From mboxrd@z Thu Jan 1 00:00:00 1970 From: jamal Subject: Re: [GIT PULL v2] Open vSwitch Date: Wed, 30 Nov 2011 08:23:05 -0500 Message-ID: <1322659385.2243.23.camel@mojatatu> References: <20111123075433.GA7928@gondor.apana.org.au> <1322050976.2039.125.camel@mojatatu> <20111128130409.GB16828@gondor.apana.org.au> Reply-To: jhs-jkUAjuhPggJWk0Htik3J/w@public.gmane.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: dev-yBygre7rU0TnMu66kgdUjQ@public.gmane.org, netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Herbert Xu , David Miller To: Jesse Gross Return-path: In-Reply-To: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: dev-bounces-yBygre7rU0TnMu66kgdUjQ@public.gmane.org Errors-To: dev-bounces-yBygre7rU0TnMu66kgdUjQ@public.gmane.org List-Id: netdev.vger.kernel.org On Tue, 2011-11-29 at 22:18 -0800, Jesse Gross wrote: > As Jamal alluded to above, it's > actually the bridge code which is more conceptually similar. Either you misread what i said or i miscommunicated. The exact similarity is in classifier action in the datapath. The bridge, as i suggested, could have had at least two features added to it in regards to learning to achieve what you wanted it to. But as pointed out the bridge - which is a victim of combining policy and mechanism in one spot - already has too many features. If we cleanly separate out those things, then i dont see why we need two bridge implementations. Ok, so here's a digression: I am uncomfortable with the fact i have to use ovs as the way to configure things in a 48 port Gige switch. In Linux we have netdevs; if you expose things as netdevs, for starters i can use standard tools to do things to them. But this is a side discussion I started with Justin - so you may have no pony in this race. cheers, jamal