On Fri, 2006-02-24 at 11:34 -0800, Sean Hefty wrote:
> Tom Tucker wrote:
> > +struct iw_cm_verbs;
> > struct ib_device {
> > struct device *dma_device;
> >
> > @@ -840,6 +844,8 @@
> >
> > u32 flags;
> >
> > + struct iw_cm_verbs* iwcm;
> > +
>
> Does anyone object to adding this to ib_device? I'm not thrilled about this,
> but I don't see another alternative, and I'm not sure it's any worse than
> having
> a 'process_mad' function.
>
> Maybe we need a more generic way of providing transport/device specific
> extensions? Something like:
>
> struct ib_device {
> ...
> union {
> struct iw_verbs *iw;
> struct ib_verbs *ib;
> } ext_verbs;
I like this... It is consistent with the CMA as well.
> ...
> };
>
> Thoughts?
>
> - Sean
_______________________________________________
openib-general mailing list
[email protected]
http://openib.org/mailman/listinfo/openib-general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
