Messages in this thread | | | Date | Wed, 13 Jul 2011 09:06:29 +0300 | Subject | Re: [PATCH] hfsplus: Add record offset check | From | Pekka Enberg <> |
| |
On Wed, Jul 13, 2011 at 6:58 AM, Hin-Tak Leung <hintak_leung@yahoo.co.uk> wrote: >> Corrupted disk may return record offset which is larger >> than node size >> and cause general protection fault like below: > > <snipped> > >> This patch add guard for this situation. >> >> Signed-off-by: Naohiro Aota <naota@elisp.net> > > Nacked. This isn't acceptable. Explained above.
'recoff' is read from disk which can be easily fuzzed to have an offset that's larger than node_size, no? The kernel shouldn't oops in that cases so what's the problem with the patch? (The changelog is terribly vague, though, and needs to be fixed).
Pekka
| |