[openib-general] Re: User MAD support for cancel MAD

[shaharf]

> > Are you saying to have other forms of MAD like cancel by mgmt_class,
> cancel
> > by SGID and/or SLID in addition to cancel by TID ? We could do this
if
> there
> > is a need. I don't think that need has been demonstrated (at least
yet).
> 
> No - I'm not suggesting that we have more ways to cancel a MAD.  I was
> simply pointing out that we _could_ have cancel_mad work based on TID
+
> mgmt_class + SGID/SLID (which would match uniqueness as defined by the
> spec), rather than just TID.
> 
> I think canceling MADs based on the wr_id is sufficient.
> 
> - Sean

Sean, I totally agrees, I too think canceling mads should be based on
wr_id, but that requires to expose the we_id to the user mode level.
This may overcome some problems that may result using TIDS.

Shahar