[openib-general] MAD receive work completion

[Sean Hefty]

Roland Dreier wrote:
> Now I'm not sure whether it makes sense to change the wc member of
> struct ib_mad_recv_wc from a struct ib_wc * to just a struct ib_wc.
> On the one hand it makes the API slightly cleaner, but on the other
> hand it is an incompatible change that may limit the internal
> implementation of handling MAD receives.

Right now, I'm leaning towards no API change, but changing the 
implementation in the MAD layer to ensure that the ib_wc is valid after 
the callback returns.  This would avoid limiting the MAD layer 
implementation, while also preventing changes to the existing users.

I'll generate a patch for this to clarify the idea.

- Sean