netdev - Re: [PATCH V1 net-next 2/2] net: Add UDP GRO support for vxlan traffic

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: <CA+mtBx_B+7CU3GDiod+n7tHfwBcA_3DEhD7tvPYhsZMebiwY8w@mail.gmail.com>
Date:	Mon, 6 Jan 2014 08:50:12 -0800
From:	Tom Herbert <therbert@...gle.com>
To:	Or Gerlitz <or.gerlitz@...il.com>
Cc:	Or Gerlitz <ogerlitz@...lanox.com>, Jerry Chu <hkchu@...gle.com>,
	Eric Dumazet <edumazet@...gle.com>,
	Herbert Xu <herbert@...dor.apana.org.au>,
	Linux Netdev List <netdev@...r.kernel.org>,
	David Miller <davem@...emloft.net>,
	Yan Burman <yanb@...lanox.com>,
	Shlomo Pongratz <shlomop@...lanox.com>
Subject: Re: [PATCH V1 net-next 2/2] net: Add UDP GRO support for vxlan traffic

On Mon, Jan 6, 2014 at 8:38 AM, Or Gerlitz <or.gerlitz@...il.com> wrote:
> On Mon, Jan 6, 2014 at 3:32 PM, Or Gerlitz <or.gerlitz@...il.com> wrote:
>> As the GUE RFC states, with UDP encapsulation, the encap protocol is
>> actually derived from the UDP destination port. With that assumption
>> at hand, we can have the udp gro code to export callbacks for upper
>> uncapsulation protocol to install a gro handler which has the
>> following trivial strucure:
>>
>> key = <PORT>  data = <gro receive/complete callbacks>
>
>> which plugs nicely into the general framework of the gro stack. The
>> udp gro receive code would act as follows
>> -- if there's no gro handler set for the packet udp dest port, flush
>> - do the normal loop to find match on the packet dest/src udp ports
>> - call the protocol (e.g VXLAN) gro receive callback
>> This would avoid dealing with sockets etc and be very efficient, makes sense?
>
> Or in short -- the destination udp port takes a similar role to the
> ethertype for gro l3 handlers or the ip protocol for gro l4 handlers
> and we enable encapsulating protocols to plug into the existing
> offload_callbacks framework, simple.

I think this would be a good start. We can further optimize the encap
path later on.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html