[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Conflicting Priorities in occam



On Thu, 4 Mar 1999, Adrian Lawrence wrote:

> Specifying the meaning of SKIP guards
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> Does this pseudo-occam capture what we mean by SKIP guards in PRI ALT?
> 
> [N]CHAN skip:
> PAR
>   PAR FOR i = 1 TO N                     PRI ALT FOR i = 1 TO N
>     skip[i] !                              SKIP
>   PRI ALT FOR i = 1 TO N        =            P[i]
>     skip[i] ?
>       P[i]

I don't think so.  The lack of a PRI on either of the PARs allows any
relative ordering between the output commits and the PRI ALT, so we get
           
	 _ N
	| |   P(i)
	 i=1

Denis A Nicole                      WWW:   http://www.hpcc.ecs.soton.ac.uk/~dan 
High Performance Computing Centre   Email: dan@xxxxxxxxxxxxxxx                  
University of Southampton           Phone: +44 1703 592703                      
UK.                                 Fax:   +44 1703 593903