sfrench/cifs-2.6.git
5 years agoMerge branch 'next' into for-linus
Dmitry Torokhov [Fri, 16 Dec 2016 17:31:17 +0000 (09:31 -0800)]
Merge branch 'next' into for-linus

Prepare input updates for 4.10 merge window.

5 years agoMerge branch 'synaptics-rmi4' into next
Dmitry Torokhov [Fri, 16 Dec 2016 05:36:09 +0000 (21:36 -0800)]
Merge branch 'synaptics-rmi4' into next

Merge updated Synaptics RMI4 support, including support for SMBus
controllers and flashing firmware.

5 years agoInput: imx6ul_tsc - generalize the averaging property
Guy Shapiro [Fri, 16 Dec 2016 05:23:02 +0000 (21:23 -0800)]
Input: imx6ul_tsc - generalize the averaging property

Make the avarage-samples property a general touchscreen property
rather than imx6ul device specific.

Signed-off-by: Guy Shapiro <guy.shapiro@mobi-wize.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: drv260x - use generic device properties
Jingkui Wang [Sun, 11 Dec 2016 06:44:10 +0000 (22:44 -0800)]
Input: drv260x - use generic device properties

Update driver drv260x to use generic device properties so that it can be
used on non-DT systems. We also remove platform data as generic device
properties work on static board code as well.

Signed-off-by: Jingkui Wang <jkwang@google.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: drv260x - use temporary for &client->dev
Dmitry Torokhov [Sun, 11 Dec 2016 07:03:33 +0000 (23:03 -0800)]
Input: drv260x - use temporary for &client->dev

Let's introduce a temporary for "client->dev" is probe() as we use
it quite a few times and "dev" is shorter.

Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: drv260x - fix input device's parent assignment
Jingkui Wang [Mon, 12 Dec 2016 21:51:46 +0000 (13:51 -0800)]
Input: drv260x - fix input device's parent assignment

We were assigning I2C bus controller instead of client as parent device.
Besides being logically wrong, it messed up with devm handling of input
device. As a result we were leaving input device and event node behind
after rmmod-ing the driver, which lead to a kernel oops if one were to
access the event node later.

Let's remove the assignment and rely on devm_input_allocate_device() to
set it up properly for us.

Signed-off-by: Jingkui Wang <jkwang@google.com>
Fixes: 7132fe4f5687 ("Input: drv260x - add TI drv260x haptics driver")
Cc: stable@vger.kernel.org
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - add support for F34 V7 bootloader
Nick Dyer [Sun, 11 Dec 2016 07:27:32 +0000 (23:27 -0800)]
Input: synaptics-rmi4 - add support for F34 V7 bootloader

Port firmware update code from Samsung Galaxy S7 driver into
mainline framework.

This patch has been tested on Synaptics S7813.

Signed-off-by: Nick Dyer <nick@shmanahar.org>
Tested-by: Chris Healy <cphealy@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: drv260x - fix initializing overdrive voltage
Dmitry Torokhov [Sun, 11 Dec 2016 06:56:21 +0000 (22:56 -0800)]
Input: drv260x - fix initializing overdrive voltage

We were accidentally initializing haptics->rated_voltage twice, and did not
initialize overdrive voltage.

Acked-by: Dan Murphy <dmurphy@ti.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: ALPS - fix protcol -> protocol
Marcos Paulo de Souza [Mon, 12 Dec 2016 18:59:56 +0000 (10:59 -0800)]
Input: ALPS - fix protcol -> protocol

Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
Reviewed-by: Pali Rohár <pali.rohar@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: i8042 - comment #else/#endif of CONFIG_PNP
Marcos Paulo de Souza [Sat, 10 Dec 2016 01:59:56 +0000 (17:59 -0800)]
Input: i8042 - comment #else/#endif of CONFIG_PNP

As this define check if huge, this makes easier to read the code.

Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: lpc32xx-keys - fix invalid error handling of a requested irq
Vladimir Zapolskiy [Thu, 8 Dec 2016 01:22:36 +0000 (17:22 -0800)]
Input: lpc32xx-keys - fix invalid error handling of a requested irq

Semantics of NR_IRQS is different on machines with SPARSE_IRQ option
disabled or enabled, in the latter case IRQs are allocated starting
at least from the value specified by NR_IRQS and going upwards, so
the check of (irq >= NR_IRQ) to decide about an error code returned by
platform_get_irq() is completely invalid, don't attempt to overrule
irq subsystem in the driver.

The change fixes lpc32xx_keys driver initialization on boot:

  lpc32xx_keys 40050000.key: failed to get platform irq
  lpc32xx_keys: probe of 40050000.key failed with error -22

Signed-off-by: Vladimir Zapolskiy <vz@mleia.com>
Acked-by: Sylvain Lemieux <slemieux.tyco@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - fix debug for sensor clip
Nick Dyer [Thu, 8 Dec 2016 01:20:06 +0000 (17:20 -0800)]
Input: synaptics-rmi4 - fix debug for sensor clip

The debug would only ever output zero for the clip information.

Signed-off-by: Nick Dyer <nick@shmanahar.org>
Reviewed-by: Andrew Duggan <aduggan@synaptics.com>
Tested-by: Chris Healy <cphealy@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - store the attn data in the driver
Benjamin Tissoires [Sat, 3 Dec 2016 01:49:10 +0000 (17:49 -0800)]
Input: synaptics-rmi4 - store the attn data in the driver

Now that we have a proper API to set the attention data, there is
no point in keeping it in the transport driver.

Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Reviewed-by: Andrew Duggan <aduggan@synaptics.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - allow to add attention data
Benjamin Tissoires [Sat, 3 Dec 2016 01:48:51 +0000 (17:48 -0800)]
Input: synaptics-rmi4 - allow to add attention data

The HID implementation of RMI4 provides the data during
the interrupt (in the input report). We need to provide
a way for this transport driver to provide the attention
data while calling an IRQ.

We use a fifo in rmi_core to not lose any incoming event.

Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Reviewed-by: Andrew Duggan <aduggan@synaptics.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - f03 - grab data passed by transport device
Dennis Wassenberg [Sat, 3 Dec 2016 01:45:29 +0000 (17:45 -0800)]
Input: synaptics-rmi4 - f03 - grab data passed by transport device

First check if there are data available passed by the transport device.
If data available use these data. If there are no data available
try to read the rmi block if dsata are passed this way.

This is the way the other rmi function handlers will do this.

This patch is needed on HID devices because the firmware reads F03 data
registers and adds them to the HID attention report. Reading those
registers from the driver after the firmware read them will result in
invalid data.

Reviewed-by: Andrew Duggan <aduggan@synaptics.com>
Signed-off-by: Dennis Wassenberg <dennis.wassenberg@secunet.com>
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - add support for F03
Lyude Paul [Sat, 3 Dec 2016 00:59:07 +0000 (16:59 -0800)]
Input: synaptics-rmi4 - add support for F03

This adds basic functionality for PS/2 passthrough on Synaptics
Touchpads using RMI4 through smbus.

Reviewed-by: Andrew Duggan <aduggan@synaptics.com>
Signed-off-by: Lyude Paul <thatslyude@gmail.com>
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: imx6ul_tsc - convert int to u32
Guy Shapiro [Wed, 30 Nov 2016 18:25:11 +0000 (10:25 -0800)]
Input: imx6ul_tsc - convert int to u32

The code uses of_property_read_u32 and expects positive values.  However,
the values are stored in signed int variables.  Additionally, the registers
values are also stored in signed variables without a good reason
(readl/writel expect u32).

The only time this caused a real bug was in the new average-samples
property, in which the property is numerically compared and implicitly
expected to be positive.

I believe it's better to change all the properties and registers to u32,
for consistency and warnings reduction.

Signed-off-by: Guy Shapiro <guy.shapiro@mobi-wize.com>
Reported-by: Bjørn Forsman <bjorn.forsman@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: imx6ul_tsc - add mask when set REG_ADC_CFG
Haibo Chen [Wed, 30 Nov 2016 17:02:06 +0000 (09:02 -0800)]
Input: imx6ul_tsc - add mask when set REG_ADC_CFG

Add mask of each function bits of REG_ADC_CFG, and clear these
function bits first, otherwise use '|=' operation may get the
wrong setting which depends on the original value of REG_ADC_CFG.

Signed-off-by: Haibo Chen <haibo.chen@nxp.com>
Reviewed-by: Guy Shapiro <guy.shapiro@mobi-wize.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - have only one struct platform data
Benjamin Tissoires [Thu, 1 Dec 2016 01:01:50 +0000 (17:01 -0800)]
Input: synaptics-rmi4 - have only one struct platform data

If struct rmi_device_platform_data contains pointers to other struct,
it gets difficult to allocate a fixed size struct and copy it over between
drivers.

Change the pointers into a struct and change the code in rmi4 accordingly.

Reviewed-by: Andrew Duggan <aduggan@synaptics.com>
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - remove EXPORT_SYMBOL_GPL for internal functions
Benjamin Tissoires [Thu, 1 Dec 2016 01:00:28 +0000 (17:00 -0800)]
Input: synaptics-rmi4 - remove EXPORT_SYMBOL_GPL for internal functions

those functions should not be used outside of rmi_core.ko.
There is no point in exporting them to the world.

It looks like rmi_read_pdt_entry() should be static too.

Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - remove mutex calls while updating the firmware
Benjamin Tissoires [Thu, 1 Dec 2016 00:59:30 +0000 (16:59 -0800)]
Input: synaptics-rmi4 - remove mutex calls while updating the firmware

This partially reverts commit 29fd0ec2bdbe ("Input: synaptics-rmi4 -
add support for F34 device reflash")

irq_mutex should be used only to protect data->current_irq_mask, not
preventing incoming input to be processed while the upgrade of the
firmware is happening. We can simply disable the irqs when we don't
want them to interfere with the upgrade process.

Tested on S7300 and S7800 (with F34 v7 patch added)

Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Nick Dyer <nick@shmanahar.org>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: drv2667 - fix misuse of regmap_update_bits
Florian Vaussard [Wed, 30 Nov 2016 01:39:25 +0000 (17:39 -0800)]
Input: drv2667 - fix misuse of regmap_update_bits

Using regmap_update_bits(..., mask, 1) with 'mask' following (1 << k)
and k greater than 0 is wrong. Indeed, _regmap_update_bits will perform
(mask & 1), which results in 0 if LSB of mask is 0. Thus the call
regmap_update_bits(..., mask, 1) is in reality equivalent to
regmap_update_bits(..., mask, 0).

In such a case, the correct use is regmap_update_bits(..., mask, mask).

This driver is performing such a mistake with the DRV2667_STANDBY mask,
which equals (1 << 6). Fix the driver to make it consistent with the
API, and fix the alignment problem at the same time. Please note that
this change is untested, as I do not have this piece of hardware.
Testers are welcome!

Signed-off-by: Florian Vaussard <florian.vaussard@heig-vd.ch>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: drv2665 - fix misuse of regmap_update_bits
Florian Vaussard [Wed, 30 Nov 2016 01:38:58 +0000 (17:38 -0800)]
Input: drv2665 - fix misuse of regmap_update_bits

Using regmap_update_bits(..., mask, 1) with 'mask' following (1 << k)
and k greater than 0 is wrong. Indeed, _regmap_update_bits will perform
(mask & 1), which results in 0 if LSB of mask is 0. Thus the call
regmap_update_bits(..., mask, 1) is in reality equivalent to
regmap_update_bits(..., mask, 0).

In such a case, the correct use is regmap_update_bits(..., mask, mask).

This driver is performing such a mistake with the DRV2665_STANDBY mask,
which equals BIT(6). Fix the driver to make it consistent with the API,
and fix the alignment problem at the same time. Please note that this
change is untested, as I do not have this piece of hardware. Testers
are welcome!

Signed-off-by: Florian Vaussard <florian.vaussard@heig-vd.ch>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - add rmi_enable/disable_irq
Benjamin Tissoires [Wed, 30 Nov 2016 01:42:13 +0000 (17:42 -0800)]
Input: synaptics-rmi4 - add rmi_enable/disable_irq

Set the .enabled boolean and trigger an event processing when enabling
for edge-triggered systems.

Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - remove unused fields in struct rmi_driver_data
Benjamin Tissoires [Wed, 30 Nov 2016 01:42:01 +0000 (17:42 -0800)]
Input: synaptics-rmi4 - remove unused fields in struct rmi_driver_data

These fields are not used anywhere, there is no point in carrying them.

Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - fix documentation of rmi_2d_sensor_platform_data
Benjamin Tissoires [Wed, 30 Nov 2016 01:41:50 +0000 (17:41 -0800)]
Input: synaptics-rmi4 - fix documentation of rmi_2d_sensor_platform_data

Typos...

Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: change KEY_DATA from 0x275 to 0x277
Ping Cheng [Mon, 28 Nov 2016 21:20:51 +0000 (13:20 -0800)]
Input: change KEY_DATA from 0x275 to 0x277

0x275 is used by KEY_FASTREVERSE.

Fixes: 488326947cd1 ("Input: add HDMI CEC specific keycodes")
Signed-off-by: Ping Cheng <ping.cheng@wacom.com>
Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Cc: stable@vger.kernel.org
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: ALPS - clean up code for SS5 hardware
Paul Donohue [Tue, 29 Nov 2016 04:16:21 +0000 (20:16 -0800)]
Input: ALPS - clean up code for SS5 hardware

The return value of alps_get_pkt_id_ss4_v2() should really be "enum
SS4_PACKET_ID", not "unsigned char".  Correct this.

Also, most of the Alps SS5 (SS4 v2) packet byte parsing code is implemented
using macros, but there are a few places where bytes are directly
manipulated in alps.c.  For consistency, migrate the rest of these to
macros.

Reviewed-by: Pali Rohár <pali.rohar@gmail.com>
Signed-off-by: Paul Donohue <linux-kernel@PaulSD.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: ALPS - clean up TrackStick handling for SS5 hardware
Paul Donohue [Tue, 29 Nov 2016 04:13:47 +0000 (20:13 -0800)]
Input: ALPS - clean up TrackStick handling for SS5 hardware

For consistency and clarity, the input_report_*() functions should be
called by alps_process_packet_ss4_v2() instead of by alps_decode_ss4_v2().

Reviewed-by: Pali Rohár <pali.rohar@gmail.com>
Signed-off-by: Paul Donohue <linux-kernel@PaulSD.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: ALPS - fix TrackStick support for SS5 hardware
Paul Donohue [Tue, 29 Nov 2016 04:11:25 +0000 (20:11 -0800)]
Input: ALPS - fix TrackStick support for SS5 hardware

The current Alps SS5 (SS4 v2) code generates bogus TouchPad events when
TrackStick packets are processed.

This causes the xorg synaptics driver to print
"unable to find touch point 0" and
"BUG: triggered 'if (priv->num_active_touches > priv->num_slots)'"
messages.  It also causes unexpected TouchPad button release and re-click
event sequences if the TrackStick is moved while holding a TouchPad
button.

This commit corrects the problem by adjusting alps_process_packet_ss4_v2()
so that it only sends TrackStick reports when processing TrackStick
packets.

Reviewed-by: Pali Rohár <pali.rohar@gmail.com>
Signed-off-by: Paul Donohue <linux-kernel@PaulSD.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: imx6ul_tsc - add support for sample averaging
Guy Shapiro [Mon, 28 Nov 2016 04:40:39 +0000 (20:40 -0800)]
Input: imx6ul_tsc - add support for sample averaging

The i.MX6UL internal touchscreen controller contains an option to
average upon samples. This feature reduces noise from the produced
touch locations.

This patch adds sample averaging support to the imx6ul_tsc device
driver.

Signed-off-by: Guy Shapiro <guy.shapiro@mobi-wize.com>
Reviewed-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: xpad - fix Xbox One rumble stopping after 2.5 secs
Cameron Gutman [Mon, 28 Nov 2016 04:37:56 +0000 (20:37 -0800)]
Input: xpad - fix Xbox One rumble stopping after 2.5 secs

Unlike previous Xbox pads, the Xbox One pad doesn't have "sticky" rumble
packets. The duration is encoded into the command and expiration is handled
by the pad firmware.

ff-memless needs pseudo-sticky behavior for rumble effects to behave
properly for long duration effects. We already specify the maximum rumble
on duration in the command packets, but it's still only good for about 2.5
seconds of rumble. This is easily reproducible running fftest's sine
vibration test.

It turns out there's a repeat count encoded in the rumble command. We can
abuse that to get the pseudo-sticky behavior needed for rumble to behave as
expected for effects with long duration.

By my math, this change should allow a single ff_effect to rumble for 10
minutes straight, which should be more than enough for most needs.

Signed-off-by: Cameron Gutman <aicommander@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: xpad - add product ID for Xbox One S pad
Cameron Gutman [Mon, 28 Nov 2016 04:37:31 +0000 (20:37 -0800)]
Input: xpad - add product ID for Xbox One S pad

This is the new gamepad that ships with the Xbox One S which
includes Bluetooth functionality.

Signed-off-by: Cameron Gutman <aicommander@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: soc_button_array - bail out earlier if gpiod_count is zero
Benjamin Tissoires [Fri, 25 Nov 2016 19:36:01 +0000 (11:36 -0800)]
Input: soc_button_array - bail out earlier if gpiod_count is zero

The PNP0C40 device of the Surface 3 doesn't have any GPIO attached to it.
Instead of trying to access the GPIO, request the count beforehand and
bail out if it is null or if an error is returned.

Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: soc_button_array - use gpio_is_valid()
Benjamin Tissoires [Fri, 25 Nov 2016 19:20:43 +0000 (11:20 -0800)]
Input: soc_button_array - use gpio_is_valid()

gpio_keys will later use gpio_is_valid(). To match the actual
behavior, we should use it here too.

Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: i8042 - fix typo from i8042_aux_close to i8042_port_close
Marcos Paulo de Souza [Wed, 23 Nov 2016 02:03:35 +0000 (18:03 -0800)]
Input: i8042 - fix typo from i8042_aux_close to i8042_port_close

Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - propagate correct number of rx and tx electrodes to F54
Guenter Roeck [Wed, 23 Nov 2016 01:57:02 +0000 (17:57 -0800)]
Input: synaptics-rmi4 - propagate correct number of rx and tx electrodes to F54

F54 diagnostics report functions provide data based on the number of
enabled rx and tx electrodes, which is not identical to the number of
electrodes reported with F54:Query0 and F54:Query1. Those values report
the number of supported electrodes, not the number of enabled electrodes.
The number of enabled electrodes can be determined by analyzing F55:Ctrl1
(sensor receiver assignment) and F55:Ctrl2 (sensor transmitter assignment).

Propagate the number of enabled electrodes from F55 to F54 to avoid
corrupted output if not all electrodes are enabled.

Fixes: 3a762dbd5347 ("[media] Input: synaptics-rmi4 - add support for F54 ...")
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Tested-by: Nick Dyer <nick@shmanahar.org>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - add support for F55 sensor tuning
Guenter Roeck [Wed, 23 Nov 2016 01:53:26 +0000 (17:53 -0800)]
Input: synaptics-rmi4 - add support for F55 sensor tuning

Sensor tuning support is needed to determine the number of enabled
tx and rx electrodes for use in F54 functions.

The number of enabled electrodes is not identical to the total number
of electrodes as reported with F55:Query0 and F55:Query1. It has to be
calculated by analyzing F55:Ctrl1 (sensor receiver assignment) and
F55:Ctrl2 (sensor transmitter assignment).

Support for additional sensor tuning functions may be added later.

Fixes: 3a762dbd5347 ("[media] Input: synaptics-rmi4 - add support for F54 ...")
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Tested-by: Nick Dyer <nick@shmanahar.org>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - add support for F34 device reflash
Nick Dyer [Wed, 23 Nov 2016 01:44:12 +0000 (17:44 -0800)]
Input: synaptics-rmi4 - add support for F34 device reflash

Add support for updating firmware, triggered by a sysfs attribute.

This patch has been tested on Synaptics S7300.

Signed-off-by: Nick Dyer <nick@shmanahar.org>
Tested-by: Chris Healy <cphealy@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: gpio_keys - set input direction explicitly
Sudeep Holla [Thu, 17 Nov 2016 01:01:26 +0000 (17:01 -0800)]
Input: gpio_keys - set input direction explicitly

Commit 700a38b27eef ("Input: gpio_keys - switch to using generic device
properties") switched to use generic device properties for GPIO keys and
commit 5feeca3c1e39 ("Input: gpio_keys - add support for GPIO descriptors")
switched from legacy GPIO numbers to GPIO descriptors.

Previously devm_gpio_request_one was explicitly passed GPIOF_DIR_IN flag
to set the GPIO direction as input. However devm_get_gpiod_from_child
doesn't have such provisions and hence fwnode_get_named_gpiod can't set
it as input.

This breaks few platforms with the following error:
" gpiochip_lock_as_irq: tried to flag a GPIO set as output for IRQ
  unable to lock HW IRQ <n> for IRQ
  genirq: Failed to request resources for POWER (irq <x>) on irqchip
  gpio_keys: Unable to claim irq <x>; error -22
  gpio-keys: probe failed with error -22 "

This patch fixes the issue by setting input direction explicitly for
gpio lines described by generic properties.

Fixes: 700a38b27eef ("Input: gpio_keys - switch to using generic device properties")
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - unlock on error
Dan Carpenter [Thu, 17 Nov 2016 01:23:22 +0000 (17:23 -0800)]
Input: synaptics-rmi4 - unlock on error

We should unlock before returning on this error path.

Fixes: 3a762dbd5347 ('[media] Input: synaptics-rmi4 - add support for F54 diagnostics')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: psmouse - disable automatic probing of BYD touchpads
Dmitry Torokhov [Sat, 12 Nov 2016 18:45:48 +0000 (10:45 -0800)]
Input: psmouse - disable automatic probing of BYD touchpads

BYD automatic protocol detection is extremely unreliable and is often
triggers false positives on regular mice, Sentelic touchpads, and other
devices. BYD has several documents that have recommended detection
sequence, but they conflict with each other and, as far as I can see, still
would not produce unique enough output to reliably differentiate BYD from
other PS/2 devices.

OEMs sourcing BYD devices also do not do us any favors by not supplying any
reasonable DMI data and instead leaving turds like "To Be Filled By O.E.M."
in place of vendor data, or "System Serial Number" as serial number.

On top of that BYD is not truly modern multitouch controller, but rather a
single-touch transitional device that only reports absolute coordinates at
the beginning of finger contact and then reverts to reporting
displacements, and thus not very precise; the only benefit from using BYD
mode vs the legacy PS/2 mode is possibility of edge scrolling.

Given the above, and the fact that BYD devices are somewhat uncommon, let's
disable automatic detection of BYD devices. Users who know they have BYD
trackpads or want to experiment can attempt to activate BYD protocol via
sysfs:

echo -n "byd" > /sys/bus/serio/devices/serio1/drvctl

Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=151691
Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=175421
Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=120781
Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=121281
Fixes: 98ee37714493 ("Input: byd - add BYD PS/2 touchpad driver")
Cc: stable@vger.kernel.org # 4.6+
Reviewed-by: Pali Rohár <pali.rohar@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: raydium_i2c_ts - fix spelling mistake in dev_err message
Colin Ian King [Thu, 17 Nov 2016 01:19:57 +0000 (17:19 -0800)]
Input: raydium_i2c_ts - fix spelling mistake in dev_err message

Trivial fix to spelling mistake "failied" to "failed" in
dev_err message.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: silead - add regulator support
Hans de Goede [Wed, 16 Nov 2016 17:37:08 +0000 (09:37 -0800)]
Input: silead - add regulator support

On some tablets the touchscreen controller is powered by separate
regulators, add support for this.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: bma150 - avoid binding to bma180 if IIO bma180 driver present
Hans de Goede [Mon, 14 Nov 2016 18:51:36 +0000 (10:51 -0800)]
Input: bma150 - avoid binding to bma180 if IIO bma180 driver present

commit ef3714fdbc8d ("Input: bma150 - extend chip detection for bma180"),
adds bma180 chip-ids to the input bma150 driver, assuming that they are
100% compatible, but the bma180 is not compatible with the bma150 at all,
it has 14 bits resolution instead of 10, and it has quite different
control registers too.

Treating the bma180 as a bma150 wrt its data registers will just result
in throwing away the lowest 4 bits, which is not too bad. But the ctrl
registers are a different story. Things happen to just work but supporting
that certainly does not make treating the bma180 the same as the bma150
right.

Since some setups depend on the evdev interface the bma150 driver offers
on top of the bma180, we cannot simply remove the bma180 ids.

So this commit only removes the bma180 id when the bma180 iio driver,
which does treat the bma180 properly, is enabled.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - add SMBus support
Benjamin Tissoires [Wed, 9 Nov 2016 01:05:58 +0000 (17:05 -0800)]
Input: synaptics-rmi4 - add SMBus support

Code obtained from https://raw.githubusercontent.com/mightybigcar/synaptics-rmi4/jf/drivers/input/rmi4/rmi_smbus.c
and updated to match upstream. And fixed to make it work.

Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Andrew Duggan <aduggan@synaptics.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - set the ABS_MT_TOOL_TYPE bit to report tool type
Andrew Duggan [Wed, 9 Nov 2016 01:03:16 +0000 (17:03 -0800)]
Input: synaptics-rmi4 - set the ABS_MT_TOOL_TYPE bit to report tool type

The rmi4 2D sensor functions report the tool type via
input_mt_report_slot_state(), but the abs parameter bit has not been
set so the tool type is not reported to userspace. This patch set
the ABS_MT_TOOL_TYPE bit.

Signed-off-by: Andrew Duggan <aduggan@synaptics.com>
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - add support for controlling dribble packets in F12
Andrew Duggan [Wed, 9 Nov 2016 00:47:58 +0000 (16:47 -0800)]
Input: synaptics-rmi4 - add support for controlling dribble packets in F12

Implements reading and setting the dribble bit in F12's control registers.

Signed-off-by: Andrew Duggan <aduggan@synaptics.com>
Acked-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - add parameters for dribble packets and palm detect gesture
Andrew Duggan [Wed, 9 Nov 2016 00:48:48 +0000 (16:48 -0800)]
Input: synaptics-rmi4 - add parameters for dribble packets and palm detect gesture

The rmi_f11 driver currently disables dribble packets and the palm detect
gesture for all devices. This patch creates a parameter in the 2d sensor
platform data for controlling this functionality on a per device basis.

For more information on dribble packets:
Commit 05ba999fcabb ("HID: rmi: disable dribble packets on Synaptics
touchpads")

For more information on the palm detect gesture:
Commit f097deef59a6 ("HID: rmi: disable palm detect gesture when present")

Signed-off-by: Andrew Duggan <aduggan@synaptics.com>
Reviewed-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - handle incomplete input data
Andrew Duggan [Wed, 9 Nov 2016 00:46:20 +0000 (16:46 -0800)]
Input: synaptics-rmi4 - handle incomplete input data

Commit 5b65c2a02966 ("HID: rmi: check sanity of the incoming report") added
support for handling incomplete HID reports do to the input data being
corrupted in transit. This patch reimplements this functionality in the
function drivers so they can handle getting less valid data then they
expect.

Signed-off-by: Andrew Duggan <aduggan@synaptics.com>
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - move IRQ handling to rmi_driver
Bjorn Andersson [Wed, 9 Nov 2016 00:34:57 +0000 (16:34 -0800)]
Input: synaptics-rmi4 - move IRQ handling to rmi_driver

The attn IRQ is related to the chip, rather than the transport, so move
all handling of interrupts to the core driver. This also makes sure that
there are no races between interrupts and availability of the resources
used by the core driver.

Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: da9062 - update bindings docs to account for da9061
Steve Twiss [Wed, 9 Nov 2016 00:15:31 +0000 (16:15 -0800)]
Input: da9062 - update bindings docs to account for da9061

Add binding information for DA9061 onkey.

This patch updates the compatible string "dlg,da9061-onkey" to support
DA9061, removes the reference to KEY_SLEEP (which the driver no longer
supports) and fixes a typo in the example for DA9063.

Supporting KEY_SLEEP was not the general convention and the typical
solution should have been for KEY_POWER to support both cases of suspend
and S/W power off. This change was sent to the DA9063 ONKEY device
driver in a separate patch, but the documentation was not updated at
that time.
f889bea Report KEY_POWER instead of KEY_SLEEP during power key-press

This patch also adds two new examples, one for DA9062 and one for DA9061.
The DA9061 examples uses a fall-back compatible string for the DA9062
onkey driver.

Signed-off-by: Steve Twiss <stwiss.opensource@diasemi.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: melfas_mip4 - add product ID reporting
Sangwon Jee [Wed, 9 Nov 2016 00:07:05 +0000 (16:07 -0800)]
Input: melfas_mip4 - add product ID reporting

Add reporting product ID through input_id.

Signed-off-by: Sangwon Jee <jeesw@melfas.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - add a couple of debug lines
Nick Dyer [Tue, 8 Nov 2016 01:36:57 +0000 (17:36 -0800)]
Input: synaptics-rmi4 - add a couple of debug lines

Signed-off-by: Nick Dyer <nick@shmanahar.org>
Tested-by: Chris Healy <cphealy@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - factor out functions from probe
Nick Dyer [Tue, 8 Nov 2016 01:35:15 +0000 (17:35 -0800)]
Input: synaptics-rmi4 - factor out functions from probe

Signed-off-by: Nick Dyer <nick@shmanahar.org>
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Tested-by: Chris Healy <cphealy@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: synaptics-rmi4 - stop scanning PDT after two empty pages
Nick Dyer [Tue, 8 Nov 2016 01:33:07 +0000 (17:33 -0800)]
Input: synaptics-rmi4 - stop scanning PDT after two empty pages

We have encountered some RMI4 firmwares where there are blank pages in
between PDT pages which contain functions. This change makes them
correctly enumerate all functions on the device.

