I have checked the etherlink model and etherint model which has an etherint peer. Once the peer is set, two etherints on both ends of the etherlink are binded together. So a direct point-to-point link is set. As you said, I can just create a switch model with some interfaces and make each etherint bind with a switch interface instead of the etherints on other systems. However, this network topology is a kind of star network which is different from the fully-connected network because the etherint on one system is shared to connect with other systems, sometimes causing the transmission delay.
I think each system should have N-1 etherints in order to implement a fully-connected network of N systems. How do you think of that? I know there is a limit on PCI devices. Do you think it possible to modify this limit? If yes, where should I start with? Is this about the pci bus address limit? Xin ---- Original message ---- >Date: Sat, 10 Nov 2007 11:45:32 -0500 (EST) >From: Nathan Binkert <[EMAIL PROTECTED]> >Subject: Re: [m5-users] simulating multicomputer system with M5 >To: M5 users mailing list <m5-users@m5sim.org> > >> Thanks for your reply. The number of systems on the network >> is less than 20. Could you provide a sample for the switch >> model and the configuration for the network issue(e.g. fs.py)? > >I think the best you'd do with the fully connected network is 7. I can't >remember what our limit on PCI devices is, but it's something like that. >YOu may be able to "share" pci slots and share irqs and but use different >PCI functions for multiple NICs. That works in real systems, but I >haven't ever needed to do that. > >As for providing a switch model, sorry, I don't have the time right now. >You could certainly start with the etherlink model to see how it models a >wire, and just create a new model that has many interfaces with some >switching logic in the middle. It really wouldn't be that hard. (And if >you do go through the trouble, please contribute the code back to the M5 >team.) > > Nate >_______________________________________________ >m5-users mailing list >m5-users@m5sim.org >http://m5sim.org/cgi-bin/mailman/listinfo/m5-users _______________________________________________ m5-users mailing list m5-users@m5sim.org http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
