🐛 Bug
Currently experiencing a core dump when running smoke tests for CNN on python 3.9:
Nov 11 09:07:54 + python /builder/test_example_code/cnn_smoke.py
Nov 11 09:07:54 Checking that basic CNN works
Nov 11 09:08:00 tensor([0.0817], device='cuda:0', grad_fn=<ViewBackward>)
Nov 11 09:08:00 terminate called without an active exception
Nov 11 09:08:00 terminate called recursively
Nov 11 09:08:01 /builder/check_binary.sh: line 374: 453 Aborted (core dumped) python ${TEST_CODE_DIR}/cnn_smoke.py
Link to CircleCI logs: https://app.circleci.com/pipelines/github/pytorch/pytorch/237660/workflows/5021f68c-07d3-4fcb-8a2a-c1b14b87d9e2/jobs/8843626
To Reproduce
#!/usr/bin/env bash
docker run --rm -it --gpus all pytorch/manylinux-cuda101 bash -s <<EOF
export PATH=/opt/python/cp39-cp39/bin/:${PATH}
pip install https://8842687-65600975-gh.circle-artifacts.com/0/final_pkgs/torch-1.8.0.dev20201111%2Bcu101-cp39-cp39-linux_x86_64.whl
git clone https://github.com/pytorch/builder
pushd builder
python test_example_code/cnn_smoke.py
EOFExpected behavior
Smoke tests don't fail
Environment
Please copy and paste the output from our
environment collection script
(or fill out the checklist below manually).
You can get the script and run it with:
wget https://raw.githubusercontent.com/pytorch/pytorch/master/torch/utils/collect_env.py
# For security purposes, please check the contents of collect_env.py before running it.
python collect_env.py
PyTorch version: 1.8.0.dev20201111+cu101
Is debug build: False
CUDA used to build PyTorch: 10.1
ROCM used to build PyTorch: N/A
OS: CentOS Linux 7 (Core) (x86_64)
GCC version: (GCC) 7.3.1 20180303 (Red Hat 7.3.1-5)
Clang version: Could not collect
CMake version: Could not collect
Python version: 3.9 (64-bit runtime)
Is CUDA available: True
CUDA runtime version: 10.1.243
GPU models and configuration:
GPU 0: Quadro GP100
GPU 1: Quadro GP100
Nvidia driver version: 418.116.00
cuDNN version: /usr/local/cuda-10.1/targets/x86_64-linux/lib/libcudnn.so.7.6.3
HIP runtime version: N/A
MIOpen runtime version: N/A
Versions of relevant libraries:
[pip3] numpy==1.19.4
[pip3] torch==1.8.0.dev20201111+cu101
[conda] Could not collect
Additional context
This issue didn't present itself when doing my initial testing for Python 3.9 here: #47238
cc @ezyang @gchanan @zou3519 @bdhirsh @albanD @gqchen @pearu @nikitaved
🐛 Bug
Currently experiencing a core dump when running smoke tests for CNN on python 3.9:
Link to CircleCI logs: https://app.circleci.com/pipelines/github/pytorch/pytorch/237660/workflows/5021f68c-07d3-4fcb-8a2a-c1b14b87d9e2/jobs/8843626
To Reproduce
Expected behavior
Smoke tests don't fail
Environment
Please copy and paste the output from our
environment collection script
(or fill out the checklist below manually).
You can get the script and run it with:
Additional context
This issue didn't present itself when doing my initial testing for Python 3.9 here: #47238
cc @ezyang @gchanan @zou3519 @bdhirsh @albanD @gqchen @pearu @nikitaved