在使用libtorch_npuc++插件时，运行pytorch/ci/libtorch_resnet.sh 失败
设备型号 910B
CANN 版本 8.2.RC1
torch和torch_npu版本 2.5.1
应该是在执行torch_npu::init_npu("npu:0");时错误
终端报错：
terminate called after throwing an instance of 'c10::Error'
what(): SetPrecisionMode:build/CMakeFiles/torch_npu.dir/compiler_depend.ts:155 NPU function error: at_npu::native::AclSetCompileopt(aclCompileOpt::ACL_PRECISION_MODE, precision_mode), error code is 500001
[ERROR] 2025-11-23-07:23:44 (PID:15264, Device:0, RankID:-1) ERR00100 PTA call acl api failed
[Error]: The internal ACL of the system is incorrect.
Rectify the fault based on the error information in the ascend log.
E40001: [PID: 15264] 2025-11-23-07:23:44.330.046 Value [/usr/local/Ascend/ascend-toolkit/latest/tools/aoe/bin:/usr/local/Ascend/ascend-toolkit/latest/bin:/usr/local/Ascend/ascend-toolkit/latest/compiler/ccec_compiler/bin:/usr/local/Ascend/ascend-toolkit/latest/tools/ccec_compiler/bin:/usr/local/Ascend/ascend-toolkit/8.2.RC1/tools/ncs/bin:/usr/local/Ascend/ascend-toolkit/latest/toolkit/tools/show_kernel_debug_data/:/usr/local/Ascend/ascend-toolkit/latest/toolkit/tools/msobjdump/:/usr/local/Ascend/ascend-toolkit/latest/toolkit/tools/ccec_compiler/bin:/usr/local/Ascend/ascend-toolkit/latest/toolkit/tools/ascend_system_advisor/asys:/usr/local/Ascend/ascend-toolkit/latest/toolkit/tools/profiler/bin/:/usr/local/Ascend/ascend-toolkit/latest/toolkit/python/site-packages/bin/:/usr/local/Ascend/ascend-toolkit/latest/compiler/ccec_compiler/bin:/usr/local/Ascend/ascend-toolkit/latest/compiler/bin:/usr/local/Ascend/ascend-toolkit/latest/bin:/usr/local/Ascend/ascend-toolkit/latest/compiler/ccec_compiler/bin:/usr/local/Ascend/ascend-toolkit/latest/tools/ccec_compiler/bin:/usr/local/Python-3.11/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin] for environment variable [PATH] is invalid when executing the cmd python3-config --prefix.
Possible Cause: The current value of environment variable is not effective.
Solution: Reset the environment variable by referring to the installation guide.
TraceBack (most recent call last):
Fail to initialize handle manager.[FUNC:Initialize][FILE:python_adapter_manager.cc][LINE:40]
Failed to initialize TeConfigInfo.
[GraphOpt][InitializeInner][InitTbeFunc] Failed to init tbe.[FUNC:InitializeTeFusion][FILE:tbe_op_store_adapter.cc][LINE:1921]
[GraphOpt][InitializeInner][InitTeFusion]: Failed to initialize TeFusion.[FUNC:InitializeInner][FILE:tbe_op_store_adapter.cc][LINE:1888]
[SubGraphOpt][PreCompileOp][InitAdapter] InitializeAdapter adapter [tbe_op_adapter] failed! Ret [4294967295][FUNC:InitializeAdapter][FILE:op_store_adapter_manager.cc][LINE:79]
[SubGraphOpt][PreCompileOp][Init] Initialize op store adapter failed, OpsStoreName[tbe-custom].[FUNC:Initialize][FILE:op_store_adapter_manager.cc][LINE:120]
[FusionMngr][Init] Op store adapter manager init failed.[FUNC:Initialize][FILE:fusion_manager.cc][LINE:115]
PluginManager InvokeAll failed.[FUNC:Initialize][FILE:ops_kernel_manager.cc][LINE:83]
OpsManager initialize failed.[FUNC:InnerInitialize][FILE:gelib.cc][LINE:239]
GELib::InnerInitialize failed.[FUNC:Initialize][FILE:gelib.cc][LINE:164]
GEInitialize failed.[FUNC:GEInitialize][FILE:ge_api.cc][LINE:382]
[Initialize][Ge]GEInitialize failed. ge result = 4294967295[FUNC:ReportCallError][FILE:log_inner.cpp][LINE:161]
[Init][Compiler]Init compiler failed[FUNC:ReportInnerError][FILE:log_inner.cpp][LINE:145]
[Set][Options]OpCompileProcessor init failed![FUNC:ReportInnerError][FILE:log_inner.cpp][LINE:145]
Exception raised from SetPrecisionMode at build/CMakeFiles/torch_npu.dir/compiler_depend.ts:155 (most recent call first):
frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0xb8 (0xffff90a0c908 in /usr/local/Python-3.11/lib/python3.11/site-packages/torch/lib/libc10.so)
frame #1: c10::detail::torchCheckFail(char const*, char const*, unsigned int, std::string const&) + 0x6c (0xffff909bb404 in /usr/local/Python-3.11/lib/python3.11/site-packages/torch/lib/libc10.so)
frame #2: <unknown function> + 0x1fd54e4 (0xffff901a54e4 in /workspace/pytorch/libtorch_npu/lib/libtorch_npu.so)
frame #3: <unknown function> + 0x1fd68a8 (0xffff901a68a8 in /workspace/pytorch/libtorch_npu/lib/libtorch_npu.so)
frame #4: <unknown function> + 0x7d19a0 (0xffff8e9a19a0 in /workspace/pytorch/libtorch_npu/lib/libtorch_npu.so)
frame #5: torch_npu::init_npu(signed char) + 0x34 (0xffff9020c524 in /workspace/pytorch/libtorch_npu/lib/libtorch_npu.so)
frame #6: torch_npu::init_npu(std::string const&) + 0x48 (0xffff9020c968 in /workspace/pytorch/libtorch_npu/lib/libtorch_npu.so)
frame #7: ./libtorch_resnet() [0x408748]
frame #8: <unknown function> + 0x2b000 (0xffff834fb000 in /usr/lib64/libc.so.6)
frame #9: __libc_start_main + 0x94 (0xffff834fb0d8 in /usr/lib64/libc.so.6)
frame #10: ./libtorch_resnet() [0x4083b0]
libtorch_resnet.sh: line 25: 15264 Aborted (core dumped) ./libtorch_resnet ../resnet_model.pt > out.log
Failed.
日志plog：
[ERROR] TEFUSION(10652,libtorch_resnet):2025-11-23-06:35:25.347.268 [py_decouple.cc:47]10652 Initialize Launch dynamic-handle failed.
[ERROR] TEFUSION(10652,libtorch_resnet):2025-11-23-06:35:25.347.319 [python_adapter_manager.cc:40]10652 Initialize Fail to initialize handle manager.
[ERROR] FE(10652,libtorch_resnet):2025-11-23-06:35:25.347.347 [tbe_op_store_adapter.cc:1921]10652 InitializeTeFusion:"[GraphOpt][InitializeInner][InitTbeFunc] Failed to init tbe."
[ERROR] FE(10652,libtorch_resnet):2025-11-23-06:35:25.347.371 [tbe_op_store_adapter.cc:1888]10652 InitializeInner:"[GraphOpt][InitializeInner][InitTeFusion]: Failed to initialize TeFusion."
[ERROR] FE(10652,libtorch_resnet):2025-11-23-06:35:25.347.388 [op_store_adapter_manager.cc:79]10652 InitializeAdapter:"[SubGraphOpt][PreCompileOp][InitAdapter] InitializeAdapter adapter [tbe_op_adapter] failed! Ret [4294967295]"
[ERROR] FE(10652,libtorch_resnet):2025-11-23-06:35:25.347.474 [op_store_adapter_manager.cc:120]10652 Initialize:"[SubGraphOpt][PreCompileOp][Init] Initialize op store adapter failed, OpsStoreName[tbe-custom]."
[ERROR] FE(10652,libtorch_resnet):2025-11-23-06:35:25.347.489 [fusion_manager.cc:115]10652 Initialize:"[FusionMngr][Init] Op store adapter manager init failed."
[ERROR] GE(10652,libtorch_resnet):2025-11-23-06:35:25.347.631 [ops_kernel_manager.cc:83]10652 Initialize: ErrorNo: 1343250441(There is no valid so about OpsKernelInfoStore or GraphOptimizer.) [INIT][OPS_KER]PluginManager InvokeAll failed.
[ERROR] GE(10652,libtorch_resnet):2025-11-23-06:35:25.347.650 [gelib.cc:238]10652 InnerInitialize: ErrorNo: 1343250441(There is no valid so about OpsKernelInfoStore or GraphOptimizer.) [INIT][OPS_KER][Init][OpsManager]GE ops manager initial failed.
[ERROR] GE(10652,libtorch_resnet):2025-11-23-06:35:25.362.658 [gelib.cc:163]10652 Initialize: ErrorNo: 1343250441(There is no valid so about OpsKernelInfoStore or GraphOptimizer.) [INIT][OPS_KER][Init][GeLib]GeLib initial failed.
[ERROR] GE(10652,libtorch_resnet):2025-11-23-06:35:25.362.731 [ge_api.cc:299]10652 GEInitializeImpl: ErrorNo: 1343229953(GEInitialize Failed.) [INIT][OPS_KER][Init][GELib]Failed, error code = 1343250441
[ERROR] GE(10652,libtorch_resnet):2025-11-23-06:35:25.362.766 [ge_api.cc:381]10652 GEInitialize: ErrorNo: 4294967295(failed) [INIT][OPS_KER][GEInit] GEInitialize failed, error code:4294967295.
[ERROR] ASCENDCL(10652,libtorch_resnet):2025-11-23-06:35:25.362.790 [local_compiler.cpp:76]10652 Init: [INIT][OPS_KER][Initialize][Ge]GEInitialize failed. ge result = 4294967295
[ERROR] ASCENDCL(10652,libtorch_resnet):2025-11-23-06:35:25.362.812 [op_compile_service.cpp:73]10652 SetCompileStrategy: [INIT][OPS_KER][Init][Compiler]Init compiler failed
[ERROR] ASCENDCL(10652,libtorch_resnet):2025-11-23-06:35:25.362.830 [op_compile_processor.cpp:67]10652 Init: [INIT][OPS_KER][Set][Options]OpCompileProcessor init failed!
[ERROR] ASCENDCL(10652,libtorch_resnet):2025-11-23-06:35:25.362.843 [op_compile_processor.cpp:111]10652 SetCompileOpt: [INIT][OPS_KER][Init][Env]init env failed!
点赞
收藏
回复
分享
举报
已解决
在使用libtorch_npuc++插件时，运行pytorch/ci/libtorch_resnet.sh 失败
设备型号 910B
CANN 版本 8.2.RC1
torch和torch_npu版本 2.5.1
应该是在执行torch_npu::init_npu("npu:0");时错误
终端报错：
terminate called after throwing an instance of 'c10::Error'
what(): SetPrecisionMode:build/CMakeFiles/torch_npu.dir/compiler_depend.ts:155 NPU function error: at_npu::native::AclSetCompileopt(aclCompileOpt::ACL_PRECISION_MODE, precision_mode), error code is 500001
[ERROR] 2025-11-23-07:23:44 (PID:15264, Device:0, RankID:-1) ERR00100 PTA call acl api failed
[Error]: The internal ACL of the system is incorrect.
Rectify the fault based on the error information in the ascend log.
E40001: [PID: 15264] 2025-11-23-07:23:44.330.046 Value [/usr/local/Ascend/ascend-toolkit/latest/tools/aoe/bin:/usr/local/Ascend/ascend-toolkit/latest/bin:/usr/local/Ascend/ascend-toolkit/latest/compiler/ccec_compiler/bin:/usr/local/Ascend/ascend-toolkit/latest/tools/ccec_compiler/bin:/usr/local/Ascend/ascend-toolkit/8.2.RC1/tools/ncs/bin:/usr/local/Ascend/ascend-toolkit/latest/toolkit/tools/show_kernel_debug_data/:/usr/local/Ascend/ascend-toolkit/latest/toolkit/tools/msobjdump/:/usr/local/Ascend/ascend-toolkit/latest/toolkit/tools/ccec_compiler/bin:/usr/local/Ascend/ascend-toolkit/latest/toolkit/tools/ascend_system_advisor/asys:/usr/local/Ascend/ascend-toolkit/latest/toolkit/tools/profiler/bin/:/usr/local/Ascend/ascend-toolkit/latest/toolkit/python/site-packages/bin/:/usr/local/Ascend/ascend-toolkit/latest/compiler/ccec_compiler/bin:/usr/local/Ascend/ascend-toolkit/latest/compiler/bin:/usr/local/Ascend/ascend-toolkit/latest/bin:/usr/local/Ascend/ascend-toolkit/latest/compiler/ccec_compiler/bin:/usr/local/Ascend/ascend-toolkit/latest/tools/ccec_compiler/bin:/usr/local/Python-3.11/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin] for environment variable [PATH] is invalid when executing the cmd python3-config --prefix.
