============================= test session starts ============================== platform linux -- Python 3.7.5, pytest-5.4.3, py-1.8.1, pluggy-0.13.1 rootdir: /home/jenkins/mindspore/testcases/testcases/tests/st/dyn_shape_dev, inifile: /home/jenkins/sault/virtual_test/virtualenv_008/sault/config/pytest.ini plugins: anyio-3.7.1, forked-1.1.3, xdist-1.32.0 [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:31.623.370 [trace_attr.c:105](tid:73092) platform is 1. [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:31.623.531 [trace_recorder.c:114](tid:73092) use root path: /home/jenkins/ascend/atrace [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:31.623.556 [trace_signal.c:133](tid:73092) register signal handler for signo 2 succeed. [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:31.623.567 [trace_signal.c:133](tid:73092) register signal handler for signo 15 succeed. [INFO] RUNTIME(73092,python3.7):2024-01-11-05:37:32.011.543 [runtime.cc:1159] 73092 GetAicoreNumByLevel: workingDev_=0 [INFO] RUNTIME(73092,python3.7):2024-01-11-05:37:32.011.611 [runtime.cc:4719] 73092 GetVisibleDevices: ASCEND_RT_VISIBLE_DEVICES param was not set collected 1 item test_roll.py [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.348.083 [process_mode_manager.cpp:109][OpenProcess][tid:73092] [ProcessModeManager] enter into open process deviceId[7] rankSize[0] [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.349.815 [process_mode_manager.cpp:379][InitTsdClient][tid:73092] [TsdClient] deviceId[7] begin to init hdc client [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.349.944 [version_verify.cpp:34][SetVersionInfo][tid:73092] VersionVerify: send client version to server [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.349.971 [version_verify.cpp:50][SetVersionInfo][tid:73092] send feature_info:{msg_type:35, features:{check before send aicpu package,}} [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.349.983 [version_verify.cpp:50][SetVersionInfo][tid:73092] send feature_info:{msg_type:37, features:{check before send open qs message,}} [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.350.374 [version_verify.cpp:66][PeerVersionCheck][tid:73092] VersionVerify: Check client version info, server[1230], client[1230] [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.350.388 [version_verify.cpp:87][ParseVersionInfo][tid:73092] VersionVerify: pass client version info success [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.350.397 [hdc_client.cpp:276][CheckHdcConnection][tid:73092] Service[2] create hdc success [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.350.410 [version_verify.cpp:120][SpecialFeatureCheck][tid:73092] VersionVerify: new type[35], supported [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.350.452 [process_mode_manager.cpp:748][GetDeviceCheckCode][tid:73092] [TsdClient][deviceId=7] [sessionId=1] wait package info respond [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.350.619 [process_mode_manager.cpp:379][InitTsdClient][tid:73092] [TsdClient] deviceId[7] begin to init hdc client [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.350.803 [version_verify.cpp:34][SetVersionInfo][tid:73092] VersionVerify: send client version to server [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.350.815 [version_verify.cpp:50][SetVersionInfo][tid:73092] send feature_info:{msg_type:35, features:{check before send aicpu package,}} [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.350.825 [version_verify.cpp:50][SetVersionInfo][tid:73092] send feature_info:{msg_type:37, features:{check before send open qs message,}} [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.350.983 [version_verify.cpp:66][PeerVersionCheck][tid:73092] VersionVerify: Check client version info, server[1230], client[1230] [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.350.995 [version_verify.cpp:87][ParseVersionInfo][tid:73092] VersionVerify: pass client version info success [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.351.004 [hdc_client.cpp:276][CheckHdcConnection][tid:73092] Service[2] create hdc success [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.351.014 [process_mode_manager.cpp:426][ConstructOpenMsg][tid:73092] [TsdClient] tsd get process sign successfully, procpid[73092] signSize[48] [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.351.045 [version_verify.cpp:112][SpecialFeatureCheck][tid:73092] VersionVerify: previous type[6], supported [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.351.065 [process_mode_manager.cpp:126][OpenProcess][tid:73092] [ProcessModeManager] deviceId[7] sessionId[1] rankSize[0], wait sub process start respond [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.754.514 [stub_process_mode_nowin.cpp:63][ProcessQueueForMdc][tid:73092] [TsdClient] it is unnecessary of current mode[0] chiptype[1] to grant queue auth to aicpusd [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.754.575 [stub_process_mode_nowin.cpp:101][OpenInHost][tid:73092] enter into OpenInHost deviceid[7] [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.754.586 [stub_process_mode_nowin.cpp:105][OpenInHost][tid:73092] host cpu not support [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.754.594 [process_mode_manager.cpp:156][OpenProcess][tid:73092] [TsdClient][deviceId=7] [sessionId=1] start hccp and computer process success [INFO] RUNTIME(73092,python3.7):2024-01-11-05:37:36.757.315 [device.cc:340] 73092 Init: isDoubledie:0, topologytype:0 [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:36.772.644 [atrace_api.c:28](tid:73092) AtraceCreate start [INFO] RUNTIME(73092,python3.7):2024-01-11-05:37:36.772.633 [npu_driver.cc:5428] 74292 GetDeviceStatus: GetDeviceStatus status=1. [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:36.772.899 [trace_rb_log.c:84](tid:73092) [RUNTIME_ATRACE_DEV7_TS0] create ring buffer success, buffer size : 131152. [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:36.772.917 [atrace_api.c:32](tid:73092) AtraceCreate end [INFO] TDT(73092,python3.7):2024-01-11-05:37:36.772.931 [client_manager.cpp:157][SetProfilingCallback][tid:73092] [TsdClient] set profiling callback success [TRACE] GE(73092,python3.7):2024-01-11-05:37:36.925.087 [status:INIT] [ge_api.cc:144]73092 GEInitializeImpl:GEInitialize start [INFO] PROFILING(73092,python3.7):2024-01-11-05:37:37.144.290 [msprofiler_impl.cpp:156] >>> (tid:73092) ProfNotifySetDevice called, is open: 1, devId: 7 [INFO] PROFILING(73092,python3.7):2024-01-11-05:37:37.144.426 [platform.cpp:38] >>> (tid:73092) Profiling platform version: 1.0. [INFO] PROFILING(73092,python3.7):2024-01-11-05:37:37.144.444 [ai_drv_dev_api.cpp:384] >>> (tid:73092) Succeeded to DrvGetApiVersion version: 0x72313 [TRACE] GE(73092,python3.7):2024-01-11-05:37:37.196.667 [status:RUNNING] [ge_api.cc:211]73092 GEInitializeImpl:Initializing environment [INFO] GE(73092,python3.7):2024-01-11-05:37:37.196.745 [gelib.cc:98][EVENT]73092 Initialize:[GEPERFTRACE] GE Init Start [INFO] GE(73092,python3.7):2024-01-11-05:37:37.197.019 [gelib.cc:307][EVENT]73092 SystemInitialize:Online infer init GELib success, device id :7 [INFO] DVPP(73092,python3.7):2024-01-11-05:37:37.557.037 [dvpp_engine.cc:41][ENGINE][Initialize:41][tid:73092]dvpp engine do not support [INFO] TUNE(73092,python3.7):2024-01-11-05:37:37.560.218 [cann_kb_pyfunc_mgr.cpp:72][CANNKB][Tid:73092]"CannKbPyfuncMgr: Enter PyObjectInit, reference_ is 0!" [INFO] TUNE(73092,python3.7):2024-01-11-05:37:37.560.258 [handle_manager.cpp:115][CANNKB][Tid:73092]"Start to run init functions to load dynamic python lib!" [INFO] TUNE(73092,python3.7):2024-01-11-05:37:37.560.315 [handle_manager.cpp:407][CANNKB][Tid:73092]"Init functions of loading dynamic python lib end!" [INFO] TUNE(73092,python3.7):2024-01-11-05:37:37.560.326 [cann_kb_pyfunc_mgr.cpp:24][CANNKB][Tid:73092]"CANN_KB_Py has already been initialized." [INFO] TUNE(73092,python3.7):2024-01-11-05:37:37.560.431 [cann_kb_pyfunc_mgr.cpp:117][CANNKB][Tid:73092]"CannKbPyfuncMgr: Run PyObjectInit successfully!" [INFO] HCCL(73092,python3.7):2024-01-11-05:37:49.618.279 [plugin_manager.cc:42][73092]hcom running normal mode. [INFO] DVPP(73092,python3.7):2024-01-11-05:37:49.618.855 [dvpp_engine.cc:92][ENGINE][GetOpsKernelInfoStores:92][tid:73092]dvpp ops kernel info store do not support [INFO] DVPP(73092,python3.7):2024-01-11-05:37:49.619.005 [dvpp_engine.cc:69][ENGINE][GetGraphOptimizerObjs:69][tid:73092]dvpp graph optimizer do not support [INFO] DVPP(73092,python3.7):2024-01-11-05:37:50.131.520 [dvpp_ops_kernel_builder.cc:48][ENGINE][Initialize:48][tid:73092]dvpp ops kernel builder do not support [INFO] GE(73092,python3.7):2024-01-11-05:37:50.139.884 [gelib.cc:169][EVENT]73092 Initialize:[GEPERFTRACE] The time cost of GELib::Initialize is [12943091] micro second. [TRACE] GE(73092,python3.7):2024-01-11-05:37:50.228.213 [status:STOP] [ge_api.cc:255]73092 GEInitializeImpl:GEInitialize finished [TRACE] GE(73092,python3.7):2024-01-11-05:37:50.228.345 [status:INIT] [ge_api.cc:398]73092 Session:Start to construct session. [TRACE] GE(73092,python3.7):2024-01-11-05:37:50.228.363 [status:RUNNING] [ge_api.cc:408]73092 Session:Creating session [INFO] GE(73092,python3.7):2024-01-11-05:37:50.228.743 [graph_var_manager.cc:1445][EVENT]73092 SetMemoryMallocSize:Total memory size is 34359738368 [INFO] GE(73092,python3.7):2024-01-11-05:37:50.228.759 [graph_var_manager.cc:1424][EVENT]73092 SetAllMemoryMaxValue:The graph_mem_max_size is 27917287424 and the var_mem_max_size is 5368709120 [INFO] PROFILING(73092,python3.7):2024-01-11-05:37:50.229.061 [msprofiler_impl.cpp:156] >>> (tid:73092) ProfNotifySetDevice called, is open: 1, devId: 7 [TRACE] GE(73092,python3.7):2024-01-11-05:37:50.229.925 [status:RUNNING] [ge_api.cc:411]73092 Session:Session id is 0 [TRACE] GE(73092,python3.7):2024-01-11-05:37:50.229.950 [status:STOP] [ge_api.cc:420]73092 Session:Session Constructor finished [INFO] PROFILING(73092,python3.7):2024-01-11-05:37:50.239.608 [platform.cpp:38] >>> (tid:73092) Profiling platform version: 1.0. [INFO] PROFILING(73092,python3.7):2024-01-11-05:37:50.239.638 [ai_drv_dev_api.cpp:384] >>> (tid:73092) Succeeded to DrvGetApiVersion version: 0x72313 [TRACE] GE(73092,python3.7):2024-01-11-05:37:50.239.810 [status:INIT] [ge_api.cc:144]73092 GEInitializeImpl:GEInitialize start TotalTime = 0.0620519, [20] [parse]: 0.0130899 [symbol_resolve]: 0.0289855, [1] [Cycle 1]: 0.0289161, [1] [resolve]: 0.0288916 [combine_like_graphs]: 8.80005e-07 [graph_reusing]: 2.85e-06 [meta_unpack_prepare]: 7.262e-05 [pre_cconv]: 4.09e-06 [abstract_specialize]: 0.004913 [pack_expand]: 1.182e-05 [auto_monad]: 8.056e-05 [inline]: 1.83e-06 [pre_auto_parallel]: 1.401e-05 [pipeline_split]: 3.62e-06 [optimize]: 0.00812843, [35] [py_interpret_to_execute]: 3.2e-06 [rewriter_before_opt_a]: 0.00012188 [opt_a]: 0.00748005, [2] [Cycle 1]: 0.00101892, [30] [expand_dump_flag]: 3.34e-06 [switch_simplify]: 1.92e-05 [a_1]: 0.00025395 [recompute_prepare]: 3.38e-06 [updatestate_depend_eliminate]: 6.42001e-06 [updatestate_assign_eliminate]: 4.09999e-06 [updatestate_loads_eliminate]: 3.11e-06 [parameter_eliminate]: 3.25e-06 [a_2]: 3.56e-05 [accelerated_algorithm]: 3.29e-06 [pynative_shard]: 1.51e-06 [auto_parallel]: 3.36e-06 [parallel]: 2e-05 [merge_comm]: 1.157e-05 [allreduce_fusion]: 1.91e-06 [virtual_dataset]: 3.06e-06 [get_grad_eliminate_]: 2.38e-06 [virtual_output]: 2.09e-06 [merge_forward]: 5.01e-06 [cell_reuse_recompute_pass]: 7.10002e-07 [cell_reuse_handle_not_recompute_node_pass]: 6.78e-06 [meta_fg_expand]: 3.28e-06 [after_resolve]: 5.39e-06 [a_after_grad]: 3.21e-06 [renormalize]: 0.0004046 [real_op_eliminate]: 6.2e-06 [auto_monad_grad]: 4.12e-06 [auto_monad_eliminator]: 1.088e-05 [cse]: 2.805e-05 [a_3]: 1.885e-05 [Cycle 2]: 0.00024277, [30] [expand_dump_flag]: 1.12e-06 [switch_simplify]: 2.6e-06 [a_1]: 1.875e-05 [recompute_prepare]: 2.13e-06 [updatestate_depend_eliminate]: 3.6e-06 [updatestate_assign_eliminate]: 2.65e-06 [updatestate_loads_eliminate]: 2.31e-06 [parameter_eliminate]: 9.89996e-07 [a_2]: 3.159e-05 [accelerated_algorithm]: 2.58e-06 [pynative_shard]: 1.18e-06 [auto_parallel]: 3.21e-06 [parallel]: 3.59e-06 [merge_comm]: 1.91999e-06 [allreduce_fusion]: 1.19e-06 [virtual_dataset]: 2.50999e-06 [get_grad_eliminate_]: 2.07e-06 [virtual_output]: 1.91e-06 [merge_forward]: 3.05e-06 [cell_reuse_recompute_pass]: 4.00003e-07 [cell_reuse_handle_not_recompute_node_pass]: 6.28e-06 [meta_fg_expand]: 2.15e-06 [after_resolve]: 3.97e-06 [a_after_grad]: 2.64e-06 [renormalize]: 7.99992e-08 [real_op_eliminate]: 2.18e-06 [auto_monad_grad]: 8.09996e-07 [auto_monad_eliminator]: 4.96e-06 [cse]: 1.136e-05 [a_3]: 1.556e-05 [py_interpret_to_execute_after_opt_a]: 3.63e-06 [slice_cell_reuse_recomputed_activation]: 2.26e-06 [rewriter_after_opt_a]: 2.525e-05 [convert_after_rewriter]: 6.78e-06 [order_py_execute_after_rewriter]: 5.17e-06 [opt_b]: 9.846e-05, [1] [Cycle 1]: 9.365e-05, [7] [b_1]: 4.551e-05 [b_2]: 3.16e-06 [updatestate_depend_eliminate]: 2.76e-06 [updatestate_assign_eliminate]: 2.57e-06 [updatestate_loads_eliminate]: 2.67e-06 [renormalize]: 3.49995e-07 [cse]: 1.082e-05 [cconv]: 2.173e-05 [opt_after_cconv]: 5.2e-05, [1] [Cycle 1]: 4.796e-05, [7] [c_1]: 5.89e-06 [parameter_eliminate]: 6.29996e-07 [updatestate_depend_eliminate]: 2.5e-06 [updatestate_assign_eliminate]: 2.03e-06 [updatestate_loads_eliminate]: 1.96999e-06 [cse]: 9.41e-06 [renormalize]: 2.3e-07 [remove_dup_value]: 1.294e-05 [tuple_transform]: 3.776e-05, [1] [Cycle 1]: 3.406e-05, [3] [d_1]: 1.553e-05 [d_2]: 7.14e-06 [renormalize]: 1.09998e-07 [add_cache_embedding]: 1.073e-05 [add_recomputation]: 4.631e-05 [cse_after_recomputation]: 1.93e-05, [1] [Cycle 1]: 1.503e-05, [1] [cse]: 1.043e-05 [environ_conv]: 1.855e-05 [label_micro_interleaved_index]: 2.03e-06 [label_fine_grained_interleaved_index]: 2.08e-06 [assign_add_opt]: 3.05e-06 [slice_recompute_activation]: 2.66e-06 [micro_interleaved_order_control]: 1.75e-06 [full_micro_interleaved_order_control]: 1.76e-06 [comp_comm_scheduling]: 2.14e-06 [reorder_send_recv_between_fp_bp]: 2.27999e-06 [comm_op_add_attrs]: 1.01e-06 [add_comm_op_reuse_tag]: 9e-07 [overlap_opt_shard_in_pipeline]: 1e-06 [grouped_pairwise_exchange_alltoall]: 1.53e-06 [overlap_recompute_and_grad_model_parallel]: 1.54e-06 [overlap_grad_matmul_and_grad_allreduce]: 6.99998e-07 [split_matmul_comm_elemetwise]: 1.99e-06 [split_layernorm_comm]: 2.04e-06 [process_send_recv_for_ge]: 2.13e-06 [handle_group_info]: 9e-07 [auto_monad_reorder]: 2.168e-05 [get_jit_bprop_graph]: 4.30002e-07 [eliminate_special_op_node]: 0.00046778 [validate]: 4.762e-05 [distribtued_split]: 1.13e-06 [task_emit]: 0.00598467 [execute]: 7.78e-06 Sums parse : 0.013090s : 23.77% symbol_resolve.resolve : 0.028892s : 52.46% combine_like_graphs : 0.000001s : 0.00% graph_reusing : 0.000003s : 0.01% meta_unpack_prepare : 0.000073s : 0.13% pre_cconv : 0.000004s : 0.01% abstract_specialize : 0.004913s : 8.92% pack_expand : 0.000012s : 0.02% auto_monad : 0.000081s : 0.15% inline : 0.000002s : 0.00% pre_auto_parallel : 0.000014s : 0.03% pipeline_split : 0.000004s : 0.01% optimize.py_interpret_to_execute : 0.000003s : 0.01% optimize.rewriter_before_opt_a : 0.000122s : 0.22% optimize.opt_a.expand_dump_flag : 0.000004s : 0.01% optimize.opt_a.switch_simplify : 0.000022s : 0.04% optimize.opt_a.a_1 : 0.000273s : 0.50% optimize.opt_a.recompute_prepare : 0.000006s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000010s : 0.02% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000005s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.01% optimize.opt_a.a_2 : 0.000067s : 0.12% optimize.opt_a.accelerated_algorithm : 0.000006s : 0.01% optimize.opt_a.pynative_shard : 0.000003s : 0.00% optimize.opt_a.auto_parallel : 0.000007s : 0.01% optimize.opt_a.parallel : 0.000024s : 0.04% optimize.opt_a.merge_comm : 0.000013s : 0.02% optimize.opt_a.allreduce_fusion : 0.000003s : 0.01% optimize.opt_a.virtual_dataset : 0.000006s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000004s : 0.01% optimize.opt_a.virtual_output : 0.000004s : 0.01% optimize.opt_a.merge_forward : 0.000008s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000001s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000013s : 0.02% optimize.opt_a.meta_fg_expand : 0.000005s : 0.01% optimize.opt_a.after_resolve : 0.000009s : 0.02% optimize.opt_a.a_after_grad : 0.000006s : 0.01% optimize.opt_a.renormalize : 0.000405s : 0.73% optimize.opt_a.real_op_eliminate : 0.000008s : 0.02% optimize.opt_a.auto_monad_grad : 0.000005s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000016s : 0.03% optimize.opt_a.cse : 0.000039s : 0.07% optimize.opt_a.a_3 : 0.000034s : 0.06% optimize.py_interpret_to_execute_after_opt_a : 0.000004s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000025s : 0.05% optimize.convert_after_rewriter : 0.000007s : 0.01% optimize.order_py_execute_after_rewriter : 0.000005s : 0.01% optimize.opt_b.b_1 : 0.000046s : 0.08% optimize.opt_b.b_2 : 0.000003s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000003s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.00% optimize.opt_b.renormalize : 0.000000s : 0.00% optimize.opt_b.cse : 0.000011s : 0.02% optimize.cconv : 0.000022s : 0.04% optimize.opt_after_cconv.c_1 : 0.000006s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000001s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000002s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000002s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000002s : 0.00% optimize.opt_after_cconv.cse : 0.000009s : 0.02% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000013s : 0.02% optimize.tuple_transform.d_1 : 0.000016s : 0.03% optimize.tuple_transform.d_2 : 0.000007s : 0.01% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.add_cache_embedding : 0.000011s : 0.02% optimize.add_recomputation : 0.000046s : 0.08% optimize.cse_after_recomputation.cse : 0.000010s : 0.02% optimize.environ_conv : 0.000019s : 0.03% optimize.label_micro_interleaved_index : 0.000002s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000002s : 0.00% optimize.assign_add_opt : 0.000003s : 0.01% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.comp_comm_scheduling : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000002s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000002s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.process_send_recv_for_ge : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% auto_monad_reorder : 0.000022s : 0.04% get_jit_bprop_graph : 0.000000s : 0.00% eliminate_special_op_node : 0.000468s : 0.85% validate : 0.000048s : 0.09% distribtued_split : 0.000001s : 0.00% task_emit : 0.005985s : 10.87% execute : 0.000008s : 0.01% Time group info: ------[substitution.] 