From: Jian Shen <shenjia...@huawei.com>
Currently, when adding flow director rule, it missed to set
rule state. Which may cause the rule state in software is
unconsistent with hardware.
Fixes: fc4243b8de8b ("net: hns3: refactor flow director configuration")
Signed-off-by: Jian Shen <shenjia...@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazh...@huawei.com>
---
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index d639519..08da3e2 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -6439,7 +6439,8 @@ static int hclge_add_fd_entry_common(struct hclge_dev
*hdev,
if (ret)
goto out;
- hclge_update_fd_list(hdev, HCLGE_FD_ACTIVE, rule->location, rule);
+ rule->state = HCLGE_FD_ACTIVE;
+ hclge_update_fd_list(hdev, rule->state, rule->location, rule);
hdev->fd_active_type = rule->rule_type;
out:
@@ -7002,6 +7003,7 @@ static void hclge_fd_build_arfs_rule(const struct
hclge_fd_rule_tuples *tuples,
rule->action = 0;
rule->vf_id = 0;
rule->rule_type = HCLGE_FD_ARFS_ACTIVE;
+ rule->state = HCLGE_FD_TO_ADD;
if (tuples->ether_proto == ETH_P_IP) {
if (tuples->ip_proto == IPPROTO_TCP)
rule->flow_type = TCP_V4_FLOW;
@@ -7064,8 +7066,7 @@ static int hclge_add_fd_entry_by_arfs(struct hnae3_handle
*handle, u16 queue_id,
rule->arfs.flow_id = flow_id;
rule->queue_id = queue_id;
hclge_fd_build_arfs_rule(&new_tuples, rule);
- hclge_update_fd_list(hdev, HCLGE_FD_TO_ADD, rule->location,
- rule);
+ hclge_update_fd_list(hdev, rule->state, rule->location, rule);
hdev->fd_active_type = HCLGE_FD_ARFS_ACTIVE;
} else if (rule->queue_id != queue_id) {
rule->queue_id = queue_id;
--
2.7.4
