diff mbox

[3/3] ACPI / PCI / PM: Rework acpi_pci_propagate_wakeup()

Message ID 4853865.5aRGCM4CdQ@aspire.rjw.lan (mailing list archive)
State New, archived
Delegated to: Bjorn Helgaas
Headers show

Commit Message

Rafael J. Wysocki July 21, 2017, 12:42 p.m. UTC
From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

The acpi_pci_propagate_wakeup() routine is there to handle cases in
which PCI bridges (or PCIe ports) are expected to signal wakeup
for devices below them, but currently it doesn't do that correctly.

The problem is that acpi_pci_propagate_wakeup() uses
acpi_pm_set_device_wakeup() for bridges and if that routine is
called for multiple times to disable wakeup for the same device,
it will disable it on the first invocation and the next calls
will have no effect (it works analogously when called to enable
wakeup, but that is not a problem).

Now, say acpi_pci_propagate_wakeup() has been called for two
different devices under the same bridge and it has called
acpi_pm_set_device_wakeup() for that bridge each time.  The
bridge is now enabled to generate wakeup signals.  Next,
suppose that one of the devices below it resumes and
acpi_pci_propagate_wakeup() is called to disable wakeup for that
device.  It will then call acpi_pm_set_device_wakeup() for the bridge
and that will effectively disable remote wakeup for all devices under
it even though some of them may still be suspended and remote wakeup
may be expected to work for them.

To address this (arguably theoretical) issue, allow
wakeup.enable_count under struct acpi_device to grow beyond 1 in
certain situations.  In particular, allow that to happen in
acpi_pci_propagate_wakeup() when wakeup is enabled or disabled
for PCI bridges, so that wakeup is actually disabled for the
bridge when all devices under it resume and not when just one
of them does that.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
 drivers/acpi/device_pm.c |   49 +++++++++++++++++++++++++++++------------------
 drivers/pci/pci-acpi.c   |    4 +--
 include/acpi/acpi_bus.h  |   14 +++++++++++--
 3 files changed, 45 insertions(+), 22 deletions(-)

Comments

Andy Shevchenko July 21, 2017, 3:45 p.m. UTC | #1
On Fri, Jul 21, 2017 at 3:42 PM, Rafael J. Wysocki <rjw@rjwysocki.net> wrote:

> The acpi_pci_propagate_wakeup() routine is there to handle cases in
> which PCI bridges (or PCIe ports) are expected to signal wakeup
> for devices below them, but currently it doesn't do that correctly.
>
> The problem is that acpi_pci_propagate_wakeup() uses
> acpi_pm_set_device_wakeup() for bridges and if that routine is
> called for multiple times to disable wakeup for the same device,
> it will disable it on the first invocation and the next calls
> will have no effect (it works analogously when called to enable
> wakeup, but that is not a problem).
>
> Now, say acpi_pci_propagate_wakeup() has been called for two
> different devices under the same bridge and it has called
> acpi_pm_set_device_wakeup() for that bridge each time.  The
> bridge is now enabled to generate wakeup signals.  Next,
> suppose that one of the devices below it resumes and
> acpi_pci_propagate_wakeup() is called to disable wakeup for that
> device.  It will then call acpi_pm_set_device_wakeup() for the bridge
> and that will effectively disable remote wakeup for all devices under
> it even though some of them may still be suspended and remote wakeup
> may be expected to work for them.
>
> To address this (arguably theoretical) issue, allow
> wakeup.enable_count under struct acpi_device to grow beyond 1 in
> certain situations.  In particular, allow that to happen in
> acpi_pci_propagate_wakeup() when wakeup is enabled or disabled
> for PCI bridges, so that wakeup is actually disabled for the
> bridge when all devices under it resume and not when just one
> of them does that.

> -       if (wakeup->enable_count > 0)
> -               goto out;
> +       if (wakeup->enable_count > 0) {
> +               if (wakeup->enable_count < max_count)
> +                       goto inc;
> +               else
> +                       goto out;
> +       }

Wouldn't be simpler

    if (wakeup->enable_count >= max_count)
      goto out;

    if (wakeup->enable_count > 0)
      goto inc;

If max_count can be <= 0,

    if (max_count > 0 && wakeup->enable_count >= max_count)
      goto out;


> +inc:
>         wakeup->enable_count++;
>
>  out:
Rafael J. Wysocki July 21, 2017, 8:44 p.m. UTC | #2
On Friday, July 21, 2017 06:45:03 PM Andy Shevchenko wrote:
> On Fri, Jul 21, 2017 at 3:42 PM, Rafael J. Wysocki <rjw@rjwysocki.net> wrote:
> 
> > The acpi_pci_propagate_wakeup() routine is there to handle cases in
> > which PCI bridges (or PCIe ports) are expected to signal wakeup
> > for devices below them, but currently it doesn't do that correctly.
> >
> > The problem is that acpi_pci_propagate_wakeup() uses
> > acpi_pm_set_device_wakeup() for bridges and if that routine is
> > called for multiple times to disable wakeup for the same device,
> > it will disable it on the first invocation and the next calls
> > will have no effect (it works analogously when called to enable
> > wakeup, but that is not a problem).
> >
> > Now, say acpi_pci_propagate_wakeup() has been called for two
> > different devices under the same bridge and it has called
> > acpi_pm_set_device_wakeup() for that bridge each time.  The
> > bridge is now enabled to generate wakeup signals.  Next,
> > suppose that one of the devices below it resumes and
> > acpi_pci_propagate_wakeup() is called to disable wakeup for that
> > device.  It will then call acpi_pm_set_device_wakeup() for the bridge
> > and that will effectively disable remote wakeup for all devices under
> > it even though some of them may still be suspended and remote wakeup
> > may be expected to work for them.
> >
> > To address this (arguably theoretical) issue, allow
> > wakeup.enable_count under struct acpi_device to grow beyond 1 in
> > certain situations.  In particular, allow that to happen in
> > acpi_pci_propagate_wakeup() when wakeup is enabled or disabled
> > for PCI bridges, so that wakeup is actually disabled for the
> > bridge when all devices under it resume and not when just one
> > of them does that.
> 
> > -       if (wakeup->enable_count > 0)
> > -               goto out;
> > +       if (wakeup->enable_count > 0) {
> > +               if (wakeup->enable_count < max_count)
> > +                       goto inc;
> > +               else
> > +                       goto out;
> > +       }
> 
> Wouldn't be simpler

I'm not really sure what you mean.

In general, ->

>     if (wakeup->enable_count >= max_count)
>       goto out;

-> this is unlikely and ->>

>     if (wakeup->enable_count > 0)
>       goto inc;

->> this isn't.

Why would checking an unlikely condition before a likely one covering it
ever be better?

> If max_count can be <= 0,

No, it can't be.

Thanks,
Rafael
Rafael J. Wysocki July 21, 2017, 9:02 p.m. UTC | #3
On Friday, July 21, 2017 10:44:30 PM Rafael J. Wysocki wrote:
> On Friday, July 21, 2017 06:45:03 PM Andy Shevchenko wrote:
> > On Fri, Jul 21, 2017 at 3:42 PM, Rafael J. Wysocki <rjw@rjwysocki.net> wrote:
> > 
> > > The acpi_pci_propagate_wakeup() routine is there to handle cases in
> > > which PCI bridges (or PCIe ports) are expected to signal wakeup
> > > for devices below them, but currently it doesn't do that correctly.
> > >
> > > The problem is that acpi_pci_propagate_wakeup() uses
> > > acpi_pm_set_device_wakeup() for bridges and if that routine is
> > > called for multiple times to disable wakeup for the same device,
> > > it will disable it on the first invocation and the next calls
> > > will have no effect (it works analogously when called to enable
> > > wakeup, but that is not a problem).
> > >
> > > Now, say acpi_pci_propagate_wakeup() has been called for two
> > > different devices under the same bridge and it has called
> > > acpi_pm_set_device_wakeup() for that bridge each time.  The
> > > bridge is now enabled to generate wakeup signals.  Next,
> > > suppose that one of the devices below it resumes and
> > > acpi_pci_propagate_wakeup() is called to disable wakeup for that
> > > device.  It will then call acpi_pm_set_device_wakeup() for the bridge
> > > and that will effectively disable remote wakeup for all devices under
> > > it even though some of them may still be suspended and remote wakeup
> > > may be expected to work for them.
> > >
> > > To address this (arguably theoretical) issue, allow
> > > wakeup.enable_count under struct acpi_device to grow beyond 1 in
> > > certain situations.  In particular, allow that to happen in
> > > acpi_pci_propagate_wakeup() when wakeup is enabled or disabled
> > > for PCI bridges, so that wakeup is actually disabled for the
> > > bridge when all devices under it resume and not when just one
> > > of them does that.
> > 
> > > -       if (wakeup->enable_count > 0)
> > > -               goto out;
> > > +       if (wakeup->enable_count > 0) {
> > > +               if (wakeup->enable_count < max_count)
> > > +                       goto inc;
> > > +               else
> > > +                       goto out;
> > > +       }
> > 
> > Wouldn't be simpler
> 
> I'm not really sure what you mean.
> 
> In general, ->
> 
> >     if (wakeup->enable_count >= max_count)
> >       goto out;
> 
> -> this is unlikely and ->>
> 
> >     if (wakeup->enable_count > 0)
> >       goto inc;
> 
> ->> this isn't.
> 
> Why would checking an unlikely condition before a likely one covering it
> ever be better?

