Hi all, I've managed to fix the bug that was in the first beta release. It turned out to be merely a problem with two prototypes in gmp.h/mpir.h. It only occurred in the alpha and beta releases of mpir-2.6.0 and as it only involved internal functions, probably didn't affect any projects except flint, which happens to use them.
I've uploaded a new MPIR-2.6.0 beta which fixes this problem. The latest git repository version of flint2 now passes its test suite when built against MPIR-2.6.0. Bill. On 27 October 2012 14:29, Bill Hart <goodwillh...@googlemail.com> wrote: > Hi all, > > We are finding a test failure in flint (fmpz_poly_compose) which > appears only with MPIR 2.6.0 and not with MPIR-2.4.x or below. > > We are looking into finding the bug and will hopefully find it soon, > but at this stage the MPIR-2.6.0 beta should be considered suspect. > > There are also a couple of minor bugs in gmp-h.in. > > The lines: > > #define mpn_mulmod_Bexpp1 __MPN(mulmod_2expp1) > __GMP_DECLSPEC int mpn_mulmod_Bexpp1 __GMP_PROTO ((mp_ptr, mp_srcptr, > mp_srcptr, mp_size_t, mp_ptr)); > > should be replaced with: > > #define mpn_mulmod_Bexpp1 __MPN(mulmod_Bexpp1) > __GMP_DECLSPEC int mpn_mulmod_Bexpp1 __GMP_PROTO ((mp_ptr, mp_srcptr, > mp_srcptr, mp_size_t, mp_ptr)); > > #define mpn_mulmod_2expp1 __MPN(mulmod_2expp1) > __GMP_DECLSPEC int mpn_mulmod_2expp1 __GMP_PROTO ((mp_ptr, mp_srcptr, > mp_srcptr,int,unsigned long, mp_ptr)); > > I will issue a new beta once we have sorted out the bugs. > > Bill. -- You received this message because you are subscribed to the Google Groups "mpir-devel" group. To post to this group, send email to mpir-devel@googlegroups.com. To unsubscribe from this group, send email to mpir-devel+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/mpir-devel?hl=en.
