The CPER error record has a reset bit that indicates that the platform has reset the bus. The reset bit can be set for any severity error including recoverable. From the AER code path's perspective, any error is fatal if the bus has been reset. This patch upgrades the severity of the AER recovery to AER_FATAL whenever the CPER error record indicates that the bus has been reset. Changes since v1: Fixed a typo in comment. Signed-off-by: Betty Dall <betty.dall@xxxxxx> --- drivers/acpi/apei/ghes.c | 21 ++++++++++++++++++++- 1 files changed, 20 insertions(+), 1 deletions(-) diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c index d668a8a..1c67d5a 100644 --- a/drivers/acpi/apei/ghes.c +++ b/drivers/acpi/apei/ghes.c @@ -451,7 +451,26 @@ static void ghes_do_proc(struct ghes *ghes, int aer_severity; devfn = PCI_DEVFN(pcie_err->device_id.device, pcie_err->device_id.function); - aer_severity = cper_severity_to_aer(sev); + /* + * Some Firmware First implementations + * put the device in SBR to contain + * the error. This is indicated by the + * CPER Section Descriptor Flags reset + * bit which means the component must + * be re-initialized or re-enabled + * prior to use. Promoting the AER + * serverity to FATAL will cause the + * AER code to link_reset and allow + * drivers to reprogram their cards. + */ + if (gdata->flags & CPER_SEC_RESET) + aer_severity = cper_severity_to_aer( + CPER_SEV_FATAL); + else + aer_severity = + cper_severity_to_aer(sev); + + aer_recover_queue(pcie_err->device_id.segment, pcie_err->device_id.bus, devfn, aer_severity); -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
