From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [PATCH] new UDPCP Communication Protocol Date: Tue, 11 Jan 2011 22:46:26 +0100 Message-ID: <1294782386.3447.21.camel@edumazet-laptop> References: <1294779020.16410.2.camel@wall-e> <20110111.125254.133402589.davem@davemloft.net> <1294780480.17388.2.camel@wall-e> <20110111.131953.180400101.davem@davemloft.net> <1294782117.17531.18.camel@wall-e> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: David Miller , linux-kernel@vger.kernel.org, akpm@linux-foundation.org, netdev@vger.kernel.org, shemminger@vyatta.com, jj@chaosbits.net, daniel.baluta@gmail.com, jochen@jochen.org, hagen@jauu.net, torvalds@linux-foundation.org, pavel@ucw.cz To: Stefani Seibold Return-path: In-Reply-To: <1294782117.17531.18.camel@wall-e> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Le mardi 11 janvier 2011 =C3=A0 22:41 +0100, Stefani Seibold a =C3=A9cr= it : > Second, the design is may in your opinion poor. I like it. What is > really poor is the kernel_...() socket functions, which are only simp= le > wrapper of the system calls without any performance improvement, skb > support and memory saving. >=20 The only thing you want is to have a callback to your own code to deliver an decapsulated skb to your state machine. Take a look at other layers on top of UDP (L2TP comes to mind)