[PATCH 03/19] net/hns3: fix the lock protection of RSS flow rule

[Dongdong Liu]

From: Huisong Li <lihuis...@huawei.com>

RSS flow rules are saved in RSS filter linked list. The linked
list is modified by rte_flow API and is used to restore RSS rules
during reset process. So this patch uses 'hw->flows_lock' to protect
the configuration and recovery of RSS rule.

Fixes: c37ca66f2b27 ("net/hns3: support RSS")
Cc: sta...@dpdk.org

Signed-off-by: Huisong Li <lihuis...@huawei.com>
Signed-off-by: Dongdong Liu <liudongdo...@huawei.com>
---
 drivers/net/hns3/hns3_flow.c | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)

diff --git a/drivers/net/hns3/hns3_flow.c b/drivers/net/hns3/hns3_flow.c
index 2fb83f756a..162a48e590 100644
--- a/drivers/net/hns3/hns3_flow.c
+++ b/drivers/net/hns3/hns3_flow.c
@@ -1596,27 +1596,20 @@ hns3_config_rss_filter(struct rte_eth_dev *dev,
                hns3_warn(hw, "Config queue numbers %u are beyond the scope of 
truncated",
                          rss_flow_conf.queue_num);
        hns3_info(hw, "Max of contiguous %u PF queues are configured", num);
-
-       rte_spinlock_lock(&hw->lock);
        if (num) {
                ret = hns3_update_indir_table(dev, &rss_flow_conf, num);
                if (ret)
-                       goto rss_config_err;
+                       return ret;
        }
 
        /* Set hash algorithm and flow types by the user's config */
        ret = hns3_hw_rss_hash_set(hw, &rss_flow_conf);
        if (ret)
-               goto rss_config_err;
+               return ret;
 
        ret = hns3_rss_conf_copy(rss_info, &rss_flow_conf);
-       if (ret) {
+       if (ret)
                hns3_err(hw, "RSS config init fail(%d)", ret);
-               goto rss_config_err;
-       }
-
-rss_config_err:
-       rte_spinlock_unlock(&hw->lock);
 
        return ret;
 }
@@ -1661,6 +1654,7 @@ hns3_restore_rss_filter(struct rte_eth_dev *dev)
        struct hns3_hw *hw = &hns->hw;
        int ret = 0;
 
+       pthread_mutex_lock(&hw->flows_lock);
        TAILQ_FOREACH(filter, &hw->flow_rss_list, entries) {
                if (!filter->filter_info.valid)
                        continue;
@@ -1673,6 +1667,8 @@ hns3_restore_rss_filter(struct rte_eth_dev *dev)
        }
 
 out:
+       pthread_mutex_unlock(&hw->flows_lock);
+
        return ret;
 }
 
-- 
2.22.0