Date: Sun, 2 May 1999 10:32:21 -0400
Message-Id: <199905021432.KAA05556@envy.delorie.com>
From: DJ Delorie <dj AT delorie DOT com>
To: djgpp-workers AT delorie DOT com
CC: sandmann AT clio DOT rice DOT edu
In-reply-to: <Pine.SUN.3.91.990502114645.563j-100000@is> (message from Eli
	Zaretskii on Sun, 2 May 1999 11:50:02 +0300 (IDT))
Subject: Re: __dpmi_int causes trouble in signals
References:  <Pine DOT SUN DOT 3 DOT 91 DOT 990502114645 DOT 563j-100000 AT is>
Reply-To: djgpp-workers AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


> Can somebody please tell what was the reason for this change in
> __dpmi_int?  (I cannot find anything pertaining to it in the v2.02
> changes' description inside wc202.txi.)

From: sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id: <9712300251 DOT AA14537 AT clio DOT rice DOT edu>
Subject: Re: Win95 problem solved! (again! ;) )
To: 44699 AT ef DOT gc DOT maricopa DOT edu (Joshua James Turpen)
Date: Mon, 29 Dec 1997 20:51:44 -0600 (CST)
Cc: eliz AT is DOT elta DOT co DOT il, dj AT delorie DOT com (DJ at Home)

Thanks for your effort in looking into this.  On a first review your fix
below sounds like a very good idea.  As you've probably noticed, I've been
on vacation and don't have much time in the next few days to look at it
much harder.  I would like to see a bit of testing with this fix - probably
putting it into the next alpha release would be a good idea to see if it
breaks anything.  It certainly fixes something, and your explaination
is very likely.  You get an A+ for investigation on this!

I'm including DJ on this too...

> Regarding ___djgpp_hw_exception and Win95:
> 
> Eureka!  :)  It turns out that Win95 is using your pmode SS for some of
> it's code when building the stack frame on a change from pmode->V86 mode.  
> Below is a very simple, hopefully safe solution to the problem.  Charles,
> please correct my code if it's wrong.
> 
> I haven't checked, but a similar fix may be needed for other functions
> that change from pmode->V86 mode, functions like
> call_real_mode_procedure_with_far_ret_frame, and
> call_real_mode_procedure_with_iret_frame.
> 
> 
> I've tested the code below and it seems to work great, even with the more
> complex test programs. 
> 
> 
> -----------file D0300_Z.S----------------------------------------
> /* Copyright (C) 1995 DJ Delorie, see COPYING.DJ for details */
> #define USE_EBX
> #define USE_EDI
> #include "dpmidefs.h"
> 
>         .comm   ___dpmi_int_ss, 2
>         .comm   ___dpmi_int_sp, 2
>         .comm   ___dpmi_int_flags, 2
>         .lcomm  dpmi_busy, 1
> 
>         .text
> 
>         FUNC(___dpmi_int)
>         ENTER
> 
>         pushl   %es
>       /* Win95 problem:  Add this next line */
>       pushl   %ss
>         movw    ___djgpp_ds_alias, %ax
>         movw    %ax, %es
>       /* Win95 Problem:  Add this next line */
>       /* This should be ok since the app's SS and ___djgpp_ds_alias */
>       /* have the same linear address space */
>       movw    %ax, %ss
>         movl    ARG1, %ebx
>         movl    $0, %ecx
>         movl    ARG2, %edi
>         incb    dpmi_busy
> 
>         cmpb    $1, dpmi_busy
>         jne     L_is_busy
> 
>         movw    ___dpmi_int_flags, %ax
>         movw    %ax, 0x20(%edi) /* r.x.flags */
>         movw    ___dpmi_int_sp, %ax
>         movw    %ax, 0x2e(%edi) /* r.x.sp */
>         movw    ___dpmi_int_ss, %ax
>         movw    %ax, 0x30(%edi) /* r.x.ss */
>         jmp     L_not_busy
> 
> L_is_busy:
>         movw    $0, 0x20(%edi)  /* r.x.flags */
>         movw    $0, 0x2e(%edi)  /* r.x.sp */
>         movw    $0, 0x30(%edi)  /* r.x.ss */
> 
> L_not_busy:
>         DPMI(0x0300)
>         xorl    %eax,%eax
> 
>       /* win95 problem:  Change this next line to pop %ss */
>         LEAVEP(decb dpmi_busy; popl %ss;  popl %es)
> 
> 
>