Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755834AbYGVM41 (ORCPT ); Tue, 22 Jul 2008 08:56:27 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753803AbYGVM4T (ORCPT ); Tue, 22 Jul 2008 08:56:19 -0400 Received: from smtp-out.google.com ([216.239.33.17]:33870 "EHLO smtp-out.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752873AbYGVM4Q (ORCPT ); Tue, 22 Jul 2008 08:56:16 -0400 DomainKey-Signature: a=rsa-sha1; s=beta; d=google.com; c=nofws; q=dns; h=received:message-id:date:from:to:subject:cc:in-reply-to: mime-version:content-type:content-transfer-encoding: content-disposition:references; b=NV37xaZFP8rbceLse57vnQtLb6SpocgViRVKSZ3t3irXBx1qTrilEXI+Yl3jKCLAG 2N4e6m+J0ibEWduToT2pA== Message-ID: <6599ad830807220556i3c2afc77j2e68d7179830b3ff@mail.gmail.com> Date: Tue, 22 Jul 2008 08:56:04 -0400 From: "Paul Menage" To: "Patrick McHardy" Subject: Re: [PATCH] Traffic control cgroups subsystem Cc: "Ranjit Manomohan" , linux-kernel@vger.kernel.org, netdev@vger.kernel.org, lizf@cn.fujitsu.com In-Reply-To: <4885D72A.1060106@trash.net> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <4885B7DD.7070507@trash.net> <6599ad830807220514m5afea0fcsb99bf2e95f401c36@mail.gmail.com> <4885D72A.1060106@trash.net> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1022 Lines: 26 On Tue, Jul 22, 2008 at 8:48 AM, Patrick McHardy wrote: > > Using skb->priority for classification would be fine, but it would > probably interfere with the default initialization to sk->sk_priority. Well internally we just overloaded sk->sk_priority to be the classid, and let skb->priority be inherited from that. > >> Regarding skb->sk->sk_cgroup_classid, is it always the case that the >> original sk is still available when we're making traffic control >> decisions? I'd thought that there were cases (e.g. cloning skbs in the >> TCP retransmit path) where the pointer to the original sk is lost. > > After cloning, TCP sets the owner of the skb to the socket, so > that should work fine. > OK, so maybe we don't need a per-skb field after all. Paul -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/