Closed atabak-cve closed 3 years ago
I installed it, but I do not know how. Doing a lot of your suggestions. I leave it open in case you want to diagnose it but feel free to close it if you like.
I still have this error.
My environment is nvidia-docker, and there have not /usr/lib/x86_64-linux-gnu/libEGL_nvidia.so.0,
and Python3.7, tensorflow1.15.0, cuda10.0.
Can I use the nvidia-docker ?????
in addition,the libEGL_nvidia.so.0 need to match to the cuda-driver of the host or the cuda-runtime of docker????
I still have this problem, could you help me provide some suggestions?
2023-03-20 02:49:22.708577: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1304] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 20237 MB memory) -> physical GPU (device: 0, name: Tesla V100-PCIE-32GB, pci bus id: 0000:41:00.0, compute capability: 7.0) 2023-03-20 02:49:24.350874: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1 2023-03-20 02:49:24.353094: F /home/gqz/gqzwork/tvcg/dirt/csrc/gl_common.h:65] none of 1 egl devices matches the active cuda device [1] 30182 abort (core dumped) CUDA_VISIBLE_DEVICES=7 python tests/square_test.py
I also have issues running the test, both with master and this PR. The info that you usually ask in other similar issues are as follows:
Output of
ls -l /usr/lib*/*/*GL*
:And the patch that you suggested here
And finally the output of
nvidia-smi -q
: