From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Ahern Subject: Re: [PATCH net-next] vrf: Add VRF_F_BYPASS_RCV_NF flag to vrf device Date: Thu, 10 Jan 2019 10:00:06 -0700 Message-ID: <3595b09c-0000-0d67-e080-0341bdbf307e@gmail.com> References: <1545896287-7282-1-git-send-email-wenxu@ucloud.cn> <2e98a763-664a-6a95-53ad-a150c3c199b8@gmail.com> <493e5cb3-224c-f83b-4fbd-caa1703bb379@ucloud.cn> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org To: wenxu , David Miller Return-path: Received: from mail-pl1-f196.google.com ([209.85.214.196]:40348 "EHLO mail-pl1-f196.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728266AbfAJRAK (ORCPT ); Thu, 10 Jan 2019 12:00:10 -0500 Received: by mail-pl1-f196.google.com with SMTP id u18so5441941plq.7 for ; Thu, 10 Jan 2019 09:00:09 -0800 (PST) In-Reply-To: <493e5cb3-224c-f83b-4fbd-caa1703bb379@ucloud.cn> Content-Language: en-US Sender: netdev-owner@vger.kernel.org List-ID: On 1/10/19 8:21 AM, wenxu wrote: > > How about the status of this patch? Should I resubmit it? > I do not like the need for a flag when the VRF is created. If something changes with the firewall rules, it means a user has to delete and re-create the VRF which is really expensive. It would be better to detect this on the fly - similar to how it detects the default qdisc and avoids the recirculation on Tx when the qdisc is the default.