Re: [PATCH 2/5] PM / Wakeirq: Add automated device wake IRQ handling

[Tony Lindgren <tony@atomide.com>]

* Rafael J. Wysocki <rjw@rjwysocki.net> [150514 14:36]:
> On Thursday, May 14, 2015 02:15:01 PM Tony Lindgren wrote:
> > * Tony Lindgren <tony@atomide.com> [150514 09:01]:
> > > * Felipe Balbi <balbi@ti.com> [150513 19:09]:
> > > > > +void dev_pm_enable_wake_irq(struct device *dev)
> > > > > +{
> > > > > +	struct wake_irq *wirq = dev->power.wakeirq;
> > > > > +
> > > > > +	if (wirq && wirq->manage_irq)
> > > > > +		enable_irq(wirq->irq);
> > > > > +}
> > > > > +EXPORT_SYMBOL_GPL(dev_pm_enable_wake_irq);
> > > > 
> > > > you probably want to enable dev_pm_enable_wake_irq() automatically for
> > > > from rpm_suspend(). According to runtime_pm documentation, wakeup should
> > > > always be enabled for runtime suspended devices. I didn't really look
> > > > through the whole thing yet to know if you did call it or not.
> > > 
> > > Yes I think we can also automate that part, I've been playing with an
> > > additional patch doing that for pm runtime. Been still thinking if
> > > there's any need to manage that in the consomer driver, I guess not.
> > 
> > Here's what that would roughly look and now I also remember why
> > I did not include it. It adds currently extra checks also for
> > devices not using dedicated wakeirqs. But basically this would
> > allow leaving out the enable/disable PM runtime calls from drivers.
...

> > @@ -1427,6 +1432,7 @@ int pm_runtime_force_suspend(struct device *dev)
> >  		goto err;
> >  	}
> >  
> > +	dev_pm_enable_wake_irq(dev);
> 
> Why here?  This is for system suspend.
> 
> >  	ret = callback(dev);
> >  	if (ret)
> >  		goto err;
> > @@ -1434,6 +1440,7 @@ int pm_runtime_force_suspend(struct device *dev)
> >  	pm_runtime_set_suspended(dev);
> >  	return 0;
> >  err:
> > +	dev_pm_disable_wake_irq(dev);
> >  	pm_runtime_enable(dev);
> >  	return ret;
> >  }
> > @@ -1463,6 +1470,7 @@ int pm_runtime_force_resume(struct device *dev)
> >  		goto out;
> >  	}
> >  
> > +	dev_pm_disable_wake_irq(dev);
> 
> This one too.

Oh you're right, seems to work fine for me without those. So it would
be the following additional patch then (only quickly tested).

Regards,

Tony

8< ----------------
--- a/drivers/base/power/runtime.c
+++ b/drivers/base/power/runtime.c
@@ -10,6 +10,7 @@
 #include <linux/sched.h>
 #include <linux/export.h>
 #include <linux/pm_runtime.h>
+#include <linux/pm_wakeirq.h>
 #include <trace/events/rpm.h>
 #include "power.h"
 
@@ -514,6 +515,7 @@ static int rpm_suspend(struct device *dev, int rpmflags)
 
 	callback = RPM_GET_CALLBACK(dev, runtime_suspend);
 
+	dev_pm_enable_wake_irq(dev);
 	retval = rpm_callback(callback, dev);
 	if (retval)
 		goto fail;
@@ -552,6 +554,7 @@ static int rpm_suspend(struct device *dev, int rpmflags)
 	return retval;
 
  fail:
+	dev_pm_disable_wake_irq(dev);
 	__update_runtime_status(dev, RPM_ACTIVE);
 	dev->power.deferred_resume = false;
 	wake_up_all(&dev->power.wait_queue);
@@ -734,10 +737,12 @@ static int rpm_resume(struct device *dev, int rpmflags)
 
 	callback = RPM_GET_CALLBACK(dev, runtime_resume);
 
+	dev_pm_disable_wake_irq(dev);
 	retval = rpm_callback(callback, dev);
 	if (retval) {
 		__update_runtime_status(dev, RPM_SUSPENDED);
 		pm_runtime_cancel_pending(dev);
+		dev_pm_enable_wake_irq(dev);
 	} else {
  no_callback:
 		__update_runtime_status(dev, RPM_ACTIVE);