Skip to contentSkip to conversation

osm_ucast_cache.c: Rewrite p_sw->lft allocation in osm_ucast_cache_process

PATCH New Hal Rosenstock 1 participant 1 message Jan 8, 2015 InfiniBand

lore
Share & tools
lore
#0 lore raw

PATCH 4/5 osm_ucast_cache.c: Rewrite p_sw->lft allocation in osm_ucast_cache_process

From: Daniel Klein <danielk-VPRAkNaXOzVWk0Htik3J/[email protected]>

Performed the following changes in the code that allocates p_sw->lft buffer in osm_ucast_cache_process:

1. remove allocation of p_sw->new_lft and replace it with assert that checks that p_sw->is already allocated.
p_sw->new_lft, should already be allocated and defined by either ucast_cache or the ucast_mgr.

2. replace p_sw->need_update check, with checking if p_sw->lft is not allocated.
p_sw->need update could indicate the switches LFT are not updated, and when p_sw->lft is already allocated, it cause memory leak.

---
 opensm/osm_ucast_cache.c |    9 ++-------
 1 files changed, 2 insertions(+), 7 deletions(-)
opensm/osm_ucast_cache.c +2 -7
index 2fab643..3caf318 100644
@@ -1040,13 +1040,8 @@ int osm_ucast_cache_process(osm_ucast_mgr_t * p_mgr)
 	for (item = cl_qmap_head(tbl); item != cl_qmap_end(tbl);
 	     item = cl_qmap_next(item)) {
 		p_sw = (osm_switch_t *) item;
-
-		if (p_sw->need_update) {
-			if (!p_sw->new_lft)
-				/* no new routing was recently calculated for this
-				   switch, but the LFT needs to be updated anyway */
-				p_sw->new_lft = p_sw->lft;
-
+		CL_ASSERT(p_sw->new_lft);
+		if (!p_sw->lft) {
 			lft_size = (p_sw->max_lid_ho / IB_SMP_DATA_SIZE + 1)
 				   * IB_SMP_DATA_SIZE;
 			p_sw->lft = malloc(lft_size);
Signature · 6 lines
1.7.1

--
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