From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262130AbUDGJMg (ORCPT ); Wed, 7 Apr 2004 05:12:36 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262351AbUDGJMg (ORCPT ); Wed, 7 Apr 2004 05:12:36 -0400 Received: from mtvcafw.sgi.com ([192.48.171.6]:62069 "EHLO omx3.sgi.com") by vger.kernel.org with ESMTP id S262130AbUDGJKD (ORCPT ); Wed, 7 Apr 2004 05:10:03 -0400 Date: Wed, 7 Apr 2004 02:08:45 -0700 From: Jeremy Higdon To: Andrew Morton Cc: linux-kernel@vger.kernel.org, jbarnes@sgi.com, davidm@hpl.hp.com Subject: Re: 2.6.5-mm2 (build error in arch/ia64/kernel/setup.c) Message-ID: <20040407090845.GA790466@sgi.com> References: <20040406223321.704682ed.akpm@osdl.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20040406223321.704682ed.akpm@osdl.org> User-Agent: Mutt/1.4.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org I get a build error in arch/ia64/kernel/setup.c. This patch fixes it for SN2 machines, but I don't claim it is Correct. In fact, I think it is Wrong. There have been changes to setup_arch(), including, apparently, the elimination of the cmdline_p argument. Unforunately, that argument was not completely purged from the function. platform_setup on SN2 is #define'd to sn_setup, which still takes the argument, but does not use it. So this hack works. I don't know about the other IA64s. I'm not sure exactly what was intended with these changes, so I haven't attempted a Correct patch. jeremy --- arch/ia64/kernel/setup.c.old 2004-04-07 02:00:01.000000000 -0700 +++ arch/ia64/kernel/setup.c 2004-04-07 01:37:15.000000000 -0700 @@ -361,7 +361,7 @@ /* enable IA-64 Machine Check Abort Handling */ ia64_mca_init(); - platform_setup(cmdline_p); + platform_setup((void *) 0); paging_init(); }