summaryrefslogtreecommitdiff
path: root/arch/powerpc/kernel
diff options
context:
space:
mode:
authorGavin Shan <gwshan@linux.vnet.ibm.com>2014-04-24 08:00:12 (GMT)
committerBenjamin Herrenschmidt <benh@kernel.crashing.org>2014-04-28 07:34:02 (GMT)
commitd0914f503f7ba2cd078b123983562be8951296d3 (patch)
tree1b8ddd75171e8ff07f8e52208906fd7a825170e8 /arch/powerpc/kernel
parent7b401850a1c9b64b010b3d107321df4e7d066995 (diff)
downloadlinux-d0914f503f7ba2cd078b123983562be8951296d3.tar.xz
powerpc/eeh: Block PCI-CFG access during PE reset
We've observed multiple PE reset failures because of PCI-CFG access during that period. Potentially, some device drivers can't support EEH very well and they can't put the device to motionless state before PE reset. So those device drivers might produce PCI-CFG accesses during PE reset. Also, we could have PCI-CFG access from user space (e.g. "lspci"). Since access to frozen PE should return 0xFF's, we can block PCI-CFG access during the period of PE reset so that we won't get recrusive EEH errors. The patch adds flag EEH_PE_RESET, which is kept during PE reset. The PowerNV/pSeries PCI-CFG accessors reuse the flag to block PCI-CFG accordingly. Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Diffstat (limited to 'arch/powerpc/kernel')
-rw-r--r--arch/powerpc/kernel/eeh_driver.c13
-rw-r--r--arch/powerpc/kernel/rtas_pci.c66
2 files changed, 61 insertions, 18 deletions
diff --git a/arch/powerpc/kernel/eeh_driver.c b/arch/powerpc/kernel/eeh_driver.c
index 1ddc046..6d91b51 100644
--- a/arch/powerpc/kernel/eeh_driver.c
+++ b/arch/powerpc/kernel/eeh_driver.c
@@ -451,19 +451,28 @@ static int eeh_reset_device(struct eeh_pe *pe, struct pci_bus *bus)
eeh_pe_dev_traverse(pe, eeh_rmv_device, &removed);
}
- /* Reset the pci controller. (Asserts RST#; resets config space).
+ /*
+ * 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.
+ *
+ * During the reset, it's very dangerous to have uncontrolled PCI
+ * 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_RESET);
rc = eeh_reset_pe(pe);
- if (rc)
+ if (rc) {
+ eeh_pe_state_clear(pe, EEH_PE_RESET);
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_RESET);
/* Give the system 5 seconds to finish running the user-space
* hotplug shutdown scripts, e.g. ifdown for ethernet. Yes,
diff --git a/arch/powerpc/kernel/rtas_pci.c b/arch/powerpc/kernel/rtas_pci.c
index 7d4c717..c168337 100644
--- a/arch/powerpc/kernel/rtas_pci.c
+++ b/arch/powerpc/kernel/rtas_pci.c
@@ -80,10 +80,6 @@ int rtas_read_config(struct pci_dn *pdn, int where, int size, u32 *val)
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;
}
@@ -92,18 +88,39 @@ static int rtas_pci_read_config(struct pci_bus *bus,
int where, int size, u32 *val)
{
struct device_node *busdn, *dn;
-
- busdn = pci_bus_to_OF_node(bus);
+ struct pci_dn *pdn;
+ bool found = false;
+#ifdef CONFIG_EEH
+ struct eeh_dev *edev;
+#endif
+ int ret;
/* Search only direct children of the bus */
+ *val = 0xFFFFFFFF;
+ busdn = pci_bus_to_OF_node(bus);
for (dn = busdn->child; dn; dn = dn->sibling) {
- struct pci_dn *pdn = PCI_DN(dn);
+ pdn = PCI_DN(dn);
if (pdn && pdn->devfn == devfn
- && of_device_is_available(dn))
- return rtas_read_config(pdn, where, size, val);
+ && of_device_is_available(dn)) {
+ found = true;
+ break;
+ }
}
- return PCIBIOS_DEVICE_NOT_FOUND;
+ if (!found)
+ return PCIBIOS_DEVICE_NOT_FOUND;
+#ifdef CONFIG_EEH
+ edev = of_node_to_eeh_dev(dn);
+ if (edev && edev->pe && edev->pe->state & EEH_PE_RESET)
+ return PCIBIOS_DEVICE_NOT_FOUND;
+#endif
+
+ ret = rtas_read_config(pdn, where, size, val);
+ if (*val == EEH_IO_ERROR_VALUE(size) &&
+ eeh_dev_check_failure(of_node_to_eeh_dev(dn)))
+ return PCIBIOS_DEVICE_NOT_FOUND;
+
+ return ret;
}
int rtas_write_config(struct pci_dn *pdn, int where, int size, u32 val)
@@ -136,17 +153,34 @@ static int rtas_pci_write_config(struct pci_bus *bus,
int where, int size, u32 val)
{
struct device_node *busdn, *dn;
-
- busdn = pci_bus_to_OF_node(bus);
+ struct pci_dn *pdn;
+ bool found = false;
+#ifdef CONFIG_EEH
+ struct eeh_dev *edev;
+#endif
+ int ret;
/* Search only direct children of the bus */
+ busdn = pci_bus_to_OF_node(bus);
for (dn = busdn->child; dn; dn = dn->sibling) {
- struct pci_dn *pdn = PCI_DN(dn);
+ pdn = PCI_DN(dn);
if (pdn && pdn->devfn == devfn
- && of_device_is_available(dn))
- return rtas_write_config(pdn, where, size, val);
+ && of_device_is_available(dn)) {
+ found = true;
+ break;
+ }
}
- return PCIBIOS_DEVICE_NOT_FOUND;
+
+ if (!found)
+ return PCIBIOS_DEVICE_NOT_FOUND;
+#ifdef CONFIG_EEH
+ edev = of_node_to_eeh_dev(dn);
+ if (edev && edev->pe && (edev->pe->state & EEH_PE_RESET))
+ return PCIBIOS_DEVICE_NOT_FOUND;
+#endif
+ ret = rtas_write_config(pdn, where, size, val);
+
+ return ret;
}
static struct pci_ops rtas_pci_ops = {