From 731815e720ae7e47a19753e00ea80651b2d52b3b Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Wed, 22 Apr 2020 06:16:07 -0700 Subject: [PATCH] qede: Add support for handling the pcie errors. The error recovery is handled by management firmware (MFW) with the help of qed/qede drivers. Upon detecting the errors, driver informs MFW about this event which in turn starts a recovery process. MFW sends ERROR_RECOVERY notification to the driver which performs the required cleanup/recovery from the driver side. Signed-off-by: Sudarsana Reddy Kalluru Acked-by: Jakub Kicinski Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qede/qede.h | 1 + drivers/net/ethernet/qlogic/qede/qede_main.c | 68 +++++++++++++++++++- 2 files changed, 68 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h index 234c6f30effb..1a708f95ce94 100644 --- a/drivers/net/ethernet/qlogic/qede/qede.h +++ b/drivers/net/ethernet/qlogic/qede/qede.h @@ -485,6 +485,7 @@ struct qede_fastpath { #define QEDE_SP_RECOVERY 0 #define QEDE_SP_RX_MODE 1 +#define QEDE_SP_AER 7 #ifdef CONFIG_RFS_ACCEL int qede_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb, diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index 34fa3917eb33..9b456198cb50 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -60,6 +60,7 @@ #include #include #include +#include #include "qede.h" #include "qede_ptp.h" @@ -124,6 +125,8 @@ static const struct pci_device_id qede_pci_tbl[] = { MODULE_DEVICE_TABLE(pci, qede_pci_tbl); static int qede_probe(struct pci_dev *pdev, const struct pci_device_id *id); +static pci_ers_result_t +qede_io_error_detected(struct pci_dev *pdev, pci_channel_state_t state); #define TX_TIMEOUT (5 * HZ) @@ -203,6 +206,10 @@ static int qede_sriov_configure(struct pci_dev *pdev, int num_vfs_param) } #endif +static const struct pci_error_handlers qede_err_handler = { + .error_detected = qede_io_error_detected, +}; + static struct pci_driver qede_pci_driver = { .name = "qede", .id_table = qede_pci_tbl, @@ -212,6 +219,7 @@ static struct pci_driver qede_pci_driver = { #ifdef CONFIG_QED_SRIOV .sriov_configure = qede_sriov_configure, #endif + .err_handler = &qede_err_handler, }; static struct qed_eth_cb_ops qede_ll_ops = { @@ -974,7 +982,8 @@ static void qede_sp_task(struct work_struct *work) /* SRIOV must be disabled outside the lock to avoid a deadlock. * The recovery of the active VFs is currently not supported. */ - qede_sriov_configure(edev->pdev, 0); + if (pci_num_vf(edev->pdev)) + qede_sriov_configure(edev->pdev, 0); #endif qede_lock(edev); qede_recovery_handler(edev); @@ -994,6 +1003,17 @@ static void qede_sp_task(struct work_struct *work) } #endif __qede_unlock(edev); + + if (test_and_clear_bit(QEDE_SP_AER, &edev->sp_flags)) { +#ifdef CONFIG_QED_SRIOV + /* SRIOV must be disabled outside the lock to avoid a deadlock. + * The recovery of the active VFs is currently not supported. + */ + if (pci_num_vf(edev->pdev)) + qede_sriov_configure(edev->pdev, 0); +#endif + edev->ops->common->recovery_process(edev->cdev); + } } static void qede_update_pf_params(struct qed_dev *cdev) @@ -2579,3 +2599,49 @@ static void qede_get_eth_tlv_data(void *dev, void *data) etlv->num_txqs_full_set = true; etlv->num_rxqs_full_set = true; } + +/** + * qede_io_error_detected - called when PCI error is detected + * @pdev: Pointer to PCI device + * @state: The current pci connection state + * + * This function is called after a PCI bus error affecting + * this device has been detected. + */ +static pci_ers_result_t +qede_io_error_detected(struct pci_dev *pdev, pci_channel_state_t state) +{ + struct net_device *dev = pci_get_drvdata(pdev); + struct qede_dev *edev = netdev_priv(dev); + + if (!edev) + return PCI_ERS_RESULT_NONE; + + DP_NOTICE(edev, "IO error detected [%d]\n", state); + + __qede_lock(edev); + if (edev->state == QEDE_STATE_RECOVERY) { + DP_NOTICE(edev, "Device already in the recovery state\n"); + __qede_unlock(edev); + return PCI_ERS_RESULT_NONE; + } + + /* PF handles the recovery of its VFs */ + if (IS_VF(edev)) { + DP_VERBOSE(edev, QED_MSG_IOV, + "VF recovery is handled by its PF\n"); + __qede_unlock(edev); + return PCI_ERS_RESULT_RECOVERED; + } + + /* Close OS Tx */ + netif_tx_disable(edev->ndev); + netif_carrier_off(edev->ndev); + + set_bit(QEDE_SP_AER, &edev->sp_flags); + schedule_delayed_work(&edev->sp_task, 0); + + __qede_unlock(edev); + + return PCI_ERS_RESULT_CAN_RECOVER; +} -- 2.25.1