Atlas300I duo卡启动服务报错EZ9999: [PID: 28542] 2024-12-21-15:15:33.083.805 Parse dynamic kernel config fail
收藏回复举报
Atlas300I duo卡启动服务报错EZ9999: [PID: 28542] 2024-12-21-15:15:33.083.805 Parse dynamic kernel config fail
t('forum.solved') 已解决
新人帖
发表于2024-12-21 15:27:09
0 查看
 

2024-12-21 15:15:28,558 [INFO] [pid: 28544] flash_causal_qwen2.py-103: >>>> qwen_DecoderModel is called. 

2024-12-21 15:15:32,938 [INFO] [pid: 28542] logging.py-180: model: 

 FlashQwen2ForCausalLM( 

  (rotary_embedding): PositionRotaryEmbedding() 

  (attn_mask): AttentionMask() 

  (transformer): FlashQwenModel( 

    (wte): TensorParallelEmbedding() 

    (h): ModuleList( 

      (0-27): 28 x FlashQwenLayer( 

        (attn): FlashQwenAttention( 

          (rotary_emb): PositionRotaryEmbedding() 

          (c_attn): TensorParallelColumnLinear( 

            (linear): FastLinear() 

          ) 

          (c_proj): TensorParallelRowLinear( 

            (linear): FastLinear() 

          ) 

        ) 

        (mlp): QwenMLP( 

          (act): SiLU() 

          (w2_w1): TensorParallelColumnLinear( 

            (linear): FastLinear() 

          ) 

          (c_proj): TensorParallelRowLinear( 

            (linear): FastLinear() 

          ) 

        ) 

        (ln_1): QwenRMSNorm() 

        (ln_2): QwenRMSNorm() 

      ) 

    ) 

    (ln_f): QwenRMSNorm() 

  ) 

  (lm_head): TensorParallelHead( 

    (linear): FastLinear() 

  ) 

[ERROR] OP(28542,mindie_llm_backend_connector):2024-12-21-15:15:33.083.735 [kernel_mgr.cpp:227][NNOP][ParseDynamicKernelConfig][29254] errno[561108] OpName:[aclnnInplaceZero_3_ZerosLikeAiCore] Parse dynamic kernel config fail. 

[ERROR] OP(28542,mindie_llm_backend_connector):2024-12-21-15:15:33.083.940 [kernel_workspace.cpp:90][NNOP][GetWorkspace][29254] errno[561000] OpName:[aclnnInplaceZero_3_ZerosLikeAiCore] AclOpKernelInit failed opType 

[ERROR] OP(28542,mindie_llm_backend_connector):2024-12-21-15:15:33.083.994 [zero_op.cpp:62][NNOP][ZerosLikeAiCore][29254] errno[561118] OpName:[aclnnInplaceZero_3_ZerosLikeAiCore] ZerosLike ADD_TO_LAUNCHER_LIST_AICORE failed. 

[ERROR] OP(28542,mindie_llm_backend_connector):2024-12-21-15:15:33.084.022 [aclnn_zero.cpp:93][NNOP][aclnnInplaceZeroGetWorkspaceSize][29254] errno[561000] OpName:[aclnnInplaceZero_3] check zeroOut != nullptr failed. 

[ERROR] OP(28542,mindie_llm_backend_connector):2024-12-21-15:15:33.084.035 [aclnn_zero.cpp:164][NNOP][~UniqueExecutor][29254] errno[561102] OpName:[aclnnInplaceZero_3] When aclnnInplaceZeroGetWorkspaceSize do success, ReleaseTo(executor) should be called before return. 

2024-12-21 15:15:33,089 [ERROR] model.py:39 - [Model]   >>> Exception:call aclnnInplaceZero failed, detail:EZ9999: Inner Error! 

EZ9999: [PID: 28542] 2024-12-21-15:15:33.083.805 Parse dynamic kernel config fail. 

        TraceBack (most recent call last): 

       AclOpKernelInit failed opType 

       ZerosLike ADD_TO_LAUNCHER_LIST_AICORE failed. 

 

[ERROR] 2024-12-21-15:15:33 (PID:28542, Device:0, RankID:-1) ERR01100 OPS call acl api failed 

Traceback (most recent call last): 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/model_wrapper/model.py", line 37, in initialize 

    return self.python_model.initialize(config) 

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/model_wrapper/standard_model.py", line 144, in initialize 

    self.generator = Generator( 

                     ^^^^^^^^^^ 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/mindie_llm/text_generator/generator.py", line 101, in __init__ 

    self.warm_up(max_prefill_tokens, max_seq_len, max_input_len, max_iter_times, inference_mode) 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/mindie_llm/text_generator/generator.py", line 245, in warm_up 

    self.generator_backend.warm_up(model_inputs) 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/mindie_llm/text_generator/adapter/generator_backend.py", line 128, in warm_up 

    _ = self.forward(model_inputs, **kwargs) 

        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/mindie_llm/utils/decorators/time_decorator.py", line 38, in wrapper 

    return func(*args, **kwargs) 

           ^^^^^^^^^^^^^^^^^^^^^ 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/mindie_llm/text_generator/adapter/generator_torch.py", line 92, in forward 

    logits = self.model_wrapper.forward(model_inputs, self.cache_pool.npu_cache, **kwargs) 

             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/mindie_llm/modeling/model_wrapper/atb/atb_model_wrapper.py", line 65, in forward 

    logits = self.forward_tensor( 

             ^^^^^^^^^^^^^^^^^^^^ 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/mindie_llm/modeling/model_wrapper/atb/atb_model_wrapper.py", line 92, in forward_tensor 

    logits = self.model_runner.forward( 

             ^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/data/mindie/atb-models/atb_llm/runner/model_runner.py", line 157, in forward 

    return self.model.forward(**kwargs) 

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/data/mindie/atb-models/atb_llm/models/base/flash_causal_lm.py", line 380, in forward 

    self.init_ascend_weight() 

  File "/data/mindie/atb-models/atb_llm/models/qwen2/flash_causal_qwen2.py", line 136, in init_ascend_weight 

    weight_wrapper = self.get_weights() 

                     ^^^^^^^^^^^^^^^^^^ 

  File "/data/mindie/atb-models/atb_llm/models/qwen2/flash_causal_qwen2.py", line 120, in get_weights 

    weight_wrapper = WeightWrapper(self.soc_info, self.tp_rank, attn_wrapper, mlp_wrapper) 

                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/data/mindie/atb-models/atb_llm/utils/data/weight_wrapper.py", line 49, in __init__ 

    self.placeholder = torch.zeros(1, dtype=torch.float16, device="npu") 

                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

RuntimeError: call aclnnInplaceZero failed, detail:EZ9999: Inner Error! 

EZ9999: [PID: 28542] 2024-12-21-15:15:33.083.805 Parse dynamic kernel config fail. 

        TraceBack (most recent call last): 

       AclOpKernelInit failed opType 

       ZerosLike ADD_TO_LAUNCHER_LIST_AICORE failed. 

 

[ERROR] 2024-12-21-15:15:33 (PID:28542, Device:0, RankID:-1) ERR01100 OPS call acl api failed 

2024-12-21 15:15:33,093 [ERROR] model.py:42 - [Model]   >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0'} 

[ERROR] OP(28544,mindie_llm_backend_connector):2024-12-21-15:15:33.882.025 [kernel_mgr.cpp:227][NNOP][ParseDynamicKernelConfig][29256] errno[561108] OpName:[aclnnInplaceZero_3_ZerosLikeAiCore] Parse dynamic kernel config fail. 

[ERROR] OP(28544,mindie_llm_backend_connector):2024-12-21-15:15:33.882.177 [kernel_workspace.cpp:90][NNOP][GetWorkspace][29256] errno[561000] OpName:[aclnnInplaceZero_3_ZerosLikeAiCore] AclOpKernelInit failed opType 

[ERROR] OP(28544,mindie_llm_backend_connector):2024-12-21-15:15:33.882.225 [zero_op.cpp:62][NNOP][ZerosLikeAiCore][29256] errno[561118] OpName:[aclnnInplaceZero_3_ZerosLikeAiCore] ZerosLike ADD_TO_LAUNCHER_LIST_AICORE failed. 

[ERROR] OP(28544,mindie_llm_backend_connector):2024-12-21-15:15:33.882.254 [aclnn_zero.cpp:93][NNOP][aclnnInplaceZeroGetWorkspaceSize][29256] errno[561000] OpName:[aclnnInplaceZero_3] check zeroOut != nullptr failed. 

[ERROR] OP(28544,mindie_llm_backend_connector):2024-12-21-15:15:33.882.266 [aclnn_zero.cpp:164][NNOP][~UniqueExecutor][29256] errno[561102] OpName:[aclnnInplaceZero_3] When aclnnInplaceZeroGetWorkspaceSize do success, ReleaseTo(executor) should be called before return. 

2024-12-21 15:15:33,886 [ERROR] model.py:39 - [Model]   >>> Exception:call aclnnInplaceZero failed, detail:EZ9999: Inner Error! 

EZ9999: [PID: 28544] 2024-12-21-15:15:33.882.083 Parse dynamic kernel config fail. 

        TraceBack (most recent call last): 

       AclOpKernelInit failed opType 

       ZerosLike ADD_TO_LAUNCHER_LIST_AICORE failed. 

 

[ERROR] 2024-12-21-15:15:33 (PID:28544, Device:1, RankID:-1) ERR01100 OPS call acl api failed 

Traceback (most recent call last): 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/model_wrapper/model.py", line 37, in initialize 

    return self.python_model.initialize(config) 

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/model_wrapper/standard_model.py", line 144, in initialize 

    self.generator = Generator( 

                     ^^^^^^^^^^ 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/mindie_llm/text_generator/generator.py", line 101, in __init__ 

    self.warm_up(max_prefill_tokens, max_seq_len, max_input_len, max_iter_times, inference_mode) 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/mindie_llm/text_generator/generator.py", line 245, in warm_up 

    self.generator_backend.warm_up(model_inputs) 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/mindie_llm/text_generator/adapter/generator_backend.py", line 128, in warm_up 

    _ = self.forward(model_inputs, **kwargs) 

        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/mindie_llm/utils/decorators/time_decorator.py", line 38, in wrapper 

    return func(*args, **kwargs) 

           ^^^^^^^^^^^^^^^^^^^^^ 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/mindie_llm/text_generator/adapter/generator_torch.py", line 92, in forward 

    logits = self.model_wrapper.forward(model_inputs, self.cache_pool.npu_cache, **kwargs) 

             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/mindie_llm/modeling/model_wrapper/atb/atb_model_wrapper.py", line 65, in forward 

    logits = self.forward_tensor( 

             ^^^^^^^^^^^^^^^^^^^^ 

  File "/data/anaconda3/envs/mindie/lib/python3.11/site-packages/mindie_llm/modeling/model_wrapper/atb/atb_model_wrapper.py", line 92, in forward_tensor 

    logits = self.model_runner.forward( 

             ^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/data/mindie/atb-models/atb_llm/runner/model_runner.py", line 157, in forward 

    return self.model.forward(**kwargs) 

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/data/mindie/atb-models/atb_llm/models/base/flash_causal_lm.py", line 380, in forward 

    self.init_ascend_weight() 

  File "/data/mindie/atb-models/atb_llm/models/qwen2/flash_causal_qwen2.py", line 136, in init_ascend_weight 

    weight_wrapper = self.get_weights() 

                     ^^^^^^^^^^^^^^^^^^ 

  File "/data/mindie/atb-models/atb_llm/models/qwen2/flash_causal_qwen2.py", line 120, in get_weights 

    weight_wrapper = WeightWrapper(self.soc_info, self.tp_rank, attn_wrapper, mlp_wrapper) 

                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

  File "/data/mindie/atb-models/atb_llm/utils/data/weight_wrapper.py", line 49, in __init__ 

    self.placeholder = torch.zeros(1, dtype=torch.float16, device="npu") 

                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

RuntimeError: call aclnnInplaceZero failed, detail:EZ9999: Inner Error! 

EZ9999: [PID: 28544] 2024-12-21-15:15:33.882.083 Parse dynamic kernel config fail. 

        TraceBack (most recent call last): 

       AclOpKernelInit failed opType 

       ZerosLike ADD_TO_LAUNCHER_LIST_AICORE failed. 

 

[ERROR] 2024-12-21-15:15:33 (PID:28544, Device:1, RankID:-1) ERR01100 OPS call acl api failed 

2024-12-21 15:15:33,889 [ERROR] model.py:42 - [Model]   >>> return initialize error result: {'status': 'error', 'npuBlockNum': '0', 'cpuBlockNum': '0'} 

2024-12-21 15:15:33.940539 28270 error [llm_infer_model_instance.cpp:45] Initialize modelBackends_ failed. 

2024-12-21 15:15:33.940666 28270 error [llm_infer_model.cpp:20] Init model instance failed. 

LLMInferEngine failed to init LLMInferModels 

ERR: Failed to init endpoint! Please check the service log or console output. 

已杀死 cke_2190.png

我要发帖子