From mboxrd@z Thu Jan 1 00:00:00 1970 From: Roland Dreier Subject: Re: IPoIB: Fix multicast packet drops before join is complete Date: Tue, 09 Jun 2009 22:04:21 -0700 Message-ID: References: <20090609.174504.114047526.davem@davemloft.net> <20090609.215719.214969588.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: cl@linux-foundation.org, netdev@vger.kernel.org, yosefe@Voltaire.COM To: David Miller Return-path: Received: from sj-iport-2.cisco.com ([171.71.176.71]:58969 "EHLO sj-iport-2.cisco.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751448AbZFJFET (ORCPT ); Wed, 10 Jun 2009 01:04:19 -0400 In-Reply-To: <20090609.215719.214969588.davem@davemloft.net> (David Miller's message of "Tue, 09 Jun 2009 21:57:19 -0700 (PDT)") Sender: netdev-owner@vger.kernel.org List-ID: > If a facility doesn't exist yet, we're going to have to create > one. It would need to do a downcall to the device when the > user joins a multicast group on a socket, and then the device > can do whatever magic is necessary to speak multicast immediately > and sleep until it really is available for immediate use. But a send-only membership is created when a multicast packet is sent, which an application can do just with sendto() -- do we want to plumb all the way from sendto() down to the device the packet is ultimately going to be sent on? There isn't any corresponding code to block an app during ARP resolution, is there? - R.