From mboxrd@z Thu Jan 1 00:00:00 1970 From: Steve Wise Subject: Re: [PATCH 2/3] cxgb4: function namespace cleanup (v2) Date: Sat, 16 Oct 2010 22:30:16 -0500 Message-ID: <4CBA6DC8.4010104@opengridcomputing.com> References: <20101015224309.179898823@vyatta.com> <20101015224523.633775810@vyatta.com> <20101015171057.409db0f2@nehalam> <4CB8FBCE.3090401@chelsio.com> <20101015212310.47a74d86@nehalam> <4CB9434D.1060702@chelsio.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Stephen Hemminger , Divy Le Ray , "David S. Miller" , Casey Leedom , netdev@vger.kernel.org To: Dimitris Michailidis Return-path: Received: from smtp.opengridcomputing.com ([209.198.142.2]:59079 "EHLO smtp.opengridcomputing.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751260Ab0JQDaL (ORCPT ); Sat, 16 Oct 2010 23:30:11 -0400 In-Reply-To: <4CB9434D.1060702@chelsio.com> Sender: netdev-owner@vger.kernel.org List-ID: On 10/16/2010 1:16 AM, Dimitris Michailidis wrote: > Stephen Hemminger wrote: >> On Fri, 15 Oct 2010 18:11:42 -0700 >> Dimitris Michailidis wrote: >> >>> Stephen Hemminger wrote: >>>> Make functions only used in one file local. >>>> Remove lots of dead code, relating to unsupported functions >>>> in mainline driver like RSS, IPv6, and TCP offload. >>> Thanks, this looks OK. One exception, cxgb4_get_tcp_stats was >>> intended to be used by the rdma driver. I see that driver doesn't >>> call it presently but if you don't mind can we give Steve a few >>> hours to tell us if he has any imminent plans to use it. If he >>> doesn't offer to do something to use it for .37 it goes. >> >> The kernel source tree is not your development place holder tree. >> At least #ifdef the code out for now. > > I am trying to protect Stephen Rothwell's time by checking that the IB > folks don't plan to add a call to this in their tree while we remove > the function in net-next. There's supposed to be a call in the IB > driver. I don't know why there isn't one or whether they are planning > to fix it for .37. I see the potential for a linux-next conflict and > I am trying to avoid it. #ifdef doesn't help, if it's not needed we > can remove it for good. I'll add a patch this week to utilize the tcp stats. Steve.