[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Help-gsl] gsl libraries by Brian Gladman

From: Brian Gladman
Subject: Re: [Help-gsl] gsl libraries by Brian Gladman
Date: Fri, 7 Sep 2012 15:01:51 +0100

There is now only one out of order declaration in the repository code as

ode-initval2\test.c (2185 ...) out of order declarations:
------------------------------------  /* Test driver_apply_fixed_step */

 gsl_odeiv2_driver_reset_hstart (d, 1e-3);
 const unsigned long int nfsteps = 100000;
 const double hfixed = 0.000025;
 s = gsl_odeiv2_driver_apply_fixed_step (d, &t, hfixed, nfsteps, y);

 if (s != GSL_SUCCESS)

Please consider dropping this into a bug report at

Thanks Rhys,

I will do this.

There are, however, quite a few test errors as follows:


I would be most grateful to know which of these are 'work in progress' and
which are unexpected and might hence be specific to a Visual Studio GSL

How many of the errors you report persist when you compile in debug
mode (i.e. -O0)?  I am aware of test tolerances being problematic on
Intel compilers at -O2 or -O3.  I suspect Microsoft compilers will
similarly perturb the numerical results when optimization is enabled.
Anything that breaks at -O0 is certainly suspect.


All the results i listed are without optimisation turned on - i.e in debug mode

Since a large number of the results are a very long way out, I was hoping that at least some might be 'work in progress' :-(

Thanks again for your comment.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]