Error in `python3': free() occur on "trt.utils.uff_to_trt_engine" (on NVIDIA's sample "2.3.2.1. TensorFlow Python Workflow")

*** The program -------------------------------------------

MNIST_DATASETS = input_data.read_data_sets(’/tmp/tensorflow/mnist/input_data’)
tf_model = run_training(MNIST_DATASETS)

uff_model = uff.from_tensorflow(tf_model, [“fc2/Relu”])
G_LOGGER = trt.infer.ConsoleLogger(trt.infer.LogSeverity.ERROR)

parser = uffparser.create_uff_parser()
parser.register_input(“Placeholder”, (1,28,28),0)
parser.register_output(“fc2/Relu”)
engine = trt.utils.uff_to_trt_engine(G_LOGGER, uff_model, parser, 1, 1 << 20)

*** Error Log -------------------------------------------

*** Error in `python3’: free(): invalid next size (fast): 0x0000000005a7c0f0 ***
======= Backtrace: =========
/lib/x86_64-linux-gnu/libc.so.6(+0x777e5)[0x7f8c3887d7e5]
/lib/x86_64-linux-gnu/libc.so.6(+0x8037a)[0x7f8c3888637a]
/lib/x86_64-linux-gnu/libc.so.6(cfree+0x4c)[0x7f8c3888a53c]
/usr/local/cuda-8.0/targets/x86_64-linux/lib/libcudnn.so.6(cudnnDestroyConvolutionDescriptor+0x9)[0x7f8bf83a4c69]
/usr/lib/x86_64-linux-gnu/libnvinfer.so.4(_ZN8nvinfer15cudnn21CudnnConvolutionLayer15getValidTacticsERKNS0_18EngineBuildContextE+0x1cc)[0x7f8bb6b2a63c]
/usr/lib/x86_64-linux-gnu/libnvinfer.so.4(_ZN8nvinfer15cudnn33selectFastestLayerAndDeleteOthersERNS0_18EngineBuildContextERKSt6vectorIPNS0_5LayerESaIS5_EE+0x7a9)[0x7f8bb6add6a9]
/usr/lib/x86_64-linux-gnu/libnvinfer.so.4(_ZN8nvinfer17builder16buildSingleLayerERNS_5cudnn18EngineBuildContextERNS0_4NodeERKSt13unordered_mapISsSt10unique_ptrINS1_6RegionESt14default_deleteIS8_EESt4hashISsESt8equal_toISsESaISt4pairIKSsSB_EEERNS_14CpuMemoryGroupEPS6_ISsSt6vectorIfSaIfEESD_SF_SaISG_ISH_SR_EEEb+0x509)[0x7f8bb6ac1119]
/usr/lib/x86_64-linux-gnu/libnvinfer.so.4(_ZN8nvinfer17builder18EngineTacticSupply13getBestTacticERNS0_4NodeERKNS_5query5PortsINS_13RegionFormatLEEEb+0x2e9)[0x7f8bb6ac3499]
/usr/lib/x86_64-linux-gnu/libnvinfer.so.4(+0x398d04)[0x7f8bb6b02d04]
/usr/lib/x86_64-linux-gnu/libnvinfer.so.4(_ZN8nvinfer17builder23chooseFormatsAndTacticsERNS0_5GraphERNS0_12TacticSupplyEPSt13unordered_mapISsSt6vectorIfSaIfEESt4hashISsESt8equal_toISsESaISt4pairIKSsS8_EEE+0xba0)[0x7f8bb6b05740]
/usr/lib/x86_64-linux-gnu/libnvinfer.so.4(_ZN8nvinfer17builder19makeEngineFromGraphERKNS_21CudaEngineBuildConfigERKNS_5cudnn15HardwareContextERNS0_5GraphEPSt13unordered_mapISsSt6vectorIfSaIfEESt4hashISsESt8equal_toISsESaISt4pairIKSsSD_EEEi+0x26e)[0x7f8bb6ac470e]
/usr/lib/x86_64-linux-gnu/libnvinfer.so.4(_ZN8nvinfer17builder11buildEngineERNS_21CudaEngineBuildConfigERKNS_5cudnn15HardwareContextERKNS_7NetworkE+0x5a3)[0x7f8bb6ac8183]
/usr/lib/x86_64-linux-gnu/libnvinfer.so.4(_ZN8nvinfer17Builder15buildCudaEngineERNS_18INetworkDefinitionE+0x11)[0x7f8bb6aa4dd1]
/usr/lib/python3.5/dist-packages/tensorrt/infer/_nv_infer_bindings.cpython-35m-x86_64-linux-gnu.so(+0x20161)[0x7f8bcccdd161]
python3(PyCFunction_Call+0x4f)[0x4e9b7f]
python3(PyEval_EvalFrameEx+0x614)[0x5372f4]
python3[0x540199]
python3(PyEval_EvalFrameEx+0x50b2)[0x53bd92]
python3[0x540199]
python3(PyEval_EvalCode+0x1f)[0x540e4f]
python3[0x60c272]
python3(PyRun_FileExFlags+0x9a)[0x60e71a]
python3(PyRun_SimpleFileExFlags+0x1bc)[0x60ef0c]
python3(Py_Main+0x456)[0x63fb26]
python3(main+0xe1)[0x4cfeb1]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf0)[0x7f8c38826830]
python3(_start+0x29)[0x5d6049]
======= Memory map: ========
00400000-007a9000 r-xp 00000000 08:21 28049422 /usr/bin/python3.5
009a9000-009ab000 r–p 003a9000 08:21 28049422 /usr/bin/python3.5
009ab000-00a42000 rw-p 003ab000 08:21 28049422 /usr/bin/python3.5
00a42000-00a73000 rw-p 00000000 00:00 0
01bce000-0dddf000 rw-p 00000000 00:00 0 [heap]
10000000-10001000 rw-s 00000000 00:06 558 /dev/nvidia0
10001000-10002000 rw-s 00000000 00:06 558 /dev/nvidia0
10002000-10003000 rw-s 00000000 00:06 558 /dev/nvidia0
10003000-10004000 rw-s 00000000 00:06 558 /dev/nvidia0
10004000-10005000 rw-s 00000000 00:06 558 /dev/nvidia0
10005000-10006000 rw-s 00000000 00:06 558 /dev/nvidia0
10006000-10007000 rw-s 00000000 00:06 558 /dev/nvidia0
10007000-10008000 rw-s 00000000 00:06 558 /dev/nvidia0
10008000-10009000 rw-s 00000000 00:06 558 /dev/nvidia0
10009000-1000a000 rw-s 00000000 00:06 558 /dev/nvidia0
1000a000-1000b000 rw-s 00000000 00:06 558 /dev/nvidia0
1000b000-1000c000 rw-s 00000000 00:06 558 /dev/nvidia0