[PATCH 0/13] ACPI / PM: Rework power resources management

From: Rafael J. Wysocki
Date: Wed Nov 24 2010 - 18:19:07 EST


Hi,

Unfortunately there are a few problems with ACPI power resources management
that should be fixed by the following series of patches.

Patch [1/13] is necessary to fix a recent regression in the mainline kernel and
patch [2/13] also should go into 2.6.37 IMO. The remaining patches are not so
urgent, but they all are necessary.

[1/13] - Fix regression if fan resume resulting from the rework of power
resources reference counting.

[2/13] - Do not reference count power resources that can't be turned on.

[3/13] - Rework acpi_power_get_inferred_state() so that it doesn't update
device->power.state behind the caller's back.

[4/13] - Introduce functions for handling lists of power resources.

[5/13] - Introduce helper function for reference counting power resources
for a given device and power state.

[6/13] - Introduce function for reading device power state without modifying
the device object.

[7/13] - Add function for device power state initialization.

[8/13] - Add function for updating device power state in a consistent way.

[9/13] - Register acpi_power_driver before scanning the root of the namespace.

[10/13] - Add power resource ACPI device objects as soon as they are needed.

[11/13] - Rework the management of ACPI power resources in the ACPI fan driver.

[12/13] - Drop acpi_bus_get_power() that has no users now.

[13/13] - Drop acpi_power_nocheck which isn't necessary any more.

The patches have been tested on my HP nx6325 (which is the only machine using
ACPI power resources I have) and a combo patch functionally equivalen to this
series has been tested by Maciej Rutecki.

Please apply.

Thanks,
Rafael

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/