==================================================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/infer/ops/test_internal_ops, configfile: ../../../../../../../../sault/virtual_test/virtualenv_004/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 8 items test_paged_attention.py [WARNING] ME(170442:281473882689328,MainProcess):2026-01-29-17:37:46.850.258 [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 = 0.0895195, [21] [bootstrap]: 0.00086127 [type_inference]: 0.0718023 [event_method]: 4.111e-05 [auto_monad]: 8.899e-05 [graph_reusing]: 5.17e-06 [inline]: 3.48e-06 [add_attr]: 0.00921094, [1] [add_attr_with_inline]: 0.009194, [1] [Cycle 1]: 0.00010639, [2] [tag_attr]: 2.132e-05 [meta_addattr_fg_expand]: 3.86999e-06 [parallel-infer-symbol]: 4.07e-06 [pre_auto_parallel]: 4.205e-05 [insert-virtual-dataset]: 2.41e-06 [parallel-infer-symbol-second]: 7.80012e-07 [dataset_repeat_opt]: 2.26e-06 [pipeline_split]: 1.86e-06 [optimize]: 0.00660858, [53] [py_interpret_to_execute]: 2.782e-05 [rewriter_before_opt_a]: 9.704e-05 [opt_a]: 0.00356527, [2] [Cycle 1]: 0.00247627, [45] [expand_dump_flag]: 3.32002e-06 [switch_simplify]: 3.383e-05 [loop_unroll]: 1.874e-05 [a_1]: 0.00041452 [with_stream_mark]: 2.543e-05 [recompute_prepare]: 1.293e-05 [updatestate_depend_eliminate]: 6.00002e-06 [updatestate_assign_eliminate]: 5.81e-06 [updatestate_loads_eliminate]: 5.04e-06 [parameter_eliminate]: 1.86998e-06 [a_2]: 0.00013263 [accelerated_algorithm]: 1.072e-05 [shard]: 3.03e-06 [meta_shard_fg_expand]: 2.66999e-06 [shard_inline]: 9.92999e-06 [merge_send_recv]: 1.041e-05 [auto_parallel]: 9.42999e-06 [parallel]: 5.962e-05 [flash_sp]: 1.284e-05 [merge_comm]: 6.09001e-06 [allreduce_fusion]: 5.08002e-06 [matmul_add_comm_reduction]: 1.305e-05 [allreduce_slice_to_reducescatter]: 1.17e-06 [virtual_shard_identity]: 1.321e-05 [virtual_dataset]: 1.002e-05 [get_grad_eliminate_]: 1.005e-05 [virtual_output]: 9.64e-06 [merge_forward]: 5.52001e-06 [cell_reuse_recompute_pass]: 1.95001e-06 [offload_activation]: 1.208e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.869e-05 [merge_recompute_call_nodes]: 1.57001e-06 [before_grad]: 1.374e-05 [set_forward_comm_id_for_comm_node_pass]: 5.09e-06 [meta_fg_expand]: 3.45e-06 [flash_sp_send_recv_attached]: 2.94999e-06 [receive_attached]: 2.46e-06 [after_resolve]: 1.979e-05 [a_after_grad]: 1.604e-05 [renormalize]: 0.00104958 [add_forward_monad_depend]: 1.689e-05 [auto_monad_grad]: 2.63e-06 [auto_monad_eliminator]: 2.272e-05 [cse]: 8.702e-05 [a_3]: 8.06e-05 [Cycle 2]: 0.00107462, [45] [expand_dump_flag]: 2.07999e-06 [switch_simplify]: 1.328e-05 [loop_unroll]: 9.58002e-06 [a_1]: 0.00022863 [with_stream_mark]: 2.176e-05 [recompute_prepare]: 9.49e-06 [updatestate_depend_eliminate]: 6.05002e-06 [updatestate_assign_eliminate]: 5.00001e-06 [updatestate_loads_eliminate]: 3.96001e-06 [parameter_eliminate]: 2.70002e-06 [a_2]: 0.00016105 [accelerated_algorithm]: 1.101e-05 [shard]: 2.16e-06 [meta_shard_fg_expand]: 2.91e-06 [shard_inline]: 1.043e-05 [merge_send_recv]: 1.047e-05 [auto_parallel]: 1.281e-05 [parallel]: 9.44e-06 [flash_sp]: 5.69999e-06 [merge_comm]: 5.81e-06 [allreduce_fusion]: 4.90999e-06 [matmul_add_comm_reduction]: 1.28e-05 [allreduce_slice_to_reducescatter]: 8.79983e-07 [virtual_shard_identity]: 1.326e-05 [virtual_dataset]: 9.66998e-06 [get_grad_eliminate_]: 9.17999e-06 [virtual_output]: 9.19998e-06 [merge_forward]: 6.60997e-06 [cell_reuse_recompute_pass]: 3.21999e-06 [offload_activation]: 1.365e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.335e-05 [merge_recompute_call_nodes]: 1.20999e-06 [before_grad]: 1.473e-05 [set_forward_comm_id_for_comm_node_pass]: 6.50002e-06 [meta_fg_expand]: 3.7e-06 [flash_sp_send_recv_attached]: 1.82001e-06 [receive_attached]: 2.88998e-06 [after_resolve]: 2.209e-05 [a_after_grad]: 1.535e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 4.65001e-06 [auto_monad_grad]: 2.22001e-06 [auto_monad_eliminator]: 1.839e-05 [cse]: 4.395e-05 [a_3]: 6.567e-05 [py_interpret_to_execute_after_opt_a]: 2.291e-05 [slice_cell_reuse_recomputed_activation]: 2.60002e-06 [rewriter_after_opt_a]: 9.897e-05 [convert_after_rewriter]: 1.166e-05 [order_py_execute_after_rewriter]: 8.28001e-06 [mutable_eliminate]: 0.00075049 [opt_b]: 0.0003296, [1] [Cycle 1]: 0.00031807, [7] [b_1]: 0.00020035 [b_2]: 1.18e-05 [updatestate_depend_eliminate]: 1.158e-05 [updatestate_assign_eliminate]: 4.97e-06 [updatestate_loads_eliminate]: 4.61002e-06 [renormalize]: 8.59989e-07 [cse]: 4.491e-05 [optimize_parallel_all_gather_comm]: 2.39e-05 [overlap_param_gather]: 1.84e-06 [cconv]: 3.409e-05 [loop_unroll]: 0.000473 [opt_after_cconv]: 0.00015005, [1] [Cycle 1]: 0.00014257, [7] [c_1]: 4.811e-05 [parameter_eliminate]: 4.82e-06 [updatestate_depend_eliminate]: 7.36999e-06 [updatestate_assign_eliminate]: 4.34997e-06 [updatestate_loads_eliminate]: 3.83999e-06 [cse]: 3.945e-05 [renormalize]: 6.00005e-07 [remove_dup_value]: 6.789e-05 [tuple_transform]: 0.00014612, [1] [Cycle 1]: 0.00014019, [4] [d_1]: 6.977e-05 [none_parameter_eliminate]: 1.55999e-06 [renormalize]: 2.59985e-07 [switch_simplify]: 4.917e-05 [partial_unused_args_eliminate]: 2.03997e-06 [add_recomputation]: 6.67e-05 [cse_after_recomputation]: 3.369e-05, [1] [Cycle 1]: 2.883e-05, [1] [cse]: 2.321e-05 [environ_conv]: 2.106e-05 [swap_dp_allreduce_reducescatter]: 6.76999e-06 [bias_add_comm_swap]: 2.69001e-06 [label_micro_interleaved_index]: 5.05001e-06 [label_fine_grained_interleaved_index]: 2.72001e-06 [merge_cast_opt]: 1.37e-06 [slice_recompute_activation]: 2.46e-06 [micro_interleaved_order_control]: 2.76e-06 [assign_add_opt]: 1.47999e-06 [ForceFp32Comm]: 1.14e-06 [remove_cast_before_assign_add]: 1.24e-06 [full_micro_interleaved_order_control]: 2.04e-06 [reorder_send_recv_between_fp_bp]: 2.84001e-06 [comm_op_add_attrs]: 9.89996e-07 [add_comm_op_reuse_tag]: 9.80013e-07 [interleave_split_concat_branches]: 1.09e-06 [interleave_parallel_branches]: 1.01002e-06 [overlap_opt_shard_in_pipeline]: 2.308e-05 [overlap_opt_shard_grad_in_pipeline]: 2.06e-06 [control_data_broadcast_order]: 1.795e-05 [grouped_pairwise_exchange_alltoall]: 1.62001e-06 [offloading_packed_experts]: 5.46e-06 [overlap_recompute_and_grad_model_parallel]: 5.69e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.29e-06 [overlap_recompute_allgather_and_fa_grad]: 1.62999e-06 [overlap_recompute_comm]: 2.12999e-06 [overlap_grad_ring_attention]: 4.94998e-06 [overlap_grad_flash_sp]: 3.657e-05 [begin_end_overlap_inline]: 5.09986e-07 [split_matmul_comm_elemetwise]: 2.21998e-06 [split_layernorm_comm]: 1.82999e-06 [handle_group_info]: 1.34998e-06 [symbol_engine_optimizer]: 0.0002768, [1] [Cycle 1]: 0.00027141, [6] [build]: 0.00015833 [elim_shapecalc]: 1.756e-05 [elim_not_effective]: 2.993e-05 [opt_reshape]: 1.028e-05 [fold_const_symbol]: 1.966e-05 [renormalize]: 2.00002e-07 [detach_backward]: 2.36e-06 [pipeline_parallel_scheduler]: 1.62001e-06 [auto_monad_reorder]: 3.025e-05 [get_jit_bprop_graph]: 1.89e-06 [rewriter_after_jit_bprop_graph]: 5.44e-06 [opt_after_jit_grad]: 0.00053588 [validate]: 7.129e-05 Sums bootstrap : 0.000861s : 1.09% type_inference : 0.071802s : 90.65% event_method : 0.000041s : 0.05% auto_monad : 0.000089s : 0.11% graph_reusing : 0.000005s : 0.01% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000021s : 0.03% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000004s : 0.01% pre_auto_parallel : 0.000042s : 0.05% 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 : 0.000028s : 0.04% optimize.rewriter_before_opt_a : 0.000097s : 0.12% optimize.opt_a.expand_dump_flag : 0.000005s : 0.01% optimize.opt_a.switch_simplify : 0.000047s : 0.06% optimize.opt_a.loop_unroll : 0.000028s : 0.04% optimize.opt_a.a_1 : 0.000643s : 0.81% optimize.opt_a.with_stream_mark : 0.000047s : 0.06% optimize.opt_a.recompute_prepare : 0.000022s : 0.03% optimize.opt_a.updatestate_depend_eliminate : 0.000012s : 0.02% optimize.opt_a.updatestate_assign_eliminate : 0.000011s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000009s : 0.01% optimize.opt_a.parameter_eliminate : 0.000005s : 0.01% optimize.opt_a.a_2 : 0.000294s : 0.37% optimize.opt_a.accelerated_algorithm : 0.000022s : 0.03% optimize.opt_a.shard : 0.000005s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000006s : 0.01% optimize.opt_a.shard_inline : 0.000020s : 0.03% optimize.opt_a.merge_send_recv : 0.000021s : 0.03% optimize.opt_a.auto_parallel : 0.000022s : 0.03% optimize.opt_a.parallel : 0.000069s : 0.09% optimize.opt_a.flash_sp : 0.000019s : 0.02% optimize.opt_a.merge_comm : 0.000012s : 0.02% optimize.opt_a.allreduce_fusion : 0.000010s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000026s : 0.03% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000026s : 0.03% optimize.opt_a.virtual_dataset : 0.000020s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000019s : 0.02% optimize.opt_a.virtual_output : 0.000019s : 0.02% optimize.opt_a.merge_forward : 0.000012s : 0.02% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.01% optimize.opt_a.offload_activation : 0.000026s : 0.03% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000042s : 0.05% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000028s : 0.04% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000012s : 0.01% optimize.opt_a.meta_fg_expand : 0.000007s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.01% optimize.opt_a.receive_attached : 0.000005s : 0.01% optimize.opt_a.after_resolve : 0.000042s : 0.05% optimize.opt_a.a_after_grad : 0.000031s : 0.04% optimize.opt_a.renormalize : 0.001050s : 1.33% optimize.opt_a.add_forward_monad_depend : 0.000022s : 0.03% optimize.opt_a.auto_monad_grad : 0.000005s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000041s : 0.05% optimize.opt_a.cse : 0.000131s : 0.17% optimize.opt_a.a_3 : 0.000146s : 0.18% optimize.py_interpret_to_execute_after_opt_a : 0.000023s : 0.03% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000099s : 0.12% optimize.convert_after_rewriter : 0.000012s : 0.01% optimize.order_py_execute_after_rewriter : 0.000008s : 0.01% optimize.mutable_eliminate : 0.000750s : 0.95% optimize.opt_b.b_1 : 0.000200s : 0.25% optimize.opt_b.b_2 : 0.000012s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000012s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000005s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000005s : 0.01% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000045s : 0.06% optimize.optimize_parallel_all_gather_comm : 0.000024s : 0.03% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000034s : 0.04% optimize.loop_unroll : 0.000473s : 0.60% optimize.opt_after_cconv.c_1 : 0.000048s : 0.06% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.01% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000007s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000004s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000004s : 0.00% optimize.opt_after_cconv.cse : 0.000039s : 0.05% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000068s : 0.09% optimize.tuple_transform.d_1 : 0.000070s : 0.09% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000049s : 0.06% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000067s : 0.08% optimize.cse_after_recomputation.cse : 0.000023s : 0.03% optimize.environ_conv : 0.000021s : 0.03% optimize.swap_dp_allreduce_reducescatter : 0.000007s : 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 : 0.000001s : 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 : 0.000023s : 0.03% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000018s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000005s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000006s : 0.01% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000005s : 0.01% optimize.overlap_grad_flash_sp : 0.000037s : 0.05% 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 : 0.000158s : 0.20% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000018s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000030s : 0.04% optimize.symbol_engine_optimizer.opt_reshape : 0.000010s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000020s : 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 : 0.000030s : 0.04% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000005s : 0.01% opt_after_jit_grad : 0.000536s : 0.68% validate : 0.000071s : 0.09% Time group info: ------[substitution.] 0.000159 33 8.52% : 0.000014s : 2: substitution.elim_not_effective 4.63% : 0.000007s : 2: substitution.fold_const_symbol 5.12% : 0.000008s : 8: substitution.graph_param_transform 68.94% : 0.000109s : 1: substitution.inline 2.89% : 0.000005s : 4: substitution.j_node_and_user_rematch 3.60% : 0.000006s : 4: substitution.remove_not_recompute_node 6.30% : 0.000010s : 12: substitution.replace_old_param ------[type_inference.] 0.071690 2 99.02% : 0.070984s : 1: type_inference.infer 0.98% : 0.000705s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000108 1 100.00% : 0.000108s : 1: match.inline ------[predicate.] 0.000261 1879 0.79% : 0.000002s : 17: predicate.accumulaten_eliminater 0.84% : 0.000002s : 8: predicate.ad_related_special_op_eliminate 0.59% : 0.000002s : 16: predicate.addn_check_dump 0.74% : 0.000002s : 17: predicate.addn_zero_filter 0.62% : 0.000002s : 17: predicate.adjust_all_reduce_mul_add 1.71% : 0.000004s : 33: predicate.arithmetic_simplify 0.70% : 0.000002s : 17: predicate.cast_eliminate 0.70% : 0.000002s : 16: predicate.check_bprop_eliminate 0.57% : 0.000002s : 16: predicate.compare_switch_simplify 0.23% : 0.000001s : 8: predicate.const_output_eliminate 0.62% : 0.000002s : 16: predicate.depend_value_elim 0.75% : 0.000002s : 17: predicate.dict_get_item_const_eliminator 0.75% : 0.000002s : 17: predicate.dict_get_item_eliminator 0.68% : 0.000002s : 17: predicate.dict_set_item_eliminator 1.11% : 0.000003s : 16: predicate.dumpgradient_eliminate 0.26% : 0.000001s : 8: predicate.elim_not_effective 0.47% : 0.000001s : 8: predicate.elim_shapecalc_of_broadcastargs 0.98% : 0.000003s : 25: predicate.environ_add_const_eliminate 0.89% : 0.000002s : 25: predicate.environ_get_add_eliminate 0.91% : 0.000002s : 25: predicate.environ_get_depend_swap 1.54% : 0.000004s : 41: predicate.environ_get_eliminate 0.91% : 0.000002s : 25: predicate.environ_get_set_eliminate 0.66% : 0.000002s : 18: predicate.exchange_switch_depend_value 1.29% : 0.000003s : 18: predicate.float_depend_g_call 0.67% : 0.000002s : 16: predicate.float_environ_get_switch 0.90% : 0.000002s : 24: predicate.float_tuple_getitem_switch 0.23% : 0.000001s : 8: predicate.fold_const_symbol 0.80% : 0.000002s : 16: predicate.get_grad_eliminate 0.34% : 0.000001s : 8: predicate.graph_param_transform 0.59% : 0.000002s : 16: predicate.incorporate_call 0.55% : 0.000001s : 16: predicate.incorporate_call_switch 4.63% : 0.000012s : 83: predicate.inline 0.84% : 0.000002s : 16: predicate.inline_without_move 0.42% : 0.000001s : 16: predicate.j_node_and_user_rematch 0.83% : 0.000002s : 16: predicate.less_batch_normalization 1.56% : 0.000004s : 33: predicate.list_to_tuple_eliminator_ 1.87% : 0.000005s : 50: predicate.load_eliminater 1.14% : 0.000003s : 8: predicate.loop_unroll_after_grad 1.14% : 0.000003s : 25: predicate.loop_unroll_before_grad 1.61% : 0.000004s : 33: predicate.make_slice_get_slice_eliminator 0.71% : 0.000002s : 16: predicate.merge_addn 0.65% : 0.000002s : 16: predicate.micro_step_allgather_replace 0.67% : 0.000002s : 16: predicate.mini_step_allgather_replace 0.59% : 0.000002s : 17: predicate.minmaximum_grad 1.59% : 0.000004s : 8: predicate.mutable_eliminate 0.35% : 0.000001s : 8: predicate.opt_reshape 0.46% : 0.000001s : 8: predicate.parallel_virtual_node 0.96% : 0.000002s : 18: predicate.partial_defer_inline 1.01% : 0.000003s : 25: predicate.partial_eliminate 0.67% : 0.000002s : 17: predicate.print_const_string_wrapper 0.66% : 0.000002s : 16: predicate.reduce_all_const_elim 1.16% : 0.000003s : 17: predicate.reduce_eliminate 1.92% : 0.000005s : 50: predicate.redundant_stop_gradient_eliminater 0.92% : 0.000002s : 16: predicate.remove_not_recompute_node 1.46% : 0.000004s : 33: predicate.replace_applicator 0.67% : 0.000002s : 16: predicate.replace_old_param 0.33% : 0.000001s : 8: predicate.reset_defer_inline 0.67% : 0.000002s : 17: predicate.reshape_eliminate 0.66% : 0.000002s : 16: predicate.row_tensor_add_zeros_like 0.39% : 0.000001s : 8: predicate.row_tensor_eliminate 0.93% : 0.000002s : 16: predicate.same_eliminate 0.49% : 0.000001s : 16: predicate.set_cell_output_no_recompute 1.45% : 0.000004s : 16: predicate.shard_identity_eliminate 0.69% : 0.000002s : 16: predicate.special_op_eliminate 0.71% : 0.000002s : 16: predicate.specialize_transform 1.22% : 0.000003s : 16: predicate.split_environ_get_set_with_tuple_value 1.11% : 0.000003s : 16: predicate.stack_unstack_eliminate 0.41% : 0.000001s : 8: predicate.switch_call_monad_eliminater 0.73% : 0.000002s : 18: predicate.switch_defer_inline 1.41% : 0.000004s : 34: predicate.switch_layer_defer_inline 17.64% : 0.000046s : 67: predicate.switch_simplify 0.69% : 0.000002s : 17: predicate.tile_eliminate 0.63% : 0.000002s : 17: predicate.transpose_eliminate 1.40% : 0.000004s : 33: predicate.tuple_list_convert_item_index_to_positive 1.43% : 0.000004s : 33: predicate.tuple_list_get_item_const_eliminator 1.22% : 0.000003s : 33: predicate.tuple_list_get_item_depend_reorder 2.70% : 0.000007s : 49: predicate.tuple_list_get_item_eliminator 1.14% : 0.000003s : 33: predicate.tuple_list_get_set_item_eliminator 1.91% : 0.000005s : 49: predicate.tuple_list_set_item_eliminator 1.31% : 0.000003s : 33: predicate.tuple_to_list_eliminator_ 1.75% : 0.000005s : 50: predicate.updatestate_pure_node_eliminater 2.53% : 0.000007s : 66: predicate.updatestate_useless_node_eliminater 0.33% : 0.000001s : 8: predicate.value_based_eliminate 0.74% : 0.000002s : 16: predicate.virtual_dataset_eliminate 0.74% : 0.000002s : 16: predicate.virtual_output_eliminate 0.30% : 0.000001s : 8: predicate.virtual_view_grad_eliminate 0.46% : 0.000001s : 8: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000586 6 49.25% : 0.000288s : 3: func_graph_cloner_run.FuncGraphClonerGraph 50.75% : 0.000297s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.108022 192 0.00% : 0.000004s : 1: ForceFp32Comm 8.53% : 0.009217s : 1: add_attr 8.52% : 0.009200s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.07% : 0.000072s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.09% : 0.000095s : 1: auto_monad 0.03% : 0.000034s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.01% : 0.000006s : 1: bias_add_comm_swap 0.84% : 0.000903s : 1: bootstrap 0.04% : 0.000038s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.02% : 0.000021s : 1: control_data_broadcast_order 0.01% : 0.000015s : 1: convert_after_rewriter 0.03% : 0.000037s : 1: cse_after_recomputation 0.01% : 0.000005s : 1: dataset_repeat_opt 0.01% : 0.000006s : 1: detach_backward 0.02% : 0.000025s : 1: environ_conv 0.05% : 0.000051s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.01% : 0.000010s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.01% : 0.000007s : 1: inline 0.01% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.01% : 0.000006s : 1: label_fine_grained_interleaved_index 0.01% : 0.000008s : 1: label_micro_interleaved_index 0.45% : 0.000485s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.01% : 0.000006s : 1: micro_interleaved_order_control 0.71% : 0.000766s : 1: mutable_eliminate 0.01% : 0.000008s : 1: offloading_packed_experts 0.02% : 0.000021s : 1: opt.transform.loop_unroll_optimizer 0.03% : 0.000027s : 1: opt.transform.mutable_eliminate 1.21% : 0.001311s : 78: opt.transform.opt_a 0.04% : 0.000047s : 1: opt.transform.opt_after_cconv 0.03% : 0.000035s : 1: opt.transform.opt_after_jit_grad 0.17% : 0.000184s : 28: opt.transform.opt_b 0.11% : 0.000116s : 2: opt.transform.opt_trans_graph 0.07% : 0.000073s : 4: opt.transform.symbol_engine_opt 3.30% : 0.003569s : 1: opt_a 0.14% : 0.000154s : 1: opt_after_cconv 0.51% : 0.000550s : 1: opt_after_jit_grad 0.31% : 0.000335s : 1: opt_b 6.12% : 0.006614s : 1: optimize 0.03% : 0.000028s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000012s : 1: order_py_execute_after_rewriter 0.04% : 0.000040s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000008s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.02% : 0.000027s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000009s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.01% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.01% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.04% : 0.000047s : 1: pre_auto_parallel 0.03% : 0.000032s : 1: py_interpret_to_execute 0.03% : 0.000028s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.07% : 0.000073s : 1: remove_dup_value 0.38% : 0.000415s : 1: renormalize.infer 0.58% : 0.000624s : 1: renormalize.specialize 0.01% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.10% : 0.000106s : 1: rewriter_after_opt_a 0.09% : 0.000102s : 1: rewriter_before_opt_a 0.01% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000010s : 1: swap_dp_allreduce_reducescatter 0.26% : 0.000280s : 1: symbol_engine_optimizer 0.14% : 0.000149s : 1: tuple_transform 66.50% : 0.071831s : 1: type_inference .