[savannah-help-public] [sr #108923] Code working with GSL 1.16 does not work with GSL 2.0

[Cyril]

URL:
  <http://savannah.gnu.org/support/?108923>

                 Summary: Code working with GSL 1.16 does not work with GSL
2.0
                 Project: Savannah Administration
            Submitted by: lock042
            Submitted on: lun. 09 nov. 2015 15:31:31 GMT
                Category: None
                Priority: 5 - Normal
                Severity: 3 - Normal
                  Status: None
             Assigned to: None
        Originator Email: address@hidden
        Operating System: None
             Open/Closed: Open
         Discussion Lock: Any

    _______________________________________________________

Details:

Hello, I have a problem with my code.
Indeed, it is working very well with gs 1.16 but since I've tried with the new
release, I have no warnings, no errors at compilation but the application
crashes.

My code (you can read all the code at:
https://free-astro.vinvin.tf/svn/siril/trunk/src/gradient.c (sorry for the bad
certificat) is something like that:
        J = gsl_matrix_alloc (n, VarParam);
        y = gsl_vector_alloc (n);
        w = gsl_vector_alloc (n);
        c = gsl_vector_alloc (VarParam);
        cov = gsl_matrix_alloc (VarParam, VarParam);
[...]
        gsl_multifit_linear_workspace * work = gsl_multifit_linear_alloc (n,
VarParam);
        gsl_multifit_wlinear (J, w, y, c, cov, &chisq, work);
        gsl_multifit_linear_free (work);
[...]

When I run the code I have:
gsl: multiwlinear.c:71: ERROR: size of workspace does not match size of
observation matrix
Default GSL error handler invoked.

So, gdb shows that the bug is on the line 204, i.e,
gsl_multifit_wlinear (J, w, y, c, cov, &chisq, work);

I have no clue to explain where the problem come from. Could help me please ?

Best regards,





    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/support/?108923>

_______________________________________________
  Message posté via/par Savannah
  http://savannah.gnu.org/