==================================================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/mint, configfile: ../../../../../../sault/virtual_test/virtualenv_001/sault/config/pytest.ini plugins: mock-3.14.0, hydra-core-1.3.2, forked-1.6.0, anyio-4.9.0, xdist-1.32.0 collected 8 items test_repeat_interleave.py . [hook] pytest_runtest_teardown:test_repeat_interleave_forward_backward_tensor[True-0-pynative] tests/st/mint/test_repeat_interleave.py::test_repeat_interleave_forward_backward_tensor[True-0-pynative],max_mem:18.0M TotalTime = 0.63888, [21] [bootstrap]: 0.00063772 [type_inference]: 0.548894 [event_method]: 0.0001224 [auto_monad]: 0.00013219 [graph_reusing]: 7.46999e-06 [inline]: 2.90998e-06 [add_attr]: 0.0805754, [1] [add_attr_with_inline]: 0.0805597, [1] [Cycle 1]: 0.00013176, [2] [tag_attr]: 4.45e-05 [meta_addattr_fg_expand]: 8.85001e-06 [parallel-infer-symbol]: 3.66001e-06 [pre_auto_parallel]: 6.953e-05 [insert-virtual-dataset]: 3.01001e-06 [parallel-infer-symbol-second]: 8.29983e-07 [dataset_repeat_opt]: 2.16998e-06 [pipeline_split]: 1.97001e-06 [optimize]: 0.00752975, [53] [py_interpret_to_execute]: 8.99e-06 [rewriter_before_opt_a]: 0.00042775 [opt_a]: 0.00424409, [2] [Cycle 1]: 0.00354297, [45] [expand_dump_flag]: 4.26001e-06 [switch_simplify]: 7.033e-05 [loop_unroll]: 5.07e-05 [a_1]: 0.00106811 [with_stream_mark]: 2.816e-05 [recompute_prepare]: 1.246e-05 [updatestate_depend_eliminate]: 5.05001e-06 [updatestate_assign_eliminate]: 3.41001e-06 [updatestate_loads_eliminate]: 3.31001e-06 [parameter_eliminate]: 2.88e-06 [a_2]: 9.036e-05 [accelerated_algorithm]: 8.23999e-06 [shard]: 2.41e-06 [meta_shard_fg_expand]: 2.01e-06 [shard_inline]: 6.56999e-06 [merge_send_recv]: 9.29e-06 [auto_parallel]: 9.17001e-06 [parallel]: 4.582e-05 [flash_sp]: 1.223e-05 [merge_comm]: 4.80001e-06 [allreduce_fusion]: 3.63e-06 [matmul_add_comm_reduction]: 1.063e-05 [allreduce_slice_to_reducescatter]: 7.80012e-07 [virtual_shard_identity]: 9.61e-06 [virtual_dataset]: 6.71e-06 [get_grad_eliminate_]: 6.20002e-06 [virtual_output]: 6.49999e-06 [merge_forward]: 4.38001e-06 [cell_reuse_recompute_pass]: 1.47999e-06 [offload_activation]: 1.074e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.656e-05 [merge_recompute_call_nodes]: 1.58002e-06 [before_grad]: 1.173e-05 [set_forward_comm_id_for_comm_node_pass]: 4.60999e-06 [meta_fg_expand]: 2.99999e-06 [flash_sp_send_recv_attached]: 2.96999e-06 [receive_attached]: 1.502e-05 [after_resolve]: 1.273e-05 [a_after_grad]: 1.007e-05 [renormalize]: 0.00153821 [add_forward_monad_depend]: 8.09002e-06 [auto_monad_grad]: 3.57002e-06 [auto_monad_eliminator]: 1.931e-05 [cse]: 5.249e-05 [a_3]: 5.933e-05 [Cycle 2]: 0.00068487, [45] [expand_dump_flag]: 2.27001e-06 [switch_simplify]: 9.17001e-06 [loop_unroll]: 6.74001e-06 [a_1]: 0.00012727 [with_stream_mark]: 1.763e-05 [recompute_prepare]: 6.33e-06 [updatestate_depend_eliminate]: 4.48999e-06 [updatestate_assign_eliminate]: 2.73003e-06 [updatestate_loads_eliminate]: 3.18998e-06 [parameter_eliminate]: 1.61002e-06 [a_2]: 7.468e-05 [accelerated_algorithm]: 6.53998e-06 [shard]: 2.15002e-06 [meta_shard_fg_expand]: 1.93002e-06 [shard_inline]: 5.99e-06 [merge_send_recv]: 8.53001e-06 [auto_parallel]: 8.90999e-06 [parallel]: 7.97003e-06 [flash_sp]: 4.32998e-06 [merge_comm]: 4.08001e-06 [allreduce_fusion]: 3.31999e-06 [matmul_add_comm_reduction]: 9.31e-06 [allreduce_slice_to_reducescatter]: 8.80013e-07 [virtual_shard_identity]: 7.14001e-06 [virtual_dataset]: 5.93002e-06 [get_grad_eliminate_]: 5.60001e-06 [virtual_output]: 5.83002e-06 [merge_forward]: 4.20999e-06 [cell_reuse_recompute_pass]: 3.62002e-06 [offload_activation]: 9.36e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.711e-05 [merge_recompute_call_nodes]: 1.92999e-06 [before_grad]: 9.55001e-06 [set_forward_comm_id_for_comm_node_pass]: 4.72998e-06 [meta_fg_expand]: 2.68e-06 [flash_sp_send_recv_attached]: 1.12999e-06 [receive_attached]: 2.19001e-06 [after_resolve]: 1.244e-05 [a_after_grad]: 9.45001e-06 [renormalize]: 8.00064e-08 [add_forward_monad_depend]: 1.27e-06 [auto_monad_grad]: 1.83002e-06 [auto_monad_eliminator]: 8.82999e-06 [cse]: 1.955e-05 [a_3]: 3.753e-05 [py_interpret_to_execute_after_opt_a]: 8.07998e-06 [slice_cell_reuse_recomputed_activation]: 2.29999e-06 [rewriter_after_opt_a]: 3.338e-05 [convert_after_rewriter]: 1.96e-06 [order_py_execute_after_rewriter]: 1.20001e-06 [mutable_eliminate]: 0.0007612 [opt_b]: 0.00026337, [1] [Cycle 1]: 0.00025445, [7] [b_1]: 0.00015532 [b_2]: 8.53001e-06 [updatestate_depend_eliminate]: 1.046e-05 [updatestate_assign_eliminate]: 3.56999e-06 [updatestate_loads_eliminate]: 2.94001e-06 [renormalize]: 5.90022e-07 [cse]: 3.365e-05 [optimize_parallel_all_gather_comm]: 2.12e-05 [overlap_param_gather]: 5.46e-06 [cconv]: 3.669e-05 [loop_unroll]: 0.00049198 [opt_after_cconv]: 0.0003234, [1] [Cycle 1]: 0.00031595, [7] [c_1]: 0.00022501 [parameter_eliminate]: 6.03002e-06 [updatestate_depend_eliminate]: 7.81001e-06 [updatestate_assign_eliminate]: 2.68e-06 [updatestate_loads_eliminate]: 2.51e-06 [cse]: 2.68e-05 [renormalize]: 8.70001e-07 [remove_dup_value]: 1.923e-05 [tuple_transform]: 8.4e-05, [1] [Cycle 1]: 7.792e-05, [4] [d_1]: 4.997e-05 [none_parameter_eliminate]: 1.60001e-06 [renormalize]: 2.20025e-07 [switch_simplify]: 7.33e-06 [partial_unused_args_eliminate]: 2.37999e-06 [add_recomputation]: 6.376e-05 [cse_after_recomputation]: 2.509e-05, [1] [Cycle 1]: 2.032e-05, [1] [cse]: 1.466e-05 [environ_conv]: 2.007e-05 [swap_dp_allreduce_reducescatter]: 6.36e-06 [bias_add_comm_swap]: 3.25e-06 [label_micro_interleaved_index]: 5.11997e-06 [label_fine_grained_interleaved_index]: 2.98e-06 [merge_cast_opt]: 1.34e-06 [slice_recompute_activation]: 2.20002e-06 [micro_interleaved_order_control]: 2.49999e-06 [assign_add_opt]: 1.50999e-06 [ForceFp32Comm]: 1.14e-06 [remove_cast_before_assign_add]: 1.27e-06 [full_micro_interleaved_order_control]: 2.79001e-06 [reorder_send_recv_between_fp_bp]: 2.79999e-06 [comm_op_add_attrs]: 1.10999e-06 [add_comm_op_reuse_tag]: 1.29e-06 [interleave_split_concat_branches]: 1.44998e-06 [interleave_parallel_branches]: 1.05001e-06 [overlap_opt_shard_in_pipeline]: 2.653e-05 [overlap_opt_shard_grad_in_pipeline]: 2.32999e-06 [control_data_broadcast_order]: 1.568e-05 [grouped_pairwise_exchange_alltoall]: 1.71e-06 [offloading_packed_experts]: 4.05998e-06 [overlap_recompute_and_grad_model_parallel]: 4.80999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.22e-06 [overlap_recompute_allgather_and_fa_grad]: 1.55001e-06 [overlap_recompute_comm]: 2.32999e-06 [overlap_grad_ring_attention]: 4.39002e-06 [overlap_grad_flash_sp]: 4.793e-05 [begin_end_overlap_inline]: 5.20027e-07 [split_matmul_comm_elemetwise]: 2.32999e-06 [split_layernorm_comm]: 1.74e-06 [handle_group_info]: 1.44e-06 [symbol_engine_optimizer]: 0.00026733, [1] [Cycle 1]: 0.00026233, [6] [build]: 0.0001567 [elim_shapecalc]: 1.718e-05 [elim_not_effective]: 2.5e-05 [opt_reshape]: 7.92998e-06 [fold_const_symbol]: 1.634e-05 [renormalize]: 2.00002e-07 [detach_backward]: 2.84001e-06 [pipeline_parallel_scheduler]: 1.77001e-06 [auto_monad_reorder]: 2.74e-05 [get_jit_bprop_graph]: 1.96e-06 [rewriter_after_jit_bprop_graph]: 5.51998e-06 [opt_after_jit_grad]: 0.00060627 [validate]: 6.539e-05 Sums bootstrap : 0.000638s : 0.11% type_inference : 0.548894s : 98.50% event_method : 0.000122s : 0.02% auto_monad : 0.000132s : 0.02% graph_reusing : 0.000007s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000045s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000009s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000070s : 0.01% 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 : 0.000009s : 0.00% optimize.rewriter_before_opt_a : 0.000428s : 0.08% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000080s : 0.01% optimize.opt_a.loop_unroll : 0.000057s : 0.01% optimize.opt_a.a_1 : 0.001195s : 0.21% optimize.opt_a.with_stream_mark : 0.000046s : 0.01% optimize.opt_a.recompute_prepare : 0.000019s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000010s : 0.00% 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 : 0.000165s : 0.03% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.00% 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 : 0.00% optimize.opt_a.merge_send_recv : 0.000018s : 0.00% optimize.opt_a.auto_parallel : 0.000018s : 0.00% optimize.opt_a.parallel : 0.000054s : 0.01% optimize.opt_a.flash_sp : 0.000017s : 0.00% optimize.opt_a.merge_comm : 0.000009s : 0.00% optimize.opt_a.allreduce_fusion : 0.000007s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000020s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000017s : 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 : 0.000012s : 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 : 0.000020s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000034s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000021s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000009s : 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 : 0.000017s : 0.00% optimize.opt_a.after_resolve : 0.000025s : 0.00% optimize.opt_a.a_after_grad : 0.000020s : 0.00% optimize.opt_a.renormalize : 0.001538s : 0.28% optimize.opt_a.add_forward_monad_depend : 0.000009s : 0.00% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000028s : 0.01% optimize.opt_a.cse : 0.000072s : 0.01% optimize.opt_a.a_3 : 0.000097s : 0.02% optimize.py_interpret_to_execute_after_opt_a : 0.000008s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000033s : 0.01% optimize.convert_after_rewriter : 0.000002s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000761s : 0.14% optimize.opt_b.b_1 : 0.000155s : 0.03% optimize.opt_b.b_2 : 0.000009s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000010s : 0.00% 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 : 0.000034s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000021s : 0.00% optimize.overlap_param_gather : 0.000005s : 0.00% optimize.cconv : 0.000037s : 0.01% optimize.loop_unroll : 0.000492s : 0.09% optimize.opt_after_cconv.c_1 : 0.000225s : 0.04% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000008s : 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 : 0.000027s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000019s : 0.00% optimize.tuple_transform.d_1 : 0.000050s : 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 : 0.000064s : 0.01% optimize.cse_after_recomputation.cse : 0.000015s : 0.00% optimize.environ_conv : 0.000020s : 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 : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 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 : 0.000027s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000016s : 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 : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000004s : 0.00% optimize.overlap_grad_flash_sp : 0.000048s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000157s : 0.03% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000017s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000025s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000016s : 0.00% 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 : 0.000027s : 0.00% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.000606s : 0.11% validate : 0.000065s : 0.01% Time group info: ------[substitution.] 0.000310 35 3.33% : 0.000010s : 2: substitution.elim_not_effective 2.27% : 0.000007s : 2: substitution.fold_const_symbol 2.14% : 0.000007s : 4: substitution.graph_param_transform 75.53% : 0.000234s : 7: substitution.inline 1.60% : 0.000005s : 4: substitution.j_node_and_user_rematch 1.65% : 0.000005s : 4: substitution.remove_not_recompute_node 2.18% : 0.000007s : 4: substitution.replace_old_param 11.30% : 0.000035s : 8: substitution.tuple_list_get_item_eliminator ------[type_inference.] 0.548777 2 99.58% : 0.546473s : 1: type_inference.infer 0.42% : 0.002304s : 1: type_inference.specialize ------[replace.] 0.000122 15 57.63% : 0.000070s : 7: replace.inline 42.37% : 0.000052s : 8: replace.tuple_list_get_item_eliminator ------[match.] 0.000261 15 87.82% : 0.000229s : 7: match.inline 12.18% : 0.000032s : 8: match.tuple_list_get_item_eliminator ------[predicate.] 0.000423 1733 0.68% : 0.000003s : 20: predicate.accumulaten_eliminater 0.46% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.26% : 0.000001s : 8: predicate.addn_check_dump 0.65% : 0.000003s : 20: predicate.addn_zero_filter 0.60% : 0.000003s : 20: predicate.adjust_all_reduce_mul_add 1.34% : 0.000006s : 28: predicate.arithmetic_simplify 0.72% : 0.000003s : 20: predicate.cast_eliminate 0.28% : 0.000001s : 8: predicate.check_bprop_eliminate 0.26% : 0.000001s : 8: predicate.compare_switch_simplify 0.09% : 0.000000s : 4: predicate.const_output_eliminate 0.36% : 0.000002s : 8: predicate.depend_value_elim 0.68% : 0.000003s : 20: predicate.dict_get_item_const_eliminator 0.73% : 0.000003s : 20: predicate.dict_get_item_eliminator 0.67% : 0.000003s : 20: predicate.dict_set_item_eliminator 0.52% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.10% : 0.000000s : 4: predicate.elim_not_effective 0.26% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 0.78% : 0.000003s : 24: predicate.environ_add_const_eliminate 0.71% : 0.000003s : 24: predicate.environ_get_add_eliminate 0.76% : 0.000003s : 24: predicate.environ_get_depend_swap 1.02% : 0.000004s : 32: predicate.environ_get_eliminate 0.71% : 0.000003s : 24: predicate.environ_get_set_eliminate 1.19% : 0.000005s : 35: predicate.exchange_switch_depend_value 1.75% : 0.000007s : 35: predicate.float_depend_g_call 0.24% : 0.000001s : 8: predicate.float_environ_get_switch 0.33% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.09% : 0.000000s : 4: predicate.fold_const_symbol 0.30% : 0.000001s : 8: predicate.get_grad_eliminate 0.13% : 0.000001s : 4: predicate.graph_param_transform 0.26% : 0.000001s : 8: predicate.incorporate_call 0.21% : 0.000001s : 8: predicate.incorporate_call_switch 3.82% : 0.000016s : 79: predicate.inline 0.31% : 0.000001s : 8: predicate.inline_without_move 0.15% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.47% : 0.000002s : 8: predicate.less_batch_normalization 1.23% : 0.000005s : 36: predicate.list_to_tuple_eliminator_ 1.94% : 0.000008s : 56: predicate.load_eliminater 0.53% : 0.000002s : 4: predicate.loop_unroll_after_grad 1.85% : 0.000008s : 58: predicate.loop_unroll_before_grad 1.12% : 0.000005s : 28: predicate.make_slice_get_slice_eliminator 0.26% : 0.000001s : 8: predicate.merge_addn 0.24% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.27% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.59% : 0.000002s : 20: predicate.minmaximum_grad 0.68% : 0.000003s : 4: predicate.mutable_eliminate 0.17% : 0.000001s : 4: predicate.opt_reshape 0.17% : 0.000001s : 4: predicate.parallel_virtual_node 1.56% : 0.000007s : 35: predicate.partial_defer_inline 1.19% : 0.000005s : 32: predicate.partial_eliminate 0.66% : 0.000003s : 20: predicate.print_const_string_wrapper 0.27% : 0.000001s : 8: predicate.reduce_all_const_elim 1.01% : 0.000004s : 20: predicate.reduce_eliminate 1.74% : 0.000007s : 56: predicate.redundant_stop_gradient_eliminater 0.30% : 0.000001s : 8: predicate.remove_not_recompute_node 0.94% : 0.000004s : 36: predicate.replace_applicator 0.21% : 0.000001s : 8: predicate.replace_old_param 0.12% : 0.000001s : 4: predicate.reset_defer_inline 0.64% : 0.000003s : 20: predicate.reshape_eliminate 0.28% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.16% : 0.000001s : 4: predicate.row_tensor_eliminate 0.40% : 0.000002s : 8: predicate.same_eliminate 0.34% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.38% : 0.000002s : 8: predicate.shard_identity_eliminate 0.28% : 0.000001s : 8: predicate.special_op_eliminate 0.35% : 0.000001s : 8: predicate.specialize_transform 0.49% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.45% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.22% : 0.000001s : 4: predicate.switch_call_monad_eliminater 1.23% : 0.000005s : 35: predicate.switch_defer_inline 1.43% : 0.000006s : 43: predicate.switch_layer_defer_inline 3.71% : 0.000016s : 105: predicate.switch_simplify 0.67% : 0.000003s : 20: predicate.tile_eliminate 0.69% : 0.000003s : 20: predicate.transpose_eliminate 0.95% : 0.000004s : 28: predicate.tuple_list_convert_item_index_to_positive 0.96% : 0.000004s : 28: predicate.tuple_list_get_item_const_eliminator 0.86% : 0.000004s : 28: predicate.tuple_list_get_item_depend_reorder 2.15% : 0.000009s : 44: predicate.tuple_list_get_item_eliminator 0.87% : 0.000004s : 28: predicate.tuple_list_get_set_item_eliminator 1.37% : 0.000006s : 36: predicate.tuple_list_set_item_eliminator 1.29% : 0.000005s : 36: predicate.tuple_to_list_eliminator_ 37.69% : 0.000160s : 56: predicate.updatestate_pure_node_eliminater 1.98% : 0.000008s : 64: predicate.updatestate_useless_node_eliminater 0.22% : 0.000001s : 4: predicate.value_based_eliminate 0.32% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.33% : 0.000001s : 8: predicate.virtual_output_eliminate 0.14% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.19% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.001459 18 54.51% : 0.000795s : 9: func_graph_cloner_run.FuncGraphClonerGraph 45.49% : 0.000664s : 9: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.730566 192 0.00% : 0.000004s : 1: ForceFp32Comm 11.03% : 0.080582s : 1: add_attr 11.03% : 0.080564s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000068s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.02% : 0.000142s : 1: auto_monad 0.00% : 0.000032s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.09% : 0.000669s : 1: bootstrap 0.01% : 0.000040s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000019s : 1: control_data_broadcast_order 0.00% : 0.000005s : 1: convert_after_rewriter 0.00% : 0.000028s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000024s : 1: environ_conv 0.02% : 0.000133s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000012s : 1: graph_reusing 0.00% : 0.000006s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000005s : 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 0.07% : 0.000502s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.11% : 0.000775s : 1: mutable_eliminate 0.00% : 0.000007s : 1: offloading_packed_experts 0.00% : 0.000017s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000021s : 1: opt.transform.mutable_eliminate 0.23% : 0.001694s : 78: opt.transform.opt_a 0.03% : 0.000223s : 1: opt.transform.opt_after_cconv 0.00% : 0.000028s : 1: opt.transform.opt_after_jit_grad 0.02% : 0.000133s : 28: opt.transform.opt_b 0.01% : 0.000055s : 2: opt.transform.opt_trans_graph 0.01% : 0.000061s : 4: opt.transform.symbol_engine_opt 0.58% : 0.004248s : 1: opt_a 0.04% : 0.000327s : 1: opt_after_cconv 0.09% : 0.000621s : 1: opt_after_jit_grad 0.04% : 0.000267s : 1: opt_b 1.03% : 0.007536s : 1: optimize 0.00% : 0.000026s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 1: order_py_execute_after_rewriter 0.01% : 0.000052s : 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% : 0.000030s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000009s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000008s : 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 0.01% : 0.000074s : 1: pre_auto_parallel 0.00% : 0.000013s : 1: py_interpret_to_execute 0.00% : 0.000011s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000023s : 1: remove_dup_value 0.11% : 0.000819s : 1: renormalize.infer 0.10% : 0.000707s : 1: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000037s : 1: rewriter_after_opt_a 0.06% : 0.000438s : 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% : 0.000010s : 1: swap_dp_allreduce_reducescatter 0.04% : 0.000271s : 1: symbol_engine_optimizer 0.01% : 0.000087s : 1: tuple_transform 75.14% : 0.548919s : 1: type_inference TotalTime = 1.09744, [21] [bootstrap]: 0.00054358 [type_inference]: 0.540935 [event_method]: 0.00029123 [auto_monad]: 0.00026872 [graph_reusing]: 1.082e-05 [inline]: 3.13e-06 [add_attr]: 0.00428924, [1] [add_attr_with_inline]: 0.00427658, [1] [Cycle 1]: 0.00012686, [2] [tag_attr]: 6.673e-05 [meta_addattr_fg_expand]: 1.474e-05 [parallel-infer-symbol]: 3.96001e-06 [pre_auto_parallel]: 9.184e-05 [insert-virtual-dataset]: 3.21001e-06 [parallel-infer-symbol-second]: 1.03001e-06 [dataset_repeat_opt]: 2.10002e-06 [pipeline_split]: 1.69e-06 [optimize]: 0.549792, [53] [py_interpret_to_execute]: 5.35001e-06 [rewriter_before_opt_a]: 0.00055251 [opt_a]: 0.409353, [3] [Cycle 1]: 0.225708, [45] [expand_dump_flag]: 6.24001e-06 [switch_simplify]: 0.00020243 [loop_unroll]: 9.941e-05 [a_1]: 0.00218827 [with_stream_mark]: 3.719e-05 [recompute_prepare]: 3e-05 [updatestate_depend_eliminate]: 1.053e-05 [updatestate_assign_eliminate]: 8.28999e-06 [updatestate_loads_eliminate]: 7.87e-06 [parameter_eliminate]: 3.28998e-06 [a_2]: 0.204398 [accelerated_algorithm]: 2.732e-05 [shard]: 5.34998e-06 [meta_shard_fg_expand]: 1.079e-05 [shard_inline]: 1.745e-05 [merge_send_recv]: 4.1e-05 [auto_parallel]: 2.172e-05 [parallel]: 3.275e-05 [flash_sp]: 1.679e-05 [merge_comm]: 1.063e-05 [allreduce_fusion]: 9.53002e-06 [matmul_add_comm_reduction]: 4.071e-05 [allreduce_slice_to_reducescatter]: 9.30013e-07 [virtual_shard_identity]: 1.964e-05 [virtual_dataset]: 1.648e-05 [get_grad_eliminate_]: 1.614e-05 [virtual_output]: 1.666e-05 [merge_forward]: 1.087e-05 [cell_reuse_recompute_pass]: 4.13001e-06 [offload_activation]: 2.003e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.628e-05 [merge_recompute_call_nodes]: 1.55001e-06 [before_grad]: 3.101e-05 [set_forward_comm_id_for_comm_node_pass]: 1.048e-05 [meta_fg_expand]: 0.00221799 [flash_sp_send_recv_attached]: 5.29e-06 [receive_attached]: 2.91999e-06 [after_resolve]: 8.287e-05 [a_after_grad]: 0.0001 [renormalize]: 0.0145711 [add_forward_monad_depend]: 1.515e-05 [auto_monad_grad]: 7.38999e-06 [auto_monad_eliminator]: 7.194e-05 [cse]: 0.00039613 [a_3]: 0.00039242 [Cycle 2]: 0.182551, [45] [expand_dump_flag]: 3.14999e-06 [switch_simplify]: 5.433e-05 [loop_unroll]: 4.852e-05 [a_1]: 0.00154997 [with_stream_mark]: 2.966e-05 [recompute_prepare]: 1.408e-05 [updatestate_depend_eliminate]: 7.20998e-06 [updatestate_assign_eliminate]: 5.69e-06 [updatestate_loads_eliminate]: 5.44e-06 [parameter_eliminate]: 2.34001e-06 [a_2]: 0.00014109 [accelerated_algorithm]: 1.2e-05 [shard]: 2.56e-06 [meta_shard_fg_expand]: 3.31001e-06 [shard_inline]: 1.157e-05 [merge_send_recv]: 1.173e-05 [auto_parallel]: 1.303e-05 [parallel]: 1.02e-05 [flash_sp]: 4.75001e-06 [merge_comm]: 6.45002e-06 [allreduce_fusion]: 6.21e-06 [matmul_add_comm_reduction]: 1.337e-05 [allreduce_slice_to_reducescatter]: 9.50007e-07 [virtual_shard_identity]: 1.419e-05 [virtual_dataset]: 1.045e-05 [get_grad_eliminate_]: 1.078e-05 [virtual_output]: 1.026e-05 [merge_forward]: 6.26e-06 [cell_reuse_recompute_pass]: 2.22001e-06 [offload_activation]: 1.445e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.391e-05 [merge_recompute_call_nodes]: 1.57999e-06 [before_grad]: 1.781e-05 [set_forward_comm_id_for_comm_node_pass]: 6.38e-06 [meta_fg_expand]: 0.00028583 [flash_sp_send_recv_attached]: 2.09e-06 [receive_attached]: 2.68e-06 [after_resolve]: 2.592e-05 [a_after_grad]: 1.882e-05 [renormalize]: 0.179557 [add_forward_monad_depend]: 1.261e-05 [auto_monad_grad]: 3.01999e-06 [auto_monad_eliminator]: 3.428e-05 [cse]: 0.00011871 [a_3]: 9.356e-05 [Cycle 3]: 0.0010684, [45] [expand_dump_flag]: 2.46e-06 [switch_simplify]: 1.441e-05 [loop_unroll]: 1.064e-05 [a_1]: 0.00027449 [with_stream_mark]: 2.258e-05 [recompute_prepare]: 1.133e-05 [updatestate_depend_eliminate]: 6.36e-06 [updatestate_assign_eliminate]: 4.98001e-06 [updatestate_loads_eliminate]: 4.70999e-06 [parameter_eliminate]: 2.04e-06 [a_2]: 0.00013013 [accelerated_algorithm]: 1.096e-05 [shard]: 3.07002e-06 [meta_shard_fg_expand]: 2.90002e-06 [shard_inline]: 1.018e-05 [merge_send_recv]: 1.221e-05 [auto_parallel]: 1.203e-05 [parallel]: 9.55001e-06 [flash_sp]: 1.88002e-06 [merge_comm]: 5.73002e-06 [allreduce_fusion]: 5.17999e-06 [matmul_add_comm_reduction]: 1.278e-05 [allreduce_slice_to_reducescatter]: 1.00999e-06 [virtual_shard_identity]: 1.285e-05 [virtual_dataset]: 1.003e-05 [get_grad_eliminate_]: 1.01e-05 [virtual_output]: 1.008e-05 [merge_forward]: 5.87999e-06 [cell_reuse_recompute_pass]: 3.52002e-06 [offload_activation]: 1.528e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.088e-05 [merge_recompute_call_nodes]: 1.79998e-06 [before_grad]: 1.757e-05 [set_forward_comm_id_for_comm_node_pass]: 6.41e-06 [meta_fg_expand]: 3.48999e-06 [flash_sp_send_recv_attached]: 1.69998e-06 [receive_attached]: 2.27001e-06 [after_resolve]: 1.621e-05 [a_after_grad]: 1.563e-05 [renormalize]: 6.00121e-08 [add_forward_monad_depend]: 2.37999e-06 [auto_monad_grad]: 1.52001e-06 [auto_monad_eliminator]: 1.723e-05 [cse]: 3.653e-05 [a_3]: 6.871e-05 [py_interpret_to_execute_after_opt_a]: 1.115e-05 [slice_cell_reuse_recomputed_activation]: 2.73e-06 [rewriter_after_opt_a]: 3.495e-05 [convert_after_rewriter]: 1.87001e-06 [order_py_execute_after_rewriter]: 1.27999e-06 [mutable_eliminate]: 0.00083699 [opt_b]: 0.00035613, [1] [Cycle 1]: 0.00034659, [7] [b_1]: 0.0002195 [b_2]: 1.214e-05 [updatestate_depend_eliminate]: 1.342e-05 [updatestate_assign_eliminate]: 5.04998e-06 [updatestate_loads_eliminate]: 5.18002e-06 [renormalize]: 7.7e-07 [cse]: 5.153e-05 [optimize_parallel_all_gather_comm]: 2.803e-05 [overlap_param_gather]: 2.29001e-06 [cconv]: 3.985e-05 [loop_unroll]: 0.0006204 [opt_after_cconv]: 0.00016316, [1] [Cycle 1]: 0.00015551, [7] [c_1]: 5.292e-05 [parameter_eliminate]: 7.08e-06 [updatestate_depend_eliminate]: 1.044e-05 [updatestate_assign_eliminate]: 4.11001e-06 [updatestate_loads_eliminate]: 4.21001e-06 [cse]: 3.948e-05 [renormalize]: 8.09989e-07 [remove_dup_value]: 2.618e-05 [tuple_transform]: 0.00012223, [1] [Cycle 1]: 0.00011666, [4] [d_1]: 8.237e-05 [none_parameter_eliminate]: 2.04999e-06 [renormalize]: 2.00002e-07 [switch_simplify]: 1.098e-05 [partial_unused_args_eliminate]: 2.09e-06 [add_recomputation]: 8.635e-05 [cse_after_recomputation]: 4.112e-05, [1] [Cycle 1]: 3.512e-05, [1] [cse]: 2.729e-05 [environ_conv]: 1.334e-05 [swap_dp_allreduce_reducescatter]: 9.02e-06 [bias_add_comm_swap]: 3.25e-06 [label_micro_interleaved_index]: 5.36002e-06 [label_fine_grained_interleaved_index]: 2.79001e-06 [merge_cast_opt]: 1.29e-06 [slice_recompute_activation]: 2.44001e-06 [micro_interleaved_order_control]: 2.44001e-06 [assign_add_opt]: 1.69e-06 [ForceFp32Comm]: 1.04e-06 [remove_cast_before_assign_add]: 1.12e-06 [full_micro_interleaved_order_control]: 2.88e-06 [reorder_send_recv_between_fp_bp]: 2.79999e-06 [comm_op_add_attrs]: 1.42e-06 [add_comm_op_reuse_tag]: 9.80013e-07 [interleave_split_concat_branches]: 1.59998e-06 [interleave_parallel_branches]: 1.05999e-06 [overlap_opt_shard_in_pipeline]: 1.27e-06 [overlap_opt_shard_grad_in_pipeline]: 1.69e-06 [control_data_broadcast_order]: 2.164e-05 [grouped_pairwise_exchange_alltoall]: 1.76e-06 [offloading_packed_experts]: 6.12999e-06 [overlap_recompute_and_grad_model_parallel]: 5.97999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.46002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.42e-06 [overlap_recompute_comm]: 2.93998e-06 [overlap_grad_ring_attention]: 5.44e-06 [overlap_grad_flash_sp]: 3.153e-05 [begin_end_overlap_inline]: 5.19998e-07 [split_matmul_comm_elemetwise]: 2.65997e-06 [split_layernorm_comm]: 1.96e-06 [handle_group_info]: 1.00001e-06 [symbol_engine_optimizer]: 0.137096, [1] [Cycle 1]: 0.137088, [6] [build]: 0.136864 [elim_shapecalc]: 5.304e-05 [elim_not_effective]: 3.9e-05 [opt_reshape]: 2.149e-05 [fold_const_symbol]: 3.13e-05 [renormalize]: 1.04998e-06 [detach_backward]: 4.42e-06 [pipeline_parallel_scheduler]: 2.35002e-06 [auto_monad_reorder]: 4.595e-05 [get_jit_bprop_graph]: 2.97002e-06 [rewriter_after_jit_bprop_graph]: 6.73e-06 [opt_after_jit_grad]: 0.0008516 [validate]: 7.804e-05 Sums bootstrap : 0.000544s : 0.05% type_inference : 0.540935s : 49.56% event_method : 0.000291s : 0.03% auto_monad : 0.000269s : 0.02% graph_reusing : 0.000011s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000067s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000015s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000092s : 0.01% 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 : 0.000005s : 0.00% optimize.rewriter_before_opt_a : 0.000553s : 0.05% optimize.opt_a.expand_dump_flag : 0.000012s : 0.00% optimize.opt_a.switch_simplify : 0.000271s : 0.02% optimize.opt_a.loop_unroll : 0.000159s : 0.01% optimize.opt_a.a_1 : 0.004013s : 0.37% optimize.opt_a.with_stream_mark : 0.000089s : 0.01% optimize.opt_a.recompute_prepare : 0.000055s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000024s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000019s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000018s : 0.00% optimize.opt_a.parameter_eliminate : 0.000008s : 0.00% optimize.opt_a.a_2 : 0.204670s : 18.75% optimize.opt_a.accelerated_algorithm : 0.000050s : 0.00% optimize.opt_a.shard : 0.000011s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000017s : 0.00% optimize.opt_a.shard_inline : 0.000039s : 0.00% optimize.opt_a.merge_send_recv : 0.000065s : 0.01% optimize.opt_a.auto_parallel : 0.000047s : 0.00% optimize.opt_a.parallel : 0.000053s : 0.00% optimize.opt_a.flash_sp : 0.000023s : 0.00% optimize.opt_a.merge_comm : 0.000023s : 0.00% optimize.opt_a.allreduce_fusion : 0.000021s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000067s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000003s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000047s : 0.00% optimize.opt_a.virtual_dataset : 0.000037s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000037s : 0.00% optimize.opt_a.virtual_output : 0.000037s : 0.00% optimize.opt_a.merge_forward : 0.000023s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000010s : 0.00% optimize.opt_a.offload_activation : 0.000050s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000081s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000066s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000023s : 0.00% optimize.opt_a.meta_fg_expand : 0.002507s : 0.23% optimize.opt_a.flash_sp_send_recv_attached : 0.000009s : 0.00% optimize.opt_a.receive_attached : 0.000008s : 0.00% optimize.opt_a.after_resolve : 0.000125s : 0.01% optimize.opt_a.a_after_grad : 0.000134s : 0.01% optimize.opt_a.renormalize : 0.194128s : 17.78% optimize.opt_a.add_forward_monad_depend : 0.000030s : 0.00% optimize.opt_a.auto_monad_grad : 0.000012s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000123s : 0.01% optimize.opt_a.cse : 0.000551s : 0.05% optimize.opt_a.a_3 : 0.000555s : 0.05% optimize.py_interpret_to_execute_after_opt_a : 0.000011s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000035s : 0.00% optimize.convert_after_rewriter : 0.000002s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000837s : 0.08% optimize.opt_b.b_1 : 0.000219s : 0.02% optimize.opt_b.b_2 : 0.000012s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000013s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000005s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000052s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000028s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000040s : 0.00% optimize.loop_unroll : 0.000620s : 0.06% optimize.opt_after_cconv.c_1 : 0.000053s : 0.00% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000010s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000004s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000004s : 0.00% optimize.opt_after_cconv.cse : 0.000039s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000026s : 0.00% optimize.tuple_transform.d_1 : 0.000082s : 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 : 0.000011s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000086s : 0.01% optimize.cse_after_recomputation.cse : 0.000027s : 0.00% optimize.environ_conv : 0.000013s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000009s : 0.00% optimize.bias_add_comm_swap : 0.000003s : 0.00% optimize.label_micro_interleaved_index : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 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 : 0.000022s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000006s : 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 : 0.000005s : 0.00% optimize.overlap_grad_flash_sp : 0.000032s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.136864s : 12.54% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000053s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000039s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000021s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000031s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000001s : 0.00% detach_backward : 0.000004s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000046s : 0.00% get_jit_bprop_graph : 0.000003s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.00% opt_after_jit_grad : 0.000852s : 0.08% validate : 0.000078s : 0.01% Time group info: ------[substitution.] 0.001105 215 0.94% : 0.000010s : 5: substitution.elim_not_effective 0.92% : 0.000010s : 14: substitution.float_depend_g_call 0.78% : 0.000009s : 3: substitution.float_tuple_getitem_switch 1.41% : 0.000016s : 5: substitution.fold_const_symbol 0.86% : 0.000010s : 8: substitution.graph_param_transform 0.49% : 0.000005s : 2: substitution.incorporate_call 0.23% : 0.000003s : 2: substitution.incorporate_call_switch 64.88% : 0.000717s : 25: substitution.inline 2.09% : 0.000023s : 2: substitution.inline_without_move 1.31% : 0.000015s : 20: substitution.j_node_and_user_rematch 1.10% : 0.000012s : 7: substitution.minmaximum_grad 0.44% : 0.000005s : 1: substitution.opt_reshape 1.04% : 0.000011s : 14: substitution.partial_eliminate 1.52% : 0.000017s : 20: substitution.remove_not_recompute_node 2.76% : 0.000030s : 9: substitution.replace_applicator 1.50% : 0.000017s : 16: substitution.replace_old_param 2.02% : 0.000022s : 3: substitution.reshape_eliminate 0.32% : 0.000003s : 1: substitution.set_cell_output_no_recompute 1.29% : 0.000014s : 3: substitution.switch_simplify 2.31% : 0.000026s : 7: substitution.tuple_list_convert_item_index_to_positive 0.97% : 0.000011s : 7: substitution.tuple_list_get_item_const_eliminator 1.41% : 0.000016s : 7: substitution.tuple_list_get_item_depend_reorder 8.00% : 0.000088s : 27: substitution.tuple_list_get_item_eliminator 1.42% : 0.000016s : 7: substitution.tuple_list_get_set_item_eliminator ------[type_inference.] 0.540787 2 99.15% : 0.536178s : 1: type_inference.infer 0.85% : 0.004608s : 1: type_inference.specialize ------[replace.] 0.000384 45 54.38% : 0.000209s : 25: replace.inline 12.56% : 0.000048s : 3: replace.switch_simplify 33.06% : 0.000127s : 17: replace.tuple_list_get_item_eliminator ------[match.] 0.000758 45 92.81% : 0.000703s : 25: match.inline 1.59% : 0.000012s : 3: match.switch_simplify 5.60% : 0.000042s : 17: match.tuple_list_get_item_eliminator ------[predicate.] 0.000903 6131 1.09% : 0.000010s : 73: predicate.accumulaten_eliminater 0.40% : 0.000004s : 8: predicate.ad_related_special_op_eliminate 2.16% : 0.000020s : 31: predicate.addn_check_dump 1.04% : 0.000009s : 73: predicate.addn_zero_filter 1.01% : 0.000009s : 73: predicate.adjust_all_reduce_mul_add 1.95% : 0.000018s : 104: predicate.arithmetic_simplify 1.10% : 0.000010s : 73: predicate.cast_eliminate 0.97% : 0.000009s : 67: predicate.check_bprop_eliminate 0.47% : 0.000004s : 31: predicate.compare_switch_simplify 0.08% : 0.000001s : 8: predicate.const_output_eliminate 0.48% : 0.000004s : 31: predicate.depend_value_elim 1.13% : 0.000010s : 73: predicate.dict_get_item_const_eliminator 1.20% : 0.000011s : 73: predicate.dict_get_item_eliminator 1.08% : 0.000010s : 73: predicate.dict_set_item_eliminator 0.35% : 0.000003s : 16: predicate.dumpgradient_eliminate 0.17% : 0.000002s : 8: predicate.elim_not_effective 0.53% : 0.000005s : 8: predicate.elim_shapecalc_of_broadcastargs 1.17% : 0.000011s : 81: predicate.environ_add_const_eliminate 1.10% : 0.000010s : 81: predicate.environ_get_add_eliminate 1.14% : 0.000010s : 81: predicate.environ_get_depend_swap 1.60% : 0.000014s : 112: predicate.environ_get_eliminate 1.14% : 0.000010s : 81: predicate.environ_get_set_eliminate 1.73% : 0.000016s : 115: predicate.exchange_switch_depend_value 2.40% : 0.000022s : 115: predicate.float_depend_g_call 0.50% : 0.000005s : 31: predicate.float_environ_get_switch 0.69% : 0.000006s : 39: predicate.float_tuple_getitem_switch 0.08% : 0.000001s : 8: predicate.fold_const_symbol 0.52% : 0.000005s : 31: predicate.get_grad_eliminate 0.10% : 0.000001s : 8: predicate.graph_param_transform 0.49% : 0.000004s : 31: predicate.incorporate_call 0.42% : 0.000004s : 31: predicate.incorporate_call_switch 5.74% : 0.000052s : 266: predicate.inline 1.27% : 0.000011s : 60: predicate.inline_without_move 0.25% : 0.000002s : 31: predicate.j_node_and_user_rematch 0.64% : 0.000006s : 31: predicate.less_batch_normalization 1.57% : 0.000014s : 106: predicate.list_to_tuple_eliminator_ 2.49% : 0.000022s : 179: predicate.load_eliminater 0.42% : 0.000004s : 8: predicate.loop_unroll_after_grad 2.62% : 0.000024s : 175: predicate.loop_unroll_before_grad 1.34% : 0.000012s : 89: predicate.make_slice_get_slice_eliminator 0.48% : 0.000004s : 31: predicate.merge_addn 0.98% : 0.000009s : 67: predicate.micro_step_allgather_replace 0.96% : 0.000009s : 67: predicate.mini_step_allgather_replace 1.02% : 0.000009s : 73: predicate.minmaximum_grad 0.50% : 0.000004s : 8: predicate.mutable_eliminate 0.24% : 0.000002s : 8: predicate.opt_reshape 0.17% : 0.000002s : 8: predicate.parallel_virtual_node 2.14% : 0.000019s : 115: predicate.partial_defer_inline 1.62% : 0.000015s : 98: predicate.partial_eliminate 1.04% : 0.000009s : 73: predicate.print_const_string_wrapper 0.47% : 0.000004s : 31: predicate.reduce_all_const_elim 1.33% : 0.000012s : 73: predicate.reduce_eliminate 2.54% : 0.000023s : 179: predicate.redundant_stop_gradient_eliminater 0.31% : 0.000003s : 31: predicate.remove_not_recompute_node 1.77% : 0.000016s : 157: predicate.replace_applicator 0.64% : 0.000006s : 60: predicate.replace_old_param 0.11% : 0.000001s : 8: predicate.reset_defer_inline 1.12% : 0.000010s : 73: predicate.reshape_eliminate 1.02% : 0.000009s : 67: predicate.row_tensor_add_zeros_like 0.15% : 0.000001s : 8: predicate.row_tensor_eliminate 1.23% : 0.000011s : 67: predicate.same_eliminate 0.34% : 0.000003s : 31: predicate.set_cell_output_no_recompute 0.68% : 0.000006s : 31: predicate.shard_identity_eliminate 0.27% : 0.000002s : 16: predicate.special_op_eliminate 0.53% : 0.000005s : 31: predicate.specialize_transform 1.09% : 0.000010s : 67: predicate.split_environ_get_set_with_tuple_value 1.21% : 0.000011s : 60: predicate.stack_unstack_eliminate 0.15% : 0.000001s : 8: predicate.switch_call_monad_eliminater 1.89% : 0.000017s : 115: predicate.switch_defer_inline 2.76% : 0.000025s : 182: predicate.switch_layer_defer_inline 5.54% : 0.000050s : 335: predicate.switch_simplify 1.04% : 0.000009s : 73: predicate.tile_eliminate 1.06% : 0.000010s : 73: predicate.transpose_eliminate 1.35% : 0.000012s : 89: predicate.tuple_list_convert_item_index_to_positive 1.42% : 0.000013s : 89: predicate.tuple_list_get_item_const_eliminator 1.36% : 0.000012s : 89: predicate.tuple_list_get_item_depend_reorder 2.84% : 0.000026s : 137: predicate.tuple_list_get_item_eliminator 1.46% : 0.000013s : 89: predicate.tuple_list_get_set_item_eliminator 2.03% : 0.000018s : 120: predicate.tuple_list_set_item_eliminator 1.56% : 0.000014s : 106: predicate.tuple_to_list_eliminator_ 2.41% : 0.000022s : 179: predicate.updatestate_pure_node_eliminater 3.02% : 0.000027s : 210: predicate.updatestate_useless_node_eliminater 0.14% : 0.000001s : 8: predicate.value_based_eliminate 0.56% : 0.000005s : 31: predicate.virtual_dataset_eliminate 0.55% : 0.000005s : 31: predicate.virtual_output_eliminate 0.14% : 0.000001s : 8: predicate.virtual_view_grad_eliminate 0.15% : 0.000001s : 8: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.004518 54 60.55% : 0.002736s : 25: func_graph_cloner_run.FuncGraphClonerGraph 39.45% : 0.001782s : 29: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 2.056187 233 0.00% : 0.000004s : 1: ForceFp32Comm 0.21% : 0.004295s : 1: add_attr 0.21% : 0.004281s : 1: add_attr_with_inline 0.00% : 0.000005s : 1: add_comm_op_reuse_tag 0.00% : 0.000093s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.01% : 0.000283s : 1: auto_monad 0.00% : 0.000050s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.03% : 0.000576s : 1: bootstrap 0.00% : 0.000044s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000025s : 1: control_data_broadcast_order 0.00% : 0.000005s : 1: convert_after_rewriter 0.00% : 0.000045s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000008s : 1: detach_backward 0.00% : 0.000017s : 1: environ_conv 0.01% : 0.000308s : 1: event_method 0.00% : 0.000006s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000016s : 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% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.03% : 0.000633s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.04% : 0.000856s : 1: mutable_eliminate 0.00% : 0.000009s : 1: offloading_packed_experts 0.00% : 0.000024s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000031s : 1: opt.transform.mutable_eliminate 10.22% : 0.210173s : 117: opt.transform.opt_a 0.00% : 0.000051s : 1: opt.transform.opt_after_cconv 0.00% : 0.000045s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000200s : 28: opt.transform.opt_b 0.00% : 0.000091s : 2: opt.transform.opt_trans_graph 0.01% : 0.000134s : 4: opt.transform.symbol_engine_opt 19.91% : 0.409358s : 1: opt_a 0.01% : 0.000167s : 1: opt_after_cconv 0.04% : 0.000864s : 1: opt_after_jit_grad 0.02% : 0.000360s : 1: opt_b 26.74% : 0.549799s : 1: optimize 0.00% : 0.000032s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 1: order_py_execute_after_rewriter 0.00% : 0.000037s : 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% : 0.000010s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000008s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.00% : 0.000097s : 1: pre_auto_parallel 0.00% : 0.000009s : 1: py_interpret_to_execute 0.00% : 0.000015s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000030s : 1: remove_dup_value 9.29% : 0.191019s : 2: renormalize.infer 0.15% : 0.003078s : 2: 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% : 0.000039s : 1: rewriter_after_opt_a 0.03% : 0.000563s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000005s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000012s : 1: swap_dp_allreduce_reducescatter 6.67% : 0.137101s : 1: symbol_engine_optimizer 0.01% : 0.000125s : 1: tuple_transform 26.31% : 0.540969s : 1: type_inference TotalTime = 0.370741, [21] [bootstrap]: 0.00059099 [type_inference]: 0.239735 [event_method]: 0.00015509 [auto_monad]: 9.652e-05 [graph_reusing]: 7.78001e-06 [inline]: 3.15002e-06 [add_attr]: 0.00573335, [1] [add_attr_with_inline]: 0.00571588, [1] [Cycle 1]: 0.00010498, [2] [tag_attr]: 4.731e-05 [meta_addattr_fg_expand]: 8.33999e-06 [parallel-infer-symbol]: 3.87998e-06 [pre_auto_parallel]: 7.328e-05 [insert-virtual-dataset]: 3.07002e-06 [parallel-infer-symbol-second]: 8.79983e-07 [dataset_repeat_opt]: 2.04999e-06 [pipeline_split]: 1.91e-06 [optimize]: 0.12352, [53] [py_interpret_to_execute]: 7.95e-06 [rewriter_before_opt_a]: 0.00036937 [opt_a]: 0.120632, [2] [Cycle 1]: 0.00437056, [45] [expand_dump_flag]: 4.33999e-06 [switch_simplify]: 7.102e-05 [loop_unroll]: 5.177e-05 [a_1]: 0.00109164 [with_stream_mark]: 2.929e-05 [recompute_prepare]: 1.045e-05 [updatestate_depend_eliminate]: 5.37999e-06 [updatestate_assign_eliminate]: 3.4e-06 [updatestate_loads_eliminate]: 3.5e-06 [parameter_eliminate]: 2.84999e-06 [a_2]: 8.677e-05 [accelerated_algorithm]: 7.71999e-06 [shard]: 2.14e-06 [meta_shard_fg_expand]: 2.12001e-06 [shard_inline]: 7.31999e-06 [merge_send_recv]: 9.54e-06 [auto_parallel]: 9.67001e-06 [parallel]: 