Use the same codepath, which has the added advantage of being able to
import dmabufs.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Tested-by: Emre Ucan <eucan@de.adit-jv.com>
... in order to be able to use it from scanout as well.
In doing this, the check for format compatibility is moved from after
selecting a plane to before selecting a plane. If different planes have
disjoint format support, this ensures that we don't reject the view from
all overlay consideration, just because the first plane we found didn't
support its format.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Tested-by: Emre Ucan <eucan@de.adit-jv.com>
Use the new helper to populate the cursor state as well, with some
special-case handling to account for how we always upload a full-size
BO.
As this now fully takes care of buffer transformations, HiDPI client
cursors work, and we also clip the cursor plane completely to CRTC
bounds.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reported-by: Derek Foreman <derekf@osg.samsung.com>
Tested-by: Emre Ucan <eucan@de.adit-jv.com>
Fixes: https://gitlab.freedesktop.org/wayland/weston/issues/118
Now that we have a helper to fill the plane state co-ordinates from a
view, use this for the scanout plane.
We now explicitly check that the view fills exactly the fullscreen area
and nothing else. We then use the new helper to fill out the plane state
values, and do further checks against the filled-in co-ordinates, i.e.
that we're not trying to show an offset into the buffer, or to scale the
image.
This now allows cases where the buffer -> surface -> view -> output
transform chain cancels each other out for scaling: previously, we would
never consider a buffer for scanout unless its scale matched the
output's. We now only look at the final result of the buffer -> output
transformation, to check that this does not result in translation or
scaling.
An audit of the error paths found some places where we would leave a
plane state hanging; this makes them all consistent.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Tested-by: Emre Ucan <eucan@de.adit-jv.com>
When considering a view for placement into an overlay plane, we
previously checked that the buffer's transform and scale were identical
to the output's, and that there were no transformations applied.
We now use a more consistent set of checks through
drm_plane_state_coords_for_view. This checks the complete transformation
chain, allowing only translation and scaling; at the end, we check if
the total buffer -> surface -> view -> output chain requires scaling or
rotation, and disallow it if so.
This allows scaling in the cases where the transformation chain cancels
itself out to produce a 1:1 buffer -> output pixel scale.
An erroneously disallowed case is where buffer -> view -> output
rotations cancel each other out; we prevent a view from being on an
overlay plane if rotation is involved at all. Fixing this would require
a complete analysis of the overall transformation matrix.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Tested-by: Emre Ucan <eucan@de.adit-jv.com>
In our new and improved helper to determine the src/dest values for a
buffer on a given plane, make sure we account for all buffer
transformations, including viewport clipping.
Rather than badly open-coding it ourselves, just use the helper which
does exactly this.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reported-by: Tiago Gomes <tiago.gomes@codethink.co.uk>
Tested-by: Emre Ucan <eucan@de.adit-jv.com>
Pull this into a helper function, so we can use it everywhere.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Tested-by: Emre Ucan <eucan@de.adit-jv.com>
Rather than a hardcoded ARGB8888 -> XRGB8888 translation inside a
GBM-specific helper, just determine whether or not the view is opaque,
and use the generic helpers to implement the format translation.
As a consequence of reordering the calls in
drm_output_prepare_overlay_view(), we move the GBM BO dereference into a
different failure path, before it gets captured by the plane state.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Tested-by: Emre Ucan <eucan@de.adit-jv.com>
e2e8013633 fixed the same issue as df573031d0 in a different way.
The latter commit (applied earlier in the upstream tree) adds a variable
to assign_planes to keep track of when we successfully assign a view to
the scanout plane, and doesn't call prepare_scanout_view if we have.
The former commit adds this checking inside prepare_scanout_view: if the
pending output state already has a framebuffer assigned to the scanout
plane, we drop out of prepare_scanout_view early. The picked_scanout
variable inside assign_planes can thus be removed.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Tested-by: Emre Ucan <eucan@de.adit-jv.com>
Since it doesn't write to the parameter, we can make it const.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
The flag bits 19-22 of the connector modes, provide the aspect-ratio
information. This information can be stored in flags bits of the
weston mode structure, so that it can used for setting a mode with a
particular aspect-ratio.
Currently, DRM layer supports aspect-ratio with atomic-modesetting by
default. For legacy modeset path, the user-space needs to set the
drm client cap for aspect-ratio, if it wants aspect-ratio information
in modes.
This patch:
- preserves aspect-ratio flags from kernel video modes and
accommodates it in wayland mode.
