> > fs/ext2/inode.c, it looks like a "should never happen" condition.
> >
> > Remy? Do you agree?
>
> Yes. This is a ``should never happen'' condition which happens
> anyway :-) I do not understand how (and why) it happens though. BTW, a
> guy in Paris has the very same problem and I will soon try to make some
> testing on his machine. Finding and fixing the problem will be easier this
> way. I expect to be able to do it next week.
>
> > - Ted
> Remy
>
As you may have noticed, gnu.gcc.announce'd GCC-2.6.3 with bounds checking
today. This may find some problems if Linus and other kernel developers
haven't been using Checker or Purify, etc. It stated:
"You will need to have a fresh source tree for GCC 2.6.3. The patches
are located in
ftp://dse.doc.ic.ac.uk/pub/misc/bcc
There is a README file there which you should read first."
Note, it is "Version 0.1 alpha."
Hopefully this is useful information, otherwise please ignore.
-- Don Parsons