==================================================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_002/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 24 items test_batch_matmul.py [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:37:46.907.833 [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 = 0.968319, [21] [bootstrap]: 0.032758 [type_inference]: 0.836801 [event_method]: 1.674e-05 [auto_monad]: 9.105e-05 [graph_reusing]: 4.95001e-06 [inline]: 3.43e-06 [add_attr]: 0.00880036, [1] [add_attr_with_inline]: 0.00878655, [1] [Cycle 1]: 8.518e-05, [2] [tag_attr]: 1.511e-05 [meta_addattr_fg_expand]: 3.52002e-06 [parallel-infer-symbol]: 3.25002e-06 [pre_auto_parallel]: 4.231e-05 [insert-virtual-dataset]: 2.67001e-06 [parallel-infer-symbol-second]: 7.89994e-07 [dataset_repeat_opt]: 2.14999e-06 [pipeline_split]: 1.59e-06 [optimize]: 0.0890055, [53] [py_interpret_to_execute]: 1.692e-05 [rewriter_before_opt_a]: 6.52e-05 [opt_a]: 0.0865857, [2] [Cycle 1]: 0.00158855, [45] [expand_dump_flag]: 2.91999e-06 [switch_simplify]: 2.46e-05 [loop_unroll]: 1.09e-05 [a_1]: 0.00026824 [with_stream_mark]: 1.96e-05 [recompute_prepare]: 7.68001e-06 [updatestate_depend_eliminate]: 3.76999e-06 [updatestate_assign_eliminate]: 3.61999e-06 [updatestate_loads_eliminate]: 3.3e-06 [parameter_eliminate]: 2.06003e-06 [a_2]: 7.854e-05 [accelerated_algorithm]: 6.96001e-06 [shard]: 2.19001e-06 [meta_shard_fg_expand]: 1.66998e-06 [shard_inline]: 6.57002e-06 [merge_send_recv]: 8.78001e-06 [auto_parallel]: 6.96001e-06 [parallel]: 4.684e-05 [flash_sp]: 8.62e-06 [merge_comm]: 4.18999e-06 [allreduce_fusion]: 3.66001e-06 [matmul_add_comm_reduction]: 9.81e-06 [allreduce_slice_to_reducescatter]: 6.19999e-07 [virtual_shard_identity]: 7.90998e-06 [virtual_dataset]: 6.49999e-06 [get_grad_eliminate_]: 6.43e-06 [virtual_output]: 5.81998e-06 [merge_forward]: 4.38001e-06 [cell_reuse_recompute_pass]: 1.27999e-06 [offload_activation]: 1.001e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.231e-05 [merge_recompute_call_nodes]: 1.60999e-06 [before_grad]: 9.97001e-06 [set_forward_comm_id_for_comm_node_pass]: 3.56999e-06 [meta_fg_expand]: 3.01999e-06 [flash_sp_send_recv_attached]: 2.59999e-06 [receive_attached]: 2.08002e-06 [after_resolve]: 1.128e-05 [a_after_grad]: 9.19e-06 [renormalize]: 0.00062105 [add_forward_monad_depend]: 4.80999e-06 [auto_monad_grad]: 2.71e-06 [auto_monad_eliminator]: 1.36e-05 [cse]: 4.577e-05 [a_3]: 4.402e-05 [Cycle 2]: 0.0849858, [45] [expand_dump_flag]: 9.70002e-07 [switch_simplify]: 8.33999e-06 [loop_unroll]: 6.06998e-06 [a_1]: 0.0842548 [with_stream_mark]: 4.074e-05 [recompute_prepare]: 1.618e-05 [updatestate_depend_eliminate]: 5.19e-06 [updatestate_assign_eliminate]: 3.51001e-06 [updatestate_loads_eliminate]: 3.73999e-06 [parameter_eliminate]: 2.75002e-06 [a_2]: 8.055e-05 [accelerated_algorithm]: 6.87002e-06 [shard]: 3.3e-06 [meta_shard_fg_expand]: 3.35e-06 [shard_inline]: 6.56e-06 [merge_send_recv]: 9.76e-06 [auto_parallel]: 1.075e-05 [parallel]: 1.029e-05 [flash_sp]: 4.73001e-06 [merge_comm]: 3.80998e-06 [allreduce_fusion]: 3.51999e-06 [matmul_add_comm_reduction]: 1.441e-05 [allreduce_slice_to_reducescatter]: 1.10999e-06 [virtual_shard_identity]: 8e-06 [virtual_dataset]: 6.34999e-06 [get_grad_eliminate_]: 6.06e-06 [virtual_output]: 6.69999e-06 [merge_forward]: 4.74e-06 [cell_reuse_recompute_pass]: 3.7e-06 [offload_activation]: 1.335e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.17e-05 [merge_recompute_call_nodes]: 1.55999e-06 [before_grad]: 1.067e-05 [set_forward_comm_id_for_comm_node_pass]: 3.65e-06 [meta_fg_expand]: 2.84999e-06 [flash_sp_send_recv_attached]: 1.74e-06 [receive_attached]: 2.53998e-06 [after_resolve]: 1.427e-05 [a_after_grad]: 9.12001e-06 [renormalize]: 8.00064e-08 [add_forward_monad_depend]: 3.13e-06 [auto_monad_grad]: 3.22002e-06 [auto_monad_eliminator]: 2.009e-05 [cse]: 4.582e-05 [a_3]: 3.968e-05 [py_interpret_to_execute_after_opt_a]: 2.293e-05 [slice_cell_reuse_recomputed_activation]: 2.61e-06 [rewriter_after_opt_a]: 8.01e-05 [convert_after_rewriter]: 7.35998e-06 [order_py_execute_after_rewriter]: 5.81003e-06 [mutable_eliminate]: 0.00072603 [opt_b]: 0.0002114, [1] [Cycle 1]: 0.00020184, [7] [b_1]: 0.00012153 [b_2]: 8.04002e-06 [updatestate_depend_eliminate]: 7.98999e-06 [updatestate_assign_eliminate]: 2.86e-06 [updatestate_loads_eliminate]: 2.49001e-06 [renormalize]: 6.39993e-07 [cse]: 2.556e-05 [optimize_parallel_all_gather_comm]: 2.17e-05 [overlap_param_gather]: 3.19001e-06 [cconv]: 3.744e-05 [loop_unroll]: 0.00044988 [opt_after_cconv]: 0.00010278, [1] [Cycle 1]: 9.667e-05, [7] [c_1]: 2.915e-05 [parameter_eliminate]: 3.08e-06 [updatestate_depend_eliminate]: 5.57999e-06 [updatestate_assign_eliminate]: 2.86e-06 [updatestate_loads_eliminate]: 2.63998e-06 [cse]: 2.06e-05 [renormalize]: 3.7998e-07 [remove_dup_value]: 1.729e-05 [tuple_transform]: 8.994e-05, [1] [Cycle 1]: 8.515e-05, [4] [d_1]: 4.414e-05 [none_parameter_eliminate]: 1.89e-06 [renormalize]: 2.3999e-07 [switch_simplify]: 6.71e-06 [partial_unused_args_eliminate]: 2.09e-06 [add_recomputation]: 5.73e-05 [cse_after_recomputation]: 2.602e-05, [1] [Cycle 1]: 2.109e-05, [1] [cse]: 1.521e-05 [environ_conv]: 1.736e-05 [swap_dp_allreduce_reducescatter]: 5.72001e-06 [bias_add_comm_swap]: 2.51e-06 [label_micro_interleaved_index]: 4.37e-06 [label_fine_grained_interleaved_index]: 2.94999e-06 [merge_cast_opt]: 1.76e-06 [slice_recompute_activation]: 2.21998e-06 [micro_interleaved_order_control]: 2.24001e-06 [assign_add_opt]: 1.32999e-06 [ForceFp32Comm]: 9.69972e-07 [remove_cast_before_assign_add]: 1.12e-06 [full_micro_interleaved_order_control]: 2.16998e-06 [reorder_send_recv_between_fp_bp]: 2.85002e-06 [comm_op_add_attrs]: 1.09e-06 [add_comm_op_reuse_tag]: 1.00001e-06 [interleave_split_concat_branches]: 1.22e-06 [interleave_parallel_branches]: 1.09e-06 [overlap_opt_shard_in_pipeline]: 2.614e-05 [overlap_opt_shard_grad_in_pipeline]: 1.84e-06 [control_data_broadcast_order]: 1.368e-05 [grouped_pairwise_exchange_alltoall]: 1.55999e-06 [offloading_packed_experts]: 4.08001e-06 [overlap_recompute_and_grad_model_parallel]: 5.69e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.20999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.49e-06 [overlap_recompute_comm]: 2.64999e-06 [overlap_grad_ring_attention]: 4.42998e-06 [overlap_grad_flash_sp]: 3.645e-05 [begin_end_overlap_inline]: 5.89993e-07 [split_matmul_comm_elemetwise]: 2.22001e-06 [split_layernorm_comm]: 2.02999e-06 [handle_group_info]: 1.20001e-06 [symbol_engine_optimizer]: 7.777e-05, [1] [Cycle 1]: 7.274e-05, [6] [build]: 3.41999e-06 [elim_shapecalc]: 1.067e-05 [elim_not_effective]: 1.5e-05 [opt_reshape]: 6.81999e-06 [fold_const_symbol]: 9.52999e-06 [renormalize]: 2.40019e-07 [detach_backward]: 2.46998e-06 [pipeline_parallel_scheduler]: 2.04e-06 [auto_monad_reorder]: 2.611e-05 [get_jit_bprop_graph]: 2.16e-06 [rewriter_after_jit_bprop_graph]: 4.42998e-06 [opt_after_jit_grad]: 0.00048297 [validate]: 6.323e-05 Sums bootstrap : 0.032758s : 3.42% type_inference : 0.836801s : 87.31% event_method : 0.000017s : 0.00% auto_monad : 0.000091s : 0.01% graph_reusing : 0.000005s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000015s : 0.00% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000003s : 0.00% pre_auto_parallel : 0.000042s : 0.00% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000017s : 0.00% optimize.rewriter_before_opt_a : 0.000065s : 0.01% optimize.opt_a.expand_dump_flag : 0.000004s : 0.00% optimize.opt_a.switch_simplify : 0.000033s : 0.00% optimize.opt_a.loop_unroll : 0.000017s : 0.00% optimize.opt_a.a_1 : 0.084523s : 8.82% optimize.opt_a.with_stream_mark : 0.000060s : 0.01% optimize.opt_a.recompute_prepare : 0.000024s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000009s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.00% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000159s : 0.02% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.00% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000005s : 0.00% optimize.opt_a.shard_inline : 0.000013s : 0.00% optimize.opt_a.merge_send_recv : 0.000019s : 0.00% optimize.opt_a.auto_parallel : 0.000018s : 0.00% optimize.opt_a.parallel : 0.000057s : 0.01% optimize.opt_a.flash_sp : 0.000013s : 0.00% optimize.opt_a.merge_comm : 0.000008s : 0.00% optimize.opt_a.allreduce_fusion : 0.000007s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000024s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000016s : 0.00% optimize.opt_a.virtual_dataset : 0.000013s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000012s : 0.00% optimize.opt_a.virtual_output : 0.000013s : 0.00% optimize.opt_a.merge_forward : 0.000009s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000023s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000034s : 0.00% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000021s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000007s : 0.00% optimize.opt_a.meta_fg_expand : 0.000006s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000026s : 0.00% optimize.opt_a.a_after_grad : 0.000018s : 0.00% optimize.opt_a.renormalize : 0.000621s : 0.06% optimize.opt_a.add_forward_monad_depend : 0.000008s : 0.00% optimize.opt_a.auto_monad_grad : 0.000006s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000034s : 0.00% optimize.opt_a.cse : 0.000092s : 0.01% optimize.opt_a.a_3 : 0.000084s : 0.01% optimize.py_interpret_to_execute_after_opt_a : 0.000023s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000080s : 0.01% optimize.convert_after_rewriter : 0.000007s : 0.00% optimize.order_py_execute_after_rewriter : 0.000006s : 0.00% optimize.mutable_eliminate : 0.000726s : 0.08% optimize.opt_b.b_1 : 0.000122s : 0.01% optimize.opt_b.b_2 : 0.000008s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000008s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000002s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000026s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000022s : 0.00% optimize.overlap_param_gather : 0.000003s : 0.00% optimize.cconv : 0.000037s : 0.00% optimize.loop_unroll : 0.000450s : 0.05% optimize.opt_after_cconv.c_1 : 0.000029s : 0.00% optimize.opt_after_cconv.parameter_eliminate : 0.000003s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000006s : 0.00% 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 : 0.000021s : 0.00% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000017s : 0.00% optimize.tuple_transform.d_1 : 0.000044s : 0.00% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000057s : 0.01% optimize.cse_after_recomputation.cse : 0.000015s : 0.00% optimize.environ_conv : 0.000017s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000004s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000001s : 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 : 0.000026s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000014s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000004s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000006s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000004s : 0.00% optimize.overlap_grad_flash_sp : 0.000036s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000003s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000011s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000015s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000026s : 0.00% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000004s : 0.00% opt_after_jit_grad : 0.000483s : 0.05% validate : 0.000063s : 0.01% Time group info: ------[substitution.] 0.000104 21 4.00% : 0.000004s : 2: substitution.elim_not_effective 1.42% : 0.000001s : 2: substitution.fold_const_symbol 5.94% : 0.000006s : 4: substitution.graph_param_transform 71.18% : 0.000074s : 1: substitution.inline 4.48% : 0.000005s : 4: substitution.j_node_and_user_rematch 5.51% : 0.000006s : 4: substitution.remove_not_recompute_node 7.47% : 0.000008s : 4: substitution.replace_old_param ------[type_inference.] 0.836724 2 99.93% : 0.836141s : 1: type_inference.infer 0.07% : 0.000583s : 1: type_inference.specialize ------[replace.] 0.000015 1 100.00% : 0.000015s : 1: replace.inline ------[match.] 0.000073 1 100.00% : 0.000073s : 1: match.inline ------[predicate.] 0.000169 967 0.87% : 0.000001s : 9: predicate.accumulaten_eliminater 0.81% : 0.000001s : 4: predicate.ad_related_special_op_eliminate 0.59% : 0.000001s : 8: predicate.addn_check_dump 0.66% : 0.000001s : 9: predicate.addn_zero_filter 4.95% : 0.000008s : 9: predicate.adjust_all_reduce_mul_add 2.24% : 0.000004s : 17: predicate.arithmetic_simplify 0.82% : 0.000001s : 9: predicate.cast_eliminate 0.86% : 0.000001s : 8: predicate.check_bprop_eliminate 0.63% : 0.000001s : 8: predicate.compare_switch_simplify 0.23% : 0.000000s : 4: predicate.const_output_eliminate 0.86% : 0.000001s : 8: predicate.depend_value_elim 0.72% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 1.01% : 0.000002s : 9: predicate.dict_get_item_eliminator 0.91% : 0.000002s : 9: predicate.dict_set_item_eliminator 1.45% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.24% : 0.000000s : 4: predicate.elim_not_effective 0.45% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.05% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.94% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.93% : 0.000002s : 13: predicate.environ_get_depend_swap 1.72% : 0.000003s : 21: predicate.environ_get_eliminate 0.91% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.71% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.58% : 0.000003s : 10: predicate.float_depend_g_call 0.59% : 0.000001s : 8: predicate.float_environ_get_switch 0.91% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.24% : 0.000000s : 4: predicate.fold_const_symbol 0.81% : 0.000001s : 8: predicate.get_grad_eliminate 0.22% : 0.000000s : 4: predicate.graph_param_transform 0.65% : 0.000001s : 8: predicate.incorporate_call 0.52% : 0.000001s : 8: predicate.incorporate_call_switch 7.24% : 0.000012s : 43: predicate.inline 0.90% : 0.000002s : 8: predicate.inline_without_move 0.42% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.90% : 0.000002s : 8: predicate.less_batch_normalization 1.54% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.03% : 0.000003s : 26: predicate.load_eliminater 1.03% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.07% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.74% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.63% : 0.000001s : 8: predicate.merge_addn 0.68% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.71% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.63% : 0.000001s : 9: predicate.minmaximum_grad 1.59% : 0.000003s : 4: predicate.mutable_eliminate 0.47% : 0.000001s : 4: predicate.opt_reshape 0.44% : 0.000001s : 4: predicate.parallel_virtual_node 1.39% : 0.000002s : 10: predicate.partial_defer_inline 1.06% : 0.000002s : 13: predicate.partial_eliminate 0.91% : 0.000002s : 9: predicate.print_const_string_wrapper 0.72% : 0.000001s : 8: predicate.reduce_all_const_elim 1.07% : 0.000002s : 9: predicate.reduce_eliminate 2.37% : 0.000004s : 26: predicate.redundant_stop_gradient_eliminater 0.72% : 0.000001s : 8: predicate.remove_not_recompute_node 1.09% : 0.000002s : 17: predicate.replace_applicator 0.96% : 0.000002s : 8: predicate.replace_old_param 0.33% : 0.000001s : 4: predicate.reset_defer_inline 0.82% : 0.000001s : 9: predicate.reshape_eliminate 0.78% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.52% : 0.000001s : 4: predicate.row_tensor_eliminate 1.04% : 0.000002s : 8: predicate.same_eliminate 0.55% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.88% : 0.000001s : 8: predicate.shard_identity_eliminate 0.84% : 0.000001s : 8: predicate.special_op_eliminate 0.84% : 0.000001s : 8: predicate.specialize_transform 1.40% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.86% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.36% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.80% : 0.000001s : 10: predicate.switch_defer_inline 2.09% : 0.000004s : 18: predicate.switch_layer_defer_inline 3.27% : 0.000006s : 35: predicate.switch_simplify 0.76% : 0.000001s : 9: predicate.tile_eliminate 0.74% : 0.000001s : 9: predicate.transpose_eliminate 1.45% : 0.000002s : 17: predicate.tuple_list_convert_item_index_to_positive 1.46% : 0.000002s : 17: predicate.tuple_list_get_item_const_eliminator 1.43% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.81% : 0.000006s : 25: predicate.tuple_list_get_item_eliminator 1.40% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.37% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.56% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.78% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.66% : 0.000004s : 34: predicate.updatestate_useless_node_eliminater 0.49% : 0.000001s : 4: predicate.value_based_eliminate 0.81% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.72% : 0.000001s : 8: predicate.virtual_output_eliminate 0.29% : 0.000000s : 4: predicate.virtual_view_grad_eliminate 0.49% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000259 4 9.22% : 0.000024s : 1: func_graph_cloner_run.FuncGraphClonerGraph 90.78% : 0.000235s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.151770 192 0.00% : 0.000004s : 1: ForceFp32Comm 0.76% : 0.008806s : 1: add_attr 0.76% : 0.008791s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000061s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.01% : 0.000098s : 1: auto_monad 0.00% : 0.000030s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 2.85% : 0.032812s : 1: bootstrap 0.00% : 0.000041s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000017s : 1: control_data_broadcast_order 0.00% : 0.000011s : 1: convert_after_rewriter 0.00% : 0.000029s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000021s : 1: environ_conv 0.00% : 0.000025s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000010s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000005s : 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% : 0.000007s : 1: label_micro_interleaved_index 0.04% : 0.000459s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.06% : 0.000736s : 1: mutable_eliminate 0.00% : 0.000007s : 1: offloading_packed_experts 0.00% : 0.000014s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000017s : 1: opt.transform.mutable_eliminate 7.37% : 0.084924s : 78: opt.transform.opt_a 0.00% : 0.000028s : 1: opt.transform.opt_after_cconv 0.00% : 0.000024s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000102s : 28: opt.transform.opt_b 0.00% : 0.000049s : 2: opt.transform.opt_trans_graph 0.00% : 0.000038s : 4: opt.transform.symbol_engine_opt 7.52% : 0.086589s : 1: opt_a 0.01% : 0.000106s : 1: opt_after_cconv 0.04% : 0.000493s : 1: opt_after_jit_grad 0.02% : 0.000215s : 1: opt_b 7.73% : 0.089011s : 1: optimize 0.00% : 0.000025s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000009s : 1: order_py_execute_after_rewriter 0.00% : 0.000040s : 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% : 0.000032s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000009s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000007s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000012s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000004s : 1: pipeline_split 0.00% : 0.000046s : 1: pre_auto_parallel 0.00% : 0.000021s : 1: py_interpret_to_execute 0.00% : 0.000026s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000021s : 1: remove_dup_value 0.03% : 0.000386s : 1: renormalize.infer 0.02% : 0.000227s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000007s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000085s : 1: rewriter_after_opt_a 0.01% : 0.000069s : 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 0.01% : 0.000080s : 1: symbol_engine_optimizer 0.01% : 0.000093s : 1: tuple_transform 72.66% : 0.836831s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[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_fp16[dim_type0-True-batch_size_list0-shape0-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:38:26.193.319 [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 = 0.237493, [21] [bootstrap]: 0.00064024 [type_inference]: 0.00446872 [event_method]: 1.915e-05 [auto_monad]: 0.00012803 [graph_reusing]: 7.65e-06 [inline]: 3.5e-06 [add_attr]: 0.00500241, [1] [add_attr_with_inline]: 0.00498185, [1] [Cycle 1]: 7.814e-05, [2] [tag_attr]: 2.094e-05 [meta_addattr_fg_expand]: 3.95e-06 [parallel-infer-symbol]: 4.55999e-06 [pre_auto_parallel]: 3.935e-05 [insert-virtual-dataset]: 2.88e-06 [parallel-infer-symbol-second]: 1.02998e-06 [dataset_repeat_opt]: 2.39001e-06 [pipeline_split]: 2.14e-06 [optimize]: 0.00563789, [53] [py_interpret_to_execute]: 2.615e-05 [rewriter_before_opt_a]: 5.463e-05 [opt_a]: 0.00275811, [2] [Cycle 1]: 0.00200267, [45] [expand_dump_flag]: 3.33e-06 [switch_simplify]: 2.673e-05 [loop_unroll]: 1.15e-05 [a_1]: 0.00033936 [with_stream_mark]: 2.493e-05 [recompute_prepare]: 2.351e-05 [updatestate_depend_eliminate]: 4.69998e-06 [updatestate_assign_eliminate]: 3.61999e-06 [updatestate_loads_eliminate]: 4.32e-06 [parameter_eliminate]: 2.59999e-06 [a_2]: 8.553e-05 [accelerated_algorithm]: 6.91001e-06 [shard]: 2.19001e-06 [meta_shard_fg_expand]: 1.90001e-06 [shard_inline]: 6.53003e-06 [merge_send_recv]: 1.063e-05 [auto_parallel]: 7.80998e-06 [parallel]: 7.266e-05 [flash_sp]: 1.249e-05 [merge_comm]: 3.82998e-06 [allreduce_fusion]: 3.78001e-06 [matmul_add_comm_reduction]: 1.094e-05 [allreduce_slice_to_reducescatter]: 1.10001e-06 [virtual_shard_identity]: 9.58002e-06 [virtual_dataset]: 7.80998e-06 [get_grad_eliminate_]: 6.79001e-06 [virtual_output]: 6.36998e-06 [merge_forward]: 6.11998e-06 [cell_reuse_recompute_pass]: 1.45001e-06 [offload_activation]: 1.41e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.507e-05 [merge_recompute_call_nodes]: 2.46e-06 [before_grad]: 1.112e-05 [set_forward_comm_id_for_comm_node_pass]: 4.35e-06 [meta_fg_expand]: 3.53e-06 [flash_sp_send_recv_attached]: 2.75002e-06 [receive_attached]: 3.35998e-06 [after_resolve]: 1.288e-05 [a_after_grad]: 9.60001e-06 [renormalize]: 0.00081569 [add_forward_monad_depend]: 7.03e-06 [auto_monad_grad]: 4.01001e-06 [auto_monad_eliminator]: 1.755e-05 [cse]: 4.641e-05 [a_3]: 5.779e-05 [Cycle 2]: 0.00074154, [45] [expand_dump_flag]: 2.59001e-06 [switch_simplify]: 9.14998e-06 [loop_unroll]: 6.69999e-06 [a_1]: 0.00012624 [with_stream_mark]: 1.764e-05 [recompute_prepare]: 6.51e-06 [updatestate_depend_eliminate]: 5.51e-06 [updatestate_assign_eliminate]: 2.89001e-06 [updatestate_loads_eliminate]: 2.78003e-06 [parameter_eliminate]: 1.66998e-06 [a_2]: 7.747e-05 [accelerated_algorithm]: 6.24001e-06 [shard]: 1.59e-06 [meta_shard_fg_expand]: 1.66998e-06 [shard_inline]: 5.93002e-06 [merge_send_recv]: 6.60997e-06 [auto_parallel]: 8.84003e-06 [parallel]: 8.3e-06 [flash_sp]: 3.78999e-06 [merge_comm]: 3.66001e-06 [allreduce_fusion]: 3.85e-06 [matmul_add_comm_reduction]: 7.36999e-06 [allreduce_slice_to_reducescatter]: 4.50003e-07 [virtual_shard_identity]: 7.81001e-06 [virtual_dataset]: 4.071e-05 [get_grad_eliminate_]: 6.71e-06 [virtual_output]: 6.09001e-06 [merge_forward]: 5.57001e-06 [cell_reuse_recompute_pass]: 2.66999e-06 [offload_activation]: 1.333e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.604e-05 [merge_recompute_call_nodes]: 9.20001e-07 [before_grad]: 1.014e-05 [set_forward_comm_id_for_comm_node_pass]: 4.35999e-06 [meta_fg_expand]: 2.34999e-06 [flash_sp_send_recv_attached]: 1.64e-06 [receive_attached]: 2.38002e-06 [after_resolve]: 1.357e-05 [a_after_grad]: 1.078e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 1.90001e-06 [auto_monad_grad]: 1.61002e-06 [auto_monad_eliminator]: 9.69e-06 [cse]: 2.3e-05 [a_3]: 3.682e-05 [py_interpret_to_execute_after_opt_a]: 1.873e-05 [slice_cell_reuse_recomputed_activation]: 2.68998e-06 [rewriter_after_opt_a]: 5.572e-05 [convert_after_rewriter]: 1.013e-05 [order_py_execute_after_rewriter]: 7.55e-06 [mutable_eliminate]: 0.00080187 [opt_b]: 0.00025268, [1] [Cycle 1]: 0.00024351, [7] [b_1]: 0.00013307 [b_2]: 9.64e-06 [updatestate_depend_eliminate]: 1.129e-05 [updatestate_assign_eliminate]: 3.79002e-06 [updatestate_loads_eliminate]: 3.55003e-06 [renormalize]: 1.00999e-06 [cse]: 4.172e-05 [optimize_parallel_all_gather_comm]: 3.044e-05 [overlap_param_gather]: 2.40997e-06 [cconv]: 3.857e-05 [loop_unroll]: 0.00065273 [opt_after_cconv]: 0.00014389, [1] [Cycle 1]: 0.00013458, [7] [c_1]: 3.568e-05 [parameter_eliminate]: 6.37001e-06 [updatestate_depend_eliminate]: 1.044e-05 [updatestate_assign_eliminate]: 3.61999e-06 [updatestate_loads_eliminate]: 2.65002e-06 [cse]: 3.616e-05 [renormalize]: 7.89994e-07 [remove_dup_value]: 2.21e-05 [tuple_transform]: 9.95e-05, [1] [Cycle 1]: 9.389e-05, [4] [d_1]: 6.149e-05 [none_parameter_eliminate]: 2.43e-06 [renormalize]: 2.3999e-07 [switch_simplify]: 8.28999e-06 [partial_unused_args_eliminate]: 2.30002e-06 [add_recomputation]: 7.253e-05 [cse_after_recomputation]: 3.143e-05, [1] [Cycle 1]: 2.534e-05, [1] [cse]: 1.832e-05 [environ_conv]: 9.17001e-06 [swap_dp_allreduce_reducescatter]: 6.89001e-06 [bias_add_comm_swap]: 4.87e-06 [label_micro_interleaved_index]: 7.90998e-06 [label_fine_grained_interleaved_index]: 3.14001e-06 [merge_cast_opt]: 1.64e-06 [slice_recompute_activation]: 2.89001e-06 [micro_interleaved_order_control]: 3.51999e-06 [assign_add_opt]: 1.83997e-06 [ForceFp32Comm]: 1.02998e-06 [remove_cast_before_assign_add]: 1.55999e-06 [full_micro_interleaved_order_control]: 2.71999e-06 [reorder_send_recv_between_fp_bp]: 3.55e-06 [comm_op_add_attrs]: 1.40001e-06 [add_comm_op_reuse_tag]: 1.15999e-06 [interleave_split_concat_branches]: 1.83002e-06 [interleave_parallel_branches]: 1.24e-06 [overlap_opt_shard_in_pipeline]: 3.89002e-06 [overlap_opt_shard_grad_in_pipeline]: 2.20002e-06 [control_data_broadcast_order]: 2.027e-05 [grouped_pairwise_exchange_alltoall]: 3.07002e-06 [offloading_packed_experts]: 5.19e-06 [overlap_recompute_and_grad_model_parallel]: 6.85998e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.40999e-06 [overlap_recompute_allgather_and_fa_grad]: 2.45002e-06 [overlap_recompute_comm]: 2.75002e-06 [overlap_grad_ring_attention]: 4.96002e-06 [overlap_grad_flash_sp]: 3.053e-05 [begin_end_overlap_inline]: 5.79981e-07 [split_matmul_comm_elemetwise]: 2.91999e-06 [split_layernorm_comm]: 2.06e-06 [handle_group_info]: 1.21997e-06 [symbol_engine_optimizer]: 0.00010778, [1] [Cycle 1]: 0.00010222, [6] [build]: 5.95002e-06 [elim_shapecalc]: 2.277e-05 [elim_not_effective]: 1.907e-05 [opt_reshape]: 9.05001e-06 [fold_const_symbol]: 1.183e-05 [renormalize]: 2.09984e-07 [detach_backward]: 3.26001e-06 [pipeline_parallel_scheduler]: 2.01e-06 [auto_monad_reorder]: 3.406e-05 [get_jit_bprop_graph]: 2.46e-06 [rewriter_after_jit_bprop_graph]: 8.52e-06 [opt_after_jit_grad]: 0.00078099 [validate]: 0.0922446 Sums bootstrap : 0.000640s : 0.62% type_inference : 0.004469s : 4.33% event_method : 0.000019s : 0.02% auto_monad : 0.000128s : 0.12% graph_reusing : 0.000008s : 0.01% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000021s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000005s : 0.00% pre_auto_parallel : 0.000039s : 0.04% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000026s : 0.03% optimize.rewriter_before_opt_a : 0.000055s : 0.05% optimize.opt_a.expand_dump_flag : 0.000006s : 0.01% optimize.opt_a.switch_simplify : 0.000036s : 0.03% optimize.opt_a.loop_unroll : 0.000018s : 0.02% optimize.opt_a.a_1 : 0.000466s : 0.45% optimize.opt_a.with_stream_mark : 0.000043s : 0.04% optimize.opt_a.recompute_prepare : 0.000030s : 0.03% optimize.opt_a.updatestate_depend_eliminate : 0.000010s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.00% optimize.opt_a.a_2 : 0.000163s : 0.16% optimize.opt_a.accelerated_algorithm : 0.000013s : 0.01% optimize.opt_a.shard : 0.000004s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.00% optimize.opt_a.shard_inline : 0.000012s : 0.01% optimize.opt_a.merge_send_recv : 0.000017s : 0.02% optimize.opt_a.auto_parallel : 0.000017s : 0.02% optimize.opt_a.parallel : 0.000081s : 0.08% optimize.opt_a.flash_sp : 0.000016s : 0.02% optimize.opt_a.merge_comm : 0.000007s : 0.01% optimize.opt_a.allreduce_fusion : 0.000008s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000018s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000017s : 0.02% optimize.opt_a.virtual_dataset : 0.000049s : 0.05% optimize.opt_a.get_grad_eliminate_ : 0.000014s : 0.01% optimize.opt_a.virtual_output : 0.000012s : 0.01% optimize.opt_a.merge_forward : 0.000012s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.00% optimize.opt_a.offload_activation : 0.000027s : 0.03% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000031s : 0.03% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000021s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000009s : 0.01% optimize.opt_a.meta_fg_expand : 0.000006s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.00% optimize.opt_a.receive_attached : 0.000006s : 0.01% optimize.opt_a.after_resolve : 0.000026s : 0.03% optimize.opt_a.a_after_grad : 0.000020s : 0.02% optimize.opt_a.renormalize : 0.000816s : 0.79% optimize.opt_a.add_forward_monad_depend : 0.000009s : 0.01% optimize.opt_a.auto_monad_grad : 0.000006s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000027s : 0.03% optimize.opt_a.cse : 0.000069s : 0.07% optimize.opt_a.a_3 : 0.000095s : 0.09% optimize.py_interpret_to_execute_after_opt_a : 0.000019s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000056s : 0.05% optimize.convert_after_rewriter : 0.000010s : 0.01% optimize.order_py_execute_after_rewriter : 0.000008s : 0.01% optimize.mutable_eliminate : 0.000802s : 0.78% optimize.opt_b.b_1 : 0.000133s : 0.13% optimize.opt_b.b_2 : 0.000010s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000011s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000004s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000004s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000042s : 0.04% optimize.optimize_parallel_all_gather_comm : 0.000030s : 0.03% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000039s : 0.04% optimize.loop_unroll : 0.000653s : 0.63% optimize.opt_after_cconv.c_1 : 0.000036s : 0.03% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.01% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000010s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000004s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000003s : 0.00% optimize.opt_after_cconv.cse : 0.000036s : 0.04% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000022s : 0.02% optimize.tuple_transform.d_1 : 0.000061s : 0.06% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000008s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000073s : 0.07% optimize.cse_after_recomputation.cse : 0.000018s : 0.02% optimize.environ_conv : 0.000009s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000007s : 0.01% optimize.bias_add_comm_swap : 0.000005s : 0.00% optimize.label_micro_interleaved_index : 0.000008s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000004s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000002s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000004s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000002s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000004s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000020s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000003s : 0.00% optimize.offloading_packed_experts : 0.000005s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 0.01% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000005s : 0.00% optimize.overlap_grad_flash_sp : 0.000031s : 0.03% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000006s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000023s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000019s : 0.02% optimize.symbol_engine_optimizer.opt_reshape : 0.000009s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000012s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000034s : 0.03% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000009s : 0.01% opt_after_jit_grad : 0.000781s : 0.76% validate : 0.092245s : 89.47% Time group info: ------[substitution.] 0.000137 21 2.00% : 0.000003s : 2: substitution.elim_not_effective 1.27% : 0.000002s : 2: substitution.fold_const_symbol 5.05% : 0.000007s : 4: substitution.graph_param_transform 78.99% : 0.000109s : 1: substitution.inline 3.25% : 0.000004s : 4: substitution.j_node_and_user_rematch 4.18% : 0.000006s : 4: substitution.remove_not_recompute_node 5.25% : 0.000007s : 4: substitution.replace_old_param ------[type_inference.] 0.004387 2 88.52% : 0.003883s : 1: type_inference.infer 11.48% : 0.000504s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000107 1 100.00% : 0.000107s : 1: match.inline ------[predicate.] 0.000170 967 0.68% : 0.000001s : 9: predicate.accumulaten_eliminater 1.79% : 0.000003s : 4: predicate.ad_related_special_op_eliminate 0.67% : 0.000001s : 8: predicate.addn_check_dump 0.74% : 0.000001s : 9: predicate.addn_zero_filter 0.61% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.24% : 0.000004s : 17: predicate.arithmetic_simplify 0.81% : 0.000001s : 9: predicate.cast_eliminate 0.72% : 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 0.78% : 0.000001s : 8: predicate.depend_value_elim 0.77% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.78% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.69% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.71% : 0.000003s : 8: predicate.dumpgradient_eliminate 0.43% : 0.000001s : 4: predicate.elim_not_effective 0.48% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.11% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.04% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.93% : 0.000002s : 13: predicate.environ_get_depend_swap 1.62% : 0.000003s : 21: predicate.environ_get_eliminate 0.88% : 0.000001s : 13: predicate.environ_get_set_eliminate 0.70% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.76% : 0.000003s : 10: predicate.float_depend_g_call 0.67% : 0.000001s : 8: predicate.float_environ_get_switch 0.85% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.25% : 0.000000s : 4: predicate.fold_const_symbol 0.86% : 0.000001s : 8: predicate.get_grad_eliminate 0.35% : 0.000001s : 4: predicate.graph_param_transform 0.69% : 0.000001s : 8: predicate.incorporate_call 0.58% : 0.000001s : 8: predicate.incorporate_call_switch 6.35% : 0.000011s : 43: predicate.inline 1.40% : 0.000002s : 8: predicate.inline_without_move 0.39% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.92% : 0.000002s : 8: predicate.less_batch_normalization 1.75% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 1.96% : 0.000003s : 26: predicate.load_eliminater 2.36% : 0.000004s : 4: predicate.loop_unroll_after_grad 1.16% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.90% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.79% : 0.000001s : 8: predicate.merge_addn 0.72% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.74% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.84% : 0.000001s : 9: predicate.minmaximum_grad 1.88% : 0.000003s : 4: predicate.mutable_eliminate 0.54% : 0.000001s : 4: predicate.opt_reshape 0.74% : 0.000001s : 4: predicate.parallel_virtual_node 1.12% : 0.000002s : 10: predicate.partial_defer_inline 1.06% : 0.000002s : 13: predicate.partial_eliminate 0.68% : 0.000001s : 9: predicate.print_const_string_wrapper 0.73% : 0.000001s : 8: predicate.reduce_all_const_elim 1.08% : 0.000002s : 9: predicate.reduce_eliminate 1.98% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.86% : 0.000001s : 8: predicate.remove_not_recompute_node 1.13% : 0.000002s : 17: predicate.replace_applicator 0.91% : 0.000002s : 8: predicate.replace_old_param 0.47% : 0.000001s : 4: predicate.reset_defer_inline 0.72% : 0.000001s : 9: predicate.reshape_eliminate 0.75% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.68% : 0.000001s : 4: predicate.row_tensor_eliminate 1.23% : 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 1.23% : 0.000002s : 8: predicate.special_op_eliminate 0.84% : 0.000001s : 8: predicate.specialize_transform 1.21% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.90% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.52% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.77% : 0.000001s : 10: predicate.switch_defer_inline 1.51% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.74% : 0.000006s : 35: predicate.switch_simplify 0.64% : 0.000001s : 9: predicate.tile_eliminate 0.65% : 0.000001s : 9: predicate.transpose_eliminate 1.60% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.77% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.21% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.61% : 0.000006s : 25: predicate.tuple_list_get_item_eliminator 1.37% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.39% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.38% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 2.06% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 3.14% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.58% : 0.000001s : 4: predicate.value_based_eliminate 0.91% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.75% : 0.000001s : 8: predicate.virtual_output_eliminate 0.33% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.58% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000283 4 8.72% : 0.000025s : 1: func_graph_cloner_run.FuncGraphClonerGraph 91.28% : 0.000258s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.157798 192 0.00% : 0.000004s : 1: ForceFp32Comm 3.18% : 0.005011s : 1: add_attr 3.16% : 0.004987s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.05% : 0.000078s : 1: add_recomputation 0.00% : 0.000006s : 1: assign_add_opt 0.10% : 0.000151s : 1: auto_monad 0.03% : 0.000040s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.01% : 0.000008s : 1: bias_add_comm_swap 0.44% : 0.000695s : 1: bootstrap 0.03% : 0.000043s : 1: cconv 0.00% : 0.000005s : 1: comm_op_add_attrs 0.02% : 0.000024s : 1: control_data_broadcast_order 0.01% : 0.000014s : 1: convert_after_rewriter 0.02% : 0.000035s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000008s : 1: detach_backward 0.01% : 0.000013s : 1: environ_conv 81.24% : 0.128191s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000007s : 1: get_jit_bprop_graph 0.01% : 0.000012s : 1: graph_reusing 0.00% : 0.000006s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.01% : 0.000011s : 1: label_micro_interleaved_index 0.42% : 0.000669s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000007s : 1: micro_interleaved_order_control 0.52% : 0.000818s : 1: mutable_eliminate 0.01% : 0.000009s : 1: offloading_packed_experts 0.02% : 0.000024s : 1: opt.transform.loop_unroll_optimizer 0.02% : 0.000025s : 1: opt.transform.mutable_eliminate 0.59% : 0.000930s : 78: opt.transform.opt_a 0.02% : 0.000034s : 1: opt.transform.opt_after_cconv 0.02% : 0.000039s : 1: opt.transform.opt_after_jit_grad 0.07% : 0.000112s : 28: opt.transform.opt_b 0.04% : 0.000067s : 2: opt.transform.opt_trans_graph 0.04% : 0.000058s : 4: opt.transform.symbol_engine_opt 1.75% : 0.002763s : 1: opt_a 0.09% : 0.000148s : 1: opt_after_cconv 0.51% : 0.000802s : 1: opt_after_jit_grad 0.16% : 0.000257s : 1: opt_b 3.58% : 0.005645s : 1: optimize 0.02% : 0.000036s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000011s : 1: order_py_execute_after_rewriter 0.02% : 0.000035s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000008s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000007s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000006s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000010s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.01% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.03% : 0.000045s : 1: pre_auto_parallel 0.02% : 0.000031s : 1: py_interpret_to_execute 0.01% : 0.000022s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.02% : 0.000026s : 1: remove_dup_value 0.31% : 0.000495s : 1: renormalize.infer 0.20% : 0.000311s : 1: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.04% : 0.000063s : 1: rewriter_after_opt_a 0.04% : 0.000060s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.01% : 0.000010s : 1: swap_dp_allreduce_reducescatter 0.07% : 0.000111s : 1: symbol_engine_optimizer 0.06% : 0.000103s : 1: tuple_transform 2.85% : 0.004499s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[dim_type0-True-batch_size_list0-shape1-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_fp16[dim_type0-True-batch_size_list0-shape1-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:38:29.914.475 [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 = 0.0540509, [21] [bootstrap]: 0.00071761 [type_inference]: 0.0423145 [event_method]: 1.716e-05 [auto_monad]: 6.738e-05 [graph_reusing]: 5.87999e-06 [inline]: 3.4e-06 [add_attr]: 0.00474169, [1] [add_attr_with_inline]: 0.00472543, [1] [Cycle 1]: 6.786e-05, [2] [tag_attr]: 1.909e-05 [meta_addattr_fg_expand]: 3.55e-06 [parallel-infer-symbol]: 4.1e-06 [pre_auto_parallel]: 4.058e-05 [insert-virtual-dataset]: 3.66999e-06 [parallel-infer-symbol-second]: 6.69999e-07 [dataset_repeat_opt]: 2.27999e-06 [pipeline_split]: 1.69e-06 [optimize]: 0.00520299, [53] [py_interpret_to_execute]: 2.393e-05 [rewriter_before_opt_a]: 6.894e-05 [opt_a]: 0.00258427, [2] [Cycle 1]: 0.00187355, [45] [expand_dump_flag]: 3.26001e-06 [switch_simplify]: 2.504e-05 [loop_unroll]: 1.151e-05 [a_1]: 0.00031151 [with_stream_mark]: 2.505e-05 [recompute_prepare]: 1.006e-05 [updatestate_depend_eliminate]: 4.87998e-06 [updatestate_assign_eliminate]: 3.95998e-06 [updatestate_loads_eliminate]: 3.90998e-06 [parameter_eliminate]: 2.09e-06 [a_2]: 8.293e-05 [accelerated_algorithm]: 8.15999e-06 [shard]: 2.41e-06 [meta_shard_fg_expand]: 1.72999e-06 [shard_inline]: 7.00998e-06 [merge_send_recv]: 9.94999e-06 [auto_parallel]: 7.4e-06 [parallel]: 3.474e-05 [flash_sp]: 1.046e-05 [merge_comm]: 3.84002e-06 [allreduce_fusion]: 3.86999e-06 [matmul_add_comm_reduction]: 1.139e-05 [allreduce_slice_to_reducescatter]: 6.00005e-07 [virtual_shard_identity]: 9.72999e-06 [virtual_dataset]: 7.05998e-06 [get_grad_eliminate_]: 6.93e-06 [virtual_output]: 6.55002e-06 [merge_forward]: 4.83001e-06 [cell_reuse_recompute_pass]: 1.45999e-06 [offload_activation]: 1.294e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.438e-05 [merge_recompute_call_nodes]: 1.62001e-06 [before_grad]: 1.127e-05 [set_forward_comm_id_for_comm_node_pass]: 3.85e-06 [meta_fg_expand]: 2.93e-06 [flash_sp_send_recv_attached]: 2.79001e-06 [receive_attached]: 2.41e-06 [after_resolve]: 1.212e-05 [a_after_grad]: 9.56e-06 [renormalize]: 0.00079708 [add_forward_monad_depend]: 8.74998e-06 [auto_monad_grad]: 3.41999e-06 [auto_monad_eliminator]: 1.801e-05 [cse]: 4.125e-05 [a_3]: 5.711e-05 [Cycle 2]: 0.00069736, [45] [expand_dump_flag]: 1.89999e-06 [switch_simplify]: 8.51002e-06 [loop_unroll]: 7.72002e-06 [a_1]: 0.00012598 [with_stream_mark]: 1.789e-05 [recompute_prepare]: 6.72002e-06 [updatestate_depend_eliminate]: 3.48999e-06 [updatestate_assign_eliminate]: 3.09999e-06 [updatestate_loads_eliminate]: 2.81e-06 [parameter_eliminate]: 1.90001e-06 [a_2]: 7.163e-05 [accelerated_algorithm]: 6.44999e-06 [shard]: 1.86998e-06 [meta_shard_fg_expand]: 2.24999e-06 [shard_inline]: 6.38003e-06 [merge_send_recv]: 8.09002e-06 [auto_parallel]: 7.93999e-06 [parallel]: 7.61999e-06 [flash_sp]: 3.85e-06 [merge_comm]: 3.96001e-06 [allreduce_fusion]: 3.35e-06 [matmul_add_comm_reduction]: 8.72e-06 [allreduce_slice_to_reducescatter]: 3.00002e-07 [virtual_shard_identity]: 7.63001e-06 [virtual_dataset]: 6.31e-06 [get_grad_eliminate_]: 6.22001e-06 [virtual_output]: 5.67001e-06 [merge_forward]: 4.09002e-06 [cell_reuse_recompute_pass]: 2.03002e-06 [offload_activation]: 1.013e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.836e-05 [merge_recompute_call_nodes]: 8.70001e-07 [before_grad]: 9.92001e-06 [set_forward_comm_id_for_comm_node_pass]: 4.15999e-06 [meta_fg_expand]: 2.70002e-06 [flash_sp_send_recv_attached]: 1.25999e-06 [receive_attached]: 1.55001e-06 [after_resolve]: 1.233e-05 [a_after_grad]: 9.29e-06 [renormalize]: 1.09983e-07 [add_forward_monad_depend]: 2.07999e-06 [auto_monad_grad]: 1.24e-06 [auto_monad_eliminator]: 1.078e-05 [cse]: 2.188e-05 [a_3]: 3.701e-05 [py_interpret_to_execute_after_opt_a]: 1.921e-05 [slice_cell_reuse_recomputed_activation]: 2.76999e-06 [rewriter_after_opt_a]: 5.127e-05 [convert_after_rewriter]: 7.63999e-06 [order_py_execute_after_rewriter]: 6.14001e-06 [mutable_eliminate]: 0.0007422 [opt_b]: 0.00023413, [1] [Cycle 1]: 0.00022498, [7] [b_1]: 0.00012993 [b_2]: 9.22999e-06 [updatestate_depend_eliminate]: 1.022e-05 [updatestate_assign_eliminate]: 3.44001e-06 [updatestate_loads_eliminate]: 3.17002e-06 [renormalize]: 9.70002e-07 [cse]: 3.199e-05 [optimize_parallel_all_gather_comm]: 2.456e-05 [overlap_param_gather]: 2.34999e-06 [cconv]: 3.494e-05 [loop_unroll]: 0.00053805 [opt_after_cconv]: 0.00011756, [1] [Cycle 1]: 0.00011045, [7] [c_1]: 3.185e-05 [parameter_eliminate]: 5.39e-06 [updatestate_depend_eliminate]: 7.52002e-06 [updatestate_assign_eliminate]: 2.70002e-06 [updatestate_loads_eliminate]: 3.14001e-06 [cse]: 2.317e-05 [renormalize]: 8.80013e-07 [remove_dup_value]: 1.837e-05 [tuple_transform]: 7.9e-05, [1] [Cycle 1]: 7.429e-05, [4] [d_1]: 4.773e-05 [none_parameter_eliminate]: 1.63002e-06 [renormalize]: 2.19996e-07 [switch_simplify]: 6.81999e-06 [partial_unused_args_eliminate]: 1.84e-06 [add_recomputation]: 0.00015025 [cse_after_recomputation]: 3.05e-05, [1] [Cycle 1]: 2.509e-05, [1] [cse]: 1.8e-05 [environ_conv]: 7.78999e-06 [swap_dp_allreduce_reducescatter]: 6.18998e-06 [bias_add_comm_swap]: 3.44001e-06 [label_micro_interleaved_index]: 4.83001e-06 [label_fine_grained_interleaved_index]: 3.26999e-06 [merge_cast_opt]: 1.52001e-06 [slice_recompute_activation]: 2.53003e-06 [micro_interleaved_order_control]: 2.81e-06 [assign_add_opt]: 1.34998e-06 [ForceFp32Comm]: 1.12e-06 [remove_cast_before_assign_add]: 1.09e-06 [full_micro_interleaved_order_control]: 2.81e-06 [reorder_send_recv_between_fp_bp]: 3.11001e-06 [comm_op_add_attrs]: 1.19e-06 [add_comm_op_reuse_tag]: 1.17e-06 [interleave_split_concat_branches]: 1.25999e-06 [interleave_parallel_branches]: 1.27999e-06 [overlap_opt_shard_in_pipeline]: 1.49998e-06 [overlap_opt_shard_grad_in_pipeline]: 2.31e-06 [control_data_broadcast_order]: 1.511e-05 [grouped_pairwise_exchange_alltoall]: 1.73002e-06 [offloading_packed_experts]: 4.60001e-06 [overlap_recompute_and_grad_model_parallel]: 5.86e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.27999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.54e-06 [overlap_recompute_comm]: 3.11001e-06 [overlap_grad_ring_attention]: 4.87e-06 [overlap_grad_flash_sp]: 2.689e-05 [begin_end_overlap_inline]: 5.19998e-07 [split_matmul_comm_elemetwise]: 2.96001e-06 [split_layernorm_comm]: 2.77002e-06 [handle_group_info]: 1.25999e-06 [symbol_engine_optimizer]: 8.902e-05, [1] [Cycle 1]: 8.287e-05, [6] [build]: 4.03999e-06 [elim_shapecalc]: 1.306e-05 [elim_not_effective]: 1.549e-05 [opt_reshape]: 7.30998e-06 [fold_const_symbol]: 1.048e-05 [renormalize]: 2.29978e-07 [detach_backward]: 2.23998e-06 [pipeline_parallel_scheduler]: 2.02001e-06 [auto_monad_reorder]: 2.779e-05 [get_jit_bprop_graph]: 2.16e-06 [rewriter_after_jit_bprop_graph]: 6.78e-06 [opt_after_jit_grad]: 0.00057053 [validate]: 5.277e-05 Sums bootstrap : 0.000718s : 1.49% type_inference : 0.042315s : 87.83% event_method : 0.000017s : 0.04% auto_monad : 0.000067s : 0.14% graph_reusing : 0.000006s : 0.01% inline : 0.000003s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000019s : 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 : 0.000041s : 0.08% insert-virtual-dataset : 0.000004s : 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 : 0.000024s : 0.05% optimize.rewriter_before_opt_a : 0.000069s : 0.14% optimize.opt_a.expand_dump_flag : 0.000005s : 0.01% optimize.opt_a.switch_simplify : 0.000034s : 0.07% optimize.opt_a.loop_unroll : 0.000019s : 0.04% optimize.opt_a.a_1 : 0.000437s : 0.91% optimize.opt_a.with_stream_mark : 0.000043s : 0.09% 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 : 0.000007s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.01% optimize.opt_a.a_2 : 0.000155s : 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 : 0.000004s : 0.01% optimize.opt_a.shard_inline : 0.000013s : 0.03% optimize.opt_a.merge_send_recv : 0.000018s : 0.04% optimize.opt_a.auto_parallel : 0.000015s : 0.03% optimize.opt_a.parallel : 0.000042s : 0.09% 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 : 0.000020s : 0.04% 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 : 0.000013s : 0.03% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.03% optimize.opt_a.virtual_output : 0.000012s : 0.03% optimize.opt_a.merge_forward : 0.000009s : 0.02% optimize.opt_a.cell_reuse_recompute_pass : 0.000003s : 0.01% optimize.opt_a.offload_activation : 0.000023s : 0.05% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000033s : 0.07% optimize.opt_a.merge_recompute_call_nodes : 0.000002s : 0.01% optimize.opt_a.before_grad : 0.000021s : 0.04% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 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 : 0.000024s : 0.05% optimize.opt_a.a_after_grad : 0.000019s : 0.04% optimize.opt_a.renormalize : 0.000797s : 1.65% optimize.opt_a.add_forward_monad_depend : 0.000011s : 0.02% optimize.opt_a.auto_monad_grad : 0.000005s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000029s : 0.06% optimize.opt_a.cse : 0.000063s : 0.13% optimize.opt_a.a_3 : 0.000094s : 0.20% optimize.py_interpret_to_execute_after_opt_a : 0.000019s : 0.04% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.01% optimize.rewriter_after_opt_a : 0.000051s : 0.11% optimize.convert_after_rewriter : 0.000008s : 0.02% optimize.order_py_execute_after_rewriter : 0.000006s : 0.01% optimize.mutable_eliminate : 0.000742s : 1.54% optimize.opt_b.b_1 : 0.000130s : 0.27% optimize.opt_b.b_2 : 0.000009s : 0.02% optimize.opt_b.updatestate_depend_eliminate : 0.000010s : 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 : 0.000032s : 0.07% optimize.optimize_parallel_all_gather_comm : 0.000025s : 0.05% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000035s : 0.07% optimize.loop_unroll : 0.000538s : 1.12% optimize.opt_after_cconv.c_1 : 0.000032s : 0.07% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 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 : 0.000023s : 0.05% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000018s : 0.04% optimize.tuple_transform.d_1 : 0.000048s : 0.10% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000150s : 0.31% optimize.cse_after_recomputation.cse : 0.000018s : 0.04% optimize.environ_conv : 0.000008s : 0.02% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.01% optimize.bias_add_comm_swap : 0.000003s : 0.01% optimize.label_micro_interleaved_index : 0.000005s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.01% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.01% optimize.micro_interleaved_order_control : 0.000003s : 0.01% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.01% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.01% optimize.comm_op_add_attrs : 0.000001s : 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 : 0.000005s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000006s : 0.01% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.01% optimize.overlap_grad_ring_attention : 0.000005s : 0.01% optimize.overlap_grad_flash_sp : 0.000027s : 0.06% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.01% optimize.split_layernorm_comm : 0.000003s : 0.01% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000004s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000013s : 0.03% optimize.symbol_engine_optimizer.elim_not_effective : 0.000015s : 0.03% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.02% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 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 : 0.000028s : 0.06% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.01% opt_after_jit_grad : 0.000571s : 1.18% validate : 0.000053s : 0.11% Time group info: ------[substitution.] 0.000122 21 1.66% : 0.000002s : 2: substitution.elim_not_effective 1.18% : 0.000001s : 2: substitution.fold_const_symbol 5.24% : 0.000006s : 4: substitution.graph_param_transform 79.32% : 0.000097s : 1: substitution.inline 3.15% : 0.000004s : 4: substitution.j_node_and_user_rematch 4.54% : 0.000006s : 4: substitution.remove_not_recompute_node 4.91% : 0.000006s : 4: substitution.replace_old_param ------[type_inference.] 0.042238 2 98.81% : 0.041736s : 1: type_inference.infer 1.19% : 0.000503s : 1: type_inference.specialize ------[replace.] 0.000020 1 100.00% : 0.000020s : 1: replace.inline ------[match.] 0.000095 1 100.00% : 0.000095s : 1: match.inline ------[predicate.] 