==================================================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_001/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 12 items test_batch_matmul.py [WARNING] ME(152945:281473835274032,MainProcess):2026-01-29-17:37:19.181.017 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.0530953, [21] [bootstrap]: 0.00063854 [type_inference]: 0.0164842 [event_method]: 1.138e-05 [auto_monad]: 8.189e-05 [graph_reusing]: 5.25001e-06 [inline]: 2.39001e-06 [add_attr]: 0.0110447, [1] [add_attr_with_inline]: 0.0110303, [1] [Cycle 1]: 0.00010018, [2] [tag_attr]: 1.651e-05 [meta_addattr_fg_expand]: 4.10998e-06 [parallel-infer-symbol]: 4e-06 [pre_auto_parallel]: 5.167e-05 [insert-virtual-dataset]: 2.82002e-06 [parallel-infer-symbol-second]: 1.00001e-06 [dataset_repeat_opt]: 1.97999e-06 [pipeline_split]: 2.01998e-06 [optimize]: 0.0239037, [53] [py_interpret_to_execute]: 2.258e-05 [rewriter_before_opt_a]: 6.06e-05 [opt_a]: 0.0212662, [2] [Cycle 1]: 0.00178729, [45] [expand_dump_flag]: 3.39001e-06 [switch_simplify]: 2.331e-05 [loop_unroll]: 1.194e-05 [a_1]: 0.00029038 [with_stream_mark]: 2.169e-05 [recompute_prepare]: 9.53002e-06 [updatestate_depend_eliminate]: 4.44998e-06 [updatestate_assign_eliminate]: 3.33998e-06 [updatestate_loads_eliminate]: 3.45e-06 [parameter_eliminate]: 2.09999e-06 [a_2]: 9.096e-05 [accelerated_algorithm]: 7.41999e-06 [shard]: 2.32001e-06 [meta_shard_fg_expand]: 1.87001e-06 [shard_inline]: 6.60002e-06 [merge_send_recv]: 2.659e-05 [auto_parallel]: 8.16002e-06 [parallel]: 4.148e-05 [flash_sp]: 1.795e-05 [merge_comm]: 4.22e-06 [allreduce_fusion]: 4.13999e-06 [matmul_add_comm_reduction]: 1.067e-05 [allreduce_slice_to_reducescatter]: 9.30013e-07 [virtual_shard_identity]: 9.17999e-06 [virtual_dataset]: 7.2e-06 [get_grad_eliminate_]: 6.27001e-06 [virtual_output]: 6.39999e-06 [merge_forward]: 4.27e-06 [cell_reuse_recompute_pass]: 1.76e-06 [offload_activation]: 1.034e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.713e-05 [merge_recompute_call_nodes]: 1.54e-06 [before_grad]: 1.08e-05 [set_forward_comm_id_for_comm_node_pass]: 4.26001e-06 [meta_fg_expand]: 2.51e-06 [flash_sp_send_recv_attached]: 3.22002e-06 [receive_attached]: 1.375e-05 [after_resolve]: 1.382e-05 [a_after_grad]: 9.94999e-06 [renormalize]: 0.00066762 [add_forward_monad_depend]: 1.153e-05 [auto_monad_grad]: 2.63e-06 [auto_monad_eliminator]: 1.747e-05 [cse]: 4.895e-05 [a_3]: 5.427e-05 [Cycle 2]: 0.0194656, [45] [expand_dump_flag]: 2.31e-06 [switch_simplify]: 8.69e-06 [loop_unroll]: 6.46e-06 [a_1]: 0.00013208 [with_stream_mark]: 1.972e-05 [recompute_prepare]: 7.93001e-06 [updatestate_depend_eliminate]: 3.88001e-06 [updatestate_assign_eliminate]: 3.33998e-06 [updatestate_loads_eliminate]: 2.48e-06 [parameter_eliminate]: 1.71e-06 [a_2]: 7.889e-05 [accelerated_algorithm]: 7.75e-06 [shard]: 1.70001e-06 [meta_shard_fg_expand]: 1.89999e-06 [shard_inline]: 7.1e-06 [merge_send_recv]: 6.98998e-06 [auto_parallel]: 8.42e-06 [parallel]: 7.22002e-06 [flash_sp]: 3.91999e-06 [merge_comm]: 4.04002e-06 [allreduce_fusion]: 3.69002e-06 [matmul_add_comm_reduction]: 9.08002e-06 [allreduce_slice_to_reducescatter]: 3.7998e-07 [virtual_shard_identity]: 7.5e-06 [virtual_dataset]: 6.56999e-06 [get_grad_eliminate_]: 6.24001e-06 [virtual_output]: 6.18002e-06 [merge_forward]: 4.67998e-06 [cell_reuse_recompute_pass]: 2.28998e-06 [offload_activation]: 1.005e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.726e-05 [merge_recompute_call_nodes]: 1.44e-06 [before_grad]: 1.058e-05 [set_forward_comm_id_for_comm_node_pass]: 4.65001e-06 [meta_fg_expand]: 2.92002e-06 [flash_sp_send_recv_attached]: 1.44998e-06 [receive_attached]: 2.33998e-06 [after_resolve]: 1.261e-05 [a_after_grad]: 9.89999e-06 [renormalize]: 6.00121e-08 [add_forward_monad_depend]: 2.96999e-06 [auto_monad_grad]: 2.01998e-06 [auto_monad_eliminator]: 9.57999e-06 [cse]: 2.282e-05 [a_3]: 0.0187478 [py_interpret_to_execute_after_opt_a]: 2.894e-05 [slice_cell_reuse_recomputed_activation]: 2.98e-06 [rewriter_after_opt_a]: 9.668e-05 [convert_after_rewriter]: 2.31e-05 [order_py_execute_after_rewriter]: 5.63002e-06 [mutable_eliminate]: 0.00073639 [opt_b]: 0.00024816, [1] [Cycle 1]: 0.00023861, [7] [b_1]: 0.00013435 [b_2]: 8.62998e-06 [updatestate_depend_eliminate]: 8.84998e-06 [updatestate_assign_eliminate]: 3.82002e-06 [updatestate_loads_eliminate]: 3.08e-06 [renormalize]: 9.09989e-07 [cse]: 4.372e-05 [optimize_parallel_all_gather_comm]: 2.209e-05 [overlap_param_gather]: 1.92999e-06 [cconv]: 3.635e-05 [loop_unroll]: 0.00048793 [opt_after_cconv]: 0.0001169, [1] [Cycle 1]: 0.00011021, [7] [c_1]: 3.072e-05 [parameter_eliminate]: 4.69998e-06 [updatestate_depend_eliminate]: 6.91999e-06 [updatestate_assign_eliminate]: 3.01999e-06 [updatestate_loads_eliminate]: 2.86e-06 [cse]: 2.623e-05 [renormalize]: 6.50005e-07 [remove_dup_value]: 1.771e-05 [tuple_transform]: 8.926e-05, [1] [Cycle 1]: 8.386e-05, [4] [d_1]: 5.44e-05 [none_parameter_eliminate]: 2.20002e-06 [renormalize]: 4.09986e-07 [switch_simplify]: 7.63001e-06 [partial_unused_args_eliminate]: 1.82999e-06 [add_recomputation]: 6.438e-05 [cse_after_recomputation]: 2.688e-05, [1] [Cycle 1]: 2.196e-05, [1] [cse]: 1.576e-05 [environ_conv]: 1.828e-05 [swap_dp_allreduce_reducescatter]: 5.71e-06 [bias_add_comm_swap]: 3.56001e-06 [label_micro_interleaved_index]: 4.65001e-06 [label_fine_grained_interleaved_index]: 2.93e-06 [merge_cast_opt]: 1.32e-06 [slice_recompute_activation]: 2.49999e-06 [micro_interleaved_order_control]: 3.2e-06 [assign_add_opt]: 1.25999e-06 [ForceFp32Comm]: 1.32e-06 [remove_cast_before_assign_add]: 1.24998e-06 [full_micro_interleaved_order_control]: 2.05002e-06 [reorder_send_recv_between_fp_bp]: 2.44999e-06 [comm_op_add_attrs]: 1.25999e-06 [add_comm_op_reuse_tag]: 1.16002e-06 [interleave_split_concat_branches]: 1.67001e-06 [interleave_parallel_branches]: 1.05999e-06 [overlap_opt_shard_in_pipeline]: 2.252e-05 [overlap_opt_shard_grad_in_pipeline]: 2.37999e-06 [control_data_broadcast_order]: 1.435e-05 [grouped_pairwise_exchange_alltoall]: 2.12001e-06 [offloading_packed_experts]: 4.15e-06 [overlap_recompute_and_grad_model_parallel]: 4.97999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.18001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.24e-06 [overlap_recompute_comm]: 2.43e-06 [overlap_grad_ring_attention]: 4.25e-06 [overlap_grad_flash_sp]: 3.629e-05 [begin_end_overlap_inline]: 5.69999e-07 [split_matmul_comm_elemetwise]: 2.60002e-06 [split_layernorm_comm]: 1.60001e-06 [handle_group_info]: 1.09003e-06 [symbol_engine_optimizer]: 8.755e-05, [1] [Cycle 1]: 8.222e-05, [6] [build]: 3.72002e-06 [elim_shapecalc]: 1.46e-05 [elim_not_effective]: 1.367e-05 [opt_reshape]: 7.52002e-06 [fold_const_symbol]: 1.06e-05 [renormalize]: 3.10014e-07 [detach_backward]: 2.59001e-06 [pipeline_parallel_scheduler]: 1.57001e-06 [auto_monad_reorder]: 2.582e-05 [get_jit_bprop_graph]: 2.07001e-06 [rewriter_after_jit_bprop_graph]: 5.40999e-06 [opt_after_jit_grad]: 0.00057409 [validate]: 6.617e-05 Sums bootstrap : 0.000639s : 1.56% type_inference : 0.016484s : 40.30% event_method : 0.000011s : 0.03% auto_monad : 0.000082s : 0.20% graph_reusing : 0.000005s : 0.01% inline : 0.000002s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000017s : 0.04% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.01% parallel-infer-symbol : 0.000004s : 0.01% pre_auto_parallel : 0.000052s : 0.13% insert-virtual-dataset : 0.000003s : 0.01% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000023s : 0.06% optimize.rewriter_before_opt_a : 0.000061s : 0.15% optimize.opt_a.expand_dump_flag : 0.000006s : 0.01% optimize.opt_a.switch_simplify : 0.000032s : 0.08% optimize.opt_a.loop_unroll : 0.000018s : 0.04% optimize.opt_a.a_1 : 0.000422s : 1.03% optimize.opt_a.with_stream_mark : 0.000041s : 0.10% optimize.opt_a.recompute_prepare : 0.000017s : 0.04% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.02% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.02% optimize.opt_a.updatestate_loads_eliminate : 0.000006s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.01% optimize.opt_a.a_2 : 0.000170s : 0.42% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.04% optimize.opt_a.shard : 0.000004s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.01% optimize.opt_a.shard_inline : 0.000014s : 0.03% optimize.opt_a.merge_send_recv : 0.000034s : 0.08% optimize.opt_a.auto_parallel : 0.000017s : 0.04% optimize.opt_a.parallel : 0.000049s : 0.12% optimize.opt_a.flash_sp : 0.000022s : 0.05% optimize.opt_a.merge_comm : 0.000008s : 0.02% optimize.opt_a.allreduce_fusion : 0.000008s : 0.02% optimize.opt_a.matmul_add_comm_reduction : 0.000020s : 0.05% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000017s : 0.04% optimize.opt_a.virtual_dataset : 0.000014s : 0.03% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.03% optimize.opt_a.virtual_output : 0.000013s : 0.03% optimize.opt_a.merge_forward : 0.000009s : 0.02% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.01% optimize.opt_a.offload_activation : 0.000020s : 0.05% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000034s : 0.08% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.01% optimize.opt_a.before_grad : 0.000021s : 0.05% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000009s : 0.02% optimize.opt_a.meta_fg_expand : 0.000005s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.01% optimize.opt_a.receive_attached : 0.000016s : 0.04% optimize.opt_a.after_resolve : 0.000026s : 0.06% optimize.opt_a.a_after_grad : 0.000020s : 0.05% optimize.opt_a.renormalize : 0.000668s : 1.63% optimize.opt_a.add_forward_monad_depend : 0.000014s : 0.04% optimize.opt_a.auto_monad_grad : 0.000005s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000027s : 0.07% optimize.opt_a.cse : 0.000072s : 0.18% optimize.opt_a.a_3 : 0.018802s : 45.97% optimize.py_interpret_to_execute_after_opt_a : 0.000029s : 0.07% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.01% optimize.rewriter_after_opt_a : 0.000097s : 0.24% optimize.convert_after_rewriter : 0.000023s : 0.06% optimize.order_py_execute_after_rewriter : 0.000006s : 0.01% optimize.mutable_eliminate : 0.000736s : 1.80% optimize.opt_b.b_1 : 0.000134s : 0.33% optimize.opt_b.b_2 : 0.000009s : 0.02% optimize.opt_b.updatestate_depend_eliminate : 0.000009s : 0.02% optimize.opt_b.updatestate_assign_eliminate : 0.000004s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.01% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000044s : 0.11% optimize.optimize_parallel_all_gather_comm : 0.000022s : 0.05% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000036s : 0.09% optimize.loop_unroll : 0.000488s : 1.19% optimize.opt_after_cconv.c_1 : 0.000031s : 0.08% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.01% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000007s : 0.02% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000003s : 0.01% optimize.opt_after_cconv.cse : 0.000026s : 0.06% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000018s : 0.04% optimize.tuple_transform.d_1 : 0.000054s : 0.13% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.01% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000008s : 0.02% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000064s : 0.16% optimize.cse_after_recomputation.cse : 0.000016s : 0.04% optimize.environ_conv : 0.000018s : 0.04% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.01% optimize.bias_add_comm_swap : 0.000004s : 0.01% optimize.label_micro_interleaved_index : 0.000005s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.01% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.01% optimize.micro_interleaved_order_control : 0.000003s : 0.01% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.01% optimize.reorder_send_recv_between_fp_bp : 0.000002s : 0.01% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000002s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000023s : 0.06% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.01% optimize.control_data_broadcast_order : 0.000014s : 0.04% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.01% optimize.offloading_packed_experts : 0.000004s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.01% 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 : 0.01% optimize.overlap_grad_ring_attention : 0.000004s : 0.01% optimize.overlap_grad_flash_sp : 0.000036s : 0.09% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.01% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000004s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000015s : 0.04% optimize.symbol_engine_optimizer.elim_not_effective : 0.000014s : 0.03% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.02% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000011s : 0.03% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.01% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000026s : 0.06% get_jit_bprop_graph : 0.000002s : 0.01% rewriter_after_jit_bprop_graph : 0.000005s : 0.01% opt_after_jit_grad : 0.000574s : 1.40% validate : 0.000066s : 0.16% Time group info: ------[substitution.] 0.000111 21 1.66% : 0.000002s : 2: substitution.elim_not_effective 1.28% : 0.000001s : 2: substitution.fold_const_symbol 6.20% : 0.000007s : 4: substitution.graph_param_transform 73.28% : 0.000081s : 1: substitution.inline 4.29% : 0.000005s : 4: substitution.j_node_and_user_rematch 7.54% : 0.000008s : 4: substitution.remove_not_recompute_node 5.76% : 0.000006s : 4: substitution.replace_old_param ------[type_inference.] 0.016423 2 97.75% : 0.016054s : 1: type_inference.infer 2.25% : 0.000369s : 1: type_inference.specialize ------[replace.] 0.000016 1 100.00% : 0.000016s : 1: replace.inline ------[match.] 0.000080 1 100.00% : 0.000080s : 1: match.inline ------[predicate.] 0.000173 967 0.80% : 0.000001s : 9: predicate.accumulaten_eliminater 1.47% : 0.000003s : 4: predicate.ad_related_special_op_eliminate 0.62% : 0.000001s : 8: predicate.addn_check_dump 0.71% : 0.000001s : 9: predicate.addn_zero_filter 0.62% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.03% : 0.000004s : 17: predicate.arithmetic_simplify 0.78% : 0.000001s : 9: predicate.cast_eliminate 0.69% : 0.000001s : 8: predicate.check_bprop_eliminate 0.58% : 0.000001s : 8: predicate.compare_switch_simplify 0.22% : 0.000000s : 4: predicate.const_output_eliminate 0.63% : 0.000001s : 8: predicate.depend_value_elim 0.83% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.77% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.72% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.90% : 0.000003s : 8: predicate.dumpgradient_eliminate 0.27% : 0.000000s : 4: predicate.elim_not_effective 0.70% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.02% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.09% : 0.000002s : 13: predicate.environ_get_add_eliminate 1.12% : 0.000002s : 13: predicate.environ_get_depend_swap 1.72% : 0.000003s : 21: predicate.environ_get_eliminate 1.00% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.71% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.51% : 0.000003s : 10: predicate.float_depend_g_call 0.57% : 0.000001s : 8: predicate.float_environ_get_switch 0.91% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.23% : 0.000000s : 4: predicate.fold_const_symbol 0.72% : 0.000001s : 8: predicate.get_grad_eliminate 0.37% : 0.000001s : 4: predicate.graph_param_transform 0.70% : 0.000001s : 8: predicate.incorporate_call 0.60% : 0.000001s : 8: predicate.incorporate_call_switch 5.71% : 0.000010s : 43: predicate.inline 1.16% : 0.000002s : 8: predicate.inline_without_move 0.36% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.01% : 0.000002s : 8: predicate.less_batch_normalization 1.56% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 1.95% : 0.000003s : 26: predicate.load_eliminater 1.41% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.10% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.90% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.71% : 0.000001s : 8: predicate.merge_addn 0.75% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.86% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.62% : 0.000001s : 9: predicate.minmaximum_grad 1.72% : 0.000003s : 4: predicate.mutable_eliminate 0.47% : 0.000001s : 4: predicate.opt_reshape 0.49% : 0.000001s : 4: predicate.parallel_virtual_node 1.27% : 0.000002s : 10: predicate.partial_defer_inline 1.10% : 0.000002s : 13: predicate.partial_eliminate 0.64% : 0.000001s : 9: predicate.print_const_string_wrapper 0.68% : 0.000001s : 8: predicate.reduce_all_const_elim 1.06% : 0.000002s : 9: predicate.reduce_eliminate 2.06% : 0.000004s : 26: predicate.redundant_stop_gradient_eliminater 0.72% : 0.000001s : 8: predicate.remove_not_recompute_node 1.80% : 0.000003s : 17: predicate.replace_applicator 0.72% : 0.000001s : 8: predicate.replace_old_param 0.43% : 0.000001s : 4: predicate.reset_defer_inline 0.86% : 0.000001s : 9: predicate.reshape_eliminate 0.78% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.61% : 0.000001s : 4: predicate.row_tensor_eliminate 0.98% : 0.000002s : 8: predicate.same_eliminate 0.53% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.94% : 0.000002s : 8: predicate.shard_identity_eliminate 0.82% : 0.000001s : 8: predicate.special_op_eliminate 0.95% : 0.000002s : 8: predicate.specialize_transform 2.08% : 0.000004s : 8: predicate.split_environ_get_set_with_tuple_value 0.93% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.49% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.81% : 0.000001s : 10: predicate.switch_defer_inline 5.49% : 0.000009s : 18: predicate.switch_layer_defer_inline 3.41% : 0.000006s : 35: predicate.switch_simplify 0.73% : 0.000001s : 9: predicate.tile_eliminate 0.70% : 0.000001s : 9: predicate.transpose_eliminate 1.39% : 0.000002s : 17: predicate.tuple_list_convert_item_index_to_positive 1.43% : 0.000002s : 17: predicate.tuple_list_get_item_const_eliminator 1.31% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.04% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.51% : 0.000003s : 17: predicate.tuple_list_get_set_item_eliminator 2.52% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.47% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.86% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.65% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.54% : 0.000001s : 4: predicate.value_based_eliminate 0.72% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.68% : 0.000001s : 8: predicate.virtual_output_eliminate 0.33% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.57% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000206 4 10.58% : 0.000022s : 1: func_graph_cloner_run.FuncGraphClonerGraph 89.42% : 0.000184s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.089729 192 0.00% : 0.000004s : 1: ForceFp32Comm 12.32% : 0.011052s : 1: add_attr 12.30% : 0.011035s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.08% : 0.000070s : 1: add_recomputation 0.01% : 0.000005s : 1: assign_add_opt 0.10% : 0.000087s : 1: auto_monad 0.03% : 0.000031s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.01% : 0.000006s : 1: bias_add_comm_swap 0.76% : 0.000684s : 1: bootstrap 0.05% : 0.000041s : 1: cconv 0.01% : 0.000007s : 1: comm_op_add_attrs 0.02% : 0.000018s : 1: control_data_broadcast_order 0.03% : 0.000027s : 1: convert_after_rewriter 0.03% : 0.000030s : 1: cse_after_recomputation 0.01% : 0.000005s : 1: dataset_repeat_opt 0.01% : 0.000006s : 1: detach_backward 0.02% : 0.000022s : 1: environ_conv 0.02% : 0.000017s : 1: event_method 0.01% : 0.000005s : 1: full_micro_interleaved_order_control 0.01% : 0.000005s : 1: get_jit_bprop_graph 0.01% : 0.000009s : 1: graph_reusing 0.01% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.01% : 0.000005s : 1: inline 0.01% : 0.000006s : 1: insert-virtual-dataset 0.01% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.01% : 0.000006s : 1: label_fine_grained_interleaved_index 0.01% : 0.000008s : 1: label_micro_interleaved_index 0.55% : 0.000497s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.01% : 0.000006s : 1: micro_interleaved_order_control 0.83% : 0.000747s : 1: mutable_eliminate 0.01% : 0.000007s : 1: offloading_packed_experts 0.02% : 0.000017s : 1: opt.transform.loop_unroll_optimizer 0.02% : 0.000021s : 1: opt.transform.mutable_eliminate 0.98% : 0.000877s : 78: opt.transform.opt_a 0.03% : 0.000029s : 1: opt.transform.opt_after_cconv 0.04% : 0.000032s : 1: opt.transform.opt_after_jit_grad 0.13% : 0.000113s : 28: opt.transform.opt_b 0.07% : 0.000060s : 2: opt.transform.opt_trans_graph 0.05% : 0.000042s : 4: opt.transform.symbol_engine_opt 23.71% : 0.021271s : 1: opt_a 0.13% : 0.000121s : 1: opt_after_cconv 0.65% : 0.000587s : 1: opt_after_jit_grad 0.28% : 0.000252s : 1: opt_b 26.65% : 0.023909s : 1: optimize 0.03% : 0.000026s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000009s : 1: order_py_execute_after_rewriter 0.04% : 0.000040s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000007s : 1: overlap_grad_ring_attention 0.01% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.03% : 0.000026s : 1: overlap_opt_shard_in_pipeline 0.07% : 0.000064s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.01% : 0.000005s : 1: overlap_recompute_comm 0.01% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.01% : 0.000005s : 1: partial_unused_args_eliminate 0.01% : 0.000005s : 1: pipeline_parallel_scheduler 0.01% : 0.000005s : 1: pipeline_split 0.06% : 0.000057s : 1: pre_auto_parallel 0.03% : 0.000027s : 1: py_interpret_to_execute 0.04% : 0.000035s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.02% : 0.000021s : 1: remove_dup_value 0.44% : 0.000391s : 1: renormalize.infer 0.30% : 0.000267s : 1: renormalize.specialize 0.01% : 0.000005s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.11% : 0.000101s : 1: rewriter_after_opt_a 0.07% : 0.000065s : 1: rewriter_before_opt_a 0.01% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.01% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000004s : 1: split_layernorm_comm 0.01% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.10% : 0.000090s : 1: symbol_engine_optimizer 0.10% : 0.000092s : 1: tuple_transform 18.39% : 0.016503s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_bf16[dim_type0-True-batch_size_list0-shape0-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_bf16[dim_type0-True-batch_size_list0-shape0-ms_dtype0],max_mem:68.0M [WARNING] ME(152945:281473835274032,MainProcess):2026-01-29-17:38:05.745.262 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.0996392, [21] [bootstrap]: 0.00058723 [type_inference]: 0.00350839 [event_method]: 1.116e-05 [auto_monad]: 5.482e-05 [graph_reusing]: 5.02999e-06 [inline]: 2.34001e-06 [add_attr]: 0.00443504, [1] [add_attr_with_inline]: 0.0044186, [1] [Cycle 1]: 6.634e-05, [2] [tag_attr]: 1.498e-05 [meta_addattr_fg_expand]: 3.23998e-06 [parallel-infer-symbol]: 3.76001e-06 [pre_auto_parallel]: 3.316e-05 [insert-virtual-dataset]: 2.58e-06 [parallel-infer-symbol-second]: 9.00007e-07 [dataset_repeat_opt]: 4.33001e-06 [pipeline_split]: 1.66998e-06 [optimize]: 0.0900021, [53] [py_interpret_to_execute]: 2.075e-05 [rewriter_before_opt_a]: 5.131e-05 [opt_a]: 0.0875072, [2] [Cycle 1]: 0.0866686, [45] [expand_dump_flag]: 2.73998e-06 [switch_simplify]: 2.425e-05 [loop_unroll]: 1.194e-05 [a_1]: 0.00031828 [with_stream_mark]: 2.245e-05 [recompute_prepare]: 8.28001e-06 [updatestate_depend_eliminate]: 5.02999e-06 [updatestate_assign_eliminate]: 3.59002e-06 [updatestate_loads_eliminate]: 3.51999e-06 [parameter_eliminate]: 2.07999e-06 [a_2]: 8.913e-05 [accelerated_algorithm]: 7.48e-06 [shard]: 2.34001e-06 [meta_shard_fg_expand]: 1.81e-06 [shard_inline]: 6.79001e-06 [merge_send_recv]: 8.64003e-06 [auto_parallel]: 6.72002e-06 [parallel]: 3.096e-05 [flash_sp]: 9.38002e-06 [merge_comm]: 4.08001e-06 [allreduce_fusion]: 3.85998e-06 [matmul_add_comm_reduction]: 1.025e-05 [allreduce_slice_to_reducescatter]: 6.39993e-07 [virtual_shard_identity]: 1.193e-05 [virtual_dataset]: 7.03e-06 [get_grad_eliminate_]: 6.54999e-06 [virtual_output]: 6.69999e-06 [merge_forward]: 4.89003e-06 [cell_reuse_recompute_pass]: 1.24998e-06 [offload_activation]: 1.025e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.466e-05 [merge_recompute_call_nodes]: 2.16e-06 [before_grad]: 1.207e-05 [set_forward_comm_id_for_comm_node_pass]: 4.03001e-06 [meta_fg_expand]: 2.94999e-06 [flash_sp_send_recv_attached]: 2.93e-06 [receive_attached]: 2.13002e-06 [after_resolve]: 1.216e-05 [a_after_grad]: 1.033e-05 [renormalize]: 0.0855216 [add_forward_monad_depend]: 1.847e-05 [auto_monad_grad]: 3.11999e-06 [auto_monad_eliminator]: 2.584e-05 [cse]: 4.384e-05 [a_3]: 7.487e-05 [Cycle 2]: 0.00082296, [45] [expand_dump_flag]: 2.36e-06 [switch_simplify]: 1.073e-05 [loop_unroll]: 8.38999e-06 [a_1]: 0.00016188 [with_stream_mark]: 2.08e-05 [recompute_prepare]: 9.27001e-06 [updatestate_depend_eliminate]: 4.95001e-06 [updatestate_assign_eliminate]: 4.08999e-06 [updatestate_loads_eliminate]: 3.65e-06 [parameter_eliminate]: 2.54001e-06 [a_2]: 8.468e-05 [accelerated_algorithm]: 7.19001e-06 [shard]: 2.79001e-06 [meta_shard_fg_expand]: 2.84999e-06 [shard_inline]: 6.98e-06 [merge_send_recv]: 9.29e-06 [auto_parallel]: 1.174e-05 [parallel]: 1.126e-05 [flash_sp]: 4.67998e-06 [merge_comm]: 4.57998e-06 [allreduce_fusion]: 4.45999e-06 [matmul_add_comm_reduction]: 1.19e-05 [allreduce_slice_to_reducescatter]: 9.50007e-07 [virtual_shard_identity]: 8.05999e-06 [virtual_dataset]: 7.46999e-06 [get_grad_eliminate_]: 7.18e-06 [virtual_output]: 7.01999e-06 [merge_forward]: 5.27001e-06 [cell_reuse_recompute_pass]: 3.68999e-06 [offload_activation]: 1.179e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.852e-05 [merge_recompute_call_nodes]: 1.60999e-06 [before_grad]: 1.226e-05 [set_forward_comm_id_for_comm_node_pass]: 4.13999e-06 [meta_fg_expand]: 3.34001e-06 [flash_sp_send_recv_attached]: 2.47001e-06 [receive_attached]: 2.69001e-06 [after_resolve]: 1.306e-05 [a_after_grad]: 1.126e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 2.39999e-06 [auto_monad_grad]: 2.04e-06 [auto_monad_eliminator]: 9.52999e-06 [cse]: 2.392e-05 [a_3]: 4.34e-05 [py_interpret_to_execute_after_opt_a]: 2.101e-05 [slice_cell_reuse_recomputed_activation]: 2.21998e-06 [rewriter_after_opt_a]: 4.782e-05 [convert_after_rewriter]: 8.2e-06 [order_py_execute_after_rewriter]: 5.68002e-06 [mutable_eliminate]: 0.00077473 [opt_b]: 0.00024897, [1] [Cycle 1]: 0.00023912, [7] [b_1]: 0.00014052 [b_2]: 1.014e-05 [updatestate_depend_eliminate]: 8.73001e-06 [updatestate_assign_eliminate]: 2.93998e-06 [updatestate_loads_eliminate]: 2.66e-06 [renormalize]: 1.12e-06 [cse]: 3.379e-05 [optimize_parallel_all_gather_comm]: 2.024e-05 [overlap_param_gather]: 6.38e-06 [cconv]: 3.384e-05 [loop_unroll]: 0.00049549 [opt_after_cconv]: 0.00011911, [1] [Cycle 1]: 0.00011106, [7] [c_1]: 3.265e-05 [parameter_eliminate]: 5.61e-06 [updatestate_depend_eliminate]: 6.33e-06 [updatestate_assign_eliminate]: 2.65002e-06 [updatestate_loads_eliminate]: 3.05998e-06 [cse]: 2.493e-05 [renormalize]: 3.69997e-07 [remove_dup_value]: 1.868e-05 [tuple_transform]: 8.475e-05, [1] [Cycle 1]: 7.947e-05, [4] [d_1]: 4.999e-05 [none_parameter_eliminate]: 1.77999e-06 [renormalize]: 6.19999e-07 [switch_simplify]: 7.25003e-06 [partial_unused_args_eliminate]: 1.75001e-06 [add_recomputation]: 5.614e-05 [cse_after_recomputation]: 2.962e-05, [1] [Cycle 1]: 2.294e-05, [1] [cse]: 1.58e-05 [environ_conv]: 7.11001e-06 [swap_dp_allreduce_reducescatter]: 5.75001e-06 [bias_add_comm_swap]: 3.09001e-06 [label_micro_interleaved_index]: 5.10999e-06 [label_fine_grained_interleaved_index]: 2.59001e-06 [merge_cast_opt]: 1.38002e-06 [slice_recompute_activation]: 2.16e-06 [micro_interleaved_order_control]: 2.12999e-06 [assign_add_opt]: 1.42e-06 [ForceFp32Comm]: 9.89996e-07 [remove_cast_before_assign_add]: 1.24e-06 [full_micro_interleaved_order_control]: 2.07001e-06 [reorder_send_recv_between_fp_bp]: 2.68998e-06 [comm_op_add_attrs]: 1.02e-06 [add_comm_op_reuse_tag]: 1.41998e-06 [interleave_split_concat_branches]: 1.20999e-06 [interleave_parallel_branches]: 1.05001e-06 [overlap_opt_shard_in_pipeline]: 3.4e-06 [overlap_opt_shard_grad_in_pipeline]: 1.86e-06 [control_data_broadcast_order]: 1.584e-05 [grouped_pairwise_exchange_alltoall]: 1.59e-06 [offloading_packed_experts]: 4.62e-06 [overlap_recompute_and_grad_model_parallel]: 5.00001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.30999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.30999e-06 [overlap_recompute_comm]: 2.15002e-06 [overlap_grad_ring_attention]: 4.28001e-06 [overlap_grad_flash_sp]: 2.251e-05 [begin_end_overlap_inline]: 5.29981e-07 [split_matmul_comm_elemetwise]: 2.44999e-06 [split_layernorm_comm]: 1.89e-06 [handle_group_info]: 1.25001e-06 [symbol_engine_optimizer]: 8.769e-05, [1] [Cycle 1]: 8.314e-05, [6] [build]: 3.8e-06 [elim_shapecalc]: 1.476e-05 [elim_not_effective]: 1.335e-05 [opt_reshape]: 7.84002e-06 [fold_const_symbol]: 1.12e-05 [renormalize]: 2.19996e-07 [detach_backward]: 2.39999e-06 [pipeline_parallel_scheduler]: 1.44e-06 [auto_monad_reorder]: 2.04e-05 [get_jit_bprop_graph]: 1.84998e-06 [rewriter_after_jit_bprop_graph]: 6.46e-06 [opt_after_jit_grad]: 0.00066955 [validate]: 5.778e-05 Sums bootstrap : 0.000587s : 0.62% type_inference : 0.003508s : 3.73% event_method : 0.000011s : 0.01% auto_monad : 0.000055s : 0.06% graph_reusing : 0.000005s : 0.01% inline : 0.000002s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000015s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000033s : 0.04% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000004s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000021s : 0.02% optimize.rewriter_before_opt_a : 0.000051s : 0.05% optimize.opt_a.expand_dump_flag : 0.000005s : 0.01% optimize.opt_a.switch_simplify : 0.000035s : 0.04% optimize.opt_a.loop_unroll : 0.000020s : 0.02% optimize.opt_a.a_1 : 0.000480s : 0.51% optimize.opt_a.with_stream_mark : 0.000043s : 0.05% optimize.opt_a.recompute_prepare : 0.000018s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000010s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000008s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.01% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000174s : 0.18% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.02% optimize.opt_a.shard : 0.000005s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000005s : 0.00% optimize.opt_a.shard_inline : 0.000014s : 0.01% optimize.opt_a.merge_send_recv : 0.000018s : 0.02% optimize.opt_a.auto_parallel : 0.000018s : 0.02% optimize.opt_a.parallel : 0.000042s : 0.04% optimize.opt_a.flash_sp : 0.000014s : 0.01% optimize.opt_a.merge_comm : 0.000009s : 0.01% optimize.opt_a.allreduce_fusion : 0.000008s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000022s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000020s : 0.02% optimize.opt_a.virtual_dataset : 0.000014s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000014s : 0.01% optimize.opt_a.virtual_output : 0.000014s : 0.01% optimize.opt_a.merge_forward : 0.000010s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.01% optimize.opt_a.offload_activation : 0.000022s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000033s : 0.04% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000024s : 0.03% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.01% optimize.opt_a.meta_fg_expand : 0.000006s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.01% optimize.opt_a.receive_attached : 0.000005s : 0.01% optimize.opt_a.after_resolve : 0.000025s : 0.03% optimize.opt_a.a_after_grad : 0.000022s : 0.02% optimize.opt_a.renormalize : 0.085522s : 90.93% optimize.opt_a.add_forward_monad_depend : 0.000021s : 0.02% optimize.opt_a.auto_monad_grad : 0.000005s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000035s : 0.04% optimize.opt_a.cse : 0.000068s : 0.07% optimize.opt_a.a_3 : 0.000118s : 0.13% optimize.py_interpret_to_execute_after_opt_a : 0.000021s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000048s : 0.05% optimize.convert_after_rewriter : 0.000008s : 0.01% optimize.order_py_execute_after_rewriter : 0.000006s : 0.01% optimize.mutable_eliminate : 0.000775s : 0.82% optimize.opt_b.b_1 : 0.000141s : 0.15% optimize.opt_b.b_2 : 0.000010s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000009s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000034s : 0.04% optimize.optimize_parallel_all_gather_comm : 0.000020s : 0.02% optimize.overlap_param_gather : 0.000006s : 0.01% optimize.cconv : 0.000034s : 0.04% optimize.loop_unroll : 0.000495s : 0.53% optimize.opt_after_cconv.c_1 : 0.000033s : 0.03% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.01% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000006s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000003s : 0.00% optimize.opt_after_cconv.cse : 0.000025s : 0.03% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000019s : 0.02% optimize.tuple_transform.d_1 : 0.000050s : 0.05% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000001s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000056s : 0.06% optimize.cse_after_recomputation.cse : 0.000016s : 0.02% optimize.environ_conv : 0.000007s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.01% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000005s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000003s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000016s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000005s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.01% 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 : 0.000004s : 0.00% optimize.overlap_grad_flash_sp : 0.000023s : 0.02% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000004s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000015s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000013s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000011s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000001s : 0.00% auto_monad_reorder : 0.000020s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.01% opt_after_jit_grad : 0.000670s : 0.71% validate : 0.000058s : 0.06% Time group info: ------[substitution.] 