From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCHv2 net-next] xen-netback: remove unconditional __pskb_pull_tail() in guest Tx path Date: Thu, 06 Nov 2014 14:40:57 -0500 (EST) Message-ID: <20141106.144057.541605315482589279.davem@davemloft.net> References: <1415184622-19421-1-git-send-email-david.vrabel@citrix.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, xen-devel@lists.xenproject.org, ian.campbell@citrix.com, wei.liu2@citrix.com, malcolm.crossley@citrix.com To: david.vrabel@citrix.com Return-path: Received: from shards.monkeyblade.net ([149.20.54.216]:57953 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751133AbaKFTk7 (ORCPT ); Thu, 6 Nov 2014 14:40:59 -0500 In-Reply-To: <1415184622-19421-1-git-send-email-david.vrabel@citrix.com> Sender: netdev-owner@vger.kernel.org List-ID: From: David Vrabel Date: Wed, 5 Nov 2014 10:50:22 +0000 > From: Malcolm Crossley > > Unconditionally pulling 128 bytes into the linear area is not required > for: > > - security: Every protocol demux starts with pskb_may_pull() to pull > frag data into the linear area, if necessary, before looking at > headers. > > - performance: Netback has already grant copied up-to 128 bytes from > the first slot of a packet into the linear area. The first slot > normally contain all the IPv4/IPv6 and TCP/UDP headers. > > The unconditional pull would often copy frag data unnecessarily. This > is a performance problem when running on a version of Xen where grant > unmap avoids TLB flushes for pages which are not accessed. TLB > flushes can now be avoided for > 99% of unmaps (it was 0% before). > > Grant unmap TLB flush avoidance will be available in a future version > of Xen (probably 4.6). > > Signed-off-by: Malcolm Crossley > Signed-off-by: David Vrabel Applied, thanks.