<I sent an another email from sundar.iyer@xxxxxxxxx, but I think it
hasnt got through>
Hi,
We need to count/trace the runtime PM transitions for all components.
Is this small snippet of change valid? This can be extended to include
a transition to active state, get exposed in sysfs, controls to reset
the value like other existing sysfs entries.
Or is the wakeup_count the correct means to figure out the resume events?
diff --git a/drivers/base/power/runtime.c
b/drivers/base/power/runtime.c index 1f339d5..233f8f5 100644
--- a/drivers/base/power/runtime.c
+++ b/drivers/base/power/runtime.c
@@ -52,6 +52,8 @@ static void __update_runtime_status(struct device
*dev, enum rpm_status status)
update_pm_runtime_accounting(dev);
trace_runtime_pm_status(dev, status);
dev->power.runtime_status = status;
+ if (status == RPM_SUSPENDED)
+ atomic_inc(&dev->power.transition_suspend_count);
}
/**
siyer5@pnp-buildsrv-4:~/ICS/hardware/intel/linux-2.6$ git diff
include/linux/pm.h diff --git a/include/linux/pm.h
b/include/linux/pm.h index 411e4f4..076a0e3 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -442,6 +442,7 @@ struct dev_pm_info {
wait_queue_head_t wait_queue;
atomic_t usage_count;
atomic_t child_count;
+ atomic_t suspend_transition_count;
unsigned int disable_depth:3;
unsigned int ignore_children:1;
unsigned int idle_notification:1;
Cheers!
--
---------
The views expressed in this email are personal and do not necessarily
echo my employers.