0.000156 967 0.75% : 0.000001s : 9: predicate.accumulaten_eliminater 1.32% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.59% : 0.000001s : 8: predicate.addn_check_dump 0.91% : 0.000001s : 9: predicate.addn_zero_filter 0.67% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.15% : 0.000003s : 17: predicate.arithmetic_simplify 0.88% : 0.000001s : 9: predicate.cast_eliminate 0.83% : 0.000001s : 8: predicate.check_bprop_eliminate 0.85% : 0.000001s : 8: predicate.compare_switch_simplify 0.24% : 0.000000s : 4: predicate.const_output_eliminate 0.81% : 0.000001s : 8: predicate.depend_value_elim 0.84% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.84% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.70% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.26% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.28% : 0.000000s : 4: predicate.elim_not_effective 0.47% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.08% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.01% : 0.000002s : 13: predicate.environ_get_add_eliminate 1.03% : 0.000002s : 13: predicate.environ_get_depend_swap 1.67% : 0.000003s : 21: predicate.environ_get_eliminate 1.00% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.76% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.68% : 0.000003s : 10: predicate.float_depend_g_call 0.65% : 0.000001s : 8: predicate.float_environ_get_switch 0.92% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.25% : 0.000000s : 4: predicate.fold_const_symbol 0.91% : 0.000001s : 8: predicate.get_grad_eliminate 0.38% : 0.000001s : 4: predicate.graph_param_transform 0.72% : 0.000001s : 8: predicate.incorporate_call 0.58% : 0.000001s : 8: predicate.incorporate_call_switch 6.36% : 0.000010s : 43: predicate.inline 1.05% : 0.000002s : 8: predicate.inline_without_move 0.41% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.95% : 0.000001s : 8: predicate.less_batch_normalization 1.57% : 0.000002s : 17: predicate.list_to_tuple_eliminator_ 2.19% : 0.000003s : 26: predicate.load_eliminater 1.32% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.27% : 0.000002s : 13: predicate.loop_unroll_before_grad 2.01% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.74% : 0.000001s : 8: predicate.merge_addn 0.71% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.76% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.67% : 0.000001s : 9: predicate.minmaximum_grad 1.89% : 0.000003s : 4: predicate.mutable_eliminate 0.51% : 0.000001s : 4: predicate.opt_reshape 0.46% : 0.000001s : 4: predicate.parallel_virtual_node 1.10% : 0.000002s : 10: predicate.partial_defer_inline 1.13% : 0.000002s : 13: predicate.partial_eliminate 0.69% : 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 2.05% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.80% : 0.000001s : 8: predicate.remove_not_recompute_node 1.53% : 0.000002s : 17: predicate.replace_applicator 0.86% : 0.000001s : 8: predicate.replace_old_param 0.50% : 0.000001s : 4: predicate.reset_defer_inline 0.88% : 0.000001s : 9: predicate.reshape_eliminate 0.95% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.55% : 0.000001s : 4: predicate.row_tensor_eliminate 1.29% : 0.000002s : 8: predicate.same_eliminate 0.58% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.31% : 0.000002s : 8: predicate.shard_identity_eliminate 1.01% : 0.000002s : 8: predicate.special_op_eliminate 0.79% : 0.000001s : 8: predicate.specialize_transform 1.37% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 1.03% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.38% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.86% : 0.000001s : 10: predicate.switch_defer_inline 1.63% : 0.000003s : 18: predicate.switch_layer_defer_inline 4.06% : 0.000006s : 35: predicate.switch_simplify 0.82% : 0.000001s : 9: predicate.tile_eliminate 0.74% : 0.000001s : 9: predicate.transpose_eliminate 1.95% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.58% : 0.000002s : 17: predicate.tuple_list_get_item_const_eliminator 1.59% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.22% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.49% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.22% : 0.000003s : 25: predicate.tuple_list_set_item_eliminator 1.42% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 1.85% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.66% : 0.000004s : 34: predicate.updatestate_useless_node_eliminater 0.49% : 0.000001s : 4: predicate.value_based_eliminate 1.26% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.81% : 0.000001s : 8: predicate.virtual_output_eliminate 0.32% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.52% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000288 4 8.35% : 0.000024s : 1: func_graph_cloner_run.FuncGraphClonerGraph 91.65% : 0.000264s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.065696 192 0.01% : 0.000004s : 1: ForceFp32Comm 7.23% : 0.004749s : 1: add_attr 7.20% : 0.004730s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.24% : 0.000158s : 1: add_recomputation 0.01% : 0.000005s : 1: assign_add_opt 0.11% : 0.000073s : 1: auto_monad 0.05% : 0.000032s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.01% : 0.000007s : 1: bias_add_comm_swap 1.16% : 0.000763s : 1: bootstrap 0.06% : 0.000039s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.03% : 0.000019s : 1: control_data_broadcast_order 0.02% : 0.000011s : 1: convert_after_rewriter 0.05% : 0.000034s : 1: cse_after_recomputation 0.01% : 0.000005s : 1: dataset_repeat_opt 0.01% : 0.000006s : 1: detach_backward 0.02% : 0.000011s : 1: environ_conv 0.04% : 0.000025s : 1: event_method 0.01% : 0.000006s : 1: full_micro_interleaved_order_control 0.01% : 0.000005s : 1: get_jit_bprop_graph 0.01% : 0.000009s : 1: graph_reusing 0.01% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000004s : 1: handle_group_info 0.01% : 0.000007s : 1: inline 0.01% : 0.000007s : 1: insert-virtual-dataset 0.01% : 0.000005s : 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 0.84% : 0.000549s : 1: loop_unroll 0.01% : 0.000004s : 1: merge_cast_opt 0.01% : 0.000005s : 1: micro_interleaved_order_control 1.15% : 0.000756s : 1: mutable_eliminate 0.01% : 0.000008s : 1: offloading_packed_experts 0.03% : 0.000016s : 1: opt.transform.loop_unroll_optimizer 0.03% : 0.000022s : 1: opt.transform.mutable_eliminate 1.29% : 0.000848s : 78: opt.transform.opt_a 0.05% : 0.000030s : 1: opt.transform.opt_after_cconv 0.04% : 0.000028s : 1: opt.transform.opt_after_jit_grad 0.16% : 0.000108s : 28: opt.transform.opt_b 0.08% : 0.000053s : 2: opt.transform.opt_trans_graph 0.06% : 0.000042s : 4: opt.transform.symbol_engine_opt 3.94% : 0.002588s : 1: opt_a 0.18% : 0.000121s : 1: opt_after_cconv 0.89% : 0.000581s : 1: opt_after_jit_grad 0.36% : 0.000238s : 1: opt_b 7.93% : 0.005210s : 1: optimize 0.04% : 0.000029s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000010s : 1: order_py_execute_after_rewriter 0.05% : 0.000030s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 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% : 0.000006s : 1: overlap_param_gather 0.01% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000009s : 1: overlap_recompute_and_grad_model_parallel 0.01% : 0.000006s : 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% : 0.000005s : 1: pipeline_split 0.07% : 0.000046s : 1: pre_auto_parallel 0.04% : 0.000028s : 1: py_interpret_to_execute 0.03% : 0.000023s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000004s : 1: remove_cast_before_assign_add 0.03% : 0.000022s : 1: remove_dup_value 0.72% : 0.000474s : 1: renormalize.infer 0.47% : 0.000312s : 1: renormalize.specialize 0.01% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.09% : 0.000057s : 1: rewriter_after_opt_a 0.11% : 0.000074s : 1: rewriter_before_opt_a 0.01% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.01% : 0.000005s : 1: slice_recompute_activation 0.01% : 0.000006s : 1: split_layernorm_comm 0.01% : 0.000006s : 1: split_matmul_comm_elemetwise 0.01% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.14% : 0.000092s : 1: symbol_engine_optimizer 0.12% : 0.000082s : 1: tuple_transform 64.45% : 0.042340s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[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_fp16[dim_type0-True-batch_size_list1-shape0-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:38:32.536.270 [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 = 0.0291938, [21] [bootstrap]: 0.00120912 [type_inference]: 0.00463669 [event_method]: 1.86e-05 [auto_monad]: 5.692e-05 [graph_reusing]: 5.44998e-06 [inline]: 2.78e-06 [add_attr]: 0.00480197, [1] [add_attr_with_inline]: 0.00478786, [1] [Cycle 1]: 6.548e-05, [2] [tag_attr]: 1.636e-05 [meta_addattr_fg_expand]: 3.50998e-06 [parallel-infer-symbol]: 3.76999e-06 [pre_auto_parallel]: 3.653e-05 [insert-virtual-dataset]: 2.81999e-06 [parallel-infer-symbol-second]: 8.89995e-07 [dataset_repeat_opt]: 2.09e-06 [pipeline_split]: 1.66e-06 [optimize]: 0.0175106, [53] [py_interpret_to_execute]: 2.162e-05 [rewriter_before_opt_a]: 4.476e-05 [opt_a]: 0.0149607, [2] [Cycle 1]: 0.0142014, [45] [expand_dump_flag]: 2.79001e-06 [switch_simplify]: 2.378e-05 [loop_unroll]: 1.153e-05 [a_1]: 0.0003324 [with_stream_mark]: 2.171e-05 [recompute_prepare]: 8.85001e-06 [updatestate_depend_eliminate]: 4.37998e-06 [updatestate_assign_eliminate]: 3.90998e-06 [updatestate_loads_eliminate]: 4e-06 [parameter_eliminate]: 2.20002e-06 [a_2]: 8.593e-05 [accelerated_algorithm]: 7.05002e-06 [shard]: 2.56e-06 [meta_shard_fg_expand]: 2.02001e-06 [shard_inline]: 7.19001e-06 [merge_send_recv]: 9.27001e-06 [auto_parallel]: 9.12999e-06 [parallel]: 3.002e-05 [flash_sp]: 1.005e-05 [merge_comm]: 3.97002e-06 [allreduce_fusion]: 3.58999e-06 [matmul_add_comm_reduction]: 1.078e-05 [allreduce_slice_to_reducescatter]: 1.14003e-06 [virtual_shard_identity]: 8.40001e-06 [virtual_dataset]: 6.58e-06 [get_grad_eliminate_]: 6.07001e-06 [virtual_output]: 6.39001e-06 [merge_forward]: 4.94e-06 [cell_reuse_recompute_pass]: 1.12999e-06 [offload_activation]: 1.124e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.614e-05 [merge_recompute_call_nodes]: 1.54998e-06 [before_grad]: 1.043e-05 [set_forward_comm_id_for_comm_node_pass]: 3.73001e-06 [meta_fg_expand]: 2.67001e-06 [flash_sp_send_recv_attached]: 2.71e-06 [receive_attached]: 2.13998e-06 [after_resolve]: 1.223e-05 [a_after_grad]: 1.014e-05 [renormalize]: 0.0130844 [add_forward_monad_depend]: 1.008e-05 [auto_monad_grad]: 3.05002e-06 [auto_monad_eliminator]: 2.002e-05 [cse]: 3.758e-05 [a_3]: 6.383e-05 [Cycle 2]: 0.00074652, [45] [expand_dump_flag]: 2.42001e-06 [switch_simplify]: 1.013e-05 [loop_unroll]: 6.31e-06 [a_1]: 0.00013204 [with_stream_mark]: 2.002e-05 [recompute_prepare]: 6.56999e-06 [updatestate_depend_eliminate]: 4.62e-06 [updatestate_assign_eliminate]: 3.73001e-06 [updatestate_loads_eliminate]: 3.51999e-06 [parameter_eliminate]: 1.94e-06 [a_2]: 7.297e-05 [accelerated_algorithm]: 7.56999e-06 [shard]: 2.68e-06 [meta_shard_fg_expand]: 2.56e-06 [shard_inline]: 6.63e-06 [merge_send_recv]: 9.97001e-06 [auto_parallel]: 1.119e-05 [parallel]: 9.14e-06 [flash_sp]: 4.82998e-06 [merge_comm]: 3.89002e-06 [allreduce_fusion]: 3.65e-06 [matmul_add_comm_reduction]: 1.463e-05 [allreduce_slice_to_reducescatter]: 5.89993e-07 [virtual_shard_identity]: 7.95e-06 [virtual_dataset]: 6.83e-06 [get_grad_eliminate_]: 6.25002e-06 [virtual_output]: 6.00002e-06 [merge_forward]: 5.14998e-06 [cell_reuse_recompute_pass]: 3.20998e-06 [offload_activation]: 1.768e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.649e-05 [merge_recompute_call_nodes]: 1.66998e-06 [before_grad]: 1.112e-05 [set_forward_comm_id_for_comm_node_pass]: 4.28999e-06 [meta_fg_expand]: 3.31001e-06 [flash_sp_send_recv_attached]: 2.48998e-06 [receive_attached]: 2.43e-06 [after_resolve]: 1.33e-05 [a_after_grad]: 1.011e-05 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 1.52999e-06 [auto_monad_grad]: 1.13001e-06 [auto_monad_eliminator]: 8.13001e-06 [cse]: 2.226e-05 [a_3]: 3.746e-05 [py_interpret_to_execute_after_opt_a]: 1.896e-05 [slice_cell_reuse_recomputed_activation]: 2.20002e-06 [rewriter_after_opt_a]: 4.33e-05 [convert_after_rewriter]: 7.61001e-06 [order_py_execute_after_rewriter]: 5.84999e-06 [mutable_eliminate]: 0.00077259 [opt_b]: 0.00023854, [1] [Cycle 1]: 0.00022799, [7] [b_1]: 0.00013083 [b_2]: 8.60001e-06 [updatestate_depend_eliminate]: 8.95001e-06 [updatestate_assign_eliminate]: 3.41001e-06 [updatestate_loads_eliminate]: 3.01999e-06 [renormalize]: 1.10001e-06 [cse]: 3.593e-05 [optimize_parallel_all_gather_comm]: 2.055e-05 [overlap_param_gather]: 1.94e-06 [cconv]: 3.589e-05 [loop_unroll]: 0.0005037 [opt_after_cconv]: 0.00012214, [1] [Cycle 1]: 0.00011533, [7] [c_1]: 3.175e-05 [parameter_eliminate]: 5.02e-06 [updatestate_depend_eliminate]: 7.13e-06 [updatestate_assign_eliminate]: 2.68998e-06 [updatestate_loads_eliminate]: 2.54001e-06 [cse]: 2.377e-05 [renormalize]: 9.39996e-07 [remove_dup_value]: 1.842e-05 [tuple_transform]: 8.025e-05, [1] [Cycle 1]: 7.579e-05, [4] [d_1]: 4.952e-05 [none_parameter_eliminate]: 1.76e-06 [renormalize]: 1.50001e-07 [switch_simplify]: 7.17997e-06 [partial_unused_args_eliminate]: 2.21e-06 [add_recomputation]: 0.00011825 [cse_after_recomputation]: 3.301e-05, [1] [Cycle 1]: 2.833e-05, [1] [cse]: 2.271e-05 [environ_conv]: 8.13999e-06 [swap_dp_allreduce_reducescatter]: 6.05002e-06 [bias_add_comm_swap]: 2.84999e-06 [label_micro_interleaved_index]: 5.63002e-06 [label_fine_grained_interleaved_index]: 2.63e-06 [merge_cast_opt]: 1.49e-06 [slice_recompute_activation]: 2.79001e-06 [micro_interleaved_order_control]: 2.81999e-06 [assign_add_opt]: 1.77001e-06 [ForceFp32Comm]: 1.35001e-06 [remove_cast_before_assign_add]: 1.34003e-06 [full_micro_interleaved_order_control]: 2.34001e-06 [reorder_send_recv_between_fp_bp]: 3.01001e-06 [comm_op_add_attrs]: 1.14998e-06 [add_comm_op_reuse_tag]: 1.00001e-06 [interleave_split_concat_branches]: 1.47999e-06 [interleave_parallel_branches]: 1.05001e-06 [overlap_opt_shard_in_pipeline]: 1.38002e-06 [overlap_opt_shard_grad_in_pipeline]: 2.10002e-06 [control_data_broadcast_order]: 1.395e-05 [grouped_pairwise_exchange_alltoall]: 1.81e-06 [offloading_packed_experts]: 4.38001e-06 [overlap_recompute_and_grad_model_parallel]: 5.37999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.35001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.40001e-06 [overlap_recompute_comm]: 2.69999e-06 [overlap_grad_ring_attention]: 4.84e-06 [overlap_grad_flash_sp]: 2.301e-05 [begin_end_overlap_inline]: 5.50004e-07 [split_matmul_comm_elemetwise]: 2.21e-06 [split_layernorm_comm]: 2.00002e-06 [handle_group_info]: 1.19e-06 [symbol_engine_optimizer]: 8.518e-05, [1] [Cycle 1]: 8.066e-05, [6] [build]: 3.54002e-06 [elim_shapecalc]: 1.186e-05 [elim_not_effective]: 1.364e-05 [opt_reshape]: 8.42e-06 [fold_const_symbol]: 9.70002e-06 [renormalize]: 2.10013e-07 [detach_backward]: 2.61999e-06 [pipeline_parallel_scheduler]: 1.42e-06 [auto_monad_reorder]: 2.194e-05 [get_jit_bprop_graph]: 2.04e-06 [rewriter_after_jit_bprop_graph]: 7.77e-06 [opt_after_jit_grad]: 0.00058752 [validate]: 5.458e-05 Sums bootstrap : 0.001209s : 5.20% type_inference : 0.004637s : 19.94% event_method : 0.000019s : 0.08% auto_monad : 0.000057s : 0.24% graph_reusing : 0.000005s : 0.02% inline : 0.000003s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000016s : 0.07% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.02% parallel-infer-symbol : 0.000004s : 0.02% pre_auto_parallel : 0.000037s : 0.16% 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 : 0.000022s : 0.09% optimize.rewriter_before_opt_a : 0.000045s : 0.19% optimize.opt_a.expand_dump_flag : 0.000005s : 0.02% optimize.opt_a.switch_simplify : 0.000034s : 0.15% optimize.opt_a.loop_unroll : 0.000018s : 0.08% optimize.opt_a.a_1 : 0.000464s : 2.00% optimize.opt_a.with_stream_mark : 0.000042s : 0.18% optimize.opt_a.recompute_prepare : 0.000015s : 0.07% optimize.opt_a.updatestate_depend_eliminate : 0.000009s : 0.04% optimize.opt_a.updatestate_assign_eliminate : 0.000008s : 0.03% optimize.opt_a.updatestate_loads_eliminate : 0.000008s : 0.03% optimize.opt_a.parameter_eliminate : 0.000004s : 0.02% optimize.opt_a.a_2 : 0.000159s : 0.68% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.06% optimize.opt_a.shard : 0.000005s : 0.02% optimize.opt_a.meta_shard_fg_expand : 0.000005s : 0.02% optimize.opt_a.shard_inline : 0.000014s : 0.06% optimize.opt_a.merge_send_recv : 0.000019s : 0.08% optimize.opt_a.auto_parallel : 0.000020s : 0.09% optimize.opt_a.parallel : 0.000039s : 0.17% optimize.opt_a.flash_sp : 0.000015s : 0.06% optimize.opt_a.merge_comm : 0.000008s : 0.03% optimize.opt_a.allreduce_fusion : 0.000007s : 0.03% optimize.opt_a.matmul_add_comm_reduction : 0.000025s : 0.11% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000016s : 0.07% optimize.opt_a.virtual_dataset : 0.000013s : 0.06% optimize.opt_a.get_grad_eliminate_ : 0.000012s : 0.05% optimize.opt_a.virtual_output : 0.000012s : 0.05% optimize.opt_a.merge_forward : 0.000010s : 0.04% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.02% optimize.opt_a.offload_activation : 0.000029s : 0.12% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000033s : 0.14% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.01% optimize.opt_a.before_grad : 0.000022s : 0.09% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.03% optimize.opt_a.meta_fg_expand : 0.000006s : 0.03% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.02% optimize.opt_a.receive_attached : 0.000005s : 0.02% optimize.opt_a.after_resolve : 0.000026s : 0.11% optimize.opt_a.a_after_grad : 0.000020s : 0.09% optimize.opt_a.renormalize : 0.013085s : 56.26% optimize.opt_a.add_forward_monad_depend : 0.000012s : 0.05% optimize.opt_a.auto_monad_grad : 0.000004s : 0.02% optimize.opt_a.auto_monad_eliminator : 0.000028s : 0.12% optimize.opt_a.cse : 0.000060s : 0.26% optimize.opt_a.a_3 : 0.000101s : 0.44% optimize.py_interpret_to_execute_after_opt_a : 0.000019s : 0.08% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.01% optimize.rewriter_after_opt_a : 0.000043s : 0.19% optimize.convert_after_rewriter : 0.000008s : 0.03% optimize.order_py_execute_after_rewriter : 0.000006s : 0.03% optimize.mutable_eliminate : 0.000773s : 3.32% optimize.opt_b.b_1 : 0.000131s : 0.56% optimize.opt_b.b_2 : 0.000009s : 0.04% optimize.opt_b.updatestate_depend_eliminate : 0.000009s : 0.04% 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 : 0.000036s : 0.15% optimize.optimize_parallel_all_gather_comm : 0.000021s : 0.09% optimize.overlap_param_gather : 0.000002s : 0.01% optimize.cconv : 0.000036s : 0.15% optimize.loop_unroll : 0.000504s : 2.17% optimize.opt_after_cconv.c_1 : 0.000032s : 0.14% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.02% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000007s : 0.03% 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 : 0.10% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000018s : 0.08% optimize.tuple_transform.d_1 : 0.000050s : 0.21% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.01% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.03% optimize.partial_unused_args_eliminate : 0.000002s : 0.01% optimize.add_recomputation : 0.000118s : 0.51% optimize.cse_after_recomputation.cse : 0.000023s : 0.10% optimize.environ_conv : 0.000008s : 0.04% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.03% optimize.bias_add_comm_swap : 0.000003s : 0.01% optimize.label_micro_interleaved_index : 0.000006s : 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 : 0.000003s : 0.01% optimize.assign_add_opt : 0.000002s : 0.01% optimize.ForceFp32Comm : 0.000001s : 0.01% optimize.remove_cast_before_assign_add : 0.000001s : 0.01% optimize.full_micro_interleaved_order_control : 0.000002s : 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 : 0.000014s : 0.06% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 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 : 0.000003s : 0.01% optimize.overlap_grad_ring_attention : 0.000005s : 0.02% optimize.overlap_grad_flash_sp : 0.000023s : 0.10% 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 : 0.000014s : 0.06% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.04% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 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 : 0.000022s : 0.09% get_jit_bprop_graph : 0.000002s : 0.01% rewriter_after_jit_bprop_graph : 0.000008s : 0.03% opt_after_jit_grad : 0.000588s : 2.53% validate : 0.000055s : 0.23% Time group info: ------[substitution.] 0.000116 21 1.73% : 0.000002s : 2: substitution.elim_not_effective 1.10% : 0.000001s : 2: substitution.fold_const_symbol 5.38% : 0.000006s : 4: substitution.graph_param_transform 77.47% : 0.000090s : 1: substitution.inline 3.96% : 0.000005s : 4: substitution.j_node_and_user_rematch 4.88% : 0.000006s : 4: substitution.remove_not_recompute_node 5.48% : 0.000006s : 4: substitution.replace_old_param ------[type_inference.] 0.004575 2 87.49% : 0.004002s : 1: type_inference.infer 12.51% : 0.000573s : 1: type_inference.specialize ------[replace.] 0.000018 1 100.00% : 0.000018s : 1: replace.inline ------[match.] 0.000089 1 100.00% : 0.000089s : 1: match.inline ------[predicate.] 0.000167 967 0.78% : 0.000001s : 9: predicate.accumulaten_eliminater 1.43% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.65% : 0.000001s : 8: predicate.addn_check_dump 0.84% : 0.000001s : 9: predicate.addn_zero_filter 0.66% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.21% : 0.000004s : 17: predicate.arithmetic_simplify 0.90% : 0.000002s : 9: predicate.cast_eliminate 0.71% : 0.000001s : 8: predicate.check_bprop_eliminate 0.63% : 0.000001s : 8: predicate.compare_switch_simplify 0.23% : 0.000000s : 4: predicate.const_output_eliminate 0.71% : 0.000001s : 8: predicate.depend_value_elim 0.74% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.90% : 0.000002s : 9: predicate.dict_get_item_eliminator 0.96% : 0.000002s : 9: predicate.dict_set_item_eliminator 1.42% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.32% : 0.000001s : 4: predicate.elim_not_effective 0.48% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.38% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.97% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.97% : 0.000002s : 13: predicate.environ_get_depend_swap 1.71% : 0.000003s : 21: predicate.environ_get_eliminate 0.99% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.78% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.77% : 0.000003s : 10: predicate.float_depend_g_call 0.62% : 0.000001s : 8: predicate.float_environ_get_switch 0.98% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.23% : 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 0.67% : 0.000001s : 8: predicate.incorporate_call 0.55% : 0.000001s : 8: predicate.incorporate_call_switch 5.93% : 0.000010s : 43: predicate.inline 1.23% : 0.000002s : 8: predicate.inline_without_move 0.42% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.01% : 0.000002s : 8: predicate.less_batch_normalization 1.58% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 1.97% : 0.000003s : 26: predicate.load_eliminater 1.36% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.18% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.88% : 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 0.68% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.66% : 0.000001s : 9: predicate.minmaximum_grad 2.21% : 0.000004s : 4: predicate.mutable_eliminate 0.50% : 0.000001s : 4: predicate.opt_reshape 0.42% : 0.000001s : 4: predicate.parallel_virtual_node 1.10% : 0.000002s : 10: predicate.partial_defer_inline 1.05% : 0.000002s : 13: predicate.partial_eliminate 0.81% : 0.000001s : 9: predicate.print_const_string_wrapper 0.71% : 0.000001s : 8: predicate.reduce_all_const_elim 1.04% : 0.000002s : 9: predicate.reduce_eliminate 2.00% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.75% : 0.000001s : 8: predicate.remove_not_recompute_node 1.28% : 0.000002s : 17: predicate.replace_applicator 0.82% : 0.000001s : 8: predicate.replace_old_param 0.32% : 0.000001s : 4: predicate.reset_defer_inline 0.89% : 0.000001s : 9: predicate.reshape_eliminate 0.84% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.50% : 0.000001s : 4: predicate.row_tensor_eliminate 1.31% : 0.000002s : 8: predicate.same_eliminate 0.55% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.99% : 0.000002s : 8: predicate.shard_identity_eliminate 0.89% : 0.000001s : 8: predicate.special_op_eliminate 0.84% : 0.000001s : 8: predicate.specialize_transform 1.42% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 1.17% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.47% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.90% : 0.000002s : 10: predicate.switch_defer_inline 1.55% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.49% : 0.000006s : 35: predicate.switch_simplify 0.71% : 0.000001s : 9: predicate.tile_eliminate 1.08% : 0.000002s : 9: predicate.transpose_eliminate 2.34% : 0.000004s : 17: predicate.tuple_list_convert_item_index_to_positive 1.59% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.29% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.24% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.71% : 0.000003s : 17: predicate.tuple_list_get_set_item_eliminator 2.60% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.83% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.94% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.69% : 0.000004s : 34: predicate.updatestate_useless_node_eliminater 0.62% : 0.000001s : 4: predicate.value_based_eliminate 0.92% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.84% : 0.000001s : 8: predicate.virtual_output_eliminate 0.34% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.63% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000403 4 5.58% : 0.000022s : 1: func_graph_cloner_run.FuncGraphClonerGraph 94.42% : 0.000381s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.065585 192 0.01% : 0.000004s : 1: ForceFp32Comm 7.33% : 0.004808s : 1: add_attr 7.31% : 0.004792s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.19% : 0.000123s : 1: add_recomputation 0.01% : 0.000008s : 1: assign_add_opt 0.09% : 0.000062s : 1: auto_monad 0.04% : 0.000026s : 1: auto_monad_reorder 0.01% : 0.000004s : 1: begin_end_overlap_inline 0.01% : 0.000006s : 1: bias_add_comm_swap 1.94% : 0.001272s : 1: bootstrap 0.06% : 0.000040s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.03% : 0.000017s : 1: control_data_broadcast_order 0.02% : 0.000011s : 1: convert_after_rewriter 0.05% : 0.000036s : 1: cse_after_recomputation 0.01% : 0.000005s : 1: dataset_repeat_opt 0.01% : 0.000010s : 1: detach_backward 0.02% : 0.000011s : 1: environ_conv 0.04% : 0.000026s : 1: event_method 0.01% : 0.000006s : 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 0.01% : 0.000004s : 1: handle_group_info 0.01% : 0.000006s : 1: inline 0.01% : 0.000007s : 1: insert-virtual-dataset 0.01% : 0.000008s : 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% : 0.000009s : 1: label_micro_interleaved_index 0.78% : 0.000513s : 1: loop_unroll 0.01% : 0.000004s : 1: merge_cast_opt 0.01% : 0.000005s : 1: micro_interleaved_order_control 1.20% : 0.000788s : 1: mutable_eliminate 0.01% : 0.000007s : 1: offloading_packed_experts 0.02% : 0.000016s : 1: opt.transform.loop_unroll_optimizer 0.03% : 0.000023s : 1: opt.transform.mutable_eliminate 1.34% : 0.000879s : 78: opt.transform.opt_a 0.05% : 0.000030s : 1: opt.transform.opt_after_cconv 0.05% : 0.000030s : 1: opt.transform.opt_after_jit_grad 0.16% : 0.000108s : 28: opt.transform.opt_b 0.08% : 0.000054s : 2: opt.transform.opt_trans_graph 0.06% : 0.000040s : 4: opt.transform.symbol_engine_opt 22.82% : 0.014965s : 1: opt_a 0.19% : 0.000126s : 1: opt_after_cconv 0.92% : 0.000601s : 1: opt_after_jit_grad 0.37% : 0.000242s : 1: opt_b 26.71% : 0.017517s : 1: optimize 0.04% : 0.000024s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000009s : 1: order_py_execute_after_rewriter 0.04% : 0.000027s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 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% : 0.000006s : 1: overlap_param_gather 0.02% : 0.000011s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.01% : 0.000006s : 1: overlap_recompute_comm 0.01% : 0.000007s : 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 0.06% : 0.000042s : 1: pre_auto_parallel 0.04% : 0.000026s : 1: py_interpret_to_execute 0.04% : 0.000023s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000005s : 1: remove_cast_before_assign_add 0.03% : 0.000022s : 1: remove_dup_value 19.31% : 0.012662s : 1: renormalize.infer 0.62% : 0.000409s : 1: renormalize.specialize 0.01% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.02% : 0.000011s : 1: rewriter_after_jit_bprop_graph 0.07% : 0.000048s : 1: rewriter_after_opt_a 0.07% : 0.000049s : 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% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.13% : 0.000088s : 1: symbol_engine_optimizer 0.13% : 0.000083s : 1: tuple_transform 7.10% : 0.004659s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[dim_type0-True-batch_size_list1-shape1-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_fp16[dim_type0-True-batch_size_list1-shape1-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:38:38.800.378 [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 = 0.789861, [21] [bootstrap]: 0.011521 [type_inference]: 0.243223 [event_method]: 1.14e-05 [auto_monad]: 5.442e-05 [graph_reusing]: 5.59998e-06 [inline]: 2.39001e-06 [add_attr]: 0.429107, [1] [add_attr_with_inline]: 0.428972, [1] [Cycle 1]: 6.689e-05, [2] [tag_attr]: 1.629e-05 [meta_addattr_fg_expand]: 3.35e-06 [parallel-infer-symbol]: 3.71001e-06 [pre_auto_parallel]: 0.00013439 [insert-virtual-dataset]: 2.58998e-06 [parallel-infer-symbol-second]: 7.40023e-07 [dataset_repeat_opt]: 0.00070737 [pipeline_split]: 1.92001e-06 [optimize]: 0.103493, [53] [py_interpret_to_execute]: 2.272e-05 [rewriter_before_opt_a]: 4.96e-05 [opt_a]: 0.00634497, [2] [Cycle 1]: 0.00510729, [45] [expand_dump_flag]: 2.59001e-06 [switch_simplify]: 2.408e-05 [loop_unroll]: 1.16e-05 [a_1]: 0.00049275 [with_stream_mark]: 2.015e-05 [recompute_prepare]: 7.88001e-06 [updatestate_depend_eliminate]: 4.57e-06 [updatestate_assign_eliminate]: 3.56001e-06 [updatestate_loads_eliminate]: 3.36001e-06 [parameter_eliminate]: 1.94999e-06 [a_2]: 7.942e-05 [accelerated_algorithm]: 7e-06 [shard]: 2.54999e-06 [meta_shard_fg_expand]: 1.87001e-06 [shard_inline]: 7.25998e-06 [merge_send_recv]: 9.23002e-06 [auto_parallel]: 6.78998e-06 [parallel]: 0.0006839 [flash_sp]: 1.066e-05 [merge_comm]: 4.85999e-06 [allreduce_fusion]: 4.08001e-06 [matmul_add_comm_reduction]: 1.027e-05 [allreduce_slice_to_reducescatter]: 9.10019e-07 [virtual_shard_identity]: 8.25e-06 [virtual_dataset]: 6.25002e-06 [get_grad_eliminate_]: 6.01e-06 [virtual_output]: 6.06e-06 [merge_forward]: 4.50001e-06 [cell_reuse_recompute_pass]: 1.72001e-06 [offload_activation]: 1.101e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.454e-05 [merge_recompute_call_nodes]: 1.72999e-06 [before_grad]: 1.05e-05 [set_forward_comm_id_for_comm_node_pass]: 4.18001e-06 [meta_fg_expand]: 2.96999e-06 [flash_sp_send_recv_attached]: 2.71e-06 [receive_attached]: 2.83e-06 [after_resolve]: 1.194e-05 [a_after_grad]: 1.011e-05 [renormalize]: 0.00234814 [add_forward_monad_depend]: 6.37001e-06 [auto_monad_grad]: 2.72001e-06 [auto_monad_eliminator]: 1.562e-05 [cse]: 3.945e-05 [a_3]: 5.29e-05 [Cycle 2]: 0.00122631, [45] [expand_dump_flag]: 1.58002e-06 [switch_simplify]: 7.77e-06 [loop_unroll]: 6.64999e-06 [a_1]: 0.00012243 [with_stream_mark]: 1.501e-05 [recompute_prepare]: 6.51999e-06 [updatestate_depend_eliminate]: 5.96998e-06 [updatestate_assign_eliminate]: 3.75998e-06 [updatestate_loads_eliminate]: 3.16999e-06 [parameter_eliminate]: 1.84e-06 [a_2]: 7.121e-05 [accelerated_algorithm]: 6.76e-06 [shard]: 1.44998e-06 [meta_shard_fg_expand]: 2.41e-06 [shard_inline]: 6.71999e-06 [merge_send_recv]: 8.31002e-06 [auto_parallel]: 8.44002e-06 [parallel]: 8.15999e-06 [flash_sp]: 3.30998e-06 [merge_comm]: 4.13999e-06 [allreduce_fusion]: 3.6e-06 [matmul_add_comm_reduction]: 0.00012859 [allreduce_slice_to_reducescatter]: 2.89991e-07 [virtual_shard_identity]: 7.92e-06 [virtual_dataset]: 5.62999e-06 [get_grad_eliminate_]: 5.81e-06 [virtual_output]: 5.90002e-06 [merge_forward]: 3.70998e-06 [cell_reuse_recompute_pass]: 2.49001e-06 [offload_activation]: 0.0001495 [cell_reuse_handle_not_recompute_node_pass]: 1.481e-05 [merge_recompute_call_nodes]: 1.03001e-06 [before_grad]: 1.017e-05 [set_forward_comm_id_for_comm_node_pass]: 4.55999e-06 [meta_fg_expand]: 2.40002e-06 [flash_sp_send_recv_attached]: 1.25999e-06 [receive_attached]: 1.57001e-06 [after_resolve]: 1.085e-05 [a_after_grad]: 9.46e-06 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 2.02001e-06 [auto_monad_grad]: 1.19e-06 [auto_monad_eliminator]: 8.60999e-06 [cse]: 8.396e-05 [a_3]: 3.571e-05 [py_interpret_to_execute_after_opt_a]: 1.651e-05 [slice_cell_reuse_recomputed_activation]: 2.05002e-06 [rewriter_after_opt_a]: 4.226e-05 [convert_after_rewriter]: 7.60998e-06 [order_py_execute_after_rewriter]: 6.21e-06 [mutable_eliminate]: 0.0937034 [opt_b]: 0.00039587, [1] [Cycle 1]: 0.00038489, [7] [b_1]: 0.00013929 [b_2]: 8.53001e-06 [updatestate_depend_eliminate]: 1.136e-05 [updatestate_assign_eliminate]: 3.98999e-06 [updatestate_loads_eliminate]: 3.66001e-06 [renormalize]: 5.50004e-07 [cse]: 0.00017926 [optimize_parallel_all_gather_comm]: 2.609e-05 [overlap_param_gather]: 1.94e-06 [cconv]: 4.258e-05 [loop_unroll]: 0.00057766 [opt_after_cconv]: 0.00061562, [1] [Cycle 1]: 0.00060882, [7] [c_1]: 3.117e-05 [parameter_eliminate]: 6.16998e-06 [updatestate_depend_eliminate]: 6.81001e-06 [updatestate_assign_eliminate]: 2.73e-06 [updatestate_loads_eliminate]: 2.41e-06 [cse]: 0.00018167 [renormalize]: 5.50004e-07 [remove_dup_value]: 1.899e-05 [tuple_transform]: 8.168e-05, [1] [Cycle 1]: 7.651e-05, [4] [d_1]: 4.947e-05 [none_parameter_eliminate]: 2.21e-06 [renormalize]: 1.39989e-07 [switch_simplify]: 7.01001e-06 [partial_unused_args_eliminate]: 2.18998e-06 [add_recomputation]: 5.639e-05 [cse_after_recomputation]: 0.00011265, [1] [Cycle 1]: 0.00010779, [1] [cse]: 0.00010236 [environ_conv]: 7.97e-06 [swap_dp_allreduce_reducescatter]: 5.44e-06 [bias_add_comm_swap]: 3.33e-06 [label_micro_interleaved_index]: 4.48001e-06 [label_fine_grained_interleaved_index]: 2.63998e-06 [merge_cast_opt]: 1.32e-06 [slice_recompute_activation]: 2.52001e-06 [micro_interleaved_order_control]: 2.44001e-06 [assign_add_opt]: 1.86e-06 [ForceFp32Comm]: 9.00007e-07 [remove_cast_before_assign_add]: 1.19e-06 [full_micro_interleaved_order_control]: 2.21998e-06 [reorder_send_recv_between_fp_bp]: 3.24001e-06 [comm_op_add_attrs]: 1.07e-06 [add_comm_op_reuse_tag]: 1.07e-06 [interleave_split_concat_branches]: 1.20001e-06 [interleave_parallel_branches]: 1.15999e-06 [overlap_opt_shard_in_pipeline]: 1.54e-06 [overlap_opt_shard_grad_in_pipeline]: 2.10002e-06 [control_data_broadcast_order]: 1.382e-05 [grouped_pairwise_exchange_alltoall]: 1.91e-06 [offloading_packed_experts]: 3.98999e-06 [overlap_recompute_and_grad_model_parallel]: 4.93001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.15999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.52999e-06 [overlap_recompute_comm]: 3.19001e-06 [overlap_grad_ring_attention]: 4.64002e-06 [overlap_grad_flash_sp]: 2.359e-05 [begin_end_overlap_inline]: 5.09986e-07 [split_matmul_comm_elemetwise]: 2.58e-06 [split_layernorm_comm]: 2.23998e-06 [handle_group_info]: 1.12999e-06 [symbol_engine_optimizer]: 0.00030105, [1] [Cycle 1]: 0.00027936, [6] [build]: 3.7e-06 [elim_shapecalc]: 1.059e-05 [elim_not_effective]: 1.394e-05 [opt_reshape]: 7.33e-06 [fold_const_symbol]: 9.92001e-06 [renormalize]: 1.69995e-07 [detach_backward]: 2.43e-06 [pipeline_parallel_scheduler]: 1.86998e-06 [auto_monad_reorder]: 0.00013528 [get_jit_bprop_graph]: 2.16998e-06 [rewriter_after_jit_bprop_graph]: 6.74001e-06 [opt_after_jit_grad]: 0.00075974 [validate]: 4.594e-05 Sums bootstrap : 0.011521s : 3.23% type_inference : 0.243223s : 68.15% event_method : 0.000011s : 0.00% auto_monad : 0.000054s : 0.02% graph_reusing : 0.000006s : 0.00% inline : 0.000002s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000016s : 0.00% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000134s : 0.04% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000707s : 0.20% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000023s : 0.01% optimize.rewriter_before_opt_a : 0.000050s : 0.01% optimize.opt_a.expand_dump_flag : 0.000004s : 0.00% optimize.opt_a.switch_simplify : 0.000032s : 0.01% optimize.opt_a.loop_unroll : 0.000018s : 0.01% optimize.opt_a.a_1 : 0.000615s : 0.17% optimize.opt_a.with_stream_mark : 0.000035s : 0.01% optimize.opt_a.recompute_prepare : 0.000014s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000011s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.00% optimize.opt_a.parameter_eliminate : 0.000004s : 0.00% optimize.opt_a.a_2 : 0.000151s : 0.04% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.00% optimize.opt_a.shard : 0.000004s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.00% optimize.opt_a.shard_inline : 0.000014s : 0.00% optimize.opt_a.merge_send_recv : 0.000018s : 0.00% optimize.opt_a.auto_parallel : 0.000015s : 0.00% optimize.opt_a.parallel : 0.000692s : 0.19% optimize.opt_a.flash_sp : 0.000014s : 0.00% optimize.opt_a.merge_comm : 0.000009s : 0.00% optimize.opt_a.allreduce_fusion : 0.000008s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000139s : 0.04% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000016s : 0.00% optimize.opt_a.virtual_dataset : 0.000012s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000012s : 0.00% optimize.opt_a.virtual_output : 0.000012s : 0.00% optimize.opt_a.merge_forward : 0.000008s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.00% optimize.opt_a.offload_activation : 0.000161s : 0.04% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000029s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000021s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000009s : 0.00% optimize.opt_a.meta_fg_expand : 0.000005s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.00% optimize.opt_a.receive_attached : 0.000004s : 0.00% optimize.opt_a.after_resolve : 0.000023s : 0.01% optimize.opt_a.a_after_grad : 0.000020s : 0.01% optimize.opt_a.renormalize : 0.002348s : 0.66% optimize.opt_a.add_forward_monad_depend : 0.000008s : 0.00% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000024s : 0.01% optimize.opt_a.cse : 0.000123s : 0.03% optimize.opt_a.a_3 : 0.000089s : 0.02% optimize.py_interpret_to_execute_after_opt_a : 0.000017s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000042s : 0.01% optimize.convert_after_rewriter : 0.000008s : 0.00% optimize.order_py_execute_after_rewriter : 0.000006s : 0.00% optimize.mutable_eliminate : 0.093703s : 26.26% optimize.opt_b.b_1 : 0.000139s : 0.04% optimize.opt_b.b_2 : 0.000009s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000011s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000004s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000004s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000179s : 0.05% optimize.optimize_parallel_all_gather_comm : 0.000026s : 0.01% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000043s : 0.01% optimize.loop_unroll : 0.000578s : 0.16% optimize.opt_after_cconv.c_1 : 0.000031s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000002s : 0.00% optimize.opt_after_cconv.cse : 0.000182s : 0.05% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000019s : 0.01% optimize.tuple_transform.d_1 : 0.000049s : 0.01% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000056s : 0.02% optimize.cse_after_recomputation.cse : 0.000102s : 0.03% optimize.environ_conv : 0.000008s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000005s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000004s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000014s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000004s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000005s : 0.00% optimize.overlap_grad_flash_sp : 0.000024s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000004s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000011s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000014s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000135s : 0.04% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.00% opt_after_jit_grad : 0.000760s : 0.21% validate : 0.000046s : 0.01% Time group info: ------[substitution.] 0.000111 21 1.90% : 0.000002s : 2: substitution.elim_not_effective 1.20% : 0.000001s : 2: substitution.fold_const_symbol 5.90% : 0.000007s : 4: substitution.graph_param_transform 78.02% : 0.000087s : 1: substitution.inline 3.91% : 0.000004s : 4: substitution.j_node_and_user_rematch 5.00% : 0.000006s : 4: substitution.remove_not_recompute_node 4.07% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.242934 2 99.28% : 0.241182s : 1: type_inference.infer 0.72% : 0.001752s : 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 0.74% : 0.000001s : 9: predicate.accumulaten_eliminater 1.14% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.58% : 0.000001s : 8: predicate.addn_check_dump 1.00% : 0.000002s : 9: predicate.addn_zero_filter 0.66% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.32% : 0.000004s : 17: predicate.arithmetic_simplify 0.82% : 0.000001s : 9: predicate.cast_eliminate 0.78% : 0.000001s : 8: predicate.check_bprop_eliminate 0.71% : 0.000001s : 8: predicate.compare_switch_simplify 0.23% : 0.000000s : 4: predicate.const_output_eliminate 0.80% : 0.000001s : 8: predicate.depend_value_elim 0.78% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.98% : 0.000002s : 9: predicate.dict_get_item_eliminator 0.79% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.22% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.28% : 0.000000s : 4: predicate.elim_not_effective 0.50% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.16% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.23% : 0.000002s : 13: predicate.environ_get_add_eliminate 1.01% : 0.000002s : 13: predicate.environ_get_depend_swap 1.79% : 0.000003s : 21: predicate.environ_get_eliminate 1.14% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.83% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.60% : 0.000003s : 10: predicate.float_depend_g_call 0.65% : 0.000001s : 8: predicate.float_environ_get_switch 1.02% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.24% : 0.000000s : 4: predicate.fold_const_symbol 0.85% : 0.000001s : 8: predicate.get_grad_eliminate 0.37% : 0.000001s : 4: predicate.graph_param_transform 0.67% : 0.000001s : 8: predicate.incorporate_call 0.57% : 0.000001s : 8: predicate.incorporate_call_switch 5.54% : 0.000009s : 43: predicate.inline 0.85% : 0.000001s : 8: predicate.inline_without_move 0.42% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.07% : 0.000002s : 8: predicate.less_batch_normalization 1.79% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.21% : 0.000004s : 26: predicate.load_eliminater 1.08% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.43% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.86% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.80% : 0.000001s : 8: predicate.merge_addn 0.69% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.67% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.65% : 0.000001s : 9: predicate.minmaximum_grad 2.58% : 0.000004s : 4: predicate.mutable_eliminate 0.63% : 0.000001s : 4: predicate.opt_reshape 0.49% : 0.000001s : 4: predicate.parallel_virtual_node 1.04% : 0.000002s : 10: predicate.partial_defer_inline 1.10% : 0.000002s : 13: predicate.partial_eliminate 0.76% : 0.000001s : 9: predicate.print_const_string_wrapper 0.73% : 0.000001s : 8: predicate.reduce_all_const_elim 1.22% : 0.000002s : 9: predicate.reduce_eliminate 2.08% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.76% : 0.000001s : 8: predicate.remove_not_recompute_node 1.27% : 0.000002s : 17: predicate.replace_applicator 0.81% : 0.000001s : 8: predicate.replace_old_param 0.73% : 0.000001s : 4: predicate.reset_defer_inline 0.79% : 0.000001s : 9: predicate.reshape_eliminate 0.92% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.51% : 0.000001s : 4: predicate.row_tensor_eliminate 1.12% : 0.000002s : 8: predicate.same_eliminate 0.56% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.86% : 0.000001s : 8: predicate.shard_identity_eliminate 0.80% : 0.000001s : 8: predicate.special_op_eliminate 0.81% : 0.000001s : 8: predicate.specialize_transform 1.53% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 1.13% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.49% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.84% : 0.000001s : 10: predicate.switch_defer_inline 1.50% : 0.000002s : 18: predicate.switch_layer_defer_inline 3.72% : 0.000006s : 35: predicate.switch_simplify 0.78% : 0.000001s : 9: predicate.tile_eliminate 0.83% : 0.000001s : 9: predicate.transpose_eliminate 1.65% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.72% : 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 1.34% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.58% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.52% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 1.87% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 3.15% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.48% : 0.000001s : 4: predicate.value_based_eliminate 0.77% : 0.000001s : 8: predicate.virtual_dataset_eliminate 1.00% : 0.000002s : 8: predicate.virtual_output_eliminate 0.33% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.61% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000927 4 2.79% : 0.000026s : 1: func_graph_cloner_run.FuncGraphClonerGraph 97.21% : 0.000901s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.325537 192 0.00% : 0.000004s : 1: ForceFp32Comm 32.37% : 0.429113s : 1: add_attr 32.36% : 0.428976s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.00% : 0.000060s : 1: add_recomputation 0.01% : 0.000159s : 1: assign_add_opt 0.00% : 0.000061s : 1: auto_monad 0.01% : 0.000139s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.89% : 0.011734s : 1: bootstrap 0.00% : 0.000047s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000017s : 1: control_data_broadcast_order 0.00% : 0.000011s : 1: convert_after_rewriter 0.01% : 0.000115s : 1: cse_after_recomputation 0.05% : 0.000714s : 1: dataset_repeat_opt 0.02% : 0.000234s : 1: detach_backward 0.00% : 0.000011s : 1: environ_conv 0.00% : 0.000019s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000010s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000005s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.01% : 0.000120s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000005s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.04% : 0.000587s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 7.07% : 0.093729s : 1: mutable_eliminate 0.00% : 0.000007s : 1: offloading_packed_experts 0.00% : 0.000015s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000036s : 1: opt.transform.mutable_eliminate 0.08% : 0.001008s : 78: opt.transform.opt_a 0.00% : 0.000029s : 1: opt.transform.opt_after_cconv 0.00% : 0.000024s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000113s : 28: opt.transform.opt_b 0.00% : 0.000054s : 2: opt.transform.opt_trans_graph 0.00% : 0.000038s : 4: opt.transform.symbol_engine_opt 0.48% : 0.006348s : 1: opt_a 0.05% : 0.000619s : 1: opt_after_cconv 0.06% : 0.000769s : 1: opt_after_jit_grad 0.03% : 0.000400s : 1: opt_b 7.81% : 0.103499s : 1: optimize 0.00% : 0.000030s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000010s : 1: order_py_execute_after_rewriter 0.00% : 0.000027s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000008s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.01% : 0.000175s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000139s : 1: pre_auto_parallel 0.00% : 0.000027s : 1: py_interpret_to_execute 0.00% : 0.000020s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000022s : 1: remove_dup_value 0.16% : 0.002055s : 1: renormalize.infer 0.02% : 0.000284s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.00% : 0.000048s : 1: rewriter_after_opt_a 0.00% : 0.000054s : 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% : 0.000008s : 1: swap_dp_allreduce_reducescatter 0.02% : 0.000304s : 1: symbol_engine_optimizer 0.01% : 0.000085s : 1: tuple_transform 18.35% : 0.243248s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[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_fp16[dim_type0-True-batch_size_list2-shape0-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:38:43.225.54 [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 = 0.0707831, [21] [bootstrap]: 0.00066472 [type_inference]: 0.0596149 [event_method]: 1.828e-05 [auto_monad]: 5.888e-05 [graph_reusing]: 5.20999e-06 [inline]: 2.93e-06 [add_attr]: 0.00475418, [1] [add_attr_with_inline]: 0.00473943, [1] [Cycle 1]: 6.406e-05, [2] [tag_attr]: 1.635e-05 [meta_addattr_fg_expand]: 3.22997e-06 [parallel-infer-symbol]: 3.93999e-06 [pre_auto_parallel]: 3.411e-05 [insert-virtual-dataset]: 2.53e-06 [parallel-infer-symbol-second]: 9.00007e-07 [dataset_repeat_opt]: 2.16e-06 [pipeline_split]: 2.06998e-06 [optimize]: 0.00475699, [53] [py_interpret_to_execute]: 2.064e-05 [rewriter_before_opt_a]: 4.662e-05 [opt_a]: 0.00241983, [2] [Cycle 1]: 0.0017301, [45] [expand_dump_flag]: 2.61999e-06 [switch_simplify]: 2.328e-05 [loop_unroll]: 1.122e-05 [a_1]: 0.00029335 [with_stream_mark]: 2.062e-05 [recompute_prepare]: 8.05e-06 [updatestate_depend_eliminate]: 4.29002e-06 [updatestate_assign_eliminate]: 3.72002e-06 [updatestate_loads_eliminate]: 3.55998e-06 [parameter_eliminate]: 2.01e-06 [a_2]: 7.937e-05 [accelerated_algorithm]: 6.89001e-06 [shard]: 2.51e-06 [meta_shard_fg_expand]: 1.74e-06 [shard_inline]: 6.58e-06 [merge_send_recv]: 8.91002e-06 [auto_parallel]: 7.08e-06 [parallel]: 3.096e-05 [flash_sp]: 1.001e-05 [merge_comm]: 3.66001e-06 [allreduce_fusion]: 3.51001e-06 [matmul_add_comm_reduction]: 1.095e-05 [allreduce_slice_to_reducescatter]: 8.29983e-07 [virtual_shard_identity]: 7.68001e-06 [virtual_dataset]: 6.43e-06 [get_grad_eliminate_]: 5.81e-06 [virtual_output]: 6.05002e-06 [merge_forward]: 4.4e-06 [cell_reuse_recompute_pass]: 1.55001e-06 [offload_activation]: 1.142e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.366e-05 [merge_recompute_call_nodes]: 1.55001e-06 [before_grad]: 1.019e-05 [set_forward_comm_id_for_comm_node_pass]: 4.08001e-06 [meta_fg_expand]: 2.71999e-06 [flash_sp_send_recv_attached]: 2.66e-06 [receive_attached]: 2.73e-06 [after_resolve]: 1.154e-05 [a_after_grad]: 9.34e-06 [renormalize]: 0.00070175 [add_forward_monad_depend]: 6.63998e-06 [auto_monad_grad]: 2.56e-06 [auto_monad_eliminator]: 1.694e-05 [cse]: 3.888e-05 [a_3]: 5.07e-05 [Cycle 2]: 0.00067802, [45] [expand_dump_flag]: 1.71e-06 [switch_simplify]: 7.59002e-06 [loop_unroll]: 6.37001e-06 [a_1]: 0.00012261 [with_stream_mark]: 1.433e-05 [recompute_prepare]: 6.88e-06 [updatestate_depend_eliminate]: 3.74002e-06 [updatestate_assign_eliminate]: 2.79999e-06 [updatestate_loads_eliminate]: 3.04999e-06 [parameter_eliminate]: 1.55999e-06 [a_2]: 7.016e-05 [accelerated_algorithm]: 6.53998e-06 [shard]: 1.79e-06 [meta_shard_fg_expand]: 1.82001e-06 [shard_inline]: 6.33998e-06 [merge_send_recv]: 6.26e-06 [auto_parallel]: 6.21998e-06 [parallel]: 5.27999e-06 [flash_sp]: 4.3e-06 [merge_comm]: 3.56999e-06 [allreduce_fusion]: 3.16001e-06 [matmul_add_comm_reduction]: 1.381e-05 [allreduce_slice_to_reducescatter]: 5.09986e-07 [virtual_shard_identity]: 7.06001e-06 [virtual_dataset]: 5.97999e-06 [get_grad_eliminate_]: 6.04999e-06 [virtual_output]: 5.64998e-06 [merge_forward]: 4.07998e-06 [cell_reuse_recompute_pass]: 2.16e-06 [offload_activation]: 1.09e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.384e-05 [merge_recompute_call_nodes]: 1.00999e-06 [before_grad]: 9.94999e-06 [set_forward_comm_id_for_comm_node_pass]: 3.76999e-06 [meta_fg_expand]: 2.27999e-06 [flash_sp_send_recv_attached]: 1.30001e-06 [receive_attached]: 1.50001e-06 [after_resolve]: 1.116e-05 [a_after_grad]: 9.55001e-06 [renormalize]: 7.00238e-08 [add_forward_monad_depend]: 1.30999e-06 [auto_monad_grad]: 1.13001e-06 [auto_monad_eliminator]: 8.64998e-06 [cse]: 2.262e-05 [a_3]: 3.574e-05 [py_interpret_to_execute_after_opt_a]: 1.364e-05 [slice_cell_reuse_recomputed_activation]: 2.68998e-06 [rewriter_after_opt_a]: 4.106e-05 [convert_after_rewriter]: 8.17e-06 [order_py_execute_after_rewriter]: 5.76e-06 [mutable_eliminate]: 0.00066056 [opt_b]: 0.00023226, [1] [Cycle 1]: 0.00022336, [7] [b_1]: 0.00012543 [b_2]: 8.3e-06 [updatestate_depend_eliminate]: 8.65001e-06 [updatestate_assign_eliminate]: 3.76999e-06 [updatestate_loads_eliminate]: 2.41e-06 [renormalize]: 5.49975e-07 [cse]: 3.683e-05 [optimize_parallel_all_gather_comm]: 2.028e-05 [overlap_param_gather]: 2.04999e-06 [cconv]: 3.542e-05 [loop_unroll]: 0.00049312 [opt_after_cconv]: 0.00012534, [1] [Cycle 1]: 0.00011785, [7] [c_1]: 3e-05 [parameter_eliminate]: 5.64e-06 [updatestate_depend_eliminate]: 6.46999e-06 [updatestate_assign_eliminate]: 3.08e-06 [updatestate_loads_eliminate]: 2.70002e-06 [cse]: 2.944e-05 [renormalize]: 4.89992e-07 [remove_dup_value]: 1.975e-05 [tuple_transform]: 7.988e-05, [1] [Cycle 1]: 7.52e-05, [4] [d_1]: 4.843e-05 [none_parameter_eliminate]: 2.04999e-06 [renormalize]: 1.50001e-07 [switch_simplify]: 6.68e-06 [partial_unused_args_eliminate]: 2.27999e-06 [add_recomputation]: 5.613e-05 [cse_after_recomputation]: 3.008e-05, [1] [Cycle 1]: 2.5e-05, [1] [cse]: 1.881e-05 [environ_conv]: 7.13e-06 [swap_dp_allreduce_reducescatter]: 5.50001e-06 [bias_add_comm_swap]: 3.03e-06 [label_micro_interleaved_index]: 5.84999e-06 [label_fine_grained_interleaved_index]: 2.71e-06 [merge_cast_opt]: 1.46002e-06 [slice_recompute_activation]: 2.31e-06 [micro_interleaved_order_control]: 2.83998e-06 [assign_add_opt]: 1.37e-06 [ForceFp32Comm]: 8.79983e-07 [remove_cast_before_assign_add]: 1.11002e-06 [full_micro_interleaved_order_control]: 2.21e-06 [reorder_send_recv_between_fp_bp]: 2.88e-06 [comm_op_add_attrs]: 1.27e-06 [add_comm_op_reuse_tag]: 1.07998e-06 [interleave_split_concat_branches]: 1.54998e-06 [interleave_parallel_branches]: 1.10001e-06 [overlap_opt_shard_in_pipeline]: 1.37999e-06 [overlap_opt_shard_grad_in_pipeline]: 1.90001e-06 [control_data_broadcast_order]: 1.582e-05 [grouped_pairwise_exchange_alltoall]: 1.74e-06 [offloading_packed_experts]: 4.49998e-06 [overlap_recompute_and_grad_model_parallel]: 6.07001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.47999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.39998e-06 [overlap_recompute_comm]: 2.53e-06 [overlap_grad_ring_attention]: 4.14002e-06 [overlap_grad_flash_sp]: 2.283e-05 [begin_end_overlap_inline]: 5.90022e-07 [split_matmul_comm_elemetwise]: 2.62001e-06 [split_layernorm_comm]: 1.91e-06 [handle_group_info]: 1.24e-06 [symbol_engine_optimizer]: 8.018e-05, [1] [Cycle 1]: 7.606e-05, [6] [build]: 3.39001e-06 [elim_shapecalc]: 1.063e-05 [elim_not_effective]: 1.332e-05 [opt_reshape]: 6.98e-06 [fold_const_symbol]: 1.012e-05 [renormalize]: 1.60013e-07 [detach_backward]: 2.27999e-06 [pipeline_parallel_scheduler]: 1.55999e-06 [auto_monad_reorder]: 2.255e-05 [get_jit_bprop_graph]: 2.00002e-06 [rewriter_after_jit_bprop_graph]: 5.81998e-06 [opt_after_jit_grad]: 0.00056362 [validate]: 5.242e-05 Sums bootstrap : 0.000665s : 1.02% type_inference : 0.059615s : 91.80% event_method : 0.000018s : 0.03% auto_monad : 0.000059s : 0.09% graph_reusing : 0.000005s : 0.01% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000016s : 0.03% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.00% parallel-infer-symbol : 0.000004s : 0.01% pre_auto_parallel : 0.000034s : 0.05% 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 : 0.03% optimize.rewriter_before_opt_a : 0.000047s : 0.07% optimize.opt_a.expand_dump_flag : 0.000004s : 0.01% optimize.opt_a.switch_simplify : 0.000031s : 0.05% optimize.opt_a.loop_unroll : 0.000018s : 0.03% optimize.opt_a.a_1 : 0.000416s : 0.64% optimize.opt_a.with_stream_mark : 0.000035s : 0.05% optimize.opt_a.recompute_prepare : 0.000015s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.01% optimize.opt_a.a_2 : 0.000150s : 0.23% optimize.opt_a.accelerated_algorithm : 0.000013s : 0.02% optimize.opt_a.shard : 0.000004s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.01% optimize.opt_a.shard_inline : 0.000013s : 0.02% optimize.opt_a.merge_send_recv : 0.000015s : 0.02% optimize.opt_a.auto_parallel : 0.000013s : 0.02% optimize.opt_a.parallel : 0.000036s : 0.06% optimize.opt_a.flash_sp : 0.000014s : 0.02% optimize.opt_a.merge_comm : 0.000007s : 0.01% optimize.opt_a.allreduce_fusion : 0.000007s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000025s : 0.04% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000015s : 0.02% optimize.opt_a.virtual_dataset : 0.000012s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000012s : 0.02% optimize.opt_a.virtual_output : 0.000012s : 0.02% optimize.opt_a.merge_forward : 0.000008s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.01% optimize.opt_a.offload_activation : 0.000022s : 0.03% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000027s : 0.04% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000020s : 0.03% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.01% optimize.opt_a.meta_fg_expand : 0.000005s : 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 : 0.000023s : 0.03% optimize.opt_a.a_after_grad : 0.000019s : 0.03% optimize.opt_a.renormalize : 0.000702s : 1.08% optimize.opt_a.add_forward_monad_depend : 0.000008s : 0.01% optimize.opt_a.auto_monad_grad : 0.000004s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000026s : 0.04% optimize.opt_a.cse : 0.000062s : 0.09% optimize.opt_a.a_3 : 0.000086s : 0.13% optimize.py_interpret_to_execute_after_opt_a : 0.000014s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000041s : 0.06% optimize.convert_after_rewriter : 0.000008s : 0.01% optimize.order_py_execute_after_rewriter : 0.000006s : 0.01% optimize.mutable_eliminate : 0.000661s : 1.02% optimize.opt_b.b_1 : 0.000125s : 0.19% optimize.opt_b.b_2 : 0.000008s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000009s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000004s : 0.01% optimize.opt_b.updatestate_loads_eliminate : 0.000002s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000037s : 0.06% optimize.optimize_parallel_all_gather_comm : 0.000020s : 0.03% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000035s : 0.05% optimize.loop_unroll : 0.000493s : 0.76% optimize.opt_after_cconv.c_1 : 0.000030s : 0.05% 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 : 0.000029s : 0.05% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000020s : 0.03% optimize.tuple_transform.d_1 : 0.000048s : 0.07% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000056s : 0.09% optimize.cse_after_recomputation.cse : 0.000019s : 0.03% 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 : 0.000006s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000002s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000016s : 0.02% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000004s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000006s : 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 : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000004s : 0.01% optimize.overlap_grad_flash_sp : 0.000023s : 0.04% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000003s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000011s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000013s : 0.02% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 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 : 0.000023s : 0.03% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.01% opt_after_jit_grad : 0.000564s : 0.87% validate : 0.000052s : 0.08% Time group info: ------[substitution.] 0.000111 21 1.75% : 0.000002s : 2: substitution.elim_not_effective 1.37% : 0.000002s : 2: substitution.fold_const_symbol 6.00% : 0.000007s : 4: substitution.graph_param_transform 78.52% : 0.000087s : 1: substitution.inline 3.43% : 0.000004s : 4: substitution.j_node_and_user_rematch 4.97% : 0.000006s : 4: substitution.remove_not_recompute_node 3.95% : 0.000004s : 4: substitution.replace_old_param ------[type_inference.] 0.059552 2 99.26% : 0.059111s : 1: type_inference.infer 0.74% : 0.000441s : 1: type_inference.specialize ------[replace.] 0.000017 1 100.00% : 0.000017s : 1: replace.inline ------[match.] 0.000086 1 100.00% : 0.000086s : 1: match.inline ------[predicate.] 0.000151 967 0.73% : 0.000001s : 9: predicate.accumulaten_eliminater 1.45% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.62% : 0.000001s : 8: predicate.addn_check_dump 0.78% : 0.000001s : 9: predicate.addn_zero_filter 0.67% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.23% : 0.000003s : 17: predicate.arithmetic_simplify 0.86% : 0.000001s : 9: predicate.cast_eliminate 0.82% : 0.000001s : 8: predicate.check_bprop_eliminate 0.61% : 0.000001s : 8: predicate.compare_switch_simplify 0.25% : 0.000000s : 4: predicate.const_output_eliminate 0.69% : 0.000001s : 8: predicate.depend_value_elim 0.76% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.85% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.81% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.21% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.29% : 0.000000s : 4: predicate.elim_not_effective 0.57% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.11% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.05% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.99% : 0.000001s : 13: predicate.environ_get_depend_swap 1.80% : 0.000003s : 21: predicate.environ_get_eliminate 1.06% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.78% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.62% : 0.000002s : 10: predicate.float_depend_g_call 0.73% : 0.000001s : 8: predicate.float_environ_get_switch 0.97% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.25% : 0.000000s : 4: predicate.fold_const_symbol 0.86% : 0.000001s : 8: predicate.get_grad_eliminate 0.39% : 0.000001s : 4: predicate.graph_param_transform 0.69% : 0.000001s : 8: predicate.incorporate_call 0.61% : 0.000001s : 8: predicate.incorporate_call_switch 5.97% : 0.000009s : 43: predicate.inline 1.13% : 0.000002s : 8: predicate.inline_without_move 0.44% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.19% : 0.000002s : 8: predicate.less_batch_normalization 1.90% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.07% : 0.000003s : 26: predicate.load_eliminater 1.51% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.47% : 0.000002s : 13: predicate.loop_unroll_before_grad 2.02% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.77% : 0.000001s : 8: predicate.merge_addn 0.73% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.81% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.67% : 0.000001s : 9: predicate.minmaximum_grad 1.96% : 0.000003s : 4: predicate.mutable_eliminate 0.50% : 0.000001s : 4: predicate.opt_reshape 0.60% : 0.000001s : 4: predicate.parallel_virtual_node 1.35% : 0.000002s : 10: predicate.partial_defer_inline 1.16% : 0.000002s : 13: predicate.partial_eliminate 0.79% : 0.000001s : 9: predicate.print_const_string_wrapper 0.75% : 0.000001s : 8: predicate.reduce_all_const_elim 1.14% : 0.000002s : 9: predicate.reduce_eliminate 2.01% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.75% : 0.000001s : 8: predicate.remove_not_recompute_node 1.39% : 0.000002s : 17: predicate.replace_applicator 0.90% : 0.000001s : 8: predicate.replace_old_param 0.54% : 0.000001s : 4: predicate.reset_defer_inline 0.72% : 0.000001s : 9: predicate.reshape_eliminate 0.93% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.51% : 0.000001s : 4: predicate.row_tensor_eliminate 1.08% : 0.000002s : 8: predicate.same_eliminate 0.71% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.00% : 0.000002s : 8: predicate.shard_identity_eliminate 0.84% : 0.000001s : 8: predicate.special_op_eliminate 0.88% : 0.000001s : 8: predicate.specialize_transform 1.09% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 1.04% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.45% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.86% : 0.000001s : 10: predicate.switch_defer_inline 1.78% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.66% : 0.000006s : 35: predicate.switch_simplify 0.71% : 0.000001s : 9: predicate.tile_eliminate 0.73% : 0.000001s : 9: predicate.transpose_eliminate 1.63% : 0.000002s : 17: predicate.tuple_list_convert_item_index_to_positive 1.64% : 0.000002s : 17: predicate.tuple_list_get_item_const_eliminator 1.43% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.43% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.35% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.32% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.59% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 2.00% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.85% : 0.000004s : 34: predicate.updatestate_useless_node_eliminater 0.50% : 0.000001s : 4: predicate.value_based_eliminate 0.89% : 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 0.60% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000246 4 8.87% : 0.000022s : 1: func_graph_cloner_run.FuncGraphClonerGraph 91.13% : 0.000224s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.081925 192 0.00% : 0.000004s : 1: ForceFp32Comm 5.81% : 0.004761s : 1: add_attr 5.79% : 0.004744s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.07% : 0.000061s : 1: add_recomputation 0.01% : 0.000005s : 1: assign_add_opt 0.08% : 0.000064s : 1: auto_monad 0.03% : 0.000027s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.01% : 0.000006s : 1: bias_add_comm_swap 0.88% : 0.000719s : 1: bootstrap 0.05% : 0.000040s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.02% : 0.000019s : 1: control_data_broadcast_order 0.01% : 0.000012s : 1: convert_after_rewriter 0.04% : 0.000033s : 1: cse_after_recomputation 0.01% : 0.000005s : 1: dataset_repeat_opt 0.02% : 0.000013s : 1: detach_backward 0.01% : 0.000010s : 1: environ_conv 0.03% : 0.000026s : 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% : 0.000004s : 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% : 0.000008s : 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% : 0.000009s : 1: label_micro_interleaved_index 0.62% : 0.000506s : 1: loop_unroll 0.01% : 0.000004s : 1: merge_cast_opt 0.01% : 0.000005s : 1: micro_interleaved_order_control 0.82% : 0.000675s : 1: mutable_eliminate 0.01% : 0.000008s : 1: offloading_packed_experts 0.02% : 0.000016s : 1: opt.transform.loop_unroll_optimizer 0.02% : 0.000020s : 1: opt.transform.mutable_eliminate 0.98% : 0.000799s : 78: opt.transform.opt_a 0.03% : 0.000029s : 1: opt.transform.opt_after_cconv 0.04% : 0.000031s : 1: opt.transform.opt_after_jit_grad 0.13% : 0.000105s : 28: opt.transform.opt_b 0.06% : 0.000053s : 2: opt.transform.opt_trans_graph 0.05% : 0.000037s : 4: opt.transform.symbol_engine_opt 2.96% : 0.002423s : 1: opt_a 0.16% : 0.000129s : 1: opt_after_cconv 0.71% : 0.000580s : 1: opt_after_jit_grad 0.29% : 0.000236s : 1: opt_b 5.81% : 0.004763s : 1: optimize 0.03% : 0.000024s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000009s : 1: order_py_execute_after_rewriter 0.03% : 0.000026s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000007s : 1: overlap_grad_ring_attention 0.01% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.01% : 0.000006s : 1: overlap_param_gather 0.01% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000009s : 1: overlap_recompute_and_grad_model_parallel 0.01% : 0.000006s : 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% : 0.000005s : 1: pipeline_split 0.05% : 0.000039s : 1: pre_auto_parallel 0.03% : 0.000025s : 1: py_interpret_to_execute 0.02% : 0.000018s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.03% : 0.000023s : 1: remove_dup_value 0.52% : 0.000423s : 1: renormalize.infer 0.33% : 0.000270s : 1: renormalize.specialize 0.01% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.06% : 0.000046s : 1: rewriter_after_opt_a 0.06% : 0.000050s : 1: rewriter_before_opt_a 0.01% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.01% : 0.000005s : 1: slice_recompute_activation 0.01% : 0.000005s : 1: split_layernorm_comm 0.01% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000008s : 1: swap_dp_allreduce_reducescatter 0.10% : 0.000083s : 1: symbol_engine_optimizer 0.10% : 0.000083s : 1: tuple_transform 72.80% : 0.059639s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[dim_type0-True-batch_size_list2-shape1-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_fp16[dim_type0-True-batch_size_list2-shape1-ms_dtype0],max_mem:8.