[arch-releng] 2010.04.05 snapshots ready for testing

[Dieter Plaetinck]

On Tue, 06 Apr 2010 22:38:24 +0200
Thomas Bächler <thomas at archlinux.org> wrote:

> Am 06.04.2010 22:31, schrieb Dieter Plaetinck:
> > thanks, but need more info.
> > which image is this ? netinstall or core?
> > when does this happen? which step exactly in the init process?
> > screenshot?
> > what happens if you pick the normal x86, or the normal x86_64?
> > 
> > i just tried netinstall-dual, picked x86 and it works fine.
> > I only have 32bit systems so i cannot try the other options.
> > (well i can select them, but they refuse to boot, obviously :)
> > 
> > also, please use http://bugs.archlinux.org/index/proj6
> 
> That problem is known (I saw Gerardo posting about it either here or
> on the bugtracker). The 64 bit kernel/aufs lacks the 32 bit
> compatibility system call, so remounting aufs on 32 bit userland with
> a 64 bit kernel fails. We can't do anything about it, and it is
> read-write anyway.
> 

hm you're right
http://mailman.archlinux.org/pipermail/arch-releng/2010-February/000883.html

is this a system call which is not enabled in the arch kernels? or just
does not exist? what exactly do you mean with "it is read-write
anyway"? what is? and how is that relevant?
If i understood ionut's mail correctly, this is something that breaks
the boot process/environment. or not?
maybe we should just disable the "64bit kernel with 32bit userspace"
option.

Dieter