diff mbox

[v3,3/3] mce, acpi/apei: Soft-offline a page on firmware GHES notification

Message ID 20130710092225.20744.63823.stgit@localhost.localdomain (mailing list archive)
State Not Applicable, archived
Headers show

Commit Message

Naveen N. Rao July 10, 2013, 9:27 a.m. UTC
On 07/09/2013 12:30 AM, Tony Luck wrote:
> I was off on vacation last week - looks like you got lots done without me
> :-)
> 
> I have parts 1 & 2 applied to an internal tree. Looks like parts 3 & 4 need
> a few final polishes to get an Ack from Boris.

Cool :)
Here is the updated patch for part 3 incorporating Boris's comments.

Thanks,
Naveen
--

If the firmware indicates in GHES error data entry that the error threshold
has exceeded for a corrected error event, then we try to soft-offline the
page. This could be called in interrupt context, so we queue this up similar
to how we handle memory failure scenarios.


Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
---
 drivers/acpi/apei/ghes.c |   38 +++++++++++++++++++++++++++++---------
 include/linux/mm.h       |    1 +
 mm/memory-failure.c      |    5 ++++-
 3 files changed, 34 insertions(+), 10 deletions(-)


--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Comments

Borislav Petkov July 10, 2013, 9:38 a.m. UTC | #1
On Wed, Jul 10, 2013 at 02:57:01PM +0530, Naveen N. Rao wrote:
> On 07/09/2013 12:30 AM, Tony Luck wrote:
> > I was off on vacation last week - looks like you got lots done without me
> > :-)
> > 
> > I have parts 1 & 2 applied to an internal tree. Looks like parts 3 & 4 need
> > a few final polishes to get an Ack from Boris.
> 
> Cool :)
> Here is the updated patch for part 3 incorporating Boris's comments.
> 
> Thanks,
> Naveen
> --
> 
> If the firmware indicates in GHES error data entry that the error threshold
> has exceeded for a corrected error event, then we try to soft-offline the
> page. This could be called in interrupt context, so we queue this up similar
> to how we handle memory failure scenarios.
> 
> 
> Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>

Acked-by: Borislav Petkov <bp@suse.de>

Thanks.
Tony Luck July 10, 2013, 8:05 p.m. UTC | #2
>> Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
>
> Acked-by: Borislav Petkov <bp@suse.de>

Applied-by: Tony Luck :-)

Naveen: Thanks for having this idea, implementing it, and sticking
with it through the review process.

Once 3.11-rc1 is out I'll ask Ingo to pull this series to the tip tree
... and then on to 3.12

-Tony
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c
index fcd7d91..ba99d32 100644
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -409,6 +409,34 @@  static void ghes_clear_estatus(struct ghes *ghes)
 	ghes->flags &= ~GHES_TO_CLEAR;
 }
 
+static void ghes_handle_memory_failure(struct acpi_hest_generic_data *gdata, int sev)
+{
+#ifdef CONFIG_ACPI_APEI_MEMORY_FAILURE
+	unsigned long pfn;
+	int sec_sev = ghes_severity(gdata->error_severity);
+	struct cper_sec_mem_err *mem_err;
+	mem_err = (struct cper_sec_mem_err *)(gdata + 1);
+
+	if (sec_sev == GHES_SEV_CORRECTED &&
+	    (gdata->flags & CPER_SEC_ERROR_THRESHOLD_EXCEEDED) &&
+	    (mem_err->validation_bits & CPER_MEM_VALID_PHYSICAL_ADDRESS)) {
+		pfn = mem_err->physical_addr >> PAGE_SHIFT;
+		if (pfn_valid(pfn))
+			memory_failure_queue(pfn, 0, MF_SOFT_OFFLINE);
+		else if (printk_ratelimit())
+			pr_warning(FW_WARN GHES_PFX
+			"Invalid address in generic error data: %#llx\n",
+			mem_err->physical_addr);
+	}
+	if (sev == GHES_SEV_RECOVERABLE &&
+	    sec_sev == GHES_SEV_RECOVERABLE &&
+	    mem_err->validation_bits & CPER_MEM_VALID_PHYSICAL_ADDRESS) {
+		pfn = mem_err->physical_addr >> PAGE_SHIFT;
+		memory_failure_queue(pfn, 0, 0);
+	}
+#endif
+}
+
 static void ghes_do_proc(struct ghes *ghes,
 			 const struct acpi_hest_generic_status *estatus)
 {
@@ -428,15 +456,7 @@  static void ghes_do_proc(struct ghes *ghes,
 			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
+			ghes_handle_memory_failure(gdata, sev);
 		}
 #ifdef CONFIG_ACPI_APEI_PCIEAER
 		else if (!uuid_le_cmp(*(uuid_le *)gdata->section_type,
diff --git a/include/linux/mm.h b/include/linux/mm.h
index e0c8528..958e9efd 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1784,6 +1784,7 @@  enum mf_flags {
 	MF_COUNT_INCREASED = 1 << 0,
 	MF_ACTION_REQUIRED = 1 << 1,
 	MF_MUST_KILL = 1 << 2,
+	MF_SOFT_OFFLINE = 1 << 3,
 };
 extern int memory_failure(unsigned long pfn, int trapno, int flags);
 extern void memory_failure_queue(unsigned long pfn, int trapno, int flags);
diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index ceb0c7f..0d6717e 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -1286,7 +1286,10 @@  static void memory_failure_work_func(struct work_struct *work)
 		spin_unlock_irqrestore(&mf_cpu->lock, proc_flags);
 		if (!gotten)
 			break;
-		memory_failure(entry.pfn, entry.trapno, entry.flags);
+		if (entry.flags & MF_SOFT_OFFLINE)
+			soft_offline_page(pfn_to_page(entry.pfn), entry.flags);
+		else
+			memory_failure(entry.pfn, entry.trapno, entry.flags);
 	}
 }