netdev - Re: [PATCH] netfilter: nft_set_hash: fix potential NULL deref in nft_rhash

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] [day] [month] [year] [list]

Message-ID: <aTA9y3iH1JjMPwQ1@strlen.de>
Date: Wed, 3 Dec 2025 14:40:27 +0100
From: Florian Westphal <fw@...len.de>
To: Melbin K Mathew <mlbnkm1@...il.com>
Cc: pablo@...filter.org, kadlec@...filter.org, phil@....cc,
	netfilter-devel@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: [PATCH] netfilter: nft_set_hash: fix potential NULL deref in
 nft_rhash_deactivate

Melbin K Mathew <mlbnkm1@...il.com> wrote:
> In nft_rhash_deactivate(), rhashtable_lookup() may return NULL when the
> set element is not found, but the function unconditionally returns
> &he->priv.

Which is equal to 'return NULL' in that case.

> Dereferencing a member of a NULL pointer is undefined behavior in C.

&he->priv doesn't dereference he, it returns the address of the member.

> Although the current struct layout places 'priv' at offset 0 (making
> this behave like returning NULL), this is fragile and relies on
> implementation details.

Its not fragile, this file has:

        BUILD_BUG_ON(offsetof(struct nft_rhash_elem, priv) != 0);

to ensure 'priv is first member' requirement.