On Thu, 20 Jan 2005, Wayne Davison <[EMAIL PROTECTED]> wrote:
> On Thu, Jan 20, 2005 at 01:54:20AM -0600, John Van Essen wrote:
>> (Note to Wayne - isn't it dangerous to assume that off_t is 64 bits?
>
> Yeah, it probably is -- that code was always a kluge that doesn't really
> get used. I looked around at the rest of the rsync code, and it looks
> to me like the various uses of uint64 could each be replaced by int64.
I agree, provided that int64 is really 64 bits. If uint64 is less
than 64 bits, and it is absolutely vital, shouldn't the compiler abort
with a meaningful error message in the #else of the #if sequence?
#else
#error "Unable to define an unsigned 64-bit integer type"
#endif
> The root problem is that rsync's configure script uses AC_TRY_RUN in
> bogus ways -- i.e. it totally punts a lot of checks when it is cross-
> compiling, so it doesn't even check if "long long" works. It looks
> like configure.in is really in need of a bunch of cleanup in that
> regard.
At least for "long long" for now to get 64 bits better implemented
for cross-compilers in 2.6.4.
--
John Van Essen Univ of MN Alumnus <[EMAIL PROTECTED]>
--
To unsubscribe or change options: https://lists.samba.org/mailman/listinfo/rsync
Before posting, read: http://www.catb.org/~esr/faqs/smart-questions.html
