==================================================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_005/sault/config/pytest.ini plugins: mock-3.14.0, hydra-core-1.3.2, forked-1.6.0, anyio-4.9.0, xdist-1.32.0 collected 4 items test_paged_attention.py [WARNING] ME(163462:281473583890224,MainProcess):2026-01-29-17:37:25.796.451 [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 = 0.0384678, [21] [bootstrap]: 0.00078945 [type_inference]: 0.0210923 [event_method]: 1.532e-05 [auto_monad]: 9.351e-05 [graph_reusing]: 5.40999e-06 [inline]: 2.89999e-06 [add_attr]: 0.00877306, [1] [add_attr_with_inline]: 0.00875942, [1] [Cycle 1]: 0.00010245, [2] [tag_attr]: 2.121e-05 [meta_addattr_fg_expand]: 4.43999e-06 [parallel-infer-symbol]: 3.62998e-06 [pre_auto_parallel]: 4.297e-05 [insert-virtual-dataset]: 2.79999e-06 [parallel-infer-symbol-second]: 7.40023e-07 [dataset_repeat_opt]: 2.02001e-06 [pipeline_split]: 1.98002e-06 [optimize]: 0.0067006, [53] [py_interpret_to_execute]: 2.923e-05 [rewriter_before_opt_a]: 8.135e-05 [opt_a]: 0.00363608, [2] [Cycle 1]: 0.00242406, [45] [expand_dump_flag]: 2.61999e-06 [switch_simplify]: 3.59e-05 [loop_unroll]: 2.093e-05 [a_1]: 0.00050893 [with_stream_mark]: 2.353e-05 [recompute_prepare]: 1.51e-05 [updatestate_depend_eliminate]: 5.58002e-06 [updatestate_assign_eliminate]: 5.15001e-06 [updatestate_loads_eliminate]: 5.04998e-06 [parameter_eliminate]: 1.84e-06 [a_2]: 0.00016842 [accelerated_algorithm]: 1.463e-05 [shard]: 2.71e-06 [meta_shard_fg_expand]: 2.91e-06 [shard_inline]: 1.229e-05 [merge_send_recv]: 2.08e-05 [auto_parallel]: 1.034e-05 [parallel]: 5.379e-05 [flash_sp]: 2.096e-05 [merge_comm]: 6.46999e-06 [allreduce_fusion]: 5.25001e-06 [matmul_add_comm_reduction]: 1.191e-05 [allreduce_slice_to_reducescatter]: 8.2e-07 [virtual_shard_identity]: 1.636e-05 [virtual_dataset]: 1.245e-05 [get_grad_eliminate_]: 1.211e-05 [virtual_output]: 1.354e-05 [merge_forward]: 5.79999e-06 [cell_reuse_recompute_pass]: 1.76e-06 [offload_activation]: 1.297e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.81e-05 [merge_recompute_call_nodes]: 1.52001e-06 [before_grad]: 1.734e-05 [set_forward_comm_id_for_comm_node_pass]: 5.57001e-06 [meta_fg_expand]: 4.60999e-06 [flash_sp_send_recv_attached]: 2.66999e-06 [receive_attached]: 2.02999e-06 [after_resolve]: 2.619e-05 [a_after_grad]: 2.018e-05 [renormalize]: 0.00082157 [add_forward_monad_depend]: 5.78997e-06 [auto_monad_grad]: 3.09999e-06 [auto_monad_eliminator]: 1.893e-05 [cse]: 8.284e-05 [a_3]: 8.976e-05 [Cycle 2]: 0.0012001, [45] [expand_dump_flag]: 1.32999e-06 [switch_simplify]: 1.378e-05 [loop_unroll]: 1.191e-05 [a_1]: 0.00038798 [with_stream_mark]: 1.875e-05 [recompute_prepare]: 1.274e-05 [updatestate_depend_eliminate]: 5.72001e-06 [updatestate_assign_eliminate]: 4.47998e-06 [updatestate_loads_eliminate]: 4.85999e-06 [parameter_eliminate]: 1.42e-06 [a_2]: 0.00015356 [accelerated_algorithm]: 1.195e-05 [shard]: 1.37e-06 [meta_shard_fg_expand]: 2.38002e-06 [shard_inline]: 1.218e-05 [merge_send_recv]: 8.06001e-06 [auto_parallel]: 9.22001e-06 [parallel]: 6.98e-06 [flash_sp]: 4.10998e-06 [merge_comm]: 5.35999e-06 [allreduce_fusion]: 4.88001e-06 [matmul_add_comm_reduction]: 1.015e-05 [allreduce_slice_to_reducescatter]: 4.50003e-07 [virtual_shard_identity]: 1.411e-05 [virtual_dataset]: 1.156e-05 [get_grad_eliminate_]: 1.186e-05 [virtual_output]: 1.112e-05 [merge_forward]: 5.38002e-06 [cell_reuse_recompute_pass]: 1.97001e-06 [offload_activation]: 1.04e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.953e-05 [merge_recompute_call_nodes]: 1.21997e-06 [before_grad]: 1.597e-05 [set_forward_comm_id_for_comm_node_pass]: 5.22e-06 [meta_fg_expand]: 3.91001e-06 [flash_sp_send_recv_attached]: 1.29e-06 [receive_attached]: 1.94e-06 [after_resolve]: 2.556e-05 [a_after_grad]: 1.803e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 1.75001e-06 [auto_monad_grad]: 1.73997e-06 [auto_monad_eliminator]: 1.168e-05 [cse]: 3.512e-05 [a_3]: 7.567e-05 [py_interpret_to_execute_after_opt_a]: 1.806e-05 [slice_cell_reuse_recomputed_activation]: 2.43e-06 [rewriter_after_opt_a]: 8.425e-05 [convert_after_rewriter]: 1.203e-05 [order_py_execute_after_rewriter]: 8.08999e-06 [mutable_eliminate]: 0.00071201 [opt_b]: 0.00038195, [1] [Cycle 1]: 0.00037321, [7] [b_1]: 0.00025374 [b_2]: 1.486e-05 [updatestate_depend_eliminate]: 1.045e-05 [updatestate_assign_eliminate]: 4.28999e-06 [updatestate_loads_eliminate]: 4.24002e-06 [renormalize]: 9.10019e-07 [cse]: 4.568e-05 [optimize_parallel_all_gather_comm]: 2.274e-05 [overlap_param_gather]: 4.17e-06 [cconv]: 3.474e-05 [loop_unroll]: 0.00053842 [opt_after_cconv]: 0.00016971, [1] [Cycle 1]: 0.00016201, [7] [c_1]: 6.452e-05 [parameter_eliminate]: 4.52e-06 [updatestate_depend_eliminate]: 8.60999e-06 [updatestate_assign_eliminate]: 4.27998e-06 [updatestate_loads_eliminate]: 4.08999e-06 [cse]: 3.907e-05 [renormalize]: 4.69998e-07 [remove_dup_value]: 5.62e-05 [tuple_transform]: 0.00013316, [1] [Cycle 1]: 0.00012761, [4] [d_1]: 9.308e-05 [none_parameter_eliminate]: 1.81003e-06 [renormalize]: 2.3999e-07 [switch_simplify]: 1.328e-05 [partial_unused_args_eliminate]: 2.29001e-06 [add_recomputation]: 6.448e-05 [cse_after_recomputation]: 3.608e-05, [1] [Cycle 1]: 3.126e-05, [1] [cse]: 2.497e-05 [environ_conv]: 2.031e-05 [swap_dp_allreduce_reducescatter]: 7.36001e-06 [bias_add_comm_swap]: 3.29001e-06 [label_micro_interleaved_index]: 6.63998e-06 [label_fine_grained_interleaved_index]: 2.72001e-06 [merge_cast_opt]: 1.47001e-06 [slice_recompute_activation]: 2.12001e-06 [micro_interleaved_order_control]: 2.41e-06 [assign_add_opt]: 1.35001e-06 [ForceFp32Comm]: 1.08001e-06 [remove_cast_before_assign_add]: 1.49e-06 [full_micro_interleaved_order_control]: 2.56e-06 [reorder_send_recv_between_fp_bp]: 3.4e-06 [comm_op_add_attrs]: 1.09e-06 [add_comm_op_reuse_tag]: 1.00001e-06 [interleave_split_concat_branches]: 1.22e-06 [interleave_parallel_branches]: 1.35001e-06 [overlap_opt_shard_in_pipeline]: 2.265e-05 [overlap_opt_shard_grad_in_pipeline]: 1.87001e-06 [control_data_broadcast_order]: 1.772e-05 [grouped_pairwise_exchange_alltoall]: 1.69e-06 [offloading_packed_experts]: 5.31002e-06 [overlap_recompute_and_grad_model_parallel]: 5.86e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.24e-06 [overlap_recompute_allgather_and_fa_grad]: 1.45001e-06 [overlap_recompute_comm]: 2.32999e-06 [overlap_grad_ring_attention]: 5.61e-06 [overlap_grad_flash_sp]: 2.513e-05 [begin_end_overlap_inline]: 8.30012e-07 [split_matmul_comm_elemetwise]: 2.42001e-06 [split_layernorm_comm]: 1.87999e-06 [handle_group_info]: 1.08001e-06 [symbol_engine_optimizer]: 0.00026234, [1] [Cycle 1]: 0.00025691, [6] [build]: 9.277e-05 [elim_shapecalc]: 1.968e-05 [elim_not_effective]: 2.98e-05 [opt_reshape]: 1.287e-05 [fold_const_symbol]: 6.148e-05 [renormalize]: 2.60014e-07 [detach_backward]: 2.56e-06 [pipeline_parallel_scheduler]: 1.59e-06 [auto_monad_reorder]: 3.075e-05 [get_jit_bprop_graph]: 1.94e-06 [rewriter_after_jit_bprop_graph]: 4.29002e-06 [opt_after_jit_grad]: 0.00064411 [validate]: 7.528e-05 Sums bootstrap : 0.000789s : 2.76% type_inference : 0.021092s : 73.70% event_method : 0.000015s : 0.05% auto_monad : 0.000094s : 0.33% graph_reusing : 0.000005s : 0.02% inline : 0.000003s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000021s : 0.07% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.02% parallel-infer-symbol : 0.000004s : 0.01% pre_auto_parallel : 0.000043s : 0.15% 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 : 0.01% optimize.py_interpret_to_execute : 0.000029s : 0.10% optimize.rewriter_before_opt_a : 0.000081s : 0.28% optimize.opt_a.expand_dump_flag : 0.000004s : 0.01% optimize.opt_a.switch_simplify : 0.000050s : 0.17% optimize.opt_a.loop_unroll : 0.000033s : 0.11% optimize.opt_a.a_1 : 0.000897s : 3.13% optimize.opt_a.with_stream_mark : 0.000042s : 0.15% optimize.opt_a.recompute_prepare : 0.000028s : 0.10% optimize.opt_a.updatestate_depend_eliminate : 0.000011s : 0.04% 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 : 0.000322s : 1.13% optimize.opt_a.accelerated_algorithm : 0.000027s : 0.09% optimize.opt_a.shard : 0.000004s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000005s : 0.02% optimize.opt_a.shard_inline : 0.000024s : 0.09% optimize.opt_a.merge_send_recv : 0.000029s : 0.10% optimize.opt_a.auto_parallel : 0.000020s : 0.07% optimize.opt_a.parallel : 0.000061s : 0.21% optimize.opt_a.flash_sp : 0.000025s : 0.09% optimize.opt_a.merge_comm : 0.000012s : 0.04% optimize.opt_a.allreduce_fusion : 0.000010s : 0.04% optimize.opt_a.matmul_add_comm_reduction : 0.000022s : 0.08% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000030s : 0.11% optimize.opt_a.virtual_dataset : 0.000024s : 0.08% optimize.opt_a.get_grad_eliminate_ : 0.000024s : 0.08% optimize.opt_a.virtual_output : 0.000025s : 0.09% optimize.opt_a.merge_forward : 0.000011s : 0.04% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.01% optimize.opt_a.offload_activation : 0.000023s : 0.08% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000048s : 0.17% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.01% optimize.opt_a.before_grad : 0.000033s : 0.12% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000011s : 0.04% optimize.opt_a.meta_fg_expand : 0.000009s : 0.03% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.01% optimize.opt_a.receive_attached : 0.000004s : 0.01% optimize.opt_a.after_resolve : 0.000052s : 0.18% optimize.opt_a.a_after_grad : 0.000038s : 0.13% optimize.opt_a.renormalize : 0.000822s : 2.87% optimize.opt_a.add_forward_monad_depend : 0.000008s : 0.03% optimize.opt_a.auto_monad_grad : 0.000005s : 0.02% optimize.opt_a.auto_monad_eliminator : 0.000031s : 0.11% optimize.opt_a.cse : 0.000118s : 0.41% optimize.opt_a.a_3 : 0.000165s : 0.58% optimize.py_interpret_to_execute_after_opt_a : 0.000018s : 0.06% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.01% optimize.rewriter_after_opt_a : 0.000084s : 0.29% optimize.convert_after_rewriter : 0.000012s : 0.04% optimize.order_py_execute_after_rewriter : 0.000008s : 0.03% optimize.mutable_eliminate : 0.000712s : 2.49% optimize.opt_b.b_1 : 0.000254s : 0.89% optimize.opt_b.b_2 : 0.000015s : 0.05% optimize.opt_b.updatestate_depend_eliminate : 0.000010s : 0.04% optimize.opt_b.updatestate_assign_eliminate : 0.000004s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000004s : 0.01% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000046s : 0.16% optimize.optimize_parallel_all_gather_comm : 0.000023s : 0.08% optimize.overlap_param_gather : 0.000004s : 0.01% optimize.cconv : 0.000035s : 0.12% optimize.loop_unroll : 0.000538s : 1.88% optimize.opt_after_cconv.c_1 : 0.000065s : 0.23% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.02% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000009s : 0.03% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000004s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000004s : 0.01% optimize.opt_after_cconv.cse : 0.000039s : 0.14% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000056s : 0.20% optimize.tuple_transform.d_1 : 0.000093s : 0.33% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.01% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000013s : 0.05% optimize.partial_unused_args_eliminate : 0.000002s : 0.01% optimize.add_recomputation : 0.000064s : 0.23% optimize.cse_after_recomputation.cse : 0.000025s : 0.09% optimize.environ_conv : 0.000020s : 0.07% optimize.swap_dp_allreduce_reducescatter : 0.000007s : 0.03% optimize.bias_add_comm_swap : 0.000003s : 0.01% optimize.label_micro_interleaved_index : 0.000007s : 0.02% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.01% optimize.merge_cast_opt : 0.000001s : 0.01% 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 : 0.01% 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 : 0.000023s : 0.08% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.01% optimize.control_data_broadcast_order : 0.000018s : 0.06% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.01% optimize.offloading_packed_experts : 0.000005s : 0.02% 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 : 0.000001s : 0.01% optimize.overlap_recompute_comm : 0.000002s : 0.01% optimize.overlap_grad_ring_attention : 0.000006s : 0.02% optimize.overlap_grad_flash_sp : 0.000025s : 0.09% 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 : 0.000093s : 0.32% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000020s : 0.07% optimize.symbol_engine_optimizer.elim_not_effective : 0.000030s : 0.10% optimize.symbol_engine_optimizer.opt_reshape : 0.000013s : 0.04% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000061s : 0.21% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.01% pipeline_parallel_scheduler : 0.000002s : 0.01% auto_monad_reorder : 0.000031s : 0.11% get_jit_bprop_graph : 0.000002s : 0.01% rewriter_after_jit_bprop_graph : 0.000004s : 0.01% opt_after_jit_grad : 0.000644s : 2.25% validate : 0.000075s : 0.26% Time group info: ------[substitution.] 0.000167 39 6.86% : 0.000011s : 2: substitution.elim_not_effective 5.07% : 0.000008s : 2: substitution.fold_const_symbol 4.90% : 0.000008s : 10: substitution.graph_param_transform 65.05% : 0.000109s : 1: substitution.inline 2.74% : 0.000005s : 4: substitution.j_node_and_user_rematch 8.26% : 0.000014s : 4: substitution.remove_not_recompute_node 7.12% : 0.000012s : 16: substitution.replace_old_param ------[type_inference.] 0.020990 2 97.01% : 0.020363s : 1: type_inference.infer 2.99% : 0.000628s : 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.] 0.000381 2335 0.69% : 0.000003s : 21: predicate.accumulaten_eliminater 0.79% : 0.000003s : 10: predicate.ad_related_special_op_eliminate 0.64% : 0.000002s : 20: predicate.addn_check_dump 0.72% : 0.000003s : 21: predicate.addn_zero_filter 0.66% : 0.000003s : 21: predicate.adjust_all_reduce_mul_add 1.50% : 0.000006s : 41: predicate.arithmetic_simplify 0.76% : 0.000003s : 21: predicate.cast_eliminate 0.72% : 0.000003s : 20: predicate.check_bprop_eliminate 0.65% : 0.000002s : 20: predicate.compare_switch_simplify 0.26% : 0.000001s : 10: predicate.const_output_eliminate 0.67% : 0.000003s : 20: predicate.depend_value_elim 0.71% : 0.000003s : 21: predicate.dict_get_item_const_eliminator 18.20% : 0.000069s : 21: predicate.dict_get_item_eliminator 0.72% : 0.000003s : 21: predicate.dict_set_item_eliminator 0.83% : 0.000003s : 20: predicate.dumpgradient_eliminate 0.31% : 0.000001s : 10: predicate.elim_not_effective 0.49% : 0.000002s : 10: predicate.elim_shapecalc_of_broadcastargs 1.01% : 0.000004s : 31: predicate.environ_add_const_eliminate 0.99% : 0.000004s : 31: predicate.environ_get_add_eliminate 0.99% : 0.000004s : 31: predicate.environ_get_depend_swap 1.73% : 0.000007s : 51: predicate.environ_get_eliminate 0.97% : 0.000004s : 31: predicate.environ_get_set_eliminate 0.70% : 0.000003s : 22: predicate.exchange_switch_depend_value 1.12% : 0.000004s : 22: predicate.float_depend_g_call 0.64% : 0.000002s : 20: predicate.float_environ_get_switch 0.96% : 0.000004s : 30: predicate.float_tuple_getitem_switch 0.23% : 0.000001s : 10: predicate.fold_const_symbol 0.75% : 0.000003s : 20: predicate.get_grad_eliminate 0.40% : 0.000002s : 10: predicate.graph_param_transform 0.68% : 0.000003s : 20: predicate.incorporate_call 0.56% : 0.000002s : 20: predicate.incorporate_call_switch 4.15% : 0.000016s : 103: predicate.inline 0.77% : 0.000003s : 20: predicate.inline_without_move 0.48% : 0.000002s : 20: predicate.j_node_and_user_rematch 0.85% : 0.000003s : 20: predicate.less_batch_normalization 1.47% : 0.000006s : 41: predicate.list_to_tuple_eliminator_ 2.03% : 0.000008s : 62: predicate.load_eliminater 0.82% : 0.000003s : 10: predicate.loop_unroll_after_grad 1.09% : 0.000004s : 31: predicate.loop_unroll_before_grad 1.49% : 0.000006s : 41: predicate.make_slice_get_slice_eliminator 0.67% : 0.000003s : 20: predicate.merge_addn 0.65% : 0.000002s : 20: predicate.micro_step_allgather_replace 0.66% : 0.000003s : 20: predicate.mini_step_allgather_replace 0.63% : 0.000002s : 21: predicate.minmaximum_grad 0.81% : 0.000003s : 10: predicate.mutable_eliminate 0.37% : 0.000001s : 10: predicate.opt_reshape 0.41% : 0.000002s : 10: predicate.parallel_virtual_node 0.79% : 0.000003s : 22: predicate.partial_defer_inline 1.06% : 0.000004s : 31: predicate.partial_eliminate 0.68% : 0.000003s : 21: predicate.print_const_string_wrapper 0.67% : 0.000003s : 20: predicate.reduce_all_const_elim 0.90% : 0.000003s : 21: predicate.reduce_eliminate 1.97% : 0.000008s : 62: predicate.redundant_stop_gradient_eliminater 0.59% : 0.000002s : 20: predicate.remove_not_recompute_node 1.21% : 0.000005s : 41: predicate.replace_applicator 0.64% : 0.000002s : 20: predicate.replace_old_param 0.28% : 0.000001s : 10: predicate.reset_defer_inline 0.68% : 0.000003s : 21: predicate.reshape_eliminate 0.75% : 0.000003s : 20: predicate.row_tensor_add_zeros_like 0.35% : 0.000001s : 10: predicate.row_tensor_eliminate 0.84% : 0.000003s : 20: predicate.same_eliminate 0.56% : 0.000002s : 20: predicate.set_cell_output_no_recompute 0.86% : 0.000003s : 20: predicate.shard_identity_eliminate 0.75% : 0.000003s : 20: predicate.special_op_eliminate 0.72% : 0.000003s : 20: predicate.specialize_transform 0.75% : 0.000003s : 20: predicate.split_environ_get_set_with_tuple_value 0.85% : 0.000003s : 20: predicate.stack_unstack_eliminate 0.36% : 0.000001s : 10: predicate.switch_call_monad_eliminater 0.76% : 0.000003s : 22: predicate.switch_defer_inline 1.33% : 0.000005s : 42: predicate.switch_layer_defer_inline 3.00% : 0.000011s : 83: predicate.switch_simplify 0.65% : 0.000002s : 21: predicate.tile_eliminate 0.70% : 0.000003s : 21: predicate.transpose_eliminate 1.21% : 0.000005s : 41: predicate.tuple_list_convert_item_index_to_positive 1.27% : 0.000005s : 41: predicate.tuple_list_get_item_const_eliminator 1.20% : 0.000005s : 41: predicate.tuple_list_get_item_depend_reorder 2.22% : 0.000008s : 61: predicate.tuple_list_get_item_eliminator 1.18% : 0.000005s : 41: predicate.tuple_list_get_set_item_eliminator 1.99% : 0.000008s : 61: predicate.tuple_list_set_item_eliminator 1.70% : 0.000006s : 41: predicate.tuple_to_list_eliminator_ 1.89% : 0.000007s : 62: predicate.updatestate_pure_node_eliminater 2.66% : 0.000010s : 82: predicate.updatestate_useless_node_eliminater 1.33% : 0.000005s : 10: predicate.value_based_eliminate 0.74% : 0.000003s : 20: predicate.virtual_dataset_eliminate 0.78% : 0.000003s : 20: predicate.virtual_output_eliminate 0.29% : 0.000001s : 10: predicate.virtual_view_grad_eliminate 0.40% : 0.000002s : 10: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000538 6 54.59% : 0.000294s : 3: func_graph_cloner_run.FuncGraphClonerGraph 45.41% : 0.000244s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.056913 192 0.01% : 0.000004s : 1: ForceFp32Comm 15.43% : 0.008779s : 1: add_attr 15.40% : 0.008764s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.12% : 0.000070s : 1: add_recomputation 0.01% : 0.000004s : 1: assign_add_opt 0.17% : 0.000099s : 1: auto_monad 0.06% : 0.000035s : 1: auto_monad_reorder 0.01% : 0.000004s : 1: begin_end_overlap_inline 0.01% : 0.000006s : 1: bias_add_comm_swap 1.46% : 0.000828s : 1: bootstrap 0.07% : 0.000039s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.04% : 0.000021s : 1: control_data_broadcast_order 0.03% : 0.000016s : 1: convert_after_rewriter 0.07% : 0.000039s : 1: cse_after_recomputation 0.01% : 0.000005s : 1: dataset_repeat_opt 0.01% : 0.000006s : 1: detach_backward 0.04% : 0.000024s : 1: environ_conv 0.04% : 0.000023s : 1: event_method 0.01% : 0.000006s : 1: full_micro_interleaved_order_control 0.01% : 0.000005s : 1: get_jit_bprop_graph 0.02% : 0.000009s : 1: graph_reusing 0.01% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000004s : 1: handle_group_info 0.01% : 0.000006s : 1: inline 0.01% : 0.000006s : 1: insert-virtual-dataset 0.01% : 0.000004s : 1: interleave_parallel_branches 0.01% : 0.000004s : 1: interleave_split_concat_branches 0.01% : 0.000006s : 1: label_fine_grained_interleaved_index 0.02% : 0.000010s : 1: label_micro_interleaved_index 0.96% : 0.000549s : 1: loop_unroll 0.01% : 0.000005s : 1: merge_cast_opt 0.01% : 0.000005s : 1: micro_interleaved_order_control 1.27% : 0.000723s : 1: mutable_eliminate 0.01% : 0.000009s : 1: offloading_packed_experts 0.04% : 0.000022s : 1: opt.transform.loop_unroll_optimizer 0.04% : 0.000025s : 1: opt.transform.mutable_eliminate 3.03% : 0.001722s : 78: opt.transform.opt_a 0.11% : 0.000063s : 1: opt.transform.opt_after_cconv 0.08% : 0.000044s : 1: opt.transform.opt_after_jit_grad 0.42% : 0.000238s : 28: opt.transform.opt_b 0.18% : 0.000103s : 2: opt.transform.opt_trans_graph 0.21% : 0.000119s : 4: opt.transform.symbol_engine_opt 6.40% : 0.003640s : 1: opt_a 0.30% : 0.000173s : 1: opt_after_cconv 1.15% : 0.000657s : 1: opt_after_jit_grad 0.68% : 0.000385s : 1: opt_b 11.78% : 0.006705s : 1: optimize 0.05% : 0.000027s : 1: optimize_parallel_all_gather_comm 0.02% : 0.000011s : 1: order_py_execute_after_rewriter 0.05% : 0.000029s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.02% : 0.000010s : 1: overlap_grad_ring_attention 0.01% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.05% : 0.000027s : 1: overlap_opt_shard_in_pipeline 0.01% : 0.000007s : 1: overlap_param_gather 0.01% : 0.000005s : 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% : 0.000007s : 1: parallel-infer-symbol 0.01% : 0.000004s : 1: parallel-infer-symbol-second 0.01% : 0.000006s : 1: partial_unused_args_eliminate 0.01% : 0.000005s : 1: pipeline_parallel_scheduler 0.01% : 0.000005s : 1: pipeline_split 0.08% : 0.000047s : 1: pre_auto_parallel 0.06% : 0.000033s : 1: py_interpret_to_execute 0.04% : 0.000022s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000004s : 1: remove_cast_before_assign_add 0.11% : 0.000062s : 1: remove_dup_value 0.76% : 0.000435s : 1: renormalize.infer 0.66% : 0.000376s : 1: renormalize.specialize 0.01% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000007s : 1: rewriter_after_jit_bprop_graph 0.16% : 0.000091s : 1: rewriter_after_opt_a 0.15% : 0.000086s : 1: rewriter_before_opt_a 0.01% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.01% : 0.000005s : 1: slice_recompute_activation 0.01% : 0.000005s : 1: split_layernorm_comm 0.01% : 0.000005s : 1: split_matmul_comm_elemetwise 0.02% : 0.000010s : 1: swap_dp_allreduce_reducescatter 0.47% : 0.000265s : 1: symbol_engine_optimizer 0.24% : 0.000137s : 1: tuple_transform 37.10% : 0.021116s : 1: type_inference . [hook] pytest_runtest_teardown:test_paged_attention_large_gsq_pertoken[QuantMethod.FP16_VEC-0] tests/st/infer/ops/test_internal_ops/test_paged_attention.py::test_paged_attention_large_gsq_pertoken[QuantMethod.FP16_VEC-0],max_mem:520.0M [WARNING] ME(163462:281473583890224,MainProcess):2026-01-29-17:47:05.215.650 [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 = 0.0213719, [21] [bootstrap]: 0.00088446 [type_inference]: 0.00708063 [event_method]: 1.481e-05 [auto_monad]: 6.389e-05 [graph_reusing]: 5.14e-06 [inline]: 2.01e-06 [add_attr]: 0.00502553, [1] [add_attr_with_inline]: 0.00500564, [1] [Cycle 1]: 6.195e-05, [2] [tag_attr]: 1.936e-05 [meta_addattr_fg_expand]: 3.94997e-06 [parallel-infer-symbol]: 3.83001e-06 [pre_auto_parallel]: 3.18e-05 [insert-virtual-dataset]: 2.36e-06 [parallel-infer-symbol-second]: 7.39994e-07 [dataset_repeat_opt]: 2.09e-06 [pipeline_split]: 1.69e-06 [optimize]: 0.00729293, [53] [py_interpret_to_execute]: 2.774e-05 [rewriter_before_opt_a]: 7.048e-05 [opt_a]: 0.00379255, [2] [Cycle 1]: 0.00238779, [45] [expand_dump_flag]: 2.81999e-06 [switch_simplify]: 3.497e-05 [loop_unroll]: 2.173e-05 [a_1]: 0.00050607 [with_stream_mark]: 3.733e-05 [recompute_prepare]: 1.481e-05 [updatestate_depend_eliminate]: 7.11999e-06 [updatestate_assign_eliminate]: 5.81998e-06 [updatestate_loads_eliminate]: 5.01002e-06 [parameter_eliminate]: 2.09999e-06 [a_2]: 0.00016033 [accelerated_algorithm]: 1.266e-05 [shard]: 2.22999e-06 [meta_shard_fg_expand]: 2.94999e-06 [shard_inline]: 2.861e-05 [merge_send_recv]: 1.034e-05 [auto_parallel]: 9.02e-06 [parallel]: 2.957e-05 [flash_sp]: 9.99001e-06 [merge_comm]: 5.76e-06 [allreduce_fusion]: 5.04998e-06 [matmul_add_comm_reduction]: 1.165e-05 [allreduce_slice_to_reducescatter]: 8.30012e-07 [virtual_shard_identity]: 1.502e-05 [virtual_dataset]: 1.283e-05 [get_grad_eliminate_]: 1.172e-05 [virtual_output]: 1.184e-05 [merge_forward]: 5.73997e-06 [cell_reuse_recompute_pass]: 1.52999e-06 [offload_activation]: 1.199e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.969e-05 [merge_recompute_call_nodes]: 2.02001e-06 [before_grad]: 1.635e-05 [set_forward_comm_id_for_comm_node_pass]: 5.42001e-06 [meta_fg_expand]: 3.93999e-06 [flash_sp_send_recv_attached]: 2.63998e-06 [receive_attached]: 2.31e-06 [after_resolve]: 2.511e-05 [a_after_grad]: 1.83e-05 [renormalize]: 0.00076075 [add_forward_monad_depend]: 5.21998e-06 [auto_monad_grad]: 2.41e-06 [auto_monad_eliminator]: 1.821e-05 [cse]: 5.38e-05 [a_3]: 8.745e-05 [Cycle 2]: 0.00139218, [45] [expand_dump_flag]: 1.35999e-06 [switch_simplify]: 1.541e-05 [loop_unroll]: 1.225e-05 [a_1]: 0.00032033 [with_stream_mark]: 1.45e-05 [recompute_prepare]: 1.217e-05 [updatestate_depend_eliminate]: 5.09998e-06 [updatestate_assign_eliminate]: 4.31002e-06 [updatestate_loads_eliminate]: 4.99003e-06 [parameter_eliminate]: 9.89996e-07 [a_2]: 0.00015206 [accelerated_algorithm]: 1.195e-05 [shard]: 1.44e-06 [meta_shard_fg_expand]: 2.37999e-06 [shard_inline]: 1.145e-05 [merge_send_recv]: 6.95002e-06 [auto_parallel]: 8.01001e-06 [parallel]: 4.89e-06 [flash_sp]: 3.53e-06 [merge_comm]: 5.62001e-06 [allreduce_fusion]: 4.92e-06 [matmul_add_comm_reduction]: 7.96001e-06 [allreduce_slice_to_reducescatter]: 3.50003e-07 [virtual_shard_identity]: 1.353e-05 [virtual_dataset]: 1.169e-05 [get_grad_eliminate_]: 1.192e-05 [virtual_output]: 1.149e-05 [merge_forward]: 5.44998e-06 [cell_reuse_recompute_pass]: 1.72001e-06 [offload_activation]: 1.01e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.974e-05 [merge_recompute_call_nodes]: 9.70002e-07 [before_grad]: 1.601e-05 [set_forward_comm_id_for_comm_node_pass]: 5.42999e-06 [meta_fg_expand]: 3.58999e-06 [flash_sp_send_recv_attached]: 1.06002e-06 [receive_attached]: 1.38002e-06 [after_resolve]: 3.557e-05 [a_after_grad]: 1.926e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 1.75001e-06 [auto_monad_grad]: 1.25001e-06 [auto_monad_eliminator]: 1.051e-05 [cse]: 4.791e-05 [a_3]: 7.7e-05 [py_interpret_to_execute_after_opt_a]: 2.685e-05 [slice_cell_reuse_recomputed_activation]: 2.68e-06 [rewriter_after_opt_a]: 6.734e-05 [convert_after_rewriter]: 2.069e-05 [order_py_execute_after_rewriter]: 7.71999e-06 [mutable_eliminate]: 0.00071779 [opt_b]: 0.00042257, [1] [Cycle 1]: 0.00041462, [7] [b_1]: 0.00024856 [b_2]: 3.04e-05 [updatestate_depend_eliminate]: 8.01001e-06 [updatestate_assign_eliminate]: 4.84998e-06 [updatestate_loads_eliminate]: 4.15e-06 [renormalize]: 5.8001e-07 [cse]: 5.124e-05 [optimize_parallel_all_gather_comm]: 3.087e-05 [overlap_param_gather]: 3.06999e-06 [cconv]: 2.807e-05 [loop_unroll]: 0.00062172 [opt_after_cconv]: 0.00021442, [1] [Cycle 1]: 0.00020769, [7] [c_1]: 6.509e-05 [parameter_eliminate]: 2.67001e-06 [updatestate_depend_eliminate]: 7.68999e-06 [updatestate_assign_eliminate]: 4.82e-06 [updatestate_loads_eliminate]: 4.33001e-06 [cse]: 4.673e-05 [renormalize]: 3.9002e-07 [remove_dup_value]: 5.878e-05 [tuple_transform]: 0.00016156, [1] [Cycle 1]: 0.00014749, [4] [d_1]: 0.00010133 [none_parameter_eliminate]: 1.96998e-06 [renormalize]: 3.89991e-07 [switch_simplify]: 1.332e-05 [partial_unused_args_eliminate]: 2.37001e-06 [add_recomputation]: 0.00015781 [cse_after_recomputation]: 4.058e-05, [1] [Cycle 1]: 3.438e-05, [1] [cse]: 2.743e-05 [environ_conv]: 8.47e-06 [swap_dp_allreduce_reducescatter]: 9.01998e-06 [bias_add_comm_swap]: 3.09999e-06 [label_micro_interleaved_index]: 5.32999e-06 [label_fine_grained_interleaved_index]: 3.09999e-06 [merge_cast_opt]: 1.47999e-06 [slice_recompute_activation]: 2.35997e-06 [micro_interleaved_order_control]: 2.59999e-06 [assign_add_opt]: 1.40001e-06 [ForceFp32Comm]: 1.10001e-06 [remove_cast_before_assign_add]: 1.17e-06 [full_micro_interleaved_order_control]: 2.61999e-06 [reorder_send_recv_between_fp_bp]: 3.20002e-06 [comm_op_add_attrs]: 1.11002e-06 [add_comm_op_reuse_tag]: 1.01002e-06 [interleave_split_concat_branches]: 1.35001e-06 [interleave_parallel_branches]: 1.42999e-06 [overlap_opt_shard_in_pipeline]: 2.54001e-06 [overlap_opt_shard_grad_in_pipeline]: 1.86998e-06 [control_data_broadcast_order]: 5.708e-05 [grouped_pairwise_exchange_alltoall]: 1.54998e-06 [offloading_packed_experts]: 5.47001e-06 [overlap_recompute_and_grad_model_parallel]: 6.63998e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.57999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.53002e-06 [overlap_recompute_comm]: 2.59999e-06 [overlap_grad_ring_attention]: 6.02001e-06 [overlap_grad_flash_sp]: 2.45e-05 [begin_end_overlap_inline]: 5.50004e-07 [split_matmul_comm_elemetwise]: 2.26e-06 [split_layernorm_comm]: 1.79e-06 [handle_group_info]: 1.11002e-06 [symbol_engine_optimizer]: 0.00023644, [1] [Cycle 1]: 0.00022116, [6] [build]: 7.675e-05 [elim_shapecalc]: 1.871e-05 [elim_not_effective]: 2.204e-05 [opt_reshape]: 2.227e-05 [fold_const_symbol]: 2.172e-05 [renormalize]: 3.10014e-07 [detach_backward]: 1.72001e-06 [pipeline_parallel_scheduler]: 1.52999e-06 [auto_monad_reorder]: 3.246e-05 [get_jit_bprop_graph]: 1.20999e-06 [rewriter_after_jit_bprop_graph]: 4.18001e-06 [opt_after_jit_grad]: 0.00065415 [validate]: 6.847e-05 Sums bootstrap : 0.000884s : 6.05% type_inference : 0.007081s : 48.44% event_method : 0.000015s : 0.10% auto_monad : 0.000064s : 0.44% graph_reusing : 0.000005s : 0.04% inline : 0.000002s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000019s : 0.13% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.03% parallel-infer-symbol : 0.000004s : 0.03% pre_auto_parallel : 0.000032s : 0.22% insert-virtual-dataset : 0.000002s : 0.02% parallel-infer-symbol-second : 0.000001s : 0.01% dataset_repeat_opt : 0.000002s : 0.01% pipeline_split : 0.000002s : 0.01% optimize.py_interpret_to_execute : 0.000028s : 0.19% optimize.rewriter_before_opt_a : 0.000070s : 0.48% optimize.opt_a.expand_dump_flag : 0.000004s : 0.03% optimize.opt_a.switch_simplify : 0.000050s : 0.34% optimize.opt_a.loop_unroll : 0.000034s : 0.23% optimize.opt_a.a_1 : 0.000826s : 5.65% optimize.opt_a.with_stream_mark : 0.000052s : 0.35% 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 : 0.000010s : 0.07% optimize.opt_a.updatestate_loads_eliminate : 0.000010s : 0.07% optimize.opt_a.parameter_eliminate : 0.000003s : 0.02% optimize.opt_a.a_2 : 0.000312s : 2.14% optimize.opt_a.accelerated_algorithm : 0.000025s : 0.17% optimize.opt_a.shard : 0.000004s : 0.03% optimize.opt_a.meta_shard_fg_expand : 0.000005s : 0.04% optimize.opt_a.shard_inline : 0.000040s : 0.27% optimize.opt_a.merge_send_recv : 0.000017s : 0.12% optimize.opt_a.auto_parallel : 0.000017s : 0.12% optimize.opt_a.parallel : 0.000034s : 0.24% optimize.opt_a.flash_sp : 0.000014s : 0.09% optimize.opt_a.merge_comm : 0.000011s : 0.08% optimize.opt_a.allreduce_fusion : 0.000010s : 0.07% optimize.opt_a.matmul_add_comm_reduction : 0.000020s : 0.13% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000029s : 0.20% optimize.opt_a.virtual_dataset : 0.000025s : 0.17% optimize.opt_a.get_grad_eliminate_ : 0.000024s : 0.16% optimize.opt_a.virtual_output : 0.000023s : 0.16% optimize.opt_a.merge_forward : 0.000011s : 0.08% optimize.opt_a.cell_reuse_recompute_pass : 0.000003s : 0.02% optimize.opt_a.offload_activation : 0.000022s : 0.15% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000039s : 0.27% 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 : 0.000011s : 0.07% optimize.opt_a.meta_fg_expand : 0.000008s : 0.05% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.03% optimize.opt_a.receive_attached : 0.000004s : 0.03% optimize.opt_a.after_resolve : 0.000061s : 0.42% optimize.opt_a.a_after_grad : 0.000038s : 0.26% optimize.opt_a.renormalize : 0.000761s : 5.20% optimize.opt_a.add_forward_monad_depend : 0.000007s : 0.05% optimize.opt_a.auto_monad_grad : 0.000004s : 0.03% optimize.opt_a.auto_monad_eliminator : 0.000029s : 0.20% optimize.opt_a.cse : 0.000102s : 0.70% optimize.opt_a.a_3 : 0.000164s : 1.12% optimize.py_interpret_to_execute_after_opt_a : 0.000027s : 0.18% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.02% optimize.rewriter_after_opt_a : 0.000067s : 0.46% optimize.convert_after_rewriter : 0.000021s : 0.14% optimize.order_py_execute_after_rewriter : 0.000008s : 0.05% optimize.mutable_eliminate : 0.000718s : 4.91% optimize.opt_b.b_1 : 0.000249s : 1.70% optimize.opt_b.b_2 : 0.000030s : 0.21% optimize.opt_b.updatestate_depend_eliminate : 0.000008s : 0.05% optimize.opt_b.updatestate_assign_eliminate : 0.000005s : 0.03% optimize.opt_b.updatestate_loads_eliminate : 0.000004s : 0.03% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000051s : 0.35% optimize.optimize_parallel_all_gather_comm : 0.000031s : 0.21% optimize.overlap_param_gather : 0.000003s : 0.02% optimize.cconv : 0.000028s : 0.19% optimize.loop_unroll : 0.000622s : 4.25% optimize.opt_after_cconv.c_1 : 0.000065s : 0.45% optimize.opt_after_cconv.parameter_eliminate : 0.000003s : 0.02% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000008s : 0.05% 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 : 0.000047s : 0.32% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000059s : 0.40% optimize.tuple_transform.d_1 : 0.000101s : 0.69% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.01% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000013s : 0.09% optimize.partial_unused_args_eliminate : 0.000002s : 0.02% optimize.add_recomputation : 0.000158s : 1.08% optimize.cse_after_recomputation.cse : 0.000027s : 0.19% optimize.environ_conv : 0.000008s : 0.06% optimize.swap_dp_allreduce_reducescatter : 0.000009s : 0.06% optimize.bias_add_comm_swap : 0.000003s : 0.02% optimize.label_micro_interleaved_index : 0.000005s : 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 : 0.02% 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 : 0.000003s : 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 : 0.000003s : 0.02% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.01% optimize.control_data_broadcast_order : 0.000057s : 0.39% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.01% optimize.offloading_packed_experts : 0.000005s : 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 : 0.000002s : 0.01% optimize.overlap_recompute_comm : 0.000003s : 0.02% optimize.overlap_grad_ring_attention : 0.000006s : 0.04% optimize.overlap_grad_flash_sp : 0.000025s : 0.17% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.02% optimize.split_layernorm_comm : 0.000002s : 0.01% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000077s : 0.53% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000019s : 0.13% optimize.symbol_engine_optimizer.elim_not_effective : 0.000022s : 0.15% optimize.symbol_engine_optimizer.opt_reshape : 0.000022s : 0.15% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000022s : 0.15% 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 : 0.000032s : 0.22% get_jit_bprop_graph : 0.000001s : 0.01% rewriter_after_jit_bprop_graph : 0.000004s : 0.03% opt_after_jit_grad : 0.000654s : 4.47% validate : 0.000068s : 0.47% Time group info: ------[substitution.] 0.000144 39 3.00% : 0.000004s : 2: substitution.elim_not_effective 4.92% : 0.000007s : 2: substitution.fold_const_symbol 6.40% : 0.000009s : 10: substitution.graph_param_transform 71.98% : 0.000104s : 1: substitution.inline 2.73% : 0.000004s : 4: substitution.j_node_and_user_rematch 3.74% : 0.000005s : 4: substitution.remove_not_recompute_node 7.23% : 0.000010s : 16: substitution.replace_old_param ------[type_inference.] 0.006957 2 89.42% : 0.006221s : 1: type_inference.infer 10.58% : 0.000736s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000102 1 100.00% : 0.000102s : 1: match.inline ------[predicate.] 0.000306 2335 0.88% : 0.000003s : 21: predicate.accumulaten_eliminater 0.84% : 0.000003s : 10: predicate.ad_related_special_op_eliminate 0.77% : 0.000002s : 20: predicate.addn_check_dump 0.91% : 0.000003s : 21: predicate.addn_zero_filter 0.78% : 0.000002s : 21: predicate.adjust_all_reduce_mul_add 1.91% : 0.000006s : 41: predicate.arithmetic_simplify 0.90% : 0.000003s : 21: predicate.cast_eliminate 0.85% : 0.000003s : 20: predicate.check_bprop_eliminate 0.75% : 0.000002s : 20: predicate.compare_switch_simplify 0.32% : 0.000001s : 10: predicate.const_output_eliminate 0.84% : 0.000003s : 20: predicate.depend_value_elim 0.85% : 0.000003s : 21: predicate.dict_get_item_const_eliminator 0.88% : 0.000003s : 21: predicate.dict_get_item_eliminator 0.88% : 0.000003s : 21: predicate.dict_set_item_eliminator 1.10% : 0.000003s : 20: predicate.dumpgradient_eliminate 0.39% : 0.000001s : 10: predicate.elim_not_effective 0.51% : 0.000002s : 10: predicate.elim_shapecalc_of_broadcastargs 1.26% : 0.000004s : 31: predicate.environ_add_const_eliminate 1.21% : 0.000004s : 31: predicate.environ_get_add_eliminate 1.16% : 0.000004s : 31: predicate.environ_get_depend_swap 1.95% : 0.000006s : 51: predicate.environ_get_eliminate 1.17% : 0.000004s : 31: predicate.environ_get_set_eliminate 0.86% : 0.000003s : 22: predicate.exchange_switch_depend_value 1.34% : 0.000004s : 22: predicate.float_depend_g_call 0.79% : 0.000002s : 20: predicate.float_environ_get_switch 1.13% : 0.000003s : 30: predicate.float_tuple_getitem_switch 0.31% : 0.000001s : 10: predicate.fold_const_symbol 0.90% : 0.000003s : 20: predicate.get_grad_eliminate 0.46% : 0.000001s : 10: predicate.graph_param_transform 0.74% : 0.000002s : 20: predicate.incorporate_call 0.70% : 0.000002s : 20: predicate.incorporate_call_switch 5.12% : 0.000016s : 103: predicate.inline 0.94% : 0.000003s : 20: predicate.inline_without_move 0.57% : 0.000002s : 20: predicate.j_node_and_user_rematch 0.92% : 0.000003s : 20: predicate.less_batch_normalization 1.74% : 0.000005s : 41: predicate.list_to_tuple_eliminator_ 2.45% : 0.000008s : 62: predicate.load_eliminater 0.83% : 0.000003s : 10: predicate.loop_unroll_after_grad 1.30% : 0.000004s : 31: predicate.loop_unroll_before_grad 1.74% : 0.000005s : 41: predicate.make_slice_get_slice_eliminator 0.75% : 0.000002s : 20: predicate.merge_addn 0.81% : 0.000002s : 20: predicate.micro_step_allgather_replace 0.82% : 0.000003s : 20: predicate.mini_step_allgather_replace 0.80% : 0.000002s : 21: predicate.minmaximum_grad 0.93% : 0.000003s : 10: predicate.mutable_eliminate 0.46% : 0.000001s : 10: predicate.opt_reshape 0.40% : 0.000001s : 10: predicate.parallel_virtual_node 1.07% : 0.000003s : 22: predicate.partial_defer_inline 1.32% : 0.000004s : 31: predicate.partial_eliminate 0.87% : 0.000003s : 21: predicate.print_const_string_wrapper 0.81% : 0.000002s : 20: predicate.reduce_all_const_elim 1.03% : 0.000003s : 21: predicate.reduce_eliminate 2.45% : 0.000007s : 62: predicate.redundant_stop_gradient_eliminater 0.77% : 0.000002s : 20: predicate.remove_not_recompute_node 1.51% : 0.000005s : 41: predicate.replace_applicator 0.72% : 0.000002s : 20: predicate.replace_old_param 0.37% : 0.000001s : 10: predicate.reset_defer_inline 0.82% : 0.000003s : 21: predicate.reshape_eliminate 0.85% : 0.000003s : 20: predicate.row_tensor_add_zeros_like 0.46% : 0.000001s : 10: predicate.row_tensor_eliminate 0.96% : 0.000003s : 20: predicate.same_eliminate 0.68% : 0.000002s : 20: predicate.set_cell_output_no_recompute 0.95% : 0.000003s : 20: predicate.shard_identity_eliminate 0.88% : 0.000003s : 20: predicate.special_op_eliminate 0.95% : 0.000003s : 20: predicate.specialize_transform 1.00% : 0.000003s : 20: predicate.split_environ_get_set_with_tuple_value 1.03% : 0.000003s : 20: predicate.stack_unstack_eliminate 0.45% : 0.000001s : 10: predicate.switch_call_monad_eliminater 0.94% : 0.000003s : 22: predicate.switch_defer_inline 1.74% : 0.000005s : 42: predicate.switch_layer_defer_inline 3.80% : 0.000012s : 83: predicate.switch_simplify 0.86% : 0.000003s : 21: predicate.tile_eliminate 0.81% : 0.000002s : 21: predicate.transpose_eliminate 1.56% : 0.000005s : 41: predicate.tuple_list_convert_item_index_to_positive 1.63% : 0.000005s : 41: predicate.tuple_list_get_item_const_eliminator 1.50% : 0.000005s : 41: predicate.tuple_list_get_item_depend_reorder 2.65% : 0.000008s : 61: predicate.tuple_list_get_item_eliminator 1.54% : 0.000005s : 41: predicate.tuple_list_get_set_item_eliminator 2.43% : 0.000007s : 61: predicate.tuple_list_set_item_eliminator 2.13% : 0.000007s : 41: predicate.tuple_to_list_eliminator_ 2.34% : 0.000007s : 62: predicate.updatestate_pure_node_eliminater 3.39% : 0.000010s : 82: predicate.updatestate_useless_node_eliminater 1.16% : 0.000004s : 10: predicate.value_based_eliminate 0.85% : 0.000003s : 20: predicate.virtual_dataset_eliminate 0.85% : 0.000003s : 20: predicate.virtual_output_eliminate 0.45% : 0.000001s : 10: predicate.virtual_view_grad_eliminate 0.47% : 0.000001s : 10: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000737 6 61.21% : 0.000451s : 3: func_graph_cloner_run.FuncGraphClonerGraph 38.79% : 0.000286s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.036468 192 0.01% : 0.000004s : 1: ForceFp32Comm 13.80% : 0.005031s : 1: add_attr 13.74% : 0.005009s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.45% : 0.000164s : 1: add_recomputation 0.04% : 0.000014s : 1: assign_add_opt 0.19% : 0.000070s : 1: auto_monad 0.10% : 0.000037s : 1: auto_monad_reorder 0.04% : 0.000014s : 1: begin_end_overlap_inline 0.02% : 0.000007s : 1: bias_add_comm_swap 2.54% : 0.000927s : 1: bootstrap 0.09% : 0.000032s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.17% : 0.000061s : 1: control_data_broadcast_order 0.07% : 0.000025s : 1: convert_after_rewriter 0.12% : 0.000044s : 1: cse_after_recomputation 0.01% : 0.000005s : 1: dataset_repeat_opt 0.01% : 0.000005s : 1: detach_backward 0.03% : 0.000012s : 1: environ_conv 0.06% : 0.000022s : 1: event_method 0.08% : 0.000028s : 1: full_micro_interleaved_order_control 0.02% : 0.000006s : 1: get_jit_bprop_graph 0.03% : 0.000009s : 1: graph_reusing 0.02% : 0.000006s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000004s : 1: handle_group_info 0.02% : 0.000006s : 1: inline 0.02% : 0.000006s : 1: insert-virtual-dataset 0.01% : 0.000005s : 1: interleave_parallel_branches 0.04% : 0.000014s : 1: interleave_split_concat_branches 0.02% : 0.000008s : 1: label_fine_grained_interleaved_index 0.02% : 0.000009s : 1: label_micro_interleaved_index 1.74% : 0.000633s : 1: loop_unroll 0.01% : 0.000005s : 1: merge_cast_opt 0.02% : 0.000006s : 1: micro_interleaved_order_control 2.00% : 0.000729s : 1: mutable_eliminate 0.02% : 0.000009s : 1: offloading_packed_experts 0.09% : 0.000032s : 1: opt.transform.loop_unroll_optimizer 0.06% : 0.000022s : 1: opt.transform.mutable_eliminate 4.54% : 0.001657s : 78: opt.transform.opt_a 0.17% : 0.000064s : 1: opt.transform.opt_after_cconv 0.13% : 0.000046s : 1: opt.transform.opt_after_jit_grad 0.69% : 0.000250s : 28: opt.transform.opt_b 0.31% : 0.000112s : 2: opt.transform.opt_trans_graph 0.22% : 0.000080s : 4: opt.transform.symbol_engine_opt 10.45% : 0.003813s : 1: opt_a 0.60% : 0.000218s : 1: opt_after_cconv 1.83% : 0.000666s : 1: opt_after_jit_grad 1.17% : 0.000426s : 1: opt_b 20.01% : 0.007298s : 1: optimize 0.10% : 0.000035s : 1: optimize_parallel_all_gather_comm 0.03% : 0.000011s : 1: order_py_execute_after_rewriter 0.08% : 0.000028s : 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 0.01% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.02% : 0.000006s : 1: overlap_opt_shard_in_pipeline 0.05% : 0.000018s : 1: overlap_param_gather 0.01% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.03% : 0.000010s : 1: overlap_recompute_and_grad_model_parallel 0.04% : 0.000015s : 1: overlap_recompute_comm 0.02% : 0.000008s : 1: parallel-infer-symbol 0.01% : 0.000004s : 1: parallel-infer-symbol-second 0.02% : 0.000006s : 1: partial_unused_args_eliminate 0.01% : 0.000005s : 1: pipeline_parallel_scheduler 0.01% : 0.000005s : 1: pipeline_split 0.10% : 0.000036s : 1: pre_auto_parallel 0.09% : 0.000033s : 1: py_interpret_to_execute 0.08% : 0.000031s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000004s : 1: remove_cast_before_assign_add 0.17% : 0.000063s : 1: remove_dup_value 0.98% : 0.000359s : 1: renormalize.infer 1.08% : 0.000393s : 1: renormalize.specialize 0.02% : 0.000008s : 1: reorder_send_recv_between_fp_bp 0.02% : 0.000008s : 1: rewriter_after_jit_bprop_graph 0.20% : 0.000072s : 1: rewriter_after_opt_a 0.21% : 0.000075s : 1: rewriter_before_opt_a 0.02% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.01% : 0.000005s : 1: slice_recompute_activation 0.01% : 0.000005s : 1: split_layernorm_comm 0.02% : 0.000006s : 1: split_matmul_comm_elemetwise 0.07% : 0.000026s : 1: swap_dp_allreduce_reducescatter 0.66% : 0.000240s : 1: symbol_engine_optimizer 0.45% : 0.000165s : 1: tuple_transform 19.47% : 0.007102s : 1: type_inference . [hook] pytest_runtest_teardown:test_paged_attention_large_gsq_pertoken[QuantMethod.FP16_VEC-1] tests/st/infer/ops/test_internal_ops/test_paged_attention.py::test_paged_attention_large_gsq_pertoken[QuantMethod.FP16_VEC-1],max_mem:520.0M [WARNING] ME(163462:281473583890224,MainProcess):2026-01-29-17:47:44.798.879 [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 = 0.0164384, [21] [bootstrap]: 0.00074568 [type_inference]: 0.00510767 [event_method]: 1.29e-05 [auto_monad]: 5.797e-05 [graph_reusing]: 5.03002e-06 [inline]: 1.67001e-06 [add_attr]: 0.00390718, [1] [add_attr_with_inline]: 0.00389611, [1] [Cycle 1]: 5.542e-05, [2] [tag_attr]: 1.655e-05 [meta_addattr_fg_expand]: 4.00998e-06 [parallel-infer-symbol]: 2.63e-06 [pre_auto_parallel]: 3.14e-05 [insert-virtual-dataset]: 2.58998e-06 [parallel-infer-symbol-second]: 7.09988e-07 [dataset_repeat_opt]: 2.07999e-06 [pipeline_split]: 1.71e-06 [optimize]: 0.00579698, [53] [py_interpret_to_execute]: 2.431e-05 [rewriter_before_opt_a]: 6.341e-05 [opt_a]: 0.00317325, [2] [Cycle 1]: 0.00205206, [45] [expand_dump_flag]: 3.27002e-06 [switch_simplify]: 3.293e-05 [loop_unroll]: 2.122e-05 [a_1]: 0.00050514 [with_stream_mark]: 1.881e-05 [recompute_prepare]: 1.411e-05 [updatestate_depend_eliminate]: 5.80002e-06 [updatestate_assign_eliminate]: 4.99e-06 [updatestate_loads_eliminate]: 5.13002e-06 [parameter_eliminate]: 2.01998e-06 [a_2]: 0.00015751 [accelerated_algorithm]: 1.261e-05 [shard]: 2.51e-06 [meta_shard_fg_expand]: 2.21e-06 [shard_inline]: 1.191e-05 [merge_send_recv]: 1.031e-05 [auto_parallel]: 8.35001e-06 [parallel]: 2.694e-05 [flash_sp]: 1.057e-05 [merge_comm]: 5.59998e-06 [allreduce_fusion]: 5.35001e-06 [matmul_add_comm_reduction]: 1.143e-05 [allreduce_slice_to_reducescatter]: 6.09987e-07 [virtual_shard_identity]: 1.488e-05 [virtual_dataset]: 1.2e-05 [get_grad_eliminate_]: 1.165e-05 [virtual_output]: 1.147e-05 [merge_forward]: 5.44e-06 [cell_reuse_recompute_pass]: 1.30001e-06 [offload_activation]: 1.113e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.814e-05 [merge_recompute_call_nodes]: 1.47001e-06 [before_grad]: 2.006e-05 [set_forward_comm_id_for_comm_node_pass]: 5.43002e-06 [meta_fg_expand]: 3.78001e-06 [flash_sp_send_recv_attached]: 2.44001e-06 [receive_attached]: 2.12999e-06 [after_resolve]: 2.541e-05 [a_after_grad]: 1.901e-05 [renormalize]: 0.00056871 [add_forward_monad_depend]: 4.72e-06 [auto_monad_grad]: 2.06998e-06 [auto_monad_eliminator]: 1.647e-05 [cse]: 6.573e-05 [a_3]: 8.434e-05 [Cycle 2]: 0.0011111, [45] [expand_dump_flag]: 9.80013e-07 [switch_simplify]: 1.349e-05 [loop_unroll]: 1.206e-05 [a_1]: 0.00031093 [with_stream_mark]: 1.245e-05 [recompute_prepare]: 1.206e-05 [updatestate_depend_eliminate]: 5.25999e-06 [updatestate_assign_eliminate]: 4.23001e-06 [updatestate_loads_eliminate]: 1.018e-05 [parameter_eliminate]: 1.03001e-06 [a_2]: 0.00015501 [accelerated_algorithm]: 1.19e-05 [shard]: 1.36002e-06 [meta_shard_fg_expand]: 1.69e-06 [shard_inline]: 1.157e-05 [merge_send_recv]: 6.17001e-06 [auto_parallel]: 7.77998e-06 [parallel]: 4.38001e-06 [flash_sp]: 3.18998e-06 [merge_comm]: 5.05001e-06 [allreduce_fusion]: 4.42e-06 [matmul_add_comm_reduction]: 7.2e-06 [allreduce_slice_to_reducescatter]: 3.09985e-07 [virtual_shard_identity]: 1.267e-05 [virtual_dataset]: 1.174e-05 [get_grad_eliminate_]: 1.173e-05 [virtual_output]: 1.167e-05 [merge_forward]: 4.43999e-06 [cell_reuse_recompute_pass]: 1.25001e-06 [offload_activation]: 8.28999e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.761e-05 [merge_recompute_call_nodes]: 7.2e-07 [before_grad]: 1.54e-05 [set_forward_comm_id_for_comm_node_pass]: 5.09e-06 [meta_fg_expand]: 3.18e-06 [flash_sp_send_recv_attached]: 8.90024e-07 [receive_attached]: 1.07e-06 [after_resolve]: 2.459e-05 [a_after_grad]: 1.744e-05 [renormalize]: 7.99773e-08 [add_forward_monad_depend]: 1.30999e-06 [auto_monad_grad]: 1.04e-06 [auto_monad_eliminator]: 8.94e-06 [cse]: 4.112e-05 [a_3]: 7.588e-05 [py_interpret_to_execute_after_opt_a]: 1.245e-05 [slice_cell_reuse_recomputed_activation]: 1.99999e-06 [rewriter_after_opt_a]: 5.287e-05 [convert_after_rewriter]: 1.06e-05 [order_py_execute_after_rewriter]: 7.48999e-06 [mutable_eliminate]: 0.00049533 [opt_b]: 0.00037273, [1] [Cycle 1]: 0.00036563, [7] [b_1]: 0.00024909 [b_2]: 1.347e-05 [updatestate_depend_eliminate]: 7.50998e-06 [updatestate_assign_eliminate]: 4.50001e-06 [updatestate_loads_eliminate]: 4.10998e-06 [renormalize]: 4.50003e-07 [cse]: 4.597e-05 [optimize_parallel_all_gather_comm]: 2.056e-05 [overlap_param_gather]: 2.96999e-06 [cconv]: 2.522e-05 [loop_unroll]: 0.00052008 [opt_after_cconv]: 0.00016535, [1] [Cycle 1]: 0.00015905, [7] [c_1]: 6.348e-05 [parameter_eliminate]: 2.27001e-06 [updatestate_depend_eliminate]: 7.31001e-06 [updatestate_assign_eliminate]: 4.12998e-06 [updatestate_loads_eliminate]: 3.88001e-06 [cse]: 3.96e-05 [renormalize]: 5.50004e-07 [remove_dup_value]: 4.281e-05 [tuple_transform]: 0.0001292, [1] [Cycle 1]: 0.00012411, [4] [d_1]: 8.843e-05 [none_parameter_eliminate]: 1.76e-06 [renormalize]: 2.09984e-07 [switch_simplify]: 1.263e-05 [partial_unused_args_eliminate]: 1.99e-06 [add_recomputation]: 6.325e-05 [cse_after_recomputation]: 3.492e-05, [1] [Cycle 1]: 3.035e-05, [1] [cse]: 2.437e-05 [environ_conv]: 8.12998e-06 [swap_dp_allreduce_reducescatter]: 3.245e-05 [bias_add_comm_swap]: 2.74001e-06 [label_micro_interleaved_index]: 4.75999e-06 [label_fine_grained_interleaved_index]: 2.70002e-06 [merge_cast_opt]: 1.40001e-06 [slice_recompute_activation]: 2.41e-06 [micro_interleaved_order_control]: 2.48e-06 [assign_add_opt]: 1.27e-06 [ForceFp32Comm]: 7.79983e-07 [remove_cast_before_assign_add]: 1.12e-06 [full_micro_interleaved_order_control]: 2.63998e-06 [reorder_send_recv_between_fp_bp]: 3.36001e-06 [comm_op_add_attrs]: 1.07e-06 [add_comm_op_reuse_tag]: 1.00999e-06 [interleave_split_concat_branches]: 1.60999e-06 [interleave_parallel_branches]: 1.34e-06 [overlap_opt_shard_in_pipeline]: 2.29999e-06 [overlap_opt_shard_grad_in_pipeline]: 2.02001e-06 [control_data_broadcast_order]: 1.708e-05 [grouped_pairwise_exchange_alltoall]: 1.89999e-06 [offloading_packed_experts]: 4.97e-06 [overlap_recompute_and_grad_model_parallel]: 5.99999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.51998e-06 [overlap_recompute_allgather_and_fa_grad]: 1.54998e-06 [overlap_recompute_comm]: 2.73e-06 [overlap_grad_ring_attention]: 5.40999e-06 [overlap_grad_flash_sp]: 2.161e-05 [begin_end_overlap_inline]: 7.40023e-07 [split_matmul_comm_elemetwise]: 2.46e-06 [split_layernorm_comm]: 1.82999e-06 [handle_group_info]: 1.44e-06 [symbol_engine_optimizer]: 0.00018403, [1] [Cycle 1]: 0.00017925, [6] [build]: 7.366e-05 [elim_shapecalc]: 1.667e-05 [elim_not_effective]: 2.162e-05 [opt_reshape]: 1.263e-05 [fold_const_symbol]: 2.007e-05 [renormalize]: 2.60014e-07 [detach_backward]: 1.77001e-06 [pipeline_parallel_scheduler]: 1.50999e-06 [auto_monad_reorder]: 2.895e-05 [get_jit_bprop_graph]: 1.00001e-06 [rewriter_after_jit_bprop_graph]: 3.28998e-06 [opt_after_jit_grad]: 0.00051385 [validate]: 5.363e-05 Sums bootstrap : 0.000746s : 6.50% type_inference : 0.005108s : 44.53% event_method : 0.000013s : 0.11% auto_monad : 0.000058s : 0.51% graph_reusing : 0.000005s : 0.04% inline : 0.000002s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000017s : 0.14% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.03% parallel-infer-symbol : 0.000003s : 0.02% pre_auto_parallel : 0.000031s : 0.27% insert-virtual-dataset : 0.000003s : 0.02% parallel-infer-symbol-second : 0.000001s : 0.01% dataset_repeat_opt : 0.000002s : 0.02% pipeline_split : 0.000002s : 0.01% optimize.py_interpret_to_execute : 0.000024s : 0.21% optimize.rewriter_before_opt_a : 0.000063s : 0.55% optimize.opt_a.expand_dump_flag : 0.000004s : 0.04% optimize.opt_a.switch_simplify : 0.000046s : 0.40% optimize.opt_a.loop_unroll : 0.000033s : 0.29% optimize.opt_a.a_1 : 0.000816s : 7.11% optimize.opt_a.with_stream_mark : 0.000031s : 0.27% optimize.opt_a.recompute_prepare : 0.000026s : 0.23% optimize.opt_a.updatestate_depend_eliminate : 0.000011s : 0.10% optimize.opt_a.updatestate_assign_eliminate : 0.000009s : 0.08% optimize.opt_a.updatestate_loads_eliminate : 0.000015s : 0.13% optimize.opt_a.parameter_eliminate : 0.000003s : 0.03% optimize.opt_a.a_2 : 0.000313s : 2.72% optimize.opt_a.accelerated_algorithm : 0.000025s : 0.21% optimize.opt_a.shard : 0.000004s : 0.03% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.03% optimize.opt_a.shard_inline : 0.000023s : 0.20% optimize.opt_a.merge_send_recv : 0.000016s : 0.14% optimize.opt_a.auto_parallel : 0.000016s : 0.14% optimize.opt_a.parallel : 0.000031s : 0.27% optimize.opt_a.flash_sp : 0.000014s : 0.12% optimize.opt_a.merge_comm : 0.000011s : 0.09% optimize.opt_a.allreduce_fusion : 0.000010s : 0.09% optimize.opt_a.matmul_add_comm_reduction : 0.000019s : 0.16% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000028s : 0.24% optimize.opt_a.virtual_dataset : 0.000024s : 0.21% optimize.opt_a.get_grad_eliminate_ : 0.000023s : 0.20% optimize.opt_a.virtual_output : 0.000023s : 0.20% optimize.opt_a.merge_forward : 0.000010s : 0.09% optimize.opt_a.cell_reuse_recompute_pass : 0.000003s : 0.02% optimize.opt_a.offload_activation : 0.000019s : 0.17% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000036s : 0.31% optimize.opt_a.merge_recompute_call_nodes : 0.000002s : 0.02% optimize.opt_a.before_grad : 0.000035s : 0.31% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000011s : 0.09% optimize.opt_a.meta_fg_expand : 0.000007s : 0.06% optimize.opt_a.flash_sp_send_recv_attached : 0.000003s : 0.03% optimize.opt_a.receive_attached : 0.000003s : 0.03% optimize.opt_a.after_resolve : 0.000050s : 0.44% optimize.opt_a.a_after_grad : 0.000036s : 0.32% optimize.opt_a.renormalize : 0.000569s : 4.96% optimize.opt_a.add_forward_monad_depend : 0.000006s : 0.05% optimize.opt_a.auto_monad_grad : 0.000003s : 0.03% optimize.opt_a.auto_monad_eliminator : 0.000025s : 0.22% optimize.opt_a.cse : 0.000107s : 0.93% optimize.opt_a.a_3 : 0.000160s : 1.40% optimize.py_interpret_to_execute_after_opt_a : 0.000012s : 0.11% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.02% optimize.rewriter_after_opt_a : 0.000053s : 0.46% optimize.convert_after_rewriter : 0.000011s : 0.09% optimize.order_py_execute_after_rewriter : 0.000007s : 0.07% optimize.mutable_eliminate : 0.000495s : 4.32% optimize.opt_b.b_1 : 0.000249s : 2.17% optimize.opt_b.b_2 : 0.000013s : 0.12% optimize.opt_b.updatestate_depend_eliminate : 0.000008s : 0.07% optimize.opt_b.updatestate_assign_eliminate : 0.000005s : 0.04% optimize.opt_b.updatestate_loads_eliminate : 0.000004s : 0.04% optimize.opt_b.renormalize : 0.000000s : 0.00% optimize.opt_b.cse : 0.000046s : 0.40% optimize.optimize_parallel_all_gather_comm : 0.000021s : 0.18% optimize.overlap_param_gather : 0.000003s : 0.03% optimize.cconv : 0.000025s : 0.22% optimize.loop_unroll : 0.000520s : 4.53% optimize.opt_after_cconv.c_1 : 0.000063s : 0.55% optimize.opt_after_cconv.parameter_eliminate : 0.000002s : 0.02% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000007s : 0.06% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000004s : 0.04% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000004s : 0.03% optimize.opt_after_cconv.cse : 0.000040s : 0.35% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000043s : 0.37% optimize.tuple_transform.d_1 : 0.000088s : 0.77% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.02% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000013s : 0.11% optimize.partial_unused_args_eliminate : 0.000002s : 0.02% optimize.add_recomputation : 0.000063s : 0.55% optimize.cse_after_recomputation.cse : 0.000024s : 0.21% optimize.environ_conv : 0.000008s : 0.07% optimize.swap_dp_allreduce_reducescatter : 0.000032s : 0.28% optimize.bias_add_comm_swap : 0.000003s : 0.02% optimize.label_micro_interleaved_index : 0.000005s : 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 : 0.02% optimize.micro_interleaved_order_control : 0.000002s : 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 : 0.000003s : 0.02% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.03% optimize.comm_op_add_attrs : 0.000001s : 0.01% optimize.add_comm_op_reuse_tag : 0.000001s : 0.01% optimize.interleave_split_concat_branches : 0.000002s : 0.01% optimize.interleave_parallel_branches : 0.000001s : 0.01% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.02% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.02% optimize.control_data_broadcast_order : 0.000017s : 0.15% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.02% optimize.offloading_packed_experts : 0.000005s : 0.04% optimize.overlap_recompute_and_grad_model_parallel : 0.000006s : 0.05% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 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 : 0.000005s : 0.05% optimize.overlap_grad_flash_sp : 0.000022s : 0.19% optimize.begin_end_overlap_inline : 0.000001s : 0.01% 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 : 0.000074s : 0.64% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000017s : 0.15% optimize.symbol_engine_optimizer.elim_not_effective : 0.000022s : 0.19% optimize.symbol_engine_optimizer.opt_reshape : 0.000013s : 0.11% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000020s : 0.17% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.02% pipeline_parallel_scheduler : 0.000002s : 0.01% auto_monad_reorder : 0.000029s : 0.25% get_jit_bprop_graph : 0.000001s : 0.01% rewriter_after_jit_bprop_graph : 0.000003s : 0.03% opt_after_jit_grad : 0.000514s : 4.48% validate : 0.000054s : 0.47% Time group info: ------[substitution.] 