Date:	Mon, 7 Apr 1997 13:05:48 -0600 (MDT)
From:	Ramesh Nallur <nalluri AT nmt DOT edu>
To:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
cc:	djgpp AT delorie DOT com
Subject:	Re: Interrupts (Hardware)
In-Reply-To:	<Pine.SUN.3.91.970407133910.5175F-100000@is>
Message-ID:	<Pine.SUN.3.95.970407120356.24230A-100000@rainbow>
MIME-Version:	1.0

First I want to thank you for replying.  

> It would help if you also tell why do you need to know that.  This
> issue is immensly complex and full of gory details; knowing what
> practical use do you have in mind will help to present manageable
> amount of info without losing important details.
> 
I am doing my  masters in  CS and currently I am  working on an OS project
and I want to write a server similar to the DPMI host. For this I started
looking at the cwsdpmi code, but lot of things confused me. Hence the
query. I really want to know all the gory details of interrupt handling
and switching between modes. I need to know the setup of the IDT etc...  

> The IDT is not ``replaced'', at least not necessarily.  The DPMI host
> usually changes some entries in the IDT to point the hardware
> interrupts to its own handlers,
If you look at the code in CWSDPMI, there is a whole setup for the PM. The
host sets up the IDT, GDT and switches modes (RM to PM) and then runs your
program. (Ofcourse the RM IDT has to be replaced to switch to PM).

> but AFAIK this is so it could reissue
> the interrupts to real-mode handlers if the DPMI client (the DJGPP
> program you run) didn't install its own handler.  There is no general
Yes, thats what I thought. Let me repeat what you just said, in PM if
there is any interrupt, the DPMI host checks if there is any PM
interrupt handler installed. If its not then switches to RM and executes
the RM handlers. So I logically assumed that in PM if I replace an IDT
entry by any of my own handker then an interrupt should be caught
there . For eg the clock
interrupt occurs at vector number 8, so replacing the IDT at vector 8
(base+8*8,base+8*8+4) should catch the clock interrupts. Well if PIC
has been programmed and its not 8 then it should have been caught some
where. So if I replace all the 256 entries then somewhere
the clock interrupt should have been caught. I am talking about only PM
here. Am I missing something very obvious here??.
Additionally I would
make sure that my interrupt handler does switch modes and run the RM
handler. 

> > Why cant I just overwrite an IDT entry in PM and expect interrupts to be
> > caught?
> 
> I think if you do that, the reflection of the interrupts from RM might
> be broken.  Using the appropriate DPMI function makes your program
> play by the rules.
> 
> Why would you need to do that, anyway?
> 
Write my own server and also I am very curious to know why when I replace
an IDT entry in PM is the hardware interrupt not caught (in PM). 

> This is the entire code of the function that sets PM interrupt
> handler.  Where do you see anything that's connected with RM handlers
> or 0000:0000?
> 
>     #define USE_EBX
>     #include "dpmidefs.h"
> 
> 	    .text
> 
> 	    FUNC(___dpmi_set_protected_mode_interrupt_vector)
> 	    ENTER
> 
> 	    movb	ARG1, %bl
> 	    movl	ARG2, %eax
> 	    movl	(%eax), %edx
> 	    movw	4(%eax), %cx
> 	    DPMI(0x0205)
> 	    xorl	%eax,%eax
> 
> 	    LEAVE

I should have been more specific here. Well, I was talking about the code
for DPMI(0x205) and not the API. Basically the handler for int 31 and
ax=0205. 

Your answers have cleared all of my doubts except for one. Why can't I
replace an IDT (PM) entry and expect the hardware interrupts to be
caught(in PM only)???. 

Thanks a lot.

- Raw text -