From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH v2 net-next 0/2] net: preserve sock reference when scrubbing the skb. Date: Thu, 28 Jun 2018 22:20:40 +0900 (KST) Message-ID: <20180628.222040.2117056805629988850.davem@davemloft.net> References: <20180627133426.3858-1-fbl@redhat.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: fbl@redhat.com, netdev@vger.kernel.org, eric.dumazet@gmail.com, pabeni@redhat.com, fw@strlen.de, netfilter-devel@vger.kernel.org To: xiyou.wangcong@gmail.com Return-path: Received: from shards.monkeyblade.net ([23.128.96.9]:52558 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752056AbeF1NUp (ORCPT ); Thu, 28 Jun 2018 09:20:45 -0400 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: From: Cong Wang Date: Wed, 27 Jun 2018 12:39:01 -0700 > Let me rephrase why I don't like this patchset: Cong, I don't think you are seeing the situation clearly and I am certainly going to apply this patch series even in the face of your objections. Suggesting that solving the lack of back pressure on a UDP socket caused by this problem by using cgroups or cpu usage controllers is just complete and utter madness.