net: hns3: fix aRFS FD rules leftover after add a user FD rule

When user had created a FD rule, all the aRFS rules should be clear up.
HNS3 process flow as below:
1.get spin lock of fd_ruls_list
2.clear up all aRFS rules
3.release lock
4.get spin lock of fd_ruls_list
5.creat a rules
6.release lock;

There is a short period of time between step 3 and step 4, which would
creatting some new aRFS FD rules if driver was receiving packet.
So refactor the fd_rule_lock to fix it.

Fixes: 4412288 ("net: hns3: refine the flow director handle")
Signed-off-by: Guojia Liao <[email protected]>
Signed-off-by: Huazhong Tan <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

Author: Guojia Liao

drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c

@@ -5806,9 +5806,9 @@ static int hclge_add_fd_entry(struct hnae3_handle *handle,
 	/* to avoid rule conflict, when user configure rule by ethtool,
 	 * we need to clear all arfs rules
 	 */
+	spin_lock_bh(&hdev->fd_rule_lock);
 	hclge_clear_arfs_rules(handle);
 
-	spin_lock_bh(&hdev->fd_rule_lock);
 	ret = hclge_fd_config_rule(hdev, rule);
 
 	spin_unlock_bh(&hdev->fd_rule_lock);
@@ -5851,6 +5851,7 @@ static int hclge_del_fd_entry(struct hnae3_handle *handle,
 	return ret;
 }
 
+/* make sure being called after lock up with fd_rule_lock */
 static void hclge_del_all_fd_entries(struct hnae3_handle *handle,
 				     bool clear_list)
 {
@@ -5863,7 +5864,6 @@ static void hclge_del_all_fd_entries(struct hnae3_handle *handle,
 	if (!hnae3_dev_fd_supported(hdev))
 		return;
 
-	spin_lock_bh(&hdev->fd_rule_lock);
 	for_each_set_bit(location, hdev->fd_bmap,
 			 hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1])
 		hclge_fd_tcam_config(hdev, HCLGE_FD_STAGE_1, true, location,
@@ -5880,8 +5880,6 @@ static void hclge_del_all_fd_entries(struct hnae3_handle *handle,
 		bitmap_zero(hdev->fd_bmap,
 			    hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1]);
 	}
-
-	spin_unlock_bh(&hdev->fd_rule_lock);
 }
 
 static int hclge_restore_fd_entries(struct hnae3_handle *handle)
@@ -6263,7 +6261,7 @@ static int hclge_add_fd_entry_by_arfs(struct hnae3_handle *handle, u16 queue_id,
 				      u16 flow_id, struct flow_keys *fkeys)
 {
 	struct hclge_vport *vport = hclge_get_vport(handle);
-	struct hclge_fd_rule_tuples new_tuples;
+	struct hclge_fd_rule_tuples new_tuples = {};
 	struct hclge_dev *hdev = vport->back;
 	struct hclge_fd_rule *rule;
 	u16 tmp_queue_id;
@@ -6273,19 +6271,17 @@ static int hclge_add_fd_entry_by_arfs(struct hnae3_handle *handle, u16 queue_id,
 	if (!hnae3_dev_fd_supported(hdev))
 		return -EOPNOTSUPP;
 
-	memset(&new_tuples, 0, sizeof(new_tuples));
-	hclge_fd_get_flow_tuples(fkeys, &new_tuples);
-
-	spin_lock_bh(&hdev->fd_rule_lock);
-
 	/* when there is already fd rule existed add by user,
 	 * arfs should not work
 	 */
+	spin_lock_bh(&hdev->fd_rule_lock);
 	if (hdev->fd_active_type == HCLGE_FD_EP_ACTIVE) {
 		spin_unlock_bh(&hdev->fd_rule_lock);
 		return -EOPNOTSUPP;
 	}
 
+	hclge_fd_get_flow_tuples(fkeys, &new_tuples);
+
 	/* check is there flow director filter existed for this flow,
 	 * if not, create a new filter for it;
 	 * if filter exist with different queue id, modify the filter;
@@ -6368,6 +6364,7 @@ static void hclge_rfs_filter_expire(struct hclge_dev *hdev)
 #endif
 }
 
+/* make sure being called after lock up with fd_rule_lock */
 static void hclge_clear_arfs_rules(struct hnae3_handle *handle)
 {
 #ifdef CONFIG_RFS_ACCEL
@@ -6420,10 +6417,14 @@ static void hclge_enable_fd(struct hnae3_handle *handle, bool enable)
 
 	hdev->fd_en = enable;
 	clear = hdev->fd_active_type == HCLGE_FD_ARFS_ACTIVE;
-	if (!enable)
+
+	if (!enable) {
+		spin_lock_bh(&hdev->fd_rule_lock);
 		hclge_del_all_fd_entries(handle, clear);
-	else
+		spin_unlock_bh(&hdev->fd_rule_lock);
+	} else {
 		hclge_restore_fd_entries(handle);
+	}
 }
 
 static void hclge_cfg_mac_mode(struct hclge_dev *hdev, bool enable)
@@ -6886,8 +6887,9 @@ static void hclge_ae_stop(struct hnae3_handle *handle)
 	int i;
 
 	set_bit(HCLGE_STATE_DOWN, &hdev->state);
-
+	spin_lock_bh(&hdev->fd_rule_lock);
 	hclge_clear_arfs_rules(handle);
+	spin_unlock_bh(&hdev->fd_rule_lock);
 
 	/* If it is not PF reset, the firmware will disable the MAC,
 	 * so it only need to stop phy here.