From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter P Waskiewicz Jr Subject: Re: [net-next-2.6 PATCH 06/13] igb: move all multicast addresses into multicast table array Date: Fri, 24 Jul 2009 18:41:29 -0700 Message-ID: <1248486089.3462.1.camel@localhost.localdomain> References: <20090724040700.30709.31473.stgit@localhost.localdomain> <20090724040853.30709.97954.stgit@localhost.localdomain> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: "Duyck, Alexander H" , "David S. Miller" , netdev , "Kirsher, Jeffrey T" , "gospo@redhat.com" To: Jesper Dangaard Brouer Return-path: Received: from mga01.intel.com ([192.55.52.88]:44093 "EHLO mga01.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752247AbZGYBmE (ORCPT ); Fri, 24 Jul 2009 21:42:04 -0400 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: On Fri, 2009-07-24 at 01:57 -0700, Jesper Dangaard Brouer wrote: > On Thu, 23 Jul 2009, Jeff Kirsher wrote: > > > From: Alexander Duyck > > > > This patch moves all of the multicast addresses out of the free Receive > > address registers and instead programs them all into the multicast table > > array. As a result the multicast filtering may not be as precise, but it > > also greatly reduces the overhead for multicast addresses. > > What do you mean by "the multicast filtering may not be as precise" ? > > I'm planning to use these NICs for multicast traffic (both routing and > monitoring). Is there any precausing I need to be aware of? > The receive address registers, or RARs, are perfect match filters. The multicast table array, or MTA, contains hashed values of the multicast addresses. Therefore, it isn't a perfect match for the multicast addresses anymore. But to support a larger number of multicast addresses, they can't be written to the RARs, since those resources are small and are needed for MAC and unicast address filtering. Cheers, -PJ Waskiewicz