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:38:57.122.129 [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 = 0.207941, [21] [bootstrap]: 0.00067477 [type_inference]: 0.0040504 [event_method]: 1.386e-05 [auto_monad]: 5.723e-05 [graph_reusing]: 5.79999e-06 [inline]: 2.69001e-06 [add_attr]: 0.196995, [1] [add_attr_with_inline]: 0.196977, [1] [Cycle 1]: 7.33e-05, [2] [tag_attr]: 2.059e-05 [meta_addattr_fg_expand]: 3.25e-06 [parallel-infer-symbol]: 4.33999e-06 [pre_auto_parallel]: 4.065e-05 [insert-virtual-dataset]: 2.67001e-06 [parallel-infer-symbol-second]: 9.10019e-07 [dataset_repeat_opt]: 3.03998e-06 [pipeline_split]: 1.69e-06 [optimize]: 0.00516186, [53] [py_interpret_to_execute]: 2.91e-05 [rewriter_before_opt_a]: 5.374e-05 [opt_a]: 0.00251662, [2] [Cycle 1]: 0.00180541, [45] [expand_dump_flag]: 3.06001e-06 [switch_simplify]: 2.315e-05 [loop_unroll]: 1.105e-05 [a_1]: 0.00031481 [with_stream_mark]: 2.294e-05 [recompute_prepare]: 9.99999e-06 [updatestate_depend_eliminate]: 4.84e-06 [updatestate_assign_eliminate]: 4.08001e-06 [updatestate_loads_eliminate]: 3.66999e-06 [parameter_eliminate]: 2.71999e-06 [a_2]: 8.063e-05 [accelerated_algorithm]: 7.25e-06 [shard]: 2.91999e-06 [meta_shard_fg_expand]: 1.90001e-06 [shard_inline]: 6.47001e-06 [merge_send_recv]: 9.74999e-06 [auto_parallel]: 7.46001e-06 [parallel]: 3.029e-05 [flash_sp]: 1.021e-05 [merge_comm]: 3.98001e-06 [allreduce_fusion]: 3.87998e-06 [matmul_add_comm_reduction]: 1.1e-05 [allreduce_slice_to_reducescatter]: 8.70001e-07 [virtual_shard_identity]: 8.37e-06 [virtual_dataset]: 6.56999e-06 [get_grad_eliminate_]: 6.13002e-06 [virtual_output]: 6.12999e-06 [merge_forward]: 4.33001e-06 [cell_reuse_recompute_pass]: 1.44e-06 [offload_activation]: 1.157e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.429e-05 [merge_recompute_call_nodes]: 1.77999e-06 [before_grad]: 1.06e-05 [set_forward_comm_id_for_comm_node_pass]: 3.58e-06 [meta_fg_expand]: 3.16001e-06 [flash_sp_send_recv_attached]: 2.54999e-06 [receive_attached]: 2.46e-06 [after_resolve]: 1.302e-05 [a_after_grad]: 1.084e-05 [renormalize]: 0.00073688 [add_forward_monad_depend]: 6.66e-06 [auto_monad_grad]: 2.84001e-06 [auto_monad_eliminator]: 1.71e-05 [cse]: 5.697e-05 [a_3]: 5.043e-05 [Cycle 2]: 0.00069887, [45] [expand_dump_flag]: 2.59999e-06 [switch_simplify]: 7.96001e-06 [loop_unroll]: 8.05e-06 [a_1]: 0.00012554 [with_stream_mark]: 1.57e-05 [recompute_prepare]: 6.94999e-06 [updatestate_depend_eliminate]: 4.67e-06 [updatestate_assign_eliminate]: 2.66999e-06 [updatestate_loads_eliminate]: 3.33e-06 [parameter_eliminate]: 1.22e-06 [a_2]: 7.187e-05 [accelerated_algorithm]: 7.54002e-06 [shard]: 1.88002e-06 [meta_shard_fg_expand]: 2.53998e-06 [shard_inline]: 6.58e-06 [merge_send_recv]: 8.13001e-06 [auto_parallel]: 7.38e-06 [parallel]: 7.21999e-06 [flash_sp]: 4.4e-06 [merge_comm]: 3.88999e-06 [allreduce_fusion]: 3.43999e-06 [matmul_add_comm_reduction]: 8.47998e-06 [allreduce_slice_to_reducescatter]: 6.00005e-07 [virtual_shard_identity]: 7.80998e-06 [virtual_dataset]: 6.27001e-06 [get_grad_eliminate_]: 6.03998e-06 [virtual_output]: 5.84999e-06 [merge_forward]: 4.62998e-06 [cell_reuse_recompute_pass]: 2.48998e-06 [offload_activation]: 1.308e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.686e-05 [merge_recompute_call_nodes]: 9.10019e-07 [before_grad]: 1.041e-05 [set_forward_comm_id_for_comm_node_pass]: 3.84002e-06 [meta_fg_expand]: 2.54999e-06 [flash_sp_send_recv_attached]: 1.25001e-06 [receive_attached]: 2.24001e-06 [after_resolve]: 1.088e-05 [a_after_grad]: 9.30001e-06 [renormalize]: 6.00121e-08 [add_forward_monad_depend]: 1.64e-06 [auto_monad_grad]: 1.42999e-06 [auto_monad_eliminator]: 8.25e-06 [cse]: 2.424e-05 [a_3]: 3.639e-05 [py_interpret_to_execute_after_opt_a]: 1.568e-05 [slice_cell_reuse_recomputed_activation]: 1.94999e-06 [rewriter_after_opt_a]: 4.585e-05 [convert_after_rewriter]: 7.38e-06 [order_py_execute_after_rewriter]: 6.21e-06 [mutable_eliminate]: 0.00078578 [opt_b]: 0.00024306, [1] [Cycle 1]: 0.00023441, [7] [b_1]: 0.00013189 [b_2]: 8.96998e-06 [updatestate_depend_eliminate]: 1.028e-05 [updatestate_assign_eliminate]: 3.15998e-06 [updatestate_loads_eliminate]: 2.81e-06 [renormalize]: 1.04e-06 [cse]: 3.91e-05 [optimize_parallel_all_gather_comm]: 2.078e-05 [overlap_param_gather]: 2.57001e-06 [cconv]: 3.967e-05 [loop_unroll]: 0.00054499 [opt_after_cconv]: 0.00013522, [1] [Cycle 1]: 0.00012722, [7] [c_1]: 3.295e-05 [parameter_eliminate]: 6.22001e-06 [updatestate_depend_eliminate]: 8.31002e-06 [updatestate_assign_eliminate]: 2.91e-06 [updatestate_loads_eliminate]: 2.56e-06 [cse]: 3.028e-05 [renormalize]: 7.7e-07 [remove_dup_value]: 5.425e-05 [tuple_transform]: 9.356e-05, [1] [Cycle 1]: 8.39e-05, [4] [d_1]: 5.488e-05 [none_parameter_eliminate]: 1.98002e-06 [renormalize]: 2.30008e-07 [switch_simplify]: 7.25998e-06 [partial_unused_args_eliminate]: 2.62001e-06 [add_recomputation]: 5.918e-05 [cse_after_recomputation]: 3.083e-05, [1] [Cycle 1]: 2.593e-05, [1] [cse]: 1.994e-05 [environ_conv]: 7.23e-06 [swap_dp_allreduce_reducescatter]: 5.97001e-06 [bias_add_comm_swap]: 3.81999e-06 [label_micro_interleaved_index]: 6.76999e-06 [label_fine_grained_interleaved_index]: 2.86999e-06 [merge_cast_opt]: 1.74e-06 [slice_recompute_activation]: 2.88e-06 [micro_interleaved_order_control]: 2.39999e-06 [assign_add_opt]: 1.45999e-06 [ForceFp32Comm]: 1.38002e-06 [remove_cast_before_assign_add]: 1.05001e-06 [full_micro_interleaved_order_control]: 2.51998e-06 [reorder_send_recv_between_fp_bp]: 2.98e-06 [comm_op_add_attrs]: 1.42e-06 [add_comm_op_reuse_tag]: 1.31002e-06 [interleave_split_concat_branches]: 1.32999e-06 [interleave_parallel_branches]: 1.09e-06 [overlap_opt_shard_in_pipeline]: 1.49e-06 [overlap_opt_shard_grad_in_pipeline]: 2.21e-06 [control_data_broadcast_order]: 1.672e-05 [grouped_pairwise_exchange_alltoall]: 1.79998e-06 [offloading_packed_experts]: 4.80999e-06 [overlap_recompute_and_grad_model_parallel]: 5.43002e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.22e-06 [overlap_recompute_allgather_and_fa_grad]: 1.84e-06 [overlap_recompute_comm]: 2.24001e-06 [overlap_grad_ring_attention]: 4.84e-06 [overlap_grad_flash_sp]: 2.277e-05 [begin_end_overlap_inline]: 5.59987e-07 [split_matmul_comm_elemetwise]: 2.73e-06 [split_layernorm_comm]: 2.63998e-06 [handle_group_info]: 1.12999e-06 [symbol_engine_optimizer]: 8.998e-05, [1] [Cycle 1]: 8.411e-05, [6] [build]: 3.74002e-06 [elim_shapecalc]: 1.228e-05 [elim_not_effective]: 1.488e-05 [opt_reshape]: 8.03999e-06 [fold_const_symbol]: 1.05e-05 [renormalize]: 1.90019e-07 [detach_backward]: 2.54999e-06 [pipeline_parallel_scheduler]: 1.86e-06 [auto_monad_reorder]: 2.177e-05 [get_jit_bprop_graph]: 2.32001e-06 [rewriter_after_jit_bprop_graph]: 7.36001e-06 [opt_after_jit_grad]: 0.00063472 [validate]: 5.473e-05 Sums bootstrap : 0.000675s : 6.85% type_inference : 0.004050s : 41.09% event_method : 0.000014s : 0.14% auto_monad : 0.000057s : 0.58% graph_reusing : 0.000006s : 0.06% inline : 0.000003s : 0.03% add_attr.add_attr_with_inline.tag_attr : 0.000021s : 0.21% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.03% parallel-infer-symbol : 0.000004s : 0.04% pre_auto_parallel : 0.000041s : 0.41% insert-virtual-dataset : 0.000003s : 0.03% parallel-infer-symbol-second : 0.000001s : 0.01% dataset_repeat_opt : 0.000003s : 0.03% pipeline_split : 0.000002s : 0.02% optimize.py_interpret_to_execute : 0.000029s : 0.30% optimize.rewriter_before_opt_a : 0.000054s : 0.55% optimize.opt_a.expand_dump_flag : 0.000006s : 0.06% optimize.opt_a.switch_simplify : 0.000031s : 0.32% optimize.opt_a.loop_unroll : 0.000019s : 0.19% optimize.opt_a.a_1 : 0.000440s : 4.47% optimize.opt_a.with_stream_mark : 0.000039s : 0.39% optimize.opt_a.recompute_prepare : 0.000017s : 0.17% optimize.opt_a.updatestate_depend_eliminate : 0.000010s : 0.10% 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 : 0.000152s : 1.55% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.15% optimize.opt_a.shard : 0.000005s : 0.05% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.05% optimize.opt_a.shard_inline : 0.000013s : 0.13% optimize.opt_a.merge_send_recv : 0.000018s : 0.18% optimize.opt_a.auto_parallel : 0.000015s : 0.15% optimize.opt_a.parallel : 0.000038s : 0.38% optimize.opt_a.flash_sp : 0.000015s : 0.15% optimize.opt_a.merge_comm : 0.000008s : 0.08% optimize.opt_a.allreduce_fusion : 0.000007s : 0.07% optimize.opt_a.matmul_add_comm_reduction : 0.000019s : 0.20% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000016s : 0.16% optimize.opt_a.virtual_dataset : 0.000013s : 0.13% optimize.opt_a.get_grad_eliminate_ : 0.000012s : 0.12% optimize.opt_a.virtual_output : 0.000012s : 0.12% optimize.opt_a.merge_forward : 0.000009s : 0.09% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.04% optimize.opt_a.offload_activation : 0.000025s : 0.25% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000031s : 0.32% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.03% optimize.opt_a.before_grad : 0.000021s : 0.21% 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 : 0.000005s : 0.05% optimize.opt_a.after_resolve : 0.000024s : 0.24% optimize.opt_a.a_after_grad : 0.000020s : 0.20% optimize.opt_a.renormalize : 0.000737s : 7.48% optimize.opt_a.add_forward_monad_depend : 0.000008s : 0.08% optimize.opt_a.auto_monad_grad : 0.000004s : 0.04% optimize.opt_a.auto_monad_eliminator : 0.000025s : 0.26% optimize.opt_a.cse : 0.000081s : 0.82% optimize.opt_a.a_3 : 0.000087s : 0.88% optimize.py_interpret_to_execute_after_opt_a : 0.000016s : 0.16% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.02% optimize.rewriter_after_opt_a : 0.000046s : 0.47% optimize.convert_after_rewriter : 0.000007s : 0.07% optimize.order_py_execute_after_rewriter : 0.000006s : 0.06% optimize.mutable_eliminate : 0.000786s : 7.97% optimize.opt_b.b_1 : 0.000132s : 1.34% optimize.opt_b.b_2 : 0.000009s : 0.09% optimize.opt_b.updatestate_depend_eliminate : 0.000010s : 0.10% 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 : 0.000039s : 0.40% optimize.optimize_parallel_all_gather_comm : 0.000021s : 0.21% optimize.overlap_param_gather : 0.000003s : 0.03% optimize.cconv : 0.000040s : 0.40% optimize.loop_unroll : 0.000545s : 5.53% optimize.opt_after_cconv.c_1 : 0.000033s : 0.33% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.06% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000008s : 0.08% 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 : 0.000030s : 0.31% optimize.opt_after_cconv.renormalize : 0.000001s : 0.01% optimize.remove_dup_value : 0.000054s : 0.55% optimize.tuple_transform.d_1 : 0.000055s : 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 : 0.07% optimize.partial_unused_args_eliminate : 0.000003s : 0.03% optimize.add_recomputation : 0.000059s : 0.60% optimize.cse_after_recomputation.cse : 0.000020s : 0.20% optimize.environ_conv : 0.000007s : 0.07% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.06% optimize.bias_add_comm_swap : 0.000004s : 0.04% optimize.label_micro_interleaved_index : 0.000007s : 0.07% 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 : 0.000001s : 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 : 0.02% optimize.control_data_broadcast_order : 0.000017s : 0.17% 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 : 0.06% 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 : 0.02% optimize.overlap_grad_ring_attention : 0.000005s : 0.05% optimize.overlap_grad_flash_sp : 0.000023s : 0.23% optimize.begin_end_overlap_inline : 0.000001s : 0.01% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.03% optimize.split_layernorm_comm : 0.000003s : 0.03% 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 : 0.15% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.08% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000011s : 0.11% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.03% pipeline_parallel_scheduler : 0.000002s : 0.02% auto_monad_reorder : 0.000022s : 0.22% get_jit_bprop_graph : 0.000002s : 0.02% rewriter_after_jit_bprop_graph : 0.000007s : 0.07% opt_after_jit_grad : 0.000635s : 6.44% validate : 0.000055s : 0.56% Time group info: ------[substitution.] 0.000127 21 1.98% : 0.000003s : 2: substitution.elim_not_effective 1.12% : 0.000001s : 2: substitution.fold_const_symbol 5.59% : 0.000007s : 4: substitution.graph_param_transform 78.82% : 0.000100s : 1: substitution.inline 3.33% : 0.000004s : 4: substitution.j_node_and_user_rematch 4.53% : 0.000006s : 4: substitution.remove_not_recompute_node 4.63% : 0.000006s : 4: substitution.replace_old_param ------[type_inference.] 0.003985 2 89.20% : 0.003554s : 1: type_inference.infer 10.80% : 0.000430s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000099 1 100.00% : 0.000099s : 1: match.inline ------[predicate.] 0.000166 967 0.70% : 0.000001s : 9: predicate.accumulaten_eliminater 1.00% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.65% : 0.000001s : 8: predicate.addn_check_dump 0.78% : 0.000001s : 9: predicate.addn_zero_filter 0.66% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.24% : 0.000004s : 17: predicate.arithmetic_simplify 0.84% : 0.000001s : 9: predicate.cast_eliminate 0.76% : 0.000001s : 8: predicate.check_bprop_eliminate 0.80% : 0.000001s : 8: predicate.compare_switch_simplify 0.23% : 0.000000s : 4: predicate.const_output_eliminate 0.78% : 0.000001s : 8: predicate.depend_value_elim 0.79% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.96% : 0.000002s : 9: predicate.dict_get_item_eliminator 0.84% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.30% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.60% : 0.000001s : 4: predicate.elim_not_effective 0.54% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 0.99% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.93% : 0.000002s : 13: predicate.environ_get_add_eliminate 1.03% : 0.000002s : 13: predicate.environ_get_depend_swap 1.76% : 0.000003s : 21: predicate.environ_get_eliminate 1.21% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.74% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.48% : 0.000002s : 10: predicate.float_depend_g_call 0.64% : 0.000001s : 8: predicate.float_environ_get_switch 0.89% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.26% : 0.000000s : 4: predicate.fold_const_symbol 0.90% : 0.000002s : 8: predicate.get_grad_eliminate 0.42% : 0.000001s : 4: predicate.graph_param_transform 0.68% : 0.000001s : 8: predicate.incorporate_call 0.58% : 0.000001s : 8: predicate.incorporate_call_switch 5.52% : 0.000009s : 43: predicate.inline 1.20% : 0.000002s : 8: predicate.inline_without_move 0.37% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.44% : 0.000002s : 8: predicate.less_batch_normalization 1.87% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.06% : 0.000003s : 26: predicate.load_eliminater 1.52% : 0.000003s : 4: predicate.loop_unroll_after_grad 1.38% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.71% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.68% : 0.000001s : 8: predicate.merge_addn 0.72% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.68% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.65% : 0.000001s : 9: predicate.minmaximum_grad 2.21% : 0.000004s : 4: predicate.mutable_eliminate 0.93% : 0.000002s : 4: predicate.opt_reshape 0.57% : 0.000001s : 4: predicate.parallel_virtual_node 1.09% : 0.000002s : 10: predicate.partial_defer_inline 1.07% : 0.000002s : 13: predicate.partial_eliminate 0.83% : 0.000001s : 9: predicate.print_const_string_wrapper 0.89% : 0.000001s : 8: predicate.reduce_all_const_elim 1.33% : 0.000002s : 9: predicate.reduce_eliminate 2.18% : 0.000004s : 26: predicate.redundant_stop_gradient_eliminater 0.71% : 0.000001s : 8: predicate.remove_not_recompute_node 1.19% : 0.000002s : 17: predicate.replace_applicator 0.83% : 0.000001s : 8: predicate.replace_old_param 0.30% : 0.000000s : 4: predicate.reset_defer_inline 0.87% : 0.000001s : 9: predicate.reshape_eliminate 0.74% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.58% : 0.000001s : 4: predicate.row_tensor_eliminate 1.07% : 0.000002s : 8: predicate.same_eliminate 0.62% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.01% : 0.000002s : 8: predicate.shard_identity_eliminate 0.86% : 0.000001s : 8: predicate.special_op_eliminate 0.79% : 0.000001s : 8: predicate.specialize_transform 1.79% : 0.000003s : 8: predicate.split_environ_get_set_with_tuple_value 1.32% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.47% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.93% : 0.000002s : 10: predicate.switch_defer_inline 1.55% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.51% : 0.000006s : 35: predicate.switch_simplify 0.69% : 0.000001s : 9: predicate.tile_eliminate 0.71% : 0.000001s : 9: predicate.transpose_eliminate 1.54% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.67% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.48% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.17% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.80% : 0.000003s : 17: predicate.tuple_list_get_set_item_eliminator 2.38% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.89% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.88% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.73% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.46% : 0.000001s : 4: predicate.value_based_eliminate 0.97% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.83% : 0.000001s : 8: predicate.virtual_output_eliminate 0.31% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.47% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000242 4 10.37% : 0.000025s : 1: func_graph_cloner_run.FuncGraphClonerGraph 89.63% : 0.000217s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.411817 192 0.00% : 0.000004s : 1: ForceFp32Comm 47.84% : 0.197003s : 1: add_attr 47.83% : 0.196981s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.02% : 0.000064s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.02% : 0.000062s : 1: auto_monad 0.01% : 0.000027s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.18% : 0.000726s : 1: bootstrap 0.01% : 0.000045s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000021s : 1: control_data_broadcast_order 0.00% : 0.000011s : 1: convert_after_rewriter 0.01% : 0.000034s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000010s : 1: environ_conv 0.01% : 0.000021s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000010s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000010s : 1: label_micro_interleaved_index 0.14% : 0.000558s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.19% : 0.000800s : 1: mutable_eliminate 0.00% : 0.000008s : 1: offloading_packed_experts 0.00% : 0.000020s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000023s : 1: opt.transform.mutable_eliminate 0.20% : 0.000841s : 78: opt.transform.opt_a 0.01% : 0.000031s : 1: opt.transform.opt_after_cconv 0.01% : 0.000030s : 1: opt.transform.opt_after_jit_grad 0.03% : 0.000109s : 28: opt.transform.opt_b 0.01% : 0.000060s : 2: opt.transform.opt_trans_graph 0.01% : 0.000041s : 4: opt.transform.symbol_engine_opt 0.61% : 0.002521s : 1: opt_a 0.03% : 0.000139s : 1: opt_after_cconv 0.16% : 0.000650s : 1: opt_after_jit_grad 0.06% : 0.000247s : 1: opt_b 1.26% : 0.005169s : 1: optimize 0.01% : 0.000025s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000010s : 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% : 0.000009s : 1: overlap_grad_ring_attention 0.00% : 0.000008s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000009s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000004s : 1: pipeline_split 0.01% : 0.000045s : 1: pre_auto_parallel 0.01% : 0.000033s : 1: py_interpret_to_execute 0.00% : 0.000019s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000008s : 1: remove_cast_before_assign_add 0.01% : 0.000059s : 1: remove_dup_value 0.11% : 0.000442s : 1: renormalize.infer 0.07% : 0.000285s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000011s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000051s : 1: rewriter_after_opt_a 0.01% : 0.000058s : 1: rewriter_before_opt_a 0.00% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000006s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.02% : 0.000093s : 1: symbol_engine_optimizer 0.02% : 0.000097s : 1: tuple_transform 0.99% : 0.004073s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[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_fp16[dim_type0-False-batch_size_list0-shape0-ms_dtype0],max_mem:8.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:39:00.282.534 [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 = 0.0158938, [21] [bootstrap]: 0.00066887 [type_inference]: 0.00426233 [event_method]: 1.569e-05 [auto_monad]: 6.172e-05 [graph_reusing]: 5.94999e-06 [inline]: 3.13e-06 [add_attr]: 0.00466388, [1] [add_attr_with_inline]: 0.0046446, [1] [Cycle 1]: 7.231e-05, [2] [tag_attr]: 1.887e-05 [meta_addattr_fg_expand]: 3.66999e-06 [parallel-infer-symbol]: 4.58999e-06 [pre_auto_parallel]: 3.943e-05 [insert-virtual-dataset]: 3.14999e-06 [parallel-infer-symbol-second]: 8.29983e-07 [dataset_repeat_opt]: 2.57001e-06 [pipeline_split]: 1.99e-06 [optimize]: 0.005365, [53] [py_interpret_to_execute]: 2.559e-05 [rewriter_before_opt_a]: 5.276e-05 [opt_a]: 0.00280286, [2] [Cycle 1]: 0.00203192, [45] [expand_dump_flag]: 2.66999e-06 [switch_simplify]: 2.579e-05 [loop_unroll]: 1.189e-05 [a_1]: 0.00031433 [with_stream_mark]: 2.528e-05 [recompute_prepare]: 9.37001e-06 [updatestate_depend_eliminate]: 4.3e-06 [updatestate_assign_eliminate]: 4.43001e-06 [updatestate_loads_eliminate]: 3.6e-06 [parameter_eliminate]: 2.16e-06 [a_2]: 0.00016724 [accelerated_algorithm]: 7.96001e-06 [shard]: 3.18998e-06 [meta_shard_fg_expand]: 2.16e-06 [shard_inline]: 7.15e-06 [merge_send_recv]: 1.061e-05 [auto_parallel]: 8.38001e-06 [parallel]: 3.027e-05 [flash_sp]: 1.109e-05 [merge_comm]: 3.8e-06 [allreduce_fusion]: 4.70999e-06 [matmul_add_comm_reduction]: 1.164e-05 [allreduce_slice_to_reducescatter]: 6.50005e-07 [virtual_shard_identity]: 9.25999e-06 [virtual_dataset]: 7.03998e-06 [get_grad_eliminate_]: 6.39999e-06 [virtual_output]: 6.59999e-06 [merge_forward]: 4.80001e-06 [cell_reuse_recompute_pass]: 1.79e-06 [offload_activation]: 1.306e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.608e-05 [merge_recompute_call_nodes]: 1.84998e-06 [before_grad]: 1.241e-05 [set_forward_comm_id_for_comm_node_pass]: 4.41002e-06 [meta_fg_expand]: 3.06001e-06 [flash_sp_send_recv_attached]: 3.66001e-06 [receive_attached]: 2.15002e-06 [after_resolve]: 1.491e-05 [a_after_grad]: 1.065e-05 [renormalize]: 0.00080822 [add_forward_monad_depend]: 8.70001e-06 [auto_monad_grad]: 3.25998e-06 [auto_monad_eliminator]: 2.306e-05 [cse]: 4.574e-05 [a_3]: 5.553e-05 [Cycle 2]: 0.0007571, [45] [expand_dump_flag]: 1.99e-06 [switch_simplify]: 9.19e-06 [loop_unroll]: 6.24999e-06 [a_1]: 0.00012558 [with_stream_mark]: 1.879e-05 [recompute_prepare]: 7.39002e-06 [updatestate_depend_eliminate]: 4.52e-06 [updatestate_assign_eliminate]: 3.34001e-06 [updatestate_loads_eliminate]: 3.52997e-06 [parameter_eliminate]: 2.29001e-06 [a_2]: 7.107e-05 [accelerated_algorithm]: 6.95002e-06 [shard]: 3.06001e-06 [meta_shard_fg_expand]: 3.13e-06 [shard_inline]: 6.45002e-06 [merge_send_recv]: 1.027e-05 [auto_parallel]: 1.057e-05 [parallel]: 9.71e-06 [flash_sp]: 4.88001e-06 [merge_comm]: 4.60001e-06 [allreduce_fusion]: 4.28999e-06 [matmul_add_comm_reduction]: 1.928e-05 [allreduce_slice_to_reducescatter]: 9.20001e-07 [virtual_shard_identity]: 8.28001e-06 [virtual_dataset]: 5.75001e-06 [get_grad_eliminate_]: 5.80002e-06 [virtual_output]: 6.03002e-06 [merge_forward]: 4.46002e-06 [cell_reuse_recompute_pass]: 2.99001e-06 [offload_activation]: 1.732e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.599e-05 [merge_recompute_call_nodes]: 1.60001e-06 [before_grad]: 9.97001e-06 [set_forward_comm_id_for_comm_node_pass]: 4.05e-06 [meta_fg_expand]: 2.32999e-06 [flash_sp_send_recv_attached]: 1.81e-06 [receive_attached]: 1.86e-06 [after_resolve]: 1.139e-05 [a_after_grad]: 9.92001e-06 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 1.32e-06 [auto_monad_grad]: 2.14e-06 [auto_monad_eliminator]: 1.151e-05 [cse]: 2.706e-05 [a_3]: 3.668e-05 [py_interpret_to_execute_after_opt_a]: 1.673e-05 [slice_cell_reuse_recomputed_activation]: 2.06e-06 [rewriter_after_opt_a]: 4.933e-05 [convert_after_rewriter]: 8.05999e-06 [order_py_execute_after_rewriter]: 5.68002e-06 [mutable_eliminate]: 0.00075348 [opt_b]: 0.00024344, [1] [Cycle 1]: 0.00023324, [7] [b_1]: 0.00012518 [b_2]: 9.32999e-06 [updatestate_depend_eliminate]: 1.087e-05 [updatestate_assign_eliminate]: 3.75e-06 [updatestate_loads_eliminate]: 2.90998e-06 [renormalize]: 9.20001e-07 [cse]: 4.224e-05 [optimize_parallel_all_gather_comm]: 2.516e-05 [overlap_param_gather]: 2.37001e-06 [cconv]: 3.854e-05 [loop_unroll]: 0.00051709 [opt_after_cconv]: 0.00012545, [1] [Cycle 1]: 0.00011811, [7] [c_1]: 3.126e-05 [parameter_eliminate]: 6.07999e-06 [updatestate_depend_eliminate]: 7.8e-06 [updatestate_assign_eliminate]: 2.81999e-06 [updatestate_loads_eliminate]: 2.52001e-06 [cse]: 2.765e-05 [renormalize]: 6.30011e-07 [remove_dup_value]: 4.766e-05 [tuple_transform]: 8.668e-05, [1] [Cycle 1]: 7.656e-05, [4] [d_1]: 4.998e-05 [none_parameter_eliminate]: 1.81e-06 [renormalize]: 1.79978e-07 [switch_simplify]: 6.75002e-06 [partial_unused_args_eliminate]: 2.27999e-06 [add_recomputation]: 5.978e-05 [cse_after_recomputation]: 3.365e-05, [1] [Cycle 1]: 2.847e-05, [1] [cse]: 2.197e-05 [environ_conv]: 7.14001e-06 [swap_dp_allreduce_reducescatter]: 6.04001e-06 [bias_add_comm_swap]: 3.03e-06 [label_micro_interleaved_index]: 5.67999e-06 [label_fine_grained_interleaved_index]: 2.76e-06 [merge_cast_opt]: 1.55999e-06 [slice_recompute_activation]: 2.54999e-06 [micro_interleaved_order_control]: 2.24001e-06 [assign_add_opt]: 1.39998e-06 [ForceFp32Comm]: 8.29983e-07 [remove_cast_before_assign_add]: 1.10001e-06 [full_micro_interleaved_order_control]: 2.29001e-06 [reorder_send_recv_between_fp_bp]: 2.92002e-06 [comm_op_add_attrs]: 1.15999e-06 [add_comm_op_reuse_tag]: 1.02998e-06 [interleave_split_concat_branches]: 1.14e-06 [interleave_parallel_branches]: 1.49e-06 [overlap_opt_shard_in_pipeline]: 1.40999e-06 [overlap_opt_shard_grad_in_pipeline]: 2.16e-06 [control_data_broadcast_order]: 1.574e-05 [grouped_pairwise_exchange_alltoall]: 1.59998e-06 [offloading_packed_experts]: 4.43999e-06 [overlap_recompute_and_grad_model_parallel]: 5.56e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.17e-06 [overlap_recompute_allgather_and_fa_grad]: 1.42999e-06 [overlap_recompute_comm]: 2.51998e-06 [overlap_grad_ring_attention]: 4.79e-06 [overlap_grad_flash_sp]: 2.353e-05 [begin_end_overlap_inline]: 5.79981e-07 [split_matmul_comm_elemetwise]: 2.61e-06 [split_layernorm_comm]: 1.81e-06 [handle_group_info]: 1.26002e-06 [symbol_engine_optimizer]: 7.847e-05, [1] [Cycle 1]: 7.353e-05, [6] [build]: 4.62e-06 [elim_shapecalc]: 1.064e-05 [elim_not_effective]: 1.302e-05 [opt_reshape]: 7.08e-06 [fold_const_symbol]: 9.94999e-06 [renormalize]: 2.19996e-07 [detach_backward]: 2.19999e-06 [pipeline_parallel_scheduler]: 1.50999e-06 [auto_monad_reorder]: 2.297e-05 [get_jit_bprop_graph]: 2.01e-06 [rewriter_after_jit_bprop_graph]: 5.44998e-06 [opt_after_jit_grad]: 0.00051104 [validate]: 5.016e-05 Sums bootstrap : 0.000669s : 6.64% type_inference : 0.004262s : 42.32% event_method : 0.000016s : 0.16% auto_monad : 0.000062s : 0.61% graph_reusing : 0.000006s : 0.06% inline : 0.000003s : 0.03% add_attr.add_attr_with_inline.tag_attr : 0.000019s : 0.19% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.04% parallel-infer-symbol : 0.000005s : 0.05% pre_auto_parallel : 0.000039s : 0.39% insert-virtual-dataset : 0.000003s : 0.03% parallel-infer-symbol-second : 0.000001s : 0.01% dataset_repeat_opt : 0.000003s : 0.03% pipeline_split : 0.000002s : 0.02% optimize.py_interpret_to_execute : 0.000026s : 0.25% optimize.rewriter_before_opt_a : 0.000053s : 0.52% optimize.opt_a.expand_dump_flag : 0.000005s : 0.05% optimize.opt_a.switch_simplify : 0.000035s : 0.35% optimize.opt_a.loop_unroll : 0.000018s : 0.18% optimize.opt_a.a_1 : 0.000440s : 4.37% optimize.opt_a.with_stream_mark : 0.000044s : 0.44% optimize.opt_a.recompute_prepare : 0.000017s : 0.17% optimize.opt_a.updatestate_depend_eliminate : 0.000009s : 0.09% optimize.opt_a.updatestate_assign_eliminate : 0.000008s : 0.08% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.07% optimize.opt_a.parameter_eliminate : 0.000004s : 0.04% optimize.opt_a.a_2 : 0.000238s : 2.37% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.15% optimize.opt_a.shard : 0.000006s : 0.06% optimize.opt_a.meta_shard_fg_expand : 0.000005s : 0.05% optimize.opt_a.shard_inline : 0.000014s : 0.14% optimize.opt_a.merge_send_recv : 0.000021s : 0.21% optimize.opt_a.auto_parallel : 0.000019s : 0.19% optimize.opt_a.parallel : 0.000040s : 0.40% optimize.opt_a.flash_sp : 0.000016s : 0.16% optimize.opt_a.merge_comm : 0.000008s : 0.08% optimize.opt_a.allreduce_fusion : 0.000009s : 0.09% optimize.opt_a.matmul_add_comm_reduction : 0.000031s : 0.31% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.02% optimize.opt_a.virtual_shard_identity : 0.000018s : 0.17% optimize.opt_a.virtual_dataset : 0.000013s : 0.13% optimize.opt_a.get_grad_eliminate_ : 0.000012s : 0.12% 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 : 0.000005s : 0.05% optimize.opt_a.offload_activation : 0.000030s : 0.30% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000032s : 0.32% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.03% optimize.opt_a.before_grad : 0.000022s : 0.22% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.08% optimize.opt_a.meta_fg_expand : 0.000005s : 0.05% 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 : 0.26% optimize.opt_a.a_after_grad : 0.000021s : 0.20% optimize.opt_a.renormalize : 0.000808s : 8.03% optimize.opt_a.add_forward_monad_depend : 0.000010s : 0.10% optimize.opt_a.auto_monad_grad : 0.000005s : 0.05% optimize.opt_a.auto_monad_eliminator : 0.000035s : 0.34% optimize.opt_a.cse : 0.000073s : 0.72% optimize.opt_a.a_3 : 0.000092s : 0.92% 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 : 0.000049s : 0.49% optimize.convert_after_rewriter : 0.000008s : 0.08% optimize.order_py_execute_after_rewriter : 0.000006s : 0.06% optimize.mutable_eliminate : 0.000753s : 7.48% optimize.opt_b.b_1 : 0.000125s : 1.24% 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 : 0.000004s : 0.04% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.03% optimize.opt_b.renormalize : 0.000001s : 0.01% optimize.opt_b.cse : 0.000042s : 0.42% optimize.optimize_parallel_all_gather_comm : 0.000025s : 0.25% optimize.overlap_param_gather : 0.000002s : 0.02% optimize.cconv : 0.000039s : 0.38% optimize.loop_unroll : 0.000517s : 5.13% optimize.opt_after_cconv.c_1 : 0.000031s : 0.31% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.06% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000008s : 0.08% 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 : 0.000028s : 0.27% optimize.opt_after_cconv.renormalize : 0.000001s : 0.01% optimize.remove_dup_value : 0.000048s : 0.47% optimize.tuple_transform.d_1 : 0.000050s : 0.50% 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 : 0.07% optimize.partial_unused_args_eliminate : 0.000002s : 0.02% optimize.add_recomputation : 0.000060s : 0.59% optimize.cse_after_recomputation.cse : 0.000022s : 0.22% optimize.environ_conv : 0.000007s : 0.07% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.06% optimize.bias_add_comm_swap : 0.000003s : 0.03% optimize.label_micro_interleaved_index : 0.000006s : 0.06% 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 : 0.000002s : 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 : 0.01% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.02% optimize.control_data_broadcast_order : 0.000016s : 0.16% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.02% optimize.offloading_packed_experts : 0.000004s : 0.04% optimize.overlap_recompute_and_grad_model_parallel : 0.000006s : 0.06% 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 : 0.000005s : 0.05% optimize.overlap_grad_flash_sp : 0.000024s : 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 : 0.000005s : 0.05% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000011s : 0.11% optimize.symbol_engine_optimizer.elim_not_effective : 0.000013s : 0.13% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.07% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.10% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.02% pipeline_parallel_scheduler : 0.000002s : 0.01% auto_monad_reorder : 0.000023s : 0.23% get_jit_bprop_graph : 0.000002s : 0.02% rewriter_after_jit_bprop_graph : 0.000005s : 0.05% opt_after_jit_grad : 0.000511s : 5.07% validate : 0.000050s : 0.50% Time group info: ------[substitution.] 0.000128 21 1.66% : 0.000002s : 2: substitution.elim_not_effective 1.46% : 0.000002s : 2: substitution.fold_const_symbol 5.61% : 0.000007s : 4: substitution.graph_param_transform 77.86% : 0.000099s : 1: substitution.inline 4.07% : 0.000005s : 4: substitution.j_node_and_user_rematch 5.15% : 0.000007s : 4: substitution.remove_not_recompute_node 4.18% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.004190 2 87.59% : 0.003669s : 1: type_inference.infer 12.41% : 0.000520s : 1: type_inference.specialize ------[replace.] 0.000020 1 100.00% : 0.000020s : 1: replace.inline ------[match.] 0.000098 1 100.00% : 0.000098s : 1: match.inline ------[predicate.] 0.000162 967 0.72% : 0.000001s : 9: predicate.accumulaten_eliminater 1.11% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.57% : 0.000001s : 8: predicate.addn_check_dump 1.10% : 0.000002s : 9: predicate.addn_zero_filter 0.68% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.49% : 0.000004s : 17: predicate.arithmetic_simplify 0.79% : 0.000001s : 9: predicate.cast_eliminate 0.85% : 0.000001s : 8: predicate.check_bprop_eliminate 0.58% : 0.000001s : 8: predicate.compare_switch_simplify 0.25% : 0.000000s : 4: predicate.const_output_eliminate 0.64% : 0.000001s : 8: predicate.depend_value_elim 0.96% : 0.000002s : 9: predicate.dict_get_item_const_eliminator 0.84% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.80% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.25% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.37% : 0.000001s : 4: predicate.elim_not_effective 0.67% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.08% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.98% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.93% : 0.000002s : 13: predicate.environ_get_depend_swap 1.81% : 0.000003s : 21: predicate.environ_get_eliminate 0.96% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.85% : 0.000001s : 10: predicate.exchange_switch_depend_value 2.02% : 0.000003s : 10: predicate.float_depend_g_call 0.61% : 0.000001s : 8: predicate.float_environ_get_switch 0.89% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.25% : 0.000000s : 4: predicate.fold_const_symbol 0.96% : 0.000002s : 8: predicate.get_grad_eliminate 0.37% : 0.000001s : 4: predicate.graph_param_transform 0.69% : 0.000001s : 8: predicate.incorporate_call 0.57% : 0.000001s : 8: predicate.incorporate_call_switch 5.65% : 0.000009s : 43: predicate.inline 1.13% : 0.000002s : 8: predicate.inline_without_move 0.41% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.96% : 0.000002s : 8: predicate.less_batch_normalization 1.62% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 1.94% : 0.000003s : 26: predicate.load_eliminater 1.42% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.18% : 0.000002s : 13: predicate.loop_unroll_before_grad 2.05% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.60% : 0.000001s : 8: predicate.merge_addn 0.69% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.67% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.72% : 0.000001s : 9: predicate.minmaximum_grad 2.28% : 0.000004s : 4: predicate.mutable_eliminate 0.58% : 0.000001s : 4: predicate.opt_reshape 0.50% : 0.000001s : 4: predicate.parallel_virtual_node 1.42% : 0.000002s : 10: predicate.partial_defer_inline 1.10% : 0.000002s : 13: predicate.partial_eliminate 0.76% : 0.000001s : 9: predicate.print_const_string_wrapper 0.89% : 0.000001s : 8: predicate.reduce_all_const_elim 1.06% : 0.000002s : 9: predicate.reduce_eliminate 2.38% : 0.000004s : 26: predicate.redundant_stop_gradient_eliminater 0.66% : 0.000001s : 8: predicate.remove_not_recompute_node 1.21% : 0.000002s : 17: predicate.replace_applicator 0.93% : 0.000002s : 8: predicate.replace_old_param 0.58% : 0.000001s : 4: predicate.reset_defer_inline 0.79% : 0.000001s : 9: predicate.reshape_eliminate 0.98% : 0.000002s : 8: predicate.row_tensor_add_zeros_like 0.64% : 0.000001s : 4: predicate.row_tensor_eliminate 1.18% : 0.000002s : 8: predicate.same_eliminate 0.67% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.57% : 0.000003s : 8: predicate.shard_identity_eliminate 0.87% : 0.000001s : 8: predicate.special_op_eliminate 0.85% : 0.000001s : 8: predicate.specialize_transform 1.33% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 1.04% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.41% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.79% : 0.000001s : 10: predicate.switch_defer_inline 1.50% : 0.000002s : 18: predicate.switch_layer_defer_inline 3.60% : 0.000006s : 35: predicate.switch_simplify 0.80% : 0.000001s : 9: predicate.tile_eliminate 0.70% : 0.000001s : 9: predicate.transpose_eliminate 1.60% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.85% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.39% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.41% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.42% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.48% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.41% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 1.94% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.63% : 0.000004s : 34: predicate.updatestate_useless_node_eliminater 0.53% : 0.000001s : 4: predicate.value_based_eliminate 0.82% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.79% : 0.000001s : 8: predicate.virtual_output_eliminate 0.35% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.64% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000298 4 8.17% : 0.000024s : 1: func_graph_cloner_run.FuncGraphClonerGraph 91.83% : 0.000274s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.027780 192 0.01% : 0.000004s : 1: ForceFp32Comm 16.81% : 0.004671s : 1: add_attr 16.74% : 0.004650s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.23% : 0.000064s : 1: add_recomputation 0.01% : 0.000004s : 1: assign_add_opt 0.24% : 0.000068s : 1: auto_monad 0.11% : 0.000030s : 1: auto_monad_reorder 0.01% : 0.000004s : 1: begin_end_overlap_inline 0.02% : 0.000006s : 1: bias_add_comm_swap 2.60% : 0.000723s : 1: bootstrap 0.15% : 0.000042s : 1: cconv 0.02% : 0.000004s : 1: comm_op_add_attrs 0.07% : 0.000019s : 1: control_data_broadcast_order 0.05% : 0.000013s : 1: convert_after_rewriter 0.13% : 0.000037s : 1: cse_after_recomputation 0.02% : 0.000006s : 1: dataset_repeat_opt 0.02% : 0.000005s : 1: detach_backward 0.04% : 0.000010s : 1: environ_conv 0.08% : 0.000024s : 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% : 0.000005s : 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 0.02% : 0.000004s : 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% : 0.000009s : 1: label_micro_interleaved_index 1.91% : 0.000531s : 1: loop_unroll 0.02% : 0.000004s : 1: merge_cast_opt 0.02% : 0.000005s : 1: micro_interleaved_order_control 2.78% : 0.000772s : 1: mutable_eliminate 0.03% : 0.000007s : 1: offloading_packed_experts 0.06% : 0.000017s : 1: opt.transform.loop_unroll_optimizer 0.08% : 0.000023s : 1: opt.transform.mutable_eliminate 3.36% : 0.000933s : 78: opt.transform.opt_a 0.11% : 0.000030s : 1: opt.transform.opt_after_cconv 0.09% : 0.000024s : 1: opt.transform.opt_after_jit_grad 0.38% : 0.000105s : 28: opt.transform.opt_b 0.20% : 0.000054s : 2: opt.transform.opt_trans_graph 0.13% : 0.000037s : 4: opt.transform.symbol_engine_opt 10.10% : 0.002806s : 1: opt_a 0.46% : 0.000129s : 1: opt_after_cconv 1.88% : 0.000522s : 1: opt_after_jit_grad 0.89% : 0.000247s : 1: opt_b 19.34% : 0.005372s : 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 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.05% : 0.000015s : 1: overlap_grad_ring_attention 0.03% : 0.000008s : 1: overlap_opt_shard_grad_in_pipeline 0.01% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.02% : 0.000006s : 1: overlap_param_gather 0.01% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.03% : 0.000009s : 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% : 0.000005s : 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 0.16% : 0.000045s : 1: pre_auto_parallel 0.11% : 0.000030s : 1: py_interpret_to_execute 0.08% : 0.000021s : 1: py_interpret_to_execute_after_opt_a 0.02% : 0.000007s : 1: remove_cast_before_assign_add 0.19% : 0.000052s : 1: remove_dup_value 1.78% : 0.000495s : 1: renormalize.infer 1.08% : 0.000300s : 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% : 0.000058s : 1: rewriter_after_opt_a 0.21% : 0.000057s : 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% : 0.000081s : 1: symbol_engine_optimizer 0.32% : 0.000090s : 1: tuple_transform 15.42% : 0.004285s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[dim_type0-False-batch_size_list0-shape1-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_fp16[dim_type0-False-batch_size_list0-shape1-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:39:03.780.656 [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 = 0.0333989, [21] [bootstrap]: 0.00067649 [type_inference]: 0.00443771 [event_method]: 1.458e-05 [auto_monad]: 5.744e-05 [graph_reusing]: 5.24998e-06 [inline]: 2.53e-06 [add_attr]: 0.00475211, [1] [add_attr_with_inline]: 0.00473524, [1] [Cycle 1]: 7.201e-05, [2] [tag_attr]: 1.964e-05 [meta_addattr_fg_expand]: 3.83001e-06 [parallel-infer-symbol]: 3.76999e-06 [pre_auto_parallel]: 3.745e-05 [insert-virtual-dataset]: 2.81999e-06 [parallel-infer-symbol-second]: 8.2e-07 [dataset_repeat_opt]: 2.78998e-06 [pipeline_split]: 1.87999e-06 [optimize]: 0.0224804, [53] [py_interpret_to_execute]: 2.338e-05 [rewriter_before_opt_a]: 5.193e-05 [opt_a]: 0.0197631, [2] [Cycle 1]: 0.0189859, [45] [expand_dump_flag]: 2.92002e-06 [switch_simplify]: 2.377e-05 [loop_unroll]: 1.117e-05 [a_1]: 0.00321397 [with_stream_mark]: 3.698e-05 [recompute_prepare]: 1.511e-05 [updatestate_depend_eliminate]: 4.99e-06 [updatestate_assign_eliminate]: 3.54002e-06 [updatestate_loads_eliminate]: 3.38999e-06 [parameter_eliminate]: 2.76999e-06 [a_2]: 8.293e-05 [accelerated_algorithm]: 7.77e-06 [shard]: 2.93998e-06 [meta_shard_fg_expand]: 3.41999e-06 [shard_inline]: 7.36999e-06 [merge_send_recv]: 1.105e-05 [auto_parallel]: 1.114e-05 [parallel]: 0.00078978 [flash_sp]: 3.774e-05 [merge_comm]: 2.176e-05 [allreduce_fusion]: 5.42999e-06 [matmul_add_comm_reduction]: 1.503e-05 [allreduce_slice_to_reducescatter]: 9.50007e-07 [virtual_shard_identity]: 3.738e-05 [virtual_dataset]: 9.40001e-06 [get_grad_eliminate_]: 7.31999e-06 [virtual_output]: 6.67002e-06 [merge_forward]: 5.65001e-06 [cell_reuse_recompute_pass]: 4.4e-06 [offload_activation]: 1.286e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.748e-05 [merge_recompute_call_nodes]: 1.55001e-06 [before_grad]: 1.226e-05 [set_forward_comm_id_for_comm_node_pass]: 4.45e-06 [meta_fg_expand]: 5.82001e-06 [flash_sp_send_recv_attached]: 3.62998e-06 [receive_attached]: 2.43e-06 [after_resolve]: 1.631e-05 [a_after_grad]: 1.169e-05 [renormalize]: 0.00085863 [add_forward_monad_depend]: 8.86002e-06 [auto_monad_grad]: 2.94001e-06 [auto_monad_eliminator]: 2.017e-05 [cse]: 4.669e-05 [a_3]: 6.143e-05 [Cycle 2]: 0.00076315, [45] [expand_dump_flag]: 2.37999e-06 [switch_simplify]: 9.52999e-06 [loop_unroll]: 8.11002e-06 [a_1]: 0.00014183 [with_stream_mark]: 1.791e-05 [recompute_prepare]: 7.73999e-06 [updatestate_depend_eliminate]: 4.75001e-06 [updatestate_assign_eliminate]: 3.98001e-06 [updatestate_loads_eliminate]: 3.86999e-06 [parameter_eliminate]: 2.02999e-06 [a_2]: 8.001e-05 [accelerated_algorithm]: 7.08998e-06 [shard]: 2.59001e-06 [meta_shard_fg_expand]: 2.19999e-06 [shard_inline]: 7.16999e-06 [merge_send_recv]: 7.66999e-06 [auto_parallel]: 1.197e-05 [parallel]: 9.25001e-06 [flash_sp]: 4.28001e-06 [merge_comm]: 4.33001e-06 [allreduce_fusion]: 3.36001e-06 [matmul_add_comm_reduction]: 1.301e-05 [allreduce_slice_to_reducescatter]: 6.39993e-07 [virtual_shard_identity]: 7.95e-06 [virtual_dataset]: 6.59001e-06 [get_grad_eliminate_]: 6.44999e-06 [virtual_output]: 5.92001e-06 [merge_forward]: 3.83999e-06 [cell_reuse_recompute_pass]: 2.63e-06 [offload_activation]: 1.78e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.548e-05 [merge_recompute_call_nodes]: 1.56998e-06 [before_grad]: 9.99001e-06 [set_forward_comm_id_for_comm_node_pass]: 3.88001e-06 [meta_fg_expand]: 2.83998e-06 [flash_sp_send_recv_attached]: 1.25999e-06 [receive_attached]: 1.87999e-06 [after_resolve]: 1.139e-05 [a_after_grad]: 9.17999e-06 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 1.69998e-06 [auto_monad_grad]: 1.26002e-06 [auto_monad_eliminator]: 8.28001e-06 [cse]: 2.334e-05 [a_3]: 3.749e-05 [py_interpret_to_execute_after_opt_a]: 1.852e-05 [slice_cell_reuse_recomputed_activation]: 2.34001e-06 [rewriter_after_opt_a]: 4.244e-05 [convert_after_rewriter]: 7.43999e-06 [order_py_execute_after_rewriter]: 5.83002e-06 [mutable_eliminate]: 0.000876 [opt_b]: 0.00024352, [1] [Cycle 1]: 0.00023425, [7] [b_1]: 0.00013134 [b_2]: 9.30001e-06 [updatestate_depend_eliminate]: 1.011e-05 [updatestate_assign_eliminate]: 2.86e-06 [updatestate_loads_eliminate]: 2.41e-06 [renormalize]: 5.29981e-07 [cse]: 3.945e-05 [optimize_parallel_all_gather_comm]: 2.095e-05 [overlap_param_gather]: 2.11e-06 [cconv]: 3.542e-05 [loop_unroll]: 0.00053565 [opt_after_cconv]: 0.00013133, [1] [Cycle 1]: 0.00012354, [7] [c_1]: 3.215e-05 [parameter_eliminate]: 6.07999e-06 [updatestate_depend_eliminate]: 7.63999e-06 [updatestate_assign_eliminate]: 2.69999e-06 [updatestate_loads_eliminate]: 2.67001e-06 [cse]: 2.966e-05 [renormalize]: 5.39992e-07 [remove_dup_value]: 5.252e-05 [tuple_transform]: 8.778e-05, [1] [Cycle 1]: 7.81e-05, [4] [d_1]: 4.946e-05 [none_parameter_eliminate]: 2.76e-06 [renormalize]: 1.80007e-07 [switch_simplify]: 7.15e-06 [partial_unused_args_eliminate]: 1.99e-06 [add_recomputation]: 5.694e-05 [cse_after_recomputation]: 3.247e-05, [1] [Cycle 1]: 2.739e-05, [1] [cse]: 2.018e-05 [environ_conv]: 7.16999e-06 [swap_dp_allreduce_reducescatter]: 6.21e-06 [bias_add_comm_swap]: 3.46999e-06 [label_micro_interleaved_index]: 5.74999e-06 [label_fine_grained_interleaved_index]: 2.69001e-06 [merge_cast_opt]: 1.64e-06 [slice_recompute_activation]: 2.75002e-06 [micro_interleaved_order_control]: 2.32999e-06 [assign_add_opt]: 2.03997e-06 [ForceFp32Comm]: 1.10999e-06 [remove_cast_before_assign_add]: 1.14e-06 [full_micro_interleaved_order_control]: 2.84001e-06 [reorder_send_recv_between_fp_bp]: 2.88998e-06 [comm_op_add_attrs]: 1.22e-06 [add_comm_op_reuse_tag]: 1.25999e-06 [interleave_split_concat_branches]: 1.26997e-06 [interleave_parallel_branches]: 1.05999e-06 [overlap_opt_shard_in_pipeline]: 1.51998e-06 [overlap_opt_shard_grad_in_pipeline]: 2.62001e-06 [control_data_broadcast_order]: 1.686e-05 [grouped_pairwise_exchange_alltoall]: 1.74e-06 [offloading_packed_experts]: 5.05999e-06 [overlap_recompute_and_grad_model_parallel]: 5.37001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.19e-06 [overlap_recompute_allgather_and_fa_grad]: 1.40001e-06 [overlap_recompute_comm]: 2.49999e-06 [overlap_grad_ring_attention]: 4.38999e-06 [overlap_grad_flash_sp]: 2.359e-05 [begin_end_overlap_inline]: 4.89992e-07 [split_matmul_comm_elemetwise]: 2.83e-06 [split_layernorm_comm]: 2.03002e-06 [handle_group_info]: 1.04e-06 [symbol_engine_optimizer]: 8.686e-05, [1] [Cycle 1]: 8.231e-05, [6] [build]: 4.15999e-06 [elim_shapecalc]: 1.243e-05 [elim_not_effective]: 1.461e-05 [opt_reshape]: 7.33e-06 [fold_const_symbol]: 1.038e-05 [renormalize]: 1.79978e-07 [detach_backward]: 2.26e-06 [pipeline_parallel_scheduler]: 1.67001e-06 [auto_monad_reorder]: 2.246e-05 [get_jit_bprop_graph]: 2.68e-06 [rewriter_after_jit_bprop_graph]: 6.04001e-06 [opt_after_jit_grad]: 0.00062484 [validate]: 5.037e-05 Sums bootstrap : 0.000676s : 4.74% type_inference : 0.004438s : 31.13% event_method : 0.000015s : 0.10% auto_monad : 0.000057s : 0.40% graph_reusing : 0.000005s : 0.04% inline : 0.000003s : 0.02% add_attr.add_attr_with_inline.tag_attr : 0.000020s : 0.14% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.03% parallel-infer-symbol : 0.000004s : 0.03% pre_auto_parallel : 0.000037s : 0.26% insert-virtual-dataset : 0.000003s : 0.02% parallel-infer-symbol-second : 0.000001s : 0.01% dataset_repeat_opt : 0.000003s : 0.02% pipeline_split : 0.000002s : 0.01% optimize.py_interpret_to_execute : 0.000023s : 0.16% optimize.rewriter_before_opt_a : 0.000052s : 0.36% optimize.opt_a.expand_dump_flag : 0.000005s : 0.04% optimize.opt_a.switch_simplify : 0.000033s : 0.23% optimize.opt_a.loop_unroll : 0.000019s : 0.14% optimize.opt_a.a_1 : 0.003356s : 23.54% optimize.opt_a.with_stream_mark : 0.000055s : 0.39% optimize.opt_a.recompute_prepare : 0.000023s : 0.16% optimize.opt_a.updatestate_depend_eliminate : 0.000010s : 0.07% optimize.opt_a.updatestate_assign_eliminate : 0.000008s : 0.05% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.05% optimize.opt_a.parameter_eliminate : 0.000005s : 0.03% optimize.opt_a.a_2 : 0.000163s : 1.14% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.10% optimize.opt_a.shard : 0.000006s : 0.04% optimize.opt_a.meta_shard_fg_expand : 0.000006s : 0.04% optimize.opt_a.shard_inline : 0.000015s : 0.10% optimize.opt_a.merge_send_recv : 0.000019s : 0.13% optimize.opt_a.auto_parallel : 0.000023s : 0.16% optimize.opt_a.parallel : 0.000799s : 5.60% optimize.opt_a.flash_sp : 0.000042s : 0.29% optimize.opt_a.merge_comm : 0.000026s : 0.18% optimize.opt_a.allreduce_fusion : 0.000009s : 0.06% optimize.opt_a.matmul_add_comm_reduction : 0.000028s : 0.20% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000045s : 0.32% optimize.opt_a.virtual_dataset : 0.000016s : 0.11% optimize.opt_a.get_grad_eliminate_ : 0.000014s : 0.10% optimize.opt_a.virtual_output : 0.000013s : 0.09% optimize.opt_a.merge_forward : 0.000009s : 0.07% optimize.opt_a.cell_reuse_recompute_pass : 0.000007s : 0.05% optimize.opt_a.offload_activation : 0.000031s : 0.22% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000043s : 0.30% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.02% optimize.opt_a.before_grad : 0.000022s : 0.16% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.06% optimize.opt_a.meta_fg_expand : 0.000009s : 0.06% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.03% optimize.opt_a.receive_attached : 0.000004s : 0.03% optimize.opt_a.after_resolve : 0.000028s : 0.19% optimize.opt_a.a_after_grad : 0.000021s : 0.15% optimize.opt_a.renormalize : 0.000859s : 6.02% optimize.opt_a.add_forward_monad_depend : 0.000011s : 0.07% optimize.opt_a.auto_monad_grad : 0.000004s : 0.03% optimize.opt_a.auto_monad_eliminator : 0.000028s : 0.20% optimize.opt_a.cse : 0.000070s : 0.49% optimize.opt_a.a_3 : 0.000099s : 0.69% optimize.py_interpret_to_execute_after_opt_a : 0.000019s : 0.13% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.02% optimize.rewriter_after_opt_a : 0.000042s : 0.30% optimize.convert_after_rewriter : 0.000007s : 0.05% optimize.order_py_execute_after_rewriter : 0.000006s : 0.04% optimize.mutable_eliminate : 0.000876s : 6.14% optimize.opt_b.b_1 : 0.000131s : 0.92% optimize.opt_b.b_2 : 0.000009s : 0.07% optimize.opt_b.updatestate_depend_eliminate : 0.000010s : 0.07% optimize.opt_b.updatestate_assign_eliminate : 0.000003s : 0.02% optimize.opt_b.updatestate_loads_eliminate : 0.000002s : 0.02% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000039s : 0.28% optimize.optimize_parallel_all_gather_comm : 0.000021s : 0.15% optimize.overlap_param_gather : 0.000002s : 0.01% optimize.cconv : 0.000035s : 0.25% optimize.loop_unroll : 0.000536s : 3.76% optimize.opt_after_cconv.c_1 : 0.000032s : 0.23% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.04% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000008s : 0.05% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.02% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000003s : 0.02% optimize.opt_after_cconv.cse : 0.000030s : 0.21% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000053s : 0.37% optimize.tuple_transform.d_1 : 0.000049s : 0.35% optimize.tuple_transform.none_parameter_eliminate : 0.000003s : 0.02% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.05% optimize.partial_unused_args_eliminate : 0.000002s : 0.01% optimize.add_recomputation : 0.000057s : 0.40% optimize.cse_after_recomputation.cse : 0.000020s : 0.14% optimize.environ_conv : 0.000007s : 0.05% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.04% optimize.bias_add_comm_swap : 0.000003s : 0.02% optimize.label_micro_interleaved_index : 0.000006s : 0.04% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.02% optimize.merge_cast_opt : 0.000002s : 0.01% optimize.slice_recompute_activation : 0.000003s : 0.02% optimize.micro_interleaved_order_control : 0.000002s : 0.02% optimize.assign_add_opt : 0.000002s : 0.01% optimize.ForceFp32Comm : 0.000001s : 0.01% optimize.remove_cast_before_assign_add : 0.000001s : 0.01% optimize.full_micro_interleaved_order_control : 0.000003s : 0.02% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.02% optimize.comm_op_add_attrs : 0.000001s : 0.01% optimize.add_comm_op_reuse_tag : 0.000001s : 0.01% optimize.interleave_split_concat_branches : 0.000001s : 0.01% optimize.interleave_parallel_branches : 0.000001s : 0.01% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.01% optimize.overlap_opt_shard_grad_in_pipeline : 0.000003s : 0.02% optimize.control_data_broadcast_order : 0.000017s : 0.12% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.01% optimize.offloading_packed_experts : 0.000005s : 0.04% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.04% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.01% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.01% optimize.overlap_recompute_comm : 0.000002s : 0.02% optimize.overlap_grad_ring_attention : 0.000004s : 0.03% optimize.overlap_grad_flash_sp : 0.000024s : 0.17% optimize.begin_end_overlap_inline : 0.000000s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.02% optimize.split_layernorm_comm : 0.000002s : 0.01% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000004s : 0.03% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000012s : 0.09% optimize.symbol_engine_optimizer.elim_not_effective : 0.000015s : 0.10% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.05% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.07% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.02% pipeline_parallel_scheduler : 0.000002s : 0.01% auto_monad_reorder : 0.000022s : 0.16% get_jit_bprop_graph : 0.000003s : 0.02% rewriter_after_jit_bprop_graph : 0.000006s : 0.04% opt_after_jit_grad : 0.000625s : 4.38% validate : 0.000050s : 0.35% Time group info: ------[substitution.] 0.002978 21 0.08% : 0.000002s : 2: substitution.elim_not_effective 0.05% : 0.000001s : 2: substitution.fold_const_symbol 0.22% : 0.000006s : 4: substitution.graph_param_transform 99.00% : 0.002948s : 1: substitution.inline 0.17% : 0.000005s : 4: substitution.j_node_and_user_rematch 0.21% : 0.000006s : 4: substitution.remove_not_recompute_node 0.27% : 0.000008s : 4: substitution.replace_old_param ------[type_inference.] 0.004375 2 89.82% : 0.003929s : 1: type_inference.infer 10.18% : 0.000445s : 1: type_inference.specialize ------[replace.] 0.000023 1 100.00% : 0.000023s : 1: replace.inline ------[match.] 0.002943 1 100.00% : 0.002943s : 1: match.inline ------[predicate.] 0.000191 967 0.83% : 0.000002s : 9: predicate.accumulaten_eliminater 1.20% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.59% : 0.000001s : 8: predicate.addn_check_dump 0.99% : 0.000002s : 9: predicate.addn_zero_filter 0.54% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.55% : 0.000005s : 17: predicate.arithmetic_simplify 0.99% : 0.000002s : 9: predicate.cast_eliminate 1.04% : 0.000002s : 8: predicate.check_bprop_eliminate 0.57% : 0.000001s : 8: predicate.compare_switch_simplify 0.21% : 0.000000s : 4: predicate.const_output_eliminate 0.70% : 0.000001s : 8: predicate.depend_value_elim 0.75% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 1.12% : 0.000002s : 9: predicate.dict_get_item_eliminator 1.02% : 0.000002s : 9: predicate.dict_set_item_eliminator 1.91% : 0.000004s : 8: predicate.dumpgradient_eliminate 0.28% : 0.000001s : 4: predicate.elim_not_effective 0.52% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.17% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.11% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.92% : 0.000002s : 13: predicate.environ_get_depend_swap 1.62% : 0.000003s : 21: predicate.environ_get_eliminate 1.20% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.74% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.88% : 0.000004s : 10: predicate.float_depend_g_call 0.59% : 0.000001s : 8: predicate.float_environ_get_switch 0.92% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.20% : 0.000000s : 4: predicate.fold_const_symbol 0.96% : 0.000002s : 8: predicate.get_grad_eliminate 0.31% : 0.000001s : 4: predicate.graph_param_transform 0.58% : 0.000001s : 8: predicate.incorporate_call 0.51% : 0.000001s : 8: predicate.incorporate_call_switch 5.64% : 0.000011s : 43: predicate.inline 1.09% : 0.000002s : 8: predicate.inline_without_move 0.34% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.17% : 0.000002s : 8: predicate.less_batch_normalization 1.73% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 1.86% : 0.000004s : 26: predicate.load_eliminater 1.39% : 0.000003s : 4: predicate.loop_unroll_after_grad 1.40% : 0.000003s : 13: predicate.loop_unroll_before_grad 2.15% : 0.000004s : 17: predicate.make_slice_get_slice_eliminator 0.65% : 0.000001s : 8: predicate.merge_addn 0.70% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.71% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.62% : 0.000001s : 9: predicate.minmaximum_grad 1.75% : 0.000003s : 4: predicate.mutable_eliminate 0.62% : 0.000001s : 4: predicate.opt_reshape 0.48% : 0.000001s : 4: predicate.parallel_virtual_node 1.55% : 0.000003s : 10: predicate.partial_defer_inline 0.94% : 0.000002s : 13: predicate.partial_eliminate 0.94% : 0.000002s : 9: predicate.print_const_string_wrapper 1.02% : 0.000002s : 8: predicate.reduce_all_const_elim 1.16% : 0.000002s : 9: predicate.reduce_eliminate 1.88% : 0.000004s : 26: predicate.redundant_stop_gradient_eliminater 0.84% : 0.000002s : 8: predicate.remove_not_recompute_node 1.19% : 0.000002s : 17: predicate.replace_applicator 0.67% : 0.000001s : 8: predicate.replace_old_param 0.31% : 0.000001s : 4: predicate.reset_defer_inline 0.98% : 0.000002s : 9: predicate.reshape_eliminate 0.82% : 0.000002s : 8: predicate.row_tensor_add_zeros_like 0.46% : 0.000001s : 4: predicate.row_tensor_eliminate 1.06% : 0.000002s : 8: predicate.same_eliminate 0.44% : 0.000001s : 8: predicate.set_cell_output_no_recompute 2.24% : 0.000004s : 8: predicate.shard_identity_eliminate 0.85% : 0.000002s : 8: predicate.special_op_eliminate 0.77% : 0.000001s : 8: predicate.specialize_transform 1.11% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 1.14% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.43% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.88% : 0.000002s : 10: predicate.switch_defer_inline 1.74% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.35% : 0.000006s : 35: predicate.switch_simplify 0.82% : 0.000002s : 9: predicate.tile_eliminate 0.94% : 0.000002s : 9: predicate.transpose_eliminate 1.51% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.72% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.51% : 0.000003s : 17: predicate.tuple_list_get_item_depend_reorder 3.39% : 0.000006s : 25: predicate.tuple_list_get_item_eliminator 1.29% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.32% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.73% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.69% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.43% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.38% : 0.000001s : 4: predicate.value_based_eliminate 1.00% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.83% : 0.000002s : 8: predicate.virtual_output_eliminate 0.28% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.56% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000260 4 8.73% : 0.000023s : 1: func_graph_cloner_run.FuncGraphClonerGraph 91.27% : 0.000237s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.065441 192 0.01% : 0.000004s : 1: ForceFp32Comm 7.27% : 0.004759s : 1: add_attr 7.24% : 0.004740s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.09% : 0.000062s : 1: add_recomputation 0.01% : 0.000005s : 1: assign_add_opt 0.10% : 0.000063s : 1: auto_monad 0.05% : 0.000031s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.01% : 0.000006s : 1: bias_add_comm_swap 1.13% : 0.000740s : 1: bootstrap 0.06% : 0.000039s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.03% : 0.000021s : 1: control_data_broadcast_order 0.02% : 0.000011s : 1: convert_after_rewriter 0.05% : 0.000035s : 1: cse_after_recomputation 0.01% : 0.000006s : 1: dataset_repeat_opt 0.01% : 0.000005s : 1: detach_backward 0.02% : 0.000011s : 1: environ_conv 0.03% : 0.000021s : 1: event_method 0.01% : 0.000006s : 1: full_micro_interleaved_order_control 0.01% : 0.000006s : 1: get_jit_bprop_graph 0.01% : 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% : 0.000005s : 1: interleave_split_concat_branches 0.01% : 0.000006s : 1: label_fine_grained_interleaved_index 0.01% : 0.000009s : 1: label_micro_interleaved_index 0.84% : 0.000548s : 1: loop_unroll 0.01% : 0.000005s : 1: merge_cast_opt 0.01% : 0.000005s : 1: micro_interleaved_order_control 1.36% : 0.000893s : 1: mutable_eliminate 0.01% : 0.000008s : 1: offloading_packed_experts 0.03% : 0.000017s : 1: opt.transform.loop_unroll_optimizer 0.03% : 0.000022s : 1: opt.transform.mutable_eliminate 5.83% : 0.003815s : 78: opt.transform.opt_a 0.05% : 0.000030s : 1: opt.transform.opt_after_cconv 0.05% : 0.000030s : 1: opt.transform.opt_after_jit_grad 0.17% : 0.000110s : 28: opt.transform.opt_b 0.08% : 0.000054s : 2: opt.transform.opt_trans_graph 0.06% : 0.000041s : 4: opt.transform.symbol_engine_opt 30.20% : 0.019766s : 1: opt_a 0.21% : 0.000135s : 1: opt_after_cconv 0.98% : 0.000640s : 1: opt_after_jit_grad 0.38% : 0.000247s : 1: opt_b 34.36% : 0.022487s : 1: optimize 0.04% : 0.000025s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000009s : 1: order_py_execute_after_rewriter 0.04% : 0.000027s : 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% : 0.000009s : 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% : 0.000005s : 1: pipeline_split 0.06% : 0.000042s : 1: pre_auto_parallel 0.04% : 0.000027s : 1: py_interpret_to_execute 0.03% : 0.000022s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000008s : 1: remove_cast_before_assign_add 0.09% : 0.000057s : 1: remove_dup_value 0.80% : 0.000524s : 1: renormalize.infer 0.49% : 0.000323s : 1: renormalize.specialize 0.01% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.07% : 0.000047s : 1: rewriter_after_opt_a 0.09% : 0.000056s : 1: rewriter_before_opt_a 0.01% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.01% : 0.000006s : 1: slice_recompute_activation 0.01% : 0.000006s : 1: split_layernorm_comm 0.04% : 0.000024s : 1: split_matmul_comm_elemetwise 0.01% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.14% : 0.000089s : 1: symbol_engine_optimizer 0.14% : 0.000091s : 1: tuple_transform 6.82% : 0.004464s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[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_fp16[dim_type0-False-batch_size_list1-shape0-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:39:06.772.594 [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 = 0.127473, [21] [bootstrap]: 0.00074698 [type_inference]: 0.115323 [event_method]: 1.879e-05 [auto_monad]: 6.688e-05 [graph_reusing]: 5.97001e-06 [inline]: 3.12002e-06 [add_attr]: 0.0053042, [1] [add_attr_with_inline]: 0.00528779, [1] [Cycle 1]: 7.017e-05, [2] [tag_attr]: 1.741e-05 [meta_addattr_fg_expand]: 3.58999e-06 [parallel-infer-symbol]: 3.45e-06 [pre_auto_parallel]: 3.601e-05 [insert-virtual-dataset]: 2.73e-06 [parallel-infer-symbol-second]: 8.39995e-07 [dataset_repeat_opt]: 2.56e-06 [pipeline_split]: 2.12001e-06 [optimize]: 0.00511786, [53] [py_interpret_to_execute]: 2.407e-05 [rewriter_before_opt_a]: 5.418e-05 [opt_a]: 0.00273257, [2] [Cycle 1]: 0.00199138, [45] [expand_dump_flag]: 2.64999e-06 [switch_simplify]: 2.359e-05 [loop_unroll]: 1.163e-05 [a_1]: 0.00036932 [with_stream_mark]: 2.341e-05 [recompute_prepare]: 9.02999e-06 [updatestate_depend_eliminate]: 4.42e-06 [updatestate_assign_eliminate]: 3.78001e-06 [updatestate_loads_eliminate]: 3.21999e-06 [parameter_eliminate]: 1.96e-06 [a_2]: 8.037e-05 [accelerated_algorithm]: 6.84001e-06 [shard]: 3.4e-06 [meta_shard_fg_expand]: 1.82999e-06 [shard_inline]: 6.74999e-06 [merge_send_recv]: 8.94e-06 [auto_parallel]: 7.48999e-06 [parallel]: 3.207e-05 [flash_sp]: 9.91998e-06 [merge_comm]: 3.98001e-06 [allreduce_fusion]: 4.05998e-06 [matmul_add_comm_reduction]: 1.511e-05 [allreduce_slice_to_reducescatter]: 9.89996e-07 [virtual_shard_identity]: 8.50999e-06 [virtual_dataset]: 7.01001e-06 [get_grad_eliminate_]: 6.86001e-06 [virtual_output]: 6.49001e-06 [merge_forward]: 5.00999e-06 [cell_reuse_recompute_pass]: 1.57999e-06 [offload_activation]: 1.169e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.533e-05 [merge_recompute_call_nodes]: 1.76e-06 [before_grad]: 1.025e-05 [set_forward_comm_id_for_comm_node_pass]: 4.72e-06 [meta_fg_expand]: 3.06001e-06 [flash_sp_send_recv_attached]: 2.73e-06 [receive_attached]: 2.53998e-06 [after_resolve]: 1.281e-05 [a_after_grad]: 1.003e-05 [renormalize]: 0.00076981 [add_forward_monad_depend]: 7.15998e-06 [auto_monad_grad]: 2.48e-06 [auto_monad_eliminator]: 1.708e-05 [cse]: 4.559e-05 [a_3]: 0.00013088 [Cycle 2]: 0.00072878, [45] [expand_dump_flag]: 2.02001e-06 [switch_simplify]: 9.55001e-06 [loop_unroll]: 7.78999e-06 [a_1]: 0.00012587 [with_stream_mark]: 1.752e-05 [recompute_prepare]: 7.11999e-06 [updatestate_depend_eliminate]: 3.95998e-06 [updatestate_assign_eliminate]: 3.56999e-06 [updatestate_loads_eliminate]: 3.07002e-06 [parameter_eliminate]: 1.79e-06 [a_2]: 7.239e-05 [accelerated_algorithm]: 6.66e-06 [shard]: 1.39e-06 [meta_shard_fg_expand]: 1.87999e-06 [shard_inline]: 6.49001e-06 [merge_send_recv]: 7.13998e-06 [auto_parallel]: 9.31002e-06 [parallel]: 7.87e-06 [flash_sp]: 4.11001e-06 [merge_comm]: 4.40999e-06 [allreduce_fusion]: 3.91999e-06 [matmul_add_comm_reduction]: 1.372e-05 [allreduce_slice_to_reducescatter]: 3.80009e-07 [virtual_shard_identity]: 8.38999e-06 [virtual_dataset]: 6.30002e-06 [get_grad_eliminate_]: 5.77001e-06 [virtual_output]: 5.51e-06 [merge_forward]: 3.68e-06 [cell_reuse_recompute_pass]: 2.48e-06 [offload_activation]: 1.584e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.575e-05 [merge_recompute_call_nodes]: 9.60019e-07 [before_grad]: 9.31e-06 [set_forward_comm_id_for_comm_node_pass]: 3.87998e-06 [meta_fg_expand]: 2.40002e-06 [flash_sp_send_recv_attached]: 1.83997e-06 [receive_attached]: 1.95001e-06 [after_resolve]: 1.081e-05 [a_after_grad]: 9.41e-06 [renormalize]: 8.00064e-08 [add_forward_monad_depend]: 1.62999e-06 [auto_monad_grad]: 1.67001e-06 [auto_monad_eliminator]: 9.59e-06 [cse]: 2.829e-05 [a_3]: 3.905e-05 [py_interpret_to_execute_after_opt_a]: 1.513e-05 [slice_cell_reuse_recomputed_activation]: 2.93e-06 [rewriter_after_opt_a]: 4.333e-05 [convert_after_rewriter]: 7.8e-06 [order_py_execute_after_rewriter]: 6.14001e-06 [mutable_eliminate]: 0.00070035 [opt_b]: 0.00022199, [1] [Cycle 1]: 0.00021282, [7] [b_1]: 0.00012517 [b_2]: 8.13999e-06 [updatestate_depend_eliminate]: 9.44998e-06 [updatestate_assign_eliminate]: 2.79999e-06 [updatestate_loads_eliminate]: 2.51e-06 [renormalize]: 5.50004e-07 [cse]: 2.97e-05 [optimize_parallel_all_gather_comm]: 1.954e-05 [overlap_param_gather]: 3.78001e-06 [cconv]: 3.558e-05 [loop_unroll]: 0.00046095 [opt_after_cconv]: 0.00011865, [1] [Cycle 1]: 0.00011169, [7] [c_1]: 3.112e-05 [parameter_eliminate]: 5.00999e-06 [updatestate_depend_eliminate]: 6.78e-06 [updatestate_assign_eliminate]: 2.63e-06 [updatestate_loads_eliminate]: 2.34001e-06 [cse]: 2.385e-05 [renormalize]: 6.69999e-07 [remove_dup_value]: 4.811e-05 [tuple_transform]: 8.477e-05, [1] [Cycle 1]: 7.532e-05, [4] [d_1]: 4.747e-05 [none_parameter_eliminate]: 2.03002e-06 [renormalize]: 1.50001e-07 [switch_simplify]: 7.49002e-06 [partial_unused_args_eliminate]: 2.44001e-06 [add_recomputation]: 5.649e-05 [cse_after_recomputation]: 3.532e-05, [1] [Cycle 1]: 3.03e-05, [1] [cse]: 2.464e-05 [environ_conv]: 6.68e-06 [swap_dp_allreduce_reducescatter]: 5.47999e-06 [bias_add_comm_swap]: 3.57997e-06 [label_micro_interleaved_index]: 4.68999e-06 [label_fine_grained_interleaved_index]: 2.58e-06 [merge_cast_opt]: 1.19e-06 [slice_recompute_activation]: 2.18998e-06 [micro_interleaved_order_control]: 2.53e-06 [assign_add_opt]: 1.50001e-06 [ForceFp32Comm]: 8.50006e-07 [remove_cast_before_assign_add]: 1.14e-06 [full_micro_interleaved_order_control]: 2.34001e-06 [reorder_send_recv_between_fp_bp]: 2.93e-06 [comm_op_add_attrs]: 1.02998e-06 [add_comm_op_reuse_tag]: 1.03001e-06 [interleave_split_concat_branches]: 1.30001e-06 [interleave_parallel_branches]: 1.05001e-06 [overlap_opt_shard_in_pipeline]: 2.42001e-06 [overlap_opt_shard_grad_in_pipeline]: 2.08002e-06 [control_data_broadcast_order]: 1.45e-05 [grouped_pairwise_exchange_alltoall]: 1.66e-06 [offloading_packed_experts]: 4.05e-06 [overlap_recompute_and_grad_model_parallel]: 5.40999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.17999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.44e-06 [overlap_recompute_comm]: 2.32001e-06 [overlap_grad_ring_attention]: 4.17e-06 [overlap_grad_flash_sp]: 2.243e-05 [begin_end_overlap_inline]: 5.89993e-07 [split_matmul_comm_elemetwise]: 2.40002e-06 [split_layernorm_comm]: 2.43002e-06 [handle_group_info]: 1.13001e-06 [symbol_engine_optimizer]: 8.364e-05, [1] [Cycle 1]: 7.894e-05, [6] [build]: 4.02e-06 [elim_shapecalc]: 1.064e-05 [elim_not_effective]: 1.355e-05 [opt_reshape]: 7.45998e-06 [fold_const_symbol]: 1.019e-05 [renormalize]: 1.50001e-07 [detach_backward]: 2.26998e-06 [pipeline_parallel_scheduler]: 1.46002e-06 [auto_monad_reorder]: 2.033e-05 [get_jit_bprop_graph]: 2.17999e-06 [rewriter_after_jit_bprop_graph]: 5.32001e-06 [opt_after_jit_grad]: 0.00053649 [validate]: 4.944e-05 Sums bootstrap : 0.000747s : 0.62% type_inference : 0.115323s : 95.27% event_method : 0.000019s : 0.02% auto_monad : 0.000067s : 0.06% graph_reusing : 0.000006s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000017s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000003s : 0.00% pre_auto_parallel : 0.000036s : 0.03% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000024s : 0.02% optimize.rewriter_before_opt_a : 0.000054s : 0.04% optimize.opt_a.expand_dump_flag : 0.000005s : 0.00% optimize.opt_a.switch_simplify : 0.000033s : 0.03% optimize.opt_a.loop_unroll : 0.000019s : 0.02% optimize.opt_a.a_1 : 0.000495s : 0.41% optimize.opt_a.with_stream_mark : 0.000041s : 0.03% optimize.opt_a.recompute_prepare : 0.000016s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.01% 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 : 0.00% optimize.opt_a.a_2 : 0.000153s : 0.13% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.01% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.00% optimize.opt_a.shard_inline : 0.000013s : 0.01% optimize.opt_a.merge_send_recv : 0.000016s : 0.01% optimize.opt_a.auto_parallel : 0.000017s : 0.01% optimize.opt_a.parallel : 0.000040s : 0.03% optimize.opt_a.flash_sp : 0.000014s : 0.01% optimize.opt_a.merge_comm : 0.000008s : 0.01% optimize.opt_a.allreduce_fusion : 0.000008s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000029s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000017s : 0.01% optimize.opt_a.virtual_dataset : 0.000013s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.01% optimize.opt_a.virtual_output : 0.000012s : 0.01% optimize.opt_a.merge_forward : 0.000009s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.00% optimize.opt_a.offload_activation : 0.000028s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000031s : 0.03% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000020s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000009s : 0.01% optimize.opt_a.meta_fg_expand : 0.000005s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.00% optimize.opt_a.receive_attached : 0.000004s : 0.00% optimize.opt_a.after_resolve : 0.000024s : 0.02% optimize.opt_a.a_after_grad : 0.000019s : 0.02% optimize.opt_a.renormalize : 0.000770s : 0.64% optimize.opt_a.add_forward_monad_depend : 0.000009s : 0.01% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000027s : 0.02% optimize.opt_a.cse : 0.000074s : 0.06% optimize.opt_a.a_3 : 0.000170s : 0.14% optimize.py_interpret_to_execute_after_opt_a : 0.000015s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000043s : 0.04% optimize.convert_after_rewriter : 0.000008s : 0.01% optimize.order_py_execute_after_rewriter : 0.000006s : 0.01% optimize.mutable_eliminate : 0.000700s : 0.58% optimize.opt_b.b_1 : 0.000125s : 0.10% optimize.opt_b.b_2 : 0.000008s : 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 : 0.000030s : 0.02% optimize.optimize_parallel_all_gather_comm : 0.000020s : 0.02% optimize.overlap_param_gather : 0.000004s : 0.00% optimize.cconv : 0.000036s : 0.03% optimize.loop_unroll : 0.000461s : 0.38% optimize.opt_after_cconv.c_1 : 0.000031s : 0.03% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000007s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000002s : 0.00% optimize.opt_after_cconv.cse : 0.000024s : 0.02% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000048s : 0.04% optimize.tuple_transform.d_1 : 0.000047s : 0.04% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000056s : 0.05% optimize.cse_after_recomputation.cse : 0.000025s : 0.02% optimize.environ_conv : 0.000007s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000005s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000014s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000004s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000004s : 0.00% optimize.overlap_grad_flash_sp : 0.000022s : 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 : 0.000011s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000014s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 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 : 0.000005s : 0.00% opt_after_jit_grad : 0.000536s : 0.44% validate : 0.000049s : 0.04% Time group info: ------[substitution.] 0.000121 21 1.98% : 0.000002s : 2: substitution.elim_not_effective 1.20% : 0.000001s : 2: substitution.fold_const_symbol 5.45% : 0.000007s : 4: substitution.graph_param_transform 79.23% : 0.000096s : 1: substitution.inline 3.19% : 0.000004s : 4: substitution.j_node_and_user_rematch 4.36% : 0.000005s : 4: substitution.remove_not_recompute_node 4.59% : 0.000006s : 4: substitution.replace_old_param ------[type_inference.] 0.115247 2 99.46% : 0.114627s : 1: type_inference.infer 0.54% : 0.000620s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000094 1 100.00% : 0.000094s : 1: match.inline ------[predicate.] 0.000221 967 0.52% : 0.000001s : 9: predicate.accumulaten_eliminater 0.74% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.52% : 0.000001s : 8: predicate.addn_check_dump 0.50% : 0.000001s : 9: predicate.addn_zero_filter 0.48% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 1.49% : 0.000003s : 17: predicate.arithmetic_simplify 0.54% : 0.000001s : 9: predicate.cast_eliminate 0.60% : 0.000001s : 8: predicate.check_bprop_eliminate 0.52% : 0.000001s : 8: predicate.compare_switch_simplify 0.17% : 0.000000s : 4: predicate.const_output_eliminate 0.50% : 0.000001s : 8: predicate.depend_value_elim 0.53% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.69% : 0.000002s : 9: predicate.dict_get_item_eliminator 0.52% : 0.000001s : 9: predicate.dict_set_item_eliminator 0.75% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.21% : 0.000000s : 4: predicate.elim_not_effective 0.37% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 0.82% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.71% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.67% : 0.000001s : 13: predicate.environ_get_depend_swap 1.36% : 0.000003s : 21: predicate.environ_get_eliminate 0.82% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.52% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.26% : 0.000003s : 10: predicate.float_depend_g_call 0.43% : 0.000001s : 8: predicate.float_environ_get_switch 0.65% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.18% : 0.000000s : 4: predicate.fold_const_symbol 0.69% : 0.000002s : 8: predicate.get_grad_eliminate 0.16% : 0.000000s : 4: predicate.graph_param_transform 0.48% : 0.000001s : 8: predicate.incorporate_call 0.43% : 0.000001s : 8: predicate.incorporate_call_switch 4.61% : 0.000010s : 43: predicate.inline 0.93% : 0.000002s : 8: predicate.inline_without_move 0.29% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.78% : 0.000002s : 8: predicate.less_batch_normalization 1.23% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 1.44% : 0.000003s : 26: predicate.load_eliminater 0.87% : 0.000002s : 4: predicate.loop_unroll_after_grad 0.90% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.25% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.54% : 0.000001s : 8: predicate.merge_addn 0.50% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.52% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.48% : 0.000001s : 9: predicate.minmaximum_grad 1.11% : 0.000002s : 4: predicate.mutable_eliminate 0.36% : 0.000001s : 4: predicate.opt_reshape 0.42% : 0.000001s : 4: predicate.parallel_virtual_node 0.81% : 0.000002s : 10: predicate.partial_defer_inline 0.82% : 0.000002s : 13: predicate.partial_eliminate 0.50% : 0.000001s : 9: predicate.print_const_string_wrapper 0.61% : 0.000001s : 8: predicate.reduce_all_const_elim 0.72% : 0.000002s : 9: predicate.reduce_eliminate 1.65% : 0.000004s : 26: predicate.redundant_stop_gradient_eliminater 0.56% : 0.000001s : 8: predicate.remove_not_recompute_node 1.00% : 0.000002s : 17: predicate.replace_applicator 0.57% : 0.000001s : 8: predicate.replace_old_param 0.43% : 0.000001s : 4: predicate.reset_defer_inline 0.64% : 0.000001s : 9: predicate.reshape_eliminate 0.69% : 0.000002s : 8: predicate.row_tensor_add_zeros_like 0.36% : 0.000001s : 4: predicate.row_tensor_eliminate 0.90% : 0.000002s : 8: predicate.same_eliminate 0.41% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.05% : 0.000002s : 8: predicate.shard_identity_eliminate 0.70% : 0.000002s : 8: predicate.special_op_eliminate 0.73% : 0.000002s : 8: predicate.specialize_transform 0.91% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.63% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.29% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.63% : 0.000001s : 10: predicate.switch_defer_inline 1.27% : 0.000003s : 18: predicate.switch_layer_defer_inline 2.60% : 0.000006s : 35: predicate.switch_simplify 0.53% : 0.000001s : 9: predicate.tile_eliminate 0.61% : 0.000001s : 9: predicate.transpose_eliminate 1.17% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.11% : 0.000002s : 17: predicate.tuple_list_get_item_const_eliminator 1.16% : 0.000003s : 17: predicate.tuple_list_get_item_depend_reorder 2.33% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 30.07% : 0.000066s : 17: predicate.tuple_list_get_set_item_eliminator 1.72% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.19% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.32% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.07% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.33% : 0.000001s : 4: predicate.value_based_eliminate 0.64% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.58% : 0.000001s : 8: predicate.virtual_output_eliminate 0.26% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.37% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000312 4 7.12% : 0.000022s : 1: func_graph_cloner_run.FuncGraphClonerGraph 92.88% : 0.000290s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.139726 192 0.00% : 0.000003s : 1: ForceFp32Comm 3.80% : 0.005311s : 1: add_attr 3.79% : 0.005292s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.04% : 0.000062s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.05% : 0.000073s : 1: auto_monad 0.02% : 0.000030s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.58% : 0.000809s : 1: bootstrap 0.03% : 0.000040s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000018s : 1: control_data_broadcast_order 0.01% : 0.000011s : 1: convert_after_rewriter 0.03% : 0.000038s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000005s : 1: detach_backward 0.01% : 0.000010s : 1: environ_conv 0.02% : 0.000027s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.01% : 0.000009s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.01% : 0.000007s : 1: label_micro_interleaved_index 0.34% : 0.000472s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.51% : 0.000716s : 1: mutable_eliminate 0.01% : 0.000007s : 1: offloading_packed_experts 0.01% : 0.000016s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000019s : 1: opt.transform.mutable_eliminate 0.64% : 0.000900s : 78: opt.transform.opt_a 0.02% : 0.000029s : 1: opt.transform.opt_after_cconv 0.04% : 0.000061s : 1: opt.transform.opt_after_jit_grad 0.07% : 0.000104s : 28: opt.transform.opt_b 0.04% : 0.000053s : 2: opt.transform.opt_trans_graph 0.03% : 0.000038s : 4: opt.transform.symbol_engine_opt 1.96% : 0.002735s : 1: opt_a 0.09% : 0.000122s : 1: opt_after_cconv 0.39% : 0.000547s : 1: opt_after_jit_grad 0.16% : 0.000225s : 1: opt_b 3.67% : 0.005124s : 1: optimize 0.02% : 0.000023s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000010s : 1: order_py_execute_after_rewriter 0.02% : 0.000026s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000008s : 1: overlap_grad_ring_attention 0.01% : 0.000010s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.01% : 0.000007s : 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 0.00% : 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% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000004s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.03% : 0.000041s : 1: pre_auto_parallel 0.02% : 0.000029s : 1: py_interpret_to_execute 0.01% : 0.000019s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000008s : 1: remove_cast_before_assign_add 0.04% : 0.000052s : 1: remove_dup_value 0.32% : 0.000446s : 1: renormalize.infer 0.22% : 0.000314s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000008s : 1: rewriter_after_jit_bprop_graph 0.03% : 0.000048s : 1: rewriter_after_opt_a 0.04% : 0.000059s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000008s : 1: swap_dp_allreduce_reducescatter 0.06% : 0.000086s : 1: symbol_engine_optimizer 0.06% : 0.000088s : 1: tuple_transform 82.56% : 0.115359s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[dim_type0-False-batch_size_list1-shape1-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_fp16[dim_type0-False-batch_size_list1-shape1-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:39:13.976.396 [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 = 0.106253, [21] [bootstrap]: 0.00060819 [type_inference]: 0.0036275 [event_method]: 1.135e-05 [auto_monad]: 5.459e-05 [graph_reusing]: 5.39e-06 [inline]: 2.48e-06 [add_attr]: 0.0962764, [1] [add_attr_with_inline]: 0.0962572, [1] [Cycle 1]: 6.786e-05, [2] [tag_attr]: 1.771e-05 [meta_addattr_fg_expand]: 3.93999e-06 [parallel-infer-symbol]: 3.98999e-06 [pre_auto_parallel]: 3.842e-05 [insert-virtual-dataset]: 2.76e-06 [parallel-infer-symbol-second]: 7.89994e-07 [dataset_repeat_opt]: 2.34001e-06 [pipeline_split]: 2.17999e-06 [optimize]: 0.00486539, [53] [py_interpret_to_execute]: 2.621e-05 [rewriter_before_opt_a]: 5.168e-05 [opt_a]: 0.00246427, [2] [Cycle 1]: 0.00176652, [45] [expand_dump_flag]: 3.14001e-06 [switch_simplify]: 2.394e-05 [loop_unroll]: 1.241e-05 [a_1]: 0.00031298 [with_stream_mark]: 2.267e-05 [recompute_prepare]: 9.62999e-06 [updatestate_depend_eliminate]: 4.76997e-06 [updatestate_assign_eliminate]: 4.22998e-06 [updatestate_loads_eliminate]: 3.78001e-06 [parameter_eliminate]: 2.37999e-06 [a_2]: 8.109e-05 [accelerated_algorithm]: 7.55998e-06 [shard]: 2.27001e-06 [meta_shard_fg_expand]: 1.94999e-06 [shard_inline]: 6.21e-06 [merge_send_recv]: 1.037e-05 [auto_parallel]: 7.27002e-06 [parallel]: 2.954e-05 [flash_sp]: 1.002e-05 [merge_comm]: 3.97e-06 [allreduce_fusion]: 3.75998e-06 [matmul_add_comm_reduction]: 1.057e-05 [allreduce_slice_to_reducescatter]: 8.80013e-07 [virtual_shard_identity]: 8.68001e-06 [virtual_dataset]: 6.53998e-06 [get_grad_eliminate_]: 6.88e-06 [virtual_output]: 6.83e-06 [merge_forward]: 5.04e-06 [cell_reuse_recompute_pass]: 2.24999e-06 [offload_activation]: 1.064e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.394e-05 [merge_recompute_call_nodes]: 1.69e-06 [before_grad]: 1.039e-05 [set_forward_comm_id_for_comm_node_pass]: 3.98001e-06 [meta_fg_expand]: 3.08998e-06 [flash_sp_send_recv_attached]: 2.82002e-06 [receive_attached]: 2.07001e-06 [after_resolve]: 1.205e-05 [a_after_grad]: 9.89001e-06 [renormalize]: 0.00070762 [add_forward_monad_depend]: 7.16001e-06 [auto_monad_grad]: 2.60002e-06 [auto_monad_eliminator]: 1.721e-05 [cse]: 4.468e-05 [a_3]: 5.196e-05 [Cycle 2]: 0.00068445, [45] [expand_dump_flag]: 1.79e-06 [switch_simplify]: 7.98999e-06 [loop_unroll]: 6.16e-06 [a_1]: 0.00012577 [with_stream_mark]: 1.621e-05 [recompute_prepare]: 7.46001e-06 [updatestate_depend_eliminate]: 3.68999e-06 [updatestate_assign_eliminate]: 3.20002e-06 [updatestate_loads_eliminate]: 3.32997e-06 [parameter_eliminate]: 1.31002e-06 [a_2]: 6.966e-05 [accelerated_algorithm]: 6.65002e-06 [shard]: 1.55999e-06 [meta_shard_fg_expand]: 1.84998e-06 [shard_inline]: 5.91e-06 [merge_send_recv]: 6.81001e-06 [auto_parallel]: 7.61999e-06 [parallel]: 5.89999e-06 [flash_sp]: 3.54002e-06 [merge_comm]: 3.41999e-06 [allreduce_fusion]: 3.25e-06 [matmul_add_comm_reduction]: 9.94001e-06 [allreduce_slice_to_reducescatter]: 7.10017e-07 [virtual_shard_identity]: 8.65999e-06 [virtual_dataset]: 5.84999e-06 [get_grad_eliminate_]: 5.86e-06 [virtual_output]: 6.09001e-06 [merge_forward]: 3.58e-06 [cell_reuse_recompute_pass]: 2.25002e-06 [offload_activation]: 9.37999e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.436e-05 [merge_recompute_call_nodes]: 8.50006e-07 [before_grad]: 1.008e-05 [set_forward_comm_id_for_comm_node_pass]: 3.91999e-06 [meta_fg_expand]: 2.30002e-06 [flash_sp_send_recv_attached]: 1.04e-06 [receive_attached]: 1.54e-06 [after_resolve]: 1.062e-05 [a_after_grad]: 1.014e-05 [renormalize]: 6.00121e-08 [add_forward_monad_depend]: 2.41e-06 [auto_monad_grad]: 1.33002e-06 [auto_monad_eliminator]: 8.08001e-06 [cse]: 2.467e-05 [a_3]: 3.678e-05 [py_interpret_to_execute_after_opt_a]: 1.317e-05 [slice_cell_reuse_recomputed_activation]: 1.95001e-06 [rewriter_after_opt_a]: 4.432e-05 [convert_after_rewriter]: 8.17e-06 [order_py_execute_after_rewriter]: 5.94e-06 [mutable_eliminate]: 0.00072496 [opt_b]: 0.00025698, [1] [Cycle 1]: 0.00024857, [7] [b_1]: 0.00014964 [b_2]: 8.64e-06 [updatestate_depend_eliminate]: 9.69999e-06 [updatestate_assign_eliminate]: 2.99999e-06 [updatestate_loads_eliminate]: 2.72001e-06 [renormalize]: 5.90022e-07 [cse]: 3.811e-05 [optimize_parallel_all_gather_comm]: 1.969e-05 [overlap_param_gather]: 2.04e-06 [cconv]: 3.41e-05 [loop_unroll]: 0.00044716 [opt_after_cconv]: 0.00011809, [1] [Cycle 1]: 0.00011199, [7] [c_1]: 3.039e-05 [parameter_eliminate]: 4.72998e-06 [updatestate_depend_eliminate]: 5.86e-06 [updatestate_assign_eliminate]: 2.88e-06 [updatestate_loads_eliminate]: 2.46e-06 [cse]: 2.345e-05 [renormalize]: 5.50004e-07 [remove_dup_value]: 4.431e-05 [tuple_transform]: 8.434e-05, [1] [Cycle 1]: 7.27e-05, [4] [d_1]: 4.61e-05 [none_parameter_eliminate]: 1.55999e-06 [renormalize]: 2.10013e-07 [switch_simplify]: 6.83998e-06 [partial_unused_args_eliminate]: 2.13002e-06 [add_recomputation]: 5.484e-05 [cse_after_recomputation]: 2.88e-05, [1] [Cycle 1]: 2.433e-05, [1] [cse]: 1.792e-05 [environ_conv]: 7.08e-06 [swap_dp_allreduce_reducescatter]: 6.06e-06 [bias_add_comm_swap]: 2.96001e-06 [label_micro_interleaved_index]: 4.34002e-06 [label_fine_grained_interleaved_index]: 2.69001e-06 [merge_cast_opt]: 1.39e-06 [slice_recompute_activation]: 2.16e-06 [micro_interleaved_order_control]: 2.16998e-06 [assign_add_opt]: 1.33002e-06 [ForceFp32Comm]: 1.22999e-06 [remove_cast_before_assign_add]: 1.09e-06 [full_micro_interleaved_order_control]: 2.17999e-06 [reorder_send_recv_between_fp_bp]: 2.76999e-06 [comm_op_add_attrs]: 1.15999e-06 [add_comm_op_reuse_tag]: 1.02e-06 [interleave_split_concat_branches]: 1.26997e-06 [interleave_parallel_branches]: 1.09e-06 [overlap_opt_shard_in_pipeline]: 1.27e-06 [overlap_opt_shard_grad_in_pipeline]: 1.99e-06 [control_data_broadcast_order]: 1.385e-05 [grouped_pairwise_exchange_alltoall]: 1.64998e-06 [offloading_packed_experts]: 4.35e-06 [overlap_recompute_and_grad_model_parallel]: 5.25001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.20001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.62999e-06 [overlap_recompute_comm]: 2.47001e-06 [overlap_grad_ring_attention]: 3.76999e-06 [overlap_grad_flash_sp]: 2.285e-05 [begin_end_overlap_inline]: 5.19998e-07 [split_matmul_comm_elemetwise]: 2.71999e-06 [split_layernorm_comm]: 1.99e-06 [handle_group_info]: 1.02998e-06 [symbol_engine_optimizer]: 7.986e-05, [1] [Cycle 1]: 7.559e-05, [6] [build]: 4.06001e-06 [elim_shapecalc]: 1.042e-05 [elim_not_effective]: 1.431e-05 [opt_reshape]: 7.3e-06 [fold_const_symbol]: 9.81e-06 [renormalize]: 1.80007e-07 [detach_backward]: 2.35002e-06 [pipeline_parallel_scheduler]: 1.64e-06 [auto_monad_reorder]: 2.063e-05 [get_jit_bprop_graph]: 2.10002e-06 [rewriter_after_jit_bprop_graph]: 5.01002e-06 [opt_after_jit_grad]: 0.00048357 [validate]: 4.326e-05 Sums bootstrap : 0.000608s : 6.82% type_inference : 0.003628s : 40.68% event_method : 0.000011s : 0.13% auto_monad : 0.000055s : 0.61% graph_reusing : 0.000005s : 0.06% inline : 0.000002s : 0.03% add_attr.add_attr_with_inline.tag_attr : 0.000018s : 0.20% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.04% parallel-infer-symbol : 0.000004s : 0.04% pre_auto_parallel : 0.000038s : 0.43% 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 : 0.000026s : 0.29% optimize.rewriter_before_opt_a : 0.000052s : 0.58% optimize.opt_a.expand_dump_flag : 0.000005s : 0.06% optimize.opt_a.switch_simplify : 0.000032s : 0.36% optimize.opt_a.loop_unroll : 0.000019s : 0.21% optimize.opt_a.a_1 : 0.000439s : 4.92% optimize.opt_a.with_stream_mark : 0.000039s : 0.44% optimize.opt_a.recompute_prepare : 0.000017s : 0.19% 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 : 0.000004s : 0.04% optimize.opt_a.a_2 : 0.000151s : 1.69% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.16% optimize.opt_a.shard : 0.000004s : 0.04% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.04% optimize.opt_a.shard_inline : 0.000012s : 0.14% optimize.opt_a.merge_send_recv : 0.000017s : 0.19% optimize.opt_a.auto_parallel : 0.000015s : 0.17% optimize.opt_a.parallel : 0.000035s : 0.40% optimize.opt_a.flash_sp : 0.000014s : 0.15% optimize.opt_a.merge_comm : 0.000007s : 0.08% optimize.opt_a.allreduce_fusion : 0.000007s : 0.08% optimize.opt_a.matmul_add_comm_reduction : 0.000021s : 0.23% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.02% optimize.opt_a.virtual_shard_identity : 0.000017s : 0.19% optimize.opt_a.virtual_dataset : 0.000012s : 0.14% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.14% optimize.opt_a.virtual_output : 0.000013s : 0.14% optimize.opt_a.merge_forward : 0.000009s : 0.10% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.05% optimize.opt_a.offload_activation : 0.000020s : 0.22% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000028s : 0.32% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.03% optimize.opt_a.before_grad : 0.000020s : 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 : 0.04% optimize.opt_a.after_resolve : 0.000023s : 0.25% optimize.opt_a.a_after_grad : 0.000020s : 0.22% optimize.opt_a.renormalize : 0.000708s : 7.94% optimize.opt_a.add_forward_monad_depend : 0.000010s : 0.11% optimize.opt_a.auto_monad_grad : 0.000004s : 0.04% optimize.opt_a.auto_monad_eliminator : 0.000025s : 0.28% optimize.opt_a.cse : 0.000069s : 0.78% optimize.opt_a.a_3 : 0.000089s : 1.00% optimize.py_interpret_to_execute_after_opt_a : 0.000013s : 0.15% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.02% optimize.rewriter_after_opt_a : 0.000044s : 0.50% optimize.convert_after_rewriter : 0.000008s : 0.09% optimize.order_py_execute_after_rewriter : 0.000006s : 0.07% optimize.mutable_eliminate : 0.000725s : 8.13% optimize.opt_b.b_1 : 0.000150s : 1.68% optimize.opt_b.b_2 : 0.000009s : 0.10% optimize.opt_b.updatestate_depend_eliminate : 0.000010s : 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 : 0.000038s : 0.43% optimize.optimize_parallel_all_gather_comm : 0.000020s : 0.22% optimize.overlap_param_gather : 0.000002s : 0.02% optimize.cconv : 0.000034s : 0.38% optimize.loop_unroll : 0.000447s : 5.01% optimize.opt_after_cconv.c_1 : 0.000030s : 0.34% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.05% 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 : 0.000002s : 0.03% optimize.opt_after_cconv.cse : 0.000023s : 0.26% optimize.opt_after_cconv.renormalize : 0.000001s : 0.01% optimize.remove_dup_value : 0.000044s : 0.50% optimize.tuple_transform.d_1 : 0.000046s : 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 : 0.000055s : 0.61% optimize.cse_after_recomputation.cse : 0.000018s : 0.20% 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 : 0.05% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.03% optimize.merge_cast_opt : 0.000001s : 0.02% optimize.slice_recompute_activation : 0.000002s : 0.02% optimize.micro_interleaved_order_control : 0.000002s : 0.02% optimize.assign_add_opt : 0.000001s : 0.01% optimize.ForceFp32Comm : 0.000001s : 0.01% optimize.remove_cast_before_assign_add : 0.000001s : 0.01% optimize.full_micro_interleaved_order_control : 0.000002s : 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 : 0.01% 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 : 0.000004s : 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 : 0.000004s : 0.04% optimize.overlap_grad_flash_sp : 0.000023s : 0.26% 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 : 0.000004s : 0.05% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000010s : 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 : 0.000002s : 0.03% pipeline_parallel_scheduler : 0.000002s : 0.02% auto_monad_reorder : 0.000021s : 0.23% get_jit_bprop_graph : 0.000002s : 0.02% rewriter_after_jit_bprop_graph : 0.000005s : 0.06% opt_after_jit_grad : 0.000484s : 5.42% validate : 0.000043s : 0.49% Time group info: ------[substitution.] 0.000119 21 1.65% : 0.000002s : 2: substitution.elim_not_effective 1.19% : 0.000001s : 2: substitution.fold_const_symbol 4.96% : 0.000006s : 4: substitution.graph_param_transform 79.98% : 0.000095s : 1: substitution.inline 3.54% : 0.000004s : 4: substitution.j_node_and_user_rematch 4.47% : 0.000005s : 4: substitution.remove_not_recompute_node 4.21% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.003501 2 88.90% : 0.003112s : 1: type_inference.infer 11.10% : 0.000389s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000094 1 100.00% : 0.000094s : 1: match.inline ------[predicate.] 0.000159 967 0.86% : 0.000001s : 9: predicate.accumulaten_eliminater 0.97% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.63% : 0.000001s : 8: predicate.addn_check_dump 0.89% : 0.000001s : 9: predicate.addn_zero_filter 0.73% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 1.99% : 0.000003s : 17: predicate.arithmetic_simplify 0.98% : 0.000002s : 9: predicate.cast_eliminate 0.77% : 0.000001s : 8: predicate.check_bprop_eliminate 0.61% : 0.000001s : 8: predicate.compare_switch_simplify 0.23% : 0.000000s : 4: predicate.const_output_eliminate 0.77% : 0.000001s : 8: predicate.depend_value_elim 0.78% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 1.06% : 0.000002s : 9: predicate.dict_get_item_eliminator 0.76% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.33% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.27% : 0.000000s : 4: predicate.elim_not_effective 0.51% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.40% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.11% : 0.000002s : 13: predicate.environ_get_add_eliminate 1.04% : 0.000002s : 13: predicate.environ_get_depend_swap 1.78% : 0.000003s : 21: predicate.environ_get_eliminate 0.94% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.80% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.66% : 0.000003s : 10: predicate.float_depend_g_call 0.61% : 0.000001s : 8: predicate.float_environ_get_switch 0.92% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.24% : 0.000000s : 4: predicate.fold_const_symbol 0.88% : 0.000001s : 8: predicate.get_grad_eliminate 0.34% : 0.000001s : 4: predicate.graph_param_transform 0.79% : 0.000001s : 8: predicate.incorporate_call 0.56% : 0.000001s : 8: predicate.incorporate_call_switch 6.25% : 0.000010s : 43: predicate.inline 1.39% : 0.000002s : 8: predicate.inline_without_move 0.44% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.22% : 0.000002s : 8: predicate.less_batch_normalization 1.83% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.04% : 0.000003s : 26: predicate.load_eliminater 1.45% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.24% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.80% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.66% : 0.000001s : 8: predicate.merge_addn 0.80% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.75% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.66% : 0.000001s : 9: predicate.minmaximum_grad 1.54% : 0.000002s : 4: predicate.mutable_eliminate 0.48% : 0.000001s : 4: predicate.opt_reshape 0.51% : 0.000001s : 4: predicate.parallel_virtual_node 1.39% : 0.000002s : 10: predicate.partial_defer_inline 1.08% : 0.000002s : 13: predicate.partial_eliminate 0.80% : 0.000001s : 9: predicate.print_const_string_wrapper 0.78% : 0.000001s : 8: predicate.reduce_all_const_elim 1.46% : 0.000002s : 9: predicate.reduce_eliminate 1.95% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.80% : 0.000001s : 8: predicate.remove_not_recompute_node 1.20% : 0.000002s : 17: predicate.replace_applicator 0.83% : 0.000001s : 8: predicate.replace_old_param 0.45% : 0.000001s : 4: predicate.reset_defer_inline 0.89% : 0.000001s : 9: predicate.reshape_eliminate 0.87% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.60% : 0.000001s : 4: predicate.row_tensor_eliminate 1.08% : 0.000002s : 8: predicate.same_eliminate 0.68% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.28% : 0.000002s : 8: predicate.shard_identity_eliminate 0.85% : 0.000001s : 8: predicate.special_op_eliminate 0.81% : 0.000001s : 8: predicate.specialize_transform 1.42% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.86% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.39% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.90% : 0.000001s : 10: predicate.switch_defer_inline 1.59% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.81% : 0.000006s : 35: predicate.switch_simplify 0.85% : 0.000001s : 9: predicate.tile_eliminate 0.97% : 0.000002s : 9: predicate.transpose_eliminate 1.56% : 0.000002s : 17: predicate.tuple_list_convert_item_index_to_positive 1.57% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.37% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.11% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.83% : 0.000003s : 17: predicate.tuple_list_get_set_item_eliminator 2.24% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.74% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 2.01% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.70% : 0.000004s : 34: predicate.updatestate_useless_node_eliminater 0.52% : 0.000001s : 4: predicate.value_based_eliminate 0.89% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.82% : 0.000001s : 8: predicate.virtual_output_eliminate 0.33% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.46% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000226 4 10.03% : 0.000023s : 1: func_graph_cloner_run.FuncGraphClonerGraph 89.97% : 0.000203s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.209064 192 0.00% : 0.000004s : 1: ForceFp32Comm 46.05% : 0.096284s : 1: add_attr 46.04% : 0.096262s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.03% : 0.000060s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.03% : 0.000060s : 1: auto_monad 0.01% : 0.000028s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.32% : 0.000664s : 1: bootstrap 0.02% : 0.000037s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000017s : 1: control_data_broadcast_order 0.01% : 0.000012s : 1: convert_after_rewriter 0.02% : 0.000032s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000010s : 1: environ_conv 0.01% : 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% : 0.000009s : 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% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000007s : 1: label_micro_interleaved_index 0.22% : 0.000457s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.35% : 0.000741s : 1: mutable_eliminate 0.00% : 0.000007s : 1: offloading_packed_experts 0.01% : 0.000016s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000021s : 1: opt.transform.mutable_eliminate 0.40% : 0.000833s : 78: opt.transform.opt_a 0.01% : 0.000029s : 1: opt.transform.opt_after_cconv 0.01% : 0.000025s : 1: opt.transform.opt_after_jit_grad 0.05% : 0.000104s : 28: opt.transform.opt_b 0.02% : 0.000051s : 2: opt.transform.opt_trans_graph 0.02% : 0.000038s : 4: opt.transform.symbol_engine_opt 1.18% : 0.002468s : 1: opt_a 0.06% : 0.000121s : 1: opt_after_cconv 0.24% : 0.000494s : 1: opt_after_jit_grad 0.12% : 0.000261s : 1: opt_b 2.33% : 0.004871s : 1: optimize 0.01% : 0.000023s : 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% : 0.000009s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000004s : 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% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.02% : 0.000043s : 1: pre_auto_parallel 0.01% : 0.000030s : 1: py_interpret_to_execute 0.01% : 0.000017s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000007s : 1: remove_cast_before_assign_add 0.02% : 0.000048s : 1: remove_dup_value 0.20% : 0.000427s : 1: renormalize.infer 0.13% : 0.000272s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000008s : 1: rewriter_after_jit_bprop_graph 0.02% : 0.000049s : 1: rewriter_after_opt_a 0.03% : 0.000056s : 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 0.04% : 0.000083s : 1: symbol_engine_optimizer 0.04% : 0.000088s : 1: tuple_transform 1.74% : 0.003647s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[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_fp16[dim_type0-False-batch_size_list2-shape0-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:39:16.543.377 [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 = 0.0760789, [21] [bootstrap]: 0.00059571 [type_inference]: 0.0260331 [event_method]: 1.668e-05 [auto_monad]: 7.095e-05 [graph_reusing]: 5.88002e-06 [inline]: 3.3e-06 [add_attr]: 0.00478773, [1] [add_attr_with_inline]: 0.00477202, [1] [Cycle 1]: 7.082e-05, [2] [tag_attr]: 1.856e-05 [meta_addattr_fg_expand]: 3.58e-06 [parallel-infer-symbol]: 4.16001e-06 [pre_auto_parallel]: 3.934e-05 [insert-virtual-dataset]: 2.79001e-06 [parallel-infer-symbol-second]: 8.39995e-07 [dataset_repeat_opt]: 1.99e-06 [pipeline_split]: 1.72999e-06 [optimize]: 0.0434271, [53] [py_interpret_to_execute]: 2.77e-05 [rewriter_before_opt_a]: 5.365e-05 [opt_a]: 0.0025153, [2] [Cycle 1]: 0.00178327, [45] [expand_dump_flag]: 2.88e-06 [switch_simplify]: 2.387e-05 [loop_unroll]: 1.154e-05 [a_1]: 0.00030564 [with_stream_mark]: 2.433e-05 [recompute_prepare]: 8.95999e-06 [updatestate_depend_eliminate]: 4.33999e-06 [updatestate_assign_eliminate]: 3.41001e-06 [updatestate_loads_eliminate]: 3.58e-06 [parameter_eliminate]: 2.36e-06 [a_2]: 8.155e-05 [accelerated_algorithm]: 7.36001e-06 [shard]: 2.61e-06 [meta_shard_fg_expand]: 2.11e-06 [shard_inline]: 6.50002e-06 [merge_send_recv]: 9.56e-06 [auto_parallel]: 8.07e-06 [parallel]: 3.166e-05 [flash_sp]: 9.86e-06 [merge_comm]: 4.03999e-06 [allreduce_fusion]: 3.56001e-06 [matmul_add_comm_reduction]: 1.082e-05 [allreduce_slice_to_reducescatter]: 6.30011e-07 [virtual_shard_identity]: 7.72998e-06 [virtual_dataset]: 6.41e-06 [get_grad_eliminate_]: 6.05002e-06 [virtual_output]: 6.85998e-06 [merge_forward]: 4.68999e-06 [cell_reuse_recompute_pass]: 1.62001e-06 [offload_activation]: 1.128e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.468e-05 [merge_recompute_call_nodes]: 1.57999e-06 [before_grad]: 1.082e-05 [set_forward_comm_id_for_comm_node_pass]: 3.68e-06 [meta_fg_expand]: 2.52001e-06 [flash_sp_send_recv_attached]: 2.94001e-06 [receive_attached]: 2.34001e-06 [after_resolve]: 1.298e-05 [a_after_grad]: 1.06e-05 [renormalize]: 0.00071916 [add_forward_monad_depend]: 6.68e-06 [auto_monad_grad]: 2.69999e-06 [auto_monad_eliminator]: 1.794e-05 [cse]: 4.287e-05 [a_3]: 4.866e-05 [Cycle 2]: 0.00072014, [45] [expand_dump_flag]: 2.54999e-06 [switch_simplify]: 8.42e-06 [loop_unroll]: 5.94e-06 [a_1]: 0.00012082 [with_stream_mark]: 1.522e-05 [recompute_prepare]: 6.15002e-06 [updatestate_depend_eliminate]: 3.73001e-06 [updatestate_assign_eliminate]: 3.23e-06 [updatestate_loads_eliminate]: 3.36999e-06 [parameter_eliminate]: 2.08998e-06 [a_2]: 7.453e-05 [accelerated_algorithm]: 6.66999e-06 [shard]: 1.05001e-06 [meta_shard_fg_expand]: 2.30002e-06 [shard_inline]: 6.45002e-06 [merge_send_recv]: 7.46999e-06 [auto_parallel]: 7.79997e-06 [parallel]: 6.07999e-06 [flash_sp]: 3.86999e-06 [merge_comm]: 3.93001e-06 [allreduce_fusion]: 3.53e-06 [matmul_add_comm_reduction]: 1.137e-05 [allreduce_slice_to_reducescatter]: 2.89991e-07 [virtual_shard_identity]: 7.73001e-06 [virtual_dataset]: 6.19001e-06 [get_grad_eliminate_]: 5.97999e-06 [virtual_output]: 5.82001e-06 [merge_forward]: 3.42002e-06 [cell_reuse_recompute_pass]: 2.04e-06 [offload_activation]: 9.71e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.403e-05 [merge_recompute_call_nodes]: 1.38002e-06 [before_grad]: 1.047e-05 [set_forward_comm_id_for_comm_node_pass]: 3.63999e-06 [meta_fg_expand]: 2.66e-06 [flash_sp_send_recv_attached]: 1.59998e-06 [receive_attached]: 2.18002e-06 [after_resolve]: 1.188e-05 [a_after_grad]: 9.50001e-06 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 1.64e-06 [auto_monad_grad]: 1.85001e-06 [auto_monad_eliminator]: 9.19998e-06 [cse]: 2.952e-05 [a_3]: 3.711e-05 [py_interpret_to_execute_after_opt_a]: 1.544e-05 [slice_cell_reuse_recomputed_activation]: 2.18002e-06 [rewriter_after_opt_a]: 4.781e-05 [convert_after_rewriter]: 7.66999e-06 [order_py_execute_after_rewriter]: 6.58e-06 [mutable_eliminate]: 0.00067436 [opt_b]: 0.00024196, [1] [Cycle 1]: 0.00023212, [7] [b_1]: 0.00012382 [b_2]: 8.90001e-06 [updatestate_depend_eliminate]: 1.211e-05 [updatestate_assign_eliminate]: 3.58999e-06 [updatestate_loads_eliminate]: 2.83e-06 [renormalize]: 1.14e-06 [cse]: 4.192e-05 [optimize_parallel_all_gather_comm]: 2.296e-05 [overlap_param_gather]: 2.99001e-06 [cconv]: 3.868e-05 [loop_unroll]: 0.00048445 [opt_after_cconv]: 0.0385131, [1] [Cycle 1]: 0.0385036, [7] [c_1]: 0.0383179 [parameter_eliminate]: 8.55999e-06 [updatestate_depend_eliminate]: 1.608e-05 [updatestate_assign_eliminate]: 3.95998e-06 [updatestate_loads_eliminate]: 3.93001e-06 [cse]: 5.808e-05 [renormalize]: 5.39992e-07 [remove_dup_value]: 5.447e-05 [tuple_transform]: 0.00011594, [1] [Cycle 1]: 9.58e-05, [4] [d_1]: 6.555e-05 [none_parameter_eliminate]: 2.20002e-06 [renormalize]: 1.69995e-07 [switch_simplify]: 7.93001e-06 [partial_unused_args_eliminate]: 2.41e-06 [add_recomputation]: 6.948e-05 [cse_after_recomputation]: 3.179e-05, [1] [Cycle 1]: 2.65e-05, [1] [cse]: 2.012e-05 [environ_conv]: 8.43999e-06 [swap_dp_allreduce_reducescatter]: 6.26e-06 [bias_add_comm_swap]: 3.63e-06 [label_micro_interleaved_index]: 9.65002e-06 [label_fine_grained_interleaved_index]: 2.95002e-06 [merge_cast_opt]: 2.01998e-06 [slice_recompute_activation]: 2.69001e-06 [micro_interleaved_order_control]: 2.51998e-06 [assign_add_opt]: 1.72999e-06 [ForceFp32Comm]: 9.09989e-07 [remove_cast_before_assign_add]: 1.39e-06 [full_micro_interleaved_order_control]: 2.61e-06 [reorder_send_recv_between_fp_bp]: 3.09999e-06 [comm_op_add_attrs]: 1.19e-06 [add_comm_op_reuse_tag]: 1.30999e-06 [interleave_split_concat_branches]: 1.30999e-06 [interleave_parallel_branches]: 1.15999e-06 [overlap_opt_shard_in_pipeline]: 2.59999e-06 [overlap_opt_shard_grad_in_pipeline]: 1.96e-06 [control_data_broadcast_order]: 1.798e-05 [grouped_pairwise_exchange_alltoall]: 2.17001e-06 [offloading_packed_experts]: 4.77e-06 [overlap_recompute_and_grad_model_parallel]: 5.35999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.29e-06 [overlap_recompute_allgather_and_fa_grad]: 1.69e-06 [overlap_recompute_comm]: 2.45002e-06 [overlap_grad_ring_attention]: 5.04e-06 [overlap_grad_flash_sp]: 2.777e-05 [begin_end_overlap_inline]: 5.59987e-07 [split_matmul_comm_elemetwise]: 2.84999e-06 [split_layernorm_comm]: 1.92999e-06 [handle_group_info]: 1.12e-06 [symbol_engine_optimizer]: 9.054e-05, [1] [Cycle 1]: 8.558e-05, [6] [build]: 4.20999e-06 [elim_shapecalc]: 1.319e-05 [elim_not_effective]: 1.586e-05 [opt_reshape]: 8.43999e-06 [fold_const_symbol]: 1.003e-05 [renormalize]: 3.29979e-07 [detach_backward]: 3.26999e-06 [pipeline_parallel_scheduler]: 1.57999e-06 [auto_monad_reorder]: 2.164e-05 [get_jit_bprop_graph]: 2.34001e-06 [rewriter_after_jit_bprop_graph]: 6.73998e-06 [opt_after_jit_grad]: 0.00079244 [validate]: 5.538e-05 Sums bootstrap : 0.000596s : 0.85% type_inference : 0.026033s : 37.15% event_method : 0.000017s : 0.02% auto_monad : 0.000071s : 0.10% graph_reusing : 0.000006s : 0.01% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000019s : 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 : 0.000039s : 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 : 0.000028s : 0.04% optimize.rewriter_before_opt_a : 0.000054s : 0.08% optimize.opt_a.expand_dump_flag : 0.000005s : 0.01% optimize.opt_a.switch_simplify : 0.000032s : 0.05% optimize.opt_a.loop_unroll : 0.000017s : 0.02% optimize.opt_a.a_1 : 0.000426s : 0.61% optimize.opt_a.with_stream_mark : 0.000040s : 0.06% optimize.opt_a.recompute_prepare : 0.000015s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.01% optimize.opt_a.a_2 : 0.000156s : 0.22% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.02% optimize.opt_a.shard : 0.000004s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.01% optimize.opt_a.shard_inline : 0.000013s : 0.02% optimize.opt_a.merge_send_recv : 0.000017s : 0.02% optimize.opt_a.auto_parallel : 0.000016s : 0.02% optimize.opt_a.parallel : 0.000038s : 0.05% optimize.opt_a.flash_sp : 0.000014s : 0.02% optimize.opt_a.merge_comm : 0.000008s : 0.01% optimize.opt_a.allreduce_fusion : 0.000007s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000022s : 0.03% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000015s : 0.02% optimize.opt_a.virtual_dataset : 0.000013s : 0.02% optimize.opt_a.get_grad_eliminate_ : 0.000012s : 0.02% optimize.opt_a.virtual_output : 0.000013s : 0.02% optimize.opt_a.merge_forward : 0.000008s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.01% optimize.opt_a.offload_activation : 0.000021s : 0.03% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000029s : 0.04% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000021s : 0.03% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000007s : 0.01% 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 : 0.000005s : 0.01% optimize.opt_a.after_resolve : 0.000025s : 0.04% optimize.opt_a.a_after_grad : 0.000020s : 0.03% optimize.opt_a.renormalize : 0.000719s : 1.03% optimize.opt_a.add_forward_monad_depend : 0.000008s : 0.01% optimize.opt_a.auto_monad_grad : 0.000005s : 0.01% optimize.opt_a.auto_monad_eliminator : 0.000027s : 0.04% optimize.opt_a.cse : 0.000072s : 0.10% optimize.opt_a.a_3 : 0.000086s : 0.12% optimize.py_interpret_to_execute_after_opt_a : 0.000015s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000048s : 0.07% optimize.convert_after_rewriter : 0.000008s : 0.01% optimize.order_py_execute_after_rewriter : 0.000007s : 0.01% optimize.mutable_eliminate : 0.000674s : 0.96% optimize.opt_b.b_1 : 0.000124s : 0.18% optimize.opt_b.b_2 : 0.000009s : 0.01% 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 : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000042s : 0.06% optimize.optimize_parallel_all_gather_comm : 0.000023s : 0.03% optimize.overlap_param_gather : 0.000003s : 0.00% optimize.cconv : 0.000039s : 0.06% optimize.loop_unroll : 0.000484s : 0.69% optimize.opt_after_cconv.c_1 : 0.038318s : 54.68% optimize.opt_after_cconv.parameter_eliminate : 0.000009s : 0.01% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000016s : 0.02% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000004s : 0.01% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000004s : 0.01% optimize.opt_after_cconv.cse : 0.000058s : 0.08% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000054s : 0.08% optimize.tuple_transform.d_1 : 0.000066s : 0.09% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000008s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000069s : 0.10% optimize.cse_after_recomputation.cse : 0.000020s : 0.03% optimize.environ_conv : 0.000008s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.01% optimize.bias_add_comm_swap : 0.000004s : 0.01% optimize.label_micro_interleaved_index : 0.000010s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000003s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000018s : 0.03% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000005s : 0.01% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 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 : 0.000028s : 0.04% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000004s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000013s : 0.02% optimize.symbol_engine_optimizer.elim_not_effective : 0.000016s : 0.02% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000022s : 0.03% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.01% opt_after_jit_grad : 0.000792s : 1.13% validate : 0.000055s : 0.08% Time group info: ------[substitution.] 0.000121 21 1.87% : 0.000002s : 2: substitution.elim_not_effective 1.20% : 0.000001s : 2: substitution.fold_const_symbol 6.27% : 0.000008s : 4: substitution.graph_param_transform 79.12% : 0.000095s : 1: substitution.inline 3.31% : 0.000004s : 4: substitution.j_node_and_user_rematch 4.22% : 0.000005s : 4: substitution.remove_not_recompute_node 4.00% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.025951 2 97.67% : 0.025346s : 1: type_inference.infer 2.33% : 0.000605s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000094 1 100.00% : 0.000094s : 1: match.inline ------[predicate.] 0.038390 967 0.00% : 0.000001s : 9: predicate.accumulaten_eliminater 0.01% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.00% : 0.000001s : 8: predicate.addn_check_dump 0.00% : 0.000001s : 9: predicate.addn_zero_filter 0.00% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 0.01% : 0.000003s : 17: predicate.arithmetic_simplify 0.00% : 0.000001s : 9: predicate.cast_eliminate 0.00% : 0.000001s : 8: predicate.check_bprop_eliminate 0.00% : 0.000001s : 8: predicate.compare_switch_simplify 0.00% : 0.000000s : 4: predicate.const_output_eliminate 0.00% : 0.000001s : 8: predicate.depend_value_elim 0.00% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.00% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.00% : 0.000001s : 9: predicate.dict_set_item_eliminator 0.01% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.00% : 0.000001s : 4: predicate.elim_not_effective 0.00% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 0.00% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.00% : 0.000001s : 13: predicate.environ_get_add_eliminate 0.00% : 0.000002s : 13: predicate.environ_get_depend_swap 0.01% : 0.000003s : 21: predicate.environ_get_eliminate 0.00% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.00% : 0.000001s : 10: predicate.exchange_switch_depend_value 0.01% : 0.000003s : 10: predicate.float_depend_g_call 0.00% : 0.000001s : 8: predicate.float_environ_get_switch 0.00% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.00% : 0.000000s : 4: predicate.fold_const_symbol 0.00% : 0.000001s : 8: predicate.get_grad_eliminate 0.00% : 0.000001s : 4: predicate.graph_param_transform 0.00% : 0.000001s : 8: predicate.incorporate_call 0.00% : 0.000001s : 8: predicate.incorporate_call_switch 99.60% : 0.038236s : 43: predicate.inline 0.00% : 0.000002s : 8: predicate.inline_without_move 0.00% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.00% : 0.000002s : 8: predicate.less_batch_normalization 0.01% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 0.01% : 0.000004s : 26: predicate.load_eliminater 0.01% : 0.000002s : 4: predicate.loop_unroll_after_grad 0.00% : 0.000002s : 13: predicate.loop_unroll_before_grad 0.01% : 0.000004s : 17: predicate.make_slice_get_slice_eliminator 0.00% : 0.000001s : 8: predicate.merge_addn 0.00% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.00% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.00% : 0.000001s : 9: predicate.minmaximum_grad 0.01% : 0.000003s : 4: predicate.mutable_eliminate 0.00% : 0.000001s : 4: predicate.opt_reshape 0.00% : 0.000001s : 4: predicate.parallel_virtual_node 0.00% : 0.000002s : 10: predicate.partial_defer_inline 0.00% : 0.000002s : 13: predicate.partial_eliminate 0.00% : 0.000001s : 9: predicate.print_const_string_wrapper 0.00% : 0.000001s : 8: predicate.reduce_all_const_elim 0.00% : 0.000001s : 9: predicate.reduce_eliminate 0.01% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.00% : 0.000001s : 8: predicate.remove_not_recompute_node 0.00% : 0.000002s : 17: predicate.replace_applicator 0.00% : 0.000001s : 8: predicate.replace_old_param 0.00% : 0.000001s : 4: predicate.reset_defer_inline 0.00% : 0.000001s : 9: predicate.reshape_eliminate 0.00% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.00% : 0.000001s : 4: predicate.row_tensor_eliminate 0.00% : 0.000002s : 8: predicate.same_eliminate 0.00% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.00% : 0.000002s : 8: predicate.shard_identity_eliminate 0.00% : 0.000001s : 8: predicate.special_op_eliminate 0.00% : 0.000001s : 8: predicate.specialize_transform 0.00% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.00% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.01% : 0.000003s : 4: predicate.switch_call_monad_eliminater 0.00% : 0.000001s : 10: predicate.switch_defer_inline 0.01% : 0.000003s : 18: predicate.switch_layer_defer_inline 0.01% : 0.000006s : 35: predicate.switch_simplify 0.00% : 0.000001s : 9: predicate.tile_eliminate 0.00% : 0.000001s : 9: predicate.transpose_eliminate 0.01% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 0.01% : 0.000002s : 17: predicate.tuple_list_get_item_const_eliminator 0.01% : 0.000003s : 17: predicate.tuple_list_get_item_depend_reorder 0.01% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 0.01% : 0.000003s : 17: predicate.tuple_list_get_set_item_eliminator 0.01% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 0.01% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 0.01% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 0.02% : 0.000009s : 34: predicate.updatestate_useless_node_eliminater 0.00% : 0.000001s : 4: predicate.value_based_eliminate 0.00% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.00% : 0.000002s : 8: predicate.virtual_output_eliminate 0.00% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.00% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000307 4 7.97% : 0.000024s : 1: func_graph_cloner_run.FuncGraphClonerGraph 92.03% : 0.000282s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.164264 192 0.00% : 0.000004s : 1: ForceFp32Comm 2.92% : 0.004794s : 1: add_attr 2.91% : 0.004776s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.05% : 0.000074s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.05% : 0.000077s : 1: auto_monad 0.02% : 0.000028s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.39% : 0.000635s : 1: bootstrap 0.03% : 0.000042s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000022s : 1: control_data_broadcast_order 0.01% : 0.000012s : 1: convert_after_rewriter 0.02% : 0.000035s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.01% : 0.000012s : 1: environ_conv 0.02% : 0.000025s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 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% : 0.000006s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.01% : 0.000013s : 1: label_micro_interleaved_index 0.30% : 0.000495s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.42% : 0.000691s : 1: mutable_eliminate 0.00% : 0.000008s : 1: offloading_packed_experts 0.01% : 0.000018s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000022s : 1: opt.transform.mutable_eliminate 0.50% : 0.000824s : 78: opt.transform.opt_a 23.32% : 0.038314s : 1: opt.transform.opt_after_cconv 0.02% : 0.000030s : 1: opt.transform.opt_after_jit_grad 0.06% : 0.000104s : 28: opt.transform.opt_b 0.04% : 0.000069s : 2: opt.transform.opt_trans_graph 0.03% : 0.000043s : 4: opt.transform.symbol_engine_opt 1.53% : 0.002518s : 1: opt_a 23.45% : 0.038519s : 1: opt_after_cconv 0.49% : 0.000808s : 1: opt_after_jit_grad 0.15% : 0.000246s : 1: opt_b 26.44% : 0.043434s : 1: optimize 0.02% : 0.000027s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000011s : 1: order_py_execute_after_rewriter 0.02% : 0.000032s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000012s : 1: overlap_grad_ring_attention 0.01% : 0.000010s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000006s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.03% : 0.000044s : 1: pre_auto_parallel 0.02% : 0.000032s : 1: py_interpret_to_execute 0.01% : 0.000019s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000008s : 1: remove_cast_before_assign_add 0.04% : 0.000061s : 1: remove_dup_value 0.27% : 0.000440s : 1: renormalize.infer 0.16% : 0.000270s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.03% : 0.000054s : 1: rewriter_after_opt_a 0.04% : 0.000059s : 1: rewriter_before_opt_a 0.00% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.01% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.06% : 0.000093s : 1: symbol_engine_optimizer 0.07% : 0.000119s : 1: tuple_transform 15.87% : 0.026070s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[dim_type0-False-batch_size_list2-shape1-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_fp16[dim_type0-False-batch_size_list2-shape1-ms_dtype0],max_mem:8.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:39:25.449.841 [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 = 0.131415, [21] [bootstrap]: 0.00062683 [type_inference]: 0.118395 [event_method]: 1.464e-05 [auto_monad]: 5.636e-05 [graph_reusing]: 5.19e-06 [inline]: 3.28e-06 [add_attr]: 0.00443242, [1] [add_attr_with_inline]: 0.00441751, [1] [Cycle 1]: 6.55e-05, [2] [tag_attr]: 1.675e-05 [meta_addattr_fg_expand]: 3.26999e-06 [parallel-infer-symbol]: 4.25e-06 [pre_auto_parallel]: 3.443e-05 [insert-virtual-dataset]: 2.43998e-06 [parallel-infer-symbol-second]: 8.59989e-07 [dataset_repeat_opt]: 2.16e-06 [pipeline_split]: 1.59e-06 [optimize]: 0.00706553, [53] [py_interpret_to_execute]: 2.052e-05 [rewriter_before_opt_a]: 4.789e-05 [opt_a]: 0.00467624, [2] [Cycle 1]: 0.00394201, [45] [expand_dump_flag]: 2.37999e-06 [switch_simplify]: 2.489e-05 [loop_unroll]: 1.099e-05 [a_1]: 0.0002918 [with_stream_mark]: 2.29e-05 [recompute_prepare]: 9.24e-06 [updatestate_depend_eliminate]: 4.53001e-06 [updatestate_assign_eliminate]: 4.12e-06 [updatestate_loads_eliminate]: 3.95e-06 [parameter_eliminate]: 1.97999e-06 [a_2]: 8.029e-05 [accelerated_algorithm]: 7.11999e-06 [shard]: 2.26e-06 [meta_shard_fg_expand]: 1.81e-06 [shard_inline]: 6.71999e-06 [merge_send_recv]: 8.83001e-06 [auto_parallel]: 7.27002e-06 [parallel]: 2.927e-05 [flash_sp]: 1.136e-05 [merge_comm]: 3.75e-06 [allreduce_fusion]: 3.43999e-06 [matmul_add_comm_reduction]: 1.017e-05 [allreduce_slice_to_reducescatter]: 7.39994e-07 [virtual_shard_identity]: 8.85001e-06 [virtual_dataset]: 6.49999e-06 [get_grad_eliminate_]: 6.31998e-06 [virtual_output]: 6.16e-06 [merge_forward]: 4.50001e-06 [cell_reuse_recompute_pass]: 1.28002e-06 [offload_activation]: 1.138e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.481e-05 [merge_recompute_call_nodes]: 1.50001e-06 [before_grad]: 1.104e-05 [set_forward_comm_id_for_comm_node_pass]: 3.66999e-06 [meta_fg_expand]: 2.64999e-06 [flash_sp_send_recv_attached]: 2.78e-06 [receive_attached]: 2.16e-06 [after_resolve]: 1.208e-05 [a_after_grad]: 9.17999e-06 [renormalize]: 0.00287284 [add_forward_monad_depend]: 1.176e-05 [auto_monad_grad]: 2.55997e-06 [auto_monad_eliminator]: 2.563e-05 [cse]: 4.243e-05 [a_3]: 6.357e-05 [Cycle 2]: 0.00071866, [45] [expand_dump_flag]: 2.71999e-06 [switch_simplify]: 1.017e-05 [loop_unroll]: 6.73e-06 [a_1]: 0.00013369 [with_stream_mark]: 2.128e-05 [recompute_prepare]: 6.76999e-06 [updatestate_depend_eliminate]: 4.33999e-06 [updatestate_assign_eliminate]: 3.30003e-06 [updatestate_loads_eliminate]: 3.41001e-06 [parameter_eliminate]: 2.49001e-06 [a_2]: 7.255e-05 [accelerated_algorithm]: 6.69999e-06 [shard]: 2.37999e-06 [meta_shard_fg_expand]: 2.50002e-06 [shard_inline]: 6.04001e-06 [merge_send_recv]: 8.64e-06 [auto_parallel]: 1.001e-05 [parallel]: 1.025e-05 [flash_sp]: 4.28999e-06 [merge_comm]: 4.03999e-06 [allreduce_fusion]: 3.45003e-06 [matmul_add_comm_reduction]: 1.055e-05 [allreduce_slice_to_reducescatter]: 9.29984e-07 [virtual_shard_identity]: 7.83001e-06 [virtual_dataset]: 6.66e-06 [get_grad_eliminate_]: 5.79999e-06 [virtual_output]: 6.43e-06 [merge_forward]: 4.62e-06 [cell_reuse_recompute_pass]: 3.3e-06 [offload_activation]: 1.225e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.893e-05 [merge_recompute_call_nodes]: 1.47001e-06 [before_grad]: 1.069e-05 [set_forward_comm_id_for_comm_node_pass]: 4.45e-06 [meta_fg_expand]: 2.69999e-06 [flash_sp_send_recv_attached]: 2.22001e-06 [receive_attached]: 2.44999e-06 [after_resolve]: 1.338e-05 [a_after_grad]: 9.42001e-06 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 1.76e-06 [auto_monad_grad]: 1.45001e-06 [auto_monad_eliminator]: 8.07e-06 [cse]: 1.984e-05 [a_3]: 3.711e-05 [py_interpret_to_execute_after_opt_a]: 1.919e-05 [slice_cell_reuse_recomputed_activation]: 2.59001e-06 [rewriter_after_opt_a]: 4.789e-05 [convert_after_rewriter]: 7.53999e-06 [order_py_execute_after_rewriter]: 5.57001e-06 [mutable_eliminate]: 0.00076171 [opt_b]: 0.0002247, [1] [Cycle 1]: 0.00021601, [7] [b_1]: 0.00013006 [b_2]: 8.40001e-06 [updatestate_depend_eliminate]: 8.60999e-06 [updatestate_assign_eliminate]: 2.78998e-06 [updatestate_loads_eliminate]: 2.47001e-06 [renormalize]: 6.10016e-07 [cse]: 2.708e-05 [optimize_parallel_all_gather_comm]: 1.935e-05 [overlap_param_gather]: 1.96e-06 [cconv]: 3.577e-05 [loop_unroll]: 0.00046466 [opt_after_cconv]: 0.00011245, [1] [Cycle 1]: 0.00010611, [7] [c_1]: 3.068e-05 [parameter_eliminate]: 4.74e-06 [updatestate_depend_eliminate]: 6.31e-06 [updatestate_assign_eliminate]: 2.93e-06 [updatestate_loads_eliminate]: 2.63003e-06 [cse]: 2.331e-05 [renormalize]: 3.89991e-07 [remove_dup_value]: 1.843e-05 [tuple_transform]: 8.014e-05, [1] [Cycle 1]: 7.524e-05, [4] [d_1]: 4.751e-05 [none_parameter_eliminate]: 1.80001e-06 [renormalize]: 1.60013e-07 [switch_simplify]: 7.16001e-06 [partial_unused_args_eliminate]: 2.00002e-06 [add_recomputation]: 5.591e-05 [cse_after_recomputation]: 2.518e-05, [1] [Cycle 1]: 2.059e-05, [1] [cse]: 1.516e-05 [environ_conv]: 8.1e-06 [swap_dp_allreduce_reducescatter]: 5.58002e-06 [bias_add_comm_swap]: 2.81999e-06 [label_micro_interleaved_index]: 5.17e-06 [label_fine_grained_interleaved_index]: 2.90998e-06 [merge_cast_opt]: 1.66e-06 [slice_recompute_activation]: 2.46e-06 [micro_interleaved_order_control]: 2.34001e-06 [assign_add_opt]: 1.45001e-06 [ForceFp32Comm]: 9.09989e-07 [remove_cast_before_assign_add]: 1.07998e-06 [full_micro_interleaved_order_control]: 2.23002e-06 [reorder_send_recv_between_fp_bp]: 3.20002e-06 [comm_op_add_attrs]: 1.14998e-06 [add_comm_op_reuse_tag]: 1.04e-06 [interleave_split_concat_branches]: 1.44e-06 [interleave_parallel_branches]: 1.08001e-06 [overlap_opt_shard_in_pipeline]: 1.62001e-06 [overlap_opt_shard_grad_in_pipeline]: 1.86e-06 [control_data_broadcast_order]: 1.489e-05 [grouped_pairwise_exchange_alltoall]: 2.19001e-06 [offloading_packed_experts]: 4.78001e-06 [overlap_recompute_and_grad_model_parallel]: 5.26002e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.14e-06 [overlap_recompute_allgather_and_fa_grad]: 1.49e-06 [overlap_recompute_comm]: 2.09e-06 [overlap_grad_ring_attention]: 4.28001e-06 [overlap_grad_flash_sp]: 2.212e-05 [begin_end_overlap_inline]: 5.19998e-07 [split_matmul_comm_elemetwise]: 2.43e-06 [split_layernorm_comm]: 1.70001e-06 [handle_group_info]: 1.17e-06 [symbol_engine_optimizer]: 7.771e-05, [1] [Cycle 1]: 7.331e-05, [6] [build]: 3.85e-06 [elim_shapecalc]: 1.039e-05 [elim_not_effective]: 1.322e-05 [opt_reshape]: 7.33999e-06 [fold_const_symbol]: 9.44e-06 [renormalize]: 1.8999e-07 [detach_backward]: 2.07001e-06 [pipeline_parallel_scheduler]: 2.01e-06 [auto_monad_reorder]: 2.044e-05 [get_jit_bprop_graph]: 2.32001e-06 [rewriter_after_jit_bprop_graph]: 4.74e-06 [opt_after_jit_grad]: 0.00050753 [validate]: 4.746e-05 Sums bootstrap : 0.000627s : 0.50% type_inference : 0.118395s : 94.01% event_method : 0.000015s : 0.01% auto_monad : 0.000056s : 0.04% graph_reusing : 0.000005s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000017s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000034s : 0.03% insert-virtual-dataset : 0.000002s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000021s : 0.02% optimize.rewriter_before_opt_a : 0.000048s : 0.04% optimize.opt_a.expand_dump_flag : 0.000005s : 0.00% optimize.opt_a.switch_simplify : 0.000035s : 0.03% optimize.opt_a.loop_unroll : 0.000018s : 0.01% optimize.opt_a.a_1 : 0.000425s : 0.34% optimize.opt_a.with_stream_mark : 0.000044s : 0.04% optimize.opt_a.recompute_prepare : 0.000016s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000009s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.00% optimize.opt_a.a_2 : 0.000153s : 0.12% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.01% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.00% optimize.opt_a.shard_inline : 0.000013s : 0.01% optimize.opt_a.merge_send_recv : 0.000017s : 0.01% optimize.opt_a.auto_parallel : 0.000017s : 0.01% optimize.opt_a.parallel : 0.000040s : 0.03% optimize.opt_a.flash_sp : 0.000016s : 0.01% optimize.opt_a.merge_comm : 0.000008s : 0.01% optimize.opt_a.allreduce_fusion : 0.000007s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000021s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000017s : 0.01% optimize.opt_a.virtual_dataset : 0.000013s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000012s : 0.01% optimize.opt_a.virtual_output : 0.000013s : 0.01% optimize.opt_a.merge_forward : 0.000009s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000024s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000034s : 0.03% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000022s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.01% optimize.opt_a.meta_fg_expand : 0.000005s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000025s : 0.02% optimize.opt_a.a_after_grad : 0.000019s : 0.01% optimize.opt_a.renormalize : 0.002873s : 2.28% optimize.opt_a.add_forward_monad_depend : 0.000014s : 0.01% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000034s : 0.03% optimize.opt_a.cse : 0.000062s : 0.05% optimize.opt_a.a_3 : 0.000101s : 0.08% optimize.py_interpret_to_execute_after_opt_a : 0.000019s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000048s : 0.04% optimize.convert_after_rewriter : 0.000008s : 0.01% optimize.order_py_execute_after_rewriter : 0.000006s : 0.00% optimize.mutable_eliminate : 0.000762s : 0.60% optimize.opt_b.b_1 : 0.000130s : 0.10% optimize.opt_b.b_2 : 0.000008s : 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 : 0.000002s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000027s : 0.02% optimize.optimize_parallel_all_gather_comm : 0.000019s : 0.02% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000036s : 0.03% optimize.loop_unroll : 0.000465s : 0.37% optimize.opt_after_cconv.c_1 : 0.000031s : 0.02% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.00% 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 : 0.000023s : 0.02% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000018s : 0.01% optimize.tuple_transform.d_1 : 0.000048s : 0.04% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000056s : 0.04% optimize.cse_after_recomputation.cse : 0.000015s : 0.01% optimize.environ_conv : 0.000008s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000015s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000005s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000004s : 0.00% optimize.overlap_grad_flash_sp : 0.000022s : 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 : 0.000010s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000013s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000009s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000020s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000005s : 0.00% opt_after_jit_grad : 0.000508s : 0.40% validate : 0.000047s : 0.04% Time group info: ------[substitution.] 0.000114 21 1.94% : 0.000002s : 2: substitution.elim_not_effective 1.09% : 0.000001s : 2: substitution.fold_const_symbol 5.92% : 0.000007s : 4: substitution.graph_param_transform 76.04% : 0.000087s : 1: substitution.inline 3.72% : 0.000004s : 4: substitution.j_node_and_user_rematch 5.24% : 0.000006s : 4: substitution.remove_not_recompute_node 6.04% : 0.000007s : 4: substitution.replace_old_param ------[type_inference.] 0.118333 2 99.62% : 0.117879s : 1: type_inference.infer 0.38% : 0.000454s : 1: type_inference.specialize ------[replace.] 0.000018 1 100.00% : 0.000018s : 1: replace.inline ------[match.] 0.000086 1 100.00% : 0.000086s : 1: match.inline ------[predicate.] 0.000163 967 1.04% : 0.000002s : 9: predicate.accumulaten_eliminater 1.03% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.63% : 0.000001s : 8: predicate.addn_check_dump 1.06% : 0.000002s : 9: predicate.addn_zero_filter 0.62% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.05% : 0.000003s : 17: predicate.arithmetic_simplify 0.78% : 0.000001s : 9: predicate.cast_eliminate 0.74% : 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 0.76% : 0.000001s : 8: predicate.depend_value_elim 0.76% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 1.01% : 0.000002s : 9: predicate.dict_get_item_eliminator 0.87% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.29% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.37% : 0.000001s : 4: predicate.elim_not_effective 0.49% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.13% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.10% : 0.000002s : 13: predicate.environ_get_add_eliminate 1.12% : 0.000002s : 13: predicate.environ_get_depend_swap 1.89% : 0.000003s : 21: predicate.environ_get_eliminate 0.99% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.74% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.81% : 0.000003s : 10: predicate.float_depend_g_call 0.62% : 0.000001s : 8: predicate.float_environ_get_switch 1.03% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.22% : 0.000000s : 4: predicate.fold_const_symbol 0.94% : 0.000002s : 8: predicate.get_grad_eliminate 0.39% : 0.000001s : 4: predicate.graph_param_transform 0.67% : 0.000001s : 8: predicate.incorporate_call 0.58% : 0.000001s : 8: predicate.incorporate_call_switch 5.79% : 0.000009s : 43: predicate.inline 1.00% : 0.000002s : 8: predicate.inline_without_move 0.40% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.19% : 0.000002s : 8: predicate.less_batch_normalization 1.71% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.43% : 0.000004s : 26: predicate.load_eliminater 1.16% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.26% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.88% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.68% : 0.000001s : 8: predicate.merge_addn 0.72% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.78% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.86% : 0.000001s : 9: predicate.minmaximum_grad 1.88% : 0.000003s : 4: predicate.mutable_eliminate 0.78% : 0.000001s : 4: predicate.opt_reshape 0.66% : 0.000001s : 4: predicate.parallel_virtual_node 1.31% : 0.000002s : 10: predicate.partial_defer_inline 1.13% : 0.000002s : 13: predicate.partial_eliminate 0.78% : 0.000001s : 9: predicate.print_const_string_wrapper 0.71% : 0.000001s : 8: predicate.reduce_all_const_elim 1.28% : 0.000002s : 9: predicate.reduce_eliminate 2.02% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.78% : 0.000001s : 8: predicate.remove_not_recompute_node 1.30% : 0.000002s : 17: predicate.replace_applicator 1.12% : 0.000002s : 8: predicate.replace_old_param 0.61% : 0.000001s : 4: predicate.reset_defer_inline 0.86% : 0.000001s : 9: predicate.reshape_eliminate 0.79% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.70% : 0.000001s : 4: predicate.row_tensor_eliminate 1.42% : 0.000002s : 8: predicate.same_eliminate 0.53% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.95% : 0.000002s : 8: predicate.shard_identity_eliminate 0.88% : 0.000001s : 8: predicate.special_op_eliminate 1.10% : 0.000002s : 8: predicate.specialize_transform 1.31% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.87% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.39% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.79% : 0.000001s : 10: predicate.switch_defer_inline 1.61% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.63% : 0.000006s : 35: predicate.switch_simplify 0.74% : 0.000001s : 9: predicate.tile_eliminate 0.78% : 0.000001s : 9: predicate.transpose_eliminate 1.53% : 0.000002s : 17: predicate.tuple_list_convert_item_index_to_positive 1.61% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.26% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.03% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.57% : 0.000003s : 17: predicate.tuple_list_get_set_item_eliminator 2.29% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.56% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.93% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.79% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.60% : 0.000001s : 4: predicate.value_based_eliminate 0.87% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.92% : 0.000001s : 8: predicate.virtual_output_eliminate 0.36% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.55% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000242 4 9.47% : 0.000023s : 1: func_graph_cloner_run.FuncGraphClonerGraph 90.53% : 0.000219s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.146742 192 0.00% : 0.000004s : 1: ForceFp32Comm 3.02% : 0.004438s : 1: add_attr 3.01% : 0.004422s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.04% : 0.000060s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.04% : 0.000061s : 1: auto_monad 0.02% : 0.000024s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.46% : 0.000668s : 1: bootstrap 0.03% : 0.000039s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000018s : 1: control_data_broadcast_order 0.01% : 0.000011s : 1: convert_after_rewriter 0.02% : 0.000028s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.01% : 0.000012s : 1: environ_conv 0.02% : 0.000023s : 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% : 0.000006s : 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 0.01% : 0.000008s : 1: label_micro_interleaved_index 0.32% : 0.000475s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.53% : 0.000775s : 1: mutable_eliminate 0.01% : 0.000008s : 1: offloading_packed_experts 0.01% : 0.000015s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000018s : 1: opt.transform.mutable_eliminate 0.57% : 0.000838s : 78: opt.transform.opt_a 0.02% : 0.000029s : 1: opt.transform.opt_after_cconv 0.02% : 0.000027s : 1: opt.transform.opt_after_jit_grad 0.07% : 0.000108s : 28: opt.transform.opt_b 0.04% : 0.000052s : 2: opt.transform.opt_trans_graph 0.03% : 0.000037s : 4: opt.transform.symbol_engine_opt 3.19% : 0.004679s : 1: opt_a 0.08% : 0.000116s : 1: opt_after_cconv 0.35% : 0.000520s : 1: opt_after_jit_grad 0.16% : 0.000229s : 1: opt_b 4.82% : 0.007071s : 1: optimize 0.02% : 0.000023s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000009s : 1: order_py_execute_after_rewriter 0.02% : 0.000025s : 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% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000008s : 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% : 0.000004s : 1: pipeline_split 0.03% : 0.000039s : 1: pre_auto_parallel 0.02% : 0.000024s : 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 0.01% : 0.000022s : 1: remove_dup_value 0.28% : 0.000409s : 1: renormalize.infer 1.67% : 0.002452s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000008s : 1: rewriter_after_jit_bprop_graph 0.04% : 0.000052s : 1: rewriter_after_opt_a 0.04% : 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 0.05% : 0.000080s : 1: symbol_engine_optimizer 0.06% : 0.000083s : 1: tuple_transform 80.70% : 0.118418s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[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_fp16[dim_type1-True-batch_size_list0-shape0-ms_dtype0],max_mem:8.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:39:28.507.949 [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 = 0.0350618, [21] [bootstrap]: 0.00065699 [type_inference]: 0.0142569 [event_method]: 1.691e-05 [auto_monad]: 5.826e-05 [graph_reusing]: 5.87001e-06 [inline]: 2.94001e-06 [add_attr]: 0.00496655, [1] [add_attr_with_inline]: 0.00494991, [1] [Cycle 1]: 6.809e-05, [2] [tag_attr]: 1.731e-05 [meta_addattr_fg_expand]: 3.57002e-06 [parallel-infer-symbol]: 4.12e-06 [pre_auto_parallel]: 3.682e-05 [insert-virtual-dataset]: 2.87002e-06 [parallel-infer-symbol-second]: 6.69999e-07 [dataset_repeat_opt]: 2.17001e-06 [pipeline_split]: 1.71e-06 [optimize]: 0.0141025, [53] [py_interpret_to_execute]: 2.242e-05 [rewriter_before_opt_a]: 5.068e-05 [opt_a]: 0.0116108, [2] [Cycle 1]: 0.0017639, [45] [expand_dump_flag]: 2.36e-06 [switch_simplify]: 2.508e-05 [loop_unroll]: 1.151e-05 [a_1]: 0.00029955 [with_stream_mark]: 2.046e-05 [recompute_prepare]: 8.40999e-06 [updatestate_depend_eliminate]: 3.96001e-06 [updatestate_assign_eliminate]: 3.7e-06 [updatestate_loads_eliminate]: 3.44001e-06 [parameter_eliminate]: 2.12001e-06 [a_2]: 8.153e-05 [accelerated_algorithm]: 8.03999e-06 [shard]: 2.74999e-06 [meta_shard_fg_expand]: 2.00002e-06 [shard_inline]: 6.31998e-06 [merge_send_recv]: 8.92999e-06 [auto_parallel]: 7.14001e-06 [parallel]: 2.953e-05 [flash_sp]: 9.99001e-06 [merge_comm]: 4.08999e-06 [allreduce_fusion]: 3.75998e-06 [matmul_add_comm_reduction]: 1.117e-05 [allreduce_slice_to_reducescatter]: 7.09988e-07 [virtual_shard_identity]: 9.50001e-06 [virtual_dataset]: 6.84999e-06 [get_grad_eliminate_]: 6.28e-06 [virtual_output]: 6.27001e-06 [merge_forward]: 4.67998e-06 [cell_reuse_recompute_pass]: 1.30001e-06 [offload_activation]: 1.101e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.484e-05 [merge_recompute_call_nodes]: 1.82999e-06 [before_grad]: 1.077e-05 [set_forward_comm_id_for_comm_node_pass]: 3.99002e-06 [meta_fg_expand]: 2.81e-06 [flash_sp_send_recv_attached]: 3.28e-06 [receive_attached]: 2.14e-06 [after_resolve]: 1.635e-05 [a_after_grad]: 9.92999e-06 [renormalize]: 0.00070081 [add_forward_monad_depend]: 6.89999e-06 [auto_monad_grad]: 3.20998e-06 [auto_monad_eliminator]: 1.698e-05 [cse]: 3.987e-05 [a_3]: 5.16e-05 [Cycle 2]: 0.00983385, [45] [expand_dump_flag]: 1.60999e-06 [switch_simplify]: 8.72e-06 [loop_unroll]: 6.34999e-06 [a_1]: 0.00012294 [with_stream_mark]: 1.494e-05 [recompute_prepare]: 8.12e-06 [updatestate_depend_eliminate]: 4.12e-06 [updatestate_assign_eliminate]: 3.03e-06 [updatestate_loads_eliminate]: 3.06001e-06 [parameter_eliminate]: 1.55999e-06 [a_2]: 0.00902958 [accelerated_algorithm]: 1.45e-05 [shard]: 3.67998e-06 [meta_shard_fg_expand]: 4.60001e-06 [shard_inline]: 8.17e-06 [merge_send_recv]: 1.909e-05 [auto_parallel]: 1.52e-05 [parallel]: 1.175e-05 [flash_sp]: 6.34999e-06 [merge_comm]: 3.93999e-06 [allreduce_fusion]: 3.28998e-06 [matmul_add_comm_reduction]: 1.874e-05 [allreduce_slice_to_reducescatter]: 8.89995e-07 [virtual_shard_identity]: 8.45999e-06 [virtual_dataset]: 6.49999e-06 [get_grad_eliminate_]: 6.53e-06 [virtual_output]: 6.48e-06 [merge_forward]: 4.87998e-06 [cell_reuse_recompute_pass]: 3.48e-06 [offload_activation]: 1.975e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.139e-05 [merge_recompute_call_nodes]: 1.50999e-06 [before_grad]: 1.134e-05 [set_forward_comm_id_for_comm_node_pass]: 3.92998e-06 [meta_fg_expand]: 3.28e-06 [flash_sp_send_recv_attached]: 1.79e-06 [receive_attached]: 2.89999e-06 [after_resolve]: 1.481e-05 [a_after_grad]: 9.62999e-06 [renormalize]: 9.00181e-08 [add_forward_monad_depend]: 2.71e-06 [auto_monad_grad]: 1.59998e-06 [auto_monad_eliminator]: 1.088e-05 [cse]: 5.861e-05 [a_3]: 4.033e-05 [py_interpret_to_execute_after_opt_a]: 2.104e-05 [slice_cell_reuse_recomputed_activation]: 1.97999e-06 [rewriter_after_opt_a]: 4.734e-05 [convert_after_rewriter]: 7.43999e-06 [order_py_execute_after_rewriter]: 5.94999e-06 [mutable_eliminate]: 0.00078955 [opt_b]: 0.00022622, [1] [Cycle 1]: 0.00021756, [7] [b_1]: 0.00012586 [b_2]: 8.92e-06 [updatestate_depend_eliminate]: 6.78998e-06 [updatestate_assign_eliminate]: 3.85e-06 [updatestate_loads_eliminate]: 3.21001e-06 [renormalize]: 6.00005e-07 [cse]: 3.362e-05 [optimize_parallel_all_gather_comm]: 1.984e-05 [overlap_param_gather]: 1.94e-06 [cconv]: 3.607e-05 [loop_unroll]: 0.00047217 [opt_after_cconv]: 0.00014953, [1] [Cycle 1]: 0.00014237, [7] [c_1]: 2.999e-05 [parameter_eliminate]: 4.08001e-06 [updatestate_depend_eliminate]: 6.44001e-06 [updatestate_assign_eliminate]: 2.82002e-06 [updatestate_loads_eliminate]: 2.73e-06 [cse]: 4.103e-05 [renormalize]: 2.19996e-07 [remove_dup_value]: 1.793e-05 [tuple_transform]: 7.686e-05, [1] [Cycle 1]: 7.22e-05, [4] [d_1]: 4.566e-05 [none_parameter_eliminate]: 1.70001e-06 [renormalize]: 2.00002e-07 [switch_simplify]: 6.98e-06 [partial_unused_args_eliminate]: 2.01e-06 [add_recomputation]: 5.195e-05 [cse_after_recomputation]: 3.813e-05, [1] [Cycle 1]: 3.355e-05, [1] [cse]: 2.756e-05 [environ_conv]: 6.73e-06 [swap_dp_allreduce_reducescatter]: 5.44e-06 [bias_add_comm_swap]: 2.59999e-06 [label_micro_interleaved_index]: 4.72998e-06 [label_fine_grained_interleaved_index]: 2.89999e-06 [merge_cast_opt]: 1.34998e-06 [slice_recompute_activation]: 2.09999e-06 [micro_interleaved_order_control]: 2.68e-06 [assign_add_opt]: 1.22e-06 [ForceFp32Comm]: 8.50006e-07 [remove_cast_before_assign_add]: 1.32999e-06 [full_micro_interleaved_order_control]: 2.53e-06 [reorder_send_recv_between_fp_bp]: 2.88e-06 [comm_op_add_attrs]: 1.15001e-06 [add_comm_op_reuse_tag]: 1.35999e-06 [interleave_split_concat_branches]: 1.25001e-06 [interleave_parallel_branches]: 1.40999e-06 [overlap_opt_shard_in_pipeline]: 1.37e-06 [overlap_opt_shard_grad_in_pipeline]: 1.81e-06 [control_data_broadcast_order]: 1.401e-05 [grouped_pairwise_exchange_alltoall]: 1.89999e-06 [offloading_packed_experts]: 4.58999e-06 [overlap_recompute_and_grad_model_parallel]: 4.93001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.37999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.52001e-06 [overlap_recompute_comm]: 2.31e-06 [overlap_grad_ring_attention]: 4.48999e-06 [overlap_grad_flash_sp]: 2.257e-05 [begin_end_overlap_inline]: 5.39992e-07 [split_matmul_comm_elemetwise]: 2.58998e-06 [split_layernorm_comm]: 1.92001e-06 [handle_group_info]: 1.02e-06 [symbol_engine_optimizer]: 7.925e-05, [1] [Cycle 1]: 7.456e-05, [6] [build]: 3.61001e-06 [elim_shapecalc]: 1.063e-05 [elim_not_effective]: 1.338e-05 [opt_reshape]: 7.28999e-06 [fold_const_symbol]: 1.034e-05 [renormalize]: 1.50001e-07 [detach_backward]: 2.27001e-06 [pipeline_parallel_scheduler]: 1.94999e-06 [auto_monad_reorder]: 2.364e-05 [get_jit_bprop_graph]: 2.06998e-06 [rewriter_after_jit_bprop_graph]: 5.81e-06 [opt_after_jit_grad]: 0.00057086 [validate]: 5.16e-05 Sums bootstrap : 0.000657s : 2.28% type_inference : 0.014257s : 49.46% event_method : 0.000017s : 0.06% auto_monad : 0.000058s : 0.20% graph_reusing : 0.000006s : 0.02% inline : 0.000003s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000017s : 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 : 0.000037s : 0.13% 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 : 0.000022s : 0.08% optimize.rewriter_before_opt_a : 0.000051s : 0.18% optimize.opt_a.expand_dump_flag : 0.000004s : 0.01% optimize.opt_a.switch_simplify : 0.000034s : 0.12% optimize.opt_a.loop_unroll : 0.000018s : 0.06% optimize.opt_a.a_1 : 0.000422s : 1.47% optimize.opt_a.with_stream_mark : 0.000035s : 0.12% optimize.opt_a.recompute_prepare : 0.000017s : 0.06% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.03% 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 : 0.009111s : 31.61% optimize.opt_a.accelerated_algorithm : 0.000023s : 0.08% optimize.opt_a.shard : 0.000006s : 0.02% optimize.opt_a.meta_shard_fg_expand : 0.000007s : 0.02% optimize.opt_a.shard_inline : 0.000014s : 0.05% optimize.opt_a.merge_send_recv : 0.000028s : 0.10% optimize.opt_a.auto_parallel : 0.000022s : 0.08% optimize.opt_a.parallel : 0.000041s : 0.14% optimize.opt_a.flash_sp : 0.000016s : 0.06% optimize.opt_a.merge_comm : 0.000008s : 0.03% optimize.opt_a.allreduce_fusion : 0.000007s : 0.02% optimize.opt_a.matmul_add_comm_reduction : 0.000030s : 0.10% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000018s : 0.06% optimize.opt_a.virtual_dataset : 0.000013s : 0.05% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.04% optimize.opt_a.virtual_output : 0.000013s : 0.04% optimize.opt_a.merge_forward : 0.000010s : 0.03% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.02% optimize.opt_a.offload_activation : 0.000031s : 0.11% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000036s : 0.13% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.01% optimize.opt_a.before_grad : 0.000022s : 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 : 0.000005s : 0.02% optimize.opt_a.receive_attached : 0.000005s : 0.02% optimize.opt_a.after_resolve : 0.000031s : 0.11% optimize.opt_a.a_after_grad : 0.000020s : 0.07% optimize.opt_a.renormalize : 0.000701s : 2.43% optimize.opt_a.add_forward_monad_depend : 0.000010s : 0.03% optimize.opt_a.auto_monad_grad : 0.000005s : 0.02% optimize.opt_a.auto_monad_eliminator : 0.000028s : 0.10% optimize.opt_a.cse : 0.000098s : 0.34% optimize.opt_a.a_3 : 0.000092s : 0.32% optimize.py_interpret_to_execute_after_opt_a : 0.000021s : 0.07% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.01% optimize.rewriter_after_opt_a : 0.000047s : 0.16% optimize.convert_after_rewriter : 0.000007s : 0.03% optimize.order_py_execute_after_rewriter : 0.000006s : 0.02% optimize.mutable_eliminate : 0.000790s : 2.74% optimize.opt_b.b_1 : 0.000126s : 0.44% optimize.opt_b.b_2 : 0.000009s : 0.03% optimize.opt_b.updatestate_depend_eliminate : 0.000007s : 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 : 0.000034s : 0.12% optimize.optimize_parallel_all_gather_comm : 0.000020s : 0.07% optimize.overlap_param_gather : 0.000002s : 0.01% optimize.cconv : 0.000036s : 0.13% optimize.loop_unroll : 0.000472s : 1.64% optimize.opt_after_cconv.c_1 : 0.000030s : 0.10% optimize.opt_after_cconv.parameter_eliminate : 0.000004s : 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 : 0.000041s : 0.14% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000018s : 0.06% optimize.tuple_transform.d_1 : 0.000046s : 0.16% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.01% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.02% optimize.partial_unused_args_eliminate : 0.000002s : 0.01% optimize.add_recomputation : 0.000052s : 0.18% optimize.cse_after_recomputation.cse : 0.000028s : 0.10% optimize.environ_conv : 0.000007s : 0.02% optimize.swap_dp_allreduce_reducescatter : 0.000005s : 0.02% 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 : 0.00% 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 : 0.000003s : 0.01% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.01% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.01% optimize.control_data_broadcast_order : 0.000014s : 0.05% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.01% optimize.offloading_packed_experts : 0.000005s : 0.02% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.02% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.01% optimize.overlap_recompute_comm : 0.000002s : 0.01% optimize.overlap_grad_ring_attention : 0.000004s : 0.02% optimize.overlap_grad_flash_sp : 0.000023s : 0.08% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.01% optimize.split_layernorm_comm : 0.000002s : 0.01% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000004s : 0.01% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000011s : 0.04% optimize.symbol_engine_optimizer.elim_not_effective : 0.000013s : 0.05% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.03% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.04% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.01% pipeline_parallel_scheduler : 0.000002s : 0.01% auto_monad_reorder : 0.000024s : 0.08% get_jit_bprop_graph : 0.000002s : 0.01% rewriter_after_jit_bprop_graph : 0.000006s : 0.02% opt_after_jit_grad : 0.000571s : 1.98% validate : 0.000052s : 0.18% Time group info: ------[substitution.] 0.000117 21 2.06% : 0.000002s : 2: substitution.elim_not_effective 1.25% : 0.000001s : 2: substitution.fold_const_symbol 5.18% : 0.000006s : 4: substitution.graph_param_transform 76.64% : 0.000090s : 1: substitution.inline 3.90% : 0.000005s : 4: substitution.j_node_and_user_rematch 4.91% : 0.000006s : 4: substitution.remove_not_recompute_node 6.05% : 0.000007s : 4: substitution.replace_old_param ------[type_inference.] 0.014197 2 97.07% : 0.013781s : 1: type_inference.infer 2.93% : 0.000416s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000089 1 100.00% : 0.000089s : 1: match.inline ------[predicate.] 0.009080 967 0.01% : 0.000001s : 9: predicate.accumulaten_eliminater 0.03% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.01% : 0.000001s : 8: predicate.addn_check_dump 0.01% : 0.000001s : 9: predicate.addn_zero_filter 0.01% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 0.06% : 0.000005s : 17: predicate.arithmetic_simplify 0.01% : 0.000001s : 9: predicate.cast_eliminate 0.01% : 0.000001s : 8: predicate.check_bprop_eliminate 0.01% : 0.000001s : 8: predicate.compare_switch_simplify 0.00% : 0.000000s : 4: predicate.const_output_eliminate 0.02% : 0.000002s : 8: predicate.depend_value_elim 0.01% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.01% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.01% : 0.000001s : 9: predicate.dict_set_item_eliminator 0.03% : 0.000003s : 8: predicate.dumpgradient_eliminate 0.01% : 0.000000s : 4: predicate.elim_not_effective 0.01% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 0.02% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.02% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.02% : 0.000001s : 13: predicate.environ_get_depend_swap 0.03% : 0.000003s : 21: predicate.environ_get_eliminate 0.02% : 0.000001s : 13: predicate.environ_get_set_eliminate 0.01% : 0.000001s : 10: predicate.exchange_switch_depend_value 0.03% : 0.000003s : 10: predicate.float_depend_g_call 98.18% : 0.008916s : 8: predicate.float_environ_get_switch 0.02% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.00% : 0.000000s : 4: predicate.fold_const_symbol 0.02% : 0.000002s : 8: predicate.get_grad_eliminate 0.00% : 0.000000s : 4: predicate.graph_param_transform 0.01% : 0.000001s : 8: predicate.incorporate_call 0.01% : 0.000001s : 8: predicate.incorporate_call_switch 0.13% : 0.000012s : 43: predicate.inline 0.02% : 0.000001s : 8: predicate.inline_without_move 0.01% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.03% : 0.000003s : 8: predicate.less_batch_normalization 0.04% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 0.04% : 0.000003s : 26: predicate.load_eliminater 0.02% : 0.000002s : 4: predicate.loop_unroll_after_grad 0.03% : 0.000002s : 13: predicate.loop_unroll_before_grad 0.03% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.01% : 0.000001s : 8: predicate.merge_addn 0.02% : 0.000002s : 8: predicate.micro_step_allgather_replace 0.01% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.01% : 0.000001s : 9: predicate.minmaximum_grad 0.03% : 0.000003s : 4: predicate.mutable_eliminate 0.01% : 0.000001s : 4: predicate.opt_reshape 0.01% : 0.000001s : 4: predicate.parallel_virtual_node 0.02% : 0.000002s : 10: predicate.partial_defer_inline 0.02% : 0.000002s : 13: predicate.partial_eliminate 0.02% : 0.000001s : 9: predicate.print_const_string_wrapper 0.01% : 0.000001s : 8: predicate.reduce_all_const_elim 0.02% : 0.000002s : 9: predicate.reduce_eliminate 0.04% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.02% : 0.000001s : 8: predicate.remove_not_recompute_node 0.02% : 0.000002s : 17: predicate.replace_applicator 0.02% : 0.000001s : 8: predicate.replace_old_param 0.01% : 0.000001s : 4: predicate.reset_defer_inline 0.01% : 0.000001s : 9: predicate.reshape_eliminate 0.01% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.01% : 0.000001s : 4: predicate.row_tensor_eliminate 0.02% : 0.000002s : 8: predicate.same_eliminate 0.01% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.02% : 0.000002s : 8: predicate.shard_identity_eliminate 0.02% : 0.000002s : 8: predicate.special_op_eliminate 0.02% : 0.000002s : 8: predicate.specialize_transform 0.03% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.02% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.01% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.01% : 0.000001s : 10: predicate.switch_defer_inline 0.03% : 0.000003s : 18: predicate.switch_layer_defer_inline 0.06% : 0.000006s : 35: predicate.switch_simplify 0.01% : 0.000001s : 9: predicate.tile_eliminate 0.01% : 0.000001s : 9: predicate.transpose_eliminate 0.03% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 0.03% : 0.000002s : 17: predicate.tuple_list_get_item_const_eliminator 0.03% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 0.06% : 0.000006s : 25: predicate.tuple_list_get_item_eliminator 0.02% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 0.05% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 0.03% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 0.04% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 0.05% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.01% : 0.000001s : 4: predicate.value_based_eliminate 0.02% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.02% : 0.000001s : 8: predicate.virtual_output_eliminate 0.00% : 0.000000s : 4: predicate.virtual_view_grad_eliminate 0.01% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000247 4 9.23% : 0.000023s : 1: func_graph_cloner_run.FuncGraphClonerGraph 90.77% : 0.000224s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.064637 192 0.01% : 0.000004s : 1: ForceFp32Comm 7.69% : 0.004972s : 1: add_attr 7.66% : 0.004954s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.09% : 0.000057s : 1: add_recomputation 0.03% : 0.000017s : 1: assign_add_opt 0.10% : 0.000064s : 1: auto_monad 0.04% : 0.000028s : 1: auto_monad_reorder 0.01% : 0.000003s : 1: begin_end_overlap_inline 0.01% : 0.000006s : 1: bias_add_comm_swap 1.07% : 0.000690s : 1: bootstrap 0.06% : 0.000039s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.03% : 0.000017s : 1: control_data_broadcast_order 0.02% : 0.000011s : 1: convert_after_rewriter 0.06% : 0.000041s : 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% : 0.000024s : 1: event_method 0.01% : 0.000005s : 1: full_micro_interleaved_order_control 0.01% : 0.000006s : 1: get_jit_bprop_graph 0.01% : 0.000009s : 1: graph_reusing 0.01% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000004s : 1: handle_group_info 0.01% : 0.000006s : 1: inline 0.01% : 0.000006s : 1: insert-virtual-dataset 0.02% : 0.000010s : 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 0.75% : 0.000482s : 1: loop_unroll 0.01% : 0.000004s : 1: merge_cast_opt 0.01% : 0.000006s : 1: micro_interleaved_order_control 1.24% : 0.000799s : 1: mutable_eliminate 0.01% : 0.000007s : 1: offloading_packed_experts 0.02% : 0.000015s : 1: opt.transform.loop_unroll_optimizer 0.03% : 0.000020s : 1: opt.transform.mutable_eliminate 15.14% : 0.009788s : 78: opt.transform.opt_a 0.04% : 0.000028s : 1: opt.transform.opt_after_cconv 0.05% : 0.000031s : 1: opt.transform.opt_after_jit_grad 0.16% : 0.000105s : 28: opt.transform.opt_b 0.08% : 0.000051s : 2: opt.transform.opt_trans_graph 0.06% : 0.000038s : 4: opt.transform.symbol_engine_opt 17.97% : 0.011614s : 1: opt_a 0.24% : 0.000154s : 1: opt_after_cconv 0.90% : 0.000585s : 1: opt_after_jit_grad 0.36% : 0.000230s : 1: opt_b 21.83% : 0.014108s : 1: optimize 0.04% : 0.000023s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000009s : 1: order_py_execute_after_rewriter 0.04% : 0.000026s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000008s : 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 0.02% : 0.000013s : 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% : 0.000003s : 1: parallel-infer-symbol-second 0.01% : 0.000005s : 1: partial_unused_args_eliminate 0.01% : 0.000008s : 1: pipeline_parallel_scheduler 0.01% : 0.000005s : 1: pipeline_split 0.06% : 0.000042s : 1: pre_auto_parallel 0.04% : 0.000027s : 1: py_interpret_to_execute 0.04% : 0.000025s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000004s : 1: remove_cast_before_assign_add 0.03% : 0.000022s : 1: remove_dup_value 0.64% : 0.000417s : 1: renormalize.infer 0.43% : 0.000275s : 1: renormalize.specialize 0.01% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.08% : 0.000052s : 1: rewriter_after_opt_a 0.09% : 0.000055s : 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% : 0.000006s : 1: split_matmul_comm_elemetwise 0.01% : 0.000008s : 1: swap_dp_allreduce_reducescatter 0.13% : 0.000082s : 1: symbol_engine_optimizer 0.12% : 0.000080s : 1: tuple_transform 22.10% : 0.014282s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[dim_type1-True-batch_size_list0-shape1-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_fp16[dim_type1-True-batch_size_list0-shape1-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:39:30.909.739 [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 = 0.0871558, [21] [bootstrap]: 0.00061088 [type_inference]: 0.00403385 [event_method]: 1.631e-05 [auto_monad]: 5.378e-05 [graph_reusing]: 5.46002e-06 [inline]: 2.63e-06 [add_attr]: 0.0771385, [1] [add_attr_with_inline]: 0.0771235, [1] [Cycle 1]: 4.593e-05, [2] [tag_attr]: 1.497e-05 [meta_addattr_fg_expand]: 3.36999e-06 [parallel-infer-symbol]: 3.65e-06 [pre_auto_parallel]: 3.221e-05 [insert-virtual-dataset]: 2.96999e-06 [parallel-infer-symbol-second]: 6.79982e-07 [dataset_repeat_opt]: 2.58998e-06 [pipeline_split]: 1.71e-06 [optimize]: 0.00454284, [53] [py_interpret_to_execute]: 2.067e-05 [rewriter_before_opt_a]: 4.58e-05 [opt_a]: 0.00227614, [2] [Cycle 1]: 0.00163407, [45] [expand_dump_flag]: 2.81e-06 [switch_simplify]: 2.079e-05 [loop_unroll]: 1.198e-05 [a_1]: 0.0002854 [with_stream_mark]: 1.818e-05 [recompute_prepare]: 8.61002e-06 [updatestate_depend_eliminate]: 4.05e-06 [updatestate_assign_eliminate]: 4.35e-06 [updatestate_loads_eliminate]: 3.39001e-06 [parameter_eliminate]: 1.59e-06 [a_2]: 8.151e-05 [accelerated_algorithm]: 7.43e-06 [shard]: 2.27999e-06 [meta_shard_fg_expand]: 1.76003e-06 [shard_inline]: 6.81999e-06 [merge_send_recv]: 9.32999e-06 [auto_parallel]: 7.4e-06 [parallel]: 3.651e-05 [flash_sp]: 1.018e-05 [merge_comm]: 3.78001e-06 [allreduce_fusion]: 3.65998e-06 [matmul_add_comm_reduction]: 7.82e-06 [allreduce_slice_to_reducescatter]: 1.35001e-06 [virtual_shard_identity]: 8.47e-06 [virtual_dataset]: 6.46999e-06 [get_grad_eliminate_]: 7.56999e-06 [virtual_output]: 7.28999e-06 [merge_forward]: 4.35e-06 [cell_reuse_recompute_pass]: 1.46002e-06 [offload_activation]: 9.89001e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.33e-05 [merge_recompute_call_nodes]: 1.59e-06 [before_grad]: 1.054e-05 [set_forward_comm_id_for_comm_node_pass]: 3.53e-06 [meta_fg_expand]: 2.92002e-06 [flash_sp_send_recv_attached]: 2.48998e-06 [receive_attached]: 2.12001e-06 [after_resolve]: 1.192e-05 [a_after_grad]: 9.80002e-06 [renormalize]: 0.00066974 [add_forward_monad_depend]: 5.72999e-06 [auto_monad_grad]: 2.90002e-06 [auto_monad_eliminator]: 1.583e-05 [cse]: 2.387e-05 [a_3]: 4.829e-05 [Cycle 2]: 0.00063244, [45] [expand_dump_flag]: 1.22999e-06 [switch_simplify]: 7.23e-06 [loop_unroll]: 5.94999e-06 [a_1]: 0.000127 [with_stream_mark]: 1.034e-05 [recompute_prepare]: 5.98002e-06 [updatestate_depend_eliminate]: 3.67002e-06 [updatestate_assign_eliminate]: 3.13e-06 [updatestate_loads_eliminate]: 2.43e-06 [parameter_eliminate]: 1.00999e-06 [a_2]: 7.076e-05 [accelerated_algorithm]: 5.89e-06 [shard]: 9.89996e-07 [meta_shard_fg_expand]: 1.55999e-06 [shard_inline]: 6.09001e-06 [merge_send_recv]: 5.61e-06 [auto_parallel]: 6.32001e-06 [parallel]: 4.95001e-06 [flash_sp]: 4.14002e-06 [merge_comm]: 3.52002e-06 [allreduce_fusion]: 3.30998e-06 [matmul_add_comm_reduction]: 7.14001e-06 [allreduce_slice_to_reducescatter]: 4.69998e-07 [virtual_shard_identity]: 7e-06 [virtual_dataset]: 6.38e-06 [get_grad_eliminate_]: 6.04999e-06 [virtual_output]: 5.84999e-06 [merge_forward]: 3.61001e-06 [cell_reuse_recompute_pass]: 1.74e-06 [offload_activation]: 8.37e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.282e-05 [merge_recompute_call_nodes]: 1.09e-06 [before_grad]: 9.72999e-06 [set_forward_comm_id_for_comm_node_pass]: 3.29001e-06 [meta_fg_expand]: 2.58e-06 [flash_sp_send_recv_attached]: 1.12999e-06 [receive_attached]: 1.50999e-06 [after_resolve]: 1.064e-05 [a_after_grad]: 8.80001e-06 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 1.50001e-06 [auto_monad_grad]: 9.09989e-07 [auto_monad_eliminator]: 6.92002e-06 [cse]: 1.54e-05 [a_3]: 3.559e-05 [py_interpret_to_execute_after_opt_a]: 1.172e-05 [slice_cell_reuse_recomputed_activation]: 2.39001e-06 [rewriter_after_opt_a]: 3.752e-05 [convert_after_rewriter]: 7.15e-06 [order_py_execute_after_rewriter]: 5.89e-06 [mutable_eliminate]: 0.0007515 [opt_b]: 0.00021494, [1] [Cycle 1]: 0.00020791, [7] [b_1]: 0.00012562 [b_2]: 8.25e-06 [updatestate_depend_eliminate]: 7.43e-06 [updatestate_assign_eliminate]: 3.02002e-06 [updatestate_loads_eliminate]: 2.68e-06 [renormalize]: 3.80009e-07 [cse]: 2.578e-05 [optimize_parallel_all_gather_comm]: 1.621e-05 [overlap_param_gather]: 2.24001e-06 [cconv]: 3.049e-05 [loop_unroll]: 0.00043744 [opt_after_cconv]: 0.00010503, [1] [Cycle 1]: 9.909e-05, [7] [c_1]: 2.988e-05 [parameter_eliminate]: 3.41001e-06 [updatestate_depend_eliminate]: 5.74e-06 [updatestate_assign_eliminate]: 2.88e-06 [updatestate_loads_eliminate]: 2.56e-06 [cse]: 2.13e-05 [renormalize]: 3.00002e-07 [remove_dup_value]: 1.663e-05 [tuple_transform]: 7.783e-05, [1] [Cycle 1]: 7.321e-05, [4] [d_1]: 4.552e-05 [none_parameter_eliminate]: 1.72001e-06 [renormalize]: 2.10013e-07 [switch_simplify]: 7.02002e-06 [partial_unused_args_eliminate]: 2.12001e-06 [add_recomputation]: 4.764e-05 [cse_after_recomputation]: 2.345e-05, [1] [Cycle 1]: 1.916e-05, [1] [cse]: 1.387e-05 [environ_conv]: 6.64001e-06 [swap_dp_allreduce_reducescatter]: 5.89e-06 [bias_add_comm_swap]: 3.08e-06 [label_micro_interleaved_index]: 5.30001e-06 [label_fine_grained_interleaved_index]: 2.74001e-06 [merge_cast_opt]: 1.45999e-06 [slice_recompute_activation]: 2.41e-06 [micro_interleaved_order_control]: 2.51e-06 [assign_add_opt]: 1.24e-06 [ForceFp32Comm]: 8.29983e-07 [remove_cast_before_assign_add]: 9.30013e-07 [full_micro_interleaved_order_control]: 2.22999e-06 [reorder_send_recv_between_fp_bp]: 3.11999e-06 [comm_op_add_attrs]: 1.07998e-06 [add_comm_op_reuse_tag]: 8.30012e-07 [interleave_split_concat_branches]: 1.29998e-06 [interleave_parallel_branches]: 9.20001e-07 [overlap_opt_shard_in_pipeline]: 1.30999e-06 [overlap_opt_shard_grad_in_pipeline]: 2.13998e-06 [control_data_broadcast_order]: 1.317e-05 [grouped_pairwise_exchange_alltoall]: 1.60001e-06 [offloading_packed_experts]: 4.35e-06 [overlap_recompute_and_grad_model_parallel]: 5.20999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.21002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.42e-06 [overlap_recompute_comm]: 2.36998e-06 [overlap_grad_ring_attention]: 5.13002e-06 [overlap_grad_flash_sp]: 2.065e-05 [begin_end_overlap_inline]: 5.49975e-07 [split_matmul_comm_elemetwise]: 2.31e-06 [split_layernorm_comm]: 1.74e-06 [handle_group_info]: 1.04e-06 [symbol_engine_optimizer]: 7.57e-05, [1] [Cycle 1]: 7.103e-05, [6] [build]: 3.13e-06 [elim_shapecalc]: 9.91e-06 [elim_not_effective]: 1.298e-05 [opt_reshape]: 6.94001e-06 [fold_const_symbol]: 1e-05 [renormalize]: 1.50001e-07 [detach_backward]: 2.07999e-06 [pipeline_parallel_scheduler]: 1.82999e-06 [auto_monad_reorder]: 1.694e-05 [get_jit_bprop_graph]: 1.32999e-06 [rewriter_after_jit_bprop_graph]: 4.92e-06 [opt_after_jit_grad]: 0.00046601 [validate]: 3.355e-05 Sums bootstrap : 0.000611s : 6.75% type_inference : 0.004034s : 44.55% event_method : 0.000016s : 0.18% auto_monad : 0.000054s : 0.59% graph_reusing : 0.000005s : 0.06% inline : 0.000003s : 0.03% add_attr.add_attr_with_inline.tag_attr : 0.000015s : 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 : 0.000032s : 0.36% insert-virtual-dataset : 0.000003s : 0.03% parallel-infer-symbol-second : 0.000001s : 0.01% dataset_repeat_opt : 0.000003s : 0.03% pipeline_split : 0.000002s : 0.02% optimize.py_interpret_to_execute : 0.000021s : 0.23% optimize.rewriter_before_opt_a : 0.000046s : 0.51% optimize.opt_a.expand_dump_flag : 0.000004s : 0.04% optimize.opt_a.switch_simplify : 0.000028s : 0.31% optimize.opt_a.loop_unroll : 0.000018s : 0.20% optimize.opt_a.a_1 : 0.000412s : 4.55% optimize.opt_a.with_stream_mark : 0.000029s : 0.31% optimize.opt_a.recompute_prepare : 0.000015s : 0.16% 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 : 0.000006s : 0.06% optimize.opt_a.parameter_eliminate : 0.000003s : 0.03% optimize.opt_a.a_2 : 0.000152s : 1.68% optimize.opt_a.accelerated_algorithm : 0.000013s : 0.15% optimize.opt_a.shard : 0.000003s : 0.04% optimize.opt_a.meta_shard_fg_expand : 0.000003s : 0.04% optimize.opt_a.shard_inline : 0.000013s : 0.14% optimize.opt_a.merge_send_recv : 0.000015s : 0.16% optimize.opt_a.auto_parallel : 0.000014s : 0.15% optimize.opt_a.parallel : 0.000041s : 0.46% optimize.opt_a.flash_sp : 0.000014s : 0.16% optimize.opt_a.merge_comm : 0.000007s : 0.08% optimize.opt_a.allreduce_fusion : 0.000007s : 0.08% optimize.opt_a.matmul_add_comm_reduction : 0.000015s : 0.17% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.02% optimize.opt_a.virtual_shard_identity : 0.000015s : 0.17% optimize.opt_a.virtual_dataset : 0.000013s : 0.14% optimize.opt_a.get_grad_eliminate_ : 0.000014s : 0.15% 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 : 0.04% optimize.opt_a.offload_activation : 0.000018s : 0.20% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000026s : 0.29% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.03% optimize.opt_a.before_grad : 0.000020s : 0.22% 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 : 0.000023s : 0.25% optimize.opt_a.a_after_grad : 0.000019s : 0.21% optimize.opt_a.renormalize : 0.000670s : 7.40% optimize.opt_a.add_forward_monad_depend : 0.000007s : 0.08% optimize.opt_a.auto_monad_grad : 0.000004s : 0.04% optimize.opt_a.auto_monad_eliminator : 0.000023s : 0.25% optimize.opt_a.cse : 0.000039s : 0.43% optimize.opt_a.a_3 : 0.000084s : 0.93% optimize.py_interpret_to_execute_after_opt_a : 0.000012s : 0.13% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.03% optimize.rewriter_after_opt_a : 0.000038s : 0.41% optimize.convert_after_rewriter : 0.000007s : 0.08% optimize.order_py_execute_after_rewriter : 0.000006s : 0.07% optimize.mutable_eliminate : 0.000751s : 8.30% optimize.opt_b.b_1 : 0.000126s : 1.39% optimize.opt_b.b_2 : 0.000008s : 0.09% optimize.opt_b.updatestate_depend_eliminate : 0.000007s : 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 : 0.000000s : 0.00% optimize.opt_b.cse : 0.000026s : 0.28% optimize.optimize_parallel_all_gather_comm : 0.000016s : 0.18% optimize.overlap_param_gather : 0.000002s : 0.02% optimize.cconv : 0.000030s : 0.34% optimize.loop_unroll : 0.000437s : 4.83% optimize.opt_after_cconv.c_1 : 0.000030s : 0.33% optimize.opt_after_cconv.parameter_eliminate : 0.000003s : 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 : 0.000003s : 0.03% optimize.opt_after_cconv.cse : 0.000021s : 0.24% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000017s : 0.18% optimize.tuple_transform.d_1 : 0.000046s : 0.50% 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 : 0.000014s : 0.15% optimize.environ_conv : 0.000007s : 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 : 0.06% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.03% optimize.merge_cast_opt : 0.000001s : 0.02% optimize.slice_recompute_activation : 0.000002s : 0.03% optimize.micro_interleaved_order_control : 0.000003s : 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 : 0.000001s : 0.01% 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 : 0.06% 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 : 0.000002s : 0.03% optimize.overlap_grad_ring_attention : 0.000005s : 0.06% optimize.overlap_grad_flash_sp : 0.000021s : 0.23% 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 : 0.000002s : 0.02% auto_monad_reorder : 0.000017s : 0.19% get_jit_bprop_graph : 0.000001s : 0.01% rewriter_after_jit_bprop_graph : 0.000005s : 0.05% opt_after_jit_grad : 0.000466s : 5.15% validate : 0.000034s : 0.37% Time group info: ------[substitution.] 0.000096 21 2.03% : 0.000002s : 2: substitution.elim_not_effective 1.52% : 0.000001s : 2: substitution.fold_const_symbol 7.05% : 0.000007s : 4: substitution.graph_param_transform 74.59% : 0.000071s : 1: substitution.inline 3.90% : 0.000004s : 4: substitution.j_node_and_user_rematch 5.36% : 0.000005s : 4: substitution.remove_not_recompute_node 5.56% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.003974 2 89.39% : 0.003553s : 1: type_inference.infer 10.61% : 0.000422s : 1: type_inference.specialize ------[replace.] 0.000013 1 100.00% : 0.000013s : 1: replace.inline ------[match.] 0.000070 1 100.00% : 0.000070s : 1: match.inline ------[predicate.] 0.000162 967 1.06% : 0.000002s : 9: predicate.accumulaten_eliminater 0.86% : 0.000001s : 4: predicate.ad_related_special_op_eliminate 0.86% : 0.000001s : 8: predicate.addn_check_dump 1.08% : 0.000002s : 9: predicate.addn_zero_filter 0.75% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.04% : 0.000003s : 17: predicate.arithmetic_simplify 0.80% : 0.000001s : 9: predicate.cast_eliminate 1.03% : 0.000002s : 8: predicate.check_bprop_eliminate 0.62% : 0.000001s : 8: predicate.compare_switch_simplify 0.23% : 0.000000s : 4: predicate.const_output_eliminate 0.73% : 0.000001s : 8: predicate.depend_value_elim 0.98% : 0.000002s : 9: predicate.dict_get_item_const_eliminator 0.86% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.74% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.24% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.27% : 0.000000s : 4: predicate.elim_not_effective 0.66% : 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 0.93% : 0.000002s : 13: predicate.environ_get_depend_swap 1.65% : 0.000003s : 21: predicate.environ_get_eliminate 1.01% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.85% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.44% : 0.000002s : 10: predicate.float_depend_g_call 0.65% : 0.000001s : 8: predicate.float_environ_get_switch 1.05% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.23% : 0.000000s : 4: predicate.fold_const_symbol 0.97% : 0.000002s : 8: predicate.get_grad_eliminate 0.37% : 0.000001s : 4: predicate.graph_param_transform 0.65% : 0.000001s : 8: predicate.incorporate_call 0.56% : 0.000001s : 8: predicate.incorporate_call_switch 5.38% : 0.000009s : 43: predicate.inline 0.99% : 0.000002s : 8: predicate.inline_without_move 0.41% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.15% : 0.000002s : 8: predicate.less_batch_normalization 1.55% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.40% : 0.000004s : 26: predicate.load_eliminater 1.15% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.32% : 0.000002s : 13: predicate.loop_unroll_before_grad 2.04% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.67% : 0.000001s : 8: predicate.merge_addn 0.72% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.73% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.78% : 0.000001s : 9: predicate.minmaximum_grad 1.39% : 0.000002s : 4: predicate.mutable_eliminate 0.49% : 0.000001s : 4: predicate.opt_reshape 0.58% : 0.000001s : 4: predicate.parallel_virtual_node 1.18% : 0.000002s : 10: predicate.partial_defer_inline 1.10% : 0.000002s : 13: predicate.partial_eliminate 1.02% : 0.000002s : 9: predicate.print_const_string_wrapper 0.91% : 0.000001s : 8: predicate.reduce_all_const_elim 1.30% : 0.000002s : 9: predicate.reduce_eliminate 2.19% : 0.000004s : 26: predicate.redundant_stop_gradient_eliminater 0.63% : 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 0.99% : 0.000002s : 9: predicate.reshape_eliminate 0.85% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.53% : 0.000001s : 4: predicate.row_tensor_eliminate 1.01% : 0.000002s : 8: predicate.same_eliminate 0.56% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.46% : 0.000002s : 8: predicate.shard_identity_eliminate 1.02% : 0.000002s : 8: predicate.special_op_eliminate 0.83% : 0.000001s : 8: predicate.specialize_transform 1.31% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 1.01% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.48% : 0.000001s : 4: predicate.switch_call_monad_eliminater 1.00% : 0.000002s : 10: predicate.switch_defer_inline 1.74% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.48% : 0.000006s : 35: predicate.switch_simplify 0.91% : 0.000001s : 9: predicate.tile_eliminate 1.04% : 0.000002s : 9: predicate.transpose_eliminate 1.52% : 0.000002s : 17: predicate.tuple_list_convert_item_index_to_positive 1.56% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.56% : 0.000003s : 17: predicate.tuple_list_get_item_depend_reorder 3.64% : 0.000006s : 25: predicate.tuple_list_get_item_eliminator 1.57% : 0.000003s : 17: predicate.tuple_list_get_set_item_eliminator 2.49% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.46% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 2.16% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.72% : 0.000004s : 34: predicate.updatestate_useless_node_eliminater 0.53% : 0.000001s : 4: predicate.value_based_eliminate 0.97% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.93% : 0.000002s : 8: predicate.virtual_output_eliminate 0.31% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.55% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000243 4 8.49% : 0.000021s : 1: func_graph_cloner_run.FuncGraphClonerGraph 91.51% : 0.000222s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.170434 192 0.00% : 0.000004s : 1: ForceFp32Comm 45.26% : 0.077145s : 1: add_attr 45.25% : 0.077128s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.03% : 0.000052s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.03% : 0.000059s : 1: auto_monad 0.01% : 0.000021s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.38% : 0.000655s : 1: bootstrap 0.02% : 0.000034s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000016s : 1: control_data_broadcast_order 0.01% : 0.000010s : 1: convert_after_rewriter 0.02% : 0.000026s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.01% : 0.000010s : 1: environ_conv 0.01% : 0.000023s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000004s : 1: get_jit_bprop_graph 0.01% : 0.000009s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.26% : 0.000446s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.45% : 0.000762s : 1: mutable_eliminate 0.00% : 0.000007s : 1: offloading_packed_experts 0.01% : 0.000014s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000016s : 1: opt.transform.mutable_eliminate 0.47% : 0.000796s : 78: opt.transform.opt_a 0.02% : 0.000029s : 1: opt.transform.opt_after_cconv 0.01% : 0.000024s : 1: opt.transform.opt_after_jit_grad 0.06% : 0.000105s : 28: opt.transform.opt_b 0.03% : 0.000050s : 2: opt.transform.opt_trans_graph 0.02% : 0.000036s : 4: opt.transform.symbol_engine_opt 1.34% : 0.002279s : 1: opt_a 0.06% : 0.000108s : 1: opt_after_cconv 0.28% : 0.000476s : 1: opt_after_jit_grad 0.13% : 0.000218s : 1: opt_b 2.67% : 0.004547s : 1: optimize 0.01% : 0.000020s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000009s : 1: order_py_execute_after_rewriter 0.01% : 0.000024s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000008s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000008s : 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 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.02% : 0.000037s : 1: pre_auto_parallel 0.01% : 0.000025s : 1: py_interpret_to_execute 0.01% : 0.000015s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.01% : 0.000020s : 1: remove_dup_value 0.23% : 0.000392s : 1: renormalize.infer 0.16% : 0.000269s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000008s : 1: rewriter_after_jit_bprop_graph 0.02% : 0.000041s : 1: rewriter_after_opt_a 0.03% : 0.000049s : 1: rewriter_before_opt_a 0.00% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000004s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.05% : 0.000078s : 1: symbol_engine_optimizer 0.05% : 0.000081s : 1: tuple_transform 2.38% : 0.004056s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[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_fp16[dim_type1-True-batch_size_list1-shape0-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:39:34.117.191 [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 = 0.0397952, [21] [bootstrap]: 0.0006915 [type_inference]: 0.00420121 [event_method]: 1.456e-05 [auto_monad]: 6.018e-05 [graph_reusing]: 5.93998e-06 [inline]: 2.49999e-06 [add_attr]: 0.0290283, [1] [add_attr_with_inline]: 0.0290122, [1] [Cycle 1]: 7.255e-05, [2] [tag_attr]: 1.752e-05 [meta_addattr_fg_expand]: 3.55e-06 [parallel-infer-symbol]: 4.41002e-06 [pre_auto_parallel]: 3.919e-05 [insert-virtual-dataset]: 2.79001e-06 [parallel-infer-symbol-second]: 7.2e-07 [dataset_repeat_opt]: 1.97999e-06 [pipeline_split]: 1.73002e-06 [optimize]: 0.00490533, [53] [py_interpret_to_execute]: 2.247e-05 [rewriter_before_opt_a]: 5.717e-05 [opt_a]: 0.00258243, [2] [Cycle 1]: 0.0019038, [45] [expand_dump_flag]: 2.73e-06 [switch_simplify]: 2.481e-05 [loop_unroll]: 1.273e-05 [a_1]: 0.00031737 [with_stream_mark]: 2.291e-05 [recompute_prepare]: 8.80999e-06 [updatestate_depend_eliminate]: 4.75999e-06 [updatestate_assign_eliminate]: 3.65e-06 [updatestate_loads_eliminate]: 3.69002e-06 [parameter_eliminate]: 1.97001e-06 [a_2]: 8.687e-05 [accelerated_algorithm]: 7.78001e-06 [shard]: 2.84001e-06 [meta_shard_fg_expand]: 1.97001e-06 [shard_inline]: 7.1e-06 [merge_send_recv]: 8.64e-06 [auto_parallel]: 8.70999e-06 [parallel]: 3.442e-05 [flash_sp]: 9.36e-06 [merge_comm]: 4.19002e-06 [allreduce_fusion]: 3.93999e-06 [matmul_add_comm_reduction]: 1.172e-05 [allreduce_slice_to_reducescatter]: 7.60017e-07 [virtual_shard_identity]: 8.98002e-06 [virtual_dataset]: 7.48e-06 [get_grad_eliminate_]: 6.41e-06 [virtual_output]: 6.84001e-06 [merge_forward]: 5.34e-06 [cell_reuse_recompute_pass]: 1.34e-06 [offload_activation]: 1.172e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.613e-05 [merge_recompute_call_nodes]: 1.52001e-06 [before_grad]: 1.266e-05 [set_forward_comm_id_for_comm_node_pass]: 4.35999e-06 [meta_fg_expand]: 3.50998e-06 [flash_sp_send_recv_attached]: 2.91e-06 [receive_attached]: 2.45002e-06 [after_resolve]: 1.257e-05 [a_after_grad]: 1.024e-05 [renormalize]: 0.00082619 [add_forward_monad_depend]: 6.49001e-06 [auto_monad_grad]: 3.05002e-06 [auto_monad_eliminator]: 1.778e-05 [cse]: 3.972e-05 [a_3]: 5.249e-05 [Cycle 2]: 0.00066656, [45] [expand_dump_flag]: 1.29998e-06 [switch_simplify]: 7.7e-06 [loop_unroll]: 6.33e-06 [a_1]: 0.00012636 [with_stream_mark]: 1.507e-05 [recompute_prepare]: 6.54001e-06 [updatestate_depend_eliminate]: 4.16001e-06 [updatestate_assign_eliminate]: 2.74999e-06 [updatestate_loads_eliminate]: 3.14001e-06 [parameter_eliminate]: 1.29e-06 [a_2]: 7.278e-05 [accelerated_algorithm]: 6.44999e-06 [shard]: 1.39998e-06 [meta_shard_fg_expand]: 2.01998e-06 [shard_inline]: 6.23e-06 [merge_send_recv]: 5.44e-06 [auto_parallel]: 6.58e-06 [parallel]: 6.54999e-06 [flash_sp]: 3.45e-06 [merge_comm]: 3.76999e-06 [allreduce_fusion]: 3.63e-06 [matmul_add_comm_reduction]: 1.466e-05 [allreduce_slice_to_reducescatter]: 3.39991e-07 [virtual_shard_identity]: 7.63001e-06 [virtual_dataset]: 6.31e-06 [get_grad_eliminate_]: 5.79999e-06 [virtual_output]: 6.20002e-06 [merge_forward]: 3.55e-06 [cell_reuse_recompute_pass]: 1.71e-06 [offload_activation]: 8.79003e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.4e-05 [merge_recompute_call_nodes]: 1.50001e-06 [before_grad]: 9.86e-06 [set_forward_comm_id_for_comm_node_pass]: 3.56999e-06 [meta_fg_expand]: 2.23998e-06 [flash_sp_send_recv_attached]: 1.41998e-06 [receive_attached]: 1.55001e-06 [after_resolve]: 1.102e-05 [a_after_grad]: 9.79e-06 [renormalize]: 8.00064e-08 [add_forward_monad_depend]: 1.19e-06 [auto_monad_grad]: 1.40999e-06 [auto_monad_eliminator]: 6.97002e-06 [cse]: 1.82e-05 [a_3]: 3.791e-05 [py_interpret_to_execute_after_opt_a]: 1.419e-05 [slice_cell_reuse_recomputed_activation]: 2.39999e-06 [rewriter_after_opt_a]: 4.116e-05 [convert_after_rewriter]: 8.38001e-06 [order_py_execute_after_rewriter]: 5.87001e-06 [mutable_eliminate]: 0.00069779 [opt_b]: 0.00021956, [1] [Cycle 1]: 0.00021187, [7] [b_1]: 0.00012839 [b_2]: 8.01001e-06 [updatestate_depend_eliminate]: 7.92e-06 [updatestate_assign_eliminate]: 2.91999e-06 [updatestate_loads_eliminate]: 2.57001e-06 [renormalize]: 7.79983e-07 [cse]: 2.572e-05 [optimize_parallel_all_gather_comm]: 2.275e-05 [overlap_param_gather]: 3.5e-06 [cconv]: 3.264e-05 [loop_unroll]: 0.00046345 [opt_after_cconv]: 0.00010965, [1] [Cycle 1]: 0.00010323, [7] [c_1]: 3.053e-05 [parameter_eliminate]: 4.12e-06 [updatestate_depend_eliminate]: 6.06998e-06 [updatestate_assign_eliminate]: 3.03e-06 [updatestate_loads_eliminate]: 2.62001e-06 [cse]: 2.174e-05 [renormalize]: 3.80009e-07 [remove_dup_value]: 1.744e-05 [tuple_transform]: 7.571e-05, [1] [Cycle 1]: 7.133e-05, [4] [d_1]: 4.459e-05 [none_parameter_eliminate]: 1.71e-06 [renormalize]: 2.3999e-07 [switch_simplify]: 7.02002e-06 [partial_unused_args_eliminate]: 2.20002e-06 [add_recomputation]: 5.887e-05 [cse_after_recomputation]: 2.57e-05, [1] [Cycle 1]: 2.078e-05, [1] [cse]: 1.514e-05 [environ_conv]: 6.79001e-06 [swap_dp_allreduce_reducescatter]: 5.84e-06 [bias_add_comm_swap]: 3.01999e-06 [label_micro_interleaved_index]: 5.05001e-06 [label_fine_grained_interleaved_index]: 2.95998e-06 [merge_cast_opt]: 1.32e-06 [slice_recompute_activation]: 2.32001e-06 [micro_interleaved_order_control]: 2.48002e-06 [assign_add_opt]: 1.20999e-06 [ForceFp32Comm]: 1.18001e-06 [remove_cast_before_assign_add]: 1.35001e-06 [full_micro_interleaved_order_control]: 2.26998e-06 [reorder_send_recv_between_fp_bp]: 2.86999e-06 [comm_op_add_attrs]: 1.08001e-06 [add_comm_op_reuse_tag]: 1.08001e-06 [interleave_split_concat_branches]: 1.25001e-06 [interleave_parallel_branches]: 1.06002e-06 [overlap_opt_shard_in_pipeline]: 2.74001e-06 [overlap_opt_shard_grad_in_pipeline]: 2.12999e-06 [control_data_broadcast_order]: 1.48e-05 [grouped_pairwise_exchange_alltoall]: 1.59e-06 [offloading_packed_experts]: 4.35e-06 [overlap_recompute_and_grad_model_parallel]: 5.14003e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.22e-06 [overlap_recompute_allgather_and_fa_grad]: 1.40999e-06 [overlap_recompute_comm]: 2.80997e-06 [overlap_grad_ring_attention]: 4.94e-06 [overlap_grad_flash_sp]: 2.405e-05 [begin_end_overlap_inline]: 5.8001e-07 [split_matmul_comm_elemetwise]: 2.19999e-06 [split_layernorm_comm]: 1.69998e-06 [handle_group_info]: 1.27e-06 [symbol_engine_optimizer]: 8.067e-05, [1] [Cycle 1]: 7.626e-05, [6] [build]: 4.33999e-06 [elim_shapecalc]: 1.038e-05 [elim_not_effective]: 1.376e-05 [opt_reshape]: 7.46999e-06 [fold_const_symbol]: 1.154e-05 [renormalize]: 1.50001e-07 [detach_backward]: 2.26998e-06 [pipeline_parallel_scheduler]: 1.53002e-06 [auto_monad_reorder]: 2.579e-05 [get_jit_bprop_graph]: 2.73998e-06 [rewriter_after_jit_bprop_graph]: 5.20999e-06 [opt_after_jit_grad]: 0.000543 [validate]: 5.346e-05 Sums bootstrap : 0.000691s : 7.11% type_inference : 0.004201s : 43.19% event_method : 0.000015s : 0.15% auto_monad : 0.000060s : 0.62% graph_reusing : 0.000006s : 0.06% inline : 0.000002s : 0.03% add_attr.add_attr_with_inline.tag_attr : 0.000018s : 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 : 0.000039s : 0.40% 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 : 0.000022s : 0.23% optimize.rewriter_before_opt_a : 0.000057s : 0.59% optimize.opt_a.expand_dump_flag : 0.000004s : 0.04% optimize.opt_a.switch_simplify : 0.000033s : 0.33% optimize.opt_a.loop_unroll : 0.000019s : 0.20% optimize.opt_a.a_1 : 0.000444s : 4.56% optimize.opt_a.with_stream_mark : 0.000038s : 0.39% optimize.opt_a.recompute_prepare : 0.000015s : 0.16% optimize.opt_a.updatestate_depend_eliminate : 0.000009s : 0.09% 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 : 0.000160s : 1.64% 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 : 0.000013s : 0.14% optimize.opt_a.merge_send_recv : 0.000014s : 0.14% optimize.opt_a.auto_parallel : 0.000015s : 0.16% optimize.opt_a.parallel : 0.000041s : 0.42% optimize.opt_a.flash_sp : 0.000013s : 0.13% 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 : 0.000026s : 0.27% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000017s : 0.17% optimize.opt_a.virtual_dataset : 0.000014s : 0.14% 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 : 0.000030s : 0.31% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.03% optimize.opt_a.before_grad : 0.000023s : 0.23% 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 : 0.000004s : 0.04% optimize.opt_a.receive_attached : 0.000004s : 0.04% optimize.opt_a.after_resolve : 0.000024s : 0.24% optimize.opt_a.a_after_grad : 0.000020s : 0.21% optimize.opt_a.renormalize : 0.000826s : 8.49% optimize.opt_a.add_forward_monad_depend : 0.000008s : 0.08% optimize.opt_a.auto_monad_grad : 0.000004s : 0.05% optimize.opt_a.auto_monad_eliminator : 0.000025s : 0.25% optimize.opt_a.cse : 0.000058s : 0.60% optimize.opt_a.a_3 : 0.000090s : 0.93% optimize.py_interpret_to_execute_after_opt_a : 0.000014s : 0.15% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.02% optimize.rewriter_after_opt_a : 0.000041s : 0.42% optimize.convert_after_rewriter : 0.000008s : 0.09% optimize.order_py_execute_after_rewriter : 0.000006s : 0.06% optimize.mutable_eliminate : 0.000698s : 7.17% optimize.opt_b.b_1 : 0.000128s : 1.32% optimize.opt_b.b_2 : 0.000008s : 0.08% 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 : 0.000026s : 0.26% optimize.optimize_parallel_all_gather_comm : 0.000023s : 0.23% optimize.overlap_param_gather : 0.000003s : 0.04% optimize.cconv : 0.000033s : 0.34% optimize.loop_unroll : 0.000463s : 4.76% optimize.opt_after_cconv.c_1 : 0.000031s : 0.31% 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 : 0.000003s : 0.03% optimize.opt_after_cconv.cse : 0.000022s : 0.22% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000017s : 0.18% optimize.tuple_transform.d_1 : 0.000045s : 0.46% 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 : 0.07% optimize.partial_unused_args_eliminate : 0.000002s : 0.02% optimize.add_recomputation : 0.000059s : 0.61% optimize.cse_after_recomputation.cse : 0.000015s : 0.16% optimize.environ_conv : 0.000007s : 0.07% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.06% 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 : 0.000001s : 0.01% 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 : 0.000003s : 0.03% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.02% optimize.control_data_broadcast_order : 0.000015s : 0.15% 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 : 0.000005s : 0.05% optimize.overlap_grad_flash_sp : 0.000024s : 0.25% optimize.begin_end_overlap_inline : 0.000001s : 0.01% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.02% optimize.split_layernorm_comm : 0.000002s : 0.02% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000004s : 0.04% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000010s : 0.11% optimize.symbol_engine_optimizer.elim_not_effective : 0.000014s : 0.14% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.08% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000012s : 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 : 0.000026s : 0.27% get_jit_bprop_graph : 0.000003s : 0.03% rewriter_after_jit_bprop_graph : 0.000005s : 0.05% opt_after_jit_grad : 0.000543s : 5.58% validate : 0.000053s : 0.55% Time group info: ------[substitution.] 0.000118 21 1.92% : 0.000002s : 2: substitution.elim_not_effective 1.43% : 0.000002s : 2: substitution.fold_const_symbol 5.70% : 0.000007s : 4: substitution.graph_param_transform 78.98% : 0.000093s : 1: substitution.inline 3.76% : 0.000004s : 4: substitution.j_node_and_user_rematch 4.28% : 0.000005s : 4: substitution.remove_not_recompute_node 3.92% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.004136 2 88.78% : 0.003672s : 1: type_inference.infer 11.22% : 0.000464s : 1: type_inference.specialize ------[replace.] 0.000018 1 100.00% : 0.000018s : 1: replace.inline ------[match.] 0.000091 1 100.00% : 0.000091s : 1: match.inline ------[predicate.] 0.000157 967 0.74% : 0.000001s : 9: predicate.accumulaten_eliminater 1.36% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.62% : 0.000001s : 8: predicate.addn_check_dump 0.73% : 0.000001s : 9: predicate.addn_zero_filter 0.63% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.11% : 0.000003s : 17: predicate.arithmetic_simplify 0.85% : 0.000001s : 9: predicate.cast_eliminate 0.84% : 0.000001s : 8: predicate.check_bprop_eliminate 0.62% : 0.000001s : 8: predicate.compare_switch_simplify 0.25% : 0.000000s : 4: predicate.const_output_eliminate 0.72% : 0.000001s : 8: predicate.depend_value_elim 0.80% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.85% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.78% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.67% : 0.000003s : 8: predicate.dumpgradient_eliminate 0.27% : 0.000000s : 4: predicate.elim_not_effective 0.53% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.20% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.97% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.98% : 0.000002s : 13: predicate.environ_get_depend_swap 1.81% : 0.000003s : 21: predicate.environ_get_eliminate 1.00% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.75% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.62% : 0.000003s : 10: predicate.float_depend_g_call 0.69% : 0.000001s : 8: predicate.float_environ_get_switch 0.92% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.25% : 0.000000s : 4: predicate.fold_const_symbol 0.87% : 0.000001s : 8: predicate.get_grad_eliminate 0.29% : 0.000000s : 4: predicate.graph_param_transform 0.68% : 0.000001s : 8: predicate.incorporate_call 0.65% : 0.000001s : 8: predicate.incorporate_call_switch 6.55% : 0.000010s : 43: predicate.inline 1.01% : 0.000002s : 8: predicate.inline_without_move 0.43% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.99% : 0.000002s : 8: predicate.less_batch_normalization 1.69% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.12% : 0.000003s : 26: predicate.load_eliminater 1.17% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.40% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.85% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.66% : 0.000001s : 8: predicate.merge_addn 0.76% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.76% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.66% : 0.000001s : 9: predicate.minmaximum_grad 1.94% : 0.000003s : 4: predicate.mutable_eliminate 0.43% : 0.000001s : 4: predicate.opt_reshape 0.45% : 0.000001s : 4: predicate.parallel_virtual_node 1.33% : 0.000002s : 10: predicate.partial_defer_inline 1.17% : 0.000002s : 13: predicate.partial_eliminate 0.81% : 0.000001s : 9: predicate.print_const_string_wrapper 0.76% : 0.000001s : 8: predicate.reduce_all_const_elim 1.36% : 0.000002s : 9: predicate.reduce_eliminate 2.01% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.66% : 0.000001s : 8: predicate.remove_not_recompute_node 1.27% : 0.000002s : 17: predicate.replace_applicator 0.89% : 0.000001s : 8: predicate.replace_old_param 0.59% : 0.000001s : 4: predicate.reset_defer_inline 0.83% : 0.000001s : 9: predicate.reshape_eliminate 0.80% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.55% : 0.000001s : 4: predicate.row_tensor_eliminate 1.04% : 0.000002s : 8: predicate.same_eliminate 0.57% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.04% : 0.000002s : 8: predicate.shard_identity_eliminate 1.01% : 0.000002s : 8: predicate.special_op_eliminate 0.79% : 0.000001s : 8: predicate.specialize_transform 1.40% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.90% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.41% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.90% : 0.000001s : 10: predicate.switch_defer_inline 1.57% : 0.000002s : 18: predicate.switch_layer_defer_inline 3.86% : 0.000006s : 35: predicate.switch_simplify 0.75% : 0.000001s : 9: predicate.tile_eliminate 0.88% : 0.000001s : 9: predicate.transpose_eliminate 1.45% : 0.000002s : 17: predicate.tuple_list_convert_item_index_to_positive 1.74% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.33% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.28% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.34% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.51% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.98% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.97% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.94% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.46% : 0.000001s : 4: predicate.value_based_eliminate 0.93% : 0.000001s : 8: predicate.virtual_dataset_eliminate 1.05% : 0.000002s : 8: predicate.virtual_output_eliminate 0.34% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.58% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000274 4 8.09% : 0.000022s : 1: func_graph_cloner_run.FuncGraphClonerGraph 91.91% : 0.000252s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.075519 192 0.01% : 0.000004s : 1: ForceFp32Comm 38.45% : 0.029035s : 1: add_attr 38.42% : 0.029017s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.08% : 0.000063s : 1: add_recomputation 0.01% : 0.000005s : 1: assign_add_opt 0.09% : 0.000066s : 1: auto_monad 0.04% : 0.000031s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.01% : 0.000006s : 1: bias_add_comm_swap 0.97% : 0.000734s : 1: bootstrap 0.05% : 0.000036s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.02% : 0.000019s : 1: control_data_broadcast_order 0.02% : 0.000012s : 1: convert_after_rewriter 0.04% : 0.000029s : 1: cse_after_recomputation 0.01% : 0.000005s : 1: dataset_repeat_opt 0.01% : 0.000006s : 1: detach_backward 0.01% : 0.000010s : 1: environ_conv 0.03% : 0.000022s : 1: event_method 0.01% : 0.000005s : 1: full_micro_interleaved_order_control 0.01% : 0.000006s : 1: get_jit_bprop_graph 0.01% : 0.000009s : 1: graph_reusing 0.01% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000004s : 1: handle_group_info 0.01% : 0.000005s : 1: inline 0.01% : 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% : 0.000006s : 1: label_fine_grained_interleaved_index 0.01% : 0.000008s : 1: label_micro_interleaved_index 0.63% : 0.000473s : 1: loop_unroll 0.01% : 0.000004s : 1: merge_cast_opt 0.01% : 0.000005s : 1: micro_interleaved_order_control 0.94% : 0.000709s : 1: mutable_eliminate 0.01% : 0.000008s : 1: offloading_packed_experts 0.02% : 0.000014s : 1: opt.transform.loop_unroll_optimizer 0.03% : 0.000020s : 1: opt.transform.mutable_eliminate 1.13% : 0.000853s : 78: opt.transform.opt_a 0.04% : 0.000029s : 1: opt.transform.opt_after_cconv 0.04% : 0.000028s : 1: opt.transform.opt_after_jit_grad 0.14% : 0.000109s : 28: opt.transform.opt_b 0.07% : 0.000050s : 2: opt.transform.opt_trans_graph 0.05% : 0.000039s : 4: opt.transform.symbol_engine_opt 3.42% : 0.002586s : 1: opt_a 0.15% : 0.000113s : 1: opt_after_cconv 0.73% : 0.000555s : 1: opt_after_jit_grad 0.30% : 0.000223s : 1: opt_b 6.50% : 0.004911s : 1: optimize 0.04% : 0.000027s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000009s : 1: order_py_execute_after_rewriter 0.04% : 0.000028s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000008s : 1: overlap_grad_ring_attention 0.01% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.01% : 0.000006s : 1: overlap_opt_shard_in_pipeline 0.01% : 0.000007s : 1: overlap_param_gather 0.01% : 0.000010s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.01% : 0.000006s : 1: overlap_recompute_comm 0.01% : 0.000009s : 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 0.06% : 0.000044s : 1: pre_auto_parallel 0.04% : 0.000027s : 1: py_interpret_to_execute 0.02% : 0.000018s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000004s : 1: remove_cast_before_assign_add 0.03% : 0.000022s : 1: remove_dup_value 0.67% : 0.000505s : 1: renormalize.infer 0.41% : 0.000313s : 1: renormalize.specialize 0.01% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.06% : 0.000046s : 1: rewriter_after_opt_a 0.08% : 0.000062s : 1: rewriter_before_opt_a 0.01% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.01% : 0.000006s : 1: slice_recompute_activation 0.01% : 0.000004s : 1: split_layernorm_comm 0.01% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.11% : 0.000083s : 1: symbol_engine_optimizer 0.10% : 0.000079s : 1: tuple_transform 5.59% : 0.004225s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[dim_type1-True-batch_size_list1-shape1-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_fp16[dim_type1-True-batch_size_list1-shape1-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:39:43.729.856 [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 = 0.121457, [21] [bootstrap]: 0.00052182 [type_inference]: 0.108843 [event_method]: 1.723e-05 [auto_monad]: 6.495e-05 [graph_reusing]: 5.56e-06 [inline]: 3.50998e-06 [add_attr]: 0.00521876, [1] [add_attr_with_inline]: 0.00519988, [1] [Cycle 1]: 7.613e-05, [2] [tag_attr]: 2.13e-05 [meta_addattr_fg_expand]: 3.76999e-06 [parallel-infer-symbol]: 4.08999e-06 [pre_auto_parallel]: 4.061e-05 [insert-virtual-dataset]: 2.79001e-06 [parallel-infer-symbol-second]: 1.07998e-06 [dataset_repeat_opt]: 1.94e-06 [pipeline_split]: 1.72001e-06 [optimize]: 0.00560406, [53] [py_interpret_to_execute]: 2.838e-05 [rewriter_before_opt_a]: 5.698e-05 [opt_a]: 0.00275861, [2] [Cycle 1]: 0.00198013, [45] [expand_dump_flag]: 2.99999e-06 [switch_simplify]: 2.622e-05 [loop_unroll]: 1.233e-05 [a_1]: 0.00032826 [with_stream_mark]: 2.656e-05 [recompute_prepare]: 1.043e-05 [updatestate_depend_eliminate]: 4.03001e-06 [updatestate_assign_eliminate]: 3.66001e-06 [updatestate_loads_eliminate]: 3.89002e-06 [parameter_eliminate]: 2.36998e-06 [a_2]: 8.169e-05 [accelerated_algorithm]: 7.7e-06 [shard]: 2.88e-06 [meta_shard_fg_expand]: 2.43998e-06 [shard_inline]: 7.75e-06 [merge_send_recv]: 1.04e-05 [auto_parallel]: 7.85e-06 [parallel]: 3.123e-05 [flash_sp]: 1.23e-05 [merge_comm]: 4.67e-06 [allreduce_fusion]: 4.03001e-06 [matmul_add_comm_reduction]: 1.096e-05 [allreduce_slice_to_reducescatter]: 9.89996e-07 [virtual_shard_identity]: 1.07e-05 [virtual_dataset]: 6.88998e-06 [get_grad_eliminate_]: 7.31999e-06 [virtual_output]: 6.54999e-06 [merge_forward]: 5.24e-06 [cell_reuse_recompute_pass]: 1.81e-06 [offload_activation]: 1.115e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.766e-05 [merge_recompute_call_nodes]: 1.82999e-06 [before_grad]: 1.225e-05 [set_forward_comm_id_for_comm_node_pass]: 3.97002e-06 [meta_fg_expand]: 3.41001e-06 [flash_sp_send_recv_attached]: 3.35e-06 [receive_attached]: 2.44999e-06 [after_resolve]: 1.493e-05 [a_after_grad]: 1.08e-05 [renormalize]: 0.00080833 [add_forward_monad_depend]: 9.59e-06 [auto_monad_grad]: 3.08998e-06 [auto_monad_eliminator]: 2.165e-05 [cse]: 4.162e-05 [a_3]: 6.002e-05 [Cycle 2]: 0.00076327, [45] [expand_dump_flag]: 2.38998e-06 [switch_simplify]: 9.05999e-06 [loop_unroll]: 6.36e-06 [a_1]: 0.00013511 [with_stream_mark]: 2.134e-05 [recompute_prepare]: 7.7e-06 [updatestate_depend_eliminate]: 4.28999e-06 [updatestate_assign_eliminate]: 3.21001e-06 [updatestate_loads_eliminate]: 3.28e-06 [parameter_eliminate]: 1.57999e-06 [a_2]: 7.459e-05 [accelerated_algorithm]: 7.01999e-06 [shard]: 2.21e-06 [meta_shard_fg_expand]: 2.34001e-06 [shard_inline]: 6.41998e-06 [merge_send_recv]: 7.88999e-06 [auto_parallel]: 1.008e-05 [parallel]: 9.52001e-06 [flash_sp]: 5.30001e-06 [merge_comm]: 3.76001e-06 [allreduce_fusion]: 3.38e-06 [matmul_add_comm_reduction]: 1.198e-05 [allreduce_slice_to_reducescatter]: 6.50005e-07 [virtual_shard_identity]: 9.89001e-06 [virtual_dataset]: 7.35998e-06 [get_grad_eliminate_]: 6.66999e-06 [virtual_output]: 5.94e-06 [merge_forward]: 5.04e-06 [cell_reuse_recompute_pass]: 2.64001e-06 [offload_activation]: 1.154e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.034e-05 [merge_recompute_call_nodes]: 1.23002e-06 [before_grad]: 1.203e-05 [set_forward_comm_id_for_comm_node_pass]: 5.08002e-06 [meta_fg_expand]: 3.35003e-06 [flash_sp_send_recv_attached]: 1.72001e-06 [receive_attached]: 2.08002e-06 [after_resolve]: 1.245e-05 [a_after_grad]: 1.015e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 2.78e-06 [auto_monad_grad]: 1.67999e-06 [auto_monad_eliminator]: 9.17999e-06 [cse]: 2.668e-05 [a_3]: 3.843e-05 [py_interpret_to_execute_after_opt_a]: 2.191e-05 [slice_cell_reuse_recomputed_activation]: 3.33998e-06 [rewriter_after_opt_a]: 5.093e-05 [convert_after_rewriter]: 8.51002e-06 [order_py_execute_after_rewriter]: 6.59999e-06 [mutable_eliminate]: 0.00083917 [opt_b]: 0.00026191, [1] [Cycle 1]: 0.00025152, [7] [b_1]: 0.00013862 [b_2]: 9.57001e-06 [updatestate_depend_eliminate]: 1.206e-05 [updatestate_assign_eliminate]: 3.74002e-06 [updatestate_loads_eliminate]: 3.18998e-06 [renormalize]: 1.20999e-06 [cse]: 4.381e-05 [optimize_parallel_all_gather_comm]: 2.664e-05 [overlap_param_gather]: 2.37999e-06 [cconv]: 4.11e-05 [loop_unroll]: 0.00064065 [opt_after_cconv]: 0.00014034, [1] [Cycle 1]: 0.00013069, [7] [c_1]: 3.404e-05 [parameter_eliminate]: 6.49999e-06 [updatestate_depend_eliminate]: 9.93002e-06 [updatestate_assign_eliminate]: 3.56001e-06 [updatestate_loads_eliminate]: 2.56e-06 [cse]: 3.601e-05 [renormalize]: 9.29984e-07 [remove_dup_value]: 1.984e-05 [tuple_transform]: 9.215e-05, [1] [Cycle 1]: 8.703e-05, [4] [d_1]: 5.626e-05 [none_parameter_eliminate]: 2.04e-06 [renormalize]: 1.69995e-07 [switch_simplify]: 7.11001e-06 [partial_unused_args_eliminate]: 2.37999e-06 [add_recomputation]: 6.314e-05 [cse_after_recomputation]: 4.256e-05, [1] [Cycle 1]: 3.667e-05, [1] [cse]: 2.987e-05 [environ_conv]: 8.90001e-06 [swap_dp_allreduce_reducescatter]: 6.53e-06 [bias_add_comm_swap]: 3.92998e-06 [label_micro_interleaved_index]: 7.39002e-06 [label_fine_grained_interleaved_index]: 3.06999e-06 [merge_cast_opt]: 1.57001e-06 [slice_recompute_activation]: 2.77002e-06 [micro_interleaved_order_control]: 2.38998e-06 [assign_add_opt]: 1.57999e-06 [ForceFp32Comm]: 1.44998e-06 [remove_cast_before_assign_add]: 1.52001e-06 [full_micro_interleaved_order_control]: 2.54999e-06 [reorder_send_recv_between_fp_bp]: 3.13e-06 [comm_op_add_attrs]: 1.39e-06 [add_comm_op_reuse_tag]: 1.04e-06 [interleave_split_concat_branches]: 1.59e-06 [interleave_parallel_branches]: 1.10999e-06 [overlap_opt_shard_in_pipeline]: 1.32999e-06 [overlap_opt_shard_grad_in_pipeline]: 1.92999e-06 [control_data_broadcast_order]: 1.724e-05 [grouped_pairwise_exchange_alltoall]: 1.69e-06 [offloading_packed_experts]: 4.74e-06 [overlap_recompute_and_grad_model_parallel]: 5.74999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.59e-06 [overlap_recompute_allgather_and_fa_grad]: 1.69e-06 [overlap_recompute_comm]: 2.22001e-06 [overlap_grad_ring_attention]: 4.15e-06 [overlap_grad_flash_sp]: 2.605e-05 [begin_end_overlap_inline]: 5.3001e-07 [split_matmul_comm_elemetwise]: 2.14999e-06 [split_layernorm_comm]: 2.22999e-06 [handle_group_info]: 1.10001e-06 [symbol_engine_optimizer]: 9.187e-05, [1] [Cycle 1]: 8.661e-05, [6] [build]: 4.34002e-06 [elim_shapecalc]: 1.346e-05 [elim_not_effective]: 1.559e-05 [opt_reshape]: 7.4e-06 [fold_const_symbol]: 1.011e-05 [renormalize]: 1.60013e-07 [detach_backward]: 2.29999e-06 [pipeline_parallel_scheduler]: 1.59998e-06 [auto_monad_reorder]: 2.868e-05 [get_jit_bprop_graph]: 1.74998e-06 [rewriter_after_jit_bprop_graph]: 9.05999e-06 [opt_after_jit_grad]: 0.00078974 [validate]: 6.33e-05 Sums bootstrap : 0.000522s : 0.45% type_inference : 0.108843s : 94.62% event_method : 0.000017s : 0.01% auto_monad : 0.000065s : 0.06% graph_reusing : 0.000006s : 0.00% inline : 0.000004s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000021s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000041s : 0.04% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000028s : 0.02% optimize.rewriter_before_opt_a : 0.000057s : 0.05% optimize.opt_a.expand_dump_flag : 0.000005s : 0.00% optimize.opt_a.switch_simplify : 0.000035s : 0.03% optimize.opt_a.loop_unroll : 0.000019s : 0.02% optimize.opt_a.a_1 : 0.000463s : 0.40% optimize.opt_a.with_stream_mark : 0.000048s : 0.04% optimize.opt_a.recompute_prepare : 0.000018s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.00% optimize.opt_a.a_2 : 0.000156s : 0.14% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.01% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000005s : 0.00% optimize.opt_a.shard_inline : 0.000014s : 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 : 0.000041s : 0.04% optimize.opt_a.flash_sp : 0.000018s : 0.02% optimize.opt_a.merge_comm : 0.000008s : 0.01% optimize.opt_a.allreduce_fusion : 0.000007s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000023s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000021s : 0.02% optimize.opt_a.virtual_dataset : 0.000014s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000014s : 0.01% optimize.opt_a.virtual_output : 0.000012s : 0.01% optimize.opt_a.merge_forward : 0.000010s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.00% optimize.opt_a.offload_activation : 0.000023s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000038s : 0.03% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000024s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000009s : 0.01% optimize.opt_a.meta_fg_expand : 0.000007s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000027s : 0.02% optimize.opt_a.a_after_grad : 0.000021s : 0.02% optimize.opt_a.renormalize : 0.000808s : 0.70% optimize.opt_a.add_forward_monad_depend : 0.000012s : 0.01% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000031s : 0.03% optimize.opt_a.cse : 0.000068s : 0.06% optimize.opt_a.a_3 : 0.000098s : 0.09% optimize.py_interpret_to_execute_after_opt_a : 0.000022s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000051s : 0.04% optimize.convert_after_rewriter : 0.000009s : 0.01% optimize.order_py_execute_after_rewriter : 0.000007s : 0.01% optimize.mutable_eliminate : 0.000839s : 0.73% optimize.opt_b.b_1 : 0.000139s : 0.12% optimize.opt_b.b_2 : 0.000010s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000012s : 0.01% optimize.opt_b.updatestate_assign_eliminate : 0.000004s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000044s : 0.04% optimize.optimize_parallel_all_gather_comm : 0.000027s : 0.02% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000041s : 0.04% optimize.loop_unroll : 0.000641s : 0.56% optimize.opt_after_cconv.c_1 : 0.000034s : 0.03% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.01% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000010s : 0.01% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000004s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000003s : 0.00% optimize.opt_after_cconv.cse : 0.000036s : 0.03% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000020s : 0.02% optimize.tuple_transform.d_1 : 0.000056s : 0.05% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000063s : 0.05% optimize.cse_after_recomputation.cse : 0.000030s : 0.03% optimize.environ_conv : 0.000009s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000007s : 0.01% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000007s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000002s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000002s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000017s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000005s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000006s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000004s : 0.00% optimize.overlap_grad_flash_sp : 0.000026s : 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 : 0.000013s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000016s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000029s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000009s : 0.01% opt_after_jit_grad : 0.000790s : 0.69% validate : 0.000063s : 0.06% Time group info: ------[substitution.] 0.000134 21 1.75% : 0.000002s : 2: substitution.elim_not_effective 1.07% : 0.000001s : 2: substitution.fold_const_symbol 5.11% : 0.000007s : 4: substitution.graph_param_transform 76.90% : 0.000103s : 1: substitution.inline 4.63% : 0.000006s : 4: substitution.j_node_and_user_rematch 5.17% : 0.000007s : 4: substitution.remove_not_recompute_node 5.37% : 0.000007s : 4: substitution.replace_old_param ------[type_inference.] 0.108765 2 2.89% : 0.003146s : 1: type_inference.infer 97.11% : 0.105619s : 1: type_inference.specialize ------[replace.] 0.000020 1 100.00% : 0.000020s : 1: replace.inline ------[match.] 0.000102 1 100.00% : 0.000102s : 1: match.inline ------[predicate.] 0.000175 967 0.81% : 0.000001s : 9: predicate.accumulaten_eliminater 1.99% : 0.000003s : 4: predicate.ad_related_special_op_eliminate 0.64% : 0.000001s : 8: predicate.addn_check_dump 0.83% : 0.000001s : 9: predicate.addn_zero_filter 0.65% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.18% : 0.000004s : 17: predicate.arithmetic_simplify 0.88% : 0.000002s : 9: predicate.cast_eliminate 0.75% : 0.000001s : 8: predicate.check_bprop_eliminate 0.65% : 0.000001s : 8: predicate.compare_switch_simplify 0.22% : 0.000000s : 4: predicate.const_output_eliminate 0.69% : 0.000001s : 8: predicate.depend_value_elim 0.77% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.88% : 0.000002s : 9: predicate.dict_get_item_eliminator 0.71% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.93% : 0.000003s : 8: predicate.dumpgradient_eliminate 0.22% : 0.000000s : 4: predicate.elim_not_effective 0.48% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.09% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.20% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.98% : 0.000002s : 13: predicate.environ_get_depend_swap 1.63% : 0.000003s : 21: predicate.environ_get_eliminate 0.99% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.76% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.46% : 0.000003s : 10: predicate.float_depend_g_call 0.64% : 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 0.85% : 0.000002s : 8: predicate.get_grad_eliminate 0.35% : 0.000001s : 4: predicate.graph_param_transform 0.63% : 0.000001s : 8: predicate.incorporate_call 0.54% : 0.000001s : 8: predicate.incorporate_call_switch 5.90% : 0.000010s : 43: predicate.inline 1.01% : 0.000002s : 8: predicate.inline_without_move 0.39% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.30% : 0.000002s : 8: predicate.less_batch_normalization 1.53% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.09% : 0.000004s : 26: predicate.load_eliminater 1.62% : 0.000003s : 4: predicate.loop_unroll_after_grad 1.28% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.76% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.72% : 0.000001s : 8: predicate.merge_addn 0.64% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.70% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.67% : 0.000001s : 9: predicate.minmaximum_grad 2.57% : 0.000005s : 4: predicate.mutable_eliminate 0.50% : 0.000001s : 4: predicate.opt_reshape 0.52% : 0.000001s : 4: predicate.parallel_virtual_node 1.14% : 0.000002s : 10: predicate.partial_defer_inline 1.06% : 0.000002s : 13: predicate.partial_eliminate 0.96% : 0.000002s : 9: predicate.print_const_string_wrapper 0.76% : 0.000001s : 8: predicate.reduce_all_const_elim 1.07% : 0.000002s : 9: predicate.reduce_eliminate 2.28% : 0.000004s : 26: predicate.redundant_stop_gradient_eliminater 0.72% : 0.000001s : 8: predicate.remove_not_recompute_node 1.13% : 0.000002s : 17: predicate.replace_applicator 0.62% : 0.000001s : 8: predicate.replace_old_param 0.30% : 0.000001s : 4: predicate.reset_defer_inline 1.04% : 0.000002s : 9: predicate.reshape_eliminate 1.07% : 0.000002s : 8: predicate.row_tensor_add_zeros_like 0.67% : 0.000001s : 4: predicate.row_tensor_eliminate 0.95% : 0.000002s : 8: predicate.same_eliminate 0.48% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.97% : 0.000002s : 8: predicate.shard_identity_eliminate 0.97% : 0.000002s : 8: predicate.special_op_eliminate 0.88% : 0.000002s : 8: predicate.specialize_transform 1.46% : 0.000003s : 8: predicate.split_environ_get_set_with_tuple_value 0.84% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.38% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.84% : 0.000001s : 10: predicate.switch_defer_inline 1.51% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.58% : 0.000006s : 35: predicate.switch_simplify 0.84% : 0.000001s : 9: predicate.tile_eliminate 0.80% : 0.000001s : 9: predicate.transpose_eliminate 1.76% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.42% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.50% : 0.000003s : 17: predicate.tuple_list_get_item_depend_reorder 3.16% : 0.000006s : 25: predicate.tuple_list_get_item_eliminator 1.69% : 0.000003s : 17: predicate.tuple_list_get_set_item_eliminator 2.80% : 0.000005s : 25: predicate.tuple_list_set_item_eliminator 1.60% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.89% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.79% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.56% : 0.000001s : 4: predicate.value_based_eliminate 0.91% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.88% : 0.000002s : 8: predicate.virtual_output_eliminate 0.28% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.58% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000359 4 7.30% : 0.000026s : 1: func_graph_cloner_run.FuncGraphClonerGraph 92.70% : 0.000333s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.134120 192 0.00% : 0.000004s : 1: ForceFp32Comm 3.90% : 0.005226s : 1: add_attr 3.88% : 0.005205s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.05% : 0.000069s : 1: add_recomputation 0.01% : 0.000011s : 1: assign_add_opt 0.05% : 0.000070s : 1: auto_monad 0.03% : 0.000035s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.01% : 0.000007s : 1: bias_add_comm_swap 0.42% : 0.000567s : 1: bootstrap 0.03% : 0.000045s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.02% : 0.000021s : 1: control_data_broadcast_order 0.01% : 0.000012s : 1: convert_after_rewriter 0.03% : 0.000046s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.01% : 0.000007s : 1: detach_backward 0.01% : 0.000012s : 1: environ_conv 0.02% : 0.000026s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.01% : 0.000009s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.01% : 0.000010s : 1: label_micro_interleaved_index 0.49% : 0.000657s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.64% : 0.000859s : 1: mutable_eliminate 0.01% : 0.000008s : 1: offloading_packed_experts 0.02% : 0.000020s : 1: opt.transform.loop_unroll_optimizer 0.02% : 0.000030s : 1: opt.transform.mutable_eliminate 0.66% : 0.000892s : 78: opt.transform.opt_a 0.02% : 0.000032s : 1: opt.transform.opt_after_cconv 0.03% : 0.000040s : 1: opt.transform.opt_after_jit_grad 0.08% : 0.000114s : 28: opt.transform.opt_b 0.05% : 0.000061s : 2: opt.transform.opt_trans_graph 0.03% : 0.000042s : 4: opt.transform.symbol_engine_opt 2.06% : 0.002762s : 1: opt_a 0.11% : 0.000144s : 1: opt_after_cconv 0.61% : 0.000813s : 1: opt_after_jit_grad 0.20% : 0.000266s : 1: opt_b 4.18% : 0.005610s : 1: optimize 0.02% : 0.000031s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000011s : 1: order_py_execute_after_rewriter 0.02% : 0.000031s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000008s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000005s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000009s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.01% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.03% : 0.000046s : 1: pre_auto_parallel 0.02% : 0.000033s : 1: py_interpret_to_execute 0.02% : 0.000027s : 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 0.36% : 0.000481s : 1: renormalize.infer 0.24% : 0.000315s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000013s : 1: rewriter_after_jit_bprop_graph 0.04% : 0.000057s : 1: rewriter_after_opt_a 0.05% : 0.000062s : 1: rewriter_before_opt_a 0.00% : 0.000007s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000010s : 1: swap_dp_allreduce_reducescatter 0.07% : 0.000095s : 1: symbol_engine_optimizer 0.07% : 0.000095s : 1: tuple_transform 81.18% : 0.108874s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[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_fp16[dim_type1-True-batch_size_list2-shape0-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:39:46.252.062 [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 = 0.126576, [21] [bootstrap]: 0.00067076 [type_inference]: 0.114583 [event_method]: 1.822e-05 [auto_monad]: 6.334e-05 [graph_reusing]: 5.12999e-06 [inline]: 3.25e-06 [add_attr]: 0.00527208, [1] [add_attr_with_inline]: 0.00525571, [1] [Cycle 1]: 7.032e-05, [2] [tag_attr]: 1.92e-05 [meta_addattr_fg_expand]: 3.8e-06 [parallel-infer-symbol]: 4.16001e-06 [pre_auto_parallel]: 3.966e-05 [insert-virtual-dataset]: 2.86999e-06 [parallel-infer-symbol-second]: 8.70001e-07 [dataset_repeat_opt]: 1.053e-05 [pipeline_split]: 1.67999e-06 [optimize]: 0.00509967, [53] [py_interpret_to_execute]: 2.496e-05 [rewriter_before_opt_a]: 5.417e-05 [opt_a]: 0.00277687, [2] [Cycle 1]: 0.00201961, [45] [expand_dump_flag]: 2.75997e-06 [switch_simplify]: 2.463e-05 [loop_unroll]: 1.138e-05 [a_1]: 0.00035193 [with_stream_mark]: 2.697e-05 [recompute_prepare]: 1.154e-05 [updatestate_depend_eliminate]: 4.29002e-06 [updatestate_assign_eliminate]: 3.52002e-06 [updatestate_loads_eliminate]: 3.33e-06 [parameter_eliminate]: 2.45002e-06 [a_2]: 8.079e-05 [accelerated_algorithm]: 8.43999e-06 [shard]: 2.49999e-06 [meta_shard_fg_expand]: 2.12999e-06 [shard_inline]: 7.1e-06 [merge_send_recv]: 9.92001e-06 [auto_parallel]: 8.02998e-06 [parallel]: 3.236e-05 [flash_sp]: 1.072e-05 [merge_comm]: 5.03002e-06 [allreduce_fusion]: 3.68e-06 [matmul_add_comm_reduction]: 1.071e-05 [allreduce_slice_to_reducescatter]: 6.19999e-07 [virtual_shard_identity]: 1.036e-05 [virtual_dataset]: 6.43998e-06 [get_grad_eliminate_]: 6.24999e-06 [virtual_output]: 1.362e-05 [merge_forward]: 6.84001e-06 [cell_reuse_recompute_pass]: 1.73002e-06 [offload_activation]: 1.23e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.892e-05 [merge_recompute_call_nodes]: 2.17001e-06 [before_grad]: 1.196e-05 [set_forward_comm_id_for_comm_node_pass]: 5.37999e-06 [meta_fg_expand]: 2.59999e-06 [flash_sp_send_recv_attached]: 3.80998e-06 [receive_attached]: 2.44999e-06 [after_resolve]: 1.42e-05 [a_after_grad]: 1.077e-05 [renormalize]: 0.00074729 [add_forward_monad_depend]: 6.45997e-06 [auto_monad_grad]: 2.63e-06 [auto_monad_eliminator]: 1.857e-05 [cse]: 4.048e-05 [a_3]: 5.427e-05 [Cycle 2]: 0.00074506, [45] [expand_dump_flag]: 2.23998e-06 [switch_simplify]: 9.86e-06 [loop_unroll]: 6.41e-06 [a_1]: 0.0001241 [with_stream_mark]: 1.538e-05 [recompute_prepare]: 6.65002e-06 [updatestate_depend_eliminate]: 4e-06 [updatestate_assign_eliminate]: 3.60998e-06 [updatestate_loads_eliminate]: 2.68e-06 [parameter_eliminate]: 1.57001e-06 [a_2]: 7.132e-05 [accelerated_algorithm]: 6.02999e-06 [shard]: 1.25999e-06 [meta_shard_fg_expand]: 2.016e-05 [shard_inline]: 9.92999e-06 [merge_send_recv]: 7.68001e-06 [auto_parallel]: 9.00001e-06 [parallel]: 7.64002e-06 [flash_sp]: 3.93001e-06 [merge_comm]: 3.63e-06 [allreduce_fusion]: 3.35998e-06 [matmul_add_comm_reduction]: 8.71002e-06 [allreduce_slice_to_reducescatter]: 5.50004e-07 [virtual_shard_identity]: 8e-06 [virtual_dataset]: 6.14001e-06 [get_grad_eliminate_]: 5.79e-06 [virtual_output]: 5.52001e-06 [merge_forward]: 4.18001e-06 [cell_reuse_recompute_pass]: 1.87001e-06 [offload_activation]: 8.74e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.549e-05 [merge_recompute_call_nodes]: 1.10999e-06 [before_grad]: 9.47001e-06 [set_forward_comm_id_for_comm_node_pass]: 3.89002e-06 [meta_fg_expand]: 2.96999e-06 [flash_sp_send_recv_attached]: 1.29e-06 [receive_attached]: 1.50001e-06 [after_resolve]: 1.135e-05 [a_after_grad]: 9.10999e-06 [renormalize]: 1.00001e-07 [add_forward_monad_depend]: 1.87001e-06 [auto_monad_grad]: 1.55001e-06 [auto_monad_eliminator]: 9.42999e-06 [cse]: 2.284e-05 [a_3]: 3.946e-05 [py_interpret_to_execute_after_opt_a]: 1.508e-05 [slice_cell_reuse_recomputed_activation]: 2.22999e-06 [rewriter_after_opt_a]: 5.171e-05 [convert_after_rewriter]: 9.02999e-06 [order_py_execute_after_rewriter]: 5.89e-06 [mutable_eliminate]: 0.00066814 [opt_b]: 0.00022237, [1] [Cycle 1]: 0.00021424, [7] [b_1]: 0.00012474 [b_2]: 7.77002e-06 [updatestate_depend_eliminate]: 9.19e-06 [updatestate_assign_eliminate]: 2.86e-06 [updatestate_loads_eliminate]: 2.61e-06 [renormalize]: 6.00005e-07 [cse]: 3.221e-05 [optimize_parallel_all_gather_comm]: 1.996e-05 [overlap_param_gather]: 1.84e-06 [cconv]: 3.258e-05 [loop_unroll]: 0.00045416 [opt_after_cconv]: 0.0001107, [1] [Cycle 1]: 0.00010465, [7] [c_1]: 2.973e-05 [parameter_eliminate]: 4.77e-06 [updatestate_depend_eliminate]: 5.78997e-06 [updatestate_assign_eliminate]: 2.99001e-06 [updatestate_loads_eliminate]: 2.66e-06 [cse]: 2.533e-05 [renormalize]: 8.2e-07 [remove_dup_value]: 1.747e-05 [tuple_transform]: 9.615e-05, [1] [Cycle 1]: 9.17e-05, [4] [d_1]: 6.164e-05 [none_parameter_eliminate]: 2.27001e-06 [renormalize]: 1.8999e-07 [switch_simplify]: 7.23e-06 [partial_unused_args_eliminate]: 2.11e-06 [add_recomputation]: 5.492e-05 [cse_after_recomputation]: 2.985e-05, [1] [Cycle 1]: 2.537e-05, [1] [cse]: 1.992e-05 [environ_conv]: 7.11001e-06 [swap_dp_allreduce_reducescatter]: 5.78002e-06 [bias_add_comm_swap]: 2.44001e-06 [label_micro_interleaved_index]: 5.00999e-06 [label_fine_grained_interleaved_index]: 2.63e-06 [merge_cast_opt]: 1.76e-06 [slice_recompute_activation]: 2.14999e-06 [micro_interleaved_order_control]: 2.13998e-06 [assign_add_opt]: 1.54998e-06 [ForceFp32Comm]: 8.80013e-07 [remove_cast_before_assign_add]: 1.02998e-06 [full_micro_interleaved_order_control]: 2.81e-06 [reorder_send_recv_between_fp_bp]: 2.80002e-06 [comm_op_add_attrs]: 1.25999e-06 [add_comm_op_reuse_tag]: 1.35999e-06 [interleave_split_concat_branches]: 1.29e-06 [interleave_parallel_branches]: 1.35001e-06 [overlap_opt_shard_in_pipeline]: 1.77999e-06 [overlap_opt_shard_grad_in_pipeline]: 1.79e-06 [control_data_broadcast_order]: 1.367e-05 [grouped_pairwise_exchange_alltoall]: 1.61002e-06 [offloading_packed_experts]: 4.23999e-06 [overlap_recompute_and_grad_model_parallel]: 5.39e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.35001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.45999e-06 [overlap_recompute_comm]: 2.29999e-06 [overlap_grad_ring_attention]: 4.3e-06 [overlap_grad_flash_sp]: 2.33e-05 [begin_end_overlap_inline]: 5.39992e-07 [split_matmul_comm_elemetwise]: 2.26e-06 [split_layernorm_comm]: 1.91e-06 [handle_group_info]: 1.05999e-06 [symbol_engine_optimizer]: 8.492e-05, [1] [Cycle 1]: 7.428e-05, [6] [build]: 3.94002e-06 [elim_shapecalc]: 1.061e-05 [elim_not_effective]: 1.305e-05 [opt_reshape]: 7.01999e-06 [fold_const_symbol]: 1.125e-05 [renormalize]: 1.69995e-07 [detach_backward]: 2.07999e-06 [pipeline_parallel_scheduler]: 1.57999e-06 [auto_monad_reorder]: 2.188e-05 [get_jit_bprop_graph]: 1.87999e-06 [rewriter_after_jit_bprop_graph]: 5.94999e-06 [opt_after_jit_grad]: 0.0005216 [validate]: 4.803e-05 Sums bootstrap : 0.000671s : 0.56% type_inference : 0.114583s : 95.43% event_method : 0.000018s : 0.02% auto_monad : 0.000063s : 0.05% graph_reusing : 0.000005s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000019s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000040s : 0.03% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000011s : 0.01% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000025s : 0.02% optimize.rewriter_before_opt_a : 0.000054s : 0.05% optimize.opt_a.expand_dump_flag : 0.000005s : 0.00% optimize.opt_a.switch_simplify : 0.000034s : 0.03% optimize.opt_a.loop_unroll : 0.000018s : 0.01% optimize.opt_a.a_1 : 0.000476s : 0.40% optimize.opt_a.with_stream_mark : 0.000042s : 0.04% optimize.opt_a.recompute_prepare : 0.000018s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.01% 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 : 0.00% optimize.opt_a.a_2 : 0.000152s : 0.13% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.01% optimize.opt_a.shard : 0.000004s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000022s : 0.02% optimize.opt_a.shard_inline : 0.000017s : 0.01% optimize.opt_a.merge_send_recv : 0.000018s : 0.01% optimize.opt_a.auto_parallel : 0.000017s : 0.01% optimize.opt_a.parallel : 0.000040s : 0.03% optimize.opt_a.flash_sp : 0.000015s : 0.01% optimize.opt_a.merge_comm : 0.000009s : 0.01% optimize.opt_a.allreduce_fusion : 0.000007s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000019s : 0.02% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000018s : 0.02% optimize.opt_a.virtual_dataset : 0.000013s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000012s : 0.01% optimize.opt_a.virtual_output : 0.000019s : 0.02% optimize.opt_a.merge_forward : 0.000011s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.00% optimize.opt_a.offload_activation : 0.000021s : 0.02% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000034s : 0.03% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000021s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000009s : 0.01% optimize.opt_a.meta_fg_expand : 0.000006s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000005s : 0.00% optimize.opt_a.receive_attached : 0.000004s : 0.00% optimize.opt_a.after_resolve : 0.000026s : 0.02% optimize.opt_a.a_after_grad : 0.000020s : 0.02% optimize.opt_a.renormalize : 0.000747s : 0.62% optimize.opt_a.add_forward_monad_depend : 0.000008s : 0.01% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000028s : 0.02% optimize.opt_a.cse : 0.000063s : 0.05% optimize.opt_a.a_3 : 0.000094s : 0.08% optimize.py_interpret_to_execute_after_opt_a : 0.000015s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000052s : 0.04% optimize.convert_after_rewriter : 0.000009s : 0.01% optimize.order_py_execute_after_rewriter : 0.000006s : 0.00% optimize.mutable_eliminate : 0.000668s : 0.56% optimize.opt_b.b_1 : 0.000125s : 0.10% optimize.opt_b.b_2 : 0.000008s : 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 : 0.000032s : 0.03% optimize.optimize_parallel_all_gather_comm : 0.000020s : 0.02% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000033s : 0.03% optimize.loop_unroll : 0.000454s : 0.38% optimize.opt_after_cconv.c_1 : 0.000030s : 0.02% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000006s : 0.00% 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 : 0.02% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000017s : 0.01% optimize.tuple_transform.d_1 : 0.000062s : 0.05% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000055s : 0.05% optimize.cse_after_recomputation.cse : 0.000020s : 0.02% optimize.environ_conv : 0.000007s : 0.01% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.00% optimize.bias_add_comm_swap : 0.000002s : 0.00% optimize.label_micro_interleaved_index : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000014s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000004s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000004s : 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 : 0.000011s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000013s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 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 : 0.000002s : 0.00% auto_monad_reorder : 0.000022s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.000522s : 0.43% validate : 0.000048s : 0.04% Time group info: ------[substitution.] 0.000150 21 1.24% : 0.000002s : 2: substitution.elim_not_effective 1.14% : 0.000002s : 2: substitution.fold_const_symbol 4.33% : 0.000006s : 4: substitution.graph_param_transform 82.10% : 0.000123s : 1: substitution.inline 3.01% : 0.000005s : 4: substitution.j_node_and_user_rematch 3.91% : 0.000006s : 4: substitution.remove_not_recompute_node 4.28% : 0.000006s : 4: substitution.replace_old_param ------[type_inference.] 0.114508 2 99.58% : 0.114025s : 1: type_inference.infer 0.42% : 0.000483s : 1: type_inference.specialize ------[replace.] 0.000021 1 100.00% : 0.000021s : 1: replace.inline ------[match.] 0.000121 1 100.00% : 0.000121s : 1: match.inline ------[predicate.] 0.000154 967 0.73% : 0.000001s : 9: predicate.accumulaten_eliminater 0.87% : 0.000001s : 4: predicate.ad_related_special_op_eliminate 0.71% : 0.000001s : 8: predicate.addn_check_dump 0.81% : 0.000001s : 9: predicate.addn_zero_filter 0.65% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.33% : 0.000004s : 17: predicate.arithmetic_simplify 0.73% : 0.000001s : 9: predicate.cast_eliminate 0.79% : 0.000001s : 8: predicate.check_bprop_eliminate 0.69% : 0.000001s : 8: predicate.compare_switch_simplify 0.25% : 0.000000s : 4: predicate.const_output_eliminate 0.69% : 0.000001s : 8: predicate.depend_value_elim 0.73% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.92% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.80% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.18% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.26% : 0.000000s : 4: predicate.elim_not_effective 0.71% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.02% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.97% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.97% : 0.000002s : 13: predicate.environ_get_depend_swap 1.96% : 0.000003s : 21: predicate.environ_get_eliminate 0.94% : 0.000001s : 13: predicate.environ_get_set_eliminate 0.81% : 0.000001s : 10: predicate.exchange_switch_depend_value 2.03% : 0.000003s : 10: predicate.float_depend_g_call 0.63% : 0.000001s : 8: predicate.float_environ_get_switch 0.99% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.24% : 0.000000s : 4: predicate.fold_const_symbol 0.81% : 0.000001s : 8: predicate.get_grad_eliminate 0.38% : 0.000001s : 4: predicate.graph_param_transform 0.73% : 0.000001s : 8: predicate.incorporate_call 0.61% : 0.000001s : 8: predicate.incorporate_call_switch 7.13% : 0.000011s : 43: predicate.inline 1.27% : 0.000002s : 8: predicate.inline_without_move 0.42% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.94% : 0.000001s : 8: predicate.less_batch_normalization 1.56% : 0.000002s : 17: predicate.list_to_tuple_eliminator_ 1.96% : 0.000003s : 26: predicate.load_eliminater 1.16% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.33% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.94% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.66% : 0.000001s : 8: predicate.merge_addn 0.73% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.75% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.66% : 0.000001s : 9: predicate.minmaximum_grad 1.97% : 0.000003s : 4: predicate.mutable_eliminate 0.55% : 0.000001s : 4: predicate.opt_reshape 0.63% : 0.000001s : 4: predicate.parallel_virtual_node 1.14% : 0.000002s : 10: predicate.partial_defer_inline 1.14% : 0.000002s : 13: predicate.partial_eliminate 0.77% : 0.000001s : 9: predicate.print_const_string_wrapper 0.70% : 0.000001s : 8: predicate.reduce_all_const_elim 1.23% : 0.000002s : 9: predicate.reduce_eliminate 2.01% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.64% : 0.000001s : 8: predicate.remove_not_recompute_node 1.29% : 0.000002s : 17: predicate.replace_applicator 0.99% : 0.000002s : 8: predicate.replace_old_param 0.51% : 0.000001s : 4: predicate.reset_defer_inline 0.76% : 0.000001s : 9: predicate.reshape_eliminate 0.75% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.41% : 0.000001s : 4: predicate.row_tensor_eliminate 1.18% : 0.000002s : 8: predicate.same_eliminate 0.57% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.21% : 0.000002s : 8: predicate.shard_identity_eliminate 1.03% : 0.000002s : 8: predicate.special_op_eliminate 1.17% : 0.000002s : 8: predicate.specialize_transform 1.39% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.95% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.37% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.85% : 0.000001s : 10: predicate.switch_defer_inline 1.63% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.77% : 0.000006s : 35: predicate.switch_simplify 0.71% : 0.000001s : 9: predicate.tile_eliminate 0.81% : 0.000001s : 9: predicate.transpose_eliminate 1.66% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.56% : 0.000002s : 17: predicate.tuple_list_get_item_const_eliminator 1.31% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.19% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.51% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.38% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.54% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 1.94% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.79% : 0.000004s : 34: predicate.updatestate_useless_node_eliminater 0.47% : 0.000001s : 4: predicate.value_based_eliminate 0.86% : 0.000001s : 8: predicate.virtual_dataset_eliminate 1.34% : 0.000002s : 8: predicate.virtual_output_eliminate 0.39% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.48% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000261 4 8.33% : 0.000022s : 1: func_graph_cloner_run.FuncGraphClonerGraph 91.67% : 0.000239s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.138710 192 0.00% : 0.000004s : 1: ForceFp32Comm 3.81% : 0.005279s : 1: add_attr 3.79% : 0.005260s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.04% : 0.000059s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.05% : 0.000071s : 1: auto_monad 0.02% : 0.000026s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000005s : 1: bias_add_comm_swap 0.51% : 0.000708s : 1: bootstrap 0.03% : 0.000036s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000017s : 1: control_data_broadcast_order 0.01% : 0.000013s : 1: convert_after_rewriter 0.02% : 0.000033s : 1: cse_after_recomputation 0.01% : 0.000014s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.01% : 0.000010s : 1: environ_conv 0.02% : 0.000026s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.01% : 0.000009s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000011s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000005s : 1: label_fine_grained_interleaved_index 0.01% : 0.000008s : 1: label_micro_interleaved_index 0.34% : 0.000465s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.49% : 0.000681s : 1: mutable_eliminate 0.01% : 0.000007s : 1: offloading_packed_experts 0.01% : 0.000015s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000020s : 1: opt.transform.mutable_eliminate 0.64% : 0.000894s : 78: opt.transform.opt_a 0.02% : 0.000029s : 1: opt.transform.opt_after_cconv 0.02% : 0.000025s : 1: opt.transform.opt_after_jit_grad 0.08% : 0.000104s : 28: opt.transform.opt_b 0.05% : 0.000067s : 2: opt.transform.opt_trans_graph 0.03% : 0.000039s : 4: opt.transform.symbol_engine_opt 2.00% : 0.002780s : 1: opt_a 0.08% : 0.000114s : 1: opt_after_cconv 0.38% : 0.000532s : 1: opt_after_jit_grad 0.16% : 0.000226s : 1: opt_b 3.68% : 0.005106s : 1: optimize 0.02% : 0.000024s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000009s : 1: order_py_execute_after_rewriter 0.02% : 0.000027s : 1: overlap_grad_flash_sp 0.01% : 0.000008s : 1: overlap_grad_matmul_and_grad_allreduce 0.01% : 0.000007s : 1: overlap_grad_ring_attention 0.00% : 0.000004s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000008s : 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% : 0.000044s : 1: pre_auto_parallel 0.02% : 0.000029s : 1: py_interpret_to_execute 0.01% : 0.000019s : 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 0.32% : 0.000451s : 1: renormalize.infer 0.21% : 0.000287s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.04% : 0.000056s : 1: rewriter_after_opt_a 0.04% : 0.000059s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.06% : 0.000088s : 1: symbol_engine_optimizer 0.07% : 0.000099s : 1: tuple_transform 82.62% : 0.114607s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[dim_type1-True-batch_size_list2-shape1-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_fp16[dim_type1-True-batch_size_list2-shape1-ms_dtype0],max_mem:6.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:39:53.728.106 [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 = 0.109331, [21] [bootstrap]: 0.00070252 [type_inference]: 0.0972511 [event_method]: 1.533e-05 [auto_monad]: 6.894e-05 [graph_reusing]: 5.42999e-06 [inline]: 3.67998e-06 [add_attr]: 0.00502356, [1] [add_attr_with_inline]: 0.00500707, [1] [Cycle 1]: 7.879e-05, [2] [tag_attr]: 1.998e-05 [meta_addattr_fg_expand]: 4.11001e-06 [parallel-infer-symbol]: 3.92002e-06 [pre_auto_parallel]: 4.105e-05 [insert-virtual-dataset]: 2.63998e-06 [parallel-infer-symbol-second]: 8.79983e-07 [dataset_repeat_opt]: 2.57001e-06 [pipeline_split]: 1.76e-06 [optimize]: 0.00530926, [53] [py_interpret_to_execute]: 2.635e-05 [rewriter_before_opt_a]: 5.806e-05 [opt_a]: 0.00268066, [2] [Cycle 1]: 0.00194931, [45] [expand_dump_flag]: 3.11001e-06 [switch_simplify]: 2.578e-05 [loop_unroll]: 1.181e-05 [a_1]: 0.00031799 [with_stream_mark]: 2.907e-05 [recompute_prepare]: 1.064e-05 [updatestate_depend_eliminate]: 4.97e-06 [updatestate_assign_eliminate]: 3.78001e-06 [updatestate_loads_eliminate]: 3.65e-06 [parameter_eliminate]: 2.06e-06 [a_2]: 8.799e-05 [accelerated_algorithm]: 7.43e-06 [shard]: 2.45002e-06 [meta_shard_fg_expand]: 1.96e-06 [shard_inline]: 6.31998e-06 [merge_send_recv]: 1.138e-05 [auto_parallel]: 7.14001e-06 [parallel]: 3.527e-05 [flash_sp]: 1.154e-05 [merge_comm]: 3.98001e-06 [allreduce_fusion]: 3.53999e-06 [matmul_add_comm_reduction]: 1.271e-05 [allreduce_slice_to_reducescatter]: 6.19999e-07 [virtual_shard_identity]: 1.055e-05 [virtual_dataset]: 6.44999e-06 [get_grad_eliminate_]: 6.23e-06 [virtual_output]: 5.77999e-06 [merge_forward]: 5.20999e-06 [cell_reuse_recompute_pass]: 1.67999e-06 [offload_activation]: 1.418e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.88e-05 [merge_recompute_call_nodes]: 2.07999e-06 [before_grad]: 1.339e-05 [set_forward_comm_id_for_comm_node_pass]: 4.47e-06 [meta_fg_expand]: 3.06999e-06 [flash_sp_send_recv_attached]: 2.98e-06 [receive_attached]: 2.43998e-06 [after_resolve]: 1.58e-05 [a_after_grad]: 1.082e-05 [renormalize]: 0.0007954 [add_forward_monad_depend]: 8.65999e-06 [auto_monad_grad]: 2.62001e-06 [auto_monad_eliminator]: 2.004e-05 [cse]: 4.72e-05 [a_3]: 5.683e-05 [Cycle 2]: 0.00071673, [45] [expand_dump_flag]: 1.97999e-06 [switch_simplify]: 8.87e-06 [loop_unroll]: 6.60997e-06 [a_1]: 0.00012214 [with_stream_mark]: 2.006e-05 [recompute_prepare]: 6.57002e-06 [updatestate_depend_eliminate]: 3.75e-06 [updatestate_assign_eliminate]: 2.62001e-06 [updatestate_loads_eliminate]: 3.4e-06 [parameter_eliminate]: 1.99e-06 [a_2]: 7.238e-05 [accelerated_algorithm]: 7e-06 [shard]: 3.17002e-06 [meta_shard_fg_expand]: 1.57999e-06 [shard_inline]: 6.98e-06 [merge_send_recv]: 7.48e-06 [auto_parallel]: 7.98001e-06 [parallel]: 8.08999e-06 [flash_sp]: 3.78999e-06 [merge_comm]: 3.63999e-06 [allreduce_fusion]: 3.35e-06 [matmul_add_comm_reduction]: 1.413e-05 [allreduce_slice_to_reducescatter]: 6.30011e-07 [virtual_shard_identity]: 7.4e-06 [virtual_dataset]: 5.75001e-06 [get_grad_eliminate_]: 5.81998e-06 [virtual_output]: 5.92999e-06 [merge_forward]: 4.80999e-06 [cell_reuse_recompute_pass]: 3.09001e-06 [offload_activation]: 1.171e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.698e-05 [merge_recompute_call_nodes]: 1.34e-06 [before_grad]: 1.101e-05 [set_forward_comm_id_for_comm_node_pass]: 4.79e-06 [meta_fg_expand]: 2.37999e-06 [flash_sp_send_recv_attached]: 1.39e-06 [receive_attached]: 2.11e-06 [after_resolve]: 1.237e-05 [a_after_grad]: 9.16998e-06 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 1.63002e-06 [auto_monad_grad]: 1.02e-06 [auto_monad_eliminator]: 8.81002e-06 [cse]: 2.373e-05 [a_3]: 3.662e-05 [py_interpret_to_execute_after_opt_a]: 1.875e-05 [slice_cell_reuse_recomputed_activation]: 2.87002e-06 [rewriter_after_opt_a]: 4.994e-05 [convert_after_rewriter]: 7.73001e-06 [order_py_execute_after_rewriter]: 5.76e-06 [mutable_eliminate]: 0.00079405 [opt_b]: 0.00023444, [1] [Cycle 1]: 0.00022528, [7] [b_1]: 0.00012785 [b_2]: 8.47998e-06 [updatestate_depend_eliminate]: 1.025e-05 [updatestate_assign_eliminate]: 3.14001e-06 [updatestate_loads_eliminate]: 3.11999e-06 [renormalize]: 5.89993e-07 [cse]: 3.63e-05 [optimize_parallel_all_gather_comm]: 2.28e-05 [overlap_param_gather]: 2.01e-06 [cconv]: 3.97e-05 [loop_unroll]: 0.0004798 [opt_after_cconv]: 0.00011717, [1] [Cycle 1]: 0.00011028, [7] [c_1]: 3.019e-05 [parameter_eliminate]: 6.50002e-06 [updatestate_depend_eliminate]: 6.34001e-06 [updatestate_assign_eliminate]: 2.61e-06 [updatestate_loads_eliminate]: 2.37001e-06 [cse]: 2.348e-05 [renormalize]: 3.60014e-07 [remove_dup_value]: 4.961e-05 [tuple_transform]: 9.36e-05, [1] [Cycle 1]: 8.392e-05, [4] [d_1]: 5.299e-05 [none_parameter_eliminate]: 2.17001e-06 [renormalize]: 1.90019e-07 [switch_simplify]: 7.94997e-06 [partial_unused_args_eliminate]: 2.49999e-06 [add_recomputation]: 6.191e-05 [cse_after_recomputation]: 2.523e-05, [1] [Cycle 1]: 2.022e-05, [1] [cse]: 1.437e-05 [environ_conv]: 7.2e-06 [swap_dp_allreduce_reducescatter]: 6.03998e-06 [bias_add_comm_swap]: 3.03e-06 [label_micro_interleaved_index]: 5.79999e-06 [label_fine_grained_interleaved_index]: 3.51999e-06 [merge_cast_opt]: 1.50999e-06 [slice_recompute_activation]: 2.86e-06 [micro_interleaved_order_control]: 2.53e-06 [assign_add_opt]: 1.32999e-06 [ForceFp32Comm]: 9.10019e-07 [remove_cast_before_assign_add]: 1.17e-06 [full_micro_interleaved_order_control]: 2.44999e-06 [reorder_send_recv_between_fp_bp]: 3.32002e-06 [comm_op_add_attrs]: 1.14e-06 [add_comm_op_reuse_tag]: 1.08001e-06 [interleave_split_concat_branches]: 1.74e-06 [interleave_parallel_branches]: 1.21997e-06 [overlap_opt_shard_in_pipeline]: 2.56e-06 [overlap_opt_shard_grad_in_pipeline]: 1.96e-06 [control_data_broadcast_order]: 1.532e-05 [grouped_pairwise_exchange_alltoall]: 1.59e-06 [offloading_packed_experts]: 4.99998e-06 [overlap_recompute_and_grad_model_parallel]: 6.59001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.22999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.44e-06 [overlap_recompute_comm]: 2.81999e-06 [overlap_grad_ring_attention]: 5.17e-06 [overlap_grad_flash_sp]: 2.653e-05 [begin_end_overlap_inline]: 5.50004e-07 [split_matmul_comm_elemetwise]: 2.42001e-06 [split_layernorm_comm]: 2.04999e-06 [handle_group_info]: 1.37999e-06 [symbol_engine_optimizer]: 8.65e-05, [1] [Cycle 1]: 8.107e-05, [6] [build]: 4.83001e-06 [elim_shapecalc]: 1.076e-05 [elim_not_effective]: 1.497e-05 [opt_reshape]: 7.93999e-06 [fold_const_symbol]: 1.024e-05 [renormalize]: 2.10013e-07 [detach_backward]: 2.71999e-06 [pipeline_parallel_scheduler]: 1.94999e-06 [auto_monad_reorder]: 2.296e-05 [get_jit_bprop_graph]: 2.48e-06 [rewriter_after_jit_bprop_graph]: 6.61e-06 [opt_after_jit_grad]: 0.0005946 [validate]: 5.318e-05 Sums bootstrap : 0.000703s : 0.68% type_inference : 0.097251s : 94.33% event_method : 0.000015s : 0.01% auto_monad : 0.000069s : 0.07% graph_reusing : 0.000005s : 0.01% inline : 0.000004s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000020s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000041s : 0.04% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000026s : 0.03% optimize.rewriter_before_opt_a : 0.000058s : 0.06% optimize.opt_a.expand_dump_flag : 0.000005s : 0.00% optimize.opt_a.switch_simplify : 0.000035s : 0.03% optimize.opt_a.loop_unroll : 0.000018s : 0.02% optimize.opt_a.a_1 : 0.000440s : 0.43% optimize.opt_a.with_stream_mark : 0.000049s : 0.05% optimize.opt_a.recompute_prepare : 0.000017s : 0.02% optimize.opt_a.updatestate_depend_eliminate : 0.000009s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000006s : 0.01% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.01% optimize.opt_a.parameter_eliminate : 0.000004s : 0.00% optimize.opt_a.a_2 : 0.000160s : 0.16% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.01% optimize.opt_a.shard : 0.000006s : 0.01% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.00% optimize.opt_a.shard_inline : 0.000013s : 0.01% optimize.opt_a.merge_send_recv : 0.000019s : 0.02% optimize.opt_a.auto_parallel : 0.000015s : 0.01% optimize.opt_a.parallel : 0.000043s : 0.04% optimize.opt_a.flash_sp : 0.000015s : 0.01% optimize.opt_a.merge_comm : 0.000008s : 0.01% optimize.opt_a.allreduce_fusion : 0.000007s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000027s : 0.03% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000018s : 0.02% optimize.opt_a.virtual_dataset : 0.000012s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000012s : 0.01% optimize.opt_a.virtual_output : 0.000012s : 0.01% optimize.opt_a.merge_forward : 0.000010s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000026s : 0.03% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000036s : 0.03% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000024s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000009s : 0.01% optimize.opt_a.meta_fg_expand : 0.000005s : 0.01% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000028s : 0.03% optimize.opt_a.a_after_grad : 0.000020s : 0.02% optimize.opt_a.renormalize : 0.000795s : 0.77% optimize.opt_a.add_forward_monad_depend : 0.000010s : 0.01% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000029s : 0.03% optimize.opt_a.cse : 0.000071s : 0.07% optimize.opt_a.a_3 : 0.000093s : 0.09% optimize.py_interpret_to_execute_after_opt_a : 0.000019s : 0.02% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000050s : 0.05% optimize.convert_after_rewriter : 0.000008s : 0.01% optimize.order_py_execute_after_rewriter : 0.000006s : 0.01% optimize.mutable_eliminate : 0.000794s : 0.77% optimize.opt_b.b_1 : 0.000128s : 0.12% optimize.opt_b.b_2 : 0.000008s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000010s : 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 : 0.000036s : 0.04% optimize.optimize_parallel_all_gather_comm : 0.000023s : 0.02% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000040s : 0.04% optimize.loop_unroll : 0.000480s : 0.47% optimize.opt_after_cconv.c_1 : 0.000030s : 0.03% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 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 : 0.000002s : 0.00% optimize.opt_after_cconv.cse : 0.000023s : 0.02% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000050s : 0.05% optimize.tuple_transform.d_1 : 0.000053s : 0.05% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000008s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000062s : 0.06% optimize.cse_after_recomputation.cse : 0.000014s : 0.01% 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 : 0.000006s : 0.01% optimize.label_fine_grained_interleaved_index : 0.000004s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000003s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000002s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 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 : 0.000015s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000005s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 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 : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000005s : 0.01% optimize.overlap_grad_flash_sp : 0.000027s : 0.03% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000005s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000011s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000015s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000023s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.01% opt_after_jit_grad : 0.000595s : 0.58% validate : 0.000053s : 0.05% Time group info: ------[substitution.] 