On 2014-01-03 12:19:26 +0100, Marc Glisse wrote: > On Fri, 3 Jan 2014, Torbjorn Granlund wrote: > > >Vincent Lefevre <vinc...@vinc17.net> writes: > > > > Note that if you want to care of non-two's-complement > > implementations, you should write: > > Isn't the notion of 2's complement restricted to signed types?
Yes, but in ~0, 0 is of signed type. In non-two's-complement implementations, 0 may have several representations, so that ~0 is not guaranteed to work. > > #define __GMP_USHRT_MAX (0 + (unsigned short) -1) > > The previous form has worked so far, and I'd like to remove the macro > completely soon, so it doesn't seem worth it. > > >Is that out-of-range conversion really Proper C? > > unsigned = modular arithmetic, so yes, it is fine. Anyway if -1 didn't work, then ~0 wouldn't have worked either, because the result of ~0 is -1. Not to be confused with ~0U or ~ (unsigned int) 0. -- Vincent Lefèvre <vinc...@vinc17.net> - Web: <http://www.vinc17.net/> 100% accessible validated (X)HTML - Blog: <http://www.vinc17.net/blog/> Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon) _______________________________________________ gmp-devel mailing list gmp-devel@gmplib.org https://gmplib.org/mailman/listinfo/gmp-devel
