BMC/Intel-BMC/linux.git/drivers/hid/wacom_sys.c, branch dev-4.6

HID: wacom: Support switching from vendor-defined device mode on G9 and G11

2016-04-05T15:20:32+00:00

A tablet PC booted into Windows may have its pen/touch hardware switched into "Wacom mode" similar to what we do with explicitly-supported hardware. Some devices appear to maintain this state across reboots, preventing their use with the generic HID driver. This patch adds support for detecting the presence of the mode switch feature report used by devices based on the G9 and G11 chips and has the HID codepath always attempt to reset the device back to sending standard HID reports. Fixes: https://sourceforge.net/p/linuxwacom/bugs/307/ Fixes: https://sourceforge.net/p/linuxwacom/bugs/310/ Fixes: https://github.com/linuxwacom/input-wacom/issues/15 Co-authored-by: Benjamin Tissoires Signed-off-by: Jason Gerecke Reviewed-by: Benjamin Tissoires Signed-off-by: Jiri Kosina

HID: wacom: Initialize hid_data.inputmode to -1

2016-04-05T15:20:31+00:00

Commit 5ae6e89 introduced hid_data.inputmode with a comment that it would have the value -1 if undefined, but then forgot to actually perform the initialization. Although this doesn't appear to have caused any problems in practice, it should still be remedied. Signed-off-by: Jason Gerecke Reviewed-by: Benjamin Tissoires Signed-off-by: Jiri Kosina

HID: wacom: close the wireless receiver on remove()

2016-02-16T19:40:38+00:00

rmmod/insmod the wacom.ko module does not work for the receiver because it was not previously closed. Now, we can hack with the wireless receiver without having to unplug/replug it. Signed-off-by: Benjamin Tissoires Acked-by: Ping Cheng Signed-off-by: Jiri Kosina

HID: wacom: cleanup input devices

2016-02-16T19:40:38+00:00

Just some cleaning up when the input devices are unregistered. Signed-off-by: Benjamin Tissoires Acked-by: Ping Cheng Signed-off-by: Jiri Kosina

HID: wacom: reuse wacom_parse_and_register() in wireless_work

2016-02-16T19:40:37+00:00

Removes duplicated code. The only difference is that we now need to stop and start the attached hid device, but this is a small cost. Signed-off-by: Benjamin Tissoires Acked-by: Ping Cheng Signed-off-by: Jiri Kosina

HID: wacom: move down wireless_work()

2016-02-16T19:40:37+00:00

If wireless_work() wants to reuse parse_and_register(), we need to have it declared after this function. No functional changes. Signed-off-by: Benjamin Tissoires Acked-by: Ping Cheng Signed-off-by: Jiri Kosina

HID: wacom: break out parsing of device and registering of input

2016-02-16T19:40:37+00:00

Simplifies the .probe() and will allow to reuse this path in the future. Few things are reshuffled in .probe(): - init wacom struct earlier - then retrieve the report descriptor - then parse it and allocate/register inputs. Signed-off-by: Benjamin Tissoires Acked-by: Ping Cheng Signed-off-by: Jiri Kosina

Merge branches 'for-4.4/upstream-fixes', 'for-4.5/async-suspend', 'for-4.5/container-of-cleanups', 'for-4.5/core', 'for-4.5/i2c-hid', 'for-4.5/logitech', 'for-4.5/multitouch', 'for-4.5/sony', 'for-4.5/upstream' and 'for-4.5/wacom' into for-linus

2016-01-14T15:11:06+00:00

HID: use kobj_to_dev()

2015-12-28T12:41:51+00:00

Use kobj_to_dev() instead of open-coding it. Signed-off-by: Geliang Tang Signed-off-by: Jiri Kosina

HID: use to_hid_device()

2015-12-28T12:41:44+00:00

Use to_hid_device() instead of container_of(). Signed-off-by: Geliang Tang Signed-off-by: Jiri Kosina