diff options
Diffstat (limited to 'arch/powerpc/kernel/eeh_driver.c')
| -rw-r--r-- | arch/powerpc/kernel/eeh_driver.c | 280 |
1 files changed, 224 insertions, 56 deletions
diff --git a/arch/powerpc/kernel/eeh_driver.c b/arch/powerpc/kernel/eeh_driver.c index 89623962c727..d9279d0ee9f5 100644 --- a/arch/powerpc/kernel/eeh_driver.c +++ b/arch/powerpc/kernel/eeh_driver.c @@ -27,6 +27,7 @@ #include <linux/irq.h> #include <linux/module.h> #include <linux/pci.h> +#include <linux/pci_hotplug.h> #include <asm/eeh.h> #include <asm/eeh_event.h> #include <asm/ppc-pci.h> @@ -81,23 +82,6 @@ static const char *pci_ers_result_name(enum pci_ers_result result) } }; -static __printf(2, 3) void eeh_edev_info(const struct eeh_dev *edev, - const char *fmt, ...) -{ - struct va_format vaf; - va_list args; - - va_start(args, fmt); - - vaf.fmt = fmt; - vaf.va = &args; - - printk(KERN_INFO "EEH: PE#%x (PCI %s): %pV\n", edev->pe_config_addr, - edev->pdev ? dev_name(&edev->pdev->dev) : "none", &vaf); - - va_end(args); -} - static enum pci_ers_result pci_ers_merge_result(enum pci_ers_result old, enum pci_ers_result new) { @@ -113,8 +97,16 @@ static bool eeh_dev_removed(struct eeh_dev *edev) static bool eeh_edev_actionable(struct eeh_dev *edev) { - return (edev->pdev && !eeh_dev_removed(edev) && - !eeh_pe_passed(edev->pe)); + if (!edev->pdev) + return false; + if (edev->pdev->error_state == pci_channel_io_perm_failure) + return false; + if (eeh_dev_removed(edev)) + return false; + if (eeh_pe_passed(edev->pe)) + return false; + + return true; } /** @@ -214,12 +206,12 @@ static void eeh_enable_irq(struct eeh_dev *edev) } } -static void *eeh_dev_save_state(struct eeh_dev *edev, void *userdata) +static void eeh_dev_save_state(struct eeh_dev *edev, void *userdata) { struct pci_dev *pdev; if (!edev) - return NULL; + return; /* * We cannot access the config space on some adapters. @@ -229,14 +221,13 @@ static void *eeh_dev_save_state(struct eeh_dev *edev, void *userdata) * device is created. */ if (edev->pe && (edev->pe->state & EEH_PE_CFG_RESTRICTED)) - return NULL; + return; pdev = eeh_dev_to_pci_dev(edev); if (!pdev) - return NULL; + return; pci_save_state(pdev); - return NULL; } static void eeh_set_channel_state(struct eeh_pe *root, enum pci_channel_state s) @@ -274,20 +265,27 @@ static void eeh_set_irq_state(struct eeh_pe *root, bool enable) } typedef enum pci_ers_result (*eeh_report_fn)(struct eeh_dev *, + struct pci_dev *, struct pci_driver *); static void eeh_pe_report_edev(struct eeh_dev *edev, eeh_report_fn fn, enum pci_ers_result *result) { + struct pci_dev *pdev; struct pci_driver *driver; enum pci_ers_result new_result; - if (!edev->pdev) { + pci_lock_rescan_remove(); + pdev = edev->pdev; + if (pdev) + get_device(&pdev->dev); + pci_unlock_rescan_remove(); + if (!pdev) { eeh_edev_info(edev, "no device"); return; } - device_lock(&edev->pdev->dev); + device_lock(&pdev->dev); if (eeh_edev_actionable(edev)) { - driver = eeh_pcid_get(edev->pdev); + driver = eeh_pcid_get(pdev); if (!driver) eeh_edev_info(edev, "no driver"); @@ -296,7 +294,7 @@ static void eeh_pe_report_edev(struct eeh_dev *edev, eeh_report_fn fn, else if (edev->mode & EEH_DEV_NO_HANDLER) eeh_edev_info(edev, "driver bound too late"); else { - new_result = fn(edev, driver); + new_result = fn(edev, pdev, driver); eeh_edev_info(edev, "%s driver reports: '%s'", driver->name, pci_ers_result_name(new_result)); @@ -305,12 +303,15 @@ static void eeh_pe_report_edev(struct eeh_dev *edev, eeh_report_fn fn, new_result); } if (driver) - eeh_pcid_put(edev->pdev); + eeh_pcid_put(pdev); } else { - eeh_edev_info(edev, "not actionable (%d,%d,%d)", !!edev->pdev, + eeh_edev_info(edev, "not actionable (%d,%d,%d)", !!pdev, !eeh_dev_removed(edev), !eeh_pe_passed(edev->pe)); } - device_unlock(&edev->pdev->dev); + device_unlock(&pdev->dev); + if (edev->pdev != pdev) + eeh_edev_warn(edev, "Device changed during processing!\n"); + put_device(&pdev->dev); } static void eeh_pe_report(const char *name, struct eeh_pe *root, @@ -337,20 +338,20 @@ static void eeh_pe_report(const char *name, struct eeh_pe *root, * Report an EEH error to each device driver. */ static enum pci_ers_result eeh_report_error(struct eeh_dev *edev, + struct pci_dev *pdev, struct pci_driver *driver) { enum pci_ers_result rc; - struct pci_dev *dev = edev->pdev; if (!driver->err_handler->error_detected) return PCI_ERS_RESULT_NONE; eeh_edev_info(edev, "Invoking %s->error_detected(IO frozen)", driver->name); - rc = driver->err_handler->error_detected(dev, pci_channel_io_frozen); + rc = driver->err_handler->error_detected(pdev, pci_channel_io_frozen); edev->in_error = true; - pci_uevent_ers(dev, PCI_ERS_RESULT_NONE); + pci_uevent_ers(pdev, PCI_ERS_RESULT_NONE); return rc; } @@ -363,12 +364,13 @@ static enum pci_ers_result eeh_report_error(struct eeh_dev *edev, * are now enabled. */ static enum pci_ers_result eeh_report_mmio_enabled(struct eeh_dev *edev, + struct pci_dev *pdev, struct pci_driver *driver) { if (!driver->err_handler->mmio_enabled) return PCI_ERS_RESULT_NONE; eeh_edev_info(edev, "Invoking %s->mmio_enabled()", driver->name); - return driver->err_handler->mmio_enabled(edev->pdev); + return driver->err_handler->mmio_enabled(pdev); } /** @@ -382,20 +384,21 @@ static enum pci_ers_result eeh_report_mmio_enabled(struct eeh_dev *edev, * driver can work again while the device is recovered. */ static enum pci_ers_result eeh_report_reset(struct eeh_dev *edev, + struct pci_dev *pdev, struct pci_driver *driver) { if (!driver->err_handler->slot_reset || !edev->in_error) return PCI_ERS_RESULT_NONE; eeh_edev_info(edev, "Invoking %s->slot_reset()", driver->name); - return driver->err_handler->slot_reset(edev->pdev); + return driver->err_handler->slot_reset(pdev); } -static void *eeh_dev_restore_state(struct eeh_dev *edev, void *userdata) +static void eeh_dev_restore_state(struct eeh_dev *edev, void *userdata) { struct pci_dev *pdev; if (!edev) - return NULL; + return; /* * The content in the config space isn't saved because @@ -407,15 +410,14 @@ static void *eeh_dev_restore_state(struct eeh_dev *edev, void *userdata) if (list_is_last(&edev->entry, &edev->pe->edevs)) eeh_pe_restore_bars(edev->pe); - return