From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stephen Hemminger Subject: Re: [PATCH] gso: Handle Trans-Ether-Bridging protocol in skb_network_protocol() Date: Wed, 8 May 2013 09:23:13 -0700 Message-ID: <20130508092313.677f9717@nehalam.linuxnetplumber.net> References: <1367995267-23443-1-git-send-email-pshelar@nicira.com> <518A01F9.8090403@linux.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: Pravin B Shelar , netdev@vger.kernel.org, Eric Dumazet , Cong Wang , "David S. Miller" To: PJ Waskiewicz Return-path: Received: from mail-pb0-f47.google.com ([209.85.160.47]:33084 "EHLO mail-pb0-f47.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755526Ab3EHQ2U (ORCPT ); Wed, 8 May 2013 12:28:20 -0400 Received: by mail-pb0-f47.google.com with SMTP id uo1so1332174pbc.34 for ; Wed, 08 May 2013 09:28:19 -0700 (PDT) In-Reply-To: <518A01F9.8090403@linux.intel.com> Sender: netdev-owner@vger.kernel.org List-ID: On Wed, 08 May 2013 00:42:49 -0700 PJ Waskiewicz wrote: > On 5/7/2013 11:41 PM, Pravin B Shelar wrote: > > Rather than having logic to calculate inner protocol in every > > tunnel gso handler move it to gso code. This simplifies code. > > Moving code that isn't used on a regular basis to hotpath for most > networking to simplify code isn't always a good thing. > > Have you run tests on non-tunneled streams using GSO to make sure there > isn't an adverse impact to performance due to the new, untaken branch? > > -PJ > -- > To unsubscribe from this list: send the line "unsubscribe netdev" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html Maybe use unlikely here? It looks like more and more of this tunnel unwrapping needs to be handled here (vxlan? ipip?) probably want more general switch statement and per-protocol functions.