* [PATCH] net/mlx5: fix sync meter action
@ 2024-03-18 18:10 Gregory Etelson
2024-03-19 11:24 ` [PATCH v2] " Gregory Etelson
0 siblings, 1 reply; 3+ messages in thread
From: Gregory Etelson @ 2024-03-18 18:10 UTC (permalink / raw)
To: dev
Cc: getelson, mkashani, rasland, Dariusz Sosnowski,
Viacheslav Ovsiienko, Ori Kam, Suanming Mou, Matan Azrad
PMD implements sync METER flow action as async.
Queue selected for sync operations is `MLX5_HW_INV_QUEUE`.
That dummy queue value is translated into `CTRL_QUEUE_ID(priv)`.
Async job allocation converts INV queue into the real value, but
job release does not.
This patch fixes the queue value provided to `flow_hw_job_put()`.
This patch also removes dead code found in METER_MARK
destroy handler.
Coverity issue: tag: 415806: Memory - corruptions (OVERRUN)
Coverity issue: tag: 415804: Control flow issues (DEADCODE)
Fixes: 4359d9d1f76b ("net/mlx5: fix sync meter processing in HWS")
Signed-off-by: Gregory Etelson <getelson@nvidia.com>
Acked-by: Dariusz Sosnowski <dsosnowski@nvidia.com>
---
drivers/net/mlx5/mlx5_flow_hw.c | 5 +----
drivers/net/mlx5/mlx5_flow_meter.c | 2 +-
2 files changed, 2 insertions(+), 5 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow_hw.c b/drivers/net/mlx5/mlx5_flow_hw.c
index 35f1ed7a03..9ebbe664d1 100644
--- a/drivers/net/mlx5/mlx5_flow_hw.c
+++ b/drivers/net/mlx5/mlx5_flow_hw.c
@@ -11494,10 +11494,7 @@ flow_hw_action_handle_destroy(struct rte_eth_dev *dev, uint32_t queue,
NULL, "Unable to wait for ASO meter CQE");
break;
}
- if (!job)
- mlx5_ipool_free(pool->idx_pool, idx);
- else
- aso = true;
+ aso = true;
break;
case MLX5_INDIRECT_ACTION_TYPE_RSS:
ret = flow_dv_action_destroy(dev, handle, error);
diff --git a/drivers/net/mlx5/mlx5_flow_meter.c b/drivers/net/mlx5/mlx5_flow_meter.c
index 4045c4c249..ca361f7efa 100644
--- a/drivers/net/mlx5/mlx5_flow_meter.c
+++ b/drivers/net/mlx5/mlx5_flow_meter.c
@@ -2265,7 +2265,7 @@ mlx5_flow_meter_hws_create(struct rte_eth_dev *dev, uint32_t meter_id,
ret = mlx5_aso_meter_update_by_wqe(priv, MLX5_HW_INV_QUEUE, aso_mtr,
&priv->mtr_bulk, job, true);
if (ret) {
- flow_hw_job_put(priv, job, MLX5_HW_INV_QUEUE);
+ flow_hw_job_put(priv, job, CTRL_QUEUE_ID(priv));
return -rte_mtr_error_set(error, ENOTSUP,
RTE_MTR_ERROR_TYPE_UNSPECIFIED,
NULL, "Failed to create devx meter.");
--
2.39.2
^ permalink raw reply [flat|nested] 3+ messages in thread* [PATCH v2] net/mlx5: fix sync meter action
2024-03-18 18:10 [PATCH] net/mlx5: fix sync meter action Gregory Etelson
@ 2024-03-19 11:24 ` Gregory Etelson
2024-03-24 22:25 ` Thomas Monjalon
0 siblings, 1 reply; 3+ messages in thread
From: Gregory Etelson @ 2024-03-19 11:24 UTC (permalink / raw)
To: dev
Cc: getelson, mkashani, rasland, Dariusz Sosnowski,
Viacheslav Ovsiienko, Ori Kam, Suanming Mou, Matan Azrad
PMD implements sync METER flow action as async.
Queue selected for sync operations is `MLX5_HW_INV_QUEUE`.
That dummy queue value is translated into `CTRL_QUEUE_ID(priv)`.
Async job allocation converts INV queue into the real value, but
job release does not.
This patch fixes the queue value provided to `flow_hw_job_put()`.
This patch also removes dead code found in METER_MARK
destroy handler.
Coverity issue: 415806
Coverity issue: 415804
Fixes: 4359d9d1f76b ("net/mlx5: fix sync meter processing in HWS")
Signed-off-by: Gregory Etelson <getelson@nvidia.com>
Acked-by: Dariusz Sosnowski <dsosnowski@nvidia.com>
---
v2: Fixed Coverity tag.
---
drivers/net/mlx5/mlx5_flow_hw.c | 5 +----
drivers/net/mlx5/mlx5_flow_meter.c | 2 +-
2 files changed, 2 insertions(+), 5 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow_hw.c b/drivers/net/mlx5/mlx5_flow_hw.c
index 35f1ed7a03..9ebbe664d1 100644
--- a/drivers/net/mlx5/mlx5_flow_hw.c
+++ b/drivers/net/mlx5/mlx5_flow_hw.c
@@ -11494,10 +11494,7 @@ flow_hw_action_handle_destroy(struct rte_eth_dev *dev, uint32_t queue,
NULL, "Unable to wait for ASO meter CQE");
break;
}
- if (!job)
- mlx5_ipool_free(pool->idx_pool, idx);
- else
- aso = true;
+ aso = true;
break;
case MLX5_INDIRECT_ACTION_TYPE_RSS:
ret = flow_dv_action_destroy(dev, handle, error);
diff --git a/drivers/net/mlx5/mlx5_flow_meter.c b/drivers/net/mlx5/mlx5_flow_meter.c
index 4045c4c249..ca361f7efa 100644
--- a/drivers/net/mlx5/mlx5_flow_meter.c
+++ b/drivers/net/mlx5/mlx5_flow_meter.c
@@ -2265,7 +2265,7 @@ mlx5_flow_meter_hws_create(struct rte_eth_dev *dev, uint32_t meter_id,
ret = mlx5_aso_meter_update_by_wqe(priv, MLX5_HW_INV_QUEUE, aso_mtr,
&priv->mtr_bulk, job, true);
if (ret) {
- flow_hw_job_put(priv, job, MLX5_HW_INV_QUEUE);
+ flow_hw_job_put(priv, job, CTRL_QUEUE_ID(priv));
return -rte_mtr_error_set(error, ENOTSUP,
RTE_MTR_ERROR_TYPE_UNSPECIFIED,
NULL, "Failed to create devx meter.");
--
2.39.2
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [PATCH v2] net/mlx5: fix sync meter action
2024-03-19 11:24 ` [PATCH v2] " Gregory Etelson
@ 2024-03-24 22:25 ` Thomas Monjalon
0 siblings, 0 replies; 3+ messages in thread
From: Thomas Monjalon @ 2024-03-24 22:25 UTC (permalink / raw)
To: Gregory Etelson
Cc: dev, mkashani, rasland, Dariusz Sosnowski, Viacheslav Ovsiienko,
Ori Kam, Suanming Mou, Matan Azrad
19/03/2024 12:24, Gregory Etelson:
> PMD implements sync METER flow action as async.
> Queue selected for sync operations is `MLX5_HW_INV_QUEUE`.
> That dummy queue value is translated into `CTRL_QUEUE_ID(priv)`.
> Async job allocation converts INV queue into the real value, but
> job release does not.
>
> This patch fixes the queue value provided to `flow_hw_job_put()`.
>
> This patch also removes dead code found in METER_MARK
> destroy handler.
>
> Coverity issue: 415806
> Coverity issue: 415804
>
> Fixes: 4359d9d1f76b ("net/mlx5: fix sync meter processing in HWS")
>
> Signed-off-by: Gregory Etelson <getelson@nvidia.com>
> Acked-by: Dariusz Sosnowski <dsosnowski@nvidia.com>
Applied, thanks.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2024-03-24 22:25 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2024-03-18 18:10 [PATCH] net/mlx5: fix sync meter action Gregory Etelson
2024-03-19 11:24 ` [PATCH v2] " Gregory Etelson
2024-03-24 22:25 ` Thomas Monjalon
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).