-
-
Notifications
You must be signed in to change notification settings - Fork 11k
Numpy 1.19.5, Python 3.6.9: Import results in segmentation fault #18468
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Comments
Someone needs to debug this on the platform in question. Is there a way to get access to this hardware, or could you try a debug build? There was no stack trace or hint as to which instruction failed? |
@mattip Thanks for responding so quickly!
I'm afraid I only have this one device and can't open it up for anyone else. Is there a fast way to install a debug build?
No. I could provide a coredump, though, in case that helps? |
If you could generate a core dump, I wonder if you could take one step further by examining the traceback in gdb? Even if it's not a debug build, experienced NumPy developers (not me!) could probably still guess the source of failure. |
IIRC, there have been other reports on Jetson. |
@charris I just searched for other issues again and, this time, also included closed tickets. I suppose you're referring to #18131 ? It seems to be exactly the same issue that I am having but I can't tell which Numpy release includes the fix – in any case, I suppose I'll have to upgrade Numpy v1.20.x to get it? |
Numpy 1.19.4 ,python 3.6.9 , opencv-python and opencv_contrib_python 4.5.1.48 work, tks~~~ |
Had to rollback to 1.19.4 as well. |
Reproducing code example:
I installed Numpy through
pip3 install --user numpy==1.19.5
on an ARM64 device (Nvidia Jetson Nano) running Ubuntu 20.04 and cannot import it in Python. Interestingly, Numpy 1.16.1 and 1.19.4 work.Error message:
NumPy/Python version information:
Related issues:
The text was updated successfully, but these errors were encountered: