> I forgot to mention, maybe the RUNNING bit was added as part of this bug > fix, > which means no LLTX driver need to really check for this condition again in > their xmit handler. That might explain the problem you faced then.
Yes, that is the missing piece. OK, I think your patch is correct and I'll apply it if I can't get rid of LLTX for 2.6.25. - R. _______________________________________________ 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