Tested on S7817 (has empty page 2).

Signed-off-by: Nick Dyer <nick@shmanahar.org>
[Tested successfully on S7817 and S7300 Synaptics touch controllers]
Tested-by: Chris Healy <cphealy@gmail.com>
Reviewed-by: Andrew Duggan <aduggan@synaptics.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoMerge tag 'media/v4.9-3' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab...
Dmitry Torokhov [Tue, 8 Nov 2016 00:35:10 +0000 (16:35 -0800)]
Merge tag 'media/v4.9-3' of git://git./linux/kernel/git/mchehab/linux-media into synaptics-rmi4

Sync up with media tree to bring up latest Synaptics and Atmel changes.

5 years agoInput: gpio_keys - switch to using generic device properties
Dmitry Torokhov [Thu, 20 Oct 2016 02:34:48 +0000 (19:34 -0700)]
Input: gpio_keys - switch to using generic device properties

Make use of the device property API in this driver so that both OF based
systems and ACPI based systems can use this driver.

Suggested-by: Geert Uytterhoeven <geert+renesas@glider.be>
Suggested-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: gpio_keys - add support for GPIO descriptors
Geert Uytterhoeven [Wed, 19 Oct 2016 22:47:21 +0000 (15:47 -0700)]
Input: gpio_keys - add support for GPIO descriptors

GPIO descriptors are the preferred way over legacy GPIO numbers
nowadays. Convert the driver to use GPIO descriptors internally but
still allow passing legacy GPIO numbers from platform data to support
existing platforms.

Based on commits 633a21d80b4a2cd6 ("input: gpio_keys_polled: Add support
for GPIO descriptors") and 1ae5ddb6f8837558 ("Input: gpio_keys_polled -
request GPIO pin as input.").

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: gpio_keys - fix leaking DT node references
Dmitry Torokhov [Thu, 20 Oct 2016 22:11:45 +0000 (15:11 -0700)]
Input: gpio_keys - fix leaking DT node references

for_each_available_child_of_node(node, pp) takes reference to 'pp' and
drops it when attempting next iteration. However if we exit the loop early
we need to drop the reference ourselves.

Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: gpio_keys - annotate PM methods as __maybe_unused
Dmitry Torokhov [Wed, 19 Oct 2016 22:38:26 +0000 (15:38 -0700)]
Input: gpio_keys - annotate PM methods as __maybe_unused

Instead of using #ifdef, let's mark suspend and resume methods as
__maybe_unused to provide better compile coverage.

Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: gpio_keys_polled - always use gpiod_get_value_cansleep
Dmitry Torokhov [Wed, 19 Oct 2016 23:36:19 +0000 (16:36 -0700)]
Input: gpio_keys_polled - always use gpiod_get_value_cansleep

It does not matter if given GPIO may sleep or not when reading state,
polling is always done in a non-atomic context, so we should always
be able to simply use gpiod_get_value_cansleep().

Also let's note in the logs when we fail to read gpio state.

Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: gpio_keys_polled - keep button data constant
Dmitry Torokhov [Tue, 23 Feb 2016 23:32:14 +0000 (15:32 -0800)]
Input: gpio_keys_polled - keep button data constant

Commit 633a21d80b4a ("input: gpio_keys_polled: Add support for GPIO
descriptors") placed gpio descriptor into gpio_keys_button structure, which
is supposed to be part of platform data and not modifiable by the driver.
To keep the data constant, let's move the descriptor to
gpio_keys_button_data structure instead.

Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: da9063 - fix module autoload when registered via OF
Javier Martinez Canillas [Wed, 26 Oct 2016 23:46:41 +0000 (16:46 -0700)]
Input: da9063 - fix module autoload when registered via OF

If the driver is built as a module, autoload won't work because the module
alias information is not filled. So user-space can't match the registered
device with the corresponding module.

Export the module alias information using the MODULE_DEVICE_TABLE() macro.

Before this patch:

$ modinfo drivers/input/misc/da9063_onkey.ko | grep alias
alias:          platform:da9063-onkey

After this patch:

$ modinfo drivers/input/misc/da9063_onkey.ko | grep alias
alias:          platform:da9063-onkey
alias:          of:N*T*Cdlg,da9062-onkeyC*
alias:          of:N*T*Cdlg,da9062-onkey
alias:          of:N*T*Cdlg,da9063-onkeyC*
alias:          of:N*T*Cdlg,da9063-onkey

Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: fsl-imx25-tcq - fix module autoload when registered via OF
Javier Martinez Canillas [Wed, 26 Oct 2016 23:46:20 +0000 (16:46 -0700)]
Input: fsl-imx25-tcq - fix module autoload when registered via OF

If the driver is built as a module, autoload won't work because the module
alias information is not filled. So user-space can't match the registered
device with the corresponding module.

Export the module alias information using the MODULE_DEVICE_TABLE() macro.

Before this patch:

$ modinfo drivers/input/touchscreen/fsl-imx25-tcq.ko | grep alias

After this patch:

$ modinfo drivers/input/touchscreen/fsl-imx25-tcq.ko | grep alias
alias:          of:N*T*Cfsl,imx25-tcqC*
alias:          of:N*T*Cfsl,imx25-tcq

Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: elan_i2c - always output the device information
Benjamin Tissoires [Wed, 26 Oct 2016 23:26:19 +0000 (16:26 -0700)]
Input: elan_i2c - always output the device information

it's always easier to retrieve these information in bug reports when
it is always printed in the dmesg.

Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: tca8418_keypad - hide gcc-4.9 -Wmaybe-uninitialized warning
Arnd Bergmann [Wed, 26 Oct 2016 22:55:02 +0000 (15:55 -0700)]
Input: tca8418_keypad - hide gcc-4.9 -Wmaybe-uninitialized warning

Older versions of gcc warn about the tca8418_irq_handler function
as they can't keep track of the variable assignment inside of the
loop when using the -Wmaybe-unintialized flag:

drivers/input/keyboard/tca8418_keypad.c: In function ‘tca8418_irq_handler’:
drivers/input/keyboard/tca8418_keypad.c:172:9: error: ‘reg’ may be used uninitialized in this function [-Werror=maybe-uninitialized]
drivers/input/keyboard/tca8418_keypad.c:165:5: note: ‘reg’ was declared here

This is fixed in gcc-6, but it's possible to rearrange the code
in a way that avoids the warning on older compilers as well.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: melfas_mip4 - use product id for firmware name
Sangwon Jee [Wed, 26 Oct 2016 22:53:13 +0000 (15:53 -0700)]
Input: melfas_mip4 - use product id for firmware name

Use product id for firmware name to request compatible firmware.

Signed-off-by: Sangwon Jee <jeesw@melfas.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: da9061 - onkey driver
Steve Twiss [Wed, 26 Oct 2016 22:49:13 +0000 (15:49 -0700)]
Input: da9061 - onkey driver

Copyright header is updated to add DA9061 in its description and the module
description macro is extended to include DA9061.

Minor change to the code, alters dev_dbg() statements to report a generic
"PMIC" instead of DA9063. This device driver is compatible with DA9061,
DA9062 and DA9063.

Kconfig is updated to reflect support for DA9061/62/63.

Signed-off-by: Steve Twiss <stwiss.opensource@diasemi.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: pxa27x_keypad - fix typo "debpunce" -> "debounce"
Colin Ian King [Tue, 25 Oct 2016 16:35:40 +0000 (09:35 -0700)]
Input: pxa27x_keypad - fix typo "debpunce" -> "debounce"

Trivial fix to typo in dev_err message

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: psmouse - cleanup Focaltech code
Dmitry Tunin [Mon, 24 Oct 2016 23:35:09 +0000 (16:35 -0700)]
Input: psmouse - cleanup Focaltech code

psmouse->name "Focaltech Touchpad" is an overkill. In xinput it is too long
as "FocaltechPS/2 Focaltech Focaltech Touchpad"

In focaltech_report_state() pointer to psmouse->dev is already stored as
*dev

Signed-off-by: Dmitry Tunin <hanipouspilot@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: i8042 - use chassis info to skip selftest on Asus laptops
Marcos Paulo de Souza [Fri, 21 Oct 2016 22:35:33 +0000 (15:35 -0700)]
Input: i8042 - use chassis info to skip selftest on Asus laptops

Instead of relying on this model zoo let's skip selftest on all newer Asus
laptops (newer as in when they changed "Computer" -> "COMPUTER" in their
DMI data).

Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: melfas_mip4 - add product_id sysfs attribute
Sangwon Jee [Fri, 21 Oct 2016 22:28:59 +0000 (15:28 -0700)]
Input: melfas_mip4 - add product_id sysfs attribute

Add product_id sysfs attribute and update protocol version to support it.

Signed-off-by: Sangwon Jee <jeesw@melfas.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years agoInput: i8042 - add XMG C504 to keyboard reset table
Patrick Scheuring [Wed, 19 Oct 2016 19:04:02 +0000 (12:04 -0700)]
Input: i8042 - add XMG C504 to keyboard reset table

The Schenker XMG C504 is a rebranded Gigabyte P35 v2 laptop.
Therefore it also needs a keyboard reset to detect the Elantech touchpad.
Otherwise the touchpad appears to be dead.

With this patch the touchpad is detected:

$ dmesg | grep -E "(i8042|Elantech|elantech)"

[    2.675399] i8042: PNP: PS/2 Controller [PNP0303:PS2K,PNP0f13:PS2M] at 0x60,0x64 irq 1,12
[    2.680372] i8042: Attempting to reset device connected to KBD port
[    2.789037] serio: i8042 KBD port at 0x60,0x64 irq 1
[    2.791586] serio: i8042 AUX port at 0x60,0x64 irq 12
[    2.813840] input: AT Translated Set 2 keyboard as /devices/platform/i8042/serio0/input/input4
[    3.811431] psmouse serio1: elantech: assuming hardware version 4 (with firmware version 0x361f0e)
[    3.825424] psmouse serio1: elantech: Synaptics capabilities query result 0x00, 0x15, 0x0f.
[    3.839424] psmouse serio1: elantech: Elan sample query result 03, 58, 74
[    3.911349] input: ETPS/2 Elantech Touchpad as /devices/platform/i8042/serio1/input/input6

Signed-off-by: Patrick Scheuring <patrick.scheuring.dev@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
5 years ago[media] radio-bcm2048: don't ignore errors
Mauro Carvalho Chehab [Wed, 12 Oct 2016 11:21:43 +0000 (08:21 -0300)]
[media] radio-bcm2048: don't ignore errors

Remove this warning:

drivers/staging/media/bcm2048/radio-bcm2048.c: In function 'bcm2048_set_rds_no_lock':
drivers/staging/media/bcm2048/radio-bcm2048.c:467:6: warning: variable 'err' set but not used [-Wunused-but-set-variable]
  int err;
      ^~~

By returning the error code.

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] pctv452e: fix semicolon.cocci warnings
kbuild test robot [Tue, 11 Oct 2016 13:22:08 +0000 (10:22 -0300)]
[media] pctv452e: fix semicolon.cocci warnings

drivers/media/usb/dvb-usb/pctv452e.c:115:2-3: Unneeded semicolon

 Remove unneeded semicolon.

Generated by: scripts/coccinelle/misc/semicolon.cocci

Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] flexcop-usb: don't use stack for DMA
Mauro Carvalho Chehab [Mon, 10 Oct 2016 10:55:54 +0000 (07:55 -0300)]
[media] flexcop-usb: don't use stack for DMA

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

While here, remove a dead function calling usb_control_msg().

