/**
* eeh_slot_error_detail - Generate combined log including driver log and error log
* @pe: EEH PE
* @severity: temporary or permanent error log
*
* This routine should be called to generate the combined log, which
* is comprised of driver log and error log. The driver log is figured
* out from the config space of the corresponding PCI device, while
* the error log is fetched through platform dependent function call.
*/
void eeh_slot_error_detail(struct eeh_pe *pe, int severity)
{
size_t loglen = 0;
struct eeh_dev *edev, *tmp;
/*
* When the PHB is fenced or dead, it's pointless to collect
* the data from PCI config space because it should return
* 0xFF's. For ER, we still retrieve the data from the PCI
* config space.
*
* For pHyp, we have to enable IO for log retrieval. Otherwise,
* 0xFF's is always returned from PCI config space.
*/
if (!(pe->type & EEH_PE_PHB)) {
if (eeh_has_flag(EEH_ENABLE_IO_FOR_LOG))
eeh_pci_enable(pe, EEH_OPT_THAW_MMIO);
eeh_ops->configure_bridge(pe);
eeh_pe_restore_bars(pe);
pci_regs_buf[0] = 0;
eeh_pe_for_each_dev(pe, edev, tmp) {
loglen += eeh_gather_pci_data(edev, pci_regs_buf + loglen,
EEH_PCI_REGS_LOG_LEN - loglen);
}
}
/*
* Explicitly clear PE's frozen state for PowerNV where
* we have frozen PE until BAR restore is completed. It's
* harmless to clear it for pSeries. To be consistent with
* PE reset (for 3 times), we try to clear the frozen state
* for 3 times as well.
*/
static void *__eeh_clear_pe_frozen_state(void *data, void *flag)
{
struct eeh_pe *pe = (struct eeh_pe *)data;
int i, rc;
for (i = 0; i < 3; i++) {
rc = eeh_pci_enable(pe, EEH_OPT_THAW_MMIO);
if (rc)
continue;
rc = eeh_pci_enable(pe, EEH_OPT_THAW_DMA);
if (!rc)
break;
}
/* The PE has been isolated, clear it */
if (rc) {
pr_warn("%s: Can't clear frozen PHB#%x-PE#%x (%d)\n",
__func__, pe->phb->global_number, pe->addr, rc);
return (void *)pe;
}
return NULL;
}
/**
* eeh_slot_error_detail - Generate combined log including driver log and error log
* @pe: EEH PE
* @severity: temporary or permanent error log
*
* This routine should be called to generate the combined log, which
* is comprised of driver log and error log. The driver log is figured
* out from the config space of the corresponding PCI device, while
* the error log is fetched through platform dependent function call.
*/
void eeh_slot_error_detail(struct eeh_pe *pe, int severity)
{
size_t loglen = 0;
/*
* When the PHB is fenced or dead, it's pointless to collect
* the data from PCI config space because it should return
* 0xFF's. For ER, we still retrieve the data from the PCI
* config space.
*
* For pHyp, we have to enable IO for log retrieval. Otherwise,
* 0xFF's is always returned from PCI config space.
*/
if (!(pe->type & EEH_PE_PHB)) {
if (eeh_has_flag(EEH_ENABLE_IO_FOR_LOG))
eeh_pci_enable(pe, EEH_OPT_THAW_MMIO);
/*
* The config space of some PCI devices can't be accessed
* when their PEs are in frozen state. Otherwise, fenced
* PHB might be seen. Those PEs are identified with flag
* EEH_PE_CFG_RESTRICTED, indicating EEH_PE_CFG_BLOCKED
* is set automatically when the PE is put to EEH_PE_ISOLATED.
*
* Restoring BARs possibly triggers PCI config access in
* (OPAL) firmware and then causes fenced PHB. If the
* PCI config is blocked with flag EEH_PE_CFG_BLOCKED, it's
* pointless to restore BARs and dump config space.
*/
eeh_ops->configure_bridge(pe);
if (!(pe->state & EEH_PE_CFG_BLOCKED)) {
eeh_pe_restore_bars(pe);
pci_regs_buf[0] = 0;
eeh_pe_traverse(pe, eeh_dump_pe_log, &loglen);
}
}
eeh_ops->get_log(pe, severity, pci_regs_buf, loglen);
}
struct eeh_dev *handle_eeh_events(struct eeh_event *event)
{
struct device_node *frozen_dn;
struct eeh_dev *frozen_edev;
struct pci_bus *frozen_bus;
int rc = 0;
enum pci_ers_result result = PCI_ERS_RESULT_NONE;
const char *location, *pci_str, *drv_str, *bus_pci_str, *bus_drv_str;
frozen_dn = eeh_find_device_pe(eeh_dev_to_of_node(event->edev));
if (!frozen_dn) {
location = of_get_property(eeh_dev_to_of_node(event->edev), "ibm,loc-code", NULL);
location = location ? location : "unknown";
printk(KERN_ERR "EEH: Error: Cannot find partition endpoint "
"for location=%s pci addr=%s\n",
location, eeh_pci_name(eeh_dev_to_pci_dev(event->edev)));
return NULL;
}
frozen_bus = pcibios_find_pci_bus(frozen_dn);
location = of_get_property(frozen_dn, "ibm,loc-code", NULL);
location = location ? location : "unknown";
if (!frozen_bus)
frozen_bus = pcibios_find_pci_bus(frozen_dn->parent);
if (!frozen_bus) {
printk(KERN_ERR "EEH: Cannot find PCI bus "
"for location=%s dn=%s\n",
location, frozen_dn->full_name);
return NULL;
}
frozen_edev = of_node_to_eeh_dev(frozen_dn);
frozen_edev->freeze_count++;
pci_str = eeh_pci_name(eeh_dev_to_pci_dev(event->edev));
drv_str = eeh_pcid_name(eeh_dev_to_pci_dev(event->edev));
if (frozen_edev->freeze_count > EEH_MAX_ALLOWED_FREEZES)
goto excess_failures;
printk(KERN_WARNING
"EEH: This PCI device has failed %d times in the last hour:\n",
frozen_edev->freeze_count);
if (frozen_edev->pdev) {
bus_pci_str = pci_name(frozen_edev->pdev);
bus_drv_str = eeh_pcid_name(frozen_edev->pdev);
printk(KERN_WARNING
"EEH: Bus location=%s driver=%s pci addr=%s\n",
location, bus_drv_str, bus_pci_str);
}
printk(KERN_WARNING
"EEH: Device location=%s driver=%s pci addr=%s\n",
location, drv_str, pci_str);
pci_walk_bus(frozen_bus, eeh_report_error, &result);
rc = eeh_ops->wait_state(eeh_dev_to_of_node(frozen_edev), MAX_WAIT_FOR_RECOVERY*1000);
if (rc < 0 || rc == EEH_STATE_NOT_SUPPORT) {
printk(KERN_WARNING "EEH: Permanent failure\n");
goto hard_fail;
}
eeh_slot_error_detail(frozen_edev, EEH_LOG_TEMP);
if (result == PCI_ERS_RESULT_NONE) {
rc = eeh_reset_device(frozen_edev, frozen_bus);
if (rc) {
printk(KERN_WARNING "EEH: Unable to reset, rc=%d\n", rc);
goto hard_fail;
}
}
if (result == PCI_ERS_RESULT_CAN_RECOVER) {
rc = eeh_pci_enable(frozen_edev, EEH_OPT_THAW_MMIO);
if (rc < 0)
goto hard_fail;
if (rc) {
result = PCI_ERS_RESULT_NEED_RESET;
} else {
result = PCI_ERS_RESULT_NONE;
pci_walk_bus(frozen_bus, eeh_report_mmio_enabled, &result);
}
}
if (result == PCI_ERS_RESULT_CAN_RECOVER) {
rc = eeh_pci_enable(frozen_edev, EEH_OPT_THAW_DMA);
if (rc < 0)
goto hard_fail;
if (rc)
result = PCI_ERS_RESULT_NEED_RESET;
else
result = PCI_ERS_RESULT_RECOVERED;
}
if (result == PCI_ERS_RESULT_DISCONNECT) {
printk(KERN_WARNING "EEH: Device driver gave up\n");
goto hard_fail;
}
if (result == PCI_ERS_RESULT_NEED_RESET) {
rc = eeh_reset_device(frozen_edev, NULL);
if (rc) {
printk(KERN_WARNING "EEH: Cannot reset, rc=%d\n", rc);
goto hard_fail;
}
result = PCI_ERS_RESULT_NONE;
pci_walk_bus(frozen_bus, eeh_report_reset, &result);
}
if ((result != PCI_ERS_RESULT_RECOVERED) &&
(result != PCI_ERS_RESULT_NONE)) {
printk(KERN_WARNING "EEH: Not recovered\n");
goto hard_fail;
}
pci_walk_bus(frozen_bus, eeh_report_resume, NULL);
return frozen_edev;
excess_failures:
printk(KERN_ERR
"EEH: PCI device at location=%s driver=%s pci addr=%s\n"
"has failed %d times in the last hour "
"and has been permanently disabled.\n"
"Please try reseating this device or replacing it.\n",
location, drv_str, pci_str, frozen_edev->freeze_count);
goto perm_error;
hard_fail:
printk(KERN_ERR
"EEH: Unable to recover from failure of PCI device "
"at location=%s driver=%s pci addr=%s\n"
"Please try reseating this device or replacing it.\n",
location, drv_str, pci_str);
perm_error:
eeh_slot_error_detail(frozen_edev, EEH_LOG_PERM);
pci_walk_bus(frozen_bus, eeh_report_failure, NULL);
pcibios_remove_pci_devices(frozen_bus);
return NULL;
}
/**
* eeh_handle_event - Reset a PCI device after hard lockup.
