Per-controller spinlock needs to be properly initialized during device probe.

[jkosina@suse.cz: massage changelog]
[jkosina@suse.cz: drop hunk that has already been applied by previous
 patch]
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

sony_dev_list_lock spinlock (which was introduced in d2d782fc ("HID: sony:
Prevent duplicate controller connections") is not being initialized properly.
Fix that.
Reported-by: NPavel Machek <pavel@ucw.cz>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

ida_destroy() must be called _after_ all the devices have been
unregistered; otherwise, when calling "rmmod hid_sony" with devices
still plugged in, the following warning would show up because of calls
to ida_simple_remove() on a destroyed ID allocator:

------------[ cut here ]------------
WARNING: CPU: 0 PID: 5509 at lib/idr.c:1052 ida_simple_remove+0x26/0x50()
ida_remove called for id=0 which is not allocated.
Modules linked in: ...
CPU: 0 PID: 5509 Comm: rmmod Not tainted 3.19.0-rc6-ao2 #35
Hardware name: System manufacturer System Product Name/M2N-MX SE, BIOS 0501    03/20/2008
 0000000000000000 ffffffff8176320d ffffffff815b3a88 ffff880036f7fdd8
 ffffffff8106ce01 0000000000000000 ffffffffa07658e0 0000000000000246
 ffff88005077d8b8 ffff88005077d8d0 ffffffff8106ce7a ffffffff81763260
Call Trace:
 [<ffffffff815b3a88>] ? dump_stack+0x40/0x50
 [<ffffffff8106ce01>] ? warn_slowpath_common+0x81/0xb0
 [<ffffffff8106ce7a>] ? warn_slowpath_fmt+0x4a/0x50
 [<ffffffff812ccb86>] ? ida_simple_remove+0x26/0x50
 [<ffffffffa0762dc8>] ? sony_remove+0x58/0xe0 [hid_sony]
 [<ffffffffa00fff15>] ? hid_device_remove+0x65/0xd0 [hid]
 [<ffffffff8140425e>] ? __device_release_driver+0x7e/0x100
 [<ffffffff81404c70>] ? driver_detach+0xa0/0xb0
 [<ffffffff81403ee5>] ? bus_remove_driver+0x55/0xe0
 [<ffffffffa01000ff>] ? hid_unregister_driver+0x2f/0xa0 [hid]
 [<ffffffff810e45bf>] ? SyS_delete_module+0x1bf/0x270
 [<ffffffff81014089>] ? do_notify_resume+0x69/0xa0
 [<ffffffff815b952d>] ? system_call_fastpath+0x16/0x1b
---[ end trace bc794b3d22c30ede ]---
Signed-off-by: NAntonio Ospite <ao2@ao2.it>
Acked-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Add vid/pid for the SMK branded third-party PS3 Bluetooth remote and enable
support in the hid-sony driver.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Replace stack buffers with kernel allocated buffers for sending
and receiving HID reports to prevent issues with DMA transfers
on certain hardware.

Output report buffers are allocated at initialization time to avoid
excessive calls to kmalloc and kfree.

Link: https://bugzilla.kernel.org/show_bug.cgi?id=87991Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Reviewed-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

The DualShock 4 touchpad has been measured to have a resolution of
44.86 dots/mm which equates to 1920x942.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Set the DualShock4 touchpad bits in the input_configured callback
so that they are registered properly for any input devices created
during hid_hw_start.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Update the file header and correct an outdated comment block.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Fix a few minor issues in the HID descriptor:
 - A 6 bit entry had a logical maximum of 255 when the largest it can be is 63.
 - A logical max value was incorrectly being set to -1 instead of 255.
 - Set the min/max of the gyroscopes to -8192/8191 as that is the range of
   values which represent the true controller orientation.  Any values beyond
   those extents are just noise.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Bit 3 in byte 31 of the Sixaxis report indicates whether the battery is
charging or not charging as opposed to whether or not the cable is plugged in.
As a result, when connected via USB and fully charged, the power_supply status
is wrongly reported as 'Discharging' instead of 'Full'.

Use the battery level value to set the cable state so that the power status
is reported correctly as that seems to be the only reliable way to determine the
cable status on the Sixaxis.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Previously only the first element of the array was initialized to 1
leading to potential incorrect max brightness values for the LEDs
on the Dualshock 3 and buzzer controllers.

Use a designated initializer to initialize the whole array to the
correct value.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Remove an old redundant comment before sony_report_fixup(), it must have
been a leftover from the first version of the driver:
sony_report_fixup() now handles all the supported devices, not only the
Sony Vaio VGX.

The comment is also redundant as the same information provided by it is
also present in the body of the function.
Signed-off-by: NAntonio Ospite <ao2@ao2.it>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Reformat sixaxis_rdesc to reflect its HID structure, and comment each
field. This will make it easier to validate changes to the descriptor in
the future.

No functional changes are introduced, the descriptor is exactly the same
as before byte by byte.

The heavy lifting has been done with the help of hidrd-convert:
https://github.com/DIGImend/hidrdSigned-off-by: NAntonio Ospite <ao2@ao2.it>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Simplify the logic of overriding the Sixaxis HID descriptor, this will
make it easier to amend the descriptor in future commits.

The current code does this:

	if (original sixaxis via USB or BT)
		fixup only a part of the descriptor
	else if (sixaxis compatible controller)
		override the _whole_ descriptor

but the end result is exactly the same, except for the trailing zero in
the case of original BT controllers.

So let's just regularize the process, and always override the HID
descriptor.

Always overriding the descriptor changes the current semantic a little
bit, before this change the BT descriptor still had the trailing zero
byte, while now it is exactly the same as the descriptor of the
controller via USB, but that does not affect proper operation of the
device.

Note that overriding the whole descriptor for original devices is not
strictly necessary for now, but it simplifies the code and in the future
the report descriptor will be patched further and keys will be remapped,
so it's handy to have only one place to patch.
Signed-off-by: NAntonio Ospite <ao2@ao2.it>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Instead of checking for SIXAXIS_CONTROLLER_USB and SIXAXIS_CONTROLLER_BT
separately, a check on SIXAXIS_CONTROLLER can be used when setting
connect_mask.
Signed-off-by: NAntonio Ospite <ao2@ao2.it>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Without this patch the 3rd party INTEC (PS3) controller will blink all
leds when user turns them off, it appears to require an extra flag set.
Signed-off-by: NSimon Wood <simon@mungewell.org>
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Add support for setting the blink rate of the LEDs.  The Sixaxis allows control
over each individual LED, but the Dualshock 4 only has one global control for
controlling the hardware blink rate so individual colors will fall back to
software timers.

Setting the brightness cancels the blinking as per the LED class specifications.

The Sixaxis and Dualshock 4 controllers accept delays in decisecond increments
from 0 to 255 (2550 milliseconds).

The value at index 1 of the DualShock 4 USB output report must be 0xFF or the
light bar won't blink.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Add an IDA id allocator to assign unique, sequential device ids to Sixaxis and
DualShock 4 controllers.

Use the device ID to initialize the Sixaxis and DualShock 4 controller LEDs to
default values.  The number or color of the controller is set relative to other
connected Sony controllers.

Set the LED class brightness values to the initial values and add the new led to
the array before calling led_classdev_register so that the correct brightness
value shows up in the LED sysfs entry.

Use explicit module init and exit functions since the IDA allocator must be
manually destroyed when the module is unloaded.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Use the controller Bluetooth MAC address as the unique identifier in the
battery name string instead of the atomic integer that was used before.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Convert all of the local initialization and shutdown functions to take a
parameter type of struct sony_sc* instead of using a mix of struct sony_sc* and
struct hid_device*.

Allows for the removal of some calls to hid_get_drvdata().
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Use a struct for the Sixaxis output report that uses named members to set the
report fields.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Use inliners to make sure that the work queue initialization flag is always
checked and set correctly when initializing or cancelling the work queue.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Remove redundant cancel_work_sync() call caused by mismerge.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Fix unfortunate mismerge between the fixes and sony branch causing
code duplication and unterminated basic block.
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

The Sixaxis and DualShock 4 want HID output reports sent on the control
endpoint when connected via Bluetooth.  Set the
HID_QUIRK_NO_OUTPUT_REPORTS_ON_INTR_EP flag for these devices so hidraw write()
works properly.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Byte 31 of the Sixaxis report can change depending on whether or not the
controller is rumbling.  Using bit 3 is the only reliable way to detect the
state of the cable regardless of rumble activity.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

hid_out_raw_report is going to be obsoleted as it is not part of the
unified HID low level transport documentation
(Documentation/hid/hid-transport.txt)

To do so, we need to introduce two new quirks:
* HID_QUIRK_NO_OUTPUT_REPORTS_ON_INTR_EP: this quirks prevents the
  transport driver to use the interrupt channel to send output report
  (and thus force to use HID_REQ_SET_REPORT command)
* HID_QUIRK_SKIP_OUTPUT_REPORT_ID: this one forces usbhid to not
  include the report ID in the buffer it sends to the device through
  HID_REQ_SET_REPORT in case of an output report

This also fixes a regression introduced in commit 3a75b249
(HID: hidraw: replace hid_output_raw_report() calls by appropriates ones).
The hidraw API was not able to communicate with the PS3 SixAxis
controllers in USB mode.
Reviewed-by: NDavid Herrmann <dh.herrmann@gmail.com>
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Tested-by: NAntonio Ospite <ao2@ao2.it>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

If a Sixaxis or Dualshock 4 controller is connected via USB while already
connected via Bluetooth it will cause duplicate devices to be added to the
input device list.

To prevent this a global list of controllers and their MAC addresses is
maintained and new controllers are checked against this list.  If a duplicate
is found, the probe function will exit with -EEXIST.

On USB the MAC is retrieved via a feature report.  On Bluetooth neither
controller reports the MAC address in a feature report so the MAC is parsed from
the uniq string.  As uniq cannot be guaranteed to be a MAC address in every case
(uHID or the behavior of HIDP changing) a parsing failure will not prevent the
connection.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Reviewed-by: NDavid Herrmann <dh.herrmann@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Make sure that an out-of-bounds read doesn't occur in the Sixaxis battery level
lookup table in the event that the controller sends an invalid battery status
value in the report.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Reviewed-by: NDavid Herrmann <dh.herrmann@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Only initialize force feedback for devices that actually support it (Sixaxis and
Dualshock 4) to prevent calls to schedule_work() with an uninitialized work
queue.

Move the cancel_work_sync() call out of sony_destroy_ff() since the state worker
is used for the LEDs even when force-feedback is disabled.

Remove the sony_destroy_ff() function since it is no longer used.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Reviewed-by: NDavid Herrmann <dh.herrmann@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Convert multi-line comments to comply with the kernel coding style.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Reviewed-by: NDavid Herrmann <dh.herrmann@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Don't initialize force-feedback for devices that don't support it to avoid calls
to schedule_work() with an uninitialized work_struct.

Move the cancel_work_sync() call out of sony_destroy_ff() since the state worker
is used for the LEDs even when force-feedback is disabled.

Remove sony_destroy_ff() to avoid a compiler warning since it is no longer used.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Add a SIXAXIS_CONTROLLER macro to simplify conditionals where the
connection type is irrelevant.

Enable the LED and force feedback controls for Sixaxis controllers connected via
Bluetooth.

Send Sixaxis Bluetooth output reports on the control channel.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Reviewed-by: NDavid Herrmann <dh.herrmann@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

The battery_charging and cable_state flags were backwards on the Sixaxis.
The low bit of report byte 30 is 0 when charging and 1 when not.
Bit 5 of byte 31 is 0 when a USB cable is connected and 1 when not.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

  ret = hid_output_raw_report(A, B, C, HID_FEATURE_REPORT);
is equivalent to
  ret = hid_hw_raw_request(A, B[0], B, C, HID_FEATURE_REPORT, HID_REQ_SET_REPORT);
whatever the transport layer is.

So use the new API where available
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Add the conditionals to enable rumble, battery reporting, LED and touchpad
support for the Dualshock 4 in Bluetooth mode.

Add dualshock4_set_operational_bt to initialize the controller to the proper
operational state.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Reviewed-by: NDavid Herrmann <dh.herrmann@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Set the initial battery level to 100% to avoid false low battery warnings if
the battery state is polled before a report with the actual battery level is
received.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Reviewed-by: NDavid Herrmann <dh.herrmann@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Add Dualshock 4 battery and touchpad parsing for Bluetooth reports.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Reviewed-by: NDavid Herrmann <dh.herrmann@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Add formating for the Dualshock 4 output report data in Bluetooth mode.

In Bluetooth mode the Dualshock 4 wants output reports sent on the control
channel.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Reviewed-by: NDavid Herrmann <dh.herrmann@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

By default, the Dualshock 4 sends controller data via report 1. Once a valid
output report 0x11 is received or a feature report of type 0x02 is requested the
controller changes from sending data in report 1 to sending data in report 17,
which is unmapped in the default descriptor. The mappings have to be moved to
report 17 to let the HID driver properly process the incoming reports.
Signed-off-by: NFrank Praznik <frank.praznik@oh.rr.com>
Reviewed-by: NDavid Herrmann <dh.herrmann@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>