net/mlx5: fix rte flow items size calculation

Message ID 1594901695-29973-1-git-send-email-rasland@mellanox.com (mailing list archive)
State Accepted, archived
Delegated to: Raslan Darawsheh
Headers
Series net/mlx5: fix rte flow items size calculation |

Checks

Context Check Description
ci/checkpatch success coding style OK
ci/Intel-compilation success Compilation OK
ci/travis-robot success Travis build: passed
ci/iol-broadcom-Performance success Performance Testing PASS
ci/iol-intel-Performance success Performance Testing PASS
ci/iol-testing success Testing PASS
ci/iol-mellanox-Performance success Performance Testing PASS

Commit Message

Raslan Darawsheh July 16, 2020, 12:14 p.m. UTC
  flow_dv_get_item_len returns the actual header size of
an rte_flow item.

Changing any of the structs for rte_flow items by adding
or removing some extra fields will break this function.

This fixes the behavior by retrning the actual header size
of each item.

Fixes: 34d41b7aa3bf ("net/mlx5: add VXLAN encap action to Direct Verbs")
Cc: dekelp@mellanox.com
Cc: stable@dpdk.org

Signed-off-by: Raslan Darawsheh <rasland@mellanox.com>
---
 drivers/net/mlx5/mlx5_flow_dv.c | 31 ++++++++++++++-----------------
 1 file changed, 14 insertions(+), 17 deletions(-)
  

Comments

Slava Ovsiienko July 19, 2020, 8:26 a.m. UTC | #1
> -----Original Message-----
> From: dev <dev-bounces@dpdk.org> On Behalf Of Raslan Darawsheh
> Sent: Thursday, July 16, 2020 15:15
> To: dev@dpdk.org
> Cc: Dekel Peled <dekelp@mellanox.com>; stable@dpdk.org
> Subject: [dpdk-dev] [PATCH] net/mlx5: fix rte flow items size calculation
> 
> flow_dv_get_item_len returns the actual header size of an rte_flow item.
> 
> Changing any of the structs for rte_flow items by adding or removing some
> extra fields will break this function.
> 
> This fixes the behavior by retrning the actual header size of each item.
> 
> Fixes: 34d41b7aa3bf ("net/mlx5: add VXLAN encap action to Direct Verbs")
> Cc: dekelp@mellanox.com
> Cc: stable@dpdk.org
> 
> Signed-off-by: Raslan Darawsheh <rasland@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
  
Raslan Darawsheh July 22, 2020, 9:11 a.m. UTC | #2
Hi,

> -----Original Message-----
> From: Raslan Darawsheh <rasland@mellanox.com>
> Sent: Thursday, July 16, 2020 3:15 PM
> To: dev@dpdk.org
> Cc: Dekel Peled <dekelp@mellanox.com>; stable@dpdk.org
> Subject: [PATCH] net/mlx5: fix rte flow items size calculation
> 
> flow_dv_get_item_len returns the actual header size of
> an rte_flow item.
> 
> Changing any of the structs for rte_flow items by adding
> or removing some extra fields will break this function.
> 
> This fixes the behavior by retrning the actual header size
> of each item.
> 
> Fixes: 34d41b7aa3bf ("net/mlx5: add VXLAN encap action to Direct Verbs")
> Cc: dekelp@mellanox.com
> Cc: stable@dpdk.org
> 
> Signed-off-by: Raslan Darawsheh <rasland@mellanox.com>
> ---
>  drivers/net/mlx5/mlx5_flow_dv.c | 31 ++++++++++++++-----------------
>  1 file changed, 14 insertions(+), 17 deletions(-)
> 


Patch applied to next-net-mlx,

Kindest regards,
Raslan Darawsheh
  

Patch

diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c
index 8b5b683..6f61cf1 100644
--- a/drivers/net/mlx5/mlx5_flow_dv.c
+++ b/drivers/net/mlx5/mlx5_flow_dv.c
@@ -29,6 +29,7 @@ 
 #include <rte_gre.h>
 #include <rte_vxlan.h>
 #include <rte_gtp.h>
+#include <rte_mpls.h>
 
