Deal with mutable materials such as a gas mixture.

To deal with mutable materials such as a gas mixture, the GeoLogVol now has a method to update the material

@jojungge MR is created, can you please add the protections you were recommending?

added 1 commit

• b3360a46 - Apply 1 suggestion(s) to 1 file(s)

Compare with previous version

added 1 commit

• 5968d14a - Apply 1 suggestion(s) to 1 file(s)

Compare with previous version

added 1 commit

• 607779b9 - Apply 1 suggestion(s) to 1 file(s)

Compare with previous version

resolved all threads

added 1 commit

• cb0da76c - Apply 1 suggestion(s) to 1 file(s)

Compare with previous version

enabled an automatic merge when the pipeline for cb0da76c succeeds

aborted the automatic merge because source branch was updated

added 1 commit

• 4115c6b6 - fix compilation error

Compare with previous version

added 1 commit

• 87decf42 - TEST the ATHENA pipeline without mutex in the modified GeoLogVol

Compare with previous version

The athena pipelines fails when the FCAL construction downcasts a pointer, here:

https://gitlab.cern.ch/atlas/athena/-/blob/main/LArCalorimeter/LArGeoModel/LArReadoutGeometry/src/FCALModule.cxx?ref_type=heads#L34

So I this latest commit is just a test, to see whether the issue is coming from the mutex or from the new setMaterial method. When understood we can hopefully find a solution.

In the meantime, the FCALModule on the Athena side should be updated to remove this downcasting.

marked this merge request as draft

added 1 commit

• 87118012 - Add back the header file inclusion and retest

Compare with previous version

added 1 commit

• c697380d - Add back mutex member data

Compare with previous version

The down cast shouldn't be the problem. It's probably the changed memory layout of the class and that somehow the athena integration pipeline broke.

I think so too. But, from the log file I cannot see where it would pickup up the old, unmodified header file from, the include directives all seem correct to me. I was going to test the downcasting locally anyway, though I agree it's probably futile.