# Tensorflow-TRT integration not working

**URL:** <https://forums.developer.nvidia.com/t/tensorflow-trt-integration-not-working/62770>\
**Category:** TensorRT\
**Created:** [June 29, 2018, 10:52pm UTC](https://forums.developer.nvidia.com/t/tensorflow-trt-integration-not-working/62770 "2018-06-29T22:52:30Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![wuxiekeji](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/wuxiekeji/32/75304_2.png) [@wuxiekeji](https://forums.developer.nvidia.com/u/wuxiekeji)\
**Post date:** [June 29, 2018, 10:52pm UTC](https://forums.developer.nvidia.com/t/tensorflow-trt-integration-not-working/62770/1 "2018-06-29T22:52:30Z")

</div>

The default ./run\_all.sh in the TFTRT example gives the following error. Any ideas what I need to do to solve this?

Tensorflow 1.8 + TensorRT 4 on Ubuntu 16.04, GTX1070, CUDA 9.0, cuDNN 7.0.5. Thanks!

```auto
2018-06-29 22:50:47.065185: I tensorflow/core/grappler/devices.cc:51] Number of eligible GPUs (core count >= 8): 1
2018-06-29 22:50:47.656448: I tensorflow/contrib/tensorrt/convert/convert_nodes.cc:2660] Max batch size= 4 max workspace size= 2147483648
2018-06-29 22:50:47.656480: I tensorflow/contrib/tensorrt/convert/convert_nodes.cc:2666] starting build engine
***Error in `python': munmap_chunk(): invalid pointer: 0x00007ffcd22d42d0***
======= Backtrace: =========
/lib/x86_64-linux-gnu/libc.so.6(+0x777e5)[0x7fb7af7097e5]
/lib/x86_64-linux-gnu/libc.so.6(cfree+0x1a8)[0x7fb7af716698]
/usr/local/lib/python2.7/dist-packages/tensorflow/python/../libtensorflow_framework.so(_ZNSt10_HashtableISsSsSaISsENSt8__detail9_IdentityESt8equal_toISsESt4hashISsENS1_18_Mod_range_hashingENS1_20_Default_ranged_hashENS1_20_Prime_rehash_policyENS1_17_Hashtable_traitsILb1ELb1ELb1EEEE21_M_insert_unique_nodeEmmPNS1_10_Hash_nodeISsLb1EEE+0xfc)[0x7fb772c2cd3c]
/usr/lib/x86_64-linux-gnu/libnvinfer.so.4(_ZNSt10_HashtableISsSsSaISsENSt8__detail9_IdentityESt8equal_toISsESt4hashISsENS1_18_Mod_range_hashingENS1_20_Default_ranged_hashENS1_20_Prime_rehash_policyENS1_17_Hashtable_traitsILb1ELb1ELb1EEEE9_M_insertIRKSsNS1_10_AllocNodeISaINS1_10_Hash_nodeISsLb1EEEEEEEESt4pairINS1_14_Node_iteratorISsLb1ELb1EEEbEOT_RKT0_St17integral_constantIbLb1EE+0x96)[0x7fb72a45ed46]
/usr/lib/x86_64-linux-gnu/libnvinfer.so.4(_ZNK8nvinfer17Network8validateERKNS_5cudnn15HardwareContextEbbi+0x131)[0x7fb72a45aab1]
/usr/lib/x86_64-linux-gnu/libnvinfer.so.4(_ZN8nvinfer17builder11buildEngineERNS_21CudaEngineBuildConfigERKNS_5cudnn15HardwareContextERKNS_7NetworkE+0x46)[0x7fb72a443606]
/usr/lib/x86_64-linux-gnu/libnvinfer.so.4(+0x481ce1)[0x7fb72a494ce1]
/usr/local/lib/python2.7/dist-packages/tensorflow/contrib/tensorrt/_wrap_conversion.so(_ZN10tensorflow8tensorrt7convert32ConvertSubGraphToTensorRTNodeDefERNS1_14SubGraphParamsE+0x1b2b)[0x7fb729bf73bb]
/usr/local/lib/python2.7/dist-packages/tensorflow/contrib/tensorrt/_wrap_conversion.so(+0x5a49a)[0x7fb729bd449a]
/usr/local/lib/python2.7/dist-packages/tensorflow/contrib/tensorrt/_wrap_conversion.so(_ZN10tensorflow8tensorrt7convert25ConvertGraphDefToTensorRTERKNS_8GraphDefERKSt6vectorISsSaISsEEmmPS2_ii+0x160c)[0x7fb729bd62ec]
/usr/local/lib/python2.7/dist-packages/tensorflow/contrib/tensorrt/_wrap_conversion.so(+0x53b8f)[0x7fb729bcdb8f]
/usr/local/lib/python2.7/dist-packages/tensorflow/contrib/tensorrt/_wrap_conversion.so(+0x5421a)[0x7fb729bce21a]

```