0.028733 51 99.42% : 0.028567s : 12: substitution.getattr_setattr_resolve 0.02% : 0.000005s : 5: substitution.graph_param_transform 0.43% : 0.000123s : 4: substitution.inline 0.08% : 0.000024s : 17: substitution.meta_unpack_prepare 0.00% : 0.000001s : 5: substitution.partial_unused_args_eliminate 0.00% : 0.000001s : 4: substitution.remove_not_recompute_node 0.01% : 0.000002s : 2: substitution.replace_old_param 0.03% : 0.000010s : 2: substitution.tuple_list_get_item_eliminator ------[renormalize.] 0.000398 2 58.10% : 0.000231s : 1: renormalize.infer 41.90% : 0.000167s : 1: renormalize.specialize ------[replace.] 0.000224 16 77.90% : 0.000175s : 10: replace.getattr_setattr_resolve 16.32% : 0.000037s : 4: replace.inline 5.78% : 0.000013s : 2: replace.tuple_list_get_item_eliminator ------[match.] 0.028603 16 99.54% : 0.028471s : 10: match.getattr_setattr_resolve 0.43% : 0.000123s : 4: match.inline 0.03% : 0.000010s : 2: match.tuple_list_get_item_eliminator ------[func_graph_cloner_run.] 0.000737 13 69.32% : 0.000511s : 7: func_graph_cloner_run.FuncGraphClonerGraph 30.68% : 0.000226s : 6: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.029420 105 0.32% : 0.000094s : 52: opt.transform.opt_a 0.13% : 0.000037s : 23: opt.transform.opt_b 98.14% : 0.028872s : 2: opt.transform.opt_resolve 0.17% : 0.000049s : 1: opt.transforms.meta_unpack_prepare 1.12% : 0.000329s : 20: opt.transforms.opt_a 0.02% : 0.000005s : 1: opt.transforms.opt_after_cconv 0.01% : 0.000002s : 1: opt.transforms.opt_b 0.07% : 0.000021s : 2: opt.transforms.opt_trans_graph 0.03% : 0.000010s : 3: opt.transforms.special_op_eliminate [INFO] GE(73092,python3.7):2024-01-11-05:37:50.586.626 [scalable_config.cc:55][EVENT]77563 ScalableConfig:device total max size: 34359738368, page_mem_size_total_thresold: 32641751449, uncacheable_size_threshold: 17179869184 [INFO] GE(73092,python3.7):2024-01-11-05:37:50.665.574 [graph_var_manager.cc:1424][EVENT]77563 SetAllMemoryMaxValue:The graph_mem_max_size is 27917287424 and the var_mem_max_size is 5368709120 [INFO] GE(73092,python3.7):2024-01-11-05:37:50.665.694 [graph_manager.cc:1248][EVENT]77563 PreRun:PreRun start: graph node size 3, session id 1, graph id 0, graph name online. [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:50.666.533 [atrace_api.c:28](tid:77563) AtraceCreate start [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:50.666.602 [trace_rb_log.c:84](tid:77563) [RUNTIME_ATRACE_DEV64_TS0] create ring buffer success, buffer size : 131152. [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:50.666.615 [atrace_api.c:32](tid:77563) AtraceCreate end [INFO] TDT(73092,python3.7):2024-01-11-05:37:50.666.640 [client_manager.cpp:157][SetProfilingCallback][tid:77563] [TsdClient] set profiling callback success [INFO] GE(73092,python3.7):2024-01-11-05:37:50.667.530 [parallel_partitioner.cc:165][EVENT]77563 DoPipelinePartition:[GEPERFTRACE] The time cost of OptimizeSubgraph::PipelinePartition is [14] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.667.572 [parallel_partitioner.cc:178][EVENT]77563 DoFlowGraphPartition:[GEPERFTRACE] The time cost of OptimizeSubgraph::FlowGraphPartition is [14] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.667.619 [graph_prepare.cc:1378][EVENT]77563 Init:[GEPERFTRACE] The time cost of FileConstantUtils::ConvertFileConstToConst is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.668.278 [graph_manager.cc:1050][EVENT]77563 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.preparer.PrepareInit is [676] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.668.303 [graph_manager.cc:1052][EVENT]77563 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.HandleSummaryOp is [5] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.668.415 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of ForToWhilePass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.668.441 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of ProcessNetOutput::SavePass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.668.500 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of ProcessNetOutput::NetOutputPass is [47] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.668.513 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of ProcessNetOutput::DataPass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.668.591 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of CreateSubGraphWithScopePass is [13] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.668.608 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of SubgraphMultiDimsClonePass is [5] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.668.623 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of MultiBatchClonePass is [4] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.668.719 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of SplitVariableIntoSubgraphPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.668.738 [graph_manager.cc:1054][EVENT]77563 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.preparer.NormalizeGraph is [422] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.675.958 [graph_manager.cc:1055][EVENT]77563 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.OptimizeGraphInit is [7205] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.676.922 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of AssertPass is [0] micro second, call num is [6] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.676.951 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of SwitchDeadBranchElimination is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.676.962 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of MergePass is [4] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.676.984 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of InferShapePass is [286] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.676.994 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of ReplaceWithEmptyConstPass is [12] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.677.003 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of SplitShapeNPass is [0] micro second, call num is [6] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.677.012 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of DimensionComputePass is [19] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.677.021 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of ConstantFoldingPass is [22] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.677.029 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of InferValuePass is [8] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.678.864 [graph_manager.cc:1056][EVENT]77563 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.OptimizeOriginalGraphForQuantize is [2862] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.678.926 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of CondRemovePass is [5] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.678.945 [graph_prepare.cc:1982][EVENT]77563 PrepareDynShape:[GEPERFTRACE] The time cost of Prepare::ProcessBeforeInfershape is [46] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.256 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of AssertPass is [1] micro second, call num is [6] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.282 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of SwitchDeadBranchElimination is [3] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.292 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of MergePass is [0] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.302 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of InferShapePass is [161] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.310 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of ReplaceWithEmptyConstPass is [8] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.319 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of SplitShapeNPass is [1] micro second, call num is [6] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.327 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of DimensionComputePass is [5] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.335 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of ConstantFoldingPass is [7] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.344 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of InferValuePass is [3] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.365 [graph_prepare.cc:1983][EVENT]77563 PrepareDynShape:[GEPERFTRACE] The time cost of Prepare::FormatAndShapeProcess is [408] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.387 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of PreRun::MarkForceUnknownForCondPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.402 [graph_prepare.cc:1984][EVENT]77563 PrepareDynShape:[GEPERFTRACE] The time cost of Prepare::CtrlFlowPreProcess is [21] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.414 [graph_prepare.cc:1985][EVENT]77563 PrepareDynShape:[GEPERFTRACE] The time cost of Prepare::multibatch::GetDynamicOutputShape is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.449 [graph_prepare.cc:1986][EVENT]77563 PrepareDynShape:[GEPERFTRACE] The time cost of Prepare::InsertAippOpUtil::Instance().UpdateDataNodeByAipp is [15] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.462 [graph_prepare.cc:1987][EVENT]77563 PrepareDynShape:[GEPERFTRACE] The time cost of Prepare::SaveOriginalGraphToOmModel is [0] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.479 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of PrepareOptimize::ShapeOperateOpRemovePass is [4] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.491 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of PrepareOptimize::ReplaceTransShapePass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.503 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of PrepareOptimize::MarkAgnosticPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.568 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of EnterPass is [0] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.581 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of CondPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.590 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of PrintOpPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.598 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of NoUseReshapeRemovePass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.607 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of DropOutPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.615 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of AssertPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.623 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of TransposeRemovePass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.631 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of UnusedConstPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.640 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of StopGradientPass is [0] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.650 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of PreventGradientPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.659 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of PlaceholderWithDefaultPass is [0] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.669 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of SnapshotPass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.678 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of GuaranteeConstPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.686 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of VarIsInitializedOpPass is [4] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.694 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of ParallelConcatStartOpPass is [0] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.702 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of IdentityPass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.724 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of PrepareOptimize::PrunePass is [7] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.738 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of PrepareOptimize::HcclMemcpyPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.774 [graph_prepare.cc:1988][EVENT]77563 PrepareDynShape:[GEPERFTRACE] The time cost of Prepare::PrepareOptimize is [302] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.679.788 [graph_manager.cc:1065][EVENT]77563 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.preparer.PrepareDynShape is [889] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.692.287 [graph_manager.cc:1077][EVENT]77563 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.OptimizeOriginalGraph is [12479] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.692.351 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of PrepareRunningFormatRefiner::VariablePrepareOpPass is [5] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.692.393 [graph_manager.cc:1080][EVENT]77563 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.preparer.PrepareRunningFormatRefiner is [69] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.695.755 [graph_manager.cc:1081][EVENT]77563 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.OptimizeOriginalGraphJudgeInsert is [3346] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.695.794 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of SubexpressionMigrationPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.695.809 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of UnusedArgsCleanPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.695.820 [graph_manager.cc:1082][EVENT]77563 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::SubexpressionMigration is [33] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.695.847 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::MergeInputMemcpyPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.695.863 