On Fri, 2007-10-19 at 09:26 -0700, Ralph Campbell wrote: > On Fri, 2007-10-19 at 09:19 -0700, Sean Hefty wrote: > > > This looks like it could be a problem, but it is not directly > > > related to the original patch; i.e. loop back of a DR SMP > > > response being sent from userspace. > > > > > > I think you could probably combine the routines as you suggest > > > or call them both as is done to solve the original problem. > > > If you want to combine this change with the original patch > > > to make a more uniform change across both problems that > > > would be fine with me. Hal, Sean, do you have an opinion? > > > > I think Hal pointed this out, but the mad layer doesn't currently > > support mixing DR and LID routed SMPs. I would like patches to add/fix > > that to be separate. > > > > - Sean > > I agree. I just wanted to review the original patch enough to be > sure I understood the limitations. Now that my memory of DR SMPs > is refreshed, if one of us resubmits the V3 patch plus the ipath > change plus the extended description, I will ACK it.
Yes, Steve should resubmit the V3 patch updated with the extended description. I think your patch to eliminate the unneeded if clause can be a separate patch (and I acked that already). Is that what you mean by the ipath change ? Is there something else right now ? -- Hal > > _______________________________________________ > 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 _______________________________________________ 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