- uses aspect-ratio to pick the appropriate mode during modeset.
- changes the mode format in configuration file weston.ini to
accommodate aspect-ratio information as:
WIDTHxHEIGHT@REFRESH-RATE ASPECT-RATIO
The aspect-ratio can take the following values :
4:3, 16:9, 64:27, 256:135.
v2: As per recommendation from Pekka Paalanen, Quentin Glidic,
Daniel Stone, dropped the aspect-ratio info from wayland protocol,
thereby avoiding exposure of aspect-ratio to the client.
v3: As suggested by Pekka Paalanen, added aspect_ratio field to store
aspect-ratio information from the drm. Also added drm client
capability for aspect-ratio, as recommended by Daniel Vetter.
v4: Minor modifications and fixes as suggested by Pekka Paalanen.
v5: Rebased, fixed some styling issues, and added aspect-ratio
information while printing weston_modes.
v6: Moved the man pages changes to a different patch. Minor
reorganization of code as suggested by Pekka Paalanen.
Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
[Pekka: replace ARRAY_SIZE with ARRAY_LENGTH]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
This is regression apparently introduced in
0de859ede4, which accidentally swapped the
sign of 'delta_width' in the original call site. If one removes an
output, the remaining outputs on the right are getting moved even
further to the right.
The outputs to the right should be moved to the left instead, to close
the gap left by the removed output.
Reported-by: Tomasz Olszak <olszak.tomasz@gmail.com>
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Ian Ray <ian.ray@ge.com>
Some modeline generators put out e.g. +HSync instead of +hsync. Accept
that too since it's not ambigous.
Signed-off-by: Guido Günther <agx@sigxcpu.org>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
This adds a function to detect the first framebuffer device in the
current seat. Instead of hardcoding /dev/fb0, detect the device
with udev, favoring the boot_vga device, and falling back to the
first framebuffer device in the seat if there is none. This is very
similar to what compositor-drm does to find display devices
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
This attempts to wake up secondary framebuffer devices
(/dev/fb1 and up) as usually these devices start powered off, and
the FBIOPUT_VSCREENINFO ioctl turns it on. This was tested on a
qemu system with the options:
-vga none -device VGA,id=video0 -device secondary-vga,id=video1 \
-device secondary-vga,id=video2
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
As only seat0 supports TTYs, this changes the logind launcher where
it detects a TTY, only if the seat is seat0. This has only been
tested for logind
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
This allows the fbdev backend to run on, and use devices from the
specified seat, similar to the drm backend.
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
This will allow the seat to be set by the environment as pam_systemd typically
sets the XDG_SEAT variable
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Rather than having a hardcoded dependency on the build-directory layout,
use an explicit module-map environment variable, which rewrites requests
for modules and helper/libexec binaries to specific paths.
Pekka: This will help with migration to Meson where setting up the paths
according to autotools would be painful and unnecessary.
Emre: This should also help setting up the test suite after a
cross-compile.
Pekka: A caveat here is that this patch makes it slightly easier to load
external backends by abusing the module map. External backends are
specifically not supported in libweston.
Signed-off-by: Daniel Stone <daniels@collabora.com>
v2:
Fixed ivi_layout-test-plugin.c:wet_module_init().
Do not change the lookup name of ivi-layout.ivi.
Improved documentation of weston_module_path_from_env() and made it cope
with map strings that a) do not end with a semicolon, and b) have
multiple consecutive semicolons.
Let WESTON_MODULE_MAP be printed into the test log so that it is easier
to run tests manually.
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Emre Ucan <eucan@de.adit-jv.com>
Suggested by Emil: Use a variable for strlen(name).
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
If the user is in group 0, we'd exit the loop early with a failure. Make sure
we run through all groups.
https://gitlab.freedesktop.org/wayland/weston/issues/86
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
[Pekka: fix one whitespace]
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
All outputs is canceled repaint when a output repaint is failed. At that
time, the output whose repaint is success is not scheduled because the
repaint status of that is still REPAINT_AWAITING_COMPLETION. Therefore,
we need to reset repaint schedule for all repainted outputs.
Signed-off-by: Tomohito Esaki <etom@igel.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
This implements a new global interface weston_touch_calibration, which
allows one client at a time to perform touchscreen calibration. This
also implements the calibrator window management.
A client asks to calibrate a specific physical touch device (not a
wl_seat which may have several physical touch devices aggregated).
Libweston grabs all touch devices and prevents normal touch event
handling during the calibation sequence.
API is added to enable this new global interface, but it not yet called
by anything. Since the implementation allows clients to grab touch devices
arbitrarily, it is not enabled by default. The compositor should take
measures to prevent unexpected access to the interface.
A client may upload a new calibration to the compositor. There is a
vfunc to allow the compositor to reject/accept it and save it to
persistent storage. The persistent storage could be a udev rule
setting LIBINPUT_CALIBRATION_MATRIX, so that all display server would
load the new calibration automatically.
Co-developed by Louis-Francis and Pekka.
v2:
- use struct weston_point2d_device_normalized
- use syspath instead of devpath
- wrong_touch was renamed to invalid_touch
- rename weston_touch_calibrator::cancelled to calibration_cancelled
- send invalid_touch on out-of-bounds touch-down
- cancel touch sequence and send invalid_touch on motion going
out-of-bounds
- rename calcoord_from_double() to wire_uint_from_double()
- send bad_coordinates error in touch_calibrator_convert()
- conversion results in 0,0 if cancelled
Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
v1 Tested-by: Matt Hoosier <matt.hoosier@gmail.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
In addition to the normal touch event processing mode, introduce a new
mode for calibrating a touchscreen input device.
In the calibration mode, normal touch event processing is skipped, and
the raw events are forwarded to the calibrator instead. The calibrator
is not yet implemented, so the calls will be added in a following patch.
To switch between modes, two functions are added, one for entering each
mode. The mode switch happens only when no touches are down on any touch
device, to avoid confusing touch grabs and clients. To realise this, the
state machine has four states: prepare and actual state for both normal
and calibrator modes.
At this point nothing will attempt to change the touch event mode.
The new calibrator mode is necessary, because when calibrating a
touchscreen, the touch events must be routed to the calibration client
directly. The touch coordinates are expected to be wrong, so they cannot
go through the normal focus surface picking. The calibrator code also
cannot use the normal touch grab interface, because it needs to be able
to distinguish between different physical touch input devices, even if
they are part of the same weston_seat. This requirement makes
calibration special enough to warrant the new mode, a sort of "super
grab".
Co-developed by Louis-Francis and Pekka.
Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
v1 Tested-by: Matt Hoosier <matt.hoosier@gmail.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Commit a30e29af2e introduced the code to
deal with a touchscreen with touches already down when Weston starts
using it. It fixed the touchpoint counting problem.
However, Weston still should not forward or process the unmatched
touch-ups either. Code inspection says it would confuse the
idle-inhibit counting, and it could probably confuse clients as well.
Hence, just drop unmatched touch-ups.
Enhance the warning message to allow identifying where the event came
from.
v2:
- use syspath instead of devpath
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
v1 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
v1 Tested-by: Matt Hoosier <matt.hoosier@gmail.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
The touchpoint counting is needed regardless of what we do with the
touch events, so move it out of process_touch_normal() into the caller
notify_touch_normalized().
This is pure refactoring.
Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
v1 Tested-by: Matt Hoosier <matt.hoosier@gmail.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
notify_touch_normalized() is an extended form of notify_touch(), adding
normalized touch coordinates which are necessary for calibrating a
touchscreen.
It would be possible to invert the transformation and convert from
global coordinates to normalized device coordinates in input.c without
adding this API, but this way it is more robust against code changes.
Recovering normalized device coordinates is necessary because libinput
calibration matrix must be given in normalized units, and it would be
difficult to compute otherwise. Libinput API does not offer normalized
coordinates directly either, but those can be fetched by pretending the
output resolution is 1x1.
Anticipating touch calibration mode, the old notify_touch() is renamed
into a private process_touch_normal(), and the new
notify_touch_normalized() delegates to it.
Co-developed by Louis-Francis and Pekka.
v2:
- introduce struct weston_point2d_device_normalized
- rename notify_touch_cal() to notify_touch_normalized()
- remove WESTON_INVALID_TOUCH_COORDINATE
Cc: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
v1 Tested-by: Matt Hoosier <matt.hoosier@gmail.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Relay touch input events into libweston core through the
weston_touch_device, so that the core can tell which individual physical
device they come from.
This is necessary for supporting touchscreen calibration, where one
needs to process a single physical device at a time instead of the
aggregate of all touch devices on the weston_seat.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
v1 Tested-by: Matt Hoosier <matt.hoosier@gmail.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Move calibration printing here and call do_set_calibration() from
evdev_device_set_calibration() so that all matrix setting paths print
the same way.
