Multicast group pre-creation for TopSpin old stack

There's a hack in the SM to deal with TopSpin's non-compliant join compmask for IPoIB v4 multicast group ff12:401b:<pkey>::1 The group is pre-defined and maintained as a well-known multicast group.

opensm/osm_prtn.c:

   232          /* workaround for TS */
   233          /* FIXME: remove this upon TS fixes */
   234          mc_rec.mgid = osm_ts_ipoib_mgid;
   235          memcpy(&mc_rec.mgid.raw[4], &pkey, sizeof(pkey));
   236          /* Scope in MCMemberRecord (if present) needs to be consistent with MGID */
   237          mc_rec.scope_state = ib_member_set_scope_state(scope, IB_MC_REC_STATE_FULL_MEMBER);
   238          ib_mgid_set_scope(&mc_rec.mgid, scope);
   239
   240          status = osm_mcmr_rcv_find_or_create_new_mgrp(p_sa, comp_mask, &mc_rec,
   241                                                        &p_mgrp);
   242          if (p_mgrp) {
   243                  p_mgrp->well_known = TRUE;
   244                  if (!p->mgrp)
   245                          p->mgrp = p_mgrp;
   246          }

As far as I can tell, this was added before git history. Any idea if it's still needed?

Hi Yevgeny,

Quoted 24 lines · Yevgeny Kliteynik

There's a hack in the SM to deal with TopSpin's non-compliant join compmask for IPoIB v4 multicast group ff12:401b:<pkey>::1 The group is pre-defined and maintained as a well-known multicast group.

opensm/osm_prtn.c:

   232          /* workaround for TS */
   233          /* FIXME: remove this upon TS fixes */
   234          mc_rec.mgid = osm_ts_ipoib_mgid;
   235          memcpy(&mc_rec.mgid.raw[4], &pkey, sizeof(pkey));
   236          /* Scope in MCMemberRecord (if present) needs to be consistent with MGID */ 237          mc_rec.scope_state = ib_member_set_scope_state(scope, IB_MC_REC_STATE_FULL_MEMBER); 238          ib_mgid_set_scope(&mc_rec.mgid, scope);
   239
   240          status = osm_mcmr_rcv_find_or_create_new_mgrp(p_sa, comp_mask, &mc_rec, 241                                                        &p_mgrp);
   242          if (p_mgrp) {
   243                  p_mgrp->well_known = TRUE;
   244                  if (!p->mgrp)
   245                          p->mgrp = p_mgrp;
   246          }

Quoted 2 lines

As far as I can tell, this was added before git history. Any idea if it's still needed?

I don't know, just remember that we have asked about this during partition manager development (somewhere in 2006), and have got an answer like "it is probably needed".

Personally I've never seen this used. Maybe Hal may remember more.

Sasha

It's been carried forward just in case anyone was still running the old TS stack. It's always hard to determine the negative (when no one is using it any longer) (with the latest OpenSM) but it may be high time to deprecate this. Is anyone still using the old TS stack ? If so, is it used with OpenSM ?

Maybe Roland has insight into whether the old TS stack might be being used with (the latest) OpenSM.

Quoted 6 lines

It's been carried forward just in case anyone was still running the old TS stack. It's always hard to determine the negative (when no one is using it any longer) (with the latest OpenSM) but it may be high time to deprecate this. Is anyone still using the old TS stack ? If so, is it used with OpenSM ?

Quoted 2 lines

Maybe Roland has insight into whether the old TS stack might be being used with (the latest) OpenSM.

Seems a fairly remote possibility but on the other hand I guess I could see someone buying old Topspin equipment on ebay and wanting to use it with modern software.

In any case, if carrying this workaround forward
costs anything, I think it could be dropped.

As far as I could understand this TS multicast group only impacts IPoIB communication between two TS stack clients. If so then it is likely that just using old Topspin switch will not activate this flow anyway. Right?