The patch titled
Subject: watchdog: update comments to explain some code
has been removed from the -mm tree. Its filename was
watchdog-implement-error-handling-for-failure-to-set-up-hardware-perf-events-fix.patch
This patch was dropped because it was folded into watchdog-implement-error-handling-for-failure-to-set-up-hardware-perf-events.patch
------------------------------------------------------
From: Don Zickus <dzickus@xxxxxxxxxx>
Subject: watchdog: update comments to explain some code
This patch was written to update some comments and concerns from Andrew
and is expected to be folded into patch
watchdog-implement-error-handling-for-failure-to-set-up-hardware-perf-events.patch
Resolves:
- comments around barriers
- comments around blind shutdown of hardware lockup detector
- printk letting someone know hardlockup detector is being shut down
Signed-off-by: Don Zickus <dzickus@xxxxxxxxxx>
Suggested-by: Ulrich Obergfell <uobergfe@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
kernel/watchdog.c | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff -puN kernel/watchdog.c~watchdog-implement-error-handling-for-failure-to-set-up-hardware-perf-events-fix kernel/watchdog.c
--- a/kernel/watchdog.c~watchdog-implement-error-handling-for-failure-to-set-up-hardware-perf-events-fix
+++ a/kernel/watchdog.c
@@ -508,6 +508,12 @@ static void watchdog(unsigned int cpu)
* failure path. Check for failures that can occur asynchronously -
* for example, when CPUs are on-lined - and shut down the hardware
* perf event on each CPU accordingly.
+ *
+ * The only non-obvious place this bit can be cleared is through
+ * watchdog_nmi_enable(), so a pr_info() is placed there. Placing a
+ * pr_info here would be too noisy as it would result in a message
+ * every few seconds if the hardlockup was disabled but the softlockup
+ * enabled.
*/
if (!(watchdog_enabled & NMI_WATCHDOG_ENABLED))
watchdog_nmi_disable(cpu);
@@ -565,6 +571,9 @@ handle_err:
* Disable the hard lockup detector if _any_ CPU fails to set up
* set up the hardware perf event. The watchdog() function checks
* the NMI_WATCHDOG_ENABLED bit periodically.
+ *
+ * The barriers are for syncing up watchdog_enabled across all the
+ * cpus, as clear_bit() does not use barriers.
*/
smp_mb__before_atomic();
clear_bit(NMI_WATCHDOG_ENABLED_BIT, &watchdog_enabled);
@@ -583,6 +592,9 @@ handle_err:
else
pr_err("disabled (cpu%i): unable to create perf event: %ld\n",
cpu, PTR_ERR(event));
+
+ pr_info("Shutting down hard lockup detector on all cpus\n");
+
return PTR_ERR(event);
/* success path */
_
Patches currently in -mm which might be from dzickus@xxxxxxxxxx are
watchdog-new-definitions-and-variables-initialization.patch
watchdog-introduce-the-proc_watchdog_update-function.patch
watchdog-move-definition-of-watchdog_proc_mutex-outside-of-proc_dowatchdog.patch
watchdog-introduce-the-proc_watchdog_common-function.patch
watchdog-introduce-separate-handlers-for-parameters-in-proc-sys-kernel.patch
watchdog-implement-error-handling-for-failure-to-set-up-hardware-perf-events.patch
watchdog-enable-the-new-user-interface-of-the-watchdog-mechanism.patch
watchdog-enable-the-new-user-interface-of-the-watchdog-mechanism-fix.patch
watchdog-clean-up-some-function-names-and-arguments.patch
watchdog-introduce-the-hardlockup_detector_disable-function.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
