On Mon, Nov 17, 2014 at 00:25, David Gwynne wrote: > im in two minds about this. > > on the one hand, its not like this can be attacked remotely. on the > other hand, maybe someone would create a metric buttload of files > and open all the ones with inodes whose values are 4096 apart. > > discussion? tests? ok?
ok > > part of me thinks we shoudl apply siphash to all the kernel hashes > as a matter of course rather than waste the effort trying to figure > out if they are attackable or not. Almost replied to tell you that, then I finished reading your email. :)
