==================================================Ascend ============================= test session starts ============================== platform linux -- Python 3.9.19, pytest-6.2.5, py-1.11.0, pluggy-1.5.0 rootdir: /home/jenkins/mindspore/testcases/testcases/tests/st/tensor, configfile: ../../../../../../sault/virtual_test/virtualenv_005/sault/config/pytest.ini plugins: mock-3.14.0, hydra-core-1.3.2, forked-1.6.0, anyio-4.9.0, xdist-1.32.0 collected 16 items test_sigmoid_.py [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:40:26.799.986 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 50.7579, [24] [bootstrap]: 0.00094627 [type_inference]: 0.197097 [event_method]: 1.906e-05 [auto_monad]: 0.00021739 [graph_reusing]: 6.36e-06 [inline]: 4e-06 [add_attr]: 0.00966862, [1] [add_attr_with_inline]: 0.00964901, [1] [Cycle 1]: 9.581e-05, [2] [tag_attr]: 2.095e-05 [meta_addattr_fg_expand]: 3.58999e-06 [parallel-infer-symbol]: 4.66002e-06 [pre_auto_parallel]: 4.984e-05 [insert-virtual-dataset]: 2.62001e-06 [parallel-infer-symbol-second]: 6.89994e-07 [dataset_repeat_opt]: 2.41e-06 [pipeline_split]: 2.39999e-06 [optimize]: 0.111089, [53] [py_interpret_to_execute]: 2.319e-05 [rewriter_before_opt_a]: 6.996e-05 [opt_a]: 0.107773, [2] [Cycle 1]: 0.106813, [45] [expand_dump_flag]: 3.04999e-06 [switch_simplify]: 2.678e-05 [loop_unroll]: 1.263e-05 [a_1]: 0.104602 [with_stream_mark]: 4.62e-05 [recompute_prepare]: 2.185e-05 [updatestate_depend_eliminate]: 7.18e-06 [updatestate_assign_eliminate]: 6.14001e-06 [updatestate_loads_eliminate]: 5.22e-06 [parameter_eliminate]: 3.67002e-06 [a_2]: 0.00015709 [accelerated_algorithm]: 1.246e-05 [shard]: 4.70001e-06 [meta_shard_fg_expand]: 5.98002e-06 [shard_inline]: 9.80002e-06 [merge_send_recv]: 1.38e-05 [auto_parallel]: 1.393e-05 [parallel]: 4.473e-05 [flash_sp]: 2.779e-05 [merge_comm]: 5.97001e-06 [allreduce_fusion]: 5.67001e-06 [matmul_add_comm_reduction]: 1.594e-05 [allreduce_slice_to_reducescatter]: 1.00001e-06 [virtual_shard_identity]: 1.586e-05 [virtual_dataset]: 1.005e-05 [get_grad_eliminate_]: 9.77001e-06 [virtual_output]: 1.024e-05 [merge_forward]: 6.94001e-06 [cell_reuse_recompute_pass]: 3.51001e-06 [offload_activation]: 1.498e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.021e-05 [merge_recompute_call_nodes]: 1.82001e-06 [before_grad]: 1.7e-05 [set_forward_comm_id_for_comm_node_pass]: 6.44001e-06 [meta_fg_expand]: 5.57001e-06 [flash_sp_send_recv_attached]: 3.98001e-06 [receive_attached]: 1.299e-05 [after_resolve]: 1.689e-05 [a_after_grad]: 1.429e-05 [renormalize]: 0.00101788 [add_forward_monad_depend]: 1.421e-05 [auto_monad_grad]: 2.75002e-06 [auto_monad_eliminator]: 2.906e-05 [cse]: 6.505e-05 [a_3]: 7.745e-05 [Cycle 2]: 0.00094219, [45] [expand_dump_flag]: 3.39001e-06 [switch_simplify]: 1.344e-05 [loop_unroll]: 9.94999e-06 [a_1]: 0.00021611 [with_stream_mark]: 2.383e-05 [recompute_prepare]: 9.42999e-06 [updatestate_depend_eliminate]: 5.80002e-06 [updatestate_assign_eliminate]: 5.05001e-06 [updatestate_loads_eliminate]: 5.59e-06 [parameter_eliminate]: 1.77999e-06 [a_2]: 0.00011578 [accelerated_algorithm]: 1.015e-05 [shard]: 1.62999e-06 [meta_shard_fg_expand]: 2.61e-06 [shard_inline]: 8.52998e-06 [merge_send_recv]: 1.072e-05 [auto_parallel]: 1.169e-05 [parallel]: 7.88999e-06 [flash_sp]: 4.38999e-06 [merge_comm]: 5.86e-06 [allreduce_fusion]: 6.04001e-06 [matmul_add_comm_reduction]: 1.009e-05 [allreduce_slice_to_reducescatter]: 6.80011e-07 [virtual_shard_identity]: 1.068e-05 [virtual_dataset]: 8.55001e-06 [get_grad_eliminate_]: 8.64e-06 [virtual_output]: 8.3e-06 [merge_forward]: 5.55001e-06 [cell_reuse_recompute_pass]: 2.51e-06 [offload_activation]: 1.237e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.933e-05 [merge_recompute_call_nodes]: 1.20001e-06 [before_grad]: 1.495e-05 [set_forward_comm_id_for_comm_node_pass]: 5.40999e-06 [meta_fg_expand]: 3.56999e-06 [flash_sp_send_recv_attached]: 1.77999e-06 [receive_attached]: 1.91e-06 [after_resolve]: 1.28e-05 [a_after_grad]: 1.252e-05 [renormalize]: 7.00238e-08 [add_forward_monad_depend]: 1.67999e-06 [auto_monad_grad]: 1.62999e-06 [auto_monad_eliminator]: 1.761e-05 [cse]: 2.544e-05 [a_3]: 5.336e-05 [py_interpret_to_execute_after_opt_a]: 2.291e-05 [slice_cell_reuse_recomputed_activation]: 2.63003e-06 [rewriter_after_opt_a]: 0.00024417 [convert_after_rewriter]: 1.83e-05 [order_py_execute_after_rewriter]: 6.98e-06 [mutable_eliminate]: 0.00091643 [opt_b]: 0.00033855, [1] [Cycle 1]: 0.00032877, [7] [b_1]: 0.00020711 [b_2]: 1.205e-05 [updatestate_depend_eliminate]: 1.631e-05 [updatestate_assign_eliminate]: 5.04e-06 [updatestate_loads_eliminate]: 5.35001e-06 [renormalize]: 9.79984e-07 [cse]: 4.187e-05 [optimize_parallel_all_gather_comm]: 2.725e-05 [overlap_param_gather]: 2.06e-06 [cconv]: 3.929e-05 [loop_unroll]: 0.00062591 [opt_after_cconv]: 0.00016486, [1] [Cycle 1]: 0.00015643, [7] [c_1]: 6.039e-05 [parameter_eliminate]: 6.36e-06 [updatestate_depend_eliminate]: 1.057e-05 [updatestate_assign_eliminate]: 5.33002e-06 [updatestate_loads_eliminate]: 4.45999e-06 [cse]: 3.084e-05 [renormalize]: 7.39994e-07 [remove_dup_value]: 1.953e-05 [tuple_transform]: 9.715e-05, [1] [Cycle 1]: 9.092e-05, [4] [d_1]: 5.95e-05 [none_parameter_eliminate]: 1.87999e-06 [renormalize]: 2.09984e-07 [switch_simplify]: 9.99001e-06 [partial_unused_args_eliminate]: 2.11e-06 [add_recomputation]: 8.996e-05 [cse_after_recomputation]: 2.74e-05, [1] [Cycle 1]: 2.223e-05, [1] [cse]: 1.609e-05 [environ_conv]: 2.396e-05 [swap_dp_allreduce_reducescatter]: 9.24998e-06 [bias_add_comm_swap]: 3.84002e-06 [label_micro_interleaved_index]: 6.57002e-06 [label_fine_grained_interleaved_index]: 3.22002e-06 [merge_cast_opt]: 1.50001e-06 [slice_recompute_activation]: 2.84999e-06 [micro_interleaved_order_control]: 2.63003e-06 [assign_add_opt]: 1.32999e-06 [ForceFp32Comm]: 1.46002e-06 [remove_cast_before_assign_add]: 1.04e-06 [full_micro_interleaved_order_control]: 2.86999e-06 [reorder_send_recv_between_fp_bp]: 3.18e-06 [comm_op_add_attrs]: 1.34e-06 [add_comm_op_reuse_tag]: 1.08001e-06 [interleave_split_concat_branches]: 1.26997e-06 [interleave_parallel_branches]: 1.05999e-06 [overlap_opt_shard_in_pipeline]: 2.432e-05 [overlap_opt_shard_grad_in_pipeline]: 2.02999e-06 [control_data_broadcast_order]: 2.131e-05 [grouped_pairwise_exchange_alltoall]: 1.70001e-06 [offloading_packed_experts]: 6.23e-06 [overlap_recompute_and_grad_model_parallel]: 6.51999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.61998e-06 [overlap_recompute_allgather_and_fa_grad]: 1.44e-06 [overlap_recompute_comm]: 2.79001e-06 [overlap_grad_ring_attention]: 6.63e-06 [overlap_grad_flash_sp]: 4.847e-05 [begin_end_overlap_inline]: 5.8001e-07 [split_matmul_comm_elemetwise]: 3.22002e-06 [split_layernorm_comm]: 2.09e-06 [handle_group_info]: 1.11002e-06 [symbol_engine_optimizer]: 0.00010239, [1] [Cycle 1]: 9.713e-05, [6] [build]: 5.17e-06 [elim_shapecalc]: 1.415e-05 [elim_not_effective]: 1.926e-05 [opt_reshape]: 9.59999e-06 [fold_const_symbol]: 1.533e-05 [renormalize]: 1.79978e-07 [detach_backward]: 2.80997e-06 [pipeline_parallel_scheduler]: 1.97999e-06 [auto_monad_reorder]: 3.966e-05 [get_jit_bprop_graph]: 1.92001e-06 [rewriter_after_jit_bprop_graph]: 6.44999e-06 [opt_after_jit_grad]: 0.0007286 [validate]: 8.431e-05 [backend_pass]: 1.05001e-06 [task_emit]: 50.4376 [execute]: 1.185e-05 Sums bootstrap : 0.000946s : 0.00% type_inference : 0.197097s : 0.39% event_method : 0.000019s : 0.00% auto_monad : 0.000217s : 0.00% graph_reusing : 0.000006s : 0.00% inline : 0.000004s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000021s : 0.00% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000005s : 0.00% pre_auto_parallel : 0.000050s : 0.00% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000023s : 0.00% optimize.rewriter_before_opt_a : 0.000070s : 0.00% optimize.opt_a.expand_dump_flag : 0.000006s : 0.00% optimize.opt_a.switch_simplify : 0.000040s : 0.00% optimize.opt_a.loop_unroll : 0.000023s : 0.00% optimize.opt_a.a_1 : 0.104818s : 0.21% optimize.opt_a.with_stream_mark : 0.000070s : 0.00% optimize.opt_a.recompute_prepare : 0.000031s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000013s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000011s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000011s : 0.00% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000273s : 0.00% optimize.opt_a.accelerated_algorithm : 0.000023s : 0.00% optimize.opt_a.shard : 0.000006s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000009s : 0.00% optimize.opt_a.shard_inline : 0.000018s : 0.00% optimize.opt_a.merge_send_recv : 0.000025s : 0.00% optimize.opt_a.auto_parallel : 0.000026s : 0.00% optimize.opt_a.parallel : 0.000053s : 0.00% optimize.opt_a.flash_sp : 0.000032s : 0.00% optimize.opt_a.merge_comm : 0.000012s : 0.00% optimize.opt_a.allreduce_fusion : 0.000012s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000026s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000027s : 0.00% optimize.opt_a.virtual_dataset : 0.000019s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000018s : 0.00% optimize.opt_a.virtual_output : 0.000019s : 0.00% optimize.opt_a.merge_forward : 0.000012s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000006s : 0.00% optimize.opt_a.offload_activation : 0.000027s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000040s : 0.00% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000032s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000012s : 0.00% optimize.opt_a.meta_fg_expand : 0.000009s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000006s : 0.00% optimize.opt_a.receive_attached : 0.000015s : 0.00% optimize.opt_a.after_resolve : 0.000030s : 0.00% optimize.opt_a.a_after_grad : 0.000027s : 0.00% optimize.opt_a.renormalize : 0.001018s : 0.00% optimize.opt_a.add_forward_monad_depend : 0.000016s : 0.00% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000047s : 0.00% optimize.opt_a.cse : 0.000090s : 0.00% optimize.opt_a.a_3 : 0.000131s : 0.00% optimize.py_interpret_to_execute_after_opt_a : 0.000023s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000244s : 0.00% optimize.convert_after_rewriter : 0.000018s : 0.00% optimize.order_py_execute_after_rewriter : 0.000007s : 0.00% optimize.mutable_eliminate : 0.000916s : 0.00% optimize.opt_b.b_1 : 0.000207s : 0.00% optimize.opt_b.b_2 : 0.000012s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000016s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000005s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000042s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000027s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000039s : 0.00% optimize.loop_unroll : 0.000626s : 0.00% optimize.opt_after_cconv.c_1 : 0.000060s : 0.00% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000011s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000004s : 0.00% optimize.opt_after_cconv.cse : 0.000031s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000020s : 0.00% optimize.tuple_transform.d_1 : 0.000059s : 0.00% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000010s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000090s : 0.00% optimize.cse_after_recomputation.cse : 0.000016s : 0.00% optimize.environ_conv : 0.000024s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000009s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000007s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000024s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000021s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000006s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000007s : 0.00% optimize.overlap_grad_flash_sp : 0.000048s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000005s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000014s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000019s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000010s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000015s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000040s : 0.00% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.000729s : 0.00% validate : 0.000084s : 0.00% backend_pass : 0.000001s : 0.00% task_emit : 50.437585s : 99.39% execute : 0.000012s : 0.00% Time group info: ------[substitution.] 0.000206 55 5.61% : 0.000012s : 2: substitution.depend_value_elim 1.50% : 0.000003s : 5: substitution.elim_not_effective 1.06% : 0.000002s : 5: substitution.fold_const_symbol 4.12% : 0.000008s : 6: substitution.graph_param_transform 55.01% : 0.000113s : 1: substitution.inline 3.12% : 0.000006s : 10: substitution.j_node_and_user_rematch 4.41% : 0.000009s : 10: substitution.remove_not_recompute_node 3.36% : 0.000007s : 2: substitution.replace_old_param 13.00% : 0.000027s : 6: substitution.updatestate_pure_node_eliminater 8.80% : 0.000018s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.197021 2 99.40% : 0.195842s : 1: type_inference.infer 0.60% : 0.001180s : 1: type_inference.specialize ------[replace.] 0.000086 1 100.00% : 0.000086s : 1: replace.inline ------[match.] 0.000112 1 100.00% : 0.000112s : 1: match.inline ------[predicate.] 0.000274 1419 1.07% : 0.000003s : 13: predicate.accumulaten_eliminater 1.33% : 0.000004s : 6: predicate.ad_related_special_op_eliminate 0.65% : 0.000002s : 12: predicate.addn_check_dump 0.98% : 0.000003s : 13: predicate.addn_zero_filter 0.71% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 2.26% : 0.000006s : 25: predicate.arithmetic_simplify 0.99% : 0.000003s : 13: predicate.cast_eliminate 0.73% : 0.000002s : 12: predicate.check_bprop_eliminate 0.68% : 0.000002s : 12: predicate.compare_switch_simplify 0.17% : 0.000000s : 6: predicate.const_output_eliminate 0.82% : 0.000002s : 12: predicate.depend_value_elim 0.77% : 0.000002s : 13: predicate.dict_get_item_const_eliminator 0.88% : 0.000002s : 13: predicate.dict_get_item_eliminator 0.75% : 0.000002s : 13: predicate.dict_set_item_eliminator 1.37% : 0.000004s : 12: predicate.dumpgradient_eliminate 0.29% : 0.000001s : 6: predicate.elim_not_effective 0.50% : 0.000001s : 6: predicate.elim_shapecalc_of_broadcastargs 1.14% : 0.000003s : 19: predicate.environ_add_const_eliminate 1.01% : 0.000003s : 19: predicate.environ_get_add_eliminate 1.16% : 0.000003s : 19: predicate.environ_get_depend_swap 1.81% : 0.000005s : 31: predicate.environ_get_eliminate 1.00% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.81% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.76% : 0.000005s : 14: predicate.float_depend_g_call 0.64% : 0.000002s : 12: predicate.float_environ_get_switch 1.01% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.20% : 0.000001s : 6: predicate.fold_const_symbol 0.78% : 0.000002s : 12: predicate.get_grad_eliminate 0.20% : 0.000001s : 6: predicate.graph_param_transform 0.72% : 0.000002s : 12: predicate.incorporate_call 0.57% : 0.000002s : 12: predicate.incorporate_call_switch 5.29% : 0.000015s : 63: predicate.inline 0.90% : 0.000002s : 12: predicate.inline_without_move 0.32% : 0.000001s : 12: predicate.j_node_and_user_rematch 1.03% : 0.000003s : 12: predicate.less_batch_normalization 1.80% : 0.000005s : 25: predicate.list_to_tuple_eliminator_ 2.34% : 0.000006s : 38: predicate.load_eliminater 1.22% : 0.000003s : 6: predicate.loop_unroll_after_grad 0.94% : 0.000003s : 17: predicate.loop_unroll_before_grad 1.66% : 0.000005s : 25: predicate.make_slice_get_slice_eliminator 0.70% : 0.000002s : 12: predicate.merge_addn 0.65% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.73% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.69% : 0.000002s : 13: predicate.minmaximum_grad 1.90% : 0.000005s : 6: predicate.mutable_eliminate 0.39% : 0.000001s : 6: predicate.opt_reshape 0.52% : 0.000001s : 6: predicate.parallel_virtual_node 2.21% : 0.000006s : 14: predicate.partial_defer_inline 1.07% : 0.000003s : 19: predicate.partial_eliminate 0.78% : 0.000002s : 13: predicate.print_const_string_wrapper 0.78% : 0.000002s : 12: predicate.reduce_all_const_elim 1.51% : 0.000004s : 13: predicate.reduce_eliminate 2.09% : 0.000006s : 38: predicate.redundant_stop_gradient_eliminater 0.67% : 0.000002s : 12: predicate.remove_not_recompute_node 1.16% : 0.000003s : 25: predicate.replace_applicator 0.61% : 0.000002s : 12: predicate.replace_old_param 0.35% : 0.000001s : 6: predicate.reset_defer_inline 0.83% : 0.000002s : 13: predicate.reshape_eliminate 0.75% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.40% : 0.000001s : 6: predicate.row_tensor_eliminate 0.90% : 0.000002s : 12: predicate.same_eliminate 0.45% : 0.000001s : 12: predicate.set_cell_output_no_recompute 1.75% : 0.000005s : 12: predicate.shard_identity_eliminate 0.91% : 0.000002s : 12: predicate.special_op_eliminate 0.84% : 0.000002s : 12: predicate.specialize_transform 1.22% : 0.000003s : 12: predicate.split_environ_get_set_with_tuple_value 0.95% : 0.000003s : 12: predicate.stack_unstack_eliminate 0.39% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.86% : 0.000002s : 14: predicate.switch_defer_inline 1.52% : 0.000004s : 26: predicate.switch_layer_defer_inline 3.27% : 0.000009s : 49: predicate.switch_simplify 0.91% : 0.000003s : 13: predicate.tile_eliminate 0.92% : 0.000003s : 13: predicate.transpose_eliminate 1.79% : 0.000005s : 25: predicate.tuple_list_convert_item_index_to_positive 1.69% : 0.000005s : 25: predicate.tuple_list_get_item_const_eliminator 1.74% : 0.000005s : 25: predicate.tuple_list_get_item_depend_reorder 3.48% : 0.000010s : 37: predicate.tuple_list_get_item_eliminator 2.04% : 0.000006s : 25: predicate.tuple_list_get_set_item_eliminator 2.51% : 0.000007s : 37: predicate.tuple_list_set_item_eliminator 1.75% : 0.000005s : 25: predicate.tuple_to_list_eliminator_ 2.08% : 0.000006s : 38: predicate.updatestate_pure_node_eliminater 3.10% : 0.000009s : 50: predicate.updatestate_useless_node_eliminater 0.43% : 0.000001s : 6: predicate.value_based_eliminate 0.82% : 0.000002s : 12: predicate.virtual_dataset_eliminate 0.82% : 0.000002s : 12: predicate.virtual_output_eliminate 0.32% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.51% : 0.000001s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000905 4 2.66% : 0.000024s : 1: func_graph_cloner_run.FuncGraphClonerGraph 97.34% : 0.000881s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 50.985460 196 0.00% : 0.000005s : 1: ForceFp32Comm 0.02% : 0.009676s : 1: add_attr 0.02% : 0.009654s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.00% : 0.000095s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.00% : 0.000225s : 1: auto_monad 0.00% : 0.000044s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.00% : 0.001006s : 1: bootstrap 0.00% : 0.000043s : 1: cconv 0.00% : 0.000005s : 1: comm_op_add_attrs 0.00% : 0.000024s : 1: control_data_broadcast_order 0.00% : 0.000024s : 1: convert_after_rewriter 0.00% : 0.000030s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000029s : 1: environ_conv 0.00% : 0.000026s : 1: event_method 0.00% : 0.000026s : 1: execute 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000011s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000010s : 1: label_micro_interleaved_index 0.00% : 0.000640s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.00% : 0.000935s : 1: mutable_eliminate 0.00% : 0.000011s : 1: offloading_packed_experts 0.00% : 0.000022s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000029s : 1: opt.transform.mutable_eliminate 0.21% : 0.105445s : 78: opt.transform.opt_a 0.00% : 0.000058s : 1: opt.transform.opt_after_cconv 0.00% : 0.000042s : 1: opt.transform.opt_after_jit_grad 0.00% : 0.000186s : 28: opt.transform.opt_b 0.00% : 0.000067s : 2: opt.transform.opt_trans_graph 0.00% : 0.000054s : 4: opt.transform.symbol_engine_opt 0.21% : 0.107777s : 1: opt_a 0.00% : 0.000169s : 1: opt_after_cconv 0.00% : 0.000745s : 1: opt_after_jit_grad 0.00% : 0.000343s : 1: opt_b 0.22% : 0.111095s : 1: optimize 0.00% : 0.000031s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000010s : 1: order_py_execute_after_rewriter 0.00% : 0.000053s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000010s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000028s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000009s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.00% : 0.000054s : 1: pre_auto_parallel 0.00% : 0.000028s : 1: py_interpret_to_execute 0.00% : 0.000026s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000023s : 1: remove_dup_value 0.00% : 0.000616s : 1: renormalize.infer 0.00% : 0.000391s : 1: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.00% : 0.000255s : 1: rewriter_after_opt_a 0.00% : 0.000074s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000007s : 1: split_matmul_comm_elemetwise 0.00% : 0.000012s : 1: swap_dp_allreduce_reducescatter 0.00% : 0.000105s : 1: symbol_engine_optimizer 98.93% : 50.437626s : 1: task_emit 0.00% : 0.000100s : 1: tuple_transform 0.39% : 0.197130s : 1: type_inference 0.00% : 0.000131s : 1: validate TotalTime = 0.326558, [24] [bootstrap]: 0.00056879 [type_inference]: 0.00477704 [event_method]: 1.918e-05 [auto_monad]: 0.00017077 [graph_reusing]: 6.75002e-06 [inline]: 3.11999e-06 [add_attr]: 0.145369, [1] [add_attr_with_inline]: 0.145357, [1] [Cycle 1]: 7.441e-05, [2] [tag_attr]: 1.812e-05 [meta_addattr_fg_expand]: 3.68999e-06 [parallel-infer-symbol]: 4.38999e-06 [pre_auto_parallel]: 4.824e-05 [insert-virtual-dataset]: 2.51998e-06 [parallel-infer-symbol-second]: 9.50007e-07 [dataset_repeat_opt]: 2.16e-06 [pipeline_split]: 1.92001e-06 [optimize]: 0.00617083, [53] [py_interpret_to_execute]: 2.428e-05 [rewriter_before_opt_a]: 6.476e-05 [opt_a]: 0.00328302, [2] [Cycle 1]: 0.00231646, [45] [expand_dump_flag]: 3.7e-06 [switch_simplify]: 2.668e-05 [loop_unroll]: 1.337e-05 [a_1]: 0.00040157 [with_stream_mark]: 2.152e-05 [recompute_prepare]: 1.038e-05 [updatestate_depend_eliminate]: 6.85002e-06 [updatestate_assign_eliminate]: 5.53997e-06 [updatestate_loads_eliminate]: 6.44999e-06 [parameter_eliminate]: 1.86e-06 [a_2]: 0.00012418 [accelerated_algorithm]: 9.68002e-06 [shard]: 2.41998e-06 [meta_shard_fg_expand]: 2.50002e-06 [shard_inline]: 8.77999e-06 [merge_send_recv]: 1.11e-05 [auto_parallel]: 8.70999e-06 [parallel]: 3.611e-05 [flash_sp]: 1.021e-05 [merge_comm]: 5.97001e-06 [allreduce_fusion]: 5.41002e-06 [matmul_add_comm_reduction]: 1.311e-05 [allreduce_slice_to_reducescatter]: 7.30011e-07 [virtual_shard_identity]: 1.229e-05 [virtual_dataset]: 8.62e-06 [get_grad_eliminate_]: 8.53001e-06 [virtual_output]: 8.58001e-06 [merge_forward]: 6.13002e-06 [cell_reuse_recompute_pass]: 1.39003e-06 [offload_activation]: 1.542e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.002e-05 [merge_recompute_call_nodes]: 1.81998e-06 [before_grad]: 1.69e-05 [set_forward_comm_id_for_comm_node_pass]: 5.84999e-06 [meta_fg_expand]: 4.21001e-06 [flash_sp_send_recv_attached]: 3.41999e-06 [receive_attached]: 2.90998e-06 [after_resolve]: 1.189e-05 [a_after_grad]: 1.274e-05 [renormalize]: 0.00100523 [add_forward_monad_depend]: 1.233e-05 [auto_monad_grad]: 2.81e-06 [auto_monad_eliminator]: 2.35e-05 [cse]: 4.094e-05 [a_3]: 6.702e-05 [Cycle 2]: 0.00095283, [45] [expand_dump_flag]: 1.67001e-06 [switch_simplify]: 1.102e-05 [loop_unroll]: 8.60999e-06 [a_1]: 0.00020729 [with_stream_mark]: 1.546e-05 [recompute_prepare]: 9.32001e-06 [updatestate_depend_eliminate]: 5.72001e-06 [updatestate_assign_eliminate]: 5.24e-06 [updatestate_loads_eliminate]: 5.61e-06 [parameter_eliminate]: 1.35999e-06 [a_2]: 0.00013209 [accelerated_algorithm]: 9.22001e-06 [shard]: 1.72999e-06 [meta_shard_fg_expand]: 2.54999e-06 [shard_inline]: 8.90001e-06 [merge_send_recv]: 8.45001e-06 [auto_parallel]: 8.48001e-06 [parallel]: 5.58002e-06 [flash_sp]: 3.71001e-06 [merge_comm]: 5.19e-06 [allreduce_fusion]: 5.42999e-06 [matmul_add_comm_reduction]: 1.007e-05 [allreduce_slice_to_reducescatter]: 3.10014e-07 [virtual_shard_identity]: 9.80002e-06 [virtual_dataset]: 8.45999e-06 [get_grad_eliminate_]: 8.05e-06 [virtual_output]: 7.97e-06 [merge_forward]: 4.80001e-06 [cell_reuse_recompute_pass]: 2.22001e-06 [offload_activation]: 1.161e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.768e-05 [merge_recompute_call_nodes]: 9.09989e-07 [before_grad]: 1.579e-05 [set_forward_comm_id_for_comm_node_pass]: 5.52001e-06 [meta_fg_expand]: 3.13998e-06 [flash_sp_send_recv_attached]: 1.53002e-06 [receive_attached]: 1.44e-06 [after_resolve]: 1.318e-05 [a_after_grad]: 1.289e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 1.61998e-06 [auto_monad_grad]: 1.09e-06 [auto_monad_eliminator]: 1.198e-05 [cse]: 4.159e-05 [a_3]: 5.668e-05 [py_interpret_to_execute_after_opt_a]: 1.465e-05 [slice_cell_reuse_recomputed_activation]: 2.57001e-06 [rewriter_after_opt_a]: 0.00019119 [convert_after_rewriter]: 1.356e-05 [order_py_execute_after_rewriter]: 7.49002e-06 [mutable_eliminate]: 0.0007408 [opt_b]: 0.00031525, [1] [Cycle 1]: 0.00030659, [7] [b_1]: 0.00019521 [b_2]: 1.188e-05 [updatestate_depend_eliminate]: 9.51e-06 [updatestate_assign_eliminate]: 4.90001e-06 [updatestate_loads_eliminate]: 4.63999e-06 [renormalize]: 1.33002e-06 [cse]: 2.777e-05 [optimize_parallel_all_gather_comm]: 2.287e-05 [overlap_param_gather]: 5.46e-06 [cconv]: 3.475e-05 [loop_unroll]: 0.00051829 [opt_after_cconv]: 0.00015386, [1] [Cycle 1]: 0.00014589, [7] [c_1]: 5.472e-05 [parameter_eliminate]: 3.23e-06 [updatestate_depend_eliminate]: 8.3e-06 [updatestate_assign_eliminate]: 5.00999e-06 [updatestate_loads_eliminate]: 4.83001e-06 [cse]: 2.326e-05 [renormalize]: 2.9002e-07 [remove_dup_value]: 1.77e-05 [tuple_transform]: 0.00010292, [1] [Cycle 1]: 9.683e-05, [4] [d_1]: 6.039e-05 [none_parameter_eliminate]: 2.11e-06 [renormalize]: 3.69997e-07 [switch_simplify]: 9.93998e-06 [partial_unused_args_eliminate]: 2.44999e-06 [add_recomputation]: 8.698e-05 [cse_after_recomputation]: 3.05e-05, [1] [Cycle 1]: 2.422e-05, [1] [cse]: 1.663e-05 [environ_conv]: 6.94999e-06 [swap_dp_allreduce_reducescatter]: 8.96002e-06 [bias_add_comm_swap]: 3.14999e-06 [label_micro_interleaved_index]: 4.89998e-06 [label_fine_grained_interleaved_index]: 3.03e-06 [merge_cast_opt]: 1.36998e-06 [slice_recompute_activation]: 2.52001e-06 [micro_interleaved_order_control]: 3.4e-06 [assign_add_opt]: 1.33002e-06 [ForceFp32Comm]: 8.50006e-07 [remove_cast_before_assign_add]: 1.05999e-06 [full_micro_interleaved_order_control]: 2.46e-06 [reorder_send_recv_between_fp_bp]: 2.76e-06 [comm_op_add_attrs]: 1.14e-06 [add_comm_op_reuse_tag]: 1.06002e-06 [interleave_split_concat_branches]: 1.42e-06 [interleave_parallel_branches]: 1.84e-06 [overlap_opt_shard_in_pipeline]: 2.84001e-06 [overlap_opt_shard_grad_in_pipeline]: 2.12999e-06 [control_data_broadcast_order]: 1.952e-05 [grouped_pairwise_exchange_alltoall]: 1.80001e-06 [offloading_packed_experts]: 5.17e-06 [overlap_recompute_and_grad_model_parallel]: 6.37001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.80001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.79e-06 [overlap_recompute_comm]: 3.09999e-06 [overlap_grad_ring_attention]: 5.74999e-06 [overlap_grad_flash_sp]: 3.168e-05 [begin_end_overlap_inline]: 5.19998e-07 [split_matmul_comm_elemetwise]: 3.18e-06 [split_layernorm_comm]: 2.14999e-06 [handle_group_info]: 1.00999e-06 [symbol_engine_optimizer]: 0.00011228, [1] [Cycle 1]: 0.00010664, [6] [build]: 5.35001e-06 [elim_shapecalc]: 1.388e-05 [elim_not_effective]: 1.963e-05 [opt_reshape]: 1.116e-05 [fold_const_symbol]: 1.669e-05 [renormalize]: 2.00002e-07 [detach_backward]: 2.38998e-06 [pipeline_parallel_scheduler]: 1.42999e-06 [auto_monad_reorder]: 3.051e-05 [get_jit_bprop_graph]: 1.57001e-06 [rewriter_after_jit_bprop_graph]: 6.09001e-06 [opt_after_jit_grad]: 0.00055994 [validate]: 6.114e-05 [backend_pass]: 1.00001e-06 [task_emit]: 0.168439 [execute]: 1.128e-05 Sums bootstrap : 0.000569s : 0.32% type_inference : 0.004777s : 2.66% event_method : 0.000019s : 0.01% auto_monad : 0.000171s : 0.09% graph_reusing : 0.000007s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000018s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000048s : 0.03% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000024s : 0.01% optimize.rewriter_before_opt_a : 0.000065s : 0.04% optimize.opt_a.expand_dump_flag : 0.000005s : 0.00% optimize.opt_a.switch_simplify : 0.000038s : 0.02% optimize.opt_a.loop_unroll : 0.000022s : 0.01% optimize.opt_a.a_1 : 0.000609s : 0.34% optimize.opt_a.with_stream_mark : 0.000037s : 0.02% optimize.opt_a.recompute_prepare : 0.000020s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000013s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000011s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000012s : 0.01% optimize.opt_a.parameter_eliminate : 0.000003s : 0.00% optimize.opt_a.a_2 : 0.000256s : 0.14% optimize.opt_a.accelerated_algorithm : 0.000019s : 0.01% optimize.opt_a.shard : 0.000004s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000005s : 0.00% optimize.opt_a.shard_inline : 0.000018s : 0.01% optimize.opt_a.merge_send_recv : 0.000020s : 0.01% optimize.opt_a.auto_parallel : 0.000017s : 0.01% optimize.opt_a.parallel : 0.000042s : 0.02% optimize.opt_a.flash_sp : 0.000014s : 0.01% optimize.opt_a.merge_comm : 0.000011s : 0.01% optimize.opt_a.allreduce_fusion : 0.000011s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000023s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000022s : 0.01% optimize.opt_a.virtual_dataset : 0.000017s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000017s : 0.01% optimize.opt_a.virtual_output : 0.000017s : 0.01% optimize.opt_a.merge_forward : 0.000011s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.00% optimize.opt_a.offload_activation : 0.000027s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000038s : 0.02% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000033s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000011s : 0.01% optimize.opt_a.meta_fg_expand : 0.000007s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.00% optimize.opt_a.receive_attached : 0.000004s : 0.00% optimize.opt_a.after_resolve : 0.000025s : 0.01% optimize.opt_a.a_after_grad : 0.000026s : 0.01% optimize.opt_a.renormalize : 0.001005s : 0.56% optimize.opt_a.add_forward_monad_depend : 0.000014s : 0.01% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000035s : 0.02% optimize.opt_a.cse : 0.000083s : 0.05% optimize.opt_a.a_3 : 0.000124s : 0.07% optimize.py_interpret_to_execute_after_opt_a : 0.000015s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000191s : 0.11% optimize.convert_after_rewriter : 0.000014s : 0.01% optimize.order_py_execute_after_rewriter : 0.000007s : 0.00% optimize.mutable_eliminate : 0.000741s : 0.41% optimize.opt_b.b_1 : 0.000195s : 0.11% optimize.opt_b.b_2 : 0.000012s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000010s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000005s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000028s : 0.02% optimize.optimize_parallel_all_gather_comm : 0.000023s : 0.01% optimize.overlap_param_gather : 0.000005s : 0.00% optimize.cconv : 0.000035s : 0.02% optimize.loop_unroll : 0.000518s : 0.29% optimize.opt_after_cconv.c_1 : 0.000055s : 0.03% optimize.opt_after_cconv.parameter_eliminate : 0.000003s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000008s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.cse : 0.000023s : 0.01% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000018s : 0.01% optimize.tuple_transform.d_1 : 0.000060s : 0.03% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000010s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000087s : 0.05% optimize.cse_after_recomputation.cse : 0.000017s : 0.01% optimize.environ_conv : 0.000007s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000009s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000002s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000003s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000020s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000005s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000006s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000006s : 0.00% optimize.overlap_grad_flash_sp : 0.000032s : 0.02% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000005s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000014s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000020s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000011s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000017s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000001s : 0.00% auto_monad_reorder : 0.000031s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.000560s : 0.31% validate : 0.000061s : 0.03% backend_pass : 0.000001s : 0.00% task_emit : 0.168439s : 93.64% execute : 0.000011s : 0.01% Time group info: ------[substitution.] 0.000183 55 4.66% : 0.000009s : 2: substitution.depend_value_elim 1.79% : 0.000003s : 5: substitution.elim_not_effective 1.31% : 0.000002s : 5: substitution.fold_const_symbol 4.37% : 0.000008s : 6: substitution.graph_param_transform 63.65% : 0.000117s : 1: substitution.inline 3.03% : 0.000006s : 10: substitution.j_node_and_user_rematch 4.53% : 0.000008s : 10: substitution.remove_not_recompute_node 1.95% : 0.000004s : 2: substitution.replace_old_param 7.87% : 0.000014s : 6: substitution.updatestate_pure_node_eliminater 6.85% : 0.000013s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.004647 2 88.39% : 0.004107s : 1: type_inference.infer 11.61% : 0.000540s : 1: type_inference.specialize ------[replace.] 0.000017 1 100.00% : 0.000017s : 1: replace.inline ------[match.] 0.000115 1 100.00% : 0.000115s : 1: match.inline ------[predicate.] 0.000235 1419 0.81% : 0.000002s : 13: predicate.accumulaten_eliminater 1.20% : 0.000003s : 6: predicate.ad_related_special_op_eliminate 0.66% : 0.000002s : 12: predicate.addn_check_dump 0.95% : 0.000002s : 13: predicate.addn_zero_filter 0.72% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 1.97% : 0.000005s : 25: predicate.arithmetic_simplify 0.86% : 0.000002s : 13: predicate.cast_eliminate 0.77% : 0.000002s : 12: predicate.check_bprop_eliminate 0.70% : 0.000002s : 12: predicate.compare_switch_simplify 0.22% : 0.000001s : 6: predicate.const_output_eliminate 0.84% : 0.000002s : 12: predicate.depend_value_elim 0.76% : 0.000002s : 13: predicate.dict_get_item_const_eliminator 0.83% : 0.000002s : 13: predicate.dict_get_item_eliminator 0.74% : 0.000002s : 13: predicate.dict_set_item_eliminator 1.16% : 0.000003s : 12: predicate.dumpgradient_eliminate 0.47% : 0.000001s : 6: predicate.elim_not_effective 0.66% : 0.000002s : 6: predicate.elim_shapecalc_of_broadcastargs 1.14% : 0.000003s : 19: predicate.environ_add_const_eliminate 1.08% : 0.000003s : 19: predicate.environ_get_add_eliminate 1.06% : 0.000003s : 19: predicate.environ_get_depend_swap 1.90% : 0.000004s : 31: predicate.environ_get_eliminate 1.05% : 0.000002s : 19: predicate.environ_get_set_eliminate 0.99% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.41% : 0.000003s : 14: predicate.float_depend_g_call 0.68% : 0.000002s : 12: predicate.float_environ_get_switch 1.03% : 0.000002s : 18: predicate.float_tuple_getitem_switch 0.21% : 0.000000s : 6: predicate.fold_const_symbol 0.80% : 0.000002s : 12: predicate.get_grad_eliminate 0.25% : 0.000001s : 6: predicate.graph_param_transform 0.80% : 0.000002s : 12: predicate.incorporate_call 0.66% : 0.000002s : 12: predicate.incorporate_call_switch 5.44% : 0.000013s : 63: predicate.inline 0.90% : 0.000002s : 12: predicate.inline_without_move 0.36% : 0.000001s : 12: predicate.j_node_and_user_rematch 1.00% : 0.000002s : 12: predicate.less_batch_normalization 1.75% : 0.000004s : 25: predicate.list_to_tuple_eliminator_ 2.27% : 0.000005s : 38: predicate.load_eliminater 1.10% : 0.000003s : 6: predicate.loop_unroll_after_grad 1.12% : 0.000003s : 17: predicate.loop_unroll_before_grad 1.72% : 0.000004s : 25: predicate.make_slice_get_slice_eliminator 0.92% : 0.000002s : 12: predicate.merge_addn 0.76% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.76% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.74% : 0.000002s : 13: predicate.minmaximum_grad 1.66% : 0.000004s : 6: predicate.mutable_eliminate 0.79% : 0.000002s : 6: predicate.opt_reshape 0.77% : 0.000002s : 6: predicate.parallel_virtual_node 1.41% : 0.000003s : 14: predicate.partial_defer_inline 1.22% : 0.000003s : 19: predicate.partial_eliminate 0.84% : 0.000002s : 13: predicate.print_const_string_wrapper 0.72% : 0.000002s : 12: predicate.reduce_all_const_elim 1.03% : 0.000002s : 13: predicate.reduce_eliminate 2.26% : 0.000005s : 38: predicate.redundant_stop_gradient_eliminater 0.53% : 0.000001s : 12: predicate.remove_not_recompute_node 1.11% : 0.000003s : 25: predicate.replace_applicator 0.67% : 0.000002s : 12: predicate.replace_old_param 0.39% : 0.000001s : 6: predicate.reset_defer_inline 0.78% : 0.000002s : 13: predicate.reshape_eliminate 0.89% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.61% : 0.000001s : 6: predicate.row_tensor_eliminate 1.21% : 0.000003s : 12: predicate.same_eliminate 0.52% : 0.000001s : 12: predicate.set_cell_output_no_recompute 1.28% : 0.000003s : 12: predicate.shard_identity_eliminate 1.20% : 0.000003s : 12: predicate.special_op_eliminate 0.87% : 0.000002s : 12: predicate.specialize_transform 1.27% : 0.000003s : 12: predicate.split_environ_get_set_with_tuple_value 0.91% : 0.000002s : 12: predicate.stack_unstack_eliminate 0.53% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.93% : 0.000002s : 14: predicate.switch_defer_inline 1.80% : 0.000004s : 26: predicate.switch_layer_defer_inline 3.60% : 0.000008s : 49: predicate.switch_simplify 0.82% : 0.000002s : 13: predicate.tile_eliminate 0.86% : 0.000002s : 13: predicate.transpose_eliminate 1.66% : 0.000004s : 25: predicate.tuple_list_convert_item_index_to_positive 1.66% : 0.000004s : 25: predicate.tuple_list_get_item_const_eliminator 1.45% : 0.000003s : 25: predicate.tuple_list_get_item_depend_reorder 2.90% : 0.000007s : 37: predicate.tuple_list_get_item_eliminator 1.51% : 0.000004s : 25: predicate.tuple_list_get_set_item_eliminator 2.58% : 0.000006s : 37: predicate.tuple_list_set_item_eliminator 1.60% : 0.000004s : 25: predicate.tuple_to_list_eliminator_ 2.39% : 0.000006s : 38: predicate.updatestate_pure_node_eliminater 3.40% : 0.000008s : 50: predicate.updatestate_useless_node_eliminater 0.62% : 0.000001s : 6: predicate.value_based_eliminate 0.83% : 0.000002s : 12: predicate.virtual_dataset_eliminate 0.84% : 0.000002s : 12: predicate.virtual_output_eliminate 0.34% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.55% : 0.000001s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000327 4 7.01% : 0.000023s : 1: func_graph_cloner_run.FuncGraphClonerGraph 92.99% : 0.000305s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.480626 196 0.00% : 0.000004s : 1: ForceFp32Comm 30.25% : 0.145377s : 1: add_attr 30.24% : 0.145361s : 1: add_attr_with_inline 0.00% : 0.000005s : 1: add_comm_op_reuse_tag 0.02% : 0.000093s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.04% : 0.000182s : 1: auto_monad 0.01% : 0.000035s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.13% : 0.000610s : 1: bootstrap 0.01% : 0.000040s : 1: cconv 0.00% : 0.000006s : 1: comm_op_add_attrs 0.00% : 0.000024s : 1: control_data_broadcast_order 0.00% : 0.000018s : 1: convert_after_rewriter 0.01% : 0.000034s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000011s : 1: environ_conv 0.01% : 0.000027s : 1: event_method 0.00% : 0.000020s : 1: execute 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000012s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000005s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000006s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000009s : 1: label_micro_interleaved_index 0.11% : 0.000529s : 1: loop_unroll 0.00% : 0.000006s : 1: merge_cast_opt 0.00% : 0.000007s : 1: micro_interleaved_order_control 0.16% : 0.000753s : 1: mutable_eliminate 0.00% : 0.000010s : 1: offloading_packed_experts 0.00% : 0.000018s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000022s : 1: opt.transform.mutable_eliminate 0.25% : 0.001209s : 78: opt.transform.opt_a 0.01% : 0.000053s : 1: opt.transform.opt_after_cconv 0.01% : 0.000035s : 1: opt.transform.opt_after_jit_grad 0.04% : 0.000176s : 28: opt.transform.opt_b 0.01% : 0.000068s : 2: opt.transform.opt_trans_graph 0.01% : 0.000058s : 4: opt.transform.symbol_engine_opt 0.68% : 0.003287s : 1: opt_a 0.03% : 0.000158s : 1: opt_after_cconv 0.12% : 0.000571s : 1: opt_after_jit_grad 0.07% : 0.000320s : 1: opt_b 1.29% : 0.006177s : 1: optimize 0.01% : 0.000027s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000012s : 1: order_py_execute_after_rewriter 0.01% : 0.000036s : 1: overlap_grad_flash_sp 0.00% : 0.000006s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000010s : 1: overlap_grad_ring_attention 0.00% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000007s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000009s : 1: overlap_param_gather 0.00% : 0.000008s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000011s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000007s : 1: overlap_recompute_comm 0.00% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000053s : 1: pre_auto_parallel 0.01% : 0.000028s : 1: py_interpret_to_execute 0.00% : 0.000019s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.00% : 0.000022s : 1: remove_dup_value 0.13% : 0.000619s : 1: renormalize.infer 0.08% : 0.000376s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.04% : 0.000199s : 1: rewriter_after_opt_a 0.01% : 0.000070s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000006s : 1: split_layernorm_comm 0.00% : 0.000008s : 1: split_matmul_comm_elemetwise 0.00% : 0.000013s : 1: swap_dp_allreduce_reducescatter 0.02% : 0.000116s : 1: symbol_engine_optimizer 35.05% : 0.168463s : 1: task_emit 0.02% : 0.000107s : 1: tuple_transform 1.00% : 0.004810s : 1: type_inference 0.02% : 0.000107s : 1: validate TotalTime = 0.974208, [24] [bootstrap]: 0.110718 [type_inference]: 0.148871 [event_method]: 9.627e-05 [auto_monad]: 0.00033196 [graph_reusing]: 1.149e-05 [inline]: 2.65002e-06 [add_attr]: 0.00484025, [1] [add_attr_with_inline]: 0.00482344, [1] [Cycle 1]: 0.0001598, [2] [tag_attr]: 9.791e-05 [meta_addattr_fg_expand]: 1.071e-05 [parallel-infer-symbol]: 4.00998e-06 [pre_auto_parallel]: 7.231e-05 [insert-virtual-dataset]: 2.34001e-06 [parallel-infer-symbol-second]: 6.80011e-07 [dataset_repeat_opt]: 2.14999e-06 [pipeline_split]: 2.14999e-06 [optimize]: 0.558512, [53] [py_interpret_to_execute]: 5.964e-05 [rewriter_before_opt_a]: 0.00018131 [opt_a]: 0.555055, [3] [Cycle 1]: 0.394162, [45] [expand_dump_flag]: 9.71e-06 [switch_simplify]: 7.542e-05 [loop_unroll]: 8.616e-05 [a_1]: 0.00280588 [with_stream_mark]: 5.368e-05 [recompute_prepare]: 4.228e-05 [updatestate_depend_eliminate]: 1.656e-05 [updatestate_assign_eliminate]: 1.296e-05 [updatestate_loads_eliminate]: 1.249e-05 [parameter_eliminate]: 3.88001e-06 [a_2]: 0.00035311 [accelerated_algorithm]: 4.948e-05 [shard]: 2.27999e-06 [meta_shard_fg_expand]: 9.00999e-06 [shard_inline]: 2.151e-05 [merge_send_recv]: 2.334e-05 [auto_parallel]: 2.02e-05 [parallel]: 5.587e-05 [flash_sp]: 2.128e-05 [merge_comm]: 1.748e-05 [allreduce_fusion]: 1.467e-05 [matmul_add_comm_reduction]: 4.958e-05 [allreduce_slice_to_reducescatter]: 9.49978e-07 [virtual_shard_identity]: 3.524e-05 [virtual_dataset]: 2.498e-05 [get_grad_eliminate_]: 2.372e-05 [virtual_output]: 2.321e-05 [merge_forward]: 1.617e-05 [cell_reuse_recompute_pass]: 2.01e-06 [offload_activation]: 3.119e-05 [cell_reuse_handle_not_recompute_node_pass]: 5.318e-05 [merge_recompute_call_nodes]: 1.99e-06 [before_grad]: 4.549e-05 [set_forward_comm_id_for_comm_node_pass]: 1.511e-05 [meta_fg_expand]: 0.313997 [flash_sp_send_recv_attached]: 7.65e-06 [receive_attached]: 2.62001e-06 [after_resolve]: 0.0001654 [a_after_grad]: 0.00022081 [renormalize]: 0.0728644 [add_forward_monad_depend]: 2.95e-05 [auto_monad_grad]: 1.816e-05 [auto_monad_eliminator]: 0.00016538 [cse]: 0.00035206 [a_3]: 0.00175314 [Cycle 2]: 0.159568, [45] [expand_dump_flag]: 6.26e-06 [switch_simplify]: 0.0001135 [loop_unroll]: 0.00010511 [a_1]: 0.00344969 [with_stream_mark]: 3.626e-05 [recompute_prepare]: 2.033e-05 [updatestate_depend_eliminate]: 3.823e-05 [updatestate_assign_eliminate]: 7.26999e-06 [updatestate_loads_eliminate]: 5.93998e-06 [parameter_eliminate]: 2.56e-06 [a_2]: 0.00020508 [accelerated_algorithm]: 2.025e-05 [shard]: 2.39999e-06 [meta_shard_fg_expand]: 5.15999e-06 [shard_inline]: 1.294e-05 [merge_send_recv]: 1.477e-05 [auto_parallel]: 1.374e-05 [parallel]: 1.287e-05 [flash_sp]: 5.25001e-06 [merge_comm]: 6.86999e-06 [allreduce_fusion]: 6.41e-06 [matmul_add_comm_reduction]: 1.542e-05 [allreduce_slice_to_reducescatter]: 1.27e-06 [virtual_shard_identity]: 1.525e-05 [virtual_dataset]: 1.239e-05 [get_grad_eliminate_]: 1.146e-05 [virtual_output]: 1.1e-05 [merge_forward]: 7.23e-06 [cell_reuse_recompute_pass]: 1.77999e-06 [offload_activation]: 1.639e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.507e-05 [merge_recompute_call_nodes]: 2.01998e-06 [before_grad]: 2.088e-05 [set_forward_comm_id_for_comm_node_pass]: 7.38999e-06 [meta_fg_expand]: 0.00012402 [flash_sp_send_recv_attached]: 2.29001e-06 [receive_attached]: 3.01001e-06 [after_resolve]: 2.365e-05 [a_after_grad]: 1.98e-05 [renormalize]: 0.154489 [add_forward_monad_depend]: 1.129e-05 [auto_monad_grad]: 2.73e-06 [auto_monad_eliminator]: 3.145e-05 [cse]: 0.00015129 [a_3]: 0.00010948 [Cycle 3]: 0.0012986, [45] [expand_dump_flag]: 2.70002e-06 [switch_simplify]: 1.485e-05 [loop_unroll]: 1.183e-05 [a_1]: 0.0003318 [with_stream_mark]: 2.24e-05 [recompute_prepare]: 1.243e-05 [updatestate_depend_eliminate]: 8.42e-06 [updatestate_assign_eliminate]: 7.41001e-06 [updatestate_loads_eliminate]: 6.02999e-06 [parameter_eliminate]: 2.01e-06 [a_2]: 0.00015852 [accelerated_algorithm]: 1.902e-05 [shard]: 2.16e-06 [meta_shard_fg_expand]: 3.41999e-06 [shard_inline]: 1.275e-05 [merge_send_recv]: 1.304e-05 [auto_parallel]: 1.39e-05 [parallel]: 1.061e-05 [flash_sp]: 1.79e-06 [merge_comm]: 7.46999e-06 [allreduce_fusion]: 6.45002e-06 [matmul_add_comm_reduction]: 1.562e-05 [allreduce_slice_to_reducescatter]: 8.99978e-07 [virtual_shard_identity]: 1.385e-05 [virtual_dataset]: 1.258e-05 [get_grad_eliminate_]: 1.112e-05 [virtual_output]: 1.057e-05 [merge_forward]: 7.55e-06 [cell_reuse_recompute_pass]: 3.18e-06 [offload_activation]: 1.607e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.462e-05 [merge_recompute_call_nodes]: 1.76e-06 [before_grad]: 2.108e-05 [set_forward_comm_id_for_comm_node_pass]: 7.88999e-06 [meta_fg_expand]: 5.53997e-06 [flash_sp_send_recv_attached]: 1.95001e-06 [receive_attached]: 2.62001e-06 [after_resolve]: 2.065e-05 [a_after_grad]: 1.838e-05 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 2.14e-06 [auto_monad_grad]: 2.06e-06 [auto_monad_eliminator]: 1.924e-05 [cse]: 3.535e-05 [a_3]: 0.00014808 [py_interpret_to_execute_after_opt_a]: 2.463e-05 [slice_cell_reuse_recomputed_activation]: 2.58e-06 [rewriter_after_opt_a]: 8.911e-05 [convert_after_rewriter]: 1.189e-05 [order_py_execute_after_rewriter]: 9.84001e-06 [mutable_eliminate]: 0.00086263 [opt_b]: 0.00043183, [1] [Cycle 1]: 0.00042007, [7] [b_1]: 0.00026793 [b_2]: 1.491e-05 [updatestate_depend_eliminate]: 1.543e-05 [updatestate_assign_eliminate]: 6.92002e-06 [updatestate_loads_eliminate]: 6.71e-06 [renormalize]: 1.45001e-06 [cse]: 5.746e-05 [optimize_parallel_all_gather_comm]: 3.548e-05 [overlap_param_gather]: 2.33002e-06 [cconv]: 4.38e-05 [loop_unroll]: 0.00057278 [opt_after_cconv]: 0.00020499, [1] [Cycle 1]: 0.00019678, [7] [c_1]: 7.698e-05 [parameter_eliminate]: 6.94999e-06 [updatestate_depend_eliminate]: 1.084e-05 [updatestate_assign_eliminate]: 6.93e-06 [updatestate_loads_eliminate]: 6.15002e-06 [cse]: 4.419e-05 [renormalize]: 6.50005e-07 [remove_dup_value]: 2.346e-05 [tuple_transform]: 0.00013567, [1] [Cycle 1]: 0.00012979, [4] [d_1]: 9.108e-05 [none_parameter_eliminate]: 2.29999e-06 [renormalize]: 2.30008e-07 [switch_simplify]: 1.266e-05 [partial_unused_args_eliminate]: 2.09e-06 [add_recomputation]: 0.00012012 [cse_after_recomputation]: 4.718e-05, [1] [Cycle 1]: 4.115e-05, [1] [cse]: 3.297e-05 [environ_conv]: 1.591e-05 [swap_dp_allreduce_reducescatter]: 1.11e-05 [bias_add_comm_swap]: 2.86e-06 [label_micro_interleaved_index]: 6.55997e-06 [label_fine_grained_interleaved_index]: 3.65e-06 [merge_cast_opt]: 1.53002e-06 [slice_recompute_activation]: 2.78998e-06 [micro_interleaved_order_control]: 2.81999e-06 [assign_add_opt]: 1.40001e-06 [ForceFp32Comm]: 8.70001e-07 [remove_cast_before_assign_add]: 1.60999e-06 [full_micro_interleaved_order_control]: 2.37999e-06 [reorder_send_recv_between_fp_bp]: 3.19001e-06 [comm_op_add_attrs]: 1.19998e-06 [add_comm_op_reuse_tag]: 1.05999e-06 [interleave_split_concat_branches]: 1.36998e-06 [interleave_parallel_branches]: 1.22e-06 [overlap_opt_shard_in_pipeline]: 1.45999e-06 [overlap_opt_shard_grad_in_pipeline]: 2.17999e-06 [control_data_broadcast_order]: 2.533e-05 [grouped_pairwise_exchange_alltoall]: 1.80001e-06 [offloading_packed_experts]: 6.97002e-06 [overlap_recompute_and_grad_model_parallel]: 6.99001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.65001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.86e-06 [overlap_recompute_comm]: 2.96001e-06 [overlap_grad_ring_attention]: 6.66e-06 [overlap_grad_flash_sp]: 4.038e-05 [begin_end_overlap_inline]: 5.50004e-07 [split_matmul_comm_elemetwise]: 2.63e-06 [split_layernorm_comm]: 1.99e-06 [handle_group_info]: 1.00001e-06 [symbol_engine_optimizer]: 0.00015169, [1] [Cycle 1]: 0.00014548, [6] [build]: 3.022e-05 [elim_shapecalc]: 2.009e-05 [elim_not_effective]: 2.485e-05 [opt_reshape]: 1.2e-05 [fold_const_symbol]: 1.986e-05 [renormalize]: 2.50002e-07 [detach_backward]: 2.35002e-06 [pipeline_parallel_scheduler]: 1.68002e-06 [auto_monad_reorder]: 4.338e-05 [get_jit_bprop_graph]: 2.66e-06 [rewriter_after_jit_bprop_graph]: 7.80998e-06 [opt_after_jit_grad]: 0.00071511 [validate]: 8.034e-05 [backend_pass]: 1.14e-06 [task_emit]: 0.149525 [execute]: 1.317e-05 Sums bootstrap : 0.110718s : 11.44% type_inference : 0.148871s : 15.39% event_method : 0.000096s : 0.01% auto_monad : 0.000332s : 0.03% graph_reusing : 0.000011s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000098s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000011s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000072s : 0.01% insert-virtual-dataset : 0.000002s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000060s : 0.01% optimize.rewriter_before_opt_a : 0.000181s : 0.02% optimize.opt_a.expand_dump_flag : 0.000019s : 0.00% optimize.opt_a.switch_simplify : 0.000204s : 0.02% optimize.opt_a.loop_unroll : 0.000203s : 0.02% optimize.opt_a.a_1 : 0.006587s : 0.68% optimize.opt_a.with_stream_mark : 0.000112s : 0.01% optimize.opt_a.recompute_prepare : 0.000075s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000063s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000028s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000024s : 0.00% optimize.opt_a.parameter_eliminate : 0.000008s : 0.00% optimize.opt_a.a_2 : 0.000717s : 0.07% optimize.opt_a.accelerated_algorithm : 0.000089s : 0.01% optimize.opt_a.shard : 0.000007s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000018s : 0.00% optimize.opt_a.shard_inline : 0.000047s : 0.00% optimize.opt_a.merge_send_recv : 0.000051s : 0.01% optimize.opt_a.auto_parallel : 0.000048s : 0.00% optimize.opt_a.parallel : 0.000079s : 0.01% optimize.opt_a.flash_sp : 0.000028s : 0.00% optimize.opt_a.merge_comm : 0.000032s : 0.00% optimize.opt_a.allreduce_fusion : 0.000028s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000081s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000003s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000064s : 0.01% optimize.opt_a.virtual_dataset : 0.000050s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000046s : 0.00% optimize.opt_a.virtual_output : 0.000045s : 0.00% optimize.opt_a.merge_forward : 0.000031s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000007s : 0.00% optimize.opt_a.offload_activation : 0.000064s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000103s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000006s : 0.00% optimize.opt_a.before_grad : 0.000087s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000030s : 0.00% optimize.opt_a.meta_fg_expand : 0.314126s : 32.47% optimize.opt_a.flash_sp_send_recv_attached : 0.000012s : 0.00% optimize.opt_a.receive_attached : 0.000008s : 0.00% optimize.opt_a.after_resolve : 0.000210s : 0.02% optimize.opt_a.a_after_grad : 0.000259s : 0.03% optimize.opt_a.renormalize : 0.227354s : 23.50% optimize.opt_a.add_forward_monad_depend : 0.000043s : 0.00% optimize.opt_a.auto_monad_grad : 0.000023s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000216s : 0.02% optimize.opt_a.cse : 0.000539s : 0.06% optimize.opt_a.a_3 : 0.002011s : 0.21% optimize.py_interpret_to_execute_after_opt_a : 0.000025s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000089s : 0.01% optimize.convert_after_rewriter : 0.000012s : 0.00% optimize.order_py_execute_after_rewriter : 0.000010s : 0.00% optimize.mutable_eliminate : 0.000863s : 0.09% optimize.opt_b.b_1 : 0.000268s : 0.03% optimize.opt_b.b_2 : 0.000015s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000015s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000007s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000007s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000057s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000035s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000044s : 0.00% optimize.loop_unroll : 0.000573s : 0.06% optimize.opt_after_cconv.c_1 : 0.000077s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000011s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.cse : 0.000044s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000023s : 0.00% optimize.tuple_transform.d_1 : 0.000091s : 0.01% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000013s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000120s : 0.01% optimize.cse_after_recomputation.cse : 0.000033s : 0.00% optimize.environ_conv : 0.000016s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000011s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000007s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000004s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000002s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000025s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000007s : 0.00% optimize.overlap_grad_flash_sp : 0.000040s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000030s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000020s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000025s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000012s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000020s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000043s : 0.00% get_jit_bprop_graph : 0.000003s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.000715s : 0.07% validate : 0.000080s : 0.01% backend_pass : 0.000001s : 0.00% task_emit : 0.149525s : 15.46% execute : 0.000013s : 0.00% Time group info: ------[substitution.] 0.002968 310 0.64% : 0.000019s : 4: substitution.depend_value_elim 0.12% : 0.000004s : 7: substitution.elim_not_effective 0.26% : 0.000008s : 7: substitution.float_depend_g_call 0.16% : 0.000005s : 2: substitution.float_tuple_getitem_switch 0.10% : 0.000003s : 7: substitution.fold_const_symbol 33.27% : 0.000987s : 4: substitution.getattr_setattr_resolve 0.34% : 0.000010s : 9: substitution.graph_param_transform 0.11% : 0.000003s : 2: substitution.incorporate_call 0.07% : 0.000002s : 2: substitution.incorporate_call_switch 44.02% : 0.001307s : 24: substitution.inline 1.74% : 0.000052s : 5: substitution.inline_without_move 0.53% : 0.000016s : 28: substitution.j_node_and_user_rematch 0.87% : 0.000026s : 3: substitution.less_batch_normalization 0.60% : 0.000018s : 9: substitution.minmaximum_grad 0.31% : 0.000009s : 7: substitution.partial_eliminate 0.69% : 0.000020s : 28: substitution.remove_not_recompute_node 3.74% : 0.000111s : 38: substitution.replace_applicator 0.76% : 0.000023s : 19: substitution.replace_old_param 0.17% : 0.000005s : 1: substitution.set_cell_output_no_recompute 1.27% : 0.000038s : 9: substitution.tuple_list_convert_item_index_to_positive 0.62% : 0.000018s : 9: substitution.tuple_list_get_item_const_eliminator 0.83% : 0.000025s : 9: substitution.tuple_list_get_item_depend_reorder 5.04% : 0.000150s : 31: substitution.tuple_list_get_item_eliminator 0.81% : 0.000024s : 9: substitution.tuple_list_get_set_item_eliminator 1.08% : 0.000032s : 16: substitution.updatestate_pure_node_eliminater 1.86% : 0.000055s : 21: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.148745 2 98.50% : 0.146515s : 1: type_inference.infer 1.50% : 0.002229s : 1: type_inference.specialize ------[replace.] 0.000701 49 10.68% : 0.000075s : 3: replace.getattr_setattr_resolve 37.20% : 0.000261s : 24: replace.inline 14.24% : 0.000100s : 1: replace.replace_applicator 31.15% : 0.000218s : 20: replace.tuple_list_get_item_eliminator 6.73% : 0.000047s : 1: replace.updatestate_useless_node_eliminater ------[match.] 0.002353 49 39.65% : 0.000933s : 3: match.getattr_setattr_resolve 54.63% : 0.001285s : 24: match.inline 0.80% : 0.000019s : 1: match.replace_applicator 4.14% : 0.000097s : 20: match.tuple_list_get_item_eliminator 0.78% : 0.000018s : 1: match.updatestate_useless_node_eliminater ------[predicate.] 0.001396 8683 0.86% : 0.000012s : 85: predicate.accumulaten_eliminater 0.33% : 0.000005s : 9: predicate.ad_related_special_op_eliminate 0.41% : 0.000006s : 37: predicate.addn_check_dump 0.87% : 0.000012s : 85: predicate.addn_zero_filter 0.78% : 0.000011s : 85: predicate.adjust_all_reduce_mul_add 1.65% : 0.000023s : 122: predicate.arithmetic_simplify 0.85% : 0.000012s : 85: predicate.cast_eliminate 2.50% : 0.000035s : 260: predicate.check_bprop_eliminate 0.37% : 0.000005s : 37: predicate.compare_switch_simplify 0.05% : 0.000001s : 9: predicate.const_output_eliminate 0.40% : 0.000006s : 37: predicate.depend_value_elim 0.88% : 0.000012s : 85: predicate.dict_get_item_const_eliminator 0.95% : 0.000013s : 85: predicate.dict_get_item_eliminator 0.85% : 0.000012s : 85: predicate.dict_set_item_eliminator 0.28% : 0.000004s : 18: predicate.dumpgradient_eliminate 0.08% : 0.000001s : 9: predicate.elim_not_effective 0.12% : 0.000002s : 9: predicate.elim_shapecalc_of_broadcastargs 1.49% : 0.000021s : 94: predicate.environ_add_const_eliminate 0.87% : 0.000012s : 94: predicate.environ_get_add_eliminate 3.46% : 0.000048s : 94: predicate.environ_get_depend_swap 1.31% : 0.000018s : 131: predicate.environ_get_eliminate 0.90% : 0.000013s : 94: predicate.environ_get_set_eliminate 1.28% : 0.000018s : 130: predicate.exchange_switch_depend_value 1.94% : 0.000027s : 130: predicate.float_depend_g_call 0.38% : 0.000005s : 37: predicate.float_environ_get_switch 0.45% : 0.000006s : 46: predicate.float_tuple_getitem_switch 0.06% : 0.000001s : 9: predicate.fold_const_symbol 0.44% : 0.000006s : 37: predicate.get_grad_eliminate 0.53% : 0.000007s : 20: predicate.getattr_setattr_resolve 0.07% : 0.000001s : 9: predicate.graph_param_transform 0.38% : 0.000005s : 37: predicate.incorporate_call 0.37% : 0.000005s : 37: predicate.incorporate_call_switch 4.29% : 0.000060s : 307: predicate.inline 1.60% : 0.000022s : 111: predicate.inline_without_move 0.19% : 0.000003s : 37: predicate.j_node_and_user_rematch 0.59% : 0.000008s : 37: predicate.less_batch_normalization 1.55% : 0.000022s : 123: predicate.list_to_tuple_eliminator_ 2.00% : 0.000028s : 208: predicate.load_eliminater 0.29% : 0.000004s : 9: predicate.loop_unroll_after_grad 1.98% : 0.000028s : 197: predicate.loop_unroll_before_grad 1.18% : 0.000016s : 103: predicate.make_slice_get_slice_eliminator 0.40% : 0.000006s : 37: predicate.merge_addn 2.50% : 0.000035s : 256: predicate.micro_step_allgather_replace 2.40% : 0.000033s : 256: predicate.mini_step_allgather_replace 0.87% : 0.000012s : 85: predicate.minmaximum_grad 0.28% : 0.000004s : 9: predicate.mutable_eliminate 0.11% : 0.000002s : 9: predicate.opt_reshape 0.15% : 0.000002s : 9: predicate.parallel_virtual_node 2.21% : 0.000031s : 130: predicate.partial_defer_inline 1.28% : 0.000018s : 114: predicate.partial_eliminate 0.83% : 0.000012s : 85: predicate.print_const_string_wrapper 0.43% : 0.000006s : 37: predicate.reduce_all_const_elim 1.16% : 0.000016s : 85: predicate.reduce_eliminate 2.09% : 0.000029s : 208: predicate.redundant_stop_gradient_eliminater 0.27% : 0.000004s : 37: predicate.remove_not_recompute_node 2.36% : 0.000033s : 363: predicate.replace_applicator 0.76% : 0.000011s : 111: predicate.replace_old_param 0.15% : 0.000002s : 9: predicate.reset_defer_inline 0.94% : 0.000013s : 85: predicate.reshape_eliminate 2.57% : 0.000036s : 256: predicate.row_tensor_add_zeros_like 0.12% : 0.000002s : 9: predicate.row_tensor_eliminate 2.67% : 0.000037s : 260: predicate.same_eliminate 0.27% : 0.000004s : 39: predicate.set_cell_output_no_recompute 0.76% : 0.000011s : 37: predicate.shard_identity_eliminate 0.25% : 0.000003s : 18: predicate.special_op_eliminate 0.43% : 0.000006s : 37: predicate.specialize_transform 2.66% : 0.000037s : 256: predicate.split_environ_get_set_with_tuple_value 1.40% : 0.000020s : 111: predicate.stack_unstack_eliminate 0.12% : 0.000002s : 9: predicate.switch_call_monad_eliminater 1.44% : 0.000020s : 130: predicate.switch_defer_inline 3.78% : 0.000053s : 390: predicate.switch_layer_defer_inline 3.99% : 0.000056s : 373: predicate.switch_simplify 0.94% : 0.000013s : 85: predicate.tile_eliminate 0.89% : 0.000012s : 85: predicate.transpose_eliminate 1.25% : 0.000017s : 103: predicate.tuple_list_convert_item_index_to_positive 1.15% : 0.000016s : 103: predicate.tuple_list_get_item_const_eliminator 1.19% : 0.000017s : 103: predicate.tuple_list_get_item_depend_reorder 2.31% : 0.000032s : 160: predicate.tuple_list_get_item_eliminator 1.18% : 0.000017s : 103: predicate.tuple_list_get_set_item_eliminator 1.67% : 0.000023s : 140: predicate.tuple_list_set_item_eliminator 1.42% : 0.000020s : 123: predicate.tuple_to_list_eliminator_ 2.10% : 0.000029s : 208: predicate.updatestate_pure_node_eliminater 5.21% : 0.000073s : 246: predicate.updatestate_useless_node_eliminater 0.13% : 0.000002s : 9: predicate.value_based_eliminate 0.41% : 0.000006s : 37: predicate.virtual_dataset_eliminate 0.42% : 0.000006s : 37: predicate.virtual_output_eliminate 0.12% : 0.000002s : 9: predicate.virtual_view_grad_eliminate 0.13% : 0.000002s : 9: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.004971 53 53.61% : 0.002665s : 23: func_graph_cloner_run.FuncGraphClonerGraph 46.39% : 0.002306s : 30: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.777045 247 0.00% : 0.000004s : 1: ForceFp32Comm 0.27% : 0.004848s : 1: add_attr 0.27% : 0.004828s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000126s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.02% : 0.000346s : 1: auto_monad 0.00% : 0.000048s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 6.23% : 0.110767s : 1: bootstrap 0.00% : 0.000048s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000029s : 1: control_data_broadcast_order 0.00% : 0.000016s : 1: convert_after_rewriter 0.00% : 0.000051s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000019s : 1: environ_conv 0.01% : 0.000106s : 1: event_method 0.00% : 0.000022s : 1: execute 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000017s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000006s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000010s : 1: label_micro_interleaved_index 0.03% : 0.000585s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.05% : 0.000878s : 1: mutable_eliminate 0.00% : 0.000010s : 1: offloading_packed_experts 0.00% : 0.000022s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000027s : 1: opt.transform.mutable_eliminate 0.59% : 0.010567s : 125: opt.transform.opt_a 0.00% : 0.000075s : 1: opt.transform.opt_after_cconv 0.00% : 0.000054s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000243s : 28: opt.transform.opt_b 0.06% : 0.001136s : 2: opt.transform.opt_resolve 0.01% : 0.000100s : 2: opt.transform.opt_trans_graph 0.00% : 0.000072s : 4: opt.transform.symbol_engine_opt 31.23% : 0.555060s : 1: opt_a 0.01% : 0.000209s : 1: opt_after_cconv 0.04% : 0.000730s : 1: opt_after_jit_grad 0.02% : 0.000437s : 1: opt_b 31.43% : 0.558519s : 1: optimize 0.00% : 0.000040s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000013s : 1: order_py_execute_after_rewriter 0.00% : 0.000045s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000010s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000006s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000010s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.00% : 0.000077s : 1: pre_auto_parallel 0.00% : 0.000065s : 1: py_interpret_to_execute 0.00% : 0.000029s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.00% : 0.000028s : 1: remove_dup_value 12.44% : 0.221010s : 2: renormalize.infer 0.35% : 0.006307s : 2: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000011s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000095s : 1: rewriter_after_opt_a 0.01% : 0.000187s : 1: rewriter_before_opt_a 0.00% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000015s : 1: swap_dp_allreduce_reducescatter 0.01% : 0.000155s : 1: symbol_engine_optimizer 8.42% : 0.149550s : 1: task_emit 0.01% : 0.000139s : 1: tuple_transform 8.38% : 0.148912s : 1: type_inference 0.01% : 0.000139s : 1: validate . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[0-shape0] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[0-shape0],max_mem:4.