Greetings! Yes, I have a branch doing this. It forces a bunch of other garbage in cmpinclude.h, which we try to keep minimal, but otherwise works. Right now I just have a configure probe set calculate the standard int with the same size as void*.
The real problem is gmp, the functions of which use long int. We've always set fixnum to long for this reason. So it appears we have to either build a gmp with a long long limb on win64, rewrite the gmp interface, or deal with 32bit fixnums on win64, which kind of defeats the whole point IMHO. I'll try to purse this on the gmp-devel mailing list. Take care, Gabriel Dos Reis <[email protected]> writes: > Or more portably, use intptr_t from <stdint.h> supported by all C compilers > and platforms. > The unsigned equivalent is uintptr_t. > > -- Gaby > > On Thu, Nov 20, 2014 at 8:08 AM, Camm Maguire <[email protected]> wrote: > > Greetings! And thanks again so much for your help. This appears as if > it will take a little time after all, as MS, alone among all other > common oses, defines 64bit long as a 32bit int. GCL code has long > equated pointer types with long integers. I'm hoping a simple typedef > at the top somewhere might take care of this. > > Take care, > -- > Camm Maguire [email protected] > ========================================================================== > "The earth is but one country, and mankind its citizens." -- Baha'u'llah > > _______________________________________________ > Gcl-devel mailing list > [email protected] > https://lists.gnu.org/mailman/listinfo/gcl-devel > -- Camm Maguire [email protected] ========================================================================== "The earth is but one country, and mankind its citizens." -- Baha'u'llah _______________________________________________ Gcl-devel mailing list [email protected] https://lists.gnu.org/mailman/listinfo/gcl-devel