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::SwitchDataEdgesBypass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.695.875 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::ConstantFuseSamePass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.695.903 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::CSEBeforeFuseDataNodesWithCommonInputPass is [18] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.695.917 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::FuseDataNodesWithCommonInputPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.695.931 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::CommonSubexpressionEliminationPass is [4] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.695.943 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::PermutePass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.695.979 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::SameTransdataBreadthFusionPass is [26] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.695.996 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::VariableOpPass is [6] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.012 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::TransOpWithoutReshapeFusionPass is [6] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.055 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::TransOpBreadthFusionPass is [33] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.073 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::DataFlowPreparePass is [6] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.095 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::MergeUnknownShapeNPass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.105 [graph_manager.cc:2700][EVENT]77563 OptimizeStage1:[GEPERFTRACE] The time cost of GraphManager::OptimizeStage1_1 is [262] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.205 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of EnterPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.219 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of AddNPass is [0] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.229 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of SwitchDeadBranchElimination is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.237 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of SwitchLogicRemovePass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.246 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of MergePass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.254 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of CastRemovePass is [6] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.263 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of TransposeTransDataPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.271 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of ReshapeRemovePass is [4] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.279 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of TransOpSymmetryEliminationPass is [3] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.287 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of TransOpNearbyAllreduceFusionPass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.295 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of ReplaceWithEmptyConstPass is [6] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.304 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of DimensionComputePass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.312 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of ConstantFoldingPass is [7] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.320 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of DimensionAdjustPass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.328 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of UselessControlOutRemovePass is [3] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.337 [graph_manager.cc:2741][EVENT]77563 OptimizeStage1:[GEPERFTRACE] The time cost of GraphManager::OptimizeStage1_2 is [216] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.346 [graph_manager.cc:2752][EVENT]77563 OptimizeStage1:[GEPERFTRACE] The time cost of extern constant folding is [0] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.365 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::Migration is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.376 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::ArgsClean is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.393 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::PrunePass is [7] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.405 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::NextIterationPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.425 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::ControlTriggerPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.437 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::MergeToStreamMergePass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.457 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::SwitchToStreamSwitchPass is [10] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.470 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::AttachStreamLabelPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.481 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::MultiBatchPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.490 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::SubgraphMultiDimsPass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.502 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::IteratorOpPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.512 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::VariableRefUselessControlOutDeletePass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.527 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::ReshapeRecoveryPass is [7] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.538 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::RemoveSameConstPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.547 [graph_manager.cc:2810][EVENT]77563 OptimizeStage1:[GEPERFTRACE] The time cost of GraphManager::OptimizeStage1_3 is [185] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.571 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of IdentityPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.583 [graph_manager.cc:2821][EVENT]77563 OptimizeStage1:[GEPERFTRACE] The time cost of GraphPrepare::node_pass is [29] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.606 [graph_manager.cc:1087][EVENT]77563 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::OptimizeStage1 is [769] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.728 [graph_manager.cc:1088][EVENT]77563 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.OptimizeAfterStage1 is [110] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.761 [graph_manager.cc:1089][EVENT]77563 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::GraphUtilsEx::InferShapeInNeed is [14] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.779 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of PreRun::CtrlEdgeTransferPass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.792 [graph_manager.cc:1097][EVENT]77563 PreRunOptimizeOriginalGraph:PreRun:PreRunOptimizeOriginalGraph success. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.696.811 [graph_manager.cc:3325][EVENT]77563 OptimizeSubgraph:[GEPERFTRACE] The time cost of OptimizeSubgraph::StagePartition is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.697.199 [engine_place.cc:144][EVENT]77563 Run:The time cost of AIcoreEngine::CheckSupported is [269] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.697.226 [engine_place.cc:144][EVENT]77563 Run:The time cost of DNN_VM_GE_LOCAL_OP_STORE::CheckSupported is [7] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.697.236 [engine_place.cc:144][EVENT]77563 Run:The time cost of DNN_VM_RTS_OP_STORE::CheckSupported is [8] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.697.321 [graph_manager.cc:3351][EVENT]77563 OptimizeSubgraph:[GEPERFTRACE] The time cost of OptimizeSubgraph::GraphPartitionDynamicShape is [497] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.697.340 [graph_manager.cc:3364][EVENT]77563 OptimizeSubgraph:[GEPERFTRACE] The time cost of OptimizeSubgraph::SubgraphPartitionAndOptimization::CompositeEngine is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.697.400 [engine_partitioner.cc:1139][EVENT]77563 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionInitialize is [15] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.697.414 [engine_partitioner.cc:1142][EVENT]77563 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionMarkClusters is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.697.529 [engine_partitioner.cc:1148][EVENT]77563 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionSplitSubGraphs is [106] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.697.562 [engine_partitioner.cc:1155][EVENT]77563 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionSortSubGraphs is [19] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.697.609 [engine_partitioner.cc:1164][EVENT]77563 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionAddPartitionsToGraphNode is [35] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.697.635 [graph_manager.cc:3405][EVENT]77563 SubgraphPartitionAndOptimization:[GEPERFTRACE] The time cost of OptimizeSubgraph::Partition1 is [281] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.697.652 [graph_manager.cc:3412][EVENT]77563 SubgraphPartitionAndOptimization:[GEPERFTRACE] The time cost of OptimizeSubgraph::SetSubgraphPreProc is [5] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.708.936 [graph_manager.cc:3422][EVENT]77563 SubgraphPartitionAndOptimization:[GEPERFTRACE] The time cost of OptimizeSubgraph::SetSubGraph is [11271] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.708.974 [graph_manager.cc:3428][EVENT]77563 SubgraphPartitionAndOptimization:[GEPERFTRACE] The time cost of OptimizeSubgraph::SetSubgraphPostProc is [7] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.077 [graph_manager.cc:3467][EVENT]77563 SubgraphPartitionAndOptimization:[GEPERFTRACE] The time cost of OptimizeSubgraph::MergeSubGraph is [83] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.094 [graph_manager.cc:3377][EVENT]77563 OptimizeSubgraph:[GEPERFTRACE] The time cost of OptimizeSubgraph::SubgraphPartitionAndOptimization::AtomicEngine is [11742] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.109 [graph_manager.cc:1106][EVENT]77563 PreRunOptimizeSubGraph:[GEPERFTRACE] The time cost of GraphManager::OptimizeSubgraph is [12303] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.177 [graph_manager.cc:1115][EVENT]77563 PreRunOptimizeSubGraph:PreRun:PreRunOptimizeSubGraph success. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.197 [graph_manager.cc:1130][EVENT]77563 PreRunAfterOptimizeSubGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.ReplacePrecompiledNodeWithOmGraph is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.227 [graph_manager.cc:1131][EVENT]77563 PreRunAfterOptimizeSubGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.OptimizeWholeGraph is [18] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.253 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses::LinkGenMaskNodesPass is [9] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.269 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage2::HcclContinuousMemcpyPass is [4] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.279 [graph_manager.cc:2837][EVENT]77563 OptimizeStage2:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses is [37] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.354 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of ConstantFoldingPass is [11] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.367 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of ReshapeRemovePass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.377 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of CondRemovePass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.385 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of BitcastPass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.394 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of AssignRemovePass is [4] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.402 [base_pass.cc:339][EVENT]77563 Run:[GEPERFTRACE] The time cost of DimensionAdjustPass is [3] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.412 [graph_manager.cc:2864][EVENT]77563 OptimizeStage2:[GEPERFTRACE] The time cost of OptimizeStage2::MergedGraphNameToPasses is [108] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.422 [graph_manager.cc:2872][EVENT]77563 OptimizeStage2:[GEPERFTRACE] The time cost of OptimizeStage2::RemoveIsolatedConst is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.439 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage2::ControlAttrOptimize::MultiBatchPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.452 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses::RefIdentityDeleteOpPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.465 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses::VariableRefDeleteOpPass is [4] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.478 [compile_nodes_pass.cc:88][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage2::ControlAttrOptimize::CompileNodesPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.488 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage2::ControlAttrOptimize::CompileNodesPass is [14] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.497 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses::SwapSpacePass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.572 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses::InputOutputConnectionIdentifyPass is [67] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.600 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses::AtomicAddrCleanPass is [14] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.612 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses::EndOfSequenceAddControlPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.624 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage2::ControlAttrOptimize::SubgraphPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.636 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage2::ControlAttrOptimize::AttachStreamLabelPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.645 [graph_manager.cc:2927][EVENT]77563 OptimizeStage2:[GEPERFTRACE] The time cost of OptimizeStage2::ControlAttrOptimize is [208] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.661 [graph_manager.cc:2937][EVENT]77563 OptimizeStage2:[GEPERFTRACE] The time cost of ModelBuilder::AssignFunctionalLabels is [7] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.682 [graph_manager.cc:2943][EVENT]77563 OptimizeStage2:[GEPERFTRACE] The time cost of MemcpyAddrAsyncPass::Run. is [6] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.709.693 [graph_manager.cc:2950][EVENT]77563 OptimizeStage2:[GEPERFTRACE] The time cost of BufferPoolMemoryPass::Run. is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.243 [graph_manager.cc:2958][EVENT]77563 OptimizeStage2:[GEPERFTRACE] The time cost of ParallelGroupPass::Run. is [37] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.289 [graph_manager.cc:1132][EVENT]77563 PreRunAfterOptimizeSubGraph:[GEPERFTRACE] The time cost of GraphManager::OptimizeStage2 is [10048] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.371 [graph_manager.cc:1135][EVENT]77563 PreRunAfterOptimizeSubGraph:[GEPERFTRACE] The time cost of GraphManager::GetCompilerStages(graph_node->GetGraphId()).optimizer.OptimizeGraphBeforeBuild is [66] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.415 [graph_manager.cc:2975][EVENT]77563 MemConflictProc:[GEPERFTRACE] The time cost of HandleMemoryRWConflict is [26] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.453 [graph_manager.cc:2981][EVENT]77563 MemConflictProc:[GEPERFTRACE] The time cost of MemLayoutConflictOptimizer::Run. is [25] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.468 [pass_manager.cc:82][EVENT]77563 Run:[GEPERFTRACE] The time cost of OptimizeStage2::SetFftsPlusAttrPass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.479 [graph_manager.cc:2986][EVENT]77563 MemConflictProc:[GEPERFTRACE] The time cost of SetFftsPlusAttrPass::last_passes.Run is [15] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.488 [graph_manager.cc:1136][EVENT]77563 PreRunAfterOptimizeSubGraph:[GEPERFTRACE] The time cost of GraphManager::MemConflictProc is [101] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.625 [graph_manager.cc:3555][EVENT]77563 Build:[GEPERFTRACE] The time cost of GraphManager::RecoverIrDefinitionAndModifyAippData is [106] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.717 [engine_partitioner.cc:1139][EVENT]77563 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionInitialize is [16] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.734 [engine_partitioner.cc:1142][EVENT]77563 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionMarkClusters is [4] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.843 [engine_partitioner.cc:1148][EVENT]77563 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionSplitSubGraphs is [99] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.875 [engine_partitioner.cc:1155][EVENT]77563 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionSortSubGraphs is [19] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.916 [engine_partitioner.cc:1164][EVENT]77563 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionAddPartitionsToGraphNode is [30] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.719.939 [graph_builder.cc:865][EVENT]77563 SecondPartition:[GEPERFTRACE] The time cost of EnginePartitioner::Partition2 is [257] micro second. [INFO] RUNTIME(73092,python3.7):2024-01-11-05:37:50.720.370 [logger.cc:1071] 77563 ModelBindStream: model_id=832, stream_id=1089, flag=0. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.720.413 [task_generator.cc:804][EVENT]77563 GenerateTask:[GEPERFTRACE] The time cost of TaskGenerator::SetStreamCtx is [170] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.720.474 [task_generator.cc:805][EVENT]77563 GenerateTask:[GEPERFTRACE] The time cost of TaskGenerator::PrepareForGenerateTask is [47] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.721.290 [task_generator.cc:814][EVENT]77563 GenerateTask:[GEPERFTRACE] The time cost of TaskGenerator::DoGenerateTask is [801] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.721.315 [task_generator.cc:954][EVENT]77563 GetTaskInfo:[GEPERFTRACE] The time cost of TaskGenerator::GenerateTask is [1072] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.721.361 [task_generator.cc:967][EVENT]77563 GetTaskInfo:[GEPERFTRACE] The time cost of TaskGenerator::AddModelTaskToModel is [23] micro second. [INFO] RUNTIME(73092,python3.7):2024-01-11-05:37:50.721.380 [logger.cc:1084] 77563 ModelUnbindStream: model_id=832, stream_id=1089, [INFO] GE(73092,python3.7):2024-01-11-05:37:50.721.530 [graph_manager.cc:1152][EVENT]77563 PreRunAfterOptimizeSubGraph:[GEPERFTRACE] The time cost of GraphManager::Build is [2019] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.721.549 [graph_manager.cc:1164][EVENT]77563 PreRunAfterOptimizeSubGraph:PreRun:PreRunAfterOptimizeSubGraph success. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.721.578 [graph_manager.cc:1271][EVENT]77563 PreRun:[GEPERFTRACE] The time cost of FlowModelBuild is [54139] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.721.589 [graph_manager.cc:1272][EVENT]77563 PreRun:[GEPERFTRACE] GE PreRun End [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:50.721.896 [atrace_api.c:93](tid:77563) AtraceDestroy start [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:50.721.919 [atrace_api.c:95](tid:77563) AtraceDestroy end [INFO] GE(73092,python3.7):2024-01-11-05:37:50.726.578 [graph_converter.cc:838][EVENT]77563 ConvertComputeGraphToExecuteGraph:[GEPERFTRACE] The time cost of ConvertComputeGraphToExecuteGraph::CreateMainNode is [1304] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.726.719 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of ZeroCopy is [99] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.727.169 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of CEM is [427] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.727.352 [copy_flow_launch_fuse.cc:395][EVENT]77563 Run:[GEPERFTRACE] The time cost of Pass::CopyFlowLaunchFuse is [159] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.727.373 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of CopyFlowLaunch is [182] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.727.575 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of TrustOutTensor is [190] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.727.601 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of AicpuFuseHostInputs is [8] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.727.634 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of ZeroCopy is [21] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.727.808 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of CEM is [160] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.727.886 [copy_flow_launch_fuse.cc:395][EVENT]77563 Run:[GEPERFTRACE] The time cost of Pass::CopyFlowLaunchFuse is [60] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.727.899 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of CopyFlowLaunch is [73] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.727.926 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of TrustOutTensor is [18] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.727.937 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of AicpuFuseHostInputs is [0] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.727.961 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of ZeroCopy is [16] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.728.027 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of CEM is [55] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.728.089 [copy_flow_launch_fuse.cc:395][EVENT]77563 Run:[GEPERFTRACE] The time cost of Pass::CopyFlowLaunchFuse is [52] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.728.110 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of CopyFlowLaunch is [73] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.728.136 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of TrustOutTensor is [16] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.728.145 [base_optimizer.cc:70][EVENT]77563 Run:[GEPERFTRACE] The time cost of AicpuFuseHostInputs is [0] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.728.157 [graph_converter.cc:849][EVENT]77563 ConvertComputeGraphToExecuteGraph:[GEPERFTRACE] The time cost of ConvertComputeGraphToExecuteGraph::RunAllPass is [1540] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.728.358 [graph_converter.cc:853][EVENT]77563 ConvertComputeGraphToExecuteGraph:[GEPERFTRACE] The time cost of ConvertComputeGraphToExecuteGraph::TopologicalSorting is [193] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.729.014 [graph_converter.cc:857][EVENT]77563 ConvertComputeGraphToExecuteGraph:[GEPERFTRACE] The time cost of ConvertComputeGraphToExecuteGraph::AppendGraphLevelData is [643] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:50.729.165 [graph_converter.cc:862][EVENT]77563 ConvertComputeGraphToExecuteGraph:[GEPERFTRACE] The time cost of ConvertComputeGraphToExecuteGraph::CalculatePriority is [127] micro second. TotalTime = 0.0615739, [20] [parse]: 0.00136392 [symbol_resolve]: 0.0301847, [1] [Cycle 1]: 0.0301112, [1] [resolve]: 0.0300902 [combine_like_graphs]: 8.39995e-07 [graph_reusing]: 3.25e-06 [meta_unpack_prepare]: 0.00012045 [pre_cconv]: 8.49999e-07 [abstract_specialize]: 0.0151275 [pack_expand]: 1.864e-05 [auto_monad]: 0.00017675 [inline]: 1.73e-06 [pre_auto_parallel]: 1.181e-05 [pipeline_split]: 2.79e-06 [optimize]: 0.00907042, [35] [py_interpret_to_execute]: 4.27999e-06 [rewriter_before_opt_a]: 0.00023056 [opt_a]: 0.00837353, [3] [Cycle 1]: 0.00457694, [30] [expand_dump_flag]: 4.94e-06 [switch_simplify]: 9.443e-05 [a_1]: 0.00056238 [recompute_prepare]: 8.66e-06 [updatestate_depend_eliminate]: 1.03e-05 [updatestate_assign_eliminate]: 7.49e-06 [updatestate_loads_eliminate]: 6.55e-06 [parameter_eliminate]: 4.6e-06 [a_2]: 9.775e-05 [accelerated_algorithm]: 6.46e-06 [pynative_shard]: 1.75e-06 [auto_parallel]: 3.67e-06 [parallel]: 8.35e-06 [merge_comm]: 8.02999e-06 [allreduce_fusion]: 2.97e-06 [virtual_dataset]: 5.65e-06 [get_grad_eliminate_]: 4.96e-06 [virtual_output]: 4.55001e-06 [merge_forward]: 1.003e-05 [cell_reuse_recompute_pass]: 6.69999e-07 [cell_reuse_handle_not_recompute_node_pass]: 1.303e-05 [meta_fg_expand]: 0.00083726 [after_resolve]: 2.614e-05 [a_after_grad]: 3.673e-05 [renormalize]: 0.00224503 [real_op_eliminate]: 4.583e-05 [auto_monad_grad]: 1.37e-05 [auto_monad_eliminator]: 4.087e-05 [cse]: 0.0001516 [a_3]: 0.00013389 [Cycle 2]: 0.00120305, [30] [expand_dump_flag]: 1.65001e-06 [switch_simplify]: 1.547e-05 [a_1]: 0.00052703 [recompute_prepare]: 3.11e-06 [updatestate_depend_eliminate]: 4.66e-06 [updatestate_assign_eliminate]: 2.73e-06 [updatestate_loads_eliminate]: 2.39e-06 [parameter_eliminate]: 2.43e-06 [a_2]: 3.467e-05 [accelerated_algorithm]: 2.93e-06 [pynative_shard]: 1.50999e-06 [auto_parallel]: 3.29e-06 [parallel]: 3.50999e-06 [merge_comm]: 2.58e-06 [allreduce_fusion]: 1.72e-06 [virtual_dataset]: 2.77e-06 [get_grad_eliminate_]: 2.27e-06 [virtual_output]: 2.12e-06 [merge_forward]: 3.41e-06 [cell_reuse_recompute_pass]: 3.69997e-07 [cell_reuse_handle_not_recompute_node_pass]: 5.45e-06 [meta_fg_expand]: 1.334e-05 [after_resolve]: 2.34001e-06 [a_after_grad]: 2.86e-06 [renormalize]: 0.00037259 [real_op_eliminate]: 5.89e-06 [auto_monad_grad]: 3.44e-06 [auto_monad_eliminator]: 7.83e-06 [cse]: 1.899e-05 [a_3]: 1.858e-05 [Cycle 3]: 0.00023901, [30] [expand_dump_flag]: 1.15e-06 [switch_simplify]: 2.64e-06 [a_1]: 1.732e-05 [recompute_prepare]: 2.16e-06 [updatestate_depend_eliminate]: 3.5e-06 [updatestate_assign_eliminate]: 2.56e-06 [updatestate_loads_eliminate]: 2.26e-06 [parameter_eliminate]: 1.01e-06 [a_2]: 3.183e-05 [accelerated_algorithm]: 2.65001e-06 [pynative_shard]: 1.14e-06 [auto_parallel]: 3.53e-06 [parallel]: 3.41e-06 [merge_comm]: 1.95e-06 [allreduce_fusion]: 1.21e-06 [virtual_dataset]: 2.45e-06 [get_grad_eliminate_]: 2.4e-06 [virtual_output]: 1.99e-06 [merge_forward]: 2.9e-06 [cell_reuse_recompute_pass]: 3.40005e-07 [cell_reuse_handle_not_recompute_node_pass]: 6.65e-06 [meta_fg_expand]: 2.09e-06 [after_resolve]: 1.98e-06 [a_after_grad]: 2.48e-06 [renormalize]: 7.0002e-08 [real_op_eliminate]: 2.3e-06 [auto_monad_grad]: 9.09997e-07 [auto_monad_eliminator]: 4.9e-06 [cse]: 1.188e-05 [a_3]: 1.561e-05 [py_interpret_to_execute_after_opt_a]: 3.14e-06 [slice_cell_reuse_recomputed_activation]: 2.59e-06 [rewriter_after_opt_a]: 2.184e-05 [convert_after_rewriter]: 5.71e-06 [order_py_execute_after_rewriter]: 4.53e-06 [opt_b]: 0.00010139, [1] [Cycle 1]: 9.653e-05, [7] [b_1]: 4.59e-05 [b_2]: 3.24e-06 [updatestate_depend_eliminate]: 2.82e-06 [updatestate_assign_eliminate]: 4.99e-06 [updatestate_loads_eliminate]: 2.13e-06 [renormalize]: 4.1e-07 [cse]: 1.058e-05 [cconv]: 1.923e-05 [opt_after_cconv]: 5.288e-05, [1] [Cycle 1]: 4.877e-05, [7] [c_1]: 5.48e-06 [parameter_eliminate]: 7.09995e-07 [updatestate_depend_eliminate]: 2.68e-06 [updatestate_assign_eliminate]: 2.23e-06 [updatestate_loads_eliminate]: 1.95e-06 [cse]: 9.24e-06 [renormalize]: 2.09999e-07 [remove_dup_value]: 1.407e-05 [tuple_transform]: 3.642e-05, [1] [Cycle 1]: 3.304e-05, [3] [d_1]: 1.52e-05 [d_2]: 6.72e-06 [renormalize]: 1.50001e-07 [add_cache_embedding]: 1.085e-05 [add_recomputation]: 3.232e-05 [cse_after_recomputation]: 1.823e-05, [1] [Cycle 1]: 1.432e-05, [1] [cse]: 1.014e-05 [environ_conv]: 7.27e-06 [label_micro_interleaved_index]: 2.32e-06 [label_fine_grained_interleaved_index]: 2.27e-06 [assign_add_opt]: 2.02e-06 [slice_recompute_activation]: 1.99e-06 [micro_interleaved_order_control]: 1.52e-06 [full_micro_interleaved_order_control]: 1.81e-06 [comp_comm_scheduling]: 1.93e-06 [reorder_send_recv_between_fp_bp]: 2.09e-06 [comm_op_add_attrs]: 9.59997e-07 [add_comm_op_reuse_tag]: 8.70001e-07 [overlap_opt_shard_in_pipeline]: 1.07e-06 [grouped_pairwise_exchange_alltoall]: 1.29e-06 [overlap_recompute_and_grad_model_parallel]: 1.55999e-06 [overlap_grad_matmul_and_grad_allreduce]: 7.00005e-07 [split_matmul_comm_elemetwise]: 2.28e-06 [split_layernorm_comm]: 1.62e-06 [process_send_recv_for_ge]: 1.1e-06 [handle_group_info]: 9.09997e-07 [auto_monad_reorder]: 1.648e-05 [get_jit_bprop_graph]: 3.69997e-07 [eliminate_special_op_node]: 0.00047857 [validate]: 2.871e-05 [distribtued_split]: 1.22e-06 [task_emit]: 0.00475382 [execute]: 7.22e-06 Sums parse : 0.001364s : 2.34% symbol_resolve.resolve : 0.030090s : 51.59% combine_like_graphs : 0.000001s : 0.00% graph_reusing : 0.000003s : 0.01% meta_unpack_prepare : 0.000120s : 0.21% pre_cconv : 0.000001s : 0.00% abstract_specialize : 0.015128s : 25.94% pack_expand : 0.000019s : 0.03% auto_monad : 0.000177s : 0.30% inline : 0.000002s : 0.00% pre_auto_parallel : 0.000012s : 0.02% pipeline_split : 0.000003s : 0.00% optimize.py_interpret_to_execute : 0.000004s : 0.01% optimize.rewriter_before_opt_a : 0.000231s : 0.40% optimize.opt_a.expand_dump_flag : 0.000008s : 0.01% optimize.opt_a.switch_simplify : 0.000113s : 0.19% optimize.opt_a.a_1 : 0.001107s : 1.90% optimize.opt_a.recompute_prepare : 0.000014s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000018s : 0.03% optimize.opt_a.updatestate_assign_eliminate : 0.000013s : 0.02% optimize.opt_a.updatestate_loads_eliminate : 0.000011s : 0.02% optimize.opt_a.parameter_eliminate : 0.000008s : 0.01% optimize.opt_a.a_2 : 0.000164s : 0.28% optimize.opt_a.accelerated_algorithm : 0.000012s : 0.02% optimize.opt_a.pynative_shard : 0.000004s : 0.01% optimize.opt_a.auto_parallel : 0.000010s : 0.02% optimize.opt_a.parallel : 0.000015s : 0.03% optimize.opt_a.merge_comm : 0.000013s : 0.02% optimize.opt_a.allreduce_fusion : 0.000006s : 0.01% optimize.opt_a.virtual_dataset : 0.000011s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000010s : 0.02% optimize.opt_a.virtual_output : 0.000009s : 0.01% optimize.opt_a.merge_forward : 0.000016s : 0.03% optimize.opt_a.cell_reuse_recompute_pass : 0.000001s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000025s : 0.04% optimize.opt_a.meta_fg_expand : 0.000853s : 1.46% optimize.opt_a.after_resolve : 0.000030s : 0.05% optimize.opt_a.a_after_grad : 0.000042s : 0.07% optimize.opt_a.renormalize : 0.002618s : 4.49% optimize.opt_a.real_op_eliminate : 0.000054s : 0.09% optimize.opt_a.auto_monad_grad : 0.000018s : 0.03% optimize.opt_a.auto_monad_eliminator : 0.000054s : 0.09% optimize.opt_a.cse : 0.000182s : 0.31% optimize.opt_a.a_3 : 0.000168s : 0.29% optimize.py_interpret_to_execute_after_opt_a : 0.000003s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000022s : 0.04% optimize.convert_after_rewriter : 0.000006s : 0.01% optimize.order_py_execute_after_rewriter : 0.000005s : 0.01% optimize.opt_b.b_1 : 0.000046s : 0.08% optimize.opt_b.b_2 : 0.000003s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000003s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000005s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000002s : 0.00% optimize.opt_b.renormalize : 0.000000s : 0.00% optimize.opt_b.cse : 0.000011s : 0.02% optimize.cconv : 0.000019s : 0.03% optimize.opt_after_cconv.c_1 : 0.000005s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000001s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000003s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000002s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000002s : 0.00% optimize.opt_after_cconv.cse : 0.000009s : 0.02% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000014s : 0.02% optimize.tuple_transform.d_1 : 0.000015s : 0.03% optimize.tuple_transform.d_2 : 0.000007s : 0.01% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.add_cache_embedding : 0.000011s : 0.02% optimize.add_recomputation : 0.000032s : 0.06% optimize.cse_after_recomputation.cse : 0.000010s : 0.02% optimize.environ_conv : 0.000007s : 0.01% optimize.label_micro_interleaved_index : 0.000002s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000002s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.comp_comm_scheduling : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000002s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000001s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000002s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.process_send_recv_for_ge : 0.000001s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% auto_monad_reorder : 0.000016s : 0.03% get_jit_bprop_graph : 0.000000s : 0.00% eliminate_special_op_node : 0.000479s : 0.82% validate : 0.000029s : 0.05% distribtued_split : 0.000001s : 0.00% task_emit : 0.004754s : 8.15% execute : 0.000007s : 0.01% Time group info: ------[substitution.] 