[Maxima-discuss] invert with GCL segfaulting (was: ratsimp not working as expected)

[andre m. <and...@gm...>]

On 03/08/2014 04:23 PM, Thomas D. Dean wrote:
> On 03/08/14 00:51, andre maute wrote:
>
> Have you tried building gcl and maxima exactly as I did?  Only configure
> for one lisp and only have one lisp installed?  And, only one maxima.
I have now tried to install maxima 5.25.5, 5.28.0, 5.31.3, 5.32.1
with the lisps combinations (sbcl,gcl,clisp,sbcl and gcl and clisp)
All lisps are provided by Fedora, I use Fedora 20.

All versions get installed in their seperate local directory.
I always create a symlink called maxima for the directory I would like 
to use.

The following segfaults for maxima-5.32.1 installed only with GCL.

------- invert-bug1-gcl.max -----------
display2d : false;
A : matrix([a11,a12],[a21,a22]);
Ainv : invert(A);
------- invert-bug1-gcl.max -----------

Though Robert Dodier gave the workaround to use
load("invert")
which indeed works.

Regards Andre

> Sorry, I lost track.  What OS are you running?  Is it possible you have
> more than one version of the various pieces installed?
>
> Tom Dean