[WARNING] PRE_ACT(170442,ffffbeca8f30,python3.9):2026-01-29-17:38:38.020.559 [mindspore/ccsrc/runtime/memory/mem_pool/abstract_dynamic_mem_pool.cc:1109] FreeIdleMemsByEagerFree] Eager free count : 1, free memory : 1383031808, real free : 1384120320, not free : 0. [hook] pytest_runtest_teardown:test_paged_attention_trapezoidal[0-MASK_DEFAULT-float16] tests/st/infer/ops/test_internal_ops/test_paged_attention.py::test_paged_attention_trapezoidal[0-MASK_DEFAULT-float16],max_mem:1416.0M [WARNING] ME(170442:281473882689328,MainProcess):2026-01-29-17:38:41.823.840 [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 = 0.185701, [21] [bootstrap]: 0.00100914 [type_inference]: 0.0076744 [event_method]: 1.882e-05 [auto_monad]: 9.967e-05 [graph_reusing]: 6.01e-06 [inline]: 2.74999e-06 [add_attr]: 0.168567, [1] [add_attr_with_inline]: 0.168549, [1] [Cycle 1]: 8.39e-05, [2] [tag_attr]: 2.648e-05 [meta_addattr_fg_expand]: 4.28999e-06 [parallel-infer-symbol]: 4.2e-06 [pre_auto_parallel]: 4.124e-05 [insert-virtual-dataset]: 4.35999e-06 [parallel-infer-symbol-second]: 7.49977e-07 [dataset_repeat_opt]: 1.91998e-06 [pipeline_split]: 2.39001e-06 [optimize]: 0.00713462, [53] [py_interpret_to_execute]: 4.069e-05 [rewriter_before_opt_a]: 8.174e-05 [opt_a]: 0.00367419, [2] [Cycle 1]: 0.00252516, [45] [expand_dump_flag]: 3.51001e-06 [switch_simplify]: 3.212e-05 [loop_unroll]: 1.784e-05 [a_1]: 0.00046402 [with_stream_mark]: 2.517e-05 [recompute_prepare]: 1.334e-05 [updatestate_depend_eliminate]: 7.02002e-06 [updatestate_assign_eliminate]: 5.59e-06 [updatestate_loads_eliminate]: 5.19e-06 [parameter_eliminate]: 2.20002e-06 [a_2]: 0.00013934 [accelerated_algorithm]: 1.279e-05 [shard]: 2.73e-06 [meta_shard_fg_expand]: 3.2e-06 [shard_inline]: 1.141e-05 [merge_send_recv]: 1.285e-05 [auto_parallel]: 9.61e-06 [parallel]: 2.977e-05 [flash_sp]: 1.141e-05 [merge_comm]: 5.87999e-06 [allreduce_fusion]: 4.97999e-06 [matmul_add_comm_reduction]: 1.317e-05 [allreduce_slice_to_reducescatter]: 1.18001e-06 [virtual_shard_identity]: 1.489e-05 [virtual_dataset]: 9.96e-06 [get_grad_eliminate_]: 9.80002e-06 [virtual_output]: 9.87999e-06 [merge_forward]: 6.66e-06 [cell_reuse_recompute_pass]: 1.71e-06 [offload_activation]: 1.234e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.093e-05 [merge_recompute_call_nodes]: 1.77001e-06 [before_grad]: 3.345e-05 [set_forward_comm_id_for_comm_node_pass]: 6.73e-06 [meta_fg_expand]: 3.34001e-06 [flash_sp_send_recv_attached]: 3.14999e-06 [receive_attached]: 2.73e-06 [after_resolve]: 2.283e-05 [a_after_grad]: 1.734e-05 [renormalize]: 0.00099363 [add_forward_monad_depend]: 1.509e-05 [auto_monad_grad]: 2.35002e-06 [auto_monad_eliminator]: 2.25e-05 [cse]: 6.377e-05 [a_3]: 9.28e-05 [Cycle 2]: 0.00113372, [45] [expand_dump_flag]: 2.27999e-06 [switch_simplify]: 1.372e-05 [loop_unroll]: 1.172e-05 [a_1]: 0.00024684 [with_stream_mark]: 2.054e-05 [recompute_prepare]: 1.197e-05 [updatestate_depend_eliminate]: 5.85002e-06 [updatestate_assign_eliminate]: 5.72999e-06 [updatestate_loads_eliminate]: 6.02001e-06 [parameter_eliminate]: 1.94e-06 [a_2]: 0.00012733 [accelerated_algorithm]: 1.373e-05 [shard]: 2.11e-06 [meta_shard_fg_expand]: 2.78e-06 [shard_inline]: 1.019e-05 [merge_send_recv]: 1.004e-05 [auto_parallel]: 1.199e-05 [parallel]: 9.00001e-06 [flash_sp]: 3.71999e-06 [merge_comm]: 5.40999e-06 [allreduce_fusion]: 5.35999e-06 [matmul_add_comm_reduction]: 1.08e-05 [allreduce_slice_to_reducescatter]: 6.10016e-07 [virtual_shard_identity]: 1.301e-05 [virtual_dataset]: 1.173e-05 [get_grad_eliminate_]: 1.137e-05 [virtual_output]: 1.121e-05 [merge_forward]: 8.21002e-06 [cell_reuse_recompute_pass]: 3.21001e-06 [offload_activation]: 1.448e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.547e-05 [merge_recompute_call_nodes]: 1.07e-06 [before_grad]: 1.617e-05 [set_forward_comm_id_for_comm_node_pass]: 7.99002e-06 [meta_fg_expand]: 3.17002e-06 [flash_sp_send_recv_attached]: 1.91e-06 [receive_attached]: 2.66e-06 [after_resolve]: 2.27e-05 [a_after_grad]: 1.852e-05 [renormalize]: 7.00238e-08 [add_forward_monad_depend]: 4e-06 [auto_monad_grad]: 2.04999e-06 [auto_monad_eliminator]: 1.73e-05 [cse]: 5.533e-05 [a_3]: 6.814e-05 [py_interpret_to_execute_after_opt_a]: 3.585e-05 [slice_cell_reuse_recomputed_activation]: 2.88e-06 [rewriter_after_opt_a]: 6.768e-05 [convert_after_rewriter]: 9.96e-06 [order_py_execute_after_rewriter]: 7.83999e-06 [mutable_eliminate]: 0.00080968 [opt_b]: 0.00037804, [1] [Cycle 1]: 0.0003666, [7] [b_1]: 0.00021556 [b_2]: 1.272e-05 [updatestate_depend_eliminate]: 1.305e-05 [updatestate_assign_eliminate]: 4.87e-06 [updatestate_loads_eliminate]: 4.52998e-06 [renormalize]: 8.60018e-07 [cse]: 6.697e-05 [optimize_parallel_all_gather_comm]: 2.466e-05 [overlap_param_gather]: 1.93002e-06 [cconv]: 4.128e-05 [loop_unroll]: 0.00061451 [opt_after_cconv]: 0.00019047, [1] [Cycle 1]: 0.00018159, [7] [c_1]: 5.28e-05 [parameter_eliminate]: 5.99999e-06 [updatestate_depend_eliminate]: 1.18e-05 [updatestate_assign_eliminate]: 5.16002e-06 [updatestate_loads_eliminate]: 4.42e-06 [cse]: 5.688e-05 [renormalize]: 5.8001e-07 [remove_dup_value]: 0.00023309 [tuple_transform]: 0.0001364, [1] [Cycle 1]: 0.00012786, [4] [d_1]: 8.426e-05 [none_parameter_eliminate]: 3.2e-06 [renormalize]: 6.19999e-07 [switch_simplify]: 1.334e-05 [partial_unused_args_eliminate]: 2.93e-06 [add_recomputation]: 7.725e-05 [cse_after_recomputation]: 4.884e-05, [1] [Cycle 1]: 4.326e-05, [1] [cse]: 3.425e-05 [environ_conv]: 8.32e-06 [swap_dp_allreduce_reducescatter]: 8.10999e-06 [bias_add_comm_swap]: 3.78999e-06 [label_micro_interleaved_index]: 6.63e-06 [label_fine_grained_interleaved_index]: 2.99999e-06 [merge_cast_opt]: 1.34998e-06 [slice_recompute_activation]: 2.17999e-06 [micro_interleaved_order_control]: 2.77002e-06 [assign_add_opt]: 1.50999e-06 [ForceFp32Comm]: 1.07998e-06 [remove_cast_before_assign_add]: 9.99979e-07 [full_micro_interleaved_order_control]: 2.90998e-06 [reorder_send_recv_between_fp_bp]: 2.62001e-06 [comm_op_add_attrs]: 1.12e-06 [add_comm_op_reuse_tag]: 1.24998e-06 [interleave_split_concat_branches]: 1.29998e-06 [interleave_parallel_branches]: 1.24e-06 [overlap_opt_shard_in_pipeline]: 1.59998e-06 [overlap_opt_shard_grad_in_pipeline]: 1.84e-06 [control_data_broadcast_order]: 2.369e-05 [grouped_pairwise_exchange_alltoall]: 1.62001e-06 [offloading_packed_experts]: 6.14001e-06 [overlap_recompute_and_grad_model_parallel]: 6.19001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.37999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.74998e-06 [overlap_recompute_comm]: 2.79999e-06 [overlap_grad_ring_attention]: 6.06e-06 [overlap_grad_flash_sp]: 2.637e-05 [begin_end_overlap_inline]: 6.30011e-07 [split_matmul_comm_elemetwise]: 2.24001e-06 [split_layernorm_comm]: 2.01e-06 [handle_group_info]: 1.04e-06 [symbol_engine_optimizer]: 0.00022524, [1] [Cycle 1]: 0.00021845, [6] [build]: 9.433e-05 [elim_shapecalc]: 2.062e-05 [elim_not_effective]: 2.171e-05 [opt_reshape]: 1.092e-05 [fold_const_symbol]: 2.225e-05 [renormalize]: 7.00005e-07 [detach_backward]: 2.91999e-06 [pipeline_parallel_scheduler]: 1.62001e-06 [auto_monad_reorder]: 2.875e-05 [get_jit_bprop_graph]: 2.26998e-06 [rewriter_after_jit_bprop_graph]: 7.42998e-06 [opt_after_jit_grad]: 0.00080626 [validate]: 6.662e-05 Sums bootstrap : 0.001009s : 6.38% type_inference : 0.007674s : 48.50% event_method : 0.000019s : 0.12% auto_monad : 0.000100s : 0.63% graph_reusing : 0.000006s : 0.04% inline : 0.000003s : 0.02% add_attr.add_attr_with_inline.tag_attr : 0.000026s : 0.17% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.03% parallel-infer-symbol : 0.000004s : 0.03% pre_auto_parallel : 0.000041s : 0.26% insert-virtual-dataset : 0.000004s : 0.03% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.01% pipeline_split : 0.000002s : 0.02% optimize.py_interpret_to_execute : 0.000041s : 0.26% optimize.rewriter_before_opt_a : 0.000082s : 0.52% optimize.opt_a.expand_dump_flag : 0.000006s : 0.04% optimize.opt_a.switch_simplify : 0.000046s : 0.29% optimize.opt_a.loop_unroll : 0.000030s : 0.19% optimize.opt_a.a_1 : 0.000711s : 4.49% optimize.opt_a.with_stream_mark : 0.000046s : 0.29% optimize.opt_a.recompute_prepare : 0.000025s : 0.16% optimize.opt_a.updatestate_depend_eliminate : 0.000013s : 0.08% optimize.opt_a.updatestate_assign_eliminate : 0.000011s : 0.07% optimize.opt_a.updatestate_loads_eliminate : 0.000011s : 0.07% optimize.opt_a.parameter_eliminate : 0.000004s : 0.03% optimize.opt_a.a_2 : 0.000267s : 1.69% optimize.opt_a.accelerated_algorithm : 0.000027s : 0.17% optimize.opt_a.shard : 0.000005s : 0.03% optimize.opt_a.meta_shard_fg_expand : 0.000006s : 0.04% optimize.opt_a.shard_inline : 0.000022s : 0.14% optimize.opt_a.merge_send_recv : 0.000023s : 0.14% optimize.opt_a.auto_parallel : 0.000022s : 0.14% optimize.opt_a.parallel : 0.000039s : 0.25% optimize.opt_a.flash_sp : 0.000015s : 0.10% optimize.opt_a.merge_comm : 0.000011s : 0.07% optimize.opt_a.allreduce_fusion : 0.000010s : 0.07% optimize.opt_a.matmul_add_comm_reduction : 0.000024s : 0.15% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000028s : 0.18% optimize.opt_a.virtual_dataset : 0.000022s : 0.14% optimize.opt_a.get_grad_eliminate_ : 0.000021s : 0.13% optimize.opt_a.virtual_output : 0.000021s : 0.13% optimize.opt_a.merge_forward : 0.000015s : 0.09% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.03% optimize.opt_a.offload_activation : 0.000027s : 0.17% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000046s : 0.29% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.02% optimize.opt_a.before_grad : 0.000050s : 0.31% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000015s : 0.09% optimize.opt_a.meta_fg_expand : 0.000007s : 0.04% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.03% optimize.opt_a.receive_attached : 0.000005s : 0.03% optimize.opt_a.after_resolve : 0.000046s : 0.29% optimize.opt_a.a_after_grad : 0.000036s : 0.23% optimize.opt_a.renormalize : 0.000994s : 6.28% optimize.opt_a.add_forward_monad_depend : 0.000019s : 0.12% optimize.opt_a.auto_monad_grad : 0.000004s : 0.03% optimize.opt_a.auto_monad_eliminator : 0.000040s : 0.25% optimize.opt_a.cse : 0.000119s : 0.75% optimize.opt_a.a_3 : 0.000161s : 1.02% optimize.py_interpret_to_execute_after_opt_a : 0.000036s : 0.23% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.02% optimize.rewriter_after_opt_a : 0.000068s : 0.43% optimize.convert_after_rewriter : 0.000010s : 0.06% optimize.order_py_execute_after_rewriter : 0.000008s : 0.05% optimize.mutable_eliminate : 0.000810s : 5.12% optimize.opt_b.b_1 : 0.000216s : 1.36% optimize.opt_b.b_2 : 0.000013s : 0.08% optimize.opt_b.updatestate_depend_eliminate : 0.000013s : 0.08% optimize.opt_b.updatestate_assign_eliminate : 0.000005s : 0.03% optimize.opt_b.updatestate_loads_eliminate : 0.000005s : 0.03% optimize.opt_b.renormalize : 0.000001s : 0.01% optimize.opt_b.cse : 0.000067s : 0.42% optimize.optimize_parallel_all_gather_comm : 0.000025s : 0.16% optimize.overlap_param_gather : 0.000002s : 0.01% optimize.cconv : 0.000041s : 0.26% optimize.loop_unroll : 0.000615s : 3.88% optimize.opt_after_cconv.c_1 : 0.000053s : 0.33% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.04% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000012s : 0.07% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000005s : 0.03% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000004s : 0.03% optimize.opt_after_cconv.cse : 0.000057s : 0.36% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000233s : 1.47% optimize.tuple_transform.d_1 : 0.000084s : 0.53% optimize.tuple_transform.none_parameter_eliminate : 0.000003s : 0.02% optimize.tuple_transform.renormalize : 0.000001s : 0.00% optimize.tuple_transform.switch_simplify : 0.000013s : 0.08% optimize.partial_unused_args_eliminate : 0.000003s : 0.02% optimize.add_recomputation : 0.000077s : 0.49% optimize.cse_after_recomputation.cse : 0.000034s : 0.22% optimize.environ_conv : 0.000008s : 0.05% optimize.swap_dp_allreduce_reducescatter : 0.000008s : 0.05% optimize.bias_add_comm_swap : 0.000004s : 0.02% optimize.label_micro_interleaved_index : 0.000007s : 0.04% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.02% optimize.merge_cast_opt : 0.000001s : 0.01% optimize.slice_recompute_activation : 0.000002s : 0.01% optimize.micro_interleaved_order_control : 0.000003s : 0.02% optimize.assign_add_opt : 0.000002s : 0.01% optimize.ForceFp32Comm : 0.000001s : 0.01% optimize.remove_cast_before_assign_add : 0.000001s : 0.01% optimize.full_micro_interleaved_order_control : 0.000003s : 0.02% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.02% optimize.comm_op_add_attrs : 0.000001s : 0.01% optimize.add_comm_op_reuse_tag : 0.000001s : 0.01% optimize.interleave_split_concat_branches : 0.000001s : 0.01% optimize.interleave_parallel_branches : 0.000001s : 0.01% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.01% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.01% optimize.control_data_broadcast_order : 0.000024s : 0.15% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.01% optimize.offloading_packed_experts : 0.000006s : 0.04% optimize.overlap_recompute_and_grad_model_parallel : 0.000006s : 0.04% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.01% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.01% optimize.overlap_recompute_comm : 0.000003s : 0.02% optimize.overlap_grad_ring_attention : 0.000006s : 0.04% optimize.overlap_grad_flash_sp : 0.000026s : 0.17% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.01% optimize.split_layernorm_comm : 0.000002s : 0.01% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000094s : 0.60% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000021s : 0.13% optimize.symbol_engine_optimizer.elim_not_effective : 0.000022s : 0.14% optimize.symbol_engine_optimizer.opt_reshape : 0.000011s : 0.07% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000022s : 0.14% optimize.symbol_engine_optimizer.renormalize : 0.000001s : 0.00% detach_backward : 0.000003s : 0.02% pipeline_parallel_scheduler : 0.000002s : 0.01% auto_monad_reorder : 0.000029s : 0.18% get_jit_bprop_graph : 0.000002s : 0.01% rewriter_after_jit_bprop_graph : 0.000007s : 0.05% opt_after_jit_grad : 0.000806s : 5.10% validate : 0.000067s : 0.42% Time group info: ------[substitution.] 0.000177 33 2.22% : 0.000004s : 2: substitution.elim_not_effective 5.24% : 0.000009s : 2: substitution.fold_const_symbol 4.73% : 0.000008s : 8: substitution.graph_param_transform 76.38% : 0.000135s : 1: substitution.inline 3.14% : 0.000006s : 4: substitution.j_node_and_user_rematch 3.07% : 0.000005s : 4: substitution.remove_not_recompute_node 5.21% : 0.000009s : 12: substitution.replace_old_param ------[type_inference.] 0.007521 2 87.42% : 0.006575s : 1: type_inference.infer 12.58% : 0.000946s : 1: type_inference.specialize ------[replace.] 0.000024 1 100.00% : 0.000024s : 1: replace.inline ------[match.] 0.000133 1 100.00% : 0.000133s : 1: match.inline ------[predicate.] 0.000241 1879 0.82% : 0.000002s : 17: predicate.accumulaten_eliminater 1.39% : 0.000003s : 8: predicate.ad_related_special_op_eliminate 0.68% : 0.000002s : 16: predicate.addn_check_dump 0.83% : 0.000002s : 17: predicate.addn_zero_filter 0.86% : 0.000002s : 17: predicate.adjust_all_reduce_mul_add 1.85% : 0.000004s : 33: predicate.arithmetic_simplify 0.80% : 0.000002s : 17: predicate.cast_eliminate 0.78% : 0.000002s : 16: predicate.check_bprop_eliminate 0.69% : 0.000002s : 16: predicate.compare_switch_simplify 0.26% : 0.000001s : 8: predicate.const_output_eliminate 0.67% : 0.000002s : 16: predicate.depend_value_elim 0.76% : 0.000002s : 17: predicate.dict_get_item_const_eliminator 0.86% : 0.000002s : 17: predicate.dict_get_item_eliminator 0.79% : 0.000002s : 17: predicate.dict_set_item_eliminator 1.55% : 0.000004s : 16: predicate.dumpgradient_eliminate 0.40% : 0.000001s : 8: predicate.elim_not_effective 0.46% : 0.000001s : 8: predicate.elim_shapecalc_of_broadcastargs 1.19% : 0.000003s : 25: predicate.environ_add_const_eliminate 1.01% : 0.000002s : 25: predicate.environ_get_add_eliminate 1.39% : 0.000003s : 25: predicate.environ_get_depend_swap 1.78% : 0.000004s : 41: predicate.environ_get_eliminate 1.00% : 0.000002s : 25: predicate.environ_get_set_eliminate 0.76% : 0.000002s : 18: predicate.exchange_switch_depend_value 1.45% : 0.000003s : 18: predicate.float_depend_g_call 0.70% : 0.000002s : 16: predicate.float_environ_get_switch 1.05% : 0.000003s : 24: predicate.float_tuple_getitem_switch 0.28% : 0.000001s : 8: predicate.fold_const_symbol 0.85% : 0.000002s : 16: predicate.get_grad_eliminate 0.39% : 0.000001s : 8: predicate.graph_param_transform 0.69% : 0.000002s : 16: predicate.incorporate_call 0.61% : 0.000001s : 16: predicate.incorporate_call_switch 5.49% : 0.000013s : 83: predicate.inline 1.38% : 0.000003s : 16: predicate.inline_without_move 0.56% : 0.000001s : 16: predicate.j_node_and_user_rematch 1.24% : 0.000003s : 16: predicate.less_batch_normalization 1.64% : 0.000004s : 33: predicate.list_to_tuple_eliminator_ 2.06% : 0.000005s : 50: predicate.load_eliminater 1.38% : 0.000003s : 8: predicate.loop_unroll_after_grad 1.26% : 0.000003s : 25: predicate.loop_unroll_before_grad 1.73% : 0.000004s : 33: predicate.make_slice_get_slice_eliminator 0.75% : 0.000002s : 16: predicate.merge_addn 0.84% : 0.000002s : 16: predicate.micro_step_allgather_replace 0.74% : 0.000002s : 16: predicate.mini_step_allgather_replace 0.63% : 0.000002s : 17: predicate.minmaximum_grad 1.44% : 0.000003s : 8: predicate.mutable_eliminate 0.50% : 0.000001s : 8: predicate.opt_reshape 0.49% : 0.000001s : 8: predicate.parallel_virtual_node 0.98% : 0.000002s : 18: predicate.partial_defer_inline 1.23% : 0.000003s : 25: predicate.partial_eliminate 0.83% : 0.000002s : 17: predicate.print_const_string_wrapper 0.76% : 0.000002s : 16: predicate.reduce_all_const_elim 1.04% : 0.000002s : 17: predicate.reduce_eliminate 2.16% : 0.000005s : 50: predicate.redundant_stop_gradient_eliminater 0.91% : 0.000002s : 16: predicate.remove_not_recompute_node 1.36% : 0.000003s : 33: predicate.replace_applicator 0.68% : 0.000002s : 16: predicate.replace_old_param 0.41% : 0.000001s : 8: predicate.reset_defer_inline 0.83% : 0.000002s : 17: predicate.reshape_eliminate 0.87% : 0.000002s : 16: predicate.row_tensor_add_zeros_like 0.58% : 0.000001s : 8: predicate.row_tensor_eliminate 1.14% : 0.000003s : 16: predicate.same_eliminate 0.79% : 0.000002s : 16: predicate.set_cell_output_no_recompute 1.15% : 0.000003s : 16: predicate.shard_identity_eliminate 0.91% : 0.000002s : 16: predicate.special_op_eliminate 0.78% : 0.000002s : 16: predicate.specialize_transform 1.20% : 0.000003s : 16: predicate.split_environ_get_set_with_tuple_value 1.09% : 0.000003s : 16: predicate.stack_unstack_eliminate 0.53% : 0.000001s : 8: predicate.switch_call_monad_eliminater 0.80% : 0.000002s : 18: predicate.switch_defer_inline 1.67% : 0.000004s : 34: predicate.switch_layer_defer_inline 3.85% : 0.000009s : 67: predicate.switch_simplify 0.79% : 0.000002s : 17: predicate.tile_eliminate 0.70% : 0.000002s : 17: predicate.transpose_eliminate 1.70% : 0.000004s : 33: predicate.tuple_list_convert_item_index_to_positive 1.73% : 0.000004s : 33: predicate.tuple_list_get_item_const_eliminator 1.38% : 0.000003s : 33: predicate.tuple_list_get_item_depend_reorder 3.30% : 0.000008s : 49: predicate.tuple_list_get_item_eliminator 1.50% : 0.000004s : 33: predicate.tuple_list_get_set_item_eliminator 2.72% : 0.000007s : 49: predicate.tuple_list_set_item_eliminator 1.64% : 0.000004s : 33: predicate.tuple_to_list_eliminator_ 2.02% : 0.000005s : 50: predicate.updatestate_pure_node_eliminater 3.09% : 0.000007s : 66: predicate.updatestate_useless_node_eliminater 0.45% : 0.000001s : 8: predicate.value_based_eliminate 1.03% : 0.000002s : 16: predicate.virtual_dataset_eliminate 0.88% : 0.000002s : 16: predicate.virtual_output_eliminate 0.39% : 0.000001s : 8: predicate.virtual_view_grad_eliminate 0.43% : 0.000001s : 8: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.001071 6 60.05% : 0.000643s : 3: func_graph_cloner_run.FuncGraphClonerGraph 39.95% : 0.000428s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.364173 192 0.00% : 0.000004s : 1: ForceFp32Comm 46.29% : 0.168575s : 1: add_attr 46.28% : 0.168555s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.02% : 0.000083s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.03% : 0.000106s : 1: auto_monad 0.01% : 0.000034s : 1: auto_monad_reorder 0.00% : 0.000005s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.29% : 0.001050s : 1: bootstrap 0.01% : 0.000047s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000030s : 1: control_data_broadcast_order 0.00% : 0.000015s : 1: convert_after_rewriter 0.01% : 0.000052s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000012s : 1: environ_conv 0.01% : 0.000027s : 1: event_method 0.00% : 0.000006s : 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% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000008s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000008s : 1: label_fine_grained_interleaved_index 0.00% : 0.000010s : 1: label_micro_interleaved_index 0.17% : 0.000631s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.23% : 0.000828s : 1: mutable_eliminate 0.00% : 0.000010s : 1: offloading_packed_experts 0.01% : 0.000028s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000030s : 1: opt.transform.mutable_eliminate 0.40% : 0.001451s : 78: opt.transform.opt_a 0.01% : 0.000051s : 1: opt.transform.opt_after_cconv 0.01% : 0.000052s : 1: opt.transform.opt_after_jit_grad 0.05% : 0.000196s : 28: opt.transform.opt_b 0.03% : 0.000094s : 2: opt.transform.opt_trans_graph 0.02% : 0.000070s : 4: opt.transform.symbol_engine_opt 1.01% : 0.003679s : 1: opt_a 0.05% : 0.000195s : 1: opt_after_cconv 0.23% : 0.000827s : 1: opt_after_jit_grad 0.11% : 0.000382s : 1: opt_b 1.96% : 0.007142s : 1: optimize 0.01% : 0.000030s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000012s : 1: order_py_execute_after_rewriter 0.01% : 0.000031s : 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% : 0.000006s : 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% : 0.000007s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000046s : 1: pre_auto_parallel 0.01% : 0.000045s : 1: py_interpret_to_execute 0.01% : 0.000042s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.07% : 0.000246s : 1: remove_dup_value 0.14% : 0.000503s : 1: renormalize.infer 0.13% : 0.000477s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.02% : 0.000075s : 1: rewriter_after_opt_a 0.02% : 0.000087s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000012s : 1: swap_dp_allreduce_reducescatter 0.06% : 0.000228s : 1: symbol_engine_optimizer 0.04% : 0.000140s : 1: tuple_transform 2.11% : 0.007701s : 1: type_inference .