On Aug 3, 1:27 am, "William Stein" <[EMAIL PROTECTED]> wrote:
> Hi Clement,
>
> Any chance you could look into this. It's an 8x8 full matrix where
> linbox (via SAGE)
> computes the wrong determinant?
>
> Here's the SAGE session that gives the bad result:
>
> sage: M = matrix(
> [
> [-3821257660, -3821257669, -1736935303, -2779096486, -1736935306,
> -2779096486, -2779096489, -2779096483],
> [-3821257669, -3821257660, -1736935303, -2779096486, -1736935303,
> -2779096489, -2779096486, -2779096489],
> [-1736935303, -1736935303, -789516040, -1263225676, -789516049,
> -1263225676, -1263225679, -1263225676],
> [-2779096486, -2779096486, -1263225676, -2021161072, -1263225676,
> -2021161081, -2021161081, -2021161084],
> [-1736935306, -1736935303, -789516049, -1263225676, -789516040,
> -1263225676, -1263225676, -1263225676],
> [-2779096486, -2779096489, -1263225676, -2021161081, -1263225676,
> -2021161072, -2021161081, -2021161081],
> [-2779096489, -2779096486, -1263225679, -2021161081, -1263225676,
> -2021161081, -2021161072, -2021161081],
> [-2779096483, -2779096489, -1263225676, -2021161084, -1263225676,
> -2021161081, -2021161081,-2021161072] ])
> sage: M._det_linbox()
> ERROR in reconstruction ?
> 0
>
> The C++ code against linbox that's used for _det_linbox is:
>
> void linbox_integer_dense_det(mpz_t ans, mpz_t** matrix, size_t nrows,
> size_t ncols) {
> commentator.setMaxDetailLevel(0);
> commentator.setMaxDepth (0);
>
> DenseMatrix<Integers> A(new_matrix_integers(matrix, nrows, ncols));
> GMP_Integers::Element d;
> det(d, A);
> mpz_set(ans, spy.get_mpz(d));
>
> }
>
> Thoughts?
I think it is the same problem as http://linalg.org/projects/linalg/ticket/3
- I found similiar issues and tracked it down to an access violation
in the FFLAS code, but I haven't really debugged this.
Cheers,
Michael
>
> --
> William Stein
> Associate Professor of Mathematics
> University of Washingtonhttp://www.williamstein.org
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
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/
-~----------~----~----~----~------~----~------~--~---
