jfm2@xxxxxxxxxxxxxxxx wrote:
>
> Seth Mos a écrit :
> >
> > At 21:26 18-9-2001 +0200, Jean Francois Martinez wrote:
> > >Seth Mos a écrit :
> > >
> > > > At 13:02 18-9-2001 -0400, Arun Ramakrishnan wrote:
> > > > >Hi,
> > > > > I heard that 2.96 is again a devel version of gcc which is sorta
> > > > > unstable.I
> > > >
> > > > It was a CVS snapshot.
> > > >
> > >
> > >It was a CVS snapshot some 18 months ago. Eighteen months of bug hunting
> > >later you can tell nothing about its stability. In fact because Gcc 2.96
> > >was
> > >frozen
> > >over a year before gcc 3.0 and has been far more dceployed I trust it far
> > >m
> > >ore than gcc 3.0+
> >
> > True, but I gather 3.0+ will pop up in a lot more distributions then just
> > redhat and mandrake.
> > 3.0 is a official release which means it is not distribution specific. The
> > reason that mandrake adopted is was more or less because a lot of mandrake
> > is still redhat based. (no flame intended)
> >
>
> I don't care which one is official what I care is which one crashes o
> and which one miscompiles less often
> For now gcc 3.0 is unproven and at least gcc 3.0 (don't know about gcc
> 3.0.1)
> has some horrific bugs far worse than those gcc 2.96 had in september
> 2000.
>
> Also RedHat and Mandrake have far more users than all the remaining
> distributions
> combined. Both RedHat ad Mandrake are poised to ship gcc 3 but gcc 2.96
> will be
> the default compiler.
>
> Mandrake is no longer a clone of RedHat. Factors were gcc 2.96 superior
> C++
> respective to gcc 2.95, faster executables (benchmarked) and perhaps
> support of
> Itanium.
Funny, I have been using gcc 3+ for months for building kernels, xfs
whatever... and havent got any problems yet.
one system is running semi-production with 1TByte of storage attached to
it.(a 2 CPU 4Gig machine)
runnin an xfs kernel (mid august CVS) for 4 weeks built with gcc3 no
problems yet... although we tested the system thoroughly on hardware
failures. pulling power in the midst of a performance test ... etc.
and had no barfs till date... only 1 problem but there I found a
workaround ... increasing size of xfs fs that had been grown already,
unmounting and xfs_repairing, mounting and extending ... worked
|