0.000136 21 1.72% : 0.000002s : 2: substitution.elim_not_effective 1.16% : 0.000002s : 2: substitution.fold_const_symbol 5.10% : 0.000007s : 4: substitution.graph_param_transform 78.92% : 0.000107s : 1: substitution.inline 3.55% : 0.000005s : 4: substitution.j_node_and_user_rematch 4.12% : 0.000006s : 4: substitution.remove_not_recompute_node 5.43% : 0.000007s : 4: substitution.replace_old_param ------[type_inference.] 0.097177 2 99.44% : 0.096631s : 1: type_inference.infer 0.56% : 0.000546s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000105 1 100.00% : 0.000105s : 1: match.inline ------[predicate.] 0.000158 967 1.01% : 0.000002s : 9: predicate.accumulaten_eliminater 1.15% : 0.000002s : 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 0.66% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.16% : 0.000003s : 17: predicate.arithmetic_simplify 1.00% : 0.000002s : 9: predicate.cast_eliminate 0.80% : 0.000001s : 8: predicate.check_bprop_eliminate 0.74% : 0.000001s : 8: predicate.compare_switch_simplify 0.26% : 0.000000s : 4: predicate.const_output_eliminate 0.64% : 0.000001s : 8: predicate.depend_value_elim 0.73% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.79% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.82% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.37% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.30% : 0.000000s : 4: predicate.elim_not_effective 0.52% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.12% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.24% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.99% : 0.000002s : 13: predicate.environ_get_depend_swap 1.65% : 0.000003s : 21: predicate.environ_get_eliminate 0.99% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.85% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.83% : 0.000003s : 10: predicate.float_depend_g_call 0.61% : 0.000001s : 8: predicate.float_environ_get_switch 0.92% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.25% : 0.000000s : 4: predicate.fold_const_symbol 0.94% : 0.000001s : 8: predicate.get_grad_eliminate 0.62% : 0.000001s : 4: predicate.graph_param_transform 0.69% : 0.000001s : 8: predicate.incorporate_call 0.58% : 0.000001s : 8: predicate.incorporate_call_switch 6.13% : 0.000010s : 43: predicate.inline 1.03% : 0.000002s : 8: predicate.inline_without_move 0.39% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.18% : 0.000002s : 8: predicate.less_batch_normalization 1.73% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 1.96% : 0.000003s : 26: predicate.load_eliminater 1.16% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.27% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.69% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.68% : 0.000001s : 8: predicate.merge_addn 0.67% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.69% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.63% : 0.000001s : 9: predicate.minmaximum_grad 2.03% : 0.000003s : 4: predicate.mutable_eliminate 0.80% : 0.000001s : 4: predicate.opt_reshape 0.75% : 0.000001s : 4: predicate.parallel_virtual_node 1.28% : 0.000002s : 10: predicate.partial_defer_inline 1.13% : 0.000002s : 13: predicate.partial_eliminate 0.71% : 0.000001s : 9: predicate.print_const_string_wrapper 0.78% : 0.000001s : 8: predicate.reduce_all_const_elim 0.89% : 0.000001s : 9: predicate.reduce_eliminate 2.19% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.97% : 0.000002s : 8: predicate.remove_not_recompute_node 1.73% : 0.000003s : 17: predicate.replace_applicator 0.74% : 0.000001s : 8: predicate.replace_old_param 0.58% : 0.000001s : 4: predicate.reset_defer_inline 0.78% : 0.000001s : 9: predicate.reshape_eliminate 0.72% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.57% : 0.000001s : 4: predicate.row_tensor_eliminate 1.02% : 0.000002s : 8: predicate.same_eliminate 0.56% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.27% : 0.000002s : 8: predicate.shard_identity_eliminate 0.92% : 0.000001s : 8: predicate.special_op_eliminate 0.82% : 0.000001s : 8: predicate.specialize_transform 1.62% : 0.000003s : 8: predicate.split_environ_get_set_with_tuple_value 0.84% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.44% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.82% : 0.000001s : 10: predicate.switch_defer_inline 1.53% : 0.000002s : 18: predicate.switch_layer_defer_inline 3.89% : 0.000006s : 35: predicate.switch_simplify 0.68% : 0.000001s : 9: predicate.tile_eliminate 0.96% : 0.000002s : 9: predicate.transpose_eliminate 1.48% : 0.000002s : 17: predicate.tuple_list_convert_item_index_to_positive 1.70% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.39% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.20% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.28% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.35% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.69% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.89% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.91% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.62% : 0.000001s : 4: predicate.value_based_eliminate 0.77% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.73% : 0.000001s : 8: predicate.virtual_output_eliminate 0.38% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.83% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000310 4 8.43% : 0.000026s : 1: func_graph_cloner_run.FuncGraphClonerGraph 91.57% : 0.000284s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.121457 192 0.00% : 0.000004s : 1: ForceFp32Comm 4.14% : 0.005031s : 1: add_attr 4.13% : 0.005012s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.05% : 0.000066s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.06% : 0.000075s : 1: auto_monad 0.03% : 0.000034s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.62% : 0.000751s : 1: bootstrap 0.04% : 0.000044s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.02% : 0.000019s : 1: control_data_broadcast_order 0.01% : 0.000011s : 1: convert_after_rewriter 0.02% : 0.000028s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.01% : 0.000006s : 1: detach_backward 0.01% : 0.000010s : 1: environ_conv 0.02% : 0.000023s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.01% : 0.000006s : 1: get_jit_bprop_graph 0.01% : 0.000009s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000005s : 1: handle_group_info 0.01% : 0.000007s : 1: inline 0.01% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.01% : 0.000007s : 1: label_fine_grained_interleaved_index 0.01% : 0.000009s : 1: label_micro_interleaved_index 0.40% : 0.000491s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.67% : 0.000808s : 1: mutable_eliminate 0.01% : 0.000008s : 1: offloading_packed_experts 0.01% : 0.000017s : 1: opt.transform.loop_unroll_optimizer 0.02% : 0.000022s : 1: opt.transform.mutable_eliminate 0.71% : 0.000861s : 78: opt.transform.opt_a 0.02% : 0.000029s : 1: opt.transform.opt_after_cconv 0.03% : 0.000033s : 1: opt.transform.opt_after_jit_grad 0.09% : 0.000107s : 28: opt.transform.opt_b 0.05% : 0.000058s : 2: opt.transform.opt_trans_graph 0.03% : 0.000040s : 4: opt.transform.symbol_engine_opt 2.21% : 0.002684s : 1: opt_a 0.10% : 0.000120s : 1: opt_after_cconv 0.50% : 0.000610s : 1: opt_after_jit_grad 0.20% : 0.000238s : 1: opt_b 4.38% : 0.005316s : 1: optimize 0.02% : 0.000027s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000009s : 1: order_py_execute_after_rewriter 0.02% : 0.000030s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.06% : 0.000078s : 1: overlap_grad_ring_attention 0.01% : 0.000010s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.01% : 0.000010s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.01% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.04% : 0.000046s : 1: pre_auto_parallel 0.03% : 0.000031s : 1: py_interpret_to_execute 0.02% : 0.000022s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000008s : 1: remove_cast_before_assign_add 0.04% : 0.000054s : 1: remove_dup_value 0.39% : 0.000471s : 1: renormalize.infer 0.26% : 0.000315s : 1: renormalize.specialize 0.01% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.01% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.05% : 0.000055s : 1: rewriter_after_opt_a 0.05% : 0.000064s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000006s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.01% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.07% : 0.000089s : 1: symbol_engine_optimizer 0.08% : 0.000097s : 1: tuple_transform 80.10% : 0.097283s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[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_fp16[dim_type1-False-batch_size_list0-shape0-ms_dtype0],max_mem:6.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:39:56.665.177 [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 = 0.0159221, [21] [bootstrap]: 0.00064918 [type_inference]: 0.00392838 [event_method]: 1.357e-05 [auto_monad]: 5.627e-05 [graph_reusing]: 5.28002e-06 [inline]: 2.66999e-06 [add_attr]: 0.0048497, [1] [add_attr_with_inline]: 0.00483496, [1] [Cycle 1]: 6.063e-05, [2] [tag_attr]: 1.62e-05 [meta_addattr_fg_expand]: 3.45e-06 [parallel-infer-symbol]: 3.53e-06 [pre_auto_parallel]: 3.755e-05 [insert-virtual-dataset]: 2.79999e-06 [parallel-infer-symbol-second]: 8.70001e-07 [dataset_repeat_opt]: 2.35002e-06 [pipeline_split]: 2.24001e-06 [optimize]: 0.00558764, [53] [py_interpret_to_execute]: 2.089e-05 [rewriter_before_opt_a]: 5.072e-05 [opt_a]: 0.00315239, [2] [Cycle 1]: 0.00245041, [45] [expand_dump_flag]: 2.60002e-06 [switch_simplify]: 2.537e-05 [loop_unroll]: 1.08e-05 [a_1]: 0.00028646 [with_stream_mark]: 2.207e-05 [recompute_prepare]: 8.1e-06 [updatestate_depend_eliminate]: 4.18001e-06 [updatestate_assign_eliminate]: 3.43e-06 [updatestate_loads_eliminate]: 3.41999e-06 [parameter_eliminate]: 1.94e-06 [a_2]: 7.987e-05 [accelerated_algorithm]: 6.94999e-06 [shard]: 2.26e-06 [meta_shard_fg_expand]: 1.77001e-06 [shard_inline]: 6.52001e-06 [merge_send_recv]: 1.051e-05 [auto_parallel]: 6.93e-06 [parallel]: 4.471e-05 [flash_sp]: 9.34998e-06 [merge_comm]: 4.22998e-06 [allreduce_fusion]: 3.63e-06 [matmul_add_comm_reduction]: 1.105e-05 [allreduce_slice_to_reducescatter]: 7.2e-07 [virtual_shard_identity]: 0.00059446 [virtual_dataset]: 1.355e-05 [get_grad_eliminate_]: 7.19001e-06 [virtual_output]: 7.18998e-06 [merge_forward]: 1.022e-05 [cell_reuse_recompute_pass]: 1.84998e-06 [offload_activation]: 1.448e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.851e-05 [merge_recompute_call_nodes]: 2.07001e-06 [before_grad]: 1.133e-05 [set_forward_comm_id_for_comm_node_pass]: 6.21e-06 [meta_fg_expand]: 2.83e-06 [flash_sp_send_recv_attached]: 4.23999e-06 [receive_attached]: 2.07001e-06 [after_resolve]: 1.394e-05 [a_after_grad]: 1.077e-05 [renormalize]: 0.0007413 [add_forward_monad_depend]: 7.47998e-06 [auto_monad_grad]: 3.06999e-06 [auto_monad_eliminator]: 3.485e-05 [cse]: 4.62e-05 [a_3]: 4.87e-05 [Cycle 2]: 0.00069057, [45] [expand_dump_flag]: 1.61002e-06 [switch_simplify]: 7.78999e-06 [loop_unroll]: 6.12999e-06 [a_1]: 0.00012544 [with_stream_mark]: 1.809e-05 [recompute_prepare]: 6.54001e-06 [updatestate_depend_eliminate]: 4.08001e-06 [updatestate_assign_eliminate]: 3.18998e-06 [updatestate_loads_eliminate]: 3.06001e-06 [parameter_eliminate]: 2.05002e-06 [a_2]: 7.211e-05 [accelerated_algorithm]: 6.34999e-06 [shard]: 1.55999e-06 [meta_shard_fg_expand]: 2.25002e-06 [shard_inline]: 6.06e-06 [merge_send_recv]: 6.71e-06 [auto_parallel]: 8.60001e-06 [parallel]: 8.10999e-06 [flash_sp]: 4.1e-06 [merge_comm]: 3.32002e-06 [allreduce_fusion]: 3.53999e-06 [matmul_add_comm_reduction]: 1.115e-05 [allreduce_slice_to_reducescatter]: 4.69998e-07 [virtual_shard_identity]: 6.96001e-06 [virtual_dataset]: 6.29999e-06 [get_grad_eliminate_]: 5.76e-06 [virtual_output]: 5.82001e-06 [merge_forward]: 4.05e-06 [cell_reuse_recompute_pass]: 1.77001e-06 [offload_activation]: 8.64003e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.351e-05 [merge_recompute_call_nodes]: 1.10001e-06 [before_grad]: 9.61e-06 [set_forward_comm_id_for_comm_node_pass]: 4.28999e-06 [meta_fg_expand]: 2.57001e-06 [flash_sp_send_recv_attached]: 1.30001e-06 [receive_attached]: 1.72001e-06 [after_resolve]: 1.123e-05 [a_after_grad]: 9.07999e-06 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 1.62001e-06 [auto_monad_grad]: 1.44e-06 [auto_monad_eliminator]: 8.90999e-06 [cse]: 2.259e-05 [a_3]: 3.691e-05 [py_interpret_to_execute_after_opt_a]: 1.477e-05 [slice_cell_reuse_recomputed_activation]: 2.39999e-06 [rewriter_after_opt_a]: 4.057e-05 [convert_after_rewriter]: 7.2e-06 [order_py_execute_after_rewriter]: 5.99e-06 [mutable_eliminate]: 0.00072467 [opt_b]: 0.00022892, [1] [Cycle 1]: 0.00022015, [7] [b_1]: 0.00012493 [b_2]: 8.47e-06 [updatestate_depend_eliminate]: 8.19002e-06 [updatestate_assign_eliminate]: 2.86e-06 [updatestate_loads_eliminate]: 2.71999e-06 [renormalize]: 7.89994e-07 [cse]: 3.664e-05 [optimize_parallel_all_gather_comm]: 2.1e-05 [overlap_param_gather]: 2.42001e-06 [cconv]: 3.664e-05 [loop_unroll]: 0.00051924 [opt_after_cconv]: 0.00011846, [1] [Cycle 1]: 0.00011114, [7] [c_1]: 3.021e-05 [parameter_eliminate]: 4.85001e-06 [updatestate_depend_eliminate]: 6.16998e-06 [updatestate_assign_eliminate]: 2.71e-06 [updatestate_loads_eliminate]: 2.58e-06 [cse]: 2.513e-05 [renormalize]: 7.09988e-07 [remove_dup_value]: 4.647e-05 [tuple_transform]: 8.012e-05, [1] [Cycle 1]: 7.433e-05, [4] [d_1]: 4.777e-05 [none_parameter_eliminate]: 1.62999e-06 [renormalize]: 2.19996e-07 [switch_simplify]: 6.98998e-06 [partial_unused_args_eliminate]: 2.31998e-06 [add_recomputation]: 5.345e-05 [cse_after_recomputation]: 2.417e-05, [1] [Cycle 1]: 1.995e-05, [1] [cse]: 1.426e-05 [environ_conv]: 6.58e-06 [swap_dp_allreduce_reducescatter]: 5.67001e-06 [bias_add_comm_swap]: 3.60998e-06 [label_micro_interleaved_index]: 5.31002e-06 [label_fine_grained_interleaved_index]: 2.84999e-06 [merge_cast_opt]: 1.24e-06 [slice_recompute_activation]: 2.19001e-06 [micro_interleaved_order_control]: 2.29001e-06 [assign_add_opt]: 1.66998e-06 [ForceFp32Comm]: 8.90024e-07 [remove_cast_before_assign_add]: 1.23002e-06 [full_micro_interleaved_order_control]: 2.91e-06 [reorder_send_recv_between_fp_bp]: 2.86e-06 [comm_op_add_attrs]: 1.31998e-06 [add_comm_op_reuse_tag]: 1.30999e-06 [interleave_split_concat_branches]: 1.42e-06 [interleave_parallel_branches]: 1.07e-06 [overlap_opt_shard_in_pipeline]: 1.25999e-06 [overlap_opt_shard_grad_in_pipeline]: 1.77001e-06 [control_data_broadcast_order]: 1.385e-05 [grouped_pairwise_exchange_alltoall]: 1.99e-06 [offloading_packed_experts]: 4.71002e-06 [overlap_recompute_and_grad_model_parallel]: 4.88001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.19998e-06 [overlap_recompute_allgather_and_fa_grad]: 1.35001e-06 [overlap_recompute_comm]: 2.75002e-06 [overlap_grad_ring_attention]: 4.25999e-06 [overlap_grad_flash_sp]: 2.337e-05 [begin_end_overlap_inline]: 5.40022e-07 [split_matmul_comm_elemetwise]: 2.35002e-06 [split_layernorm_comm]: 1.86e-06 [handle_group_info]: 1.39998e-06 [symbol_engine_optimizer]: 7.97e-05, [1] [Cycle 1]: 7.524e-05, [6] [build]: 4.05e-06 [elim_shapecalc]: 1.015e-05 [elim_not_effective]: 1.287e-05 [opt_reshape]: 7.43999e-06 [fold_const_symbol]: 1.026e-05 [renormalize]: 2.00002e-07 [detach_backward]: 1.856e-05 [pipeline_parallel_scheduler]: 1.70001e-06 [auto_monad_reorder]: 2.077e-05 [get_jit_bprop_graph]: 1.84e-06 [rewriter_after_jit_bprop_graph]: 5.92999e-06 [opt_after_jit_grad]: 0.00048957 [validate]: 4.509e-05 Sums bootstrap : 0.000649s : 6.51% type_inference : 0.003928s : 39.37% event_method : 0.000014s : 0.14% auto_monad : 0.000056s : 0.56% graph_reusing : 0.000005s : 0.05% inline : 0.000003s : 0.03% add_attr.add_attr_with_inline.tag_attr : 0.000016s : 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 : 0.000038s : 0.38% 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 : 0.21% optimize.rewriter_before_opt_a : 0.000051s : 0.51% optimize.opt_a.expand_dump_flag : 0.000004s : 0.04% optimize.opt_a.switch_simplify : 0.000033s : 0.33% optimize.opt_a.loop_unroll : 0.000017s : 0.17% optimize.opt_a.a_1 : 0.000412s : 4.13% optimize.opt_a.with_stream_mark : 0.000040s : 0.40% optimize.opt_a.recompute_prepare : 0.000015s : 0.15% 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 : 0.06% optimize.opt_a.parameter_eliminate : 0.000004s : 0.04% optimize.opt_a.a_2 : 0.000152s : 1.52% optimize.opt_a.accelerated_algorithm : 0.000013s : 0.13% 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 : 0.000017s : 0.17% optimize.opt_a.auto_parallel : 0.000016s : 0.16% optimize.opt_a.parallel : 0.000053s : 0.53% optimize.opt_a.flash_sp : 0.000013s : 0.13% optimize.opt_a.merge_comm : 0.000008s : 0.08% optimize.opt_a.allreduce_fusion : 0.000007s : 0.07% optimize.opt_a.matmul_add_comm_reduction : 0.000022s : 0.22% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000601s : 6.03% optimize.opt_a.virtual_dataset : 0.000020s : 0.20% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.13% optimize.opt_a.virtual_output : 0.000013s : 0.13% optimize.opt_a.merge_forward : 0.000014s : 0.14% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.04% optimize.opt_a.offload_activation : 0.000023s : 0.23% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000032s : 0.32% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.03% optimize.opt_a.before_grad : 0.000021s : 0.21% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000010s : 0.11% optimize.opt_a.meta_fg_expand : 0.000005s : 0.05% optimize.opt_a.flash_sp_send_recv_attached : 0.000006s : 0.06% optimize.opt_a.receive_attached : 0.000004s : 0.04% optimize.opt_a.after_resolve : 0.000025s : 0.25% optimize.opt_a.a_after_grad : 0.000020s : 0.20% optimize.opt_a.renormalize : 0.000741s : 7.43% optimize.opt_a.add_forward_monad_depend : 0.000009s : 0.09% optimize.opt_a.auto_monad_grad : 0.000005s : 0.05% optimize.opt_a.auto_monad_eliminator : 0.000044s : 0.44% optimize.opt_a.cse : 0.000069s : 0.69% optimize.opt_a.a_3 : 0.000086s : 0.86% optimize.py_interpret_to_execute_after_opt_a : 0.000015s : 0.15% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.02% optimize.rewriter_after_opt_a : 0.000041s : 0.41% optimize.convert_after_rewriter : 0.000007s : 0.07% optimize.order_py_execute_after_rewriter : 0.000006s : 0.06% optimize.mutable_eliminate : 0.000725s : 7.26% optimize.opt_b.b_1 : 0.000125s : 1.25% optimize.opt_b.b_2 : 0.000008s : 0.08% 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 : 0.000037s : 0.37% optimize.optimize_parallel_all_gather_comm : 0.000021s : 0.21% optimize.overlap_param_gather : 0.000002s : 0.02% optimize.cconv : 0.000037s : 0.37% optimize.loop_unroll : 0.000519s : 5.20% optimize.opt_after_cconv.c_1 : 0.000030s : 0.30% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.05% 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 : 0.000003s : 0.03% optimize.opt_after_cconv.cse : 0.000025s : 0.25% optimize.opt_after_cconv.renormalize : 0.000001s : 0.01% optimize.remove_dup_value : 0.000046s : 0.47% optimize.tuple_transform.d_1 : 0.000048s : 0.48% 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 : 0.07% optimize.partial_unused_args_eliminate : 0.000002s : 0.02% optimize.add_recomputation : 0.000053s : 0.54% optimize.cse_after_recomputation.cse : 0.000014s : 0.14% optimize.environ_conv : 0.000007s : 0.07% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.06% optimize.bias_add_comm_swap : 0.000004s : 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 : 0.000001s : 0.01% optimize.slice_recompute_activation : 0.000002s : 0.02% optimize.micro_interleaved_order_control : 0.000002s : 0.02% optimize.assign_add_opt : 0.000002s : 0.02% optimize.ForceFp32Comm : 0.000001s : 0.01% optimize.remove_cast_before_assign_add : 0.000001s : 0.01% optimize.full_micro_interleaved_order_control : 0.000003s : 0.03% 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 : 0.01% 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 : 0.000003s : 0.03% optimize.overlap_grad_ring_attention : 0.000004s : 0.04% optimize.overlap_grad_flash_sp : 0.000023s : 0.23% optimize.begin_end_overlap_inline : 0.000001s : 0.01% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.02% optimize.split_layernorm_comm : 0.000002s : 0.02% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000004s : 0.04% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000010s : 0.10% optimize.symbol_engine_optimizer.elim_not_effective : 0.000013s : 0.13% optimize.symbol_engine_optimizer.opt_reshape : 0.000007s : 0.07% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.10% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000019s : 0.19% pipeline_parallel_scheduler : 0.000002s : 0.02% auto_monad_reorder : 0.000021s : 0.21% get_jit_bprop_graph : 0.000002s : 0.02% rewriter_after_jit_bprop_graph : 0.000006s : 0.06% opt_after_jit_grad : 0.000490s : 4.91% validate : 0.000045s : 0.45% Time group info: ------[substitution.] 0.000107 21 1.82% : 0.000002s : 2: substitution.elim_not_effective 1.47% : 0.000002s : 2: substitution.fold_const_symbol 6.40% : 0.000007s : 4: substitution.graph_param_transform 76.17% : 0.000082s : 1: substitution.inline 3.78% : 0.000004s : 4: substitution.j_node_and_user_rematch 4.95% : 0.000005s : 4: substitution.remove_not_recompute_node 5.41% : 0.000006s : 4: substitution.replace_old_param ------[type_inference.] 0.003874 2 89.59% : 0.003471s : 1: type_inference.infer 10.41% : 0.000403s : 1: type_inference.specialize ------[replace.] 0.000016 1 100.00% : 0.000016s : 1: replace.inline ------[match.] 0.000081 1 100.00% : 0.000081s : 1: match.inline ------[predicate.] 0.000727 967 0.17% : 0.000001s : 9: predicate.accumulaten_eliminater 0.18% : 0.000001s : 4: predicate.ad_related_special_op_eliminate 0.13% : 0.000001s : 8: predicate.addn_check_dump 0.17% : 0.000001s : 9: predicate.addn_zero_filter 0.15% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 0.46% : 0.000003s : 17: predicate.arithmetic_simplify 0.18% : 0.000001s : 9: predicate.cast_eliminate 0.18% : 0.000001s : 8: predicate.check_bprop_eliminate 0.16% : 0.000001s : 8: predicate.compare_switch_simplify 0.05% : 0.000000s : 4: predicate.const_output_eliminate 0.16% : 0.000001s : 8: predicate.depend_value_elim 0.16% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.18% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.18% : 0.000001s : 9: predicate.dict_set_item_eliminator 0.24% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.06% : 0.000000s : 4: predicate.elim_not_effective 0.11% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 0.26% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.20% : 0.000001s : 13: predicate.environ_get_add_eliminate 0.20% : 0.000001s : 13: predicate.environ_get_depend_swap 0.36% : 0.000003s : 21: predicate.environ_get_eliminate 0.22% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.17% : 0.000001s : 10: predicate.exchange_switch_depend_value 0.36% : 0.000003s : 10: predicate.float_depend_g_call 0.13% : 0.000001s : 8: predicate.float_environ_get_switch 0.20% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.05% : 0.000000s : 4: predicate.fold_const_symbol 0.19% : 0.000001s : 8: predicate.get_grad_eliminate 0.08% : 0.000001s : 4: predicate.graph_param_transform 0.16% : 0.000001s : 8: predicate.incorporate_call 0.13% : 0.000001s : 8: predicate.incorporate_call_switch 1.21% : 0.000009s : 43: predicate.inline 0.25% : 0.000002s : 8: predicate.inline_without_move 0.09% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.22% : 0.000002s : 8: predicate.less_batch_normalization 0.32% : 0.000002s : 17: predicate.list_to_tuple_eliminator_ 0.44% : 0.000003s : 26: predicate.load_eliminater 0.30% : 0.000002s : 4: predicate.loop_unroll_after_grad 0.27% : 0.000002s : 13: predicate.loop_unroll_before_grad 0.39% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.20% : 0.000001s : 8: predicate.merge_addn 0.15% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.16% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.14% : 0.000001s : 9: predicate.minmaximum_grad 0.42% : 0.000003s : 4: predicate.mutable_eliminate 0.12% : 0.000001s : 4: predicate.opt_reshape 0.10% : 0.000001s : 4: predicate.parallel_virtual_node 0.24% : 0.000002s : 10: predicate.partial_defer_inline 0.24% : 0.000002s : 13: predicate.partial_eliminate 0.18% : 0.000001s : 9: predicate.print_const_string_wrapper 0.17% : 0.000001s : 8: predicate.reduce_all_const_elim 0.24% : 0.000002s : 9: predicate.reduce_eliminate 0.44% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.20% : 0.000001s : 8: predicate.remove_not_recompute_node 0.27% : 0.000002s : 17: predicate.replace_applicator 0.20% : 0.000001s : 8: predicate.replace_old_param 0.08% : 0.000001s : 4: predicate.reset_defer_inline 0.17% : 0.000001s : 9: predicate.reshape_eliminate 0.16% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.11% : 0.000001s : 4: predicate.row_tensor_eliminate 0.24% : 0.000002s : 8: predicate.same_eliminate 0.12% : 0.000001s : 8: predicate.set_cell_output_no_recompute 79.13% : 0.000575s : 8: predicate.shard_identity_eliminate 0.21% : 0.000001s : 8: predicate.special_op_eliminate 0.20% : 0.000001s : 8: predicate.specialize_transform 0.31% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.21% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.09% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.19% : 0.000001s : 10: predicate.switch_defer_inline 0.32% : 0.000002s : 18: predicate.switch_layer_defer_inline 0.82% : 0.000006s : 35: predicate.switch_simplify 0.16% : 0.000001s : 9: predicate.tile_eliminate 0.15% : 0.000001s : 9: predicate.transpose_eliminate 0.36% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 0.34% : 0.000002s : 17: predicate.tuple_list_get_item_const_eliminator 0.29% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 0.68% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 0.32% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 0.53% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 0.31% : 0.000002s : 17: predicate.tuple_to_list_eliminator_ 0.41% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 0.61% : 0.000004s : 34: predicate.updatestate_useless_node_eliminater 0.10% : 0.000001s : 4: predicate.value_based_eliminate 0.38% : 0.000003s : 8: predicate.virtual_dataset_eliminate 0.17% : 0.000001s : 8: predicate.virtual_output_eliminate 0.07% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.17% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000239 4 9.45% : 0.000023s : 1: func_graph_cloner_run.FuncGraphClonerGraph 90.55% : 0.000216s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.028625 192 0.01% : 0.000004s : 1: ForceFp32Comm 16.96% : 0.004856s : 1: add_attr 16.90% : 0.004839s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.20% : 0.000058s : 1: add_recomputation 0.01% : 0.000004s : 1: assign_add_opt 0.22% : 0.000062s : 1: auto_monad 0.11% : 0.000030s : 1: auto_monad_reorder 0.01% : 0.000003s : 1: begin_end_overlap_inline 0.02% : 0.000007s : 1: bias_add_comm_swap 2.44% : 0.000698s : 1: bootstrap 0.14% : 0.000041s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.06% : 0.000017s : 1: control_data_broadcast_order 0.04% : 0.000010s : 1: convert_after_rewriter 0.09% : 0.000027s : 1: cse_after_recomputation 0.02% : 0.000005s : 1: dataset_repeat_opt 0.09% : 0.000026s : 1: detach_backward 0.03% : 0.000010s : 1: environ_conv 0.07% : 0.000020s : 1: event_method 0.02% : 0.000006s : 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% : 0.000006s : 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% : 0.000008s : 1: label_micro_interleaved_index 1.85% : 0.000530s : 1: loop_unroll 0.01% : 0.000004s : 1: merge_cast_opt 0.02% : 0.000005s : 1: micro_interleaved_order_control 2.58% : 0.000739s : 1: mutable_eliminate 0.03% : 0.000007s : 1: offloading_packed_experts 0.05% : 0.000015s : 1: opt.transform.loop_unroll_optimizer 0.07% : 0.000019s : 1: opt.transform.mutable_eliminate 4.89% : 0.001400s : 78: opt.transform.opt_a 0.10% : 0.000029s : 1: opt.transform.opt_after_cconv 0.08% : 0.000024s : 1: opt.transform.opt_after_jit_grad 0.36% : 0.000104s : 28: opt.transform.opt_b 0.18% : 0.000053s : 2: opt.transform.opt_trans_graph 0.13% : 0.000037s : 4: opt.transform.symbol_engine_opt 11.03% : 0.003156s : 1: opt_a 0.43% : 0.000122s : 1: opt_after_cconv 1.74% : 0.000499s : 1: opt_after_jit_grad 0.81% : 0.000233s : 1: opt_b 19.54% : 0.005593s : 1: optimize 0.09% : 0.000025s : 1: optimize_parallel_all_gather_comm 0.03% : 0.000009s : 1: order_py_execute_after_rewriter 0.09% : 0.000027s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.05% : 0.000014s : 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% : 0.000006s : 1: overlap_param_gather 0.01% : 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% : 0.000007s : 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% : 0.000005s : 1: pipeline_split 0.15% : 0.000042s : 1: pre_auto_parallel 0.09% : 0.000024s : 1: py_interpret_to_execute 0.07% : 0.000019s : 1: py_interpret_to_execute_after_opt_a 0.02% : 0.000005s : 1: remove_cast_before_assign_add 0.18% : 0.000051s : 1: remove_dup_value 1.57% : 0.000448s : 1: renormalize.infer 0.99% : 0.000284s : 1: renormalize.specialize 0.02% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.03% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.16% : 0.000045s : 1: rewriter_after_opt_a 0.19% : 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% : 0.000005s : 1: split_layernorm_comm 0.02% : 0.000005s : 1: split_matmul_comm_elemetwise 0.03% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.29% : 0.000082s : 1: symbol_engine_optimizer 0.29% : 0.000083s : 1: tuple_transform 13.80% : 0.003949s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[dim_type1-False-batch_size_list0-shape1-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_fp16[dim_type1-False-batch_size_list0-shape1-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:39:59.229.728 [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 = 0.0930199, [21] [bootstrap]: 0.00061974 [type_inference]: 0.00415256 [event_method]: 1.444e-05 [auto_monad]: 5.847e-05 [graph_reusing]: 5.06002e-06 [inline]: 2.72001e-06 [add_attr]: 0.0692652, [1] [add_attr_with_inline]: 0.069248, [1] [Cycle 1]: 6.899e-05, [2] [tag_attr]: 2.119e-05 [meta_addattr_fg_expand]: 3.41999e-06 [parallel-infer-symbol]: 4.57e-06 [pre_auto_parallel]: 3.775e-05 [insert-virtual-dataset]: 2.89999e-06 [parallel-infer-symbol-second]: 6.80011e-07 [dataset_repeat_opt]: 2.24999e-06 [pipeline_split]: 1.65001e-06 [optimize]: 0.0179788, [53] [py_interpret_to_execute]: 2.432e-05 [rewriter_before_opt_a]: 5.447e-05 [opt_a]: 0.0153398, [2] [Cycle 1]: 0.00181356, [45] [expand_dump_flag]: 2.58998e-06 [switch_simplify]: 2.535e-05 [loop_unroll]: 1.15e-05 [a_1]: 0.00031711 [with_stream_mark]: 2.447e-05 [recompute_prepare]: 9.30001e-06 [updatestate_depend_eliminate]: 4.59002e-06 [updatestate_assign_eliminate]: 3.98999e-06 [updatestate_loads_eliminate]: 3.83001e-06 [parameter_eliminate]: 2.27001e-06 [a_2]: 8.111e-05 [accelerated_algorithm]: 7.37002e-06 [shard]: 2.60002e-06 [meta_shard_fg_expand]: 2.05002e-06 [shard_inline]: 6.65998e-06 [merge_send_recv]: 9.29e-06 [auto_parallel]: 9.14e-06 [parallel]: 3.425e-05 [flash_sp]: 1.112e-05 [merge_comm]: 3.56001e-06 [allreduce_fusion]: 3.32997e-06 [matmul_add_comm_reduction]: 1.112e-05 [allreduce_slice_to_reducescatter]: 8.30012e-07 [virtual_shard_identity]: 8.65001e-06 [virtual_dataset]: 6.38e-06 [get_grad_eliminate_]: 6.71e-06 [virtual_output]: 6.01998e-06 [merge_forward]: 4.38999e-06 [cell_reuse_recompute_pass]: 1.39998e-06 [offload_activation]: 1.139e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.329e-05 [merge_recompute_call_nodes]: 1.49998e-06 [before_grad]: 1.056e-05 [set_forward_comm_id_for_comm_node_pass]: 3.91999e-06 [meta_fg_expand]: 2.78e-06 [flash_sp_send_recv_attached]: 3.06001e-06 [receive_attached]: 2.52001e-06 [after_resolve]: 1.514e-05 [a_after_grad]: 1.042e-05 [renormalize]: 0.00074133 [add_forward_monad_depend]: 6.64001e-06 [auto_monad_grad]: 2.99001e-06 [auto_monad_eliminator]: 1.651e-05 [cse]: 4.473e-05 [a_3]: 5.035e-05 [Cycle 2]: 0.0135121, [45] [expand_dump_flag]: 1.62999e-06 [switch_simplify]: 8.81002e-06 [loop_unroll]: 6.14001e-06 [a_1]: 0.00012547 [with_stream_mark]: 1.586e-05 [recompute_prepare]: 7.62002e-06 [updatestate_depend_eliminate]: 3.81999e-06 [updatestate_assign_eliminate]: 2.99001e-06 [updatestate_loads_eliminate]: 2.70997e-06 [parameter_eliminate]: 2.31e-06 [a_2]: 7.188e-05 [accelerated_algorithm]: 6.88e-06 [shard]: 1.47999e-06 [meta_shard_fg_expand]: 1.94e-06 [shard_inline]: 7.08e-06 [merge_send_recv]: 7.23e-06 [auto_parallel]: 8.32e-06 [parallel]: 6.74999e-06 [flash_sp]: 5.29e-06 [merge_comm]: 3.48e-06 [allreduce_fusion]: 3.28998e-06 [matmul_add_comm_reduction]: 1.608e-05 [allreduce_slice_to_reducescatter]: 5.59987e-07 [virtual_shard_identity]: 7.27002e-06 [virtual_dataset]: 6.04999e-06 [get_grad_eliminate_]: 6.12001e-06 [virtual_output]: 5.72999e-06 [merge_forward]: 4.57e-06 [cell_reuse_recompute_pass]: 1.68997e-06 [offload_activation]: 1.255e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.565e-05 [merge_recompute_call_nodes]: 1.15999e-06 [before_grad]: 9.66998e-06 [set_forward_comm_id_for_comm_node_pass]: 3.86001e-06 [meta_fg_expand]: 2.54999e-06 [flash_sp_send_recv_attached]: 1.05001e-06 [receive_attached]: 1.35999e-06 [after_resolve]: 1.119e-05 [a_after_grad]: 9.26998e-06 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 0.0126235 [auto_monad_grad]: 6.53e-06 [auto_monad_eliminator]: 3.457e-05 [cse]: 5.59e-05 [a_3]: 7.575e-05 [py_interpret_to_execute_after_opt_a]: 2.472e-05 [slice_cell_reuse_recomputed_activation]: 3.43999e-06 [rewriter_after_opt_a]: 5.745e-05 [convert_after_rewriter]: 8.59998e-06 [order_py_execute_after_rewriter]: 5.99999e-06 [mutable_eliminate]: 0.00085016 [opt_b]: 0.00024104, [1] [Cycle 1]: 0.00023004, [7] [b_1]: 0.00013456 [b_2]: 9.40001e-06 [updatestate_depend_eliminate]: 8.78001e-06 [updatestate_assign_eliminate]: 3.97e-06 [updatestate_loads_eliminate]: 3.41001e-06 [renormalize]: 6.19999e-07 [cse]: 3.288e-05 [optimize_parallel_all_gather_comm]: 2.299e-05 [overlap_param_gather]: 2.02001e-06 [cconv]: 4.041e-05 [loop_unroll]: 0.00048512 [opt_after_cconv]: 0.00011823, [1] [Cycle 1]: 0.0001114, [7] [c_1]: 3.215e-05 [parameter_eliminate]: 4.60001e-06 [updatestate_depend_eliminate]: 6.51e-06 [updatestate_assign_eliminate]: 3.08e-06 [updatestate_loads_eliminate]: 2.51e-06 [cse]: 2.616e-05 [renormalize]: 6.09987e-07 [remove_dup_value]: 5.39e-05 [tuple_transform]: 8.361e-05, [1] [Cycle 1]: 7.742e-05, [4] [d_1]: 5.033e-05 [none_parameter_eliminate]: 1.92999e-06 [renormalize]: 1.80007e-07 [switch_simplify]: 6.93e-06 [partial_unused_args_eliminate]: 1.99999e-06 [add_recomputation]: 5.73e-05 [cse_after_recomputation]: 3.211e-05, [1] [Cycle 1]: 2.655e-05, [1] [cse]: 2.002e-05 [environ_conv]: 7.16001e-06 [swap_dp_allreduce_reducescatter]: 5.49e-06 [bias_add_comm_swap]: 3.56999e-06 [label_micro_interleaved_index]: 5.58997e-06 [label_fine_grained_interleaved_index]: 2.68e-06 [merge_cast_opt]: 1.39998e-06 [slice_recompute_activation]: 2.31e-06 [micro_interleaved_order_control]: 2.74999e-06 [assign_add_opt]: 1.31998e-06 [ForceFp32Comm]: 9.69972e-07 [remove_cast_before_assign_add]: 1.15001e-06 [full_micro_interleaved_order_control]: 2.46e-06 [reorder_send_recv_between_fp_bp]: 3.08e-06 [comm_op_add_attrs]: 1.10001e-06 [add_comm_op_reuse_tag]: 1.14e-06 [interleave_split_concat_branches]: 1.34e-06 [interleave_parallel_branches]: 1.10001e-06 [overlap_opt_shard_in_pipeline]: 1.99999e-06 [overlap_opt_shard_grad_in_pipeline]: 2.03002e-06 [control_data_broadcast_order]: 1.468e-05 [grouped_pairwise_exchange_alltoall]: 1.98002e-06 [offloading_packed_experts]: 4.72e-06 [overlap_recompute_and_grad_model_parallel]: 5.25999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.28002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.59998e-06 [overlap_recompute_comm]: 2.66999e-06 [overlap_grad_ring_attention]: 4.79e-06 [overlap_grad_flash_sp]: 2.431e-05 [begin_end_overlap_inline]: 5.60016e-07 [split_matmul_comm_elemetwise]: 2.32999e-06 [split_layernorm_comm]: 2.19001e-06 [handle_group_info]: 1.45001e-06 [symbol_engine_optimizer]: 8.785e-05, [1] [Cycle 1]: 8.221e-05, [6] [build]: 4.22e-06 [elim_shapecalc]: 1.156e-05 [elim_not_effective]: 1.563e-05 [opt_reshape]: 7.92e-06 [fold_const_symbol]: 1.048e-05 [renormalize]: 1.69995e-07 [detach_backward]: 2.64999e-06 [pipeline_parallel_scheduler]: 1.78997e-06 [auto_monad_reorder]: 2.264e-05 [get_jit_bprop_graph]: 2.32999e-06 [rewriter_after_jit_bprop_graph]: 4.61002e-06 [opt_after_jit_grad]: 0.00057868 [validate]: 4.936e-05 Sums bootstrap : 0.000620s : 2.75% type_inference : 0.004153s : 18.40% event_method : 0.000014s : 0.06% auto_monad : 0.000058s : 0.26% graph_reusing : 0.000005s : 0.02% inline : 0.000003s : 0.01% add_attr.add_attr_with_inline.tag_attr : 0.000021s : 0.09% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000003s : 0.02% parallel-infer-symbol : 0.000005s : 0.02% pre_auto_parallel : 0.000038s : 0.17% 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 : 0.000024s : 0.11% optimize.rewriter_before_opt_a : 0.000054s : 0.24% optimize.opt_a.expand_dump_flag : 0.000004s : 0.02% optimize.opt_a.switch_simplify : 0.000034s : 0.15% optimize.opt_a.loop_unroll : 0.000018s : 0.08% optimize.opt_a.a_1 : 0.000443s : 1.96% optimize.opt_a.with_stream_mark : 0.000040s : 0.18% optimize.opt_a.recompute_prepare : 0.000017s : 0.07% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.04% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.03% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.03% optimize.opt_a.parameter_eliminate : 0.000005s : 0.02% optimize.opt_a.a_2 : 0.000153s : 0.68% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.06% optimize.opt_a.shard : 0.000004s : 0.02% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.02% optimize.opt_a.shard_inline : 0.000014s : 0.06% optimize.opt_a.merge_send_recv : 0.000017s : 0.07% optimize.opt_a.auto_parallel : 0.000017s : 0.08% optimize.opt_a.parallel : 0.000041s : 0.18% optimize.opt_a.flash_sp : 0.000016s : 0.07% 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 : 0.000027s : 0.12% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000016s : 0.07% optimize.opt_a.virtual_dataset : 0.000012s : 0.06% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.06% optimize.opt_a.virtual_output : 0.000012s : 0.05% optimize.opt_a.merge_forward : 0.000009s : 0.04% optimize.opt_a.cell_reuse_recompute_pass : 0.000003s : 0.01% optimize.opt_a.offload_activation : 0.000024s : 0.11% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000029s : 0.13% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.01% optimize.opt_a.before_grad : 0.000020s : 0.09% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.03% optimize.opt_a.meta_fg_expand : 0.000005s : 0.02% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.02% optimize.opt_a.receive_attached : 0.000004s : 0.02% optimize.opt_a.after_resolve : 0.000026s : 0.12% optimize.opt_a.a_after_grad : 0.000020s : 0.09% optimize.opt_a.renormalize : 0.000741s : 3.29% optimize.opt_a.add_forward_monad_depend : 0.012630s : 55.96% optimize.opt_a.auto_monad_grad : 0.000010s : 0.04% optimize.opt_a.auto_monad_eliminator : 0.000051s : 0.23% optimize.opt_a.cse : 0.000101s : 0.45% optimize.opt_a.a_3 : 0.000126s : 0.56% optimize.py_interpret_to_execute_after_opt_a : 0.000025s : 0.11% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.02% optimize.rewriter_after_opt_a : 0.000057s : 0.25% optimize.convert_after_rewriter : 0.000009s : 0.04% optimize.order_py_execute_after_rewriter : 0.000006s : 0.03% optimize.mutable_eliminate : 0.000850s : 3.77% optimize.opt_b.b_1 : 0.000135s : 0.60% optimize.opt_b.b_2 : 0.000009s : 0.04% optimize.opt_b.updatestate_depend_eliminate : 0.000009s : 0.04% optimize.opt_b.updatestate_assign_eliminate : 0.000004s : 0.02% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.02% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000033s : 0.15% optimize.optimize_parallel_all_gather_comm : 0.000023s : 0.10% optimize.overlap_param_gather : 0.000002s : 0.01% optimize.cconv : 0.000040s : 0.18% optimize.loop_unroll : 0.000485s : 2.15% optimize.opt_after_cconv.c_1 : 0.000032s : 0.14% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.02% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000007s : 0.03% 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 : 0.12% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000054s : 0.24% optimize.tuple_transform.d_1 : 0.000050s : 0.22% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.01% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.03% optimize.partial_unused_args_eliminate : 0.000002s : 0.01% optimize.add_recomputation : 0.000057s : 0.25% optimize.cse_after_recomputation.cse : 0.000020s : 0.09% optimize.environ_conv : 0.000007s : 0.03% optimize.swap_dp_allreduce_reducescatter : 0.000005s : 0.02% optimize.bias_add_comm_swap : 0.000004s : 0.02% optimize.label_micro_interleaved_index : 0.000006s : 0.02% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.01% optimize.merge_cast_opt : 0.000001s : 0.01% optimize.slice_recompute_activation : 0.000002s : 0.01% optimize.micro_interleaved_order_control : 0.000003s : 0.01% optimize.assign_add_opt : 0.000001s : 0.01% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.01% 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 : 0.01% optimize.interleave_split_concat_branches : 0.000001s : 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 : 0.000015s : 0.07% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.01% optimize.offloading_packed_experts : 0.000005s : 0.02% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.02% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.01% optimize.overlap_recompute_allgather_and_fa_grad : 0.000002s : 0.01% optimize.overlap_recompute_comm : 0.000003s : 0.01% optimize.overlap_grad_ring_attention : 0.000005s : 0.02% optimize.overlap_grad_flash_sp : 0.000024s : 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 : 0.000016s : 0.07% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.04% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.05% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.01% pipeline_parallel_scheduler : 0.000002s : 0.01% auto_monad_reorder : 0.000023s : 0.10% get_jit_bprop_graph : 0.000002s : 0.01% rewriter_after_jit_bprop_graph : 0.000005s : 0.02% opt_after_jit_grad : 0.000579s : 2.56% validate : 0.000049s : 0.22% Time group info: ------[substitution.] 0.000122 21 1.83% : 0.000002s : 2: substitution.elim_not_effective 1.44% : 0.000002s : 2: substitution.fold_const_symbol 5.62% : 0.000007s : 4: substitution.graph_param_transform 79.17% : 0.000097s : 1: substitution.inline 3.41% : 0.000004s : 4: substitution.j_node_and_user_rematch 4.39% : 0.000005s : 4: substitution.remove_not_recompute_node 4.13% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.004091 2 89.13% : 0.003646s : 1: type_inference.infer 10.87% : 0.000445s : 1: type_inference.specialize ------[replace.] 0.000020 1 100.00% : 0.000020s : 1: replace.inline ------[match.] 0.000096 1 100.00% : 0.000096s : 1: match.inline ------[predicate.] 0.000166 967 0.82% : 0.000001s : 9: predicate.accumulaten_eliminater 1.05% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.60% : 0.000001s : 8: predicate.addn_check_dump 0.71% : 0.000001s : 9: predicate.addn_zero_filter 0.65% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.28% : 0.000004s : 17: predicate.arithmetic_simplify 0.89% : 0.000001s : 9: predicate.cast_eliminate 1.09% : 0.000002s : 8: predicate.check_bprop_eliminate 0.59% : 0.000001s : 8: predicate.compare_switch_simplify 0.22% : 0.000000s : 4: predicate.const_output_eliminate 0.66% : 0.000001s : 8: predicate.depend_value_elim 0.77% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.76% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.73% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.45% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.40% : 0.000001s : 4: predicate.elim_not_effective 0.64% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.19% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.97% : 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 1.00% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.69% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.67% : 0.000003s : 10: predicate.float_depend_g_call 0.58% : 0.000001s : 8: predicate.float_environ_get_switch 1.01% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.25% : 0.000000s : 4: predicate.fold_const_symbol 0.95% : 0.000002s : 8: predicate.get_grad_eliminate 0.38% : 0.000001s : 4: predicate.graph_param_transform 0.66% : 0.000001s : 8: predicate.incorporate_call 0.58% : 0.000001s : 8: predicate.incorporate_call_switch 6.03% : 0.000010s : 43: predicate.inline 0.89% : 0.000001s : 8: predicate.inline_without_move 0.40% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.14% : 0.000002s : 8: predicate.less_batch_normalization 1.63% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.21% : 0.000004s : 26: predicate.load_eliminater 1.41% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.40% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.94% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.67% : 0.000001s : 8: predicate.merge_addn 0.95% : 0.000002s : 8: predicate.micro_step_allgather_replace 0.91% : 0.000002s : 8: predicate.mini_step_allgather_replace 0.60% : 0.000001s : 9: predicate.minmaximum_grad 1.35% : 0.000002s : 4: predicate.mutable_eliminate 0.66% : 0.000001s : 4: predicate.opt_reshape 0.56% : 0.000001s : 4: predicate.parallel_virtual_node 1.32% : 0.000002s : 10: predicate.partial_defer_inline 1.07% : 0.000002s : 13: predicate.partial_eliminate 0.69% : 0.000001s : 9: predicate.print_const_string_wrapper 0.66% : 0.000001s : 8: predicate.reduce_all_const_elim 0.95% : 0.000002s : 9: predicate.reduce_eliminate 1.93% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.64% : 0.000001s : 8: predicate.remove_not_recompute_node 1.84% : 0.000003s : 17: predicate.replace_applicator 0.79% : 0.000001s : 8: predicate.replace_old_param 0.52% : 0.000001s : 4: predicate.reset_defer_inline 0.83% : 0.000001s : 9: predicate.reshape_eliminate 1.00% : 0.000002s : 8: predicate.row_tensor_add_zeros_like 0.51% : 0.000001s : 4: predicate.row_tensor_eliminate 1.44% : 0.000002s : 8: predicate.same_eliminate 0.62% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.86% : 0.000001s : 8: predicate.shard_identity_eliminate 0.98% : 0.000002s : 8: predicate.special_op_eliminate 0.76% : 0.000001s : 8: predicate.specialize_transform 1.96% : 0.000003s : 8: predicate.split_environ_get_set_with_tuple_value 1.04% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.56% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.82% : 0.000001s : 10: predicate.switch_defer_inline 1.59% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.61% : 0.000006s : 35: predicate.switch_simplify 0.70% : 0.000001s : 9: predicate.tile_eliminate 0.84% : 0.000001s : 9: predicate.transpose_eliminate 1.96% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.55% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.99% : 0.000003s : 17: predicate.tuple_list_get_item_depend_reorder 3.22% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.50% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 2.28% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.84% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.82% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.60% : 0.000004s : 34: predicate.updatestate_useless_node_eliminater 0.47% : 0.000001s : 4: predicate.value_based_eliminate 0.76% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.73% : 0.000001s : 8: predicate.virtual_output_eliminate 0.29% : 0.000000s : 4: predicate.virtual_view_grad_eliminate 0.71% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000268 4 9.28% : 0.000025s : 1: func_graph_cloner_run.FuncGraphClonerGraph 90.72% : 0.000244s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.182014 192 0.00% : 0.000004s : 1: ForceFp32Comm 38.06% : 0.069272s : 1: add_attr 38.05% : 0.069252s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.03% : 0.000061s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.04% : 0.000064s : 1: auto_monad 0.02% : 0.000034s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.37% : 0.000669s : 1: bootstrap 0.02% : 0.000044s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.01% : 0.000019s : 1: control_data_broadcast_order 0.01% : 0.000012s : 1: convert_after_rewriter 0.02% : 0.000035s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.01% : 0.000011s : 1: environ_conv 0.01% : 0.000021s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000009s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000009s : 1: label_micro_interleaved_index 0.27% : 0.000495s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.47% : 0.000861s : 1: mutable_eliminate 0.00% : 0.000008s : 1: offloading_packed_experts 0.02% : 0.000037s : 1: opt.transform.loop_unroll_optimizer 0.01% : 0.000020s : 1: opt.transform.mutable_eliminate 0.48% : 0.000867s : 78: opt.transform.opt_a 0.02% : 0.000031s : 1: opt.transform.opt_after_cconv 0.01% : 0.000027s : 1: opt.transform.opt_after_jit_grad 0.06% : 0.000112s : 28: opt.transform.opt_b 0.03% : 0.000055s : 2: opt.transform.opt_trans_graph 0.02% : 0.000042s : 4: opt.transform.symbol_engine_opt 8.43% : 0.015344s : 1: opt_a 0.07% : 0.000123s : 1: opt_after_cconv 0.32% : 0.000591s : 1: opt_after_jit_grad 0.13% : 0.000245s : 1: opt_b 9.88% : 0.017985s : 1: optimize 0.01% : 0.000026s : 1: optimize_parallel_all_gather_comm 0.01% : 0.000010s : 1: order_py_execute_after_rewriter 0.02% : 0.000028s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000009s : 1: overlap_grad_ring_attention 0.01% : 0.000017s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000005s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000006s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000008s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.02% : 0.000043s : 1: pre_auto_parallel 0.02% : 0.000029s : 1: py_interpret_to_execute 0.02% : 0.000028s : 1: py_interpret_to_execute_after_opt_a 0.01% : 0.000014s : 1: remove_cast_before_assign_add 0.03% : 0.000059s : 1: remove_dup_value 0.24% : 0.000436s : 1: renormalize.infer 0.16% : 0.000296s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000008s : 1: rewriter_after_jit_bprop_graph 0.03% : 0.000062s : 1: rewriter_after_opt_a 0.03% : 0.000059s : 1: rewriter_before_opt_a 0.00% : 0.000007s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.05% : 0.000091s : 1: symbol_engine_optimizer 0.05% : 0.000087s : 1: tuple_transform 2.29% : 0.004175s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[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_fp16[dim_type1-False-batch_size_list1-shape0-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:40:02.375.77 [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 = 0.282595, [21] [bootstrap]: 0.00646161 [type_inference]: 0.119463 [event_method]: 4.954e-05 [auto_monad]: 9.268e-05 [graph_reusing]: 5.92999e-06 [inline]: 3.31001e-06 [add_attr]: 0.142875, [1] [add_attr_with_inline]: 0.142802, [1] [Cycle 1]: 8.494e-05, [2] [tag_attr]: 2.572e-05 [meta_addattr_fg_expand]: 6.11e-06 [parallel-infer-symbol]: 4.1e-06 [pre_auto_parallel]: 0.00010638 [insert-virtual-dataset]: 2.74001e-06 [parallel-infer-symbol-second]: 7.79983e-07 [dataset_repeat_opt]: 2.07001e-06 [pipeline_split]: 2.13998e-06 [optimize]: 0.0118195, [53] [py_interpret_to_execute]: 0.00029456 [rewriter_before_opt_a]: 5.488e-05 [opt_a]: 0.00628565, [2] [Cycle 1]: 0.00518431, [45] [expand_dump_flag]: 2.87002e-06 [switch_simplify]: 2.589e-05 [loop_unroll]: 1.211e-05 [a_1]: 0.00049494 [with_stream_mark]: 2.26e-05 [recompute_prepare]: 8.28001e-06 [updatestate_depend_eliminate]: 4.4e-06 [updatestate_assign_eliminate]: 3.39001e-06 [updatestate_loads_eliminate]: 3.71001e-06 [parameter_eliminate]: 2.81e-06 [a_2]: 8.039e-05 [accelerated_algorithm]: 7.11999e-06 [shard]: 2.27999e-06 [meta_shard_fg_expand]: 1.81998e-06 [shard_inline]: 6.76e-06 [merge_send_recv]: 8.58001e-06 [auto_parallel]: 7.86001e-06 [parallel]: 0.0006208 [flash_sp]: 9.99001e-06 [merge_comm]: 4.37e-06 [allreduce_fusion]: 4.72e-06 [matmul_add_comm_reduction]: 1.159e-05 [allreduce_slice_to_reducescatter]: 8.60018e-07 [virtual_shard_identity]: 9.65002e-06 [virtual_dataset]: 6.56e-06 [get_grad_eliminate_]: 6.37001e-06 [virtual_output]: 6.34999e-06 [merge_forward]: 4.21001e-06 [cell_reuse_recompute_pass]: 1.47999e-06 [offload_activation]: 1.101e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.469e-05 [merge_recompute_call_nodes]: 1.99999e-06 [before_grad]: 1.162e-05 [set_forward_comm_id_for_comm_node_pass]: 4.42998e-06 [meta_fg_expand]: 3.21001e-06 [flash_sp_send_recv_attached]: 2.65002e-06 [receive_attached]: 2.36998e-06 [after_resolve]: 1.373e-05 [a_after_grad]: 1.061e-05 [renormalize]: 0.00235596 [add_forward_monad_depend]: 6.19001e-06 [auto_monad_grad]: 3.01001e-06 [auto_monad_eliminator]: 1.683e-05 [cse]: 4.278e-05 [a_3]: 4.957e-05 [Cycle 2]: 0.0010899, [45] [expand_dump_flag]: 1.14e-06 [switch_simplify]: 7.58999e-06 [loop_unroll]: 6.19999e-06 [a_1]: 0.00012655 [with_stream_mark]: 1.483e-05 [recompute_prepare]: 6.24999e-06 [updatestate_depend_eliminate]: 3.51001e-06 [updatestate_assign_eliminate]: 2.58998e-06 [updatestate_loads_eliminate]: 2.64001e-06 [parameter_eliminate]: 1.26002e-06 [a_2]: 7.016e-05 [accelerated_algorithm]: 5.96e-06 [shard]: 1.02e-06 [meta_shard_fg_expand]: 1.75001e-06 [shard_inline]: 6.28e-06 [merge_send_recv]: 5.82999e-06 [auto_parallel]: 5.61998e-06 [parallel]: 5.67001e-06 [flash_sp]: 3.85e-06 [merge_comm]: 3.24001e-06 [allreduce_fusion]: 3.44001e-06 [matmul_add_comm_reduction]: 8.177e-05 [allreduce_slice_to_reducescatter]: 3.59985e-07 [virtual_shard_identity]: 7.86001e-06 [virtual_dataset]: 6.34001e-06 [get_grad_eliminate_]: 5.84999e-06 [virtual_output]: 5.81e-06 [merge_forward]: 3.19001e-06 [cell_reuse_recompute_pass]: 1.80001e-06 [offload_activation]: 9.648e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.378e-05 [merge_recompute_call_nodes]: 9.30013e-07 [before_grad]: 9.94999e-06 [set_forward_comm_id_for_comm_node_pass]: 3.61999e-06 [meta_fg_expand]: 2.41e-06 [flash_sp_send_recv_attached]: 9.20001e-07 [receive_attached]: 1.21997e-06 [after_resolve]: 9.95002e-06 [a_after_grad]: 9.42001e-06 [renormalize]: 8.00064e-08 [add_forward_monad_depend]: 1.30999e-06 [auto_monad_grad]: 8.39995e-07 [auto_monad_eliminator]: 7.36001e-06 [cse]: 1.922e-05 [a_3]: 3.553e-05 [py_interpret_to_execute_after_opt_a]: 1.251e-05 [slice_cell_reuse_recomputed_activation]: 2.17999e-06 [rewriter_after_opt_a]: 3.904e-05 [convert_after_rewriter]: 7.03e-06 [order_py_execute_after_rewriter]: 6.35002e-06 [mutable_eliminate]: 0.00083392 [opt_b]: 0.00040387, [1] [Cycle 1]: 0.00039726, [7] [b_1]: 0.00012197 [b_2]: 7.93001e-06 [updatestate_depend_eliminate]: 6.33998e-06 [updatestate_assign_eliminate]: 2.68e-06 [updatestate_loads_eliminate]: 2.79001e-06 [renormalize]: 5.8001e-07 [cse]: 0.00021908 [optimize_parallel_all_gather_comm]: 1.782e-05 [overlap_param_gather]: 1.92001e-06 [cconv]: 2.8e-05 [loop_unroll]: 0.0004496 [opt_after_cconv]: 0.00124691, [1] [Cycle 1]: 0.0012402, [7] [c_1]: 2.978e-05 [parameter_eliminate]: 3.36999e-06 [updatestate_depend_eliminate]: 6.44001e-06 [updatestate_assign_eliminate]: 2.81999e-06 [updatestate_loads_eliminate]: 2.35002e-06 [cse]: 2.847e-05 [renormalize]: 7.89994e-07 [remove_dup_value]: 4.213e-05 [tuple_transform]: 0.00048729, [1] [Cycle 1]: 7.691e-05, [4] [d_1]: 4.789e-05 [none_parameter_eliminate]: 3.06001e-06 [renormalize]: 2.29978e-07 [switch_simplify]: 7.19001e-06 [partial_unused_args_eliminate]: 2.49999e-06 [add_recomputation]: 8.214e-05 [cse_after_recomputation]: 0.0001429, [1] [Cycle 1]: 0.00013773, [1] [cse]: 0.00013144 [environ_conv]: 6.21e-06 [swap_dp_allreduce_reducescatter]: 5.44e-06 [bias_add_comm_swap]: 3.37002e-06 [label_micro_interleaved_index]: 4.65999e-06 [label_fine_grained_interleaved_index]: 2.64001e-06 [merge_cast_opt]: 1.45999e-06 [slice_recompute_activation]: 2.24001e-06 [micro_interleaved_order_control]: 2.36e-06 [assign_add_opt]: 1.30001e-06 [ForceFp32Comm]: 1.06002e-06 [remove_cast_before_assign_add]: 1.02e-06 [full_micro_interleaved_order_control]: 2.54999e-06 [reorder_send_recv_between_fp_bp]: 2.93998e-06 [comm_op_add_attrs]: 1.10001e-06 [add_comm_op_reuse_tag]: 1.00001e-06 [interleave_split_concat_branches]: 1.22999e-06 [interleave_parallel_branches]: 1.07e-06 [overlap_opt_shard_in_pipeline]: 1.59998e-06 [overlap_opt_shard_grad_in_pipeline]: 1.79998e-06 [control_data_broadcast_order]: 1.7e-05 [grouped_pairwise_exchange_alltoall]: 1.71e-06 [offloading_packed_experts]: 4.67998e-06 [overlap_recompute_and_grad_model_parallel]: 5.33002e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.19e-06 [overlap_recompute_allgather_and_fa_grad]: 1.45999e-06 [overlap_recompute_comm]: 2.10002e-06 [overlap_grad_ring_attention]: 3.91999e-06 [overlap_grad_flash_sp]: 2.226e-05 [begin_end_overlap_inline]: 5.29981e-07 [split_matmul_comm_elemetwise]: 2.07999e-06 [split_layernorm_comm]: 1.76e-06 [handle_group_info]: 1.20999e-06 [symbol_engine_optimizer]: 8.26e-05, [1] [Cycle 1]: 7.754e-05, [6] [build]: 3.83001e-06 [elim_shapecalc]: 1.273e-05 [elim_not_effective]: 1.413e-05 [opt_reshape]: 7.66999e-06 [fold_const_symbol]: 9.89001e-06 [renormalize]: 1.69995e-07 [detach_backward]: 2.10002e-06 [pipeline_parallel_scheduler]: 1.47999e-06 [auto_monad_reorder]: 2.213e-05 [get_jit_bprop_graph]: 1.83002e-06 [rewriter_after_jit_bprop_graph]: 4.50999e-06 [opt_after_jit_grad]: 0.0007648 [validate]: 0.00010878 Sums bootstrap : 0.006462s : 4.81% type_inference : 0.119463s : 88.93% event_method : 0.000050s : 0.04% auto_monad : 0.000093s : 0.07% graph_reusing : 0.000006s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000026s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000006s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000106s : 0.08% 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 : 0.000295s : 0.22% optimize.rewriter_before_opt_a : 0.000055s : 0.04% optimize.opt_a.expand_dump_flag : 0.000004s : 0.00% optimize.opt_a.switch_simplify : 0.000033s : 0.02% optimize.opt_a.loop_unroll : 0.000018s : 0.01% optimize.opt_a.a_1 : 0.000621s : 0.46% optimize.opt_a.with_stream_mark : 0.000037s : 0.03% optimize.opt_a.recompute_prepare : 0.000015s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000008s : 0.01% optimize.opt_a.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_a.parameter_eliminate : 0.000004s : 0.00% optimize.opt_a.a_2 : 0.000151s : 0.11% optimize.opt_a.accelerated_algorithm : 0.000013s : 0.01% optimize.opt_a.shard : 0.000003s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000004s : 0.00% optimize.opt_a.shard_inline : 0.000013s : 0.01% optimize.opt_a.merge_send_recv : 0.000014s : 0.01% optimize.opt_a.auto_parallel : 0.000013s : 0.01% optimize.opt_a.parallel : 0.000626s : 0.47% optimize.opt_a.flash_sp : 0.000014s : 0.01% optimize.opt_a.merge_comm : 0.000008s : 0.01% optimize.opt_a.allreduce_fusion : 0.000008s : 0.01% optimize.opt_a.matmul_add_comm_reduction : 0.000093s : 0.07% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000018s : 0.01% optimize.opt_a.virtual_dataset : 0.000013s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000012s : 0.01% optimize.opt_a.virtual_output : 0.000012s : 0.01% optimize.opt_a.merge_forward : 0.000007s : 0.01% optimize.opt_a.cell_reuse_recompute_pass : 0.000003s : 0.00% optimize.opt_a.offload_activation : 0.000107s : 0.08% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000028s : 0.02% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000022s : 0.02% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.01% optimize.opt_a.meta_fg_expand : 0.000006s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.00% optimize.opt_a.receive_attached : 0.000004s : 0.00% optimize.opt_a.after_resolve : 0.000024s : 0.02% optimize.opt_a.a_after_grad : 0.000020s : 0.01% optimize.opt_a.renormalize : 0.002356s : 1.75% optimize.opt_a.add_forward_monad_depend : 0.000007s : 0.01% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000024s : 0.02% optimize.opt_a.cse : 0.000062s : 0.05% optimize.opt_a.a_3 : 0.000085s : 0.06% optimize.py_interpret_to_execute_after_opt_a : 0.000013s : 0.01% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000039s : 0.03% optimize.convert_after_rewriter : 0.000007s : 0.01% optimize.order_py_execute_after_rewriter : 0.000006s : 0.00% optimize.mutable_eliminate : 0.000834s : 0.62% optimize.opt_b.b_1 : 0.000122s : 0.09% optimize.opt_b.b_2 : 0.000008s : 0.01% optimize.opt_b.updatestate_depend_eliminate : 0.000006s : 0.00% 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 : 0.000219s : 0.16% optimize.optimize_parallel_all_gather_comm : 0.000018s : 0.01% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000028s : 0.02% optimize.loop_unroll : 0.000450s : 0.33% optimize.opt_after_cconv.c_1 : 0.000030s : 0.02% optimize.opt_after_cconv.parameter_eliminate : 0.000003s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000002s : 0.00% optimize.opt_after_cconv.cse : 0.000028s : 0.02% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000042s : 0.03% optimize.tuple_transform.d_1 : 0.000048s : 0.04% optimize.tuple_transform.none_parameter_eliminate : 0.000003s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000007s : 0.01% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000082s : 0.06% optimize.cse_after_recomputation.cse : 0.000131s : 0.10% optimize.environ_conv : 0.000006s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000005s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000002s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000017s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000005s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000004s : 0.00% optimize.overlap_grad_flash_sp : 0.000022s : 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 : 0.000013s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000014s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.01% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 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 : 0.000022s : 0.02% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000005s : 0.00% opt_after_jit_grad : 0.000765s : 0.57% validate : 0.000109s : 0.08% Time group info: ------[substitution.] 0.000208 21 0.91% : 0.000002s : 2: substitution.elim_not_effective 0.62% : 0.000001s : 2: substitution.fold_const_symbol 2.91% : 0.000006s : 4: substitution.graph_param_transform 88.92% : 0.000185s : 1: substitution.inline 2.00% : 0.000004s : 4: substitution.j_node_and_user_rematch 2.44% : 0.000005s : 4: substitution.remove_not_recompute_node 2.20% : 0.000005s : 4: substitution.replace_old_param ------[type_inference.] 0.119275 2 97.44% : 0.116222s : 1: type_inference.infer 2.56% : 0.003052s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000183 1 100.00% : 0.000183s : 1: match.inline ------[predicate.] 0.000157 967 0.82% : 0.000001s : 9: predicate.accumulaten_eliminater 0.76% : 0.000001s : 4: predicate.ad_related_special_op_eliminate 0.63% : 0.000001s : 8: predicate.addn_check_dump 1.00% : 0.000002s : 9: predicate.addn_zero_filter 0.66% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.36% : 0.000004s : 17: predicate.arithmetic_simplify 0.83% : 0.000001s : 9: predicate.cast_eliminate 0.81% : 0.000001s : 8: predicate.check_bprop_eliminate 0.59% : 0.000001s : 8: predicate.compare_switch_simplify 0.26% : 0.000000s : 4: predicate.const_output_eliminate 0.76% : 0.000001s : 8: predicate.depend_value_elim 0.76% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.94% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.92% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.70% : 0.000003s : 8: predicate.dumpgradient_eliminate 0.31% : 0.000000s : 4: predicate.elim_not_effective 0.77% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.10% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.06% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.99% : 0.000002s : 13: predicate.environ_get_depend_swap 1.70% : 0.000003s : 21: predicate.environ_get_eliminate 1.01% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.86% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.60% : 0.000003s : 10: predicate.float_depend_g_call 0.67% : 0.000001s : 8: predicate.float_environ_get_switch 0.94% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.24% : 0.000000s : 4: predicate.fold_const_symbol 0.89% : 0.000001s : 8: predicate.get_grad_eliminate 0.45% : 0.000001s : 4: predicate.graph_param_transform 0.70% : 0.000001s : 8: predicate.incorporate_call 0.60% : 0.000001s : 8: predicate.incorporate_call_switch 5.89% : 0.000009s : 43: predicate.inline 1.00% : 0.000002s : 8: predicate.inline_without_move 0.42% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.07% : 0.000002s : 8: predicate.less_batch_normalization 1.95% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.08% : 0.000003s : 26: predicate.load_eliminater 0.92% : 0.000001s : 4: predicate.loop_unroll_after_grad 1.19% : 0.000002s : 13: predicate.loop_unroll_before_grad 2.15% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.68% : 0.000001s : 8: predicate.merge_addn 0.72% : 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 1.10% : 0.000002s : 4: predicate.mutable_eliminate 0.66% : 0.000001s : 4: predicate.opt_reshape 0.44% : 0.000001s : 4: predicate.parallel_virtual_node 1.23% : 0.000002s : 10: predicate.partial_defer_inline 1.16% : 0.000002s : 13: predicate.partial_eliminate 0.82% : 0.000001s : 9: predicate.print_const_string_wrapper 0.73% : 0.000001s : 8: predicate.reduce_all_const_elim 1.34% : 0.000002s : 9: predicate.reduce_eliminate 1.99% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.60% : 0.000001s : 8: predicate.remove_not_recompute_node 1.18% : 0.000002s : 17: predicate.replace_applicator 0.71% : 0.000001s : 8: predicate.replace_old_param 0.38% : 0.000001s : 4: predicate.reset_defer_inline 1.12% : 0.000002s : 9: predicate.reshape_eliminate 0.81% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.52% : 0.000001s : 4: predicate.row_tensor_eliminate 1.03% : 0.000002s : 8: predicate.same_eliminate 0.59% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.37% : 0.000002s : 8: predicate.shard_identity_eliminate 1.00% : 0.000002s : 8: predicate.special_op_eliminate 0.88% : 0.000001s : 8: predicate.specialize_transform 1.19% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.95% : 0.000001s : 8: predicate.stack_unstack_eliminate 0.42% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.86% : 0.000001s : 10: predicate.switch_defer_inline 1.60% : 0.000003s : 18: predicate.switch_layer_defer_inline 3.50% : 0.000005s : 35: predicate.switch_simplify 0.73% : 0.000001s : 9: predicate.tile_eliminate 0.98% : 0.000002s : 9: predicate.transpose_eliminate 1.79% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.76% : 0.000003s : 17: predicate.tuple_list_get_item_const_eliminator 1.56% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.45% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.94% : 0.000003s : 17: predicate.tuple_list_get_set_item_eliminator 2.52% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.79% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.95% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.78% : 0.000004s : 34: predicate.updatestate_useless_node_eliminater 0.73% : 0.000001s : 4: predicate.value_based_eliminate 0.84% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.84% : 0.000001s : 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.] 0.002507 4 0.77% : 0.000019s : 1: func_graph_cloner_run.FuncGraphClonerGraph 99.23% : 0.002487s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.440202 192 0.00% : 0.000004s : 1: ForceFp32Comm 32.46% : 0.142883s : 1: add_attr 32.44% : 0.142807s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.02% : 0.000089s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.02% : 0.000099s : 1: auto_monad 0.06% : 0.000263s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 1.54% : 0.006773s : 1: bootstrap 0.01% : 0.000032s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000021s : 1: control_data_broadcast_order 0.00% : 0.000010s : 1: convert_after_rewriter 0.03% : 0.000146s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000009s : 1: detach_backward 0.00% : 0.000009s : 1: environ_conv 0.01% : 0.000057s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000010s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.10% : 0.000460s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.19% : 0.000844s : 1: mutable_eliminate 0.00% : 0.000008s : 1: offloading_packed_experts 0.00% : 0.000013s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000015s : 1: opt.transform.mutable_eliminate 0.23% : 0.001014s : 78: opt.transform.opt_a 0.01% : 0.000028s : 1: opt.transform.opt_after_cconv 0.01% : 0.000028s : 1: opt.transform.opt_after_jit_grad 0.02% : 0.000102s : 28: opt.transform.opt_b 0.01% : 0.000053s : 2: opt.transform.opt_trans_graph 0.01% : 0.000040s : 4: opt.transform.symbol_engine_opt 1.43% : 0.006289s : 1: opt_a 0.28% : 0.001251s : 1: opt_after_cconv 0.18% : 0.000781s : 1: opt_after_jit_grad 0.09% : 0.000408s : 1: opt_b 2.69% : 0.011826s : 1: optimize 0.00% : 0.000022s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000010s : 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 0.04% : 0.000190s : 1: overlap_grad_ring_attention 0.02% : 0.000099s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000004s : 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% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000004s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.03% : 0.000112s : 1: pre_auto_parallel 0.07% : 0.000300s : 1: py_interpret_to_execute 0.00% : 0.000016s : 1: py_interpret_to_execute_after_opt_a 0.07% : 0.000313s : 1: remove_cast_before_assign_add 0.01% : 0.000047s : 1: remove_dup_value 0.37% : 0.001636s : 1: renormalize.infer 0.16% : 0.000704s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000007s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000043s : 1: rewriter_after_opt_a 0.01% : 0.000060s : 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% : 0.000008s : 1: swap_dp_allreduce_reducescatter 0.02% : 0.000086s : 1: symbol_engine_optimizer 0.11% : 0.000491s : 1: tuple_transform 27.14% : 0.119490s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[dim_type1-False-batch_size_list1-shape1-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_fp16[dim_type1-False-batch_size_list1-shape1-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:40:10.473.90 [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 = 0.0172764, [21] [bootstrap]: 0.00063551 [type_inference]: 0.00451296 [event_method]: 2.253e-05 [auto_monad]: 6.848e-05 [graph_reusing]: 6.13002e-06 [inline]: 3.6e-06 [add_attr]: 0.00548764, [1] [add_attr_with_inline]: 0.00546912, [1] [Cycle 1]: 7.403e-05, [2] [tag_attr]: 1.992e-05 [meta_addattr_fg_expand]: 3.94002e-06 [parallel-infer-symbol]: 4.43999e-06 [pre_auto_parallel]: 4.868e-05 [insert-virtual-dataset]: 2.96999e-06 [parallel-infer-symbol-second]: 7.90023e-07 [dataset_repeat_opt]: 1.92999e-06 [pipeline_split]: 1.66e-06 [optimize]: 0.00557671, [53] [py_interpret_to_execute]: 2.705e-05 [rewriter_before_opt_a]: 5.411e-05 [opt_a]: 0.00273113, [2] [Cycle 1]: 0.00196746, [45] [expand_dump_flag]: 2.95998e-06 [switch_simplify]: 3.034e-05 [loop_unroll]: 1.202e-05 [a_1]: 0.00031816 [with_stream_mark]: 2.464e-05 [recompute_prepare]: 9.32001e-06 [updatestate_depend_eliminate]: 4.47e-06 [updatestate_assign_eliminate]: 4.17998e-06 [updatestate_loads_eliminate]: 3.65998e-06 [parameter_eliminate]: 1.98002e-06 [a_2]: 8.61e-05 [accelerated_algorithm]: 7.73999e-06 [shard]: 2.36e-06 [meta_shard_fg_expand]: 1.91e-06 [shard_inline]: 6.49999e-06 [merge_send_recv]: 9.85002e-06 [auto_parallel]: 7.68999e-06 [parallel]: 3.276e-05 [flash_sp]: 1.152e-05 [merge_comm]: 4.87998e-06 [allreduce_fusion]: 3.62002e-06 [matmul_add_comm_reduction]: 1.169e-05 [allreduce_slice_to_reducescatter]: 1.04e-06 [virtual_shard_identity]: 9.29e-06 [virtual_dataset]: 7.14001e-06 [get_grad_eliminate_]: 6.66e-06 [virtual_output]: 6.88e-06 [merge_forward]: 5.51998e-06 [cell_reuse_recompute_pass]: 1.45001e-06 [offload_activation]: 1.198e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.582e-05 [merge_recompute_call_nodes]: 2.03002e-06 [before_grad]: 1.232e-05 [set_forward_comm_id_for_comm_node_pass]: 3.81999e-06 [meta_fg_expand]: 3.13e-06 [flash_sp_send_recv_attached]: 3.3e-06 [receive_attached]: 2.46e-06 [after_resolve]: 1.405e-05 [a_after_grad]: 1.062e-05 [renormalize]: 0.00083603 [add_forward_monad_depend]: 7.95e-06 [auto_monad_grad]: 3.73999e-06 [auto_monad_eliminator]: 2.215e-05 [cse]: 4.789e-05 [a_3]: 5.871e-05 [Cycle 2]: 0.00074999, [45] [expand_dump_flag]: 2.12999e-06 [switch_simplify]: 1.081e-05 [loop_unroll]: 1.315e-05 [a_1]: 0.00012555 [with_stream_mark]: 2.095e-05 [recompute_prepare]: 7.33999e-06 [updatestate_depend_eliminate]: 4.62e-06 [updatestate_assign_eliminate]: 3.25998e-06 [updatestate_loads_eliminate]: 2.99999e-06 [parameter_eliminate]: 1.54998e-06 [a_2]: 7.244e-05 [accelerated_algorithm]: 6.56e-06 [shard]: 2.13002e-06 [meta_shard_fg_expand]: 2.54999e-06 [shard_inline]: 6.31998e-06 [merge_send_recv]: 8.57998e-06 [auto_parallel]: 8.49998e-06 [parallel]: 7.23e-06 [flash_sp]: 4.37e-06 [merge_comm]: 4.07e-06 [allreduce_fusion]: 3.54002e-06 [matmul_add_comm_reduction]: 1.331e-05 [allreduce_slice_to_reducescatter]: 5.39992e-07 [virtual_shard_identity]: 7.66999e-06 [virtual_dataset]: 6.41998e-06 [get_grad_eliminate_]: 6.81999e-06 [virtual_output]: 6.11e-06 [merge_forward]: 4.84998e-06 [cell_reuse_recompute_pass]: 2.48998e-06 [offload_activation]: 1.181e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.541e-05 [merge_recompute_call_nodes]: 1.10999e-06 [before_grad]: 1.042e-05 [set_forward_comm_id_for_comm_node_pass]: 4.4e-06 [meta_fg_expand]: 2.48e-06 [flash_sp_send_recv_attached]: 1.67999e-06 [receive_attached]: 1.87999e-06 [after_resolve]: 1.212e-05 [a_after_grad]: 9.49e-06 [renormalize]: 1.00001e-07 [add_forward_monad_depend]: 2.15002e-06 [auto_monad_grad]: 1.37999e-06 [auto_monad_eliminator]: 9.04e-06 [cse]: 2.921e-05 [a_3]: 3.832e-05 [py_interpret_to_execute_after_opt_a]: 1.673e-05 [slice_cell_reuse_recomputed_activation]: 2.65997e-06 [rewriter_after_opt_a]: 4.607e-05 [convert_after_rewriter]: 7.78999e-06 [order_py_execute_after_rewriter]: 5.92001e-06 [mutable_eliminate]: 0.00096078 [opt_b]: 0.00025012, [1] [Cycle 1]: 0.00023961, [7] [b_1]: 0.00013089 [b_2]: 8.08001e-06 [updatestate_depend_eliminate]: 1.126e-05 [updatestate_assign_eliminate]: 3.13e-06 [updatestate_loads_eliminate]: 3.06999e-06 [renormalize]: 1.04e-06 [cse]: 4.489e-05 [optimize_parallel_all_gather_comm]: 4.251e-05 [overlap_param_gather]: 3.04999e-06 [cconv]: 4.168e-05 [loop_unroll]: 0.00050992 [opt_after_cconv]: 0.00013008, [1] [Cycle 1]: 0.00012273, [7] [c_1]: 3.061e-05 [parameter_eliminate]: 5.91e-06 [updatestate_depend_eliminate]: 8.05e-06 [updatestate_assign_eliminate]: 2.66e-06 [updatestate_loads_eliminate]: 2.51e-06 [cse]: 2.795e-05 [renormalize]: 2.89991e-07 [remove_dup_value]: 5.352e-05 [tuple_transform]: 9.587e-05, [1] [Cycle 1]: 7.888e-05, [4] [d_1]: 5.061e-05 [none_parameter_eliminate]: 2.46e-06 [renormalize]: 1.80007e-07 [switch_simplify]: 7.43999e-06 [partial_unused_args_eliminate]: 2.31e-06 [add_recomputation]: 5.878e-05 [cse_after_recomputation]: 3.656e-05, [1] [Cycle 1]: 3.145e-05, [1] [cse]: 2.559e-05 [environ_conv]: 7.75e-06 [swap_dp_allreduce_reducescatter]: 5.66003e-06 [bias_add_comm_swap]: 2.96001e-06 [label_micro_interleaved_index]: 5.79999e-06 [label_fine_grained_interleaved_index]: 3.21001e-06 [merge_cast_opt]: 1.77001e-06 [slice_recompute_activation]: 2.135e-05 [micro_interleaved_order_control]: 3.05002e-06 [assign_add_opt]: 1.71e-06 [ForceFp32Comm]: 1.25001e-06 [remove_cast_before_assign_add]: 1.47001e-06 [full_micro_interleaved_order_control]: 2.39001e-06 [reorder_send_recv_between_fp_bp]: 2.79999e-06 [comm_op_add_attrs]: 1.39998e-06 [add_comm_op_reuse_tag]: 1.37999e-06 [interleave_split_concat_branches]: 1.67001e-06 [interleave_parallel_branches]: 1.15001e-06 [overlap_opt_shard_in_pipeline]: 1.45999e-06 [overlap_opt_shard_grad_in_pipeline]: 1.89999e-06 [control_data_broadcast_order]: 1.531e-05 [grouped_pairwise_exchange_alltoall]: 2.19999e-06 [offloading_packed_experts]: 5.14e-06 [overlap_recompute_and_grad_model_parallel]: 5.23002e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.26002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.62999e-06 [overlap_recompute_comm]: 2.61e-06 [overlap_grad_ring_attention]: 4.49998e-06 [overlap_grad_flash_sp]: 2.457e-05 [begin_end_overlap_inline]: 6.39993e-07 [split_matmul_comm_elemetwise]: 2.26e-06 [split_layernorm_comm]: 1.84e-06 [handle_group_info]: 1.12999e-06 [symbol_engine_optimizer]: 8.977e-05, [1] [Cycle 1]: 8.472e-05, [6] [build]: 4.42998e-06 [elim_shapecalc]: 1.587e-05 [elim_not_effective]: 1.503e-05 [opt_reshape]: 7.95998e-06 [fold_const_symbol]: 1.014e-05 [renormalize]: 2.00002e-07 [detach_backward]: 2.56e-06 [pipeline_parallel_scheduler]: 1.66e-06 [auto_monad_reorder]: 2.195e-05 [get_jit_bprop_graph]: 2.24001e-06 [rewriter_after_jit_bprop_graph]: 7.52998e-06 [opt_after_jit_grad]: 0.00058011 [validate]: 5.068e-05 Sums bootstrap : 0.000636s : 5.99% type_inference : 0.004513s : 42.57% event_method : 0.000023s : 0.21% auto_monad : 0.000068s : 0.65% graph_reusing : 0.000006s : 0.06% inline : 0.000004s : 0.03% add_attr.add_attr_with_inline.tag_attr : 0.000020s : 0.19% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.04% parallel-infer-symbol : 0.000004s : 0.04% pre_auto_parallel : 0.000049s : 0.46% 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 : 0.000027s : 0.26% optimize.rewriter_before_opt_a : 0.000054s : 0.51% optimize.opt_a.expand_dump_flag : 0.000005s : 0.05% optimize.opt_a.switch_simplify : 0.000041s : 0.39% optimize.opt_a.loop_unroll : 0.000025s : 0.24% optimize.opt_a.a_1 : 0.000444s : 4.19% optimize.opt_a.with_stream_mark : 0.000046s : 0.43% optimize.opt_a.recompute_prepare : 0.000017s : 0.16% optimize.opt_a.updatestate_depend_eliminate : 0.000009s : 0.09% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.07% optimize.opt_a.updatestate_loads_eliminate : 0.000007s : 0.06% optimize.opt_a.parameter_eliminate : 0.000004s : 0.03% optimize.opt_a.a_2 : 0.000159s : 1.50% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.13% 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 : 0.12% optimize.opt_a.merge_send_recv : 0.000018s : 0.17% optimize.opt_a.auto_parallel : 0.000016s : 0.15% optimize.opt_a.parallel : 0.000040s : 0.38% optimize.opt_a.flash_sp : 0.000016s : 0.15% optimize.opt_a.merge_comm : 0.000009s : 0.08% optimize.opt_a.allreduce_fusion : 0.000007s : 0.07% optimize.opt_a.matmul_add_comm_reduction : 0.000025s : 0.24% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000017s : 0.16% optimize.opt_a.virtual_dataset : 0.000014s : 0.13% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.13% optimize.opt_a.virtual_output : 0.000013s : 0.12% optimize.opt_a.merge_forward : 0.000010s : 0.10% optimize.opt_a.cell_reuse_recompute_pass : 0.000004s : 0.04% optimize.opt_a.offload_activation : 0.000024s : 0.22% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000031s : 0.29% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.03% optimize.opt_a.before_grad : 0.000023s : 0.21% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.08% optimize.opt_a.meta_fg_expand : 0.000006s : 0.05% 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 : 0.25% optimize.opt_a.a_after_grad : 0.000020s : 0.19% optimize.opt_a.renormalize : 0.000836s : 7.89% optimize.opt_a.add_forward_monad_depend : 0.000010s : 0.10% optimize.opt_a.auto_monad_grad : 0.000005s : 0.05% optimize.opt_a.auto_monad_eliminator : 0.000031s : 0.29% optimize.opt_a.cse : 0.000077s : 0.73% optimize.opt_a.a_3 : 0.000097s : 0.92% optimize.py_interpret_to_execute_after_opt_a : 0.000017s : 0.16% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.03% optimize.rewriter_after_opt_a : 0.000046s : 0.43% optimize.convert_after_rewriter : 0.000008s : 0.07% optimize.order_py_execute_after_rewriter : 0.000006s : 0.06% optimize.mutable_eliminate : 0.000961s : 9.06% optimize.opt_b.b_1 : 0.000131s : 1.23% optimize.opt_b.b_2 : 0.000008s : 0.08% 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 : 0.000045s : 0.42% optimize.optimize_parallel_all_gather_comm : 0.000043s : 0.40% optimize.overlap_param_gather : 0.000003s : 0.03% optimize.cconv : 0.000042s : 0.39% optimize.loop_unroll : 0.000510s : 4.81% optimize.opt_after_cconv.c_1 : 0.000031s : 0.29% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.06% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000008s : 0.08% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000003s : 0.03% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000003s : 0.02% optimize.opt_after_cconv.cse : 0.000028s : 0.26% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000054s : 0.50% optimize.tuple_transform.d_1 : 0.000051s : 0.48% 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 : 0.07% optimize.partial_unused_args_eliminate : 0.000002s : 0.02% optimize.add_recomputation : 0.000059s : 0.55% optimize.cse_after_recomputation.cse : 0.000026s : 0.24% optimize.environ_conv : 0.000008s : 0.07% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.05% optimize.bias_add_comm_swap : 0.000003s : 0.03% optimize.label_micro_interleaved_index : 0.000006s : 0.05% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.03% optimize.merge_cast_opt : 0.000002s : 0.02% optimize.slice_recompute_activation : 0.000021s : 0.20% 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 : 0.02% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.03% optimize.comm_op_add_attrs : 0.000001s : 0.01% optimize.add_comm_op_reuse_tag : 0.000001s : 0.01% optimize.interleave_split_concat_branches : 0.000002s : 0.02% 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 : 0.000015s : 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 : 0.000002s : 0.02% optimize.overlap_recompute_comm : 0.000003s : 0.02% optimize.overlap_grad_ring_attention : 0.000004s : 0.04% optimize.overlap_grad_flash_sp : 0.000025s : 0.23% optimize.begin_end_overlap_inline : 0.000001s : 0.01% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.02% optimize.split_layernorm_comm : 0.000002s : 0.02% optimize.handle_group_info : 0.000001s : 0.01% optimize.symbol_engine_optimizer.build : 0.000004s : 0.04% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000016s : 0.15% optimize.symbol_engine_optimizer.elim_not_effective : 0.000015s : 0.14% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.08% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000010s : 0.10% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.02% pipeline_parallel_scheduler : 0.000002s : 0.02% auto_monad_reorder : 0.000022s : 0.21% get_jit_bprop_graph : 0.000002s : 0.02% rewriter_after_jit_bprop_graph : 0.000008s : 0.07% opt_after_jit_grad : 0.000580s : 5.47% validate : 0.000051s : 0.48% Time group info: ------[substitution.] 0.000131 21 2.06% : 0.000003s : 2: substitution.elim_not_effective 0.97% : 0.000001s : 2: substitution.fold_const_symbol 5.36% : 0.000007s : 4: substitution.graph_param_transform 78.43% : 0.000102s : 1: substitution.inline 4.17% : 0.000005s : 4: substitution.j_node_and_user_rematch 4.26% : 0.000006s : 4: substitution.remove_not_recompute_node 4.75% : 0.000006s : 4: substitution.replace_old_param ------[type_inference.] 0.004437 2 87.47% : 0.003881s : 1: type_inference.infer 12.53% : 0.000556s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000101 1 100.00% : 0.000101s : 1: match.inline ------[predicate.] 0.000169 967 0.76% : 0.000001s : 9: predicate.accumulaten_eliminater 1.52% : 0.000003s : 4: predicate.ad_related_special_op_eliminate 0.63% : 0.000001s : 8: predicate.addn_check_dump 0.73% : 0.000001s : 9: predicate.addn_zero_filter 0.68% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 2.08% : 0.000004s : 17: predicate.arithmetic_simplify 0.80% : 0.000001s : 9: predicate.cast_eliminate 0.79% : 0.000001s : 8: predicate.check_bprop_eliminate 0.63% : 0.000001s : 8: predicate.compare_switch_simplify 0.23% : 0.000000s : 4: predicate.const_output_eliminate 0.64% : 0.000001s : 8: predicate.depend_value_elim 0.77% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 1.05% : 0.000002s : 9: predicate.dict_get_item_eliminator 0.71% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.33% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.30% : 0.000001s : 4: predicate.elim_not_effective 0.50% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.02% : 0.000002s : 13: predicate.environ_add_const_eliminate 1.07% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.96% : 0.000002s : 13: predicate.environ_get_depend_swap 1.58% : 0.000003s : 21: predicate.environ_get_eliminate 0.93% : 0.000002s : 13: predicate.environ_get_set_eliminate 1.02% : 0.000002s : 10: predicate.exchange_switch_depend_value 1.96% : 0.000003s : 10: predicate.float_depend_g_call 0.60% : 0.000001s : 8: predicate.float_environ_get_switch 0.99% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.27% : 0.000000s : 4: predicate.fold_const_symbol 1.05% : 0.000002s : 8: predicate.get_grad_eliminate 0.24% : 0.000000s : 4: predicate.graph_param_transform 0.63% : 0.000001s : 8: predicate.incorporate_call 0.54% : 0.000001s : 8: predicate.incorporate_call_switch 5.91% : 0.000010s : 43: predicate.inline 0.91% : 0.000002s : 8: predicate.inline_without_move 0.42% : 0.000001s : 8: predicate.j_node_and_user_rematch 1.39% : 0.000002s : 8: predicate.less_batch_normalization 1.64% : 0.000003s : 17: predicate.list_to_tuple_eliminator_ 2.01% : 0.000003s : 26: predicate.load_eliminater 1.25% : 0.000002s : 4: predicate.loop_unroll_after_grad 4.74% : 0.000008s : 13: predicate.loop_unroll_before_grad 1.64% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.63% : 0.000001s : 8: predicate.merge_addn 0.69% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.70% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.73% : 0.000001s : 9: predicate.minmaximum_grad 2.79% : 0.000005s : 4: predicate.mutable_eliminate 0.49% : 0.000001s : 4: predicate.opt_reshape 0.44% : 0.000001s : 4: predicate.parallel_virtual_node 0.98% : 0.000002s : 10: predicate.partial_defer_inline 1.03% : 0.000002s : 13: predicate.partial_eliminate 0.64% : 0.000001s : 9: predicate.print_const_string_wrapper 0.74% : 0.000001s : 8: predicate.reduce_all_const_elim 0.95% : 0.000002s : 9: predicate.reduce_eliminate 1.87% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.76% : 0.000001s : 8: predicate.remove_not_recompute_node 1.22% : 0.000002s : 17: predicate.replace_applicator 0.72% : 0.000001s : 8: predicate.replace_old_param 0.33% : 0.000001s : 4: predicate.reset_defer_inline 0.68% : 0.000001s : 9: predicate.reshape_eliminate 0.77% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.50% : 0.000001s : 4: predicate.row_tensor_eliminate 0.94% : 0.000002s : 8: predicate.same_eliminate 0.51% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.34% : 0.000002s : 8: predicate.shard_identity_eliminate 1.07% : 0.000002s : 8: predicate.special_op_eliminate 0.78% : 0.000001s : 8: predicate.specialize_transform 1.33% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.90% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.36% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.76% : 0.000001s : 10: predicate.switch_defer_inline 1.55% : 0.000003s : 18: predicate.switch_layer_defer_inline 4.07% : 0.000007s : 35: predicate.switch_simplify 0.72% : 0.000001s : 9: predicate.tile_eliminate 0.68% : 0.000001s : 9: predicate.transpose_eliminate 1.68% : 0.000003s : 17: predicate.tuple_list_convert_item_index_to_positive 1.33% : 0.000002s : 17: predicate.tuple_list_get_item_const_eliminator 1.22% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 3.21% : 0.000005s : 25: predicate.tuple_list_get_item_eliminator 1.74% : 0.000003s : 17: predicate.tuple_list_get_set_item_eliminator 2.22% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.59% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.76% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.71% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.42% : 0.000001s : 4: predicate.value_based_eliminate 0.87% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.86% : 0.000001s : 8: predicate.virtual_output_eliminate 0.27% : 0.000000s : 4: predicate.virtual_view_grad_eliminate 0.55% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000299 4 9.02% : 0.000027s : 1: func_graph_cloner_run.FuncGraphClonerGraph 90.98% : 0.000272s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.030176 192 0.01% : 0.000004s : 1: ForceFp32Comm 18.21% : 0.005496s : 1: add_attr 18.14% : 0.005473s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.21% : 0.000063s : 1: add_recomputation 0.01% : 0.000004s : 1: assign_add_opt 0.25% : 0.000075s : 1: auto_monad 0.10% : 0.000031s : 1: auto_monad_reorder 0.01% : 0.000003s : 1: begin_end_overlap_inline 0.02% : 0.000006s : 1: bias_add_comm_swap 2.24% : 0.000677s : 1: bootstrap 0.15% : 0.000045s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.06% : 0.000019s : 1: control_data_broadcast_order 0.04% : 0.000012s : 1: convert_after_rewriter 0.13% : 0.000039s : 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 0.10% : 0.000031s : 1: event_method 0.02% : 0.000005s : 1: full_micro_interleaved_order_control 0.02% : 0.000006s : 1: get_jit_bprop_graph 0.03% : 0.000010s : 1: graph_reusing 0.02% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000004s : 1: handle_group_info 0.02% : 0.000007s : 1: inline 0.02% : 0.000006s : 1: insert-virtual-dataset 0.01% : 0.000004s : 1: interleave_parallel_branches 0.02% : 0.000005s : 1: interleave_split_concat_branches 0.02% : 0.000006s : 1: label_fine_grained_interleaved_index 0.03% : 0.000009s : 1: label_micro_interleaved_index 1.73% : 0.000521s : 1: loop_unroll 0.01% : 0.000005s : 1: merge_cast_opt 0.02% : 0.000007s : 1: micro_interleaved_order_control 3.24% : 0.000979s : 1: mutable_eliminate 0.03% : 0.000008s : 1: offloading_packed_experts 0.05% : 0.000016s : 1: opt.transform.loop_unroll_optimizer 0.09% : 0.000028s : 1: opt.transform.mutable_eliminate 2.89% : 0.000871s : 78: opt.transform.opt_a 0.10% : 0.000029s : 1: opt.transform.opt_after_cconv 0.10% : 0.000031s : 1: opt.transform.opt_after_jit_grad 0.36% : 0.000108s : 28: opt.transform.opt_b 0.18% : 0.000056s : 2: opt.transform.opt_trans_graph 0.15% : 0.000045s : 4: opt.transform.symbol_engine_opt 9.06% : 0.002735s : 1: opt_a 0.44% : 0.000134s : 1: opt_after_cconv 1.98% : 0.000598s : 1: opt_after_jit_grad 0.85% : 0.000255s : 1: opt_b 18.50% : 0.005583s : 1: optimize 0.16% : 0.000049s : 1: optimize_parallel_all_gather_comm 0.03% : 0.000010s : 1: order_py_execute_after_rewriter 0.10% : 0.000029s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.04% : 0.000013s : 1: overlap_grad_ring_attention 0.02% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.01% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.02% : 0.000007s : 1: overlap_param_gather 0.01% : 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% : 0.000009s : 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 0.18% : 0.000055s : 1: pre_auto_parallel 0.10% : 0.000032s : 1: py_interpret_to_execute 0.07% : 0.000021s : 1: py_interpret_to_execute_after_opt_a 0.02% : 0.000005s : 1: remove_cast_before_assign_add 0.19% : 0.000058s : 1: remove_dup_value 1.66% : 0.000501s : 1: renormalize.infer 1.07% : 0.000323s : 1: renormalize.specialize 0.02% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.10% : 0.000029s : 1: rewriter_after_jit_bprop_graph 0.17% : 0.000052s : 1: rewriter_after_opt_a 0.20% : 0.000060s : 1: rewriter_before_opt_a 0.02% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.09% : 0.000027s : 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 0.31% : 0.000092s : 1: symbol_engine_optimizer 0.33% : 0.000099s : 1: tuple_transform 15.06% : 0.004544s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[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_fp16[dim_type1-False-batch_size_list2-shape0-ms_dtype0],max_mem:4.0M [WARNING] ME(168578:281472845774640,MainProcess):2026-01-29-17:40:11.707.555 [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 = 0.0168982, [21] [bootstrap]: 0.00061632 [type_inference]: 0.004558 [event_method]: 1.834e-05 [auto_monad]: 6.246e-05 [graph_reusing]: 5.56e-06 [inline]: 2.92002e-06 [add_attr]: 0.00527112, [1] [add_attr_with_inline]: 0.00525436, [1] [Cycle 1]: 7.309e-05, [2] [tag_attr]: 1.895e-05 [meta_addattr_fg_expand]: 3.93001e-06 [parallel-infer-symbol]: 4e-06 [pre_auto_parallel]: 3.935e-05 [insert-virtual-dataset]: 2.98998e-06 [parallel-infer-symbol-second]: 8.89995e-07 [dataset_repeat_opt]: 2.91e-06 [pipeline_split]: 1.92001e-06 [optimize]: 0.00542361, [53] [py_interpret_to_execute]: 2.429e-05 [rewriter_before_opt_a]: 5.42e-05 [opt_a]: 0.0026529, [2] [Cycle 1]: 0.00192789, [45] [expand_dump_flag]: 2.74001e-06 [switch_simplify]: 2.787e-05 [loop_unroll]: 1.203e-05 [a_1]: 0.00031915 [with_stream_mark]: 2.373e-05 [recompute_prepare]: 9.57999e-06 [updatestate_depend_eliminate]: 4.99e-06 [updatestate_assign_eliminate]: 3.98999e-06 [updatestate_loads_eliminate]: 3.61001e-06 [parameter_eliminate]: 2.15002e-06 [a_2]: 8.944e-05 [accelerated_algorithm]: 7.45e-06 [shard]: 2.22001e-06 [meta_shard_fg_expand]: 1.76e-06 [shard_inline]: 6.65002e-06 [merge_send_recv]: 9.51e-06 [auto_parallel]: 8.59e-06 [parallel]: 3.731e-05 [flash_sp]: 1.12e-05 [merge_comm]: 4.39002e-06 [allreduce_fusion]: 4.38001e-06 [matmul_add_comm_reduction]: 1.136e-05 [allreduce_slice_to_reducescatter]: 6.59988e-07 [virtual_shard_identity]: 9.54e-06 [virtual_dataset]: 6.54999e-06 [get_grad_eliminate_]: 6.02001e-06 [virtual_output]: 7.63999e-06 [merge_forward]: 5.32999e-06 [cell_reuse_recompute_pass]: 1.31002e-06 [offload_activation]: 1.089e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.424e-05 [merge_recompute_call_nodes]: 2.37001e-06 [before_grad]: 1.073e-05 [set_forward_comm_id_for_comm_node_pass]: 3.66999e-06 [meta_fg_expand]: 3.45e-06 [flash_sp_send_recv_attached]: 3.08e-06 [receive_attached]: 2.59001e-06 [after_resolve]: 1.199e-05 [a_after_grad]: 1.052e-05 [renormalize]: 0.00078917 [add_forward_monad_depend]: 8.03999e-06 [auto_monad_grad]: 3.04999e-06 [auto_monad_eliminator]: 1.889e-05 [cse]: 4.519e-05 [a_3]: 5.308e-05 [Cycle 2]: 0.00071265, [45] [expand_dump_flag]: 2.12001e-06 [switch_simplify]: 8.59e-06 [loop_unroll]: 6.64999e-06 [a_1]: 0.00012318 [with_stream_mark]: 1.625e-05 [recompute_prepare]: 8.05e-06 [updatestate_depend_eliminate]: 4.06001e-06 [updatestate_assign_eliminate]: 3.27002e-06 [updatestate_loads_eliminate]: 2.81999e-06 [parameter_eliminate]: 1.29e-06 [a_2]: 8.204e-05 [accelerated_algorithm]: 6.50002e-06 [shard]: 1.42e-06 [meta_shard_fg_expand]: 2.13002e-06 [shard_inline]: 6.53e-06 [merge_send_recv]: 6.49001e-06 [auto_parallel]: 9.27001e-06 [parallel]: 6.48003e-06 [flash_sp]: 4.38001e-06 [merge_comm]: 3.61001e-06 [allreduce_fusion]: 3.3e-06 [matmul_add_comm_reduction]: 8.37e-06 [allreduce_slice_to_reducescatter]: 4.59986e-07 [virtual_shard_identity]: 8.55001e-06 [virtual_dataset]: 6.09999e-06 [get_grad_eliminate_]: 5.86e-06 [virtual_output]: 5.84e-06 [merge_forward]: 3.70998e-06 [cell_reuse_recompute_pass]: 1.96e-06 [offload_activation]: 1.008e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.445e-05 [merge_recompute_call_nodes]: 9.50007e-07 [before_grad]: 1.026e-05 [set_forward_comm_id_for_comm_node_pass]: 4.3e-06 [meta_fg_expand]: 2.73e-06 [flash_sp_send_recv_attached]: 1.24e-06 [receive_attached]: 1.87001e-06 [after_resolve]: 1.221e-05 [a_after_grad]: 9.27001e-06 [renormalize]: 5.9983e-08 [add_forward_monad_depend]: 2.19999e-06 [auto_monad_grad]: 2.29999e-06 [auto_monad_eliminator]: 9.54e-06 [cse]: 2.31e-05 [a_3]: 3.846e-05 [py_interpret_to_execute_after_opt_a]: 1.792e-05 [slice_cell_reuse_recomputed_activation]: 2.49999e-06 [rewriter_after_opt_a]: 4.494e-05 [convert_after_rewriter]: 8.07e-06 [order_py_execute_after_rewriter]: 5.81e-06 [mutable_eliminate]: 0.00075563 [opt_b]: 0.00033356, [1] [Cycle 1]: 0.00032387, [7] [b_1]: 0.00020054 [b_2]: 8.99e-06 [updatestate_depend_eliminate]: 9.34e-06 [updatestate_assign_eliminate]: 3.57997e-06 [updatestate_loads_eliminate]: 3.39001e-06 [renormalize]: 7.79983e-07 [cse]: 5.666e-05 [optimize_parallel_all_gather_comm]: 2.112e-05 [overlap_param_gather]: 2.66e-06 [cconv]: 3.74e-05 [loop_unroll]: 0.00052243 [opt_after_cconv]: 0.0001445, [1] [Cycle 1]: 0.00013684, [7] [c_1]: 3.142e-05 [parameter_eliminate]: 5.66e-06 [updatestate_depend_eliminate]: 8.37998e-06 [updatestate_assign_eliminate]: 3.13998e-06 [updatestate_loads_eliminate]: 2.66999e-06 [cse]: 2.928e-05 [renormalize]: 6.00005e-07 [remove_dup_value]: 5.046e-05 [tuple_transform]: 0.0001035, [1] [Cycle 1]: 7.852e-05, [4] [d_1]: 4.993e-05 [none_parameter_eliminate]: 2.01e-06 [renormalize]: 2.10013e-07 [switch_simplify]: 7.03e-06 [partial_unused_args_eliminate]: 2.22001e-06 [add_recomputation]: 6.02e-05 [cse_after_recomputation]: 4.246e-05, [1] [Cycle 1]: 3.763e-05, [1] [cse]: 3.121e-05 [environ_conv]: 7.64002e-06 [swap_dp_allreduce_reducescatter]: 5.62999e-06 [bias_add_comm_swap]: 2.93e-06 [label_micro_interleaved_index]: 6.07001e-06 [label_fine_grained_interleaved_index]: 2.81e-06 [merge_cast_opt]: 1.44e-06 [slice_recompute_activation]: 2.24001e-06 [micro_interleaved_order_control]: 2.73e-06 [assign_add_opt]: 1.40001e-06 [ForceFp32Comm]: 8.80013e-07 [remove_cast_before_assign_add]: 1.09998e-06 [full_micro_interleaved_order_control]: 2.76999e-06 [reorder_send_recv_between_fp_bp]: 2.91999e-06 [comm_op_add_attrs]: 1.10999e-06 [add_comm_op_reuse_tag]: 1.02e-06 [interleave_split_concat_branches]: 1.29e-06 [interleave_parallel_branches]: 1.09003e-06 [overlap_opt_shard_in_pipeline]: 1.76e-06 [overlap_opt_shard_grad_in_pipeline]: 2.50002e-06 [control_data_broadcast_order]: 1.634e-05 [grouped_pairwise_exchange_alltoall]: 1.64e-06 [offloading_packed_experts]: 5.24e-06 [overlap_recompute_and_grad_model_parallel]: 5.47001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.32999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.42e-06 [overlap_recompute_comm]: 2.81999e-06 [overlap_grad_ring_attention]: 4.52e-06 [overlap_grad_flash_sp]: 2.325e-05 [begin_end_overlap_inline]: 5.10016e-07 [split_matmul_comm_elemetwise]: 2.81e-06 [split_layernorm_comm]: 1.97999e-06 [handle_group_info]: 1.55999e-06 [symbol_engine_optimizer]: 8.676e-05, [1] [Cycle 1]: 8.21e-05, [6] [build]: 4.43001e-06 [elim_shapecalc]: 1.049e-05 [elim_not_effective]: 1.367e-05 [opt_reshape]: 8.22e-06 [fold_const_symbol]: 1.087e-05 [renormalize]: 4.19997e-07 [detach_backward]: 2.85998e-06 [pipeline_parallel_scheduler]: 1.57999e-06 [auto_monad_reorder]: 2.161e-05 [get_jit_bprop_graph]: 2.68003e-06 [rewriter_after_jit_bprop_graph]: 6.14001e-06 [opt_after_jit_grad]: 0.00058776 [validate]: 5.285e-05 Sums bootstrap : 0.000616s : 5.94% type_inference : 0.004558s : 43.96% event_method : 0.000018s : 0.18% auto_monad : 0.000062s : 0.60% graph_reusing : 0.000006s : 0.05% inline : 0.000003s : 0.03% add_attr.add_attr_with_inline.tag_attr : 0.000019s : 0.18% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000004s : 0.04% parallel-infer-symbol : 0.000004s : 0.04% pre_auto_parallel : 0.000039s : 0.38% insert-virtual-dataset : 0.000003s : 0.03% parallel-infer-symbol-second : 0.000001s : 0.01% dataset_repeat_opt : 0.000003s : 0.03% pipeline_split : 0.000002s : 0.02% optimize.py_interpret_to_execute : 0.000024s : 0.23% optimize.rewriter_before_opt_a : 0.000054s : 0.52% optimize.opt_a.expand_dump_flag : 0.000005s : 0.05% optimize.opt_a.switch_simplify : 0.000036s : 0.35% optimize.opt_a.loop_unroll : 0.000019s : 0.18% optimize.opt_a.a_1 : 0.000442s : 4.27% optimize.opt_a.with_stream_mark : 0.000040s : 0.39% optimize.opt_a.recompute_prepare : 0.000018s : 0.17% optimize.opt_a.updatestate_depend_eliminate : 0.000009s : 0.09% optimize.opt_a.updatestate_assign_eliminate : 0.000007s : 0.07% optimize.opt_a.updatestate_loads_eliminate : 0.000006s : 0.06% optimize.opt_a.parameter_eliminate : 0.000003s : 0.03% optimize.opt_a.a_2 : 0.000171s : 1.65% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.13% 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 : 0.15% optimize.opt_a.auto_parallel : 0.000018s : 0.17% optimize.opt_a.parallel : 0.000044s : 0.42% optimize.opt_a.flash_sp : 0.000016s : 0.15% optimize.opt_a.merge_comm : 0.000008s : 0.08% optimize.opt_a.allreduce_fusion : 0.000008s : 0.07% optimize.opt_a.matmul_add_comm_reduction : 0.000020s : 0.19% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000001s : 0.01% optimize.opt_a.virtual_shard_identity : 0.000018s : 0.17% optimize.opt_a.virtual_dataset : 0.000013s : 0.12% optimize.opt_a.get_grad_eliminate_ : 0.000012s : 0.11% 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 : 0.20% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000029s : 0.28% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.03% optimize.opt_a.before_grad : 0.000021s : 0.20% 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 : 0.000004s : 0.04% optimize.opt_a.receive_attached : 0.000004s : 0.04% optimize.opt_a.after_resolve : 0.000024s : 0.23% optimize.opt_a.a_after_grad : 0.000020s : 0.19% optimize.opt_a.renormalize : 0.000789s : 7.61% optimize.opt_a.add_forward_monad_depend : 0.000010s : 0.10% optimize.opt_a.auto_monad_grad : 0.000005s : 0.05% optimize.opt_a.auto_monad_eliminator : 0.000028s : 0.27% optimize.opt_a.cse : 0.000068s : 0.66% optimize.opt_a.a_3 : 0.000092s : 0.88% optimize.py_interpret_to_execute_after_opt_a : 0.000018s : 0.17% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.02% optimize.rewriter_after_opt_a : 0.000045s : 0.43% optimize.convert_after_rewriter : 0.000008s : 0.08% optimize.order_py_execute_after_rewriter : 0.000006s : 0.06% optimize.mutable_eliminate : 0.000756s : 7.29% optimize.opt_b.b_1 : 0.000201s : 1.93% 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 : 0.000004s : 0.03% optimize.opt_b.updatestate_loads_eliminate : 0.000003s : 0.03% optimize.opt_b.renormalize : 0.000001s : 0.01% optimize.opt_b.cse : 0.000057s : 0.55% optimize.optimize_parallel_all_gather_comm : 0.000021s : 0.20% optimize.overlap_param_gather : 0.000003s : 0.03% optimize.cconv : 0.000037s : 0.36% optimize.loop_unroll : 0.000522s : 5.04% optimize.opt_after_cconv.c_1 : 0.000031s : 0.30% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.05% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000008s : 0.08% 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 : 0.000029s : 0.28% optimize.opt_after_cconv.renormalize : 0.000001s : 0.01% optimize.remove_dup_value : 0.000050s : 0.49% optimize.tuple_transform.d_1 : 0.000050s : 0.48% 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 : 0.07% optimize.partial_unused_args_eliminate : 0.000002s : 0.02% optimize.add_recomputation : 0.000060s : 0.58% optimize.cse_after_recomputation.cse : 0.000031s : 0.30% optimize.environ_conv : 0.000008s : 0.07% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.05% optimize.bias_add_comm_swap : 0.000003s : 0.03% optimize.label_micro_interleaved_index : 0.000006s : 0.06% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.03% optimize.merge_cast_opt : 0.000001s : 0.01% optimize.slice_recompute_activation : 0.000002s : 0.02% optimize.micro_interleaved_order_control : 0.000003s : 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 : 0.000003s : 0.03% 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 : 0.000003s : 0.02% optimize.control_data_broadcast_order : 0.000016s : 0.16% 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 : 0.000003s : 0.03% optimize.overlap_grad_ring_attention : 0.000005s : 0.04% optimize.overlap_grad_flash_sp : 0.000023s : 0.22% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.03% optimize.split_layernorm_comm : 0.000002s : 0.02% optimize.handle_group_info : 0.000002s : 0.02% optimize.symbol_engine_optimizer.build : 0.000004s : 0.04% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000010s : 0.10% optimize.symbol_engine_optimizer.elim_not_effective : 0.000014s : 0.13% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.08% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000011s : 0.10% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.03% pipeline_parallel_scheduler : 0.000002s : 0.02% auto_monad_reorder : 0.000022s : 0.21% get_jit_bprop_graph : 0.000003s : 0.03% rewriter_after_jit_bprop_graph : 0.000006s : 0.06% opt_after_jit_grad : 0.000588s : 5.67% validate : 0.000053s : 0.51% Time group info: ------[substitution.] 0.000126 21 1.80% : 0.000002s : 2: substitution.elim_not_effective 1.04% : 0.000001s : 2: substitution.fold_const_symbol 5.75% : 0.000007s : 4: substitution.graph_param_transform 78.68% : 0.000099s : 1: substitution.inline 3.64% : 0.000005s : 4: substitution.j_node_and_user_rematch 4.71% : 0.000006s : 4: substitution.remove_not_recompute_node 4.37% : 0.000006s : 4: substitution.replace_old_param ------[type_inference.] 0.004494 2 89.58% : 0.004026s : 1: type_inference.infer 10.42% : 0.000468s : 1: type_inference.specialize ------[replace.] 0.000019 1 100.00% : 0.000019s : 1: replace.inline ------[match.] 0.000098 1 100.00% : 0.000098s : 1: match.inline ------[predicate.] 0.000233 967 0.56% : 0.000001s : 9: predicate.accumulaten_eliminater 1.03% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.42% : 0.000001s : 8: predicate.addn_check_dump 0.63% : 0.000001s : 9: predicate.addn_zero_filter 0.45% : 0.000001s : 9: predicate.adjust_all_reduce_mul_add 1.49% : 0.000003s : 17: predicate.arithmetic_simplify 0.51% : 0.000001s : 9: predicate.cast_eliminate 0.61% : 0.000001s : 8: predicate.check_bprop_eliminate 0.57% : 0.000001s : 8: predicate.compare_switch_simplify 0.18% : 0.000000s : 4: predicate.const_output_eliminate 0.46% : 0.000001s : 8: predicate.depend_value_elim 0.49% : 0.000001s : 9: predicate.dict_get_item_const_eliminator 0.59% : 0.000001s : 9: predicate.dict_get_item_eliminator 0.51% : 0.000001s : 9: predicate.dict_set_item_eliminator 1.00% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.18% : 0.000000s : 4: predicate.elim_not_effective 0.53% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 0.74% : 0.000002s : 13: predicate.environ_add_const_eliminate 0.74% : 0.000002s : 13: predicate.environ_get_add_eliminate 0.73% : 0.000002s : 13: predicate.environ_get_depend_swap 1.15% : 0.000003s : 21: predicate.environ_get_eliminate 0.67% : 0.000002s : 13: predicate.environ_get_set_eliminate 0.53% : 0.000001s : 10: predicate.exchange_switch_depend_value 1.07% : 0.000002s : 10: predicate.float_depend_g_call 0.44% : 0.000001s : 8: predicate.float_environ_get_switch 0.66% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.18% : 0.000000s : 4: predicate.fold_const_symbol 0.54% : 0.000001s : 8: predicate.get_grad_eliminate 0.24% : 0.000001s : 4: predicate.graph_param_transform 0.46% : 0.000001s : 8: predicate.incorporate_call 0.40% : 0.000001s : 8: predicate.incorporate_call_switch 4.61% : 0.000011s : 43: predicate.inline 0.61% : 0.000001s : 8: predicate.inline_without_move 0.27% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.67% : 0.000002s : 8: predicate.less_batch_normalization 0.97% : 0.000002s : 17: predicate.list_to_tuple_eliminator_ 1.53% : 0.000004s : 26: predicate.load_eliminater 1.08% : 0.000003s : 4: predicate.loop_unroll_after_grad 0.97% : 0.000002s : 13: predicate.loop_unroll_before_grad 1.30% : 0.000003s : 17: predicate.make_slice_get_slice_eliminator 0.46% : 0.000001s : 8: predicate.merge_addn 0.46% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.69% : 0.000002s : 8: predicate.mini_step_allgather_replace 0.52% : 0.000001s : 9: predicate.minmaximum_grad 1.75% : 0.000004s : 4: predicate.mutable_eliminate 0.58% : 0.000001s : 4: predicate.opt_reshape 0.36% : 0.000001s : 4: predicate.parallel_virtual_node 0.86% : 0.000002s : 10: predicate.partial_defer_inline 0.81% : 0.000002s : 13: predicate.partial_eliminate 0.49% : 0.000001s : 9: predicate.print_const_string_wrapper 0.48% : 0.000001s : 8: predicate.reduce_all_const_elim 0.74% : 0.000002s : 9: predicate.reduce_eliminate 1.40% : 0.000003s : 26: predicate.redundant_stop_gradient_eliminater 0.49% : 0.000001s : 8: predicate.remove_not_recompute_node 0.93% : 0.000002s : 17: predicate.replace_applicator 0.60% : 0.000001s : 8: predicate.replace_old_param 0.35% : 0.000001s : 4: predicate.reset_defer_inline 0.47% : 0.000001s : 9: predicate.reshape_eliminate 0.46% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.36% : 0.000001s : 4: predicate.row_tensor_eliminate 0.74% : 0.000002s : 8: predicate.same_eliminate 0.38% : 0.000001s : 8: predicate.set_cell_output_no_recompute 1.01% : 0.000002s : 8: predicate.shard_identity_eliminate 0.67% : 0.000002s : 8: predicate.special_op_eliminate 0.61% : 0.000001s : 8: predicate.specialize_transform 1.08% : 0.000003s : 8: predicate.split_environ_get_set_with_tuple_value 0.79% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.29% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.56% : 0.000001s : 10: predicate.switch_defer_inline 1.23% : 0.000003s : 18: predicate.switch_layer_defer_inline 2.73% : 0.000006s : 35: predicate.switch_simplify 0.50% : 0.000001s : 9: predicate.tile_eliminate 0.50% : 0.000001s : 9: predicate.transpose_eliminate 31.01% : 0.000072s : 17: predicate.tuple_list_convert_item_index_to_positive 1.05% : 0.000002s : 17: predicate.tuple_list_get_item_const_eliminator 0.88% : 0.000002s : 17: predicate.tuple_list_get_item_depend_reorder 2.46% : 0.000006s : 25: predicate.tuple_list_get_item_eliminator 0.97% : 0.000002s : 17: predicate.tuple_list_get_set_item_eliminator 1.61% : 0.000004s : 25: predicate.tuple_list_set_item_eliminator 1.23% : 0.000003s : 17: predicate.tuple_to_list_eliminator_ 1.28% : 0.000003s : 26: predicate.updatestate_pure_node_eliminater 2.04% : 0.000005s : 34: predicate.updatestate_useless_node_eliminater 0.30% : 0.000001s : 4: predicate.value_based_eliminate 0.56% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.74% : 0.000002s : 8: predicate.virtual_output_eliminate 0.23% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.48% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.000272 4 8.55% : 0.000023s : 1: func_graph_cloner_run.FuncGraphClonerGraph 91.45% : 0.000249s : 3: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.029433 192 0.01% : 0.000004s : 1: ForceFp32Comm 17.94% : 0.005279s : 1: add_attr 17.87% : 0.005259s : 1: add_attr_with_inline 0.01% : 0.000004s : 1: add_comm_op_reuse_tag 0.22% : 0.000065s : 1: add_recomputation 0.01% : 0.000004s : 1: assign_add_opt 0.23% : 0.000067s : 1: auto_monad 0.10% : 0.000031s : 1: auto_monad_reorder 0.01% : 0.000004s : 1: begin_end_overlap_inline 0.02% : 0.000006s : 1: bias_add_comm_swap 2.24% : 0.000660s : 1: bootstrap 0.14% : 0.000041s : 1: cconv 0.01% : 0.000004s : 1: comm_op_add_attrs 0.07% : 0.000020s : 1: control_data_broadcast_order 0.04% : 0.000012s : 1: convert_after_rewriter 0.15% : 0.000045s : 1: cse_after_recomputation 0.02% : 0.000006s : 1: dataset_repeat_opt 0.02% : 0.000006s : 1: detach_backward 0.04% : 0.000011s : 1: environ_conv 0.09% : 0.000026s : 1: event_method 0.02% : 0.000006s : 1: full_micro_interleaved_order_control 0.02% : 0.000006s : 1: get_jit_bprop_graph 0.03% : 0.000009s : 1: graph_reusing 0.02% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.01% : 0.000004s : 1: handle_group_info 0.02% : 0.000006s : 1: inline 0.02% : 0.000007s : 1: insert-virtual-dataset 0.01% : 0.000004s : 1: interleave_parallel_branches 0.02% : 0.000005s : 1: interleave_split_concat_branches 0.02% : 0.000006s : 1: label_fine_grained_interleaved_index 0.03% : 0.000009s : 1: label_micro_interleaved_index 1.81% : 0.000533s : 1: loop_unroll 0.01% : 0.000004s : 1: merge_cast_opt 0.02% : 0.000006s : 1: micro_interleaved_order_control 2.62% : 0.000772s : 1: mutable_eliminate 0.03% : 0.000008s : 1: offloading_packed_experts 0.06% : 0.000019s : 1: opt.transform.loop_unroll_optimizer 0.08% : 0.000024s : 1: opt.transform.mutable_eliminate 2.94% : 0.000867s : 78: opt.transform.opt_a 0.10% : 0.000030s : 1: opt.transform.opt_after_cconv 0.11% : 0.000031s : 1: opt.transform.opt_after_jit_grad 0.61% : 0.000179s : 28: opt.transform.opt_b 0.19% : 0.000055s : 2: opt.transform.opt_trans_graph 0.13% : 0.000040s : 4: opt.transform.symbol_engine_opt 9.02% : 0.002656s : 1: opt_a 0.51% : 0.000149s : 1: opt_after_cconv 2.06% : 0.000605s : 1: opt_after_jit_grad 1.15% : 0.000338s : 1: opt_b 18.45% : 0.005431s : 1: optimize 0.09% : 0.000025s : 1: optimize_parallel_all_gather_comm 0.03% : 0.000010s : 1: order_py_execute_after_rewriter 0.09% : 0.000027s : 1: overlap_grad_flash_sp 0.01% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.10% : 0.000030s : 1: overlap_grad_ring_attention 0.07% : 0.000020s : 1: overlap_opt_shard_grad_in_pipeline 0.02% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.03% : 0.000008s : 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% : 0.000006s : 1: overlap_recompute_comm 0.03% : 0.000009s : 1: parallel-infer-symbol 0.02% : 0.000004s : 1: parallel-infer-symbol-second 0.02% : 0.000006s : 1: partial_unused_args_eliminate 0.02% : 0.000005s : 1: pipeline_parallel_scheduler 0.02% : 0.000005s : 1: pipeline_split 0.15% : 0.000044s : 1: pre_auto_parallel 0.10% : 0.000029s : 1: py_interpret_to_execute 0.08% : 0.000022s : 1: py_interpret_to_execute_after_opt_a 0.09% : 0.000026s : 1: remove_cast_before_assign_add 0.19% : 0.000056s : 1: remove_dup_value 1.62% : 0.000476s : 1: renormalize.infer 1.03% : 0.000304s : 1: renormalize.specialize 0.02% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.03% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.17% : 0.000050s : 1: rewriter_after_opt_a 0.20% : 0.000058s : 1: rewriter_before_opt_a 0.02% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.02% : 0.000005s : 1: slice_recompute_activation 0.02% : 0.000005s : 1: split_layernorm_comm 0.02% : 0.000006s : 1: split_matmul_comm_elemetwise 0.03% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.30% : 0.000089s : 1: symbol_engine_optimizer 0.36% : 0.000107s : 1: tuple_transform 15.58% : 0.004585s : 1: type_inference . [hook] pytest_runtest_teardown:test_batch_matmul_3dx_ndw_fp16[dim_type1-False-batch_size_list2-shape1-ms_dtype0] tests/st/infer/ops/test_internal_ops/test_batch_matmul.py::test_batch_matmul_3dx_ndw_fp16[dim_type1-False-batch_size_list2-shape1-ms_dtype0],max_mem:6.0M =============================== warnings summary =============================== ../../../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/classifier/transdata/transdata_classifier.py:222 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/classifier/transdata/transdata_classifier.py:222: DeprecationWarning: invalid escape sequence \B """ ../../../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:143 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:143: DeprecationWarning: invalid escape sequence \c """ ../../../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:170 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:170: DeprecationWarning: invalid escape sequence \c """ ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for type is zero. setattr(self, word, getattr(machar, word).flat[0]) ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for type is zero. return self._float_to_str(self.smallest_subnormal) ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for type is zero. setattr(self, word, getattr(machar, word).flat[0]) ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for type is zero. return self._float_to_str(self.smallest_subnormal) ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2.py:57 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2.py:57: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad.py:56 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad.py:56: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2_grad") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad_reduce.py:48 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad_reduce.py:48: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2_grad_reduce") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul.py:51 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul.py:51: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:51 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:51: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul_grad") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:143 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:143: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul_grad_reduce") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad.py:92 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad.py:92: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer_grad_d") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad_reduce.py:49 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad_reduce.py:49: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer_grad_d_reduce") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad.py:91 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad.py:91: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel_grad_d") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad_reduce.py:48 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad_reduce.py:48: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel_grad_d_reduce") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel.py:52 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel.py:52: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_perchannel") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel_grad.py:81 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel_grad.py:81: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_perchannel_grad") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer.py:54 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer.py:54: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_per_layer") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer_grad.py:81 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer_grad.py:81: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_per_layer_grad") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perchannel.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perchannel.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("minmax_update_perchannel") ../../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perlayer.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perlayer.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("minmax_update_perlayer") -- Docs: https://docs.pytest.org/en/stable/warnings.html ================= 24 passed, 25 warnings in 205.63s (0:03:25) ==================