In case of an error when an asynchronous flow create operation was
enqueued, test-pmd attempted to enqueue a flow destroy operation
of that flow rule.
However, this was incorrect because:
- Flow rule index was used to enqueue a flow destroy operation.
This flow rule index was not yet initialized, so flow rule number 0
was always destroyed as a result.
- Since rte_flow_async_create() does not return a handle on error,
then there is no flow rule to destroy.
test-pmd only needs to free internal memory allocated for
storing a flow rule. Any flow destroy operation is not needed
in this case.
Fixes: ecdc927b99f2 ("app/testpmd: add async flow create/destroy operations")
Cc: akozy...@nvidia.com
Cc: sta...@dpdk.org
Signed-off-by: Dariusz Sosnowski <dsosnow...@nvidia.com>
Acked-by: Ori Kam <or...@nvidia.com>
---
app/test-pmd/config.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
index 08d7d0e2f8..8bf2e411b5 100644
--- a/app/test-pmd/config.c
+++ b/app/test-pmd/config.c
@@ -2856,8 +2856,7 @@ port_queue_flow_create(portid_t port_id, queueid_t
queue_id,
flow = rte_flow_async_create_by_index(port_id, queue_id,
&op_attr, pt->table,
rule_idx, actions, actions_idx, job, &error);
if (!flow) {
- uint64_t flow_id = pf->id;
- port_queue_flow_destroy(port_id, queue_id, true, 1, &flow_id);
+ free(pf);
free(job);
return port_flow_complain(&error);
}
--
2.39.2
