Re: [Maxima-bugs] [ maxima-Bugs-988701 ] incorrect matrix inversion



On Wed, 2004-07-14 at 11:16, Raymond Toy wrote:
> >>>>> "Camm" == Camm Maguire <camm@enhanced.com> writes:
> 
>     Camm> Greetings!  I think the problem is here (opers.lisp):
>     Camm> (defmfun mul2* (x y)
>   
>     Camm>   (cond
>     Camm>     #+cl ((and (numberp x) (numberp y)) (f* x y))
>     Camm>     ((=1 x) (simplifya y nil))
>     Camm> 	((=1 y) (simplifya x nil))
>     Camm> 	(t (simplifya `((mtimes) ,x ,y) nil))))
> 
> FWIW, I replaced (f* x y) with (* x y) here, and recompiled maxima.
> invert now returns the correct answer.

OK. Thanks to everyone's help, I think we now understand:

	1) What changed between 5.9.0 and 5.9.1.
	2) How the problem can be avoided.

I don't completely understand why the problem only appears in compiled
GCL code, but I am long past caring.

Please commit the fix.

--Jim