The default multitouch protocol class in use since the kernel v3.9 is
working quite well. Since its inclusion, the only devices we had to tweak
were those who really need quirks (GeneralTouch, FocalTech and Wistron,
the 3 of them are Win 7 certified ones).
The flow of new unhandled devices has stopped, which is great and I think
it's time to reduce the list of registered device.

This commit removes only the registration in the kernel of devices that
use the class MT_CLS_DEFAULT, or that can use it. By that, I mean that I
checked all the recordings I have, and the produced input device and
events are the same before and after applying the patch.

This gives two benefits:
- remove a bunch of lines of codes
- prevent bad handling of existing registered devices which are using a
different protocol while using the same VID/PID (I got the case of a
Quanta 3008 recently).

I also removed the associated classes (MT_CLS*). I kept their #define in
case people use the new_id sysfs node with a non standard class (their
should be really few people now, but we never now). This is why there
are /* reserved .... */.

Last, I add a comment on top of mt_devices[] definition to remember people
(and myself) not to include devices for the beauty of it.

To people still trying to add devices with the default class:
"""
Guys, try out your device under a kernel greater or equal to v3.9. If it
works, you are all set. Adding your VID/PID to the kernel only brings us
overload and you won't get anything from it _because_ even a backport of
this shiny patch will _not_ make the device work under 3.0, 3.2, 3.4 or
even 3.8.
So if it works, it works.
If it does not work, then yes, submit a patch or call for help.
In any cases, if you want me to do regression tests, I'd be happy to
get some traces of your device. But I won't patch the kernel if it works.
"""
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

[jkosina@suse.cz: refresh to apply after SIS quirk merging]
Signed-off-by: NEmanuel Krenz <emanuelkrenz@web.de>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Since commit 765e5fbd merged the sis quirk,
then USB_VENDOR_ID_SIS2_TOUCH remains undefined.
Reported-by: NFengguang Wu <fengguang.wu@intel.com>
Signed-off-by: NWanlong Gao <gaowanlong@cn.fujitsu.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This patch adds PID VID to support for the Wistron Inc. Optical touch panel.
Signed-off-by: NKaiChung Cheng <kenny_cheng@wistron.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Add support for SiS multitouch panels.
Signed-off-by: NForest Bond <forest.bond@rapidrollout.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

GeneralTouch products should use the quirk SLOT_IS_CONTACTID
instead of SLOT_IS_CONTACTNUMBER.

Adding PIDs 0101,e100,0102,0106,010a from the new products.

Tested on new and older products by GeneralTouch engineers.
Signed-off-by: NLuosong <android@generaltouch.com>
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

When working on report indexes, always validate that they are in bounds.
Without this, a HID device could report a malicious feature report that
could trick the driver into a heap overflow:

[  634.885003] usb 1-1: New USB device found, idVendor=0596, idProduct=0500
...
[  676.469629] BUG kmalloc-192 (Tainted: G        W   ): Redzone overwritten

Note that we need to change the indexes from s8 to s16 as they can
be between -1 and 255.

CVE-2013-2897

Cc: stable@vger.kernel.org
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Acked-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Some multitouch screens do not like to be polled for input reports.
However, the Win8 spec says that all touches should be sent during
each report, making the initialization of reports unnecessary.
The Win7 spec is less precise, so do not use this for those devices.

Add the quirk HID_QUIRK_NO_INIT_INPUT_REPORTS so that we do not have to
introduce a quirk for each problematic device. This quirk makes the driver
behave the same way the Win 8 does. It actually retrieves the features,
but not the inputs.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Reviewed-by: NHenrik Rydberg <rydberg@euromail.se>
Tested-by: Srinivas Pandruvada<srinivas.pandruvada@linux.intel.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Detecting Win 8 multitouch devices in core allows us to set quirks
before the device is parsed through hid_hw_start().
It also simplifies the detection of those devices in hid-multitouch and
makes the handling of those devices cleaner.

As Win 8 multitouch panels are in the group multitouch and rely on a
special feature to be detected, this patch adds a bitfield in the parser.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Reviewed-by: NHenrik Rydberg <rydberg@euromail.se>
Tested-by: Srinivas Pandruvada<srinivas.pandruvada@linux.intel.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

HID special drivers can use safely the devres API.
Use it to remove 25 lines of code and to clean up a little the error paths.

Besides the basic kzalloc -> devm_kzalloc conversions, I changed the
place of the allocation of the new name. Doing this right in
mt_input_configured() removes the kstrdup call which was not very helpful
and the new way is simpler to understand (and to debug).
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

mt_free_input_name() was never called during .remove():
hid_hw_stop() removes the hid_input items in hdev->inputs, and so the
list is therefore empty after the call. In the end, we never free the
special names that has been allocated during .probe().

Restore the original name before freeing it to avoid acessing already
freed pointer.

This fixes a regression introduced by 49a5a827 ("HID: multitouch: append " Pen" to
the name of the stylus input")
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This patch also removes the duplicate entry in the hid-multitouch.c as
suggested by Benjamin Tissoires.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Just sort the list by IDs. There is no functional change.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-By: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Add vendor ID for Data Modul and their easyMaxTouch device.
The device has to be configured to multitouch mode prior to using this driver.
Signed-off-by: NSteffen Trumtrar <s.trumtrar@pengutronix.de>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Commit "HID: multitouch: use the callback "report" instead..." breaks the
buttons of touchpads following the HID multitouch specification.
The buttons were emmitted through hid-input, but as now the events
are generated only in hid-multitouch, the buttons are not emmitted anymore.

The input_event() call is far much simpler than the hid-input one as
many of the different tests do not apply to multitouch touchpads.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

The "tablet" udev rule relies on BTN_STYLUS to be set.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This is not just cosmetics, it can help to write udev and X.org
rules.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Dual sensors devices reports pen and touch on two different reports.
Using the quirk HID_QUIRK_MULTI_INPUT allows us to create a new input
device to forward pen events.

The quirk HID_QUIRK_NO_EMPTY_INPUT avoids the creation of input devices
for the not used mouse emulation present on Win7 certified devices.

Since hid-multitouch sets the quirk HID_QUIRK_NO_INPUT_SYNC, we need
to manually send SYN events for pen report too.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

To implement different methods for pen and touch, the previous
implementation has to be reworked.

This detection of the input attached to the touch sensor is the same
than the one used in mt_report().
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

hid-multitouch only handles touch events, so there is no point in
mapping other kind of events.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This will allow easier integration of hybrid pen and touch devices.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

hid-multitouch used to direclty call for a set_idle in the .resume()
callback. With the new hid_hw_idle(), this can be dropped and
hid-multitouch is now freed from its transport layer.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Reviewed-by: NDavid Herrmann <dh.herrmann@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

The aim of last_field_index was to detect the end of the report.
With the introduction of .report(), it is not required anymore to
detect it on the fly as we can put it in the right place during the
.report().

The resulting code path is simpler to read because the terminating
condition is not evaluated after each field.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This reflects my new company and officialize the regression tests I'm
conducting between each change.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Acked-by: NHenrik Rydberg <rydberg@euromail.se>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This patch is part of the cleanup of the HID drivers
against their low-level transport layer.
With new touchscreens relying on I2C, it's better to
handle now these dependencies before users get kernel oopses.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This allows the hid drivers to be independent from the transport layer.

The patch was constructed by replacing all occurences of
usbhid_submit_report() by its hid_hw_request() counterpart.
Then, drivers not requiring USB_HID anymore have their USB_HID
dependency cleaned in the Kconfig file.

Finally, few drivers still depends on USB_HID. Many of them
are requiring the io wait callback. They are found in the next patch.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>

For the sensor-hub part:
Tested-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

The previous implementation registered a pointer towards hid-core
to the value of contact count. This is not safe and may be difficult
to debug if hid-core ever changes its implementation.
The use of regular indexes is a better choice.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

By running a test on all the traces of the devices I have,
I noticed that the class MT_CLS_ALWAYS_TRUE could handle all
the devices I've seen so far without any other quirks.
I guess this is the behavior Win 7 requires in its driver.

We can change the default class then and keep the existing classes
for backward compatibility and performances for some of them.

Two operations have been done:
 * replaced MT_CLS_DEFAULT by MT_CLS_NSMU
 * then replaced MT_CLS_ALWAYS_TRUE by MT_CLS_DEFAULT
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

The previous protocol was nearly working, but when several fingers were
present on the sensor, those that were not moving were not updated
in the next report, introducing a lot of releases.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>

Conflicts:
	drivers/hid/hid-multitouch.c
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Cando 2087:0a02 was broken, this fixes it.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Since the inclusion of this device in hid-multitouch, the device
did not forward any events. Using the serial class makes it working
again.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

The serial protocol makes contact count a redondant information, and
sometimes it is not reliable (TRS-Star are in this case).
Disabling the use of contact count for these devices is thus safer.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

The Sharp LC-20FE1-W screen (04dd:9681) behaves like the Nexio 42".
It may report out of ranges values that are filtered out by relying
on the Contact Count HID field.
Adding the quirk MT_QUIRK_CONTACT_CNT_ACCURATE makes hid-multitouch
strongest against this kind of device, without breaking the current
devices.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This device is the worst device I saw. It keeps TipSwitch and InRange
at 1 for fingers that are not touching the panel.
The solution is to rely on the field ContactCount, which is accurate
as the correct information are packed at the begining of the frame.

Unfortunately, CountactCount is most of the time at the end of the report.
The solution is to pick it when we have the whole report in raw_event.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Nexio 42" devices requires to rely on the HID field Contact
Count to compute the valid values. However, this field is
most of the time at the end of the report, meaning that we
need to get the all report parsed before processing it.

This patch does not introduce functional changes.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Use the new module_hid_driver macro in all HID drivers that have
a simple register/unregister init/exit.

This also converts the hid drivers that test for a failure of
hid_register_driver() and report the failure. Using module_hid_driver
in those drivers removes the failure message.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

The Win 8 protocol specify the fact that each valid touch must be reported
within a frame until it is released.
We can therefore use the always_valid quirk and dismiss reports when we see
duplicate contacts ID.

We recognize Win8 certified devices from their vendor feature 0xff0000c5
where Microsoft put a signed blob in the report to check if the device
passed the certification.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Reviewed-by: NHenrik Rydberg <rydberg@euromail.se>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Win8 devices supporting hovering must provides InRange HID field.
The information that the finger is here but is not touching the surface
is sent to the user space through ABS_MT_DISTANCE as required by the
multitouch protocol.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Reviewed-by: NHenrik Rydberg <rydberg@euromail.se>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This quirk allows a device to reuse a contact id when sending garbage
inactive contacts at the end of a report.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Reviewed-by: NHenrik Rydberg <rydberg@euromail.se>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Win 8 device specification changed the requirements for the hid usages
of the multitouch devices. Now InRange is optional and must be only
used when the device supports hovering.

This ensures that the quirk ALWAYS_VALID is taken into account and
also ensures its precedence over the other VALID* quirks.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@gmail.com>
Reviewed-by: NHenrik Rydberg <rydberg@euromail.se>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>