Skip to content

Place the L3 coil turns explicitly instead of dividing the polyhedra - #15767

Open
sawenzel wants to merge 1 commit into
AliceO2Group:devfrom
sawenzel:swenzel/l3-coil-explicit-turns
Open

Place the L3 coil turns explicitly instead of dividing the polyhedra#15767
sawenzel wants to merge 1 commit into
AliceO2Group:devfrom
sawenzel:swenzel/l3-coil-explicit-turns

Conversation

@sawenzel

@sawenzel sawenzel commented Sep 4, 2026

Copy link
Copy Markdown
Collaborator

This fixes a problem in navigation with multithreaded native Geant4 and leaves the material unchanged.

  • The 168 coil turns were made with TGeoPgon::Divide. Geant4's G4ParameterisationPolyhedraZ rebuilds the shared master solid while navigating and allocates the new sides' thread-split data on worker threads, so a multithreaded run segfaults in the G4PolyhedraSide constructor.
  • L3CM is the only divided polyhedra in ALICE. The other 24 divisions are box divisions in HMPID, ZDC and FIT and are not affected.

https://its.cern.ch/jira/browse/O2-7153

This fixes a problem in navigation with multithreaded native Geant4 and leaves
the material unchanged.

- The 168 coil turns were made with TGeoPgon::Divide. Geant4's
  G4ParameterisationPolyhedraZ rebuilds the shared master solid while navigating
  and allocates the new sides' thread-split data on worker threads, so a
  multithreaded run segfaults in the G4PolyhedraSide constructor.
- L3CM is the only divided polyhedra in ALICE. The other 24 divisions are box
  divisions in HMPID, ZDC and FIT and are not affected.

https://its.cern.ch/jira/browse/O2-7153

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Development

Successfully merging this pull request may close these issues.

1 participant