]> git.itanic.dy.fi Git - linux-stable/commitdiff
ACPI / PM: Make acpi_pm_device_sleep_state() follow the specification
authorRafael J. Wysocki <rjw@sisk.pl>
Tue, 29 May 2012 19:21:07 +0000 (21:21 +0200)
committerBen Hutchings <ben@decadent.org.uk>
Wed, 25 Jul 2012 03:11:45 +0000 (04:11 +0100)
commit dbe9a2edd17d843d80faf2b99f20a691c1853418 upstream.

The comparison between the system sleep state being entered
and the lowest system sleep state the given device may wake up
from in acpi_pm_device_sleep_state() is reversed, because the
specification (ACPI 5.0) says that for wakeup to work:

"The sleeping state being entered must be less than or equal to the
 power state declared in element 1 of the _PRW object."

In other words, the state returned by _PRW is the deepest
(lowest-power) system sleep state the device is capable of waking up
the system from.

Moreover, acpi_pm_device_sleep_state() also should check if the
wakeup capability is supported through ACPI, because in principle it
may be done via native PCIe PME, for example, in which case _SxW
should not be evaluated.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
drivers/acpi/sleep.c

index ca191ff978444c2fedf09aa950f8b26d327befbd..ed6bc52c82ade178f94b8d2889462fe14458de51 100644 (file)
@@ -702,8 +702,8 @@ int acpi_pm_device_sleep_state(struct device *dev, int *d_min_p)
         * can wake the system.  _S0W may be valid, too.
         */
        if (acpi_target_sleep_state == ACPI_STATE_S0 ||
-           (device_may_wakeup(dev) &&
-            adev->wakeup.sleep_state <= acpi_target_sleep_state)) {
+           (device_may_wakeup(dev) && adev->wakeup.flags.valid &&
+            adev->wakeup.sleep_state >= acpi_target_sleep_state)) {
                acpi_status status;
 
                acpi_method[3] = 'W';