Multiwriter insert function was using a fixed value for
the bucket size, instead of using the
RTE_HASH_BUCKET_ENTRIES macro, which value was changed
recently (making it inconsistent in this case).
Fixes: be856325cba3 ("hash: add scalable multi-writer insertion with Intel TSX")
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch at intel.com>
---
lib/librte_hash/rte_cuckoo_hash_x86.h | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/lib/librte_hash/rte_cuckoo_hash_x86.h
b/lib/librte_hash/rte_cuckoo_hash_x86.h
index 47aec6d..0c94244 100644
--- a/lib/librte_hash/rte_cuckoo_hash_x86.h
+++ b/lib/librte_hash/rte_cuckoo_hash_x86.h
@@ -167,7 +167,8 @@ rte_hash_cuckoo_make_space_mw_tm(const struct rte_hash *h,
/* Cuckoo bfs Search */
while (likely(tail != head && head <
- queue + RTE_HASH_BFS_QUEUE_MAX_LEN -
4)) {
+ queue + RTE_HASH_BFS_QUEUE_MAX_LEN -
+ RTE_HASH_BUCKET_ENTRIES)) {
curr_bkt = tail->bkt;
for (i = 0; i < RTE_HASH_BUCKET_ENTRIES; i++) {
if (curr_bkt->key_idx[i] == EMPTY_SLOT) {
--
2.7.4
