If a device is positively identified as an accelerometer, pointing stick
or clickpad, then we don't need to second-guess it.
In practice this does not change the result for any device in our
test data, so add some artificial records that exercise this.
Signed-off-by: Simon McVittie <smcv@collabora.com>
In newer kernels, devices that can be positively identified as a
particular device type (for example accelerometers) get a property
bit set. Plumb this information through into the function, but don't
use it for anything just yet.
Signed-off-by: Simon McVittie <smcv@collabora.com>
The props array was too small for the highest property bits to be set,
although in practice this didn't matter since only the lower-order bits
have a meaning. Make it consistent with all the others.
Signed-off-by: Simon McVittie <smcv@collabora.com>
udev distinguishes between ID_INPUT_KEY, a device with any keyboard keys
at all (including for example USB numeric keypads, Bluetooth headsets
with volume control buttons, and some game controllers; and
ID_INPUT_KEYBOARD, a reasonably fully-featured keyboard that you could
use for general-purpose text entry. If we do the same here, then it's
useful input to our heuristics for identifying devices: for example,
a device with ID_INPUT_KEY could reasonably be a gamepad, but a device
with ID_INPUT_KEYBOARD certainly isn't.
Resolves: https://github.com/libsdl-org/SDL/issues/7827
Signed-off-by: Simon McVittie <smcv@collabora.com>
We don't currently use these for anything, but we might start using
them as input to our heuristics as part of #7697.
Signed-off-by: Simon McVittie <smcv@collabora.com>
We don't currently use these in our device-classification heuristic,
but it could be a useful input in future.
Thanks to Sam Lantinga, Ben Fradella, kevenwyld and schlegp for
providing some of these.
Signed-off-by: Simon McVittie <smcv@collabora.com>
The comments here disagreed with the actual bytes. According to
https://elixir.bootlin.com/linux/v6.3.7/source/drivers/hid/hid-wiimote-modules.c,
the Balance Board reports BTN_A and ABS_HAT0X, HAT0Y, HAT1X and HAT1Y.
This means the comments here were correct, but the .abs bits shown
were in the wrong byte.
Matching the Wii U Pro Controller against the same kernel source, it
appears to be correct: it's the same representation as a PS3 gamepad,
except that it lacks the Z and RZ axes for analogue triggers.
Signed-off-by: Simon McVittie <smcv@collabora.com>
Some of the test data previously seen here was guessed from kernel
source code, and not all of it was correct. The following devices have
now been verified against `evemu-describe` output with Linux 6.3
(thanks to Jeremy Whiting for collecting this):
- basic Wiimote
- buttons
- 3-axis accelerometer
- infra-red sensor for Sensor Bar location (precise aim)
- Motion Plus accessory (3-axis gyroscope)
- Nunchuck accessory (joystick, 2 buttons, second 3-axis accelerometer)
- Classic Controller accessory (a complete traditional gamepad)
Signed-off-by: Simon McVittie <smcv@collabora.com>
This slightly newer device than the one from #7814 is functionally
equivalent when connected via USB. When connected via Bluetooth, it has
a different button mapping.
Thanks to Sam Lantinga.
Signed-off-by: Simon McVittie <smcv@collabora.com>
We can't actually tell yet whether a controller has paddles, so this code isn't effective, but I'll file an upstream issue and see if we can get that resolved.
This heuristic for gamepads without a more specific mapping already
tried two incompatible conventions for handling triggers: the Linux
Gamepad Specification uses hat switch 2 for the triggers (for whatever
reason), but the de facto standard set by the drivers for older Xbox
and Playstation controllers represents each trigger as the Z-axis of
the nearest analog stick.
Android documentation encourages Bluetooth gamepad manufacturers to use
a third incompatible convention where the left and right triggers are
represented as the brake and gas pedals of a driving simulator
controller. The Android convention also changes the representation of
the right stick: instead of using X and Y rotation as a second pair
of axes, Android uses Z position as a second horizontal axis, and
Z rotation as a second vertical axis.
Try to cope gracefully with all of these. This will hopefully resolve
the issue described in #5406 (when using unpatched kernels).
Signed-off-by: Simon McVittie <smcv@collabora.com>
The bitfield `mapped` has two different sets of meanings, depending
whether we're setting up the triggers or the d-pad. Represent them
as symbolic constants rather than opaque integers.
Signed-off-by: Simon McVittie <smcv@collabora.com>
C SDLmain() thread might have started (mSDLThread.start() called)
while the SDL_Init() might not have been called yet,
and so the previous QUIT event will be discarded by SDL_Init() and app is running, not exiting.
This is reprocible by adding instrumentation code in the SDLActivity.
And hopefully, this could fix an ANR, where SDLActivity is in WAITING state (in thread.join()):
at java.lang.Thread.join (Thread.java:1519)
at org.libsdl.app.SDLActivity.onDestroy (SDLActivity.java)
while SDLThread seems to be running
This prevents the case where the mouse might be at the edge of the
window when enabling relative mode, which confuses macOS, at it
might believe the user is attempting to resize the window.
Fixes#6994.
(cherry picked from commit 2afb49ba9a2edc361a3b35d8e30be977a5bcc0c4)
Sam Lantinga
Ozkan Sezer
Simon McVittie
SDL Wiki Bot
Anonymous Maarten
Sylvain
Mathieu Eyraud
Linus Probert
Ryan C. Gordon