net: hns3: delay setting of reset level for hw errors until slot_reset is called

Presently the error handling code sets the reset level required
for the recovery of the hw errors to the reset framework in the
error_detected AER callback. However the rest_event would be
called later from the slot_reset callback. This can cause issue
of using the wrong reset_level if a high priority reset request
occur before the slot_reset is called.

This patch delays setting of the reset level, required
for the hw errors, to the reset framework until the
slot_reset is called.

Reported-by: Salil Mehta <salil.mehta@huawei.com>
Signed-off-by: Shiju Jose <shiju.jose@huawei.com>
Signed-off-by: Weihang Li <liweihang@hisilicon.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 10 insertions, 4 deletions

diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
index b7ba8932d30c..f3e90304f2fb 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
@@ -41,6 +41,8 @@ static int hclge_set_umv_space(struct hclge_dev *hdev, u16 space_size,
 			       u16 *allocated_size, bool is_alloc);
 static void hclge_rfs_filter_expire(struct hclge_dev *hdev);
 static void hclge_clear_arfs_rules(struct hnae3_handle *handle);
+static enum hnae3_reset_type hclge_get_reset_level(struct hnae3_ae_dev *ae_dev,
+						   unsigned long *addr);
 
 static struct hnae3_ae_algo ae_algo;
 
@@ -3066,10 +3068,11 @@ static void hclge_do_reset(struct hclge_dev *hdev)
 	}
 }
 
-static enum hnae3_reset_type hclge_get_reset_level(struct hclge_dev *hdev,
+static enum hnae3_reset_type hclge_get_reset_level(struct hnae3_ae_dev *ae_dev,
 						   unsigned long *addr)
 {
 	enum hnae3_reset_type rst_level = HNAE3_NONE_RESET;
+	struct hclge_dev *hdev = ae_dev->priv;
 
 	/* first, resolve any unknown reset type to the known type(s) */
 	if (test_bit(HNAE3_UNKNOWN_RESET, addr)) {
@@ -3398,7 +3401,7 @@ static void hclge_reset_event(struct pci_dev *pdev, struct hnae3_handle *handle)
 		return;
 	else if (hdev->default_reset_request)
 		hdev->reset_level =
-			hclge_get_reset_level(hdev,
+			hclge_get_reset_level(ae_dev,
 					      &hdev->default_reset_request);
 	else if (time_after(jiffies, (hdev->last_reset_time + 4 * 5 * HZ)))
 		hdev->reset_level = HNAE3_FUNC_RESET;
@@ -3434,6 +3437,8 @@ static void hclge_reset_timer(struct timer_list *t)
 
 static void hclge_reset_subtask(struct hclge_dev *hdev)
 {
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev);
+
 	/* check if there is any ongoing reset in the hardware. This status can
 	 * be checked from reset_pending. If there is then, we need to wait for
 	 * hardware to complete reset.
@@ -3444,12 +3449,12 @@ static void hclge_reset_subtask(struct hclge_dev *hdev)
 	 *       now.
 	 */
 	hdev->last_reset_time = jiffies;
-	hdev->reset_type = hclge_get_reset_level(hdev, &hdev->reset_pending);
+	hdev->reset_type = hclge_get_reset_level(ae_dev, &hdev->reset_pending);
 	if (hdev->reset_type != HNAE3_NONE_RESET)
 		hclge_reset(hdev);
 
 	/* check if we got any *new* reset requests to be honored */
-	hdev->reset_type = hclge_get_reset_level(hdev, &hdev->reset_request);
+	hdev->reset_type = hclge_get_reset_level(ae_dev, &hdev->reset_request);
 	if (hdev->reset_type != HNAE3_NONE_RESET)
 		hclge_do_reset(hdev);
 
@@ -9231,6 +9236,7 @@ static const struct hnae3_ae_ops hclge_ops = {
 	.set_vf_vlan_filter = hclge_set_vf_vlan_filter,
 	.enable_hw_strip_rxvtag = hclge_en_hw_strip_rxvtag,
 	.reset_event = hclge_reset_event,
+	.get_reset_level = hclge_get_reset_level,
 	.set_default_reset_request = hclge_set_def_reset_request,
 	.get_tqps_and_rss_info = hclge_get_tqps_and_rss_info,
 	.set_channels = hclge_set_channels,