This simplifies error unwinding and device teardown.
Tested-by: NHeiko Stuebner <heiko@sntech.de>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Sometimes drivers need to execute one-off actions in their error handling
or device teardown paths. An example would be toggling a GPIO line to
reset the controlled device into predefined state.

To allow performing such actions when using managed resources let's allow
adding them to stack/group of devres resources.
Acked-by: NTejun Heo <tj@kernel.org>
Acked-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Add the necessary code to create the needed platformdata from devicetree
informations.

The interrupt mode of the chip is not set via devicetree, as it is not
a property of the hardware but instead only a preferred type of operation.
This should probably be made settable via configfs in the future.
The option set as default is the mode the datasheet mentions as default.
Signed-off-by: NHeiko Stuebner <heiko@sntech.de>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

When supporting devicetree the platformdata may not necessarily come
from the dev but may be generated in the driver instead.

Therefore keep the pointer in the driver struct instead of using
dev.platform_data.
Signed-off-by: NHeiko Stuebner <heiko@sntech.de>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Devicetree based platforms don't handle device callbacks very well
and until now no board has come along that needs more extended hwinit
than pulling the rst gpio high.

Therefore pull the reset handling directly into the driver and remove
the callbacks from the driver.

If extended device setup is needed at some later point, power-sequences
would probably be the solution of choice.
Signed-off-by: NHeiko Stuebner <heiko@sntech.de>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Previously the gpio was not configured at all.
Signed-off-by: NHeiko Stuebner <heiko@sntech.de>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

This enables autoloading of tsc2005 driver when is compiled as a module.
Signed-off-by: NPali Rohár <pali.rohar@gmail.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Tegra only supports, and always enables, device tree. Remove all ifdefs
and runtime checks for DT support from the driver. Platform data is
therefore no longer required. Delete the header that defines it, and
rework the driver to parse the device tree directly into struct
tegra_kbc.
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

To properly setup event parameters for emulated events, pass
the appropriate flag to the slot initialization function. Also,
all MT-related events should be setup before initialization.

Incidentally, this solves the issue of doubly filtered pointer
events.
Reported-by: NDaniel Kurtz <djkurtz@chromium.org>
Signed-off-by: NHenrik Rydberg <rydberg@euromail.se>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

The pointer emulation events are derived from contact values that
have already been filtered, so send the emulated events as is.
Reported-by: NDaniel Kurtz <djkurtz@chromium.org>
Signed-off-by: NHenrik Rydberg <rydberg@euromail.se>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Make the constants referring to range and bandwidth public so they can
be used when initializing the platform data fields in the platform code.

Fix also some comments regarding the unit of measurement to use for the
range and bandwidth fields, the values are not actually expected to be
in G or HZ, the code in bma150.c just uses the BMA150_RANGE_xxx and
BMA150_BW_xxx constants like they are with no translation from actual
values in G or HZ.
Signed-off-by: NMichael Trimarchi <michael@amarulasolutions.com>
Signed-off-by: NAntonio Ospite <ao2@amarulasolutions.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

When PM_RUNTIME is not defined, pm_runtime_get_sync() returns 1, see
include/linux/pm_runtime.c::__pm_runtime_resume(), and the check of the
return value was overlooking this, in this case bma150_open() would
return 1 which is not expected by upper layers.

Maybe the check for != -ENOSYS (Function not implemented) was meant to
cover this, but pm_runtime_get_sync() does not return this value.

For now fix the issue locally by checking explicitly for negative return
values.
Signed-off-by: NMichael Trimarchi <michael@amarulasolutions.com>
Signed-off-by: NAntonio Ospite <ao2@amarulasolutions.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Separate out the common trackstick probe/setup sequences, then call them
from each of the v3 init functions.

Credits: Emmanual Thome furnished the information on the trackstick init
and how it affected the report format.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Tested-by: NDave Turvene <dturvene@dahetral.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Rushmore touchpads are found on Dell E6230/E6430/E6530.  They use the V3
protocol with slightly tweaked init sequences and report formats.

The E7 report is 73 03 0a, and the EC report is 88 08 1d

Credits: Emmanuel Thome reported the MT bitmap changes.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Tested-by: NDave Turvene <dturvene@dahetral.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

A number of different ALPS touchpad protocols can reuse
alps_process_touchpad_packet_v3() with small tweaks to the bitfield
decoding.  Create a new priv->decode_fields() callback that handles the
per-model differences.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Tested-by: NDave Turvene <dturvene@dahetral.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Newer touchpads use different constants, so make them runtime-
configurable.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Tested-by: NDave Turvene <dturvene@dahetral.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Pinnacle class devices should return "88 07 xx" or "88 08 xx" when
entering command mode.  If either the first byte or the second byte is
invalid, return an error.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Tested-by: NDave Turvene <dturvene@dahetral.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

The official ALPS driver uses the EC report, not the E7 report, to detect
these devices.  Also, they check for a range of values; the original
table-based code only checked for two specific ones.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Tested-by: NDave Turvene <dturvene@dahetral.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

This allows alps_identify() to override these settings based on the
device characteristics, if it is ever necessary.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Tested-by: NDave Turvene <dturvene@dahetral.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

In anticipation of adding more ALPS protocols and more per-device quirks,
use function pointers instead of switch statements to call functions that
differ from one device to the next.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Tested-by: NDave Turvene <dturvene@dahetral.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

If the E6 report test passes, get the E7 and EC reports right away and
then try to match an entry in the table.

Pass in the alps_data struct, so that the detection code will be able to
set operating parameters based on information found during detection.

Change the version (psmouse->model) to report the protocol version only,
in preparation for supporting models that do not show up in the ID table.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Tested-by: NDave Turvene <dturvene@dahetral.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Several ALPS driver init sequences repeat a command three times, then
issue PSMOUSE_CMD_GETINFO to read the result.  Move this into a helper
function to simplify the code.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Tested-by: NDave Turvene <dturvene@dahetral.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

This will minimize the number of forward declarations needed when
alps_get_model() starts assigning function pointers.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Tested-by: NDave Turvene <dturvene@dahetral.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Not every type of ALPS touchpad is well-suited to table-based detection.
Start moving the various alps_model_data attributes into the alps_data
struct so that we don't need a unique table entry for every possible
permutation of protocol version, flags, byte0/mask0, etc.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Tested-by: NDave Turvene <dturvene@dahetral.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Add kernel-doc markup.
Signed-off-by: NKevin Cernekee <cernekee@gmail.com>
Tested-by: NDave Turvene <dturvene@dahetral.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

It is a pen with 10 finger touch device.
Signed-off-by: NPing Cheng <pingc@wacom.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

This patch adds support for the Cypress APA Smbus Trackpad type,
which uses a modified register map that fits within the
limitations of the smbus protocol.

Devices that use this protocol include:
CYTRA-116001-00 - Samsung Series 5 550 Chromebook trackpad
CYTRA-103002-00 - Acer C7 Chromebook trackpad
CYTRA-101003-00 - HP Pavilion 14 Chromebook trackpad
Signed-off-by: NDudley Du <dudl@cypress.com>
Signed-off-by: NBenson Leung <bleung@chromium.org>
Reviewed-by: NDaniel Kurtz <djkurtz@chromium.org>
Reviewed-by: NHenrik Rydberg <rydberg@euromail.se>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Investigating the following gesture highlighted two slight implementation
errors with choosing which slots to report in which slot when multiple
contacts are present:

Action                 SGM  AGM (MTB slot:Contact)
1. Touch contact 0    (0:0)
2. Touch contact 1    (0:0, 1:1)
3. Lift  contact 0    (1:1)
4. Touch contacts 2,3 (0:2, 1:3)

In step 4, slot 1 was not being cleared first, which means the same
tracking ID was being used for reporting both the old contact 1 and the
new contact 3.  This could result in "drumroll", where the old contact 1
would appear to suddenly jump to new finger 3 position.

