Re: [ROOT] Memory

From: Rene Brun (Rene.Brun@cern.ch)
Date: Thu Jul 18 2002 - 08:52:53 MEST


I have no problems running your test with the file you sent to me.

When you have a problem like this one, run under gdb and send
the output of the traceback (gdb "bt" or "info stack" command).

Rene Brun

On Tue, 16 Jul 2002, Stanley Forrester wrote:

> 
> How does ROOT allocate memory?  I keep getting a  "*** Break *** 
> segmentation violation" message.  I've tried everything I can think of 
> within root and now suspect that there is a problem with how I have 
> root set up (for instance making some buffer larger).  I'm currently 
> running root from within a Gnome terminal on a Linux 7.0 box (fermilab 
> version of Linux).
> 
> I'm running the following script
> 
> void HistoComp() {
>   
>   // Debugging.
>   cout << "Made it to step 1 \n";
>   // Open .root file then change to appropriate directory 
>   // within .root file 
>   TFile *f = new TFile("/home/sforrest/dump/sum_hist_147724.root");
>   cout << "Made it to step 2 \n";
>   // Pull and draw the Reference Histograms
>   TH1F * hist1 = (TH1F*) 
> f.Get("TauTriggerValidation/CEM8Track5Validation_TAU_ELECTRON8_TRACK5_ISO_v_NPassingElectrons");
>   TH1F * hist2 = (TH1F*) 
> f.Get("TauTriggerValidation/CEM8Track5Validation_TAU_ELECTRON8_TRACK5_ISO_v_NPassingTracks");
>   hist1->Draw();  // Debugging:  Just to see if I actually have the right histo.
>   std::cout << "Made it to step 6 \n";
>   TH1F * hist3 = (TH1F*)  
> f.Get("TauTriggerValidation/CEM8Track5Validation_TAU_ELECTRON8_TRACK5_ISO_v_electronCESDz");
>  
> }
> 
> I get (Usually) an output of hist1 to the screen and the messages below on 
> the terminal screen
> 
> root [0] .L HistoCompairison3.C
> root [1] HistoComp()           
> Made it to step 1 
> Made it to step 2 
> <TCanvas::MakeDefCanvas>: created default TCanvas with name c1
> Made it to step 6 
> 
>  *** Break *** segmentation violation
> Root > Function HistoComp() busy flag cleared
> 
> 
> 
> 
> I'm currently running on linux 7.0 (fermilab version)
> 
> 



This archive was generated by hypermail 2b29 : Sat Jan 04 2003 - 23:51:00 MET