
	Are you having trouble locating the culprit who is zapping
	an executive or pool space location ? Sometimes, by the time
	you get an XDT or crash, the program, driver or directive
	which caused the problem is long gone. If you happen to have 
	SPM support in your SYSGEN, you can possibly track the problem
	down a little easier. 

	If the location being zapped is consistent, you can monitor
	its contents by patching the SPM hook to branch off and check if it changed.
	The SPM hook is dispatched for almost every executive function
	(directives, interrupts, etc.) When SPM is turned off, the dispatch
	hook ($HKEXT) points to essentially a return sequence. When SPM
	is turned on, $HKEXT points to a lower location in the SPM executive
	routine which tracks the function being processed. 
	All we have to do is to patch in a monitoring routine just above the
	hook return and point $HKEXT to the start of this routine. 
	The following is an example of a monitoring routine to check for
	a location being changed:


		actual instruction		octal equivalent
		CMP	13276,#66044		022737,013276,066044		; see if location has changed
		BEQ	.+1			001401
		BPT				000003				; yes, XDT so we can look at it
;										  here we continue with normal SPM return code.




	open the following locations in listed order to monitor
	locations getting zapped. 

location	from		to					actual instruction
066016		132540		022737		(location #1 check) 	CMP	AAAAAA,#BBBBBB
066020		000006		BBBBBB		(location #1 expected contents)
066022		017646		AAAAAA		(location #1 address)
066024		000006		001005					BNE	.+5
066026		004736		022737		(location #2 check)	CMP	CCCCCC,#DDDDDD
066030		012667		DDDDDD		(location #2 contents)
066032		104336		CCCCCC		(location #2 address)
066034		012667		001001					BNE	.+1		
066036		105414		000003		(breakpoint)		BPT	
						normal return code follows

001326		066040		066016		(hook dispatch address)






