This allows maximum flexibility for configuring the direct GPIO based
interrupts.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

If the WM831x pen down and data IRQs run in parallel it is possible for the
data and pen down IRQs to deadlock themselves as one is part way through
disabling its operation while the other is part way through enabling. Fix
this by always disabling the pen down interrupt while data is active and
vice versa.  When a changeover is required we disable the IRQ that is to
be stopped then schedule work that will enable the new IRQ.

We need to handle the data flow in the data IRQ as the readback from the
device needs to be ordered correctly with the IRQ for robust operation.

This also fixes an issue when using the built in IRQs due to enable_irq()
not being valid from interrupt context on an interrupt controller with bus
operations like the built in IRQ controller - this issue may also have
affected other interrupt controllers.  We can't rely on having the data
and pen down IRQs available via GPIOs on the CPU on every system.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Mouse gets "stuck" after restore of PV guest but buttons are in working
condition.

If driver has been configured for ABS coordinates at start it will get
XENKBD_TYPE_POS events and then suddenly after restore it'll start getting
XENKBD_TYPE_MOTION events, that will be dropped later and they won't get
into user-space.

Regression was introduced by hunk 5 and 6 of
5ea5254aa0ad269cfbd2875c973ef25ab5b5e9db
("Input: xen-kbdfront - advertise either absolute or relative
coordinates").

Driver on restore should ask xen for request-abs-pointer again if it is
available. So restore parts that did it before 5ea5254.
Acked-by: NOlaf Hering <olaf@aepfle.de>
Signed-off-by: NIgor Mammedov <imammedo@redhat.com>
[v1: Expanded the commit description]
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Calculate a default based on the number of ABS axes, REL axes,
and MT slots for the device during input device registration.
Signed-off-by: NJeff Brown <jeffbrown@android.com>
Reviewed-by: NHenrik Rydberg <rydberg@euromail.se>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Add a new EV_SYN code, SYN_DROPPED, to inform the client when input
events have been dropped from the evdev input buffer due to a
buffer overrun.  The client should use this event as a hint to
reset its state or ignore all following events until the next
packet begins.
Signed-off-by: NJeff Brown <jeffbrown@android.com>
[dtor@mail.ru: Implement Henrik's suggestion and drop old events in
 case of overflow.]
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

This commit adds the file Documentation/input/event-codes.txt.
Acked-by: NHenrik Rydberg <rydberg@euromail.se>
Reviewed-by: NPeter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: NChase Douglas <chase.douglas@canonical.com>
Signed-off-by: NChris Bagwell <chris@cnpbagwell.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Many media center remotes have buttons intended for jumping straight to
one type of media browser or another -- commonly, images/photos/pictures,
audio/music, television, and movies. At present, remotes with an images
or photos or pictures button use any number of different keycodes which
sort of maybe fit. I've seen at least KEY_MEDIA, KEY_CAMERA,
KEY_GRAPHICSEDITOR and KEY_PRESENTATION. None of those seem quite right.
In my mind, KEY_MEDIA should be something more like a media center
application launcher (and I'd like to standardize on that for things
like the windows media center button on the mce remotes). KEY_CAMERA is
used in a lot of webcams, and typically means "take a picture now".
KEY_GRAPHICSEDITOR implies an editor, not a browser. KEY_PRESENTATION
might be the closest fit here, if you think "photo slide show", but it
may well be more intended for "run application in full-screen
presentation mode" or to launch something like magicpoint, I dunno.
And thus, I'd like to have a KEY_IMAGES, which matches the HID Usage AL
Image Browser, the meaning of which I think is crystal-clear. I believe
AL Audio Browser is already covered by KEY_AUDIO, and AL Movie Browser
by KEY_VIDEO, so I'm also adding appropriate comments next to those
keys.
Signed-off-by: NJarod Wilson <jarod@redhat.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

We should first check whether platform data is NULL or not, before
dereferencing it to get the keymap.
Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Reviewed-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

In case of an error in h3600ts_connect(), deconstruct in correct order
and with the right calls.
Signed-off-by: NChristoph Fritz <chf.fritz@googlemail.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Signed-off-by: NShubhrajyoti D <a0393217@india.ti.com>
Acked-by: NAxel Lin <axel.lin@gmail.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

In rpckbd_open prror path, free_irq() was using NULL rather than the
driver data as the data pointer so free_irq() wouldn't have matched.
Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Signed-off-by: NManoj Iyer <manoj.iyer@canonical.com>
Acked-by: NPing Cheng <pingc@wacom.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

On a "really fragile" laptop I noticed a single
    i8042.c: i8042 controller selftest failed. (0x1 != 0x55)

error in the log. But there's no reason to print this message at
KERN_ERR level each time that loop fails, especially since the message
telling about the overall selftest failure is printed at KERN_INFO level
(on X86).

Add an actual error message for non-X86 systems, where a selftest
failure is (apparently) more serious. Remove a space in an another error
message.
Signed-off-by: NPaul Bolle <pebolle@tiscali.nl>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

'struct dmi_system_id' arrays must always have a terminator to keep
dmi_check_system() from looking at data (and possibly crashing) it
isn't supposed to look at.

The issue went unnoticed until ef8313bb,
but was introduced about a year earlier with
7705d548 (which also similarly changed
lifebook.c, but the problem there got eliminated shortly afterwards).

The first hunk therefore is a stable candidate back to 2.6.33, while
the full change is needed only on 2.6.38.
Signed-off-by: NJan Beulich <jbeulich@novell.com>
Cc: stable@kernel.org
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

We should return IRQ_NONE from interrupt handler in case keyboard
does not report DATA_AVAIL condition.
Signed-off-by: NRajeev Kumar <rajeev-dlh.kumar@st.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Some devices provide absolute axes with min/max of 0/0 (e.g. wacom's
ABS_MISC axis). Current uinput restrictions do not allow duplication of
these devices and require hacks in userspace to work around this.

If the kernel accepts physical devices with a min/max of 0/0, uinput
shouldn't disallow the same range.
Signed-off-by: NPeter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

This allows for debugging non-functional keys easily from
userspace.
Signed-off-by: NSeth Forshee <seth.forshee@canonical.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Scancodes are useful debugging aids when incorrect keycodes
are being sent, as is common with laptop hotkeys.
Signed-off-by: NSeth Forshee <seth.forshee@canonical.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Signed-off-by: NSylvestre Ledru <sylvestre.ledru@scilab.org>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Touch resolution is reported to the userland by retrieving the value
from the HID descriptor. But pen resolution is not since it can not
be retrieved. The current Wacom X driver has a resolution table.
To centralize the source of these values, the resolution entries are
added in the wacom_features struct for x and y coordinates respectively.
The values are then reported to the userland.
Signed-off-by: NPing Cheng <pingc@wacom.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Signed-off-by: NPing Cheng <pingc@wacom.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Commit 0b950d3d (Input: tsc2005 - add open/close) introduced a
locking issue with the ESD watchdog: __tsc2005_disable() is calling
cancel_delayed_work_sync() with mutex held, and the work also needs the
same mutex.

Fix the problem by using mutex_trylock() in tsc2005_esd_work(). If the
mutex is taken, we know we are in the middle of disable or enable and
the watchdog check can be skipped.
Signed-off-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Use relative jiffies to schedule the watchdog. Otherwise it will run
like a mad one.
Signed-off-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

drivers/input/touchscreen/tsc2005.c: In function ‘tsc2005_probe’:
drivers/input/touchscreen/tsc2005.c:666: error: implicit declaration of function ‘set_irq_wake’

In addition, migrate from set_irq_wake() (marked "do not use" as of commit
a0cd9ca2 ("genirq: Namespace cleanup")) to
irq_set_irq_wake().
Signed-off-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

I believe that enable/disable functionality should not be implemented on
the individual driver level but rather in device core, potentially
reusing parts of PM framework. Therefore the driver-specific "disable"
attribute is removed from the mainline driver.
Tested-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Introduce open and close methods for the input device to keep the device
powered down when it is not in use. Also rework interaction between
interrupt thread and starting/shutting off/resetting the device: instead
of taking a mutex in the intterrupt thread and elsewhere disable interrupts
before transitioning the device in a new state.

The ESD handling is also separated from the IRQ thread; we poll regularly
at a given interval and simply skip reads if we see that valid interrupt
happened not so long ago. This allows us not cancel and reschedule ESD
work from interrupt context all the time.
Tested-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Tested-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

We will most likely rearm it yet again the IRQ thread so doing it here
is pointless.
Tested-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

We do not need process context to send input events so let's switch to
a regular timer. I am going to get rid of taking ts->mutex in
tsc2005_irq_thread() later.
Tested-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Sparse in unhappy when people use 0 instead of NULL for pointers so
let's rework the way we initialize spi_transfer structure in
tsc2005_cmd() and tsc2005_write().
Tested-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Tested-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

If implementation to perform self-test/reset has not been provided by the
platform code hide 'selftest' sysfs attribute instead of returning error
when someone tries to use it.
Tested-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

We need to make sure we have time/work initialized before requesting and
enabling interrupts, otherwise we might start using them way too early.
Tested-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

We know what bus we are residing on (SPI) so let's make this data
available to the users.
Tested-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Set up SPI device as parent of the input device so it gets placed into
proper place in sysfs tree.
Tested-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

We should not leave garbage pointers in driver structure after we unbind
it from the device or if bind fails.
Tested-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

Add proper module description so that it would show in 'modinfo'
output.
Tested-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>

The boot process is noisy as it is and input core already announces
all new device so let's get rid of the banner message in the driver.
Tested-by: NAaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: NDmitry Torokhov <dtor@mail.ru>