From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1762483AbYEPADu (ORCPT ); Thu, 15 May 2008 20:03:50 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755934AbYEPADm (ORCPT ); Thu, 15 May 2008 20:03:42 -0400 Received: from relay2.sgi.com ([192.48.171.30]:52670 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1755350AbYEPADl (ORCPT ); Thu, 15 May 2008 20:03:41 -0400 Date: Thu, 15 May 2008 19:03:40 -0500 From: Paul Jackson To: "Huang, Ying" , "H. Peter Anvin" , "Andi Kleen" , "Ingo Molnar" Cc: akpm@linux-foundation.org, tglx@linutronix.de, steiner@sgi.com, travis@sgi.com, linux-kernel@vger.kernel.org Subject: Re: [PATCH 10/10] x86 boot: add code to add BIOS provided EFI memory entries to kernel Message-Id: <20080515190340.b232ca7d.pj@sgi.com> In-Reply-To: <20080514151558.148486.91282.sendpatchset@jackhammer.engr.sgi.com> References: <20080514151504.148486.32236.sendpatchset@jackhammer.engr.sgi.com> <20080514151558.148486.91282.sendpatchset@jackhammer.engr.sgi.com> Organization: SGI X-Mailer: Sylpheed version 2.2.4 (GTK+ 2.12.0; i686-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Ying Huang, H. Peter Anvin, Andi Kleen, Ingo Molnar (and anyone else interested): I invite your review of these patches, especially 7, 9 and 10 of the set of 10, that I posted yesterday. In combination, I believe that they allow passing more than E820MAX (128) memory map entries from EFI firmware to a booting kernel, beyond the E820 constraints of non-EFI legacy BIOS's, without changing the EFI interface interface. My understanding of the EFI interface is that it already allows for this, and that we just needed to add the ability, as done in these patches, for the kernel to make use of this. The separate work of Ying Huang, to leverage Andi's reserve_early() and to revise the EFI interface, to pass multiple pages of information from the EFI firmware to the kernel at boot is, I presume, still needed for other extensions, such as for passing more hard disk drive entries. But the particular extension of interest to me, passing more memory map entries (for future x86_64 systems with more than 128 memory nodes) can I believe be done fairly easily, with these patches, with no need of revising the EFI interface or reserving early memory, and with code that is compatible across both x86_32 and x86_64 arch's. -- I won't rest till it's the best ... Programmer, Linux Scalability Paul Jackson 1.940.382.4214