Eric Biggers [Tue, 30 May 2017 02:57:19 +0000 (19:57 -0700)]
Input: synaptics - clear device info before filling in
synaptics_query_hardware() was being passed a 'struct synaptics_device_info'
in uninitialized stack memory, then not always initializing all fields.
This caused garbage to show up in certain fields, making the touchpad
unusable.
Fix by zeroing the device info, so all fields default to 0.
Fixes: 6c53694fb222 ("Input: synaptics - split device info into a separate structure") Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Hans de Goede [Fri, 26 May 2017 23:09:48 +0000 (16:09 -0700)]
Input: silead - disable interrupt during suspend
When we put the touchscreen controller in low-power mode the irq
pin may trigger (float) and if we then try to read a data packet we
get the following error in dmesg:
[ 478.801017] silead_ts i2c-MSSL1680:00: Data read error -121
This commit disables the irq during suspend/resume fixing this error.
Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
KT Liao [Tue, 23 May 2017 20:41:47 +0000 (13:41 -0700)]
Input: elan_i2c - ignore signals when finishing updating firmware
Use wait_for_completion_timeout() instead of
wait_for_completion_interruptible_timeout() to avoid stray signals ruining
firmware update. Our timeout is only 300 msec so we are fine simply letting
it expire in case device misbehaves.
KT Liao [Thu, 25 May 2017 17:06:21 +0000 (10:06 -0700)]
Input: elan_i2c - clear INT before resetting controller
Some old touchpad FWs need to have interrupt cleared before issuing reset
command after updating firmware. We clear interrupt by attempting to read
full report from the controller, and discarding any data read.
Input: edt-ft5x06 - increase allowed data range for threshold parameter
The datasheet and application note does not mention an allowed range for
the M09_REGISTER_THRESHOLD parameter. One of our customers needs to set
lower values than 20 and they seem to work just fine on EDT EP0xx0M09 with
T5x06 touch.
So, lacking a known lower limit, we increase the range for thresholds,
and set the lower limit to 0. The documentation is updated accordingly.
Signed-off-by: Schoefegger Stefan <stefan.schoefegger@ginzinger.com> Signed-off-by: Manfred Schlaegl <manfred.schlaegl@ginzinger.com> Signed-off-by: Martin Kepplinger <martin.kepplinger@ginzinger.com> Acked-by: Rob Herring <robh@kernel.org> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
The userspace xboxdrv driver [0] contains some USB IDs unknown to the
kernel driver. I have created a simple script [1] to extract the missing
devices and add them to xpad.
A quick google search confirmed that all the new devices called
Fightstick/pad are Arcade-type devices [2] where the
MAP_TRIGGERS_TO_BUTTONS option should apply.
There are some similar devices in the existing device table where this
flag is not set, but I did refrain from changing those.
The device could as well be in command mode, in which this driver cannot
handle the device. When opening the device, let's make sure the device
will be in the mode we expect it to be for this driver.
Signed-off-by: Martin Kepplinger <martin.kepplinger@ginzinger.com> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Input: xen-kbdfront - add module parameter for setting resolution
Add a parameter for setting the resolution of xen-kbdfront in order to
be able to cope with a (virtual) frame buffer of arbitrary resolution.
While at it remove the pointless second reading of parameters from
Xenstore in the device connection phase: all parameters are available
during device probing already and that is where they should be read.
Input: xpad - don't use literal blocks inside footnotes
Unfortunately, Sphinx (or LaTeX) can't handle literal blocks inside
footnotes. So, just use normal text for the two literal code-blocks that
documents the output of /sys/kernel/debug/usb/devices for xpad devices.
Stop saying that API is experimental and that only USB is supported,
acknowledge that evdev is the preferred interface, and remove paragraph
encouraging people sending snail mail to Vojtech :) along with his email.
Input: docs - split input docs into kernel- and user-facing
Split input documentation into several groups: kernel- and user-facing, and
notes about individual device drivers. Move device drivers docs into a
separate subdirectory.
Nick Dyer [Fri, 14 Apr 2017 21:43:25 +0000 (14:43 -0700)]
Input: synaptics-rmi4 - change F12 clip to inactive border debug
The data in F12_2D_Ctrl8 corresponds to the inactive border width used by
the RMI device. It is not in pixel units and should not be treated as a
clip value.
Signed-off-by: Nick Dyer <nick@shmanahar.org> Tested-by: Chris Healy <cphealy@gmail.com> Acked-by: Benjamin Tissoires <benjamin.tissoires@redhat.com> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Nick Dyer [Fri, 14 Apr 2017 21:42:44 +0000 (14:42 -0700)]
Input: synaptics-rmi4 - use dev_driver_string when registering interrupt
When IRQ handling was moved to rmi_driver in 3aeed5b the naming of the
interrupt changed from "rmi4_i2c" to "2-0020" (or similar). This patch
restores the previous behaviour and makes the interrupt easier to identify
in /proc/interrupts.
Signed-off-by: Nick Dyer <nick@shmanahar.org> Tested-by: Chris Healy <cphealy@gmail.com> Acked-by: Benjamin Tissoires <benjamin.tissoires@redhat.com> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Input: ar1021_i2c - do not force raising edge IRQ trigger
We should not be forcing edge triggered interrupt, but rather let platform
decide the kind of trigger it needs to use. Also, the driver is not quite
safe with regard to edge-triggered interrupts as it does not try to kick
the controller after requesting/enabling IRQ.
