Sounds good, thanks!
Quoting Niket Agarwal <niketa@xxxxxxxxxxxxx>:
> I don't think you need to change the standard_SMP-protocol.sm to
> standard_CMP-protocol.sm in the .slicc file.
> The SMP protocol simulates a multi-chip protocol. You can hack it to
>
> be a single CMP private L2 protocol by
> still simulating the protocol with multiple chips, but lowering the
>
> network latencies between the chips.
>
> You might also want to look at adding directory caches at the
> directory node. Otherwise, all directory access incur the expensive
>
> memory latency.
>
> Cheers,
> Niket
>
> On Jan 15, 2009, at 1:37 PM, hb166307@xxxxxxxx wrote:
>
> > Hello users,
> >
> > I have been reading through some of the threads about creating a
> CMP
> > system
> > with a private L1 & private L2 configuration that I should use the
> > MOESI_SMP_directory and change the latencies. I have also seen that
>
> > I should
> > add standard_CMP-protocol.sm in place of standard_SMP-protocol.sm
> in
> > MOESI_SMP_direcory.slicc! Can anybody help to verify any of this?
> >
> > Thanks!
> > _______________________________________________
> > Gems-users mailing list
> > Gems-users@xxxxxxxxxxx
> > https://lists.cs.wisc.edu/mailman/listinfo/gems-users
> > Use Google to search the GEMS Users mailing list by adding
> "site:https://lists.cs.wisc.edu/archive/gems-users/
> > " to your search.
> >
>
> _______________________________________________
> Gems-users mailing list
> Gems-users@xxxxxxxxxxx
> https://lists.cs.wisc.edu/mailman/listinfo/gems-users
> Use Google to search the GEMS Users mailing list by adding
> "site:https://lists.cs.wisc.edu/archive/gems-users/" to your search.
>
>
|