netdev - Re: [PATCH net-next] netlink: make getters tolerate NULL nla arg

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 16 Jan 2020 17:23:31 +0100
From:   Toke Høiland-Jørgensen <toke@...hat.com>
To:     Florian Westphal <fw@...len.de>
Cc:     Florian Westphal <fw@...len.de>, netdev@...r.kernel.org
Subject: Re: [PATCH net-next] netlink: make getters tolerate NULL nla arg

Florian Westphal <fw@...len.de> writes:

> Toke Høiland-Jørgensen <toke@...hat.com> wrote:
>> Florian Westphal <fw@...len.de> writes:
>> 
>> > One recurring bug pattern triggered by syzbot is NULL dereference in
>> > netlink code paths due to a missing "tb[NL_ARG_FOO] != NULL" test.
>> >
>> > At least some of these missing checks would not have crashed the kernel if
>> > the various nla_get_XXX helpers would return 0 in case of missing arg.
>> 
>> Won't this risk just papering over the issue and lead to subtly wrong
>> behaviour instead? At least a crash is somewhat visible :)
>
> How?  Its no different than tb[X] being set with a 0 value.

I was thinking that at lack of NULL check could also imply a lack of
proper bounds checking. And that the crashes at least shine a light on
them forcing people to consider whether that is indeed the case?

(IDK if that's actually the case, I'm asking :))

-Toke