This adds a new joystick backend, implemented on top of HID and without
any host dependencies. This will be progressively implementated, and
it's not going to be usable until at least a few more patches.
Because of that, and because it may also introduce regressions compared
to the existing backends, it is disabled by default and is optionally
enabled using the following global registry key:
[HKCU\\Software\\Wine\\DirectInput\\Joysticks]
"HID"="enabled"
Or using the corresponding AppDefaults registry key:
[HKCU\\Software\\Wine\\AppDefaults\\<app.exe>\\DirectInput\\Joysticks]
"HID"="enabled"
This setting will be removed later, when it becomes usable enough, to
use the individual device disable mechanism available in joy.cpl.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
It's accessed on Release, and this will make it easier to handle device
creation failure cleanup.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Both Acquire() and event processing with DirectInput seem to be
asynchronous. In most cases we can just keep hammering GetDeviceData()
until the event gets processed.
Things get pretty racy around Acquire() though. If we fire event right
after the device is acquired we can find ourselves in one of the three
situations:
1. Event happened after acquiring has completed - the wait will suffice.
2. Event happened before acquiring did any real work - the device will
pick up the state as if the event was processed, but there's nothing
in GetDeviceData(). Because of that we cannot fail on wait.
3. Event happened somewhere in the middle of acquiring - we ended up
both missing the event for GetDeviceData() and we have outdated
state. Sending event again will register as if the button was not
already pressed.
This change covers all three scenarios.
Signed-off-by: Arkadiusz Hiler <ahiler@codeweavers.com>
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This helps when dealing with the quirky c_dfDIJoystick2 format,
which defines all the sliders as overlapping, i.e. rglSlider[2],
rglVSlider[2], rglASlider[2] and rglFSlider[2] all share the same
offset.
With the Linux backends, if the device exposes enough axes (e.g. some
HOTAS) we end up using rglSlider.
Let's make sure we don't unnecessarily override the values there.
Signed-off-by: Arkadiusz Hiler <ahiler@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
To factor out device allocation and initialization.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Instead of a separate duplicate implementation.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Instead of sometimes using IDirectInputDevice8A.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
To make it more likely it gets focused on X11 and fix spurious failures.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
FlatOut2 demo calls FreeLibrary on the dinput8 module more times than
it has loaded it, and the module reference that the internal thread has
is getting decremented before its exit.
Having an internal window to destroy also makes it more likely to crash
as it takes more time to exit calls the window procedure function.
Waiting for the thread to complete when all the dinput instances are
destroyed ensures that the thread isn't alive when the game frees the
DLL, and prevents the thread from crashing.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=50673
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This fixes not working analog inputs on game controllers in Slay the Spire.
Signed-off-by: Arkadiusz Hiler <ahiler@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
LL hooks are heavy and using them cause performance hit with high
polling rate mice. We don't need them anymore since we now use rawinput
API for mouse device.
This also uses a separate list for rawinput mouse devices.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This adds a message window that will be used as the rawinput target
window for WM_INPUT messages.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Instead of using function pointer, which is also not very CPU friendly.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Instead of going through devices list of dinputs list, and checking
their acquired field, which is not CPU friendly.
This also removes the now unused IDirectInputImpl critical section and
devices list.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This message triggers screensaver notification in the user drivers, but
it is not useful to send it on every input message, and it even hurts
performance when high polling mouse is used.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=33643
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=46976
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Grand Theft Auto IV injects VK_F8 and scancode=0, and expects DirectInput not
to report that F8 is pressed.
Signed-off-by: Brendan Shanks <bshanks@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
The buffer size was previously limited to 20 in:
8d05256172
Testing shows that it is somewhere around 1024 on Windows, and using a
smaller buffer causes some input data to be lost with high polling rate
mouses. It can be spotted in Wolfenstein II: New Colossus menus.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
To prevent msvcrt linking its own importlib into DLL.
Signed-off-by: Jacek Caban <jacek@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
When a program calls SetProperty with DIPROP_BUFFERSIZE, dinput records
this value for GetProperty but only uses it when the device can support
that number of buffers otherwise a max value.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=45732
Signed-off-by: Alistair Leslie-Hughes <leslie_alistair@hotmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Also fixes a few test strings.
Signed-off-by: Alistair Leslie-Hughes <leslie_alistair@hotmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Tablet/Stylus devices have a js* entry which we are
treating as a joystick. Any testbox VM that is configured
with a tablet is currently failing the joystick tests.
Signed-off-by: Alistair Leslie-Hughes <leslie_alistair@hotmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Windows uses an undocumented GUID (ending in ASCII "PIDVID") for
non-keyboard/mouse devices. Various games (DiRT Rally 2) and libraries
(SDL, GLFW) look for this GUID to match specific devices or know that
the PID/VID can be extracted.
Signed-off-by: Brendan Shanks <bshanks@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Some Ubisoft titles use the property to check if a dinput gamepad is
also an xinput device.
Signed-off-by: Alexey Prokhin <alexey@prokhin.ru>
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This prevents crashes if a controller disconnects from OS/X.
Signed-off-by: Aric Stewart <aric@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Star Wars: Rogue 3D requests a DirectX 3.0 version
0108:trace:dinput:DirectInputCreateEx (0x400000,0300,{9a4cb684-236d-11d3-8e9d-00c04f6844ae},0x7acfa0,(nil))
and then tries to interate over all Joysticks.
When using modern windows, the current tests show that this is the correct behaviour.
However, it's possible to use a shim to allow older games, like the
one above, to interate over joysticks.
Signed-off-by: Alistair Leslie-Hughes <leslie_alistair@hotmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
The axes wheel, gas, and brake are just differently named replacements
for X, Y, and Z when using wheel-like joysticks.
Make those accessible by remapping them to X, Y, and Z internally.
Signed-off-by: Tim Schumacher <timschumi@gmx.de>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This matches what Windows does with common controllers like xbox 360.
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This fix another deadlock issue with dinput.
The old implementation of dinput waits for a hook thread to exit
during its uninitialization, which will never return during DLL_THREAD_DETACH.
Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Based on a patch by Andrew Wesie.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=45568
Signed-off-by: Zebediah Figura <z.figura12@gmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This fixes "keyboard.c:365: Tests skipped: failed to queue keyboard event"
message on my Win10 PC (real hardware).
Since some thread messages and window messages, including IME related,
are observed on the thread, they sometimes confuse the message handler.
Signed-off-by: Akihiro Sagawa <sagawa.aki@gmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
For the further improvments, subtype is a member of SysKeyboardImpl structure.
Signed-off-by: Akihiro Sagawa <sagawa.aki@gmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This is already done on the linux backends.
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Avoids letting the display go to sleep while a joystick is being used.
Signed-off-by: Zebediah Figura <zfigura@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This causes intermittent failures due to implementation of our window drivers.
Signed-off-by: Zebediah Figura <z.figura12@gmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
The wiimote is a well known problematic device, mainly because it is not
a joystick. It is a USB device. But the classic controller is at least
advertised as a joystick with 6 axes and 11 buttons (js driver only, NO
event).
Signed-off-by: Bruno Jesus <bjesus@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
EnumDevicesBySemantics enums only devices with given username when
DIEDBSFL_THISUSER is set and only unowned devices when
DIEDBSFL_AVAILABLEDEVICES is set.
Signed-off-by: Jetro Jormalainen <jje-wine@jv.jetro.fi>
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Rémi Bernon