net: hns3: add barrier in vf mailbox reply process

[ Upstream commit ac92c0a ]

In hclgevf_mbx_handler() and hclgevf_get_mbx_resp() functions,
there is a typical store-store and load-load scenario between
received_resp and additional_info. This patch adds barrier
to fix the problem.

Fixes: 4671042 ("net: hns3: add match_id to check mailbox response from PF to VF")
Signed-off-by: Yonglong Liu <liuyonglong@huawei.com>
Signed-off-by: Jijie Shao <shaojijie@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c

@@ -63,6 +63,9 @@ static int hclgevf_get_mbx_resp(struct hclgevf_dev *hdev, u16 code0, u16 code1,
 		i++;
 	}
 
+	/* ensure additional_info will be seen after received_resp */
+	smp_rmb();
+
 	if (i >= HCLGEVF_MAX_TRY_TIMES) {
 		dev_err(&hdev->pdev->dev,
 			"VF could not get mbx(%u,%u) resp(=%d) from PF in %d tries\n",
@@ -178,6 +181,10 @@ static void hclgevf_handle_mbx_response(struct hclgevf_dev *hdev,
 	resp->resp_status = hclgevf_resp_to_errno(resp_status);
 	memcpy(resp->additional_info, req->msg.resp_data,
 	       HCLGE_MBX_MAX_RESP_DATA_SIZE * sizeof(u8));
+
+	/* ensure additional_info will be seen before setting received_resp */
+	smp_wmb();
+
 	if (match_id) {
 		/* If match_id is not zero, it means PF support match_id.
 		 * if the match_id is right, VF get the right response, or