Thread

1. CVS compile on AIX 4.3.2

   Samuel A Horwitz <horwitz@argoscomp.com> — 2003-07-15T15:20:17Z

   I am getting the following error
   
   gmake[4]: Entering directory
   `/usr/local/postgres/pgsql/src/backend/access/commo
   n'
   gcc -O2 -pipe -Wall -Wmissing-prototypes -Wmissing-declarations
   -I../../../../sr
   c/include -I/usr/local/ssl/include  -c -o printtup.o printtup.c
   In file included from ../../../../src/include/libpq/libpq-be.h:22,
                    from ../../../../src/include/libpq/libpq.h:21,
                    from printtup.c:20:
   ../../../../src/include/libpq/pqcomm.h:60: `int64_t' undeclared here (not
   in a f
   unction)
   ../../../../src/include/libpq/pqcomm.h:64: parse error before "int64_t"
   ../../../../src/include/libpq/pqcomm.h:64: warning: no semicolon at end of
   struc
   t or union
   ../../../../src/include/libpq/pqcomm.h:67: `int64_t' undeclared here (not
   in a f
   unction)
   ../../../../src/include/libpq/pqcomm.h:67: `int64_t' undeclared here (not
   in a f
   unction)
   ../../../../src/include/libpq/pqcomm.h:71: parse error before '}' token
   ../../../../src/include/libpq/pqcomm.h:81: field `addr' has incomplete
   type
   gmake[4]: *** [printtup.o] Error 1
   gmake[4]: Leaving directory
   `/usr/local/postgres/pgsql/src/backend/access/common
   '
   gmake[3]: *** [common-recursive] Error 2
   gmake[3]: Leaving directory `/usr/local/postgres/pgsql/src/backend/access'
   gmake[2]: *** [access-recursive] Error 2
   gmake[2]: Leaving directory `/usr/local/postgres/pgsql/src/backend'
   
   
   horwitz@argoscomp.com (Samuel A Horwitz)
   
   
   
   

2. Re: CVS compile on AIX 4.3.2

   Tom Lane <tgl@sss.pgh.pa.us> — 2003-07-15T16:49:11Z

   Samuel A Horwitz <horwitz@argoscomp.com> writes:
   > I am getting the following error
   > ../../../../src/include/libpq/pqcomm.h:60: `int64_t' undeclared here (not
   > in a function)
   
   There's a patch floating around the list to remove use of int64_t, which
   is evidently not too portable.  I'll try to get it applied soon.
   
   			regards, tom lane
   
   

3. Re: CVS compile on AIX 4.3.2

   Sean Chittenden <sean@chittenden.org> — 2003-07-15T17:55:11Z

   > > I am getting the following error
   > > ../../../../src/include/libpq/pqcomm.h:60: `int64_t' undeclared here (not
   > > in a function)
   > 
   > There's a patch floating around the list to remove use of int64_t, which
   > is evidently not too portable.  I'll try to get it applied soon.
   
   int64_t is a C99 data type that hasn't existed prior to recent
   versions of gcc, but is quite valid/correct.  I'd think that int64_t
   support should be fudged on platforms where in64_t isn't defined as
   long long.  IIRC, this is what FreeBSD has done for ages to get around
   the lack of a 64bit type.
   
   /* From machine/_types.h */
   #if defined(lint)
   /* LONGLONG */
   typedef long long               __int64_t;
   /* LONGLONG */
   typedef unsigned long long      __uint64_t;
   #elif defined(__GNUC__)
   typedef int __attribute__((__mode__(__DI__)))           __int64_t;
   typedef unsigned int __attribute__((__mode__(__DI__)))  __uint64_t;
   #else
   /* LONGLONG */
   typedef long long               __int64_t;
   /* LONGLONG */
   typedef unsigned long long      __uint64_t;
   #endif
   
   -sc
   
   -- 
   Sean Chittenden
   
   

4. Re: CVS compile on AIX 4.3.2

   Tom Lane <tgl@sss.pgh.pa.us> — 2003-07-15T17:59:34Z

   Sean Chittenden <sean@chittenden.org> writes:
   > int64_t is a C99 data type that hasn't existed prior to recent
   > versions of gcc, but is quite valid/correct.  I'd think that int64_t
   > support should be fudged on platforms where in64_t isn't defined as
   > long long.
   
   We have int64 defined (and well tested) already; I see no reason to muck
   with it.
   
   			regards, tom lane
   
   

5. Re: CVS compile on AIX 4.3.2

   Stephan Szabo <sszabo@megazone.bigpanda.com> — 2003-07-15T18:02:31Z

   On Tue, 15 Jul 2003, Sean Chittenden wrote:
   
   > > > I am getting the following error
   > > > ../../../../src/include/libpq/pqcomm.h:60: `int64_t' undeclared here (not
   > > > in a function)
   > >
   > > There's a patch floating around the list to remove use of int64_t, which
   > > is evidently not too portable.  I'll try to get it applied soon.
   >
   > int64_t is a C99 data type that hasn't existed prior to recent
   > versions of gcc, but is quite valid/correct.  I'd think that int64_t
   
   It's also not guaranteed to exist on any given machine if there isn't a
   native exactly 64 bit integer type.  I'd think that if you're going to use
   one of the C99 integer types, int_least64_t or int_fast64_t would be a
   better choice unless you really must guarantee 64 bits since those appear
   to be required.