0.000113 21 2.20% : 0.000002s : 2: substitution.elim_not_effective 1.23% : 0.000001s : 2: substitution.fold_const_symbol 5.72% : 0.000006s : 4: substitution.graph_param_transform 74.68% : 0.000084s : 1: substitution.inline 4.50% : 0.000005s : 4: substitution.j_node_and_user_rematch 6.07% : 0.000007s : 4: substitution.remove_not_recompute_node 5.59% : 0.000006s : 4: substitution.replace_old_param ------[type_inference.] 0.003453 2 89.38% : 0.003086s : 1: type_inference.infer 10.62% : 0.000367s : 1: type_inference.specialize ------[replace.] 0.000017 1 100.00% : 0.000017s : 1: replace.inline ------[match.] 0.000083 1 100.00% : 0.000083s : 1: match.inline ------[predicate.] 0.000176 967 0.79% : 0.000001s : 9: predicate.accumulaten_eliminater 1.40% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.65% : 0.000001s : 8: predicate.addn_check_dump 0.77% : 0.000001s : 9: predicate.addn_zero_filter 0.92% : 0.000002s : 9: predicate.adjust_all_reduce_mul_add 2.30% : 0.000004s : 17: predicate.arithmetic_simplify 1.02% : 0.000002s : 9: predicate.cast_eliminate 0.84% : 0.000001s : 8: predicate.check_bprop_eliminate 0.64% : 0.000001s : 8: predicate.compare_switch_simplify 0.23% : 0.000000s : 4: predicate.const_output_eliminate 0.89% : 0.000002s : 8: predicate.depend_value_elim 0.73% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.96% : 0.000002s : 9: predicate.dict_get_item_eliminator 0.93% : 0.000002s : 9: predicate.dict_set_item_eliminator 1.52% : 0.000003s : 8: predicate.dumpgradient_eliminate 0.26% : 0.000000s : 4: predicate.elim_not_effective 0.62% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.07% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.24% : 0.000002s : 13: predicate.environ_get_add_eliminate 1.00% : 0.000002s : 13: predicate.environ_get_depend_swap 1.67% : 0.000003s : 21: predicate.environ_get_eliminate 1.01% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.77% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.60% : 0.000003s : 10: predicate.float_depend_g_call 0.61% : 0.000001s : 8: predicate.float_environ_get_switch 0.94% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.23% : 0.000000s : 4: predicate.fold_const_symbol 0.96% : 0.000002s : 8: predicate.get_grad_eliminate 0.26% : 0.000000s : 4: predicate.graph_param_transform 0.69% : 0.000001s : 8: predicate.incorporate_call 0.57% : 0.000001s : 8: predicate.incorporate_call_switch 5.78% : 0.000010s : 43: predicate.inline 0.92% : 0.000002s : 8: predicate.inline_without_move 0.42% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.00% : 0.000002s : 8: predicate.less_batch_normalization 1.73% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.16% : 0.000004s : 26: predicate.load_eliminater 1.06% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.30% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.80% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.63% : 0.000001s : 8: predicate.merge_addn 0.77% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.77% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.70% : 0.000001s : 9: predicate.minmaximum_grad 1.51% : 0.000003s : 4: predicate.mutable_eliminate 0.48% : 0.000001s : 4: predicate.opt_reshape 0.60% : 0.000001s : 4: predicate.parallel_virtual_node 1.30% : 0.000002s : 10: predicate.partial_defer_inline 1.19% : 0.000002s : 13: predicate.partial_eliminate 0.79% : 0.000001s : 9: predicate.print_const_string_wrapper 0.88% : 0.000002s : 8: predicate.reduce_all_const_elim 1.12% : 0.000002s : 9: predicate.reduce_eliminate 2.05% : 0.000004s : 26: predicate.redundant_stop_gradient_eliminater 0.72% : 0.000001s : 8: predicate.remove_not_recompute_node 1.55% : 0.000003s : 17: predicate.replace_applicator 0.76% : 0.000001s : 8: predicate.replace_old_param 0.42% : 0.000001s : 4: predicate.reset_defer_inline 0.90% : 0.000002s : 9: predicate.reshape_eliminate 0.83% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.58% : 0.000001s : 4: predicate.row_tensor_eliminate 1.25% : 0.000002s : 8: predicate.same_eliminate 0.51% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.89% : 0.000002s : 8: predicate.shard_identity_eliminate 0.89% : 0.000002s : 8: predicate.special_op_eliminate 0.97% : 0.000002s : 8: predicate.specialize_transform 1.76% : 0.000003s : 8: predicate.split_environ_get_set_with_tuple_value 0.92% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.47% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.82% : 0.000001s : 10: predicate.switch_defer_inline 1.58% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.34% : 0.000006s : 35: predicate.switch_simplify 0.91% : 0.000002s : 9: predicate.tile_eliminate 0.83% : 0.000001s : 9: predicate.transpose_eliminate 2.00% : 0.000004s : 17: predicate.tuple_list_convert_item_index_to_positive 1.69% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.31% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.29% : 0.000006s : 25: predicate.tuple_list_get_item_eliminator 1.61% : 0.000003s : 17: predicate.tuple_list_get_set_item_eliminator 2.73% : 0.000005s : 25: predicate.tuple_list_set_item_eliminator 1.54% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.91% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.83% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.43% : 0.000001s : 4: predicate.value_based_eliminate 0.97% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.94% : 0.000002s : 8: predicate.virtual_output_eliminate 0.34% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.52% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000270 4 7.63% : 0.000021s : 1: func_graph_cloner_run.FuncGraphClonerGraph 92.37% : 0.000249s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.280673 192 0.00% : 0.000004s : 1: ForceFp32Comm 1.58% : 0.004443s : 1: add_attr 1.58% : 0.004422s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.02% : 0.000061s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.02% : 0.000060s : 1: auto_monad 0.01% : 0.000025s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.22% : 0.000625s : 1: bootstrap 0.01% : 0.000038s : 1: cconv 0.00% : 0.000010s : 1: comm_op_add_attrs 0.01% : 0.000019s : 1: control_data_broadcast_order 0.00% : 0.000012s : 1: convert_after_rewriter 0.01% : 0.000033s : 1: cse_after_recomputation 0.00% : 0.000007s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000010s : 1: environ_conv 0.01% : 0.000018s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000008s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000005s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.18% : 0.000505s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.28% : 0.000789s : 1: mutable_eliminate 0.00% : 0.000008s : 1: offloading_packed_experts 0.01% : 0.000017s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000023s : 1: opt.transform.mutable_eliminate 0.33% : 0.000931s : 78: opt.transform.opt_a 0.01% : 0.000031s : 1: opt.transform.opt_after_cconv 0.01% : 0.000034s : 1: opt.transform.opt_after_jit_grad 0.04% : 0.000116s : 28: opt.transform.opt_b 0.02% : 0.000055s : 2: opt.transform.opt_trans_graph 0.02% : 0.000043s : 4: opt.transform.symbol_engine_opt 31.18% : 0.087511s : 1: opt_a 0.04% : 0.000123s : 1: opt_after_cconv 0.24% : 0.000686s : 1: opt_after_jit_grad 0.09% : 0.000252s : 1: opt_b 32.07% : 0.090007s : 1: optimize 0.01% : 0.000024s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000009s : 1: order_py_execute_after_rewriter 0.01% : 0.000026s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000007s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000006s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000010s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000008s : 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 0.02% : 0.000055s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000038s : 1: pre_auto_parallel 0.01% : 0.000025s : 1: py_interpret_to_execute 0.01% : 0.000025s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.01% : 0.000022s : 1: remove_dup_value 30.32% : 0.085111s : 1: renormalize.infer 0.14% : 0.000393s : 1: renormalize.specialize 0.00% : 0.000005s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.02% : 0.000054s : 1: rewriter_after_opt_a 0.02% : 0.000055s : 1: rewriter_before_opt_a 0.00% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.03% : 0.000091s : 1: symbol_engine_optimizer 0.03% : 0.000088s : 1: tuple_transform 1.26% : 0.003527s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_bf16[dim_type0-True-batch_size_list1-shape0-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_bf16[dim_type0-True-batch_size_list1-shape0-ms_dtype0],max_mem:196.0M [WARNING] ME(152945:281473835274032,MainProcess):2026-01-29-17:38:50.344.858 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.0156743, [21] [bootstrap]: 0.00080639 [type_inference]: 0.00394845 [event_method]: 1.244e-05 [auto_monad]: 5.676e-05 [graph_reusing]: 5.28002e-06 [inline]: 2.40002e-06 [add_attr]: 0.0048602, [1] [add_attr_with_inline]: 0.00484395, [1] [Cycle 1]: 5.895e-05, [2] [tag_attr]: 1.61e-05 [meta_addattr_fg_expand]: 3.18e-06 [parallel-infer-symbol]: 4.02998e-06 [pre_auto_parallel]: 3.325e-05 [insert-virtual-dataset]: 2.91999e-06 [parallel-infer-symbol-second]: 9.20001e-07 [dataset_repeat_opt]: 1.94999e-06 [pipeline_split]: 1.94999e-06 [optimize]: 0.00503956, [53] [py_interpret_to_execute]: 2.098e-05 [rewriter_before_opt_a]: 5.056e-05 [opt_a]: 0.00256154, [2] [Cycle 1]: 0.00182893, [45] [expand_dump_flag]: 2.98998e-06 [switch_simplify]: 2.432e-05 [loop_unroll]: 1.213e-05 [a_1]: 0.00030926 [with_stream_mark]: 2.213e-05 [recompute_prepare]: 8.62e-06 [updatestate_depend_eliminate]: 4.13999e-06 [updatestate_assign_eliminate]: 3.75998e-06 [updatestate_loads_eliminate]: 3.58e-06 [parameter_eliminate]: 2.27001e-06 [a_2]: 8.572e-05 [accelerated_algorithm]: 7.53e-06 [shard]: 2.14e-06 [meta_shard_fg_expand]: 1.74e-06 [shard_inline]: 6.66e-06 [merge_send_recv]: 7.93001e-06 [auto_parallel]: 7.80998e-06 [parallel]: 2.853e-05 [flash_sp]: 1.016e-05 [merge_comm]: 4.38999e-06 [allreduce_fusion]: 3.73999e-06 [matmul_add_comm_reduction]: 9.86998e-06 [allreduce_slice_to_reducescatter]: 1.03001e-06 [virtual_shard_identity]: 1.035e-05 [virtual_dataset]: 6.63e-06 [get_grad_eliminate_]: 6.33998e-06 [virtual_output]: 6.49001e-06 [merge_forward]: 4.18999e-06 [cell_reuse_recompute_pass]: 1.68997e-06 [offload_activation]: 1.099e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.602e-05 [merge_recompute_call_nodes]: 1.54e-06 [before_grad]: 1.252e-05 [set_forward_comm_id_for_comm_node_pass]: 4.28001e-06 [meta_fg_expand]: 3.18e-06 [flash_sp_send_recv_attached]: 3.41001e-06 [receive_attached]: 2.84001e-06 [after_resolve]: 1.351e-05 [a_after_grad]: 1.025e-05 [renormalize]: 0.00076416 [add_forward_monad_depend]: 1.345e-05 [auto_monad_grad]: 2.63998e-06 [auto_monad_eliminator]: 1.711e-05 [cse]: 3.914e-05 [a_3]: 5.62e-05 [Cycle 2]: 0.00072138, [45] [expand_dump_flag]: 1.37999e-06 [switch_simplify]: 8.69e-06 [loop_unroll]: 6.37001e-06 [a_1]: 0.0001333 [with_stream_mark]: 1.687e-05 [recompute_prepare]: 7.61999e-06 [updatestate_depend_eliminate]: 4.2e-06 [updatestate_assign_eliminate]: 3.46001e-06 [updatestate_loads_eliminate]: 3.2e-06 [parameter_eliminate]: 1.75001e-06 [a_2]: 7.665e-05 [accelerated_algorithm]: 7.53999e-06 [shard]: 1.89999e-06 [meta_shard_fg_expand]: 2.32001e-06 [shard_inline]: 6.39999e-06 [merge_send_recv]: 7.78001e-06 [auto_parallel]: 7.86001e-06 [parallel]: 6.65998e-06 [flash_sp]: 3.54002e-06 [merge_comm]: 3.61001e-06 [allreduce_fusion]: 4.30999e-06 [matmul_add_comm_reduction]: 7.58999e-06 [allreduce_slice_to_reducescatter]: 4.39992e-07 [virtual_shard_identity]: 8.27e-06 [virtual_dataset]: 6.41998e-06 [get_grad_eliminate_]: 6.03998e-06 [virtual_output]: 6.39999e-06 [merge_forward]: 4.63001e-06 [cell_reuse_recompute_pass]: 1.55999e-06 [offload_activation]: 9.77999e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.687e-05 [merge_recompute_call_nodes]: 1.37e-06 [before_grad]: 1.064e-05 [set_forward_comm_id_for_comm_node_pass]: 3.73001e-06 [meta_fg_expand]: 2.37999e-06 [flash_sp_send_recv_attached]: 1.40999e-06 [receive_attached]: 1.37e-06 [after_resolve]: 1.259e-05 [a_after_grad]: 1.02e-05 [renormalize]: 1.00001e-07 [add_forward_monad_depend]: 2.69001e-06 [auto_monad_grad]: 1.67999e-06 [auto_monad_eliminator]: 9.78002e-06 [cse]: 2.323e-05 [a_3]: 4.027e-05 [py_interpret_to_execute_after_opt_a]: 1.665e-05 [slice_cell_reuse_recomputed_activation]: 2.18998e-06 [rewriter_after_opt_a]: 5.13e-05 [convert_after_rewriter]: 9.38002e-06 [order_py_execute_after_rewriter]: 5.97001e-06 [mutable_eliminate]: 0.00073419 [opt_b]: 0.0002393, [1] [Cycle 1]: 0.00023063, [7] [b_1]: 0.00013369 [b_2]: 8.93002e-06 [updatestate_depend_eliminate]: 1.074e-05 [updatestate_assign_eliminate]: 3.25002e-06 [updatestate_loads_eliminate]: 3.18e-06 [renormalize]: 9.39996e-07 [cse]: 3.283e-05 [optimize_parallel_all_gather_comm]: 2.478e-05 [overlap_param_gather]: 1.90001e-06 [cconv]: 3.636e-05 [loop_unroll]: 0.00048583 [opt_after_cconv]: 0.00011604, [1] [Cycle 1]: 0.00010917, [7] [c_1]: 3.271e-05 [parameter_eliminate]: 5.61e-06 [updatestate_depend_eliminate]: 7.05e-06 [updatestate_assign_eliminate]: 2.82002e-06 [updatestate_loads_eliminate]: 2.58003e-06 [cse]: 2.318e-05 [renormalize]: 6.30011e-07 [remove_dup_value]: 1.939e-05 [tuple_transform]: 8.454e-05, [1] [Cycle 1]: 7.991e-05, [4] [d_1]: 5.068e-05 [none_parameter_eliminate]: 1.89e-06 [renormalize]: 2.20025e-07 [switch_simplify]: 7.61999e-06 [partial_unused_args_eliminate]: 1.97001e-06 [add_recomputation]: 5.854e-05 [cse_after_recomputation]: 2.606e-05, [1] [Cycle 1]: 2.11e-05, [1] [cse]: 1.509e-05 [environ_conv]: 8.31002e-06 [swap_dp_allreduce_reducescatter]: 6.91001e-06 [bias_add_comm_swap]: 3.42002e-06 [label_micro_interleaved_index]: 4.55999e-06 [label_fine_grained_interleaved_index]: 2.66e-06 [merge_cast_opt]: 1.50001e-06 [slice_recompute_activation]: 2.64999e-06 [micro_interleaved_order_control]: 2.22001e-06 [assign_add_opt]: 1.28002e-06 [ForceFp32Comm]: 7.89994e-07 [remove_cast_before_assign_add]: 1.12e-06 [full_micro_interleaved_order_control]: 2.75997e-06 [reorder_send_recv_between_fp_bp]: 2.86999e-06 [comm_op_add_attrs]: 1.65001e-06 [add_comm_op_reuse_tag]: 1.32e-06 [interleave_split_concat_branches]: 1.40999e-06 [interleave_parallel_branches]: 9.80013e-07 [overlap_opt_shard_in_pipeline]: 1.53002e-06 [overlap_opt_shard_grad_in_pipeline]: 1.81998e-06 [control_data_broadcast_order]: 1.401e-05 [grouped_pairwise_exchange_alltoall]: 1.84998e-06 [offloading_packed_experts]: 4.59002e-06 [overlap_recompute_and_grad_model_parallel]: 4.87e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.20999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.32999e-06 [overlap_recompute_comm]: 2.09999e-06 [overlap_grad_ring_attention]: 4.30999e-06 [overlap_grad_flash_sp]: 2.461e-05 [begin_end_overlap_inline]: 5.00004e-07 [split_matmul_comm_elemetwise]: 2.47001e-06 [split_layernorm_comm]: 1.68002e-06 [handle_group_info]: 1.10001e-06 [symbol_engine_optimizer]: 0.00011091, [1] [Cycle 1]: 0.00010596, [6] [build]: 3.83001e-06 [elim_shapecalc]: 1.163e-05 [elim_not_effective]: 1.517e-05 [opt_reshape]: 7.38e-06 [fold_const_symbol]: 3.728e-05 [renormalize]: 1.90019e-07 [detach_backward]: 2.17999e-06 [pipeline_parallel_scheduler]: 1.74e-06 [auto_monad_reorder]: 2.015e-05 [get_jit_bprop_graph]: 2.09999e-06 [rewriter_after_jit_bprop_graph]: 5.77001e-06 [opt_after_jit_grad]: 0.00061601 [validate]: 5.196e-05 Sums bootstrap : 0.000806s : 8.29% type_inference : 0.003948s : 40.57% event_method : 0.000012s : 0.13% auto_monad : 0.000057s : 0.58% graph_reusing : 0.000005s : 0.05% inline : 0.000002s : 0.02% add_attr.add_attr_with_inline.tag_attr : 0.000016s : 0.17% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.03% parallel-infer-symbol : 0.000004s : 0.04% pre_auto_parallel : 0.000033s : 0.34% insert-virtual-dataset : 0.000003s : 0.03% parallel-infer-symbol-second : 0.000001s : 0.01% dataset_repeat_opt : 0.000002s : 0.02% pipeline_split : 0.000002s : 0.02% optimize.py_interpret_to_execute : 0.000021s : 0.22% optimize.rewriter_before_opt_a : 0.000051s : 0.52% optimize.opt_a.expand_dump_flag : 0.000004s : 0.04% optimize.opt_a.switch_simplify : 0.000033s : 0.34% optimize.opt_a.loop_unroll : 0.000018s : 0.19% optimize.opt_a.a_1 : 0.000443s : 4.55% optimize.opt_a.with_stream_mark : 0.000039s : 0.40% optimize.opt_a.recompute_prepare : 0.000016s : 0.17% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.09% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.07% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.07% optimize.opt_a.parameter_eliminate : 0.000004s : 0.04% optimize.opt_a.a_2 : 0.000162s : 1.67% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.15% optimize.opt_a.shard : 0.000004s : 0.04% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.04% optimize.opt_a.shard_inline : 0.000013s : 0.13% optimize.opt_a.merge_send_recv : 0.000016s : 0.16% optimize.opt_a.auto_parallel : 0.000016s : 0.16% optimize.opt_a.parallel : 0.000035s : 0.36% optimize.opt_a.flash_sp : 0.000014s : 0.14% optimize.opt_a.merge_comm : 0.000008s : 0.08% optimize.opt_a.allreduce_fusion : 0.000008s : 0.08% optimize.opt_a.matmul_add_comm_reduction : 0.000017s : 0.18% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.02% optimize.opt_a.virtual_shard_identity : 0.000019s : 0.19% optimize.opt_a.virtual_dataset : 0.000013s : 0.13% optimize.opt_a.get_grad_eliminate_ : 0.000012s : 0.13% optimize.opt_a.virtual_output : 0.000013s : 0.13% optimize.opt_a.merge_forward : 0.000009s : 0.09% optimize.opt_a.cell_reuse_recompute_pass : 0.000003s : 0.03% optimize.opt_a.offload_activation : 0.000021s : 0.21% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000033s : 0.34% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.03% optimize.opt_a.before_grad : 0.000023s : 0.24% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.08% optimize.opt_a.meta_fg_expand : 0.000006s : 0.06% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.05% optimize.opt_a.receive_attached : 0.000004s : 0.04% optimize.opt_a.after_resolve : 0.000026s : 0.27% optimize.opt_a.a_after_grad : 0.000020s : 0.21% optimize.opt_a.renormalize : 0.000764s : 7.85% optimize.opt_a.add_forward_monad_depend : 0.000016s : 0.17% optimize.opt_a.auto_monad_grad : 0.000004s : 0.04% optimize.opt_a.auto_monad_eliminator : 0.000027s : 0.28% optimize.opt_a.cse : 0.000062s : 0.64% optimize.opt_a.a_3 : 0.000096s : 0.99% optimize.py_interpret_to_execute_after_opt_a : 0.000017s : 0.17% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.02% optimize.rewriter_after_opt_a : 0.000051s : 0.53% optimize.convert_after_rewriter : 0.000009s : 0.10% optimize.order_py_execute_after_rewriter : 0.000006s : 0.06% optimize.mutable_eliminate : 0.000734s : 7.54% optimize.opt_b.b_1 : 0.000134s : 1.37% optimize.opt_b.b_2 : 0.000009s : 0.09% optimize.opt_b.updatestate_depend_eliminate : 0.000011s : 0.11% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.03% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.03% optimize.opt_b.renormalize : 0.000001s : 0.01% optimize.opt_b.cse : 0.000033s : 0.34% optimize.optimize_parallel_all_gather_comm : 0.000025s : 0.25% optimize.overlap_param_gather : 0.000002s : 0.02% optimize.cconv : 0.000036s : 0.37% optimize.loop_unroll : 0.000486s : 4.99% optimize.opt_after_cconv.c_1 : 0.000033s : 0.34% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.06% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000007s : 0.07% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.03% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000003s : 0.03% optimize.opt_after_cconv.cse : 0.000023s : 0.24% optimize.opt_after_cconv.renormalize : 0.000001s : 0.01% optimize.remove_dup_value : 0.000019s : 0.20% optimize.tuple_transform.d_1 : 0.000051s : 0.52% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.02% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000008s : 0.08% optimize.partial_unused_args_eliminate : 0.000002s : 0.02% optimize.add_recomputation : 0.000059s : 0.60% optimize.cse_after_recomputation.cse : 0.000015s : 0.16% optimize.environ_conv : 0.000008s : 0.09% optimize.swap_dp_allreduce_reducescatter : 0.000007s : 0.07% optimize.bias_add_comm_swap : 0.000003s : 0.04% optimize.label_micro_interleaved_index : 0.000005s : 0.05% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.03% optimize.merge_cast_opt : 0.000002s : 0.02% optimize.slice_recompute_activation : 0.000003s : 0.03% optimize.micro_interleaved_order_control : 0.000002s : 0.02% optimize.assign_add_opt : 0.000001s : 0.01% optimize.ForceFp32Comm : 0.000001s : 0.01% optimize.remove_cast_before_assign_add : 0.000001s : 0.01% optimize.full_micro_interleaved_order_control : 0.000003s : 0.03% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.03% optimize.comm_op_add_attrs : 0.000002s : 0.02% optimize.add_comm_op_reuse_tag : 0.000001s : 0.01% optimize.interleave_split_concat_branches : 0.000001s : 0.01% optimize.interleave_parallel_branches : 0.000001s : 0.01% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.02% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.02% optimize.control_data_broadcast_order : 0.000014s : 0.14% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.02% optimize.offloading_packed_experts : 0.000005s : 0.05% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.05% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.01% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.01% optimize.overlap_recompute_comm : 0.000002s : 0.02% optimize.overlap_grad_ring_attention : 0.000004s : 0.04% optimize.overlap_grad_flash_sp : 0.000025s : 0.25% optimize.begin_end_overlap_inline : 0.000001s : 0.01% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.03% optimize.split_layernorm_comm : 0.000002s : 0.02% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000004s : 0.04% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000012s : 0.12% optimize.symbol_engine_optimizer.elim_not_effective : 0.000015s : 0.16% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.08% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000037s : 0.38% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.02% pipeline_parallel_scheduler : 0.000002s : 0.02% auto_monad_reorder : 0.000020s : 0.21% get_jit_bprop_graph : 0.000002s : 0.02% rewriter_after_jit_bprop_graph : 0.000006s : 0.06% opt_after_jit_grad : 0.000616s : 6.33% validate : 0.000052s : 0.53% Time group info: ------[substitution.] 0.000114 21 1.79% : 0.000002s : 2: substitution.elim_not_effective 1.33% : 0.000002s : 2: substitution.fold_const_symbol 6.14% : 0.000007s : 4: substitution.graph_param_transform 76.52% : 0.000087s : 1: substitution.inline 4.04% : 0.000005s : 4: substitution.j_node_and_user_rematch 4.67% : 0.000005s : 4: substitution.remove_not_recompute_node 5.50% : 0.000006s : 4: substitution.replace_old_param ------[type_inference.] 0.003884 2 89.66% : 0.003482s : 1: type_inference.infer 10.34% : 0.000402s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000086 1 100.00% : 0.000086s : 1: match.inline ------[predicate.] 0.000158 967 0.79% : 0.000001s : 9: predicate.accumulaten_eliminater 1.11% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.62% : 0.000001s : 8: predicate.addn_check_dump 0.72% : 0.000001s : 9: predicate.addn_zero_filter 0.70% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.04% : 0.000003s : 17: predicate.arithmetic_simplify 0.87% : 0.000001s : 9: predicate.cast_eliminate 0.98% : 0.000002s : 8: predicate.check_bprop_eliminate 0.76% : 0.000001s : 8: predicate.compare_switch_simplify 0.25% : 0.000000s : 4: predicate.const_output_eliminate 0.71% : 0.000001s : 8: predicate.depend_value_elim 0.78% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.84% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.75% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.26% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.41% : 0.000001s : 4: predicate.elim_not_effective 0.54% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.19% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.99% : 0.000002s : 13: predicate.environ_get_add_eliminate 1.02% : 0.000002s : 13: predicate.environ_get_depend_swap 1.78% : 0.000003s : 21: predicate.environ_get_eliminate 1.09% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.89% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.78% : 0.000003s : 10: predicate.float_depend_g_call 0.70% : 0.000001s : 8: predicate.float_environ_get_switch 0.98% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.24% : 0.000000s : 4: predicate.fold_const_symbol 0.90% : 0.000001s : 8: predicate.get_grad_eliminate 0.30% : 0.000000s : 4: predicate.graph_param_transform 0.79% : 0.000001s : 8: predicate.incorporate_call 0.63% : 0.000001s : 8: predicate.incorporate_call_switch 6.73% : 0.000011s : 43: predicate.inline 0.90% : 0.000001s : 8: predicate.inline_without_move 0.41% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.20% : 0.000002s : 8: predicate.less_batch_normalization 1.68% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.18% : 0.000003s : 26: predicate.load_eliminater 1.37% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.29% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.92% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.70% : 0.000001s : 8: predicate.merge_addn 0.97% : 0.000002s : 8: predicate.micro_step_allgather_replace 0.76% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.69% : 0.000001s : 9: predicate.minmaximum_grad 1.44% : 0.000002s : 4: predicate.mutable_eliminate 0.49% : 0.000001s : 4: predicate.opt_reshape 0.60% : 0.000001s : 4: predicate.parallel_virtual_node 1.10% : 0.000002s : 10: predicate.partial_defer_inline 1.29% : 0.000002s : 13: predicate.partial_eliminate 0.92% : 0.000001s : 9: predicate.print_const_string_wrapper 0.72% : 0.000001s : 8: predicate.reduce_all_const_elim 1.05% : 0.000002s : 9: predicate.reduce_eliminate 2.03% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 1.11% : 0.000002s : 8: predicate.remove_not_recompute_node 1.21% : 0.000002s : 17: predicate.replace_applicator 0.70% : 0.000001s : 8: predicate.replace_old_param 0.48% : 0.000001s : 4: predicate.reset_defer_inline 0.79% : 0.000001s : 9: predicate.reshape_eliminate 0.81% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.54% : 0.000001s : 4: predicate.row_tensor_eliminate 0.91% : 0.000001s : 8: predicate.same_eliminate 0.58% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.95% : 0.000002s : 8: predicate.shard_identity_eliminate 0.93% : 0.000001s : 8: predicate.special_op_eliminate 1.07% : 0.000002s : 8: predicate.specialize_transform 1.36% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 1.15% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.55% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.88% : 0.000001s : 10: predicate.switch_defer_inline 1.57% : 0.000002s : 18: predicate.switch_layer_defer_inline 3.67% : 0.000006s : 35: predicate.switch_simplify 0.74% : 0.000001s : 9: predicate.tile_eliminate 0.81% : 0.000001s : 9: predicate.transpose_eliminate 1.50% : 0.000002s : 17: predicate.tuple_list_convert_item_index_to_positive 1.62% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.31% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.15% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.47% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.47% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.51% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 2.01% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.88% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.71% : 0.000001s : 4: predicate.value_based_eliminate 0.79% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.96% : 0.000002s : 8: predicate.virtual_output_eliminate 0.32% : 0.000000s : 4: predicate.virtual_view_grad_eliminate 0.60% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000235 4 9.93% : 0.000023s : 1: func_graph_cloner_run.FuncGraphClonerGraph 90.07% : 0.000212s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.027323 192 0.01% : 0.000004s : 1: ForceFp32Comm 17.81% : 0.004867s : 1: add_attr 17.74% : 0.004848s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.23% : 0.000063s : 1: add_recomputation 0.02% : 0.000005s : 1: assign_add_opt 0.23% : 0.000062s : 1: auto_monad 0.09% : 0.000024s : 1: auto_monad_reorder 0.01% : 0.000003s : 1: begin_end_overlap_inline 0.02% : 0.000006s : 1: bias_add_comm_swap 3.10% : 0.000848s : 1: bootstrap 0.15% : 0.000040s : 1: cconv 0.02% : 0.000004s : 1: comm_op_add_attrs 0.06% : 0.000017s : 1: control_data_broadcast_order 0.05% : 0.000013s : 1: convert_after_rewriter 0.11% : 0.000029s : 1: cse_after_recomputation 0.02% : 0.000005s : 1: dataset_repeat_opt 0.02% : 0.000006s : 1: detach_backward 0.04% : 0.000011s : 1: environ_conv 0.07% : 0.000019s : 1: event_method 0.03% : 0.000007s : 1: full_micro_interleaved_order_control 0.02% : 0.000005s : 1: get_jit_bprop_graph 0.03% : 0.000009s : 1: graph_reusing 0.02% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000004s : 1: handle_group_info 0.02% : 0.000005s : 1: inline 0.02% : 0.000006s : 1: insert-virtual-dataset 0.02% : 0.000004s : 1: interleave_parallel_branches 0.02% : 0.000004s : 1: interleave_split_concat_branches 0.02% : 0.000005s : 1: label_fine_grained_interleaved_index 0.03% : 0.000008s : 1: label_micro_interleaved_index 1.81% : 0.000496s : 1: loop_unroll 0.02% : 0.000004s : 1: merge_cast_opt 0.02% : 0.000005s : 1: micro_interleaved_order_control 2.74% : 0.000749s : 1: mutable_eliminate 0.03% : 0.000007s : 1: offloading_packed_experts 0.06% : 0.000016s : 1: opt.transform.loop_unroll_optimizer 0.08% : 0.000023s : 1: opt.transform.mutable_eliminate 3.15% : 0.000861s : 78: opt.transform.opt_a 0.11% : 0.000031s : 1: opt.transform.opt_after_cconv 0.10% : 0.000028s : 1: opt.transform.opt_after_jit_grad 0.41% : 0.000111s : 28: opt.transform.opt_b 0.20% : 0.000056s : 2: opt.transform.opt_trans_graph 0.25% : 0.000067s : 4: opt.transform.symbol_engine_opt 9.39% : 0.002565s : 1: opt_a 0.44% : 0.000119s : 1: opt_after_cconv 2.30% : 0.000629s : 1: opt_after_jit_grad 0.89% : 0.000243s : 1: opt_b 18.46% : 0.005045s : 1: optimize 0.11% : 0.000029s : 1: optimize_parallel_all_gather_comm 0.03% : 0.000009s : 1: order_py_execute_after_rewriter 0.10% : 0.000028s : 1: overlap_grad_flash_sp 0.02% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.03% : 0.000007s : 1: overlap_grad_ring_attention 0.02% : 0.000004s : 1: overlap_opt_shard_grad_in_pipeline 0.02% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.02% : 0.000005s : 1: overlap_param_gather 0.02% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.03% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.02% : 0.000005s : 1: overlap_recompute_comm 0.03% : 0.000008s : 1: parallel-infer-symbol 0.01% : 0.000004s : 1: parallel-infer-symbol-second 0.02% : 0.000005s : 1: partial_unused_args_eliminate 0.02% : 0.000005s : 1: pipeline_parallel_scheduler 0.02% : 0.000005s : 1: pipeline_split 0.14% : 0.000037s : 1: pre_auto_parallel 0.09% : 0.000025s : 1: py_interpret_to_execute 0.08% : 0.000021s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000004s : 1: remove_cast_before_assign_add 0.08% : 0.000023s : 1: remove_dup_value 1.77% : 0.000484s : 1: renormalize.infer 0.99% : 0.000271s : 1: renormalize.specialize 0.02% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.03% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.21% : 0.000057s : 1: rewriter_after_opt_a 0.20% : 0.000055s : 1: rewriter_before_opt_a 0.02% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.02% : 0.000005s : 1: slice_recompute_activation 0.02% : 0.000004s : 1: split_layernorm_comm 0.02% : 0.000005s : 1: split_matmul_comm_elemetwise 0.04% : 0.000010s : 1: swap_dp_allreduce_reducescatter 0.42% : 0.000114s : 1: symbol_engine_optimizer 0.32% : 0.000087s : 1: tuple_transform 14.53% : 0.003970s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_bf16[dim_type0-True-batch_size_list2-shape0-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_bf16[dim_type0-True-batch_size_list2-shape0-ms_dtype0],max_mem:516.0M [WARNING] ME(152945:281473835274032,MainProcess):2026-01-29-17:41:39.500.813 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.0146244, [21] [bootstrap]: 0.00063653 [type_inference]: 0.00365572 [event_method]: 1.121e-05 [auto_monad]: 5.182e-05 [graph_reusing]: 4.82998e-06 [inline]: 2.64001e-06 [add_attr]: 0.0045247, [1] [add_attr_with_inline]: 0.00450863, [1] [Cycle 1]: 6.464e-05, [2] [tag_attr]: 1.618e-05 [meta_addattr_fg_expand]: 3.35e-06 [parallel-infer-symbol]: 4e-06 [pre_auto_parallel]: 3.592e-05 [insert-virtual-dataset]: 2.62001e-06 [parallel-infer-symbol-second]: 7.39994e-07 [dataset_repeat_opt]: 2.32001e-06 [pipeline_split]: 1.67999e-06 [optimize]: 0.00493567, [53] [py_interpret_to_execute]: 2.066e-05 [rewriter_before_opt_a]: 5.063e-05 [opt_a]: 0.00248204, [2] [Cycle 1]: 0.00174517, [45] [expand_dump_flag]: 2.48e-06 [switch_simplify]: 2.369e-05 [loop_unroll]: 1.231e-05 [a_1]: 0.00030455 [with_stream_mark]: 2.189e-05 [recompute_prepare]: 8.42e-06 [updatestate_depend_eliminate]: 4.21001e-06 [updatestate_assign_eliminate]: 3.61999e-06 [updatestate_loads_eliminate]: 3.21001e-06 [parameter_eliminate]: 2.17001e-06 [a_2]: 8.543e-05 [accelerated_algorithm]: 7.98999e-06 [shard]: 2.58e-06 [meta_shard_fg_expand]: 1.81998e-06 [shard_inline]: 6.64999e-06 [merge_send_recv]: 8.99998e-06 [auto_parallel]: 7.33e-06 [parallel]: 4.004e-05 [flash_sp]: 9.55001e-06 [merge_comm]: 4.11001e-06 [allreduce_fusion]: 3.88001e-06 [matmul_add_comm_reduction]: 1.033e-05 [allreduce_slice_to_reducescatter]: 6.50005e-07 [virtual_shard_identity]: 8.95999e-06 [virtual_dataset]: 6.86001e-06 [get_grad_eliminate_]: 6.63e-06 [virtual_output]: 6.68e-06 [merge_forward]: 4.17e-06 [cell_reuse_recompute_pass]: 1.62999e-06 [offload_activation]: 1.036e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.473e-05 [merge_recompute_call_nodes]: 1.45001e-06 [before_grad]: 1.099e-05 [set_forward_comm_id_for_comm_node_pass]: 4.03001e-06 [meta_fg_expand]: 3.04001e-06 [flash_sp_send_recv_attached]: 2.95002e-06 [receive_attached]: 2.56e-06 [after_resolve]: 