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: Wed, 16 Aug 2006 08:36:31 -0400 Message-ID: <1155731791.5122.115.camel@jzny2> References: <20060815122300.GV14627@postel.suug.ch> <20060816105856.GW14627@postel.suug.ch> <20060816111240.GA28846@gondor.apana.org.au> <20060816113950.GA29003@gondor.apana.org.au> <1155729864.5122.95.camel@jzny2> <20060816120808.GA14627@postel.suug.ch> Reply-To: hadi@cyberus.ca Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: Herbert Xu , hasso@estpak.ee, kuznet@ms2.inr.ac.ru, davem@davemloft.net, netdev@vger.kernel.org Return-path: Received: from mx03.cybersurf.com ([209.197.145.106]:27084 "EHLO mx03.cybersurf.com") by vger.kernel.org with ESMTP id S1751149AbWHPMgq (ORCPT ); Wed, 16 Aug 2006 08:36:46 -0400 Received: from mail.cyberus.ca ([209.197.145.21]) by mx03.cybersurf.com with esmtp (Exim 4.30) id 1GDKdc-0005kr-4Y for netdev@vger.kernel.org; Wed, 16 Aug 2006 08:36:48 -0400 To: Thomas Graf In-Reply-To: <20060816120808.GA14627@postel.suug.ch> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Wed, 2006-16-08 at 14:08 +0200, Thomas Graf wrote: > * jamal 2006-08-16 08:04 > > current->pid i think is coming out to be a bad idea. Thomas' patches > > revert it out. Again this has everything to do with the original idea > > what maps to pid now changing to socketid. > > It probably developed from autobind using current->tid. In one conversation with Alexey he told me there was some inspiration from pfkey in the semantics of it i.e processid. Obviously with many sockets on the same process etc, that assumption is no longer valid. On Wed, 2006-16-08 at 22:08 +1000, Herbert Xu wrote: > On Wed, Aug 16, 2006 at 08:04:24AM -0400, jamal wrote: > > > > What do you think of the idea of infact rewriting the pid to be that of > > the socket id? > > Rewriting it with the netlink socket address? That's fine by me as > long as there is a clear 1-to-1 relationship between the request > and the notification. you would have to call getpeername() to get a correct 1-1 mapping as is today when in doubt. What i was suggesting is notifications using the pid that would id the socket and would therefore require a getpeername() which identify the real socket it came from; if you are fine with what Thomas is doing, then this unnecessary since i was suggesting it as a compromise for consistency you pointed was lacking. cheers, jamal