0.030064 222 0.01% : 0.000004s : 10: substitution.float_depend_g_call 0.01% : 0.000003s : 2: substitution.float_tuple_getitem_switch 98.09% : 0.029489s : 23: substitution.getattr_setattr_resolve 0.02% : 0.000005s : 4: substitution.graph_param_transform 0.01% : 0.000002s : 2: substitution.incorporate_call 0.00% : 0.000001s : 2: substitution.incorporate_call_switch 1.25% : 0.000376s : 21: substitution.inline 0.10% : 0.000030s : 60: substitution.meta_unpack_prepare 0.02% : 0.000007s : 7: substitution.minmaximum_grad 0.07% : 0.000020s : 10: substitution.partial_eliminate 0.00% : 0.000001s : 4: substitution.partial_unused_args_eliminate 0.02% : 0.000006s : 1: substitution.real_op_eliminate 0.01% : 0.000003s : 12: substitution.remove_not_recompute_node 0.04% : 0.000013s : 9: substitution.replace_applicator 0.01% : 0.000002s : 7: substitution.replace_old_param 0.00% : 0.000001s : 1: substitution.set_cell_output_no_recompute 0.02% : 0.000006s : 3: substitution.switch_simplify 0.09% : 0.000027s : 7: substitution.tuple_list_convert_item_index_to_positive 0.03% : 0.000008s : 7: substitution.tuple_list_get_item_const_eliminator 0.04% : 0.000012s : 7: substitution.tuple_list_get_item_depend_reorder 0.12% : 0.000036s : 16: substitution.tuple_list_get_item_eliminator 0.04% : 0.000012s : 7: substitution.tuple_list_get_set_item_eliminator ------[renormalize.] 0.002609 4 57.58% : 0.001502s : 2: renormalize.infer 42.42% : 0.001107s : 2: renormalize.specialize ------[replace.] 0.000624 50 51.21% : 0.000319s : 20: replace.getattr_setattr_resolve 31.35% : 0.000195s : 19: replace.inline 1.89% : 0.000012s : 1: replace.real_op_eliminate 6.28% : 0.000039s : 3: replace.switch_simplify 9.26% : 0.000058s : 7: replace.tuple_list_get_item_eliminator ------[match.] 0.029741 50 98.68% : 0.029349s : 20: match.getattr_setattr_resolve 1.21% : 0.000359s : 19: match.inline 0.02% : 0.000006s : 1: match.real_op_eliminate 0.02% : 0.000006s : 3: match.switch_simplify 0.07% : 0.000021s : 7: match.tuple_list_get_item_eliminator ------[func_graph_cloner_run.] 0.002228 39 73.09% : 0.001628s : 16: func_graph_cloner_run.FuncGraphClonerGraph 26.91% : 0.000599s : 23: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.031960 141 1.02% : 0.000325s : 78: opt.transform.opt_a 0.12% : 0.000038s : 23: opt.transform.opt_b 94.13% : 0.030083s : 2: opt.transform.opt_resolve 0.31% : 0.000099s : 1: opt.transforms.meta_unpack_prepare 4.31% : 0.001378s : 30: opt.transforms.opt_a 0.01% : 0.000004s : 1: opt.transforms.opt_after_cconv 0.01% : 0.000002s : 1: opt.transforms.opt_b 0.06% : 0.000020s : 2: opt.transforms.opt_trans_graph 0.03% : 0.000010s : 3: opt.transforms.special_op_eliminate [INFO] GE(73092,python3.7):2024-01-11-05:37:51.080.719 [graph_var_manager.cc:1424][EVENT]77564 SetAllMemoryMaxValue:The graph_mem_max_size is 27917287424 and the var_mem_max_size is 5368709120 [INFO] GE(73092,python3.7):2024-01-11-05:37:51.080.815 [graph_manager.cc:1248][EVENT]77564 PreRun:PreRun start: graph node size 3, session id 2, graph id 1, graph name online. [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:51.081.742 [atrace_api.c:28](tid:77564) AtraceCreate start [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:51.081.819 [trace_rb_log.c:84](tid:77564) [RUNTIME_ATRACE_DEV64_TS0] create ring buffer success, buffer size : 131152. [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:51.081.833 [atrace_api.c:32](tid:77564) AtraceCreate end [INFO] TDT(73092,python3.7):2024-01-11-05:37:51.081.845 [client_manager.cpp:157][SetProfilingCallback][tid:77564] [TsdClient] set profiling callback success [INFO] GE(73092,python3.7):2024-01-11-05:37:51.082.653 [parallel_partitioner.cc:165][EVENT]77564 DoPipelinePartition:[GEPERFTRACE] The time cost of OptimizeSubgraph::PipelinePartition is [14] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.082.691 [parallel_partitioner.cc:178][EVENT]77564 DoFlowGraphPartition:[GEPERFTRACE] The time cost of OptimizeSubgraph::FlowGraphPartition is [10] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.082.737 [graph_prepare.cc:1378][EVENT]77564 Init:[GEPERFTRACE] The time cost of FileConstantUtils::ConvertFileConstToConst is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.083.368 [graph_manager.cc:1050][EVENT]77564 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.preparer.PrepareInit is [651] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.083.396 [graph_manager.cc:1052][EVENT]77564 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.HandleSummaryOp is [5] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.083.495 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of ForToWhilePass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.083.521 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of ProcessNetOutput::SavePass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.083.578 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of ProcessNetOutput::NetOutputPass is [25] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.083.590 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of ProcessNetOutput::DataPass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.083.629 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of CreateSubGraphWithScopePass is [8] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.083.642 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of SubgraphMultiDimsClonePass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.083.659 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of MultiBatchClonePass is [5] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.083.747 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of SplitVariableIntoSubgraphPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.083.766 [graph_manager.cc:1054][EVENT]77564 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.preparer.NormalizeGraph is [355] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.083.935 [graph_manager.cc:1055][EVENT]77564 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.OptimizeGraphInit is [156] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.084.691 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of AssertPass is [1] micro second, call num is [6] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.084.719 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of SwitchDeadBranchElimination is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.084.730 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of MergePass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.084.740 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of InferShapePass is [218] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.084.749 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of ReplaceWithEmptyConstPass is [9] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.084.757 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of SplitShapeNPass is [1] micro second, call num is [6] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.084.765 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of DimensionComputePass is [6] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.084.774 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of ConstantFoldingPass is [12] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.084.782 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of InferValuePass is [5] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.086.560 [graph_manager.cc:1056][EVENT]77564 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.OptimizeOriginalGraphForQuantize is [2605] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.086.619 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of CondRemovePass is [4] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.086.637 [graph_prepare.cc:1982][EVENT]77564 PrepareDynShape:[GEPERFTRACE] The time cost of Prepare::ProcessBeforeInfershape is [42] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.086.943 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of AssertPass is [1] micro second, call num is [6] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.086.967 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of SwitchDeadBranchElimination is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.086.978 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of MergePass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.001 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of InferShapePass is [158] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.011 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of ReplaceWithEmptyConstPass is [7] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.020 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of SplitShapeNPass is [1] micro second, call num is [6] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.028 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of DimensionComputePass is [4] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.036 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of ConstantFoldingPass is [8] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.044 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of InferValuePass is [4] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.066 [graph_prepare.cc:1983][EVENT]77564 PrepareDynShape:[GEPERFTRACE] The time cost of Prepare::FormatAndShapeProcess is [417] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.086 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of PreRun::MarkForceUnknownForCondPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.097 [graph_prepare.cc:1984][EVENT]77564 PrepareDynShape:[GEPERFTRACE] The time cost of Prepare::CtrlFlowPreProcess is [17] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.109 [graph_prepare.cc:1985][EVENT]77564 PrepareDynShape:[GEPERFTRACE] The time cost of Prepare::multibatch::GetDynamicOutputShape is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.127 [graph_prepare.cc:1986][EVENT]77564 PrepareDynShape:[GEPERFTRACE] The time cost of Prepare::InsertAippOpUtil::Instance().UpdateDataNodeByAipp is [8] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.139 [graph_prepare.cc:1987][EVENT]77564 PrepareDynShape:[GEPERFTRACE] The time cost of Prepare::SaveOriginalGraphToOmModel is [0] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.153 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of PrepareOptimize::ShapeOperateOpRemovePass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.164 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of PrepareOptimize::ReplaceTransShapePass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.176 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of PrepareOptimize::MarkAgnosticPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.240 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of EnterPass is [0] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.252 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of CondPass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.261 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of PrintOpPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.269 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of NoUseReshapeRemovePass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.277 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of DropOutPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.285 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of AssertPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.294 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of TransposeRemovePass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.313 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of UnusedConstPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.322 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of StopGradientPass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.330 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of PreventGradientPass is [0] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.338 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of PlaceholderWithDefaultPass is [0] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.346 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of SnapshotPass is [3] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.355 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of GuaranteeConstPass is [0] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.363 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of VarIsInitializedOpPass is [4] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.371 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of ParallelConcatStartOpPass is [0] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.379 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of IdentityPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.398 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of PrepareOptimize::PrunePass is [7] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.409 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of PrepareOptimize::HcclMemcpyPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.435 [graph_prepare.cc:1988][EVENT]77564 PrepareDynShape:[GEPERFTRACE] The time cost of Prepare::PrepareOptimize is [286] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.087.448 [graph_manager.cc:1065][EVENT]77564 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.preparer.PrepareDynShape is [853] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.099.279 [graph_manager.cc:1077][EVENT]77564 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.OptimizeOriginalGraph is [11811] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.099.347 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of PrepareRunningFormatRefiner::VariablePrepareOpPass is [4] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.099.404 [graph_manager.cc:1080][EVENT]77564 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.preparer.PrepareRunningFormatRefiner is [82] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.568 [graph_manager.cc:1081][EVENT]77564 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.OptimizeOriginalGraphJudgeInsert is [3147] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.609 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of SubexpressionMigrationPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.624 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of UnusedArgsCleanPass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.635 [graph_manager.cc:1082][EVENT]77564 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::SubexpressionMigration is [32] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.662 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::MergeInputMemcpyPass is [4] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.678 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::SwitchDataEdgesBypass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.703 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::ConstantFuseSamePass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.726 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::CSEBeforeFuseDataNodesWithCommonInputPass is [14] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.739 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::FuseDataNodesWithCommonInputPass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.753 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::CommonSubexpressionEliminationPass is [4] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.766 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::PermutePass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.796 