Crash on arthur

[padsley]

Crash with 2 ADC modules.

This is for my own notes. Go away.

The lines below might hint at the cause of the crash. If they do not help you then please submit a bug report at http://root.cern.ch/bugs. Please post the ENTIRE stack trace from above as an attachment in addition to anything else

that might help us fixing this issue.

6 0x000000000042b4f6 in CloverEnergyCalc (Channel=Channel

entry=22, ADCValue=167) at CloverSort.c:153

7 0x000000000042b7e4 in CloverSort (ADC_import=0x2ac61d0, ntdc=ntdc

entry=27, TDC_channel_import=TDC_channel_import entry=0x3339670, TDC_value_import=TDC_value_import entry=0x3339af0, gammy=0x3339b70) at CloverSort.c:60

8 0x0000000000433d9d in main_event (pheader=, pevent=) at main.c:1884

9 0x000000000041828b in process_event (par=0x68b980 , pevent=0x7fb38df05010) at src/mana.c:3157

10 0x0000000000418822 in analyze_run (run_number=1083, input_file_name=0x7ffcd4a11d70 "/experiment-data/PR251/2016/run01083.mid.gz", output_file_name=) at src/mana.c:4178

11 0x0000000000418d52 in loop_runs_offline () at src/mana.c:4366

12 0x0000000000419585 in main (argc=, argv=) at src/mana.c:5579

[padsley]

The problem is in 'ADCCalibrationParameters' somewhere...

[padsley]

Oh!

Found it.

  for(int i=1;i<npars+1;i++)
  {
    result += ADCCalibrationParameters[Channel][i] * pow(RandyADCValue,(double)i-1.);
  }

The problem happens in here.

[padsley]

Ignore that last message. Was nonsense. It failed there because I'd stopped it failing earlier. Doh.

[padsley]

Fixed.

Problem was in CloverSort.c.

In Parameters.c, the declaration is: double **ADCCalibrationParameters;.

There is, on the line above, an outdated declaration: // std::vector<std::vector<double> > ADCCalibrationParameters;.

This is left over from my first attempt to add non-linear calibrations.

In CloverSort.c, the declaration was //extern std::vector<std::vector<double> > ADCCalibrationParameters;.

For some reason, this allowed the compilation. Je ne comprends pas. Anyway, it caused the array behaviour to be different (in some manner which I do not understand) which caused the code to crash.

Obviously, the problem was in the memory allocation of ADCCalibrationParameters. Hence why the problem only occurred when people attempted to run with only 2 ADC modules and not with a higher number.

[padsley]

I just checked, the only other place with a non-linear calibration implemented is in RawData.c.

Here, the declaration looks fine. C'est bon.

[padsley]

Definitely my fault: https://github.com/padsley/k600analyser/commit/01ead185d645576bc1d468d43232aa5f460ccfa3