* [PATCH v1] net/mlx5: fix indirect meter index leak
@ 2025-11-05 12:54 Rongwei Liu
0 siblings, 0 replies; only message in thread
From: Rongwei Liu @ 2025-11-05 12:54 UTC (permalink / raw)
To: dev, matan, viacheslavo, orika, suanmingm, thomas
Cc: akozyrev, stable, Dariusz Sosnowski, Bing Zhao
When destroying the meter_mark indirect action, PMD
didn't recycle the ipool resources.
Fixes: 48fbb0e93d06 ("net/mlx5: support flow meter mark indirect action with HWS")
Cc: akozyrev@nvidia.com
Cc: stable@dpdk.org
Signed-off-by: Rongwei Liu <rongweil@nvidia.com>
Acked-by: Dariusz Sosnowski <dsosnowski@nvidia.com>
---
drivers/net/mlx5/mlx5_flow_hw.c | 17 ++++++++++-------
1 file changed, 10 insertions(+), 7 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow_hw.c b/drivers/net/mlx5/mlx5_flow_hw.c
index ff68483a40..22503f49d5 100644
--- a/drivers/net/mlx5/mlx5_flow_hw.c
+++ b/drivers/net/mlx5/mlx5_flow_hw.c
@@ -13011,13 +13011,16 @@ flow_hw_action_handle_destroy(struct rte_eth_dev *dev, uint32_t queue,
break;
}
/* Wait for ASO object completion. */
- if (queue == MLX5_HW_INV_QUEUE &&
- mlx5_aso_mtr_wait(priv, aso_mtr, true)) {
- ret = -EINVAL;
- rte_flow_error_set(error, EINVAL,
- RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
- NULL, "Unable to wait for ASO meter CQE");
- break;
+ if (queue == MLX5_HW_INV_QUEUE) {
+ if (mlx5_aso_mtr_wait(priv, aso_mtr, true)) {
+ ret = -EINVAL;
+ rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
+ NULL, "Unable to wait for ASO meter CQE");
+ }
+ mlx5_ipool_free(pool->idx_pool, idx);
+ if (ret < 0)
+ break;
}
aso = true;
break;
--
2.27.0
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2025-11-05 12:56 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2025-11-05 12:54 [PATCH v1] net/mlx5: fix indirect meter index leak Rongwei Liu
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).