www.delorie.com/archives/browse.cgi   search  
Mail Archives: djgpp/1998/04/21/08:38:39

Message-ID: <4D61B534C6B1D011B21400805FBEB42F61D616@pca03.pca.ml.com>
From: "Castro, Edison M. (PCA)" <edison_castro AT ml DOT com>
To: "'Nate Eldredge'" <nate AT cartsys DOT com>,
"Castro, Edison M. (PCA)"
<edison_castro AT ml DOT com>, djgpp AT delorie DOT com
Subject: RE: Problem with biosdisk
Date: Tue, 21 Apr 1998 08:37:06 -0400

Nate, the v1.x version works in these machines, and I really certain that
the called to int 13h function 8 is the failure point.

Another twist to the story: If I use the version that fails (v2.x) and I
disable emm386.exe from being loaded on my config.sys, suddenly everything
works as expected. So this tells me that the problem is interaction between
the new v2.x DMPI server cwsdpmi or cwsdpr0 and emm386 in these machines
(Compaq Deskpro 6000-PII 333 two 4GB Drives and Dell GXPRO -same config)

My config.sys is:

device=a:\dos\himem.sys 
device=a:\dos\emm386.exe noems
buffers=15
lastdrive=z
dos=high,umb
device=a:\dos\ifshlp.sys

-----Original Message-----
From: Nate Eldredge [mailto:nate AT cartsys DOT com]
Sent: Monday, April 20, 1998 9:11 PM
To: Edison M. Castro; djgpp AT delorie DOT com
Subject: Re: Problem with biosdisk


At 10:02  4/20/1998 -0400, Edison M. Castro wrote:
>I have written an fdisk replacement program to manage disk partitions. This
>program works without problem when compiled with djgcc v1.x.
>
>I decided to migrate it to version 2.x, I compile it , put it in a dos
>floppy, boot up the machine and when my code executed in some machines, it
>frozed.

Does the version compiled with v1 work on these machines? Because otherwise
it is possibly a BIOS bug.

>
>I can not, for the life of me, discern the possible causes for this
problem.
>It is not tie up to any special type of hardware or processor.
>
>I have replaced the call to biosdisk with __int86x, __dpmi_int, __go32_int,
>etc with the same results. I think I have found some type of problem with
>the dpmi host or something like that.

You are *certain* that the biosdisk Get Parameters is what fails? If you
actually write to the disk, that could do bad things (like deleting the
partition on which lives the DPMI swap file). biosdisk function 8 certainly
works on my machine.

Nate Eldredge
nate AT cartsys DOT com


- Raw text -


  webmaster     delorie software   privacy  
  Copyright © 2019   by DJ Delorie     Updated Jul 2019