Re: [Gems-users] Virtual Networks in MOESI_CMP_Directory

[Mike Marty <mikem@xxxxxxxxxxx>]

Yes, I think you can remove them.

--Mike



> We have seen that in the MOESI_CMP_Directory protocol
> foo and goo message buffers are defined and also their
> corresponding IN_ports and OUT_ports. But these ports
> don't produce any event or change of state in the
> protocol and they are never used by any action.
>
> In the L1 memory:
>
>   in_port(goo_in, RequestMsg, goo) {
>     if (goo_in.isReady()) {
>       peek(goo_in, RequestMsg) {
>         assert(false);
>       }
>     }
>   }
>
> In the directory:
>
>   in_port(foo1_in, ResponseMsg, foo1) {
>
>   }
>
> Does that mean that we can remove them?
>
> Thank you
>
> Marco and Enric
>
>
>  --- Enric Herrero <enricherrero@xxxxxxxx> escribió:
>
> > We are studying the use of the virtual networks in
> > the MOESI_CMP_Directory protocol. We saw that in
> > this protocol there are two message buffers called
> > 'foo' and 'goo'. We don't understand which purpose
> > have and between which components are they
> > connected. They are defined in the L1 object using
> > the VN(1) and in the directory using VN(0). Why?
> >
> >  Also in the network parameters we assume that link
> > bandwith is specified in bytes/(1000*time_cycle) and
> > link latency in cycles. Is this correct?
> >
> >  Thank you
> >
> >  Marco and Enric
>
>
>
> ______________________________________________
> LLama Gratis a cualquier PC del Mundo.
> Llamadas a fijos y móviles desde 1 céntimo por minuto.
> http://es.voice.yahoo.com
> _______________________________________________
> Gems-users mailing list
> Gems-users@xxxxxxxxxxx
> https://lists.cs.wisc.edu/mailman/listinfo/gems-users
>