Sean wrote: > This is similar in concept to what I have in my latest patch. > A difference is > that your discriminator is located at the start of the > private data, whereas I > was trying to use a mask.
Actually our discrimator is an offset/len into the private data, adding an optional mask to that concept would be useful, this allowed a contiguous portion of the private data to be tested, but it did not need to occur at the start of the private data. > > Did you find a use for listening on the sender GUID? We have not yet used that feature, but having the ability to key off of all the assorted addressing info in the REQ seemed sensible. I'm sure some applications can come up with a use, perhaps as a security feature? Todd Rimmer _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
