> Quoting Roland Dreier <[EMAIL PROTECTED]>: > Subject: Re: [PATCHv3 for-2.6.21] IB/mthca: fix race in QP destroy > > > Yes but here we also must make sure completion events and async events are > > flushed out: once QP is in reset no events should be generated. > > Completion events are fine -- at worst the consumer gets a spurious > event but doesn't find any CQEs.
This would be a spec violation - once no active QPs are attached, the consumer should be able to assume he will get no new events. But I agree we can regard this as a separate issue, since it does not affect IPoIB/CM. So for now I'm going for your proposed change: merge all QP associated events and the CMD IFC events to a single EQ. -- MST _______________________________________________ general mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