NULL; + return; } pdev = eeh_dev_to_pci_dev(edev); if (!pdev) - return NULL; + return; pci_restore_state(pdev); - return NULL; } /** @@ -428,13 +430,14 @@ static void *eeh_dev_restore_state(struct eeh_dev *edev, void *userdata) * to make the recovered device work again. */ static enum pci_ers_result eeh_report_resume(struct eeh_dev *edev, + struct pci_dev *pdev, struct pci_driver *driver) { if (!driver->err_handler->resume || !edev->in_error) return PCI_ERS_RESULT_NONE; eeh_edev_info(edev, "Invoking %s->resume()", driver->name); - driver->err_handler->resume(edev->pdev); + driver->err_handler->resume(pdev); pci_uevent_ers(edev->pdev, PCI_ERS_RESULT_RECOVERED); #ifdef CONFIG_PCI_IOV @@ -453,6 +456,7 @@ static enum pci_ers_result eeh_report_resume(struct eeh_dev *edev, * dead, and that no further recovery attempts will be made on it. */ static enum pci_ers_result eeh_report_failure(struct eeh_dev *edev, + struct pci_dev *pdev, struct pci_driver *driver) { enum pci_ers_result rc; @@ -462,10 +466,10 @@ static enum pci_ers_result eeh_report_failure(struct eeh_dev *edev, eeh_edev_info(edev, "Invoking %s->error_detected(permanent failure)", driver->name); - rc = driver->err_handler->error_detected(edev->pdev, + rc = driver->err_handler->error_detected(pdev, pci_channel_io_perm_failure); - pci_uevent_ers(edev->pdev, PCI_ERS_RESULT_DISCONNECT); + pci_uevent_ers(pdev, PCI_ERS_RESULT_DISCONNECT); return rc; } @@ -473,12 +477,9 @@ static void *eeh_add_virt_device(struct eeh_dev *edev) { struct pci_driver *driver; struct pci_dev *dev = eeh_dev_to_pci_dev(edev); - struct pci_dn *pdn = eeh_dev_to_pdn(edev); if (!(edev->physfn)) { - pr_warn("%s: EEH dev %04x:%02x:%02x.%01x not for VF\n", - __func__, pdn->phb->global_number, pdn->busno, - PCI_SLOT(pdn->devfn), PCI_FUNC(pdn->devfn)); + eeh_edev_warn(edev, "Not for VF\n"); return NULL; } @@ -492,12 +493,12 @@ static void *eeh_add_virt_device(struct eeh_dev *edev) } #ifdef CONFIG_PCI_IOV - pci_iov_add_virtfn(edev->physfn, pdn->vf_index); + pci_iov_add_virtfn(edev->physfn, eeh_dev_to_pdn(edev)->vf_index); #endif return NULL; } -static void *eeh_rmv_device(struct eeh_dev *edev, void *userdata) +static void eeh_rmv_device(struct eeh_dev *edev, void *userdata) { struct pci_driver *driver; struct pci_dev *dev = eeh_dev_to_pci_dev(edev); @@ -512,7 +513,7 @@ static void *eeh_rmv_device(struct eeh_dev *edev, void *userdata) */ if (!eeh_edev_actionable(edev) || (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE)) - return NULL; + return; if (rmv_data) { driver = eeh_pcid_get(dev); @@ -521,7 +522,7 @@ static void *eeh_rmv_device(struct eeh_dev *edev, void *userdata) driver->err_handler->error_detected && driver->err_handler->slot_reset) { eeh_pcid_put(dev); - return NULL; + return; } eeh_pcid_put(dev); } @@ -554,8 +555,6 @@ static void *eeh_rmv_device(struct eeh_dev *edev, void *userdata) pci_stop_and_remove_bus_device(dev); pci_unlock_rescan_remove(); } - - return NULL; } static void *eeh_pe_detach_dev(struct eeh_pe *pe, void *userdata) @@ -744,6 +743,99 @@ static int eeh_reset_device(struct eeh_pe *pe, struct pci_bus *bus, */ #define MAX_WAIT_FOR_RECOVERY 300 + +/* Walks the PE tree after processing an event to remove any stale PEs. + * + * NB: This needs to be recursive to ensure the leaf PEs get removed + * before their parents do. Although this is possible to do recursively + * we don't since this is easier to read and we need to garantee + * the leaf nodes will be handled first. + */ +static void eeh_pe_cleanup(struct eeh_pe *pe) +{ + struct eeh_pe *child_pe, *tmp; + + list_for_each_entry_safe(child_pe, tmp, &pe->child_list, child) + eeh_pe_cleanup(child_pe); + + if (pe->state & EEH_PE_KEEP) + return; + + if (!(pe->state & EEH_PE_INVALID)) + return; + + if (list_empty(&pe->edevs) && list_empty(&pe->child_list)) { + list_del(&pe->child); + kfree(pe); + } +} + +/** + * eeh_check_slot_presence - Check if a device is still present in a slot + * @pdev: pci_dev to check + * + * This function may return a false positive if we can't determine the slot's + * presence state. This might happen for for PCIe slots if the PE containing + * the upstream bridge is also frozen, or the bridge is part of the same PE + * as the device. + * + * This shouldn't happen often, but you might see it if you hotplug a PCIe + * switch. + */ +static bool eeh_slot_presence_check(struct pci_dev *pdev) +{ + const struct hotplug_slot_ops *ops; + struct pci_slot *slot; + u8 state; + int rc; + + if (!pdev) + return false; + + if (pdev->error_state == pci_channel_io_perm_failure) + return false; + + slot = pdev->slot; + if (!slot || !slot->hotplug) + return true; + + ops = slot->hotplug->ops; + if (!ops || !ops->get_adapter_status) + return true; + + /* set the attention indicator while we've got the slot ops */ + if (ops->set_attention_status) + ops->set_attention_status(slot->hotplug, 1); + + rc = ops->get_adapter_status(slot->hotplug, &state); + if (rc) + return true; + + return !!state; +} + +static void eeh_clear_slot_attention(struct pci_dev *pdev) +{ + const struct hotplug_slot_ops *ops; + struct pci_slot *slot; + + if (!pdev) + return; + + if (pdev->error_state == pci_channel_io_perm_failure) + return; + + slot = pdev->slot; + if (!slot || !slot->hotplug) + return; + + ops = slot->hotplug->ops; + if (!ops || !ops->set_attention_status) + return; + + ops->set_attention_status(slot->hotplug, 0); +} + /** * eeh_handle_normal_event - Handle EEH events on a specific PE * @pe: EEH PE - which should not be used after we return, as it may @@ -774,6 +866,7 @@ void eeh_handle_normal_event(struct eeh_pe *pe) enum pci_ers_result result = PCI_ERS_RESULT_NONE; struct eeh_rmv_data rmv_data = {LIST_HEAD_INIT(rmv_data.removed_vf_list), 0}; + int devices = 0; bus = eeh_pe_bus_get(pe); if (!bus) { @@ -782,7 +875,59 @@ void eeh_handle_normal_event(struct eeh_pe *pe) return; } - eeh_pe_state_mark(pe, EEH_PE_RECOVERING); + /* + * When devices are hot-removed we might get an EEH due to + * a driver attempting to touch the MMIO space of a removed + * device. In this case we don't have a device to recover + * so suppress the event if we can't find any present devices. + * + * The hotplug driver should take care of tearing down the + * device itself. + */ + eeh_for_each_pe(pe, tmp_pe) + eeh_pe_for_each_dev(tmp_pe, edev, tmp) + if (eeh_slot_presence_check(edev->pdev)) + devices++; + + if (!devices) { + pr_debug("EEH: Frozen PHB#%x-PE#%x is empty!\n", + pe->phb->global_number, pe->addr); + goto out; /* nothing to recover */ + } + + /* Log the event */ + if (pe->type & EEH_PE_PHB) { + pr_err("EEH: PHB#%x failure detected, location: %s\n", + pe->phb->global_number, eeh_pe_loc_get(pe)); + } else { + struct eeh_pe *phb_pe = eeh_phb_pe_get(pe->phb); + + pr_err("EEH: Frozen PHB#%x-PE#%x detected\n", + pe->phb->global_number, pe->addr); + pr_err("EEH: PE location: %s, PHB location: %s\n", + eeh_pe_loc_get(pe), eeh_pe_loc_get(phb_pe)); + } + +#ifdef CONFIG_STACKTRACE + /* + * Print the saved stack trace now that we've verified there's + * something to recover. + */ + if (pe->trace_entries) { + void **ptrs = (void **) pe->stack_trace; + int i; + + pr_err("EEH: Frozen PHB#%x-PE#%x detected\n", + pe->phb->global_number, pe->addr); + + /* FIXME: Use the same format as dump_stack() */ + pr_err("EEH: Call Trace:\n"); + for (i = 0; i < pe->trace_entries; i++) + pr_err("EEH: [%pK] %pS\n", ptrs[i], ptrs[i]); + + pe->trace_entries = 0; + } +#endif /* CONFIG_STACKTRACE */ eeh_pe_update_time_stamp(pe); pe->freeze_count++; @@ -793,6 +938,10 @@ void eeh_handle_normal_event(struct eeh_pe *pe) result = PCI_ERS_RESULT_DISCONNECT; } + eeh_for_each_pe(pe, tmp_pe) + eeh_pe_for_each_dev(tmp_pe, edev, tmp) + edev->mode &= ~EEH_DEV_NO_HANDLER; + /* Walk the various device drivers attached to this slot through * a reset sequence, giving each an opportunity to do what it needs * to accomplish the reset. Each child gets a report of the @@ -969,6 +1118,19 @@ void eeh_handle_normal_event(struct eeh_pe *pe) return; } } + +out: + /* + * Clean up any PEs without devices. While marked as EEH_PE_RECOVERYING + * we don't want to modify the PE tree structure so we do it here. + */ + eeh_pe_cleanup(pe); + + /* clear the slot attention LED for all recovered devices */ + eeh_for_each_pe(pe, tmp_pe) + eeh_pe_for_each_dev(tmp_pe, edev, tmp) + eeh_clear_slot_attention(edev->pdev); + eeh_pe_state_clear(pe, EEH_PE_RECOVERING, true); } @@ -981,7 +1143,8 @@ void eeh_handle_normal_event(struct eeh_pe *pe) */ void eeh_handle_special_event(void) { - struct eeh_pe *pe, *phb_pe; + struct eeh_pe *pe, *phb_pe, *tmp_pe; + struct eeh_dev *edev, *tmp_edev; struct pci_bus *bus; struct pci_controller *hose; unsigned long flags; @@ -1040,6 +1203,7 @@ void eeh_handle_special_event(void) */ if (rc == EEH_NEXT_ERR_FROZEN_PE || rc == EEH_NEXT_ERR_FENCED_PHB) { + eeh_pe_state_mark(pe, EEH_PE_RECOVERING); eeh_handle_normal_event(pe); } else { pci_lock_rescan_remove(); @@ -1050,6 +1214,10 @@ void eeh_handle_special_event(void) (phb_pe->state & EEH_PE_RECOVERING)) continue; + eeh_for_each_pe(pe, tmp_pe) + eeh_pe_for_each_dev(tmp_pe, edev, tmp_edev) + edev->mode &= ~EEH_DEV_NO_HANDLER; + /* Notify all devices to be down */ eeh_pe_state_clear(pe, EEH_PE_PRI_BUS, true); eeh_set_channel_state(pe, pci_channel_io_perm_failure); |