Jack wrote, >Since the change you request (which is a kernel-level change) affects many >products, not just IntelMPI, it is not possible to just include it at the >last minute and hope for the best (due to lack of QA).
>- Jack At this point if the 1.0-pre1 tar ball goes gold this Friday as scheduled, then I guess we will have to live with what we have and people running larger clusters will need to use the trunk until OFED 1.1. If however, it is decided that there needs to be another RC to fix the other problems with things like pathscale not working with SDP or uDAPL, then I think we should allow the setops fixes in also, but I understand that would mean going through another QA cycle. my 2 cents. woody _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