Similarly, if contacts 2 & 3 are not detected at the same sample, step 4
is split into two:

Action                SGM  AGM  (MTB slot:contact)
1. Touch contact 0   (0:0)
2. Touch contact 1   (0:0, 1:1)
3. Lift  contact 0   (1:1)
4. Touch contact 2   (0:2, 1:1)
5. Touch contact 3   (0:2, 1:3)

In this case, there was also a bug.  In step 4, when contact 1 moves from
SGM to AGM and contact 2 is first reported in SGM, slot 0 was actually
empty.  So slot 0 can be used to report the new SGM (contact 0),
immediately.  Since it was empty, contact 2 in slot 0 will get a new
tracking ID.
Signed-off-by: NDaniel Kurtz <djkurtz@chromium.org>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

When removing the !S390 dependency from drivers/input/Kconfig
a couple of drivers don't compile because they have a dependency
on GENERIC_HARDIRQS.  So add the missing dependencies.
Fixes e.g. this one:

drivers/input/keyboard/lm8323.c: In function ‘lm8323_suspend’:
drivers/input/keyboard/lm8323.c:801:2: error: implicit declaration of function ‘irq_set_irq_wake’
	[-Werror=implicit-function-declaration]
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

It is time to switch to system wq instead creating a queue for the driver.
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Convert the probe to use devm_*. At the same time reorder the calls
so we will register the input device as the last step when the driver
is loaded.
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

This device is a direct pipe from "hardware" to the input event subsystem,
allowing us to avoid having to route "keypad" style events through an
AT keyboard driver (gross!).

As with the other submissions this driver is cross architecture.
Signed-off-by: NMike A. Chan <mikechan@google.com>
[Tided up to work on x86]
Signed-off-by: NSheng Yang <sheng@linux.intel.com>
Signed-off-by: NYunhong Jiang <yunhong.jiang@intel.com>
Signed-off-by: NXiaohui Xin <xiaohui.xin@intel.com>
Signed-off-by: NJun Nakajima <jun.nakajima@intel.com>
Signed-off-by: NBruce Beare <bruce.j.beare@intel.com>
[Ported to 3.4]
Signed-off-by: NTom Keel <thomas.keel@intel.com>
[Cleaned up for 3.7 and submission]
Signed-off-by: NAlan Cox <alan@linux.intel.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Postpone claiming the port until the device is opened, instead of doing
that when the driver is loaded.
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

We have been using -EBUSY where we should have used -EIO or -ENOMEM,
so let's fix that and also add some diagnostic messages.
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

This will place the joystick's input device into propoer place in sysfs
hierarchy as long as th port has device assigned to it (i.e. it is not
legacy port).
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

On resume from suspend there is a possibility for multi-byte scancodes
to be handled incorrectly. atkbd_reconnect disables the processing of
scancodes in software by calling atkbd_disable, but the keyboard may
still be active because no disconnect command was sent. Later, software
handling is re-enabled. If a multi-byte scancode sent from the keyboard
straddles the re-enable, only the latter byte(s) will be handled.

In practice, this leads to cases where multi-byte break codes (ex. "e0
4d" - break code for right-arrow) are misread as make codes ("4d" - make
code for numeric 6), leading to one or more unwanted, untyped characters
being interpreted.

The solution implemented here involves sending command f5 (reset
disable) to the keyboard prior to disabling software handling of codes.
Later, the command to re-enable the keyboard is sent only after we are
prepared to handle scancodes.
Signed-off-by: NShawn Nematbakhsh <shawnn@chromium.org>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

It is time to switch to system wq instead creating a queue for the driver.
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

This simplifies error handling and eliminates the need for implementing
remove() method.
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>