* @event: EEH event
*
* While PHB detects address or data parity errors on particular PCI
* slot, the associated PE will be frozen. Besides, DMA's occurring
* to wild addresses (which usually happen due to bugs in device
* drivers or in PCI adapter firmware) can cause EEH error. #SERR,
* #PERR or other misc PCI-related errors also can trigger EEH errors.
*
* Recovery process consists of unplugging the device driver (which
* generated hotplug events to userspace), then issuing a PCI #RST to
* the device, then reconfiguring the PCI config space for all bridges
* & devices under this slot, and then finally restarting the device
* drivers (which cause a second set of hotplug events to go out to
* userspace).
*/
struct eeh_dev *handle_eeh_events(struct eeh_event *event)
{
struct device_node *frozen_dn;
struct eeh_dev *frozen_edev;
struct pci_bus *frozen_bus;
int rc = 0;
enum pci_ers_result result = PCI_ERS_RESULT_NONE;
const char *location, *pci_str, *drv_str, *bus_pci_str, *bus_drv_str;
frozen_dn = eeh_find_device_pe(eeh_dev_to_of_node(event->edev));
if (!frozen_dn) {
location = of_get_property(eeh_dev_to_of_node(event->edev), "ibm,loc-code", NULL);
location = location ? location : "unknown";
printk(KERN_ERR "EEH: Error: Cannot find partition endpoint "
"for location=%s pci addr=%s\n",
location, eeh_pci_name(eeh_dev_to_pci_dev(event->edev)));
return NULL;
}
frozen_bus = pcibios_find_pci_bus(frozen_dn);
location = of_get_property(frozen_dn, "ibm,loc-code", NULL);
location = location ? location : "unknown";
/* There are two different styles for coming up with the PE.
* In the old style, it was the highest EEH-capable device
* which was always an EADS pci bridge. In the new style,
* there might not be any EADS bridges, and even when there are,
* the firmware marks them as "EEH incapable". So another
* two-step is needed to find the pci bus..
*/
if (!frozen_bus)
frozen_bus = pcibios_find_pci_bus(frozen_dn->parent);
if (!frozen_bus) {
printk(KERN_ERR "EEH: Cannot find PCI bus "
"for location=%s dn=%s\n",
location, frozen_dn->full_name);
return NULL;
}
frozen_edev = of_node_to_eeh_dev(frozen_dn);
frozen_edev->freeze_count++;
pci_str = eeh_pci_name(eeh_dev_to_pci_dev(event->edev));
drv_str = eeh_pcid_name(eeh_dev_to_pci_dev(event->edev));
if (frozen_edev->freeze_count > EEH_MAX_ALLOWED_FREEZES)
goto excess_failures;
printk(KERN_WARNING
"EEH: This PCI device has failed %d times in the last hour:\n",
frozen_edev->freeze_count);
if (frozen_edev->pdev) {
bus_pci_str = pci_name(frozen_edev->pdev);
bus_drv_str = eeh_pcid_name(frozen_edev->pdev);
printk(KERN_WARNING
"EEH: Bus location=%s driver=%s pci addr=%s\n",
location, bus_drv_str, bus_pci_str);
}
printk(KERN_WARNING
"EEH: Device location=%s driver=%s pci addr=%s\n",
location, drv_str, pci_str);
/* 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
* status ... if any child can't handle the reset, then the entire
* slot is dlpar removed and added.
