From: Long Wu <long...@corigine.com>
If the flow rule dose not have 'spec' in IP layer, NFP will
set all IP flow control message type to 0.
Move up the IP flow control message setting code to fix it.
Fixes: 42eabda06b0f ("net/nfp: support IPv4 flow item")
Fixes: 9f27cb889246 ("net/nfp: support IPv6 flow item")
Cc: Chaoyong He <chaoyong...@corigine.com>
Cc: sta...@dpdk.org
Signed-off-by: Long Wu <long...@corigine.com>
Reviewed-by: Chaoyong He <chaoyong...@corigine.com>
---
drivers/net/nfp/nfp_net_flow.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/net/nfp/nfp_net_flow.c b/drivers/net/nfp/nfp_net_flow.c
index 3b33f3b6e9..bd983aaf6a 100644
--- a/drivers/net/nfp/nfp_net_flow.c
+++ b/drivers/net/nfp/nfp_net_flow.c
@@ -237,6 +237,8 @@ nfp_net_flow_merge_ipv4(struct rte_flow *nfp_flow,
const struct rte_flow_item_ipv4 *mask;
const struct rte_flow_item_ipv4 *spec;
+ nfp_flow->payload.cmsg_type = NFP_NET_CFG_MBOX_CMD_FS_ADD_V4;
+
spec = item->spec;
if (spec == NULL) {
PMD_DRV_LOG(DEBUG, "NFP flow merge ipv4: no item->spec!");
@@ -245,7 +247,6 @@ nfp_net_flow_merge_ipv4(struct rte_flow *nfp_flow,
mask = (item->mask != NULL) ? item->mask : proc->mask_default;
- nfp_flow->payload.cmsg_type = NFP_NET_CFG_MBOX_CMD_FS_ADD_V4;
ipv4 = (struct nfp_net_cmsg_match_v4 *)nfp_flow->payload.match_data;
ipv4->l4_protocol_mask = mask->hdr.next_proto_id;
@@ -269,6 +270,8 @@ nfp_net_flow_merge_ipv6(struct rte_flow *nfp_flow,
const struct rte_flow_item_ipv6 *mask;
const struct rte_flow_item_ipv6 *spec;
+ nfp_flow->payload.cmsg_type = NFP_NET_CFG_MBOX_CMD_FS_ADD_V6;
+
spec = item->spec;
if (spec == NULL) {
PMD_DRV_LOG(DEBUG, "NFP flow merge ipv6: no item->spec!");
@@ -277,7 +280,6 @@ nfp_net_flow_merge_ipv6(struct rte_flow *nfp_flow,
mask = (item->mask != NULL) ? item->mask : proc->mask_default;
- nfp_flow->payload.cmsg_type = NFP_NET_CFG_MBOX_CMD_FS_ADD_V6;
ipv6 = (struct nfp_net_cmsg_match_v6 *)nfp_flow->payload.match_data;
ipv6->l4_protocol_mask = mask->hdr.proto;
--
2.39.1
