From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S267507AbUHZEod (ORCPT ); Thu, 26 Aug 2004 00:44:33 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S267529AbUHZEod (ORCPT ); Thu, 26 Aug 2004 00:44:33 -0400 Received: from umhlanga.stratnet.net ([12.162.17.40]:19323 "EHLO umhlanga.STRATNET.NET") by vger.kernel.org with ESMTP id S267507AbUHZEoa (ORCPT ); Thu, 26 Aug 2004 00:44:30 -0400 To: jmerkey@comcast.net Cc: linux-kernel@vger.kernel.org, jmerkey@drdos.com Subject: Re: 1GB/2GB/3GB User Space Splitting Patch 2.6.8.1 (PSEUDO SPAM) X-Message-Flag: Warning: May contain useful information References: <082620040421.9849.412D655C000690BA000026792200735446970A059D0A0306@comcast.net> From: Roland Dreier Date: Wed, 25 Aug 2004 21:42:28 -0700 In-Reply-To: <082620040421.9849.412D655C000690BA000026792200735446970A059D0A0306@comcast.net> (jmerkey@comcast.net's message of "Thu, 26 Aug 2004 04:21:48 +0000") Message-ID: <52n00ibjd7.fsf@topspin.com> User-Agent: Gnus/5.1006 (Gnus v5.10.6) XEmacs/21.4 (Security Through Obscurity, linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-OriginalArrivalTime: 26 Aug 2004 04:42:28.0354 (UTC) FILETIME=[17441620:01C48B27] Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org jmerkey> That incredibly useful patch for 2.4.X that Andrea wrote jmerkey> that splits the kernel user space into 1GB/2GB/3GB jmerkey> sections I ported to 2.6.8.1 and posted it to: jmerkey> ftp.kernel.org:/pub/linux/kernel/people/jmerkey/patches/linux-2.6.8.1-highmem-split-08-25-04.patch This is indeed pretty useful. A few comments on your version of the patch: - might as well post a patch this small inline - In Kconfig, what happens if someone turns on highmem? It seems all the USER_XXX choices depend on NOHIGHMEM. also, the config option probably needs at least some help text. - the change to vmlinux.ld.S can be dropped from future versions (Linus merged this post-2.6.8) - why create PAGE_OFFSET_RAW in asm-generic, when it depends on i386-only config symbols and is only used in i386? - what's the reason for the odd rewrite of free_one_pgd()? it looks equivalent (and misindented) Thanks, Roland