*/
pci_walk_bus(frozen_bus, eeh_report_error, &result);
/* Get the current PCI slot state. This can take a long time,
* sometimes over 3 seconds for certain systems.
*/
rc = eeh_ops->wait_state(eeh_dev_to_of_node(frozen_edev), MAX_WAIT_FOR_RECOVERY*1000);
if (rc < 0 || rc == EEH_STATE_NOT_SUPPORT) {
printk(KERN_WARNING "EEH: Permanent failure\n");
goto hard_fail;
}
/* Since rtas may enable MMIO when posting the error log,
* don't post the error log until after all dev drivers
* have been informed.
*/
eeh_slot_error_detail(frozen_edev, EEH_LOG_TEMP);
/* If all device drivers were EEH-unaware, then shut
* down all of the device drivers, and hope they
* go down willingly, without panicing the system.
*/
if (result == PCI_ERS_RESULT_NONE) {
rc = eeh_reset_device(frozen_edev, frozen_bus);
if (rc) {
printk(KERN_WARNING "EEH: Unable to reset, rc=%d\n", rc);
goto hard_fail;
}
}
/* If all devices reported they can proceed, then re-enable MMIO */
if (result == PCI_ERS_RESULT_CAN_RECOVER) {
rc = eeh_pci_enable(frozen_edev, EEH_OPT_THAW_MMIO);
if (rc < 0)
goto hard_fail;
if (rc) {
result = PCI_ERS_RESULT_NEED_RESET;
} else {
result = PCI_ERS_RESULT_NONE;
pci_walk_bus(frozen_bus, eeh_report_mmio_enabled, &result);
}
}
/* If all devices reported they can proceed, then re-enable DMA */
if (result == PCI_ERS_RESULT_CAN_RECOVER) {
rc = eeh_pci_enable(frozen_edev, EEH_OPT_THAW_DMA);
if (rc < 0)
goto hard_fail;
if (rc)
result = PCI_ERS_RESULT_NEED_RESET;
else
result = PCI_ERS_RESULT_RECOVERED;
}
/* If any device has a hard failure, then shut off everything. */
if (result == PCI_ERS_RESULT_DISCONNECT) {
printk(KERN_WARNING "EEH: Device driver gave up\n");
goto hard_fail;
}
/* If any device called out for a reset, then reset the slot */
if (result == PCI_ERS_RESULT_NEED_RESET) {
rc = eeh_reset_device(frozen_edev, NULL);
if (rc) {
printk(KERN_WARNING "EEH: Cannot reset, rc=%d\n", rc);
goto hard_fail;
}
result = PCI_ERS_RESULT_NONE;
pci_walk_bus(frozen_bus, eeh_report_reset, &result);
}
/* All devices should claim they have recovered by now. */
if ((result != PCI_ERS_RESULT_RECOVERED) &&
(result != PCI_ERS_RESULT_NONE)) {
printk(KERN_WARNING "EEH: Not recovered\n");
goto hard_fail;
}
/* Tell all device drivers that they can resume operations */
pci_walk_bus(frozen_bus, eeh_report_resume, NULL);
return frozen_edev;
excess_failures:
/*
* About 90% of all real-life EEH failures in the field
* are due to poorly seated PCI cards. Only 10% or so are
* due to actual, failed cards.
*/
printk(KERN_ERR
"EEH: PCI device at location=%s driver=%s pci addr=%s\n"
"has failed %d times in the last hour "
"and has been permanently disabled.\n"
"Please try reseating this device or replacing it.\n",
location, drv_str, pci_str, frozen_edev->freeze_count);
goto perm_error;
hard_fail:
printk(KERN_ERR
"EEH: Unable to recover from failure of PCI device "
"at location=%s driver=%s pci addr=%s\n"
"Please try reseating this device or replacing it.\n",
location, drv_str, pci_str);
perm_error:
eeh_slot_error_detail(frozen_edev, EEH_LOG_PERM);
/* Notify all devices that they're about to go down. */
pci_walk_bus(frozen_bus, eeh_report_failure, NULL);
/* Shut down the device drivers for good. */
pcibios_remove_pci_devices(frozen_bus);
return NULL;
}
