From owner-freebsd-hackers Thu Jan 2 02:46:14 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.4/8.8.4) id CAA27922 for hackers-outgoing; Thu, 2 Jan 1997 02:46:14 -0800 (PST) Received: from cheops.anu.edu.au (avalon@cheops.anu.edu.au [150.203.76.24]) by freefall.freebsd.org (8.8.4/8.8.4) with ESMTP id CAA27917 for ; Thu, 2 Jan 1997 02:46:10 -0800 (PST) Message-Id: <199701021046.CAA27917@freefall.freebsd.org> Received: by cheops.anu.edu.au (1.37.109.16/16.2) id AA165621948; Thu, 2 Jan 1997 21:45:48 +1100 From: Darren Reed Subject: Re: bootloader & memory test... To: hans@brandinnovators.com (Hans Zuidam) Date: Thu, 2 Jan 1997 21:45:48 +1100 (EDT) Cc: freebsd-hackers@freebsd.org In-Reply-To: <199612271048.KAA05227@truk.brandinnovators.com> from "Hans Zuidam" at Dec 27, 96 10:48:59 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In some mail from Hans Zuidam, sie said: > > Hi, > > After about a year of sig-11 plague I decided to find exactly what was > wrong with the memory of my system and found a program called MemTest-86 > by Chris Brady. It is the only one that's indeed reporting errors, none > of the DOS based ones found anything! > > To narrow the search space I would like to make some modifications, but > the sources use a modified version of the Linux boot loader. To make > things worse apparently two different assembler syntaxes (AT&T and Intel) > are used and I am not well versed (to say the least) in either. > > The memory test sets the CPU for flat 32 bit addressing and loads itself > to address 0x100. It uses the area from 0x0 to 0xFF as it's stack. Could > someone give me a hand (or pointers) to achieve the same using the FreeBSD > boot loader? Thanks in advance, I haven't delved too deeply, but how hard would it be to have the kernel's idle loop do a memory test ? Darren