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:41:30.179.289 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.329824, [24] [bootstrap]: 0.00069014 [type_inference]: 0.109019 [event_method]: 2.846e-05 [auto_monad]: 0.00025212 [graph_reusing]: 6.79001e-06 [inline]: 3.26001e-06 [add_attr]: 0.00727641, [1] [add_attr_with_inline]: 0.00725798, [1] [Cycle 1]: 7.319e-05, [2] [tag_attr]: 1.926e-05 [meta_addattr_fg_expand]: 4e-06 [parallel-infer-symbol]: 4.63001e-06 [pre_auto_parallel]: 3.83e-05 [insert-virtual-dataset]: 2.79999e-06 [parallel-infer-symbol-second]: 6.50005e-07 [dataset_repeat_opt]: 2.17001e-06 [pipeline_split]: 1.72001e-06 [optimize]: 0.109121, [53] [py_interpret_to_execute]: 2.82e-05 [rewriter_before_opt_a]: 0.00011475 [opt_a]: 0.00468559, [2] [Cycle 1]: 0.00334766, [45] [expand_dump_flag]: 5.99999e-06 [switch_simplify]: 2.848e-05 [loop_unroll]: 1.611e-05 [a_1]: 0.0007714 [with_stream_mark]: 2.659e-05 [recompute_prepare]: 1.891e-05 [updatestate_depend_eliminate]: 9.29998e-06 [updatestate_assign_eliminate]: 3.756e-05 [updatestate_loads_eliminate]: 1.05e-05 [parameter_eliminate]: 2.51e-06 [a_2]: 0.00019228 [accelerated_algorithm]: 1.407e-05 [shard]: 3.14999e-06 [meta_shard_fg_expand]: 2.26998e-06 [shard_inline]: 9.32999e-06 [merge_send_recv]: 1.141e-05 [auto_parallel]: 1.764e-05 [parallel]: 3.546e-05 [flash_sp]: 1.33e-05 [merge_comm]: 8.51002e-06 [allreduce_fusion]: 8.482e-05 [matmul_add_comm_reduction]: 2.631e-05 [allreduce_slice_to_reducescatter]: 9.39996e-07 [virtual_shard_identity]: 1.886e-05 [virtual_dataset]: 1.934e-05 [get_grad_eliminate_]: 1.893e-05 [virtual_output]: 2.398e-05 [merge_forward]: 9.71998e-06 [cell_reuse_recompute_pass]: 1.70001e-06 [offload_activation]: 2.647e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.075e-05 [merge_recompute_call_nodes]: 1.67001e-06 [before_grad]: 2.054e-05 [set_forward_comm_id_for_comm_node_pass]: 6.66e-06 [meta_fg_expand]: 4.12e-06 [flash_sp_send_recv_attached]: 2.86999e-06 [receive_attached]: 2.22001e-06 [after_resolve]: 2.01e-05 [a_after_grad]: 3.018e-05 [renormalize]: 0.00117399 [add_forward_monad_depend]: 1.463e-05 [auto_monad_grad]: 2.61e-06 [auto_monad_eliminator]: 2.51e-05 [cse]: 4.392e-05 [a_3]: 6.95e-05 [Cycle 2]: 0.00131402, [45] [expand_dump_flag]: 2.17999e-06 [switch_simplify]: 1.053e-05 [loop_unroll]: 9.22999e-06 [a_1]: 0.00021116 [with_stream_mark]: 1.851e-05 [recompute_prepare]: 8.99e-06 [updatestate_depend_eliminate]: 1.038e-05 [updatestate_assign_eliminate]: 5.79999e-06 [updatestate_loads_eliminate]: 8.33999e-06 [parameter_eliminate]: 1.94999e-06 [a_2]: 0.00011995 [accelerated_algorithm]: 9.40001e-06 [shard]: 5.20999e-06 [meta_shard_fg_expand]: 2.99999e-06 [shard_inline]: 8.79e-06 [merge_send_recv]: 8.99e-06 [auto_parallel]: 1.26e-05 [parallel]: 1.114e-05 [flash_sp]: 4.07003e-06 [merge_comm]: 8.48001e-06 [allreduce_fusion]: 1.126e-05 [matmul_add_comm_reduction]: 1.176e-05 [allreduce_slice_to_reducescatter]: 6.30011e-07 [virtual_shard_identity]: 1.269e-05 [virtual_dataset]: 1.222e-05 [get_grad_eliminate_]: 1.212e-05 [virtual_output]: 9.23002e-06 [merge_forward]: 6.50002e-06 [cell_reuse_recompute_pass]: 1.86e-06 [offload_activation]: 1.662e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.227e-05 [merge_recompute_call_nodes]: 1.25999e-06 [before_grad]: 2.496e-05 [set_forward_comm_id_for_comm_node_pass]: 6.44001e-06 [meta_fg_expand]: 4.52e-06 [flash_sp_send_recv_attached]: 1.19e-06 [receive_attached]: 2.24999e-06 [after_resolve]: 1.339e-05 [a_after_grad]: 1.509e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 2.07001e-06 [auto_monad_grad]: 1.55999e-06 [auto_monad_eliminator]: 1.439e-05 [cse]: 2.548e-05 [a_3]: 5.448e-05 [py_interpret_to_execute_after_opt_a]: 2.346e-05 [slice_cell_reuse_recomputed_activation]: 1.99e-06 [rewriter_after_opt_a]: 0.0002239 [convert_after_rewriter]: 1.181e-05 [order_py_execute_after_rewriter]: 7.46999e-06 [mutable_eliminate]: 0.00094751 [opt_b]: 0.00048981, [1] [Cycle 1]: 0.0004802, [7] [b_1]: 0.00031688 [b_2]: 4.971e-05 [updatestate_depend_eliminate]: 1.19e-05 [updatestate_assign_eliminate]: 1.099e-05 [updatestate_loads_eliminate]: 5.17e-06 [renormalize]: 8.40024e-07 [cse]: 3.492e-05 [optimize_parallel_all_gather_comm]: 2.394e-05 [overlap_param_gather]: 1.89e-06 [cconv]: 3.669e-05 [loop_unroll]: 0.00054982 [opt_after_cconv]: 0.00017025, [1] [Cycle 1]: 0.00016231, [7] [c_1]: 7.161e-05 [parameter_eliminate]: 4.1e-06 [updatestate_depend_eliminate]: 8.12998e-06 [updatestate_assign_eliminate]: 4.65999e-06 [updatestate_loads_eliminate]: 4.74e-06 [cse]: 2.389e-05 [renormalize]: 1.42999e-06 [remove_dup_value]: 1.767e-05 [tuple_transform]: 0.100808, [1] [Cycle 1]: 0.100798, [4] [d_1]: 0.100666 [none_parameter_eliminate]: 8.39998e-06 [renormalize]: 9.99979e-07 [switch_simplify]: 2.572e-05 [partial_unused_args_eliminate]: 4.34002e-06 [add_recomputation]: 0.00013224 [cse_after_recomputation]: 7.503e-05, [1] [Cycle 1]: 6.166e-05, [1] [cse]: 5.276e-05 [environ_conv]: 9.62001e-06 [swap_dp_allreduce_reducescatter]: 1.041e-05 [bias_add_comm_swap]: 3.91001e-06 [label_micro_interleaved_index]: 9.36002e-06 [label_fine_grained_interleaved_index]: 3.44001e-06 [merge_cast_opt]: 1.67999e-06 [slice_recompute_activation]: 2.61999e-06 [micro_interleaved_order_control]: 2.81e-06 [assign_add_opt]: 1.77999e-06 [ForceFp32Comm]: 8.89995e-07 [remove_cast_before_assign_add]: 9.80013e-07 [full_micro_interleaved_order_control]: 2.91e-06 [reorder_send_recv_between_fp_bp]: 3.13e-06 [comm_op_add_attrs]: 1.15001e-06 [add_comm_op_reuse_tag]: 1.11002e-06 [interleave_split_concat_branches]: 1.37e-06 [interleave_parallel_branches]: 1.22999e-06 [overlap_opt_shard_in_pipeline]: 1.27999e-06 [overlap_opt_shard_grad_in_pipeline]: 1.89e-06 [control_data_broadcast_order]: 2.666e-05 [grouped_pairwise_exchange_alltoall]: 1.78002e-06 [offloading_packed_experts]: 6.96999e-06 [overlap_recompute_and_grad_model_parallel]: 1.108e-05 [overlap_grad_matmul_and_grad_allreduce]: 1.68002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.97999e-06 [overlap_recompute_comm]: 2.84001e-06 [overlap_grad_ring_attention]: 9.31998e-06 [overlap_grad_flash_sp]: 3.658e-05 [begin_end_overlap_inline]: 8.30012e-07 [split_matmul_comm_elemetwise]: 2.63e-06 [split_layernorm_comm]: 1.84998e-06 [handle_group_info]: 1.05999e-06 [symbol_engine_optimizer]: 0.00025505, [1] [Cycle 1]: 0.00024875, [6] [build]: 7.70998e-06 [elim_shapecalc]: 3.503e-05 [elim_not_effective]: 5.262e-05 [opt_reshape]: 1.352e-05 [fold_const_symbol]: 4.384e-05 [renormalize]: 5.10016e-07 [detach_backward]: 2.37999e-06 [pipeline_parallel_scheduler]: 2.00002e-06 [auto_monad_reorder]: 4.587e-05 [get_jit_bprop_graph]: 2.28998e-06 [rewriter_after_jit_bprop_graph]: 9.10001e-06 [opt_after_jit_grad]: 0.00131288 [validate]: 8.777e-05 [backend_pass]: 1.27e-06 [task_emit]: 0.101495 [execute]: 2.322e-05 Sums bootstrap : 0.000690s : 0.22% type_inference : 0.109019s : 34.02% event_method : 0.000028s : 0.01% auto_monad : 0.000252s : 0.08% graph_reusing : 0.000007s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000019s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000005s : 0.00% pre_auto_parallel : 0.000038s : 0.01% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000028s : 0.01% optimize.rewriter_before_opt_a : 0.000115s : 0.04% optimize.opt_a.expand_dump_flag : 0.000008s : 0.00% optimize.opt_a.switch_simplify : 0.000039s : 0.01% optimize.opt_a.loop_unroll : 0.000025s : 0.01% optimize.opt_a.a_1 : 0.000983s : 0.31% optimize.opt_a.with_stream_mark : 0.000045s : 0.01% optimize.opt_a.recompute_prepare : 0.000028s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000020s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000043s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000019s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.00% optimize.opt_a.a_2 : 0.000312s : 0.10% optimize.opt_a.accelerated_algorithm : 0.000023s : 0.01% optimize.opt_a.shard : 0.000008s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000005s : 0.00% optimize.opt_a.shard_inline : 0.000018s : 0.01% optimize.opt_a.merge_send_recv : 0.000020s : 0.01% optimize.opt_a.auto_parallel : 0.000030s : 0.01% optimize.opt_a.parallel : 0.000047s : 0.01% optimize.opt_a.flash_sp : 0.000017s : 0.01% optimize.opt_a.merge_comm : 0.000017s : 0.01% optimize.opt_a.allreduce_fusion : 0.000096s : 0.03% optimize.opt_a.matmul_add_comm_reduction : 0.000038s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000032s : 0.01% optimize.opt_a.virtual_dataset : 0.000032s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000031s : 0.01% optimize.opt_a.virtual_output : 0.000033s : 0.01% optimize.opt_a.merge_forward : 0.000016s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.00% optimize.opt_a.offload_activation : 0.000043s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000063s : 0.02% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000046s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000013s : 0.00% optimize.opt_a.meta_fg_expand : 0.000009s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.00% optimize.opt_a.receive_attached : 0.000004s : 0.00% optimize.opt_a.after_resolve : 0.000033s : 0.01% optimize.opt_a.a_after_grad : 0.000045s : 0.01% optimize.opt_a.renormalize : 0.001174s : 0.37% optimize.opt_a.add_forward_monad_depend : 0.000017s : 0.01% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000039s : 0.01% optimize.opt_a.cse : 0.000069s : 0.02% optimize.opt_a.a_3 : 0.000124s : 0.04% optimize.py_interpret_to_execute_after_opt_a : 0.000023s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000224s : 0.07% optimize.convert_after_rewriter : 0.000012s : 0.00% optimize.order_py_execute_after_rewriter : 0.000007s : 0.00% optimize.mutable_eliminate : 0.000948s : 0.30% optimize.opt_b.b_1 : 0.000317s : 0.10% optimize.opt_b.b_2 : 0.000050s : 0.02% optimize.opt_b.updatestate_depend_eliminate : 0.000012s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000011s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000035s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000024s : 0.01% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000037s : 0.01% optimize.loop_unroll : 0.000550s : 0.17% optimize.opt_after_cconv.c_1 : 0.000072s : 0.02% optimize.opt_after_cconv.parameter_eliminate : 0.000004s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000008s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.cse : 0.000024s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000018s : 0.01% optimize.tuple_transform.d_1 : 0.100666s : 31.41% optimize.tuple_transform.none_parameter_eliminate : 0.000008s : 0.00% optimize.tuple_transform.renormalize : 0.000001s : 0.00% optimize.tuple_transform.switch_simplify : 0.000026s : 0.01% optimize.partial_unused_args_eliminate : 0.000004s : 0.00% optimize.add_recomputation : 0.000132s : 0.04% optimize.cse_after_recomputation.cse : 0.000053s : 0.02% optimize.environ_conv : 0.000010s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000010s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000009s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000027s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000011s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000009s : 0.00% optimize.overlap_grad_flash_sp : 0.000037s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000008s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000035s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000053s : 0.02% optimize.symbol_engine_optimizer.opt_reshape : 0.000014s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000044s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000001s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000046s : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000009s : 0.00% opt_after_jit_grad : 0.001313s : 0.41% validate : 0.000088s : 0.03% backend_pass : 0.000001s : 0.00% task_emit : 0.101495s : 31.67% execute : 0.000023s : 0.01% Time group info: ------[substitution.] 0.000245 55 5.16% : 0.000013s : 2: substitution.depend_value_elim 1.61% : 0.000004s : 5: substitution.elim_not_effective 0.98% : 0.000002s : 5: substitution.fold_const_symbol 7.82% : 0.000019s : 6: substitution.graph_param_transform 63.78% : 0.000156s : 1: substitution.inline 2.39% : 0.000006s : 10: substitution.j_node_and_user_rematch 3.46% : 0.000008s : 10: substitution.remove_not_recompute_node 1.88% : 0.000005s : 2: substitution.replace_old_param 6.32% : 0.000015s : 6: substitution.updatestate_pure_node_eliminater 6.60% : 0.000016s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.108936 2 98.97% : 0.107819s : 1: type_inference.infer 1.03% : 0.001117s : 1: type_inference.specialize ------[replace.] 0.000024 1 100.00% : 0.000024s : 1: replace.inline ------[match.] 0.000153 1 100.00% : 0.000153s : 1: match.inline ------[predicate.] 0.000259 1419 0.78% : 0.000002s : 13: predicate.accumulaten_eliminater 2.08% : 0.000005s : 6: predicate.ad_related_special_op_eliminate 0.66% : 0.000002s : 12: predicate.addn_check_dump 0.78% : 0.000002s : 13: predicate.addn_zero_filter 0.74% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 2.06% : 0.000005s : 25: predicate.arithmetic_simplify 0.76% : 0.000002s : 13: predicate.cast_eliminate 0.78% : 0.000002s : 12: predicate.check_bprop_eliminate 0.63% : 0.000002s : 12: predicate.compare_switch_simplify 0.20% : 0.000001s : 6: predicate.const_output_eliminate 0.68% : 0.000002s : 12: predicate.depend_value_elim 0.74% : 0.000002s : 13: predicate.dict_get_item_const_eliminator 0.91% : 0.000002s : 13: predicate.dict_get_item_eliminator 0.69% : 0.000002s : 13: predicate.dict_set_item_eliminator 1.64% : 0.000004s : 12: predicate.dumpgradient_eliminate 0.34% : 0.000001s : 6: predicate.elim_not_effective 0.81% : 0.000002s : 6: predicate.elim_shapecalc_of_broadcastargs 1.10% : 0.000003s : 19: predicate.environ_add_const_eliminate 1.07% : 0.000003s : 19: predicate.environ_get_add_eliminate 1.17% : 0.000003s : 19: predicate.environ_get_depend_swap 1.82% : 0.000005s : 31: predicate.environ_get_eliminate 1.04% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.75% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.30% : 0.000003s : 14: predicate.float_depend_g_call 0.78% : 0.000002s : 12: predicate.float_environ_get_switch 0.95% : 0.000002s : 18: predicate.float_tuple_getitem_switch 0.19% : 0.000000s : 6: predicate.fold_const_symbol 0.75% : 0.000002s : 12: predicate.get_grad_eliminate 0.56% : 0.000001s : 6: predicate.graph_param_transform 0.71% : 0.000002s : 12: predicate.incorporate_call 0.62% : 0.000002s : 12: predicate.incorporate_call_switch 5.25% : 0.000014s : 63: predicate.inline 1.28% : 0.000003s : 12: predicate.inline_without_move 0.38% : 0.000001s : 12: predicate.j_node_and_user_rematch 0.94% : 0.000002s : 12: predicate.less_batch_normalization 2.00% : 0.000005s : 25: predicate.list_to_tuple_eliminator_ 2.05% : 0.000005s : 38: predicate.load_eliminater 1.16% : 0.000003s : 6: predicate.loop_unroll_after_grad 1.04% : 0.000003s : 17: predicate.loop_unroll_before_grad 1.75% : 0.000005s : 25: predicate.make_slice_get_slice_eliminator 0.72% : 0.000002s : 12: predicate.merge_addn 0.79% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.68% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.78% : 0.000002s : 13: predicate.minmaximum_grad 1.52% : 0.000004s : 6: predicate.mutable_eliminate 0.71% : 0.000002s : 6: predicate.opt_reshape 0.49% : 0.000001s : 6: predicate.parallel_virtual_node 1.01% : 0.000003s : 14: predicate.partial_defer_inline 1.18% : 0.000003s : 19: predicate.partial_eliminate 0.77% : 0.000002s : 13: predicate.print_const_string_wrapper 0.73% : 0.000002s : 12: predicate.reduce_all_const_elim 1.07% : 0.000003s : 13: predicate.reduce_eliminate 2.05% : 0.000005s : 38: predicate.redundant_stop_gradient_eliminater 0.65% : 0.000002s : 12: predicate.remove_not_recompute_node 1.12% : 0.000003s : 25: predicate.replace_applicator 0.52% : 0.000001s : 12: predicate.replace_old_param 0.43% : 0.000001s : 6: predicate.reset_defer_inline 0.76% : 0.000002s : 13: predicate.reshape_eliminate 0.79% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.58% : 0.000001s : 6: predicate.row_tensor_eliminate 0.93% : 0.000002s : 12: predicate.same_eliminate 0.43% : 0.000001s : 12: predicate.set_cell_output_no_recompute 0.87% : 0.000002s : 12: predicate.shard_identity_eliminate 0.90% : 0.000002s : 12: predicate.special_op_eliminate 0.80% : 0.000002s : 12: predicate.specialize_transform 1.18% : 0.000003s : 12: predicate.split_environ_get_set_with_tuple_value 0.93% : 0.000002s : 12: predicate.stack_unstack_eliminate 0.47% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.79% : 0.000002s : 14: predicate.switch_defer_inline 1.47% : 0.000004s : 26: predicate.switch_layer_defer_inline 4.35% : 0.000011s : 49: predicate.switch_simplify 0.77% : 0.000002s : 13: predicate.tile_eliminate 0.78% : 0.000002s : 13: predicate.transpose_eliminate 1.74% : 0.000005s : 25: predicate.tuple_list_convert_item_index_to_positive 1.88% : 0.000005s : 25: predicate.tuple_list_get_item_const_eliminator 2.47% : 0.000006s : 25: predicate.tuple_list_get_item_depend_reorder 3.54% : 0.000009s : 37: predicate.tuple_list_get_item_eliminator 1.55% : 0.000004s : 25: predicate.tuple_list_get_set_item_eliminator 2.59% : 0.000007s : 37: predicate.tuple_list_set_item_eliminator 1.69% : 0.000004s : 25: predicate.tuple_to_list_eliminator_ 2.22% : 0.000006s : 38: predicate.updatestate_pure_node_eliminater 3.10% : 0.000008s : 50: predicate.updatestate_useless_node_eliminater 0.47% : 0.000001s : 6: predicate.value_based_eliminate 1.00% : 0.000003s : 12: predicate.virtual_dataset_eliminate 0.80% : 0.000002s : 12: predicate.virtual_output_eliminate 0.41% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.58% : 0.000002s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000627 4 3.65% : 0.000023s : 1: func_graph_cloner_run.FuncGraphClonerGraph 96.35% : 0.000604s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.550312 196 0.00% : 0.000004s : 1: ForceFp32Comm 1.32% : 0.007288s : 1: add_attr 1.32% : 0.007263s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.03% : 0.000139s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.05% : 0.000264s : 1: auto_monad 0.01% : 0.000054s : 1: auto_monad_reorder 0.00% : 0.000014s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.13% : 0.000728s : 1: bootstrap 0.01% : 0.000040s : 1: cconv 0.00% : 0.000005s : 1: comm_op_add_attrs 0.01% : 0.000033s : 1: control_data_broadcast_order 0.00% : 0.000017s : 1: convert_after_rewriter 0.01% : 0.000079s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000014s : 1: environ_conv 0.01% : 0.000036s : 1: event_method 0.01% : 0.000034s : 1: execute 0.00% : 0.000007s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000016s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000012s : 1: handle_group_info 0.00% : 0.000011s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000013s : 1: label_micro_interleaved_index 0.10% : 0.000561s : 1: loop_unroll 0.00% : 0.000010s : 1: merge_cast_opt 0.00% : 0.000009s : 1: micro_interleaved_order_control 0.17% : 0.000962s : 1: mutable_eliminate 0.00% : 0.000016s : 1: offloading_packed_experts 0.00% : 0.000022s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000024s : 1: opt.transform.mutable_eliminate 0.32% : 0.001761s : 78: opt.transform.opt_a 0.01% : 0.000070s : 1: opt.transform.opt_after_cconv 0.01% : 0.000067s : 1: opt.transform.opt_after_jit_grad 0.06% : 0.000331s : 28: opt.transform.opt_b 18.30% : 0.100680s : 2: opt.transform.opt_trans_graph 0.03% : 0.000140s : 4: opt.transform.symbol_engine_opt 0.85% : 0.004694s : 1: opt_a 0.03% : 0.000175s : 1: opt_after_cconv 0.24% : 0.001347s : 1: opt_after_jit_grad 0.09% : 0.000494s : 1: opt_b 19.83% : 0.109135s : 1: optimize 0.01% : 0.000028s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000011s : 1: order_py_execute_after_rewriter 0.01% : 0.000041s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000013s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000014s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000010s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000016s : 1: partial_unused_args_eliminate 0.00% : 0.000009s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000046s : 1: pre_auto_parallel 0.01% : 0.000033s : 1: py_interpret_to_execute 0.01% : 0.000028s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000022s : 1: remove_dup_value 0.12% : 0.000661s : 1: renormalize.infer 0.09% : 0.000468s : 1: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000013s : 1: rewriter_after_jit_bprop_graph 0.04% : 0.000232s : 1: rewriter_after_opt_a 0.02% : 0.000122s : 1: rewriter_before_opt_a 0.00% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000013s : 1: split_matmul_comm_elemetwise 0.00% : 0.000019s : 1: swap_dp_allreduce_reducescatter 0.05% : 0.000262s : 1: symbol_engine_optimizer 18.45% : 0.101526s : 1: task_emit 18.32% : 0.100814s : 1: tuple_transform 19.82% : 0.109046s : 1: type_inference 0.03% : 0.000163s : 1: validate TotalTime = 0.15845, [24] [bootstrap]: 0.00046917 [type_inference]: 0.004286 [event_method]: 2.547e-05 [auto_monad]: 0.00027442 [graph_reusing]: 7.5e-06 [inline]: 2.56e-06 [add_attr]: 0.0050725, [1] [add_attr_with_inline]: 0.00505903, [1] [Cycle 1]: 7.299e-05, [2] [tag_attr]: 1.969e-05 [meta_addattr_fg_expand]: 3.80998e-06 [parallel-infer-symbol]: 3.7e-06 [pre_auto_parallel]: 3.777e-05 [insert-virtual-dataset]: 2.54001e-06 [parallel-infer-symbol-second]: 7.10017e-07 [dataset_repeat_opt]: 2.32999e-06 [pipeline_split]: 1.66e-06 [optimize]: 0.135016, [53] [py_interpret_to_execute]: 2.756e-05 [rewriter_before_opt_a]: 0.00011163 [opt_a]: 0.131382, [2] [Cycle 1]: 0.130184, [45] [expand_dump_flag]: 3.56999e-06 [switch_simplify]: 2.936e-05 [loop_unroll]: 1.468e-05 [a_1]: 0.00080738 [with_stream_mark]: 3.18e-05 [recompute_prepare]: 3.252e-05 [updatestate_depend_eliminate]: 1.458e-05 [updatestate_assign_eliminate]: 7.28e-06 [updatestate_loads_eliminate]: 6.77002e-06 [parameter_eliminate]: 2.84999e-06 [a_2]: 0.00020222 [accelerated_algorithm]: 1.4e-05 [shard]: 3.02002e-06 [meta_shard_fg_expand]: 4.38999e-06 [shard_inline]: 1.161e-05 [merge_send_recv]: 1.289e-05 [auto_parallel]: 1.436e-05 [parallel]: 2.278e-05 [flash_sp]: 1.504e-05 [merge_comm]: 6.86001e-06 [allreduce_fusion]: 6.09001e-06 [matmul_add_comm_reduction]: 1.606e-05 [allreduce_slice_to_reducescatter]: 1.14003e-06 [virtual_shard_identity]: 2.595e-05 [virtual_dataset]: 2.061e-05 [get_grad_eliminate_]: 1.572e-05 [virtual_output]: 1.822e-05 [merge_forward]: 1.334e-05 [cell_reuse_recompute_pass]: 2.61e-06 [offload_activation]: 1.712e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.799e-05 [merge_recompute_call_nodes]: 1.58002e-06 [before_grad]: 3.62e-05 [set_forward_comm_id_for_comm_node_pass]: 7.20998e-06 [meta_fg_expand]: 5.49e-06 [flash_sp_send_recv_attached]: 3.41999e-06 [receive_attached]: 3.48999e-06 [after_resolve]: 1.95e-05 [a_after_grad]: 2.61e-05 [renormalize]: 0.127942 [add_forward_monad_depend]: 1.412e-05 [auto_monad_grad]: 3.95e-06 [auto_monad_eliminator]: 4.117e-05 [cse]: 4.944e-05 [a_3]: 9.935e-05 [Cycle 2]: 0.00117708, [45] [expand_dump_flag]: 3.2e-06 [switch_simplify]: 1.433e-05 [loop_unroll]: 1.007e-05 [a_1]: 0.00025064 [with_stream_mark]: 2.778e-05 [recompute_prepare]: 1.173e-05 [updatestate_depend_eliminate]: 6.58e-06 [updatestate_assign_eliminate]: 6.01e-06 [updatestate_loads_eliminate]: 7.23999e-06 [parameter_eliminate]: 2.58e-06 [a_2]: 0.00012683 [accelerated_algorithm]: 1.087e-05 [shard]: 3.41001e-06 [meta_shard_fg_expand]: 3.23e-06 [shard_inline]: 1.001e-05 [merge_send_recv]: 1.296e-05 [auto_parallel]: 1.409e-05 [parallel]: 1.214e-05 [flash_sp]: 4.80001e-06 [merge_comm]: 6.44999e-06 [allreduce_fusion]: 6.26e-06 [matmul_add_comm_reduction]: 1.508e-05 [allreduce_slice_to_reducescatter]: 1.14e-06 [virtual_shard_identity]: 1.192e-05 [virtual_dataset]: 1.088e-05 [get_grad_eliminate_]: 9.39998e-06 [virtual_output]: 9.23002e-06 [merge_forward]: 6.71e-06 [cell_reuse_recompute_pass]: 3.34001e-06 [offload_activation]: 1.882e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.253e-05 [merge_recompute_call_nodes]: 1.52001e-06 [before_grad]: 1.866e-05 [set_forward_comm_id_for_comm_node_pass]: 6.16e-06 [meta_fg_expand]: 5.05999e-06 [flash_sp_send_recv_attached]: 2.30002e-06 [receive_attached]: 2.59001e-06 [after_resolve]: 1.79e-05 [a_after_grad]: 1.472e-05 [renormalize]: 8.00064e-08 [add_forward_monad_depend]: 1.80001e-06 [auto_monad_grad]: 1.49e-06 [auto_monad_eliminator]: 1.842e-05 [cse]: 2.538e-05 [a_3]: 5.746e-05 [py_interpret_to_execute_after_opt_a]: 2.921e-05 [slice_cell_reuse_recomputed_activation]: 2.06998e-06 [rewriter_after_opt_a]: 0.00024901 [convert_after_rewriter]: 1.326e-05 [order_py_execute_after_rewriter]: 8e-06 [mutable_eliminate]: 0.00096836 [opt_b]: 0.00044405, [1] [Cycle 1]: 0.00043306, [7] [b_1]: 0.00029517 [b_2]: 1.325e-05 [updatestate_depend_eliminate]: 1.442e-05 [updatestate_assign_eliminate]: 6.11e-06 [updatestate_loads_eliminate]: 6.09001e-06 [renormalize]: 6.80011e-07 [cse]: 3.879e-05 [optimize_parallel_all_gather_comm]: 2.726e-05 [overlap_param_gather]: 2.03002e-06 [cconv]: 4.308e-05 [loop_unroll]: 0.00062206 [opt_after_cconv]: 0.00019001, [1] [Cycle 1]: 0.00018127, [7] [c_1]: 6.787e-05 [parameter_eliminate]: 6.29999e-06 [updatestate_depend_eliminate]: 9.25999e-06 [updatestate_assign_eliminate]: 5.57001e-06 [updatestate_loads_eliminate]: 5.39e-06 [cse]: 2.619e-05 [renormalize]: 9.20001e-07 [remove_dup_value]: 2.1e-05 [tuple_transform]: 0.00012419, [1] [Cycle 1]: 0.00011645, [4] [d_1]: 7.498e-05 [none_parameter_eliminate]: 2.17999e-06 [renormalize]: 2.10013e-07 [switch_simplify]: 1.099e-05 [partial_unused_args_eliminate]: 2.32001e-06 [add_recomputation]: 9.91e-05 [cse_after_recomputation]: 3.313e-05, [1] [Cycle 1]: 2.722e-05, [1] [cse]: 1.859e-05 [environ_conv]: 9.46e-06 [swap_dp_allreduce_reducescatter]: 1.025e-05 [bias_add_comm_swap]: 3.35e-06 [label_micro_interleaved_index]: 6.21998e-06 [label_fine_grained_interleaved_index]: 2.96999e-06 [merge_cast_opt]: 1.90001e-06 [slice_recompute_activation]: 2.81999e-06 [micro_interleaved_order_control]: 2.94001e-06 [assign_add_opt]: 1.45999e-06 [ForceFp32Comm]: 1.14998e-06 [remove_cast_before_assign_add]: 9.09989e-07 [full_micro_interleaved_order_control]: 2.41e-06 [reorder_send_recv_between_fp_bp]: 3.08e-06 [comm_op_add_attrs]: 1.14e-06 [add_comm_op_reuse_tag]: 1.30999e-06 [interleave_split_concat_branches]: 1.67999e-06 [interleave_parallel_branches]: 2.04999e-06 [overlap_opt_shard_in_pipeline]: 1.44998e-06 [overlap_opt_shard_grad_in_pipeline]: 1.81998e-06 [control_data_broadcast_order]: 2.125e-05 [grouped_pairwise_exchange_alltoall]: 1.98002e-06 [offloading_packed_experts]: 5.89999e-06 [overlap_recompute_and_grad_model_parallel]: 7.25e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.64e-06 [overlap_recompute_allgather_and_fa_grad]: 1.69998e-06 [overlap_recompute_comm]: 3.03e-06 [overlap_grad_ring_attention]: 6.01e-06 [overlap_grad_flash_sp]: 3.126e-05 [begin_end_overlap_inline]: 5.50004e-07 [split_matmul_comm_elemetwise]: 2.40002e-06 [split_layernorm_comm]: 1.74998e-06 [handle_group_info]: 1.57999e-06 [symbol_engine_optimizer]: 0.00012014, [1] [Cycle 1]: 0.00011418, [6] [build]: 5.77999e-06 [elim_shapecalc]: 1.585e-05 [elim_not_effective]: 1.985e-05 [opt_reshape]: 1.133e-05 [fold_const_symbol]: 1.644e-05 [renormalize]: 2.50002e-07 [detach_backward]: 2.51998e-06 [pipeline_parallel_scheduler]: 1.44e-06 [auto_monad_reorder]: 3.247e-05 [get_jit_bprop_graph]: 1.86998e-06 [rewriter_after_jit_bprop_graph]: 6.09001e-06 [opt_after_jit_grad]: 0.00068349 [validate]: 6.596e-05 [backend_pass]: 1.17999e-06 [task_emit]: 0.0121144 [execute]: 1.174e-05 Sums bootstrap : 0.000469s : 0.31% type_inference : 0.004286s : 2.83% event_method : 0.000025s : 0.02% auto_monad : 0.000274s : 0.18% graph_reusing : 0.000007s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000020s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000038s : 0.02% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000028s : 0.02% optimize.rewriter_before_opt_a : 0.000112s : 0.07% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000044s : 0.03% optimize.opt_a.loop_unroll : 0.000025s : 0.02% optimize.opt_a.a_1 : 0.001058s : 0.70% optimize.opt_a.with_stream_mark : 0.000060s : 0.04% optimize.opt_a.recompute_prepare : 0.000044s : 0.03% optimize.opt_a.updatestate_depend_eliminate : 0.000021s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000013s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000014s : 0.01% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000329s : 0.22% optimize.opt_a.accelerated_algorithm : 0.000025s : 0.02% optimize.opt_a.shard : 0.000006s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000008s : 0.01% optimize.opt_a.shard_inline : 0.000022s : 0.01% optimize.opt_a.merge_send_recv : 0.000026s : 0.02% optimize.opt_a.auto_parallel : 0.000028s : 0.02% optimize.opt_a.parallel : 0.000035s : 0.02% optimize.opt_a.flash_sp : 0.000020s : 0.01% optimize.opt_a.merge_comm : 0.000013s : 0.01% optimize.opt_a.allreduce_fusion : 0.000012s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000031s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000038s : 0.02% optimize.opt_a.virtual_dataset : 0.000031s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000025s : 0.02% optimize.opt_a.virtual_output : 0.000027s : 0.02% optimize.opt_a.merge_forward : 0.000020s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000006s : 0.00% optimize.opt_a.offload_activation : 0.000036s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000051s : 0.03% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000055s : 0.04% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000013s : 0.01% optimize.opt_a.meta_fg_expand : 0.000011s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000006s : 0.00% optimize.opt_a.receive_attached : 0.000006s : 0.00% optimize.opt_a.after_resolve : 0.000037s : 0.02% optimize.opt_a.a_after_grad : 0.000041s : 0.03% optimize.opt_a.renormalize : 0.127942s : 84.39% optimize.opt_a.add_forward_monad_depend : 0.000016s : 0.01% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000060s : 0.04% optimize.opt_a.cse : 0.000075s : 0.05% optimize.opt_a.a_3 : 0.000157s : 0.10% optimize.py_interpret_to_execute_after_opt_a : 0.000029s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000249s : 0.16% optimize.convert_after_rewriter : 0.000013s : 0.01% optimize.order_py_execute_after_rewriter : 0.000008s : 0.01% optimize.mutable_eliminate : 0.000968s : 0.64% optimize.opt_b.b_1 : 0.000295s : 0.19% optimize.opt_b.b_2 : 0.000013s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000014s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000039s : 0.03% optimize.optimize_parallel_all_gather_comm : 0.000027s : 0.02% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000043s : 0.03% optimize.loop_unroll : 0.000622s : 0.41% optimize.opt_after_cconv.c_1 : 0.000068s : 0.04% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000009s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.cse : 0.000026s : 0.02% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000021s : 0.01% optimize.tuple_transform.d_1 : 0.000075s : 0.05% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000011s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000099s : 0.07% optimize.cse_after_recomputation.cse : 0.000019s : 0.01% optimize.environ_conv : 0.000009s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000010s : 0.01% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000006s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000002s : 0.00% optimize.interleave_parallel_branches : 0.000002s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000021s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000006s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000006s : 0.00% optimize.overlap_grad_flash_sp : 0.000031s : 0.02% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000002s : 0.00% optimize.symbol_engine_optimizer.build : 0.000006s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000016s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000020s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000011s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000016s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000001s : 0.00% auto_monad_reorder : 0.000032s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.000683s : 0.45% validate : 0.000066s : 0.04% backend_pass : 0.000001s : 0.00% task_emit : 0.012114s : 7.99% execute : 0.000012s : 0.01% Time group info: ------[substitution.] 0.000225 55 4.84% : 0.000011s : 2: substitution.depend_value_elim 1.59% : 0.000004s : 5: substitution.elim_not_effective 1.01% : 0.000002s : 5: substitution.fold_const_symbol 3.40% : 0.000008s : 6: substitution.graph_param_transform 60.26% : 0.000136s : 1: substitution.inline 2.92% : 0.000007s : 10: substitution.j_node_and_user_rematch 3.70% : 0.000008s : 10: substitution.remove_not_recompute_node 3.07% : 0.000007s : 2: substitution.replace_old_param 9.66% : 0.000022s : 6: substitution.updatestate_pure_node_eliminater 9.55% : 0.000022s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.004218 2 83.11% : 0.003505s : 1: type_inference.infer 16.89% : 0.000712s : 1: type_inference.specialize ------[replace.] 0.000022 1 100.00% : 0.000022s : 1: replace.inline ------[match.] 0.000135 1 100.00% : 0.000135s : 1: match.inline ------[predicate.] 0.000261 1419 0.83% : 0.000002s : 13: predicate.accumulaten_eliminater 1.49% : 0.000004s : 6: predicate.ad_related_special_op_eliminate 0.69% : 0.000002s : 12: predicate.addn_check_dump 0.78% : 0.000002s : 13: predicate.addn_zero_filter 0.85% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 2.21% : 0.000006s : 25: predicate.arithmetic_simplify 0.82% : 0.000002s : 13: predicate.cast_eliminate 0.79% : 0.000002s : 12: predicate.check_bprop_eliminate 0.61% : 0.000002s : 12: predicate.compare_switch_simplify 0.20% : 0.000001s : 6: predicate.const_output_eliminate 0.82% : 0.000002s : 12: predicate.depend_value_elim 0.78% : 0.000002s : 13: predicate.dict_get_item_const_eliminator 0.92% : 0.000002s : 13: predicate.dict_get_item_eliminator 0.91% : 0.000002s : 13: predicate.dict_set_item_eliminator 1.21% : 0.000003s : 12: predicate.dumpgradient_eliminate 0.25% : 0.000001s : 6: predicate.elim_not_effective 0.50% : 0.000001s : 6: predicate.elim_shapecalc_of_broadcastargs 1.10% : 0.000003s : 19: predicate.environ_add_const_eliminate 1.01% : 0.000003s : 19: predicate.environ_get_add_eliminate 1.00% : 0.000003s : 19: predicate.environ_get_depend_swap 1.76% : 0.000005s : 31: predicate.environ_get_eliminate 1.05% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.80% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.56% : 0.000004s : 14: predicate.float_depend_g_call 0.63% : 0.000002s : 12: predicate.float_environ_get_switch 1.10% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.19% : 0.000000s : 6: predicate.fold_const_symbol 1.12% : 0.000003s : 12: predicate.get_grad_eliminate 0.29% : 0.000001s : 6: predicate.graph_param_transform 0.69% : 0.000002s : 12: predicate.incorporate_call 0.60% : 0.000002s : 12: predicate.incorporate_call_switch 5.63% : 0.000015s : 63: predicate.inline 1.14% : 0.000003s : 12: predicate.inline_without_move 0.33% : 0.000001s : 12: predicate.j_node_and_user_rematch 1.08% : 0.000003s : 12: predicate.less_batch_normalization 1.69% : 0.000004s : 25: predicate.list_to_tuple_eliminator_ 2.11% : 0.000006s : 38: predicate.load_eliminater 1.53% : 0.000004s : 6: predicate.loop_unroll_after_grad 1.00% : 0.000003s : 17: predicate.loop_unroll_before_grad 1.67% : 0.000004s : 25: predicate.make_slice_get_slice_eliminator 0.66% : 0.000002s : 12: predicate.merge_addn 0.70% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.72% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.74% : 0.000002s : 13: predicate.minmaximum_grad 2.26% : 0.000006s : 6: predicate.mutable_eliminate 0.37% : 0.000001s : 6: predicate.opt_reshape 0.52% : 0.000001s : 6: predicate.parallel_virtual_node 1.31% : 0.000003s : 14: predicate.partial_defer_inline 1.11% : 0.000003s : 19: predicate.partial_eliminate 0.76% : 0.000002s : 13: predicate.print_const_string_wrapper 0.83% : 0.000002s : 12: predicate.reduce_all_const_elim 1.36% : 0.000004s : 13: predicate.reduce_eliminate 2.01% : 0.000005s : 38: predicate.redundant_stop_gradient_eliminater 0.56% : 0.000001s : 12: predicate.remove_not_recompute_node 1.33% : 0.000003s : 25: predicate.replace_applicator 0.69% : 0.000002s : 12: predicate.replace_old_param 0.26% : 0.000001s : 6: predicate.reset_defer_inline 1.16% : 0.000003s : 13: predicate.reshape_eliminate 0.79% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.51% : 0.000001s : 6: predicate.row_tensor_eliminate 0.93% : 0.000002s : 12: predicate.same_eliminate 0.72% : 0.000002s : 12: predicate.set_cell_output_no_recompute 1.11% : 0.000003s : 12: predicate.shard_identity_eliminate 0.92% : 0.000002s : 12: predicate.special_op_eliminate 0.76% : 0.000002s : 12: predicate.specialize_transform 1.28% : 0.000003s : 12: predicate.split_environ_get_set_with_tuple_value 0.90% : 0.000002s : 12: predicate.stack_unstack_eliminate 0.49% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.82% : 0.000002s : 14: predicate.switch_defer_inline 1.44% : 0.000004s : 26: predicate.switch_layer_defer_inline 3.38% : 0.000009s : 49: predicate.switch_simplify 0.84% : 0.000002s : 13: predicate.tile_eliminate 0.93% : 0.000002s : 13: predicate.transpose_eliminate 1.61% : 0.000004s : 25: predicate.tuple_list_convert_item_index_to_positive 1.74% : 0.000005s : 25: predicate.tuple_list_get_item_const_eliminator 1.62% : 0.000004s : 25: predicate.tuple_list_get_item_depend_reorder 3.26% : 0.000009s : 37: predicate.tuple_list_get_item_eliminator 1.85% : 0.000005s : 25: predicate.tuple_list_get_set_item_eliminator 2.54% : 0.000007s : 37: predicate.tuple_list_set_item_eliminator 1.59% : 0.000004s : 25: predicate.tuple_to_list_eliminator_ 2.16% : 0.000006s : 38: predicate.updatestate_pure_node_eliminater 3.41% : 0.000009s : 50: predicate.updatestate_useless_node_eliminater 0.52% : 0.000001s : 6: predicate.value_based_eliminate 1.01% : 0.000003s : 12: predicate.virtual_dataset_eliminate 0.91% : 0.000002s : 12: predicate.virtual_output_eliminate 0.35% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.52% : 0.000001s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000598 4 4.69% : 0.000028s : 1: func_graph_cloner_run.FuncGraphClonerGraph 95.31% : 0.000570s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.428787 196 0.00% : 0.000005s : 1: ForceFp32Comm 1.19% : 0.005081s : 1: add_attr 1.18% : 0.005065s : 1: add_attr_with_inline 0.00% : 0.000005s : 1: add_comm_op_reuse_tag 0.02% : 0.000105s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.07% : 0.000293s : 1: auto_monad 0.01% : 0.000037s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000008s : 1: begin_end_overlap_inline 0.00% : 0.000008s : 1: bias_add_comm_swap 0.12% : 0.000500s : 1: bootstrap 0.01% : 0.000048s : 1: cconv 0.00% : 0.000006s : 1: comm_op_add_attrs 0.01% : 0.000026s : 1: control_data_broadcast_order 0.00% : 0.000019s : 1: convert_after_rewriter 0.01% : 0.000038s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000014s : 1: environ_conv 0.01% : 0.000034s : 1: event_method 0.00% : 0.000020s : 1: execute 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000007s : 1: get_jit_bprop_graph 0.00% : 0.000013s : 1: graph_reusing 0.00% : 0.000006s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000007s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000007s : 1: interleave_parallel_branches 0.00% : 0.000006s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000011s : 1: label_micro_interleaved_index 0.15% : 0.000634s : 1: loop_unroll 0.00% : 0.000007s : 1: merge_cast_opt 0.00% : 0.000007s : 1: micro_interleaved_order_control 0.23% : 0.000984s : 1: mutable_eliminate 0.00% : 0.000011s : 1: offloading_packed_experts 0.01% : 0.000030s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000039s : 1: opt.transform.mutable_eliminate 0.44% : 0.001869s : 78: opt.transform.opt_a 0.02% : 0.000066s : 1: opt.transform.opt_after_cconv 0.01% : 0.000043s : 1: opt.transform.opt_after_jit_grad 0.06% : 0.000271s : 28: opt.transform.opt_b 0.02% : 0.000083s : 2: opt.transform.opt_trans_graph 0.01% : 0.000059s : 4: opt.transform.symbol_engine_opt 30.64% : 0.131394s : 1: opt_a 0.05% : 0.000195s : 1: opt_after_cconv 0.16% : 0.000697s : 1: opt_after_jit_grad 0.10% : 0.000450s : 1: opt_b 31.49% : 0.135025s : 1: optimize 0.01% : 0.000032s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000013s : 1: order_py_execute_after_rewriter 0.01% : 0.000035s : 1: overlap_grad_flash_sp 0.00% : 0.000006s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000010s : 1: overlap_grad_ring_attention 0.00% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000007s : 1: overlap_param_gather 0.00% : 0.000006s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000011s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000007s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000005s : 1: parallel-infer-symbol-second 0.00% : 0.000007s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000006s : 1: pipeline_split 0.01% : 0.000043s : 1: pre_auto_parallel 0.01% : 0.000033s : 1: py_interpret_to_execute 0.01% : 0.000034s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.01% : 0.000026s : 1: remove_dup_value 0.16% : 0.000703s : 1: renormalize.infer 29.67% : 0.127219s : 1: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.06% : 0.000259s : 1: rewriter_after_opt_a 0.03% : 0.000121s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000007s : 1: slice_recompute_activation 0.00% : 0.000008s : 1: split_layernorm_comm 0.00% : 0.000010s : 1: split_matmul_comm_elemetwise 0.00% : 0.000015s : 1: swap_dp_allreduce_reducescatter 0.03% : 0.000124s : 1: symbol_engine_optimizer 2.83% : 0.012142s : 1: task_emit 0.03% : 0.000128s : 1: tuple_transform 1.01% : 0.004311s : 1: type_inference 0.03% : 0.000118s : 1: validate TotalTime = 0.810574, [24] [bootstrap]: 0.00065189 [type_inference]: 0.152796 [event_method]: 7.376e-05 [auto_monad]: 0.0004633 [graph_reusing]: 1.147e-05 [inline]: 3.28998e-06 [add_attr]: 0.00526397, [1] [add_attr_with_inline]: 0.00524986, [1] [Cycle 1]: 0.00011709, [2] [tag_attr]: 5.678e-05 [meta_addattr_fg_expand]: 1.084e-05 [parallel-infer-symbol]: 3.514e-05 [pre_auto_parallel]: 7.182e-05 [insert-virtual-dataset]: 2.58e-06 [parallel-infer-symbol-second]: 6.50005e-07 [dataset_repeat_opt]: 2.48e-06 [pipeline_split]: 1.84998e-06 [optimize]: 0.458834, [53] [py_interpret_to_execute]: 6.156e-05 [rewriter_before_opt_a]: 0.00020716 [opt_a]: 0.454849, [3] [Cycle 1]: 0.44521, [45] [expand_dump_flag]: 9.13002e-06 [switch_simplify]: 0.00010182 [loop_unroll]: 7.71e-05 [a_1]: 0.163789 [with_stream_mark]: 5.748e-05 [recompute_prepare]: 5.608e-05 [updatestate_depend_eliminate]: 1.747e-05 [updatestate_assign_eliminate]: 1.227e-05 [updatestate_loads_eliminate]: 1.294e-05 [parameter_eliminate]: 4.30999e-06 [a_2]: 0.00037995 [accelerated_algorithm]: 5.427e-05 [shard]: 2.99001e-06 [meta_shard_fg_expand]: 1.195e-05 [shard_inline]: 2.105e-05 [merge_send_recv]: 2.53e-05 [auto_parallel]: 2.279e-05 [parallel]: 2.696e-05 [flash_sp]: 1.976e-05 [merge_comm]: 1.511e-05 [allreduce_fusion]: 1.32e-05 [matmul_add_comm_reduction]: 5.047e-05 [allreduce_slice_to_reducescatter]: 9.00007e-07 [virtual_shard_identity]: 3.066e-05 [virtual_dataset]: 3.082e-05 [get_grad_eliminate_]: 2.559e-05 [virtual_output]: 2.408e-05 [merge_forward]: 1.67e-05 [cell_reuse_recompute_pass]: 1.72001e-06 [offload_activation]: 3.776e-05 [cell_reuse_handle_not_recompute_node_pass]: 5.109e-05 [merge_recompute_call_nodes]: 1.79998e-06 [before_grad]: 4.483e-05 [set_forward_comm_id_for_comm_node_pass]: 1.739e-05 [meta_fg_expand]: 0.00854865 [flash_sp_send_recv_attached]: 6.60002e-06 [receive_attached]: 3.06999e-06 [after_resolve]: 0.00016727 [a_after_grad]: 0.00026771 [renormalize]: 0.174061 [add_forward_monad_depend]: 4.198e-05 [auto_monad_grad]: 3.128e-05 [auto_monad_eliminator]: 0.00020266 [cse]: 0.00041042 [a_3]: 0.0957533 [Cycle 2]: 0.00827746, [45] [expand_dump_flag]: 1.965e-05 [switch_simplify]: 0.00017754 [loop_unroll]: 0.00010852 [a_1]: 0.00385372 [with_stream_mark]: 4.095e-05 [recompute_prepare]: 2.711e-05 [updatestate_depend_eliminate]: 4.881e-05 [updatestate_assign_eliminate]: 7.68001e-06 [updatestate_loads_eliminate]: 6.60002e-06 [parameter_eliminate]: 2.51e-06 [a_2]: 0.00017409 [accelerated_algorithm]: 2.026e-05 [shard]: 2.98e-06 [meta_shard_fg_expand]: 8.35001e-06 [shard_inline]: 1.313e-05 [merge_send_recv]: 1.37e-05 [auto_parallel]: 1.548e-05 [parallel]: 1.299e-05 [flash_sp]: 5.20999e-06 [merge_comm]: 7.00998e-06 [allreduce_fusion]: 6.25002e-06 [matmul_add_comm_reduction]: 1.568e-05 [allreduce_slice_to_reducescatter]: 9.09989e-07 [virtual_shard_identity]: 1.731e-05 [virtual_dataset]: 1.193e-05 [get_grad_eliminate_]: 1.106e-05 [virtual_output]: 1.149e-05 [merge_forward]: 7.16999e-06 [cell_reuse_recompute_pass]: 2.42001e-06 [offload_activation]: 1.711e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.54e-05 [merge_recompute_call_nodes]: 2.02001e-06 [before_grad]: 2.017e-05 [set_forward_comm_id_for_comm_node_pass]: 8.42e-06 [meta_fg_expand]: 0.00011072 [flash_sp_send_recv_attached]: 1.76e-06 [receive_attached]: 3.04001e-06 [after_resolve]: 2.305e-05 [a_after_grad]: 1.924e-05 [renormalize]: 0.00266856 [add_forward_monad_depend]: 1.168e-05 [auto_monad_grad]: 2.75997e-06 [auto_monad_eliminator]: 3.296e-05 [cse]: 0.00018548 [a_3]: 0.00010724 [Cycle 3]: 0.00133082, [45] [expand_dump_flag]: 2.79999e-06 [switch_simplify]: 1.496e-05 [loop_unroll]: 1.25e-05 [a_1]: 0.00037175 [with_stream_mark]: 2.463e-05 [recompute_prepare]: 1.349e-05 [updatestate_depend_eliminate]: 7.77998e-06 [updatestate_assign_eliminate]: 6.24999e-06 [updatestate_loads_eliminate]: 6.16e-06 [parameter_eliminate]: 2.22001e-06 [a_2]: 0.00016296 [accelerated_algorithm]: 1.928e-05 [shard]: 2.48e-06 [meta_shard_fg_expand]: 3.41001e-06 [shard_inline]: 1.286e-05 [merge_send_recv]: 1.43e-05 [auto_parallel]: 1.407e-05 [parallel]: 1.072e-05 [flash_sp]: 1.87001e-06 [merge_comm]: 8.38001e-06 [allreduce_fusion]: 6.70998e-06 [matmul_add_comm_reduction]: 1.53e-05 [allreduce_slice_to_reducescatter]: 8.29983e-07 [virtual_shard_identity]: 1.531e-05 [virtual_dataset]: 1.411e-05 [get_grad_eliminate_]: 1.129e-05 [virtual_output]: 1.068e-05 [merge_forward]: 7.51001e-06 [cell_reuse_recompute_pass]: 2.98e-06 [offload_activation]: 1.629e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.72e-05 [merge_recompute_call_nodes]: 1.50001e-06 [before_grad]: 2.123e-05 [set_forward_comm_id_for_comm_node_pass]: 9.14e-06 [meta_fg_expand]: 5.84999e-06 [flash_sp_send_recv_attached]: 1.91e-06 [receive_attached]: 2.74001e-06 [after_resolve]: 1.968e-05 [a_after_grad]: 1.92e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 3.06999e-06 [auto_monad_grad]: 2.14e-06 [auto_monad_eliminator]: 2.66e-05 [cse]: 4.582e-05 [a_3]: 7.363e-05 [py_interpret_to_execute_after_opt_a]: 2.445e-05 [slice_cell_reuse_recomputed_activation]: 2.17001e-06 [rewriter_after_opt_a]: 0.0001293 [convert_after_rewriter]: 1.141e-05 [order_py_execute_after_rewriter]: 9.34998e-06 [mutable_eliminate]: 0.00097127 [opt_b]: 0.00045761, [1] [Cycle 1]: 0.00044597, [7] [b_1]: 0.00028043 [b_2]: 1.435e-05 [updatestate_depend_eliminate]: 1.504e-05 [updatestate_assign_eliminate]: 7.23999e-06 [updatestate_loads_eliminate]: 7.50998e-06 [renormalize]: 2.30008e-07 [cse]: 6.671e-05 [optimize_parallel_all_gather_comm]: 3.275e-05 [overlap_param_gather]: 1.90001e-06 [cconv]: 4.53e-05 [loop_unroll]: 0.00068514 [opt_after_cconv]: 0.00024507, [1] [Cycle 1]: 0.00023395, [7] [c_1]: 8.705e-05 [parameter_eliminate]: 7.16001e-06 [updatestate_depend_eliminate]: 1.438e-05 [updatestate_assign_eliminate]: 7.46999e-06 [updatestate_loads_eliminate]: 6.84001e-06 [cse]: 5.386e-05 [renormalize]: 6.60017e-07 [remove_dup_value]: 2.48e-05 [tuple_transform]: 0.00015356, [1] [Cycle 1]: 0.00014627, [4] [d_1]: 9.867e-05 [none_parameter_eliminate]: 2.18002e-06 [renormalize]: 2.50002e-07 [switch_simplify]: 1.423e-05 [partial_unused_args_eliminate]: 2.46e-06 [add_recomputation]: 0.00012844 [cse_after_recomputation]: 5.775e-05, [1] [Cycle 1]: 5.106e-05, [1] [cse]: 4.141e-05 [environ_conv]: 1.669e-05 [swap_dp_allreduce_reducescatter]: 1.325e-05 [bias_add_comm_swap]: 3.28998e-06 [label_micro_interleaved_index]: 7.05e-06 [label_fine_grained_interleaved_index]: 3.93001e-06 [merge_cast_opt]: 1.69e-06 [slice_recompute_activation]: 2.34001e-06 [micro_interleaved_order_control]: 2.63998e-06 [assign_add_opt]: 1.70001e-06 [ForceFp32Comm]: 1.32e-06 [remove_cast_before_assign_add]: 1.22e-06 [full_micro_interleaved_order_control]: 2.81999e-06 [reorder_send_recv_between_fp_bp]: 3.4e-06 [comm_op_add_attrs]: 1.34e-06 [add_comm_op_reuse_tag]: 1.04003e-06 [interleave_split_concat_branches]: 1.57999e-06 [interleave_parallel_branches]: 1.46002e-06 [overlap_opt_shard_in_pipeline]: 2.55002e-06 [overlap_opt_shard_grad_in_pipeline]: 2.01e-06 [control_data_broadcast_order]: 3.093e-05 [grouped_pairwise_exchange_alltoall]: 1.76e-06 [offloading_packed_experts]: 8.39002e-06 [overlap_recompute_and_grad_model_parallel]: 8.76002e-06 [overlap_grad_matmul_and_grad_allreduce]: 2.13998e-06 [overlap_recompute_allgather_and_fa_grad]: 1.70001e-06 [overlap_recompute_comm]: 3.02002e-06 [overlap_grad_ring_attention]: 8.11002e-06 [overlap_grad_flash_sp]: 4.014e-05 [begin_end_overlap_inline]: 5.89993e-07 [split_matmul_comm_elemetwise]: 2.64001e-06 [split_layernorm_comm]: 2.09e-06 [handle_group_info]: 1.06002e-06 [symbol_engine_optimizer]: 0.00018209, [1] [Cycle 1]: 0.0001728, [6] [build]: 1.973e-05 [elim_shapecalc]: 3.262e-05 [elim_not_effective]: 2.939e-05 [opt_reshape]: 1.594e-05 [fold_const_symbol]: 2.406e-05 [renormalize]: 4.80009e-07 [detach_backward]: 2.29001e-06 [pipeline_parallel_scheduler]: 1.92999e-06 [auto_monad_reorder]: 5.065e-05 [get_jit_bprop_graph]: 2.25002e-06 [rewriter_after_jit_bprop_graph]: 8.65001e-06 [opt_after_jit_grad]: 0.0009466 [validate]: 0.00023112 [backend_pass]: 1.24998e-06 [task_emit]: 0.190694 [execute]: 1.167e-05 Sums bootstrap : 0.000652s : 0.08% type_inference : 0.152796s : 19.03% event_method : 0.000074s : 0.01% auto_monad : 0.000463s : 0.06% graph_reusing : 0.000011s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000057s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000011s : 0.00% parallel-infer-symbol : 0.000035s : 0.00% pre_auto_parallel : 0.000072s : 0.01% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000062s : 0.01% optimize.rewriter_before_opt_a : 0.000207s : 0.03% optimize.opt_a.expand_dump_flag : 0.000032s : 0.00% optimize.opt_a.switch_simplify : 0.000294s : 0.04% optimize.opt_a.loop_unroll : 0.000198s : 0.02% optimize.opt_a.a_1 : 0.168015s : 20.92% optimize.opt_a.with_stream_mark : 0.000123s : 0.02% optimize.opt_a.recompute_prepare : 0.000097s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000074s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000026s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000026s : 0.00% optimize.opt_a.parameter_eliminate : 0.000009s : 0.00% optimize.opt_a.a_2 : 0.000717s : 0.09% optimize.opt_a.accelerated_algorithm : 0.000094s : 0.01% optimize.opt_a.shard : 0.000008s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000024s : 0.00% optimize.opt_a.shard_inline : 0.000047s : 0.01% optimize.opt_a.merge_send_recv : 0.000053s : 0.01% optimize.opt_a.auto_parallel : 0.000052s : 0.01% optimize.opt_a.parallel : 0.000051s : 0.01% optimize.opt_a.flash_sp : 0.000027s : 0.00% optimize.opt_a.merge_comm : 0.000030s : 0.00% optimize.opt_a.allreduce_fusion : 0.000026s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000081s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000003s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000063s : 0.01% optimize.opt_a.virtual_dataset : 0.000057s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000048s : 0.01% optimize.opt_a.virtual_output : 0.000046s : 0.01% optimize.opt_a.merge_forward : 0.000031s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000007s : 0.00% optimize.opt_a.offload_activation : 0.000071s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000104s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000086s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000035s : 0.00% optimize.opt_a.meta_fg_expand : 0.008665s : 1.08% optimize.opt_a.flash_sp_send_recv_attached : 0.000010s : 0.00% optimize.opt_a.receive_attached : 0.000009s : 0.00% optimize.opt_a.after_resolve : 0.000210s : 0.03% optimize.opt_a.a_after_grad : 0.000306s : 0.04% optimize.opt_a.renormalize : 0.176730s : 22.01% optimize.opt_a.add_forward_monad_depend : 0.000057s : 0.01% optimize.opt_a.auto_monad_grad : 0.000036s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000262s : 0.03% optimize.opt_a.cse : 0.000642s : 0.08% optimize.opt_a.a_3 : 0.095934s : 11.95% optimize.py_interpret_to_execute_after_opt_a : 0.000024s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000129s : 0.02% optimize.convert_after_rewriter : 0.000011s : 0.00% optimize.order_py_execute_after_rewriter : 0.000009s : 0.00% optimize.mutable_eliminate : 0.000971s : 0.12% optimize.opt_b.b_1 : 0.000280s : 0.03% optimize.opt_b.b_2 : 0.000014s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000015s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000007s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000008s : 0.00% optimize.opt_b.renormalize : 0.000000s : 0.00% optimize.opt_b.cse : 0.000067s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000033s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000045s : 0.01% optimize.loop_unroll : 0.000685s : 0.09% optimize.opt_after_cconv.c_1 : 0.000087s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000014s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.cse : 0.000054s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000025s : 0.00% optimize.tuple_transform.d_1 : 0.000099s : 0.01% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000014s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000128s : 0.02% optimize.cse_after_recomputation.cse : 0.000041s : 0.01% optimize.environ_conv : 0.000017s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000013s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000007s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000004s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000002s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000003s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000031s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000008s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000009s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000008s : 0.00% optimize.overlap_grad_flash_sp : 0.000040s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000020s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000033s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000029s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000016s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000024s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000051s : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000009s : 0.00% opt_after_jit_grad : 0.000947s : 0.12% validate : 0.000231s : 0.03% backend_pass : 0.000001s : 0.00% task_emit : 0.190694s : 23.75% execute : 0.000012s : 0.00% Time group info: ------[substitution.] 0.164391 310 0.01% : 0.000019s : 4: substitution.depend_value_elim 0.00% : 0.000004s : 7: substitution.elim_not_effective 0.01% : 0.000009s : 7: substitution.float_depend_g_call 0.00% : 0.000005s : 2: substitution.float_tuple_getitem_switch 0.00% : 0.000003s : 7: substitution.fold_const_symbol 0.61% : 0.001008s : 4: substitution.getattr_setattr_resolve 0.01% : 0.000011s : 9: substitution.graph_param_transform 0.00% : 0.000003s : 2: substitution.incorporate_call 0.00% : 0.000002s : 2: substitution.incorporate_call_switch 98.95% : 0.162662s : 24: substitution.inline 0.03% : 0.000052s : 5: substitution.inline_without_move 0.01% : 0.000015s : 28: substitution.j_node_and_user_rematch 0.02% : 0.000028s : 3: substitution.less_batch_normalization 0.01% : 0.000018s : 9: substitution.minmaximum_grad 0.01% : 0.000011s : 7: substitution.partial_eliminate 0.01% : 0.000019s : 28: substitution.remove_not_recompute_node 0.10% : 0.000157s : 38: substitution.replace_applicator 0.01% : 0.000020s : 19: substitution.replace_old_param 0.00% : 0.000005s : 1: substitution.set_cell_output_no_recompute 0.02% : 0.000037s : 9: substitution.tuple_list_convert_item_index_to_positive 0.01% : 0.000017s : 9: substitution.tuple_list_get_item_const_eliminator 0.02% : 0.000026s : 9: substitution.tuple_list_get_item_depend_reorder 0.08% : 0.000133s : 31: substitution.tuple_list_get_item_eliminator 0.02% : 0.000027s : 9: substitution.tuple_list_get_set_item_eliminator 0.02% : 0.000041s : 16: substitution.updatestate_pure_node_eliminater 0.04% : 0.000059s : 21: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.152671 2 98.41% : 0.150243s : 1: type_inference.infer 1.59% : 0.002428s : 1: type_inference.specialize ------[replace.] 0.000864 49 9.53% : 0.000082s : 3: replace.getattr_setattr_resolve 44.89% : 0.000388s : 24: replace.inline 10.33% : 0.000089s : 1: replace.replace_applicator 33.56% : 0.000290s : 20: replace.tuple_list_get_item_eliminator 1.69% : 0.000015s : 1: replace.updatestate_useless_node_eliminater ------[match.] 0.163718 49 0.58% : 0.000956s : 3: match.getattr_setattr_resolve 99.34% : 0.162636s : 24: match.inline 0.02% : 0.000027s : 1: match.replace_applicator 0.05% : 0.000077s : 20: match.tuple_list_get_item_eliminator 0.01% : 0.000020s : 1: match.updatestate_useless_node_eliminater ------[predicate.] 0.001463 8683 0.84% : 0.000012s : 85: predicate.accumulaten_eliminater 0.40% : 0.000006s : 9: predicate.ad_related_special_op_eliminate 0.36% : 0.000005s : 37: predicate.addn_check_dump 0.88% : 0.000013s : 85: predicate.addn_zero_filter 0.85% : 0.000012s : 85: predicate.adjust_all_reduce_mul_add 1.94% : 0.000028s : 122: predicate.arithmetic_simplify 0.93% : 0.000014s : 85: predicate.cast_eliminate 2.68% : 0.000039s : 260: predicate.check_bprop_eliminate 0.35% : 0.000005s : 37: predicate.compare_switch_simplify 0.05% : 0.000001s : 9: predicate.const_output_eliminate 0.41% : 0.000006s : 37: predicate.depend_value_elim 0.89% : 0.000013s : 85: predicate.dict_get_item_const_eliminator 0.97% : 0.000014s : 85: predicate.dict_get_item_eliminator 0.87% : 0.000013s : 85: predicate.dict_set_item_eliminator 0.35% : 0.000005s : 18: predicate.dumpgradient_eliminate 0.08% : 0.000001s : 9: predicate.elim_not_effective 0.13% : 0.000002s : 9: predicate.elim_shapecalc_of_broadcastargs 0.88% : 0.000013s : 94: predicate.environ_add_const_eliminate 0.87% : 0.000013s : 94: predicate.environ_get_add_eliminate 0.85% : 0.000012s : 94: predicate.environ_get_depend_swap 1.32% : 0.000019s : 131: predicate.environ_get_eliminate 0.87% : 0.000013s : 94: predicate.environ_get_set_eliminate 1.24% : 0.000018s : 130: predicate.exchange_switch_depend_value 2.04% : 0.000030s : 130: predicate.float_depend_g_call 0.38% : 0.000006s : 37: predicate.float_environ_get_switch 0.47% : 0.000007s : 46: predicate.float_tuple_getitem_switch 0.07% : 0.000001s : 9: predicate.fold_const_symbol 0.41% : 0.000006s : 37: predicate.get_grad_eliminate 0.51% : 0.000007s : 20: predicate.getattr_setattr_resolve 0.08% : 0.000001s : 9: predicate.graph_param_transform 0.35% : 0.000005s : 37: predicate.incorporate_call 0.32% : 0.000005s : 37: predicate.incorporate_call_switch 4.38% : 0.000064s : 307: predicate.inline 1.55% : 0.000023s : 111: predicate.inline_without_move 0.22% : 0.000003s : 37: predicate.j_node_and_user_rematch 0.62% : 0.000009s : 37: predicate.less_batch_normalization 1.27% : 0.000019s : 123: predicate.list_to_tuple_eliminator_ 1.97% : 0.000029s : 208: predicate.load_eliminater 0.38% : 0.000006s : 9: predicate.loop_unroll_after_grad 1.93% : 0.000028s : 197: predicate.loop_unroll_before_grad 1.07% : 0.000016s : 103: predicate.make_slice_get_slice_eliminator 0.36% : 0.000005s : 37: predicate.merge_addn 2.51% : 0.000037s : 256: predicate.micro_step_allgather_replace 2.48% : 0.000036s : 256: predicate.mini_step_allgather_replace 0.87% : 0.000013s : 85: predicate.minmaximum_grad 2.72% : 0.000040s : 9: predicate.mutable_eliminate 0.12% : 0.000002s : 9: predicate.opt_reshape 0.11% : 0.000002s : 9: predicate.parallel_virtual_node 2.55% : 0.000037s : 130: predicate.partial_defer_inline 1.22% : 0.000018s : 114: predicate.partial_eliminate 0.81% : 0.000012s : 85: predicate.print_const_string_wrapper 0.38% : 0.000006s : 37: predicate.reduce_all_const_elim 1.15% : 0.000017s : 85: predicate.reduce_eliminate 2.01% : 0.000029s : 208: predicate.redundant_stop_gradient_eliminater 0.29% : 0.000004s : 37: predicate.remove_not_recompute_node 2.34% : 0.000034s : 363: predicate.replace_applicator 0.69% : 0.000010s : 111: predicate.replace_old_param 0.11% : 0.000002s : 9: predicate.reset_defer_inline 0.85% : 0.000012s : 85: predicate.reshape_eliminate 2.92% : 0.000043s : 256: predicate.row_tensor_add_zeros_like 0.12% : 0.000002s : 9: predicate.row_tensor_eliminate 2.77% : 0.000041s : 260: predicate.same_eliminate 0.31% : 0.000005s : 39: predicate.set_cell_output_no_recompute 0.50% : 0.000007s : 37: predicate.shard_identity_eliminate 0.22% : 0.000003s : 18: predicate.special_op_eliminate 0.43% : 0.000006s : 37: predicate.specialize_transform 3.80% : 0.000056s : 256: predicate.split_environ_get_set_with_tuple_value 1.33% : 0.000019s : 111: predicate.stack_unstack_eliminate 0.12% : 0.000002s : 9: predicate.switch_call_monad_eliminater 1.41% : 0.000021s : 130: predicate.switch_defer_inline 3.95% : 0.000058s : 390: predicate.switch_layer_defer_inline 6.08% : 0.000089s : 373: predicate.switch_simplify 0.86% : 0.000013s : 85: predicate.tile_eliminate 0.90% : 0.000013s : 85: predicate.transpose_eliminate 1.35% : 0.000020s : 103: predicate.tuple_list_convert_item_index_to_positive 1.18% : 0.000017s : 103: predicate.tuple_list_get_item_const_eliminator 1.22% : 0.000018s : 103: predicate.tuple_list_get_item_depend_reorder 2.39% : 0.000035s : 160: predicate.tuple_list_get_item_eliminator 1.25% : 0.000018s : 103: predicate.tuple_list_get_set_item_eliminator 1.63% : 0.000024s : 140: predicate.tuple_list_set_item_eliminator 1.21% : 0.000018s : 123: predicate.tuple_to_list_eliminator_ 1.90% : 0.000028s : 208: predicate.updatestate_pure_node_eliminater 2.45% : 0.000036s : 246: predicate.updatestate_useless_node_eliminater 0.12% : 0.000002s : 9: predicate.value_based_eliminate 0.41% : 0.000006s : 37: predicate.virtual_dataset_eliminate 0.39% : 0.000006s : 37: predicate.virtual_output_eliminate 0.10% : 0.000001s : 9: predicate.virtual_view_grad_eliminate 0.14% : 0.000002s : 9: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.006020 53 50.59% : 0.003046s : 23: func_graph_cloner_run.FuncGraphClonerGraph 49.41% : 0.002974s : 30: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.719085 247 0.00% : 0.000005s : 1: ForceFp32Comm 0.31% : 0.005272s : 1: add_attr 0.31% : 0.005254s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000137s : 1: add_recomputation 0.00% : 0.000006s : 1: assign_add_opt 0.03% : 0.000484s : 1: auto_monad 0.00% : 0.000057s : 1: auto_monad_reorder 0.00% : 0.000009s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.04% : 0.000685s : 1: bootstrap 0.00% : 0.000050s : 1: cconv 0.00% : 0.000005s : 1: comm_op_add_attrs 0.00% : 0.000036s : 1: control_data_broadcast_order 0.00% : 0.000016s : 1: convert_after_rewriter 0.00% : 0.000062s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000021s : 1: environ_conv 0.00% : 0.000085s : 1: event_method 0.00% : 0.000021s : 1: execute 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000018s : 1: graph_reusing 0.00% : 0.000006s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000008s : 1: insert-virtual-dataset 0.00% : 0.000007s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000008s : 1: label_fine_grained_interleaved_index 0.00% : 0.000011s : 1: label_micro_interleaved_index 0.04% : 0.000701s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.06% : 0.000989s : 1: mutable_eliminate 0.00% : 0.000012s : 1: offloading_packed_experts 0.00% : 0.000032s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000072s : 1: opt.transform.mutable_eliminate 15.48% : 0.266049s : 125: opt.transform.opt_a 0.00% : 0.000085s : 1: opt.transform.opt_after_cconv 0.00% : 0.000071s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000251s : 28: opt.transform.opt_b 0.07% : 0.001162s : 2: opt.transform.opt_resolve 0.01% : 0.000110s : 2: opt.transform.opt_trans_graph 0.01% : 0.000096s : 4: opt.transform.symbol_engine_opt 26.46% : 0.454854s : 1: opt_a 0.01% : 0.000251s : 1: opt_after_cconv 0.06% : 0.000970s : 1: opt_after_jit_grad 0.03% : 0.000463s : 1: opt_b 26.69% : 0.458841s : 1: optimize 0.00% : 0.000037s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000013s : 1: order_py_execute_after_rewriter 0.00% : 0.000045s : 1: overlap_grad_flash_sp 0.00% : 0.000006s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000012s : 1: overlap_grad_ring_attention 0.00% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000012s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000007s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000013s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000044s : 1: parallel-infer-symbol 0.00% : 0.000005s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.00% : 0.000078s : 1: pre_auto_parallel 0.00% : 0.000067s : 1: py_interpret_to_execute 0.00% : 0.000029s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.00% : 0.000029s : 1: remove_dup_value 9.80% : 0.168497s : 2: renormalize.infer 0.48% : 0.008196s : 2: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000013s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000141s : 1: rewriter_after_opt_a 0.01% : 0.000214s : 1: rewriter_before_opt_a 0.00% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000007s : 1: split_matmul_comm_elemetwise 0.00% : 0.000018s : 1: swap_dp_allreduce_reducescatter 0.01% : 0.000185s : 1: symbol_engine_optimizer 11.09% : 0.190721s : 1: task_emit 0.01% : 0.000157s : 1: tuple_transform 8.89% : 0.152831s : 1: type_inference 0.02% : 0.000310s : 1: validate . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[0-shape1] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[0-shape1],max_mem:4.