gcc-4.1.1

Terry Chan tpchan at comcast.net
Sat Jan 27 15:42:32 CET 2007


When apps break due to gcc changes in the ABI, it's almost always C++ apps.

So a pure C++ app that was compiled with gcc-3.4.x can/will have problems
with the ABI in gcc-4.x.x.  Usually a recompile of the C++ app with the
newer compiler is all that is necessary.  Or you might try recompiling
the dependency tree of the C++ app as more C++ parts can/will be broken as
well.

If all that fails, then you'll need to check with the devs of the said C++
app to see if they have a svn/cvs version that fixes your problem(s).

So yes, if you only have ONE lunar box, then don't feel compelled as a lunar
dev to possibly wreck you're only working box by installing a newer gcc-4.x.

Terry Chan
------------------------------------------------------------------
On Sat, Jan 27, 2007 at 01:40:38PM +0100, samuel verstraete wrote:
> I was wrong... R b0rked :)
> 
> On 1/27/07, samuel verstraete <samuel.verstraete at gmail.com> wrote:
> > i just lin -w 4.1.1 gcc
> > and i had no problems at all with running any binary that was compiled
> > with 3.4.6 .
> > Is that clear enough?
> >
> > I upgraded now two boxes already and stillnone of them b0rked.
> >
> > cheers,S.
> >
> >
> >
> > On 1/27/07, Jan Eidtmann <cmak at lunar-linux.org> wrote:
> > > On 27.01.07, Moritz Heiber wrote:
> > > > On Sat, 27 Jan 2007 12:37:15 +0100
> > > > Jan Eidtmann <cmak at lunar-linux.org> wrote:
> > > >
> > > > > which where compiled with gcc 3.4.6?
> > > > > when you say "the update" did that just involve the lin -w
> > > > > 4.1.1 gcc part, or anything else? :>
> > > >
> > > > God cmak, I'm telling you right now: ITS SAFE, okay?
> > > >
> > > > Jesus!
> > >
> > > im not sure about that...why dont you just answer my
> > > question? i shouldnt have used the ":>" i guess. it looks
> > > like its saying "gotcha, and you know that". but what it
> > > really sais is "please help, please waste your time for me,
> > > please hmmkay, im clueless"
> > >
> > >  ///
> > > cmak


More information about the Lunar-dev mailing list