Problem 137 - G4LEVELGAMMADATA ? / segmentation fault when inactivating PiMinusAbsorptionAtRest and Decay
Summary: G4LEVELGAMMADATA ? / segmentation fault when inactivating PiMinusAbsorptionAt...
Status: RESOLVED WONTFIX
Alias: None
Product: Geant4
Classification: Unclassified
Component: processes/hadronic (show other problems)
Version: 2.0
Hardware: PC Linux
: P2 normal
Assignee: kurasige
URL:
Depends on:
Blocks:
 
Reported: 2000-07-21 08:52 CEST by V.Hejny
Modified: 2004-05-11 00:44 CEST (History)
1 user (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this problem.
Description V.Hejny 2000-07-21 08:52:18 CEST
When using the PhysicsList of example N04, which contains a 'mostly
complete' set of processes, one question and one problem arise.

(1) I've been told to set G4LEVELGAMMADATA, but nowhere is written what
this variable means and where it should point to. I assume (!) that I
have to download one of the optional data sets, but I don't know ...

(2) When turning off both, absorption at rest and decay, using
/processes/inactivate PiMinusAbsorptionAtRest
/processes/inactivate Decay
Geant4 crashes when starting a run.

By the way: I saw, that there are two methods for handling absorption
at rest (for pi- and K-). What's the difference, because they give very
different results?
Comment 1 Hans-Peter.Wellisch 2000-07-22 10:03:59 CEST
ad 1) The G4LEVELGAMMADATA variable should point to the PhotonEvaporation data.
ad 2) Hisaya will be able to help in this.
ad 3) There are two classes. One type includes the prompt nucleons, the other
one not.
Comment 2 kurasige 2000-08-22 22:40:59 CEST
(Problem 2)

  It is dangerous to inactivate "decay" without any "special" care.
  Crashes easily occur due to the following reason.
   1) Suppose a pion (or any kind of particles which can decay) stop in a
material.
   2) Normally, it will decay at rest. But decay is inactivated in this case
and no process can take care of it.
   3) G4SteppingManager::InvokeAtRestDoItProcs assumes that at lease one
process can be invoked at rest step. Without process, its behaviour is not
defined and sometimes it crashes.

   I think that crashes can be avoided by fixing
G4SteppingManager::InvokeAtRestDoItProcs method.
   However, the Stepping Manager can only exit Geant4 with warning messages
because NO behaviour is defined for particles at rest.
Comment 3 Takashi.Sasaki 2004-05-11 00:44:59 CEST
*** Bug 622 has been marked as a duplicate of this bug. ***
*** 'kurasige@phys.sci.kobe-u.ac.jp' will be responsible for bug 622. ***