Problem 645

Summary:	a crash for neutron induced reaction in copper
Product:	Geant4	Reporter:	Hans-Peter.Wellisch
Component:	processes/hadronic/models	Assignee:	Hans-Peter.Wellisch
Status:	RESOLVED FIXED
Severity:	normal	CC:	Gabriele.Cosmo
Priority:	P2
Version:	6.2
Hardware:	PC
OS:	Linux

Description Hans-Peter.Wellisch 2004-07-12 02:58:15 CEST

Hi,

  we have a crash (segmentation fault) in the Bertini cascade, where the GHAD
signal handler is then dumping the following as reaction initial state:

GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD
Dumping the registered hadronic state information
Nucleus A, Z = 63 29
Projectile was a neutron
projectile momentum (px, py, pz) = (1.28521e-15, 3.69843e-15, 35.9839)
Projectile energy = 940.254
GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD_GHAD

This happened when trying the new G4Allocator. Gabriele knows the details.

Many greetings,

Hans-Peter.

Comment 1 Hans-Peter.Wellisch 2004-09-10 02:56:59 CEST

Likely solves by commenting lines 391 and 392 in
hadronic/models/cascade/cascade/G4CascadeInterface.cc.
Can you please verify?

Many greetings,

Hans-Peter.

Comment 2 Gabriele Cosmo 2004-09-10 04:58:59 CEST

By commenting the two lines in G4CascadeInterface.cc (revision 1.54) as
suggested, the crash observed in test25 is resolved. Also the run-time error
reported by Valgrind is fixed.
Please, provide a tag including the fix which I see you've committed on the
HEAD (revision 1.55 of that file).

Comment 3 Gabriele Cosmo 2004-09-20 02:29:59 CEST

The fix is included in tag "ghad-bert-cas-V06-02-01".