1.261e-05 [a_after_grad]: 1.038e-05 [renormalize]: 0.00068066 [add_forward_monad_depend]: 1.011e-05 [auto_monad_grad]: 2.51e-06 [auto_monad_eliminator]: 1.525e-05 [cse]: 5.584e-05 [a_3]: 5.33e-05 [Cycle 2]: 0.00072533, [45] [expand_dump_flag]: 1.69e-06 [switch_simplify]: 7.90998e-06 [loop_unroll]: 6.79001e-06 [a_1]: 0.00012894 [with_stream_mark]: 1.374e-05 [recompute_prepare]: 6.96999e-06 [updatestate_depend_eliminate]: 3.43999e-06 [updatestate_assign_eliminate]: 3.00002e-06 [updatestate_loads_eliminate]: 2.76999e-06 [parameter_eliminate]: 1.16997e-06 [a_2]: 0.00011073 [accelerated_algorithm]: 6.59001e-06 [shard]: 1.70001e-06 [meta_shard_fg_expand]: 1.37e-06 [shard_inline]: 6.33998e-06 [merge_send_recv]: 6.53e-06 [auto_parallel]: 8.03001e-06 [parallel]: 6.18998e-06 [flash_sp]: 3.8e-06 [merge_comm]: 3.71001e-06 [allreduce_fusion]: 3.61001e-06 [matmul_add_comm_reduction]: 7.51001e-06 [allreduce_slice_to_reducescatter]: 3.4002e-07 [virtual_shard_identity]: 7.6e-06 [virtual_dataset]: 6.68e-06 [get_grad_eliminate_]: 6.34001e-06 [virtual_output]: 6.19999e-06 [merge_forward]: 3.85e-06 [cell_reuse_recompute_pass]: 2.00002e-06 [offload_activation]: 9.09e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.427e-05 [merge_recompute_call_nodes]: 1.10999e-06 [before_grad]: 9.97999e-06 [set_forward_comm_id_for_comm_node_pass]: 4.12e-06 [meta_fg_expand]: 2.37999e-06 [flash_sp_send_recv_attached]: 1.05999e-06 [receive_attached]: 1.55999e-06 [after_resolve]: 1.149e-05 [a_after_grad]: 9.46998e-06 [renormalize]: 9.00181e-08 [add_forward_monad_depend]: 1.44e-06 [auto_monad_grad]: 1.19e-06 [auto_monad_eliminator]: 7.51001e-06 [cse]: 2.326e-05 [a_3]: 3.992e-05 [py_interpret_to_execute_after_opt_a]: 1.715e-05 [slice_cell_reuse_recomputed_activation]: 2.43e-06 [rewriter_after_opt_a]: 4.125e-05 [convert_after_rewriter]: 8.1e-06 [order_py_execute_after_rewriter]: 5.86e-06 [mutable_eliminate]: 0.00078544 [opt_b]: 0.00023943, [1] [Cycle 1]: 0.00023125, [7] [b_1]: 0.00013784 [b_2]: 9.19e-06 [updatestate_depend_eliminate]: 8.77999e-06 [updatestate_assign_eliminate]: 3.21999e-06 [updatestate_loads_eliminate]: 3.24001e-06 [renormalize]: 9.00007e-07 [cse]: 3.176e-05 [optimize_parallel_all_gather_comm]: 1.946e-05 [overlap_param_gather]: 2.12999e-06 [cconv]: 3.425e-05 [loop_unroll]: 0.00048435 [opt_after_cconv]: 0.00011407, [1] [Cycle 1]: 0.00010803, [7] [c_1]: 3.334e-05 [parameter_eliminate]: 4.52e-06 [updatestate_depend_eliminate]: 6.58e-06 [updatestate_assign_eliminate]: 2.84999e-06 [updatestate_loads_eliminate]: 2.64001e-06 [cse]: 2.32e-05 [renormalize]: 6.39993e-07 [remove_dup_value]: 4.583e-05 [tuple_transform]: 8.345e-05, [1] [Cycle 1]: 7.735e-05, [4] [d_1]: 4.865e-05 [none_parameter_eliminate]: 2.21e-06 [renormalize]: 1.8999e-07 [switch_simplify]: 7.23e-06 [partial_unused_args_eliminate]: 1.81e-06 [add_recomputation]: 5.374e-05 [cse_after_recomputation]: 2.466e-05, [1] [Cycle 1]: 1.98e-05, [1] [cse]: 1.409e-05 [environ_conv]: 6.81999e-06 [swap_dp_allreduce_reducescatter]: 6.38e-06 [bias_add_comm_swap]: 2.79999e-06 [label_micro_interleaved_index]: 3.93999e-06 [label_fine_grained_interleaved_index]: 2.89001e-06 [merge_cast_opt]: 1.33002e-06 [slice_recompute_activation]: 2.29001e-06 [micro_interleaved_order_control]: 2.84999e-06 [assign_add_opt]: 1.40001e-06 [ForceFp32Comm]: 8.80013e-07 [remove_cast_before_assign_add]: 1.34998e-06 [full_micro_interleaved_order_control]: 2.26e-06 [reorder_send_recv_between_fp_bp]: 2.74001e-06 [comm_op_add_attrs]: 1.12e-06 [add_comm_op_reuse_tag]: 1.17999e-06 [interleave_split_concat_branches]: 1.29e-06 [interleave_parallel_branches]: 1.01002e-06 [overlap_opt_shard_in_pipeline]: 5.36998e-06 [overlap_opt_shard_grad_in_pipeline]: 2.06e-06 [control_data_broadcast_order]: 1.359e-05 [grouped_pairwise_exchange_alltoall]: 1.74e-06 [offloading_packed_experts]: 4.43001e-06 [overlap_recompute_and_grad_model_parallel]: 4.65999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.09998e-06 [overlap_recompute_allgather_and_fa_grad]: 1.34003e-06 [overlap_recompute_comm]: 2.63e-06 [overlap_grad_ring_attention]: 4.4e-06 [overlap_grad_flash_sp]: 2.123e-05 [begin_end_overlap_inline]: 5.60016e-07 [split_matmul_comm_elemetwise]: 2.76e-06 [split_layernorm_comm]: 2.04999e-06 [handle_group_info]: 1.00999e-06 [symbol_engine_optimizer]: 7.882e-05, [1] [Cycle 1]: 7.367e-05, [6] [build]: 3.45e-06 [elim_shapecalc]: 1.023e-05 [elim_not_effective]: 1.352e-05 [opt_reshape]: 7.4e-06 [fold_const_symbol]: 1.002e-05 [renormalize]: 3.09985e-07 [detach_backward]: 2.42001e-06 [pipeline_parallel_scheduler]: 1.67001e-06 [auto_monad_reorder]: 1.816e-05 [get_jit_bprop_graph]: 1.72001e-06 [rewriter_after_jit_bprop_graph]: 6.55002e-06 [opt_after_jit_grad]: 0.0005008 [validate]: 4.607e-05 Sums bootstrap : 0.000637s : 7.00% type_inference : 0.003656s : 40.21% event_method : 0.000011s : 0.12% auto_monad : 0.000052s : 0.57% graph_reusing : 0.000005s : 0.05% inline : 0.000003s : 0.03% add_attr.add_attr_with_inline.tag_attr : 0.000016s : 0.18% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.04% parallel-infer-symbol : 0.000004s : 0.04% pre_auto_parallel : 0.000036s : 0.40% insert-virtual-dataset : 0.000003s : 0.03% parallel-infer-symbol-second : 0.000001s : 0.01% dataset_repeat_opt : 0.000002s : 0.03% pipeline_split : 0.000002s : 0.02% optimize.py_interpret_to_execute : 0.000021s : 0.23% optimize.rewriter_before_opt_a : 0.000051s : 0.56% optimize.opt_a.expand_dump_flag : 0.000004s : 0.05% optimize.opt_a.switch_simplify : 0.000032s : 0.35% optimize.opt_a.loop_unroll : 0.000019s : 0.21% optimize.opt_a.a_1 : 0.000433s : 4.77% optimize.opt_a.with_stream_mark : 0.000036s : 0.39% optimize.opt_a.recompute_prepare : 0.000015s : 0.17% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.08% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.07% optimize.opt_a.updatestate_loads_eliminate : 0.000006s : 0.07% optimize.opt_a.parameter_eliminate : 0.000003s : 0.04% optimize.opt_a.a_2 : 0.000196s : 2.16% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.16% optimize.opt_a.shard : 0.000004s : 0.05% optimize.opt_a.meta_shard_fg_expand : 0.000003s : 0.04% optimize.opt_a.shard_inline : 0.000013s : 0.14% optimize.opt_a.merge_send_recv : 0.000016s : 0.17% optimize.opt_a.auto_parallel : 0.000015s : 0.17% optimize.opt_a.parallel : 0.000046s : 0.51% optimize.opt_a.flash_sp : 0.000013s : 0.15% optimize.opt_a.merge_comm : 0.000008s : 0.09% optimize.opt_a.allreduce_fusion : 0.000007s : 0.08% optimize.opt_a.matmul_add_comm_reduction : 0.000018s : 0.20% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000017s : 0.18% optimize.opt_a.virtual_dataset : 0.000014s : 0.15% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.14% optimize.opt_a.virtual_output : 0.000013s : 0.14% optimize.opt_a.merge_forward : 0.000008s : 0.09% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.04% optimize.opt_a.offload_activation : 0.000019s : 0.21% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000029s : 0.32% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.03% optimize.opt_a.before_grad : 0.000021s : 0.23% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.09% optimize.opt_a.meta_fg_expand : 0.000005s : 0.06% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.04% optimize.opt_a.receive_attached : 0.000004s : 0.05% optimize.opt_a.after_resolve : 0.000024s : 0.27% optimize.opt_a.a_after_grad : 0.000020s : 0.22% optimize.opt_a.renormalize : 0.000681s : 7.49% optimize.opt_a.add_forward_monad_depend : 0.000012s : 0.13% optimize.opt_a.auto_monad_grad : 0.000004s : 0.04% optimize.opt_a.auto_monad_eliminator : 0.000023s : 0.25% optimize.opt_a.cse : 0.000079s : 0.87% optimize.opt_a.a_3 : 0.000093s : 1.03% optimize.py_interpret_to_execute_after_opt_a : 0.000017s : 0.19% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.03% optimize.rewriter_after_opt_a : 0.000041s : 0.45% optimize.convert_after_rewriter : 0.000008s : 0.09% optimize.order_py_execute_after_rewriter : 0.000006s : 0.06% optimize.mutable_eliminate : 0.000785s : 8.64% optimize.opt_b.b_1 : 0.000138s : 1.52% optimize.opt_b.b_2 : 0.000009s : 0.10% optimize.opt_b.updatestate_depend_eliminate : 0.000009s : 0.10% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.04% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.04% optimize.opt_b.renormalize : 0.000001s : 0.01% optimize.opt_b.cse : 0.000032s : 0.35% optimize.optimize_parallel_all_gather_comm : 0.000019s : 0.21% optimize.overlap_param_gather : 0.000002s : 0.02% optimize.cconv : 0.000034s : 0.38% optimize.loop_unroll : 0.000484s : 5.33% optimize.opt_after_cconv.c_1 : 0.000033s : 0.37% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.05% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000007s : 0.07% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.03% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000003s : 0.03% optimize.opt_after_cconv.cse : 0.000023s : 0.26% optimize.opt_after_cconv.renormalize : 0.000001s : 0.01% optimize.remove_dup_value : 0.000046s : 0.50% optimize.tuple_transform.d_1 : 0.000049s : 0.54% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.02% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.08% optimize.partial_unused_args_eliminate : 0.000002s : 0.02% optimize.add_recomputation : 0.000054s : 0.59% optimize.cse_after_recomputation.cse : 0.000014s : 0.15% optimize.environ_conv : 0.000007s : 0.08% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.07% optimize.bias_add_comm_swap : 0.000003s : 0.03% optimize.label_micro_interleaved_index : 0.000004s : 0.04% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.03% optimize.merge_cast_opt : 0.000001s : 0.01% optimize.slice_recompute_activation : 0.000002s : 0.03% optimize.micro_interleaved_order_control : 0.000003s : 0.03% optimize.assign_add_opt : 0.000001s : 0.02% optimize.ForceFp32Comm : 0.000001s : 0.01% optimize.remove_cast_before_assign_add : 0.000001s : 0.01% optimize.full_micro_interleaved_order_control : 0.000002s : 0.02% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.03% optimize.comm_op_add_attrs : 0.000001s : 0.01% optimize.add_comm_op_reuse_tag : 0.000001s : 0.01% optimize.interleave_split_concat_branches : 0.000001s : 0.01% optimize.interleave_parallel_branches : 0.000001s : 0.01% optimize.overlap_opt_shard_in_pipeline : 0.000005s : 0.06% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.02% optimize.control_data_broadcast_order : 0.000014s : 0.15% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.02% optimize.offloading_packed_experts : 0.000004s : 0.05% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.05% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.01% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.01% optimize.overlap_recompute_comm : 0.000003s : 0.03% optimize.overlap_grad_ring_attention : 0.000004s : 0.05% optimize.overlap_grad_flash_sp : 0.000021s : 0.23% optimize.begin_end_overlap_inline : 0.000001s : 0.01% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.03% optimize.split_layernorm_comm : 0.000002s : 0.02% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000003s : 0.04% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000010s : 0.11% optimize.symbol_engine_optimizer.elim_not_effective : 0.000014s : 0.15% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.08% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.11% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.03% pipeline_parallel_scheduler : 0.000002s : 0.02% auto_monad_reorder : 0.000018s : 0.20% get_jit_bprop_graph : 0.000002s : 0.02% rewriter_after_jit_bprop_graph : 0.000007s : 0.07% opt_after_jit_grad : 0.000501s : 5.51% validate : 0.000046s : 0.51% Time group info: ------[substitution.] 0.000111 21 1.63% : 0.000002s : 2: substitution.elim_not_effective 1.24% : 0.000001s : 2: substitution.fold_const_symbol 5.91% : 0.000007s : 4: substitution.graph_param_transform 78.07% : 0.000087s : 1: substitution.inline 3.72% : 0.000004s : 4: substitution.j_node_and_user_rematch 4.89% : 0.000005s : 4: substitution.remove_not_recompute_node 4.54% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.003596 2 89.49% : 0.003219s : 1: type_inference.infer 10.51% : 0.000378s : 1: type_inference.specialize ------[replace.] 0.000016 1 100.00% : 0.000016s : 1: replace.inline ------[match.] 0.000086 1 100.00% : 0.000086s : 1: match.inline ------[predicate.] 0.000160 967 0.73% : 0.000001s : 9: predicate.accumulaten_eliminater 0.76% : 0.000001s : 4: predicate.ad_related_special_op_eliminate 0.61% : 0.000001s : 8: predicate.addn_check_dump 0.73% : 0.000001s : 9: predicate.addn_zero_filter 0.69% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.58% : 0.000004s : 17: predicate.arithmetic_simplify 1.03% : 0.000002s : 9: predicate.cast_eliminate 0.75% : 0.000001s : 8: predicate.check_bprop_eliminate 0.61% : 0.000001s : 8: predicate.compare_switch_simplify 0.24% : 0.000000s : 4: predicate.const_output_eliminate 0.64% : 0.000001s : 8: predicate.depend_value_elim 0.84% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.93% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.83% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.41% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.56% : 0.000001s : 4: predicate.elim_not_effective 0.58% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.08% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.03% : 0.000002s : 13: predicate.environ_get_add_eliminate 1.06% : 0.000002s : 13: predicate.environ_get_depend_swap 1.73% : 0.000003s : 21: predicate.environ_get_eliminate 1.01% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.78% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.60% : 0.000003s : 10: predicate.float_depend_g_call 0.63% : 0.000001s : 8: predicate.float_environ_get_switch 0.93% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.24% : 0.000000s : 4: predicate.fold_const_symbol 0.79% : 0.000001s : 8: predicate.get_grad_eliminate 0.53% : 0.000001s : 4: predicate.graph_param_transform 0.80% : 0.000001s : 8: predicate.incorporate_call 0.61% : 0.000001s : 8: predicate.incorporate_call_switch 6.31% : 0.000010s : 43: predicate.inline 1.19% : 0.000002s : 8: predicate.inline_without_move 0.41% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.10% : 0.000002s : 8: predicate.less_batch_normalization 1.66% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.09% : 0.000003s : 26: predicate.load_eliminater 0.95% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.36% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.92% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.77% : 0.000001s : 8: predicate.merge_addn 0.74% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.77% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.70% : 0.000001s : 9: predicate.minmaximum_grad 1.48% : 0.000002s : 4: predicate.mutable_eliminate 0.53% : 0.000001s : 4: predicate.opt_reshape 0.92% : 0.000001s : 4: predicate.parallel_virtual_node 1.23% : 0.000002s : 10: predicate.partial_defer_inline 1.25% : 0.000002s : 13: predicate.partial_eliminate 0.84% : 0.000001s : 9: predicate.print_const_string_wrapper 0.67% : 0.000001s : 8: predicate.reduce_all_const_elim 1.31% : 0.000002s : 9: predicate.reduce_eliminate 2.15% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.84% : 0.000001s : 8: predicate.remove_not_recompute_node 1.23% : 0.000002s : 17: predicate.replace_applicator 0.72% : 0.000001s : 8: predicate.replace_old_param 0.43% : 0.000001s : 4: predicate.reset_defer_inline 0.80% : 0.000001s : 9: predicate.reshape_eliminate 0.75% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.60% : 0.000001s : 4: predicate.row_tensor_eliminate 1.11% : 0.000002s : 8: predicate.same_eliminate 0.74% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.00% : 0.000002s : 8: predicate.shard_identity_eliminate 0.87% : 0.000001s : 8: predicate.special_op_eliminate 0.88% : 0.000001s : 8: predicate.specialize_transform 1.31% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.99% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.44% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.86% : 0.000001s : 10: predicate.switch_defer_inline 1.54% : 0.000002s : 18: predicate.switch_layer_defer_inline 3.51% : 0.000006s : 35: predicate.switch_simplify 0.80% : 0.000001s : 9: predicate.tile_eliminate 0.75% : 0.000001s : 9: predicate.transpose_eliminate 1.65% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.58% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.38% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.69% : 0.000006s : 25: predicate.tuple_list_get_item_eliminator 1.65% : 0.000003s : 17: predicate.tuple_list_get_set_item_eliminator 2.35% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.41% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 2.41% : 0.000004s : 26: predicate.updatestate_pure_node_eliminater 2.88% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.45% : 0.000001s : 4: predicate.value_based_eliminate 0.93% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.83% : 0.000001s : 8: predicate.virtual_output_eliminate 0.35% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.58% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000219 4 10.37% : 0.000023s : 1: func_graph_cloner_run.FuncGraphClonerGraph 89.63% : 0.000196s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.025794 192 0.01% : 0.000003s : 1: ForceFp32Comm 17.57% : 0.004531s : 1: add_attr 17.50% : 0.004514s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.23% : 0.000058s : 1: add_recomputation 0.02% : 0.000004s : 1: assign_add_opt 0.22% : 0.000057s : 1: auto_monad 0.09% : 0.000023s : 1: auto_monad_reorder 0.01% : 0.000003s : 1: begin_end_overlap_inline 0.02% : 0.000006s : 1: bias_add_comm_swap 2.64% : 0.000682s : 1: bootstrap 0.15% : 0.000038s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.06% : 0.000017s : 1: control_data_broadcast_order 0.05% : 0.000012s : 1: convert_after_rewriter 0.11% : 0.000028s : 1: cse_after_recomputation 0.02% : 0.000005s : 1: dataset_repeat_opt 0.02% : 0.000006s : 1: detach_backward 0.04% : 0.000010s : 1: environ_conv 0.07% : 0.000018s : 1: event_method 0.02% : 0.000005s : 1: full_micro_interleaved_order_control 0.02% : 0.000005s : 1: get_jit_bprop_graph 0.03% : 0.000008s : 1: graph_reusing 0.02% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000004s : 1: handle_group_info 0.02% : 0.000006s : 1: inline 0.02% : 0.000006s : 1: insert-virtual-dataset 0.01% : 0.000004s : 1: interleave_parallel_branches 0.02% : 0.000004s : 1: interleave_split_concat_branches 0.02% : 0.000006s : 1: label_fine_grained_interleaved_index 0.03% : 0.000007s : 1: label_micro_interleaved_index 1.92% : 0.000494s : 1: loop_unroll 0.02% : 0.000004s : 1: merge_cast_opt 0.02% : 0.000006s : 1: micro_interleaved_order_control 3.09% : 0.000797s : 1: mutable_eliminate 0.03% : 0.000007s : 1: offloading_packed_experts 0.06% : 0.000015s : 1: opt.transform.loop_unroll_optimizer 0.09% : 0.000023s : 1: opt.transform.mutable_eliminate 3.39% : 0.000875s : 78: opt.transform.opt_a 0.12% : 0.000032s : 1: opt.transform.opt_after_cconv 0.10% : 0.000026s : 1: opt.transform.opt_after_jit_grad 0.44% : 0.000114s : 28: opt.transform.opt_b 0.21% : 0.000054s : 2: opt.transform.opt_trans_graph 0.15% : 0.000037s : 4: opt.transform.symbol_engine_opt 9.64% : 0.002486s : 1: opt_a 0.46% : 0.000118s : 1: opt_after_cconv 1.99% : 0.000512s : 1: opt_after_jit_grad 0.94% : 0.000243s : 1: opt_b 19.16% : 0.004941s : 1: optimize 0.09% : 0.000023s : 1: optimize_parallel_all_gather_comm 0.03% : 0.000009s : 1: order_py_execute_after_rewriter 0.10% : 0.000025s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.03% : 0.000008s : 1: overlap_grad_ring_attention 0.02% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.03% : 0.000008s : 1: overlap_opt_shard_in_pipeline 0.02% : 0.000005s : 1: overlap_param_gather 0.02% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.03% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.02% : 0.000006s : 1: overlap_recompute_comm 0.03% : 0.000007s : 1: parallel-infer-symbol 0.01% : 0.000004s : 1: parallel-infer-symbol-second 0.02% : 0.000005s : 1: partial_unused_args_eliminate 0.02% : 0.000005s : 1: pipeline_parallel_scheduler 0.02% : 0.000004s : 1: pipeline_split 0.16% : 0.000041s : 1: pre_auto_parallel 0.10% : 0.000025s : 1: py_interpret_to_execute 0.08% : 0.000021s : 1: py_interpret_to_execute_after_opt_a 0.02% : 0.000005s : 1: remove_cast_before_assign_add 0.19% : 0.000050s : 1: remove_dup_value 1.62% : 0.000417s : 1: renormalize.infer 0.99% : 0.000255s : 1: renormalize.specialize 0.02% : 0.000005s : 1: reorder_send_recv_between_fp_bp 0.04% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.18% : 0.000045s : 1: rewriter_after_opt_a 0.21% : 0.000054s : 1: rewriter_before_opt_a 0.02% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.02% : 0.000005s : 1: slice_recompute_activation 0.02% : 0.000005s : 1: split_layernorm_comm 0.02% : 0.000005s : 1: split_matmul_comm_elemetwise 0.04% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.32% : 0.000081s : 1: symbol_engine_optimizer 0.33% : 0.000086s : 1: tuple_transform 14.25% : 0.003676s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_bf16[dim_type0-False-batch_size_list0-shape0-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_bf16[dim_type0-False-batch_size_list0-shape0-ms_dtype0],max_mem:516.0M [WARNING] ME(152945:281473835274032,MainProcess):2026-01-29-17:42:01.557.667 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.0275224, [21] [bootstrap]: 0.00058999 [type_inference]: 0.00394307 [event_method]: 1.112e-05 [auto_monad]: 5.373e-05 [graph_reusing]: 5.52001e-06 [inline]: 2.61e-06 [add_attr]: 0.0173076, [1] [add_attr_with_inline]: 0.0172915, [1] [Cycle 1]: 6.202e-05, [2] [tag_attr]: 1.602e-05 [meta_addattr_fg_expand]: 3.71999e-06 [parallel-infer-symbol]: 4.25999e-06 [pre_auto_parallel]: 3.351e-05 [insert-virtual-dataset]: 2.36998e-06 [parallel-infer-symbol-second]: 7.79983e-07 [dataset_repeat_opt]: 2.00002e-06 [pipeline_split]: 1.81e-06 [optimize]: 0.00480177, [53] [py_interpret_to_execute]: 1.992e-05 [rewriter_before_opt_a]: 5.155e-05 [opt_a]: 0.00244837, [2] [Cycle 1]: 0.00174589, [45] [expand_dump_flag]: 3.09001e-06 [switch_simplify]: 2.432e-05 [loop_unroll]: 1.238e-05 [a_1]: 0.00030851 [with_stream_mark]: 2.336e-05 [recompute_prepare]: 9.53002e-06 [updatestate_depend_eliminate]: 4.25e-06 [updatestate_assign_eliminate]: 3.55e-06 [updatestate_loads_eliminate]: 3.26001e-06 [parameter_eliminate]: 1.77001e-06 [a_2]: 8.833e-05 [accelerated_algorithm]: 7.43e-06 [shard]: 2.41e-06 [meta_shard_fg_expand]: 1.99999e-06 [shard_inline]: 7e-06 [merge_send_recv]: 9.99999e-06 [auto_parallel]: 7.51999e-06 [parallel]: 2.8e-05 [flash_sp]: 1.061e-05 [merge_comm]: 4.42e-06 [allreduce_fusion]: 3.7e-06 [matmul_add_comm_reduction]: 9.88002e-06 [allreduce_slice_to_reducescatter]: 8.50006e-07 [virtual_shard_identity]: 8.99e-06 [virtual_dataset]: 6.64001e-06 [get_grad_eliminate_]: 6.58998e-06 [virtual_output]: 7.05002e-06 [merge_forward]: 4.67e-06 [cell_reuse_recompute_pass]: 1.56002e-06 [offload_activation]: 1.083e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.378e-05 [merge_recompute_call_nodes]: 1.45999e-06 [before_grad]: 1.14e-05 [set_forward_comm_id_for_comm_node_pass]: 4.4e-06 [meta_fg_expand]: 3.25002e-06 [flash_sp_send_recv_attached]: 2.52001e-06 [receive_attached]: 2.69999e-06 [after_resolve]: 1.26e-05 [a_after_grad]: 1e-05 [renormalize]: 0.00067831 [add_forward_monad_depend]: 1.101e-05 [auto_monad_grad]: 2.22999e-06 [auto_monad_eliminator]: 1.581e-05 [cse]: 4.003e-05 [a_3]: 5.114e-05 [Cycle 2]: 0.00069094, [45] [expand_dump_flag]: 1.17e-06 [switch_simplify]: 7.78001e-06 [loop_unroll]: 7.35e-06 [a_1]: 0.00012932 [with_stream_mark]: 1.159e-05 [recompute_prepare]: 6.46999e-06 [updatestate_depend_eliminate]: 3.44001e-06 [updatestate_assign_eliminate]: 2.89001e-06 [updatestate_loads_eliminate]: 2.83e-06 [parameter_eliminate]: 1.16002e-06 [a_2]: 7.593e-05 [accelerated_algorithm]: 6.46e-06 [shard]: 1.33002e-06 [meta_shard_fg_expand]: 1.72999e-06 [shard_inline]: 6.84999e-06 [merge_send_recv]: 5.53002e-06 [auto_parallel]: 7.02002e-06 [parallel]: 5.34e-06 [flash_sp]: 3.55998e-06 [merge_comm]: 3.63999e-06 [allreduce_fusion]: 3.39001e-06 [matmul_add_comm_reduction]: 1.027e-05 [allreduce_slice_to_reducescatter]: 3.69997e-07 [virtual_shard_identity]: 7.58999e-06 [virtual_dataset]: 6.59001e-06 [get_grad_eliminate_]: 6.29001e-06 [virtual_output]: 6.23998e-06 [merge_forward]: 3.65998e-06 [cell_reuse_recompute_pass]: 1.62999e-06 [offload_activation]: 1.095e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.503e-05 [merge_recompute_call_nodes]: 9.80013e-07 [before_grad]: 1.044e-05 [set_forward_comm_id_for_comm_node_pass]: 3.91999e-06 [meta_fg_expand]: 2.24999e-06 [flash_sp_send_recv_attached]: 9.70002e-07 [receive_attached]: 1.74998e-06 [after_resolve]: 1.094e-05 [a_after_grad]: 9.55001e-06 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 1.51998e-06 [auto_monad_grad]: 1.07e-06 [auto_monad_eliminator]: 7.5e-06 [cse]: 2.378e-05 [a_3]: 3.952e-05 [py_interpret_to_execute_after_opt_a]: 1.26e-05 [slice_cell_reuse_recomputed_activation]: 2.26e-06 [rewriter_after_opt_a]: 4.664e-05 [convert_after_rewriter]: 8.08999e-06 [order_py_execute_after_rewriter]: 5.52001e-06 [mutable_eliminate]: 0.00065711 [opt_b]: 0.0002293, [1] [Cycle 1]: 0.00022185, [7] [b_1]: 0.00013272 [b_2]: 8.54998e-06 [updatestate_depend_eliminate]: 7.73001e-06 [updatestate_assign_eliminate]: 2.69999e-06 [updatestate_loads_eliminate]: 2.75002e-06 [renormalize]: 7.79983e-07 [cse]: 3.161e-05 [optimize_parallel_all_gather_comm]: 1.921e-05 [overlap_param_gather]: 2.19001e-06 [cconv]: 2.886e-05 [loop_unroll]: 0.00047639 [opt_after_cconv]: 0.00011932, [1] [Cycle 1]: 0.00011328, [7] [c_1]: 3.226e-05 [parameter_eliminate]: 3.72998e-06 [updatestate_depend_eliminate]: 6.81001e-06 [updatestate_assign_eliminate]: 2.89001e-06 [updatestate_loads_eliminate]: 2.49999e-06 [cse]: 2.378e-05 [renormalize]: 9.39996e-07 [remove_dup_value]: 4.164e-05 [tuple_transform]: 8.483e-05, [1] [Cycle 1]: 7.6e-05, [4] [d_1]: 4.782e-05 [none_parameter_eliminate]: 2.12001e-06 [renormalize]: 2.19996e-07 [switch_simplify]: 7.25e-06 [partial_unused_args_eliminate]: 2.29001e-06 [add_recomputation]: 5.044e-05 [cse_after_recomputation]: 2.824e-05, [1] [Cycle 1]: 2.379e-05, [1] [cse]: 1.813e-05 [environ_conv]: 6.43e-06 [swap_dp_allreduce_reducescatter]: 6.29001e-06 [bias_add_comm_swap]: 2.79999e-06 [label_micro_interleaved_index]: 4.99e-06 [label_fine_grained_interleaved_index]: 2.43e-06 [merge_cast_opt]: 1.46002e-06 [slice_recompute_activation]: 2.49999e-06 [micro_interleaved_order_control]: 2.45002e-06 [assign_add_opt]: 1.60999e-06 [ForceFp32Comm]: 8.39995e-07 [remove_cast_before_assign_add]: 1.32e-06 [full_micro_interleaved_order_control]: 2.02001e-06 [reorder_send_recv_between_fp_bp]: 2.79001e-06 [comm_op_add_attrs]: 1.21997e-06 [add_comm_op_reuse_tag]: 1.27e-06 [interleave_split_concat_branches]: 1.17e-06 [interleave_parallel_branches]: 1.06002e-06 [overlap_opt_shard_in_pipeline]: 1.45999e-06 [overlap_opt_shard_grad_in_pipeline]: 2.37001e-06 [control_data_broadcast_order]: 1.394e-05 [grouped_pairwise_exchange_alltoall]: 1.49e-06 [offloading_packed_experts]: 4.19002e-06 [overlap_recompute_and_grad_model_parallel]: 5.10999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.15001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.84e-06 [overlap_recompute_comm]: 2.41e-06 [overlap_grad_ring_attention]: 4.72e-06 [overlap_grad_flash_sp]: 2.19e-05 [begin_end_overlap_inline]: 4.89992e-07 [split_matmul_comm_elemetwise]: 2.14e-06 [split_layernorm_comm]: 1.87001e-06 [handle_group_info]: 9.39996e-07 [symbol_engine_optimizer]: 8.156e-05, [1] [Cycle 1]: 7.739e-05, [6] [build]: 3.12002e-06 [elim_shapecalc]: 1.061e-05 [elim_not_effective]: 1.435e-05 [opt_reshape]: 7.37002e-06 [fold_const_symbol]: 1.031e-05 [renormalize]: 2.29978e-07 [detach_backward]: 2.57001e-06 [pipeline_parallel_scheduler]: 1.43002e-06 [auto_monad_reorder]: 1.964e-05 [get_jit_bprop_graph]: 2.21e-06 [rewriter_after_jit_bprop_graph]: 5.92001e-06 [opt_after_jit_grad]: 0.00050676 [validate]: 4.169e-05 Sums bootstrap : 0.000590s : 6.46% type_inference : 0.003943s : 43.21% event_method : 0.000011s : 0.12% auto_monad : 0.000054s : 0.59% graph_reusing : 0.000006s : 0.06% inline : 0.000003s : 0.03% add_attr.add_attr_with_inline.tag_attr : 0.000016s : 0.18% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.04% parallel-infer-symbol : 0.000004s : 0.05% pre_auto_parallel : 0.000034s : 0.37% insert-virtual-dataset : 0.000002s : 0.03% parallel-infer-symbol-second : 0.000001s : 0.01% dataset_repeat_opt : 0.000002s : 0.02% pipeline_split : 0.000002s : 0.02% optimize.py_interpret_to_execute : 0.000020s : 0.22% optimize.rewriter_before_opt_a : 0.000052s : 0.56% optimize.opt_a.expand_dump_flag : 0.000004s : 0.05% optimize.opt_a.switch_simplify : 0.000032s : 0.35% optimize.opt_a.loop_unroll : 0.000020s : 0.22% optimize.opt_a.a_1 : 0.000438s : 4.80% optimize.opt_a.with_stream_mark : 0.000035s : 0.38% optimize.opt_a.recompute_prepare : 0.000016s : 0.18% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.08% optimize.opt_a.updatestate_assign_eliminate : 0.000006s : 0.07% optimize.opt_a.updatestate_loads_eliminate : 0.000006s : 0.07% optimize.opt_a.parameter_eliminate : 0.000003s : 0.03% optimize.opt_a.a_2 : 0.000164s : 1.80% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.15% optimize.opt_a.shard : 0.000004s : 0.04% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.04% optimize.opt_a.shard_inline : 0.000014s : 0.15% optimize.opt_a.merge_send_recv : 0.000016s : 0.17% optimize.opt_a.auto_parallel : 0.000015s : 0.16% optimize.opt_a.parallel : 0.000033s : 0.37% optimize.opt_a.flash_sp : 0.000014s : 0.16% optimize.opt_a.merge_comm : 0.000008s : 0.09% optimize.opt_a.allreduce_fusion : 0.000007s : 0.08% optimize.opt_a.matmul_add_comm_reduction : 0.000020s : 0.22% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000017s : 0.18% optimize.opt_a.virtual_dataset : 0.000013s : 0.14% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.14% optimize.opt_a.virtual_output : 0.000013s : 0.15% optimize.opt_a.merge_forward : 0.000008s : 0.09% optimize.opt_a.cell_reuse_recompute_pass : 0.000003s : 0.03% optimize.opt_a.offload_activation : 0.000022s : 0.24% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000029s : 0.32% optimize.opt_a.merge_recompute_call_nodes : 0.000002s : 0.03% optimize.opt_a.before_grad : 0.000022s : 0.24% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.09% optimize.opt_a.meta_fg_expand : 0.000006s : 0.06% optimize.opt_a.flash_sp_send_recv_attached : 0.000003s : 0.04% optimize.opt_a.receive_attached : 0.000004s : 0.05% optimize.opt_a.after_resolve : 0.000024s : 0.26% optimize.opt_a.a_after_grad : 0.000020s : 0.21% optimize.opt_a.renormalize : 0.000678s : 7.43% optimize.opt_a.add_forward_monad_depend : 0.000013s : 0.14% optimize.opt_a.auto_monad_grad : 0.000003s : 0.04% optimize.opt_a.auto_monad_eliminator : 0.000023s : 0.26% optimize.opt_a.cse : 0.000064s : 0.70% optimize.opt_a.a_3 : 0.000091s : 0.99% optimize.py_interpret_to_execute_after_opt_a : 0.000013s : 0.14% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.02% optimize.rewriter_after_opt_a : 0.000047s : 0.51% optimize.convert_after_rewriter : 0.000008s : 0.09% optimize.order_py_execute_after_rewriter : 0.000006s : 0.06% optimize.mutable_eliminate : 0.000657s : 7.20% optimize.opt_b.b_1 : 0.000133s : 1.45% optimize.opt_b.b_2 : 0.000009s : 0.09% optimize.opt_b.updatestate_depend_eliminate : 0.000008s : 0.08% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.03% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.03% optimize.opt_b.renormalize : 0.000001s : 0.01% optimize.opt_b.cse : 0.000032s : 0.35% optimize.optimize_parallel_all_gather_comm : 0.000019s : 0.21% optimize.overlap_param_gather : 0.000002s : 0.02% optimize.cconv : 0.000029s : 0.32% optimize.loop_unroll : 0.000476s : 5.22% optimize.opt_after_cconv.c_1 : 0.000032s : 0.35% optimize.opt_after_cconv.parameter_eliminate : 0.000004s : 0.04% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000007s : 0.07% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.03% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000002s : 0.03% optimize.opt_after_cconv.cse : 0.000024s : 0.26% optimize.opt_after_cconv.renormalize : 0.000001s : 0.01% optimize.remove_dup_value : 0.000042s : 0.46% optimize.tuple_transform.d_1 : 0.000048s : 0.52% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.02% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.08% optimize.partial_unused_args_eliminate : 0.000002s : 0.03% optimize.add_recomputation : 0.000050s : 0.55% optimize.cse_after_recomputation.cse : 0.000018s : 0.20% optimize.environ_conv : 0.000006s : 0.07% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.07% optimize.bias_add_comm_swap : 0.000003s : 0.03% optimize.label_micro_interleaved_index : 0.000005s : 0.05% optimize.label_fine_grained_interleaved_index : 0.000002s : 0.03% optimize.merge_cast_opt : 0.000001s : 0.02% optimize.slice_recompute_activation : 0.000002s : 0.03% optimize.micro_interleaved_order_control : 0.000002s : 0.03% optimize.assign_add_opt : 0.000002s : 0.02% optimize.ForceFp32Comm : 0.000001s : 0.01% optimize.remove_cast_before_assign_add : 0.000001s : 0.01% optimize.full_micro_interleaved_order_control : 0.000002s : 0.02% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.03% optimize.comm_op_add_attrs : 0.000001s : 0.01% optimize.add_comm_op_reuse_tag : 0.000001s : 0.01% optimize.interleave_split_concat_branches : 0.000001s : 0.01% optimize.interleave_parallel_branches : 0.000001s : 0.01% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.02% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.03% optimize.control_data_broadcast_order : 0.000014s : 0.15% optimize.grouped_pairwise_exchange_alltoall : 0.000001s : 0.02% optimize.offloading_packed_experts : 0.000004s : 0.05% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.06% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.01% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.02% optimize.overlap_recompute_comm : 0.000002s : 0.03% optimize.overlap_grad_ring_attention : 0.000005s : 0.05% optimize.overlap_grad_flash_sp : 0.000022s : 0.24% optimize.begin_end_overlap_inline : 0.000000s : 0.01% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.02% optimize.split_layernorm_comm : 0.000002s : 0.02% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000003s : 0.03% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000011s : 0.12% optimize.symbol_engine_optimizer.elim_not_effective : 0.000014s : 0.16% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.08% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.11% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.03% pipeline_parallel_scheduler : 0.000001s : 0.02% auto_monad_reorder : 0.000020s : 0.22% get_jit_bprop_graph : 0.000002s : 0.02% rewriter_after_jit_bprop_graph : 0.000006s : 0.06% opt_after_jit_grad : 0.000507s : 5.55% validate : 0.000042s : 0.46% Time group info: ------[substitution.] 0.000110 21 1.84% : 0.000002s : 2: substitution.elim_not_effective 1.31% : 0.000001s : 2: substitution.fold_const_symbol 5.72% : 0.000006s : 4: substitution.graph_param_transform 78.18% : 0.000086s : 1: substitution.inline 3.80% : 0.000004s : 4: substitution.j_node_and_user_rematch 4.72% : 0.000005s : 4: substitution.remove_not_recompute_node 4.43% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.003882 2 89.97% : 0.003492s : 1: type_inference.infer 10.03% : 0.000390s : 1: type_inference.specialize ------[replace.] 0.000018 1 100.00% : 0.000018s : 1: replace.inline ------[match.] 0.000085 1 100.00% : 0.000085s : 1: match.inline ------[predicate.] 0.000156 967 0.77% : 0.000001s : 9: predicate.accumulaten_eliminater 1.07% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.71% : 0.000001s : 8: predicate.addn_check_dump 0.74% : 0.000001s : 9: predicate.addn_zero_filter 0.68% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.15% : 0.000003s : 17: predicate.arithmetic_simplify 0.79% : 0.000001s : 9: predicate.cast_eliminate 1.06% : 0.000002s : 8: predicate.check_bprop_eliminate 0.88% : 0.000001s : 8: predicate.compare_switch_simplify 0.24% : 0.000000s : 4: predicate.const_output_eliminate 0.72% : 0.000001s : 8: predicate.depend_value_elim 0.79% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.92% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.77% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.07% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.33% : 0.000001s : 4: predicate.elim_not_effective 0.54% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.15% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.02% : 0.000002s : 13: predicate.environ_get_add_eliminate 1.02% : 0.000002s : 13: predicate.environ_get_depend_swap 1.81% : 0.000003s : 21: predicate.environ_get_eliminate 1.02% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.81% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.61% : 0.000003s : 10: predicate.float_depend_g_call 0.69% : 0.000001s : 8: predicate.float_environ_get_switch 1.01% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.24% : 0.000000s : 4: predicate.fold_const_symbol 0.90% : 0.000001s : 8: predicate.get_grad_eliminate 0.31% : 0.000000s : 4: predicate.graph_param_transform 0.76% : 0.000001s : 8: predicate.incorporate_call 0.63% : 0.000001s : 8: predicate.incorporate_call_switch 6.55% : 0.000010s : 43: predicate.inline 1.12% : 0.000002s : 8: predicate.inline_without_move 0.45% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.13% : 0.000002s : 8: predicate.less_batch_normalization 1.81% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.33% : 0.000004s : 26: predicate.load_eliminater 1.07% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.27% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.94% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.72% : 0.000001s : 8: predicate.merge_addn 0.77% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.75% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.74% : 0.000001s : 9: predicate.minmaximum_grad 1.30% : 0.000002s : 4: predicate.mutable_eliminate 0.45% : 0.000001s : 4: predicate.opt_reshape 0.51% : 0.000001s : 4: predicate.parallel_virtual_node 1.15% : 0.000002s : 10: predicate.partial_defer_inline 1.29% : 0.000002s : 13: predicate.partial_eliminate 0.70% : 0.000001s : 9: predicate.print_const_string_wrapper 0.77% : 0.000001s : 8: predicate.reduce_all_const_elim 1.15% : 0.000002s : 9: predicate.reduce_eliminate 2.15% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 1.07% : 0.000002s : 8: predicate.remove_not_recompute_node 1.34% : 0.000002s : 17: predicate.replace_applicator 0.71% : 0.000001s : 8: predicate.replace_old_param 0.53% : 0.000001s : 4: predicate.reset_defer_inline 1.00% : 0.000002s : 9: predicate.reshape_eliminate 0.89% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.45% : 0.000001s : 4: predicate.row_tensor_eliminate 1.11% : 0.000002s : 8: predicate.same_eliminate 0.68% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.11% : 0.000002s : 8: predicate.shard_identity_eliminate 0.93% : 0.000001s : 8: predicate.special_op_eliminate 0.90% : 0.000001s : 8: predicate.specialize_transform 1.42% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.88% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.45% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.88% : 0.000001s : 10: predicate.switch_defer_inline 1.60% : 0.000002s : 18: predicate.switch_layer_defer_inline 3.61% : 0.000006s : 35: predicate.switch_simplify 0.76% : 0.000001s : 9: predicate.tile_eliminate 0.85% : 0.000001s : 9: predicate.transpose_eliminate 1.52% : 0.000002s : 17: predicate.tuple_list_convert_item_index_to_positive 1.61% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.40% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.23% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.57% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.50% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.44% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 2.11% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.90% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.45% : 0.000001s : 4: predicate.value_based_eliminate 0.90% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.83% : 0.000001s : 8: predicate.virtual_output_eliminate 0.37% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.61% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000211 4 8.80% : 0.000019s : 1: func_graph_cloner_run.FuncGraphClonerGraph 91.20% : 0.000193s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.051303 192 0.01% : 0.000003s : 1: ForceFp32Comm 33.75% : 0.017314s : 1: add_attr 33.71% : 0.017296s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.11% : 0.000054s : 1: add_recomputation 0.01% : 0.000004s : 1: assign_add_opt 0.12% : 0.000059s : 1: auto_monad 0.05% : 0.000027s : 1: auto_monad_reorder 0.01% : 0.000003s : 1: begin_end_overlap_inline 0.01% : 0.000006s : 1: bias_add_comm_swap 1.24% : 0.000637s : 1: bootstrap 0.06% : 0.000032s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.03% : 0.000017s : 1: control_data_broadcast_order 0.02% : 0.000012s : 1: convert_after_rewriter 0.06% : 0.000031s : 1: cse_after_recomputation 0.01% : 0.000005s : 1: dataset_repeat_opt 0.01% : 0.000006s : 1: detach_backward 0.02% : 0.000010s : 1: environ_conv 0.04% : 0.000019s : 1: event_method 0.01% : 0.000005s : 1: full_micro_interleaved_order_control 0.01% : 0.000005s : 1: get_jit_bprop_graph 0.02% : 0.000009s : 1: graph_reusing 0.01% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000004s : 1: handle_group_info 0.01% : 0.000006s : 1: inline 0.01% : 0.000006s : 1: insert-virtual-dataset 0.01% : 0.000004s : 1: interleave_parallel_branches 0.01% : 0.000004s : 1: interleave_split_concat_branches 0.01% : 0.000005s : 1: label_fine_grained_interleaved_index 0.02% : 0.000008s : 1: label_micro_interleaved_index 0.95% : 0.000485s : 1: loop_unroll 0.01% : 0.000004s : 1: merge_cast_opt 0.01% : 0.000005s : 1: micro_interleaved_order_control 1.38% : 0.000709s : 1: mutable_eliminate 0.01% : 0.000007s : 1: offloading_packed_experts 0.03% : 0.000014s : 1: opt.transform.loop_unroll_optimizer 0.03% : 0.000016s : 1: opt.transform.mutable_eliminate 1.66% : 0.000849s : 78: opt.transform.opt_a 0.06% : 0.000031s : 1: opt.transform.opt_after_cconv 0.05% : 0.000025s : 1: opt.transform.opt_after_jit_grad 0.21% : 0.000110s : 28: opt.transform.opt_b 0.10% : 0.000053s : 2: opt.transform.opt_trans_graph 0.08% : 0.000039s : 4: opt.transform.symbol_engine_opt 4.78% : 0.002452s : 1: opt_a 0.24% : 0.000123s : 1: opt_after_cconv 1.01% : 0.000517s : 1: opt_after_jit_grad 0.45% : 0.000233s : 1: opt_b 9.37% : 0.004807s : 1: optimize 0.04% : 0.000023s : 1: optimize_parallel_all_gather_comm 0.02% : 0.000008s : 1: order_py_execute_after_rewriter 0.05% : 0.000025s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.02% : 0.000011s : 1: overlap_grad_ring_attention 0.01% : 0.000008s : 1: overlap_opt_shard_grad_in_pipeline 0.01% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.01% : 0.000005s : 1: overlap_param_gather 0.01% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.02% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.01% : 0.000005s : 1: overlap_recompute_comm 0.02% : 0.000008s : 1: parallel-infer-symbol 0.01% : 0.000004s : 1: parallel-infer-symbol-second 0.01% : 0.000005s : 1: partial_unused_args_eliminate 0.01% : 0.000005s : 1: pipeline_parallel_scheduler 0.01% : 0.000005s : 1: pipeline_split 0.07% : 0.000038s : 1: pre_auto_parallel 0.05% : 0.000024s : 1: py_interpret_to_execute 0.03% : 0.000016s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000007s : 1: remove_cast_before_assign_add 0.09% : 0.000046s : 1: remove_dup_value 0.81% : 0.000414s : 1: renormalize.infer 0.50% : 0.000255s : 1: renormalize.specialize 0.01% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.02% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.10% : 0.000051s : 1: rewriter_after_opt_a 0.11% : 0.000056s : 1: rewriter_before_opt_a 0.01% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.01% : 0.000005s : 1: slice_recompute_activation 0.01% : 0.000005s : 1: split_layernorm_comm 0.01% : 0.000005s : 1: split_matmul_comm_elemetwise 0.02% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.16% : 0.000084s : 1: symbol_engine_optimizer 0.17% : 0.000088s : 1: tuple_transform 7.73% : 0.003964s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_bf16[dim_type0-False-batch_size_list1-shape0-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_bf16[dim_type0-False-batch_size_list1-shape0-ms_dtype0],max_mem:196.0M [WARNING] ME(152945:281473835274032,MainProcess):2026-01-29-17:42:39.686.275 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.0594846, [21] [bootstrap]: 0.0399234 [type_inference]: 0.00367315 [event_method]: 1.178e-05 [auto_monad]: 5.684e-05 [graph_reusing]: 5.50001e-06 [inline]: 2.74001e-06 [add_attr]: 0.00586713, [1] [add_attr_with_inline]: 0.0058536, [1] [Cycle 1]: 6.309e-05, [2] [tag_attr]: 1.708e-05 [meta_addattr_fg_expand]: 3.58e-06 [parallel-infer-symbol]: 4.13999e-06 [pre_auto_parallel]: 3.169e-05 [insert-virtual-dataset]: 2.54001e-06 [parallel-infer-symbol-second]: 9.89996e-07 [dataset_repeat_opt]: 1.86003e-06 [pipeline_split]: 1.67001e-06 [optimize]: 0.00894246, [53] [py_interpret_to_execute]: 2.062e-05 [rewriter_before_opt_a]: 4.537e-05 [opt_a]: 0.00252278, [2] [Cycle 1]: 0.00174345, [45] [expand_dump_flag]: 2.74999e-06 [switch_simplify]: 2.352e-05 [loop_unroll]: 1.383e-05 [a_1]: 0.00030148 [with_stream_mark]: 2.021e-05 [recompute_prepare]: 9.39e-06 [updatestate_depend_eliminate]: 4.12e-06 [updatestate_assign_eliminate]: 3.85998e-06 [updatestate_loads_eliminate]: 3.48e-06 [parameter_eliminate]: 1.83997e-06 [a_2]: 8.843e-05 [accelerated_algorithm]: 7.73001e-06 [shard]: 2.09e-06 [meta_shard_fg_expand]: 1.74e-06 [shard_inline]: 6.83998e-06 [merge_send_recv]: 8.62e-06 [auto_parallel]: 7.43e-06 [parallel]: 2.742e-05 [flash_sp]: 9.85002e-06 [merge_comm]: 4.46002e-06 [allreduce_fusion]: 3.78999e-06 [matmul_add_comm_reduction]: 1.056e-05 [allreduce_slice_to_reducescatter]: 6.89994e-07 [virtual_shard_identity]: 9.56998e-06 [virtual_dataset]: 7.29001e-06 [get_grad_eliminate_]: 6.53e-06 [virtual_output]: 7.22002e-06 [merge_forward]: 4.06001e-06 [cell_reuse_recompute_pass]: 1.69998e-06 [offload_activation]: 9.92001e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.509e-05 [merge_recompute_call_nodes]: 1.47999e-06 [before_grad]: 1.195e-05 [set_forward_comm_id_for_comm_node_pass]: 3.97e-06 [meta_fg_expand]: 3.18e-06 [flash_sp_send_recv_attached]: 2.79999e-06 [receive_attached]: 2.55002e-06 [after_resolve]: 1.376e-05 [a_after_grad]: 1.044e-05 [renormalize]: 0.00067949 [add_forward_monad_depend]: 1.231e-05 [auto_monad_grad]: 2.37999e-06 [auto_monad_eliminator]: 1.539e-05 [cse]: 4.22e-05 [a_3]: 5.381e-05 [Cycle 2]: 0.00076704, [45] [expand_dump_flag]: 2.24001e-06 [switch_simplify]: 9.00001e-06 [loop_unroll]: 6.66e-06 [a_1]: 0.00013318 [with_stream_mark]: 1.593e-05 [recompute_prepare]: 7.92e-06 [updatestate_depend_eliminate]: 4.03999e-06 [updatestate_assign_eliminate]: 2.91999e-06 [updatestate_loads_eliminate]: 2.71e-06 [parameter_eliminate]: 1.87999e-06 [a_2]: 7.955e-05 [accelerated_algorithm]: 7.43e-06 [shard]: 1.86e-06 [meta_shard_fg_expand]: 1.76e-06 [shard_inline]: 7.30998e-06 [merge_send_recv]: 7.87e-06 [auto_parallel]: 7.93001e-06 [parallel]: 7.4e-06 [flash_sp]: 3.7e-06 [merge_comm]: 3.79002e-06 [allreduce_fusion]: 3.70003e-06 [matmul_add_comm_reduction]: 1.405e-05 [allreduce_slice_to_reducescatter]: 5.49975e-07 [virtual_shard_identity]: 9.28002e-06 [virtual_dataset]: 6.74001e-06 [get_grad_eliminate_]: 6.25002e-06 [virtual_output]: 6.33e-06 [merge_forward]: 4.23999e-06 [cell_reuse_recompute_pass]: 1.81e-06 [offload_activation]: 1.073e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.68e-05 [merge_recompute_call_nodes]: 1.40001e-06 [before_grad]: 1.182e-05 [set_forward_comm_id_for_comm_node_pass]: 5.10001e-06 [meta_fg_expand]: 2.45002e-06 [flash_sp_send_recv_attached]: 1.40001e-06 [receive_attached]: 1.69998e-06 [after_resolve]: 1.209e-05 [a_after_grad]: 1.007e-05 [renormalize]: 9.00181e-08 [add_forward_monad_depend]: 3.22002e-06 [auto_monad_grad]: 1.74e-06 [auto_monad_eliminator]: 9.23997e-06 [cse]: 3.115e-05 [a_3]: 4.134e-05 [py_interpret_to_execute_after_opt_a]: 1.379e-05 [slice_cell_reuse_recomputed_activation]: 1.82999e-06 [rewriter_after_opt_a]: 4.627e-05 [convert_after_rewriter]: 8.80001e-06 [order_py_execute_after_rewriter]: 5.59998e-06 [mutable_eliminate]: 0.00441721 [opt_b]: 0.00029164, [1] [Cycle 1]: 0.00028008, [7] [b_1]: 0.00015158 [b_2]: 1.017e-05 [updatestate_depend_eliminate]: 1.178e-05 [updatestate_assign_eliminate]: 3.61999e-06 [updatestate_loads_eliminate]: 3.16001e-06 [renormalize]: 9.10019e-07 [cse]: 5.834e-05 [optimize_parallel_all_gather_comm]: 2.486e-05 [overlap_param_gather]: 1.81998e-06 [cconv]: 4.116e-05 [loop_unroll]: 0.00067493 [opt_after_cconv]: 0.00013422, [1] [Cycle 1]: 0.00012691, [7] [c_1]: 3.485e-05 [parameter_eliminate]: 5.86e-06 [updatestate_depend_eliminate]: 8.2e-06 [updatestate_assign_eliminate]: 2.81999e-06 [updatestate_loads_eliminate]: 3.06001e-06 [cse]: 3.016e-05 [renormalize]: 7.2e-07 [remove_dup_value]: 4.771e-05 [tuple_transform]: 9.439e-05, [1] [Cycle 1]: 8.496e-05, [4] [d_1]: 5.464e-05 [none_parameter_eliminate]: 2.56998e-06 [renormalize]: 1.80007e-07 [switch_simplify]: 8.41002e-06 [partial_unused_args_eliminate]: 2.14e-06 [add_recomputation]: 5.829e-05 [cse_after_recomputation]: 3.147e-05, [1] [Cycle 1]: 2.668e-05, [1] [cse]: 2.044e-05 [environ_conv]: 6.88e-06 [swap_dp_allreduce_reducescatter]: 5.22999e-06 [bias_add_comm_swap]: 3.86999e-06 [label_micro_interleaved_index]: 5.10001e-06 [label_fine_grained_interleaved_index]: 2.64999e-06 [merge_cast_opt]: 1.44e-06 [slice_recompute_activation]: 2.34999e-06 [micro_interleaved_order_control]: 2.57001e-06 [assign_add_opt]: 1.32999e-06 [ForceFp32Comm]: 1.04998e-06 [remove_cast_before_assign_add]: 1.04e-06 [full_micro_interleaved_order_control]: 2.11e-06 [reorder_send_recv_between_fp_bp]: 2.72001e-06 [comm_op_add_attrs]: 1.13001e-06 [add_comm_op_reuse_tag]: 9.50007e-07 [interleave_split_concat_branches]: 1.22e-06 [interleave_parallel_branches]: 1.01997e-06 [overlap_opt_shard_in_pipeline]: 1.18001e-06 [overlap_opt_shard_grad_in_pipeline]: 2.06e-06 [control_data_broadcast_order]: 1.472e-05 [grouped_pairwise_exchange_alltoall]: 1.55001e-06 [offloading_packed_experts]: 4.07998e-06 [overlap_recompute_and_grad_model_parallel]: 4.90999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.37999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.60001e-06 [overlap_recompute_comm]: 2.22001e-06 [overlap_grad_ring_attention]: 4.61002e-06 [overlap_grad_flash_sp]: 2.099e-05 [begin_end_overlap_inline]: 5.19998e-07 [split_matmul_comm_elemetwise]: 2.11e-06 [split_layernorm_comm]: 2.02001e-06 [handle_group_info]: 1.05999e-06 [symbol_engine_optimizer]: 8.797e-05, [1] [Cycle 1]: 8.352e-05, [6] [build]: 3.08e-06 [elim_shapecalc]: 1.274e-05 [elim_not_effective]: 1.444e-05 [opt_reshape]: 9.81e-06 [fold_const_symbol]: 1.066e-05 [renormalize]: 2.30008e-07 [detach_backward]: 2.29001e-06 [pipeline_parallel_scheduler]: 1.60001e-06 [auto_monad_reorder]: 2.096e-05 [get_jit_bprop_graph]: 2.22999e-06 [rewriter_after_jit_bprop_graph]: 6.07999e-06 [opt_after_jit_grad]: 0.0006787 [validate]: 5.068e-05 Sums bootstrap : 0.039923s : 76.06% type_inference : 0.003673s : 7.00% event_method : 0.000012s : 0.02% auto_monad : 0.000057s : 0.11% graph_reusing : 0.000006s : 0.01% inline : 0.000003s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000017s : 0.03% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.01% parallel-infer-symbol : 0.000004s : 0.01% pre_auto_parallel : 0.000032s : 0.06% 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 : 0.000021s : 0.04% optimize.rewriter_before_opt_a : 0.000045s : 0.09% optimize.opt_a.expand_dump_flag : 0.000005s : 0.01% optimize.opt_a.switch_simplify : 0.000033s : 0.06% optimize.opt_a.loop_unroll : 0.000020s : 0.04% optimize.opt_a.a_1 : 0.000435s : 0.83% optimize.opt_a.with_stream_mark : 0.000036s : 0.07% optimize.opt_a.recompute_prepare : 0.000017s : 0.03% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.02% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000006s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.01% optimize.opt_a.a_2 : 0.000168s : 0.32% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.03% optimize.opt_a.shard : 0.000004s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000003s : 0.01% optimize.opt_a.shard_inline : 0.000014s : 0.03% optimize.opt_a.merge_send_recv : 0.000016s : 0.03% optimize.opt_a.auto_parallel : 0.000015s : 0.03% optimize.opt_a.parallel : 0.000035s : 0.07% optimize.opt_a.flash_sp : 0.000014s : 0.03% optimize.opt_a.merge_comm : 0.000008s : 0.02% optimize.opt_a.allreduce_fusion : 0.000007s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000025s : 0.05% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000019s : 0.04% optimize.opt_a.virtual_dataset : 0.000014s : 0.03% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.02% optimize.opt_a.virtual_output : 0.000014s : 0.03% optimize.opt_a.merge_forward : 0.000008s : 0.02% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.01% optimize.opt_a.offload_activation : 0.000021s : 0.04% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000032s : 0.06% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.01% optimize.opt_a.before_grad : 0.000024s : 0.05% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000009s : 0.02% optimize.opt_a.meta_fg_expand : 0.000006s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.01% optimize.opt_a.receive_attached : 0.000004s : 0.01% optimize.opt_a.after_resolve : 0.000026s : 0.05% optimize.opt_a.a_after_grad : 0.000021s : 0.04% optimize.opt_a.renormalize : 0.000680s : 1.29% optimize.opt_a.add_forward_monad_depend : 0.000016s : 0.03% optimize.opt_a.auto_monad_grad : 0.000004s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000025s : 0.05% optimize.opt_a.cse : 0.000073s : 0.14% optimize.opt_a.a_3 : 0.000095s : 0.18% optimize.py_interpret_to_execute_after_opt_a : 0.000014s : 0.03% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000046s : 0.09% optimize.convert_after_rewriter : 0.000009s : 0.02% optimize.order_py_execute_after_rewriter : 0.000006s : 0.01% optimize.mutable_eliminate : 0.004417s : 8.42% optimize.opt_b.b_1 : 0.000152s : 0.29% optimize.opt_b.b_2 : 0.000010s : 0.02% optimize.opt_b.updatestate_depend_eliminate : 0.000012s : 0.02% optimize.opt_b.updatestate_assign_eliminate : 0.000004s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.01% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000058s : 0.11% optimize.optimize_parallel_all_gather_comm : 0.000025s : 0.05% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000041s : 0.08% optimize.loop_unroll : 0.000675s : 1.29% optimize.opt_after_cconv.c_1 : 0.000035s : 0.07% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.01% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000008s : 0.02% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000003s : 0.01% optimize.opt_after_cconv.cse : 0.000030s : 0.06% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000048s : 0.09% optimize.tuple_transform.d_1 : 0.000055s : 0.10% optimize.tuple_transform.none_parameter_eliminate : 0.000003s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000008s : 0.02% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000058s : 0.11% optimize.cse_after_recomputation.cse : 0.000020s : 0.04% optimize.environ_conv : 0.000007s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000005s : 0.01% optimize.bias_add_comm_swap : 0.000004s : 0.01% optimize.label_micro_interleaved_index : 0.000005s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.01% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.01% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000015s : 0.03% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000004s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.01% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000005s : 0.01% optimize.overlap_grad_flash_sp : 0.000021s : 0.04% 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 : 0.000003s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000013s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000014s : 0.03% optimize.symbol_engine_optimizer.opt_reshape : 0.000010s : 0.02% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000011s : 0.02% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000021s : 0.04% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.01% opt_after_jit_grad : 0.000679s : 1.29% validate : 0.000051s : 0.10% Time group info: ------[substitution.] 0.000111 21 2.08% : 0.000002s : 2: substitution.elim_not_effective 1.11% : 0.000001s : 2: substitution.fold_const_symbol 6.14% : 0.000007s : 4: substitution.graph_param_transform 76.83% : 0.000085s : 1: substitution.inline 3.84% : 0.000004s : 4: substitution.j_node_and_user_rematch 5.58% : 0.000006s : 4: substitution.remove_not_recompute_node 4.41% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.003613 2 89.28% : 0.003225s : 1: type_inference.infer 10.72% : 0.000387s : 1: type_inference.specialize ------[replace.] 0.000017 1 100.00% : 0.000017s : 1: replace.inline ------[match.] 0.000084 1 100.00% : 0.000084s : 1: match.inline ------[predicate.] 0.000165 967 0.75% : 0.000001s : 9: predicate.accumulaten_eliminater 1.21% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.66% : 0.000001s : 8: predicate.addn_check_dump 0.69% : 0.000001s : 9: predicate.addn_zero_filter 0.63% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 1.96% : 0.000003s : 17: predicate.arithmetic_simplify 0.75% : 0.000001s : 9: predicate.cast_eliminate 0.80% : 0.000001s : 8: predicate.check_bprop_eliminate 0.61% : 0.000001s : 8: predicate.compare_switch_simplify 0.26% : 0.000000s : 4: predicate.const_output_eliminate 0.62% : 0.000001s : 8: predicate.depend_value_elim 0.76% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.83% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.73% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.54% : 0.000003s : 8: predicate.dumpgradient_eliminate 0.35% : 0.000001s : 4: predicate.elim_not_effective 0.67% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.18% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.06% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.95% : 0.000002s : 13: predicate.environ_get_depend_swap 1.75% : 0.000003s : 21: predicate.environ_get_eliminate 1.03% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.78% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.48% : 0.000002s : 10: predicate.float_depend_g_call 0.58% : 0.000001s : 8: predicate.float_environ_get_switch 1.00% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.26% : 0.000000s : 4: predicate.fold_const_symbol 0.74% : 0.000001s : 8: predicate.get_grad_eliminate 0.36% : 0.000001s : 4: predicate.graph_param_transform 0.72% : 0.000001s : 8: predicate.incorporate_call 0.61% : 0.000001s : 8: predicate.incorporate_call_switch 6.66% : 0.000011s : 43: predicate.inline 0.95% : 0.000002s : 8: predicate.inline_without_move 0.41% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.00% : 0.000002s : 8: predicate.less_batch_normalization 1.75% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.15% : 0.000004s : 26: predicate.load_eliminater 1.21% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.25% : 0.000002s : 13: predicate.loop_unroll_before_grad 2.01% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.69% : 0.000001s : 8: predicate.merge_addn 0.69% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.79% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.67% : 0.000001s : 9: predicate.minmaximum_grad 3.02% : 0.000005s : 4: predicate.mutable_eliminate 0.67% : 0.000001s : 4: predicate.opt_reshape 0.52% : 0.000001s : 4: predicate.parallel_virtual_node 1.07% : 0.000002s : 10: predicate.partial_defer_inline 1.21% : 0.000002s : 13: predicate.partial_eliminate 0.96% : 0.000002s : 9: predicate.print_const_string_wrapper 0.62% : 0.000001s : 8: predicate.reduce_all_const_elim 0.94% : 0.000002s : 9: predicate.reduce_eliminate 2.12% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.94% : 0.000002s : 8: predicate.remove_not_recompute_node 1.19% : 0.000002s : 17: predicate.replace_applicator 0.65% : 0.000001s : 8: predicate.replace_old_param 0.60% : 0.000001s : 4: predicate.reset_defer_inline 0.80% : 0.000001s : 9: predicate.reshape_eliminate 0.70% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.84% : 0.000001s : 4: predicate.row_tensor_eliminate 1.18% : 0.000002s : 8: predicate.same_eliminate 0.56% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.87% : 0.000001s : 8: predicate.shard_identity_eliminate 0.97% : 0.000002s : 8: predicate.special_op_eliminate 1.00% : 0.000002s : 8: predicate.specialize_transform 1.37% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.84% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.49% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.86% : 0.000001s : 10: predicate.switch_defer_inline 1.52% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.77% : 0.000006s : 35: predicate.switch_simplify 0.72% : 0.000001s : 9: predicate.tile_eliminate 0.76% : 0.000001s : 9: predicate.transpose_eliminate 1.54% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.51% : 0.000002s : 17: predicate.tuple_list_get_item_const_eliminator 1.44% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.60% : 0.000006s : 25: predicate.tuple_list_get_item_eliminator 1.46% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.53% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.57% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.97% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.82% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.53% : 0.000001s : 4: predicate.value_based_eliminate 0.89% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.80% : 0.000001s : 8: predicate.virtual_output_eliminate 0.32% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.71% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000209 4 12.31% : 0.000026s : 1: func_graph_cloner_run.FuncGraphClonerGraph 87.69% : 0.000183s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.076032 192 0.00% : 0.000004s : 1: ForceFp32Comm 7.73% : 0.005873s : 1: add_attr 7.70% : 0.005858s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.08% : 0.000063s : 1: add_recomputation 0.01% : 0.000004s : 1: assign_add_opt 0.08% : 0.000063s : 1: auto_monad 0.04% : 0.000029s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.01% : 0.000007s : 1: bias_add_comm_swap 52.58% : 0.039977s : 1: bootstrap 0.06% : 0.000045s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.02% : 0.000018s : 1: control_data_broadcast_order 0.02% : 0.000013s : 1: convert_after_rewriter 0.05% : 0.000035s : 1: cse_after_recomputation 0.01% : 0.000005s : 1: dataset_repeat_opt 0.01% : 0.000005s : 1: detach_backward 0.01% : 0.000010s : 1: environ_conv 0.02% : 0.000018s : 1: event_method 0.01% : 0.000005s : 1: full_micro_interleaved_order_control 0.01% : 0.000005s : 1: get_jit_bprop_graph 0.01% : 0.000010s : 1: graph_reusing 0.01% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.01% : 0.000006s : 1: inline 0.01% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.01% : 0.000004s : 1: interleave_split_concat_branches 0.01% : 0.000006s : 1: label_fine_grained_interleaved_index 0.01% : 0.000008s : 1: label_micro_interleaved_index 0.90% : 0.000687s : 1: loop_unroll 0.01% : 0.000004s : 1: merge_cast_opt 0.01% : 0.000005s : 1: micro_interleaved_order_control 5.85% : 0.004445s : 1: mutable_eliminate 0.01% : 0.000007s : 1: offloading_packed_experts 0.02% : 0.000018s : 1: opt.transform.loop_unroll_optimizer 0.05% : 0.000038s : 1: opt.transform.mutable_eliminate 1.14% : 0.000865s : 78: opt.transform.opt_a 0.04% : 0.000033s : 1: opt.transform.opt_after_cconv 0.04% : 0.000031s : 1: opt.transform.opt_after_jit_grad 0.16% : 0.000124s : 28: opt.transform.opt_b 0.08% : 0.000060s : 2: opt.transform.opt_trans_graph 0.06% : 0.000043s : 4: opt.transform.symbol_engine_opt 3.32% : 0.002526s : 1: opt_a 0.18% : 0.000138s : 1: opt_after_cconv 0.91% : 0.000692s : 1: opt_after_jit_grad 0.39% : 0.000296s : 1: opt_b 11.77% : 0.008948s : 1: optimize 0.04% : 0.000028s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000009s : 1: order_py_execute_after_rewriter 0.03% : 0.000024s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.02% : 0.000013s : 1: overlap_grad_ring_attention 0.01% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.01% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.01% : 0.000005s : 1: overlap_param_gather 0.01% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.01% : 0.000005s : 1: overlap_recompute_comm 0.01% : 0.000008s : 1: parallel-infer-symbol 0.01% : 0.000004s : 1: parallel-infer-symbol-second 0.01% : 0.000005s : 1: partial_unused_args_eliminate 0.01% : 0.000005s : 1: pipeline_parallel_scheduler 0.01% : 0.000004s : 1: pipeline_split 0.05% : 0.000037s : 1: pre_auto_parallel 0.03% : 0.000025s : 1: py_interpret_to_execute 0.02% : 0.000017s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000008s : 1: remove_cast_before_assign_add 0.07% : 0.000052s : 1: remove_dup_value 0.48% : 0.000368s : 1: renormalize.infer 0.40% : 0.000303s : 1: renormalize.specialize 0.01% : 0.000005s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.07% : 0.000052s : 1: rewriter_after_opt_a 0.07% : 0.000050s : 1: rewriter_before_opt_a 0.01% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.01% : 0.000005s : 1: slice_recompute_activation 0.01% : 0.000005s : 1: split_layernorm_comm 0.01% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000008s : 1: swap_dp_allreduce_reducescatter 0.12% : 0.000091s : 1: symbol_engine_optimizer 0.13% : 0.000097s : 1: tuple_transform 4.86% : 0.003695s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_bf16[dim_type0-False-batch_size_list2-shape0-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_bf16[dim_type0-False-batch_size_list2-shape0-ms_dtype0],max_mem:516.0M [WARNING] ME(152945:281473835274032,MainProcess):2026-01-29-17:44:00.564.281 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.0154459, [21] [bootstrap]: 0.00058993 [type_inference]: 0.00409056 [event_method]: 1.333e-05 [auto_monad]: 5.491e-05 [graph_reusing]: 5.04e-06 [inline]: 2.91e-06 [add_attr]: 0.00518584, [1] [add_attr_with_inline]: 0.0051712, [1] [Cycle 1]: 6.085e-05, [2] [tag_attr]: 1.599e-05 [meta_addattr_fg_expand]: 3.24001e-06 [parallel-infer-symbol]: 4.12e-06 [pre_auto_parallel]: 3.33e-05 [insert-virtual-dataset]: 2.58e-06 [parallel-infer-symbol-second]: 8.89995e-07 [dataset_repeat_opt]: 1.91003e-06 [pipeline_split]: 1.86e-06 [optimize]: 0.00473453, [53] [py_interpret_to_execute]: 2.121e-05 [rewriter_before_opt_a]: 4.718e-05 [opt_a]: 0.00257428, [2] [Cycle 1]: 0.00180628, [45] [expand_dump_flag]: 2.74001e-06 [switch_simplify]: 2.489e-05 [loop_unroll]: 1.293e-05 [a_1]: 0.00035147 [with_stream_mark]: 2.267e-05 [recompute_prepare]: 9.76e-06 [updatestate_depend_eliminate]: 4.11001e-06 [updatestate_assign_eliminate]: 3.75e-06 [updatestate_loads_eliminate]: 3.64002e-06 [parameter_eliminate]: 1.92001e-06 [a_2]: 9.178e-05 [accelerated_algorithm]: 7.85e-06 [shard]: 2.63998e-06 [meta_shard_fg_expand]: 1.76998e-06 [shard_inline]: 7.35e-06 [merge_send_recv]: 8.52e-06 [auto_parallel]: 7.59002e-06 [parallel]: 3.177e-05 [flash_sp]: 1.006e-05 [merge_comm]: 4.45999e-06 [allreduce_fusion]: 4.10998e-06 [matmul_add_comm_reduction]: 1.036e-05 [allreduce_slice_to_reducescatter]: 7.39994e-07 [virtual_shard_identity]: 9.87999e-06 [virtual_dataset]: 8.05e-06 [get_grad_eliminate_]: 6.61e-06 [virtual_output]: 6.83e-06 [merge_forward]: 4.25e-06 [cell_reuse_recompute_pass]: 1.50999e-06 [offload_activation]: 1.068e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.467e-05 [merge_recompute_call_nodes]: 1.86998e-06 [before_grad]: 1.143e-05 [set_forward_comm_id_for_comm_node_pass]: 4.25e-06 [meta_fg_expand]: 2.99001e-06 [flash_sp_send_recv_attached]: 2.37001e-06 [receive_attached]: 2.88e-06 [after_resolve]: 1.175e-05 [a_after_grad]: 1.08e-05 [renormalize]: 0.00068377 [add_forward_monad_depend]: 9.44e-06 [auto_monad_grad]: 2.81e-06 [auto_monad_eliminator]: 1.658e-05 [cse]: 3.638e-05 [a_3]: 5.421e-05 [Cycle 2]: 0.00075687, [45] [expand_dump_flag]: 1.14998e-06 [switch_simplify]: 9.26002e-06 [loop_unroll]: 7.25e-06 [a_1]: 0.00014123 [with_stream_mark]: 1.337e-05 [recompute_prepare]: 6.34001e-06 [updatestate_depend_eliminate]: 3.51001e-06 [updatestate_assign_eliminate]: 2.74999e-06 [updatestate_loads_eliminate]: 2.98998e-06 [parameter_eliminate]: 1.27e-06 [a_2]: 7.688e-05 [accelerated_algorithm]: 2.254e-05 [shard]: 1.34e-06 [meta_shard_fg_expand]: 1.87999e-06 [shard_inline]: 7.51001e-06 [merge_send_recv]: 6.21e-06 [auto_parallel]: 7.20998e-06 [parallel]: 6.94999e-06 [flash_sp]: 4.03999e-06 [merge_comm]: 4.08999e-06 [allreduce_fusion]: 3.81001e-06 [matmul_add_comm_reduction]: 1.197e-05 [allreduce_slice_to_reducescatter]: 3.59985e-07 [virtual_shard_identity]: 7.35003e-06 [virtual_dataset]: 6.66e-06 [get_grad_eliminate_]: 6.44001e-06 [virtual_output]: 6.12999e-06 [merge_forward]: 3.72998e-06 [cell_reuse_recompute_pass]: 1.99e-06 [offload_activation]: 1.126e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.466e-05 [merge_recompute_call_nodes]: 1.04998e-06 [before_grad]: 1.024e-05 [set_forward_comm_id_for_comm_node_pass]: 4.48999e-06 [meta_fg_expand]: 2.58003e-06 [flash_sp_send_recv_attached]: 9.89996e-07 [receive_attached]: 1.28002e-06 [after_resolve]: 1.11e-05 [a_after_grad]: 9.39e-06 [renormalize]: 1.00001e-07 [add_forward_monad_depend]: 1.93002e-06 [auto_monad_grad]: 1.67999e-06 [auto_monad_eliminator]: 7.65e-06 [cse]: 2.547e-05 [a_3]: 4.054e-05 [py_interpret_to_execute_after_opt_a]: 1.277e-05 [slice_cell_reuse_recomputed_activation]: 1.98002e-06 [rewriter_after_opt_a]: 4.307e-05 [convert_after_rewriter]: 7.23999e-06 [order_py_execute_after_rewriter]: 5.52999e-06 [mutable_eliminate]: 0.00060749 [opt_b]: 0.00022418, [1] [Cycle 1]: 0.00021724, [7] [b_1]: 0.0001305 [b_2]: 8.55001e-06 [updatestate_depend_eliminate]: 8.50001e-06 [updatestate_assign_eliminate]: 2.99001e-06 [updatestate_loads_eliminate]: 2.71999e-06 [renormalize]: 6.10016e-07 [cse]: 2.766e-05 [optimize_parallel_all_gather_comm]: 1.801e-05 [overlap_param_gather]: 1.81e-06 [cconv]: 2.772e-05 [loop_unroll]: 0.00044079 [opt_after_cconv]: 0.00011435, [1] [Cycle 1]: 0.0001087, [7] [c_1]: 3.044e-05 [parameter_eliminate]: 3.58e-06 [updatestate_depend_eliminate]: 5.92001e-06 [updatestate_assign_eliminate]: 2.66e-06 [updatestate_loads_eliminate]: 2.44999e-06 [cse]: 2.282e-05 [renormalize]: 3.89991e-07 [remove_dup_value]: 1.603e-05 [tuple_transform]: 7.753e-05, [1] [Cycle 1]: 7.311e-05, [4] [d_1]: 4.546e-05 [none_parameter_eliminate]: 1.76e-06 [renormalize]: 2.60014e-07 [switch_simplify]: 7.08e-06 [partial_unused_args_eliminate]: 1.67999e-06 [add_recomputation]: 4.83e-05 [cse_after_recomputation]: 2.789e-05, [1] [Cycle 1]: 2.362e-05, [1] [cse]: 1.82e-05 [environ_conv]: 6.12999e-06 [swap_dp_allreduce_reducescatter]: 6.16e-06 [bias_add_comm_swap]: 2.49001e-06 [label_micro_interleaved_index]: 4.48001e-06 [label_fine_grained_interleaved_index]: 2.63e-06 [merge_cast_opt]: 1.51998e-06 [slice_recompute_activation]: 2.34001e-06 [micro_interleaved_order_control]: 2.56998e-06 [assign_add_opt]: 1.50001e-06 [ForceFp32Comm]: 9.70002e-07 [remove_cast_before_assign_add]: 1.10001e-06 [full_micro_interleaved_order_control]: 2.44999e-06 [reorder_send_recv_between_fp_bp]: 2.41e-06 [comm_op_add_attrs]: 9.89996e-07 [add_comm_op_reuse_tag]: 1.02e-06 [interleave_split_concat_branches]: 1.12e-06 [interleave_parallel_branches]: 1.00001e-06 [overlap_opt_shard_in_pipeline]: 1.10001e-06 [overlap_opt_shard_grad_in_pipeline]: 1.65001e-06 [control_data_broadcast_order]: 1.209e-05 [grouped_pairwise_exchange_alltoall]: 1.66e-06 [offloading_packed_experts]: 3.95e-06 [overlap_recompute_and_grad_model_parallel]: 4.88001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.10001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.35999e-06 [overlap_recompute_comm]: 2.56e-06 [overlap_grad_ring_attention]: 4.37998e-06 [overlap_grad_flash_sp]: 1.939e-05 [begin_end_overlap_inline]: 5.19998e-07 [split_matmul_comm_elemetwise]: 2.48e-06 [split_layernorm_comm]: 1.99999e-06 [handle_group_info]: 1.01002e-06 [symbol_engine_optimizer]: 7.688e-05, [1] [Cycle 1]: 7.251e-05, [6] [build]: 3.06001e-06 [elim_shapecalc]: 9.89001e-06 [elim_not_effective]: 1.251e-05 [opt_reshape]: 6.91999e-06 [fold_const_symbol]: 9.88998e-06 [renormalize]: 2.30008e-07 [detach_backward]: 1.91e-06 [pipeline_parallel_scheduler]: 1.49e-06 [auto_monad_reorder]: 1.866e-05 [get_jit_bprop_graph]: 1.52999e-06 [rewriter_after_jit_bprop_graph]: 4.69002e-06 [opt_after_jit_grad]: 0.00048049 [validate]: 4.169e-05 Sums bootstrap : 0.000590s : 6.41% type_inference : 0.004091s : 44.47% event_method : 0.000013s : 0.14% auto_monad : 0.000055s : 0.60% graph_reusing : 0.000005s : 0.05% inline : 0.000003s : 0.03% add_attr.add_attr_with_inline.tag_attr : 0.000016s : 0.17% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.04% parallel-infer-symbol : 0.000004s : 0.04% pre_auto_parallel : 0.000033s : 0.36% insert-virtual-dataset : 0.000003s : 0.03% parallel-infer-symbol-second : 0.000001s : 0.01% dataset_repeat_opt : 0.000002s : 0.02% pipeline_split : 0.000002s : 0.02% optimize.py_interpret_to_execute : 0.000021s : 0.23% optimize.rewriter_before_opt_a : 0.000047s : 0.51% optimize.opt_a.expand_dump_flag : 0.000004s : 0.04% optimize.opt_a.switch_simplify : 0.000034s : 0.37% optimize.opt_a.loop_unroll : 0.000020s : 0.22% optimize.opt_a.a_1 : 0.000493s : 5.36% optimize.opt_a.with_stream_mark : 0.000036s : 0.39% optimize.opt_a.recompute_prepare : 0.000016s : 0.18% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.08% optimize.opt_a.updatestate_assign_eliminate : 0.000006s : 0.07% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.07% optimize.opt_a.parameter_eliminate : 0.000003s : 0.03% optimize.opt_a.a_2 : 0.000169s : 1.83% optimize.opt_a.accelerated_algorithm : 0.000030s : 0.33% optimize.opt_a.shard : 0.000004s : 0.04% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.04% optimize.opt_a.shard_inline : 0.000015s : 0.16% optimize.opt_a.merge_send_recv : 0.000015s : 0.16% optimize.opt_a.auto_parallel : 0.000015s : 0.16% optimize.opt_a.parallel : 0.000039s : 0.42% optimize.opt_a.flash_sp : 0.000014s : 0.15% optimize.opt_a.merge_comm : 0.000009s : 0.09% optimize.opt_a.allreduce_fusion : 0.000008s : 0.09% optimize.opt_a.matmul_add_comm_reduction : 0.000022s : 0.24% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000017s : 0.19% optimize.opt_a.virtual_dataset : 0.000015s : 0.16% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.14% optimize.opt_a.virtual_output : 0.000013s : 0.14% optimize.opt_a.merge_forward : 0.000008s : 0.09% optimize.opt_a.cell_reuse_recompute_pass : 0.000003s : 0.04% optimize.opt_a.offload_activation : 0.000022s : 0.24% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000029s : 0.32% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.03% optimize.opt_a.before_grad : 0.000022s : 0.24% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000009s : 0.10% optimize.opt_a.meta_fg_expand : 0.000006s : 0.06% optimize.opt_a.flash_sp_send_recv_attached : 0.000003s : 0.04% optimize.opt_a.receive_attached : 0.000004s : 0.05% optimize.opt_a.after_resolve : 0.000023s : 0.25% optimize.opt_a.a_after_grad : 0.000020s : 0.22% optimize.opt_a.renormalize : 0.000684s : 7.43% optimize.opt_a.add_forward_monad_depend : 0.000011s : 0.12% optimize.opt_a.auto_monad_grad : 0.000004s : 0.05% optimize.opt_a.auto_monad_eliminator : 0.000024s : 0.26% optimize.opt_a.cse : 0.000062s : 0.67% optimize.opt_a.a_3 : 0.000095s : 1.03% optimize.py_interpret_to_execute_after_opt_a : 0.000013s : 0.14% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.02% optimize.rewriter_after_opt_a : 0.000043s : 0.47% optimize.convert_after_rewriter : 0.000007s : 0.08% optimize.order_py_execute_after_rewriter : 0.000006s : 0.06% optimize.mutable_eliminate : 0.000607s : 6.60% optimize.opt_b.b_1 : 0.000131s : 1.42% optimize.opt_b.b_2 : 0.000009s : 0.09% optimize.opt_b.updatestate_depend_eliminate : 0.000009s : 0.09% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.03% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.03% optimize.opt_b.renormalize : 0.000001s : 0.01% optimize.opt_b.cse : 0.000028s : 0.30% optimize.optimize_parallel_all_gather_comm : 0.000018s : 0.20% optimize.overlap_param_gather : 0.000002s : 0.02% optimize.cconv : 0.000028s : 0.30% optimize.loop_unroll : 0.000441s : 4.79% optimize.opt_after_cconv.c_1 : 0.000030s : 0.33% optimize.opt_after_cconv.parameter_eliminate : 0.000004s : 0.04% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000006s : 0.06% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.03% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000002s : 0.03% optimize.opt_after_cconv.cse : 0.000023s : 0.25% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000016s : 0.17% optimize.tuple_transform.d_1 : 0.000045s : 0.49% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.02% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.08% optimize.partial_unused_args_eliminate : 0.000002s : 0.02% optimize.add_recomputation : 0.000048s : 0.53% optimize.cse_after_recomputation.cse : 0.000018s : 0.20% optimize.environ_conv : 0.000006s : 0.07% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.07% optimize.bias_add_comm_swap : 0.000002s : 0.03% optimize.label_micro_interleaved_index : 0.000004s : 0.05% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.03% optimize.merge_cast_opt : 0.000002s : 0.02% optimize.slice_recompute_activation : 0.000002s : 0.03% optimize.micro_interleaved_order_control : 0.000003s : 0.03% optimize.assign_add_opt : 0.000002s : 0.02% optimize.ForceFp32Comm : 0.000001s : 0.01% optimize.remove_cast_before_assign_add : 0.000001s : 0.01% optimize.full_micro_interleaved_order_control : 0.000002s : 0.03% optimize.reorder_send_recv_between_fp_bp : 0.000002s : 0.03% optimize.comm_op_add_attrs : 0.000001s : 0.01% optimize.add_comm_op_reuse_tag : 0.000001s : 0.01% optimize.interleave_split_concat_branches : 0.000001s : 0.01% optimize.interleave_parallel_branches : 0.000001s : 0.01% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.01% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.02% optimize.control_data_broadcast_order : 0.000012s : 0.13% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.02% optimize.offloading_packed_experts : 0.000004s : 0.04% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.05% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.01% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.01% optimize.overlap_recompute_comm : 0.000003s : 0.03% optimize.overlap_grad_ring_attention : 0.000004s : 0.05% optimize.overlap_grad_flash_sp : 0.000019s : 0.21% optimize.begin_end_overlap_inline : 0.000001s : 0.01% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.03% optimize.split_layernorm_comm : 0.000002s : 0.02% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000003s : 0.03% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000010s : 0.11% optimize.symbol_engine_optimizer.elim_not_effective : 0.000013s : 0.14% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.08% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.11% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.02% pipeline_parallel_scheduler : 0.000001s : 0.02% auto_monad_reorder : 0.000019s : 0.20% get_jit_bprop_graph : 0.000002s : 0.02% rewriter_after_jit_bprop_graph : 0.000005s : 0.05% opt_after_jit_grad : 0.000480s : 5.22% validate : 0.000042s : 0.45% Time group info: ------[substitution.] 0.000107 21 1.66% : 0.000002s : 2: substitution.elim_not_effective 1.24% : 0.000001s : 2: substitution.fold_const_symbol 5.43% : 0.000006s : 4: substitution.graph_param_transform 78.88% : 0.000085s : 1: substitution.inline 3.94% : 0.000004s : 4: substitution.j_node_and_user_rematch 4.46% : 0.000005s : 4: substitution.remove_not_recompute_node 4.39% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.004026 2 87.61% : 0.003528s : 1: type_inference.infer 12.39% : 0.000499s : 1: type_inference.specialize ------[replace.] 0.000018 1 100.00% : 0.000018s : 1: replace.inline ------[match.] 0.000084 1 100.00% : 0.000084s : 1: match.inline ------[predicate.] 0.000156 967 0.74% : 0.000001s : 9: predicate.accumulaten_eliminater 0.91% : 0.000001s : 4: predicate.ad_related_special_op_eliminate 0.67% : 0.000001s : 8: predicate.addn_check_dump 0.88% : 0.000001s : 9: predicate.addn_zero_filter 0.74% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.19% : 0.000003s : 17: predicate.arithmetic_simplify 0.82% : 0.000001s : 9: predicate.cast_eliminate 0.98% : 0.000002s : 8: predicate.check_bprop_eliminate 0.68% : 0.000001s : 8: predicate.compare_switch_simplify 0.25% : 0.000000s : 4: predicate.const_output_eliminate 0.66% : 0.000001s : 8: predicate.depend_value_elim 0.83% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.93% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.88% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.35% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.30% : 0.000000s : 4: predicate.elim_not_effective 0.45% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.07% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.01% : 0.000002s : 13: predicate.environ_get_add_eliminate 1.06% : 0.000002s : 13: predicate.environ_get_depend_swap 1.80% : 0.000003s : 21: predicate.environ_get_eliminate 1.09% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.84% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.72% : 0.000003s : 10: predicate.float_depend_g_call 0.75% : 0.000001s : 8: predicate.float_environ_get_switch 0.95% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.27% : 0.000000s : 4: predicate.fold_const_symbol 0.83% : 0.000001s : 8: predicate.get_grad_eliminate 0.24% : 0.000000s : 4: predicate.graph_param_transform 0.82% : 0.000001s : 8: predicate.incorporate_call 0.63% : 0.000001s : 8: predicate.incorporate_call_switch 6.26% : 0.000010s : 43: predicate.inline 1.18% : 0.000002s : 8: predicate.inline_without_move 0.42% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.27% : 0.000002s : 8: predicate.less_batch_normalization 1.77% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.43% : 0.000004s : 26: predicate.load_eliminater 1.09% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.46% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.77% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.78% : 0.000001s : 8: predicate.merge_addn 0.76% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.74% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.72% : 0.000001s : 9: predicate.minmaximum_grad 1.27% : 0.000002s : 4: predicate.mutable_eliminate 0.43% : 0.000001s : 4: predicate.opt_reshape 0.51% : 0.000001s : 4: predicate.parallel_virtual_node 1.11% : 0.000002s : 10: predicate.partial_defer_inline 1.26% : 0.000002s : 13: predicate.partial_eliminate 0.75% : 0.000001s : 9: predicate.print_const_string_wrapper 0.74% : 0.000001s : 8: predicate.reduce_all_const_elim 1.08% : 0.000002s : 9: predicate.reduce_eliminate 2.24% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.77% : 0.000001s : 8: predicate.remove_not_recompute_node 1.44% : 0.000002s : 17: predicate.replace_applicator 0.80% : 0.000001s : 8: predicate.replace_old_param 0.40% : 0.000001s : 4: predicate.reset_defer_inline 0.84% : 0.000001s : 9: predicate.reshape_eliminate 0.95% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.45% : 0.000001s : 4: predicate.row_tensor_eliminate 1.07% : 0.000002s : 8: predicate.same_eliminate 0.61% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.00% : 0.000002s : 8: predicate.shard_identity_eliminate 1.06% : 0.000002s : 8: predicate.special_op_eliminate 0.91% : 0.000001s : 8: predicate.specialize_transform 1.09% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.92% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.40% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.95% : 0.000001s : 10: predicate.switch_defer_inline 1.67% : 0.000003s : 18: predicate.switch_layer_defer_inline 4.19% : 0.000007s : 35: predicate.switch_simplify 1.01% : 0.000002s : 9: predicate.tile_eliminate 0.84% : 0.000001s : 9: predicate.transpose_eliminate 1.62% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.60% : 0.000002s : 17: predicate.tuple_list_get_item_const_eliminator 1.33% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.22% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.40% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.35% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.59% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 2.16% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.98% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.49% : 0.000001s : 4: predicate.value_based_eliminate 0.89% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.80% : 0.000001s : 8: predicate.virtual_output_eliminate 0.33% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.53% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000228 4 7.87% : 0.000018s : 1: func_graph_cloner_run.FuncGraphClonerGraph 92.13% : 0.000210s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.027118 192 0.01% : 0.000004s : 1: ForceFp32Comm 19.15% : 0.005192s : 1: add_attr 19.09% : 0.005176s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.19% : 0.000052s : 1: add_recomputation 0.04% : 0.000011s : 1: assign_add_opt 0.22% : 0.000060s : 1: auto_monad 0.08% : 0.000022s : 1: auto_monad_reorder 0.01% : 0.000003s : 1: begin_end_overlap_inline 0.02% : 0.000005s : 1: bias_add_comm_swap 2.31% : 0.000626s : 1: bootstrap 0.11% : 0.000031s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.06% : 0.000015s : 1: control_data_broadcast_order 0.04% : 0.000011s : 1: convert_after_rewriter 0.11% : 0.000031s : 1: cse_after_recomputation 0.02% : 0.000005s : 1: dataset_repeat_opt 0.03% : 0.000009s : 1: detach_backward 0.03% : 0.000009s : 1: environ_conv 0.07% : 0.000020s : 1: event_method 0.02% : 0.000005s : 1: full_micro_interleaved_order_control 0.02% : 0.000005s : 1: get_jit_bprop_graph 0.03% : 0.000009s : 1: graph_reusing 0.02% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000004s : 1: handle_group_info 0.02% : 0.000006s : 1: inline 0.02% : 0.000006s : 1: insert-virtual-dataset 0.03% : 0.000007s : 1: interleave_parallel_branches 0.01% : 0.000004s : 1: interleave_split_concat_branches 0.02% : 0.000006s : 1: label_fine_grained_interleaved_index 0.03% : 0.000007s : 1: label_micro_interleaved_index 1.66% : 0.000449s : 1: loop_unroll 0.02% : 0.000005s : 1: merge_cast_opt 0.02% : 0.000006s : 1: micro_interleaved_order_control 2.28% : 0.000619s : 1: mutable_eliminate 0.03% : 0.000007s : 1: offloading_packed_experts 0.05% : 0.000014s : 1: opt.transform.loop_unroll_optimizer 0.06% : 0.000016s : 1: opt.transform.mutable_eliminate 3.43% : 0.000931s : 78: opt.transform.opt_a 0.11% : 0.000029s : 1: opt.transform.opt_after_cconv 0.09% : 0.000024s : 1: opt.transform.opt_after_jit_grad 0.40% : 0.000110s : 28: opt.transform.opt_b 0.19% : 0.000051s : 2: opt.transform.opt_trans_graph 0.13% : 0.000035s : 4: opt.transform.symbol_engine_opt 9.51% : 0.002578s : 1: opt_a 0.43% : 0.000118s : 1: opt_after_cconv 1.81% : 0.000490s : 1: opt_after_jit_grad 0.84% : 0.000228s : 1: opt_b 17.47% : 0.004739s : 1: optimize 0.08% : 0.000022s : 1: optimize_parallel_all_gather_comm 0.03% : 0.000009s : 1: order_py_execute_after_rewriter 0.09% : 0.000023s : 1: overlap_grad_flash_sp 0.02% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.03% : 0.000007s : 1: overlap_grad_ring_attention 0.02% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.01% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.02% : 0.000005s : 1: overlap_param_gather 0.03% : 0.000007s : 1: overlap_recompute_allgather_and_fa_grad 0.03% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.02% : 0.000005s : 1: overlap_recompute_comm 0.03% : 0.000008s : 1: parallel-infer-symbol 0.01% : 0.000004s : 1: parallel-infer-symbol-second 0.02% : 0.000005s : 1: partial_unused_args_eliminate 0.02% : 0.000005s : 1: pipeline_parallel_scheduler 0.02% : 0.000005s : 1: pipeline_split 0.14% : 0.000038s : 1: pre_auto_parallel 0.09% : 0.000026s : 1: py_interpret_to_execute 0.06% : 0.000016s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000004s : 1: remove_cast_before_assign_add 0.07% : 0.000020s : 1: remove_dup_value 1.53% : 0.000416s : 1: renormalize.infer 0.95% : 0.000259s : 1: renormalize.specialize 0.02% : 0.000005s : 1: reorder_send_recv_between_fp_bp 0.03% : 0.000008s : 1: rewriter_after_jit_bprop_graph 0.18% : 0.000048s : 1: rewriter_after_opt_a 0.19% : 0.000051s : 1: rewriter_before_opt_a 0.02% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.02% : 0.000005s : 1: slice_recompute_activation 0.02% : 0.000005s : 1: split_layernorm_comm 0.02% : 0.000005s : 1: split_matmul_comm_elemetwise 0.03% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.29% : 0.000080s : 1: symbol_engine_optimizer 0.30% : 0.000081s : 1: tuple_transform 15.17% : 0.004114s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_bf16[dim_type1-True-batch_size_list0-shape0-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_bf16[dim_type1-True-batch_size_list0-shape0-ms_dtype0],max_mem:516.0M [WARNING] ME(152945:281473835274032,MainProcess):2026-01-29-17:44:08.631.515 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.0274408, [21] [bootstrap]: 0.00062131 [type_inference]: 0.0163296 [event_method]: 1.4e-05 [auto_monad]: 5.43e-05 [graph_reusing]: 5.32001e-06 [inline]: 2.64999e-06 [add_attr]: 0.0044678, [1] [add_attr_with_inline]: 0.00445394, [1] [Cycle 1]: 6.008e-05, [2] [tag_attr]: 1.655e-05 [meta_addattr_fg_expand]: 3.22002e-06 [parallel-infer-symbol]: 3.80998e-06 [pre_auto_parallel]: 3.255e-05 [insert-virtual-dataset]: 2.44999e-06 [parallel-infer-symbol-second]: 9.50007e-07 [dataset_repeat_opt]: 1.79998e-06 [pipeline_split]: 1.61998e-06 [optimize]: 0.00503307, [53] [py_interpret_to_execute]: 2.031e-05 [rewriter_before_opt_a]: 4.709e-05 [opt_a]: 0.00265002, [2] [Cycle 1]: 0.00167263, [45] [expand_dump_flag]: 2.61999e-06 [switch_simplify]: 2.55e-05 [loop_unroll]: 1.253e-05 [a_1]: 0.00029755 [with_stream_mark]: 2.069e-05 [recompute_prepare]: 9.15001e-06 [updatestate_depend_eliminate]: 4.47e-06 [updatestate_assign_eliminate]: 3.41001e-06 [updatestate_loads_eliminate]: 3.13e-06 [parameter_eliminate]: 2.51e-06 [a_2]: 8.403e-05 [accelerated_algorithm]: 7.36999e-06 [shard]: 2.04e-06 [meta_shard_fg_expand]: 1.69e-06 [shard_inline]: 7.41999e-06 [merge_send_recv]: 8.23999e-06 [auto_parallel]: 5.95002e-06 [parallel]: 2.922e-05 [flash_sp]: 8.97e-06 [merge_comm]: 4.07e-06 [allreduce_fusion]: 3.83999e-06 [matmul_add_comm_reduction]: 9.41003e-06 [allreduce_slice_to_reducescatter]: 8.50006e-07 [virtual_shard_identity]: 9.00999e-06 [virtual_dataset]: 6.84001e-06 [get_grad_eliminate_]: 6.61999e-06 [virtual_output]: 6.68e-06 [merge_forward]: 4.36002e-06 [cell_reuse_recompute_pass]: 1.72999e-06 [offload_activation]: 1.071e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.472e-05 [merge_recompute_call_nodes]: 1.36998e-06 [before_grad]: 1.109e-05 [set_forward_comm_id_for_comm_node_pass]: 3.70998e-06 [meta_fg_expand]: 2.84999e-06 [flash_sp_send_recv_attached]: 2.63e-06 [receive_attached]: 2.78e-06 [after_resolve]: 1.242e-05 [a_after_grad]: 9.81e-06 [renormalize]: 0.00063767 [add_forward_monad_depend]: 1.068e-05 [auto_monad_grad]: 2.42001e-06 [auto_monad_eliminator]: 1.549e-05 [cse]: 3.576e-05 [a_3]: 5.326e-05 [Cycle 2]: 0.00096574, [45] [expand_dump_flag]: 1.96e-06 [switch_simplify]: 1.149e-05 [loop_unroll]: 7.45998e-06 [a_1]: 0.00022334 [with_stream_mark]: 1.82e-05 [recompute_prepare]: 8.62998e-06 [updatestate_depend_eliminate]: 4.65999e-06 [updatestate_assign_eliminate]: 3.12002e-06 [updatestate_loads_eliminate]: 2.99001e-06 [parameter_eliminate]: 1.70001e-06 [a_2]: 7.976e-05 [accelerated_algorithm]: 7.88999e-06 [shard]: 1.29e-06 [meta_shard_fg_expand]: 2.21e-06 [shard_inline]: 7.2e-06 [merge_send_recv]: 6.70002e-06 [auto_parallel]: 8.75999e-06 [parallel]: 6.23002e-06 [flash_sp]: 3.68999e-06 [merge_comm]: 3.75e-06 [allreduce_fusion]: 3.43999e-06 [matmul_add_comm_reduction]: 1.046e-05 [allreduce_slice_to_reducescatter]: 3.50003e-07 [virtual_shard_identity]: 7.57998e-06 [virtual_dataset]: 6.68e-06 [get_grad_eliminate_]: 6.23e-06 [virtual_output]: 6.13998e-06 [merge_forward]: 3.97e-06 [cell_reuse_recompute_pass]: 1.78002e-06 [offload_activation]: 1.508e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.726e-05 [merge_recompute_call_nodes]: 1.24e-06 [before_grad]: 1.052e-05 [set_forward_comm_id_for_comm_node_pass]: 4.12e-06 [meta_fg_expand]: 2.93998e-06 [flash_sp_send_recv_attached]: 1.27999e-06 [receive_attached]: 1.87001e-06 [after_resolve]: 1.166e-05 [a_after_grad]: 1.024e-05 [renormalize]: 1.30007e-07 [add_forward_monad_depend]: 2.23998e-06 [auto_monad_grad]: 2.51e-06 [auto_monad_eliminator]: 9.20001e-06 [cse]: 2.48e-05 [a_3]: 0.0001194 [py_interpret_to_execute_after_opt_a]: 1.629e-05 [slice_cell_reuse_recomputed_activation]: 2.19999e-06 [rewriter_after_opt_a]: 4.44e-05 [convert_after_rewriter]: 7.07997e-06 [order_py_execute_after_rewriter]: 5.32001e-06 [mutable_eliminate]: 0.00066624 [opt_b]: 0.00023954, [1] [Cycle 1]: 0.0002311, [7] [b_1]: 0.00013368 [b_2]: 8.78001e-06 [updatestate_depend_eliminate]: 9.76998e-06 [updatestate_assign_eliminate]: 3.46999e-06 [updatestate_loads_eliminate]: 2.81999e-06 [renormalize]: 1.01002e-06 [cse]: 3.44e-05 [optimize_parallel_all_gather_comm]: 2.108e-05 [overlap_param_gather]: 1.81e-06 [cconv]: 3.451e-05 [loop_unroll]: 0.00050831 [opt_after_cconv]: 0.00013083, [1] [Cycle 1]: 0.00012345, [7] [c_1]: 3.422e-05 [parameter_eliminate]: 6.12001e-06 [updatestate_depend_eliminate]: 7.88999e-06 [updatestate_assign_eliminate]: 2.74001e-06 [updatestate_loads_eliminate]: 2.44001e-06 [cse]: 2.781e-05 [renormalize]: 7.80012e-07 [remove_dup_value]: 1.672e-05 [tuple_transform]: 8.464e-05, [1] [Cycle 1]: 7.897e-05, [4] [d_1]: 5.008e-05 [none_parameter_eliminate]: 1.69998e-06 [renormalize]: 2.30008e-07 [switch_simplify]: 7.69002e-06 [partial_unused_args_eliminate]: 2.07001e-06 [add_recomputation]: 5.602e-05 [cse_after_recomputation]: 3.249e-05, [1] [Cycle 1]: 2.653e-05, [1] [cse]: 2.068e-05 [environ_conv]: 6.78e-06 [swap_dp_allreduce_reducescatter]: 5.76998e-06 [bias_add_comm_swap]: 2.78e-06 [label_micro_interleaved_index]: 4.60001e-06 [label_fine_grained_interleaved_index]: 2.66e-06 [merge_cast_opt]: 1.39e-06 [slice_recompute_activation]: 2.55002e-06 [micro_interleaved_order_control]: 2.32999e-06 [assign_add_opt]: 1.66e-06 [ForceFp32Comm]: 9.39996e-07 [remove_cast_before_assign_add]: 1.09e-06 [full_micro_interleaved_order_control]: 2.22001e-06 [reorder_send_recv_between_fp_bp]: 3.06999e-06 [comm_op_add_attrs]: 1.00999e-06 [add_comm_op_reuse_tag]: 1.02e-06 [interleave_split_concat_branches]: 1.35001e-06 [interleave_parallel_branches]: 1.09e-06 [overlap_opt_shard_in_pipeline]: 1.12e-06 [overlap_opt_shard_grad_in_pipeline]: 2.22001e-06 [control_data_broadcast_order]: 1.514e-05 [grouped_pairwise_exchange_alltoall]: 1.45999e-06 [offloading_packed_experts]: 4.18001e-06 [overlap_recompute_and_grad_model_parallel]: 5.07e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.14e-06 [overlap_recompute_allgather_and_fa_grad]: 1.27e-06 [overlap_recompute_comm]: 2.44001e-06 [overlap_grad_ring_attention]: 4.89e-06 [overlap_grad_flash_sp]: 2.296e-05 [begin_end_overlap_inline]: 7.10017e-07 [split_matmul_comm_elemetwise]: 1.99e-06 [split_layernorm_comm]: 2.16e-06 [handle_group_info]: 1.47001e-06 [symbol_engine_optimizer]: 8.623e-05, [1] [Cycle 1]: 8.131e-05, [6] [build]: 4.38001e-06 [elim_shapecalc]: 1.18e-05 [elim_not_effective]: 1.506e-05 [opt_reshape]: 7.43e-06 [fold_const_symbol]: 1.078e-05 [renormalize]: 1.8999e-07 [detach_backward]: 2.22999e-06 [pipeline_parallel_scheduler]: 1.47001e-06 [auto_monad_reorder]: 2.081e-05 [get_jit_bprop_graph]: 1.99999e-06 [rewriter_after_jit_bprop_graph]: 5.97001e-06 [opt_after_jit_grad]: 0.00060509 [validate]: 5.276e-05 Sums bootstrap : 0.000621s : 2.84% type_inference : 0.016330s : 74.72% event_method : 0.000014s : 0.06% auto_monad : 0.000054s : 0.25% graph_reusing : 0.000005s : 0.02% inline : 0.000003s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000017s : 0.08% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.01% parallel-infer-symbol : 0.000004s : 0.02% pre_auto_parallel : 0.000033s : 0.15% insert-virtual-dataset : 0.000002s : 0.01% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.01% pipeline_split : 0.000002s : 0.01% optimize.py_interpret_to_execute : 0.000020s : 0.09% optimize.rewriter_before_opt_a : 0.000047s : 0.22% optimize.opt_a.expand_dump_flag : 0.000005s : 0.02% optimize.opt_a.switch_simplify : 0.000037s : 0.17% optimize.opt_a.loop_unroll : 0.000020s : 0.09% optimize.opt_a.a_1 : 0.000521s : 2.38% optimize.opt_a.with_stream_mark : 0.000039s : 0.18% optimize.opt_a.recompute_prepare : 0.000018s : 0.08% optimize.opt_a.updatestate_depend_eliminate : 0.000009s : 0.04% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.03% optimize.opt_a.updatestate_loads_eliminate : 0.000006s : 0.03% optimize.opt_a.parameter_eliminate : 0.000004s : 0.02% optimize.opt_a.a_2 : 0.000164s : 0.75% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.07% optimize.opt_a.shard : 0.000003s : 0.02% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.02% optimize.opt_a.shard_inline : 0.000015s : 0.07% optimize.opt_a.merge_send_recv : 0.000015s : 0.07% optimize.opt_a.auto_parallel : 0.000015s : 0.07% optimize.opt_a.parallel : 0.000035s : 0.16% optimize.opt_a.flash_sp : 0.000013s : 0.06% optimize.opt_a.merge_comm : 0.000008s : 0.04% optimize.opt_a.allreduce_fusion : 0.000007s : 0.03% optimize.opt_a.matmul_add_comm_reduction : 0.000020s : 0.09% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000017s : 0.08% optimize.opt_a.virtual_dataset : 0.000014s : 0.06% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.06% optimize.opt_a.virtual_output : 0.000013s : 0.06% optimize.opt_a.merge_forward : 0.000008s : 0.04% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.02% optimize.opt_a.offload_activation : 0.000026s : 0.12% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000032s : 0.15% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.01% optimize.opt_a.before_grad : 0.000022s : 0.10% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.04% optimize.opt_a.meta_fg_expand : 0.000006s : 0.03% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.02% optimize.opt_a.receive_attached : 0.000005s : 0.02% optimize.opt_a.after_resolve : 0.000024s : 0.11% optimize.opt_a.a_after_grad : 0.000020s : 0.09% optimize.opt_a.renormalize : 0.000638s : 2.92% optimize.opt_a.add_forward_monad_depend : 0.000013s : 0.06% optimize.opt_a.auto_monad_grad : 0.000005s : 0.02% optimize.opt_a.auto_monad_eliminator : 0.000025s : 0.11% optimize.opt_a.cse : 0.000061s : 0.28% optimize.opt_a.a_3 : 0.000173s : 0.79% optimize.py_interpret_to_execute_after_opt_a : 0.000016s : 0.07% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.01% optimize.rewriter_after_opt_a : 0.000044s : 0.20% optimize.convert_after_rewriter : 0.000007s : 0.03% optimize.order_py_execute_after_rewriter : 0.000005s : 0.02% optimize.mutable_eliminate : 0.000666s : 3.05% optimize.opt_b.b_1 : 0.000134s : 0.61% optimize.opt_b.b_2 : 0.000009s : 0.04% optimize.opt_b.updatestate_depend_eliminate : 0.000010s : 0.04% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.02% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.01% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000034s : 0.16% optimize.optimize_parallel_all_gather_comm : 0.000021s : 0.10% optimize.overlap_param_gather : 0.000002s : 0.01% optimize.cconv : 0.000035s : 0.16% optimize.loop_unroll : 0.000508s : 2.33% optimize.opt_after_cconv.c_1 : 0.000034s : 0.16% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.03% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000008s : 0.04% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000002s : 0.01% optimize.opt_after_cconv.cse : 0.000028s : 0.13% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000017s : 0.08% optimize.tuple_transform.d_1 : 0.000050s : 0.23% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.01% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000008s : 0.04% optimize.partial_unused_args_eliminate : 0.000002s : 0.01% optimize.add_recomputation : 0.000056s : 0.26% optimize.cse_after_recomputation.cse : 0.000021s : 0.09% optimize.environ_conv : 0.000007s : 0.03% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.03% optimize.bias_add_comm_swap : 0.000003s : 0.01% optimize.label_micro_interleaved_index : 0.000005s : 0.02% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.01% optimize.merge_cast_opt : 0.000001s : 0.01% optimize.slice_recompute_activation : 0.000003s : 0.01% optimize.micro_interleaved_order_control : 0.000002s : 0.01% optimize.assign_add_opt : 0.000002s : 0.01% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.01% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.01% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.01% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.01% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.01% optimize.control_data_broadcast_order : 0.000015s : 0.07% optimize.grouped_pairwise_exchange_alltoall : 0.000001s : 0.01% optimize.offloading_packed_experts : 0.000004s : 0.02% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.02% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.01% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.01% optimize.overlap_recompute_comm : 0.000002s : 0.01% optimize.overlap_grad_ring_attention : 0.000005s : 0.02% optimize.overlap_grad_flash_sp : 0.000023s : 0.11% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.01% optimize.split_layernorm_comm : 0.000002s : 0.01% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000004s : 0.02% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000012s : 0.05% optimize.symbol_engine_optimizer.elim_not_effective : 0.000015s : 0.07% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.03% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000011s : 0.05% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.01% pipeline_parallel_scheduler : 0.000001s : 0.01% auto_monad_reorder : 0.000021s : 0.10% get_jit_bprop_graph : 0.000002s : 0.01% rewriter_after_jit_bprop_graph : 0.000006s : 0.03% opt_after_jit_grad : 0.000605s : 2.77% validate : 0.000053s : 0.24% Time group info: ------[substitution.] 