Paddle gpu 1.7.0 安装提示Error: Cannot load cudnn shared library

Created by: zengqi0730

1）PaddlePaddle版本：paddlepaddle-gpu 1.7.0.post107 2）CPU：Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz 3）GPU：1080Ti 4）系统环境：Ubuntu Kylin 18.04.3 LTS、Python3.6.10

安装方式信息： 1）pip安装：pip install paddlepaddle-gpu==1.7.0.post107 -i https://mirror.baidu.com/pypi/simple
复现信息：
conda 已有环境 base * /home/edu/anaconda3 paddle /home/edu/anaconda3/envs/paddle py36 /home/edu/anaconda3/envs/py36 # 其中安装了pytorch1.4.0 tf1.12 /home/edu/anaconda3/envs/tf1.12 edu@edu-Alienware-Aurora-R7:~/Desktop$ source activate paddle (paddle) edu@edu-Alienware-Aurora-R7:~/Desktop$ conda list

packages in environment at /home/edu/anaconda3/envs/paddle:

Name Version Build Channel

import paddle.fluid paddle.fluid.install_check.run_check() Running Verify Paddle Program ... W0313 11:13:35.911985 11802 device_context.cc:237] Please NOTE: device: 0, CUDA Capability: 61, Driver API Version: 10.1, Runtime API Version: 10.0 W0313 11:13:35.912094 11802 dynamic_loader.cc:120] Can not find library: libcudnn.so. The process maybe hang. Please try to add the lib path to LD_LIBRARY_PATH. W0313 11:13:35.912115 11802 dynamic_loader.cc:179] Failed to find dynamic library: libcudnn.so ( libcudnn.so: cannot open shared object file: No such file or directory ) Please specify its path correctly using following ways: Method. set environment variable LD_LIBRARY_PATH on Linux or DYLD_LIBRARY_PATH on Mac OS. For instance, issue command: export LD_LIBRARY_PATH=... Note: After Mac OS 10.11, using the DYLD_LIBRARY_PATH is impossible unless System Integrity Protection (SIP) is disabled. /home/edu/anaconda3/envs/paddle/lib/python3.6/site-packages/paddle/fluid/executor.py:782: UserWarning: The following exception is not an EOF exception. "The following exception is not an EOF exception.") Traceback (most recent call last): File "", line 1, in File "/home/edu/anaconda3/envs/paddle/lib/python3.6/site-packages/paddle/fluid/install_check.py", line 124, in run_check test_simple_exe() File "/home/edu/anaconda3/envs/paddle/lib/python3.6/site-packages/paddle/fluid/install_check.py", line 120, in test_simple_exe exe0.run(startup_prog) File "/home/edu/anaconda3/envs/paddle/lib/python3.6/site-packages/paddle/fluid/executor.py", line 783, in run six.reraise(*sys.exc_info()) File "/home/edu/anaconda3/envs/paddle/lib/python3.6/site-packages/six.py", line 703, in reraise raise value File "/home/edu/anaconda3/envs/paddle/lib/python3.6/site-packages/paddle/fluid/executor.py", line 778, in run use_program_cache=use_program_cache) File "/home/edu/anaconda3/envs/paddle/lib/python3.6/site-packages/paddle/fluid/executor.py", line 831, in _run_impl use_program_cache=use_program_cache) File "/home/edu/anaconda3/envs/paddle/lib/python3.6/site-packages/paddle/fluid/executor.py", line 905, in _run_program fetch_var_name) paddle.fluid.core_avx.EnforceNotMet:

C++ Call Stacks (More useful to developers):

0 std::string paddle::platform::GetTraceBackString<char const*>(char const*&&, char const*, int) 1 paddle::platform::EnforceNotMet::EnforceNotMet(std::__exception_ptr::exception_ptr, char const*, int) 2 paddle::platform::dynload::EnforceCUDNNLoaded(char const*) 3 paddle::platform::CUDADeviceContext::CUDADeviceContext(paddle::platform::CUDAPlace) 4 std::_Function_handler<std::unique_ptr<paddle::platform::DeviceContext, std::default_deletepaddle::platform::DeviceContext > (), std::reference_wrapper<std::_Bind_simple<paddle::platform::EmplaceDeviceContext<paddle::platform::CUDADeviceContext, paddle::platform::CUDAPlace>(std::map<paddle::platform::Place, std::shared_future<std::unique_ptr<paddle::platform::DeviceContext, std::default_deletepaddle::platform::DeviceContext > >, std::lesspaddle::platform::Place, std::allocator<std::pair<paddle::platform::Place const, std::shared_future<std::unique_ptr<paddle::platform::DeviceContext, std::default_deletepaddle::platform::DeviceContext > > > > >, paddle::platform::Place)::{lambda()#1 (closed)} ()> > >::_M_invoke(std::_Any_data const&) 5 std::_Function_handler<std::unique_ptr<std::__future_base::_Result_base, std::__future_base::_Result_base::_Deleter> (), std::__future_base::_Task_setter<std::unique_ptr<std::__future_base::_Result<std::unique_ptr<paddle::platform::DeviceContext, std::default_deletepaddle::platform::DeviceContext > >, std::__future_base::_Result_base::_Deleter>, std::unique_ptr<paddle::platform::DeviceContext, std::default_deletepaddle::platform::DeviceContext > > >::_M_invoke(std::_Any_data const&) 6 std::__future_base::_State_base::_M_do_set(std::function<std::unique_ptr<std::__future_base::_Result_base, std::__future_base::_Result_base::_Deleter> ()>&, bool&) 7 std::__future_base::_Deferred_state<std::_Bind_simple<paddle::platform::EmplaceDeviceContext<paddle::platform::CUDADeviceContext, paddle::platform::CUDAPlace>(std::map<paddle::platform::Place, std::shared_future<std::unique_ptr<paddle::platform::DeviceContext, std::default_deletepaddle::platform::DeviceContext > >, std::lesspaddle::platform::Place, std::allocator<std::pair<paddle::platform::Place const, std::shared_future<std::unique_ptr<paddle::platform::DeviceContext, std::default_deletepaddle::platform::DeviceContext > > > > >, paddle::platform::Place)::{lambda()#1 (closed)} ()>, std::unique_ptr<paddle::platform::DeviceContext, std::default_deletepaddle::platform::DeviceContext > >::_M_run_deferred() 8 paddle::platform::DeviceContextPool::Get(paddle::platform::Place const&) 9 paddle::framework::GarbageCollector::GarbageCollector(paddle::platform::Place const&, unsigned long) 10 paddle::framework::UnsafeFastGPUGarbageCollector::UnsafeFastGPUGarbageCollector(paddle::platform::CUDAPlace const&, unsigned long) 11 paddle::framework::Executor::RunPreparedContext(paddle::framework::ExecutorPrepareContext*, paddle::framework::Scope*, bool, bool, bool) 12 paddle::framework::Executor::Run(paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool, std::vector<std::string, std::allocatorstd::string > const&, bool, bool)

Error Message Summary:

Error: Cannot load cudnn shared library. Cannot invoke method cudnnGetVersion at (/paddle/paddle/fluid/platform/dynload/cudnn.cc:63)

import tensorflow as tf print(tf.version) 1.14.0 tf.test.gpu_device_name() 2020-03-13 11:17:03.429640: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 FMA 2020-03-13 11:17:03.460899: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 3699850000 Hz 2020-03-13 11:17:03.461319: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x562f07660ec0 executing computations on platform Host. Devices: 2020-03-13 11:17:03.461343: I tensorflow/compiler/xla/service/service.cc:175] StreamExecutor device (0): , 2020-03-13 11:17:03.461449: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcuda.so.1 2020-03-13 11:17:03.461552: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero 2020-03-13 11:17:03.461941: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1640] Found device 0 with properties: name: GeForce GTX 1080 Ti major: 6 minor: 1 memoryClockRate(GHz): 1.582 pciBusID: 0000:01:00.0 2020-03-13 11:17:03.480250: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcudart.so.10.0 2020-03-13 11:17:03.654387: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcublas.so.10.0 2020-03-13 11:17:03.749657: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcufft.so.10.0 2020-03-13 11:17:03.781214: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcurand.so.10.0 2020-03-13 11:17:03.948525: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcusolver.so.10.0 2020-03-13 11:17:04.061082: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcusparse.so.10.0 2020-03-13 11:17:04.071400: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcudnn.so.7 2020-03-13 11:17:04.071702: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero 2020-03-13 11:17:04.073268: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero 2020-03-13 11:17:04.074585: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1763] Adding visible gpu devices: 0 2020-03-13 11:17:04.074718: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcudart.so.10.0 2020-03-13 11:17:04.097656: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1181] Device interconnect StreamExecutor with strength 1 edge matrix: 2020-03-13 11:17:04.097738: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1187] 0 2020-03-13 11:17:04.097764: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1200] 0: N 2020-03-13 11:17:04.098118: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero 2020-03-13 11:17:04.099641: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero 2020-03-13 11:17:04.101200: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero 2020-03-13 11:17:04.102656: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1326] Created TensorFlow device (/device:GPU:0 with 10087 MB memory) -> physical GPU (device: 0, name: GeForce GTX 1080 Ti, pci bus id: 0000:01:00.0, compute capability: 6.1) 2020-03-13 11:17:04.107436: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x562f07661600 executing computations on platform CUDA. Devices: 2020-03-13 11:17:04.107505: I tensorflow/compiler/xla/service/service.cc:175] StreamExecutor device (0): GeForce GTX 1080 Ti, Compute Capability 6.1 '/device:GPU:0'

(paddle) edu@edu-Alienware-Aurora-R7:~/Downloads$ cat /usr/local/cuda/version.txt CUDA Version 10.1.243 (paddle) edu@edu-Alienware-Aurora-R7:~/Downloads$ cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2 #define CUDNN_MAJOR 7 #define CUDNN_MINOR 6 #define CUDNN_PATCHLEVEL 5

#define CUDNN_VERSION (CUDNN_MAJOR * 1000 + CUDNN_MINOR * 100 + CUDNN_PATCHLEVEL)

#include "driver_types.h"

PaddlePaddle / Paddle 大约 2 年 前同步成功

Paddle gpu 1.7.0 安装 提示Error: Cannot load cudnn shared library