Sean Hefty wrote: > Caitlin Bestler wrote: >> How about something along the lines of stating that the transport >> specific CM is responsible for translating 0 to a number that does >> not conflict with the host stack, and then leave the details to be >> implemented as required. > > That doesn't quite work. If there are multiple transports, > you want the same translation. > > - Sean
I can see how an offload device can share an IP address with the host stack and hence need to prevent conflicting uses of the TCP port. But how do *two* offload devices share the same IP address with each other and/or with the host stack? And if "0" means "don't care - pick something", why is there a need for it to be consistent/predictable? _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
