Created by: jiaozhusos

如下： cuda 版本为8.0，cudnn版本为5.1.10，报了以下的错

paddle.init(use_gpu=True, trainer_count=1) I0605 01:44:45.392531 26917 Util.cpp:166] commandline: --use_gpu=True --trainer_count=1 F0605 01:44:46.962916 26917 hl_cuda_cudnn.cc:171] Check failed: (cudnn_cuh_major < 4 && cudnn_dso_major < 4) || (cudnn_cuh_major == cudnn_dso_major) [cudnn init] libcudnn v5 with header v7 unmatched! PaddlePaddle Requirement: (header v[2-3] with libcudnn v[2-3]) Or (header v4 with libcudnn v4) Or (header v5 with libcudnn v5) Or(header v6 with libcudnn v6).

* Check failure stack trace: *

@ 0x7f4f6549ef8d google::LogMessage::Fail() @ 0x7f4f654a2a3c google::LogMessage::SendToLog() @ 0x7f4f6549eab3 google::LogMessage::Flush() @ 0x7f4f654a3f4e google::LogMessageFatal::~LogMessageFatal() @ 0x7f4f65454eaf hl_cudnn_init() @ 0x7f4f6545ced5 hl_create_global_resources() @ 0x7f4f6545d681 hl_specify_devices_start() @ 0x7f4f6545da3d hl_start() @ 0x7f4f653e7406 paddle::initMain() @ 0x7f4f65486ac1 initPaddle() @ 0x7f4f650331f9 _wrap_initPaddle @ 0x7f4fa4152210 PyEval_EvalFrameEx @ 0x7f4fa41554e9 PyEval_EvalCodeEx @ 0x7f4fa41529b8 PyEval_EvalFrameEx @ 0x7f4fa41554e9 PyEval_EvalCodeEx @ 0x7f4fa40de28a function_call @ 0x7f4fa40b97a3 PyObject_Call @ 0x7f4fa414e4be PyEval_EvalFrameEx @ 0x7f4fa41554e9 PyEval_EvalCodeEx @ 0x7f4fa41529b8 PyEval_EvalFrameEx @ 0x7f4fa41554e9 PyEval_EvalCodeEx @ 0x7f4fa415570a PyEval_EvalCode @ 0x7f4fa416e9cd run_mod @ 0x7f4fa4170a0e PyRun_InteractiveOneFlags @ 0x7f4fa4170baf PyRun_InteractiveLoopFlags @ 0x7f4fa41711f0 PyRun_AnyFileExFlags @ 0x7f4fa4182f8c Py_Main @ 0x7f4fa339a830 __libc_start_main @ 0x55af81fa08bf (unknown) Aborted (core dumped)