I guess I answered my own question on this one. W15 is the stack 
pointer. By initializing this to 0 the chip always resets.
> Another question. Why do they leave out W15? I have been trying to get 
> external INT0 working on this chip.
> All associated regs set correctly etc., but when the interrupt occurs, 
> the PIC goes to the reset vector. The docs say that
> any modification of an uninitialized W reg will cause a reset. Well 
> the docs also say that the interrupt handler uses W15 as
> part of the context save. Should the below code be modified to include 
> W15?
>>   W0 thru W15 are memory locations 0..31.
>>    
>>   The above loop clears the memory locations associated with W1 thru W13
>>    
>>   How, you ask?
>>    
>>   CLR W0             W0<-0
>>   MOV W0,W14   W14<-0
>>   REPEAT #12     Repeat next instruction 13! times
>>   MOV W0,[++W14] Does the following:
>>       Adds 2 to W14
>>       Moves W0 (which is zero) to the memory location pointed to by W14
>>   CLR W14            W14<-0
>>    
>>    
>>   --- Bob Ammerman
>>   RAm Systems
>>    
>>   >> 
>>   
>>> It prevents buggy code from generating an uninitialized W register
>>> reset, which is caused by using an uninitialized W register as an
>>> address. This code pretty much defeats the whole point of the
>>> uninitialized W register reset feature.
>>>
>>> --
>>> John W. Temples, III
>>>
>>>     
>> But this is only doing something to W0 and W14...or am I wrong? I'm 
>> really trying to
>> get my head around this!
>>
>>   
>
>
> -- 
> Regards,
> Stephen D. Barnes
>   


-- 
Regards,
Stephen D. Barnes


-- 
Regards,
Stephen D. Barnes

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist