- 13 Oct, 2011 1 commit
-
-
Dan Carpenter authored
The problem here is that max_effects can wrap on 32 bits systems. We'd allocate a smaller amount of data than sizeof(struct ff_device). The call to kcalloc() on the next line would fail but it would write the NULL return outside of the memory we just allocated causing data corruption. The call path is that uinput_setup_device() get ->ff_effects_max from the user and sets the value in the ->private_data struct. From there it is: -> uinput_ioctl_handler() -> uinput_create_device() -> input_ff_create(dev, udev->ff_effects_max); I've also changed ff_effects_max so it's an unsigned int instead of a signed int as a cleanup. Signed-off-by:Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
Dmitry Torokhov <dtor@mail.ru>
-
- 12 Oct, 2011 1 commit
-
-
Philip Rakity authored
Abort driver initialization if X plate resistance was not specified in platform data as it will cause pressure to be always calculated as 0, and making userspace ignore touch coordinates. Signed-off-by:
Philip Rakity <prakity@marvell.com> Signed-off-by:
-
- 11 Oct, 2011 10 commits
-
-
Dmitry Torokhov authored
Apparently we never freed memory allocated when users open our char devices nor removed old users from the list of connected clients. Also unregister misc device immediately upon disconnecting the port instead of waiting until last user drops off (refcounting in misc device code will make sure needed pieces stay around while they are needed) and make sure we are not holing holding serio_raw_mutex when registering/unregistering misc device. This should fix potential deadlock between serio_raw and misc device code uncovered by lockdep and reported by Thomas Tuttle. Reviewed-by:
Wanlong Gao <gaowanlong@cn.fujitsu.com> Signed-off-by:
-
Dmitry Torokhov authored
Send SIGIO/POLL_HUP and otherwise wake up waiters when corresponding serio port is being disconnected. Also check if port is dead in serio_raw_poll and signal POLLHUP|POLLERR. This should speed up process of releasing dead devices by userspace applications. Reviewed-by:
-
Dmitry Torokhov authored
Instead of relying on setting serio_raw->serio to NULL upon disconnecting ports mark them explicitly as "dead". Also take and carry reference to underlying serio port to make sure it does not go away until we are done with it. Reviewed-by:
-
Dmitry Torokhov authored
This makes checkpatch.pl happy with the driver Reviewed-by:
-
Dmitry Torokhov authored
This will ensure our reporting is consistent with the rest of the system. Reviewed-by:
-
Dmitry Torokhov authored
Reviewed-by:
-
Dmitry Torokhov authored
Make sure we hold serio lock when adding clients to client list so that we do not race with serio_raw_release() removing clients from the same list. Reviewed-by:
-
Dmitry Torokhov authored
'serio_raw_list' and 'list' names do not accurately represent their objects and are extremely confusing when reading the code. Let's use better suited names. Reviewed-by:
-
Dmitry Torokhov authored
Reviewed-by:
-
Dmitry Torokhov authored
This will ensure our reporting is consistent with the rest of the system and we do not refer to obsolete source file names. Reviewed-by:
JJ Ding <dgdunix@gmail.com> Reviewed-by:
Daniel Kurtz <djkurtz@chromium.org> Signed-off-by:
-
- 10 Oct, 2011 3 commits
-
-
Chris Bagwell authored
Medium size Bamboo P&T driver reused max X/Y form older Bamboo 1 medium size tablets and never updated to real value. Actual active area of tablet is slightly larger in Y direction. Signed-off-by:
Chris Bagwell <chris@cnpbagwell.com> Acked-by:
Ping Cheng <pingc@wacom.com> Signed-off-by:
-
Chris Bagwell authored
Tablet reports a distance of 0 right at highest point possible to be in proximity and distance_max when touching tablet. Inverse the distance since user land has no way of knowing ABS_DISTANCE is not distance from tablet. Signed-off-by:
-
Chris Bagwell authored
These were left in during removal of touch pressure reports but not needed now. Signed-off-by:
-
- 07 Oct, 2011 5 commits
-
-
Dmitry Torokhov authored
CONFIG_PM is defined when CONFIG_PM_SLEEP or CONFIG_PM_RUNTIME is defined, however suspend and resume methods are only valid in the context of CONFIG_PM_SLEEP. If only CONFIG_PM_RUNTIME is defined we get the following warning (courtesy of Geerts randconfig builds): lm8323.c: warning: 'lm8323_resume' defined but not used Signed-off-by: -
Dmitry Torokhov authored
CONFIG_PM is defined when CONFIG_PM_SLEEP or CONFIG_PM_RUNTIME is defined, however suspend and resume methods are only valid in the context of CONFIG_PM_SLEEP. If only CONFIG_PM_RUNTIME is defined we get the following warning (courtesy of Geerts randconfig builds): ad7879-i2c.c: warning: 'ad7879_i2c_resume' defined but not used Signed-off-by: -
Dmitry Torokhov authored