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:41:40.748.920 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.316128, [24] [bootstrap]: 0.00078208 [type_inference]: 0.00705504 [event_method]: 2.683e-05 [auto_monad]: 0.00025768 [graph_reusing]: 6.81001e-06 [inline]: 2.73998e-06 [add_attr]: 0.191545, [1] [add_attr_with_inline]: 0.191526, [1] [Cycle 1]: 9.86e-05, [2] [tag_attr]: 2.007e-05 [meta_addattr_fg_expand]: 4.4e-06 [parallel-infer-symbol]: 4.1e-06 [pre_auto_parallel]: 5.017e-05 [insert-virtual-dataset]: 2.69001e-06 [parallel-infer-symbol-second]: 7.30011e-07 [dataset_repeat_opt]: 2.53e-06 [pipeline_split]: 1.95001e-06 [optimize]: 0.00908005, [53] [py_interpret_to_execute]: 3.798e-05 [rewriter_before_opt_a]: 8.752e-05 [opt_a]: 0.00496616, [2] [Cycle 1]: 0.00336218, [45] [expand_dump_flag]: 7.65998e-06 [switch_simplify]: 3.322e-05 [loop_unroll]: 1.55e-05 [a_1]: 0.00068927 [with_stream_mark]: 2.582e-05 [recompute_prepare]: 1.401e-05 [updatestate_depend_eliminate]: 6.91001e-06 [updatestate_assign_eliminate]: 5.72999e-06 [updatestate_loads_eliminate]: 5.47999e-06 [parameter_eliminate]: 2.16e-06 [a_2]: 0.00014685 [accelerated_algorithm]: 1.737e-05 [shard]: 4.13999e-06 [meta_shard_fg_expand]: 2.68e-06 [shard_inline]: 3.3e-05 [merge_send_recv]: 1.405e-05 [auto_parallel]: 1.259e-05 [parallel]: 4.232e-05 [flash_sp]: 1.197e-05 [merge_comm]: 6.26e-06 [allreduce_fusion]: 6.38e-06 [matmul_add_comm_reduction]: 2.562e-05 [allreduce_slice_to_reducescatter]: 9.09989e-07 [virtual_shard_identity]: 1.981e-05 [virtual_dataset]: 1.14e-05 [get_grad_eliminate_]: 1.444e-05 [virtual_output]: 1.395e-05 [merge_forward]: 1.002e-05 [cell_reuse_recompute_pass]: 2.81e-06 [offload_activation]: 4.355e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.969e-05 [merge_recompute_call_nodes]: 1.64e-06 [before_grad]: 2.655e-05 [set_forward_comm_id_for_comm_node_pass]: 8.22e-06 [meta_fg_expand]: 5.56e-06 [flash_sp_send_recv_attached]: 3.97002e-06 [receive_attached]: 3.04999e-06 [after_resolve]: 2.157e-05 [a_after_grad]: 2.03e-05 [renormalize]: 0.00128409 [add_forward_monad_depend]: 1.215e-05 [auto_monad_grad]: 2.74001e-06 [auto_monad_eliminator]: 2.785e-05 [cse]: 4.563e-05 [a_3]: 9.041e-05 [Cycle 2]: 0.0015845, [45] [expand_dump_flag]: 3.11001e-06 [switch_simplify]: 1.325e-05 [loop_unroll]: 1.156e-05 [a_1]: 0.00025419 [with_stream_mark]: 2.018e-05 [recompute_prepare]: 1.118e-05 [updatestate_depend_eliminate]: 1.159e-05 [updatestate_assign_eliminate]: 9.39998e-06 [updatestate_loads_eliminate]: 1.246e-05 [parameter_eliminate]: 1.96e-06 [a_2]: 0.00014388 [accelerated_algorithm]: 1.204e-05 [shard]: 5.56e-06 [meta_shard_fg_expand]: 3.35e-06 [shard_inline]: 1.381e-05 [merge_send_recv]: 1.957e-05 [auto_parallel]: 2.279e-05 [parallel]: 1.131e-05 [flash_sp]: 5.37999e-06 [merge_comm]: 7.28e-06 [allreduce_fusion]: 6.17001e-06 [matmul_add_comm_reduction]: 2.217e-05 [allreduce_slice_to_reducescatter]: 9.29984e-07 [virtual_shard_identity]: 1.502e-05 [virtual_dataset]: 1.438e-05 [get_grad_eliminate_]: 1.525e-05 [virtual_output]: 1.357e-05 [merge_forward]: 8.20999e-06 [cell_reuse_recompute_pass]: 2.99999e-06 [offload_activation]: 2.398e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.078e-05 [merge_recompute_call_nodes]: 1.45999e-06 [before_grad]: 3.755e-05 [set_forward_comm_id_for_comm_node_pass]: 1.01e-05 [meta_fg_expand]: 4.32e-06 [flash_sp_send_recv_attached]: 2.19001e-06 [receive_attached]: 7.01999e-06 [after_resolve]: 2.571e-05 [a_after_grad]: 2.038e-05 [renormalize]: 9.00181e-08 [add_forward_monad_depend]: 3.31001e-06 [auto_monad_grad]: 2.02999e-06 [auto_monad_eliminator]: 2.081e-05 [cse]: 3.642e-05 [a_3]: 6.449e-05 [py_interpret_to_execute_after_opt_a]: 2.655e-05 [slice_cell_reuse_recomputed_activation]: 2.93e-06 [rewriter_after_opt_a]: 0.00027423 [convert_after_rewriter]: 1.387e-05 [order_py_execute_after_rewriter]: 9.11998e-06 [mutable_eliminate]: 0.00089514 [opt_b]: 0.00054437, [1] [Cycle 1]: 0.0005246, [7] [b_1]: 0.00034519 [b_2]: 2.018e-05 [updatestate_depend_eliminate]: 1.013e-05 [updatestate_assign_eliminate]: 1.102e-05 [updatestate_loads_eliminate]: 5.91998e-06 [renormalize]: 7.39994e-07 [cse]: 4.21e-05 [optimize_parallel_all_gather_comm]: 2.791e-05 [overlap_param_gather]: 2.04999e-06 [cconv]: 3.28e-05 [loop_unroll]: 0.00066288 [opt_after_cconv]: 0.00019712, [1] [Cycle 1]: 0.000188, [7] [c_1]: 8.418e-05 [parameter_eliminate]: 4.43999e-06 [updatestate_depend_eliminate]: 8.78001e-06 [updatestate_assign_eliminate]: 5.02999e-06 [updatestate_loads_eliminate]: 5.07e-06 [cse]: 2.931e-05 [renormalize]: 6.59988e-07 [remove_dup_value]: 1.985e-05 [tuple_transform]: 0.0001874, [1] [Cycle 1]: 0.00018069, [4] [d_1]: 0.00010131 [none_parameter_eliminate]: 2.04999e-06 [renormalize]: 3.30008e-07 [switch_simplify]: 1.265e-05 [partial_unused_args_eliminate]: 2.66e-06 [add_recomputation]: 0.00012482 [cse_after_recomputation]: 3.964e-05, [1] [Cycle 1]: 3.334e-05, [1] [cse]: 2.031e-05 [environ_conv]: 9.25001e-06 [swap_dp_allreduce_reducescatter]: 1.069e-05 [bias_add_comm_swap]: 3.71001e-06 [label_micro_interleaved_index]: 1.043e-05 [label_fine_grained_interleaved_index]: 3.23e-06 [merge_cast_opt]: 1.92999e-06 [slice_recompute_activation]: 2.54001e-06 [micro_interleaved_order_control]: 2.84999e-06 [assign_add_opt]: 1.66e-06 [ForceFp32Comm]: 1.32e-06 [remove_cast_before_assign_add]: 9.5999e-07 [full_micro_interleaved_order_control]: 5.66e-06 [reorder_send_recv_between_fp_bp]: 3.46001e-06 [comm_op_add_attrs]: 1.59998e-06 [add_comm_op_reuse_tag]: 1.29e-06 [interleave_split_concat_branches]: 5.64e-06 [interleave_parallel_branches]: 4.74e-06 [overlap_opt_shard_in_pipeline]: 1.44e-06 [overlap_opt_shard_grad_in_pipeline]: 1.89e-06 [control_data_broadcast_order]: 2.347e-05 [grouped_pairwise_exchange_alltoall]: 2.34999e-06 [offloading_packed_experts]: 1.671e-05 [overlap_recompute_and_grad_model_parallel]: 1.084e-05 [overlap_grad_matmul_and_grad_allreduce]: 1.69e-06 [overlap_recompute_allgather_and_fa_grad]: 1.85001e-06 [overlap_recompute_comm]: 2.89999e-06 [overlap_grad_ring_attention]: 1.417e-05 [overlap_grad_flash_sp]: 3.85e-05 [begin_end_overlap_inline]: 7.79983e-07 [split_matmul_comm_elemetwise]: 2.49999e-06 [split_layernorm_comm]: 1.92999e-06 [handle_group_info]: 1.85001e-06 [symbol_engine_optimizer]: 0.00022816, [1] [Cycle 1]: 0.00021983, [6] [build]: 7.18e-06 [elim_shapecalc]: 4.286e-05 [elim_not_effective]: 2.946e-05 [opt_reshape]: 2.508e-05 [fold_const_symbol]: 2.74e-05 [renormalize]: 2.59985e-07 [detach_backward]: 2.41e-06 [pipeline_parallel_scheduler]: 1.54e-06 [auto_monad_reorder]: 4.036e-05 [get_jit_bprop_graph]: 1.99e-06 [rewriter_after_jit_bprop_graph]: 4.79998e-06 [opt_after_jit_grad]: 0.0913422 [validate]: 9.071e-05 [backend_pass]: 1.14e-06 [task_emit]: 0.0154027 [execute]: 1.03e-05 Sums bootstrap : 0.000782s : 0.64% type_inference : 0.007055s : 5.77% event_method : 0.000027s : 0.02% auto_monad : 0.000258s : 0.21% graph_reusing : 0.000007s : 0.01% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000020s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000050s : 0.04% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000038s : 0.03% optimize.rewriter_before_opt_a : 0.000088s : 0.07% optimize.opt_a.expand_dump_flag : 0.000011s : 0.01% optimize.opt_a.switch_simplify : 0.000046s : 0.04% optimize.opt_a.loop_unroll : 0.000027s : 0.02% optimize.opt_a.a_1 : 0.000943s : 0.77% optimize.opt_a.with_stream_mark : 0.000046s : 0.04% optimize.opt_a.recompute_prepare : 0.000025s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000019s : 0.02% optimize.opt_a.updatestate_assign_eliminate : 0.000015s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000018s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.00% optimize.opt_a.a_2 : 0.000291s : 0.24% optimize.opt_a.accelerated_algorithm : 0.000029s : 0.02% optimize.opt_a.shard : 0.000010s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000006s : 0.00% optimize.opt_a.shard_inline : 0.000047s : 0.04% optimize.opt_a.merge_send_recv : 0.000034s : 0.03% optimize.opt_a.auto_parallel : 0.000035s : 0.03% optimize.opt_a.parallel : 0.000054s : 0.04% optimize.opt_a.flash_sp : 0.000017s : 0.01% optimize.opt_a.merge_comm : 0.000014s : 0.01% optimize.opt_a.allreduce_fusion : 0.000013s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000048s : 0.04% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000035s : 0.03% optimize.opt_a.virtual_dataset : 0.000026s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000030s : 0.02% optimize.opt_a.virtual_output : 0.000028s : 0.02% optimize.opt_a.merge_forward : 0.000018s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000006s : 0.00% optimize.opt_a.offload_activation : 0.000068s : 0.06% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000070s : 0.06% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000064s : 0.05% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000018s : 0.01% optimize.opt_a.meta_fg_expand : 0.000010s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000006s : 0.01% optimize.opt_a.receive_attached : 0.000010s : 0.01% optimize.opt_a.after_resolve : 0.000047s : 0.04% optimize.opt_a.a_after_grad : 0.000041s : 0.03% optimize.opt_a.renormalize : 0.001284s : 1.05% optimize.opt_a.add_forward_monad_depend : 0.000015s : 0.01% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000049s : 0.04% optimize.opt_a.cse : 0.000082s : 0.07% optimize.opt_a.a_3 : 0.000155s : 0.13% optimize.py_interpret_to_execute_after_opt_a : 0.000027s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000274s : 0.22% optimize.convert_after_rewriter : 0.000014s : 0.01% optimize.order_py_execute_after_rewriter : 0.000009s : 0.01% optimize.mutable_eliminate : 0.000895s : 0.73% optimize.opt_b.b_1 : 0.000345s : 0.28% optimize.opt_b.b_2 : 0.000020s : 0.02% optimize.opt_b.updatestate_depend_eliminate : 0.000010s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000011s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000042s : 0.03% optimize.optimize_parallel_all_gather_comm : 0.000028s : 0.02% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000033s : 0.03% optimize.loop_unroll : 0.000663s : 0.54% optimize.opt_after_cconv.c_1 : 0.000084s : 0.07% optimize.opt_after_cconv.parameter_eliminate : 0.000004s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000009s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.cse : 0.000029s : 0.02% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000020s : 0.02% optimize.tuple_transform.d_1 : 0.000101s : 0.08% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000013s : 0.01% optimize.partial_unused_args_eliminate : 0.000003s : 0.00% optimize.add_recomputation : 0.000125s : 0.10% optimize.cse_after_recomputation.cse : 0.000020s : 0.02% optimize.environ_conv : 0.000009s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000011s : 0.01% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000010s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000006s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000002s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000006s : 0.00% optimize.interleave_parallel_branches : 0.000005s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000023s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000017s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000011s : 0.01% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000014s : 0.01% optimize.overlap_grad_flash_sp : 0.000039s : 0.03% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000002s : 0.00% optimize.symbol_engine_optimizer.build : 0.000007s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000043s : 0.04% optimize.symbol_engine_optimizer.elim_not_effective : 0.000029s : 0.02% optimize.symbol_engine_optimizer.opt_reshape : 0.000025s : 0.02% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000027s : 0.02% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000040s : 0.03% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000005s : 0.00% opt_after_jit_grad : 0.091342s : 74.75% validate : 0.000091s : 0.07% backend_pass : 0.000001s : 0.00% task_emit : 0.015403s : 12.61% execute : 0.000010s : 0.01% Time group info: ------[substitution.] 0.000246 55 4.40% : 0.000011s : 2: substitution.depend_value_elim 1.32% : 0.000003s : 5: substitution.elim_not_effective 1.03% : 0.000003s : 5: substitution.fold_const_symbol 3.53% : 0.000009s : 6: substitution.graph_param_transform 64.49% : 0.000159s : 1: substitution.inline 2.97% : 0.000007s : 10: substitution.j_node_and_user_rematch 4.24% : 0.000010s : 10: substitution.remove_not_recompute_node 2.28% : 0.000006s : 2: substitution.replace_old_param 7.84% : 0.000019s : 6: substitution.updatestate_pure_node_eliminater 7.89% : 0.000019s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.006968 2 83.12% : 0.005792s : 1: type_inference.infer 16.88% : 0.001177s : 1: type_inference.specialize ------[replace.] 0.000032 1 100.00% : 0.000032s : 1: replace.inline ------[match.] 0.000157 1 100.00% : 0.000157s : 1: match.inline ------[predicate.] 0.000293 1419 0.82% : 0.000002s : 13: predicate.accumulaten_eliminater 1.99% : 0.000006s : 6: predicate.ad_related_special_op_eliminate 0.69% : 0.000002s : 12: predicate.addn_check_dump 1.02% : 0.000003s : 13: predicate.addn_zero_filter 0.82% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 2.14% : 0.000006s : 25: predicate.arithmetic_simplify 0.81% : 0.000002s : 13: predicate.cast_eliminate 0.84% : 0.000002s : 12: predicate.check_bprop_eliminate 0.75% : 0.000002s : 12: predicate.compare_switch_simplify 0.19% : 0.000001s : 6: predicate.const_output_eliminate 0.69% : 0.000002s : 12: predicate.depend_value_elim 0.74% : 0.000002s : 13: predicate.dict_get_item_const_eliminator 0.89% : 0.000003s : 13: predicate.dict_get_item_eliminator 0.76% : 0.000002s : 13: predicate.dict_set_item_eliminator 1.45% : 0.000004s : 12: predicate.dumpgradient_eliminate 0.41% : 0.000001s : 6: predicate.elim_not_effective 0.62% : 0.000002s : 6: predicate.elim_shapecalc_of_broadcastargs 1.15% : 0.000003s : 19: predicate.environ_add_const_eliminate 1.18% : 0.000003s : 19: predicate.environ_get_add_eliminate 1.05% : 0.000003s : 19: predicate.environ_get_depend_swap 1.72% : 0.000005s : 31: predicate.environ_get_eliminate 1.18% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.92% : 0.000003s : 14: predicate.exchange_switch_depend_value 1.29% : 0.000004s : 14: predicate.float_depend_g_call 0.86% : 0.000003s : 12: predicate.float_environ_get_switch 1.09% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.21% : 0.000001s : 6: predicate.fold_const_symbol 1.19% : 0.000003s : 12: predicate.get_grad_eliminate 0.25% : 0.000001s : 6: predicate.graph_param_transform 0.62% : 0.000002s : 12: predicate.incorporate_call 0.56% : 0.000002s : 12: predicate.incorporate_call_switch 5.07% : 0.000015s : 63: predicate.inline 1.14% : 0.000003s : 12: predicate.inline_without_move 0.31% : 0.000001s : 12: predicate.j_node_and_user_rematch 1.11% : 0.000003s : 12: predicate.less_batch_normalization 1.67% : 0.000005s : 25: predicate.list_to_tuple_eliminator_ 2.06% : 0.000006s : 38: predicate.load_eliminater 1.03% : 0.000003s : 6: predicate.loop_unroll_after_grad 1.20% : 0.000004s : 17: predicate.loop_unroll_before_grad 1.88% : 0.000006s : 25: predicate.make_slice_get_slice_eliminator 0.72% : 0.000002s : 12: predicate.merge_addn 0.72% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.72% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.80% : 0.000002s : 13: predicate.minmaximum_grad 1.04% : 0.000003s : 6: predicate.mutable_eliminate 0.76% : 0.000002s : 6: predicate.opt_reshape 0.59% : 0.000002s : 6: predicate.parallel_virtual_node 1.14% : 0.000003s : 14: predicate.partial_defer_inline 1.03% : 0.000003s : 19: predicate.partial_eliminate 0.78% : 0.000002s : 13: predicate.print_const_string_wrapper 0.81% : 0.000002s : 12: predicate.reduce_all_const_elim 0.93% : 0.000003s : 13: predicate.reduce_eliminate 2.13% : 0.000006s : 38: predicate.redundant_stop_gradient_eliminater 0.70% : 0.000002s : 12: predicate.remove_not_recompute_node 0.92% : 0.000003s : 25: predicate.replace_applicator 0.56% : 0.000002s : 12: predicate.replace_old_param 0.39% : 0.000001s : 6: predicate.reset_defer_inline 0.91% : 0.000003s : 13: predicate.reshape_eliminate 0.83% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.58% : 0.000002s : 6: predicate.row_tensor_eliminate 0.95% : 0.000003s : 12: predicate.same_eliminate 0.47% : 0.000001s : 12: predicate.set_cell_output_no_recompute 1.30% : 0.000004s : 12: predicate.shard_identity_eliminate 0.99% : 0.000003s : 12: predicate.special_op_eliminate 0.72% : 0.000002s : 12: predicate.specialize_transform 1.61% : 0.000005s : 12: predicate.split_environ_get_set_with_tuple_value 1.21% : 0.000004s : 12: predicate.stack_unstack_eliminate 0.39% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.91% : 0.000003s : 14: predicate.switch_defer_inline 1.42% : 0.000004s : 26: predicate.switch_layer_defer_inline 3.90% : 0.000011s : 49: predicate.switch_simplify 0.89% : 0.000003s : 13: predicate.tile_eliminate 0.94% : 0.000003s : 13: predicate.transpose_eliminate 1.64% : 0.000005s : 25: predicate.tuple_list_convert_item_index_to_positive 1.60% : 0.000005s : 25: predicate.tuple_list_get_item_const_eliminator 1.86% : 0.000005s : 25: predicate.tuple_list_get_item_depend_reorder 2.94% : 0.000009s : 37: predicate.tuple_list_get_item_eliminator 1.68% : 0.000005s : 25: predicate.tuple_list_get_set_item_eliminator 2.30% : 0.000007s : 37: predicate.tuple_list_set_item_eliminator 1.87% : 0.000005s : 25: predicate.tuple_to_list_eliminator_ 2.28% : 0.000007s : 38: predicate.updatestate_pure_node_eliminater 3.17% : 0.000009s : 50: predicate.updatestate_useless_node_eliminater 0.72% : 0.000002s : 6: predicate.value_based_eliminate 0.88% : 0.000003s : 12: predicate.virtual_dataset_eliminate 1.03% : 0.000003s : 12: predicate.virtual_output_eliminate 0.34% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.54% : 0.000002s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000797 4 2.66% : 0.000021s : 1: func_graph_cloner_run.FuncGraphClonerGraph 97.34% : 0.000776s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.520360 196 0.00% : 0.000010s : 1: ForceFp32Comm 36.81% : 0.191553s : 1: add_attr 36.81% : 0.191531s : 1: add_attr_with_inline 0.00% : 0.000012s : 1: add_comm_op_reuse_tag 0.03% : 0.000134s : 1: add_recomputation 0.00% : 0.000012s : 1: assign_add_opt 0.05% : 0.000272s : 1: auto_monad 0.01% : 0.000046s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000008s : 1: bias_add_comm_swap 0.16% : 0.000851s : 1: bootstrap 0.01% : 0.000038s : 1: cconv 0.00% : 0.000013s : 1: comm_op_add_attrs 0.01% : 0.000032s : 1: control_data_broadcast_order 0.00% : 0.000020s : 1: convert_after_rewriter 0.01% : 0.000048s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000013s : 1: environ_conv 0.01% : 0.000042s : 1: event_method 0.00% : 0.000018s : 1: execute 0.00% : 0.000016s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000012s : 1: graph_reusing 0.00% : 0.000020s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000009s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000011s : 1: insert-virtual-dataset 0.00% : 0.000016s : 1: interleave_parallel_branches 0.00% : 0.000016s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000015s : 1: label_micro_interleaved_index 0.13% : 0.000678s : 1: loop_unroll 0.00% : 0.000007s : 1: merge_cast_opt 0.00% : 0.000007s : 1: micro_interleaved_order_control 0.18% : 0.000911s : 1: mutable_eliminate 0.01% : 0.000027s : 1: offloading_packed_experts 0.00% : 0.000023s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000029s : 1: opt.transform.mutable_eliminate 0.34% : 0.001767s : 78: opt.transform.opt_a 0.02% : 0.000082s : 1: opt.transform.opt_after_cconv 0.01% : 0.000066s : 1: opt.transform.opt_after_jit_grad 0.06% : 0.000315s : 28: opt.transform.opt_b 0.02% : 0.000111s : 2: opt.transform.opt_trans_graph 0.02% : 0.000120s : 4: opt.transform.symbol_engine_opt 0.96% : 0.004979s : 1: opt_a 0.04% : 0.000202s : 1: opt_after_cconv 17.56% : 0.091371s : 1: opt_after_jit_grad 0.11% : 0.000550s : 1: opt_b 1.75% : 0.009088s : 1: optimize 0.01% : 0.000033s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000028s : 1: order_py_execute_after_rewriter 0.01% : 0.000043s : 1: overlap_grad_flash_sp 0.00% : 0.000009s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000018s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000010s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000019s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000008s : 1: parallel-infer-symbol-second 0.00% : 0.000016s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000006s : 1: pipeline_split 0.01% : 0.000056s : 1: pre_auto_parallel 0.01% : 0.000043s : 1: py_interpret_to_execute 0.01% : 0.000033s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.01% : 0.000029s : 1: remove_dup_value 0.12% : 0.000648s : 1: renormalize.infer 0.12% : 0.000623s : 1: renormalize.specialize 0.00% : 0.000010s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000014s : 1: rewriter_after_jit_bprop_graph 0.06% : 0.000289s : 1: rewriter_after_opt_a 0.02% : 0.000094s : 1: rewriter_before_opt_a 0.00% : 0.000007s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000010s : 1: slice_recompute_activation 0.00% : 0.000006s : 1: split_layernorm_comm 0.00% : 0.000009s : 1: split_matmul_comm_elemetwise 0.00% : 0.000019s : 1: swap_dp_allreduce_reducescatter 0.05% : 0.000236s : 1: symbol_engine_optimizer 2.96% : 0.015429s : 1: task_emit 0.04% : 0.000191s : 1: tuple_transform 1.36% : 0.007091s : 1: type_inference 0.03% : 0.000150s : 1: validate TotalTime = 0.391774, [24] [bootstrap]: 0.00047729 [type_inference]: 0.0039252 [event_method]: 1.618e-05 [auto_monad]: 0.00015936 [graph_reusing]: 6.23998e-06 [inline]: 2.37999e-06 [add_attr]: 0.0291235, [1] [add_attr_with_inline]: 0.0291072, [1] [Cycle 1]: 8.117e-05, [2] [tag_attr]: 1.921e-05 [meta_addattr_fg_expand]: 4.80001e-06 [parallel-infer-symbol]: 3.61999e-06 [pre_auto_parallel]: 3.789e-05 [insert-virtual-dataset]: 2.57001e-06 [parallel-infer-symbol-second]: 7.10017e-07 [dataset_repeat_opt]: 2.14e-06 [pipeline_split]: 1.84e-06 [optimize]: 0.198044, [53] [py_interpret_to_execute]: 3.994e-05 [rewriter_before_opt_a]: 8.448e-05 [opt_a]: 0.193805, [2] [Cycle 1]: 0.192374, [45] [expand_dump_flag]: 3.8e-06 [switch_simplify]: 2.765e-05 [loop_unroll]: 1.461e-05 [a_1]: 0.189506 [with_stream_mark]: 5.168e-05 [recompute_prepare]: 3.018e-05 [updatestate_depend_eliminate]: 9.90002e-06 [updatestate_assign_eliminate]: 6.47001e-06 [updatestate_loads_eliminate]: 6.48e-06 [parameter_eliminate]: 3.17002e-06 [a_2]: 0.00017907 [accelerated_algorithm]: 1.758e-05 [shard]: 5.00999e-06 [meta_shard_fg_expand]: 8.89e-06 [shard_inline]: 1.289e-05 [merge_send_recv]: 1.598e-05 [auto_parallel]: 1.578e-05 [parallel]: 3.562e-05 [flash_sp]: 1.713e-05 [merge_comm]: 1.074e-05 [allreduce_fusion]: 6.61999e-06 [matmul_add_comm_reduction]: 1.718e-05 [allreduce_slice_to_reducescatter]: 1.18001e-06 [virtual_shard_identity]: 2.27e-05 [virtual_dataset]: 1.675e-05 [get_grad_eliminate_]: 1.327e-05 [virtual_output]: 1.381e-05 [merge_forward]: 9.62001e-06 [cell_reuse_recompute_pass]: 4.75001e-06 [offload_activation]: 1.92e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.584e-05 [merge_recompute_call_nodes]: 1.99e-06 [before_grad]: 2.4e-05 [set_forward_comm_id_for_comm_node_pass]: 9.69e-06 [meta_fg_expand]: 6.74999e-06 [flash_sp_send_recv_attached]: 4.45999e-06 [receive_attached]: 2.81e-06 [after_resolve]: 2.64e-05 [a_after_grad]: 1.877e-05 [renormalize]: 0.00137531 [add_forward_monad_depend]: 1.129e-05 [auto_monad_grad]: 2.88e-06 [auto_monad_eliminator]: 3.581e-05 [cse]: 4.752e-05 [a_3]: 9.13e-05 [Cycle 2]: 0.00141146, [45] [expand_dump_flag]: 3.22997e-06 [switch_simplify]: 1.516e-05 [loop_unroll]: 1.08e-05 [a_1]: 0.00024518 [with_stream_mark]: 2.668e-05 [recompute_prepare]: 1.434e-05 [updatestate_depend_eliminate]: 8.60001e-06 [updatestate_assign_eliminate]: 5.75001e-06 [updatestate_loads_eliminate]: 6.60002e-06 [parameter_eliminate]: 3.55998e-06 [a_2]: 0.00013054 [accelerated_algorithm]: 2.111e-05 [shard]: 4.52e-06 [meta_shard_fg_expand]: 6.585e-05 [shard_inline]: 1.499e-05 [merge_send_recv]: 1.374e-05 [auto_parallel]: 1.612e-05 [parallel]: 1.241e-05 [flash_sp]: 5.29e-06 [merge_comm]: 6.07001e-06 [allreduce_fusion]: 6.33e-06 [matmul_add_comm_reduction]: 1.388e-05 [allreduce_slice_to_reducescatter]: 7.89994e-07 [virtual_shard_identity]: 1.696e-05 [virtual_dataset]: 1.11e-05 [get_grad_eliminate_]: 1.11e-05 [virtual_output]: 9.76998e-06 [merge_forward]: 7.72998e-06 [cell_reuse_recompute_pass]: 3.28e-06 [offload_activation]: 1.887e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.273e-05 [merge_recompute_call_nodes]: 1.84e-06 [before_grad]: 2.083e-05 [set_forward_comm_id_for_comm_node_pass]: 8.3e-06 [meta_fg_expand]: 4.42998e-06 [flash_sp_send_recv_attached]: 2.29001e-06 [receive_attached]: 2.53998e-06 [after_resolve]: 1.874e-05 [a_after_grad]: 1.67e-05 [renormalize]: 7.99773e-08 [add_forward_monad_depend]: 5.94e-06 [auto_monad_grad]: 2.74001e-06 [auto_monad_eliminator]: 2.669e-05 [cse]: 4.081e-05 [a_3]: 6.987e-05 [py_interpret_to_execute_after_opt_a]: 2.584e-05 [slice_cell_reuse_recomputed_activation]: 2.49999e-06 [rewriter_after_opt_a]: 0.00030139 [convert_after_rewriter]: 1.909e-05 [order_py_execute_after_rewriter]: 8.46002e-06 [mutable_eliminate]: 0.00095797 [opt_b]: 0.00046853, [1] [Cycle 1]: 0.00045577, [7] [b_1]: 0.00029174 [b_2]: 1.779e-05 [updatestate_depend_eliminate]: 1.654e-05 [updatestate_assign_eliminate]: 6.83e-06 [updatestate_loads_eliminate]: 5.97001e-06 [renormalize]: 1.25001e-06 [cse]: 4.773e-05 [optimize_parallel_all_gather_comm]: 3.202e-05 [overlap_param_gather]: 4.97e-06 [cconv]: 4.568e-05 [loop_unroll]: 0.00077887 [opt_after_cconv]: 0.00022363, [1] [Cycle 1]: 0.00021238, [7] [c_1]: 8.299e-05 [parameter_eliminate]: 6.51999e-06 [updatestate_depend_eliminate]: 1.504e-05 [updatestate_assign_eliminate]: 5.64e-06 [updatestate_loads_eliminate]: 5.57001e-06 [cse]: 4.236e-05 [renormalize]: 8.59989e-07 [remove_dup_value]: 2.775e-05 [tuple_transform]: 0.00014492, [1] [Cycle 1]: 0.00013742, [4] [d_1]: 8.979e-05 [none_parameter_eliminate]: 2.64001e-06 [renormalize]: 2.19996e-07 [switch_simplify]: 1.291e-05 [partial_unused_args_eliminate]: 2.93e-06 [add_recomputation]: 0.00010255 [cse_after_recomputation]: 4.495e-05, [1] [Cycle 1]: 3.686e-05, [1] [cse]: 2.593e-05 [environ_conv]: 1.121e-05 [swap_dp_allreduce_reducescatter]: 1.058e-05 [bias_add_comm_swap]: 3.43999e-06 [label_micro_interleaved_index]: 8.37e-06 [label_fine_grained_interleaved_index]: 3.11001e-06 [merge_cast_opt]: 1.77999e-06 [slice_recompute_activation]: 2.39001e-06 [micro_interleaved_order_control]: 3.31001e-06 [assign_add_opt]: 1.87001e-06 [ForceFp32Comm]: 1.45001e-06 [remove_cast_before_assign_add]: 9.00007e-07 [full_micro_interleaved_order_control]: 2.99001e-06 [reorder_send_recv_between_fp_bp]: 3.68e-06 [comm_op_add_attrs]: 1.47001e-06 [add_comm_op_reuse_tag]: 1.42e-06 [interleave_split_concat_branches]: 1.55999e-06 [interleave_parallel_branches]: 1.54e-06 [overlap_opt_shard_in_pipeline]: 2.91999e-06 [overlap_opt_shard_grad_in_pipeline]: 2.34999e-06 [control_data_broadcast_order]: 2.735e-05 [grouped_pairwise_exchange_alltoall]: 2.34001e-06 [offloading_packed_experts]: 6.95002e-06 [overlap_recompute_and_grad_model_parallel]: 7.78999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.90001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.77001e-06 [overlap_recompute_comm]: 2.90002e-06 [overlap_grad_ring_attention]: 6.28e-06 [overlap_grad_flash_sp]: 3.607e-05 [begin_end_overlap_inline]: 8.10018e-07 [split_matmul_comm_elemetwise]: 2.59999e-06 [split_layernorm_comm]: 2.24001e-06 [handle_group_info]: 1.24e-06 [symbol_engine_optimizer]: 0.00028326, [1] [Cycle 1]: 0.00027307, [6] [build]: 9.07999e-06 [elim_shapecalc]: 4.883e-05 [elim_not_effective]: 2.902e-05 [opt_reshape]: 2.179e-05 [fold_const_symbol]: 2.984e-05 [renormalize]: 8.50006e-07 [detach_backward]: 3.27002e-06 [pipeline_parallel_scheduler]: 2.39999e-06 [auto_monad_reorder]: 4.956e-05 [get_jit_bprop_graph]: 2.53003e-06 [rewriter_after_jit_bprop_graph]: 9.29e-06 [opt_after_jit_grad]: 0.00107256 [validate]: 7.673e-05 [backend_pass]: 1.20001e-06 [task_emit]: 0.1584 [execute]: 1.433e-05 Sums bootstrap : 0.000477s : 0.13% type_inference : 0.003925s : 1.09% event_method : 0.000016s : 0.00% auto_monad : 0.000159s : 0.04% graph_reusing : 0.000006s : 0.00% inline : 0.000002s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000019s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000005s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000038s : 0.01% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000040s : 0.01% optimize.rewriter_before_opt_a : 0.000084s : 0.02% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000043s : 0.01% optimize.opt_a.loop_unroll : 0.000025s : 0.01% optimize.opt_a.a_1 : 0.189752s : 52.64% optimize.opt_a.with_stream_mark : 0.000078s : 0.02% optimize.opt_a.recompute_prepare : 0.000045s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000019s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000012s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000013s : 0.00% optimize.opt_a.parameter_eliminate : 0.000007s : 0.00% optimize.opt_a.a_2 : 0.000310s : 0.09% optimize.opt_a.accelerated_algorithm : 0.000039s : 0.01% optimize.opt_a.shard : 0.000010s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000075s : 0.02% optimize.opt_a.shard_inline : 0.000028s : 0.01% optimize.opt_a.merge_send_recv : 0.000030s : 0.01% optimize.opt_a.auto_parallel : 0.000032s : 0.01% optimize.opt_a.parallel : 0.000048s : 0.01% optimize.opt_a.flash_sp : 0.000022s : 0.01% optimize.opt_a.merge_comm : 0.000017s : 0.00% optimize.opt_a.allreduce_fusion : 0.000013s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000031s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000040s : 0.01% optimize.opt_a.virtual_dataset : 0.000028s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000024s : 0.01% optimize.opt_a.virtual_output : 0.000024s : 0.01% optimize.opt_a.merge_forward : 0.000017s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000008s : 0.00% optimize.opt_a.offload_activation : 0.000038s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000069s : 0.02% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000045s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000018s : 0.00% optimize.opt_a.meta_fg_expand : 0.000011s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000007s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000045s : 0.01% optimize.opt_a.a_after_grad : 0.000035s : 0.01% optimize.opt_a.renormalize : 0.001375s : 0.38% optimize.opt_a.add_forward_monad_depend : 0.000017s : 0.00% optimize.opt_a.auto_monad_grad : 0.000006s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000062s : 0.02% optimize.opt_a.cse : 0.000088s : 0.02% optimize.opt_a.a_3 : 0.000161s : 0.04% optimize.py_interpret_to_execute_after_opt_a : 0.000026s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000301s : 0.08% optimize.convert_after_rewriter : 0.000019s : 0.01% optimize.order_py_execute_after_rewriter : 0.000008s : 0.00% optimize.mutable_eliminate : 0.000958s : 0.27% optimize.opt_b.b_1 : 0.000292s : 0.08% optimize.opt_b.b_2 : 0.000018s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000017s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000007s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000048s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000032s : 0.01% optimize.overlap_param_gather : 0.000005s : 0.00% optimize.cconv : 0.000046s : 0.01% optimize.loop_unroll : 0.000779s : 0.22% optimize.opt_after_cconv.c_1 : 0.000083s : 0.02% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000015s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.cse : 0.000042s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000028s : 0.01% optimize.tuple_transform.d_1 : 0.000090s : 0.02% optimize.tuple_transform.none_parameter_eliminate : 0.000003s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000013s : 0.00% optimize.partial_unused_args_eliminate : 0.000003s : 0.00% optimize.add_recomputation : 0.000103s : 0.03% optimize.cse_after_recomputation.cse : 0.000026s : 0.01% optimize.environ_conv : 0.000011s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000011s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000008s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000004s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000002s : 0.00% optimize.interleave_parallel_branches : 0.000002s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000003s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000027s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000008s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000006s : 0.00% optimize.overlap_grad_flash_sp : 0.000036s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000009s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000049s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000029s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000022s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000030s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000001s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000050s : 0.01% get_jit_bprop_graph : 0.000003s : 0.00% rewriter_after_jit_bprop_graph : 0.000009s : 0.00% opt_after_jit_grad : 0.001073s : 0.30% validate : 0.000077s : 0.02% backend_pass : 0.000001s : 0.00% task_emit : 0.158400s : 43.94% execute : 0.000014s : 0.00% Time group info: ------[substitution.] 0.000254 55 4.65% : 0.000012s : 2: substitution.depend_value_elim 1.56% : 0.000004s : 5: substitution.elim_not_effective 1.19% : 0.000003s : 5: substitution.fold_const_symbol 3.18% : 0.000008s : 6: substitution.graph_param_transform 56.96% : 0.000145s : 1: substitution.inline 3.04% : 0.000008s : 10: substitution.j_node_and_user_rematch 4.16% : 0.000011s : 10: substitution.remove_not_recompute_node 3.07% : 0.000008s : 2: substitution.replace_old_param 12.61% : 0.000032s : 6: substitution.updatestate_pure_node_eliminater 9.59% : 0.000024s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.003867 2 85.96% : 0.003324s : 1: type_inference.infer 14.04% : 0.000543s : 1: type_inference.specialize ------[replace.] 0.188841 1 100.00% : 0.188841s : 1: replace.inline ------[match.] 0.000143 1 100.00% : 0.000143s : 1: match.inline ------[predicate.] 0.000327 1419 0.79% : 0.000003s : 13: predicate.accumulaten_eliminater 1.92% : 0.000006s : 6: predicate.ad_related_special_op_eliminate 0.61% : 0.000002s : 12: predicate.addn_check_dump 1.06% : 0.000003s : 13: predicate.addn_zero_filter 0.78% : 0.000003s : 13: predicate.adjust_all_reduce_mul_add 2.13% : 0.000007s : 25: predicate.arithmetic_simplify 0.89% : 0.000003s : 13: predicate.cast_eliminate 0.70% : 0.000002s : 12: predicate.check_bprop_eliminate 0.61% : 0.000002s : 12: predicate.compare_switch_simplify 0.17% : 0.000001s : 6: predicate.const_output_eliminate 0.71% : 0.000002s : 12: predicate.depend_value_elim 0.75% : 0.000002s : 13: predicate.dict_get_item_const_eliminator 0.95% : 0.000003s : 13: predicate.dict_get_item_eliminator 1.02% : 0.000003s : 13: predicate.dict_set_item_eliminator 1.39% : 0.000005s : 12: predicate.dumpgradient_eliminate 0.26% : 0.000001s : 6: predicate.elim_not_effective 0.65% : 0.000002s : 6: predicate.elim_shapecalc_of_broadcastargs 1.08% : 0.000004s : 19: predicate.environ_add_const_eliminate 1.11% : 0.000004s : 19: predicate.environ_get_add_eliminate 1.07% : 0.000004s : 19: predicate.environ_get_depend_swap 1.69% : 0.000006s : 31: predicate.environ_get_eliminate 1.04% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.73% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.57% : 0.000005s : 14: predicate.float_depend_g_call 0.67% : 0.000002s : 12: predicate.float_environ_get_switch 1.06% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.23% : 0.000001s : 6: predicate.fold_const_symbol 0.99% : 0.000003s : 12: predicate.get_grad_eliminate 0.19% : 0.000001s : 6: predicate.graph_param_transform 0.63% : 0.000002s : 12: predicate.incorporate_call 0.48% : 0.000002s : 12: predicate.incorporate_call_switch 5.66% : 0.000019s : 63: predicate.inline 1.18% : 0.000004s : 12: predicate.inline_without_move 0.28% : 0.000001s : 12: predicate.j_node_and_user_rematch 1.33% : 0.000004s : 12: predicate.less_batch_normalization 1.65% : 0.000005s : 25: predicate.list_to_tuple_eliminator_ 1.94% : 0.000006s : 38: predicate.load_eliminater 1.45% : 0.000005s : 6: predicate.loop_unroll_after_grad 0.90% : 0.000003s : 17: predicate.loop_unroll_before_grad 1.83% : 0.000006s : 25: predicate.make_slice_get_slice_eliminator 0.66% : 0.000002s : 12: predicate.merge_addn 0.67% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.78% : 0.000003s : 12: predicate.mini_step_allgather_replace 0.73% : 0.000002s : 13: predicate.minmaximum_grad 1.79% : 0.000006s : 6: predicate.mutable_eliminate 0.42% : 0.000001s : 6: predicate.opt_reshape 0.38% : 0.000001s : 6: predicate.parallel_virtual_node 3.48% : 0.000011s : 14: predicate.partial_defer_inline 0.88% : 0.000003s : 19: predicate.partial_eliminate 0.75% : 0.000002s : 13: predicate.print_const_string_wrapper 0.76% : 0.000002s : 12: predicate.reduce_all_const_elim 1.57% : 0.000005s : 13: predicate.reduce_eliminate 1.81% : 0.000006s : 38: predicate.redundant_stop_gradient_eliminater 0.59% : 0.000002s : 12: predicate.remove_not_recompute_node 0.85% : 0.000003s : 25: predicate.replace_applicator 0.53% : 0.000002s : 12: predicate.replace_old_param 0.27% : 0.000001s : 6: predicate.reset_defer_inline 0.94% : 0.000003s : 13: predicate.reshape_eliminate 0.69% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.57% : 0.000002s : 6: predicate.row_tensor_eliminate 0.92% : 0.000003s : 12: predicate.same_eliminate 0.51% : 0.000002s : 12: predicate.set_cell_output_no_recompute 1.22% : 0.000004s : 12: predicate.shard_identity_eliminate 0.86% : 0.000003s : 12: predicate.special_op_eliminate 0.77% : 0.000003s : 12: predicate.specialize_transform 1.32% : 0.000004s : 12: predicate.split_environ_get_set_with_tuple_value 0.83% : 0.000003s : 12: predicate.stack_unstack_eliminate 0.38% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.86% : 0.000003s : 14: predicate.switch_defer_inline 1.63% : 0.000005s : 26: predicate.switch_layer_defer_inline 3.01% : 0.000010s : 49: predicate.switch_simplify 0.83% : 0.000003s : 13: predicate.tile_eliminate 0.83% : 0.000003s : 13: predicate.transpose_eliminate 1.78% : 0.000006s : 25: predicate.tuple_list_convert_item_index_to_positive 1.66% : 0.000005s : 25: predicate.tuple_list_get_item_const_eliminator 1.78% : 0.000006s : 25: predicate.tuple_list_get_item_depend_reorder 3.22% : 0.000011s : 37: predicate.tuple_list_get_item_eliminator 1.96% : 0.000006s : 25: predicate.tuple_list_get_set_item_eliminator 2.82% : 0.000009s : 37: predicate.tuple_list_set_item_eliminator 1.72% : 0.000006s : 25: predicate.tuple_to_list_eliminator_ 1.83% : 0.000006s : 38: predicate.updatestate_pure_node_eliminater 2.92% : 0.000010s : 50: predicate.updatestate_useless_node_eliminater 0.39% : 0.000001s : 6: predicate.value_based_eliminate 0.80% : 0.000003s : 12: predicate.virtual_dataset_eliminate 0.97% : 0.000003s : 12: predicate.virtual_output_eliminate 0.29% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.64% : 0.000002s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000527 4 5.33% : 0.000028s : 1: func_graph_cloner_run.FuncGraphClonerGraph 94.67% : 0.000499s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.811385 196 0.00% : 0.000005s : 1: ForceFp32Comm 3.59% : 0.029134s : 1: add_attr 3.59% : 0.029113s : 1: add_attr_with_inline 0.00% : 0.000005s : 1: add_comm_op_reuse_tag 0.01% : 0.000113s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.02% : 0.000168s : 1: auto_monad 0.01% : 0.000058s : 1: auto_monad_reorder 0.00% : 0.000009s : 1: backend_pass 0.00% : 0.000005s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.06% : 0.000515s : 1: bootstrap 0.01% : 0.000054s : 1: cconv 0.00% : 0.000011s : 1: comm_op_add_attrs 0.00% : 0.000036s : 1: control_data_broadcast_order 0.00% : 0.000027s : 1: convert_after_rewriter 0.01% : 0.000049s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000016s : 1: environ_conv 0.00% : 0.000023s : 1: event_method 0.00% : 0.000024s : 1: execute 0.00% : 0.000009s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000011s : 1: graph_reusing 0.00% : 0.000006s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000006s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000010s : 1: interleave_parallel_branches 0.00% : 0.000006s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000012s : 1: label_micro_interleaved_index 0.10% : 0.000806s : 1: loop_unroll 0.00% : 0.000007s : 1: merge_cast_opt 0.00% : 0.000008s : 1: micro_interleaved_order_control 0.12% : 0.000980s : 1: mutable_eliminate 0.00% : 0.000015s : 1: offloading_packed_experts 0.00% : 0.000033s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000034s : 1: opt.transform.mutable_eliminate 23.48% : 0.190548s : 78: opt.transform.opt_a 0.01% : 0.000081s : 1: opt.transform.opt_after_cconv 0.01% : 0.000064s : 1: opt.transform.opt_after_jit_grad 0.03% : 0.000263s : 28: opt.transform.opt_b 0.01% : 0.000098s : 2: opt.transform.opt_trans_graph 0.01% : 0.000121s : 4: opt.transform.symbol_engine_opt 23.89% : 0.193812s : 1: opt_a 0.03% : 0.000229s : 1: opt_after_cconv 0.14% : 0.001103s : 1: opt_after_jit_grad 0.06% : 0.000475s : 1: opt_b 24.41% : 0.198053s : 1: optimize 0.01% : 0.000041s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000014s : 1: order_py_execute_after_rewriter 0.01% : 0.000042s : 1: overlap_grad_flash_sp 0.00% : 0.000006s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000011s : 1: overlap_grad_ring_attention 0.00% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000009s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000010s : 1: overlap_param_gather 0.00% : 0.000006s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000014s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000032s : 1: partial_unused_args_eliminate 0.00% : 0.000007s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000043s : 1: pre_auto_parallel 0.01% : 0.000052s : 1: py_interpret_to_execute 0.00% : 0.000033s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000016s : 1: remove_cast_before_assign_add 0.00% : 0.000033s : 1: remove_dup_value 0.09% : 0.000716s : 1: renormalize.infer 0.08% : 0.000641s : 1: renormalize.specialize 0.00% : 0.000008s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000014s : 1: rewriter_after_jit_bprop_graph 0.04% : 0.000319s : 1: rewriter_after_opt_a 0.01% : 0.000096s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000006s : 1: split_layernorm_comm 0.00% : 0.000008s : 1: split_matmul_comm_elemetwise 0.00% : 0.000016s : 1: swap_dp_allreduce_reducescatter 0.04% : 0.000288s : 1: symbol_engine_optimizer 19.53% : 0.158428s : 1: task_emit 0.02% : 0.000149s : 1: tuple_transform 0.49% : 0.003946s : 1: type_inference 0.02% : 0.000139s : 1: validate TotalTime = 0.762357, [24] [bootstrap]: 0.00084669 [type_inference]: 0.210521 [event_method]: 0.00012042 [auto_monad]: 0.0004711 [graph_reusing]: 9.95997e-06 [inline]: 2.66999e-06 [add_attr]: 0.00437948, [1] [add_attr_with_inline]: 0.00436521, [1] [Cycle 1]: 0.00010981, [2] [tag_attr]: 5.04e-05 [meta_addattr_fg_expand]: 1.045e-05 [parallel-infer-symbol]: 3.99002e-06 [pre_auto_parallel]: 6.459e-05 [insert-virtual-dataset]: 2.58e-06 [parallel-infer-symbol-second]: 6.49976e-07 [dataset_repeat_opt]: 1.98002e-06 [pipeline_split]: 1.64e-06 [optimize]: 0.369928, [53] [py_interpret_to_execute]: 6.061e-05 [rewriter_before_opt_a]: 0.000162 [opt_a]: 0.366336, [3] [Cycle 1]: 0.357778, [45] [expand_dump_flag]: 1.013e-05 [switch_simplify]: 9.257e-05 [loop_unroll]: 8.404e-05 [a_1]: 0.00189122 [with_stream_mark]: 4.331e-05 [recompute_prepare]: 3.502e-05 [updatestate_depend_eliminate]: 1.274e-05 [updatestate_assign_eliminate]: 1.068e-05 [updatestate_loads_eliminate]: 1.034e-05 [parameter_eliminate]: 3.6e-06 [a_2]: 0.00036354 [accelerated_algorithm]: 4.56e-05 [shard]: 1.97001e-06 [meta_shard_fg_expand]: 6.93e-06 [shard_inline]: 2.207e-05 [merge_send_recv]: 2.408e-05 [auto_parallel]: 1.943e-05 [parallel]: 2.353e-05 [flash_sp]: 1.94e-05 [merge_comm]: 1.271e-05 [allreduce_fusion]: 1.142e-05 [matmul_add_comm_reduction]: 4.353e-05 [allreduce_slice_to_reducescatter]: 1.02e-06 [virtual_shard_identity]: 2.836e-05 [virtual_dataset]: 2.764e-05 [get_grad_eliminate_]: 2.991e-05 [virtual_output]: 2.679e-05 [merge_forward]: 1.36e-05 [cell_reuse_recompute_pass]: 1.71998e-06 [offload_activation]: 2.645e-05 [cell_reuse_handle_not_recompute_node_pass]: 5.399e-05 [merge_recompute_call_nodes]: 1.66e-06 [before_grad]: 4.983e-05 [set_forward_comm_id_for_comm_node_pass]: 1.566e-05 [meta_fg_expand]: 0.192998 [flash_sp_send_recv_attached]: 7.48999e-06 [receive_attached]: 2.54001e-06 [after_resolve]: 0.0001893 [a_after_grad]: 0.00029509 [renormalize]: 0.0123224 [add_forward_monad_depend]: 2.622e-05 [auto_monad_grad]: 2.13e-05 [auto_monad_eliminator]: 0.145926 [cse]: 0.0004358 [a_3]: 0.00188252 [Cycle 2]: 0.00708719, [45] [expand_dump_flag]: 7.05998e-06 [switch_simplify]: 0.00011364 [loop_unroll]: 0.00010385 [a_1]: 0.00317313 [with_stream_mark]: 3.614e-05 [recompute_prepare]: 2.152e-05 [updatestate_depend_eliminate]: 4.027e-05 [updatestate_assign_eliminate]: 7.45e-06 [updatestate_loads_eliminate]: 5.86998e-06 [parameter_eliminate]: 2.56998e-06 [a_2]: 0.00016861 [accelerated_algorithm]: 1.841e-05 [shard]: 2.02999e-06 [meta_shard_fg_expand]: 6.29999e-06 [shard_inline]: 1.173e-05 [merge_send_recv]: 1.3e-05 [auto_parallel]: 1.396e-05 [parallel]: 1.164e-05 [flash_sp]: 4.57e-06 [merge_comm]: 6.61e-06 [allreduce_fusion]: 6.31998e-06 [matmul_add_comm_reduction]: 1.481e-05 [allreduce_slice_to_reducescatter]: 8.2e-07 [virtual_shard_identity]: 1.463e-05 [virtual_dataset]: 1.129e-05 [get_grad_eliminate_]: 1.096e-05 [virtual_output]: 1.083e-05 [merge_forward]: 7.32002e-06 [cell_reuse_recompute_pass]: 1.57001e-06 [offload_activation]: 1.548e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.305e-05 [merge_recompute_call_nodes]: 1.50001e-06 [before_grad]: 2.015e-05 [set_forward_comm_id_for_comm_node_pass]: 7.34002e-06 [meta_fg_expand]: 0.00010277 [flash_sp_send_recv_attached]: 2.05002e-06 [receive_attached]: 2.92002e-06 [after_resolve]: 1.989e-05 [a_after_grad]: 1.842e-05 [renormalize]: 0.00227375 [add_forward_monad_depend]: 1.053e-05 [auto_monad_grad]: 2.76e-06 [auto_monad_eliminator]: 3.217e-05 [cse]: 0.00027111 [a_3]: 0.00010042 [Cycle 3]: 0.00144511, [45] [expand_dump_flag]: 2.91e-06 [switch_simplify]: 1.406e-05 [loop_unroll]: 1.199e-05 [a_1]: 0.00032468 [with_stream_mark]: 2.111e-05 [recompute_prepare]: 1.218e-05 [updatestate_depend_eliminate]: 7.53e-06 [updatestate_assign_eliminate]: 6.39001e-06 [updatestate_loads_eliminate]: 6.54999e-06 [parameter_eliminate]: 1.93002e-06 [a_2]: 0.00018716 [accelerated_algorithm]: 1.781e-05 [shard]: 2.58e-06 [meta_shard_fg_expand]: 3.38e-06 [shard_inline]: 1.264e-05 [merge_send_recv]: 1.315e-05 [auto_parallel]: 1.337e-05 [parallel]: 9.92001e-06 [flash_sp]: 1.66e-06 [merge_comm]: 8.13001e-06 [allreduce_fusion]: 6.24001e-06 [matmul_add_comm_reduction]: 1.446e-05 [allreduce_slice_to_reducescatter]: 6.89994e-07 [virtual_shard_identity]: 1.318e-05 [virtual_dataset]: 1.26e-05 [get_grad_eliminate_]: 1.086e-05 [virtual_output]: 1.037e-05 [merge_forward]: 7.35e-06 [cell_reuse_recompute_pass]: 3.04999e-06 [offload_activation]: 1.531e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.376e-05 [merge_recompute_call_nodes]: 1.49e-06 [before_grad]: 2.02e-05 [set_forward_comm_id_for_comm_node_pass]: 9.07999e-06 [meta_fg_expand]: 5.34e-06 [flash_sp_send_recv_attached]: 2.20002e-06 [receive_attached]: 2.24001e-06 [after_resolve]: 1.97e-05 [a_after_grad]: 1.853e-05 [renormalize]: 7.99773e-08 [add_forward_monad_depend]: 3.2e-06 [auto_monad_grad]: 2.43e-06 [auto_monad_eliminator]: 2.007e-05 [cse]: 0.00022297 [a_3]: 8.336e-05 [py_interpret_to_execute_after_opt_a]: 2.724e-05 [slice_cell_reuse_recomputed_activation]: 2.84999e-06 [rewriter_after_opt_a]: 0.0001081 [convert_after_rewriter]: 1.505e-05 [order_py_execute_after_rewriter]: 8.24998e-06 [mutable_eliminate]: 0.00091693 [opt_b]: 0.00052099, [1] [Cycle 1]: 0.00051001, [7] [b_1]: 0.00034112 [b_2]: 1.635e-05 [updatestate_depend_eliminate]: 1.22e-05 [updatestate_assign_eliminate]: 7.13e-06 [updatestate_loads_eliminate]: 7.75e-06 [renormalize]: 4.19997e-07 [cse]: 6.2e-05 [optimize_parallel_all_gather_comm]: 2.655e-05 [overlap_param_gather]: 1.84998e-06 [cconv]: 3.884e-05 [loop_unroll]: 0.00060536 [opt_after_cconv]: 0.00018797, [1] [Cycle 1]: 0.00018029, [7] [c_1]: 7.752e-05 [parameter_eliminate]: 5.14998e-06 [updatestate_depend_eliminate]: 9.87001e-06 [updatestate_assign_eliminate]: 5.66e-06 [updatestate_loads_eliminate]: 6.19999e-06 [cse]: 3.674e-05 [renormalize]: 1.27e-06 [remove_dup_value]: 2.183e-05 [tuple_transform]: 0.00014299, [1] [Cycle 1]: 0.00013778, [4] [d_1]: 0.00010137 [none_parameter_eliminate]: 2.27001e-06 [renormalize]: 1.69995e-07 [switch_simplify]: 1.483e-05 [partial_unused_args_eliminate]: 2.00002e-06 [add_recomputation]: 0.00011155 [cse_after_recomputation]: 4.743e-05, [1] [Cycle 1]: 4.052e-05, [1] [cse]: 3.41e-05 [environ_conv]: 1.352e-05 [swap_dp_allreduce_reducescatter]: 9.32001e-06 [bias_add_comm_swap]: 2.83e-06 [label_micro_interleaved_index]: 5.58997e-06 [label_fine_grained_interleaved_index]: 3.31999e-06 [merge_cast_opt]: 1.41998e-06 [slice_recompute_activation]: 2.31e-06 [micro_interleaved_order_control]: 2.68e-06 [assign_add_opt]: 1.34e-06 [ForceFp32Comm]: 9.09989e-07 [remove_cast_before_assign_add]: 9.09989e-07 [full_micro_interleaved_order_control]: 2.46e-06 [reorder_send_recv_between_fp_bp]: 2.71999e-06 [comm_op_add_attrs]: 1.30999e-06 [add_comm_op_reuse_tag]: 1.12e-06 [interleave_split_concat_branches]: 1.25999e-06 [interleave_parallel_branches]: 1.07e-06 [overlap_opt_shard_in_pipeline]: 1.24e-06 [overlap_opt_shard_grad_in_pipeline]: 1.79e-06 [control_data_broadcast_order]: 2.459e-05 [grouped_pairwise_exchange_alltoall]: 1.97999e-06 [offloading_packed_experts]: 6.52001e-06 [overlap_recompute_and_grad_model_parallel]: 7.25e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.35999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.41002e-06 [overlap_recompute_comm]: 2.50002e-06 [overlap_grad_ring_attention]: 7.06001e-06 [overlap_grad_flash_sp]: 3.493e-05 [begin_end_overlap_inline]: 5.3001e-07 [split_matmul_comm_elemetwise]: 2.94001e-06 [split_layernorm_comm]: 2.06e-06 [handle_group_info]: 1.37999e-06 [symbol_engine_optimizer]: 0.00013687, [1] [Cycle 1]: 0.00012898, [6] [build]: 1.654e-05 [elim_shapecalc]: 2.19e-05 [elim_not_effective]: 2.35e-05 [opt_reshape]: 1.599e-05 [fold_const_symbol]: 1.874e-05 [renormalize]: 2.80008e-07 [detach_backward]: 2.72001e-06 [pipeline_parallel_scheduler]: 1.59e-06 [auto_monad_reorder]: 4.309e-05 [get_jit_bprop_graph]: 2.43e-06 [rewriter_after_jit_bprop_graph]: 7.06001e-06 [opt_after_jit_grad]: 0.00066746 [validate]: 6.8e-05 [backend_pass]: 1.40001e-06 [task_emit]: 0.174848 [execute]: 1.327e-05 Sums bootstrap : 0.000847s : 0.11% type_inference : 0.210521s : 27.85% event_method : 0.000120s : 0.02% auto_monad : 0.000471s : 0.06% graph_reusing : 0.000010s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000050s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000010s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000065s : 0.01% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000061s : 0.01% optimize.rewriter_before_opt_a : 0.000162s : 0.02% optimize.opt_a.expand_dump_flag : 0.000020s : 0.00% optimize.opt_a.switch_simplify : 0.000220s : 0.03% optimize.opt_a.loop_unroll : 0.000200s : 0.03% optimize.opt_a.a_1 : 0.005389s : 0.71% optimize.opt_a.with_stream_mark : 0.000101s : 0.01% optimize.opt_a.recompute_prepare : 0.000069s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000061s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000025s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000023s : 0.00% optimize.opt_a.parameter_eliminate : 0.000008s : 0.00% optimize.opt_a.a_2 : 0.000719s : 0.10% optimize.opt_a.accelerated_algorithm : 0.000082s : 0.01% optimize.opt_a.shard : 0.000007s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000017s : 0.00% optimize.opt_a.shard_inline : 0.000046s : 0.01% optimize.opt_a.merge_send_recv : 0.000050s : 0.01% optimize.opt_a.auto_parallel : 0.000047s : 0.01% optimize.opt_a.parallel : 0.000045s : 0.01% optimize.opt_a.flash_sp : 0.000026s : 0.00% optimize.opt_a.merge_comm : 0.000027s : 0.00% optimize.opt_a.allreduce_fusion : 0.000024s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000073s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000003s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000056s : 0.01% optimize.opt_a.virtual_dataset : 0.000052s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000052s : 0.01% optimize.opt_a.virtual_output : 0.000048s : 0.01% optimize.opt_a.merge_forward : 0.000028s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000006s : 0.00% optimize.opt_a.offload_activation : 0.000057s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000101s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000090s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000032s : 0.00% optimize.opt_a.meta_fg_expand : 0.193106s : 25.54% optimize.opt_a.flash_sp_send_recv_attached : 0.000012s : 0.00% optimize.opt_a.receive_attached : 0.000008s : 0.00% optimize.opt_a.after_resolve : 0.000229s : 0.03% optimize.opt_a.a_after_grad : 0.000332s : 0.04% optimize.opt_a.renormalize : 0.014596s : 1.93% optimize.opt_a.add_forward_monad_depend : 0.000040s : 0.01% optimize.opt_a.auto_monad_grad : 0.000026s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.145978s : 19.31% optimize.opt_a.cse : 0.000930s : 0.12% optimize.opt_a.a_3 : 0.002066s : 0.27% optimize.py_interpret_to_execute_after_opt_a : 0.000027s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000108s : 0.01% optimize.convert_after_rewriter : 0.000015s : 0.00% optimize.order_py_execute_after_rewriter : 0.000008s : 0.00% optimize.mutable_eliminate : 0.000917s : 0.12% optimize.opt_b.b_1 : 0.000341s : 0.05% optimize.opt_b.b_2 : 0.000016s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000012s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000007s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000008s : 0.00% optimize.opt_b.renormalize : 0.000000s : 0.00% optimize.opt_b.cse : 0.000062s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000027s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000039s : 0.01% optimize.loop_unroll : 0.000605s : 0.08% optimize.opt_after_cconv.c_1 : 0.000078s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000010s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.cse : 0.000037s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000022s : 0.00% optimize.tuple_transform.d_1 : 0.000101s : 0.01% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000015s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000112s : 0.01% optimize.cse_after_recomputation.cse : 0.000034s : 0.00% optimize.environ_conv : 0.000014s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000009s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000006s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000025s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000007s : 0.00% optimize.overlap_grad_flash_sp : 0.000035s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000017s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000022s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000024s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000016s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000019s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000043s : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.00% opt_after_jit_grad : 0.000667s : 0.09% validate : 0.000068s : 0.01% backend_pass : 0.000001s : 0.00% task_emit : 0.174848s : 23.13% execute : 0.000013s : 0.00% Time group info: ------[substitution.] 0.003064 310 0.58% : 0.000018s : 4: substitution.depend_value_elim 0.11% : 0.000003s : 7: substitution.elim_not_effective 0.20% : 0.000006s : 7: substitution.float_depend_g_call 0.18% : 0.000005s : 2: substitution.float_tuple_getitem_switch 0.09% : 0.000003s : 7: substitution.fold_const_symbol 40.02% : 0.001226s : 4: substitution.getattr_setattr_resolve 0.31% : 0.000009s : 9: substitution.graph_param_transform 0.09% : 0.000003s : 2: substitution.incorporate_call 0.06% : 0.000002s : 2: substitution.incorporate_call_switch 38.06% : 0.001166s : 24: substitution.inline 2.08% : 0.000064s : 5: substitution.inline_without_move 0.55% : 0.000017s : 28: substitution.j_node_and_user_rematch 0.77% : 0.000024s : 3: substitution.less_batch_normalization 0.48% : 0.000015s : 9: substitution.minmaximum_grad 0.29% : 0.000009s : 7: substitution.partial_eliminate 0.73% : 0.000022s : 28: substitution.remove_not_recompute_node 6.22% : 0.000191s : 38: substitution.replace_applicator 0.69% : 0.000021s : 19: substitution.replace_old_param 0.12% : 0.000004s : 1: substitution.set_cell_output_no_recompute 1.01% : 0.000031s : 9: substitution.tuple_list_convert_item_index_to_positive 0.41% : 0.000013s : 9: substitution.tuple_list_get_item_const_eliminator 0.66% : 0.000020s : 9: substitution.tuple_list_get_item_depend_reorder 3.04% : 0.000093s : 31: substitution.tuple_list_get_item_eliminator 0.61% : 0.000019s : 9: substitution.tuple_list_get_set_item_eliminator 0.97% : 0.000030s : 16: substitution.updatestate_pure_node_eliminater 1.67% : 0.000051s : 21: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.210399 2 98.95% : 0.208193s : 1: type_inference.infer 1.05% : 0.002205s : 1: type_inference.specialize ------[replace.] 0.000632 49 12.85% : 0.000081s : 3: replace.getattr_setattr_resolve 37.05% : 0.000234s : 24: replace.inline 14.35% : 0.000091s : 1: replace.replace_applicator 33.67% : 0.000213s : 20: replace.tuple_list_get_item_eliminator 2.07% : 0.000013s : 1: replace.updatestate_useless_node_eliminater ------[match.] 0.002467 49 47.21% : 0.001165s : 3: match.getattr_setattr_resolve 46.53% : 0.001148s : 24: match.inline 3.37% : 0.000083s : 1: match.replace_applicator 2.15% : 0.000053s : 20: match.tuple_list_get_item_eliminator 0.74% : 0.000018s : 1: match.updatestate_useless_node_eliminater ------[predicate.] 0.001342 8683 0.90% : 0.000012s : 85: predicate.accumulaten_eliminater 0.28% : 0.000004s : 9: predicate.ad_related_special_op_eliminate 0.40% : 0.000005s : 37: predicate.addn_check_dump 1.10% : 0.000015s : 85: predicate.addn_zero_filter 0.82% : 0.000011s : 85: predicate.adjust_all_reduce_mul_add 1.60% : 0.000021s : 122: predicate.arithmetic_simplify 0.94% : 0.000013s : 85: predicate.cast_eliminate 2.66% : 0.000036s : 260: predicate.check_bprop_eliminate 0.46% : 0.000006s : 37: predicate.compare_switch_simplify 0.06% : 0.000001s : 9: predicate.const_output_eliminate 0.42% : 0.000006s : 37: predicate.depend_value_elim 0.93% : 0.000012s : 85: predicate.dict_get_item_const_eliminator 1.06% : 0.000014s : 85: predicate.dict_get_item_eliminator 0.90% : 0.000012s : 85: predicate.dict_set_item_eliminator 0.31% : 0.000004s : 18: predicate.dumpgradient_eliminate 0.06% : 0.000001s : 9: predicate.elim_not_effective 0.16% : 0.000002s : 9: predicate.elim_shapecalc_of_broadcastargs 1.03% : 0.000014s : 94: predicate.environ_add_const_eliminate 0.99% : 0.000013s : 94: predicate.environ_get_add_eliminate 1.00% : 0.000013s : 94: predicate.environ_get_depend_swap 1.40% : 0.000019s : 131: predicate.environ_get_eliminate 0.94% : 0.000013s : 94: predicate.environ_get_set_eliminate 1.32% : 0.000018s : 130: predicate.exchange_switch_depend_value 1.92% : 0.000026s : 130: predicate.float_depend_g_call 0.42% : 0.000006s : 37: predicate.float_environ_get_switch 0.54% : 0.000007s : 46: predicate.float_tuple_getitem_switch 0.06% : 0.000001s : 9: predicate.fold_const_symbol 0.46% : 0.000006s : 37: predicate.get_grad_eliminate 0.59% : 0.000008s : 20: predicate.getattr_setattr_resolve 0.07% : 0.000001s : 9: predicate.graph_param_transform 0.38% : 0.000005s : 37: predicate.incorporate_call 0.35% : 0.000005s : 37: predicate.incorporate_call_switch 4.49% : 0.000060s : 307: predicate.inline 1.62% : 0.000022s : 111: predicate.inline_without_move 0.21% : 0.000003s : 37: predicate.j_node_and_user_rematch 0.59% : 0.000008s : 37: predicate.less_batch_normalization 1.30% : 0.000018s : 123: predicate.list_to_tuple_eliminator_ 2.08% : 0.000028s : 208: predicate.load_eliminater 0.35% : 0.000005s : 9: predicate.loop_unroll_after_grad 2.07% : 0.000028s : 197: predicate.loop_unroll_before_grad 1.15% : 0.000015s : 103: predicate.make_slice_get_slice_eliminator 0.43% : 0.000006s : 37: predicate.merge_addn 2.53% : 0.000034s : 256: predicate.micro_step_allgather_replace 2.67% : 0.000036s : 256: predicate.mini_step_allgather_replace 0.93% : 0.000012s : 85: predicate.minmaximum_grad 0.28% : 0.000004s : 9: predicate.mutable_eliminate 0.19% : 0.000003s : 9: predicate.opt_reshape 0.15% : 0.000002s : 9: predicate.parallel_virtual_node 2.22% : 0.000030s : 130: predicate.partial_defer_inline 1.33% : 0.000018s : 114: predicate.partial_eliminate 0.91% : 0.000012s : 85: predicate.print_const_string_wrapper 0.46% : 0.000006s : 37: predicate.reduce_all_const_elim 1.18% : 0.000016s : 85: predicate.reduce_eliminate 2.17% : 0.000029s : 208: predicate.redundant_stop_gradient_eliminater 0.25% : 0.000003s : 37: predicate.remove_not_recompute_node 2.43% : 0.000033s : 363: predicate.replace_applicator 2.25% : 0.000030s : 111: predicate.replace_old_param 0.07% : 0.000001s : 9: predicate.reset_defer_inline 0.96% : 0.000013s : 85: predicate.reshape_eliminate 2.96% : 0.000040s : 256: predicate.row_tensor_add_zeros_like 0.17% : 0.000002s : 9: predicate.row_tensor_eliminate 3.00% : 0.000040s : 260: predicate.same_eliminate 0.26% : 0.000003s : 39: predicate.set_cell_output_no_recompute 0.51% : 0.000007s : 37: predicate.shard_identity_eliminate 0.23% : 0.000003s : 18: predicate.special_op_eliminate 0.45% : 0.000006s : 37: predicate.specialize_transform 2.81% : 0.000038s : 256: predicate.split_environ_get_set_with_tuple_value 1.50% : 0.000020s : 111: predicate.stack_unstack_eliminate 0.13% : 0.000002s : 9: predicate.switch_call_monad_eliminater 1.57% : 0.000021s : 130: predicate.switch_defer_inline 4.22% : 0.000057s : 390: predicate.switch_layer_defer_inline 4.20% : 0.000056s : 373: predicate.switch_simplify 0.96% : 0.000013s : 85: predicate.tile_eliminate 0.93% : 0.000013s : 85: predicate.transpose_eliminate 1.24% : 0.000017s : 103: predicate.tuple_list_convert_item_index_to_positive 1.20% : 0.000016s : 103: predicate.tuple_list_get_item_const_eliminator 1.23% : 0.000017s : 103: predicate.tuple_list_get_item_depend_reorder 2.34% : 0.000031s : 160: predicate.tuple_list_get_item_eliminator 1.23% : 0.000017s : 103: predicate.tuple_list_get_set_item_eliminator 1.62% : 0.000022s : 140: predicate.tuple_list_set_item_eliminator 1.33% : 0.000018s : 123: predicate.tuple_to_list_eliminator_ 2.17% : 0.000029s : 208: predicate.updatestate_pure_node_eliminater 2.65% : 0.000036s : 246: predicate.updatestate_useless_node_eliminater 0.14% : 0.000002s : 9: predicate.value_based_eliminate 0.47% : 0.000006s : 37: predicate.virtual_dataset_eliminate 0.46% : 0.000006s : 37: predicate.virtual_output_eliminate 0.10% : 0.000001s : 9: predicate.virtual_view_grad_eliminate 0.14% : 0.000002s : 9: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.005391 53 55.85% : 0.003011s : 23: func_graph_cloner_run.FuncGraphClonerGraph 44.15% : 0.002380s : 30: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.162704 247 0.00% : 0.000005s : 1: ForceFp32Comm 0.38% : 0.004386s : 1: add_attr 0.38% : 0.004370s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000117s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.04% : 0.000488s : 1: auto_monad 0.00% : 0.000048s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.08% : 0.000891s : 1: bootstrap 0.00% : 0.000043s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000028s : 1: control_data_broadcast_order 0.00% : 0.000019s : 1: convert_after_rewriter 0.00% : 0.000051s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000008s : 1: detach_backward 0.00% : 0.000017s : 1: environ_conv 0.01% : 0.000134s : 1: event_method 0.00% : 0.000021s : 1: execute 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000008s : 1: get_jit_bprop_graph 0.00% : 0.000015s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000009s : 1: label_micro_interleaved_index 0.05% : 0.000617s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.08% : 0.000932s : 1: mutable_eliminate 0.00% : 0.000010s : 1: offloading_packed_experts 0.00% : 0.000025s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000025s : 1: opt.transform.mutable_eliminate 0.82% : 0.009534s : 125: opt.transform.opt_a 0.01% : 0.000075s : 1: opt.transform.opt_after_cconv 0.00% : 0.000055s : 1: opt.transform.opt_after_jit_grad 0.03% : 0.000320s : 28: opt.transform.opt_b 0.12% : 0.001406s : 2: opt.transform.opt_resolve 0.01% : 0.000114s : 2: opt.transform.opt_trans_graph 0.01% : 0.000075s : 4: opt.transform.symbol_engine_opt 31.51% : 0.366340s : 1: opt_a 0.02% : 0.000194s : 1: opt_after_cconv 0.06% : 0.000684s : 1: opt_after_jit_grad 0.05% : 0.000525s : 1: opt_b 31.82% : 0.369939s : 1: optimize 0.00% : 0.000031s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000012s : 1: order_py_execute_after_rewriter 0.00% : 0.000040s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000010s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000010s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000004s : 1: pipeline_split 0.01% : 0.000070s : 1: pre_auto_parallel 0.01% : 0.000065s : 1: py_interpret_to_execute 0.00% : 0.000031s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000025s : 1: remove_dup_value 0.69% : 0.008012s : 2: renormalize.infer 0.56% : 0.006550s : 2: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000011s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000115s : 1: rewriter_after_opt_a 0.01% : 0.000168s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000007s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000007s : 1: split_matmul_comm_elemetwise 0.00% : 0.000013s : 1: swap_dp_allreduce_reducescatter 0.01% : 0.000140s : 1: symbol_engine_optimizer 15.04% : 0.174874s : 1: task_emit 0.01% : 0.000146s : 1: tuple_transform 18.11% : 0.210562s : 1: type_inference 0.01% : 0.000120s : 1: validate . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[0-shape2] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[0-shape2],max_mem:4.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:41:53.537.73 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.58786, [24] [bootstrap]: 0.00236165 [type_inference]: 0.186425 [event_method]: 2.125e-05 [auto_monad]: 0.00207682 [graph_reusing]: 1.466e-05 [inline]: 3.20998e-06 [add_attr]: 0.117289, [1] [add_attr_with_inline]: 0.117263, [1] [Cycle 1]: 0.00012207, [2] [tag_attr]: 5.862e-05 [meta_addattr_fg_expand]: 4.27998e-06 [parallel-infer-symbol]: 4.33001e-06 [pre_auto_parallel]: 7.638e-05 [insert-virtual-dataset]: 2.72001e-06 [parallel-infer-symbol-second]: 1.25001e-06 [dataset_repeat_opt]: 2.56e-06 [pipeline_split]: 1.57999e-06 [optimize]: 0.133544, [53] [py_interpret_to_execute]: 0.0002028 [rewriter_before_opt_a]: 0.00014808 [opt_a]: 0.128125, [2] [Cycle 1]: 0.125999, [45] [expand_dump_flag]: 3.33998e-06 [switch_simplify]: 3.333e-05 [loop_unroll]: 4.82e-05 [a_1]: 0.00269356 [with_stream_mark]: 7.18e-05 [recompute_prepare]: 5.192e-05 [updatestate_depend_eliminate]: 9.18002e-06 [updatestate_assign_eliminate]: 1.417e-05 [updatestate_loads_eliminate]: 1.215e-05 [parameter_eliminate]: 2.94001e-06 [a_2]: 0.00031583 [accelerated_algorithm]: 1.477e-05 [shard]: 4.35999e-06 [meta_shard_fg_expand]: 4.90999e-06 [shard_inline]: 1.166e-05 [merge_send_recv]: 1.59e-05 [auto_parallel]: 1.607e-05 [parallel]: 8.363e-05 [flash_sp]: 1.678e-05 [merge_comm]: 8.23001e-06 [allreduce_fusion]: 5.97999e-06 [matmul_add_comm_reduction]: 7.589e-05 [allreduce_slice_to_reducescatter]: 1.00999e-06 [virtual_shard_identity]: 0.00013461 [virtual_dataset]: 5.848e-05 [get_grad_eliminate_]: 3.771e-05 [virtual_output]: 6.316e-05 [merge_forward]: 1.558e-05 [cell_reuse_recompute_pass]: 3.61001e-06 [offload_activation]: 4.132e-05 [cell_reuse_handle_not_recompute_node_pass]: 0.00013806 [merge_recompute_call_nodes]: 2.09e-06 [before_grad]: 0.00014281 [set_forward_comm_id_for_comm_node_pass]: 1.092e-05 [meta_fg_expand]: 6.17001e-06 [flash_sp_send_recv_attached]: 5.78997e-06 [receive_attached]: 2.99999e-06 [after_resolve]: 3.876e-05 [a_after_grad]: 5.163e-05 [renormalize]: 0.119721 [add_forward_monad_depend]: 2.042e-05 [auto_monad_grad]: 2.99001e-06 [auto_monad_eliminator]: 3.766e-05 [cse]: 4.741e-05 [a_3]: 0.00011241 [Cycle 2]: 0.00205613, [45] [expand_dump_flag]: 2.86999e-06 [switch_simplify]: 1.386e-05 [loop_unroll]: 1.03e-05 [a_1]: 0.00023514 [with_stream_mark]: 2.436e-05 [recompute_prepare]: 1.108e-05 [updatestate_depend_eliminate]: 7.46001e-06 [updatestate_assign_eliminate]: 3.506e-05 [updatestate_loads_eliminate]: 6.57002e-06 [parameter_eliminate]: 2.14e-06 [a_2]: 0.00012958 [accelerated_algorithm]: 1.138e-05 [shard]: 4.01001e-06 [meta_shard_fg_expand]: 3.35e-06 [shard_inline]: 9.50001e-06 [merge_send_recv]: 1.221e-05 [auto_parallel]: 1.715e-05 [parallel]: 3.953e-05 [flash_sp]: 5.16002e-06 [merge_comm]: 6.99001e-06 [allreduce_fusion]: 3.801e-05 [matmul_add_comm_reduction]: 1.652e-05 [allreduce_slice_to_reducescatter]: 1.19e-06 [virtual_shard_identity]: 1.266e-05 [virtual_dataset]: 1.04e-05 [get_grad_eliminate_]: 1.351e-05 [virtual_output]: 9.65002e-06 [merge_forward]: 4.181e-05 [cell_reuse_recompute_pass]: 3.56999e-06 [offload_activation]: 4.825e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.494e-05 [merge_recompute_call_nodes]: 1.81e-06 [before_grad]: 7.967e-05 [set_forward_comm_id_for_comm_node_pass]: 6.84999e-06 [meta_fg_expand]: 4.11001e-06 [flash_sp_send_recv_attached]: 1.86e-06 [receive_attached]: 2.80002e-06 [after_resolve]: 1.742e-05 [a_after_grad]: 1.477e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 4.01001e-06 [auto_monad_grad]: 3.10998e-06 [auto_monad_eliminator]: 2.526e-05 [cse]: 6.225e-05 [a_3]: 9.228e-05 [py_interpret_to_execute_after_opt_a]: 2.831e-05 [slice_cell_reuse_recomputed_activation]: 2.39999e-06 [rewriter_after_opt_a]: 0.00038757 [convert_after_rewriter]: 1.847e-05 [order_py_execute_after_rewriter]: 7.65e-06 [mutable_eliminate]: 0.00102497 [opt_b]: 0.00061792, [1] [Cycle 1]: 0.00060281, [7] [b_1]: 0.00035038 [b_2]: 1.761e-05 [updatestate_depend_eliminate]: 4.28e-05 [updatestate_assign_eliminate]: 6.04001e-06 [updatestate_loads_eliminate]: 5.30999e-06 [renormalize]: 7.99977e-07 [cse]: 4.257e-05 [optimize_parallel_all_gather_comm]: 3.105e-05 [overlap_param_gather]: 2.59999e-06 [cconv]: 6.615e-05 [loop_unroll]: 0.00098507 [opt_after_cconv]: 0.00025589, [1] [Cycle 1]: 0.00024727, [7] [c_1]: 6.637e-05 [parameter_eliminate]: 4.90999e-06 [updatestate_depend_eliminate]: 9.98998e-06 [updatestate_assign_eliminate]: 4.95001e-06 [updatestate_loads_eliminate]: 4.98001e-06 [cse]: 6.104e-05 [renormalize]: 8.29983e-07 [remove_dup_value]: 1.793e-05 [tuple_transform]: 0.0001731, [1] [Cycle 1]: 0.00016145, [4] [d_1]: 8.462e-05 [none_parameter_eliminate]: 2.58998e-06 [renormalize]: 1.79978e-07 [switch_simplify]: 1.003e-05 [partial_unused_args_eliminate]: 2.09999e-06 [add_recomputation]: 0.00018489 [cse_after_recomputation]: 3.274e-05, [1] [Cycle 1]: 2.706e-05, [1] [cse]: 1.988e-05 [environ_conv]: 8.18999e-06 [swap_dp_allreduce_reducescatter]: 9.49e-06 [bias_add_comm_swap]: 3.17002e-06 [label_micro_interleaved_index]: 6.28e-06 [label_fine_grained_interleaved_index]: 3.58e-06 [merge_cast_opt]: 3.09001e-06 [slice_recompute_activation]: 2.49999e-06 [micro_interleaved_order_control]: 4.52998e-06 [assign_add_opt]: 1.64998e-06 [ForceFp32Comm]: 1.58002e-06 [remove_cast_before_assign_add]: 1.00999e-06 [full_micro_interleaved_order_control]: 3.08e-06 [reorder_send_recv_between_fp_bp]: 3.09001e-06 [comm_op_add_attrs]: 1.27e-06 [add_comm_op_reuse_tag]: 1.02998e-06 [interleave_split_concat_branches]: 1.39e-06 [interleave_parallel_branches]: 5.66003e-06 [overlap_opt_shard_in_pipeline]: 1.24e-06 [overlap_opt_shard_grad_in_pipeline]: 2.32999e-06 [control_data_broadcast_order]: 2.696e-05 [grouped_pairwise_exchange_alltoall]: 1.85001e-06 [offloading_packed_experts]: 6.43998e-06 [overlap_recompute_and_grad_model_parallel]: 7.34002e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.40999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.54e-06 [overlap_recompute_comm]: 2.62001e-06 [overlap_grad_ring_attention]: 1.059e-05 [overlap_grad_flash_sp]: 6.463e-05 [begin_end_overlap_inline]: 6.39993e-07 [split_matmul_comm_elemetwise]: 2.43e-06 [split_layernorm_comm]: 1.92001e-06 [handle_group_info]: 1.15999e-06 [symbol_engine_optimizer]: 0.00019306, [1] [Cycle 1]: 0.00018449, [6] [build]: 7.87998e-06 [elim_shapecalc]: 7.541e-05 [elim_not_effective]: 2.518e-05 [opt_reshape]: 1.168e-05 [fold_const_symbol]: 1.716e-05 [renormalize]: 4.09986e-07 [detach_backward]: 2.84001e-06 [pipeline_parallel_scheduler]: 1.54e-06 [auto_monad_reorder]: 3.509e-05 [get_jit_bprop_graph]: 2.32999e-06 [rewriter_after_jit_bprop_graph]: 8.43001e-06 [opt_after_jit_grad]: 0.001061 [validate]: 6.758e-05 [backend_pass]: 2.851e-05 [task_emit]: 0.144276 [execute]: 1.345e-05 Sums bootstrap : 0.002362s : 0.51% type_inference : 0.186425s : 40.00% event_method : 0.000021s : 0.00% auto_monad : 0.002077s : 0.45% graph_reusing : 0.000015s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000059s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000076s : 0.02% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000203s : 0.04% optimize.rewriter_before_opt_a : 0.000148s : 0.03% optimize.opt_a.expand_dump_flag : 0.000006s : 0.00% optimize.opt_a.switch_simplify : 0.000047s : 0.01% optimize.opt_a.loop_unroll : 0.000059s : 0.01% optimize.opt_a.a_1 : 0.002929s : 0.63% optimize.opt_a.with_stream_mark : 0.000096s : 0.02% optimize.opt_a.recompute_prepare : 0.000063s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000017s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000049s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000019s : 0.00% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000445s : 0.10% optimize.opt_a.accelerated_algorithm : 0.000026s : 0.01% optimize.opt_a.shard : 0.000008s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000008s : 0.00% optimize.opt_a.shard_inline : 0.000021s : 0.00% optimize.opt_a.merge_send_recv : 0.000028s : 0.01% optimize.opt_a.auto_parallel : 0.000033s : 0.01% optimize.opt_a.parallel : 0.000123s : 0.03% optimize.opt_a.flash_sp : 0.000022s : 0.00% optimize.opt_a.merge_comm : 0.000015s : 0.00% optimize.opt_a.allreduce_fusion : 0.000044s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000092s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000147s : 0.03% optimize.opt_a.virtual_dataset : 0.000069s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000051s : 0.01% optimize.opt_a.virtual_output : 0.000073s : 0.02% optimize.opt_a.merge_forward : 0.000057s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000007s : 0.00% optimize.opt_a.offload_activation : 0.000090s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000163s : 0.03% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000222s : 0.05% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000018s : 0.00% optimize.opt_a.meta_fg_expand : 0.000010s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000008s : 0.00% optimize.opt_a.receive_attached : 0.000006s : 0.00% optimize.opt_a.after_resolve : 0.000056s : 0.01% optimize.opt_a.a_after_grad : 0.000066s : 0.01% optimize.opt_a.renormalize : 0.119722s : 25.69% optimize.opt_a.add_forward_monad_depend : 0.000024s : 0.01% optimize.opt_a.auto_monad_grad : 0.000006s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000063s : 0.01% optimize.opt_a.cse : 0.000110s : 0.02% optimize.opt_a.a_3 : 0.000205s : 0.04% optimize.py_interpret_to_execute_after_opt_a : 0.000028s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000388s : 0.08% optimize.convert_after_rewriter : 0.000018s : 0.00% optimize.order_py_execute_after_rewriter : 0.000008s : 0.00% optimize.mutable_eliminate : 0.001025s : 0.22% optimize.opt_b.b_1 : 0.000350s : 0.08% optimize.opt_b.b_2 : 0.000018s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000043s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000043s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000031s : 0.01% optimize.overlap_param_gather : 0.000003s : 0.00% optimize.cconv : 0.000066s : 0.01% optimize.loop_unroll : 0.000985s : 0.21% optimize.opt_after_cconv.c_1 : 0.000066s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000010s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.cse : 0.000061s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000018s : 0.00% optimize.tuple_transform.d_1 : 0.000085s : 0.02% optimize.tuple_transform.none_parameter_eliminate : 0.000003s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000010s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000185s : 0.04% optimize.cse_after_recomputation.cse : 0.000020s : 0.00% optimize.environ_conv : 0.000008s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000009s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000006s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000004s : 0.00% optimize.merge_cast_opt : 0.000003s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000005s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000002s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000006s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000027s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000006s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000011s : 0.00% optimize.overlap_grad_flash_sp : 0.000065s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000008s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000075s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000025s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000012s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000017s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000035s : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.001061s : 0.23% validate : 0.000068s : 0.01% backend_pass : 0.000029s : 0.01% task_emit : 0.144276s : 30.96% execute : 0.000013s : 0.00% Time group info: ------[substitution.] 0.000539 55 2.06% : 0.000011s : 2: substitution.depend_value_elim 0.70% : 0.000004s : 5: substitution.elim_not_effective 0.51% : 0.000003s : 5: substitution.fold_const_symbol 1.66% : 0.000009s : 6: substitution.graph_param_transform 80.60% : 0.000434s : 1: substitution.inline 1.42% : 0.000008s : 10: substitution.j_node_and_user_rematch 2.10% : 0.000011s : 10: substitution.remove_not_recompute_node 1.49% : 0.000008s : 2: substitution.replace_old_param 5.52% : 0.000030s : 6: substitution.updatestate_pure_node_eliminater 3.93% : 0.000021s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.186282 2 97.55% : 0.181714s : 1: type_inference.infer 2.45% : 0.004568s : 1: type_inference.specialize ------[replace.] 0.000055 1 100.00% : 0.000055s : 1: replace.inline ------[match.] 0.000430 1 100.00% : 0.000430s : 1: match.inline ------[predicate.] 0.000295 1419 0.96% : 0.000003s : 13: predicate.accumulaten_eliminater 1.20% : 0.000004s : 6: predicate.ad_related_special_op_eliminate 0.62% : 0.000002s : 12: predicate.addn_check_dump 0.93% : 0.000003s : 13: predicate.addn_zero_filter 0.68% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 2.25% : 0.000007s : 25: predicate.arithmetic_simplify 0.75% : 0.000002s : 13: predicate.cast_eliminate 0.62% : 0.000002s : 12: predicate.check_bprop_eliminate 1.03% : 0.000003s : 12: predicate.compare_switch_simplify 0.18% : 0.000001s : 6: predicate.const_output_eliminate 0.72% : 0.000002s : 12: predicate.depend_value_elim 0.98% : 0.000003s : 13: predicate.dict_get_item_const_eliminator 0.85% : 0.000002s : 13: predicate.dict_get_item_eliminator 0.94% : 0.000003s : 13: predicate.dict_set_item_eliminator 1.29% : 0.000004s : 12: predicate.dumpgradient_eliminate 0.33% : 0.000001s : 6: predicate.elim_not_effective 0.62% : 0.000002s : 6: predicate.elim_shapecalc_of_broadcastargs 1.14% : 0.000003s : 19: predicate.environ_add_const_eliminate 0.91% : 0.000003s : 19: predicate.environ_get_add_eliminate 1.67% : 0.000005s : 19: predicate.environ_get_depend_swap 1.71% : 0.000005s : 31: predicate.environ_get_eliminate 1.00% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.74% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.59% : 0.000005s : 14: predicate.float_depend_g_call 0.62% : 0.000002s : 12: predicate.float_environ_get_switch 0.95% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.18% : 0.000001s : 6: predicate.fold_const_symbol 0.76% : 0.000002s : 12: predicate.get_grad_eliminate 0.26% : 0.000001s : 6: predicate.graph_param_transform 0.62% : 0.000002s : 12: predicate.incorporate_call 0.53% : 0.000002s : 12: predicate.incorporate_call_switch 6.02% : 0.000018s : 63: predicate.inline 1.25% : 0.000004s : 12: predicate.inline_without_move 0.33% : 0.000001s : 12: predicate.j_node_and_user_rematch 1.09% : 0.000003s : 12: predicate.less_batch_normalization 1.99% : 0.000006s : 25: predicate.list_to_tuple_eliminator_ 2.20% : 0.000006s : 38: predicate.load_eliminater 1.41% : 0.000004s : 6: predicate.loop_unroll_after_grad 1.21% : 0.000004s : 17: predicate.loop_unroll_before_grad 2.11% : 0.000006s : 25: predicate.make_slice_get_slice_eliminator 0.64% : 0.000002s : 12: predicate.merge_addn 0.61% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.73% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.68% : 0.000002s : 13: predicate.minmaximum_grad 1.43% : 0.000004s : 6: predicate.mutable_eliminate 0.38% : 0.000001s : 6: predicate.opt_reshape 0.43% : 0.000001s : 6: predicate.parallel_virtual_node 1.88% : 0.000006s : 14: predicate.partial_defer_inline 1.05% : 0.000003s : 19: predicate.partial_eliminate 0.79% : 0.000002s : 13: predicate.print_const_string_wrapper 0.74% : 0.000002s : 12: predicate.reduce_all_const_elim 1.08% : 0.000003s : 13: predicate.reduce_eliminate 2.19% : 0.000006s : 38: predicate.redundant_stop_gradient_eliminater 0.55% : 0.000002s : 12: predicate.remove_not_recompute_node 0.97% : 0.000003s : 25: predicate.replace_applicator 0.54% : 0.000002s : 12: predicate.replace_old_param 0.40% : 0.000001s : 6: predicate.reset_defer_inline 0.96% : 0.000003s : 13: predicate.reshape_eliminate 0.69% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.62% : 0.000002s : 6: predicate.row_tensor_eliminate 1.14% : 0.000003s : 12: predicate.same_eliminate 0.45% : 0.000001s : 12: predicate.set_cell_output_no_recompute 1.68% : 0.000005s : 12: predicate.shard_identity_eliminate 0.83% : 0.000002s : 12: predicate.special_op_eliminate 0.78% : 0.000002s : 12: predicate.specialize_transform 1.07% : 0.000003s : 12: predicate.split_environ_get_set_with_tuple_value 1.20% : 0.000004s : 12: predicate.stack_unstack_eliminate 0.42% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.92% : 0.000003s : 14: predicate.switch_defer_inline 1.39% : 0.000004s : 26: predicate.switch_layer_defer_inline 3.39% : 0.000010s : 49: predicate.switch_simplify 0.82% : 0.000002s : 13: predicate.tile_eliminate 0.92% : 0.000003s : 13: predicate.transpose_eliminate 1.57% : 0.000005s : 25: predicate.tuple_list_convert_item_index_to_positive 1.54% : 0.000005s : 25: predicate.tuple_list_get_item_const_eliminator 1.55% : 0.000005s : 25: predicate.tuple_list_get_item_depend_reorder 3.58% : 0.000011s : 37: predicate.tuple_list_get_item_eliminator 1.55% : 0.000005s : 25: predicate.tuple_list_get_set_item_eliminator 2.37% : 0.000007s : 37: predicate.tuple_list_set_item_eliminator 1.68% : 0.000005s : 25: predicate.tuple_to_list_eliminator_ 2.22% : 0.000007s : 38: predicate.updatestate_pure_node_eliminater 2.88% : 0.000009s : 50: predicate.updatestate_useless_node_eliminater 0.42% : 0.000001s : 6: predicate.value_based_eliminate 1.02% : 0.000003s : 12: predicate.virtual_dataset_eliminate 1.17% : 0.000003s : 12: predicate.virtual_output_eliminate 0.35% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.49% : 0.000001s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.003040 4 0.86% : 0.000026s : 1: func_graph_cloner_run.FuncGraphClonerGraph 99.14% : 0.003014s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.963427 196 0.00% : 0.000008s : 1: ForceFp32Comm 12.18% : 0.117298s : 1: add_attr 12.17% : 0.117271s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.02% : 0.000198s : 1: add_recomputation 0.01% : 0.000143s : 1: assign_add_opt 0.22% : 0.002155s : 1: auto_monad 0.00% : 0.000040s : 1: auto_monad_reorder 0.00% : 0.000039s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.26% : 0.002459s : 1: bootstrap 0.01% : 0.000071s : 1: cconv 0.00% : 0.000006s : 1: comm_op_add_attrs 0.00% : 0.000035s : 1: control_data_broadcast_order 0.00% : 0.000028s : 1: convert_after_rewriter 0.00% : 0.000036s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000016s : 1: environ_conv 0.01% : 0.000062s : 1: event_method 0.00% : 0.000026s : 1: execute 0.00% : 0.000007s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000024s : 1: graph_reusing 0.00% : 0.000009s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000009s : 1: interleave_parallel_branches 0.00% : 0.000008s : 1: interleave_split_concat_branches 0.01% : 0.000137s : 1: label_fine_grained_interleaved_index 0.00% : 0.000013s : 1: label_micro_interleaved_index 0.10% : 0.001003s : 1: loop_unroll 0.01% : 0.000050s : 1: merge_cast_opt 0.00% : 0.000013s : 1: micro_interleaved_order_control 0.11% : 0.001042s : 1: mutable_eliminate 0.00% : 0.000016s : 1: offloading_packed_experts 0.00% : 0.000025s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000031s : 1: opt.transform.mutable_eliminate 0.46% : 0.004476s : 78: opt.transform.opt_a 0.01% : 0.000065s : 1: opt.transform.opt_after_cconv 0.00% : 0.000047s : 1: opt.transform.opt_after_jit_grad 0.03% : 0.000313s : 28: opt.transform.opt_b 0.01% : 0.000092s : 2: opt.transform.opt_trans_graph 0.01% : 0.000124s : 4: opt.transform.symbol_engine_opt 13.30% : 0.128131s : 1: opt_a 0.03% : 0.000263s : 1: opt_after_cconv 0.11% : 0.001078s : 1: opt_after_jit_grad 0.06% : 0.000623s : 1: opt_b 13.86% : 0.133551s : 1: optimize 0.00% : 0.000035s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000011s : 1: order_py_execute_after_rewriter 0.01% : 0.000070s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000043s : 1: overlap_grad_ring_attention 0.00% : 0.000013s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000041s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000041s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000011s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000010s : 1: overlap_recompute_comm 0.00% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000004s : 1: pipeline_split 0.01% : 0.000082s : 1: pre_auto_parallel 0.02% : 0.000239s : 1: py_interpret_to_execute 0.00% : 0.000038s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.00% : 0.000023s : 1: remove_dup_value 12.29% : 0.118371s : 1: renormalize.infer 0.14% : 0.001330s : 1: renormalize.specialize 0.00% : 0.000014s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.04% : 0.000399s : 1: rewriter_after_opt_a 0.02% : 0.000156s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000013s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000007s : 1: split_matmul_comm_elemetwise 0.00% : 0.000013s : 1: swap_dp_allreduce_reducescatter 0.02% : 0.000197s : 1: symbol_engine_optimizer 14.98% : 0.144297s : 1: task_emit 0.02% : 0.000185s : 1: tuple_transform 19.36% : 0.186508s : 1: type_inference 0.01% : 0.000123s : 1: validate TotalTime = 0.130691, [24] [bootstrap]: 0.00048516 [type_inference]: 0.00404749 [event_method]: 5.476e-05 [auto_monad]: 0.00015128 [graph_reusing]: 5.56002e-06 [inline]: 2.76e-06 [add_attr]: 0.0853233, [1] [add_attr_with_inline]: 0.0853086, [1] [Cycle 1]: 7.883e-05, [2] [tag_attr]: 1.84e-05 [meta_addattr_fg_expand]: 4.18999e-06 [parallel-infer-symbol]: 3.81999e-06 [pre_auto_parallel]: 3.544e-05 [insert-virtual-dataset]: 2.71e-06 [parallel-infer-symbol-second]: 6.10016e-07 [dataset_repeat_opt]: 2.05002e-06 [pipeline_split]: 1.80001e-06 [optimize]: 0.00840221, [53] [py_interpret_to_execute]: 2.685e-05 [rewriter_before_opt_a]: 6.369e-05 [opt_a]: 0.00402609, [2] [Cycle 1]: 0.00284711, [45] [expand_dump_flag]: 3.08998e-06 [switch_simplify]: 2.839e-05 [loop_unroll]: 1.593e-05 [a_1]: 0.00055045 [with_stream_mark]: 2.245e-05 [recompute_prepare]: 1.566e-05 [updatestate_depend_eliminate]: 6.91999e-06 [updatestate_assign_eliminate]: 6.26998e-06 [updatestate_loads_eliminate]: 5.74999e-06 [parameter_eliminate]: 1.91998e-06 [a_2]: 0.0002049 [accelerated_algorithm]: 1.285e-05 [shard]: 3.14999e-06 [meta_shard_fg_expand]: 3.78001e-06 [shard_inline]: 1.029e-05 [merge_send_recv]: 1.199e-05 [auto_parallel]: 1.179e-05 [parallel]: 2.737e-05 [flash_sp]: 1.198e-05 [merge_comm]: 6.95002e-06 [allreduce_fusion]: 6.07999e-06 [matmul_add_comm_reduction]: 1.391e-05 [allreduce_slice_to_reducescatter]: 9.5999e-07 [virtual_shard_identity]: 1.231e-05 [virtual_dataset]: 1.278e-05 [get_grad_eliminate_]: 1.044e-05 [virtual_output]: 1.115e-05 [merge_forward]: 5.84e-06 [cell_reuse_recompute_pass]: 1.81e-06 [offload_activation]: 1.727e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.366e-05 [merge_recompute_call_nodes]: 1.44e-06 [before_grad]: 1.978e-05 [set_forward_comm_id_for_comm_node_pass]: 9.98998e-06 [meta_fg_expand]: 4.87e-06 [flash_sp_send_recv_attached]: 3.38e-06 [receive_attached]: 3.21001e-06 [after_resolve]: 1.644e-05 [a_after_grad]: 1.652e-05 [renormalize]: 0.00117771 [add_forward_monad_depend]: 8.28999e-06 [auto_monad_grad]: 2.92002e-06 [auto_monad_eliminator]: 2.708e-05 [cse]: 4.399e-05 [a_3]: 7.678e-05 [Cycle 2]: 0.00116202, [45] [expand_dump_flag]: 2.56e-06 [switch_simplify]: 1.167e-05 [loop_unroll]: 9.37999e-06 [a_1]: 0.00027558 [with_stream_mark]: 2.45e-05 [recompute_prepare]: 1.112e-05 [updatestate_depend_eliminate]: 6.26998e-06 [updatestate_assign_eliminate]: 5.88002e-06 [updatestate_loads_eliminate]: 5.72999e-06 [parameter_eliminate]: 1.56002e-06 [a_2]: 0.00012305 [accelerated_algorithm]: 1.151e-05 [shard]: 2.12999e-06 [meta_shard_fg_expand]: 3.88999e-06 [shard_inline]: 9.47001e-06 [merge_send_recv]: 1.159e-05 [auto_parallel]: 1.18e-05 [parallel]: 1.037e-05 [flash_sp]: 4.21001e-06 [merge_comm]: 5.89e-06 [allreduce_fusion]: 5.70001e-06 [matmul_add_comm_reduction]: 1.281e-05 [allreduce_slice_to_reducescatter]: 8.39995e-07 [virtual_shard_identity]: 1.229e-05 [virtual_dataset]: 1.018e-05 [get_grad_eliminate_]: 9.00999e-06 [virtual_output]: 9.34e-06 [merge_forward]: 9.32001e-06 [cell_reuse_recompute_pass]: 2.91999e-06 [offload_activation]: 1.591e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.194e-05 [merge_recompute_call_nodes]: 1.32e-06 [before_grad]: 1.88e-05 [set_forward_comm_id_for_comm_node_pass]: 6.59999e-06 [meta_fg_expand]: 4.2e-06 [flash_sp_send_recv_attached]: 1.59e-06 [receive_attached]: 2.49999e-06 [after_resolve]: 1.712e-05 [a_after_grad]: 1.523e-05 [renormalize]: 7.00238e-08 [add_forward_monad_depend]: 3.52002e-06 [auto_monad_grad]: 1.92999e-06 [auto_monad_eliminator]: 1.83e-05 [cse]: 3.411e-05 [a_3]: 5.817e-05 [py_interpret_to_execute_after_opt_a]: 2.295e-05 [slice_cell_reuse_recomputed_activation]: 2.41998e-06 [rewriter_after_opt_a]: 0.0002667 [convert_after_rewriter]: 1.801e-05 [order_py_execute_after_rewriter]: 7.48999e-06 [mutable_eliminate]: 0.00100709 [opt_b]: 0.00046637, [1] [Cycle 1]: 0.00045399, [7] [b_1]: 0.00025915 [b_2]: 6.028e-05 [updatestate_depend_eliminate]: 1.504e-05 [updatestate_assign_eliminate]: 6.19999e-06 [updatestate_loads_eliminate]: 5.32001e-06 [renormalize]: 9.10019e-07 [cse]: 4.6e-05 [optimize_parallel_all_gather_comm]: 3.208e-05 [overlap_param_gather]: 7.73999e-06 [cconv]: 3.894e-05 [loop_unroll]: 0.00112536 [opt_after_cconv]: 0.00022482, [1] [Cycle 1]: 0.00021084, [7] [c_1]: 8.847e-05 [parameter_eliminate]: 5.99999e-06 [updatestate_depend_eliminate]: 1.309e-05 [updatestate_assign_eliminate]: 6.40002e-06 [updatestate_loads_eliminate]: 4.85001e-06 [cse]: 4.313e-05 [renormalize]: 8.40024e-07 [remove_dup_value]: 1.975e-05 [tuple_transform]: 0.00015919, [1] [Cycle 1]: 0.0001529, [4] [d_1]: 8.132e-05 [none_parameter_eliminate]: 2.40002e-06 [renormalize]: 3.00002e-07 [switch_simplify]: 4.076e-05 [partial_unused_args_eliminate]: 2.81e-06 [add_recomputation]: 9.615e-05 [cse_after_recomputation]: 3.657e-05, [1] [Cycle 1]: 3.033e-05, [1] [cse]: 2.232e-05 [environ_conv]: 9.52001e-06 [swap_dp_allreduce_reducescatter]: 9.51998e-06 [bias_add_comm_swap]: 3.04999e-06 [label_micro_interleaved_index]: 8.49998e-06 [label_fine_grained_interleaved_index]: 3.11001e-06 [merge_cast_opt]: 1.53002e-06 [slice_recompute_activation]: 2.55002e-06 [micro_interleaved_order_control]: 2.79001e-06 [assign_add_opt]: 1.46998e-06 [ForceFp32Comm]: 1.27e-06 [remove_cast_before_assign_add]: 9.29984e-07 [full_micro_interleaved_order_control]: 2.62001e-06 [reorder_send_recv_between_fp_bp]: 2.64001e-06 [comm_op_add_attrs]: 1.52001e-06 [add_comm_op_reuse_tag]: 1.20001e-06 [interleave_split_concat_branches]: 1.36998e-06 [interleave_parallel_branches]: 1.32e-06 [overlap_opt_shard_in_pipeline]: 2.96001e-06 [overlap_opt_shard_grad_in_pipeline]: 1.86e-06 [control_data_broadcast_order]: 2.433e-05 [grouped_pairwise_exchange_alltoall]: 1.69998e-06 [offloading_packed_experts]: 7.38e-06 [overlap_recompute_and_grad_model_parallel]: 7.25998e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.49e-06 [overlap_recompute_allgather_and_fa_grad]: 1.50999e-06 [overlap_recompute_comm]: 2.84999e-06 [overlap_grad_ring_attention]: 6.68998e-06 [overlap_grad_flash_sp]: 3.142e-05 [begin_end_overlap_inline]: 5.19998e-07 [split_matmul_comm_elemetwise]: 2.37999e-06 [split_layernorm_comm]: 1.71002e-06 [handle_group_info]: 1.16997e-06 [symbol_engine_optimizer]: 0.00013556, [1] [Cycle 1]: 0.00012885, [6] [build]: 6.48e-06 [elim_shapecalc]: 2.003e-05 [elim_not_effective]: 2.314e-05 [opt_reshape]: 1.285e-05 [fold_const_symbol]: 1.712e-05 [renormalize]: 3.29979e-07 [detach_backward]: 2.70997e-06 [pipeline_parallel_scheduler]: 1.73997e-06 [auto_monad_reorder]: 3.667e-05 [get_jit_bprop_graph]: 2.20002e-06 [rewriter_after_jit_bprop_graph]: 6.83e-06 [opt_after_jit_grad]: 0.00103032 [validate]: 6.727e-05 [backend_pass]: 1.17e-06 [task_emit]: 0.0307127 [execute]: 1.163e-05 Sums bootstrap : 0.000485s : 1.11% type_inference : 0.004047s : 9.26% event_method : 0.000055s : 0.13% auto_monad : 0.000151s : 0.35% graph_reusing : 0.000006s : 0.01% inline : 0.000003s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000018s : 0.04% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.01% parallel-infer-symbol : 0.000004s : 0.01% pre_auto_parallel : 0.000035s : 0.08% insert-virtual-dataset : 0.000003s : 0.01% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000027s : 0.06% optimize.rewriter_before_opt_a : 0.000064s : 0.15% optimize.opt_a.expand_dump_flag : 0.000006s : 0.01% optimize.opt_a.switch_simplify : 0.000040s : 0.09% optimize.opt_a.loop_unroll : 0.000025s : 0.06% optimize.opt_a.a_1 : 0.000826s : 1.89% optimize.opt_a.with_stream_mark : 0.000047s : 0.11% optimize.opt_a.recompute_prepare : 0.000027s : 0.06% optimize.opt_a.updatestate_depend_eliminate : 0.000013s : 0.03% optimize.opt_a.updatestate_assign_eliminate : 0.000012s : 0.03% optimize.opt_a.updatestate_loads_eliminate : 0.000011s : 0.03% optimize.opt_a.parameter_eliminate : 0.000003s : 0.01% optimize.opt_a.a_2 : 0.000328s : 0.75% optimize.opt_a.accelerated_algorithm : 0.000024s : 0.06% optimize.opt_a.shard : 0.000005s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000008s : 0.02% optimize.opt_a.shard_inline : 0.000020s : 0.05% optimize.opt_a.merge_send_recv : 0.000024s : 0.05% optimize.opt_a.auto_parallel : 0.000024s : 0.05% optimize.opt_a.parallel : 0.000038s : 0.09% optimize.opt_a.flash_sp : 0.000016s : 0.04% optimize.opt_a.merge_comm : 0.000013s : 0.03% optimize.opt_a.allreduce_fusion : 0.000012s : 0.03% optimize.opt_a.matmul_add_comm_reduction : 0.000027s : 0.06% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000025s : 0.06% optimize.opt_a.virtual_dataset : 0.000023s : 0.05% optimize.opt_a.get_grad_eliminate_ : 0.000019s : 0.04% optimize.opt_a.virtual_output : 0.000020s : 0.05% optimize.opt_a.merge_forward : 0.000015s : 0.03% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.01% optimize.opt_a.offload_activation : 0.000033s : 0.08% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000046s : 0.10% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.01% optimize.opt_a.before_grad : 0.000039s : 0.09% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000017s : 0.04% optimize.opt_a.meta_fg_expand : 0.000009s : 0.02% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.01% optimize.opt_a.receive_attached : 0.000006s : 0.01% optimize.opt_a.after_resolve : 0.000034s : 0.08% optimize.opt_a.a_after_grad : 0.000032s : 0.07% optimize.opt_a.renormalize : 0.001178s : 2.70% optimize.opt_a.add_forward_monad_depend : 0.000012s : 0.03% optimize.opt_a.auto_monad_grad : 0.000005s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000045s : 0.10% optimize.opt_a.cse : 0.000078s : 0.18% optimize.opt_a.a_3 : 0.000135s : 0.31% optimize.py_interpret_to_execute_after_opt_a : 0.000023s : 0.05% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.01% optimize.rewriter_after_opt_a : 0.000267s : 0.61% optimize.convert_after_rewriter : 0.000018s : 0.04% optimize.order_py_execute_after_rewriter : 0.000007s : 0.02% optimize.mutable_eliminate : 0.001007s : 2.30% optimize.opt_b.b_1 : 0.000259s : 0.59% optimize.opt_b.b_2 : 0.000060s : 0.14% optimize.opt_b.updatestate_depend_eliminate : 0.000015s : 0.03% optimize.opt_b.updatestate_assign_eliminate : 0.000006s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000005s : 0.01% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000046s : 0.11% optimize.optimize_parallel_all_gather_comm : 0.000032s : 0.07% optimize.overlap_param_gather : 0.000008s : 0.02% optimize.cconv : 0.000039s : 0.09% optimize.loop_unroll : 0.001125s : 2.58% optimize.opt_after_cconv.c_1 : 0.000088s : 0.20% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.01% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000013s : 0.03% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000006s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.01% optimize.opt_after_cconv.cse : 0.000043s : 0.10% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000020s : 0.05% optimize.tuple_transform.d_1 : 0.000081s : 0.19% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.01% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000041s : 0.09% optimize.partial_unused_args_eliminate : 0.000003s : 0.01% optimize.add_recomputation : 0.000096s : 0.22% optimize.cse_after_recomputation.cse : 0.000022s : 0.05% optimize.environ_conv : 0.000010s : 0.02% optimize.swap_dp_allreduce_reducescatter : 0.000010s : 0.02% optimize.bias_add_comm_swap : 0.000003s : 0.01% optimize.label_micro_interleaved_index : 0.000008s : 0.02% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.01% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.01% optimize.micro_interleaved_order_control : 0.000003s : 0.01% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.01% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.01% optimize.comm_op_add_attrs : 0.000002s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000003s : 0.01% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000024s : 0.06% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.02% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 0.02% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.01% optimize.overlap_grad_ring_attention : 0.000007s : 0.02% optimize.overlap_grad_flash_sp : 0.000031s : 0.07% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.01% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000006s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000020s : 0.05% optimize.symbol_engine_optimizer.elim_not_effective : 0.000023s : 0.05% optimize.symbol_engine_optimizer.opt_reshape : 0.000013s : 0.03% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000017s : 0.04% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.01% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000037s : 0.08% get_jit_bprop_graph : 0.000002s : 0.01% rewriter_after_jit_bprop_graph : 0.000007s : 0.02% opt_after_jit_grad : 0.001030s : 2.36% validate : 0.000067s : 0.15% backend_pass : 0.000001s : 0.00% task_emit : 0.030713s : 70.29% execute : 0.000012s : 0.03% Time group info: ------[substitution.] 0.000210 55 5.25% : 0.000011s : 2: substitution.depend_value_elim 1.62% : 0.000003s : 5: substitution.elim_not_effective 1.09% : 0.000002s : 5: substitution.fold_const_symbol 4.17% : 0.000009s : 6: substitution.graph_param_transform 57.57% : 0.000121s : 1: substitution.inline 3.06% : 0.000006s : 10: substitution.j_node_and_user_rematch 4.40% : 0.000009s : 10: substitution.remove_not_recompute_node 2.95% : 0.000006s : 2: substitution.replace_old_param 11.10% : 0.000023s : 6: substitution.updatestate_pure_node_eliminater 8.78% : 0.000018s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.003989 2 85.05% : 0.003393s : 1: type_inference.infer 14.95% : 0.000597s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000120 1 100.00% : 0.000120s : 1: match.inline ------[predicate.] 0.000272 1419 0.78% : 0.000002s : 13: predicate.accumulaten_eliminater 1.29% : 0.000003s : 6: predicate.ad_related_special_op_eliminate 0.65% : 0.000002s : 12: predicate.addn_check_dump 0.85% : 0.000002s : 13: predicate.addn_zero_filter 0.70% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 2.14% : 0.000006s : 25: predicate.arithmetic_simplify 0.85% : 0.000002s : 13: predicate.cast_eliminate 0.75% : 0.000002s : 12: predicate.check_bprop_eliminate 0.67% : 0.000002s : 12: predicate.compare_switch_simplify 0.19% : 0.000001s : 6: predicate.const_output_eliminate 0.70% : 0.000002s : 12: predicate.depend_value_elim 0.77% : 0.000002s : 13: predicate.dict_get_item_const_eliminator 0.94% : 0.000003s : 13: predicate.dict_get_item_eliminator 0.74% : 0.000002s : 13: predicate.dict_set_item_eliminator 1.77% : 0.000005s : 12: predicate.dumpgradient_eliminate 0.24% : 0.000001s : 6: predicate.elim_not_effective 0.64% : 0.000002s : 6: predicate.elim_shapecalc_of_broadcastargs 1.14% : 0.000003s : 19: predicate.environ_add_const_eliminate 1.03% : 0.000003s : 19: predicate.environ_get_add_eliminate 1.03% : 0.000003s : 19: predicate.environ_get_depend_swap 1.67% : 0.000005s : 31: predicate.environ_get_eliminate 0.99% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.79% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.36% : 0.000004s : 14: predicate.float_depend_g_call 0.73% : 0.000002s : 12: predicate.float_environ_get_switch 0.94% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.18% : 0.000000s : 6: predicate.fold_const_symbol 0.86% : 0.000002s : 12: predicate.get_grad_eliminate 0.21% : 0.000001s : 6: predicate.graph_param_transform 0.68% : 0.000002s : 12: predicate.incorporate_call 0.62% : 0.000002s : 12: predicate.incorporate_call_switch 5.49% : 0.000015s : 63: predicate.inline 0.76% : 0.000002s : 12: predicate.inline_without_move 0.33% : 0.000001s : 12: predicate.j_node_and_user_rematch 1.21% : 0.000003s : 12: predicate.less_batch_normalization 1.80% : 0.000005s : 25: predicate.list_to_tuple_eliminator_ 2.09% : 0.000006s : 38: predicate.load_eliminater 2.00% : 0.000005s : 6: predicate.loop_unroll_after_grad 1.08% : 0.000003s : 17: predicate.loop_unroll_before_grad 1.93% : 0.000005s : 25: predicate.make_slice_get_slice_eliminator 0.71% : 0.000002s : 12: predicate.merge_addn 0.70% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.75% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.70% : 0.000002s : 13: predicate.minmaximum_grad 1.79% : 0.000005s : 6: predicate.mutable_eliminate 0.47% : 0.000001s : 6: predicate.opt_reshape 0.51% : 0.000001s : 6: predicate.parallel_virtual_node 1.17% : 0.000003s : 14: predicate.partial_defer_inline 1.04% : 0.000003s : 19: predicate.partial_eliminate 0.87% : 0.000002s : 13: predicate.print_const_string_wrapper 0.74% : 0.000002s : 12: predicate.reduce_all_const_elim 1.18% : 0.000003s : 13: predicate.reduce_eliminate 2.20% : 0.000006s : 38: predicate.redundant_stop_gradient_eliminater 0.67% : 0.000002s : 12: predicate.remove_not_recompute_node 0.99% : 0.000003s : 25: predicate.replace_applicator 0.59% : 0.000002s : 12: predicate.replace_old_param 0.39% : 0.000001s : 6: predicate.reset_defer_inline 0.94% : 0.000003s : 13: predicate.reshape_eliminate 0.83% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.71% : 0.000002s : 6: predicate.row_tensor_eliminate 1.18% : 0.000003s : 12: predicate.same_eliminate 0.43% : 0.000001s : 12: predicate.set_cell_output_no_recompute 1.27% : 0.000003s : 12: predicate.shard_identity_eliminate 1.01% : 0.000003s : 12: predicate.special_op_eliminate 0.75% : 0.000002s : 12: predicate.specialize_transform 1.16% : 0.000003s : 12: predicate.split_environ_get_set_with_tuple_value 1.04% : 0.000003s : 12: predicate.stack_unstack_eliminate 0.44% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.81% : 0.000002s : 14: predicate.switch_defer_inline 1.65% : 0.000004s : 26: predicate.switch_layer_defer_inline 3.74% : 0.000010s : 49: predicate.switch_simplify 0.72% : 0.000002s : 13: predicate.tile_eliminate 0.96% : 0.000003s : 13: predicate.transpose_eliminate 1.72% : 0.000005s : 25: predicate.tuple_list_convert_item_index_to_positive 1.72% : 0.000005s : 25: predicate.tuple_list_get_item_const_eliminator 1.64% : 0.000004s : 25: predicate.tuple_list_get_item_depend_reorder 3.34% : 0.000009s : 37: predicate.tuple_list_get_item_eliminator 1.70% : 0.000005s : 25: predicate.tuple_list_get_set_item_eliminator 2.63% : 0.000007s : 37: predicate.tuple_list_set_item_eliminator 1.67% : 0.000005s : 25: predicate.tuple_to_list_eliminator_ 2.03% : 0.000006s : 38: predicate.updatestate_pure_node_eliminater 3.50% : 0.000010s : 50: predicate.updatestate_useless_node_eliminater 0.54% : 0.000001s : 6: predicate.value_based_eliminate 0.93% : 0.000003s : 12: predicate.virtual_dataset_eliminate 0.91% : 0.000002s : 12: predicate.virtual_output_eliminate 0.39% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.60% : 0.000002s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000482 4 5.14% : 0.000025s : 1: func_graph_cloner_run.FuncGraphClonerGraph 94.86% : 0.000457s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.227716 196 0.00% : 0.000005s : 1: ForceFp32Comm 37.47% : 0.085332s : 1: add_attr 37.46% : 0.085314s : 1: add_attr_with_inline 0.00% : 0.000005s : 1: add_comm_op_reuse_tag 0.04% : 0.000102s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.07% : 0.000159s : 1: auto_monad 0.02% : 0.000042s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.23% : 0.000517s : 1: bootstrap 0.02% : 0.000049s : 1: cconv 0.00% : 0.000006s : 1: comm_op_add_attrs 0.01% : 0.000031s : 1: control_data_broadcast_order 0.01% : 0.000024s : 1: convert_after_rewriter 0.02% : 0.000040s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.01% : 0.000013s : 1: environ_conv 0.03% : 0.000064s : 1: event_method 0.01% : 0.000020s : 1: execute 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000010s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.01% : 0.000023s : 1: label_micro_interleaved_index 0.53% : 0.001200s : 1: loop_unroll 0.00% : 0.000007s : 1: merge_cast_opt 0.00% : 0.000007s : 1: micro_interleaved_order_control 0.45% : 0.001027s : 1: mutable_eliminate 0.01% : 0.000012s : 1: offloading_packed_experts 0.01% : 0.000031s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000029s : 1: opt.transform.mutable_eliminate 0.68% : 0.001551s : 78: opt.transform.opt_a 0.04% : 0.000086s : 1: opt.transform.opt_after_cconv 0.05% : 0.000108s : 1: opt.transform.opt_after_jit_grad 0.12% : 0.000270s : 28: opt.transform.opt_b 0.05% : 0.000119s : 2: opt.transform.opt_trans_graph 0.03% : 0.000069s : 4: opt.transform.symbol_engine_opt 1.77% : 0.004031s : 1: opt_a 0.10% : 0.000235s : 1: opt_after_cconv 0.46% : 0.001053s : 1: opt_after_jit_grad 0.21% : 0.000473s : 1: opt_b 3.69% : 0.008409s : 1: optimize 0.02% : 0.000037s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000012s : 1: order_py_execute_after_rewriter 0.02% : 0.000036s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000011s : 1: overlap_grad_ring_attention 0.02% : 0.000050s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000007s : 1: overlap_opt_shard_in_pipeline 0.01% : 0.000012s : 1: overlap_param_gather 0.00% : 0.000010s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000011s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000011s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000007s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.02% : 0.000040s : 1: pre_auto_parallel 0.01% : 0.000031s : 1: py_interpret_to_execute 0.01% : 0.000028s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.01% : 0.000025s : 1: remove_dup_value 0.27% : 0.000610s : 1: renormalize.infer 0.24% : 0.000554s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.12% : 0.000278s : 1: rewriter_after_opt_a 0.03% : 0.000069s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000008s : 1: split_matmul_comm_elemetwise 0.01% : 0.000014s : 1: swap_dp_allreduce_reducescatter 0.06% : 0.000140s : 1: symbol_engine_optimizer 13.50% : 0.030737s : 1: task_emit 0.07% : 0.000163s : 1: tuple_transform 1.79% : 0.004073s : 1: type_inference 0.05% : 0.000120s : 1: validate TotalTime = 0.692374, [24] [bootstrap]: 0.00053426 [type_inference]: 0.100951 [event_method]: 6.451e-05 [auto_monad]: 0.00033343 [graph_reusing]: 9.76e-06 [inline]: 3.23998e-06 [add_attr]: 0.00417938, [1] [add_attr_with_inline]: 0.00416566, [1] [Cycle 1]: 0.00010144, [2] [tag_attr]: 4.685e-05 [meta_addattr_fg_expand]: 9.51e-06 [parallel-infer-symbol]: 4.07e-06 [pre_auto_parallel]: 6.445e-05 [insert-virtual-dataset]: 2.72001e-06 [parallel-infer-symbol-second]: 1.64e-06 [dataset_repeat_opt]: 2.29001e-06 [pipeline_split]: 1.71e-06 [optimize]: 0.438843, [53] [py_interpret_to_execute]: 5.568e-05 [rewriter_before_opt_a]: 0.00014928 [opt_a]: 0.435597, [3] [Cycle 1]: 0.32302, [45] [expand_dump_flag]: 7.51999e-06 [switch_simplify]: 8.371e-05 [loop_unroll]: 6.993e-05 [a_1]: 0.00167776 [with_stream_mark]: 4.154e-05 [recompute_prepare]: 3.434e-05 [updatestate_depend_eliminate]: 1.368e-05 [updatestate_assign_eliminate]: 1.113e-05 [updatestate_loads_eliminate]: 1.025e-05 [parameter_eliminate]: 3.63e-06 [a_2]: 0.0003169 [accelerated_algorithm]: 3.904e-05 [shard]: 2.51998e-06 [meta_shard_fg_expand]: 5.80002e-06 [shard_inline]: 2.167e-05 [merge_send_recv]: 2.451e-05 [auto_parallel]: 1.832e-05 [parallel]: 2.339e-05 [flash_sp]: 1.887e-05 [merge_comm]: 1.302e-05 [allreduce_fusion]: 1.236e-05 [matmul_add_comm_reduction]: 4.329e-05 [allreduce_slice_to_reducescatter]: 1.02e-06 [virtual_shard_identity]: 2.997e-05 [virtual_dataset]: 2.289e-05 [get_grad_eliminate_]: 2.357e-05 [virtual_output]: 2.35e-05 [merge_forward]: 1.366e-05 [cell_reuse_recompute_pass]: 1.78997e-06 [offload_activation]: 2.453e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.992e-05 [merge_recompute_call_nodes]: 1.92999e-06 [before_grad]: 4.457e-05 [set_forward_comm_id_for_comm_node_pass]: 1.442e-05 [meta_fg_expand]: 0.176294 [flash_sp_send_recv_attached]: 7.02002e-06 [receive_attached]: 2.59999e-06 [after_resolve]: 0.00015206 [a_after_grad]: 0.0002249 [renormalize]: 0.140796 [add_forward_monad_depend]: 2.993e-05 [auto_monad_grad]: 1.945e-05 [auto_monad_eliminator]: 0.00017595 [cse]: 0.00037367 [a_3]: 0.00171189 [Cycle 2]: 0.111339, [45] [expand_dump_flag]: 7.82998e-06 [switch_simplify]: 0.00011351 [loop_unroll]: 0.0001068 [a_1]: 0.107711 [with_stream_mark]: 3.729e-05 [recompute_prepare]: 2.066e-05 [updatestate_depend_eliminate]: 4.383e-05 [updatestate_assign_eliminate]: 7.34002e-06 [updatestate_loads_eliminate]: 6.10002e-06 [parameter_eliminate]: 2.48e-06 [a_2]: 0.00016614 [accelerated_algorithm]: 2.033e-05 [shard]: 2.35002e-06 [meta_shard_fg_expand]: 6.59999e-06 [shard_inline]: 1.127e-05 [merge_send_recv]: 1.338e-05 [auto_parallel]: 1.318e-05 [parallel]: 1.127e-05 [flash_sp]: 4.87998e-06 [merge_comm]: 6.78e-06 [allreduce_fusion]: 6.43998e-06 [matmul_add_comm_reduction]: 1.48e-05 [allreduce_slice_to_reducescatter]: 9.80013e-07 [virtual_shard_identity]: 1.543e-05 [virtual_dataset]: 1.159e-05 [get_grad_eliminate_]: 1.06e-05 [virtual_output]: 1.059e-05 [merge_forward]: 7.33999e-06 [cell_reuse_recompute_pass]: 1.71e-06 [offload_activation]: 1.583e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.453e-05 [merge_recompute_call_nodes]: 2.46e-06 [before_grad]: 2.08e-05 [set_forward_comm_id_for_comm_node_pass]: 7.86001e-06 [meta_fg_expand]: 9.345e-05 [flash_sp_send_recv_attached]: 2.02999e-06 [receive_attached]: 2.44001e-06 [after_resolve]: 2.438e-05 [a_after_grad]: 1.975e-05 [renormalize]: 0.00206633 [add_forward_monad_depend]: 1.118e-05 [auto_monad_grad]: 2.83e-06 [auto_monad_eliminator]: 3.116e-05 [cse]: 0.00017328 [a_3]: 0.00010322 [Cycle 3]: 0.00121349, [45] [expand_dump_flag]: 3.77998e-06 [switch_simplify]: 1.59e-05 [loop_unroll]: 1.208e-05 [a_1]: 0.00032569 [with_stream_mark]: 2.29e-05 [recompute_prepare]: 1.185e-05 [updatestate_depend_eliminate]: 8.12e-06 [updatestate_assign_eliminate]: 7.36999e-06 [updatestate_loads_eliminate]: 7.25e-06 [parameter_eliminate]: 1.96e-06 [a_2]: 0.00015828 [accelerated_algorithm]: 1.831e-05 [shard]: 2.14999e-06 [meta_shard_fg_expand]: 3.48e-06 [shard_inline]: 1.171e-05 [merge_send_recv]: 1.188e-05 [auto_parallel]: 1.381e-05 [parallel]: 1.032e-05 [flash_sp]: 1.79e-06 [merge_comm]: 8.15e-06 [allreduce_fusion]: 6.94001e-06 [matmul_add_comm_reduction]: 1.454e-05 [allreduce_slice_to_reducescatter]: 6.99976e-07 [virtual_shard_identity]: 1.474e-05 [virtual_dataset]: 1.327e-05 [get_grad_eliminate_]: 1.206e-05 [virtual_output]: 1.149e-05 [merge_forward]: 7.46999e-06 [cell_reuse_recompute_pass]: 3.10998e-06 [offload_activation]: 1.45e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.594e-05 [merge_recompute_call_nodes]: 1.54e-06 [before_grad]: 2.245e-05 [set_forward_comm_id_for_comm_node_pass]: 7.86001e-06 [meta_fg_expand]: 5.70001e-06 [flash_sp_send_recv_attached]: 2.48e-06 [receive_attached]: 2.39001e-06 [after_resolve]: 1.945e-05 [a_after_grad]: 1.826e-05 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 2.34999e-06 [auto_monad_grad]: 1.85001e-06 [auto_monad_eliminator]: 1.72e-05 [cse]: 3.443e-05 [a_3]: 7.329e-05 [py_interpret_to_execute_after_opt_a]: 2.514e-05 [slice_cell_reuse_recomputed_activation]: 2.39001e-06 [rewriter_after_opt_a]: 7.371e-05 [convert_after_rewriter]: 1.181e-05 [order_py_execute_after_rewriter]: 7.57002e-06 [mutable_eliminate]: 0.00080309 [opt_b]: 0.00039149, [1] [Cycle 1]: 0.00038267, [7] [b_1]: 0.00026034 [b_2]: 1.45e-05 [updatestate_depend_eliminate]: 1.152e-05 [updatestate_assign_eliminate]: 6.33998e-06 [updatestate_loads_eliminate]: 5.99e-06 [renormalize]: 8.70001e-07 [cse]: 4.302e-05 [optimize_parallel_all_gather_comm]: 2.651e-05 [overlap_param_gather]: 1.78002e-06 [cconv]: 3.788e-05 [loop_unroll]: 0.00062658 [opt_after_cconv]: 0.00018237, [1] [Cycle 1]: 0.00017478, [7] [c_1]: 7.497e-05 [parameter_eliminate]: 4.80001e-06 [updatestate_depend_eliminate]: 1.058e-05 [updatestate_assign_eliminate]: 5.44998e-06 [updatestate_loads_eliminate]: 5.42001e-06 [cse]: 3.566e-05 [renormalize]: 5.20027e-07 [remove_dup_value]: 2.03e-05 [tuple_transform]: 0.00012581, [1] [Cycle 1]: 0.00012007, [4] [d_1]: 8.518e-05 [none_parameter_eliminate]: 2.01e-06 [renormalize]: 1.69995e-07 [switch_simplify]: 1.227e-05 [partial_unused_args_eliminate]: 2.09e-06 [add_recomputation]: 0.00010419 [cse_after_recomputation]: 4.534e-05, [1] [Cycle 1]: 3.864e-05, [1] [cse]: 3.106e-05 [environ_conv]: 1.337e-05 [swap_dp_allreduce_reducescatter]: 1.03e-05 [bias_add_comm_swap]: 3.51999e-06 [label_micro_interleaved_index]: 5.49e-06 [label_fine_grained_interleaved_index]: 2.84001e-06 [merge_cast_opt]: 1.65001e-06 [slice_recompute_activation]: 2.24001e-06 [micro_interleaved_order_control]: 2.98e-06 [assign_add_opt]: 1.47001e-06 [ForceFp32Comm]: 1.11002e-06 [remove_cast_before_assign_add]: 1.43002e-06 [full_micro_interleaved_order_control]: 2.46e-06 [reorder_send_recv_between_fp_bp]: 2.66e-06 [comm_op_add_attrs]: 1.10001e-06 [add_comm_op_reuse_tag]: 1.02998e-06 [interleave_split_concat_branches]: 1.44e-06 [interleave_parallel_branches]: 1.04e-06 [overlap_opt_shard_in_pipeline]: 1.68002e-06 [overlap_opt_shard_grad_in_pipeline]: 2.24001e-06 [control_data_broadcast_order]: 2.564e-05 [grouped_pairwise_exchange_alltoall]: 2.17999e-06 [offloading_packed_experts]: 6.53e-06 [overlap_recompute_and_grad_model_parallel]: 6.95998e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.25001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.39e-06 [overlap_recompute_comm]: 2.58e-06 [overlap_grad_ring_attention]: 7.15e-06 [overlap_grad_flash_sp]: 3.449e-05 [begin_end_overlap_inline]: 6.10016e-07 [split_matmul_comm_elemetwise]: 2.83e-06 [split_layernorm_comm]: 2.40002e-06 [handle_group_info]: 9.89996e-07 [symbol_engine_optimizer]: 0.00013817, [1] [Cycle 1]: 0.00013074, [6] [build]: 1.495e-05 [elim_shapecalc]: 2.147e-05 [elim_not_effective]: 2.454e-05 [opt_reshape]: 1.499e-05 [fold_const_symbol]: 1.911e-05 [renormalize]: 4.30009e-07 [detach_backward]: 2.81e-06 [pipeline_parallel_scheduler]: 2.78e-06 [auto_monad_reorder]: 4.078e-05 [get_jit_bprop_graph]: 1.87001e-06 [rewriter_after_jit_bprop_graph]: 6.41e-06 [opt_after_jit_grad]: 0.00067648 [validate]: 6.807e-05 [backend_pass]: 1.06002e-06 [task_emit]: 0.146241 [execute]: 1.049e-05 Sums bootstrap : 0.000534s : 0.08% type_inference : 0.100951s : 14.71% event_method : 0.000065s : 0.01% auto_monad : 0.000333s : 0.05% graph_reusing : 0.000010s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000047s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000010s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000064s : 0.01% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000002s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000056s : 0.01% optimize.rewriter_before_opt_a : 0.000149s : 0.02% optimize.opt_a.expand_dump_flag : 0.000019s : 0.00% optimize.opt_a.switch_simplify : 0.000213s : 0.03% optimize.opt_a.loop_unroll : 0.000189s : 0.03% optimize.opt_a.a_1 : 0.109714s : 15.98% optimize.opt_a.with_stream_mark : 0.000102s : 0.01% optimize.opt_a.recompute_prepare : 0.000067s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000066s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000026s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000024s : 0.00% optimize.opt_a.parameter_eliminate : 0.000008s : 0.00% optimize.opt_a.a_2 : 0.000641s : 0.09% optimize.opt_a.accelerated_algorithm : 0.000078s : 0.01% optimize.opt_a.shard : 0.000007s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000016s : 0.00% optimize.opt_a.shard_inline : 0.000045s : 0.01% optimize.opt_a.merge_send_recv : 0.000050s : 0.01% optimize.opt_a.auto_parallel : 0.000045s : 0.01% optimize.opt_a.parallel : 0.000045s : 0.01% optimize.opt_a.flash_sp : 0.000026s : 0.00% optimize.opt_a.merge_comm : 0.000028s : 0.00% optimize.opt_a.allreduce_fusion : 0.000026s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000073s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000003s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000060s : 0.01% optimize.opt_a.virtual_dataset : 0.000048s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000046s : 0.01% optimize.opt_a.virtual_output : 0.000046s : 0.01% optimize.opt_a.merge_forward : 0.000028s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000007s : 0.00% optimize.opt_a.offload_activation : 0.000055s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000100s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000006s : 0.00% optimize.opt_a.before_grad : 0.000088s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000030s : 0.00% optimize.opt_a.meta_fg_expand : 0.176393s : 25.70% optimize.opt_a.flash_sp_send_recv_attached : 0.000012s : 0.00% optimize.opt_a.receive_attached : 0.000007s : 0.00% optimize.opt_a.after_resolve : 0.000196s : 0.03% optimize.opt_a.a_after_grad : 0.000263s : 0.04% optimize.opt_a.renormalize : 0.142862s : 20.81% optimize.opt_a.add_forward_monad_depend : 0.000043s : 0.01% optimize.opt_a.auto_monad_grad : 0.000024s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000224s : 0.03% optimize.opt_a.cse : 0.000581s : 0.08% optimize.opt_a.a_3 : 0.001888s : 0.28% optimize.py_interpret_to_execute_after_opt_a : 0.000025s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000074s : 0.01% optimize.convert_after_rewriter : 0.000012s : 0.00% optimize.order_py_execute_after_rewriter : 0.000008s : 0.00% optimize.mutable_eliminate : 0.000803s : 0.12% optimize.opt_b.b_1 : 0.000260s : 0.04% optimize.opt_b.b_2 : 0.000014s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000012s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000043s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000027s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000038s : 0.01% optimize.loop_unroll : 0.000627s : 0.09% optimize.opt_after_cconv.c_1 : 0.000075s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000011s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.cse : 0.000036s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000020s : 0.00% optimize.tuple_transform.d_1 : 0.000085s : 0.01% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000012s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000104s : 0.02% optimize.cse_after_recomputation.cse : 0.000031s : 0.00% optimize.environ_conv : 0.000013s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000010s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000026s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000007s : 0.00% optimize.overlap_grad_flash_sp : 0.000034s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000015s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000021s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000025s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000015s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000019s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000003s : 0.00% auto_monad_reorder : 0.000041s : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.000676s : 0.10% validate : 0.000068s : 0.01% backend_pass : 0.000001s : 0.00% task_emit : 0.146241s : 21.31% execute : 0.000010s : 0.00% Time group info: ------[substitution.] 0.002737 310 0.63% : 0.000017s : 4: substitution.depend_value_elim 0.15% : 0.000004s : 7: substitution.elim_not_effective 0.26% : 0.000007s : 7: substitution.float_depend_g_call 0.18% : 0.000005s : 2: substitution.float_tuple_getitem_switch 0.11% : 0.000003s : 7: substitution.fold_const_symbol 38.34% : 0.001050s : 4: substitution.getattr_setattr_resolve 0.36% : 0.000010s : 9: substitution.graph_param_transform 0.09% : 0.000003s : 2: substitution.incorporate_call 0.07% : 0.000002s : 2: substitution.incorporate_call_switch 39.44% : 0.001080s : 24: substitution.inline 2.05% : 0.000056s : 5: substitution.inline_without_move 0.60% : 0.000016s : 28: substitution.j_node_and_user_rematch 0.79% : 0.000022s : 3: substitution.less_batch_normalization 0.54% : 0.000015s : 9: substitution.minmaximum_grad 0.31% : 0.000009s : 7: substitution.partial_eliminate 0.79% : 0.000022s : 28: substitution.remove_not_recompute_node 4.27% : 0.000117s : 38: substitution.replace_applicator 0.79% : 0.000021s : 19: substitution.replace_old_param 0.17% : 0.000005s : 1: substitution.set_cell_output_no_recompute 1.07% : 0.000029s : 9: substitution.tuple_list_convert_item_index_to_positive 0.50% : 0.000014s : 9: substitution.tuple_list_get_item_const_eliminator 0.68% : 0.000019s : 9: substitution.tuple_list_get_item_depend_reorder 4.03% : 0.000110s : 31: substitution.tuple_list_get_item_eliminator 0.68% : 0.000019s : 9: substitution.tuple_list_get_set_item_eliminator 1.22% : 0.000034s : 16: substitution.updatestate_pure_node_eliminater 1.91% : 0.000052s : 21: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.100833 2 98.17% : 0.098990s : 1: type_inference.infer 1.83% : 0.001842s : 1: type_inference.specialize ------[replace.] 0.000662 49 10.95% : 0.000072s : 3: replace.getattr_setattr_resolve 35.51% : 0.000235s : 24: replace.inline 10.81% : 0.000072s : 1: replace.replace_applicator 40.66% : 0.000269s : 20: replace.tuple_list_get_item_eliminator 2.08% : 0.000014s : 1: replace.updatestate_useless_node_eliminater ------[match.] 0.002169 49 46.23% : 0.001003s : 3: match.getattr_setattr_resolve 48.89% : 0.001061s : 24: match.inline 0.98% : 0.000021s : 1: match.replace_applicator 3.04% : 0.000066s : 20: match.tuple_list_get_item_eliminator 0.86% : 0.000019s : 1: match.updatestate_useless_node_eliminater ------[predicate.] 0.105596 8683 0.01% : 0.000012s : 85: predicate.accumulaten_eliminater 0.00% : 0.000003s : 9: predicate.ad_related_special_op_eliminate 0.00% : 0.000005s : 37: predicate.addn_check_dump 0.01% : 0.000013s : 85: predicate.addn_zero_filter 0.01% : 0.000011s : 85: predicate.adjust_all_reduce_mul_add 0.02% : 0.000026s : 122: predicate.arithmetic_simplify 0.01% : 0.000013s : 85: predicate.cast_eliminate 0.03% : 0.000034s : 260: predicate.check_bprop_eliminate 0.00% : 0.000005s : 37: predicate.compare_switch_simplify 0.00% : 0.000001s : 9: predicate.const_output_eliminate 0.00% : 0.000005s : 37: predicate.depend_value_elim 0.01% : 0.000012s : 85: predicate.dict_get_item_const_eliminator 0.01% : 0.000014s : 85: predicate.dict_get_item_eliminator 0.01% : 0.000012s : 85: predicate.dict_set_item_eliminator 0.00% : 0.000004s : 18: predicate.dumpgradient_eliminate 0.00% : 0.000001s : 9: predicate.elim_not_effective 0.00% : 0.000002s : 9: predicate.elim_shapecalc_of_broadcastargs 0.01% : 0.000012s : 94: predicate.environ_add_const_eliminate 0.01% : 0.000012s : 94: predicate.environ_get_add_eliminate 0.01% : 0.000012s : 94: predicate.environ_get_depend_swap 0.02% : 0.000018s : 131: predicate.environ_get_eliminate 0.01% : 0.000012s : 94: predicate.environ_get_set_eliminate 0.02% : 0.000018s : 130: predicate.exchange_switch_depend_value 0.02% : 0.000026s : 130: predicate.float_depend_g_call 0.00% : 0.000005s : 37: predicate.float_environ_get_switch 0.01% : 0.000006s : 46: predicate.float_tuple_getitem_switch 0.00% : 0.000001s : 9: predicate.fold_const_symbol 0.01% : 0.000006s : 37: predicate.get_grad_eliminate 0.01% : 0.000006s : 20: predicate.getattr_setattr_resolve 0.00% : 0.000001s : 9: predicate.graph_param_transform 0.01% : 0.000005s : 37: predicate.incorporate_call 0.00% : 0.000005s : 37: predicate.incorporate_call_switch 0.06% : 0.000061s : 307: predicate.inline 0.02% : 0.000022s : 111: predicate.inline_without_move 0.00% : 0.000003s : 37: predicate.j_node_and_user_rematch 0.01% : 0.000008s : 37: predicate.less_batch_normalization 0.02% : 0.000018s : 123: predicate.list_to_tuple_eliminator_ 0.03% : 0.000028s : 208: predicate.load_eliminater 0.00% : 0.000004s : 9: predicate.loop_unroll_after_grad 0.03% : 0.000027s : 197: predicate.loop_unroll_before_grad 0.01% : 0.000015s : 103: predicate.make_slice_get_slice_eliminator 0.01% : 0.000006s : 37: predicate.merge_addn 0.03% : 0.000033s : 256: predicate.micro_step_allgather_replace 0.03% : 0.000033s : 256: predicate.mini_step_allgather_replace 0.01% : 0.000012s : 85: predicate.minmaximum_grad 0.00% : 0.000005s : 9: predicate.mutable_eliminate 0.00% : 0.000002s : 9: predicate.opt_reshape 0.00% : 0.000002s : 9: predicate.parallel_virtual_node 0.03% : 0.000031s : 130: predicate.partial_defer_inline 0.02% : 0.000020s : 114: predicate.partial_eliminate 0.01% : 0.000011s : 85: predicate.print_const_string_wrapper 0.01% : 0.000005s : 37: predicate.reduce_all_const_elim 0.01% : 0.000015s : 85: predicate.reduce_eliminate 0.03% : 0.000028s : 208: predicate.redundant_stop_gradient_eliminater 0.00% : 0.000004s : 37: predicate.remove_not_recompute_node 0.03% : 0.000033s : 363: predicate.replace_applicator 0.01% : 0.000010s : 111: predicate.replace_old_param 0.00% : 0.000001s : 9: predicate.reset_defer_inline 0.01% : 0.000012s : 85: predicate.reshape_eliminate 0.03% : 0.000035s : 256: predicate.row_tensor_add_zeros_like 0.00% : 0.000002s : 9: predicate.row_tensor_eliminate 0.04% : 0.000037s : 260: predicate.same_eliminate 0.00% : 0.000003s : 39: predicate.set_cell_output_no_recompute 0.01% : 0.000007s : 37: predicate.shard_identity_eliminate 0.00% : 0.000003s : 18: predicate.special_op_eliminate 0.01% : 0.000006s : 37: predicate.specialize_transform 0.04% : 0.000037s : 256: predicate.split_environ_get_set_with_tuple_value 0.02% : 0.000020s : 111: predicate.stack_unstack_eliminate 0.00% : 0.000002s : 9: predicate.switch_call_monad_eliminater 0.02% : 0.000020s : 130: predicate.switch_defer_inline 0.05% : 0.000053s : 390: predicate.switch_layer_defer_inline 0.05% : 0.000056s : 373: predicate.switch_simplify 0.01% : 0.000013s : 85: predicate.tile_eliminate 0.01% : 0.000013s : 85: predicate.transpose_eliminate 0.02% : 0.000016s : 103: predicate.tuple_list_convert_item_index_to_positive 0.02% : 0.000017s : 103: predicate.tuple_list_get_item_const_eliminator 0.01% : 0.000015s : 103: predicate.tuple_list_get_item_depend_reorder 0.03% : 0.000031s : 160: predicate.tuple_list_get_item_eliminator 0.01% : 0.000016s : 103: predicate.tuple_list_get_set_item_eliminator 0.02% : 0.000024s : 140: predicate.tuple_list_set_item_eliminator 0.02% : 0.000019s : 123: predicate.tuple_to_list_eliminator_ 0.03% : 0.000028s : 208: predicate.updatestate_pure_node_eliminater 98.81% : 0.104340s : 246: predicate.updatestate_useless_node_eliminater 0.00% : 0.000002s : 9: predicate.value_based_eliminate 0.01% : 0.000006s : 37: predicate.virtual_dataset_eliminate 0.01% : 0.000006s : 37: predicate.virtual_output_eliminate 0.00% : 0.000001s : 9: predicate.virtual_view_grad_eliminate 0.00% : 0.000002s : 9: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.173102 53 98.79% : 0.171002s : 23: func_graph_cloner_run.FuncGraphClonerGraph 1.21% : 0.002099s : 30: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.393362 247 0.00% : 0.000005s : 1: ForceFp32Comm 0.30% : 0.004186s : 1: add_attr 0.30% : 0.004170s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000110s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.03% : 0.000350s : 1: auto_monad 0.00% : 0.000046s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.04% : 0.000575s : 1: bootstrap 0.00% : 0.000041s : 1: cconv 0.00% : 0.000005s : 1: comm_op_add_attrs 0.00% : 0.000031s : 1: control_data_broadcast_order 0.00% : 0.000015s : 1: convert_after_rewriter 0.00% : 0.000048s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000016s : 1: detach_backward 0.00% : 0.000017s : 1: environ_conv 0.01% : 0.000075s : 1: event_method 0.00% : 0.000018s : 1: execute 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000007s : 1: get_jit_bprop_graph 0.00% : 0.000015s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000008s : 1: insert-virtual-dataset 0.00% : 0.000006s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000010s : 1: label_micro_interleaved_index 0.05% : 0.000638s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.06% : 0.000815s : 1: mutable_eliminate 0.00% : 0.000010s : 1: offloading_packed_experts 0.00% : 0.000022s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000028s : 1: opt.transform.mutable_eliminate 8.14% : 0.113476s : 125: opt.transform.opt_a 0.01% : 0.000073s : 1: opt.transform.opt_after_cconv 0.00% : 0.000052s : 1: opt.transform.opt_after_jit_grad 0.02% : 0.000241s : 28: opt.transform.opt_b 0.09% : 0.001190s : 2: opt.transform.opt_resolve 0.01% : 0.000095s : 2: opt.transform.opt_trans_graph 0.01% : 0.000076s : 4: opt.transform.symbol_engine_opt 31.26% : 0.435600s : 1: opt_a 0.01% : 0.000186s : 1: opt_after_cconv 0.05% : 0.000691s : 1: opt_after_jit_grad 0.03% : 0.000395s : 1: opt_b 31.50% : 0.438850s : 1: optimize 0.00% : 0.000030s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000011s : 1: order_py_execute_after_rewriter 0.00% : 0.000038s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000010s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000010s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000005s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000070s : 1: pre_auto_parallel 0.00% : 0.000060s : 1: py_interpret_to_execute 0.00% : 0.000029s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.00% : 0.000025s : 1: remove_dup_value 9.81% : 0.136669s : 2: renormalize.infer 0.44% : 0.006158s : 2: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000079s : 1: rewriter_after_opt_a 0.01% : 0.000154s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000013s : 1: swap_dp_allreduce_reducescatter 0.01% : 0.000141s : 1: symbol_engine_optimizer 10.50% : 0.146265s : 1: task_emit 0.01% : 0.000130s : 1: tuple_transform 7.25% : 0.100986s : 1: type_inference 0.01% : 0.000119s : 1: validate . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[0-shape3] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[0-shape3],max_mem:4.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:42:05.540.805 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.195437, [24] [bootstrap]: 0.00123876 [type_inference]: 0.090678 [event_method]: 4.208e-05 [auto_monad]: 0.00127502 [graph_reusing]: 1.189e-05 [inline]: 3.58999e-06 [add_attr]: 0.00878407, [1] [add_attr_with_inline]: 0.00875532, [1] [Cycle 1]: 7.409e-05, [2] [tag_attr]: 1.942e-05 [meta_addattr_fg_expand]: 4.20999e-06 [parallel-infer-symbol]: 4.60001e-06 [pre_auto_parallel]: 4.504e-05 [insert-virtual-dataset]: 2.83e-06 [parallel-infer-symbol-second]: 9.20001e-07 [dataset_repeat_opt]: 2.60997e-06 [pipeline_split]: 2.02001e-06 [optimize]: 0.0189135, [53] [py_interpret_to_execute]: 6.568e-05 [rewriter_before_opt_a]: 8.004e-05 [opt_a]: 0.00520656, [2] [Cycle 1]: 0.00383108, [45] [expand_dump_flag]: 6.57002e-06 [switch_simplify]: 2.857e-05 [loop_unroll]: 1.888e-05 [a_1]: 0.00082919 [with_stream_mark]: 2.834e-05 [recompute_prepare]: 1.828e-05 [updatestate_depend_eliminate]: 7.16999e-06 [updatestate_assign_eliminate]: 9.46003e-06 [updatestate_loads_eliminate]: 8.65999e-06 [parameter_eliminate]: 2.64001e-06 [a_2]: 0.00015063 [accelerated_algorithm]: 1.184e-05 [shard]: 2.99999e-06 [meta_shard_fg_expand]: 3.13e-06 [shard_inline]: 1.563e-05 [merge_send_recv]: 1.44e-05 [auto_parallel]: 1.599e-05 [parallel]: 5.356e-05 [flash_sp]: 1.401e-05 [merge_comm]: 1.135e-05 [allreduce_fusion]: 5.97001e-06 [matmul_add_comm_reduction]: 1.328e-05 [allreduce_slice_to_reducescatter]: 9.30013e-07 [virtual_shard_identity]: 1.644e-05 [virtual_dataset]: 1.535e-05 [get_grad_eliminate_]: 1.869e-05 [virtual_output]: 1.089e-05 [merge_forward]: 1.225e-05 [cell_reuse_recompute_pass]: 1.78002e-06 [offload_activation]: 1.826e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.63e-05 [merge_recompute_call_nodes]: 1.60001e-06 [before_grad]: 2.992e-05 [set_forward_comm_id_for_comm_node_pass]: 7.05998e-06 [meta_fg_expand]: 4.26001e-06 [flash_sp_send_recv_attached]: 3.28e-06 [receive_attached]: 3.07002e-06 [after_resolve]: 1.951e-05 [a_after_grad]: 2.655e-05 [renormalize]: 0.00135751 [add_forward_monad_depend]: 1.774e-05 [auto_monad_grad]: 3.23998e-06 [auto_monad_eliminator]: 5.589e-05 [cse]: 4.951e-05 [a_3]: 8.318e-05 [Cycle 2]: 0.00135505, [45] [expand_dump_flag]: 2.48e-06 [switch_simplify]: 1.207e-05 [loop_unroll]: 9.61998e-06 [a_1]: 0.00023162 [with_stream_mark]: 2.351e-05 [recompute_prepare]: 1.374e-05 [updatestate_depend_eliminate]: 7.07002e-06 [updatestate_assign_eliminate]: 9.47001e-06 [updatestate_loads_eliminate]: 6.21e-06 [parameter_eliminate]: 2.11e-06 [a_2]: 0.00012671 [accelerated_algorithm]: 9.90002e-06 [shard]: 3.23998e-06 [meta_shard_fg_expand]: 3.14999e-06 [shard_inline]: 9.22999e-06 [merge_send_recv]: 1.19e-05 [auto_parallel]: 1.248e-05 [parallel]: 1.054e-05 [flash_sp]: 4.77e-06 [merge_comm]: 5.67999e-06 [allreduce_fusion]: 8.87e-06 [matmul_add_comm_reduction]: 1.965e-05 [allreduce_slice_to_reducescatter]: 9.5999e-07 [virtual_shard_identity]: 1.766e-05 [virtual_dataset]: 9.54e-06 [get_grad_eliminate_]: 9.24e-06 [virtual_output]: 1.166e-05 [merge_forward]: 7.69002e-06 [cell_reuse_recompute_pass]: 2.52001e-06 [offload_activation]: 1.974e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.692e-05 [merge_recompute_call_nodes]: 1.69e-06 [before_grad]: 2.711e-05 [set_forward_comm_id_for_comm_node_pass]: 7.15003e-06 [meta_fg_expand]: 4.30999e-06 [flash_sp_send_recv_attached]: 1.81e-06 [receive_attached]: 2.96999e-06 [after_resolve]: 1.592e-05 [a_after_grad]: 1.746e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 3.5e-06 [auto_monad_grad]: 2.36998e-06 [auto_monad_eliminator]: 2.179e-05 [cse]: 3.639e-05 [a_3]: 5.705e-05 [py_interpret_to_execute_after_opt_a]: 2.915e-05 [slice_cell_reuse_recomputed_activation]: 2.82002e-06 [rewriter_after_opt_a]: 0.00027223 [convert_after_rewriter]: 1.847e-05 [order_py_execute_after_rewriter]: 1.44e-05 [mutable_eliminate]: 0.0010215 [opt_b]: 0.00054255, [1] [Cycle 1]: 0.0005269, [7] [b_1]: 0.00034019 [b_2]: 2.094e-05 [updatestate_depend_eliminate]: 1.716e-05 [updatestate_assign_eliminate]: 6.13002e-06 [updatestate_loads_eliminate]: 5.52999e-06 [renormalize]: 1.02e-06 [cse]: 6.081e-05 [optimize_parallel_all_gather_comm]: 4.737e-05 [overlap_param_gather]: 6.02001e-06 [cconv]: 4.537e-05 [loop_unroll]: 0.00723627 [opt_after_cconv]: 0.00027737, [1] [Cycle 1]: 0.00026613, [7] [c_1]: 0.00010149 [parameter_eliminate]: 7.95e-06 [updatestate_depend_eliminate]: 1.251e-05 [updatestate_assign_eliminate]: 6.19999e-06 [updatestate_loads_eliminate]: 9.46998e-06 [cse]: 5.633e-05 [renormalize]: 1.31002e-06 [remove_dup_value]: 2.466e-05 [tuple_transform]: 0.00015715, [1] [Cycle 1]: 0.0001511, [4] [d_1]: 0.00010711 [none_parameter_eliminate]: 1.77999e-06 [renormalize]: 1.80007e-07 [switch_simplify]: 1.142e-05 [partial_unused_args_eliminate]: 2.42001e-06 [add_recomputation]: 0.00012886 [cse_after_recomputation]: 3.489e-05, [1] [Cycle 1]: 2.916e-05, [1] [cse]: 2.067e-05 [environ_conv]: 1.094e-05 [swap_dp_allreduce_reducescatter]: 1.298e-05 [bias_add_comm_swap]: 3.58999e-06 [label_micro_interleaved_index]: 1.798e-05 [label_fine_grained_interleaved_index]: 3.85e-06 [merge_cast_opt]: 2.00002e-06 [slice_recompute_activation]: 2.54001e-06 [micro_interleaved_order_control]: 6.90002e-06 [assign_add_opt]: 1.84e-06 [ForceFp32Comm]: 1.24998e-06 [remove_cast_before_assign_add]: 1.57001e-06 [full_micro_interleaved_order_control]: 2.86e-06 [reorder_send_recv_between_fp_bp]: 2.92002e-06 [comm_op_add_attrs]: 1.47001e-06 [add_comm_op_reuse_tag]: 1.09998e-06 [interleave_split_concat_branches]: 1.30001e-06 [interleave_parallel_branches]: 1.42e-06 [overlap_opt_shard_in_pipeline]: 1.28002e-06 [overlap_opt_shard_grad_in_pipeline]: 1.92999e-06 [control_data_broadcast_order]: 5.174e-05 [grouped_pairwise_exchange_alltoall]: 2.31998e-06 [offloading_packed_experts]: 7.01999e-06 [overlap_recompute_and_grad_model_parallel]: 7.48e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.71e-06 [overlap_recompute_allgather_and_fa_grad]: 1.74e-06 [overlap_recompute_comm]: 2.76999e-06 [overlap_grad_ring_attention]: 5.73002e-06 [overlap_grad_flash_sp]: 4.719e-05 [begin_end_overlap_inline]: 6.79982e-07 [split_matmul_comm_elemetwise]: 2.94999e-06 [split_layernorm_comm]: 1.94e-06 [handle_group_info]: 1.20001e-06 [symbol_engine_optimizer]: 0.0003937, [1] [Cycle 1]: 0.00037725, [6] [build]: 1.142e-05 [elim_shapecalc]: 5.253e-05 [elim_not_effective]: 3.985e-05 [opt_reshape]: 1.665e-05 [fold_const_symbol]: 1.817e-05 [renormalize]: 6.80011e-07 [detach_backward]: 2.39001e-06 [pipeline_parallel_scheduler]: 1.69998e-06 [auto_monad_reorder]: 4.473e-05 [get_jit_bprop_graph]: 1.99999e-06 [rewriter_after_jit_bprop_graph]: 9.11002e-06 [opt_after_jit_grad]: 0.00100689 [validate]: 7.743e-05 [backend_pass]: 7.25e-06 [task_emit]: 0.0637811 [execute]: 1.075e-05 Sums bootstrap : 0.001239s : 0.72% type_inference : 0.090678s : 52.62% event_method : 0.000042s : 0.02% auto_monad : 0.001275s : 0.74% graph_reusing : 0.000012s : 0.01% inline : 0.000004s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000019s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000005s : 0.00% pre_auto_parallel : 0.000045s : 0.03% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000066s : 0.04% optimize.rewriter_before_opt_a : 0.000080s : 0.05% optimize.opt_a.expand_dump_flag : 0.000009s : 0.01% optimize.opt_a.switch_simplify : 0.000041s : 0.02% optimize.opt_a.loop_unroll : 0.000028s : 0.02% optimize.opt_a.a_1 : 0.001061s : 0.62% optimize.opt_a.with_stream_mark : 0.000052s : 0.03% optimize.opt_a.recompute_prepare : 0.000032s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000014s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000019s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000015s : 0.01% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000277s : 0.16% optimize.opt_a.accelerated_algorithm : 0.000022s : 0.01% optimize.opt_a.shard : 0.000006s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000006s : 0.00% optimize.opt_a.shard_inline : 0.000025s : 0.01% optimize.opt_a.merge_send_recv : 0.000026s : 0.02% optimize.opt_a.auto_parallel : 0.000028s : 0.02% optimize.opt_a.parallel : 0.000064s : 0.04% optimize.opt_a.flash_sp : 0.000019s : 0.01% optimize.opt_a.merge_comm : 0.000017s : 0.01% optimize.opt_a.allreduce_fusion : 0.000015s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000033s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000034s : 0.02% optimize.opt_a.virtual_dataset : 0.000025s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000028s : 0.02% optimize.opt_a.virtual_output : 0.000023s : 0.01% optimize.opt_a.merge_forward : 0.000020s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.00% optimize.opt_a.offload_activation : 0.000038s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000063s : 0.04% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000057s : 0.03% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000014s : 0.01% optimize.opt_a.meta_fg_expand : 0.000009s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.00% optimize.opt_a.receive_attached : 0.000006s : 0.00% optimize.opt_a.after_resolve : 0.000035s : 0.02% optimize.opt_a.a_after_grad : 0.000044s : 0.03% optimize.opt_a.renormalize : 0.001358s : 0.79% optimize.opt_a.add_forward_monad_depend : 0.000021s : 0.01% optimize.opt_a.auto_monad_grad : 0.000006s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000078s : 0.05% optimize.opt_a.cse : 0.000086s : 0.05% optimize.opt_a.a_3 : 0.000140s : 0.08% optimize.py_interpret_to_execute_after_opt_a : 0.000029s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000272s : 0.16% optimize.convert_after_rewriter : 0.000018s : 0.01% optimize.order_py_execute_after_rewriter : 0.000014s : 0.01% optimize.mutable_eliminate : 0.001022s : 0.59% optimize.opt_b.b_1 : 0.000340s : 0.20% optimize.opt_b.b_2 : 0.000021s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000017s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000061s : 0.04% optimize.optimize_parallel_all_gather_comm : 0.000047s : 0.03% optimize.overlap_param_gather : 0.000006s : 0.00% optimize.cconv : 0.000045s : 0.03% optimize.loop_unroll : 0.007236s : 4.20% optimize.opt_after_cconv.c_1 : 0.000101s : 0.06% optimize.opt_after_cconv.parameter_eliminate : 0.000008s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000013s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000009s : 0.01% optimize.opt_after_cconv.cse : 0.000056s : 0.03% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000025s : 0.01% optimize.tuple_transform.d_1 : 0.000107s : 0.06% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000011s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000129s : 0.07% optimize.cse_after_recomputation.cse : 0.000021s : 0.01% optimize.environ_conv : 0.000011s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000013s : 0.01% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000018s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000004s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000007s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000002s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000052s : 0.03% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000006s : 0.00% optimize.overlap_grad_flash_sp : 0.000047s : 0.03% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000011s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000053s : 0.03% optimize.symbol_engine_optimizer.elim_not_effective : 0.000040s : 0.02% optimize.symbol_engine_optimizer.opt_reshape : 0.000017s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000018s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000001s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000045s : 0.03% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000009s : 0.01% opt_after_jit_grad : 0.001007s : 0.58% validate : 0.000077s : 0.04% backend_pass : 0.000007s : 0.00% task_emit : 0.063781s : 37.01% execute : 0.000011s : 0.01% Time group info: ------[substitution.] 0.000269 55 4.16% : 0.000011s : 2: substitution.depend_value_elim 1.33% : 0.000004s : 5: substitution.elim_not_effective 0.90% : 0.000002s : 5: substitution.fold_const_symbol 3.07% : 0.000008s : 6: substitution.graph_param_transform 63.72% : 0.000172s : 1: substitution.inline 2.47% : 0.000007s : 10: substitution.j_node_and_user_rematch 3.45% : 0.000009s : 10: substitution.remove_not_recompute_node 2.84% : 0.000008s : 2: substitution.replace_old_param 9.90% : 0.000027s : 6: substitution.updatestate_pure_node_eliminater 8.15% : 0.000022s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.090567 2 98.17% : 0.088905s : 1: type_inference.infer 1.83% : 0.001661s : 1: type_inference.specialize ------[replace.] 0.000052 1 100.00% : 0.000052s : 1: replace.inline ------[match.] 0.000170 1 100.00% : 0.000170s : 1: match.inline ------[predicate.] 0.000273 1419 0.79% : 0.000002s : 13: predicate.accumulaten_eliminater 1.44% : 0.000004s : 6: predicate.ad_related_special_op_eliminate 0.66% : 0.000002s : 12: predicate.addn_check_dump 0.86% : 0.000002s : 13: predicate.addn_zero_filter 0.72% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 1.96% : 0.000005s : 25: predicate.arithmetic_simplify 0.82% : 0.000002s : 13: predicate.cast_eliminate 0.77% : 0.000002s : 12: predicate.check_bprop_eliminate 0.64% : 0.000002s : 12: predicate.compare_switch_simplify 0.19% : 0.000001s : 6: predicate.const_output_eliminate 0.79% : 0.000002s : 12: predicate.depend_value_elim 0.78% : 0.000002s : 13: predicate.dict_get_item_const_eliminator 0.82% : 0.000002s : 13: predicate.dict_get_item_eliminator 0.81% : 0.000002s : 13: predicate.dict_set_item_eliminator 1.79% : 0.000005s : 12: predicate.dumpgradient_eliminate 0.44% : 0.000001s : 6: predicate.elim_not_effective 1.02% : 0.000003s : 6: predicate.elim_shapecalc_of_broadcastargs 1.24% : 0.000003s : 19: predicate.environ_add_const_eliminate 1.06% : 0.000003s : 19: predicate.environ_get_add_eliminate 1.02% : 0.000003s : 19: predicate.environ_get_depend_swap 1.90% : 0.000005s : 31: predicate.environ_get_eliminate 0.98% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.75% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.46% : 0.000004s : 14: predicate.float_depend_g_call 0.68% : 0.000002s : 12: predicate.float_environ_get_switch 1.00% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.17% : 0.000000s : 6: predicate.fold_const_symbol 0.80% : 0.000002s : 12: predicate.get_grad_eliminate 0.29% : 0.000001s : 6: predicate.graph_param_transform 0.64% : 0.000002s : 12: predicate.incorporate_call 0.56% : 0.000002s : 12: predicate.incorporate_call_switch 5.87% : 0.000016s : 63: predicate.inline 1.11% : 0.000003s : 12: predicate.inline_without_move 0.35% : 0.000001s : 12: predicate.j_node_and_user_rematch 1.10% : 0.000003s : 12: predicate.less_batch_normalization 1.64% : 0.000004s : 25: predicate.list_to_tuple_eliminator_ 2.13% : 0.000006s : 38: predicate.load_eliminater 2.56% : 0.000007s : 6: predicate.loop_unroll_after_grad 1.07% : 0.000003s : 17: predicate.loop_unroll_before_grad 1.91% : 0.000005s : 25: predicate.make_slice_get_slice_eliminator 0.69% : 0.000002s : 12: predicate.merge_addn 0.70% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.70% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.77% : 0.000002s : 13: predicate.minmaximum_grad 1.53% : 0.000004s : 6: predicate.mutable_eliminate 0.44% : 0.000001s : 6: predicate.opt_reshape 0.45% : 0.000001s : 6: predicate.parallel_virtual_node 1.01% : 0.000003s : 14: predicate.partial_defer_inline 1.13% : 0.000003s : 19: predicate.partial_eliminate 0.75% : 0.000002s : 13: predicate.print_const_string_wrapper 0.62% : 0.000002s : 12: predicate.reduce_all_const_elim 1.03% : 0.000003s : 13: predicate.reduce_eliminate 2.20% : 0.000006s : 38: predicate.redundant_stop_gradient_eliminater 0.56% : 0.000002s : 12: predicate.remove_not_recompute_node 1.07% : 0.000003s : 25: predicate.replace_applicator 0.71% : 0.000002s : 12: predicate.replace_old_param 0.36% : 0.000001s : 6: predicate.reset_defer_inline 0.78% : 0.000002s : 13: predicate.reshape_eliminate 0.78% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.48% : 0.000001s : 6: predicate.row_tensor_eliminate 1.02% : 0.000003s : 12: predicate.same_eliminate 0.63% : 0.000002s : 12: predicate.set_cell_output_no_recompute 1.47% : 0.000004s : 12: predicate.shard_identity_eliminate 0.89% : 0.000002s : 12: predicate.special_op_eliminate 0.74% : 0.000002s : 12: predicate.specialize_transform 1.39% : 0.000004s : 12: predicate.split_environ_get_set_with_tuple_value 1.13% : 0.000003s : 12: predicate.stack_unstack_eliminate 0.45% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.83% : 0.000002s : 14: predicate.switch_defer_inline 1.48% : 0.000004s : 26: predicate.switch_layer_defer_inline 3.15% : 0.000009s : 49: predicate.switch_simplify 0.80% : 0.000002s : 13: predicate.tile_eliminate 0.87% : 0.000002s : 13: predicate.transpose_eliminate 1.65% : 0.000004s : 25: predicate.tuple_list_convert_item_index_to_positive 1.62% : 0.000004s : 25: predicate.tuple_list_get_item_const_eliminator 1.69% : 0.000005s : 25: predicate.tuple_list_get_item_depend_reorder 2.97% : 0.000008s : 37: predicate.tuple_list_get_item_eliminator 1.64% : 0.000004s : 25: predicate.tuple_list_get_set_item_eliminator 2.72% : 0.000007s : 37: predicate.tuple_list_set_item_eliminator 1.73% : 0.000005s : 25: predicate.tuple_to_list_eliminator_ 2.26% : 0.000006s : 38: predicate.updatestate_pure_node_eliminater 3.49% : 0.000010s : 50: predicate.updatestate_useless_node_eliminater 0.45% : 0.000001s : 6: predicate.value_based_eliminate 0.80% : 0.000002s : 12: predicate.virtual_dataset_eliminate 0.94% : 0.000003s : 12: predicate.virtual_output_eliminate 0.33% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.52% : 0.000001s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000979 4 2.70% : 0.000026s : 1: func_graph_cloner_run.FuncGraphClonerGraph 97.30% : 0.000952s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.226910 196 0.00% : 0.000008s : 1: ForceFp32Comm 3.87% : 0.008791s : 1: add_attr 3.86% : 0.008764s : 1: add_attr_with_inline 0.00% : 0.000005s : 1: add_comm_op_reuse_tag 0.06% : 0.000139s : 1: add_recomputation 0.00% : 0.000006s : 1: assign_add_opt 0.58% : 0.001309s : 1: auto_monad 0.02% : 0.000050s : 1: auto_monad_reorder 0.02% : 0.000054s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 1.15% : 0.002601s : 1: bias_add_comm_swap 0.57% : 0.001305s : 1: bootstrap 0.02% : 0.000051s : 1: cconv 0.00% : 0.000006s : 1: comm_op_add_attrs 0.03% : 0.000060s : 1: control_data_broadcast_order 0.01% : 0.000028s : 1: convert_after_rewriter 0.02% : 0.000039s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.01% : 0.000016s : 1: environ_conv 0.02% : 0.000050s : 1: event_method 0.01% : 0.000019s : 1: execute 0.00% : 0.000011s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.01% : 0.000027s : 1: graph_reusing 0.01% : 0.000014s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.01% : 0.000013s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000008s : 1: interleave_split_concat_branches 0.01% : 0.000032s : 1: label_fine_grained_interleaved_index 0.02% : 0.000047s : 1: label_micro_interleaved_index 3.20% : 0.007257s : 1: loop_unroll 0.00% : 0.000009s : 1: merge_cast_opt 0.00% : 0.000011s : 1: micro_interleaved_order_control 0.46% : 0.001046s : 1: mutable_eliminate 0.01% : 0.000017s : 1: offloading_packed_experts 0.02% : 0.000042s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000028s : 1: opt.transform.mutable_eliminate 0.80% : 0.001817s : 78: opt.transform.opt_a 0.04% : 0.000099s : 1: opt.transform.opt_after_cconv 0.03% : 0.000064s : 1: opt.transform.opt_after_jit_grad 0.14% : 0.000312s : 28: opt.transform.opt_b 0.05% : 0.000116s : 2: opt.transform.opt_trans_graph 0.05% : 0.000118s : 4: opt.transform.symbol_engine_opt 2.30% : 0.005216s : 1: opt_a 0.12% : 0.000282s : 1: opt_after_cconv 0.45% : 0.001029s : 1: opt_after_jit_grad 0.24% : 0.000552s : 1: opt_b 8.34% : 0.018922s : 1: optimize 0.02% : 0.000055s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000019s : 1: order_py_execute_after_rewriter 0.03% : 0.000058s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000013s : 1: overlap_grad_ring_attention 0.00% : 0.000009s : 1: overlap_opt_shard_grad_in_pipeline 0.01% : 0.000012s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000010s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000011s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000007s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.02% : 0.000049s : 1: pre_auto_parallel 0.03% : 0.000076s : 1: py_interpret_to_execute 0.02% : 0.000037s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.01% : 0.000029s : 1: remove_dup_value 0.32% : 0.000715s : 1: renormalize.infer 0.28% : 0.000628s : 1: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000013s : 1: rewriter_after_jit_bprop_graph 0.12% : 0.000283s : 1: rewriter_after_opt_a 0.04% : 0.000092s : 1: rewriter_before_opt_a 0.00% : 0.000009s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000010s : 1: split_layernorm_comm 0.01% : 0.000013s : 1: split_matmul_comm_elemetwise 0.01% : 0.000021s : 1: swap_dp_allreduce_reducescatter 0.18% : 0.000397s : 1: symbol_engine_optimizer 28.12% : 0.063805s : 1: task_emit 0.07% : 0.000161s : 1: tuple_transform 39.98% : 0.090711s : 1: type_inference 4.05% : 0.009182s : 1: validate TotalTime = 0.200251, [24] [bootstrap]: 0.0004556 [type_inference]: 0.00390488 [event_method]: 1.554e-05 [auto_monad]: 0.00014948 [graph_reusing]: 6.29001e-06 [inline]: 2.67001e-06 [add_attr]: 0.0050666, [1] [add_attr_with_inline]: 0.00505542, [1] [Cycle 1]: 5.234e-05, [2] [tag_attr]: 1.676e-05 [meta_addattr_fg_expand]: 3.12002e-06 [parallel-infer-symbol]: 3.51999e-06 [pre_auto_parallel]: 2.886e-05 [insert-virtual-dataset]: 3.17002e-06 [parallel-infer-symbol-second]: 7.2e-07 [dataset_repeat_opt]: 2.16e-06 [pipeline_split]: 1.94e-06 [optimize]: 0.177192, [53] [py_interpret_to_execute]: 1.935e-05 [rewriter_before_opt_a]: 5.091e-05 [opt_a]: 0.00372309, [2] [Cycle 1]: 0.00264815, [45] [expand_dump_flag]: 3.15998e-06 [switch_simplify]: 2.24e-05 [loop_unroll]: 1.497e-05 [a_1]: 0.00052188 [with_stream_mark]: 2.181e-05 [recompute_prepare]: 1.253e-05 [updatestate_depend_eliminate]: 6.92002e-06 [updatestate_assign_eliminate]: 5.97999e-06 [updatestate_loads_eliminate]: 5.84e-06 [parameter_eliminate]: 1.70001e-06 [a_2]: 0.00015992 [accelerated_algorithm]: 1.098e-05 [shard]: 2.93e-06 [meta_shard_fg_expand]: 2.29999e-06 [shard_inline]: 9.84999e-06 [merge_send_recv]: 1.167e-05 [auto_parallel]: 1.071e-05 [parallel]: 2.102e-05 [flash_sp]: 1.024e-05 [merge_comm]: 6.30002e-06 [allreduce_fusion]: 5.635e-05 [matmul_add_comm_reduction]: 1.538e-05 [allreduce_slice_to_reducescatter]: 6.50005e-07 [virtual_shard_identity]: 1.215e-05 [virtual_dataset]: 1.085e-05 [get_grad_eliminate_]: 1.046e-05 [virtual_output]: 1.085e-05 [merge_forward]: 6.26e-06 [cell_reuse_recompute_pass]: 1.49e-06 [offload_activation]: 1.661e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.264e-05 [merge_recompute_call_nodes]: 1.57999e-06 [before_grad]: 1.852e-05 [set_forward_comm_id_for_comm_node_pass]: 6.26998e-06 [meta_fg_expand]: 5.16002e-06 [flash_sp_send_recv_attached]: 3.06001e-06 [receive_attached]: 3.00998e-06 [after_resolve]: 1.341e-05 [a_after_grad]: 2.013e-05 [renormalize]: 0.00105962 [add_forward_monad_depend]: 7.48e-06 [auto_monad_grad]: 2.74001e-06 [auto_monad_eliminator]: 2.279e-05 [cse]: 5.943e-05 [a_3]: 7.123e-05 [Cycle 2]: 0.00106137, [45] [expand_dump_flag]: 3.27002e-06 [switch_simplify]: 1.05e-05 [loop_unroll]: 9.06002e-06 [a_1]: 0.00021907 [with_stream_mark]: 2.132e-05 [recompute_prepare]: 9.84001e-06 [updatestate_depend_eliminate]: 6.19999e-06 [updatestate_assign_eliminate]: 5.25999e-06 [updatestate_loads_eliminate]: 5.52001e-06 [parameter_eliminate]: 1.30999e-06 [a_2]: 0.00011732 [accelerated_algorithm]: 9.70002e-06 [shard]: 2.24999e-06 [meta_shard_fg_expand]: 2.15002e-06 [shard_inline]: 8.87999e-06 [merge_send_recv]: 9.96e-06 [auto_parallel]: 1.042e-05 [parallel]: 7.86001e-06 [flash_sp]: 3.80998e-06 [merge_comm]: 5.61003e-06 [allreduce_fusion]: 5.40999e-06 [matmul_add_comm_reduction]: 1.232e-05 [allreduce_slice_to_reducescatter]: 6.90023e-07 [virtual_shard_identity]: 1.054e-05 [virtual_dataset]: 9.54e-06 [get_grad_eliminate_]: 8.74998e-06 [virtual_output]: 8.43999e-06 [merge_forward]: 5.82999e-06 [cell_reuse_recompute_pass]: 2.69999e-06 [offload_activation]: 1.358e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.976e-05 [merge_recompute_call_nodes]: 1.17999e-06 [before_grad]: 1.77e-05 [set_forward_comm_id_for_comm_node_pass]: 6.49999e-06 [meta_fg_expand]: 4.05e-06 [flash_sp_send_recv_attached]: 1.59998e-06 [receive_attached]: 2.39001e-06 [after_resolve]: 1.321e-05 [a_after_grad]: 1.362e-05 [renormalize]: 1.90019e-07 [add_forward_monad_depend]: 1.74e-06 [auto_monad_grad]: 2.04e-06 [auto_monad_eliminator]: 1.793e-05 [cse]: 5.716e-05 [a_3]: 5.961e-05 [py_interpret_to_execute_after_opt_a]: 1.745e-05 [slice_cell_reuse_recomputed_activation]: 2.24999e-06 [rewriter_after_opt_a]: 0.00022127 [convert_after_rewriter]: 1.348e-05 [order_py_execute_after_rewriter]: 7.4e-06 [mutable_eliminate]: 0.170681 [opt_b]: 0.00045558, [1] [Cycle 1]: 0.00044137, [7] [b_1]: 0.00027784 [b_2]: 1.458e-05 [updatestate_depend_eliminate]: 1.86e-05 [updatestate_assign_eliminate]: 6.53e-06 [updatestate_loads_eliminate]: 6.44999e-06 [renormalize]: 9.39996e-07 [cse]: 5.929e-05 [optimize_parallel_all_gather_comm]: 3.406e-05 [overlap_param_gather]: 2.44001e-06 [cconv]: 4.432e-05 [loop_unroll]: 0.00079679 [opt_after_cconv]: 0.00019498, [1] [Cycle 1]: 0.00018607, [7] [c_1]: 7.117e-05 [parameter_eliminate]: 6.12999e-06 [updatestate_depend_eliminate]: 1.001e-05 [updatestate_assign_eliminate]: 5.86e-06 [updatestate_loads_eliminate]: 5.06997e-06 [cse]: 3.885e-05 [renormalize]: 4.89992e-07 [remove_dup_value]: 2.158e-05 [tuple_transform]: 0.00013282, [1] [Cycle 1]: 0.00012568, [4] [d_1]: 8.469e-05 [none_parameter_eliminate]: 2.48002e-06 [renormalize]: 2.30008e-07 [switch_simplify]: 1.102e-05 [partial_unused_args_eliminate]: 2.12001e-06 [add_recomputation]: 9.457e-05 [cse_after_recomputation]: 3.328e-05, [1] [Cycle 1]: 2.712e-05, [1] [cse]: 1.902e-05 [environ_conv]: 1.058e-05 [swap_dp_allreduce_reducescatter]: 8.69e-06 [bias_add_comm_swap]: 3.43999e-06 [label_micro_interleaved_index]: 7.19001e-06 [label_fine_grained_interleaved_index]: 3.7e-06 [merge_cast_opt]: 1.71e-06 [slice_recompute_activation]: 2.27999e-06 [micro_interleaved_order_control]: 3.52002e-06 [assign_add_opt]: 1.62001e-06 [ForceFp32Comm]: 9.89996e-07 [remove_cast_before_assign_add]: 9.39996e-07 [full_micro_interleaved_order_control]: 2.78998e-06 [reorder_send_recv_between_fp_bp]: 3.31999e-06 [comm_op_add_attrs]: 1.49e-06 [add_comm_op_reuse_tag]: 1.08001e-06 [interleave_split_concat_branches]: 1.35999e-06 [interleave_parallel_branches]: 1.33002e-06 [overlap_opt_shard_in_pipeline]: 1.20999e-06 [overlap_opt_shard_grad_in_pipeline]: 1.85001e-06 [control_data_broadcast_order]: 3.735e-05 [grouped_pairwise_exchange_alltoall]: 1.91e-06 [offloading_packed_experts]: 7.46001e-06 [overlap_recompute_and_grad_model_parallel]: 1.574e-05 [overlap_grad_matmul_and_grad_allreduce]: 1.49e-06 [overlap_recompute_allgather_and_fa_grad]: 1.70001e-06 [overlap_recompute_comm]: 2.84999e-06 [overlap_grad_ring_attention]: 5.77999e-06 [overlap_grad_flash_sp]: 3.218e-05 [begin_end_overlap_inline]: 7.89994e-07 [split_matmul_comm_elemetwise]: 2.53998e-06 [split_layernorm_comm]: 2.29999e-06 [handle_group_info]: 1.37999e-06 [symbol_engine_optimizer]: 0.00013042, [1] [Cycle 1]: 0.00012391, [6] [build]: 6.74001e-06 [elim_shapecalc]: 1.901e-05 [elim_not_effective]: 2.149e-05 [opt_reshape]: 1.129e-05 [fold_const_symbol]: 1.733e-05 [renormalize]: 2.19996e-07 [detach_backward]: 2.28998e-06 [pipeline_parallel_scheduler]: 1.55001e-06 [auto_monad_reorder]: 3.584e-05 [get_jit_bprop_graph]: 2.01998e-06 [rewriter_after_jit_bprop_graph]: 9.25999e-06 [opt_after_jit_grad]: 0.0007207 [validate]: 6.843e-05 [backend_pass]: 1.04998e-06 [task_emit]: 0.0122723 [execute]: 1.273e-05 Sums bootstrap : 0.000456s : 0.24% type_inference : 0.003905s : 2.02% event_method : 0.000016s : 0.01% auto_monad : 0.000149s : 0.08% graph_reusing : 0.000006s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000017s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000029s : 0.01% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000019s : 0.01% optimize.rewriter_before_opt_a : 0.000051s : 0.03% optimize.opt_a.expand_dump_flag : 0.000006s : 0.00% optimize.opt_a.switch_simplify : 0.000033s : 0.02% optimize.opt_a.loop_unroll : 0.000024s : 0.01% optimize.opt_a.a_1 : 0.000741s : 0.38% optimize.opt_a.with_stream_mark : 0.000043s : 0.02% optimize.opt_a.recompute_prepare : 0.000022s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000013s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000011s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000011s : 0.01% optimize.opt_a.parameter_eliminate : 0.000003s : 0.00% optimize.opt_a.a_2 : 0.000277s : 0.14% optimize.opt_a.accelerated_algorithm : 0.000021s : 0.01% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.00% optimize.opt_a.shard_inline : 0.000019s : 0.01% optimize.opt_a.merge_send_recv : 0.000022s : 0.01% optimize.opt_a.auto_parallel : 0.000021s : 0.01% optimize.opt_a.parallel : 0.000029s : 0.01% optimize.opt_a.flash_sp : 0.000014s : 0.01% optimize.opt_a.merge_comm : 0.000012s : 0.01% optimize.opt_a.allreduce_fusion : 0.000062s : 0.03% optimize.opt_a.matmul_add_comm_reduction : 0.000028s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000023s : 0.01% optimize.opt_a.virtual_dataset : 0.000020s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000019s : 0.01% optimize.opt_a.virtual_output : 0.000019s : 0.01% optimize.opt_a.merge_forward : 0.000012s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.00% optimize.opt_a.offload_activation : 0.000030s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000042s : 0.02% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000036s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000013s : 0.01% optimize.opt_a.meta_fg_expand : 0.000009s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000027s : 0.01% optimize.opt_a.a_after_grad : 0.000034s : 0.02% optimize.opt_a.renormalize : 0.001060s : 0.55% optimize.opt_a.add_forward_monad_depend : 0.000009s : 0.00% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000041s : 0.02% optimize.opt_a.cse : 0.000117s : 0.06% optimize.opt_a.a_3 : 0.000131s : 0.07% optimize.py_interpret_to_execute_after_opt_a : 0.000017s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000221s : 0.11% optimize.convert_after_rewriter : 0.000013s : 0.01% optimize.order_py_execute_after_rewriter : 0.000007s : 0.00% optimize.mutable_eliminate : 0.170681s : 88.11% optimize.opt_b.b_1 : 0.000278s : 0.14% optimize.opt_b.b_2 : 0.000015s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000019s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000007s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000059s : 0.03% optimize.optimize_parallel_all_gather_comm : 0.000034s : 0.02% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000044s : 0.02% optimize.loop_unroll : 0.000797s : 0.41% optimize.opt_after_cconv.c_1 : 0.000071s : 0.04% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000010s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.cse : 0.000039s : 0.02% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000022s : 0.01% optimize.tuple_transform.d_1 : 0.000085s : 0.04% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000011s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000095s : 0.05% optimize.cse_after_recomputation.cse : 0.000019s : 0.01% optimize.environ_conv : 0.000011s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000009s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000007s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000004s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000004s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000037s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000016s : 0.01% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000006s : 0.00% optimize.overlap_grad_flash_sp : 0.000032s : 0.02% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000007s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000019s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000021s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000011s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000017s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000036s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000009s : 0.00% opt_after_jit_grad : 0.000721s : 0.37% validate : 0.000068s : 0.04% backend_pass : 0.000001s : 0.00% task_emit : 0.012272s : 6.34% execute : 0.000013s : 0.01% Time group info: ------[substitution.] 0.000194 55 5.13% : 0.000010s : 2: substitution.depend_value_elim 1.53% : 0.000003s : 5: substitution.elim_not_effective 1.31% : 0.000003s : 5: substitution.fold_const_symbol 4.79% : 0.000009s : 6: substitution.graph_param_transform 56.23% : 0.000109s : 1: substitution.inline 2.94% : 0.000006s : 10: substitution.j_node_and_user_rematch 4.41% : 0.000009s : 10: substitution.remove_not_recompute_node 2.39% : 0.000005s : 2: substitution.replace_old_param 12.33% : 0.000024s : 6: substitution.updatestate_pure_node_eliminater 8.93% : 0.000017s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.003844 2 87.33% : 0.003357s : 1: type_inference.infer 12.67% : 0.000487s : 1: type_inference.specialize ------[replace.] 0.000018 1 100.00% : 0.000018s : 1: replace.inline ------[match.] 0.000108 1 100.00% : 0.000108s : 1: match.inline ------[predicate.] 0.000281 1419 0.74% : 0.000002s : 13: predicate.accumulaten_eliminater 0.98% : 0.000003s : 6: predicate.ad_related_special_op_eliminate 0.60% : 0.000002s : 12: predicate.addn_check_dump 0.82% : 0.000002s : 13: predicate.addn_zero_filter 0.64% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 1.98% : 0.000006s : 25: predicate.arithmetic_simplify 0.87% : 0.000002s : 13: predicate.cast_eliminate 0.70% : 0.000002s : 12: predicate.check_bprop_eliminate 0.59% : 0.000002s : 12: predicate.compare_switch_simplify 0.23% : 0.000001s : 6: predicate.const_output_eliminate 0.68% : 0.000002s : 12: predicate.depend_value_elim 0.79% : 0.000002s : 13: predicate.dict_get_item_const_eliminator 0.81% : 0.000002s : 13: predicate.dict_get_item_eliminator 0.78% : 0.000002s : 13: predicate.dict_set_item_eliminator 1.14% : 0.000003s : 12: predicate.dumpgradient_eliminate 0.30% : 0.000001s : 6: predicate.elim_not_effective 0.58% : 0.000002s : 6: predicate.elim_shapecalc_of_broadcastargs 1.07% : 0.000003s : 19: predicate.environ_add_const_eliminate 1.03% : 0.000003s : 19: predicate.environ_get_add_eliminate 1.07% : 0.000003s : 19: predicate.environ_get_depend_swap 1.69% : 0.000005s : 31: predicate.environ_get_eliminate 0.99% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.73% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.11% : 0.000003s : 14: predicate.float_depend_g_call 0.62% : 0.000002s : 12: predicate.float_environ_get_switch 1.00% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.21% : 0.000001s : 6: predicate.fold_const_symbol 0.75% : 0.000002s : 12: predicate.get_grad_eliminate 0.30% : 0.000001s : 6: predicate.graph_param_transform 7.43% : 0.000021s : 12: predicate.incorporate_call 0.55% : 0.000002s : 12: predicate.incorporate_call_switch 4.96% : 0.000014s : 63: predicate.inline 1.00% : 0.000003s : 12: predicate.inline_without_move 0.32% : 0.000001s : 12: predicate.j_node_and_user_rematch 0.96% : 0.000003s : 12: predicate.less_batch_normalization 1.47% : 0.000004s : 25: predicate.list_to_tuple_eliminator_ 1.93% : 0.000005s : 38: predicate.load_eliminater 1.66% : 0.000005s : 6: predicate.loop_unroll_after_grad 0.98% : 0.000003s : 17: predicate.loop_unroll_before_grad 1.71% : 0.000005s : 25: predicate.make_slice_get_slice_eliminator 0.60% : 0.000002s : 12: predicate.merge_addn 0.72% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.64% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.71% : 0.000002s : 13: predicate.minmaximum_grad 3.38% : 0.000009s : 6: predicate.mutable_eliminate 0.43% : 0.000001s : 6: predicate.opt_reshape 0.52% : 0.000001s : 6: predicate.parallel_virtual_node 0.94% : 0.000003s : 14: predicate.partial_defer_inline 1.09% : 0.000003s : 19: predicate.partial_eliminate 0.79% : 0.000002s : 13: predicate.print_const_string_wrapper 0.74% : 0.000002s : 12: predicate.reduce_all_const_elim 1.08% : 0.000003s : 13: predicate.reduce_eliminate 2.31% : 0.000006s : 38: predicate.redundant_stop_gradient_eliminater 0.43% : 0.000001s : 12: predicate.remove_not_recompute_node 0.94% : 0.000003s : 25: predicate.replace_applicator 0.42% : 0.000001s : 12: predicate.replace_old_param 0.52% : 0.000001s : 6: predicate.reset_defer_inline 0.71% : 0.000002s : 13: predicate.reshape_eliminate 0.68% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.49% : 0.000001s : 6: predicate.row_tensor_eliminate 1.08% : 0.000003s : 12: predicate.same_eliminate 0.44% : 0.000001s : 12: predicate.set_cell_output_no_recompute 0.90% : 0.000003s : 12: predicate.shard_identity_eliminate 0.85% : 0.000002s : 12: predicate.special_op_eliminate 0.73% : 0.000002s : 12: predicate.specialize_transform 1.26% : 0.000004s : 12: predicate.split_environ_get_set_with_tuple_value 0.88% : 0.000002s : 12: predicate.stack_unstack_eliminate 0.43% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.78% : 0.000002s : 14: predicate.switch_defer_inline 1.36% : 0.000004s : 26: predicate.switch_layer_defer_inline 3.09% : 0.000009s : 49: predicate.switch_simplify 0.76% : 0.000002s : 13: predicate.tile_eliminate 0.79% : 0.000002s : 13: predicate.transpose_eliminate 1.56% : 0.000004s : 25: predicate.tuple_list_convert_item_index_to_positive 2.01% : 0.000006s : 25: predicate.tuple_list_get_item_const_eliminator 1.54% : 0.000004s : 25: predicate.tuple_list_get_item_depend_reorder 3.00% : 0.000008s : 37: predicate.tuple_list_get_item_eliminator 1.60% : 0.000005s : 25: predicate.tuple_list_get_set_item_eliminator 2.60% : 0.000007s : 37: predicate.tuple_list_set_item_eliminator 1.68% : 0.000005s : 25: predicate.tuple_to_list_eliminator_ 2.01% : 0.000006s : 38: predicate.updatestate_pure_node_eliminater 3.03% : 0.000009s : 50: predicate.updatestate_useless_node_eliminater 0.52% : 0.000001s : 6: predicate.value_based_eliminate 0.76% : 0.000002s : 12: predicate.virtual_dataset_eliminate 0.73% : 0.000002s : 12: predicate.virtual_output_eliminate 0.29% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.48% : 0.000001s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000413 4 6.77% : 0.000028s : 1: func_graph_cloner_run.FuncGraphClonerGraph 93.23% : 0.000385s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.385430 196 0.00% : 0.000004s : 1: ForceFp32Comm 1.32% : 0.005074s : 1: add_attr 1.31% : 0.005060s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.03% : 0.000101s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.04% : 0.000159s : 1: auto_monad 0.01% : 0.000041s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000005s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.13% : 0.000497s : 1: bootstrap 0.01% : 0.000049s : 1: cconv 0.00% : 0.000006s : 1: comm_op_add_attrs 0.01% : 0.000044s : 1: control_data_broadcast_order 0.01% : 0.000020s : 1: convert_after_rewriter 0.01% : 0.000037s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000014s : 1: environ_conv 0.01% : 0.000022s : 1: event_method 0.01% : 0.000020s : 1: execute 0.00% : 0.000007s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000011s : 1: graph_reusing 0.00% : 0.000006s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000005s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000008s : 1: label_fine_grained_interleaved_index 0.00% : 0.000011s : 1: label_micro_interleaved_index 0.21% : 0.000810s : 1: loop_unroll 0.00% : 0.000007s : 1: merge_cast_opt 0.00% : 0.000007s : 1: micro_interleaved_order_control 44.29% : 0.170716s : 1: mutable_eliminate 0.00% : 0.000012s : 1: offloading_packed_experts 0.01% : 0.000026s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000050s : 1: opt.transform.mutable_eliminate 0.36% : 0.001391s : 78: opt.transform.opt_a 0.02% : 0.000069s : 1: opt.transform.opt_after_cconv 0.01% : 0.000049s : 1: opt.transform.opt_after_jit_grad 0.06% : 0.000248s : 28: opt.transform.opt_b 0.02% : 0.000093s : 2: opt.transform.opt_trans_graph 0.02% : 0.000066s : 4: opt.transform.symbol_engine_opt 0.97% : 0.003728s : 1: opt_a 0.05% : 0.000200s : 1: opt_after_cconv 0.19% : 0.000735s : 1: opt_after_jit_grad 0.12% : 0.000461s : 1: opt_b 45.97% : 0.177199s : 1: optimize 0.01% : 0.000039s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000012s : 1: order_py_execute_after_rewriter 0.01% : 0.000037s : 1: overlap_grad_flash_sp 0.00% : 0.000006s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000009s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000007s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000021s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000033s : 1: pre_auto_parallel 0.01% : 0.000024s : 1: py_interpret_to_execute 0.01% : 0.000022s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.01% : 0.000026s : 1: remove_dup_value 0.15% : 0.000575s : 1: renormalize.infer 0.12% : 0.000474s : 1: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000013s : 1: rewriter_after_jit_bprop_graph 0.06% : 0.000230s : 1: rewriter_after_opt_a 0.01% : 0.000055s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000006s : 1: split_layernorm_comm 0.00% : 0.000007s : 1: split_matmul_comm_elemetwise 0.00% : 0.000013s : 1: swap_dp_allreduce_reducescatter 0.03% : 0.000134s : 1: symbol_engine_optimizer 3.19% : 0.012298s : 1: task_emit 0.04% : 0.000136s : 1: tuple_transform 1.02% : 0.003927s : 1: type_inference 0.03% : 0.000120s : 1: validate TotalTime = 0.780755, [24] [bootstrap]: 0.00053702 [type_inference]: 0.204912 [event_method]: 7.88e-05 [auto_monad]: 0.00030802 [graph_reusing]: 1.225e-05 [inline]: 3.4e-06 [add_attr]: 0.00441999, [1] [add_attr_with_inline]: 0.00440586, [1] [Cycle 1]: 0.0001107, [2] [tag_attr]: 4.997e-05 [meta_addattr_fg_expand]: 1.094e-05 [parallel-infer-symbol]: 3.81999e-06 [pre_auto_parallel]: 6.423e-05 [insert-virtual-dataset]: 2.51e-06 [parallel-infer-symbol-second]: 6.90023e-07 [dataset_repeat_opt]: 2.02999e-06 [pipeline_split]: 1.62001e-06 [optimize]: 0.475882, [53] [py_interpret_to_execute]: 6.876e-05 [rewriter_before_opt_a]: 0.00023922 [opt_a]: 0.341639, [3] [Cycle 1]: 0.332504, [45] [expand_dump_flag]: 9.27999e-06 [switch_simplify]: 8.645e-05 [loop_unroll]: 7.083e-05 [a_1]: 0.165989 [with_stream_mark]: 6.027e-05 [recompute_prepare]: 3.922e-05 [updatestate_depend_eliminate]: 1.55e-05 [updatestate_assign_eliminate]: 1.121e-05 [updatestate_loads_eliminate]: 1.041e-05 [parameter_eliminate]: 3.81999e-06 [a_2]: 0.0003372 [accelerated_algorithm]: 4.263e-05 [shard]: 2.04999e-06 [meta_shard_fg_expand]: 1.144e-05 [shard_inline]: 1.998e-05 [merge_send_recv]: 2.244e-05 [auto_parallel]: 1.873e-05 [parallel]: 2.342e-05 [flash_sp]: 1.687e-05 [merge_comm]: 1.211e-05 [allreduce_fusion]: 1.125e-05 [matmul_add_comm_reduction]: 4.606e-05 [allreduce_slice_to_reducescatter]: 1.43002e-06 [virtual_shard_identity]: 2.673e-05 [virtual_dataset]: 2.6e-05 [get_grad_eliminate_]: 2.25e-05 [virtual_output]: 2.292e-05 [merge_forward]: 1.49e-05 [cell_reuse_recompute_pass]: 1.92999e-06 [offload_activation]: 2.487e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.699e-05 [merge_recompute_call_nodes]: 1.92999e-06 [before_grad]: 4.199e-05 [set_forward_comm_id_for_comm_node_pass]: 1.297e-05 [meta_fg_expand]: 0.00759051 [flash_sp_send_recv_attached]: 5.59998e-06 [receive_attached]: 4e-06 [after_resolve]: 0.00014254 [a_after_grad]: 0.00024169 [renormalize]: 0.154432 [add_forward_monad_depend]: 3.341e-05 [auto_monad_grad]: 2.085e-05 [auto_monad_eliminator]: 0.00019869 [cse]: 0.00036562 [a_3]: 0.00185469 [Cycle 2]: 0.0077394, [45] [expand_dump_flag]: 8.40001e-06 [switch_simplify]: 0.00011191 [loop_unroll]: 0.00010544 [a_1]: 0.00387121 [with_stream_mark]: 4.389e-05 [recompute_prepare]: 2.725e-05 [updatestate_depend_eliminate]: 4.527e-05 [updatestate_assign_eliminate]: 8.34998e-06 [updatestate_loads_eliminate]: 6.07001e-06 [parameter_eliminate]: 2.98e-06 [a_2]: 0.00017184 [accelerated_algorithm]: 2.166e-05 [shard]: 3.03e-06 [meta_shard_fg_expand]: 7.06001e-06 [shard_inline]: 1.25e-05 [merge_send_recv]: 1.406e-05 [auto_parallel]: 1.634e-05 [parallel]: 1.419e-05 [flash_sp]: 5.65001e-06 [merge_comm]: 7.88999e-06 [allreduce_fusion]: 6.58e-06 [matmul_add_comm_reduction]: 1.501e-05 [allreduce_slice_to_reducescatter]: 9.20001e-07 [virtual_shard_identity]: 1.611e-05 [virtual_dataset]: 1.192e-05 [get_grad_eliminate_]: 1.152e-05 [virtual_output]: 1.125e-05 [merge_forward]: 8.09002e-06 [cell_reuse_recompute_pass]: 2.43e-06 [offload_activation]: 1.766e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.892e-05 [merge_recompute_call_nodes]: 1.53002e-06 [before_grad]: 3.81e-05 [set_forward_comm_id_for_comm_node_pass]: 9.37999e-06 [meta_fg_expand]: 0.00012711 [flash_sp_send_recv_attached]: 2.57001e-06 [receive_attached]: 3.05998e-06 [after_resolve]: 2.907e-05 [a_after_grad]: 1.976e-05 [renormalize]: 0.00214505 [add_forward_monad_depend]: 1.236e-05 [auto_monad_grad]: 3.03e-06 [auto_monad_eliminator]: 3.24e-05 [cse]: 0.00017403 [a_3]: 0.00010116 [Cycle 3]: 0.00136635, [45] [expand_dump_flag]: 2.32001e-06 [switch_simplify]: 1.475e-05 [loop_unroll]: 1.253e-05 [a_1]: 0.00032207 [with_stream_mark]: 2.359e-05 [recompute_prepare]: 1.406e-05 [updatestate_depend_eliminate]: 7.4e-06 [updatestate_assign_eliminate]: 6.74001e-06 [updatestate_loads_eliminate]: 7.06999e-06 [parameter_eliminate]: 2.64999e-06 [a_2]: 0.00019735 [accelerated_algorithm]: 2.296e-05 [shard]: 2.36e-06 [meta_shard_fg_expand]: 4.33999e-06 [shard_inline]: 1.31e-05 [merge_send_recv]: 1.476e-05 [auto_parallel]: 1.365e-05 [parallel]: 1.241e-05 [flash_sp]: 2.06e-06 [merge_comm]: 8.54e-06 [allreduce_fusion]: 6.51999e-06 [matmul_add_comm_reduction]: 1.528e-05 [allreduce_slice_to_reducescatter]: 1.45999e-06 [virtual_shard_identity]: 1.598e-05 [virtual_dataset]: 1.398e-05 [get_grad_eliminate_]: 1.189e-05 [virtual_output]: 1.039e-05 [merge_forward]: 6.86999e-06 [cell_reuse_recompute_pass]: 3.71999e-06 [offload_activation]: 1.697e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.736e-05 [merge_recompute_call_nodes]: 1.59e-06 [before_grad]: 2.152e-05 [set_forward_comm_id_for_comm_node_pass]: 9.39998e-06 [meta_fg_expand]: 4.95999e-06 [flash_sp_send_recv_attached]: 1.90001e-06 [receive_attached]: 2.27999e-06 [after_resolve]: 2.22e-05 [a_after_grad]: 1.876e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 4.20999e-06 [auto_monad_grad]: 2.31e-06 [auto_monad_eliminator]: 4.561e-05 [cse]: 4.762e-05 [a_3]: 7.628e-05 [py_interpret_to_execute_after_opt_a]: 2.72e-05 [slice_cell_reuse_recomputed_activation]: 2.46998e-06 [rewriter_after_opt_a]: 8.054e-05 [convert_after_rewriter]: 1.107e-05 [order_py_execute_after_rewriter]: 7.58001e-06 [mutable_eliminate]: 0.00086681 [opt_b]: 0.00046258, [1] [Cycle 1]: 0.00045195, [7] [b_1]: 0.00030608 [b_2]: 1.47e-05 [updatestate_depend_eliminate]: 1.527e-05 [updatestate_assign_eliminate]: 6.41e-06 [updatestate_loads_eliminate]: 6.94999e-06 [renormalize]: 9.99979e-07 [cse]: 5.432e-05 [optimize_parallel_all_gather_comm]: 3.122e-05 [overlap_param_gather]: 2.15002e-06 [cconv]: 4.306e-05 [loop_unroll]: 0.00059312 [opt_after_cconv]: 0.130812, [1] [Cycle 1]: 0.130799, [7] [c_1]: 0.130549 [parameter_eliminate]: 9.15001e-06 [updatestate_depend_eliminate]: 3.505e-05 [updatestate_assign_eliminate]: 7.14001e-06 [updatestate_loads_eliminate]: 7.24001e-06 [cse]: 7.332e-05 [renormalize]: 1.01002e-06 [remove_dup_value]: 2.68e-05 [tuple_transform]: 0.00017555, [1] [Cycle 1]: 0.00016247, [4] [d_1]: 0.00012266 [none_parameter_eliminate]: 2.02001e-06 [renormalize]: 1.8999e-07 [switch_simplify]: 1.361e-05 [partial_unused_args_eliminate]: 2.51998e-06 [add_recomputation]: 0.00012189 [cse_after_recomputation]: 4.48e-05, [1] [Cycle 1]: 3.731e-05, [1] [cse]: 3.132e-05 [environ_conv]: 1.63e-05 [swap_dp_allreduce_reducescatter]: 1.034e-05 [bias_add_comm_swap]: 3.09001e-06 [label_micro_interleaved_index]: 9.50001e-06 [label_fine_grained_interleaved_index]: 2.69999e-06 [merge_cast_opt]: 1.63002e-06 [slice_recompute_activation]: 2.41e-06 [micro_interleaved_order_control]: 2.57001e-06 [assign_add_opt]: 2.04999e-06 [ForceFp32Comm]: 1.10001e-06 [remove_cast_before_assign_add]: 1.12999e-06 [full_micro_interleaved_order_control]: 2.19001e-06 [reorder_send_recv_between_fp_bp]: 3.03e-06 [comm_op_add_attrs]: 1.15999e-06 [add_comm_op_reuse_tag]: 1.40999e-06 [interleave_split_concat_branches]: 1.14e-06 [interleave_parallel_branches]: 1.17e-06 [overlap_opt_shard_in_pipeline]: 1.20999e-06 [overlap_opt_shard_grad_in_pipeline]: 2.14999e-06 [control_data_broadcast_order]: 2.67e-05 [grouped_pairwise_exchange_alltoall]: 1.95001e-06 [offloading_packed_experts]: 7.40003e-06 [overlap_recompute_and_grad_model_parallel]: 6.86001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.63002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.60999e-06 [overlap_recompute_comm]: 2.39001e-06 [overlap_grad_ring_attention]: 7.77998e-06 [overlap_grad_flash_sp]: 3.517e-05 [begin_end_overlap_inline]: 5.10016e-07 [split_matmul_comm_elemetwise]: 2.73e-06 [split_layernorm_comm]: 2.20002e-06 [handle_group_info]: 1.03001e-06 [symbol_engine_optimizer]: 0.00015126, [1] [Cycle 1]: 0.00014513, [6] [build]: 1.572e-05 [elim_shapecalc]: 3.342e-05 [elim_not_effective]: 2.527e-05 [opt_reshape]: 1.504e-05 [fold_const_symbol]: 1.908e-05 [renormalize]: 2.50002e-07 [detach_backward]: 2.48e-06 [pipeline_parallel_scheduler]: 3.04999e-06 [auto_monad_reorder]: 4.088e-05 [get_jit_bprop_graph]: 2.21e-06 [rewriter_after_jit_bprop_graph]: 7.71001e-06 [opt_after_jit_grad]: 0.0008865 [validate]: 7.085e-05 [backend_pass]: 1.34998e-06 [task_emit]: 0.0931374 [execute]: 1.174e-05 Sums bootstrap : 0.000537s : 0.07% type_inference : 0.204912s : 26.47% event_method : 0.000079s : 0.01% auto_monad : 0.000308s : 0.04% graph_reusing : 0.000012s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000050s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000011s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000064s : 0.01% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000069s : 0.01% optimize.rewriter_before_opt_a : 0.000239s : 0.03% optimize.opt_a.expand_dump_flag : 0.000020s : 0.00% optimize.opt_a.switch_simplify : 0.000213s : 0.03% optimize.opt_a.loop_unroll : 0.000189s : 0.02% optimize.opt_a.a_1 : 0.170182s : 21.98% optimize.opt_a.with_stream_mark : 0.000128s : 0.02% optimize.opt_a.recompute_prepare : 0.000081s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000068s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000026s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000024s : 0.00% optimize.opt_a.parameter_eliminate : 0.000009s : 0.00% optimize.opt_a.a_2 : 0.000706s : 0.09% optimize.opt_a.accelerated_algorithm : 0.000087s : 0.01% optimize.opt_a.shard : 0.000007s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000023s : 0.00% optimize.opt_a.shard_inline : 0.000046s : 0.01% optimize.opt_a.merge_send_recv : 0.000051s : 0.01% optimize.opt_a.auto_parallel : 0.000049s : 0.01% optimize.opt_a.parallel : 0.000050s : 0.01% optimize.opt_a.flash_sp : 0.000025s : 0.00% optimize.opt_a.merge_comm : 0.000029s : 0.00% optimize.opt_a.allreduce_fusion : 0.000024s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000076s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000004s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000059s : 0.01% optimize.opt_a.virtual_dataset : 0.000052s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000046s : 0.01% optimize.opt_a.virtual_output : 0.000045s : 0.01% optimize.opt_a.merge_forward : 0.000030s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000008s : 0.00% optimize.opt_a.offload_activation : 0.000059s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000103s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000102s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000032s : 0.00% optimize.opt_a.meta_fg_expand : 0.007723s : 1.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000010s : 0.00% optimize.opt_a.receive_attached : 0.000009s : 0.00% optimize.opt_a.after_resolve : 0.000194s : 0.03% optimize.opt_a.a_after_grad : 0.000280s : 0.04% optimize.opt_a.renormalize : 0.156577s : 20.22% optimize.opt_a.add_forward_monad_depend : 0.000050s : 0.01% optimize.opt_a.auto_monad_grad : 0.000026s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000277s : 0.04% optimize.opt_a.cse : 0.000587s : 0.08% optimize.opt_a.a_3 : 0.002032s : 0.26% optimize.py_interpret_to_execute_after_opt_a : 0.000027s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000081s : 0.01% optimize.convert_after_rewriter : 0.000011s : 0.00% optimize.order_py_execute_after_rewriter : 0.000008s : 0.00% optimize.mutable_eliminate : 0.000867s : 0.11% optimize.opt_b.b_1 : 0.000306s : 0.04% optimize.opt_b.b_2 : 0.000015s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000015s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000007s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000054s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000031s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000043s : 0.01% optimize.loop_unroll : 0.000593s : 0.08% optimize.opt_after_cconv.c_1 : 0.130549s : 16.86% optimize.opt_after_cconv.parameter_eliminate : 0.000009s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000035s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.cse : 0.000073s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000027s : 0.00% optimize.tuple_transform.d_1 : 0.000123s : 0.02% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000014s : 0.00% optimize.partial_unused_args_eliminate : 0.000003s : 0.00% optimize.add_recomputation : 0.000122s : 0.02% optimize.cse_after_recomputation.cse : 0.000031s : 0.00% optimize.environ_conv : 0.000016s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000010s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000010s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000027s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000008s : 0.00% optimize.overlap_grad_flash_sp : 0.000035s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000016s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000033s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000025s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000015s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000019s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000003s : 0.00% auto_monad_reorder : 0.000041s : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.000886s : 0.11% validate : 0.000071s : 0.01% backend_pass : 0.000001s : 0.00% task_emit : 0.093137s : 12.03% execute : 0.000012s : 0.00% Time group info: ------[substitution.] 0.002946 310 0.67% : 0.000020s : 4: substitution.depend_value_elim 0.12% : 0.000004s : 7: substitution.elim_not_effective 0.31% : 0.000009s : 7: substitution.float_depend_g_call 0.17% : 0.000005s : 2: substitution.float_tuple_getitem_switch 0.10% : 0.000003s : 7: substitution.fold_const_symbol 33.99% : 0.001001s : 4: substitution.getattr_setattr_resolve 0.40% : 0.000012s : 9: substitution.graph_param_transform 0.10% : 0.000003s : 2: substitution.incorporate_call 0.07% : 0.000002s : 2: substitution.incorporate_call_switch 42.71% : 0.001258s : 24: substitution.inline 1.51% : 0.000044s : 5: substitution.inline_without_move 0.57% : 0.000017s : 28: substitution.j_node_and_user_rematch 0.82% : 0.000024s : 3: substitution.less_batch_normalization 0.63% : 0.000018s : 9: substitution.minmaximum_grad 0.42% : 0.000012s : 7: substitution.partial_eliminate 0.71% : 0.000021s : 28: substitution.remove_not_recompute_node 4.80% : 0.000141s : 38: substitution.replace_applicator 0.70% : 0.000021s : 19: substitution.replace_old_param 0.19% : 0.000006s : 1: substitution.set_cell_output_no_recompute 1.23% : 0.000036s : 9: substitution.tuple_list_convert_item_index_to_positive 0.51% : 0.000015s : 9: substitution.tuple_list_get_item_const_eliminator 0.76% : 0.000022s : 9: substitution.tuple_list_get_item_depend_reorder 4.35% : 0.000128s : 31: substitution.tuple_list_get_item_eliminator 0.72% : 0.000021s : 9: substitution.tuple_list_get_set_item_eliminator 1.40% : 0.000041s : 16: substitution.updatestate_pure_node_eliminater 2.04% : 0.000060s : 21: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.204796 2 99.09% : 0.202926s : 1: type_inference.infer 0.91% : 0.001870s : 1: type_inference.specialize ------[replace.] 0.000728 49 9.20% : 0.000067s : 3: replace.getattr_setattr_resolve 43.16% : 0.000314s : 24: replace.inline 11.20% : 0.000081s : 1: replace.replace_applicator 34.24% : 0.000249s : 20: replace.tuple_list_get_item_eliminator 2.21% : 0.000016s : 1: replace.updatestate_useless_node_eliminater ------[match.] 0.002297 49 41.38% : 0.000950s : 3: match.getattr_setattr_resolve 53.85% : 0.001237s : 24: match.inline 0.96% : 0.000022s : 1: match.replace_applicator 2.93% : 0.000067s : 20: match.tuple_list_get_item_eliminator 0.89% : 0.000020s : 1: match.updatestate_useless_node_eliminater ------[predicate.] 0.001556 8683 0.76% : 0.000012s : 85: predicate.accumulaten_eliminater 0.24% : 0.000004s : 9: predicate.ad_related_special_op_eliminate 0.35% : 0.000005s : 37: predicate.addn_check_dump 0.80% : 0.000013s : 85: predicate.addn_zero_filter 0.79% : 0.000012s : 85: predicate.adjust_all_reduce_mul_add 2.33% : 0.000036s : 122: predicate.arithmetic_simplify 0.83% : 0.000013s : 85: predicate.cast_eliminate 2.20% : 0.000034s : 260: predicate.check_bprop_eliminate 0.32% : 0.000005s : 37: predicate.compare_switch_simplify 0.05% : 0.000001s : 9: predicate.const_output_eliminate 0.35% : 0.000005s : 37: predicate.depend_value_elim 0.80% : 0.000012s : 85: predicate.dict_get_item_const_eliminator 0.90% : 0.000014s : 85: predicate.dict_get_item_eliminator 0.76% : 0.000012s : 85: predicate.dict_set_item_eliminator 0.29% : 0.000005s : 18: predicate.dumpgradient_eliminate 0.07% : 0.000001s : 9: predicate.elim_not_effective 0.13% : 0.000002s : 9: predicate.elim_shapecalc_of_broadcastargs 0.80% : 0.000012s : 94: predicate.environ_add_const_eliminate 0.86% : 0.000013s : 94: predicate.environ_get_add_eliminate 0.79% : 0.000012s : 94: predicate.environ_get_depend_swap 1.24% : 0.000019s : 131: predicate.environ_get_eliminate 0.84% : 0.000013s : 94: predicate.environ_get_set_eliminate 1.17% : 0.000018s : 130: predicate.exchange_switch_depend_value 1.94% : 0.000030s : 130: predicate.float_depend_g_call 0.34% : 0.000005s : 37: predicate.float_environ_get_switch 0.44% : 0.000007s : 46: predicate.float_tuple_getitem_switch 0.05% : 0.000001s : 9: predicate.fold_const_symbol 0.40% : 0.000006s : 37: predicate.get_grad_eliminate 0.47% : 0.000007s : 20: predicate.getattr_setattr_resolve 0.07% : 0.000001s : 9: predicate.graph_param_transform 0.33% : 0.000005s : 37: predicate.incorporate_call 0.30% : 0.000005s : 37: predicate.incorporate_call_switch 4.12% : 0.000064s : 307: predicate.inline 1.43% : 0.000022s : 111: predicate.inline_without_move 0.18% : 0.000003s : 37: predicate.j_node_and_user_rematch 0.53% : 0.000008s : 37: predicate.less_batch_normalization 1.21% : 0.000019s : 123: predicate.list_to_tuple_eliminator_ 1.83% : 0.000028s : 208: predicate.load_eliminater 0.24% : 0.000004s : 9: predicate.loop_unroll_after_grad 1.76% : 0.000027s : 197: predicate.loop_unroll_before_grad 1.03% : 0.000016s : 103: predicate.make_slice_get_slice_eliminator 0.34% : 0.000005s : 37: predicate.merge_addn 2.20% : 0.000034s : 256: predicate.micro_step_allgather_replace 2.37% : 0.000037s : 256: predicate.mini_step_allgather_replace 0.76% : 0.000012s : 85: predicate.minmaximum_grad 0.28% : 0.000004s : 9: predicate.mutable_eliminate 0.14% : 0.000002s : 9: predicate.opt_reshape 0.13% : 0.000002s : 9: predicate.parallel_virtual_node 2.38% : 0.000037s : 130: predicate.partial_defer_inline 1.33% : 0.000021s : 114: predicate.partial_eliminate 1.35% : 0.000021s : 85: predicate.print_const_string_wrapper 0.35% : 0.000005s : 37: predicate.reduce_all_const_elim 1.00% : 0.000016s : 85: predicate.reduce_eliminate 1.85% : 0.000029s : 208: predicate.redundant_stop_gradient_eliminater 0.23% : 0.000004s : 37: predicate.remove_not_recompute_node 2.12% : 0.000033s : 363: predicate.replace_applicator 0.63% : 0.000010s : 111: predicate.replace_old_param 0.08% : 0.000001s : 9: predicate.reset_defer_inline 0.77% : 0.000012s : 85: predicate.reshape_eliminate 2.27% : 0.000035s : 256: predicate.row_tensor_add_zeros_like 0.11% : 0.000002s : 9: predicate.row_tensor_eliminate 2.41% : 0.000037s : 260: predicate.same_eliminate 0.25% : 0.000004s : 39: predicate.set_cell_output_no_recompute 0.42% : 0.000007s : 37: predicate.shard_identity_eliminate 0.20% : 0.000003s : 18: predicate.special_op_eliminate 0.41% : 0.000006s : 37: predicate.specialize_transform 2.34% : 0.000036s : 256: predicate.split_environ_get_set_with_tuple_value 1.22% : 0.000019s : 111: predicate.stack_unstack_eliminate 0.10% : 0.000002s : 9: predicate.switch_call_monad_eliminater 12.43% : 0.000193s : 130: predicate.switch_defer_inline 3.48% : 0.000054s : 390: predicate.switch_layer_defer_inline 5.53% : 0.000086s : 373: predicate.switch_simplify 0.79% : 0.000012s : 85: predicate.tile_eliminate 0.85% : 0.000013s : 85: predicate.transpose_eliminate 1.12% : 0.000017s : 103: predicate.tuple_list_convert_item_index_to_positive 1.12% : 0.000017s : 103: predicate.tuple_list_get_item_const_eliminator 1.09% : 0.000017s : 103: predicate.tuple_list_get_item_depend_reorder 2.28% : 0.000035s : 160: predicate.tuple_list_get_item_eliminator 1.07% : 0.000017s : 103: predicate.tuple_list_get_set_item_eliminator 1.38% : 0.000022s : 140: predicate.tuple_list_set_item_eliminator 1.22% : 0.000019s : 123: predicate.tuple_to_list_eliminator_ 1.83% : 0.000028s : 208: predicate.updatestate_pure_node_eliminater 2.33% : 0.000036s : 246: predicate.updatestate_useless_node_eliminater 0.11% : 0.000002s : 9: predicate.value_based_eliminate 0.43% : 0.000007s : 37: predicate.virtual_dataset_eliminate 0.37% : 0.000006s : 37: predicate.virtual_output_eliminate 0.09% : 0.000001s : 9: predicate.virtual_view_grad_eliminate 0.13% : 0.000002s : 9: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.004511 53 49.28% : 0.002223s : 23: func_graph_cloner_run.FuncGraphClonerGraph 50.72% : 0.002288s : 30: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.723954 247 0.00% : 0.000005s : 1: ForceFp32Comm 0.26% : 0.004427s : 1: add_attr 0.26% : 0.004411s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000129s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.02% : 0.000344s : 1: auto_monad 0.00% : 0.000046s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.03% : 0.000582s : 1: bootstrap 0.00% : 0.000048s : 1: cconv 0.00% : 0.000005s : 1: comm_op_add_attrs 0.00% : 0.000030s : 1: control_data_broadcast_order 0.00% : 0.000015s : 1: convert_after_rewriter 0.00% : 0.000048s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000020s : 1: environ_conv 0.01% : 0.000089s : 1: event_method 0.00% : 0.000020s : 1: execute 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000020s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000008s : 1: insert-virtual-dataset 0.00% : 0.000006s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000014s : 1: label_micro_interleaved_index 0.04% : 0.000608s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.05% : 0.000889s : 1: mutable_eliminate 0.00% : 0.000011s : 1: offloading_packed_experts 0.00% : 0.000025s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000032s : 1: opt.transform.mutable_eliminate 10.10% : 0.174185s : 125: opt.transform.opt_a 7.57% : 0.130540s : 1: opt.transform.opt_after_cconv 0.00% : 0.000059s : 1: opt.transform.opt_after_jit_grad 0.02% : 0.000282s : 28: opt.transform.opt_b 0.07% : 0.001141s : 2: opt.transform.opt_resolve 0.01% : 0.000132s : 2: opt.transform.opt_trans_graph 0.01% : 0.000089s : 4: opt.transform.symbol_engine_opt 19.82% : 0.341644s : 1: opt_a 7.59% : 0.130818s : 1: opt_after_cconv 0.05% : 0.000906s : 1: opt_after_jit_grad 0.03% : 0.000468s : 1: opt_b 27.60% : 0.475890s : 1: optimize 0.00% : 0.000036s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000011s : 1: order_py_execute_after_rewriter 0.00% : 0.000039s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000011s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000010s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000004s : 1: pipeline_split 0.00% : 0.000070s : 1: pre_auto_parallel 0.00% : 0.000074s : 1: py_interpret_to_execute 0.00% : 0.000036s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000033s : 1: remove_dup_value 0.40% : 0.006861s : 2: renormalize.infer 8.68% : 0.149683s : 2: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.00% : 0.000086s : 1: rewriter_after_opt_a 0.01% : 0.000250s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000014s : 1: swap_dp_allreduce_reducescatter 0.01% : 0.000154s : 1: symbol_engine_optimizer 5.40% : 0.093161s : 1: task_emit 0.01% : 0.000180s : 1: tuple_transform 11.89% : 0.204951s : 1: type_inference 0.01% : 0.000128s : 1: validate . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[0-shape4] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[0-shape4],max_mem:4.