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::SameTransdataBreadthFusionPass is [21] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.814 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::VariableOpPass is [6] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.830 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::TransOpWithoutReshapeFusionPass is [6] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.851 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::TransOpBreadthFusionPass is [11] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.864 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::DataFlowPreparePass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.876 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_1::MergeUnknownShapeNPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.885 [graph_manager.cc:2700][EVENT]77564 OptimizeStage1:[GEPERFTRACE] The time cost of GraphManager::OptimizeStage1_1 is [228] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.978 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of EnterPass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.102.991 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of AddNPass is [0] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.001 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of SwitchDeadBranchElimination is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.010 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of SwitchLogicRemovePass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.018 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of MergePass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.027 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of CastRemovePass is [7] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.035 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of TransposeTransDataPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.043 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of ReshapeRemovePass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.052 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of TransOpSymmetryEliminationPass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.060 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of TransOpNearbyAllreduceFusionPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.076 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of ReplaceWithEmptyConstPass is [6] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.085 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of DimensionComputePass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.093 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of ConstantFoldingPass is [9] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.101 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of DimensionAdjustPass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.110 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of UselessControlOutRemovePass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.119 [graph_manager.cc:2741][EVENT]77564 OptimizeStage1:[GEPERFTRACE] The time cost of GraphManager::OptimizeStage1_2 is [218] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.128 [graph_manager.cc:2752][EVENT]77564 OptimizeStage1:[GEPERFTRACE] The time cost of extern constant folding is [0] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.147 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::Migration is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.160 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::ArgsClean is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.176 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::PrunePass is [7] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.191 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::NextIterationPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.202 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::ControlTriggerPass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.213 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::MergeToStreamMergePass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.228 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::SwitchToStreamSwitchPass is [6] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.240 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::AttachStreamLabelPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.253 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::MultiBatchPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.263 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::SubgraphMultiDimsPass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.275 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::IteratorOpPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.286 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::VariableRefUselessControlOutDeletePass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.301 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::ReshapeRecoveryPass is [6] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.312 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage1_3::RemoveSameConstPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.322 [graph_manager.cc:2810][EVENT]77564 OptimizeStage1:[GEPERFTRACE] The time cost of GraphManager::OptimizeStage1_3 is [178] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.346 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of IdentityPass is [2] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.365 [graph_manager.cc:2821][EVENT]77564 OptimizeStage1:[GEPERFTRACE] The time cost of GraphPrepare::node_pass is [35] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.387 [graph_manager.cc:1087][EVENT]77564 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::OptimizeStage1 is [736] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.507 [graph_manager.cc:1088][EVENT]77564 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.OptimizeAfterStage1 is [108] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.540 [graph_manager.cc:1089][EVENT]77564 PreRunOptimizeOriginalGraph:[GEPERFTRACE] The time cost of GraphManager::GraphUtilsEx::InferShapeInNeed is [14] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.556 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of PreRun::CtrlEdgeTransferPass is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.570 [graph_manager.cc:1097][EVENT]77564 PreRunOptimizeOriginalGraph:PreRun:PreRunOptimizeOriginalGraph success. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.589 [graph_manager.cc:3325][EVENT]77564 OptimizeSubgraph:[GEPERFTRACE] The time cost of OptimizeSubgraph::StagePartition is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.908 [engine_place.cc:144][EVENT]77564 Run:The time cost of AIcoreEngine::CheckSupported is [238] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.933 [engine_place.cc:144][EVENT]77564 Run:The time cost of DNN_VM_GE_LOCAL_OP_STORE::CheckSupported is [9] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.943 [engine_place.cc:144][EVENT]77564 Run:The time cost of DNN_VM_RTS_OP_STORE::CheckSupported is [8] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.103.998 [graph_manager.cc:3351][EVENT]77564 OptimizeSubgraph:[GEPERFTRACE] The time cost of OptimizeSubgraph::GraphPartitionDynamicShape is [396] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.104.015 [graph_manager.cc:3364][EVENT]77564 OptimizeSubgraph:[GEPERFTRACE] The time cost of OptimizeSubgraph::SubgraphPartitionAndOptimization::CompositeEngine is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.104.067 [engine_partitioner.cc:1139][EVENT]77564 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionInitialize is [12] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.104.082 [engine_partitioner.cc:1142][EVENT]77564 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionMarkClusters is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.104.197 [engine_partitioner.cc:1148][EVENT]77564 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionSplitSubGraphs is [105] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.104.229 [engine_partitioner.cc:1155][EVENT]77564 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionSortSubGraphs is [19] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.104.266 [engine_partitioner.cc:1164][EVENT]77564 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionAddPartitionsToGraphNode is [27] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.104.293 [graph_manager.cc:3405][EVENT]77564 SubgraphPartitionAndOptimization:[GEPERFTRACE] The time cost of OptimizeSubgraph::Partition1 is [266] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.104.310 [graph_manager.cc:3412][EVENT]77564 SubgraphPartitionAndOptimization:[GEPERFTRACE] The time cost of OptimizeSubgraph::SetSubgraphPreProc is [5] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.577 [graph_manager.cc:3422][EVENT]77564 SubgraphPartitionAndOptimization:[GEPERFTRACE] The time cost of OptimizeSubgraph::SetSubGraph is [7255] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.613 [graph_manager.cc:3428][EVENT]77564 SubgraphPartitionAndOptimization:[GEPERFTRACE] The time cost of OptimizeSubgraph::SetSubgraphPostProc is [8] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.730 [graph_manager.cc:3467][EVENT]77564 SubgraphPartitionAndOptimization:[GEPERFTRACE] The time cost of OptimizeSubgraph::MergeSubGraph is [90] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.747 [graph_manager.cc:3377][EVENT]77564 OptimizeSubgraph:[GEPERFTRACE] The time cost of OptimizeSubgraph::SubgraphPartitionAndOptimization::AtomicEngine is [7721] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.762 [graph_manager.cc:1106][EVENT]77564 PreRunOptimizeSubGraph:[GEPERFTRACE] The time cost of GraphManager::OptimizeSubgraph is [8178] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.773 [graph_manager.cc:1115][EVENT]77564 PreRunOptimizeSubGraph:PreRun:PreRunOptimizeSubGraph success. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.792 [graph_manager.cc:1130][EVENT]77564 PreRunAfterOptimizeSubGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.ReplacePrecompiledNodeWithOmGraph is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.821 [graph_manager.cc:1131][EVENT]77564 PreRunAfterOptimizeSubGraph:[GEPERFTRACE] The time cost of GraphManager::stages.optimizer.OptimizeWholeGraph is [17] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.844 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses::LinkGenMaskNodesPass is [5] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.860 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage2::HcclContinuousMemcpyPass is [4] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.870 [graph_manager.cc:2837][EVENT]77564 OptimizeStage2:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses is [34] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.940 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of ConstantFoldingPass is [13] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.952 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of ReshapeRemovePass is [3] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.961 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of CondRemovePass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.969 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of BitcastPass is [1] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.978 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of AssignRemovePass is [4] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.986 [base_pass.cc:339][EVENT]77564 Run:[GEPERFTRACE] The time cost of DimensionAdjustPass is [3] micro second, call num is [3] [INFO] GE(73092,python3.7):2024-01-11-05:37:51.111.996 [graph_manager.cc:2864][EVENT]77564 OptimizeStage2:[GEPERFTRACE] The time cost of OptimizeStage2::MergedGraphNameToPasses is [111] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.007 [graph_manager.cc:2872][EVENT]77564 OptimizeStage2:[GEPERFTRACE] The time cost of OptimizeStage2::RemoveIsolatedConst is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.024 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage2::ControlAttrOptimize::MultiBatchPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.037 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses::RefIdentityDeleteOpPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.050 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses::VariableRefDeleteOpPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.062 [compile_nodes_pass.cc:88][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage2::ControlAttrOptimize::CompileNodesPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.081 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage2::ControlAttrOptimize::CompileNodesPass is [21] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.091 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses::SwapSpacePass is [0] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.147 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses::InputOutputConnectionIdentifyPass is [47] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.172 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses::AtomicAddrCleanPass is [13] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.184 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage2::AfterMergePasses::EndOfSequenceAddControlPass is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.195 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage2::ControlAttrOptimize::SubgraphPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.207 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage2::ControlAttrOptimize::AttachStreamLabelPass is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.215 [graph_manager.cc:2927][EVENT]77564 OptimizeStage2:[GEPERFTRACE] The time cost of OptimizeStage2::ControlAttrOptimize is [195] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.226 [graph_manager.cc:2937][EVENT]77564 OptimizeStage2:[GEPERFTRACE] The time cost of ModelBuilder::AssignFunctionalLabels is [3] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.240 [graph_manager.cc:2943][EVENT]77564 OptimizeStage2:[GEPERFTRACE] The time cost of MemcpyAddrAsyncPass::Run. is [5] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.251 [graph_manager.cc:2950][EVENT]77564 OptimizeStage2:[GEPERFTRACE] The time cost of BufferPoolMemoryPass::Run. is [1] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.408 [graph_manager.cc:2958][EVENT]77564 OptimizeStage2:[GEPERFTRACE] The time cost of ParallelGroupPass::Run. is [26] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.436 [graph_manager.cc:1132][EVENT]77564 PreRunAfterOptimizeSubGraph:[GEPERFTRACE] The time cost of GraphManager::OptimizeStage2 is [601] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.499 [graph_manager.cc:1135][EVENT]77564 PreRunAfterOptimizeSubGraph:[GEPERFTRACE] The time cost of GraphManager::GetCompilerStages(graph_node->GetGraphId()).optimizer.OptimizeGraphBeforeBuild is [51] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.532 [graph_manager.cc:2975][EVENT]77564 MemConflictProc:[GEPERFTRACE] The time cost of HandleMemoryRWConflict is [17] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.560 [graph_manager.cc:2981][EVENT]77564 MemConflictProc:[GEPERFTRACE] The time cost of MemLayoutConflictOptimizer::Run. is [14] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.573 [pass_manager.cc:82][EVENT]77564 Run:[GEPERFTRACE] The time cost of OptimizeStage2::SetFftsPlusAttrPass is [0] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.582 [graph_manager.cc:2986][EVENT]77564 MemConflictProc:[GEPERFTRACE] The time cost of SetFftsPlusAttrPass::last_passes.Run is [12] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.591 [graph_manager.cc:1136][EVENT]77564 PreRunAfterOptimizeSubGraph:[GEPERFTRACE] The time cost of GraphManager::MemConflictProc is [77] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.693 [graph_manager.cc:3555][EVENT]77564 Build:[GEPERFTRACE] The time cost of GraphManager::RecoverIrDefinitionAndModifyAippData is [75] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.787 [engine_partitioner.cc:1139][EVENT]77564 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionInitialize is [14] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.803 [engine_partitioner.cc:1142][EVENT]77564 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionMarkClusters is [2] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.899 [engine_partitioner.cc:1148][EVENT]77564 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionSplitSubGraphs is [87] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.928 [engine_partitioner.cc:1155][EVENT]77564 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionSortSubGraphs is [17] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.968 [engine_partitioner.cc:1164][EVENT]77564 PartitionSubGraph:[GEPERFTRACE] The time cost of EnginePartitioner::PartitionAddPartitionsToGraphNode is [28] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.112.987 [graph_builder.cc:865][EVENT]77564 SecondPartition:[GEPERFTRACE] The time cost of EnginePartitioner::Partition2 is [235] micro second. [INFO] RUNTIME(73092,python3.7):2024-01-11-05:37:51.113.408 [logger.cc:1071] 77564 ModelBindStream: model_id=832, stream_id=1089, flag=0. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.113.439 [task_generator.cc:804][EVENT]77564 GenerateTask:[GEPERFTRACE] The time cost of TaskGenerator::SetStreamCtx is [167] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.113.489 [task_generator.cc:805][EVENT]77564 GenerateTask:[GEPERFTRACE] The time cost of TaskGenerator::PrepareForGenerateTask is [37] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.114.097 [task_generator.cc:814][EVENT]77564 GenerateTask:[GEPERFTRACE] The time cost of TaskGenerator::DoGenerateTask is [595] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.114.112 [task_generator.cc:954][EVENT]77564 GetTaskInfo:[GEPERFTRACE] The time cost of TaskGenerator::GenerateTask is [839] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.114.157 [task_generator.cc:967][EVENT]77564 GetTaskInfo:[GEPERFTRACE] The time cost of TaskGenerator::AddModelTaskToModel is [24] micro second. [INFO] RUNTIME(73092,python3.7):2024-01-11-05:37:51.114.174 [logger.cc:1084] 77564 ModelUnbindStream: model_id=832, stream_id=1089, [INFO] GE(73092,python3.7):2024-01-11-05:37:51.114.311 [graph_manager.cc:1152][EVENT]77564 PreRunAfterOptimizeSubGraph:[GEPERFTRACE] The time cost of GraphManager::Build is [1700] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.114.328 [graph_manager.cc:1164][EVENT]77564 PreRunAfterOptimizeSubGraph:PreRun:PreRunAfterOptimizeSubGraph success. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.114.355 [graph_manager.cc:1271][EVENT]77564 PreRun:[GEPERFTRACE] The time cost of FlowModelBuild is [31766] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.114.366 [graph_manager.cc:1272][EVENT]77564 PreRun:[GEPERFTRACE] GE PreRun End [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:51.114.673 [atrace_api.c:93](tid:77564) AtraceDestroy start [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:51.114.687 [atrace_api.c:95](tid:77564) AtraceDestroy end [INFO] GE(73092,python3.7):2024-01-11-05:37:51.118.894 [graph_converter.cc:838][EVENT]77564 ConvertComputeGraphToExecuteGraph:[GEPERFTRACE] The time cost of ConvertComputeGraphToExecuteGraph::CreateMainNode is [1144] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.119.032 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of ZeroCopy is [94] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.119.464 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of CEM is [408] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.119.637 [copy_flow_launch_fuse.cc:395][EVENT]77564 Run:[GEPERFTRACE] The time cost of Pass::CopyFlowLaunchFuse is [150] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.119.658 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of CopyFlowLaunch is [172] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.119.859 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of TrustOutTensor is [178] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.119.878 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of AicpuFuseHostInputs is [0] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.119.910 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of ZeroCopy is [21] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.120.083 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of CEM is [160] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.120.162 [copy_flow_launch_fuse.cc:395][EVENT]77564 Run:[GEPERFTRACE] The time cost of Pass::CopyFlowLaunchFuse is [60] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.120.175 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of CopyFlowLaunch is [75] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.120.203 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of TrustOutTensor is [18] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.120.214 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of AicpuFuseHostInputs is [0] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.120.238 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of ZeroCopy is [16] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.120.303 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of CEM is [55] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.120.366 [copy_flow_launch_fuse.cc:395][EVENT]77564 Run:[GEPERFTRACE] The time cost of Pass::CopyFlowLaunchFuse is [53] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.120.376 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of CopyFlowLaunch is [63] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.120.401 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of TrustOutTensor is [16] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.120.410 [base_optimizer.cc:70][EVENT]77564 Run:[GEPERFTRACE] The time cost of AicpuFuseHostInputs is [0] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.120.422 [graph_converter.cc:849][EVENT]77564 ConvertComputeGraphToExecuteGraph:[GEPERFTRACE] The time cost of ConvertComputeGraphToExecuteGraph::RunAllPass is [1487] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.120.624 [graph_converter.cc:853][EVENT]77564 ConvertComputeGraphToExecuteGraph:[GEPERFTRACE] The time cost of ConvertComputeGraphToExecuteGraph::TopologicalSorting is [193] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.121.259 [graph_converter.cc:857][EVENT]77564 ConvertComputeGraphToExecuteGraph:[GEPERFTRACE] The time cost of ConvertComputeGraphToExecuteGraph::AppendGraphLevelData is [621] micro second. [INFO] GE(73092,python3.7):2024-01-11-05:37:51.121.383 [graph_converter.cc:862][EVENT]77564 ConvertComputeGraphToExecuteGraph:[GEPERFTRACE] The time cost of ConvertComputeGraphToExecuteGraph::CalculatePriority is [98] micro second. TotalTime = 0.0261733, [20] [parse]: 0.00151441 [symbol_resolve]: 0.0117613, [1] [Cycle 1]: 0.0117103, [1] [resolve]: 0.0116897 [combine_like_graphs]: 8.60004e-07 [graph_reusing]: 2.69e-06 [meta_unpack_prepare]: 5.614e-05 [pre_cconv]: 5.9e-07 [abstract_specialize]: 0.00314309 [pack_expand]: 1.23e-05 [auto_monad]: 5.103e-05 [inline]: 1.89e-06 [pre_auto_parallel]: 9.04e-06 [pipeline_split]: 2.97e-06 [optimize]: 0.00417034, [35] [py_interpret_to_execute]: 3.4e-06 [rewriter_before_opt_a]: 0.00010986 [opt_a]: 0.00359126, [2] [Cycle 1]: 0.00097703, [30] [expand_dump_flag]: 3.89e-06 [switch_simplify]: 1.748e-05 [a_1]: 0.00025012 [recompute_prepare]: 3.13e-06 [updatestate_depend_eliminate]: 6.48e-06 [updatestate_assign_eliminate]: 3.98e-06 [updatestate_loads_eliminate]: 3.45e-06 [parameter_eliminate]: 2.97e-06 [a_2]: 3.549e-05 [accelerated_algorithm]: 3.08e-06 [pynative_shard]: 1.6e-06 [auto_parallel]: 3.07e-06 [parallel]: 8.47e-06 [merge_comm]: 4.11e-06 [allreduce_fusion]: 1.81e-06 [virtual_dataset]: 2.96999e-06 [get_grad_eliminate_]: 2.62e-06 [virtual_output]: 2.01001e-06 [merge_forward]: 4.82e-06 [cell_reuse_recompute_pass]: 6.80004e-07 [cell_reuse_handle_not_recompute_node_pass]: 6.74e-06 [meta_fg_expand]: 3.63e-06 [after_resolve]: 5.76e-06 [a_after_grad]: 3.02e-06 [renormalize]: 0.00039243 [real_op_eliminate]: 6.02e-06 [auto_monad_grad]: 4.1e-06 [auto_monad_eliminator]: 1.1e-05 [cse]: 2.781e-05 [a_3]: 1.875e-05 [Cycle 2]: 0.00024228, [30] [expand_dump_flag]: 1.16001e-06 [switch_simplify]: 2.66e-06 [a_1]: 1.864e-05 [recompute_prepare]: 2.01e-06 [updatestate_depend_eliminate]: 3.35e-06 [updatestate_assign_eliminate]: 2.55999e-06 [updatestate_loads_eliminate]: 2.36e-06 [parameter_eliminate]: 1.08e-06 [a_2]: 3.159e-05 [accelerated_algorithm]: 2.55e-06 [pynative_shard]: 1.04e-06 [auto_parallel]: 3.11e-06 [parallel]: 3.26e-06 [merge_comm]: 1.85e-06 [allreduce_fusion]: 1.3e-06 [virtual_dataset]: 2.33e-06 [get_grad_eliminate_]: 2.14e-06 [virtual_output]: 2e-06 [merge_forward]: 3.22e-06 [cell_reuse_recompute_pass]: 3.70004e-07 [cell_reuse_handle_not_recompute_node_pass]: 6.52e-06 [meta_fg_expand]: 2.07e-06 [after_resolve]: 4.2e-06 [a_after_grad]: 2.65e-06 [renormalize]: 7.0002e-08 [real_op_eliminate]: 2.05e-06 [auto_monad_grad]: 7.30004e-07 [auto_monad_eliminator]: 4.89e-06 [cse]: 1.08e-05 [a_3]: 1.563e-05 [py_interpret_to_execute_after_opt_a]: 3.38e-06 [slice_cell_reuse_recomputed_activation]: 2.94e-06 [rewriter_after_opt_a]: 2.105e-05 [convert_after_rewriter]: 6.08e-06 [order_py_execute_after_rewriter]: 4.91e-06 [opt_b]: 9.727e-05, [1] [Cycle 1]: 9.245e-05, [7] [b_1]: 4.549e-05 [b_2]: 3.55e-06 [updatestate_depend_eliminate]: 2.65e-06 [updatestate_assign_eliminate]: 2.78e-06 [updatestate_loads_eliminate]: 2.29e-06 [renormalize]: 2.89998e-07 [cse]: 9.58001e-06 [cconv]: 2.283e-05 [opt_after_cconv]: 5.272e-05, [1] [Cycle 1]: 4.851e-05, [7] [c_1]: 5.87e-06 [parameter_eliminate]: 6.69999e-07 [updatestate_depend_eliminate]: 2.52e-06 [updatestate_assign_eliminate]: 2.16e-06 [updatestate_loads_eliminate]: 1.92e-06 [cse]: 9.17e-06 [renormalize]: 1.89997e-07 [remove_dup_value]: 1.211e-05 [tuple_transform]: 3.73e-05, [1] [Cycle 1]: 3.396e-05, [3] [d_1]: 1.539e-05 [d_2]: 7.20999e-06 [renormalize]: 1.49994e-07 [add_cache_embedding]: 1.011e-05 [add_recomputation]: 3.975e-05 [cse_after_recomputation]: 1.77e-05, [1] [Cycle 1]: 1.36e-05, [1] [cse]: 9.2e-06 [environ_conv]: 7.3e-06 [label_micro_interleaved_index]: 1.91e-06 [label_fine_grained_interleaved_index]: 2.66e-06 [assign_add_opt]: 1.51e-06 [slice_recompute_activation]: 2.11e-06 [micro_interleaved_order_control]: 1.54e-06 [full_micro_interleaved_order_control]: 1.78999e-06 [comp_comm_scheduling]: 2.08e-06 [reorder_send_recv_between_fp_bp]: 2.11e-06 [comm_op_add_attrs]: 9.59997e-07 [add_comm_op_reuse_tag]: 8.70001e-07 [overlap_opt_shard_in_pipeline]: 1.28e-06 [grouped_pairwise_exchange_alltoall]: 1.37e-06 [overlap_recompute_and_grad_model_parallel]: 1.92e-06 [overlap_grad_matmul_and_grad_allreduce]: 7.2e-07 [split_matmul_comm_elemetwise]: 2.67e-06 [split_layernorm_comm]: 1.68e-06 [process_send_recv_for_ge]: 8.60004e-07 [handle_group_info]: 1.18e-06 [auto_monad_reorder]: 1.71e-05 [get_jit_bprop_graph]: 3.49995e-07 [eliminate_special_op_node]: 0.00051174 [validate]: 2.816e-05 [distribtued_split]: 1.11001e-06 [task_emit]: 0.00469475 [execute]: 7.89999e-06 Sums parse : 0.001514s : 6.55% symbol_resolve.resolve : 0.011690s : 50.56% combine_like_graphs : 0.000001s : 0.00% graph_reusing : 0.000003s : 0.01% meta_unpack_prepare : 0.000056s : 0.24% pre_cconv : 0.000001s : 0.00% abstract_specialize : 0.003143s : 13.59% pack_expand : 0.000012s : 0.05% auto_monad : 0.000051s : 0.22% inline : 0.000002s : 0.01% pre_auto_parallel : 0.000009s : 0.04% pipeline_split : 0.000003s : 0.01% optimize.py_interpret_to_execute : 0.000003s : 0.01% optimize.rewriter_before_opt_a : 0.000110s : 0.48% optimize.opt_a.expand_dump_flag : 0.000005s : 0.02% optimize.opt_a.switch_simplify : 0.000020s : 0.09% optimize.opt_a.a_1 : 0.000269s : 1.16% optimize.opt_a.recompute_prepare : 0.000005s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000010s : 0.04% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.03% optimize.opt_a.updatestate_loads_eliminate : 0.000006s : 0.03% optimize.opt_a.parameter_eliminate : 0.000004s : 0.02% optimize.opt_a.a_2 : 0.000067s : 0.29% optimize.opt_a.accelerated_algorithm : 0.000006s : 0.02% optimize.opt_a.pynative_shard : 0.000003s : 0.01% optimize.opt_a.auto_parallel : 0.000006s : 0.03% optimize.opt_a.parallel : 0.000012s : 0.05% optimize.opt_a.merge_comm : 0.000006s : 0.03% optimize.opt_a.allreduce_fusion : 0.000003s : 0.01% optimize.opt_a.virtual_dataset : 0.000005s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000005s : 0.02% optimize.opt_a.virtual_output : 0.000004s : 0.02% optimize.opt_a.merge_forward : 0.000008s : 0.03% optimize.opt_a.cell_reuse_recompute_pass : 0.000001s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000013s : 0.06% optimize.opt_a.meta_fg_expand : 0.000006s : 0.02% optimize.opt_a.after_resolve : 0.000010s : 0.04% optimize.opt_a.a_after_grad : 0.000006s : 0.02% optimize.opt_a.renormalize : 0.000392s : 1.70% optimize.opt_a.real_op_eliminate : 0.000008s : 0.03% optimize.opt_a.auto_monad_grad : 0.000005s : 0.02% optimize.opt_a.auto_monad_eliminator : 0.000016s : 0.07% optimize.opt_a.cse : 0.000039s : 0.17% optimize.opt_a.a_3 : 0.000034s : 0.15% optimize.py_interpret_to_execute_after_opt_a : 0.000003s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.01% optimize.rewriter_after_opt_a : 0.000021s : 0.09% optimize.convert_after_rewriter : 0.000006s : 0.03% optimize.order_py_execute_after_rewriter : 0.000005s : 0.02% optimize.opt_b.b_1 : 0.000045s : 0.20% optimize.opt_b.b_2 : 0.000004s : 0.02% optimize.opt_b.updatestate_depend_eliminate : 0.000003s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000002s : 0.01% optimize.opt_b.renormalize : 0.000000s : 0.00% optimize.opt_b.cse : 0.000010s : 0.04% optimize.cconv : 0.000023s : 0.10% optimize.opt_after_cconv.c_1 : 0.000006s : 0.03% optimize.opt_after_cconv.parameter_eliminate : 0.000001s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000003s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000002s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000002s : 0.01% optimize.opt_after_cconv.cse : 0.000009s : 0.04% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000012s : 0.05% optimize.tuple_transform.d_1 : 0.000015s : 0.07% optimize.tuple_transform.d_2 : 0.000007s : 0.03% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.add_cache_embedding : 0.000010s : 0.04% optimize.add_recomputation : 0.000040s : 0.17% optimize.cse_after_recomputation.cse : 0.000009s : 0.04% optimize.environ_conv : 0.000007s : 0.03% optimize.label_micro_interleaved_index : 0.000002s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.01% optimize.assign_add_opt : 0.000002s : 0.01% optimize.slice_recompute_activation : 0.000002s : 0.01% optimize.micro_interleaved_order_control : 0.000002s : 0.01% optimize.full_micro_interleaved_order_control : 0.000002s : 0.01% optimize.comp_comm_scheduling : 0.000002s : 0.01% optimize.reorder_send_recv_between_fp_bp : 0.000002s : 0.01% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000001s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000002s : 0.01% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.01% optimize.split_layernorm_comm : 0.000002s : 0.01% optimize.process_send_recv_for_ge : 0.000001s : 0.00% optimize.handle_group_info : 0.000001s : 0.01% auto_monad_reorder : 0.000017s : 0.07% get_jit_bprop_graph : 0.000000s : 0.00% eliminate_special_op_node : 0.000512s : 2.21% validate : 0.000028s : 0.12% distribtued_split : 0.000001s : 0.00% task_emit : 0.004695s : 20.31% execute : 0.000008s : 0.03% Time group info: ------[substitution.] 0.011533 51 98.69% : 0.011382s : 12: substitution.getattr_setattr_resolve 0.04% : 0.000005s : 5: substitution.graph_param_transform 1.03% : 0.000119s : 4: substitution.inline 0.10% : 0.000012s : 17: substitution.meta_unpack_prepare 0.01% : 0.000001s : 5: substitution.partial_unused_args_eliminate 0.01% : 0.000001s : 4: substitution.remove_not_recompute_node 0.02% : 0.000003s : 2: substitution.replace_old_param 0.09% : 0.000010s : 2: substitution.tuple_list_get_item_eliminator ------[renormalize.] 0.000386 2 57.00% : 0.000220s : 1: renormalize.infer 43.00% : 0.000166s : 1: renormalize.specialize ------[replace.] 0.000217 16 77.65% : 0.000169s : 10: replace.getattr_setattr_resolve 15.82% : 0.000034s : 4: replace.inline 6.53% : 0.000014s : 2: replace.tuple_list_get_item_eliminator ------[match.] 