Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752134AbdHPQa4 (ORCPT ); Wed, 16 Aug 2017 12:30:56 -0400 Received: from shards.monkeyblade.net ([184.105.139.130]:60720 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751678AbdHPQaz (ORCPT ); Wed, 16 Aug 2017 12:30:55 -0400 Date: Wed, 16 Aug 2017 09:30:54 -0700 (PDT) Message-Id: <20170816.093054.959220457289425520.davem@davemloft.net> To: jasowang@redhat.com Cc: mst@redhat.com, eric.dumazet@gmail.com, netdev@vger.kernel.org, linux-kernel@vger.kernel.org, kubakici@wp.pl Subject: Re: [PATCH net-next V2 1/3] tap: use build_skb() for small packet From: David Miller In-Reply-To: <3b66193a-2df1-0191-0785-20123e38460a@redhat.com> References: <20170816065837-mutt-send-email-mst@kernel.org> <3b24805d-b489-2dfc-f930-0518ba1a6ea0@redhat.com> <3b66193a-2df1-0191-0785-20123e38460a@redhat.com> X-Mailer: Mew version 6.7 on Emacs 25.2 / Mule 6.0 (HANACHIRUSATO) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.5.12 (shards.monkeyblade.net [149.20.54.216]); Wed, 16 Aug 2017 09:30:54 -0700 (PDT) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 349 Lines: 11 From: Jason Wang Date: Wed, 16 Aug 2017 17:17:45 +0800 > It looks like full page allocation just produce too much stress on the > page allocator. > > I get 1.58Mpps (full page) vs 1.95Mpps (page frag) with the patches > attached. Yes, this is why drivers doing XDP tend to drift towards implementing a local cache of pages.