Problem 82

Summary: memory leak and missing energy in hadronic processes
Product: Geant4 Reporter: Sanya.Solodkov
Component: processes/hadronicAssignee: Hans-Peter.Wellisch
Status: CLOSED FIXED    
Severity: normal    
Priority: P2    
Version: 1.1   
Hardware: PC   
OS: Linux   

Description Sanya.Solodkov 2000-04-07 12:10:15 CEST 
In my simulation program I use physics list from example N04
I have no problems with electrons,
but with pions there is a huge memory leak.
Initial size of the program in memory is 26 Mb,
after 1000 50GeV pion events it is 104 Mb  (RedHat5.1, geant4.1.0)
after 904 200GeV pion events it is 318 Mb  (Redhat6.1, geant4.1.1)
Total deposited energy is less than expected.
For 100GeV pions it is only 55GeV
The program is in /afs/cern.ch/user/t/tilecal/geant4/tileTB
Comment 1 Hans-Peter.Wellisch 2000-04-10 13:43:59 CEST 
Seems I identified the problem. The usage of G4FastVector is not consistant with
its memory management policy, consistently throughout the LE code. I changed the
main usage, and the problem is diminished. There still seems to be something
smaller left, but I am down to ~1MB/2000GeV. I am trying to nail this down next.

As to the occasional energy non-conservation: This is a known bug. I stems from
Harms code and he is investigating.
Comment 2 Hans-Peter.Wellisch 2000-04-15 07:46:59 CEST 
I have run 100TeV (1.5 days of running on my solaris box) of hadronic flux
through the tile set-up, without significant growth in memory usage. There is
still something small that I have to nail down, but this should be usable as is.
I'll tag, and start the release cycle.

As for the energy non conservation, Harm is still working on it.
Comment 3 Hans-Peter.Wellisch 2000-07-03 07:46:59 CEST 
Received a fix from Harm, now verifying that it actually fixes the problem.
Comment 4 Hans-Peter.Wellisch 2000-07-20 02:59:59 CEST 
Hi folks,

  the energy non-conservation finally is fixed as well. I'll tag for the next
reference tag. For the meanwhile I have put a tar file of the high_energy code
in /afs/cern.ch/user/h/hpw/public/high_energy.tar.gz

Many greetings,

Hans-Peter.