0.000105 21 1.94% : 0.000002s : 2: substitution.elim_not_effective 1.84% : 0.000002s : 2: substitution.fold_const_symbol 6.16% : 0.000006s : 4: substitution.graph_param_transform 76.30% : 0.000080s : 1: substitution.inline 3.85% : 0.000004s : 4: substitution.j_node_and_user_rematch 5.18% : 0.000005s : 4: substitution.remove_not_recompute_node 4.73% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.016269 2 97.40% : 0.015847s : 1: type_inference.infer 2.60% : 0.000422s : 1: type_inference.specialize ------[replace.] 0.000017 1 100.00% : 0.000017s : 1: replace.inline ------[match.] 0.000079 1 100.00% : 0.000079s : 1: match.inline ------[predicate.] 0.000162 967 0.77% : 0.000001s : 9: predicate.accumulaten_eliminater 1.43% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.70% : 0.000001s : 8: predicate.addn_check_dump 0.94% : 0.000002s : 9: predicate.addn_zero_filter 0.64% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 1.98% : 0.000003s : 17: predicate.arithmetic_simplify 0.74% : 0.000001s : 9: predicate.cast_eliminate 0.73% : 0.000001s : 8: predicate.check_bprop_eliminate 0.60% : 0.000001s : 8: predicate.compare_switch_simplify 0.25% : 0.000000s : 4: predicate.const_output_eliminate 0.76% : 0.000001s : 8: predicate.depend_value_elim 0.77% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.93% : 0.000002s : 9: predicate.dict_get_item_eliminator 0.72% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.64% : 0.000003s : 8: predicate.dumpgradient_eliminate 0.36% : 0.000001s : 4: predicate.elim_not_effective 0.53% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.07% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.06% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.98% : 0.000002s : 13: predicate.environ_get_depend_swap 1.68% : 0.000003s : 21: predicate.environ_get_eliminate 0.98% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.77% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.80% : 0.000003s : 10: predicate.float_depend_g_call 0.62% : 0.000001s : 8: predicate.float_environ_get_switch 0.94% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.26% : 0.000000s : 4: predicate.fold_const_symbol 0.80% : 0.000001s : 8: predicate.get_grad_eliminate 0.27% : 0.000000s : 4: predicate.graph_param_transform 0.73% : 0.000001s : 8: predicate.incorporate_call 0.67% : 0.000001s : 8: predicate.incorporate_call_switch 7.30% : 0.000012s : 43: predicate.inline 1.20% : 0.000002s : 8: predicate.inline_without_move 0.44% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.33% : 0.000002s : 8: predicate.less_batch_normalization 1.98% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.17% : 0.000004s : 26: predicate.load_eliminater 1.31% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.43% : 0.000002s : 13: predicate.loop_unroll_before_grad 2.09% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.64% : 0.000001s : 8: predicate.merge_addn 0.73% : 0.000001s : 8: predicate.micro_step_allgather_replace 1.06% : 0.000002s : 8: predicate.mini_step_allgather_replace 0.68% : 0.000001s : 9: predicate.minmaximum_grad 1.55% : 0.000003s : 4: predicate.mutable_eliminate 0.43% : 0.000001s : 4: predicate.opt_reshape 0.46% : 0.000001s : 4: predicate.parallel_virtual_node 1.29% : 0.000002s : 10: predicate.partial_defer_inline 1.17% : 0.000002s : 13: predicate.partial_eliminate 0.77% : 0.000001s : 9: predicate.print_const_string_wrapper 0.65% : 0.000001s : 8: predicate.reduce_all_const_elim 1.09% : 0.000002s : 9: predicate.reduce_eliminate 2.09% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.73% : 0.000001s : 8: predicate.remove_not_recompute_node 1.33% : 0.000002s : 17: predicate.replace_applicator 0.68% : 0.000001s : 8: predicate.replace_old_param 0.63% : 0.000001s : 4: predicate.reset_defer_inline 0.77% : 0.000001s : 9: predicate.reshape_eliminate 1.09% : 0.000002s : 8: predicate.row_tensor_add_zeros_like 0.57% : 0.000001s : 4: predicate.row_tensor_eliminate 1.07% : 0.000002s : 8: predicate.same_eliminate 0.57% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.96% : 0.000002s : 8: predicate.shard_identity_eliminate 1.06% : 0.000002s : 8: predicate.special_op_eliminate 0.89% : 0.000001s : 8: predicate.specialize_transform 0.98% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.93% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.56% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.86% : 0.000001s : 10: predicate.switch_defer_inline 1.51% : 0.000002s : 18: predicate.switch_layer_defer_inline 3.59% : 0.000006s : 35: predicate.switch_simplify 0.70% : 0.000001s : 9: predicate.tile_eliminate 0.83% : 0.000001s : 9: predicate.transpose_eliminate 1.44% : 0.000002s : 17: predicate.tuple_list_convert_item_index_to_positive 1.63% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.30% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.19% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.40% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.33% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.54% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 2.01% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.95% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.49% : 0.000001s : 4: predicate.value_based_eliminate 0.89% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.77% : 0.000001s : 8: predicate.virtual_output_eliminate 0.35% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.45% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000220 4 10.21% : 0.000022s : 1: func_graph_cloner_run.FuncGraphClonerGraph 89.79% : 0.000197s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.038751 192 0.01% : 0.000004s : 1: ForceFp32Comm 11.54% : 0.004474s : 1: add_attr 11.51% : 0.004459s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.16% : 0.000061s : 1: add_recomputation 0.02% : 0.000008s : 1: assign_add_opt 0.15% : 0.000060s : 1: auto_monad 0.06% : 0.000025s : 1: auto_monad_reorder 0.01% : 0.000003s : 1: begin_end_overlap_inline 0.01% : 0.000006s : 1: bias_add_comm_swap 1.71% : 0.000662s : 1: bootstrap 0.10% : 0.000038s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.05% : 0.000018s : 1: control_data_broadcast_order 0.03% : 0.000011s : 1: convert_after_rewriter 0.09% : 0.000035s : 1: cse_after_recomputation 0.01% : 0.000005s : 1: dataset_repeat_opt 0.03% : 0.000012s : 1: detach_backward 0.03% : 0.000010s : 1: environ_conv 0.05% : 0.000021s : 1: event_method 0.01% : 0.000005s : 1: full_micro_interleaved_order_control 0.01% : 0.000005s : 1: get_jit_bprop_graph 0.02% : 0.000009s : 1: graph_reusing 0.01% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000004s : 1: handle_group_info 0.01% : 0.000006s : 1: inline 0.02% : 0.000006s : 1: insert-virtual-dataset 0.01% : 0.000005s : 1: interleave_parallel_branches 0.01% : 0.000004s : 1: interleave_split_concat_branches 0.01% : 0.000005s : 1: label_fine_grained_interleaved_index 0.02% : 0.000007s : 1: label_micro_interleaved_index 1.34% : 0.000519s : 1: loop_unroll 0.01% : 0.000004s : 1: merge_cast_opt 0.01% : 0.000005s : 1: micro_interleaved_order_control 1.75% : 0.000678s : 1: mutable_eliminate 0.02% : 0.000007s : 1: offloading_packed_experts 0.04% : 0.000016s : 1: opt.transform.loop_unroll_optimizer 0.05% : 0.000019s : 1: opt.transform.mutable_eliminate 2.63% : 0.001021s : 78: opt.transform.opt_a 0.08% : 0.000033s : 1: opt.transform.opt_after_cconv 0.08% : 0.000032s : 1: opt.transform.opt_after_jit_grad 0.29% : 0.000111s : 28: opt.transform.opt_b 0.14% : 0.000056s : 2: opt.transform.opt_trans_graph 0.11% : 0.000041s : 4: opt.transform.symbol_engine_opt 6.85% : 0.002653s : 1: opt_a 0.35% : 0.000135s : 1: opt_after_cconv 1.60% : 0.000618s : 1: opt_after_jit_grad 0.63% : 0.000243s : 1: opt_b 13.00% : 0.005038s : 1: optimize 0.06% : 0.000025s : 1: optimize_parallel_all_gather_comm 0.02% : 0.000008s : 1: order_py_execute_after_rewriter 0.07% : 0.000027s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.02% : 0.000008s : 1: overlap_grad_ring_attention 0.01% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.01% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.01% : 0.000005s : 1: overlap_param_gather 0.03% : 0.000011s : 1: overlap_recompute_allgather_and_fa_grad 0.02% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.01% : 0.000005s : 1: overlap_recompute_comm 0.02% : 0.000008s : 1: parallel-infer-symbol 0.01% : 0.000004s : 1: parallel-infer-symbol-second 0.01% : 0.000005s : 1: partial_unused_args_eliminate 0.01% : 0.000005s : 1: pipeline_parallel_scheduler 0.01% : 0.000005s : 1: pipeline_split 0.10% : 0.000037s : 1: pre_auto_parallel 0.06% : 0.000024s : 1: py_interpret_to_execute 0.05% : 0.000020s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000004s : 1: remove_cast_before_assign_add 0.05% : 0.000021s : 1: remove_dup_value 0.95% : 0.000367s : 1: renormalize.infer 0.68% : 0.000263s : 1: renormalize.specialize 0.01% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.02% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.12% : 0.000048s : 1: rewriter_after_opt_a 0.13% : 0.000051s : 1: rewriter_before_opt_a 0.01% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.01% : 0.000005s : 1: slice_recompute_activation 0.01% : 0.000005s : 1: split_layernorm_comm 0.01% : 0.000005s : 1: split_matmul_comm_elemetwise 0.02% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.23% : 0.000089s : 1: symbol_engine_optimizer 0.23% : 0.000088s : 1: tuple_transform 42.19% : 0.016349s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_bf16[dim_type1-True-batch_size_list1-shape0-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_bf16[dim_type1-True-batch_size_list1-shape0-ms_dtype0],max_mem:196.0M [WARNING] ME(152945:281473835274032,MainProcess):2026-01-29-17:44:25.835.560 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.0446272, [21] [bootstrap]: 0.00069718 [type_inference]: 0.00674357 [event_method]: 1.184e-05 [auto_monad]: 5.11e-05 [graph_reusing]: 5.47001e-06 [inline]: 2.42001e-06 [add_attr]: 0.00439836, [1] [add_attr_with_inline]: 0.00438481, [1] [Cycle 1]: 5.812e-05, [2] [tag_attr]: 1.455e-05 [meta_addattr_fg_expand]: 3.22002e-06 [parallel-infer-symbol]: 4e-06 [pre_auto_parallel]: 3.122e-05 [insert-virtual-dataset]: 2.49999e-06 [parallel-infer-symbol-second]: 8.39995e-07 [dataset_repeat_opt]: 2.32999e-06 [pipeline_split]: 1.67999e-06 [optimize]: 0.0319499, [53] [py_interpret_to_execute]: 1.853e-05 [rewriter_before_opt_a]: 4.758e-05 [opt_a]: 0.0294525, [2] [Cycle 1]: 0.0286685, [45] [expand_dump_flag]: 2.51e-06 [switch_simplify]: 2.432e-05 [loop_unroll]: 1.241e-05 [a_1]: 0.00029385 [with_stream_mark]: 1.834e-05 [recompute_prepare]: 8.85999e-06 [updatestate_depend_eliminate]: 4.15999e-06 [updatestate_assign_eliminate]: 3.32002e-06 [updatestate_loads_eliminate]: 3.31999e-06 [parameter_eliminate]: 1.84e-06 [a_2]: 8.607e-05 [accelerated_algorithm]: 7.20998e-06 [shard]: 1.93002e-06 [meta_shard_fg_expand]: 1.98002e-06 [shard_inline]: 6.74001e-06 [merge_send_recv]: 8.70001e-06 [auto_parallel]: 6.33e-06 [parallel]: 3.017e-05 [flash_sp]: 9.52999e-06 [merge_comm]: 3.93001e-06 [allreduce_fusion]: 4.07e-06 [matmul_add_comm_reduction]: 9.55001e-06 [allreduce_slice_to_reducescatter]: 5.8001e-07 [virtual_shard_identity]: 8.74e-06 [virtual_dataset]: 7.25998e-06 [get_grad_eliminate_]: 6.46999e-06 [virtual_output]: 6.76e-06 [merge_forward]: 3.97e-06 [cell_reuse_recompute_pass]: 1.79e-06 [offload_activation]: 9.71e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.464e-05 [merge_recompute_call_nodes]: 1.45999e-06 [before_grad]: 1.095e-05 [set_forward_comm_id_for_comm_node_pass]: 3.83001e-06 [meta_fg_expand]: 3.08998e-06 [flash_sp_send_recv_attached]: 2.39001e-06 [receive_attached]: 2.35002e-06 [after_resolve]: 1.236e-05 [a_after_grad]: 1.009e-05 [renormalize]: 0.0275754 [add_forward_monad_depend]: 1.771e-05 [auto_monad_grad]: 2.89001e-06 [auto_monad_eliminator]: 2.484e-05 [cse]: 3.908e-05 [a_3]: 6.994e-05 [Cycle 2]: 0.00077118, [45] [expand_dump_flag]: 2.08998e-06 [switch_simplify]: 1.003e-05 [loop_unroll]: 6.81001e-06 [a_1]: 0.00014786 [with_stream_mark]: 1.917e-05 [recompute_prepare]: 7.11001e-06 [updatestate_depend_eliminate]: 4.12e-06 [updatestate_assign_eliminate]: 3.63999e-06 [updatestate_loads_eliminate]: 3.61999e-06 [parameter_eliminate]: 1.91998e-06 [a_2]: 8.035e-05 [accelerated_algorithm]: 7.53e-06 [shard]: 2.25002e-06 [meta_shard_fg_expand]: 2.33998e-06 [shard_inline]: 7.01999e-06 [merge_send_recv]: 9.68002e-06 [auto_parallel]: 1.043e-05 [parallel]: 9.44e-06 [flash_sp]: 4.63001e-06 [merge_comm]: 4.20999e-06 [allreduce_fusion]: 3.86001e-06 [matmul_add_comm_reduction]: 1.409e-05 [allreduce_slice_to_reducescatter]: 6.29982e-07 [virtual_shard_identity]: 8.08001e-06 [virtual_dataset]: 6.93e-06 [get_grad_eliminate_]: 7.08998e-06 [virtual_output]: 6.48998e-06 [merge_forward]: 4.62e-06 [cell_reuse_recompute_pass]: 3.97e-06 [offload_activation]: 1.498e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.67e-05 [merge_recompute_call_nodes]: 1.54e-06 [before_grad]: 1.163e-05 [set_forward_comm_id_for_comm_node_pass]: 4.03001e-06 [meta_fg_expand]: 3.27002e-06 [flash_sp_send_recv_attached]: 1.85001e-06 [receive_attached]: 1.98997e-06 [after_resolve]: 1.299e-05 [a_after_grad]: 1.095e-05 [renormalize]: 9.00181e-08 [add_forward_monad_depend]: 1.30999e-06 [auto_monad_grad]: 9.09989e-07 [auto_monad_eliminator]: 7.90998e-06 [cse]: 2.024e-05 [a_3]: 3.977e-05 [py_interpret_to_execute_after_opt_a]: 1.99e-05 [slice_cell_reuse_recomputed_activation]: 1.98997e-06 [rewriter_after_opt_a]: 4.227e-05 [convert_after_rewriter]: 7.35003e-06 [order_py_execute_after_rewriter]: 6.01e-06 [mutable_eliminate]: 0.00073515 [opt_b]: 0.0002308, [1] [Cycle 1]: 0.000223, [7] [b_1]: 0.00013607 [b_2]: 8.32e-06 [updatestate_depend_eliminate]: 6.63e-06 [updatestate_assign_eliminate]: 2.89999e-06 [updatestate_loads_eliminate]: 2.72001e-06 [renormalize]: 9.60019e-07 [cse]: 2.912e-05 [optimize_parallel_all_gather_comm]: 1.802e-05 [overlap_param_gather]: 4.22e-06 [cconv]: 2.968e-05 [loop_unroll]: 0.00046544 [opt_after_cconv]: 0.00011695, [1] [Cycle 1]: 0.00011069, [7] [c_1]: 3.207e-05 [parameter_eliminate]: 3.20998e-06 [updatestate_depend_eliminate]: 6.04001e-06 [updatestate_assign_eliminate]: 2.89001e-06 [updatestate_loads_eliminate]: 2.53e-06 [cse]: 2.361e-05 [renormalize]: 4.19997e-07 [remove_dup_value]: 1.863e-05 [tuple_transform]: 0.00010164, [1] [Cycle 1]: 9.612e-05, [4] [d_1]: 4.986e-05 [none_parameter_eliminate]: 2.47001e-06 [renormalize]: 2.80008e-07 [switch_simplify]: 8.64e-06 [partial_unused_args_eliminate]: 1.97001e-06 [add_recomputation]: 6.062e-05 [cse_after_recomputation]: 3.242e-05, [1] [Cycle 1]: 2.702e-05, [1] [cse]: 2.087e-05 [environ_conv]: 6.29001e-06 [swap_dp_allreduce_reducescatter]: 5.37001e-06 [bias_add_comm_swap]: 2.61e-06 [label_micro_interleaved_index]: 4.07998e-06 [label_fine_grained_interleaved_index]: 2.79999e-06 [merge_cast_opt]: 1.22999e-06 [slice_recompute_activation]: 2.05002e-06 [micro_interleaved_order_control]: 2.34001e-06 [assign_add_opt]: 1.42e-06 [ForceFp32Comm]: 1.14e-06 [remove_cast_before_assign_add]: 1.07e-06 [full_micro_interleaved_order_control]: 2.11998e-06 [reorder_send_recv_between_fp_bp]: 2.86999e-06 [comm_op_add_attrs]: 1.04e-06 [add_comm_op_reuse_tag]: 9.70002e-07 [interleave_split_concat_branches]: 1.14998e-06 [interleave_parallel_branches]: 1.03001e-06 [overlap_opt_shard_in_pipeline]: 3.25e-06 [overlap_opt_shard_grad_in_pipeline]: 1.94999e-06 [control_data_broadcast_order]: 1.342e-05 [grouped_pairwise_exchange_alltoall]: 1.53002e-06 [offloading_packed_experts]: 4.45e-06 [overlap_recompute_and_grad_model_parallel]: 4.51002e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.23002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.35001e-06 [overlap_recompute_comm]: 2.02001e-06 [overlap_grad_ring_attention]: 4.58999e-06 [overlap_grad_flash_sp]: 2.01e-05 [begin_end_overlap_inline]: 4.89992e-07 [split_matmul_comm_elemetwise]: 2.59001e-06 [split_layernorm_comm]: 1.89999e-06 [handle_group_info]: 9.70002e-07 [symbol_engine_optimizer]: 8.222e-05, [1] [Cycle 1]: 7.79e-05, [6] [build]: 3.00002e-06 [elim_shapecalc]: 1.156e-05 [elim_not_effective]: 1.359e-05 [opt_reshape]: 6.88e-06 [fold_const_symbol]: 9.99999e-06 [renormalize]: 1.99972e-07 [detach_backward]: 2.26998e-06 [pipeline_parallel_scheduler]: 1.52999e-06 [auto_monad_reorder]: 1.929e-05 [get_jit_bprop_graph]: 1.72999e-06 [rewriter_after_jit_bprop_graph]: 3.70998e-06 [opt_after_jit_grad]: 0.00049257 [validate]: 4.099e-05 Sums bootstrap : 0.000697s : 1.79% type_inference : 0.006744s : 17.28% event_method : 0.000012s : 0.03% auto_monad : 0.000051s : 0.13% graph_reusing : 0.000005s : 0.01% inline : 0.000002s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000015s : 0.04% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.01% parallel-infer-symbol : 0.000004s : 0.01% pre_auto_parallel : 0.000031s : 0.08% insert-virtual-dataset : 0.000002s : 0.01% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.01% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000019s : 0.05% optimize.rewriter_before_opt_a : 0.000048s : 0.12% optimize.opt_a.expand_dump_flag : 0.000005s : 0.01% optimize.opt_a.switch_simplify : 0.000034s : 0.09% optimize.opt_a.loop_unroll : 0.000019s : 0.05% optimize.opt_a.a_1 : 0.000442s : 1.13% optimize.opt_a.with_stream_mark : 0.000038s : 0.10% optimize.opt_a.recompute_prepare : 0.000016s : 0.04% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.02% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.02% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.02% optimize.opt_a.parameter_eliminate : 0.000004s : 0.01% optimize.opt_a.a_2 : 0.000166s : 0.43% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.04% optimize.opt_a.shard : 0.000004s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.01% optimize.opt_a.shard_inline : 0.000014s : 0.04% optimize.opt_a.merge_send_recv : 0.000018s : 0.05% optimize.opt_a.auto_parallel : 0.000017s : 0.04% optimize.opt_a.parallel : 0.000040s : 0.10% optimize.opt_a.flash_sp : 0.000014s : 0.04% optimize.opt_a.merge_comm : 0.000008s : 0.02% optimize.opt_a.allreduce_fusion : 0.000008s : 0.02% optimize.opt_a.matmul_add_comm_reduction : 0.000024s : 0.06% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000017s : 0.04% optimize.opt_a.virtual_dataset : 0.000014s : 0.04% optimize.opt_a.get_grad_eliminate_ : 0.000014s : 0.03% optimize.opt_a.virtual_output : 0.000013s : 0.03% optimize.opt_a.merge_forward : 0.000009s : 0.02% optimize.opt_a.cell_reuse_recompute_pass : 0.000006s : 0.01% optimize.opt_a.offload_activation : 0.000025s : 0.06% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000031s : 0.08% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.01% optimize.opt_a.before_grad : 0.000023s : 0.06% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.02% optimize.opt_a.meta_fg_expand : 0.000006s : 0.02% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.01% optimize.opt_a.receive_attached : 0.000004s : 0.01% optimize.opt_a.after_resolve : 0.000025s : 0.06% optimize.opt_a.a_after_grad : 0.000021s : 0.05% optimize.opt_a.renormalize : 0.027575s : 70.64% optimize.opt_a.add_forward_monad_depend : 0.000019s : 0.05% optimize.opt_a.auto_monad_grad : 0.000004s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000033s : 0.08% optimize.opt_a.cse : 0.000059s : 0.15% optimize.opt_a.a_3 : 0.000110s : 0.28% optimize.py_interpret_to_execute_after_opt_a : 0.000020s : 0.05% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.01% optimize.rewriter_after_opt_a : 0.000042s : 0.11% optimize.convert_after_rewriter : 0.000007s : 0.02% optimize.order_py_execute_after_rewriter : 0.000006s : 0.02% optimize.mutable_eliminate : 0.000735s : 1.88% optimize.opt_b.b_1 : 0.000136s : 0.35% optimize.opt_b.b_2 : 0.000008s : 0.02% optimize.opt_b.updatestate_depend_eliminate : 0.000007s : 0.02% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.01% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000029s : 0.07% optimize.optimize_parallel_all_gather_comm : 0.000018s : 0.05% optimize.overlap_param_gather : 0.000004s : 0.01% optimize.cconv : 0.000030s : 0.08% optimize.loop_unroll : 0.000465s : 1.19% optimize.opt_after_cconv.c_1 : 0.000032s : 0.08% optimize.opt_after_cconv.parameter_eliminate : 0.000003s : 0.01% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000006s : 0.02% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000003s : 0.01% optimize.opt_after_cconv.cse : 0.000024s : 0.06% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000019s : 0.05% optimize.tuple_transform.d_1 : 0.000050s : 0.13% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.01% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000009s : 0.02% optimize.partial_unused_args_eliminate : 0.000002s : 0.01% optimize.add_recomputation : 0.000061s : 0.16% optimize.cse_after_recomputation.cse : 0.000021s : 0.05% optimize.environ_conv : 0.000006s : 0.02% optimize.swap_dp_allreduce_reducescatter : 0.000005s : 0.01% optimize.bias_add_comm_swap : 0.000003s : 0.01% optimize.label_micro_interleaved_index : 0.000004s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.01% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.01% optimize.micro_interleaved_order_control : 0.000002s : 0.01% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.01% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.01% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000003s : 0.01% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000013s : 0.03% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000004s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.01% 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 : 0.01% optimize.overlap_grad_ring_attention : 0.000005s : 0.01% optimize.overlap_grad_flash_sp : 0.000020s : 0.05% optimize.begin_end_overlap_inline : 0.000000s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.01% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000003s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000012s : 0.03% optimize.symbol_engine_optimizer.elim_not_effective : 0.000014s : 0.03% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.02% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.03% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.01% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000019s : 0.05% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000004s : 0.01% opt_after_jit_grad : 0.000493s : 1.26% validate : 0.000041s : 0.11% Time group info: ------[substitution.] 0.000106 21 2.07% : 0.000002s : 2: substitution.elim_not_effective 1.29% : 0.000001s : 2: substitution.fold_const_symbol 6.14% : 0.000007s : 4: substitution.graph_param_transform 74.62% : 0.000079s : 1: substitution.inline 4.25% : 0.000005s : 4: substitution.j_node_and_user_rematch 5.79% : 0.000006s : 4: substitution.remove_not_recompute_node 5.84% : 0.000006s : 4: substitution.replace_old_param ------[type_inference.] 0.006687 2 94.05% : 0.006289s : 1: type_inference.infer 5.95% : 0.000398s : 1: type_inference.specialize ------[replace.] 0.000015 1 100.00% : 0.000015s : 1: replace.inline ------[match.] 0.000078 1 100.00% : 0.000078s : 1: match.inline ------[predicate.] 0.000165 967 1.09% : 0.000002s : 9: predicate.accumulaten_eliminater 0.93% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.65% : 0.000001s : 8: predicate.addn_check_dump 0.83% : 0.000001s : 9: predicate.addn_zero_filter 0.67% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.60% : 0.000004s : 17: predicate.arithmetic_simplify 0.93% : 0.000002s : 9: predicate.cast_eliminate 0.89% : 0.000001s : 8: predicate.check_bprop_eliminate 0.68% : 0.000001s : 8: predicate.compare_switch_simplify 0.23% : 0.000000s : 4: predicate.const_output_eliminate 0.83% : 0.000001s : 8: predicate.depend_value_elim 0.74% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 1.07% : 0.000002s : 9: predicate.dict_get_item_eliminator 1.03% : 0.000002s : 9: predicate.dict_set_item_eliminator 1.10% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.33% : 0.000001s : 4: predicate.elim_not_effective 0.81% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.29% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.09% : 0.000002s : 13: predicate.environ_get_add_eliminate 1.00% : 0.000002s : 13: predicate.environ_get_depend_swap 1.75% : 0.000003s : 21: predicate.environ_get_eliminate 1.01% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.82% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.66% : 0.000003s : 10: predicate.float_depend_g_call 0.66% : 0.000001s : 8: predicate.float_environ_get_switch 1.01% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.23% : 0.000000s : 4: predicate.fold_const_symbol 0.86% : 0.000001s : 8: predicate.get_grad_eliminate 0.58% : 0.000001s : 4: predicate.graph_param_transform 0.75% : 0.000001s : 8: predicate.incorporate_call 0.63% : 0.000001s : 8: predicate.incorporate_call_switch 5.85% : 0.000010s : 43: predicate.inline 0.96% : 0.000002s : 8: predicate.inline_without_move 0.42% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.34% : 0.000002s : 8: predicate.less_batch_normalization 1.71% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.09% : 0.000003s : 26: predicate.load_eliminater 1.05% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.27% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.94% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.65% : 0.000001s : 8: predicate.merge_addn 0.73% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.86% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.70% : 0.000001s : 9: predicate.minmaximum_grad 1.04% : 0.000002s : 4: predicate.mutable_eliminate 0.44% : 0.000001s : 4: predicate.opt_reshape 0.57% : 0.000001s : 4: predicate.parallel_virtual_node 1.10% : 0.000002s : 10: predicate.partial_defer_inline 1.18% : 0.000002s : 13: predicate.partial_eliminate 0.85% : 0.000001s : 9: predicate.print_const_string_wrapper 0.78% : 0.000001s : 8: predicate.reduce_all_const_elim 1.09% : 0.000002s : 9: predicate.reduce_eliminate 2.00% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.66% : 0.000001s : 8: predicate.remove_not_recompute_node 1.43% : 0.000002s : 17: predicate.replace_applicator 0.77% : 0.000001s : 8: predicate.replace_old_param 0.34% : 0.000001s : 4: predicate.reset_defer_inline 0.97% : 0.000002s : 9: predicate.reshape_eliminate 0.84% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.52% : 0.000001s : 4: predicate.row_tensor_eliminate 1.41% : 0.000002s : 8: predicate.same_eliminate 0.56% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.99% : 0.000002s : 8: predicate.shard_identity_eliminate 0.86% : 0.000001s : 8: predicate.special_op_eliminate 0.86% : 0.000001s : 8: predicate.specialize_transform 1.39% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.92% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.45% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.91% : 0.000002s : 10: predicate.switch_defer_inline 1.71% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.73% : 0.000006s : 35: predicate.switch_simplify 0.90% : 0.000001s : 9: predicate.tile_eliminate 0.83% : 0.000001s : 9: predicate.transpose_eliminate 1.52% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.72% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.31% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.17% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.60% : 0.000003s : 17: predicate.tuple_list_get_set_item_eliminator 2.63% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.58% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 2.04% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.86% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.44% : 0.000001s : 4: predicate.value_based_eliminate 0.94% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.93% : 0.000002s : 8: predicate.virtual_output_eliminate 0.40% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.47% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000259 4 7.53% : 0.000019s : 1: func_graph_cloner_run.FuncGraphClonerGraph 92.47% : 0.000239s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.109568 192 0.00% : 0.000004s : 1: ForceFp32Comm 4.02% : 0.004404s : 1: add_attr 4.01% : 0.004390s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.06% : 0.000065s : 1: add_recomputation 0.01% : 0.000008s : 1: assign_add_opt 0.05% : 0.000056s : 1: auto_monad 0.02% : 0.000023s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000005s : 1: bias_add_comm_swap 0.67% : 0.000732s : 1: bootstrap 0.03% : 0.000033s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.02% : 0.000017s : 1: control_data_broadcast_order 0.01% : 0.000011s : 1: convert_after_rewriter 0.03% : 0.000035s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.01% : 0.000006s : 1: detach_backward 0.01% : 0.000009s : 1: environ_conv 0.02% : 0.000018s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.01% : 0.000009s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000005s : 1: inline 0.01% : 0.000006s : 1: insert-virtual-dataset 0.01% : 0.000007s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.01% : 0.000006s : 1: label_fine_grained_interleaved_index 0.01% : 0.000007s : 1: label_micro_interleaved_index 0.43% : 0.000475s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.68% : 0.000746s : 1: mutable_eliminate 0.01% : 0.000007s : 1: offloading_packed_experts 0.01% : 0.000015s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000016s : 1: opt.transform.mutable_eliminate 0.80% : 0.000875s : 78: opt.transform.opt_a 0.03% : 0.000031s : 1: opt.transform.opt_after_cconv 0.02% : 0.000025s : 1: opt.transform.opt_after_jit_grad 0.10% : 0.000112s : 28: opt.transform.opt_b 0.05% : 0.000055s : 2: opt.transform.opt_trans_graph 0.03% : 0.000038s : 4: opt.transform.symbol_engine_opt 26.88% : 0.029457s : 1: opt_a 0.21% : 0.000227s : 1: opt_after_cconv 0.46% : 0.000502s : 1: opt_after_jit_grad 0.21% : 0.000234s : 1: opt_b 29.17% : 0.031956s : 1: optimize 0.02% : 0.000021s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000009s : 1: order_py_execute_after_rewriter 0.02% : 0.000024s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000007s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.01% : 0.000006s : 1: overlap_opt_shard_in_pipeline 0.01% : 0.000008s : 1: overlap_param_gather 0.01% : 0.000007s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000007s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.01% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.03% : 0.000036s : 1: pre_auto_parallel 0.03% : 0.000033s : 1: py_interpret_to_execute 0.02% : 0.000023s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.02% : 0.000024s : 1: remove_dup_value 24.78% : 0.027154s : 1: renormalize.infer 0.37% : 0.000404s : 1: renormalize.specialize 0.01% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000007s : 1: rewriter_after_jit_bprop_graph 0.04% : 0.000046s : 1: rewriter_after_opt_a 0.05% : 0.000052s : 1: rewriter_before_opt_a 0.00% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000008s : 1: swap_dp_allreduce_reducescatter 0.08% : 0.000085s : 1: symbol_engine_optimizer 0.10% : 0.000104s : 1: tuple_transform 6.17% : 0.006763s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_bf16[dim_type1-True-batch_size_list2-shape0-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_bf16[dim_type1-True-batch_size_list2-shape0-ms_dtype0],max_mem:292.0M [WARNING] ME(152945:281473835274032,MainProcess):2026-01-29-17:44:40.652.871 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.0140542, [21] [bootstrap]: 0.00053617 [type_inference]: 0.00354856 [event_method]: 9.30001e-06 [auto_monad]: 4.834e-05 [graph_reusing]: 5.33002e-06 [inline]: 2.26e-06 [add_attr]: 0.00434194, [1] [add_attr_with_inline]: 0.00432791, [1] [Cycle 1]: 5.224e-05, [2] [tag_attr]: 1.371e-05 [meta_addattr_fg_expand]: 3.23e-06 [parallel-infer-symbol]: 4.17998e-06 [pre_auto_parallel]: 3.27e-05 [insert-virtual-dataset]: 2.88e-06 [parallel-infer-symbol-second]: 7.59988e-07 [dataset_repeat_opt]: 2.21998e-06 [pipeline_split]: 1.73002e-06 [optimize]: 0.00475474, [53] [py_interpret_to_execute]: 1.778e-05 [rewriter_before_opt_a]: 4.661e-05 [opt_a]: 0.00244719, [2] [Cycle 1]: 0.00167155, [45] [expand_dump_flag]: 2.63e-06 [switch_simplify]: 2.363e-05 [loop_unroll]: 1.237e-05 [a_1]: 0.00029592 [with_stream_mark]: 1.856e-05 [recompute_prepare]: 8.3e-06 [updatestate_depend_eliminate]: 4.55001e-06 [updatestate_assign_eliminate]: 3.58e-06 [updatestate_loads_eliminate]: 3.48999e-06 [parameter_eliminate]: 1.67999e-06 [a_2]: 8.562e-05 [accelerated_algorithm]: 7.23e-06 [shard]: 2.07001e-06 [meta_shard_fg_expand]: 1.72001e-06 [shard_inline]: 6.82002e-06 [merge_send_recv]: 9.22999e-06 [auto_parallel]: 7.00998e-06 [parallel]: 2.653e-05 [flash_sp]: 9.58002e-06 [merge_comm]: 3.88001e-06 [allreduce_fusion]: 4.12998e-06 [matmul_add_comm_reduction]: 9.87001e-06 [allreduce_slice_to_reducescatter]: 8.39995e-07 [virtual_shard_identity]: 9.49e-06 [virtual_dataset]: 7.15e-06 [get_grad_eliminate_]: 6.71e-06 [virtual_output]: 6.29001e-06 [merge_forward]: 4.01001e-06 [cell_reuse_recompute_pass]: 1.25999e-06 [offload_activation]: 1.024e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.407e-05 [merge_recompute_call_nodes]: 1.49998e-06 [before_grad]: 1.082e-05 [set_forward_comm_id_for_comm_node_pass]: 3.95e-06 [meta_fg_expand]: 2.64001e-06 [flash_sp_send_recv_attached]: 2.68e-06 [receive_attached]: 2.46e-06 [after_resolve]: 1.182e-05 [a_after_grad]: 1.047e-05 [renormalize]: 0.00064145 [add_forward_monad_depend]: 