Re: [ROOT] Bug with fitting....

From: Rene Brun (Rene.Brun@cern.ch)
Date: Tue Jan 21 2003 - 22:41:44 MET


Hi Ray,

It is possible to do this, but you must use the right cast.
see your interactive fit routine slightly modified in the 
attachement.

Rene Brun

On 
Tue, 21 Jan 
2003, Ray Fliller III wrote:

> 
> 
>    Hello rooters,
> 
>     I am using ROOT version 3.05 CVSed yesterday on LINUX. 
>  I am tryingto do a fit to a graph with a function other than
> GraphFitChisquare().  Attached is a script that will work when ACLiC
> compiled (note that GraphFitChisquareNoEX is just GraphFitChisquare for
> debugging).  But if GraphFitChisquareNoEX is compiled and testfit is in
> interpreted mode, I get a SegV.  I can trace it to the line  
> 
>    fu   = grF1->EvalPar(x,u);
> 
>  and it seems that the segV is called because of u.  If I do a
> cout<<u[0]<<endl; in the function, I get a segV at that point.
> 
> 





This archive was generated by hypermail 2b29 : Thu Jan 01 2004 - 17:50:08 MET