/* Check for an eeh failure at the given token address.
* The given value has been read and it should be 1's (0xff, 0xffff or
* 0xffffffff).
*
* Probe to determine if an error actually occurred. If not return val.
* Otherwise panic.
*/
unsigned long eeh_check_failure(void *token, unsigned long val)
{
unsigned long addr;
struct pci_dev *dev;
struct device_node *dn;
unsigned long ret, rets[2];
/* IO BAR access could get us here...or if we manually force EEH
* operation on even if the hardware won't support it.
*/
if (!eeh_implemented || ibm_read_slot_reset_state == RTAS_UNKNOWN_SERVICE)
return val;
/* Finding the phys addr + pci device is quite expensive.
* However, the RTAS call is MUCH slower.... :(
*/
addr = eeh_token_to_phys((unsigned long)token);
dev = pci_find_dev_by_addr(addr);
if (!dev) {
printk("EEH: no pci dev found for addr=0x%lx\n", addr);
return val;
}
dn = pci_device_to_OF_node(dev);
if (!dn) {
printk("EEH: no pci dn found for addr=0x%lx\n", addr);
return val;
}
/* Access to IO BARs might get this far and still not want checking. */
if (!(dn->eeh_mode & EEH_MODE_SUPPORTED) || dn->eeh_mode & EEH_MODE_NOCHECK)
return val;
/* Now test for an EEH failure. This is VERY expensive.
* Note that the eeh_config_addr may be a parent device
* in the case of a device behind a bridge, or it may be
* function zero of a multi-function device.
* In any case they must share a common PHB.
*/
if (dn->eeh_config_addr) {
ret = rtas_call(ibm_read_slot_reset_state, 3, 3, rets,
dn->eeh_config_addr, BUID_HI(dn->phb->buid), BUID_LO(dn->phb->buid));
if (ret == 0 && rets[1] == 1 && rets[0] >= 2) {
/*
* XXX We should create a separate sysctl for this.
*
* Since the panic_on_oops sysctl is used to halt
* the system in light of potential corruption, we
* can use it here.
*/
if (panic_on_oops)
panic("EEH: MMIO failure (%ld) on device:\n%s\n", rets[0], pci_name(dev));
else
printk("EEH: MMIO failure (%ld) on device:\n%s\n", rets[0], pci_name(dev));
}
}
eeh_false_positives++;
return val; /* good case */
}
/**
* eeh_check_failure - check if all 1's data is due to EEH slot freeze
* @token i/o token, should be address in the form 0xA....
* @val value, should be all 1's (XXX why do we need this arg??)
*
* Check for an EEH failure at the given token address. Call this
* routine if the result of a read was all 0xff's and you want to
* find out if this is due to an EEH slot freeze event. This routine
* will query firmware for the EEH status.
*
* Note this routine is safe to call in an interrupt context.
*/
unsigned long eeh_check_failure(const volatile void __iomem *token, unsigned long val)
{
unsigned long addr;
struct pci_dev *dev;
struct device_node *dn;
/* Finding the phys addr + pci device; this is pretty quick. */
addr = eeh_token_to_phys((unsigned long __force) token);
dev = pci_get_device_by_addr(addr);
if (!dev) {
no_device++;
return val;
}
dn = pci_device_to_OF_node(dev);
eeh_dn_check_failure (dn, dev);
pci_dev_put(dev);
return val;
}
/* Check for an eeh failure at the given token address.
* The given value has been read and it should be 1's (0xff, 0xffff or
* 0xffffffff).
*
* Probe to determine if an error actually occurred. If not return val.
* Otherwise panic.
*/
unsigned long eeh_check_failure(void *token, unsigned long val)
{
unsigned long addr;
struct pci_dev *dev;
struct device_node *dn;
unsigned long ret, rets[2];
/* IO BAR access could get us here...or if we manually force EEH
* operation on even if the hardware won't support it.
*/
if (!eeh_implemented || ibm_read_slot_reset_state == RTAS_UNKNOWN_SERVICE)
return val;
/* Finding the phys addr + pci device is quite expensive.
* However, the RTAS call is MUCH slower.... :(
*/
addr = eeh_token_to_phys((unsigned long)token);
dev = pci_find_dev_by_addr(addr);
if (!dev) {
printk("EEH: no pci dev found for addr=0x%lx\n", addr);
return val;
}
dn = pci_device_to_OF_node(dev);
if (!dn) {
printk("EEH: no pci dn found for addr=0x%lx\n", addr);
return val;
}
/* Access to IO BARs might get this far and still not want checking. */
if (!(dn->eeh_mode & EEH_MODE_SUPPORTED) || dn->eeh_mode & EEH_MODE_NOCHECK)
return val;
/* Now test for an EEH failure. This is VERY expensive.
* Note that the eeh_config_addr may be a parent device
* in the case of a device behind a bridge, or it may be
* function zero of a multi-function device.
* In any case they must share a common PHB.
*/
if (dn->eeh_config_addr) {
ret = rtas_call(ibm_read_slot_reset_state, 3, 3, rets,
dn->eeh_config_addr, BUID_HI(dn->phb->buid), BUID_LO(dn->phb->buid));
if (ret == 0 && rets[1] == 1 && rets[0] >= 2) {
unsigned char slot_err_buf[RTAS_ERROR_LOG_MAX];
unsigned long slot_err_ret;
memset(slot_err_buf, 0, RTAS_ERROR_LOG_MAX);
slot_err_ret = rtas_call(rtas_token("ibm,slot-error-detail"),
8, 1, dn->eeh_config_addr,
BUID_HI(dn->phb->buid), BUID_LO(dn->phb->buid),
NULL, 0, __pa(slot_err_buf), RTAS_ERROR_LOG_MAX,
2 /* Permanent Error */);
if (slot_err_ret == 0)
log_error(slot_err_buf, ERR_TYPE_RTAS_LOG, 1 /* Fatal */);
panic("EEH: MMIO failure (%ld) on device:\n %s %s\n",
rets[0], dev->slot_name, dev->name);
}
}
eeh_false_positives++;
return val; /* good case */
}
