Problem 2491 - Rare crashes of NeutronHP in Geant4 11.0.p02 and 11.0.ref05
Summary: Rare crashes of NeutronHP in Geant4 11.0.p02 and 11.0.ref05
Status: RESOLVED FIXED
Alias: None
Product: Geant4
Classification: Unclassified
Component: processes/hadronic/models/neutron_hp (show other problems)
Version: 11.0
Hardware: All All
: P5 critical
Assignee: Alberto.Ribon
URL:
Depends on:
Blocks:
 
Reported: 2022-06-09 20:13 CEST by Alberto.Ribon
Modified: 2022-06-28 15:15 CEST (History)
0 users

See Also:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this problem.
Description Alberto.Ribon 2022-06-09 20:13:48 CEST 
Rare cases of crashes when using physics lists with _HP (e.g. FTFP_BERT_HP, QGSP_BERT_HP), using Geant4 11.0.p02 or 11.0.ref05.
The crashes messages are per-se useless, but show that there are rare cases with huge energy violations, with ions (likely nuclear residuals) with huge energy (thousands of TeV, with a few GeV primary particle impinging on a calorimeter).

The problem is very difficult to reproduce, and it is not seen at process-level.
Comment 1 Alberto.Ribon 2022-06-28 15:15:15 CEST 
The problem has been understood and fixed: it was due to access to arrays at negative index (-1) in the method G4ParticleHPContAngularPar::Sample.
The fix will appear in G4 11.1.beta and 11.0.p03.