<<< Date Index >>>     <<< Thread Index >>>

Re: 1.5.8 build failure on Solaris 8



On Friday, 18 February 2005 at 18:25, Lars Hecking wrote:
> 
> > > #define _CS_XBS5_LP64_OFF64_CFLAGS 709 
> > > #define _CS_XBS5_LP64_OFF64_LIBS 711 
> > > #define _SC_XBS5_LP64_OFF64 722 
> > > #define _CS_XBS5_LP64_OFF64_LDFLAGS 710 
> > > #define _CS_XBS5_LP64_OFF64_LINTFLAGS 712 
> > > #define IOC_LP64 DATAMODEL_LP64 
> > > #define _FILE_OFFSET_BITS 64 
> > 
> > Hmm. I just tried this on my brother's Solaris 8 sparc32 system, with
> > gcc 3.4.2, and got:
> > 
> > #define _CS_XBS5_LP64_OFF64_CFLAGS 709
> > #define _CS_XBS5_LP64_OFF64_LDFLAGS 710
> > #define IOC_LP64 DATAMODEL_LP64
> > #define _CS_XBS5_LP64_OFF64_LIBS 711
> > #define _CS_XBS5_LP64_OFF64_LINTFLAGS 712
> > #define _SC_XBS5_LP64_OFF64 722
> > #define __PRAGMA_REDEFINE_EXTNAME 1
> > #define _FILE_OFFSET_BITS 64
> > 
> > and it compiled without problems. So I guess the problem is that
> > __PRAGMA_REDEFINE_EXTNAME isn't getting defined. I wonder what
> > triggers this difference?
> 
>  No idea. In any case, I am not in a position to change anything about the
>  OS rev or patch level on these machines here.

yes. I was more wondering if there were a simple define that could
fix it, rather than a large search-and-replace. For one thing, no
search-and-replace on mutt could fix the DB_ENV pointer, so either
DB4 doesn't work on your system or there's a define that could fix
it.

Attachment: pgpdf22tj6p4N.pgp
Description: PGP signature