[openib-general] matching an RMPP NACK to send or receive?

[Hal Rosenstock]

Hi Sean,

On Thu, 2005-07-14 at 17:01, Sean Hefty wrote:
> Eitan Zahavi wrote:
> > EZ: TID should consist of a client selected part and process ID. For kernels
> > I would select some part of the GUID too.
> > We need to minimize cases of same TID use. But in such case - abort both of
> > them.
> 
> I don't think that the receive handling code should make any assumptions 
> about the format of the TID.  

Agreed.

> Aborting both transactions seems a little 
> harsh, but not sure if there's another way.

I don't see another way either :-( All that could be done is to narrow
the window by checking on initiation of a send transaction that there is
not a matching TID in progress but this is not worth it IMO.

> For now, the code will only 
> abort a send, which should be the more common case.  I will need to add 
> aborting receives later and assume that actual TID collisions would be 
> extremely rare.

What status code will you use for this ?

> The RMPP status does help indicate whether a send or receive is being 
> aborted, but not in all cases.

What cases are not covered ?

-- Hal

> 
> - 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