Possible Cause: The current value of environment variable is not effective.
Solution: Reset the environment variable by referring to the installation guide.
TraceBack (most recent call last):
Fail to initialize handle manager.[FUNC:Initialize][FILE:python_adapter_manager.cc][LINE:40]
Failed to initialize TeConfigInfo.
[GraphOpt][InitializeInner][InitTbeFunc] Failed to init tbe.[FUNC:InitializeTeFusion][FILE:tbe_op_store_adapter.cc][LINE:1921]
[GraphOpt][InitializeInner][InitTeFusion]: Failed to initialize TeFusion.[FUNC:InitializeInner][FILE:tbe_op_store_adapter.cc][LINE:1888]
[SubGraphOpt][PreCompileOp][InitAdapter] InitializeAdapter adapter [tbe_op_adapter] failed! Ret [4294967295][FUNC:InitializeAdapter][FILE:op_store_adapter_manager.cc][LINE:79]
[SubGraphOpt][PreCompileOp][Init] Initialize op store adapter failed, OpsStoreName[tbe-custom].[FUNC:Initialize][FILE:op_store_adapter_manager.cc][LINE:120]
[FusionMngr][Init] Op store adapter manager init failed.[FUNC:Initialize][FILE:fusion_manager.cc][LINE:115]
PluginManager InvokeAll failed.[FUNC:Initialize][FILE:ops_kernel_manager.cc][LINE:83]
OpsManager initialize failed.[FUNC:InnerInitialize][FILE:gelib.cc][LINE:239]
GELib::InnerInitialize failed.[FUNC:Initialize][FILE:gelib.cc][LINE:164]
GEInitialize failed.[FUNC:GEInitialize][FILE:ge_api.cc][LINE:382]
[Initialize][Ge]GEInitialize failed. ge result = 4294967295[FUNC:ReportCallError][FILE:log_inner.cpp][LINE:161]
[Init][Compiler]Init compiler failed[FUNC:ReportInnerError][FILE:log_inner.cpp][LINE:145]
[Set][Options]OpCompileProcessor init failed![FUNC:ReportInnerError][FILE:log_inner.cpp][LINE:145]
Exception raised from SetPrecisionMode at build/CMakeFiles/torch_npu.dir/compiler_depend.ts:155 (most recent call first):
frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0xb8 (0xffff90a0c908 in /usr/local/Python-3.11/lib/python3.11/site-packages/torch/lib/libc10.so)
frame #1: c10::detail::torchCheckFail(char const*, char const*, unsigned int, std::string const&) + 0x6c (0xffff909bb404 in /usr/local/Python-3.11/lib/python3.11/site-packages/torch/lib/libc10.so)
frame #2: <unknown function> + 0x1fd54e4 (0xffff901a54e4 in /workspace/pytorch/libtorch_npu/lib/libtorch_npu.so)
frame #3: <unknown function> + 0x1fd68a8 (0xffff901a68a8 in /workspace/pytorch/libtorch_npu/lib/libtorch_npu.so)
frame #4: <unknown function> + 0x7d19a0 (0xffff8e9a19a0 in /workspace/pytorch/libtorch_npu/lib/libtorch_npu.so)
frame #5: torch_npu::init_npu(signed char) + 0x34 (0xffff9020c524 in /workspace/pytorch/libtorch_npu/lib/libtorch_npu.so)
frame #6: torch_npu::init_npu(std::string const&) + 0x48 (0xffff9020c968 in /workspace/pytorch/libtorch_npu/lib/libtorch_npu.so)
frame #7: ./libtorch_resnet() [0x408748]
frame #8: <unknown function> + 0x2b000 (0xffff834fb000 in /usr/lib64/libc.so.6)
frame #9: __libc_start_main + 0x94 (0xffff834fb0d8 in /usr/lib64/libc.so.6)
frame #10: ./libtorch_resnet() [0x4083b0]
libtorch_resnet.sh: line 25: 15264 Aborted (core dumped) ./libtorch_resnet ../resnet_model.pt > out.log
Failed.
