-
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
Segfault in HitPairEDProducer #36804
Comments
assign reconstruction |
New categories assigned: reconstruction @slava77,@jpata,@clacaputo you have been requested to review this Pull request/Issue and eventually sign? Thanks |
A new Issue was created by @makortel Matti Kortelainen. @Dr15Jones, @perrotta, @dpiparo, @makortel, @smuzaffar, @qliphy can you please review it and eventually sign/assign? Thanks. cms-bot commands are listed here |
Possible culprit could be #36184 |
Crash was also seen in #36801 (comment) workflow 1040.0 step 2
|
@AdrianoDee |
Also CAHitQuadrupletEDProducer is showing crashes (#36808). |
I'm not being able to reproduce this (both for
with four threads and streams. Everything is running smoothly. |
Could valgrind reveal anything? (#36801 (comment) showed a crash on first event in 1040.0 step 2 on one thread) |
In the traceback, the problem comes from this call
and is somewhere in this function cmssw/RecoTracker/TkHitPairs/plugins/HitPairEDProducer.cc Lines 172 to 176 in ec53ab6
|
Looking at the code, I think the most likely problem is the |
The call chain from the cmssw/RecoTracker/TkHitPairs/interface/RegionsSeedingHitSets.h Lines 36 to 39 in ec53ab6
cmssw/RecoTracker/TkSeedingLayers/interface/SeedingHitSet.h Lines 45 to 50 in a8490c8
The last loop does (with theRecHits having two elements)
|
Should move it to |
How about filtering out the argument pointers after first |
Indeed, this method was there while the container had fixed size. Now doesn't make too much sense to stick to it. Let me prepare a PR. |
The latest ASAN builds have a large number of heap-buffer-overflow errors that I suspect are related (I don't see any other obvious candidates). It seems to be highly reproducible in the ASAN build (803 failures):
|
@dan131riley, thanks. Indeed it was reproducible with the ASAN build (e.g. |
this is solved by #36819 right? |
@cms-sw/reconstruction-l2 , I think this has been resolved, can you please confirm? |
+reconstruction |
This issue is fully signed and ready to be closed. |
Workflow 1030.0 crashed in CMSSW_12_3_X_2022-01-25-1700
https://cmssdt.cern.ch/SDT/cgi-bin/logreader/slc7_amd64_gcc10/CMSSW_12_3_X_2022-01-25-1700/pyRelValMatrixLogs/run/1030.0_RunHLTPhy2017B+RunHLTPhy2017B+TIER0EXPHPBS+ALCASPLITHPBS+ALCAHARVDHPBS+ALCAHARVDHPBSLOWPU/step2_RunHLTPhy2017B+RunHLTPhy2017B+TIER0EXPHPBS+ALCASPLITHPBS+ALCAHARVDHPBS+ALCAHARVDHPBSLOWPU.log#/
The text was updated successfully, but these errors were encountered: