alexlamfromhome
commented
1 year ago I have upgraded to OpenVINO 2022.2, but the problem still exists.
(ovtf) alexlam@vct-adl-p:~/projects/openvino_tensorflow$ python3 examples/classification_sample.py --no_show --backend GPU
2022-11-23 15:57:37.335409: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2022-11-23 15:57:37.338252: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /opt/intel/openvino_2022/tools/compile_tool:/opt/intel/openvino_2022/runtime/3rdparty/tbb/lib:/opt/intel/openvino_2022/runtime/3rdparty/hddl/lib:/opt/intel/openvino_2022/runtime/lib/intel64
2022-11-23 15:57:37.338266: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
2022-11-23 15:57:38.395115: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcuda.so.1'; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /home/alexlam/projects/openvino_tensorflow/venv/lib/python3.8/site-packages/cv2/../../lib64:/opt/intel/openvino_2022/tools/compile_tool:/opt/intel/openvino_2022/runtime/3rdparty/tbb/lib:/opt/intel/openvino_2022/runtime/3rdparty/hddl/lib:/opt/intel/openvino_2022/runtime/lib/intel64
2022-11-23 15:57:38.395136: W tensorflow/stream_executor/cuda/cuda_driver.cc:269] failed call to cuInit: UNKNOWN ERROR (303)
2022-11-23 15:57:38.395147: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (vct-adl-p): /proc/driver/nvidia/version does not exist
2022-11-23 15:57:38.395293: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 AVX_VNNI FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
Available Backends:
CPU
GPU
VAD-M
2022-11-23 15:57:39.733766: OVTF Summary -> 138 out of 899 nodes in the graph (15%) are now running with OpenVINO™ backend
Segmentation fault (core dumped)But when I test the same on 11th Gen, everything works
(ovtf) alexlam@vct-tgl2:~/projects/openvino_tensorflow$ python3 examples/classification_sample.py --no_show --backend GPU
2022-11-23 16:04:12.426052: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2022-11-23 16:04:12.428993: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory
2022-11-23 16:04:12.429008: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
2022-11-23 16:04:13.590567: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcuda.so.1'; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /home/alexlam/projects/openvino_tensorflow/venv/lib/python3.8/site-packages/cv2/../../lib64:
2022-11-23 16:04:13.590591: W tensorflow/stream_executor/cuda/cuda_driver.cc:269] failed call to cuInit: UNKNOWN ERROR (303)
2022-11-23 16:04:13.590605: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (alexlam-vss-tgl2): /proc/driver/nvidia/version does not exist
2022-11-23 16:04:13.590963: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 AVX512F AVX512_VNNI FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
Available Backends:
CPU
GPU
2022-11-23 16:04:15.407457: OVTF Summary -> 138 out of 899 nodes in the graph (15%) are now running with OpenVINO™ backend
Inference time in ms: 20.30
military uniform 0.79601717
mortarboard 0.020910148
academic gown 0.014557097
suit 0.009166191
comic book 0.007978324The segfault added this in my dmesg
[2005709.437153] python3[1646147]: segfault at 2f2a30303b ip 00007f52351a7d8b sp 00007f50ab7fb800 error 4 in libc-2.31.so[7f5235132000+178000]
[2005709.437173] Code: 0f b7 02 83 c0 01 66 89 02 0f b7 c0 48 3b 05 6c 45 15 00 0f 83 86 00 00 00 48 8b 56 10 48 85 d2 74 7d 64 8b 04 25 18 00 00 00 <4c> 8b 52 10 85 c0 74 9d eb 4a 0f 1f 00 48 8b 50 10 49 89 c0 4c 89
I was able to complete the example on 12th Gen Core Processor using CPU backend. But when I change to GPU backend, the example SegFault.
I have verified that OpenVINO 2022.1 was correctly installed by running the OpenVINO benchmark_app against the GPU device.