From: Daniel Klein <danielk@xxxxxxxxxxxx>
Bug description:
When PerfMgr starts discovery, it resets the discovery_count of ports and
node, but it doesn't reset the physp_discoved buffers.
As a result, the drop manager frees osm_port_t and osm_physp_t objects
of nodes without removing the nodes themselves.
On the next discovery process, when OpenSM receives responses for
NodeInfo from a switch, it treats it as a NodeInfo from an existing
node and assumes that osm_physp_t objects of the switch exists and tries
to access them.
This change adds code that clears the physp_discoverd buffers of all nodes
at the beginning of the discovery process in PerfMgr.
Signed-off-by: Daniel Klein <danielk@xxxxxxxxxxxx>
Signed-off-by: Hal Rosenstock <hal@xxxxxxxxxxxx>
---
opensm/osm_perfmgr.c | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/opensm/osm_perfmgr.c b/opensm/osm_perfmgr.c
index 4ab654b..993fdb7 100644
--- a/opensm/osm_perfmgr.c
+++ b/opensm/osm_perfmgr.c
@@ -985,6 +985,9 @@ static void reset_node_count(cl_map_item_t * p_map_item, void *cxt)
{
osm_node_t *p_node = (osm_node_t *) p_map_item;
p_node->discovery_count = 0;
+
+ memset(p_node->physp_discovered, 0,
+ sizeof(uint8_t) * p_node->physp_tbl_size);
}
static void reset_port_count(cl_map_item_t * p_map_item, void *cxt)
--
1.7.8.2
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
