-
-
Notifications
You must be signed in to change notification settings - Fork 10.9k
Segfault ufunc_object.c:1750 #2843
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
Hard to say much from the back trace... Can you provide a minimal failing
|
I don't really know exactly where it is segfaulting in my python scripts. After it has segfaulted is there a way to find out where it is failing in my script? Like a line number/file? |
Ok figured out how to get line numbers and files in versions of gdb < 7.0 which already has support for this
|
Hmmm... Think it might be a thread safety issue in matplotlib - Can I safely assume that matplotlib is thread safe? |
Ok I've put together a "stack trace" - I manually extracted all the files and function calls and have split them into the gdb part of the stack trace and the python extracted section. Unfortunately it doesn't give the exact line just the start of the function. I also tried writing something to reproduce the issue but I can't seem to come up with anything. It looks as though its a problem with writing matplotlib plots to disk. It seems to be an issue after about 2000+ plots. I have also previously had problems with 'too many files open' using matplotlib - see bug matplotlib/matplotlib#1466 -This was solved by increasing the number of files a user can have open to 200000. C Calls
Python calls
|
Not sure what to tell you, but you might try asking the matplotlib folks? On Wed, Dec 19, 2012 at 12:16 AM, hankhank notifications@github.com wrote:
|
@njsmith: We're happy to have a look over at matplotlib. I didn't realise numpy was having a problem with developer person-power. I always look from afar and am amazed at the productivity over here ;) |
Oh, well, it may just be a grass-looks-greener thing :-). On Thu, Dec 20, 2012 at 1:20 PM, Michael Droettboom <
|
(Closing this for now so it doesn't clutter up the issue list, feel free to re-open if you get more information pointing at numpy or just want to poke us again :-).) |
Hi repeatedly getting a segfault in umath. Its an internal test tool that performs some analysis on PCM voice streams using scipy. I don't really know how to debug the issue but here is the information I collected from the last segfault. I can provide more info if necessary, just need to be pointed to some instructions.
Cheers,
Hank
OS: Linux localhost.localhost 2.6.18-308.16.1.el5 #1 SMP Tue Oct 2 22:01:37 EDT 2012 i686 i686 i386 GNU/Linux - CentOS release 5.8 (Final)
Python: Python2.7
Numpy:
NAME
numpy.version - # THIS FILE IS GENERATED FROM NUMPY SETUP.PY
FILE
/usr/local/lib/python2.7/site-packages/numpy/version.py
DATA
full_version = '1.6.2'
git_revision = 'ca07bce202ae26b6f0a73870eb2ef0b88e0210c5'
release = True
short_version = '1.6.2'
version = '1.6.2'
SciPy:
Help on module scipy.version in scipy:
NAME
scipy.version - # THIS FILE IS GENERATED FROM SCIPY SETUP.PY
FILE
/usr/local/lib/python2.7/site-packages/scipy/version.py
DATA
full_version = '0.11.0'
git_revision = '827679b6f0149abb2c0e121749887012946b99c3'
release = True
short_version = '0.11.0'
version = '0.11.0'
Matplotlib
version = '1.2.0'
version__numpy = '1.4'
The text was updated successfully, but these errors were encountered: