Dan Riley

> Hi @dan131riley , I noted a detail in the ZS unpacking in your cuda core, which I would like to have clarified. > > In the unpacked Digi, there...

There are some [UBSAN hits](https://cmssdt.cern.ch/SDT/cgi-bin/logreader/el8_amd64_gcc13/CMSSW_16_0_UBSAN_X_2025-12-01-2300/pyRelValMatrixLogs/run/30834.0_TTbar_14TeV+Run4D113/step2_TTbar_14TeV+Run4D113.log#/76-76) in that neighborhood, of the form ``` src/L1Trigger/TrackFindingTracklet/interface/TripletEngineUnit.h:27:9: runtime error: load of value 224, which is not a valid value for type 'bool' #0 0x14c29798d48e...

There are a couple of exception messages just before the segfault, but it isn't clear if they're associated with the crash: ``` %MSG-e TrackProducer: TrackRefitter:refittedForPixelDQM 09-Dec-2025 06:37:10 CET Run: 1...

Run it under gdb. When it gets stuck, interrupt with a control-c and do "thread apply all bt".

If the job finished, I'd expect to see an excessive time warning saying that a module took too long (if that's enabled). Also, if you're running these jobs multi-threaded, try...

* Thread 1, 28 and 29 are waiting on the MemoryPool lock. * Thread 27 is holding the MemoryPool lock and waiting on the device lock. * 2, 4, and...