> Is it is sufficient to make the
> lookup inode operation et al. and the hash and compare
> dcache operations case insensetive to make a foolproof
> race-free, kernel-panic free case-insensetive filesystem?
Yes. Assuming that you took care of the normal fs stuff and
are sure that fs you are mounting doesn't contain /FooBar and /fooBAR
simultaneously, indeed.
> Just a quick answer please. This is for an assignment
> so *no* detailed solutions. A 'yes' would be nice.
> CC any replies to me please...
>
> While I have your attention, I have a few one-liners:
>
> A debug version of linux that would allow you to force
> unload a module with registered devices etc and not
> have the whole system fall to pieces.
Eh? One-liner? You managed to do do one-line revoke()? Share!
> Can you change the permissions of files in /proc without
> patching and recompiling? (BTW, IMHO it's annoying that
> ipchains -L only works for root. Seems wrong)
In principle it can be done, but that will not be persistent over reboots,
indeed. Another problem being that you can't guarantee persistent
permissions for files registered by modules, unless you have clever
insmod that will set them on each loading.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/