On Thu, 2018-01-11 at 13:27 +0200, Moni Shoua wrote: > Maybe I am missing something but I think that the race is when qp is > in ERROR state and the following functions run in parallel > * rxe_drain_req_pkts (called from rxe_requester after post_send) > * rxe_drain_resp_pkts (called from rxe_completer after modify to ERROR) > > Am I right? Hello Moni, I think that's a real race and a race that has to be fixed but not the race that caused the missing completions in the tests I ran myself. Best regards, Bart.