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] stk-webcam: don't use stack for DMA
Mauro Carvalho Chehab [Mon, 10 Oct 2016 14:17:15 +0000 (11:17 -0300)]
[media] stk-webcam: don't use stack for DMA

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] s2255drv: don't use stack for DMA
Mauro Carvalho Chehab [Mon, 10 Oct 2016 14:11:13 +0000 (11:11 -0300)]
[media] s2255drv: don't use stack for DMA

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] cpia2_usb: don't use stack for DMA
Mauro Carvalho Chehab [Mon, 10 Oct 2016 11:28:26 +0000 (08:28 -0300)]
[media] cpia2_usb: don't use stack for DMA

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] digitv: handle error code on RC query
Mauro Carvalho Chehab [Fri, 7 Oct 2016 16:52:31 +0000 (13:52 -0300)]
[media] digitv: handle error code on RC query

There's no sense on decoding and generating a RC key code if
there was an error on the URB control message.

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] dw2102: return error if su3000_power_ctrl() fails
Mauro Carvalho Chehab [Fri, 7 Oct 2016 16:55:53 +0000 (13:55 -0300)]
[media] dw2102: return error if su3000_power_ctrl() fails

Instead of silently ignoring the error, return it.

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] nova-t-usb2: handle error code on RC query
Mauro Carvalho Chehab [Fri, 7 Oct 2016 16:50:21 +0000 (13:50 -0300)]
[media] nova-t-usb2: handle error code on RC query

There's no sense on decoding and generating a RC key code if
there was an error on the URB control message.

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] technisat-usb2: use DMA buffers for I2C transfers
Mauro Carvalho Chehab [Fri, 7 Oct 2016 15:04:38 +0000 (12:04 -0300)]
[media] technisat-usb2: use DMA buffers for I2C transfers

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

On this driver, most of the transfers are OK, but the I2C
one was using stack.

Reviewed-by: Patrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] pctv452e: don't call BUG_ON() on non-fatal error
Mauro Carvalho Chehab [Fri, 7 Oct 2016 17:15:53 +0000 (14:15 -0300)]
[media] pctv452e: don't call BUG_ON() on non-fatal error

There are some conditions on this driver that are tested with
BUG_ON() with are not serious enough to hang a machine.

So, just return an error if this happens.

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] pctv452e: don't do DMA on stack
Mauro Carvalho Chehab [Fri, 7 Oct 2016 14:59:44 +0000 (11:59 -0300)]
[media] pctv452e: don't do DMA on stack

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

Reviewed-by: Patrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] nova-t-usb2: don't do DMA on stack
Mauro Carvalho Chehab [Fri, 7 Oct 2016 14:38:46 +0000 (11:38 -0300)]
[media] nova-t-usb2: don't do DMA on stack

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

Reviewed-by: Patrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] gp8psk: don't go past the buffer size
Mauro Carvalho Chehab [Fri, 7 Oct 2016 17:12:48 +0000 (14:12 -0300)]
[media] gp8psk: don't go past the buffer size

Add checks to avoid going out of the buffer.

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] gp8psk: don't do DMA on stack
Mauro Carvalho Chehab [Fri, 7 Oct 2016 14:24:21 +0000 (11:24 -0300)]
[media] gp8psk: don't do DMA on stack

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] dtv5100: don't do DMA on stack
Mauro Carvalho Chehab [Fri, 7 Oct 2016 11:51:23 +0000 (08:51 -0300)]
[media] dtv5100: don't do DMA on stack

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] dtt200u: handle USB control message errors
Mauro Carvalho Chehab [Fri, 7 Oct 2016 16:37:50 +0000 (13:37 -0300)]
[media] dtt200u: handle USB control message errors

If something bad happens while an USB control message is
transfered, return an error code.

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] dtt200u: don't do DMA on stack
Mauro Carvalho Chehab [Fri, 7 Oct 2016 11:40:43 +0000 (08:40 -0300)]
[media] dtt200u: don't do DMA on stack

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

Reviewed-by: Patrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] dtt200u-fe: handle errors on USB control messages
Mauro Carvalho Chehab [Fri, 7 Oct 2016 16:44:09 +0000 (13:44 -0300)]
[media] dtt200u-fe: handle errors on USB control messages

If something goes wrong, return an error code, instead of
assuming that everything went fine.

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] dtt200u-fe: don't do DMA on stack
Mauro Carvalho Chehab [Fri, 7 Oct 2016 11:06:57 +0000 (08:06 -0300)]
[media] dtt200u-fe: don't do DMA on stack

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] dtt200u-fe: don't keep waiting for lock at set_frontend()
Mauro Carvalho Chehab [Mon, 10 Oct 2016 18:34:13 +0000 (15:34 -0300)]
[media] dtt200u-fe: don't keep waiting for lock at set_frontend()

It is up to the frontend kthread to wait for lock.

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] digitv: don't do DMA on stack
Mauro Carvalho Chehab [Fri, 7 Oct 2016 10:55:38 +0000 (07:55 -0300)]
[media] digitv: don't do DMA on stack

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

Reviewed-by: Patrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] dibusb: handle error code on RC query
Mauro Carvalho Chehab [Fri, 7 Oct 2016 16:40:48 +0000 (13:40 -0300)]
[media] dibusb: handle error code on RC query

There's no sense on decoding and generating a RC key code if
there was an error on the URB control message.

Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] dibusb: don't do DMA on stack
Mauro Carvalho Chehab [Fri, 7 Oct 2016 10:34:36 +0000 (07:34 -0300)]
[media] dibusb: don't do DMA on stack

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

Reviewed-by: Patrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] dib0700_core: don't use stack on I2C reads
Mauro Carvalho Chehab [Fri, 7 Oct 2016 09:53:51 +0000 (06:53 -0300)]
[media] dib0700_core: don't use stack on I2C reads

Be sure that I2C reads won't use stack by passing
a pointer to the state buffer, that we know it was
allocated via kmalloc, instead of relying on the buffer
allocated by an I2C client.

Reviewed-by: Patrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
5 years ago[media] dib0700: be sure that dib0700_ctrl_rd() users can do DMA
Mauro Carvalho Chehab [Fri, 7 Oct 2016 09:40:24 +0000 (06:40 -0300)]
[media] dib0700: be sure that dib0700_ctrl_rd() users can do DMA

dib0700_ctrl_rd() takes a RX and a TX pointer. Be sure that
both will point to a memory allocated via kmalloc().

Reviewed-by: Patrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>