CONFIG_PM is defined when CONFIG_PM_SLEEP or CONFIG_PM_RUNTIME is defined, however suspend and resume methods are only valid in context of CONFIG_PM_SLEEP. If only CONFIG_PM_RUNTIME is defined we get the following warning (courtesy of Geerts randconfig builds): synaptics_i2c.c: warning: 'synaptics_i2c_resume' defined but not used Signed-off-by: -
Dmitry Torokhov authored
Sometimes GCC is not smart enough to recognize that x, y and z are always used properly initialized in mma8450_poll(). Let's rearrange the code a bit to help GCC. Signed-off-by: -
Daniel Kurtz authored
Sysfs attribute show methods are always passed a buffer of length PAGE_SIZE. To keep from overwriting this buffer and causing havoc, use snprintf() to guarantee we never write more than the buffer can hold. In addition, at least for my touchscreen, the number and size of objects was far too big to fit in a single 4K page. Therefore, this patch also trims some redundant framing text to leave more room for actual data. Signed-off-by:
Nick Dyer <nick.dyer@itdev.co.uk> Signed-off-by:
-
- 05 Oct, 2011 4 commits
-
-
Ping Cheng authored
Reviewed-by:
Eduard Hasenleithner <eduard@hasenleithner.at> Tested-by:
-
Ping Cheng authored
Cintiq 21ux2 has two sets of four LEDs on right and left side of the tablet, respectively. Reviewed-by:
-
Ping Cheng authored
The LED also indicates the status of the tablet. Don't turn it off. Reviewed-by:
-
Ping Cheng authored
The LED luminance level is normally lower when no button is pressed. Reviewed-by:
-
- 28 Sep, 2011 1 commit
-
-
Sebastian Andrzej Siewior authored
This patch adds the TSC-40 serial touchscreen driver and should be compatible with TSC-10 and TSC-25. The driver was written by Linutronix on behalf of Bachmann electronic GmbH. Signed-off-by:
Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by:
Christian Gmeiner <christian.gmeiner@gmail.com> Signed-off-by:
-
- 21 Sep, 2011 10 commits
-
-
Michael Tandy authored
This patch clarifies a few bits of documentation in the header file for the adxl34x driver. Signed-off-by:
Michael Tandy <lkml@mkt.me.uk> Acked-by:
Michael Hennerich <michael.hennerich@analog.com> Signed-off-by:
-
Tobias Klauser authored
Use the of_property_read_u32() helper function to retrieve u32 values from the device tree. Also do not pass the len parameter to of_get_property if it isn't checked afterwards. Signed-off-by:
Tobias Klauser <tklauser@distanz.ch> Signed-off-by:
-
Dmitry Torokhov authored
Since touchscreen driver does not handle any events to be sent to the device we can close serio port first and then unregister the input device. Tested-by:
John Sung <penmount.touch@gmail.com> Signed-off-by:
-
Dmitry Torokhov authored
Instead of having one large switch based on product ID use pointer to function actually doing protocol decoding. Tested-by:
-
John Sung authored
Add multi touch support for PenMount 6250 touch controller. Signed-off-by:
-
John Sung authored
Add dual touch support for PenMount 3000 touch controller. Signed-off-by:
-
John Sung authored
Add support for PenMount 6000 touch controller. Signed-off-by:
-
John Sung authored
The MSB and LSB of the XY axis value are switched according to the PenMount 9000 protocol. The driver name is also changed from penmountlpc, since it is not for LPC interface at all. Signed-off-by:
-
JJ Ding authored
This essentially reverts commit f81bc788. With recent work on elantech driver, I believe we now have complete support for all elantech touchpads. So remove this hack. Signed-off-by:
JJ Ding <jj_ding@emc.com.tw> Reviewed-by:
Éric Piel <eric.piel@tremplin-utc.net> Signed-off-by:
-
JJ Ding authored
V2 hardware has many variants. This patch adddresses two issues: - some model also has debounce packets, but with a different signature than v3. Now we just check debounce for all v2 hardware. - due to different scanning methods the hardware uses, x and y ranges have to be calculated differently. And for some specific versions, we can just see them as custom-made, so set {x, y} the same values as Windows driver does. Signed-off-by:Richard Schütz <r.schtz@t-online.de> Reviewed-by:
-
- 09 Sep, 2011 5 commits
-
-
JJ Ding authored
v4 hardware is a true multitouch capable touchpad (up to 5 fingers). The packet format is quite complex, please see protocol document for reference. Signed-off-by:
-
JJ Ding authored
v3 hardware's packet format is almost identical to v2 (one/three finger touch), except when sensing two finger touch, the hardware sends 12 bytes of data. Signed-off-by:
-
JJ Ding authored
Group property setting code into elantech_set_properties. Signed-off-by:
-
JJ Ding authored
For v2 hardware, there is no real parity check, but we can still check some constant bits for data integrity. Also rename elantech_check_parity_v1 to elantech_packet_check_v1 to make these packet checking function names consistent. Signed-off-by:
-
JJ Ding authored
Don't try to be too clever and remove ETP_EDGE_FUZZ_V2. X, Y ranges should be just the raw resolution of the device. Otherwise, they can cause underflow on the Y axis. Suggested-by:
-
