Sean Hefty wrote: > Or Gerlitz wrote: >> + * In the case of error, a reject message is sent to the remote side >> and the >> + * state of the qp associated with the id is modified to error, such >> that any >> + * previously posted receive buffers would be flushed. > > Hmm... this makes me question whether this is what it should be doing. > Is there any reason not to reject the connection if accept fails?
I think this (sending REJ, modifying the QP to ERROR) is exactly what it should be doing. Why would someone count/expect that a REJ would not be sent in this case? Even if for some reason which i don't see now we will do some change here (eg let the ULP send the REJ etc), lets have this patch which document what we have now merged for 2.6.19. Or. _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
