netdev - Re: [PATCH net] ipv6: set lwtstate for ns and na dst

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]

Date:   Tue, 26 Nov 2019 16:47:11 +0800
From:   Xin Long <lucien.xin@...il.com>
To:     David Miller <davem@...emloft.net>
Cc:     network dev <netdev@...r.kernel.org>,
        David Ahern <dsahern@...il.com>, Jiri Benc <jbenc@...hat.com>,
        Roopa Prabhu <roopa@...ulusnetworks.com>
Subject: Re: [PATCH net] ipv6: set lwtstate for ns and na dst

On Tue, Nov 26, 2019 at 6:53 AM David Miller <davem@...emloft.net> wrote:
>
> From: Xin Long <lucien.xin@...il.com>
> Date: Mon, 25 Nov 2019 14:23:12 +0800
>
> > This patch is to fix an issue that ipv6 ns and na can't go out with the
> > correct tunnel info, and it can be reproduced by:
>
> And why shouldn't RS and redirects get this treatment too?
RS daddr is using IN6ADDR_LINKLOCAL_ALLROUTERS_INIT.
I didn't think it could be an address that users may use to configure
a lwtunnel state.

>
> And then, at that point, all callers of ndisc_send_skb() have this
> early route lookup code (and thus the "!dst" code path is unused), and
> the question ultimately becomes why doesn't ndisc_send_skb() itself
> have the dst lookup modification?
>
> I'm not too sure about this change and will not apply it.
I'm not so sure about this fix either, but I couldn't see other fixes for it.
Hope to get more developer's opinions, especially from router's.
besides the above, do you have other concerns?

Thanks.