[PATCH] malloc: harden removal from unsorted list
DJ Delorie
dj@redhat.com
Fri Feb 23 22:17:00 GMT 2018
While I have no problems with the patch itself (i.e. LGTM), I have two
side notes:
1. We probably could use better error messages now that we have three
numbered ones. ERROR 458104 - CONSULT MANUAL.
2. At what point do these attack vectors move from "something a hacker
could do" to "something a developer could do" ? If a hacker can do this
type of computations, a hacker can do pretty much whatever they want
already. Given how critical malloc performance is to applications these
days, we need to be careful to not harden against things that are only
possible through the developer's malice.
More information about the Libc-alpha
mailing list