From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ruby.spiritone.com (ruby.spiritone.com [216.99.193.130]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTP id 85F00DDE3C for ; Sun, 12 Aug 2007 01:30:11 +1000 (EST) Received: from [192.168.0.4] (216-99-213-225.dsl.aracnet.com [216.99.213.225]) (authenticated bits=0) by ruby.spiritone.com (8.13.6/8.13.4) with ESMTP id l7BFMaHq012057 for ; Sat, 11 Aug 2007 08:22:37 -0700 Message-ID: <46BDD447.5040903@BitWagon.com> Date: Sat, 11 Aug 2007 08:22:47 -0700 From: John Reiser MIME-Version: 1.0 To: linuxppc-embedded@ozlabs.org Subject: Re: Is it possible to move Kernel Virtual Space to 0xE0000000 References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1 List-Id: Linux on Embedded PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , > Giving Kernel virtual space to 0xD0000000 and it boots without any problems, > however when I increase it more to 0xE0000000 it throws Program Exception. I > will be glad if someone helps me out with refernces for Linux Porting on > MPCs. Search for: linux "4G/4G split" This work was active a couple years ago on x86. The kernel patches are not trivial, and they cost about 15% in performance of system calls. But they do provide almost a full 32-bits of address space for user mode. The problem of "not enough user address space" also is solvable with money. Get a 64-bit machine. It costs more per board but saves development time, so costs less in the long run unless you are making more than 10,000 systems. --