From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pablo Neira Ayuso Subject: Re: [PATCH libnftables v2] Add support for ct set Date: Fri, 10 Jan 2014 11:37:58 +0100 Message-ID: <20140110103758.GA4732@localhost> References: <1389170211-7024-1-git-send-email-kristian.evensen@gmail.com> <20140110005015.GA18032@localhost> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Netfilter Development Mailing list To: Kristian Evensen Return-path: Received: from mail.us.es ([193.147.175.20]:58494 "EHLO mail.us.es" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756695AbaAJKiN (ORCPT ); Fri, 10 Jan 2014 05:38:13 -0500 Content-Disposition: inline In-Reply-To: Sender: netfilter-devel-owner@vger.kernel.org List-ID: On Fri, Jan 10, 2014 at 09:26:12AM +0100, Kristian Evensen wrote: > Hi Pablo, > > On Fri, Jan 10, 2014 at 1:50 AM, Pablo Neira Ayuso wrote: > > Please, rework this. It would be good to rework the meta/set part > > available in libnftables next-3.14. If you cannot make it, let me know > > and I'll schedule time to fix that. Thanks. > > Thank you for your feedback, I will try to rework the patch today and > if not, then over the weekend. > > Btw, during development I noticed that the dreg in ct (in libnftables) > is store as a uint32, while meta uses a uint8 for dreg/sreg. I use a > uint32 for sreg to be consistent with what is already there, but after > looking more into the code this seems not be needed as kernel > sreg/dreg is only 8 bits wide. Should I change the storage type for > sreg/dreg at the same time, or does it qualify as obscure behavior > too? That's inconsistent and needs to be fixed. My suggestion is to fix it by using u32 for registers. Thanks.