Igor Schein on Fri, 14 Jun 2002 10:51:26 -0400


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

Re: nffactor() regression


On Fri, Jun 14, 2002 at 03:41:23AM +0200, Karim BELABAS wrote:
> On Thu, 13 Jun 2002, Igor Schein wrote:
> > the command below enters an infinite loop in 2.2.4 on Alpha Linux.  It worked
> > fine in 2.2.3
> [...]
> 
> I and Xavier Roblot have been refurbishing my relative variant of van Hoeij's
> method (to factor polynomials over number fields). We had not yet tested it
> on 64 bit machines yet !
> 
> It broke in a trivial place. It is fixed (typo in nf_factor_bound, due to
> 3 == DEFAULTPREC on 64bit).

It's still not fixed yet, it goes further, but for the last hour I'm seeing
these messages at \g5:

LLL_cmbf: b = 251, r = 22
K2 K3 K4 K5 K6 K7 K8 K9 K10 K11 K12 K13 K14 K15 K16 K17 K18 K19 K20 K21 K22
LLL_cmbf: b = 251, r = 22
K2 K3 K4 K5 K6 K7 K8 K9 K10 K11 K12 K13 K14 K15 K16 K17 K18 K19 K20 K21 K22

Thanks

Igor