/**
* eeh_reset_device - Perform actual reset of a pci slot
* @edev: PE associated EEH device
* @bus: PCI bus corresponding to the isolcated slot
*
* This routine must be called to do reset on the indicated PE.
* During the reset, udev might be invoked because those affected
* PCI devices will be removed and then added.
*/
static int eeh_reset_device(struct eeh_dev *edev, struct pci_bus *bus)
{
struct device_node *dn;
int cnt, rc;
/* pcibios will clear the counter; save the value */
cnt = edev->freeze_count;
if (bus)
pcibios_remove_pci_devices(bus);
/* Reset the pci controller. (Asserts RST#; resets config space).
* Reconfigure bridges and devices. Don't try to bring the system
* up if the reset failed for some reason.
*/
rc = eeh_reset_pe(edev);
if (rc)
return rc;
/* Walk over all functions on this device. */
dn = eeh_dev_to_of_node(edev);
if (!pcibios_find_pci_bus(dn) && of_node_to_eeh_dev(dn->parent))
dn = dn->parent->child;
while (dn) {
struct eeh_dev *pedev = of_node_to_eeh_dev(dn);
/* On Power4, always true because eeh_pe_config_addr=0 */
if (edev->pe_config_addr == pedev->pe_config_addr) {
eeh_ops->configure_bridge(dn);
eeh_restore_bars(pedev);
}
dn = dn->sibling;
}
/* Give the system 5 seconds to finish running the user-space
* hotplug shutdown scripts, e.g. ifdown for ethernet. Yes,
* this is a hack, but if we don't do this, and try to bring
* the device up before the scripts have taken it down,
* potentially weird things happen.
*/
if (bus) {
ssleep(5);
pcibios_add_pci_devices(bus);
}
edev->freeze_count = cnt;
return 0;
}
/* When a PCI device is isolated from the bus, a subsequent MMIO read is
* required for the kernel EEH mechanisms to notice. As the Solarflare driver
* was written to minimise MMIO read (for latency) then a periodic call to check
* the EEH status of the device is required so that device recovery can happen
* in a timely fashion.
*/
static void siena_monitor(struct efx_nic *efx)
{
struct eeh_dev *eehdev =
of_node_to_eeh_dev(pci_device_to_OF_node(efx->pci_dev));
eeh_dev_check_failure(eehdev);
}
int rtas_read_config(struct pci_dn *pdn, int where, int size, u32 *val)
{
int returnval = -1;
unsigned long buid, addr;
int ret;
if (!pdn)
return PCIBIOS_DEVICE_NOT_FOUND;
if (!config_access_valid(pdn, where))
return PCIBIOS_BAD_REGISTER_NUMBER;
addr = rtas_config_addr(pdn->busno, pdn->devfn, where);
buid = pdn->phb->buid;
if (buid) {
ret = rtas_call(ibm_read_pci_config, 4, 2, &returnval,
addr, BUID_HI(buid), BUID_LO(buid), size);
} else {
ret = rtas_call(read_pci_config, 2, 2, &returnval, addr, size);
}
*val = returnval;
if (ret)
return PCIBIOS_DEVICE_NOT_FOUND;
if (returnval == EEH_IO_ERROR_VALUE(size) &&
eeh_dev_check_failure(of_node_to_eeh_dev(pdn->node)))
return PCIBIOS_DEVICE_NOT_FOUND;
return PCIBIOS_SUCCESSFUL;
}
/**
* pseries_eeh_configure_bridge - Configure PCI bridges in the indicated PE
* @dn: PE associated device node
*
* The function will be called to reconfigure the bridges included
* in the specified PE so that the mulfunctional PE would be recovered
* again.
*/
static int pseries_eeh_configure_bridge(struct device_node *dn)
{
struct eeh_dev *edev;
int config_addr;
int ret;
/* Figure out the PE address */
edev = of_node_to_eeh_dev(dn);
config_addr = edev->config_addr;
if (edev->pe_config_addr)
config_addr = edev->pe_config_addr;
/* Use new configure-pe function, if supported */
if (ibm_configure_pe != RTAS_UNKNOWN_SERVICE) {
ret = rtas_call(ibm_configure_pe, 3, 1, NULL,
config_addr, BUID_HI(edev->phb->buid),
BUID_LO(edev->phb->buid));
} else if (ibm_configure_bridge != RTAS_UNKNOWN_SERVICE) {
ret = rtas_call(ibm_configure_bridge, 3, 1, NULL,
config_addr, BUID_HI(edev->phb->buid),
BUID_LO(edev->phb->buid));
} else {
return -EFAULT;
}
if (ret)
pr_warning("%s: Unable to configure bridge %d for %s\n",
__func__, ret, dn->full_name);
return ret;
}
/**
* pseries_eeh_get_log - Retrieve error log
* @dn: device node
* @severity: temporary or permanent error log
* @drv_log: driver log to be combined with retrieved error log
* @len: length of driver log
*
* Retrieve the temporary or permanent error from the PE.