[WARNING] PRE_ACT(170442,ffffbeca8f30,python3.9):2026-01-29-17:39:33.649.151 [mindspore/ccsrc/runtime/memory/mem_pool/abstract_dynamic_mem_pool.cc:1109] FreeIdleMemsByEagerFree] Eager free count : 2, free memory : 2725209088, real free : 2726297600, not free : 0. [hook] pytest_runtest_teardown:test_paged_attention_trapezoidal[0-MASK_DEFAULT-float32] tests/st/infer/ops/test_internal_ops/test_paged_attention.py::test_paged_attention_trapezoidal[0-MASK_DEFAULT-float32],max_mem:2696.0M [WARNING] ME(170442:281473882689328,MainProcess):2026-01-29-17:39:34.894.870 [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 = 0.160503, [21] [bootstrap]: 0.00063864 [type_inference]: 0.00543434 [event_method]: 1.354e-05 [auto_monad]: 5.762e-05 [graph_reusing]: 5.60001e-06 [inline]: 1.99e-06 [add_attr]: 0.0044867, [1] [add_attr_with_inline]: 0.0044739, [1] [Cycle 1]: 5.443e-05, [2] [tag_attr]: 1.901e-05 [meta_addattr_fg_expand]: 3.86999e-06 [parallel-infer-symbol]: 3.53e-06 [pre_auto_parallel]: 3.44e-05 [insert-virtual-dataset]: 2.49999e-06 [parallel-infer-symbol-second]: 6.89994e-07 [dataset_repeat_opt]: 1.92001e-06 [pipeline_split]: 1.48002e-06 [optimize]: 0.148998, [53] [py_interpret_to_execute]: 2.987e-05 [rewriter_before_opt_a]: 6.987e-05 [opt_a]: 0.016157, [2] [Cycle 1]: 0.0150746, [45] [expand_dump_flag]: 2.82002e-06 [switch_simplify]: 2.777e-05 [loop_unroll]: 1.737e-05 [a_1]: 0.00047146 [with_stream_mark]: 1.929e-05 [recompute_prepare]: 1.495e-05 [updatestate_depend_eliminate]: 6.27001e-06 [updatestate_assign_eliminate]: 5.51002e-06 [updatestate_loads_eliminate]: 4.89e-06 [parameter_eliminate]: 1.66e-06 [a_2]: 0.00016337 [accelerated_algorithm]: 1.195e-05 [shard]: 2.36e-06 [meta_shard_fg_expand]: 2.29001e-06 [shard_inline]: 1.01e-05 [merge_send_recv]: 1.089e-05 [auto_parallel]: 8.94998e-06 [parallel]: 2.964e-05 [flash_sp]: 9.96998e-06 [merge_comm]: 5.39998e-06 [allreduce_fusion]: 5.35999e-06 [matmul_add_comm_reduction]: 1.233e-05 [allreduce_slice_to_reducescatter]: 6.39993e-07 [virtual_shard_identity]: 1.284e-05 [virtual_dataset]: 1.188e-05 [get_grad_eliminate_]: 1.154e-05 [virtual_output]: 1.161e-05 [merge_forward]: 6.04999e-06 [cell_reuse_recompute_pass]: 1.90001e-06 [offload_activation]: 1.367e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.935e-05 [merge_recompute_call_nodes]: 5.04e-06 [before_grad]: 8.393e-05 [set_forward_comm_id_for_comm_node_pass]: 2.903e-05 [meta_fg_expand]: 1.083e-05 [flash_sp_send_recv_attached]: 1.158e-05 [receive_attached]: 3.59002e-06 [after_resolve]: 3.189e-05 [a_after_grad]: 2.065e-05 [renormalize]: 0.00138199 [add_forward_monad_depend]: 1.428e-05 [auto_monad_grad]: 2.84001e-06 [auto_monad_eliminator]: 2.185e-05 [cse]: 7.867e-05 [a_3]: 9.058e-05 [Cycle 2]: 0.00106911, [45] [expand_dump_flag]: 2.01998e-06 [switch_simplify]: 1.287e-05 [loop_unroll]: 1.655e-05 [a_1]: 0.00026287 [with_stream_mark]: 2.163e-05 [recompute_prepare]: 1.079e-05 [updatestate_depend_eliminate]: 6.46e-06 [updatestate_assign_eliminate]: 5.10999e-06 [updatestate_loads_eliminate]: 5.57001e-06 [parameter_eliminate]: 2.66e-06 [a_2]: 0.00012608 [accelerated_algorithm]: 9.92999e-06 [shard]: 2.82002e-06 [meta_shard_fg_expand]: 2.66e-06 [shard_inline]: 9.59e-06 [merge_send_recv]: 9.12999e-06 [auto_parallel]: 1.235e-05 [parallel]: 1.637e-05 [flash_sp]: 4.80001e-06 [merge_comm]: 5.15999e-06 [allreduce_fusion]: 4.74998e-06 [matmul_add_comm_reduction]: 1.16e-05 [allreduce_slice_to_reducescatter]: 1.05001e-06 [virtual_shard_identity]: 1.154e-05 [virtual_dataset]: 9.56e-06 [get_grad_eliminate_]: 9.92001e-06 [virtual_output]: 1.336e-05 [merge_forward]: 6.76e-06 [cell_reuse_recompute_pass]: 3.08e-06 [offload_activation]: 1.235e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.372e-05 [merge_recompute_call_nodes]: 1.25999e-06 [before_grad]: 1.304e-05 [set_forward_comm_id_for_comm_node_pass]: 5.51002e-06 [meta_fg_expand]: 3.59002e-06 [flash_sp_send_recv_attached]: 9.29984e-07 [receive_attached]: 1.60999e-06 [after_resolve]: 1.899e-05 [a_after_grad]: 1.486e-05 [renormalize]: 1.80007e-07 [add_forward_monad_depend]: 1.25001e-06 [auto_monad_grad]: 1.55999e-06 [auto_monad_eliminator]: 1.037e-05 [cse]: 4.232e-05 [a_3]: 6.213e-05 [py_interpret_to_execute_after_opt_a]: 2.243e-05 [slice_cell_reuse_recomputed_activation]: 2.02999e-06 [rewriter_after_opt_a]: 0.129559 [convert_after_rewriter]: 4.042e-05 [order_py_execute_after_rewriter]: 7.83001e-06 [mutable_eliminate]: 0.00090893 [opt_b]: 0.00045269, [1] [Cycle 1]: 0.00043872, [7] [b_1]: 0.00025034 [b_2]: 1.723e-05 [updatestate_depend_eliminate]: 1.034e-05 [updatestate_assign_eliminate]: 5.87001e-06 [updatestate_loads_eliminate]: 5.72999e-06 [renormalize]: 1.12e-06 [cse]: 8.641e-05 [optimize_parallel_all_gather_comm]: 2.73e-05 [overlap_param_gather]: 2.49001e-06 [cconv]: 4.109e-05 [loop_unroll]: 0.00055719 [opt_after_cconv]: 0.00018343, [1] [Cycle 1]: 0.0001753, [7] [c_1]: 5.371e-05 [parameter_eliminate]: 6.46e-06 [updatestate_depend_eliminate]: 8.43001e-06 [updatestate_assign_eliminate]: 4.82998e-06 [updatestate_loads_eliminate]: 4.43999e-06 [cse]: 5.08e-05 [renormalize]: 6.59988e-07 [remove_dup_value]: 7.359e-05 [tuple_transform]: 0.00012286, [1] [Cycle 1]: 0.00011763, [4] [d_1]: 8.48e-05 [none_parameter_eliminate]: 2.46e-06 [renormalize]: 1.79978e-07 [switch_simplify]: 1.104e-05 [partial_unused_args_eliminate]: 1.68002e-06 [add_recomputation]: 6.882e-05 [cse_after_recomputation]: 3.668e-05, [1] [Cycle 1]: 3.177e-05, [1] [cse]: 2.524e-05 [environ_conv]: 7.56999e-06 [swap_dp_allreduce_reducescatter]: 8.01001e-06 [bias_add_comm_swap]: 3.17002e-06 [label_micro_interleaved_index]: 4.92e-06 [label_fine_grained_interleaved_index]: 2.69001e-06 [merge_cast_opt]: 1.54e-06 [slice_recompute_activation]: 2.37999e-06 [micro_interleaved_order_control]: 2.53003e-06 [assign_add_opt]: 1.40001e-06 [ForceFp32Comm]: 8.40024e-07 [remove_cast_before_assign_add]: 1.34e-06 [full_micro_interleaved_order_control]: 2.23998e-06 [reorder_send_recv_between_fp_bp]: 2.81e-06 [comm_op_add_attrs]: 9.89996e-07 [add_comm_op_reuse_tag]: 9.99979e-07 [interleave_split_concat_branches]: 1.24003e-06 [interleave_parallel_branches]: 1.01002e-06 [overlap_opt_shard_in_pipeline]: 1.29e-06 [overlap_opt_shard_grad_in_pipeline]: 2.46e-06 [control_data_broadcast_order]: 1.821e-05 [grouped_pairwise_exchange_alltoall]: 1.80001e-06 [offloading_packed_experts]: 5.37001e-06 [overlap_recompute_and_grad_model_parallel]: 6.21e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.12999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.54e-06 [overlap_recompute_comm]: 2.26e-06 [overlap_grad_ring_attention]: 5.19e-06 [overlap_grad_flash_sp]: 2.347e-05 [begin_end_overlap_inline]: 5.3001e-07 [split_matmul_comm_elemetwise]: 2.14999e-06 [split_layernorm_comm]: 1.92999e-06 [handle_group_info]: 8.70001e-07 [symbol_engine_optimizer]: 0.00020141, [1] [Cycle 1]: 0.00019623, [6] [build]: 9.489e-05 [elim_shapecalc]: 1.745e-05 [elim_not_effective]: 2.026e-05 [opt_reshape]: 1.068e-05 [fold_const_symbol]: 1.947e-05 [renormalize]: 1.8999e-07 [detach_backward]: 2.32001e-06 [pipeline_parallel_scheduler]: 1.55999e-06 [auto_monad_reorder]: 2.448e-05 [get_jit_bprop_graph]: 2.07999e-06 [rewriter_after_jit_bprop_graph]: 4.87e-06 [opt_after_jit_grad]: 0.00056665 [validate]: 5.549e-05 Sums bootstrap : 0.000639s : 0.45% type_inference : 0.005434s : 3.81% event_method : 0.000014s : 0.01% auto_monad : 0.000058s : 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 : 0.000004s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000034s : 0.02% insert-virtual-dataset : 0.000002s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000001s : 0.00% optimize.py_interpret_to_execute : 0.000030s : 0.02% optimize.rewriter_before_opt_a : 0.000070s : 0.05% optimize.opt_a.expand_dump_flag : 0.000005s : 0.00% optimize.opt_a.switch_simplify : 0.000041s : 0.03% optimize.opt_a.loop_unroll : 0.000034s : 0.02% optimize.opt_a.a_1 : 0.000734s : 0.51% optimize.opt_a.with_stream_mark : 0.000041s : 0.03% optimize.opt_a.recompute_prepare : 0.000026s : 0.02% 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 : 0.000010s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.00% optimize.opt_a.a_2 : 0.000289s : 0.20% optimize.opt_a.accelerated_algorithm : 0.000022s : 0.02% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000005s : 0.00% optimize.opt_a.shard_inline : 0.000020s : 0.01% optimize.opt_a.merge_send_recv : 0.000020s : 0.01% optimize.opt_a.auto_parallel : 0.000021s : 0.01% optimize.opt_a.parallel : 0.000046s : 0.03% optimize.opt_a.flash_sp : 0.000015s : 0.01% optimize.opt_a.merge_comm : 0.000011s : 0.01% optimize.opt_a.allreduce_fusion : 0.000010s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000024s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000024s : 0.02% optimize.opt_a.virtual_dataset : 0.000021s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000021s : 0.02% optimize.opt_a.virtual_output : 0.000025s : 0.02% optimize.opt_a.merge_forward : 0.000013s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000026s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000043s : 0.03% optimize.opt_a.merge_recompute_call_nodes : 0.000006s : 0.00% optimize.opt_a.before_grad : 0.000097s : 0.07% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000035s : 0.02% optimize.opt_a.meta_fg_expand : 0.000014s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000013s : 0.01% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000051s : 0.04% optimize.opt_a.a_after_grad : 0.000036s : 0.02% optimize.opt_a.renormalize : 0.001382s : 0.97% optimize.opt_a.add_forward_monad_depend : 0.000016s : 0.01% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000032s : 0.02% optimize.opt_a.cse : 0.000121s : 0.08% optimize.opt_a.a_3 : 0.000153s : 0.11% optimize.py_interpret_to_execute_after_opt_a : 0.000022s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.129559s : 90.76% optimize.convert_after_rewriter : 0.000040s : 0.03% optimize.order_py_execute_after_rewriter : 0.000008s : 0.01% optimize.mutable_eliminate : 0.000909s : 0.64% optimize.opt_b.b_1 : 0.000250s : 0.18% optimize.opt_b.b_2 : 0.000017s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000010s : 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 : 0.000086s : 0.06% optimize.optimize_parallel_all_gather_comm : 0.000027s : 0.02% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000041s : 0.03% optimize.loop_unroll : 0.000557s : 0.39% optimize.opt_after_cconv.c_1 : 0.000054s : 0.04% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000008s : 0.01% 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 : 0.000051s : 0.04% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000074s : 0.05% optimize.tuple_transform.d_1 : 0.000085s : 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 : 0.000069s : 0.05% optimize.cse_after_recomputation.cse : 0.000025s : 0.02% optimize.environ_conv : 0.000008s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000008s : 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 : 0.000002s : 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 : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000018s : 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 : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000005s : 0.00% optimize.overlap_grad_flash_sp : 0.000023s : 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 : 0.000095s : 0.07% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000017s : 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 : 0.000019s : 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 : 0.000024s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000005s : 0.00% opt_after_jit_grad : 0.000567s : 0.40% validate : 0.000055s : 0.04% Time group info: ------[substitution.] 0.000146 33 2.81% : 0.000004s : 2: substitution.elim_not_effective 5.22% : 0.000008s : 2: substitution.fold_const_symbol 5.62% : 0.000008s : 8: substitution.graph_param_transform 64.23% : 0.000094s : 1: substitution.inline 6.94% : 0.000010s : 4: substitution.j_node_and_user_rematch 6.29% : 0.000009s : 4: substitution.remove_not_recompute_node 8.89% : 0.000013s : 12: substitution.replace_old_param ------[type_inference.] 0.005349 2 90.73% : 0.004853s : 1: type_inference.infer 9.27% : 0.000496s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000093 1 100.00% : 0.000093s : 1: match.inline ------[predicate.] 0.000234 1879 0.82% : 0.000002s : 17: predicate.accumulaten_eliminater 0.94% : 0.000002s : 8: predicate.ad_related_special_op_eliminate 0.70% : 0.000002s : 16: predicate.addn_check_dump 0.85% : 0.000002s : 17: predicate.addn_zero_filter 0.78% : 0.000002s : 17: predicate.adjust_all_reduce_mul_add 1.87% : 0.000004s : 33: predicate.arithmetic_simplify 0.80% : 0.000002s : 17: predicate.cast_eliminate 0.80% : 0.000002s : 16: predicate.check_bprop_eliminate 0.67% : 0.000002s : 16: predicate.compare_switch_simplify 0.27% : 0.000001s : 8: predicate.const_output_eliminate 0.79% : 0.000002s : 16: predicate.depend_value_elim 0.77% : 0.000002s : 17: predicate.dict_get_item_const_eliminator 0.92% : 0.000002s : 17: predicate.dict_get_item_eliminator 0.99% : 0.000002s : 17: predicate.dict_set_item_eliminator 1.26% : 0.000003s : 16: predicate.dumpgradient_eliminate 0.41% : 0.000001s : 8: predicate.elim_not_effective 0.48% : 0.000001s : 8: predicate.elim_shapecalc_of_broadcastargs 1.17% : 0.000003s : 25: predicate.environ_add_const_eliminate 1.06% : 0.000002s : 25: predicate.environ_get_add_eliminate 1.06% : 0.000002s : 25: predicate.environ_get_depend_swap 1.87% : 0.000004s : 41: predicate.environ_get_eliminate 1.06% : 0.000002s : 25: predicate.environ_get_set_eliminate 0.90% : 0.000002s : 18: predicate.exchange_switch_depend_value 1.44% : 0.000003s : 18: predicate.float_depend_g_call 0.73% : 0.000002s : 16: predicate.float_environ_get_switch 1.12% : 0.000003s : 24: predicate.float_tuple_getitem_switch 0.28% : 0.000001s : 8: predicate.fold_const_symbol 1.07% : 0.000003s : 16: predicate.get_grad_eliminate 0.36% : 0.000001s : 8: predicate.graph_param_transform 0.73% : 0.000002s : 16: predicate.incorporate_call 0.64% : 0.000002s : 16: predicate.incorporate_call_switch 5.37% : 0.000013s : 83: predicate.inline 1.18% : 0.000003s : 16: predicate.inline_without_move 1.28% : 0.000003s : 16: predicate.j_node_and_user_rematch 0.93% : 0.000002s : 16: predicate.less_batch_normalization 1.90% : 0.000004s : 33: predicate.list_to_tuple_eliminator_ 2.19% : 0.000005s : 50: predicate.load_eliminater 1.23% : 0.000003s : 8: predicate.loop_unroll_after_grad 1.24% : 0.000003s : 25: predicate.loop_unroll_before_grad 1.81% : 0.000004s : 33: predicate.make_slice_get_slice_eliminator 0.70% : 0.000002s : 16: predicate.merge_addn 0.85% : 0.000002s : 16: predicate.micro_step_allgather_replace 0.80% : 0.000002s : 16: predicate.mini_step_allgather_replace 0.68% : 0.000002s : 17: predicate.minmaximum_grad 1.11% : 0.000003s : 8: predicate.mutable_eliminate 0.58% : 0.000001s : 8: predicate.opt_reshape 0.45% : 0.000001s : 8: predicate.parallel_virtual_node 1.02% : 0.000002s : 18: predicate.partial_defer_inline 1.15% : 0.000003s : 25: predicate.partial_eliminate 0.78% : 0.000002s : 17: predicate.print_const_string_wrapper 0.84% : 0.000002s : 16: predicate.reduce_all_const_elim 1.13% : 0.000003s : 17: predicate.reduce_eliminate 2.24% : 0.000005s : 50: predicate.redundant_stop_gradient_eliminater 0.73% : 0.000002s : 16: predicate.remove_not_recompute_node 1.43% : 0.000003s : 33: predicate.replace_applicator 0.71% : 0.000002s : 16: predicate.replace_old_param 0.62% : 0.000001s : 8: predicate.reset_defer_inline 0.82% : 0.000002s : 17: predicate.reshape_eliminate 0.81% : 0.000002s : 16: predicate.row_tensor_add_zeros_like 0.47% : 0.000001s : 8: predicate.row_tensor_eliminate 1.08% : 0.000003s : 16: predicate.same_eliminate 0.62% : 0.000001s : 16: predicate.set_cell_output_no_recompute 1.04% : 0.000002s : 16: predicate.shard_identity_eliminate 0.87% : 0.000002s : 16: predicate.special_op_eliminate 0.79% : 0.000002s : 16: predicate.specialize_transform 1.19% : 0.000003s : 16: predicate.split_environ_get_set_with_tuple_value 1.13% : 0.000003s : 16: predicate.stack_unstack_eliminate 0.41% : 0.000001s : 8: predicate.switch_call_monad_eliminater 0.81% : 0.000002s : 18: predicate.switch_defer_inline 1.58% : 0.000004s : 34: predicate.switch_layer_defer_inline 3.38% : 0.000008s : 67: predicate.switch_simplify 0.79% : 0.000002s : 17: predicate.tile_eliminate 0.76% : 0.000002s : 17: predicate.transpose_eliminate 1.89% : 0.000004s : 33: predicate.tuple_list_convert_item_index_to_positive 1.67% : 0.000004s : 33: predicate.tuple_list_get_item_const_eliminator 1.65% : 0.000004s : 33: predicate.tuple_list_get_item_depend_reorder 3.20% : 0.000007s : 49: predicate.tuple_list_get_item_eliminator 1.69% : 0.000004s : 33: predicate.tuple_list_get_set_item_eliminator 2.56% : 0.000006s : 49: predicate.tuple_list_set_item_eliminator 1.89% : 0.000004s : 33: predicate.tuple_to_list_eliminator_ 2.05% : 0.000005s : 50: predicate.updatestate_pure_node_eliminater 2.97% : 0.000007s : 66: predicate.updatestate_useless_node_eliminater 0.45% : 0.000001s : 8: predicate.value_based_eliminate 0.98% : 0.000002s : 16: predicate.virtual_dataset_eliminate 0.84% : 0.000002s : 16: predicate.virtual_output_eliminate 0.36% : 0.000001s : 8: predicate.virtual_view_grad_eliminate 0.73% : 0.000002s : 8: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000707 6 40.49% : 0.000286s : 3: func_graph_cloner_run.FuncGraphClonerGraph 59.51% : 0.000421s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.317232 192 0.00% : 0.000003s : 1: ForceFp32Comm 1.42% : 0.004493s : 1: add_attr 1.41% : 0.004478s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.02% : 0.000073s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.02% : 0.000063s : 1: auto_monad 0.01% : 0.000028s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.21% : 0.000673s : 1: bootstrap 0.02% : 0.000049s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000021s : 1: control_data_broadcast_order 0.02% : 0.000049s : 1: convert_after_rewriter 0.01% : 0.000040s : 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% : 0.000019s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000010s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000003s : 1: handle_group_info 0.00% : 0.000005s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.18% : 0.000567s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.29% : 0.000922s : 1: mutable_eliminate 0.00% : 0.000009s : 1: offloading_packed_experts 0.01% : 0.000023s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000034s : 1: opt.transform.mutable_eliminate 0.48% : 0.001513s : 78: opt.transform.opt_a 0.02% : 0.000052s : 1: opt.transform.opt_after_cconv 0.01% : 0.000041s : 1: opt.transform.opt_after_jit_grad 0.07% : 0.000229s : 28: opt.transform.opt_b 0.03% : 0.000094s : 2: opt.transform.opt_trans_graph 0.02% : 0.000064s : 4: opt.transform.symbol_engine_opt 5.09% : 0.016161s : 1: opt_a 0.06% : 0.000188s : 1: opt_after_cconv 0.18% : 0.000579s : 1: opt_after_jit_grad 0.14% : 0.000457s : 1: opt_b 46.97% : 0.149004s : 1: optimize 0.01% : 0.000032s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000012s : 1: order_py_execute_after_rewriter 0.01% : 0.000027s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000014s : 1: overlap_grad_ring_attention 0.00% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000009s : 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% : 0.000007s : 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% : 0.000039s : 1: pre_auto_parallel 0.01% : 0.000034s : 1: py_interpret_to_execute 0.01% : 0.000026s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.02% : 0.000079s : 1: remove_dup_value 0.18% : 0.000586s : 1: renormalize.infer 0.25% : 0.000779s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000008s : 1: rewriter_after_jit_bprop_graph 40.85% : 0.129594s : 1: rewriter_after_opt_a 0.02% : 0.000074s : 1: rewriter_before_opt_a 0.00% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000004s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000011s : 1: swap_dp_allreduce_reducescatter 0.06% : 0.000204s : 1: symbol_engine_optimizer 0.04% : 0.000126s : 1: tuple_transform 1.72% : 0.005453s : 1: type_inference .[WARNING] PRE_ACT(170442,ffffbeca8f30,python3.9):2026-01-29-17:40:21.330.586 [mindspore/ccsrc/runtime/memory/mem_pool/abstract_dynamic_mem_pool.cc:1109] FreeIdleMemsByEagerFree] Eager free count : 3, free memory : 1383032320, real free : 1384120320, not free : 0. [hook] pytest_runtest_teardown:test_paged_attention_trapezoidal[0-TRAPEZOIDAL-float16] tests/st/infer/ops/test_internal_ops/test_paged_attention.py::test_paged_attention_trapezoidal[0-TRAPEZOIDAL-float16],max_mem:2696.0M [WARNING] ME(170442:281473882689328,MainProcess):2026-01-29-17:40:22.236.650 [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 = 0.742891, [21] [bootstrap]: 0.00265626 [type_inference]: 0.358561 [event_method]: 2.313e-05 [auto_monad]: 0.00044092 [graph_reusing]: 5.14998e-06 [inline]: 3.5e-06 [add_attr]: 0.36742, [1] [add_attr_with_inline]: 0.367267, [1] [Cycle 1]: 7.585e-05, [2] [tag_attr]: 2.302e-05 [meta_addattr_fg_expand]: 4.28999e-06 [parallel-infer-symbol]: 3.56001e-06 [pre_auto_parallel]: 3.439e-05 [insert-virtual-dataset]: 2.36e-06 [parallel-infer-symbol-second]: 6.69999e-07 [dataset_repeat_opt]: 1.84e-06 [pipeline_split]: 1.84998e-06 [optimize]: 0.0128757, [53] [py_interpret_to_execute]: 0.00010587 [rewriter_before_opt_a]: 7.819e-05 [opt_a]: 0.00913397, [2] [Cycle 1]: 0.00727828, [45] [expand_dump_flag]: 3.63e-06 [switch_simplify]: 3.038e-05 [loop_unroll]: 1.788e-05 [a_1]: 0.00060489 [with_stream_mark]: 2.182e-05 [recompute_prepare]: 1.405e-05 [updatestate_depend_eliminate]: 6.11e-06 [updatestate_assign_eliminate]: 6.04001e-06 [updatestate_loads_eliminate]: 4.82e-06 [parameter_eliminate]: 1.74998e-06 [a_2]: 0.00014337 [accelerated_algorithm]: 1.179e-05 [shard]: 2.12999e-06 [meta_shard_fg_expand]: 2.31e-06 [shard_inline]: 1.017e-05 [merge_send_recv]: 1.016e-05 [auto_parallel]: 8.23999e-06 [parallel]: 0.00026347 [flash_sp]: 1.054e-05 [merge_comm]: 5.39e-06 [allreduce_fusion]: 5.25999e-06 [matmul_add_comm_reduction]: 1.153e-05 [allreduce_slice_to_reducescatter]: 6.50005e-07 [virtual_shard_identity]: 1.215e-05 [virtual_dataset]: 1.116e-05 [get_grad_eliminate_]: 1.098e-05 [virtual_output]: 1.135e-05 [merge_forward]: 5.79999e-06 [cell_reuse_recompute_pass]: 1.64e-06 [offload_activation]: 1.345e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.819e-05 [merge_recompute_call_nodes]: 1.74e-06 [before_grad]: 1.507e-05 [set_forward_comm_id_for_comm_node_pass]: 5.66e-06 [meta_fg_expand]: 3.64002e-06 [flash_sp_send_recv_attached]: 2.51e-06 [receive_attached]: 2.50002e-06 [after_resolve]: 2.093e-05 [a_after_grad]: 1.824e-05 [renormalize]: 0.0053535 [add_forward_monad_depend]: 0.00010399 [auto_monad_grad]: 2.67001e-06 [auto_monad_eliminator]: 2.01e-05 [cse]: 7.719e-05 [a_3]: 8.097e-05 [Cycle 2]: 0.00184137, [45] [expand_dump_flag]: 2.14e-06 [switch_simplify]: 1.199e-05 [loop_unroll]: 9.79e-06 [a_1]: 0.00023178 [with_stream_mark]: 1.806e-05 [recompute_prepare]: 9.78002e-06 [updatestate_depend_eliminate]: 5.87999e-06 [updatestate_assign_eliminate]: 5.17e-06 [updatestate_loads_eliminate]: 4.69998e-06 [parameter_eliminate]: 1.54998e-06 [a_2]: 0.00018636 [accelerated_algorithm]: 1.1e-05 [shard]: 1.92999e-06 [meta_shard_fg_expand]: 2.37999e-06 [shard_inline]: 9.77999e-06 [merge_send_recv]: 1.053e-05 [auto_parallel]: 6.75e-05 [parallel]: 8.20999e-06 [flash_sp]: 3.97e-06 [merge_comm]: 5.46e-06 [allreduce_fusion]: 5.04e-06 [matmul_add_comm_reduction]: 1.042e-05 [allreduce_slice_to_reducescatter]: 9.60019e-07 [virtual_shard_identity]: 1.065e-05 [virtual_dataset]: 9.51998e-06 [get_grad_eliminate_]: 9.32001e-06 [virtual_output]: 9.14998e-06 [merge_forward]: 5.34e-06 [cell_reuse_recompute_pass]: 2.72001e-06 [offload_activation]: 1.138e-05 [cell_reuse_handle_not_recompute_node_pass]: 8.163e-05 [merge_recompute_call_nodes]: 1.20999e-06 [before_grad]: 1.447e-05 [set_forward_comm_id_for_comm_node_pass]: 5.69e-06 [meta_fg_expand]: 3.24001e-06 [flash_sp_send_recv_attached]: 1.20999e-06 [receive_attached]: 1.89e-06 [after_resolve]: 2.113e-05 [a_after_grad]: 1.393e-05 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 1.79e-06 [auto_monad_grad]: 1.13001e-06 [auto_monad_eliminator]: 9.67999e-06 [cse]: 4.878e-05 [a_3]: 0.00022509 [py_interpret_to_execute_after_opt_a]: 1.747e-05 [slice_cell_reuse_recomputed_activation]: 2.31e-06 [rewriter_after_opt_a]: 5.764e-05 [convert_after_rewriter]: 9.71e-06 [order_py_execute_after_rewriter]: 6.91001e-06 [mutable_eliminate]: 0.0008155 [opt_b]: 0.00046016, [1] [Cycle 1]: 0.00045148, [7] [b_1]: 0.00023532 [b_2]: 1.281e-05 [updatestate_depend_eliminate]: 8.69e-06 [updatestate_assign_eliminate]: 4.45999e-06 [updatestate_loads_eliminate]: 4.25e-06 [renormalize]: 6.19999e-07 [cse]: 6.306e-05 [optimize_parallel_all_gather_comm]: 2.073e-05 [overlap_param_gather]: 2.44001e-06 [cconv]: 2.759e-05 [loop_unroll]: 0.00068703 [opt_after_cconv]: 0.00016793, [1] [Cycle 1]: 0.0001613, [7] [c_1]: 5.457e-05 [parameter_eliminate]: 3.01001e-06 [updatestate_depend_eliminate]: 7.86001e-06 [updatestate_assign_eliminate]: 4.52003e-06 [updatestate_loads_eliminate]: 4.20999e-06 [cse]: 4.777e-05 [renormalize]: 3.30008e-07 [remove_dup_value]: 6.862e-05 [tuple_transform]: 0.00011708, [1] [Cycle 1]: 0.00011206, [4] [d_1]: 7.909e-05 [none_parameter_eliminate]: 2.04999e-06 [renormalize]: 2.50002e-07 [switch_simplify]: 1.073e-05 [partial_unused_args_eliminate]: 1.77001e-06 [add_recomputation]: 6.429e-05 [cse_after_recomputation]: 3.607e-05, [1] [Cycle 1]: 3.139e-05, [1] [cse]: 2.491e-05 [environ_conv]: 7.13e-06 [swap_dp_allreduce_reducescatter]: 7.63001e-06 [bias_add_comm_swap]: 2.58e-06 [label_micro_interleaved_index]: 4.20999e-06 [label_fine_grained_interleaved_index]: 2.68003e-06 [merge_cast_opt]: 1.32e-06 [slice_recompute_activation]: 1.99e-06 [micro_interleaved_order_control]: 2.64001e-06 [assign_add_opt]: 1.24e-06 [ForceFp32Comm]: 8.10018e-07 [remove_cast_before_assign_add]: 1.10001e-06 [full_micro_interleaved_order_control]: 2.55002e-06 [reorder_send_recv_between_fp_bp]: 2.69001e-06 [comm_op_add_attrs]: 1.02e-06 [add_comm_op_reuse_tag]: 1.38002e-06 [interleave_split_concat_branches]: 1.35001e-06 [interleave_parallel_branches]: 1.12e-06 [overlap_opt_shard_in_pipeline]: 1.28002e-06 [overlap_opt_shard_grad_in_pipeline]: 1.79e-06 [control_data_broadcast_order]: 1.719e-05 [grouped_pairwise_exchange_alltoall]: 1.86e-06 [offloading_packed_experts]: 5.42999e-06 [overlap_recompute_and_grad_model_parallel]: 6.54001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.55999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.50999e-06 [overlap_recompute_comm]: 2.83e-06 [overlap_grad_ring_attention]: 5.09e-06 [overlap_grad_flash_sp]: 2.229e-05 [begin_end_overlap_inline]: 6.69999e-07 [split_matmul_comm_elemetwise]: 2.38998e-06 [split_layernorm_comm]: 1.84998e-06 [handle_group_info]: 1.05999e-06 [symbol_engine_optimizer]: 0.00039485, [1] [Cycle 1]: 0.00038968, [6] [build]: 0.00016483 [elim_shapecalc]: 1.552e-05 [elim_not_effective]: 1.874e-05 [opt_reshape]: 1.068e-05 [fold_const_symbol]: 0.00014141 [renormalize]: 2.09984e-07 [detach_backward]: 2.41e-06 [pipeline_parallel_scheduler]: 1.60999e-06 [auto_monad_reorder]: 2.595e-05 [get_jit_bprop_graph]: 1.65001e-06 [rewriter_after_jit_bprop_graph]: 4.50999e-06 [opt_after_jit_grad]: 0.00050677 [validate]: 5.058e-05 Sums bootstrap : 0.002656s : 0.71% type_inference : 0.358561s : 96.01% event_method : 0.000023s : 0.01% auto_monad : 0.000441s : 0.12% graph_reusing : 0.000005s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000023s : 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 : 0.000034s : 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 : 0.000106s : 0.03% optimize.rewriter_before_opt_a : 0.000078s : 0.02% optimize.opt_a.expand_dump_flag : 0.000006s : 0.00% optimize.opt_a.switch_simplify : 0.000042s : 0.01% optimize.opt_a.loop_unroll : 0.000028s : 0.01% optimize.opt_a.a_1 : 0.000837s : 0.22% optimize.opt_a.with_stream_mark : 0.000040s : 0.01% optimize.opt_a.recompute_prepare : 0.000024s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000012s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000011s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000010s : 0.00% optimize.opt_a.parameter_eliminate : 0.000003s : 0.00% optimize.opt_a.a_2 : 0.000330s : 0.09% optimize.opt_a.accelerated_algorithm : 0.000023s : 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 : 0.000020s : 0.01% optimize.opt_a.merge_send_recv : 0.000021s : 0.01% optimize.opt_a.auto_parallel : 0.000076s : 0.02% optimize.opt_a.parallel : 0.000272s : 0.07% optimize.opt_a.flash_sp : 0.000015s : 0.00% optimize.opt_a.merge_comm : 0.000011s : 0.00% optimize.opt_a.allreduce_fusion : 0.000010s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000022s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000023s : 0.01% optimize.opt_a.virtual_dataset : 0.000021s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000020s : 0.01% optimize.opt_a.virtual_output : 0.000020s : 0.01% optimize.opt_a.merge_forward : 0.000011s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.00% optimize.opt_a.offload_activation : 0.000025s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000100s : 0.03% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000030s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000011s : 0.00% optimize.opt_a.meta_fg_expand : 0.000007s : 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 : 0.000042s : 0.01% optimize.opt_a.a_after_grad : 0.000032s : 0.01% optimize.opt_a.renormalize : 0.005354s : 1.43% optimize.opt_a.add_forward_monad_depend : 0.000106s : 0.03% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000030s : 0.01% optimize.opt_a.cse : 0.000126s : 0.03% optimize.opt_a.a_3 : 0.000306s : 0.08% optimize.py_interpret_to_execute_after_opt_a : 0.000017s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000058s : 0.02% optimize.convert_after_rewriter : 0.000010s : 0.00% optimize.order_py_execute_after_rewriter : 0.000007s : 0.00% optimize.mutable_eliminate : 0.000815s : 0.22% optimize.opt_b.b_1 : 0.000235s : 0.06% optimize.opt_b.b_2 : 0.000013s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000009s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000004s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000004s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000063s : 0.02% optimize.optimize_parallel_all_gather_comm : 0.000021s : 0.01% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000028s : 0.01% optimize.loop_unroll : 0.000687s : 0.18% optimize.opt_after_cconv.c_1 : 0.000055s : 0.01% 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 : 0.000004s : 0.00% optimize.opt_after_cconv.cse : 0.000048s : 0.01% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000069s : 0.02% optimize.tuple_transform.d_1 : 0.000079s : 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 : 0.000011s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000064s : 0.02% optimize.cse_after_recomputation.cse : 0.000025s : 0.01% optimize.environ_conv : 0.000007s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000008s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000004s : 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 : 0.000003s : 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 : 0.000017s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000005s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 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 : 0.000005s : 0.00% optimize.overlap_grad_flash_sp : 0.000022s : 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 : 0.000165s : 0.04% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000016s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000019s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000011s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000141s : 0.04% 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 : 0.000026s : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000005s : 0.00% opt_after_jit_grad : 0.000507s : 0.14% validate : 0.000051s : 0.01% Time group info: ------[substitution.] 0.000147 33 2.35% : 0.000003s : 2: substitution.elim_not_effective 4.40% : 0.000006s : 2: substitution.fold_const_symbol 5.09% : 0.000008s : 8: substitution.graph_param_transform 75.29% : 0.000111s : 1: substitution.inline 2.84% : 0.000004s : 4: substitution.j_node_and_user_rematch 3.55% : 0.000005s : 4: substitution.remove_not_recompute_node 6.49% : 0.000010s : 12: substitution.replace_old_param ------[type_inference.] 0.358246 2 50.93% : 0.182464s : 1: type_inference.infer 49.07% : 0.175781s : 1: type_inference.specialize ------[replace.] 0.000022 1 100.00% : 0.000022s : 1: replace.inline ------[match.] 0.000110 1 100.00% : 0.000110s : 1: match.inline ------[predicate.] 0.000385 1879 0.50% : 0.000002s : 17: predicate.accumulaten_eliminater 0.51% : 0.000002s : 8: predicate.ad_related_special_op_eliminate 0.43% : 0.000002s : 16: predicate.addn_check_dump 0.52% : 0.000002s : 17: predicate.addn_zero_filter 0.42% : 0.000002s : 17: predicate.adjust_all_reduce_mul_add 1.38% : 0.000005s : 33: predicate.arithmetic_simplify 0.51% : 0.000002s : 17: predicate.cast_eliminate 0.46% : 0.000002s : 16: predicate.check_bprop_eliminate 0.41% : 0.000002s : 16: predicate.compare_switch_simplify 0.17% : 0.000001s : 8: predicate.const_output_eliminate 0.45% : 0.000002s : 16: predicate.depend_value_elim 0.48% : 0.000002s : 17: predicate.dict_get_item_const_eliminator 0.53% : 0.000002s : 17: predicate.dict_get_item_eliminator 0.47% : 0.000002s : 17: predicate.dict_set_item_eliminator 0.64% : 0.000002s : 16: predicate.dumpgradient_eliminate 0.25% : 0.000001s : 8: predicate.elim_not_effective 0.30% : 0.000001s : 8: predicate.elim_shapecalc_of_broadcastargs 0.70% : 0.000003s : 25: predicate.environ_add_const_eliminate 0.62% : 0.000002s : 25: predicate.environ_get_add_eliminate 0.63% : 0.000002s : 25: predicate.environ_get_depend_swap 1.10% : 0.000004s : 41: predicate.environ_get_eliminate 0.63% : 0.000002s : 25: predicate.environ_get_set_eliminate 0.45% : 0.000002s : 18: predicate.exchange_switch_depend_value 0.83% : 0.000003s : 18: predicate.float_depend_g_call 0.42% : 0.000002s : 16: predicate.float_environ_get_switch 0.62% : 0.000002s : 24: predicate.float_tuple_getitem_switch 0.18% : 0.000001s : 8: predicate.fold_const_symbol 0.51% : 0.000002s : 16: predicate.get_grad_eliminate 0.17% : 0.000001s : 8: predicate.graph_param_transform 0.44% : 0.000002s : 16: predicate.incorporate_call 0.38% : 0.000001s : 16: predicate.incorporate_call_switch 3.82% : 0.000015s : 83: predicate.inline 0.54% : 0.000002s : 16: predicate.inline_without_move 0.30% : 0.000001s : 16: predicate.j_node_and_user_rematch 0.58% : 0.000002s : 16: predicate.less_batch_normalization 0.94% : 0.000004s : 33: predicate.list_to_tuple_eliminator_ 1.45% : 0.000006s : 50: predicate.load_eliminater 0.59% : 0.000002s : 8: predicate.loop_unroll_after_grad 0.82% : 0.000003s : 25: predicate.loop_unroll_before_grad 1.03% : 0.000004s : 33: predicate.make_slice_get_slice_eliminator 0.46% : 0.000002s : 16: predicate.merge_addn 0.44% : 0.000002s : 16: predicate.micro_step_allgather_replace 0.49% : 0.000002s : 16: predicate.mini_step_allgather_replace 0.43% : 0.000002s : 17: predicate.minmaximum_grad 0.67% : 0.000003s : 8: predicate.mutable_eliminate 0.27% : 0.000001s : 8: predicate.opt_reshape 0.27% : 0.000001s : 8: predicate.parallel_virtual_node 0.61% : 0.000002s : 18: predicate.partial_defer_inline 0.71% : 0.000003s : 25: predicate.partial_eliminate 0.50% : 0.000002s : 17: predicate.print_const_string_wrapper 0.44% : 0.000002s : 16: predicate.reduce_all_const_elim 0.58% : 0.000002s : 17: predicate.reduce_eliminate 1.31% : 0.000005s : 50: predicate.redundant_stop_gradient_eliminater 0.52% : 0.000002s : 16: predicate.remove_not_recompute_node 0.92% : 0.000004s : 33: predicate.replace_applicator 0.47% : 0.000002s : 16: predicate.replace_old_param 0.26% : 0.000001s : 8: predicate.reset_defer_inline 0.50% : 0.000002s : 17: predicate.reshape_eliminate 0.48% : 0.000002s : 16: predicate.row_tensor_add_zeros_like 0.29% : 0.000001s : 8: predicate.row_tensor_eliminate 42.11% : 0.000162s : 16: predicate.same_eliminate 0.35% : 0.000001s : 16: predicate.set_cell_output_no_recompute 0.54% : 0.000002s : 16: predicate.shard_identity_eliminate 0.51% : 0.000002s : 16: predicate.special_op_eliminate 0.51% : 0.000002s : 16: predicate.specialize_transform 0.71% : 0.000003s : 16: predicate.split_environ_get_set_with_tuple_value 0.55% : 0.000002s : 16: predicate.stack_unstack_eliminate 0.25% : 0.000001s : 8: predicate.switch_call_monad_eliminater 0.49% : 0.000002s : 18: predicate.switch_defer_inline 0.99% : 0.000004s : 34: predicate.switch_layer_defer_inline 2.29% : 0.000009s : 67: predicate.switch_simplify 0.45% : 0.000002s : 17: predicate.tile_eliminate 0.47% : 0.000002s : 17: predicate.transpose_eliminate 0.97% : 0.000004s : 33: predicate.tuple_list_convert_item_index_to_positive 1.01% : 0.000004s : 33: predicate.tuple_list_get_item_const_eliminator 0.83% : 0.000003s : 33: predicate.tuple_list_get_item_depend_reorder 1.94% : 0.000007s : 49: predicate.tuple_list_get_item_eliminator 0.90% : 0.000003s : 33: predicate.tuple_list_get_set_item_eliminator 1.44% : 0.000006s : 49: predicate.tuple_list_set_item_eliminator 1.00% : 0.000004s : 33: predicate.tuple_to_list_eliminator_ 1.21% : 0.000005s : 50: predicate.updatestate_pure_node_eliminater 1.78% : 0.000007s : 66: predicate.updatestate_useless_node_eliminater 0.27% : 0.000001s : 8: predicate.value_based_eliminate 0.52% : 0.000002s : 16: predicate.virtual_dataset_eliminate 0.50% : 0.000002s : 16: predicate.virtual_output_eliminate 0.23% : 0.000001s : 8: predicate.virtual_view_grad_eliminate 0.35% : 0.000001s : 8: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.009800 6 12.65% : 0.001240s : 3: func_graph_cloner_run.FuncGraphClonerGraph 87.35% : 0.008560s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.130552 192 0.00% : 0.000004s : 1: ForceFp32Comm 32.50% : 0.367427s : 1: add_attr 32.49% : 0.367272s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000068s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.04% : 0.000449s : 1: auto_monad 0.00% : 0.000030s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000005s : 1: bias_add_comm_swap 0.24% : 0.002760s : 1: bootstrap 0.00% : 0.000032s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000021s : 1: control_data_broadcast_order 0.00% : 0.000014s : 1: convert_after_rewriter 0.00% : 0.000039s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000010s : 1: environ_conv 0.00% : 0.000031s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000010s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000007s : 1: label_micro_interleaved_index 0.06% : 0.000698s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.09% : 0.001002s : 1: mutable_eliminate 0.00% : 0.000008s : 1: offloading_packed_experts 0.00% : 0.000020s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000030s : 1: opt.transform.mutable_eliminate 0.15% : 0.001744s : 78: opt.transform.opt_a 0.00% : 0.000053s : 1: opt.transform.opt_after_cconv 0.00% : 0.000039s : 1: opt.transform.opt_after_jit_grad 0.02% : 0.000215s : 28: opt.transform.opt_b 0.01% : 0.000088s : 2: opt.transform.opt_trans_graph 0.02% : 0.000183s : 4: opt.transform.symbol_engine_opt 0.81% : 0.009138s : 1: opt_a 0.02% : 0.000172s : 1: opt_after_cconv 0.05% : 0.000516s : 1: opt_after_jit_grad 0.04% : 0.000465s : 1: opt_b 1.14% : 0.012882s : 1: optimize 0.00% : 0.000025s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000010s : 1: order_py_execute_after_rewriter 0.00% : 0.000026s : 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% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000004s : 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% : 0.000007s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.00% : 0.000039s : 1: pre_auto_parallel 0.01% : 0.000111s : 1: py_interpret_to_execute 0.00% : 0.000022s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.01% : 0.000073s : 1: remove_dup_value 0.41% : 0.004615s : 1: renormalize.infer 0.06% : 0.000726s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000008s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000062s : 1: rewriter_after_opt_a 0.01% : 0.000083s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000011s : 1: swap_dp_allreduce_reducescatter 0.04% : 0.000398s : 1: symbol_engine_optimizer 0.01% : 0.000120s : 1: tuple_transform 31.72% : 0.358602s : 1: type_inference .[WARNING] PRE_ACT(170442,ffffbeca8f30,python3.9):2026-01-29-17:41:16.774.238 [mindspore/ccsrc/runtime/memory/mem_pool/abstract_dynamic_mem_pool.cc:1109] FreeIdleMemsByEagerFree] Eager free count : 4, free memory : 2725209600, real free : 2726297600, not free : 0. [hook] pytest_runtest_teardown:test_paged_attention_trapezoidal[0-TRAPEZOIDAL-float32] tests/st/infer/ops/test_internal_ops/test_paged_attention.py::test_paged_attention_trapezoidal[0-TRAPEZOIDAL-float32],max_mem:2696.0M [WARNING] ME(170442:281473882689328,MainProcess):2026-01-29-17:41:18.197.043 [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 = 0.320151, [21] [bootstrap]: 0.00077255 [type_inference]: 0.00663233 [event_method]: 1.675e-05 [auto_monad]: 6.04e-05 [graph_reusing]: 5.57001e-06 [inline]: 2.51e-06 [add_attr]: 0.30404, [1] [add_attr_with_inline]: 0.304023, [1] [Cycle 1]: 7.733e-05, [2] [tag_attr]: 2.397e-05 [meta_addattr_fg_expand]: 3.78001e-06 [parallel-infer-symbol]: 3.68999e-06 [pre_auto_parallel]: 3.586e-05 [insert-virtual-dataset]: 2.83e-06 [parallel-infer-symbol-second]: 6.90023e-07 [dataset_repeat_opt]: 2.26e-06 [pipeline_split]: 1.55001e-06 [optimize]: 0.00768494, [53] [py_interpret_to_execute]: 4.548e-05 [rewriter_before_opt_a]: 7.905e-05 [opt_a]: 0.00449937, [2] [Cycle 1]: 0.00327558, [45] [expand_dump_flag]: 2.71e-06 [switch_simplify]: 9.928e-05 [loop_unroll]: 1.956e-05 [a_1]: 0.00057085 [with_stream_mark]: 3.129e-05 [recompute_prepare]: 1.877e-05 [updatestate_depend_eliminate]: 7e-06 [updatestate_assign_eliminate]: 5.85002e-06 [updatestate_loads_eliminate]: 5.19003e-06 [parameter_eliminate]: 2.64001e-06 [a_2]: 0.0001558 [accelerated_algorithm]: 1.458e-05 [shard]: 3.31999e-06 [meta_shard_fg_expand]: 3.03e-06 [shard_inline]: 1.122e-05 [merge_send_recv]: 1.143e-05 [auto_parallel]: 1.169e-05 [parallel]: 4.853e-05 [flash_sp]: 1.379e-05 [merge_comm]: 6.52001e-06 [allreduce_fusion]: 5.34e-06 [matmul_add_comm_reduction]: 1.38e-05 [allreduce_slice_to_reducescatter]: 7.30011e-07 [virtual_shard_identity]: 1.74e-05 [virtual_dataset]: 1.225e-05 [get_grad_eliminate_]: 1.194e-05 [virtual_output]: 1.18e-05 [merge_forward]: 6.19999e-06 [cell_reuse_recompute_pass]: 2.39001e-06 [offload_activation]: 1.421e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.372e-05 [merge_recompute_call_nodes]: 1.67001e-06 [before_grad]: 1.741e-05 [set_forward_comm_id_for_comm_node_pass]: 5.45001e-06 [meta_fg_expand]: 3.56001e-06 [flash_sp_send_recv_attached]: 2.94999e-06 [receive_attached]: 2.86e-06 [after_resolve]: 6.582e-05 [a_after_grad]: 3.04e-05 [renormalize]: 0.00137331 [add_forward_monad_depend]: 2.423e-05 [auto_monad_grad]: 2.49999e-06 [auto_monad_eliminator]: 2.466e-05 [cse]: 9.599e-05 [a_3]: 9.122e-05 [Cycle 2]: 0.00117515, [45] [expand_dump_flag]: 2.88e-06 [switch_simplify]: 1.432e-05 [loop_unroll]: 1.175e-05 [a_1]: 0.00024396 [with_stream_mark]: 2.114e-05 [recompute_prepare]: 1.267e-05 [updatestate_depend_eliminate]: 6.79999e-06 [updatestate_assign_eliminate]: 5.47999e-06 [updatestate_loads_eliminate]: 5.30001e-06 [parameter_eliminate]: 2.14e-06 [a_2]: 0.00012948 [accelerated_algorithm]: 1.22e-05 [shard]: 2.84001e-06 [meta_shard_fg_expand]: 3.16001e-06 [shard_inline]: 1.075e-05 [merge_send_recv]: 1.253e-05 [auto_parallel]: 1.231e-05 [parallel]: 1.079e-05 [flash_sp]: 4.07e-06 [merge_comm]: 6.21998e-06 [allreduce_fusion]: 5.19e-06 [matmul_add_comm_reduction]: 1.213e-05 [allreduce_slice_to_reducescatter]: 6.80011e-07 [virtual_shard_identity]: 1.299e-05 [virtual_dataset]: 1.072e-05 [get_grad_eliminate_]: 9.83002e-06 [virtual_output]: 9.69e-06 [merge_forward]: 6.66999e-06 [cell_reuse_recompute_pass]: 2.69999e-06 [offload_activation]: 1.313e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.486e-05 [merge_recompute_call_nodes]: 1.40999e-06 [before_grad]: 1.684e-05 [set_forward_comm_id_for_comm_node_pass]: 7.16001e-06 [meta_fg_expand]: 3.69002e-06 [flash_sp_send_recv_attached]: 1.69998e-06 [receive_attached]: 2.71e-06 [after_resolve]: 3.166e-05 [a_after_grad]: 1.722e-05 [renormalize]: 1.00001e-07 [add_forward_monad_depend]: 2.89001e-06 [auto_monad_grad]: 2.59999e-06 [auto_monad_eliminator]: 1.425e-05 [cse]: 5.855e-05 [a_3]: 6.423e-05 [py_interpret_to_execute_after_opt_a]: 2.566e-05 [slice_cell_reuse_recomputed_activation]: 2.69999e-06 [rewriter_after_opt_a]: 6.961e-05 [convert_after_rewriter]: 1.102e-05 [order_py_execute_after_rewriter]: 7.02002e-06 [mutable_eliminate]: 0.00083486 [opt_b]: 0.00041536, [1] [Cycle 1]: 0.00040407, [7] [b_1]: 0.0002395 [b_2]: 1.299e-05 [updatestate_depend_eliminate]: 1.149e-05 [updatestate_assign_eliminate]: 5.04e-06 [updatestate_loads_eliminate]: 4.72e-06 [renormalize]: 1.34e-06 [cse]: 8.257e-05 [optimize_parallel_all_gather_comm]: 3.684e-05 [overlap_param_gather]: 2.32999e-06 [cconv]: 3.78e-05 [loop_unroll]: 0.0005301 [opt_after_cconv]: 0.00017607, [1] [Cycle 1]: 0.00016833, [7] [c_1]: 5.528e-05 [parameter_eliminate]: 4.42e-06 [updatestate_depend_eliminate]: 7.97998e-06 [updatestate_assign_eliminate]: 4.31002e-06 [updatestate_loads_eliminate]: 4.22e-06 [cse]: 5.102e-05 [renormalize]: 5.79981e-07 [remove_dup_value]: 7.494e-05 [tuple_transform]: 0.00012552, [1] [Cycle 1]: 0.00011966, [4] [d_1]: 8.556e-05 [none_parameter_eliminate]: 1.91e-06 [renormalize]: 1.69995e-07 [switch_simplify]: 1.098e-05 [partial_unused_args_eliminate]: 1.75001e-06 [add_recomputation]: 7.275e-05 [cse_after_recomputation]: 3.701e-05, [1] [Cycle 1]: 3.169e-05, [1] [cse]: 2.494e-05 [environ_conv]: 8.05e-06 [swap_dp_allreduce_reducescatter]: 7.6e-06 [bias_add_comm_swap]: 2.36998e-06 [label_micro_interleaved_index]: 4.92e-06 [label_fine_grained_interleaved_index]: 2.97002e-06 [merge_cast_opt]: 1.55001e-06 [slice_recompute_activation]: 2.17999e-06 [micro_interleaved_order_control]: 2.69001e-06 [assign_add_opt]: 1.21997e-06 [ForceFp32Comm]: 8.50006e-07 [remove_cast_before_assign_add]: 1.17999e-06 [full_micro_interleaved_order_control]: 2.37001e-06 [reorder_send_recv_between_fp_bp]: 3.10998e-06 [comm_op_add_attrs]: 1.41002e-06 [add_comm_op_reuse_tag]: 1.26997e-06 [interleave_split_concat_branches]: 1.33002e-06 [interleave_parallel_branches]: 1.19e-06 [overlap_opt_shard_in_pipeline]: 4.68999e-06 [overlap_opt_shard_grad_in_pipeline]: 1.69e-06 [control_data_broadcast_order]: 1.812e-05 [grouped_pairwise_exchange_alltoall]: 1.56002e-06 [offloading_packed_experts]: 5.64e-06 [overlap_recompute_and_grad_model_parallel]: 5.77001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.32e-06 [overlap_recompute_allgather_and_fa_grad]: 1.40999e-06 [overlap_recompute_comm]: 2.75002e-06 [overlap_grad_ring_attention]: 4.74e-06 [overlap_grad_flash_sp]: 2.797e-05 [begin_end_overlap_inline]: 6.60017e-07 [split_matmul_comm_elemetwise]: 2.71e-06 [split_layernorm_comm]: 1.66e-06 [handle_group_info]: 1.00001e-06 [symbol_engine_optimizer]: 0.00021135, [1] [Cycle 1]: 0.00020584, [6] [build]: 9.719e-05 [elim_shapecalc]: 1.801e-05 [elim_not_effective]: 2.03e-05 [opt_reshape]: 1.099e-05 [fold_const_symbol]: 2.034e-05 [renormalize]: 4.10015e-07 [detach_backward]: 2.19999e-06 [pipeline_parallel_scheduler]: 1.59e-06 [auto_monad_reorder]: 2.72e-05 [get_jit_bprop_graph]: 1.96e-06 [rewriter_after_jit_bprop_graph]: 6.27001e-06 [opt_after_jit_grad]: 0.00060285 [validate]: 5.659e-05 Sums bootstrap : 0.000773s : 5.23% type_inference : 0.006632s : 44.91% event_method : 0.000017s : 0.11% auto_monad : 0.000060s : 0.41% graph_reusing : 0.000006s : 0.04% inline : 0.000003s : 0.02% add_attr.add_attr_with_inline.tag_attr : 0.000024s : 0.16% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.03% parallel-infer-symbol : 0.000004s : 0.02% pre_auto_parallel : 0.000036s : 0.24% insert-virtual-dataset : 0.000003s : 0.02% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.02% pipeline_split : 0.000002s : 0.01% optimize.py_interpret_to_execute : 0.000045s : 0.31% optimize.rewriter_before_opt_a : 0.000079s : 0.54% optimize.opt_a.expand_dump_flag : 0.000006s : 0.04% optimize.opt_a.switch_simplify : 0.000114s : 0.77% optimize.opt_a.loop_unroll : 0.000031s : 0.21% optimize.opt_a.a_1 : 0.000815s : 5.52% optimize.opt_a.with_stream_mark : 0.000052s : 0.36% optimize.opt_a.recompute_prepare : 0.000031s : 0.21% optimize.opt_a.updatestate_depend_eliminate : 0.000014s : 0.09% optimize.opt_a.updatestate_assign_eliminate : 0.000011s : 0.08% optimize.opt_a.updatestate_loads_eliminate : 0.000010s : 0.07% optimize.opt_a.parameter_eliminate : 0.000005s : 0.03% optimize.opt_a.a_2 : 0.000285s : 1.93% optimize.opt_a.accelerated_algorithm : 0.000027s : 0.18% optimize.opt_a.shard : 0.000006s : 0.04% optimize.opt_a.meta_shard_fg_expand : 0.000006s : 0.04% optimize.opt_a.shard_inline : 0.000022s : 0.15% optimize.opt_a.merge_send_recv : 0.000024s : 0.16% optimize.opt_a.auto_parallel : 0.000024s : 0.16% optimize.opt_a.parallel : 0.000059s : 0.40% optimize.opt_a.flash_sp : 0.000018s : 0.12% optimize.opt_a.merge_comm : 0.000013s : 0.09% optimize.opt_a.allreduce_fusion : 0.000011s : 0.07% optimize.opt_a.matmul_add_comm_reduction : 0.000026s : 0.18% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000030s : 0.21% optimize.opt_a.virtual_dataset : 0.000023s : 0.16% optimize.opt_a.get_grad_eliminate_ : 0.000022s : 0.15% optimize.opt_a.virtual_output : 0.000021s : 0.15% optimize.opt_a.merge_forward : 0.000013s : 0.09% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.03% optimize.opt_a.offload_activation : 0.000027s : 0.19% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000049s : 0.33% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.02% optimize.opt_a.before_grad : 0.000034s : 0.23% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000013s : 0.09% optimize.opt_a.meta_fg_expand : 0.000007s : 0.05% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.03% optimize.opt_a.receive_attached : 0.000006s : 0.04% optimize.opt_a.after_resolve : 0.000097s : 0.66% optimize.opt_a.a_after_grad : 0.000048s : 0.32% optimize.opt_a.renormalize : 0.001373s : 9.30% optimize.opt_a.add_forward_monad_depend : 0.000027s : 0.18% optimize.opt_a.auto_monad_grad : 0.000005s : 0.03% optimize.opt_a.auto_monad_eliminator : 0.000039s : 0.26% optimize.opt_a.cse : 0.000155s : 1.05% optimize.opt_a.a_3 : 0.000155s : 1.05% optimize.py_interpret_to_execute_after_opt_a : 0.000026s : 0.17% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.02% optimize.rewriter_after_opt_a : 0.000070s : 0.47% optimize.convert_after_rewriter : 0.000011s : 0.07% optimize.order_py_execute_after_rewriter : 0.000007s : 0.05% optimize.mutable_eliminate : 0.000835s : 5.65% optimize.opt_b.b_1 : 0.000240s : 1.62% optimize.opt_b.b_2 : 0.000013s : 0.09% optimize.opt_b.updatestate_depend_eliminate : 0.000011s : 0.08% optimize.opt_b.updatestate_assign_eliminate : 0.000005s : 0.03% optimize.opt_b.updatestate_loads_eliminate : 0.000005s : 0.03% optimize.opt_b.renormalize : 0.000001s : 0.01% optimize.opt_b.cse : 0.000083s : 0.56% optimize.optimize_parallel_all_gather_comm : 0.000037s : 0.25% optimize.overlap_param_gather : 0.000002s : 0.02% optimize.cconv : 0.000038s : 0.26% optimize.loop_unroll : 0.000530s : 3.59% optimize.opt_after_cconv.c_1 : 0.000055s : 0.37% optimize.opt_after_cconv.parameter_eliminate : 0.000004s : 0.03% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000008s : 0.05% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000004s : 0.03% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000004s : 0.03% optimize.opt_after_cconv.cse : 0.000051s : 0.35% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000075s : 0.51% optimize.tuple_transform.d_1 : 0.000086s : 0.58% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.01% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000011s : 0.07% optimize.partial_unused_args_eliminate : 0.000002s : 0.01% optimize.add_recomputation : 0.000073s : 0.49% optimize.cse_after_recomputation.cse : 0.000025s : 0.17% optimize.environ_conv : 0.000008s : 0.05% optimize.swap_dp_allreduce_reducescatter : 0.000008s : 0.05% optimize.bias_add_comm_swap : 0.000002s : 0.02% optimize.label_micro_interleaved_index : 0.000005s : 0.03% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.02% optimize.merge_cast_opt : 0.000002s : 0.01% optimize.slice_recompute_activation : 0.000002s : 0.01% optimize.micro_interleaved_order_control : 0.000003s : 0.02% optimize.assign_add_opt : 0.000001s : 0.01% optimize.ForceFp32Comm : 0.000001s : 0.01% optimize.remove_cast_before_assign_add : 0.000001s : 0.01% optimize.full_micro_interleaved_order_control : 0.000002s : 0.02% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.02% optimize.comm_op_add_attrs : 0.000001s : 0.01% optimize.add_comm_op_reuse_tag : 0.000001s : 0.01% optimize.interleave_split_concat_branches : 0.000001s : 0.01% optimize.interleave_parallel_branches : 0.000001s : 0.01% optimize.overlap_opt_shard_in_pipeline : 0.000005s : 0.03% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.01% optimize.control_data_broadcast_order : 0.000018s : 0.12% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.01% optimize.offloading_packed_experts : 0.000006s : 0.04% optimize.overlap_recompute_and_grad_model_parallel : 0.000006s : 0.04% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.01% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.01% optimize.overlap_recompute_comm : 0.000003s : 0.02% optimize.overlap_grad_ring_attention : 0.000005s : 0.03% optimize.overlap_grad_flash_sp : 0.000028s : 0.19% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.02% optimize.split_layernorm_comm : 0.000002s : 0.01% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000097s : 0.66% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000018s : 0.12% optimize.symbol_engine_optimizer.elim_not_effective : 0.000020s : 0.14% optimize.symbol_engine_optimizer.opt_reshape : 0.000011s : 0.07% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000020s : 0.14% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.01% pipeline_parallel_scheduler : 0.000002s : 0.01% auto_monad_reorder : 0.000027s : 0.18% get_jit_bprop_graph : 0.000002s : 0.01% rewriter_after_jit_bprop_graph : 0.000006s : 0.04% opt_after_jit_grad : 0.000603s : 4.08% validate : 0.000057s : 0.38% Time group info: ------[substitution.] 0.000173 33 2.63% : 0.000005s : 2: substitution.elim_not_effective 4.51% : 0.000008s : 2: substitution.fold_const_symbol 4.73% : 0.000008s : 8: substitution.graph_param_transform 75.59% : 0.000131s : 1: substitution.inline 2.99% : 0.000005s : 4: substitution.j_node_and_user_rematch 3.57% : 0.000006s : 4: substitution.remove_not_recompute_node 5.98% : 0.000010s : 12: substitution.replace_old_param ------[type_inference.] 0.006519 2 89.04% : 0.005804s : 1: type_inference.infer 10.96% : 0.000715s : 1: type_inference.specialize ------[replace.] 0.000024 1 100.00% : 0.000024s : 1: replace.inline ------[match.] 0.000130 1 100.00% : 0.000130s : 1: match.inline ------[predicate.] 0.000237 1879 0.97% : 0.000002s : 17: predicate.accumulaten_eliminater 0.81% : 0.000002s : 8: predicate.ad_related_special_op_eliminate 0.71% : 0.000002s : 16: predicate.addn_check_dump 0.83% : 0.000002s : 17: predicate.addn_zero_filter 0.79% : 0.000002s : 17: predicate.adjust_all_reduce_mul_add 2.17% : 0.000005s : 33: predicate.arithmetic_simplify 0.86% : 0.000002s : 17: predicate.cast_eliminate 0.80% : 0.000002s : 16: predicate.check_bprop_eliminate 0.70% : 0.000002s : 16: predicate.compare_switch_simplify 0.27% : 0.000001s : 8: predicate.const_output_eliminate 0.68% : 0.000002s : 16: predicate.depend_value_elim 0.77% : 0.000002s : 17: predicate.dict_get_item_const_eliminator 0.88% : 0.000002s : 17: predicate.dict_get_item_eliminator 0.96% : 0.000002s : 17: predicate.dict_set_item_eliminator 1.68% : 0.000004s : 16: predicate.dumpgradient_eliminate 0.31% : 0.000001s : 8: predicate.elim_not_effective 0.49% : 0.000001s : 8: predicate.elim_shapecalc_of_broadcastargs 1.11% : 0.000003s : 25: predicate.environ_add_const_eliminate 1.03% : 0.000002s : 25: predicate.environ_get_add_eliminate 1.02% : 0.000002s : 25: predicate.environ_get_depend_swap 1.79% : 0.000004s : 41: predicate.environ_get_eliminate 1.06% : 0.000003s : 25: predicate.environ_get_set_eliminate 0.72% : 0.000002s : 18: predicate.exchange_switch_depend_value 1.44% : 0.000003s : 18: predicate.float_depend_g_call 0.74% : 0.000002s : 16: predicate.float_environ_get_switch 1.27% : 0.000003s : 24: predicate.float_tuple_getitem_switch 0.24% : 0.000001s : 8: predicate.fold_const_symbol 0.91% : 0.000002s : 16: predicate.get_grad_eliminate 0.40% : 0.000001s : 8: predicate.graph_param_transform 0.75% : 0.000002s : 16: predicate.incorporate_call 0.66% : 0.000002s : 16: predicate.incorporate_call_switch 5.47% : 0.000013s : 83: predicate.inline 1.02% : 0.000002s : 16: predicate.inline_without_move 0.50% : 0.000001s : 16: predicate.j_node_and_user_rematch 1.04% : 0.000002s : 16: predicate.less_batch_normalization 1.78% : 0.000004s : 33: predicate.list_to_tuple_eliminator_ 2.10% : 0.000005s : 50: predicate.load_eliminater 1.17% : 0.000003s : 8: predicate.loop_unroll_after_grad 1.57% : 0.000004s : 25: predicate.loop_unroll_before_grad 1.91% : 0.000005s : 33: predicate.make_slice_get_slice_eliminator 0.86% : 0.000002s : 16: predicate.merge_addn 0.84% : 0.000002s : 16: predicate.micro_step_allgather_replace 0.84% : 0.000002s : 16: predicate.mini_step_allgather_replace 0.69% : 0.000002s : 17: predicate.minmaximum_grad 1.76% : 0.000004s : 8: predicate.mutable_eliminate 0.55% : 0.000001s : 8: predicate.opt_reshape 0.44% : 0.000001s : 8: predicate.parallel_virtual_node 1.10% : 0.000003s : 18: predicate.partial_defer_inline 1.18% : 0.000003s : 25: predicate.partial_eliminate 0.82% : 0.000002s : 17: predicate.print_const_string_wrapper 0.81% : 0.000002s : 16: predicate.reduce_all_const_elim 1.17% : 0.000003s : 17: predicate.reduce_eliminate 2.38% : 0.000006s : 50: predicate.redundant_stop_gradient_eliminater 0.71% : 0.000002s : 16: predicate.remove_not_recompute_node 1.69% : 0.000004s : 33: predicate.replace_applicator 0.78% : 0.000002s : 16: predicate.replace_old_param 0.30% : 0.000001s : 8: predicate.reset_defer_inline 0.78% : 0.000002s : 17: predicate.reshape_eliminate 0.86% : 0.000002s : 16: predicate.row_tensor_add_zeros_like 0.46% : 0.000001s : 8: predicate.row_tensor_eliminate 1.03% : 0.000002s : 16: predicate.same_eliminate 0.68% : 0.000002s : 16: predicate.set_cell_output_no_recompute 0.98% : 0.000002s : 16: predicate.shard_identity_eliminate 0.83% : 0.000002s : 16: predicate.special_op_eliminate 0.99% : 0.000002s : 16: predicate.specialize_transform 1.44% : 0.000003s : 16: predicate.split_environ_get_set_with_tuple_value 1.22% : 0.000003s : 16: predicate.stack_unstack_eliminate 0.43% : 0.000001s : 8: predicate.switch_call_monad_eliminater 0.81% : 0.000002s : 18: predicate.switch_defer_inline 1.56% : 0.000004s : 34: predicate.switch_layer_defer_inline 3.91% : 0.000009s : 67: predicate.switch_simplify 0.82% : 0.000002s : 17: predicate.tile_eliminate 0.77% : 0.000002s : 17: predicate.transpose_eliminate 1.49% : 0.000004s : 33: predicate.tuple_list_convert_item_index_to_positive 1.53% : 0.000004s : 33: predicate.tuple_list_get_item_const_eliminator 1.39% : 0.000003s : 33: predicate.tuple_list_get_item_depend_reorder 3.37% : 0.000008s : 49: predicate.tuple_list_get_item_eliminator 1.47% : 0.000003s : 33: predicate.tuple_list_get_set_item_eliminator 2.33% : 0.000006s : 49: predicate.tuple_list_set_item_eliminator 1.63% : 0.000004s : 33: predicate.tuple_to_list_eliminator_ 2.05% : 0.000005s : 50: predicate.updatestate_pure_node_eliminater 2.98% : 0.000007s : 66: predicate.updatestate_useless_node_eliminater 0.42% : 0.000001s : 8: predicate.value_based_eliminate 0.82% : 0.000002s : 16: predicate.virtual_dataset_eliminate 0.79% : 0.000002s : 16: predicate.virtual_output_eliminate 0.37% : 0.000001s : 8: predicate.virtual_view_grad_eliminate 0.46% : 0.000001s : 8: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000829 6 44.13% : 0.000366s : 3: func_graph_cloner_run.FuncGraphClonerGraph 55.87% : 0.000463s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.635305 192 0.00% : 0.000004s : 1: ForceFp32Comm 47.86% : 0.304047s : 1: add_attr 47.86% : 0.304028s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000077s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.01% : 0.000066s : 1: auto_monad 0.01% : 0.000032s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000005s : 1: bias_add_comm_swap 0.13% : 0.000817s : 1: bootstrap 0.01% : 0.000042s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000022s : 1: control_data_broadcast_order 0.00% : 0.000015s : 1: convert_after_rewriter 0.01% : 0.000040s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000011s : 1: environ_conv 0.00% : 0.000024s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 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% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.09% : 0.000541s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.13% : 0.000851s : 1: mutable_eliminate 0.00% : 0.000009s : 1: offloading_packed_experts 0.00% : 0.000022s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000027s : 1: opt.transform.mutable_eliminate 0.27% : 0.001700s : 78: opt.transform.opt_a 0.01% : 0.000054s : 1: opt.transform.opt_after_cconv 0.01% : 0.000042s : 1: opt.transform.opt_after_jit_grad 0.03% : 0.000218s : 28: opt.transform.opt_b 0.01% : 0.000094s : 2: opt.transform.opt_trans_graph 0.01% : 0.000066s : 4: opt.transform.symbol_engine_opt 0.71% : 0.004504s : 1: opt_a 0.03% : 0.000180s : 1: opt_after_cconv 0.10% : 0.000617s : 1: opt_after_jit_grad 0.07% : 0.000420s : 1: opt_b 1.21% : 0.007692s : 1: optimize 0.01% : 0.000042s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000011s : 1: order_py_execute_after_rewriter 0.01% : 0.000032s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000009s : 1: overlap_grad_ring_attention 0.00% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000008s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000009s : 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 0.01% : 0.000040s : 1: pre_auto_parallel 0.01% : 0.000050s : 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 0.01% : 0.000080s : 1: remove_dup_value 0.12% : 0.000765s : 1: renormalize.infer 0.09% : 0.000593s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000076s : 1: rewriter_after_opt_a 0.01% : 0.000084s : 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% : 0.000004s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000011s : 1: swap_dp_allreduce_reducescatter 0.03% : 0.000215s : 1: symbol_engine_optimizer 0.02% : 0.000129s : 1: tuple_transform 1.05% : 0.006657s : 1: type_inference .