GSI Forum: Analysis » Different results for same information extracted in different ways

Home » PANDA » PandaRoot » Analysis » Different results for same information extracted in different ways

Show: Today's Messages :: Polls :: Message Navigator

Re: Different results for same information extracted in different ways [message #17493 is a reply to message #17489]

Mon, 10 November 2014 17:41

Stefan Pflueger
Messages: 99
Registered: February 2012

continuous participant

From: *kph.uni-mainz.de

Hi Mamen,

so when I run your macro in the compiled mode (.C+) it runs fine (sometimes I get errors but all the way at the end, when everything is clean up). From what I can see in your Output file, your crashes appear also all the way at the end so all results etc should be fine. The segfault is annoying nevertheless... I don't see why it crashes at this point, but since it is all the way at the end it has to do with cleanup of your objects (destructors are called). I would guess that somehow the memory management of root screws up there.
When I run the macros in interpreter mode (.C) it runs fine on some files and crashes on other files. In my case it crashes when filling the Reco2 histogram at some defined entry (not the last) of the tree. Debugging this nearly impossible (afaik), as more or less non logical stuff happens. However I traced it back to the #include statements. So removing the them should make your macro run just just like in the compiled mode (crashes at the end of the macro when cleanup happens, but that is no biggie). Hope that helps...

Small remark: I would recommend to replace the char array and sprint statements with stringstreams as they are a bit safer, so something like this:

  int W2=10;
  std::string FWBW="bw";
  int NBINS=100;
  int BINMIN=-2;
  int BINMAX=2;

  std::stringstream ss;
  ss<<"epempi0-W2-"<<W2<<"-Delta0-"<<FWBW<<"-LargeQ2-Merged.root";

Cheers,

Stefan

Report message to a moderator

[Message index]

		Different results for same information extracted in different ways By: Mamen on Fri, 07 November 2014 13:42
		Re: Different results for same information extracted in different ways By: Klaus Götzen on Fri, 07 November 2014 13:55
		Re: Different results for same information extracted in different ways By: Mamen on Fri, 07 November 2014 14:59
		Re: Different results for same information extracted in different ways By: Klaus Götzen on Fri, 07 November 2014 17:14
		Re: Different results for same information extracted in different ways By: Mamen on Fri, 07 November 2014 17:26
		Re: Different results for same information extracted in different ways By: Mamen on Mon, 10 November 2014 15:30
		Re: Different results for same information extracted in different ways By: Stefan Pflueger on Mon, 10 November 2014 17:41
		Re: Different results for same information extracted in different ways By: Mamen on Mon, 10 November 2014 17:54
		Re: Different results for same information extracted in different ways By: Mamen on Mon, 10 November 2014 16:41
		Re: Different results for same information extracted in different ways By: Klaus Götzen on Mon, 10 November 2014 17:13
		Re: Different results for same information extracted in different ways By: Mamen on Mon, 10 November 2014 17:51

Previous Topic:	vertex reconstruction without fitting?
Next Topic:	Vertex fitter for two consecutive decays

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Mon Sep 09 19:58:15 CEST 2024

Total time taken to generate the page: 0.00847 seconds