2012/5/29 Charles Manning <cdhmanning@gmail.com>:
> I expect there has been some shuffling of the mtd interface structures
> that should be reasonably easy to untangle.
>
> Please try give it a go and send patches.
>
> -- Charles

Well, here you have, a quick and completely irresponsible hack ;)

Note that I didn't really stop too much to understand what's going one
while doing the modifications (I knew nothing about mtd beforehand
so...). This is just a quick hack, to see how difficult it would be
the port. The logic can be corrected later, I guess, if needed (for
sure). Too much ifdefs if you ask me, but it seems to work for now.

One thing that should really be checked is the kernel version. I used
3.4 as the cutter because that's what I have right now, but I am sure
that that version should be lowered. Probably all the 3.x releases and
some of the latest 2.6 might fail to compile. I have some kernels
around and can give this a try, but can't really tell you when I will
get the time to do so.

Most of the modifications seem to consist of a migration from
mtd->whateverfunc to globals in the fashion of mtd_whateverfunc, same
argument list. So, it was trivial even for me. That is, until bugs
appear hehe.

Well, the module compiles, and after moving it to /lib/modules and
doing depmod -a, it even modprobes fine, and I can see yaffs and
yaffs2 listed in /proc/filesystems. I haven't managed to mount
anything. So I am now sure that my android phone images are not in
this file system format. But that's another much longer story. So, I
think I've achieved my goal for today, even if it did serve no real
purpose, and It's time to sleep.

Good night. :)

-- 
Jesús Guerrero Botella