Re: [pci] about simulation

[random_user@163.com]

    Thank you very much for your kind help!
    But,taking the difference between  P-T-P and sharded bus into 
account, would you give me some advices about how to modify the 
testbench to be used in  a cardbus to wishbone bridge ?
    Is the configuration cycle initiated by the host or one of the MASTER
card?

regards,
wangc


----- Original Message ----- 
From: "Miha Dolenc" <mihad@o... > 
To: <pci@o... > 
Date: Fri, 26 Jul 2002 09:20:02 +0200 
Subject: Re: [pci] about simulation 

> 
> 
> Point to Point means that every configuration cycle the device 
> detects is 
> meant for it, so it has to be accepted. Therefore I suppose, you do 
> not need 
> an IDSEL pin. You should hardwire IDSEL pin of your "CardBus" 
> target to 
> constant 1. 
> 
> Regards, 
> Miha Dolenc 
> 
> ----- Original Message ----- 
> From: <random_user@1... > 
> To: <pci@o... > 
> Sent: Thursday, July 25, 2002 7:59 AM 
> Subject: [pci] about simulation 
> 
> 
> > Hi all, 
> >      I am now working at the testbench and I have met with 
> some 
> > problems. 
> >      1.Why there are two pci_behaviorial_device modules in the 
> > testbench? Isn't one enough? 
> >      2.I am trying to modify the testbench to be used in a 
> CardBus 
> > project.But in the CardBus spec.there is no idsel pin,so when 
> I use this 
> > testbench,an error occured:more than one device drive devsel 
> > simultaneously! When I read the CardBus spec.,I noticed that 
> the 
> > Cardbus is a point to point bus,while PCI is a shared bus.What 
> on earth 
> > does this mean and HOW would this influence the configuration 
> access? 
> > 
> > Thanks! 
> > 
> > regards, 
> > wangc 
> > 
> > 
> 
--
To unsubscribe from pci mailing list please visit http://www.opencores.org/mailinglists.shtml