opensm/osm_lid_mgr: use 'first_time_master_sweep' flag
PATCH 1/1 opensm/osm_lid_mgr: use 'first_time_master_sweep' flag
Use 'first_time_master_sweep' as initial sweep indication instead of 'coming_out_of_standby'. The last flag will be set only when OpenSM previously was in standby state, this doesn't cover some other cases such as temporary SM port disconnection.
--- opensm/opensm/osm_lid_mgr.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
opensm/opensm/osm_lid_mgr.c +1 -1
@@ -304,7 +304,7 @@ static int lid_mgr_init_sweep(IN osm_lid_mgr_t * p_mgr)info we might have. Do this only if the honor_guid2lid_file option is FALSE. If not, then need to honor this file. */ -if (p_mgr->p_subn->coming_out_of_standby == TRUE) { +if (p_mgr->p_subn->first_time_master_sweep == TRUE) {osm_db_clear (p_mgr->p_g2l);if (p_mgr->p_subn->opt.honor_guid2lid_file == FALSE) { OSM_LOG(p_mgr->p_log, OSM_LOG_DEBUG,
Signature · 6 lines
1.7.0.4 -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
PATCH 1/1 opensm/osm_lid_mgr: fix couple of duplicate LIDs bugs
Fix couple of cases where duplicated LID generation is possible:
1) When guid2lid db is reloaded after standby or disconnected SM port state (discovering) it must be validated, so content of used_lids[] array (used as reference during LID assignment) will be consistent with loaded guid2lid db.
2) When port for which LIDs were assigned exists in guid2lid db already its record should be updated due to possibly changed LIDs, LMC, etc..
There still be more potential cases where LIDs could be duplicated. So this patch fixes just part of them. I'm continuing an investigation.
--- opensm/opensm/osm_lid_mgr.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-)
opensm/opensm/osm_lid_mgr.c +2 -2
@@ -319,6 +319,7 @@ static int lid_mgr_init_sweep(IN osm_lid_mgr_t * p_mgr)"ERR 0306: " "Error restoring Guid-to-Lid " "persistent database. Ignoring it\n"); +lid_mgr_validate_db (p_mgr); } }
@@ -710,7 +711,7 @@ static int lid_mgr_get_port_lid(IN osm_lid_mgr_t * p_mgr,/* we still need to send the setting to the target port */ lid_changed =1 ; } - goto Exit; + goto NewLidSet; }else OSM_LOG(p_mgr->p_log, OSM_LOG_DEBUG, "0x%016" PRIx64 " has no persistent lid assigned\n",
@@ -764,7 +765,6 @@ NewLidSet:for (lid = *p_min_lid; lid <= *p_max_lid; lid++) p_mgr->used_lids[lid] =1 ; -Exit:/* make sure the assigned lids are marked in port_lid_tbl */ for (lid = *p_min_lid; lid <= *p_max_lid; lid++)cl_ptr_vector_set (&p_mgr->p_subn->port_lid_tbl, lid, p_port);
Signature · 6 lines
1.7.0.4 -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
PATCH v2 1/1 opensm/osm_lid_mgr: fix couple of duplicate LIDs bugs
Fix couple of cases where duplicated LID generation is possible:
1) When guid2lid db is reloaded after standby or disconnected SM port state (discovering) it must be validated, so content of used_lids[] array (used as reference during LID assignment) will be consistent with loaded guid2lid db.
2) When port for which LIDs were assigned exists in guid2lid db already its record should be updated due to possibly changed LIDs, LMC, etc..
There still be more potential cases where LIDs could be duplicated. So this patch fixes just part of them. I'm continuing an investigation.
--- opensm/opensm/osm_lid_mgr.c | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-)
v2: With guid2lid db reset (osm_db_clear()) clean also user_lids[] array.
opensm/opensm/osm_lid_mgr.c +5 -5
@@ -306,11 +306,11 @@ static int lid_mgr_init_sweep(IN osm_lid_mgr_t * p_mgr)need to honor this file. */ if (p_mgr->p_subn->first_time_master_sweep == TRUE) {osm_db_clear (p_mgr->p_g2l); -if (p_mgr->p_subn->opt.honor_guid2lid_file == FALSE) { +memset (p_mgr->used_lids,0 ,sizeof (p_mgr->used_lids)); +if (p_mgr->p_subn->opt.honor_guid2lid_file == FALSE)OSM_LOG (p_mgr->p_log,OSM_LOG_DEBUG ,"Ignore guid2lid file when coming out of standby\n" ); -memset (p_mgr->used_lids,0 ,sizeof (p_mgr->used_lids)); - }else { +else {OSM_LOG (p_mgr->p_log,OSM_LOG_DEBUG ,"Honor current guid2lid file when coming out " "of standby\n" );
@@ -319,6 +319,7 @@ static int lid_mgr_init_sweep(IN osm_lid_mgr_t * p_mgr)"ERR 0306: " "Error restoring Guid-to-Lid " "persistent database. Ignoring it\n"); +lid_mgr_validate_db (p_mgr); } }
@@ -710,7 +711,7 @@ static int lid_mgr_get_port_lid(IN osm_lid_mgr_t * p_mgr,/* we still need to send the setting to the target port */ lid_changed =1 ; } - goto Exit; + goto NewLidSet; }else OSM_LOG(p_mgr->p_log, OSM_LOG_DEBUG, "0x%016" PRIx64 " has no persistent lid assigned\n",
@@ -764,7 +765,6 @@ NewLidSet:for (lid = *p_min_lid; lid <= *p_max_lid; lid++) p_mgr->used_lids[lid] =1 ; -Exit:/* make sure the assigned lids are marked in port_lid_tbl */ for (lid = *p_min_lid; lid <= *p_max_lid; lid++)cl_ptr_vector_set (&p_mgr->p_subn->port_lid_tbl, lid, p_port);
Signature · 6 lines
1.7.0.4 -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 16 lines · Sasha Khapyorsky
Use '
first_time_master_sweep' as initial sweep indication instead of 'coming_out_of_standby'. The last flag will be set only when OpenSM previously was in standby state, this doesn't cover some other cases such as temporary SM port disconnection.--- opensm/opensm/osm_lid_mgr.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)opensm/opensm/osm_lid_mgr.c +0
-0
opensm/opensm/osm_lid_mgr.c static int lid_mgr_init_sweep(IN osm_lid_mgr_t * p_mgr) · line 307
Quoted 5 lines
opensm/opensm/osm_lid_mgr.c +1
-1@@ -304,7 +304,7 @@ static int lid_mgr_init_sweep(IN osm_lid_mgr_t * p_mgr)info we might have. Do this only if the honor_guid2lid_file option is FALSE. If not, then need to honor this file. */ -if (p_mgr->p_subn->coming_out_of_standby == TRUE) { +if (p_mgr->p_subn->first_time_master_sweep == TRUE) {
Actually this changes the current behavior. Flag 'coming_out_of_standby' is not set on first initial sweep and 'first_time_master_sweep' is. So I'm reverting this patch yet up to more cleanup will be performed.
Sasha
Quoted 6 lines
opensm/opensm/osm_lid_mgr.c +0
-0osm_db_clear (p_mgr->p_g2l);if (p_mgr->p_subn->opt.honor_guid2lid_file == FALSE) { OSM_LOG(p_mgr->p_log, OSM_LOG_DEBUG,Signature · 1 line
1.7.0.4
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
PATCH 1/1 opensm/osm_lid_mgr: revert 'coming_out_of_standby' replacement patch
This reverts the patch where 'coming_out_of_standby' flag was replaced by 'first_time_master_sweep'. It is probably a good idea in general to merge those, but right now it will change the default lid manager behavior, since it will ignore guid2lid file in initial sweep.
--- opensm/opensm/osm_lid_mgr.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
opensm/opensm/osm_lid_mgr.c +1 -1
@@ -304,7 +304,7 @@ static int lid_mgr_init_sweep(IN osm_lid_mgr_t * p_mgr)info we might have. Do this only if the honor_guid2lid_file option is FALSE. If not, then need to honor this file. */ -if (p_mgr->p_subn->first_time_master_sweep == TRUE) { +if (p_mgr->p_subn->coming_out_of_standby == TRUE) {osm_db_clear (p_mgr->p_g2l);memset (p_mgr->used_lids,0 ,sizeof (p_mgr->used_lids));if (p_mgr->p_subn->opt.honor_guid2lid_file == FALSE)
Signature · 6 lines
1.7.0.4 -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html