Re: 3.3.4 build report: powerpc-ibm-aix4.3.3.0

[Gabriel Dos Reis <gdr at integrable-solutions dot net>]

Joe Buck <Joe.Buck@synopsys.COM> writes:

| I built 3.3.4 for AIX 4.3.3 (powerpc-ibm-aix4.3.3.0), using
| the vendor cc, as, and ld.  Results are at
| http://gcc.gnu.org/ml/gcc-testresults/2004-06/msg00101.html .

Thanks for the report.

| Java doesn't build on this platform.
| 
| I had to use --disable-nls to get a successful build.  Without this flag,
| I got the following failure in the gcc/intl directory:
| 
| cc -c -DIN_GCC -DLOCALEDIR=\"/u/jbuck/cvs.aix/3.3.4-pre/share/locale\" -DLOCALE_ALIAS_PATH=\"/u/jbuck/cvs.aix/3.3.4-pre/share/locale\" -DLIBDIR=\"/u/jbuck/cvs.aix/3.3.4-pre/lib\" -DHAVE_CONFIG_H -I.. -I. -I/remote/dtg103/jbuck/gnu/src/gcc-3.3.4-20040530/gcc/intl -I/remote/dtg103/jbuck/gnu/src/gcc-3.3.4-20040530/gcc -I/remote/dtg103/jbuck/gnu/src/gcc-3.3.4-20040530/gcc/config -I/remote/dtg103/jbuck/gnu/src/gcc-3.3.4-20040530/gcc/../include  -g    /remote/dtg103/jbuck/gnu/src/gcc-3.3.4-20040530/gcc/intl/plural.c -o plural.o
| "plural.y", line 264.1: 1506-343 (S) Redeclaration of __gettextlex differs from previous declaration on line 73 of "plural.y".
| "plural.y", line 264.1: 1506-381 (I) The type "const unsigned char**" of parameter 2 in the prototype declaration is not compatible with the corresponding parameter type "unsigned char**" in the nonprototype declaration.
| 

Interesting.  From intl/ChangeLog, files in gcc/intl have not been
changed since 2003-05-13, the release of 3.3.0.  The last change that
touched gcc/intl/pluark.y was 

  2001-11-03  Kaveh R. Ghazi  <ghazi@caip.rutgers.edu>

          * l10nflist.c (_nl_normalize_codeset): Cast argument of ctype
          function to `unsigned char'.
          * loadmsgcat.c (_nl_load_domain): Likewise.
          * localealias.c (read_alias_file, alias_compare): Likewise.
          * localcharset.c: Include "gettextP.h".
          * plural.y (__attribute__): Define.
          (yyerror): Mark parameter with __attribute__ ((__unused__)).
          * plural.c: Regenerate.

but I don't see anything obvious there that would have introduced the
problem.  I'm more worried about why GCC itself did not catch the
problem in the first place.

-- Gaby