* Actually, the error will be retrieved through the dedicated
* RTAS call.
*/
static int pseries_eeh_get_log(struct device_node *dn, int severity, char *drv_log, unsigned long len)
{
struct eeh_dev *edev;
int config_addr;
unsigned long flags;
int ret;
edev = of_node_to_eeh_dev(dn);
spin_lock_irqsave(&slot_errbuf_lock, flags);
memset(slot_errbuf, 0, eeh_error_buf_size);
/* Figure out the PE address */
config_addr = edev->config_addr;
if (edev->pe_config_addr)
config_addr = edev->pe_config_addr;
ret = rtas_call(ibm_slot_error_detail, 8, 1, NULL, config_addr,
BUID_HI(edev->phb->buid), BUID_LO(edev->phb->buid),
virt_to_phys(drv_log), len,
virt_to_phys(slot_errbuf), eeh_error_buf_size,
severity);
if (!ret)
log_error(slot_errbuf, ERR_TYPE_RTAS_LOG, 0);
spin_unlock_irqrestore(&slot_errbuf_lock, flags);
return ret;
}
/**
* pseries_eeh_reset - Reset the specified PE
* @dn: PE associated device node
* @option: reset option
*
* Reset the specified PE
*/
static int pseries_eeh_reset(struct device_node *dn, int option)
{
struct eeh_dev *edev;
int config_addr;
int ret;
/* Figure out PE address */
edev = of_node_to_eeh_dev(dn);
config_addr = edev->config_addr;
if (edev->pe_config_addr)
config_addr = edev->pe_config_addr;
/* Reset PE through RTAS call */
ret = rtas_call(ibm_set_slot_reset, 4, 1, NULL,
config_addr, BUID_HI(edev->phb->buid),
BUID_LO(edev->phb->buid), option);
/* If fundamental-reset not supported, try hot-reset */
if (option == EEH_RESET_FUNDAMENTAL &&
ret == -8) {
ret = rtas_call(ibm_set_slot_reset, 4, 1, NULL,
config_addr, BUID_HI(edev->phb->buid),
BUID_LO(edev->phb->buid), EEH_RESET_HOT);
}
return ret;
}
static int eeh_reset_device(struct eeh_dev *edev, struct pci_bus *bus)
{
struct device_node *dn;
int cnt, rc;
cnt = edev->freeze_count;
if (bus)
pcibios_remove_pci_devices(bus);
rc = eeh_reset_pe(edev);
if (rc)
return rc;
dn = eeh_dev_to_of_node(edev);
if (!pcibios_find_pci_bus(dn) && of_node_to_eeh_dev(dn->parent))
dn = dn->parent->child;
while (dn) {
struct eeh_dev *pedev = of_node_to_eeh_dev(dn);
if (edev->pe_config_addr == pedev->pe_config_addr) {
eeh_ops->configure_bridge(dn);
eeh_restore_bars(pedev);
}
dn = dn->sibling;
}
if (bus) {
ssleep(5);
pcibios_add_pci_devices(bus);
}
edev->freeze_count = cnt;
return 0;
}
/**
* pseries_eeh_get_pe_addr - Retrieve PE address
* @dn: device node
*
* Retrieve the assocated PE address. Actually, there're 2 RTAS
* function calls dedicated for the purpose. We need implement
* it through the new function and then the old one. Besides,
* you should make sure the config address is figured out from
* FDT node before calling the function.
*
* It's notable that zero'ed return value means invalid PE config
* address.
*/
static int pseries_eeh_get_pe_addr(struct device_node *dn)
{
struct eeh_dev *edev;
int ret = 0;
int rets[3];
edev = of_node_to_eeh_dev(dn);
if (ibm_get_config_addr_info2 != RTAS_UNKNOWN_SERVICE) {
/*
* First of all, we need to make sure there has one PE
* associated with the device. Otherwise, PE address is
* meaningless.
*/
ret = rtas_call(ibm_get_config_addr_info2, 4, 2, rets,
edev->config_addr, BUID_HI(edev->phb->buid),
BUID_LO(edev->phb->buid), 1);
if (ret || (rets[0] == 0))
return 0;
/* Retrieve the associated PE config address */
ret = rtas_call(ibm_get_config_addr_info2, 4, 2, rets,
edev->config_addr, BUID_HI(edev->phb->buid),
BUID_LO(edev->phb->buid), 0);
if (ret) {
pr_warning("%s: Failed to get PE address for %s\n",
__func__, dn->full_name);
return 0;
}
return rets[0];
}
if (ibm_get_config_addr_info != RTAS_UNKNOWN_SERVICE) {
ret = rtas_call(ibm_get_config_addr_info, 4, 2, rets,
edev->config_addr, BUID_HI(edev->phb->buid),
BUID_LO(edev->phb->buid), 0);
if (ret) {
pr_warning("%s: Failed to get PE address for %s\n",
__func__, dn->full_name);
return 0;
}
return rets[0];
}
return ret;
}
/**
* pseries_eeh_set_option - Initialize EEH or MMIO/DMA reenable
* @dn: device node
* @option: operation to be issued
*
* The function is used to control the EEH functionality globally.
* Currently, following options are support according to PAPR:
* Enable EEH, Disable EEH, Enable MMIO and Enable DMA
*/
static int pseries_eeh_set_option(struct device_node *dn, int option)
{
int ret = 0;
struct eeh_dev *edev;
const u32 *reg;
int config_addr;
edev = of_node_to_eeh_dev(dn);
/*
* When we're enabling or disabling EEH functioality on
* the particular PE, the PE config address is possibly
* unavailable. Therefore, we have to figure it out from
* the FDT node.
*/
switch (option) {
case EEH_OPT_DISABLE:
case EEH_OPT_ENABLE:
reg = of_get_property(dn, "reg", NULL);
config_addr = reg[0];
break;
case EEH_OPT_THAW_MMIO:
case EEH_OPT_THAW_DMA:
config_addr = edev->config_addr;
if (edev->pe_config_addr)
config_addr = edev->pe_config_addr;
break;
default:
pr_err("%s: Invalid option %d\n",
__func__, option);
return -EINVAL;
}
ret = rtas_call(ibm_set_eeh_option, 4, 1, NULL,
config_addr, BUID_HI(edev->phb->buid),
BUID_LO(edev->phb->buid), option);
return ret;
}
/**
* pseries_eeh_of_probe - EEH probe on the given device
* @dn: OF node
* @flag: Unused
*
* When EEH module is installed during system boot, all PCI devices
* are checked one by one to see if it supports EEH. The function
* is introduced for the purpose.
*/
static void *pseries_eeh_of_probe(struct device_node *dn, void *flag)
{
struct eeh_dev *edev;
struct eeh_pe pe;
const u32 *class_code, *vendor_id, *device_id;
const u32 *regs;
int enable = 0;
int ret;
/* Retrieve OF node and eeh device */
edev = of_node_to_eeh_dev(dn);
if (!of_device_is_available(dn))
return NULL;
/* Retrieve class/vendor/device IDs */
class_code = of_get_property(dn, "class-code", NULL);
vendor_id = of_get_property(dn, "vendor-id", NULL);
device_id = of_get_property(dn, "device-id", NULL);
/* Skip for bad OF node or PCI-ISA bridge */
if (!class_code || !vendor_id || !device_id)
return NULL;
if (dn->type && !strcmp(dn->type, "isa"))
return NULL;
/* Update class code and mode of eeh device */
edev->class_code = *class_code;
edev->mode = 0;
/* Retrieve the device address */
regs = of_get_property(dn, "reg", NULL);
if (!regs) {
pr_warning("%s: OF node property %s::reg not found\n",
__func__, dn->full_name);
return NULL;
}
/* Initialize the fake PE */
memset(&pe, 0, sizeof(struct eeh_pe));
pe.phb = edev->phb;
pe.config_addr = regs[0];
/* Enable EEH on the device */
ret = eeh_ops->set_option(&pe, EEH_OPT_ENABLE);
if (!ret) {
edev->config_addr = regs[0];
/* Retrieve PE address */
edev->pe_config_addr = eeh_ops->get_pe_addr(&pe);
pe.addr = edev->pe_config_addr;
/* Some older systems (Power4) allow the ibm,set-eeh-option
* call to succeed even on nodes where EEH is not supported.
