From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262767AbTLKFtq (ORCPT ); Thu, 11 Dec 2003 00:49:46 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264260AbTLKFtq (ORCPT ); Thu, 11 Dec 2003 00:49:46 -0500 Received: from mail-06.iinet.net.au ([203.59.3.38]:56248 "HELO mail.iinet.net.au") by vger.kernel.org with SMTP id S262767AbTLKFto (ORCPT ); Thu, 11 Dec 2003 00:49:44 -0500 Message-ID: <3FD80543.8040905@cyberone.com.au> Date: Thu, 11 Dec 2003 16:48:51 +1100 From: Nick Piggin User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030827 Debian/1.4-3 X-Accept-Language: en MIME-Version: 1.0 To: Raul Miller CC: William Lee Irwin III , Ed Sweetman , Donald Maner , linux-kernel@vger.kernel.org Subject: Re: PROBLEM: Linux 2.6.0-test11 only lets me use 1GB out of 2GB ram. References: <3FD7FCF5.7030109@cyberone.com.au> <3FD801B3.7080604@wmich.edu> <20031211054111.GX8039@holomorphy.com> In-Reply-To: <20031211054111.GX8039@holomorphy.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org William Lee Irwin III wrote: >On Thu, Dec 11, 2003 at 12:33:39AM -0500, Ed Sweetman wrote: > >>I thought highmem wasn't necesarily needed for memory <=2GB? Highmem >>incurs some performance hits doesn't it and so the urge to move to it >>with only 2GB is not very attractive. Anyways i'm just interested in if >>that's the case or not since 2GB is easy to get to these days and i had >>heard that highmem could be avoided passed the 1GB barrier. >> > >You're probably thinking of 2:2 split patches. > >2:2 splits are at least technically ABI violations, which is probably >why this isn't merged etc. Applications sensitive to it are uncommon. > >Yes, the SVR4 i386 ELF/ABI spec literally mandates 0xC0000000 as the >top of the process address space. > At any rate, Raul, highmem shouldn't hurt your performance significantly with the 2.6 kernel. If it does then send a note to the list. Your other options are a different user/kernel split, or a 64-bit kernel, both of which should have less overhead than highmem.