[WARNING] PRE_ACT(170442,ffffbeca8f30,python3.9):2026-01-29-17:41:54.633.160 [mindspore/ccsrc/runtime/memory/mem_pool/abstract_dynamic_mem_pool.cc:1109] FreeIdleMemsByEagerFree] Eager free count : 5, free memory : 1383031808, real free : 1384120320, not free : 0. [hook] pytest_runtest_teardown:test_paged_attention_trapezoidal[1-MASK_DEFAULT-float16] tests/st/infer/ops/test_internal_ops/test_paged_attention.py::test_paged_attention_trapezoidal[1-MASK_DEFAULT-float16],max_mem:2696.0M [WARNING] ME(170442:281473882689328,MainProcess):2026-01-29-17:41:55.541.716 [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 = 0.154448, [21] [bootstrap]: 0.00084012 [type_inference]: 0.0310596 [event_method]: 1.86e-05 [auto_monad]: 6.078e-05 [graph_reusing]: 4.56002e-06 [inline]: 3.33998e-06 [add_attr]: 0.0182181, [1] [add_attr_with_inline]: 0.0181985, [1] [Cycle 1]: 8.617e-05, [2] [tag_attr]: 2.709e-05 [meta_addattr_fg_expand]: 3.76999e-06 [parallel-infer-symbol]: 3.46999e-06 [pre_auto_parallel]: 3.694e-05 [insert-virtual-dataset]: 2.62001e-06 [parallel-infer-symbol-second]: 7.30011e-07 [dataset_repeat_opt]: 2.04999e-06 [pipeline_split]: 1.60999e-06 [optimize]: 0.103203, [53] [py_interpret_to_execute]: 3.767e-05 [rewriter_before_opt_a]: 8.304e-05 [opt_a]: 0.0999289, [2] [Cycle 1]: 0.0987412, [45] [expand_dump_flag]: 2.68e-06 [switch_simplify]: 2.977e-05 [loop_unroll]: 1.813e-05 [a_1]: 0.0006565 [with_stream_mark]: 3.548e-05 [recompute_prepare]: 1.69e-05 [updatestate_depend_eliminate]: 6.56999e-06 [updatestate_assign_eliminate]: 5.49e-06 [updatestate_loads_eliminate]: 4.71002e-06 [parameter_eliminate]: 2.83e-06 [a_2]: 0.00014901 [accelerated_algorithm]: 1.397e-05 [shard]: 2.75002e-06 [meta_shard_fg_expand]: 5.83002e-06 [shard_inline]: 1.192e-05 [merge_send_recv]: 1.187e-05 [auto_parallel]: 1.323e-05 [parallel]: 3.527e-05 [flash_sp]: 1.378e-05 [merge_comm]: 6.06003e-06 [allreduce_fusion]: 5.51e-06 [matmul_add_comm_reduction]: 1.336e-05 [allreduce_slice_to_reducescatter]: 9.49978e-07 [virtual_shard_identity]: 1.576e-05 [virtual_dataset]: 1.261e-05 [get_grad_eliminate_]: 1.172e-05 [virtual_output]: 1.532e-05 [merge_forward]: 9.74e-06 [cell_reuse_recompute_pass]: 1.86998e-06 [offload_activation]: 1.413e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.579e-05 [merge_recompute_call_nodes]: 1.76e-06 [before_grad]: 2.293e-05 [set_forward_comm_id_for_comm_node_pass]: 7.19001e-06 [meta_fg_expand]: 3.91001e-06 [flash_sp_send_recv_attached]: 3.00002e-06 [receive_attached]: 2.41998e-06 [after_resolve]: 3.255e-05 [a_after_grad]: 2.463e-05 [renormalize]: 0.00106425 [add_forward_monad_depend]: 1.389e-05 [auto_monad_grad]: 2.67001e-06 [auto_monad_eliminator]: 2.962e-05 [cse]: 8.66e-05 [a_3]: 8.469e-05 [Cycle 2]: 0.00116887, [45] [expand_dump_flag]: 2.56e-06 [switch_simplify]: 1.339e-05 [loop_unroll]: 1.051e-05 [a_1]: 0.00025778 [with_stream_mark]: 2.311e-05 [recompute_prepare]: 1.25e-05 [updatestate_depend_eliminate]: 6.73e-06 [updatestate_assign_eliminate]: 5.56e-06 [updatestate_loads_eliminate]: 5.37999e-06 [parameter_eliminate]: 2.24001e-06 [a_2]: 0.00012534 [accelerated_algorithm]: 1.052e-05 [shard]: 2.74001e-06 [meta_shard_fg_expand]: 3.01001e-06 [shard_inline]: 9.75002e-06 [merge_send_recv]: 1.049e-05 [auto_parallel]: 1.189e-05 [parallel]: 1.012e-05 [flash_sp]: 4.09002e-06 [merge_comm]: 5.47001e-06 [allreduce_fusion]: 4.89e-06 [matmul_add_comm_reduction]: 1.181e-05 [allreduce_slice_to_reducescatter]: 1.00999e-06 [virtual_shard_identity]: 1.356e-05 [virtual_dataset]: 1.101e-05 [get_grad_eliminate_]: 1.004e-05 [virtual_output]: 9.62001e-06 [merge_forward]: 6.12999e-06 [cell_reuse_recompute_pass]: 2.93998e-06 [offload_activation]: 1.295e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.42e-05 [merge_recompute_call_nodes]: 1.52001e-06 [before_grad]: 1.652e-05 [set_forward_comm_id_for_comm_node_pass]: 6.31e-06 [meta_fg_expand]: 3.90998e-06 [flash_sp_send_recv_attached]: 2.09e-06 [receive_attached]: 2.39001e-06 [after_resolve]: 2.311e-05 [a_after_grad]: 1.622e-05 [renormalize]: 8.00064e-08 [add_forward_monad_depend]: 3.93001e-06 [auto_monad_grad]: 1.68002e-06 [auto_monad_eliminator]: 1.563e-05 [cse]: 6.405e-05 [a_3]: 6.637e-05 [py_interpret_to_execute_after_opt_a]: 2.604e-05 [slice_cell_reuse_recomputed_activation]: 2.46e-06 [rewriter_after_opt_a]: 6.638e-05 [convert_after_rewriter]: 1.019e-05 [order_py_execute_after_rewriter]: 7.74002e-06 [mutable_eliminate]: 0.00081609 [opt_b]: 0.00042043, [1] [Cycle 1]: 0.00041, [7] [b_1]: 0.00024313 [b_2]: 1.361e-05 [updatestate_depend_eliminate]: 1.24e-05 [updatestate_assign_eliminate]: 5.52999e-06 [updatestate_loads_eliminate]: 5.49e-06 [renormalize]: 6.00005e-07 [cse]: 8.294e-05 [optimize_parallel_all_gather_comm]: 2.591e-05 [overlap_param_gather]: 5.56998e-06 [cconv]: 3.755e-05 [loop_unroll]: 0.00056167 [opt_after_cconv]: 0.00018562, [1] [Cycle 1]: 0.00017706, [7] [c_1]: 5.672e-05 [parameter_eliminate]: 6.91001e-06 [updatestate_depend_eliminate]: 8.62e-06 [updatestate_assign_eliminate]: 4.53999e-06 [updatestate_loads_eliminate]: 4.4e-06 [cse]: 5.471e-05 [renormalize]: 3.00002e-07 [remove_dup_value]: 7.72e-05 [tuple_transform]: 0.00012983, [1] [Cycle 1]: 0.00012365, [4] [d_1]: 8.793e-05 [none_parameter_eliminate]: 1.77001e-06 [renormalize]: 2.00002e-07 [switch_simplify]: 1.222e-05 [partial_unused_args_eliminate]: 1.71e-06 [add_recomputation]: 7.143e-05 [cse_after_recomputation]: 3.81e-05, [1] [Cycle 1]: 3.257e-05, [1] [cse]: 2.532e-05 [environ_conv]: 8.47e-06 [swap_dp_allreduce_reducescatter]: 7.48999e-06 [bias_add_comm_swap]: 4.55001e-06 [label_micro_interleaved_index]: 5.72001e-06 [label_fine_grained_interleaved_index]: 3.06999e-06 [merge_cast_opt]: 1.40001e-06 [slice_recompute_activation]: 2.43e-06 [micro_interleaved_order_control]: 2.32001e-06 [assign_add_opt]: 1.40999e-06 [ForceFp32Comm]: 9.20001e-07 [remove_cast_before_assign_add]: 1.00999e-06 [full_micro_interleaved_order_control]: 2.92002e-06 [reorder_send_recv_between_fp_bp]: 2.90002e-06 [comm_op_add_attrs]: 1.03001e-06 [add_comm_op_reuse_tag]: 9.89996e-07 [interleave_split_concat_branches]: 1.35999e-06 [interleave_parallel_branches]: 1.05001e-06 [overlap_opt_shard_in_pipeline]: 5.71e-06 [overlap_opt_shard_grad_in_pipeline]: 1.67999e-06 [control_data_broadcast_order]: 1.88e-05 [grouped_pairwise_exchange_alltoall]: 1.99999e-06 [offloading_packed_experts]: 5.47999e-06 [overlap_recompute_and_grad_model_parallel]: 6.20002e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.27999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.78002e-06 [overlap_recompute_comm]: 2.41998e-06 [overlap_grad_ring_attention]: 5.34998e-06 [overlap_grad_flash_sp]: 2.474e-05 [begin_end_overlap_inline]: 6.40022e-07 [split_matmul_comm_elemetwise]: 2.44001e-06 [split_layernorm_comm]: 1.96e-06 [handle_group_info]: 1.06002e-06 [symbol_engine_optimizer]: 0.00026616, [1] [Cycle 1]: 0.0002607, [6] [build]: 9.121e-05 [elim_shapecalc]: 2.023e-05 [elim_not_effective]: 2.252e-05 [opt_reshape]: 1.098e-05 [fold_const_symbol]: 7.181e-05 [renormalize]: 2.50002e-07 [detach_backward]: 2.86e-06 [pipeline_parallel_scheduler]: 1.55999e-06 [auto_monad_reorder]: 3.299e-05 [get_jit_bprop_graph]: 2.12999e-06 [rewriter_after_jit_bprop_graph]: 7.27002e-06 [opt_after_jit_grad]: 0.00069593 [validate]: 5.904e-05 Sums bootstrap : 0.000840s : 2.15% type_inference : 0.031060s : 79.46% event_method : 0.000019s : 0.05% auto_monad : 0.000061s : 0.16% graph_reusing : 0.000005s : 0.01% inline : 0.000003s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000027s : 0.07% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.01% parallel-infer-symbol : 0.000003s : 0.01% pre_auto_parallel : 0.000037s : 0.09% insert-virtual-dataset : 0.000003s : 0.01% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.01% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000038s : 0.10% optimize.rewriter_before_opt_a : 0.000083s : 0.21% optimize.opt_a.expand_dump_flag : 0.000005s : 0.01% optimize.opt_a.switch_simplify : 0.000043s : 0.11% optimize.opt_a.loop_unroll : 0.000029s : 0.07% optimize.opt_a.a_1 : 0.000914s : 2.34% optimize.opt_a.with_stream_mark : 0.000059s : 0.15% optimize.opt_a.recompute_prepare : 0.000029s : 0.08% optimize.opt_a.updatestate_depend_eliminate : 0.000013s : 0.03% optimize.opt_a.updatestate_assign_eliminate : 0.000011s : 0.03% optimize.opt_a.updatestate_loads_eliminate : 0.000010s : 0.03% optimize.opt_a.parameter_eliminate : 0.000005s : 0.01% optimize.opt_a.a_2 : 0.000274s : 0.70% optimize.opt_a.accelerated_algorithm : 0.000024s : 0.06% optimize.opt_a.shard : 0.000005s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000009s : 0.02% optimize.opt_a.shard_inline : 0.000022s : 0.06% optimize.opt_a.merge_send_recv : 0.000022s : 0.06% optimize.opt_a.auto_parallel : 0.000025s : 0.06% optimize.opt_a.parallel : 0.000045s : 0.12% optimize.opt_a.flash_sp : 0.000018s : 0.05% optimize.opt_a.merge_comm : 0.000012s : 0.03% optimize.opt_a.allreduce_fusion : 0.000010s : 0.03% optimize.opt_a.matmul_add_comm_reduction : 0.000025s : 0.06% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000029s : 0.08% optimize.opt_a.virtual_dataset : 0.000024s : 0.06% optimize.opt_a.get_grad_eliminate_ : 0.000022s : 0.06% optimize.opt_a.virtual_output : 0.000025s : 0.06% optimize.opt_a.merge_forward : 0.000016s : 0.04% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.01% optimize.opt_a.offload_activation : 0.000027s : 0.07% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000050s : 0.13% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.01% optimize.opt_a.before_grad : 0.000039s : 0.10% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000014s : 0.03% optimize.opt_a.meta_fg_expand : 0.000008s : 0.02% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.01% optimize.opt_a.receive_attached : 0.000005s : 0.01% optimize.opt_a.after_resolve : 0.000056s : 0.14% optimize.opt_a.a_after_grad : 0.000041s : 0.10% optimize.opt_a.renormalize : 0.001064s : 2.72% optimize.opt_a.add_forward_monad_depend : 0.000018s : 0.05% optimize.opt_a.auto_monad_grad : 0.000004s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000045s : 0.12% optimize.opt_a.cse : 0.000151s : 0.39% optimize.opt_a.a_3 : 0.000151s : 0.39% optimize.py_interpret_to_execute_after_opt_a : 0.000026s : 0.07% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.01% optimize.rewriter_after_opt_a : 0.000066s : 0.17% optimize.convert_after_rewriter : 0.000010s : 0.03% optimize.order_py_execute_after_rewriter : 0.000008s : 0.02% optimize.mutable_eliminate : 0.000816s : 2.09% optimize.opt_b.b_1 : 0.000243s : 0.62% optimize.opt_b.b_2 : 0.000014s : 0.03% optimize.opt_b.updatestate_depend_eliminate : 0.000012s : 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 : 0.000083s : 0.21% optimize.optimize_parallel_all_gather_comm : 0.000026s : 0.07% optimize.overlap_param_gather : 0.000006s : 0.01% optimize.cconv : 0.000038s : 0.10% optimize.loop_unroll : 0.000562s : 1.44% optimize.opt_after_cconv.c_1 : 0.000057s : 0.15% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.02% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000009s : 0.02% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000005s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000004s : 0.01% optimize.opt_after_cconv.cse : 0.000055s : 0.14% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000077s : 0.20% optimize.tuple_transform.d_1 : 0.000088s : 0.22% 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 : 0.03% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000071s : 0.18% optimize.cse_after_recomputation.cse : 0.000025s : 0.06% optimize.environ_conv : 0.000008s : 0.02% optimize.swap_dp_allreduce_reducescatter : 0.000007s : 0.02% optimize.bias_add_comm_swap : 0.000005s : 0.01% optimize.label_micro_interleaved_index : 0.000006s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.01% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.01% optimize.micro_interleaved_order_control : 0.000002s : 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 : 0.000001s : 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 : 0.000006s : 0.01% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000019s : 0.05% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.01% optimize.offloading_packed_experts : 0.000005s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000006s : 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 : 0.000002s : 0.01% optimize.overlap_grad_ring_attention : 0.000005s : 0.01% optimize.overlap_grad_flash_sp : 0.000025s : 0.06% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.01% optimize.split_layernorm_comm : 0.000002s : 0.01% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000091s : 0.23% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000020s : 0.05% optimize.symbol_engine_optimizer.elim_not_effective : 0.000023s : 0.06% optimize.symbol_engine_optimizer.opt_reshape : 0.000011s : 0.03% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000072s : 0.18% 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 : 0.000033s : 0.08% get_jit_bprop_graph : 0.000002s : 0.01% rewriter_after_jit_bprop_graph : 0.000007s : 0.02% opt_after_jit_grad : 0.000696s : 1.78% validate : 0.000059s : 0.15% Time group info: ------[substitution.] 0.000214 33 2.66% : 0.000006s : 2: substitution.elim_not_effective 4.32% : 0.000009s : 2: substitution.fold_const_symbol 3.70% : 0.000008s : 8: substitution.graph_param_transform 79.02% : 0.000169s : 1: substitution.inline 2.66% : 0.000006s : 4: substitution.j_node_and_user_rematch 2.71% : 0.000006s : 4: substitution.remove_not_recompute_node 4.92% : 0.000011s : 12: substitution.replace_old_param ------[type_inference.] 0.030940 2 97.63% : 0.030205s : 1: type_inference.infer 2.37% : 0.000735s : 1: type_inference.specialize ------[replace.] 0.000039 1 100.00% : 0.000039s : 1: replace.inline ------[match.] 0.000168 1 100.00% : 0.000168s : 1: match.inline ------[predicate.] 0.000246 1879 0.86% : 0.000002s : 17: predicate.accumulaten_eliminater 1.25% : 0.000003s : 8: predicate.ad_related_special_op_eliminate 0.67% : 0.000002s : 16: predicate.addn_check_dump 0.73% : 0.000002s : 17: predicate.addn_zero_filter 0.64% : 0.000002s : 17: predicate.adjust_all_reduce_mul_add 1.87% : 0.000005s : 33: predicate.arithmetic_simplify 0.87% : 0.000002s : 17: predicate.cast_eliminate 0.75% : 0.000002s : 16: predicate.check_bprop_eliminate 0.74% : 0.000002s : 16: predicate.compare_switch_simplify 0.26% : 0.000001s : 8: predicate.const_output_eliminate 0.70% : 0.000002s : 16: predicate.depend_value_elim 0.77% : 0.000002s : 17: predicate.dict_get_item_const_eliminator 0.82% : 0.000002s : 17: predicate.dict_get_item_eliminator 0.75% : 0.000002s : 17: predicate.dict_set_item_eliminator 1.59% : 0.000004s : 16: predicate.dumpgradient_eliminate 0.31% : 0.000001s : 8: predicate.elim_not_effective 0.68% : 0.000002s : 8: predicate.elim_shapecalc_of_broadcastargs 1.21% : 0.000003s : 25: predicate.environ_add_const_eliminate 1.02% : 0.000003s : 25: predicate.environ_get_add_eliminate 1.06% : 0.000003s : 25: predicate.environ_get_depend_swap 1.73% : 0.000004s : 41: predicate.environ_get_eliminate 1.05% : 0.000003s : 25: predicate.environ_get_set_eliminate 0.75% : 0.000002s : 18: predicate.exchange_switch_depend_value 1.99% : 0.000005s : 18: predicate.float_depend_g_call 0.66% : 0.000002s : 16: predicate.float_environ_get_switch 0.99% : 0.000002s : 24: predicate.float_tuple_getitem_switch 0.26% : 0.000001s : 8: predicate.fold_const_symbol 0.98% : 0.000002s : 16: predicate.get_grad_eliminate 0.42% : 0.000001s : 8: predicate.graph_param_transform 0.70% : 0.000002s : 16: predicate.incorporate_call 0.61% : 0.000001s : 16: predicate.incorporate_call_switch 5.38% : 0.000013s : 83: predicate.inline 0.94% : 0.000002s : 16: predicate.inline_without_move 0.48% : 0.000001s : 16: predicate.j_node_and_user_rematch 1.22% : 0.000003s : 16: predicate.less_batch_normalization 1.96% : 0.000005s : 33: predicate.list_to_tuple_eliminator_ 2.03% : 0.000005s : 50: predicate.load_eliminater 1.40% : 0.000003s : 8: predicate.loop_unroll_after_grad 1.29% : 0.000003s : 25: predicate.loop_unroll_before_grad 1.63% : 0.000004s : 33: predicate.make_slice_get_slice_eliminator 0.76% : 0.000002s : 16: predicate.merge_addn 0.70% : 0.000002s : 16: predicate.micro_step_allgather_replace 0.72% : 0.000002s : 16: predicate.mini_step_allgather_replace 0.74% : 0.000002s : 17: predicate.minmaximum_grad 1.75% : 0.000004s : 8: predicate.mutable_eliminate 0.42% : 0.000001s : 8: predicate.opt_reshape 0.43% : 0.000001s : 8: predicate.parallel_virtual_node 3.42% : 0.000008s : 18: predicate.partial_defer_inline 1.13% : 0.000003s : 25: predicate.partial_eliminate 0.81% : 0.000002s : 17: predicate.print_const_string_wrapper 0.86% : 0.000002s : 16: predicate.reduce_all_const_elim 1.10% : 0.000003s : 17: predicate.reduce_eliminate 2.30% : 0.000006s : 50: predicate.redundant_stop_gradient_eliminater 0.83% : 0.000002s : 16: predicate.remove_not_recompute_node 1.36% : 0.000003s : 33: predicate.replace_applicator 0.71% : 0.000002s : 16: predicate.replace_old_param 0.48% : 0.000001s : 8: predicate.reset_defer_inline 1.11% : 0.000003s : 17: predicate.reshape_eliminate 0.83% : 0.000002s : 16: predicate.row_tensor_add_zeros_like 0.48% : 0.000001s : 8: predicate.row_tensor_eliminate 0.92% : 0.000002s : 16: predicate.same_eliminate 0.61% : 0.000001s : 16: predicate.set_cell_output_no_recompute 1.08% : 0.000003s : 16: predicate.shard_identity_eliminate 0.79% : 0.000002s : 16: predicate.special_op_eliminate 0.78% : 0.000002s : 16: predicate.specialize_transform 1.27% : 0.000003s : 16: predicate.split_environ_get_set_with_tuple_value 1.04% : 0.000003s : 16: predicate.stack_unstack_eliminate 0.45% : 0.000001s : 8: predicate.switch_call_monad_eliminater 0.87% : 0.000002s : 18: predicate.switch_defer_inline 1.57% : 0.000004s : 34: predicate.switch_layer_defer_inline 3.31% : 0.000008s : 67: predicate.switch_simplify 0.78% : 0.000002s : 17: predicate.tile_eliminate 0.76% : 0.000002s : 17: predicate.transpose_eliminate 1.49% : 0.000004s : 33: predicate.tuple_list_convert_item_index_to_positive 1.73% : 0.000004s : 33: predicate.tuple_list_get_item_const_eliminator 1.42% : 0.000003s : 33: predicate.tuple_list_get_item_depend_reorder 3.15% : 0.000008s : 49: predicate.tuple_list_get_item_eliminator 1.36% : 0.000003s : 33: predicate.tuple_list_get_set_item_eliminator 2.28% : 0.000006s : 49: predicate.tuple_list_set_item_eliminator 1.65% : 0.000004s : 33: predicate.tuple_to_list_eliminator_ 2.02% : 0.000005s : 50: predicate.updatestate_pure_node_eliminater 2.84% : 0.000007s : 66: predicate.updatestate_useless_node_eliminater 0.46% : 0.000001s : 8: predicate.value_based_eliminate 0.86% : 0.000002s : 16: predicate.virtual_dataset_eliminate 0.80% : 0.000002s : 16: predicate.virtual_output_eliminate 0.35% : 0.000001s : 8: predicate.virtual_view_grad_eliminate 0.43% : 0.000001s : 8: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000789 6 44.60% : 0.000352s : 3: func_graph_cloner_run.FuncGraphClonerGraph 55.40% : 0.000437s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.279014 192 0.00% : 0.000004s : 1: ForceFp32Comm 6.53% : 0.018225s : 1: add_attr 6.52% : 0.018204s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.03% : 0.000077s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.02% : 0.000066s : 1: auto_monad 0.01% : 0.000038s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000008s : 1: bias_add_comm_swap 0.31% : 0.000875s : 1: bootstrap 0.02% : 0.000042s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000023s : 1: control_data_broadcast_order 0.01% : 0.000015s : 1: convert_after_rewriter 0.01% : 0.000042s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000012s : 1: environ_conv 0.01% : 0.000028s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000008s : 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% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000010s : 1: label_micro_interleaved_index 0.20% : 0.000572s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.30% : 0.000831s : 1: mutable_eliminate 0.00% : 0.000009s : 1: offloading_packed_experts 0.01% : 0.000024s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000028s : 1: opt.transform.mutable_eliminate 0.59% : 0.001658s : 78: opt.transform.opt_a 0.02% : 0.000055s : 1: opt.transform.opt_after_cconv 0.02% : 0.000049s : 1: opt.transform.opt_after_jit_grad 0.08% : 0.000221s : 28: opt.transform.opt_b 0.04% : 0.000098s : 2: opt.transform.opt_trans_graph 0.04% : 0.000121s : 4: opt.transform.symbol_engine_opt 35.82% : 0.099933s : 1: opt_a 0.07% : 0.000189s : 1: opt_after_cconv 0.26% : 0.000713s : 1: opt_after_jit_grad 0.15% : 0.000425s : 1: opt_b 36.99% : 0.103210s : 1: optimize 0.01% : 0.000031s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000012s : 1: order_py_execute_after_rewriter 0.01% : 0.000029s : 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% : 0.000006s : 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% : 0.000005s : 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% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000004s : 1: pipeline_split 0.01% : 0.000042s : 1: pre_auto_parallel 0.02% : 0.000042s : 1: py_interpret_to_execute 0.01% : 0.000032s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.03% : 0.000082s : 1: remove_dup_value 0.17% : 0.000471s : 1: renormalize.infer 0.21% : 0.000580s : 1: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000011s : 1: rewriter_after_jit_bprop_graph 0.03% : 0.000073s : 1: rewriter_after_opt_a 0.03% : 0.000088s : 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% : 0.000011s : 1: swap_dp_allreduce_reducescatter 0.10% : 0.000270s : 1: symbol_engine_optimizer 0.05% : 0.000133s : 1: tuple_transform 11.14% : 0.031086s : 1: type_inference .[WARNING] PRE_ACT(170442,ffffbeca8f30,python3.9):2026-01-29-17:42:22.797.076 [mindspore/ccsrc/runtime/memory/mem_pool/abstract_dynamic_mem_pool.cc:1109] FreeIdleMemsByEagerFree] Eager free count : 6, free memory : 2725209088, real free : 2726297600, not free : 0. [hook] pytest_runtest_teardown:test_paged_attention_trapezoidal[1-MASK_DEFAULT-float32] tests/st/infer/ops/test_internal_ops/test_paged_attention.py::test_paged_attention_trapezoidal[1-MASK_DEFAULT-float32],max_mem:2696.0M [WARNING] ME(170442:281473882689328,MainProcess):2026-01-29-17:42:23.713.090 [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 = 0.0207366, [21] [bootstrap]: 0.00078153 [type_inference]: 0.00579371 [event_method]: 1.406e-05 [auto_monad]: 5.983e-05 [graph_reusing]: 5.30001e-06 [inline]: 2.32001e-06 [add_attr]: 0.00462693, [1] [add_attr_with_inline]: 0.00460936, [1] [Cycle 1]: 7.711e-05, [2] [tag_attr]: 2.372e-05 [meta_addattr_fg_expand]: 4.14002e-06 [parallel-infer-symbol]: 3.47997e-06 [pre_auto_parallel]: 4.127e-05 [insert-virtual-dataset]: 2.39001e-06 [parallel-infer-symbol-second]: 6.60017e-07 [dataset_repeat_opt]: 1.92999e-06 [pipeline_split]: 2.15002e-06 [optimize]: 0.00815191, [53] [py_interpret_to_execute]: 3.338e-05 [rewriter_before_opt_a]: 8.315e-05 [opt_a]: 0.00396954, [2] [Cycle 1]: 0.00280023, [45] [expand_dump_flag]: 3.43e-06 [switch_simplify]: 3.064e-05 [loop_unroll]: 1.763e-05 [a_1]: 0.00054234 [with_stream_mark]: 2.922e-05 [recompute_prepare]: 1.555e-05 [updatestate_depend_eliminate]: 6.54001e-06 [updatestate_assign_eliminate]: 6.15002e-06 [updatestate_loads_eliminate]: 4.79e-06 [parameter_eliminate]: 2.42001e-06 [a_2]: 0.00014802 [accelerated_algorithm]: 1.335e-05 [shard]: 2.58e-06 [meta_shard_fg_expand]: 2.91999e-06 [shard_inline]: 1.042e-05 [merge_send_recv]: 1.097e-05 [auto_parallel]: 1.093e-05 [parallel]: 3.105e-05 [flash_sp]: 1.084e-05 [merge_comm]: 6.12999e-06 [allreduce_fusion]: 5.14e-06 [matmul_add_comm_reduction]: 1.214e-05 [allreduce_slice_to_reducescatter]: 7.79983e-07 [virtual_shard_identity]: 1.35e-05 [virtual_dataset]: 1.196e-05 [get_grad_eliminate_]: 1.09e-05 [virtual_output]: 1.103e-05 [merge_forward]: 5.84999e-06 [cell_reuse_recompute_pass]: 2.06e-06 [offload_activation]: 1.383e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.213e-05 [merge_recompute_call_nodes]: 1.43002e-06 [before_grad]: 1.772e-05 [set_forward_comm_id_for_comm_node_pass]: 5.47001e-06 [meta_fg_expand]: 4.03999e-06 [flash_sp_send_recv_attached]: 2.76e-06 [receive_attached]: 3.01999e-06 [after_resolve]: 2.4e-05 [a_after_grad]: 1.928e-05 [renormalize]: 0.00113502 [add_forward_monad_depend]: 1.5e-05 [auto_monad_grad]: 2.26e-06 [auto_monad_eliminator]: 2.529e-05 [cse]: 9.379e-05 [a_3]: 8.809e-05 [Cycle 2]: 0.00115245, [45] [expand_dump_flag]: 3.06001e-06 [switch_simplify]: 1.332e-05 [loop_unroll]: 1.1e-05 [a_1]: 0.00024533 [with_stream_mark]: 2.337e-05 [recompute_prepare]: 1.16e-05 [updatestate_depend_eliminate]: 5.87999e-06 [updatestate_assign_eliminate]: 5.10999e-06 [updatestate_loads_eliminate]: 5.14e-06 [parameter_eliminate]: 3.00998e-06 [a_2]: 0.00012561 [accelerated_algorithm]: 1.119e-05 [shard]: 2.78e-06 [meta_shard_fg_expand]: 4.20999e-06 [shard_inline]: 1.021e-05 [merge_send_recv]: 1.07e-05 [auto_parallel]: 1.265e-05 [parallel]: 9.57999e-06 [flash_sp]: 4.79e-06 [merge_comm]: 6.11e-06 [allreduce_fusion]: 5.27001e-06 [matmul_add_comm_reduction]: 1.233e-05 [allreduce_slice_to_reducescatter]: 5.3001e-07 [virtual_shard_identity]: 1.473e-05 [virtual_dataset]: 9.82999e-06 [get_grad_eliminate_]: 1.025e-05 [virtual_output]: 1.01e-05 [merge_forward]: 7.21999e-06 [cell_reuse_recompute_pass]: 3.29001e-06 [offload_activation]: 1.388e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.546e-05 [merge_recompute_call_nodes]: 1.27999e-06 [before_grad]: 1.46e-05 [set_forward_comm_id_for_comm_node_pass]: 6.18002e-06 [meta_fg_expand]: 3.58999e-06 [flash_sp_send_recv_attached]: 2.06998e-06 [receive_attached]: 2.24001e-06 [after_resolve]: 2.352e-05 [a_after_grad]: 1.66e-05 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 3.37002e-06 [auto_monad_grad]: 2.44999e-06 [auto_monad_eliminator]: 1.803e-05 [cse]: 6.401e-05 [a_3]: 6.648e-05 [py_interpret_to_execute_after_opt_a]: 2.767e-05 [slice_cell_reuse_recomputed_activation]: 2.16e-06 [rewriter_after_opt_a]: 7.579e-05 [convert_after_rewriter]: 1.161e-05 [order_py_execute_after_rewriter]: 7.53999e-06 [mutable_eliminate]: 0.00085859 [opt_b]: 0.0004621, [1] [Cycle 1]: 0.00045057, [7] [b_1]: 0.00025364 [b_2]: 1.422e-05 [updatestate_depend_eliminate]: 1.498e-05 [updatestate_assign_eliminate]: 5.19e-06 [updatestate_loads_eliminate]: 4.72e-06 [renormalize]: 1.02998e-06 [cse]: 0.00010305 [optimize_parallel_all_gather_comm]: 3.175e-05 [overlap_param_gather]: 2.54001e-06 [cconv]: 4.153e-05 [loop_unroll]: 0.00076377 [opt_after_cconv]: 0.00058101, [1] [Cycle 1]: 0.000571, [7] [c_1]: 6.481e-05 [parameter_eliminate]: 7.16999e-06 [updatestate_depend_eliminate]: 1.495e-05 [updatestate_assign_eliminate]: 5.25999e-06 [updatestate_loads_eliminate]: 4.37998e-06 [cse]: 0.00040438 [renormalize]: 9.19972e-07 [remove_dup_value]: 8.736e-05 [tuple_transform]: 0.00018593, [1] [Cycle 1]: 0.00017709, [4] [d_1]: 0.00013182 [none_parameter_eliminate]: 4.4e-06 [renormalize]: 1.40019e-07 [switch_simplify]: 1.302e-05 [partial_unused_args_eliminate]: 3.31001e-06 [add_recomputation]: 0.00011884 [cse_after_recomputation]: 6.387e-05, [1] [Cycle 1]: 5.716e-05, [1] [cse]: 4.635e-05 [environ_conv]: 1.158e-05 [swap_dp_allreduce_reducescatter]: 8.69e-06 [bias_add_comm_swap]: 4.47e-06 [label_micro_interleaved_index]: 9.06002e-06 [label_fine_grained_interleaved_index]: 3.48e-06 [merge_cast_opt]: 1.59e-06 [slice_recompute_activation]: 2.37001e-06 [micro_interleaved_order_control]: 2.75997e-06 [assign_add_opt]: 1.75001e-06 [ForceFp32Comm]: 1.46002e-06 [remove_cast_before_assign_add]: 1.17e-06 [full_micro_interleaved_order_control]: 2.54001e-06 [reorder_send_recv_between_fp_bp]: 2.96001e-06 [comm_op_add_attrs]: 1.07e-06 [add_comm_op_reuse_tag]: 1.27e-06 [interleave_split_concat_branches]: 1.27999e-06 [interleave_parallel_branches]: 1.17e-06 [overlap_opt_shard_in_pipeline]: 1.81e-06 [overlap_opt_shard_grad_in_pipeline]: 2.24999e-06 [control_data_broadcast_order]: 2.394e-05 [grouped_pairwise_exchange_alltoall]: 1.60001e-06 [offloading_packed_experts]: 5.76e-06 [overlap_recompute_and_grad_model_parallel]: 6.66999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.58002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.43002e-06 [overlap_recompute_comm]: 2.42001e-06 [overlap_grad_ring_attention]: 5.49e-06 [overlap_grad_flash_sp]: 2.951e-05 [begin_end_overlap_inline]: 5.3001e-07 [split_matmul_comm_elemetwise]: 2.43e-06 [split_layernorm_comm]: 2.46e-06 [handle_group_info]: 1.20001e-06 [symbol_engine_optimizer]: 0.00026165, [1] [Cycle 1]: 0.000255, [6] [build]: 0.0001063 [elim_shapecalc]: 2.471e-05 [elim_not_effective]: 2.599e-05 [opt_reshape]: 1.276e-05 [fold_const_symbol]: 2.253e-05 [renormalize]: 3.29979e-07 [detach_backward]: 2.99999e-06 [pipeline_parallel_scheduler]: 1.84e-06 [auto_monad_reorder]: 3.957e-05 [get_jit_bprop_graph]: 2.30002e-06 [rewriter_after_jit_bprop_graph]: 7.30998e-06 [opt_after_jit_grad]: 0.00093649 [validate]: 7.105e-05 Sums bootstrap : 0.000782s : 5.32% type_inference : 0.005794s : 39.42% event_method : 0.000014s : 0.10% auto_monad : 0.000060s : 0.41% graph_reusing : 0.000005s : 0.04% inline : 0.000002s : 0.02% add_attr.add_attr_with_inline.tag_attr : 0.000024s : 0.16% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.03% parallel-infer-symbol : 0.000003s : 0.02% pre_auto_parallel : 0.000041s : 0.28% insert-virtual-dataset : 0.000002s : 0.02% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.01% pipeline_split : 0.000002s : 0.01% optimize.py_interpret_to_execute : 0.000033s : 0.23% optimize.rewriter_before_opt_a : 0.000083s : 0.57% optimize.opt_a.expand_dump_flag : 0.000006s : 0.04% optimize.opt_a.switch_simplify : 0.000044s : 0.30% optimize.opt_a.loop_unroll : 0.000029s : 0.19% optimize.opt_a.a_1 : 0.000788s : 5.36% optimize.opt_a.with_stream_mark : 0.000053s : 0.36% optimize.opt_a.recompute_prepare : 0.000027s : 0.18% optimize.opt_a.updatestate_depend_eliminate : 0.000012s : 0.08% optimize.opt_a.updatestate_assign_eliminate : 0.000011s : 0.08% optimize.opt_a.updatestate_loads_eliminate : 0.000010s : 0.07% optimize.opt_a.parameter_eliminate : 0.000005s : 0.04% optimize.opt_a.a_2 : 0.000274s : 1.86% optimize.opt_a.accelerated_algorithm : 0.000025s : 0.17% optimize.opt_a.shard : 0.000005s : 0.04% optimize.opt_a.meta_shard_fg_expand : 0.000007s : 0.05% optimize.opt_a.shard_inline : 0.000021s : 0.14% optimize.opt_a.merge_send_recv : 0.000022s : 0.15% optimize.opt_a.auto_parallel : 0.000024s : 0.16% optimize.opt_a.parallel : 0.000041s : 0.28% optimize.opt_a.flash_sp : 0.000016s : 0.11% optimize.opt_a.merge_comm : 0.000012s : 0.08% optimize.opt_a.allreduce_fusion : 0.000010s : 0.07% optimize.opt_a.matmul_add_comm_reduction : 0.000024s : 0.17% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000028s : 0.19% optimize.opt_a.virtual_dataset : 0.000022s : 0.15% optimize.opt_a.get_grad_eliminate_ : 0.000021s : 0.14% optimize.opt_a.virtual_output : 0.000021s : 0.14% optimize.opt_a.merge_forward : 0.000013s : 0.09% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.04% optimize.opt_a.offload_activation : 0.000028s : 0.19% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000048s : 0.32% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.02% optimize.opt_a.before_grad : 0.000032s : 0.22% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000012s : 0.08% optimize.opt_a.meta_fg_expand : 0.000008s : 0.05% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.03% optimize.opt_a.receive_attached : 0.000005s : 0.04% optimize.opt_a.after_resolve : 0.000048s : 0.32% optimize.opt_a.a_after_grad : 0.000036s : 0.24% optimize.opt_a.renormalize : 0.001135s : 7.72% optimize.opt_a.add_forward_monad_depend : 0.000018s : 0.12% optimize.opt_a.auto_monad_grad : 0.000005s : 0.03% optimize.opt_a.auto_monad_eliminator : 0.000043s : 0.29% optimize.opt_a.cse : 0.000158s : 1.07% optimize.opt_a.a_3 : 0.000155s : 1.05% optimize.py_interpret_to_execute_after_opt_a : 0.000028s : 0.19% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.01% optimize.rewriter_after_opt_a : 0.000076s : 0.52% optimize.convert_after_rewriter : 0.000012s : 0.08% optimize.order_py_execute_after_rewriter : 0.000008s : 0.05% optimize.mutable_eliminate : 0.000859s : 5.84% optimize.opt_b.b_1 : 0.000254s : 1.73% optimize.opt_b.b_2 : 0.000014s : 0.10% optimize.opt_b.updatestate_depend_eliminate : 0.000015s : 0.10% optimize.opt_b.updatestate_assign_eliminate : 0.000005s : 0.04% optimize.opt_b.updatestate_loads_eliminate : 0.000005s : 0.03% optimize.opt_b.renormalize : 0.000001s : 0.01% optimize.opt_b.cse : 0.000103s : 0.70% optimize.optimize_parallel_all_gather_comm : 0.000032s : 0.22% optimize.overlap_param_gather : 0.000003s : 0.02% optimize.cconv : 0.000042s : 0.28% optimize.loop_unroll : 0.000764s : 5.20% optimize.opt_after_cconv.c_1 : 0.000065s : 0.44% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.05% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000015s : 0.10% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000005s : 0.04% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000004s : 0.03% optimize.opt_after_cconv.cse : 0.000404s : 2.75% optimize.opt_after_cconv.renormalize : 0.000001s : 0.01% optimize.remove_dup_value : 0.000087s : 0.59% optimize.tuple_transform.d_1 : 0.000132s : 0.90% optimize.tuple_transform.none_parameter_eliminate : 0.000004s : 0.03% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000013s : 0.09% optimize.partial_unused_args_eliminate : 0.000003s : 0.02% optimize.add_recomputation : 0.000119s : 0.81% optimize.cse_after_recomputation.cse : 0.000046s : 0.32% optimize.environ_conv : 0.000012s : 0.08% optimize.swap_dp_allreduce_reducescatter : 0.000009s : 0.06% optimize.bias_add_comm_swap : 0.000004s : 0.03% optimize.label_micro_interleaved_index : 0.000009s : 0.06% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.02% optimize.merge_cast_opt : 0.000002s : 0.01% optimize.slice_recompute_activation : 0.000002s : 0.02% optimize.micro_interleaved_order_control : 0.000003s : 0.02% optimize.assign_add_opt : 0.000002s : 0.01% optimize.ForceFp32Comm : 0.000001s : 0.01% optimize.remove_cast_before_assign_add : 0.000001s : 0.01% optimize.full_micro_interleaved_order_control : 0.000003s : 0.02% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.02% optimize.comm_op_add_attrs : 0.000001s : 0.01% optimize.add_comm_op_reuse_tag : 0.000001s : 0.01% optimize.interleave_split_concat_branches : 0.000001s : 0.01% optimize.interleave_parallel_branches : 0.000001s : 0.01% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.01% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.02% optimize.control_data_broadcast_order : 0.000024s : 0.16% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.01% optimize.offloading_packed_experts : 0.000006s : 0.04% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 0.05% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.01% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.01% optimize.overlap_recompute_comm : 0.000002s : 0.02% optimize.overlap_grad_ring_attention : 0.000005s : 0.04% optimize.overlap_grad_flash_sp : 0.000030s : 0.20% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.02% optimize.split_layernorm_comm : 0.000002s : 0.02% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000106s : 0.72% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000025s : 0.17% optimize.symbol_engine_optimizer.elim_not_effective : 0.000026s : 0.18% optimize.symbol_engine_optimizer.opt_reshape : 0.000013s : 0.09% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000023s : 0.15% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.02% pipeline_parallel_scheduler : 0.000002s : 0.01% auto_monad_reorder : 0.000040s : 0.27% get_jit_bprop_graph : 0.000002s : 0.02% rewriter_after_jit_bprop_graph : 0.000007s : 0.05% opt_after_jit_grad : 0.000936s : 6.37% validate : 0.000071s : 0.48% Time group info: ------[substitution.] 0.000184 33 2.91% : 0.000005s : 2: substitution.elim_not_effective 4.98% : 0.000009s : 2: substitution.fold_const_symbol 5.06% : 0.000009s : 8: substitution.graph_param_transform 75.58% : 0.000139s : 1: substitution.inline 2.75% : 0.000005s : 4: substitution.j_node_and_user_rematch 2.79% : 0.000005s : 4: substitution.remove_not_recompute_node 5.93% : 0.000011s : 12: substitution.replace_old_param ------[type_inference.] 0.005689 2 87.99% : 0.005006s : 1: type_inference.infer 12.01% : 0.000683s : 1: type_inference.specialize ------[replace.] 0.000024 1 100.00% : 0.000024s : 1: replace.inline ------[match.] 0.000138 1 100.00% : 0.000138s : 1: match.inline ------[predicate.] 0.000257 1879 0.91% : 0.000002s : 17: predicate.accumulaten_eliminater 1.34% : 0.000003s : 8: predicate.ad_related_special_op_eliminate 0.69% : 0.000002s : 16: predicate.addn_check_dump 0.98% : 0.000003s : 17: predicate.addn_zero_filter 0.67% : 0.000002s : 17: predicate.adjust_all_reduce_mul_add 2.21% : 0.000006s : 33: predicate.arithmetic_simplify 0.85% : 0.000002s : 17: predicate.cast_eliminate 0.74% : 0.000002s : 16: predicate.check_bprop_eliminate 0.67% : 0.000002s : 16: predicate.compare_switch_simplify 0.27% : 0.000001s : 8: predicate.const_output_eliminate 0.80% : 0.000002s : 16: predicate.depend_value_elim 0.75% : 0.000002s : 17: predicate.dict_get_item_const_eliminator 0.83% : 0.000002s : 17: predicate.dict_get_item_eliminator 0.79% : 0.000002s : 17: predicate.dict_set_item_eliminator 1.75% : 0.000005s : 16: predicate.dumpgradient_eliminate 0.30% : 0.000001s : 8: predicate.elim_not_effective 0.64% : 0.000002s : 8: predicate.elim_shapecalc_of_broadcastargs 1.34% : 0.000003s : 25: predicate.environ_add_const_eliminate 1.00% : 0.000003s : 25: predicate.environ_get_add_eliminate 0.99% : 0.000003s : 25: predicate.environ_get_depend_swap 1.78% : 0.000005s : 41: predicate.environ_get_eliminate 0.99% : 0.000003s : 25: predicate.environ_get_set_eliminate 0.74% : 0.000002s : 18: predicate.exchange_switch_depend_value 1.48% : 0.000004s : 18: predicate.float_depend_g_call 0.65% : 0.000002s : 16: predicate.float_environ_get_switch 1.01% : 0.000003s : 24: predicate.float_tuple_getitem_switch 0.26% : 0.000001s : 8: predicate.fold_const_symbol 0.92% : 0.000002s : 16: predicate.get_grad_eliminate 0.40% : 0.000001s : 8: predicate.graph_param_transform 0.65% : 0.000002s : 16: predicate.incorporate_call 0.58% : 0.000001s : 16: predicate.incorporate_call_switch 5.47% : 0.000014s : 83: predicate.inline 0.92% : 0.000002s : 16: predicate.inline_without_move 0.43% : 0.000001s : 16: predicate.j_node_and_user_rematch 1.11% : 0.000003s : 16: predicate.less_batch_normalization 2.12% : 0.000005s : 33: predicate.list_to_tuple_eliminator_ 2.34% : 0.000006s : 50: predicate.load_eliminater 2.02% : 0.000005s : 8: predicate.loop_unroll_after_grad 1.18% : 0.000003s : 25: predicate.loop_unroll_before_grad 1.80% : 0.000005s : 33: predicate.make_slice_get_slice_eliminator 0.69% : 0.000002s : 16: predicate.merge_addn 0.72% : 0.000002s : 16: predicate.micro_step_allgather_replace 0.75% : 0.000002s : 16: predicate.mini_step_allgather_replace 0.65% : 0.000002s : 17: predicate.minmaximum_grad 1.58% : 0.000004s : 8: predicate.mutable_eliminate 0.44% : 0.000001s : 8: predicate.opt_reshape 0.45% : 0.000001s : 8: predicate.parallel_virtual_node 1.09% : 0.000003s : 18: predicate.partial_defer_inline 1.07% : 0.000003s : 25: predicate.partial_eliminate 0.74% : 0.000002s : 17: predicate.print_const_string_wrapper 0.69% : 0.000002s : 16: predicate.reduce_all_const_elim 1.02% : 0.000003s : 17: predicate.reduce_eliminate 2.09% : 0.000005s : 50: predicate.redundant_stop_gradient_eliminater 0.76% : 0.000002s : 16: predicate.remove_not_recompute_node 1.33% : 0.000003s : 33: predicate.replace_applicator 0.69% : 0.000002s : 16: predicate.replace_old_param 0.48% : 0.000001s : 8: predicate.reset_defer_inline 0.77% : 0.000002s : 17: predicate.reshape_eliminate 0.71% : 0.000002s : 16: predicate.row_tensor_add_zeros_like 0.43% : 0.000001s : 8: predicate.row_tensor_eliminate 1.00% : 0.000003s : 16: predicate.same_eliminate 0.54% : 0.000001s : 16: predicate.set_cell_output_no_recompute 0.99% : 0.000003s : 16: predicate.shard_identity_eliminate 0.86% : 0.000002s : 16: predicate.special_op_eliminate 0.74% : 0.000002s : 16: predicate.specialize_transform 1.16% : 0.000003s : 16: predicate.split_environ_get_set_with_tuple_value 0.96% : 0.000002s : 16: predicate.stack_unstack_eliminate 0.53% : 0.000001s : 8: predicate.switch_call_monad_eliminater 0.76% : 0.000002s : 18: predicate.switch_defer_inline 1.43% : 0.000004s : 34: predicate.switch_layer_defer_inline 3.58% : 0.000009s : 67: predicate.switch_simplify 0.84% : 0.000002s : 17: predicate.tile_eliminate 0.75% : 0.000002s : 17: predicate.transpose_eliminate 1.59% : 0.000004s : 33: predicate.tuple_list_convert_item_index_to_positive 3.74% : 0.000010s : 33: predicate.tuple_list_get_item_const_eliminator 1.53% : 0.000004s : 33: predicate.tuple_list_get_item_depend_reorder 3.33% : 0.000009s : 49: predicate.tuple_list_get_item_eliminator 1.53% : 0.000004s : 33: predicate.tuple_list_get_set_item_eliminator 2.34% : 0.000006s : 49: predicate.tuple_list_set_item_eliminator 1.55% : 0.000004s : 33: predicate.tuple_to_list_eliminator_ 2.01% : 0.000005s : 50: predicate.updatestate_pure_node_eliminater 2.91% : 0.000007s : 66: predicate.updatestate_useless_node_eliminater 0.48% : 0.000001s : 8: predicate.value_based_eliminate 0.77% : 0.000002s : 16: predicate.virtual_dataset_eliminate 0.79% : 0.000002s : 16: predicate.virtual_output_eliminate 0.31% : 0.000001s : 8: predicate.virtual_view_grad_eliminate 0.43% : 0.000001s : 8: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000885 6 41.72% : 0.000369s : 3: func_graph_cloner_run.FuncGraphClonerGraph 58.28% : 0.000516s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.036613 192 0.01% : 0.000005s : 1: ForceFp32Comm 12.66% : 0.004634s : 1: add_attr 12.60% : 0.004615s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.35% : 0.000127s : 1: add_recomputation 0.01% : 0.000005s : 1: assign_add_opt 0.18% : 0.000066s : 1: auto_monad 0.13% : 0.000047s : 1: auto_monad_reorder 0.01% : 0.000004s : 1: begin_end_overlap_inline 0.02% : 0.000008s : 1: bias_add_comm_swap 2.22% : 0.000813s : 1: bootstrap 0.13% : 0.000047s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.08% : 0.000030s : 1: control_data_broadcast_order 0.05% : 0.000017s : 1: convert_after_rewriter 0.19% : 0.000068s : 1: cse_after_recomputation 0.01% : 0.000005s : 1: dataset_repeat_opt 0.02% : 0.000006s : 1: detach_backward 0.04% : 0.000016s : 1: environ_conv 0.06% : 0.000021s : 1: event_method 0.02% : 0.000006s : 1: full_micro_interleaved_order_control 0.02% : 0.000006s : 1: get_jit_bprop_graph 0.03% : 0.000010s : 1: graph_reusing 0.01% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000005s : 1: handle_group_info 0.02% : 0.000006s : 1: inline 0.02% : 0.000006s : 1: insert-virtual-dataset 0.01% : 0.000004s : 1: interleave_parallel_branches 0.01% : 0.000005s : 1: interleave_split_concat_branches 0.02% : 0.000007s : 1: label_fine_grained_interleaved_index 0.04% : 0.000013s : 1: label_micro_interleaved_index 2.15% : 0.000788s : 1: loop_unroll 0.01% : 0.000005s : 1: merge_cast_opt 0.02% : 0.000006s : 1: micro_interleaved_order_control 2.41% : 0.000882s : 1: mutable_eliminate 0.03% : 0.000010s : 1: offloading_packed_experts 0.10% : 0.000038s : 1: opt.transform.loop_unroll_optimizer 0.10% : 0.000035s : 1: opt.transform.mutable_eliminate 4.12% : 0.001510s : 78: opt.transform.opt_a 0.17% : 0.000063s : 1: opt.transform.opt_after_cconv 0.16% : 0.000059s : 1: opt.transform.opt_after_jit_grad 0.62% : 0.000229s : 28: opt.transform.opt_b 0.38% : 0.000140s : 2: opt.transform.opt_trans_graph 0.22% : 0.000081s : 4: opt.transform.symbol_engine_opt 10.85% : 0.003974s : 1: opt_a 1.61% : 0.000588s : 1: opt_after_cconv 2.62% : 0.000958s : 1: opt_after_jit_grad 1.28% : 0.000468s : 1: opt_b 22.28% : 0.008159s : 1: optimize 0.10% : 0.000038s : 1: optimize_parallel_all_gather_comm 0.03% : 0.000012s : 1: order_py_execute_after_rewriter 0.10% : 0.000035s : 1: overlap_grad_flash_sp 0.01% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.03% : 0.000010s : 1: overlap_grad_ring_attention 0.02% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.01% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.02% : 0.000006s : 1: overlap_param_gather 0.01% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.03% : 0.000010s : 1: overlap_recompute_and_grad_model_parallel 0.02% : 0.000007s : 1: overlap_recompute_comm 0.02% : 0.000008s : 1: parallel-infer-symbol 0.01% : 0.000004s : 1: parallel-infer-symbol-second 0.02% : 0.000008s : 1: partial_unused_args_eliminate 0.02% : 0.000007s : 1: pipeline_parallel_scheduler 0.01% : 0.000005s : 1: pipeline_split 0.13% : 0.000046s : 1: pre_auto_parallel 0.11% : 0.000039s : 1: py_interpret_to_execute 0.09% : 0.000034s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000005s : 1: remove_cast_before_assign_add 0.25% : 0.000093s : 1: remove_dup_value 1.27% : 0.000463s : 1: renormalize.infer 1.80% : 0.000659s : 1: renormalize.specialize 0.02% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.03% : 0.000011s : 1: rewriter_after_jit_bprop_graph 0.23% : 0.000085s : 1: rewriter_after_opt_a 0.24% : 0.000089s : 1: rewriter_before_opt_a 0.02% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.02% : 0.000006s : 1: slice_recompute_activation 0.01% : 0.000005s : 1: split_layernorm_comm 0.02% : 0.000006s : 1: split_matmul_comm_elemetwise 0.04% : 0.000013s : 1: swap_dp_allreduce_reducescatter 0.73% : 0.000266s : 1: symbol_engine_optimizer 0.52% : 0.000189s : 1: tuple_transform 15.88% : 0.005815s : 1: type_inference .