* Verify support explicitly.
*/
ret = eeh_ops->get_state(&pe, NULL);
if (ret > 0 && ret != EEH_STATE_NOT_SUPPORT)
enable = 1;
if (enable) {
eeh_subsystem_enabled = 1;
eeh_add_to_parent_pe(edev);
pr_debug("%s: EEH enabled on %s PHB#%d-PE#%x, config addr#%x\n",
__func__, dn->full_name, pe.phb->global_number,
pe.addr, pe.config_addr);
} else if (dn->parent && of_node_to_eeh_dev(dn->parent) &&
(of_node_to_eeh_dev(dn->parent))->pe) {
/* This device doesn't support EEH, but it may have an
* EEH parent, in which case we mark it as supported.
*/
edev->config_addr = of_node_to_eeh_dev(dn->parent)->config_addr;
edev->pe_config_addr = of_node_to_eeh_dev(dn->parent)->pe_config_addr;
eeh_add_to_parent_pe(edev);
}
}
/* Save memory bars */
eeh_save_bars(edev);
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;
}
/**
* pseries_eeh_get_state - Retrieve PE state
* @dn: PE associated device node
* @state: return value
*
* Retrieve the state of the specified PE. On RTAS compliant
* pseries platform, there already has one dedicated RTAS function
* for the purpose. It's notable that the associated PE config address
* might be ready when calling the function. Therefore, endeavour to
* use the PE config address if possible. Further more, there're 2
* RTAS calls for the purpose, we need to try the new one and back
* to the old one if the new one couldn't work properly.
*/
static int pseries_eeh_get_state(struct device_node *dn, int *state)
{
struct eeh_dev *edev;
int config_addr;
int ret;
int rets[4];
int result;
/* Figure out PE config address if possible */
edev = of_node_to_eeh_dev(dn);
config_addr = edev->config_addr;
if (edev->pe_config_addr)
config_addr = edev->pe_config_addr;
if (ibm_read_slot_reset_state2 != RTAS_UNKNOWN_SERVICE) {
ret = rtas_call(ibm_read_slot_reset_state2, 3, 4, rets,
config_addr, BUID_HI(edev->phb->buid),
BUID_LO(edev->phb->buid));
} else if (ibm_read_slot_reset_state != RTAS_UNKNOWN_SERVICE) {
/* Fake PE unavailable info */
rets[2] = 0;
ret = rtas_call(ibm_read_slot_reset_state, 3, 3, rets,
config_addr, BUID_HI(edev->phb->buid),
BUID_LO(edev->phb->buid));
} else {
return EEH_STATE_NOT_SUPPORT;
}
if (ret)
return ret;
/* Parse the result out */
result = 0;
if (rets[1]) {
switch(rets[0]) {
case 0:
result &= ~EEH_STATE_RESET_ACTIVE;
result |= EEH_STATE_MMIO_ACTIVE;
result |= EEH_STATE_DMA_ACTIVE;
break;
case 1:
result |= EEH_STATE_RESET_ACTIVE;
result |= EEH_STATE_MMIO_ACTIVE;
result |= EEH_STATE_DMA_ACTIVE;
break;
case 2:
result &= ~EEH_STATE_RESET_ACTIVE;
result &= ~EEH_STATE_MMIO_ACTIVE;
result &= ~EEH_STATE_DMA_ACTIVE;
break;
case 4:
result &= ~EEH_STATE_RESET_ACTIVE;
result &= ~EEH_STATE_MMIO_ACTIVE;
result &= ~EEH_STATE_DMA_ACTIVE;
result |= EEH_STATE_MMIO_ENABLED;
break;
case 5:
if (rets[2]) {
if (state) *state = rets[2];
result = EEH_STATE_UNAVAILABLE;
} else {
result = EEH_STATE_NOT_SUPPORT;
}
default:
result = EEH_STATE_NOT_SUPPORT;
}
} else {
result = EEH_STATE_NOT_SUPPORT;
}
return result;
}
/**
* 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;
}
