Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!clyde!rutgers!husc6!cmcl2!nrl-cmf!ames!elroy!mahendo!jplgodo!wlbr!etn-rad!jru
From: jru@etn-rad.UUCP
Newsgroups: comp.sys.ibm.pc
Subject: Re: TSR interrupt handlers
Message-ID: <304@etn-rad.UUCP>
Date: Thu, 12-Nov-87 18:32:08 EST
Article-I.D.: etn-rad.304
Posted: Thu Nov 12 18:32:08 1987
Date-Received: Sat, 14-Nov-87 16:38:55 EST
References: <4106@watdcsu.waterloo.edu>
Reply-To: jru@etn-rad.UUCP (John Unekis)
Distribution: comp
Organization: Eaton Inc. IMSD, Westlake Village, CA
Lines: 46
Keywords: TSR interrupt handler

In article <4106@watdcsu.waterloo.edu> ryders@watdcsu.waterloo.edu (Steve Ryder                                                                     Independent Studies) writes:
>Anyone experienced with TSR programs and interrupts?
>
>What I am trying to do is create a program to disable the PrtSc. This is
...
>'Print-Screen disabled.' message, unusual characters print to the screen and
>then the machine hangs. Any help would be greatly appreciated. ...
...
>pscreen	proc 	near
>
>	sti
*problem #1 - you are not saving the register you are using on the
	      stack. This will corrupt the values in dx and ax, which
	      may be in the process of being used by someone when
	      the interrupt occurs.

>	mov	dx,offset warning	;interupt request for PrtSc
*problem #2 - the data segemnt register ds probably does not contain the
	      value which you are  expecting it to contain. It will have
	      the value that was being used by whoever was running when
	      the interrupt occurred,maybe DOS, maybe applications code?
	      For code like this you should either refer to everything
	      as offsets from cs which is set up correctly by the 
	      interrupt hardware, or use the large model and load the 
	      segment of warning into ds explicitly (after saving ds
	      on the stack of course).
>	mov	ah,9			;and print 'disabled' message
>	int	21h
>	iret
>
>pscreen	endp
>

Dont you just love segmented architectures? Wouldnt you like to have the
head of the Intel engineer responsible for them on a plate? Life would have
been so nice if ibm had used motorola.
The people who bring you the famous phoenix bios for clone PCs have now 
written a program that emulates the 8086 instruction set as a task on the
motorola 68030, and its faster than the real thing. Wouldnt you like to 
see a 68030 based PS/2 running REAL UNIX with DOS available as an optional
sub-process?
How about it IBM, its not to late to change the PS/2 from a desktop toy
into a real departmental multi-user computer. Or maybe IBM is afraid that
a multi-user PS/2 would smother the infant model 9370 ? (What a joke, a
departmental minicomputer running MVS... 4 users and 4 systems programmers
to keep MVS from eating itself) but I digress.