Ubutun 安装gpu版本无异常，但是paddle.fluid.install_check.run_check()报错

版本、环境信息： 1）PaddlePaddle版本：PaddlePaddle版本1.4.1 2）GPU：CUDA版本=9，CUDNN版本=7。GPU设备为2080Ti 3）系统环境：系统=16.04.1-Ubuntu、Python版本=Python3.5.5
安装方式信息： 1）pip安装：pip3 install paddlepaddle-gpu
问题描述：安装过程无错误警告，安装后输入import paddle.fluid ，再输入 paddle.fluid.install_check.run_check()报错如下：
Running Verify Fluid Program ... 
W0520 16:52:06.161931 27216 device_context.cc:261] Please NOTE: device: 0, CUDA Capability: 75, Driver API Version: 10.0, Runtime API Version: 9.0
W0520 16:52:06.187935 27216 device_context.cc:269] device: 0, cuDNN Version: 7.0.
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/lab/liujiepeng/MachineComprehension/DuReader/paddle_py35_even/lib/python3.5/site-packages/paddle/fluid/install_check.py", line 65, in run_check
fetch_list=[out.name, param_grads[1].name])
File "/home/lab/liujiepeng/MachineComprehension/DuReader/paddle_py35_even/lib/python3.5/site-packages/paddle/fluid/executor.py", line 565, in run
use_program_cache=use_program_cache)
File "/home/lab/liujiepeng/MachineComprehension/DuReader/paddle_py35_even/lib/python3.5/site-packages/paddle/fluid/executor.py", line 642, in _run
exe.run(program.desc, scope, 0, True, True, fetch_var_name)
paddle.fluid.core.EnforceNotMet: Invoke operator mul error.
Python Callstacks: 
File "/home/lab/liujiepeng/MachineComprehension/DuReader/paddle_py35_even/lib/python3.5/site-packages/paddle/fluid/framework.py", line 1654, in append_op
attrs=kwargs.get("attrs", None))
File "/home/lab/liujiepeng/MachineComprehension/DuReader/paddle_py35_even/lib/python3.5/site-packages/paddle/fluid/dygraph/layer_object_helper.py", line 52, in append_op
stop_gradient=stop_gradient)
File "/home/lab/liujiepeng/MachineComprehension/DuReader/paddle_py35_even/lib/python3.5/site-packages/paddle/fluid/dygraph/nn.py", line 919, in forward
"y_num_col_dims": 1
File "/home/lab/liujiepeng/MachineComprehension/DuReader/paddle_py35_even/lib/python3.5/site-packages/paddle/fluid/dygraph/layers.py", line 158, in __call__
outputs = self.forward(*inputs)
File "/home/lab/liujiepeng/MachineComprehension/DuReader/paddle_py35_even/lib/python3.5/site-packages/paddle/fluid/install_check.py", line 37, in forward
x = self._fc1(inputs)
File "/home/lab/liujiepeng/MachineComprehension/DuReader/paddle_py35_even/lib/python3.5/site-packages/paddle/fluid/dygraph/layers.py", line 158, in __call__
outputs = self.forward(*inputs)
File "/home/lab/liujiepeng/MachineComprehension/DuReader/paddle_py35_even/lib/python3.5/site-packages/paddle/fluid/install_check.py", line 58, in run_check
out = simple_layer(inp)
File "<stdin>", line 1, in <module>
C++ Callstacks: 
CUBLAS: execution failed,  at [/paddle/paddle/fluid/operators/math/blas_impl.cu.h:34]
PaddlePaddle Call Stacks: 
0       0x7fb372fe4c90p void paddle::platform::EnforceNotMet::Init<char const*>(char const*, char const*, int) + 352
1       0x7fb372fe5009p paddle::platform::EnforceNotMet::EnforceNotMet(std::__exception_ptr::exception_ptr, char const*, int) + 137
2       0x7fb373250097p void paddle::operators::math::Blas<paddle::platform::CUDADeviceContext>::GEMM<float>(CBLAS_TRANSPOSE, CBLAS_TRANSPOSE, int, int, int, float, float const*, float const*, float, float*) const + 1815
3       0x7fb3736caa6cp void paddle::operators::math::Blas<paddle::platform::CUDADeviceContext>::MatMul<float>(paddle::framework::Tensor const&, bool, paddle::framework::Tensor const&, bool, float, paddle::framework::Tensor*, float) const + 844
4       0x7fb3736caea6p paddle::operators::MulKernel<paddle::platform::CUDADeviceContext, float>::Compute(paddle::framework::ExecutionContext const&) const + 662
5       0x7fb3736cb093p std::_Function_handler<void (paddle::framework::ExecutionContext const&), paddle::framework::OpKernelRegistrarFunctor<paddle::platform::CUDAPlace, false, 0ul, paddle::operators::MulKernel<paddle::platform::CUDADeviceContext, float>, paddle::operators::MulKernel<paddle::platform::CUDADeviceContext, double>, paddle::operators::MulKernel<paddle::platform::CUDADeviceContext, paddle::platform::float16> >::operator()(char const*, char const*, int) const::{lambda(paddle::framework::ExecutionContext const&)#1}>::_M_invoke(std::_Any_data const&, paddle::framework::ExecutionContext const&) + 35
6       0x7fb374c74326p paddle::framework::OperatorWithKernel::RunImpl(paddle::framework::Scope const&, boost::variant<paddle::platform::CUDAPlace, paddle::platform::CPUPlace, paddle::platform::CUDAPinnedPlace, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_> const&, paddle::framework::RuntimeContext*) const + 662
7       0x7fb374c74a94p paddle::framework::OperatorWithKernel::RunImpl(paddle::framework::Scope const&, boost::variant<paddle::platform::CUDAPlace, paddle::platform::CPUPlace, paddle::platform::CUDAPinnedPlace, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_> const&) const + 292
8       0x7fb374c723bcp paddle::framework::OperatorBase::Run(paddle::framework::Scope const&, boost::variant<paddle::platform::CUDAPlace, paddle::platform::CPUPlace, paddle::platform::CUDAPinnedPlace, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_, boost::detail::variant::void_> const&) + 332
9       0x7fb3731573aep paddle::framework::Executor::RunPreparedContext(paddle::framework::ExecutorPrepareContext*, paddle::framework::Scope*, bool, bool, bool) + 382
10      0x7fb3731581efp paddle::framework::Executor::Run(paddle::framework::ProgramDesc const&, paddle::framework::Scope*, int, bool, bool, std::vector<std::string, std::allocator<std::string> > const&, bool) + 143
11      0x7fb372fd489ep
12      0x7fb37301777ep
13      0x55c1828a34e8p PyCFunction_Call + 120
14      0x55c1828ffec4p PyEval_EvalFrameEx + 23284
15      0x55c1828ff490p PyEval_EvalFrameEx + 20672
16      0x55c1828ff490p PyEval_EvalFrameEx + 20672
17      0x55c1828fae90p PyEval_EvalFrameEx + 2752
18      0x55c18290503dp PyEval_EvalCodeEx + 525
19      0x55c182905f5cp PyEval_EvalCode + 28
20      0x55c182962454p
21      0x55c18282eb5ep PyRun_InteractiveOneObject + 709
22      0x55c18282ed01p PyRun_InteractiveLoopFlags + 239
23      0x55c18282ed62p
24      0x55c18282f080p
25      0x55c18282f871p main + 225
26      0x7fb3b31d2830p __libc_start_main + 240
27      0x55c182907808p
PS：本机TensorFlow-gpu和PyTorch-gpu都可以正常运行。
PaddlePaddle / Paddle

Ubutun 安装gpu版本无异常，但是paddle.fluid.install_check.run_check()报错 #17496