Reviewed-By: Christian Gmeiner <christian.gmeiner@gmail.com> Tested-by: Martin Kepplinger <martin.kepplinger@ginzinger.com> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Dmitry Torokhov [Mon, 12 Dec 2016 23:32:57 +0000 (15:32 -0800)]
Input: ar1021_i2c - fix too long name in driver's device table
The name field in structure i2c_device_id is 20 characters, and we expect
it to be NULL-terminated, however we are trying to stuff it with 21 bytes
and thus NULL-terminator is lost. This causes issues when one creates
device with name "MICROCHIP_AR1021_I2C" as i2c core cuts off the last "C",
and automatic module loading by alias does not work as result.
The -I2C suffix in the device name is superfluous, we know what bus we are
dealing with, so let's drop it. Also, no other driver uses capitals, and
the manufacturer name is normally not included, except in very rare cases
of incompatible name collisions.
According to the previous error handling code, it is likely that 'goto
err_free_keymap' is expected here in order to avoid a memory leak in error
handling path.
There are several quirky Xbox One pads that depend on initialization
packets that the Microsoft pads don't require. To deal with these,
I've added a mechanism for issuing device-specific initialization
packets using a VID/PID-based quirks list.
For the initial set of init quirks, I have added quirk handling from
Valve's Steam Link xpad driver[0] and the 360Controller project[1] for
macOS to enable some new pads to work properly.
This should enable full functionality on the following quirky pads:
0x0e6f:0x0165 - Titanfall 2 gamepad (previously fully non-functional)
0x0f0d:0x0067 - Hori Horipad (analog sticks previously non-functional)
0x24c6:0x541a - PowerA Xbox One pad (previously fully non-functional)
0x24c6:0x542a - PowerA Xbox One pad (previously fully non-functional)
0x24c6:0x543a - PowerA Xbox One pad (previously fully non-functional)
Input: yealink - define packet offset __be16 instead of u16
sparse says
warning: incorrect type in assignment (different base types)
expected unsigned short [unsigned] [usertype] offset
got restricted __be16 [usertype] <noident>
for every usage of cpu_to_be16 in yealink.c. Defining it __be16 in the
first place shouldn't hurt.
Signed-off-by: Martin Kepplinger <martin.kepplinger@ginzinger.com> Signed-off-by: Henk.Vergonet@gmail.com Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Hans de Goede [Tue, 11 Apr 2017 03:34:19 +0000 (20:34 -0700)]
Input: soc_button_array - properly map usage 0x07/0xe3 to KEY_LEFTMETA
When submitting the support for the ACPI0011 windows tablet keys device I
mapped the "windows" logo homekey to KEY_HOMEPAGE. But this is inconsistent
with how it is done on windows tablets using the old PNP0C40 ACPI device
and it does not match the HUT spec, which says that usage-page 7 usage 0xe3
is "Keyboard Left GUI".
This commit maps usage-page 7 usage 0xe3 to KEY_LEFTMETA fixing this.
Fixes: 4c3362f44980 ("Input: soc_button_array - add support for ACPI 6.0...") Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Input: use svg files instead of xfig in force feedback documentation
Now that the images got converted to svg, add them at the
ff.rst file.
NOTE: After merging upstream, the "figure" tag should be
replaced by the new "kernel-figure" tag, in order for the
SVG images to be included at the PDF files.
Input: iforce - convert documentation into ReST format
This file seems to be using some other markup language, (maybe mediawiki?).
Manually convert it to ReST markup, with is the one we're using along the
Kernel documentaiton.
Input: pm8xxx-vib - handle separate enable register
Some PMIC vibrator IPs use a separate enable register to turn the
vibrator on and off. To detect if a vibrator uses this feature, rely on
the enable_mask being non-zero.
In order to prepare this driver to support other vibrators of the same
kind, move some hardcoded values to a structure holding register
parameters (address, mask, shit of the control register).
Dmitry Torokhov [Mon, 20 Feb 2017 01:21:56 +0000 (17:21 -0800)]
Input: eeti_ts - switch to gpiod API
gpiod API allows standard way of specifying GPIO polarity and takes it into
account when reading or setting GPIO state. It also allows us to switch to
common way of obtaining GPIO descriptor and away form legacy platform data.
Reviewed-by: Daniel Mack <daniel@zonque.org> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Dmitry Torokhov [Mon, 20 Feb 2017 01:23:47 +0000 (17:23 -0800)]
Input: eeti_ts - switch to using threaded interrupt
Instead of having standard interrupt handler and manually firing work item
to perform I2C reads, let's switch to threaded interrupts, which were
designed specifically for this purpose.
Reviewed-by: Daniel Mack <daniel@zonque.org> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Dmitry Torokhov [Tue, 21 Feb 2017 07:13:45 +0000 (23:13 -0800)]
Input: eeti_ts - use gpio_get_value_cansleep
We are reading GPIO state in a non-atomic context (workqueue), so we can
use "cansleep" variant, and thus make the driver available on systems where
GPIO controllers require sleeping when reading GPIO state.
Reviewed-by: Daniel Mack <daniel@zonque.org> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Dmitry Torokhov [Tue, 21 Feb 2017 05:50:13 +0000 (21:50 -0800)]
Input: eeti_ts - use get_unaligned_be16() to retrieve data
Instead of manually converting big endian data on wire into host
endianness, let's use helpers to do that for us. It might save us
a few cycles if host endianness matches what's on wire.
Reviewed-by: Daniel Mack <daniel@zonque.org> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
projects / mirror_ubuntu-eoan-kernel.git / log