Re: Does this break binary compatability on 64bit architectures?
On Thu, 2005-02-24 at 20:52 +0100, Kurt Roeckx wrote:
> On Thu, Feb 24, 2005 at 05:53:02PM +0300, Nikita V. Youshchenko wrote:
> > Hello.
> >
> > Upstream of a library package that I maintain changed function prototypes
> > in the followinf way:
> >
> > >
> > > -int mailpop3_retr(mailpop3 * f, uint32_t index, char ** result,
> > > +int mailpop3_retr(mailpop3 * f, unsigned int index, char ** result,
> > > size_t * result_len);
> >
> > That is, 'uint32_t' was changed to 'unsigned int'.
> >
> > Does this break binary compatability on any of debian architectures (so
> > soname change is needed)?
>
> Afaik, on all 64 bit arches in debian an int is still 32 bit.
Just out of curiosity, why would upstream change a library fun from
uint32_t, which will *always* be 32 bits, to int, which may or may
not be 32 bits, depending on the arch and compiler?
--
-----------------------------------------------------------------
Ron Johnson, Jr.
Jefferson, LA USA
PGP Key ID 8834C06B I prefer encrypted mail.
$ python -c 'print len(str(2**3000000))'
903090
Reply to: