config.set10 is much more convenient and nicer to read but can provide
false positive if the value is 0 and #ifdef is used instead of #if. So
let's switch everything to use #ifdef instead, that way we cannot get
false positives if the value is unset.
Closes#1162
Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1277>
Tablets may have different ABS_PRESSURE ranges with the oldest tablets
having 1k pressure range, then 2k, and the newer ones 8k.
If the same tool is used across two tablets with different ABS_PRESSURE
ranges, the first tablet in proximity calculated the range on where to
normalize to. As a result the other tablet either couldn't reach the
full pressure (2k pressure first, then 8k) or the full pressure range
was reached at a fraction of the full range (8k pressure first, then
2k).
Fix this by moving the threshold handling into a separate struct and
hardcoding up to 4 of those per tool. That is 2 more than the more
complicated setups I've heard of (and this only applies to tracking the
same stylus across those tablets anyway).
This duplicates the pressure offset heuristics but that's easier than
figuring out how to handle heuristics across potentially two tablets.
The range configuration is left as-is on the assumption that this one is
per tool, not per tablet.
Closes#1089
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1143>
If a tablet has an area configured and the pen goes into proximity
outside this area, ignore all events from this sequence. This truly
deactivates that area so it can even be used for e.g. placing a pen
there.
For simplicity, a sequence that starts outside the configured area will
be completely ignored, i.e. moving into the tablet area will not trigger
any fake proximity events as we cross into the allowed area. This
requires quite a bit of effort and it's unclear if it's really needed by
users - we can reconsider when we get complaints.
We do however accept a proximity event within within 3% of the
configured area. This gives us 6mm on a 200mm tablet where we can move
in from the area and still have events work, i.e. some error margin for
where a user needs both an area and work closes to the edge of that
area.
Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1013>
For external tablets like the Intuos series we now expose the area
rectangle configuration and the (minimum) implementation required to
make this work.
Because an area configuration may apply late and tablet events usually
get scaled by the compositor we need to store the current axis extents
in each event. This is to behave correctly in this events sequence:
1. tool proximity in
2. caller changes config, config is pending
3. tool moves, generates events
4. tool goes out of prox, new config applies
5. caller processes motion events from step 3
If the caller in step five uses any of the get_x_transformed calls these
need to be scaled relative to the original area, not the one set in
step 2.
The current implementation merely clips into the area so moving a stylus
outside the area will be equivalent to moving it along the respective
edge of the area. It's not a true dead zone yet.
Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1013>
Instead of re-creating the the libwacom device from the database every
time we need it let's create it once during tablet|pad_init and pass it
down to the functions.
This allows us to have one point per tablet/pad where we can log an
error if the device is not supported by libwacom - previously this was
printed during the left-handed setup.
Part-of: <https://gitlab.freedesktop.org/libinput/libinput/-/merge_requests/1073>
Data in
https://gitlab.freedesktop.org/libinput/libinput/-/issues/225#note_379034
suggests that AES devices have lower noise than the older EMR
devices, so let's try disabling it for those devices.
We can't directly get the AES devices in libinput unless we want to add a
whole bunch of quirks for the various vid/pid combinations. But we can get
that info from libwacom, primarily because we know that libwacom will list all
known AES pens for any device. So we can check for one that we know of (0x11)
and if it's in the list, the tablet is an AES tablet.
Setting the history size to 1 means we never do any actual smoothing.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
This quirk was introduced for #248 was caused by buggy input-wacom drivers,
not by actual firmware, see
https://gitlab.freedesktop.org/libinput/libinput/issues/381#note_279371
This appears to be the only tablet where this fix was needed, but we've been
playing whack-a-mole ever since to work around the various other tablets that
break with this behavior in place.
So let's revert that fix and hope there aren't any other tablets out there
(and if they are, we can probably quirk those). The revert makes the ISDV4 pen
quirk obsolete (see 9cb089f2b6), so this was
folded into this commit.
This reverts commit 4f63345b60.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
While buttons are down, don't let a forced proximity out happen. If the tablet
goes out of proximity normally that's fine but we don't force a proximity out.
Remains to be seen if this causes stuck buttons now on devices that rely on
the forced proximity out...
Fixes#403
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
This pen has random timeouts, often when a button is pressed. This causes a
forced proximity out (and the button release) and makes the whole device a
tad unusable.
Nothing we can detect by heuristics since it looks like other devices that
don't send proximity out events. And the timeout can be quite high, the
recording in #304 has over 800ms for one sequence.
Fixes#304
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Where the proximity out event is delayed by the kernel, libinput would cause
an extra proxmity in-out after the forced proximity out event.
Event sequence is basically (k: kernel, l: libinput)
k: tablet axis events
l: tablet axis events
k: nothing for $proximity timer milliseconds
l: tablet proximity out
k: proximity out event
l: proximity in event
l: proximity out event
Fixes#306
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
This is the public API only, not the internal bits, so nothing will work just
yet.
This interface addition is for the Dell Canvas Totem tool, so let's go with
the same name because options like "Rotary" are too ambiguous.
The totem is a knob that can be placed on the surface, it provides us with
location and rotation data. The touch major/minor fields are filled in by the
current totem, but they're always the same size.
The totem exports BTN_0 as well, so let's add that to the debug-events output.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Follow-up to 6229df184e
We must not rely on the caller to toggle the left-handed bits correctly since
they may not know which devices belong together (despite device groups). Let's
do the right thing here, if the touchpad is set to left-handed, rotate the
tablet accordingly.
Note that the left-handed setting of the touchpad is left as-is
(right-handed). Until we have notifications about configuration changes, this
is the best we can do.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
There are tablets out there that *sometimes* send the right event sequence,
but are generally broken. So let's not disable that quirk even if we do get a
right sequence.
Affected devices: Lenovo Flex 5
Fixes#248Fixes#290
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Unlike virtually everything else, the tablet tool was processed at the time
the event was read rather than when the subsequent EV_SYN came in. This causes
difficulties with tablets that send the wrong BTN_TOOL_PEN events.
Moving the tool change processing to tablet_flush() makes the injection of the
BTN_TOOL_PEN event a lot easier, simply flipping the matching bit does the
job. It also makes it easier to ignore duplicate tool updates like we've seen
in #259.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
We expect the kernel to transition properly for us, e.g. BTN_TOOL_PEN goes to
0, BTN_TOOL_ERASER goes to 1. Two cases have surfaced recently where this
doesn't happen and debugging this takes time - so let's warn about it to make
it obvious.
Example 1: https://github.com/linuxwacom/libwacom/issues/70
Example 2: https://gitlab.freedesktop.org/libinput/libinput/issues/259
This is just a warning, nothing more. We should just handle that case
accordingly but that requires more effort.
Fixes#260
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
If the tilt angle on tip down is not 0 set the touch arbitration to a
rectangle around the assumed position of the hand. This assumed position is
right of the tip for a rightwards tilt and left of the tip for a leftwards
tilt (i.e. left-handed mode). The rectangle is 200x200mm with a 20x50mm
NW of the tip or NE for left-handed. In other words, if the period below is
the tip, the rectangle looks like this:
+-----------+ +-----------+
| . | <- for rightwards tilt | . |
| | | |
| | | |
| | for leftwards tilt -> | |
+-----------+ +-----------+
Touches within that rectangle are canceled, new touches are ignored. As the
tip moves around the rectangle is updated but touches are only cancelled on
the original tip down. While the tip is down, new touches are ignored in the
exclusion area but pre-existing touches are not cancelled.
This is currently only implemented in the fallback interface, i.e. it will
only work for Cintiqs.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Use the same blurb everywhere, changing from the old style MIT to the Expat
license we're using everywhere else.
Similar to bc9f16b40e
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
And instead disable it when we do get a proximity out.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Benjamin Tissoires <benjamin.tissoires@gmail.com>
Could be fixed in the kernel, but these tablets are effectively abandoned and
fixing them is a one-by-one issue. Let's put the infrastructure in place to
have this fixed once for this type of device and move on.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Yay-by: Benjamin Tissoires <benjamin.tissoires@gmail.com>
commit 3925936 introduced changes to container_of, this is hopefully the
last part of it.
In the linux kernel, container_of() takes a type name, and not a
variable. Without this, in some cases it is needed to declare an unused
variable in order to call container_of().
example:
return container_of(dispatch, struct fallback_dispatch, base);
instead of:
struct fallback_dispatch *p;
return container_of(dispatch, p, base);
This introduce also list_first_entry(), a simple wrapper around
container_of() to retrieve the first element of a non empty list. It
allows to simplify list_for_each() and list_for_each_safe().
Signed-off-by: Gabriel Laskar <gabriel@lse.epita.fr>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Taking the tablet events as-is produces the occasional wobble in what should
be a straight line. Bug 99961 has a jpg attachment to illustrate that.
Emulate the wacom driver behavior and average x/y across the last 4 values to
smoothen out these dents.
https://bugs.freedesktop.org/show_bug.cgi?id=99961
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Acked-by: Ping Cheng <pingc@wacom.com>
Stores the processed axes values in a history 4 events deep. Currently unused
but will be used to smoothen out axis values to avoid transducer-caused axis
wobbles.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Acked-by: Ping Cheng <pingc@wacom.com>
Set the dispatch type on creation, then check that whenever we try to get the
dispatch struct. This avoids a potential mismatch between the backends.
Plus, use of container_of means we're not dependent on the exact layout
anymore.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
So far we've relied on the wacom kernel module to do touch arbitration for us
but that won't be the case in upcoming kernels. Implement touch arbitration in
userspace by pairing the two devices and suspending the touch device whenever
a tool comes into proximity.
In the future more sophisticated arbitration can be done (e.g. only touches
which are close to the pen) but let's burn that bridge when we have to cross
it.
Note that touch arbitration is "device suspend light", i.e. we leave the
device enabled and the fd is active. Tablet interactions are comparatively
short-lived, so closing the fd and asking logind for a new one every time the
pen changes proximity is suboptimal. Instead, we just keep a boolean around
and discard all events while it is set.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Jason Gerecke <jason.gerecke@wacom.com>
This is especially a problem for the cursor tool which can be legitimately
left on the tablet. It wobbles by a couple of device units, resulting in
continuous axis updates to the caller. Pre-filter any of these events by the
axis' fuzz value so we don't even process them.
For ABS_DISTANCE which doesn't have a fuzz we hard-code a minimum fuzz of 2.
This should eventually land in the kernel though.
https://bugs.freedesktop.org/show_bug.cgi?id=94892
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Jason Gerecke <jason.gerecke@wacom.com>
Makes the code less generic, but more expressive. No visible functional
changes.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Acked-by: Jason Gerecke <jason.gerecke@wacom.com>
For the puck/lens cursor tool we need to artificially reduce proximity
detection. These tools are usually used in a relative mode (i.e. like a mouse)
and thus require lifting and resetting the tool multiple times to move across
the screen. The tablets' distance detection goes too far, requiring the user
to lift the device several cm on every move. This is uncomfortable.
Introduce an artificial distance threshold for the devices with the default
value taken from the X.Org wacom driver. If a tool is in proximity but outside
of this range, fake proximity events accordingly.
If a button was pressed while we were out of range we discard that event and
send it later when we enter proximity again.
This is the simple implementation that only takes one proximity out value (the
one from the wacom driver) and applies it to all. Those devices that support a
button/lens tool and have a different default threshold are well out of date.
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
[rebased, tests updated for new axis percentage behavior (8d76734f)]
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
If a tool wears out, it may have a pre-loaded pressure offset. In that case,
even when the tool is not physically in contact with the tablet surface it
will send pressure events.
Use automatic pressure offset detection, similar to what the X.Org wacom
driver does. On proximity-in, check the pressure and if the distance is above
50% of the range and the pressure is nonzero but below 20% of the range, use
that value as pressure offset.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Ping Cheng <pingc@wacom.com>
If it's a finger, it's a touchscreen or a touchpad, not a tablet.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Behavior for axis events in the same event frame as the BTN_TOUCH is to
always send axis events before any tip state.
Behavior for button events in the same event frame as the BTN_TOUCH is to
order button events to happen when the tip is in proximity, i.e. after the tip
event on tip down and before the tip event on tip up.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
If a tool is in proximity when we init, send a proximity event immediately.
This is only partially reliable due to the current kernel behavior:
* if the tool comes into proximity when there is no evdev client, the device
won't send any events and must be lifted out-of-proximity first. Patch is in
the works, see https://patchwork.kernel.org/patch/5924611/
* before 3.19, if the tool was in proximity (with an evdev client attached),
but goes out of proximity and back in with no client connected, we get an
immediate proximity out event from the kernel once we connect to the device
and no further events after that.
See kernel commit b905811a49bcd6e6726ce5bbb591f57aaddfd3be
Otherwise, things work as expected. The above should be fixed in the kernel
anyway.
Note that this changes the order of events during a udev seat init, before we
had all DEVICE_ADDED events in a row, now the proximity event may be
interspersed.
Reported-by: Jason Gerecke <killertofu@gmail.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Acked-by: Benjamin Tissoires <benjamin.tissoires@gmail.com>
Providing a relative axis in the axis_get_value() is inconsistent with the
other axes, this will be fixed in a follow-up commit.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Benjamin Tissoires <benjamin.tissoires@gmail.com>
Reviewed-by: Stephen Chandler Paul <thatslyude@gmail.com>