5.652e-05 [flash_sp]: 1.531e-05 [merge_comm]: 5.35999e-06 [allreduce_fusion]: 3.71999e-06 [matmul_add_comm_reduction]: 1.247e-05 [allreduce_slice_to_reducescatter]: 9.50007e-07 [virtual_shard_identity]: 1.021e-05 [virtual_dataset]: 7.99002e-06 [get_grad_eliminate_]: 6.91999e-06 [virtual_output]: 7.5e-06 [merge_forward]: 4.59998e-06 [cell_reuse_recompute_pass]: 1.67999e-06 [offload_activation]: 1.187e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.63e-05 [merge_recompute_call_nodes]: 1.78002e-06 [before_grad]: 1.174e-05 [set_forward_comm_id_for_comm_node_pass]: 3.9e-06 [meta_fg_expand]: 3.98001e-06 [flash_sp_send_recv_attached]: 2.98e-06 [receive_attached]: 2.96999e-06 [after_resolve]: 1.363e-05 [a_after_grad]: 1.045e-05 [renormalize]: 0.00227654 [add_forward_monad_depend]: 1.357e-05 [auto_monad_grad]: 3.18e-06 [auto_monad_eliminator]: 3.001e-05 [cse]: 4.847e-05 [a_3]: 7.194e-05 [Cycle 2]: 0.116242, [45] [expand_dump_flag]: 3.06999e-06 [switch_simplify]: 1.115e-05 [loop_unroll]: 7.95e-06 [a_1]: 0.0001448 [with_stream_mark]: 2.59e-05 [recompute_prepare]: 6.81001e-06 [updatestate_depend_eliminate]: 5.13002e-06 [updatestate_assign_eliminate]: 3.76001e-06 [updatestate_loads_eliminate]: 4.15e-06 [parameter_eliminate]: 2.04e-06 [a_2]: 7.856e-05 [accelerated_algorithm]: 7.43999e-06 [shard]: 2.44001e-06 [meta_shard_fg_expand]: 2.73e-06 [shard_inline]: 5.387e-05 [merge_send_recv]: 2.497e-05 [auto_parallel]: 2.16e-05 [parallel]: 1.176e-05 [flash_sp]: 6.36e-06 [merge_comm]: 4.2e-06 [allreduce_fusion]: 3.51999e-06 [matmul_add_comm_reduction]: 1.538e-05 [allreduce_slice_to_reducescatter]: 9.80013e-07 [virtual_shard_identity]: 1.154e-05 [virtual_dataset]: 8.48999e-06 [get_grad_eliminate_]: 8.3e-06 [virtual_output]: 6.86001e-06 [merge_forward]: 5.61998e-06 [cell_reuse_recompute_pass]: 4.45999e-06 [offload_activation]: 1.173e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.048e-05 [merge_recompute_call_nodes]: 1.82999e-06 [before_grad]: 1.093e-05 [set_forward_comm_id_for_comm_node_pass]: 4.57998e-06 [meta_fg_expand]: 5.10001e-06 [flash_sp_send_recv_attached]: 1.97999e-06 [receive_attached]: 2.48e-06 [after_resolve]: 1.461e-05 [a_after_grad]: 1.059e-05 [renormalize]: 8.00064e-08 [add_forward_monad_depend]: 7.98999e-06 [auto_monad_grad]: 3.41999e-06 [auto_monad_eliminator]: 2.417e-05 [cse]: 4.766e-05 [a_3]: 4.645e-05 [py_interpret_to_execute_after_opt_a]: 1.021e-05 [slice_cell_reuse_recomputed_activation]: 2.64999e-06 [rewriter_after_opt_a]: 2.51e-05 [convert_after_rewriter]: 2.306e-05 [order_py_execute_after_rewriter]: 1.24e-06 [mutable_eliminate]: 0.00080729 [opt_b]: 0.00023691, [1] [Cycle 1]: 0.00022626, [7] [b_1]: 0.00013876 [b_2]: 9.31e-06 [updatestate_depend_eliminate]: 8.79e-06 [updatestate_assign_eliminate]: 3.28998e-06 [updatestate_loads_eliminate]: 3.45e-06 [renormalize]: 7.7e-07 [cse]: 2.517e-05 [optimize_parallel_all_gather_comm]: 1.811e-05 [overlap_param_gather]: 1.94e-06 [cconv]: 3.87e-05 [loop_unroll]: 0.00046143 [opt_after_cconv]: 0.00010845, [1] [Cycle 1]: 0.00010152, [7] [c_1]: 3.099e-05 [parameter_eliminate]: 4.04002e-06 [updatestate_depend_eliminate]: 5.96998e-06 [updatestate_assign_eliminate]: 2.64001e-06 [updatestate_loads_eliminate]: 2.60002e-06 [cse]: 2.158e-05 [renormalize]: 6.19999e-07 [remove_dup_value]: 1.652e-05 [tuple_transform]: 7.942e-05, [1] [Cycle 1]: 7.442e-05, [4] [d_1]: 4.674e-05 [none_parameter_eliminate]: 1.58002e-06 [renormalize]: 2.00002e-07 [switch_simplify]: 7.69002e-06 [partial_unused_args_eliminate]: 2.07999e-06 [add_recomputation]: 5.472e-05 [cse_after_recomputation]: 2.452e-05, [1] [Cycle 1]: 2.005e-05, [1] [cse]: 1.439e-05 [environ_conv]: 7.66999e-06 [swap_dp_allreduce_reducescatter]: 5.84999e-06 [bias_add_comm_swap]: 3.55e-06 [label_micro_interleaved_index]: 4.92e-06 [label_fine_grained_interleaved_index]: 3.06999e-06 [merge_cast_opt]: 1.28002e-06 [slice_recompute_activation]: 1.97001e-06 [micro_interleaved_order_control]: 2.83e-06 [assign_add_opt]: 1.25001e-06 [ForceFp32Comm]: 8.50006e-07 [remove_cast_before_assign_add]: 1.04e-06 [full_micro_interleaved_order_control]: 2.11003e-06 [reorder_send_recv_between_fp_bp]: 3.23e-06 [comm_op_add_attrs]: 9.99979e-07 [add_comm_op_reuse_tag]: 1.05999e-06 [interleave_split_concat_branches]: 1.21002e-06 [interleave_parallel_branches]: 1.25001e-06 [overlap_opt_shard_in_pipeline]: 1.55001e-06 [overlap_opt_shard_grad_in_pipeline]: 1.81998e-06 [control_data_broadcast_order]: 1.36e-05 [grouped_pairwise_exchange_alltoall]: 1.48002e-06 [offloading_packed_experts]: 3.98001e-06 [overlap_recompute_and_grad_model_parallel]: 4.90999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.32e-06 [overlap_recompute_allgather_and_fa_grad]: 1.29998e-06 [overlap_recompute_comm]: 2.31998e-06 [overlap_grad_ring_attention]: 3.98999e-06 [overlap_grad_flash_sp]: 2.264e-05 [begin_end_overlap_inline]: 5.79981e-07 [split_matmul_comm_elemetwise]: 2.11e-06 [split_layernorm_comm]: 1.99e-06 [handle_group_info]: 1.15999e-06 [symbol_engine_optimizer]: 0.00024884, [1] [Cycle 1]: 0.00024327, [6] [build]: 0.00015782 [elim_shapecalc]: 1.304e-05 [elim_not_effective]: 1.609e-05 [opt_reshape]: 7.76001e-06 [fold_const_symbol]: 1.55e-05 [renormalize]: 2.10013e-07 [detach_backward]: 2.80997e-06 [pipeline_parallel_scheduler]: 1.67999e-06 [auto_monad_reorder]: 1.984e-05 [get_jit_bprop_graph]: 2.25002e-06 [rewriter_after_jit_bprop_graph]: 5.17e-06 [opt_after_jit_grad]: 0.0005237 [validate]: 4.807e-05 Sums bootstrap : 0.000591s : 0.24% type_inference : 0.239735s : 96.44% event_method : 0.000155s : 0.06% auto_monad : 0.000097s : 0.04% graph_reusing : 0.000008s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000047s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000008s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000073s : 0.03% 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 : 0.000008s : 0.00% optimize.rewriter_before_opt_a : 0.000369s : 0.15% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000082s : 0.03% optimize.opt_a.loop_unroll : 0.000060s : 0.02% optimize.opt_a.a_1 : 0.001236s : 0.50% optimize.opt_a.with_stream_mark : 0.000055s : 0.02% optimize.opt_a.recompute_prepare : 0.000017s : 0.01% 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 : 0.000008s : 0.00% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000165s : 0.07% 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 : 0.000061s : 0.02% optimize.opt_a.merge_send_recv : 0.000035s : 0.01% optimize.opt_a.auto_parallel : 0.000031s : 0.01% optimize.opt_a.parallel : 0.000068s : 0.03% optimize.opt_a.flash_sp : 0.000022s : 0.01% optimize.opt_a.merge_comm : 0.000010s : 0.00% optimize.opt_a.allreduce_fusion : 0.000007s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000028s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000022s : 0.01% optimize.opt_a.virtual_dataset : 0.000016s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000015s : 0.01% optimize.opt_a.virtual_output : 0.000014s : 0.01% optimize.opt_a.merge_forward : 0.000010s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000006s : 0.00% optimize.opt_a.offload_activation : 0.000024s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000047s : 0.02% optimize.opt_a.merge_recompute_call_nodes : 0.000004s : 0.00% optimize.opt_a.before_grad : 0.000023s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.00% optimize.opt_a.meta_fg_expand : 0.000009s : 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 : 0.000028s : 0.01% optimize.opt_a.a_after_grad : 0.000021s : 0.01% optimize.opt_a.renormalize : 0.002277s : 0.92% optimize.opt_a.add_forward_monad_depend : 0.000022s : 0.01% optimize.opt_a.auto_monad_grad : 0.000007s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000054s : 0.02% optimize.opt_a.cse : 0.000096s : 0.04% optimize.opt_a.a_3 : 0.000118s : 0.05% optimize.py_interpret_to_execute_after_opt_a : 0.000010s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000025s : 0.01% optimize.convert_after_rewriter : 0.000023s : 0.01% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000807s : 0.32% optimize.opt_b.b_1 : 0.000139s : 0.06% optimize.opt_b.b_2 : 0.000009s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000009s : 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 : 0.000025s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000018s : 0.01% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000039s : 0.02% optimize.loop_unroll : 0.000461s : 0.19% optimize.opt_after_cconv.c_1 : 0.000031s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000004s : 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 : 0.000022s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000017s : 0.01% optimize.tuple_transform.d_1 : 0.000047s : 0.02% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000008s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000055s : 0.02% optimize.cse_after_recomputation.cse : 0.000014s : 0.01% optimize.environ_conv : 0.000008s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 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 : 0.000001s : 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 : 0.000014s : 0.01% optimize.grouped_pairwise_exchange_alltoall : 0.000001s : 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 : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000158s : 0.06% 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 : 0.000008s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000016s : 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 : 0.000020s : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000005s : 0.00% opt_after_jit_grad : 0.000524s : 0.21% validate : 0.000048s : 0.02% Time group info: ------[substitution.] 0.000314 35 1.34% : 0.000004s : 2: substitution.elim_not_effective 2.01% : 0.000006s : 2: substitution.fold_const_symbol 2.07% : 0.000006s : 4: substitution.graph_param_transform 80.41% : 0.000253s : 7: substitution.inline 1.65% : 0.000005s : 4: substitution.j_node_and_user_rematch 2.35% : 0.000007s : 4: substitution.remove_not_recompute_node 2.86% : 0.000009s : 4: substitution.replace_old_param 7.32% : 0.000023s : 8: substitution.tuple_list_get_item_eliminator ------[type_inference.] 0.239638 2 98.74% : 0.236628s : 1: type_inference.infer 1.26% : 0.003010s : 1: type_inference.specialize ------[replace.] 0.000123 15 58.21% : 0.000071s : 7: replace.inline 41.79% : 0.000051s : 8: replace.tuple_list_get_item_eliminator ------[match.] 0.000268 15 92.62% : 0.000249s : 7: match.inline 7.38% : 0.000020s : 8: match.tuple_list_get_item_eliminator ------[predicate.] 0.000295 1733 1.14% : 0.000003s : 20: predicate.accumulaten_eliminater 0.78% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.35% : 0.000001s : 8: predicate.addn_check_dump 0.93% : 0.000003s : 20: predicate.addn_zero_filter 0.89% : 0.000003s : 20: predicate.adjust_all_reduce_mul_add 2.31% : 0.000007s : 28: predicate.arithmetic_simplify 1.09% : 0.000003s : 20: predicate.cast_eliminate 0.47% : 0.000001s : 8: predicate.check_bprop_eliminate 0.36% : 0.000001s : 8: predicate.compare_switch_simplify 0.12% : 0.000000s : 4: predicate.const_output_eliminate 0.44% : 0.000001s : 8: predicate.depend_value_elim 1.02% : 0.000003s : 20: predicate.dict_get_item_const_eliminator 1.11% : 0.000003s : 20: predicate.dict_get_item_eliminator 1.00% : 0.000003s : 20: predicate.dict_set_item_eliminator 0.59% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.16% : 0.000000s : 4: predicate.elim_not_effective 0.34% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.11% : 0.000003s : 24: predicate.environ_add_const_eliminate 1.34% : 0.000004s : 24: predicate.environ_get_add_eliminate 1.02% : 0.000003s : 24: predicate.environ_get_depend_swap 1.49% : 0.000004s : 32: predicate.environ_get_eliminate 1.05% : 0.000003s : 24: predicate.environ_get_set_eliminate 1.80% : 0.000005s : 35: predicate.exchange_switch_depend_value 2.51% : 0.000007s : 35: predicate.float_depend_g_call 0.48% : 0.000001s : 8: predicate.float_environ_get_switch 0.57% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.14% : 0.000000s : 4: predicate.fold_const_symbol 0.52% : 0.000002s : 8: predicate.get_grad_eliminate 0.20% : 0.000001s : 4: predicate.graph_param_transform 0.38% : 0.000001s : 8: predicate.incorporate_call 0.32% : 0.000001s : 8: predicate.incorporate_call_switch 8.20% : 0.000024s : 79: predicate.inline 0.55% : 0.000002s : 8: predicate.inline_without_move 0.25% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.63% : 0.000002s : 8: predicate.less_batch_normalization 1.80% : 0.000005s : 36: predicate.list_to_tuple_eliminator_ 2.63% : 0.000008s : 56: predicate.load_eliminater 0.61% : 0.000002s : 4: predicate.loop_unroll_after_grad 2.97% : 0.000009s : 58: predicate.loop_unroll_before_grad 1.59% : 0.000005s : 28: predicate.make_slice_get_slice_eliminator 0.42% : 0.000001s : 8: predicate.merge_addn 0.52% : 0.000002s : 8: predicate.micro_step_allgather_replace 0.50% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.96% : 0.000003s : 20: predicate.minmaximum_grad 0.93% : 0.000003s : 4: predicate.mutable_eliminate 0.38% : 0.000001s : 4: predicate.opt_reshape 0.28% : 0.000001s : 4: predicate.parallel_virtual_node 2.10% : 0.000006s : 35: predicate.partial_defer_inline 1.65% : 0.000005s : 32: predicate.partial_eliminate 0.93% : 0.000003s : 20: predicate.print_const_string_wrapper 0.49% : 0.000001s : 8: predicate.reduce_all_const_elim 1.32% : 0.000004s : 20: predicate.reduce_eliminate 2.61% : 0.000008s : 56: predicate.redundant_stop_gradient_eliminater 0.43% : 0.000001s : 8: predicate.remove_not_recompute_node 1.60% : 0.000005s : 36: predicate.replace_applicator 0.31% : 0.000001s : 8: predicate.replace_old_param 0.28% : 0.000001s : 4: predicate.reset_defer_inline 0.95% : 0.000003s : 20: predicate.reshape_eliminate 0.55% : 0.000002s : 8: predicate.row_tensor_add_zeros_like 0.29% : 0.000001s : 4: predicate.row_tensor_eliminate 0.65% : 0.000002s : 8: predicate.same_eliminate 0.31% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.68% : 0.000002s : 8: predicate.shard_identity_eliminate 0.49% : 0.000001s : 8: predicate.special_op_eliminate 0.53% : 0.000002s : 8: predicate.specialize_transform 0.92% : 0.000003s : 8: predicate.split_environ_get_set_with_tuple_value 0.71% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.27% : 0.000001s : 4: predicate.switch_call_monad_eliminater 1.74% : 0.000005s : 35: predicate.switch_defer_inline 2.20% : 0.000006s : 43: predicate.switch_layer_defer_inline 5.49% : 0.000016s : 105: predicate.switch_simplify 1.28% : 0.000004s : 20: predicate.tile_eliminate 1.20% : 0.000004s : 20: predicate.transpose_eliminate 1.57% : 0.000005s : 28: predicate.tuple_list_convert_item_index_to_positive 1.57% : 0.000005s : 28: predicate.tuple_list_get_item_const_eliminator 1.46% : 0.000004s : 28: predicate.tuple_list_get_item_depend_reorder 2.95% : 0.000009s : 44: predicate.tuple_list_get_item_eliminator 1.62% : 0.000005s : 28: predicate.tuple_list_get_set_item_eliminator 2.10% : 0.000006s : 36: predicate.tuple_list_set_item_eliminator 1.99% : 0.000006s : 36: predicate.tuple_to_list_eliminator_ 2.60% : 0.000008s : 56: predicate.updatestate_pure_node_eliminater 2.92% : 0.000009s : 64: predicate.updatestate_useless_node_eliminater 0.34% : 0.000001s : 4: predicate.value_based_eliminate 0.65% : 0.000002s : 8: predicate.virtual_dataset_eliminate 0.54% : 0.000002s : 8: predicate.virtual_output_eliminate 0.20% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.26% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.001921 18 48.80% : 0.000938s : 9: func_graph_cloner_run.FuncGraphClonerGraph 51.20% : 0.000984s : 9: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.504227 192 0.00% : 0.000004s : 1: ForceFp32Comm 1.14% : 0.005740s : 1: add_attr 1.13% : 0.005721s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000059s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.02% : 0.000106s : 1: auto_monad 0.00% : 0.000024s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.13% : 0.000640s : 1: bootstrap 0.01% : 0.000042s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000017s : 1: control_data_broadcast_order 0.01% : 0.000026s : 1: convert_after_rewriter 0.01% : 0.000027s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000011s : 1: environ_conv 0.03% : 0.000168s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000012s : 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% : 0.000007s : 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 0.09% : 0.000471s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.16% : 0.000819s : 1: mutable_eliminate 0.00% : 0.000007s : 1: offloading_packed_experts 0.00% : 0.000015s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000018s : 1: opt.transform.mutable_eliminate 0.36% : 0.001823s : 78: opt.transform.opt_a 0.01% : 0.000030s : 1: opt.transform.opt_after_cconv 0.01% : 0.000026s : 1: opt.transform.opt_after_jit_grad 0.02% : 0.000114s : 28: opt.transform.opt_b 0.01% : 0.000052s : 2: opt.transform.opt_trans_graph 0.01% : 0.000048s : 4: opt.transform.symbol_engine_opt 23.92% : 0.120636s : 1: opt_a 0.02% : 0.000112s : 1: opt_after_cconv 0.11% : 0.000534s : 1: opt_after_jit_grad 0.05% : 0.000240s : 1: opt_b 24.50% : 0.123526s : 1: optimize 0.00% : 0.000022s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 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% : 0.000008s : 1: overlap_grad_ring_attention 0.00% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000004s : 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% : 0.000007s : 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% : 0.000078s : 1: pre_auto_parallel 0.00% : 0.000012s : 1: py_interpret_to_execute 0.00% : 0.000013s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.00% : 0.000020s : 1: remove_dup_value 0.21% : 0.001061s : 1: renormalize.infer 0.24% : 0.001198s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000008s : 1: rewriter_after_jit_bprop_graph 0.01% : 0.000029s : 1: rewriter_after_opt_a 0.08% : 0.000378s : 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 0.05% : 0.000252s : 1: symbol_engine_optimizer 0.02% : 0.000082s : 1: tuple_transform 47.55% : 0.239768s : 1: type_inference TotalTime = 0.716455, [21] [bootstrap]: 0.00053544 [type_inference]: 0.326443 [event_method]: 0.00030446 [auto_monad]: 0.00021945 [graph_reusing]: 1.169e-05 [inline]: 3.13e-06 [add_attr]: 0.15776, [1] [add_attr_with_inline]: 0.157746, [1] [Cycle 1]: 0.00014434, [2] [tag_attr]: 7.87e-05 [meta_addattr_fg_expand]: 1.671e-05 [parallel-infer-symbol]: 4.35e-06 [pre_auto_parallel]: 9.633e-05 [insert-virtual-dataset]: 2.63e-06 [parallel-infer-symbol-second]: 9.89996e-07 [dataset_repeat_opt]: 2.22999e-06 [pipeline_split]: 2.11e-06 [optimize]: 0.229773, [53] [py_interpret_to_execute]: 8.87999e-06 [rewriter_before_opt_a]: 0.00059139 [opt_a]: 0.132621, [3] [Cycle 1]: 0.0206356, [45] [expand_dump_flag]: 6.17999e-06 [switch_simplify]: 0.00020868 [loop_unroll]: 0.00010199 [a_1]: 0.00243559 [with_stream_mark]: 4.283e-05 [recompute_prepare]: 3.456e-05 [updatestate_depend_eliminate]: 1.133e-05 [updatestate_assign_eliminate]: 9.02999e-06 [updatestate_loads_eliminate]: 9.01002e-06 [parameter_eliminate]: 4.42e-06 [a_2]: 0.0002722 [accelerated_algorithm]: 2.027e-05 [shard]: 2.46998e-06 [meta_shard_fg_expand]: 5.87999e-06 [shard_inline]: 1.74e-05 [merge_send_recv]: 2.296e-05 [auto_parallel]: 1.714e-05 [parallel]: 2.166e-05 [flash_sp]: 1.445e-05 [merge_comm]: 1.117e-05 [allreduce_fusion]: 9.54e-06 [matmul_add_comm_reduction]: 3.72e-05 [allreduce_slice_to_reducescatter]: 9.99979e-07 [virtual_shard_identity]: 2.163e-05 [virtual_dataset]: 1.686e-05 [get_grad_eliminate_]: 1.667e-05 [virtual_output]: 1.644e-05 [merge_forward]: 1.154e-05 [cell_reuse_recompute_pass]: 1.78002e-06 [offload_activation]: 2.015e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.538e-05 [merge_recompute_call_nodes]: 1.60999e-06 [before_grad]: 3.292e-05 [set_forward_comm_id_for_comm_node_pass]: 1.112e-05 [meta_fg_expand]: 0.00238574 [flash_sp_send_recv_attached]: 5.74e-06 [receive_attached]: 2.66999e-06 [after_resolve]: 9.082e-05 [a_after_grad]: 0.00010705 [renormalize]: 0.0130804 [add_forward_monad_depend]: 2.387e-05 [auto_monad_grad]: 8.62998e-06 [auto_monad_eliminator]: 8.556e-05 [cse]: 0.00047576 [a_3]: 0.00044787 [Cycle 2]: 0.110857, [45] [expand_dump_flag]: 3.5e-06 [switch_simplify]: 6.022e-05 [loop_unroll]: 5.377e-05 [a_1]: 0.00183117 [with_stream_mark]: 3.815e-05 [recompute_prepare]: 1.898e-05 [updatestate_depend_eliminate]: 8.08999e-06 [updatestate_assign_eliminate]: 5.49e-06 [updatestate_loads_eliminate]: 5.74e-06 [parameter_eliminate]: 3.5e-06 [a_2]: 0.00015823 [accelerated_algorithm]: 1.377e-05 [shard]: 2.95002e-06 [meta_shard_fg_expand]: 4.25e-06 [shard_inline]: 1.168e-05 [merge_send_recv]: 1.307e-05 [auto_parallel]: 1.442e-05 [parallel]: 1.016e-05 [flash_sp]: 4.39998e-06 [merge_comm]: 7.08e-06 [allreduce_fusion]: 5.86e-06 [matmul_add_comm_reduction]: 1.404e-05 [allreduce_slice_to_reducescatter]: 9.30013e-07 [virtual_shard_identity]: 1.506e-05 [virtual_dataset]: 1.119e-05 [get_grad_eliminate_]: 1.128e-05 [virtual_output]: 1.108e-05 [merge_forward]: 7.23999e-06 [cell_reuse_recompute_pass]: 2.17001e-06 [offload_activation]: 1.509e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.755e-05 [merge_recompute_call_nodes]: 1.52999e-06 [before_grad]: 1.986e-05 [set_forward_comm_id_for_comm_node_pass]: 6.96001e-06 [meta_fg_expand]: 0.00033149 [flash_sp_send_recv_attached]: 3.02002e-06 [receive_attached]: 3.23e-06 [after_resolve]: 3.114e-05 [a_after_grad]: 1.972e-05 [renormalize]: 0.107409 [add_forward_monad_depend]: 1.285e-05 [auto_monad_grad]: 3.14999e-06 [auto_monad_eliminator]: 3.667e-05 [cse]: 0.00012647 [a_3]: 9.832e-05 [Cycle 3]: 0.00110449, [45] [expand_dump_flag]: 2.96999e-06 [switch_simplify]: 1.391e-05 [loop_unroll]: 1.085e-05 [a_1]: 0.00028133 [with_stream_mark]: 2.279e-05 [recompute_prepare]: 1.203e-05 [updatestate_depend_eliminate]: 6.31e-06 [updatestate_assign_eliminate]: 5.14e-06 [updatestate_loads_eliminate]: 5.04e-06 [parameter_eliminate]: 2.26e-06 [a_2]: 0.00013358 [accelerated_algorithm]: 1.086e-05 [shard]: 2.51998e-06 [meta_shard_fg_expand]: 3.86001e-06 [shard_inline]: 1.023e-05 [merge_send_recv]: 1.416e-05 [auto_parallel]: 1.238e-05 [parallel]: 1.054e-05 [flash_sp]: 1.94999e-06 [merge_comm]: 7.7e-06 [allreduce_fusion]: 6.10002e-06 [matmul_add_comm_reduction]: 1.33e-05 [allreduce_slice_to_reducescatter]: 7.59988e-07 [virtual_shard_identity]: 1.371e-05 [virtual_dataset]: 1.007e-05 [get_grad_eliminate_]: 1.031e-05 [virtual_output]: 1.001e-05 [merge_forward]: 6.81001e-06 [cell_reuse_recompute_pass]: 4.06001e-06 [offload_activation]: 1.627e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.238e-05 [merge_recompute_call_nodes]: 1.83997e-06 [before_grad]: 1.93e-05 [set_forward_comm_id_for_comm_node_pass]: 7.18e-06 [meta_fg_expand]: 4.20999e-06 [flash_sp_send_recv_attached]: 2.10002e-06 [receive_attached]: 2.77002e-06 [after_resolve]: 1.831e-05 [a_after_grad]: 1.661e-05 [renormalize]: 7.00238e-08 [add_forward_monad_depend]: 2.73e-06 [auto_monad_grad]: 2.59999e-06 [auto_monad_eliminator]: 1.61e-05 [cse]: 4.193e-05 [a_3]: 6.573e-05 [py_interpret_to_execute_after_opt_a]: 1.059e-05 [slice_cell_reuse_recomputed_activation]: 2.59001e-06 [rewriter_after_opt_a]: 3.415e-05 [convert_after_rewriter]: 1.20001e-06 [order_py_execute_after_rewriter]: 1.17999e-06 [mutable_eliminate]: 0.00085642 [opt_b]: 0.00036976, [1] [Cycle 1]: 0.00035999, [7] [b_1]: 0.0002358 [b_2]: 1.186e-05 [updatestate_depend_eliminate]: 1.311e-05 [updatestate_assign_eliminate]: 5.70001e-06 [updatestate_loads_eliminate]: 5.53002e-06 [renormalize]: 6.80011e-07 [cse]: 4.604e-05 [optimize_parallel_all_gather_comm]: 2.697e-05 [overlap_param_gather]: 2.41e-06 [cconv]: 5.846e-05 [loop_unroll]: 0.00058968 [opt_after_cconv]: 0.00019257, [1] [Cycle 1]: 0.0001845, [7] [c_1]: 6.687e-05 [parameter_eliminate]: 7.16001e-06 [updatestate_depend_eliminate]: 1.181e-05 [updatestate_assign_eliminate]: 5.72999e-06 [updatestate_loads_eliminate]: 5.52001e-06 [cse]: 4.622e-05 [renormalize]: 1.00999e-06 [remove_dup_value]: 2.86e-05 [tuple_transform]: 0.00013503, [1] [Cycle 1]: 0.00012943, [4] [d_1]: 9.117e-05 [none_parameter_eliminate]: 2.47001e-06 [renormalize]: 1.8999e-07 [switch_simplify]: 1.311e-05 [partial_unused_args_eliminate]: 2.71e-06 [add_recomputation]: 9.101e-05 [cse_after_recomputation]: 4.44e-05, [1] [Cycle 1]: 3.874e-05, [1] [cse]: 3.18e-05 [environ_conv]: 1.342e-05 [swap_dp_allreduce_reducescatter]: 1.066e-05 [bias_add_comm_swap]: 3.53999e-06 [label_micro_interleaved_index]: 5.61e-06 [label_fine_grained_interleaved_index]: 3.13998e-06 [merge_cast_opt]: 1.43002e-06 [slice_recompute_activation]: 2.21e-06 [micro_interleaved_order_control]: 2.24999e-06 [assign_add_opt]: 1.29998e-06 [ForceFp32Comm]: 8.39995e-07 [remove_cast_before_assign_add]: 1.12e-06 [full_micro_interleaved_order_control]: 2.29001e-06 [reorder_send_recv_between_fp_bp]: 2.73003e-06 [comm_op_add_attrs]: 1.07e-06 [add_comm_op_reuse_tag]: 1.02e-06 [interleave_split_concat_branches]: 1.43002e-06 [interleave_parallel_branches]: 1.08001e-06 [overlap_opt_shard_in_pipeline]: 1.25001e-06 [overlap_opt_shard_grad_in_pipeline]: 1.94e-06 [control_data_broadcast_order]: 2.208e-05 [grouped_pairwise_exchange_alltoall]: 2.17999e-06 [offloading_packed_experts]: 6.66e-06 [overlap_recompute_and_grad_model_parallel]: 7.68999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.54e-06 [overlap_recompute_allgather_and_fa_grad]: 1.32999e-06 [overlap_recompute_comm]: 2.43e-06 [overlap_grad_ring_attention]: 6.68e-06 [overlap_grad_flash_sp]: 3.17e-05 [begin_end_overlap_inline]: 5.00004e-07 [split_matmul_comm_elemetwise]: 2.27999e-06 [split_layernorm_comm]: 1.99e-06 [handle_group_info]: 1.06002e-06 [symbol_engine_optimizer]: 0.0936816, [1] [Cycle 1]: 0.0936725, [6] [build]: 0.0934458 [elim_shapecalc]: 5.508e-05 [elim_not_effective]: 4.15e-05 [opt_reshape]: 2.08e-05 [fold_const_symbol]: 3.151e-05 [renormalize]: 5.09986e-07 [detach_backward]: 4.41002e-06 [pipeline_parallel_scheduler]: 2.45002e-06 [auto_monad_reorder]: 4.693e-05 [get_jit_bprop_graph]: 2.93e-06 [rewriter_after_jit_bprop_graph]: 7.05e-06 [opt_after_jit_grad]: 0.00090152 [validate]: 7.87e-05 Sums bootstrap : 0.000535s : 0.10% type_inference : 0.326443s : 58.61% event_method : 0.000304s : 0.05% auto_monad : 0.000219s : 0.04% graph_reusing : 0.000012s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000079s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000017s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000096s : 0.02% 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 : 0.000009s : 0.00% optimize.rewriter_before_opt_a : 0.000591s : 0.11% optimize.opt_a.expand_dump_flag : 0.000013s : 0.00% optimize.opt_a.switch_simplify : 0.000283s : 0.05% optimize.opt_a.loop_unroll : 0.000167s : 0.03% optimize.opt_a.a_1 : 0.004548s : 0.82% optimize.opt_a.with_stream_mark : 0.000104s : 0.02% optimize.opt_a.recompute_prepare : 0.000066s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000026s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000020s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000020s : 0.00% optimize.opt_a.parameter_eliminate : 0.000010s : 0.00% optimize.opt_a.a_2 : 0.000564s : 0.10% optimize.opt_a.accelerated_algorithm : 0.000045s : 0.01% optimize.opt_a.shard : 0.000008s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000014s : 0.00% optimize.opt_a.shard_inline : 0.000039s : 0.01% optimize.opt_a.merge_send_recv : 0.000050s : 0.01% optimize.opt_a.auto_parallel : 0.000044s : 0.01% optimize.opt_a.parallel : 0.000042s : 0.01% optimize.opt_a.flash_sp : 0.000021s : 0.00% optimize.opt_a.merge_comm : 0.000026s : 0.00% optimize.opt_a.allreduce_fusion : 0.000022s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000065s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000003s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000050s : 0.01% optimize.opt_a.virtual_dataset : 0.000038s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000038s : 0.01% optimize.opt_a.virtual_output : 0.000038s : 0.01% optimize.opt_a.merge_forward : 0.000026s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000008s : 0.00% optimize.opt_a.offload_activation : 0.000052s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000085s : 0.02% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000072s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000025s : 0.00% optimize.opt_a.meta_fg_expand : 0.002721s : 0.49% optimize.opt_a.flash_sp_send_recv_attached : 0.000011s : 0.00% optimize.opt_a.receive_attached : 0.000009s : 0.00% optimize.opt_a.after_resolve : 0.000140s : 0.03% optimize.opt_a.a_after_grad : 0.000143s : 0.03% optimize.opt_a.renormalize : 0.120490s : 21.63% optimize.opt_a.add_forward_monad_depend : 0.000039s : 0.01% optimize.opt_a.auto_monad_grad : 0.000014s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000138s : 0.02% optimize.opt_a.cse : 0.000644s : 0.12% optimize.opt_a.a_3 : 0.000612s : 0.11% optimize.py_interpret_to_execute_after_opt_a : 0.000011s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000034s : 0.01% optimize.convert_after_rewriter : 0.000001s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000856s : 0.15% optimize.opt_b.b_1 : 0.000236s : 0.04% optimize.opt_b.b_2 : 0.000012s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000013s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000046s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000027s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000058s : 0.01% optimize.loop_unroll : 0.000590s : 0.11% optimize.opt_after_cconv.c_1 : 0.000067s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000012s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.cse : 0.000046s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000029s : 0.01% optimize.tuple_transform.d_1 : 0.000091s : 0.02% optimize.tuple_transform.none_parameter_eliminate : 0.000002s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000013s : 0.00% optimize.partial_unused_args_eliminate : 0.000003s : 0.00% optimize.add_recomputation : 0.000091s : 0.02% optimize.cse_after_recomputation.cse : 0.000032s : 0.01% optimize.environ_conv : 0.000013s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000011s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000006s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000002s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000022s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000008s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000002s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000007s : 0.00% optimize.overlap_grad_flash_sp : 0.000032s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.093446s : 16.78% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000055s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000041s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000021s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000032s : 0.01% optimize.symbol_engine_optimizer.renormalize : 0.000001s : 0.00% detach_backward : 0.000004s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000047s : 0.01% get_jit_bprop_graph : 0.000003s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.00% opt_after_jit_grad : 0.000902s : 0.16% validate : 0.000079s : 0.01% Time group info: ------[substitution.] 0.001334 215 0.79% : 0.000011s : 5: substitution.elim_not_effective 0.85% : 0.000011s : 14: substitution.float_depend_g_call 0.42% : 0.000006s : 3: substitution.float_tuple_getitem_switch 1.10% : 0.000015s : 5: substitution.fold_const_symbol 0.87% : 0.000012s : 8: substitution.graph_param_transform 0.25% : 0.000003s : 2: substitution.incorporate_call 0.18% : 0.000002s : 2: substitution.incorporate_call_switch 69.76% : 0.000930s : 25: substitution.inline 2.00% : 0.000027s : 2: substitution.inline_without_move 1.13% : 0.000015s : 20: substitution.j_node_and_user_rematch 0.98% : 0.000013s : 7: substitution.minmaximum_grad 0.37% : 0.000005s : 1: substitution.opt_reshape 0.88% : 0.000012s : 14: substitution.partial_eliminate 1.32% : 0.000018s : 20: substitution.remove_not_recompute_node 2.61% : 0.000035s : 9: substitution.replace_applicator 1.37% : 0.000018s : 16: substitution.replace_old_param 1.64% : 0.000022s : 3: substitution.reshape_eliminate 0.34% : 0.000005s : 1: substitution.set_cell_output_no_recompute 1.07% : 0.000014s : 3: substitution.switch_simplify 2.05% : 0.000027s : 7: substitution.tuple_list_convert_item_index_to_positive 0.89% : 0.000012s : 7: substitution.tuple_list_get_item_const_eliminator 1.36% : 0.000018s : 7: substitution.tuple_list_get_item_depend_reorder 6.57% : 0.000088s : 27: substitution.tuple_list_get_item_eliminator 1.18% : 0.000016s : 7: substitution.tuple_list_get_set_item_eliminator ------[type_inference.] 0.326301 2 98.51% : 0.321440s : 1: type_inference.infer 1.49% : 0.004861s : 1: type_inference.specialize ------[replace.] 0.000464 45 56.41% : 0.000262s : 25: replace.inline 10.88% : 0.000050s : 3: replace.switch_simplify 32.70% : 0.000152s : 17: replace.tuple_list_get_item_eliminator ------[match.] 0.000974 45 93.97% : 0.000916s : 25: match.inline 1.18% : 0.000012s : 3: match.switch_simplify 4.85% : 0.000047s : 17: match.tuple_list_get_item_eliminator ------[predicate.] 0.001009 6131 1.03% : 0.000010s : 73: predicate.accumulaten_eliminater 0.31% : 0.000003s : 8: predicate.ad_related_special_op_eliminate 0.43% : 0.000004s : 31: predicate.addn_check_dump 0.97% : 0.000010s : 73: predicate.addn_zero_filter 0.99% : 0.000010s : 73: predicate.adjust_all_reduce_mul_add 1.94% : 0.000020s : 104: predicate.arithmetic_simplify 1.06% : 0.000011s : 73: predicate.cast_eliminate 0.94% : 0.000009s : 67: predicate.check_bprop_eliminate 0.44% : 0.000004s : 31: predicate.compare_switch_simplify 0.07% : 0.000001s : 8: predicate.const_output_eliminate 0.43% : 0.000004s : 31: predicate.depend_value_elim 1.07% : 0.000011s : 73: predicate.dict_get_item_const_eliminator 1.07% : 0.000011s : 73: predicate.dict_get_item_eliminator 1.03% : 0.000010s : 73: predicate.dict_set_item_eliminator 7.71% : 0.000078s : 16: predicate.dumpgradient_eliminate 0.12% : 0.000001s : 8: predicate.elim_not_effective 0.38% : 0.000004s : 8: predicate.elim_shapecalc_of_broadcastargs 1.10% : 0.000011s : 81: predicate.environ_add_const_eliminate 1.05% : 0.000011s : 81: predicate.environ_get_add_eliminate 1.11% : 0.000011s : 81: predicate.environ_get_depend_swap 1.56% : 0.000016s : 112: predicate.environ_get_eliminate 1.15% : 0.000012s : 81: predicate.environ_get_set_eliminate 1.64% : 0.000017s : 115: predicate.exchange_switch_depend_value 2.51% : 0.000025s : 115: predicate.float_depend_g_call 0.42% : 0.000004s : 31: predicate.float_environ_get_switch 0.59% : 0.000006s : 39: predicate.float_tuple_getitem_switch 0.07% : 0.000001s : 8: predicate.fold_const_symbol 0.48% : 0.000005s : 31: predicate.get_grad_eliminate 0.08% : 0.000001s : 8: predicate.graph_param_transform 0.42% : 0.000004s : 31: predicate.incorporate_call 0.39% : 0.000004s : 31: predicate.incorporate_call_switch 5.25% : 0.000053s : 266: predicate.inline 1.21% : 0.000012s : 60: predicate.inline_without_move 0.23% : 0.000002s : 31: predicate.j_node_and_user_rematch 0.55% : 0.000006s : 31: predicate.less_batch_normalization 1.47% : 0.000015s : 106: predicate.list_to_tuple_eliminator_ 2.36% : 0.000024s : 179: predicate.load_eliminater 0.33% : 0.000003s : 8: predicate.loop_unroll_after_grad 2.43% : 0.000025s : 175: predicate.loop_unroll_before_grad 1.24% : 0.000012s : 89: predicate.make_slice_get_slice_eliminator 0.46% : 0.000005s : 31: predicate.merge_addn 0.94% : 0.000009s : 67: predicate.micro_step_allgather_replace 0.98% : 0.000010s : 67: predicate.mini_step_allgather_replace 0.96% : 0.000010s : 73: predicate.minmaximum_grad 0.37% : 0.000004s : 8: predicate.mutable_eliminate 0.15% : 0.000002s : 8: predicate.opt_reshape 0.17% : 0.000002s : 8: predicate.parallel_virtual_node 2.21% : 0.000022s : 115: predicate.partial_defer_inline 1.51% : 0.000015s : 98: predicate.partial_eliminate 0.99% : 0.000010s : 73: predicate.print_const_string_wrapper 0.42% : 0.000004s : 31: predicate.reduce_all_const_elim 1.25% : 0.000013s : 73: predicate.reduce_eliminate 2.30% : 0.000023s : 179: predicate.redundant_stop_gradient_eliminater 0.27% : 0.000003s : 31: predicate.remove_not_recompute_node 1.65% : 0.000017s : 157: predicate.replace_applicator 0.64% : 0.000006s : 60: predicate.replace_old_param 0.11% : 0.000001s : 8: predicate.reset_defer_inline 1.06% : 0.000011s : 73: predicate.reshape_eliminate 0.97% : 0.000010s : 67: predicate.row_tensor_add_zeros_like 0.13% : 0.000001s : 8: predicate.row_tensor_eliminate 1.17% : 0.000012s : 67: predicate.same_eliminate 0.29% : 0.000003s : 31: predicate.set_cell_output_no_recompute 0.59% : 0.000006s : 31: predicate.shard_identity_eliminate 0.30% : 0.000003s : 16: predicate.special_op_eliminate 0.50% : 0.000005s : 31: predicate.specialize_transform 1.09% : 0.000011s : 67: predicate.split_environ_get_set_with_tuple_value 1.09% : 0.000011s : 60: predicate.stack_unstack_eliminate 0.14% : 0.000001s : 8: predicate.switch_call_monad_eliminater 1.87% : 0.000019s : 115: predicate.switch_defer_inline 2.63% : 0.000027s : 182: predicate.switch_layer_defer_inline 5.29% : 0.000053s : 335: predicate.switch_simplify 0.95% : 0.000010s : 73: predicate.tile_eliminate 1.02% : 0.000010s : 73: predicate.transpose_eliminate 1.27% : 0.000013s : 89: predicate.tuple_list_convert_item_index_to_positive 1.34% : 0.000014s : 89: predicate.tuple_list_get_item_const_eliminator 1.28% : 0.000013s : 89: predicate.tuple_list_get_item_depend_reorder 2.63% : 0.000026s : 137: predicate.tuple_list_get_item_eliminator 1.32% : 0.000013s : 89: predicate.tuple_list_get_set_item_eliminator 1.97% : 0.000020s : 120: predicate.tuple_list_set_item_eliminator 1.48% : 0.000015s : 106: predicate.tuple_to_list_eliminator_ 2.40% : 0.000024s : 179: predicate.updatestate_pure_node_eliminater 2.89% : 0.000029s : 210: predicate.updatestate_useless_node_eliminater 0.14% : 0.000001s : 8: predicate.value_based_eliminate 0.46% : 0.000005s : 31: predicate.virtual_dataset_eliminate 0.46% : 0.000005s : 31: predicate.virtual_output_eliminate 0.12% : 0.000001s : 8: predicate.virtual_view_grad_eliminate 0.16% : 0.000002s : 8: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.005018 54 59.89% : 0.003005s : 25: func_graph_cloner_run.FuncGraphClonerGraph 40.11% : 0.002013s : 29: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.231683 233 0.00% : 0.000004s : 1: ForceFp32Comm 12.81% : 0.157767s : 1: add_attr 12.81% : 0.157752s : 1: add_attr_with_inline 0.00% : 0.000005s : 1: add_comm_op_reuse_tag 0.01% : 0.000099s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.02% : 0.000233s : 1: auto_monad 0.00% : 0.000052s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.05% : 0.000612s : 1: bootstrap 0.01% : 0.000064s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000026s : 1: control_data_broadcast_order 0.00% : 0.000004s : 1: convert_after_rewriter 0.00% : 0.000048s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000008s : 1: detach_backward 0.00% : 0.000017s : 1: environ_conv 0.03% : 0.000319s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000017s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.05% : 0.000604s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.07% : 0.000873s : 1: mutable_eliminate 0.00% : 0.000010s : 1: offloading_packed_experts 0.00% : 0.000028s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000032s : 1: opt.transform.mutable_eliminate 0.55% : 0.006738s : 117: opt.transform.opt_a 0.01% : 0.000066s : 1: opt.transform.opt_after_cconv 0.01% : 0.000122s : 1: opt.transform.opt_after_jit_grad 0.02% : 0.000213s : 28: opt.transform.opt_b 0.01% : 0.000102s : 2: opt.transform.opt_trans_graph 0.01% : 0.000136s : 4: opt.transform.symbol_engine_opt 10.77% : 0.132625s : 1: opt_a 0.02% : 0.000197s : 1: opt_after_cconv 0.07% : 0.000914s : 1: opt_after_jit_grad 0.03% : 0.000374s : 1: opt_b 18.66% : 0.229779s : 1: optimize 0.00% : 0.000031s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 1: order_py_execute_after_rewriter 0.00% : 0.000036s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000010s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 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% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000013s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000005s : 1: overlap_recompute_comm 0.00% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000102s : 1: pre_auto_parallel 0.00% : 0.000012s : 1: py_interpret_to_execute 0.00% : 0.000014s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000033s : 1: remove_dup_value 9.50% : 0.116963s : 2: renormalize.infer 0.28% : 0.003490s : 2: renormalize.specialize 0.00% : 0.000005s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.00% : 0.000038s : 1: rewriter_after_opt_a 0.05% : 0.000601s : 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% : 0.000006s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000014s : 1: swap_dp_allreduce_reducescatter 7.61% : 0.093687s : 1: symbol_engine_optimizer 0.01% : 0.000138s : 1: tuple_transform 26.51% : 0.326473s : 1: type_inference . [hook] pytest_runtest_teardown:test_repeat_interleave_forward_backward_tensor[True-0-KBK] tests/st/mint/test_repeat_interleave.py::test_repeat_interleave_forward_backward_tensor[True-0-KBK],max_mem:20.0M . [hook] pytest_runtest_teardown:test_repeat_interleave_forward_backward_tensor[True-None-pynative] tests/st/mint/test_repeat_interleave.py::test_repeat_interleave_forward_backward_tensor[True-None-pynative],max_mem:20.0M TotalTime = 0.4008, [21] [bootstrap]: 0.00072792 [type_inference]: 0.209695 [event_method]: 7.886e-05 [auto_monad]: 8.526e-05 [graph_reusing]: 7.61999e-06 [inline]: 2.68e-06 [add_attr]: 0.00520584, [1] [add_attr_with_inline]: 0.00518888, [1] [Cycle 1]: 0.00011578, [2] [tag_attr]: 5.982e-05 [meta_addattr_fg_expand]: 9.05999e-06 [parallel-infer-symbol]: 4.60001e-06 [pre_auto_parallel]: 6.757e-05 [insert-virtual-dataset]: 2.59999e-06 [parallel-infer-symbol-second]: 9.09989e-07 [dataset_repeat_opt]: 2.44001e-06 [pipeline_split]: 1.67999e-06 [optimize]: 0.184055, [53] [py_interpret_to_execute]: 8.28999e-06 [rewriter_before_opt_a]: 0.00036307 [opt_a]: 0.181016, [2] [Cycle 1]: 0.180272, [45] [expand_dump_flag]: 3.93999e-06 [switch_simplify]: 6.783e-05 [loop_unroll]: 5.039e-05 [a_1]: 0.00111549 [with_stream_mark]: 2.53e-05 [recompute_prepare]: 1.293e-05 [updatestate_depend_eliminate]: 4.46002e-06 [updatestate_assign_eliminate]: 3.25998e-06 [updatestate_loads_eliminate]: 2.94999e-06 [parameter_eliminate]: 2.09e-06 [a_2]: 8.364e-05 [accelerated_algorithm]: 8.69998e-06 [shard]: 1.86e-06 [meta_shard_fg_expand]: 2.86999e-06 [shard_inline]: 7.11999e-06 [merge_send_recv]: 9.66998e-06 [auto_parallel]: 9.76e-06 [parallel]: 3.315e-05 [flash_sp]: 1.112e-05 [merge_comm]: 4.32e-06 [allreduce_fusion]: 3.87998e-06 [matmul_add_comm_reduction]: 1.055e-05 [allreduce_slice_to_reducescatter]: 1.09e-06 [virtual_shard_identity]: 1.007e-05 [virtual_dataset]: 7.01999e-06 [get_grad_eliminate_]: 6.17001e-06 [virtual_output]: 6.21e-06 [merge_forward]: 4.57e-06 [cell_reuse_recompute_pass]: 1.28002e-06 [offload_activation]: 1.031e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.464e-05 [merge_recompute_call_nodes]: 1.69e-06 [before_grad]: 1.11e-05 [set_forward_comm_id_for_comm_node_pass]: 3.56001e-06 [meta_fg_expand]: 3.13e-06 [flash_sp_send_recv_attached]: 3.16001e-06 [receive_attached]: 2.40002e-06 [after_resolve]: 1.4e-05 [a_after_grad]: 9.48002e-06 [renormalize]: 0.178233 [add_forward_monad_depend]: 9.02e-06 [auto_monad_grad]: 2.92002e-06 [auto_monad_eliminator]: 2.446e-05 [cse]: 4.491e-05 [a_3]: 6.935e-05 [Cycle 2]: 0.0007262, [45] [expand_dump_flag]: 2.52001e-06 [switch_simplify]: 9.47001e-06 [loop_unroll]: 6.70002e-06 [a_1]: 0.00013894 [with_stream_mark]: 1.884e-05 [recompute_prepare]: 7.52998e-06 [updatestate_depend_eliminate]: 4.08999e-06 [updatestate_assign_eliminate]: 3.52002e-06 [updatestate_loads_eliminate]: 3.31001e-06 [parameter_eliminate]: 2.54999e-06 [a_2]: 7.673e-05 [accelerated_algorithm]: 6.68e-06 [shard]: 2.64999e-06 [meta_shard_fg_expand]: 3.08e-06 [shard_inline]: 6.09001e-06 [merge_send_recv]: 1.071e-05 [auto_parallel]: 1.036e-05 [parallel]: 9.52999e-06 [flash_sp]: 3.98001e-06 [merge_comm]: 3.97e-06 [allreduce_fusion]: 3.53e-06 [matmul_add_comm_reduction]: 1.113e-05 [allreduce_slice_to_reducescatter]: 7.29982e-07 [virtual_shard_identity]: 9.09e-06 [virtual_dataset]: 6.09999e-06 [get_grad_eliminate_]: 5.68002e-06 [virtual_output]: 5.82001e-06 [merge_forward]: 4.18001e-06 [cell_reuse_recompute_pass]: 3.90998e-06 [offload_activation]: 9.86998e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.499e-05 [merge_recompute_call_nodes]: 1.77999e-06 [before_grad]: 1.016e-05 [set_forward_comm_id_for_comm_node_pass]: 3.86999e-06 [meta_fg_expand]: 2.48e-06 [flash_sp_send_recv_attached]: 1.79e-06 [receive_attached]: 2.68e-06 [after_resolve]: 1.275e-05 [a_after_grad]: 9.84001e-06 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 1.96998e-06 [auto_monad_grad]: 1.49e-06 [auto_monad_eliminator]: 7.97e-06 [cse]: 2.098e-05 [a_3]: 3.618e-05 [py_interpret_to_execute_after_opt_a]: 7.95998e-06 [slice_cell_reuse_recomputed_activation]: 2.04999e-06 [rewriter_after_opt_a]: 2.285e-05 [convert_after_rewriter]: 2.03997e-06 [order_py_execute_after_rewriter]: 1.26002e-06 [mutable_eliminate]: 0.00076873 [opt_b]: 0.00022084, [1] [Cycle 1]: 0.00021201, [7] [b_1]: 0.00012659 [b_2]: 9.01002e-06 [updatestate_depend_eliminate]: 9.31998e-06 [updatestate_assign_eliminate]: 3.21001e-06 [updatestate_loads_eliminate]: 3.11001e-06 [renormalize]: 5.69999e-07 [cse]: 2.527e-05 [optimize_parallel_all_gather_comm]: 1.94e-05 [overlap_param_gather]: 1.94999e-06 [cconv]: 2.731e-05 [loop_unroll]: 0.00047394 [opt_after_cconv]: 0.00010605, [1] [Cycle 1]: 9.933e-05, [7] [c_1]: 2.973e-05 [parameter_eliminate]: 3.19001e-06 [updatestate_depend_eliminate]: 5.34e-06 [updatestate_assign_eliminate]: 2.54999e-06 [updatestate_loads_eliminate]: 2.31e-06 [cse]: 2.231e-05 [renormalize]: 3.39991e-07 [remove_dup_value]: 4.083e-05 [tuple_transform]: 7.829e-05, [1] [Cycle 1]: 7.129e-05, [4] [d_1]: 4.423e-05 [none_parameter_eliminate]: 1.66e-06 [renormalize]: 1.39989e-07 [switch_simplify]: 7.03e-06 [partial_unused_args_eliminate]: 1.77999e-06 [add_recomputation]: 5.433e-05 [cse_after_recomputation]: 2.512e-05, [1] [Cycle 1]: 2.051e-05, [1] [cse]: 1.434e-05 [environ_conv]: 5.90002e-06 [swap_dp_allreduce_reducescatter]: 5.45001e-06 [bias_add_comm_swap]: 3.26001e-06 [label_micro_interleaved_index]: 4.68001e-06 [label_fine_grained_interleaved_index]: 2.83998e-06 [merge_cast_opt]: 1.57999e-06 [slice_recompute_activation]: 2.36998e-06 [micro_interleaved_order_control]: 2.90002e-06 [assign_add_opt]: 1.37e-06 [ForceFp32Comm]: 1.07e-06 [remove_cast_before_assign_add]: 9.89996e-07 [full_micro_interleaved_order_control]: 2.66999e-06 [reorder_send_recv_between_fp_bp]: 3.10998e-06 [comm_op_add_attrs]: 1.11002e-06 [add_comm_op_reuse_tag]: 1.12e-06 [interleave_split_concat_branches]: 1.37999e-06 [interleave_parallel_branches]: 1.10999e-06 [overlap_opt_shard_in_pipeline]: 1.62999e-06 [overlap_opt_shard_grad_in_pipeline]: 1.87001e-06 [control_data_broadcast_order]: 1.502e-05 [grouped_pairwise_exchange_alltoall]: 1.47999e-06 [offloading_packed_experts]: 5.39e-06 [overlap_recompute_and_grad_model_parallel]: 5.17e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.36998e-06 [overlap_recompute_allgather_and_fa_grad]: 1.29003e-06 [overlap_recompute_comm]: 2.56e-06 [overlap_grad_ring_attention]: 4.75001e-06 [overlap_grad_flash_sp]: 2.337e-05 [begin_end_overlap_inline]: 5.60016e-07 [split_matmul_comm_elemetwise]: 2.62001e-06 [split_layernorm_comm]: 1.85001e-06 [handle_group_info]: 1.18001e-06 [symbol_engine_optimizer]: 0.0003694, [1] [Cycle 1]: 0.00036274, [6] [build]: 0.00026348 [elim_shapecalc]: 1.595e-05 [elim_not_effective]: 1.741e-05 [opt_reshape]: 7.73001e-06 [fold_const_symbol]: 1.616e-05 [renormalize]: 3.39991e-07 [detach_backward]: 3.01999e-06 [pipeline_parallel_scheduler]: 1.92999e-06 [auto_monad_reorder]: 2.314e-05 [get_jit_bprop_graph]: 2.02999e-06 [rewriter_after_jit_bprop_graph]: 4.32e-06 [opt_after_jit_grad]: 0.00056429 [validate]: 4.214e-05 Sums bootstrap : 0.000728s : 0.18% type_inference : 0.209695s : 53.17% event_method : 0.000079s : 0.02% auto_monad : 0.000085s : 0.02% graph_reusing : 0.000008s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000060s : 0.02% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000009s : 0.00% parallel-infer-symbol : 0.000005s : 0.00% pre_auto_parallel : 0.000068s : 0.02% 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 : 0.000008s : 0.00% optimize.rewriter_before_opt_a : 0.000363s : 0.09% optimize.opt_a.expand_dump_flag : 0.000006s : 0.00% optimize.opt_a.switch_simplify : 0.000077s : 0.02% optimize.opt_a.loop_unroll : 0.000057s : 0.01% optimize.opt_a.a_1 : 0.001254s : 0.32% optimize.opt_a.with_stream_mark : 0.000044s : 0.01% optimize.opt_a.recompute_prepare : 0.000020s : 0.01% 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 : 0.000006s : 0.00% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000160s : 0.04% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.00% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000006s : 0.00% optimize.opt_a.shard_inline : 0.000013s : 0.00% optimize.opt_a.merge_send_recv : 0.000020s : 0.01% optimize.opt_a.auto_parallel : 0.000020s : 0.01% optimize.opt_a.parallel : 0.000043s : 0.01% optimize.opt_a.flash_sp : 0.000015s : 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 : 0.000022s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000019s : 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 : 0.000012s : 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 : 0.000020s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000030s : 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 : 0.000007s : 0.00% 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 : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000027s : 0.01% optimize.opt_a.a_after_grad : 0.000019s : 0.00% optimize.opt_a.renormalize : 0.178234s : 45.19% optimize.opt_a.add_forward_monad_depend : 0.000011s : 0.00% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000032s : 0.01% optimize.opt_a.cse : 0.000066s : 0.02% optimize.opt_a.a_3 : 0.000106s : 0.03% optimize.py_interpret_to_execute_after_opt_a : 0.000008s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000023s : 0.01% optimize.convert_after_rewriter : 0.000002s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000769s : 0.19% optimize.opt_b.b_1 : 0.000127s : 0.03% optimize.opt_b.b_2 : 0.000009s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000009s : 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 : 0.000025s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000019s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000027s : 0.01% optimize.loop_unroll : 0.000474s : 0.12% optimize.opt_after_cconv.c_1 : 0.000030s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000003s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000005s : 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 : 0.000022s : 0.01% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000041s : 0.01% optimize.tuple_transform.d_1 : 0.000044s : 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 : 0.000054s : 0.01% optimize.cse_after_recomputation.cse : 0.000014s : 0.00% 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 : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000003s : 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 : 0.000015s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000001s : 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 : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000005s : 0.00% optimize.overlap_grad_flash_sp : 0.000023s : 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 : 0.000263s : 0.07% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000016s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000017s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000016s : 0.00% 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 : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000004s : 0.00% opt_after_jit_grad : 0.000564s : 0.14% validate : 0.000042s : 0.01% Time group info: ------[substitution.] 0.000311 35 1.39% : 0.000004s : 2: substitution.elim_not_effective 1.98% : 0.000006s : 2: substitution.fold_const_symbol 2.28% : 0.000007s : 4: substitution.graph_param_transform 80.32% : 0.000250s : 7: substitution.inline 1.49% : 0.000005s : 4: substitution.j_node_and_user_rematch 1.76% : 0.000005s : 4: substitution.remove_not_recompute_node 2.06% : 0.000006s : 4: substitution.replace_old_param 8.72% : 0.000027s : 8: substitution.tuple_list_get_item_eliminator ------[type_inference.] 0.209602 2 99.10% : 0.207724s : 1: type_inference.infer 0.90% : 0.001877s : 1: type_inference.specialize ------[replace.] 0.000130 15 57.00% : 0.000074s : 7: replace.inline 43.00% : 0.000056s : 8: replace.tuple_list_get_item_eliminator ------[match.] 0.000269 15 91.45% : 0.000246s : 7: match.inline 8.55% : 0.000023s : 8: match.tuple_list_get_item_eliminator ------[predicate.] 0.000281 1733 1.24% : 0.000003s : 20: predicate.accumulaten_eliminater 0.55% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.34% : 0.000001s : 8: predicate.addn_check_dump 1.01% : 0.000003s : 20: predicate.addn_zero_filter 0.90% : 0.000003s : 20: predicate.adjust_all_reduce_mul_add 2.09% : 0.000006s : 28: predicate.arithmetic_simplify 1.20% : 0.000003s : 20: predicate.cast_eliminate 0.42% : 0.000001s : 8: predicate.check_bprop_eliminate 0.46% : 0.000001s : 8: predicate.compare_switch_simplify 0.13% : 0.000000s : 4: predicate.const_output_eliminate 0.42% : 0.000001s : 8: predicate.depend_value_elim 1.08% : 0.000003s : 20: predicate.dict_get_item_const_eliminator 1.17% : 0.000003s : 20: predicate.dict_get_item_eliminator 1.06% : 0.000003s : 20: predicate.dict_set_item_eliminator 0.58% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.19% : 0.000001s : 4: predicate.elim_not_effective 0.38% : 0.000001s : 4: predicate.elim_shapecalc_of_broadcastargs 1.22% : 0.000003s : 24: predicate.environ_add_const_eliminate 1.17% : 0.000003s : 24: predicate.environ_get_add_eliminate 1.26% : 0.000004s : 24: predicate.environ_get_depend_swap 1.58% : 0.000004s : 32: predicate.environ_get_eliminate 1.26% : 0.000004s : 24: predicate.environ_get_set_eliminate 1.79% : 0.000005s : 35: predicate.exchange_switch_depend_value 2.82% : 0.000008s : 35: predicate.float_depend_g_call 0.32% : 0.000001s : 8: predicate.float_environ_get_switch 0.59% : 0.000002s : 12: predicate.float_tuple_getitem_switch 0.12% : 0.000000s : 4: predicate.fold_const_symbol 0.49% : 0.000001s : 8: predicate.get_grad_eliminate 0.14% : 0.000000s : 4: predicate.graph_param_transform 0.40% : 0.000001s : 8: predicate.incorporate_call 0.32% : 0.000001s : 8: predicate.incorporate_call_switch 6.00% : 0.000017s : 79: predicate.inline 0.48% : 0.000001s : 8: predicate.inline_without_move 0.24% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.74% : 0.000002s : 8: predicate.less_batch_normalization 1.88% : 0.000005s : 36: predicate.list_to_tuple_eliminator_ 2.81% : 0.000008s : 56: predicate.load_eliminater 0.46% : 0.000001s : 4: predicate.loop_unroll_after_grad 2.82% : 0.000008s : 58: predicate.loop_unroll_before_grad 1.76% : 0.000005s : 28: predicate.make_slice_get_slice_eliminator 0.49% : 0.000001s : 8: predicate.merge_addn 0.36% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.36% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.94% : 0.000003s : 20: predicate.minmaximum_grad 0.97% : 0.000003s : 4: predicate.mutable_eliminate 0.22% : 0.000001s : 4: predicate.opt_reshape 0.32% : 0.000001s : 4: predicate.parallel_virtual_node 2.32% : 0.000007s : 35: predicate.partial_defer_inline 1.79% : 0.000005s : 32: predicate.partial_eliminate 1.22% : 0.000003s : 20: predicate.print_const_string_wrapper 0.39% : 0.000001s : 8: predicate.reduce_all_const_elim 1.30% : 0.000004s : 20: predicate.reduce_eliminate 2.80% : 0.000008s : 56: predicate.redundant_stop_gradient_eliminater 0.34% : 0.000001s : 8: predicate.remove_not_recompute_node 1.65% : 0.000005s : 36: predicate.replace_applicator 0.35% : 0.000001s : 8: predicate.replace_old_param 0.20% : 0.000001s : 4: predicate.reset_defer_inline 1.23% : 0.000003s : 20: predicate.reshape_eliminate 0.44% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.30% : 0.000001s : 4: predicate.row_tensor_eliminate 0.59% : 0.000002s : 8: predicate.same_eliminate 0.40% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.78% : 0.000002s : 8: predicate.shard_identity_eliminate 0.65% : 0.000002s : 8: predicate.special_op_eliminate 0.45% : 0.000001s : 8: predicate.specialize_transform 0.71% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.56% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.21% : 0.000001s : 4: predicate.switch_call_monad_eliminater 1.95% : 0.000005s : 35: predicate.switch_defer_inline 2.31% : 0.000006s : 43: predicate.switch_layer_defer_inline 6.01% : 0.000017s : 105: predicate.switch_simplify 0.99% : 0.000003s : 20: predicate.tile_eliminate 1.12% : 0.000003s : 20: predicate.transpose_eliminate 1.59% : 0.000004s : 28: predicate.tuple_list_convert_item_index_to_positive 1.59% : 0.000004s : 28: predicate.tuple_list_get_item_const_eliminator 1.47% : 0.000004s : 28: predicate.tuple_list_get_item_depend_reorder 3.43% : 0.000010s : 44: predicate.tuple_list_get_item_eliminator 1.66% : 0.000005s : 28: predicate.tuple_list_get_set_item_eliminator 2.15% : 0.000006s : 36: predicate.tuple_list_set_item_eliminator 2.00% : 0.000006s : 36: predicate.tuple_to_list_eliminator_ 2.78% : 0.000008s : 56: predicate.updatestate_pure_node_eliminater 3.14% : 0.000009s : 64: predicate.updatestate_useless_node_eliminater 0.25% : 0.000001s : 4: predicate.value_based_eliminate 0.45% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.45% : 0.000001s : 8: predicate.virtual_output_eliminate 0.21% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.24% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.001780 18 57.41% : 0.001022s : 9: func_graph_cloner_run.FuncGraphClonerGraph 42.59% : 0.000758s : 9: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.770118 192 0.00% : 0.000004s : 1: ForceFp32Comm 0.68% : 0.005213s : 1: add_attr 0.67% : 0.005194s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000058s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.01% : 0.000093s : 1: auto_monad 0.00% : 0.000028s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.10% : 0.000787s : 1: bootstrap 0.00% : 0.000031s : 1: cconv 0.00% : 0.000023s : 1: comm_op_add_attrs 0.00% : 0.000018s : 1: control_data_broadcast_order 0.00% : 0.000005s : 1: convert_after_rewriter 0.00% : 0.000029s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000009s : 1: environ_conv 0.01% : 0.000090s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000012s : 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% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.06% : 0.000484s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.10% : 0.000783s : 1: mutable_eliminate 0.00% : 0.000010s : 1: offloading_packed_experts 0.00% : 0.000015s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000021s : 1: opt.transform.mutable_eliminate 0.23% : 0.001753s : 78: opt.transform.opt_a 0.00% : 0.000028s : 1: opt.transform.opt_after_cconv 0.00% : 0.000030s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000104s : 28: opt.transform.opt_b 0.01% : 0.000049s : 2: opt.transform.opt_trans_graph 0.01% : 0.000052s : 4: opt.transform.symbol_engine_opt 23.51% : 0.181019s : 1: opt_a 0.01% : 0.000109s : 1: opt_after_cconv 0.07% : 0.000575s : 1: opt_after_jit_grad 0.03% : 0.000224s : 1: opt_b 23.90% : 0.184062s : 1: optimize 0.00% : 0.000023s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 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% : 0.000005s : 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% : 0.000013s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000073s : 1: pre_auto_parallel 0.00% : 0.000012s : 1: py_interpret_to_execute 0.00% : 0.000012s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.01% : 0.000045s : 1: remove_dup_value 23.02% : 0.177306s : 1: renormalize.infer 0.12% : 0.000909s : 1: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000008s : 1: rewriter_after_jit_bprop_graph 0.00% : 0.000027s : 1: rewriter_after_opt_a 0.05% : 0.000372s : 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% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000008s : 1: swap_dp_allreduce_reducescatter 0.05% : 0.000373s : 1: symbol_engine_optimizer 0.01% : 0.000081s : 1: tuple_transform 27.23% : 0.209724s : 1: type_inference TotalTime = 0.72404, [21] [bootstrap]: 0.00051397 [type_inference]: 0.355708 [event_method]: 0.00025083 [auto_monad]: 0.00022556 [graph_reusing]: 1.148e-05 [inline]: 3.24001e-06 [add_attr]: 0.00434096, [1] [add_attr_with_inline]: 0.00432489, [1] [Cycle 1]: 0.00015085, [2] [tag_attr]: 8.235e-05 [meta_addattr_fg_expand]: 1.714e-05 [parallel-infer-symbol]: 3.43999e-06 [pre_auto_parallel]: 0.00010091 [insert-virtual-dataset]: 2.97002e-06 [parallel-infer-symbol-second]: 8.59989e-07 [dataset_repeat_opt]: 2.86e-06 [pipeline_split]: 2.14e-06 [optimize]: 0.361616, [53] [py_interpret_to_execute]: 8.97999e-06 [rewriter_before_opt_a]: 0.00059915 [opt_a]: 0.346774, [3] [Cycle 1]: 0.18487, [45] [expand_dump_flag]: 7.23e-06 [switch_simplify]: 0.169733 [loop_unroll]: 0.0001908 [a_1]: 0.00246853 [with_stream_mark]: 4.395e-05 [recompute_prepare]: 3.422e-05 [updatestate_depend_eliminate]: 1.265e-05 [updatestate_assign_eliminate]: 8.39002e-06 [updatestate_loads_eliminate]: 8.40999e-06 [parameter_eliminate]: 4.28001e-06 [a_2]: 0.00026825 [accelerated_algorithm]: 1.88e-05 [shard]: 2.69999e-06 [meta_shard_fg_expand]: 7.55998e-06 [shard_inline]: 1.69e-05 [merge_send_recv]: 2.147e-05 [auto_parallel]: 1.801e-05 [parallel]: 2.368e-05 [flash_sp]: 1.439e-05 [merge_comm]: 1.126e-05 [allreduce_fusion]: 9.61e-06 [matmul_add_comm_reduction]: 3.704e-05 [allreduce_slice_to_reducescatter]: 1.15999e-06 [virtual_shard_identity]: 2.066e-05 [virtual_dataset]: 1.682e-05 [get_grad_eliminate_]: 1.646e-05 [virtual_output]: 1.679e-05 [merge_forward]: 1.074e-05 [cell_reuse_recompute_pass]: 1.57999e-06 [offload_activation]: 2.002e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.413e-05 [merge_recompute_call_nodes]: 1.49e-06 [before_grad]: 3.11e-05 [set_forward_comm_id_for_comm_node_pass]: 1.187e-05 [meta_fg_expand]: 0.00221952 [flash_sp_send_recv_attached]: 4.63001e-06 [receive_attached]: 2.89999e-06 [after_resolve]: 8.549e-05 [a_after_grad]: 0.00010139 [renormalize]: 0.00787462 [add_forward_monad_depend]: 1.599e-05 [auto_monad_grad]: 7.55e-06 [auto_monad_eliminator]: 7.692e-05 [cse]: 0.00048443 [a_3]: 0.00040145 [Cycle 2]: 0.16077, [45] [expand_dump_flag]: 2.81e-06 [switch_simplify]: 5.357e-05 [loop_unroll]: 4.852e-05 [a_1]: 0.001663 [with_stream_mark]: 9.623e-05 [recompute_prepare]: 2.163e-05 [updatestate_depend_eliminate]: 8.22e-06 [updatestate_assign_eliminate]: 5.21002e-06 [updatestate_loads_eliminate]: 5.07999e-06 [parameter_eliminate]: 2.76999e-06 [a_2]: 0.00014043 [accelerated_algorithm]: 1.167e-05 [shard]: 2.61e-06 [meta_shard_fg_expand]: 3.28998e-06 [shard_inline]: 1.067e-05 [merge_send_recv]: 1.276e-05 [auto_parallel]: 1.419e-05 [parallel]: 9.74999e-06 [flash_sp]: 4.33001e-06 [merge_comm]: 6.78998e-06 [allreduce_fusion]: 5.56e-06 [matmul_add_comm_reduction]: 1.357e-05 [allreduce_slice_to_reducescatter]: 8.40024e-07 [virtual_shard_identity]: 1.406e-05 [virtual_dataset]: 9.94999e-06 [get_grad_eliminate_]: 1.056e-05 [virtual_output]: 1.083e-05 [merge_forward]: 6.42001e-06 [cell_reuse_recompute_pass]: 1.96998e-06 [offload_activation]: 1.572e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.549e-05 [merge_recompute_call_nodes]: 1.67001e-06 [before_grad]: 1.792e-05 [set_forward_comm_id_for_comm_node_pass]: 5.89e-06 [meta_fg_expand]: 0.00031615 [flash_sp_send_recv_attached]: 2.78e-06 [receive_attached]: 2.71999e-06 [after_resolve]: 2.791e-05 [a_after_grad]: 1.685e-05 [renormalize]: 0.15755 [add_forward_monad_depend]: 1.32e-05 [auto_monad_grad]: 2.84999e-06 [auto_monad_eliminator]: 3.462e-05 [cse]: 0.00010103 [a_3]: 9.867e-05 [Cycle 3]: 0.00110691, [45] [expand_dump_flag]: 2.73998e-06 [switch_simplify]: 1.499e-05 [loop_unroll]: 1.071e-05 [a_1]: 0.00030026 [with_stream_mark]: 2.247e-05 [recompute_prepare]: 1.162e-05 [updatestate_depend_eliminate]: 6.70002e-06 [updatestate_assign_eliminate]: 5.51998e-06 [updatestate_loads_eliminate]: 4.92e-06 [parameter_eliminate]: 2.11e-06 [a_2]: 0.00013292 [accelerated_algorithm]: 1.14e-05 [shard]: 2.36998e-06 [meta_shard_fg_expand]: 3.14999e-06 [shard_inline]: 1.016e-05 [merge_send_recv]: 1.151e-05 [auto_parallel]: 1.324e-05 [parallel]: 9.72001e-06 [flash_sp]: 1.59e-06 [merge_comm]: 6.88e-06 [allreduce_fusion]: 5.69e-06 [matmul_add_comm_reduction]: 1.344e-05 [allreduce_slice_to_reducescatter]: 1.08001e-06 [virtual_shard_identity]: 1.261e-05 [virtual_dataset]: 1.097e-05 [get_grad_eliminate_]: 9.71998e-06 [virtual_output]: 1.063e-05 [merge_forward]: 7.05002e-06 [cell_reuse_recompute_pass]: 3.81001e-06 [offload_activation]: 1.538e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.331e-05 [merge_recompute_call_nodes]: 1.50999e-06 [before_grad]: 1.86e-05 [set_forward_comm_id_for_comm_node_pass]: 7.13e-06 [meta_fg_expand]: 4.95999e-06 [flash_sp_send_recv_attached]: 2.61e-06 [receive_attached]: 2.61e-06 [after_resolve]: 1.664e-05 [a_after_grad]: 1.608e-05 [renormalize]: 9.00181e-08 [add_forward_monad_depend]: 2.12001e-06 [auto_monad_grad]: 2.05002e-06 [auto_monad_eliminator]: 1.43e-05 [cse]: 3.509e-05 [a_3]: 6.666e-05 [py_interpret_to_execute_after_opt_a]: 9.86e-06 [slice_cell_reuse_recomputed_activation]: 2.39001e-06 [rewriter_after_opt_a]: 3.486e-05 [convert_after_rewriter]: 1.27999e-06 [order_py_execute_after_rewriter]: 1.44e-06 [mutable_eliminate]: 0.00080735 [opt_b]: 0.00036272, [1] [Cycle 1]: 0.00035306, [7] [b_1]: 0.00022314 [b_2]: 1.446e-05 [updatestate_depend_eliminate]: 1.217e-05 [updatestate_assign_eliminate]: 6.11e-06 [updatestate_loads_eliminate]: 5.39e-06 [renormalize]: 1.51002e-06 [cse]: 4.957e-05 [optimize_parallel_all_gather_comm]: 2.761e-05 [overlap_param_gather]: 2.09999e-06 [cconv]: 3.919e-05 [loop_unroll]: 0.00063001 [opt_after_cconv]: 0.00018256, [1] [Cycle 1]: 0.00017426, [7] [c_1]: 5.919e-05 [parameter_eliminate]: 6.34999e-06 [updatestate_depend_eliminate]: 1.128e-05 [updatestate_assign_eliminate]: 4.85999e-06 [updatestate_loads_eliminate]: 4.56002e-06 [cse]: 4.809e-05 [renormalize]: 1.04998e-06 [remove_dup_value]: 6.172e-05 [tuple_transform]: 0.00012589, [1] [Cycle 1]: 0.0001204, [4] [d_1]: 8.485e-05 [none_parameter_eliminate]: 1.71e-06 [renormalize]: 2.10013e-07 [switch_simplify]: 1.309e-05 [partial_unused_args_eliminate]: 2.11e-06 [add_recomputation]: 9.259e-05 [cse_after_recomputation]: 6.521e-05, [1] [Cycle 1]: 5.392e-05, [1] [cse]: 4.143e-05 [environ_conv]: 1.386e-05 [swap_dp_allreduce_reducescatter]: 1.053e-05 [bias_add_comm_swap]: 3.57002e-06 [label_micro_interleaved_index]: 6.30002e-06 [label_fine_grained_interleaved_index]: 2.91e-06 [merge_cast_opt]: 1.55999e-06 [slice_recompute_activation]: 2.29999e-06 [micro_interleaved_order_control]: 2.56e-06 [assign_add_opt]: 1.27e-06 [ForceFp32Comm]: 1.37999e-06 [remove_cast_before_assign_add]: 1.15999e-06 [full_micro_interleaved_order_control]: 2.16e-06 [reorder_send_recv_between_fp_bp]: 2.87002e-06 [comm_op_add_attrs]: 1.15001e-06 [add_comm_op_reuse_tag]: 1.42e-06 [interleave_split_concat_branches]: 1.57999e-06 [interleave_parallel_branches]: 1.44e-06 [overlap_opt_shard_in_pipeline]: 1.36002e-06 [overlap_opt_shard_grad_in_pipeline]: 2.10002e-06 [control_data_broadcast_order]: 2.319e-05 [grouped_pairwise_exchange_alltoall]: 1.76998e-06 [offloading_packed_experts]: 6.56999e-06 [overlap_recompute_and_grad_model_parallel]: 7.02002e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.47001e-06 [overlap_recompute_allgather_and_fa_grad]: 1.43002e-06 [overlap_recompute_comm]: 2.36e-06 [overlap_grad_ring_attention]: 6.69999e-06 [overlap_grad_flash_sp]: 3.396e-05 [begin_end_overlap_inline]: 5.69999e-07 [split_matmul_comm_elemetwise]: 2.90998e-06 [split_layernorm_comm]: 1.97999e-06 [handle_group_info]: 1.39e-06 [symbol_engine_optimizer]: 0.0111951, [1] [Cycle 1]: 0.0111857, [6] [build]: 0.0109636 [elim_shapecalc]: 5.238e-05 [elim_not_effective]: 4.157e-05 [opt_reshape]: 2.085e-05 [fold_const_symbol]: 2.959e-05 [renormalize]: 6.50005e-07 [detach_backward]: 4.55999e-06 [pipeline_parallel_scheduler]: 2.49001e-06 [auto_monad_reorder]: 4.628e-05 [get_jit_bprop_graph]: 2.56998e-06 [rewriter_after_jit_bprop_graph]: 6.48003e-06 [opt_after_jit_grad]: 0.00086566 [validate]: 7.171e-05 Sums bootstrap : 0.000514s : 0.07% type_inference : 0.355708s : 49.55% event_method : 0.000251s : 0.03% auto_monad : 0.000226s : 0.03% graph_reusing : 0.000011s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000082s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000017s : 0.00% parallel-infer-symbol : 0.000003s : 0.00% pre_auto_parallel : 0.000101s : 0.01% 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 : 0.000009s : 0.00% optimize.rewriter_before_opt_a : 0.000599s : 0.08% optimize.opt_a.expand_dump_flag : 0.000013s : 0.00% optimize.opt_a.switch_simplify : 0.169801s : 23.65% optimize.opt_a.loop_unroll : 0.000250s : 0.03% optimize.opt_a.a_1 : 0.004432s : 0.62% optimize.opt_a.with_stream_mark : 0.000163s : 0.02% optimize.opt_a.recompute_prepare : 0.000067s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000028s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000019s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000018s : 0.00% optimize.opt_a.parameter_eliminate : 0.000009s : 0.00% optimize.opt_a.a_2 : 0.000542s : 0.08% optimize.opt_a.accelerated_algorithm : 0.000042s : 0.01% optimize.opt_a.shard : 0.000008s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000014s : 0.00% optimize.opt_a.shard_inline : 0.000038s : 0.01% optimize.opt_a.merge_send_recv : 0.000046s : 0.01% optimize.opt_a.auto_parallel : 0.000045s : 0.01% optimize.opt_a.parallel : 0.000043s : 0.01% optimize.opt_a.flash_sp : 0.000020s : 0.00% optimize.opt_a.merge_comm : 0.000025s : 0.00% optimize.opt_a.allreduce_fusion : 0.000021s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000064s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000003s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000047s : 0.01% optimize.opt_a.virtual_dataset : 0.000038s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000037s : 0.01% optimize.opt_a.virtual_output : 0.000038s : 0.01% optimize.opt_a.merge_forward : 0.000024s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000007s : 0.00% optimize.opt_a.offload_activation : 0.000051s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000083s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000068s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000025s : 0.00% optimize.opt_a.meta_fg_expand : 0.002541s : 0.35% optimize.opt_a.flash_sp_send_recv_attached : 0.000010s : 0.00% optimize.opt_a.receive_attached : 0.000008s : 0.00% optimize.opt_a.after_resolve : 0.000130s : 0.02% optimize.opt_a.a_after_grad : 0.000134s : 0.02% optimize.opt_a.renormalize : 0.165425s : 23.04% optimize.opt_a.add_forward_monad_depend : 0.000031s : 0.00% optimize.opt_a.auto_monad_grad : 0.000012s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000126s : 0.02% optimize.opt_a.cse : 0.000621s : 0.09% optimize.opt_a.a_3 : 0.000567s : 0.08% optimize.py_interpret_to_execute_after_opt_a : 0.000010s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000035s : 0.00% optimize.convert_after_rewriter : 0.000001s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000807s : 0.11% optimize.opt_b.b_1 : 0.000223s : 0.03% optimize.opt_b.b_2 : 0.000014s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000012s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000006s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_b.renormalize : 0.000002s : 0.00% optimize.opt_b.cse : 0.000050s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000028s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000039s : 0.01% optimize.loop_unroll : 0.000630s : 0.09% optimize.opt_after_cconv.c_1 : 0.000059s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000006s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000011s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.cse : 0.000048s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000062s : 0.01% optimize.tuple_transform.d_1 : 0.000085s : 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 : 0.000013s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000093s : 0.01% optimize.cse_after_recomputation.cse : 0.000041s : 0.01% optimize.environ_conv : 0.000014s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000011s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000006s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 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 : 0.000023s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 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 : 0.000007s : 0.00% optimize.overlap_grad_flash_sp : 0.000034s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.010964s : 1.53% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000052s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000042s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000021s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000030s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000001s : 0.00% detach_backward : 0.000005s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000046s : 0.01% get_jit_bprop_graph : 0.000003s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.000866s : 0.12% validate : 0.000072s : 0.01% Time group info: ------[substitution.] 0.001220 215 0.85% : 0.000010s : 5: substitution.elim_not_effective 0.85% : 0.000010s : 14: substitution.float_depend_g_call 0.53% : 0.000006s : 3: substitution.float_tuple_getitem_switch 1.10% : 0.000013s : 5: substitution.fold_const_symbol 0.75% : 0.000009s : 8: substitution.graph_param_transform 0.29% : 0.000004s : 2: substitution.incorporate_call 0.23% : 0.000003s : 2: substitution.incorporate_call_switch 67.76% : 0.000827s : 25: substitution.inline 1.81% : 0.000022s : 2: substitution.inline_without_move 1.15% : 0.000014s : 20: substitution.j_node_and_user_rematch 1.10% : 0.000013s : 7: substitution.minmaximum_grad 0.42% : 0.000005s : 1: substitution.opt_reshape 0.95% : 0.000012s : 14: substitution.partial_eliminate 1.45% : 0.000018s : 20: substitution.remove_not_recompute_node 2.87% : 0.000035s : 9: substitution.replace_applicator 1.29% : 0.000016s : 16: substitution.replace_old_param 2.01% : 0.000024s : 3: substitution.reshape_eliminate 0.36% : 0.000004s : 1: substitution.set_cell_output_no_recompute 1.29% : 0.000016s : 3: substitution.switch_simplify 2.12% : 0.000026s : 7: substitution.tuple_list_convert_item_index_to_positive 0.93% : 0.000011s : 7: substitution.tuple_list_get_item_const_eliminator 1.40% : 0.000017s : 7: substitution.tuple_list_get_item_depend_reorder 7.13% : 0.000087s : 27: substitution.tuple_list_get_item_eliminator 1.36% : 0.000017s : 7: substitution.tuple_list_get_set_item_eliminator ------[type_inference.] 0.355551 2 98.61% : 0.350601s : 1: type_inference.infer 1.39% : 0.004950s : 1: type_inference.specialize ------[replace.] 0.000427 45 54.06% : 0.000231s : 25: replace.inline 12.28% : 0.000052s : 3: replace.switch_simplify 33.67% : 0.000144s : 17: replace.tuple_list_get_item_eliminator ------[match.] 0.000874 45 92.93% : 0.000813s : 25: match.inline 1.43% : 0.000012s : 3: match.switch_simplify 5.65% : 0.000049s : 17: match.tuple_list_get_item_eliminator ------[predicate.] 0.000942 6131 1.08% : 0.000010s : 73: predicate.accumulaten_eliminater 0.40% : 0.000004s : 8: predicate.ad_related_special_op_eliminate 0.43% : 0.000004s : 31: predicate.addn_check_dump 1.04% : 0.000010s : 73: predicate.addn_zero_filter 0.95% : 0.000009s : 73: predicate.adjust_all_reduce_mul_add 2.10% : 0.000020s : 104: predicate.arithmetic_simplify 1.02% : 0.000010s : 73: predicate.cast_eliminate 0.99% : 0.000009s : 67: predicate.check_bprop_eliminate 0.48% : 0.000005s : 31: predicate.compare_switch_simplify 0.07% : 0.000001s : 8: predicate.const_output_eliminate 0.45% : 0.000004s : 31: predicate.depend_value_elim 1.04% : 0.000010s : 73: predicate.dict_get_item_const_eliminator 1.17% : 0.000011s : 73: predicate.dict_get_item_eliminator 1.01% : 0.000010s : 73: predicate.dict_set_item_eliminator 0.31% : 0.000003s : 16: predicate.dumpgradient_eliminate 0.15% : 0.000001s : 8: predicate.elim_not_effective 0.37% : 0.000003s : 8: predicate.elim_shapecalc_of_broadcastargs 1.16% : 0.000011s : 81: predicate.environ_add_const_eliminate 1.12% : 0.000011s : 81: predicate.environ_get_add_eliminate 1.09% : 0.000010s : 81: predicate.environ_get_depend_swap 1.59% : 0.000015s : 112: predicate.environ_get_eliminate 1.14% : 0.000011s : 81: predicate.environ_get_set_eliminate 1.68% : 0.000016s : 115: predicate.exchange_switch_depend_value 2.54% : 0.000024s : 115: predicate.float_depend_g_call 0.48% : 0.000004s : 31: predicate.float_environ_get_switch 0.60% : 0.000006s : 39: predicate.float_tuple_getitem_switch 0.07% : 0.000001s : 8: predicate.fold_const_symbol 0.49% : 0.000005s : 31: predicate.get_grad_eliminate 0.08% : 0.000001s : 8: predicate.graph_param_transform 0.48% : 0.000005s : 31: predicate.incorporate_call 0.40% : 0.000004s : 31: predicate.incorporate_call_switch 5.34% : 0.000050s : 266: predicate.inline 1.14% : 0.000011s : 60: predicate.inline_without_move 0.24% : 0.000002s : 31: predicate.j_node_and_user_rematch 0.62% : 0.000006s : 31: predicate.less_batch_normalization 1.57% : 0.000015s : 106: predicate.list_to_tuple_eliminator_ 2.67% : 0.000025s : 179: predicate.load_eliminater 0.49% : 0.000005s : 8: predicate.loop_unroll_after_grad 5.20% : 0.000049s : 175: predicate.loop_unroll_before_grad 1.44% : 0.000014s : 89: predicate.make_slice_get_slice_eliminator 0.47% : 0.000004s : 31: predicate.merge_addn 0.96% : 0.000009s : 67: predicate.micro_step_allgather_replace 0.93% : 0.000009s : 67: predicate.mini_step_allgather_replace 0.99% : 0.000009s : 73: predicate.minmaximum_grad 0.36% : 0.000003s : 8: predicate.mutable_eliminate 0.19% : 0.000002s : 8: predicate.opt_reshape 0.14% : 0.000001s : 8: predicate.parallel_virtual_node 2.18% : 0.000021s : 115: predicate.partial_defer_inline 1.57% : 0.000015s : 98: predicate.partial_eliminate 1.02% : 0.000010s : 73: predicate.print_const_string_wrapper 0.45% : 0.000004s : 31: predicate.reduce_all_const_elim 1.24% : 0.000012s : 73: predicate.reduce_eliminate 2.45% : 0.000023s : 179: predicate.redundant_stop_gradient_eliminater 0.29% : 0.000003s : 31: predicate.remove_not_recompute_node 1.75% : 0.000016s : 157: predicate.replace_applicator 0.65% : 0.000006s : 60: predicate.replace_old_param 0.11% : 0.000001s : 8: predicate.reset_defer_inline 1.11% : 0.000010s : 73: predicate.reshape_eliminate 0.97% : 0.000009s : 67: predicate.row_tensor_add_zeros_like 0.15% : 0.000001s : 8: predicate.row_tensor_eliminate 1.26% : 0.000012s : 67: predicate.same_eliminate 0.34% : 0.000003s : 31: predicate.set_cell_output_no_recompute 0.65% : 0.000006s : 31: predicate.shard_identity_eliminate 0.36% : 0.000003s : 16: predicate.special_op_eliminate 0.50% : 0.000005s : 31: predicate.specialize_transform 1.24% : 0.000012s : 67: predicate.split_environ_get_set_with_tuple_value 1.10% : 0.000010s : 60: predicate.stack_unstack_eliminate 0.15% : 0.000001s : 8: predicate.switch_call_monad_eliminater 1.88% : 0.000018s : 115: predicate.switch_defer_inline 2.72% : 0.000026s : 182: predicate.switch_layer_defer_inline 6.54% : 0.000062s : 335: predicate.switch_simplify 1.01% : 0.000010s : 73: predicate.tile_eliminate 1.00% : 0.000009s : 73: predicate.transpose_eliminate 1.38% : 0.000013s : 89: predicate.tuple_list_convert_item_index_to_positive 1.42% : 0.000013s : 89: predicate.tuple_list_get_item_const_eliminator 1.35% : 0.000013s : 89: predicate.tuple_list_get_item_depend_reorder 2.77% : 0.000026s : 137: predicate.tuple_list_get_item_eliminator 1.35% : 0.000013s : 89: predicate.tuple_list_get_set_item_eliminator 1.91% : 0.000018s : 120: predicate.tuple_list_set_item_eliminator 1.53% : 0.000014s : 106: predicate.tuple_to_list_eliminator_ 2.37% : 0.000022s : 179: predicate.updatestate_pure_node_eliminater 2.96% : 0.000028s : 210: predicate.updatestate_useless_node_eliminater 0.18% : 0.000002s : 8: predicate.value_based_eliminate 0.53% : 0.000005s : 31: predicate.virtual_dataset_eliminate 0.56% : 0.000005s : 31: predicate.virtual_output_eliminate 0.11% : 0.000001s : 8: predicate.virtual_view_grad_eliminate 0.15% : 0.000001s : 8: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.004731 54 56.67% : 0.002681s : 25: func_graph_cloner_run.FuncGraphClonerGraph 43.33% : 0.002050s : 29: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.431750 233 0.00% : 0.000004s : 1: ForceFp32Comm 0.30% : 0.004349s : 1: add_attr 0.30% : 0.004329s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000098s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.02% : 0.000239s : 1: auto_monad 0.00% : 0.000051s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.04% : 0.000591s : 1: bootstrap 0.00% : 0.000044s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000027s : 1: control_data_broadcast_order 0.00% : 0.000004s : 1: convert_after_rewriter 0.00% : 0.000069s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000008s : 1: detach_backward 0.00% : 0.000017s : 1: environ_conv 0.02% : 0.000268s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000016s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000005s : 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% : 0.000006s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000010s : 1: label_micro_interleaved_index 0.05% : 0.000646s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.06% : 0.000823s : 1: mutable_eliminate 0.00% : 0.000009s : 1: offloading_packed_experts 0.00% : 0.000029s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000027s : 1: opt.transform.mutable_eliminate 12.30% : 0.176116s : 117: opt.transform.opt_a 0.00% : 0.000058s : 1: opt.transform.opt_after_cconv 0.00% : 0.000047s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000206s : 28: opt.transform.opt_b 0.01% : 0.000095s : 2: opt.transform.opt_trans_graph 0.01% : 0.000133s : 4: opt.transform.symbol_engine_opt 24.22% : 0.346778s : 1: opt_a 0.01% : 0.000187s : 1: opt_after_cconv 0.06% : 0.000879s : 1: opt_after_jit_grad 0.03% : 0.000366s : 1: opt_b 25.26% : 0.361624s : 1: optimize 0.00% : 0.000032s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 1: order_py_execute_after_rewriter 0.00% : 0.000038s : 1: overlap_grad_flash_sp 0.00% : 0.000006s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000010s : 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% : 0.000011s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000007s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000106s : 1: pre_auto_parallel 0.00% : 0.000013s : 1: py_interpret_to_execute 0.00% : 0.000013s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000067s : 1: remove_dup_value 11.33% : 0.162218s : 2: renormalize.infer 0.22% : 0.003175s : 2: 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% : 0.000039s : 1: rewriter_after_opt_a 0.04% : 0.000610s : 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% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000014s : 1: swap_dp_allreduce_reducescatter 0.78% : 0.011200s : 1: symbol_engine_optimizer 0.01% : 0.000129s : 1: tuple_transform 24.85% : 0.355736s : 1: type_inference . TotalTime = 0.441907, [21] [bootstrap]: 0.0004845 [type_inference]: 0.194872 [event_method]: 7.727e-05 [auto_monad]: 9.308e-05 [graph_reusing]: 8e-06 [inline]: 3.28998e-06 [add_attr]: 0.00479144, [1] [add_attr_with_inline]: 0.00477598, [1] [Cycle 1]: 9.587e-05, [2] [tag_attr]: 4.148e-05 [meta_addattr_fg_expand]: 8.64998e-06 [parallel-infer-symbol]: 3.61001e-06 [pre_auto_parallel]: 5.799e-05 [insert-virtual-dataset]: 3.35003e-06 [parallel-infer-symbol-second]: 9.70002e-07 [dataset_repeat_opt]: 2.26998e-06 [pipeline_split]: 1.89e-06 [optimize]: 0.240296, [53] [py_interpret_to_execute]: 8.22e-06 [rewriter_before_opt_a]: 0.0003594 [opt_a]: 0.00425508, [2] [Cycle 1]: 0.00356065, [45] [expand_dump_flag]: 3.76001e-06 [switch_simplify]: 6.53e-05 [loop_unroll]: 5.07e-05 [a_1]: 0.001101 [with_stream_mark]: 3.09e-05 [recompute_prepare]: 1.501e-05 [updatestate_depend_eliminate]: 5.36998e-06 [updatestate_assign_eliminate]: 3.35003e-06 [updatestate_loads_eliminate]: 3.28e-06 [parameter_eliminate]: 2.42001e-06 [a_2]: 8.452e-05 [accelerated_algorithm]: 7.75998e-06 [shard]: 2.12999e-06 [meta_shard_fg_expand]: 2.54999e-06 [shard_inline]: 6.36e-06 [merge_send_recv]: 9.62001e-06 [auto_parallel]: 9.76e-06 [parallel]: 2.05e-05 [flash_sp]: 1.2e-05 [merge_comm]: 4.23001e-06 [allreduce_fusion]: 3.4e-06 [matmul_add_comm_reduction]: 1.017e-05 [allreduce_slice_to_reducescatter]: 9.09989e-07 [virtual_shard_identity]: 9.59e-06 [virtual_dataset]: 6.86001e-06 [get_grad_eliminate_]: 6.02999e-06 [virtual_output]: 6.36e-06 [merge_forward]: 4.67e-06 [cell_reuse_recompute_pass]: 2.09999e-06 [offload_activation]: 1.108e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.653e-05 [merge_recompute_call_nodes]: 1.66e-06 [before_grad]: 1.118e-05 [set_forward_comm_id_for_comm_node_pass]: 3.42997e-06 [meta_fg_expand]: 2.84999e-06 [flash_sp_send_recv_attached]: 3.31999e-06 [receive_attached]: 2.27999e-06 [after_resolve]: 1.304e-05 [a_after_grad]: 9.92001e-06 [renormalize]: 0.00157399 [add_forward_monad_depend]: 9.89001e-06 [auto_monad_grad]: 2.91999e-06 [auto_monad_eliminator]: 2.197e-05 [cse]: 4.29e-05 [a_3]: 5.882e-05 [Cycle 2]: 0.00067905, [45] [expand_dump_flag]: 2.29999e-06 [switch_simplify]: 9.14998e-06 [loop_unroll]: 6.58e-06 [a_1]: 0.0001259 [with_stream_mark]: 1.769e-05 [recompute_prepare]: 7.02002e-06 [updatestate_depend_eliminate]: 3.86001e-06 [updatestate_assign_eliminate]: 3.06001e-06 [updatestate_loads_eliminate]: 3.31001e-06 [parameter_eliminate]: 1.80001e-06 [a_2]: 7.1e-05 [accelerated_algorithm]: 7.10998e-06 [shard]: 2.63e-06 [meta_shard_fg_expand]: 2.67001e-06 [shard_inline]: 6.69999e-06 [merge_send_recv]: 8.79003e-06 [auto_parallel]: 9.36e-06 [parallel]: 8.64e-06 [flash_sp]: 3.55e-06 [merge_comm]: 3.42997e-06 [allreduce_fusion]: 3.16999e-06 [matmul_add_comm_reduction]: 9.13002e-06 [allreduce_slice_to_reducescatter]: 7.30011e-07 [virtual_shard_identity]: 6.60997e-06 [virtual_dataset]: 5.96998e-06 [get_grad_eliminate_]: 5.89999e-06 [virtual_output]: 5.47999e-06 [merge_forward]: 4.37e-06 [cell_reuse_recompute_pass]: 3.49001e-06 [offload_activation]: 1.02e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.707e-05 [merge_recompute_call_nodes]: 1.45999e-06 [before_grad]: 9.91e-06 [set_forward_comm_id_for_comm_node_pass]: 3.92998e-06 [meta_fg_expand]: 2.48002e-06 [flash_sp_send_recv_attached]: 1.98997e-06 [receive_attached]: 2.22001e-06 [after_resolve]: 1.182e-05 [a_after_grad]: 9.02999e-06 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 1.75001e-06 [auto_monad_grad]: 1.54998e-06 [auto_monad_eliminator]: 7.35003e-06 [cse]: 1.722e-05 [a_3]: 3.565e-05 [py_interpret_to_execute_after_opt_a]: 7.6e-06 [slice_cell_reuse_recomputed_activation]: 2.55002e-06 [rewriter_after_opt_a]: 2.565e-05 [convert_after_rewriter]: 1.79e-06 [order_py_execute_after_rewriter]: 1.43002e-06 [mutable_eliminate]: 0.00076311 [opt_b]: 0.00021894, [1] [Cycle 1]: 0.00020853, [7] [b_1]: 0.00012236 [b_2]: 8.96002e-06 [updatestate_depend_eliminate]: 8.01001e-06 [updatestate_assign_eliminate]: 