netdev - Re: [PATCH net-next 12/14] gtp: Configuration for zero UDP checksum

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, 21 Sep 2017 15:41:03 -0700
From:   Tom Herbert <tom@...ntonium.net>
To:     Harald Welte <laforge@...monks.org>
Cc:     Tom Herbert <tom@...bertland.com>,
        David Miller <davem@...emloft.net>,
        Linux Kernel Network Developers <netdev@...r.kernel.org>,
        Pablo Neira Ayuso <pablo@...filter.org>,
        Rohit Seth <rohit@...ntonium.net>
Subject: Re: [PATCH net-next 12/14] gtp: Configuration for zero UDP checksum

On Wed, Sep 20, 2017 at 6:55 PM, Harald Welte <laforge@...monks.org> wrote:
> Hi Tom,
>
> On Wed, Sep 20, 2017 at 11:09:29AM -0700, Tom Herbert wrote:
>> On Mon, Sep 18, 2017 at 9:24 PM, David Miller <davem@...emloft.net> wrote:
>> > From: Tom Herbert <tom@...ntonium.net>
>> >> Add configuration to control use of zero checksums on transmit for both
>> >> IPv4 and IPv6, and control over accepting zero IPv6 checksums on
>> >> receive.
>> >
>> > I thought we were trying to move away from this special case of allowing
>> > zero UDP checksums with tunnels, especially for ipv6.
>>
>> I don't have a strong preference either way. I like consistency with
>> VXLAN and foo/UDP, but I guess it's not required. Interestingly, since
>> GTP only carries IP, IPv6 zero checksums are actually safer here than
>> VXLAN or GRE/UDP.
>
> Just for the record: I don't care either way and I defer to the kernel
> networking developers to decide if they want to have zero UDP checksum
> in GTP or not.
>
> The 3GPP specs don't say anything about UDP checksums.  So there's no
> requirement to use them, and hence operation without UDP checksums
> should be compliant.  Cisco GTP implementation has udp checksumming
> configurable, so other implementations also seem to provide both ways.
>
> In general, I would argue one wants UDP checksumming of GTP in all
> setups, as while the inner IP packet might be protected, the GTP header
> itself is not, and that's what contains important data suhc as the TEID
> (Tunnel Endpoint ID).  But that's of course just my personal opinion,
> and I'm not saying we should prevent people from using lower protection
> if that's what they want.
>
The tradeoffs and requirements of zero UDP6 checksums are discussed at
length in RFC6935 and RFC6936. Given other implementations make it
configurable it should also be here.

Tom

> --
> - Harald Welte <laforge@...monks.org>           http://laforge.gnumonks.org/
> ============================================================================
> "Privacy in residential applications is a desirable marketing option."
>                                                   (ETSI EN 300 175-7 Ch. A6)