日志plog：
[ERROR] TEFUSION(10652,libtorch_resnet):2025-11-23-06:35:25.347.268 [py_decouple.cc:47]10652 Initialize Launch dynamic-handle failed.
[ERROR] TEFUSION(10652,libtorch_resnet):2025-11-23-06:35:25.347.319 [python_adapter_manager.cc:40]10652 Initialize Fail to initialize handle manager.
[ERROR] FE(10652,libtorch_resnet):2025-11-23-06:35:25.347.347 [tbe_op_store_adapter.cc:1921]10652 InitializeTeFusion:"[GraphOpt][InitializeInner][InitTbeFunc] Failed to init tbe."
[ERROR] FE(10652,libtorch_resnet):2025-11-23-06:35:25.347.371 [tbe_op_store_adapter.cc:1888]10652 InitializeInner:"[GraphOpt][InitializeInner][InitTeFusion]: Failed to initialize TeFusion."
[ERROR] FE(10652,libtorch_resnet):2025-11-23-06:35:25.347.388 [op_store_adapter_manager.cc:79]10652 InitializeAdapter:"[SubGraphOpt][PreCompileOp][InitAdapter] InitializeAdapter adapter [tbe_op_adapter] failed! Ret [4294967295]"
[ERROR] FE(10652,libtorch_resnet):2025-11-23-06:35:25.347.474 [op_store_adapter_manager.cc:120]10652 Initialize:"[SubGraphOpt][PreCompileOp][Init] Initialize op store adapter failed, OpsStoreName[tbe-custom]."
[ERROR] FE(10652,libtorch_resnet):2025-11-23-06:35:25.347.489 [fusion_manager.cc:115]10652 Initialize:"[FusionMngr][Init] Op store adapter manager init failed."
[ERROR] GE(10652,libtorch_resnet):2025-11-23-06:35:25.347.631 [ops_kernel_manager.cc:83]10652 Initialize: ErrorNo: 1343250441(There is no valid so about OpsKernelInfoStore or GraphOptimizer.) [INIT][OPS_KER]PluginManager InvokeAll failed.
[ERROR] GE(10652,libtorch_resnet):2025-11-23-06:35:25.347.650 [gelib.cc:238]10652 InnerInitialize: ErrorNo: 1343250441(There is no valid so about OpsKernelInfoStore or GraphOptimizer.) [INIT][OPS_KER][Init][OpsManager]GE ops manager initial failed.
[ERROR] GE(10652,libtorch_resnet):2025-11-23-06:35:25.362.658 [gelib.cc:163]10652 Initialize: ErrorNo: 1343250441(There is no valid so about OpsKernelInfoStore or GraphOptimizer.) [INIT][OPS_KER][Init][GeLib]GeLib initial failed.
[ERROR] GE(10652,libtorch_resnet):2025-11-23-06:35:25.362.731 [ge_api.cc:299]10652 GEInitializeImpl: ErrorNo: 1343229953(GEInitialize Failed.) [INIT][OPS_KER][Init][GELib]Failed, error code = 1343250441
[ERROR] GE(10652,libtorch_resnet):2025-11-23-06:35:25.362.766 [ge_api.cc:381]10652 GEInitialize: ErrorNo: 4294967295(failed) [INIT][OPS_KER][GEInit] GEInitialize failed, error code:4294967295.
[ERROR] ASCENDCL(10652,libtorch_resnet):2025-11-23-06:35:25.362.790 [local_compiler.cpp:76]10652 Init: [INIT][OPS_KER][Initialize][Ge]GEInitialize failed. ge result = 4294967295
[ERROR] ASCENDCL(10652,libtorch_resnet):2025-11-23-06:35:25.362.812 [op_compile_service.cpp:73]10652 SetCompileStrategy: [INIT][OPS_KER][Init][Compiler]Init compiler failed
[ERROR] ASCENDCL(10652,libtorch_resnet):2025-11-23-06:35:25.362.830 [op_compile_processor.cpp:67]10652 Init: [INIT][OPS_KER][Set][Options]OpCompileProcessor init failed!
[ERROR] ASCENDCL(10652,libtorch_resnet):2025-11-23-06:35:25.362.843 [op_compile_processor.cpp:111]10652 SetCompileOpt: [INIT][OPS_KER][Init][Env]init env failed!