Guys, I'm worried that several critical IMO issues reported (with patches) a long time ago still aren't fixed even on svn trunk. I agree the issues are non-trivial races, but I wander whether you could give these a bit higher priority?
Roland, could we get the discussion going again on my pile of ipoib patches? IPoIB at stress causes system hangs for me unless I apply them, and I dont cherish the perspective of debugging these all over again, so I'm always running with all of them applied. Its rock solid with them. I can repost them one by one, adding detailed description and failure scenarious if needed. If we could get, at least, a patch per day discussed, we'll be in a good shape in 2.6.17 time scale. Would that work? Sean, there's a patch https://openib.org/svn/trunk/contrib/mellanox/patches/mad_port_close.patch which corrects an oops condition in mad.c. You had efficiency concerns with respect to this patch, but no better solution surfaced. Could you try fixing this if you have a better idea? -- Michael S. Tsirkin Staff Engineer, Mellanox Technologies _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
