better memory management in Geo2G4Svc

Aims to address a memory leak spotted by valgrind:

==178886== 954,008 (18,616 direct, 935,392 indirect) bytes in 179 blocks are definitely lost in loss record 55,963 of 56,023
==178886== at 0x4C2808C: operator new(unsigned long) (vg_replace_malloc.c:333)
==178886== by 0x2FB56303: Geo2G4AssemblyFactory::Build(GeoPVConstLink, bool&) const (in /cvmfs/atlas-nightlies.cern.ch/repo/sw/21.0/2019-03-18T2158/Athena/21.0.95/InstallArea/x86_64-slc6-gcc62-opt/lib/libGeo2G4Lib.so)

Removes static initializations of AssemblyFactory and further static map in Geo2G4AssemblyFactory.cxx.

An AssemblyFactory is now a member object of Geo2G4Svc. Some functions could no longer be const because Geo2G4AssemblyFactory::Build modifies the member map. Before this was possible because the map was not a member object but rather a static object.

Jenkins please retry a build

This merge request affects 1 package:

• Simulation/G4Utilities/Geo2G4

Adding @jchapman ,@vpascuzz as watchers

added 21.0 Simulation labels

CI Result FAILURE

	Athena	AthDataQuality	AthSimulation
externals
cmake
make
required tests
optional tests

Full details available at NICOS MR-22123-2019-03-26-00-44
Athena: number of compilation errors 0, warnings 165
AthDataQuality: number of compilation errors 0, warnings 1
AthSimulation: number of compilation errors 0, warnings 6
For experts only: Jenkins output [CI-MERGE-REQUEST 35955]

added 1 commit

• 74f3b1f8 - better memory management in Geo2G4Svc

Compare with previous version

Jenkins please retry a build

This merge request affects 1 package:

• Simulation/G4Utilities/Geo2G4

Adding @jchapman ,@vpascuzz as watchers

CI Result SUCCESS

	Athena	AthDataQuality	AthSimulation
externals
cmake
make
required tests
optional tests

Full details available at NICOS MR-22123-2019-03-26-12-31
Athena: number of compilation errors 0, warnings 0
AthDataQuality: number of compilation errors 0, warnings 0
AthSimulation: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST 35982]

Thanks for this Miha. Can you test the changes out in master in particular in a G4MT job before dropping the WIP status please?

mentioned in merge request !22149 (merged)

Hi @jchapman , I made a manual sweep: !22149 (merged) , since there was a small conflict.

I ran test_AtlasG4_TTbarSim_MT.sh locally and here is the result:

PyJobTransforms.main 2019-03-26 15:29:09,712 INFO /cvmfs/atlas-nightlies.cern.ch/repo/sw/master/2019-03-24T2154/Athena/22.0.1/InstallArea/x86_64-slc6-gcc8-opt/bin/AtlasG4_tf.py stopped at Tue Mar 26 15:29:09 2019, trf exit code 0

unmarked as a Work In Progress

added sweep:ignore label

Jenkins please retry a build

This merge request affects 1 package:

• Simulation/G4Utilities/Geo2G4

Adding @jchapman ,@vpascuzz as watchers

added review-pending-level-1 label

CI Result SUCCESS

	Athena	AthDataQuality	AthSimulation
externals
cmake
make
required tests
optional tests

Full details available at NICOS MR-22123-2019-04-09-06-39
Athena: number of compilation errors 0, warnings 164
AthDataQuality: number of compilation errors 0, warnings 1
AthSimulation: number of compilation errors 0, warnings 6
For experts only: Jenkins output [CI-MERGE-REQUEST 36731]

Hi,

the changes look ok, but please consider updating the end date in the copyright statement for future MRs.

Cheers, Tomas (L1)

added review-approved label and removed review-pending-level-1 label

merged