[WARNING] PRE_ACT(170442,ffffbeca8f30,python3.9):2026-01-29-17:42:39.076.171 [mindspore/ccsrc/runtime/memory/mem_pool/abstract_dynamic_mem_pool.cc:1109] FreeIdleMemsByEagerFree] Eager free count : 7, free memory : 1383032320, real free : 1384120320, not free : 0. [hook] pytest_runtest_teardown:test_paged_attention_trapezoidal[1-TRAPEZOIDAL-float16] tests/st/infer/ops/test_internal_ops/test_paged_attention.py::test_paged_attention_trapezoidal[1-TRAPEZOIDAL-float16],max_mem:2696.0M [WARNING] ME(170442:281473882689328,MainProcess):2026-01-29-17:42:39.623.967 [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 = 0.0431349, [21] [bootstrap]: 0.00072998 [type_inference]: 0.0300504 [event_method]: 1.433e-05 [auto_monad]: 5.189e-05 [graph_reusing]: 4.45e-06 [inline]: 2.26e-06 [add_attr]: 0.00495831, [1] [add_attr_with_inline]: 0.00494534, [1] [Cycle 1]: 6.359e-05, [2] [tag_attr]: 1.866e-05 [meta_addattr_fg_expand]: 3.36999e-06 [parallel-infer-symbol]: 3.53e-06 [pre_auto_parallel]: 2.873e-05 [insert-virtual-dataset]: 2.40002e-06 [parallel-infer-symbol-second]: 7.2e-07 [dataset_repeat_opt]: 1.90001e-06 [pipeline_split]: 1.76e-06 [optimize]: 0.00647368, [53] [py_interpret_to_execute]: 2.595e-05 [rewriter_before_opt_a]: 8.279e-05 [opt_a]: 0.00336432, [2] [Cycle 1]: 0.002346, [45] [expand_dump_flag]: 3.14999e-06 [switch_simplify]: 2.934e-05 [loop_unroll]: 1.756e-05 [a_1]: 0.00048164 [with_stream_mark]: 1.892e-05 [recompute_prepare]: 1.366e-05 [updatestate_depend_eliminate]: 5.95002e-06 [updatestate_assign_eliminate]: 5.88002e-06 [updatestate_loads_eliminate]: 5.59998e-06 [parameter_eliminate]: 1.86e-06 [a_2]: 0.00017596 [accelerated_algorithm]: 1.257e-05 [shard]: 2.53998e-06 [meta_shard_fg_expand]: 2.51e-06 [shard_inline]: 1.054e-05 [merge_send_recv]: 1.119e-05 [auto_parallel]: 8.92999e-06 [parallel]: 2.934e-05 [flash_sp]: 1.029e-05 [merge_comm]: 5.62999e-06 [allreduce_fusion]: 5.57999e-06 [matmul_add_comm_reduction]: 1.136e-05 [allreduce_slice_to_reducescatter]: 6.39993e-07 [virtual_shard_identity]: 1.27e-05 [virtual_dataset]: 1.132e-05 [get_grad_eliminate_]: 1.122e-05 [virtual_output]: 1.156e-05 [merge_forward]: 5.96e-06 [cell_reuse_recompute_pass]: 2.04e-06 [offload_activation]: 1.188e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.97e-05 [merge_recompute_call_nodes]: 1.37999e-06 [before_grad]: 1.571e-05 [set_forward_comm_id_for_comm_node_pass]: 5.87999e-06 [meta_fg_expand]: 3.25e-06 [flash_sp_send_recv_attached]: 2.39999e-06 [receive_attached]: 2.83e-06 [after_resolve]: 2.119e-05 [a_after_grad]: 1.766e-05 [renormalize]: 0.00084159 [add_forward_monad_depend]: 9.36002e-06 [auto_monad_grad]: 2.61e-06 [auto_monad_eliminator]: 1.835e-05 [cse]: 7.394e-05 [a_3]: 7.447e-05 [Cycle 2]: 0.00100281, [45] [expand_dump_flag]: 1.27e-06 [switch_simplify]: 1.198e-05 [loop_unroll]: 9.85002e-06 [a_1]: 0.0002367 [with_stream_mark]: 1.571e-05 [recompute_prepare]: 1.014e-05 [updatestate_depend_eliminate]: 5.32999e-06 [updatestate_assign_eliminate]: 4.57e-06 [updatestate_loads_eliminate]: 4.48999e-06 [parameter_eliminate]: 1.34e-06 [a_2]: 0.00012252 [accelerated_algorithm]: 1.014e-05 [shard]: 1.66998e-06 [meta_shard_fg_expand]: 1.84e-06 [shard_inline]: 9.42999e-06 [merge_send_recv]: 1.161e-05 [auto_parallel]: 8.39002e-06 [parallel]: 6.49999e-06 [flash_sp]: 3.58999e-06 [merge_comm]: 5.63002e-06 [allreduce_fusion]: 4.69002e-06 [matmul_add_comm_reduction]: 8.87999e-06 [allreduce_slice_to_reducescatter]: 4.59986e-07 [virtual_shard_identity]: 1.089e-05 [virtual_dataset]: 9.82001e-06 [get_grad_eliminate_]: 9.44998e-06 [virtual_output]: 9.64e-06 [merge_forward]: 4.84e-06 [cell_reuse_recompute_pass]: 1.44e-06 [offload_activation]: 1.023e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.865e-05 [merge_recompute_call_nodes]: 1.38002e-06 [before_grad]: 1.341e-05 [set_forward_comm_id_for_comm_node_pass]: 5.07e-06 [meta_fg_expand]: 3.08e-06 [flash_sp_send_recv_attached]: 1.08001e-06 [receive_attached]: 1.44998e-06 [after_resolve]: 1.927e-05 [a_after_grad]: 1.412e-05 [renormalize]: 8.00064e-08 [add_forward_monad_depend]: 1.59e-06 [auto_monad_grad]: 1.44998e-06 [auto_monad_eliminator]: 1.078e-05 [cse]: 4.742e-05 [a_3]: 6.154e-05 [py_interpret_to_execute_after_opt_a]: 2.458e-05 [slice_cell_reuse_recomputed_activation]: 2.52001e-06 [rewriter_after_opt_a]: 8.295e-05 [convert_after_rewriter]: 9.39e-06 [order_py_execute_after_rewriter]: 1.062e-05 [mutable_eliminate]: 0.00070079 [opt_b]: 0.00038987, [1] [Cycle 1]: 0.00038165, [7] [b_1]: 0.00023714 [b_2]: 1.38e-05 [updatestate_depend_eliminate]: 8.28999e-06 [updatestate_assign_eliminate]: 4.75999e-06 [updatestate_loads_eliminate]: 4.50001e-06 [renormalize]: 6.00005e-07 [cse]: 6.391e-05 [optimize_parallel_all_gather_comm]: 2.588e-05 [overlap_param_gather]: 2.17001e-06 [cconv]: 3.034e-05 [loop_unroll]: 0.00061825 [opt_after_cconv]: 0.00018877, [1] [Cycle 1]: 0.00018079, [7] [c_1]: 5.689e-05 [parameter_eliminate]: 3.46999e-06 [updatestate_depend_eliminate]: 8e-06 [updatestate_assign_eliminate]: 4.57e-06 [updatestate_loads_eliminate]: 4.3e-06 [cse]: 4.975e-05 [renormalize]: 4.30009e-07 [remove_dup_value]: 6.857e-05 [tuple_transform]: 0.00014226, [1] [Cycle 1]: 0.00013617, [4] [d_1]: 0.0001008 [none_parameter_eliminate]: 2.22001e-06 [renormalize]: 1.70025e-07 [switch_simplify]: 1.11e-05 [partial_unused_args_eliminate]: 2.24001e-06 [add_recomputation]: 6.674e-05 [cse_after_recomputation]: 3.848e-05, [1] [Cycle 1]: 3.353e-05, [1] [cse]: 2.653e-05 [environ_conv]: 7.06999e-06 [swap_dp_allreduce_reducescatter]: 7.00002e-06 [bias_add_comm_swap]: 3.4e-06 [label_micro_interleaved_index]: 5.41998e-06 [label_fine_grained_interleaved_index]: 2.91999e-06 [merge_cast_opt]: 1.24e-06 [slice_recompute_activation]: 1.99e-06 [micro_interleaved_order_control]: 2.16998e-06 [assign_add_opt]: 1.30999e-06 [ForceFp32Comm]: 8.30012e-07 [remove_cast_before_assign_add]: 1.15001e-06 [full_micro_interleaved_order_control]: 2.39999e-06 [reorder_send_recv_between_fp_bp]: 2.56e-06 [comm_op_add_attrs]: 1.05001e-06 [add_comm_op_reuse_tag]: 9.50007e-07 [interleave_split_concat_branches]: 1.22e-06 [interleave_parallel_branches]: 1.14e-06 [overlap_opt_shard_in_pipeline]: 1.24003e-06 [overlap_opt_shard_grad_in_pipeline]: 2.02999e-06 [control_data_broadcast_order]: 1.835e-05 [grouped_pairwise_exchange_alltoall]: 1.74998e-06 [offloading_packed_experts]: 5.23002e-06 [overlap_recompute_and_grad_model_parallel]: 5.87001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.27999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.55999e-06 [overlap_recompute_comm]: 2.10002e-06 [overlap_grad_ring_attention]: 5.37999e-06 [overlap_grad_flash_sp]: 2.189e-05 [begin_end_overlap_inline]: 5.3001e-07 [split_matmul_comm_elemetwise]: 2.31998e-06 [split_layernorm_comm]: 1.78002e-06 [handle_group_info]: 1.03001e-06 [symbol_engine_optimizer]: 0.00020076, [1] [Cycle 1]: 0.00019568, [6] [build]: 7.933e-05 [elim_shapecalc]: 1.724e-05 [elim_not_effective]: 2.392e-05 [opt_reshape]: 1.11e-05 [fold_const_symbol]: 1.908e-05 [renormalize]: 2.3999e-07 [detach_backward]: 2.06e-06 [pipeline_parallel_scheduler]: 1.54e-06 [auto_monad_reorder]: 2.897e-05 [get_jit_bprop_graph]: 1.55999e-06 [rewriter_after_jit_bprop_graph]: 5.07999e-06 [opt_after_jit_grad]: 0.00054329 [validate]: 5.172e-05 Sums bootstrap : 0.000730s : 1.97% type_inference : 0.030050s : 81.28% event_method : 0.000014s : 0.04% auto_monad : 0.000052s : 0.14% graph_reusing : 0.000004s : 0.01% inline : 0.000002s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000019s : 0.05% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.01% parallel-infer-symbol : 0.000004s : 0.01% pre_auto_parallel : 0.000029s : 0.08% insert-virtual-dataset : 0.000002s : 0.01% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.01% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000026s : 0.07% optimize.rewriter_before_opt_a : 0.000083s : 0.22% optimize.opt_a.expand_dump_flag : 0.000004s : 0.01% optimize.opt_a.switch_simplify : 0.000041s : 0.11% optimize.opt_a.loop_unroll : 0.000027s : 0.07% optimize.opt_a.a_1 : 0.000718s : 1.94% optimize.opt_a.with_stream_mark : 0.000035s : 0.09% optimize.opt_a.recompute_prepare : 0.000024s : 0.06% optimize.opt_a.updatestate_depend_eliminate : 0.000011s : 0.03% optimize.opt_a.updatestate_assign_eliminate : 0.000010s : 0.03% optimize.opt_a.updatestate_loads_eliminate : 0.000010s : 0.03% optimize.opt_a.parameter_eliminate : 0.000003s : 0.01% optimize.opt_a.a_2 : 0.000298s : 0.81% optimize.opt_a.accelerated_algorithm : 0.000023s : 0.06% optimize.opt_a.shard : 0.000004s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.01% optimize.opt_a.shard_inline : 0.000020s : 0.05% optimize.opt_a.merge_send_recv : 0.000023s : 0.06% optimize.opt_a.auto_parallel : 0.000017s : 0.05% optimize.opt_a.parallel : 0.000036s : 0.10% optimize.opt_a.flash_sp : 0.000014s : 0.04% optimize.opt_a.merge_comm : 0.000011s : 0.03% optimize.opt_a.allreduce_fusion : 0.000010s : 0.03% optimize.opt_a.matmul_add_comm_reduction : 0.000020s : 0.05% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000024s : 0.06% optimize.opt_a.virtual_dataset : 0.000021s : 0.06% optimize.opt_a.get_grad_eliminate_ : 0.000021s : 0.06% optimize.opt_a.virtual_output : 0.000021s : 0.06% optimize.opt_a.merge_forward : 0.000011s : 0.03% optimize.opt_a.cell_reuse_recompute_pass : 0.000003s : 0.01% optimize.opt_a.offload_activation : 0.000022s : 0.06% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000038s : 0.10% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.01% optimize.opt_a.before_grad : 0.000029s : 0.08% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000011s : 0.03% optimize.opt_a.meta_fg_expand : 0.000006s : 0.02% optimize.opt_a.flash_sp_send_recv_attached : 0.000003s : 0.01% optimize.opt_a.receive_attached : 0.000004s : 0.01% optimize.opt_a.after_resolve : 0.000040s : 0.11% optimize.opt_a.a_after_grad : 0.000032s : 0.09% optimize.opt_a.renormalize : 0.000842s : 2.28% optimize.opt_a.add_forward_monad_depend : 0.000011s : 0.03% optimize.opt_a.auto_monad_grad : 0.000004s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000029s : 0.08% optimize.opt_a.cse : 0.000121s : 0.33% optimize.opt_a.a_3 : 0.000136s : 0.37% optimize.py_interpret_to_execute_after_opt_a : 0.000025s : 0.07% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.01% optimize.rewriter_after_opt_a : 0.000083s : 0.22% optimize.convert_after_rewriter : 0.000009s : 0.03% optimize.order_py_execute_after_rewriter : 0.000011s : 0.03% optimize.mutable_eliminate : 0.000701s : 1.90% optimize.opt_b.b_1 : 0.000237s : 0.64% optimize.opt_b.b_2 : 0.000014s : 0.04% optimize.opt_b.updatestate_depend_eliminate : 0.000008s : 0.02% optimize.opt_b.updatestate_assign_eliminate : 0.000005s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000005s : 0.01% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000064s : 0.17% optimize.optimize_parallel_all_gather_comm : 0.000026s : 0.07% optimize.overlap_param_gather : 0.000002s : 0.01% optimize.cconv : 0.000030s : 0.08% optimize.loop_unroll : 0.000618s : 1.67% optimize.opt_after_cconv.c_1 : 0.000057s : 0.15% optimize.opt_after_cconv.parameter_eliminate : 0.000003s : 0.01% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000008s : 0.02% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000005s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000004s : 0.01% optimize.opt_after_cconv.cse : 0.000050s : 0.13% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000069s : 0.19% optimize.tuple_transform.d_1 : 0.000101s : 0.27% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.01% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000011s : 0.03% optimize.partial_unused_args_eliminate : 0.000002s : 0.01% optimize.add_recomputation : 0.000067s : 0.18% optimize.cse_after_recomputation.cse : 0.000027s : 0.07% optimize.environ_conv : 0.000007s : 0.02% optimize.swap_dp_allreduce_reducescatter : 0.000007s : 0.02% optimize.bias_add_comm_swap : 0.000003s : 0.01% optimize.label_micro_interleaved_index : 0.000005s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.01% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.01% optimize.micro_interleaved_order_control : 0.000002s : 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 : 0.000002s : 0.01% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.01% 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 : 0.01% optimize.control_data_broadcast_order : 0.000018s : 0.05% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000005s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000006s : 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 : 0.000002s : 0.01% optimize.overlap_grad_ring_attention : 0.000005s : 0.01% optimize.overlap_grad_flash_sp : 0.000022s : 0.06% 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 : 0.000079s : 0.21% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000017s : 0.05% optimize.symbol_engine_optimizer.elim_not_effective : 0.000024s : 0.06% optimize.symbol_engine_optimizer.opt_reshape : 0.000011s : 0.03% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000019s : 0.05% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.01% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000029s : 0.08% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000005s : 0.01% opt_after_jit_grad : 0.000543s : 1.47% validate : 0.000052s : 0.14% Time group info: ------[substitution.] 0.000130 33 2.25% : 0.000003s : 2: substitution.elim_not_effective 4.82% : 0.000006s : 2: substitution.fold_const_symbol 5.98% : 0.000008s : 8: substitution.graph_param_transform 74.70% : 0.000097s : 1: substitution.inline 2.84% : 0.000004s : 4: substitution.j_node_and_user_rematch 3.69% : 0.000005s : 4: substitution.remove_not_recompute_node 5.72% : 0.000007s : 12: substitution.replace_old_param ------[type_inference.] 0.029956 2 98.06% : 0.029376s : 1: type_inference.infer 1.94% : 0.000581s : 1: type_inference.specialize ------[replace.] 0.000024 1 100.00% : 0.000024s : 1: replace.inline ------[match.] 0.000096 1 100.00% : 0.000096s : 1: match.inline ------[predicate.] 0.000212 1879 0.83% : 0.000002s : 17: predicate.accumulaten_eliminater 0.81% : 0.000002s : 8: predicate.ad_related_special_op_eliminate 0.72% : 0.000002s : 16: predicate.addn_check_dump 0.82% : 0.000002s : 17: predicate.addn_zero_filter 0.74% : 0.000002s : 17: predicate.adjust_all_reduce_mul_add 2.03% : 0.000004s : 33: predicate.arithmetic_simplify 0.96% : 0.000002s : 17: predicate.cast_eliminate 0.77% : 0.000002s : 16: predicate.check_bprop_eliminate 0.73% : 0.000002s : 16: predicate.compare_switch_simplify 0.33% : 0.000001s : 8: predicate.const_output_eliminate 0.79% : 0.000002s : 16: predicate.depend_value_elim 0.86% : 0.000002s : 17: predicate.dict_get_item_const_eliminator 0.93% : 0.000002s : 17: predicate.dict_get_item_eliminator 0.84% : 0.000002s : 17: predicate.dict_set_item_eliminator 1.14% : 0.000002s : 16: predicate.dumpgradient_eliminate 0.35% : 0.000001s : 8: predicate.elim_not_effective 0.51% : 0.000001s : 8: predicate.elim_shapecalc_of_broadcastargs 1.23% : 0.000003s : 25: predicate.environ_add_const_eliminate 1.16% : 0.000002s : 25: predicate.environ_get_add_eliminate 1.10% : 0.000002s : 25: predicate.environ_get_depend_swap 1.92% : 0.000004s : 41: predicate.environ_get_eliminate 1.14% : 0.000002s : 25: predicate.environ_get_set_eliminate 0.81% : 0.000002s : 18: predicate.exchange_switch_depend_value 1.48% : 0.000003s : 18: predicate.float_depend_g_call 0.73% : 0.000002s : 16: predicate.float_environ_get_switch 1.10% : 0.000002s : 24: predicate.float_tuple_getitem_switch 0.29% : 0.000001s : 8: predicate.fold_const_symbol 0.88% : 0.000002s : 16: predicate.get_grad_eliminate 0.34% : 0.000001s : 8: predicate.graph_param_transform 0.80% : 0.000002s : 16: predicate.incorporate_call 0.72% : 0.000002s : 16: predicate.incorporate_call_switch 5.50% : 0.000012s : 83: predicate.inline 0.96% : 0.000002s : 16: predicate.inline_without_move 0.54% : 0.000001s : 16: predicate.j_node_and_user_rematch 0.94% : 0.000002s : 16: predicate.less_batch_normalization 1.73% : 0.000004s : 33: predicate.list_to_tuple_eliminator_ 2.44% : 0.000005s : 50: predicate.load_eliminater 1.05% : 0.000002s : 8: predicate.loop_unroll_after_grad 1.35% : 0.000003s : 25: predicate.loop_unroll_before_grad 1.70% : 0.000004s : 33: predicate.make_slice_get_slice_eliminator 0.76% : 0.000002s : 16: predicate.merge_addn 0.76% : 0.000002s : 16: predicate.micro_step_allgather_replace 0.82% : 0.000002s : 16: predicate.mini_step_allgather_replace 0.76% : 0.000002s : 17: predicate.minmaximum_grad 1.25% : 0.000003s : 8: predicate.mutable_eliminate 0.51% : 0.000001s : 8: predicate.opt_reshape 0.45% : 0.000001s : 8: predicate.parallel_virtual_node 1.14% : 0.000002s : 18: predicate.partial_defer_inline 1.29% : 0.000003s : 25: predicate.partial_eliminate 0.92% : 0.000002s : 17: predicate.print_const_string_wrapper 0.76% : 0.000002s : 16: predicate.reduce_all_const_elim 1.04% : 0.000002s : 17: predicate.reduce_eliminate 2.40% : 0.000005s : 50: predicate.redundant_stop_gradient_eliminater 0.85% : 0.000002s : 16: predicate.remove_not_recompute_node 1.56% : 0.000003s : 33: predicate.replace_applicator 0.75% : 0.000002s : 16: predicate.replace_old_param 0.39% : 0.000001s : 8: predicate.reset_defer_inline 0.85% : 0.000002s : 17: predicate.reshape_eliminate 0.87% : 0.000002s : 16: predicate.row_tensor_add_zeros_like 0.45% : 0.000001s : 8: predicate.row_tensor_eliminate 0.98% : 0.000002s : 16: predicate.same_eliminate 0.72% : 0.000002s : 16: predicate.set_cell_output_no_recompute 0.88% : 0.000002s : 16: predicate.shard_identity_eliminate 0.96% : 0.000002s : 16: predicate.special_op_eliminate 0.88% : 0.000002s : 16: predicate.specialize_transform 1.02% : 0.000002s : 16: predicate.split_environ_get_set_with_tuple_value 1.01% : 0.000002s : 16: predicate.stack_unstack_eliminate 0.46% : 0.000001s : 8: predicate.switch_call_monad_eliminater 0.86% : 0.000002s : 18: predicate.switch_defer_inline 1.73% : 0.000004s : 34: predicate.switch_layer_defer_inline 3.68% : 0.000008s : 67: predicate.switch_simplify 0.83% : 0.000002s : 17: predicate.tile_eliminate 0.81% : 0.000002s : 17: predicate.transpose_eliminate 1.53% : 0.000003s : 33: predicate.tuple_list_convert_item_index_to_positive 1.66% : 0.000004s : 33: predicate.tuple_list_get_item_const_eliminator 1.46% : 0.000003s : 33: predicate.tuple_list_get_item_depend_reorder 3.20% : 0.000007s : 49: predicate.tuple_list_get_item_eliminator 1.54% : 0.000003s : 33: predicate.tuple_list_get_set_item_eliminator 2.50% : 0.000005s : 49: predicate.tuple_list_set_item_eliminator 1.76% : 0.000004s : 33: predicate.tuple_to_list_eliminator_ 2.19% : 0.000005s : 50: predicate.updatestate_pure_node_eliminater 3.20% : 0.000007s : 66: predicate.updatestate_useless_node_eliminater 0.53% : 0.000001s : 8: predicate.value_based_eliminate 0.99% : 0.000002s : 16: predicate.virtual_dataset_eliminate 0.90% : 0.000002s : 16: predicate.virtual_output_eliminate 0.42% : 0.000001s : 8: predicate.virtual_view_grad_eliminate 0.59% : 0.000001s : 8: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000799 6 58.28% : 0.000466s : 3: func_graph_cloner_run.FuncGraphClonerGraph 41.72% : 0.000333s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.057216 192 0.01% : 0.000004s : 1: ForceFp32Comm 8.68% : 0.004964s : 1: add_attr 8.65% : 0.004949s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.12% : 0.000071s : 1: add_recomputation 0.01% : 0.000004s : 1: assign_add_opt 0.10% : 0.000058s : 1: auto_monad 0.06% : 0.000033s : 1: auto_monad_reorder 0.01% : 0.000004s : 1: begin_end_overlap_inline 0.01% : 0.000007s : 1: bias_add_comm_swap 1.35% : 0.000770s : 1: bootstrap 0.06% : 0.000035s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.04% : 0.000023s : 1: control_data_broadcast_order 0.02% : 0.000013s : 1: convert_after_rewriter 0.07% : 0.000042s : 1: cse_after_recomputation 0.01% : 0.000005s : 1: dataset_repeat_opt 0.01% : 0.000006s : 1: detach_backward 0.02% : 0.000011s : 1: environ_conv 0.04% : 0.000021s : 1: event_method 0.01% : 0.000005s : 1: full_micro_interleaved_order_control 0.01% : 0.000005s : 1: get_jit_bprop_graph 0.01% : 0.000008s : 1: graph_reusing 0.01% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000004s : 1: handle_group_info 0.01% : 0.000005s : 1: inline 0.01% : 0.000006s : 1: insert-virtual-dataset 0.01% : 0.000004s : 1: interleave_parallel_branches 0.01% : 0.000005s : 1: interleave_split_concat_branches 0.01% : 0.000006s : 1: label_fine_grained_interleaved_index 0.02% : 0.000009s : 1: label_micro_interleaved_index 1.10% : 0.000631s : 1: loop_unroll 0.01% : 0.000004s : 1: merge_cast_opt 0.01% : 0.000005s : 1: micro_interleaved_order_control 1.24% : 0.000712s : 1: mutable_eliminate 0.02% : 0.000009s : 1: offloading_packed_experts 0.04% : 0.000021s : 1: opt.transform.loop_unroll_optimizer 0.04% : 0.000020s : 1: opt.transform.mutable_eliminate 2.49% : 0.001426s : 78: opt.transform.opt_a 0.10% : 0.000056s : 1: opt.transform.opt_after_cconv 0.07% : 0.000039s : 1: opt.transform.opt_after_jit_grad 0.38% : 0.000218s : 28: opt.transform.opt_b 0.19% : 0.000110s : 2: opt.transform.opt_trans_graph 0.12% : 0.000068s : 4: opt.transform.symbol_engine_opt 5.89% : 0.003368s : 1: opt_a 0.35% : 0.000197s : 1: opt_after_cconv 0.97% : 0.000554s : 1: opt_after_jit_grad 0.69% : 0.000394s : 1: opt_b 11.33% : 0.006480s : 1: optimize 0.06% : 0.000036s : 1: optimize_parallel_all_gather_comm 0.03% : 0.000014s : 1: order_py_execute_after_rewriter 0.04% : 0.000025s : 1: overlap_grad_flash_sp 0.01% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.02% : 0.000009s : 1: overlap_grad_ring_attention 0.01% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.01% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.01% : 0.000006s : 1: overlap_param_gather 0.01% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.02% : 0.000009s : 1: overlap_recompute_and_grad_model_parallel 0.01% : 0.000005s : 1: overlap_recompute_comm 0.01% : 0.000008s : 1: parallel-infer-symbol 0.01% : 0.000004s : 1: parallel-infer-symbol-second 0.01% : 0.000006s : 1: partial_unused_args_eliminate 0.01% : 0.000006s : 1: pipeline_parallel_scheduler 0.01% : 0.000005s : 1: pipeline_split 0.06% : 0.000033s : 1: pre_auto_parallel 0.05% : 0.000031s : 1: py_interpret_to_execute 0.06% : 0.000032s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000004s : 1: remove_cast_before_assign_add 0.13% : 0.000076s : 1: remove_dup_value 0.68% : 0.000389s : 1: renormalize.infer 0.78% : 0.000444s : 1: renormalize.specialize 0.01% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.02% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.16% : 0.000091s : 1: rewriter_after_opt_a 0.15% : 0.000088s : 1: rewriter_before_opt_a 0.02% : 0.000014s : 1: slice_cell_reuse_recomputed_activation 0.01% : 0.000005s : 1: slice_recompute_activation 0.01% : 0.000005s : 1: split_layernorm_comm 0.01% : 0.000006s : 1: split_matmul_comm_elemetwise 0.02% : 0.000011s : 1: swap_dp_allreduce_reducescatter 0.36% : 0.000204s : 1: symbol_engine_optimizer 0.25% : 0.000145s : 1: tuple_transform 52.56% : 0.030071s : 1: type_inference .[WARNING] PRE_ACT(170442,ffffbeca8f30,python3.9):2026-01-29-17:42:53.436.975 [mindspore/ccsrc/runtime/memory/mem_pool/abstract_dynamic_mem_pool.cc:1109] FreeIdleMemsByEagerFree] Eager free count : 8, free memory : 2725209600, real free : 2726297600, not free : 0. [hook] pytest_runtest_teardown:test_paged_attention_trapezoidal[1-TRAPEZOIDAL-float32] tests/st/infer/ops/test_internal_ops/test_paged_attention.py::test_paged_attention_trapezoidal[1-TRAPEZOIDAL-float32],max_mem:2696.0M =============================== 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 ================== 8 passed, 25 warnings in 363.19s (0:06:03) ==================