Re: [partial-reconfig] Bus macro problem

[Sean Durkin <durkinsn@xxxxxxxxxxxxxxxxx>]

John Williams wrote:
> Hi John,
> 
> jedward@lme.usp.br wrote:
> 
>      I have a doubt about this problem. If I use a hard bus macro to
>      communicate static modules A and C, while B is reconfiguring. How 
> can
>      i be sure that a hard bus macro inside module B is not part of
>      reconfiguration frame?. Is it possivel? or always that I remove the
>      module B, a hard bus macro across the module B is removed too?
> 
> 
> My understanding is that the hard macro *will* be part of the 
> reconfiguration frame, however since the reconfiguration is 
> "glitch-free", and the original and new configuration for that resource 
> are the same, the communications between the two modules should be 
> uninterrupted.  I've never tried it, but that's the theory.
Exactly: All nets that have the same routing before and after 
reconfiguration will stay intact during reconfiguration. I.e. they will 
not be disconnected during reconfiguration and wired up identically 
afterwards but will remain uninterrupted the entire time. And this 
actually works as promised.

The problem is to force the tools (par, in this case) to use a specific 
routing for a specific net, as there are currently no constraints for 
this. At the moment bus macros are the only way to do this, although 
theoretically there are other possibilities.

When you remove "module B", the bus macro will not be removed, since it 
actually belongs to the top level of the design, not to the module.

-- 
Best regards,

Sean Durkin
Fraunhofer Institute for Integrated Circuits IIS
Dep. Electronic Imaging - HIS (Bildsensorik - HIS)
Am Wolfsmantel 33
91058 Erlangen
Germany

Fon: +49 9131 776 503
Fax: +49 9131 776 598

Web: http://www.iis.fraunhofer.de
mailto:durkinsn@iis.fraunhofer.de
___________________________
partial-reconfig mailing list
partial-reconfig@itee.uq.edu.au
Mailing List Archive : http://www.itee.uq.edu.au/~listarch/partial-reconfig/