net: ena: Add a counter for driver's reset failures

This patch adds a counter to the ena_adapter struct in
order to keep track of reset failures.
The counter is incremented every time either ena_restore_device()
or ena_destroy_device() fail.
Signed-off-by: Osama Abboud <osamaabb@amazon.com>
Signed-off-by: David Arinzon <darinzon@amazon.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240512134637.25299-2-darinzon@amazon.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

drivers/net/ethernet/amazon/ena/ena_ethtool.c

@@ -49,6 +49,7 @@ static const struct ena_stats ena_stats_global_strings[] = {
 	ENA_STAT_GLOBAL_ENTRY(interface_up),
 	ENA_STAT_GLOBAL_ENTRY(interface_down),
 	ENA_STAT_GLOBAL_ENTRY(admin_q_pause),
+	ENA_STAT_GLOBAL_ENTRY(reset_fail),
 };
 
 static const struct ena_stats ena_stats_eni_strings[] = {

drivers/net/ethernet/amazon/ena/ena_netdev.c

@@ -42,7 +42,7 @@ MODULE_DEVICE_TABLE(pci, ena_pci_tbl);
 
 static int ena_rss_init_default(struct ena_adapter *adapter);
 static void check_for_admin_com_state(struct ena_adapter *adapter);
-static void ena_destroy_device(struct ena_adapter *adapter, bool graceful);
+static int ena_destroy_device(struct ena_adapter *adapter, bool graceful);
 static int ena_restore_device(struct ena_adapter *adapter);
 
 static void ena_tx_timeout(struct net_device *dev, unsigned int txqueue)
@@ -3235,14 +3235,15 @@ static int ena_enable_msix_and_set_admin_interrupts(struct ena_adapter *adapter)
 	return rc;
 }
 
-static void ena_destroy_device(struct ena_adapter *adapter, bool graceful)
+static int ena_destroy_device(struct ena_adapter *adapter, bool graceful)
 {
 	struct net_device *netdev = adapter->netdev;
 	struct ena_com_dev *ena_dev = adapter->ena_dev;
 	bool dev_up;
+	int rc = 0;
 
 	if (!test_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags))
-		return;
+		return 0;
 
 	netif_carrier_off(netdev);
 
@@ -3260,7 +3261,7 @@ static void ena_destroy_device(struct ena_adapter *adapter, bool graceful)
 	 *  and device is up, ena_down() already reset the device.
 	 */
 	if (!(test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags) && dev_up))
-		ena_com_dev_reset(adapter->ena_dev, adapter->reset_reason);
+		rc = ena_com_dev_reset(adapter->ena_dev, adapter->reset_reason);
 
 	ena_free_mgmnt_irq(adapter);
 
@@ -3279,6 +3280,8 @@ static void ena_destroy_device(struct ena_adapter *adapter, bool graceful)
 
 	clear_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags);
 	clear_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags);
+
+	return rc;
 }
 
 static int ena_restore_device(struct ena_adapter *adapter)
@@ -3355,14 +3358,17 @@ static int ena_restore_device(struct ena_adapter *adapter)
 
 static void ena_fw_reset_device(struct work_struct *work)
 {
+	int rc = 0;
+
 	struct ena_adapter *adapter =
 		container_of(work, struct ena_adapter, reset_task);
 
 	rtnl_lock();
 
 	if (likely(test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags))) {
-		ena_destroy_device(adapter, false);
-		ena_restore_device(adapter);
+		rc |= ena_destroy_device(adapter, false);
+		rc |= ena_restore_device(adapter);
+		adapter->dev_stats.reset_fail += !!rc;
 
 		dev_err(&adapter->pdev->dev, "Device reset completed successfully\n");
 	}

drivers/net/ethernet/amazon/ena/ena_netdev.h

@@ -290,6 +290,7 @@ struct ena_stats_dev {
 	u64 admin_q_pause;
 	u64 rx_drops;
 	u64 tx_drops;
+	u64 reset_fail;
 };
 
 enum ena_flags_t {