0.000125 39 3.42% : 0.000004s : 2: substitution.elim_not_effective 4.46% : 0.000006s : 2: substitution.fold_const_symbol 6.67% : 0.000008s : 10: substitution.graph_param_transform 71.14% : 0.000089s : 1: substitution.inline 2.61% : 0.000003s : 4: substitution.j_node_and_user_rematch 3.72% : 0.000005s : 4: substitution.remove_not_recompute_node 7.97% : 0.000010s : 16: substitution.replace_old_param ------[type_inference.] 0.005026 2 89.77% : 0.004512s : 1: type_inference.infer 10.23% : 0.000514s : 1: type_inference.specialize ------[replace.] 0.000016 1 100.00% : 0.000016s : 1: replace.inline ------[match.] 0.000088 1 100.00% : 0.000088s : 1: match.inline ------[predicate.] 0.000297 2335 0.92% : 0.000003s : 21: predicate.accumulaten_eliminater 0.73% : 0.000002s : 10: predicate.ad_related_special_op_eliminate 0.78% : 0.000002s : 20: predicate.addn_check_dump 0.85% : 0.000003s : 21: predicate.addn_zero_filter 0.79% : 0.000002s : 21: predicate.adjust_all_reduce_mul_add 1.75% : 0.000005s : 41: predicate.arithmetic_simplify 0.86% : 0.000003s : 21: predicate.cast_eliminate 0.84% : 0.000002s : 20: predicate.check_bprop_eliminate 0.77% : 0.000002s : 20: predicate.compare_switch_simplify 0.34% : 0.000001s : 10: predicate.const_output_eliminate 0.80% : 0.000002s : 20: predicate.depend_value_elim 0.89% : 0.000003s : 21: predicate.dict_get_item_const_eliminator 0.90% : 0.000003s : 21: predicate.dict_get_item_eliminator 0.84% : 0.000002s : 21: predicate.dict_set_item_eliminator 1.06% : 0.000003s : 20: predicate.dumpgradient_eliminate 0.35% : 0.000001s : 10: predicate.elim_not_effective 0.48% : 0.000001s : 10: predicate.elim_shapecalc_of_broadcastargs 1.23% : 0.000004s : 31: predicate.environ_add_const_eliminate 1.19% : 0.000004s : 31: predicate.environ_get_add_eliminate 1.19% : 0.000004s : 31: predicate.environ_get_depend_swap 2.06% : 0.000006s : 51: predicate.environ_get_eliminate 1.18% : 0.000003s : 31: predicate.environ_get_set_eliminate 0.87% : 0.000003s : 22: predicate.exchange_switch_depend_value 1.35% : 0.000004s : 22: predicate.float_depend_g_call 0.77% : 0.000002s : 20: predicate.float_environ_get_switch 1.15% : 0.000003s : 30: predicate.float_tuple_getitem_switch 0.32% : 0.000001s : 10: predicate.fold_const_symbol 0.87% : 0.000003s : 20: predicate.get_grad_eliminate 0.44% : 0.000001s : 10: predicate.graph_param_transform 0.80% : 0.000002s : 20: predicate.incorporate_call 0.70% : 0.000002s : 20: predicate.incorporate_call_switch 5.28% : 0.000016s : 103: predicate.inline 0.91% : 0.000003s : 20: predicate.inline_without_move 0.60% : 0.000002s : 20: predicate.j_node_and_user_rematch 0.91% : 0.000003s : 20: predicate.less_batch_normalization 1.76% : 0.000005s : 41: predicate.list_to_tuple_eliminator_ 2.44% : 0.000007s : 62: predicate.load_eliminater 0.83% : 0.000002s : 10: predicate.loop_unroll_after_grad 1.36% : 0.000004s : 31: predicate.loop_unroll_before_grad 1.79% : 0.000005s : 41: predicate.make_slice_get_slice_eliminator 0.79% : 0.000002s : 20: predicate.merge_addn 0.80% : 0.000002s : 20: predicate.micro_step_allgather_replace 0.80% : 0.000002s : 20: predicate.mini_step_allgather_replace 0.81% : 0.000002s : 21: predicate.minmaximum_grad 0.99% : 0.000003s : 10: predicate.mutable_eliminate 0.45% : 0.000001s : 10: predicate.opt_reshape 0.38% : 0.000001s : 10: predicate.parallel_virtual_node 1.02% : 0.000003s : 22: predicate.partial_defer_inline 1.39% : 0.000004s : 31: predicate.partial_eliminate 0.85% : 0.000003s : 21: predicate.print_const_string_wrapper 0.85% : 0.000003s : 20: predicate.reduce_all_const_elim 0.97% : 0.000003s : 21: predicate.reduce_eliminate 2.48% : 0.000007s : 62: predicate.redundant_stop_gradient_eliminater 0.72% : 0.000002s : 20: predicate.remove_not_recompute_node 1.59% : 0.000005s : 41: predicate.replace_applicator 0.73% : 0.000002s : 20: predicate.replace_old_param 0.38% : 0.000001s : 10: predicate.reset_defer_inline 0.87% : 0.000003s : 21: predicate.reshape_eliminate 0.84% : 0.000002s : 20: predicate.row_tensor_add_zeros_like 0.42% : 0.000001s : 10: predicate.row_tensor_eliminate 0.91% : 0.000003s : 20: predicate.same_eliminate 0.71% : 0.000002s : 20: predicate.set_cell_output_no_recompute 0.94% : 0.000003s : 20: predicate.shard_identity_eliminate 0.79% : 0.000002s : 20: predicate.special_op_eliminate 0.90% : 0.000003s : 20: predicate.specialize_transform 0.90% : 0.000003s : 20: predicate.split_environ_get_set_with_tuple_value 0.94% : 0.000003s : 20: predicate.stack_unstack_eliminate 0.46% : 0.000001s : 10: predicate.switch_call_monad_eliminater 0.92% : 0.000003s : 22: predicate.switch_defer_inline 1.72% : 0.000005s : 42: predicate.switch_layer_defer_inline 3.81% : 0.000011s : 83: predicate.switch_simplify 0.84% : 0.000002s : 21: predicate.tile_eliminate 0.85% : 0.000003s : 21: predicate.transpose_eliminate 1.53% : 0.000005s : 41: predicate.tuple_list_convert_item_index_to_positive 1.61% : 0.000005s : 41: predicate.tuple_list_get_item_const_eliminator 1.50% : 0.000004s : 41: predicate.tuple_list_get_item_depend_reorder 2.73% : 0.000008s : 61: predicate.tuple_list_get_item_eliminator 1.55% : 0.000005s : 41: predicate.tuple_list_get_set_item_eliminator 2.32% : 0.000007s : 61: predicate.tuple_list_set_item_eliminator 2.10% : 0.000006s : 41: predicate.tuple_to_list_eliminator_ 2.37% : 0.000007s : 62: predicate.updatestate_pure_node_eliminater 3.33% : 0.000010s : 82: predicate.updatestate_useless_node_eliminater 1.57% : 0.000005s : 10: predicate.value_based_eliminate 0.88% : 0.000003s : 20: predicate.virtual_dataset_eliminate 0.86% : 0.000003s : 20: predicate.virtual_output_eliminate 0.44% : 0.000001s : 10: predicate.virtual_view_grad_eliminate 0.49% : 0.000001s : 10: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000495 6 52.53% : 0.000260s : 3: func_graph_cloner_run.FuncGraphClonerGraph 47.47% : 0.000235s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.028723 192 0.01% : 0.000004s : 1: ForceFp32Comm 13.62% : 0.003912s : 1: add_attr 13.58% : 0.003900s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.24% : 0.000068s : 1: add_recomputation 0.01% : 0.000004s : 1: assign_add_opt 0.22% : 0.000064s : 1: auto_monad 0.12% : 0.000033s : 1: auto_monad_reorder 0.01% : 0.000004s : 1: begin_end_overlap_inline 0.02% : 0.000006s : 1: bias_add_comm_swap 2.74% : 0.000786s : 1: bootstrap 0.10% : 0.000029s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.07% : 0.000021s : 1: control_data_broadcast_order 0.05% : 0.000015s : 1: convert_after_rewriter 0.13% : 0.000038s : 1: cse_after_recomputation 0.02% : 0.000005s : 1: dataset_repeat_opt 0.02% : 0.000005s : 1: detach_backward 0.04% : 0.000012s : 1: environ_conv 0.07% : 0.000019s : 1: event_method 0.02% : 0.000006s : 1: full_micro_interleaved_order_control 0.01% : 0.000004s : 1: get_jit_bprop_graph 0.03% : 0.000009s : 1: graph_reusing 0.02% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.02% : 0.000004s : 1: handle_group_info 0.02% : 0.000005s : 1: inline 0.02% : 0.000006s : 1: insert-virtual-dataset 0.02% : 0.000005s : 1: interleave_parallel_branches 0.02% : 0.000005s : 1: interleave_split_concat_branches 0.02% : 0.000007s : 1: label_fine_grained_interleaved_index 0.03% : 0.000008s : 1: label_micro_interleaved_index 1.84% : 0.000530s : 1: loop_unroll 0.02% : 0.000005s : 1: merge_cast_opt 0.02% : 0.000006s : 1: micro_interleaved_order_control 1.76% : 0.000505s : 1: mutable_eliminate 0.03% : 0.000008s : 1: offloading_packed_experts 0.07% : 0.000021s : 1: opt.transform.loop_unroll_optimizer 0.07% : 0.000021s : 1: opt.transform.mutable_eliminate 5.60% : 0.001608s : 78: opt.transform.opt_a 0.22% : 0.000062s : 1: opt.transform.opt_after_cconv 0.15% : 0.000042s : 1: opt.transform.opt_after_jit_grad 0.81% : 0.000233s : 28: opt.transform.opt_b 0.34% : 0.000099s : 2: opt.transform.opt_trans_graph 0.23% : 0.000067s : 4: opt.transform.symbol_engine_opt 11.06% : 0.003177s : 1: opt_a 0.59% : 0.000169s : 1: opt_after_cconv 1.82% : 0.000524s : 1: opt_after_jit_grad 1.31% : 0.000377s : 1: opt_b 20.20% : 0.005801s : 1: optimize 0.09% : 0.000025s : 1: optimize_parallel_all_gather_comm 0.04% : 0.000011s : 1: order_py_execute_after_rewriter 0.09% : 0.000025s : 1: overlap_grad_flash_sp 0.02% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.03% : 0.000010s : 1: overlap_grad_ring_attention 0.02% : 0.000007s : 1: overlap_opt_shard_grad_in_pipeline 0.02% : 0.000006s : 1: overlap_opt_shard_in_pipeline 0.02% : 0.000007s : 1: overlap_param_gather 0.02% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.03% : 0.000009s : 1: overlap_recompute_and_grad_model_parallel 0.02% : 0.000006s : 1: overlap_recompute_comm 0.02% : 0.000006s : 1: parallel-infer-symbol 0.01% : 0.000004s : 1: parallel-infer-symbol-second 0.02% : 0.000005s : 1: partial_unused_args_eliminate 0.02% : 0.000005s : 1: pipeline_parallel_scheduler 0.02% : 0.000005s : 1: pipeline_split 0.13% : 0.000036s : 1: pre_auto_parallel 0.10% : 0.000028s : 1: py_interpret_to_execute 0.06% : 0.000016s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000004s : 1: remove_cast_before_assign_add 0.16% : 0.000047s : 1: remove_dup_value 0.92% : 0.000265s : 1: renormalize.infer 1.03% : 0.000295s : 1: renormalize.specialize 0.02% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.02% : 0.000007s : 1: rewriter_after_jit_bprop_graph 0.20% : 0.000057s : 1: rewriter_after_opt_a 0.24% : 0.000068s : 1: rewriter_before_opt_a 0.02% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.02% : 0.000005s : 1: slice_recompute_activation 0.02% : 0.000005s : 1: split_layernorm_comm 0.02% : 0.000006s : 1: split_matmul_comm_elemetwise 0.13% : 0.000037s : 1: swap_dp_allreduce_reducescatter 0.65% : 0.000187s : 1: symbol_engine_optimizer 0.46% : 0.000132s : 1: tuple_transform 17.83% : 0.005121s : 1: type_inference . [hook] pytest_runtest_teardown:test_paged_attention_large_gsq_pertoken[QuantMethod.INT_CUBE-0] tests/st/infer/ops/test_internal_ops/test_paged_attention.py::test_paged_attention_large_gsq_pertoken[QuantMethod.INT_CUBE-0],max_mem:520.0M [WARNING] ME(163462:281473583890224,MainProcess):2026-01-29-17:48:50.787.982 [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 = 0.0177346, [21] [bootstrap]: 0.0007061 [type_inference]: 0.00542864 [event_method]: 1.399e-05 [auto_monad]: 5.591e-05 [graph_reusing]: 5.03002e-06 [inline]: 1.78002e-06 [add_attr]: 0.00404124, [1] [add_attr_with_inline]: 