>>> +#ifdef OSM_VENDOR_INTF_AL >>> + dlid = p_osmt->local_port.sm_lid; // already in correct >>> byte-order +#else dlid = cl_ntoh16(p_osmt->local_port.sm_lid); >>> +#endif >> >> osmtest looks like the wrong layer to apply this sort of >> work-around/fix. > >Applying this fix to lower levels breaks the test in other ways. >Given that vendor IBAL will eventually go away, I'm less inclined to rewrite >osmtest to fully address the differences between IBAL and umad.
This seems like an issue in one of the vendor layers, if they report the LID in different orders. It's hard to tell if this is the case from looking at just this patch. _______________________________________________ ofw mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ofw
