linux-toradex.git/drivers/platform/x86/wmi.c, branch v3.0.87

WMI: properly cleanup devices to avoid crashes

2011-11-11T17:36:09+00:00

commit 023b9565972a4a5e0f01b9aa32680af6e9b5c388 upstream.

We need to remove devices that we destroy from the list, otherwise
we'll crash if there are more than one "_WDG" methods in DSDT.

This fixes https://bugzilla.kernel.org/show_bug.cgi?id=32052

Tested-by: Ilya Tumaykin 
Signed-off-by: Dmitry Torokhov 
Acked-by: Carlos Corbacho 
Signed-off-by: Matthew Garrett 
Signed-off-by: Greg Kroah-Hartman

wmi: Removed trailing whitespace from logging message.

2011-05-27T16:35:53+00:00

Just neatening.

Signed-off-by: Joe Perches 
Signed-off-by: Matthew Garrett

WMI: return error if wmi_create_device() fails

2011-01-07T22:03:45+00:00

The break resets the retval to 0 but we want to return an error code.
This was introduced in c64eefd48c4 "WMI: embed struct device directly
into wmi_block"

Signed-off-by: Dan Carpenter 
Signed-off-by: Matthew Garrett 
Acked-by: Dmitry Torokhov

WMI: Cater for multiple events with same GUID

2011-01-07T22:03:41+00:00

WMI data blocks can contain WMI events with the same GUID but with
different notifiy_ids, for example volume up/down hotkeys.
This patch enables a single event handler to be registered and
unregistered against all events with same GUID but different
notify_ids.  Since an event handler is passed the notify_id of
an event it can can differentiate between the different events.

The patch also ensures we only register and unregister a device per
unique GUID.

Signed-off-by: Colin Ian King 
Signed-off-by: Matthew Garrett

wmi: use memcmp instead of strncmp to compare GUIDs

2010-12-06T22:19:13+00:00

While looking for the duplicates in /sys/class/wmi/, I couldn't find
them. The code that looks for duplicates uses strncmp in a binary GUID,
which may contain zero bytes. The right function is memcmp, which is
also used in another section of wmi code.

It was finding 49142400-C6A3-40FA-BADB-8A2652834100 as a duplicate of
39142400-C6A3-40FA-BADB-8A2652834100. Since the first byte is the fourth
printed, they were found as equal by strncmp.

Signed-off-by: Thadeu Lima de Souza Cascardo 
Signed-off-by: Matthew Garrett 
Cc: stable@kernel.org

WMI: embed struct device directly into wmi_block

2010-10-21T13:36:48+00:00

Instead of creating wmi_blocks and then register corresponding devices
on a separate pass do it all in one shot, since lifetime rules for both
objects are the same. This also takes care of leaking devices when
device_create fails for one of them.

Signed-off-by: Dmitry Torokhov 
Signed-off-by: Matthew Garrett

WMI: make use of class device's attributres

2010-10-21T13:36:48+00:00

Instead of adding modalias attribute manually set it up as class's
device attribute so driver core will create and remove it for us.

Signed-off-by: Dmitry Torokhov 
Signed-off-by: Matthew Garrett

WMI: use pr_err() and friends

2010-10-21T13:36:48+00:00

This makes source more concise and easier to read.

Signed-off-by: Dmitry Torokhov 
Signed-off-by: Matthew Garrett

WMI: use separate list head for storing wmi blocks

2010-10-21T13:36:47+00:00

Do not abuse wmi_block structure to hold the head of list
of blocks, use separate list_head for that.

Signed-off-by: Dmitry Torokhov 
Signed-off-by: Matthew Garrett

WMI: simplify handling of returned WMI blocks in parse_wdg()

2010-10-21T13:36:47+00:00

There is no reason why we allocate memory and copy data into an
intermediate buffer, it is not like we are working with data coming
from userspace.

Signed-off-by: Dmitry Torokhov 
Signed-off-by: Matthew Garrett