0.011441 16 98.87% : 0.011312s : 10: match.getattr_setattr_resolve 1.04% : 0.000119s : 4: match.inline 0.09% : 0.000010s : 2: match.tuple_list_get_item_eliminator ------[func_graph_cloner_run.] 0.000654 13 70.90% : 0.000464s : 7: func_graph_cloner_run.FuncGraphClonerGraph 29.10% : 0.000190s : 6: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.012213 105 0.78% : 0.000095s : 52: opt.transform.opt_a 0.31% : 0.000038s : 23: opt.transform.opt_b 95.66% : 0.011683s : 2: opt.transform.opt_resolve 0.29% : 0.000035s : 1: opt.transforms.meta_unpack_prepare 2.65% : 0.000324s : 20: opt.transforms.opt_a 0.04% : 0.000005s : 1: opt.transforms.opt_after_cconv 0.02% : 0.000003s : 1: opt.transforms.opt_b 0.17% : 0.000021s : 2: opt.transforms.opt_trans_graph 0.08% : 0.000010s : 3: opt.transforms.special_op_eliminate TotalTime = 0.0472588, [20] [parse]: 0.00147826 [symbol_resolve]: 0.0204601, [1] [Cycle 1]: 0.0203803, [1] [resolve]: 0.0203591 [combine_like_graphs]: 1.01e-06 [graph_reusing]: 3.45999e-06 [meta_unpack_prepare]: 0.00012259 [pre_cconv]: 6.19999e-07 [abstract_specialize]: 0.0106372 [pack_expand]: 1.732e-05 [auto_monad]: 0.00017591 [inline]: 1.84e-06 [pre_auto_parallel]: 1.054e-05 [pipeline_split]: 2.77e-06 [optimize]: 0.00891468, [35] [py_interpret_to_execute]: 3.97e-06 [rewriter_before_opt_a]: 0.0002269 [opt_a]: 0.00821924, [3] [Cycle 1]: 0.00441417, [30] [expand_dump_flag]: 4.8e-06 [switch_simplify]: 9.258e-05 [a_1]: 0.00054316 [recompute_prepare]: 8.2e-06 [updatestate_depend_eliminate]: 1.024e-05 [updatestate_assign_eliminate]: 7.45e-06 [updatestate_loads_eliminate]: 6.8e-06 [parameter_eliminate]: 5.31e-06 [a_2]: 9.856e-05 [accelerated_algorithm]: 6.24e-06 [pynative_shard]: 1.91e-06 [auto_parallel]: 3.45e-06 [parallel]: 8.48e-06 [merge_comm]: 8.34e-06 [allreduce_fusion]: 3.4e-06 [virtual_dataset]: 5.95e-06 [get_grad_eliminate_]: 4.99e-06 [virtual_output]: 4.61e-06 [merge_forward]: 9.55e-06 [cell_reuse_recompute_pass]: 7.99999e-07 [cell_reuse_handle_not_recompute_node_pass]: 1.263e-05 [meta_fg_expand]: 0.00074146 [after_resolve]: 2.561e-05 [a_after_grad]: 5.273e-05 [renormalize]: 0.00218175 [real_op_eliminate]: 4.625e-05 [auto_monad_grad]: 1.289e-05 [auto_monad_eliminator]: 4.24e-05 [cse]: 0.00015282 [a_3]: 0.00013422 [Cycle 2]: 0.00113559, [30] [expand_dump_flag]: 1.58e-06 [switch_simplify]: 1.557e-05 [a_1]: 0.00046295 [recompute_prepare]: 3.06e-06 [updatestate_depend_eliminate]: 4.61e-06 [updatestate_assign_eliminate]: 2.8e-06 [updatestate_loads_eliminate]: 2.47e-06 [parameter_eliminate]: 2.44e-06 [a_2]: 3.536e-05 [accelerated_algorithm]: 3.02e-06 [pynative_shard]: 1.02e-06 [auto_parallel]: 3.18e-06 [parallel]: 3.50999e-06 [merge_comm]: 2.4e-06 [allreduce_fusion]: 1.77e-06 [virtual_dataset]: 3.12e-06 [get_grad_eliminate_]: 2.38e-06 [virtual_output]: 2.26e-06 [merge_forward]: 3.56e-06 [cell_reuse_recompute_pass]: 3.50003e-07 [cell_reuse_handle_not_recompute_node_pass]: 5.8e-06 [meta_fg_expand]: 1.352e-05 [after_resolve]: 2.52e-06 [a_after_grad]: 2.88e-06 [renormalize]: 0.00036849 [real_op_eliminate]: 5.82e-06 [auto_monad_grad]: 3.39e-06 [auto_monad_eliminator]: 7.71999e-06 [cse]: 1.875e-05 [a_3]: 1.854e-05 [Cycle 3]: 0.00026166, [30] [expand_dump_flag]: 1.1e-06 [switch_simplify]: 2.61e-06 [a_1]: 1.681e-05 [recompute_prepare]: 2.11e-06 [updatestate_depend_eliminate]: 3.38e-06 [updatestate_assign_eliminate]: 2.56e-06 [updatestate_loads_eliminate]: 2.53e-06 [parameter_eliminate]: 1.04e-06 [a_2]: 3.205e-05 [accelerated_algorithm]: 2.59e-06 [pynative_shard]: 1.1e-06 [auto_parallel]: 3.15e-06 [parallel]: 3.27e-06 [merge_comm]: 2.06001e-06 [allreduce_fusion]: 1.28e-06 [virtual_dataset]: 2.39e-06 [get_grad_eliminate_]: 2.13e-06 [virtual_output]: 1.97e-06 [merge_forward]: 3.18e-06 [cell_reuse_recompute_pass]: 4.00003e-07 [cell_reuse_handle_not_recompute_node_pass]: 6.52e-06 [meta_fg_expand]: 2.33e-06 [after_resolve]: 1.99e-06 [a_after_grad]: 2.46e-06 [renormalize]: 7.0002e-08 [real_op_eliminate]: 2.03e-06 [auto_monad_grad]: 9.20001e-07 [auto_monad_eliminator]: 4.79999e-06 [cse]: 3.364e-05 [a_3]: 1.603e-05 [py_interpret_to_execute_after_opt_a]: 3.61e-06 [slice_cell_reuse_recomputed_activation]: 2.4e-06 [rewriter_after_opt_a]: 2.273e-05 [convert_after_rewriter]: 5.98e-06 [order_py_execute_after_rewriter]: 4.47e-06 [opt_b]: 0.00010449, [1] [Cycle 1]: 9.941e-05, [7] [b_1]: 4.768e-05 [b_2]: 3.39001e-06 [updatestate_depend_eliminate]: 2.95e-06 [updatestate_assign_eliminate]: 2.63e-06 [updatestate_loads_eliminate]: 2.29e-06 [renormalize]: 4.20005e-07 [cse]: 1.143e-05 [cconv]: 1.891e-05 [opt_after_cconv]: 5.255e-05, [1] [Cycle 1]: 4.858e-05, [7] [c_1]: 5.58e-06 [parameter_eliminate]: 7.40001e-07 [updatestate_depend_eliminate]: 2.76e-06 [updatestate_assign_eliminate]: 2.27e-06 [updatestate_loads_eliminate]: 2.14e-06 [cse]: 9.07999e-06 [renormalize]: 2.10006e-07 [remove_dup_value]: 1.298e-05 [tuple_transform]: 3.656e-05, [1] [Cycle 1]: 3.298e-05, [3] [d_1]: 1.48e-05 [d_2]: 7.09e-06 [renormalize]: 1.49994e-07 [add_cache_embedding]: 1.079e-05 [add_recomputation]: 3.152e-05 [cse_after_recomputation]: 1.776e-05, [1] [Cycle 1]: 1.384e-05, [1] [cse]: 9.59e-06 [environ_conv]: 7.24e-06 [label_micro_interleaved_index]: 2.65e-06 [label_fine_grained_interleaved_index]: 2.4e-06 [assign_add_opt]: 1.35e-06 [slice_recompute_activation]: 2.03e-06 [micro_interleaved_order_control]: 1.56e-06 [full_micro_interleaved_order_control]: 1.76e-06 [comp_comm_scheduling]: 2.2e-06 [reorder_send_recv_between_fp_bp]: 2.19e-06 [comm_op_add_attrs]: 9.70002e-07 [add_comm_op_reuse_tag]: 1.36e-06 [overlap_opt_shard_in_pipeline]: 1.03e-06 [grouped_pairwise_exchange_alltoall]: 1.28e-06 [overlap_recompute_and_grad_model_parallel]: 1.64e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.08e-06 [split_matmul_comm_elemetwise]: 2.38e-06 [split_layernorm_comm]: 1.81e-06 [process_send_recv_for_ge]: 1.05e-06 [handle_group_info]: 8.70001e-07 [auto_monad_reorder]: 1.594e-05 [get_jit_bprop_graph]: 4.80002e-07 [eliminate_special_op_node]: 0.00048038 [validate]: 2.94e-05 [distribtued_split]: 1.12e-06 [task_emit]: 0.0046999 [execute]: 7.37e-06 Sums parse : 0.001478s : 3.36% symbol_resolve.resolve : 0.020359s : 46.32% combine_like_graphs : 0.000001s : 0.00% graph_reusing : 0.000003s : 0.01% meta_unpack_prepare : 0.000123s : 0.28% pre_cconv : 0.000001s : 0.00% abstract_specialize : 0.010637s : 24.20% pack_expand : 0.000017s : 0.04% auto_monad : 0.000176s : 0.40% inline : 0.000002s : 0.00% pre_auto_parallel : 0.000011s : 0.02% pipeline_split : 0.000003s : 0.01% optimize.py_interpret_to_execute : 0.000004s : 0.01% optimize.rewriter_before_opt_a : 0.000227s : 0.52% optimize.opt_a.expand_dump_flag : 0.000007s : 0.02% optimize.opt_a.switch_simplify : 0.000111s : 0.25% optimize.opt_a.a_1 : 0.001023s : 2.33% optimize.opt_a.recompute_prepare : 0.000013s : 0.03% optimize.opt_a.updatestate_depend_eliminate : 0.000018s : 0.04% optimize.opt_a.updatestate_assign_eliminate : 0.000013s : 0.03% optimize.opt_a.updatestate_loads_eliminate : 0.000012s : 0.03% optimize.opt_a.parameter_eliminate : 0.000009s : 0.02% optimize.opt_a.a_2 : 0.000166s : 0.38% optimize.opt_a.accelerated_algorithm : 0.000012s : 0.03% optimize.opt_a.pynative_shard : 0.000004s : 0.01% optimize.opt_a.auto_parallel : 0.000010s : 0.02% optimize.opt_a.parallel : 0.000015s : 0.03% optimize.opt_a.merge_comm : 0.000013s : 0.03% optimize.opt_a.allreduce_fusion : 0.000006s : 0.01% optimize.opt_a.virtual_dataset : 0.000011s : 0.03% optimize.opt_a.get_grad_eliminate_ : 0.000009s : 0.02% optimize.opt_a.virtual_output : 0.000009s : 0.02% optimize.opt_a.merge_forward : 0.000016s : 0.04% optimize.opt_a.cell_reuse_recompute_pass : 0.000002s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000025s : 0.06% optimize.opt_a.meta_fg_expand : 0.000757s : 1.72% optimize.opt_a.after_resolve : 0.000030s : 0.07% optimize.opt_a.a_after_grad : 0.000058s : 0.13% optimize.opt_a.renormalize : 0.002550s : 5.80% optimize.opt_a.real_op_eliminate : 0.000054s : 0.12% optimize.opt_a.auto_monad_grad : 0.000017s : 0.04% optimize.opt_a.auto_monad_eliminator : 0.000055s : 0.12% optimize.opt_a.cse : 0.000205s : 0.47% optimize.opt_a.a_3 : 0.000169s : 0.38% optimize.py_interpret_to_execute_after_opt_a : 0.000004s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.01% optimize.rewriter_after_opt_a : 0.000023s : 0.05% optimize.convert_after_rewriter : 0.000006s : 0.01% optimize.order_py_execute_after_rewriter : 0.000004s : 0.01% optimize.opt_b.b_1 : 0.000048s : 0.11% optimize.opt_b.b_2 : 0.000003s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000003s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000002s : 0.01% optimize.opt_b.renormalize : 0.000000s : 0.00% optimize.opt_b.cse : 0.000011s : 0.03% optimize.cconv : 0.000019s : 0.04% optimize.opt_after_cconv.c_1 : 0.000006s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000001s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000003s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000002s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000002s : 0.00% optimize.opt_after_cconv.cse : 0.000009s : 0.02% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000013s : 0.03% optimize.tuple_transform.d_1 : 0.000015s : 0.03% optimize.tuple_transform.d_2 : 0.000007s : 0.02% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.add_cache_embedding : 0.000011s : 0.02% optimize.add_recomputation : 0.000032s : 0.07% optimize.cse_after_recomputation.cse : 0.000010s : 0.02% optimize.environ_conv : 0.000007s : 0.02% optimize.label_micro_interleaved_index : 0.000003s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000002s : 0.01% optimize.assign_add_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.comp_comm_scheduling : 0.000002s : 0.01% optimize.reorder_send_recv_between_fp_bp : 0.000002s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000001s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000002s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.01% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.process_send_recv_for_ge : 0.000001s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% auto_monad_reorder : 0.000016s : 0.04% get_jit_bprop_graph : 0.000000s : 0.00% eliminate_special_op_node : 0.000480s : 1.09% validate : 0.000029s : 0.07% distribtued_split : 0.000001s : 0.00% task_emit : 0.004700s : 10.69% execute : 0.000007s : 0.02% Time group info: ------[substitution.] 0.020306 222 0.02% : 0.000004s : 10: substitution.float_depend_g_call 0.01% : 0.000002s : 2: substitution.float_tuple_getitem_switch 97.26% : 0.019750s : 23: substitution.getattr_setattr_resolve 0.02% : 0.000005s : 4: substitution.graph_param_transform 0.01% : 0.000002s : 2: substitution.incorporate_call 0.00% : 0.000001s : 2: substitution.incorporate_call_switch 1.83% : 0.000372s : 21: substitution.inline 0.15% : 0.000030s : 60: substitution.meta_unpack_prepare 0.03% : 0.000007s : 7: substitution.minmaximum_grad 0.02% : 0.000004s : 10: substitution.partial_eliminate 0.01% : 0.000002s : 4: substitution.partial_unused_args_eliminate 0.03% : 0.000006s : 1: substitution.real_op_eliminate 0.01% : 0.000002s : 12: substitution.remove_not_recompute_node 0.06% : 0.000013s : 9: substitution.replace_applicator 0.01% : 0.000003s : 7: substitution.replace_old_param 0.01% : 0.000001s : 1: substitution.set_cell_output_no_recompute 0.03% : 0.000006s : 3: substitution.switch_simplify 0.13% : 0.000026s : 7: substitution.tuple_list_convert_item_index_to_positive 0.04% : 0.000008s : 7: substitution.tuple_list_get_item_const_eliminator 0.06% : 0.000012s : 7: substitution.tuple_list_get_item_depend_reorder 0.18% : 0.000037s : 16: substitution.tuple_list_get_item_eliminator 0.05% : 0.000011s : 7: substitution.tuple_list_get_set_item_eliminator ------[renormalize.] 0.002541 4 56.63% : 0.001439s : 2: renormalize.infer 43.37% : 0.001102s : 2: renormalize.specialize ------[replace.] 0.000561 50 56.87% : 0.000319s : 20: replace.getattr_setattr_resolve 23.75% : 0.000133s : 19: replace.inline 2.07% : 0.000012s : 1: replace.real_op_eliminate 6.95% : 0.000039s : 3: replace.switch_simplify 10.35% : 0.000058s : 7: replace.tuple_list_get_item_eliminator ------[match.] 0.019995 50 98.06% : 0.019607s : 20: match.getattr_setattr_resolve 1.78% : 0.000355s : 19: match.inline 0.03% : 0.000006s : 1: match.real_op_eliminate 0.03% : 0.000006s : 3: match.switch_simplify 0.10% : 0.000020s : 7: match.tuple_list_get_item_eliminator ------[func_graph_cloner_run.] 0.002136 39 73.45% : 0.001569s : 16: func_graph_cloner_run.FuncGraphClonerGraph 26.55% : 0.000567s : 23: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.022163 141 1.48% : 0.000327s : 78: opt.transform.opt_a 0.18% : 0.000039s : 23: opt.transform.opt_b 91.83% : 0.020351s : 2: opt.transform.opt_resolve 0.45% : 0.000100s : 1: opt.transforms.meta_unpack_prepare 5.90% : 0.001308s : 30: opt.transforms.opt_a 0.02% : 0.000004s : 1: opt.transforms.opt_after_cconv 0.01% : 0.000003s : 1: opt.transforms.opt_b 0.09% : 0.000020s : 2: opt.transforms.opt_trans_graph 0.05% : 0.000010s : 3: opt.transforms.special_op_eliminate . ============================== 1 passed in 21.22s ============================== [TRACE] GE(73092,python3.7):2024-01-11-05:37:53.275.692 [status:INIT] [ge_api.cc:463]73092 ~Session:Start to destruct session. [TRACE] GE(73092,python3.7):2024-01-11-05:37:53.276.141 [status:RUNNING] [ge_api.cc:475]73092 ~Session:Session id is 0 [TRACE] GE(73092,python3.7):2024-01-11-05:37:53.276.163 [status:RUNNING] [ge_api.cc:476]73092 ~Session:Destroying session [TRACE] GE(73092,python3.7):2024-01-11-05:37:53.277.042 [status:STOP] [ge_api.cc:491]73092 ~Session:Session Destructor finished [TRACE] GE(73092,python3.7):2024-01-11-05:37:53.277.072 [status:INIT] [ge_api.cc:301]73092 GEFinalize:GEFinalize start [INFO] GE(73092,python3.7):2024-01-11-05:37:53.277.183 [execution_runtime.cc:80][EVENT]73092 FinalizeExecutionRuntime:Execution runtime finalize begin. [INFO] GE(73092,python3.7):2024-01-11-05:37:53.277.201 [execution_runtime.cc:92][EVENT]73092 FinalizeExecutionRuntime:Execution runtime finalized. [TRACE] GE(73092,python3.7):2024-01-11-05:37:53.277.212 [status:RUNNING] [ge_api.cc:313]73092 GEFinalize:Finalizing environment [INFO] TUNE(73092,python3.7):2024-01-11-05:37:53.567.436 [cann_kb_pyfunc_mgr.cpp:127][CANNKB][Tid:73092]"CannKbPyfuncMgr: enter PyObjectDeinit function, reference_[1]" [INFO] TUNE(73092,python3.7):2024-01-11-05:37:53.567.485 [cann_kb_pyfunc_mgr.cpp:138][CANNKB][Tid:73092]"CannKbPyfuncMgr: PyObjectDeinit function end successfully!" [INFO] GE(73092,python3.7):2024-01-11-05:37:53.568.808 [gelib.cc:324][EVENT]73092 SystemFinalize:Online infer finalize GELib success. [TRACE] GE(73092,python3.7):2024-01-11-05:37:54.499.715 [status:STOP] [ge_api.cc:341]73092 GEFinalize:GEFinalize finished [INFO] TDT(73092,python3.7):2024-01-11-05:37:54.911.437 [process_mode_manager.cpp:184][Close][tid:73092] [TsdClient] Close [deviceId=7][sessionId=1] hccp and computer enter [INFO] TDT(73092,python3.7):2024-01-11-05:37:54.911.488 [version_verify.cpp:112][SpecialFeatureCheck][tid:73092] VersionVerify: previous type[7], supported [INFO] TDT(73092,python3.7):2024-01-11-05:37:54.911.532 [process_mode_manager.cpp:192][Close][tid:73092] [TsdClient][deviceId=7] [sessionId=1] wait hccp and computer process close respond [INFO] TDT(73092,python3.7):2024-01-11-05:37:54.932.957 [process_mode_manager.cpp:197][Close][tid:73092] [TsdClient][logicDeviceId_=7]has recv close hccp and computer process respond [INFO] TDT(73092,python3.7):2024-01-11-05:37:54.932.975 [stub_process_mode_nowin.cpp:151][CloseInHost][tid:73092] enter into CloseInHost deviceid[7] [INFO] TDT(73092,python3.7):2024-01-11-05:37:54.932.986 [stub_process_mode_nowin.cpp:154][CloseInHost][tid:73092] host cpu not support [INFO] TDT(73092,python3.7):2024-01-11-05:37:54.933.050 [process_mode_manager.cpp:208][Close][tid:73092] [TsdClient][deviceId=7] [sessionId=1] close hccp and computer process success [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:54.933.063 [atrace_api.c:93](tid:73092) AtraceDestroy start [INFO] ATRACE(73092,python3.7):2024-01-11-05:37:54.933.078 [atrace_api.c:95](tid:73092) AtraceDestroy end [INFO] PROFILING(73092,python3.7):2024-01-11-05:37:54.933.098 [msprofiler_impl.cpp:156] >>> (tid:73092) ProfNotifySetDevice called, is open: 0, devId: 7 [INFO] RUNTIME(73092,python3.7):2024-01-11-05:37:56.813.550 [runtime.cc:1737] 73092 ~Runtime: deconstruct runtime.