1.05e-05 [auto_monad_grad]: 2.85002e-06 [auto_monad_eliminator]: 1.449e-05 [cse]: 3.956e-05 [a_3]: 4.964e-05 [Cycle 2]: 0.00076485, [45] [expand_dump_flag]: 1.00001e-06 [switch_simplify]: 8e-06 [loop_unroll]: 6.44999e-06 [a_1]: 0.0001339 [with_stream_mark]: 1.306e-05 [recompute_prepare]: 7.18e-06 [updatestate_depend_eliminate]: 3.40998e-06 [updatestate_assign_eliminate]: 3.06001e-06 [updatestate_loads_eliminate]: 3.21999e-06 [parameter_eliminate]: 1.12e-06 [a_2]: 0.00012471 [accelerated_algorithm]: 7.03e-06 [shard]: 1.19998e-06 [meta_shard_fg_expand]: 1.83002e-06 [shard_inline]: 6.71e-06 [merge_send_recv]: 7.18e-06 [auto_parallel]: 7.2e-06 [parallel]: 5.69e-06 [flash_sp]: 3.41001e-06 [merge_comm]: 3.73999e-06 [allreduce_fusion]: 3.34001e-06 [matmul_add_comm_reduction]: 1.031e-05 [allreduce_slice_to_reducescatter]: 3.20026e-07 [virtual_shard_identity]: 7.93999e-06 [virtual_dataset]: 6.50002e-06 [get_grad_eliminate_]: 6.01998e-06 [virtual_output]: 5.97999e-06 [merge_forward]: 3.88999e-06 [cell_reuse_recompute_pass]: 1.33002e-06 [offload_activation]: 1.366e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.441e-05 [merge_recompute_call_nodes]: 8.79983e-07 [before_grad]: 9.94999e-06 [set_forward_comm_id_for_comm_node_pass]: 4.67e-06 [meta_fg_expand]: 2.36e-06 [flash_sp_send_recv_attached]: 1.13001e-06 [receive_attached]: 1.49998e-06 [after_resolve]: 1.088e-05 [a_after_grad]: 1.048e-05 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 2.07001e-06 [auto_monad_grad]: 1.50001e-06 [auto_monad_eliminator]: 9.30001e-06 [cse]: 2.397e-05 [a_3]: 4.005e-05 [py_interpret_to_execute_after_opt_a]: 1.354e-05 [slice_cell_reuse_recomputed_activation]: 2.57001e-06 [rewriter_after_opt_a]: 4.15e-05 [convert_after_rewriter]: 7.29001e-06 [order_py_execute_after_rewriter]: 6.30002e-06 [mutable_eliminate]: 0.00064921 [opt_b]: 0.00023578, [1] [Cycle 1]: 0.00022793, [7] [b_1]: 0.00013646 [b_2]: 8.1e-06 [updatestate_depend_eliminate]: 8.82e-06 [updatestate_assign_eliminate]: 3.16999e-06 [updatestate_loads_eliminate]: 2.49001e-06 [renormalize]: 6.30011e-07 [cse]: 3.331e-05 [optimize_parallel_all_gather_comm]: 1.837e-05 [overlap_param_gather]: 2.02999e-06 [cconv]: 3.227e-05 [loop_unroll]: 0.0004719 [opt_after_cconv]: 0.00011959, [1] [Cycle 1]: 0.00011287, [7] [c_1]: 3.313e-05 [parameter_eliminate]: 3.78999e-06 [updatestate_depend_eliminate]: 5.86998e-06 [updatestate_assign_eliminate]: 2.68e-06 [updatestate_loads_eliminate]: 2.72001e-06 [cse]: 2.232e-05 [renormalize]: 5.69999e-07 [remove_dup_value]: 4.386e-05 [tuple_transform]: 8.704e-05, [1] [Cycle 1]: 7.687e-05, [4] [d_1]: 4.86e-05 [none_parameter_eliminate]: 1.72999e-06 [renormalize]: 1.69995e-07 [switch_simplify]: 7.35e-06 [partial_unused_args_eliminate]: 1.72999e-06 [add_recomputation]: 5.235e-05 [cse_after_recomputation]: 2.903e-05, [1] [Cycle 1]: 2.436e-05, [1] [cse]: 1.87e-05 [environ_conv]: 6.41e-06 [swap_dp_allreduce_reducescatter]: 5.84e-06 [bias_add_comm_swap]: 2.80002e-06 [label_micro_interleaved_index]: 3.81999e-06 [label_fine_grained_interleaved_index]: 2.84999e-06 [merge_cast_opt]: 1.29998e-06 [slice_recompute_activation]: 2.27999e-06 [micro_interleaved_order_control]: 2.12001e-06 [assign_add_opt]: 1.19e-06 [ForceFp32Comm]: 1.30999e-06 [remove_cast_before_assign_add]: 1.42e-06 [full_micro_interleaved_order_control]: 2.16e-06 [reorder_send_recv_between_fp_bp]: 2.79999e-06 [comm_op_add_attrs]: 1.07998e-06 [add_comm_op_reuse_tag]: 1.12e-06 [interleave_split_concat_branches]: 1.37e-06 [interleave_parallel_branches]: 9.89996e-07 [overlap_opt_shard_in_pipeline]: 1.19e-06 [overlap_opt_shard_grad_in_pipeline]: 1.67001e-06 [control_data_broadcast_order]: 1.358e-05 [grouped_pairwise_exchange_alltoall]: 1.76998e-06 [offloading_packed_experts]: 4.78001e-06 [overlap_recompute_and_grad_model_parallel]: 4.80001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.28002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.39e-06 [overlap_recompute_comm]: 2.54001e-06 [overlap_grad_ring_attention]: 4.54002e-06 [overlap_grad_flash_sp]: 2.058e-05 [begin_end_overlap_inline]: 7.09988e-07 [split_matmul_comm_elemetwise]: 2.40002e-06 [split_layernorm_comm]: 1.81998e-06 [handle_group_info]: 1.27e-06 [symbol_engine_optimizer]: 8.535e-05, [1] [Cycle 1]: 8.012e-05, [6] [build]: 3.5e-06 [elim_shapecalc]: 1.445e-05 [elim_not_effective]: 1.301e-05 [opt_reshape]: 7.58999e-06 [fold_const_symbol]: 1.019e-05 [renormalize]: 1.59984e-07 [detach_backward]: 1.99999e-06 [pipeline_parallel_scheduler]: 1.60001e-06 [auto_monad_reorder]: 1.916e-05 [get_jit_bprop_graph]: 2.21e-06 [rewriter_after_jit_bprop_graph]: 5.45001e-06 [opt_after_jit_grad]: 0.00052313 [validate]: 4.78e-05 Sums bootstrap : 0.000536s : 6.18% type_inference : 0.003549s : 40.93% event_method : 0.000009s : 0.11% auto_monad : 0.000048s : 0.56% graph_reusing : 0.000005s : 0.06% inline : 0.000002s : 0.03% add_attr.add_attr_with_inline.tag_attr : 0.000014s : 0.16% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.04% parallel-infer-symbol : 0.000004s : 0.05% pre_auto_parallel : 0.000033s : 0.38% insert-virtual-dataset : 0.000003s : 0.03% parallel-infer-symbol-second : 0.000001s : 0.01% dataset_repeat_opt : 0.000002s : 0.03% pipeline_split : 0.000002s : 0.02% optimize.py_interpret_to_execute : 0.000018s : 0.21% optimize.rewriter_before_opt_a : 0.000047s : 0.54% optimize.opt_a.expand_dump_flag : 0.000004s : 0.04% optimize.opt_a.switch_simplify : 0.000032s : 0.36% optimize.opt_a.loop_unroll : 0.000019s : 0.22% optimize.opt_a.a_1 : 0.000430s : 4.96% optimize.opt_a.with_stream_mark : 0.000032s : 0.36% optimize.opt_a.recompute_prepare : 0.000015s : 0.18% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.09% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.08% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.08% optimize.opt_a.parameter_eliminate : 0.000003s : 0.03% optimize.opt_a.a_2 : 0.000210s : 2.43% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.16% optimize.opt_a.shard : 0.000003s : 0.04% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.04% optimize.opt_a.shard_inline : 0.000014s : 0.16% optimize.opt_a.merge_send_recv : 0.000016s : 0.19% optimize.opt_a.auto_parallel : 0.000014s : 0.16% optimize.opt_a.parallel : 0.000032s : 0.37% optimize.opt_a.flash_sp : 0.000013s : 0.15% optimize.opt_a.merge_comm : 0.000008s : 0.09% optimize.opt_a.allreduce_fusion : 0.000007s : 0.09% optimize.opt_a.matmul_add_comm_reduction : 0.000020s : 0.23% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000017s : 0.20% optimize.opt_a.virtual_dataset : 0.000014s : 0.16% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.15% optimize.opt_a.virtual_output : 0.000012s : 0.14% optimize.opt_a.merge_forward : 0.000008s : 0.09% optimize.opt_a.cell_reuse_recompute_pass : 0.000003s : 0.03% optimize.opt_a.offload_activation : 0.000024s : 0.28% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000028s : 0.33% optimize.opt_a.merge_recompute_call_nodes : 0.000002s : 0.03% optimize.opt_a.before_grad : 0.000021s : 0.24% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000009s : 0.10% optimize.opt_a.meta_fg_expand : 0.000005s : 0.06% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.04% optimize.opt_a.receive_attached : 0.000004s : 0.05% optimize.opt_a.after_resolve : 0.000023s : 0.26% optimize.opt_a.a_after_grad : 0.000021s : 0.24% optimize.opt_a.renormalize : 0.000642s : 7.40% optimize.opt_a.add_forward_monad_depend : 0.000013s : 0.14% optimize.opt_a.auto_monad_grad : 0.000004s : 0.05% optimize.opt_a.auto_monad_eliminator : 0.000024s : 0.27% optimize.opt_a.cse : 0.000064s : 0.73% optimize.opt_a.a_3 : 0.000090s : 1.03% optimize.py_interpret_to_execute_after_opt_a : 0.000014s : 0.16% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.03% optimize.rewriter_after_opt_a : 0.000041s : 0.48% optimize.convert_after_rewriter : 0.000007s : 0.08% optimize.order_py_execute_after_rewriter : 0.000006s : 0.07% optimize.mutable_eliminate : 0.000649s : 7.49% optimize.opt_b.b_1 : 0.000136s : 1.57% optimize.opt_b.b_2 : 0.000008s : 0.09% optimize.opt_b.updatestate_depend_eliminate : 0.000009s : 0.10% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.04% optimize.opt_b.updatestate_loads_eliminate : 0.000002s : 0.03% optimize.opt_b.renormalize : 0.000001s : 0.01% optimize.opt_b.cse : 0.000033s : 0.38% optimize.optimize_parallel_all_gather_comm : 0.000018s : 0.21% optimize.overlap_param_gather : 0.000002s : 0.02% optimize.cconv : 0.000032s : 0.37% optimize.loop_unroll : 0.000472s : 5.44% optimize.opt_after_cconv.c_1 : 0.000033s : 0.38% optimize.opt_after_cconv.parameter_eliminate : 0.000004s : 0.04% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000006s : 0.07% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.03% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000003s : 0.03% optimize.opt_after_cconv.cse : 0.000022s : 0.26% optimize.opt_after_cconv.renormalize : 0.000001s : 0.01% optimize.remove_dup_value : 0.000044s : 0.51% optimize.tuple_transform.d_1 : 0.000049s : 0.56% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.02% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.08% optimize.partial_unused_args_eliminate : 0.000002s : 0.02% optimize.add_recomputation : 0.000052s : 0.60% optimize.cse_after_recomputation.cse : 0.000019s : 0.22% optimize.environ_conv : 0.000006s : 0.07% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.07% optimize.bias_add_comm_swap : 0.000003s : 0.03% optimize.label_micro_interleaved_index : 0.000004s : 0.04% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.03% optimize.merge_cast_opt : 0.000001s : 0.01% optimize.slice_recompute_activation : 0.000002s : 0.03% optimize.micro_interleaved_order_control : 0.000002s : 0.02% optimize.assign_add_opt : 0.000001s : 0.01% optimize.ForceFp32Comm : 0.000001s : 0.02% optimize.remove_cast_before_assign_add : 0.000001s : 0.02% optimize.full_micro_interleaved_order_control : 0.000002s : 0.02% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.03% optimize.comm_op_add_attrs : 0.000001s : 0.01% optimize.add_comm_op_reuse_tag : 0.000001s : 0.01% optimize.interleave_split_concat_branches : 0.000001s : 0.02% optimize.interleave_parallel_branches : 0.000001s : 0.01% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.01% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.02% optimize.control_data_broadcast_order : 0.000014s : 0.16% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.02% optimize.offloading_packed_experts : 0.000005s : 0.06% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.06% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.01% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.02% optimize.overlap_recompute_comm : 0.000003s : 0.03% optimize.overlap_grad_ring_attention : 0.000005s : 0.05% optimize.overlap_grad_flash_sp : 0.000021s : 0.24% optimize.begin_end_overlap_inline : 0.000001s : 0.01% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.03% optimize.split_layernorm_comm : 0.000002s : 0.02% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000003s : 0.04% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000014s : 0.17% optimize.symbol_engine_optimizer.elim_not_effective : 0.000013s : 0.15% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.09% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.12% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.02% pipeline_parallel_scheduler : 0.000002s : 0.02% auto_monad_reorder : 0.000019s : 0.22% get_jit_bprop_graph : 0.000002s : 0.03% rewriter_after_jit_bprop_graph : 0.000005s : 0.06% opt_after_jit_grad : 0.000523s : 6.03% validate : 0.000048s : 0.55% Time group info: ------[substitution.] 0.000101 21 1.98% : 0.000002s : 2: substitution.elim_not_effective 1.50% : 0.000002s : 2: substitution.fold_const_symbol 6.24% : 0.000006s : 4: substitution.graph_param_transform 77.40% : 0.000078s : 1: substitution.inline 3.46% : 0.000003s : 4: substitution.j_node_and_user_rematch 4.97% : 0.000005s : 4: substitution.remove_not_recompute_node 4.44% : 0.000004s : 4: substitution.replace_old_param ------[type_inference.] 0.003499 2 89.40% : 0.003128s : 1: type_inference.infer 10.60% : 0.000371s : 1: type_inference.specialize ------[replace.] 0.000016 1 100.00% : 0.000016s : 1: replace.inline ------[match.] 0.000077 1 100.00% : 0.000077s : 1: match.inline ------[predicate.] 0.000156 967 0.76% : 0.000001s : 9: predicate.accumulaten_eliminater 1.14% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.63% : 0.000001s : 8: predicate.addn_check_dump 0.79% : 0.000001s : 9: predicate.addn_zero_filter 0.69% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.05% : 0.000003s : 17: predicate.arithmetic_simplify 0.77% : 0.000001s : 9: predicate.cast_eliminate 0.77% : 0.000001s : 8: predicate.check_bprop_eliminate 0.65% : 0.000001s : 8: predicate.compare_switch_simplify 0.26% : 0.000000s : 4: predicate.const_output_eliminate 0.81% : 0.000001s : 8: predicate.depend_value_elim 0.81% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.95% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.85% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.50% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.25% : 0.000000s : 4: predicate.elim_not_effective 0.67% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.13% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.18% : 0.000002s : 13: predicate.environ_get_add_eliminate 1.08% : 0.000002s : 13: predicate.environ_get_depend_swap 1.78% : 0.000003s : 21: predicate.environ_get_eliminate 1.05% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.79% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.73% : 0.000003s : 10: predicate.float_depend_g_call 0.63% : 0.000001s : 8: predicate.float_environ_get_switch 0.97% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.22% : 0.000000s : 4: predicate.fold_const_symbol 0.83% : 0.000001s : 8: predicate.get_grad_eliminate 0.37% : 0.000001s : 4: predicate.graph_param_transform 0.80% : 0.000001s : 8: predicate.incorporate_call 0.65% : 0.000001s : 8: predicate.incorporate_call_switch 6.12% : 0.000010s : 43: predicate.inline 1.04% : 0.000002s : 8: predicate.inline_without_move 0.44% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.09% : 0.000002s : 8: predicate.less_batch_normalization 1.70% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.18% : 0.000003s : 26: predicate.load_eliminater 1.25% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.42% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.97% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.67% : 0.000001s : 8: predicate.merge_addn 0.75% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.75% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.68% : 0.000001s : 9: predicate.minmaximum_grad 1.56% : 0.000002s : 4: predicate.mutable_eliminate 0.52% : 0.000001s : 4: predicate.opt_reshape 0.66% : 0.000001s : 4: predicate.parallel_virtual_node 1.08% : 0.000002s : 10: predicate.partial_defer_inline 1.22% : 0.000002s : 13: predicate.partial_eliminate 0.78% : 0.000001s : 9: predicate.print_const_string_wrapper 0.72% : 0.000001s : 8: predicate.reduce_all_const_elim 1.02% : 0.000002s : 9: predicate.reduce_eliminate 2.20% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.72% : 0.000001s : 8: predicate.remove_not_recompute_node 1.19% : 0.000002s : 17: predicate.replace_applicator 0.70% : 0.000001s : 8: predicate.replace_old_param 0.30% : 0.000000s : 4: predicate.reset_defer_inline 0.95% : 0.000001s : 9: predicate.reshape_eliminate 0.76% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.66% : 0.000001s : 4: predicate.row_tensor_eliminate 0.95% : 0.000001s : 8: predicate.same_eliminate 0.58% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.15% : 0.000002s : 8: predicate.shard_identity_eliminate 1.00% : 0.000002s : 8: predicate.special_op_eliminate 1.13% : 0.000002s : 8: predicate.specialize_transform 1.25% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.95% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.56% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.90% : 0.000001s : 10: predicate.switch_defer_inline 1.59% : 0.000002s : 18: predicate.switch_layer_defer_inline 3.64% : 0.000006s : 35: predicate.switch_simplify 0.74% : 0.000001s : 9: predicate.tile_eliminate 0.83% : 0.000001s : 9: predicate.transpose_eliminate 1.61% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.64% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.37% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.50% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.58% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.47% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.50% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 2.04% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.99% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.53% : 0.000001s : 4: predicate.value_based_eliminate 1.08% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.76% : 0.000001s : 8: predicate.virtual_output_eliminate 0.38% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.65% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000209 4 9.86% : 0.000021s : 1: func_graph_cloner_run.FuncGraphClonerGraph 90.14% : 0.000188s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.024783 192 0.02% : 0.000004s : 1: ForceFp32Comm 17.54% : 0.004348s : 1: add_attr 17.48% : 0.004332s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.23% : 0.000056s : 1: add_recomputation 0.02% : 0.000004s : 1: assign_add_opt 0.22% : 0.000053s : 1: auto_monad 0.11% : 0.000026s : 1: auto_monad_reorder 0.01% : 0.000003s : 1: begin_end_overlap_inline 0.02% : 0.000006s : 1: bias_add_comm_swap 2.29% : 0.000567s : 1: bootstrap 0.14% : 0.000036s : 1: cconv 0.02% : 0.000004s : 1: comm_op_add_attrs 0.07% : 0.000017s : 1: control_data_broadcast_order 0.04% : 0.000011s : 1: convert_after_rewriter 0.13% : 0.000032s : 1: cse_after_recomputation 0.02% : 0.000005s : 1: dataset_repeat_opt 0.02% : 0.000005s : 1: detach_backward 0.04% : 0.000009s : 1: environ_conv 0.07% : 0.000016s : 1: event_method 0.02% : 0.000005s : 1: full_micro_interleaved_order_control 0.02% : 0.000006s : 1: get_jit_bprop_graph 0.04% : 0.000009s : 1: graph_reusing 0.02% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.02% : 0.000004s : 1: handle_group_info 0.02% : 0.000005s : 1: inline 0.03% : 0.000006s : 1: insert-virtual-dataset 0.02% : 0.000004s : 1: interleave_parallel_branches 0.02% : 0.000004s : 1: interleave_split_concat_branches 0.02% : 0.000006s : 1: label_fine_grained_interleaved_index 0.03% : 0.000007s : 1: label_micro_interleaved_index 1.94% : 0.000481s : 1: loop_unroll 0.02% : 0.000004s : 1: merge_cast_opt 0.02% : 0.000005s : 1: micro_interleaved_order_control 2.66% : 0.000660s : 1: mutable_eliminate 0.03% : 0.000008s : 1: offloading_packed_experts 0.06% : 0.000015s : 1: opt.transform.loop_unroll_optimizer 0.08% : 0.000020s : 1: opt.transform.mutable_eliminate 3.37% : 0.000836s : 78: opt.transform.opt_a 0.13% : 0.000032s : 1: opt.transform.opt_after_cconv 0.11% : 0.000028s : 1: opt.transform.opt_after_jit_grad 0.46% : 0.000114s : 28: opt.transform.opt_b 0.22% : 0.000054s : 2: opt.transform.opt_trans_graph 0.16% : 0.000039s : 4: opt.transform.symbol_engine_opt 9.89% : 0.002450s : 1: opt_a 0.50% : 0.000123s : 1: opt_after_cconv 2.16% : 0.000534s : 1: opt_after_jit_grad 0.97% : 0.000240s : 1: opt_b 19.21% : 0.004760s : 1: optimize 0.09% : 0.000022s : 1: optimize_parallel_all_gather_comm 0.04% : 0.000009s : 1: order_py_execute_after_rewriter 0.10% : 0.000024s : 1: overlap_grad_flash_sp 0.02% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.06% : 0.000015s : 1: overlap_grad_ring_attention 0.02% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.02% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.02% : 0.000005s : 1: overlap_param_gather 0.02% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.03% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.02% : 0.000006s : 1: overlap_recompute_comm 0.03% : 0.000008s : 1: parallel-infer-symbol 0.02% : 0.000004s : 1: parallel-infer-symbol-second 0.02% : 0.000005s : 1: partial_unused_args_eliminate 0.02% : 0.000005s : 1: pipeline_parallel_scheduler 0.02% : 0.000005s : 1: pipeline_split 0.15% : 0.000037s : 1: pre_auto_parallel 0.09% : 0.000021s : 1: py_interpret_to_execute 0.07% : 0.000017s : 1: py_interpret_to_execute_after_opt_a 0.04% : 0.000010s : 1: remove_cast_before_assign_add 0.19% : 0.000048s : 1: remove_dup_value 1.53% : 0.000379s : 1: renormalize.infer 1.03% : 0.000255s : 1: renormalize.specialize 0.02% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.03% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.19% : 0.000047s : 1: rewriter_after_opt_a 0.20% : 0.000051s : 1: rewriter_before_opt_a 0.02% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.02% : 0.000005s : 1: slice_recompute_activation 0.02% : 0.000004s : 1: split_layernorm_comm 0.02% : 0.000005s : 1: split_matmul_comm_elemetwise 0.03% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.35% : 0.000088s : 1: symbol_engine_optimizer 0.36% : 0.000090s : 1: tuple_transform 14.39% : 0.003566s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_bf16[dim_type1-False-batch_size_list0-shape0-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_bf16[dim_type1-False-batch_size_list0-shape0-ms_dtype0],max_mem:292.0M [WARNING] ME(152945:281473835274032,MainProcess):2026-01-29-17:44:47.632.280 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.030701, [21] [bootstrap]: 0.00059716 [type_inference]: 0.00354513 [event_method]: 1.069e-05 [auto_monad]: 5.352e-05 [graph_reusing]: 5.00999e-06 [inline]: 2.71e-06 [add_attr]: 0.00443427, [1] [add_attr_with_inline]: 0.00442028, [1] [Cycle 1]: 5.568e-05, [2] [tag_attr]: 1.423e-05 [meta_addattr_fg_expand]: 3.51001e-06 [parallel-infer-symbol]: 3.67998e-06 [pre_auto_parallel]: 3.02e-05 [insert-virtual-dataset]: 2.65002e-06 [parallel-infer-symbol-second]: 9.5999e-07 [dataset_repeat_opt]: 2.19999e-06 [pipeline_split]: 2.02999e-06 [optimize]: 0.0209536, [53] [py_interpret_to_execute]: 1.828e-05 [rewriter_before_opt_a]: 6.178e-05 [opt_a]: 0.00232543, [2] [Cycle 1]: 0.00161228, [45] [expand_dump_flag]: 2.76999e-06 [switch_simplify]: 2.428e-05 [loop_unroll]: 1.271e-05 [a_1]: 0.0002995 [with_stream_mark]: 2.16e-05 [recompute_prepare]: 8.80001e-06 [updatestate_depend_eliminate]: 4.17e-06 [updatestate_assign_eliminate]: 4.33001e-06 [updatestate_loads_eliminate]: 3.46999e-06 [parameter_eliminate]: 1.97001e-06 [a_2]: 8.628e-05 [accelerated_algorithm]: 7.18e-06 [shard]: 2.27999e-06 [meta_shard_fg_expand]: 1.98002e-06 [shard_inline]: 6.68e-06 [merge_send_recv]: 8.38999e-06 [auto_parallel]: 6.63e-06 [parallel]: 2.597e-05 [flash_sp]: 8.70999e-06 [merge_comm]: 3.91001e-06 [allreduce_fusion]: 3.7e-06 [matmul_add_comm_reduction]: 9.63002e-06 [allreduce_slice_to_reducescatter]: 1.00999e-06 [virtual_shard_identity]: 8.02998e-06 [virtual_dataset]: 7.41001e-06 [get_grad_eliminate_]: 6.89999e-06 [virtual_output]: 7.05e-06 [merge_forward]: 4.07e-06 [cell_reuse_recompute_pass]: 1.67999e-06 [offload_activation]: 1.043e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.403e-05 [merge_recompute_call_nodes]: 1.39998e-06 [before_grad]: 1.07e-05 [set_forward_comm_id_for_comm_node_pass]: 3.91001e-06 [meta_fg_expand]: 2.77002e-06 [flash_sp_send_recv_attached]: 2.66e-06 [receive_attached]: 2.39999e-06 [after_resolve]: 1.207e-05 [a_after_grad]: 1.056e-05 [renormalize]: 0.00059117 [add_forward_monad_depend]: 8.60001e-06 [auto_monad_grad]: 2.48998e-06 [auto_monad_eliminator]: 1.456e-05 [cse]: 3.606e-05 [a_3]: 4.959e-05 [Cycle 2]: 0.00070215, [45] [expand_dump_flag]: 1.20001e-06 [switch_simplify]: 7.95e-06 [loop_unroll]: 6.51999e-06 [a_1]: 0.00013577 [with_stream_mark]: 1.244e-05 [recompute_prepare]: 7.13e-06 [updatestate_depend_eliminate]: 3.73001e-06 [updatestate_assign_eliminate]: 2.74001e-06 [updatestate_loads_eliminate]: 2.39999e-06 [parameter_eliminate]: 9.80013e-07 [a_2]: 7.847e-05 [accelerated_algorithm]: 6.43e-06 [shard]: 1.15001e-06 [meta_shard_fg_expand]: 1.67001e-06 [shard_inline]: 6.39001e-06 [merge_send_recv]: 5.65001e-06 [auto_parallel]: 5.59998e-06 [parallel]: 5.37001e-06 [flash_sp]: 3.26001e-06 [merge_comm]: 3.47002e-06 [allreduce_fusion]: 3.14001e-06 [matmul_add_comm_reduction]: 9.32999e-06 [allreduce_slice_to_reducescatter]: 5.50004e-07 [virtual_shard_identity]: 7.26999e-06 [virtual_dataset]: 6.11998e-06 [get_grad_eliminate_]: 6.24999e-06 [virtual_output]: 5.91e-06 [merge_forward]: 3.06001e-06 [cell_reuse_recompute_pass]: 1.29e-06 [offload_activation]: 1.305e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.371e-05 [merge_recompute_call_nodes]: 8.30012e-07 [before_grad]: 9.64e-06 [set_forward_comm_id_for_comm_node_pass]: 4.23999e-06 [meta_fg_expand]: 2.88e-06 [flash_sp_send_recv_attached]: 9.50007e-07 [receive_attached]: 1.05999e-06 [after_resolve]: 1.089e-05 [a_after_grad]: 9.39e-06 [renormalize]: 5.9983e-08 [add_forward_monad_depend]: 1.15999e-06 [auto_monad_grad]: 8.80013e-07 [auto_monad_eliminator]: 6.89001e-06 [cse]: 2.644e-05 [a_3]: 3.905e-05 [py_interpret_to_execute_after_opt_a]: 1.047e-05 [slice_cell_reuse_recomputed_activation]: 2.44001e-06 [rewriter_after_opt_a]: 4.231e-05 [convert_after_rewriter]: 8.35999e-06 [order_py_execute_after_rewriter]: 5.33002e-06 [mutable_eliminate]: 0.000625 [opt_b]: 0.00022303, [1] [Cycle 1]: 0.000216, [7] [b_1]: 0.00013255 [b_2]: 8.37e-06 [updatestate_depend_eliminate]: 5.79e-06 [updatestate_assign_eliminate]: 2.60002e-06 [updatestate_loads_eliminate]: 2.54999e-06 [renormalize]: 4.80009e-07 [cse]: 2.864e-05 [optimize_parallel_all_gather_comm]: 1.895e-05 [overlap_param_gather]: 1.88002e-06 [cconv]: 2.734e-05 [loop_unroll]: 0.00046042 [opt_after_cconv]: 0.00011834, [1] [Cycle 1]: 0.00011227, [7] [c_1]: 3.347e-05 [parameter_eliminate]: 3.31001e-06 [updatestate_depend_eliminate]: 5.96998e-06 [updatestate_assign_eliminate]: 2.71e-06 [updatestate_loads_eliminate]: 2.46e-06 [cse]: 2.31e-05 [renormalize]: 4.89992e-07 [remove_dup_value]: 0.0162353 [tuple_transform]: 0.00014886, [1] [Cycle 1]: 0.00012718, [4] [d_1]: 8.22e-05 [none_parameter_eliminate]: 5.29998e-06 [renormalize]: 1.13001e-06 [switch_simplify]: 9.47001e-06 [partial_unused_args_eliminate]: 2.44001e-06 [add_recomputation]: 7.376e-05 [cse_after_recomputation]: 5.395e-05, [1] [Cycle 1]: 4.783e-05, [1] [cse]: 4.093e-05 [environ_conv]: 7.62998e-06 [swap_dp_allreduce_reducescatter]: 7.82998e-06 [bias_add_comm_swap]: 3.31999e-06 [label_micro_interleaved_index]: 6.41e-06 [label_fine_grained_interleaved_index]: 3.21001e-06 [merge_cast_opt]: 1.50001e-06 [slice_recompute_activation]: 2.21e-06 [micro_interleaved_order_control]: 2.98e-06 [assign_add_opt]: 1.74e-06 [ForceFp32Comm]: 9.5999e-07 [remove_cast_before_assign_add]: 1.03001e-06 [full_micro_interleaved_order_control]: 2.18002e-06 [reorder_send_recv_between_fp_bp]: 2.79001e-06 [comm_op_add_attrs]: 1.35001e-06 [add_comm_op_reuse_tag]: 9.90025e-07 [interleave_split_concat_branches]: 1.57999e-06 [interleave_parallel_branches]: 1.01002e-06 [overlap_opt_shard_in_pipeline]: 1.79e-06 [overlap_opt_shard_grad_in_pipeline]: 1.84e-06 [control_data_broadcast_order]: 1.87e-05 [grouped_pairwise_exchange_alltoall]: 1.38002e-06 [offloading_packed_experts]: 4.78001e-06 [overlap_recompute_and_grad_model_parallel]: 5.02e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.21002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.19e-06 [overlap_recompute_comm]: 2.24001e-06 [overlap_grad_ring_attention]: 4.60999e-06 [overlap_grad_flash_sp]: 2.345e-05 [begin_end_overlap_inline]: 5.09986e-07 [split_matmul_comm_elemetwise]: 2.22001e-06 [split_layernorm_comm]: 1.81e-06 [handle_group_info]: 1.49003e-06 [symbol_engine_optimizer]: 9.633e-05, [1] [Cycle 1]: 8.994e-05, [6] [build]: 3.56001e-06 [elim_shapecalc]: 1.481e-05 [elim_not_effective]: 1.577e-05 [opt_reshape]: 1.112e-05 [fold_const_symbol]: 1.12e-05 [renormalize]: 4.39992e-07 [detach_backward]: 2.65002e-06 [pipeline_parallel_scheduler]: 1.39e-06 [auto_monad_reorder]: 2.072e-05 [get_jit_bprop_graph]: 1.84e-06 [rewriter_after_jit_bprop_graph]: 6.36e-06 [opt_after_jit_grad]: 0.00080736 [validate]: 4.845e-05 Sums bootstrap : 0.000597s : 2.37% type_inference : 0.003545s : 14.08% event_method : 0.000011s : 0.04% auto_monad : 0.000054s : 0.21% graph_reusing : 0.000005s : 0.02% inline : 0.000003s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000014s : 0.06% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.01% parallel-infer-symbol : 0.000004s : 0.01% pre_auto_parallel : 0.000030s : 0.12% insert-virtual-dataset : 0.000003s : 0.01% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.01% pipeline_split : 0.000002s : 0.01% optimize.py_interpret_to_execute : 0.000018s : 0.07% optimize.rewriter_before_opt_a : 0.000062s : 0.25% optimize.opt_a.expand_dump_flag : 0.000004s : 0.02% optimize.opt_a.switch_simplify : 0.000032s : 0.13% optimize.opt_a.loop_unroll : 0.000019s : 0.08% optimize.opt_a.a_1 : 0.000435s : 1.73% optimize.opt_a.with_stream_mark : 0.000034s : 0.14% optimize.opt_a.recompute_prepare : 0.000016s : 0.06% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.03% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.03% optimize.opt_a.updatestate_loads_eliminate : 0.000006s : 0.02% optimize.opt_a.parameter_eliminate : 0.000003s : 0.01% optimize.opt_a.a_2 : 0.000165s : 0.65% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.05% optimize.opt_a.shard : 0.000003s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.01% optimize.opt_a.shard_inline : 0.000013s : 0.05% optimize.opt_a.merge_send_recv : 0.000014s : 0.06% optimize.opt_a.auto_parallel : 0.000012s : 0.05% optimize.opt_a.parallel : 0.000031s : 0.12% optimize.opt_a.flash_sp : 0.000012s : 0.05% optimize.opt_a.merge_comm : 0.000007s : 0.03% optimize.opt_a.allreduce_fusion : 0.000007s : 0.03% optimize.opt_a.matmul_add_comm_reduction : 0.000019s : 0.08% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000015s : 0.06% optimize.opt_a.virtual_dataset : 0.000014s : 0.05% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.05% optimize.opt_a.virtual_output : 0.000013s : 0.05% optimize.opt_a.merge_forward : 0.000007s : 0.03% optimize.opt_a.cell_reuse_recompute_pass : 0.000003s : 0.01% optimize.opt_a.offload_activation : 0.000023s : 0.09% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000028s : 0.11% optimize.opt_a.merge_recompute_call_nodes : 0.000002s : 0.01% optimize.opt_a.before_grad : 0.000020s : 0.08% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.03% optimize.opt_a.meta_fg_expand : 0.000006s : 0.02% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.01% optimize.opt_a.receive_attached : 0.000003s : 0.01% optimize.opt_a.after_resolve : 0.000023s : 0.09% optimize.opt_a.a_after_grad : 0.000020s : 0.08% optimize.opt_a.renormalize : 0.000591s : 2.35% optimize.opt_a.add_forward_monad_depend : 0.000010s : 0.04% optimize.opt_a.auto_monad_grad : 0.000003s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000021s : 0.09% optimize.opt_a.cse : 0.000063s : 0.25% optimize.opt_a.a_3 : 0.000089s : 0.35% optimize.py_interpret_to_execute_after_opt_a : 0.000010s : 0.04% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.01% optimize.rewriter_after_opt_a : 0.000042s : 0.17% optimize.convert_after_rewriter : 0.000008s : 0.03% optimize.order_py_execute_after_rewriter : 0.000005s : 0.02% optimize.mutable_eliminate : 0.000625s : 2.48% optimize.opt_b.b_1 : 0.000133s : 0.53% optimize.opt_b.b_2 : 0.000008s : 0.03% optimize.opt_b.updatestate_depend_eliminate : 0.000006s : 0.02% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.01% optimize.opt_b.renormalize : 0.000000s : 0.00% optimize.opt_b.cse : 0.000029s : 0.11% optimize.optimize_parallel_all_gather_comm : 0.000019s : 0.08% optimize.overlap_param_gather : 0.000002s : 0.01% optimize.cconv : 0.000027s : 0.11% optimize.loop_unroll : 0.000460s : 1.83% optimize.opt_after_cconv.c_1 : 0.000033s : 0.13% optimize.opt_after_cconv.parameter_eliminate : 0.000003s : 0.01% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000006s : 0.02% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000002s : 0.01% optimize.opt_after_cconv.cse : 0.000023s : 0.09% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.016235s : 64.50% optimize.tuple_transform.d_1 : 0.000082s : 0.33% optimize.tuple_transform.none_parameter_eliminate : 0.000005s : 0.02% optimize.tuple_transform.renormalize : 0.000001s : 0.00% optimize.tuple_transform.switch_simplify : 0.000009s : 0.04% optimize.partial_unused_args_eliminate : 0.000002s : 0.01% optimize.add_recomputation : 0.000074s : 0.29% optimize.cse_after_recomputation.cse : 0.000041s : 0.16% optimize.environ_conv : 0.000008s : 0.03% optimize.swap_dp_allreduce_reducescatter : 0.000008s : 0.03% optimize.bias_add_comm_swap : 0.000003s : 0.01% optimize.label_micro_interleaved_index : 0.000006s : 0.03% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.01% optimize.merge_cast_opt : 0.000002s : 0.01% optimize.slice_recompute_activation : 0.000002s : 0.01% optimize.micro_interleaved_order_control : 0.000003s : 0.01% optimize.assign_add_opt : 0.000002s : 0.01% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.01% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.01% optimize.comm_op_add_attrs : 0.000001s : 0.01% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000002s : 0.01% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.01% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.01% optimize.control_data_broadcast_order : 0.000019s : 0.07% optimize.grouped_pairwise_exchange_alltoall : 0.000001s : 0.01% optimize.offloading_packed_experts : 0.000005s : 0.02% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.02% 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 : 0.01% optimize.overlap_grad_ring_attention : 0.000005s : 0.02% optimize.overlap_grad_flash_sp : 0.000023s : 0.09% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.01% optimize.split_layernorm_comm : 0.000002s : 0.01% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000004s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000015s : 0.06% optimize.symbol_engine_optimizer.elim_not_effective : 0.000016s : 0.06% optimize.symbol_engine_optimizer.opt_reshape : 0.000011s : 0.04% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000011s : 0.04% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.01% pipeline_parallel_scheduler : 0.000001s : 0.01% auto_monad_reorder : 0.000021s : 0.08% get_jit_bprop_graph : 0.000002s : 0.01% rewriter_after_jit_bprop_graph : 0.000006s : 0.03% opt_after_jit_grad : 0.000807s : 3.21% validate : 0.000048s : 0.19% Time group info: ------[substitution.] 0.000104 21 2.08% : 0.000002s : 2: substitution.elim_not_effective 1.36% : 0.000001s : 2: substitution.fold_const_symbol 8.36% : 0.000009s : 4: substitution.graph_param_transform 75.26% : 0.000079s : 1: substitution.inline 3.28% : 0.000003s : 4: substitution.j_node_and_user_rematch 5.00% : 0.000005s : 4: substitution.remove_not_recompute_node 4.66% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.003492 2 89.35% : 0.003120s : 1: type_inference.infer 10.65% : 0.000372s : 1: type_inference.specialize ------[replace.] 0.000016 1 100.00% : 0.000016s : 1: replace.inline ------[match.] 0.000077 1 100.00% : 0.000077s : 1: match.inline ------[predicate.] 0.000164 967 0.78% : 0.000001s : 9: predicate.accumulaten_eliminater 1.34% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.63% : 0.000001s : 8: predicate.addn_check_dump 0.67% : 0.000001s : 9: predicate.addn_zero_filter 0.73% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.03% : 0.000003s : 17: predicate.arithmetic_simplify 0.79% : 0.000001s : 9: predicate.cast_eliminate 0.73% : 0.000001s : 8: predicate.check_bprop_eliminate 0.73% : 0.000001s : 8: predicate.compare_switch_simplify 0.24% : 0.000000s : 4: predicate.const_output_eliminate 0.70% : 0.000001s : 8: predicate.depend_value_elim 0.81% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 1.00% : 0.000002s : 9: predicate.dict_get_item_eliminator 0.72% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.25% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.56% : 0.000001s : 4: predicate.elim_not_effective 0.71% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.20% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.99% : 0.000002s : 13: predicate.environ_get_add_eliminate 1.01% : 0.000002s : 13: predicate.environ_get_depend_swap 1.70% : 0.000003s : 21: predicate.environ_get_eliminate 0.98% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.84% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.53% : 0.000002s : 10: predicate.float_depend_g_call 0.85% : 0.000001s : 8: predicate.float_environ_get_switch 0.90% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.26% : 0.000000s : 4: predicate.fold_const_symbol 1.11% : 0.000002s : 8: predicate.get_grad_eliminate 0.65% : 0.000001s : 4: predicate.graph_param_transform 0.74% : 0.000001s : 8: predicate.incorporate_call 0.63% : 0.000001s : 8: predicate.incorporate_call_switch 6.34% : 0.000010s : 43: predicate.inline 0.95% : 0.000002s : 8: predicate.inline_without_move 0.42% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.03% : 0.000002s : 8: predicate.less_batch_normalization 1.93% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.30% : 0.000004s : 26: predicate.load_eliminater 1.22% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.28% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.97% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.74% : 0.000001s : 8: predicate.merge_addn 0.70% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.78% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.66% : 0.000001s : 9: predicate.minmaximum_grad 1.58% : 0.000003s : 4: predicate.mutable_eliminate 0.53% : 0.000001s : 4: predicate.opt_reshape 0.42% : 0.000001s : 4: predicate.parallel_virtual_node 1.06% : 0.000002s : 10: predicate.partial_defer_inline 1.39% : 0.000002s : 13: predicate.partial_eliminate 0.79% : 0.000001s : 9: predicate.print_const_string_wrapper 0.73% : 0.000001s : 8: predicate.reduce_all_const_elim 1.01% : 0.000002s : 9: predicate.reduce_eliminate 2.03% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.62% : 0.000001s : 8: predicate.remove_not_recompute_node 1.23% : 0.000002s : 17: predicate.replace_applicator 0.73% : 0.000001s : 8: predicate.replace_old_param 0.31% : 0.000001s : 4: predicate.reset_defer_inline 0.93% : 0.000002s : 9: predicate.reshape_eliminate 0.89% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.69% : 0.000001s : 4: predicate.row_tensor_eliminate 0.90% : 0.000001s : 8: predicate.same_eliminate 0.60% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.93% : 0.000002s : 8: predicate.shard_identity_eliminate 1.15% : 0.000002s : 8: predicate.special_op_eliminate 0.89% : 0.000001s : 8: predicate.specialize_transform 1.21% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.98% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.43% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.83% : 0.000001s : 10: predicate.switch_defer_inline 1.51% : 0.000002s : 18: predicate.switch_layer_defer_inline 3.81% : 0.000006s : 35: predicate.switch_simplify 0.73% : 0.000001s : 9: predicate.tile_eliminate 0.76% : 0.000001s : 9: predicate.transpose_eliminate 1.57% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.83% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.45% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.53% : 0.000006s : 25: predicate.tuple_list_get_item_eliminator 1.54% : 0.000003s : 17: predicate.tuple_list_get_set_item_eliminator 2.59% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.71% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.95% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.80% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.48% : 0.000001s : 4: predicate.value_based_eliminate 0.96% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.81% : 0.000001s : 8: predicate.virtual_output_eliminate 0.31% : 0.000000s : 4: predicate.virtual_view_grad_eliminate 0.70% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000195 4 9.44% : 0.000018s : 1: func_graph_cloner_run.FuncGraphClonerGraph 90.56% : 0.000177s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.057703 192 0.01% : 0.000004s : 1: ForceFp32Comm 7.69% : 0.004440s : 1: add_attr 7.67% : 0.004425s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.14% : 0.000079s : 1: add_recomputation 0.01% : 0.000004s : 1: assign_add_opt 0.10% : 0.000059s : 1: auto_monad 0.05% : 0.000032s : 1: auto_monad_reorder 0.01% : 0.000003s : 1: begin_end_overlap_inline 0.01% : 0.000006s : 1: bias_add_comm_swap 1.09% : 0.000627s : 1: bootstrap 0.05% : 0.000031s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.04% : 0.000022s : 1: control_data_broadcast_order 0.02% : 0.000012s : 1: convert_after_rewriter 0.10% : 0.000058s : 1: cse_after_recomputation 0.01% : 0.000005s : 1: dataset_repeat_opt 0.01% : 0.000006s : 1: detach_backward 0.02% : 0.000011s : 1: environ_conv 0.03% : 0.000016s : 1: event_method 0.01% : 0.000005s : 1: full_micro_interleaved_order_control 0.01% : 0.000005s : 1: get_jit_bprop_graph 0.01% : 0.000008s : 1: graph_reusing 0.01% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000005s : 1: handle_group_info 0.01% : 0.000006s : 1: inline 0.01% : 0.000006s : 1: insert-virtual-dataset 0.01% : 0.000004s : 1: interleave_parallel_branches 0.01% : 0.000004s : 1: interleave_split_concat_branches 0.01% : 0.000007s : 1: label_fine_grained_interleaved_index 0.02% : 0.000009s : 1: label_micro_interleaved_index 0.81% : 0.000470s : 1: loop_unroll 0.01% : 0.000004s : 1: merge_cast_opt 0.01% : 0.000006s : 1: micro_interleaved_order_control 1.10% : 0.000635s : 1: mutable_eliminate 0.01% : 0.000008s : 1: offloading_packed_experts 0.03% : 0.000015s : 1: opt.transform.loop_unroll_optimizer 0.03% : 0.000018s : 1: opt.transform.mutable_eliminate 1.46% : 0.000840s : 78: opt.transform.opt_a 0.06% : 0.000032s : 1: opt.transform.opt_after_cconv 0.05% : 0.000031s : 1: opt.transform.opt_after_jit_grad 0.19% : 0.000111s : 28: opt.transform.opt_b 0.14% : 0.000083s : 2: opt.transform.opt_trans_graph 0.08% : 0.000049s : 4: opt.transform.symbol_engine_opt 4.04% : 0.002328s : 1: opt_a 0.21% : 0.000122s : 1: opt_after_cconv 1.42% : 0.000821s : 1: opt_after_jit_grad 0.39% : 0.000226s : 1: opt_b 36.32% : 0.020960s : 1: optimize 0.04% : 0.000023s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000008s : 1: order_py_execute_after_rewriter 0.05% : 0.000027s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.02% : 0.000012s : 1: overlap_grad_ring_attention 0.01% : 0.000007s : 1: overlap_opt_shard_grad_in_pipeline 0.01% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.01% : 0.000005s : 1: overlap_param_gather 0.01% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.01% : 0.000005s : 1: overlap_recompute_comm 0.01% : 0.000008s : 1: parallel-infer-symbol 0.01% : 0.000004s : 1: parallel-infer-symbol-second 0.01% : 0.000006s : 1: partial_unused_args_eliminate 0.01% : 0.000005s : 1: pipeline_parallel_scheduler 0.01% : 0.000005s : 1: pipeline_split 0.06% : 0.000035s : 1: pre_auto_parallel 0.04% : 0.000022s : 1: py_interpret_to_execute 0.02% : 0.000014s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000008s : 1: remove_cast_before_assign_add 28.18% : 0.016261s : 1: remove_dup_value 0.60% : 0.000347s : 1: renormalize.infer 0.41% : 0.000237s : 1: renormalize.specialize 0.01% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.02% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.08% : 0.000046s : 1: rewriter_after_opt_a 0.12% : 0.000066s : 1: rewriter_before_opt_a 0.01% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.01% : 0.000005s : 1: slice_recompute_activation 0.01% : 0.000004s : 1: split_layernorm_comm 0.01% : 0.000005s : 1: split_matmul_comm_elemetwise 0.02% : 0.000011s : 1: swap_dp_allreduce_reducescatter 0.17% : 0.000099s : 1: symbol_engine_optimizer 0.27% : 0.000154s : 1: tuple_transform 6.18% : 0.003563s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_bf16[dim_type1-False-batch_size_list1-shape0-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_bf16[dim_type1-False-batch_size_list1-shape0-ms_dtype0],max_mem:196.0M [WARNING] ME(152945:281473835274032,MainProcess):2026-01-29-17:45:04.262.313 [mindspore/context.py:1334] For 'context.set_context', the parameter 'device_target' will be deprecated and removed in a future version. Please use the api mindspore.set_device() instead. TotalTime = 0.0145427, [21] [bootstrap]: 0.00060236 [type_inference]: 0.00353734 [event_method]: 1.003e-05 [auto_monad]: 5.111e-05 [graph_reusing]: 4.87998e-06 [inline]: 2.66e-06 [add_attr]: 0.00451243, [1] [add_attr_with_inline]: 0.00450003, [1] [Cycle 1]: 5.044e-05, [2] [tag_attr]: 1.454e-05 [meta_addattr_fg_expand]: 3.03998e-06 [parallel-infer-symbol]: 4.02002e-06 [pre_auto_parallel]: 2.976e-05 [insert-virtual-dataset]: 2.84999e-06 [parallel-infer-symbol-second]: 9.79984e-07 [dataset_repeat_opt]: 2.17999e-06 [pipeline_split]: 1.57001e-06 [optimize]: 0.00462787, [53] [py_interpret_to_execute]: 1.932e-05 [rewriter_before_opt_a]: 4.726e-05 [opt_a]: 0.00235142, [2] [Cycle 1]: 0.00162333, [45] [expand_dump_flag]: 2.73998e-06 [switch_simplify]: 2.409e-05 [loop_unroll]: 1.219e-05 [a_1]: 0.00029943 [with_stream_mark]: 1.778e-05 [recompute_prepare]: 9.22999e-06 [updatestate_depend_eliminate]: 3.83001e-06 [updatestate_assign_eliminate]: 3.81999e-06 [updatestate_loads_eliminate]: 3.04001e-06 [parameter_eliminate]: 2.04e-06 [a_2]: 8.483e-05 [accelerated_algorithm]: 7.54002e-06 [shard]: 1.98997e-06 [meta_shard_fg_expand]: 1.77001e-06 [shard_inline]: 7.00998e-06 [merge_send_recv]: 9.04e-06 [auto_parallel]: 7.41001e-06 [parallel]: 2.709e-05 [flash_sp]: 9.15999e-06 [merge_comm]: 3.9e-06 [allreduce_fusion]: 3.76999e-06 [matmul_add_comm_reduction]: 9.68002e-06 [allreduce_slice_to_reducescatter]: 9.00007e-07 [virtual_shard_identity]: 8.09997e-06 [virtual_dataset]: 7.56999e-06 [get_grad_eliminate_]: 6.49001e-06 [virtual_output]: 6.56e-06 [merge_forward]: 4.22e-06 [cell_reuse_recompute_pass]: 1.79e-06 [offload_activation]: 1.196e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.359e-05 [merge_recompute_call_nodes]: 1.70001e-06 [before_grad]: 1.071e-05 [set_forward_comm_id_for_comm_node_pass]: 4.00998e-06 [meta_fg_expand]: 3.08e-06 [flash_sp_send_recv_attached]: 2.60997e-06 [receive_attached]: 2.41e-06 [after_resolve]: 1.247e-05 [a_after_grad]: 1.01e-05 [renormalize]: 0.00059399 [add_forward_monad_depend]: 7.98999e-06 [auto_monad_grad]: 2.51e-06 [auto_monad_eliminator]: 1.464e-05 [cse]: 4.486e-05 [a_3]: 4.939e-05 [Cycle 2]: 0.00071703, [45] [expand_dump_flag]: 9.39996e-07 [switch_simplify]: 8.21002e-06 [loop_unroll]: 6.91001e-06 [a_1]: 0.00013549 [with_stream_mark]: 1.197e-05 [recompute_prepare]: 6.57002e-06 [updatestate_depend_eliminate]: 3.16999e-06 [updatestate_assign_eliminate]: 2.91e-06 [updatestate_loads_eliminate]: 2.59001e-06 [parameter_eliminate]: 1.12999e-06 [a_2]: 7.64e-05 [accelerated_algorithm]: 6.76999e-06 [shard]: 1.40001e-06 [meta_shard_fg_expand]: 1.58002e-06 [shard_inline]: 6.98e-06 [merge_send_recv]: 4.97e-06 [auto_parallel]: 6.51e-06 [parallel]: 4.76002e-06 [flash_sp]: 3.67002e-06 [merge_comm]: 3.7e-06 [allreduce_fusion]: 3.31999e-06 [matmul_add_comm_reduction]: 1.311e-05 [allreduce_slice_to_reducescatter]: 3.69997e-07 [virtual_shard_identity]: 7.45e-06 [virtual_dataset]: 6.61999e-06 [get_grad_eliminate_]: 6.36998e-06 [virtual_output]: 6.22001e-06 [merge_forward]: 3.2e-06 [cell_reuse_recompute_pass]: 1.65001e-06 [offload_activation]: 2.131e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.316e-05 [merge_recompute_call_nodes]: 7.09988e-07 [before_grad]: 9.75002e-06 [set_forward_comm_id_for_comm_node_pass]: 3.42002e-06 [meta_fg_expand]: 2.46e-06 [flash_sp_send_recv_attached]: 9.09989e-07 [receive_attached]: 1.27999e-06 [after_resolve]: 1.129e-05 [a_after_grad]: 9.79999e-06 [renormalize]: 7.99773e-08 [add_forward_monad_depend]: 1.27e-06 [auto_monad_grad]: 1.07e-06 [auto_monad_eliminator]: 7.48999e-06 [cse]: 2.794e-05 [a_3]: 3.845e-05 [py_interpret_to_execute_after_opt_a]: 1.078e-05 [slice_cell_reuse_recomputed_activation]: 1.86e-06 [rewriter_after_opt_a]: 3.774e-05 [convert_after_rewriter]: 7.09001e-06 [order_py_execute_after_rewriter]: 5.62999e-06 [mutable_eliminate]: 0.00058359 [opt_b]: 0.00026153, [1] [Cycle 1]: 0.00025505, [7] [b_1]: 0.0001715 [b_2]: 8.73001e-06 [updatestate_depend_eliminate]: 6.15002e-06 [updatestate_assign_eliminate]: 2.79999e-06 [updatestate_loads_eliminate]: 2.79001e-06 [renormalize]: 4.10015e-07 [cse]: 2.524e-05 [optimize_parallel_all_gather_comm]: 1.746e-05 [overlap_param_gather]: 1.89e-06 [cconv]: 2.763e-05 [loop_unroll]: 0.00048741 [opt_after_cconv]: 0.00011208, [1] [Cycle 1]: 0.00010592, [7] [c_1]: 3.409e-05 [parameter_eliminate]: 3.11001e-06 [updatestate_depend_eliminate]: 5.82999e-06 [updatestate_assign_eliminate]: 2.61999e-06 [updatestate_loads_eliminate]: 2.61e-06 [cse]: 2.252e-05 [renormalize]: 4.19997e-07 [remove_dup_value]: 3.38e-05 [tuple_transform]: 7.964e-05, [1] [Cycle 1]: 7.407e-05, [4] [d_1]: 4.663e-05 [none_parameter_eliminate]: 1.54998e-06 [renormalize]: 1.30007e-07 [switch_simplify]: 7.43e-06 [partial_unused_args_eliminate]: 2.10002e-06 [add_recomputation]: 8.341e-05 [cse_after_recomputation]: 2.741e-05, [1] [Cycle 1]: 2.259e-05, [1] [cse]: 1.645e-05 [environ_conv]: 5.97001e-06 [swap_dp_allreduce_reducescatter]: 5.97999e-06 [bias_add_comm_swap]: 2.65997e-06 [label_micro_interleaved_index]: 4.67e-06 [label_fine_grained_interleaved_index]: 2.73e-06 [merge_cast_opt]: 1.59e-06 [slice_recompute_activation]: 2.18002e-06 [micro_interleaved_order_control]: 2.41998e-06 [assign_add_opt]: 1.29e-06 [ForceFp32Comm]: 1.26002e-06 [remove_cast_before_assign_add]: 1.04003e-06 [full_micro_interleaved_order_control]: 2.14e-06 [reorder_send_recv_between_fp_bp]: 2.57001e-06 [comm_op_add_attrs]: 9.70002e-07 [add_comm_op_reuse_tag]: 9.39996e-07 [interleave_split_concat_branches]: 1.18001e-06 [interleave_parallel_branches]: 1.32e-06 [overlap_opt_shard_in_pipeline]: 1.55001e-06 [overlap_opt_shard_grad_in_pipeline]: 1.55001e-06 [control_data_broadcast_order]: 1.302e-05 [grouped_pairwise_exchange_alltoall]: 1.66e-06 [offloading_packed_experts]: 4.12e-06 [overlap_recompute_and_grad_model_parallel]: 4.68001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.13001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.33002e-06 [overlap_recompute_comm]: 2.19999e-06 [overlap_grad_ring_attention]: 4.13999e-06 [overlap_grad_flash_sp]: 1.904e-05 [begin_end_overlap_inline]: 4.69998e-07 [split_matmul_comm_elemetwise]: 2.16e-06 [split_layernorm_comm]: 1.58002e-06 [handle_group_info]: 1.50001e-06 [symbol_engine_optimizer]: 8.117e-05, [1] [Cycle 1]: 7.694e-05, [6] [build]: 3.26999e-06 [elim_shapecalc]: 1.152e-05 [elim_not_effective]: 1.445e-05 [opt_reshape]: 7.61999e-06 [fold_const_symbol]: 1.101e-05 [renormalize]: 1.80007e-07 [detach_backward]: 1.98002e-06 [pipeline_parallel_scheduler]: 1.42e-06 [auto_monad_reorder]: 1.821e-05 [get_jit_bprop_graph]: 1.83002e-06 [rewriter_after_jit_bprop_graph]: 4.37e-06 [opt_after_jit_grad]: 0.00084996 [validate]: 4.586e-05 Sums bootstrap : 0.000602s : 6.74% type_inference : 0.003537s : 39.58% event_method : 0.000010s : 0.11% auto_monad : 0.000051s : 0.57% graph_reusing : 0.000005s : 0.05% inline : 0.000003s : 0.03% add_attr.add_attr_with_inline.tag_attr : 0.000015s : 0.16% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.03% parallel-infer-symbol : 0.000004s : 0.04% pre_auto_parallel : 0.000030s : 0.33% insert-virtual-dataset : 0.000003s : 0.03% parallel-infer-symbol-second : 0.000001s : 0.01% dataset_repeat_opt : 0.000002s : 0.02% pipeline_split : 0.000002s : 0.02% optimize.py_interpret_to_execute : 0.000019s : 0.22% optimize.rewriter_before_opt_a : 0.000047s : 0.53% optimize.opt_a.expand_dump_flag : 0.000004s : 0.04% optimize.opt_a.switch_simplify : 0.000032s : 0.36% optimize.opt_a.loop_unroll : 0.000019s : 0.21% optimize.opt_a.a_1 : 0.000435s : 4.87% optimize.opt_a.with_stream_mark : 0.000030s : 0.33% optimize.opt_a.recompute_prepare : 0.000016s : 0.18% optimize.opt_a.updatestate_depend_eliminate : 0.000007s : 0.08% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.08% optimize.opt_a.updatestate_loads_eliminate : 0.000006s : 0.06% optimize.opt_a.parameter_eliminate : 0.000003s : 0.04% optimize.opt_a.a_2 : 0.000161s : 1.80% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.16% optimize.opt_a.shard : 0.000003s : 0.04% optimize.opt_a.meta_shard_fg_expand : 0.000003s : 0.04% optimize.opt_a.shard_inline : 0.000014s : 0.16% optimize.opt_a.merge_send_recv : 0.000014s : 0.16% optimize.opt_a.auto_parallel : 0.000014s : 0.16% optimize.opt_a.parallel : 0.000032s : 0.36% optimize.opt_a.flash_sp : 0.000013s : 0.14% optimize.opt_a.merge_comm : 0.000008s : 0.09% optimize.opt_a.allreduce_fusion : 0.000007s : 0.08% optimize.opt_a.matmul_add_comm_reduction : 0.000023s : 0.26% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000016s : 0.17% optimize.opt_a.virtual_dataset : 0.000014s : 0.16% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.14% optimize.opt_a.virtual_output : 0.000013s : 0.14% optimize.opt_a.merge_forward : 0.000007s : 0.08% optimize.opt_a.cell_reuse_recompute_pass : 0.000003s : 0.04% optimize.opt_a.offload_activation : 0.000033s : 0.37% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000027s : 0.30% optimize.opt_a.merge_recompute_call_nodes : 0.000002s : 0.03% optimize.opt_a.before_grad : 0.000020s : 0.23% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000007s : 0.08% optimize.opt_a.meta_fg_expand : 0.000006s : 0.06% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.04% optimize.opt_a.receive_attached : 0.000004s : 0.04% optimize.opt_a.after_resolve : 0.000024s : 0.27% optimize.opt_a.a_after_grad : 0.000020s : 0.22% optimize.opt_a.renormalize : 0.000594s : 6.65% optimize.opt_a.add_forward_monad_depend : 0.000009s : 0.10% optimize.opt_a.auto_monad_grad : 0.000004s : 0.04% optimize.opt_a.auto_monad_eliminator : 0.000022s : 0.25% optimize.opt_a.cse : 0.000073s : 0.81% optimize.opt_a.a_3 : 0.000088s : 0.98% optimize.py_interpret_to_execute_after_opt_a : 0.000011s : 0.12% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.02% optimize.rewriter_after_opt_a : 0.000038s : 0.42% optimize.convert_after_rewriter : 0.000007s : 0.08% optimize.order_py_execute_after_rewriter : 0.000006s : 0.06% optimize.mutable_eliminate : 0.000584s : 6.53% optimize.opt_b.b_1 : 0.000171s : 1.92% optimize.opt_b.b_2 : 0.000009s : 0.10% optimize.opt_b.updatestate_depend_eliminate : 0.000006s : 0.07% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.03% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.03% optimize.opt_b.renormalize : 0.000000s : 0.00% optimize.opt_b.cse : 0.000025s : 0.28% optimize.optimize_parallel_all_gather_comm : 0.000017s : 0.20% optimize.overlap_param_gather : 0.000002s : 0.02% optimize.cconv : 0.000028s : 0.31% optimize.loop_unroll : 0.000487s : 5.45% optimize.opt_after_cconv.c_1 : 0.000034s : 0.38% optimize.opt_after_cconv.parameter_eliminate : 0.000003s : 0.03% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000006s : 0.07% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.03% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000003s : 0.03% optimize.opt_after_cconv.cse : 0.000023s : 0.25% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000034s : 0.38% optimize.tuple_transform.d_1 : 0.000047s : 0.52% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.02% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.08% optimize.partial_unused_args_eliminate : 0.000002s : 0.02% optimize.add_recomputation : 0.000083s : 0.93% optimize.cse_after_recomputation.cse : 0.000016s : 0.18% optimize.environ_conv : 0.000006s : 0.07% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.07% optimize.bias_add_comm_swap : 0.000003s : 0.03% optimize.label_micro_interleaved_index : 0.000005s : 0.05% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.03% optimize.merge_cast_opt : 0.000002s : 0.02% optimize.slice_recompute_activation : 0.000002s : 0.02% optimize.micro_interleaved_order_control : 0.000002s : 0.03% optimize.assign_add_opt : 0.000001s : 0.01% optimize.ForceFp32Comm : 0.000001s : 0.01% optimize.remove_cast_before_assign_add : 0.000001s : 0.01% optimize.full_micro_interleaved_order_control : 0.000002s : 0.02% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.03% optimize.comm_op_add_attrs : 0.000001s : 0.01% optimize.add_comm_op_reuse_tag : 0.000001s : 0.01% optimize.interleave_split_concat_branches : 0.000001s : 0.01% optimize.interleave_parallel_branches : 0.000001s : 0.01% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.02% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.02% optimize.control_data_broadcast_order : 0.000013s : 0.15% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.02% optimize.offloading_packed_experts : 0.000004s : 0.05% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.05% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.01% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.01% optimize.overlap_recompute_comm : 0.000002s : 0.02% optimize.overlap_grad_ring_attention : 0.000004s : 0.05% optimize.overlap_grad_flash_sp : 0.000019s : 0.21% optimize.begin_end_overlap_inline : 0.000000s : 0.01% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.02% optimize.split_layernorm_comm : 0.000002s : 0.02% optimize.handle_group_info : 0.000002s : 0.02% optimize.symbol_engine_optimizer.build : 0.000003s : 0.04% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000012s : 0.13% optimize.symbol_engine_optimizer.elim_not_effective : 0.000014s : 0.16% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.09% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000011s : 0.12% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.02% pipeline_parallel_scheduler : 0.000001s : 0.02% auto_monad_reorder : 0.000018s : 0.20% get_jit_bprop_graph : 0.000002s : 0.02% rewriter_after_jit_bprop_graph : 0.000004s : 0.05% opt_after_jit_grad : 0.000850s : 9.51% validate : 0.000046s : 0.51% Time group info: ------[substitution.] 0.000097 21 2.02% : 0.000002s : 2: substitution.elim_not_effective 1.62% : 0.000002s : 2: substitution.fold_const_symbol 5.83% : 0.000006s : 4: substitution.graph_param_transform 76.73% : 0.000074s : 1: substitution.inline 3.71% : 0.000004s : 4: substitution.j_node_and_user_rematch 5.04% : 0.000005s : 4: substitution.remove_not_recompute_node 5.05% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.003484 2 89.57% : 0.003121s : 1: type_inference.infer 10.43% : 0.000363s : 1: type_inference.specialize ------[replace.] 0.000016 1 100.00% : 0.000016s : 1: replace.inline ------[match.] 0.000073 1 100.00% : 0.000073s : 1: match.inline ------[predicate.] 0.000475 967 0.37% : 0.000002s : 9: predicate.accumulaten_eliminater 0.36% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.21% : 0.000001s : 8: predicate.addn_check_dump 0.30% : 0.000001s : 9: predicate.addn_zero_filter 0.24% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 0.71% : 0.000003s : 17: predicate.arithmetic_simplify 0.29% : 0.000001s : 9: predicate.cast_eliminate 0.25% : 0.000001s : 8: predicate.check_bprop_eliminate 0.21% : 0.000001s : 8: predicate.compare_switch_simplify 0.08% : 0.000000s : 4: predicate.const_output_eliminate 0.23% : 0.000001s : 8: predicate.depend_value_elim 0.30% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.28% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.26% : 0.000001s : 9: predicate.dict_set_item_eliminator 0.54% : 0.000003s : 8: predicate.dumpgradient_eliminate 0.10% : 0.000000s : 4: predicate.elim_not_effective 0.20% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 0.41% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.36% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.33% : 0.000002s : 13: predicate.environ_get_depend_swap 0.65% : 0.000003s : 21: predicate.environ_get_eliminate 0.43% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.28% : 0.000001s : 10: predicate.exchange_switch_depend_value 0.52% : 0.000002s : 10: predicate.float_depend_g_call 0.20% : 0.000001s : 8: predicate.float_environ_get_switch 0.35% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.09% : 0.000000s : 4: predicate.fold_const_symbol 0.36% : 0.000002s : 8: predicate.get_grad_eliminate 0.10% : 0.000000s : 4: predicate.graph_param_transform 0.25% : 0.000001s : 8: predicate.incorporate_call 0.20% : 0.000001s : 8: predicate.incorporate_call_switch 2.07% : 0.000010s : 43: predicate.inline 0.31% : 0.000001s : 8: predicate.inline_without_move 0.14% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.41% : 0.000002s : 8: predicate.less_batch_normalization 0.63% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 0.79% : 0.000004s : 26: predicate.load_eliminater 0.37% : 0.000002s : 4: predicate.loop_unroll_after_grad 0.48% : 0.000002s : 13: predicate.loop_unroll_before_grad 0.75% : 0.000004s : 17: predicate.make_slice_get_slice_eliminator 0.21% : 0.000001s : 8: predicate.merge_addn 0.23% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.25% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.24% : 0.000001s : 9: predicate.minmaximum_grad 0.45% : 0.000002s : 4: predicate.mutable_eliminate 0.16% : 0.000001s : 4: predicate.opt_reshape 0.20% : 0.000001s : 4: predicate.parallel_virtual_node 0.39% : 0.000002s : 10: predicate.partial_defer_inline 0.42% : 0.000002s : 13: predicate.partial_eliminate 0.29% : 0.000001s : 9: predicate.print_const_string_wrapper 0.25% : 0.000001s : 8: predicate.reduce_all_const_elim 0.43% : 0.000002s : 9: predicate.reduce_eliminate 0.72% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.25% : 0.000001s : 8: predicate.remove_not_recompute_node 0.43% : 0.000002s : 17: predicate.replace_applicator 0.22% : 0.000001s : 8: predicate.replace_old_param 0.11% : 0.000001s : 4: predicate.reset_defer_inline 0.28% : 0.000001s : 9: predicate.reshape_eliminate 0.26% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.17% : 0.000001s : 4: predicate.row_tensor_eliminate 0.30% : 0.000001s : 8: predicate.same_eliminate 0.20% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.33% : 0.000002s : 8: predicate.shard_identity_eliminate 66.37% : 0.000315s : 8: predicate.special_op_eliminate 0.31% : 0.000001s : 8: predicate.specialize_transform 0.38% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.32% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.16% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.31% : 0.000001s : 10: predicate.switch_defer_inline 0.56% : 0.000003s : 18: predicate.switch_layer_defer_inline 1.28% : 0.000006s : 35: predicate.switch_simplify 0.27% : 0.000001s : 9: predicate.tile_eliminate 0.37% : 0.000002s : 9: predicate.transpose_eliminate 0.50% : 0.000002s : 17: predicate.tuple_list_convert_item_index_to_positive 0.54% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 0.45% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 1.09% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 0.52% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 0.95% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 0.54% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 0.71% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 1.04% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.17% : 0.000001s : 4: predicate.value_based_eliminate 0.39% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.29% : 0.000001s : 8: predicate.virtual_output_eliminate 0.11% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.17% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000207 4 9.10% : 0.000019s : 1: func_graph_cloner_run.FuncGraphClonerGraph 90.90% : 0.000188s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.025584 192 0.02% : 0.000004s : 1: ForceFp32Comm 17.66% : 0.004517s : 1: add_attr 17.61% : 0.004504s : 1: add_attr_with_inline 0.02% : 0.000004s : 1: add_comm_op_reuse_tag 0.34% : 0.000088s : 1: add_recomputation 0.02% : 0.000004s : 1: assign_add_opt 0.22% : 0.000057s : 1: auto_monad 0.09% : 0.000023s : 1: auto_monad_reorder 0.01% : 0.000003s : 1: begin_end_overlap_inline 0.02% : 0.000006s : 1: bias_add_comm_swap 2.51% : 0.000642s : 1: bootstrap 0.12% : 0.000031s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.06% : 0.000016s : 1: control_data_broadcast_order 0.04% : 0.000010s : 1: convert_after_rewriter 0.13% : 0.000034s : 1: cse_after_recomputation 0.02% : 0.000005s : 1: dataset_repeat_opt 0.02% : 0.000005s : 1: detach_backward 0.04% : 0.000010s : 1: environ_conv 0.06% : 0.000015s : 1: event_method 0.02% : 0.000005s : 1: full_micro_interleaved_order_control 0.02% : 0.000005s : 1: get_jit_bprop_graph 0.03% : 0.000008s : 1: graph_reusing 0.02% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.02% : 0.000004s : 1: handle_group_info 0.02% : 0.000006s : 1: inline 0.02% : 0.000006s : 1: insert-virtual-dataset 0.02% : 0.000004s : 1: interleave_parallel_branches 0.02% : 0.000004s : 1: interleave_split_concat_branches 0.02% : 0.000006s : 1: label_fine_grained_interleaved_index 0.03% : 0.000007s : 1: label_micro_interleaved_index 2.00% : 0.000511s : 1: loop_unroll 0.02% : 0.000004s : 1: merge_cast_opt 0.02% : 0.000005s : 1: micro_interleaved_order_control 2.32% : 0.000594s : 1: mutable_eliminate 0.03% : 0.000007s : 1: offloading_packed_experts 0.06% : 0.000016s : 1: opt.transform.loop_unroll_optimizer 0.06% : 0.000016s : 1: opt.transform.mutable_eliminate 3.28% : 0.000839s : 78: opt.transform.opt_a 0.13% : 0.000032s : 1: opt.transform.opt_after_cconv 1.35% : 0.000346s : 1: opt.transform.opt_after_jit_grad 0.43% : 0.000111s : 28: opt.transform.opt_b 0.20% : 0.000052s : 2: opt.transform.opt_trans_graph 0.16% : 0.000040s : 4: opt.transform.symbol_engine_opt 9.20% : 0.002355s : 1: opt_a 0.45% : 0.000116s : 1: opt_after_cconv 3.37% : 0.000863s : 1: opt_after_jit_grad 1.04% : 0.000265s : 1: opt_b 18.11% : 0.004633s : 1: optimize 0.08% : 0.000021s : 1: optimize_parallel_all_gather_comm 0.03% : 0.000009s : 1: order_py_execute_after_rewriter 0.09% : 0.000022s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.03% : 0.000008s : 1: overlap_grad_ring_attention 0.02% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.02% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.02% : 0.000005s : 1: overlap_param_gather 0.02% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.03% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.02% : 0.000005s : 1: overlap_recompute_comm 0.03% : 0.000008s : 1: parallel-infer-symbol 0.02% : 0.000004s : 1: parallel-infer-symbol-second 0.02% : 0.000005s : 1: partial_unused_args_eliminate 0.02% : 0.000004s : 1: pipeline_parallel_scheduler 0.25% : 0.000064s : 1: pipeline_split 0.13% : 0.000034s : 1: pre_auto_parallel 0.10% : 0.000025s : 1: py_interpret_to_execute 0.06% : 0.000014s : 1: py_interpret_to_execute_after_opt_a 0.02% : 0.000004s : 1: remove_cast_before_assign_add 0.15% : 0.000038s : 1: remove_dup_value 1.39% : 0.000356s : 1: renormalize.infer 0.90% : 0.000230s : 1: renormalize.specialize 0.02% : 0.000005s : 1: reorder_send_recv_between_fp_bp 0.03% : 0.000008s : 1: rewriter_after_jit_bprop_graph 0.16% : 0.000042s : 1: rewriter_after_opt_a 0.20% : 0.000052s : 1: rewriter_before_opt_a 0.02% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.02% : 0.000005s : 1: slice_recompute_activation 0.02% : 0.000004s : 1: split_layernorm_comm 0.02% : 0.000005s : 1: split_matmul_comm_elemetwise 0.03% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.33% : 0.000084s : 1: symbol_engine_optimizer 0.32% : 0.000082s : 1: tuple_transform 13.89% : 0.003555s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_bf16[dim_type1-False-batch_size_list2-shape0-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_bf16[dim_type1-False-batch_size_list2-shape0-ms_dtype0],max_mem:292.0M =============================== 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_batch_matmul.py::test_batch_matmul_3dx_ndw_bf16[dim_type1-False-batch_size_list2-shape0-ms_dtype0] /home/jenkins/mindspore/testcases/testcases/tests/st/infer/ops/test_internal_ops/test_batch_matmul.py:51: 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 ================= 12 passed, 26 warnings in 521.12s (0:08:41) ==================