infer-symbol-second]: 7.09988e-07 [dataset_repeat_opt]: 2.53e-06 [pipeline_split]: 1.77001e-06 [optimize]: 0.139815, [53] [py_interpret_to_execute]: 3.8e-05 [rewriter_before_opt_a]: 9.928e-05 [opt_a]: 0.00542449, [2] [Cycle 1]: 0.0039295, [45] [expand_dump_flag]: 3.81999e-06 [switch_simplify]: 4.529e-05 [loop_unroll]: 3.331e-05 [a_1]: 0.0011834 [with_stream_mark]: 2.865e-05 [recompute_prepare]: 2.394e-05 [updatestate_depend_eliminate]: 7.379e-05 [updatestate_assign_eliminate]: 5.156e-05 [updatestate_loads_eliminate]: 1.534e-05 [parameter_eliminate]: 2.56e-06 [a_2]: 0.0002357 [accelerated_algorithm]: 3.995e-05 [shard]: 2.36e-06 [meta_shard_fg_expand]: 4.23999e-06 [shard_inline]: 1.481e-05 [merge_send_recv]: 1.464e-05 [auto_parallel]: 1.434e-05 [parallel]: 2.56e-05 [flash_sp]: 1.176e-05 [merge_comm]: 1.022e-05 [allreduce_fusion]: 8.82e-06 [matmul_add_comm_reduction]: 1.67e-05 [allreduce_slice_to_reducescatter]: 7.2e-07 [virtual_shard_identity]: 1.892e-05 [virtual_dataset]: 1.464e-05 [get_grad_eliminate_]: 1.491e-05 [virtual_output]: 1.532e-05 [merge_forward]: 9.11998e-06 [cell_reuse_recompute_pass]: 1.14e-06 [offload_activation]: 1.778e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.987e-05 [merge_recompute_call_nodes]: 1.50999e-06 [before_grad]: 2.72e-05 [set_forward_comm_id_for_comm_node_pass]: 8.82e-06 [meta_fg_expand]: 6.62002e-06 [flash_sp_send_recv_attached]: 4.48999e-06 [receive_attached]: 3.01001e-06 [after_resolve]: 2.148e-05 [a_after_grad]: 2.425e-05 [renormalize]: 0.00128517 [add_forward_monad_depend]: 9.72001e-06 [auto_monad_grad]: 2.71999e-06 [auto_monad_eliminator]: 3.182e-05 [cse]: 0.00011639 [a_3]: 0.00011551 [Cycle 2]: 0.00148059, [45] [expand_dump_flag]: 2.44999e-06 [switch_simplify]: 1.647e-05 [loop_unroll]: 1.463e-05 [a_1]: 0.00043923 [with_stream_mark]: 1.905e-05 [recompute_prepare]: 1.542e-05 [updatestate_depend_eliminate]: 9.56e-06 [updatestate_assign_eliminate]: 1.037e-05 [updatestate_loads_eliminate]: 1.393e-05 [parameter_eliminate]: 1.70001e-06 [a_2]: 0.00021839 [accelerated_algorithm]: 2.153e-05 [shard]: 2.07999e-06 [meta_shard_fg_expand]: 3.72002e-06 [shard_inline]: 1.517e-05 [merge_send_recv]: 1.413e-05 [auto_parallel]: 1.346e-05 [parallel]: 8.47e-06 [flash_sp]: 4.77e-06 [merge_comm]: 8.84e-06 [allreduce_fusion]: 8.47e-06 [matmul_add_comm_reduction]: 1.606e-05 [allreduce_slice_to_reducescatter]: 6.89994e-07 [virtual_shard_identity]: 1.649e-05 [virtual_dataset]: 1.443e-05 [get_grad_eliminate_]: 1.439e-05 [virtual_output]: 1.461e-05 [merge_forward]: 8.85001e-06 [cell_reuse_recompute_pass]: 2.34999e-06 [offload_activation]: 1.743e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.943e-05 [merge_recompute_call_nodes]: 1.25999e-06 [before_grad]: 2.465e-05 [set_forward_comm_id_for_comm_node_pass]: 8.80001e-06 [meta_fg_expand]: 6.10002e-06 [flash_sp_send_recv_attached]: 1.62001e-06 [receive_attached]: 2.16e-06 [after_resolve]: 2.253e-05 [a_after_grad]: 2.466e-05 [renormalize]: 5.9983e-08 [add_forward_monad_depend]: 3.14001e-06 [auto_monad_grad]: 1.96998e-06 [auto_monad_eliminator]: 2.249e-05 [cse]: 4.578e-05 [a_3]: 0.00010118 [py_interpret_to_execute_after_opt_a]: 2.344e-05 [slice_cell_reuse_recomputed_activation]: 2.11998e-06 [rewriter_after_opt_a]: 0.00018422 [convert_after_rewriter]: 1.823e-05 [order_py_execute_after_rewriter]: 1.118e-05 [mutable_eliminate]: 0.131276 [opt_b]: 0.00060881, [1] [Cycle 1]: 0.00059624, [7] [b_1]: 0.00039741 [b_2]: 2.051e-05 [updatestate_depend_eliminate]: 2.069e-05 [updatestate_assign_eliminate]: 1.197e-05 [updatestate_loads_eliminate]: 1.558e-05 [renormalize]: 1.54998e-06 [cse]: 7.887e-05 [optimize_parallel_all_gather_comm]: 4.263e-05 [overlap_param_gather]: 2.04999e-06 [cconv]: 4.544e-05 [loop_unroll]: 0.00070287 [opt_after_cconv]: 0.00027038, [1] [Cycle 1]: 0.00026019, [7] [c_1]: 0.00012128 [parameter_eliminate]: 7.33999e-06 [updatestate_depend_eliminate]: 1.567e-05 [updatestate_assign_eliminate]: 1.069e-05 [updatestate_loads_eliminate]: 1.211e-05 [cse]: 5.004e-05 [renormalize]: 8.89995e-07 [remove_dup_value]: 5.699e-05 [tuple_transform]: 0.00016844, [1] [Cycle 1]: 0.00016217, [4] [d_1]: 0.00011788 [none_parameter_eliminate]: 2.11e-06 [renormalize]: 3.30008e-07 [switch_simplify]: 1.832e-05 [partial_unused_args_eliminate]: 2.91e-06 [add_recomputation]: 0.00011923 [cse_after_recomputation]: 4.912e-05, [1] [Cycle 1]: 4.334e-05, [1] [cse]: 3.512e-05 [environ_conv]: 1.765e-05 [swap_dp_allreduce_reducescatter]: 1.364e-05 [bias_add_comm_swap]: 3.85e-06 [label_micro_interleaved_index]: 8.3e-06 [label_fine_grained_interleaved_index]: 2.74999e-06 [merge_cast_opt]: 1.84e-06 [slice_recompute_activation]: 2.57001e-06 [micro_interleaved_order_control]: 3.01999e-06 [assign_add_opt]: 1.45999e-06 [ForceFp32Comm]: 1.04e-06 [remove_cast_before_assign_add]: 1.17e-06 [full_micro_interleaved_order_control]: 2.52001e-06 [reorder_send_recv_between_fp_bp]: 3.97e-06 [comm_op_add_attrs]: 1.12999e-06 [add_comm_op_reuse_tag]: 1.04e-06 [interleave_split_concat_branches]: 1.27e-06 [interleave_parallel_branches]: 1.22999e-06 [overlap_opt_shard_in_pipeline]: 1.29998e-06 [overlap_opt_shard_grad_in_pipeline]: 1.81e-06 [control_data_broadcast_order]: 3.113e-05 [grouped_pairwise_exchange_alltoall]: 1.72999e-06 [offloading_packed_experts]: 9.36e-06 [overlap_recompute_and_grad_model_parallel]: 8.38999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.24e-06 [overlap_recompute_allgather_and_fa_grad]: 1.45999e-06 [overlap_recompute_comm]: 2.51e-06 [overlap_grad_ring_attention]: 7.92e-06 [overlap_grad_flash_sp]: 4.135e-05 [begin_end_overlap_inline]: 5.09986e-07 [split_matmul_comm_elemetwise]: 2.56e-06 [split_layernorm_comm]: 1.94e-06 [handle_group_info]: 1.34e-06 [symbol_engine_optimizer]: 0.00019626, [1] [Cycle 1]: 0.00018952, [6] [build]: 1.826e-05 [elim_shapecalc]: 2.824e-05 [elim_not_effective]: 5.442e-05 [opt_reshape]: 1.991e-05 [fold_const_symbol]: 2.549e-05 [renormalize]: 2.10013e-07 [detach_backward]: 2.81e-06 [pipeline_parallel_scheduler]: 1.42e-06 [auto_monad_reorder]: 5.8e-05 [get_jit_bprop_graph]: 2.48e-06 [rewriter_after_jit_bprop_graph]: 8.3e-06 [opt_after_jit_grad]: 0.00075896 [validate]: 0.00118426 Sums bootstrap : 0.000601s : 0.18% type_inference : 0.200051s : 58.53% event_method : 0.000017s : 0.00% auto_monad : 0.000226s : 0.07% graph_reusing : 0.000007s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000025s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000007s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000048s : 0.01% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000038s : 0.01% optimize.rewriter_before_opt_a : 0.000099s : 0.03% optimize.opt_a.expand_dump_flag : 0.000006s : 0.00% optimize.opt_a.switch_simplify : 0.000062s : 0.02% optimize.opt_a.loop_unroll : 0.000048s : 0.01% optimize.opt_a.a_1 : 0.001623s : 0.47% optimize.opt_a.with_stream_mark : 0.000048s : 0.01% optimize.opt_a.recompute_prepare : 0.000039s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000083s : 0.02% optimize.opt_a.updatestate_assign_eliminate : 0.000062s : 0.02% optimize.opt_a.updatestate_loads_eliminate : 0.000029s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.00% optimize.opt_a.a_2 : 0.000454s : 0.13% optimize.opt_a.accelerated_algorithm : 0.000061s : 0.02% optimize.opt_a.shard : 0.000004s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000008s : 0.00% optimize.opt_a.shard_inline : 0.000030s : 0.01% optimize.opt_a.merge_send_recv : 0.000029s : 0.01% optimize.opt_a.auto_parallel : 0.000028s : 0.01% optimize.opt_a.parallel : 0.000034s : 0.01% optimize.opt_a.flash_sp : 0.000017s : 0.00% optimize.opt_a.merge_comm : 0.000019s : 0.01% optimize.opt_a.allreduce_fusion : 0.000017s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000033s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000035s : 0.01% optimize.opt_a.virtual_dataset : 0.000029s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000029s : 0.01% optimize.opt_a.virtual_output : 0.000030s : 0.01% optimize.opt_a.merge_forward : 0.000018s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000003s : 0.00% optimize.opt_a.offload_activation : 0.000035s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000059s : 0.02% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000052s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000018s : 0.01% optimize.opt_a.meta_fg_expand : 0.000013s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000006s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000044s : 0.01% optimize.opt_a.a_after_grad : 0.000049s : 0.01% optimize.opt_a.renormalize : 0.001285s : 0.38% optimize.opt_a.add_forward_monad_depend : 0.000013s : 0.00% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000054s : 0.02% optimize.opt_a.cse : 0.000162s : 0.05% optimize.opt_a.a_3 : 0.000217s : 0.06% optimize.py_interpret_to_execute_after_opt_a : 0.000023s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000184s : 0.05% optimize.convert_after_rewriter : 0.000018s : 0.01% optimize.order_py_execute_after_rewriter : 0.000011s : 0.00% optimize.mutable_eliminate : 0.131276s : 38.41% optimize.opt_b.b_1 : 0.000397s : 0.12% optimize.opt_b.b_2 : 0.000021s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000021s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000012s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000016s : 0.00% optimize.opt_b.renormalize : 0.000002s : 0.00% optimize.opt_b.cse : 0.000079s : 0.02% optimize.optimize_parallel_all_gather_comm : 0.000043s : 0.01% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000045s : 0.01% optimize.loop_unroll : 0.000703s : 0.21% optimize.opt_after_cconv.c_1 : 0.000121s : 0.04% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000016s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000011s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000012s : 0.00% optimize.opt_after_cconv.cse : 0.000050s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000057s : 0.02% optimize.tuple_transform.d_1 : 0.000118s : 0.03% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000018s : 0.01% optimize.partial_unused_args_eliminate : 0.000003s : 0.00% optimize.add_recomputation : 0.000119s : 0.03% optimize.cse_after_recomputation.cse : 0.000035s : 0.01% optimize.environ_conv : 0.000018s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000014s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000008s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000004s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000031s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000009s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000008s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000008s : 0.00% optimize.overlap_grad_flash_sp : 0.000041s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000018s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000028s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000054s : 0.02% optimize.symbol_engine_optimizer.opt_reshape : 0.000020s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000025s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000001s : 0.00% auto_monad_reorder : 0.000058s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.000759s : 0.22% validate : 0.001184s : 0.35% Time group info: ------[substitution.] 0.000527 143 1.80% : 0.000009s : 2: substitution.depend_value_elim 0.94% : 0.000005s : 10: substitution.elim_not_effective 0.69% : 0.000004s : 10: substitution.fold_const_symbol 2.23% : 0.000012s : 13: substitution.graph_param_transform 59.65% : 0.000314s : 4: substitution.inline 1.83% : 0.000010s : 20: substitution.j_node_and_user_rematch 4.11% : 0.000022s : 4: substitution.less_batch_normalization 1.01% : 0.000005s : 6: substitution.load_eliminater 2.72% : 0.000014s : 20: substitution.remove_not_recompute_node 1.44% : 0.000008s : 6: substitution.replace_old_param 5.90% : 0.000031s : 22: substitution.updatestate_pure_node_eliminater 17.69% : 0.000093s : 26: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.199984 2 99.65% : 0.199280s : 1: type_inference.infer 0.35% : 0.000705s : 1: type_inference.specialize ------[replace.] 0.000066 4 100.00% : 0.000066s : 4: replace.inline ------[match.] 0.000311 4 100.00% : 0.000311s : 4: match.inline ------[predicate.] 0.000523 3432 0.93% : 0.000005s : 35: predicate.accumulaten_eliminater 0.91% : 0.000005s : 13: predicate.ad_related_special_op_eliminate 0.74% : 0.000004s : 26: predicate.addn_check_dump 1.05% : 0.000005s : 35: predicate.addn_zero_filter 0.89% : 0.000005s : 35: predicate.adjust_all_reduce_mul_add 2.14% : 0.000011s : 61: predicate.arithmetic_simplify 1.00% : 0.000005s : 35: predicate.cast_eliminate 0.73% : 0.000004s : 26: predicate.check_bprop_eliminate 0.72% : 0.000004s : 26: predicate.compare_switch_simplify 0.19% : 0.000001s : 13: predicate.const_output_eliminate 0.71% : 0.000004s : 26: predicate.depend_value_elim 0.97% : 0.000005s : 35: predicate.dict_get_item_const_eliminator 1.03% : 0.000005s : 35: predicate.dict_get_item_eliminator 0.91% : 0.000005s : 35: predicate.dict_set_item_eliminator 1.03% : 0.000005s : 26: predicate.dumpgradient_eliminate 0.25% : 0.000001s : 13: predicate.elim_not_effective 0.50% : 0.000003s : 13: predicate.elim_shapecalc_of_broadcastargs 1.26% : 0.000007s : 48: predicate.environ_add_const_eliminate 1.24% : 0.000006s : 48: predicate.environ_get_add_eliminate 1.23% : 0.000006s : 48: predicate.environ_get_depend_swap 1.95% : 0.000010s : 74: predicate.environ_get_eliminate 1.24% : 0.000006s : 48: predicate.environ_get_set_eliminate 0.99% : 0.000005s : 39: predicate.exchange_switch_depend_value 1.50% : 0.000008s : 39: predicate.float_depend_g_call 0.69% : 0.000004s : 26: predicate.float_environ_get_switch 1.01% : 0.000005s : 39: predicate.float_tuple_getitem_switch 0.19% : 0.000001s : 13: predicate.fold_const_symbol 0.75% : 0.000004s : 26: predicate.get_grad_eliminate 0.28% : 0.000001s : 13: predicate.graph_param_transform 0.73% : 0.000004s : 26: predicate.incorporate_call 0.63% : 0.000003s : 26: predicate.incorporate_call_switch 5.49% : 0.000029s : 152: predicate.inline 0.92% : 0.000005s : 26: predicate.inline_without_move 0.36% : 0.000002s : 26: predicate.j_node_and_user_rematch 1.14% : 0.000006s : 26: predicate.less_batch_normalization 1.65% : 0.000009s : 61: predicate.list_to_tuple_eliminator_ 2.44% : 0.000013s : 96: predicate.load_eliminater 1.02% : 0.000005s : 13: predicate.loop_unroll_after_grad 1.41% : 0.000007s : 49: predicate.loop_unroll_before_grad 1.80% : 0.000009s : 61: predicate.make_slice_get_slice_eliminator 0.72% : 0.000004s : 26: predicate.merge_addn 0.69% : 0.000004s : 26: predicate.micro_step_allgather_replace 0.69% : 0.000004s : 26: predicate.mini_step_allgather_replace 0.85% : 0.000004s : 35: predicate.minmaximum_grad 2.09% : 0.000011s : 13: predicate.mutable_eliminate 0.58% : 0.000003s : 13: predicate.opt_reshape 0.45% : 0.000002s : 13: predicate.parallel_virtual_node 1.31% : 0.000007s : 39: predicate.partial_defer_inline 1.37% : 0.000007s : 48: predicate.partial_eliminate 0.89% : 0.000005s : 35: predicate.print_const_string_wrapper 0.69% : 0.000004s : 26: predicate.reduce_all_const_elim 1.14% : 0.000006s : 35: predicate.reduce_eliminate 2.43% : 0.000013s : 96: predicate.redundant_stop_gradient_eliminater 0.54% : 0.000003s : 26: predicate.remove_not_recompute_node 1.09% : 0.000006s : 61: predicate.replace_applicator 0.47% : 0.000002s : 26: predicate.replace_old_param 0.35% : 0.000002s : 13: predicate.reset_defer_inline 0.92% : 0.000005s : 35: predicate.reshape_eliminate 0.68% : 0.000004s : 26: predicate.row_tensor_add_zeros_like 0.46% : 0.000002s : 13: predicate.row_tensor_eliminate 0.80% : 0.000004s : 26: predicate.same_eliminate 0.50% : 0.000003s : 32: predicate.set_cell_output_no_recompute 0.87% : 0.000005s : 26: predicate.shard_identity_eliminate 0.78% : 0.000004s : 26: predicate.special_op_eliminate 0.80% : 0.000004s : 26: predicate.specialize_transform 0.81% : 0.000004s : 26: predicate.split_environ_get_set_with_tuple_value 0.83% : 0.000004s : 26: predicate.stack_unstack_eliminate 0.41% : 0.000002s : 13: predicate.switch_call_monad_eliminater 1.08% : 0.000006s : 39: predicate.switch_defer_inline 1.80% : 0.000009s : 65: predicate.switch_layer_defer_inline 3.67% : 0.000019s : 127: predicate.switch_simplify 0.91% : 0.000005s : 35: predicate.tile_eliminate 0.96% : 0.000005s : 35: predicate.transpose_eliminate 1.75% : 0.000009s : 61: predicate.tuple_list_convert_item_index_to_positive 1.69% : 0.000009s : 61: predicate.tuple_list_get_item_const_eliminator 1.62% : 0.000008s : 61: predicate.tuple_list_get_item_depend_reorder 2.90% : 0.000015s : 87: predicate.tuple_list_get_item_eliminator 1.85% : 0.000010s : 61: predicate.tuple_list_get_set_item_eliminator 2.57% : 0.000013s : 87: predicate.tuple_list_set_item_eliminator 1.68% : 0.000009s : 61: predicate.tuple_to_list_eliminator_ 2.52% : 0.000013s : 96: predicate.updatestate_pure_node_eliminater 3.39% : 0.000018s : 122: predicate.updatestate_useless_node_eliminater 0.40% : 0.000002s : 13: predicate.value_based_eliminate 0.72% : 0.000004s : 26: predicate.virtual_dataset_eliminate 0.76% : 0.000004s : 26: predicate.virtual_output_eliminate 0.36% : 0.000002s : 13: predicate.virtual_view_grad_eliminate 0.53% : 0.000003s : 13: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000450 7 6.60% : 0.000030s : 1: func_graph_cloner_run.FuncGraphClonerGraph 18.28% : 0.000082s : 1: func_graph_cloner_run.FuncGraphClonerNode 75.12% : 0.000338s : 5: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.495972 192 0.00% : 0.000004s : 1: ForceFp32Comm 0.96% : 0.004771s : 1: add_attr 0.96% : 0.004756s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.03% : 0.000126s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.05% : 0.000235s : 1: auto_monad 0.01% : 0.000064s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000008s : 1: bias_add_comm_swap 0.13% : 0.000638s : 1: bootstrap 0.01% : 0.000051s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000036s : 1: control_data_broadcast_order 0.00% : 0.000022s : 1: convert_after_rewriter 0.01% : 0.000052s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000022s : 1: environ_conv 0.00% : 0.000023s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000010s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000011s : 1: label_micro_interleaved_index 0.14% : 0.000719s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 26.47% : 0.131297s : 1: mutable_eliminate 0.00% : 0.000012s : 1: offloading_packed_experts 0.01% : 0.000035s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000055s : 1: opt.transform.mutable_eliminate 0.56% : 0.002757s : 78: opt.transform.opt_a 0.02% : 0.000119s : 1: opt.transform.opt_after_cconv 0.01% : 0.000061s : 1: opt.transform.opt_after_jit_grad 0.08% : 0.000376s : 28: opt.transform.opt_b 0.03% : 0.000132s : 2: opt.transform.opt_trans_graph 0.02% : 0.000123s : 4: opt.transform.symbol_engine_opt 1.09% : 0.005428s : 1: opt_a 0.06% : 0.000275s : 1: opt_after_cconv 0.16% : 0.000775s : 1: opt_after_jit_grad 0.12% : 0.000615s : 1: opt_b 28.19% : 0.139821s : 1: optimize 0.01% : 0.000047s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000015s : 1: order_py_execute_after_rewriter 0.01% : 0.000046s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000012s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000012s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000007s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000052s : 1: pre_auto_parallel 0.01% : 0.000043s : 1: py_interpret_to_execute 0.01% : 0.000027s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.01% : 0.000063s : 1: remove_dup_value 0.15% : 0.000724s : 1: renormalize.infer 0.11% : 0.000551s : 1: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000011s : 1: rewriter_after_jit_bprop_graph 0.04% : 0.000192s : 1: rewriter_after_opt_a 0.02% : 0.000104s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000017s : 1: swap_dp_allreduce_reducescatter 0.04% : 0.000200s : 1: symbol_engine_optimizer 0.03% : 0.000172s : 1: tuple_transform 40.34% : 0.200078s : 1: type_inference [WARNING] PIPELINE(166799,ffff961eff30,python3.9):2026-01-29-17:41:59.402.883 [mindspore/ccsrc/frontend/jit/ps/validator.cc:244] CheckAssignReturnValue] Deprecated: the return value of Assign/AssignAdd/AssignSub operator will be removed in subsequent releases. You can modify the code from: a = P.Assign()(param, value) b = a * 2 to: P.Assign()(param, value) b = param * 2 Please check your code:In file /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/auto_generate/gen_ops_def.py:9741, 11~37 return assign_op(variable, value) ^~~~~~~~~~~~~~~~~~~~~~~~~~ TotalTime = 0.355939, [21] [bootstrap]: 0.00052827 [type_inference]: 0.17121 [event_method]: 1.78e-05 [auto_monad]: 0.00023044 [graph_reusing]: 6.63e-06 [inline]: 2.49999e-06 [add_attr]: 0.00421998, [1] [add_attr_with_inline]: 0.00420573, [1] [Cycle 1]: 7.524e-05, [2] [tag_attr]: 2.565e-05 [meta_addattr_fg_expand]: 6.42001e-06 [parallel-infer-symbol]: 4.36002e-06 [pre_auto_parallel]: 4.564e-05 [insert-virtual-dataset]: 2.73e-06 [parallel-infer-symbol-second]: 8.29983e-07 [dataset_repeat_opt]: 2.29001e-06 [pipeline_split]: 1.77999e-06 [optimize]: 0.00940864, [53] [py_interpret_to_execute]: 3.32e-05 [rewriter_before_opt_a]: 9.509e-05 [opt_a]: 0.00570024, [2] [Cycle 1]: 0.00419729, [45] [expand_dump_flag]: 3.43e-06 [switch_simplify]: 4.709e-05 [loop_unroll]: 3.277e-05 [a_1]: 0.00119069 [with_stream_mark]: 3.231e-05 [recompute_prepare]: 2.65e-05 [updatestate_depend_eliminate]: 7.737e-05 [updatestate_assign_eliminate]: 0.00013375 [updatestate_loads_eliminate]: 1.696e-05 [parameter_eliminate]: 2.58e-06 [a_2]: 0.00028136 [accelerated_algorithm]: 4.253e-05 [shard]: 3.52997e-06 [meta_shard_fg_expand]: 4.71002e-06 [shard_inline]: 1.554e-05 [merge_send_recv]: 1.633e-05 [auto_parallel]: 1.747e-05 [parallel]: 2.617e-05 [flash_sp]: 1.382e-05 [merge_comm]: 1e-05 [allreduce_fusion]: 9.05999e-06 [matmul_add_comm_reduction]: 1.656e-05 [allreduce_slice_to_reducescatter]: 8.80013e-07 [virtual_shard_identity]: 1.803e-05 [virtual_dataset]: 1.544e-05 [get_grad_eliminate_]: 2.106e-05 [virtual_output]: 1.733e-05 [merge_forward]: 1.058e-05 [cell_reuse_recompute_pass]: 1.98002e-06 [offload_activation]: 2.05e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.199e-05 [merge_recompute_call_nodes]: 1.98002e-06 [before_grad]: 2.82e-05 [set_forward_comm_id_for_comm_node_pass]: 1.059e-05 [meta_fg_expand]: 7.68999e-06 [flash_sp_send_recv_attached]: 7.01001e-06 [receive_attached]: 2.21e-06 [after_resolve]: 2.412e-05 [a_after_grad]: 2.509e-05 [renormalize]: 0.00129195 [add_forward_monad_depend]: 8.62e-06 [auto_monad_grad]: 3.14999e-06 [auto_monad_eliminator]: 3.354e-05 [cse]: 0.00012248 [a_3]: 0.00011796 [Cycle 2]: 0.0014893, [45] [expand_dump_flag]: 2.76999e-06 [switch_simplify]: 1.773e-05 [loop_unroll]: 1.472e-05 [a_1]: 0.00044461 [with_stream_mark]: 2.263e-05 [recompute_prepare]: 1.541e-05 [updatestate_depend_eliminate]: 9.86e-06 [updatestate_assign_eliminate]: 1.165e-05 [updatestate_loads_eliminate]: 1.256e-05 [parameter_eliminate]: 1.94999e-06 [a_2]: 0.00021857 [accelerated_algorithm]: 2.2e-05 [shard]: 1.76998e-06 [meta_shard_fg_expand]: 4.18001e-06 [shard_inline]: 1.508e-05 [merge_send_recv]: 1.468e-05 [auto_parallel]: 1.505e-05 [parallel]: 9.02999e-06 [flash_sp]: 4.80001e-06 [merge_comm]: 1.004e-05 [allreduce_fusion]: 8.64998e-06 [matmul_add_comm_reduction]: 1.649e-05 [allreduce_slice_to_reducescatter]: 1.09e-06 [virtual_shard_identity]: 1.755e-05 [virtual_dataset]: 1.453e-05 [get_grad_eliminate_]: 1.406e-05 [virtual_output]: 1.472e-05 [merge_forward]: 8.75001e-06 [cell_reuse_recompute_pass]: 2.31e-06 [offload_activation]: 1.768e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.763e-05 [merge_recompute_call_nodes]: 1.79e-06 [before_grad]: 2.536e-05 [set_forward_comm_id_for_comm_node_pass]: 9.32001e-06 [meta_fg_expand]: 6.98998e-06 [flash_sp_send_recv_attached]: 1.62999e-06 [receive_attached]: 2.63e-06 [after_resolve]: 2.104e-05 [a_after_grad]: 2.367e-05 [renormalize]: 1.19995e-07 [add_forward_monad_depend]: 2.09999e-06 [auto_monad_grad]: 2.03997e-06 [auto_monad_eliminator]: 2.162e-05 [cse]: 4.477e-05 [a_3]: 9.773e-05 [py_interpret_to_execute_after_opt_a]: 2.171e-05 [slice_cell_reuse_recomputed_activation]: 2.64999e-06 [rewriter_after_opt_a]: 0.00024853 [convert_after_rewriter]: 1.663e-05 [order_py_execute_after_rewriter]: 1.144e-05 [mutable_eliminate]: 0.00082474 [opt_b]: 0.00053421, [1] [Cycle 1]: 0.00052481, [7] [b_1]: 0.00036604 [b_2]: 1.776e-05 [updatestate_depend_eliminate]: 1.594e-05 [updatestate_assign_eliminate]: 1.06e-05 [updatestate_loads_eliminate]: 1.286e-05 [renormalize]: 7.89994e-07 [cse]: 5.79e-05 [optimize_parallel_all_gather_comm]: 0.00012219 [overlap_param_gather]: 3.25e-06 [cconv]: 3.734e-05 [loop_unroll]: 0.00057156 [opt_after_cconv]: 0.00024672, [1] [Cycle 1]: 0.00023902, [7] [c_1]: 0.00011373 [parameter_eliminate]: 5.95002e-06 [updatestate_depend_eliminate]: 1.366e-05 [updatestate_assign_eliminate]: 9.50001e-06 [updatestate_loads_eliminate]: 1.081e-05 [cse]: 4.547e-05 [renormalize]: 1.04e-06 [remove_dup_value]: 2.424e-05 [tuple_transform]: 0.00015488, [1] [Cycle 1]: 0.00014935, [4] [d_1]: 0.00011031 [none_parameter_eliminate]: 1.97001e-06 [renormalize]: 1.60013e-07 [switch_simplify]: 1.649e-05 [partial_unused_args_eliminate]: 1.91003e-06 [add_recomputation]: 0.00011185 [cse_after_recomputation]: 4.529e-05, [1] [Cycle 1]: 3.938e-05, [1] [cse]: 3.346e-05 [environ_conv]: 1.613e-05 [swap_dp_allreduce_reducescatter]: 1.217e-05 [bias_add_comm_swap]: 3.23e-06 [label_micro_interleaved_index]: 6.52001e-06 [label_fine_grained_interleaved_index]: 3.64002e-06 [merge_cast_opt]: 1.47001e-06 [slice_recompute_activation]: 2.21e-06 [micro_interleaved_order_control]: 2.64999e-06 [assign_add_opt]: 1.47001e-06 [ForceFp32Comm]: 1.03001e-06 [remove_cast_before_assign_add]: 1.04e-06 [full_micro_interleaved_order_control]: 2.38998e-06 [reorder_send_recv_between_fp_bp]: 2.78e-06 [comm_op_add_attrs]: 1.44998e-06 [add_comm_op_reuse_tag]: 1.27e-06 [interleave_split_concat_branches]: 1.16002e-06 [interleave_parallel_branches]: 1.06002e-06 [overlap_opt_shard_in_pipeline]: 1.24003e-06 [overlap_opt_shard_grad_in_pipeline]: 1.67999e-06 [control_data_broadcast_order]: 2.95e-05 [grouped_pairwise_exchange_alltoall]: 1.66e-06 [offloading_packed_experts]: 8.06001e-06 [overlap_recompute_and_grad_model_parallel]: 8.50999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.45999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.37999e-06 [overlap_recompute_comm]: 2.43e-06 [overlap_grad_ring_attention]: 7.8e-06 [overlap_grad_flash_sp]: 4.67e-05 [begin_end_overlap_inline]: 5.40022e-07 [split_matmul_comm_elemetwise]: 2.41e-06 [split_layernorm_comm]: 1.67001e-06 [handle_group_info]: 1.05999e-06 [symbol_engine_optimizer]: 0.00015926, [1] [Cycle 1]: 0.00015371, [6] [build]: 1.842e-05 [elim_shapecalc]: 2.364e-05 [elim_not_effective]: 2.961e-05 [opt_reshape]: 1.969e-05 [fold_const_symbol]: 2.62e-05 [renormalize]: 1.69995e-07 [detach_backward]: 2.42001e-06 [pipeline_parallel_scheduler]: 1.49e-06 [auto_monad_reorder]: 5.321e-05 [get_jit_bprop_graph]: 2.15002e-06 [rewriter_after_jit_bprop_graph]: 7.7e-06 [opt_after_jit_grad]: 0.168697 [validate]: 0.00126858 Sums bootstrap : 0.000528s : 0.15% type_inference : 0.171210s : 48.85% event_method : 0.000018s : 0.01% auto_monad : 0.000230s : 0.07% graph_reusing : 0.000007s : 0.00% inline : 0.000002s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000026s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000006s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000046s : 0.01% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000033s : 0.01% optimize.rewriter_before_opt_a : 0.000095s : 0.03% optimize.opt_a.expand_dump_flag : 0.000006s : 0.00% optimize.opt_a.switch_simplify : 0.000065s : 0.02% optimize.opt_a.loop_unroll : 0.000047s : 0.01% optimize.opt_a.a_1 : 0.001635s : 0.47% optimize.opt_a.with_stream_mark : 0.000055s : 0.02% optimize.opt_a.recompute_prepare : 0.000042s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000087s : 0.02% optimize.opt_a.updatestate_assign_eliminate : 0.000145s : 0.04% optimize.opt_a.updatestate_loads_eliminate : 0.000030s : 0.01% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000500s : 0.14% optimize.opt_a.accelerated_algorithm : 0.000065s : 0.02% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000009s : 0.00% optimize.opt_a.shard_inline : 0.000031s : 0.01% optimize.opt_a.merge_send_recv : 0.000031s : 0.01% optimize.opt_a.auto_parallel : 0.000033s : 0.01% optimize.opt_a.parallel : 0.000035s : 0.01% optimize.opt_a.flash_sp : 0.000019s : 0.01% optimize.opt_a.merge_comm : 0.000020s : 0.01% optimize.opt_a.allreduce_fusion : 0.000018s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000033s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000036s : 0.01% optimize.opt_a.virtual_dataset : 0.000030s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000035s : 0.01% optimize.opt_a.virtual_output : 0.000032s : 0.01% optimize.opt_a.merge_forward : 0.000019s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.00% optimize.opt_a.offload_activation : 0.000038s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000060s : 0.02% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000054s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000020s : 0.01% optimize.opt_a.meta_fg_expand : 0.000015s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000009s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000045s : 0.01% optimize.opt_a.a_after_grad : 0.000049s : 0.01% optimize.opt_a.renormalize : 0.001292s : 0.37% optimize.opt_a.add_forward_monad_depend : 0.000011s : 0.00% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000055s : 0.02% optimize.opt_a.cse : 0.000167s : 0.05% optimize.opt_a.a_3 : 0.000216s : 0.06% optimize.py_interpret_to_execute_after_opt_a : 0.000022s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000249s : 0.07% optimize.convert_after_rewriter : 0.000017s : 0.00% optimize.order_py_execute_after_rewriter : 0.000011s : 0.00% optimize.mutable_eliminate : 0.000825s : 0.24% optimize.opt_b.b_1 : 0.000366s : 0.10% optimize.opt_b.b_2 : 0.000018s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000016s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000011s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000013s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000058s : 0.02% optimize.optimize_parallel_all_gather_comm : 0.000122s : 0.03% optimize.overlap_param_gather : 0.000003s : 0.00% optimize.cconv : 0.000037s : 0.01% optimize.loop_unroll : 0.000572s : 0.16% optimize.opt_after_cconv.c_1 : 0.000114s : 0.03% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000014s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000010s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000011s : 0.00% optimize.opt_after_cconv.cse : 0.000045s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000024s : 0.01% optimize.tuple_transform.d_1 : 0.000110s : 0.03% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000016s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000112s : 0.03% optimize.cse_after_recomputation.cse : 0.000033s : 0.01% optimize.environ_conv : 0.000016s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000012s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000007s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000004s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000030s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000008s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000009s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000008s : 0.00% optimize.overlap_grad_flash_sp : 0.000047s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000018s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000024s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000030s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000020s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000026s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000001s : 0.00% auto_monad_reorder : 0.000053s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.168697s : 48.13% validate : 0.001269s : 0.36% Time group info: ------[substitution.] 0.000520 143 2.16% : 0.000011s : 2: substitution.depend_value_elim 0.82% : 0.000004s : 10: substitution.elim_not_effective 0.75% : 0.000004s : 10: substitution.fold_const_symbol 2.31% : 0.000012s : 13: substitution.graph_param_transform 60.47% : 0.000315s : 4: substitution.inline 1.89% : 0.000010s : 20: substitution.j_node_and_user_rematch 4.41% : 0.000023s : 4: substitution.less_batch_normalization 1.22% : 0.000006s : 6: substitution.load_eliminater 2.61% : 0.000014s : 20: substitution.remove_not_recompute_node 1.46% : 0.000008s : 6: substitution.replace_old_param 5.86% : 0.000031s : 22: substitution.updatestate_pure_node_eliminater 16.03% : 0.000083s : 26: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.171063 2 99.56% : 0.170317s : 1: type_inference.infer 0.44% : 0.000747s : 1: type_inference.specialize ------[replace.] 0.000062 4 100.00% : 0.000062s : 4: replace.inline ------[match.] 0.000311 4 100.00% : 0.000311s : 4: match.inline ------[predicate.] 0.000564 3432 0.88% : 0.000005s : 35: predicate.accumulaten_eliminater 2.15% : 0.000012s : 13: predicate.ad_related_special_op_eliminate 0.66% : 0.000004s : 26: predicate.addn_check_dump 0.90% : 0.000005s : 35: predicate.addn_zero_filter 0.89% : 0.000005s : 35: predicate.adjust_all_reduce_mul_add 2.09% : 0.000012s : 61: predicate.arithmetic_simplify 0.88% : 0.000005s : 35: predicate.cast_eliminate 0.65% : 0.000004s : 26: predicate.check_bprop_eliminate 0.60% : 0.000003s : 26: predicate.compare_switch_simplify 0.18% : 0.000001s : 13: predicate.const_output_eliminate 0.66% : 0.000004s : 26: predicate.depend_value_elim 0.92% : 0.000005s : 35: predicate.dict_get_item_const_eliminator 0.98% : 0.000006s : 35: predicate.dict_get_item_eliminator 0.87% : 0.000005s : 35: predicate.dict_set_item_eliminator 1.16% : 0.000007s : 26: predicate.dumpgradient_eliminate 0.23% : 0.000001s : 13: predicate.elim_not_effective 0.38% : 0.000002s : 13: predicate.elim_shapecalc_of_broadcastargs 1.13% : 0.000006s : 48: predicate.environ_add_const_eliminate 1.10% : 0.000006s : 48: predicate.environ_get_add_eliminate 1.17% : 0.000007s : 48: predicate.environ_get_depend_swap 1.75% : 0.000010s : 74: predicate.environ_get_eliminate 1.10% : 0.000006s : 48: predicate.environ_get_set_eliminate 0.95% : 0.000005s : 39: predicate.exchange_switch_depend_value 1.43% : 0.000008s : 39: predicate.float_depend_g_call 0.63% : 0.000004s : 26: predicate.float_environ_get_switch 0.97% : 0.000005s : 39: predicate.float_tuple_getitem_switch 0.18% : 0.000001s : 13: predicate.fold_const_symbol 0.87% : 0.000005s : 26: predicate.get_grad_eliminate 0.28% : 0.000002s : 13: predicate.graph_param_transform 0.72% : 0.000004s : 26: predicate.incorporate_call 0.59% : 0.000003s : 26: predicate.incorporate_call_switch 5.22% : 0.000029s : 152: predicate.inline 0.80% : 0.000005s : 26: predicate.inline_without_move 0.33% : 0.000002s : 26: predicate.j_node_and_user_rematch 1.00% : 0.000006s : 26: predicate.less_batch_normalization 1.60% : 0.000009s : 61: predicate.list_to_tuple_eliminator_ 2.30% : 0.000013s : 96: predicate.load_eliminater 0.93% : 0.000005s : 13: predicate.loop_unroll_after_grad 1.25% : 0.000007s : 49: predicate.loop_unroll_before_grad 1.58% : 0.000009s : 61: predicate.make_slice_get_slice_eliminator 0.66% : 0.000004s : 26: predicate.merge_addn 0.61% : 0.000003s : 26: predicate.micro_step_allgather_replace 0.68% : 0.000004s : 26: predicate.mini_step_allgather_replace 0.79% : 0.000004s : 35: predicate.minmaximum_grad 0.87% : 0.000005s : 13: predicate.mutable_eliminate 0.40% : 0.000002s : 13: predicate.opt_reshape 0.37% : 0.000002s : 13: predicate.parallel_virtual_node 1.34% : 0.000008s : 39: predicate.partial_defer_inline 1.24% : 0.000007s : 48: predicate.partial_eliminate 0.88% : 0.000005s : 35: predicate.print_const_string_wrapper 0.66% : 0.000004s : 26: predicate.reduce_all_const_elim 1.06% : 0.000006s : 35: predicate.reduce_eliminate 2.39% : 0.000013s : 96: predicate.redundant_stop_gradient_eliminater 0.43% : 0.000002s : 26: predicate.remove_not_recompute_node 1.02% : 0.000006s : 61: predicate.replace_applicator 0.47% : 0.000003s : 26: predicate.replace_old_param 0.24% : 0.000001s : 13: predicate.reset_defer_inline 0.88% : 0.000005s : 35: predicate.reshape_eliminate 0.66% : 0.000004s : 26: predicate.row_tensor_add_zeros_like 0.42% : 0.000002s : 13: predicate.row_tensor_eliminate 0.78% : 0.000004s : 26: predicate.same_eliminate 0.46% : 0.000003s : 32: predicate.set_cell_output_no_recompute 0.76% : 0.000004s : 26: predicate.shard_identity_eliminate 0.86% : 0.000005s : 26: predicate.special_op_eliminate 0.79% : 0.000004s : 26: predicate.specialize_transform 0.94% : 0.000005s : 26: predicate.split_environ_get_set_with_tuple_value 0.78% : 0.000004s : 26: predicate.stack_unstack_eliminate 0.34% : 0.000002s : 13: predicate.switch_call_monad_eliminater 1.00% : 0.000006s : 39: predicate.switch_defer_inline 1.62% : 0.000009s : 65: predicate.switch_layer_defer_inline 3.59% : 0.000020s : 127: predicate.switch_simplify 0.88% : 0.000005s : 35: predicate.tile_eliminate 0.87% : 0.000005s : 35: predicate.transpose_eliminate 1.60% : 0.000009s : 61: predicate.tuple_list_convert_item_index_to_positive 1.67% : 0.000009s : 61: predicate.tuple_list_get_item_const_eliminator 1.47% : 0.000008s : 61: predicate.tuple_list_get_item_depend_reorder 2.59% : 0.000015s : 87: predicate.tuple_list_get_item_eliminator 1.56% : 0.000009s : 61: predicate.tuple_list_get_set_item_eliminator 8.57% : 0.000048s : 87: predicate.tuple_list_set_item_eliminator 1.43% : 0.000008s : 61: predicate.tuple_to_list_eliminator_ 2.44% : 0.000014s : 96: predicate.updatestate_pure_node_eliminater 3.21% : 0.000018s : 122: predicate.updatestate_useless_node_eliminater 0.37% : 0.000002s : 13: predicate.value_based_eliminate 0.73% : 0.000004s : 26: predicate.virtual_dataset_eliminate 0.85% : 0.000005s : 26: predicate.virtual_output_eliminate 0.33% : 0.000002s : 13: predicate.virtual_view_grad_eliminate 0.46% : 0.000003s : 13: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000474 7 5.14% : 0.000024s : 1: func_graph_cloner_run.FuncGraphClonerGraph 16.77% : 0.000079s : 1: func_graph_cloner_run.FuncGraphClonerNode 78.09% : 0.000370s : 5: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.373135 192 0.00% : 0.000004s : 1: ForceFp32Comm 1.13% : 0.004227s : 1: add_attr 1.13% : 0.004210s : 1: add_attr_with_inline 0.00% : 0.000005s : 1: add_comm_op_reuse_tag 0.03% : 0.000117s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.06% : 0.000241s : 1: auto_monad 0.02% : 0.000060s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.15% : 0.000564s : 1: bootstrap 0.01% : 0.000041s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000033s : 1: control_data_broadcast_order 0.01% : 0.000021s : 1: convert_after_rewriter 0.01% : 0.000048s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.01% : 0.000019s : 1: environ_conv 0.01% : 0.000025s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000011s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000010s : 1: label_micro_interleaved_index 0.16% : 0.000585s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.22% : 0.000838s : 1: mutable_eliminate 0.00% : 0.000011s : 1: offloading_packed_experts 0.01% : 0.000029s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000033s : 1: opt.transform.mutable_eliminate 0.76% : 0.002834s : 78: opt.transform.opt_a 0.03% : 0.000112s : 1: opt.transform.opt_after_cconv 0.03% : 0.000096s : 1: opt.transform.opt_after_jit_grad 0.09% : 0.000349s : 28: opt.transform.opt_b 0.03% : 0.000123s : 2: opt.transform.opt_trans_graph 0.03% : 0.000095s : 4: opt.transform.symbol_engine_opt 1.53% : 0.005704s : 1: opt_a 0.07% : 0.000250s : 1: opt_after_cconv 45.22% : 0.168733s : 1: opt_after_jit_grad 0.14% : 0.000538s : 1: opt_b 2.52% : 0.009415s : 1: optimize 0.03% : 0.000127s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000015s : 1: order_py_execute_after_rewriter 0.01% : 0.000052s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000011s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000007s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000012s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000004s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000050s : 1: pre_auto_parallel 0.01% : 0.000038s : 1: py_interpret_to_execute 0.01% : 0.000027s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.01% : 0.000029s : 1: remove_dup_value 0.19% : 0.000726s : 1: renormalize.infer 0.15% : 0.000555s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000011s : 1: rewriter_after_jit_bprop_graph 0.07% : 0.000258s : 1: rewriter_after_opt_a 0.03% : 0.000100s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000015s : 1: swap_dp_allreduce_reducescatter 0.04% : 0.000162s : 1: symbol_engine_optimizer 0.04% : 0.000158s : 1: tuple_transform 45.89% : 0.171238s : 1: type_inference [WARNING] PIPELINE(166799,ffff961eff30,python3.9):2026-01-29-17:42:03.304.314 [mindspore/ccsrc/frontend/jit/ps/validator.cc:244] CheckAssignReturnValue] Deprecated: the return value of Assign/AssignAdd/AssignSub operator will be removed in subsequent releases. You can modify the code from: a = P.Assign()(param, value) b = a * 2 to: P.Assign()(param, value) b = param * 2 Please check your code:In file /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/auto_generate/gen_ops_def.py:9741, 11~37 return assign_op(variable, value) ^~~~~~~~~~~~~~~~~~~~~~~~~~ TotalTime = 0.274532, [21] [bootstrap]: 0.00107769 [type_inference]: 0.254853 [event_method]: 1.8e-05 [auto_monad]: 0.00025128 [graph_reusing]: 6.24999e-06 [inline]: 2.48002e-06 [add_attr]: 0.0063918, [1] [add_attr_with_inline]: 0.00637673, [1] [Cycle 1]: 8.097e-05, [2] [tag_attr]: 2.677e-05 [meta_addattr_fg_expand]: 6.94001e-06 [parallel-infer-symbol]: 3.82998e-06 [pre_auto_parallel]: 0.0001155 [insert-virtual-dataset]: 3.13e-06 [parallel-infer-symbol-second]: 9.5999e-07 [dataset_repeat_opt]: 2.05002e-06 [pipeline_split]: 1.92999e-06 [optimize]: 0.00958659, [53] [py_interpret_to_execute]: 3.76e-05 [rewriter_before_opt_a]: 0.00010031 [opt_a]: 0.00590641, [2] [Cycle 1]: 0.00429729, [45] [expand_dump_flag]: 3.73001e-06 [switch_simplify]: 4.674e-05 [loop_unroll]: 3.235e-05 [a_1]: 0.00131312 [with_stream_mark]: 2.778e-05 [recompute_prepare]: 2.686e-05 [updatestate_depend_eliminate]: 7.651e-05 [updatestate_assign_eliminate]: 4.796e-05 [updatestate_loads_eliminate]: 1.608e-05 [parameter_eliminate]: 3.16001e-06 [a_2]: 0.00030228 [accelerated_algorithm]: 4.396e-05 [shard]: 2.63e-06 [meta_shard_fg_expand]: 4.22003e-06 [shard_inline]: 1.596e-05 [merge_send_recv]: 1.505e-05 [auto_parallel]: 1.552e-05 [parallel]: 0.00011717 [flash_sp]: 1.319e-05 [merge_comm]: 1.002e-05 [allreduce_fusion]: 8.59e-06 [matmul_add_comm_reduction]: 1.692e-05 [allreduce_slice_to_reducescatter]: 8.80013e-07 [virtual_shard_identity]: 1.83e-05 [virtual_dataset]: 1.539e-05 [get_grad_eliminate_]: 1.485e-05 [virtual_output]: 1.516e-05 [merge_forward]: 8.85999e-06 [cell_reuse_recompute_pass]: 1.65001e-06 [offload_activation]: 1.832e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.131e-05 [merge_recompute_call_nodes]: 1.79998e-06 [before_grad]: 2.901e-05 [set_forward_comm_id_for_comm_node_pass]: 8.74e-06 [meta_fg_expand]: 7.08e-06 [flash_sp_send_recv_attached]: 4.87e-06 [receive_attached]: 2.15002e-06 [after_resolve]: 2.366e-05 [a_after_grad]: 2.469e-05 [renormalize]: 0.00122236 [add_forward_monad_depend]: 9.646e-05 [auto_monad_grad]: 2.88e-06 [auto_monad_eliminator]: 3.265e-05 [cse]: 0.0001274 [a_3]: 0.00012039 [Cycle 2]: 0.00159511, [45] [expand_dump_flag]: 2.91999e-06 [switch_simplify]: 1.651e-05 [loop_unroll]: 1.497e-05 [a_1]: 0.00046875 [with_stream_mark]: 2.353e-05 [recompute_prepare]: 1.626e-05 [updatestate_depend_eliminate]: 1.015e-05 [updatestate_assign_eliminate]: 1.161e-05 [updatestate_loads_eliminate]: 1.458e-05 [parameter_eliminate]: 2.36998e-06 [a_2]: 0.00023019 [accelerated_algorithm]: 2.252e-05 [shard]: 1.82001e-06 [meta_shard_fg_expand]: 4.40999e-06 [shard_inline]: 1.539e-05 [merge_send_recv]: 1.459e-05 [auto_parallel]: 1.583e-05 [parallel]: 1.059e-05 [flash_sp]: 4.48001e-06 [merge_comm]: 9.27001e-06 [allreduce_fusion]: 8.97e-06 [matmul_add_comm_reduction]: 1.676e-05 [allreduce_slice_to_reducescatter]: 7.80012e-07 [virtual_shard_identity]: 1.701e-05 [virtual_dataset]: 1.556e-05 [get_grad_eliminate_]: 1.511e-05 [virtual_output]: 1.474e-05 [merge_forward]: 8.68001e-06 [cell_reuse_recompute_pass]: 3.20002e-06 [offload_activation]: 1.684e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.318e-05 [merge_recompute_call_nodes]: 2.03002e-06 [before_grad]: 2.92e-05 [set_forward_comm_id_for_comm_node_pass]: 9.42001e-06 [meta_fg_expand]: 7.23e-06 [flash_sp_send_recv_attached]: 1.72999e-06 [receive_attached]: 3.09001e-06 [after_resolve]: 2.526e-05 [a_after_grad]: 2.552e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 2.70002e-06 [auto_monad_grad]: 2.71e-06 [auto_monad_eliminator]: 2.422e-05 [cse]: 5.382e-05 [a_3]: 0.00011116 [py_interpret_to_execute_after_opt_a]: 2.245e-05 [slice_cell_reuse_recomputed_activation]: 2.63e-06 [rewriter_after_opt_a]: 0.00019114 [convert_after_rewriter]: 1.659e-05 [order_py_execute_after_rewriter]: 1.179e-05 [mutable_eliminate]: 0.00078042 [opt_b]: 0.00055083, [1] [Cycle 1]: 0.00054203, [7] [b_1]: 0.00038727 [b_2]: 1.851e-05 [updatestate_depend_eliminate]: 1.412e-05 [updatestate_assign_eliminate]: 1.088e-05 [updatestate_loads_eliminate]: 1.479e-05 [renormalize]: 8.2e-07 [cse]: 5.412e-05 [optimize_parallel_all_gather_comm]: 3.295e-05 [overlap_param_gather]: 2.55997e-06 [cconv]: 3.98e-05 [loop_unroll]: 0.0005745 [opt_after_cconv]: 0.00024704, [1] [Cycle 1]: 0.00024003, [7] [c_1]: 0.00012069 [parameter_eliminate]: 5.82999e-06 [updatestate_depend_eliminate]: 1.261e-05 [updatestate_assign_eliminate]: 9.08002e-06 [updatestate_loads_eliminate]: 1.089e-05 [cse]: 4.363e-05 [renormalize]: 8.70001e-07 [remove_dup_value]: 2.396e-05 [tuple_transform]: 0.00018209, [1] [Cycle 1]: 0.00017644, [4] [d_1]: 0.00013571 [none_parameter_eliminate]: 2.24999e-06 [renormalize]: 2.19996e-07 [switch_simplify]: 1.697e-05 [partial_unused_args_eliminate]: 1.96e-06 [add_recomputation]: 0.00020104 [cse_after_recomputation]: 5.055e-05, [1] [Cycle 1]: 4.479e-05, [1] [cse]: 3.797e-05 [environ_conv]: 1.741e-05 [swap_dp_allreduce_reducescatter]: 1.295e-05 [bias_add_comm_swap]: 4.38001e-06 [label_micro_interleaved_index]: 6.58e-06 [label_fine_grained_interleaved_index]: 3.44001e-06 [merge_cast_opt]: 1.85001e-06 [slice_recompute_activation]: 2.34001e-06 [micro_interleaved_order_control]: 2.48e-06 [assign_add_opt]: 1.32e-06 [ForceFp32Comm]: 9.80013e-07 [remove_cast_before_assign_add]: 1.14998e-06 [full_micro_interleaved_order_control]: 2.19999e-06 [reorder_send_recv_between_fp_bp]: 2.87002e-06 [comm_op_add_attrs]: 1.55999e-06 [add_comm_op_reuse_tag]: 1.10999e-06 [interleave_split_concat_branches]: 1.23002e-06 [interleave_parallel_branches]: 1.14e-06 [overlap_opt_shard_in_pipeline]: 1.67001e-06 [overlap_opt_shard_grad_in_pipeline]: 1.92001e-06 [control_data_broadcast_order]: 3.049e-05 [grouped_pairwise_exchange_alltoall]: 1.69e-06 [offloading_packed_experts]: 9.19998e-06 [overlap_recompute_and_grad_model_parallel]: 8.66997e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.32e-06 [overlap_recompute_allgather_and_fa_grad]: 1.71e-06 [overlap_recompute_comm]: 2.41998e-06 [overlap_grad_ring_attention]: 8.73001e-06 [overlap_grad_flash_sp]: 4.187e-05 [begin_end_overlap_inline]: 6.39993e-07 [split_matmul_comm_elemetwise]: 2.54001e-06 [split_layernorm_comm]: 1.97999e-06 [handle_group_info]: 1.04e-06 [symbol_engine_optimizer]: 0.00016536, [1] [Cycle 1]: 0.00015935, [6] [build]: 1.937e-05 [elim_shapecalc]: 2.669e-05 [elim_not_effective]: 3.143e-05 [opt_reshape]: 2.023e-05 [fold_const_symbol]: 2.671e-05 [renormalize]: 2.50002e-07 [detach_backward]: 2.09e-06 [pipeline_parallel_scheduler]: 1.60001e-06 [auto_monad_reorder]: 5.337e-05 [get_jit_bprop_graph]: 1.76e-06 [rewriter_after_jit_bprop_graph]: 5.78002e-06 [opt_after_jit_grad]: 0.00073943 [validate]: 0.00120536 Sums bootstrap : 0.001078s : 0.40% type_inference : 0.254853s : 95.46% event_method : 0.000018s : 0.01% auto_monad : 0.000251s : 0.09% graph_reusing : 0.000006s : 0.00% inline : 0.000002s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000027s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000007s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000115s : 0.04% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000038s : 0.01% optimize.rewriter_before_opt_a : 0.000100s : 0.04% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000063s : 0.02% optimize.opt_a.loop_unroll : 0.000047s : 0.02% optimize.opt_a.a_1 : 0.001782s : 0.67% optimize.opt_a.with_stream_mark : 0.000051s : 0.02% optimize.opt_a.recompute_prepare : 0.000043s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000087s : 0.03% optimize.opt_a.updatestate_assign_eliminate : 0.000060s : 0.02% optimize.opt_a.updatestate_loads_eliminate : 0.000031s : 0.01% optimize.opt_a.parameter_eliminate : 0.000006s : 0.00% optimize.opt_a.a_2 : 0.000532s : 0.20% optimize.opt_a.accelerated_algorithm : 0.000066s : 0.02% optimize.opt_a.shard : 0.000004s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000009s : 0.00% optimize.opt_a.shard_inline : 0.000031s : 0.01% optimize.opt_a.merge_send_recv : 0.000030s : 0.01% optimize.opt_a.auto_parallel : 0.000031s : 0.01% optimize.opt_a.parallel : 0.000128s : 0.05% optimize.opt_a.flash_sp : 0.000018s : 0.01% optimize.opt_a.merge_comm : 0.000019s : 0.01% optimize.opt_a.allreduce_fusion : 0.000018s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000034s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000035s : 0.01% optimize.opt_a.virtual_dataset : 0.000031s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000030s : 0.01% optimize.opt_a.virtual_output : 0.000030s : 0.01% optimize.opt_a.merge_forward : 0.000018s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000035s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000064s : 0.02% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000058s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000018s : 0.01% optimize.opt_a.meta_fg_expand : 0.000014s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000007s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000049s : 0.02% optimize.opt_a.a_after_grad : 0.000050s : 0.02% optimize.opt_a.renormalize : 0.001222s : 0.46% optimize.opt_a.add_forward_monad_depend : 0.000099s : 0.04% optimize.opt_a.auto_monad_grad : 0.000006s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000057s : 0.02% optimize.opt_a.cse : 0.000181s : 0.07% optimize.opt_a.a_3 : 0.000232s : 0.09% optimize.py_interpret_to_execute_after_opt_a : 0.000022s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000191s : 0.07% optimize.convert_after_rewriter : 0.000017s : 0.01% optimize.order_py_execute_after_rewriter : 0.000012s : 0.00% optimize.mutable_eliminate : 0.000780s : 0.29% optimize.opt_b.b_1 : 0.000387s : 0.15% optimize.opt_b.b_2 : 0.000019s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000014s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000011s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000015s : 0.01% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000054s : 0.02% optimize.optimize_parallel_all_gather_comm : 0.000033s : 0.01% optimize.overlap_param_gather : 0.000003s : 0.00% optimize.cconv : 0.000040s : 0.01% optimize.loop_unroll : 0.000575s : 0.22% optimize.opt_after_cconv.c_1 : 0.000121s : 0.05% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000013s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000009s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000011s : 0.00% optimize.opt_after_cconv.cse : 0.000044s : 0.02% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000024s : 0.01% optimize.tuple_transform.d_1 : 0.000136s : 0.05% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000017s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000201s : 0.08% optimize.cse_after_recomputation.cse : 0.000038s : 0.01% optimize.environ_conv : 0.000017s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000013s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000007s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000002s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000030s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000009s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000009s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000009s : 0.00% optimize.overlap_grad_flash_sp : 0.000042s : 0.02% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000019s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000027s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000031s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000020s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000027s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000053s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.000739s : 0.28% validate : 0.001205s : 0.45% Time group info: ------[substitution.] 0.000590 143 1.88% : 0.000011s : 2: substitution.depend_value_elim 0.82% : 0.000005s : 10: substitution.elim_not_effective 0.62% : 0.000004s : 10: substitution.fold_const_symbol 2.18% : 0.000013s : 13: substitution.graph_param_transform 65.01% : 0.000383s : 4: substitution.inline 1.81% : 0.000011s : 20: substitution.j_node_and_user_rematch 3.71% : 0.000022s : 4: substitution.less_batch_normalization 1.22% : 0.000007s : 6: substitution.load_eliminater 2.68% : 0.000016s : 20: substitution.remove_not_recompute_node 1.36% : 0.000008s : 6: substitution.replace_old_param 4.99% : 0.000029s : 22: substitution.updatestate_pure_node_eliminater 13.72% : 0.000081s : 26: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.254778 2 99.65% : 0.253886s : 1: type_inference.infer 0.35% : 0.000892s : 1: type_inference.specialize ------[replace.] 0.000069 4 100.00% : 0.000069s : 4: replace.inline ------[match.] 0.000380 4 100.00% : 0.000380s : 4: match.inline ------[predicate.] 0.000534 3432 0.95% : 0.000005s : 35: predicate.accumulaten_eliminater 0.91% : 0.000005s : 13: predicate.ad_related_special_op_eliminate 0.69% : 0.000004s : 26: predicate.addn_check_dump 0.98% : 0.000005s : 35: predicate.addn_zero_filter 0.94% : 0.000005s : 35: predicate.adjust_all_reduce_mul_add 2.07% : 0.000011s : 61: predicate.arithmetic_simplify 0.95% : 0.000005s : 35: predicate.cast_eliminate 0.80% : 0.000004s : 26: predicate.check_bprop_eliminate 0.75% : 0.000004s : 26: predicate.compare_switch_simplify 0.19% : 0.000001s : 13: predicate.const_output_eliminate 0.80% : 0.000004s : 26: predicate.depend_value_elim 0.94% : 0.000005s : 35: predicate.dict_get_item_const_eliminator 1.05% : 0.000006s : 35: predicate.dict_get_item_eliminator 0.90% : 0.000005s : 35: predicate.dict_set_item_eliminator 1.01% : 0.000005s : 26: predicate.dumpgradient_eliminate 0.25% : 0.000001s : 13: predicate.elim_not_effective 0.45% : 0.000002s : 13: predicate.elim_shapecalc_of_broadcastargs 1.34% : 0.000007s : 48: predicate.environ_add_const_eliminate 1.19% : 0.000006s : 48: predicate.environ_get_add_eliminate 1.19% : 0.000006s : 48: predicate.environ_get_depend_swap 1.97% : 0.000011s : 74: predicate.environ_get_eliminate 1.23% : 0.000007s : 48: predicate.environ_get_set_eliminate 1.08% : 0.000006s : 39: predicate.exchange_switch_depend_value 1.52% : 0.000008s : 39: predicate.float_depend_g_call 0.72% : 0.000004s : 26: predicate.float_environ_get_switch 0.99% : 0.000005s : 39: predicate.float_tuple_getitem_switch 0.19% : 0.000001s : 13: predicate.fold_const_symbol 0.78% : 0.000004s : 26: predicate.get_grad_eliminate 0.24% : 0.000001s : 13: predicate.graph_param_transform 0.65% : 0.000003s : 26: predicate.incorporate_call 0.61% : 0.000003s : 26: predicate.incorporate_call_switch 5.44% : 0.000029s : 152: predicate.inline 1.01% : 0.000005s : 26: predicate.inline_without_move 0.36% : 0.000002s : 26: predicate.j_node_and_user_rematch 1.12% : 0.000006s : 26: predicate.less_batch_normalization 1.73% : 0.000009s : 61: predicate.list_to_tuple_eliminator_ 2.50% : 0.000013s : 96: predicate.load_eliminater 0.87% : 0.000005s : 13: predicate.loop_unroll_after_grad 1.31% : 0.000007s : 49: predicate.loop_unroll_before_grad 1.78% : 0.000010s : 61: predicate.make_slice_get_slice_eliminator 0.76% : 0.000004s : 26: predicate.merge_addn 0.73% : 0.000004s : 26: predicate.micro_step_allgather_replace 0.70% : 0.000004s : 26: predicate.mini_step_allgather_replace 0.97% : 0.000005s : 35: predicate.minmaximum_grad 0.90% : 0.000005s : 13: predicate.mutable_eliminate 0.40% : 0.000002s : 13: predicate.opt_reshape 0.50% : 0.000003s : 13: predicate.parallel_virtual_node 1.47% : 0.000008s : 39: predicate.partial_defer_inline 1.46% : 0.000008s : 48: predicate.partial_eliminate 0.88% : 0.000005s : 35: predicate.print_const_string_wrapper 0.89% : 0.000005s : 26: predicate.reduce_all_const_elim 1.10% : 0.000006s : 35: predicate.reduce_eliminate 2.46% : 0.000013s : 96: predicate.redundant_stop_gradient_eliminater 0.45% : 0.000002s : 26: predicate.remove_not_recompute_node 1.07% : 0.000006s : 61: predicate.replace_applicator 0.51% : 0.000003s : 26: predicate.replace_old_param 0.24% : 0.000001s : 13: predicate.reset_defer_inline 0.97% : 0.000005s : 35: predicate.reshape_eliminate 0.78% : 0.000004s : 26: predicate.row_tensor_add_zeros_like 0.47% : 0.000003s : 13: predicate.row_tensor_eliminate 0.95% : 0.000005s : 26: predicate.same_eliminate 0.53% : 0.000003s : 32: predicate.set_cell_output_no_recompute 0.93% : 0.000005s : 26: predicate.shard_identity_eliminate 0.77% : 0.000004s : 26: predicate.special_op_eliminate 0.87% : 0.000005s : 26: predicate.specialize_transform 0.97% : 0.000005s : 26: predicate.split_environ_get_set_with_tuple_value 0.82% : 0.000004s : 26: predicate.stack_unstack_eliminate 0.42% : 0.000002s : 13: predicate.switch_call_monad_eliminater 1.09% : 0.000006s : 39: predicate.switch_defer_inline 1.69% : 0.000009s : 65: predicate.switch_layer_defer_inline 3.75% : 0.000020s : 127: predicate.switch_simplify 0.92% : 0.000005s : 35: predicate.tile_eliminate 0.90% : 0.000005s : 35: predicate.transpose_eliminate 1.77% : 0.000009s : 61: predicate.tuple_list_convert_item_index_to_positive 1.81% : 0.000010s : 61: predicate.tuple_list_get_item_const_eliminator 1.74% : 0.000009s : 61: predicate.tuple_list_get_item_depend_reorder 2.94% : 0.000016s : 87: predicate.tuple_list_get_item_eliminator 1.73% : 0.000009s : 61: predicate.tuple_list_get_set_item_eliminator 2.52% : 0.000013s : 87: predicate.tuple_list_set_item_eliminator 1.66% : 0.000009s : 61: predicate.tuple_to_list_eliminator_ 2.68% : 0.000014s : 96: predicate.updatestate_pure_node_eliminater 3.56% : 0.000019s : 122: predicate.updatestate_useless_node_eliminater 0.41% : 0.000002s : 13: predicate.value_based_eliminate 0.75% : 0.000004s : 26: predicate.virtual_dataset_eliminate 0.79% : 0.000004s : 26: predicate.virtual_output_eliminate 0.40% : 0.000002s : 13: predicate.virtual_view_grad_eliminate 0.44% : 0.000002s : 13: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000463 7 5.61% : 0.000026s : 1: func_graph_cloner_run.FuncGraphClonerGraph 15.13% : 0.000070s : 1: func_graph_cloner_run.FuncGraphClonerNode 79.26% : 0.000367s : 5: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.294273 192 0.00% : 0.000004s : 1: ForceFp32Comm 2.17% : 0.006399s : 1: add_attr 2.17% : 0.006382s : 1: add_attr_with_inline 0.00% : 0.000005s : 1: add_comm_op_reuse_tag 0.07% : 0.000207s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.09% : 0.000260s : 1: auto_monad 0.02% : 0.000059s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000008s : 1: bias_add_comm_swap 0.38% : 0.001114s : 1: bootstrap 0.01% : 0.000044s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000035s : 1: control_data_broadcast_order 0.01% : 0.000021s : 1: convert_after_rewriter 0.02% : 0.000054s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.01% : 0.000021s : 1: environ_conv 0.01% : 0.000024s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000010s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000010s : 1: label_micro_interleaved_index 0.20% : 0.000584s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.27% : 0.000793s : 1: mutable_eliminate 0.00% : 0.000012s : 1: offloading_packed_experts 0.01% : 0.000029s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000033s : 1: opt.transform.mutable_eliminate 1.03% : 0.003031s : 78: opt.transform.opt_a 0.04% : 0.000119s : 1: opt.transform.opt_after_cconv 0.02% : 0.000063s : 1: opt.transform.opt_after_jit_grad 0.13% : 0.000368s : 28: opt.transform.opt_b 0.04% : 0.000132s : 2: opt.transform.opt_trans_graph 0.03% : 0.000100s : 4: opt.transform.symbol_engine_opt 2.01% : 0.005911s : 1: opt_a 0.09% : 0.000251s : 1: opt_after_cconv 0.26% : 0.000754s : 1: opt_after_jit_grad 0.19% : 0.000555s : 1: opt_b 3.26% : 0.009593s : 1: optimize 0.01% : 0.000037s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000016s : 1: order_py_execute_after_rewriter 0.02% : 0.000046s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000012s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000013s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.04% : 0.000122s : 1: pre_auto_parallel 0.01% : 0.000043s : 1: py_interpret_to_execute 0.01% : 0.000027s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.01% : 0.000028s : 1: remove_dup_value 0.23% : 0.000664s : 1: renormalize.infer 0.19% : 0.000546s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.07% : 0.000199s : 1: rewriter_after_opt_a 0.04% : 0.000106s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000006s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.01% : 0.000016s : 1: swap_dp_allreduce_reducescatter 0.06% : 0.000169s : 1: symbol_engine_optimizer 0.06% : 0.000185s : 1: tuple_transform 86.61% : 0.254883s : 1: type_inference TotalTime = 0.33179, [21] [bootstrap]: 0.00051456 [type_inference]: 0.314288 [event_method]: 3.009e-05 [auto_monad]: 0.0003657 [graph_reusing]: 1.014e-05 [inline]: 3.28e-06 [add_attr]: 0.00467141, [1] [add_attr_with_inline]: 0.00465769, [1] [Cycle 1]: 0.00011186, [2] [tag_attr]: 4.764e-05 [meta_addattr_fg_expand]: 1.044e-05 [parallel-infer-symbol]: 3.97998e-06 [pre_auto_parallel]: 6.714e-05 [insert-virtual-dataset]: 2.90002e-06 [parallel-infer-symbol-second]: 1.07e-06 [dataset_repeat_opt]: 2.41e-06 [pipeline_split]: 1.69e-06 [optimize]: 0.0108367, [53] [py_interpret_to_execute]: 5.206e-05 [rewriter_before_opt_a]: 0.00016463 [opt_a]: 0.00724111, [2] [Cycle 1]: 0.00559863, [45] [expand_dump_flag]: 3.75e-06 [switch_simplify]: 0.00016059 [loop_unroll]: 5.485e-05 [a_1]: 0.00169532 [with_stream_mark]: 4.503e-05 [recompute_prepare]: 3.553e-05 [updatestate_depend_eliminate]: 0.00010357 [updatestate_assign_eliminate]: 1.2e-05 [updatestate_loads_eliminate]: 5.949e-05 [parameter_eliminate]: 2.83e-06 [a_2]: 0.00025299 [accelerated_algorithm]: 4.203e-05 [shard]: 3.03e-06 [meta_shard_fg_expand]: 4.85001e-06 [shard_inline]: 1.655e-05 [merge_send_recv]: 1.526e-05 [auto_parallel]: 1.745e-05 [parallel]: 2.832e-05 [flash_sp]: 1.516e-05 [merge_comm]: 1.085e-05 [allreduce_fusion]: 8.99e-06 [matmul_add_comm_reduction]: 1.867e-05 [allreduce_slice_to_reducescatter]: 9.70002e-07 [virtual_shard_identity]: 1.926e-05 [virtual_dataset]: 1.644e-05 [get_grad_eliminate_]: 1.602e-05 [virtual_output]: 1.644e-05 [merge_forward]: 1.098e-05 [cell_reuse_recompute_pass]: 2.09999e-06 [offload_activation]: 2.142e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.556e-05 [merge_recompute_call_nodes]: 2.04e-06 [before_grad]: 2.796e-05 [set_forward_comm_id_for_comm_node_pass]: 1.131e-05 [meta_fg_expand]: 8.24002e-06 [flash_sp_send_recv_attached]: 7.13e-06 [receive_attached]: 2.89999e-06 [after_resolve]: 2.524e-05 [a_after_grad]: 2.659e-05 [renormalize]: 0.0019519 [add_forward_monad_depend]: 1.189e-05 [auto_monad_grad]: 2.88e-06 [auto_monad_eliminator]: 3.605e-05 [cse]: 0.00021076 [a_3]: 0.0001163 [Cycle 2]: 0.0016264, [45] [expand_dump_flag]: 2.76e-06 [switch_simplify]: 1.63e-05 [loop_unroll]: 1.459e-05 [a_1]: 0.0004725 [with_stream_mark]: 2.739e-05 [recompute_prepare]: 1.611e-05 [updatestate_depend_eliminate]: 1.031e-05 [updatestate_assign_eliminate]: 9.28002e-06 [updatestate_loads_eliminate]: 1.202e-05 [parameter_eliminate]: 2.43998e-06 [a_2]: 0.00020636 [accelerated_algorithm]: 2.536e-05 [shard]: 2.37001e-06 [meta_shard_fg_expand]: 4.62e-06 [shard_inline]: 1.453e-05 [merge_send_recv]: 1.511e-05 [auto_parallel]: 1.535e-05 [parallel]: 1.991e-05 [flash_sp]: 4.32e-06 [merge_comm]: 1.091e-05 [allreduce_fusion]: 9.64999e-06 [matmul_add_comm_reduction]: 1.976e-05 [allreduce_slice_to_reducescatter]: 1.49e-06 [virtual_shard_identity]: 1.848e-05 [virtual_dataset]: 1.354e-05 [get_grad_eliminate_]: 1.477e-05 [virtual_output]: 1.309e-05 [merge_forward]: 1.204e-05 [cell_reuse_recompute_pass]: 2.56e-06 [offload_activation]: 1.872e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.208e-05 [merge_recompute_call_nodes]: 1.64e-06 [before_grad]: 2.377e-05 [set_forward_comm_id_for_comm_node_pass]: 9.37001e-06 [meta_fg_expand]: 3.204e-05 [flash_sp_send_recv_attached]: 1.80001e-06 [receive_attached]: 2.48e-06 [after_resolve]: 2.85e-05 [a_after_grad]: 2.389e-05 [renormalize]: 6.00121e-08 [add_forward_monad_depend]: 3.68e-06 [auto_monad_grad]: 2.49001e-06 [auto_monad_eliminator]: 2.862e-05 [cse]: 5.899e-05 [a_3]: 9.496e-05 [py_interpret_to_execute_after_opt_a]: 2.818e-05 [slice_cell_reuse_recomputed_activation]: 2.91e-06 [rewriter_after_opt_a]: 0.0002196 [convert_after_rewriter]: 1.783e-05 [order_py_execute_after_rewriter]: 1.091e-05 [mutable_eliminate]: 0.00082311 [opt_b]: 0.00049334, [1] [Cycle 1]: 0.00048386, [7] [b_1]: 0.00032099 [b_2]: 1.783e-05 [updatestate_depend_eliminate]: 1.573e-05 [updatestate_assign_eliminate]: 9.66e-06 [updatestate_loads_eliminate]: 1.12e-05 [renormalize]: 1.79e-06 [cse]: 6.452e-05 [optimize_parallel_all_gather_comm]: 3.544e-05 [overlap_param_gather]: 2.63998e-06 [cconv]: 4.126e-05 [loop_unroll]: 0.00053627 [opt_after_cconv]: 0.00022211, [1] [Cycle 1]: 0.000215, [7] [c_1]: 9.738e-05 [parameter_eliminate]: 5.67999e-06 [updatestate_depend_eliminate]: 1.325e-05 [updatestate_assign_eliminate]: 8.07e-06 [updatestate_loads_eliminate]: 9.39998e-06 [cse]: 4.406e-05 [renormalize]: 5.79981e-07 [remove_dup_value]: 2.303e-05 [tuple_transform]: 0.00014517, [1] [Cycle 1]: 0.00014056, [4] [d_1]: 0.00010292 [none_parameter_eliminate]: 1.72999e-06 [renormalize]: 4.60015e-07 [switch_simplify]: 1.63e-05 [partial_unused_args_eliminate]: 2.19001e-06 [add_recomputation]: 0.00014659 [cse_after_recomputation]: 4.567e-05, [1] [Cycle 1]: 4.065e-05, [1] [cse]: 3.436e-05 [environ_conv]: 1.431e-05 [swap_dp_allreduce_reducescatter]: 1.24e-05 [bias_add_comm_swap]: 2.53998e-06 [label_micro_interleaved_index]: 7.52002e-06 [label_fine_grained_interleaved_index]: 3.20998e-06 [merge_cast_opt]: 1.45001e-06 [slice_recompute_activation]: 2.44999e-06 [micro_interleaved_order_control]: 2.67001e-06 [assign_add_opt]: 1.47001e-06 [ForceFp32Comm]: 8.00006e-07 [remove_cast_before_assign_add]: 1.26002e-06 [full_micro_interleaved_order_control]: 2.34001e-06 [reorder_send_recv_between_fp_bp]: 2.95998e-06 [comm_op_add_attrs]: 1.60999e-06 [add_comm_op_reuse_tag]: 1.20001e-06 [interleave_split_concat_branches]: 1.14998e-06 [interleave_parallel_branches]: 1.12e-06 [overlap_opt_shard_in_pipeline]: 1.36998e-06 [overlap_opt_shard_grad_in_pipeline]: 1.98997e-06 [control_data_broadcast_order]: 2.737e-05 [grouped_pairwise_exchange_alltoall]: 1.74e-06 [offloading_packed_experts]: 7.71001e-06 [overlap_recompute_and_grad_model_parallel]: 8.35999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.46002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.59998e-06 [overlap_recompute_comm]: 2.46e-06 [overlap_grad_ring_attention]: 7.23999e-06 [overlap_grad_flash_sp]: 3.895e-05 [begin_end_overlap_inline]: 5.29981e-07 [split_matmul_comm_elemetwise]: 2.81e-06 [split_layernorm_comm]: 1.67999e-06 [handle_group_info]: 1.26997e-06 [symbol_engine_optimizer]: 0.00014595, [1] [Cycle 1]: 0.00014069, [6] [build]: 1.241e-05 [elim_shapecalc]: 2.288e-05 [elim_not_effective]: 2.806e-05 [opt_reshape]: 1.71e-05 [fold_const_symbol]: 2.431e-05 [renormalize]: 2.00002e-07 [detach_backward]: 2.29001e-06 [pipeline_parallel_scheduler]: 1.55999e-06 [auto_monad_reorder]: 5.111e-05 [get_jit_bprop_graph]: 1.97999e-06 [rewriter_after_jit_bprop_graph]: 5.28002e-06 [opt_after_jit_grad]: 0.00063568 [validate]: 7.93e-05 Sums bootstrap : 0.000515s : 0.16% type_inference : 0.314288s : 96.45% event_method : 0.000030s : 0.01% auto_monad : 0.000366s : 0.11% graph_reusing : 0.000010s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000048s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000010s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000067s : 0.02% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000052s : 0.02% optimize.rewriter_before_opt_a : 0.000165s : 0.05% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000177s : 0.05% optimize.opt_a.loop_unroll : 0.000069s : 0.02% optimize.opt_a.a_1 : 0.002168s : 0.67% optimize.opt_a.with_stream_mark : 0.000072s : 0.02% optimize.opt_a.recompute_prepare : 0.000052s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000114s : 0.03% optimize.opt_a.updatestate_assign_eliminate : 0.000021s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000072s : 0.02% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000459s : 0.14% optimize.opt_a.accelerated_algorithm : 0.000067s : 0.02% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000009s : 0.00% optimize.opt_a.shard_inline : 0.000031s : 0.01% optimize.opt_a.merge_send_recv : 0.000030s : 0.01% optimize.opt_a.auto_parallel : 0.000033s : 0.01% optimize.opt_a.parallel : 0.000048s : 0.01% optimize.opt_a.flash_sp : 0.000019s : 0.01% optimize.opt_a.merge_comm : 0.000022s : 0.01% optimize.opt_a.allreduce_fusion : 0.000019s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000038s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000038s : 0.01% optimize.opt_a.virtual_dataset : 0.000030s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000031s : 0.01% optimize.opt_a.virtual_output : 0.000030s : 0.01% optimize.opt_a.merge_forward : 0.000023s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000040s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000068s : 0.02% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000052s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000021s : 0.01% optimize.opt_a.meta_fg_expand : 0.000040s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000009s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000054s : 0.02% optimize.opt_a.a_after_grad : 0.000050s : 0.02% optimize.opt_a.renormalize : 0.001952s : 0.60% optimize.opt_a.add_forward_monad_depend : 0.000016s : 0.00% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000065s : 0.02% optimize.opt_a.cse : 0.000270s : 0.08% optimize.opt_a.a_3 : 0.000211s : 0.06% optimize.py_interpret_to_execute_after_opt_a : 0.000028s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000220s : 0.07% optimize.convert_after_rewriter : 0.000018s : 0.01% optimize.order_py_execute_after_rewriter : 0.000011s : 0.00% optimize.mutable_eliminate : 0.000823s : 0.25% optimize.opt_b.b_1 : 0.000321s : 0.10% optimize.opt_b.b_2 : 0.000018s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000016s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000010s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000011s : 0.00% optimize.opt_b.renormalize : 0.000002s : 0.00% optimize.opt_b.cse : 0.000065s : 0.02% optimize.optimize_parallel_all_gather_comm : 0.000035s : 0.01% optimize.overlap_param_gather : 0.000003s : 0.00% optimize.cconv : 0.000041s : 0.01% optimize.loop_unroll : 0.000536s : 0.16% optimize.opt_after_cconv.c_1 : 0.000097s : 0.03% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000013s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000008s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000009s : 0.00% optimize.opt_after_cconv.cse : 0.000044s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000023s : 0.01% optimize.tuple_transform.d_1 : 0.000103s : 0.03% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000016s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000147s : 0.04% optimize.cse_after_recomputation.cse : 0.000034s : 0.01% optimize.environ_conv : 0.000014s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000012s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000008s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000002s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000027s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000008s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000008s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000007s : 0.00% optimize.overlap_grad_flash_sp : 0.000039s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000012s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000023s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000028s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000017s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000024s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000051s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000005s : 0.00% opt_after_jit_grad : 0.000636s : 0.20% validate : 0.000079s : 0.02% Time group info: ------[substitution.] 0.000582 175 7.47% : 0.000043s : 5: substitution.arithmetic_simplify 1.99% : 0.000012s : 2: substitution.depend_value_elim 0.70% : 0.000004s : 9: substitution.elim_not_effective 0.65% : 0.000004s : 9: substitution.fold_const_symbol 2.00% : 0.000012s : 12: substitution.graph_param_transform 59.52% : 0.000346s : 8: substitution.inline 1.64% : 0.000010s : 19: substitution.j_node_and_user_rematch 3.89% : 0.000023s : 4: substitution.less_batch_normalization 1.35% : 0.000008s : 8: substitution.load_eliminater 2.48% : 0.000014s : 19: substitution.remove_not_recompute_node 1.70% : 0.000010s : 6: substitution.replace_old_param 1.91% : 0.000011s : 2: substitution.switch_simplify 6.92% : 0.000040s : 34: substitution.updatestate_pure_node_eliminater 7.76% : 0.000045s : 38: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.314189 2 99.40% : 0.312294s : 1: type_inference.infer 0.60% : 0.001895s : 1: type_inference.specialize ------[replace.] 0.000139 10 55.52% : 0.000077s : 8: replace.inline 44.48% : 0.000062s : 2: replace.switch_simplify ------[match.] 0.000350 10 97.27% : 0.000340s : 8: match.inline 2.73% : 0.000010s : 2: match.switch_simplify ------[predicate.] 0.000595 3878 1.07% : 0.000006s : 45: predicate.accumulaten_eliminater 0.78% : 0.000005s : 12: predicate.ad_related_special_op_eliminate 0.59% : 0.000004s : 25: predicate.addn_check_dump 1.13% : 0.000007s : 45: predicate.addn_zero_filter 1.06% : 0.000006s : 45: predicate.adjust_all_reduce_mul_add 2.55% : 0.000015s : 70: predicate.arithmetic_simplify 1.11% : 0.000007s : 45: predicate.cast_eliminate 0.58% : 0.000003s : 24: predicate.check_bprop_eliminate 0.59% : 0.000004s : 25: predicate.compare_switch_simplify 0.15% : 0.000001s : 12: predicate.const_output_eliminate 0.70% : 0.000004s : 25: predicate.depend_value_elim 1.14% : 0.000007s : 45: predicate.dict_get_item_const_eliminator 1.15% : 0.000007s : 45: predicate.dict_get_item_eliminator 1.08% : 0.000006s : 45: predicate.dict_set_item_eliminator 0.66% : 0.000004s : 24: predicate.dumpgradient_eliminate 0.21% : 0.000001s : 12: predicate.elim_not_effective 0.38% : 0.000002s : 12: predicate.elim_shapecalc_of_broadcastargs 1.31% : 0.000008s : 57: predicate.environ_add_const_eliminate 1.31% : 0.000008s : 57: predicate.environ_get_add_eliminate 1.28% : 0.000008s : 57: predicate.environ_get_depend_swap 1.99% : 0.000012s : 82: predicate.environ_get_eliminate 1.33% : 0.000008s : 57: predicate.environ_get_set_eliminate 1.20% : 0.000007s : 53: predicate.exchange_switch_depend_value 1.92% : 0.000011s : 53: predicate.float_depend_g_call 0.64% : 0.000004s : 25: predicate.float_environ_get_switch 0.85% : 0.000005s : 37: predicate.float_tuple_getitem_switch 0.16% : 0.000001s : 12: predicate.fold_const_symbol 0.69% : 0.000004s : 25: predicate.get_grad_eliminate 0.21% : 0.000001s : 12: predicate.graph_param_transform 0.61% : 0.000004s : 25: predicate.incorporate_call 0.53% : 0.000003s : 25: predicate.incorporate_call_switch 5.49% : 0.000033s : 172: predicate.inline 0.89% : 0.000005s : 25: predicate.inline_without_move 0.29% : 0.000002s : 25: predicate.j_node_and_user_rematch 0.94% : 0.000006s : 25: predicate.less_batch_normalization 1.68% : 0.000010s : 69: predicate.list_to_tuple_eliminator_ 2.63% : 0.000016s : 114: predicate.load_eliminater 0.79% : 0.000005s : 12: predicate.loop_unroll_after_grad 1.78% : 0.000011s : 75: predicate.loop_unroll_before_grad 1.66% : 0.000010s : 69: predicate.make_slice_get_slice_eliminator 0.61% : 0.000004s : 25: predicate.merge_addn 0.56% : 0.000003s : 24: predicate.micro_step_allgather_replace 0.68% : 0.000004s : 24: predicate.mini_step_allgather_replace 1.02% : 0.000006s : 45: predicate.minmaximum_grad 1.06% : 0.000006s : 12: predicate.mutable_eliminate 0.30% : 0.000002s : 12: predicate.opt_reshape 0.32% : 0.000002s : 12: predicate.parallel_virtual_node 1.64% : 0.000010s : 53: predicate.partial_defer_inline 1.45% : 0.000009s : 57: predicate.partial_eliminate 1.07% : 0.000006s : 45: predicate.print_const_string_wrapper 0.60% : 0.000004s : 25: predicate.reduce_all_const_elim 1.46% : 0.000009s : 45: predicate.reduce_eliminate 2.56% : 0.000015s : 114: predicate.redundant_stop_gradient_eliminater 0.47% : 0.000003s : 25: predicate.remove_not_recompute_node 1.14% : 0.000007s : 69: predicate.replace_applicator 0.39% : 0.000002s : 25: predicate.replace_old_param 0.21% : 0.000001s : 12: predicate.reset_defer_inline 1.05% : 0.000006s : 45: predicate.reshape_eliminate 0.58% : 0.000003s : 24: predicate.row_tensor_add_zeros_like 0.35% : 0.000002s : 12: predicate.row_tensor_eliminate 0.79% : 0.000005s : 24: predicate.same_eliminate 0.52% : 0.000003s : 34: predicate.set_cell_output_no_recompute 0.87% : 0.000005s : 25: predicate.shard_identity_eliminate 0.60% : 0.000004s : 24: predicate.special_op_eliminate 0.71% : 0.000004s : 25: predicate.specialize_transform 0.94% : 0.000006s : 24: predicate.split_environ_get_set_with_tuple_value 0.76% : 0.000005s : 25: predicate.stack_unstack_eliminate 0.33% : 0.000002s : 12: predicate.switch_call_monad_eliminater 1.31% : 0.000008s : 53: predicate.switch_defer_inline 1.81% : 0.000011s : 77: predicate.switch_layer_defer_inline 4.48% : 0.000027s : 169: predicate.switch_simplify 1.03% : 0.000006s : 45: predicate.tile_eliminate 1.03% : 0.000006s : 45: predicate.transpose_eliminate 1.71% : 0.000010s : 69: predicate.tuple_list_convert_item_index_to_positive 1.71% : 0.000010s : 69: predicate.tuple_list_get_item_const_eliminator 1.58% : 0.000009s : 69: predicate.tuple_list_get_item_depend_reorder 2.79% : 0.000017s : 94: predicate.tuple_list_get_item_eliminator 1.70% : 0.000010s : 69: predicate.tuple_list_get_set_item_eliminator 2.49% : 0.000015s : 94: predicate.tuple_list_set_item_eliminator 1.59% : 0.000009s : 69: predicate.tuple_to_list_eliminator_ 2.74% : 0.000016s : 114: predicate.updatestate_pure_node_eliminater 3.49% : 0.000021s : 139: predicate.updatestate_useless_node_eliminater 0.32% : 0.000002s : 12: predicate.value_based_eliminate 0.70% : 0.000004s : 25: predicate.virtual_dataset_eliminate 0.66% : 0.000004s : 25: predicate.virtual_output_eliminate 0.29% : 0.000002s : 12: predicate.virtual_view_grad_eliminate 0.35% : 0.000002s : 12: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.001107 12 33.72% : 0.000373s : 2: func_graph_cloner_run.FuncGraphClonerGraph 66.28% : 0.000734s : 10: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.353229 192 0.00% : 0.000003s : 1: ForceFp32Comm 1.32% : 0.004680s : 1: add_attr 1.32% : 0.004663s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.04% : 0.000153s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.11% : 0.000380s : 1: auto_monad 0.02% : 0.000057s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000005s : 1: bias_add_comm_swap 0.16% : 0.000553s : 1: bootstrap 0.01% : 0.000046s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000031s : 1: control_data_broadcast_order 0.01% : 0.000023s : 1: convert_after_rewriter 0.01% : 0.000049s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.01% : 0.000018s : 1: environ_conv 0.01% : 0.000038s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000015s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000010s : 1: label_micro_interleaved_index 0.16% : 0.000549s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.24% : 0.000841s : 1: mutable_eliminate 0.00% : 0.000011s : 1: offloading_packed_experts 0.01% : 0.000024s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000032s : 1: opt.transform.mutable_eliminate 0.98% : 0.003467s : 78: opt.transform.opt_a 0.03% : 0.000096s : 1: opt.transform.opt_after_cconv 0.02% : 0.000053s : 1: opt.transform.opt_after_jit_grad 0.09% : 0.000305s : 28: opt.transform.opt_b 0.03% : 0.000116s : 2: opt.transform.opt_trans_graph 0.02% : 0.000088s : 4: opt.transform.symbol_engine_opt 2.05% : 0.007245s : 1: opt_a 0.06% : 0.000226s : 1: opt_after_cconv 0.18% : 0.000648s : 1: opt_after_jit_grad 0.14% : 0.000497s : 1: opt_b 3.07% : 0.010843s : 1: optimize 0.01% : 0.000040s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000015s : 1: order_py_execute_after_rewriter 0.01% : 0.000042s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000011s : 1: overlap_grad_ring_attention 0.00% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000011s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.02% : 0.000073s : 1: pre_auto_parallel 0.02% : 0.000058s : 1: py_interpret_to_execute 0.01% : 0.000033s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.01% : 0.000027s : 1: remove_dup_value 0.30% : 0.001067s : 1: renormalize.infer 0.25% : 0.000871s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.06% : 0.000229s : 1: rewriter_after_opt_a 0.05% : 0.000172s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000016s : 1: swap_dp_allreduce_reducescatter 0.04% : 0.000149s : 1: symbol_engine_optimizer 0.04% : 0.000148s : 1: tuple_transform 88.98% : 0.314321s : 1: type_inference TotalTime = 0.409247, [21] [bootstrap]: 0.0005115 [type_inference]: 0.232405 [event_method]: 2.818e-05 [auto_monad]: 0.00043559 [graph_reusing]: 9.35001e-06 [inline]: 3.28998e-06 [add_attr]: 0.00422278, [1] [add_attr_with_inline]: 0.00420974, [1] [Cycle 1]: 9.898e-05, [2] [tag_attr]: 4.426e-05 [meta_addattr_fg_expand]: 1.01e-05 [parallel-infer-symbol]: 3.77998e-06 [pre_auto_parallel]: 6.21e-05 [insert-virtual-dataset]: 2.64999e-06 [parallel-infer-symbol-second]: 8.50006e-07 [dataset_repeat_opt]: 2.16998e-06 [pipeline_split]: 1.66998e-06 [optimize]: 0.170464, [53] [py_interpret_to_execute]: 4.812e-05 [rewriter_before_opt_a]: 0.00014991 [opt_a]: 0.166584, [2] [Cycle 1]: 0.164907, [45] [expand_dump_flag]: 3.83999e-06 [switch_simplify]: 0.0001508 [loop_unroll]: 5.426e-05 [a_1]: 0.00160383 [with_stream_mark]: 3.761e-05 [recompute_prepare]: 2.961e-05 [updatestate_depend_eliminate]: 9.323e-05 [updatestate_assign_eliminate]: 1.176e-05 [updatestate_loads_eliminate]: 5.43e-05 [parameter_eliminate]: 3.13e-06 [a_2]: 0.00023039 [accelerated_algorithm]: 3.84e-05 [shard]: 1.96e-06 [meta_shard_fg_expand]: 4.59002e-06 [shard_inline]: 1.555e-05 [merge_send_recv]: 1.62e-05 [auto_parallel]: 1.457e-05 [parallel]: 2.374e-05 [flash_sp]: 1.421e-05 [merge_comm]: 9.22001e-06 [allreduce_fusion]: 8.42998e-06 [matmul_add_comm_reduction]: 1.862e-05 [allreduce_slice_to_reducescatter]: 8.70001e-07 [virtual_shard_identity]: 1.812e-05 [virtual_dataset]: 1.579e-05 [get_grad_eliminate_]: 1.485e-05 [virtual_output]: 1.583e-05 [merge_forward]: 1.019e-05 [cell_reuse_recompute_pass]: 1.90001e-06 [offload_activation]: 1.841e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.121e-05 [merge_recompute_call_nodes]: 1.52001e-06 [before_grad]: 2.831e-05 [set_forward_comm_id_for_comm_node_pass]: 1.04e-05 [meta_fg_expand]: 8.02003e-06 [flash_sp_send_recv_attached]: 5.67999e-06 [receive_attached]: 2.16e-06 [after_resolve]: 2.54e-05 [a_after_grad]: 2.682e-05 [renormalize]: 0.161402 [add_forward_monad_depend]: 1.413e-05 [auto_monad_grad]: 3.95e-06 [auto_monad_eliminator]: 4.734e-05 [cse]: 0.00025968 [a_3]: 0.00012702 [Cycle 2]: 0.00166111, [45] [expand_dump_flag]: 3.14001e-06 [switch_simplify]: 2.08e-05 [loop_unroll]: 1.742e-05 [a_1]: 0.00045391 [with_stream_mark]: 3.212e-05 [recompute_prepare]: 1.851e-05 [updatestate_depend_eliminate]: 1.099e-05 [updatestate_assign_eliminate]: 1.123e-05 [updatestate_loads_eliminate]: 1.333e-05 [parameter_eliminate]: 2.71e-06 [a_2]: 0.0002139 [accelerated_algorithm]: 2.544e-05 [shard]: 2.29001e-06 [meta_shard_fg_expand]: 3.79002e-06 [shard_inline]: 1.617e-05 [merge_send_recv]: 1.771e-05 [auto_parallel]: 1.788e-05 [parallel]: 1.131e-05 [flash_sp]: 4.69998e-06 [merge_comm]: 1.113e-05 [allreduce_fusion]: 9.87999e-06 [matmul_add_comm_reduction]: 2.136e-05 [allreduce_slice_to_reducescatter]: 1.27e-06 [virtual_shard_identity]: 2.052e-05 [virtual_dataset]: 1.562e-05 [get_grad_eliminate_]: 1.552e-05 [virtual_output]: 1.445e-05 [merge_forward]: 1.169e-05 [cell_reuse_recompute_pass]: 3.36001e-06 [offload_activation]: 2.059e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.342e-05 [merge_recompute_call_nodes]: 2.09e-06 [before_grad]: 2.492e-05 [set_forward_comm_id_for_comm_node_pass]: 1.116e-05 [meta_fg_expand]: 6.09999e-06 [flash_sp_send_recv_attached]: 1.79e-06 [receive_attached]: 2.49001e-06 [after_resolve]: 2.447e-05 [a_after_grad]: 2.47e-05 [renormalize]: 6.00121e-08 [add_forward_monad_depend]: 3.93999e-06 [auto_monad_grad]: 2.83e-06 [auto_monad_eliminator]: 3.253e-05 [cse]: 6.053e-05 [a_3]: 0.00010153 [py_interpret_to_execute_after_opt_a]: 2.736e-05 [slice_cell_reuse_recomputed_activation]: 2.76e-06 [rewriter_after_opt_a]: 0.00023057 [convert_after_rewriter]: 2.012e-05 [order_py_execute_after_rewriter]: 1.097e-05 [mutable_eliminate]: 0.00090467 [opt_b]: 0.00053457, [1] [Cycle 1]: 0.00052427, [7] [b_1]: 0.00034456 [b_2]: 1.854e-05 [updatestate_depend_eliminate]: 1.831e-05 [updatestate_assign_eliminate]: 9.34e-06 [updatestate_loads_eliminate]: 1.217e-05 [renormalize]: 8.00006e-07 [cse]: 7.36e-05 [optimize_parallel_all_gather_comm]: 3.775e-05 [overlap_param_gather]: 2.27999e-06 [cconv]: 4.3e-05 [loop_unroll]: 0.00062324 [opt_after_cconv]: 0.00023229, [1] [Cycle 1]: 0.00022397, [7] [c_1]: 9.937e-05 [parameter_eliminate]: 5.92999e-06 [updatestate_depend_eliminate]: 1.391e-05 [updatestate_assign_eliminate]: 8.15e-06 [updatestate_loads_eliminate]: 1e-05 [cse]: 4.856e-05 [renormalize]: 7.49977e-07 [remove_dup_value]: 2.328e-05 [tuple_transform]: 0.00014842, [1] [Cycle 1]: 0.00014293, [4] [d_1]: 0.00010436 [none_parameter_eliminate]: 2.14e-06 [renormalize]: 1.90019e-07 [switch_simplify]: 1.661e-05 [partial_unused_args_eliminate]: 1.95001e-06 [add_recomputation]: 0.00010961 [cse_after_recomputation]: 4.935e-05, [1] [Cycle 1]: 4.373e-05, [1] [cse]: 3.602e-05 [environ_conv]: 1.687e-05 [swap_dp_allreduce_reducescatter]: 1.281e-05 [bias_add_comm_swap]: 3.28e-06 [label_micro_interleaved_index]: 7.36999e-06 [label_fine_grained_interleaved_index]: 2.92002e-06 [merge_cast_opt]: 1.66002e-06 [slice_recompute_activation]: 2.55002e-06 [micro_interleaved_order_control]: 2.64999e-06 [assign_add_opt]: 1.24e-06 [ForceFp32Comm]: 8.70001e-07 [remove_cast_before_assign_add]: 8.99978e-07 [full_micro_interleaved_order_control]: 2.51e-06 [reorder_send_recv_between_fp_bp]: 3.11999e-06 [comm_op_add_attrs]: 1.12e-06 [add_comm_op_reuse_tag]: 1.05999e-06 [interleave_split_concat_branches]: 1.39998e-06 [interleave_parallel_branches]: 1.04e-06 [overlap_opt_shard_in_pipeline]: 1.39e-06 [overlap_opt_shard_grad_in_pipeline]: 1.87999e-06 [control_data_broadcast_order]: 0.00010255 [grouped_pairwise_exchange_alltoall]: 2.19001e-06 [offloading_packed_experts]: 8.47e-06 [overlap_recompute_and_grad_model_parallel]: 8.45001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.21002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.62999e-06 [overlap_recompute_comm]: 2.42001e-06 [overlap_grad_ring_attention]: 8.06001e-06 [overlap_grad_flash_sp]: 4.125e-05 [begin_end_overlap_inline]: 5.3001e-07 [split_matmul_comm_elemetwise]: 2.91999e-06 [split_layernorm_comm]: 1.70001e-06 [handle_group_info]: 1.04e-06 [symbol_engine_optimizer]: 0.00015634, [1] [Cycle 1]: 0.00014954, [6] [build]: 1.665e-05 [elim_shapecalc]: 2.572e-05 [elim_not_effective]: 2.833e-05 [opt_reshape]: 1.746e-05 [fold_const_symbol]: 2.417e-05 [renormalize]: 2.69996e-07 [detach_backward]: 2.06998e-06 [pipeline_parallel_scheduler]: 1.59998e-06 [auto_monad_reorder]: 5.725e-05 [get_jit_bprop_graph]: 2.57001e-06 [rewriter_after_jit_bprop_graph]: 9.04998e-06 [opt_after_jit_grad]: 0.00069774 [validate]: 8.858e-05 Sums bootstrap : 0.000511s : 0.13% type_inference : 0.232405s : 57.57% event_method : 0.000028s : 0.01% auto_monad : 0.000436s : 0.11% graph_reusing : 0.000009s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000044s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000010s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000062s : 0.02% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000048s : 0.01% optimize.rewriter_before_opt_a : 0.000150s : 0.04% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000172s : 0.04% optimize.opt_a.loop_unroll : 0.000072s : 0.02% optimize.opt_a.a_1 : 0.002058s : 0.51% optimize.opt_a.with_stream_mark : 0.000070s : 0.02% optimize.opt_a.recompute_prepare : 0.000048s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000104s : 0.03% optimize.opt_a.updatestate_assign_eliminate : 0.000023s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000068s : 0.02% optimize.opt_a.parameter_eliminate : 0.000006s : 0.00% optimize.opt_a.a_2 : 0.000444s : 0.11% optimize.opt_a.accelerated_algorithm : 0.000064s : 0.02% optimize.opt_a.shard : 0.000004s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000008s : 0.00% optimize.opt_a.shard_inline : 0.000032s : 0.01% optimize.opt_a.merge_send_recv : 0.000034s : 0.01% optimize.opt_a.auto_parallel : 0.000032s : 0.01% optimize.opt_a.parallel : 0.000035s : 0.01% optimize.opt_a.flash_sp : 0.000019s : 0.00% optimize.opt_a.merge_comm : 0.000020s : 0.01% optimize.opt_a.allreduce_fusion : 0.000018s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000040s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000039s : 0.01% optimize.opt_a.virtual_dataset : 0.000031s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000030s : 0.01% optimize.opt_a.virtual_output : 0.000030s : 0.01% optimize.opt_a.merge_forward : 0.000022s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000039s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000065s : 0.02% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000053s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000022s : 0.01% optimize.opt_a.meta_fg_expand : 0.000014s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000007s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000050s : 0.01% optimize.opt_a.a_after_grad : 0.000052s : 0.01% optimize.opt_a.renormalize : 0.161402s : 39.98% optimize.opt_a.add_forward_monad_depend : 0.000018s : 0.00% optimize.opt_a.auto_monad_grad : 0.000007s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000080s : 0.02% optimize.opt_a.cse : 0.000320s : 0.08% optimize.opt_a.a_3 : 0.000229s : 0.06% optimize.py_interpret_to_execute_after_opt_a : 0.000027s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000231s : 0.06% optimize.convert_after_rewriter : 0.000020s : 0.00% optimize.order_py_execute_after_rewriter : 0.000011s : 0.00% optimize.mutable_eliminate : 0.000905s : 0.22% optimize.opt_b.b_1 : 0.000345s : 0.09% optimize.opt_b.b_2 : 0.000019s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000018s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000009s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000012s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000074s : 0.02% optimize.optimize_parallel_all_gather_comm : 0.000038s : 0.01% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000043s : 0.01% optimize.loop_unroll : 0.000623s : 0.15% optimize.opt_after_cconv.c_1 : 0.000099s : 0.02% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000014s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000008s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000010s : 0.00% optimize.opt_after_cconv.cse : 0.000049s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000023s : 0.01% optimize.tuple_transform.d_1 : 0.000104s : 0.03% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000017s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000110s : 0.03% optimize.cse_after_recomputation.cse : 0.000036s : 0.01% optimize.environ_conv : 0.000017s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000013s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000007s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000103s : 0.03% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000008s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000008s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000008s : 0.00% optimize.overlap_grad_flash_sp : 0.000041s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000017s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000026s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000028s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000017s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000024s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000057s : 0.01% get_jit_bprop_graph : 0.000003s : 0.00% rewriter_after_jit_bprop_graph : 0.000009s : 0.00% opt_after_jit_grad : 0.000698s : 0.17% validate : 0.000089s : 0.02% Time group info: ------[substitution.] 0.000555 175 8.09% : 0.000045s : 5: substitution.arithmetic_simplify 2.00% : 0.000011s : 2: substitution.depend_value_elim 0.70% : 0.000004s : 9: substitution.elim_not_effective 0.65% : 0.000004s : 9: substitution.fold_const_symbol 2.04% : 0.000011s : 12: substitution.graph_param_transform 56.98% : 0.000316s : 8: substitution.inline 1.85% : 0.000010s : 19: substitution.j_node_and_user_rematch 3.77% : 0.000021s : 4: substitution.less_batch_normalization 1.47% : 0.000008s : 8: substitution.load_eliminater 2.50% : 0.000014s : 19: substitution.remove_not_recompute_node 1.66% : 0.000009s : 6: substitution.replace_old_param 2.06% : 0.000011s : 2: substitution.switch_simplify 8.00% : 0.000044s : 34: substitution.updatestate_pure_node_eliminater 8.20% : 0.000046s : 38: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.232316 2 99.30% : 0.230700s : 1: type_inference.infer 0.70% : 0.001617s : 1: type_inference.specialize ------[replace.] 0.000125 10 54.72% : 0.000068s : 8: replace.inline 45.28% : 0.000057s : 2: replace.switch_simplify ------[match.] 0.000320 10 96.90% : 0.000310s : 8: match.inline 3.10% : 0.000010s : 2: match.switch_simplify ------[predicate.] 0.000600 3878 1.08% : 0.000006s : 45: predicate.accumulaten_eliminater 0.82% : 0.000005s : 12: predicate.ad_related_special_op_eliminate 0.65% : 0.000004s : 25: predicate.addn_check_dump 1.15% : 0.000007s : 45: predicate.addn_zero_filter 1.04% : 0.000006s : 45: predicate.adjust_all_reduce_mul_add 2.49% : 0.000015s : 70: predicate.arithmetic_simplify 1.05% : 0.000006s : 45: predicate.cast_eliminate 0.60% : 0.000004s : 24: predicate.check_bprop_eliminate 0.56% : 0.000003s : 25: predicate.compare_switch_simplify 0.15% : 0.000001s : 12: predicate.const_output_eliminate 0.58% : 0.000004s : 25: predicate.depend_value_elim 1.10% : 0.000007s : 45: predicate.dict_get_item_const_eliminator 1.20% : 0.000007s : 45: predicate.dict_get_item_eliminator 1.05% : 0.000006s : 45: predicate.dict_set_item_eliminator 0.86% : 0.000005s : 24: predicate.dumpgradient_eliminate 0.18% : 0.000001s : 12: predicate.elim_not_effective 0.38% : 0.000002s : 12: predicate.elim_shapecalc_of_broadcastargs 1.36% : 0.000008s : 57: predicate.environ_add_const_eliminate 1.28% : 0.000008s : 57: predicate.environ_get_add_eliminate 1.26% : 0.000008s : 57: predicate.environ_get_depend_swap 1.92% : 0.000012s : 82: predicate.environ_get_eliminate 1.28% : 0.000008s : 57: predicate.environ_get_set_eliminate 1.19% : 0.000007s : 53: predicate.exchange_switch_depend_value 1.87% : 0.000011s : 53: predicate.float_depend_g_call 0.57% : 0.000003s : 25: predicate.float_environ_get_switch 0.88% : 0.000005s : 37: predicate.float_tuple_getitem_switch 0.16% : 0.000001s : 12: predicate.fold_const_symbol 0.67% : 0.000004s : 25: predicate.get_grad_eliminate 0.19% : 0.000001s : 12: predicate.graph_param_transform 0.57% : 0.000003s : 25: predicate.incorporate_call 0.52% : 0.000003s : 25: predicate.incorporate_call_switch 5.51% : 0.000033s : 172: predicate.inline 0.83% : 0.000005s : 25: predicate.inline_without_move 0.30% : 0.000002s : 25: predicate.j_node_and_user_rematch 0.86% : 0.000005s : 25: predicate.less_batch_normalization 1.64% : 0.000010s : 69: predicate.list_to_tuple_eliminator_ 2.69% : 0.000016s : 114: predicate.load_eliminater 0.81% : 0.000005s : 12: predicate.loop_unroll_after_grad 1.82% : 0.000011s : 75: predicate.loop_unroll_before_grad 1.66% : 0.000010s : 69: predicate.make_slice_get_slice_eliminator 0.61% : 0.000004s : 25: predicate.merge_addn 0.68% : 0.000004s : 24: predicate.micro_step_allgather_replace 0.58% : 0.000003s : 24: predicate.mini_step_allgather_replace 0.96% : 0.000006s : 45: predicate.minmaximum_grad 1.06% : 0.000006s : 12: predicate.mutable_eliminate 0.32% : 0.000002s : 12: predicate.opt_reshape 0.36% : 0.000002s : 12: predicate.parallel_virtual_node 1.75% : 0.000010s : 53: predicate.partial_defer_inline 1.43% : 0.000009s : 57: predicate.partial_eliminate 1.04% : 0.000006s : 45: predicate.print_const_string_wrapper 0.64% : 0.000004s : 25: predicate.reduce_all_const_elim 1.26% : 0.000008s : 45: predicate.reduce_eliminate 2.57% : 0.000015s : 114: predicate.redundant_stop_gradient_eliminater 0.40% : 0.000002s : 25: predicate.remove_not_recompute_node 1.14% : 0.000007s : 69: predicate.replace_applicator 0.42% : 0.000003s : 25: predicate.replace_old_param 0.18% : 0.000001s : 12: predicate.reset_defer_inline 1.18% : 0.000007s : 45: predicate.reshape_eliminate 0.60% : 0.000004s : 24: predicate.row_tensor_add_zeros_like 0.40% : 0.000002s : 12: predicate.row_tensor_eliminate 0.93% : 0.000006s : 24: predicate.same_eliminate 0.49% : 0.000003s : 34: predicate.set_cell_output_no_recompute 0.76% : 0.000005s : 25: predicate.shard_identity_eliminate 0.62% : 0.000004s : 24: predicate.special_op_eliminate 0.69% : 0.000004s : 25: predicate.specialize_transform 0.97% : 0.000006s : 24: predicate.split_environ_get_set_with_tuple_value 0.84% : 0.000005s : 25: predicate.stack_unstack_eliminate 0.30% : 0.000002s : 12: predicate.switch_call_monad_eliminater 1.28% : 0.000008s : 53: predicate.switch_defer_inline 1.84% : 0.000011s : 77: predicate.switch_layer_defer_inline 4.43% : 0.000027s : 169: predicate.switch_simplify 1.03% : 0.000006s : 45: predicate.tile_eliminate 1.03% : 0.000006s : 45: predicate.transpose_eliminate 1.79% : 0.000011s : 69: predicate.tuple_list_convert_item_index_to_positive 1.77% : 0.000011s : 69: predicate.tuple_list_get_item_const_eliminator 1.67% : 0.000010s : 69: predicate.tuple_list_get_item_depend_reorder 2.90% : 0.000017s : 94: predicate.tuple_list_get_item_eliminator 1.63% : 0.000010s : 69: predicate.tuple_list_get_set_item_eliminator 2.39% : 0.000014s : 94: predicate.tuple_list_set_item_eliminator 1.66% : 0.000010s : 69: predicate.tuple_to_list_eliminator_ 2.81% : 0.000017s : 114: predicate.updatestate_pure_node_eliminater 3.43% : 0.000021s : 139: predicate.updatestate_useless_node_eliminater 0.36% : 0.000002s : 12: predicate.value_based_eliminate 0.68% : 0.000004s : 25: predicate.virtual_dataset_eliminate 0.64% : 0.000004s : 25: predicate.virtual_output_eliminate 0.30% : 0.000002s : 12: predicate.virtual_view_grad_eliminate 0.38% : 0.000002s : 12: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.001122 12 35.01% : 0.000393s : 2: func_graph_cloner_run.FuncGraphClonerGraph 64.99% : 0.000729s : 10: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.749209 192 0.00% : 0.000004s : 1: ForceFp32Comm 0.56% : 0.004230s : 1: add_attr 0.56% : 0.004214s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.02% : 0.000114s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.06% : 0.000447s : 1: auto_monad 0.01% : 0.000065s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.07% : 0.000551s : 1: bootstrap 0.01% : 0.000047s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000111s : 1: control_data_broadcast_order 0.00% : 0.000026s : 1: convert_after_rewriter 0.01% : 0.000053s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000020s : 1: environ_conv 0.00% : 0.000036s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000014s : 1: graph_reusing 0.00% : 0.000006s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000010s : 1: label_micro_interleaved_index 0.08% : 0.000636s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.12% : 0.000923s : 1: mutable_eliminate 0.00% : 0.000012s : 1: offloading_packed_experts 0.00% : 0.000028s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000034s : 1: opt.transform.mutable_eliminate 0.45% : 0.003339s : 78: opt.transform.opt_a 0.01% : 0.000098s : 1: opt.transform.opt_after_cconv 0.01% : 0.000057s : 1: opt.transform.opt_after_jit_grad 0.04% : 0.000325s : 28: opt.transform.opt_b 0.02% : 0.000118s : 2: opt.transform.opt_trans_graph 0.01% : 0.000091s : 4: opt.transform.symbol_engine_opt 22.24% : 0.166588s : 1: opt_a 0.03% : 0.000236s : 1: opt_after_cconv 0.10% : 0.000734s : 1: opt_after_jit_grad 0.07% : 0.000539s : 1: opt_b 22.75% : 0.170472s : 1: optimize 0.01% : 0.000042s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000015s : 1: order_py_execute_after_rewriter 0.01% : 0.000045s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000012s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000011s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000067s : 1: pre_auto_parallel 0.01% : 0.000052s : 1: py_interpret_to_execute 0.00% : 0.000034s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000027s : 1: remove_dup_value 21.39% : 0.160226s : 1: renormalize.infer 0.15% : 0.001154s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.03% : 0.000243s : 1: rewriter_after_opt_a 0.02% : 0.000155s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000016s : 1: swap_dp_allreduce_reducescatter 0.02% : 0.000159s : 1: symbol_engine_optimizer 0.02% : 0.000151s : 1: tuple_transform 31.02% : 0.232440s : 1: type_inference TotalTime = 0.762389, [21] [bootstrap]: 0.00052547 [type_inference]: 0.579458 [event_method]: 0.00018375 [auto_monad]: 0.0003024 [graph_reusing]: 8.12e-06 [inline]: 3.43e-06 [add_attr]: 0.00462842, [1] [add_attr_with_inline]: 0.00461306, [1] [Cycle 1]: 0.00010456, [2] [tag_attr]: 4.075e-05 [meta_addattr_fg_expand]: 8.96002e-06 [parallel-infer-symbol]: 3.92002e-06 [pre_auto_parallel]: 5.845e-05 [insert-virtual-dataset]: 3.04999e-06 [parallel-infer-symbol-second]: 7.89994e-07 [dataset_repeat_opt]: 2.19001e-06 [pipeline_split]: 1.64e-06 [optimize]: 0.176034, [53] [py_interpret_to_execute]: 4.973e-05 [rewriter_before_opt_a]: 0.00013745 [opt_a]: 0.00781501, [2] [Cycle 1]: 0.00581712, [45] [expand_dump_flag]: 4.3e-06 [switch_simplify]: 6.828e-05 [loop_unroll]: 5.193e-05 [a_1]: 0.00178432 [with_stream_mark]: 3.67e-05 [recompute_prepare]: 3.31e-05 [updatestate_depend_eliminate]: 8.214e-05 [updatestate_assign_eliminate]: 1.18e-05 [updatestate_loads_eliminate]: 1.092e-05 [parameter_eliminate]: 3.21001e-06 [a_2]: 0.00029391 [accelerated_algorithm]: 5.114e-05 [shard]: 3.09999e-06 [meta_shard_fg_expand]: 6.26e-06 [shard_inline]: 2.244e-05 [merge_send_recv]: 1.943e-05 [auto_parallel]: 1.856e-05 [parallel]: 2.57e-05 [flash_sp]: 1.624e-05 [merge_comm]: 1.264e-05 [allreduce_fusion]: 1.134e-05 [matmul_add_comm_reduction]: 2.165e-05 [allreduce_slice_to_reducescatter]: 1.27999e-06 [virtual_shard_identity]: 2.537e-05 [virtual_dataset]: 1.936e-05 [get_grad_eliminate_]: 1.971e-05 [virtual_output]: 1.952e-05 [merge_forward]: 1.215e-05 [cell_reuse_recompute_pass]: 1.22e-06 [offload_activation]: 2.229e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.578e-05 [merge_recompute_call_nodes]: 2.01e-06 [before_grad]: 3.737e-05 [set_forward_comm_id_for_comm_node_pass]: 1.618e-05 [meta_fg_expand]: 