Not sure this is a memory leak, but i have my doubts that this is an intentional feature.
All the observations that follow refer to code freshly pulled from the trunk.
When running the simulation (without any digitization), the json output file gets a new line for each spill we run. The number of hits in the "mc" branch of each new line grows with each new spill. Shouldn't there be a reset? If we try to run a large number of events, the memory consumption becomes unbearable at some point. (Had to restart my laptop when trying to run 1000 spills).
- Tracker changed from Feature to Bug
Are you using the trunk version? I think this is an old bug:
bzr pull lp:maus
bzr merge lp:maus
Yes, I'm using the trunk version.
I first found this feature on my own version and then verified it in code just pulled from the trunk.
ClearSDHits() within MICEDetectorConstruction isn't being called anymore...
- Target version set to Future MAUS release
Ack - sorry... not in a test so I broke it when I updated the code. Will implement fix plus test ASAP.
Fixed and tested in rogers
- Status changed from Open to Closed
- % Done changed from 0 to 100
- Target version changed from Future MAUS release to MAUS-v0.0.4
Also available in: Atom