static pci_ers_result_t broadcast_error_message(struct pci_dev *dev,
enum pci_channel_state state,
char *error_mesg,
int (*cb)(struct pci_dev *, void *))
{
struct aer_broadcast_data result_data;
dev_printk(KERN_DEBUG, &dev->dev, "broadcast %s message\n", error_mesg);
result_data.state = state;
if (cb == report_error_detected)
result_data.result = PCI_ERS_RESULT_CAN_RECOVER;
else
result_data.result = PCI_ERS_RESULT_RECOVERED;
if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE) {
if (cb == report_error_detected)
dev->error_state = state;
pci_walk_bus(dev->subordinate, cb, &result_data);
if (cb == report_resume) {
pci_cleanup_aer_uncorrect_error_status(dev);
dev->error_state = pci_channel_io_normal;
}
} else {
pci_walk_bus(dev->bus, cb, &result_data);
}
return result_data.result;
}
static void via_no_dac(struct pci_dev *dev)
{
if (!disable_dac_quirk) {
dev_info(&dev->dev, "disabling DAC on VIA PCI bridge\n");
pci_walk_bus(dev->subordinate, via_no_dac_cb, NULL);
}
}
static int its_pci_msi_prepare(struct irq_domain *domain, struct device *dev,
int nvec, msi_alloc_info_t *info)
{
struct pci_dev *pdev, *alias_dev;
struct msi_domain_info *msi_info;
int alias_count = 0;
if (!dev_is_pci(dev))
return -EINVAL;
msi_info = msi_get_domain_info(domain->parent);
pdev = to_pci_dev(dev);
/*
* If pdev is downstream of any aliasing bridges, take an upper
* bound of how many other vectors could map to the same DevID.
*/
pci_for_each_dma_alias(pdev, its_get_pci_alias, &alias_dev);
if (alias_dev != pdev && alias_dev->subordinate)
pci_walk_bus(alias_dev->subordinate, its_pci_msi_vec_count,
&alias_count);
/* ITS specific DeviceID, as the core ITS ignores dev. */
info->scratchpad[0].ul = pci_msi_domain_get_msi_rid(domain, pdev);
return msi_info->ops->msi_prepare(domain->parent,
dev, max(nvec, alias_count), info);
}
/**
* find_source_device - search through device hierarchy for source device
* @parent: pointer to Root Port pci_dev data structure
* @e_info: including detailed error information such like id
*
* Return true if found.
*
* Invoked by DPC when error is detected at the Root Port.
* Caller of this function must set id, severity, and multi_error_valid of
* struct aer_err_info pointed by @e_info properly. This function must fill
* e_info->error_dev_num and e_info->dev[], based on the given information.
*/
static bool find_source_device(struct pci_dev *parent,
struct aer_err_info *e_info)
{
struct pci_dev *dev = parent;
int result;
/* Must reset in this function */
e_info->error_dev_num = 0;
/* Is Root Port an agent that sends error message? */
result = find_device_iter(dev, e_info);
if (result)
return true;
pci_walk_bus(parent->subordinate, find_device_iter, e_info);
if (!e_info->error_dev_num) {
#ifdef CONFIG_DEBUG_PRINTK
dev_printk(KERN_DEBUG, &parent->dev,
"can't find device of ID%04x\n",
e_info->id);
#else
dev_;
#endif
return false;
}
return true;
}
static void set_downstream_devices_error_reporting(struct pci_dev *dev,
bool enable)
{
set_device_error_reporting(dev, &enable);
if (!dev->subordinate)
return;
pci_walk_bus(dev->subordinate, set_device_error_reporting, &enable);
}
/**
* broadcast_error_message - handle message broadcast to downstream drivers
* @dev: pointer to from where in a hierarchy message is broadcasted down
* @state: error state
* @error_mesg: message to print
* @cb: callback to be broadcasted
*
* Invoked during error recovery process. Once being invoked, the content
* of error severity will be broadcasted to all downstream drivers in a
* hierarchy in question.
*/
static pci_ers_result_t broadcast_error_message(struct pci_dev *dev,
enum pci_channel_state state,
char *error_mesg,
int (*cb)(struct pci_dev *, void *))
{
struct aer_broadcast_data result_data;
#ifdef CONFIG_DEBUG_PRINTK
dev_printk(KERN_DEBUG, &dev->dev, "broadcast %s message\n", error_mesg);
#else
dev_;
#endif
result_data.state = state;
if (cb == report_error_detected)
result_data.result = PCI_ERS_RESULT_CAN_RECOVER;
else
result_data.result = PCI_ERS_RESULT_RECOVERED;
if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE) {
/*
* If the error is reported by a bridge, we think this error
* is related to the downstream link of the bridge, so we
* do error recovery on all subordinates of the bridge instead
* of the bridge and clear the error status of the bridge.
*/
if (cb == report_error_detected)
dev->error_state = state;
pci_walk_bus(dev->subordinate, cb, &result_data);
if (cb == report_resume) {
pci_cleanup_aer_uncorrect_error_status(dev);
dev->error_state = pci_channel_io_normal;
}
} else {
/*
* If the error is reported by an end point, we think this
* error is related to the upstream link of the end point.