8.94e-06 [flash_sp_send_recv_attached]: 6.39001e-06 [receive_attached]: 2.38998e-06 [after_resolve]: 3.107e-05 [a_after_grad]: 3.251e-05 [renormalize]: 0.00218571 [add_forward_monad_depend]: 1.299e-05 [auto_monad_grad]: 3.22002e-06 [auto_monad_eliminator]: 4.581e-05 [cse]: 0.00013634 [a_3]: 0.00014843 [Cycle 2]: 0.00197964, [45] [expand_dump_flag]: 2.79001e-06 [switch_simplify]: 2.267e-05 [loop_unroll]: 1.854e-05 [a_1]: 0.00056194 [with_stream_mark]: 3.274e-05 [recompute_prepare]: 2.673e-05 [updatestate_depend_eliminate]: 1.24e-05 [updatestate_assign_eliminate]: 1.074e-05 [updatestate_loads_eliminate]: 1.01e-05 [parameter_eliminate]: 2.70997e-06 [a_2]: 0.00027072 [accelerated_algorithm]: 3.408e-05 [shard]: 2.04999e-06 [meta_shard_fg_expand]: 6.36e-06 [shard_inline]: 1.913e-05 [merge_send_recv]: 1.936e-05 [auto_parallel]: 2.004e-05 [parallel]: 1.043e-05 [flash_sp]: 4.4e-06 [merge_comm]: 1.379e-05 [allreduce_fusion]: 1.235e-05 [matmul_add_comm_reduction]: 2.076e-05 [allreduce_slice_to_reducescatter]: 8.70001e-07 [virtual_shard_identity]: 2.688e-05 [virtual_dataset]: 1.852e-05 [get_grad_eliminate_]: 1.893e-05 [virtual_output]: 1.978e-05 [merge_forward]: 1.308e-05 [cell_reuse_recompute_pass]: 3.39001e-06 [offload_activation]: 2.156e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.193e-05 [merge_recompute_call_nodes]: 1.54e-06 [before_grad]: 3.345e-05 [set_forward_comm_id_for_comm_node_pass]: 1.291e-05 [meta_fg_expand]: 8.64998e-06 [flash_sp_send_recv_attached]: 2.45002e-06 [receive_attached]: 2.52001e-06 [after_resolve]: 2.927e-05 [a_after_grad]: 2.98e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 6.66e-06 [auto_monad_grad]: 2.35002e-06 [auto_monad_eliminator]: 3.702e-05 [cse]: 6.942e-05 [a_3]: 0.00012997 [py_interpret_to_execute_after_opt_a]: 3.262e-05 [slice_cell_reuse_recomputed_activation]: 2.46998e-06 [rewriter_after_opt_a]: 0.00023594 [convert_after_rewriter]: 2.204e-05 [order_py_execute_after_rewriter]: 1.205e-05 [mutable_eliminate]: 0.00084925 [opt_b]: 0.164368, [1] [Cycle 1]: 0.164354, [7] [b_1]: 0.00044712 [b_2]: 2.191e-05 [updatestate_depend_eliminate]: 1.916e-05 [updatestate_assign_eliminate]: 1.033e-05 [updatestate_loads_eliminate]: 9.93998e-06 [renormalize]: 1.97999e-06 [cse]: 0.00015468 [optimize_parallel_all_gather_comm]: 5.33e-05 [overlap_param_gather]: 2.25002e-06 [cconv]: 4.83e-05 [loop_unroll]: 0.00087105 [opt_after_cconv]: 0.00031806, [1] [Cycle 1]: 0.0003064, [7] [c_1]: 0.00014642 [parameter_eliminate]: 8.52e-06 [updatestate_depend_eliminate]: 2.054e-05 [updatestate_assign_eliminate]: 1.074e-05 [updatestate_loads_eliminate]: 9.84001e-06 [cse]: 6.772e-05 [renormalize]: 4.10015e-07 [remove_dup_value]: 7.049e-05 [tuple_transform]: 0.00020678, [1] [Cycle 1]: 0.00019875, [4] [d_1]: 0.00015003 [none_parameter_eliminate]: 3.04999e-06 [renormalize]: 4.09986e-07 [switch_simplify]: 2.105e-05 [partial_unused_args_eliminate]: 2.88e-06 [add_recomputation]: 0.00014762 [cse_after_recomputation]: 6.234e-05, [1] [Cycle 1]: 5.615e-05, [1] [cse]: 4.77e-05 [environ_conv]: 2.265e-05 [swap_dp_allreduce_reducescatter]: 1.773e-05 [bias_add_comm_swap]: 3.51001e-06 [label_micro_interleaved_index]: 8.32e-06 [label_fine_grained_interleaved_index]: 2.66e-06 [merge_cast_opt]: 1.57001e-06 [slice_recompute_activation]: 2.39001e-06 [micro_interleaved_order_control]: 2.71999e-06 [assign_add_opt]: 1.39998e-06 [ForceFp32Comm]: 1.04e-06 [remove_cast_before_assign_add]: 1.14e-06 [full_micro_interleaved_order_control]: 2.31e-06 [reorder_send_recv_between_fp_bp]: 2.89001e-06 [comm_op_add_attrs]: 1.10999e-06 [add_comm_op_reuse_tag]: 1.33002e-06 [interleave_split_concat_branches]: 1.13001e-06 [interleave_parallel_branches]: 1.39e-06 [overlap_opt_shard_in_pipeline]: 1.76e-06 [overlap_opt_shard_grad_in_pipeline]: 1.72999e-06 [control_data_broadcast_order]: 3.781e-05 [grouped_pairwise_exchange_alltoall]: 1.99e-06 [offloading_packed_experts]: 1.009e-05 [overlap_recompute_and_grad_model_parallel]: 9.46e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.17e-06 [overlap_recompute_allgather_and_fa_grad]: 1.73002e-06 [overlap_recompute_comm]: 2.49001e-06 [overlap_grad_ring_attention]: 9.15001e-06 [overlap_grad_flash_sp]: 5.02e-05 [begin_end_overlap_inline]: 6.39993e-07 [split_matmul_comm_elemetwise]: 2.66e-06 [split_layernorm_comm]: 1.89999e-06 [handle_group_info]: 1.02998e-06 [symbol_engine_optimizer]: 0.00020333, [1] [Cycle 1]: 0.00019567, [6] [build]: 2.217e-05 [elim_shapecalc]: 3.316e-05 [elim_not_effective]: 4.145e-05 [opt_reshape]: 2.186e-05 [fold_const_symbol]: 3.194e-05 [renormalize]: 4.19997e-07 [detach_backward]: 3.08998e-06 [pipeline_parallel_scheduler]: 1.64e-06 [auto_monad_reorder]: 6.001e-05 [get_jit_bprop_graph]: 2.13998e-06 [rewriter_after_jit_bprop_graph]: 8.53001e-06 [opt_after_jit_grad]: 0.00076588 [validate]: 9.988e-05 Sums bootstrap : 0.000525s : 0.09% type_inference : 0.579458s : 97.77% event_method : 0.000184s : 0.03% auto_monad : 0.000302s : 0.05% graph_reusing : 0.000008s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000041s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000009s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000058s : 0.01% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000050s : 0.01% optimize.rewriter_before_opt_a : 0.000137s : 0.02% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000091s : 0.02% optimize.opt_a.loop_unroll : 0.000070s : 0.01% optimize.opt_a.a_1 : 0.002346s : 0.40% optimize.opt_a.with_stream_mark : 0.000069s : 0.01% optimize.opt_a.recompute_prepare : 0.000060s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000095s : 0.02% optimize.opt_a.updatestate_assign_eliminate : 0.000023s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000021s : 0.00% optimize.opt_a.parameter_eliminate : 0.000006s : 0.00% optimize.opt_a.a_2 : 0.000565s : 0.10% optimize.opt_a.accelerated_algorithm : 0.000085s : 0.01% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000013s : 0.00% optimize.opt_a.shard_inline : 0.000042s : 0.01% optimize.opt_a.merge_send_recv : 0.000039s : 0.01% optimize.opt_a.auto_parallel : 0.000039s : 0.01% optimize.opt_a.parallel : 0.000036s : 0.01% optimize.opt_a.flash_sp : 0.000021s : 0.00% optimize.opt_a.merge_comm : 0.000026s : 0.00% optimize.opt_a.allreduce_fusion : 0.000024s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000042s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000052s : 0.01% optimize.opt_a.virtual_dataset : 0.000038s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000039s : 0.01% optimize.opt_a.virtual_output : 0.000039s : 0.01% optimize.opt_a.merge_forward : 0.000025s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000044s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000088s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000071s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000029s : 0.00% optimize.opt_a.meta_fg_expand : 0.000018s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000009s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000060s : 0.01% optimize.opt_a.a_after_grad : 0.000062s : 0.01% optimize.opt_a.renormalize : 0.002186s : 0.37% optimize.opt_a.add_forward_monad_depend : 0.000020s : 0.00% optimize.opt_a.auto_monad_grad : 0.000006s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000083s : 0.01% optimize.opt_a.cse : 0.000206s : 0.03% optimize.opt_a.a_3 : 0.000278s : 0.05% optimize.py_interpret_to_execute_after_opt_a : 0.000033s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000236s : 0.04% optimize.convert_after_rewriter : 0.000022s : 0.00% optimize.order_py_execute_after_rewriter : 0.000012s : 0.00% optimize.mutable_eliminate : 0.000849s : 0.14% optimize.opt_b.b_1 : 0.000447s : 0.08% optimize.opt_b.b_2 : 0.000022s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000019s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000010s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000010s : 0.00% optimize.opt_b.renormalize : 0.000002s : 0.00% optimize.opt_b.cse : 0.000155s : 0.03% optimize.optimize_parallel_all_gather_comm : 0.000053s : 0.01% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000048s : 0.01% optimize.loop_unroll : 0.000871s : 0.15% optimize.opt_after_cconv.c_1 : 0.000146s : 0.02% optimize.opt_after_cconv.parameter_eliminate : 0.000009s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000021s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000011s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000010s : 0.00% optimize.opt_after_cconv.cse : 0.000068s : 0.01% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000070s : 0.01% optimize.tuple_transform.d_1 : 0.000150s : 0.03% optimize.tuple_transform.none_parameter_eliminate : 0.000003s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000021s : 0.00% optimize.partial_unused_args_eliminate : 0.000003s : 0.00% optimize.add_recomputation : 0.000148s : 0.02% optimize.cse_after_recomputation.cse : 0.000048s : 0.01% optimize.environ_conv : 0.000023s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000018s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000008s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000038s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000010s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000009s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000009s : 0.00% optimize.overlap_grad_flash_sp : 0.000050s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000022s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000033s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000041s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000022s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000032s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000060s : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000009s : 0.00% opt_after_jit_grad : 0.000766s : 0.13% validate : 0.000100s : 0.02% Time group info: ------[substitution.] 0.000672 177 1.77% : 0.000012s : 2: substitution.depend_value_elim 0.89% : 0.000006s : 13: substitution.elim_not_effective 0.68% : 0.000005s : 13: substitution.fold_const_symbol 2.08% : 0.000014s : 17: substitution.graph_param_transform 72.19% : 0.000485s : 8: substitution.inline 1.97% : 0.000013s : 26: substitution.j_node_and_user_rematch 3.53% : 0.000024s : 2: substitution.less_batch_normalization 2.65% : 0.000018s : 26: substitution.remove_not_recompute_node 1.39% : 0.000009s : 8: substitution.replace_old_param 6.50% : 0.000044s : 28: substitution.updatestate_pure_node_eliminater 6.36% : 0.000043s : 34: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.579359 2 99.64% : 0.577245s : 1: type_inference.infer 0.36% : 0.002114s : 1: type_inference.specialize ------[replace.] 0.000113 8 100.00% : 0.000113s : 8: replace.inline ------[match.] 0.000479 8 100.00% : 0.000479s : 8: match.inline ------[predicate.] 0.000688 4588 0.97% : 0.000007s : 47: predicate.accumulaten_eliminater 0.91% : 0.000006s : 17: predicate.ad_related_special_op_eliminate 0.68% : 0.000005s : 34: predicate.addn_check_dump 0.94% : 0.000006s : 47: predicate.addn_zero_filter 0.92% : 0.000006s : 47: predicate.adjust_all_reduce_mul_add 2.16% : 0.000015s : 81: predicate.arithmetic_simplify 0.98% : 0.000007s : 47: predicate.cast_eliminate 0.67% : 0.000005s : 34: predicate.check_bprop_eliminate 0.67% : 0.000005s : 34: predicate.compare_switch_simplify 0.19% : 0.000001s : 17: predicate.const_output_eliminate 0.73% : 0.000005s : 34: predicate.depend_value_elim 0.96% : 0.000007s : 47: predicate.dict_get_item_const_eliminator 1.02% : 0.000007s : 47: predicate.dict_get_item_eliminator 1.00% : 0.000007s : 47: predicate.dict_set_item_eliminator 0.95% : 0.000007s : 34: predicate.dumpgradient_eliminate 0.30% : 0.000002s : 17: predicate.elim_not_effective 0.40% : 0.000003s : 17: predicate.elim_shapecalc_of_broadcastargs 1.33% : 0.000009s : 64: predicate.environ_add_const_eliminate 1.23% : 0.000008s : 64: predicate.environ_get_add_eliminate 1.19% : 0.000008s : 64: predicate.environ_get_depend_swap 1.93% : 0.000013s : 98: predicate.environ_get_eliminate 1.22% : 0.000008s : 64: predicate.environ_get_set_eliminate 1.07% : 0.000007s : 55: predicate.exchange_switch_depend_value 1.68% : 0.000012s : 55: predicate.float_depend_g_call 0.66% : 0.000005s : 34: predicate.float_environ_get_switch 1.02% : 0.000007s : 51: predicate.float_tuple_getitem_switch 0.18% : 0.000001s : 17: predicate.fold_const_symbol 0.81% : 0.000006s : 34: predicate.get_grad_eliminate 0.24% : 0.000002s : 17: predicate.graph_param_transform 0.74% : 0.000005s : 34: predicate.incorporate_call 0.61% : 0.000004s : 34: predicate.incorporate_call_switch 5.67% : 0.000039s : 204: predicate.inline 0.93% : 0.000006s : 34: predicate.inline_without_move 0.36% : 0.000002s : 34: predicate.j_node_and_user_rematch 1.28% : 0.000009s : 34: predicate.less_batch_normalization 1.62% : 0.000011s : 81: predicate.list_to_tuple_eliminator_ 2.45% : 0.000017s : 128: predicate.load_eliminater 0.87% : 0.000006s : 17: predicate.loop_unroll_after_grad 1.53% : 0.000011s : 77: predicate.loop_unroll_before_grad 1.75% : 0.000012s : 81: predicate.make_slice_get_slice_eliminator 0.71% : 0.000005s : 34: predicate.merge_addn 0.68% : 0.000005s : 34: predicate.micro_step_allgather_replace 0.69% : 0.000005s : 34: predicate.mini_step_allgather_replace 0.89% : 0.000006s : 47: predicate.minmaximum_grad 0.99% : 0.000007s : 17: predicate.mutable_eliminate 0.38% : 0.000003s : 17: predicate.opt_reshape 0.39% : 0.000003s : 17: predicate.parallel_virtual_node 1.61% : 0.000011s : 55: predicate.partial_defer_inline 1.43% : 0.000010s : 64: predicate.partial_eliminate 0.92% : 0.000006s : 47: predicate.print_const_string_wrapper 0.73% : 0.000005s : 34: predicate.reduce_all_const_elim 1.22% : 0.000008s : 47: predicate.reduce_eliminate 2.45% : 0.000017s : 128: predicate.redundant_stop_gradient_eliminater 0.45% : 0.000003s : 34: predicate.remove_not_recompute_node 1.17% : 0.000008s : 81: predicate.replace_applicator 0.54% : 0.000004s : 34: predicate.replace_old_param 0.26% : 0.000002s : 17: predicate.reset_defer_inline 0.96% : 0.000007s : 47: predicate.reshape_eliminate 0.68% : 0.000005s : 34: predicate.row_tensor_add_zeros_like 0.41% : 0.000003s : 17: predicate.row_tensor_eliminate 0.93% : 0.000006s : 34: predicate.same_eliminate 0.53% : 0.000004s : 40: predicate.set_cell_output_no_recompute 0.86% : 0.000006s : 34: predicate.shard_identity_eliminate 0.76% : 0.000005s : 34: predicate.special_op_eliminate 0.79% : 0.000005s : 34: predicate.specialize_transform 0.99% : 0.000007s : 34: predicate.split_environ_get_set_with_tuple_value 0.87% : 0.000006s : 34: predicate.stack_unstack_eliminate 0.39% : 0.000003s : 17: predicate.switch_call_monad_eliminater 1.16% : 0.000008s : 55: predicate.switch_defer_inline 1.83% : 0.000013s : 89: predicate.switch_layer_defer_inline 4.14% : 0.000029s : 183: predicate.switch_simplify 0.94% : 0.000006s : 47: predicate.tile_eliminate 0.94% : 0.000006s : 47: predicate.transpose_eliminate 1.72% : 0.000012s : 81: predicate.tuple_list_convert_item_index_to_positive 1.76% : 0.000012s : 81: predicate.tuple_list_get_item_const_eliminator 1.61% : 0.000011s : 81: predicate.tuple_list_get_item_depend_reorder 2.82% : 0.000019s : 115: predicate.tuple_list_get_item_eliminator 1.66% : 0.000011s : 81: predicate.tuple_list_get_set_item_eliminator 2.46% : 0.000017s : 115: predicate.tuple_list_set_item_eliminator 1.59% : 0.000011s : 81: predicate.tuple_to_list_eliminator_ 2.54% : 0.000018s : 128: predicate.updatestate_pure_node_eliminater 3.65% : 0.000025s : 162: predicate.updatestate_useless_node_eliminater 0.41% : 0.000003s : 17: predicate.value_based_eliminate 0.78% : 0.000005s : 34: predicate.virtual_dataset_eliminate 0.75% : 0.000005s : 34: predicate.virtual_output_eliminate 0.35% : 0.000002s : 17: predicate.virtual_view_grad_eliminate 0.38% : 0.000003s : 17: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.001194 14 31.22% : 0.000373s : 4: func_graph_cloner_run.FuncGraphClonerGraph 8.01% : 0.000096s : 2: func_graph_cloner_run.FuncGraphClonerNode 60.78% : 0.000726s : 8: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.949874 192 0.00% : 0.000004s : 1: ForceFp32Comm 0.49% : 0.004636s : 1: add_attr 0.49% : 0.004619s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.02% : 0.000154s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.03% : 0.000319s : 1: auto_monad 0.01% : 0.000068s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.06% : 0.000562s : 1: bootstrap 0.01% : 0.000054s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000043s : 1: control_data_broadcast_order 0.00% : 0.000028s : 1: convert_after_rewriter 0.01% : 0.000065s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000027s : 1: environ_conv 0.02% : 0.000200s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000013s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000008s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000011s : 1: label_micro_interleaved_index 0.09% : 0.000890s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.09% : 0.000867s : 1: mutable_eliminate 0.00% : 0.000013s : 1: offloading_packed_experts 0.00% : 0.000045s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000039s : 1: opt.transform.mutable_eliminate 0.41% : 0.003850s : 78: opt.transform.opt_a 0.02% : 0.000144s : 1: opt.transform.opt_after_cconv 0.01% : 0.000076s : 1: opt.transform.opt_after_jit_grad 0.05% : 0.000432s : 28: opt.transform.opt_b 0.02% : 0.000166s : 2: opt.transform.opt_trans_graph 0.01% : 0.000122s : 4: opt.transform.symbol_engine_opt 0.82% : 0.007819s : 1: opt_a 0.03% : 0.000323s : 1: opt_after_cconv 0.08% : 0.000785s : 1: opt_after_jit_grad 17.30% : 0.164375s : 1: opt_b 18.53% : 0.176041s : 1: optimize 0.01% : 0.000059s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000016s : 1: order_py_execute_after_rewriter 0.01% : 0.000056s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000013s : 1: overlap_grad_ring_attention 0.00% : 0.000004s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000013s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000007s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000064s : 1: pre_auto_parallel 0.01% : 0.000057s : 1: py_interpret_to_execute 0.00% : 0.000038s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.01% : 0.000078s : 1: remove_dup_value 0.13% : 0.001262s : 1: renormalize.infer 0.10% : 0.000908s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000013s : 1: rewriter_after_jit_bprop_graph 0.03% : 0.000248s : 1: rewriter_after_opt_a 0.02% : 0.000146s : 1: rewriter_before_opt_a 0.00% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000021s : 1: swap_dp_allreduce_reducescatter 0.02% : 0.000207s : 1: symbol_engine_optimizer 0.02% : 0.000210s : 1: tuple_transform 61.01% : 0.579492s : 1: type_inference [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:17.591.220 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceSubExt-op0, the new abstract of Expand/_InplaceSubExt/Sub-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab29f86c101, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:17.591.302 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab29f86c101, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:17.591.669 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceRemainderTensorTensor-op0, the new abstract of Expand/_InplaceRemainderTensorTensor/FloorMod-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab441298702, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:17.591.712 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab441298702, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:17.592.025 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceDiv-op0, the new abstract of Expand/_InplaceDiv/RealDiv-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab29ee76103, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:17.592.065 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab29ee76103, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) . TotalTime = 1.21786, [30] [bootstrap]: 0.00081699 [type_inference]: 0.390467 [event_method]: 0.00021547 [auto_monad]: 0.00064863 [graph_reusing]: 1.088e-05 [pre_auto_parallel]: 3.18998e-06 [py_interpret_to_execute]: 7.08e-05 [rewriter_before_opt_a]: 0.00023737 [expand_dump_flag]: 4.19002e-06 [jit_opt_a]: 0.819695, [3] [Cycle 1]: 0.673229, [27] [switch_simplify]: 0.00016065 [loop_unroll]: 0.00010207 [a_1]: 0.00354638 [with_stream_mark]: 6.142e-05 [recompute_prepare]: 6.795e-05 [updatestate_depend_eliminate]: 0.00010711 [updatestate_assign_eliminate]: 2.23e-05 [updatestate_loads_eliminate]: 1.897e-05 [parameter_eliminate]: 5.96e-06 [specialize_transform]: 4.544e-05 [updatestate_useless_node_eliminater]: 4.99e-05 [accelerated_algorithm]: 7.014e-05 [meta_shard_fg_expand]: 1.11e-05 [get_grad_eliminate_]: 3.572e-05 [merge_forward]: 2.162e-05 [cell_reuse_recompute_pass]: 1.88002e-06 [cell_reuse_handle_not_recompute_node_pass]: 6.986e-05 [j_node_and_user_rematch]: 6.246e-05 [meta_fg_expand]: 0.314039 [replace_old_param]: 0.00038106 [inline_without_move]: 0.00029306 [renormalize]: 0.351592 [add_forward_monad_depend]: 0.00010639 [auto_monad_grad]: 3.371e-05 [auto_monad_eliminator]: 0.00039453 [cse]: 0.00086956 [replace_applicator]: 0.00061541 [Cycle 2]: 0.139323, [27] [switch_simplify]: 0.00027302 [loop_unroll]: 0.00026074 [a_1]: 0.131245 [with_stream_mark]: 0.00015716 [recompute_prepare]: 6.122e-05 [updatestate_depend_eliminate]: 8.135e-05 [updatestate_assign_eliminate]: 2.55e-05 [updatestate_loads_eliminate]: 3.471e-05 [parameter_eliminate]: 3.21001e-06 [specialize_transform]: 4.51e-05 [updatestate_useless_node_eliminater]: 5.524e-05 [accelerated_algorithm]: 5.089e-05 [meta_shard_fg_expand]: 2.077e-05 [get_grad_eliminate_]: 3.972e-05 [merge_forward]: 2.394e-05 [cell_reuse_recompute_pass]: 2.44001e-06 [cell_reuse_handle_not_recompute_node_pass]: 8.759e-05 [j_node_and_user_rematch]: 7.909e-05 [meta_fg_expand]: 0.00028029 [replace_old_param]: 5.85e-05 [inline_without_move]: 4.176e-05 [renormalize]: 0.00575931 [add_forward_monad_depend]: 8.98002e-06 [auto_monad_grad]: 3.14999e-06 [auto_monad_eliminator]: 9.503e-05 [cse]: 0.0001863 [replace_applicator]: 6.327e-05 [Cycle 3]: 0.00253255, [27] [switch_simplify]: 4.161e-05 [loop_unroll]: 3.912e-05 [a_1]: 0.0013735 [with_stream_mark]: 4.189e-05 [recompute_prepare]: 4.557e-05 [updatestate_depend_eliminate]: 2.786e-05 [updatestate_assign_eliminate]: 2.298e-05 [updatestate_loads_eliminate]: 2.201e-05 [parameter_eliminate]: 2.61e-06 [specialize_transform]: 4.306e-05 [updatestate_useless_node_eliminater]: 4.987e-05 [accelerated_algorithm]: 4.887e-05 [meta_shard_fg_expand]: 9.60001e-06 [get_grad_eliminate_]: 3.949e-05 [merge_forward]: 2.336e-05 [cell_reuse_recompute_pass]: 3.15998e-06 [cell_reuse_handle_not_recompute_node_pass]: 9.344e-05 [j_node_and_user_rematch]: 8e-05 [meta_fg_expand]: 1.639e-05 [replace_old_param]: 4.81e-05 [inline_without_move]: 4.001e-05 [renormalize]: 8.00064e-08 [add_forward_monad_depend]: 4.33999e-06 [auto_monad_grad]: 2.03002e-06 [auto_monad_eliminator]: 6.141e-05 [cse]: 0.0001344 [replace_applicator]: 4.399e-05 [py_interpret_to_execute_after_opt_a]: 3.704e-05 [rewriter_after_opt_a]: 0.00020571 [convert_after_rewriter]: 3.359e-05 [order_py_execute_after_rewriter]: 2.633e-05 [mutable_eliminate]: 0.00093313 [jit_opt_b]: 0.00061036, [2] [Cycle 1]: 0.00036847, [2] [frontend_op_eliminate]: 0.00023639 [inline_after_opt_a]: 0.00011125 [Cycle 2]: 0.0002276, [2] [frontend_op_eliminate]: 0.00011844 [inline_after_opt_a]: 9.806e-05 [cconv]: 4.611e-05 [loop_unroll]: 0.00060792 [jit_opt_after_cconv]: 0.00072173, [1] [Cycle 1]: 0.00071267, [11] [c_1]: 0.00023448 [parameter_eliminate]: 6.65002e-06 [updatestate_depend_eliminate]: 4.136e-05 [updatestate_assign_eliminate]: 2.403e-05 [updatestate_loads_eliminate]: 2.614e-05 [cse]: 0.00012938 [call_graph_tuple_transform]: 0.00010524 [tuple_list_get_item_eliminator]: 3.899e-05 [none_parameter_eliminate]: 1.91e-06 [renormalize]: 1.04e-06 [switch_simplify]: 4.032e-05 [remove_dup_value]: 0.00015224 [partial_unused_args_eliminate]: 3.19001e-06 [environ_conv]: 2.842e-05 [add_recomputation]: 0.00027005 [cse_after_recomputation]: 0.00019413, [1] [Cycle 1]: 0.00018327, [1] [cse]: 0.0001641 [auto_monad_reorder]: 0.00011547 [get_jit_bprop_graph]: 2.30002e-06 [rewriter_after_jit_bprop_graph]: 9.52001e-06 [opt_after_jit_grad]: 0.00081886 [symbol_engine_optimizer]: 0.00032374, [1] [Cycle 1]: 0.00031484, [6] [build]: 3.044e-05 [elim_shapecalc]: 5.167e-05 [elim_not_effective]: 8.18e-05 [opt_reshape]: 4.622e-05 [fold_const_symbol]: 7.094e-05 [renormalize]: 8.29983e-07 [validate]: 0.00014463 Sums bootstrap : 0.000817s : 0.07% type_inference : 0.390467s : 32.22% event_method : 0.000215s : 0.02% auto_monad : 0.000649s : 0.05% graph_reusing : 0.000011s : 0.00% pre_auto_parallel : 0.000003s : 0.00% py_interpret_to_execute : 0.000071s : 0.01% rewriter_before_opt_a : 0.000237s : 0.02% expand_dump_flag : 0.000004s : 0.00% jit_opt_a.switch_simplify : 0.000475s : 0.04% jit_opt_a.loop_unroll : 0.000402s : 0.03% jit_opt_a.a_1 : 0.136165s : 11.24% jit_opt_a.with_stream_mark : 0.000260s : 0.02% jit_opt_a.recompute_prepare : 0.000175s : 0.01% jit_opt_a.updatestate_depend_eliminate : 0.000216s : 0.02% jit_opt_a.updatestate_assign_eliminate : 0.000071s : 0.01% jit_opt_a.updatestate_loads_eliminate : 0.000076s : 0.01% jit_opt_a.parameter_eliminate : 0.000012s : 0.00% jit_opt_a.specialize_transform : 0.000134s : 0.01% jit_opt_a.updatestate_useless_node_eliminater : 0.000155s : 0.01% jit_opt_a.accelerated_algorithm : 0.000170s : 0.01% jit_opt_a.meta_shard_fg_expand : 0.000041s : 0.00% jit_opt_a.get_grad_eliminate_ : 0.000115s : 0.01% jit_opt_a.merge_forward : 0.000069s : 0.01% jit_opt_a.cell_reuse_recompute_pass : 0.000007s : 0.00% jit_opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000251s : 0.02% jit_opt_a.j_node_and_user_rematch : 0.000222s : 0.02% jit_opt_a.meta_fg_expand : 0.314336s : 25.94% jit_opt_a.replace_old_param : 0.000488s : 0.04% jit_opt_a.inline_without_move : 0.000375s : 0.03% jit_opt_a.renormalize : 0.357351s : 29.49% jit_opt_a.add_forward_monad_depend : 0.000120s : 0.01% jit_opt_a.auto_monad_grad : 0.000039s : 0.00% jit_opt_a.auto_monad_eliminator : 0.000551s : 0.05% jit_opt_a.cse : 0.001190s : 0.10% jit_opt_a.replace_applicator : 0.000723s : 0.06% py_interpret_to_execute_after_opt_a : 0.000037s : 0.00% rewriter_after_opt_a : 0.000206s : 0.02% convert_after_rewriter : 0.000034s : 0.00% order_py_execute_after_rewriter : 0.000026s : 0.00% mutable_eliminate : 0.000933s : 0.08% jit_opt_b.frontend_op_eliminate : 0.000355s : 0.03% jit_opt_b.inline_after_opt_a : 0.000209s : 0.02% cconv : 0.000046s : 0.00% loop_unroll : 0.000608s : 0.05% jit_opt_after_cconv.c_1 : 0.000234s : 0.02% jit_opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% jit_opt_after_cconv.updatestate_depend_eliminate : 0.000041s : 0.00% jit_opt_after_cconv.updatestate_assign_eliminate : 0.000024s : 0.00% jit_opt_after_cconv.updatestate_loads_eliminate : 0.000026s : 0.00% jit_opt_after_cconv.cse : 0.000129s : 0.01% jit_opt_after_cconv.call_graph_tuple_transform : 0.000105s : 0.01% jit_opt_after_cconv.tuple_list_get_item_eliminator : 0.000039s : 0.00% jit_opt_after_cconv.none_parameter_eliminate : 0.000002s : 0.00% jit_opt_after_cconv.renormalize : 0.000001s : 0.00% jit_opt_after_cconv.switch_simplify : 0.000040s : 0.00% remove_dup_value : 0.000152s : 0.01% partial_unused_args_eliminate : 0.000003s : 0.00% environ_conv : 0.000028s : 0.00% add_recomputation : 0.000270s : 0.02% cse_after_recomputation.cse : 0.000164s : 0.01% auto_monad_reorder : 0.000115s : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000010s : 0.00% opt_after_jit_grad : 0.000819s : 0.07% symbol_engine_optimizer.build : 0.000030s : 0.00% symbol_engine_optimizer.elim_shapecalc : 0.000052s : 0.00% symbol_engine_optimizer.elim_not_effective : 0.000082s : 0.01% symbol_engine_optimizer.opt_reshape : 0.000046s : 0.00% symbol_engine_optimizer.fold_const_symbol : 0.000071s : 0.01% symbol_engine_optimizer.renormalize : 0.000001s : 0.00% validate : 0.000145s : 0.01% Time group info: ------[substitution.] 0.005827 842 0.63% : 0.000036s : 7: substitution.addn_check_dump 1.53% : 0.000089s : 7: substitution.addn_zero_filter 4.30% : 0.000251s : 20: substitution.arithmetic_simplify 0.25% : 0.000015s : 1: substitution.cast_eliminate 0.73% : 0.000043s : 12: substitution.depend_value_elim 0.22% : 0.000013s : 37: substitution.elim_not_effective 0.20% : 0.000012s : 37: substitution.fold_const_symbol 18.21% : 0.001061s : 4: substitution.getattr_setattr_resolve 0.50% : 0.000029s : 41: substitution.graph_param_transform 53.76% : 0.003133s : 54: substitution.inline 1.23% : 0.000072s : 12: substitution.inline_without_move 0.72% : 0.000042s : 103: substitution.j_node_and_user_rematch 0.63% : 0.000037s : 8: substitution.less_batch_normalization 0.13% : 0.000008s : 6: substitution.load_eliminater 0.47% : 0.000028s : 7: substitution.merge_addn 0.79% : 0.000046s : 33: substitution.minmaximum_grad 0.30% : 0.000017s : 12: substitution.partial_eliminate 1.09% : 0.000063s : 103: substitution.remove_not_recompute_node 3.09% : 0.000180s : 53: substitution.replace_applicator 0.56% : 0.000032s : 48: substitution.replace_old_param 0.09% : 0.000005s : 1: substitution.set_cell_output_no_recompute 1.72% : 0.000100s : 33: substitution.tuple_list_convert_item_index_to_positive 1.13% : 0.000066s : 33: substitution.tuple_list_get_item_depend_reorder 4.37% : 0.000254s : 94: substitution.tuple_list_get_item_eliminator 0.96% : 0.000056s : 30: substitution.updatestate_pure_node_eliminater 1.40% : 0.000082s : 45: substitution.updatestate_useless_node_eliminater 1.01% : 0.000059s : 1: substitution.zero_like_fill_zero ------[type_inference.] 0.390322 2 99.13% : 0.386910s : 1: type_inference.infer 0.87% : 0.003412s : 1: type_inference.specialize ------[replace.] 0.001733 128 0.89% : 0.000015s : 1: replace.arithmetic_simplify 0.93% : 0.000016s : 1: replace.cast_eliminate 4.25% : 0.000074s : 3: replace.getattr_setattr_resolve 48.24% : 0.000836s : 54: replace.inline 6.27% : 0.000109s : 6: replace.replace_applicator 36.74% : 0.000637s : 61: replace.tuple_list_get_item_eliminator 0.75% : 0.000013s : 1: replace.updatestate_useless_node_eliminater 1.93% : 0.000034s : 1: replace.zero_like_fill_zero ------[match.] 0.004476 128 3.14% : 0.000141s : 1: match.arithmetic_simplify 0.29% : 0.000013s : 1: match.cast_eliminate 21.36% : 0.000956s : 3: match.getattr_setattr_resolve 69.17% : 0.003096s : 54: match.inline 0.95% : 0.000042s : 6: match.replace_applicator 3.43% : 0.000154s : 61: match.tuple_list_get_item_eliminator 0.37% : 0.000016s : 1: match.updatestate_useless_node_eliminater 1.29% : 0.000058s : 1: match.zero_like_fill_zero ------[predicate.] 0.123054 16800 0.03% : 0.000037s : 260: predicate.accumulaten_eliminater 0.01% : 0.000010s : 41: predicate.ad_related_special_op_eliminate 0.03% : 0.000033s : 260: predicate.addn_check_dump 0.04% : 0.000043s : 260: predicate.addn_zero_filter 0.04% : 0.000046s : 261: predicate.arithmetic_simplify 0.03% : 0.000035s : 262: predicate.cast_eliminate 0.01% : 0.000012s : 83: predicate.check_bprop_eliminate 0.03% : 0.000033s : 260: predicate.compare_switch_simplify 0.03% : 0.000037s : 260: predicate.depend_value_elim 0.03% : 0.000033s : 262: predicate.dict_get_item_const_eliminator 0.03% : 0.000035s : 262: predicate.dict_get_item_eliminator 0.03% : 0.000034s : 262: predicate.dict_set_item_eliminator 0.01% : 0.000007s : 41: predicate.dumpgradient_eliminate 0.00% : 0.000003s : 41: predicate.elim_not_effective 0.01% : 0.000006s : 41: predicate.elim_shapecalc_of_broadcastargs 0.03% : 0.000034s : 262: predicate.environ_add_const_eliminate 0.03% : 0.000033s : 262: predicate.environ_get_add_eliminate 0.03% : 0.000034s : 262: predicate.environ_get_depend_swap 0.03% : 0.000035s : 262: predicate.environ_get_eliminate 0.05% : 0.000060s : 262: predicate.environ_get_set_eliminate 0.00% : 0.000003s : 41: predicate.fold_const_symbol 0.01% : 0.000017s : 125: predicate.get_grad_eliminate 0.01% : 0.000007s : 20: predicate.getattr_setattr_resolve 0.00% : 0.000003s : 41: predicate.graph_param_transform 0.10% : 0.000122s : 501: predicate.inline 0.04% : 0.000050s : 310: predicate.inline_without_move 0.01% : 0.000009s : 125: predicate.j_node_and_user_rematch 0.02% : 0.000021s : 125: predicate.less_batch_normalization 0.04% : 0.000047s : 323: predicate.list_to_tuple_eliminator_ 0.04% : 0.000050s : 364: predicate.load_eliminater 0.01% : 0.000010s : 41: predicate.loop_unroll_after_grad 0.05% : 0.000063s : 477: predicate.loop_unroll_before_grad 0.03% : 0.000040s : 303: predicate.make_slice_get_slice_eliminator 0.03% : 0.000034s : 260: predicate.merge_addn 0.03% : 0.000034s : 261: predicate.minmaximum_grad 0.01% : 0.000010s : 43: predicate.mutable_eliminate 0.00% : 0.000006s : 41: predicate.opt_reshape 0.05% : 0.000061s : 364: predicate.partial_eliminate 0.03% : 0.000034s : 260: predicate.print_const_string_wrapper 0.03% : 0.000042s : 261: predicate.reduce_eliminate 0.03% : 0.000043s : 323: predicate.redundant_stop_gradient_eliminater 0.01% : 0.000010s : 125: predicate.remove_not_recompute_node 98.06% : 0.120672s : 731: predicate.replace_applicator 0.02% : 0.000024s : 310: predicate.replace_old_param 0.00% : 0.000006s : 82: predicate.reset_defer_inline 0.03% : 0.000034s : 261: predicate.reshape_eliminate 0.03% : 0.000036s : 260: predicate.row_tensor_add_zeros_like 0.01% : 0.000013s : 83: predicate.row_tensor_eliminate 0.03% : 0.000035s : 260: predicate.same_eliminate 0.01% : 0.000012s : 135: predicate.set_cell_output_no_recompute 0.01% : 0.000018s : 124: predicate.special_op_eliminate 0.02% : 0.000021s : 125: predicate.specialize_transform 0.03% : 0.000039s : 260: predicate.split_environ_get_set_with_tuple_value 0.03% : 0.000034s : 260: predicate.stack_unstack_eliminate 0.01% : 0.000006s : 41: predicate.switch_call_monad_eliminater 0.07% : 0.000088s : 378: predicate.switch_defer_inline 0.05% : 0.000062s : 378: predicate.switch_layer_defer_inline 0.11% : 0.000129s : 896: predicate.switch_simplify 0.03% : 0.000034s : 261: predicate.tile_eliminate 0.03% : 0.000034s : 261: predicate.transpose_eliminate 0.04% : 0.000043s : 262: predicate.tuple_list_convert_item_index_to_positive 0.03% : 0.000040s : 262: predicate.tuple_list_get_item_depend_reorder 0.07% : 0.000087s : 446: predicate.tuple_list_get_item_eliminator 0.03% : 0.000040s : 262: predicate.tuple_list_set_item_eliminator 0.03% : 0.000042s : 323: predicate.tuple_to_list_eliminator_ 0.04% : 0.000050s : 364: predicate.updatestate_pure_node_eliminater 0.06% : 0.000074s : 490: predicate.updatestate_useless_node_eliminater 0.03% : 0.000042s : 260: predicate.value_based_eliminate 0.00% : 0.000006s : 41: predicate.virtual_view_grad_eliminate 0.01% : 0.000014s : 84: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.009576 122 58.93% : 0.005643s : 56: func_graph_cloner_run.FuncGraphClonerGraph 6.80% : 0.000651s : 9: func_graph_cloner_run.FuncGraphClonerNode 34.27% : 0.003282s : 57: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.717273 93 0.02% : 0.000277s : 1: add_recomputation 0.04% : 0.000660s : 1: auto_monad 0.01% : 0.000121s : 1: auto_monad_reorder 0.05% : 0.000845s : 1: bootstrap 0.00% : 0.000049s : 1: cconv 0.00% : 0.000038s : 1: convert_after_rewriter 0.01% : 0.000197s : 1: cse_after_recomputation 0.00% : 0.000031s : 1: environ_conv 0.01% : 0.000224s : 1: event_method 0.00% : 0.000007s : 1: expand_dump_flag 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000014s : 1: graph_reusing 47.73% : 0.819699s : 1: jit_opt_a 0.04% : 0.000726s : 1: jit_opt_after_cconv 0.04% : 0.000614s : 1: jit_opt_b 0.04% : 0.000623s : 1: loop_unroll 0.06% : 0.000948s : 1: mutable_eliminate 8.14% : 0.139758s : 39: opt.transform.jit_opt_a 0.02% : 0.000413s : 4: opt.transform.jit_opt_after_cconv 0.03% : 0.000549s : 8: opt.transform.jit_opt_b 0.00% : 0.000060s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000064s : 1: opt.transform.mutable_eliminate 0.01% : 0.000148s : 1: opt.transform.opt_after_jit_grad 0.07% : 0.001205s : 2: opt.transform.opt_resolve 0.01% : 0.000246s : 4: opt.transform.symbol_engine_opt 0.05% : 0.000834s : 1: opt_after_jit_grad 0.00% : 0.000029s : 1: order_py_execute_after_rewriter 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pre_auto_parallel 0.00% : 0.000074s : 1: py_interpret_to_execute 0.00% : 0.000040s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000160s : 1: remove_dup_value 10.56% : 0.181265s : 2: renormalize.infer 10.25% : 0.176052s : 2: renormalize.specialize 0.00% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000211s : 1: rewriter_after_opt_a 0.01% : 0.000241s : 1: rewriter_before_opt_a 0.02% : 0.000327s : 1: symbol_engine_optimizer 22.74% : 0.390495s : 1: type_inference [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:29.198.344 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceSubExt-op0, the new abstract of Expand/_InplaceSubExt/Sub-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab44ae5f406, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:29.198.419 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab44ae5f406, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:29.198.782 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceRemainderTensorTensor-op0, the new abstract of Expand/_InplaceRemainderTensorTensor/FloorMod-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab44ae3e907, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:29.198.819 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab44ae3e907, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:29.199.116 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceDiv-op0, the new abstract of Expand/_InplaceDiv/RealDiv-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab44a114008, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:29.199.179 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab44a114008, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) . TotalTime = 0.175578, [21] [bootstrap]: 0.00059024 [type_inference]: 0.156522 [event_method]: 0.0001723 [auto_monad]: 0.00026966 [graph_reusing]: 8.15e-06 [inline]: 3.13e-06 [add_attr]: 0.00453275, [1] [add_attr_with_inline]: 0.00451964, [1] [Cycle 1]: 9.804e-05, [2] [tag_attr]: 3.874e-05 [meta_addattr_fg_expand]: 9.19998e-06 [parallel-infer-symbol]: 3.78999e-06 [pre_auto_parallel]: 5.671e-05 [insert-virtual-dataset]: 2.64001e-06 [parallel-infer-symbol-second]: 9.00007e-07 [dataset_repeat_opt]: 2.53003e-06 [pipeline_split]: 1.74998e-06 [optimize]: 0.01212, [53] [py_interpret_to_execute]: 4.495e-05 [rewriter_before_opt_a]: 0.00013552 [opt_a]: 0.00772804, [2] [Cycle 1]: 0.00565062, [45] [expand_dump_flag]: 4.11001e-06 [switch_simplify]: 6.589e-05 [loop_unroll]: 4.915e-05 [a_1]: 0.00166049 [with_stream_mark]: 3.862e-05 [recompute_prepare]: 3.421e-05 [updatestate_depend_eliminate]: 8.241e-05 [updatestate_assign_eliminate]: 1.213e-05 [updatestate_loads_eliminate]: 9.12001e-06 [parameter_eliminate]: 3.91999e-06 [a_2]: 0.00027962 [accelerated_algorithm]: 4.888e-05 [shard]: 2.74999e-06 [meta_shard_fg_expand]: 5.62999e-06 [shard_inline]: 1.811e-05 [merge_send_recv]: 1.826e-05 [auto_parallel]: 1.92e-05 [parallel]: 2.291e-05 [flash_sp]: 1.526e-05 [merge_comm]: 1.142e-05 [allreduce_fusion]: 1.024e-05 [matmul_add_comm_reduction]: 7.221e-05 [allreduce_slice_to_reducescatter]: 9.39996e-07 [virtual_shard_identity]: 2.943e-05 [virtual_dataset]: 1.948e-05 [get_grad_eliminate_]: 1.829e-05 [virtual_output]: 2.021e-05 [merge_forward]: 1.392e-05 [cell_reuse_recompute_pass]: 2.22001e-06 [offload_activation]: 2.274e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.187e-05 [merge_recompute_call_nodes]: 1.86e-06 [before_grad]: 3.458e-05 [set_forward_comm_id_for_comm_node_pass]: 1.226e-05 [meta_fg_expand]: 9.27999e-06 [flash_sp_send_recv_attached]: 6.83998e-06 [receive_attached]: 2.66e-06 [after_resolve]: 3.016e-05 [a_after_grad]: 3.269e-05 [renormalize]: 0.0021296 [add_forward_monad_depend]: 1.155e-05 [auto_monad_grad]: 2.83e-06 [auto_monad_eliminator]: 4.497e-05 [cse]: 0.00014354 [a_3]: 0.00015776 [Cycle 2]: 0.00206101, [45] [expand_dump_flag]: 3.35e-06 [switch_simplify]: 2.284e-05 [loop_unroll]: 1.972e-05 [a_1]: 0.00060539 [with_stream_mark]: 3.165e-05 [recompute_prepare]: 2.645e-05 [updatestate_depend_eliminate]: 1.28e-05 [updatestate_assign_eliminate]: 1.087e-05 [updatestate_loads_eliminate]: 1.02e-05 [parameter_eliminate]: 2.86999e-06 [a_2]: 0.0002834 [accelerated_algorithm]: 2.985e-05 [shard]: 2.49999e-06 [meta_shard_fg_expand]: 6.29999e-06 [shard_inline]: 1.975e-05 [merge_send_recv]: 1.855e-05 [auto_parallel]: 1.844e-05 [parallel]: 1.113e-05 [flash_sp]: 4.79e-06 [merge_comm]: 1.14e-05 [allreduce_fusion]: 1.157e-05 [matmul_add_comm_reduction]: 2.016e-05 [allreduce_slice_to_reducescatter]: 1.05999e-06 [virtual_shard_identity]: 2.62e-05 [virtual_dataset]: 2.043e-05 [get_grad_eliminate_]: 1.997e-05 [virtual_output]: 2.034e-05 [merge_forward]: 1.149e-05 [cell_reuse_recompute_pass]: 2.82002e-06 [offload_activation]: 2.004e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.031e-05 [merge_recompute_call_nodes]: 1.78002e-06 [before_grad]: 3.316e-05 [set_forward_comm_id_for_comm_node_pass]: 1.189e-05 [meta_fg_expand]: 9.23002e-06 [flash_sp_send_recv_attached]: 2.05002e-06 [receive_attached]: 2.46998e-06 [after_resolve]: 2.955e-05 [a_after_grad]: 3.498e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 4.53999e-06 [auto_monad_grad]: 2.35002e-06 [auto_monad_eliminator]: 3.524e-05 [cse]: 6.71e-05 [a_3]: 0.00020446 [py_interpret_to_execute_after_opt_a]: 3.312e-05 [slice_cell_reuse_recomputed_activation]: 2.81e-06 [rewriter_after_opt_a]: 0.00022709 [convert_after_rewriter]: 2.01e-05 [order_py_execute_after_rewriter]: 1.36e-05 [mutable_eliminate]: 0.00091968 [opt_b]: 0.00068492, [1] [Cycle 1]: 0.00067375, [7] [b_1]: 0.00047314 [b_2]: 2.303e-05 [updatestate_depend_eliminate]: 1.914e-05 [updatestate_assign_eliminate]: 1.047e-05 [updatestate_loads_eliminate]: 1.081e-05 [renormalize]: 9.09989e-07 [cse]: 9.037e-05 [optimize_parallel_all_gather_comm]: 4.306e-05 [overlap_param_gather]: 2.54001e-06 [cconv]: 4.377e-05 [loop_unroll]: 0.00068433 [opt_after_cconv]: 0.00031152, [1] [Cycle 1]: 0.00030162, [7] [c_1]: 0.00014264 [parameter_eliminate]: 6.01998e-06 [updatestate_depend_eliminate]: 1.765e-05 [updatestate_assign_eliminate]: 9.67001e-06 [updatestate_loads_eliminate]: 1.016e-05 [cse]: 6.942e-05 [renormalize]: 1.16002e-06 [remove_dup_value]: 7.781e-05 [tuple_transform]: 0.00021071, [1] [Cycle 1]: 0.0002046, [4] [d_1]: 0.0001547 [none_parameter_eliminate]: 2.81999e-06 [renormalize]: 5.39992e-07 [switch_simplify]: 2.263e-05 [partial_unused_args_eliminate]: 2.54001e-06 [add_recomputation]: 0.0001519 [cse_after_recomputation]: 6.635e-05, [1] [Cycle 1]: 6.056e-05, [1] [cse]: 4.936e-05 [environ_conv]: 2.208e-05 [swap_dp_allreduce_reducescatter]: 1.725e-05 [bias_add_comm_swap]: 3.65e-06 [label_micro_interleaved_index]: 8.80999e-06 [label_fine_grained_interleaved_index]: 3.08e-06 [merge_cast_opt]: 1.61998e-06 [slice_recompute_activation]: 2.37999e-06 [micro_interleaved_order_control]: 3.28998e-06 [assign_add_opt]: 1.54998e-06 [ForceFp32Comm]: 1.06997e-06 [remove_cast_before_assign_add]: 1.66e-06 [full_micro_interleaved_order_control]: 2.74001e-06 [reorder_send_recv_between_fp_bp]: 3.11001e-06 [comm_op_add_attrs]: 1.39e-06 [add_comm_op_reuse_tag]: 1.44998e-06 [interleave_split_concat_branches]: 1.34998e-06 [interleave_parallel_branches]: 1.29e-06 [overlap_opt_shard_in_pipeline]: 1.34998e-06 [overlap_opt_shard_grad_in_pipeline]: 2.00002e-06 [control_data_broadcast_order]: 3.69e-05 [grouped_pairwise_exchange_alltoall]: 2.11e-06 [offloading_packed_experts]: 1.01e-05 [overlap_recompute_and_grad_model_parallel]: 9.44e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.47001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.60999e-06 [overlap_recompute_comm]: 2.37999e-06 [overlap_grad_ring_attention]: 8.48001e-06 [overlap_grad_flash_sp]: 5.263e-05 [begin_end_overlap_inline]: 5.8001e-07 [split_matmul_comm_elemetwise]: 2.41998e-06 [split_layernorm_comm]: 1.86e-06 [handle_group_info]: 1.17999e-06 [symbol_engine_optimizer]: 0.00020212, [1] [Cycle 1]: 0.00019443, [6] [build]: 2.171e-05 [elim_shapecalc]: 3.269e-05 [elim_not_effective]: 4.287e-05 [opt_reshape]: 2.188e-05 [fold_const_symbol]: 3.414e-05 [renormalize]: 1.80007e-07 [detach_backward]: 2.75002e-06 [pipeline_parallel_scheduler]: 2.21998e-06 [auto_monad_reorder]: 5.799e-05 [get_jit_bprop_graph]: 2.17999e-06 [rewriter_after_jit_bprop_graph]: 7.98001e-06 [opt_after_jit_grad]: 0.00089891 [validate]: 9.745e-05 Sums bootstrap : 0.000590s : 0.35% type_inference : 0.156522s : 92.24% event_method : 0.000172s : 0.10% auto_monad : 0.000270s : 0.16% graph_reusing : 0.000008s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000039s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000009s : 0.01% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000057s : 0.03% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000045s : 0.03% optimize.rewriter_before_opt_a : 0.000136s : 0.08% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000089s : 0.05% optimize.opt_a.loop_unroll : 0.000069s : 0.04% optimize.opt_a.a_1 : 0.002266s : 1.34% optimize.opt_a.with_stream_mark : 0.000070s : 0.04% optimize.opt_a.recompute_prepare : 0.000061s : 0.04% optimize.opt_a.updatestate_depend_eliminate : 0.000095s : 0.06% optimize.opt_a.updatestate_assign_eliminate : 0.000023s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000019s : 0.01% optimize.opt_a.parameter_eliminate : 0.000007s : 0.00% optimize.opt_a.a_2 : 0.000563s : 0.33% optimize.opt_a.accelerated_algorithm : 0.000079s : 0.05% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000012s : 0.01% optimize.opt_a.shard_inline : 0.000038s : 0.02% optimize.opt_a.merge_send_recv : 0.000037s : 0.02% optimize.opt_a.auto_parallel : 0.000038s : 0.02% optimize.opt_a.parallel : 0.000034s : 0.02% optimize.opt_a.flash_sp : 0.000020s : 0.01% optimize.opt_a.merge_comm : 0.000023s : 0.01% optimize.opt_a.allreduce_fusion : 0.000022s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000092s : 0.05% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000056s : 0.03% optimize.opt_a.virtual_dataset : 0.000040s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000038s : 0.02% optimize.opt_a.virtual_output : 0.000041s : 0.02% optimize.opt_a.merge_forward : 0.000025s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000043s : 0.03% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000082s : 0.05% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000068s : 0.04% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000024s : 0.01% optimize.opt_a.meta_fg_expand : 0.000019s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000009s : 0.01% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000060s : 0.04% optimize.opt_a.a_after_grad : 0.000068s : 0.04% optimize.opt_a.renormalize : 0.002130s : 1.26% optimize.opt_a.add_forward_monad_depend : 0.000016s : 0.01% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000080s : 0.05% optimize.opt_a.cse : 0.000211s : 0.12% optimize.opt_a.a_3 : 0.000362s : 0.21% optimize.py_interpret_to_execute_after_opt_a : 0.000033s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000227s : 0.13% optimize.convert_after_rewriter : 0.000020s : 0.01% optimize.order_py_execute_after_rewriter : 0.000014s : 0.01% optimize.mutable_eliminate : 0.000920s : 0.54% optimize.opt_b.b_1 : 0.000473s : 0.28% optimize.opt_b.b_2 : 0.000023s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000019s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000010s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000011s : 0.01% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000090s : 0.05% optimize.optimize_parallel_all_gather_comm : 0.000043s : 0.03% optimize.overlap_param_gather : 0.000003s : 0.00% optimize.cconv : 0.000044s : 0.03% optimize.loop_unroll : 0.000684s : 0.40% optimize.opt_after_cconv.c_1 : 0.000143s : 0.08% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000018s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000010s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000010s : 0.01% optimize.opt_after_cconv.cse : 0.000069s : 0.04% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000078s : 0.05% optimize.tuple_transform.d_1 : 0.000155s : 0.09% optimize.tuple_transform.none_parameter_eliminate : 0.000003s : 0.00% optimize.tuple_transform.renormalize : 0.000001s : 0.00% optimize.tuple_transform.switch_simplify : 0.000023s : 0.01% optimize.partial_unused_args_eliminate : 0.000003s : 0.00% optimize.add_recomputation : 0.000152s : 0.09% optimize.cse_after_recomputation.cse : 0.000049s : 0.03% optimize.environ_conv : 0.000022s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000017s : 0.01% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000009s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000002s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000037s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000010s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000009s : 0.01% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000008s : 0.00% optimize.overlap_grad_flash_sp : 0.000053s : 0.03% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000022s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000033s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000043s : 0.03% optimize.symbol_engine_optimizer.opt_reshape : 0.000022s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000034s : 0.02% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000058s : 0.03% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.000899s : 0.53% validate : 0.000097s : 0.06% Time group info: ------[substitution.] 0.000676 177 1.67% : 0.000011s : 2: substitution.depend_value_elim 0.90% : 0.000006s : 13: substitution.elim_not_effective 0.74% : 0.000005s : 13: substitution.fold_const_symbol 2.37% : 0.000016s : 17: substitution.graph_param_transform 72.67% : 0.000491s : 8: substitution.inline 1.80% : 0.000012s : 26: substitution.j_node_and_user_rematch 3.58% : 0.000024s : 2: substitution.less_batch_normalization 2.65% : 0.000018s : 26: substitution.remove_not_recompute_node 1.46% : 0.000010s : 8: substitution.replace_old_param 6.30% : 0.000043s : 28: substitution.updatestate_pure_node_eliminater 5.84% : 0.000040s : 34: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.156442 2 98.94% : 0.154779s : 1: type_inference.infer 1.06% : 0.001663s : 1: type_inference.specialize ------[replace.] 0.000094 8 100.00% : 0.000094s : 8: replace.inline ------[match.] 0.000486 8 100.00% : 0.000486s : 8: match.inline ------[predicate.] 0.000769 4588 0.86% : 0.000007s : 47: predicate.accumulaten_eliminater 0.70% : 0.000005s : 17: predicate.ad_related_special_op_eliminate 0.63% : 0.000005s : 34: predicate.addn_check_dump 0.85% : 0.000007s : 47: predicate.addn_zero_filter 0.86% : 0.000007s : 47: predicate.adjust_all_reduce_mul_add 1.92% : 0.000015s : 81: predicate.arithmetic_simplify 0.86% : 0.000007s : 47: predicate.cast_eliminate 0.69% : 0.000005s : 34: predicate.check_bprop_eliminate 0.65% : 0.000005s : 34: predicate.compare_switch_simplify 0.17% : 0.000001s : 17: predicate.const_output_eliminate 0.67% : 0.000005s : 34: predicate.depend_value_elim 0.88% : 0.000007s : 47: predicate.dict_get_item_const_eliminator 0.93% : 0.000007s : 47: predicate.dict_get_item_eliminator 0.85% : 0.000007s : 47: predicate.dict_set_item_eliminator 0.86% : 0.000007s : 34: predicate.dumpgradient_eliminate 0.22% : 0.000002s : 17: predicate.elim_not_effective 0.50% : 0.000004s : 17: predicate.elim_shapecalc_of_broadcastargs 1.18% : 0.000009s : 64: predicate.environ_add_const_eliminate 1.14% : 0.000009s : 64: predicate.environ_get_add_eliminate 1.15% : 0.000009s : 64: predicate.environ_get_depend_swap 1.76% : 0.000014s : 98: predicate.environ_get_eliminate 1.17% : 0.000009s : 64: predicate.environ_get_set_eliminate 1.00% : 0.000008s : 55: predicate.exchange_switch_depend_value 1.43% : 0.000011s : 55: predicate.float_depend_g_call 0.60% : 0.000005s : 34: predicate.float_environ_get_switch 0.94% : 0.000007s : 51: predicate.float_tuple_getitem_switch 0.18% : 0.000001s : 17: predicate.fold_const_symbol 0.73% : 0.000006s : 34: predicate.get_grad_eliminate 0.22% : 0.000002s : 17: predicate.graph_param_transform 0.62% : 0.000005s : 34: predicate.incorporate_call 0.56% : 0.000004s : 34: predicate.incorporate_call_switch 5.04% : 0.000039s : 204: predicate.inline 0.98% : 0.000007s : 34: predicate.inline_without_move 0.31% : 0.000002s : 34: predicate.j_node_and_user_rematch 1.02% : 0.000008s : 34: predicate.less_batch_normalization 1.63% : 0.000013s : 81: predicate.list_to_tuple_eliminator_ 2.35% : 0.000018s : 128: predicate.load_eliminater 0.83% : 0.000006s : 17: predicate.loop_unroll_after_grad 1.34% : 0.000010s : 77: predicate.loop_unroll_before_grad 1.58% : 0.000012s : 81: predicate.make_slice_get_slice_eliminator 0.64% : 0.000005s : 34: predicate.merge_addn 0.65% : 0.000005s : 34: predicate.micro_step_allgather_replace 0.67% : 0.000005s : 34: predicate.mini_step_allgather_replace 0.83% : 0.000006s : 47: predicate.minmaximum_grad 0.92% : 0.000007s : 17: predicate.mutable_eliminate 0.43% : 0.000003s : 17: predicate.opt_reshape 0.43% : 0.000003s : 17: predicate.parallel_virtual_node 1.47% : 0.000011s : 55: predicate.partial_defer_inline 1.25% : 0.000010s : 64: predicate.partial_eliminate 0.85% : 0.000007s : 47: predicate.print_const_string_wrapper 0.66% : 0.000005s : 34: predicate.reduce_all_const_elim 1.04% : 0.000008s : 47: predicate.reduce_eliminate 2.25% : 0.000017s : 128: predicate.redundant_stop_gradient_eliminater 0.39% : 0.000003s : 34: predicate.remove_not_recompute_node 1.01% : 0.000008s : 81: predicate.replace_applicator 0.42% : 0.000003s : 34: predicate.replace_old_param 0.24% : 0.000002s : 17: predicate.reset_defer_inline 0.88% : 0.000007s : 47: predicate.reshape_eliminate 0.73% : 0.000006s : 34: predicate.row_tensor_add_zeros_like 0.39% : 0.000003s : 17: predicate.row_tensor_eliminate 9.60% : 0.000074s : 34: predicate.same_eliminate 0.45% : 0.000003s : 40: predicate.set_cell_output_no_recompute 0.87% : 0.000007s : 34: predicate.shard_identity_eliminate 0.69% : 0.000005s : 34: predicate.special_op_eliminate 0.70% : 0.000005s : 34: predicate.specialize_transform 0.83% : 0.000006s : 34: predicate.split_environ_get_set_with_tuple_value 0.77% : 0.000006s : 34: predicate.stack_unstack_eliminate 0.37% : 0.000003s : 17: predicate.switch_call_monad_eliminater 1.05% : 0.000008s : 55: predicate.switch_defer_inline 1.70% : 0.000013s : 89: predicate.switch_layer_defer_inline 3.70% : 0.000028s : 183: predicate.switch_simplify 0.87% : 0.000007s : 47: predicate.tile_eliminate 0.84% : 0.000006s : 47: predicate.transpose_eliminate 1.50% : 0.000012s : 81: predicate.tuple_list_convert_item_index_to_positive 1.56% : 0.000012s : 81: predicate.tuple_list_get_item_const_eliminator 1.39% : 0.000011s : 81: predicate.tuple_list_get_item_depend_reorder 2.52% : 0.000019s : 115: predicate.tuple_list_get_item_eliminator 1.50% : 0.000012s : 81: predicate.tuple_list_get_set_item_eliminator 2.26% : 0.000017s : 115: predicate.tuple_list_set_item_eliminator 1.52% : 0.000012s : 81: predicate.tuple_to_list_eliminator_ 2.42% : 0.000019s : 128: predicate.updatestate_pure_node_eliminater 3.19% : 0.000025s : 162: predicate.updatestate_useless_node_eliminater 0.37% : 0.000003s : 17: predicate.value_based_eliminate 0.78% : 0.000006s : 34: predicate.virtual_dataset_eliminate 0.83% : 0.000006s : 34: predicate.virtual_output_eliminate 0.32% : 0.000002s : 17: predicate.virtual_view_grad_eliminate 0.42% : 0.000003s : 17: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.001128 14 35.36% : 0.000399s : 4: func_graph_cloner_run.FuncGraphClonerGraph 8.07% : 0.000091s : 2: func_graph_cloner_run.FuncGraphClonerNode 56.57% : 0.000638s : 8: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.199031 192 0.00% : 0.000004s : 1: ForceFp32Comm 2.28% : 0.004539s : 1: add_attr 2.27% : 0.004524s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.08% : 0.000160s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.14% : 0.000285s : 1: auto_monad 0.03% : 0.000064s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.32% : 0.000629s : 1: bootstrap 0.02% : 0.000048s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.02% : 0.000042s : 1: control_data_broadcast_order 0.01% : 0.000025s : 1: convert_after_rewriter 0.04% : 0.000070s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.01% : 0.000026s : 1: environ_conv 0.09% : 0.000186s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.01% : 0.000013s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000005s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.01% : 0.000012s : 1: label_micro_interleaved_index 0.35% : 0.000702s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000007s : 1: micro_interleaved_order_control 0.47% : 0.000938s : 1: mutable_eliminate 0.01% : 0.000014s : 1: offloading_packed_experts 0.02% : 0.000036s : 1: opt.transform.loop_unroll_optimizer 0.02% : 0.000044s : 1: opt.transform.mutable_eliminate 1.93% : 0.003851s : 78: opt.transform.opt_a 0.07% : 0.000141s : 1: opt.transform.opt_after_cconv 0.04% : 0.000075s : 1: opt.transform.opt_after_jit_grad 0.23% : 0.000457s : 28: opt.transform.opt_b 0.09% : 0.000174s : 2: opt.transform.opt_trans_graph 0.06% : 0.000126s : 4: opt.transform.symbol_engine_opt 3.89% : 0.007733s : 1: opt_a 0.16% : 0.000316s : 1: opt_after_cconv 0.46% : 0.000914s : 1: opt_after_jit_grad 0.35% : 0.000690s : 1: opt_b 6.09% : 0.012127s : 1: optimize 0.02% : 0.000048s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000017s : 1: order_py_execute_after_rewriter 0.03% : 0.000059s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000012s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000007s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000013s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000007s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.03% : 0.000061s : 1: pre_auto_parallel 0.02% : 0.000049s : 1: py_interpret_to_execute 0.02% : 0.000038s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.04% : 0.000083s : 1: remove_dup_value 0.60% : 0.001197s : 1: renormalize.infer 0.46% : 0.000915s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.12% : 0.000234s : 1: rewriter_after_opt_a 0.07% : 0.000142s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000021s : 1: swap_dp_allreduce_reducescatter 0.10% : 0.000205s : 1: symbol_engine_optimizer 0.11% : 0.000214s : 1: tuple_transform 78.66% : 0.156556s : 1: type_inference [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:39.022.400 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceSubExt-op0, the new abstract of Expand/_InplaceSubExt/Sub-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab44a7a3a010, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:39.022.481 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab44a7a3a010, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:39.022.768 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceRemainderTensorTensor-op0, the new abstract of Expand/_InplaceRemainderTensorTensor/FloorMod-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab4413e5f011, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:39.022.833 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab4413e5f011, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:39.023.082 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceDiv-op0, the new abstract of Expand/_InplaceDiv/RealDiv-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab299c19c012, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:39.023.121 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab299c19c012, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) TotalTime = 0.992831, [30] [bootstrap]: 0.00031646 [type_inference]: 0.24976 [event_method]: 0.00028443 [auto_monad]: 0.00037335 [graph_reusing]: 9.79e-06 [pre_auto_parallel]: 3.13e-06 [py_interpret_to_execute]: 6.219e-05 [rewriter_before_opt_a]: 0.0002272 [expand_dump_flag]: 4.01001e-06 [jit_opt_a]: 0.582902, [3] [Cycle 1]: 0.254419, [27] [switch_simplify]: 0.00015164 [loop_unroll]: 0.0001006 [a_1]: 0.00277312 [with_stream_mark]: 5.035e-05 [recompute_prepare]: 5.35e-05 [updatestate_depend_eliminate]: 9.99e-05 [updatestate_assign_eliminate]: 2.161e-05 [updatestate_loads_eliminate]: 1.877e-05 [parameter_eliminate]: 3.51001e-06 [specialize_transform]: 4e-05 [updatestate_useless_node_eliminater]: 4.361e-05 [accelerated_algorithm]: 6.652e-05 [meta_shard_fg_expand]: 8.80999e-06 [get_grad_eliminate_]: 3.532e-05 [merge_forward]: 2.088e-05 [cell_reuse_recompute_pass]: 1.52999e-06 [cell_reuse_handle_not_recompute_node_pass]: 6.974e-05 [j_node_and_user_rematch]: 6.182e-05 [meta_fg_expand]: 0.0959738 [replace_old_param]: 0.0003046 [inline_without_move]: 0.00029541 [renormalize]: 0.151859 [add_forward_monad_depend]: 6.318e-05 [auto_monad_grad]: 3.168e-05 [auto_monad_eliminator]: 0.00038421 [cse]: 0.00084755 [replace_applicator]: 0.00064266 [Cycle 2]: 0.195561, [27] [switch_simplify]: 0.00028366 [loop_unroll]: 0.00028047 [a_1]: 0.186708 [with_stream_mark]: 0.00013969 [recompute_prepare]: 5.864e-05 [updatestate_depend_eliminate]: 8.085e-05 [updatestate_assign_eliminate]: 2.454e-05 [updatestate_loads_eliminate]: 2.285e-05 [parameter_eliminate]: 3.01001e-06 [specialize_transform]: 4.701e-05 [updatestate_useless_node_eliminater]: 0.00019682 [accelerated_algorithm]: 5.362e-05 [meta_shard_fg_expand]: 1.844e-05 [get_grad_eliminate_]: 4.188e-05 [merge_forward]: 2.714e-05 [cell_reuse_recompute_pass]: 2.56e-06 [cell_reuse_handle_not_recompute_node_pass]: 9.336e-05 [j_node_and_user_rematch]: 8.235e-05 [meta_fg_expand]: 0.00029855 [replace_old_param]: 6.018e-05 [inline_without_move]: 4.448e-05 [renormalize]: 0.00621682 [add_forward_monad_depend]: 1.656e-05 [auto_monad_grad]: 3.58e-06 [auto_monad_eliminator]: 0.00014053 [cse]: 0.00021201 [replace_applicator]: 6.906e-05 [Cycle 3]: 0.00264023, [27] [switch_simplify]: 4.194e-05 [loop_unroll]: 3.951e-05 [a_1]: 0.00139131 [with_stream_mark]: 6.536e-05 [recompute_prepare]: 4.709e-05 [updatestate_depend_eliminate]: 2.927e-05 [updatestate_assign_eliminate]: 2.367e-05 [updatestate_loads_eliminate]: 2.274e-05 [parameter_eliminate]: 2.81e-06 [specialize_transform]: 4.312e-05 [updatestate_useless_node_eliminater]: 5.362e-05 [accelerated_algorithm]: 5.086e-05 [meta_shard_fg_expand]: 1.461e-05 [get_grad_eliminate_]: 3.991e-05 [merge_forward]: 2.427e-05 [cell_reuse_recompute_pass]: 2.72001e-06 [cell_reuse_handle_not_recompute_node_pass]: 8.869e-05 [j_node_and_user_rematch]: 7.748e-05 [meta_fg_expand]: 1.77e-05 [replace_old_param]: 5.105e-05 [inline_without_move]: 4.029e-05 [renormalize]: 1.80007e-07 [add_forward_monad_depend]: 4.03999e-06 [auto_monad_grad]: 2.84001e-06 [auto_monad_eliminator]: 8.376e-05 [cse]: 0.00013835 [replace_applicator]: 4.373e-05 [py_interpret_to_execute_after_opt_a]: 4.942e-05 [rewriter_after_opt_a]: 0.00025668 [convert_after_rewriter]: 3.689e-05 [order_py_execute_after_rewriter]: 2.723e-05 [mutable_eliminate]: 0.00090038 [jit_opt_b]: 0.00064315, [2] [Cycle 1]: 0.00040417, [2] [frontend_op_eliminate]: 0.000272 [inline_after_opt_a]: 0.0001088 [Cycle 2]: 0.00022214, [2] [frontend_op_eliminate]: 0.00011407 [inline_after_opt_a]: 9.657e-05 [cconv]: 5.124e-05 [loop_unroll]: 0.00062862 [jit_opt_after_cconv]: 0.00074134, [1] [Cycle 1]: 0.00073165, [11] [c_1]: 0.00023642 [parameter_eliminate]: 6.67002e-06 [updatestate_depend_eliminate]: 4.413e-05 [updatestate_assign_eliminate]: 2.461e-05 [updatestate_loads_eliminate]: 2.546e-05 [cse]: 0.00015089 [call_graph_tuple_transform]: 0.00010439 [tuple_list_get_item_eliminator]: 3.883e-05 [none_parameter_eliminate]: 2.85002e-06 [renormalize]: 1.35999e-06 [switch_simplify]: 3.736e-05 [remove_dup_value]: 0.0001738 [partial_unused_args_eliminate]: 2.93998e-06 [environ_conv]: 3.197e-05 [add_recomputation]: 0.00027233 [cse_after_recomputation]: 0.00012507, [1] [Cycle 1]: 0.00011463, [1] [cse]: 0.0001 [auto_monad_reorder]: 0.152875 [get_jit_bprop_graph]: 6.17001e-06 [rewriter_after_jit_bprop_graph]: 1.424e-05 [opt_after_jit_grad]: 0.00108842 [symbol_engine_optimizer]: 0.00037242, [1] [Cycle 1]: 0.00035887, [6] [build]: 4.475e-05 [elim_shapecalc]: 5.769e-05 [elim_not_effective]: 9.096e-05 [opt_reshape]: 4.506e-05 [fold_const_symbol]: 7.294e-05 [renormalize]: 9.00007e-07 [validate]: 0.00016199 Sums bootstrap : 0.000316s : 0.04% type_inference : 0.249760s : 29.01% event_method : 0.000284s : 0.03% auto_monad : 0.000373s : 0.04% graph_reusing : 0.000010s : 0.00% pre_auto_parallel : 0.000003s : 0.00% py_interpret_to_execute : 0.000062s : 0.01% rewriter_before_opt_a : 0.000227s : 0.03% expand_dump_flag : 0.000004s : 0.00% jit_opt_a.switch_simplify : 0.000477s : 0.06% jit_opt_a.loop_unroll : 0.000421s : 0.05% jit_opt_a.a_1 : 0.190872s : 22.17% jit_opt_a.with_stream_mark : 0.000255s : 0.03% jit_opt_a.recompute_prepare : 0.000159s : 0.02% jit_opt_a.updatestate_depend_eliminate : 0.000210s : 0.02% jit_opt_a.updatestate_assign_eliminate : 0.000070s : 0.01% jit_opt_a.updatestate_loads_eliminate : 0.000064s : 0.01% jit_opt_a.parameter_eliminate : 0.000009s : 0.00% jit_opt_a.specialize_transform : 0.000130s : 0.02% jit_opt_a.updatestate_useless_node_eliminater : 0.000294s : 0.03% jit_opt_a.accelerated_algorithm : 0.000171s : 0.02% jit_opt_a.meta_shard_fg_expand : 0.000042s : 0.00% jit_opt_a.get_grad_eliminate_ : 0.000117s : 0.01% jit_opt_a.merge_forward : 0.000072s : 0.01% jit_opt_a.cell_reuse_recompute_pass : 0.000007s : 0.00% jit_opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000252s : 0.03% jit_opt_a.j_node_and_user_rematch : 0.000222s : 0.03% jit_opt_a.meta_fg_expand : 0.096290s : 11.18% jit_opt_a.replace_old_param : 0.000416s : 0.05% jit_opt_a.inline_without_move : 0.000380s : 0.04% jit_opt_a.renormalize : 0.158076s : 18.36% jit_opt_a.add_forward_monad_depend : 0.000084s : 0.01% jit_opt_a.auto_monad_grad : 0.000038s : 0.00% jit_opt_a.auto_monad_eliminator : 0.000609s : 0.07% jit_opt_a.cse : 0.001198s : 0.14% jit_opt_a.replace_applicator : 0.000755s : 0.09% py_interpret_to_execute_after_opt_a : 0.000049s : 0.01% rewriter_after_opt_a : 0.000257s : 0.03% convert_after_rewriter : 0.000037s : 0.00% order_py_execute_after_rewriter : 0.000027s : 0.00% mutable_eliminate : 0.000900s : 0.10% jit_opt_b.frontend_op_eliminate : 0.000386s : 0.04% jit_opt_b.inline_after_opt_a : 0.000205s : 0.02% cconv : 0.000051s : 0.01% loop_unroll : 0.000629s : 0.07% jit_opt_after_cconv.c_1 : 0.000236s : 0.03% jit_opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% jit_opt_after_cconv.updatestate_depend_eliminate : 0.000044s : 0.01% jit_opt_after_cconv.updatestate_assign_eliminate : 0.000025s : 0.00% jit_opt_after_cconv.updatestate_loads_eliminate : 0.000025s : 0.00% jit_opt_after_cconv.cse : 0.000151s : 0.02% jit_opt_after_cconv.call_graph_tuple_transform : 0.000104s : 0.01% jit_opt_after_cconv.tuple_list_get_item_eliminator : 0.000039s : 0.00% jit_opt_after_cconv.none_parameter_eliminate : 0.000003s : 0.00% jit_opt_after_cconv.renormalize : 0.000001s : 0.00% jit_opt_after_cconv.switch_simplify : 0.000037s : 0.00% remove_dup_value : 0.000174s : 0.02% partial_unused_args_eliminate : 0.000003s : 0.00% environ_conv : 0.000032s : 0.00% add_recomputation : 0.000272s : 0.03% cse_after_recomputation.cse : 0.000100s : 0.01% auto_monad_reorder : 0.152875s : 17.76% get_jit_bprop_graph : 0.000006s : 0.00% rewriter_after_jit_bprop_graph : 0.000014s : 0.00% opt_after_jit_grad : 0.001088s : 0.13% symbol_engine_optimizer.build : 0.000045s : 0.01% symbol_engine_optimizer.elim_shapecalc : 0.000058s : 0.01% symbol_engine_optimizer.elim_not_effective : 0.000091s : 0.01% symbol_engine_optimizer.opt_reshape : 0.000045s : 0.01% symbol_engine_optimizer.fold_const_symbol : 0.000073s : 0.01% symbol_engine_optimizer.renormalize : 0.000001s : 0.00% validate : 0.000162s : 0.02% Time group info: ------[substitution.] 0.005570 842 0.67% : 0.000037s : 7: substitution.addn_check_dump 1.58% : 0.000088s : 7: substitution.addn_zero_filter 4.51% : 0.000251s : 20: substitution.arithmetic_simplify 0.22% : 0.000012s : 1: substitution.cast_eliminate 0.67% : 0.000037s : 12: substitution.depend_value_elim 0.24% : 0.000013s : 37: substitution.elim_not_effective 0.21% : 0.000012s : 37: substitution.fold_const_symbol 15.97% : 0.000890s : 4: substitution.getattr_setattr_resolve 0.52% : 0.000029s : 41: substitution.graph_param_transform 54.54% : 0.003038s : 54: substitution.inline 1.34% : 0.000075s : 12: substitution.inline_without_move 0.75% : 0.000042s : 103: substitution.j_node_and_user_rematch 0.66% : 0.000037s : 8: substitution.less_batch_normalization 0.11% : 0.000006s : 6: substitution.load_eliminater 0.48% : 0.000027s : 7: substitution.merge_addn 0.76% : 0.000043s : 33: substitution.minmaximum_grad 0.43% : 0.000024s : 12: substitution.partial_eliminate 1.07% : 0.000060s : 103: substitution.remove_not_recompute_node 3.22% : 0.000180s : 53: substitution.replace_applicator 0.65% : 0.000036s : 48: substitution.replace_old_param 0.07% : 0.000004s : 1: substitution.set_cell_output_no_recompute 1.87% : 0.000104s : 33: substitution.tuple_list_convert_item_index_to_positive 1.25% : 0.000070s : 33: substitution.tuple_list_get_item_depend_reorder 4.43% : 0.000247s : 94: substitution.tuple_list_get_item_eliminator 0.98% : 0.000055s : 30: substitution.updatestate_pure_node_eliminater 1.42% : 0.000079s : 45: substitution.updatestate_useless_node_eliminater 1.35% : 0.000075s : 1: substitution.zero_like_fill_zero ------[type_inference.] 0.249613 2 98.83% : 0.246680s : 1: type_inference.infer 1.17% : 0.002933s : 1: type_inference.specialize ------[replace.] 0.001718 128 1.07% : 0.000018s : 1: replace.arithmetic_simplify 0.64% : 0.000011s : 1: replace.cast_eliminate 3.67% : 0.000063s : 3: replace.getattr_setattr_resolve 46.90% : 0.000806s : 54: replace.inline 6.72% : 0.000115s : 6: replace.replace_applicator 37.70% : 0.000648s : 61: replace.tuple_list_get_item_eliminator 0.75% : 0.000013s : 1: replace.updatestate_useless_node_eliminater 2.55% : 0.000044s : 1: replace.zero_like_fill_zero ------[match.] 0.004199 128 3.23% : 0.000136s : 1: match.arithmetic_simplify 0.26% : 0.000011s : 1: match.cast_eliminate 18.45% : 0.000775s : 3: match.getattr_setattr_resolve 71.46% : 0.003000s : 54: match.inline 1.00% : 0.000042s : 6: match.replace_applicator 3.40% : 0.000143s : 61: match.tuple_list_get_item_eliminator 0.44% : 0.000019s : 1: match.updatestate_useless_node_eliminater 1.75% : 0.000074s : 1: match.zero_like_fill_zero ------[predicate.] 0.002440 16800 1.48% : 0.000036s : 260: predicate.accumulaten_eliminater 0.92% : 0.000022s : 41: predicate.ad_related_special_op_eliminate 1.41% : 0.000035s : 260: predicate.addn_check_dump 1.48% : 0.000036s : 260: predicate.addn_zero_filter 1.90% : 0.000046s : 261: predicate.arithmetic_simplify 1.43% : 0.000035s : 262: predicate.cast_eliminate 0.48% : 0.000012s : 83: predicate.check_bprop_eliminate 1.38% : 0.000034s : 260: predicate.compare_switch_simplify 1.47% : 0.000036s : 260: predicate.depend_value_elim 1.36% : 0.000033s : 262: predicate.dict_get_item_const_eliminator 1.45% : 0.000035s : 262: predicate.dict_get_item_eliminator 1.42% : 0.000035s : 262: predicate.dict_set_item_eliminator 0.35% : 0.000008s : 41: predicate.dumpgradient_eliminate 0.14% : 0.000003s : 41: predicate.elim_not_effective 0.31% : 0.000008s : 41: predicate.elim_shapecalc_of_broadcastargs 1.40% : 0.000034s : 262: predicate.environ_add_const_eliminate 1.39% : 0.000034s : 262: predicate.environ_get_add_eliminate 1.37% : 0.000033s : 262: predicate.environ_get_depend_swap 1.40% : 0.000034s : 262: predicate.environ_get_eliminate 1.38% : 0.000034s : 262: predicate.environ_get_set_eliminate 0.12% : 0.000003s : 41: predicate.fold_const_symbol 0.75% : 0.000018s : 125: predicate.get_grad_eliminate 0.28% : 0.000007s : 20: predicate.getattr_setattr_resolve 0.11% : 0.000003s : 41: predicate.graph_param_transform 4.51% : 0.000110s : 501: predicate.inline 2.04% : 0.000050s : 310: predicate.inline_without_move 0.37% : 0.000009s : 125: predicate.j_node_and_user_rematch 0.86% : 0.000021s : 125: predicate.less_batch_normalization 1.77% : 0.000043s : 323: predicate.list_to_tuple_eliminator_ 2.30% : 0.000056s : 364: predicate.load_eliminater 0.44% : 0.000011s : 41: predicate.loop_unroll_after_grad 2.75% : 0.000067s : 477: predicate.loop_unroll_before_grad 1.69% : 0.000041s : 303: predicate.make_slice_get_slice_eliminator 1.36% : 0.000033s : 260: predicate.merge_addn 1.39% : 0.000034s : 261: predicate.minmaximum_grad 0.58% : 0.000014s : 43: predicate.mutable_eliminate 0.32% : 0.000008s : 41: predicate.opt_reshape 2.54% : 0.000062s : 364: predicate.partial_eliminate 1.38% : 0.000034s : 260: predicate.print_const_string_wrapper 1.78% : 0.000044s : 261: predicate.reduce_eliminate 1.78% : 0.000043s : 323: predicate.redundant_stop_gradient_eliminater 0.39% : 0.000010s : 125: predicate.remove_not_recompute_node 2.78% : 0.000068s : 731: predicate.replace_applicator 0.96% : 0.000023s : 310: predicate.replace_old_param 0.25% : 0.000006s : 82: predicate.reset_defer_inline 1.45% : 0.000035s : 261: predicate.reshape_eliminate 1.39% : 0.000034s : 260: predicate.row_tensor_add_zeros_like 0.54% : 0.000013s : 83: predicate.row_tensor_eliminate 1.43% : 0.000035s : 260: predicate.same_eliminate 0.44% : 0.000011s : 135: predicate.set_cell_output_no_recompute 0.76% : 0.000018s : 124: predicate.special_op_eliminate 0.84% : 0.000021s : 125: predicate.specialize_transform 1.74% : 0.000042s : 260: predicate.split_environ_get_set_with_tuple_value 1.38% : 0.000034s : 260: predicate.stack_unstack_eliminate 0.26% : 0.000006s : 41: predicate.switch_call_monad_eliminater 3.80% : 0.000093s : 378: predicate.switch_defer_inline 2.35% : 0.000057s : 378: predicate.switch_layer_defer_inline 5.53% : 0.000135s : 896: predicate.switch_simplify 1.40% : 0.000034s : 261: predicate.tile_eliminate 1.37% : 0.000034s : 261: predicate.transpose_eliminate 1.76% : 0.000043s : 262: predicate.tuple_list_convert_item_index_to_positive 1.63% : 0.000040s : 262: predicate.tuple_list_get_item_depend_reorder 3.37% : 0.000082s : 446: predicate.tuple_list_get_item_eliminator 1.76% : 0.000043s : 262: predicate.tuple_list_set_item_eliminator 1.77% : 0.000043s : 323: predicate.tuple_to_list_eliminator_ 2.02% : 0.000049s : 364: predicate.updatestate_pure_node_eliminater 3.02% : 0.000074s : 490: predicate.updatestate_useless_node_eliminater 1.72% : 0.000042s : 260: predicate.value_based_eliminate 0.23% : 0.000006s : 41: predicate.virtual_view_grad_eliminate 0.61% : 0.000015s : 84: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.009342 122 59.58% : 0.005566s : 56: func_graph_cloner_run.FuncGraphClonerGraph 6.80% : 0.000636s : 9: func_graph_cloner_run.FuncGraphClonerNode 33.62% : 0.003141s : 57: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.347673 93 0.02% : 0.000279s : 1: add_recomputation 0.03% : 0.000382s : 1: auto_monad 11.35% : 0.152907s : 1: auto_monad_reorder 0.03% : 0.000340s : 1: bootstrap 0.00% : 0.000055s : 1: cconv 0.00% : 0.000040s : 1: convert_after_rewriter 0.01% : 0.000129s : 1: cse_after_recomputation 0.00% : 0.000035s : 1: environ_conv 0.02% : 0.000297s : 1: event_method 0.00% : 0.000006s : 1: expand_dump_flag 0.00% : 0.000013s : 1: get_jit_bprop_graph 0.00% : 0.000013s : 1: graph_reusing 43.25% : 0.582907s : 1: jit_opt_a 0.06% : 0.000745s : 1: jit_opt_after_cconv 0.05% : 0.000647s : 1: jit_opt_b 0.05% : 0.000642s : 1: loop_unroll 0.07% : 0.000917s : 1: mutable_eliminate 14.44% : 0.194585s : 39: opt.transform.jit_opt_a 0.03% : 0.000412s : 4: opt.transform.jit_opt_after_cconv 0.04% : 0.000576s : 8: opt.transform.jit_opt_b 0.00% : 0.000056s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000075s : 1: opt.transform.mutable_eliminate 0.01% : 0.000176s : 1: opt.transform.opt_after_jit_grad 0.08% : 0.001014s : 2: opt.transform.opt_resolve 0.02% : 0.000261s : 4: opt.transform.symbol_engine_opt 0.08% : 0.001110s : 1: opt_after_jit_grad 0.00% : 0.000030s : 1: order_py_execute_after_rewriter 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pre_auto_parallel 0.00% : 0.000065s : 1: py_interpret_to_execute 0.00% : 0.000053s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000180s : 1: remove_dup_value 10.81% : 0.145622s : 2: renormalize.infer 0.92% : 0.012413s : 2: renormalize.specialize 0.00% : 0.000017s : 1: rewriter_after_jit_bprop_graph 0.02% : 0.000268s : 1: rewriter_after_opt_a 0.02% : 0.000232s : 1: rewriter_before_opt_a 0.03% : 0.000376s : 1: symbol_engine_optimizer 18.53% : 0.249789s : 1: type_inference [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:40.535.816 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceSubExt-op0, the new abstract of Expand/_InplaceSubExt/Sub-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab44a37e6015, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:40.535.904 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab44a37e6015, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:40.536.214 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceRemainderTensorTensor-op0, the new abstract of Expand/_InplaceRemainderTensorTensor/FloorMod-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab44c55d4016, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:40.536.257 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab44c55d4016, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:40.536.621 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceDiv-op0, the new abstract of Expand/_InplaceDiv/RealDiv-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab44c7dec017, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:40.536.668 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab44c7dec017, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) TotalTime = 0.249794, [21] [bootstrap]: 0.00044724 [type_inference]: 0.217139 [event_method]: 0.00017183 [auto_monad]: 0.00028472 [graph_reusing]: 7.5e-06 [inline]: 3.6e-06 [add_attr]: 0.0183294, [1] [add_attr_with_inline]: 0.0183138, [1] [Cycle 1]: 0.00010006, [2] [tag_attr]: 3.954e-05 [meta_addattr_fg_expand]: 8.12e-06 [parallel-infer-symbol]: 4.23999e-06 [pre_auto_parallel]: 6.081e-05 [insert-virtual-dataset]: 3.02002e-06 [parallel-infer-symbol-second]: 6.90023e-07 [dataset_repeat_opt]: 2.92002e-06 [pipeline_split]: 1.67999e-06 [optimize]: 0.012012, [53] [py_interpret_to_execute]: 4.437e-05 [rewriter_before_opt_a]: 0.00013614 [opt_a]: 0.00741346, [2] [Cycle 1]: 0.00554104, [45] [expand_dump_flag]: 3.88999e-06 [switch_simplify]: 6.442e-05 [loop_unroll]: 4.93e-05 [a_1]: 0.00160267 [with_stream_mark]: 3.668e-05 [recompute_prepare]: 3.305e-05 [updatestate_depend_eliminate]: 8.746e-05 [updatestate_assign_eliminate]: 1.179e-05 [updatestate_loads_eliminate]: 9.79999e-06 [parameter_eliminate]: 3.78999e-06 [a_2]: 0.00034276 [accelerated_algorithm]: 4.72e-05 [shard]: 2.78e-06 [meta_shard_fg_expand]: 6.44001e-06 [shard_inline]: 2.099e-05 [merge_send_recv]: 1.694e-05 [auto_parallel]: 1.819e-05 [parallel]: 2.455e-05 [flash_sp]: 1.526e-05 [merge_comm]: 1.138e-05 [allreduce_fusion]: 1.014e-05 [matmul_add_comm_reduction]: 2.048e-05 [allreduce_slice_to_reducescatter]: 1.00999e-06 [virtual_shard_identity]: 2.007e-05 [virtual_dataset]: 1.918e-05 [get_grad_eliminate_]: 1.855e-05 [virtual_output]: 1.824e-05 [merge_forward]: 1.064e-05 [cell_reuse_recompute_pass]: 2.15002e-06 [offload_activation]: 2.032e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.301e-05 [merge_recompute_call_nodes]: 1.69e-06 [before_grad]: 3.398e-05 [set_forward_comm_id_for_comm_node_pass]: 1.083e-05 [meta_fg_expand]: 8.65001e-06 [flash_sp_send_recv_attached]: 5.62001e-06 [receive_attached]: 2.17999e-06 [after_resolve]: 2.777e-05 [a_after_grad]: 3.149e-05 [renormalize]: 0.00211705 [add_forward_monad_depend]: 1.146e-05 [auto_monad_grad]: 3.29001e-06 [auto_monad_eliminator]: 4.592e-05 [cse]: 0.00015439 [a_3]: 0.00015741 [Cycle 2]: 0.00185618, [45] [expand_dump_flag]: 3.12002e-06 [switch_simplify]: 2.019e-05 [loop_unroll]: 1.746e-05 [a_1]: 0.00054598 [with_stream_mark]: 2.709e-05 [recompute_prepare]: 2.012e-05 [updatestate_depend_eliminate]: 1.16e-05 [updatestate_assign_eliminate]: 1.026e-05 [updatestate_loads_eliminate]: 9.25001e-06 [parameter_eliminate]: 2.90002e-06 [a_2]: 0.0002637 [accelerated_algorithm]: 2.517e-05 [shard]: 2.43002e-06 [meta_shard_fg_expand]: 6.44999e-06 [shard_inline]: 1.845e-05 [merge_send_recv]: 1.747e-05 [auto_parallel]: 1.892e-05 [parallel]: 1.114e-05 [flash_sp]: 4.80001e-06 [merge_comm]: 1.085e-05 [allreduce_fusion]: 1.186e-05 [matmul_add_comm_reduction]: 1.843e-05 [allreduce_slice_to_reducescatter]: 1.04998e-06 [virtual_shard_identity]: 2.247e-05 [virtual_dataset]: 1.786e-05 [get_grad_eliminate_]: 1.827e-05 [virtual_output]: 1.795e-05 [merge_forward]: 1.037e-05 [cell_reuse_recompute_pass]: 3.20002e-06 [offload_activation]: 2.188e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.899e-05 [merge_recompute_call_nodes]: 1.94999e-06 [before_grad]: 3.227e-05 [set_forward_comm_id_for_comm_node_pass]: 1.173e-05 [meta_fg_expand]: 8.03001e-06 [flash_sp_send_recv_attached]: 1.86e-06 [receive_attached]: 2.76e-06 [after_resolve]: 2.878e-05 [a_after_grad]: 3.058e-05 [renormalize]: 1.80007e-07 [add_forward_monad_depend]: 5.44e-06 [auto_monad_grad]: 2.74999e-06 [auto_monad_eliminator]: 3.571e-05 [cse]: 6.73e-05 [a_3]: 0.00013638 [py_interpret_to_execute_after_opt_a]: 4.068e-05 [slice_cell_reuse_recomputed_activation]: 2.56998e-06 [rewriter_after_opt_a]: 0.00024825 [convert_after_rewriter]: 2.225e-05 [order_py_execute_after_rewriter]: 1.19e-05 [mutable_eliminate]: 0.00092683 [opt_b]: 0.00068541, [1] [Cycle 1]: 0.00067421, [7] [b_1]: 0.00044857 [b_2]: 2.207e-05 [updatestate_depend_eliminate]: 2.117e-05 [updatestate_assign_eliminate]: 1.033e-05 [updatestate_loads_eliminate]: 9.94001e-06 [renormalize]: 1.39e-06 [cse]: 0.00011093 [optimize_parallel_all_gather_comm]: 4.638e-05 [overlap_param_gather]: 2.37001e-06 [cconv]: 4.507e-05 [loop_unroll]: 0.00083118 [opt_after_cconv]: 0.00035007, [1] [Cycle 1]: 0.00034028, [7] [c_1]: 0.00015654 [parameter_eliminate]: 7.33e-06 [updatestate_depend_eliminate]: 2.139e-05 [updatestate_assign_eliminate]: 1.016e-05 [updatestate_loads_eliminate]: 9.15999e-06 [cse]: 9.397e-05 [renormalize]: 8.70001e-07 [remove_dup_value]: 7.476e-05 [tuple_transform]: 0.0002053, [1] [Cycle 1]: 0.00019896, [4] [d_1]: 0.00014953 [none_parameter_eliminate]: 2.44001e-06 [renormalize]: 2.59985e-07 [switch_simplify]: 2.346e-05 [partial_unused_args_eliminate]: 2.20002e-06 [add_recomputation]: 0.00014615 [cse_after_recomputation]: 6.553e-05, [1] [Cycle 1]: 5.955e-05, [1] [cse]: 5.077e-05 [environ_conv]: 2.172e-05 [swap_dp_allreduce_reducescatter]: 1.831e-05 [bias_add_comm_swap]: 3.46999e-06 [label_micro_interleaved_index]: 8.08999e-06 [label_fine_grained_interleaved_index]: 3.14999e-06 [merge_cast_opt]: 1.66002e-06 [slice_recompute_activation]: 2.38998e-06 [micro_interleaved_order_control]: 2.88e-06 [assign_add_opt]: 1.60999e-06 [ForceFp32Comm]: 1.30999e-06 [remove_cast_before_assign_add]: 1.19e-06 [full_micro_interleaved_order_control]: 2.70002e-06 [reorder_send_recv_between_fp_bp]: 2.63e-06 [comm_op_add_attrs]: 1.50999e-06 [add_comm_op_reuse_tag]: 1.36002e-06 [interleave_split_concat_branches]: 1.21002e-06 [interleave_parallel_branches]: 1.32e-06 [overlap_opt_shard_in_pipeline]: 1.74e-06 [overlap_opt_shard_grad_in_pipeline]: 2.19001e-06 [control_data_broadcast_order]: 3.561e-05 [grouped_pairwise_exchange_alltoall]: 1.79e-06 [offloading_packed_experts]: 8.50001e-06 [overlap_recompute_and_grad_model_parallel]: 9.98998e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.40001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.80001e-06 [overlap_recompute_comm]: 2.69999e-06 [overlap_grad_ring_attention]: 8.66002e-06 [overlap_grad_flash_sp]: 5.055e-05 [begin_end_overlap_inline]: 5.59987e-07 [split_matmul_comm_elemetwise]: 2.39001e-06 [split_layernorm_comm]: 1.77999e-06 [handle_group_info]: 1.12e-06 [symbol_engine_optimizer]: 0.00020174, [1] [Cycle 1]: 0.00019323, [6] [build]: 1.731e-05 [elim_shapecalc]: 3.695e-05 [elim_not_effective]: 4.144e-05 [opt_reshape]: 2.183e-05 [fold_const_symbol]: 3.376e-05 [renormalize]: 2.80008e-07 [detach_backward]: 2.71e-06 [pipeline_parallel_scheduler]: 1.57001e-06 [auto_monad_reorder]: 6.012e-05 [get_jit_bprop_graph]: 2.26e-06 [rewriter_after_jit_bprop_graph]: 8.13999e-06 [opt_after_jit_grad]: 0.00092449 [validate]: 9.962e-05 Sums bootstrap : 0.000447s : 0.19% type_inference : 0.217139s : 94.34% event_method : 0.000172s : 0.07% auto_monad : 0.000285s : 0.12% graph_reusing : 0.000007s : 0.00% inline : 0.000004s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000040s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000008s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000061s : 0.03% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000044s : 0.02% optimize.rewriter_before_opt_a : 0.000136s : 0.06% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000085s : 0.04% optimize.opt_a.loop_unroll : 0.000067s : 0.03% optimize.opt_a.a_1 : 0.002149s : 0.93% optimize.opt_a.with_stream_mark : 0.000064s : 0.03% optimize.opt_a.recompute_prepare : 0.000053s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000099s : 0.04% optimize.opt_a.updatestate_assign_eliminate : 0.000022s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000019s : 0.01% optimize.opt_a.parameter_eliminate : 0.000007s : 0.00% optimize.opt_a.a_2 : 0.000606s : 0.26% optimize.opt_a.accelerated_algorithm : 0.000072s : 0.03% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000013s : 0.01% optimize.opt_a.shard_inline : 0.000039s : 0.02% optimize.opt_a.merge_send_recv : 0.000034s : 0.01% optimize.opt_a.auto_parallel : 0.000037s : 0.02% optimize.opt_a.parallel : 0.000036s : 0.02% optimize.opt_a.flash_sp : 0.000020s : 0.01% optimize.opt_a.merge_comm : 0.000022s : 0.01% optimize.opt_a.allreduce_fusion : 0.000022s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000039s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000043s : 0.02% optimize.opt_a.virtual_dataset : 0.000037s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000037s : 0.02% optimize.opt_a.virtual_output : 0.000036s : 0.02% optimize.opt_a.merge_forward : 0.000021s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000042s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000082s : 0.04% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000066s : 0.03% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000023s : 0.01% optimize.opt_a.meta_fg_expand : 0.000017s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000007s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000057s : 0.02% optimize.opt_a.a_after_grad : 0.000062s : 0.03% optimize.opt_a.renormalize : 0.002117s : 0.92% optimize.opt_a.add_forward_monad_depend : 0.000017s : 0.01% optimize.opt_a.auto_monad_grad : 0.000006s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000082s : 0.04% optimize.opt_a.cse : 0.000222s : 0.10% optimize.opt_a.a_3 : 0.000294s : 0.13% optimize.py_interpret_to_execute_after_opt_a : 0.000041s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000248s : 0.11% optimize.convert_after_rewriter : 0.000022s : 0.01% optimize.order_py_execute_after_rewriter : 0.000012s : 0.01% optimize.mutable_eliminate : 0.000927s : 0.40% optimize.opt_b.b_1 : 0.000449s : 0.19% optimize.opt_b.b_2 : 0.000022s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000021s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000010s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000010s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000111s : 0.05% optimize.optimize_parallel_all_gather_comm : 0.000046s : 0.02% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000045s : 0.02% optimize.loop_unroll : 0.000831s : 0.36% optimize.opt_after_cconv.c_1 : 0.000157s : 0.07% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000021s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000010s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000009s : 0.00% optimize.opt_after_cconv.cse : 0.000094s : 0.04% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000075s : 0.03% optimize.tuple_transform.d_1 : 0.000150s : 0.06% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000023s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000146s : 0.06% optimize.cse_after_recomputation.cse : 0.000051s : 0.02% optimize.environ_conv : 0.000022s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000018s : 0.01% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000008s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000002s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000036s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000009s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000010s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000009s : 0.00% optimize.overlap_grad_flash_sp : 0.000051s : 0.02% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000017s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000037s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000041s : 0.02% optimize.symbol_engine_optimizer.opt_reshape : 0.000022s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000034s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000060s : 0.03% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.000924s : 0.40% validate : 0.000100s : 0.04% Time group info: ------[substitution.] 0.000620 177 1.85% : 0.000011s : 2: substitution.depend_value_elim 1.02% : 0.000006s : 13: substitution.elim_not_effective 0.82% : 0.000005s : 13: substitution.fold_const_symbol 2.45% : 0.000015s : 17: substitution.graph_param_transform 69.33% : 0.000430s : 8: substitution.inline 1.90% : 0.000012s : 26: substitution.j_node_and_user_rematch 3.78% : 0.000023s : 2: substitution.less_batch_normalization 3.23% : 0.000020s : 26: substitution.remove_not_recompute_node 1.62% : 0.000010s : 8: substitution.replace_old_param 7.50% : 0.000047s : 28: substitution.updatestate_pure_node_eliminater 6.51% : 0.000040s : 34: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.217049 2 99.06% : 0.215001s : 1: type_inference.infer 0.94% : 0.002048s : 1: type_inference.specialize ------[replace.] 0.000088 8 100.00% : 0.000088s : 8: replace.inline ------[match.] 0.000425 8 100.00% : 0.000425s : 8: match.inline ------[predicate.] 0.000685 4588 0.99% : 0.000007s : 47: predicate.accumulaten_eliminater 0.89% : 0.000006s : 17: predicate.ad_related_special_op_eliminate 0.68% : 0.000005s : 34: predicate.addn_check_dump 1.04% : 0.000007s : 47: predicate.addn_zero_filter 0.97% : 0.000007s : 47: predicate.adjust_all_reduce_mul_add 2.00% : 0.000014s : 81: predicate.arithmetic_simplify 0.95% : 0.000006s : 47: predicate.cast_eliminate 0.71% : 0.000005s : 34: predicate.check_bprop_eliminate 0.65% : 0.000004s : 34: predicate.compare_switch_simplify 0.19% : 0.000001s : 17: predicate.const_output_eliminate 0.71% : 0.000005s : 34: predicate.depend_value_elim 0.98% : 0.000007s : 47: predicate.dict_get_item_const_eliminator 1.02% : 0.000007s : 47: predicate.dict_get_item_eliminator 0.93% : 0.000006s : 47: predicate.dict_set_item_eliminator 0.95% : 0.000007s : 34: predicate.dumpgradient_eliminate 0.24% : 0.000002s : 17: predicate.elim_not_effective 0.59% : 0.000004s : 17: predicate.elim_shapecalc_of_broadcastargs 1.29% : 0.000009s : 64: predicate.environ_add_const_eliminate 1.22% : 0.000008s : 64: predicate.environ_get_add_eliminate 1.24% : 0.000009s : 64: predicate.environ_get_depend_swap 2.01% : 0.000014s : 98: predicate.environ_get_eliminate 1.28% : 0.000009s : 64: predicate.environ_get_set_eliminate 1.14% : 0.000008s : 55: predicate.exchange_switch_depend_value 1.54% : 0.000011s : 55: predicate.float_depend_g_call 0.69% : 0.000005s : 34: predicate.float_environ_get_switch 1.02% : 0.000007s : 51: predicate.float_tuple_getitem_switch 0.19% : 0.000001s : 17: predicate.fold_const_symbol 0.78% : 0.000005s : 34: predicate.get_grad_eliminate 0.22% : 0.000002s : 17: predicate.graph_param_transform 0.67% : 0.000005s : 34: predicate.incorporate_call 0.62% : 0.000004s : 34: predicate.incorporate_call_switch 5.53% : 0.000038s : 204: predicate.inline 0.92% : 0.000006s : 34: predicate.inline_without_move 0.40% : 0.000003s : 34: predicate.j_node_and_user_rematch 1.10% : 0.000008s : 34: predicate.less_batch_normalization 1.71% : 0.000012s : 81: predicate.list_to_tuple_eliminator_ 2.53% : 0.000017s : 128: predicate.load_eliminater 1.26% : 0.000009s : 17: predicate.loop_unroll_after_grad 1.48% : 0.000010s : 77: predicate.loop_unroll_before_grad 1.93% : 0.000013s : 81: predicate.make_slice_get_slice_eliminator 0.69% : 0.000005s : 34: predicate.merge_addn 0.70% : 0.000005s : 34: predicate.micro_step_allgather_replace 0.78% : 0.000005s : 34: predicate.mini_step_allgather_replace 0.92% : 0.000006s : 47: predicate.minmaximum_grad 1.29% : 0.000009s : 17: predicate.mutable_eliminate 0.50% : 0.000003s : 17: predicate.opt_reshape 0.41% : 0.000003s : 17: predicate.parallel_virtual_node 1.50% : 0.000010s : 55: predicate.partial_defer_inline 1.35% : 0.000009s : 64: predicate.partial_eliminate 0.91% : 0.000006s : 47: predicate.print_const_string_wrapper 0.77% : 0.000005s : 34: predicate.reduce_all_const_elim 1.11% : 0.000008s : 47: predicate.reduce_eliminate 2.51% : 0.000017s : 128: predicate.redundant_stop_gradient_eliminater 0.49% : 0.000003s : 34: predicate.remove_not_recompute_node 1.19% : 0.000008s : 81: predicate.replace_applicator 0.50% : 0.000003s : 34: predicate.replace_old_param 0.25% : 0.000002s : 17: predicate.reset_defer_inline 0.92% : 0.000006s : 47: predicate.reshape_eliminate 0.76% : 0.000005s : 34: predicate.row_tensor_add_zeros_like 0.41% : 0.000003s : 17: predicate.row_tensor_eliminate 0.85% : 0.000006s : 34: predicate.same_eliminate 0.47% : 0.000003s : 40: predicate.set_cell_output_no_recompute 0.86% : 0.000006s : 34: predicate.shard_identity_eliminate 0.70% : 0.000005s : 34: predicate.special_op_eliminate 0.82% : 0.000006s : 34: predicate.specialize_transform 0.95% : 0.000006s : 34: predicate.split_environ_get_set_with_tuple_value 0.83% : 0.000006s : 34: predicate.stack_unstack_eliminate 0.40% : 0.000003s : 17: predicate.switch_call_monad_eliminater 1.16% : 0.000008s : 55: predicate.switch_defer_inline 1.84% : 0.000013s : 89: predicate.switch_layer_defer_inline 3.95% : 0.000027s : 183: predicate.switch_simplify 0.93% : 0.000006s : 47: predicate.tile_eliminate 0.98% : 0.000007s : 47: predicate.transpose_eliminate 1.69% : 0.000012s : 81: predicate.tuple_list_convert_item_index_to_positive 1.66% : 0.000011s : 81: predicate.tuple_list_get_item_const_eliminator 1.59% : 0.000011s : 81: predicate.tuple_list_get_item_depend_reorder 2.67% : 0.000018s : 115: predicate.tuple_list_get_item_eliminator 1.66% : 0.000011s : 81: predicate.tuple_list_get_set_item_eliminator 2.34% : 0.000016s : 115: predicate.tuple_list_set_item_eliminator 1.75% : 0.000012s : 81: predicate.tuple_to_list_eliminator_ 2.58% : 0.000018s : 128: predicate.updatestate_pure_node_eliminater 3.52% : 0.000024s : 162: predicate.updatestate_useless_node_eliminater 0.38% : 0.000003s : 17: predicate.value_based_eliminate 0.76% : 0.000005s : 34: predicate.virtual_dataset_eliminate 0.76% : 0.000005s : 34: predicate.virtual_output_eliminate 0.33% : 0.000002s : 17: predicate.virtual_view_grad_eliminate 0.38% : 0.000003s : 17: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.001258 14 31.83% : 0.000400s : 4: func_graph_cloner_run.FuncGraphClonerGraph 7.12% : 0.000090s : 2: func_graph_cloner_run.FuncGraphClonerNode 61.05% : 0.000768s : 8: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.286738 192 0.00% : 0.000004s : 1: ForceFp32Comm 6.39% : 0.018337s : 1: add_attr 6.39% : 0.018319s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.05% : 0.000152s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.10% : 0.000298s : 1: auto_monad 0.02% : 0.000068s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.17% : 0.000481s : 1: bootstrap 0.02% : 0.000050s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000040s : 1: control_data_broadcast_order 0.01% : 0.000029s : 1: convert_after_rewriter 0.02% : 0.000069s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.01% : 0.000026s : 1: environ_conv 0.06% : 0.000185s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000007s : 1: get_jit_bprop_graph 0.00% : 0.000012s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000012s : 1: label_micro_interleaved_index 0.30% : 0.000848s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.33% : 0.000948s : 1: mutable_eliminate 0.00% : 0.000012s : 1: offloading_packed_experts 0.02% : 0.000050s : 1: opt.transform.loop_unroll_optimizer 0.02% : 0.000054s : 1: opt.transform.mutable_eliminate 1.27% : 0.003655s : 78: opt.transform.opt_a 0.05% : 0.000154s : 1: opt.transform.opt_after_cconv 0.03% : 0.000075s : 1: opt.transform.opt_after_jit_grad 0.15% : 0.000433s : 28: opt.transform.opt_b 0.06% : 0.000169s : 2: opt.transform.opt_trans_graph 0.04% : 0.000128s : 4: opt.transform.symbol_engine_opt 2.59% : 0.007419s : 1: opt_a 0.12% : 0.000355s : 1: opt_after_cconv 0.33% : 0.000945s : 1: opt_after_jit_grad 0.24% : 0.000690s : 1: opt_b 4.19% : 0.012020s : 1: optimize 0.02% : 0.000051s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000015s : 1: order_py_execute_after_rewriter 0.02% : 0.000055s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000012s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000014s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000008s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000007s : 1: pipeline_parallel_scheduler 0.00% : 0.000004s : 1: pipeline_split 0.02% : 0.000066s : 1: pre_auto_parallel 0.02% : 0.000049s : 1: py_interpret_to_execute 0.02% : 0.000048s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.03% : 0.000081s : 1: remove_dup_value 0.40% : 0.001159s : 1: renormalize.infer 0.33% : 0.000942s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.09% : 0.000260s : 1: rewriter_after_opt_a 0.05% : 0.000141s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000022s : 1: swap_dp_allreduce_reducescatter 0.07% : 0.000206s : 1: symbol_engine_optimizer 0.07% : 0.000208s : 1: tuple_transform 75.74% : 0.217173s : 1: type_inference [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:43.430.474 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceSubExt-op0, the new abstract of Expand/_InplaceSubExt/Sub-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab44de157019, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:43.430.545 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab44de157019, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:43.430.844 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceRemainderTensorTensor-op0, the new abstract of Expand/_InplaceRemainderTensorTensor/FloorMod-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab44b2b9e020, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:43.430.896 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab44b2b9e020, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:43.431.145 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:266] IbTryExpandCNode] After expanding cnode Default/InplaceDiv-op0, the new abstract of Expand/_InplaceDiv/RealDiv-op0 does not match original cnode's abstract. new: AbstractTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), old: AbstractRefTensor(key: 0xaaab44d7785021, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny, is_inplace) [WARNING] SESSION(166799,ffff961eff30,python3.9):2026-01-29-17:42:43.431.178 [mindspore/ccsrc/backend/common/expander/fallback/expander_fallback.cc:274] IbTryExpandCNode] Restore new abstract to AbstractRefTensor new:AbstractRefTensor(key: 0xaaab44d7785021, ref_value: AbstractRefTensor(shape: (3, 3, 5, 3, 4, 1, 4), element: AbstractScalar(Type: Float32, Value: ValueAny, Shape: NoShape), value_ptr: 0xaaab16fa07f0, value: ValueAny), value: ValueAny) . [hook] pytest_runtest_teardown:test_inplace tests/st/graph_kernel/test_dvm_inplace.py::test_inplace,max_mem:4.0M =============================== warnings summary =============================== ../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/classifier/transdata/transdata_classifier.py:222 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/classifier/transdata/transdata_classifier.py:222: DeprecationWarning: invalid escape sequence \B """ ../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:143 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:143: DeprecationWarning: invalid escape sequence \c """ ../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:170 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:170: DeprecationWarning: invalid escape sequence \c """ ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for type is zero. setattr(self, word, getattr(machar, word).flat[0]) ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for type is zero. return self._float_to_str(self.smallest_subnormal) ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for type is zero. setattr(self, word, getattr(machar, word).flat[0]) ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for type is zero. return self._float_to_str(self.smallest_subnormal) ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2.py:57 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2.py:57: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad.py:56 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad.py:56: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2_grad") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad_reduce.py:48 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad_reduce.py:48: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2_grad_reduce") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul.py:51 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul.py:51: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:51 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:51: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul_grad") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:143 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:143: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul_grad_reduce") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad.py:92 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad.py:92: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer_grad_d") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad_reduce.py:49 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad_reduce.py:49: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer_grad_d_reduce") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad.py:91 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad.py:91: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel_grad_d") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad_reduce.py:48 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad_reduce.py:48: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel_grad_d_reduce") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel.py:52 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel.py:52: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_perchannel") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel_grad.py:81 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel_grad.py:81: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_perchannel_grad") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer.py:54 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer.py:54: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_per_layer") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer_grad.py:81 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer_grad.py:81: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_per_layer_grad") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perchannel.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perchannel.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("minmax_update_perchannel") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perlayer.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perlayer.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("minmax_update_perlayer") test_dvm_inplace.py::test_inplace /usr/local/Ascend/cann-8.5.0/python/site-packages/asc_op_compile_base/asc_op_compiler/ascendc_compile_gen_code.py:161: DeprecationWarning: invalid escape sequence \w match = re.search(f'{option}=(\w+)', ' '.join(compile_options)) -- Docs: https://docs.pytest.org/en/stable/warnings.html ================== 1 passed, 26 warnings in 353.73s (0:05:53) ==================