[openib-general] Re: [PATCH] RFC Verbs: add support for transport specific verbs

[Tom Tucker]

On Tue, 2006-02-28 at 11:50 -0800, Sean Hefty wrote:
> Michael S. Tsirkin wrote:
> > Is it worth it to separate these things out?
> > Even within IB lots of methods are optional - so why cant an iWarp device just
> > avoid defining process_local_mad, and IB device avoid defining iWarp CM ops?
> 
> There are 7 additional function needed by iWarp.  How should these be added to 
> ib_device?  Using process_mad as an example, we would add all 7 function 
> prototypes directly to ib_device.

... And in fact in the end there will be more. This separation allows
one transport to change without impacting the other.

> 
> Tom's original proposal was to add an iWarp specific pointer to ib_device, with 
> the functions declared as part of a structure referenced by that pointer.
> 
> I'd just like consistency on how transport specific functionality is handled, 
> more than I have a specific preference at this point.

I like Sean's union approach better too. I think my original approach
was aesthetically unpleasant work around (ugly hack). 

> 
> - Sean
> _______________________________________________
> openib-general mailing list
> openib-general at openib.org
> http://openib.org/mailman/listinfo/openib-general
> 
> To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general