Public SourceMage Discussion List

[Jeremy Kolb <jkolb AT brandeis.edu>]

so i should recompile my gcc and then see if i can compile the latest
glibc?

Jeremy

On Tue, 2004-03-16 at 07:46, Arwed von Merkatz wrote:
> On Tue, Mar 16, 2004 at 10:06:09AM +0100, Arwed von Merkatz wrote:
> > On Tue, Mar 16, 2004 at 09:56:26AM +0100, Wolfgang Scheicher wrote:
> > > Am Tuesday 16 March 2004 01:14 schrieb Eric Sandall:
> > > > Quoting Jeremy Kolb <jkolb AT brandeis.edu>:
> > > 
> > > >> configure: error: working compiler support for visibility attribute 
> > > >> is
> > > >> required 
> > > >
> > > > It appears that someone else has this problem, but I saw no resolution
> > > 
> > > That was me, and as far as i suspect that you get a broken gcc if you 
> > > compile 
> > > gcc on a system with a nptl glibc. Too broken to compile any glibc.
> > 
> > I definitely compiled gcc on nptl at least once and don't have such
> > problems, will try it again now.
> 
> Ok, just recompiled gcc with the latest glibc, no problems compiling
> the test program with that. The cause must be something else, though i
> don't really have a clue what that could be.
> binutils  2.14.90.0.8
> gcc       3.3.3
> glibc     2.3.3
> 
> gcc compiled with C, C++, Java, Ada and Objective C frontends, without
> fixed includes.
> 
> > > Luckily, i had a working gcc in a other pc's /var/cache/sorcery
> > > 
> > > Attached a small test file by rycee which uses this visibility 
> > > attribute.
> > > 
> > > Note: this is propably not the only problem, i couldn't compile qt 
> > > either.
> > > 
> > > How can this be fixed? Eventually we have to use a prerelease snapshot 
> > > of gcc 
> > > too, if we use one for glibc :-)
> > > 
> > > Worf