powerpc/eeh: Set EEH_PE_RESET on PE reset

The patch introduces additional flag EEH_PE_RESET to indicate the
corresponding PE is under reset. In turn, the PE retrieval bakcend
on PowerNV platform can return unfrozen state for the EEH core to
moving forward. Flag EEH_PE_CFG_BLOCKED isn't the correct one for
the purpose.

In PCI passthrou case, the problem is more worse: Guest doesn't
recover 6th EEH error. The PE is left in isolated (frozen) and
config blocked state on Broadcom adapters. We can't retrieve the
PE's state correctly any more, even from the host side via sysfs
/sys/bus/pci/devices/xxx/eeh_pe_state.

Reported-by: Rajeshkumar Subramanian <rajeshkumars@in.ibm.com>
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This commit is contained in:
Gavin Shan 2014-11-14 10:47:29 +11:00 committed by Benjamin Herrenschmidt
parent b85743ee95
commit 28bf36f92a
4 changed files with 8 additions and 9 deletions

View file

@ -72,6 +72,7 @@ struct device_node;
#define EEH_PE_ISOLATED (1 << 0) /* Isolated PE */
#define EEH_PE_RECOVERING (1 << 1) /* Recovering PE */
#define EEH_PE_CFG_BLOCKED (1 << 2) /* Block config access */
#define EEH_PE_RESET (1 << 3) /* PE reset in progress */
#define EEH_PE_KEEP (1 << 8) /* Keep PE on hotplug */
#define EEH_PE_CFG_RESTRICTED (1 << 9) /* Block config on error */

View file

@ -760,6 +760,9 @@ int eeh_reset_pe(struct eeh_pe *pe)
int flags = (EEH_STATE_MMIO_ACTIVE | EEH_STATE_DMA_ACTIVE);
int i, state, ret;
/* Mark as reset and block config space */
eeh_pe_state_mark(pe, EEH_PE_RESET | EEH_PE_CFG_BLOCKED);
/* Take three shots at resetting the bus */
for (i = 0; i < 3; i++) {
eeh_reset_pe_once(pe);
@ -788,6 +791,7 @@ int eeh_reset_pe(struct eeh_pe *pe)
}
out:
eeh_pe_state_clear(pe, EEH_PE_RESET | EEH_PE_CFG_BLOCKED);
return ret;
}

View file

@ -528,13 +528,11 @@ int eeh_pe_reset_and_recover(struct eeh_pe *pe)
eeh_pe_dev_traverse(pe, eeh_report_error, &result);
/* Issue reset */
eeh_pe_state_mark(pe, EEH_PE_CFG_BLOCKED);
ret = eeh_reset_pe(pe);
if (ret) {
eeh_pe_state_clear(pe, EEH_PE_RECOVERING | EEH_PE_CFG_BLOCKED);
eeh_pe_state_clear(pe, EEH_PE_RECOVERING);
return ret;
}
eeh_pe_state_clear(pe, EEH_PE_CFG_BLOCKED);
/* Unfreeze the PE */
ret = eeh_clear_pe_frozen_state(pe, true);
@ -601,19 +599,15 @@ static int eeh_reset_device(struct eeh_pe *pe, struct pci_bus *bus)
* config accesses. So we prefer to block them. However, controlled
* PCI config accesses initiated from EEH itself are allowed.
*/
eeh_pe_state_mark(pe, EEH_PE_CFG_BLOCKED);
rc = eeh_reset_pe(pe);
if (rc) {
eeh_pe_state_clear(pe, EEH_PE_CFG_BLOCKED);
if (rc)
return rc;
}
pci_lock_rescan_remove();
/* Restore PE */
eeh_ops->configure_bridge(pe);
eeh_pe_restore_bars(pe);
eeh_pe_state_clear(pe, EEH_PE_CFG_BLOCKED);
/* Clear frozen state */
rc = eeh_clear_pe_frozen_state(pe, false);

View file

@ -372,7 +372,7 @@ static int ioda_eeh_get_pe_state(struct eeh_pe *pe)
* moving forward, we have to return operational
* state during PE reset.
*/
if (pe->state & EEH_PE_CFG_BLOCKED) {
if (pe->state & EEH_PE_RESET) {
result = (EEH_STATE_MMIO_ACTIVE |
EEH_STATE_DMA_ACTIVE |
EEH_STATE_MMIO_ENABLED |