Bug#308649: Useless modules loaded: confirmed on iBook G4

[Sven Luther <sven.luther@wanadoo.fr>]

On Thu, Apr 27, 2006 at 09:32:03AM -0700, Jurij Smakov wrote:
> On Thu, 27 Apr 2006, Jonas Smedegaard wrote:
> 
> >reassign 308649 initrd-tools
> >thanks
> [..]
> >>Wouldn't it make better sense to reassign the bugreport to the package
> >>actually loading those modules, instead of closing it when the kernels
> >>switch ramfs generator default?
> >
> >Sorry, Jurij - I can of course just reassign myself :-)
> >
> >Phillip: Kernels do not load modules themselves, rely on a so-called
> >"ramdisk generator" to collect and load modules needed for your
> >specific computer.
> 
> This bug was filed against 2.6.11 kernel from pre-initramfs era. In 
> kernels before .12 there was no reliable way to detect IDE controller 
> needed, so to be on the safe side all of them were loaded. Unfortunately, 
> at this point unloading of unused modules was not possible. So it's not a 
> bug but more of a feature, and we went through tons of bug reports about 
> it. Reassignment to initrd-tools is effectively a noop since I'm pretty 
> sure that it's impossible to fix in the proper sense (additionally, nobody 
> would push such fixes into stable even if it was possible). So, under the 
> circumstances closing the bug indeed feels like a sensible way.

Well, let's just remove initrd-tools from etch/sid, this will also solve the
issue :)

Friendly,

Sven Luther