From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Graf Subject: Re: [PATCH net-next] vxlan: expose COLLECT_METADATA flag to user space Date: Wed, 5 Aug 2015 16:39:15 +0200 Message-ID: <20150805143915.GA3530@pox.localdomain> References: <1438312222-7130-1-git-send-email-ast@plumgrid.com> <20150731080019.GB4738@pox.localdomain> <55BB9734.1010005@plumgrid.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Alexei Starovoitov , "David S. Miller" , Daniel Borkmann , netdev To: Jesse Gross Return-path: Received: from mail-wi0-f182.google.com ([209.85.212.182]:37861 "EHLO mail-wi0-f182.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751550AbbHEOjS (ORCPT ); Wed, 5 Aug 2015 10:39:18 -0400 Received: by wibhh20 with SMTP id hh20so27477997wib.0 for ; Wed, 05 Aug 2015 07:39:16 -0700 (PDT) Content-Disposition: inline In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: On 08/03/15 at 02:14pm, Jesse Gross wrote: > On Fri, Jul 31, 2015 at 8:41 AM, Alexei Starovoitov wrote: > > thanks. I think exposing collect_metadata for vxlan and in the future > > for other tunnel types is the clean enough way, though the other > > alternative would be to get rid of collect_metadata flag > > from the kernel and do it when flowmode flag is set. Thoughts? > > This seems like a good idea to me - I'm not sure that flow based > tunnels are all that useful without metadata collection enabled and > the fewer interfaces that we have to create for each tunnel type, the > better. The use case at hand where the flag is useful if someone is only doing flow based TX which can be handy if the tunnel id remains unused.