On Nov 28, 2007 5:45 AM, mabshoff
<[EMAIL PROTECTED]> wrote:
>
>
>
> On Nov 28, 2:34 pm, "William Stein" <[EMAIL PROTECTED]> wrote:
> > On Nov 27, 2007 4:21 PM, mabshoff
> >
> >
> >
>
> > <[EMAIL PROTECTED]> wrote:
> >
> > > Hello folks,
> >
> > > somewhere since 2.8.3 we introduced a new bug that seems to affect
> > > computing the power of matrices when the entries are multivariate
> > > polynomials with rational coefficients:
> >
> > > [EMAIL PROTECTED] sage-2.8.14$ ./sage
> > > ----------------------------------------------------------------------
> > > | SAGE Version 2.8.14, Release Date: 2007-11-24                      |
> > > | Type notebook() for the GUI, and license() for information.        |
> > > ----------------------------------------------------------------------
> >
> > > sage: R.<x,y>=MPolynomialRing(QQ,2)
> > > sage: A = matrix([[Integer(1),x], [y,Integer(1)]])
> > > sage: A
> >
> > > [1 x]
> > > [y 1]
> > > sage: A*A
> >
> > > [x*y + 1     2*x]
> > > [    2*y x*y + 1]
> > > sage: A**2
> >
> > > ------------------------------------------------------------
> > > Unhandled SIGSEGV: A segmentation fault occured in SAGE.
> > > This probably occured because a *compiled* component
> > > of SAGE has a bug in it (typically accessing invalid memory)
> > > or is not properly wrapped with _sig_on, _sig_off.
> > > You might want to run SAGE under gdb with 'sage -gdb' to debug this.
> > > SAGE will now terminate (sorry).
> > > ------------------------------------------------------------
> >
> > I cannot replicate this on either sage.math or osx 10.5 intel:
> >
> > [EMAIL PROTECTED]:~$ sage
> > ----------------------------------------------------------------------
> > | SAGE Version 2.8.14, Release Date: 2007-11-24                      |
> > | Type notebook() for the GUI, and license() for information.        |
> > ----------------------------------------------------------------------
> >
> > sage: sage: R.<x,y>=MPolynomialRing(QQ,2)
> > sage: sage: A = matrix([[Integer(1),x], [y,Integer(1)]])
> > sage: A
> >
> > [1 x]
> > [y 1]
> > sage: A*A
> >
> > [x*y + 1     2*x]
> > [    2*y x*y + 1]
> > sage: A**2
> >
> > [x*y + 1     2*x]
> > [    2*y x*y + 1]
> >
> > keyah:~ was$ sage
> > ----------------------------------------------------------------------
> > | SAGE Version 2.8.14, Release Date: 2007-11-24                      |
> > | Type notebook() for the GUI, and license() for information.        |
> > ----------------------------------------------------------------------
> >
> > sage: sage: R.<x,y>=MPolynomialRing(QQ,2)
> > sage: sage: A = matrix([[Integer(1),x], [y,Integer(1)]])
> > sage: A^3
> >
> > [  3*x*y + 1 x^2*y + 3*x]
> > [x*y^2 + 3*y   3*x*y + 1]
> > sage: A^2
> >
> > [x*y + 1     2*x]
> > [    2*y x*y + 1]
> > sage: A
> >
> > [1 x]
> > [y 1]
> > sage: A**2
> >
> > [x*y + 1     2*x]
> > [    2*y x*y + 1]
> >
> > -----
> >
> > Your subject line suggests maybe this is a solaris-only issue?    Is
> > that the case?
>
> Yep.
>
> > If so, what does "sage -gdb" have to say about it?
> >
>
> I get a pretty bt that point to __cmp__ in libSingular. Notice that it
> depends on the coefficient ring, i.e. mv polynomials over ZZ are fine,
> QQ not so much.

That's probably because polynomials over ZZ are still unfortunately *not*
implemented using libsingular (which is why they are amusing much
slower in Sage).

> I am currently building 2.8.14 on Linux PPC under
> Linux because I suspect endianess issues and I have valgrind there
> also to investigate.


The problem doesn't occur on OSX powerpc (fermat.math.harvard.edu), which
is maybe an endian data point.

> I am doing "real" work right now, but I should be done with that in
> 8-10 hours ;). Once I am done I will get back to Sage and open ticket
> for all the issues I have uncovered during the port. I have 2.8.14 on
> neron in /tmp. You need to set LD_LIBRRAY_PATH to my 4.2.1-2 gcc to
> play with it.
>
> I also created http://wiki.sagemath.org/solaris to keep track of the
> port.

Excellent.

>
> > William
>
>
> Cheers,
>
> Michael
> >
>



-- 
William Stein
Associate Professor of Mathematics
University of Washington
http://wstein.org

--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~----------~----~----~----~------~----~------~--~---

Reply via email to