-
Notifications
You must be signed in to change notification settings - Fork 4.3k
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
Possible memory corruption in 10023.0 #19638
Comments
A new Issue was created by @dan131riley Dan Riley. @davidlange6, @Dr15Jones, @smuzaffar can you please review it and eventually sign/assign? Thanks. cms-bot commands are listed here |
another one, with
in common with the previous one...time to run some valgrind
|
Looking at the stack traces I realized that this is the return of #14897 from a year ago... |
Poking further, both of these are for the slc6_amd64_gcc700 architecture, which is using jemalloc 5.0.1. Per jemalloc/jemalloc#369 jemalloc 5 does not support valgrind. Attempting valgrind memcheck or helgrind gives the error message
valgrind on an architecture with jemalloc 4.5.0 doesn't find any relevant problems (it does find lots else to complain about, though). It's possible this is actually a concurrency problem with jemalloc, or it could be a return of #14897 (which we never did understand, it just empirically went away when we reduced the InputTag churn), but our best tool for finding out doesn't work. I guess I'll try valgrind on slc6_amd64_gcc700 cmsRunGlibC. |
look like this has been fixed |
https://cmssdt.cern.ch/SDT/cgi-bin/buildlogs/slc6_amd64_gcc700/CMSSW_9_3_X_2017-07-07-2300/pyRelValMatrixLogs/run/10023.0_MinBias_13+MinBias_13TeV_pythia8_TuneCUETP8M1_2017_GenSimFullINPUT+DigiFull_2017+RecoFull_2017+ALCAFull_2017+HARVESTFull_2017/step3_MinBias_13+MinBias_13TeV_pythia8_TuneCUETP8M1_2017_GenSimFullINPUT+DigiFull_2017+RecoFull_2017+ALCAFull_2017+HARVESTFull_2017.log
segmentation fault in jemalloc, only seen once so far:
The text was updated successfully, but these errors were encountered: