Re: Invalid module format

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

* Peter_22@xxxxxx wrote:

> > * markus reichelt <ml@xxxxxxxxxxxxx> wrote:
> > Have you managed to solve the problem in the meantime?
> > 
> > If not, can you reproduce the error (maybe via a virtual machine)?
> dmesg and kernel config say loop module is built-in :-(

Welcome to the club. Slackware has loop also built-in by now. 

And a modular kernel compile takes quite some time...

> In practice a kernel recompile is needed. Seems the ubuntu jerks
> tend to adopt the same bad habits SuSE developers developed before
> I changed to ubuntu. The ubuntu distro features plenty of debian
> packages, even loop-aes related ones. In the manual of one of these
> Max Vozeler is named, the maintainer. I tried to integrate loop-aes
> via these packages with module-assistant. Following this path leads
> to the same result.

As was already pointed out, dirty hacking to make the loop-aes module
work is doable (proof of concept), but it shouldn't be done. At all.
It breaks backwards compatibility and is a hassle to maintain.

> Once again I get the impression loop-aes is somehow driven out of
> linux for the sake of luks/dm-crypt. There is no sane reason to
> build the loop module into the kernel other than hindering people
> from using loop-aes and other fancy stuff relying on loops. So far
> I only looked at the 32bit version but I have little hope for
> Jaunty Jackalope on 64bit AMD either. I don't understand why they
> do this to me.

Actively driven out is too harsh a wording, imho. It's somewhat
neglected, yes, but there are valid reasons for it.

It's just unfortunate how it ends up affecting the end-user.

left blank, right bald
loop-AES FAQ:

Attachment: pgpEBxTXvWgkf.pgp
Description: PGP signature

[Home]     [Kernel]     [Linux Crypto]     [Gnu Crypto]     [Gnu Classpath]     [Netfilter]     [Bugtraq]     [Network Security Reading]

Add to Google