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:42:14.622.061 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.527101, [24] [bootstrap]: 0.00074423 [type_inference]: 0.219292 [event_method]: 2.752e-05 [auto_monad]: 0.00025967 [graph_reusing]: 6.14001e-06 [inline]: 3.48e-06 [add_attr]: 0.007766, [1] [add_attr_with_inline]: 0.00774424, [1] [Cycle 1]: 0.00010047, [2] [tag_attr]: 2.545e-05 [meta_addattr_fg_expand]: 4.31002e-06 [parallel-infer-symbol]: 4.42e-06 [pre_auto_parallel]: 4.416e-05 [insert-virtual-dataset]: 2.56e-06 [parallel-infer-symbol-second]: 1.28002e-06 [dataset_repeat_opt]: 2.26e-06 [pipeline_split]: 1.61002e-06 [optimize]: 0.00894174, [53] [py_interpret_to_execute]: 3.87e-05 [rewriter_before_opt_a]: 8.041e-05 [opt_a]: 0.00517747, [2] [Cycle 1]: 0.00378785, [45] [expand_dump_flag]: 8.84e-06 [switch_simplify]: 4.079e-05 [loop_unroll]: 2.235e-05 [a_1]: 0.00093352 [with_stream_mark]: 4.079e-05 [recompute_prepare]: 2.979e-05 [updatestate_depend_eliminate]: 1.512e-05 [updatestate_assign_eliminate]: 6.78e-06 [updatestate_loads_eliminate]: 9.25999e-06 [parameter_eliminate]: 3.01001e-06 [a_2]: 0.00015954 [accelerated_algorithm]: 1.687e-05 [shard]: 2.76999e-06 [meta_shard_fg_expand]: 2.62001e-06 [shard_inline]: 1.149e-05 [merge_send_recv]: 1.279e-05 [auto_parallel]: 2.121e-05 [parallel]: 4.002e-05 [flash_sp]: 1.412e-05 [merge_comm]: 9.41e-06 [allreduce_fusion]: 6.02999e-06 [matmul_add_comm_reduction]: 2.306e-05 [allreduce_slice_to_reducescatter]: 6.40022e-07 [virtual_shard_identity]: 1.656e-05 [virtual_dataset]: 1.955e-05 [get_grad_eliminate_]: 1.814e-05 [virtual_output]: 1.747e-05 [merge_forward]: 6.90002e-06 [cell_reuse_recompute_pass]: 2.54001e-06 [offload_activation]: 2.098e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.722e-05 [merge_recompute_call_nodes]: 2.17999e-06 [before_grad]: 3.297e-05 [set_forward_comm_id_for_comm_node_pass]: 1.286e-05 [meta_fg_expand]: 4.80001e-06 [flash_sp_send_recv_attached]: 5.00001e-06 [receive_attached]: 5.88002e-06 [after_resolve]: 3.084e-05 [a_after_grad]: 2.444e-05 [renormalize]: 0.00129742 [add_forward_monad_depend]: 1.452e-05 [auto_monad_grad]: 2.59001e-06 [auto_monad_eliminator]: 3.224e-05 [cse]: 4.678e-05 [a_3]: 7.979e-05 [Cycle 2]: 0.0013718, [45] [expand_dump_flag]: 2.89999e-06 [switch_simplify]: 1.176e-05 [loop_unroll]: 1.025e-05 [a_1]: 0.00022418 [with_stream_mark]: 2.455e-05 [recompute_prepare]: 1.037e-05 [updatestate_depend_eliminate]: 6.66e-06 [updatestate_assign_eliminate]: 5.76e-06 [updatestate_loads_eliminate]: 8.72998e-06 [parameter_eliminate]: 1.81998e-06 [a_2]: 0.00018517 [accelerated_algorithm]: 1.134e-05 [shard]: 3.67002e-06 [meta_shard_fg_expand]: 2.82002e-06 [shard_inline]: 1.268e-05 [merge_send_recv]: 1.15e-05 [auto_parallel]: 1.297e-05 [parallel]: 1.499e-05 [flash_sp]: 4.4e-06 [merge_comm]: 5.52001e-06 [allreduce_fusion]: 7.97998e-06 [matmul_add_comm_reduction]: 1.36e-05 [allreduce_slice_to_reducescatter]: 6.19999e-07 [virtual_shard_identity]: 1.182e-05 [virtual_dataset]: 1.162e-05 [get_grad_eliminate_]: 1.017e-05 [virtual_output]: 1.046e-05 [merge_forward]: 1.185e-05 [cell_reuse_recompute_pass]: 2.93e-06 [offload_activation]: 2.323e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.874e-05 [merge_recompute_call_nodes]: 2.09e-06 [before_grad]: 2.813e-05 [set_forward_comm_id_for_comm_node_pass]: 1.22e-05 [meta_fg_expand]: 4.65001e-06 [flash_sp_send_recv_attached]: 1.88002e-06 [receive_attached]: 2.57001e-06 [after_resolve]: 1.627e-05 [a_after_grad]: 1.512e-05 [renormalize]: 7.99773e-08 [add_forward_monad_depend]: 4.08001e-06 [auto_monad_grad]: 2.35997e-06 [auto_monad_eliminator]: 2.182e-05 [cse]: 3.264e-05 [a_3]: 5.705e-05 [py_interpret_to_execute_after_opt_a]: 2.901e-05 [slice_cell_reuse_recomputed_activation]: 2.66e-06 [rewriter_after_opt_a]: 0.0002379 [convert_after_rewriter]: 1.593e-05 [order_py_execute_after_rewriter]: 7.58001e-06 [mutable_eliminate]: 0.00089937 [opt_b]: 0.00046102, [1] [Cycle 1]: 0.00044995, [7] [b_1]: 0.00029746 [b_2]: 1.302e-05 [updatestate_depend_eliminate]: 1.343e-05 [updatestate_assign_eliminate]: 8.50001e-06 [updatestate_loads_eliminate]: 8.1e-06 [renormalize]: 1.23002e-06 [cse]: 4.052e-05 [optimize_parallel_all_gather_comm]: 2.994e-05 [overlap_param_gather]: 2.64999e-06 [cconv]: 4.579e-05 [loop_unroll]: 0.00064007 [opt_after_cconv]: 0.00021748, [1] [Cycle 1]: 0.00020901, [7] [c_1]: 7.727e-05 [parameter_eliminate]: 6.02999e-06 [updatestate_depend_eliminate]: 1.263e-05 [updatestate_assign_eliminate]: 8.82e-06 [updatestate_loads_eliminate]: 8.35999e-06 [cse]: 3.228e-05 [renormalize]: 7.09988e-07 [remove_dup_value]: 2.132e-05 [tuple_transform]: 0.00014176, [1] [Cycle 1]: 0.00013568, [4] [d_1]: 9.103e-05 [none_parameter_eliminate]: 1.84e-06 [renormalize]: 1.80007e-07 [switch_simplify]: 1.043e-05 [partial_unused_args_eliminate]: 1.94e-06 [add_recomputation]: 0.00011887 [cse_after_recomputation]: 3.178e-05, [1] [Cycle 1]: 2.634e-05, [1] [cse]: 1.849e-05 [environ_conv]: 8.50999e-06 [swap_dp_allreduce_reducescatter]: 8.52e-06 [bias_add_comm_swap]: 6.299e-05 [label_micro_interleaved_index]: 5.24e-06 [label_fine_grained_interleaved_index]: 2.98e-06 [merge_cast_opt]: 1.64e-06 [slice_recompute_activation]: 2.49001e-06 [micro_interleaved_order_control]: 2.61e-06 [assign_add_opt]: 1.35999e-06 [ForceFp32Comm]: 1.48002e-06 [remove_cast_before_assign_add]: 1.49e-06 [full_micro_interleaved_order_control]: 2.46e-06 [reorder_send_recv_between_fp_bp]: 2.86999e-06 [comm_op_add_attrs]: 1.30999e-06 [add_comm_op_reuse_tag]: 1.14e-06 [interleave_split_concat_branches]: 1.42e-06 [interleave_parallel_branches]: 1.32e-06 [overlap_opt_shard_in_pipeline]: 3.93001e-06 [overlap_opt_shard_grad_in_pipeline]: 2.04e-06 [control_data_broadcast_order]: 2.264e-05 [grouped_pairwise_exchange_alltoall]: 1.67999e-06 [offloading_packed_experts]: 1.18e-05 [overlap_recompute_and_grad_model_parallel]: 8.62998e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.44998e-06 [overlap_recompute_allgather_and_fa_grad]: 1.54e-06 [overlap_recompute_comm]: 2.81e-06 [overlap_grad_ring_attention]: 6.03002e-06 [overlap_grad_flash_sp]: 3.028e-05 [begin_end_overlap_inline]: 6.19999e-07 [split_matmul_comm_elemetwise]: 2.24001e-06 [split_layernorm_comm]: 1.84e-06 [handle_group_info]: 1.07998e-06 [symbol_engine_optimizer]: 0.00014361, [1] [Cycle 1]: 0.00013634, [6] [build]: 5.59e-06 [elim_shapecalc]: 2.307e-05 [elim_not_effective]: 2.22e-05 [opt_reshape]: 1.142e-05 [fold_const_symbol]: 2.026e-05 [renormalize]: 2.99973e-07 [detach_backward]: 2.15002e-06 [pipeline_parallel_scheduler]: 1.84998e-06 [auto_monad_reorder]: 3.402e-05 [get_jit_bprop_graph]: 1.99e-06 [rewriter_after_jit_bprop_graph]: 6.76e-06 [opt_after_jit_grad]: 0.00077075 [validate]: 7.485e-05 [backend_pass]: 9.70002e-07 [task_emit]: 0.288708 [execute]: 1.104e-05 Sums bootstrap : 0.000744s : 0.14% type_inference : 0.219292s : 42.40% event_method : 0.000028s : 0.01% auto_monad : 0.000260s : 0.05% graph_reusing : 0.000006s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000025s : 0.00% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000044s : 0.01% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000039s : 0.01% optimize.rewriter_before_opt_a : 0.000080s : 0.02% optimize.opt_a.expand_dump_flag : 0.000012s : 0.00% optimize.opt_a.switch_simplify : 0.000053s : 0.01% optimize.opt_a.loop_unroll : 0.000033s : 0.01% optimize.opt_a.a_1 : 0.001158s : 0.22% optimize.opt_a.with_stream_mark : 0.000065s : 0.01% optimize.opt_a.recompute_prepare : 0.000040s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000022s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000013s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000018s : 0.00% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000345s : 0.07% optimize.opt_a.accelerated_algorithm : 0.000028s : 0.01% optimize.opt_a.shard : 0.000006s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000005s : 0.00% optimize.opt_a.shard_inline : 0.000024s : 0.00% optimize.opt_a.merge_send_recv : 0.000024s : 0.00% optimize.opt_a.auto_parallel : 0.000034s : 0.01% optimize.opt_a.parallel : 0.000055s : 0.01% optimize.opt_a.flash_sp : 0.000019s : 0.00% optimize.opt_a.merge_comm : 0.000015s : 0.00% optimize.opt_a.allreduce_fusion : 0.000014s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000037s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000028s : 0.01% optimize.opt_a.virtual_dataset : 0.000031s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000028s : 0.01% optimize.opt_a.virtual_output : 0.000028s : 0.01% optimize.opt_a.merge_forward : 0.000019s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000044s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000076s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000061s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000025s : 0.00% optimize.opt_a.meta_fg_expand : 0.000009s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000007s : 0.00% optimize.opt_a.receive_attached : 0.000008s : 0.00% optimize.opt_a.after_resolve : 0.000047s : 0.01% optimize.opt_a.a_after_grad : 0.000040s : 0.01% optimize.opt_a.renormalize : 0.001297s : 0.25% optimize.opt_a.add_forward_monad_depend : 0.000019s : 0.00% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000054s : 0.01% optimize.opt_a.cse : 0.000079s : 0.02% optimize.opt_a.a_3 : 0.000137s : 0.03% optimize.py_interpret_to_execute_after_opt_a : 0.000029s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000238s : 0.05% optimize.convert_after_rewriter : 0.000016s : 0.00% optimize.order_py_execute_after_rewriter : 0.000008s : 0.00% optimize.mutable_eliminate : 0.000899s : 0.17% optimize.opt_b.b_1 : 0.000297s : 0.06% optimize.opt_b.b_2 : 0.000013s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000013s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000009s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000008s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000041s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000030s : 0.01% optimize.overlap_param_gather : 0.000003s : 0.00% optimize.cconv : 0.000046s : 0.01% optimize.loop_unroll : 0.000640s : 0.12% optimize.opt_after_cconv.c_1 : 0.000077s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000013s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000009s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000008s : 0.00% optimize.opt_after_cconv.cse : 0.000032s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000021s : 0.00% optimize.tuple_transform.d_1 : 0.000091s : 0.02% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000010s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000119s : 0.02% optimize.cse_after_recomputation.cse : 0.000018s : 0.00% optimize.environ_conv : 0.000009s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000009s : 0.00% optimize.bias_add_comm_swap : 0.000063s : 0.01% optimize.label_micro_interleaved_index : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000004s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000023s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000012s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000009s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000006s : 0.00% optimize.overlap_grad_flash_sp : 0.000030s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000006s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000023s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000022s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000011s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000020s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000034s : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.00% opt_after_jit_grad : 0.000771s : 0.15% validate : 0.000075s : 0.01% backend_pass : 0.000001s : 0.00% task_emit : 0.288708s : 55.82% execute : 0.000011s : 0.00% Time group info: ------[substitution.] 0.000320 55 4.51% : 0.000014s : 2: substitution.depend_value_elim 0.85% : 0.000003s : 5: substitution.elim_not_effective 0.69% : 0.000002s : 5: substitution.fold_const_symbol 2.57% : 0.000008s : 6: substitution.graph_param_transform 69.81% : 0.000223s : 1: substitution.inline 2.46% : 0.000008s : 10: substitution.j_node_and_user_rematch 2.89% : 0.000009s : 10: substitution.remove_not_recompute_node 2.42% : 0.000008s : 2: substitution.replace_old_param 7.82% : 0.000025s : 6: substitution.updatestate_pure_node_eliminater 5.98% : 0.000019s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.219206 2 99.50% : 0.218110s : 1: type_inference.infer 0.50% : 0.001096s : 1: type_inference.specialize ------[replace.] 0.000030 1 100.00% : 0.000030s : 1: replace.inline ------[match.] 0.000222 1 100.00% : 0.000222s : 1: match.inline ------[predicate.] 0.000258 1419 0.80% : 0.000002s : 13: predicate.accumulaten_eliminater 1.32% : 0.000003s : 6: predicate.ad_related_special_op_eliminate 0.70% : 0.000002s : 12: predicate.addn_check_dump 0.89% : 0.000002s : 13: predicate.addn_zero_filter 0.74% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 2.30% : 0.000006s : 25: predicate.arithmetic_simplify 0.82% : 0.000002s : 13: predicate.cast_eliminate 0.75% : 0.000002s : 12: predicate.check_bprop_eliminate 0.65% : 0.000002s : 12: predicate.compare_switch_simplify 0.20% : 0.000001s : 6: predicate.const_output_eliminate 0.76% : 0.000002s : 12: predicate.depend_value_elim 0.81% : 0.000002s : 13: predicate.dict_get_item_const_eliminator 0.99% : 0.000003s : 13: predicate.dict_get_item_eliminator 0.82% : 0.000002s : 13: predicate.dict_set_item_eliminator 1.26% : 0.000003s : 12: predicate.dumpgradient_eliminate 0.35% : 0.000001s : 6: predicate.elim_not_effective 0.58% : 0.000002s : 6: predicate.elim_shapecalc_of_broadcastargs 1.15% : 0.000003s : 19: predicate.environ_add_const_eliminate 1.10% : 0.000003s : 19: predicate.environ_get_add_eliminate 1.07% : 0.000003s : 19: predicate.environ_get_depend_swap 1.94% : 0.000005s : 31: predicate.environ_get_eliminate 1.03% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.80% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.51% : 0.000004s : 14: predicate.float_depend_g_call 0.71% : 0.000002s : 12: predicate.float_environ_get_switch 1.07% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.20% : 0.000001s : 6: predicate.fold_const_symbol 0.88% : 0.000002s : 12: predicate.get_grad_eliminate 0.32% : 0.000001s : 6: predicate.graph_param_transform 0.71% : 0.000002s : 12: predicate.incorporate_call 0.60% : 0.000002s : 12: predicate.incorporate_call_switch 5.50% : 0.000014s : 63: predicate.inline 1.17% : 0.000003s : 12: predicate.inline_without_move 0.35% : 0.000001s : 12: predicate.j_node_and_user_rematch 1.32% : 0.000003s : 12: predicate.less_batch_normalization 1.65% : 0.000004s : 25: predicate.list_to_tuple_eliminator_ 2.10% : 0.000005s : 38: predicate.load_eliminater 0.99% : 0.000003s : 6: predicate.loop_unroll_after_grad 1.16% : 0.000003s : 17: predicate.loop_unroll_before_grad 1.98% : 0.000005s : 25: predicate.make_slice_get_slice_eliminator 0.73% : 0.000002s : 12: predicate.merge_addn 0.75% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.72% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.74% : 0.000002s : 13: predicate.minmaximum_grad 1.81% : 0.000005s : 6: predicate.mutable_eliminate 0.50% : 0.000001s : 6: predicate.opt_reshape 0.49% : 0.000001s : 6: predicate.parallel_virtual_node 1.46% : 0.000004s : 14: predicate.partial_defer_inline 1.13% : 0.000003s : 19: predicate.partial_eliminate 0.74% : 0.000002s : 13: predicate.print_const_string_wrapper 0.72% : 0.000002s : 12: predicate.reduce_all_const_elim 0.99% : 0.000003s : 13: predicate.reduce_eliminate 2.13% : 0.000005s : 38: predicate.redundant_stop_gradient_eliminater 0.68% : 0.000002s : 12: predicate.remove_not_recompute_node 1.08% : 0.000003s : 25: predicate.replace_applicator 0.60% : 0.000002s : 12: predicate.replace_old_param 0.45% : 0.000001s : 6: predicate.reset_defer_inline 0.78% : 0.000002s : 13: predicate.reshape_eliminate 0.80% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.50% : 0.000001s : 6: predicate.row_tensor_eliminate 1.10% : 0.000003s : 12: predicate.same_eliminate 0.64% : 0.000002s : 12: predicate.set_cell_output_no_recompute 1.11% : 0.000003s : 12: predicate.shard_identity_eliminate 0.93% : 0.000002s : 12: predicate.special_op_eliminate 0.89% : 0.000002s : 12: predicate.specialize_transform 1.28% : 0.000003s : 12: predicate.split_environ_get_set_with_tuple_value 1.20% : 0.000003s : 12: predicate.stack_unstack_eliminate 0.46% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.89% : 0.000002s : 14: predicate.switch_defer_inline 1.54% : 0.000004s : 26: predicate.switch_layer_defer_inline 3.52% : 0.000009s : 49: predicate.switch_simplify 0.91% : 0.000002s : 13: predicate.tile_eliminate 0.86% : 0.000002s : 13: predicate.transpose_eliminate 1.69% : 0.000004s : 25: predicate.tuple_list_convert_item_index_to_positive 1.56% : 0.000004s : 25: predicate.tuple_list_get_item_const_eliminator 1.58% : 0.000004s : 25: predicate.tuple_list_get_item_depend_reorder 3.11% : 0.000008s : 37: predicate.tuple_list_get_item_eliminator 1.61% : 0.000004s : 25: predicate.tuple_list_get_set_item_eliminator 2.47% : 0.000006s : 37: predicate.tuple_list_set_item_eliminator 1.64% : 0.000004s : 25: predicate.tuple_to_list_eliminator_ 2.30% : 0.000006s : 38: predicate.updatestate_pure_node_eliminater 3.49% : 0.000009s : 50: predicate.updatestate_useless_node_eliminater 0.51% : 0.000001s : 6: predicate.value_based_eliminate 0.92% : 0.000002s : 12: predicate.virtual_dataset_eliminate 1.05% : 0.000003s : 12: predicate.virtual_output_eliminate 0.34% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.55% : 0.000001s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000754 4 4.06% : 0.000031s : 1: func_graph_cloner_run.FuncGraphClonerGraph 95.94% : 0.000724s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.547636 196 0.00% : 0.000008s : 1: ForceFp32Comm 1.42% : 0.007773s : 1: add_attr 1.42% : 0.007751s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.02% : 0.000125s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.05% : 0.000274s : 1: auto_monad 0.01% : 0.000041s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000008s : 1: begin_end_overlap_inline 0.01% : 0.000068s : 1: bias_add_comm_swap 0.16% : 0.000856s : 1: bootstrap 0.01% : 0.000053s : 1: cconv 0.00% : 0.000006s : 1: comm_op_add_attrs 0.01% : 0.000030s : 1: control_data_broadcast_order 0.00% : 0.000021s : 1: convert_after_rewriter 0.01% : 0.000041s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000005s : 1: detach_backward 0.00% : 0.000012s : 1: environ_conv 0.01% : 0.000035s : 1: event_method 0.00% : 0.000019s : 1: execute 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000011s : 1: graph_reusing 0.00% : 0.000009s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000007s : 1: handle_group_info 0.00% : 0.000010s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000008s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000010s : 1: label_fine_grained_interleaved_index 0.00% : 0.000017s : 1: label_micro_interleaved_index 0.12% : 0.000653s : 1: loop_unroll 0.00% : 0.000006s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.17% : 0.000917s : 1: mutable_eliminate 0.00% : 0.000016s : 1: offloading_packed_experts 0.00% : 0.000025s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000029s : 1: opt.transform.mutable_eliminate 0.37% : 0.002036s : 78: opt.transform.opt_a 0.01% : 0.000075s : 1: opt.transform.opt_after_cconv 0.01% : 0.000054s : 1: opt.transform.opt_after_jit_grad 0.05% : 0.000266s : 28: opt.transform.opt_b 0.02% : 0.000098s : 2: opt.transform.opt_trans_graph 0.01% : 0.000073s : 4: opt.transform.symbol_engine_opt 0.95% : 0.005184s : 1: opt_a 0.04% : 0.000222s : 1: opt_after_cconv 0.14% : 0.000793s : 1: opt_after_jit_grad 0.09% : 0.000466s : 1: opt_b 1.63% : 0.008949s : 1: optimize 0.01% : 0.000038s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000018s : 1: order_py_execute_after_rewriter 0.01% : 0.000035s : 1: overlap_grad_flash_sp 0.00% : 0.000009s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000010s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000012s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000008s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000016s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000005s : 1: parallel-infer-symbol-second 0.00% : 0.000011s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000004s : 1: pipeline_split 0.01% : 0.000049s : 1: pre_auto_parallel 0.01% : 0.000043s : 1: py_interpret_to_execute 0.01% : 0.000038s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000006s : 1: remove_cast_before_assign_add 0.00% : 0.000026s : 1: remove_dup_value 0.12% : 0.000663s : 1: renormalize.infer 0.11% : 0.000622s : 1: renormalize.specialize 0.00% : 0.000009s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000013s : 1: rewriter_after_jit_bprop_graph 0.05% : 0.000253s : 1: rewriter_after_opt_a 0.02% : 0.000086s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000009s : 1: split_layernorm_comm 0.00% : 0.000007s : 1: split_matmul_comm_elemetwise 0.00% : 0.000012s : 1: swap_dp_allreduce_reducescatter 0.03% : 0.000147s : 1: symbol_engine_optimizer 52.72% : 0.288733s : 1: task_emit 0.03% : 0.000149s : 1: tuple_transform 40.05% : 0.219317s : 1: type_inference 0.02% : 0.000132s : 1: validate TotalTime = 0.4821, [24] [bootstrap]: 0.00051137 [type_inference]: 0.00405365 [event_method]: 1.767e-05 [auto_monad]: 0.0001724 [graph_reusing]: 5.82001e-06 [inline]: 3.34001e-06 [add_attr]: 0.305358, [1] [add_attr_with_inline]: 0.305341, [1] [Cycle 1]: 8.036e-05, [2] [tag_attr]: 1.983e-05 [meta_addattr_fg_expand]: 3.9e-06 [parallel-infer-symbol]: 4.23999e-06 [pre_auto_parallel]: 4.201e-05 [insert-virtual-dataset]: 3.2e-06 [parallel-infer-symbol-second]: 1.24e-06 [dataset_repeat_opt]: 1.96e-06 [pipeline_split]: 1.95001e-06 [optimize]: 0.00858718, [53] [py_interpret_to_execute]: 4.013e-05 [rewriter_before_opt_a]: 6.717e-05 [opt_a]: 0.00442493, [2] [Cycle 1]: 0.00303167, [45] [expand_dump_flag]: 3.64002e-06 [switch_simplify]: 2.963e-05 [loop_unroll]: 1.729e-05 [a_1]: 0.00056962 [with_stream_mark]: 2.767e-05 [recompute_prepare]: 1.991e-05 [updatestate_depend_eliminate]: 7.01001e-06 [updatestate_assign_eliminate]: 5.92999e-06 [updatestate_loads_eliminate]: 5.47001e-06 [parameter_eliminate]: 2.69999e-06 [a_2]: 0.0001537 [accelerated_algorithm]: 1.374e-05 [shard]: 3.2e-06 [meta_shard_fg_expand]: 3.6e-06 [shard_inline]: 1.133e-05 [merge_send_recv]: 1.284e-05 [auto_parallel]: 1.715e-05 [parallel]: 2.35e-05 [flash_sp]: 1.448e-05 [merge_comm]: 7.37002e-06 [allreduce_fusion]: 6.43e-06 [matmul_add_comm_reduction]: 1.465e-05 [allreduce_slice_to_reducescatter]: 9.49978e-07 [virtual_shard_identity]: 1.931e-05 [virtual_dataset]: 1.375e-05 [get_grad_eliminate_]: 1.094e-05 [virtual_output]: 1.212e-05 [merge_forward]: 8.58001e-06 [cell_reuse_recompute_pass]: 3.43999e-06 [offload_activation]: 2.03e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.194e-05 [merge_recompute_call_nodes]: 1.58002e-06 [before_grad]: 2.294e-05 [set_forward_comm_id_for_comm_node_pass]: 7.73001e-06 [meta_fg_expand]: 5.79e-06 [flash_sp_send_recv_attached]: 3.6e-06 [receive_attached]: 2.76999e-06 [after_resolve]: 1.816e-05 [a_after_grad]: 1.847e-05 [renormalize]: 0.00124933 [add_forward_monad_depend]: 1.221e-05 [auto_monad_grad]: 3.29001e-06 [auto_monad_eliminator]: 3.293e-05 [cse]: 4.55e-05 [a_3]: 8.592e-05 [Cycle 2]: 0.00137487, [45] [expand_dump_flag]: 2.79001e-06 [switch_simplify]: 1.341e-05 [loop_unroll]: 1.03e-05 [a_1]: 0.00023862 [with_stream_mark]: 2.776e-05 [recompute_prepare]: 1.148e-05 [updatestate_depend_eliminate]: 7.8e-06 [updatestate_assign_eliminate]: 6.39001e-06 [updatestate_loads_eliminate]: 5.81e-06 [parameter_eliminate]: 2.36e-06 [a_2]: 0.00012605 [accelerated_algorithm]: 1.39e-05 [shard]: 3.77002e-06 [meta_shard_fg_expand]: 4.33999e-06 [shard_inline]: 8.97e-06 [merge_send_recv]: 1.347e-05 [auto_parallel]: 2.108e-05 [parallel]: 1.114e-05 [flash_sp]: 4.42e-06 [merge_comm]: 6.26e-06 [allreduce_fusion]: 5.46e-06 [matmul_add_comm_reduction]: 1.384e-05 [allreduce_slice_to_reducescatter]: 7.00005e-07 [virtual_shard_identity]: 1.291e-05 [virtual_dataset]: 1.029e-05 [get_grad_eliminate_]: 9.29e-06 [virtual_output]: 9.01002e-06 [merge_forward]: 6.01998e-06 [cell_reuse_recompute_pass]: 2.94999e-06 [offload_activation]: 1.603e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.447e-05 [merge_recompute_call_nodes]: 1.60999e-06 [before_grad]: 2.067e-05 [set_forward_comm_id_for_comm_node_pass]: 7.65998e-06 [meta_fg_expand]: 4.58999e-06 [flash_sp_send_recv_attached]: 1.89e-06 [receive_attached]: 2.68e-06 [after_resolve]: 1.758e-05 [a_after_grad]: 1.433e-05 [renormalize]: 7.00238e-08 [add_forward_monad_depend]: 5.04e-06 [auto_monad_grad]: 2.44999e-06 [auto_monad_eliminator]: 2.425e-05 [cse]: 0.00016731 [a_3]: 7.043e-05 [py_interpret_to_execute_after_opt_a]: 3.35e-05 [slice_cell_reuse_recomputed_activation]: 2.99999e-06 [rewriter_after_opt_a]: 0.00031115 [convert_after_rewriter]: 1.489e-05 [order_py_execute_after_rewriter]: 7.35e-06 [mutable_eliminate]: 0.00116219 [opt_b]: 0.00044736, [1] [Cycle 1]: 0.00043556, [7] [b_1]: 0.00027276 [b_2]: 2.476e-05 [updatestate_depend_eliminate]: 1.727e-05 [updatestate_assign_eliminate]: 6.21998e-06 [updatestate_loads_eliminate]: 5.84999e-06 [renormalize]: 1.05001e-06 [cse]: 4.826e-05 [optimize_parallel_all_gather_comm]: 3.308e-05 [overlap_param_gather]: 2.01998e-06 [cconv]: 4.316e-05 [loop_unroll]: 0.00083053 [opt_after_cconv]: 0.00021734, [1] [Cycle 1]: 0.00020668, [7] [c_1]: 8.184e-05 [parameter_eliminate]: 6.95002e-06 [updatestate_depend_eliminate]: 1.378e-05 [updatestate_assign_eliminate]: 6.12001e-06 [updatestate_loads_eliminate]: 4.72e-06 [cse]: 3.994e-05 [renormalize]: 9.10019e-07 [remove_dup_value]: 1.954e-05 [tuple_transform]: 0.0001248, [1] [Cycle 1]: 0.00011814, [4] [d_1]: 7.891e-05 [none_parameter_eliminate]: 2.04e-06 [renormalize]: 1.8999e-07 [switch_simplify]: 1.047e-05 [partial_unused_args_eliminate]: 2.64999e-06 [add_recomputation]: 9.109e-05 [cse_after_recomputation]: 3.273e-05, [1] [Cycle 1]: 2.744e-05, [1] [cse]: 1.952e-05 [environ_conv]: 8.87e-06 [swap_dp_allreduce_reducescatter]: 8.84003e-06 [bias_add_comm_swap]: 3.26001e-06 [label_micro_interleaved_index]: 8.14002e-06 [label_fine_grained_interleaved_index]: 3.65e-06 [merge_cast_opt]: 1.55001e-06 [slice_recompute_activation]: 2.21e-06 [micro_interleaved_order_control]: 2.94001e-06 [assign_add_opt]: 1.33002e-06 [ForceFp32Comm]: 1.16002e-06 [remove_cast_before_assign_add]: 1.24e-06 [full_micro_interleaved_order_control]: 2.54999e-06 [reorder_send_recv_between_fp_bp]: 2.72001e-06 [comm_op_add_attrs]: 1.34e-06 [add_comm_op_reuse_tag]: 1.04e-06 [interleave_split_concat_branches]: 1.32e-06 [interleave_parallel_branches]: 1.98002e-06 [overlap_opt_shard_in_pipeline]: 1.34e-06 [overlap_opt_shard_grad_in_pipeline]: 2.01003e-06 [control_data_broadcast_order]: 2.346e-05 [grouped_pairwise_exchange_alltoall]: 2.90002e-06 [offloading_packed_experts]: 6.22001e-06 [overlap_recompute_and_grad_model_parallel]: 7.86001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.39e-06 [overlap_recompute_allgather_and_fa_grad]: 1.84e-06 [overlap_recompute_comm]: 2.88e-06 [overlap_grad_ring_attention]: 6.59001e-06 [overlap_grad_flash_sp]: 3.195e-05 [begin_end_overlap_inline]: 5.60016e-07 [split_matmul_comm_elemetwise]: 2.50997e-06 [split_layernorm_comm]: 2.19001e-06 [handle_group_info]: 1.15999e-06 [symbol_engine_optimizer]: 0.00013275, [1] [Cycle 1]: 0.00012643, [6] [build]: 7.23e-06 [elim_shapecalc]: 2.267e-05 [elim_not_effective]: 2.211e-05 [opt_reshape]: 1.163e-05 [fold_const_symbol]: 1.738e-05 [renormalize]: 2.3999e-07 [detach_backward]: 2.58e-06 [pipeline_parallel_scheduler]: 1.39998e-06 [auto_monad_reorder]: 3.799e-05 [get_jit_bprop_graph]: 3.08e-06 [rewriter_after_jit_bprop_graph]: 7.38e-06 [opt_after_jit_grad]: 0.00091046 [validate]: 7.331e-05 [backend_pass]: 1.04e-06 [task_emit]: 0.161951 [execute]: 1.444e-05 Sums bootstrap : 0.000511s : 0.29% type_inference : 0.004054s : 2.32% event_method : 0.000018s : 0.01% auto_monad : 0.000172s : 0.10% graph_reusing : 0.000006s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000020s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000042s : 0.02% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000040s : 0.02% optimize.rewriter_before_opt_a : 0.000067s : 0.04% optimize.opt_a.expand_dump_flag : 0.000006s : 0.00% optimize.opt_a.switch_simplify : 0.000043s : 0.02% optimize.opt_a.loop_unroll : 0.000028s : 0.02% optimize.opt_a.a_1 : 0.000808s : 0.46% optimize.opt_a.with_stream_mark : 0.000055s : 0.03% optimize.opt_a.recompute_prepare : 0.000031s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000015s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000012s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000011s : 0.01% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000280s : 0.16% optimize.opt_a.accelerated_algorithm : 0.000028s : 0.02% optimize.opt_a.shard : 0.000007s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000008s : 0.00% optimize.opt_a.shard_inline : 0.000020s : 0.01% optimize.opt_a.merge_send_recv : 0.000026s : 0.02% optimize.opt_a.auto_parallel : 0.000038s : 0.02% optimize.opt_a.parallel : 0.000035s : 0.02% optimize.opt_a.flash_sp : 0.000019s : 0.01% optimize.opt_a.merge_comm : 0.000014s : 0.01% optimize.opt_a.allreduce_fusion : 0.000012s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000028s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000032s : 0.02% optimize.opt_a.virtual_dataset : 0.000024s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000020s : 0.01% optimize.opt_a.virtual_output : 0.000021s : 0.01% optimize.opt_a.merge_forward : 0.000015s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000006s : 0.00% optimize.opt_a.offload_activation : 0.000036s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000056s : 0.03% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000044s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000015s : 0.01% optimize.opt_a.meta_fg_expand : 0.000010s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000036s : 0.02% optimize.opt_a.a_after_grad : 0.000033s : 0.02% optimize.opt_a.renormalize : 0.001249s : 0.71% optimize.opt_a.add_forward_monad_depend : 0.000017s : 0.01% optimize.opt_a.auto_monad_grad : 0.000006s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000057s : 0.03% optimize.opt_a.cse : 0.000213s : 0.12% optimize.opt_a.a_3 : 0.000156s : 0.09% optimize.py_interpret_to_execute_after_opt_a : 0.000033s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000311s : 0.18% optimize.convert_after_rewriter : 0.000015s : 0.01% optimize.order_py_execute_after_rewriter : 0.000007s : 0.00% optimize.mutable_eliminate : 0.001162s : 0.66% optimize.opt_b.b_1 : 0.000273s : 0.16% optimize.opt_b.b_2 : 0.000025s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000017s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000048s : 0.03% optimize.optimize_parallel_all_gather_comm : 0.000033s : 0.02% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000043s : 0.02% optimize.loop_unroll : 0.000831s : 0.47% optimize.opt_after_cconv.c_1 : 0.000082s : 0.05% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000014s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.cse : 0.000040s : 0.02% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000020s : 0.01% optimize.tuple_transform.d_1 : 0.000079s : 0.05% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000010s : 0.01% optimize.partial_unused_args_eliminate : 0.000003s : 0.00% optimize.add_recomputation : 0.000091s : 0.05% optimize.cse_after_recomputation.cse : 0.000020s : 0.01% optimize.environ_conv : 0.000009s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000009s : 0.01% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000008s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000004s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000002s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000023s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000003s : 0.00% optimize.offloading_packed_experts : 0.000006s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000008s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000007s : 0.00% optimize.overlap_grad_flash_sp : 0.000032s : 0.02% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000007s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000023s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000022s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000012s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000017s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000001s : 0.00% auto_monad_reorder : 0.000038s : 0.02% get_jit_bprop_graph : 0.000003s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.00% opt_after_jit_grad : 0.000910s : 0.52% validate : 0.000073s : 0.04% backend_pass : 0.000001s : 0.00% task_emit : 0.161951s : 92.56% execute : 0.000014s : 0.01% Time group info: ------[substitution.] 0.000244 55 4.56% : 0.000011s : 2: substitution.depend_value_elim 1.23% : 0.000003s : 5: substitution.elim_not_effective 0.93% : 0.000002s : 5: substitution.fold_const_symbol 3.41% : 0.000008s : 6: substitution.graph_param_transform 58.72% : 0.000143s : 1: substitution.inline 3.34% : 0.000008s : 10: substitution.j_node_and_user_rematch 4.19% : 0.000010s : 10: substitution.remove_not_recompute_node 3.14% : 0.000008s : 2: substitution.replace_old_param 11.37% : 0.000028s : 6: substitution.updatestate_pure_node_eliminater 9.11% : 0.000022s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.003986 2 86.89% : 0.003463s : 1: type_inference.infer 13.11% : 0.000523s : 1: type_inference.specialize ------[replace.] 0.000020 1 100.00% : 0.000020s : 1: replace.inline ------[match.] 0.000142 1 100.00% : 0.000142s : 1: match.inline ------[predicate.] 0.000280 1419 0.89% : 0.000002s : 13: predicate.accumulaten_eliminater 1.88% : 0.000005s : 6: predicate.ad_related_special_op_eliminate 0.72% : 0.000002s : 12: predicate.addn_check_dump 0.87% : 0.000002s : 13: predicate.addn_zero_filter 0.69% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 2.62% : 0.000007s : 25: predicate.arithmetic_simplify 0.82% : 0.000002s : 13: predicate.cast_eliminate 0.77% : 0.000002s : 12: predicate.check_bprop_eliminate 0.66% : 0.000002s : 12: predicate.compare_switch_simplify 0.18% : 0.000001s : 6: predicate.const_output_eliminate 0.70% : 0.000002s : 12: predicate.depend_value_elim 0.75% : 0.000002s : 13: predicate.dict_get_item_const_eliminator 0.83% : 0.000002s : 13: predicate.dict_get_item_eliminator 0.84% : 0.000002s : 13: predicate.dict_set_item_eliminator 1.42% : 0.000004s : 12: predicate.dumpgradient_eliminate 0.30% : 0.000001s : 6: predicate.elim_not_effective 0.55% : 0.000002s : 6: predicate.elim_shapecalc_of_broadcastargs 1.23% : 0.000003s : 19: predicate.environ_add_const_eliminate 0.98% : 0.000003s : 19: predicate.environ_get_add_eliminate 0.99% : 0.000003s : 19: predicate.environ_get_depend_swap 1.81% : 0.000005s : 31: predicate.environ_get_eliminate 0.97% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.72% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.51% : 0.000004s : 14: predicate.float_depend_g_call 0.64% : 0.000002s : 12: predicate.float_environ_get_switch 0.92% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.17% : 0.000000s : 6: predicate.fold_const_symbol 0.77% : 0.000002s : 12: predicate.get_grad_eliminate 0.29% : 0.000001s : 6: predicate.graph_param_transform 0.66% : 0.000002s : 12: predicate.incorporate_call 0.54% : 0.000002s : 12: predicate.incorporate_call_switch 5.33% : 0.000015s : 63: predicate.inline 1.03% : 0.000003s : 12: predicate.inline_without_move 0.34% : 0.000001s : 12: predicate.j_node_and_user_rematch 0.98% : 0.000003s : 12: predicate.less_batch_normalization 1.67% : 0.000005s : 25: predicate.list_to_tuple_eliminator_ 1.98% : 0.000006s : 38: predicate.load_eliminater 2.14% : 0.000006s : 6: predicate.loop_unroll_after_grad 1.19% : 0.000003s : 17: predicate.loop_unroll_before_grad 1.90% : 0.000005s : 25: predicate.make_slice_get_slice_eliminator 0.66% : 0.000002s : 12: predicate.merge_addn 0.71% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.69% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.75% : 0.000002s : 13: predicate.minmaximum_grad 2.75% : 0.000008s : 6: predicate.mutable_eliminate 0.44% : 0.000001s : 6: predicate.opt_reshape 0.45% : 0.000001s : 6: predicate.parallel_virtual_node 1.16% : 0.000003s : 14: predicate.partial_defer_inline 1.08% : 0.000003s : 19: predicate.partial_eliminate 0.94% : 0.000003s : 13: predicate.print_const_string_wrapper 0.71% : 0.000002s : 12: predicate.reduce_all_const_elim 1.26% : 0.000004s : 13: predicate.reduce_eliminate 2.19% : 0.000006s : 38: predicate.redundant_stop_gradient_eliminater 0.54% : 0.000002s : 12: predicate.remove_not_recompute_node 1.21% : 0.000003s : 25: predicate.replace_applicator 0.62% : 0.000002s : 12: predicate.replace_old_param 0.41% : 0.000001s : 6: predicate.reset_defer_inline 0.90% : 0.000003s : 13: predicate.reshape_eliminate 0.71% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.47% : 0.000001s : 6: predicate.row_tensor_eliminate 1.10% : 0.000003s : 12: predicate.same_eliminate 0.42% : 0.000001s : 12: predicate.set_cell_output_no_recompute 1.07% : 0.000003s : 12: predicate.shard_identity_eliminate 0.96% : 0.000003s : 12: predicate.special_op_eliminate 0.81% : 0.000002s : 12: predicate.specialize_transform 1.83% : 0.000005s : 12: predicate.split_environ_get_set_with_tuple_value 0.92% : 0.000003s : 12: predicate.stack_unstack_eliminate 0.51% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.81% : 0.000002s : 14: predicate.switch_defer_inline 1.37% : 0.000004s : 26: predicate.switch_layer_defer_inline 3.21% : 0.000009s : 49: predicate.switch_simplify 0.88% : 0.000002s : 13: predicate.tile_eliminate 0.94% : 0.000003s : 13: predicate.transpose_eliminate 1.69% : 0.000005s : 25: predicate.tuple_list_convert_item_index_to_positive 1.68% : 0.000005s : 25: predicate.tuple_list_get_item_const_eliminator 1.62% : 0.000005s : 25: predicate.tuple_list_get_item_depend_reorder 2.99% : 0.000008s : 37: predicate.tuple_list_get_item_eliminator 1.68% : 0.000005s : 25: predicate.tuple_list_get_set_item_eliminator 2.34% : 0.000007s : 37: predicate.tuple_list_set_item_eliminator 1.82% : 0.000005s : 25: predicate.tuple_to_list_eliminator_ 2.07% : 0.000006s : 38: predicate.updatestate_pure_node_eliminater 3.12% : 0.000009s : 50: predicate.updatestate_useless_node_eliminater 0.56% : 0.000002s : 6: predicate.value_based_eliminate 0.80% : 0.000002s : 12: predicate.virtual_dataset_eliminate 0.76% : 0.000002s : 12: predicate.virtual_output_eliminate 0.40% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.69% : 0.000002s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000512 4 5.43% : 0.000028s : 1: func_graph_cloner_run.FuncGraphClonerGraph 94.57% : 0.000484s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.799293 196 0.00% : 0.000004s : 1: ForceFp32Comm 38.20% : 0.305366s : 1: add_attr 38.20% : 0.305347s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000097s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.02% : 0.000181s : 1: auto_monad 0.01% : 0.000044s : 1: auto_monad_reorder 0.00% : 0.000009s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.07% : 0.000556s : 1: bootstrap 0.01% : 0.000048s : 1: cconv 0.00% : 0.000005s : 1: comm_op_add_attrs 0.00% : 0.000028s : 1: control_data_broadcast_order 0.00% : 0.000020s : 1: convert_after_rewriter 0.00% : 0.000037s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000013s : 1: environ_conv 0.00% : 0.000026s : 1: event_method 0.00% : 0.000023s : 1: execute 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000007s : 1: get_jit_bprop_graph 0.00% : 0.000010s : 1: graph_reusing 0.00% : 0.000007s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000005s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000008s : 1: insert-virtual-dataset 0.00% : 0.000007s : 1: interleave_parallel_branches 0.01% : 0.000049s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000012s : 1: label_micro_interleaved_index 0.11% : 0.000853s : 1: loop_unroll 0.00% : 0.000006s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.15% : 0.001190s : 1: mutable_eliminate 0.00% : 0.000011s : 1: offloading_packed_experts 0.00% : 0.000035s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000044s : 1: opt.transform.mutable_eliminate 0.19% : 0.001531s : 78: opt.transform.opt_a 0.01% : 0.000079s : 1: opt.transform.opt_after_cconv 0.01% : 0.000053s : 1: opt.transform.opt_after_jit_grad 0.03% : 0.000248s : 28: opt.transform.opt_b 0.01% : 0.000087s : 2: opt.transform.opt_trans_graph 0.01% : 0.000068s : 4: opt.transform.symbol_engine_opt 0.55% : 0.004430s : 1: opt_a 0.03% : 0.000221s : 1: opt_after_cconv 0.12% : 0.000935s : 1: opt_after_jit_grad 0.06% : 0.000453s : 1: opt_b 1.08% : 0.008594s : 1: optimize 0.00% : 0.000038s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000012s : 1: order_py_execute_after_rewriter 0.00% : 0.000037s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000012s : 1: overlap_grad_ring_attention 0.00% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000009s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000012s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000005s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000047s : 1: pre_auto_parallel 0.01% : 0.000045s : 1: py_interpret_to_execute 0.01% : 0.000040s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.00% : 0.000024s : 1: remove_dup_value 0.08% : 0.000619s : 1: renormalize.infer 0.08% : 0.000615s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.04% : 0.000324s : 1: rewriter_after_opt_a 0.01% : 0.000073s : 1: rewriter_before_opt_a 0.00% : 0.000007s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000006s : 1: split_layernorm_comm 0.00% : 0.000010s : 1: split_matmul_comm_elemetwise 0.00% : 0.000013s : 1: swap_dp_allreduce_reducescatter 0.02% : 0.000137s : 1: symbol_engine_optimizer 20.26% : 0.161976s : 1: task_emit 0.02% : 0.000128s : 1: tuple_transform 0.51% : 0.004084s : 1: type_inference 0.02% : 0.000133s : 1: validate TotalTime = 0.65829, [24] [bootstrap]: 0.00051672 [type_inference]: 0.161387 [event_method]: 6.286e-05 [auto_monad]: 0.00032671 [graph_reusing]: 1.035e-05 [inline]: 2.51998e-06 [add_attr]: 0.00439256, [1] [add_attr_with_inline]: 0.00437846, [1] [Cycle 1]: 0.00010982, [2] [tag_attr]: 4.713e-05 [meta_addattr_fg_expand]: 1.039e-05 [parallel-infer-symbol]: 3.58999e-06 [pre_auto_parallel]: 6.124e-05 [insert-virtual-dataset]: 2.48e-06 [parallel-infer-symbol-second]: 9.90025e-07 [dataset_repeat_opt]: 2.53e-06 [pipeline_split]: 1.67999e-06 [optimize]: 0.266078, [53] [py_interpret_to_execute]: 5.833e-05 [rewriter_before_opt_a]: 0.00015392 [opt_a]: 0.262218, [3] [Cycle 1]: 0.252818, [45] [expand_dump_flag]: 7.58001e-06 [switch_simplify]: 8.483e-05 [loop_unroll]: 7.122e-05 [a_1]: 0.00173643 [with_stream_mark]: 3.984e-05 [recompute_prepare]: 3.533e-05 [updatestate_depend_eliminate]: 1.38e-05 [updatestate_assign_eliminate]: 1.119e-05 [updatestate_loads_eliminate]: 1.014e-05 [parameter_eliminate]: 3.73001e-06 [a_2]: 0.00032 [accelerated_algorithm]: 4.421e-05 [shard]: 2.54999e-06 [meta_shard_fg_expand]: 5.77001e-06 [shard_inline]: 2.149e-05 [merge_send_recv]: 2.315e-05 [auto_parallel]: 1.645e-05 [parallel]: 2.354e-05 [flash_sp]: 1.875e-05 [merge_comm]: 1.432e-05 [allreduce_fusion]: 1.14e-05 [matmul_add_comm_reduction]: 4.153e-05 [allreduce_slice_to_reducescatter]: 6.09987e-07 [virtual_shard_identity]: 3.406e-05 [virtual_dataset]: 2.304e-05 [get_grad_eliminate_]: 2.335e-05 [virtual_output]: 2.227e-05 [merge_forward]: 1.279e-05 [cell_reuse_recompute_pass]: 1.71e-06 [offload_activation]: 2.41e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.871e-05 [merge_recompute_call_nodes]: 1.51998e-06 [before_grad]: 9.974e-05 [set_forward_comm_id_for_comm_node_pass]: 1.538e-05 [meta_fg_expand]: 0.00785739 [flash_sp_send_recv_attached]: 6.71e-06 [receive_attached]: 2.26e-06 [after_resolve]: 0.00015673 [a_after_grad]: 0.00028259 [renormalize]: 0.238799 [add_forward_monad_depend]: 2.814e-05 [auto_monad_grad]: 1.783e-05 [auto_monad_eliminator]: 0.00015515 [cse]: 0.00032529 [a_3]: 0.00180581 [Cycle 2]: 0.00810151, [45] [expand_dump_flag]: 6.38e-06 [switch_simplify]: 0.00011972 [loop_unroll]: 0.00010351 [a_1]: 0.00396659 [with_stream_mark]: 4.135e-05 [recompute_prepare]: 2.434e-05 [updatestate_depend_eliminate]: 4.438e-05 [updatestate_assign_eliminate]: 7.15e-06 [updatestate_loads_eliminate]: 5.91003e-06 [parameter_eliminate]: 3.33e-06 [a_2]: 0.00017542 [accelerated_algorithm]: 1.944e-05 [shard]: 2.17999e-06 [meta_shard_fg_expand]: 5.67999e-06 [shard_inline]: 1.238e-05 [merge_send_recv]: 1.481e-05 [auto_parallel]: 1.429e-05 [parallel]: 1.341e-05 [flash_sp]: 4.98001e-06 [merge_comm]: 6.63998e-06 [allreduce_fusion]: 6.33e-06 [matmul_add_comm_reduction]: 1.583e-05 [allreduce_slice_to_reducescatter]: 1.17e-06 [virtual_shard_identity]: 1.587e-05 [virtual_dataset]: 1.138e-05 [get_grad_eliminate_]: 1.137e-05 [virtual_output]: 1.18e-05 [merge_forward]: 7.71999e-06 [cell_reuse_recompute_pass]: 2.37001e-06 [offload_activation]: 1.751e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.393e-05 [merge_recompute_call_nodes]: 2.21998e-06 [before_grad]: 2.068e-05 [set_forward_comm_id_for_comm_node_pass]: 8.74998e-06 [meta_fg_expand]: 9.626e-05 [flash_sp_send_recv_attached]: 2.88998e-06 [receive_attached]: 3.09001e-06 [after_resolve]: 2.719e-05 [a_after_grad]: 1.856e-05 [renormalize]: 0.00219586 [add_forward_monad_depend]: 1.307e-05 [auto_monad_grad]: 3.35e-06 [auto_monad_eliminator]: 3.467e-05 [cse]: 0.0001819 [a_3]: 0.00011258 [Cycle 3]: 0.00127105, [45] [expand_dump_flag]: 3.3e-06 [switch_simplify]: 1.56e-05 [loop_unroll]: 1.285e-05 [a_1]: 0.00033458 [with_stream_mark]: 2.694e-05 [recompute_prepare]: 1.225e-05 [updatestate_depend_eliminate]: 8.19002e-06 [updatestate_assign_eliminate]: 6.34001e-06 [updatestate_loads_eliminate]: 6.33e-06 [parameter_eliminate]: 2.16e-06 [a_2]: 0.00015925 [accelerated_algorithm]: 1.827e-05 [shard]: 2.25002e-06 [meta_shard_fg_expand]: 3.88001e-06 [shard_inline]: 1.374e-05 [merge_send_recv]: 1.418e-05 [auto_parallel]: 1.465e-05 [parallel]: 1.285e-05 [flash_sp]: 1.70001e-06 [merge_comm]: 7.79002e-06 [allreduce_fusion]: 6.19999e-06 [matmul_add_comm_reduction]: 1.531e-05 [allreduce_slice_to_reducescatter]: 9.20001e-07 [virtual_shard_identity]: 1.472e-05 [virtual_dataset]: 1.361e-05 [get_grad_eliminate_]: 1.118e-05 [virtual_output]: 1.074e-05 [merge_forward]: 7.15998e-06 [cell_reuse_recompute_pass]: 3.6e-06 [offload_activation]: 1.833e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.546e-05 [merge_recompute_call_nodes]: 1.34e-06 [before_grad]: 2.175e-05 [set_forward_comm_id_for_comm_node_pass]: 7.74002e-06 [meta_fg_expand]: 5.04998e-06 [flash_sp_send_recv_attached]: 2.00002e-06 [receive_attached]: 2.21998e-06 [after_resolve]: 2.276e-05 [a_after_grad]: 1.831e-05 [renormalize]: 5.9983e-08 [add_forward_monad_depend]: 2.68e-06 [auto_monad_grad]: 2.49001e-06 [auto_monad_eliminator]: 2.547e-05 [cse]: 4.68e-05 [a_3]: 7.189e-05 [py_interpret_to_execute_after_opt_a]: 2.564e-05 [slice_cell_reuse_recomputed_activation]: 2.34001e-06 [rewriter_after_opt_a]: 9.005e-05 [convert_after_rewriter]: 1.354e-05 [order_py_execute_after_rewriter]: 8.05e-06 [mutable_eliminate]: 0.00091882 [opt_b]: 0.00044291, [1] [Cycle 1]: 0.00043071, [7] [b_1]: 0.00027328 [b_2]: 1.396e-05 [updatestate_depend_eliminate]: 1.543e-05 [updatestate_assign_eliminate]: 6.51999e-06 [updatestate_loads_eliminate]: 6.51999e-06 [renormalize]: 1.30001e-06 [cse]: 6.523e-05 [optimize_parallel_all_gather_comm]: 3.957e-05 [overlap_param_gather]: 2.46e-06 [cconv]: 4.534e-05 [loop_unroll]: 0.00080783 [opt_after_cconv]: 0.00024462, [1] [Cycle 1]: 0.00023453, [7] [c_1]: 9.508e-05 [parameter_eliminate]: 8.62e-06 [updatestate_depend_eliminate]: 1.55e-05 [updatestate_assign_eliminate]: 6.64999e-06 [updatestate_loads_eliminate]: 6.48e-06 [cse]: 5.771e-05 [renormalize]: 1.76003e-06 [remove_dup_value]: 2.346e-05 [tuple_transform]: 0.0001504, [1] [Cycle 1]: 0.00014434, [4] [d_1]: 0.00010694 [none_parameter_eliminate]: 2.06e-06 [renormalize]: 1.80007e-07 [switch_simplify]: 1.333e-05 [partial_unused_args_eliminate]: 2.61999e-06 [add_recomputation]: 0.00012162 [cse_after_recomputation]: 5.387e-05, [1] [Cycle 1]: 4.673e-05, [1] [cse]: 3.782e-05 [environ_conv]: 1.819e-05 [swap_dp_allreduce_reducescatter]: 1.082e-05 [bias_add_comm_swap]: 3.53999e-06 [label_micro_interleaved_index]: 8.48999e-06 [label_fine_grained_interleaved_index]: 2.75002e-06 [merge_cast_opt]: 1.75001e-06 [slice_recompute_activation]: 2.58e-06 [micro_interleaved_order_control]: 3.02002e-06 [assign_add_opt]: 1.62999e-06 [ForceFp32Comm]: 9.89996e-07 [remove_cast_before_assign_add]: 1.14e-06 [full_micro_interleaved_order_control]: 2.53003e-06 [reorder_send_recv_between_fp_bp]: 2.99001e-06 [comm_op_add_attrs]: 1.47001e-06 [add_comm_op_reuse_tag]: 1.21002e-06 [interleave_split_concat_branches]: 1.45001e-06 [interleave_parallel_branches]: 1.34e-06 [overlap_opt_shard_in_pipeline]: 1.51998e-06 [overlap_opt_shard_grad_in_pipeline]: 1.89999e-06 [control_data_broadcast_order]: 2.688e-05 [grouped_pairwise_exchange_alltoall]: 1.74e-06 [offloading_packed_experts]: 6.71e-06 [overlap_recompute_and_grad_model_parallel]: 8e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.36002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.49e-06 [overlap_recompute_comm]: 2.51e-06 [overlap_grad_ring_attention]: 8.22e-06 [overlap_grad_flash_sp]: 4.045e-05 [begin_end_overlap_inline]: 8.60018e-07 [split_matmul_comm_elemetwise]: 2.70002e-06 [split_layernorm_comm]: 2.21e-06 [handle_group_info]: 1.19e-06 [symbol_engine_optimizer]: 0.00016357, [1] [Cycle 1]: 0.00015418, [6] [build]: 1.963e-05 [elim_shapecalc]: 2.68e-05 [elim_not_effective]: 2.792e-05 [opt_reshape]: 1.574e-05 [fold_const_symbol]: 2.05e-05 [renormalize]: 1.80007e-07 [detach_backward]: 2.61999e-06 [pipeline_parallel_scheduler]: 1.74998e-06 [auto_monad_reorder]: 4.429e-05 [get_jit_bprop_graph]: 2.92002e-06 [rewriter_after_jit_bprop_graph]: 8.18001e-06 [opt_after_jit_grad]: 0.00090671 [validate]: 8.496e-05 [backend_pass]: 1.44e-06 [task_emit]: 0.223986 [execute]: 1.1e-05 Sums bootstrap : 0.000517s : 0.08% type_inference : 0.161387s : 24.77% event_method : 0.000063s : 0.01% auto_monad : 0.000327s : 0.05% graph_reusing : 0.000010s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000047s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000010s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000061s : 0.01% insert-virtual-dataset : 0.000002s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000058s : 0.01% optimize.rewriter_before_opt_a : 0.000154s : 0.02% optimize.opt_a.expand_dump_flag : 0.000017s : 0.00% optimize.opt_a.switch_simplify : 0.000220s : 0.03% optimize.opt_a.loop_unroll : 0.000188s : 0.03% optimize.opt_a.a_1 : 0.006038s : 0.93% optimize.opt_a.with_stream_mark : 0.000108s : 0.02% optimize.opt_a.recompute_prepare : 0.000072s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000066s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000025s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000022s : 0.00% optimize.opt_a.parameter_eliminate : 0.000009s : 0.00% optimize.opt_a.a_2 : 0.000655s : 0.10% optimize.opt_a.accelerated_algorithm : 0.000082s : 0.01% optimize.opt_a.shard : 0.000007s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000015s : 0.00% optimize.opt_a.shard_inline : 0.000048s : 0.01% optimize.opt_a.merge_send_recv : 0.000052s : 0.01% optimize.opt_a.auto_parallel : 0.000045s : 0.01% optimize.opt_a.parallel : 0.000050s : 0.01% optimize.opt_a.flash_sp : 0.000025s : 0.00% optimize.opt_a.merge_comm : 0.000029s : 0.00% optimize.opt_a.allreduce_fusion : 0.000024s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000073s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000003s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000065s : 0.01% optimize.opt_a.virtual_dataset : 0.000048s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000046s : 0.01% optimize.opt_a.virtual_output : 0.000045s : 0.01% optimize.opt_a.merge_forward : 0.000028s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000008s : 0.00% optimize.opt_a.offload_activation : 0.000060s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000098s : 0.02% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000142s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000032s : 0.00% optimize.opt_a.meta_fg_expand : 0.007959s : 1.22% optimize.opt_a.flash_sp_send_recv_attached : 0.000012s : 0.00% optimize.opt_a.receive_attached : 0.000008s : 0.00% optimize.opt_a.after_resolve : 0.000207s : 0.03% optimize.opt_a.a_after_grad : 0.000319s : 0.05% optimize.opt_a.renormalize : 0.240995s : 36.99% optimize.opt_a.add_forward_monad_depend : 0.000044s : 0.01% optimize.opt_a.auto_monad_grad : 0.000024s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000215s : 0.03% optimize.opt_a.cse : 0.000554s : 0.09% optimize.opt_a.a_3 : 0.001990s : 0.31% optimize.py_interpret_to_execute_after_opt_a : 0.000026s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000090s : 0.01% optimize.convert_after_rewriter : 0.000014s : 0.00% optimize.order_py_execute_after_rewriter : 0.000008s : 0.00% optimize.mutable_eliminate : 0.000919s : 0.14% optimize.opt_b.b_1 : 0.000273s : 0.04% optimize.opt_b.b_2 : 0.000014s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000015s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000007s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000007s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000065s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000040s : 0.01% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000045s : 0.01% optimize.loop_unroll : 0.000808s : 0.12% optimize.opt_after_cconv.c_1 : 0.000095s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000009s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000016s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.cse : 0.000058s : 0.01% optimize.opt_after_cconv.renormalize : 0.000002s : 0.00% optimize.remove_dup_value : 0.000023s : 0.00% optimize.tuple_transform.d_1 : 0.000107s : 0.02% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000013s : 0.00% optimize.partial_unused_args_eliminate : 0.000003s : 0.00% optimize.add_recomputation : 0.000122s : 0.02% optimize.cse_after_recomputation.cse : 0.000038s : 0.01% optimize.environ_conv : 0.000018s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000011s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000008s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000027s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000008s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000008s : 0.00% optimize.overlap_grad_flash_sp : 0.000040s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000020s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000027s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000028s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000016s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000021s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000044s : 0.01% get_jit_bprop_graph : 0.000003s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.000907s : 0.14% validate : 0.000085s : 0.01% backend_pass : 0.000001s : 0.00% task_emit : 0.223986s : 34.38% execute : 0.000011s : 0.00% Time group info: ------[substitution.] 0.003246 310 0.53% : 0.000017s : 4: substitution.depend_value_elim 0.12% : 0.000004s : 7: substitution.elim_not_effective 0.25% : 0.000008s : 7: substitution.float_depend_g_call 0.16% : 0.000005s : 2: substitution.float_tuple_getitem_switch 0.10% : 0.000003s : 7: substitution.fold_const_symbol 44.49% : 0.001444s : 4: substitution.getattr_setattr_resolve 0.31% : 0.000010s : 9: substitution.graph_param_transform 0.08% : 0.000003s : 2: substitution.incorporate_call 0.06% : 0.000002s : 2: substitution.incorporate_call_switch 35.96% : 0.001167s : 24: substitution.inline 1.81% : 0.000059s : 5: substitution.inline_without_move 0.53% : 0.000017s : 28: substitution.j_node_and_user_rematch 0.71% : 0.000023s : 3: substitution.less_batch_normalization 0.50% : 0.000016s : 9: substitution.minmaximum_grad 0.31% : 0.000010s : 7: substitution.partial_eliminate 0.65% : 0.000021s : 28: substitution.remove_not_recompute_node 3.65% : 0.000118s : 38: substitution.replace_applicator 0.70% : 0.000023s : 19: substitution.replace_old_param 0.13% : 0.000004s : 1: substitution.set_cell_output_no_recompute 1.01% : 0.000033s : 9: substitution.tuple_list_convert_item_index_to_positive 0.44% : 0.000014s : 9: substitution.tuple_list_get_item_const_eliminator 0.70% : 0.000023s : 9: substitution.tuple_list_get_item_depend_reorder 3.18% : 0.000103s : 31: substitution.tuple_list_get_item_eliminator 0.60% : 0.000020s : 9: substitution.tuple_list_get_set_item_eliminator 1.26% : 0.000041s : 16: substitution.updatestate_pure_node_eliminater 1.77% : 0.000058s : 21: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.161279 2 98.99% : 0.159657s : 1: type_inference.infer 1.01% : 0.001621s : 1: type_inference.specialize ------[replace.] 0.000636 49 11.98% : 0.000076s : 3: replace.getattr_setattr_resolve 39.50% : 0.000251s : 24: replace.inline 12.27% : 0.000078s : 1: replace.replace_applicator 33.79% : 0.000215s : 20: replace.tuple_list_get_item_eliminator 2.45% : 0.000016s : 1: replace.updatestate_useless_node_eliminater ------[match.] 0.002640 49 52.79% : 0.001394s : 3: match.getattr_setattr_resolve 43.43% : 0.001147s : 24: match.inline 0.82% : 0.000022s : 1: match.replace_applicator 2.26% : 0.000060s : 20: match.tuple_list_get_item_eliminator 0.70% : 0.000018s : 1: match.updatestate_useless_node_eliminater ------[predicate.] 0.001367 8683 0.84% : 0.000011s : 85: predicate.accumulaten_eliminater 0.46% : 0.000006s : 9: predicate.ad_related_special_op_eliminate 0.37% : 0.000005s : 37: predicate.addn_check_dump 0.90% : 0.000012s : 85: predicate.addn_zero_filter 0.82% : 0.000011s : 85: predicate.adjust_all_reduce_mul_add 1.75% : 0.000024s : 122: predicate.arithmetic_simplify 0.95% : 0.000013s : 85: predicate.cast_eliminate 2.52% : 0.000034s : 260: predicate.check_bprop_eliminate 0.41% : 0.000006s : 37: predicate.compare_switch_simplify 0.05% : 0.000001s : 9: predicate.const_output_eliminate 0.41% : 0.000006s : 37: predicate.depend_value_elim 0.97% : 0.000013s : 85: predicate.dict_get_item_const_eliminator 0.99% : 0.000014s : 85: predicate.dict_get_item_eliminator 0.99% : 0.000014s : 85: predicate.dict_set_item_eliminator 0.33% : 0.000004s : 18: predicate.dumpgradient_eliminate 0.10% : 0.000001s : 9: predicate.elim_not_effective 0.13% : 0.000002s : 9: predicate.elim_shapecalc_of_broadcastargs 0.90% : 0.000012s : 94: predicate.environ_add_const_eliminate 0.90% : 0.000012s : 94: predicate.environ_get_add_eliminate 0.92% : 0.000013s : 94: predicate.environ_get_depend_swap 1.31% : 0.000018s : 131: predicate.environ_get_eliminate 0.93% : 0.000013s : 94: predicate.environ_get_set_eliminate 1.33% : 0.000018s : 130: predicate.exchange_switch_depend_value 1.94% : 0.000027s : 130: predicate.float_depend_g_call 0.38% : 0.000005s : 37: predicate.float_environ_get_switch 0.50% : 0.000007s : 46: predicate.float_tuple_getitem_switch 0.06% : 0.000001s : 9: predicate.fold_const_symbol 0.44% : 0.000006s : 37: predicate.get_grad_eliminate 0.48% : 0.000006s : 20: predicate.getattr_setattr_resolve 0.07% : 0.000001s : 9: predicate.graph_param_transform 0.37% : 0.000005s : 37: predicate.incorporate_call 0.35% : 0.000005s : 37: predicate.incorporate_call_switch 4.65% : 0.000064s : 307: predicate.inline 1.66% : 0.000023s : 111: predicate.inline_without_move 0.21% : 0.000003s : 37: predicate.j_node_and_user_rematch 0.50% : 0.000007s : 37: predicate.less_batch_normalization 1.62% : 0.000022s : 123: predicate.list_to_tuple_eliminator_ 2.02% : 0.000028s : 208: predicate.load_eliminater 0.56% : 0.000008s : 9: predicate.loop_unroll_after_grad 1.95% : 0.000027s : 197: predicate.loop_unroll_before_grad 1.13% : 0.000015s : 103: predicate.make_slice_get_slice_eliminator 0.41% : 0.000006s : 37: predicate.merge_addn 2.43% : 0.000033s : 256: predicate.micro_step_allgather_replace 2.44% : 0.000033s : 256: predicate.mini_step_allgather_replace 0.86% : 0.000012s : 85: predicate.minmaximum_grad 0.36% : 0.000005s : 9: predicate.mutable_eliminate 0.12% : 0.000002s : 9: predicate.opt_reshape 0.13% : 0.000002s : 9: predicate.parallel_virtual_node 2.06% : 0.000028s : 130: predicate.partial_defer_inline 1.30% : 0.000018s : 114: predicate.partial_eliminate 0.87% : 0.000012s : 85: predicate.print_const_string_wrapper 0.41% : 0.000006s : 37: predicate.reduce_all_const_elim 1.13% : 0.000015s : 85: predicate.reduce_eliminate 2.09% : 0.000029s : 208: predicate.redundant_stop_gradient_eliminater 0.24% : 0.000003s : 37: predicate.remove_not_recompute_node 2.44% : 0.000033s : 363: predicate.replace_applicator 0.77% : 0.000011s : 111: predicate.replace_old_param 0.12% : 0.000002s : 9: predicate.reset_defer_inline 0.90% : 0.000012s : 85: predicate.reshape_eliminate 2.70% : 0.000037s : 256: predicate.row_tensor_add_zeros_like 0.14% : 0.000002s : 9: predicate.row_tensor_eliminate 2.89% : 0.000040s : 260: predicate.same_eliminate 0.27% : 0.000004s : 39: predicate.set_cell_output_no_recompute 0.76% : 0.000010s : 37: predicate.shard_identity_eliminate 0.26% : 0.000004s : 18: predicate.special_op_eliminate 0.51% : 0.000007s : 37: predicate.specialize_transform 2.75% : 0.000038s : 256: predicate.split_environ_get_set_with_tuple_value 4.87% : 0.000067s : 111: predicate.stack_unstack_eliminate 0.12% : 0.000002s : 9: predicate.switch_call_monad_eliminater 1.51% : 0.000021s : 130: predicate.switch_defer_inline 3.89% : 0.000053s : 390: predicate.switch_layer_defer_inline 4.20% : 0.000057s : 373: predicate.switch_simplify 0.88% : 0.000012s : 85: predicate.tile_eliminate 0.88% : 0.000012s : 85: predicate.transpose_eliminate 1.23% : 0.000017s : 103: predicate.tuple_list_convert_item_index_to_positive 1.28% : 0.000017s : 103: predicate.tuple_list_get_item_const_eliminator 1.08% : 0.000015s : 103: predicate.tuple_list_get_item_depend_reorder 2.41% : 0.000033s : 160: predicate.tuple_list_get_item_eliminator 1.15% : 0.000016s : 103: predicate.tuple_list_get_set_item_eliminator 1.67% : 0.000023s : 140: predicate.tuple_list_set_item_eliminator 1.29% : 0.000018s : 123: predicate.tuple_to_list_eliminator_ 2.09% : 0.000029s : 208: predicate.updatestate_pure_node_eliminater 2.68% : 0.000037s : 246: predicate.updatestate_useless_node_eliminater 0.12% : 0.000002s : 9: predicate.value_based_eliminate 0.47% : 0.000006s : 37: predicate.virtual_dataset_eliminate 0.43% : 0.000006s : 37: predicate.virtual_output_eliminate 0.11% : 0.000002s : 9: predicate.virtual_view_grad_eliminate 0.14% : 0.000002s : 9: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.004413 53 52.91% : 0.002335s : 23: func_graph_cloner_run.FuncGraphClonerGraph 47.09% : 0.002078s : 30: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.181905 247 0.00% : 0.000005s : 1: ForceFp32Comm 0.37% : 0.004401s : 1: add_attr 0.37% : 0.004383s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000130s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.03% : 0.000339s : 1: auto_monad 0.00% : 0.000052s : 1: auto_monad_reorder 0.00% : 0.000010s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.05% : 0.000556s : 1: bootstrap 0.00% : 0.000051s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000031s : 1: control_data_broadcast_order 0.00% : 0.000018s : 1: convert_after_rewriter 0.00% : 0.000057s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000008s : 1: detach_backward 0.00% : 0.000022s : 1: environ_conv 0.01% : 0.000072s : 1: event_method 0.00% : 0.000020s : 1: execute 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000009s : 1: get_jit_bprop_graph 0.00% : 0.000015s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000012s : 1: label_micro_interleaved_index 0.07% : 0.000829s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.08% : 0.000941s : 1: mutable_eliminate 0.00% : 0.000010s : 1: offloading_packed_experts 0.00% : 0.000040s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000034s : 1: opt.transform.mutable_eliminate 0.85% : 0.010048s : 125: opt.transform.opt_a 0.01% : 0.000093s : 1: opt.transform.opt_after_cconv 0.01% : 0.000067s : 1: opt.transform.opt_after_jit_grad 0.02% : 0.000250s : 28: opt.transform.opt_b 0.13% : 0.001593s : 2: opt.transform.opt_resolve 0.01% : 0.000117s : 2: opt.transform.opt_trans_graph 0.01% : 0.000083s : 4: opt.transform.symbol_engine_opt 22.19% : 0.262222s : 1: opt_a 0.02% : 0.000250s : 1: opt_after_cconv 0.08% : 0.000933s : 1: opt_after_jit_grad 0.04% : 0.000448s : 1: opt_b 22.51% : 0.266086s : 1: optimize 0.00% : 0.000045s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000011s : 1: order_py_execute_after_rewriter 0.00% : 0.000046s : 1: overlap_grad_flash_sp 0.00% : 0.000006s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000012s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000006s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000012s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000004s : 1: pipeline_split 0.01% : 0.000066s : 1: pre_auto_parallel 0.01% : 0.000063s : 1: py_interpret_to_execute 0.00% : 0.000030s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000029s : 1: remove_dup_value 1.75% : 0.020706s : 2: renormalize.infer 18.64% : 0.220252s : 2: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000098s : 1: rewriter_after_opt_a 0.01% : 0.000159s : 1: rewriter_before_opt_a 0.00% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000007s : 1: split_matmul_comm_elemetwise 0.00% : 0.000014s : 1: swap_dp_allreduce_reducescatter 0.01% : 0.000167s : 1: symbol_engine_optimizer 18.95% : 0.224011s : 1: task_emit 0.01% : 0.000155s : 1: tuple_transform 13.66% : 0.161421s : 1: type_inference 0.02% : 0.000187s : 1: validate . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[0-shape5] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[0-shape5],max_mem:4.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:42:25.772.267 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.18171, [24] [bootstrap]: 0.00085632 [type_inference]: 0.108512 [event_method]: 3.075e-05 [auto_monad]: 0.00037485 [graph_reusing]: 8.82e-06 [inline]: 2.90002e-06 [add_attr]: 0.0173935, [1] [add_attr_with_inline]: 0.0173714, [1] [Cycle 1]: 0.00010373, [2] [tag_attr]: 2.851e-05 [meta_addattr_fg_expand]: 3.95998e-06 [parallel-infer-symbol]: 4.63999e-06 [pre_auto_parallel]: 5.4e-05 [insert-virtual-dataset]: 2.39999e-06 [parallel-infer-symbol-second]: 1.14003e-06 [dataset_repeat_opt]: 2.48e-06 [pipeline_split]: 1.69e-06 [optimize]: 0.00952453, [53] [py_interpret_to_execute]: 5.078e-05 [rewriter_before_opt_a]: 8.756e-05 [opt_a]: 0.00537253, [2] [Cycle 1]: 0.00394632, [45] [expand_dump_flag]: 3.73001e-06 [switch_simplify]: 4.577e-05 [loop_unroll]: 2.223e-05 [a_1]: 0.00091644 [with_stream_mark]: 3.161e-05 [recompute_prepare]: 2.087e-05 [updatestate_depend_eliminate]: 1.332e-05 [updatestate_assign_eliminate]: 9.85002e-06 [updatestate_loads_eliminate]: 5.94e-06 [parameter_eliminate]: 2.22999e-06 [a_2]: 0.00019603 [accelerated_algorithm]: 2.126e-05 [shard]: 6.24001e-06 [meta_shard_fg_expand]: 2.61e-06 [shard_inline]: 1.67e-05 [merge_send_recv]: 1.91e-05 [auto_parallel]: 1.906e-05 [parallel]: 4.575e-05 [flash_sp]: 1.32e-05 [merge_comm]: 6.48998e-06 [allreduce_fusion]: 6.26e-06 [matmul_add_comm_reduction]: 1.52e-05 [allreduce_slice_to_reducescatter]: 8.30012e-07 [virtual_shard_identity]: 2.219e-05 [virtual_dataset]: 2.547e-05 [get_grad_eliminate_]: 1.217e-05 [virtual_output]: 1.472e-05 [merge_forward]: 7.36001e-06 [cell_reuse_recompute_pass]: 2.36e-06 [offload_activation]: 3.879e-05 [cell_reuse_handle_not_recompute_node_pass]: 5.42e-05 [merge_recompute_call_nodes]: 1.73002e-06 [before_grad]: 4.776e-05 [set_forward_comm_id_for_comm_node_pass]: 1.461e-05 [meta_fg_expand]: 4.48999e-06 [flash_sp_send_recv_attached]: 3.61001e-06 [receive_attached]: 2.96001e-06 [after_resolve]: 2.428e-05 [a_after_grad]: 2.8e-05 [renormalize]: 0.0013287 [add_forward_monad_depend]: 1.38e-05 [auto_monad_grad]: 2.91e-06 [auto_monad_eliminator]: 3.187e-05 [cse]: 4.813e-05 [a_3]: 7.973e-05 [Cycle 2]: 0.00139944, [45] [expand_dump_flag]: 2.35002e-06 [switch_simplify]: 1.648e-05 [loop_unroll]: 1.032e-05 [a_1]: 0.00023333 [with_stream_mark]: 2.448e-05 [recompute_prepare]: 1.625e-05 [updatestate_depend_eliminate]: 1.12e-05 [updatestate_assign_eliminate]: 9.32001e-06 [updatestate_loads_eliminate]: 5.89e-06 [parameter_eliminate]: 1.94e-06 [a_2]: 0.00012301 [accelerated_algorithm]: 1.198e-05 [shard]: 2.59999e-06 [meta_shard_fg_expand]: 3.04999e-06 [shard_inline]: 8.90001e-06 [merge_send_recv]: 1.768e-05 [auto_parallel]: 1.551e-05 [parallel]: 1.486e-05 [flash_sp]: 4.83001e-06 [merge_comm]: 1.077e-05 [allreduce_fusion]: 9.62001e-06 [matmul_add_comm_reduction]: 1.393e-05 [allreduce_slice_to_reducescatter]: 8.29983e-07 [virtual_shard_identity]: 1.434e-05 [virtual_dataset]: 9.37001e-06 [get_grad_eliminate_]: 8.92e-06 [virtual_output]: 9.37999e-06 [merge_forward]: 1.179e-05 [cell_reuse_recompute_pass]: 2.77002e-06 [offload_activation]: 2.45e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.413e-05 [merge_recompute_call_nodes]: 1.73002e-06 [before_grad]: 1.763e-05 [set_forward_comm_id_for_comm_node_pass]: 1.015e-05 [meta_fg_expand]: 3.64002e-06 [flash_sp_send_recv_attached]: 2.22999e-06 [receive_attached]: 2.46998e-06 [after_resolve]: 2.153e-05 [a_after_grad]: 1.438e-05 [renormalize]: 7.00238e-08 [add_forward_monad_depend]: 3.85e-06 [auto_monad_grad]: 2.41e-06 [auto_monad_eliminator]: 2.546e-05 [cse]: 3.179e-05 [a_3]: 5.8e-05 [py_interpret_to_execute_after_opt_a]: 2.925e-05 [slice_cell_reuse_recomputed_activation]: 2.27999e-06 [rewriter_after_opt_a]: 0.00025414 [convert_after_rewriter]: 1.995e-05 [order_py_execute_after_rewriter]: 7.28999e-06 [mutable_eliminate]: 0.00106743 [opt_b]: 0.00049035, [1] [Cycle 1]: 0.0004738, [7] [b_1]: 0.00027616 [b_2]: 2.192e-05 [updatestate_depend_eliminate]: 1.483e-05 [updatestate_assign_eliminate]: 6.21998e-06 [updatestate_loads_eliminate]: 5.77999e-06 [renormalize]: 7.2e-07 [cse]: 4.053e-05 [optimize_parallel_all_gather_comm]: 3.101e-05 [overlap_param_gather]: 3.77002e-06 [cconv]: 3.968e-05 [loop_unroll]: 0.00072274 [opt_after_cconv]: 0.00022534, [1] [Cycle 1]: 0.0002138, [7] [c_1]: 7.567e-05 [parameter_eliminate]: 6.23e-06 [updatestate_depend_eliminate]: 1.272e-05 [updatestate_assign_eliminate]: 6.39999e-06 [updatestate_loads_eliminate]: 5.34e-06 [cse]: 4.593e-05 [renormalize]: 7.59988e-07 [remove_dup_value]: 1.908e-05 [tuple_transform]: 0.00016128, [1] [Cycle 1]: 0.00014962, [4] [d_1]: 8.993e-05 [none_parameter_eliminate]: 1.97001e-06 [renormalize]: 1.8999e-07 [switch_simplify]: 1.187e-05 [partial_unused_args_eliminate]: 2.13002e-06 [add_recomputation]: 0.0001255 [cse_after_recomputation]: 4.449e-05, [1] [Cycle 1]: 3.759e-05, [1] [cse]: 2.209e-05 [environ_conv]: 9.13002e-06 [swap_dp_allreduce_reducescatter]: 1.45e-05 [bias_add_comm_swap]: 3.3e-06 [label_micro_interleaved_index]: 7.27002e-06 [label_fine_grained_interleaved_index]: 3.4e-06 [merge_cast_opt]: 1.53002e-06 [slice_recompute_activation]: 2.12001e-06 [micro_interleaved_order_control]: 2.41e-06 [assign_add_opt]: 1.74e-06 [ForceFp32Comm]: 9.00007e-07 [remove_cast_before_assign_add]: 9.20001e-07 [full_micro_interleaved_order_control]: 2.63e-06 [reorder_send_recv_between_fp_bp]: 2.99001e-06 [comm_op_add_attrs]: 1.11002e-06 [add_comm_op_reuse_tag]: 9.39996e-07 [interleave_split_concat_branches]: 4.94e-06 [interleave_parallel_branches]: 1.34e-06 [overlap_opt_shard_in_pipeline]: 2.61e-06 [overlap_opt_shard_grad_in_pipeline]: 2.02999e-06 [control_data_broadcast_order]: 2.643e-05 [grouped_pairwise_exchange_alltoall]: 1.69e-06 [offloading_packed_experts]: 6.36e-06 [overlap_recompute_and_grad_model_parallel]: 1.439e-05 [overlap_grad_matmul_and_grad_allreduce]: 1.72999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.69e-06 [overlap_recompute_comm]: 2.66999e-06 [overlap_grad_ring_attention]: 7e-06 [overlap_grad_flash_sp]: 3.675e-05 [begin_end_overlap_inline]: 5.00004e-07 [split_matmul_comm_elemetwise]: 2.44001e-06 [split_layernorm_comm]: 2.01998e-06 [handle_group_info]: 1.27e-06 [symbol_engine_optimizer]: 0.00017023, [1] [Cycle 1]: 0.00016268, [6] [build]: 6.29001e-06 [elim_shapecalc]: 2.571e-05 [elim_not_effective]: 2.711e-05 [opt_reshape]: 1.368e-05 [fold_const_symbol]: 2.087e-05 [renormalize]: 4.09986e-07 [detach_backward]: 2.34999e-06 [pipeline_parallel_scheduler]: 1.49e-06 [auto_monad_reorder]: 3.809e-05 [get_jit_bprop_graph]: 2.12001e-06 [rewriter_after_jit_bprop_graph]: 7.97998e-06 [opt_after_jit_grad]: 0.00082925 [validate]: 6.885e-05 [backend_pass]: 1.09e-06 [task_emit]: 0.0436047 [execute]: 1.633e-05 Sums bootstrap : 0.000856s : 0.53% type_inference : 0.108512s : 67.00% event_method : 0.000031s : 0.02% auto_monad : 0.000375s : 0.23% graph_reusing : 0.000009s : 0.01% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000029s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000005s : 0.00% pre_auto_parallel : 0.000054s : 0.03% insert-virtual-dataset : 0.000002s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000051s : 0.03% optimize.rewriter_before_opt_a : 0.000088s : 0.05% optimize.opt_a.expand_dump_flag : 0.000006s : 0.00% optimize.opt_a.switch_simplify : 0.000062s : 0.04% optimize.opt_a.loop_unroll : 0.000033s : 0.02% optimize.opt_a.a_1 : 0.001150s : 0.71% optimize.opt_a.with_stream_mark : 0.000056s : 0.03% optimize.opt_a.recompute_prepare : 0.000037s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000025s : 0.02% optimize.opt_a.updatestate_assign_eliminate : 0.000019s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000012s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.00% optimize.opt_a.a_2 : 0.000319s : 0.20% optimize.opt_a.accelerated_algorithm : 0.000033s : 0.02% optimize.opt_a.shard : 0.000009s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000006s : 0.00% optimize.opt_a.shard_inline : 0.000026s : 0.02% optimize.opt_a.merge_send_recv : 0.000037s : 0.02% optimize.opt_a.auto_parallel : 0.000035s : 0.02% optimize.opt_a.parallel : 0.000061s : 0.04% optimize.opt_a.flash_sp : 0.000018s : 0.01% optimize.opt_a.merge_comm : 0.000017s : 0.01% optimize.opt_a.allreduce_fusion : 0.000016s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000029s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000037s : 0.02% optimize.opt_a.virtual_dataset : 0.000035s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000021s : 0.01% optimize.opt_a.virtual_output : 0.000024s : 0.01% optimize.opt_a.merge_forward : 0.000019s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000063s : 0.04% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000078s : 0.05% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000065s : 0.04% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000025s : 0.02% optimize.opt_a.meta_fg_expand : 0.000008s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000006s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000046s : 0.03% optimize.opt_a.a_after_grad : 0.000042s : 0.03% optimize.opt_a.renormalize : 0.001329s : 0.82% optimize.opt_a.add_forward_monad_depend : 0.000018s : 0.01% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000057s : 0.04% optimize.opt_a.cse : 0.000080s : 0.05% optimize.opt_a.a_3 : 0.000138s : 0.09% optimize.py_interpret_to_execute_after_opt_a : 0.000029s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000254s : 0.16% optimize.convert_after_rewriter : 0.000020s : 0.01% optimize.order_py_execute_after_rewriter : 0.000007s : 0.00% optimize.mutable_eliminate : 0.001067s : 0.66% optimize.opt_b.b_1 : 0.000276s : 0.17% optimize.opt_b.b_2 : 0.000022s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000015s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000041s : 0.03% optimize.optimize_parallel_all_gather_comm : 0.000031s : 0.02% optimize.overlap_param_gather : 0.000004s : 0.00% optimize.cconv : 0.000040s : 0.02% optimize.loop_unroll : 0.000723s : 0.45% optimize.opt_after_cconv.c_1 : 0.000076s : 0.05% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000013s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.cse : 0.000046s : 0.03% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000019s : 0.01% optimize.tuple_transform.d_1 : 0.000090s : 0.06% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000012s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000125s : 0.08% optimize.cse_after_recomputation.cse : 0.000022s : 0.01% optimize.environ_conv : 0.000009s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000014s : 0.01% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000007s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000005s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000003s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000026s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000006s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000014s : 0.01% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000007s : 0.00% optimize.overlap_grad_flash_sp : 0.000037s : 0.02% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000006s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000026s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000027s : 0.02% optimize.symbol_engine_optimizer.opt_reshape : 0.000014s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000021s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000001s : 0.00% auto_monad_reorder : 0.000038s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.000829s : 0.51% validate : 0.000069s : 0.04% backend_pass : 0.000001s : 0.00% task_emit : 0.043605s : 26.92% execute : 0.000016s : 0.01% Time group info: ------[substitution.] 0.000257 55 3.91% : 0.000010s : 2: substitution.depend_value_elim 1.20% : 0.000003s : 5: substitution.elim_not_effective 0.95% : 0.000002s : 5: substitution.fold_const_symbol 3.15% : 0.000008s : 6: substitution.graph_param_transform 61.22% : 0.000157s : 1: substitution.inline 2.72% : 0.000007s : 10: substitution.j_node_and_user_rematch 4.01% : 0.000010s : 10: substitution.remove_not_recompute_node 2.60% : 0.000007s : 2: substitution.replace_old_param 10.57% : 0.000027s : 6: substitution.updatestate_pure_node_eliminater 9.67% : 0.000025s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.108420 2 98.65% : 0.106960s : 1: type_inference.infer 1.35% : 0.001460s : 1: type_inference.specialize ------[replace.] 0.000020 1 100.00% : 0.000020s : 1: replace.inline ------[match.] 0.000155 1 100.00% : 0.000155s : 1: match.inline ------[predicate.] 0.000277 1419 1.06% : 0.000003s : 13: predicate.accumulaten_eliminater 1.31% : 0.000004s : 6: predicate.ad_related_special_op_eliminate 0.72% : 0.000002s : 12: predicate.addn_check_dump 0.84% : 0.000002s : 13: predicate.addn_zero_filter 0.69% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 2.20% : 0.000006s : 25: predicate.arithmetic_simplify 0.85% : 0.000002s : 13: predicate.cast_eliminate 0.86% : 0.000002s : 12: predicate.check_bprop_eliminate 0.66% : 0.000002s : 12: predicate.compare_switch_simplify 0.21% : 0.000001s : 6: predicate.const_output_eliminate 0.79% : 0.000002s : 12: predicate.depend_value_elim 0.90% : 0.000003s : 13: predicate.dict_get_item_const_eliminator 0.90% : 0.000002s : 13: predicate.dict_get_item_eliminator 0.78% : 0.000002s : 13: predicate.dict_set_item_eliminator 1.46% : 0.000004s : 12: predicate.dumpgradient_eliminate 0.27% : 0.000001s : 6: predicate.elim_not_effective 0.62% : 0.000002s : 6: predicate.elim_shapecalc_of_broadcastargs 1.14% : 0.000003s : 19: predicate.environ_add_const_eliminate 1.27% : 0.000003s : 19: predicate.environ_get_add_eliminate 1.14% : 0.000003s : 19: predicate.environ_get_depend_swap 1.76% : 0.000005s : 31: predicate.environ_get_eliminate 1.06% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.78% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.52% : 0.000004s : 14: predicate.float_depend_g_call 0.68% : 0.000002s : 12: predicate.float_environ_get_switch 1.02% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.20% : 0.000001s : 6: predicate.fold_const_symbol 0.99% : 0.000003s : 12: predicate.get_grad_eliminate 0.21% : 0.000001s : 6: predicate.graph_param_transform 0.70% : 0.000002s : 12: predicate.incorporate_call 0.61% : 0.000002s : 12: predicate.incorporate_call_switch 5.46% : 0.000015s : 63: predicate.inline 1.20% : 0.000003s : 12: predicate.inline_without_move 0.41% : 0.000001s : 12: predicate.j_node_and_user_rematch 1.07% : 0.000003s : 12: predicate.less_batch_normalization 1.77% : 0.000005s : 25: predicate.list_to_tuple_eliminator_ 2.23% : 0.000006s : 38: predicate.load_eliminater 1.46% : 0.000004s : 6: predicate.loop_unroll_after_grad 1.09% : 0.000003s : 17: predicate.loop_unroll_before_grad 2.03% : 0.000006s : 25: predicate.make_slice_get_slice_eliminator 0.81% : 0.000002s : 12: predicate.merge_addn 0.67% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.71% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.72% : 0.000002s : 13: predicate.minmaximum_grad 1.74% : 0.000005s : 6: predicate.mutable_eliminate 0.64% : 0.000002s : 6: predicate.opt_reshape 0.44% : 0.000001s : 6: predicate.parallel_virtual_node 1.28% : 0.000004s : 14: predicate.partial_defer_inline 1.12% : 0.000003s : 19: predicate.partial_eliminate 0.91% : 0.000003s : 13: predicate.print_const_string_wrapper 0.77% : 0.000002s : 12: predicate.reduce_all_const_elim 1.06% : 0.000003s : 13: predicate.reduce_eliminate 2.06% : 0.000006s : 38: predicate.redundant_stop_gradient_eliminater 0.57% : 0.000002s : 12: predicate.remove_not_recompute_node 1.02% : 0.000003s : 25: predicate.replace_applicator 0.69% : 0.000002s : 12: predicate.replace_old_param 0.38% : 0.000001s : 6: predicate.reset_defer_inline 0.80% : 0.000002s : 13: predicate.reshape_eliminate 0.79% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.56% : 0.000002s : 6: predicate.row_tensor_eliminate 1.06% : 0.000003s : 12: predicate.same_eliminate 0.63% : 0.000002s : 12: predicate.set_cell_output_no_recompute 1.25% : 0.000003s : 12: predicate.shard_identity_eliminate 0.93% : 0.000003s : 12: predicate.special_op_eliminate 0.76% : 0.000002s : 12: predicate.specialize_transform 1.18% : 0.000003s : 12: predicate.split_environ_get_set_with_tuple_value 1.27% : 0.000004s : 12: predicate.stack_unstack_eliminate 0.43% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.96% : 0.000003s : 14: predicate.switch_defer_inline 1.60% : 0.000004s : 26: predicate.switch_layer_defer_inline 3.78% : 0.000010s : 49: predicate.switch_simplify 0.84% : 0.000002s : 13: predicate.tile_eliminate 0.85% : 0.000002s : 13: predicate.transpose_eliminate 1.70% : 0.000005s : 25: predicate.tuple_list_convert_item_index_to_positive 1.63% : 0.000005s : 25: predicate.tuple_list_get_item_const_eliminator 1.70% : 0.000005s : 25: predicate.tuple_list_get_item_depend_reorder 2.77% : 0.000008s : 37: predicate.tuple_list_get_item_eliminator 1.45% : 0.000004s : 25: predicate.tuple_list_get_set_item_eliminator 2.39% : 0.000007s : 37: predicate.tuple_list_set_item_eliminator 1.57% : 0.000004s : 25: predicate.tuple_to_list_eliminator_ 2.02% : 0.000006s : 38: predicate.updatestate_pure_node_eliminater 3.50% : 0.000010s : 50: predicate.updatestate_useless_node_eliminater 0.47% : 0.000001s : 6: predicate.value_based_eliminate 0.80% : 0.000002s : 12: predicate.virtual_dataset_eliminate 0.86% : 0.000002s : 12: predicate.virtual_output_eliminate 0.33% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.61% : 0.000002s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000890 4 2.96% : 0.000026s : 1: func_graph_cloner_run.FuncGraphClonerGraph 97.04% : 0.000864s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.212465 196 0.00% : 0.000008s : 1: ForceFp32Comm 8.19% : 0.017402s : 1: add_attr 8.18% : 0.017376s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.06% : 0.000135s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.18% : 0.000390s : 1: auto_monad 0.02% : 0.000045s : 1: auto_monad_reorder 0.00% : 0.000009s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000010s : 1: bias_add_comm_swap 0.43% : 0.000918s : 1: bootstrap 0.02% : 0.000050s : 1: cconv 0.00% : 0.000005s : 1: comm_op_add_attrs 0.02% : 0.000036s : 1: control_data_broadcast_order 0.01% : 0.000025s : 1: convert_after_rewriter 0.02% : 0.000048s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.01% : 0.000014s : 1: environ_conv 0.02% : 0.000039s : 1: event_method 0.01% : 0.000027s : 1: execute 0.00% : 0.000007s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.01% : 0.000014s : 1: graph_reusing 0.00% : 0.000010s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000005s : 1: handle_group_info 0.00% : 0.000009s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000009s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.01% : 0.000011s : 1: label_micro_interleaved_index 0.35% : 0.000746s : 1: loop_unroll 0.00% : 0.000007s : 1: merge_cast_opt 0.00% : 0.000010s : 1: micro_interleaved_order_control 0.51% : 0.001087s : 1: mutable_eliminate 0.01% : 0.000012s : 1: offloading_packed_experts 0.02% : 0.000035s : 1: opt.transform.loop_unroll_optimizer 0.02% : 0.000038s : 1: opt.transform.mutable_eliminate 0.95% : 0.002021s : 78: opt.transform.opt_a 0.03% : 0.000073s : 1: opt.transform.opt_after_cconv 0.02% : 0.000046s : 1: opt.transform.opt_after_jit_grad 0.12% : 0.000259s : 28: opt.transform.opt_b 0.05% : 0.000098s : 2: opt.transform.opt_trans_graph 0.04% : 0.000082s : 4: opt.transform.symbol_engine_opt 2.53% : 0.005384s : 1: opt_a 0.11% : 0.000230s : 1: opt_after_cconv 0.40% : 0.000850s : 1: opt_after_jit_grad 0.23% : 0.000499s : 1: opt_b 4.49% : 0.009532s : 1: optimize 0.02% : 0.000036s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000017s : 1: order_py_execute_after_rewriter 0.02% : 0.000042s : 1: overlap_grad_flash_sp 0.00% : 0.000009s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000015s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000006s : 1: overlap_opt_shard_in_pipeline 0.01% : 0.000011s : 1: overlap_param_gather 0.00% : 0.000009s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000022s : 1: overlap_recompute_and_grad_model_parallel 0.01% : 0.000011s : 1: overlap_recompute_comm 0.00% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.03% : 0.000060s : 1: pre_auto_parallel 0.03% : 0.000060s : 1: py_interpret_to_execute 0.02% : 0.000039s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000009s : 1: remove_cast_before_assign_add 0.01% : 0.000024s : 1: remove_dup_value 0.32% : 0.000689s : 1: renormalize.infer 0.29% : 0.000624s : 1: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.13% : 0.000266s : 1: rewriter_after_opt_a 0.05% : 0.000098s : 1: rewriter_before_opt_a 0.01% : 0.000015s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000007s : 1: split_matmul_comm_elemetwise 0.01% : 0.000019s : 1: swap_dp_allreduce_reducescatter 0.08% : 0.000174s : 1: symbol_engine_optimizer 20.54% : 0.043631s : 1: task_emit 0.08% : 0.000169s : 1: tuple_transform 51.09% : 0.108548s : 1: type_inference 0.06% : 0.000129s : 1: validate TotalTime = 0.106771, [24] [bootstrap]: 0.00045982 [type_inference]: 0.00400834 [event_method]: 1.88e-05 [auto_monad]: 0.00016635 [graph_reusing]: 6.53e-06 [inline]: 2.84999e-06 [add_attr]: 0.00422068, [1] [add_attr_with_inline]: 0.00420469, [1] [Cycle 1]: 8.205e-05, [2] [tag_attr]: 1.926e-05 [meta_addattr_fg_expand]: 4.23001e-06 [parallel-infer-symbol]: 4.31002e-06 [pre_auto_parallel]: 3.955e-05 [insert-virtual-dataset]: 2.69001e-06 [parallel-infer-symbol-second]: 8.70001e-07 [dataset_repeat_opt]: 2.44001e-06 [pipeline_split]: 1.76e-06 [optimize]: 0.0834027, [53] [py_interpret_to_execute]: 2.757e-05 [rewriter_before_opt_a]: 6.332e-05 [opt_a]: 0.0801447, [2] [Cycle 1]: 0.078942, [45] [expand_dump_flag]: 3.76001e-06 [switch_simplify]: 3.053e-05 [loop_unroll]: 2.232e-05 [a_1]: 0.00050909 [with_stream_mark]: 3.119e-05 [recompute_prepare]: 1.568e-05 [updatestate_depend_eliminate]: 7.42002e-06 [updatestate_assign_eliminate]: 6.71999e-06 [updatestate_loads_eliminate]: 5.62999e-06 [parameter_eliminate]: 2.05002e-06 [a_2]: 0.00014511 [accelerated_algorithm]: 1.493e-05 [shard]: 2.85998e-06 [meta_shard_fg_expand]: 3.43e-06 [shard_inline]: 1.025e-05 [merge_send_recv]: 1.281e-05 [auto_parallel]: 1.285e-05 [parallel]: 2.437e-05 [flash_sp]: 1.225e-05 [merge_comm]: 7.03e-06 [allreduce_fusion]: 6.26998e-06 [matmul_add_comm_reduction]: 1.508e-05 [allreduce_slice_to_reducescatter]: 7.00005e-07 [virtual_shard_identity]: 1.511e-05 [virtual_dataset]: 1.14e-05 [get_grad_eliminate_]: 1.005e-05 [virtual_output]: 1.032e-05 [merge_forward]: 9.60001e-06 [cell_reuse_recompute_pass]: 1.57001e-06 [offload_activation]: 1.661e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.759e-05 [merge_recompute_call_nodes]: 1.55999e-06 [before_grad]: 1.939e-05 [set_forward_comm_id_for_comm_node_pass]: 6.17001e-06 [meta_fg_expand]: 4.63001e-06 [flash_sp_send_recv_attached]: 2.69999e-06 [receive_attached]: 2.29001e-06 [after_resolve]: 1.529e-05 [a_after_grad]: 1.634e-05 [renormalize]: 0.077297 [add_forward_monad_depend]: 1.177e-05 [auto_monad_grad]: 2.84001e-06 [auto_monad_eliminator]: 3.484e-05 [cse]: 4.604e-05 [a_3]: 8.659e-05 [Cycle 2]: 0.00118409, [45] [expand_dump_flag]: 2.78998e-06 [switch_simplify]: 1.209e-05 [loop_unroll]: 9.66998e-06 [a_1]: 0.00025203 [with_stream_mark]: 2.593e-05 [recompute_prepare]: 1.094e-05 [updatestate_depend_eliminate]: 6.73e-06 [updatestate_assign_eliminate]: 5.97999e-06 [updatestate_loads_eliminate]: 6.41e-06 [parameter_eliminate]: 1.91998e-06 [a_2]: 0.00013143 [accelerated_algorithm]: 1.119e-05 [shard]: 3.26999e-06 [meta_shard_fg_expand]: 2.76999e-06 [shard_inline]: 9.15001e-06 [merge_send_recv]: 1.114e-05 [auto_parallel]: 1.364e-05 [parallel]: 1.052e-05 [flash_sp]: 4.28999e-06 [merge_comm]: 5.92999e-06 [allreduce_fusion]: 5.46e-06 [matmul_add_comm_reduction]: 1.448e-05 [allreduce_slice_to_reducescatter]: 8.10018e-07 [virtual_shard_identity]: 1.362e-05 [virtual_dataset]: 1.053e-05 [get_grad_eliminate_]: 1.027e-05 [virtual_output]: 1.067e-05 [merge_forward]: 6.63003e-06 [cell_reuse_recompute_pass]: 3.71999e-06 [offload_activation]: 1.637e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.643e-05 [merge_recompute_call_nodes]: 1.64998e-06 [before_grad]: 1.67e-05 [set_forward_comm_id_for_comm_node_pass]: 7.28e-06 [meta_fg_expand]: 4.15999e-06 [flash_sp_send_recv_attached]: 2.06e-06 [receive_attached]: 2.71e-06 [after_resolve]: 1.408e-05 [a_after_grad]: 1.374e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 2.87002e-06 [auto_monad_grad]: 1.52001e-06 [auto_monad_eliminator]: 1.65e-05 [cse]: 3.012e-05 [a_3]: 5.599e-05 [py_interpret_to_execute_after_opt_a]: 2.682e-05 [slice_cell_reuse_recomputed_activation]: 2.02999e-06 [rewriter_after_opt_a]: 0.00024138 [convert_after_rewriter]: 1.219e-05 [order_py_execute_after_rewriter]: 7.23e-06 [mutable_eliminate]: 0.00084162 [opt_b]: 0.0003817, [1] [Cycle 1]: 0.00036827, [7] [b_1]: 0.00023696 [b_2]: 1.25e-05 [updatestate_depend_eliminate]: 1.316e-05 [updatestate_assign_eliminate]: 7.05998e-06 [updatestate_loads_eliminate]: 5.97001e-06 [renormalize]: 3.4002e-07 [cse]: 3.527e-05 [optimize_parallel_all_gather_comm]: 3.155e-05 [overlap_param_gather]: 2.05002e-06 [cconv]: 3.987e-05 [loop_unroll]: 0.00058874 [opt_after_cconv]: 0.00018274, [1] [Cycle 1]: 0.00017409, [7] [c_1]: 6.56e-05 [parameter_eliminate]: 5.39e-06 [updatestate_depend_eliminate]: 1.062e-05 [updatestate_assign_eliminate]: 5.72001e-06 [updatestate_loads_eliminate]: 4.87e-06 [cse]: 2.783e-05 [renormalize]: 2.69996e-07 [remove_dup_value]: 1.967e-05 [tuple_transform]: 0.00012145, [1] [Cycle 1]: 0.00011411, [4] [d_1]: 7.241e-05 [none_parameter_eliminate]: 1.71e-06 [renormalize]: 1.99972e-07 [switch_simplify]: 1.081e-05 [partial_unused_args_eliminate]: 2.22001e-06 [add_recomputation]: 8.624e-05 [cse_after_recomputation]: 3.045e-05, [1] [Cycle 1]: 2.499e-05, [1] [cse]: 1.825e-05 [environ_conv]: 8.42e-06 [swap_dp_allreduce_reducescatter]: 8.87e-06 [bias_add_comm_swap]: 3.12002e-06 [label_micro_interleaved_index]: 5.30999e-06 [label_fine_grained_interleaved_index]: 2.97002e-06 [merge_cast_opt]: 1.52999e-06 [slice_recompute_activation]: 2.68e-06 [micro_interleaved_order_control]: 2.44001e-06 [assign_add_opt]: 1.30001e-06 [ForceFp32Comm]: 8.70001e-07 [remove_cast_before_assign_add]: 1.34e-06 [full_micro_interleaved_order_control]: 2.69999e-06 [reorder_send_recv_between_fp_bp]: 2.85002e-06 [comm_op_add_attrs]: 1.60999e-06 [add_comm_op_reuse_tag]: 1.03001e-06 [interleave_split_concat_branches]: 1.57001e-06 [interleave_parallel_branches]: 1.24e-06 [overlap_opt_shard_in_pipeline]: 1.22999e-06 [overlap_opt_shard_grad_in_pipeline]: 2.02001e-06 [control_data_broadcast_order]: 1.982e-05 [grouped_pairwise_exchange_alltoall]: 1.71e-06 [offloading_packed_experts]: 5.90002e-06 [overlap_recompute_and_grad_model_parallel]: 6.58e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.78002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.63002e-06 [overlap_recompute_comm]: 2.76e-06 [overlap_grad_ring_attention]: 5.90002e-06 [overlap_grad_flash_sp]: 3.004e-05 [begin_end_overlap_inline]: 5.19998e-07 [split_matmul_comm_elemetwise]: 3.09001e-06 [split_layernorm_comm]: 2.15002e-06 [handle_group_info]: 1.29998e-06 [symbol_engine_optimizer]: 0.00011702, [1] [Cycle 1]: 0.00011075, [6] [build]: 5.72999e-06 [elim_shapecalc]: 1.748e-05 [elim_not_effective]: 1.928e-05 [opt_reshape]: 1.181e-05 [fold_const_symbol]: 1.615e-05 [renormalize]: 4.40021e-07 [detach_backward]: 2.11e-06 [pipeline_parallel_scheduler]: 1.88002e-06 [auto_monad_reorder]: 3.149e-05 [get_jit_bprop_graph]: 1.93002e-06 [rewriter_after_jit_bprop_graph]: 6.46e-06 [opt_after_jit_grad]: 0.00075549 [validate]: 7.159e-05 [backend_pass]: 1.16002e-06 [task_emit]: 0.0132348 [execute]: 1.195e-05 Sums bootstrap : 0.000460s : 0.46% type_inference : 0.004008s : 3.97% event_method : 0.000019s : 0.02% auto_monad : 0.000166s : 0.16% graph_reusing : 0.000007s : 0.01% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000019s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000040s : 0.04% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000028s : 0.03% optimize.rewriter_before_opt_a : 0.000063s : 0.06% optimize.opt_a.expand_dump_flag : 0.000007s : 0.01% optimize.opt_a.switch_simplify : 0.000043s : 0.04% optimize.opt_a.loop_unroll : 0.000032s : 0.03% optimize.opt_a.a_1 : 0.000761s : 0.75% optimize.opt_a.with_stream_mark : 0.000057s : 0.06% optimize.opt_a.recompute_prepare : 0.000027s : 0.03% optimize.opt_a.updatestate_depend_eliminate : 0.000014s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000013s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000012s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.00% optimize.opt_a.a_2 : 0.000277s : 0.27% optimize.opt_a.accelerated_algorithm : 0.000026s : 0.03% optimize.opt_a.shard : 0.000006s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000006s : 0.01% optimize.opt_a.shard_inline : 0.000019s : 0.02% optimize.opt_a.merge_send_recv : 0.000024s : 0.02% optimize.opt_a.auto_parallel : 0.000026s : 0.03% optimize.opt_a.parallel : 0.000035s : 0.03% optimize.opt_a.flash_sp : 0.000017s : 0.02% optimize.opt_a.merge_comm : 0.000013s : 0.01% optimize.opt_a.allreduce_fusion : 0.000012s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000030s : 0.03% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000029s : 0.03% optimize.opt_a.virtual_dataset : 0.000022s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000020s : 0.02% optimize.opt_a.virtual_output : 0.000021s : 0.02% optimize.opt_a.merge_forward : 0.000016s : 0.02% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.01% optimize.opt_a.offload_activation : 0.000033s : 0.03% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000054s : 0.05% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000036s : 0.04% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000013s : 0.01% optimize.opt_a.meta_fg_expand : 0.000009s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000029s : 0.03% optimize.opt_a.a_after_grad : 0.000030s : 0.03% optimize.opt_a.renormalize : 0.077297s : 76.55% optimize.opt_a.add_forward_monad_depend : 0.000015s : 0.01% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000051s : 0.05% optimize.opt_a.cse : 0.000076s : 0.08% optimize.opt_a.a_3 : 0.000143s : 0.14% optimize.py_interpret_to_execute_after_opt_a : 0.000027s : 0.03% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000241s : 0.24% optimize.convert_after_rewriter : 0.000012s : 0.01% optimize.order_py_execute_after_rewriter : 0.000007s : 0.01% optimize.mutable_eliminate : 0.000842s : 0.83% optimize.opt_b.b_1 : 0.000237s : 0.23% optimize.opt_b.b_2 : 0.000013s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000013s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000007s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000006s : 0.01% optimize.opt_b.renormalize : 0.000000s : 0.00% optimize.opt_b.cse : 0.000035s : 0.03% optimize.optimize_parallel_all_gather_comm : 0.000032s : 0.03% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000040s : 0.04% optimize.loop_unroll : 0.000589s : 0.58% optimize.opt_after_cconv.c_1 : 0.000066s : 0.06% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.01% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000011s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000006s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.cse : 0.000028s : 0.03% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000020s : 0.02% optimize.tuple_transform.d_1 : 0.000072s : 0.07% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000011s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000086s : 0.09% optimize.cse_after_recomputation.cse : 0.000018s : 0.02% optimize.environ_conv : 0.000008s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000009s : 0.01% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000005s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000002s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000002s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000020s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000006s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 0.01% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000006s : 0.01% optimize.overlap_grad_flash_sp : 0.000030s : 0.03% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000006s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000017s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000019s : 0.02% optimize.symbol_engine_optimizer.opt_reshape : 0.000012s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000016s : 0.02% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000031s : 0.03% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.01% opt_after_jit_grad : 0.000755s : 0.75% validate : 0.000072s : 0.07% backend_pass : 0.000001s : 0.00% task_emit : 0.013235s : 13.11% execute : 0.000012s : 0.01% Time group info: ------[substitution.] 0.000212 55 5.20% : 0.000011s : 2: substitution.depend_value_elim 1.31% : 0.000003s : 5: substitution.elim_not_effective 1.03% : 0.000002s : 5: substitution.fold_const_symbol 3.68% : 0.000008s : 6: substitution.graph_param_transform 61.76% : 0.000131s : 1: substitution.inline 2.88% : 0.000006s : 10: substitution.j_node_and_user_rematch 3.89% : 0.000008s : 10: substitution.remove_not_recompute_node 2.73% : 0.000006s : 2: substitution.replace_old_param 9.58% : 0.000020s : 6: substitution.updatestate_pure_node_eliminater 7.93% : 0.000017s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.003945 2 85.63% : 0.003378s : 1: type_inference.infer 14.37% : 0.000567s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000129 1 100.00% : 0.000129s : 1: match.inline ------[predicate.] 0.000260 1419 0.94% : 0.000002s : 13: predicate.accumulaten_eliminater 1.49% : 0.000004s : 6: predicate.ad_related_special_op_eliminate 0.70% : 0.000002s : 12: predicate.addn_check_dump 0.92% : 0.000002s : 13: predicate.addn_zero_filter 0.75% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 2.30% : 0.000006s : 25: predicate.arithmetic_simplify 0.87% : 0.000002s : 13: predicate.cast_eliminate 0.80% : 0.000002s : 12: predicate.check_bprop_eliminate 0.75% : 0.000002s : 12: predicate.compare_switch_simplify 0.21% : 0.000001s : 6: predicate.const_output_eliminate 0.71% : 0.000002s : 12: predicate.depend_value_elim 0.85% : 0.000002s : 13: predicate.dict_get_item_const_eliminator 0.91% : 0.000002s : 13: predicate.dict_get_item_eliminator 0.91% : 0.000002s : 13: predicate.dict_set_item_eliminator 1.52% : 0.000004s : 12: predicate.dumpgradient_eliminate 0.30% : 0.000001s : 6: predicate.elim_not_effective 0.54% : 0.000001s : 6: predicate.elim_shapecalc_of_broadcastargs 1.12% : 0.000003s : 19: predicate.environ_add_const_eliminate 1.12% : 0.000003s : 19: predicate.environ_get_add_eliminate 1.12% : 0.000003s : 19: predicate.environ_get_depend_swap 2.15% : 0.000006s : 31: predicate.environ_get_eliminate 1.09% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.75% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.54% : 0.000004s : 14: predicate.float_depend_g_call 0.68% : 0.000002s : 12: predicate.float_environ_get_switch 1.09% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.19% : 0.000001s : 6: predicate.fold_const_symbol 0.92% : 0.000002s : 12: predicate.get_grad_eliminate 0.28% : 0.000001s : 6: predicate.graph_param_transform 0.67% : 0.000002s : 12: predicate.incorporate_call 0.59% : 0.000002s : 12: predicate.incorporate_call_switch 5.61% : 0.000015s : 63: predicate.inline 0.88% : 0.000002s : 12: predicate.inline_without_move 0.35% : 0.000001s : 12: predicate.j_node_and_user_rematch 1.19% : 0.000003s : 12: predicate.less_batch_normalization 1.81% : 0.000005s : 25: predicate.list_to_tuple_eliminator_ 2.23% : 0.000006s : 38: predicate.load_eliminater 1.32% : 0.000003s : 6: predicate.loop_unroll_after_grad 1.08% : 0.000003s : 17: predicate.loop_unroll_before_grad 1.98% : 0.000005s : 25: predicate.make_slice_get_slice_eliminator 0.70% : 0.000002s : 12: predicate.merge_addn 0.67% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.68% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.77% : 0.000002s : 13: predicate.minmaximum_grad 1.68% : 0.000004s : 6: predicate.mutable_eliminate 0.41% : 0.000001s : 6: predicate.opt_reshape 0.49% : 0.000001s : 6: predicate.parallel_virtual_node 1.15% : 0.000003s : 14: predicate.partial_defer_inline 1.12% : 0.000003s : 19: predicate.partial_eliminate 0.81% : 0.000002s : 13: predicate.print_const_string_wrapper 0.82% : 0.000002s : 12: predicate.reduce_all_const_elim 1.07% : 0.000003s : 13: predicate.reduce_eliminate 2.10% : 0.000005s : 38: predicate.redundant_stop_gradient_eliminater 0.81% : 0.000002s : 12: predicate.remove_not_recompute_node 1.15% : 0.000003s : 25: predicate.replace_applicator 0.59% : 0.000002s : 12: predicate.replace_old_param 0.40% : 0.000001s : 6: predicate.reset_defer_inline 0.96% : 0.000002s : 13: predicate.reshape_eliminate 0.79% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.55% : 0.000001s : 6: predicate.row_tensor_eliminate 1.01% : 0.000003s : 12: predicate.same_eliminate 0.44% : 0.000001s : 12: predicate.set_cell_output_no_recompute 1.09% : 0.000003s : 12: predicate.shard_identity_eliminate 1.05% : 0.000003s : 12: predicate.special_op_eliminate 0.80% : 0.000002s : 12: predicate.specialize_transform 1.21% : 0.000003s : 12: predicate.split_environ_get_set_with_tuple_value 0.92% : 0.000002s : 12: predicate.stack_unstack_eliminate 0.44% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.86% : 0.000002s : 14: predicate.switch_defer_inline 1.51% : 0.000004s : 26: predicate.switch_layer_defer_inline 3.37% : 0.000009s : 49: predicate.switch_simplify 0.88% : 0.000002s : 13: predicate.tile_eliminate 0.88% : 0.000002s : 13: predicate.transpose_eliminate 1.68% : 0.000004s : 25: predicate.tuple_list_convert_item_index_to_positive 1.58% : 0.000004s : 25: predicate.tuple_list_get_item_const_eliminator 1.66% : 0.000004s : 25: predicate.tuple_list_get_item_depend_reorder 3.09% : 0.000008s : 37: predicate.tuple_list_get_item_eliminator 1.69% : 0.000004s : 25: predicate.tuple_list_get_set_item_eliminator 2.31% : 0.000006s : 37: predicate.tuple_list_set_item_eliminator 1.69% : 0.000004s : 25: predicate.tuple_to_list_eliminator_ 2.44% : 0.000006s : 38: predicate.updatestate_pure_node_eliminater 3.28% : 0.000009s : 50: predicate.updatestate_useless_node_eliminater 0.44% : 0.000001s : 6: predicate.value_based_eliminate 0.83% : 0.000002s : 12: predicate.virtual_dataset_eliminate 0.92% : 0.000002s : 12: predicate.virtual_output_eliminate 0.43% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.55% : 0.000001s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000552 4 4.60% : 0.000025s : 1: func_graph_cloner_run.FuncGraphClonerGraph 95.40% : 0.000527s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.273526 196 0.00% : 0.000004s : 1: ForceFp32Comm 1.55% : 0.004228s : 1: add_attr 1.54% : 0.004211s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.04% : 0.000096s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.06% : 0.000175s : 1: auto_monad 0.01% : 0.000036s : 1: auto_monad_reorder 0.00% : 0.000008s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.18% : 0.000502s : 1: bootstrap 0.02% : 0.000045s : 1: cconv 0.00% : 0.000006s : 1: comm_op_add_attrs 0.01% : 0.000024s : 1: control_data_broadcast_order 0.01% : 0.000017s : 1: convert_after_rewriter 0.01% : 0.000034s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000012s : 1: environ_conv 0.01% : 0.000026s : 1: event_method 0.01% : 0.000021s : 1: execute 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000011s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000005s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000009s : 1: label_micro_interleaved_index 0.22% : 0.000602s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.31% : 0.000860s : 1: mutable_eliminate 0.00% : 0.000010s : 1: offloading_packed_experts 0.01% : 0.000022s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000027s : 1: opt.transform.mutable_eliminate 0.53% : 0.001451s : 78: opt.transform.opt_a 0.02% : 0.000064s : 1: opt.transform.opt_after_cconv 0.02% : 0.000050s : 1: opt.transform.opt_after_jit_grad 0.08% : 0.000211s : 28: opt.transform.opt_b 0.03% : 0.000081s : 2: opt.transform.opt_trans_graph 0.02% : 0.000061s : 4: opt.transform.symbol_engine_opt 29.30% : 0.080150s : 1: opt_a 0.07% : 0.000187s : 1: opt_after_cconv 0.28% : 0.000779s : 1: opt_after_jit_grad 0.14% : 0.000387s : 1: opt_b 30.49% : 0.083409s : 1: optimize 0.01% : 0.000036s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000012s : 1: order_py_execute_after_rewriter 0.01% : 0.000034s : 1: overlap_grad_flash_sp 0.00% : 0.000006s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000009s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000011s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.02% : 0.000044s : 1: pre_auto_parallel 0.01% : 0.000032s : 1: py_interpret_to_execute 0.01% : 0.000033s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.01% : 0.000024s : 1: remove_dup_value 28.00% : 0.076578s : 1: renormalize.infer 0.25% : 0.000694s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.09% : 0.000252s : 1: rewriter_after_opt_a 0.03% : 0.000069s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000007s : 1: split_matmul_comm_elemetwise 0.00% : 0.000013s : 1: swap_dp_allreduce_reducescatter 0.04% : 0.000120s : 1: symbol_engine_optimizer 4.85% : 0.013262s : 1: task_emit 0.05% : 0.000125s : 1: tuple_transform 1.48% : 0.004036s : 1: type_inference 0.05% : 0.000128s : 1: validate TotalTime = 0.432651, [24] [bootstrap]: 0.00053832 [type_inference]: 0.0126455 [event_method]: 7.404e-05 [auto_monad]: 0.00032199 [graph_reusing]: 1.053e-05 [inline]: 3.25e-06 [add_attr]: 0.172528, [1] [add_attr_with_inline]: 0.172514, [1] [Cycle 1]: 0.00012318, [2] [tag_attr]: 5.266e-05 [meta_addattr_fg_expand]: 1.089e-05 [parallel-infer-symbol]: 3.97e-06 [pre_auto_parallel]: 6.304e-05 [insert-virtual-dataset]: 2.36e-06 [parallel-infer-symbol-second]: 2.71999e-06 [dataset_repeat_opt]: 2.04999e-06 [pipeline_split]: 1.60999e-06 [optimize]: 0.0791088, [53] [py_interpret_to_execute]: 8.631e-05 [rewriter_before_opt_a]: 0.00016753 [opt_a]: 0.0755662, [3] [Cycle 1]: 0.0672525, [45] [expand_dump_flag]: 1.11e-05 [switch_simplify]: 0.00011018 [loop_unroll]: 7.427e-05 [a_1]: 0.00186373 [with_stream_mark]: 4.799e-05 [recompute_prepare]: 4.18e-05 [updatestate_depend_eliminate]: 1.446e-05 [updatestate_assign_eliminate]: 1.166e-05 [updatestate_loads_eliminate]: 1.03e-05 [parameter_eliminate]: 4.08999e-06 [a_2]: 0.00033239 [accelerated_algorithm]: 5.251e-05 [shard]: 2.26e-06 [meta_shard_fg_expand]: 8.64998e-06 [shard_inline]: 2.323e-05 [merge_send_recv]: 2.553e-05 [auto_parallel]: 1.98e-05 [parallel]: 2.495e-05 [flash_sp]: 1.816e-05 [merge_comm]: 1.371e-05 [allreduce_fusion]: 1.239e-05 [matmul_add_comm_reduction]: 4.547e-05 [allreduce_slice_to_reducescatter]: 8.89995e-07 [virtual_shard_identity]: 0.00011592 [virtual_dataset]: 2.811e-05 [get_grad_eliminate_]: 2.585e-05 [virtual_output]: 2.45e-05 [merge_forward]: 1.78e-05 [cell_reuse_recompute_pass]: 2.68e-06 [offload_activation]: 2.848e-05 [cell_reuse_handle_not_recompute_node_pass]: 5.357e-05 [merge_recompute_call_nodes]: 2.07999e-06 [before_grad]: 4.477e-05 [set_forward_comm_id_for_comm_node_pass]: 1.924e-05 [meta_fg_expand]: 0.00843821 [flash_sp_send_recv_attached]: 6.89999e-06 [receive_attached]: 2.49999e-06 [after_resolve]: 0.00052546 [a_after_grad]: 0.00035067 [renormalize]: 0.052023 [add_forward_monad_depend]: 3.158e-05 [auto_monad_grad]: 1.821e-05 [auto_monad_eliminator]: 0.00015256 [cse]: 0.00032211 [a_3]: 0.00169735 [Cycle 2]: 0.00699869, [45] [expand_dump_flag]: 4.70999e-06 [switch_simplify]: 0.00011005 [loop_unroll]: 0.00010322 [a_1]: 0.00320626 [with_stream_mark]: 3.543e-05 [recompute_prepare]: 2.184e-05 [updatestate_depend_eliminate]: 4.483e-05 [updatestate_assign_eliminate]: 7.65e-06 [updatestate_loads_eliminate]: 6.20002e-06 [parameter_eliminate]: 2.47001e-06 [a_2]: 0.00016798 [accelerated_algorithm]: 2.115e-05 [shard]: 2.22001e-06 [meta_shard_fg_expand]: 5.91e-06 [shard_inline]: 1.239e-05 [merge_send_recv]: 1.27e-05 [auto_parallel]: 1.457e-05 [parallel]: 1.16e-05 [flash_sp]: 4.53001e-06 [merge_comm]: 6.81001e-06 [allreduce_fusion]: 6.36e-06 [matmul_add_comm_reduction]: 1.473e-05 [allreduce_slice_to_reducescatter]: 1.52999e-06 [virtual_shard_identity]: 1.595e-05 [virtual_dataset]: 1.153e-05 [get_grad_eliminate_]: 1.085e-05 [virtual_output]: 1.065e-05 [merge_forward]: 6.56e-06 [cell_reuse_recompute_pass]: 1.91e-06 [offload_activation]: 1.674e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.362e-05 [merge_recompute_call_nodes]: 1.49e-06 [before_grad]: 2.034e-05 [set_forward_comm_id_for_comm_node_pass]: 8.15999e-06 [meta_fg_expand]: 9.761e-05 [flash_sp_send_recv_attached]: 2.00002e-06 [receive_attached]: 2.58003e-06 [after_resolve]: 2.309e-05 [a_after_grad]: 1.972e-05 [renormalize]: 0.00221989 [add_forward_monad_depend]: 9.86e-06 [auto_monad_grad]: 2.78e-06 [auto_monad_eliminator]: 3.536e-05 [cse]: 0.00016132 [a_3]: 9.943e-05 [Cycle 3]: 0.00128646, [45] [expand_dump_flag]: 3.46999e-06 [switch_simplify]: 1.566e-05 [loop_unroll]: 1.284e-05 [a_1]: 0.00031907 [with_stream_mark]: 2.425e-05 [recompute_prepare]: 1.436e-05 [updatestate_depend_eliminate]: 7.99997e-06 [updatestate_assign_eliminate]: 6.46e-06 [updatestate_loads_eliminate]: 6.34001e-06 [parameter_eliminate]: 2.16e-06 [a_2]: 0.00015925 [accelerated_algorithm]: 2.034e-05 [shard]: 1.99e-06 [meta_shard_fg_expand]: 3.5e-06 [shard_inline]: 1.414e-05 [merge_send_recv]: 1.305e-05 [auto_parallel]: 1.414e-05 [parallel]: 1.1e-05 [flash_sp]: 1.89999e-06 [merge_comm]: 9.26002e-06 [allreduce_fusion]: 6.31e-06 [matmul_add_comm_reduction]: 1.47e-05 [allreduce_slice_to_reducescatter]: 1.11997e-06 [virtual_shard_identity]: 1.598e-05 [virtual_dataset]: 1.357e-05 [get_grad_eliminate_]: 1.091e-05 [virtual_output]: 1.069e-05 [merge_forward]: 7.93999e-06 [cell_reuse_recompute_pass]: 3.55e-06 [offload_activation]: 1.617e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.606e-05 [merge_recompute_call_nodes]: 1.35999e-06 [before_grad]: 2.138e-05 [set_forward_comm_id_for_comm_node_pass]: 9.73002e-06 [meta_fg_expand]: 4.97e-06 [flash_sp_send_recv_attached]: 1.97001e-06 [receive_attached]: 2.06e-06 [after_resolve]: 2.181e-05 [a_after_grad]: 1.897e-05 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 3.78001e-06 [auto_monad_grad]: 2.29999e-06 [auto_monad_eliminator]: 2.76e-05 [cse]: 4.34e-05 [a_3]: 7.432e-05 [py_interpret_to_execute_after_opt_a]: 2.607e-05 [slice_cell_reuse_recomputed_activation]: 2.68e-06 [rewriter_after_opt_a]: 8.399e-05 [convert_after_rewriter]: 1.265e-05 [order_py_execute_after_rewriter]: 8.05999e-06 [mutable_eliminate]: 0.00093364 [opt_b]: 0.00042505, [1] [Cycle 1]: 0.00041395, [7] [b_1]: 0.00026529 [b_2]: 1.489e-05 [updatestate_depend_eliminate]: 1.573e-05 [updatestate_assign_eliminate]: 6.54001e-06 [updatestate_loads_eliminate]: 6.69999e-06 [renormalize]: 7.89994e-07 [cse]: 5.165e-05 [optimize_parallel_all_gather_comm]: 3.166e-05 [overlap_param_gather]: 1.75001e-06 [cconv]: 4.068e-05 [loop_unroll]: 0.00059944 [opt_after_cconv]: 0.00020466, [1] [Cycle 1]: 0.00019484, [7] [c_1]: 8.129e-05 [parameter_eliminate]: 5.49e-06 [updatestate_depend_eliminate]: 1.342e-05 [updatestate_assign_eliminate]: 5.97001e-06 [updatestate_loads_eliminate]: 5.91998e-06 [cse]: 4.123e-05 [renormalize]: 7.89994e-07 [remove_dup_value]: 2.152e-05 [tuple_transform]: 0.00013265, [1] [Cycle 1]: 0.00012622, [4] [d_1]: 8.875e-05 [none_parameter_eliminate]: 2.15002e-06 [renormalize]: 1.50001e-07 [switch_simplify]: 1.249e-05 [partial_unused_args_eliminate]: 1.69e-06 [add_recomputation]: 0.00010474 [cse_after_recomputation]: 4.446e-05, [1] [Cycle 1]: 3.73e-05, [1] [cse]: 2.897e-05 [environ_conv]: 1.438e-05 [swap_dp_allreduce_reducescatter]: 9.24e-06 [bias_add_comm_swap]: 2.88e-06 [label_micro_interleaved_index]: 6.78998e-06 [label_fine_grained_interleaved_index]: 2.69999e-06 [merge_cast_opt]: 1.15999e-06 [slice_recompute_activation]: 2.12001e-06 [micro_interleaved_order_control]: 2.43998e-06 [assign_add_opt]: 1.24998e-06 [ForceFp32Comm]: 1.11002e-06 [remove_cast_before_assign_add]: 1.09e-06 [full_micro_interleaved_order_control]: 2.29001e-06 [reorder_send_recv_between_fp_bp]: 3.17002e-06 [comm_op_add_attrs]: 1.15001e-06 [add_comm_op_reuse_tag]: 9.09989e-07 [interleave_split_concat_branches]: 1.42999e-06 [interleave_parallel_branches]: 1.05001e-06 [overlap_opt_shard_in_pipeline]: 2.29001e-06 [overlap_opt_shard_grad_in_pipeline]: 1.78002e-06 [control_data_broadcast_order]: 2.504e-05 [grouped_pairwise_exchange_alltoall]: 1.60999e-06 [offloading_packed_experts]: 7.23e-06 [overlap_recompute_and_grad_model_parallel]: 8e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.29e-06 [overlap_recompute_allgather_and_fa_grad]: 1.34e-06 [overlap_recompute_comm]: 2.47001e-06 [overlap_grad_ring_attention]: 6.04999e-06 [overlap_grad_flash_sp]: 3.492e-05 [begin_end_overlap_inline]: 5.10016e-07 [split_matmul_comm_elemetwise]: 2.73e-06 [split_layernorm_comm]: 1.99999e-06 [handle_group_info]: 9.80013e-07 [symbol_engine_optimizer]: 0.00015168, [1] [Cycle 1]: 0.00014383, [6] [build]: 1.708e-05 [elim_shapecalc]: 2.42e-05 [elim_not_effective]: 2.581e-05 [opt_reshape]: 1.314e-05 [fold_const_symbol]: 2.127e-05 [renormalize]: 2.00002e-07 [detach_backward]: 3.63999e-06 [pipeline_parallel_scheduler]: 1.37e-06 [auto_monad_reorder]: 4.251e-05 [get_jit_bprop_graph]: 2.04e-06 [rewriter_after_jit_bprop_graph]: 6.25002e-06 [opt_after_jit_grad]: 0.00068257 [validate]: 6.954e-05 [backend_pass]: 1.15999e-06 [task_emit]: 0.166189 [execute]: 1.008e-05 Sums bootstrap : 0.000538s : 0.21% type_inference : 0.012645s : 4.90% event_method : 0.000074s : 0.03% auto_monad : 0.000322s : 0.12% graph_reusing : 0.000011s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000053s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000011s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000063s : 0.02% insert-virtual-dataset : 0.000002s : 0.00% parallel-infer-symbol-second : 0.000003s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000086s : 0.03% optimize.rewriter_before_opt_a : 0.000168s : 0.06% optimize.opt_a.expand_dump_flag : 0.000019s : 0.01% optimize.opt_a.switch_simplify : 0.000236s : 0.09% optimize.opt_a.loop_unroll : 0.000190s : 0.07% optimize.opt_a.a_1 : 0.005389s : 2.09% optimize.opt_a.with_stream_mark : 0.000108s : 0.04% optimize.opt_a.recompute_prepare : 0.000078s : 0.03% optimize.opt_a.updatestate_depend_eliminate : 0.000067s : 0.03% optimize.opt_a.updatestate_assign_eliminate : 0.000026s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000023s : 0.01% optimize.opt_a.parameter_eliminate : 0.000009s : 0.00% optimize.opt_a.a_2 : 0.000660s : 0.26% optimize.opt_a.accelerated_algorithm : 0.000094s : 0.04% optimize.opt_a.shard : 0.000006s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000018s : 0.01% optimize.opt_a.shard_inline : 0.000050s : 0.02% optimize.opt_a.merge_send_recv : 0.000051s : 0.02% optimize.opt_a.auto_parallel : 0.000049s : 0.02% optimize.opt_a.parallel : 0.000048s : 0.02% optimize.opt_a.flash_sp : 0.000025s : 0.01% optimize.opt_a.merge_comm : 0.000030s : 0.01% optimize.opt_a.allreduce_fusion : 0.000025s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000075s : 0.03% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000004s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000148s : 0.06% optimize.opt_a.virtual_dataset : 0.000053s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000048s : 0.02% optimize.opt_a.virtual_output : 0.000046s : 0.02% optimize.opt_a.merge_forward : 0.000032s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000008s : 0.00% optimize.opt_a.offload_activation : 0.000061s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000103s : 0.04% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000086s : 0.03% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000037s : 0.01% optimize.opt_a.meta_fg_expand : 0.008541s : 3.31% optimize.opt_a.flash_sp_send_recv_attached : 0.000011s : 0.00% optimize.opt_a.receive_attached : 0.000007s : 0.00% optimize.opt_a.after_resolve : 0.000570s : 0.22% optimize.opt_a.a_after_grad : 0.000389s : 0.15% optimize.opt_a.renormalize : 0.054243s : 21.01% optimize.opt_a.add_forward_monad_depend : 0.000045s : 0.02% optimize.opt_a.auto_monad_grad : 0.000023s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000216s : 0.08% optimize.opt_a.cse : 0.000527s : 0.20% optimize.opt_a.a_3 : 0.001871s : 0.72% optimize.py_interpret_to_execute_after_opt_a : 0.000026s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000084s : 0.03% optimize.convert_after_rewriter : 0.000013s : 0.00% optimize.order_py_execute_after_rewriter : 0.000008s : 0.00% optimize.mutable_eliminate : 0.000934s : 0.36% optimize.opt_b.b_1 : 0.000265s : 0.10% optimize.opt_b.b_2 : 0.000015s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000016s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000007s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000007s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000052s : 0.02% optimize.optimize_parallel_all_gather_comm : 0.000032s : 0.01% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000041s : 0.02% optimize.loop_unroll : 0.000599s : 0.23% optimize.opt_after_cconv.c_1 : 0.000081s : 0.03% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000013s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.cse : 0.000041s : 0.02% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000022s : 0.01% optimize.tuple_transform.d_1 : 0.000089s : 0.03% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000012s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000105s : 0.04% optimize.cse_after_recomputation.cse : 0.000029s : 0.01% optimize.environ_conv : 0.000014s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000009s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000007s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000025s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000008s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000006s : 0.00% optimize.overlap_grad_flash_sp : 0.000035s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000017s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000024s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000026s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000013s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000021s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000004s : 0.00% pipeline_parallel_scheduler : 0.000001s : 0.00% auto_monad_reorder : 0.000043s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.000683s : 0.26% validate : 0.000070s : 0.03% backend_pass : 0.000001s : 0.00% task_emit : 0.166189s : 64.39% execute : 0.000010s : 0.00% Time group info: ------[substitution.] 0.002815 310 0.67% : 0.000019s : 4: substitution.depend_value_elim 0.17% : 0.000005s : 7: substitution.elim_not_effective 0.24% : 0.000007s : 7: substitution.float_depend_g_call 0.15% : 0.000004s : 2: substitution.float_tuple_getitem_switch 0.09% : 0.000003s : 7: substitution.fold_const_symbol 38.03% : 0.001070s : 4: substitution.getattr_setattr_resolve 0.37% : 0.000010s : 9: substitution.graph_param_transform 0.12% : 0.000003s : 2: substitution.incorporate_call 0.07% : 0.000002s : 2: substitution.incorporate_call_switch 39.79% : 0.001120s : 24: substitution.inline 2.69% : 0.000076s : 5: substitution.inline_without_move 0.56% : 0.000016s : 28: substitution.j_node_and_user_rematch 0.87% : 0.000025s : 3: substitution.less_batch_normalization 0.52% : 0.000015s : 9: substitution.minmaximum_grad 0.35% : 0.000010s : 7: substitution.partial_eliminate 0.75% : 0.000021s : 28: substitution.remove_not_recompute_node 3.86% : 0.000109s : 38: substitution.replace_applicator 0.85% : 0.000024s : 19: substitution.replace_old_param 0.17% : 0.000005s : 1: substitution.set_cell_output_no_recompute 1.07% : 0.000030s : 9: substitution.tuple_list_convert_item_index_to_positive 0.49% : 0.000014s : 9: substitution.tuple_list_get_item_const_eliminator 0.69% : 0.000019s : 9: substitution.tuple_list_get_item_depend_reorder 3.66% : 0.000103s : 31: substitution.tuple_list_get_item_eliminator 0.66% : 0.000018s : 9: substitution.tuple_list_get_set_item_eliminator 1.18% : 0.000033s : 16: substitution.updatestate_pure_node_eliminater 1.92% : 0.000054s : 21: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.012531 2 85.65% : 0.010733s : 1: type_inference.infer 14.35% : 0.001798s : 1: type_inference.specialize ------[replace.] 0.000605 49 12.31% : 0.000074s : 3: replace.getattr_setattr_resolve 40.91% : 0.000248s : 24: replace.inline 10.19% : 0.000062s : 1: replace.replace_applicator 34.58% : 0.000209s : 20: replace.tuple_list_get_item_eliminator 2.00% : 0.000012s : 1: replace.updatestate_useless_node_eliminater ------[match.] 0.002208 49 46.00% : 0.001016s : 3: match.getattr_setattr_resolve 49.88% : 0.001102s : 24: match.inline 0.77% : 0.000017s : 1: match.replace_applicator 2.54% : 0.000056s : 20: match.tuple_list_get_item_eliminator 0.81% : 0.000018s : 1: match.updatestate_useless_node_eliminater ------[predicate.] 0.001309 8683 0.87% : 0.000011s : 85: predicate.accumulaten_eliminater 0.29% : 0.000004s : 9: predicate.ad_related_special_op_eliminate 0.39% : 0.000005s : 37: predicate.addn_check_dump 0.92% : 0.000012s : 85: predicate.addn_zero_filter 0.83% : 0.000011s : 85: predicate.adjust_all_reduce_mul_add 1.77% : 0.000023s : 122: predicate.arithmetic_simplify 0.90% : 0.000012s : 85: predicate.cast_eliminate 2.61% : 0.000034s : 260: predicate.check_bprop_eliminate 0.38% : 0.000005s : 37: predicate.compare_switch_simplify 0.06% : 0.000001s : 9: predicate.const_output_eliminate 0.43% : 0.000006s : 37: predicate.depend_value_elim 0.97% : 0.000013s : 85: predicate.dict_get_item_const_eliminator 0.98% : 0.000013s : 85: predicate.dict_get_item_eliminator 0.95% : 0.000012s : 85: predicate.dict_set_item_eliminator 0.35% : 0.000005s : 18: predicate.dumpgradient_eliminate 0.07% : 0.000001s : 9: predicate.elim_not_effective 0.16% : 0.000002s : 9: predicate.elim_shapecalc_of_broadcastargs 0.98% : 0.000013s : 94: predicate.environ_add_const_eliminate 0.96% : 0.000013s : 94: predicate.environ_get_add_eliminate 0.96% : 0.000013s : 94: predicate.environ_get_depend_swap 1.38% : 0.000018s : 131: predicate.environ_get_eliminate 0.94% : 0.000012s : 94: predicate.environ_get_set_eliminate 1.36% : 0.000018s : 130: predicate.exchange_switch_depend_value 2.02% : 0.000026s : 130: predicate.float_depend_g_call 0.40% : 0.000005s : 37: predicate.float_environ_get_switch 0.50% : 0.000006s : 46: predicate.float_tuple_getitem_switch 0.06% : 0.000001s : 9: predicate.fold_const_symbol 0.46% : 0.000006s : 37: predicate.get_grad_eliminate 0.58% : 0.000008s : 20: predicate.getattr_setattr_resolve 0.08% : 0.000001s : 9: predicate.graph_param_transform 0.40% : 0.000005s : 37: predicate.incorporate_call 0.36% : 0.000005s : 37: predicate.incorporate_call_switch 4.73% : 0.000062s : 307: predicate.inline 2.06% : 0.000027s : 111: predicate.inline_without_move 0.20% : 0.000003s : 37: predicate.j_node_and_user_rematch 0.58% : 0.000008s : 37: predicate.less_batch_normalization 1.37% : 0.000018s : 123: predicate.list_to_tuple_eliminator_ 2.11% : 0.000028s : 208: predicate.load_eliminater 0.34% : 0.000004s : 9: predicate.loop_unroll_after_grad 2.08% : 0.000027s : 197: predicate.loop_unroll_before_grad 1.20% : 0.000016s : 103: predicate.make_slice_get_slice_eliminator 0.41% : 0.000005s : 37: predicate.merge_addn 2.55% : 0.000033s : 256: predicate.micro_step_allgather_replace 2.70% : 0.000035s : 256: predicate.mini_step_allgather_replace 0.95% : 0.000012s : 85: predicate.minmaximum_grad 0.38% : 0.000005s : 9: predicate.mutable_eliminate 0.12% : 0.000002s : 9: predicate.opt_reshape 0.14% : 0.000002s : 9: predicate.parallel_virtual_node 2.46% : 0.000032s : 130: predicate.partial_defer_inline 1.39% : 0.000018s : 114: predicate.partial_eliminate 0.94% : 0.000012s : 85: predicate.print_const_string_wrapper 0.43% : 0.000006s : 37: predicate.reduce_all_const_elim 1.18% : 0.000015s : 85: predicate.reduce_eliminate 2.16% : 0.000028s : 208: predicate.redundant_stop_gradient_eliminater 0.27% : 0.000003s : 37: predicate.remove_not_recompute_node 2.45% : 0.000032s : 363: predicate.replace_applicator 0.81% : 0.000011s : 111: predicate.replace_old_param 0.12% : 0.000002s : 9: predicate.reset_defer_inline 0.96% : 0.000013s : 85: predicate.reshape_eliminate 2.78% : 0.000036s : 256: predicate.row_tensor_add_zeros_like 0.16% : 0.000002s : 9: predicate.row_tensor_eliminate 3.07% : 0.000040s : 260: predicate.same_eliminate 0.29% : 0.000004s : 39: predicate.set_cell_output_no_recompute 0.64% : 0.000008s : 37: predicate.shard_identity_eliminate 0.24% : 0.000003s : 18: predicate.special_op_eliminate 0.46% : 0.000006s : 37: predicate.specialize_transform 2.78% : 0.000036s : 256: predicate.split_environ_get_set_with_tuple_value 1.60% : 0.000021s : 111: predicate.stack_unstack_eliminate 0.12% : 0.000002s : 9: predicate.switch_call_monad_eliminater 1.51% : 0.000020s : 130: predicate.switch_defer_inline 4.08% : 0.000053s : 390: predicate.switch_layer_defer_inline 4.31% : 0.000056s : 373: predicate.switch_simplify 1.05% : 0.000014s : 85: predicate.tile_eliminate 1.10% : 0.000014s : 85: predicate.transpose_eliminate 1.20% : 0.000016s : 103: predicate.tuple_list_convert_item_index_to_positive 1.28% : 0.000017s : 103: predicate.tuple_list_get_item_const_eliminator 1.19% : 0.000016s : 103: predicate.tuple_list_get_item_depend_reorder 2.45% : 0.000032s : 160: predicate.tuple_list_get_item_eliminator 1.24% : 0.000016s : 103: predicate.tuple_list_get_set_item_eliminator 1.65% : 0.000022s : 140: predicate.tuple_list_set_item_eliminator 1.34% : 0.000017s : 123: predicate.tuple_to_list_eliminator_ 2.12% : 0.000028s : 208: predicate.updatestate_pure_node_eliminater 2.63% : 0.000034s : 246: predicate.updatestate_useless_node_eliminater 0.13% : 0.000002s : 9: predicate.value_based_eliminate 0.47% : 0.000006s : 37: predicate.virtual_dataset_eliminate 0.43% : 0.000006s : 37: predicate.virtual_output_eliminate 0.10% : 0.000001s : 9: predicate.virtual_view_grad_eliminate 0.13% : 0.000002s : 9: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.004698 53 53.23% : 0.002501s : 23: func_graph_cloner_run.FuncGraphClonerGraph 46.77% : 0.002197s : 30: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.750015 247 0.00% : 0.000004s : 1: ForceFp32Comm 23.00% : 0.172536s : 1: add_attr 23.00% : 0.172519s : 1: add_attr_with_inline 0.00% : 0.000003s : 1: add_comm_op_reuse_tag 0.02% : 0.000113s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.05% : 0.000339s : 1: auto_monad 0.01% : 0.000049s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.08% : 0.000581s : 1: bootstrap 0.01% : 0.000047s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000030s : 1: control_data_broadcast_order 0.00% : 0.000017s : 1: convert_after_rewriter 0.01% : 0.000048s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000009s : 1: detach_backward 0.00% : 0.000018s : 1: environ_conv 0.01% : 0.000087s : 1: event_method 0.00% : 0.000017s : 1: execute 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000017s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000006s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000008s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000010s : 1: label_micro_interleaved_index 0.08% : 0.000618s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.13% : 0.000955s : 1: mutable_eliminate 0.00% : 0.000010s : 1: offloading_packed_experts 0.00% : 0.000029s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000033s : 1: opt.transform.mutable_eliminate 1.31% : 0.009797s : 125: opt.transform.opt_a 0.01% : 0.000079s : 1: opt.transform.opt_after_cconv 0.01% : 0.000057s : 1: opt.transform.opt_after_jit_grad 0.03% : 0.000245s : 28: opt.transform.opt_b 0.16% : 0.001225s : 2: opt.transform.opt_resolve 0.01% : 0.000098s : 2: opt.transform.opt_trans_graph 0.01% : 0.000080s : 4: opt.transform.symbol_engine_opt 10.08% : 0.075570s : 1: opt_a 0.03% : 0.000209s : 1: opt_after_cconv 0.09% : 0.000702s : 1: opt_after_jit_grad 0.06% : 0.000429s : 1: opt_b 10.55% : 0.079115s : 1: optimize 0.01% : 0.000038s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000011s : 1: order_py_execute_after_rewriter 0.01% : 0.000040s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000009s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000011s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000006s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000069s : 1: pre_auto_parallel 0.01% : 0.000093s : 1: py_interpret_to_execute 0.00% : 0.000031s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000027s : 1: remove_dup_value 1.02% : 0.007648s : 2: renormalize.infer 6.21% : 0.046561s : 2: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000094s : 1: rewriter_after_opt_a 0.02% : 0.000175s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000013s : 1: swap_dp_allreduce_reducescatter 0.02% : 0.000155s : 1: symbol_engine_optimizer 22.16% : 0.166211s : 1: task_emit 0.02% : 0.000137s : 1: tuple_transform 1.69% : 0.012679s : 1: type_inference 0.02% : 0.000123s : 1: validate . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[0-shape6] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[0-shape6],max_mem:4.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:42:33.363.325 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.160438, [24] [bootstrap]: 0.00097402 [type_inference]: 0.0880119 [event_method]: 2.956e-05 [auto_monad]: 0.00030028 [graph_reusing]: 1.02e-05 [inline]: 3.14001e-06 [add_attr]: 0.00707915, [1] [add_attr_with_inline]: 0.00705836, [1] [Cycle 1]: 9.124e-05, [2] [tag_attr]: 2.426e-05 [meta_addattr_fg_expand]: 3.4e-06 [parallel-infer-symbol]: 3.69002e-06 [pre_auto_parallel]: 4.189e-05 [insert-virtual-dataset]: 2.82002e-06 [parallel-infer-symbol-second]: 1.29998e-06 [dataset_repeat_opt]: 2.83e-06 [pipeline_split]: 1.52001e-06 [optimize]: 0.0491952, [53] [py_interpret_to_execute]: 4.005e-05 [rewriter_before_opt_a]: 6.764e-05 [opt_a]: 0.00486524, [2] [Cycle 1]: 0.00358221, [45] [expand_dump_flag]: 6.31e-06 [switch_simplify]: 4.776e-05 [loop_unroll]: 1.738e-05 [a_1]: 0.00077809 [with_stream_mark]: 3.676e-05 [recompute_prepare]: 2.234e-05 [updatestate_depend_eliminate]: 7.4e-06 [updatestate_assign_eliminate]: 6.28e-06 [updatestate_loads_eliminate]: 8.45001e-06 [parameter_eliminate]: 2.70997e-06 [a_2]: 0.00018167 [accelerated_algorithm]: 1.521e-05 [shard]: 6.76e-06 [meta_shard_fg_expand]: 3.98001e-06 [shard_inline]: 1.434e-05 [merge_send_recv]: 1.239e-05 [auto_parallel]: 1.762e-05 [parallel]: 3.791e-05 [flash_sp]: 1.211e-05 [merge_comm]: 1.029e-05 [allreduce_fusion]: 8.06001e-06 [matmul_add_comm_reduction]: 1.938e-05 [allreduce_slice_to_reducescatter]: 1.05001e-06 [virtual_shard_identity]: 2.747e-05 [virtual_dataset]: 2.024e-05 [get_grad_eliminate_]: 2.235e-05 [virtual_output]: 1.18e-05 [merge_forward]: 1.045e-05 [cell_reuse_recompute_pass]: 2.70997e-06 [offload_activation]: 2.767e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.416e-05 [merge_recompute_call_nodes]: 1.64e-06 [before_grad]: 3.185e-05 [set_forward_comm_id_for_comm_node_pass]: 1.012e-05 [meta_fg_expand]: 5.51e-06 [flash_sp_send_recv_attached]: 4.09997e-06 [receive_attached]: 2.36e-06 [after_resolve]: 2.347e-05 [a_after_grad]: 2.436e-05 [renormalize]: 0.00128664 [add_forward_monad_depend]: 1.393e-05 [auto_monad_grad]: 3.21001e-06 [auto_monad_eliminator]: 2.822e-05 [cse]: 4.237e-05 [a_3]: 7.83e-05 [Cycle 2]: 0.00125956, [45] [expand_dump_flag]: 2.41998e-06 [switch_simplify]: 1.253e-05 [loop_unroll]: 9.72001e-06 [a_1]: 0.00022718 [with_stream_mark]: 2.618e-05 [recompute_prepare]: 1.522e-05 [updatestate_depend_eliminate]: 6.96999e-06 [updatestate_assign_eliminate]: 5.81998e-06 [updatestate_loads_eliminate]: 6.23e-06 [parameter_eliminate]: 1.77001e-06 [a_2]: 0.00012415 [accelerated_algorithm]: 1.096e-05 [shard]: 2.24001e-06 [meta_shard_fg_expand]: 2.44999e-06 [shard_inline]: 1.132e-05 [merge_send_recv]: 1.014e-05 [auto_parallel]: 1.228e-05 [parallel]: 1.14e-05 [flash_sp]: 3.81001e-06 [merge_comm]: 5.55001e-06 [allreduce_fusion]: 5.30999e-06 [matmul_add_comm_reduction]: 1.816e-05 [allreduce_slice_to_reducescatter]: 5.3001e-07 [virtual_shard_identity]: 1.039e-05 [virtual_dataset]: 9.61e-06 [get_grad_eliminate_]: 1.132e-05 [virtual_output]: 1.188e-05 [merge_forward]: 6.46999e-06 [cell_reuse_recompute_pass]: 2.58e-06 [offload_activation]: 2.361e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.104e-05 [merge_recompute_call_nodes]: 1.32e-06 [before_grad]: 2.307e-05 [set_forward_comm_id_for_comm_node_pass]: 9.17999e-06 [meta_fg_expand]: 3.85998e-06 [flash_sp_send_recv_attached]: 1.55999e-06 [receive_attached]: 2.03997e-06 [after_resolve]: 2.104e-05 [a_after_grad]: 2.292e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 3.20002e-06 [auto_monad_grad]: 1.79998e-06 [auto_monad_eliminator]: 1.763e-05 [cse]: 2.955e-05 [a_3]: 6.681e-05 [py_interpret_to_execute_after_opt_a]: 3.325e-05 [slice_cell_reuse_recomputed_activation]: 2.69001e-06 [rewriter_after_opt_a]: 0.00025404 [convert_after_rewriter]: 1.417e-05 [order_py_execute_after_rewriter]: 6.94999e-06 [mutable_eliminate]: 0.00089539 [opt_b]: 0.0406899, [1] [Cycle 1]: 0.0406763, [7] [b_1]: 0.0404354 [b_2]: 2.557e-05 [updatestate_depend_eliminate]: 2.136e-05 [updatestate_assign_eliminate]: 7.18e-06 [updatestate_loads_eliminate]: 1.068e-05 [renormalize]: 1.10999e-06 [cse]: 5.148e-05 [optimize_parallel_all_gather_comm]: 3.552e-05 [overlap_param_gather]: 2.47001e-06 [cconv]: 4.23e-05 [loop_unroll]: 0.00095399 [opt_after_cconv]: 0.00024289, [1] [Cycle 1]: 0.00022851, [7] [c_1]: 8.506e-05 [parameter_eliminate]: 7.26999e-06 [updatestate_depend_eliminate]: 1.218e-05 [updatestate_assign_eliminate]: 5.82999e-06 [updatestate_loads_eliminate]: 5.43997e-06 [cse]: 4.269e-05 [renormalize]: 7.00005e-07 [remove_dup_value]: 2.168e-05 [tuple_transform]: 0.00016136, [1] [Cycle 1]: 0.00015567, [4] [d_1]: 8.889e-05 [none_parameter_eliminate]: 2.04e-06 [renormalize]: 1.69995e-07 [switch_simplify]: 1.442e-05 [partial_unused_args_eliminate]: 2.08998e-06 [add_recomputation]: 0.00011808 [cse_after_recomputation]: 3.481e-05, [1] [Cycle 1]: 2.818e-05, [1] [cse]: 2.014e-05 [environ_conv]: 9.69999e-06 [swap_dp_allreduce_reducescatter]: 1.398e-05 [bias_add_comm_swap]: 3.23e-06 [label_micro_interleaved_index]: 1.062e-05 [label_fine_grained_interleaved_index]: 2.68e-06 [merge_cast_opt]: 1.66e-06 [slice_recompute_activation]: 2.64999e-06 [micro_interleaved_order_control]: 2.81e-06 [assign_add_opt]: 1.50999e-06 [ForceFp32Comm]: 9.60019e-07 [remove_cast_before_assign_add]: 1.17e-06 [full_micro_interleaved_order_control]: 2.53e-06 [reorder_send_recv_between_fp_bp]: 2.81e-06 [comm_op_add_attrs]: 1.10999e-06 [add_comm_op_reuse_tag]: 9.39996e-07 [interleave_split_concat_branches]: 1.44e-06 [interleave_parallel_branches]: 1.46002e-06 [overlap_opt_shard_in_pipeline]: 1.35999e-06 [overlap_opt_shard_grad_in_pipeline]: 1.86998e-06 [control_data_broadcast_order]: 2.601e-05 [grouped_pairwise_exchange_alltoall]: 1.60001e-06 [offloading_packed_experts]: 6.20002e-06 [overlap_recompute_and_grad_model_parallel]: 1.017e-05 [overlap_grad_matmul_and_grad_allreduce]: 4.60001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.82001e-06 [overlap_recompute_comm]: 2.74999e-06 [overlap_grad_ring_attention]: 6.20002e-06 [overlap_grad_flash_sp]: 3.692e-05 [begin_end_overlap_inline]: 5.29981e-07 [split_matmul_comm_elemetwise]: 2.35002e-06 [split_layernorm_comm]: 1.81e-06 [handle_group_info]: 1.37e-06 [symbol_engine_optimizer]: 0.00016587, [1] [Cycle 1]: 0.00016009, [6] [build]: 6.86999e-06 [elim_shapecalc]: 3.149e-05 [elim_not_effective]: 2.074e-05 [opt_reshape]: 1.224e-05 [fold_const_symbol]: 2.052e-05 [renormalize]: 3.50003e-07 [detach_backward]: 2.23002e-06 [pipeline_parallel_scheduler]: 1.25999e-06 [auto_monad_reorder]: 3.331e-05 [get_jit_bprop_graph]: 2.37001e-06 [rewriter_after_jit_bprop_graph]: 6.84999e-06 [opt_after_jit_grad]: 0.00078518 [validate]: 7.221e-05 [backend_pass]: 9.5999e-07 [task_emit]: 0.0134886 [execute]: 1.058e-05 Sums bootstrap : 0.000974s : 0.64% type_inference : 0.088012s : 58.20% event_method : 0.000030s : 0.02% auto_monad : 0.000300s : 0.20% graph_reusing : 0.000010s : 0.01% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000024s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000042s : 0.03% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000040s : 0.03% optimize.rewriter_before_opt_a : 0.000068s : 0.04% optimize.opt_a.expand_dump_flag : 0.000009s : 0.01% optimize.opt_a.switch_simplify : 0.000060s : 0.04% optimize.opt_a.loop_unroll : 0.000027s : 0.02% optimize.opt_a.a_1 : 0.001005s : 0.66% optimize.opt_a.with_stream_mark : 0.000063s : 0.04% optimize.opt_a.recompute_prepare : 0.000038s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000014s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000012s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000015s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.00% optimize.opt_a.a_2 : 0.000306s : 0.20% optimize.opt_a.accelerated_algorithm : 0.000026s : 0.02% optimize.opt_a.shard : 0.000009s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000006s : 0.00% optimize.opt_a.shard_inline : 0.000026s : 0.02% optimize.opt_a.merge_send_recv : 0.000023s : 0.01% optimize.opt_a.auto_parallel : 0.000030s : 0.02% optimize.opt_a.parallel : 0.000049s : 0.03% optimize.opt_a.flash_sp : 0.000016s : 0.01% optimize.opt_a.merge_comm : 0.000016s : 0.01% optimize.opt_a.allreduce_fusion : 0.000013s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000038s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000038s : 0.03% optimize.opt_a.virtual_dataset : 0.000030s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000034s : 0.02% optimize.opt_a.virtual_output : 0.000024s : 0.02% optimize.opt_a.merge_forward : 0.000017s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000051s : 0.03% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000065s : 0.04% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000055s : 0.04% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000019s : 0.01% optimize.opt_a.meta_fg_expand : 0.000009s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000006s : 0.00% optimize.opt_a.receive_attached : 0.000004s : 0.00% optimize.opt_a.after_resolve : 0.000045s : 0.03% optimize.opt_a.a_after_grad : 0.000047s : 0.03% optimize.opt_a.renormalize : 0.001287s : 0.85% optimize.opt_a.add_forward_monad_depend : 0.000017s : 0.01% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000046s : 0.03% optimize.opt_a.cse : 0.000072s : 0.05% optimize.opt_a.a_3 : 0.000145s : 0.10% optimize.py_interpret_to_execute_after_opt_a : 0.000033s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000254s : 0.17% optimize.convert_after_rewriter : 0.000014s : 0.01% optimize.order_py_execute_after_rewriter : 0.000007s : 0.00% optimize.mutable_eliminate : 0.000895s : 0.59% optimize.opt_b.b_1 : 0.040435s : 26.74% optimize.opt_b.b_2 : 0.000026s : 0.02% optimize.opt_b.updatestate_depend_eliminate : 0.000021s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000007s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000011s : 0.01% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000051s : 0.03% optimize.optimize_parallel_all_gather_comm : 0.000036s : 0.02% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000042s : 0.03% optimize.loop_unroll : 0.000954s : 0.63% optimize.opt_after_cconv.c_1 : 0.000085s : 0.06% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000012s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.cse : 0.000043s : 0.03% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000022s : 0.01% optimize.tuple_transform.d_1 : 0.000089s : 0.06% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000014s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000118s : 0.08% optimize.cse_after_recomputation.cse : 0.000020s : 0.01% optimize.environ_conv : 0.000010s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000014s : 0.01% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000011s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000026s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000006s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000010s : 0.01% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000005s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000006s : 0.00% optimize.overlap_grad_flash_sp : 0.000037s : 0.02% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000007s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000031s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000021s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000012s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000021s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000001s : 0.00% auto_monad_reorder : 0.000033s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.00% opt_after_jit_grad : 0.000785s : 0.52% validate : 0.000072s : 0.05% backend_pass : 0.000001s : 0.00% task_emit : 0.013489s : 8.92% execute : 0.000011s : 0.01% Time group info: ------[substitution.] 0.000248 55 4.29% : 0.000011s : 2: substitution.depend_value_elim 1.15% : 0.000003s : 5: substitution.elim_not_effective 0.88% : 0.000002s : 5: substitution.fold_const_symbol 3.04% : 0.000008s : 6: substitution.graph_param_transform 59.45% : 0.000147s : 1: substitution.inline 2.92% : 0.000007s : 10: substitution.j_node_and_user_rematch 4.09% : 0.000010s : 10: substitution.remove_not_recompute_node 2.40% : 0.000006s : 2: substitution.replace_old_param 11.45% : 0.000028s : 6: substitution.updatestate_pure_node_eliminater 10.33% : 0.000026s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.087909 2 98.49% : 0.086585s : 1: type_inference.infer 1.51% : 0.001324s : 1: type_inference.specialize ------[replace.] 0.000022 1 100.00% : 0.000022s : 1: replace.inline ------[match.] 0.000142 1 100.00% : 0.000142s : 1: match.inline ------[predicate.] 0.000276 1419 0.79% : 0.000002s : 13: predicate.accumulaten_eliminater 1.31% : 0.000004s : 6: predicate.ad_related_special_op_eliminate 0.68% : 0.000002s : 12: predicate.addn_check_dump 0.72% : 0.000002s : 13: predicate.addn_zero_filter 0.61% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 2.13% : 0.000006s : 25: predicate.arithmetic_simplify 0.76% : 0.000002s : 13: predicate.cast_eliminate 0.71% : 0.000002s : 12: predicate.check_bprop_eliminate 0.78% : 0.000002s : 12: predicate.compare_switch_simplify 0.23% : 0.000001s : 6: predicate.const_output_eliminate 0.76% : 0.000002s : 12: predicate.depend_value_elim 0.77% : 0.000002s : 13: predicate.dict_get_item_const_eliminator 0.77% : 0.000002s : 13: predicate.dict_get_item_eliminator 0.72% : 0.000002s : 13: predicate.dict_set_item_eliminator 1.71% : 0.000005s : 12: predicate.dumpgradient_eliminate 0.23% : 0.000001s : 6: predicate.elim_not_effective 0.74% : 0.000002s : 6: predicate.elim_shapecalc_of_broadcastargs 1.07% : 0.000003s : 19: predicate.environ_add_const_eliminate 1.01% : 0.000003s : 19: predicate.environ_get_add_eliminate 0.98% : 0.000003s : 19: predicate.environ_get_depend_swap 1.73% : 0.000005s : 31: predicate.environ_get_eliminate 1.01% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.73% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.53% : 0.000004s : 14: predicate.float_depend_g_call 0.81% : 0.000002s : 12: predicate.float_environ_get_switch 1.01% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.19% : 0.000001s : 6: predicate.fold_const_symbol 0.77% : 0.000002s : 12: predicate.get_grad_eliminate 0.28% : 0.000001s : 6: predicate.graph_param_transform 0.71% : 0.000002s : 12: predicate.incorporate_call 0.57% : 0.000002s : 12: predicate.incorporate_call_switch 7.31% : 0.000020s : 63: predicate.inline 0.86% : 0.000002s : 12: predicate.inline_without_move 0.40% : 0.000001s : 12: predicate.j_node_and_user_rematch 0.96% : 0.000003s : 12: predicate.less_batch_normalization 1.59% : 0.000004s : 25: predicate.list_to_tuple_eliminator_ 2.13% : 0.000006s : 38: predicate.load_eliminater 1.58% : 0.000004s : 6: predicate.loop_unroll_after_grad 1.22% : 0.000003s : 17: predicate.loop_unroll_before_grad 1.78% : 0.000005s : 25: predicate.make_slice_get_slice_eliminator 0.70% : 0.000002s : 12: predicate.merge_addn 0.65% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.67% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.71% : 0.000002s : 13: predicate.minmaximum_grad 1.31% : 0.000004s : 6: predicate.mutable_eliminate 0.53% : 0.000001s : 6: predicate.opt_reshape 0.67% : 0.000002s : 6: predicate.parallel_virtual_node 1.34% : 0.000004s : 14: predicate.partial_defer_inline 1.09% : 0.000003s : 19: predicate.partial_eliminate 0.74% : 0.000002s : 13: predicate.print_const_string_wrapper 0.71% : 0.000002s : 12: predicate.reduce_all_const_elim 0.93% : 0.000003s : 13: predicate.reduce_eliminate 2.09% : 0.000006s : 38: predicate.redundant_stop_gradient_eliminater 0.62% : 0.000002s : 12: predicate.remove_not_recompute_node 0.98% : 0.000003s : 25: predicate.replace_applicator 0.48% : 0.000001s : 12: predicate.replace_old_param 0.93% : 0.000003s : 6: predicate.reset_defer_inline 0.74% : 0.000002s : 13: predicate.reshape_eliminate 0.75% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.72% : 0.000002s : 6: predicate.row_tensor_eliminate 0.87% : 0.000002s : 12: predicate.same_eliminate 0.53% : 0.000001s : 12: predicate.set_cell_output_no_recompute 0.97% : 0.000003s : 12: predicate.shard_identity_eliminate 1.20% : 0.000003s : 12: predicate.special_op_eliminate 0.76% : 0.000002s : 12: predicate.specialize_transform 1.28% : 0.000004s : 12: predicate.split_environ_get_set_with_tuple_value 1.27% : 0.000003s : 12: predicate.stack_unstack_eliminate 0.48% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.82% : 0.000002s : 14: predicate.switch_defer_inline 1.47% : 0.000004s : 26: predicate.switch_layer_defer_inline 3.83% : 0.000011s : 49: predicate.switch_simplify 0.67% : 0.000002s : 13: predicate.tile_eliminate 0.79% : 0.000002s : 13: predicate.transpose_eliminate 1.57% : 0.000004s : 25: predicate.tuple_list_convert_item_index_to_positive 1.60% : 0.000004s : 25: predicate.tuple_list_get_item_const_eliminator 1.51% : 0.000004s : 25: predicate.tuple_list_get_item_depend_reorder 3.10% : 0.000009s : 37: predicate.tuple_list_get_item_eliminator 1.66% : 0.000005s : 25: predicate.tuple_list_get_set_item_eliminator 2.45% : 0.000007s : 37: predicate.tuple_list_set_item_eliminator 1.51% : 0.000004s : 25: predicate.tuple_to_list_eliminator_ 2.06% : 0.000006s : 38: predicate.updatestate_pure_node_eliminater 4.31% : 0.000012s : 50: predicate.updatestate_useless_node_eliminater 0.72% : 0.000002s : 6: predicate.value_based_eliminate 0.74% : 0.000002s : 12: predicate.virtual_dataset_eliminate 0.99% : 0.000003s : 12: predicate.virtual_output_eliminate 0.33% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.54% : 0.000001s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000833 4 3.26% : 0.000027s : 1: func_graph_cloner_run.FuncGraphClonerGraph 96.74% : 0.000806s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.260446 196 0.00% : 0.000008s : 1: ForceFp32Comm 2.72% : 0.007086s : 1: add_attr 2.71% : 0.007066s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.05% : 0.000125s : 1: add_recomputation 0.00% : 0.000007s : 1: assign_add_opt 0.12% : 0.000325s : 1: auto_monad 0.01% : 0.000039s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.41% : 0.001056s : 1: bootstrap 0.02% : 0.000047s : 1: cconv 0.00% : 0.000011s : 1: comm_op_add_attrs 0.01% : 0.000033s : 1: control_data_broadcast_order 0.01% : 0.000019s : 1: convert_after_rewriter 0.01% : 0.000039s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.01% : 0.000014s : 1: environ_conv 0.02% : 0.000039s : 1: event_method 0.01% : 0.000020s : 1: execute 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.01% : 0.000017s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000008s : 1: handle_group_info 0.00% : 0.000010s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000007s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.01% : 0.000014s : 1: label_micro_interleaved_index 0.37% : 0.000973s : 1: loop_unroll 0.00% : 0.000006s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.35% : 0.000910s : 1: mutable_eliminate 0.00% : 0.000012s : 1: offloading_packed_experts 0.01% : 0.000029s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000031s : 1: opt.transform.mutable_eliminate 0.71% : 0.001847s : 78: opt.transform.opt_a 0.03% : 0.000083s : 1: opt.transform.opt_after_cconv 0.02% : 0.000047s : 1: opt.transform.opt_after_jit_grad 15.50% : 0.040368s : 28: opt.transform.opt_b 0.04% : 0.000100s : 2: opt.transform.opt_trans_graph 0.03% : 0.000081s : 4: opt.transform.symbol_engine_opt 1.87% : 0.004870s : 1: opt_a 0.09% : 0.000247s : 1: opt_after_cconv 0.31% : 0.000803s : 1: opt_after_jit_grad 15.63% : 0.040696s : 1: opt_b 18.89% : 0.049203s : 1: optimize 0.02% : 0.000044s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000013s : 1: order_py_execute_after_rewriter 0.02% : 0.000045s : 1: overlap_grad_flash_sp 0.00% : 0.000009s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000013s : 1: overlap_grad_ring_attention 0.00% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000009s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000014s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000009s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000004s : 1: pipeline_split 0.02% : 0.000047s : 1: pre_auto_parallel 0.02% : 0.000046s : 1: py_interpret_to_execute 0.02% : 0.000045s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.01% : 0.000029s : 1: remove_dup_value 0.26% : 0.000681s : 1: renormalize.infer 0.23% : 0.000589s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000011s : 1: rewriter_after_jit_bprop_graph 0.10% : 0.000265s : 1: rewriter_after_opt_a 0.03% : 0.000079s : 1: rewriter_before_opt_a 0.00% : 0.000007s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.01% : 0.000014s : 1: split_matmul_comm_elemetwise 0.01% : 0.000019s : 1: swap_dp_allreduce_reducescatter 0.06% : 0.000169s : 1: symbol_engine_optimizer 5.19% : 0.013516s : 1: task_emit 0.06% : 0.000165s : 1: tuple_transform 33.81% : 0.088045s : 1: type_inference 0.05% : 0.000125s : 1: validate TotalTime = 0.135417, [24] [bootstrap]: 0.00048275 [type_inference]: 0.00405637 [event_method]: 1.702e-05 [auto_monad]: 0.00015154 [graph_reusing]: 5.40001e-06 [inline]: 2.79001e-06 [add_attr]: 0.00442641, [1] [add_attr_with_inline]: 0.00441122, [1] [Cycle 1]: 7.56e-05, [2] [tag_attr]: 1.939e-05 [meta_addattr_fg_expand]: 3.71999e-06 [parallel-infer-symbol]: 4.06001e-06 [pre_auto_parallel]: 3.493e-05 [insert-virtual-dataset]: 2.92002e-06 [parallel-infer-symbol-second]: 7.29982e-07 [dataset_repeat_opt]: 2.26e-06 [pipeline_split]: 1.80001e-06 [optimize]: 0.0297829, [53] [py_interpret_to_execute]: 2.596e-05 [rewriter_before_opt_a]: 6.137e-05 [opt_a]: 0.0265516, [2] [Cycle 1]: 0.0254574, [45] [expand_dump_flag]: 2.83e-06 [switch_simplify]: 2.912e-05 [loop_unroll]: 1.567e-05 [a_1]: 0.0005152 [with_stream_mark]: 2.543e-05 [recompute_prepare]: 1.416e-05 [updatestate_depend_eliminate]: 7.93999e-06 [updatestate_assign_eliminate]: 6.02001e-06 [updatestate_loads_eliminate]: 5.81e-06 [parameter_eliminate]: 2.34001e-06 [a_2]: 0.00014081 [accelerated_algorithm]: 1.376e-05 [shard]: 6.47001e-06 [meta_shard_fg_expand]: 3.38e-06 [shard_inline]: 1.036e-05 [merge_send_recv]: 1.202e-05 [auto_parallel]: 1.184e-05 [parallel]: 2.463e-05 [flash_sp]: 1.317e-05 [merge_comm]: 5.94e-06 [allreduce_fusion]: 5.76998e-06 [matmul_add_comm_reduction]: 1.467e-05 [allreduce_slice_to_reducescatter]: 6.19999e-07 [virtual_shard_identity]: 1.551e-05 [virtual_dataset]: 1.104e-05 [get_grad_eliminate_]: 1.074e-05 [virtual_output]: 9.87999e-06 [merge_forward]: 6.68003e-06 [cell_reuse_recompute_pass]: 1.81e-06 [offload_activation]: 1.629e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.497e-05 [merge_recompute_call_nodes]: 1.77001e-06 [before_grad]: 1.896e-05 [set_forward_comm_id_for_comm_node_pass]: 6.91999e-06 [meta_fg_expand]: 3.91999e-06 [flash_sp_send_recv_attached]: 3.65e-06 [receive_attached]: 2.52001e-06 [after_resolve]: 1.692e-05 [a_after_grad]: 2.317e-05 [renormalize]: 0.0238153 [add_forward_monad_depend]: 1.174e-05 [auto_monad_grad]: 3.03e-06 [auto_monad_eliminator]: 3.389e-05 [cse]: 4.68e-05 [a_3]: 8.778e-05 [Cycle 2]: 0.00107629, [45] [expand_dump_flag]: 2.73e-06 [switch_simplify]: 1.202e-05 [loop_unroll]: 1.026e-05 [a_1]: 0.00023898 [with_stream_mark]: 2.604e-05 [recompute_prepare]: 1.104e-05 [updatestate_depend_eliminate]: 6.54999e-06 [updatestate_assign_eliminate]: 6.10002e-06 [updatestate_loads_eliminate]: 5.47999e-06 [parameter_eliminate]: 1.94e-06 [a_2]: 0.00012006 [accelerated_algorithm]: 1.046e-05 [shard]: 3.23998e-06 [meta_shard_fg_expand]: 2.72001e-06 [shard_inline]: 9.05001e-06 [merge_send_recv]: 1.103e-05 [auto_parallel]: 1.254e-05 [parallel]: 1.092e-05 [flash_sp]: 4.47e-06 [merge_comm]: 6.02999e-06 [allreduce_fusion]: 5.29998e-06 [matmul_add_comm_reduction]: 1.376e-05 [allreduce_slice_to_reducescatter]: 9.80013e-07 [virtual_shard_identity]: 1.051e-05 [virtual_dataset]: 8.87e-06 [get_grad_eliminate_]: 9.57001e-06 [virtual_output]: 9.15999e-06 [merge_forward]: 9.47001e-06 [cell_reuse_recompute_pass]: 3.39001e-06 [offload_activation]: 1.559e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.927e-05 [merge_recompute_call_nodes]: 1.71e-06 [before_grad]: 1.641e-05 [set_forward_comm_id_for_comm_node_pass]: 7.63001e-06 [meta_fg_expand]: 4.54002e-06 [flash_sp_send_recv_attached]: 2.05002e-06 [receive_attached]: 3.03998e-06 [after_resolve]: 1.467e-05 [a_after_grad]: 1.65e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 1.98997e-06 [auto_monad_grad]: 1.81e-06 [auto_monad_eliminator]: 1.614e-05 [cse]: 2.639e-05 [a_3]: 5.407e-05 [py_interpret_to_execute_after_opt_a]: 2.35e-05 [slice_cell_reuse_recomputed_activation]: 2.70997e-06 [rewriter_after_opt_a]: 0.00021622 [convert_after_rewriter]: 1.321e-05 [order_py_execute_after_rewriter]: 7.13998e-06 [mutable_eliminate]: 0.00081469 [opt_b]: 0.00037034, [1] [Cycle 1]: 0.00036081, [7] [b_1]: 0.00023748 [b_2]: 1.149e-05 [updatestate_depend_eliminate]: 1.477e-05 [updatestate_assign_eliminate]: 5.10999e-06 [updatestate_loads_eliminate]: 5.29998e-06 [renormalize]: 8.70001e-07 [cse]: 3.413e-05 [optimize_parallel_all_gather_comm]: 2.708e-05 [overlap_param_gather]: 2.09999e-06 [cconv]: 3.7e-05 [loop_unroll]: 0.00057947 [opt_after_cconv]: 0.00017553, [1] [Cycle 1]: 0.00016649, [7] [c_1]: 6.451e-05 [parameter_eliminate]: 4.43001e-06 [updatestate_depend_eliminate]: 1.004e-05 [updatestate_assign_eliminate]: 5.46002e-06 [updatestate_loads_eliminate]: 4.90999e-06 [cse]: 3.257e-05 [renormalize]: 1.14998e-06 [remove_dup_value]: 1.566e-05 [tuple_transform]: 0.00011732, [1] [Cycle 1]: 0.0001115, [4] [d_1]: 7.35e-05 [none_parameter_eliminate]: 2.07999e-06 [renormalize]: 2.19996e-07 [switch_simplify]: 1.059e-05 [partial_unused_args_eliminate]: 1.89e-06 [add_recomputation]: 9.106e-05 [cse_after_recomputation]: 3.111e-05, [1] [Cycle 1]: 2.381e-05, [1] [cse]: 1.683e-05 [environ_conv]: 7.6e-06 [swap_dp_allreduce_reducescatter]: 9.96e-06 [bias_add_comm_swap]: 2.91e-06 [label_micro_interleaved_index]: 4.67e-06 [label_fine_grained_interleaved_index]: 3.31001e-06 [merge_cast_opt]: 1.48002e-06 [slice_recompute_activation]: 2.14e-06 [micro_interleaved_order_control]: 2.60002e-06 [assign_add_opt]: 1.28002e-06 [ForceFp32Comm]: 9.20001e-07 [remove_cast_before_assign_add]: 1.22e-06 [full_micro_interleaved_order_control]: 2.48e-06 [reorder_send_recv_between_fp_bp]: 2.73e-06 [comm_op_add_attrs]: 1.18001e-06 [add_comm_op_reuse_tag]: 1.02e-06 [interleave_split_concat_branches]: 1.29e-06 [interleave_parallel_branches]: 1.30001e-06 [overlap_opt_shard_in_pipeline]: 1.45999e-06 [overlap_opt_shard_grad_in_pipeline]: 1.69e-06 [control_data_broadcast_order]: 1.973e-05 [grouped_pairwise_exchange_alltoall]: 1.74e-06 [offloading_packed_experts]: 5.89e-06 [overlap_recompute_and_grad_model_parallel]: 6.66e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.41002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.60999e-06 [overlap_recompute_comm]: 2.78998e-06 [overlap_grad_ring_attention]: 6.48e-06 [overlap_grad_flash_sp]: 3.044e-05 [begin_end_overlap_inline]: 5.09986e-07 [split_matmul_comm_elemetwise]: 2.84001e-06 [split_layernorm_comm]: 1.86e-06 [handle_group_info]: 1.10999e-06 [symbol_engine_optimizer]: 0.00019905, [1] [Cycle 1]: 0.00019329, [6] [build]: 6.09999e-06 [elim_shapecalc]: 2.217e-05 [elim_not_effective]: 2e-05 [opt_reshape]: 1.234e-05 [fold_const_symbol]: 1.692e-05 [renormalize]: 5.19998e-07 [detach_backward]: 2.59999e-06 [pipeline_parallel_scheduler]: 1.49e-06 [auto_monad_reorder]: 3.45e-05 [get_jit_bprop_graph]: 1.81e-06 [rewriter_after_jit_bprop_graph]: 7.42998e-06 [opt_after_jit_grad]: 0.00065602 [validate]: 6.283e-05 [backend_pass]: 1.00999e-06 [task_emit]: 0.0953663 [execute]: 1.188e-05 Sums bootstrap : 0.000483s : 0.37% type_inference : 0.004056s : 3.13% event_method : 0.000017s : 0.01% auto_monad : 0.000152s : 0.12% graph_reusing : 0.000005s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000019s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000035s : 0.03% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000026s : 0.02% optimize.rewriter_before_opt_a : 0.000061s : 0.05% optimize.opt_a.expand_dump_flag : 0.000006s : 0.00% optimize.opt_a.switch_simplify : 0.000041s : 0.03% optimize.opt_a.loop_unroll : 0.000026s : 0.02% optimize.opt_a.a_1 : 0.000754s : 0.58% optimize.opt_a.with_stream_mark : 0.000051s : 0.04% optimize.opt_a.recompute_prepare : 0.000025s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000014s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000012s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000011s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.00% optimize.opt_a.a_2 : 0.000261s : 0.20% optimize.opt_a.accelerated_algorithm : 0.000024s : 0.02% optimize.opt_a.shard : 0.000010s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000006s : 0.00% optimize.opt_a.shard_inline : 0.000019s : 0.01% optimize.opt_a.merge_send_recv : 0.000023s : 0.02% optimize.opt_a.auto_parallel : 0.000024s : 0.02% optimize.opt_a.parallel : 0.000036s : 0.03% optimize.opt_a.flash_sp : 0.000018s : 0.01% optimize.opt_a.merge_comm : 0.000012s : 0.01% optimize.opt_a.allreduce_fusion : 0.000011s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000028s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000026s : 0.02% optimize.opt_a.virtual_dataset : 0.000020s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000020s : 0.02% optimize.opt_a.virtual_output : 0.000019s : 0.01% optimize.opt_a.merge_forward : 0.000016s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000032s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000044s : 0.03% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000035s : 0.03% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000015s : 0.01% optimize.opt_a.meta_fg_expand : 0.000008s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000006s : 0.00% optimize.opt_a.receive_attached : 0.000006s : 0.00% optimize.opt_a.after_resolve : 0.000032s : 0.02% optimize.opt_a.a_after_grad : 0.000040s : 0.03% optimize.opt_a.renormalize : 0.023815s : 18.40% optimize.opt_a.add_forward_monad_depend : 0.000014s : 0.01% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000050s : 0.04% optimize.opt_a.cse : 0.000073s : 0.06% optimize.opt_a.a_3 : 0.000142s : 0.11% optimize.py_interpret_to_execute_after_opt_a : 0.000023s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000216s : 0.17% optimize.convert_after_rewriter : 0.000013s : 0.01% optimize.order_py_execute_after_rewriter : 0.000007s : 0.01% optimize.mutable_eliminate : 0.000815s : 0.63% optimize.opt_b.b_1 : 0.000237s : 0.18% optimize.opt_b.b_2 : 0.000011s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000015s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000005s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000034s : 0.03% optimize.optimize_parallel_all_gather_comm : 0.000027s : 0.02% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000037s : 0.03% optimize.loop_unroll : 0.000579s : 0.45% optimize.opt_after_cconv.c_1 : 0.000065s : 0.05% optimize.opt_after_cconv.parameter_eliminate : 0.000004s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000010s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.cse : 0.000033s : 0.03% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000016s : 0.01% optimize.tuple_transform.d_1 : 0.000074s : 0.06% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000011s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000091s : 0.07% optimize.cse_after_recomputation.cse : 0.000017s : 0.01% optimize.environ_conv : 0.000008s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000010s : 0.01% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000020s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000006s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 0.01% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000006s : 0.01% optimize.overlap_grad_flash_sp : 0.000030s : 0.02% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000006s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000022s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000020s : 0.02% optimize.symbol_engine_optimizer.opt_reshape : 0.000012s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000017s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000001s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000001s : 0.00% auto_monad_reorder : 0.000035s : 0.03% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.01% opt_after_jit_grad : 0.000656s : 0.51% validate : 0.000063s : 0.05% backend_pass : 0.000001s : 0.00% task_emit : 0.095366s : 73.67% execute : 0.000012s : 0.01% Time group info: ------[substitution.] 0.000195 55 5.89% : 0.000011s : 2: substitution.depend_value_elim 1.56% : 0.000003s : 5: substitution.elim_not_effective 1.28% : 0.000002s : 5: substitution.fold_const_symbol 4.07% : 0.000008s : 6: substitution.graph_param_transform 57.69% : 0.000113s : 1: substitution.inline 2.84% : 0.000006s : 10: substitution.j_node_and_user_rematch 4.27% : 0.000008s : 10: substitution.remove_not_recompute_node 3.08% : 0.000006s : 2: substitution.replace_old_param 10.95% : 0.000021s : 6: substitution.updatestate_pure_node_eliminater 8.38% : 0.000016s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.003996 2 87.58% : 0.003500s : 1: type_inference.infer 12.42% : 0.000496s : 1: type_inference.specialize ------[replace.] 0.000020 1 100.00% : 0.000020s : 1: replace.inline ------[match.] 0.000111 1 100.00% : 0.000111s : 1: match.inline ------[predicate.] 0.000255 1419 0.87% : 0.000002s : 13: predicate.accumulaten_eliminater 1.18% : 0.000003s : 6: predicate.ad_related_special_op_eliminate 0.65% : 0.000002s : 12: predicate.addn_check_dump 1.01% : 0.000003s : 13: predicate.addn_zero_filter 0.68% : 0.000002s : 13: predicate.adjust_all_reduce_mul_add 2.34% : 0.000006s : 25: predicate.arithmetic_simplify 0.87% : 0.000002s : 13: predicate.cast_eliminate 0.89% : 0.000002s : 12: predicate.check_bprop_eliminate 0.66% : 0.000002s : 12: predicate.compare_switch_simplify 0.20% : 0.000001s : 6: predicate.const_output_eliminate 0.80% : 0.000002s : 12: predicate.depend_value_elim 0.78% : 0.000002s : 13: predicate.dict_get_item_const_eliminator 0.94% : 0.000002s : 13: predicate.dict_get_item_eliminator 0.77% : 0.000002s : 13: predicate.dict_set_item_eliminator 1.26% : 0.000003s : 12: predicate.dumpgradient_eliminate 0.34% : 0.000001s : 6: predicate.elim_not_effective 0.49% : 0.000001s : 6: predicate.elim_shapecalc_of_broadcastargs 1.17% : 0.000003s : 19: predicate.environ_add_const_eliminate 1.04% : 0.000003s : 19: predicate.environ_get_add_eliminate 1.05% : 0.000003s : 19: predicate.environ_get_depend_swap 1.85% : 0.000005s : 31: predicate.environ_get_eliminate 1.10% : 0.000003s : 19: predicate.environ_get_set_eliminate 0.81% : 0.000002s : 14: predicate.exchange_switch_depend_value 1.60% : 0.000004s : 14: predicate.float_depend_g_call 0.64% : 0.000002s : 12: predicate.float_environ_get_switch 0.96% : 0.000002s : 18: predicate.float_tuple_getitem_switch 0.20% : 0.000000s : 6: predicate.fold_const_symbol 0.82% : 0.000002s : 12: predicate.get_grad_eliminate 0.28% : 0.000001s : 6: predicate.graph_param_transform 0.71% : 0.000002s : 12: predicate.incorporate_call 0.60% : 0.000002s : 12: predicate.incorporate_call_switch 5.34% : 0.000014s : 63: predicate.inline 0.86% : 0.000002s : 12: predicate.inline_without_move 0.36% : 0.000001s : 12: predicate.j_node_and_user_rematch 1.37% : 0.000003s : 12: predicate.less_batch_normalization 1.70% : 0.000004s : 25: predicate.list_to_tuple_eliminator_ 2.36% : 0.000006s : 38: predicate.load_eliminater 1.37% : 0.000003s : 6: predicate.loop_unroll_after_grad 1.15% : 0.000003s : 17: predicate.loop_unroll_before_grad 1.76% : 0.000004s : 25: predicate.make_slice_get_slice_eliminator 0.78% : 0.000002s : 12: predicate.merge_addn 0.70% : 0.000002s : 12: predicate.micro_step_allgather_replace 0.81% : 0.000002s : 12: predicate.mini_step_allgather_replace 0.76% : 0.000002s : 13: predicate.minmaximum_grad 1.37% : 0.000003s : 6: predicate.mutable_eliminate 0.41% : 0.000001s : 6: predicate.opt_reshape 0.49% : 0.000001s : 6: predicate.parallel_virtual_node 1.20% : 0.000003s : 14: predicate.partial_defer_inline 1.17% : 0.000003s : 19: predicate.partial_eliminate 0.96% : 0.000002s : 13: predicate.print_const_string_wrapper 0.73% : 0.000002s : 12: predicate.reduce_all_const_elim 1.07% : 0.000003s : 13: predicate.reduce_eliminate 2.16% : 0.000006s : 38: predicate.redundant_stop_gradient_eliminater 0.66% : 0.000002s : 12: predicate.remove_not_recompute_node 1.36% : 0.000003s : 25: predicate.replace_applicator 0.68% : 0.000002s : 12: predicate.replace_old_param 0.46% : 0.000001s : 6: predicate.reset_defer_inline 0.86% : 0.000002s : 13: predicate.reshape_eliminate 0.74% : 0.000002s : 12: predicate.row_tensor_add_zeros_like 0.48% : 0.000001s : 6: predicate.row_tensor_eliminate 1.15% : 0.000003s : 12: predicate.same_eliminate 0.54% : 0.000001s : 12: predicate.set_cell_output_no_recompute 1.25% : 0.000003s : 12: predicate.shard_identity_eliminate 0.91% : 0.000002s : 12: predicate.special_op_eliminate 0.87% : 0.000002s : 12: predicate.specialize_transform 1.20% : 0.000003s : 12: predicate.split_environ_get_set_with_tuple_value 1.21% : 0.000003s : 12: predicate.stack_unstack_eliminate 0.40% : 0.000001s : 6: predicate.switch_call_monad_eliminater 0.94% : 0.000002s : 14: predicate.switch_defer_inline 1.56% : 0.000004s : 26: predicate.switch_layer_defer_inline 3.53% : 0.000009s : 49: predicate.switch_simplify 0.96% : 0.000002s : 13: predicate.tile_eliminate 0.89% : 0.000002s : 13: predicate.transpose_eliminate 1.61% : 0.000004s : 25: predicate.tuple_list_convert_item_index_to_positive 1.77% : 0.000005s : 25: predicate.tuple_list_get_item_const_eliminator 1.44% : 0.000004s : 25: predicate.tuple_list_get_item_depend_reorder 3.46% : 0.000009s : 37: predicate.tuple_list_get_item_eliminator 1.69% : 0.000004s : 25: predicate.tuple_list_get_set_item_eliminator 2.49% : 0.000006s : 37: predicate.tuple_list_set_item_eliminator 1.60% : 0.000004s : 25: predicate.tuple_to_list_eliminator_ 2.32% : 0.000006s : 38: predicate.updatestate_pure_node_eliminater 3.35% : 0.000009s : 50: predicate.updatestate_useless_node_eliminater 0.49% : 0.000001s : 6: predicate.value_based_eliminate 0.84% : 0.000002s : 12: predicate.virtual_dataset_eliminate 0.88% : 0.000002s : 12: predicate.virtual_output_eliminate 0.37% : 0.000001s : 6: predicate.virtual_view_grad_eliminate 0.55% : 0.000001s : 6: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000530 4 4.47% : 0.000024s : 1: func_graph_cloner_run.FuncGraphClonerGraph 95.53% : 0.000506s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.195241 196 0.00% : 0.000004s : 1: ForceFp32Comm 2.27% : 0.004434s : 1: add_attr 2.26% : 0.004418s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.05% : 0.000096s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.08% : 0.000159s : 1: auto_monad 0.02% : 0.000040s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.27% : 0.000527s : 1: bootstrap 0.02% : 0.000041s : 1: cconv 0.00% : 0.000005s : 1: comm_op_add_attrs 0.01% : 0.000024s : 1: control_data_broadcast_order 0.01% : 0.000017s : 1: convert_after_rewriter 0.02% : 0.000034s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.01% : 0.000013s : 1: environ_conv 0.01% : 0.000024s : 1: event_method 0.01% : 0.000021s : 1: execute 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.01% : 0.000010s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.30% : 0.000593s : 1: loop_unroll 0.00% : 0.000006s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.43% : 0.000831s : 1: mutable_eliminate 0.01% : 0.000010s : 1: offloading_packed_experts 0.01% : 0.000022s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000024s : 1: opt.transform.mutable_eliminate 0.73% : 0.001419s : 78: opt.transform.opt_a 0.03% : 0.000063s : 1: opt.transform.opt_after_cconv 0.02% : 0.000044s : 1: opt.transform.opt_after_jit_grad 0.11% : 0.000210s : 28: opt.transform.opt_b 0.04% : 0.000082s : 2: opt.transform.opt_trans_graph 0.03% : 0.000066s : 4: opt.transform.symbol_engine_opt 13.60% : 0.026557s : 1: opt_a 0.09% : 0.000180s : 1: opt_after_cconv 0.35% : 0.000677s : 1: opt_after_jit_grad 0.19% : 0.000375s : 1: opt_b 15.26% : 0.029790s : 1: optimize 0.02% : 0.000032s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000011s : 1: order_py_execute_after_rewriter 0.02% : 0.000035s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000010s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000010s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.02% : 0.000040s : 1: pre_auto_parallel 0.02% : 0.000030s : 1: py_interpret_to_execute 0.01% : 0.000028s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.01% : 0.000020s : 1: remove_dup_value 11.84% : 0.023123s : 1: renormalize.infer 0.34% : 0.000667s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000011s : 1: rewriter_after_jit_bprop_graph 0.12% : 0.000227s : 1: rewriter_after_opt_a 0.04% : 0.000071s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000007s : 1: split_matmul_comm_elemetwise 0.01% : 0.000014s : 1: swap_dp_allreduce_reducescatter 0.10% : 0.000203s : 1: symbol_engine_optimizer 48.86% : 0.095390s : 1: task_emit 0.06% : 0.000120s : 1: tuple_transform 2.09% : 0.004080s : 1: type_inference 0.06% : 0.000112s : 1: validate TotalTime = 0.484344, [24] [bootstrap]: 0.00050204 [type_inference]: 0.0115782 [event_method]: 7.167e-05 [auto_monad]: 0.00031212 [graph_reusing]: 9.62999e-06 [inline]: 3.22002e-06 [add_attr]: 0.128588, [1] [add_attr_with_inline]: 0.128572, [1] [Cycle 1]: 0.00010671, [2] [tag_attr]: 4.721e-05 [meta_addattr_fg_expand]: 1.028e-05 [parallel-infer-symbol]: 3.7e-06 [pre_auto_parallel]: 6.454e-05 [insert-virtual-dataset]: 2.59999e-06 [parallel-infer-symbol-second]: 1.89e-06 [dataset_repeat_opt]: 2.11e-06 [pipeline_split]: 1.89e-06 [optimize]: 0.150337, [53] [py_interpret_to_execute]: 6.442e-05 [rewriter_before_opt_a]: 0.00016035 [opt_a]: 0.147098, [3] [Cycle 1]: 0.139079, [45] [expand_dump_flag]: 8.84e-06 [switch_simplify]: 8.317e-05 [loop_unroll]: 7.191e-05 [a_1]: 0.00174471 [with_stream_mark]: 3.828e-05 [recompute_prepare]: 3.505e-05 [updatestate_depend_eliminate]: 1.324e-05 [updatestate_assign_eliminate]: 1.086e-05 [updatestate_loads_eliminate]: 1.042e-05 [parameter_eliminate]: 3.20998e-06 [a_2]: 0.0003486 [accelerated_algorithm]: 4.728e-05 [shard]: 2.74999e-06 [meta_shard_fg_expand]: 5.47001e-06 [shard_inline]: 2.259e-05 [merge_send_recv]: 2.354e-05 [auto_parallel]: 1.847e-05 [parallel]: 2.282e-05 [flash_sp]: 1.802e-05 [merge_comm]: 1.262e-05 [allreduce_fusion]: 1.194e-05 [matmul_add_comm_reduction]: 4.15e-05 [allreduce_slice_to_reducescatter]: 6.90023e-07 [virtual_shard_identity]: 3.192e-05 [virtual_dataset]: 2.44e-05 [get_grad_eliminate_]: 2.397e-05 [virtual_output]: 2.319e-05 [merge_forward]: 1.413e-05 [cell_reuse_recompute_pass]: 1.66e-06 [offload_activation]: 2.537e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.609e-05 [merge_recompute_call_nodes]: 1.96998e-06 [before_grad]: 4.331e-05 [set_forward_comm_id_for_comm_node_pass]: 1.289e-05 [meta_fg_expand]: 0.00804456 [flash_sp_send_recv_attached]: 5.39e-06 [receive_attached]: 2.34999e-06 [after_resolve]: 0.00014015 [a_after_grad]: 0.00021048 [renormalize]: 0.00955699 [add_forward_monad_depend]: 2.685e-05 [auto_monad_grad]: 3.017e-05 [auto_monad_eliminator]: 0.00016146 [cse]: 0.0003532 [a_3]: 0.117221 [Cycle 2]: 0.00679567, [45] [expand_dump_flag]: 6.28e-06 [switch_simplify]: 0.00011686 [loop_unroll]: 0.0001036 [a_1]: 0.00335704 [with_stream_mark]: 3.521e-05 [recompute_prepare]: 2.344e-05 [updatestate_depend_eliminate]: 3.41e-05 [updatestate_assign_eliminate]: 9.12999e-06 [updatestate_loads_eliminate]: 6.39001e-06 [parameter_eliminate]: 2.91e-06 [a_2]: 0.00016619 [accelerated_algorithm]: 1.823e-05 [shard]: 2.36998e-06 [meta_shard_fg_expand]: 5.15999e-06 [shard_inline]: 1.149e-05 [merge_send_recv]: 1.21e-05 [auto_parallel]: 1.443e-05 [parallel]: 1.152e-05 [flash_sp]: 4.28001e-06 [merge_comm]: 6.59999e-06 [allreduce_fusion]: 6.33998e-06 [matmul_add_comm_reduction]: 1.424e-05 [allreduce_slice_to_reducescatter]: 8.59989e-07 [virtual_shard_identity]: 1.33e-05 [virtual_dataset]: 1.09e-05 [get_grad_eliminate_]: 1.076e-05 [virtual_output]: 1.054e-05 [merge_forward]: 6.54999e-06 [cell_reuse_recompute_pass]: 1.96e-06 [offload_activation]: 1.571e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.255e-05 [merge_recompute_call_nodes]: 1.96e-06 [before_grad]: 1.991e-05 [set_forward_comm_id_for_comm_node_pass]: 7.98999e-06 [meta_fg_expand]: 8.58e-05 [flash_sp_send_recv_attached]: 2.40002e-06 [receive_attached]: 2.59001e-06 [after_resolve]: 1.978e-05 [a_after_grad]: 1.787e-05 [renormalize]: 0.00194342 [add_forward_monad_depend]: 8.52e-06 [auto_monad_grad]: 3.4e-06 [auto_monad_eliminator]: 2.896e-05 [cse]: 0.00015813 [a_3]: 9.521e-05 [Cycle 3]: 0.00119776, [45] [expand_dump_flag]: 2.37001e-06 [switch_simplify]: 1.404e-05 [loop_unroll]: 1.192e-05 [a_1]: 0.00031499 [with_stream_mark]: 1.965e-05 [recompute_prepare]: 1.155e-05 [updatestate_depend_eliminate]: 7.87e-06 [updatestate_assign_eliminate]: 6.06e-06 [updatestate_loads_eliminate]: 6.34999e-06 [parameter_eliminate]: 1.79e-06 [a_2]: 0.00016571 [accelerated_algorithm]: 1.864e-05 [shard]: 1.93002e-06 [meta_shard_fg_expand]: 3.23e-06 [shard_inline]: 1.23e-05 [merge_send_recv]: 1.185e-05 [auto_parallel]: 1.291e-05 [parallel]: 8.70999e-06 [flash_sp]: 1.60999e-06 [merge_comm]: 7.85e-06 [allreduce_fusion]: 6.12999e-06 [matmul_add_comm_reduction]: 1.29e-05 [allreduce_slice_to_reducescatter]: 6.19999e-07 [virtual_shard_identity]: 1.335e-05 [virtual_dataset]: 1.303e-05 [get_grad_eliminate_]: 1.247e-05 [virtual_output]: 1.144e-05 [merge_forward]: 7.77e-06 [cell_reuse_recompute_pass]: 3.18e-06 [offload_activation]: 1.425e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.225e-05 [merge_recompute_call_nodes]: 1.54e-06 [before_grad]: 1.973e-05 [set_forward_comm_id_for_comm_node_pass]: 8.25e-06 [meta_fg_expand]: 4.92999e-06 [flash_sp_send_recv_attached]: 1.66e-06 [receive_attached]: 1.65001e-06 [after_resolve]: 1.915e-05 [a_after_grad]: 1.874e-05 [renormalize]: 7.99773e-08 [add_forward_monad_depend]: 2.16e-06 [auto_monad_grad]: 1.76e-06 [auto_monad_eliminator]: 1.995e-05 [cse]: 3.474e-05 [a_3]: 7.073e-05 [py_interpret_to_execute_after_opt_a]: 2.448e-05 [slice_cell_reuse_recomputed_activation]: 2.52001e-06 [rewriter_after_opt_a]: 7.884e-05 [convert_after_rewriter]: 1.309e-05 [order_py_execute_after_rewriter]: 7.67002e-06 [mutable_eliminate]: 0.00080217 [opt_b]: 0.0004316, [1] [Cycle 1]: 0.00042247, [7] [b_1]: 0.00028294 [b_2]: 1.575e-05 [updatestate_depend_eliminate]: 1.239e-05 [updatestate_assign_eliminate]: 6.38998e-06 [updatestate_loads_eliminate]: 5.81e-06 [renormalize]: 8.10018e-07 [cse]: 5.684e-05 [optimize_parallel_all_gather_comm]: 2.914e-05 [overlap_param_gather]: 2.11998e-06 [cconv]: 3.885e-05 [loop_unroll]: 0.00050347 [opt_after_cconv]: 0.00021935, [1] [Cycle 1]: 0.0002125, [7] [c_1]: 7.513e-05 [parameter_eliminate]: 4.41002e-06 [updatestate_depend_eliminate]: 9.15999e-06 [updatestate_assign_eliminate]: 5.71e-06 [updatestate_loads_eliminate]: 3.913e-05 [cse]: 3.822e-05 [renormalize]: 8.39995e-07 [remove_dup_value]: 1.93e-05 [tuple_transform]: 0.00013914, [1] [Cycle 1]: 0.000133, [4] [d_1]: 9.787e-05 [none_parameter_eliminate]: 2.01003e-06 [renormalize]: 2.30008e-07 [switch_simplify]: 1.25e-05 [partial_unused_args_eliminate]: 2.02999e-06 [add_recomputation]: 0.00010693 [cse_after_recomputation]: 5.099e-05, [1] [Cycle 1]: 4.5e-05, [1] [cse]: 3.886e-05 [environ_conv]: 1.236e-05 [swap_dp_allreduce_reducescatter]: 1.028e-05 [bias_add_comm_swap]: 2.86999e-06 [label_micro_interleaved_index]: 4.58999e-06 [label_fine_grained_interleaved_index]: 2.88e-06 [merge_cast_opt]: 1.67999e-06 [slice_recompute_activation]: 2.09999e-06 [micro_interleaved_order_control]: 2.81e-06 [assign_add_opt]: 1.24e-06 [ForceFp32Comm]: 9.29984e-07 [remove_cast_before_assign_add]: 8.2e-07 [full_micro_interleaved_order_control]: 3.06001e-06 [reorder_send_recv_between_fp_bp]: 2.66999e-06 [comm_op_add_attrs]: 1.19e-06 [add_comm_op_reuse_tag]: 1.04e-06 [interleave_split_concat_branches]: 1.18001e-06 [interleave_parallel_branches]: 1.10999e-06 [overlap_opt_shard_in_pipeline]: 1.29998e-06 [overlap_opt_shard_grad_in_pipeline]: 2.05002e-06 [control_data_broadcast_order]: 2.299e-05 [grouped_pairwise_exchange_alltoall]: 1.69e-06 [offloading_packed_experts]: 6.59001e-06 [overlap_recompute_and_grad_model_parallel]: 7.49002e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.34998e-06 [overlap_recompute_allgather_and_fa_grad]: 1.45001e-06 [overlap_recompute_comm]: 2.50002e-06 [overlap_grad_ring_attention]: 8.13001e-06 [overlap_grad_flash_sp]: 3.71e-05 [begin_end_overlap_inline]: 5.59987e-07 [split_matmul_comm_elemetwise]: 2.25002e-06 [split_layernorm_comm]: 2.36e-06 [handle_group_info]: 1.14e-06 [symbol_engine_optimizer]: 0.00013116, [1] [Cycle 1]: 0.00012405, [6] [build]: 1.65e-05 [elim_shapecalc]: 1.823e-05 [elim_not_effective]: 2.41e-05 [opt_reshape]: 1.36e-05 [fold_const_symbol]: 1.917e-05 [renormalize]: 2.60014e-07 [detach_backward]: 2.14999e-06 [pipeline_parallel_scheduler]: 2.09e-06 [auto_monad_reorder]: 3.913e-05 [get_jit_bprop_graph]: 2.16998e-06 [rewriter_after_jit_bprop_graph]: 5.61e-06 [opt_after_jit_grad]: 0.00059116 [validate]: 6.6e-05 [backend_pass]: 1.34998e-06 [task_emit]: 0.191857 [execute]: 1.182e-05 Sums bootstrap : 0.000502s : 0.14% type_inference : 0.011578s : 3.27% event_method : 0.000072s : 0.02% auto_monad : 0.000312s : 0.09% graph_reusing : 0.000010s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000047s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000010s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000065s : 0.02% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000002s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000064s : 0.02% optimize.rewriter_before_opt_a : 0.000160s : 0.05% optimize.opt_a.expand_dump_flag : 0.000017s : 0.00% optimize.opt_a.switch_simplify : 0.000214s : 0.06% optimize.opt_a.loop_unroll : 0.000187s : 0.05% optimize.opt_a.a_1 : 0.005417s : 1.53% optimize.opt_a.with_stream_mark : 0.000093s : 0.03% optimize.opt_a.recompute_prepare : 0.000070s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000055s : 0.02% optimize.opt_a.updatestate_assign_eliminate : 0.000026s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000023s : 0.01% optimize.opt_a.parameter_eliminate : 0.000008s : 0.00% optimize.opt_a.a_2 : 0.000680s : 0.19% optimize.opt_a.accelerated_algorithm : 0.000084s : 0.02% optimize.opt_a.shard : 0.000007s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000014s : 0.00% optimize.opt_a.shard_inline : 0.000046s : 0.01% optimize.opt_a.merge_send_recv : 0.000047s : 0.01% optimize.opt_a.auto_parallel : 0.000046s : 0.01% optimize.opt_a.parallel : 0.000043s : 0.01% optimize.opt_a.flash_sp : 0.000024s : 0.01% optimize.opt_a.merge_comm : 0.000027s : 0.01% optimize.opt_a.allreduce_fusion : 0.000024s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000069s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000059s : 0.02% optimize.opt_a.virtual_dataset : 0.000048s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000047s : 0.01% optimize.opt_a.virtual_output : 0.000045s : 0.01% optimize.opt_a.merge_forward : 0.000028s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000007s : 0.00% optimize.opt_a.offload_activation : 0.000055s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000091s : 0.03% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000083s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000029s : 0.01% optimize.opt_a.meta_fg_expand : 0.008135s : 2.30% optimize.opt_a.flash_sp_send_recv_attached : 0.000009s : 0.00% optimize.opt_a.receive_attached : 0.000007s : 0.00% optimize.opt_a.after_resolve : 0.000179s : 0.05% optimize.opt_a.a_after_grad : 0.000247s : 0.07% optimize.opt_a.renormalize : 0.011500s : 3.25% optimize.opt_a.add_forward_monad_depend : 0.000038s : 0.01% optimize.opt_a.auto_monad_grad : 0.000035s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000210s : 0.06% optimize.opt_a.cse : 0.000546s : 0.15% optimize.opt_a.a_3 : 0.117387s : 33.16% optimize.py_interpret_to_execute_after_opt_a : 0.000024s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000079s : 0.02% optimize.convert_after_rewriter : 0.000013s : 0.00% optimize.order_py_execute_after_rewriter : 0.000008s : 0.00% optimize.mutable_eliminate : 0.000802s : 0.23% optimize.opt_b.b_1 : 0.000283s : 0.08% optimize.opt_b.b_2 : 0.000016s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000012s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000057s : 0.02% optimize.optimize_parallel_all_gather_comm : 0.000029s : 0.01% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000039s : 0.01% optimize.loop_unroll : 0.000503s : 0.14% optimize.opt_after_cconv.c_1 : 0.000075s : 0.02% optimize.opt_after_cconv.parameter_eliminate : 0.000004s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000009s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000039s : 0.01% optimize.opt_after_cconv.cse : 0.000038s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000019s : 0.01% optimize.tuple_transform.d_1 : 0.000098s : 0.03% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000012s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000107s : 0.03% optimize.cse_after_recomputation.cse : 0.000039s : 0.01% optimize.environ_conv : 0.000012s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000010s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000023s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000008s : 0.00% optimize.overlap_grad_flash_sp : 0.000037s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000016s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000018s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000024s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000014s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000019s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000039s : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.000591s : 0.17% validate : 0.000066s : 0.02% backend_pass : 0.000001s : 0.00% task_emit : 0.191857s : 54.20% execute : 0.000012s : 0.00% Time group info: ------[substitution.] 0.002524 310 0.66% : 0.000017s : 4: substitution.depend_value_elim 0.13% : 0.000003s : 7: substitution.elim_not_effective 0.24% : 0.000006s : 7: substitution.float_depend_g_call 0.19% : 0.000005s : 2: substitution.float_tuple_getitem_switch 0.11% : 0.000003s : 7: substitution.fold_const_symbol 36.20% : 0.000913s : 4: substitution.getattr_setattr_resolve 0.37% : 0.000009s : 9: substitution.graph_param_transform 0.11% : 0.000003s : 2: substitution.incorporate_call 0.07% : 0.000002s : 2: substitution.incorporate_call_switch 40.41% : 0.001020s : 24: substitution.inline 1.88% : 0.000048s : 5: substitution.inline_without_move 0.62% : 0.000016s : 28: substitution.j_node_and_user_rematch 0.91% : 0.000023s : 3: substitution.less_batch_normalization 0.59% : 0.000015s : 9: substitution.minmaximum_grad 0.34% : 0.000009s : 7: substitution.partial_eliminate 0.79% : 0.000020s : 28: substitution.remove_not_recompute_node 5.11% : 0.000129s : 38: substitution.replace_applicator 0.80% : 0.000020s : 19: substitution.replace_old_param 0.14% : 0.000004s : 1: substitution.set_cell_output_no_recompute 1.16% : 0.000029s : 9: substitution.tuple_list_convert_item_index_to_positive 0.49% : 0.000012s : 9: substitution.tuple_list_get_item_const_eliminator 0.70% : 0.000018s : 9: substitution.tuple_list_get_item_depend_reorder 4.06% : 0.000102s : 31: substitution.tuple_list_get_item_eliminator 0.70% : 0.000018s : 9: substitution.tuple_list_get_set_item_eliminator 1.22% : 0.000031s : 16: substitution.updatestate_pure_node_eliminater 2.01% : 0.000051s : 21: substitution.updatestate_useless_node_eliminater ------[type_inference.] 0.011467 2 85.79% : 0.009837s : 1: type_inference.infer 14.21% : 0.001630s : 1: type_inference.specialize ------[replace.] 0.000607 49 11.16% : 0.000068s : 3: replace.getattr_setattr_resolve 39.07% : 0.000237s : 24: replace.inline 14.90% : 0.000090s : 1: replace.replace_applicator 32.59% : 0.000198s : 20: replace.tuple_list_get_item_eliminator 2.28% : 0.000014s : 1: replace.updatestate_useless_node_eliminater ------[match.] 0.001971 49 43.90% : 0.000865s : 3: match.getattr_setattr_resolve 50.79% : 0.001001s : 24: match.inline 1.44% : 0.000028s : 1: match.replace_applicator 2.90% : 0.000057s : 20: match.tuple_list_get_item_eliminator 0.97% : 0.000019s : 1: match.updatestate_useless_node_eliminater ------[predicate.] 0.001531 8683 0.77% : 0.000012s : 85: predicate.accumulaten_eliminater 0.21% : 0.000003s : 9: predicate.ad_related_special_op_eliminate 0.35% : 0.000005s : 37: predicate.addn_check_dump 0.80% : 0.000012s : 85: predicate.addn_zero_filter 0.76% : 0.000012s : 85: predicate.adjust_all_reduce_mul_add 1.47% : 0.000022s : 122: predicate.arithmetic_simplify 0.84% : 0.000013s : 85: predicate.cast_eliminate 6.96% : 0.000107s : 260: predicate.check_bprop_eliminate 0.35% : 0.000005s : 37: predicate.compare_switch_simplify 0.05% : 0.000001s : 9: predicate.const_output_eliminate 0.35% : 0.000005s : 37: predicate.depend_value_elim 0.95% : 0.000014s : 85: predicate.dict_get_item_const_eliminator 0.92% : 0.000014s : 85: predicate.dict_get_item_eliminator 0.75% : 0.000011s : 85: predicate.dict_set_item_eliminator 0.31% : 0.000005s : 18: predicate.dumpgradient_eliminate 0.09% : 0.000001s : 9: predicate.elim_not_effective 0.10% : 0.000002s : 9: predicate.elim_shapecalc_of_broadcastargs 0.82% : 0.000013s : 94: predicate.environ_add_const_eliminate 0.82% : 0.000013s : 94: predicate.environ_get_add_eliminate 0.79% : 0.000012s : 94: predicate.environ_get_depend_swap 1.18% : 0.000018s : 131: predicate.environ_get_eliminate 0.82% : 0.000013s : 94: predicate.environ_get_set_eliminate 1.17% : 0.000018s : 130: predicate.exchange_switch_depend_value 1.65% : 0.000025s : 130: predicate.float_depend_g_call 0.34% : 0.000005s : 37: predicate.float_environ_get_switch 0.43% : 0.000007s : 46: predicate.float_tuple_getitem_switch 0.05% : 0.000001s : 9: predicate.fold_const_symbol 0.39% : 0.000006s : 37: predicate.get_grad_eliminate 0.38% : 0.000006s : 20: predicate.getattr_setattr_resolve 0.05% : 0.000001s : 9: predicate.graph_param_transform 0.35% : 0.000005s : 37: predicate.incorporate_call 0.31% : 0.000005s : 37: predicate.incorporate_call_switch 3.94% : 0.000060s : 307: predicate.inline 1.43% : 0.000022s : 111: predicate.inline_without_move 0.20% : 0.000003s : 37: predicate.j_node_and_user_rematch 0.51% : 0.000008s : 37: predicate.less_batch_normalization 1.13% : 0.000017s : 123: predicate.list_to_tuple_eliminator_ 1.81% : 0.000028s : 208: predicate.load_eliminater 0.20% : 0.000003s : 9: predicate.loop_unroll_after_grad 1.74% : 0.000027s : 197: predicate.loop_unroll_before_grad 1.03% : 0.000016s : 103: predicate.make_slice_get_slice_eliminator 0.37% : 0.000006s : 37: predicate.merge_addn 2.21% : 0.000034s : 256: predicate.micro_step_allgather_replace 2.20% : 0.000034s : 256: predicate.mini_step_allgather_replace 0.75% : 0.000012s : 85: predicate.minmaximum_grad 0.26% : 0.000004s : 9: predicate.mutable_eliminate 0.10% : 0.000002s : 9: predicate.opt_reshape 0.14% : 0.000002s : 9: predicate.parallel_virtual_node 1.71% : 0.000026s : 130: predicate.partial_defer_inline 1.17% : 0.000018s : 114: predicate.partial_eliminate 0.76% : 0.000012s : 85: predicate.print_const_string_wrapper 0.37% : 0.000006s : 37: predicate.reduce_all_const_elim 0.96% : 0.000015s : 85: predicate.reduce_eliminate 1.81% : 0.000028s : 208: predicate.redundant_stop_gradient_eliminater 0.25% : 0.000004s : 37: predicate.remove_not_recompute_node 2.11% : 0.000032s : 363: predicate.replace_applicator 0.62% : 0.000009s : 111: predicate.replace_old_param 0.06% : 0.000001s : 9: predicate.reset_defer_inline 0.73% : 0.000011s : 85: predicate.reshape_eliminate 2.27% : 0.000035s : 256: predicate.row_tensor_add_zeros_like 0.11% : 0.000002s : 9: predicate.row_tensor_eliminate 2.53% : 0.000039s : 260: predicate.same_eliminate 0.23% : 0.000004s : 39: predicate.set_cell_output_no_recompute 0.55% : 0.000008s : 37: predicate.shard_identity_eliminate 0.23% : 0.000003s : 18: predicate.special_op_eliminate 0.41% : 0.000006s : 37: predicate.specialize_transform 2.34% : 0.000036s : 256: predicate.split_environ_get_set_with_tuple_value 1.23% : 0.000019s : 111: predicate.stack_unstack_eliminate 0.10% : 0.000002s : 9: predicate.switch_call_monad_eliminater 13.03% : 0.000199s : 130: predicate.switch_defer_inline 3.58% : 0.000055s : 390: predicate.switch_layer_defer_inline 3.69% : 0.000057s : 373: predicate.switch_simplify 0.81% : 0.000012s : 85: predicate.tile_eliminate 0.84% : 0.000013s : 85: predicate.transpose_eliminate 1.07% : 0.000016s : 103: predicate.tuple_list_convert_item_index_to_positive 1.06% : 0.000016s : 103: predicate.tuple_list_get_item_const_eliminator 0.93% : 0.000014s : 103: predicate.tuple_list_get_item_depend_reorder 2.01% : 0.000031s : 160: predicate.tuple_list_get_item_eliminator 1.01% : 0.000015s : 103: predicate.tuple_list_get_set_item_eliminator 1.45% : 0.000022s : 140: predicate.tuple_list_set_item_eliminator 1.09% : 0.000017s : 123: predicate.tuple_to_list_eliminator_ 1.91% : 0.000029s : 208: predicate.updatestate_pure_node_eliminater 2.36% : 0.000036s : 246: predicate.updatestate_useless_node_eliminater 0.11% : 0.000002s : 9: predicate.value_based_eliminate 0.37% : 0.000006s : 37: predicate.virtual_dataset_eliminate 0.37% : 0.000006s : 37: predicate.virtual_output_eliminate 0.08% : 0.000001s : 9: predicate.virtual_view_grad_eliminate 0.11% : 0.000002s : 9: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.003999 53 54.46% : 0.002178s : 23: func_graph_cloner_run.FuncGraphClonerGraph 45.54% : 0.001821s : 30: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.785640 247 0.00% : 0.000004s : 1: ForceFp32Comm 16.37% : 0.128595s : 1: add_attr 16.37% : 0.128576s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000112s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.04% : 0.000325s : 1: auto_monad 0.01% : 0.000043s : 1: auto_monad_reorder 0.00% : 0.000007s : 1: backend_pass 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.07% : 0.000534s : 1: bootstrap 0.01% : 0.000043s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000027s : 1: control_data_broadcast_order 0.00% : 0.000017s : 1: convert_after_rewriter 0.01% : 0.000055s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000016s : 1: environ_conv 0.01% : 0.000082s : 1: event_method 0.00% : 0.000019s : 1: execute 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000007s : 1: get_jit_bprop_graph 0.00% : 0.000015s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.07% : 0.000513s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.10% : 0.000816s : 1: mutable_eliminate 0.00% : 0.000010s : 1: offloading_packed_experts 0.00% : 0.000022s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000028s : 1: opt.transform.mutable_eliminate 1.19% : 0.009349s : 125: opt.transform.opt_a 0.01% : 0.000073s : 1: opt.transform.opt_after_cconv 0.01% : 0.000050s : 1: opt.transform.opt_after_jit_grad 0.03% : 0.000266s : 28: opt.transform.opt_b 0.13% : 0.001051s : 2: opt.transform.opt_resolve 0.01% : 0.000107s : 2: opt.transform.opt_trans_graph 0.01% : 0.000070s : 4: opt.transform.symbol_engine_opt 18.72% : 0.147102s : 1: opt_a 0.03% : 0.000223s : 1: opt_after_cconv 0.08% : 0.000605s : 1: opt_after_jit_grad 0.06% : 0.000436s : 1: opt_b 19.14% : 0.150344s : 1: optimize 0.00% : 0.000034s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000011s : 1: order_py_execute_after_rewriter 0.01% : 0.000041s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000011s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000011s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000005s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000069s : 1: pre_auto_parallel 0.01% : 0.000069s : 1: py_interpret_to_execute 0.00% : 0.000028s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000024s : 1: remove_dup_value 0.80% : 0.006324s : 2: renormalize.infer 0.65% : 0.005144s : 2: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000084s : 1: rewriter_after_opt_a 0.02% : 0.000165s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000007s : 1: split_matmul_comm_elemetwise 0.00% : 0.000014s : 1: swap_dp_allreduce_reducescatter 0.02% : 0.000134s : 1: symbol_engine_optimizer 24.42% : 0.191875s : 1: task_emit 0.02% : 0.000143s : 1: tuple_transform 1.48% : 0.011604s : 1: type_inference 0.02% : 0.000119s : 1: validate . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[0-shape7] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[0-shape7],max_mem:4.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:42:41.840.510 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[1-shape0] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[1-shape0],max_mem:4.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:42:42.500.869 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[1-shape1] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[1-shape1],max_mem:2.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:42:42.812.891 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[1-shape2] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[1-shape2],max_mem:2.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:42:43.297.874 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[1-shape3] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[1-shape3],max_mem:2.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:42:43.799.792 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[1-shape4] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[1-shape4],max_mem:2.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:42:44.315.343 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[1-shape5] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[1-shape5],max_mem:2.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:42:44.864.297 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[1-shape6] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[1-shape6],max_mem:2.0M [WARNING] ME(169060:281472967610160,MainProcess):2026-01-29-17:42:45.780.243 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. . [hook] pytest_runtest_teardown:test_sigmoid__different_dimensions[1-shape7] tests/st/tensor/test_sigmoid_.py::test_sigmoid__different_dimensions[1-shape7],max_mem:2.0M =============================== warnings summary =============================== ../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/classifier/transdata/transdata_classifier.py:222 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/classifier/transdata/transdata_classifier.py:222: DeprecationWarning: invalid escape sequence \B """ ../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:143 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:143: DeprecationWarning: invalid escape sequence \c """ ../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:170 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:170: DeprecationWarning: invalid escape sequence \c """ ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for type is zero. setattr(self, word, getattr(machar, word).flat[0]) ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for type is zero. return self._float_to_str(self.smallest_subnormal) ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for type is zero. setattr(self, word, getattr(machar, word).flat[0]) ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for type is zero. return self._float_to_str(self.smallest_subnormal) ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2.py:57 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2.py:57: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad.py:56 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad.py:56: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2_grad") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad_reduce.py:48 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad_reduce.py:48: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2_grad_reduce") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul.py:51 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul.py:51: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:51 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:51: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul_grad") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:143 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:143: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul_grad_reduce") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad.py:92 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad.py:92: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer_grad_d") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad_reduce.py:49 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad_reduce.py:49: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer_grad_d_reduce") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad.py:91 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad.py:91: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel_grad_d") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad_reduce.py:48 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad_reduce.py:48: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel_grad_d_reduce") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel.py:52 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel.py:52: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_perchannel") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel_grad.py:81 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel_grad.py:81: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_perchannel_grad") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer.py:54 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer.py:54: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_per_layer") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer_grad.py:81 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer_grad.py:81: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_per_layer_grad") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perchannel.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perchannel.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("minmax_update_perchannel") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perlayer.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perlayer.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("minmax_update_perlayer") -- Docs: https://docs.pytest.org/en/stable/warnings.html ================= 16 passed, 25 warnings in 352.19s (0:05:52) ==================