[dpdk-dev] [PATCH 1/5] net/mlx5: fix unused calculation in RSS expansion

[dpdk-dev] [PATCH 1/5] net/mlx5: fix unused calculation in RSS expansion

[Michael Baum]

The RSS flow expansion get a memory buffer to fill the new patterns of
the expanded flows.
This memory management saves the next address to write into the buffer
in a dedicated variable.

The calculation for the next address was wrongly also done when all the
patterns were ready.

Remove it.

Fixes: 4ed05fcd441b ("ethdev: add flow API to expand RSS flows")
Cc: stable@dpdk.org

Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
---
 drivers/net/mlx5/mlx5_flow.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/mlx5/mlx5_flow.c b/drivers/net/mlx5/mlx5_flow.c
index 92adfca..ebd79c6 100644
--- a/drivers/net/mlx5/mlx5_flow.c
+++ b/drivers/net/mlx5/mlx5_flow.c
@@ -377,8 +377,6 @@ struct mlx5_flow_expand_rss {
 				   user_pattern_size);
 			addr = (void *)(((uintptr_t)addr) + user_pattern_size);
 			rte_memcpy(addr, flow_items, elt * sizeof(*item));
-			addr = (void *)(((uintptr_t)addr) +
-					elt * sizeof(*item));
 		}
 	}
 	return lsize;
-- 
1.8.3.1

[dpdk-dev] [PATCH 2/5] net/mlx5: fix Rx queue creation error flow

[Michael Baum]

In Rx queue creation, there is a validation for the Rx configuration.

When scatter offload validation for buffer split is failed, the Rx queue
object memory was not freed what caused a memory leak.

Free it.

Fixes: a0a45e8af723 ("net/mlx5: configure Rx queue for buffer split")

Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
---
 drivers/net/mlx5/mlx5_rxq.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/mlx5/mlx5_rxq.c b/drivers/net/mlx5/mlx5_rxq.c
index 4e6b8f4..0693c92 100644
--- a/drivers/net/mlx5/mlx5_rxq.c
+++ b/drivers/net/mlx5/mlx5_rxq.c
@@ -1506,7 +1506,7 @@ struct mlx5_rxq_ctrl *
 			dev->data->port_id, idx, mb_len, max_rx_pkt_len,
 			RTE_PKTMBUF_HEADROOM);
 		rte_errno = ENOSPC;
-		return NULL;
+		goto error;
 	}
 	tmpl->type = MLX5_RXQ_TYPE_STANDARD;
 	if (mlx5_mr_btree_init(&tmpl->rxq.mr_ctrl.cache_bh,
-- 
1.8.3.1

[dpdk-dev] [PATCH 3/5] net/mlx5: fix ASO age pools resize error flow

[Michael Baum]

In ASO age pools resize, the PMD starts ASO data-path.

When starting ASO data-path is failed, the pools memory was not freed
what caused a memory leak.

Free it.

Fixes: f935ed4b645a ("net/mlx5: support flow hit action for aging")

Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
---
 drivers/net/mlx5/mlx5_flow_dv.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c
index 78c710f..8214c59 100644
--- a/drivers/net/mlx5/mlx5_flow_dv.c
+++ b/drivers/net/mlx5/mlx5_flow_dv.c
@@ -9347,8 +9347,10 @@ struct mlx5_cache_entry *
 		/* First ASO flow hit allocation - starting ASO data-path. */
 		int ret = mlx5_aso_queue_start(priv->sh);
 
-		if (ret)
+		if (ret) {
+			mlx5_free(pools);
 			return ret;
+		}
 	}
 	mng->n = resize;
 	mng->pools = pools;
-- 
1.8.3.1

[dpdk-dev] [PATCH 4/5] net/mlx5: fix compiler code arrangement in probe

[Michael Baum]

Bonding adjustment is done only when DEVX_PORT is supported in the
rdma-core.

Some bonding condition was done even when DEVX_PORT is not supported.

Remove it.

