be2net: Fix crash on 2nd invocation of PCI AER/EEH error_detected hook

During a PCI EEH/AER error recovery flow, if the device did not successfully
restart, the error_detected() hook may be called a second time with a
"perm_failure" state. This patch skips over driver cleanup for the second
invocation of the callback.

Also, Lancer error recovery code is fixed-up to handle these changes.

Signed-off-by: Kalesh AP <kalesh.purayil@emulex.com>
Signed-off-by: Somnath kotur <somnath.kotur@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c
index a236ecd27cf384dea6af9e0a992a075dab306a5d..1db2df61b8af65d5b0ca32534b8b6a96e9a2c794 100644 (file)
--- a/drivers/net/ethernet/emulex/benet/be_cmds.c
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.c
@@ -562,7 +562,7 @@ int lancer_test_and_set_rdy_state(struct be_adapter *adapter)
 
        resource_error = lancer_provisioning_error(adapter);
        if (resource_error)
-               return -1;
+               return -EAGAIN;
 
        status = lancer_wait_ready(adapter);
        if (!status) {
@@ -590,8 +590,8 @@ int lancer_test_and_set_rdy_state(struct be_adapter *adapter)
         * when PF provisions resources.
         */
        resource_error = lancer_provisioning_error(adapter);
-       if (status == -1 && !resource_error)
-               adapter->eeh_error = true;
+       if (resource_error)
+               status = -EAGAIN;
 
        return status;
 }

diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 32a6927ca977e38503f0cd240514aa271309791c..8bc1b21b1c790b44cffc347d7fe365f22f3a7ee2 100644 (file)
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -4098,6 +4098,7 @@ static int be_get_initial_config(struct be_adapter *adapter)
 
 static int lancer_recover_func(struct be_adapter *adapter)
 {
+       struct device *dev = &adapter->pdev->dev;
        int status;
 
        status = lancer_test_and_set_rdy_state(adapter);
@@ -4109,8 +4110,7 @@ static int lancer_recover_func(struct be_adapter *adapter)
 
        be_clear(adapter);
 
-       adapter->hw_error = false;
-       adapter->fw_timeout = false;
+       be_clear_all_error(adapter);
 
        status = be_setup(adapter);
        if (status)
@@ -4122,13 +4122,13 @@ static int lancer_recover_func(struct be_adapter *adapter)
                        goto err;
        }
 
-       dev_err(&adapter->pdev->dev,
-               "Adapter SLIPORT recovery succeeded\n");
+       dev_err(dev, "Error recovery successful\n");
        return 0;
 err:
-       if (adapter->eeh_error)
-               dev_err(&adapter->pdev->dev,
-                       "Adapter SLIPORT recovery failed\n");
+       if (status == -EAGAIN)
+               dev_err(dev, "Waiting for resource provisioning\n");
+       else
+               dev_err(dev, "Error recovery failed\n");
 
        return status;
 }
@@ -4137,28 +4137,27 @@ static void be_func_recovery_task(struct work_struct *work)
 {
        struct be_adapter *adapter =
                container_of(work, struct be_adapter,  func_recovery_work.work);
-       int status;
+       int status = 0;
 
        be_detect_error(adapter);
 
        if (adapter->hw_error && lancer_chip(adapter)) {
 
-               if (adapter->eeh_error)
-                       goto out;
-
                rtnl_lock();
                netif_device_detach(adapter->netdev);
                rtnl_unlock();
 
                status = lancer_recover_func(adapter);
-
                if (!status)
                        netif_device_attach(adapter->netdev);
        }
 
-out:
-       schedule_delayed_work(&adapter->func_recovery_work,
-                             msecs_to_jiffies(1000));
+       /* In Lancer, for all errors other than provisioning error (-EAGAIN),
+        * no need to attempt further recovery.
+        */
+       if (!status || status == -EAGAIN)
+               schedule_delayed_work(&adapter->func_recovery_work,
+                                     msecs_to_jiffies(1000));
 }
 
 static void be_worker(struct work_struct *work)
@@ -4441,20 +4440,19 @@ static pci_ers_result_t be_eeh_err_detected(struct pci_dev *pdev,
 
        dev_err(&adapter->pdev->dev, "EEH error detected\n");
 
-       adapter->eeh_error = true;
-
-       cancel_delayed_work_sync(&adapter->func_recovery_work);
+       if (!adapter->eeh_error) {
+               adapter->eeh_error = true;
 
-       rtnl_lock();
-       netif_device_detach(netdev);
-       rtnl_unlock();
+               cancel_delayed_work_sync(&adapter->func_recovery_work);
 
-       if (netif_running(netdev)) {
                rtnl_lock();
-               be_close(netdev);
+               netif_device_detach(netdev);
+               if (netif_running(netdev))
+                       be_close(netdev);
                rtnl_unlock();
+
+               be_clear(adapter);
        }
-       be_clear(adapter);
 
        if (state == pci_channel_io_perm_failure)
                return PCI_ERS_RESULT_DISCONNECT;
@@ -4479,7 +4477,6 @@ static pci_ers_result_t be_eeh_reset(struct pci_dev *pdev)
        int status;
 
        dev_info(&adapter->pdev->dev, "EEH reset\n");
-       be_clear_all_error(adapter);
 
        status = pci_enable_device(pdev);
        if (status)
@@ -4497,6 +4494,7 @@ static pci_ers_result_t be_eeh_reset(struct pci_dev *pdev)
                return PCI_ERS_RESULT_DISCONNECT;
 
        pci_cleanup_aer_uncorrect_error_status(pdev);
+       be_clear_all_error(adapter);
        return PCI_ERS_RESULT_RECOVERED;
 }