From mboxrd@z Thu Jan 1 00:00:00 1970 From: Huang Ying Subject: [RFC 2/4] ACPI, APEI, GHES: Add hardware memory error recovery support Date: Tue, 19 Apr 2011 09:43:37 +0800 Message-ID: <1303177419-22466-2-git-send-email-ying.huang@intel.com> References: <1303177419-22466-1-git-send-email-ying.huang@intel.com> Return-path: Received: from mga03.intel.com ([143.182.124.21]:27981 "EHLO mga03.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753571Ab1DSBnr (ORCPT ); Mon, 18 Apr 2011 21:43:47 -0400 In-Reply-To: <1303177419-22466-1-git-send-email-ying.huang@intel.com> Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: Len Brown Cc: linux-kernel@vger.kernel.org, Andi Kleen , Tony Luck , ying.huang@intel.com, linux-acpi@vger.kernel.org memory_failure_queue() is called when recoverable memory errors are notified by firmware to do the recovery work. Signed-off-by: Huang Ying --- drivers/acpi/apei/Kconfig | 7 +++++++ drivers/acpi/apei/ghes.c | 24 +++++++++++++++++------- 2 files changed, 24 insertions(+), 7 deletions(-) --- a/drivers/acpi/apei/Kconfig +++ b/drivers/acpi/apei/Kconfig @@ -31,6 +31,13 @@ config ACPI_APEI_PCIEAER PCIe AER errors may be reported via APEI firmware first mode. 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 tristate "APEI Error INJection (EINJ)" depends on ACPI_APEI && DEBUG_FS --- a/drivers/acpi/apei/ghes.c +++ b/drivers/acpi/apei/ghes.c @@ -430,20 +430,30 @@ static void ghes_clear_estatus(struct gh 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; sev = ghes_severity(estatus->error_severity); 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, CPER_SEC_PLATFORM_MEM)) { - apei_mce_report_mem_error( - sev == GHES_SEV_CORRECTED, - (struct cper_sec_mem_err *)(gdata+1)); - processed = 1; - } + struct cper_sec_mem_err *mem_err; + mem_err = (struct cper_sec_mem_err *)(gdata+1); +#ifdef CONFIG_X86_MCE + 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 + } } }