Fixes: 2eb4d0107acc ("net/mlx5: refactor PCI probing on Linux")
Cc: stable@dpdk.org

Signed-off-by: Michael Baum <michaelba@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
---
 drivers/net/mlx5/linux/mlx5_os.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/mlx5/linux/mlx5_os.c b/drivers/net/mlx5/linux/mlx5_os.c
index c78d56f..ce25108 100644
--- a/drivers/net/mlx5/linux/mlx5_os.c
+++ b/drivers/net/mlx5/linux/mlx5_os.c
@@ -1942,6 +1942,7 @@
 						(list[ns].ifindex,
 						 &list[ns].info);
 			}
+#ifdef HAVE_MLX5DV_DR_DEVX_PORT
 			if (!ret && bd >= 0) {
 				switch (list[ns].info.name_type) {
 				case MLX5_PHYS_PORT_NAME_TYPE_UPLINK:
@@ -1959,6 +1960,7 @@
 				}
 				continue;
 			}
+#endif
 			if (!ret && (list[ns].info.representor ^
 				     list[ns].info.master))
 				ns++;
-- 
1.8.3.1

[dpdk-dev] [PATCH 5/5] common/mlx5: fix flex parser creation error flow

[Michael Baum]

From: Michael Baum <michaelba@mellanox.com>

The mlx5_devx_cmd_create_flex_parser function defines a local array
(with constant size) named in, and then allocates a pointer to
mlx5_devx_obj structure by the mlx5_malloc function.

If the allocation fails, the function releases the array in and returns
NULL. However, the array has been defined locally on the stack and the
mlx5_free function is not required to release it.

Remove the call to the mlx5_free function.

Fixes: 66914d19d135 ("common/mlx5: convert control path memory to unified malloc")
Cc: stable@dpdk.org

Signed-off-by: Michael Baum <michaelba@mellanox.com>
Acked-by: Matan Azrad <matan@nvidia.com>
---
 drivers/common/mlx5/mlx5_devx_cmds.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/common/mlx5/mlx5_devx_cmds.c b/drivers/common/mlx5/mlx5_devx_cmds.c
index 5998c4b..9c1d188 100644
--- a/drivers/common/mlx5/mlx5_devx_cmds.c
+++ b/drivers/common/mlx5/mlx5_devx_cmds.c
@@ -530,15 +530,13 @@ struct mlx5_devx_obj *
 	void *sample = MLX5_ADDR_OF(parse_graph_flex, flex, sample_table);
 	void *in_arc = MLX5_ADDR_OF(parse_graph_flex, flex, input_arc);
 	void *out_arc = MLX5_ADDR_OF(parse_graph_flex, flex, output_arc);
-	struct mlx5_devx_obj *parse_flex_obj = NULL;
+	struct mlx5_devx_obj *parse_flex_obj = mlx5_malloc
+		     (MLX5_MEM_ZERO, sizeof(*parse_flex_obj), 0, SOCKET_ID_ANY);
 	uint32_t i;
 
-	parse_flex_obj = mlx5_malloc(MLX5_MEM_ZERO, sizeof(*parse_flex_obj), 0,
-				     SOCKET_ID_ANY);
 	if (!parse_flex_obj) {
-		DRV_LOG(ERR, "Failed to allocate flex parser data");
+		DRV_LOG(ERR, "Failed to allocate flex parser data.");
 		rte_errno = ENOMEM;
-		mlx5_free(in);
 		return NULL;
 	}
 	MLX5_SET(general_obj_in_cmd_hdr, hdr, opcode,
-- 
1.8.3.1

Re: [dpdk-dev] [PATCH 1/5] net/mlx5: fix unused calculation in RSS expansion

[Thomas Monjalon]

11/11/2020 08:36, Michael Baum:
> The RSS flow expansion get a memory buffer to fill the new patterns of
> the expanded flows.
> This memory management saves the next address to write into the buffer
> in a dedicated variable.
> 
> The calculation for the next address was wrongly also done when all the
> patterns were ready.
> 
> Remove it.
> 
> Fixes: 4ed05fcd441b ("ethdev: add flow API to expand RSS flows")
> Cc: stable@dpdk.org
> 
> Signed-off-by: Michael Baum <michaelba@nvidia.com>
> Acked-by: Matan Azrad <matan@nvidia.com>

Series applied in next-net-mlx, thanks.