>> I'm sure Charles will ask for this when he wakes up in New
>> Zealand... 
I guess not :-)

> Sure... But I f*cked up, because I "overwrote" the "faulty" libc with a 
> new version to solve the problem! I thought we really needed a working 
> device to get further, but we settled without, so it's still here on my 
> desk, with the bad libc in the flash but with old "inodes" (I guess)...
> 
> If it comes to it, I guess it is possible to create a special kernel and 
> yaffs-fs implementation, that removes the newly written libc-file, to 
> restore the old state. Although I reckon this requires quite the works ;-(
Can I enable some flags in the YAFFS core enabling some debug 
information that'll help me investigating this problem?
I guess what I want is to have a list of nodes stating what nodes belong 
to what file and when the node was written...

// Martin