*/
pci_walk_bus(dev->bus, cb, &result_data);
}
return result_data.result;
}
/**
* find_source_device - search through device hierarchy for source device
* @parent: pointer to Root Port pci_dev data structure
* @err_info: including detailed error information such like id
*
* Invoked when error is detected at the Root Port.
*/
static void find_source_device(struct pci_dev *parent,
struct aer_err_info *e_info)
{
struct pci_dev *dev = parent;
int result;
/* Is Root Port an agent that sends error message? */
result = find_device_iter(dev, e_info);
if (result)
return;
pci_walk_bus(parent->subordinate, find_device_iter, e_info);
}
static void find_source_device(struct pci_dev *parent,
struct aer_err_info *e_info)
{
struct pci_dev *dev = parent;
int result;
result = find_device_iter(dev, e_info);
if (result)
return;
pci_walk_bus(parent->subordinate, find_device_iter, e_info);
}
struct pci_dn * handle_eeh_events (struct eeh_event *event)
{
struct device_node *frozen_dn;
struct pci_dn *frozen_pdn;
struct pci_bus *frozen_bus;
int rc = 0;
enum pci_ers_result result = PCI_ERS_RESULT_NONE;
const char *location, *pci_str, *drv_str;
frozen_dn = find_device_pe(event->dn);
if (!frozen_dn) {
location = of_get_property(event->dn, "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, pci_name(event->dev));
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_pdn = PCI_DN(frozen_dn);
frozen_pdn->eeh_freeze_count++;
if (frozen_pdn->pcidev) {
pci_str = pci_name (frozen_pdn->pcidev);
drv_str = pcid_name (frozen_pdn->pcidev);
} else {
pci_str = pci_name (event->dev);
drv_str = pcid_name (event->dev);
}
if (frozen_pdn->eeh_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_pdn->eeh_freeze_count);
printk(KERN_WARNING
"EEH: 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_wait_for_slot_status (frozen_pdn, MAX_WAIT_FOR_RECOVERY*1000);
if (rc < 0) {
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_pdn, EEH_LOG_TEMP_FAILURE);
/* 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_pdn, 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 = rtas_pci_enable(frozen_pdn, EEH_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 = rtas_pci_enable(frozen_pdn, EEH_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_pdn, 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_pdn;
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_pdn->eeh_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_pdn, EEH_LOG_PERM_FAILURE);
/* 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;
}
struct pci_dn * handle_eeh_events (struct eeh_event *event)
{
struct device_node *frozen_dn;
struct pci_dn *frozen_pdn;
struct pci_bus *frozen_bus;
int rc = 0;
enum pci_ers_result result = PCI_ERS_RESULT_NONE;
const char *location, *pci_str, *drv_str;
frozen_dn = find_device_pe(event->dn);
frozen_bus = pcibios_find_pci_bus(frozen_dn);
if (!frozen_dn) {
location = (char *) get_property(event->dn, "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, pci_name(event->dev));
return NULL;
}
location = (char *) 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;
}
#if 0
/* We may get "permanent failure" messages on empty slots.
* These are false alarms. Empty slots have no child dn. */
if ((event->state == pci_channel_io_perm_failure) && (frozen_device == NULL))
return;
#endif
frozen_pdn = PCI_DN(frozen_dn);
frozen_pdn->eeh_freeze_count++;
if (frozen_pdn->pcidev) {
pci_str = pci_name (frozen_pdn->pcidev);
drv_str = pcid_name (frozen_pdn->pcidev);
} else {
pci_str = pci_name (event->dev);
drv_str = pcid_name (event->dev);
}
if (frozen_pdn->eeh_freeze_count > EEH_MAX_ALLOWED_FREEZES)
goto excess_failures;
/* If the reset state is a '5' and the time to reset is 0 (infinity)
* or is more then 15 seconds, then mark this as a permanent failure.
*/
if ((event->state == pci_channel_io_perm_failure) &&
((event->time_unavail <= 0) ||
(event->time_unavail > MAX_WAIT_FOR_RECOVERY*1000)))
goto hard_fail;
eeh_slot_error_detail(frozen_pdn, 1 /* Temporary Error */);
printk(KERN_WARNING
"EEH: This PCI device has failed %d times since last reboot: "
"location=%s driver=%s pci addr=%s\n",
frozen_pdn->eeh_freeze_count, 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);
/* 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_pdn, frozen_bus);
if (rc)
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_pdn, NULL);
if (rc)
goto hard_fail;
pci_walk_bus(frozen_bus, eeh_report_reset, NULL);
}
/* If all devices reported they can proceed, the re-enable PIO */
if (result == PCI_ERS_RESULT_CAN_RECOVER) {
/* XXX Not supported; we brute-force reset the device */
rc = eeh_reset_device(frozen_pdn, NULL);
if (rc)
goto hard_fail;
pci_walk_bus(frozen_bus, eeh_report_reset, NULL);
}
/* Tell all device drivers that they can resume operations */
pci_walk_bus(frozen_bus, eeh_report_resume, NULL);
return frozen_pdn;
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 and has been permanently disabled. \n"
"Please try reseating this device or replacing it.\n",
location, drv_str, pci_str, frozen_pdn->eeh_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_pdn, 2 /* Permanent Error */);
/* 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;
}
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;
}
