Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1423061AbXBBBck (ORCPT ); Thu, 1 Feb 2007 20:32:40 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1423063AbXBBBck (ORCPT ); Thu, 1 Feb 2007 20:32:40 -0500 Received: from mga05.intel.com ([192.55.52.89]:43219 "EHLO fmsmga101.fm.intel.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1423061AbXBBBcj (ORCPT ); Thu, 1 Feb 2007 20:32:39 -0500 X-ExtLoop1: 1 X-IronPort-AV: i="4.13,269,1167638400"; d="scan'208"; a="194828187:sNHT275615109" Message-ID: <45C294AA.9000400@linux.intel.com> Date: Fri, 02 Feb 2007 09:32:26 +0800 From: "bibo,mao" User-Agent: Thunderbird 1.5.0.8 (Windows/20061025) MIME-Version: 1.0 To: Frederic Riss CC: Bjorn Helgaas , linux-kernel@vger.kernel.org, artiom.myaskouvskey@intel.com, akpm@osdl.org, Linus Torvalds Subject: Re: [PATCH] EFI x86: pass firmware call parameters on the stack References: <1170183678.21603.73.camel@funkylaptop> <200701301217.47520.bjorn.helgaas@hp.com> <1170189677.21603.102.camel@funkylaptop> <45C1513F.1000706@linux.intel.com> <87d3b2040702010010h7534cc9cte1605e5cecd56025@mail.gmail.com> In-Reply-To: <87d3b2040702010010h7534cc9cte1605e5cecd56025@mail.gmail.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1457 Lines: 32 Frederic Riss wrote: > 2007/2/1, bibo,mao : >> currently x86_64 kernel does not support efi, efi convention comply to >> MS convention. On ia32 parameter is passed on stack, on x86_64 parameter >> is passed by registers but that is different from x86_64 linux >> convention. > > Is an x86_64 EFI firmware required to have 2 entry points, one x86_64 > and one in ia32 mode ? If that's not the case, it means that an i386 > kernel won't be able to work correctly on an EFI powered x86_64 box... > AFAIK x86_64 EFI firmware now has only one entry point, i386 kernel can not boot up with EFI powered x86_64 box. >> How about adding EFIAPI prefix before efi runtime service function, this >> prefix has different definition in different architecture. > > Have you any objection to the second patch I posted? It's not only a > matter of passing arguments, ia64 does quite a lot of things in its > wrapper functions. I can easily imagine that such additional work > could be added in the i386 wrappers if EFI becomes more widely used. > I do not object your second patch if there will be more wrapper functions to be added in i386 efi code. I have to give up my patch:) thanks bibo,mao - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/