From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Nicholas Knight" Subject: Re: '@' In blank space on Bochs. Date: Wed, 29 May 2002 08:32:12 -0700 Sender: linux-8086-owner@vger.kernel.org Message-ID: <000d01c20726$01377ac0$6407070a@blue> References: Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: List-Id: Content-Type: text/plain; charset="us-ascii" To: Michael McConnell , Linux 8086 On Wednesday, May 29, 2002, Michael McConnell wrote: > On Wed, 29 May 2002, Nicholas Knight wrote: > > > ELKS apparently writes out blank areas of the screen with 0's > > (arch/i86/drivers/char/dircon.c, #define A_BLANK 0x00). This creates an > > issue with Bochs because, for some reason, Bochs has '@' there instead of > > null (as it happens, MINIX runs into the same problem). This may be specific > > to the Windows version of Bochs, as I've yet to use either MINIX or ELKS in > > Bochs in any other host OS. > > > > Once I get my laptop back (grumble, stupid hdd) I can recompile ELKS to use > > 0x32 (ASCII's "space" character) instead of 0x00 on my personal systems, but > > I thought others might be interested in this. > > Space is 0x20 (decimal 32). 0x32 is '2'. Oops, read the colums in reverse :) > > Maybe a config option? CONFIG_BLANK_BOCHS? > > Are you using the LGPL VGABIOS or the Elpin one? You could try using the > other VGABIOS file. Was using Elpin. Set it to the LGPL, same thing. They're probably shooting for functionaly identicle to the Elpin BIOS, right down to the questionable "features". Or else they don't realize it's non-standard and causing problems, which seems a little odd in itself since it's been pointed out to Bochs maintainers several times by MINIX users. > > Personally, I think making a Bochs config option is not a good move, since as > far as anything in it is concerned, it's a PC complete with its hardware. > Just different hardware to what is currently being used with it. Well, Alan seems to think that it'd be OK to just set it to use 0x20 by default on everything. Speaking as a user (not being a kernel developer), I think it may make more logical sense.