From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [net-next PATCH v2 0/8] tc offload for cls_u32 on ixgbe Date: Tue, 16 Feb 2016 14:50:55 -0500 (EST) Message-ID: <20160216.145055.1581160555302651949.davem@davemloft.net> References: <20160216055819.32452.84181.stgit@john-Precision-Tower-5810> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: john.fastabend@gmail.com, amir@vadai.me, jhs@mojatatu.com, netdev@vger.kernel.org, jiri@resnulli.us, jeffrey.t.kirsher@intel.com To: gerlitz.or@gmail.com Return-path: Received: from shards.monkeyblade.net ([149.20.54.216]:39140 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751872AbcBPTu5 (ORCPT ); Tue, 16 Feb 2016 14:50:57 -0500 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: From: Or Gerlitz Date: Tue, 16 Feb 2016 21:30:18 +0200 > On Tue, Feb 16, 2016 at 8:00 AM, John Fastabend > wrote: > [...] > >> Specifically this uses a 'big hammer' feature bit >> to do the offload or not. If the bit is set you get offloaded rules >> if it is not then rules will not be offloaded. If we can agree on >> this patch series there are some more patches on my queue we can >> talk about to make the offload decision per rule using flags similar >> to how we do l2 mac updates. Additionally the error strategy can >> be improved to be hard aborting, log and continue, etc. I think >> these are nice to have improvements but shouldn't block this series. > > John, > > I thought we've got into agreement in netdev that the offloading > directive would be a tri-state creature (sw, hw and fail the op if > can't, hw and fallback to sw). Can you please elaborate what do you > mean by "make the offload decision per rule using flags similar to how > we do l2 mac updates" I think John's changes are an intermediate step in a long term evoluation. I'm going to apply John's changes as-is, and people can submit relative changes on top if they want.