Re: Release update: debian-installer, kernels, infrastructure, freeze, etch, arm

[Adrian Bunk <bunk@stusta.de>]

On Sat, Apr 02, 2005 at 12:35:55AM +0100, Matthew Garrett wrote:
> Adrian Bunk <bunk@stusta.de> wrote:
> > It doesn't need to be an exact date, but someting like
> > "third quarter of 2005" or "mid-2008" would help to avoid situations 
> > like the sarge C++ transition that was too early [1] or the more than 
> > two years old X11 that will ship with sarge (and that doesn't support 
> > all hardware supported by recent X.org releases) [2].
> 
> (snip)
> 
> > [1] the first birthday of gcc 3.4.0 is only a few days from now
> 
> The C++ ABI was broken with gcc 3.2.0. For most architectures, 3.4.0
> doesn't change anything. Is your point anything other than "The Debian
> release process is broken and you should get rid of testing"? If not,
> we've heard that several times already. It doesn't need reiterating.


Where did I say that these problems had anything to do with testing?

It seems you are one of the people who think
"You don't like testing so everything you are saying has to be wrong.".


Please check the facts:

gcc 3.4 has a different C++ ABI compared to gcc 3.2/3.3 on _all_ 
architectures [1].


cu
Adrian

[1] look at libstdc++5 <-> libstdc++6

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed