==================================================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_003/sault/config/pytest.ini plugins: mock-3.14.0, hydra-core-1.3.2, forked-1.6.0, anyio-4.9.0, xdist-1.32.0 collected 6 items test_qbmm_split.py [WARNING] ME(166984:281473373798192,MainProcess):2026-01-29-17:37:50.956.757 [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 = 1.91237, [21] [bootstrap]: 0.00058483 [type_inference]: 1.71536 [event_method]: 1.891e-05 [auto_monad]: 0.0012967 [graph_reusing]: 7.52998e-06 [inline]: 2.89999e-06 [add_attr]: 0.0677784, [1] [add_attr_with_inline]: 0.0677631, [1] [Cycle 1]: 0.00010179, [2] [tag_attr]: 2.483e-05 [meta_addattr_fg_expand]: 5.44e-06 [parallel-infer-symbol]: 3.66001e-06 [pre_auto_parallel]: 4.956e-05 [insert-virtual-dataset]: 2.78e-06 [parallel-infer-symbol-second]: 7.50006e-07 [dataset_repeat_opt]: 2.32001e-06 [pipeline_split]: 1.64998e-06 [optimize]: 0.126226, [53] [py_interpret_to_execute]: 2.902e-05 [rewriter_before_opt_a]: 0.00010185 [opt_a]: 0.121919, [2] [Cycle 1]: 0.0335537, [45] [expand_dump_flag]: 2.81e-06 [switch_simplify]: 4.052e-05 [loop_unroll]: 2.671e-05 [a_1]: 0.00080088 [with_stream_mark]: 2.658e-05 [recompute_prepare]: 2.174e-05 [updatestate_depend_eliminate]: 1.185e-05 [updatestate_assign_eliminate]: 1.348e-05 [updatestate_loads_eliminate]: 1.679e-05 [parameter_eliminate]: 2.29001e-06 [a_2]: 0.00032403 [accelerated_algorithm]: 4.735e-05 [shard]: 2.22001e-06 [meta_shard_fg_expand]: 4.2e-06 [shard_inline]: 2.032e-05 [merge_send_recv]: 0.00011479 [auto_parallel]: 1.592e-05 [parallel]: 5.311e-05 [flash_sp]: 1.244e-05 [merge_comm]: 1.197e-05 [allreduce_fusion]: 1.071e-05 [matmul_add_comm_reduction]: 1.991e-05 [allreduce_slice_to_reducescatter]: 8.50006e-07 [virtual_shard_identity]: 2.605e-05 [virtual_dataset]: 2.04e-05 [get_grad_eliminate_]: 1.984e-05 [virtual_output]: 2.036e-05 [merge_forward]: 1.126e-05 [cell_reuse_recompute_pass]: 1.60001e-06 [offload_activation]: 2.194e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.852e-05 [merge_recompute_call_nodes]: 1.84e-06 [before_grad]: 3.314e-05 [set_forward_comm_id_for_comm_node_pass]: 1.208e-05 [meta_fg_expand]: 7.7e-06 [flash_sp_send_recv_attached]: 4.73001e-06 [receive_attached]: 2.23998e-06 [after_resolve]: 2.921e-05 [a_after_grad]: 3.255e-05 [renormalize]: 0.0199955 [add_forward_monad_depend]: 1.292e-05 [auto_monad_grad]: 2.89999e-06 [auto_monad_eliminator]: 7.331e-05 [cse]: 0.00025535 [a_3]: 0.00025406 [Cycle 2]: 0.0883488, [45] [expand_dump_flag]: 5.56002e-06 [switch_simplify]: 2.628e-05 [loop_unroll]: 2.212e-05 [a_1]: 0.00066444 [with_stream_mark]: 3.968e-05 [recompute_prepare]: 2.509e-05 [updatestate_depend_eliminate]: 1.373e-05 [updatestate_assign_eliminate]: 1.109e-05 [updatestate_loads_eliminate]: 1.929e-05 [parameter_eliminate]: 2.84001e-06 [a_2]: 0.00030103 [accelerated_algorithm]: 2.883e-05 [shard]: 2.71e-06 [meta_shard_fg_expand]: 9.15999e-06 [shard_inline]: 2.008e-05 [merge_send_recv]: 2.043e-05 [auto_parallel]: 1.811e-05 [parallel]: 1.236e-05 [flash_sp]: 4.02e-06 [merge_comm]: 1.105e-05 [allreduce_fusion]: 1.025e-05 [matmul_add_comm_reduction]: 2.132e-05 [allreduce_slice_to_reducescatter]: 1.37999e-06 [virtual_shard_identity]: 2.459e-05 [virtual_dataset]: 2.064e-05 [get_grad_eliminate_]: 1.893e-05 [virtual_output]: 2.07e-05 [merge_forward]: 1.24e-05 [cell_reuse_recompute_pass]: 3.39001e-06 [offload_activation]: 2.149e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.914e-05 [merge_recompute_call_nodes]: 1.82999e-06 [before_grad]: 3.278e-05 [set_forward_comm_id_for_comm_node_pass]: 1.192e-05 [meta_fg_expand]: 8.76002e-06 [flash_sp_send_recv_attached]: 2.32999e-06 [receive_attached]: 2.64001e-06 [after_resolve]: 3.015e-05 [a_after_grad]: 0.0861504 [renormalize]: 4.10015e-07 [add_forward_monad_depend]: 1.089e-05 [auto_monad_grad]: 3.21001e-06 [auto_monad_eliminator]: 9.065e-05 [cse]: 0.00010475 [a_3]: 0.00014782 [py_interpret_to_execute_after_opt_a]: 3.622e-05 [slice_cell_reuse_recomputed_activation]: 3.03998e-06 [rewriter_after_opt_a]: 0.00023874 [convert_after_rewriter]: 1.787e-05 [order_py_execute_after_rewriter]: 1.268e-05 [mutable_eliminate]: 0.00081164 [opt_b]: 0.000704, [1] [Cycle 1]: 0.00069478, [7] [b_1]: 0.00049269 [b_2]: 2.334e-05 [updatestate_depend_eliminate]: 1.623e-05 [updatestate_assign_eliminate]: 1.166e-05 [updatestate_loads_eliminate]: 1.757e-05 [renormalize]: 8.10018e-07 [cse]: 9.297e-05 [optimize_parallel_all_gather_comm]: 3.801e-05 [overlap_param_gather]: 1.97001e-06 [cconv]: 4.029e-05 [loop_unroll]: 0.00049566 [opt_after_cconv]: 0.00028681, [1] [Cycle 1]: 0.00027935, [7] [c_1]: 0.00013245 [parameter_eliminate]: 3.88999e-06 [updatestate_depend_eliminate]: 1.396e-05 [updatestate_assign_eliminate]: 9.72999e-06 [updatestate_loads_eliminate]: 1.353e-05 [cse]: 6.72e-05 [renormalize]: 5.89993e-07 [remove_dup_value]: 9.308e-05 [tuple_transform]: 0.00019449, [1] [Cycle 1]: 0.00018833, [4] [d_1]: 0.00014344 [none_parameter_eliminate]: 2.48e-06 [renormalize]: 1.8999e-07 [switch_simplify]: 2.224e-05 [partial_unused_args_eliminate]: 2.02001e-06 [add_recomputation]: 0.00012972 [cse_after_recomputation]: 7.444e-05, [1] [Cycle 1]: 6.912e-05, [1] [cse]: 6.155e-05 [environ_conv]: 3.464e-05 [swap_dp_allreduce_reducescatter]: 1.655e-05 [bias_add_comm_swap]: 3.32002e-06 [label_micro_interleaved_index]: 4.80999e-06 [label_fine_grained_interleaved_index]: 3.43999e-06 [merge_cast_opt]: 1.47999e-06 [slice_recompute_activation]: 2.29999e-06 [micro_interleaved_order_control]: 2.31e-06 [assign_add_opt]: 1.57999e-06 [ForceFp32Comm]: 9.09989e-07 [remove_cast_before_assign_add]: 1.27999e-06 [full_micro_interleaved_order_control]: 2.16e-06 [reorder_send_recv_between_fp_bp]: 2.94001e-06 [comm_op_add_attrs]: 1.05999e-06 [add_comm_op_reuse_tag]: 9.89996e-07 [interleave_split_concat_branches]: 1.20999e-06 [interleave_parallel_branches]: 1.29e-06 [overlap_opt_shard_in_pipeline]: 2.358e-05 [overlap_opt_shard_grad_in_pipeline]: 2.09e-06 [control_data_broadcast_order]: 0.00032504 [grouped_pairwise_exchange_alltoall]: 2.07999e-06 [offloading_packed_experts]: 1.248e-05 [overlap_recompute_and_grad_model_parallel]: 1.063e-05 [overlap_grad_matmul_and_grad_allreduce]: 1.22e-06 [overlap_recompute_allgather_and_fa_grad]: 1.32e-06 [overlap_recompute_comm]: 2.50002e-06 [overlap_grad_ring_attention]: 8.79e-06 [overlap_grad_flash_sp]: 6.35e-05 [begin_end_overlap_inline]: 5.69999e-07 [split_matmul_comm_elemetwise]: 2.36998e-06 [split_layernorm_comm]: 1.78002e-06 [handle_group_info]: 1.02e-06 [symbol_engine_optimizer]: 0.00018497, [1] [Cycle 1]: 0.0001784, [6] [build]: 1.843e-05 [elim_shapecalc]: 2.994e-05 [elim_not_effective]: 3.736e-05 [opt_reshape]: 2.638e-05 [fold_const_symbol]: 3.153e-05 [renormalize]: 2.89991e-07 [detach_backward]: 2.47001e-06 [pipeline_parallel_scheduler]: 1.64e-06 [auto_monad_reorder]: 7.311e-05 [get_jit_bprop_graph]: 1.79998e-06 [rewriter_after_jit_bprop_graph]: 5.51998e-06 [opt_after_jit_grad]: 0.00063784 [validate]: 0.00011391 Sums bootstrap : 0.000585s : 0.03% type_inference : 1.715362s : 93.60% event_method : 0.000019s : 0.00% auto_monad : 0.001297s : 0.07% graph_reusing : 0.000008s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000025s : 0.00% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000005s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000050s : 0.00% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000029s : 0.00% optimize.rewriter_before_opt_a : 0.000102s : 0.01% optimize.opt_a.expand_dump_flag : 0.000008s : 0.00% optimize.opt_a.switch_simplify : 0.000067s : 0.00% optimize.opt_a.loop_unroll : 0.000049s : 0.00% optimize.opt_a.a_1 : 0.001465s : 0.08% optimize.opt_a.with_stream_mark : 0.000066s : 0.00% optimize.opt_a.recompute_prepare : 0.000047s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000026s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000025s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000036s : 0.00% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000625s : 0.03% optimize.opt_a.accelerated_algorithm : 0.000076s : 0.00% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000013s : 0.00% optimize.opt_a.shard_inline : 0.000040s : 0.00% optimize.opt_a.merge_send_recv : 0.000135s : 0.01% optimize.opt_a.auto_parallel : 0.000034s : 0.00% optimize.opt_a.parallel : 0.000065s : 0.00% optimize.opt_a.flash_sp : 0.000016s : 0.00% optimize.opt_a.merge_comm : 0.000023s : 0.00% optimize.opt_a.allreduce_fusion : 0.000021s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000041s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000051s : 0.00% optimize.opt_a.virtual_dataset : 0.000041s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000039s : 0.00% optimize.opt_a.virtual_output : 0.000041s : 0.00% optimize.opt_a.merge_forward : 0.000024s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000043s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000078s : 0.00% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000066s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000024s : 0.00% optimize.opt_a.meta_fg_expand : 0.000016s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000007s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000059s : 0.00% optimize.opt_a.a_after_grad : 0.086183s : 4.70% optimize.opt_a.renormalize : 0.019996s : 1.09% optimize.opt_a.add_forward_monad_depend : 0.000024s : 0.00% optimize.opt_a.auto_monad_grad : 0.000006s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000164s : 0.01% optimize.opt_a.cse : 0.000360s : 0.02% optimize.opt_a.a_3 : 0.000402s : 0.02% optimize.py_interpret_to_execute_after_opt_a : 0.000036s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000239s : 0.01% optimize.convert_after_rewriter : 0.000018s : 0.00% optimize.order_py_execute_after_rewriter : 0.000013s : 0.00% optimize.mutable_eliminate : 0.000812s : 0.04% optimize.opt_b.b_1 : 0.000493s : 0.03% optimize.opt_b.b_2 : 0.000023s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000016s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000012s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000018s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000093s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000038s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000040s : 0.00% optimize.loop_unroll : 0.000496s : 0.03% optimize.opt_after_cconv.c_1 : 0.000132s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000004s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000014s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000010s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000014s : 0.00% optimize.opt_after_cconv.cse : 0.000067s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000093s : 0.01% optimize.tuple_transform.d_1 : 0.000143s : 0.01% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000022s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000130s : 0.01% optimize.cse_after_recomputation.cse : 0.000062s : 0.00% optimize.environ_conv : 0.000035s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000017s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000024s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000325s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000012s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000011s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000009s : 0.00% optimize.overlap_grad_flash_sp : 0.000064s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000018s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000030s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000037s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000026s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000032s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000073s : 0.00% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.000638s : 0.03% validate : 0.000114s : 0.01% Time group info: ------[substitution.] 0.000394 135 2.95% : 0.000012s : 2: substitution.depend_value_elim 1.21% : 0.000005s : 11: substitution.elim_not_effective 1.23% : 0.000005s : 11: substitution.fold_const_symbol 3.85% : 0.000015s : 18: substitution.graph_param_transform 29.78% : 0.000117s : 1: substitution.inline 2.78% : 0.000011s : 22: substitution.j_node_and_user_rematch 6.08% : 0.000024s : 2: substitution.less_batch_normalization 2.74% : 0.000011s : 18: substitution.load_eliminater 0.62% : 0.000002s : 2: substitution.opt_reshape 4.36% : 0.000017s : 22: substitution.remove_not_recompute_node 2.40% : 0.000009s : 8: substitution.replace_old_param 7.95% : 0.000031s : 4: substitution.reshape_eliminate 2.77% : 0.000011s : 6: substitution.updatestate_pure_node_eliminater 31.31% : 0.000123s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 1.715247 2 99.75% : 1.710973s : 1: type_inference.infer 0.25% : 0.004274s : 1: type_inference.specialize ------[replace.] 0.000017 1 100.00% : 0.000017s : 1: replace.inline ------[match.] 0.000116 1 100.00% : 0.000116s : 1: match.inline ------[predicate.] 0.086739 4131 0.01% : 0.000005s : 37: predicate.accumulaten_eliminater 0.01% : 0.000005s : 18: predicate.ad_related_special_op_eliminate 0.01% : 0.000005s : 36: predicate.addn_check_dump 0.01% : 0.000004s : 37: predicate.addn_zero_filter 0.00% : 0.000004s : 37: predicate.adjust_all_reduce_mul_add 0.01% : 0.000012s : 73: predicate.arithmetic_simplify 0.01% : 0.000005s : 37: predicate.cast_eliminate 0.01% : 0.000005s : 36: predicate.check_bprop_eliminate 0.01% : 0.000004s : 36: predicate.compare_switch_simplify 0.00% : 0.000001s : 18: predicate.const_output_eliminate 0.01% : 0.000005s : 36: predicate.depend_value_elim 0.01% : 0.000005s : 37: predicate.dict_get_item_const_eliminator 0.01% : 0.000005s : 37: predicate.dict_get_item_eliminator 0.01% : 0.000005s : 37: predicate.dict_set_item_eliminator 0.01% : 0.000006s : 36: predicate.dumpgradient_eliminate 0.00% : 0.000002s : 18: predicate.elim_not_effective 0.00% : 0.000003s : 18: predicate.elim_shapecalc_of_broadcastargs 0.01% : 0.000007s : 55: predicate.environ_add_const_eliminate 0.01% : 0.000006s : 55: predicate.environ_get_add_eliminate 0.01% : 0.000007s : 55: predicate.environ_get_depend_swap 0.01% : 0.000011s : 91: predicate.environ_get_eliminate 0.01% : 0.000006s : 55: predicate.environ_get_set_eliminate 0.01% : 0.000005s : 38: predicate.exchange_switch_depend_value 0.01% : 0.000007s : 38: predicate.float_depend_g_call 0.01% : 0.000005s : 36: predicate.float_environ_get_switch 0.01% : 0.000007s : 54: predicate.float_tuple_getitem_switch 0.00% : 0.000001s : 18: predicate.fold_const_symbol 0.01% : 0.000005s : 36: predicate.get_grad_eliminate 0.00% : 0.000002s : 18: predicate.graph_param_transform 0.01% : 0.000005s : 36: predicate.incorporate_call 0.00% : 0.000004s : 36: predicate.incorporate_call_switch 0.03% : 0.000030s : 183: predicate.inline 99.23% : 0.086070s : 36: predicate.inline_without_move 0.00% : 0.000003s : 36: predicate.j_node_and_user_rematch 0.01% : 0.000007s : 36: predicate.less_batch_normalization 0.01% : 0.000009s : 73: predicate.list_to_tuple_eliminator_ 0.02% : 0.000014s : 110: predicate.load_eliminater 0.01% : 0.000005s : 18: predicate.loop_unroll_after_grad 0.01% : 0.000006s : 41: predicate.loop_unroll_before_grad 0.01% : 0.000010s : 73: predicate.make_slice_get_slice_eliminator 0.01% : 0.000005s : 36: predicate.merge_addn 0.01% : 0.000005s : 36: predicate.micro_step_allgather_replace 0.01% : 0.000005s : 36: predicate.mini_step_allgather_replace 0.00% : 0.000004s : 37: predicate.minmaximum_grad 0.00% : 0.000004s : 18: predicate.mutable_eliminate 0.00% : 0.000003s : 18: predicate.opt_reshape 0.00% : 0.000003s : 18: predicate.parallel_virtual_node 0.01% : 0.000005s : 38: predicate.partial_defer_inline 0.01% : 0.000007s : 55: predicate.partial_eliminate 0.00% : 0.000004s : 37: predicate.print_const_string_wrapper 0.01% : 0.000005s : 36: predicate.reduce_all_const_elim 0.01% : 0.000006s : 37: predicate.reduce_eliminate 0.01% : 0.000013s : 110: predicate.redundant_stop_gradient_eliminater 0.00% : 0.000003s : 36: predicate.remove_not_recompute_node 0.01% : 0.000008s : 73: predicate.replace_applicator 0.00% : 0.000003s : 36: predicate.replace_old_param 0.00% : 0.000002s : 18: predicate.reset_defer_inline 0.01% : 0.000005s : 37: predicate.reshape_eliminate 0.01% : 0.000006s : 36: predicate.row_tensor_add_zeros_like 0.00% : 0.000003s : 18: predicate.row_tensor_eliminate 0.07% : 0.000058s : 36: predicate.same_eliminate 0.00% : 0.000003s : 36: predicate.set_cell_output_no_recompute 0.01% : 0.000005s : 36: predicate.shard_identity_eliminate 0.01% : 0.000005s : 36: predicate.special_op_eliminate 0.01% : 0.000005s : 36: predicate.specialize_transform 0.01% : 0.000008s : 36: predicate.split_environ_get_set_with_tuple_value 0.01% : 0.000010s : 36: predicate.stack_unstack_eliminate 0.00% : 0.000002s : 18: predicate.switch_call_monad_eliminater 0.01% : 0.000005s : 38: predicate.switch_defer_inline 0.01% : 0.000009s : 74: predicate.switch_layer_defer_inline 0.02% : 0.000019s : 133: predicate.switch_simplify 0.01% : 0.000005s : 37: predicate.tile_eliminate 0.00% : 0.000004s : 37: predicate.transpose_eliminate 0.01% : 0.000009s : 73: predicate.tuple_list_convert_item_index_to_positive 0.01% : 0.000009s : 73: predicate.tuple_list_get_item_const_eliminator 0.05% : 0.000047s : 73: predicate.tuple_list_get_item_depend_reorder 0.02% : 0.000016s : 109: predicate.tuple_list_get_item_eliminator 0.01% : 0.000009s : 73: predicate.tuple_list_get_set_item_eliminator 0.02% : 0.000014s : 109: predicate.tuple_list_set_item_eliminator 0.01% : 0.000009s : 73: predicate.tuple_to_list_eliminator_ 0.01% : 0.000013s : 110: predicate.updatestate_pure_node_eliminater 0.05% : 0.000043s : 146: predicate.updatestate_useless_node_eliminater 0.00% : 0.000003s : 18: predicate.value_based_eliminate 0.01% : 0.000005s : 36: predicate.virtual_dataset_eliminate 0.01% : 0.000006s : 36: predicate.virtual_output_eliminate 0.00% : 0.000002s : 18: predicate.virtual_view_grad_eliminate 0.00% : 0.000003s : 18: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.010507 58 73.22% : 0.007693s : 55: func_graph_cloner_run.FuncGraphClonerGraph 26.78% : 0.002814s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 2.216347 192 0.00% : 0.000004s : 1: ForceFp32Comm 3.06% : 0.067784s : 1: add_attr 3.06% : 0.067768s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000134s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.06% : 0.001318s : 1: auto_monad 0.00% : 0.000078s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.03% : 0.000616s : 1: bootstrap 0.00% : 0.000044s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.02% : 0.000336s : 1: control_data_broadcast_order 0.00% : 0.000023s : 1: convert_after_rewriter 0.00% : 0.000078s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000039s : 1: environ_conv 0.00% : 0.000027s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000014s : 1: graph_reusing 0.00% : 0.000006s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.02% : 0.000504s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.04% : 0.000822s : 1: mutable_eliminate 0.00% : 0.000016s : 1: offloading_packed_experts 0.00% : 0.000031s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000036s : 1: opt.transform.mutable_eliminate 4.02% : 0.089193s : 78: opt.transform.opt_a 0.01% : 0.000131s : 1: opt.transform.opt_after_cconv 0.00% : 0.000067s : 1: opt.transform.opt_after_jit_grad 0.02% : 0.000480s : 28: opt.transform.opt_b 0.01% : 0.000163s : 2: opt.transform.opt_trans_graph 0.01% : 0.000120s : 4: opt.transform.symbol_engine_opt 5.50% : 0.121924s : 1: opt_a 0.01% : 0.000291s : 1: opt_after_cconv 0.03% : 0.000649s : 1: opt_after_jit_grad 0.03% : 0.000708s : 1: opt_b 5.70% : 0.126232s : 1: optimize 0.00% : 0.000042s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000016s : 1: order_py_execute_after_rewriter 0.00% : 0.000068s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000012s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000027s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000014s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000007s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.00% : 0.000054s : 1: pre_auto_parallel 0.00% : 0.000033s : 1: py_interpret_to_execute 0.00% : 0.000040s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000098s : 1: remove_dup_value 0.76% : 0.016741s : 1: renormalize.infer 0.15% : 0.003237s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000244s : 1: rewriter_after_opt_a 0.00% : 0.000106s : 1: rewriter_before_opt_a 0.00% : 0.000007s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000020s : 1: swap_dp_allreduce_reducescatter 0.01% : 0.000188s : 1: symbol_engine_optimizer 0.01% : 0.000198s : 1: tuple_transform 77.40% : 1.715392s : 1: type_inference . [hook] pytest_runtest_teardown:test_qbmm_qkv_1408_128_128[False-1024-32] tests/st/infer/ops/test_internal_ops/test_qbmm_split.py::test_qbmm_qkv_1408_128_128[False-1024-32],max_mem:100.0M [WARNING] ME(166984:281473373798192,MainProcess):2026-01-29-17:38:15.829.807 [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 = 4.61725, [21] [bootstrap]: 0.00063153 [type_inference]: 4.13253 [event_method]: 2.216e-05 [auto_monad]: 0.00049788 [graph_reusing]: 7.2e-06 [inline]: 4.17e-06 [add_attr]: 0.142849, [1] [add_attr_with_inline]: 0.142829, [1] [Cycle 1]: 8.947e-05, [2] [tag_attr]: 3.31e-05 [meta_addattr_fg_expand]: 5.28002e-06 [parallel-infer-symbol]: 3.97998e-06 [pre_auto_parallel]: 4.215e-05 [insert-virtual-dataset]: 2.96999e-06 [parallel-infer-symbol-second]: 7.30011e-07 [dataset_repeat_opt]: 2.43998e-06 [pipeline_split]: 1.96998e-06 [optimize]: 0.339163, [53] [py_interpret_to_execute]: 4.616e-05 [rewriter_before_opt_a]: 0.00014158 [opt_a]: 0.169757, [2] [Cycle 1]: 0.167246, [45] [expand_dump_flag]: 3.28998e-06 [switch_simplify]: 4.671e-05 [loop_unroll]: 2.494e-05 [a_1]: 0.00120384 [with_stream_mark]: 0.00013022 [recompute_prepare]: 6.541e-05 [updatestate_depend_eliminate]: 2.142e-05 [updatestate_assign_eliminate]: 1.69e-05 [updatestate_loads_eliminate]: 2.132e-05 [parameter_eliminate]: 3.12002e-06 [a_2]: 0.00040627 [accelerated_algorithm]: 7.023e-05 [shard]: 3.78001e-06 [meta_shard_fg_expand]: 1.152e-05 [shard_inline]: 2.216e-05 [merge_send_recv]: 1.938e-05 [auto_parallel]: 2.149e-05 [parallel]: 3.76e-05 [flash_sp]: 1.687e-05 [merge_comm]: 1.211e-05 [allreduce_fusion]: 1.167e-05 [matmul_add_comm_reduction]: 2.531e-05 [allreduce_slice_to_reducescatter]: 1.14e-06 [virtual_shard_identity]: 3.4e-05 [virtual_dataset]: 2.928e-05 [get_grad_eliminate_]: 2.97e-05 [virtual_output]: 2.855e-05 [merge_forward]: 1.478e-05 [cell_reuse_recompute_pass]: 2.96001e-06 [offload_activation]: 4.884e-05 [cell_reuse_handle_not_recompute_node_pass]: 6.205e-05 [merge_recompute_call_nodes]: 2.39001e-06 [before_grad]: 5.142e-05 [set_forward_comm_id_for_comm_node_pass]: 1.83e-05 [meta_fg_expand]: 1.07e-05 [flash_sp_send_recv_attached]: 6.48e-06 [receive_attached]: 2.57001e-06 [after_resolve]: 4.175e-05 [a_after_grad]: 5.465e-05 [renormalize]: 0.16158 [add_forward_monad_depend]: 2.06e-05 [auto_monad_grad]: 4.08999e-06 [auto_monad_eliminator]: 8.855e-05 [cse]: 0.00223268 [a_3]: 0.00020141 [Cycle 2]: 0.00249141, [45] [expand_dump_flag]: 4.62998e-06 [switch_simplify]: 2.449e-05 [loop_unroll]: 2.493e-05 [a_1]: 0.00082224 [with_stream_mark]: 4.406e-05 [recompute_prepare]: 2.75e-05 [updatestate_depend_eliminate]: 1.659e-05 [updatestate_assign_eliminate]: 1.268e-05 [updatestate_loads_eliminate]: 2.133e-05 [parameter_eliminate]: 2.93998e-06 [a_2]: 0.00037145 [accelerated_algorithm]: 3.102e-05 [shard]: 3.46999e-06 [meta_shard_fg_expand]: 9.19e-06 [shard_inline]: 2.185e-05 [merge_send_recv]: 2.165e-05 [auto_parallel]: 1.929e-05 [parallel]: 1.425e-05 [flash_sp]: 4.92e-06 [merge_comm]: 1.214e-05 [allreduce_fusion]: 1.105e-05 [matmul_add_comm_reduction]: 2.588e-05 [allreduce_slice_to_reducescatter]: 1.22e-06 [virtual_shard_identity]: 2.527e-05 [virtual_dataset]: 2.195e-05 [get_grad_eliminate_]: 2.109e-05 [virtual_output]: 2.177e-05 [merge_forward]: 1.172e-05 [cell_reuse_recompute_pass]: 3.41001e-06 [offload_activation]: 4.568e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.73e-05 [merge_recompute_call_nodes]: 2.53e-06 [before_grad]: 3.768e-05 [set_forward_comm_id_for_comm_node_pass]: 1.359e-05 [meta_fg_expand]: 1.067e-05 [flash_sp_send_recv_attached]: 2.46e-06 [receive_attached]: 2.80002e-06 [after_resolve]: 3.141e-05 [a_after_grad]: 3.445e-05 [renormalize]: 8.00064e-08 [add_forward_monad_depend]: 4.30999e-06 [auto_monad_grad]: 3.28998e-06 [auto_monad_eliminator]: 7.137e-05 [cse]: 9.905e-05 [a_3]: 0.00014284 [py_interpret_to_execute_after_opt_a]: 4.864e-05 [slice_cell_reuse_recomputed_activation]: 2.94999e-06 [rewriter_after_opt_a]: 0.00025027 [convert_after_rewriter]: 2.188e-05 [order_py_execute_after_rewriter]: 1.341e-05 [mutable_eliminate]: 0.00092159 [opt_b]: 0.00080447, [1] [Cycle 1]: 0.0007934, [7] [b_1]: 0.00056915 [b_2]: 2.896e-05 [updatestate_depend_eliminate]: 1.89e-05 [updatestate_assign_eliminate]: 1.13e-05 [updatestate_loads_eliminate]: 1.62e-05 [renormalize]: 1.10001e-06 [cse]: 9.936e-05 [optimize_parallel_all_gather_comm]: 4.433e-05 [overlap_param_gather]: 2.14999e-06 [cconv]: 4.492e-05 [loop_unroll]: 0.00166333 [opt_after_cconv]: 0.0003461, [1] [Cycle 1]: 0.00033733, [7] [c_1]: 0.00015666 [parameter_eliminate]: 7.98999e-06 [updatestate_depend_eliminate]: 1.811e-05 [updatestate_assign_eliminate]: 1.03e-05 [updatestate_loads_eliminate]: 1.618e-05 [cse]: 8.644e-05 [renormalize]: 8.30012e-07 [remove_dup_value]: 0.00012598 [tuple_transform]: 0.00021416, [1] [Cycle 1]: 0.00020763, [4] [d_1]: 0.00015864 [none_parameter_eliminate]: 2.69999e-06 [renormalize]: 2.30008e-07 [switch_simplify]: 2.306e-05 [partial_unused_args_eliminate]: 2.01e-06 [add_recomputation]: 0.163687 [cse_after_recomputation]: 0.00016413, [1] [Cycle 1]: 0.00014865, [1] [cse]: 0.00012303 [environ_conv]: 3.947e-05 [swap_dp_allreduce_reducescatter]: 1.844e-05 [bias_add_comm_swap]: 3.81001e-06 [label_micro_interleaved_index]: 9.07999e-06 [label_fine_grained_interleaved_index]: 2.78e-06 [merge_cast_opt]: 1.42e-06 [slice_recompute_activation]: 2.67001e-06 [micro_interleaved_order_control]: 2.66e-06 [assign_add_opt]: 2.73998e-06 [ForceFp32Comm]: 1.35001e-06 [remove_cast_before_assign_add]: 1.12e-06 [full_micro_interleaved_order_control]: 2.47001e-06 [reorder_send_recv_between_fp_bp]: 3.08998e-06 [comm_op_add_attrs]: 1.41002e-06 [add_comm_op_reuse_tag]: 1.07e-06 [interleave_split_concat_branches]: 1.42e-06 [interleave_parallel_branches]: 1.10001e-06 [overlap_opt_shard_in_pipeline]: 1.40999e-06 [overlap_opt_shard_grad_in_pipeline]: 2.31e-06 [control_data_broadcast_order]: 4.015e-05 [grouped_pairwise_exchange_alltoall]: 1.87001e-06 [offloading_packed_experts]: 1.039e-05 [overlap_recompute_and_grad_model_parallel]: 9.83002e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.30001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.48002e-06 [overlap_recompute_comm]: 2.59001e-06 [overlap_grad_ring_attention]: 1.175e-05 [overlap_grad_flash_sp]: 5.17e-05 [begin_end_overlap_inline]: 5.39992e-07 [split_matmul_comm_elemetwise]: 2.66e-06 [split_layernorm_comm]: 2.48e-06 [handle_group_info]: 1.24e-06 [symbol_engine_optimizer]: 0.00025313, [1] [Cycle 1]: 0.00024588, [6] [build]: 2.253e-05 [elim_shapecalc]: 5.777e-05 [elim_not_effective]: 5.366e-05 [opt_reshape]: 3.297e-05 [fold_const_symbol]: 3.559e-05 [renormalize]: 4.19997e-07 [detach_backward]: 2.29999e-06 [pipeline_parallel_scheduler]: 1.46002e-06 [auto_monad_reorder]: 8.429e-05 [get_jit_bprop_graph]: 2.22999e-06 [rewriter_after_jit_bprop_graph]: 9.52999e-06 [opt_after_jit_grad]: 0.00101023 [validate]: 0.00012167 Sums bootstrap : 0.000632s : 0.01% type_inference : 4.132531s : 92.39% event_method : 0.000022s : 0.00% auto_monad : 0.000498s : 0.01% graph_reusing : 0.000007s : 0.00% inline : 0.000004s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000033s : 0.00% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000005s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000042s : 0.00% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000046s : 0.00% optimize.rewriter_before_opt_a : 0.000142s : 0.00% optimize.opt_a.expand_dump_flag : 0.000008s : 0.00% optimize.opt_a.switch_simplify : 0.000071s : 0.00% optimize.opt_a.loop_unroll : 0.000050s : 0.00% optimize.opt_a.a_1 : 0.002026s : 0.05% optimize.opt_a.with_stream_mark : 0.000174s : 0.00% optimize.opt_a.recompute_prepare : 0.000093s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000038s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000030s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000043s : 0.00% optimize.opt_a.parameter_eliminate : 0.000006s : 0.00% optimize.opt_a.a_2 : 0.000778s : 0.02% optimize.opt_a.accelerated_algorithm : 0.000101s : 0.00% optimize.opt_a.shard : 0.000007s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000021s : 0.00% optimize.opt_a.shard_inline : 0.000044s : 0.00% optimize.opt_a.merge_send_recv : 0.000041s : 0.00% optimize.opt_a.auto_parallel : 0.000041s : 0.00% optimize.opt_a.parallel : 0.000052s : 0.00% optimize.opt_a.flash_sp : 0.000022s : 0.00% optimize.opt_a.merge_comm : 0.000024s : 0.00% optimize.opt_a.allreduce_fusion : 0.000023s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000051s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000059s : 0.00% optimize.opt_a.virtual_dataset : 0.000051s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000051s : 0.00% optimize.opt_a.virtual_output : 0.000050s : 0.00% optimize.opt_a.merge_forward : 0.000026s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000006s : 0.00% optimize.opt_a.offload_activation : 0.000095s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000109s : 0.00% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000089s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000032s : 0.00% optimize.opt_a.meta_fg_expand : 0.000021s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000009s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000073s : 0.00% optimize.opt_a.a_after_grad : 0.000089s : 0.00% optimize.opt_a.renormalize : 0.161580s : 3.61% optimize.opt_a.add_forward_monad_depend : 0.000025s : 0.00% optimize.opt_a.auto_monad_grad : 0.000007s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000160s : 0.00% optimize.opt_a.cse : 0.002332s : 0.05% optimize.opt_a.a_3 : 0.000344s : 0.01% optimize.py_interpret_to_execute_after_opt_a : 0.000049s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000250s : 0.01% optimize.convert_after_rewriter : 0.000022s : 0.00% optimize.order_py_execute_after_rewriter : 0.000013s : 0.00% optimize.mutable_eliminate : 0.000922s : 0.02% optimize.opt_b.b_1 : 0.000569s : 0.01% optimize.opt_b.b_2 : 0.000029s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000019s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000011s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000016s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000099s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000044s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000045s : 0.00% optimize.loop_unroll : 0.001663s : 0.04% optimize.opt_after_cconv.c_1 : 0.000157s : 0.00% optimize.opt_after_cconv.parameter_eliminate : 0.000008s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000018s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000010s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000016s : 0.00% optimize.opt_after_cconv.cse : 0.000086s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000126s : 0.00% optimize.tuple_transform.d_1 : 0.000159s : 0.00% optimize.tuple_transform.none_parameter_eliminate : 0.000003s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000023s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.163687s : 3.66% optimize.cse_after_recomputation.cse : 0.000123s : 0.00% optimize.environ_conv : 0.000039s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000018s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000009s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000003s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000040s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000010s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000010s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000012s : 0.00% optimize.overlap_grad_flash_sp : 0.000052s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000023s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000058s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000054s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000033s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000036s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000001s : 0.00% auto_monad_reorder : 0.000084s : 0.00% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000010s : 0.00% opt_after_jit_grad : 0.001010s : 0.02% validate : 0.000122s : 0.00% Time group info: ------[substitution.] 0.000463 135 2.93% : 0.000014s : 2: substitution.depend_value_elim 1.23% : 0.000006s : 11: substitution.elim_not_effective 0.90% : 0.000004s : 11: substitution.fold_const_symbol 3.14% : 0.000015s : 18: substitution.graph_param_transform 35.03% : 0.000162s : 1: substitution.inline 2.60% : 0.000012s : 22: substitution.j_node_and_user_rematch 6.44% : 0.000030s : 2: substitution.less_batch_normalization 2.37% : 0.000011s : 18: substitution.load_eliminater 0.65% : 0.000003s : 2: substitution.opt_reshape 4.41% : 0.000020s : 22: substitution.remove_not_recompute_node 2.23% : 0.000010s : 8: substitution.replace_old_param 8.38% : 0.000039s : 4: substitution.reshape_eliminate 2.49% : 0.000012s : 6: substitution.updatestate_pure_node_eliminater 27.19% : 0.000126s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 4.132397 2 99.79% : 4.123745s : 1: type_inference.infer 0.21% : 0.008652s : 1: type_inference.specialize ------[replace.] 0.000025 1 100.00% : 0.000025s : 1: replace.inline ------[match.] 0.000161 1 100.00% : 0.000161s : 1: match.inline ------[predicate.] 0.000638 4131 0.80% : 0.000005s : 37: predicate.accumulaten_eliminater 1.25% : 0.000008s : 18: predicate.ad_related_special_op_eliminate 0.74% : 0.000005s : 36: predicate.addn_check_dump 0.75% : 0.000005s : 37: predicate.addn_zero_filter 0.83% : 0.000005s : 37: predicate.adjust_all_reduce_mul_add 2.08% : 0.000013s : 73: predicate.arithmetic_simplify 0.84% : 0.000005s : 37: predicate.cast_eliminate 0.85% : 0.000005s : 36: predicate.check_bprop_eliminate 0.76% : 0.000005s : 36: predicate.compare_switch_simplify 0.25% : 0.000002s : 18: predicate.const_output_eliminate 0.92% : 0.000006s : 36: predicate.depend_value_elim 0.86% : 0.000005s : 37: predicate.dict_get_item_const_eliminator 0.91% : 0.000006s : 37: predicate.dict_get_item_eliminator 0.75% : 0.000005s : 37: predicate.dict_set_item_eliminator 1.19% : 0.000008s : 36: predicate.dumpgradient_eliminate 0.34% : 0.000002s : 18: predicate.elim_not_effective 0.75% : 0.000005s : 18: predicate.elim_shapecalc_of_broadcastargs 1.13% : 0.000007s : 55: predicate.environ_add_const_eliminate 1.09% : 0.000007s : 55: predicate.environ_get_add_eliminate 1.10% : 0.000007s : 55: predicate.environ_get_depend_swap 1.90% : 0.000012s : 91: predicate.environ_get_eliminate 1.11% : 0.000007s : 55: predicate.environ_get_set_eliminate 0.75% : 0.000005s : 38: predicate.exchange_switch_depend_value 1.28% : 0.000008s : 38: predicate.float_depend_g_call 0.86% : 0.000005s : 36: predicate.float_environ_get_switch 1.29% : 0.000008s : 54: predicate.float_tuple_getitem_switch 0.25% : 0.000002s : 18: predicate.fold_const_symbol 0.88% : 0.000006s : 36: predicate.get_grad_eliminate 0.27% : 0.000002s : 18: predicate.graph_param_transform 0.75% : 0.000005s : 36: predicate.incorporate_call 0.70% : 0.000004s : 36: predicate.incorporate_call_switch 5.26% : 0.000034s : 183: predicate.inline 1.14% : 0.000007s : 36: predicate.inline_without_move 0.43% : 0.000003s : 36: predicate.j_node_and_user_rematch 1.52% : 0.000010s : 36: predicate.less_batch_normalization 1.63% : 0.000010s : 73: predicate.list_to_tuple_eliminator_ 2.38% : 0.000015s : 110: predicate.load_eliminater 1.55% : 0.000010s : 18: predicate.loop_unroll_after_grad 0.97% : 0.000006s : 41: predicate.loop_unroll_before_grad 1.81% : 0.000012s : 73: predicate.make_slice_get_slice_eliminator 0.83% : 0.000005s : 36: predicate.merge_addn 0.87% : 0.000006s : 36: predicate.micro_step_allgather_replace 0.87% : 0.000006s : 36: predicate.mini_step_allgather_replace 0.70% : 0.000004s : 37: predicate.minmaximum_grad 1.22% : 0.000008s : 18: predicate.mutable_eliminate 0.56% : 0.000004s : 18: predicate.opt_reshape 0.48% : 0.000003s : 18: predicate.parallel_virtual_node 1.20% : 0.000008s : 38: predicate.partial_defer_inline 1.22% : 0.000008s : 55: predicate.partial_eliminate 0.81% : 0.000005s : 37: predicate.print_const_string_wrapper 0.83% : 0.000005s : 36: predicate.reduce_all_const_elim 1.14% : 0.000007s : 37: predicate.reduce_eliminate 2.26% : 0.000014s : 110: predicate.redundant_stop_gradient_eliminater 0.53% : 0.000003s : 36: predicate.remove_not_recompute_node 1.42% : 0.000009s : 73: predicate.replace_applicator 0.56% : 0.000004s : 36: predicate.replace_old_param 0.27% : 0.000002s : 18: predicate.reset_defer_inline 1.00% : 0.000006s : 37: predicate.reshape_eliminate 0.91% : 0.000006s : 36: predicate.row_tensor_add_zeros_like 0.46% : 0.000003s : 18: predicate.row_tensor_eliminate 1.20% : 0.000008s : 36: predicate.same_eliminate 0.68% : 0.000004s : 36: predicate.set_cell_output_no_recompute 0.91% : 0.000006s : 36: predicate.shard_identity_eliminate 0.91% : 0.000006s : 36: predicate.special_op_eliminate 0.92% : 0.000006s : 36: predicate.specialize_transform 1.09% : 0.000007s : 36: predicate.split_environ_get_set_with_tuple_value 1.52% : 0.000010s : 36: predicate.stack_unstack_eliminate 0.43% : 0.000003s : 18: predicate.switch_call_monad_eliminater 0.78% : 0.000005s : 38: predicate.switch_defer_inline 1.62% : 0.000010s : 74: predicate.switch_layer_defer_inline 3.37% : 0.000021s : 133: predicate.switch_simplify 0.76% : 0.000005s : 37: predicate.tile_eliminate 0.74% : 0.000005s : 37: predicate.transpose_eliminate 1.80% : 0.000012s : 73: predicate.tuple_list_convert_item_index_to_positive 1.63% : 0.000010s : 73: predicate.tuple_list_get_item_const_eliminator 1.44% : 0.000009s : 73: predicate.tuple_list_get_item_depend_reorder 2.72% : 0.000017s : 109: predicate.tuple_list_get_item_eliminator 1.59% : 0.000010s : 73: predicate.tuple_list_get_set_item_eliminator 2.54% : 0.000016s : 109: predicate.tuple_list_set_item_eliminator 1.52% : 0.000010s : 73: predicate.tuple_to_list_eliminator_ 2.32% : 0.000015s : 110: predicate.updatestate_pure_node_eliminater 3.38% : 0.000022s : 146: predicate.updatestate_useless_node_eliminater 0.50% : 0.000003s : 18: predicate.value_based_eliminate 0.91% : 0.000006s : 36: predicate.virtual_dataset_eliminate 0.84% : 0.000005s : 36: predicate.virtual_output_eliminate 0.38% : 0.000002s : 18: predicate.virtual_view_grad_eliminate 0.49% : 0.000003s : 18: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.058343 58 89.66% : 0.052310s : 55: func_graph_cloner_run.FuncGraphClonerGraph 10.34% : 0.006034s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 5.265667 192 0.00% : 0.000004s : 1: ForceFp32Comm 2.71% : 0.142859s : 1: add_attr 2.71% : 0.142834s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 3.11% : 0.163725s : 1: add_recomputation 0.00% : 0.000006s : 1: assign_add_opt 0.01% : 0.000514s : 1: auto_monad 0.00% : 0.000090s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.01% : 0.000680s : 1: bootstrap 0.00% : 0.000049s : 1: cconv 0.00% : 0.000005s : 1: comm_op_add_attrs 0.00% : 0.000044s : 1: control_data_broadcast_order 0.00% : 0.000026s : 1: convert_after_rewriter 0.00% : 0.000171s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000047s : 1: environ_conv 0.00% : 0.000031s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000008s : 1: get_jit_bprop_graph 0.00% : 0.000012s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000008s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000007s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000013s : 1: label_micro_interleaved_index 0.03% : 0.001680s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.02% : 0.000936s : 1: mutable_eliminate 0.00% : 0.000013s : 1: offloading_packed_experts 0.00% : 0.000049s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000044s : 1: opt.transform.mutable_eliminate 0.07% : 0.003923s : 78: opt.transform.opt_a 0.00% : 0.000154s : 1: opt.transform.opt_after_cconv 0.00% : 0.000090s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000539s : 28: opt.transform.opt_b 0.00% : 0.000178s : 2: opt.transform.opt_trans_graph 0.00% : 0.000171s : 4: opt.transform.symbol_engine_opt 3.22% : 0.169763s : 1: opt_a 0.01% : 0.000350s : 1: opt_after_cconv 0.02% : 0.001025s : 1: opt_after_jit_grad 0.02% : 0.000809s : 1: opt_b 6.44% : 0.339174s : 1: optimize 0.00% : 0.000050s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000017s : 1: order_py_execute_after_rewriter 0.00% : 0.000056s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000015s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000007s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000013s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000007s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.00% : 0.000047s : 1: pre_auto_parallel 0.00% : 0.000051s : 1: py_interpret_to_execute 0.00% : 0.000055s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000131s : 1: remove_dup_value 2.94% : 0.154685s : 1: renormalize.infer 0.13% : 0.006868s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000013s : 1: rewriter_after_jit_bprop_graph 0.00% : 0.000258s : 1: rewriter_after_opt_a 0.00% : 0.000147s : 1: rewriter_before_opt_a 0.00% : 0.000008s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000008s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000022s : 1: swap_dp_allreduce_reducescatter 0.00% : 0.000256s : 1: symbol_engine_optimizer 0.00% : 0.000217s : 1: tuple_transform 78.48% : 4.132569s : 1: type_inference . [hook] pytest_runtest_teardown:test_qbmm_qkv_1408_128_128[False-11264-32] tests/st/infer/ops/test_internal_ops/test_qbmm_split.py::test_qbmm_qkv_1408_128_128[False-11264-32],max_mem:118.0M [WARNING] ME(166984:281473373798192,MainProcess):2026-01-29-17:38:34.784.325 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 6.54207, [21] [bootstrap]: 0.00062152 [type_inference]: 5.9311 [event_method]: 2.307e-05 [auto_monad]: 0.00044736 [graph_reusing]: 7.36001e-06 [inline]: 6.25997e-06 [add_attr]: 0.179826, [1] [add_attr_with_inline]: 0.179808, [1] [Cycle 1]: 8.026e-05, [2] [tag_attr]: 2.944e-05 [meta_addattr_fg_expand]: 5.50001e-06 [parallel-infer-symbol]: 3.56001e-06 [pre_auto_parallel]: 4.468e-05 [insert-virtual-dataset]: 2.58e-06 [parallel-infer-symbol-second]: 9.50007e-07 [dataset_repeat_opt]: 2.41e-06 [pipeline_split]: 1.86e-06 [optimize]: 0.428642, [53] [py_interpret_to_execute]: 3.952e-05 [rewriter_before_opt_a]: 0.00012642 [opt_a]: 0.232725, [2] [Cycle 1]: 0.230345, [45] [expand_dump_flag]: 2.88998e-06 [switch_simplify]: 4.291e-05 [loop_unroll]: 2.47e-05 [a_1]: 0.0011044 [with_stream_mark]: 3.614e-05 [recompute_prepare]: 4.241e-05 [updatestate_depend_eliminate]: 1.621e-05 [updatestate_assign_eliminate]: 1.301e-05 [updatestate_loads_eliminate]: 2.04e-05 [parameter_eliminate]: 3.03e-06 [a_2]: 0.00039365 [accelerated_algorithm]: 6.652e-05 [shard]: 2.65002e-06 [meta_shard_fg_expand]: 6.81999e-06 [shard_inline]: 2.38e-05 [merge_send_recv]: 1.992e-05 [auto_parallel]: 1.89e-05 [parallel]: 4.324e-05 [flash_sp]: 1.557e-05 [merge_comm]: 1.343e-05 [allreduce_fusion]: 1.247e-05 [matmul_add_comm_reduction]: 2.641e-05 [allreduce_slice_to_reducescatter]: 9.30013e-07 [virtual_shard_identity]: 3.84e-05 [virtual_dataset]: 3.166e-05 [get_grad_eliminate_]: 3.221e-05 [virtual_output]: 2.96e-05 [merge_forward]: 1.526e-05 [cell_reuse_recompute_pass]: 2.43998e-06 [offload_activation]: 2.684e-05 [cell_reuse_handle_not_recompute_node_pass]: 5.967e-05 [merge_recompute_call_nodes]: 1.60999e-06 [before_grad]: 5.043e-05 [set_forward_comm_id_for_comm_node_pass]: 1.487e-05 [meta_fg_expand]: 8.50001e-06 [flash_sp_send_recv_attached]: 6.83e-06 [receive_attached]: 2.54001e-06 [after_resolve]: 4.605e-05 [a_after_grad]: 5.566e-05 [renormalize]: 0.22485 [add_forward_monad_depend]: 1.664e-05 [auto_monad_grad]: 3.5e-06 [auto_monad_eliminator]: 7.922e-05 [cse]: 0.00233246 [a_3]: 0.00024421 [Cycle 2]: 0.00236363, [45] [expand_dump_flag]: 5.29e-06 [switch_simplify]: 2.556e-05 [loop_unroll]: 2.791e-05 [a_1]: 0.00070382 [with_stream_mark]: 4.509e-05 [recompute_prepare]: 2.474e-05 [updatestate_depend_eliminate]: 1.53e-05 [updatestate_assign_eliminate]: 1.216e-05 [updatestate_loads_eliminate]: 2.13e-05 [parameter_eliminate]: 3.17002e-06 [a_2]: 0.0003126 [accelerated_algorithm]: 3.096e-05 [shard]: 2.34001e-06 [meta_shard_fg_expand]: 7.41999e-06 [shard_inline]: 2.166e-05 [merge_send_recv]: 2.278e-05 [auto_parallel]: 1.976e-05 [parallel]: 1.481e-05 [flash_sp]: 5.67001e-06 [merge_comm]: 1.352e-05 [allreduce_fusion]: 1.038e-05 [matmul_add_comm_reduction]: 2.274e-05 [allreduce_slice_to_reducescatter]: 1.30999e-06 [virtual_shard_identity]: 2.84e-05 [virtual_dataset]: 2.403e-05 [get_grad_eliminate_]: 2.257e-05 [virtual_output]: 2.503e-05 [merge_forward]: 1.258e-05 [cell_reuse_recompute_pass]: 4.13001e-06 [offload_activation]: 2.239e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.907e-05 [merge_recompute_call_nodes]: 2.26998e-06 [before_grad]: 3.494e-05 [set_forward_comm_id_for_comm_node_pass]: 1.257e-05 [meta_fg_expand]: 9.14e-06 [flash_sp_send_recv_attached]: 2.04e-06 [receive_attached]: 2.68e-06 [after_resolve]: 3.491e-05 [a_after_grad]: 3.585e-05 [renormalize]: 8.00064e-08 [add_forward_monad_depend]: 4.78001e-06 [auto_monad_grad]: 3.23e-06 [auto_monad_eliminator]: 7.783e-05 [cse]: 0.00010425 [a_3]: 0.00017552 [py_interpret_to_execute_after_opt_a]: 0.191207 [slice_cell_reuse_recomputed_activation]: 4.91002e-06 [rewriter_after_opt_a]: 0.00032477 [convert_after_rewriter]: 2.201e-05 [order_py_execute_after_rewriter]: 1.161e-05 [mutable_eliminate]: 0.00086463 [opt_b]: 0.00082525, [1] [Cycle 1]: 0.00081538, [7] [b_1]: 0.00055361 [b_2]: 3.129e-05 [updatestate_depend_eliminate]: 1.823e-05 [updatestate_assign_eliminate]: 1.173e-05 [updatestate_loads_eliminate]: 1.79e-05 [renormalize]: 9.80013e-07 [cse]: 0.00013537 [optimize_parallel_all_gather_comm]: 4.033e-05 [overlap_param_gather]: 2.26998e-06 [cconv]: 4.377e-05 [loop_unroll]: 0.00065024 [opt_after_cconv]: 0.00033777, [1] [Cycle 1]: 0.00032918, [7] [c_1]: 0.00015191 [parameter_eliminate]: 6.94999e-06 [updatestate_depend_eliminate]: 1.701e-05 [updatestate_assign_eliminate]: 1.028e-05 [updatestate_loads_eliminate]: 1.491e-05 [cse]: 8.494e-05 [renormalize]: 1.15999e-06 [remove_dup_value]: 0.00013767 [tuple_transform]: 0.00023557, [1] [Cycle 1]: 0.00022935, [4] [d_1]: 0.00017733 [none_parameter_eliminate]: 2.83e-06 [renormalize]: 2.09984e-07 [switch_simplify]: 2.378e-05 [partial_unused_args_eliminate]: 2.45002e-06 [add_recomputation]: 0.00015845 [cse_after_recomputation]: 8.359e-05, [1] [Cycle 1]: 7.497e-05, [1] [cse]: 6.674e-05 [environ_conv]: 2.948e-05 [swap_dp_allreduce_reducescatter]: 1.61e-05 [bias_add_comm_swap]: 3.67998e-06 [label_micro_interleaved_index]: 7.05998e-06 [label_fine_grained_interleaved_index]: 3.23e-06 [merge_cast_opt]: 1.62999e-06 [slice_recompute_activation]: 2.78e-06 [micro_interleaved_order_control]: 3.36001e-06 [assign_add_opt]: 1.55001e-06 [ForceFp32Comm]: 1.00999e-06 [remove_cast_before_assign_add]: 1.34e-06 [full_micro_interleaved_order_control]: 2.58e-06 [reorder_send_recv_between_fp_bp]: 3.3e-06 [comm_op_add_attrs]: 1.20001e-06 [add_comm_op_reuse_tag]: 1.02e-06 [interleave_split_concat_branches]: 1.22e-06 [interleave_parallel_branches]: 1.19e-06 [overlap_opt_shard_in_pipeline]: 1.47999e-06 [overlap_opt_shard_grad_in_pipeline]: 2.07999e-06 [control_data_broadcast_order]: 3.556e-05 [grouped_pairwise_exchange_alltoall]: 2.19001e-06 [offloading_packed_experts]: 9.40001e-06 [overlap_recompute_and_grad_model_parallel]: 9.46998e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.87999e-06 [overlap_recompute_allgather_and_fa_grad]: 2.01998e-06 [overlap_recompute_comm]: 2.25002e-06 [overlap_grad_ring_attention]: 1.115e-05 [overlap_grad_flash_sp]: 4.924e-05 [begin_end_overlap_inline]: 6.19999e-07 [split_matmul_comm_elemetwise]: 2.72001e-06 [split_layernorm_comm]: 1.97001e-06 [handle_group_info]: 1.07e-06 [symbol_engine_optimizer]: 0.00022092, [1] [Cycle 1]: 0.00021282, [6] [build]: 2.265e-05 [elim_shapecalc]: 3.468e-05 [elim_not_effective]: 4.219e-05 [opt_reshape]: 3.302e-05 [fold_const_symbol]: 3.499e-05 [renormalize]: 4.39992e-07 [detach_backward]: 2.96001e-06 [pipeline_parallel_scheduler]: 1.49998e-06 [auto_monad_reorder]: 8.21e-05 [get_jit_bprop_graph]: 2.33998e-06 [rewriter_after_jit_bprop_graph]: 8.10999e-06 [opt_after_jit_grad]: 0.00088929 [validate]: 0.00011961 Sums bootstrap : 0.000622s : 0.01% type_inference : 5.931098s : 93.25% event_method : 0.000023s : 0.00% auto_monad : 0.000447s : 0.01% graph_reusing : 0.000007s : 0.00% inline : 0.000006s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000029s : 0.00% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000006s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000045s : 0.00% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000040s : 0.00% optimize.rewriter_before_opt_a : 0.000126s : 0.00% optimize.opt_a.expand_dump_flag : 0.000008s : 0.00% optimize.opt_a.switch_simplify : 0.000068s : 0.00% optimize.opt_a.loop_unroll : 0.000053s : 0.00% optimize.opt_a.a_1 : 0.001808s : 0.03% optimize.opt_a.with_stream_mark : 0.000081s : 0.00% optimize.opt_a.recompute_prepare : 0.000067s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000032s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000025s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000042s : 0.00% optimize.opt_a.parameter_eliminate : 0.000006s : 0.00% optimize.opt_a.a_2 : 0.000706s : 0.01% optimize.opt_a.accelerated_algorithm : 0.000097s : 0.00% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000014s : 0.00% optimize.opt_a.shard_inline : 0.000045s : 0.00% optimize.opt_a.merge_send_recv : 0.000043s : 0.00% optimize.opt_a.auto_parallel : 0.000039s : 0.00% optimize.opt_a.parallel : 0.000058s : 0.00% optimize.opt_a.flash_sp : 0.000021s : 0.00% optimize.opt_a.merge_comm : 0.000027s : 0.00% optimize.opt_a.allreduce_fusion : 0.000023s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000049s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000067s : 0.00% optimize.opt_a.virtual_dataset : 0.000056s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000055s : 0.00% optimize.opt_a.virtual_output : 0.000055s : 0.00% optimize.opt_a.merge_forward : 0.000028s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000007s : 0.00% optimize.opt_a.offload_activation : 0.000049s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000109s : 0.00% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000085s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000027s : 0.00% optimize.opt_a.meta_fg_expand : 0.000018s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000009s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000081s : 0.00% optimize.opt_a.a_after_grad : 0.000092s : 0.00% optimize.opt_a.renormalize : 0.224850s : 3.53% optimize.opt_a.add_forward_monad_depend : 0.000021s : 0.00% optimize.opt_a.auto_monad_grad : 0.000007s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000157s : 0.00% optimize.opt_a.cse : 0.002437s : 0.04% optimize.opt_a.a_3 : 0.000420s : 0.01% optimize.py_interpret_to_execute_after_opt_a : 0.191207s : 3.01% optimize.slice_cell_reuse_recomputed_activation : 0.000005s : 0.00% optimize.rewriter_after_opt_a : 0.000325s : 0.01% optimize.convert_after_rewriter : 0.000022s : 0.00% optimize.order_py_execute_after_rewriter : 0.000012s : 0.00% optimize.mutable_eliminate : 0.000865s : 0.01% optimize.opt_b.b_1 : 0.000554s : 0.01% optimize.opt_b.b_2 : 0.000031s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000018s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000012s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000018s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000135s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000040s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000044s : 0.00% optimize.loop_unroll : 0.000650s : 0.01% optimize.opt_after_cconv.c_1 : 0.000152s : 0.00% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000017s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000010s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000015s : 0.00% optimize.opt_after_cconv.cse : 0.000085s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000138s : 0.00% optimize.tuple_transform.d_1 : 0.000177s : 0.00% optimize.tuple_transform.none_parameter_eliminate : 0.000003s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000024s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000158s : 0.00% optimize.cse_after_recomputation.cse : 0.000067s : 0.00% optimize.environ_conv : 0.000029s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000016s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000007s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000036s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000009s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000009s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000011s : 0.00% optimize.overlap_grad_flash_sp : 0.000049s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000023s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000035s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000042s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000033s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000035s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000001s : 0.00% auto_monad_reorder : 0.000082s : 0.00% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.000889s : 0.01% validate : 0.000120s : 0.00% Time group info: ------[substitution.] 0.000459 135 2.57% : 0.000012s : 2: substitution.depend_value_elim 0.98% : 0.000005s : 11: substitution.elim_not_effective 0.98% : 0.000004s : 11: substitution.fold_const_symbol 3.49% : 0.000016s : 18: substitution.graph_param_transform 32.69% : 0.000150s : 1: substitution.inline 2.64% : 0.000012s : 22: substitution.j_node_and_user_rematch 7.21% : 0.000033s : 2: substitution.less_batch_normalization 2.82% : 0.000013s : 18: substitution.load_eliminater 1.03% : 0.000005s : 2: substitution.opt_reshape 4.64% : 0.000021s : 22: substitution.remove_not_recompute_node 2.49% : 0.000011s : 8: substitution.replace_old_param 7.80% : 0.000036s : 4: substitution.reshape_eliminate 2.76% : 0.000013s : 6: substitution.updatestate_pure_node_eliminater 27.91% : 0.000128s : 8: substitution.updatestate_useless_node_eliminater ------[type_inference.] 5.930977 2 99.85% : 5.922366s : 1: type_inference.infer 0.15% : 0.008611s : 1: type_inference.specialize ------[replace.] 0.000023 1 100.00% : 0.000023s : 1: replace.inline ------[match.] 0.000149 1 100.00% : 0.000149s : 1: match.inline ------[predicate.] 0.000646 4131 0.75% : 0.000005s : 37: predicate.accumulaten_eliminater 0.92% : 0.000006s : 18: predicate.ad_related_special_op_eliminate 0.74% : 0.000005s : 36: predicate.addn_check_dump 0.77% : 0.000005s : 37: predicate.addn_zero_filter 0.69% : 0.000004s : 37: predicate.adjust_all_reduce_mul_add 1.88% : 0.000012s : 73: predicate.arithmetic_simplify 0.92% : 0.000006s : 37: predicate.cast_eliminate 0.86% : 0.000006s : 36: predicate.check_bprop_eliminate 0.84% : 0.000005s : 36: predicate.compare_switch_simplify 0.23% : 0.000001s : 18: predicate.const_output_eliminate 0.81% : 0.000005s : 36: predicate.depend_value_elim 0.81% : 0.000005s : 37: predicate.dict_get_item_const_eliminator 0.87% : 0.000006s : 37: predicate.dict_get_item_eliminator 0.78% : 0.000005s : 37: predicate.dict_set_item_eliminator 0.99% : 0.000006s : 36: predicate.dumpgradient_eliminate 0.26% : 0.000002s : 18: predicate.elim_not_effective 0.53% : 0.000003s : 18: predicate.elim_shapecalc_of_broadcastargs 1.14% : 0.000007s : 55: predicate.environ_add_const_eliminate 1.18% : 0.000008s : 55: predicate.environ_get_add_eliminate 1.25% : 0.000008s : 55: predicate.environ_get_depend_swap 1.79% : 0.000012s : 91: predicate.environ_get_eliminate 1.09% : 0.000007s : 55: predicate.environ_get_set_eliminate 0.72% : 0.000005s : 38: predicate.exchange_switch_depend_value 1.14% : 0.000007s : 38: predicate.float_depend_g_call 0.75% : 0.000005s : 36: predicate.float_environ_get_switch 1.12% : 0.000007s : 54: predicate.float_tuple_getitem_switch 0.21% : 0.000001s : 18: predicate.fold_const_symbol 0.93% : 0.000006s : 36: predicate.get_grad_eliminate 0.24% : 0.000002s : 18: predicate.graph_param_transform 0.72% : 0.000005s : 36: predicate.incorporate_call 0.66% : 0.000004s : 36: predicate.incorporate_call_switch 4.82% : 0.000031s : 183: predicate.inline 1.19% : 0.000008s : 36: predicate.inline_without_move 0.42% : 0.000003s : 36: predicate.j_node_and_user_rematch 1.15% : 0.000007s : 36: predicate.less_batch_normalization 1.67% : 0.000011s : 73: predicate.list_to_tuple_eliminator_ 2.20% : 0.000014s : 110: predicate.load_eliminater 0.92% : 0.000006s : 18: predicate.loop_unroll_after_grad 1.01% : 0.000007s : 41: predicate.loop_unroll_before_grad 1.75% : 0.000011s : 73: predicate.make_slice_get_slice_eliminator 0.91% : 0.000006s : 36: predicate.merge_addn 0.85% : 0.000006s : 36: predicate.micro_step_allgather_replace 0.82% : 0.000005s : 36: predicate.mini_step_allgather_replace 0.72% : 0.000005s : 37: predicate.minmaximum_grad 1.08% : 0.000007s : 18: predicate.mutable_eliminate 0.50% : 0.000003s : 18: predicate.opt_reshape 0.48% : 0.000003s : 18: predicate.parallel_virtual_node 0.99% : 0.000006s : 38: predicate.partial_defer_inline 1.17% : 0.000008s : 55: predicate.partial_eliminate 0.72% : 0.000005s : 37: predicate.print_const_string_wrapper 0.75% : 0.000005s : 36: predicate.reduce_all_const_elim 1.08% : 0.000007s : 37: predicate.reduce_eliminate 2.18% : 0.000014s : 110: predicate.redundant_stop_gradient_eliminater 0.48% : 0.000003s : 36: predicate.remove_not_recompute_node 5.93% : 0.000038s : 73: predicate.replace_applicator 0.54% : 0.000003s : 36: predicate.replace_old_param 0.33% : 0.000002s : 18: predicate.reset_defer_inline 0.89% : 0.000006s : 37: predicate.reshape_eliminate 1.00% : 0.000006s : 36: predicate.row_tensor_add_zeros_like 0.49% : 0.000003s : 18: predicate.row_tensor_eliminate 1.24% : 0.000008s : 36: predicate.same_eliminate 0.52% : 0.000003s : 36: predicate.set_cell_output_no_recompute 1.11% : 0.000007s : 36: predicate.shard_identity_eliminate 0.90% : 0.000006s : 36: predicate.special_op_eliminate 0.87% : 0.000006s : 36: predicate.specialize_transform 1.27% : 0.000008s : 36: predicate.split_environ_get_set_with_tuple_value 1.05% : 0.000007s : 36: predicate.stack_unstack_eliminate 0.48% : 0.000003s : 18: predicate.switch_call_monad_eliminater 0.77% : 0.000005s : 38: predicate.switch_defer_inline 1.53% : 0.000010s : 74: predicate.switch_layer_defer_inline 3.25% : 0.000021s : 133: predicate.switch_simplify 0.73% : 0.000005s : 37: predicate.tile_eliminate 0.96% : 0.000006s : 37: predicate.transpose_eliminate 1.67% : 0.000011s : 73: predicate.tuple_list_convert_item_index_to_positive 1.59% : 0.000010s : 73: predicate.tuple_list_get_item_const_eliminator 1.61% : 0.000010s : 73: predicate.tuple_list_get_item_depend_reorder 2.80% : 0.000018s : 109: predicate.tuple_list_get_item_eliminator 1.69% : 0.000011s : 73: predicate.tuple_list_get_set_item_eliminator 2.50% : 0.000016s : 109: predicate.tuple_list_set_item_eliminator 1.48% : 0.000010s : 73: predicate.tuple_to_list_eliminator_ 2.16% : 0.000014s : 110: predicate.updatestate_pure_node_eliminater 3.00% : 0.000019s : 146: predicate.updatestate_useless_node_eliminater 0.44% : 0.000003s : 18: predicate.value_based_eliminate 1.02% : 0.000007s : 36: predicate.virtual_dataset_eliminate 0.93% : 0.000006s : 36: predicate.virtual_output_eliminate 0.39% : 0.000003s : 18: predicate.virtual_view_grad_eliminate 0.44% : 0.000003s : 18: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.308066 58 97.87% : 0.301519s : 55: func_graph_cloner_run.FuncGraphClonerGraph 2.13% : 0.006547s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 7.379972 192 0.00% : 0.000004s : 1: ForceFp32Comm 2.44% : 0.179835s : 1: add_attr 2.44% : 0.179812s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.00% : 0.000164s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.01% : 0.000460s : 1: auto_monad 0.00% : 0.000089s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.01% : 0.000657s : 1: bootstrap 0.00% : 0.000048s : 1: cconv 0.00% : 0.000006s : 1: comm_op_add_attrs 0.00% : 0.000040s : 1: control_data_broadcast_order 0.00% : 0.000027s : 1: convert_after_rewriter 0.00% : 0.000087s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000035s : 1: environ_conv 0.00% : 0.000031s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000009s : 1: get_jit_bprop_graph 0.00% : 0.000012s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000010s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000006s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000011s : 1: label_micro_interleaved_index 0.01% : 0.000663s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.01% : 0.000877s : 1: mutable_eliminate 0.00% : 0.000013s : 1: offloading_packed_experts 0.00% : 0.000040s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000048s : 1: opt.transform.mutable_eliminate 0.05% : 0.003704s : 78: opt.transform.opt_a 0.00% : 0.000149s : 1: opt.transform.opt_after_cconv 0.00% : 0.000090s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000535s : 28: opt.transform.opt_b 0.00% : 0.000198s : 2: opt.transform.opt_trans_graph 0.00% : 0.000139s : 4: opt.transform.symbol_engine_opt 3.15% : 0.232731s : 1: opt_a 0.00% : 0.000341s : 1: opt_after_cconv 0.01% : 0.000904s : 1: opt_after_jit_grad 0.01% : 0.000830s : 1: opt_b 5.81% : 0.428652s : 1: optimize 0.00% : 0.000045s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000015s : 1: order_py_execute_after_rewriter 0.00% : 0.000053s : 1: overlap_grad_flash_sp 0.00% : 0.000006s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000014s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000007s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000013s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000007s : 1: partial_unused_args_eliminate 0.00% : 0.000007s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.00% : 0.000049s : 1: pre_auto_parallel 0.00% : 0.000044s : 1: py_interpret_to_execute 2.59% : 0.191238s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000145s : 1: remove_dup_value 2.95% : 0.217519s : 1: renormalize.infer 0.10% : 0.007307s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.00% : 0.000333s : 1: rewriter_after_opt_a 0.00% : 0.000132s : 1: rewriter_before_opt_a 0.00% : 0.000013s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000007s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000019s : 1: swap_dp_allreduce_reducescatter 0.00% : 0.000224s : 1: symbol_engine_optimizer 0.00% : 0.000239s : 1: tuple_transform 80.37% : 5.931138s : 1: type_inference . [hook] pytest_runtest_teardown:test_qbmm_qkv_1408_128_128[False-128-32] tests/st/infer/ops/test_internal_ops/test_qbmm_split.py::test_qbmm_qkv_1408_128_128[False-128-32],max_mem:118.0M [WARNING] ME(166984:281473373798192,MainProcess):2026-01-29-17:38:46.114.863 [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 = 7.25152, [21] [bootstrap]: 0.00064469 [type_inference]: 6.05971 [event_method]: 0.00032468 [auto_monad]: 0.00116453 [graph_reusing]: 4.446e-05 [inline]: 2.93e-06 [add_attr]: 0.274425, [1] [add_attr_with_inline]: 0.274349, [1] [Cycle 1]: 0.00079369, [2] [tag_attr]: 0.00067038 [meta_addattr_fg_expand]: 5.927e-05 [parallel-infer-symbol]: 4.30999e-06 [pre_auto_parallel]: 0.00030585 [insert-virtual-dataset]: 6.11e-06 [parallel-infer-symbol-second]: 8.59989e-07 [dataset_repeat_opt]: 2.81e-06 [pipeline_split]: 4.79e-06 [optimize]: 0.728405, [53] [py_interpret_to_execute]: 0.0005938 [rewriter_before_opt_a]: 0.00117021 [opt_a]: 0.719024, [2] [Cycle 1]: 0.715875, [45] [expand_dump_flag]: 5.745e-05 [switch_simplify]: 0.00503464 [loop_unroll]: 0.00069794 [a_1]: 0.234762 [with_stream_mark]: 0.00057356 [recompute_prepare]: 5.697e-05 [updatestate_depend_eliminate]: 3.227e-05 [updatestate_assign_eliminate]: 2.014e-05 [updatestate_loads_eliminate]: 2.875e-05 [parameter_eliminate]: 3.73999e-06 [a_2]: 0.00101717 [accelerated_algorithm]: 0.00015214 [shard]: 2.44001e-06 [meta_shard_fg_expand]: 3.58e-05 [shard_inline]: 2.854e-05 [merge_send_recv]: 2.353e-05 [auto_parallel]: 2.427e-05 [parallel]: 0.00011222 [flash_sp]: 1.676e-05 [merge_comm]: 1.643e-05 [allreduce_fusion]: 1.575e-05 [matmul_add_comm_reduction]: 2.846e-05 [allreduce_slice_to_reducescatter]: 1.18001e-06 [virtual_shard_identity]: 3.387e-05 [virtual_dataset]: 3.054e-05 [get_grad_eliminate_]: 3.113e-05 [virtual_output]: 2.835e-05 [merge_forward]: 1.819e-05 [cell_reuse_recompute_pass]: 2.01998e-06 [offload_activation]: 2.846e-05 [cell_reuse_handle_not_recompute_node_pass]: 5.911e-05 [merge_recompute_call_nodes]: 1.61002e-06 [before_grad]: 5.221e-05 [set_forward_comm_id_for_comm_node_pass]: 1.741e-05 [meta_fg_expand]: 2.241e-05 [flash_sp_send_recv_attached]: 5.42001e-06 [receive_attached]: 2.31e-06 [after_resolve]: 4.112e-05 [a_after_grad]: 4.998e-05 [renormalize]: 0.469193 [add_forward_monad_depend]: 1.701e-05 [auto_monad_grad]: 4.08999e-06 [auto_monad_eliminator]: 0.00017177 [cse]: 0.00229288 [a_3]: 0.00023568 [Cycle 2]: 0.00312704, [45] [expand_dump_flag]: 4.03999e-06 [switch_simplify]: 3.285e-05 [loop_unroll]: 2.973e-05 [a_1]: 0.00094901 [with_stream_mark]: 5.135e-05 [recompute_prepare]: 3.393e-05 [updatestate_depend_eliminate]: 2.069e-05 [updatestate_assign_eliminate]: 1.661e-05 [updatestate_loads_eliminate]: 4.344e-05 [parameter_eliminate]: 5.00001e-06 [a_2]: 0.00044724 [accelerated_algorithm]: 4.29e-05 [shard]: 2.84999e-06 [meta_shard_fg_expand]: 1.543e-05 [shard_inline]: 2.789e-05 [merge_send_recv]: 3.117e-05 [auto_parallel]: 2.522e-05 [parallel]: 1.167e-05 [flash_sp]: 5.38002e-06 [merge_comm]: 1.948e-05 [allreduce_fusion]: 1.615e-05 [matmul_add_comm_reduction]: 2.873e-05 [allreduce_slice_to_reducescatter]: 9.60019e-07 [virtual_shard_identity]: 3.685e-05 [virtual_dataset]: 3.179e-05 [get_grad_eliminate_]: 2.912e-05 [virtual_output]: 2.7e-05 [merge_forward]: 1.854e-05 [cell_reuse_recompute_pass]: 3.23e-06 [offload_activation]: 5.399e-05 [cell_reuse_handle_not_recompute_node_pass]: 6.294e-05 [merge_recompute_call_nodes]: 1.69e-06 [before_grad]: 4.8e-05 [set_forward_comm_id_for_comm_node_pass]: 2.308e-05 [meta_fg_expand]: 1.59e-05 [flash_sp_send_recv_attached]: 2.84999e-06 [receive_attached]: 3.18e-06 [after_resolve]: 4.153e-05 [a_after_grad]: 4.435e-05 [renormalize]: 6.00121e-08 [add_forward_monad_depend]: 5.54e-06 [auto_monad_grad]: 3.26001e-06 [auto_monad_eliminator]: 9.07e-05 [cse]: 0.00013054 [a_3]: 0.00019109 [py_interpret_to_execute_after_opt_a]: 5.092e-05 [slice_cell_reuse_recomputed_activation]: 3.02002e-06 [rewriter_after_opt_a]: 0.00106114 [convert_after_rewriter]: 8.325e-05 [order_py_execute_after_rewriter]: 3.498e-05 [mutable_eliminate]: 0.00108933 [opt_b]: 0.00120633, [1] [Cycle 1]: 0.00119354, [7] [b_1]: 0.00080869 [b_2]: 3.754e-05 [updatestate_depend_eliminate]: 3.308e-05 [updatestate_assign_eliminate]: 1.709e-05 [updatestate_loads_eliminate]: 2.856e-05 [renormalize]: 2.052e-05 [cse]: 0.00017158 [optimize_parallel_all_gather_comm]: 5.627e-05 [overlap_param_gather]: 2.24999e-06 [cconv]: 5.409e-05 [loop_unroll]: 0.00087439 [opt_after_cconv]: 0.00051259, [1] [Cycle 1]: 0.0005005, [7] [c_1]: 0.00022149 [parameter_eliminate]: 7.78999e-06 [updatestate_depend_eliminate]: 3.27e-05 [updatestate_assign_eliminate]: 1.704e-05 [updatestate_loads_eliminate]: 2.318e-05 [cse]: 0.00013537 [renormalize]: 1.39e-06 [remove_dup_value]: 0.0001865 [tuple_transform]: 0.00041438, [1] [Cycle 1]: 0.00040558, [4] [d_1]: 0.00032067 [none_parameter_eliminate]: 3.54002e-06 [renormalize]: 2.30008e-07 [switch_simplify]: 3.671e-05 [partial_unused_args_eliminate]: 3.14001e-06 [add_recomputation]: 0.00022387 [cse_after_recomputation]: 0.00013382, [1] [Cycle 1]: 0.00012607, [1] [cse]: 0.00011211 [environ_conv]: 4.688e-05 [swap_dp_allreduce_reducescatter]: 2.468e-05 [bias_add_comm_swap]: 4.25999e-06 [label_micro_interleaved_index]: 9.42001e-06 [label_fine_grained_interleaved_index]: 2.95002e-06 [merge_cast_opt]: 1.49e-06 [slice_recompute_activation]: 2.54001e-06 [micro_interleaved_order_control]: 2.63e-06 [assign_add_opt]: 1.75001e-06 [ForceFp32Comm]: 9.30013e-07 [remove_cast_before_assign_add]: 1.32999e-06 [full_micro_interleaved_order_control]: 3.02002e-06 [reorder_send_recv_between_fp_bp]: 3.13e-06 [comm_op_add_attrs]: 1.23002e-06 [add_comm_op_reuse_tag]: 1.44e-06 [interleave_split_concat_branches]: 1.74e-06 [interleave_parallel_branches]: 1.27e-06 [overlap_opt_shard_in_pipeline]: 1.51998e-06 [overlap_opt_shard_grad_in_pipeline]: 2.62001e-06 [control_data_broadcast_order]: 5.172e-05 [grouped_pairwise_exchange_alltoall]: 1.76e-06 [offloading_packed_experts]: 1.732e-05 [overlap_recompute_and_grad_model_parallel]: 1.477e-05 [overlap_grad_matmul_and_grad_allreduce]: 2.22001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.45999e-06 [overlap_recompute_comm]: 2.69001e-06 [overlap_grad_ring_attention]: 1.39e-05 [overlap_grad_flash_sp]: 9.161e-05 [begin_end_overlap_inline]: 1.02998e-06 [split_matmul_comm_elemetwise]: 2.87002e-06 [split_layernorm_comm]: 2.09e-06 [handle_group_info]: 1.12e-06 [symbol_engine_optimizer]: 0.00079948, [1] [Cycle 1]: 0.00078931, [6] [build]: 0.00039581 [elim_shapecalc]: 5.878e-05 [elim_not_effective]: 8.895e-05 [opt_reshape]: 4.534e-05 [fold_const_symbol]: 0.00013167 [renormalize]: 5.50004e-07 [detach_backward]: 3.51001e-06 [pipeline_parallel_scheduler]: 4.55001e-06 [auto_monad_reorder]: 0.00010637 [get_jit_bprop_graph]: 2.21e-06 [rewriter_after_jit_bprop_graph]: 8.00999e-06 [opt_after_jit_grad]: 0.00101993 [validate]: 0.00027976 Sums bootstrap : 0.000645s : 0.01% type_inference : 6.059714s : 89.23% event_method : 0.000325s : 0.00% auto_monad : 0.001165s : 0.02% graph_reusing : 0.000044s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000670s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000059s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000306s : 0.00% insert-virtual-dataset : 0.000006s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000005s : 0.00% optimize.py_interpret_to_execute : 0.000594s : 0.01% optimize.rewriter_before_opt_a : 0.001170s : 0.02% optimize.opt_a.expand_dump_flag : 0.000061s : 0.00% optimize.opt_a.switch_simplify : 0.005067s : 0.07% optimize.opt_a.loop_unroll : 0.000728s : 0.01% optimize.opt_a.a_1 : 0.235711s : 3.47% optimize.opt_a.with_stream_mark : 0.000625s : 0.01% optimize.opt_a.recompute_prepare : 0.000091s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000053s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000037s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000072s : 0.00% optimize.opt_a.parameter_eliminate : 0.000009s : 0.00% optimize.opt_a.a_2 : 0.001464s : 0.02% optimize.opt_a.accelerated_algorithm : 0.000195s : 0.00% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000051s : 0.00% optimize.opt_a.shard_inline : 0.000056s : 0.00% optimize.opt_a.merge_send_recv : 0.000055s : 0.00% optimize.opt_a.auto_parallel : 0.000049s : 0.00% optimize.opt_a.parallel : 0.000124s : 0.00% optimize.opt_a.flash_sp : 0.000022s : 0.00% optimize.opt_a.merge_comm : 0.000036s : 0.00% optimize.opt_a.allreduce_fusion : 0.000032s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000057s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000071s : 0.00% optimize.opt_a.virtual_dataset : 0.000062s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000060s : 0.00% optimize.opt_a.virtual_output : 0.000055s : 0.00% optimize.opt_a.merge_forward : 0.000037s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000082s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000122s : 0.00% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000100s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000040s : 0.00% optimize.opt_a.meta_fg_expand : 0.000038s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000008s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000083s : 0.00% optimize.opt_a.a_after_grad : 0.000094s : 0.00% optimize.opt_a.renormalize : 0.469193s : 6.91% optimize.opt_a.add_forward_monad_depend : 0.000023s : 0.00% optimize.opt_a.auto_monad_grad : 0.000007s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000262s : 0.00% optimize.opt_a.cse : 0.002423s : 0.04% optimize.opt_a.a_3 : 0.000427s : 0.01% optimize.py_interpret_to_execute_after_opt_a : 0.000051s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.001061s : 0.02% optimize.convert_after_rewriter : 0.000083s : 0.00% optimize.order_py_execute_after_rewriter : 0.000035s : 0.00% optimize.mutable_eliminate : 0.001089s : 0.02% optimize.opt_b.b_1 : 0.000809s : 0.01% optimize.opt_b.b_2 : 0.000038s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000033s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000017s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000029s : 0.00% optimize.opt_b.renormalize : 0.000021s : 0.00% optimize.opt_b.cse : 0.000172s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000056s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000054s : 0.00% optimize.loop_unroll : 0.000874s : 0.01% optimize.opt_after_cconv.c_1 : 0.000221s : 0.00% optimize.opt_after_cconv.parameter_eliminate : 0.000008s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000033s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000017s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000023s : 0.00% optimize.opt_after_cconv.cse : 0.000135s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000186s : 0.00% optimize.tuple_transform.d_1 : 0.000321s : 0.00% optimize.tuple_transform.none_parameter_eliminate : 0.000004s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000037s : 0.00% optimize.partial_unused_args_eliminate : 0.000003s : 0.00% optimize.add_recomputation : 0.000224s : 0.00% optimize.cse_after_recomputation.cse : 0.000112s : 0.00% optimize.environ_conv : 0.000047s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000025s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000009s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000002s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000003s : 0.00% optimize.control_data_broadcast_order : 0.000052s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000017s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000015s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000014s : 0.00% optimize.overlap_grad_flash_sp : 0.000092s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000396s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000059s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000089s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000045s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000132s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000001s : 0.00% detach_backward : 0.000004s : 0.00% pipeline_parallel_scheduler : 0.000005s : 0.00% auto_monad_reorder : 0.000106s : 0.00% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.001020s : 0.02% validate : 0.000280s : 0.00% Time group info: ------[substitution.] 0.006419 608 0.29% : 0.000019s : 8: substitution.depend_value_elim 0.54% : 0.000034s : 17: substitution.elim_not_effective 0.31% : 0.000020s : 12: substitution.float_tuple_getitem_switch 1.32% : 0.000085s : 17: substitution.fold_const_symbol 0.36% : 0.000023s : 22: substitution.graph_param_transform 67.38% : 0.004326s : 71: substitution.inline 0.26% : 0.000016s : 35: substitution.j_node_and_user_rematch 1.85% : 0.000119s : 6: substitution.less_batch_normalization 1.08% : 0.000069s : 2: substitution.list_to_tuple_eliminator_ 0.19% : 0.000012s : 18: substitution.load_eliminater 0.52% : 0.000033s : 21: substitution.minmaximum_grad 0.12% : 0.000008s : 2: substitution.opt_reshape 0.40% : 0.000026s : 35: substitution.remove_not_recompute_node 0.21% : 0.000013s : 8: substitution.replace_old_param 1.36% : 0.000087s : 18: substitution.reshape_eliminate 2.91% : 0.000187s : 24: substitution.switch_simplify 3.14% : 0.000202s : 51: substitution.tuple_list_convert_item_index_to_positive 4.75% : 0.000305s : 51: substitution.tuple_list_get_item_const_eliminator 3.42% : 0.000219s : 51: substitution.tuple_list_get_item_depend_reorder 4.39% : 0.000282s : 73: substitution.tuple_list_get_item_eliminator 2.44% : 0.000157s : 51: substitution.tuple_list_get_set_item_eliminator 0.22% : 0.000014s : 6: substitution.updatestate_pure_node_eliminater 2.57% : 0.000165s : 9: substitution.updatestate_useless_node_eliminater ------[type_inference.] 6.058959 2 88.55% : 5.365372s : 1: type_inference.infer 11.45% : 0.693587s : 1: type_inference.specialize ------[replace.] 0.003059 115 1.82% : 0.000056s : 5: replace.depend_value_elim 37.14% : 0.001136s : 71: replace.inline 1.67% : 0.000051s : 2: replace.list_to_tuple_eliminator_ 54.70% : 0.001673s : 24: replace.switch_simplify 4.67% : 0.000143s : 13: replace.tuple_list_get_item_eliminator ------[match.] 0.004554 115 0.05% : 0.000002s : 5: match.depend_value_elim 93.77% : 0.004270s : 71: match.inline 1.46% : 0.000066s : 2: match.list_to_tuple_eliminator_ 3.79% : 0.000173s : 24: match.switch_simplify 0.93% : 0.000043s : 13: match.tuple_list_get_item_eliminator ------[predicate.] 0.002820 10915 0.75% : 0.000021s : 133: predicate.accumulaten_eliminater 0.36% : 0.000010s : 22: predicate.ad_related_special_op_eliminate 0.37% : 0.000010s : 73: predicate.addn_check_dump 0.74% : 0.000021s : 133: predicate.addn_zero_filter 0.66% : 0.000019s : 133: predicate.adjust_all_reduce_mul_add 1.60% : 0.000045s : 206: predicate.arithmetic_simplify 0.85% : 0.000024s : 133: predicate.cast_eliminate 0.24% : 0.000007s : 44: predicate.check_bprop_eliminate 0.36% : 0.000010s : 73: predicate.compare_switch_simplify 0.06% : 0.000002s : 22: predicate.const_output_eliminate 0.44% : 0.000012s : 76: predicate.depend_value_elim 0.85% : 0.000024s : 133: predicate.dict_get_item_const_eliminator 0.83% : 0.000023s : 133: predicate.dict_get_item_eliminator 0.72% : 0.000020s : 133: predicate.dict_set_item_eliminator 0.34% : 0.000010s : 44: predicate.dumpgradient_eliminate 0.08% : 0.000002s : 22: predicate.elim_not_effective 0.19% : 0.000005s : 22: predicate.elim_shapecalc_of_broadcastargs 0.82% : 0.000023s : 155: predicate.environ_add_const_eliminate 0.83% : 0.000023s : 155: predicate.environ_get_add_eliminate 0.84% : 0.000024s : 155: predicate.environ_get_depend_swap 3.16% : 0.000089s : 228: predicate.environ_get_eliminate 0.89% : 0.000025s : 155: predicate.environ_get_set_eliminate 1.19% : 0.000034s : 219: predicate.exchange_switch_depend_value 1.77% : 0.000050s : 219: predicate.float_depend_g_call 0.37% : 0.000011s : 73: predicate.float_environ_get_switch 0.52% : 0.000015s : 95: predicate.float_tuple_getitem_switch 0.07% : 0.000002s : 22: predicate.fold_const_symbol 0.34% : 0.000010s : 45: predicate.get_grad_eliminate 0.07% : 0.000002s : 22: predicate.graph_param_transform 0.37% : 0.000010s : 73: predicate.incorporate_call 0.34% : 0.000010s : 73: predicate.incorporate_call_switch 4.10% : 0.000116s : 514: predicate.inline 0.31% : 0.000009s : 45: predicate.inline_without_move 0.11% : 0.000003s : 45: predicate.j_node_and_user_rematch 0.32% : 0.000009s : 45: predicate.less_batch_normalization 1.28% : 0.000036s : 192: predicate.list_to_tuple_eliminator_ 1.76% : 0.000050s : 325: predicate.load_eliminater 0.48% : 0.000014s : 22: predicate.loop_unroll_after_grad 4.71% : 0.000133s : 324: predicate.loop_unroll_before_grad 1.07% : 0.000030s : 177: predicate.make_slice_get_slice_eliminator 0.37% : 0.000010s : 73: predicate.merge_addn 0.25% : 0.000007s : 44: predicate.micro_step_allgather_replace 0.22% : 0.000006s : 44: predicate.mini_step_allgather_replace 0.80% : 0.000023s : 133: predicate.minmaximum_grad 0.59% : 0.000016s : 22: predicate.mutable_eliminate 0.16% : 0.000005s : 22: predicate.opt_reshape 0.17% : 0.000005s : 22: predicate.parallel_virtual_node 2.56% : 0.000072s : 219: predicate.partial_defer_inline 0.93% : 0.000026s : 170: predicate.partial_eliminate 0.72% : 0.000020s : 133: predicate.print_const_string_wrapper 0.34% : 0.000010s : 68: predicate.reduce_all_const_elim 0.99% : 0.000028s : 133: predicate.reduce_eliminate 1.67% : 0.000047s : 325: predicate.redundant_stop_gradient_eliminater 0.13% : 0.000004s : 45: predicate.remove_not_recompute_node 0.72% : 0.000020s : 192: predicate.replace_applicator 0.13% : 0.000004s : 45: predicate.replace_old_param 0.07% : 0.000002s : 22: predicate.reset_defer_inline 0.90% : 0.000025s : 133: predicate.reshape_eliminate 0.24% : 0.000007s : 44: predicate.row_tensor_add_zeros_like 0.17% : 0.000005s : 22: predicate.row_tensor_eliminate 0.36% : 0.000010s : 44: predicate.same_eliminate 0.15% : 0.000004s : 50: predicate.set_cell_output_no_recompute 0.26% : 0.000007s : 45: predicate.shard_identity_eliminate 0.29% : 0.000008s : 44: predicate.special_op_eliminate 0.42% : 0.000012s : 73: predicate.specialize_transform 0.34% : 0.000010s : 44: predicate.split_environ_get_set_with_tuple_value 0.27% : 0.000008s : 45: predicate.stack_unstack_eliminate 0.14% : 0.000004s : 22: predicate.switch_call_monad_eliminater 1.31% : 0.000037s : 219: predicate.switch_defer_inline 1.78% : 0.000050s : 263: predicate.switch_layer_defer_inline 31.66% : 0.000893s : 686: predicate.switch_simplify 0.74% : 0.000021s : 133: predicate.tile_eliminate 0.72% : 0.000020s : 133: predicate.transpose_eliminate 1.26% : 0.000036s : 177: predicate.tuple_list_convert_item_index_to_positive 1.16% : 0.000033s : 177: predicate.tuple_list_get_item_const_eliminator 1.11% : 0.000031s : 177: predicate.tuple_list_get_item_depend_reorder 2.06% : 0.000058s : 263: predicate.tuple_list_get_item_eliminator 1.12% : 0.000032s : 177: predicate.tuple_list_get_set_item_eliminator 1.56% : 0.000044s : 250: predicate.tuple_list_set_item_eliminator 1.20% : 0.000034s : 190: predicate.tuple_to_list_eliminator_ 1.66% : 0.000047s : 325: predicate.updatestate_pure_node_eliminater 2.11% : 0.000059s : 398: predicate.updatestate_useless_node_eliminater 0.14% : 0.000004s : 22: predicate.value_based_eliminate 0.30% : 0.000008s : 45: predicate.virtual_dataset_eliminate 0.26% : 0.000007s : 45: predicate.virtual_output_eliminate 0.12% : 0.000003s : 22: predicate.virtual_view_grad_eliminate 0.20% : 0.000006s : 22: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.197576 135 88.58% : 0.175008s : 55: func_graph_cloner_run.FuncGraphClonerGraph 11.42% : 0.022568s : 80: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 8.969004 209 0.00% : 0.000004s : 1: ForceFp32Comm 3.06% : 0.274433s : 1: add_attr 3.06% : 0.274355s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.00% : 0.000234s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.01% : 0.001180s : 1: auto_monad 0.00% : 0.000112s : 1: auto_monad_reorder 0.00% : 0.000006s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.01% : 0.000681s : 1: bootstrap 0.00% : 0.000060s : 1: cconv 0.00% : 0.000007s : 1: comm_op_add_attrs 0.00% : 0.000058s : 1: control_data_broadcast_order 0.00% : 0.000093s : 1: convert_after_rewriter 0.00% : 0.000138s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000055s : 1: environ_conv 0.00% : 0.000339s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000008s : 1: get_jit_bprop_graph 0.00% : 0.000051s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000010s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000013s : 1: label_micro_interleaved_index 0.01% : 0.000897s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.01% : 0.001108s : 1: mutable_eliminate 0.00% : 0.000021s : 1: offloading_packed_experts 0.00% : 0.000069s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000098s : 1: opt.transform.mutable_eliminate 2.72% : 0.244196s : 95: opt.transform.opt_a 0.00% : 0.000219s : 1: opt.transform.opt_after_cconv 0.00% : 0.000116s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000784s : 28: opt.transform.opt_b 0.00% : 0.000347s : 2: opt.transform.opt_trans_graph 0.00% : 0.000315s : 4: opt.transform.symbol_engine_opt 8.02% : 0.719030s : 1: opt_a 0.01% : 0.000518s : 1: opt_after_cconv 2.07% : 0.185739s : 1: opt_after_jit_grad 0.01% : 0.001213s : 1: opt_b 8.12% : 0.728415s : 1: optimize 0.00% : 0.000061s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000043s : 1: order_py_execute_after_rewriter 0.00% : 0.000103s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000017s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000007s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000019s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000011s : 1: parallel-infer-symbol 0.00% : 0.000005s : 1: parallel-infer-symbol-second 0.00% : 0.000010s : 1: partial_unused_args_eliminate 0.00% : 0.000009s : 1: pipeline_parallel_scheduler 0.00% : 0.000008s : 1: pipeline_split 0.00% : 0.000315s : 1: pre_auto_parallel 0.01% : 0.000608s : 1: py_interpret_to_execute 0.00% : 0.000057s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000200s : 1: remove_dup_value 2.54% : 0.227456s : 1: renormalize.infer 2.69% : 0.241708s : 1: renormalize.specialize 0.00% : 0.000009s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000032s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.001081s : 1: rewriter_after_opt_a 0.01% : 0.001188s : 1: rewriter_before_opt_a 0.00% : 0.000007s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000008s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000029s : 1: swap_dp_allreduce_reducescatter 0.01% : 0.000804s : 1: symbol_engine_optimizer 0.00% : 0.000419s : 1: tuple_transform 67.56% : 6.059750s : 1: type_inference . [hook] pytest_runtest_teardown:test_qbmm_qkv_1408_128_128[True-1024-32] tests/st/infer/ops/test_internal_ops/test_qbmm_split.py::test_qbmm_qkv_1408_128_128[True-1024-32],max_mem:116.0M [WARNING] ME(166984:281473373798192,MainProcess):2026-01-29-17:38:59.706.367 [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 = 5.91167, [21] [bootstrap]: 0.00066287 [type_inference]: 5.11828 [event_method]: 0.00027512 [auto_monad]: 0.00100538 [graph_reusing]: 5.04e-05 [inline]: 4.06001e-06 [add_attr]: 0.00664319, [1] [add_attr_with_inline]: 0.00662214, [1] [Cycle 1]: 0.00053551, [2] [tag_attr]: 0.00038303 [meta_addattr_fg_expand]: 8.566e-05 [parallel-infer-symbol]: 4.27998e-06 [pre_auto_parallel]: 0.00037632 [insert-virtual-dataset]: 5.86e-06 [parallel-infer-symbol-second]: 1.17e-06 [dataset_repeat_opt]: 2.49001e-06 [pipeline_split]: 4.35e-06 [optimize]: 0.783092, [53] [py_interpret_to_execute]: 0.00046398 [rewriter_before_opt_a]: 0.00105247 [opt_a]: 0.775249, [2] [Cycle 1]: 0.772438, [45] [expand_dump_flag]: 6.59e-05 [switch_simplify]: 0.201928 [loop_unroll]: 0.00057533 [a_1]: 0.0138366 [with_stream_mark]: 0.00115409 [recompute_prepare]: 7.605e-05 [updatestate_depend_eliminate]: 4.613e-05 [updatestate_assign_eliminate]: 2.252e-05 [updatestate_loads_eliminate]: 3.195e-05 [parameter_eliminate]: 3.50998e-06 [a_2]: 0.205498 [accelerated_algorithm]: 0.00010674 [shard]: 3.86001e-06 [meta_shard_fg_expand]: 5.018e-05 [shard_inline]: 3.081e-05 [merge_send_recv]: 2.833e-05 [auto_parallel]: 2.626e-05 [parallel]: 4.319e-05 [flash_sp]: 1.832e-05 [merge_comm]: 1.813e-05 [allreduce_fusion]: 1.718e-05 [matmul_add_comm_reduction]: 3.073e-05 [allreduce_slice_to_reducescatter]: 8.80013e-07 [virtual_shard_identity]: 4.205e-05 [virtual_dataset]: 3.714e-05 [get_grad_eliminate_]: 3.747e-05 [virtual_output]: 3.672e-05 [merge_forward]: 2.041e-05 [cell_reuse_recompute_pass]: 3.53e-06 [offload_activation]: 2.994e-05 [cell_reuse_handle_not_recompute_node_pass]: 7.708e-05 [merge_recompute_call_nodes]: 1.69e-06 [before_grad]: 6.378e-05 [set_forward_comm_id_for_comm_node_pass]: 1.846e-05 [meta_fg_expand]: 2.622e-05 [flash_sp_send_recv_attached]: 6.01998e-06 [receive_attached]: 2.07001e-06 [after_resolve]: 5.327e-05 [a_after_grad]: 6.473e-05 [renormalize]: 0.346305 [add_forward_monad_depend]: 1.778e-05 [auto_monad_grad]: 3.38e-06 [auto_monad_eliminator]: 0.00011749 [cse]: 0.00092791 [a_3]: 0.00022584 [Cycle 2]: 0.00278809, [45] [expand_dump_flag]: 3.83001e-06 [switch_simplify]: 3.061e-05 [loop_unroll]: 2.985e-05 [a_1]: 0.00088504 [with_stream_mark]: 4.132e-05 [recompute_prepare]: 2.863e-05 [updatestate_depend_eliminate]: 2.058e-05 [updatestate_assign_eliminate]: 1.55e-05 [updatestate_loads_eliminate]: 2.118e-05 [parameter_eliminate]: 3.07997e-06 [a_2]: 0.00046796 [accelerated_algorithm]: 4.083e-05 [shard]: 2.93e-06 [meta_shard_fg_expand]: 1.136e-05 [shard_inline]: 2.644e-05 [merge_send_recv]: 2.798e-05 [auto_parallel]: 2.262e-05 [parallel]: 1.091e-05 [flash_sp]: 5.03002e-06 [merge_comm]: 1.804e-05 [allreduce_fusion]: 1.495e-05 [matmul_add_comm_reduction]: 2.899e-05 [allreduce_slice_to_reducescatter]: 1.30999e-06 [virtual_shard_identity]: 2.888e-05 [virtual_dataset]: 2.787e-05 [get_grad_eliminate_]: 2.54e-05 [virtual_output]: 2.548e-05 [merge_forward]: 1.558e-05 [cell_reuse_recompute_pass]: 3.25e-06 [offload_activation]: 2.636e-05 [cell_reuse_handle_not_recompute_node_pass]: 5.076e-05 [merge_recompute_call_nodes]: 2.01e-06 [before_grad]: 4.553e-05 [set_forward_comm_id_for_comm_node_pass]: 1.891e-05 [meta_fg_expand]: 1.254e-05 [flash_sp_send_recv_attached]: 2.37999e-06 [receive_attached]: 2.52001e-06 [after_resolve]: 3.955e-05 [a_after_grad]: 4.338e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 4.67998e-06 [auto_monad_grad]: 2.74999e-06 [auto_monad_eliminator]: 8.993e-05 [cse]: 0.00011057 [a_3]: 0.00018294 [py_interpret_to_execute_after_opt_a]: 4.803e-05 [slice_cell_reuse_recomputed_activation]: 3.06001e-06 [rewriter_after_opt_a]: 0.0009912 [convert_after_rewriter]: 3.763e-05 [order_py_execute_after_rewriter]: 1.669e-05 [mutable_eliminate]: 0.00094147 [opt_b]: 0.0010334, [1] [Cycle 1]: 0.00102379, [7] [b_1]: 0.0007452 [b_2]: 3.347e-05 [updatestate_depend_eliminate]: 2.445e-05 [updatestate_assign_eliminate]: 1.577e-05 [updatestate_loads_eliminate]: 2.187e-05 [renormalize]: 7.39994e-07 [cse]: 0.00013071 [optimize_parallel_all_gather_comm]: 4.538e-05 [overlap_param_gather]: 2.00002e-06 [cconv]: 4.502e-05 [loop_unroll]: 0.00058382 [opt_after_cconv]: 0.00040644, [1] [Cycle 1]: 0.00039895, [7] [c_1]: 0.00017978 [parameter_eliminate]: 5.74e-06 [updatestate_depend_eliminate]: 2.088e-05 [updatestate_assign_eliminate]: 1.465e-05 [updatestate_loads_eliminate]: 1.773e-05 [cse]: 9.135e-05 [renormalize]: 3.50003e-07 [remove_dup_value]: 0.0001513 [tuple_transform]: 0.00029589, [1] [Cycle 1]: 0.00028988, [4] [d_1]: 0.00023612 [none_parameter_eliminate]: 2.14999e-06 [renormalize]: 3.4002e-07 [switch_simplify]: 2.795e-05 [partial_unused_args_eliminate]: 2.31e-06 [add_recomputation]: 0.00018506 [cse_after_recomputation]: 0.00013722, [1] [Cycle 1]: 0.00013146, [1] [cse]: 8.36e-05 [environ_conv]: 3.467e-05 [swap_dp_allreduce_reducescatter]: 2.272e-05 [bias_add_comm_swap]: 2.60002e-06 [label_micro_interleaved_index]: 4.92e-06 [label_fine_grained_interleaved_index]: 3.34001e-06 [merge_cast_opt]: 1.72999e-06 [slice_recompute_activation]: 2.55002e-06 [micro_interleaved_order_control]: 2.78998e-06 [assign_add_opt]: 1.89999e-06 [ForceFp32Comm]: 1.10999e-06 [remove_cast_before_assign_add]: 1.09e-06 [full_micro_interleaved_order_control]: 2.48e-06 [reorder_send_recv_between_fp_bp]: 2.83998e-06 [comm_op_add_attrs]: 1.14e-06 [add_comm_op_reuse_tag]: 1.10001e-06 [interleave_split_concat_branches]: 1.55001e-06 [interleave_parallel_branches]: 1.27e-06 [overlap_opt_shard_in_pipeline]: 1.50999e-06 [overlap_opt_shard_grad_in_pipeline]: 1.86e-06 [control_data_broadcast_order]: 4.328e-05 [grouped_pairwise_exchange_alltoall]: 1.62001e-06 [offloading_packed_experts]: 1.329e-05 [overlap_recompute_and_grad_model_parallel]: 1.223e-05 [overlap_grad_matmul_and_grad_allreduce]: 1.20999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.39e-06 [overlap_recompute_comm]: 2.71e-06 [overlap_grad_ring_attention]: 1.14e-05 [overlap_grad_flash_sp]: 6.185e-05 [begin_end_overlap_inline]: 5.40022e-07 [split_matmul_comm_elemetwise]: 2.32999e-06 [split_layernorm_comm]: 1.81e-06 [handle_group_info]: 1.13001e-06 [symbol_engine_optimizer]: 0.00079022, [1] [Cycle 1]: 0.00078166, [6] [build]: 0.00031206 [elim_shapecalc]: 3.969e-05 [elim_not_effective]: 6.023e-05 [opt_reshape]: 4.267e-05 [fold_const_symbol]: 0.00026677 [renormalize]: 6.49976e-07 [detach_backward]: 3.17002e-06 [pipeline_parallel_scheduler]: 3.74002e-06 [auto_monad_reorder]: 0.00010058 [get_jit_bprop_graph]: 2.56e-06 [rewriter_after_jit_bprop_graph]: 8.34002e-06 [opt_after_jit_grad]: 0.00074794 [validate]: 0.00012534 Sums bootstrap : 0.000663s : 0.01% type_inference : 5.118282s : 86.70% event_method : 0.000275s : 0.00% auto_monad : 0.001005s : 0.02% graph_reusing : 0.000050s : 0.00% inline : 0.000004s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000383s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000086s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000376s : 0.01% insert-virtual-dataset : 0.000006s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000004s : 0.00% optimize.py_interpret_to_execute : 0.000464s : 0.01% optimize.rewriter_before_opt_a : 0.001052s : 0.02% optimize.opt_a.expand_dump_flag : 0.000070s : 0.00% optimize.opt_a.switch_simplify : 0.201959s : 3.42% optimize.opt_a.loop_unroll : 0.000605s : 0.01% optimize.opt_a.a_1 : 0.014722s : 0.25% optimize.opt_a.with_stream_mark : 0.001195s : 0.02% optimize.opt_a.recompute_prepare : 0.000105s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000067s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000038s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000053s : 0.00% optimize.opt_a.parameter_eliminate : 0.000007s : 0.00% optimize.opt_a.a_2 : 0.205966s : 3.49% optimize.opt_a.accelerated_algorithm : 0.000148s : 0.00% optimize.opt_a.shard : 0.000007s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000062s : 0.00% optimize.opt_a.shard_inline : 0.000057s : 0.00% optimize.opt_a.merge_send_recv : 0.000056s : 0.00% optimize.opt_a.auto_parallel : 0.000049s : 0.00% optimize.opt_a.parallel : 0.000054s : 0.00% optimize.opt_a.flash_sp : 0.000023s : 0.00% optimize.opt_a.merge_comm : 0.000036s : 0.00% optimize.opt_a.allreduce_fusion : 0.000032s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000060s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000071s : 0.00% optimize.opt_a.virtual_dataset : 0.000065s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000063s : 0.00% optimize.opt_a.virtual_output : 0.000062s : 0.00% optimize.opt_a.merge_forward : 0.000036s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000007s : 0.00% optimize.opt_a.offload_activation : 0.000056s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000128s : 0.00% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000109s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000037s : 0.00% optimize.opt_a.meta_fg_expand : 0.000039s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000008s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000093s : 0.00% optimize.opt_a.a_after_grad : 0.000108s : 0.00% optimize.opt_a.renormalize : 0.346305s : 5.87% optimize.opt_a.add_forward_monad_depend : 0.000022s : 0.00% optimize.opt_a.auto_monad_grad : 0.000006s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000207s : 0.00% optimize.opt_a.cse : 0.001038s : 0.02% optimize.opt_a.a_3 : 0.000409s : 0.01% optimize.py_interpret_to_execute_after_opt_a : 0.000048s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000991s : 0.02% optimize.convert_after_rewriter : 0.000038s : 0.00% optimize.order_py_execute_after_rewriter : 0.000017s : 0.00% optimize.mutable_eliminate : 0.000941s : 0.02% optimize.opt_b.b_1 : 0.000745s : 0.01% optimize.opt_b.b_2 : 0.000033s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000024s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000016s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000022s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000131s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000045s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000045s : 0.00% optimize.loop_unroll : 0.000584s : 0.01% optimize.opt_after_cconv.c_1 : 0.000180s : 0.00% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000021s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000015s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000018s : 0.00% optimize.opt_after_cconv.cse : 0.000091s : 0.00% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000151s : 0.00% optimize.tuple_transform.d_1 : 0.000236s : 0.00% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000028s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000185s : 0.00% optimize.cse_after_recomputation.cse : 0.000084s : 0.00% optimize.environ_conv : 0.000035s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000023s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000002s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000043s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000013s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000012s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000011s : 0.00% optimize.overlap_grad_flash_sp : 0.000062s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000312s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000040s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000060s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000043s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000267s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000001s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000004s : 0.00% auto_monad_reorder : 0.000101s : 0.00% get_jit_bprop_graph : 0.000003s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.000748s : 0.01% validate : 0.000125s : 0.00% Time group info: ------[substitution.] 0.005683 608 0.34% : 0.000019s : 8: substitution.depend_value_elim 0.25% : 0.000014s : 17: substitution.elim_not_effective 0.58% : 0.000033s : 12: substitution.float_tuple_getitem_switch 0.61% : 0.000035s : 17: substitution.fold_const_symbol 0.31% : 0.000018s : 22: substitution.graph_param_transform 70.21% : 0.003990s : 71: substitution.inline 0.29% : 0.000017s : 35: substitution.j_node_and_user_rematch 0.97% : 0.000055s : 6: substitution.less_batch_normalization 1.33% : 0.000075s : 2: substitution.list_to_tuple_eliminator_ 0.19% : 0.000011s : 18: substitution.load_eliminater 0.67% : 0.000038s : 21: substitution.minmaximum_grad 0.15% : 0.000008s : 2: substitution.opt_reshape 0.44% : 0.000025s : 35: substitution.remove_not_recompute_node 0.23% : 0.000013s : 8: substitution.replace_old_param 1.58% : 0.000090s : 18: substitution.reshape_eliminate 1.63% : 0.000092s : 24: substitution.switch_simplify 3.60% : 0.000204s : 51: substitution.tuple_list_convert_item_index_to_positive 2.33% : 0.000133s : 51: substitution.tuple_list_get_item_const_eliminator 2.77% : 0.000157s : 51: substitution.tuple_list_get_item_depend_reorder 6.04% : 0.000343s : 73: substitution.tuple_list_get_item_eliminator 2.70% : 0.000153s : 51: substitution.tuple_list_get_set_item_eliminator 0.20% : 0.000011s : 6: substitution.updatestate_pure_node_eliminater 2.58% : 0.000147s : 9: substitution.updatestate_useless_node_eliminater ------[type_inference.] 5.117690 2 91.70% : 4.692715s : 1: type_inference.infer 8.30% : 0.424975s : 1: type_inference.specialize ------[replace.] 0.202163 115 0.05% : 0.000100s : 5: replace.depend_value_elim 0.43% : 0.000862s : 71: replace.inline 0.03% : 0.000056s : 2: replace.list_to_tuple_eliminator_ 99.40% : 0.200945s : 24: replace.switch_simplify 0.10% : 0.000200s : 13: replace.tuple_list_get_item_eliminator ------[match.] 0.004134 115 0.06% : 0.000003s : 5: match.depend_value_elim 95.23% : 0.003937s : 71: match.inline 1.73% : 0.000071s : 2: match.list_to_tuple_eliminator_ 1.78% : 0.000074s : 24: match.switch_simplify 1.20% : 0.000050s : 13: match.tuple_list_get_item_eliminator ------[predicate.] 0.002139 10915 0.91% : 0.000019s : 133: predicate.accumulaten_eliminater 0.43% : 0.000009s : 22: predicate.ad_related_special_op_eliminate 0.49% : 0.000011s : 73: predicate.addn_check_dump 1.10% : 0.000024s : 133: predicate.addn_zero_filter 0.90% : 0.000019s : 133: predicate.adjust_all_reduce_mul_add 2.22% : 0.000047s : 206: predicate.arithmetic_simplify 1.06% : 0.000023s : 133: predicate.cast_eliminate 0.34% : 0.000007s : 44: predicate.check_bprop_eliminate 0.52% : 0.000011s : 73: predicate.compare_switch_simplify 0.08% : 0.000002s : 22: predicate.const_output_eliminate 0.64% : 0.000014s : 76: predicate.depend_value_elim 0.99% : 0.000021s : 133: predicate.dict_get_item_const_eliminator 1.21% : 0.000026s : 133: predicate.dict_get_item_eliminator 0.96% : 0.000021s : 133: predicate.dict_set_item_eliminator 0.38% : 0.000008s : 44: predicate.dumpgradient_eliminate 0.08% : 0.000002s : 22: predicate.elim_not_effective 0.18% : 0.000004s : 22: predicate.elim_shapecalc_of_broadcastargs 1.11% : 0.000024s : 155: predicate.environ_add_const_eliminate 1.12% : 0.000024s : 155: predicate.environ_get_add_eliminate 1.11% : 0.000024s : 155: predicate.environ_get_depend_swap 2.65% : 0.000057s : 228: predicate.environ_get_eliminate 1.12% : 0.000024s : 155: predicate.environ_get_set_eliminate 1.57% : 0.000034s : 219: predicate.exchange_switch_depend_value 2.32% : 0.000050s : 219: predicate.float_depend_g_call 0.56% : 0.000012s : 73: predicate.float_environ_get_switch 0.73% : 0.000016s : 95: predicate.float_tuple_getitem_switch 0.09% : 0.000002s : 22: predicate.fold_const_symbol 0.34% : 0.000007s : 45: predicate.get_grad_eliminate 0.10% : 0.000002s : 22: predicate.graph_param_transform 0.52% : 0.000011s : 73: predicate.incorporate_call 0.48% : 0.000010s : 73: predicate.incorporate_call_switch 5.00% : 0.000107s : 514: predicate.inline 0.42% : 0.000009s : 45: predicate.inline_without_move 0.15% : 0.000003s : 45: predicate.j_node_and_user_rematch 0.47% : 0.000010s : 45: predicate.less_batch_normalization 1.53% : 0.000033s : 192: predicate.list_to_tuple_eliminator_ 2.44% : 0.000052s : 325: predicate.load_eliminater 0.39% : 0.000008s : 22: predicate.loop_unroll_after_grad 6.52% : 0.000139s : 324: predicate.loop_unroll_before_grad 1.34% : 0.000029s : 177: predicate.make_slice_get_slice_eliminator 0.68% : 0.000014s : 73: predicate.merge_addn 0.31% : 0.000007s : 44: predicate.micro_step_allgather_replace 0.31% : 0.000007s : 44: predicate.mini_step_allgather_replace 0.96% : 0.000021s : 133: predicate.minmaximum_grad 0.48% : 0.000010s : 22: predicate.mutable_eliminate 0.19% : 0.000004s : 22: predicate.opt_reshape 0.23% : 0.000005s : 22: predicate.parallel_virtual_node 4.20% : 0.000090s : 219: predicate.partial_defer_inline 1.24% : 0.000027s : 170: predicate.partial_eliminate 1.01% : 0.000022s : 133: predicate.print_const_string_wrapper 0.47% : 0.000010s : 68: predicate.reduce_all_const_elim 1.33% : 0.000028s : 133: predicate.reduce_eliminate 2.24% : 0.000048s : 325: predicate.redundant_stop_gradient_eliminater 0.15% : 0.000003s : 45: predicate.remove_not_recompute_node 0.91% : 0.000019s : 192: predicate.replace_applicator 0.18% : 0.000004s : 45: predicate.replace_old_param 0.10% : 0.000002s : 22: predicate.reset_defer_inline 1.22% : 0.000026s : 133: predicate.reshape_eliminate 0.32% : 0.000007s : 44: predicate.row_tensor_add_zeros_like 0.20% : 0.000004s : 22: predicate.row_tensor_eliminate 0.41% : 0.000009s : 44: predicate.same_eliminate 0.25% : 0.000005s : 50: predicate.set_cell_output_no_recompute 0.36% : 0.000008s : 45: predicate.shard_identity_eliminate 0.35% : 0.000007s : 44: predicate.special_op_eliminate 1.99% : 0.000043s : 73: predicate.specialize_transform 0.38% : 0.000008s : 44: predicate.split_environ_get_set_with_tuple_value 0.36% : 0.000008s : 45: predicate.stack_unstack_eliminate 0.15% : 0.000003s : 22: predicate.switch_call_monad_eliminater 1.75% : 0.000038s : 219: predicate.switch_defer_inline 1.98% : 0.000042s : 263: predicate.switch_layer_defer_inline 9.43% : 0.000202s : 686: predicate.switch_simplify 0.96% : 0.000020s : 133: predicate.tile_eliminate 0.94% : 0.000020s : 133: predicate.transpose_eliminate 1.56% : 0.000033s : 177: predicate.tuple_list_convert_item_index_to_positive 1.50% : 0.000032s : 177: predicate.tuple_list_get_item_const_eliminator 1.46% : 0.000031s : 177: predicate.tuple_list_get_item_depend_reorder 2.71% : 0.000058s : 263: predicate.tuple_list_get_item_eliminator 1.64% : 0.000035s : 177: predicate.tuple_list_get_set_item_eliminator 2.13% : 0.000046s : 250: predicate.tuple_list_set_item_eliminator 1.42% : 0.000030s : 190: predicate.tuple_to_list_eliminator_ 2.94% : 0.000063s : 325: predicate.updatestate_pure_node_eliminater 2.86% : 0.000061s : 398: predicate.updatestate_useless_node_eliminater 0.19% : 0.000004s : 22: predicate.value_based_eliminate 0.34% : 0.000007s : 45: predicate.virtual_dataset_eliminate 0.33% : 0.000007s : 45: predicate.virtual_output_eliminate 0.16% : 0.000003s : 22: predicate.virtual_view_grad_eliminate 0.19% : 0.000004s : 22: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.633749 135 66.51% : 0.421484s : 55: func_graph_cloner_run.FuncGraphClonerGraph 33.49% : 0.212265s : 80: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 7.269584 209 0.00% : 0.000004s : 1: ForceFp32Comm 0.09% : 0.006653s : 1: add_attr 0.09% : 0.006627s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.00% : 0.000190s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.01% : 0.001023s : 1: auto_monad 0.00% : 0.000107s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000005s : 1: bias_add_comm_swap 0.01% : 0.000717s : 1: bootstrap 0.00% : 0.000049s : 1: cconv 0.00% : 0.000006s : 1: comm_op_add_attrs 0.00% : 0.000047s : 1: control_data_broadcast_order 0.00% : 0.000046s : 1: convert_after_rewriter 0.00% : 0.000140s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000041s : 1: environ_conv 0.00% : 0.000290s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000009s : 1: get_jit_bprop_graph 0.00% : 0.000057s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000010s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.01% : 0.000595s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.01% : 0.000956s : 1: mutable_eliminate 0.00% : 0.000016s : 1: offloading_packed_experts 0.00% : 0.000048s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000067s : 1: opt.transform.mutable_eliminate 3.03% : 0.220404s : 95: opt.transform.opt_a 0.00% : 0.000177s : 1: opt.transform.opt_after_cconv 0.00% : 0.000125s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000728s : 28: opt.transform.opt_b 0.00% : 0.000259s : 2: opt.transform.opt_trans_graph 0.01% : 0.000402s : 4: opt.transform.symbol_engine_opt 10.66% : 0.775255s : 1: opt_a 0.01% : 0.000410s : 1: opt_after_cconv 0.01% : 0.000760s : 1: opt_after_jit_grad 0.01% : 0.001037s : 1: opt_b 10.77% : 0.783101s : 1: optimize 0.00% : 0.000050s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000021s : 1: order_py_execute_after_rewriter 0.00% : 0.000065s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000014s : 1: overlap_grad_ring_attention 0.00% : 0.000004s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000006s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000015s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000012s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000008s : 1: partial_unused_args_eliminate 0.00% : 0.000007s : 1: pipeline_parallel_scheduler 0.00% : 0.000007s : 1: pipeline_split 0.01% : 0.000387s : 1: pre_auto_parallel 0.01% : 0.000480s : 1: py_interpret_to_execute 0.00% : 0.000053s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000156s : 1: remove_dup_value 0.16% : 0.011764s : 1: renormalize.infer 4.60% : 0.334510s : 1: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.001007s : 1: rewriter_after_opt_a 0.01% : 0.001073s : 1: rewriter_before_opt_a 0.00% : 0.000007s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000007s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000026s : 1: swap_dp_allreduce_reducescatter 0.01% : 0.000795s : 1: symbol_engine_optimizer 0.00% : 0.000299s : 1: tuple_transform 70.41% : 5.118317s : 1: type_inference . [hook] pytest_runtest_teardown:test_qbmm_qkv_1408_128_128[True-11264-32] tests/st/infer/ops/test_internal_ops/test_qbmm_split.py::test_qbmm_qkv_1408_128_128[True-11264-32],max_mem:118.0M [WARNING] ME(166984:281473373798192,MainProcess):2026-01-29-17:39:26.321.922 [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 = 5.93393, [21] [bootstrap]: 0.00077858 [type_inference]: 5.09081 [event_method]: 0.00036186 [auto_monad]: 0.00106971 [graph_reusing]: 6.306e-05 [inline]: 3.56999e-06 [add_attr]: 0.00668052, [1] [add_attr_with_inline]: 0.00665715, [1] [Cycle 1]: 0.00043546, [2] [tag_attr]: 0.0003129 [meta_addattr_fg_expand]: 6.777e-05 [parallel-infer-symbol]: 4.47998e-06 [pre_auto_parallel]: 0.00037752 [insert-virtual-dataset]: 3.18998e-06 [parallel-infer-symbol-second]: 3.90998e-06 [dataset_repeat_opt]: 2.77002e-06 [pipeline_split]: 1.80001e-06 [optimize]: 0.586898, [53] [py_interpret_to_execute]: 0.00038926 [rewriter_before_opt_a]: 0.00110333 [opt_a]: 0.579099, [2] [Cycle 1]: 0.576352, [45] [expand_dump_flag]: 4.932e-05 [switch_simplify]: 0.00137055 [loop_unroll]: 0.00053407 [a_1]: 0.20726 [with_stream_mark]: 0.00066098 [recompute_prepare]: 6.306e-05 [updatestate_depend_eliminate]: 3.626e-05 [updatestate_assign_eliminate]: 2.132e-05 [updatestate_loads_eliminate]: 2.995e-05 [parameter_eliminate]: 3.41001e-06 [a_2]: 0.00104954 [accelerated_algorithm]: 7.249e-05 [shard]: 3.26001e-06 [meta_shard_fg_expand]: 3.985e-05 [shard_inline]: 2.821e-05 [merge_send_recv]: 2.523e-05 [auto_parallel]: 2.543e-05 [parallel]: 3.969e-05 [flash_sp]: 1.738e-05 [merge_comm]: 1.66e-05 [allreduce_fusion]: 1.601e-05 [matmul_add_comm_reduction]: 4.24e-05 [allreduce_slice_to_reducescatter]: 9.09989e-07 [virtual_shard_identity]: 3.439e-05 [virtual_dataset]: 3.081e-05 [get_grad_eliminate_]: 0.00012887 [virtual_output]: 4.49e-05 [merge_forward]: 3.927e-05 [cell_reuse_recompute_pass]: 3.4e-06 [offload_activation]: 4.862e-05 [cell_reuse_handle_not_recompute_node_pass]: 9.273e-05 [merge_recompute_call_nodes]: 3.28e-06 [before_grad]: 6.424e-05 [set_forward_comm_id_for_comm_node_pass]: 2.664e-05 [meta_fg_expand]: 6.301e-05 [flash_sp_send_recv_attached]: 1.431e-05 [receive_attached]: 2.29001e-06 [after_resolve]: 5.338e-05 [a_after_grad]: 6.103e-05 [renormalize]: 0.221244 [add_forward_monad_depend]: 1.592e-05 [auto_monad_grad]: 3.60998e-06 [auto_monad_eliminator]: 0.00010817 [cse]: 0.00049932 [a_3]: 0.00043165 [Cycle 2]: 0.00272377, [45] [expand_dump_flag]: 2.88998e-06 [switch_simplify]: 3.284e-05 [loop_unroll]: 2.912e-05 [a_1]: 0.00091598 [with_stream_mark]: 3.782e-05 [recompute_prepare]: 2.989e-05 [updatestate_depend_eliminate]: 1.832e-05 [updatestate_assign_eliminate]: 1.587e-05 [updatestate_loads_eliminate]: 2.127e-05 [parameter_eliminate]: 2.58e-06 [a_2]: 0.0004036 [accelerated_algorithm]: 3.568e-05 [shard]: 2.35002e-06 [meta_shard_fg_expand]: 9.14e-06 [shard_inline]: 4.827e-05 [merge_send_recv]: 2.58e-05 [auto_parallel]: 2.324e-05 [parallel]: 1.039e-05 [flash_sp]: 5.31002e-06 [merge_comm]: 1.723e-05 [allreduce_fusion]: 1.454e-05 [matmul_add_comm_reduction]: 2.908e-05 [allreduce_slice_to_reducescatter]: 1.19e-06 [virtual_shard_identity]: 4.633e-05 [virtual_dataset]: 2.949e-05 [get_grad_eliminate_]: 2.476e-05 [virtual_output]: 2.466e-05 [merge_forward]: 1.648e-05 [cell_reuse_recompute_pass]: 3.82002e-06 [offload_activation]: 2.772e-05 [cell_reuse_handle_not_recompute_node_pass]: 4.855e-05 [merge_recompute_call_nodes]: 1.86e-06 [before_grad]: 4.439e-05 [set_forward_comm_id_for_comm_node_pass]: 1.832e-05 [meta_fg_expand]: 1.144e-05 [flash_sp_send_recv_attached]: 1.77999e-06 [receive_attached]: 2.51998e-06 [after_resolve]: 3.817e-05 [a_after_grad]: 4.55e-05 [renormalize]: 8.00064e-08 [add_forward_monad_depend]: 3.6e-06 [auto_monad_grad]: 2.79999e-06 [auto_monad_eliminator]: 6.289e-05 [cse]: 0.00010316 [a_3]: 0.00017742 [py_interpret_to_execute_after_opt_a]: 4.508e-05 [slice_cell_reuse_recomputed_activation]: 2.91999e-06 [rewriter_after_opt_a]: 0.00110156 [convert_after_rewriter]: 3.148e-05 [order_py_execute_after_rewriter]: 1.55e-05 [mutable_eliminate]: 0.0008938 [opt_b]: 0.0010074, [1] [Cycle 1]: 0.00099821, [7] [b_1]: 0.00072613 [b_2]: 2.897e-05 [updatestate_depend_eliminate]: 2.183e-05 [updatestate_assign_eliminate]: 1.449e-05 [updatestate_loads_eliminate]: 2.29e-05 [renormalize]: 8.80013e-07 [cse]: 0.00013465 [optimize_parallel_all_gather_comm]: 4.564e-05 [overlap_param_gather]: 2.09e-06 [cconv]: 4.973e-05 [loop_unroll]: 0.00067306 [opt_after_cconv]: 0.00037729, [1] [Cycle 1]: 0.00037, [7] [c_1]: 0.00017404 [parameter_eliminate]: 6.17001e-06 [updatestate_depend_eliminate]: 2.043e-05 [updatestate_assign_eliminate]: 1.346e-05 [updatestate_loads_eliminate]: 1.785e-05 [cse]: 9.595e-05 [renormalize]: 6.59988e-07 [remove_dup_value]: 0.0001851 [tuple_transform]: 0.0003303, [1] [Cycle 1]: 0.00032347, [4] [d_1]: 0.00026582 [none_parameter_eliminate]: 2.78e-06 [renormalize]: 4.39992e-07 [switch_simplify]: 2.816e-05 [partial_unused_args_eliminate]: 2.32001e-06 [add_recomputation]: 0.00021623 [cse_after_recomputation]: 8.994e-05, [1] [Cycle 1]: 8.361e-05, [1] [cse]: 7.621e-05 [environ_conv]: 3.99e-05 [swap_dp_allreduce_reducescatter]: 2.194e-05 [bias_add_comm_swap]: 4.07e-06 [label_micro_interleaved_index]: 6.17999e-06 [label_fine_grained_interleaved_index]: 2.96001e-06 [merge_cast_opt]: 1.74e-06 [slice_recompute_activation]: 2.37999e-06 [micro_interleaved_order_control]: 2.90002e-06 [assign_add_opt]: 1.77001e-06 [ForceFp32Comm]: 9.20001e-07 [remove_cast_before_assign_add]: 1.11997e-06 [full_micro_interleaved_order_control]: 2.83e-06 [reorder_send_recv_between_fp_bp]: 2.96001e-06 [comm_op_add_attrs]: 1.22999e-06 [add_comm_op_reuse_tag]: 1.29e-06 [interleave_split_concat_branches]: 1.49998e-06 [interleave_parallel_branches]: 1.14e-06 [overlap_opt_shard_in_pipeline]: 1.68002e-06 [overlap_opt_shard_grad_in_pipeline]: 1.96e-06 [control_data_broadcast_order]: 4.672e-05 [grouped_pairwise_exchange_alltoall]: 1.71002e-06 [offloading_packed_experts]: 1.376e-05 [overlap_recompute_and_grad_model_parallel]: 1.182e-05 [overlap_grad_matmul_and_grad_allreduce]: 1.37e-06 [overlap_recompute_allgather_and_fa_grad]: 1.48002e-06 [overlap_recompute_comm]: 2.24999e-06 [overlap_grad_ring_attention]: 1.084e-05 [overlap_grad_flash_sp]: 8.25e-05 [begin_end_overlap_inline]: 6.10016e-07 [split_matmul_comm_elemetwise]: 2.73e-06 [split_layernorm_comm]: 2.30002e-06 [handle_group_info]: 1.10001e-06 [symbol_engine_optimizer]: 0.00059737, [1] [Cycle 1]: 0.00058887, [6] [build]: 0.00032084 [elim_shapecalc]: 4.31e-05 [elim_not_effective]: 6.325e-05 [opt_reshape]: 4.227e-05 [fold_const_symbol]: 6.831e-05 [renormalize]: 4.19997e-07 [detach_backward]: 2.91999e-06 [pipeline_parallel_scheduler]: 3.93001e-06 [auto_monad_reorder]: 9.255e-05 [get_jit_bprop_graph]: 2.21e-06 [rewriter_after_jit_bprop_graph]: 6.23998e-06 [opt_after_jit_grad]: 0.246227 [validate]: 0.00023296 Sums bootstrap : 0.000779s : 0.01% type_inference : 5.090806s : 88.01% event_method : 0.000362s : 0.01% auto_monad : 0.001070s : 0.02% graph_reusing : 0.000063s : 0.00% inline : 0.000004s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000313s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000068s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000378s : 0.01% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000004s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000389s : 0.01% optimize.rewriter_before_opt_a : 0.001103s : 0.02% optimize.opt_a.expand_dump_flag : 0.000052s : 0.00% optimize.opt_a.switch_simplify : 0.001403s : 0.02% optimize.opt_a.loop_unroll : 0.000563s : 0.01% optimize.opt_a.a_1 : 0.208176s : 3.60% optimize.opt_a.with_stream_mark : 0.000699s : 0.01% optimize.opt_a.recompute_prepare : 0.000093s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000055s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000037s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000051s : 0.00% optimize.opt_a.parameter_eliminate : 0.000006s : 0.00% optimize.opt_a.a_2 : 0.001453s : 0.03% optimize.opt_a.accelerated_algorithm : 0.000108s : 0.00% optimize.opt_a.shard : 0.000006s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000049s : 0.00% optimize.opt_a.shard_inline : 0.000076s : 0.00% optimize.opt_a.merge_send_recv : 0.000051s : 0.00% optimize.opt_a.auto_parallel : 0.000049s : 0.00% optimize.opt_a.parallel : 0.000050s : 0.00% optimize.opt_a.flash_sp : 0.000023s : 0.00% optimize.opt_a.merge_comm : 0.000034s : 0.00% optimize.opt_a.allreduce_fusion : 0.000031s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000071s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000081s : 0.00% optimize.opt_a.virtual_dataset : 0.000060s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000154s : 0.00% optimize.opt_a.virtual_output : 0.000070s : 0.00% optimize.opt_a.merge_forward : 0.000056s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000007s : 0.00% optimize.opt_a.offload_activation : 0.000076s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000141s : 0.00% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000109s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000045s : 0.00% optimize.opt_a.meta_fg_expand : 0.000074s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000016s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000092s : 0.00% optimize.opt_a.a_after_grad : 0.000107s : 0.00% optimize.opt_a.renormalize : 0.221244s : 3.82% optimize.opt_a.add_forward_monad_depend : 0.000020s : 0.00% optimize.opt_a.auto_monad_grad : 0.000006s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000171s : 0.00% optimize.opt_a.cse : 0.000602s : 0.01% optimize.opt_a.a_3 : 0.000609s : 0.01% optimize.py_interpret_to_execute_after_opt_a : 0.000045s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.001102s : 0.02% optimize.convert_after_rewriter : 0.000031s : 0.00% optimize.order_py_execute_after_rewriter : 0.000016s : 0.00% optimize.mutable_eliminate : 0.000894s : 0.02% optimize.opt_b.b_1 : 0.000726s : 0.01% optimize.opt_b.b_2 : 0.000029s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000022s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000014s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000023s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000135s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000046s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000050s : 0.00% optimize.loop_unroll : 0.000673s : 0.01% optimize.opt_after_cconv.c_1 : 0.000174s : 0.00% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000020s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000013s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000018s : 0.00% optimize.opt_after_cconv.cse : 0.000096s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000185s : 0.00% optimize.tuple_transform.d_1 : 0.000266s : 0.00% optimize.tuple_transform.none_parameter_eliminate : 0.000003s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000028s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000216s : 0.00% optimize.cse_after_recomputation.cse : 0.000076s : 0.00% optimize.environ_conv : 0.000040s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000022s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000006s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000047s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000014s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000012s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000011s : 0.00% optimize.overlap_grad_flash_sp : 0.000082s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000321s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000043s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000063s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000042s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000068s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000004s : 0.00% auto_monad_reorder : 0.000093s : 0.00% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.246227s : 4.26% validate : 0.000233s : 0.00% Time group info: ------[substitution.] 0.199452 608 0.01% : 0.000018s : 8: substitution.depend_value_elim 0.01% : 0.000016s : 17: substitution.elim_not_effective 0.01% : 0.000018s : 12: substitution.float_tuple_getitem_switch 0.02% : 0.000031s : 17: substitution.fold_const_symbol 0.01% : 0.000019s : 22: substitution.graph_param_transform 99.22% : 0.197897s : 71: substitution.inline 0.01% : 0.000017s : 35: substitution.j_node_and_user_rematch 0.02% : 0.000038s : 6: substitution.less_batch_normalization 0.04% : 0.000073s : 2: substitution.list_to_tuple_eliminator_ 0.01% : 0.000011s : 18: substitution.load_eliminater 0.02% : 0.000038s : 21: substitution.minmaximum_grad 0.00% : 0.000009s : 2: substitution.opt_reshape 0.01% : 0.000026s : 35: substitution.remove_not_recompute_node 0.01% : 0.000014s : 8: substitution.replace_old_param 0.05% : 0.000101s : 18: substitution.reshape_eliminate 0.04% : 0.000088s : 24: substitution.switch_simplify 0.10% : 0.000200s : 51: substitution.tuple_list_convert_item_index_to_positive 0.05% : 0.000109s : 51: substitution.tuple_list_get_item_const_eliminator 0.08% : 0.000163s : 51: substitution.tuple_list_get_item_depend_reorder 0.14% : 0.000271s : 73: substitution.tuple_list_get_item_eliminator 0.07% : 0.000138s : 51: substitution.tuple_list_get_set_item_eliminator 0.01% : 0.000013s : 6: substitution.updatestate_pure_node_eliminater 0.07% : 0.000143s : 9: substitution.updatestate_useless_node_eliminater ------[type_inference.] 5.089818 2 90.01% : 4.581455s : 1: type_inference.infer 9.99% : 0.508363s : 1: type_inference.specialize ------[replace.] 0.001696 115 3.97% : 0.000067s : 5: replace.depend_value_elim 53.54% : 0.000908s : 71: replace.inline 3.32% : 0.000056s : 2: replace.list_to_tuple_eliminator_ 30.86% : 0.000523s : 24: replace.switch_simplify 8.31% : 0.000141s : 13: replace.tuple_list_get_item_eliminator ------[match.] 0.198022 115 0.00% : 0.000002s : 5: match.depend_value_elim 99.91% : 0.197838s : 71: match.inline 0.03% : 0.000069s : 2: match.list_to_tuple_eliminator_ 0.03% : 0.000068s : 24: match.switch_simplify 0.02% : 0.000046s : 13: match.tuple_list_get_item_eliminator ------[predicate.] 0.002084 10915 0.98% : 0.000020s : 133: predicate.accumulaten_eliminater 0.75% : 0.000016s : 22: predicate.ad_related_special_op_eliminate 0.48% : 0.000010s : 73: predicate.addn_check_dump 0.97% : 0.000020s : 133: predicate.addn_zero_filter 0.98% : 0.000020s : 133: predicate.adjust_all_reduce_mul_add 2.33% : 0.000049s : 206: predicate.arithmetic_simplify 1.08% : 0.000022s : 133: predicate.cast_eliminate 0.33% : 0.000007s : 44: predicate.check_bprop_eliminate 0.51% : 0.000011s : 73: predicate.compare_switch_simplify 0.08% : 0.000002s : 22: predicate.const_output_eliminate 0.56% : 0.000012s : 76: predicate.depend_value_elim 1.06% : 0.000022s : 133: predicate.dict_get_item_const_eliminator 3.36% : 0.000070s : 133: predicate.dict_get_item_eliminator 1.01% : 0.000021s : 133: predicate.dict_set_item_eliminator 0.39% : 0.000008s : 44: predicate.dumpgradient_eliminate 0.09% : 0.000002s : 22: predicate.elim_not_effective 0.21% : 0.000004s : 22: predicate.elim_shapecalc_of_broadcastargs 1.11% : 0.000023s : 155: predicate.environ_add_const_eliminate 1.31% : 0.000027s : 155: predicate.environ_get_add_eliminate 1.05% : 0.000022s : 155: predicate.environ_get_depend_swap 1.79% : 0.000037s : 228: predicate.environ_get_eliminate 1.19% : 0.000025s : 155: predicate.environ_get_set_eliminate 1.87% : 0.000039s : 219: predicate.exchange_switch_depend_value 2.42% : 0.000050s : 219: predicate.float_depend_g_call 0.48% : 0.000010s : 73: predicate.float_environ_get_switch 0.69% : 0.000014s : 95: predicate.float_tuple_getitem_switch 0.08% : 0.000002s : 22: predicate.fold_const_symbol 1.40% : 0.000029s : 45: predicate.get_grad_eliminate 0.10% : 0.000002s : 22: predicate.graph_param_transform 0.48% : 0.000010s : 73: predicate.incorporate_call 0.45% : 0.000009s : 73: predicate.incorporate_call_switch 5.15% : 0.000107s : 514: predicate.inline 0.45% : 0.000009s : 45: predicate.inline_without_move 0.16% : 0.000003s : 45: predicate.j_node_and_user_rematch 0.45% : 0.000009s : 45: predicate.less_batch_normalization 1.61% : 0.000034s : 192: predicate.list_to_tuple_eliminator_ 2.39% : 0.000050s : 325: predicate.load_eliminater 0.44% : 0.000009s : 22: predicate.loop_unroll_after_grad 4.94% : 0.000103s : 324: predicate.loop_unroll_before_grad 1.41% : 0.000029s : 177: predicate.make_slice_get_slice_eliminator 0.53% : 0.000011s : 73: predicate.merge_addn 0.30% : 0.000006s : 44: predicate.micro_step_allgather_replace 0.37% : 0.000008s : 44: predicate.mini_step_allgather_replace 0.98% : 0.000020s : 133: predicate.minmaximum_grad 0.39% : 0.000008s : 22: predicate.mutable_eliminate 0.18% : 0.000004s : 22: predicate.opt_reshape 0.18% : 0.000004s : 22: predicate.parallel_virtual_node 3.59% : 0.000075s : 219: predicate.partial_defer_inline 1.29% : 0.000027s : 170: predicate.partial_eliminate 0.96% : 0.000020s : 133: predicate.print_const_string_wrapper 0.48% : 0.000010s : 68: predicate.reduce_all_const_elim 1.48% : 0.000031s : 133: predicate.reduce_eliminate 2.32% : 0.000048s : 325: predicate.redundant_stop_gradient_eliminater 0.20% : 0.000004s : 45: predicate.remove_not_recompute_node 0.96% : 0.000020s : 192: predicate.replace_applicator 0.17% : 0.000004s : 45: predicate.replace_old_param 0.11% : 0.000002s : 22: predicate.reset_defer_inline 1.15% : 0.000024s : 133: predicate.reshape_eliminate 0.31% : 0.000006s : 44: predicate.row_tensor_add_zeros_like 0.18% : 0.000004s : 22: predicate.row_tensor_eliminate 0.45% : 0.000009s : 44: predicate.same_eliminate 0.23% : 0.000005s : 50: predicate.set_cell_output_no_recompute 0.43% : 0.000009s : 45: predicate.shard_identity_eliminate 0.34% : 0.000007s : 44: predicate.special_op_eliminate 0.58% : 0.000012s : 73: predicate.specialize_transform 0.40% : 0.000008s : 44: predicate.split_environ_get_set_with_tuple_value 0.39% : 0.000008s : 45: predicate.stack_unstack_eliminate 0.17% : 0.000004s : 22: predicate.switch_call_monad_eliminater 1.77% : 0.000037s : 219: predicate.switch_defer_inline 2.14% : 0.000045s : 263: predicate.switch_layer_defer_inline 8.84% : 0.000184s : 686: predicate.switch_simplify 1.08% : 0.000022s : 133: predicate.tile_eliminate 0.98% : 0.000020s : 133: predicate.transpose_eliminate 1.47% : 0.000031s : 177: predicate.tuple_list_convert_item_index_to_positive 1.61% : 0.000034s : 177: predicate.tuple_list_get_item_const_eliminator 1.51% : 0.000032s : 177: predicate.tuple_list_get_item_depend_reorder 2.93% : 0.000061s : 263: predicate.tuple_list_get_item_eliminator 1.65% : 0.000034s : 177: predicate.tuple_list_get_set_item_eliminator 2.28% : 0.000048s : 250: predicate.tuple_list_set_item_eliminator 1.44% : 0.000030s : 190: predicate.tuple_to_list_eliminator_ 2.24% : 0.000047s : 325: predicate.updatestate_pure_node_eliminater 3.02% : 0.000063s : 398: predicate.updatestate_useless_node_eliminater 0.19% : 0.000004s : 22: predicate.value_based_eliminate 0.36% : 0.000008s : 45: predicate.virtual_dataset_eliminate 0.38% : 0.000008s : 45: predicate.virtual_output_eliminate 0.17% : 0.000004s : 22: predicate.virtual_view_grad_eliminate 0.20% : 0.000004s : 22: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.222491 135 5.83% : 0.012967s : 55: func_graph_cloner_run.FuncGraphClonerGraph 94.17% : 0.209524s : 80: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 6.963021 209 0.00% : 0.000004s : 1: ForceFp32Comm 0.10% : 0.006690s : 1: add_attr 0.10% : 0.006664s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.00% : 0.000223s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.02% : 0.001090s : 1: auto_monad 0.00% : 0.000098s : 1: auto_monad_reorder 0.00% : 0.000005s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.01% : 0.000824s : 1: bootstrap 0.00% : 0.000054s : 1: cconv 0.00% : 0.000006s : 1: comm_op_add_attrs 0.00% : 0.000050s : 1: control_data_broadcast_order 0.00% : 0.000038s : 1: convert_after_rewriter 0.00% : 0.000093s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000046s : 1: environ_conv 0.01% : 0.000381s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000007s : 1: get_jit_bprop_graph 0.00% : 0.000070s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000008s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000009s : 1: label_micro_interleaved_index 0.01% : 0.000685s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.01% : 0.000906s : 1: mutable_eliminate 0.00% : 0.000017s : 1: offloading_packed_experts 0.00% : 0.000048s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000052s : 1: opt.transform.mutable_eliminate 3.06% : 0.213097s : 95: opt.transform.opt_a 0.00% : 0.000172s : 1: opt.transform.opt_after_cconv 0.00% : 0.000136s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000710s : 28: opt.transform.opt_b 0.00% : 0.000288s : 2: opt.transform.opt_trans_graph 0.00% : 0.000210s : 4: opt.transform.symbol_engine_opt 8.32% : 0.579106s : 1: opt_a 0.01% : 0.000381s : 1: opt_after_cconv 3.54% : 0.246255s : 1: opt_after_jit_grad 0.01% : 0.001011s : 1: opt_b 8.43% : 0.586907s : 1: optimize 0.00% : 0.000050s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000020s : 1: order_py_execute_after_rewriter 0.00% : 0.000089s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000014s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000006s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000015s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000011s : 1: parallel-infer-symbol 0.00% : 0.000007s : 1: parallel-infer-symbol-second 0.00% : 0.000008s : 1: partial_unused_args_eliminate 0.00% : 0.000007s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000393s : 1: pre_auto_parallel 0.01% : 0.000401s : 1: py_interpret_to_execute 0.00% : 0.000049s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000194s : 1: remove_dup_value 0.15% : 0.010573s : 1: renormalize.infer 3.03% : 0.210643s : 1: renormalize.specialize 0.00% : 0.000008s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.02% : 0.001115s : 1: rewriter_after_opt_a 0.02% : 0.001129s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000007s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000025s : 1: swap_dp_allreduce_reducescatter 0.01% : 0.000601s : 1: symbol_engine_optimizer 0.00% : 0.000334s : 1: tuple_transform 73.11% : 5.090850s : 1: type_inference . [hook] pytest_runtest_teardown:test_qbmm_qkv_1408_128_128[True-128-32] tests/st/infer/ops/test_internal_ops/test_qbmm_split.py::test_qbmm_qkv_1408_128_128[True-128-32],max_mem:118.0M =============================== warnings summary =============================== ../../../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/classifier/transdata/transdata_classifier.py:222 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/classifier/transdata/transdata_classifier.py:222: DeprecationWarning: invalid escape sequence \B """ ../../../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:143 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:143: DeprecationWarning: invalid escape sequence \c """ ../../../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:170 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:170: DeprecationWarning: invalid escape sequence \c """ ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for type is zero. setattr(self, word, getattr(machar, word).flat[0]) ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for type is zero. return self._float_to_str(self.smallest_subnormal) ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for type is zero. setattr(self, word, getattr(machar, word).flat[0]) ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for type is zero. return self._float_to_str(self.smallest_subnormal) ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2.py:57 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2.py:57: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad.py:56 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad.py:56: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2_grad") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad_reduce.py:48 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad_reduce.py:48: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2_grad_reduce") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul.py:51 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul.py:51: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:51 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:51: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul_grad") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:143 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:143: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul_grad_reduce") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad.py:92 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad.py:92: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer_grad_d") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad_reduce.py:49 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad_reduce.py:49: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer_grad_d_reduce") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad.py:91 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad.py:91: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel_grad_d") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad_reduce.py:48 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad_reduce.py:48: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel_grad_d_reduce") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel.py:52 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel.py:52: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_perchannel") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel_grad.py:81 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel_grad.py:81: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_perchannel_grad") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer.py:54 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer.py:54: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_per_layer") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer_grad.py:81 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer_grad.py:81: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_per_layer_grad") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perchannel.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perchannel.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("minmax_update_perchannel") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perlayer.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perlayer.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("minmax_update_perlayer") test_qbmm_split.py::test_qbmm_qkv_1408_128_128[False-128-32] test_qbmm_split.py::test_qbmm_qkv_1408_128_128[True-128-32] /home/jenkins/mindspore/testcases/testcases/tests/st/infer/ops/test_internal_ops/st_utils.py:39: RuntimeWarning: invalid value encountered in divide err_cnt = np.sum(np.abs(out_flatten - expect_flatten) / -- Docs: https://docs.pytest.org/en/stable/warnings.html ================== 6 passed, 27 warnings in 165.68s (0:02:45) ==================