Re: [PATCH net] l2tp: reject short IPv6 address attributes

From: khj

Date: Tue Nov 17 2026 - 04:35:10 EST


The issue was found by a static analysis scan (pattern rules for
NLA_BINARY attributes that are read with a fixed size, LLM-assisted),
then confirmed by reading the code.

It was triggered at runtime on 7.4-rc1 in QEMU: an unprivileged user in
its own user+net namespace creates an L2TPv3 IPv6 tunnel with a 0 or
4 byte L2TP_ATTR_IP6_DADDR, and L2TP_CMD_TUNNEL_GET returns 16 bytes,
the missing ones being stale heap data:

sent 0 byte(s): 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41 41
sent 4 byte(s): 20 20 20 20 41 41 41 41 41 41 41 41 41 41 41 41

There is no KASAN report, since the read stays inside the skb head
allocation. The leaked bytes are non-zero only with init_on_alloc=0.

2026년 11월 17일 (화) 오후 6:27, <netdev-bot+sinfo@xxxxxxxxxx>님이 작성:
>
> Hi!
>
> This is an automated message. This series looks like a fix, but its
> commit messages seem to be missing some information:
>
> - How the issue was discovered, e.g. hit in production, hit during
> development, syzbot report, manual code inspection, LLM or static
> analysis tool scan.
>
> - Whether the issue was actually triggered, or is only theoretical
> (e.g. found by LLM and not proven at runtime). If it was triggered
> please include the symptoms, like the stack trace or kernel messages.
>
> Please do not repost the series just to address the above. Instead,
> reply to this email with the missing information, so that reviewers
> can take it into account.
>
> *IMPORTANT*: If you do repost (for other reasons) - please include
> the relevant info in the cover letter or commit messages.
> --
> The evaluation is done by an LLM so it may be wrong, if you think
> that is the case please reply and explain.