Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Sat, 18 Nov 2000 12:12:16 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Sat, 18 Nov 2000 12:11:57 -0500 Received: from ha1.rdc2.mi.home.com ([24.2.68.68]:30458 "EHLO mail.rdc2.mi.home.com") by vger.kernel.org with ESMTP id ; Sat, 18 Nov 2000 12:11:25 -0500 Message-ID: <3A16B0CA.E48EFFE7@didntduck.org> Date: Sat, 18 Nov 2000 11:39:38 -0500 From: Brian Gerst X-Mailer: Mozilla 4.75 [en] (X11; U; Linux 2.4.0-test11-pre5-mm i686) X-Accept-Language: en MIME-Version: 1.0 To: Tigran Aivazian CC: Linus Torvalds , Linux kernel mailing list Subject: Re: [PATCH] x86 mm init cleanup In-Reply-To: Content-Type: multipart/mixed; boundary="------------6BD51D271D7D7A9F9E44AC88" Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org This is a multi-part message in MIME format. --------------6BD51D271D7D7A9F9E44AC88 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Tigran Aivazian wrote: > > On Sat, 18 Nov 2000, Brian Gerst wrote: > > > Patch against test11. This patch moves the setting of %cr4 out of the > > loops and makes the code a bit more readable. Tested with standard > > pagetables, PSE, and PAE. > > > > > > Brian, > > while you were there, so close to paging_init() why not also correct the > wrong comment at the top of the function? It talks about 0-4M pagetables > whereas we really setup (see head.S) 0-8M. > > Regards, > Tigran Add this little patch then -- Brian Gerst --------------6BD51D271D7D7A9F9E44AC88 Content-Type: text/plain; charset=us-ascii; name="mminit2.diff" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="mminit2.diff" diff -urN linux-2.4.0t11p7/arch/i386/mm/init.c linux/arch/i386/mm/init.c --- linux-2.4.0t11p7/arch/i386/mm/init.c Mon Oct 23 17:42:33 2000 +++ linux/arch/i386/mm/init.c Sat Nov 18 11:38:02 2000 @@ -437,7 +437,7 @@ } /* - * paging_init() sets up the page tables - note that the first 4MB are + * paging_init() sets up the page tables - note that the first 8MB are * already mapped by head.S. * * This routines also unmaps the page at virtual kernel address 0, so --------------6BD51D271D7D7A9F9E44AC88-- - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/