 #include <mlx5_devx_cmds.h>
 #include <mlx5_prm.h>
@@ -2869,7 +2870,7 @@  flow_dv_push_vlan_action_resource_register
 	return 0;
 }
 /**
- * Get the size of specific rte_flow_item_type
+ * Get the size of specific rte_flow_item_type hdr size
  *
  * @param[in] item_type
  *   Tested rte_flow_item_type.
@@ -2878,43 +2879,39 @@  flow_dv_push_vlan_action_resource_register
  *   sizeof struct item_type, 0 if void or irrelevant.
  */
 static size_t
-flow_dv_get_item_len(const enum rte_flow_item_type item_type)
+flow_dv_get_item_hdr_len(const enum rte_flow_item_type item_type)
 {
 	size_t retval;
 
 	switch (item_type) {
 	case RTE_FLOW_ITEM_TYPE_ETH:
-		retval = sizeof(struct rte_flow_item_eth);
+		retval = sizeof(struct rte_ether_hdr);
 		break;
 	case RTE_FLOW_ITEM_TYPE_VLAN:
-		retval = sizeof(struct rte_flow_item_vlan);
+		retval = sizeof(struct rte_vlan_hdr);
 		break;
 	case RTE_FLOW_ITEM_TYPE_IPV4:
-		retval = sizeof(struct rte_flow_item_ipv4);
+		retval = sizeof(struct rte_ipv4_hdr);
 		break;
 	case RTE_FLOW_ITEM_TYPE_IPV6:
-		retval = sizeof(struct rte_flow_item_ipv6);
+		retval = sizeof(struct rte_ipv6_hdr);
 		break;
 	case RTE_FLOW_ITEM_TYPE_UDP:
-		retval = sizeof(struct rte_flow_item_udp);
+		retval = sizeof(struct rte_udp_hdr);
 		break;
 	case RTE_FLOW_ITEM_TYPE_TCP:
-		retval = sizeof(struct rte_flow_item_tcp);
+		retval = sizeof(struct rte_tcp_hdr);
 		break;
 	case RTE_FLOW_ITEM_TYPE_VXLAN:
-		retval = sizeof(struct rte_flow_item_vxlan);
+	case RTE_FLOW_ITEM_TYPE_VXLAN_GPE:
+		retval = sizeof(struct rte_vxlan_hdr);
 		break;
 	case RTE_FLOW_ITEM_TYPE_GRE:
-		retval = sizeof(struct rte_flow_item_gre);
-		break;
 	case RTE_FLOW_ITEM_TYPE_NVGRE:
-		retval = sizeof(struct rte_flow_item_nvgre);
-		break;
-	case RTE_FLOW_ITEM_TYPE_VXLAN_GPE:
-		retval = sizeof(struct rte_flow_item_vxlan_gpe);
+		retval = sizeof(struct rte_gre_hdr);
 		break;
 	case RTE_FLOW_ITEM_TYPE_MPLS:
-		retval = sizeof(struct rte_flow_item_mpls);
+		retval = sizeof(struct rte_mpls_hdr);
 		break;
 	case RTE_FLOW_ITEM_TYPE_VOID: /* Fall through. */
 	default:
@@ -2967,7 +2964,7 @@  flow_dv_convert_encap_data(const struct rte_flow_item *items, uint8_t *buf,
 					  RTE_FLOW_ERROR_TYPE_ACTION,
 					  NULL, "invalid empty data");
 	for (; items->type != RTE_FLOW_ITEM_TYPE_END; items++) {
-		len = flow_dv_get_item_len(items->type);
+		len = flow_dv_get_item_hdr_len(items->type);
 		if (len + temp_size > MLX5_ENCAP_MAX_LEN)
 			return rte_flow_error_set(error, EINVAL,
 						  RTE_FLOW_ERROR_TYPE_ACTION,