Problem 2340

Summary:	Division by zero in FTF stopping
Product:	Geant4	Reporter:	Vladimir.Ivantchenko
Component:	processes/hadronic/stopping	Assignee:	Alberto.Ribon
Status:	RESOLVED WONTFIX
Severity:	major
Priority:	P4
Version:	10.7
Hardware:	All
OS:	All

Description Vladimir.Ivantchenko 2021-03-01 09:54:50 CET

Hello,

there is a problem in continuous test64: http://cdash.cern.ch/testDetails.php?test=7660223&build=64262 

It is reproduced at pcgeant06 with gcc9.2 if random seed = 1614528527, 
it should be set by hand in the test64.cc

The problem is likely here for a long time and may affect LHC experiment productions.


Call Stack:
12: /build/jenkins/workspace/BuildProducts/lib64/libG4run.so
11: /build/jenkins/workspace/BuildProducts/lib64/libG4run.so
10: /lib64/libpthread.so.0
9: /build/jenkins/workspace/BuildProducts/lib64/libG4processes.so : G4FTFAnnihilation::Create2QuarkAntiQuarkStrings(G4VSplitableHadron*, G4VSplitableHadron*, G4FTFParameters*, G4FTFAnnihilation::CommonVariables&) const
8: /build/jenkins/workspace/BuildProducts/lib64/libG4processes.so : G4FTFAnnihilation::Annihilate(G4VSplitableHadron*, G4VSplitableHadron*, G4VSplitableHadron*&, G4FTFParameters*) const
7: /build/jenkins/workspace/BuildProducts/lib64/libG4processes.so : G4FTFModel::ExciteParticipants()
6: /build/jenkins/workspace/BuildProducts/lib64/libG4processes.so : G4FTFModel::GetStrings()
5: /build/jenkins/workspace/BuildProducts/lib64/libG4processes.so : G4VPartonStringModel::Scatter(G4Nucleus const&, G4DynamicParticle const&)
4: /build/jenkins/workspace/BuildProducts/lib64/libG4processes.so : G4TheoFSGenerator::ApplyYourself(G4HadProjectile const&, G4Nucleus&)
3: /build/jenkins/workspace/BuildProducts/lib64/libG4processes.so : G4HadronStoppingProcess::AtRestDoIt(G4Track const&, G4Step const&)
2: /build/jenkins/workspace/tests/test64/test64
1: /lib64/libc.so.6
0: /build/jenkins/workspace/tests/test64/test64

Comment 1 Vladimir.Ivantchenko 2021-03-01 10:37:56 CET

Hi Dennis,

you may assign this problem to Alberto first, who already look into it.

Cheers,
Vladimir

Comment 2 Alberto.Ribon 2021-03-11 15:47:31 CET

I have run millions of annihilation at rest with FTFP, but was unable to reproduce any crash. So I can't fix anything.

Comment 3 Vladimir.Ivantchenko 2021-03-11 15:52:48 CET

Hi Alberto,

have you set random seed to this exact value and run at gcc9.2?

Cheers,
Vladimir