netdev - Re: [PATCH/RFC net-next v2 2/4] flow dissector: return error on icmp dissection under-run

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]

Message-ID: <f1590e4d-aeba-2f52-bb74-5e905659bcf9@mojatatu.com>
Date:   Mon, 8 May 2017 07:21:29 -0400
From:   Jamal Hadi Salim <jhs@...atatu.com>
To:     Simon Horman <simon.horman@...ronome.com>,
        Jiri Pirko <jiri@...lanox.com>,
        Cong Wang <xiyou.wangcong@...il.com>
Cc:     Dinan Gunawardena <dinan.gunawardena@...ronome.com>,
        netdev@...r.kernel.org, oss-drivers@...ronome.com,
        Benjamin LaHaise <benjamin.lahaise@...ronome.com>
Subject: Re: [PATCH/RFC net-next v2 2/4] flow dissector: return error on icmp
 dissection under-run

On 17-05-05 08:47 AM, Simon Horman wrote:
> Return an error from __skb_flow_dissect() if insufficient packet data is
> present when dissecting icmp type and code.
>
> Without this patch the absence of the ICMP type and code in truncated
> ICMPv4 or IPVPv6 packets is treated the way same as the presence of a code
> and type of value of zero.  And without this patch the flower classifier is
> unable to differentiate between these two cases which may lead to
> unexpected matching of truncated packets.
>
> With this patch the flow dissector and in turn the flower classifier can
> differentiate between packets with zero ICMP type and code, and truncated
> packets.
>
> The approach taken here is to return an error if the IP protocol indicates
> ICMP but the type and code data is not present in the packet - an error
> return value from __skb_header_pointer().
>
> This should only effect the flower classifier as it is the only user of
> W_DISSECTOR_KEY_ICMP.  The behavioural update for flower only takes effect
> with a separate patch to have it refuse to match if dissection fails.
>
> Signed-off-by: Simon Horman <simon.horman@...ronome.com>
> Reviewed-by: Benjamin LaHaise <benjamin.lahaise@...ronome.com>

Reviewed-by: Jamal Hadi Salim <jhs@...atatu.com>

cheers,
jamal