2.94001e-06 [updatestate_loads_eliminate]: 2.73e-06 [renormalize]: 6.59988e-07 [cse]: 2.809e-05 [optimize_parallel_all_gather_comm]: 1.835e-05 [overlap_param_gather]: 2.18002e-06 [cconv]: 3.503e-05 [loop_unroll]: 0.00048749 [opt_after_cconv]: 0.0001116, [1] [Cycle 1]: 0.00010438, [7] [c_1]: 3.066e-05 [parameter_eliminate]: 4.79998e-06 [updatestate_depend_eliminate]: 6.61e-06 [updatestate_assign_eliminate]: 2.53003e-06 [updatestate_loads_eliminate]: 2.53998e-06 [cse]: 2.357e-05 [renormalize]: 3.9002e-07 [remove_dup_value]: 4.351e-05 [tuple_transform]: 7.962e-05, [1] [Cycle 1]: 7.389e-05, [4] [d_1]: 4.606e-05 [none_parameter_eliminate]: 2.27001e-06 [renormalize]: 1.69995e-07 [switch_simplify]: 7.93001e-06 [partial_unused_args_eliminate]: 1.94e-06 [add_recomputation]: 5.242e-05 [cse_after_recomputation]: 2.482e-05, [1] [Cycle 1]: 2.022e-05, [1] [cse]: 1.44e-05 [environ_conv]: 5.77001e-06 [swap_dp_allreduce_reducescatter]: 5.71003e-06 [bias_add_comm_swap]: 2.74001e-06 [label_micro_interleaved_index]: 4.68999e-06 [label_fine_grained_interleaved_index]: 2.81999e-06 [merge_cast_opt]: 1.67999e-06 [slice_recompute_activation]: 1.97999e-06 [micro_interleaved_order_control]: 2.66999e-06 [assign_add_opt]: 1.30001e-06 [ForceFp32Comm]: 9.10019e-07 [remove_cast_before_assign_add]: 1.64e-06 [full_micro_interleaved_order_control]: 2.22999e-06 [reorder_send_recv_between_fp_bp]: 2.79001e-06 [comm_op_add_attrs]: 1.03001e-06 [add_comm_op_reuse_tag]: 1.14e-06 [interleave_split_concat_branches]: 1.19e-06 [interleave_parallel_branches]: 1.12e-06 [overlap_opt_shard_in_pipeline]: 1.28002e-06 [overlap_opt_shard_grad_in_pipeline]: 1.86e-06 [control_data_broadcast_order]: 1.439e-05 [grouped_pairwise_exchange_alltoall]: 1.59e-06 [offloading_packed_experts]: 3.95e-06 [overlap_recompute_and_grad_model_parallel]: 4.68999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.44998e-06 [overlap_recompute_allgather_and_fa_grad]: 1.57999e-06 [overlap_recompute_comm]: 2.24001e-06 [overlap_grad_ring_attention]: 4.37e-06 [overlap_grad_flash_sp]: 2.288e-05 [begin_end_overlap_inline]: 6.30011e-07 [split_matmul_comm_elemetwise]: 2.38002e-06 [split_layernorm_comm]: 1.62001e-06 [handle_group_info]: 1.03001e-06 [symbol_engine_optimizer]: 0.233388, [1] [Cycle 1]: 0.233379, [6] [build]: 0.233172 [elim_shapecalc]: 5.371e-05 [elim_not_effective]: 3.081e-05 [opt_reshape]: 1.04e-05 [fold_const_symbol]: 2.179e-05 [renormalize]: 8.89995e-07 [detach_backward]: 4.71002e-06 [pipeline_parallel_scheduler]: 2.69999e-06 [auto_monad_reorder]: 3.772e-05 [get_jit_bprop_graph]: 3.11001e-06 [rewriter_after_jit_bprop_graph]: 1.063e-05 [opt_after_jit_grad]: 0.00091157 [validate]: 4.744e-05 Sums bootstrap : 0.000484s : 0.11% type_inference : 0.194872s : 44.70% event_method : 0.000077s : 0.02% auto_monad : 0.000093s : 0.02% graph_reusing : 0.000008s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000041s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000009s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000058s : 0.01% 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 : 0.000008s : 0.00% optimize.rewriter_before_opt_a : 0.000359s : 0.08% optimize.opt_a.expand_dump_flag : 0.000006s : 0.00% optimize.opt_a.switch_simplify : 0.000074s : 0.02% optimize.opt_a.loop_unroll : 0.000057s : 0.01% optimize.opt_a.a_1 : 0.001227s : 0.28% optimize.opt_a.with_stream_mark : 0.000049s : 0.01% optimize.opt_a.recompute_prepare : 0.000022s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000009s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000006s : 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 : 0.000156s : 0.04% optimize.opt_a.accelerated_algorithm : 0.000015s : 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 : 0.000018s : 0.00% optimize.opt_a.auto_parallel : 0.000019s : 0.00% optimize.opt_a.parallel : 0.000029s : 0.01% optimize.opt_a.flash_sp : 0.000016s : 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 : 0.000019s : 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 : 0.000012s : 0.00% optimize.opt_a.merge_forward : 0.000009s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000006s : 0.00% optimize.opt_a.offload_activation : 0.000021s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000034s : 0.01% 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 : 0.000005s : 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 : 0.01% optimize.opt_a.a_after_grad : 0.000019s : 0.00% optimize.opt_a.renormalize : 0.001574s : 0.36% optimize.opt_a.add_forward_monad_depend : 0.000012s : 0.00% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000029s : 0.01% optimize.opt_a.cse : 0.000060s : 0.01% optimize.opt_a.a_3 : 0.000094s : 0.02% optimize.py_interpret_to_execute_after_opt_a : 0.000008s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000026s : 0.01% optimize.convert_after_rewriter : 0.000002s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000763s : 0.18% optimize.opt_b.b_1 : 0.000122s : 0.03% optimize.opt_b.b_2 : 0.000009s : 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 : 0.000003s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000028s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000018s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000035s : 0.01% optimize.loop_unroll : 0.000487s : 0.11% optimize.opt_after_cconv.c_1 : 0.000031s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 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 : 0.000003s : 0.00% optimize.opt_after_cconv.cse : 0.000024s : 0.01% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000044s : 0.01% optimize.tuple_transform.d_1 : 0.000046s : 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 : 0.000008s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000052s : 0.01% optimize.cse_after_recomputation.cse : 0.000014s : 0.00% optimize.environ_conv : 0.000006s : 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 : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000002s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000014s : 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 : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000004s : 0.00% optimize.overlap_grad_flash_sp : 0.000023s : 0.01% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.233172s : 53.48% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000054s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000031s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000010s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000022s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000001s : 0.00% detach_backward : 0.000005s : 0.00% pipeline_parallel_scheduler : 0.000003s : 0.00% auto_monad_reorder : 0.000038s : 0.01% get_jit_bprop_graph : 0.000003s : 0.00% rewriter_after_jit_bprop_graph : 0.000011s : 0.00% opt_after_jit_grad : 0.000912s : 0.21% validate : 0.000047s : 0.01% Time group info: ------[substitution.] 0.000327 35 1.83% : 0.000006s : 2: substitution.elim_not_effective 3.32% : 0.000011s : 2: substitution.fold_const_symbol 1.86% : 0.000006s : 4: substitution.graph_param_transform 80.87% : 0.000264s : 7: substitution.inline 1.28% : 0.000004s : 4: substitution.j_node_and_user_rematch 1.62% : 0.000005s : 4: substitution.remove_not_recompute_node 1.98% : 0.000006s : 4: substitution.replace_old_param 7.24% : 0.000024s : 8: substitution.tuple_list_get_item_eliminator ------[type_inference.] 0.194776 2 99.11% : 0.193047s : 1: type_inference.infer 0.89% : 0.001729s : 1: type_inference.specialize ------[replace.] 0.000133 15 58.08% : 0.000077s : 7: replace.inline 41.92% : 0.000056s : 8: replace.tuple_list_get_item_eliminator ------[match.] 0.000281 15 92.73% : 0.000260s : 7: match.inline 7.27% : 0.000020s : 8: match.tuple_list_get_item_eliminator ------[predicate.] 0.000273 1733 1.01% : 0.000003s : 20: predicate.accumulaten_eliminater 0.84% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.37% : 0.000001s : 8: predicate.addn_check_dump 1.03% : 0.000003s : 20: predicate.addn_zero_filter 1.03% : 0.000003s : 20: predicate.adjust_all_reduce_mul_add 2.11% : 0.000006s : 28: predicate.arithmetic_simplify 0.99% : 0.000003s : 20: predicate.cast_eliminate 0.39% : 0.000001s : 8: predicate.check_bprop_eliminate 0.36% : 0.000001s : 8: predicate.compare_switch_simplify 0.13% : 0.000000s : 4: predicate.const_output_eliminate 0.40% : 0.000001s : 8: predicate.depend_value_elim 1.02% : 0.000003s : 20: predicate.dict_get_item_const_eliminator 1.07% : 0.000003s : 20: predicate.dict_get_item_eliminator 1.02% : 0.000003s : 20: predicate.dict_set_item_eliminator 0.72% : 0.000002s : 8: predicate.dumpgradient_eliminate 0.40% : 0.000001s : 4: predicate.elim_not_effective 2.17% : 0.000006s : 4: predicate.elim_shapecalc_of_broadcastargs 1.27% : 0.000003s : 24: predicate.environ_add_const_eliminate 1.13% : 0.000003s : 24: predicate.environ_get_add_eliminate 1.13% : 0.000003s : 24: predicate.environ_get_depend_swap 1.56% : 0.000004s : 32: predicate.environ_get_eliminate 1.11% : 0.000003s : 24: predicate.environ_get_set_eliminate 1.71% : 0.000005s : 35: predicate.exchange_switch_depend_value 2.85% : 0.000008s : 35: predicate.float_depend_g_call 0.37% : 0.000001s : 8: predicate.float_environ_get_switch 0.53% : 0.000001s : 12: predicate.float_tuple_getitem_switch 0.14% : 0.000000s : 4: predicate.fold_const_symbol 0.42% : 0.000001s : 8: predicate.get_grad_eliminate 0.21% : 0.000001s : 4: predicate.graph_param_transform 0.39% : 0.000001s : 8: predicate.incorporate_call 0.34% : 0.000001s : 8: predicate.incorporate_call_switch 6.10% : 0.000017s : 79: predicate.inline 0.46% : 0.000001s : 8: predicate.inline_without_move 0.23% : 0.000001s : 8: predicate.j_node_and_user_rematch 0.61% : 0.000002s : 8: predicate.less_batch_normalization 1.77% : 0.000005s : 36: predicate.list_to_tuple_eliminator_ 2.68% : 0.000007s : 56: predicate.load_eliminater 0.71% : 0.000002s : 4: predicate.loop_unroll_after_grad 2.93% : 0.000008s : 58: predicate.loop_unroll_before_grad 1.58% : 0.000004s : 28: predicate.make_slice_get_slice_eliminator 0.37% : 0.000001s : 8: predicate.merge_addn 0.39% : 0.000001s : 8: predicate.micro_step_allgather_replace 0.49% : 0.000001s : 8: predicate.mini_step_allgather_replace 0.94% : 0.000003s : 20: predicate.minmaximum_grad 1.30% : 0.000004s : 4: predicate.mutable_eliminate 0.45% : 0.000001s : 4: predicate.opt_reshape 0.25% : 0.000001s : 4: predicate.parallel_virtual_node 2.17% : 0.000006s : 35: predicate.partial_defer_inline 1.79% : 0.000005s : 32: predicate.partial_eliminate 0.97% : 0.000003s : 20: predicate.print_const_string_wrapper 0.43% : 0.000001s : 8: predicate.reduce_all_const_elim 1.28% : 0.000003s : 20: predicate.reduce_eliminate 2.62% : 0.000007s : 56: predicate.redundant_stop_gradient_eliminater 0.37% : 0.000001s : 8: predicate.remove_not_recompute_node 1.49% : 0.000004s : 36: predicate.replace_applicator 0.38% : 0.000001s : 8: predicate.replace_old_param 0.28% : 0.000001s : 4: predicate.reset_defer_inline 1.06% : 0.000003s : 20: predicate.reshape_eliminate 0.40% : 0.000001s : 8: predicate.row_tensor_add_zeros_like 0.26% : 0.000001s : 4: predicate.row_tensor_eliminate 0.65% : 0.000002s : 8: predicate.same_eliminate 0.33% : 0.000001s : 8: predicate.set_cell_output_no_recompute 0.66% : 0.000002s : 8: predicate.shard_identity_eliminate 0.56% : 0.000002s : 8: predicate.special_op_eliminate 0.46% : 0.000001s : 8: predicate.specialize_transform 0.88% : 0.000002s : 8: predicate.split_environ_get_set_with_tuple_value 0.55% : 0.000002s : 8: predicate.stack_unstack_eliminate 0.25% : 0.000001s : 4: predicate.switch_call_monad_eliminater 1.85% : 0.000005s : 35: predicate.switch_defer_inline 2.15% : 0.000006s : 43: predicate.switch_layer_defer_inline 5.93% : 0.000016s : 105: predicate.switch_simplify 0.98% : 0.000003s : 20: predicate.tile_eliminate 1.13% : 0.000003s : 20: predicate.transpose_eliminate 1.56% : 0.000004s : 28: predicate.tuple_list_convert_item_index_to_positive 1.48% : 0.000004s : 28: predicate.tuple_list_get_item_const_eliminator 1.43% : 0.000004s : 28: predicate.tuple_list_get_item_depend_reorder 3.24% : 0.000009s : 44: predicate.tuple_list_get_item_eliminator 1.43% : 0.000004s : 28: predicate.tuple_list_get_set_item_eliminator 2.00% : 0.000005s : 36: predicate.tuple_list_set_item_eliminator 1.86% : 0.000005s : 36: predicate.tuple_to_list_eliminator_ 2.56% : 0.000007s : 56: predicate.updatestate_pure_node_eliminater 3.21% : 0.000009s : 64: predicate.updatestate_useless_node_eliminater 0.23% : 0.000001s : 4: predicate.value_based_eliminate 0.51% : 0.000001s : 8: predicate.virtual_dataset_eliminate 0.52% : 0.000001s : 8: predicate.virtual_output_eliminate 0.23% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.39% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.001295 18 54.64% : 0.000707s : 9: func_graph_cloner_run.FuncGraphClonerGraph 45.36% : 0.000587s : 9: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 0.690437 192 0.00% : 0.000004s : 1: ForceFp32Comm 0.69% : 0.004798s : 1: add_attr 0.69% : 0.004781s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000056s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.01% : 0.000102s : 1: auto_monad 0.01% : 0.000042s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.07% : 0.000517s : 1: bootstrap 0.01% : 0.000039s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000017s : 1: control_data_broadcast_order 0.00% : 0.000005s : 1: convert_after_rewriter 0.00% : 0.000028s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000008s : 1: detach_backward 0.00% : 0.000009s : 1: environ_conv 0.01% : 0.000088s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000012s : 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% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.07% : 0.000500s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.11% : 0.000778s : 1: mutable_eliminate 0.00% : 0.000007s : 1: offloading_packed_experts 0.00% : 0.000015s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000020s : 1: opt.transform.mutable_eliminate 0.25% : 0.001718s : 78: opt.transform.opt_a 0.00% : 0.000029s : 1: opt.transform.opt_after_cconv 0.00% : 0.000033s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000102s : 28: opt.transform.opt_b 0.01% : 0.000051s : 2: opt.transform.opt_trans_graph 0.01% : 0.000103s : 4: opt.transform.symbol_engine_opt 0.62% : 0.004258s : 1: opt_a 0.02% : 0.000115s : 1: opt_after_cconv 0.13% : 0.000924s : 1: opt_after_jit_grad 0.03% : 0.000222s : 1: opt_b 34.80% : 0.240303s : 1: optimize 0.00% : 0.000022s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 1: order_py_execute_after_rewriter 0.00% : 0.000026s : 1: overlap_grad_flash_sp 0.00% : 0.000004s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000007s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 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% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000063s : 1: pre_auto_parallel 0.00% : 0.000012s : 1: py_interpret_to_execute 0.00% : 0.000031s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.01% : 0.000048s : 1: remove_dup_value 0.12% : 0.000827s : 1: renormalize.infer 0.11% : 0.000733s : 1: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000014s : 1: rewriter_after_jit_bprop_graph 0.00% : 0.000029s : 1: rewriter_after_opt_a 0.05% : 0.000367s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000004s : 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 33.80% : 0.233394s : 1: symbol_engine_optimizer 0.01% : 0.000082s : 1: tuple_transform 28.23% : 0.194901s : 1: type_inference TotalTime = 0.712498, [21] [bootstrap]: 0.00048801 [type_inference]: 0.323903 [event_method]: 0.00024759 [auto_monad]: 0.00020554 [graph_reusing]: 1.263e-05 [inline]: 3.31999e-06 [add_attr]: 0.0045244, [1] [add_attr_with_inline]: 0.00451041, [1] [Cycle 1]: 0.00014829, [2] [tag_attr]: 7.775e-05 [meta_addattr_fg_expand]: 1.717e-05 [parallel-infer-symbol]: 4.24002e-06 [pre_auto_parallel]: 0.00010121 [insert-virtual-dataset]: 2.94001e-06 [parallel-infer-symbol-second]: 9.49978e-07 [dataset_repeat_opt]: 2.68e-06 [pipeline_split]: 1.74e-06 [optimize]: 0.381716, [53] [py_interpret_to_execute]: 9.82999e-06 [rewriter_before_opt_a]: 0.00059841 [opt_a]: 0.191215, [3] [Cycle 1]: 0.185705, [45] [expand_dump_flag]: 6.54001e-06 [switch_simplify]: 0.00022806 [loop_unroll]: 0.00011046 [a_1]: 0.00254121 [with_stream_mark]: 4.392e-05 [recompute_prepare]: 3.154e-05 [updatestate_depend_eliminate]: 1.097e-05 [updatestate_assign_eliminate]: 9.02e-06 [updatestate_loads_eliminate]: 8.62e-06 [parameter_eliminate]: 4.07998e-06 [a_2]: 0.00026798 [accelerated_algorithm]: 2.03e-05 [shard]: 2.14999e-06 [meta_shard_fg_expand]: 7.61001e-06 [shard_inline]: 1.823e-05 [merge_send_recv]: 2.021e-05 [auto_parallel]: 1.63e-05 [parallel]: 2.235e-05 [flash_sp]: 1.416e-05 [merge_comm]: 1.143e-05 [allreduce_fusion]: 9.81998e-06 [matmul_add_comm_reduction]: 3.502e-05 [allreduce_slice_to_reducescatter]: 8.80013e-07 [virtual_shard_identity]: 2.177e-05 [virtual_dataset]: 1.734e-05 [get_grad_eliminate_]: 1.766e-05 [virtual_output]: 1.631e-05 [merge_forward]: 1.09e-05 [cell_reuse_recompute_pass]: 1.45001e-06 [offload_activation]: 1.941e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.454e-05 [merge_recompute_call_nodes]: 1.54e-06 [before_grad]: 3.177e-05 [set_forward_comm_id_for_comm_node_pass]: 1.147e-05 [meta_fg_expand]: 0.00224115 [flash_sp_send_recv_attached]: 4.74e-06 [receive_attached]: 3.45998e-06 [after_resolve]: 8.79e-05 [a_after_grad]: 0.00010569 [renormalize]: 0.178198 [add_forward_monad_depend]: 1.792e-05 [auto_monad_grad]: 7.88001e-06 [auto_monad_eliminator]: 7.7e-05 [cse]: 0.00039326 [a_3]: 0.00053112 [Cycle 2]: 0.00444351, [45] [expand_dump_flag]: 3.7e-06 [switch_simplify]: 5.487e-05 [loop_unroll]: 5.246e-05 [a_1]: 0.00167036 [with_stream_mark]: 3.544e-05 [recompute_prepare]: 1.631e-05 [updatestate_depend_eliminate]: 8.21002e-06 [updatestate_assign_eliminate]: 5.70001e-06 [updatestate_loads_eliminate]: 5.42999e-06 [parameter_eliminate]: 3.43e-06 [a_2]: 0.00014053 [accelerated_algorithm]: 1.268e-05 [shard]: 2.54001e-06 [meta_shard_fg_expand]: 4.15e-06 [shard_inline]: 1.072e-05 [merge_send_recv]: 1.152e-05 [auto_parallel]: 1.232e-05 [parallel]: 9.63002e-06 [flash_sp]: 4.07998e-06 [merge_comm]: 6.29999e-06 [allreduce_fusion]: 6.39999e-06 [matmul_add_comm_reduction]: 1.373e-05 [allreduce_slice_to_reducescatter]: 9.80013e-07 [virtual_shard_identity]: 1.363e-05 [virtual_dataset]: 1.059e-05 [get_grad_eliminate_]: 1.118e-05 [virtual_output]: 1.151e-05 [merge_forward]: 6.06e-06 [cell_reuse_recompute_pass]: 2.84999e-06 [offload_activation]: 1.495e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.327e-05 [merge_recompute_call_nodes]: 1.64e-06 [before_grad]: 1.83e-05 [set_forward_comm_id_for_comm_node_pass]: 7.21999e-06 [meta_fg_expand]: 0.00031199 [flash_sp_send_recv_attached]: 2.11998e-06 [receive_attached]: 2.41998e-06 [after_resolve]: 2.54e-05 [a_after_grad]: 1.875e-05 [renormalize]: 0.00137436 [add_forward_monad_depend]: 7.89002e-06 [auto_monad_grad]: 2.43998e-06 [auto_monad_eliminator]: 2.425e-05 [cse]: 8.594e-05 [a_3]: 8.306e-05 [Cycle 3]: 0.00104162, [45] [expand_dump_flag]: 2.42001e-06 [switch_simplify]: 1.324e-05 [loop_unroll]: 1.114e-05 [a_1]: 0.00027632 [with_stream_mark]: 1.884e-05 [recompute_prepare]: 1.133e-05 [updatestate_depend_eliminate]: 6.14001e-06 [updatestate_assign_eliminate]: 6.04001e-06 [updatestate_loads_eliminate]: 5.27001e-06 [parameter_eliminate]: 1.60001e-06 [a_2]: 0.00013409 [accelerated_algorithm]: 1.138e-05 [shard]: 2.21998e-06 [meta_shard_fg_expand]: 2.78e-06 [shard_inline]: 1.043e-05 [merge_send_recv]: 1.038e-05 [auto_parallel]: 1.136e-05 [parallel]: 8.02003e-06 [flash_sp]: 1.38002e-06 [merge_comm]: 6.06e-06 [allreduce_fusion]: 5.74e-06 [matmul_add_comm_reduction]: 1.015e-05 [allreduce_slice_to_reducescatter]: 6.09987e-07 [virtual_shard_identity]: 1.218e-05 [virtual_dataset]: 1.011e-05 [get_grad_eliminate_]: 1.011e-05 [virtual_output]: 9.76e-06 [merge_forward]: 5.19003e-06 [cell_reuse_recompute_pass]: 2.41998e-06 [offload_activation]: 1.298e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.243e-05 [merge_recompute_call_nodes]: 1.33002e-06 [before_grad]: 1.744e-05 [set_forward_comm_id_for_comm_node_pass]: 5.94999e-06 [meta_fg_expand]: 3.69002e-06 [flash_sp_send_recv_attached]: 1.30001e-06 [receive_attached]: 2.07999e-06 [after_resolve]: 1.505e-05 [a_after_grad]: 1.671e-05 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 1.77001e-06 [auto_monad_grad]: 1.95001e-06 [auto_monad_eliminator]: 1.334e-05 [cse]: 3.289e-05 [a_3]: 6.777e-05 [py_interpret_to_execute_after_opt_a]: 8.42998e-06 [slice_cell_reuse_recomputed_activation]: 2.02001e-06 [rewriter_after_opt_a]: 3.357e-05 [convert_after_rewriter]: 1.32e-06 [order_py_execute_after_rewriter]: 1.24e-06 [mutable_eliminate]: 0.00079118 [opt_b]: 0.00035403, [1] [Cycle 1]: 0.00034389, [7] [b_1]: 0.00022129 [b_2]: 1.463e-05 [updatestate_depend_eliminate]: 1.177e-05 [updatestate_assign_eliminate]: 4.62e-06 [updatestate_loads_eliminate]: 4.76002e-06 [renormalize]: 7.2e-07 [cse]: 4.91e-05 [optimize_parallel_all_gather_comm]: 2.769e-05 [overlap_param_gather]: 2.68e-06 [cconv]: 3.965e-05 [loop_unroll]: 0.00053594 [opt_after_cconv]: 0.00020022, [1] [Cycle 1]: 0.00019296, [7] [c_1]: 6.546e-05 [parameter_eliminate]: 5.00001e-06 [updatestate_depend_eliminate]: 9.92001e-06 [updatestate_assign_eliminate]: 5.03002e-06 [updatestate_loads_eliminate]: 4.90001e-06 [cse]: 6.352e-05 [renormalize]: 5.69999e-07 [remove_dup_value]: 5.775e-05 [tuple_transform]: 0.00011951, [1] [Cycle 1]: 0.0001136, [4] [d_1]: 8.029e-05 [none_parameter_eliminate]: 2.24001e-06 [renormalize]: 3.19997e-07 [switch_simplify]: 1.248e-05 [partial_unused_args_eliminate]: 2.12999e-06 [add_recomputation]: 7.926e-05 [cse_after_recomputation]: 4.07e-05, [1] [Cycle 1]: 3.491e-05, [1] [cse]: 2.911e-05 [environ_conv]: 1.206e-05 [swap_dp_allreduce_reducescatter]: 9.10001e-06 [bias_add_comm_swap]: 3.83001e-06 [label_micro_interleaved_index]: 5.04e-06 [label_fine_grained_interleaved_index]: 2.95002e-06 [merge_cast_opt]: 1.44e-06 [slice_recompute_activation]: 2.32001e-06 [micro_interleaved_order_control]: 2.99999e-06 [assign_add_opt]: 1.55999e-06 [ForceFp32Comm]: 1.05999e-06 [remove_cast_before_assign_add]: 1.14e-06 [full_micro_interleaved_order_control]: 2.37001e-06 [reorder_send_recv_between_fp_bp]: 2.94001e-06 [comm_op_add_attrs]: 1.09e-06 [add_comm_op_reuse_tag]: 9.89996e-07 [interleave_split_concat_branches]: 1.24e-06 [interleave_parallel_branches]: 1.32999e-06 [overlap_opt_shard_in_pipeline]: 1.15001e-06 [overlap_opt_shard_grad_in_pipeline]: 1.89999e-06 [control_data_broadcast_order]: 2.128e-05 [grouped_pairwise_exchange_alltoall]: 2.00002e-06 [offloading_packed_experts]: 6.76e-06 [overlap_recompute_and_grad_model_parallel]: 6.98998e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.42e-06 [overlap_recompute_allgather_and_fa_grad]: 1.38002e-06 [overlap_recompute_comm]: 2.48e-06 [overlap_grad_ring_attention]: 7.37997e-06 [overlap_grad_flash_sp]: 3.051e-05 [begin_end_overlap_inline]: 8.30012e-07 [split_matmul_comm_elemetwise]: 2.29999e-06 [split_layernorm_comm]: 1.74998e-06 [handle_group_info]: 9.39996e-07 [symbol_engine_optimizer]: 0.187188, [1] [Cycle 1]: 0.18718, [6] [build]: 0.186941 [elim_shapecalc]: 5.82e-05 [elim_not_effective]: 4.352e-05 [opt_reshape]: 2.304e-05 [fold_const_symbol]: 3.292e-05 [renormalize]: 1.09e-06 [detach_backward]: 5.14e-06 [pipeline_parallel_scheduler]: 2.17999e-06 [auto_monad_reorder]: 5.106e-05 [get_jit_bprop_graph]: 2.49001e-06 [rewriter_after_jit_bprop_graph]: 8.07e-06 [opt_after_jit_grad]: 0.00090457 [validate]: 7.489e-05 Sums bootstrap : 0.000488s : 0.07% type_inference : 0.323903s : 45.85% event_method : 0.000248s : 0.04% auto_monad : 0.000206s : 0.03% graph_reusing : 0.000013s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000078s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000017s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000101s : 0.01% 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 : 0.000010s : 0.00% optimize.rewriter_before_opt_a : 0.000598s : 0.08% optimize.opt_a.expand_dump_flag : 0.000013s : 0.00% optimize.opt_a.switch_simplify : 0.000296s : 0.04% optimize.opt_a.loop_unroll : 0.000174s : 0.02% optimize.opt_a.a_1 : 0.004488s : 0.64% optimize.opt_a.with_stream_mark : 0.000098s : 0.01% optimize.opt_a.recompute_prepare : 0.000059s : 0.01% optimize.opt_a.updatestate_depend_eliminate : 0.000025s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000021s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000019s : 0.00% optimize.opt_a.parameter_eliminate : 0.000009s : 0.00% optimize.opt_a.a_2 : 0.000543s : 0.08% optimize.opt_a.accelerated_algorithm : 0.000044s : 0.01% optimize.opt_a.shard : 0.000007s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000015s : 0.00% optimize.opt_a.shard_inline : 0.000039s : 0.01% optimize.opt_a.merge_send_recv : 0.000042s : 0.01% optimize.opt_a.auto_parallel : 0.000040s : 0.01% optimize.opt_a.parallel : 0.000040s : 0.01% optimize.opt_a.flash_sp : 0.000020s : 0.00% optimize.opt_a.merge_comm : 0.000024s : 0.00% optimize.opt_a.allreduce_fusion : 0.000022s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000059s : 0.01% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000048s : 0.01% optimize.opt_a.virtual_dataset : 0.000038s : 0.01% optimize.opt_a.get_grad_eliminate_ : 0.000039s : 0.01% optimize.opt_a.virtual_output : 0.000038s : 0.01% optimize.opt_a.merge_forward : 0.000022s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000007s : 0.00% optimize.opt_a.offload_activation : 0.000047s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000080s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000068s : 0.01% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000025s : 0.00% optimize.opt_a.meta_fg_expand : 0.002557s : 0.36% optimize.opt_a.flash_sp_send_recv_attached : 0.000008s : 0.00% optimize.opt_a.receive_attached : 0.000008s : 0.00% optimize.opt_a.after_resolve : 0.000128s : 0.02% optimize.opt_a.a_after_grad : 0.000141s : 0.02% optimize.opt_a.renormalize : 0.179572s : 25.42% optimize.opt_a.add_forward_monad_depend : 0.000028s : 0.00% optimize.opt_a.auto_monad_grad : 0.000012s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000115s : 0.02% optimize.opt_a.cse : 0.000512s : 0.07% optimize.opt_a.a_3 : 0.000682s : 0.10% optimize.py_interpret_to_execute_after_opt_a : 0.000008s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000034s : 0.00% optimize.convert_after_rewriter : 0.000001s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000791s : 0.11% optimize.opt_b.b_1 : 0.000221s : 0.03% optimize.opt_b.b_2 : 0.000015s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000012s : 0.00% optimize.opt_b.updatestate_assign_eliminate : 0.000005s : 0.00% optimize.opt_b.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_b.renormalize : 0.000001s : 0.00% optimize.opt_b.cse : 0.000049s : 0.01% optimize.optimize_parallel_all_gather_comm : 0.000028s : 0.00% optimize.overlap_param_gather : 0.000003s : 0.00% optimize.cconv : 0.000040s : 0.01% optimize.loop_unroll : 0.000536s : 0.08% optimize.opt_after_cconv.c_1 : 0.000065s : 0.01% optimize.opt_after_cconv.parameter_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000010s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000005s : 0.00% optimize.opt_after_cconv.cse : 0.000064s : 0.01% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000058s : 0.01% optimize.tuple_transform.d_1 : 0.000080s : 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 : 0.000012s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000079s : 0.01% optimize.cse_after_recomputation.cse : 0.000029s : 0.00% optimize.environ_conv : 0.000012s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000009s : 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 : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000021s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000007s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000007s : 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 : 0.000007s : 0.00% optimize.overlap_grad_flash_sp : 0.000031s : 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 : 0.186941s : 26.46% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000058s : 0.01% optimize.symbol_engine_optimizer.elim_not_effective : 0.000044s : 0.01% optimize.symbol_engine_optimizer.opt_reshape : 0.000023s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000033s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000001s : 0.00% detach_backward : 0.000005s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000051s : 0.01% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.000905s : 0.13% validate : 0.000075s : 0.01% Time group info: ------[substitution.] 0.001222 215 0.95% : 0.000012s : 5: substitution.elim_not_effective 0.98% : 0.000012s : 14: substitution.float_depend_g_call 0.40% : 0.000005s : 3: substitution.float_tuple_getitem_switch 1.26% : 0.000015s : 5: substitution.fold_const_symbol 0.75% : 0.000009s : 8: substitution.graph_param_transform 0.28% : 0.000003s : 2: substitution.incorporate_call 0.17% : 0.000002s : 2: substitution.incorporate_call_switch 68.51% : 0.000837s : 25: substitution.inline 1.94% : 0.000024s : 2: substitution.inline_without_move 1.14% : 0.000014s : 20: substitution.j_node_and_user_rematch 1.06% : 0.000013s : 7: substitution.minmaximum_grad 0.44% : 0.000005s : 1: substitution.opt_reshape 0.96% : 0.000012s : 14: substitution.partial_eliminate 1.34% : 0.000016s : 20: substitution.remove_not_recompute_node 2.87% : 0.000035s : 9: substitution.replace_applicator 1.22% : 0.000015s : 16: substitution.replace_old_param 1.60% : 0.000019s : 3: substitution.reshape_eliminate 0.25% : 0.000003s : 1: substitution.set_cell_output_no_recompute 1.33% : 0.000016s : 3: substitution.switch_simplify 1.98% : 0.000024s : 7: substitution.tuple_list_convert_item_index_to_positive 0.90% : 0.000011s : 7: substitution.tuple_list_get_item_const_eliminator 1.32% : 0.000016s : 7: substitution.tuple_list_get_item_depend_reorder 7.10% : 0.000087s : 27: substitution.tuple_list_get_item_eliminator 1.25% : 0.000015s : 7: substitution.tuple_list_get_set_item_eliminator ------[type_inference.] 0.323739 2 98.58% : 0.319145s : 1: type_inference.infer 1.42% : 0.004594s : 1: type_inference.specialize ------[replace.] 0.000456 45 54.57% : 0.000249s : 25: replace.inline 12.06% : 0.000055s : 3: replace.switch_simplify 33.37% : 0.000152s : 17: replace.tuple_list_get_item_eliminator ------[match.] 0.000882 45 93.26% : 0.000823s : 25: match.inline 1.47% : 0.000013s : 3: match.switch_simplify 5.27% : 0.000047s : 17: match.tuple_list_get_item_eliminator ------[predicate.] 0.000907 6131 1.16% : 0.000011s : 73: predicate.accumulaten_eliminater 0.38% : 0.000003s : 8: predicate.ad_related_special_op_eliminate 0.44% : 0.000004s : 31: predicate.addn_check_dump 1.08% : 0.000010s : 73: predicate.addn_zero_filter 0.98% : 0.000009s : 73: predicate.adjust_all_reduce_mul_add 2.03% : 0.000018s : 104: predicate.arithmetic_simplify 1.06% : 0.000010s : 73: predicate.cast_eliminate 1.20% : 0.000011s : 67: predicate.check_bprop_eliminate 0.46% : 0.000004s : 31: predicate.compare_switch_simplify 0.08% : 0.000001s : 8: predicate.const_output_eliminate 0.55% : 0.000005s : 31: predicate.depend_value_elim 1.09% : 0.000010s : 73: predicate.dict_get_item_const_eliminator 1.15% : 0.000010s : 73: predicate.dict_get_item_eliminator 1.02% : 0.000009s : 73: predicate.dict_set_item_eliminator 0.37% : 0.000003s : 16: predicate.dumpgradient_eliminate 0.15% : 0.000001s : 8: predicate.elim_not_effective 0.45% : 0.000004s : 8: predicate.elim_shapecalc_of_broadcastargs 1.15% : 0.000010s : 81: predicate.environ_add_const_eliminate 1.08% : 0.000010s : 81: predicate.environ_get_add_eliminate 1.08% : 0.000010s : 81: predicate.environ_get_depend_swap 1.61% : 0.000015s : 112: predicate.environ_get_eliminate 1.10% : 0.000010s : 81: predicate.environ_get_set_eliminate 1.73% : 0.000016s : 115: predicate.exchange_switch_depend_value 2.53% : 0.000023s : 115: predicate.float_depend_g_call 0.50% : 0.000005s : 31: predicate.float_environ_get_switch 0.55% : 0.000005s : 39: predicate.float_tuple_getitem_switch 0.08% : 0.000001s : 8: predicate.fold_const_symbol 0.54% : 0.000005s : 31: predicate.get_grad_eliminate 0.09% : 0.000001s : 8: predicate.graph_param_transform 0.46% : 0.000004s : 31: predicate.incorporate_call 0.42% : 0.000004s : 31: predicate.incorporate_call_switch 5.76% : 0.000052s : 266: predicate.inline 1.36% : 0.000012s : 60: predicate.inline_without_move 0.25% : 0.000002s : 31: predicate.j_node_and_user_rematch 0.62% : 0.000006s : 31: predicate.less_batch_normalization 1.58% : 0.000014s : 106: predicate.list_to_tuple_eliminator_ 2.52% : 0.000023s : 179: predicate.load_eliminater 0.31% : 0.000003s : 8: predicate.loop_unroll_after_grad 2.87% : 0.000026s : 175: predicate.loop_unroll_before_grad 1.42% : 0.000013s : 89: predicate.make_slice_get_slice_eliminator 0.50% : 0.000005s : 31: predicate.merge_addn 0.96% : 0.000009s : 67: predicate.micro_step_allgather_replace 0.95% : 0.000009s : 67: predicate.mini_step_allgather_replace 1.00% : 0.000009s : 73: predicate.minmaximum_grad 0.52% : 0.000005s : 8: predicate.mutable_eliminate 0.22% : 0.000002s : 8: predicate.opt_reshape 0.15% : 0.000001s : 8: predicate.parallel_virtual_node 2.51% : 0.000023s : 115: predicate.partial_defer_inline 1.67% : 0.000015s : 98: predicate.partial_eliminate 1.09% : 0.000010s : 73: predicate.print_const_string_wrapper 0.50% : 0.000005s : 31: predicate.reduce_all_const_elim 1.28% : 0.000012s : 73: predicate.reduce_eliminate 2.55% : 0.000023s : 179: predicate.redundant_stop_gradient_eliminater 0.31% : 0.000003s : 31: predicate.remove_not_recompute_node 1.84% : 0.000017s : 157: predicate.replace_applicator 0.69% : 0.000006s : 60: predicate.replace_old_param 0.12% : 0.000001s : 8: predicate.reset_defer_inline 1.06% : 0.000010s : 73: predicate.reshape_eliminate 1.03% : 0.000009s : 67: predicate.row_tensor_add_zeros_like 0.20% : 0.000002s : 8: predicate.row_tensor_eliminate 1.27% : 0.000012s : 67: predicate.same_eliminate 0.36% : 0.000003s : 31: predicate.set_cell_output_no_recompute 0.61% : 0.000006s : 31: predicate.shard_identity_eliminate 0.30% : 0.000003s : 16: predicate.special_op_eliminate 0.62% : 0.000006s : 31: predicate.specialize_transform 1.24% : 0.000011s : 67: predicate.split_environ_get_set_with_tuple_value 1.13% : 0.000010s : 60: predicate.stack_unstack_eliminate 0.15% : 0.000001s : 8: predicate.switch_call_monad_eliminater 1.91% : 0.000017s : 115: predicate.switch_defer_inline 2.79% : 0.000025s : 182: predicate.switch_layer_defer_inline 5.94% : 0.000054s : 335: predicate.switch_simplify 1.06% : 0.000010s : 73: predicate.tile_eliminate 1.03% : 0.000009s : 73: predicate.transpose_eliminate 1.47% : 0.000013s : 89: predicate.tuple_list_convert_item_index_to_positive 1.54% : 0.000014s : 89: predicate.tuple_list_get_item_const_eliminator 1.35% : 0.000012s : 89: predicate.tuple_list_get_item_depend_reorder 2.73% : 0.000025s : 137: predicate.tuple_list_get_item_eliminator 1.38% : 0.000012s : 89: predicate.tuple_list_get_set_item_eliminator 2.03% : 0.000018s : 120: predicate.tuple_list_set_item_eliminator 1.59% : 0.000014s : 106: predicate.tuple_to_list_eliminator_ 2.62% : 0.000024s : 179: predicate.updatestate_pure_node_eliminater 3.03% : 0.000027s : 210: predicate.updatestate_useless_node_eliminater 0.18% : 0.000002s : 8: predicate.value_based_eliminate 0.51% : 0.000005s : 31: predicate.virtual_dataset_eliminate 0.51% : 0.000005s : 31: predicate.virtual_output_eliminate 0.12% : 0.000001s : 8: predicate.virtual_view_grad_eliminate 0.15% : 0.000001s : 8: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.004344 54 61.93% : 0.002690s : 25: func_graph_cloner_run.FuncGraphClonerGraph 38.07% : 0.001654s : 29: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.285423 233 0.00% : 0.000004s : 1: ForceFp32Comm 0.35% : 0.004532s : 1: add_attr 0.35% : 0.004515s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.01% : 0.000084s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.02% : 0.000217s : 1: auto_monad 0.00% : 0.000057s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.04% : 0.000526s : 1: bootstrap 0.00% : 0.000044s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000026s : 1: control_data_broadcast_order 0.00% : 0.000004s : 1: convert_after_rewriter 0.00% : 0.000044s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000009s : 1: detach_backward 0.00% : 0.000015s : 1: environ_conv 0.02% : 0.000265s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000007s : 1: get_jit_bprop_graph 0.00% : 0.000018s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000005s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000007s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.04% : 0.000547s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.06% : 0.000804s : 1: mutable_eliminate 0.00% : 0.000010s : 1: offloading_packed_experts 0.00% : 0.000023s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000028s : 1: opt.transform.mutable_eliminate 0.52% : 0.006734s : 117: opt.transform.opt_a 0.00% : 0.000064s : 1: opt.transform.opt_after_cconv 0.00% : 0.000050s : 1: opt.transform.opt_after_jit_grad 0.02% : 0.000205s : 28: opt.transform.opt_b 0.01% : 0.000090s : 2: opt.transform.opt_trans_graph 0.01% : 0.000144s : 4: opt.transform.symbol_engine_opt 14.88% : 0.191219s : 1: opt_a 0.02% : 0.000204s : 1: opt_after_cconv 0.07% : 0.000917s : 1: opt_after_jit_grad 0.03% : 0.000358s : 1: opt_b 29.70% : 0.381724s : 1: optimize 0.00% : 0.000032s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 1: order_py_execute_after_rewriter 0.00% : 0.000035s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 1: overlap_grad_matmul_and_grad_allreduce 0.00% : 0.000010s : 1: overlap_grad_ring_attention 0.00% : 0.000005s : 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% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000010s : 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% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.01% : 0.000108s : 1: pre_auto_parallel 0.00% : 0.000015s : 1: py_interpret_to_execute 0.00% : 0.000012s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000062s : 1: remove_dup_value 13.74% : 0.176620s : 2: renormalize.infer 0.23% : 0.002927s : 2: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.00% : 0.000037s : 1: rewriter_after_opt_a 0.05% : 0.000611s : 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% : 0.000012s : 1: swap_dp_allreduce_reducescatter 14.56% : 0.187194s : 1: symbol_engine_optimizer 0.01% : 0.000123s : 1: tuple_transform 25.20% : 0.323935s : 1: type_inference . [hook] pytest_runtest_teardown:test_repeat_interleave_forward_backward_tensor[True-None-KBK] tests/st/mint/test_repeat_interleave.py::test_repeat_interleave_forward_backward_tensor[True-None-KBK],max_mem:20.0M . [hook] pytest_runtest_teardown:test_repeat_interleave_forward_backward_tensor[False-0-pynative] tests/st/mint/test_repeat_interleave.py::test_repeat_interleave_forward_backward_tensor[False-0-pynative],max_mem:20.0M TotalTime = 2.12521, [21] [bootstrap]: 0.00084733 [type_inference]: 1.9229 [event_method]: 0.00062072 [auto_monad]: 0.00010363 [graph_reusing]: 8e-06 [inline]: 2.55002e-06 [add_attr]: 0.00484971, [1] [add_attr_with_inline]: 0.00483011, [1] [Cycle 1]: 9.906e-05, [2] [tag_attr]: 4.29e-05 [meta_addattr_fg_expand]: 7.75e-06 [parallel-infer-symbol]: 3.5e-06 [pre_auto_parallel]: 6.031e-05 [insert-virtual-dataset]: 4.06001e-06 [parallel-infer-symbol-second]: 8.89995e-07 [dataset_repeat_opt]: 1.92999e-06 [pipeline_split]: 2.84001e-06 [optimize]: 0.19448, [53] [py_interpret_to_execute]: 8.62998e-06 [rewriter_before_opt_a]: 0.00042445 [opt_a]: 0.16625, [2] [Cycle 1]: 0.165558, [45] [expand_dump_flag]: 4.58001e-06 [switch_simplify]: 6.672e-05 [loop_unroll]: 4.955e-05 [a_1]: 0.00094946 [with_stream_mark]: 2.58e-05 [recompute_prepare]: 1.046e-05 [updatestate_depend_eliminate]: 5.34998e-06 [updatestate_assign_eliminate]: 3.45e-06 [updatestate_loads_eliminate]: 3.06999e-06 [parameter_eliminate]: 1.84e-06 [a_2]: 8.536e-05 [accelerated_algorithm]: 7.05998e-06 [shard]: 1.98997e-06 [meta_shard_fg_expand]: 2.47001e-06 [shard_inline]: 7.10998e-06 [merge_send_recv]: 9.00999e-06 [auto_parallel]: 9.05001e-06 [parallel]: 3.281e-05 [flash_sp]: 9.67999e-06 [merge_comm]: 4.12998e-06 [allreduce_fusion]: 3.96001e-06 [matmul_add_comm_reduction]: 1.078e-05 [allreduce_slice_to_reducescatter]: 1.45999e-06 [virtual_shard_identity]: 9.41e-06 [virtual_dataset]: 7.11001e-06 [get_grad_eliminate_]: 6.16e-06 [virtual_output]: 6.69999e-06 [merge_forward]: 4.03001e-06 [cell_reuse_recompute_pass]: 1.29e-06 [offload_activation]: 1.103e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.789e-05 [merge_recompute_call_nodes]: 1.50999e-06 [before_grad]: 1.158e-05 [set_forward_comm_id_for_comm_node_pass]: 4.38999e-06 [meta_fg_expand]: 3.04999e-06 [flash_sp_send_recv_attached]: 3.23e-06 [receive_attached]: 2.79999e-06 [after_resolve]: 1.128e-05 [a_after_grad]: 9.46998e-06 [renormalize]: 0.163687 [add_forward_monad_depend]: 1.327e-05 [auto_monad_grad]: 2.42001e-06 [auto_monad_eliminator]: 2.592e-05 [cse]: 4.184e-05 [a_3]: 6.831e-05 [Cycle 2]: 0.00067583, [45] [expand_dump_flag]: 2.61e-06 [switch_simplify]: 1.04e-05 [loop_unroll]: 6.30002e-06 [a_1]: 0.00011184 [with_stream_mark]: 2.072e-05 [recompute_prepare]: 6.11998e-06 [updatestate_depend_eliminate]: 4.63001e-06 [updatestate_assign_eliminate]: 3.26999e-06 [updatestate_loads_eliminate]: 3.81999e-06 [parameter_eliminate]: 2.48e-06 [a_2]: 7.229e-05 [accelerated_algorithm]: 6.89999e-06 [shard]: 2.37999e-06 [meta_shard_fg_expand]: 2.24999e-06 [shard_inline]: 5.91e-06 [merge_send_recv]: 8.80999e-06 [auto_parallel]: 1.071e-05 [parallel]: 8.75999e-06 [flash_sp]: 4.18999e-06 [merge_comm]: 3.51001e-06 [allreduce_fusion]: 3.6e-06 [matmul_add_comm_reduction]: 1.013e-05 [allreduce_slice_to_reducescatter]: 1.29e-06 [virtual_shard_identity]: 6.75998e-06 [virtual_dataset]: 5.88998e-06 [get_grad_eliminate_]: 5.32999e-06 [virtual_output]: 5.52001e-06 [merge_forward]: 4.04002e-06 [cell_reuse_recompute_pass]: 3.96001e-06 [offload_activation]: 1.017e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.731e-05 [merge_recompute_call_nodes]: 1.45999e-06 [before_grad]: 1.023e-05 [set_forward_comm_id_for_comm_node_pass]: 3.68e-06 [meta_fg_expand]: 2.96001e-06 [flash_sp_send_recv_attached]: 1.77999e-06 [receive_attached]: 2.19001e-06 [after_resolve]: 1.073e-05 [a_after_grad]: 9.75002e-06 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 2.09e-06 [auto_monad_grad]: 1.14998e-06 [auto_monad_eliminator]: 7.25e-06 [cse]: 1.838e-05 [a_3]: 3.546e-05 [py_interpret_to_execute_after_opt_a]: 1.151e-05 [slice_cell_reuse_recomputed_activation]: 2.24999e-06 [rewriter_after_opt_a]: 2.491e-05 [convert_after_rewriter]: 1.65001e-06 [order_py_execute_after_rewriter]: 1.20001e-06 [mutable_eliminate]: 0.00082354 [opt_b]: 0.00933151, [1] [Cycle 1]: 0.00037806, [7] [b_1]: 0.00020266 [b_2]: 1.28e-05 [updatestate_depend_eliminate]: 1.362e-05 [updatestate_assign_eliminate]: 3.44001e-06 [updatestate_loads_eliminate]: 3.24001e-06 [renormalize]: 8.80013e-07 [cse]: 5.344e-05 [optimize_parallel_all_gather_comm]: 2.79e-05 [overlap_param_gather]: 2.09e-06 [cconv]: 4.062e-05 [loop_unroll]: 0.00078949 [opt_after_cconv]: 0.0001191, [1] [Cycle 1]: 0.00011017, [7] [c_1]: 2.877e-05 [parameter_eliminate]: 7.15e-06 [updatestate_depend_eliminate]: 7.1e-06 [updatestate_assign_eliminate]: 2.50002e-06 [updatestate_loads_eliminate]: 2.41e-06 [cse]: 2.692e-05 [renormalize]: 6.10016e-07 [remove_dup_value]: 2.052e-05 [tuple_transform]: 0.0158223, [1] [Cycle 1]: 7.206e-05, [4] [d_1]: 4.529e-05 [none_parameter_eliminate]: 1.77001e-06 [renormalize]: 1.60013e-07 [switch_simplify]: 7.00002e-06 [partial_unused_args_eliminate]: 1.079e-05 [add_recomputation]: 0.00010137 [cse_after_recomputation]: 8.044e-05, [1] [Cycle 1]: 6.916e-05, [1] [cse]: 5.351e-05 [environ_conv]: 1.532e-05 [swap_dp_allreduce_reducescatter]: 6.26e-06 [bias_add_comm_swap]: 3.81001e-06 [label_micro_interleaved_index]: 8.38001e-06 [label_fine_grained_interleaved_index]: 2.66999e-06 [merge_cast_opt]: 1.61998e-06 [slice_recompute_activation]: 2.22001e-06 [micro_interleaved_order_control]: 2.84001e-06 [assign_add_opt]: 1.81003e-06 [ForceFp32Comm]: 9.00007e-07 [remove_cast_before_assign_add]: 1.29e-06 [full_micro_interleaved_order_control]: 2.11998e-06 [reorder_send_recv_between_fp_bp]: 3.01999e-06 [comm_op_add_attrs]: 1.35999e-06 [add_comm_op_reuse_tag]: 1.35001e-06 [interleave_split_concat_branches]: 1.20001e-06 [interleave_parallel_branches]: 1.05999e-06 [overlap_opt_shard_in_pipeline]: 1.37999e-06 [overlap_opt_shard_grad_in_pipeline]: 2.27001e-06 [control_data_broadcast_order]: 2.014e-05 [grouped_pairwise_exchange_alltoall]: 1.40999e-06 [offloading_packed_experts]: 4.12998e-06 [overlap_recompute_and_grad_model_parallel]: 4.58999e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.20999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.37999e-06 [overlap_recompute_comm]: 2.43e-06 [overlap_grad_ring_attention]: 4.43999e-06 [overlap_grad_flash_sp]: 2.401e-05 [begin_end_overlap_inline]: 5.19998e-07 [split_matmul_comm_elemetwise]: 2.71e-06 [split_layernorm_comm]: 1.69e-06 [handle_group_info]: 1.25999e-06 [symbol_engine_optimizer]: 0.00011748, [1] [Cycle 1]: 0.00011055, [6] [build]: 4.18999e-06 [elim_shapecalc]: 2.772e-05 [elim_not_effective]: 2.377e-05 [opt_reshape]: 1.034e-05 [fold_const_symbol]: 1.06e-05 [renormalize]: 4.10015e-07 [detach_backward]: 2.47001e-06 [pipeline_parallel_scheduler]: 3.17002e-06 [auto_monad_reorder]: 2.326e-05 [get_jit_bprop_graph]: 2.29999e-06 [rewriter_after_jit_bprop_graph]: 6.98998e-06 [opt_after_jit_grad]: 0.00088748 [validate]: 0.00016526 Sums bootstrap : 0.000847s : 0.04% type_inference : 1.922902s : 91.81% event_method : 0.000621s : 0.03% auto_monad : 0.000104s : 0.00% graph_reusing : 0.000008s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000043s : 0.00% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000008s : 0.00% parallel-infer-symbol : 0.000003s : 0.00% pre_auto_parallel : 0.000060s : 0.00% insert-virtual-dataset : 0.000004s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000003s : 0.00% optimize.py_interpret_to_execute : 0.000009s : 0.00% optimize.rewriter_before_opt_a : 0.000424s : 0.02% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000077s : 0.00% optimize.opt_a.loop_unroll : 0.000056s : 0.00% optimize.opt_a.a_1 : 0.001061s : 0.05% optimize.opt_a.with_stream_mark : 0.000047s : 0.00% optimize.opt_a.recompute_prepare : 0.000017s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000010s : 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 : 0.000158s : 0.01% optimize.opt_a.accelerated_algorithm : 0.000014s : 0.00% optimize.opt_a.shard : 0.000004s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000005s : 0.00% optimize.opt_a.shard_inline : 0.000013s : 0.00% optimize.opt_a.merge_send_recv : 0.000018s : 0.00% optimize.opt_a.auto_parallel : 0.000020s : 0.00% optimize.opt_a.parallel : 0.000042s : 0.00% optimize.opt_a.flash_sp : 0.000014s : 0.00% optimize.opt_a.merge_comm : 0.000008s : 0.00% optimize.opt_a.allreduce_fusion : 0.000008s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000021s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000003s : 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_ : 0.000011s : 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 : 0.000005s : 0.00% optimize.opt_a.offload_activation : 0.000021s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000035s : 0.00% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000022s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.00% 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 : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000022s : 0.00% optimize.opt_a.a_after_grad : 0.000019s : 0.00% optimize.opt_a.renormalize : 0.163687s : 7.82% optimize.opt_a.add_forward_monad_depend : 0.000015s : 0.00% optimize.opt_a.auto_monad_grad : 0.000004s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000033s : 0.00% optimize.opt_a.cse : 0.000060s : 0.00% optimize.opt_a.a_3 : 0.000104s : 0.00% optimize.py_interpret_to_execute_after_opt_a : 0.000012s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000025s : 0.00% optimize.convert_after_rewriter : 0.000002s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000824s : 0.04% optimize.opt_b.b_1 : 0.000203s : 0.01% optimize.opt_b.b_2 : 0.000013s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000014s : 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 : 0.000053s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000028s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000041s : 0.00% optimize.loop_unroll : 0.000789s : 0.04% optimize.opt_after_cconv.c_1 : 0.000029s : 0.00% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 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 : 0.000027s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000021s : 0.00% optimize.tuple_transform.d_1 : 0.000045s : 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 : 0.000011s : 0.00% optimize.add_recomputation : 0.000101s : 0.00% optimize.cse_after_recomputation.cse : 0.000054s : 0.00% optimize.environ_conv : 0.000015s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000008s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000002s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000020s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000001s : 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 : 0.000024s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000001s : 0.00% optimize.symbol_engine_optimizer.build : 0.000004s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000028s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000024s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000010s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000011s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000003s : 0.00% auto_monad_reorder : 0.000023s : 0.00% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.00% opt_after_jit_grad : 0.000887s : 0.04% validate : 0.000165s : 0.01% Time group info: ------[substitution.] 0.000297 28 0.85% : 0.000003s : 2: substitution.elim_not_effective 0.58% : 0.000002s : 2: substitution.fold_const_symbol 1.95% : 0.000006s : 3: substitution.graph_param_transform 84.31% : 0.000250s : 7: substitution.inline 1.63% : 0.000005s : 4: substitution.j_node_and_user_rematch 2.12% : 0.000006s : 4: substitution.remove_not_recompute_node 1.60% : 0.000005s : 2: substitution.replace_old_param 6.96% : 0.000021s : 4: substitution.tuple_list_get_item_eliminator ------[type_inference.] 1.922767 2 99.80% : 1.918837s : 1: type_inference.infer 0.20% : 0.003931s : 1: type_inference.specialize ------[replace.] 0.000119 11 68.43% : 0.000081s : 7: replace.inline 31.57% : 0.000037s : 4: replace.tuple_list_get_item_eliminator ------[match.] 0.000264 11 93.17% : 0.000246s : 7: match.inline 6.83% : 0.000018s : 4: match.tuple_list_get_item_eliminator ------[predicate.] 0.000248 1305 1.03% : 0.000003s : 15: predicate.accumulaten_eliminater 1.00% : 0.000002s : 3: predicate.ad_related_special_op_eliminate 0.40% : 0.000001s : 6: predicate.addn_check_dump 0.92% : 0.000002s : 15: predicate.addn_zero_filter 1.00% : 0.000002s : 15: predicate.adjust_all_reduce_mul_add 2.07% : 0.000005s : 21: predicate.arithmetic_simplify 1.03% : 0.000003s : 15: predicate.cast_eliminate 0.36% : 0.000001s : 6: predicate.check_bprop_eliminate 0.49% : 0.000001s : 6: predicate.compare_switch_simplify 0.13% : 0.000000s : 3: predicate.const_output_eliminate 0.44% : 0.000001s : 6: predicate.depend_value_elim 0.90% : 0.000002s : 15: predicate.dict_get_item_const_eliminator 1.05% : 0.000003s : 15: predicate.dict_get_item_eliminator 0.91% : 0.000002s : 15: predicate.dict_set_item_eliminator 1.25% : 0.000003s : 6: predicate.dumpgradient_eliminate 0.36% : 0.000001s : 3: predicate.elim_not_effective 0.85% : 0.000002s : 3: predicate.elim_shapecalc_of_broadcastargs 1.21% : 0.000003s : 18: predicate.environ_add_const_eliminate 1.01% : 0.000003s : 18: predicate.environ_get_add_eliminate 1.14% : 0.000003s : 18: predicate.environ_get_depend_swap 1.65% : 0.000004s : 24: predicate.environ_get_eliminate 1.03% : 0.000003s : 18: predicate.environ_get_set_eliminate 1.52% : 0.000004s : 26: predicate.exchange_switch_depend_value 2.44% : 0.000006s : 26: predicate.float_depend_g_call 0.40% : 0.000001s : 6: predicate.float_environ_get_switch 0.50% : 0.000001s : 9: predicate.float_tuple_getitem_switch 0.12% : 0.000000s : 3: predicate.fold_const_symbol 0.48% : 0.000001s : 6: predicate.get_grad_eliminate 0.23% : 0.000001s : 3: predicate.graph_param_transform 0.44% : 0.000001s : 6: predicate.incorporate_call 0.31% : 0.000001s : 6: predicate.incorporate_call_switch 5.98% : 0.000015s : 59: predicate.inline 0.58% : 0.000001s : 6: predicate.inline_without_move 0.19% : 0.000000s : 6: predicate.j_node_and_user_rematch 0.60% : 0.000001s : 6: predicate.less_batch_normalization 1.79% : 0.000004s : 25: predicate.list_to_tuple_eliminator_ 2.52% : 0.000006s : 40: predicate.load_eliminater 0.98% : 0.000002s : 3: predicate.loop_unroll_after_grad 3.13% : 0.000008s : 56: predicate.loop_unroll_before_grad 1.71% : 0.000004s : 21: predicate.make_slice_get_slice_eliminator 0.40% : 0.000001s : 6: predicate.merge_addn 0.39% : 0.000001s : 6: predicate.micro_step_allgather_replace 0.38% : 0.000001s : 6: predicate.mini_step_allgather_replace 0.89% : 0.000002s : 15: predicate.minmaximum_grad 1.06% : 0.000003s : 3: predicate.mutable_eliminate 0.31% : 0.000001s : 3: predicate.opt_reshape 0.36% : 0.000001s : 3: predicate.parallel_virtual_node 2.00% : 0.000005s : 26: predicate.partial_defer_inline 1.37% : 0.000003s : 22: predicate.partial_eliminate 0.88% : 0.000002s : 15: predicate.print_const_string_wrapper 0.44% : 0.000001s : 6: predicate.reduce_all_const_elim 1.26% : 0.000003s : 15: predicate.reduce_eliminate 2.33% : 0.000006s : 40: predicate.redundant_stop_gradient_eliminater 0.35% : 0.000001s : 6: predicate.remove_not_recompute_node 1.25% : 0.000003s : 25: predicate.replace_applicator 0.44% : 0.000001s : 6: predicate.replace_old_param 0.45% : 0.000001s : 3: predicate.reset_defer_inline 0.92% : 0.000002s : 15: predicate.reshape_eliminate 0.55% : 0.000001s : 6: predicate.row_tensor_add_zeros_like 0.55% : 0.000001s : 3: predicate.row_tensor_eliminate 0.68% : 0.000002s : 6: predicate.same_eliminate 0.27% : 0.000001s : 6: predicate.set_cell_output_no_recompute 0.56% : 0.000001s : 6: predicate.shard_identity_eliminate 0.86% : 0.000002s : 6: predicate.special_op_eliminate 0.55% : 0.000001s : 6: predicate.specialize_transform 0.89% : 0.000002s : 6: predicate.split_environ_get_set_with_tuple_value 0.79% : 0.000002s : 6: predicate.stack_unstack_eliminate 0.29% : 0.000001s : 3: predicate.switch_call_monad_eliminater 1.66% : 0.000004s : 26: predicate.switch_defer_inline 2.02% : 0.000005s : 32: predicate.switch_layer_defer_inline 5.48% : 0.000014s : 91: predicate.switch_simplify 1.08% : 0.000003s : 15: predicate.tile_eliminate 1.02% : 0.000003s : 15: predicate.transpose_eliminate 1.69% : 0.000004s : 21: predicate.tuple_list_convert_item_index_to_positive 1.42% : 0.000004s : 21: predicate.tuple_list_get_item_const_eliminator 1.45% : 0.000004s : 21: predicate.tuple_list_get_item_depend_reorder 3.44% : 0.000009s : 31: predicate.tuple_list_get_item_eliminator 1.38% : 0.000003s : 21: predicate.tuple_list_get_set_item_eliminator 2.36% : 0.000006s : 27: predicate.tuple_list_set_item_eliminator 1.82% : 0.000005s : 25: predicate.tuple_to_list_eliminator_ 2.25% : 0.000006s : 40: predicate.updatestate_pure_node_eliminater 2.74% : 0.000007s : 46: predicate.updatestate_useless_node_eliminater 0.29% : 0.000001s : 3: predicate.value_based_eliminate 0.47% : 0.000001s : 6: predicate.virtual_dataset_eliminate 0.46% : 0.000001s : 6: predicate.virtual_output_eliminate 0.21% : 0.000001s : 3: predicate.virtual_view_grad_eliminate 3.13% : 0.000008s : 3: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.057459 31 70.02% : 0.040231s : 22: func_graph_cloner_run.FuncGraphClonerGraph 29.98% : 0.017228s : 9: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 2.489819 192 0.00% : 0.000005s : 1: ForceFp32Comm 0.20% : 0.004856s : 1: add_attr 0.19% : 0.004834s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.00% : 0.000107s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.00% : 0.000114s : 1: auto_monad 0.00% : 0.000027s : 1: auto_monad_reorder 0.00% : 0.000005s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.04% : 0.000909s : 1: bootstrap 0.00% : 0.000044s : 1: cconv 0.00% : 0.000006s : 1: comm_op_add_attrs 0.00% : 0.000023s : 1: control_data_broadcast_order 0.00% : 0.000004s : 1: convert_after_rewriter 0.00% : 0.000084s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000022s : 1: environ_conv 0.03% : 0.000640s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000012s : 1: graph_reusing 0.00% : 0.000007s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000008s : 1: insert-virtual-dataset 0.00% : 0.000005s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000013s : 1: label_micro_interleaved_index 0.03% : 0.000800s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.03% : 0.000837s : 1: mutable_eliminate 0.00% : 0.000007s : 1: offloading_packed_experts 0.00% : 0.000020s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000022s : 1: opt.transform.mutable_eliminate 0.06% : 0.001549s : 78: opt.transform.opt_a 0.00% : 0.000027s : 1: opt.transform.opt_after_cconv 0.00% : 0.000033s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000150s : 28: opt.transform.opt_b 0.00% : 0.000050s : 2: opt.transform.opt_trans_graph 0.00% : 0.000066s : 4: opt.transform.symbol_engine_opt 6.68% : 0.166255s : 1: opt_a 0.00% : 0.000122s : 1: opt_after_cconv 0.04% : 0.000906s : 1: opt_after_jit_grad 0.38% : 0.009338s : 1: opt_b 7.81% : 0.194487s : 1: optimize 0.00% : 0.000031s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 1: order_py_execute_after_rewriter 0.00% : 0.000028s : 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% : 0.000006s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000004s : 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% : 0.000007s : 1: parallel-infer-symbol 0.00% : 0.000004s : 1: parallel-infer-symbol-second 0.00% : 0.000022s : 1: partial_unused_args_eliminate 0.00% : 0.000006s : 1: pipeline_parallel_scheduler 0.00% : 0.000006s : 1: pipeline_split 0.00% : 0.000065s : 1: pre_auto_parallel 0.00% : 0.000012s : 1: py_interpret_to_execute 0.00% : 0.000015s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000025s : 1: remove_dup_value 5.87% : 0.146188s : 1: renormalize.infer 0.70% : 0.017479s : 1: renormalize.specialize 0.00% : 0.000007s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000011s : 1: rewriter_after_jit_bprop_graph 0.00% : 0.000028s : 1: rewriter_after_opt_a 0.02% : 0.000433s : 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% : 0.000004s : 1: split_layernorm_comm 0.00% : 0.000006s : 1: split_matmul_comm_elemetwise 0.00% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.00% : 0.000120s : 1: symbol_engine_optimizer 0.64% : 0.015864s : 1: tuple_transform 77.23% : 1.922927s : 1: type_inference TotalTime = 2.25848, [21] [bootstrap]: 0.00053474 [type_inference]: 1.88321 [event_method]: 0.00076974 [auto_monad]: 0.00021479 [graph_reusing]: 1.145e-05 [inline]: 3.11001e-06 [add_attr]: 0.0391667, [1] [add_attr_with_inline]: 0.039151, [1] [Cycle 1]: 0.00015615, [2] [tag_attr]: 8.909e-05 [meta_addattr_fg_expand]: 1.345e-05 [parallel-infer-symbol]: 4.37998e-06 [pre_auto_parallel]: 8.929e-05 [insert-virtual-dataset]: 2.93e-06 [parallel-infer-symbol-second]: 2.04999e-06 [dataset_repeat_opt]: 2.46998e-06 [pipeline_split]: 1.98002e-06 [optimize]: 0.333244, [53] [py_interpret_to_execute]: 1.277e-05 [rewriter_before_opt_a]: 0.00067191 [opt_a]: 0.295694, [3] [Cycle 1]: 0.264314, [45] [expand_dump_flag]: 7.75e-06 [switch_simplify]: 0.00021009 [loop_unroll]: 9.12e-05 [a_1]: 0.0302536 [with_stream_mark]: 4.498e-05 [recompute_prepare]: 3.455e-05 [updatestate_depend_eliminate]: 1.116e-05 [updatestate_assign_eliminate]: 7.58001e-06 [updatestate_loads_eliminate]: 7.03e-06 [parameter_eliminate]: 4.32e-06 [a_2]: 0.00024516 [accelerated_algorithm]: 1.643e-05 [shard]: 1.77999e-06 [meta_shard_fg_expand]: 8.88002e-06 [shard_inline]: 1.537e-05 [merge_send_recv]: 2.034e-05 [auto_parallel]: 1.754e-05 [parallel]: 2.274e-05 [flash_sp]: 1.472e-05 [merge_comm]: 1.018e-05 [allreduce_fusion]: 8.79998e-06 [matmul_add_comm_reduction]: 3.395e-05 [allreduce_slice_to_reducescatter]: 1.02e-06 [virtual_shard_identity]: 2.066e-05 [virtual_dataset]: 1.814e-05 [get_grad_eliminate_]: 1.669e-05 [virtual_output]: 1.745e-05 [merge_forward]: 1.052e-05 [cell_reuse_recompute_pass]: 2.53e-06 [offload_activation]: 1.805e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.457e-05 [merge_recompute_call_nodes]: 1.84e-06 [before_grad]: 3.076e-05 [set_forward_comm_id_for_comm_node_pass]: 1.018e-05 [meta_fg_expand]: 0.00248523 [flash_sp_send_recv_attached]: 5.64e-06 [receive_attached]: 2.49001e-06 [after_resolve]: 9.577e-05 [a_after_grad]: 0.00011151 [renormalize]: 0.212437 [add_forward_monad_depend]: 2.219e-05 [auto_monad_grad]: 9.97999e-06 [auto_monad_eliminator]: 7.457e-05 [cse]: 0.00059363 [a_3]: 0.0166903 [Cycle 2]: 0.0306686, [45] [expand_dump_flag]: 5.66e-06 [switch_simplify]: 5.505e-05 [loop_unroll]: 4.389e-05 [a_1]: 0.0123312 [with_stream_mark]: 3.719e-05 [recompute_prepare]: 1.712e-05 [updatestate_depend_eliminate]: 5.19e-06 [updatestate_assign_eliminate]: 3.39001e-06 [updatestate_loads_eliminate]: 3.2e-06 [parameter_eliminate]: 2.30002e-06 [a_2]: 7.134e-05 [accelerated_algorithm]: 6.27001e-06 [shard]: 2.32001e-06 [meta_shard_fg_expand]: 3.84002e-06 [shard_inline]: 8.21002e-06 [merge_send_recv]: 1.103e-05 [auto_parallel]: 1.036e-05 [parallel]: 1.002e-05 [flash_sp]: 4.20999e-06 [merge_comm]: 4.17e-06 [allreduce_fusion]: 3.3e-06 [matmul_add_comm_reduction]: 1.249e-05 [allreduce_slice_to_reducescatter]: 9.29984e-07 [virtual_shard_identity]: 8.00999e-06 [virtual_dataset]: 5.69999e-06 [get_grad_eliminate_]: 5.30001e-06 [virtual_output]: 6.70998e-06 [merge_forward]: 4.32e-06 [cell_reuse_recompute_pass]: 2.01003e-06 [offload_activation]: 1.199e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.656e-05 [merge_recompute_call_nodes]: 1.52999e-06 [before_grad]: 1.111e-05 [set_forward_comm_id_for_comm_node_pass]: 3.91001e-06 [meta_fg_expand]: 0.00173862 [flash_sp_send_recv_attached]: 4.25999e-06 [receive_attached]: 3.6e-06 [after_resolve]: 3.058e-05 [a_after_grad]: 1.027e-05 [renormalize]: 0.0155829 [add_forward_monad_depend]: 1.327e-05 [auto_monad_grad]: 2.32999e-06 [auto_monad_eliminator]: 2.604e-05 [cse]: 4.431e-05 [a_3]: 5.892e-05 [Cycle 3]: 0.00068245, [45] [expand_dump_flag]: 2.36e-06 [switch_simplify]: 8.75999e-06 [loop_unroll]: 6.44001e-06 [a_1]: 9.499e-05 [with_stream_mark]: 1.831e-05 [recompute_prepare]: 6.53003e-06 [updatestate_depend_eliminate]: 4.25e-06 [updatestate_assign_eliminate]: 3.41001e-06 [updatestate_loads_eliminate]: 3.04999e-06 [parameter_eliminate]: 2.29001e-06 [a_2]: 6.322e-05 [accelerated_algorithm]: 5.86e-06 [shard]: 2.44999e-06 [meta_shard_fg_expand]: 2.19999e-06 [shard_inline]: 5.27001e-06 [merge_send_recv]: 8.73001e-06 [auto_parallel]: 1.024e-05 [parallel]: 9.08002e-06 [flash_sp]: 1.40999e-06 [merge_comm]: 3.88001e-06 [allreduce_fusion]: 3.73001e-06 [matmul_add_comm_reduction]: 9.01002e-06 [allreduce_slice_to_reducescatter]: 1.00999e-06 [virtual_shard_identity]: 6.73e-06 [virtual_dataset]: 5.02e-06 [get_grad_eliminate_]: 4.80999e-06 [virtual_output]: 5.32999e-06 [merge_forward]: 4.57998e-06 [cell_reuse_recompute_pass]: 3.56001e-06 [offload_activation]: 9.96e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.592e-05 [merge_recompute_call_nodes]: 1.40999e-06 [before_grad]: 9.61e-06 [set_forward_comm_id_for_comm_node_pass]: 3.88999e-06 [meta_fg_expand]: 2.56998e-06 [flash_sp_send_recv_attached]: 2.21998e-06 [receive_attached]: 2.88e-06 [after_resolve]: 6.91999e-06 [a_after_grad]: 5.197e-05 [renormalize]: 3.50003e-07 [add_forward_monad_depend]: 1.97999e-06 [auto_monad_grad]: 1.47999e-06 [auto_monad_eliminator]: 8.15e-06 [cse]: 2.103e-05 [a_3]: 3.465e-05 [py_interpret_to_execute_after_opt_a]: 1.019e-05 [slice_cell_reuse_recomputed_activation]: 2.66e-06 [rewriter_after_opt_a]: 2.525e-05 [convert_after_rewriter]: 1.71e-06 [order_py_execute_after_rewriter]: 1.23002e-06 [mutable_eliminate]: 0.0008029 [opt_b]: 0.00021653, [1] [Cycle 1]: 0.00020842, [7] [b_1]: 0.00012034 [b_2]: 6.94999e-06 [updatestate_depend_eliminate]: 7.68999e-06 [updatestate_assign_eliminate]: 2.71e-06 [updatestate_loads_eliminate]: 2.44999e-06 [renormalize]: 7.2e-07 [cse]: 2.853e-05 [optimize_parallel_all_gather_comm]: 2.052e-05 [overlap_param_gather]: 1.98002e-06 [cconv]: 5.845e-05 [loop_unroll]: 0.0005234 [opt_after_cconv]: 0.0343111, [1] [Cycle 1]: 0.0343004, [7] [c_1]: 2.294e-05 [parameter_eliminate]: 0.034106 [updatestate_depend_eliminate]: 2.332e-05 [updatestate_assign_eliminate]: 3.92002e-06 [updatestate_loads_eliminate]: 3.64002e-06 [cse]: 4.753e-05 [renormalize]: 5.50004e-07 [remove_dup_value]: 2.279e-05 [tuple_transform]: 0.00011867, [1] [Cycle 1]: 0.00010887, [4] [d_1]: 7.303e-05 [none_parameter_eliminate]: 5.04e-06 [renormalize]: 1.39989e-07 [switch_simplify]: 8.89e-06 [partial_unused_args_eliminate]: 2.14e-06 [add_recomputation]: 6.418e-05 [cse_after_recomputation]: 3.458e-05, [1] [Cycle 1]: 2.679e-05, [1] [cse]: 1.904e-05 [environ_conv]: 9.04003e-06 [swap_dp_allreduce_reducescatter]: 6.31e-06 [bias_add_comm_swap]: 3.48e-06 [label_micro_interleaved_index]: 7.73001e-06 [label_fine_grained_interleaved_index]: 3.13e-06 [merge_cast_opt]: 1.42e-06 [slice_recompute_activation]: 2.19001e-06 [micro_interleaved_order_control]: 2.51e-06 [assign_add_opt]: 1.67001e-06 [ForceFp32Comm]: 1.47999e-06 [remove_cast_before_assign_add]: 1.10999e-06 [full_micro_interleaved_order_control]: 2.77002e-06 [reorder_send_recv_between_fp_bp]: 2.66e-06 [comm_op_add_attrs]: 1.02998e-06 [add_comm_op_reuse_tag]: 9.60019e-07 [interleave_split_concat_branches]: 1.35001e-06 [interleave_parallel_branches]: 1.06002e-06 [overlap_opt_shard_in_pipeline]: 1.24e-06 [overlap_opt_shard_grad_in_pipeline]: 2.31e-06 [control_data_broadcast_order]: 2.031e-05 [grouped_pairwise_exchange_alltoall]: 1.71e-06 [offloading_packed_experts]: 4.10998e-06 [overlap_recompute_and_grad_model_parallel]: 5.02e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.42e-06 [overlap_recompute_allgather_and_fa_grad]: 1.17999e-06 [overlap_recompute_comm]: 2.37999e-06 [overlap_grad_ring_attention]: 4.42e-06 [overlap_grad_flash_sp]: 2.433e-05 [begin_end_overlap_inline]: 5.39992e-07 [split_matmul_comm_elemetwise]: 2.27999e-06 [split_layernorm_comm]: 2.69001e-06 [handle_group_info]: 1.20001e-06 [symbol_engine_optimizer]: 9.471e-05, [1] [Cycle 1]: 8.851e-05, [6] [build]: 4.50001e-06 [elim_shapecalc]: 1.665e-05 [elim_not_effective]: 1.569e-05 [opt_reshape]: 7.48999e-06 [fold_const_symbol]: 9.82999e-06 [renormalize]: 3.29979e-07 [detach_backward]: 2.31998e-06 [pipeline_parallel_scheduler]: 1.47999e-06 [auto_monad_reorder]: 2.37e-05 [get_jit_bprop_graph]: 2.12999e-06 [rewriter_after_jit_bprop_graph]: 7.06001e-06 [opt_after_jit_grad]: 0.00089651 [validate]: 5.245e-05 Sums bootstrap : 0.000535s : 0.02% type_inference : 1.883208s : 84.93% event_method : 0.000770s : 0.03% auto_monad : 0.000215s : 0.01% graph_reusing : 0.000011s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000089s : 0.00% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000013s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000089s : 0.00% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000002s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000013s : 0.00% optimize.rewriter_before_opt_a : 0.000672s : 0.03% optimize.opt_a.expand_dump_flag : 0.000016s : 0.00% optimize.opt_a.switch_simplify : 0.000274s : 0.01% optimize.opt_a.loop_unroll : 0.000142s : 0.01% optimize.opt_a.a_1 : 0.042680s : 1.92% optimize.opt_a.with_stream_mark : 0.000100s : 0.00% optimize.opt_a.recompute_prepare : 0.000058s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000021s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000014s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000013s : 0.00% optimize.opt_a.parameter_eliminate : 0.000009s : 0.00% optimize.opt_a.a_2 : 0.000380s : 0.02% optimize.opt_a.accelerated_algorithm : 0.000029s : 0.00% optimize.opt_a.shard : 0.000007s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000015s : 0.00% optimize.opt_a.shard_inline : 0.000029s : 0.00% optimize.opt_a.merge_send_recv : 0.000040s : 0.00% optimize.opt_a.auto_parallel : 0.000038s : 0.00% optimize.opt_a.parallel : 0.000042s : 0.00% optimize.opt_a.flash_sp : 0.000020s : 0.00% optimize.opt_a.merge_comm : 0.000018s : 0.00% optimize.opt_a.allreduce_fusion : 0.000016s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000055s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000003s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000035s : 0.00% optimize.opt_a.virtual_dataset : 0.000029s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000027s : 0.00% optimize.opt_a.virtual_output : 0.000029s : 0.00% optimize.opt_a.merge_forward : 0.000019s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000008s : 0.00% optimize.opt_a.offload_activation : 0.000040s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000067s : 0.00% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000051s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000018s : 0.00% optimize.opt_a.meta_fg_expand : 0.004226s : 0.19% optimize.opt_a.flash_sp_send_recv_attached : 0.000012s : 0.00% optimize.opt_a.receive_attached : 0.000009s : 0.00% optimize.opt_a.after_resolve : 0.000133s : 0.01% optimize.opt_a.a_after_grad : 0.000174s : 0.01% optimize.opt_a.renormalize : 0.228021s : 10.28% optimize.opt_a.add_forward_monad_depend : 0.000037s : 0.00% optimize.opt_a.auto_monad_grad : 0.000014s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000109s : 0.00% optimize.opt_a.cse : 0.000659s : 0.03% optimize.opt_a.a_3 : 0.016784s : 0.76% optimize.py_interpret_to_execute_after_opt_a : 0.000010s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000025s : 0.00% optimize.convert_after_rewriter : 0.000002s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000803s : 0.04% optimize.opt_b.b_1 : 0.000120s : 0.01% optimize.opt_b.b_2 : 0.000007s : 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 : 0.000029s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000021s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000058s : 0.00% optimize.loop_unroll : 0.000523s : 0.02% optimize.opt_after_cconv.c_1 : 0.000023s : 0.00% optimize.opt_after_cconv.parameter_eliminate : 0.034106s : 1.54% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000023s : 0.00% optimize.opt_after_cconv.updatestate_assign_eliminate : 0.000004s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000004s : 0.00% optimize.opt_after_cconv.cse : 0.000048s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000023s : 0.00% optimize.tuple_transform.d_1 : 0.000073s : 0.00% optimize.tuple_transform.none_parameter_eliminate : 0.000005s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000009s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000064s : 0.00% optimize.cse_after_recomputation.cse : 0.000019s : 0.00% optimize.environ_conv : 0.000009s : 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 : 0.000008s : 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 : 0.000003s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000020s : 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 : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000004s : 0.00% optimize.overlap_grad_flash_sp : 0.000024s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000002s : 0.00% optimize.split_layernorm_comm : 0.000003s : 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 : 0.000017s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000016s : 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 : 0.000001s : 0.00% auto_monad_reorder : 0.000024s : 0.00% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.00% opt_after_jit_grad : 0.000897s : 0.04% validate : 0.000052s : 0.00% Time group info: ------[substitution.] 0.007167 163 0.04% : 0.000003s : 2: substitution.elim_not_effective 0.21% : 0.000015s : 12: substitution.float_depend_g_call 0.06% : 0.000004s : 2: substitution.float_tuple_getitem_switch 0.02% : 0.000002s : 2: substitution.fold_const_symbol 0.10% : 0.000007s : 2: substitution.graph_param_transform 0.05% : 0.000004s : 2: substitution.incorporate_call 0.04% : 0.000003s : 2: substitution.incorporate_call_switch 94.55% : 0.006777s : 23: substitution.inline 0.37% : 0.000027s : 2: substitution.inline_without_move 0.15% : 0.000011s : 12: substitution.j_node_and_user_rematch 0.22% : 0.000016s : 7: substitution.minmaximum_grad 0.20% : 0.000015s : 12: substitution.partial_eliminate 0.17% : 0.000012s : 12: substitution.remove_not_recompute_node 0.67% : 0.000048s : 9: substitution.replace_applicator 0.12% : 0.000009s : 11: substitution.replace_old_param 0.06% : 0.000005s : 1: substitution.set_cell_output_no_recompute 0.26% : 0.000018s : 3: substitution.switch_simplify 0.45% : 0.000032s : 7: substitution.tuple_list_convert_item_index_to_positive 0.19% : 0.000013s : 7: substitution.tuple_list_get_item_const_eliminator 0.29% : 0.000020s : 7: substitution.tuple_list_get_item_depend_reorder 1.53% : 0.000109s : 19: substitution.tuple_list_get_item_eliminator 0.26% : 0.000019s : 7: substitution.tuple_list_get_set_item_eliminator ------[type_inference.] 1.883015 2 97.17% : 1.829663s : 1: type_inference.infer 2.83% : 0.053352s : 1: type_inference.specialize ------[replace.] 0.000557 36 56.13% : 0.000313s : 23: replace.inline 9.60% : 0.000053s : 3: replace.switch_simplify 34.27% : 0.000191s : 10: replace.tuple_list_get_item_eliminator ------[match.] 0.006835 36 98.86% : 0.006757s : 23: match.inline 0.21% : 0.000015s : 3: match.switch_simplify 0.92% : 0.000063s : 10: match.tuple_list_get_item_eliminator ------[predicate.] 0.000739 4067 1.13% : 0.000008s : 53: predicate.accumulaten_eliminater 0.25% : 0.000002s : 2: predicate.ad_related_special_op_eliminate 0.30% : 0.000002s : 16: predicate.addn_check_dump 1.52% : 0.000011s : 53: predicate.addn_zero_filter 0.96% : 0.000007s : 53: predicate.adjust_all_reduce_mul_add 2.92% : 0.000022s : 69: predicate.arithmetic_simplify 1.24% : 0.000009s : 53: predicate.cast_eliminate 1.59% : 0.000012s : 43: predicate.check_bprop_eliminate 0.32% : 0.000002s : 16: predicate.compare_switch_simplify 0.03% : 0.000000s : 2: predicate.const_output_eliminate 0.31% : 0.000002s : 16: predicate.depend_value_elim 1.09% : 0.000008s : 53: predicate.dict_get_item_const_eliminator 1.66% : 0.000012s : 53: predicate.dict_get_item_eliminator 1.06% : 0.000008s : 53: predicate.dict_set_item_eliminator 0.37% : 0.000003s : 4: predicate.dumpgradient_eliminate 0.09% : 0.000001s : 2: predicate.elim_not_effective 0.18% : 0.000001s : 2: predicate.elim_shapecalc_of_broadcastargs 1.14% : 0.000008s : 55: predicate.environ_add_const_eliminate 1.03% : 0.000008s : 55: predicate.environ_get_add_eliminate 1.03% : 0.000008s : 55: predicate.environ_get_depend_swap 1.44% : 0.000011s : 71: predicate.environ_get_eliminate 0.98% : 0.000007s : 55: predicate.environ_get_set_eliminate 1.71% : 0.000013s : 86: predicate.exchange_switch_depend_value 2.75% : 0.000020s : 86: predicate.float_depend_g_call 0.31% : 0.000002s : 16: predicate.float_environ_get_switch 0.37% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.02% : 0.000000s : 2: predicate.fold_const_symbol 0.41% : 0.000003s : 16: predicate.get_grad_eliminate 0.12% : 0.000001s : 2: predicate.graph_param_transform 0.34% : 0.000003s : 16: predicate.incorporate_call 0.28% : 0.000002s : 16: predicate.incorporate_call_switch 5.93% : 0.000044s : 175: predicate.inline 1.49% : 0.000011s : 46: predicate.inline_without_move 0.20% : 0.000002s : 16: predicate.j_node_and_user_rematch 0.49% : 0.000004s : 16: predicate.less_batch_normalization 1.49% : 0.000011s : 67: predicate.list_to_tuple_eliminator_ 2.88% : 0.000021s : 120: predicate.load_eliminater 0.35% : 0.000003s : 2: predicate.loop_unroll_after_grad 2.67% : 0.000020s : 146: predicate.loop_unroll_before_grad 1.41% : 0.000010s : 57: predicate.make_slice_get_slice_eliminator 0.36% : 0.000003s : 16: predicate.merge_addn 0.82% : 0.000006s : 43: predicate.micro_step_allgather_replace 0.82% : 0.000006s : 43: predicate.mini_step_allgather_replace 1.03% : 0.000008s : 53: predicate.minmaximum_grad 0.31% : 0.000002s : 2: predicate.mutable_eliminate 0.08% : 0.000001s : 2: predicate.opt_reshape 0.09% : 0.000001s : 2: predicate.parallel_virtual_node 3.16% : 0.000023s : 86: predicate.partial_defer_inline 1.53% : 0.000011s : 65: predicate.partial_eliminate 1.10% : 0.000008s : 53: predicate.print_const_string_wrapper 0.36% : 0.000003s : 16: predicate.reduce_all_const_elim 1.50% : 0.000011s : 53: predicate.reduce_eliminate 2.35% : 0.000017s : 120: predicate.redundant_stop_gradient_eliminater 0.26% : 0.000002s : 16: predicate.remove_not_recompute_node 1.60% : 0.000012s : 106: predicate.replace_applicator 0.76% : 0.000006s : 46: predicate.replace_old_param 0.07% : 0.000001s : 2: predicate.reset_defer_inline 1.16% : 0.000009s : 53: predicate.reshape_eliminate 0.85% : 0.000006s : 43: predicate.row_tensor_add_zeros_like 0.09% : 0.000001s : 2: predicate.row_tensor_eliminate 2.73% : 0.000020s : 43: predicate.same_eliminate 0.27% : 0.000002s : 16: predicate.set_cell_output_no_recompute 0.42% : 0.000003s : 16: predicate.shard_identity_eliminate 0.22% : 0.000002s : 4: predicate.special_op_eliminate 0.35% : 0.000003s : 16: predicate.specialize_transform 1.08% : 0.000008s : 43: predicate.split_environ_get_set_with_tuple_value 1.32% : 0.000010s : 46: predicate.stack_unstack_eliminate 0.12% : 0.000001s : 2: predicate.switch_call_monad_eliminater 1.98% : 0.000015s : 86: predicate.switch_defer_inline 2.69% : 0.000020s : 129: predicate.switch_layer_defer_inline 5.60% : 0.000041s : 256: predicate.switch_simplify 1.06% : 0.000008s : 53: predicate.tile_eliminate 1.03% : 0.000008s : 53: predicate.transpose_eliminate 1.44% : 0.000011s : 57: predicate.tuple_list_convert_item_index_to_positive 1.41% : 0.000010s : 57: predicate.tuple_list_get_item_const_eliminator 1.44% : 0.000011s : 57: predicate.tuple_list_get_item_depend_reorder 3.07% : 0.000023s : 83: predicate.tuple_list_get_item_eliminator 2.24% : 0.000017s : 57: predicate.tuple_list_get_set_item_eliminator 2.12% : 0.000016s : 73: predicate.tuple_list_set_item_eliminator 1.51% : 0.000011s : 67: predicate.tuple_to_list_eliminator_ 2.13% : 0.000016s : 120: predicate.updatestate_pure_node_eliminater 2.59% : 0.000019s : 136: predicate.updatestate_useless_node_eliminater 0.07% : 0.000001s : 2: predicate.value_based_eliminate 0.42% : 0.000003s : 16: predicate.virtual_dataset_eliminate 0.40% : 0.000003s : 16: predicate.virtual_output_eliminate 0.06% : 0.000000s : 2: predicate.virtual_view_grad_eliminate 0.09% : 0.000001s : 2: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.081201 72 97.40% : 0.079089s : 45: func_graph_cloner_run.FuncGraphClonerGraph 2.60% : 0.002111s : 27: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 2.919725 233 0.00% : 0.000006s : 1: ForceFp32Comm 1.34% : 0.039174s : 1: add_attr 1.34% : 0.039157s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.00% : 0.000069s : 1: add_recomputation 0.01% : 0.000148s : 1: assign_add_opt 0.01% : 0.000230s : 1: auto_monad 0.00% : 0.000029s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000008s : 1: bias_add_comm_swap 0.02% : 0.000571s : 1: bootstrap 0.00% : 0.000063s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000025s : 1: control_data_broadcast_order 0.00% : 0.000005s : 1: convert_after_rewriter 0.00% : 0.000037s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000012s : 1: environ_conv 0.03% : 0.000792s : 1: event_method 0.00% : 0.000007s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000016s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000011s : 1: label_micro_interleaved_index 0.02% : 0.000536s : 1: loop_unroll 0.00% : 0.000008s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.03% : 0.000814s : 1: mutable_eliminate 0.00% : 0.000007s : 1: offloading_packed_experts 0.00% : 0.000017s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000019s : 1: opt.transform.mutable_eliminate 2.08% : 0.060715s : 117: opt.transform.opt_a 0.00% : 0.000022s : 1: opt.transform.opt_after_cconv 0.00% : 0.000034s : 1: opt.transform.opt_after_jit_grad 0.00% : 0.000095s : 28: opt.transform.opt_b 0.00% : 0.000076s : 2: opt.transform.opt_trans_graph 0.00% : 0.000046s : 4: opt.transform.symbol_engine_opt 10.13% : 0.295698s : 1: opt_a 1.18% : 0.034325s : 1: opt_after_cconv 0.03% : 0.000915s : 1: opt_after_jit_grad 0.01% : 0.000220s : 1: opt_b 11.41% : 0.333251s : 1: optimize 0.00% : 0.000024s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 1: order_py_execute_after_rewriter 0.00% : 0.000030s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 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% : 0.000007s : 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% : 0.000009s : 1: parallel-infer-symbol 0.00% : 0.000005s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000006s : 1: pipeline_split 0.00% : 0.000095s : 1: pre_auto_parallel 0.00% : 0.000016s : 1: py_interpret_to_execute 0.00% : 0.000013s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000027s : 1: remove_dup_value 7.06% : 0.206163s : 2: renormalize.infer 0.75% : 0.021819s : 2: 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% : 0.000029s : 1: rewriter_after_opt_a 0.02% : 0.000685s : 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% : 0.000006s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.00% : 0.000098s : 1: symbol_engine_optimizer 0.00% : 0.000122s : 1: tuple_transform 64.50% : 1.883246s : 1: type_inference TotalTime = 1.63767, [21] [bootstrap]: 0.00054806 [type_inference]: 1.4616 [event_method]: 0.00075687 [auto_monad]: 0.00011435 [graph_reusing]: 8.55999e-06 [inline]: 3.08998e-06 [add_attr]: 0.00415645, [1] [add_attr_with_inline]: 0.00414233, [1] [Cycle 1]: 0.00010825, [2] [tag_attr]: 4.816e-05 [meta_addattr_fg_expand]: 8.13001e-06 [parallel-infer-symbol]: 3.54002e-06 [pre_auto_parallel]: 0.0643958 [insert-virtual-dataset]: 7.11999e-06 [parallel-infer-symbol-second]: 3.08998e-06 [dataset_repeat_opt]: 3.14001e-06 [pipeline_split]: 4.12e-06 [optimize]: 0.104815, [53] [py_interpret_to_execute]: 1.308e-05 [rewriter_before_opt_a]: 0.00041994 [opt_a]: 0.0843847, [2] [Cycle 1]: 0.0789563, [45] [expand_dump_flag]: 6.94999e-06 [switch_simplify]: 7.207e-05 [loop_unroll]: 4.934e-05 [a_1]: 0.00368682 [with_stream_mark]: 3.224e-05 [recompute_prepare]: 1.514e-05 [updatestate_depend_eliminate]: 4.84e-06 [updatestate_assign_eliminate]: 3.35e-06 [updatestate_loads_eliminate]: 3.03998e-06 [parameter_eliminate]: 3.06999e-06 [a_2]: 8.617e-05 [accelerated_algorithm]: 7.87e-06 [shard]: 2.24999e-06 [meta_shard_fg_expand]: 3.38999e-06 [shard_inline]: 7.65e-06 [merge_send_recv]: 9.04e-06 [auto_parallel]: 1.039e-05 [parallel]: 2.102e-05 [flash_sp]: 1.189e-05 [merge_comm]: 4.70999e-06 [allreduce_fusion]: 3.48999e-06 [matmul_add_comm_reduction]: 1.129e-05 [allreduce_slice_to_reducescatter]: 1.19003e-06 [virtual_shard_identity]: 1.027e-05 [virtual_dataset]: 6.91001e-06 [get_grad_eliminate_]: 6.23998e-06 [virtual_output]: 6.79999e-06 [merge_forward]: 4.89e-06 [cell_reuse_recompute_pass]: 1.77001e-06 [offload_activation]: 1.144e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.844e-05 [merge_recompute_call_nodes]: 1.60999e-06 [before_grad]: 1.169e-05 [set_forward_comm_id_for_comm_node_pass]: 3.70998e-06 [meta_fg_expand]: 3.04999e-06 [flash_sp_send_recv_attached]: 2.66e-06 [receive_attached]: 2.19999e-06 [after_resolve]: 1.112e-05 [a_after_grad]: 9.36e-06 [renormalize]: 0.0486244 [add_forward_monad_depend]: 1.384e-05 [auto_monad_grad]: 1.187e-05 [auto_monad_eliminator]: 4.485e-05 [cse]: 5.234e-05 [a_3]: 8.216e-05 [Cycle 2]: 0.00540431, [45] [expand_dump_flag]: 3.30998e-06 [switch_simplify]: 9.81e-06 [loop_unroll]: 7.00002e-06 [a_1]: 0.00011953 [with_stream_mark]: 2.517e-05 [recompute_prepare]: 8.13001e-06 [updatestate_depend_eliminate]: 4.97e-06 [updatestate_assign_eliminate]: 3.65e-06 [updatestate_loads_eliminate]: 4.42e-06 [parameter_eliminate]: 2.36e-06 [a_2]: 0.00456712 [accelerated_algorithm]: 1.691e-05 [shard]: 5.10999e-06 [meta_shard_fg_expand]: 4.72998e-06 [shard_inline]: 6.59999e-06 [merge_send_recv]: 1.84e-05 [auto_parallel]: 1.53e-05 [parallel]: 1.14e-05 [flash_sp]: 5.94999e-06 [merge_comm]: 4.50001e-06 [allreduce_fusion]: 3.68e-06 [matmul_add_comm_reduction]: 1.371e-05 [allreduce_slice_to_reducescatter]: 1.00001e-06 [virtual_shard_identity]: 1.179e-05 [virtual_dataset]: 6.69999e-06 [get_grad_eliminate_]: 6.87002e-06 [virtual_output]: 6.59999e-06 [merge_forward]: 4.80999e-06 [cell_reuse_recompute_pass]: 4.63999e-06 [offload_activation]: 1.076e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.382e-05 [merge_recompute_call_nodes]: 1.66e-06 [before_grad]: 1.148e-05 [set_forward_comm_id_for_comm_node_pass]: 4.25e-06 [meta_fg_expand]: 3.46999e-06 [flash_sp_send_recv_attached]: 1.80001e-06 [receive_attached]: 2.58e-06 [after_resolve]: 1.4e-05 [a_after_grad]: 1.152e-05 [renormalize]: 8.9989e-08 [add_forward_monad_depend]: 7.16999e-06 [auto_monad_grad]: 3.18e-06 [auto_monad_eliminator]: 1.948e-05 [cse]: 5.138e-05 [a_3]: 4.528e-05 [py_interpret_to_execute_after_opt_a]: 1.119e-05 [slice_cell_reuse_recomputed_activation]: 2.31e-06 [rewriter_after_opt_a]: 2.675e-05 [convert_after_rewriter]: 2.48e-06 [order_py_execute_after_rewriter]: 1.29e-06 [mutable_eliminate]: 0.00087978 [opt_b]: 0.00023806, [1] [Cycle 1]: 0.00022625, [7] [b_1]: 0.00013116 [b_2]: 8.47e-06 [updatestate_depend_eliminate]: 9.09e-06 [updatestate_assign_eliminate]: 3.26999e-06 [updatestate_loads_eliminate]: 3.23e-06 [renormalize]: 7.2e-07 [cse]: 3.278e-05 [optimize_parallel_all_gather_comm]: 2.068e-05 [overlap_param_gather]: 1.95001e-06 [cconv]: 3.759e-05 [loop_unroll]: 0.0148744 [opt_after_cconv]: 0.00017777, [1] [Cycle 1]: 0.00016542, [7] [c_1]: 3.615e-05 [parameter_eliminate]: 7.65e-06 [updatestate_depend_eliminate]: 1.48e-05 [updatestate_assign_eliminate]: 4.17e-06 [updatestate_loads_eliminate]: 3.45998e-06 [cse]: 5.87e-05 [renormalize]: 7.30011e-07 [remove_dup_value]: 2.108e-05 [tuple_transform]: 0.00011841, [1] [Cycle 1]: 0.00011272, [4] [d_1]: 8.15e-05 [none_parameter_eliminate]: 2.39001e-06 [renormalize]: 1.40019e-07 [switch_simplify]: 8.33001e-06 [partial_unused_args_eliminate]: 2.57001e-06 [add_recomputation]: 6.456e-05 [cse_after_recomputation]: 3.243e-05, [1] [Cycle 1]: 2.641e-05, [1] [cse]: 1.599e-05 [environ_conv]: 9.12999e-06 [swap_dp_allreduce_reducescatter]: 5.67001e-06 [bias_add_comm_swap]: 3.73999e-06 [label_micro_interleaved_index]: 7.85e-06 [label_fine_grained_interleaved_index]: 2.79001e-06 [merge_cast_opt]: 1.91e-06 [slice_recompute_activation]: 2.11998e-06 [micro_interleaved_order_control]: 2.77002e-06 [assign_add_opt]: 1.44e-06 [ForceFp32Comm]: 8.50006e-07 [remove_cast_before_assign_add]: 1.05999e-06 [full_micro_interleaved_order_control]: 2.38998e-06 [reorder_send_recv_between_fp_bp]: 3.20998e-06 [comm_op_add_attrs]: 1.00001e-06 [add_comm_op_reuse_tag]: 9.50007e-07 [interleave_split_concat_branches]: 1.34998e-06 [interleave_parallel_branches]: 1.03001e-06 [overlap_opt_shard_in_pipeline]: 1.25999e-06 [overlap_opt_shard_grad_in_pipeline]: 1.84e-06 [control_data_broadcast_order]: 1.662e-05 [grouped_pairwise_exchange_alltoall]: 1.76e-06 [offloading_packed_experts]: 0.00287588 [overlap_recompute_and_grad_model_parallel]: 1.586e-05 [overlap_grad_matmul_and_grad_allreduce]: 1.40999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.40001e-06 [overlap_recompute_comm]: 2.76e-06 [overlap_grad_ring_attention]: 5.10001e-06 [overlap_grad_flash_sp]: 3.152e-05 [begin_end_overlap_inline]: 7.80012e-07 [split_matmul_comm_elemetwise]: 2.76e-06 [split_layernorm_comm]: 1.82001e-06 [handle_group_info]: 1.59998e-06 [symbol_engine_optimizer]: 0.00014029, [1] [Cycle 1]: 0.00012981, [6] [build]: 8.12998e-06 [elim_shapecalc]: 3.203e-05 [elim_not_effective]: 2.513e-05 [opt_reshape]: 7.71999e-06 [fold_const_symbol]: 1.31e-05 [renormalize]: 6.89994e-07 [detach_backward]: 3.19001e-06 [pipeline_parallel_scheduler]: 1.57001e-06 [auto_monad_reorder]: 2.947e-05 [get_jit_bprop_graph]: 2.01998e-06 [rewriter_after_jit_bprop_graph]: 7.88001e-06 [opt_after_jit_grad]: 0.00085998 [validate]: 5.719e-05 Sums bootstrap : 0.000548s : 0.03% type_inference : 1.461596s : 90.98% event_method : 0.000757s : 0.05% auto_monad : 0.000114s : 0.01% graph_reusing : 0.000009s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000048s : 0.00% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000008s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.064396s : 4.01% insert-virtual-dataset : 0.000007s : 0.00% parallel-infer-symbol-second : 0.000003s : 0.00% dataset_repeat_opt : 0.000003s : 0.00% pipeline_split : 0.000004s : 0.00% optimize.py_interpret_to_execute : 0.000013s : 0.00% optimize.rewriter_before_opt_a : 0.000420s : 0.03% optimize.opt_a.expand_dump_flag : 0.000010s : 0.00% optimize.opt_a.switch_simplify : 0.000082s : 0.01% optimize.opt_a.loop_unroll : 0.000056s : 0.00% optimize.opt_a.a_1 : 0.003806s : 0.24% optimize.opt_a.with_stream_mark : 0.000057s : 0.00% optimize.opt_a.recompute_prepare : 0.000023s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000010s : 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 : 0.004653s : 0.29% optimize.opt_a.accelerated_algorithm : 0.000025s : 0.00% optimize.opt_a.shard : 0.000007s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000008s : 0.00% optimize.opt_a.shard_inline : 0.000014s : 0.00% optimize.opt_a.merge_send_recv : 0.000027s : 0.00% optimize.opt_a.auto_parallel : 0.000026s : 0.00% optimize.opt_a.parallel : 0.000032s : 0.00% optimize.opt_a.flash_sp : 0.000018s : 0.00% optimize.opt_a.merge_comm : 0.000009s : 0.00% optimize.opt_a.allreduce_fusion : 0.000007s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000025s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000022s : 0.00% optimize.opt_a.virtual_dataset : 0.000014s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.00% optimize.opt_a.virtual_output : 0.000013s : 0.00% optimize.opt_a.merge_forward : 0.000010s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000006s : 0.00% optimize.opt_a.offload_activation : 0.000022s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000042s : 0.00% optimize.opt_a.merge_recompute_call_nodes : 0.000003s : 0.00% optimize.opt_a.before_grad : 0.000023s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000008s : 0.00% optimize.opt_a.meta_fg_expand : 0.000007s : 0.00% optimize.opt_a.flash_sp_send_recv_attached : 0.000004s : 0.00% optimize.opt_a.receive_attached : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000025s : 0.00% optimize.opt_a.a_after_grad : 0.000021s : 0.00% optimize.opt_a.renormalize : 0.048625s : 3.03% optimize.opt_a.add_forward_monad_depend : 0.000021s : 0.00% optimize.opt_a.auto_monad_grad : 0.000015s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000064s : 0.00% optimize.opt_a.cse : 0.000104s : 0.01% optimize.opt_a.a_3 : 0.000127s : 0.01% optimize.py_interpret_to_execute_after_opt_a : 0.000011s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000027s : 0.00% optimize.convert_after_rewriter : 0.000002s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000880s : 0.05% optimize.opt_b.b_1 : 0.000131s : 0.01% optimize.opt_b.b_2 : 0.000008s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000009s : 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 : 0.000033s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000021s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000038s : 0.00% optimize.loop_unroll : 0.014874s : 0.93% optimize.opt_after_cconv.c_1 : 0.000036s : 0.00% optimize.opt_after_cconv.parameter_eliminate : 0.000008s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000015s : 0.00% 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 : 0.000059s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000021s : 0.00% optimize.tuple_transform.d_1 : 0.000082s : 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 : 0.000008s : 0.00% optimize.partial_unused_args_eliminate : 0.000003s : 0.00% optimize.add_recomputation : 0.000065s : 0.00% optimize.cse_after_recomputation.cse : 0.000016s : 0.00% optimize.environ_conv : 0.000009s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000006s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000008s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000017s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.002876s : 0.18% optimize.overlap_recompute_and_grad_model_parallel : 0.000016s : 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 : 0.000005s : 0.00% optimize.overlap_grad_flash_sp : 0.000032s : 0.00% optimize.begin_end_overlap_inline : 0.000001s : 0.00% optimize.split_matmul_comm_elemetwise : 0.000003s : 0.00% optimize.split_layernorm_comm : 0.000002s : 0.00% optimize.handle_group_info : 0.000002s : 0.00% optimize.symbol_engine_optimizer.build : 0.000008s : 0.00% optimize.symbol_engine_optimizer.elim_shapecalc : 0.000032s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000025s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000013s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000001s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000002s : 0.00% auto_monad_reorder : 0.000029s : 0.00% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000008s : 0.00% opt_after_jit_grad : 0.000860s : 0.05% validate : 0.000057s : 0.00% Time group info: ------[substitution.] 0.002941 28 0.12% : 0.000003s : 2: substitution.elim_not_effective 0.05% : 0.000001s : 2: substitution.fold_const_symbol 0.22% : 0.000007s : 3: substitution.graph_param_transform 98.21% : 0.002888s : 7: substitution.inline 0.18% : 0.000005s : 4: substitution.j_node_and_user_rematch 0.21% : 0.000006s : 4: substitution.remove_not_recompute_node 0.24% : 0.000007s : 2: substitution.replace_old_param 0.77% : 0.000023s : 4: substitution.tuple_list_get_item_eliminator ------[type_inference.] 1.461451 2 98.20% : 1.435143s : 1: type_inference.infer 1.80% : 0.026308s : 1: type_inference.specialize ------[replace.] 0.000136 11 69.45% : 0.000094s : 7: replace.inline 30.55% : 0.000041s : 4: replace.tuple_list_get_item_eliminator ------[match.] 0.002899 11 99.31% : 0.002879s : 7: match.inline 0.69% : 0.000020s : 4: match.tuple_list_get_item_eliminator ------[predicate.] 0.000301 1305 0.93% : 0.000003s : 15: predicate.accumulaten_eliminater 0.67% : 0.000002s : 3: predicate.ad_related_special_op_eliminate 0.41% : 0.000001s : 6: predicate.addn_check_dump 1.02% : 0.000003s : 15: predicate.addn_zero_filter 0.70% : 0.000002s : 15: predicate.adjust_all_reduce_mul_add 2.45% : 0.000007s : 21: predicate.arithmetic_simplify 0.97% : 0.000003s : 15: predicate.cast_eliminate 0.34% : 0.000001s : 6: predicate.check_bprop_eliminate 1.31% : 0.000004s : 6: predicate.compare_switch_simplify 0.10% : 0.000000s : 3: predicate.const_output_eliminate 0.43% : 0.000001s : 6: predicate.depend_value_elim 0.80% : 0.000002s : 15: predicate.dict_get_item_const_eliminator 1.12% : 0.000003s : 15: predicate.dict_get_item_eliminator 0.82% : 0.000002s : 15: predicate.dict_set_item_eliminator 0.85% : 0.000003s : 6: predicate.dumpgradient_eliminate 0.30% : 0.000001s : 3: predicate.elim_not_effective 0.35% : 0.000001s : 3: predicate.elim_shapecalc_of_broadcastargs 1.25% : 0.000004s : 18: predicate.environ_add_const_eliminate 0.94% : 0.000003s : 18: predicate.environ_get_add_eliminate 0.85% : 0.000003s : 18: predicate.environ_get_depend_swap 1.28% : 0.000004s : 24: predicate.environ_get_eliminate 0.92% : 0.000003s : 18: predicate.environ_get_set_eliminate 1.29% : 0.000004s : 26: predicate.exchange_switch_depend_value 2.63% : 0.000008s : 26: predicate.float_depend_g_call 0.35% : 0.000001s : 6: predicate.float_environ_get_switch 0.48% : 0.000001s : 9: predicate.float_tuple_getitem_switch 0.09% : 0.000000s : 3: predicate.fold_const_symbol 0.41% : 0.000001s : 6: predicate.get_grad_eliminate 0.17% : 0.000001s : 3: predicate.graph_param_transform 0.38% : 0.000001s : 6: predicate.incorporate_call 0.27% : 0.000001s : 6: predicate.incorporate_call_switch 6.16% : 0.000019s : 59: predicate.inline 0.44% : 0.000001s : 6: predicate.inline_without_move 0.18% : 0.000001s : 6: predicate.j_node_and_user_rematch 0.71% : 0.000002s : 6: predicate.less_batch_normalization 1.73% : 0.000005s : 25: predicate.list_to_tuple_eliminator_ 2.01% : 0.000006s : 40: predicate.load_eliminater 2.36% : 0.000007s : 3: predicate.loop_unroll_after_grad 2.51% : 0.000008s : 56: predicate.loop_unroll_before_grad 1.53% : 0.000005s : 21: predicate.make_slice_get_slice_eliminator 0.79% : 0.000002s : 6: predicate.merge_addn 0.36% : 0.000001s : 6: predicate.micro_step_allgather_replace 0.47% : 0.000001s : 6: predicate.mini_step_allgather_replace 0.72% : 0.000002s : 15: predicate.minmaximum_grad 1.42% : 0.000004s : 3: predicate.mutable_eliminate 0.27% : 0.000001s : 3: predicate.opt_reshape 0.22% : 0.000001s : 3: predicate.parallel_virtual_node 2.04% : 0.000006s : 26: predicate.partial_defer_inline 1.15% : 0.000003s : 22: predicate.partial_eliminate 1.12% : 0.000003s : 15: predicate.print_const_string_wrapper 0.44% : 0.000001s : 6: predicate.reduce_all_const_elim 1.50% : 0.000005s : 15: predicate.reduce_eliminate 2.15% : 0.000006s : 40: predicate.redundant_stop_gradient_eliminater 0.38% : 0.000001s : 6: predicate.remove_not_recompute_node 1.20% : 0.000004s : 25: predicate.replace_applicator 0.32% : 0.000001s : 6: predicate.replace_old_param 0.22% : 0.000001s : 3: predicate.reset_defer_inline 0.86% : 0.000003s : 15: predicate.reshape_eliminate 0.50% : 0.000002s : 6: predicate.row_tensor_add_zeros_like 0.23% : 0.000001s : 3: predicate.row_tensor_eliminate 0.77% : 0.000002s : 6: predicate.same_eliminate 0.32% : 0.000001s : 6: predicate.set_cell_output_no_recompute 0.71% : 0.000002s : 6: predicate.shard_identity_eliminate 0.43% : 0.000001s : 6: predicate.special_op_eliminate 0.61% : 0.000002s : 6: predicate.specialize_transform 0.69% : 0.000002s : 6: predicate.split_environ_get_set_with_tuple_value 0.63% : 0.000002s : 6: predicate.stack_unstack_eliminate 0.24% : 0.000001s : 3: predicate.switch_call_monad_eliminater 1.55% : 0.000005s : 26: predicate.switch_defer_inline 1.68% : 0.000005s : 32: predicate.switch_layer_defer_inline 4.96% : 0.000015s : 91: predicate.switch_simplify 0.79% : 0.000002s : 15: predicate.tile_eliminate 0.89% : 0.000003s : 15: predicate.transpose_eliminate 1.26% : 0.000004s : 21: predicate.tuple_list_convert_item_index_to_positive 1.50% : 0.000005s : 21: predicate.tuple_list_get_item_const_eliminator 1.27% : 0.000004s : 21: predicate.tuple_list_get_item_depend_reorder 2.91% : 0.000009s : 31: predicate.tuple_list_get_item_eliminator 1.14% : 0.000003s : 21: predicate.tuple_list_get_set_item_eliminator 11.37% : 0.000034s : 27: predicate.tuple_list_set_item_eliminator 1.59% : 0.000005s : 25: predicate.tuple_to_list_eliminator_ 2.04% : 0.000006s : 40: predicate.updatestate_pure_node_eliminater 2.33% : 0.000007s : 46: predicate.updatestate_useless_node_eliminater 0.21% : 0.000001s : 3: predicate.value_based_eliminate 0.43% : 0.000001s : 6: predicate.virtual_dataset_eliminate 0.41% : 0.000001s : 6: predicate.virtual_output_eliminate 0.17% : 0.000000s : 3: predicate.virtual_view_grad_eliminate 0.23% : 0.000001s : 3: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.044505 31 52.85% : 0.023523s : 22: func_graph_cloner_run.FuncGraphClonerGraph 47.15% : 0.020982s : 9: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.804269 192 0.00% : 0.000003s : 1: ForceFp32Comm 0.23% : 0.004163s : 1: add_attr 0.23% : 0.004147s : 1: add_attr_with_inline 0.00% : 0.000003s : 1: add_comm_op_reuse_tag 0.00% : 0.000069s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.01% : 0.000126s : 1: auto_monad 0.00% : 0.000034s : 1: auto_monad_reorder 0.00% : 0.000006s : 1: begin_end_overlap_inline 0.00% : 0.000008s : 1: bias_add_comm_swap 0.03% : 0.000586s : 1: bootstrap 0.00% : 0.000041s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000021s : 1: control_data_broadcast_order 0.00% : 0.000006s : 1: convert_after_rewriter 0.00% : 0.000035s : 1: cse_after_recomputation 0.00% : 0.000006s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000012s : 1: environ_conv 0.04% : 0.000781s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000013s : 1: graph_reusing 0.00% : 0.000006s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000006s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000016s : 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% : 0.000011s : 1: label_micro_interleaved_index 0.83% : 0.014906s : 1: loop_unroll 0.00% : 0.000006s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.05% : 0.000897s : 1: mutable_eliminate 0.16% : 0.002907s : 1: offloading_packed_experts 0.00% : 0.000041s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000027s : 1: opt.transform.mutable_eliminate 0.49% : 0.008814s : 78: opt.transform.opt_a 0.00% : 0.000034s : 1: opt.transform.opt_after_cconv 0.00% : 0.000033s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000107s : 28: opt.transform.opt_b 0.00% : 0.000087s : 2: opt.transform.opt_trans_graph 0.00% : 0.000070s : 4: opt.transform.symbol_engine_opt 4.68% : 0.084389s : 1: opt_a 0.01% : 0.000183s : 1: opt_after_cconv 0.05% : 0.000878s : 1: opt_after_jit_grad 0.01% : 0.000242s : 1: opt_b 5.81% : 0.104823s : 1: optimize 0.00% : 0.000025s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 1: order_py_execute_after_rewriter 0.00% : 0.000035s : 1: overlap_grad_flash_sp 0.00% : 0.000005s : 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% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000022s : 1: overlap_recompute_and_grad_model_parallel 0.00% : 0.000006s : 1: overlap_recompute_comm 0.00% : 0.000007s : 1: parallel-infer-symbol 0.00% : 0.000006s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000007s : 1: pipeline_split 3.57% : 0.064448s : 1: pre_auto_parallel 0.00% : 0.000018s : 1: py_interpret_to_execute 0.00% : 0.000015s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000024s : 1: remove_dup_value 2.43% : 0.043913s : 1: renormalize.infer 0.26% : 0.004692s : 1: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000012s : 1: rewriter_after_jit_bprop_graph 0.00% : 0.000031s : 1: rewriter_after_opt_a 0.02% : 0.000431s : 1: rewriter_before_opt_a 0.00% : 0.000005s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000004s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.01% : 0.000143s : 1: symbol_engine_optimizer 0.01% : 0.000121s : 1: tuple_transform 81.01% : 1.461634s : 1: type_inference TotalTime = 1.83718, [21] [bootstrap]: 0.00049607 [type_inference]: 1.49837 [event_method]: 0.00020155 [auto_monad]: 0.00018789 [graph_reusing]: 1.101e-05 [inline]: 3.07002e-06 [add_attr]: 0.0851213, [1] [add_attr_with_inline]: 0.0851039, [1] [Cycle 1]: 0.00013384, [2] [tag_attr]: 7.439e-05 [meta_addattr_fg_expand]: 1.344e-05 [parallel-infer-symbol]: 3.53e-06 [pre_auto_parallel]: 8.712e-05 [insert-virtual-dataset]: 2.46e-06 [parallel-infer-symbol-second]: 1.62001e-06 [dataset_repeat_opt]: 2.24001e-06 [pipeline_split]: 1.68002e-06 [optimize]: 0.251819, [53] [py_interpret_to_execute]: 1.098e-05 [rewriter_before_opt_a]: 0.0178734 [opt_a]: 0.231592, [3] [Cycle 1]: 0.207191, [45] [expand_dump_flag]: 1.115e-05 [switch_simplify]: 0.00035776 [loop_unroll]: 9.343e-05 [a_1]: 0.0250173 [with_stream_mark]: 5.499e-05 [recompute_prepare]: 4.155e-05 [updatestate_depend_eliminate]: 1.085e-05 [updatestate_assign_eliminate]: 3.381e-05 [updatestate_loads_eliminate]: 8.95999e-06 [parameter_eliminate]: 9.12001e-06 [a_2]: 0.00029743 [accelerated_algorithm]: 2.109e-05 [shard]: 2.78998e-06 [meta_shard_fg_expand]: 1.477e-05 [shard_inline]: 1.753e-05 [merge_send_recv]: 2.52e-05 [auto_parallel]: 2.13e-05 [parallel]: 2.534e-05 [flash_sp]: 1.561e-05 [merge_comm]: 9.62999e-06 [allreduce_fusion]: 8.38001e-06 [matmul_add_comm_reduction]: 3.817e-05 [allreduce_slice_to_reducescatter]: 1.29998e-06 [virtual_shard_identity]: 2.021e-05 [virtual_dataset]: 1.627e-05 [get_grad_eliminate_]: 1.554e-05 [virtual_output]: 1.529e-05 [merge_forward]: 1.168e-05 [cell_reuse_recompute_pass]: 3.71999e-06 [offload_activation]: 1.908e-05 [cell_reuse_handle_not_recompute_node_pass]: 3.269e-05 [merge_recompute_call_nodes]: 2.16e-06 [before_grad]: 2.717e-05 [set_forward_comm_id_for_comm_node_pass]: 1.013e-05 [meta_fg_expand]: 0.0024975 [flash_sp_send_recv_attached]: 4.70999e-06 [receive_attached]: 2.32001e-06 [after_resolve]: 8.79e-05 [a_after_grad]: 0.00010297 [renormalize]: 0.151821 [add_forward_monad_depend]: 1.925e-05 [auto_monad_grad]: 7.84002e-06 [auto_monad_eliminator]: 0.0212904 [cse]: 0.00064712 [a_3]: 0.00240086 [Cycle 2]: 0.0237593, [45] [expand_dump_flag]: 6.02001e-06 [switch_simplify]: 5.614e-05 [loop_unroll]: 4.428e-05 [a_1]: 0.0117357 [with_stream_mark]: 3.602e-05 [recompute_prepare]: 1.381e-05 [updatestate_depend_eliminate]: 4.60999e-06 [updatestate_assign_eliminate]: 3.2e-06 [updatestate_loads_eliminate]: 3.4e-06 [parameter_eliminate]: 2.69001e-06 [a_2]: 6.94e-05 [accelerated_algorithm]: 7.07002e-06 [shard]: 2.12001e-06 [meta_shard_fg_expand]: 3.26001e-06 [shard_inline]: 6.73e-06 [merge_send_recv]: 9.35001e-06 [auto_parallel]: 1.072e-05 [parallel]: 1.018e-05 [flash_sp]: 4.53001e-06 [merge_comm]: 3.55e-06 [allreduce_fusion]: 3.46999e-06 [matmul_add_comm_reduction]: 1.096e-05 [allreduce_slice_to_reducescatter]: 1.12e-06 [virtual_shard_identity]: 8.72e-06 [virtual_dataset]: 5.62001e-06 [get_grad_eliminate_]: 5.04e-06 [virtual_output]: 6.17999e-06 [merge_forward]: 4.13001e-06 [cell_reuse_recompute_pass]: 1.46002e-06 [offload_activation]: 1.097e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.856e-05 [merge_recompute_call_nodes]: 1.75001e-06 [before_grad]: 1.071e-05 [set_forward_comm_id_for_comm_node_pass]: 3.52002e-06 [meta_fg_expand]: 0.00035458 [flash_sp_send_recv_attached]: 3.00998e-06 [receive_attached]: 2.49999e-06 [after_resolve]: 1.704e-05 [a_after_grad]: 8.90999e-06 [renormalize]: 0.0107623 [add_forward_monad_depend]: 1.195e-05 [auto_monad_grad]: 2.58e-06 [auto_monad_eliminator]: 2.376e-05 [cse]: 4.242e-05 [a_3]: 6.069e-05 [Cycle 3]: 0.00061038, [45] [expand_dump_flag]: 2.26998e-06 [switch_simplify]: 8.27998e-06 [loop_unroll]: 5.91e-06 [a_1]: 8.581e-05 [with_stream_mark]: 1.752e-05 [recompute_prepare]: 5.57001e-06 [updatestate_depend_eliminate]: 4.09002e-06 [updatestate_assign_eliminate]: 3.11001e-06 [updatestate_loads_eliminate]: 2.79999e-06 [parameter_eliminate]: 2.21e-06 [a_2]: 6.302e-05 [accelerated_algorithm]: 5.99e-06 [shard]: 3.28e-06 [meta_shard_fg_expand]: 2.37999e-06 [shard_inline]: 5.86e-06 [merge_send_recv]: 8.62e-06 [auto_parallel]: 9.52999e-06 [parallel]: 9.67999e-06 [flash_sp]: 1.80001e-06 [merge_comm]: 3.7e-06 [allreduce_fusion]: 3.76999e-06 [matmul_add_comm_reduction]: 9.54e-06 [allreduce_slice_to_reducescatter]: 8.09989e-07 [virtual_shard_identity]: 6.60997e-06 [virtual_dataset]: 5.40001e-06 [get_grad_eliminate_]: 5.21998e-06 [virtual_output]: 5.34998e-06 [merge_forward]: 4.28999e-06 [cell_reuse_recompute_pass]: 4.06001e-06 [offload_activation]: 9.77001e-06 [cell_reuse_handle_not_recompute_node_pass]: 1.613e-05 [merge_recompute_call_nodes]: 1.52001e-06 [before_grad]: 9.72001e-06 [set_forward_comm_id_for_comm_node_pass]: 3.57002e-06 [meta_fg_expand]: 2.56e-06 [flash_sp_send_recv_attached]: 2.34001e-06 [receive_attached]: 2.58e-06 [after_resolve]: 5.38002e-06 [a_after_grad]: 7.3e-06 [renormalize]: 7.99773e-08 [add_forward_monad_depend]: 1.50999e-06 [auto_monad_grad]: 1.13001e-06 [auto_monad_eliminator]: 7.54002e-06 [cse]: 1.627e-05 [a_3]: 3.133e-05 [py_interpret_to_execute_after_opt_a]: 1.152e-05 [slice_cell_reuse_recomputed_activation]: 2.35002e-06 [rewriter_after_opt_a]: 2.44e-05 [convert_after_rewriter]: 1.44998e-06 [order_py_execute_after_rewriter]: 1.38002e-06 [mutable_eliminate]: 0.00079043 [opt_b]: 0.00020349, [1] [Cycle 1]: 0.0001947, [7] [b_1]: 0.00011218 [b_2]: 7.08e-06 [updatestate_depend_eliminate]: 8.15e-06 [updatestate_assign_eliminate]: 2.55002e-06 [updatestate_loads_eliminate]: 2.24001e-06 [renormalize]: 7.2e-07 [cse]: 2.616e-05 [optimize_parallel_all_gather_comm]: 1.791e-05 [overlap_param_gather]: 2.04e-06 [cconv]: 3.343e-05 [loop_unroll]: 0.00049234 [opt_after_cconv]: 0.00010323, [1] [Cycle 1]: 9.624e-05, [7] [c_1]: 2.376e-05 [parameter_eliminate]: 5.27999e-06 [updatestate_depend_eliminate]: 5.84999e-06 [updatestate_assign_eliminate]: 2.32999e-06 [updatestate_loads_eliminate]: 2.21e-06 [cse]: 2.077e-05 [renormalize]: 9.70002e-07 [remove_dup_value]: 1.734e-05 [tuple_transform]: 6.496e-05, [1] [Cycle 1]: 6.059e-05, [4] [d_1]: 3.555e-05 [none_parameter_eliminate]: 1.67999e-06 [renormalize]: 1.30007e-07 [switch_simplify]: 6.44999e-06 [partial_unused_args_eliminate]: 2.19001e-06 [add_recomputation]: 5.328e-05 [cse_after_recomputation]: 2.62e-05, [1] [Cycle 1]: 2.066e-05, [1] [cse]: 1.517e-05 [environ_conv]: 7.46999e-06 [swap_dp_allreduce_reducescatter]: 5.02999e-06 [bias_add_comm_swap]: 3.2e-06 [label_micro_interleaved_index]: 5.03002e-06 [label_fine_grained_interleaved_index]: 2.75997e-06 [merge_cast_opt]: 1.69e-06 [slice_recompute_activation]: 2.46e-06 [micro_interleaved_order_control]: 2.78e-06 [assign_add_opt]: 1.25999e-06 [ForceFp32Comm]: 8.49977e-07 [remove_cast_before_assign_add]: 1.00001e-06 [full_micro_interleaved_order_control]: 2.22001e-06 [reorder_send_recv_between_fp_bp]: 2.91999e-06 [comm_op_add_attrs]: 1.00001e-06 [add_comm_op_reuse_tag]: 1.18001e-06 [interleave_split_concat_branches]: 1.43002e-06 [interleave_parallel_branches]: 1.12e-06 [overlap_opt_shard_in_pipeline]: 1.29e-06 [overlap_opt_shard_grad_in_pipeline]: 1.96e-06 [control_data_broadcast_order]: 1.332e-05 [grouped_pairwise_exchange_alltoall]: 1.42e-06 [offloading_packed_experts]: 5.39e-06 [overlap_recompute_and_grad_model_parallel]: 5.99e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.15999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.22999e-06 [overlap_recompute_comm]: 2.27001e-06 [overlap_grad_ring_attention]: 4.16001e-06 [overlap_grad_flash_sp]: 2.043e-05 [begin_end_overlap_inline]: 4.69998e-07 [split_matmul_comm_elemetwise]: 2.18998e-06 [split_layernorm_comm]: 1.92001e-06 [handle_group_info]: 1.12e-06 [symbol_engine_optimizer]: 0.0001056, [1] [Cycle 1]: 0.00010071, [6] [build]: 3.5e-06 [elim_shapecalc]: 9.20999e-06 [elim_not_effective]: 1.191e-05 [opt_reshape]: 5.68997e-06 [fold_const_symbol]: 3.872e-05 [renormalize]: 2.40019e-07 [detach_backward]: 2.91999e-06 [pipeline_parallel_scheduler]: 2.09999e-06 [auto_monad_reorder]: 1.972e-05 [get_jit_bprop_graph]: 2.39999e-06 [rewriter_after_jit_bprop_graph]: 6.63e-06 [opt_after_jit_grad]: 0.00056688 [validate]: 5.1e-05 Sums bootstrap : 0.000496s : 0.03% type_inference : 1.498375s : 85.67% event_method : 0.000202s : 0.01% auto_monad : 0.000188s : 0.01% graph_reusing : 0.000011s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000074s : 0.00% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000013s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000087s : 0.00% insert-virtual-dataset : 0.000002s : 0.00% parallel-infer-symbol-second : 0.000002s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000011s : 0.00% optimize.rewriter_before_opt_a : 0.017873s : 1.02% optimize.opt_a.expand_dump_flag : 0.000019s : 0.00% optimize.opt_a.switch_simplify : 0.000422s : 0.02% optimize.opt_a.loop_unroll : 0.000144s : 0.01% optimize.opt_a.a_1 : 0.036839s : 2.11% optimize.opt_a.with_stream_mark : 0.000109s : 0.01% optimize.opt_a.recompute_prepare : 0.000061s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000020s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000040s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000015s : 0.00% optimize.opt_a.parameter_eliminate : 0.000014s : 0.00% optimize.opt_a.a_2 : 0.000430s : 0.02% optimize.opt_a.accelerated_algorithm : 0.000034s : 0.00% optimize.opt_a.shard : 0.000008s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000020s : 0.00% optimize.opt_a.shard_inline : 0.000030s : 0.00% optimize.opt_a.merge_send_recv : 0.000043s : 0.00% optimize.opt_a.auto_parallel : 0.000042s : 0.00% optimize.opt_a.parallel : 0.000045s : 0.00% optimize.opt_a.flash_sp : 0.000022s : 0.00% optimize.opt_a.merge_comm : 0.000017s : 0.00% optimize.opt_a.allreduce_fusion : 0.000016s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000059s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000003s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000036s : 0.00% optimize.opt_a.virtual_dataset : 0.000027s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000026s : 0.00% optimize.opt_a.virtual_output : 0.000027s : 0.00% optimize.opt_a.merge_forward : 0.000020s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000009s : 0.00% optimize.opt_a.offload_activation : 0.000040s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000067s : 0.00% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000048s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000017s : 0.00% optimize.opt_a.meta_fg_expand : 0.002855s : 0.16% optimize.opt_a.flash_sp_send_recv_attached : 0.000010s : 0.00% optimize.opt_a.receive_attached : 0.000007s : 0.00% optimize.opt_a.after_resolve : 0.000110s : 0.01% optimize.opt_a.a_after_grad : 0.000119s : 0.01% optimize.opt_a.renormalize : 0.162584s : 9.30% optimize.opt_a.add_forward_monad_depend : 0.000033s : 0.00% optimize.opt_a.auto_monad_grad : 0.000012s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.021322s : 1.22% optimize.opt_a.cse : 0.000706s : 0.04% optimize.opt_a.a_3 : 0.002493s : 0.14% optimize.py_interpret_to_execute_after_opt_a : 0.000012s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000024s : 0.00% optimize.convert_after_rewriter : 0.000001s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000790s : 0.05% optimize.opt_b.b_1 : 0.000112s : 0.01% optimize.opt_b.b_2 : 0.000007s : 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 : 0.000018s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000033s : 0.00% optimize.loop_unroll : 0.000492s : 0.03% optimize.opt_after_cconv.c_1 : 0.000024s : 0.00% 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 : 0.000002s : 0.00% optimize.opt_after_cconv.updatestate_loads_eliminate : 0.000002s : 0.00% optimize.opt_after_cconv.cse : 0.000021s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000017s : 0.00% optimize.tuple_transform.d_1 : 0.000036s : 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 : 0.000006s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000053s : 0.00% optimize.cse_after_recomputation.cse : 0.000015s : 0.00% optimize.environ_conv : 0.000007s : 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 : 0.000002s : 0.00% optimize.slice_recompute_activation : 0.000002s : 0.00% optimize.micro_interleaved_order_control : 0.000003s : 0.00% optimize.assign_add_opt : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000013s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000001s : 0.00% optimize.offloading_packed_experts : 0.000005s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000006s : 0.00% optimize.overlap_grad_matmul_and_grad_allreduce : 0.000001s : 0.00% optimize.overlap_recompute_allgather_and_fa_grad : 0.000001s : 0.00% optimize.overlap_recompute_comm : 0.000002s : 0.00% optimize.overlap_grad_ring_attention : 0.000004s : 0.00% optimize.overlap_grad_flash_sp : 0.000020s : 0.00% optimize.begin_end_overlap_inline : 0.000000s : 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 : 0.000009s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000012s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000006s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000039s : 0.00% 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 : 0.000020s : 0.00% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000007s : 0.00% opt_after_jit_grad : 0.000567s : 0.03% validate : 0.000051s : 0.00% Time group info: ------[substitution.] 0.028794 163 0.01% : 0.000002s : 2: substitution.elim_not_effective 0.05% : 0.000015s : 12: substitution.float_depend_g_call 0.02% : 0.000005s : 2: substitution.float_tuple_getitem_switch 0.01% : 0.000002s : 2: substitution.fold_const_symbol 0.02% : 0.000006s : 2: substitution.graph_param_transform 0.02% : 0.000006s : 2: substitution.incorporate_call 0.01% : 0.000002s : 2: substitution.incorporate_call_switch 37.88% : 0.010908s : 23: substitution.inline 0.09% : 0.000025s : 2: substitution.inline_without_move 0.04% : 0.000011s : 12: substitution.j_node_and_user_rematch 0.05% : 0.000014s : 7: substitution.minmaximum_grad 0.05% : 0.000015s : 12: substitution.partial_eliminate 0.04% : 0.000012s : 12: substitution.remove_not_recompute_node 0.23% : 0.000066s : 9: substitution.replace_applicator 0.03% : 0.000009s : 11: substitution.replace_old_param 0.03% : 0.000008s : 1: substitution.set_cell_output_no_recompute 0.08% : 0.000023s : 3: substitution.switch_simplify 0.11% : 0.000032s : 7: substitution.tuple_list_convert_item_index_to_positive 0.04% : 0.000012s : 7: substitution.tuple_list_get_item_const_eliminator 0.07% : 0.000019s : 7: substitution.tuple_list_get_item_depend_reorder 61.07% : 0.017585s : 19: substitution.tuple_list_get_item_eliminator 0.06% : 0.000017s : 7: substitution.tuple_list_get_set_item_eliminator ------[type_inference.] 1.498176 2 99.54% : 1.491273s : 1: type_inference.infer 0.46% : 0.006903s : 1: type_inference.specialize ------[replace.] 0.000657 36 47.80% : 0.000314s : 23: replace.inline 13.18% : 0.000087s : 3: replace.switch_simplify 39.02% : 0.000256s : 10: replace.tuple_list_get_item_eliminator ------[match.] 0.010976 36 99.21% : 0.010890s : 23: match.inline 0.16% : 0.000018s : 3: match.switch_simplify 0.63% : 0.000069s : 10: match.tuple_list_get_item_eliminator ------[predicate.] 0.000775 4067 1.18% : 0.000009s : 53: predicate.accumulaten_eliminater 0.21% : 0.000002s : 2: predicate.ad_related_special_op_eliminate 0.33% : 0.000003s : 16: predicate.addn_check_dump 1.08% : 0.000008s : 53: predicate.addn_zero_filter 0.84% : 0.000006s : 53: predicate.adjust_all_reduce_mul_add 2.17% : 0.000017s : 69: predicate.arithmetic_simplify 1.28% : 0.000010s : 53: predicate.cast_eliminate 0.77% : 0.000006s : 43: predicate.check_bprop_eliminate 0.26% : 0.000002s : 16: predicate.compare_switch_simplify 0.02% : 0.000000s : 2: predicate.const_output_eliminate 0.35% : 0.000003s : 16: predicate.depend_value_elim 1.00% : 0.000008s : 53: predicate.dict_get_item_const_eliminator 1.28% : 0.000010s : 53: predicate.dict_get_item_eliminator 1.15% : 0.000009s : 53: predicate.dict_set_item_eliminator 0.23% : 0.000002s : 4: predicate.dumpgradient_eliminate 0.03% : 0.000000s : 2: predicate.elim_not_effective 0.11% : 0.000001s : 2: predicate.elim_shapecalc_of_broadcastargs 1.00% : 0.000008s : 55: predicate.environ_add_const_eliminate 0.94% : 0.000007s : 55: predicate.environ_get_add_eliminate 1.00% : 0.000008s : 55: predicate.environ_get_depend_swap 1.46% : 0.000011s : 71: predicate.environ_get_eliminate 1.01% : 0.000008s : 55: predicate.environ_get_set_eliminate 1.57% : 0.000012s : 86: predicate.exchange_switch_depend_value 2.85% : 0.000022s : 86: predicate.float_depend_g_call 0.34% : 0.000003s : 16: predicate.float_environ_get_switch 0.41% : 0.000003s : 18: predicate.float_tuple_getitem_switch 0.02% : 0.000000s : 2: predicate.fold_const_symbol 0.41% : 0.000003s : 16: predicate.get_grad_eliminate 0.06% : 0.000000s : 2: predicate.graph_param_transform 0.37% : 0.000003s : 16: predicate.incorporate_call 0.27% : 0.000002s : 16: predicate.incorporate_call_switch 5.29% : 0.000041s : 175: predicate.inline 1.25% : 0.000010s : 46: predicate.inline_without_move 0.15% : 0.000001s : 16: predicate.j_node_and_user_rematch 0.54% : 0.000004s : 16: predicate.less_batch_normalization 1.66% : 0.000013s : 67: predicate.list_to_tuple_eliminator_ 2.19% : 0.000017s : 120: predicate.load_eliminater 0.22% : 0.000002s : 2: predicate.loop_unroll_after_grad 2.68% : 0.000021s : 146: predicate.loop_unroll_before_grad 1.55% : 0.000012s : 57: predicate.make_slice_get_slice_eliminator 0.32% : 0.000002s : 16: predicate.merge_addn 0.75% : 0.000006s : 43: predicate.micro_step_allgather_replace 0.77% : 0.000006s : 43: predicate.mini_step_allgather_replace 0.92% : 0.000007s : 53: predicate.minmaximum_grad 0.42% : 0.000003s : 2: predicate.mutable_eliminate 0.06% : 0.000000s : 2: predicate.opt_reshape 0.07% : 0.000001s : 2: predicate.parallel_virtual_node 2.99% : 0.000023s : 86: predicate.partial_defer_inline 1.34% : 0.000010s : 65: predicate.partial_eliminate 1.06% : 0.000008s : 53: predicate.print_const_string_wrapper 0.35% : 0.000003s : 16: predicate.reduce_all_const_elim 1.52% : 0.000012s : 53: predicate.reduce_eliminate 2.22% : 0.000017s : 120: predicate.redundant_stop_gradient_eliminater 0.29% : 0.000002s : 16: predicate.remove_not_recompute_node 1.81% : 0.000014s : 106: predicate.replace_applicator 0.67% : 0.000005s : 46: predicate.replace_old_param 0.07% : 0.000001s : 2: predicate.reset_defer_inline 1.09% : 0.000008s : 53: predicate.reshape_eliminate 1.40% : 0.000011s : 43: predicate.row_tensor_add_zeros_like 0.06% : 0.000000s : 2: predicate.row_tensor_eliminate 1.12% : 0.000009s : 43: predicate.same_eliminate 0.25% : 0.000002s : 16: predicate.set_cell_output_no_recompute 0.45% : 0.000003s : 16: predicate.shard_identity_eliminate 0.13% : 0.000001s : 4: predicate.special_op_eliminate 0.68% : 0.000005s : 16: predicate.specialize_transform 1.03% : 0.000008s : 43: predicate.split_environ_get_set_with_tuple_value 1.04% : 0.000008s : 46: predicate.stack_unstack_eliminate 0.07% : 0.000001s : 2: predicate.switch_call_monad_eliminater 1.79% : 0.000014s : 86: predicate.switch_defer_inline 2.69% : 0.000021s : 129: predicate.switch_layer_defer_inline 13.95% : 0.000108s : 256: predicate.switch_simplify 1.03% : 0.000008s : 53: predicate.tile_eliminate 1.11% : 0.000009s : 53: predicate.transpose_eliminate 1.20% : 0.000009s : 57: predicate.tuple_list_convert_item_index_to_positive 1.61% : 0.000012s : 57: predicate.tuple_list_get_item_const_eliminator 1.20% : 0.000009s : 57: predicate.tuple_list_get_item_depend_reorder 2.96% : 0.000023s : 83: predicate.tuple_list_get_item_eliminator 1.25% : 0.000010s : 57: predicate.tuple_list_get_set_item_eliminator 1.86% : 0.000014s : 73: predicate.tuple_list_set_item_eliminator 1.33% : 0.000010s : 67: predicate.tuple_to_list_eliminator_ 2.09% : 0.000016s : 120: predicate.updatestate_pure_node_eliminater 2.47% : 0.000019s : 136: predicate.updatestate_useless_node_eliminater 0.07% : 0.000001s : 2: predicate.value_based_eliminate 0.37% : 0.000003s : 16: predicate.virtual_dataset_eliminate 0.37% : 0.000003s : 16: predicate.virtual_output_eliminate 0.05% : 0.000000s : 2: predicate.virtual_view_grad_eliminate 0.10% : 0.000001s : 2: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.008253 68 72.99% : 0.006024s : 41: func_graph_cloner_run.FuncGraphClonerGraph 27.01% : 0.002229s : 27: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 2.377480 233 0.00% : 0.000004s : 1: ForceFp32Comm 3.58% : 0.085129s : 1: add_attr 3.58% : 0.085110s : 1: add_attr_with_inline 0.00% : 0.000005s : 1: add_comm_op_reuse_tag 0.00% : 0.000058s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.01% : 0.000199s : 1: auto_monad 0.00% : 0.000024s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.02% : 0.000533s : 1: bootstrap 0.00% : 0.000037s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000016s : 1: control_data_broadcast_order 0.00% : 0.000004s : 1: convert_after_rewriter 0.00% : 0.000029s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000010s : 1: environ_conv 0.01% : 0.000216s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000016s : 1: graph_reusing 0.00% : 0.000004s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000005s : 1: handle_group_info 0.00% : 0.000006s : 1: inline 0.00% : 0.000006s : 1: insert-virtual-dataset 0.00% : 0.000004s : 1: interleave_parallel_branches 0.00% : 0.000004s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.02% : 0.000503s : 1: loop_unroll 0.00% : 0.000005s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.03% : 0.000805s : 1: mutable_eliminate 0.00% : 0.000008s : 1: offloading_packed_experts 0.00% : 0.000015s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000022s : 1: opt.transform.mutable_eliminate 1.71% : 0.040699s : 117: opt.transform.opt_a 0.00% : 0.000023s : 1: opt.transform.opt_after_cconv 0.00% : 0.000023s : 1: opt.transform.opt_after_jit_grad 0.00% : 0.000090s : 28: opt.transform.opt_b 0.00% : 0.000040s : 2: opt.transform.opt_trans_graph 0.00% : 0.000062s : 4: opt.transform.symbol_engine_opt 9.74% : 0.231598s : 1: opt_a 0.00% : 0.000107s : 1: opt_after_cconv 0.02% : 0.000581s : 1: opt_after_jit_grad 0.01% : 0.000207s : 1: opt_b 10.59% : 0.251826s : 1: optimize 0.00% : 0.000022s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 1: order_py_execute_after_rewriter 0.00% : 0.000024s : 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% : 0.000004s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000010s : 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% : 0.000005s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.00% : 0.000093s : 1: pre_auto_parallel 0.00% : 0.000015s : 1: py_interpret_to_execute 0.00% : 0.000015s : 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 6.04% : 0.143546s : 2: renormalize.infer 0.80% : 0.018998s : 2: 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% : 0.000027s : 1: rewriter_after_opt_a 0.75% : 0.017913s : 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% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000008s : 1: swap_dp_allreduce_reducescatter 0.00% : 0.000109s : 1: symbol_engine_optimizer 0.00% : 0.000068s : 1: tuple_transform 63.03% : 1.498414s : 1: type_inference . [hook] pytest_runtest_teardown:test_repeat_interleave_forward_backward_tensor[False-0-KBK] tests/st/mint/test_repeat_interleave.py::test_repeat_interleave_forward_backward_tensor[False-0-KBK],max_mem:20.0M . [hook] pytest_runtest_teardown:test_repeat_interleave_forward_backward_tensor[False-None-pynative] tests/st/mint/test_repeat_interleave.py::test_repeat_interleave_forward_backward_tensor[False-None-pynative],max_mem:20.0M TotalTime = 1.65532, [21] [bootstrap]: 0.00070116 [type_inference]: 1.48175 [event_method]: 0.00059005 [auto_monad]: 0.00010347 [graph_reusing]: 8.60001e-06 [inline]: 3.78001e-06 [add_attr]: 0.0047697, [1] [add_attr_with_inline]: 0.00474808, [1] [Cycle 1]: 0.0001064, [2] [tag_attr]: 4.55e-05 [meta_addattr_fg_expand]: 8.00999e-06 [parallel-infer-symbol]: 4.03999e-06 [pre_auto_parallel]: 6.65e-05 [insert-virtual-dataset]: 4.35e-06 [parallel-infer-symbol-second]: 8.80013e-07 [dataset_repeat_opt]: 1.86e-06 [pipeline_split]: 3.01001e-06 [optimize]: 0.165945, [53] [py_interpret_to_execute]: 8.95001e-06 [rewriter_before_opt_a]: 0.00065079 [opt_a]: 0.129045, [2] [Cycle 1]: 0.128289, [45] [expand_dump_flag]: 4.41002e-06 [switch_simplify]: 7.102e-05 [loop_unroll]: 4.843e-05 [a_1]: 0.0847752 [with_stream_mark]: 3.182e-05 [recompute_prepare]: 1.567e-05 [updatestate_depend_eliminate]: 5.10001e-06 [updatestate_assign_eliminate]: 3.69002e-06 [updatestate_loads_eliminate]: 3.09999e-06 [parameter_eliminate]: 2.74001e-06 [a_2]: 9.04e-05 [accelerated_algorithm]: 8.01001e-06 [shard]: 2.04999e-06 [meta_shard_fg_expand]: 3.01999e-06 [shard_inline]: 6.60002e-06 [merge_send_recv]: 8.89e-06 [auto_parallel]: 1.171e-05 [parallel]: 3.098e-05 [flash_sp]: 1.189e-05 [merge_comm]: 4.25999e-06 [allreduce_fusion]: 3.78001e-06 [matmul_add_comm_reduction]: 1.046e-05 [allreduce_slice_to_reducescatter]: 9.79984e-07 [virtual_shard_identity]: 8.62998e-06 [virtual_dataset]: 7.58999e-06 [get_grad_eliminate_]: 7.23999e-06 [virtual_output]: 7.5e-06 [merge_forward]: 4.27e-06 [cell_reuse_recompute_pass]: 1.71998e-06 [offload_activation]: 1.07e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.663e-05 [merge_recompute_call_nodes]: 1.79e-06 [before_grad]: 1.087e-05 [set_forward_comm_id_for_comm_node_pass]: 3.9e-06 [meta_fg_expand]: 3.40998e-06 [flash_sp_send_recv_attached]: 3.45e-06 [receive_attached]: 2.24001e-06 [after_resolve]: 1.196e-05 [a_after_grad]: 1.039e-05 [renormalize]: 0.0391278 [add_forward_monad_depend]: 1.962e-05 [auto_monad_grad]: 3.33998e-06 [auto_monad_eliminator]: 4.242e-05 [cse]: 5.788e-05 [a_3]: 8.072e-05 [Cycle 2]: 0.00073847, [45] [expand_dump_flag]: 2.63003e-06 [switch_simplify]: 8.66997e-06 [loop_unroll]: 6.93998e-06 [a_1]: 0.00012706 [with_stream_mark]: 2.528e-05 [recompute_prepare]: 6.36e-06 [updatestate_depend_eliminate]: 5.46998e-06 [updatestate_assign_eliminate]: 3.56999e-06 [updatestate_loads_eliminate]: 3.23e-06 [parameter_eliminate]: 2.29001e-06 [a_2]: 7.159e-05 [accelerated_algorithm]: 7.25998e-06 [shard]: 3.39001e-06 [meta_shard_fg_expand]: 3.36999e-06 [shard_inline]: 5.27001e-06 [merge_send_recv]: 9.69e-06 [auto_parallel]: 1.041e-05 [parallel]: 1.076e-05 [flash_sp]: 3.81999e-06 [merge_comm]: 3.95e-06 [allreduce_fusion]: 3.46999e-06 [matmul_add_comm_reduction]: 1.045e-05 [allreduce_slice_to_reducescatter]: 1.03001e-06 [virtual_shard_identity]: 8.05999e-06 [virtual_dataset]: 6.46e-06 [get_grad_eliminate_]: 6.06e-06 [virtual_output]: 5.82001e-06 [merge_forward]: 4.99003e-06 [cell_reuse_recompute_pass]: 3.85998e-06 [offload_activation]: 1.085e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.299e-05 [merge_recompute_call_nodes]: 1.51002e-06 [before_grad]: 1.025e-05 [set_forward_comm_id_for_comm_node_pass]: 3.90998e-06 [meta_fg_expand]: 3.09001e-06 [flash_sp_send_recv_attached]: 1.97001e-06 [receive_attached]: 2.69001e-06 [after_resolve]: 1.207e-05 [a_after_grad]: 1.002e-05 [renormalize]: 5.00004e-08 [add_forward_monad_depend]: 1.97999e-06 [auto_monad_grad]: 1.53002e-06 [auto_monad_eliminator]: 9.03002e-06 [cse]: 2.167e-05 [a_3]: 3.544e-05 [py_interpret_to_execute_after_opt_a]: 1.247e-05 [slice_cell_reuse_recomputed_activation]: 2.33998e-06 [rewriter_after_opt_a]: 2.573e-05 [convert_after_rewriter]: 2.19999e-06 [order_py_execute_after_rewriter]: 1.37e-06 [mutable_eliminate]: 0.0185969 [opt_b]: 0.00027283, [1] [Cycle 1]: 0.00026066, [7] [b_1]: 0.00013734 [b_2]: 9.68997e-06 [updatestate_depend_eliminate]: 1.352e-05 [updatestate_assign_eliminate]: 3.88999e-06 [updatestate_loads_eliminate]: 3.15002e-06 [renormalize]: 7.99977e-07 [cse]: 5.335e-05 [optimize_parallel_all_gather_comm]: 2.522e-05 [overlap_param_gather]: 2.13998e-06 [cconv]: 4.221e-05 [loop_unroll]: 0.00062354 [opt_after_cconv]: 0.0158053, [1] [Cycle 1]: 0.0157941, [7] [c_1]: 3.181e-05 [parameter_eliminate]: 7.25e-06 [updatestate_depend_eliminate]: 9.75002e-06 [updatestate_assign_eliminate]: 3.03e-06 [updatestate_loads_eliminate]: 2.54001e-06 [cse]: 0.01565 [renormalize]: 1.43002e-06 [remove_dup_value]: 6.106e-05 [tuple_transform]: 0.00012777, [1] [Cycle 1]: 0.00011579, [4] [d_1]: 7.961e-05 [none_parameter_eliminate]: 5.36002e-06 [renormalize]: 2.80008e-07 [switch_simplify]: 8.88002e-06 [partial_unused_args_eliminate]: 2.07999e-06 [add_recomputation]: 7.877e-05 [cse_after_recomputation]: 3.865e-05, [1] [Cycle 1]: 3.27e-05, [1] [cse]: 2.597e-05 [environ_conv]: 1.006e-05 [swap_dp_allreduce_reducescatter]: 6.83e-06 [bias_add_comm_swap]: 3.61001e-06 [label_micro_interleaved_index]: 7.83999e-06 [label_fine_grained_interleaved_index]: 2.81999e-06 [merge_cast_opt]: 1.82001e-06 [slice_recompute_activation]: 2.12001e-06 [micro_interleaved_order_control]: 2.37999e-06 [assign_add_opt]: 2.17001e-06 [ForceFp32Comm]: 8.50006e-07 [remove_cast_before_assign_add]: 1.57001e-06 [full_micro_interleaved_order_control]: 2.16e-06 [reorder_send_recv_between_fp_bp]: 3.04999e-06 [comm_op_add_attrs]: 1.71998e-06 [add_comm_op_reuse_tag]: 1.02998e-06 [interleave_split_concat_branches]: 1.56002e-06 [interleave_parallel_branches]: 1.02e-06 [overlap_opt_shard_in_pipeline]: 1.25001e-06 [overlap_opt_shard_grad_in_pipeline]: 2.06e-06 [control_data_broadcast_order]: 2.003e-05 [grouped_pairwise_exchange_alltoall]: 1.91e-06 [offloading_packed_experts]: 4.58999e-06 [overlap_recompute_and_grad_model_parallel]: 4.75001e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.19998e-06 [overlap_recompute_allgather_and_fa_grad]: 1.27999e-06 [overlap_recompute_comm]: 2.22999e-06 [overlap_grad_ring_attention]: 5.94e-06 [overlap_grad_flash_sp]: 2.415e-05 [begin_end_overlap_inline]: 4.80009e-07 [split_matmul_comm_elemetwise]: 1.99e-06 [split_layernorm_comm]: 1.66e-06 [handle_group_info]: 9.79984e-07 [symbol_engine_optimizer]: 9.234e-05, [1] [Cycle 1]: 8.738e-05, [6] [build]: 4.02e-06 [elim_shapecalc]: 1.63e-05 [elim_not_effective]: 1.535e-05 [opt_reshape]: 7.86001e-06 [fold_const_symbol]: 1.071e-05 [renormalize]: 4.40021e-07 [detach_backward]: 2.79001e-06 [pipeline_parallel_scheduler]: 1.41998e-06 [auto_monad_reorder]: 2.204e-05 [get_jit_bprop_graph]: 3.5e-06 [rewriter_after_jit_bprop_graph]: 6.36998e-06 [opt_after_jit_grad]: 0.00101213 [validate]: 5.714e-05 Sums bootstrap : 0.000701s : 0.04% type_inference : 1.481749s : 90.03% event_method : 0.000590s : 0.04% auto_monad : 0.000103s : 0.01% graph_reusing : 0.000009s : 0.00% inline : 0.000004s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000045s : 0.00% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000008s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000067s : 0.00% insert-virtual-dataset : 0.000004s : 0.00% parallel-infer-symbol-second : 0.000001s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000003s : 0.00% optimize.py_interpret_to_execute : 0.000009s : 0.00% optimize.rewriter_before_opt_a : 0.000651s : 0.04% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000080s : 0.00% optimize.opt_a.loop_unroll : 0.000055s : 0.00% optimize.opt_a.a_1 : 0.084902s : 5.16% optimize.opt_a.with_stream_mark : 0.000057s : 0.00% optimize.opt_a.recompute_prepare : 0.000022s : 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 : 0.000006s : 0.00% optimize.opt_a.parameter_eliminate : 0.000005s : 0.00% optimize.opt_a.a_2 : 0.000162s : 0.01% optimize.opt_a.accelerated_algorithm : 0.000015s : 0.00% optimize.opt_a.shard : 0.000005s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000006s : 0.00% optimize.opt_a.shard_inline : 0.000012s : 0.00% optimize.opt_a.merge_send_recv : 0.000019s : 0.00% optimize.opt_a.auto_parallel : 0.000022s : 0.00% optimize.opt_a.parallel : 0.000042s : 0.00% optimize.opt_a.flash_sp : 0.000016s : 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 : 0.000021s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000002s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000017s : 0.00% optimize.opt_a.virtual_dataset : 0.000014s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 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 : 0.000006s : 0.00% optimize.opt_a.offload_activation : 0.000022s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000040s : 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 : 0.000008s : 0.00% 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 : 0.000005s : 0.00% optimize.opt_a.after_resolve : 0.000024s : 0.00% optimize.opt_a.a_after_grad : 0.000020s : 0.00% optimize.opt_a.renormalize : 0.039128s : 2.38% optimize.opt_a.add_forward_monad_depend : 0.000022s : 0.00% optimize.opt_a.auto_monad_grad : 0.000005s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000051s : 0.00% optimize.opt_a.cse : 0.000080s : 0.00% optimize.opt_a.a_3 : 0.000116s : 0.01% optimize.py_interpret_to_execute_after_opt_a : 0.000012s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000026s : 0.00% optimize.convert_after_rewriter : 0.000002s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.018597s : 1.13% optimize.opt_b.b_1 : 0.000137s : 0.01% optimize.opt_b.b_2 : 0.000010s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000014s : 0.00% 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 : 0.000053s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000025s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000042s : 0.00% optimize.loop_unroll : 0.000624s : 0.04% optimize.opt_after_cconv.c_1 : 0.000032s : 0.00% optimize.opt_after_cconv.parameter_eliminate : 0.000007s : 0.00% optimize.opt_after_cconv.updatestate_depend_eliminate : 0.000010s : 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 : 0.015650s : 0.95% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000061s : 0.00% optimize.tuple_transform.d_1 : 0.000080s : 0.00% optimize.tuple_transform.none_parameter_eliminate : 0.000005s : 0.00% optimize.tuple_transform.renormalize : 0.000000s : 0.00% optimize.tuple_transform.switch_simplify : 0.000009s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000079s : 0.00% optimize.cse_after_recomputation.cse : 0.000026s : 0.00% optimize.environ_conv : 0.000010s : 0.00% optimize.swap_dp_allreduce_reducescatter : 0.000007s : 0.00% optimize.bias_add_comm_swap : 0.000004s : 0.00% optimize.label_micro_interleaved_index : 0.000008s : 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 : 0.000002s : 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 : 0.000002s : 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 : 0.000020s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000002s : 0.00% optimize.offloading_packed_experts : 0.000005s : 0.00% optimize.overlap_recompute_and_grad_model_parallel : 0.000005s : 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 : 0.000006s : 0.00% optimize.overlap_grad_flash_sp : 0.000024s : 0.00% optimize.begin_end_overlap_inline : 0.000000s : 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 : 0.000016s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000015s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000008s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000011s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000001s : 0.00% auto_monad_reorder : 0.000022s : 0.00% get_jit_bprop_graph : 0.000003s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.001012s : 0.06% validate : 0.000057s : 0.00% Time group info: ------[substitution.] 0.000517 28 0.40% : 0.000002s : 2: substitution.elim_not_effective 0.43% : 0.000002s : 2: substitution.fold_const_symbol 1.85% : 0.000010s : 3: substitution.graph_param_transform 89.41% : 0.000462s : 7: substitution.inline 0.87% : 0.000004s : 4: substitution.j_node_and_user_rematch 1.26% : 0.000006s : 4: substitution.remove_not_recompute_node 1.31% : 0.000007s : 2: substitution.replace_old_param 4.46% : 0.000023s : 4: substitution.tuple_list_get_item_eliminator ------[type_inference.] 1.481588 2 99.74% : 1.477721s : 1: type_inference.infer 0.26% : 0.003867s : 1: type_inference.specialize ------[replace.] 0.000150 11 70.90% : 0.000107s : 7: replace.inline 29.10% : 0.000044s : 4: replace.tuple_list_get_item_eliminator ------[match.] 0.000475 11 95.76% : 0.000455s : 7: match.inline 4.24% : 0.000020s : 4: match.tuple_list_get_item_eliminator ------[predicate.] 0.000277 1305 1.21% : 0.000003s : 15: predicate.accumulaten_eliminater 1.36% : 0.000004s : 3: predicate.ad_related_special_op_eliminate 0.34% : 0.000001s : 6: predicate.addn_check_dump 1.28% : 0.000004s : 15: predicate.addn_zero_filter 0.90% : 0.000002s : 15: predicate.adjust_all_reduce_mul_add 2.56% : 0.000007s : 21: predicate.arithmetic_simplify 1.11% : 0.000003s : 15: predicate.cast_eliminate 0.53% : 0.000001s : 6: predicate.check_bprop_eliminate 0.35% : 0.000001s : 6: predicate.compare_switch_simplify 0.10% : 0.000000s : 3: predicate.const_output_eliminate 0.36% : 0.000001s : 6: predicate.depend_value_elim 1.26% : 0.000003s : 15: predicate.dict_get_item_const_eliminator 2.84% : 0.000008s : 15: predicate.dict_get_item_eliminator 1.15% : 0.000003s : 15: predicate.dict_set_item_eliminator 0.96% : 0.000003s : 6: predicate.dumpgradient_eliminate 0.29% : 0.000001s : 3: predicate.elim_not_effective 0.28% : 0.000001s : 3: predicate.elim_shapecalc_of_broadcastargs 1.38% : 0.000004s : 18: predicate.environ_add_const_eliminate 0.97% : 0.000003s : 18: predicate.environ_get_add_eliminate 0.97% : 0.000003s : 18: predicate.environ_get_depend_swap 1.55% : 0.000004s : 24: predicate.environ_get_eliminate 1.12% : 0.000003s : 18: predicate.environ_get_set_eliminate 1.43% : 0.000004s : 26: predicate.exchange_switch_depend_value 2.61% : 0.000007s : 26: predicate.float_depend_g_call 0.35% : 0.000001s : 6: predicate.float_environ_get_switch 0.56% : 0.000002s : 9: predicate.float_tuple_getitem_switch 0.10% : 0.000000s : 3: predicate.fold_const_symbol 0.49% : 0.000001s : 6: predicate.get_grad_eliminate 0.35% : 0.000001s : 3: predicate.graph_param_transform 0.34% : 0.000001s : 6: predicate.incorporate_call 0.28% : 0.000001s : 6: predicate.incorporate_call_switch 6.31% : 0.000017s : 59: predicate.inline 0.47% : 0.000001s : 6: predicate.inline_without_move 0.18% : 0.000001s : 6: predicate.j_node_and_user_rematch 0.65% : 0.000002s : 6: predicate.less_batch_normalization 1.91% : 0.000005s : 25: predicate.list_to_tuple_eliminator_ 2.24% : 0.000006s : 40: predicate.load_eliminater 0.61% : 0.000002s : 3: predicate.loop_unroll_after_grad 2.75% : 0.000008s : 56: predicate.loop_unroll_before_grad 1.90% : 0.000005s : 21: predicate.make_slice_get_slice_eliminator 0.33% : 0.000001s : 6: predicate.merge_addn 0.33% : 0.000001s : 6: predicate.micro_step_allgather_replace 0.53% : 0.000001s : 6: predicate.mini_step_allgather_replace 0.82% : 0.000002s : 15: predicate.minmaximum_grad 2.05% : 0.000006s : 3: predicate.mutable_eliminate 0.27% : 0.000001s : 3: predicate.opt_reshape 0.28% : 0.000001s : 3: predicate.parallel_virtual_node 2.71% : 0.000008s : 26: predicate.partial_defer_inline 1.23% : 0.000003s : 22: predicate.partial_eliminate 0.99% : 0.000003s : 15: predicate.print_const_string_wrapper 0.42% : 0.000001s : 6: predicate.reduce_all_const_elim 1.84% : 0.000005s : 15: predicate.reduce_eliminate 2.41% : 0.000007s : 40: predicate.redundant_stop_gradient_eliminater 0.35% : 0.000001s : 6: predicate.remove_not_recompute_node 1.26% : 0.000003s : 25: predicate.replace_applicator 0.42% : 0.000001s : 6: predicate.replace_old_param 0.48% : 0.000001s : 3: predicate.reset_defer_inline 0.96% : 0.000003s : 15: predicate.reshape_eliminate 0.38% : 0.000001s : 6: predicate.row_tensor_add_zeros_like 0.25% : 0.000001s : 3: predicate.row_tensor_eliminate 0.81% : 0.000002s : 6: predicate.same_eliminate 0.34% : 0.000001s : 6: predicate.set_cell_output_no_recompute 0.53% : 0.000001s : 6: predicate.shard_identity_eliminate 0.54% : 0.000001s : 6: predicate.special_op_eliminate 0.42% : 0.000001s : 6: predicate.specialize_transform 0.93% : 0.000003s : 6: predicate.split_environ_get_set_with_tuple_value 0.64% : 0.000002s : 6: predicate.stack_unstack_eliminate 0.27% : 0.000001s : 3: predicate.switch_call_monad_eliminater 1.47% : 0.000004s : 26: predicate.switch_defer_inline 2.14% : 0.000006s : 32: predicate.switch_layer_defer_inline 5.03% : 0.000014s : 91: predicate.switch_simplify 1.07% : 0.000003s : 15: predicate.tile_eliminate 0.99% : 0.000003s : 15: predicate.transpose_eliminate 1.38% : 0.000004s : 21: predicate.tuple_list_convert_item_index_to_positive 1.82% : 0.000005s : 21: predicate.tuple_list_get_item_const_eliminator 1.46% : 0.000004s : 21: predicate.tuple_list_get_item_depend_reorder 3.47% : 0.000010s : 31: predicate.tuple_list_get_item_eliminator 1.34% : 0.000004s : 21: predicate.tuple_list_get_set_item_eliminator 2.20% : 0.000006s : 27: predicate.tuple_list_set_item_eliminator 1.87% : 0.000005s : 25: predicate.tuple_to_list_eliminator_ 2.03% : 0.000006s : 40: predicate.updatestate_pure_node_eliminater 2.76% : 0.000008s : 46: predicate.updatestate_useless_node_eliminater 0.30% : 0.000001s : 3: predicate.value_based_eliminate 0.54% : 0.000001s : 6: predicate.virtual_dataset_eliminate 0.39% : 0.000001s : 6: predicate.virtual_output_eliminate 0.20% : 0.000001s : 3: predicate.virtual_view_grad_eliminate 0.36% : 0.000001s : 3: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.095365 31 84.80% : 0.080869s : 22: func_graph_cloner_run.FuncGraphClonerGraph 15.20% : 0.014496s : 9: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.950715 192 0.00% : 0.000004s : 1: ForceFp32Comm 0.24% : 0.004777s : 1: add_attr 0.24% : 0.004753s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.00% : 0.000084s : 1: add_recomputation 0.00% : 0.000005s : 1: assign_add_opt 0.01% : 0.000114s : 1: auto_monad 0.00% : 0.000027s : 1: auto_monad_reorder 0.00% : 0.000005s : 1: begin_end_overlap_inline 0.00% : 0.000007s : 1: bias_add_comm_swap 0.04% : 0.000751s : 1: bootstrap 0.00% : 0.000046s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000024s : 1: control_data_broadcast_order 0.00% : 0.000005s : 1: convert_after_rewriter 0.00% : 0.000044s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000007s : 1: detach_backward 0.00% : 0.000014s : 1: environ_conv 0.03% : 0.000616s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000007s : 1: get_jit_bprop_graph 0.00% : 0.000013s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000003s : 1: handle_group_info 0.00% : 0.000007s : 1: inline 0.00% : 0.000009s : 1: insert-virtual-dataset 0.00% : 0.000003s : 1: interleave_parallel_branches 0.00% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000011s : 1: label_micro_interleaved_index 0.03% : 0.000637s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.95% : 0.018618s : 1: mutable_eliminate 0.00% : 0.000007s : 1: offloading_packed_experts 0.00% : 0.000020s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000038s : 1: opt.transform.mutable_eliminate 4.38% : 0.085410s : 78: opt.transform.opt_a 0.00% : 0.000030s : 1: opt.transform.opt_after_cconv 0.00% : 0.000039s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000112s : 28: opt.transform.opt_b 0.00% : 0.000080s : 2: opt.transform.opt_trans_graph 0.00% : 0.000045s : 4: opt.transform.symbol_engine_opt 6.62% : 0.129051s : 1: opt_a 0.81% : 0.015813s : 1: opt_after_cconv 0.05% : 0.001036s : 1: opt_after_jit_grad 0.01% : 0.000278s : 1: opt_b 8.51% : 0.165953s : 1: optimize 0.00% : 0.000029s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 1: order_py_execute_after_rewriter 0.00% : 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 0.00% : 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% : 0.000004s : 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% : 0.000006s : 1: pipeline_split 0.00% : 0.000073s : 1: pre_auto_parallel 0.00% : 0.000013s : 1: py_interpret_to_execute 0.00% : 0.000016s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000068s : 1: remove_dup_value 1.25% : 0.024425s : 1: renormalize.infer 0.75% : 0.014677s : 1: renormalize.specialize 0.00% : 0.000009s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.00% : 0.000029s : 1: rewriter_after_opt_a 0.03% : 0.000665s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000004s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000010s : 1: swap_dp_allreduce_reducescatter 0.00% : 0.000095s : 1: symbol_engine_optimizer 0.01% : 0.000131s : 1: tuple_transform 75.96% : 1.481787s : 1: type_inference TotalTime = 1.89363, [21] [bootstrap]: 0.00042596 [type_inference]: 1.50188 [event_method]: 0.0443014 [auto_monad]: 0.00025187 [graph_reusing]: 1.317e-05 [inline]: 3.68999e-06 [add_attr]: 0.00461666, [1] [add_attr_with_inline]: 0.00460207, [1] [Cycle 1]: 0.00014017, [2] [tag_attr]: 7.55e-05 [meta_addattr_fg_expand]: 1.367e-05 [parallel-infer-symbol]: 4.05e-06 [pre_auto_parallel]: 8.804e-05 [insert-virtual-dataset]: 2.63998e-06 [parallel-infer-symbol-second]: 3.03998e-06 [dataset_repeat_opt]: 2.26998e-06 [pipeline_split]: 1.55999e-06 [optimize]: 0.341009, [53] [py_interpret_to_execute]: 1.174e-05 [rewriter_before_opt_a]: 0.00095235 [opt_a]: 0.337683, [3] [Cycle 1]: 0.271877, [45] [expand_dump_flag]: 9.87999e-06 [switch_simplify]: 0.000218 [loop_unroll]: 9.014e-05 [a_1]: 0.0443557 [with_stream_mark]: 3.731e-05 [recompute_prepare]: 2.834e-05 [updatestate_depend_eliminate]: 1.118e-05 [updatestate_assign_eliminate]: 7.2e-06 [updatestate_loads_eliminate]: 7.19001e-06 [parameter_eliminate]: 3.14001e-06 [a_2]: 0.00023143 [accelerated_algorithm]: 1.612e-05 [shard]: 2.61e-06 [meta_shard_fg_expand]: 7.03e-06 [shard_inline]: 1.539e-05 [merge_send_recv]: 1.855e-05 [auto_parallel]: 1.465e-05 [parallel]: 2.078e-05 [flash_sp]: 1.312e-05 [merge_comm]: 9.29e-06 [allreduce_fusion]: 8.37e-06 [matmul_add_comm_reduction]: 3.372e-05 [allreduce_slice_to_reducescatter]: 8.50006e-07 [virtual_shard_identity]: 1.792e-05 [virtual_dataset]: 1.535e-05 [get_grad_eliminate_]: 1.518e-05 [virtual_output]: 1.461e-05 [merge_forward]: 9.81998e-06 [cell_reuse_recompute_pass]: 1.81e-06 [offload_activation]: 1.87e-05 [cell_reuse_handle_not_recompute_node_pass]: 7.002e-05 [merge_recompute_call_nodes]: 1.71998e-06 [before_grad]: 2.763e-05 [set_forward_comm_id_for_comm_node_pass]: 1.054e-05 [meta_fg_expand]: 0.0231502 [flash_sp_send_recv_attached]: 5.07999e-06 [receive_attached]: 2.64001e-06 [after_resolve]: 9.281e-05 [a_after_grad]: 0.0001004 [renormalize]: 0.20174 [add_forward_monad_depend]: 1.964e-05 [auto_monad_grad]: 8.00999e-06 [auto_monad_eliminator]: 7.336e-05 [cse]: 0.00053369 [a_3]: 0.00037319 [Cycle 2]: 0.0648664, [45] [expand_dump_flag]: 3.28e-06 [switch_simplify]: 4.845e-05 [loop_unroll]: 4.345e-05 [a_1]: 0.0382657 [with_stream_mark]: 3.714e-05 [recompute_prepare]: 1.532e-05 [updatestate_depend_eliminate]: 5.92001e-06 [updatestate_assign_eliminate]: 3.91999e-06 [updatestate_loads_eliminate]: 4.37998e-06 [parameter_eliminate]: 2.16998e-06 [a_2]: 9.54e-05 [accelerated_algorithm]: 9.04e-06 [shard]: 2.24001e-06 [meta_shard_fg_expand]: 3.51999e-06 [shard_inline]: 8.17e-06 [merge_send_recv]: 9.61998e-06 [auto_parallel]: 1.054e-05 [parallel]: 1.048e-05 [flash_sp]: 4.12e-06 [merge_comm]: 4.48999e-06 [allreduce_fusion]: 4.05e-06 [matmul_add_comm_reduction]: 1.138e-05 [allreduce_slice_to_reducescatter]: 1.37e-06 [virtual_shard_identity]: 9.92001e-06 [virtual_dataset]: 7.06999e-06 [get_grad_eliminate_]: 7.00002e-06 [virtual_output]: 8.72e-06 [merge_forward]: 4.70001e-06 [cell_reuse_recompute_pass]: 2.17999e-06 [offload_activation]: 1.107e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.863e-05 [merge_recompute_call_nodes]: 1.52999e-06 [before_grad]: 1.269e-05 [set_forward_comm_id_for_comm_node_pass]: 4.69002e-06 [meta_fg_expand]: 0.00035182 [flash_sp_send_recv_attached]: 2.39999e-06 [receive_attached]: 2.91e-06 [after_resolve]: 1.561e-05 [a_after_grad]: 1.147e-05 [renormalize]: 0.022206 [add_forward_monad_depend]: 1.112e-05 [auto_monad_grad]: 2.99999e-06 [auto_monad_eliminator]: 2.738e-05 [cse]: 4.468e-05 [a_3]: 0.00305742 [Cycle 3]: 0.00091031, [45] [expand_dump_flag]: 4.12e-06 [switch_simplify]: 1.662e-05 [loop_unroll]: 8.36002e-06 [a_1]: 0.00018164 [with_stream_mark]: 2.854e-05 [recompute_prepare]: 8.13999e-06 [updatestate_depend_eliminate]: 6.84999e-06 [updatestate_assign_eliminate]: 4.4e-06 [updatestate_loads_eliminate]: 3.53999e-06 [parameter_eliminate]: 2.19001e-06 [a_2]: 8.873e-05 [accelerated_algorithm]: 7.71001e-06 [shard]: 3.25e-06 [meta_shard_fg_expand]: 3.72002e-06 [shard_inline]: 7.73999e-06 [merge_send_recv]: 1.095e-05 [auto_parallel]: 1.051e-05 [parallel]: 1.042e-05 [flash_sp]: 1.49998e-06 [merge_comm]: 4.85999e-06 [allreduce_fusion]: 3.95e-06 [matmul_add_comm_reduction]: 1.29e-05 [allreduce_slice_to_reducescatter]: 1.17999e-06 [virtual_shard_identity]: 8.60001e-06 [virtual_dataset]: 6.77002e-06 [get_grad_eliminate_]: 7.46001e-06 [virtual_output]: 7.2e-06 [merge_forward]: 5.09e-06 [cell_reuse_recompute_pass]: 3.55e-06 [offload_activation]: 1.237e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.625e-05 [merge_recompute_call_nodes]: 1.46998e-06 [before_grad]: 1.306e-05 [set_forward_comm_id_for_comm_node_pass]: 4.97999e-06 [meta_fg_expand]: 3.8e-06 [flash_sp_send_recv_attached]: 1.88997e-06 [receive_attached]: 3.09999e-06 [after_resolve]: 2.717e-05 [a_after_grad]: 1.107e-05 [renormalize]: 6.99947e-08 [add_forward_monad_depend]: 3.38e-06 [auto_monad_grad]: 3.33998e-06 [auto_monad_eliminator]: 1.758e-05 [cse]: 4.773e-05 [a_3]: 4.443e-05 [py_interpret_to_execute_after_opt_a]: 1.183e-05 [slice_cell_reuse_recomputed_activation]: 2.83998e-06 [rewriter_after_opt_a]: 2.861e-05 [convert_after_rewriter]: 1.40999e-06 [order_py_execute_after_rewriter]: 1.27e-06 [mutable_eliminate]: 0.00078913 [opt_b]: 0.00024523, [1] [Cycle 1]: 0.00023662, [7] [b_1]: 0.00014486 [b_2]: 1.012e-05 [updatestate_depend_eliminate]: 9.07999e-06 [updatestate_assign_eliminate]: 3.31999e-06 [updatestate_loads_eliminate]: 3.04999e-06 [renormalize]: 6.39993e-07 [cse]: 2.993e-05 [optimize_parallel_all_gather_comm]: 2.048e-05 [overlap_param_gather]: 1.91e-06 [cconv]: 3.685e-05 [loop_unroll]: 0.00045623 [opt_after_cconv]: 0.00011769, [1] [Cycle 1]: 0.00011135, [7] [c_1]: 3.264e-05 [parameter_eliminate]: 3.88001e-06 [updatestate_depend_eliminate]: 7.16999e-06 [updatestate_assign_eliminate]: 2.99001e-06 [updatestate_loads_eliminate]: 3.08998e-06 [cse]: 2.81e-05 [renormalize]: 3.19997e-07 [remove_dup_value]: 2.414e-05 [tuple_transform]: 7.895e-05, [1] [Cycle 1]: 7.466e-05, [4] [d_1]: 4.504e-05 [none_parameter_eliminate]: 2.06e-06 [renormalize]: 1.69995e-07 [switch_simplify]: 7.88999e-06 [partial_unused_args_eliminate]: 2.58e-06 [add_recomputation]: 5.222e-05 [cse_after_recomputation]: 2.845e-05, [1] [Cycle 1]: 2.414e-05, [1] [cse]: 1.851e-05 [environ_conv]: 1.108e-05 [swap_dp_allreduce_reducescatter]: 6.46999e-06 [bias_add_comm_swap]: 3.33998e-06 [label_micro_interleaved_index]: 4.55999e-06 [label_fine_grained_interleaved_index]: 2.71e-06 [merge_cast_opt]: 1.39e-06 [slice_recompute_activation]: 2.12999e-06 [micro_interleaved_order_control]: 2.37999e-06 [assign_add_opt]: 1.17999e-06 [ForceFp32Comm]: 8.70001e-07 [remove_cast_before_assign_add]: 1.02998e-06 [full_micro_interleaved_order_control]: 2.11e-06 [reorder_send_recv_between_fp_bp]: 2.89999e-06 [comm_op_add_attrs]: 1.19e-06 [add_comm_op_reuse_tag]: 1.00001e-06 [interleave_split_concat_branches]: 1.20001e-06 [interleave_parallel_branches]: 1.04e-06 [overlap_opt_shard_in_pipeline]: 1.22e-06 [overlap_opt_shard_grad_in_pipeline]: 2.14e-06 [control_data_broadcast_order]: 1.567e-05 [grouped_pairwise_exchange_alltoall]: 1.39998e-06 [offloading_packed_experts]: 6.09001e-06 [overlap_recompute_and_grad_model_parallel]: 5.04e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.23002e-06 [overlap_recompute_allgather_and_fa_grad]: 1.25999e-06 [overlap_recompute_comm]: 2.27999e-06 [overlap_grad_ring_attention]: 4.35999e-06 [overlap_grad_flash_sp]: 2.903e-05 [begin_end_overlap_inline]: 5.19998e-07 [split_matmul_comm_elemetwise]: 2.12001e-06 [split_layernorm_comm]: 1.95001e-06 [handle_group_info]: 1.05001e-06 [symbol_engine_optimizer]: 8.718e-05, [1] [Cycle 1]: 8.203e-05, [6] [build]: 3.46001e-06 [elim_shapecalc]: 1.175e-05 [elim_not_effective]: 1.617e-05 [opt_reshape]: 1.203e-05 [fold_const_symbol]: 1.089e-05 [renormalize]: 2.3999e-07 [detach_backward]: 2.73998e-06 [pipeline_parallel_scheduler]: 0.00016017 [auto_monad_reorder]: 2.015e-05 [get_jit_bprop_graph]: 2.44001e-06 [rewriter_after_jit_bprop_graph]: 6.18998e-06 [opt_after_jit_grad]: 0.00051588 [validate]: 6.691e-05 Sums bootstrap : 0.000426s : 0.02% type_inference : 1.501878s : 79.58% event_method : 0.044301s : 2.35% auto_monad : 0.000252s : 0.01% graph_reusing : 0.000013s : 0.00% inline : 0.000004s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000076s : 0.00% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000014s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000088s : 0.00% insert-virtual-dataset : 0.000003s : 0.00% parallel-infer-symbol-second : 0.000003s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000012s : 0.00% optimize.rewriter_before_opt_a : 0.000952s : 0.05% optimize.opt_a.expand_dump_flag : 0.000017s : 0.00% optimize.opt_a.switch_simplify : 0.000283s : 0.01% optimize.opt_a.loop_unroll : 0.000142s : 0.01% optimize.opt_a.a_1 : 0.082803s : 4.39% optimize.opt_a.with_stream_mark : 0.000103s : 0.01% optimize.opt_a.recompute_prepare : 0.000052s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000024s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000016s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000015s : 0.00% optimize.opt_a.parameter_eliminate : 0.000007s : 0.00% optimize.opt_a.a_2 : 0.000416s : 0.02% optimize.opt_a.accelerated_algorithm : 0.000033s : 0.00% optimize.opt_a.shard : 0.000008s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000014s : 0.00% optimize.opt_a.shard_inline : 0.000031s : 0.00% optimize.opt_a.merge_send_recv : 0.000039s : 0.00% optimize.opt_a.auto_parallel : 0.000036s : 0.00% optimize.opt_a.parallel : 0.000042s : 0.00% optimize.opt_a.flash_sp : 0.000019s : 0.00% optimize.opt_a.merge_comm : 0.000019s : 0.00% optimize.opt_a.allreduce_fusion : 0.000016s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000058s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000003s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000036s : 0.00% optimize.opt_a.virtual_dataset : 0.000029s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000030s : 0.00% optimize.opt_a.virtual_output : 0.000031s : 0.00% optimize.opt_a.merge_forward : 0.000020s : 0.00% optimize.opt_a.cell_reuse_recompute_pass : 0.000008s : 0.00% optimize.opt_a.offload_activation : 0.000042s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000105s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000005s : 0.00% optimize.opt_a.before_grad : 0.000053s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000020s : 0.00% optimize.opt_a.meta_fg_expand : 0.023506s : 1.25% optimize.opt_a.flash_sp_send_recv_attached : 0.000009s : 0.00% optimize.opt_a.receive_attached : 0.000009s : 0.00% optimize.opt_a.after_resolve : 0.000136s : 0.01% optimize.opt_a.a_after_grad : 0.000123s : 0.01% optimize.opt_a.renormalize : 0.223946s : 11.87% optimize.opt_a.add_forward_monad_depend : 0.000034s : 0.00% optimize.opt_a.auto_monad_grad : 0.000014s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000118s : 0.01% optimize.opt_a.cse : 0.000626s : 0.03% optimize.opt_a.a_3 : 0.003475s : 0.18% optimize.py_interpret_to_execute_after_opt_a : 0.000012s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000003s : 0.00% optimize.rewriter_after_opt_a : 0.000029s : 0.00% optimize.convert_after_rewriter : 0.000001s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000789s : 0.04% optimize.opt_b.b_1 : 0.000145s : 0.01% optimize.opt_b.b_2 : 0.000010s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000009s : 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 : 0.000030s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000020s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000037s : 0.00% optimize.loop_unroll : 0.000456s : 0.02% optimize.opt_after_cconv.c_1 : 0.000033s : 0.00% optimize.opt_after_cconv.parameter_eliminate : 0.000004s : 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 : 0.000003s : 0.00% optimize.opt_after_cconv.cse : 0.000028s : 0.00% optimize.opt_after_cconv.renormalize : 0.000000s : 0.00% optimize.remove_dup_value : 0.000024s : 0.00% optimize.tuple_transform.d_1 : 0.000045s : 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 : 0.000008s : 0.00% optimize.partial_unused_args_eliminate : 0.000003s : 0.00% optimize.add_recomputation : 0.000052s : 0.00% optimize.cse_after_recomputation.cse : 0.000019s : 0.00% optimize.environ_conv : 0.000011s : 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 : 0.000005s : 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 : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000016s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000001s : 0.00% optimize.offloading_packed_experts : 0.000006s : 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 : 0.000029s : 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 : 0.000012s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000016s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000012s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000011s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000003s : 0.00% pipeline_parallel_scheduler : 0.000160s : 0.01% auto_monad_reorder : 0.000020s : 0.00% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.000516s : 0.03% validate : 0.000067s : 0.00% Time group info: ------[substitution.] 0.001274 174 0.17% : 0.000002s : 3: substitution.elim_not_effective 0.98% : 0.000012s : 12: substitution.float_depend_g_call 0.37% : 0.000005s : 2: substitution.float_tuple_getitem_switch 0.15% : 0.000002s : 3: substitution.fold_const_symbol 0.48% : 0.000006s : 4: substitution.graph_param_transform 0.22% : 0.000003s : 2: substitution.incorporate_call 0.17% : 0.000002s : 2: substitution.incorporate_call_switch 69.17% : 0.000881s : 23: substitution.inline 1.91% : 0.000024s : 2: substitution.inline_without_move 0.90% : 0.000011s : 14: substitution.j_node_and_user_rematch 1.18% : 0.000015s : 7: substitution.minmaximum_grad 0.12% : 0.000002s : 1: substitution.opt_reshape 0.97% : 0.000012s : 12: substitution.partial_eliminate 4.29% : 0.000055s : 14: substitution.remove_not_recompute_node 2.77% : 0.000035s : 9: substitution.replace_applicator 0.70% : 0.000009s : 11: substitution.replace_old_param 1.87% : 0.000024s : 2: substitution.reshape_eliminate 0.28% : 0.000004s : 1: substitution.set_cell_output_no_recompute 1.22% : 0.000016s : 3: substitution.switch_simplify 2.31% : 0.000029s : 7: substitution.tuple_list_convert_item_index_to_positive 0.93% : 0.000012s : 7: substitution.tuple_list_get_item_const_eliminator 1.35% : 0.000017s : 7: substitution.tuple_list_get_item_depend_reorder 6.02% : 0.000077s : 19: substitution.tuple_list_get_item_eliminator 1.46% : 0.000019s : 7: substitution.tuple_list_get_set_item_eliminator ------[type_inference.] 1.501660 2 99.59% : 1.495495s : 1: type_inference.infer 0.41% : 0.006165s : 1: type_inference.specialize ------[replace.] 0.000439 36 57.69% : 0.000253s : 23: replace.inline 13.17% : 0.000058s : 3: replace.switch_simplify 29.14% : 0.000128s : 10: replace.tuple_list_get_item_eliminator ------[match.] 0.000924 36 93.93% : 0.000868s : 23: match.inline 1.50% : 0.000014s : 3: match.switch_simplify 4.58% : 0.000042s : 10: match.tuple_list_get_item_eliminator ------[predicate.] 0.042982 4535 0.02% : 0.000009s : 57: predicate.accumulaten_eliminater 0.01% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.01% : 0.000003s : 20: predicate.addn_check_dump 0.02% : 0.000008s : 57: predicate.addn_zero_filter 0.02% : 0.000007s : 57: predicate.adjust_all_reduce_mul_add 0.03% : 0.000015s : 77: predicate.arithmetic_simplify 0.02% : 0.000008s : 57: predicate.cast_eliminate 0.02% : 0.000007s : 49: predicate.check_bprop_eliminate 0.01% : 0.000003s : 20: predicate.compare_switch_simplify 0.00% : 0.000000s : 4: predicate.const_output_eliminate 0.01% : 0.000003s : 20: predicate.depend_value_elim 0.02% : 0.000008s : 57: predicate.dict_get_item_const_eliminator 0.02% : 0.000009s : 57: predicate.dict_get_item_eliminator 0.02% : 0.000008s : 57: predicate.dict_set_item_eliminator 0.00% : 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 0.02% : 0.000009s : 61: predicate.environ_add_const_eliminate 0.02% : 0.000008s : 61: predicate.environ_get_add_eliminate 0.03% : 0.000014s : 61: predicate.environ_get_depend_swap 0.03% : 0.000012s : 81: predicate.environ_get_eliminate 0.02% : 0.000008s : 61: predicate.environ_get_set_eliminate 0.03% : 0.000013s : 90: predicate.exchange_switch_depend_value 0.05% : 0.000021s : 90: predicate.float_depend_g_call 0.01% : 0.000003s : 20: predicate.float_environ_get_switch 0.01% : 0.000004s : 24: predicate.float_tuple_getitem_switch 0.00% : 0.000000s : 4: predicate.fold_const_symbol 0.01% : 0.000003s : 20: predicate.get_grad_eliminate 0.00% : 0.000000s : 4: predicate.graph_param_transform 0.01% : 0.000003s : 20: predicate.incorporate_call 0.01% : 0.000003s : 20: predicate.incorporate_call_switch 0.10% : 0.000042s : 195: predicate.inline 0.02% : 0.000010s : 50: predicate.inline_without_move 0.00% : 0.000001s : 20: predicate.j_node_and_user_rematch 0.01% : 0.000004s : 20: predicate.less_batch_normalization 0.03% : 0.000011s : 75: predicate.list_to_tuple_eliminator_ 0.04% : 0.000017s : 132: predicate.load_eliminater 0.00% : 0.000002s : 4: predicate.loop_unroll_after_grad 0.05% : 0.000020s : 150: predicate.loop_unroll_before_grad 0.03% : 0.000012s : 65: predicate.make_slice_get_slice_eliminator 0.01% : 0.000003s : 20: predicate.merge_addn 0.02% : 0.000007s : 49: predicate.micro_step_allgather_replace 0.03% : 0.000011s : 49: predicate.mini_step_allgather_replace 0.02% : 0.000008s : 57: 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 0.05% : 0.000021s : 90: predicate.partial_defer_inline 0.02% : 0.000011s : 71: predicate.partial_eliminate 0.03% : 0.000013s : 57: predicate.print_const_string_wrapper 0.01% : 0.000003s : 20: predicate.reduce_all_const_elim 0.02% : 0.000010s : 57: predicate.reduce_eliminate 0.04% : 0.000019s : 132: predicate.redundant_stop_gradient_eliminater 0.00% : 0.000002s : 20: predicate.remove_not_recompute_node 0.03% : 0.000013s : 116: predicate.replace_applicator 0.01% : 0.000006s : 50: predicate.replace_old_param 0.00% : 0.000001s : 4: predicate.reset_defer_inline 0.02% : 0.000009s : 57: predicate.reshape_eliminate 0.02% : 0.000009s : 49: predicate.row_tensor_add_zeros_like 0.00% : 0.000001s : 4: predicate.row_tensor_eliminate 0.02% : 0.000009s : 49: predicate.same_eliminate 0.01% : 0.000002s : 20: predicate.set_cell_output_no_recompute 0.01% : 0.000004s : 20: predicate.shard_identity_eliminate 0.00% : 0.000001s : 8: predicate.special_op_eliminate 0.01% : 0.000003s : 20: predicate.specialize_transform 0.02% : 0.000010s : 49: predicate.split_environ_get_set_with_tuple_value 0.02% : 0.000009s : 50: predicate.stack_unstack_eliminate 0.00% : 0.000001s : 4: predicate.switch_call_monad_eliminater 0.03% : 0.000014s : 90: predicate.switch_defer_inline 0.05% : 0.000021s : 139: predicate.switch_layer_defer_inline 0.10% : 0.000043s : 270: predicate.switch_simplify 0.02% : 0.000008s : 57: predicate.tile_eliminate 0.02% : 0.000008s : 57: predicate.transpose_eliminate 0.02% : 0.000010s : 65: predicate.tuple_list_convert_item_index_to_positive 0.02% : 0.000011s : 65: predicate.tuple_list_get_item_const_eliminator 0.02% : 0.000010s : 65: predicate.tuple_list_get_item_depend_reorder 98.35% : 0.042271s : 95: predicate.tuple_list_get_item_eliminator 0.02% : 0.000010s : 65: predicate.tuple_list_get_set_item_eliminator 0.03% : 0.000014s : 85: predicate.tuple_list_set_item_eliminator 0.03% : 0.000012s : 75: predicate.tuple_to_list_eliminator_ 0.04% : 0.000016s : 132: predicate.updatestate_pure_node_eliminater 0.05% : 0.000020s : 152: predicate.updatestate_useless_node_eliminater 0.00% : 0.000001s : 4: predicate.value_based_eliminate 0.01% : 0.000003s : 20: predicate.virtual_dataset_eliminate 0.01% : 0.000004s : 20: 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.] 0.085781 80 97.53% : 0.083665s : 53: func_graph_cloner_run.FuncGraphClonerGraph 2.47% : 0.002116s : 27: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 2.550878 233 0.00% : 0.000003s : 1: ForceFp32Comm 0.18% : 0.004623s : 1: add_attr 0.18% : 0.004608s : 1: add_attr_with_inline 0.00% : 0.000003s : 1: add_comm_op_reuse_tag 0.00% : 0.000057s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.01% : 0.000269s : 1: auto_monad 0.00% : 0.000026s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.02% : 0.000451s : 1: bootstrap 0.00% : 0.000041s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000019s : 1: control_data_broadcast_order 0.00% : 0.000005s : 1: convert_after_rewriter 0.00% : 0.000031s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000015s : 1: environ_conv 1.74% : 0.044358s : 1: event_method 0.00% : 0.000007s : 1: full_micro_interleaved_order_control 0.00% : 0.000007s : 1: get_jit_bprop_graph 0.00% : 0.000019s : 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% : 0.000004s : 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 0.02% : 0.000466s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000006s : 1: micro_interleaved_order_control 0.03% : 0.000801s : 1: mutable_eliminate 0.00% : 0.000009s : 1: offloading_packed_experts 0.00% : 0.000015s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000020s : 1: opt.transform.mutable_eliminate 3.43% : 0.087570s : 117: opt.transform.opt_a 0.00% : 0.000031s : 1: opt.transform.opt_after_cconv 0.00% : 0.000029s : 1: opt.transform.opt_after_jit_grad 0.00% : 0.000125s : 28: opt.transform.opt_b 0.00% : 0.000051s : 2: opt.transform.opt_trans_graph 0.00% : 0.000047s : 4: opt.transform.symbol_engine_opt 13.24% : 0.337688s : 1: opt_a 0.00% : 0.000121s : 1: opt_after_cconv 0.02% : 0.000528s : 1: opt_after_jit_grad 0.01% : 0.000249s : 1: opt_b 13.37% : 0.341015s : 1: optimize 0.00% : 0.000025s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 1: order_py_execute_after_rewriter 0.00% : 0.000033s : 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% : 0.000005s : 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 0.00% : 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% : 0.000006s : 1: parallel-infer-symbol-second 0.00% : 0.000006s : 1: partial_unused_args_eliminate 0.01% : 0.000165s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.00% : 0.000093s : 1: pre_auto_parallel 0.00% : 0.000015s : 1: py_interpret_to_execute 0.00% : 0.000015s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000028s : 1: remove_dup_value 8.62% : 0.219912s : 2: renormalize.infer 0.16% : 0.003997s : 2: renormalize.specialize 0.00% : 0.000005s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000010s : 1: rewriter_after_jit_bprop_graph 0.00% : 0.000032s : 1: rewriter_after_opt_a 0.04% : 0.000970s : 1: rewriter_before_opt_a 0.00% : 0.000006s : 1: slice_cell_reuse_recomputed_activation 0.00% : 0.000005s : 1: slice_recompute_activation 0.00% : 0.000007s : 1: split_layernorm_comm 0.00% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.00% : 0.000090s : 1: symbol_engine_optimizer 0.00% : 0.000082s : 1: tuple_transform 58.88% : 1.501915s : 1: type_inference TotalTime = 1.71504, [21] [bootstrap]: 0.00047674 [type_inference]: 1.56541 [event_method]: 0.00603489 [auto_monad]: 0.00012445 [graph_reusing]: 2.559e-05 [inline]: 9.69e-06 [add_attr]: 0.026273, [1] [add_attr_with_inline]: 0.026258, [1] [Cycle 1]: 9.958e-05, [2] [tag_attr]: 4.609e-05 [meta_addattr_fg_expand]: 7.75e-06 [parallel-infer-symbol]: 3.98001e-06 [pre_auto_parallel]: 5.865e-05 [insert-virtual-dataset]: 3.06001e-06 [parallel-infer-symbol-second]: 7.10017e-07 [dataset_repeat_opt]: 2.27001e-06 [pipeline_split]: 1.76e-06 [optimize]: 0.0908291, [53] [py_interpret_to_execute]: 1.216e-05 [rewriter_before_opt_a]: 0.00037587 [opt_a]: 0.0880243, [2] [Cycle 1]: 0.0787492, [45] [expand_dump_flag]: 3.63e-06 [switch_simplify]: 6.14e-05 [loop_unroll]: 4.811e-05 [a_1]: 0.0146217 [with_stream_mark]: 2.946e-05 [recompute_prepare]: 1.324e-05 [updatestate_depend_eliminate]: 5.03002e-06 [updatestate_assign_eliminate]: 4.13001e-06 [updatestate_loads_eliminate]: 3.28e-06 [parameter_eliminate]: 2.49001e-06 [a_2]: 8.262e-05 [accelerated_algorithm]: 7.4e-06 [shard]: 2.12999e-06 [meta_shard_fg_expand]: 4.31002e-06 [shard_inline]: 6.69001e-06 [merge_send_recv]: 1.005e-05 [auto_parallel]: 1.106e-05 [parallel]: 2.737e-05 [flash_sp]: 1.099e-05 [merge_comm]: 4.74998e-06 [allreduce_fusion]: 3.46999e-06 [matmul_add_comm_reduction]: 1.064e-05 [allreduce_slice_to_reducescatter]: 9.69972e-07 [virtual_shard_identity]: 8.69e-06 [virtual_dataset]: 6.83e-06 [get_grad_eliminate_]: 6.50002e-06 [virtual_output]: 6.16998e-06 [merge_forward]: 4.50999e-06 [cell_reuse_recompute_pass]: 1.53002e-06 [offload_activation]: 1.017e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.408e-05 [merge_recompute_call_nodes]: 1.34998e-06 [before_grad]: 1.076e-05 [set_forward_comm_id_for_comm_node_pass]: 4.07003e-06 [meta_fg_expand]: 3.83999e-06 [flash_sp_send_recv_attached]: 2.73998e-06 [receive_attached]: 2.16e-06 [after_resolve]: 1.068e-05 [a_after_grad]: 1.038e-05 [renormalize]: 0.0631937 [add_forward_monad_depend]: 1.298e-05 [auto_monad_grad]: 2.54001e-06 [auto_monad_eliminator]: 2.864e-05 [cse]: 4.655e-05 [a_3]: 6.515e-05 [Cycle 2]: 0.00925674, [45] [expand_dump_flag]: 3.08998e-06 [switch_simplify]: 8.73001e-06 [loop_unroll]: 6.46999e-06 [a_1]: 0.00011253 [with_stream_mark]: 2.185e-05 [recompute_prepare]: 7.28999e-06 [updatestate_depend_eliminate]: 4.63999e-06 [updatestate_assign_eliminate]: 3.55003e-06 [updatestate_loads_eliminate]: 3.66001e-06 [parameter_eliminate]: 1.95001e-06 [a_2]: 0.0044557 [accelerated_algorithm]: 2.028e-05 [shard]: 5.77001e-06 [meta_shard_fg_expand]: 5.14e-06 [shard_inline]: 9.59e-06 [merge_send_recv]: 1.79e-05 [auto_parallel]: 1.58e-05 [parallel]: 1.419e-05 [flash_sp]: 5.24e-06 [merge_comm]: 4.42e-06 [allreduce_fusion]: 3.66001e-06 [matmul_add_comm_reduction]: 1.318e-05 [allreduce_slice_to_reducescatter]: 1.12e-06 [virtual_shard_identity]: 7.71001e-06 [virtual_dataset]: 7.01999e-06 [get_grad_eliminate_]: 6.34999e-06 [virtual_output]: 6.30002e-06 [merge_forward]: 4.89003e-06 [cell_reuse_recompute_pass]: 3.34001e-06 [offload_activation]: 1.13e-05 [cell_reuse_handle_not_recompute_node_pass]: 2.268e-05 [merge_recompute_call_nodes]: 1.35999e-06 [before_grad]: 1.007e-05 [set_forward_comm_id_for_comm_node_pass]: 3.97e-06 [meta_fg_expand]: 3.6e-06 [flash_sp_send_recv_attached]: 2.14999e-06 [receive_attached]: 2.61999e-06 [after_resolve]: 1.459e-05 [a_after_grad]: 9.65002e-06 [renormalize]: 9.00181e-08 [add_forward_monad_depend]: 6.66999e-06 [auto_monad_grad]: 3.23e-06 [auto_monad_eliminator]: 2.22e-05 [cse]: 4.89e-05 [a_3]: 0.00402412 [py_interpret_to_execute_after_opt_a]: 1.407e-05 [slice_cell_reuse_recomputed_activation]: 3.95e-06 [rewriter_after_opt_a]: 3.885e-05 [convert_after_rewriter]: 2.13998e-06 [order_py_execute_after_rewriter]: 1.18001e-06 [mutable_eliminate]: 0.00079458 [opt_b]: 0.00024552, [1] [Cycle 1]: 0.00023703, [7] [b_1]: 0.00012716 [b_2]: 8e-06 [updatestate_depend_eliminate]: 1.191e-05 [updatestate_assign_eliminate]: 3.48999e-06 [updatestate_loads_eliminate]: 2.96999e-06 [renormalize]: 7.59988e-07 [cse]: 4.579e-05 [optimize_parallel_all_gather_comm]: 2.208e-05 [overlap_param_gather]: 1.87999e-06 [cconv]: 3.723e-05 [loop_unroll]: 0.00047636 [opt_after_cconv]: 0.00013154, [1] [Cycle 1]: 0.00012453, [7] [c_1]: 2.752e-05 [parameter_eliminate]: 5.49998e-06 [updatestate_depend_eliminate]: 6.23e-06 [updatestate_assign_eliminate]: 2.74001e-06 [updatestate_loads_eliminate]: 2.41e-06 [cse]: 4.36e-05 [renormalize]: 5.20027e-07 [remove_dup_value]: 4.89e-05 [tuple_transform]: 7.701e-05, [1] [Cycle 1]: 7.217e-05, [4] [d_1]: 4.375e-05 [none_parameter_eliminate]: 2.47001e-06 [renormalize]: 1.30007e-07 [switch_simplify]: 7.37997e-06 [partial_unused_args_eliminate]: 2.10002e-06 [add_recomputation]: 5.435e-05 [cse_after_recomputation]: 2.698e-05, [1] [Cycle 1]: 2.245e-05, [1] [cse]: 1.672e-05 [environ_conv]: 1.04e-05 [swap_dp_allreduce_reducescatter]: 5.50001e-06 [bias_add_comm_swap]: 2.93998e-06 [label_micro_interleaved_index]: 5.04e-06 [label_fine_grained_interleaved_index]: 2.58e-06 [merge_cast_opt]: 1.42e-06 [slice_recompute_activation]: 2.07001e-06 [micro_interleaved_order_control]: 2.55002e-06 [assign_add_opt]: 1.37999e-06 [ForceFp32Comm]: 9.5999e-07 [remove_cast_before_assign_add]: 1.05001e-06 [full_micro_interleaved_order_control]: 2.34001e-06 [reorder_send_recv_between_fp_bp]: 2.71e-06 [comm_op_add_attrs]: 1.39998e-06 [add_comm_op_reuse_tag]: 1.00001e-06 [interleave_split_concat_branches]: 1.27e-06 [interleave_parallel_branches]: 1.07e-06 [overlap_opt_shard_in_pipeline]: 1.48002e-06 [overlap_opt_shard_grad_in_pipeline]: 1.74e-06 [control_data_broadcast_order]: 1.524e-05 [grouped_pairwise_exchange_alltoall]: 1.39e-06 [offloading_packed_experts]: 5.06002e-06 [overlap_recompute_and_grad_model_parallel]: 5.12e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.20999e-06 [overlap_recompute_allgather_and_fa_grad]: 1.17e-06 [overlap_recompute_comm]: 2.54999e-06 [overlap_grad_ring_attention]: 4.15999e-06 [overlap_grad_flash_sp]: 2.172e-05 [begin_end_overlap_inline]: 4.69998e-07 [split_matmul_comm_elemetwise]: 2.13998e-06 [split_layernorm_comm]: 1.73002e-06 [handle_group_info]: 1.28002e-06 [symbol_engine_optimizer]: 7.871e-05, [1] [Cycle 1]: 7.406e-05, [6] [build]: 3.95e-06 [elim_shapecalc]: 1.234e-05 [elim_not_effective]: 1.308e-05 [opt_reshape]: 6.47001e-06 [fold_const_symbol]: 9.59e-06 [renormalize]: 1.50001e-07 [detach_backward]: 2.31998e-06 [pipeline_parallel_scheduler]: 1.46998e-06 [auto_monad_reorder]: 1.881e-05 [get_jit_bprop_graph]: 2.17999e-06 [rewriter_after_jit_bprop_graph]: 5.00001e-06 [opt_after_jit_grad]: 0.0143368 [validate]: 0.00245665 Sums bootstrap : 0.000477s : 0.03% type_inference : 1.565407s : 93.24% event_method : 0.006035s : 0.36% auto_monad : 0.000124s : 0.01% graph_reusing : 0.000026s : 0.00% inline : 0.000010s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000046s : 0.00% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000008s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000059s : 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 : 0.000012s : 0.00% optimize.rewriter_before_opt_a : 0.000376s : 0.02% optimize.opt_a.expand_dump_flag : 0.000007s : 0.00% optimize.opt_a.switch_simplify : 0.000070s : 0.00% optimize.opt_a.loop_unroll : 0.000055s : 0.00% optimize.opt_a.a_1 : 0.014734s : 0.88% optimize.opt_a.with_stream_mark : 0.000051s : 0.00% optimize.opt_a.recompute_prepare : 0.000021s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000010s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000008s : 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 : 0.004538s : 0.27% optimize.opt_a.accelerated_algorithm : 0.000028s : 0.00% optimize.opt_a.shard : 0.000008s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000009s : 0.00% optimize.opt_a.shard_inline : 0.000016s : 0.00% optimize.opt_a.merge_send_recv : 0.000028s : 0.00% optimize.opt_a.auto_parallel : 0.000027s : 0.00% optimize.opt_a.parallel : 0.000042s : 0.00% optimize.opt_a.flash_sp : 0.000016s : 0.00% optimize.opt_a.merge_comm : 0.000009s : 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 : 0.000014s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000013s : 0.00% optimize.opt_a.virtual_output : 0.000012s : 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 : 0.000021s : 0.00% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000037s : 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 : 0.000008s : 0.00% optimize.opt_a.meta_fg_expand : 0.000007s : 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 : 0.00% optimize.opt_a.a_after_grad : 0.000020s : 0.00% optimize.opt_a.renormalize : 0.063194s : 3.76% optimize.opt_a.add_forward_monad_depend : 0.000020s : 0.00% optimize.opt_a.auto_monad_grad : 0.000006s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000051s : 0.00% optimize.opt_a.cse : 0.000095s : 0.01% optimize.opt_a.a_3 : 0.004089s : 0.24% optimize.py_interpret_to_execute_after_opt_a : 0.000014s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000004s : 0.00% optimize.rewriter_after_opt_a : 0.000039s : 0.00% optimize.convert_after_rewriter : 0.000002s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000795s : 0.05% optimize.opt_b.b_1 : 0.000127s : 0.01% optimize.opt_b.b_2 : 0.000008s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000012s : 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 : 0.000046s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000022s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000037s : 0.00% optimize.loop_unroll : 0.000476s : 0.03% optimize.opt_after_cconv.c_1 : 0.000028s : 0.00% 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 : 0.000002s : 0.00% optimize.opt_after_cconv.cse : 0.000044s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000049s : 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 : 0.000054s : 0.00% optimize.cse_after_recomputation.cse : 0.000017s : 0.00% optimize.environ_conv : 0.000010s : 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 : 0.000005s : 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 : 0.000001s : 0.00% optimize.ForceFp32Comm : 0.000001s : 0.00% optimize.remove_cast_before_assign_add : 0.000001s : 0.00% optimize.full_micro_interleaved_order_control : 0.000002s : 0.00% optimize.reorder_send_recv_between_fp_bp : 0.000003s : 0.00% optimize.comm_op_add_attrs : 0.000001s : 0.00% optimize.add_comm_op_reuse_tag : 0.000001s : 0.00% optimize.interleave_split_concat_branches : 0.000001s : 0.00% optimize.interleave_parallel_branches : 0.000001s : 0.00% optimize.overlap_opt_shard_in_pipeline : 0.000001s : 0.00% optimize.overlap_opt_shard_grad_in_pipeline : 0.000002s : 0.00% optimize.control_data_broadcast_order : 0.000015s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000001s : 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 : 0.000003s : 0.00% optimize.overlap_grad_ring_attention : 0.000004s : 0.00% optimize.overlap_grad_flash_sp : 0.000022s : 0.00% optimize.begin_end_overlap_inline : 0.000000s : 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 : 0.000012s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000013s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000006s : 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 : 0.000001s : 0.00% auto_monad_reorder : 0.000019s : 0.00% get_jit_bprop_graph : 0.000002s : 0.00% rewriter_after_jit_bprop_graph : 0.000005s : 0.00% opt_after_jit_grad : 0.014337s : 0.85% validate : 0.002457s : 0.15% Time group info: ------[substitution.] 0.000344 28 0.58% : 0.000002s : 2: substitution.elim_not_effective 0.36% : 0.000001s : 2: substitution.fold_const_symbol 1.77% : 0.000006s : 3: substitution.graph_param_transform 86.59% : 0.000298s : 7: substitution.inline 1.55% : 0.000005s : 4: substitution.j_node_and_user_rematch 1.68% : 0.000006s : 4: substitution.remove_not_recompute_node 2.00% : 0.000007s : 2: substitution.replace_old_param 5.46% : 0.000019s : 4: substitution.tuple_list_get_item_eliminator ------[type_inference.] 1.565193 2 96.67% : 1.513076s : 1: type_inference.infer 3.33% : 0.052117s : 1: type_inference.specialize ------[replace.] 0.013720 11 99.70% : 0.013680s : 7: replace.inline 0.30% : 0.000040s : 4: replace.tuple_list_get_item_eliminator ------[match.] 0.000308 11 94.89% : 0.000293s : 7: match.inline 5.11% : 0.000016s : 4: match.tuple_list_get_item_eliminator ------[predicate.] 0.003913 1305 0.08% : 0.000003s : 15: predicate.accumulaten_eliminater 0.11% : 0.000004s : 3: predicate.ad_related_special_op_eliminate 0.02% : 0.000001s : 6: predicate.addn_check_dump 0.08% : 0.000003s : 15: predicate.addn_zero_filter 0.05% : 0.000002s : 15: predicate.adjust_all_reduce_mul_add 0.16% : 0.000006s : 21: predicate.arithmetic_simplify 0.07% : 0.000003s : 15: predicate.cast_eliminate 0.03% : 0.000001s : 6: predicate.check_bprop_eliminate 0.02% : 0.000001s : 6: predicate.compare_switch_simplify 0.01% : 0.000000s : 3: predicate.const_output_eliminate 0.03% : 0.000001s : 6: predicate.depend_value_elim 0.06% : 0.000002s : 15: predicate.dict_get_item_const_eliminator 0.08% : 0.000003s : 15: predicate.dict_get_item_eliminator 0.06% : 0.000003s : 15: predicate.dict_set_item_eliminator 0.11% : 0.000004s : 6: predicate.dumpgradient_eliminate 0.01% : 0.000000s : 3: predicate.elim_not_effective 0.02% : 0.000001s : 3: predicate.elim_shapecalc_of_broadcastargs 0.07% : 0.000003s : 18: predicate.environ_add_const_eliminate 0.07% : 0.000003s : 18: predicate.environ_get_add_eliminate 0.07% : 0.000003s : 18: predicate.environ_get_depend_swap 0.10% : 0.000004s : 24: predicate.environ_get_eliminate 0.06% : 0.000002s : 18: predicate.environ_get_set_eliminate 0.10% : 0.000004s : 26: predicate.exchange_switch_depend_value 0.19% : 0.000007s : 26: predicate.float_depend_g_call 0.04% : 0.000001s : 6: predicate.float_environ_get_switch 0.04% : 0.000001s : 9: predicate.float_tuple_getitem_switch 0.01% : 0.000000s : 3: predicate.fold_const_symbol 0.03% : 0.000001s : 6: predicate.get_grad_eliminate 0.01% : 0.000001s : 3: predicate.graph_param_transform 0.02% : 0.000001s : 6: predicate.incorporate_call 0.02% : 0.000001s : 6: predicate.incorporate_call_switch 0.40% : 0.000016s : 59: predicate.inline 0.04% : 0.000002s : 6: predicate.inline_without_move 0.01% : 0.000001s : 6: predicate.j_node_and_user_rematch 0.10% : 0.000004s : 6: predicate.less_batch_normalization 0.11% : 0.000004s : 25: predicate.list_to_tuple_eliminator_ 0.17% : 0.000007s : 40: predicate.load_eliminater 0.05% : 0.000002s : 3: predicate.loop_unroll_after_grad 0.19% : 0.000007s : 56: predicate.loop_unroll_before_grad 0.12% : 0.000005s : 21: predicate.make_slice_get_slice_eliminator 0.02% : 0.000001s : 6: predicate.merge_addn 0.05% : 0.000002s : 6: predicate.micro_step_allgather_replace 93.51% : 0.003659s : 6: predicate.mini_step_allgather_replace 0.05% : 0.000002s : 15: predicate.minmaximum_grad 0.06% : 0.000002s : 3: predicate.mutable_eliminate 0.02% : 0.000001s : 3: predicate.opt_reshape 0.03% : 0.000001s : 3: predicate.parallel_virtual_node 0.19% : 0.000007s : 26: predicate.partial_defer_inline 0.09% : 0.000003s : 22: predicate.partial_eliminate 0.07% : 0.000003s : 15: predicate.print_const_string_wrapper 0.07% : 0.000003s : 6: predicate.reduce_all_const_elim 0.10% : 0.000004s : 15: predicate.reduce_eliminate 0.15% : 0.000006s : 40: predicate.redundant_stop_gradient_eliminater 0.02% : 0.000001s : 6: predicate.remove_not_recompute_node 0.08% : 0.000003s : 25: predicate.replace_applicator 0.03% : 0.000001s : 6: predicate.replace_old_param 0.02% : 0.000001s : 3: predicate.reset_defer_inline 0.08% : 0.000003s : 15: predicate.reshape_eliminate 0.03% : 0.000001s : 6: predicate.row_tensor_add_zeros_like 0.02% : 0.000001s : 3: predicate.row_tensor_eliminate 0.05% : 0.000002s : 6: predicate.same_eliminate 0.02% : 0.000001s : 6: predicate.set_cell_output_no_recompute 0.03% : 0.000001s : 6: predicate.shard_identity_eliminate 0.03% : 0.000001s : 6: predicate.special_op_eliminate 0.03% : 0.000001s : 6: predicate.specialize_transform 0.08% : 0.000003s : 6: predicate.split_environ_get_set_with_tuple_value 0.04% : 0.000001s : 6: predicate.stack_unstack_eliminate 0.02% : 0.000001s : 3: predicate.switch_call_monad_eliminater 0.11% : 0.000004s : 26: predicate.switch_defer_inline 0.13% : 0.000005s : 32: predicate.switch_layer_defer_inline 0.34% : 0.000013s : 91: predicate.switch_simplify 0.07% : 0.000003s : 15: predicate.tile_eliminate 0.06% : 0.000002s : 15: predicate.transpose_eliminate 0.09% : 0.000004s : 21: predicate.tuple_list_convert_item_index_to_positive 0.09% : 0.000003s : 21: predicate.tuple_list_get_item_const_eliminator 0.09% : 0.000004s : 21: predicate.tuple_list_get_item_depend_reorder 0.19% : 0.000007s : 31: predicate.tuple_list_get_item_eliminator 0.11% : 0.000004s : 21: predicate.tuple_list_get_set_item_eliminator 0.14% : 0.000005s : 27: predicate.tuple_list_set_item_eliminator 0.12% : 0.000005s : 25: predicate.tuple_to_list_eliminator_ 0.14% : 0.000005s : 40: predicate.updatestate_pure_node_eliminater 0.17% : 0.000007s : 46: predicate.updatestate_useless_node_eliminater 0.02% : 0.000001s : 3: predicate.value_based_eliminate 0.03% : 0.000001s : 6: predicate.virtual_dataset_eliminate 0.03% : 0.000001s : 6: predicate.virtual_output_eliminate 0.02% : 0.000001s : 3: predicate.virtual_view_grad_eliminate 0.02% : 0.000001s : 3: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.005015 31 62.24% : 0.003121s : 22: func_graph_cloner_run.FuncGraphClonerGraph 37.76% : 0.001894s : 9: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.916586 192 0.00% : 0.000004s : 1: ForceFp32Comm 1.37% : 0.026280s : 1: add_attr 1.37% : 0.026263s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.00% : 0.000058s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.46% : 0.008810s : 1: auto_monad 0.00% : 0.000022s : 1: auto_monad_reorder 0.00% : 0.000004s : 1: begin_end_overlap_inline 0.00% : 0.000006s : 1: bias_add_comm_swap 0.03% : 0.000505s : 1: bootstrap 0.00% : 0.000041s : 1: cconv 0.00% : 0.000004s : 1: comm_op_add_attrs 0.00% : 0.000019s : 1: control_data_broadcast_order 0.00% : 0.000005s : 1: convert_after_rewriter 0.00% : 0.000030s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000005s : 1: detach_backward 0.00% : 0.000014s : 1: environ_conv 0.32% : 0.006078s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000005s : 1: get_jit_bprop_graph 0.00% : 0.000035s : 1: graph_reusing 0.00% : 0.000005s : 1: grouped_pairwise_exchange_alltoall 0.00% : 0.000004s : 1: handle_group_info 0.00% : 0.000014s : 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% : 0.000007s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.03% : 0.000486s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000007s : 1: micro_interleaved_order_control 0.04% : 0.000805s : 1: mutable_eliminate 0.00% : 0.000008s : 1: offloading_packed_experts 0.00% : 0.000016s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000020s : 1: opt.transform.mutable_eliminate 1.23% : 0.023562s : 78: opt.transform.opt_a 0.00% : 0.000026s : 1: opt.transform.opt_after_cconv 0.00% : 0.000053s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000104s : 28: opt.transform.opt_b 0.00% : 0.000048s : 2: opt.transform.opt_trans_graph 0.00% : 0.000038s : 4: opt.transform.symbol_engine_opt 4.59% : 0.088030s : 1: opt_a 0.01% : 0.000136s : 1: opt_after_cconv 0.75% : 0.014369s : 1: opt_after_jit_grad 0.01% : 0.000249s : 1: opt_b 4.74% : 0.090835s : 1: optimize 0.00% : 0.000026s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 1: order_py_execute_after_rewriter 0.00% : 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% : 0.000005s : 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% : 0.000006s : 1: partial_unused_args_eliminate 0.00% : 0.000005s : 1: pipeline_parallel_scheduler 0.00% : 0.000005s : 1: pipeline_split 0.00% : 0.000064s : 1: pre_auto_parallel 0.00% : 0.000016s : 1: py_interpret_to_execute 0.00% : 0.000019s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000005s : 1: remove_cast_before_assign_add 0.00% : 0.000053s : 1: remove_dup_value 2.47% : 0.047303s : 1: renormalize.infer 0.83% : 0.015868s : 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% : 0.000042s : 1: rewriter_after_opt_a 0.02% : 0.000383s : 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 0.00% : 0.000081s : 1: symbol_engine_optimizer 0.00% : 0.000080s : 1: tuple_transform 81.68% : 1.565445s : 1: type_inference TotalTime = 1.33294, [21] [bootstrap]: 0.00046508 [type_inference]: 1.05357 [event_method]: 0.00020532 [auto_monad]: 0.00018602 [graph_reusing]: 1.137e-05 [inline]: 2.76999e-06 [add_attr]: 0.0450294, [1] [add_attr_with_inline]: 0.0450152, [1] [Cycle 1]: 0.00013166, [2] [tag_attr]: 6.83e-05 [meta_addattr_fg_expand]: 1.338e-05 [parallel-infer-symbol]: 3.95e-06 [pre_auto_parallel]: 0.00013923 [insert-virtual-dataset]: 2.49001e-06 [parallel-infer-symbol-second]: 2.46e-06 [dataset_repeat_opt]: 2.46e-06 [pipeline_split]: 1.81e-06 [optimize]: 0.232424, [53] [py_interpret_to_execute]: 1.122e-05 [rewriter_before_opt_a]: 0.00056511 [opt_a]: 0.229342, [3] [Cycle 1]: 0.177238, [45] [expand_dump_flag]: 6.63e-06 [switch_simplify]: 0.00019393 [loop_unroll]: 9.141e-05 [a_1]: 0.00963539 [with_stream_mark]: 4.212e-05 [recompute_prepare]: 3.537e-05 [updatestate_depend_eliminate]: 1.167e-05 [updatestate_assign_eliminate]: 8.25e-06 [updatestate_loads_eliminate]: 7.28999e-06 [parameter_eliminate]: 3.74002e-06 [a_2]: 0.00023433 [accelerated_algorithm]: 1.723e-05 [shard]: 2.49999e-06 [meta_shard_fg_expand]: 8.67998e-06 [shard_inline]: 1.626e-05 [merge_send_recv]: 2.075e-05 [auto_parallel]: 1.766e-05 [parallel]: 2.084e-05 [flash_sp]: 1.471e-05 [merge_comm]: 9.79e-06 [allreduce_fusion]: 8.84e-06 [matmul_add_comm_reduction]: 3.348e-05 [allreduce_slice_to_reducescatter]: 1.07998e-06 [virtual_shard_identity]: 2.328e-05 [virtual_dataset]: 1.555e-05 [get_grad_eliminate_]: 1.529e-05 [virtual_output]: 1.552e-05 [merge_forward]: 0.00673984 [cell_reuse_recompute_pass]: 7.41999e-06 [offload_activation]: 4.162e-05 [cell_reuse_handle_not_recompute_node_pass]: 6.729e-05 [merge_recompute_call_nodes]: 2.91999e-06 [before_grad]: 3.118e-05 [set_forward_comm_id_for_comm_node_pass]: 2.271e-05 [meta_fg_expand]: 0.00859976 [flash_sp_send_recv_attached]: 6.90998e-06 [receive_attached]: 2.37999e-06 [after_resolve]: 0.00010825 [a_after_grad]: 0.00010831 [renormalize]: 0.149588 [add_forward_monad_depend]: 1.891e-05 [auto_monad_grad]: 8.27e-06 [auto_monad_eliminator]: 6.885e-05 [cse]: 0.00044087 [a_3]: 0.00036571 [Cycle 2]: 0.0513113, [45] [expand_dump_flag]: 3.4e-06 [switch_simplify]: 4.921e-05 [loop_unroll]: 4.277e-05 [a_1]: 0.00134429 [with_stream_mark]: 3.101e-05 [recompute_prepare]: 1.188e-05 [updatestate_depend_eliminate]: 6.16e-06 [updatestate_assign_eliminate]: 5.44e-06 [updatestate_loads_eliminate]: 3.41001e-06 [parameter_eliminate]: 2.78e-06 [a_2]: 9.094e-05 [accelerated_algorithm]: 8.09002e-06 [shard]: 2.42001e-06 [meta_shard_fg_expand]: 3.02002e-06 [shard_inline]: 7.18998e-06 [merge_send_recv]: 9.54e-06 [auto_parallel]: 1.076e-05 [parallel]: 9.25999e-06 [flash_sp]: 3.88001e-06 [merge_comm]: 5.06997e-06 [allreduce_fusion]: 4.00998e-06 [matmul_add_comm_reduction]: 1.128e-05 [allreduce_slice_to_reducescatter]: 8.30012e-07 [virtual_shard_identity]: 9.99999e-06 [virtual_dataset]: 7.08e-06 [get_grad_eliminate_]: 6.33e-06 [virtual_output]: 7.64002e-06 [merge_forward]: 5.41998e-06 [cell_reuse_recompute_pass]: 1.89e-06 [offload_activation]: 1.14e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.703e-05 [merge_recompute_call_nodes]: 1.50999e-06 [before_grad]: 1.198e-05 [set_forward_comm_id_for_comm_node_pass]: 4.21001e-06 [meta_fg_expand]: 0.0313475 [flash_sp_send_recv_attached]: 5.74e-06 [receive_attached]: 4.2e-06 [after_resolve]: 0.00078248 [a_after_grad]: 2.619e-05 [renormalize]: 0.0168612 [add_forward_monad_depend]: 1.202e-05 [auto_monad_grad]: 2.68e-06 [auto_monad_eliminator]: 2.811e-05 [cse]: 4.93e-05 [a_3]: 6.814e-05 [Cycle 3]: 0.00076911, [45] [expand_dump_flag]: 2.03002e-06 [switch_simplify]: 1.107e-05 [loop_unroll]: 7.36001e-06 [a_1]: 0.00016477 [with_stream_mark]: 1.747e-05 [recompute_prepare]: 7.32997e-06 [updatestate_depend_eliminate]: 5.64e-06 [updatestate_assign_eliminate]: 3.8e-06 [updatestate_loads_eliminate]: 3.55e-06 [parameter_eliminate]: 2.34001e-06 [a_2]: 8.406e-05 [accelerated_algorithm]: 7.3e-06 [shard]: 2.21e-06 [meta_shard_fg_expand]: 2.32001e-06 [shard_inline]: 6.93998e-06 [merge_send_recv]: 1.031e-05 [auto_parallel]: 1.029e-05 [parallel]: 9.71e-06 [flash_sp]: 1.86e-06 [merge_comm]: 4.43001e-06 [allreduce_fusion]: 4.3e-06 [matmul_add_comm_reduction]: 9.92001e-06 [allreduce_slice_to_reducescatter]: 1.22999e-06 [virtual_shard_identity]: 8.28999e-06 [virtual_dataset]: 6.59999e-06 [get_grad_eliminate_]: 6.31e-06 [virtual_output]: 7.06999e-06 [merge_forward]: 4.76002e-06 [cell_reuse_recompute_pass]: 4.12998e-06 [offload_activation]: 1.133e-05 [cell_reuse_handle_not_recompute_node_pass]: 1.56e-05 [merge_recompute_call_nodes]: 1.54e-06 [before_grad]: 1.189e-05 [set_forward_comm_id_for_comm_node_pass]: 4.80999e-06 [meta_fg_expand]: 3.53e-06 [flash_sp_send_recv_attached]: 2.09e-06 [receive_attached]: 2.37001e-06 [after_resolve]: 7.1e-06 [a_after_grad]: 9.94001e-06 [renormalize]: 9.00181e-08 [add_forward_monad_depend]: 1.44e-06 [auto_monad_grad]: 1.60001e-06 [auto_monad_eliminator]: 9.25999e-06 [cse]: 2.234e-05 [a_3]: 4.155e-05 [py_interpret_to_execute_after_opt_a]: 1.102e-05 [slice_cell_reuse_recomputed_activation]: 2.06e-06 [rewriter_after_opt_a]: 2.869e-05 [convert_after_rewriter]: 1.22e-06 [order_py_execute_after_rewriter]: 1.11997e-06 [mutable_eliminate]: 0.00082178 [opt_b]: 0.0002561, [1] [Cycle 1]: 0.0002472, [7] [b_1]: 0.00014803 [b_2]: 9.66003e-06 [updatestate_depend_eliminate]: 1.057e-05 [updatestate_assign_eliminate]: 3.30003e-06 [updatestate_loads_eliminate]: 3.16001e-06 [renormalize]: 8.89995e-07 [cse]: 3.459e-05 [optimize_parallel_all_gather_comm]: 2.067e-05 [overlap_param_gather]: 2.09999e-06 [cconv]: 3.706e-05 [loop_unroll]: 0.00049014 [opt_after_cconv]: 0.00011917, [1] [Cycle 1]: 0.00011299, [7] [c_1]: 3.316e-05 [parameter_eliminate]: 5.88002e-06 [updatestate_depend_eliminate]: 7.43999e-06 [updatestate_assign_eliminate]: 2.99999e-06 [updatestate_loads_eliminate]: 2.93e-06 [cse]: 2.702e-05 [renormalize]: 5.99975e-07 [remove_dup_value]: 2.395e-05 [tuple_transform]: 0.00013291, [1] [Cycle 1]: 0.00012827, [4] [d_1]: 9.893e-05 [none_parameter_eliminate]: 1.94999e-06 [renormalize]: 1.80007e-07 [switch_simplify]: 8.02e-06 [partial_unused_args_eliminate]: 2.19999e-06 [add_recomputation]: 5.65e-05 [cse_after_recomputation]: 3.159e-05, [1] [Cycle 1]: 2.704e-05, [1] [cse]: 2.095e-05 [environ_conv]: 1.13e-05 [swap_dp_allreduce_reducescatter]: 5.92999e-06 [bias_add_comm_swap]: 2.71e-06 [label_micro_interleaved_index]: 4.58999e-06 [label_fine_grained_interleaved_index]: 2.67001e-06 [merge_cast_opt]: 1.34003e-06 [slice_recompute_activation]: 2.58e-06 [micro_interleaved_order_control]: 2.68e-06 [assign_add_opt]: 1.40001e-06 [ForceFp32Comm]: 1.18001e-06 [remove_cast_before_assign_add]: 1.02e-06 [full_micro_interleaved_order_control]: 2.01e-06 [reorder_send_recv_between_fp_bp]: 3.09999e-06 [comm_op_add_attrs]: 1.02e-06 [add_comm_op_reuse_tag]: 1.16997e-06 [interleave_split_concat_branches]: 1.26997e-06 [interleave_parallel_branches]: 1.09e-06 [overlap_opt_shard_in_pipeline]: 1.22e-06 [overlap_opt_shard_grad_in_pipeline]: 1.74e-06 [control_data_broadcast_order]: 1.596e-05 [grouped_pairwise_exchange_alltoall]: 1.41002e-06 [offloading_packed_experts]: 5.09e-06 [overlap_recompute_and_grad_model_parallel]: 5.38002e-06 [overlap_grad_matmul_and_grad_allreduce]: 1.39e-06 [overlap_recompute_allgather_and_fa_grad]: 1.24003e-06 [overlap_recompute_comm]: 2.31998e-06 [overlap_grad_ring_attention]: 5.34e-06 [overlap_grad_flash_sp]: 2.471e-05 [begin_end_overlap_inline]: 5.29981e-07 [split_matmul_comm_elemetwise]: 2.25002e-06 [split_layernorm_comm]: 1.73997e-06 [handle_group_info]: 1.00999e-06 [symbol_engine_optimizer]: 8.942e-05, [1] [Cycle 1]: 8.479e-05, [6] [build]: 3.39001e-06 [elim_shapecalc]: 1.178e-05 [elim_not_effective]: 1.556e-05 [opt_reshape]: 1.202e-05 [fold_const_symbol]: 1.243e-05 [renormalize]: 2.00002e-07 [detach_backward]: 2.37999e-06 [pipeline_parallel_scheduler]: 1.43002e-06 [auto_monad_reorder]: 2.07e-05 [get_jit_bprop_graph]: 2.76e-06 [rewriter_after_jit_bprop_graph]: 5.76e-06 [opt_after_jit_grad]: 0.00055042 [validate]: 0.000101 Sums bootstrap : 0.000465s : 0.04% type_inference : 1.053567s : 81.91% event_method : 0.000205s : 0.02% auto_monad : 0.000186s : 0.01% graph_reusing : 0.000011s : 0.00% inline : 0.000003s : 0.00% add_attr.add_attr_with_inline.tag_attr : 0.000068s : 0.01% add_attr.add_attr_with_inline.meta_addattr_fg_expand : 0.000013s : 0.00% parallel-infer-symbol : 0.000004s : 0.00% pre_auto_parallel : 0.000139s : 0.01% insert-virtual-dataset : 0.000002s : 0.00% parallel-infer-symbol-second : 0.000002s : 0.00% dataset_repeat_opt : 0.000002s : 0.00% pipeline_split : 0.000002s : 0.00% optimize.py_interpret_to_execute : 0.000011s : 0.00% optimize.rewriter_before_opt_a : 0.000565s : 0.04% optimize.opt_a.expand_dump_flag : 0.000012s : 0.00% optimize.opt_a.switch_simplify : 0.000254s : 0.02% optimize.opt_a.loop_unroll : 0.000142s : 0.01% optimize.opt_a.a_1 : 0.011144s : 0.87% optimize.opt_a.with_stream_mark : 0.000091s : 0.01% optimize.opt_a.recompute_prepare : 0.000055s : 0.00% optimize.opt_a.updatestate_depend_eliminate : 0.000023s : 0.00% optimize.opt_a.updatestate_assign_eliminate : 0.000017s : 0.00% optimize.opt_a.updatestate_loads_eliminate : 0.000014s : 0.00% optimize.opt_a.parameter_eliminate : 0.000009s : 0.00% optimize.opt_a.a_2 : 0.000409s : 0.03% optimize.opt_a.accelerated_algorithm : 0.000033s : 0.00% optimize.opt_a.shard : 0.000007s : 0.00% optimize.opt_a.meta_shard_fg_expand : 0.000014s : 0.00% optimize.opt_a.shard_inline : 0.000030s : 0.00% optimize.opt_a.merge_send_recv : 0.000041s : 0.00% optimize.opt_a.auto_parallel : 0.000039s : 0.00% optimize.opt_a.parallel : 0.000040s : 0.00% optimize.opt_a.flash_sp : 0.000020s : 0.00% optimize.opt_a.merge_comm : 0.000019s : 0.00% optimize.opt_a.allreduce_fusion : 0.000017s : 0.00% optimize.opt_a.matmul_add_comm_reduction : 0.000055s : 0.00% optimize.opt_a.allreduce_slice_to_reducescatter : 0.000003s : 0.00% optimize.opt_a.virtual_shard_identity : 0.000042s : 0.00% optimize.opt_a.virtual_dataset : 0.000029s : 0.00% optimize.opt_a.get_grad_eliminate_ : 0.000028s : 0.00% optimize.opt_a.virtual_output : 0.000030s : 0.00% optimize.opt_a.merge_forward : 0.006750s : 0.52% optimize.opt_a.cell_reuse_recompute_pass : 0.000013s : 0.00% optimize.opt_a.offload_activation : 0.000064s : 0.01% optimize.opt_a.cell_reuse_handle_not_recompute_node_pass : 0.000100s : 0.01% optimize.opt_a.merge_recompute_call_nodes : 0.000006s : 0.00% optimize.opt_a.before_grad : 0.000055s : 0.00% optimize.opt_a.set_forward_comm_id_for_comm_node_pass : 0.000032s : 0.00% optimize.opt_a.meta_fg_expand : 0.039951s : 3.11% optimize.opt_a.flash_sp_send_recv_attached : 0.000015s : 0.00% optimize.opt_a.receive_attached : 0.000009s : 0.00% optimize.opt_a.after_resolve : 0.000898s : 0.07% optimize.opt_a.a_after_grad : 0.000144s : 0.01% optimize.opt_a.renormalize : 0.166450s : 12.94% optimize.opt_a.add_forward_monad_depend : 0.000032s : 0.00% optimize.opt_a.auto_monad_grad : 0.000013s : 0.00% optimize.opt_a.auto_monad_eliminator : 0.000106s : 0.01% optimize.opt_a.cse : 0.000513s : 0.04% optimize.opt_a.a_3 : 0.000475s : 0.04% optimize.py_interpret_to_execute_after_opt_a : 0.000011s : 0.00% optimize.slice_cell_reuse_recomputed_activation : 0.000002s : 0.00% optimize.rewriter_after_opt_a : 0.000029s : 0.00% optimize.convert_after_rewriter : 0.000001s : 0.00% optimize.order_py_execute_after_rewriter : 0.000001s : 0.00% optimize.mutable_eliminate : 0.000822s : 0.06% optimize.opt_b.b_1 : 0.000148s : 0.01% optimize.opt_b.b_2 : 0.000010s : 0.00% optimize.opt_b.updatestate_depend_eliminate : 0.000011s : 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 : 0.000035s : 0.00% optimize.optimize_parallel_all_gather_comm : 0.000021s : 0.00% optimize.overlap_param_gather : 0.000002s : 0.00% optimize.cconv : 0.000037s : 0.00% optimize.loop_unroll : 0.000490s : 0.04% optimize.opt_after_cconv.c_1 : 0.000033s : 0.00% 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 : 0.000003s : 0.00% optimize.opt_after_cconv.cse : 0.000027s : 0.00% optimize.opt_after_cconv.renormalize : 0.000001s : 0.00% optimize.remove_dup_value : 0.000024s : 0.00% optimize.tuple_transform.d_1 : 0.000099s : 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 : 0.000008s : 0.00% optimize.partial_unused_args_eliminate : 0.000002s : 0.00% optimize.add_recomputation : 0.000057s : 0.00% optimize.cse_after_recomputation.cse : 0.000021s : 0.00% optimize.environ_conv : 0.000011s : 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 : 0.000005s : 0.00% optimize.label_fine_grained_interleaved_index : 0.000003s : 0.00% optimize.merge_cast_opt : 0.000001s : 0.00% optimize.slice_recompute_activation : 0.000003s : 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 : 0.000001s : 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 : 0.000016s : 0.00% optimize.grouped_pairwise_exchange_alltoall : 0.000001s : 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 : 0.000005s : 0.00% optimize.overlap_grad_flash_sp : 0.000025s : 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 : 0.000012s : 0.00% optimize.symbol_engine_optimizer.elim_not_effective : 0.000016s : 0.00% optimize.symbol_engine_optimizer.opt_reshape : 0.000012s : 0.00% optimize.symbol_engine_optimizer.fold_const_symbol : 0.000012s : 0.00% optimize.symbol_engine_optimizer.renormalize : 0.000000s : 0.00% detach_backward : 0.000002s : 0.00% pipeline_parallel_scheduler : 0.000001s : 0.00% auto_monad_reorder : 0.000021s : 0.00% get_jit_bprop_graph : 0.000003s : 0.00% rewriter_after_jit_bprop_graph : 0.000006s : 0.00% opt_after_jit_grad : 0.000550s : 0.04% validate : 0.000101s : 0.01% Time group info: ------[substitution.] 0.001181 174 0.18% : 0.000002s : 3: substitution.elim_not_effective 1.04% : 0.000012s : 12: substitution.float_depend_g_call 0.34% : 0.000004s : 2: substitution.float_tuple_getitem_switch 0.23% : 0.000003s : 3: substitution.fold_const_symbol 4.54% : 0.000054s : 4: substitution.graph_param_transform 0.26% : 0.000003s : 2: substitution.incorporate_call 0.18% : 0.000002s : 2: substitution.incorporate_call_switch 67.46% : 0.000796s : 23: substitution.inline 2.60% : 0.000031s : 2: substitution.inline_without_move 0.97% : 0.000011s : 14: substitution.j_node_and_user_rematch 1.04% : 0.000012s : 7: substitution.minmaximum_grad 0.11% : 0.000001s : 1: substitution.opt_reshape 1.16% : 0.000014s : 12: substitution.partial_eliminate 1.11% : 0.000013s : 14: substitution.remove_not_recompute_node 2.72% : 0.000032s : 9: substitution.replace_applicator 0.81% : 0.000010s : 11: substitution.replace_old_param 1.48% : 0.000018s : 2: substitution.reshape_eliminate 0.50% : 0.000006s : 1: substitution.set_cell_output_no_recompute 1.46% : 0.000017s : 3: substitution.switch_simplify 2.12% : 0.000025s : 7: substitution.tuple_list_convert_item_index_to_positive 0.88% : 0.000010s : 7: substitution.tuple_list_get_item_const_eliminator 1.33% : 0.000016s : 7: substitution.tuple_list_get_item_depend_reorder 6.23% : 0.000074s : 19: substitution.tuple_list_get_item_eliminator 1.25% : 0.000015s : 7: substitution.tuple_list_get_set_item_eliminator ------[type_inference.] 1.053369 2 97.20% : 1.023878s : 1: type_inference.infer 2.80% : 0.029491s : 1: type_inference.specialize ------[replace.] 0.000463 36 52.62% : 0.000244s : 23: replace.inline 10.60% : 0.000049s : 3: replace.switch_simplify 36.77% : 0.000170s : 10: replace.tuple_list_get_item_eliminator ------[match.] 0.000834 36 93.66% : 0.000781s : 23: match.inline 1.50% : 0.000013s : 3: match.switch_simplify 4.83% : 0.000040s : 10: match.tuple_list_get_item_eliminator ------[predicate.] 0.000697 4535 1.19% : 0.000008s : 57: predicate.accumulaten_eliminater 0.31% : 0.000002s : 4: predicate.ad_related_special_op_eliminate 0.38% : 0.000003s : 20: predicate.addn_check_dump 1.14% : 0.000008s : 57: predicate.addn_zero_filter 0.99% : 0.000007s : 57: predicate.adjust_all_reduce_mul_add 2.08% : 0.000015s : 77: predicate.arithmetic_simplify 1.22% : 0.000008s : 57: predicate.cast_eliminate 0.93% : 0.000007s : 49: predicate.check_bprop_eliminate 0.39% : 0.000003s : 20: predicate.compare_switch_simplify 0.06% : 0.000000s : 4: predicate.const_output_eliminate 0.43% : 0.000003s : 20: predicate.depend_value_elim 1.16% : 0.000008s : 57: predicate.dict_get_item_const_eliminator 1.32% : 0.000009s : 57: predicate.dict_get_item_eliminator 1.24% : 0.000009s : 57: predicate.dict_set_item_eliminator 0.30% : 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 1.19% : 0.000008s : 61: predicate.environ_add_const_eliminate 1.09% : 0.000008s : 61: predicate.environ_get_add_eliminate 1.12% : 0.000008s : 61: predicate.environ_get_depend_swap 1.60% : 0.000011s : 81: predicate.environ_get_eliminate 1.12% : 0.000008s : 61: predicate.environ_get_set_eliminate 1.88% : 0.000013s : 90: predicate.exchange_switch_depend_value 2.81% : 0.000020s : 90: predicate.float_depend_g_call 0.40% : 0.000003s : 20: predicate.float_environ_get_switch 0.61% : 0.000004s : 24: predicate.float_tuple_getitem_switch 0.05% : 0.000000s : 4: predicate.fold_const_symbol 0.47% : 0.000003s : 20: predicate.get_grad_eliminate 0.07% : 0.000000s : 4: predicate.graph_param_transform 0.44% : 0.000003s : 20: predicate.incorporate_call 0.35% : 0.000002s : 20: predicate.incorporate_call_switch 5.47% : 0.000038s : 195: predicate.inline 1.72% : 0.000012s : 50: predicate.inline_without_move 0.20% : 0.000001s : 20: predicate.j_node_and_user_rematch 0.64% : 0.000004s : 20: predicate.less_batch_normalization 1.51% : 0.000011s : 75: predicate.list_to_tuple_eliminator_ 2.49% : 0.000017s : 132: predicate.load_eliminater 0.31% : 0.000002s : 4: predicate.loop_unroll_after_grad 3.00% : 0.000021s : 150: predicate.loop_unroll_before_grad 1.36% : 0.000010s : 65: predicate.make_slice_get_slice_eliminator 0.48% : 0.000003s : 20: predicate.merge_addn 0.95% : 0.000007s : 49: predicate.micro_step_allgather_replace 0.95% : 0.000007s : 49: predicate.mini_step_allgather_replace 1.16% : 0.000008s : 57: predicate.minmaximum_grad 0.47% : 0.000003s : 4: predicate.mutable_eliminate 0.14% : 0.000001s : 4: predicate.opt_reshape 0.12% : 0.000001s : 4: predicate.parallel_virtual_node 2.87% : 0.000020s : 90: predicate.partial_defer_inline 1.51% : 0.000011s : 71: predicate.partial_eliminate 1.09% : 0.000008s : 57: predicate.print_const_string_wrapper 0.43% : 0.000003s : 20: predicate.reduce_all_const_elim 1.33% : 0.000009s : 57: predicate.reduce_eliminate 2.42% : 0.000017s : 132: predicate.redundant_stop_gradient_eliminater 0.43% : 0.000003s : 20: predicate.remove_not_recompute_node 1.83% : 0.000013s : 116: predicate.replace_applicator 0.79% : 0.000006s : 50: predicate.replace_old_param 0.11% : 0.000001s : 4: predicate.reset_defer_inline 1.35% : 0.000009s : 57: predicate.reshape_eliminate 0.99% : 0.000007s : 49: predicate.row_tensor_add_zeros_like 0.13% : 0.000001s : 4: predicate.row_tensor_eliminate 1.28% : 0.000009s : 49: predicate.same_eliminate 0.34% : 0.000002s : 20: predicate.set_cell_output_no_recompute 0.59% : 0.000004s : 20: predicate.shard_identity_eliminate 0.23% : 0.000002s : 8: predicate.special_op_eliminate 0.54% : 0.000004s : 20: predicate.specialize_transform 1.23% : 0.000009s : 49: predicate.split_environ_get_set_with_tuple_value 1.38% : 0.000010s : 50: predicate.stack_unstack_eliminate 0.12% : 0.000001s : 4: predicate.switch_call_monad_eliminater 1.93% : 0.000013s : 90: predicate.switch_defer_inline 3.06% : 0.000021s : 139: predicate.switch_layer_defer_inline 5.89% : 0.000041s : 270: predicate.switch_simplify 1.06% : 0.000007s : 57: predicate.tile_eliminate 1.07% : 0.000007s : 57: predicate.transpose_eliminate 1.32% : 0.000009s : 65: predicate.tuple_list_convert_item_index_to_positive 1.45% : 0.000010s : 65: predicate.tuple_list_get_item_const_eliminator 1.36% : 0.000009s : 65: predicate.tuple_list_get_item_depend_reorder 3.07% : 0.000021s : 95: predicate.tuple_list_get_item_eliminator 1.37% : 0.000010s : 65: predicate.tuple_list_get_set_item_eliminator 1.98% : 0.000014s : 85: predicate.tuple_list_set_item_eliminator 1.48% : 0.000010s : 75: predicate.tuple_to_list_eliminator_ 2.34% : 0.000016s : 132: predicate.updatestate_pure_node_eliminater 2.85% : 0.000020s : 152: predicate.updatestate_useless_node_eliminater 0.14% : 0.000001s : 4: predicate.value_based_eliminate 0.47% : 0.000003s : 20: predicate.virtual_dataset_eliminate 0.50% : 0.000004s : 20: predicate.virtual_output_eliminate 0.09% : 0.000001s : 4: predicate.virtual_view_grad_eliminate 0.14% : 0.000001s : 4: predicate.zero_like_fill_zero ------[func_graph_cloner_run.] 0.034313 68 63.60% : 0.021823s : 41: func_graph_cloner_run.FuncGraphClonerGraph 36.40% : 0.012490s : 27: func_graph_cloner_run.FuncGraphSpecializer ------[meta_graph.] 0.000000 0 ------[manager.] 0.000000 0 ------[pynative] 0.000000 0 ------[others.] 1.789909 233 0.00% : 0.000005s : 1: ForceFp32Comm 2.52% : 0.045036s : 1: add_attr 2.52% : 0.045021s : 1: add_attr_with_inline 0.00% : 0.000004s : 1: add_comm_op_reuse_tag 0.00% : 0.000061s : 1: add_recomputation 0.00% : 0.000004s : 1: assign_add_opt 0.01% : 0.000197s : 1: auto_monad 0.00% : 0.000025s : 1: auto_monad_reorder 0.00% : 0.000003s : 1: begin_end_overlap_inline 0.00% : 0.000005s : 1: bias_add_comm_swap 0.03% : 0.000499s : 1: bootstrap 0.00% : 0.000041s : 1: cconv 0.00% : 0.000005s : 1: comm_op_add_attrs 0.00% : 0.000019s : 1: control_data_broadcast_order 0.00% : 0.000004s : 1: convert_after_rewriter 0.00% : 0.000034s : 1: cse_after_recomputation 0.00% : 0.000005s : 1: dataset_repeat_opt 0.00% : 0.000006s : 1: detach_backward 0.00% : 0.000015s : 1: environ_conv 0.01% : 0.000220s : 1: event_method 0.00% : 0.000005s : 1: full_micro_interleaved_order_control 0.00% : 0.000006s : 1: get_jit_bprop_graph 0.00% : 0.000017s : 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% : 0.000005s : 1: interleave_split_concat_branches 0.00% : 0.000006s : 1: label_fine_grained_interleaved_index 0.00% : 0.000008s : 1: label_micro_interleaved_index 0.03% : 0.000502s : 1: loop_unroll 0.00% : 0.000004s : 1: merge_cast_opt 0.00% : 0.000005s : 1: micro_interleaved_order_control 0.05% : 0.000839s : 1: mutable_eliminate 0.00% : 0.000008s : 1: offloading_packed_experts 0.00% : 0.000017s : 1: opt.transform.loop_unroll_optimizer 0.00% : 0.000027s : 1: opt.transform.mutable_eliminate 0.72% : 0.012925s : 117: opt.transform.opt_a 0.00% : 0.000032s : 1: opt.transform.opt_after_cconv 0.00% : 0.000029s : 1: opt.transform.opt_after_jit_grad 0.01% : 0.000123s : 28: opt.transform.opt_b 0.01% : 0.000105s : 2: opt.transform.opt_trans_graph 0.00% : 0.000048s : 4: opt.transform.symbol_engine_opt 12.81% : 0.229347s : 1: opt_a 0.01% : 0.000122s : 1: opt_after_cconv 0.03% : 0.000563s : 1: opt_after_jit_grad 0.01% : 0.000261s : 1: opt_b 12.99% : 0.232431s : 1: optimize 0.00% : 0.000024s : 1: optimize_parallel_all_gather_comm 0.00% : 0.000004s : 1: order_py_execute_after_rewriter 0.00% : 0.000028s : 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% : 0.000004s : 1: overlap_opt_shard_grad_in_pipeline 0.00% : 0.000004s : 1: overlap_opt_shard_in_pipeline 0.00% : 0.000005s : 1: overlap_param_gather 0.00% : 0.000004s : 1: overlap_recompute_allgather_and_fa_grad 0.00% : 0.000010s : 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% : 0.000006s : 1: parallel-infer-symbol-second 0.00% : 0.000005s : 1: partial_unused_args_eliminate 0.00% : 0.000004s : 1: pipeline_parallel_scheduler 0.00% : 0.000006s : 1: pipeline_split 0.01% : 0.000146s : 1: pre_auto_parallel 0.00% : 0.000015s : 1: py_interpret_to_execute 0.00% : 0.000014s : 1: py_interpret_to_execute_after_opt_a 0.00% : 0.000004s : 1: remove_cast_before_assign_add 0.00% : 0.000028s : 1: remove_dup_value 9.09% : 0.162628s : 2: renormalize.infer 0.21% : 0.003785s : 2: renormalize.specialize 0.00% : 0.000006s : 1: reorder_send_recv_between_fp_bp 0.00% : 0.000009s : 1: rewriter_after_jit_bprop_graph 0.00% : 0.000033s : 1: rewriter_after_opt_a 0.03% : 0.000580s : 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% : 0.000005s : 1: split_matmul_comm_elemetwise 0.00% : 0.000009s : 1: swap_dp_allreduce_reducescatter 0.01% : 0.000092s : 1: symbol_engine_optimizer 0.01% : 0.000137s : 1: tuple_transform 58.86% : 1.053605s : 1: type_inference . [hook] pytest_runtest_teardown:test_repeat_interleave_forward_backward_tensor[False-None-KBK] tests/st/mint/test_repeat_interleave.py::test_repeat_interleave_forward_backward_tensor[False-None-KBK],max_mem:20.0M =============================== warnings summary =============================== ../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/classifier/transdata/transdata_classifier.py:222 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/classifier/transdata/transdata_classifier.py:222: DeprecationWarning: invalid escape sequence \B """ ../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:143 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:143: DeprecationWarning: invalid escape sequence \c """ ../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:170 /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:170: DeprecationWarning: invalid escape sequence \c """ ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for type is zero. setattr(self, word, getattr(machar, word).flat[0]) ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for type is zero. return self._float_to_str(self.smallest_subnormal) ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for type is zero. setattr(self, word, getattr(machar, word).flat[0]) ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for type is zero. return self._float_to_str(self.smallest_subnormal) ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2.py:57 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2.py:57: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad.py:56 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad.py:56: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2_grad") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad_reduce.py:48 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad_reduce.py:48: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("batchnorm_fold2_grad_reduce") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul.py:51 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul.py:51: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:51 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:51: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul_grad") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:143 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:143: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("correction_mul_grad_reduce") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad.py:92 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad.py:92: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer_grad_d") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad_reduce.py:49 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad_reduce.py:49: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perlayer_grad_d_reduce") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad.py:91 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad.py:91: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel_grad_d") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad_reduce.py:48 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad_reduce.py:48: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_learned_scale_quant_perchannel_grad_d_reduce") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel.py:52 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel.py:52: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_perchannel") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel_grad.py:81 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel_grad.py:81: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_perchannel_grad") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer.py:54 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer.py:54: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_per_layer") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer_grad.py:81 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer_grad.py:81: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("fake_quant_per_layer_grad") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perchannel.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perchannel.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("minmax_update_perchannel") ../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perlayer.py:50 /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perlayer.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute @fusion_manager.register("minmax_update_perlayer") test_repeat_interleave.py::test_repeat_interleave_forward_backward_tensor[True-0-KBK] /usr/local/Ascend/cann-8.5.0/python/site-packages/asc_op_compile_base/asc_op_compiler/ascendc_compile_gen_code.py:161: DeprecationWarning: invalid escape sequence \w match = re.search(f'{option}=(\w+)', ' '.join(compile_options)) -- Docs: https://docs.pytest.org/en/stable/warnings.html ================== 8 passed, 26 warnings in 358.23s (0:05:58) ==================