From mboxrd@z Thu Jan 1 00:00:00 1970 From: jamal Subject: Re: [PATCH 09/16] [IPv6] address: Convert address notification to use rtnl_notify() Date: Mon, 14 Aug 2006 22:46:05 -0400 Message-ID: <1155609965.6946.74.camel@jzny2> References: Reply-To: hadi@cyberus.ca Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: Alexey Kuznetsov , tgraf@suug.ch, davem@davemloft.net, netdev@vger.kernel.org Return-path: Received: from mx02.cybersurf.com ([209.197.145.105]:64158 "EHLO mx02.cybersurf.com") by vger.kernel.org with ESMTP id S1752003AbWHOCqJ (ORCPT ); Mon, 14 Aug 2006 22:46:09 -0400 Received: from mail.cyberus.ca ([209.197.145.21]) by mx02.cybersurf.com with esmtp (Exim 4.30) id 1GCowZ-0005J1-M7 for netdev@vger.kernel.org; Mon, 14 Aug 2006 22:46:15 -0400 To: Herbert Xu In-Reply-To: Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Tue, 2006-15-08 at 10:56 +1000, Herbert Xu wrote: > Agreed. The pid field in the netlink header should be treated as an > opaque value. Any attempt to interpret it as the process ID is doomed > to failure. Not necessarily as a processid ("PID" is a really bad noun in that sense); but rather as something meaningful of interpretation in regards to the real origin of the executed change. The concept of "whodunnit" is invaluable. And a processid tends to be useful when nothing else is there to identify the originator. Just saying "it is the kernel" (PID=0) when the kernel just acted as a proxy of some user space app, is not useful all the times, IMO. [Routes, but not other functional blocks in rtnetlink, actually have a field (called protocol) that says who added them]. Thats all the quagga and other folks were looking for in cases where it was ambigous. cheers, jamal