OK, the common case is max_cout == 1 and it that case
enable_count >= max_count is equivalent to enable_count > 0,
so I guess fair enough.

Thanks,
Rafael
diff mbox

Patch

Index: linux-pm/drivers/acpi/device_pm.c
===================================================================
--- linux-pm.orig/drivers/acpi/device_pm.c
+++ linux-pm/drivers/acpi/device_pm.c
@@ -682,19 +682,8 @@  static void acpi_pm_notify_work_func(str
 
 static DEFINE_MUTEX(acpi_wakeup_lock);
 
-/**
- * acpi_device_wakeup_enable - Enable wakeup functionality for device.
- * @adev: ACPI device to enable wakeup functionality for.
- * @target_state: State the system is transitioning into.
- *
- * Enable the GPE associated with @adev so that it can generate wakeup signals
- * for the device in response to external (remote) events and enable wakeup
- * power for it.
- *
- * Callers must ensure that @adev is a valid ACPI device node before executing
- * this function.
- */
-static int acpi_device_wakeup_enable(struct acpi_device *adev, u32 target_state)
+static int __acpi_device_wakeup_enable(struct acpi_device *adev,
+				       u32 target_state, int max_count)
 {
 	struct acpi_device_wakeup *wakeup = &adev->wakeup;
 	acpi_status status;
@@ -702,8 +691,12 @@  static int acpi_device_wakeup_enable(str
 
 	mutex_lock(&acpi_wakeup_lock);
 
-	if (wakeup->enable_count > 0)
-		goto out;
+	if (wakeup->enable_count > 0) {
+		if (wakeup->enable_count < max_count)
+			goto inc;
+		else
+			goto out;
+	}
 
 	error = acpi_enable_wakeup_device_power(adev, target_state);
 	if (error)
@@ -716,6 +709,7 @@  static int acpi_device_wakeup_enable(str
 		goto out;
 	}
 
+inc:
 	wakeup->enable_count++;
 
 out:
@@ -724,6 +718,23 @@  out:
 }
 
 /**
+ * acpi_device_wakeup_enable - Enable wakeup functionality for device.
+ * @adev: ACPI device to enable wakeup functionality for.
+ * @target_state: State the system is transitioning into.
+ *
+ * Enable the GPE associated with @adev so that it can generate wakeup signals
+ * for the device in response to external (remote) events and enable wakeup
+ * power for it.
+ *
+ * Callers must ensure that @adev is a valid ACPI device node before executing
+ * this function.
+ */
+static int acpi_device_wakeup_enable(struct acpi_device *adev, u32 target_state)
+{
+	return __acpi_device_wakeup_enable(adev, target_state, 1);
+}
+
+/**
  * acpi_device_wakeup_disable - Disable wakeup functionality for device.
  * @adev: ACPI device to disable wakeup functionality for.
  *
@@ -754,8 +765,9 @@  out:
  * acpi_pm_set_device_wakeup - Enable/disable remote wakeup for given device.
  * @dev: Device to enable/disable to generate wakeup events.
  * @enable: Whether to enable or disable the wakeup functionality.
+ * @max_count: Maximum value of the enable reference counter.
  */
-int acpi_pm_set_device_wakeup(struct device *dev, bool enable)
+int __acpi_pm_set_device_wakeup(struct device *dev, bool enable, int max_count)
 {
 	struct acpi_device *adev;
 	int error;
@@ -775,13 +787,14 @@  int acpi_pm_set_device_wakeup(struct dev
 		return 0;
 	}
 
-	error = acpi_device_wakeup_enable(adev, acpi_target_system_state());
+	error = __acpi_device_wakeup_enable(adev, acpi_target_system_state(),
+					    max_count);
 	if (!error)
 		dev_dbg(dev, "Wakeup enabled by ACPI\n");
 
 	return error;
 }
-EXPORT_SYMBOL(acpi_pm_set_device_wakeup);
+EXPORT_SYMBOL(__acpi_pm_set_device_wakeup);
 
 /**
  * acpi_dev_pm_low_power - Put ACPI device into a low-power state.
Index: linux-pm/include/acpi/acpi_bus.h
===================================================================
--- linux-pm.orig/include/acpi/acpi_bus.h
+++ linux-pm/include/acpi/acpi_bus.h
@@ -605,7 +605,7 @@  acpi_status acpi_add_pm_notifier(struct
 acpi_status acpi_remove_pm_notifier(struct acpi_device *adev);
 bool acpi_pm_device_can_wakeup(struct device *dev);
 int acpi_pm_device_sleep_state(struct device *, int *, int);
-int acpi_pm_set_device_wakeup(struct device *dev, bool enable);
+int __acpi_pm_set_device_wakeup(struct device *dev, bool enable, int max_count);
 #else
 static inline void acpi_pm_wakeup_event(struct device *dev)
 {
@@ -632,12 +632,22 @@  static inline int acpi_pm_device_sleep_s
 	return (m >= ACPI_STATE_D0 && m <= ACPI_STATE_D3_COLD) ?
 		m : ACPI_STATE_D0;
 }
-static inline int acpi_pm_set_device_wakeup(struct device *dev, bool enable)
+static inline int __acpi_pm_set_device_wakeup(struct device *dev, bool enable, int max_count)
 {
 	return -ENODEV;
 }
 #endif
 
+static inline int acpi_pm_set_device_wakeup(struct device *dev, bool enable)
+{
+	return __acpi_pm_set_device_wakeup(dev, enable, 1);
+}
+
+static inline int acpi_pm_set_bridge_wakeup(struct device *dev, bool enable)
+{
+	return __acpi_pm_set_device_wakeup(dev, enable, INT_MAX);
+}
+
 #ifdef CONFIG_ACPI_SLEEP
 u32 acpi_target_system_state(void);
 #else
Index: linux-pm/drivers/pci/pci-acpi.c
===================================================================
--- linux-pm.orig/drivers/pci/pci-acpi.c
+++ linux-pm/drivers/pci/pci-acpi.c
@@ -573,7 +573,7 @@  static int acpi_pci_propagate_wakeup(str
 {
 	while (bus->parent) {
 		if (acpi_pm_device_can_wakeup(&bus->self->dev))
-			return acpi_pm_set_device_wakeup(&bus->self->dev, enable);
+			return acpi_pm_set_bridge_wakeup(&bus->self->dev, enable);
 
 		bus = bus->parent;
 	}
@@ -581,7 +581,7 @@  static int acpi_pci_propagate_wakeup(str
 	/* We have reached the root bus. */
 	if (bus->bridge) {
 		if (acpi_pm_device_can_wakeup(bus->bridge))
-			return acpi_pm_set_device_wakeup(bus->bridge, enable);
+			return acpi_pm_set_bridge_wakeup(bus->bridge, enable);
 	}
 	return 0;
 }