On 2006-04-16 13:39:49 -0400, Steve M. Robbins wrote: > For the record: a simple rebuild allowed my CGAL programs > to work again.
Yes, we had a bug report about the same problem, but not with much details. I already got rid of mpn_sub_nc in the trunk (fixing a potential bug at the same time); the patch will be applied later to the branch. Concerning the __gmp_rand, a patch will probably modify the PRNG. I wonder if this is acceptable (however we don't say anything about the behavior of the PRNG). BTW, I think the use of GMP internals mainly comes from the time MPFR was more or less part of GMP. -- Vincent Lefèvre <[EMAIL PROTECTED]> - Web: <http://www.vinc17.org/> 100% accessible validated (X)HTML - Blog: <http://www.vinc17.org/blog/> Work: CR INRIA - computer arithmetic / SPACES project at LORIA