0.00402944, [1] [Cycle 1]: 5.647e-05, [2] [tag_attr]: 1.897e-05 [meta_addattr_fg_expand]: 4.25e-06 [parallel-infer-symbol]: 3.52002e-06 [pre_auto_parallel]: 3.66e-05 [insert-virtual-dataset]: 2.50002e-06 [parallel-infer-symbol-second]: 8.39995e-07 [dataset_repeat_opt]: 2.07999e-06 [pipeline_split]: 2.14e-06 [optimize]: 0.00658001, [53] [py_interpret_to_execute]: 2.651e-05 [rewriter_before_opt_a]: 6.843e-05 [opt_a]: 0.00332367, [2] [Cycle 1]: 0.00217452, [45] [expand_dump_flag]: 3.27002e-06 [switch_simplify]: 3.298e-05 [loop_unroll]: 2.131e-05 [a_1]: 0.00049753 [with_stream_mark]: 1.98e-05 [recompute_prepare]: 1.393e-05 [updatestate_depend_eliminate]: 6.01e-06 [updatestate_assign_eliminate]: 4.99e-06 [updatestate_loads_eliminate]: 4.87e-06 [parameter_eliminate]: 1.76e-06 [a_2]: 0.00015948 [accelerated_algorithm]: 1.254e-05 [shard]: 2.27999e-06 [meta_shard_fg_expand]: 2.43e-06 [shard_inline]: 1.198e-05 [merge_send_recv]: 9.81e-06 [auto_parallel]: 8.01001e-06 [parallel]: 2.782e-05 [flash_sp]: 9.62001e-06 [merge_comm]: 5.64998e-06 [allreduce_fusion]: 5.47999e-06 [matmul_add_comm_reduction]: 1.165e-05 [allreduce_slice_to_reducescatter]: 6.50005e-07 [virtual_shard_identity]: 1.417e-05 [virtual_dataset]: 1.176e-05 [get_grad_eliminate_]: 1.152e-05 [virtual_output]: 1.219e-05 [merge_forward]: 6.06e-06 [cell_reuse_recompute_pass]: 1.43002e-06 [offload_activation]: 1.255e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.832e-05 [merge_recompute_call_nodes]: 1.49e-06 [before_grad]: 1.909e-05 [set_forward_comm_id_for_comm_node_pass]: 5.39e-06 [meta_fg_expand]: 3.97e-06 [flash_sp_send_recv_attached]: 2.68003e-06 [receive_attached]: 2.50997e-06 [after_resolve]: 2.531e-05 [a_after_grad]: 1.846e-05 [renormalize]: 0.00067962 [add_forward_monad_depend]: 5.36998e-06 [auto_monad_grad]: 2.05002e-06 [auto_monad_eliminator]: 1.808e-05 [cse]: 6.914e-05 [a_3]: 8.642e-05 [Cycle 2]: 0.00113737, [45] [expand_dump_flag]: 1.57999e-06 [switch_simplify]: 1.36e-05 [loop_unroll]: 1.209e-05 [a_1]: 0.00031196 [with_stream_mark]: 1.314e-05 [recompute_prepare]: 1.219e-05 [updatestate_depend_eliminate]: 5.13002e-06 [updatestate_assign_eliminate]: 4.27e-06 [updatestate_loads_eliminate]: 5.60001e-06 [parameter_eliminate]: 9.89996e-07 [a_2]: 0.00015278 [accelerated_algorithm]: 1.217e-05 [shard]: 1.54e-06 [meta_shard_fg_expand]: 1.76003e-06 [shard_inline]: 1.193e-05 [merge_send_recv]: 6.97002e-06 [auto_parallel]: 7.89002e-06 [parallel]: 4.97e-06 [flash_sp]: 3.55e-06 [merge_comm]: 4.91002e-06 [allreduce_fusion]: 4.58001e-06 [matmul_add_comm_reduction]: 7.33999e-06 [allreduce_slice_to_reducescatter]: 3.4002e-07 [virtual_shard_identity]: 1.331e-05 [virtual_dataset]: 1.177e-05 [get_grad_eliminate_]: 1.155e-05 [virtual_output]: 1.148e-05 [merge_forward]: 4.64998e-06 [cell_reuse_recompute_pass]: 1.47001e-06 [offload_activation]: 9.44998e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.857e-05 [merge_recompute_call_nodes]: 8.30012e-07 [before_grad]: 1.525e-05 [set_forward_comm_id_for_comm_node_pass]: 5.39e-06 [meta_fg_expand]: 3.27997e-06 [flash_sp_send_recv_attached]: 1.05999e-06 [receive_attached]: 1.92001e-06 [after_resolve]: 2.436e-05 [a_after_grad]: 1.825e-05 [renormalize]: 8.00064e-08 [add_forward_monad_depend]: 1.22e-06 [auto_monad_grad]: 1.10999e-06 [auto_monad_eliminator]: 9.52001e-06 [cse]: 4.637e-05 [a_3]: 7.554e-05 [py_interpret_to_execute_after_opt_a]: 1.731e-05 [slice_cell_reuse_recomputed_activation]: 2.35002e-06 [rewriter_after_opt_a]: 5.851e-05 [convert_after_rewriter]: 9.92001e-06 [order_py_execute_after_rewriter]: 6.95002e-06 [mutable_eliminate]: 0.00097849 [opt_b]: 0.00039353, [1] [Cycle 1]: 0.00038315, [7] [b_1]: 0.00025123 [b_2]: 1.379e-05 [updatestate_depend_eliminate]: 7.78999e-06 [updatestate_assign_eliminate]: 4.42e-06 [updatestate_loads_eliminate]: 4.35e-06 [renormalize]: 4.39992e-07 [cse]: 5.228e-05 [optimize_parallel_all_gather_comm]: 2.149e-05 [overlap_param_gather]: 2.94999e-06 [cconv]: 2.688e-05 [loop_unroll]: 0.00054404 [opt_after_cconv]: 0.00018426, [1] [Cycle 1]: 0.0001753, [7] [c_1]: 6.451e-05 [parameter_eliminate]: 2.50002e-06 [updatestate_depend_eliminate]: 7.46999e-06 [updatestate_assign_eliminate]: 4.45e-06 [updatestate_loads_eliminate]: 4.38001e-06 [cse]: 4.337e-05 [renormalize]: 6.39993e-07 [remove_dup_value]: 5.227e-05 [tuple_transform]: 0.00013541, [1] [Cycle 1]: 0.00012957, [4] [d_1]: 9.004e-05 [none_parameter_eliminate]: 2.16998e-06 [renormalize]: 3.30008e-07 [switch_simplify]: 1.305e-05 [partial_unused_args_eliminate]: 2.26e-06 [add_recomputation]: 6.512e-05 [cse_after_recomputation]: 3.86e-05, [1] [Cycle 1]: 3.363e-05, [1] [cse]: 2.483e-05 [environ_conv]: 7.58999e-06 [swap_dp_allreduce_reducescatter]: 1.049e-05 [bias_add_comm_swap]: 2.49001e-06 [label_micro_interleaved_index]: 4.70999e-06 [label_fine_grained_interleaved_index]: 2.93998e-06 [merge_cast_opt]: 1.66e-06 [slice_recompute_activation]: 2.17999e-06 [micro_interleaved_order_control]: 2.54001e-06 [assign_add_opt]: 1.45999e-06 [ForceFp32Comm]: 8.39995e-07 [remove_cast_before_assign_add]: 8.70001e-07 [full_micro_interleaved_order_control]: 2.44001e-06 [reorder_send_recv_between_fp_bp]: 3.12997e-06 [comm_op_add_attrs]: 1.06002e-06 [add_comm_op_reuse_tag]: 1.02e-06 [interleave_split_concat_branches]: 1.38002e-06 [interleave_parallel_branches]: 1.52001e-06 [overlap_opt_shard_in_pipeline]: 2.26e-06 [overlap_opt_shard_grad_in_pipeline]: 1.91998e-06 [control_data_broadcast_order]: 1.675e-05 [grouped_pairwise_exchange_alltoall]: 1.53002e-06 [offloading_packed_experts]: 5.25999e-06 [overlap_recompute_and_grad_model_parallel]: 5.82001e-06 [overlap_grad_matmul_and_grad_allreduce]: 3.78001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.52999e-06 [overlap_recompute_comm]: 2.59999e-06 [overlap_grad_ring_attention]: 5.30999e-06 [overlap_grad_flash_sp]: 2.315e-05 [begin_end_overlap_inline]: 5.3001e-07 [split_matmul_comm_elemetwise]: 2.66999e-06 [split_layernorm_comm]: 2.26998e-06 [handle_group_info]: 1.14e-06 [symbol_engine_optimizer]: 0.00020077, [1] [Cycle 1]: 0.00019578, [6] [build]: 7.855e-05 [elim_shapecalc]: 1.727e-05 [elim_not_effective]: 2.125e-05 [opt_reshape]: 1.27e-05 [fold_const_symbol]: 2.375e-05 [renormalize]: 3.09985e-07 [detach_backward]: 1.81003e-06 [pipeline_parallel_scheduler]: 1.49e-06 [auto_monad_reorder]: 2.948e-05 [get_jit_bprop_graph]: 1.21002e-06 [rewriter_after_jit_bprop_graph]: 3.72998e-06 [opt_after_jit_grad]: 0.00058781 [validate]: 6.147e-05 Sums bootstrap : 0.000706s : 5.64% type_inference : 0.005429s : 43.39% event_method : 0.000014s : 0.11% auto_monad : 0.000056s : 0.45% graph_reusing : 0.000005s : 0.04% inline : 0.000002s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000019s : 0.15% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.03% parallel-infer-symbol : 0.000004s : 0.03% pre_auto_parallel : 0.000037s : 0.29% insert-virtual-dataset : 0.000003s : 0.02% parallel-infer-symbol-second : 0.000001s : 0.01% dataset_repeat_opt : 0.000002s : 0.02% pipeline_split : 0.000002s : 0.02% optimize.py_interpret_to_execute : 0.000027s : 0.21% optimize.rewriter_before_opt_a : 0.000068s : 0.55% optimize.opt_a.expand_dump_flag : 0.000005s : 0.04% optimize.opt_a.switch_simplify : 0.000047s : 0.37% optimize.opt_a.loop_unroll : 0.000033s : 0.27% optimize.opt_a.a_1 : 0.000809s : 6.47% optimize.opt_a.with_stream_mark : 0.000033s : 0.26% optimize.opt_a.recompute_prepare : 0.000026s : 0.21% optimize.opt_a.updatestate_depend_eliminate : 0.000011s : 0.09% optimize.opt_a.updatestate_assign_eliminate : 0.000009s : 0.07% optimize.opt_a.updatestate_loads_eliminate : 0.000010s : 0.08% optimize.opt_a.parameter_eliminate : 0.000003s : 0.02% optimize.opt_a.a_2 : 0.000312s : 2.50% optimize.opt_a.accelerated_algorithm : 0.000025s : 0.20% optimize.opt_a.shard : 0.000004s : 0.03% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.03% optimize.opt_a.shard_inline : 0.000024s : 0.19% optimize.opt_a.merge_send_recv : 0.000017s : 0.13% optimize.opt_a.auto_parallel : 0.000016s : 0.13% optimize.opt_a.parallel : 0.000033s : 0.26% optimize.opt_a.flash_sp : 0.000013s : 0.11% optimize.opt_a.merge_comm : 0.000011s : 0.08% optimize.opt_a.allreduce_fusion : 0.000010s : 0.08% optimize.opt_a.matmul_add_comm_reduction : 0.000019s : 0.15% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000027s : 0.22% optimize.opt_a.virtual_dataset : 0.000024s : 0.19% optimize.opt_a.get_grad_eliminate_ : 0.000023s : 0.18% optimize.opt_a.virtual_output : 0.000024s : 0.19% optimize.opt_a.merge_forward : 0.000011s : 0.09% optimize.opt_a.cell_reuse_recompute_pass : 0.000003s : 0.02% optimize.opt_a.offload_activation : 0.000022s : 0.18% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000037s : 0.29% optimize.opt_a.merge_recompute_call_nodes : 0.000002s : 0.02% optimize.opt_a.before_grad : 0.000034s : 0.27% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000011s : 0.09% optimize.opt_a.meta_fg_expand : 0.000007s : 0.06% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.03% optimize.opt_a.receive_attached : 0.000004s : 0.04% optimize.opt_a.after_resolve : 0.000050s : 0.40% optimize.opt_a.a_after_grad : 0.000037s : 0.29% optimize.opt_a.renormalize : 0.000680s : 5.43% optimize.opt_a.add_forward_monad_depend : 0.000007s : 0.05% optimize.opt_a.auto_monad_grad : 0.000003s : 0.03% optimize.opt_a.auto_monad_eliminator : 0.000028s : 0.22% optimize.opt_a.cse : 0.000116s : 0.92% optimize.opt_a.a_3 : 0.000162s : 1.29% optimize.py_interpret_to_execute_after_opt_a : 0.000017s : 0.14% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.02% optimize.rewriter_after_opt_a : 0.000059s : 0.47% optimize.convert_after_rewriter : 0.000010s : 0.08% optimize.order_py_execute_after_rewriter : 0.000007s : 0.06% optimize.mutable_eliminate : 0.000978s : 7.82% optimize.opt_b.b_1 : 0.000251s : 2.01% optimize.opt_b.b_2 : 0.000014s : 0.11% optimize.opt_b.updatestate_depend_eliminate : 0.000008s : 0.06% optimize.opt_b.updatestate_assign_eliminate : 0.000004s : 0.04% optimize.opt_b.updatestate_loads_eliminate : 0.000004s : 0.03% optimize.opt_b.renormalize : 0.000000s : 0.00% optimize.opt_b.cse : 0.000052s : 0.42% optimize.optimize_parallel_all_gather_comm : 0.000021s : 0.17% optimize.overlap_param_gather : 0.000003s : 0.02% optimize.cconv : 0.000027s : 0.21% optimize.loop_unroll : 0.000544s : 4.35% optimize.opt_after_cconv.c_1 : 0.000065s : 0.52% optimize.opt_after_cconv.parameter_eliminate : 0.000003s : 0.02% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000007s : 0.06% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000004s : 0.04% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000004s : 0.04% optimize.opt_after_cconv.cse : 0.000043s : 0.35% optimize.opt_after_cconv.renormalize : 0.000001s : 0.01% optimize.remove_dup_value : 0.000052s : 0.42% optimize.tuple_transform.d_1 : 0.000090s : 0.72% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.02% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000013s : 0.10% optimize.partial_unused_args_eliminate : 0.000002s : 0.02% optimize.add_recomputation : 0.000065s : 0.52% optimize.cse_after_recomputation.cse : 0.000025s : 0.20% optimize.environ_conv : 0.000008s : 0.06% optimize.swap_dp_allreduce_reducescatter : 0.000010s : 0.08% optimize.bias_add_comm_swap : 0.000002s : 0.02% optimize.label_micro_interleaved_index : 0.000005s : 0.04% 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 : 0.000001s : 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 : 0.03% 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 : 0.000002s : 0.01% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.02% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.02% optimize.control_data_broadcast_order : 0.000017s : 0.13% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.01% optimize.offloading_packed_experts : 0.000005s : 0.04% optimize.overlap_recompute_and_grad_model_parallel : 0.000006s : 0.05% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000004s : 0.03% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.01% optimize.overlap_recompute_comm : 0.000003s : 0.02% optimize.overlap_grad_ring_attention : 0.000005s : 0.04% optimize.overlap_grad_flash_sp : 0.000023s : 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 : 0.02% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000079s : 0.63% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000017s : 0.14% optimize.symbol_engine_optimizer.elim_not_effective : 0.000021s : 0.17% optimize.symbol_engine_optimizer.opt_reshape : 0.000013s : 0.10% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000024s : 0.19% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.01% pipeline_parallel_scheduler : 0.000001s : 0.01% auto_monad_reorder : 0.000029s : 0.24% get_jit_bprop_graph : 0.000001s : 0.01% rewriter_after_jit_bprop_graph : 0.000004s : 0.03% opt_after_jit_grad : 0.000588s : 4.70% validate : 0.000061s : 0.49% Time group info: ------[substitution.] 0.000142 39 2.97% : 0.000004s : 2: substitution.elim_not_effective 4.99% : 0.000007s : 2: substitution.fold_const_symbol 6.07% : 0.000009s : 10: substitution.graph_param_transform 72.77% : 0.000103s : 1: substitution.inline 2.33% : 0.000003s : 4: substitution.j_node_and_user_rematch 3.54% : 0.000005s : 4: substitution.remove_not_recompute_node 7.32% : 0.000010s : 16: substitution.replace_old_param ------[type_inference.] 0.005305 2 87.93% : 0.004665s : 1: type_inference.infer 12.07% : 0.000640s : 1: type_inference.specialize ------[replace.] 0.000017 1 100.00% : 0.000017s : 1: replace.inline ------[match.] 0.000102 1 100.00% : 0.000102s : 1: match.inline ------[predicate.] 0.000297 2335 0.91% : 0.000003s : 21: predicate.accumulaten_eliminater 0.76% : 0.000002s : 10: predicate.ad_related_special_op_eliminate 0.78% : 0.000002s : 20: predicate.addn_check_dump 0.90% : 0.000003s : 21: predicate.addn_zero_filter 0.81% : 0.000002s : 21: predicate.adjust_all_reduce_mul_add 1.94% : 0.000006s : 41: predicate.arithmetic_simplify 0.89% : 0.000003s : 21: predicate.cast_eliminate 0.86% : 0.000003s : 20: predicate.check_bprop_eliminate 0.78% : 0.000002s : 20: predicate.compare_switch_simplify 0.32% : 0.000001s : 10: predicate.const_output_eliminate 0.81% : 0.000002s : 20: predicate.depend_value_elim 0.89% : 0.000003s : 21: predicate.dict_get_item_const_eliminator 0.91% : 0.000003s : 21: predicate.dict_get_item_eliminator 0.85% : 0.000003s : 21: predicate.dict_set_item_eliminator 1.22% : 0.000004s : 20: predicate.dumpgradient_eliminate 0.34% : 0.000001s : 10: predicate.elim_not_effective 0.47% : 0.000001s : 10: predicate.elim_shapecalc_of_broadcastargs 1.24% : 0.000004s : 31: predicate.environ_add_const_eliminate 1.16% : 0.000003s : 31: predicate.environ_get_add_eliminate 1.19% : 0.000004s : 31: predicate.environ_get_depend_swap 2.11% : 0.000006s : 51: predicate.environ_get_eliminate 1.23% : 0.000004s : 31: predicate.environ_get_set_eliminate 0.86% : 0.000003s : 22: predicate.exchange_switch_depend_value 1.46% : 0.000004s : 22: predicate.float_depend_g_call 0.76% : 0.000002s : 20: predicate.float_environ_get_switch 1.16% : 0.000003s : 30: predicate.float_tuple_getitem_switch 0.33% : 0.000001s : 10: predicate.fold_const_symbol 0.84% : 0.000002s : 20: predicate.get_grad_eliminate 0.42% : 0.000001s : 10: predicate.graph_param_transform 0.78% : 0.000002s : 20: predicate.incorporate_call 0.70% : 0.000002s : 20: predicate.incorporate_call_switch 5.20% : 0.000015s : 103: predicate.inline 0.91% : 0.000003s : 20: predicate.inline_without_move 0.59% : 0.000002s : 20: predicate.j_node_and_user_rematch 0.93% : 0.000003s : 20: predicate.less_batch_normalization 1.72% : 0.000005s : 41: predicate.list_to_tuple_eliminator_ 2.43% : 0.000007s : 62: predicate.load_eliminater 0.80% : 0.000002s : 10: predicate.loop_unroll_after_grad 1.29% : 0.000004s : 31: predicate.loop_unroll_before_grad 1.73% : 0.000005s : 41: predicate.make_slice_get_slice_eliminator 0.81% : 0.000002s : 20: predicate.merge_addn 0.79% : 0.000002s : 20: predicate.micro_step_allgather_replace 0.81% : 0.000002s : 20: predicate.mini_step_allgather_replace 0.85% : 0.000003s : 21: predicate.minmaximum_grad 0.87% : 0.000003s : 10: predicate.mutable_eliminate 0.43% : 0.000001s : 10: predicate.opt_reshape 0.39% : 0.000001s : 10: predicate.parallel_virtual_node 1.01% : 0.000003s : 22: predicate.partial_defer_inline 1.37% : 0.000004s : 31: predicate.partial_eliminate 0.89% : 0.000003s : 21: predicate.print_const_string_wrapper 0.81% : 0.000002s : 20: predicate.reduce_all_const_elim 0.97% : 0.000003s : 21: predicate.reduce_eliminate 2.41% : 0.000007s : 62: predicate.redundant_stop_gradient_eliminater 0.80% : 0.000002s : 20: predicate.remove_not_recompute_node 1.56% : 0.000005s : 41: predicate.replace_applicator 0.72% : 0.000002s : 20: predicate.replace_old_param 0.38% : 0.000001s : 10: predicate.reset_defer_inline 0.97% : 0.000003s : 21: predicate.reshape_eliminate 0.85% : 0.000003s : 20: predicate.row_tensor_add_zeros_like 0.48% : 0.000001s : 10: predicate.row_tensor_eliminate 0.97% : 0.000003s : 20: predicate.same_eliminate 0.67% : 0.000002s : 20: predicate.set_cell_output_no_recompute 0.99% : 0.000003s : 20: predicate.shard_identity_eliminate 0.83% : 0.000002s : 20: predicate.special_op_eliminate 0.89% : 0.000003s : 20: predicate.specialize_transform 0.94% : 0.000003s : 20: predicate.split_environ_get_set_with_tuple_value 0.96% : 0.000003s : 20: predicate.stack_unstack_eliminate 0.43% : 0.000001s : 10: predicate.switch_call_monad_eliminater 0.92% : 0.000003s : 22: predicate.switch_defer_inline 1.66% : 0.000005s : 42: predicate.switch_layer_defer_inline 3.73% : 0.000011s : 83: predicate.switch_simplify 0.89% : 0.000003s : 21: predicate.tile_eliminate 0.82% : 0.000002s : 21: predicate.transpose_eliminate 1.54% : 0.000005s : 41: predicate.tuple_list_convert_item_index_to_positive 1.64% : 0.000005s : 41: predicate.tuple_list_get_item_const_eliminator 1.50% : 0.000004s : 41: predicate.tuple_list_get_item_depend_reorder 2.78% : 0.000008s : 61: predicate.tuple_list_get_item_eliminator 1.58% : 0.000005s : 41: predicate.tuple_list_get_set_item_eliminator 2.33% : 0.000007s : 61: predicate.tuple_list_set_item_eliminator 2.06% : 0.000006s : 41: predicate.tuple_to_list_eliminator_ 2.33% : 0.000007s : 62: predicate.updatestate_pure_node_eliminater 3.34% : 0.000010s : 82: predicate.updatestate_useless_node_eliminater 0.90% : 0.000003s : 10: predicate.value_based_eliminate 0.85% : 0.000003s : 20: predicate.virtual_dataset_eliminate 1.03% : 0.000003s : 20: predicate.virtual_output_eliminate 0.42% : 0.000001s : 10: predicate.virtual_view_grad_eliminate 0.55% : 0.000002s : 10: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000540 6 50.13% : 0.000271s : 3: func_graph_cloner_run.FuncGraphClonerGraph 49.87% : 0.000269s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.031024 192 0.01% : 0.000004s : 1: ForceFp32Comm 13.04% : 0.004047s : 1: add_attr 13.00% : 0.004033s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.22% : 0.000070s : 1: add_recomputation 0.02% : 0.000005s : 1: assign_add_opt 0.20% : 0.000061s : 1: auto_monad 0.11% : 0.000034s : 1: auto_monad_reorder 0.02% : 0.000006s : 1: begin_end_overlap_inline 0.02% : 0.000006s : 1: bias_add_comm_swap 2.40% : 0.000745s : 1: bootstrap 0.11% : 0.000034s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.07% : 0.000020s : 1: control_data_broadcast_order 0.04% : 0.000014s : 1: convert_after_rewriter 0.14% : 0.000043s : 1: cse_after_recomputation 0.02% : 0.000005s : 1: dataset_repeat_opt 0.02% : 0.000005s : 1: detach_backward 0.04% : 0.000011s : 1: environ_conv 0.06% : 0.000020s : 1: event_method 0.02% : 0.000006s : 1: full_micro_interleaved_order_control 0.02% : 0.000007s : 1: get_jit_bprop_graph 0.03% : 0.000009s : 1: graph_reusing 0.02% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000004s : 1: handle_group_info 0.02% : 0.000005s : 1: inline 0.02% : 0.000006s : 1: insert-virtual-dataset 0.01% : 0.000005s : 1: interleave_parallel_branches 0.01% : 0.000004s : 1: interleave_split_concat_branches 0.02% : 0.000007s : 1: label_fine_grained_interleaved_index 0.03% : 0.000008s : 1: label_micro_interleaved_index 1.79% : 0.000554s : 1: loop_unroll 0.02% : 0.000007s : 1: merge_cast_opt 0.02% : 0.000006s : 1: micro_interleaved_order_control 3.19% : 0.000990s : 1: mutable_eliminate 0.03% : 0.000009s : 1: offloading_packed_experts 0.07% : 0.000021s : 1: opt.transform.loop_unroll_optimizer 0.07% : 0.000023s : 1: opt.transform.mutable_eliminate 5.17% : 0.001605s : 78: opt.transform.opt_a 0.20% : 0.000063s : 1: opt.transform.opt_after_cconv 0.15% : 0.000047s : 1: opt.transform.opt_after_jit_grad 0.76% : 0.000235s : 28: opt.transform.opt_b 0.32% : 0.000101s : 2: opt.transform.opt_trans_graph 0.23% : 0.000071s : 4: opt.transform.symbol_engine_opt 10.73% : 0.003328s : 1: opt_a 0.61% : 0.000188s : 1: opt_after_cconv 1.93% : 0.000599s : 1: opt_after_jit_grad 1.28% : 0.000398s : 1: opt_b 21.23% : 0.006585s : 1: optimize 0.09% : 0.000028s : 1: optimize_parallel_all_gather_comm 0.04% : 0.000011s : 1: order_py_execute_after_rewriter 0.09% : 0.000027s : 1: overlap_grad_flash_sp 0.02% : 0.000007s : 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 0.02% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.02% : 0.000007s : 1: overlap_param_gather 0.02% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.03% : 0.000009s : 1: overlap_recompute_and_grad_model_parallel 0.03% : 0.000010s : 1: overlap_recompute_comm 0.02% : 0.000007s : 1: parallel-infer-symbol 0.01% : 0.000004s : 1: parallel-infer-symbol-second 0.02% : 0.000006s : 1: partial_unused_args_eliminate 0.02% : 0.000005s : 1: pipeline_parallel_scheduler 0.02% : 0.000005s : 1: pipeline_split 0.13% : 0.000041s : 1: pre_auto_parallel 0.10% : 0.000031s : 1: py_interpret_to_execute 0.07% : 0.000022s : 1: py_interpret_to_execute_after_opt_a 0.02% : 0.000006s : 1: remove_cast_before_assign_add 0.19% : 0.000059s : 1: remove_dup_value 1.00% : 0.000310s : 1: renormalize.infer 1.16% : 0.000361s : 1: renormalize.specialize 0.02% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.02% : 0.000007s : 1: rewriter_after_jit_bprop_graph 0.20% : 0.000063s : 1: rewriter_after_opt_a 0.24% : 0.000073s : 1: rewriter_before_opt_a 0.02% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.02% : 0.000005s : 1: slice_recompute_activation 0.02% : 0.000006s : 1: split_layernorm_comm 0.02% : 0.000006s : 1: split_matmul_comm_elemetwise 0.05% : 0.000014s : 1: swap_dp_allreduce_reducescatter 0.66% : 0.000204s : 1: symbol_engine_optimizer 0.45% : 0.000139s : 1: tuple_transform 17.55% : 0.005445s : 1: type_inference . [hook] pytest_runtest_teardown:test_paged_attention_large_gsq_pertoken[QuantMethod.INT_CUBE-1] tests/st/infer/ops/test_internal_ops/test_paged_attention.py::test_paged_attention_large_gsq_pertoken[QuantMethod.INT_CUBE-1],max_mem:520.0M =============================== 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 ================== 4 passed, 25 warnings in 746.11s (0:12:26) ==================