This is the first step in switching away from the check framework.
Our litest macros already do almost exactly the same anyway so most of
this is a simple sed with a few compiler fixes where things mismatch
(nonnull -> notnull) and (_tol -> _epsilon).
This now generates a whole bunch of integer mismatch warnings: check
casts everything to intmax_t whereas we use typeof, so lots of warnings
especially for enums.
Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
Theoretically we should be using ck_assert_double_eq here for
consistency but this patch is part of a series eventually
replacing those calls, so let's jump to litest_assert_double
directly to avoid further rebase conflicts.
Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1059>
Wraps libinput_dispatch() with a location which will make things a bit
easier to track. Output (in --verbose) is something like:
gestures_swipe_3fg_unaccel_fn():1346 - dispatching
Which makes it easier to associate the various calls to libinput
dispatch with the other output from libinput.
This patch switches all uses of libinput_dispatch() in test cases over
but not the litest functions that may call dispatch too. Remains to be
seen if that is necessary.
Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1048>
For a device used upside-down, make sure the wheels correspond to the
new physical directions. There's a grace range of 20 degrees either way
since that seems like it makes sense.
For 90 degree rotation (or 270 degree) the wheel is left as-is, the
heuristics to guess what angle we want in this case is not clear enough.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
The key_count array for buttons records the logical button sent to the
client - for left-handed configurations that means a BTN_LEFT is
recorded as BTN_RIGHT.
When the device is suspended and we are releasing all keys we must thus
release the button code as-is without trying to map it again.
Fixes#881
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
A whitespace fix, moving a check-related #define closer to the #include
and change a test that doesn't need a device to litest_add_no_device().
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Adds a dedicated scroll movement type to the custom acceleration profile.
Supported by physical mouse and touchpad.
Other profiles remain the same by using the same unaccelerated filter for the scroll filter.
Signed-off-by: Yinon Burgansky <51504-Yinon@users.noreply.gitlab.freedesktop.org>
Adds min and max size limit for custom acceleration function's points.
Adds tests to make sure validation works properly.
Signed-off-by: Yinon Burgansky <51504-Yinon@users.noreply.gitlab.freedesktop.org>
The custom acceleration profile allow the user to define custom
acceleration functions for each movement type per device, giving
full control over accelerations behavior at different speeds.
This commit introduces 2 movement types which corresponds to the
2 profiles currently in use by libinput.
regular filter is Motion type.
constant filter is Fallback type.
This allows possible expansion of new movement types for the
different devices.
The custom pointer acceleration profile gives the user full control over the
acceleration behavior at different speeds.
The user needs to provide a custom acceleration function f(x) where
the x-axis is the device speed and the y-axis is the pointer speed.
The user should take into account the native device dpi and screen dpi in
order to achieve the desired behavior/feel of the acceleration.
The custom acceleration function is defined using n points which are spaced
uniformly along the x-axis, starting from 0 and continuing in constant steps.
There by the points defining the custom function are:
(0 * step, f[0]), (1 * step, f[1]), ..., ((n-1) * step, f[n-1])
where f is a list of n unitless values defining the acceleration
factor for each velocity.
When a velocity value does not lie exactly on those points, a linear
interpolation of the two closest points will be calculated.
When a velocity value is greater than the max point defined, a linear
extrapolation of the two biggest points will be calculated.
Signed-off-by: Yinon Burgansky <51504-Yinon@users.noreply.gitlab.freedesktop.org>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
The `debounce_bounce_high_delay` and `debounce_spurious_trigger_high_delay`
tests are failing with annoying frequency in valgrind, but that is
entirely due to valgrind being too slow for the tight timing reqirements
of these tests. Skipping them in valgrind has next to no potential to hide
memory leaks because the code paths leading to success are also covered by
other tests which are less picky about timing, and the CI test suite run
without valgrind still tests for their success.
Signed-off-by: satrmb <10471-satrmb@users.noreply.gitlab.freedesktop.org>
These two tests were identical except for the WHEEL/HWHEEL
differentiator, let's make this into a ranged test instead.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
The IBM/Lenovo Scrollpoint mouse features a trackpoint-like stick that
sends a great amount of scroll deltas.
In order to handle the device, a quirk is in place to normalize the
scroll events as they were relative motion.
However, when high-resolution scroll was implemented, we started
normalizing the hi-res events instead of the lo-res events by mistake.
Fix the quirk by normalizing the right deltas.
Fixes: 6bb02aaf30 ("High-resolution scroll wheel support")
Signed-off-by: José Expósito <jose.exposito89@gmail.com>
Tested-by: Peter Ganzhorn <peter.ganzhorn@gmail.com>
These tests gave us false positives for devices without a REL_WHEEL or
REL_HWHEEL because one of the helper functions papered over missing
events.
We have two tests here, one for horizontal, one for vertical but they
mixed WHEEL and HWHEEL in both tests. Fix this by splitting them
properly, so each test only checks that axis.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Since cd4f2f32b5 ("fallback: disable mouse scroll wheel while middle
button is pressed") the mouse wheel is inhibited while the mouse wheel
is pressed.
The original intention of this feature was to avoid unintended scroll
while pressing the scroll wheel. However, now that high-resolution
scroll is fully integrated in libinput we can improve this feature and
filter unintended scroll (below half a detent) and allow it when it is
intended (over half a detent).
Remove the "WHEEL_STATE_PRESSED" state from the wheel state machine and
let the general heuristics handle this case.
Also, remove the specific tests for this feature as now it is covered
by the general test cases.
Signed-off-by: José Expósito <jose.exposito89@gmail.com>
Most mice with high-resolution support have a mechanism in place to
adjust the wheel to a detent. When scrolling, it is possible to stop
between two detents and this mechanism could generate a small amount of
scroll in the oposite direction.
Track the scroll direction in the wheel state machine and reset it when
the direction changes to avoid this issue.
Signed-off-by: José Expósito <jose.exposito89@gmail.com>
Mice with high-resolution support can generate deltas when the finger is
put on the wheel or when the user tries to click the wheel.
To avoid sending involuntary scroll events, add an extra state the the
wheel state machine to accumulate scroll deltas.
While the accumulated scroll is lower than a certain threshold, ignore
them until the threshold is reached.
Since no finish event is sent by the mouse, reset the state machine
after a period of scroll inactivity.
Signed-off-by: José Expósito <jose.exposito89@gmail.com>
Some devices might announce support for high-resolution scroll wheel
by enabling REL_WHEEL_HI_RES and/or REL_HWHEEL_HI_RES but never send
a high-resolution scroll event.
When the first low-resolution scroll event is received without any
previous high-resolution event, print a kernel bug warning and start
emulating high-resolution scroll events.
Fix#668
Signed-off-by: José Expósito <jose.exposito89@gmail.com>
Starting with kernel v5.0 two new axes are available for high-resolution wheel
scrolling: REL_WHEEL_HI_RES and REL_HWHEEL_HI_RES. Both axes send data in
fractions of 120 where each multiple of 120 amounts to one logical scroll
event. Fractions of 120 indicate a wheel movement less than one detent.
This commit adds a new API for scroll events. Three new event types that encode
the axis source in the event type name and a new API to get a normalized-to-120
value that also used by Windows and the kernel (each multiple of 120 represents
a logical scroll click).
This addresses a main shortcoming with the existing API - it was unreliable to
calculate the click angle based on the axis value+discrete events and thus any
caller using the axis value alone would be left with some ambiguity. With the
v120 API it's now possible to (usually) calculate the click angle, but more
importantly it provides the simplest hw-independent way of scrolling by a
click or a fraction of a click.
A new event type is required, the only way to integrate the v120 value
otherwise was to start sending events with a discrete value of 0. This
would break existing xf86-input-libinput (divide by zero, fixed in 0.28.2) and
weston (general confusion). mutter, kwin are unaffected.
With the new API, the old POINTER_AXIS event are deprecated - callers should use
the new API where available and discard any POINTER_AXIS events.
Notable: REL_WHEEL/REL_HWHEEL are emulated by the kernel but there's no
guarantee that they'll come every accumulated 120 values, e.g. Logitech mice
often send events that don't add up to 120 per detent.
We use the kernel's wheel click emulation instead of doing our own.
libinput guarantees high-resolution events even on pre-5.0 kernels.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: José Expósito <jose.exposito89@gmail.com>
The group names are forced by check (they are called suite names there) but
for our test suite they provide very little benefit. Much easier to just
use the filename a test is in as group name.
This removes the pure substring match for --filter-group, it's now fnmatch
only. group names are short enough that the typing isn't an issue and we don't
want to run tests twice (e.g. 'pad' is also in 'touchpad').
This patch caused #574 until it got fixed in d838e3a3a4
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Symmetrical to litest_create_context(), this allows us to store special data
in that context that we have access to during the tests.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
This has never been supported through the stack. No device ever had the
required MOUSE_WHEEL_TILT_VERTICAL/HORIZONTAL udev property set, so
libinput never set the right axis source. Neither weston nor mutter
added the code for it. Even if we added wheel tilt for devices now, it
would break those devices. And the benefit we get from having those
separate is miniscule at best.
So let's do the long-term thing and just deprecate this axis source.
The wheel tilt mouse test device remains in the test suite, with the
udev properties set just to verify that we do indeed ignore those now.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Scroll button locking is an accessibility feature. When enabled, the scroll
button does not need to be held down, the first click holds it logically down,
to be released on the second click of that same button.
This is implemented as simple event filter, so we still get the same behavior
from the emulated logical button, i.e. a physical double click results in a
single logical click of that button provided no scrolling was triggered.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
This used to work under valgrind up to F30 but with the F31 beta something is
now a tad slower so it triggers the timeouts before the middle emulation kicks
in.
The middlebutton timeout is 50ms and the first debounce timeout is 30ms, so if
we're late by 20ms, well, there goes the timeout.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Alleged division by zero and use of an uninitialized variable. Both cannot
happen the way we call the tests, so let's just abort to make coverity happy.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
This escaped us before because the MOUSE_DPI setting on the low-dpi device was
ignored thanks to a broken udev rule (see a future commit for that).
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
If a middle-button-emulating device is removed with one button down, the timer
never gets cancelled and triggers an assert during device removal.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
An emulated button is recorded as BTN_MIDDLE in the key down mask. If the
device is removed in that state, the BTN_MIDDLE event processed triggers
an assertion when we try to send out the event twice.
Fixes#201
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
If a device was removed while a button was held down and within the timeout,
the timer was never cancelled (and removed from the timer list), triggering an
assert during device removal.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
destroy isn't called until the last libinput_device_unref(), so we may trigger
a debounce timer after the device was removed. The same fix is neded for the
touchpad interface.
Fixes https://gitlab.freedesktop.org/libinput/libinput/issues/72
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
A set of wireless devices that can scramble the timestamps, so we get
press/release within 8ms even though I doubt the user is capable of doing
this. Since they're generally good quality anyway, let's just disable
debouncing on those until someone complains and we need something more
sophisticated.
https://bugs.freedesktop.org/show_bug.cgi?id=104415
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
The current debouncing code monitors events and switches on when events are
too close together. From then on, any event can be delayed.
Vicente Bergas provided an algorithm that avoids most of these delays:
on a button state change we now forward the change without delay but start a
timer. If the button changes state during that timer, the changes are
ignored. On timer expiry, events are sent to match the hardware state
with the client's view of the device. This is only done if needed.
Thus, a press-release sequence of: PRP sends a single press event, a sequence of
PRPR sends press and then the release at the end of the timeout. The timeout
is short enough that the delay should not be noticeable.
This new mode is called the 'bounce' mode. The old mode is now referred to as
'spurious' mode and only covers the case of a button held down that loses
contact. It works as before, monitoring a button for these spurious contact
losses and switching on. When on, button release events are delayed as before.
The whole button debouncing moves to a state machine which makes debugging a
lot easier. See the accompanying SVG for the diagram.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
This is via a simple search & replace. Later auditing is needed to switch
clicks that should not be debounced (e.g. touchpads) back to a non-debounced
version.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Some devices have worn-out switches or just cheap switches that trigger
multiple button events for each press. These can be identified by unfeasably
short time deltas between the release and the next press event. In the
recordings I've seen so far, that timeout is 8ms.
We have a two-stage behavior: by default, we do not delay any events but we
monitor timestamps. The first time a bouncing button is detected we switch to
debounce mode. From then on, release events are delayed slightly to check for
subsequent button events. If one occurs, the releas and press are filtered. If
none occurs, the release event is passed to the caller.
https://bugs.freedesktop.org/show_bug.cgi?id=100057
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
If the kernel sends us a button press for a button that is thought to be down
we have lost track of the state of the button. Ignore the button press event,
in the hope that the next release makes things right again.
A release event may be masked if another process grabs the device for some
period of time, e.g. libinput debug-events --grab.
https://bugs.freedesktop.org/show_bug.cgi?id=101796
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
