Commit ba61ca4a authored by Huang Ying's avatar Huang Ying Committed by Len Brown
Browse files

ACPI, APEI, GHES: Add hardware memory error recovery support 



memory_failure_queue() is called when recoverable memory errors are
notified by firmware to do the recovery work.

Signed-off-by: default avatarHuang Ying <ying.huang@intel.com>
Signed-off-by: default avatarLen Brown <len.brown@intel.com>
parent ea8f5fb8

drivers/acpi/apei/Kconfig

+7 −0
Original line number Original line Diff line number Diff line
@@ -32,6 +32,13 @@ config ACPI_APEI_PCIEAER 
	  PCIe AER errors may be reported via APEI firmware first mode.

	  PCIe AER errors may be reported via APEI firmware first mode.

	  Turn on this option to enable the corresponding support.

	  Turn on this option to enable the corresponding support.





config ACPI_APEI_MEMORY_FAILURE

	bool "APEI memory error recovering support"

	depends on ACPI_APEI && MEMORY_FAILURE

	help

	  Memory errors may be reported via APEI firmware first mode.

	  Turn on this option to enable the memory recovering support.



config ACPI_APEI_EINJ

config ACPI_APEI_EINJ

	tristate "APEI Error INJection (EINJ)"

	tristate "APEI Error INJection (EINJ)"

	depends on ACPI_APEI && DEBUG_FS

	depends on ACPI_APEI && DEBUG_FS

drivers/acpi/apei/ghes.c

+17 −7
Original line number Original line Diff line number Diff line
@@ -451,22 +451,32 @@ static void ghes_clear_estatus(struct ghes *ghes) 




static void ghes_do_proc(const struct acpi_hest_generic_status *estatus)

static void ghes_do_proc(const struct acpi_hest_generic_status *estatus)

{

{

	int sev, processed = 0;

	int sev, sec_sev;

	struct acpi_hest_generic_data *gdata;

	struct acpi_hest_generic_data *gdata;





	sev = ghes_severity(estatus->error_severity);

	sev = ghes_severity(estatus->error_severity);

	apei_estatus_for_each_section(estatus, gdata) {

	apei_estatus_for_each_section(estatus, gdata) {

#ifdef CONFIG_X86_MCE

		sec_sev = ghes_severity(gdata->error_severity);

		if (!uuid_le_cmp(*(uuid_le *)gdata->section_type,

		if (!uuid_le_cmp(*(uuid_le *)gdata->section_type,

				 CPER_SEC_PLATFORM_MEM)) {

				 CPER_SEC_PLATFORM_MEM)) {

			apei_mce_report_mem_error(

			struct cper_sec_mem_err *mem_err;

				sev == GHES_SEV_CORRECTED,

			mem_err = (struct cper_sec_mem_err *)(gdata+1);

				(struct cper_sec_mem_err *)(gdata+1));

#ifdef CONFIG_X86_MCE

			processed = 1;

			apei_mce_report_mem_error(sev == GHES_SEV_CORRECTED,

						  mem_err);

#endif

#ifdef CONFIG_ACPI_APEI_MEMORY_FAILURE

			if (sev == GHES_SEV_RECOVERABLE &&

			    sec_sev == GHES_SEV_RECOVERABLE &&

			    mem_err->validation_bits & CPER_MEM_VALID_PHYSICAL_ADDRESS) {

				unsigned long pfn;

				pfn = mem_err->physical_addr >> PAGE_SHIFT;

				memory_failure_queue(pfn, 0, 0);

			}

			}

#endif

#endif

		}

		}

	}

	}

}





static void __ghes_print_estatus(const char *pfx,

static void __ghes_print_estatus(const char *pfx,

				 const struct acpi_hest_generic *generic,

				 const struct acpi_hest_generic *generic,