Print the matrix values in a matrix style to help readability, and
mention the input device.
v2:
- use 'cal' instead of 'calb' as variable name
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
v1 Tested-by: Matt Hoosier <matt.hoosier@gmail.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Introduce weston_touch_device for libweston core to track individual
touchscreen input devices. A weston_seat/weston_touch may be an
aggregation of several physical touchscreen input devices. Separating
the physical devices will be required for implementing touchscreen
calibration. One can only calibrate one device at a time, and we want to
make sure to handle the right one.
Both backends that support touch devices are updated to create
weston_touch_devices. Wayland-backend provides touch devices that cannot
be calibrated, because we have no access to raw touch coordinates from
the device - calibration is the responsibility of the parent display
server. Libinput backend provides touch devices that can be calibrated,
hence implementing the set and get calibration hooks.
Backends need to maintain an output pointer in any case, so we have a
get_output() hook instead of having to maintain an identical field in
weston_touch_device. The same justification applies to
get_calibration_head_name.
Also update the test plugin to manage weston_touch_device objects.
Co-developed by Louis-Francis and Pekka.
v2:
- Consistently use 'cal' instead of 'calb' or 'matrix'.
- change devpath into syspath
- update copyrights
Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
v1 Tested-by: Matt Hoosier <matt.hoosier@gmail.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Rather than segfaulting by attempting to traverse an initially
null log handler pointer, explicitly print a message and abort.
Signed-off-by: Matt Hoosier <matt.hoosier@gmail.com>
Reviewed-by: Ian Ray <ian.ray@ge.com>
[Pekka: coding style fix]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
The SURFACE_BITS_COMMAND struct has changed and some members have been moved in the
bmp field.
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Tested-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Allow global control of the pixman shadow buffers. The compositor can
choose whether all output use or do not use a shadow buffer with the
pixman renderer.
The option is added to the end of struct weston_drm_backend_config to
avoid bumping WESTON_DRM_BACKEND_CONFIG_VERSION.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Fabien Lahoudere <fabien.lahoudere@collabora.com>
Reviewed-by: Ian Ray <ian.ray@ge.com>
Add a flag to pixman-renderer for initializing the output with a shadow
framebuffer. All backends were getting the shadow implcitly, so all
backends are modified to ask for the shadow explicitly.
Using a shadow buffer is usually beneficial, because read-modify-write
cycles (blending) into a scanout-capable buffer may be very slow. The
scanout framebuffer may also have reduced color depth, making blending
and read-back produce inferior results.
In some use cases though the shadow buffer might be just an extra copy
hurting more than it helps. Whether it helps or hurts depends on the
platform and the workload. Therefore let the backends control whether
pixman-renderer uses a shadow buffer for an output or not.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Fabien Lahoudere <fabien.lahoudere@collabora.com>
Reviewed-by: Ian Ray <ian.ray@ge.com>
Pixman-renderer uses a single internal shadow buffer. It is enough to
composite the current damage into shadow, but the copy to hw buffer
needs to include the previous damage because of double-buffering in
DRM-backend.
This patch lets pixman-renderer do exactly that without compositing also
the previous damage on DRM-renderer.
Arguably weston_output should not have field previous_damage to begin
with, because it implies double-buffering, which e.g. EGL does not
guarantee. It would be better for each backend explicitly always provide
any extra damage that should be copied to hw.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Fabien Lahoudere <fabien.lahoudere@collabora.com>
Reviewed-by: Ian Ray <ian.ray@ge.com>
This issue was likely introduced by "libweston: add
weston_view_set_output()" which forgot to ensure the output destroy
listener is removed when weston_view is destroyed, leading to freed
memory being left into the list.
This was quite easy to trigger by opening and closing an application
window a few times, leading various memory corruption symptoms.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Ian Ray <ian.ray@ge.com>
Instead of desktop shell assigning view outputs directly,
use a new method, weston_view_set_output(). The method can
set up an output destroy listener to make sure that views
do not have stale output pointers.
Without this patch it is possible to end up in a scenario
where, e.g. configure_static_view() accesses memory that
has already been freed. The scenario can be provoked by
repeatedly plugging and unplugging a display. The faulty
memory accesses are reported by valgrind.
Signed-off-by: Semi Malinen <semi.malinen@ge.com>
Signed-off-by: Fabien Lahoudere <fabien.lahoudere@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Allow cloning up to 4 connectors from the same CRTC. All the
implementation bits support more than one head per output already.
Four is just an arbitary number, small but unlikely to ever be the
limiting factor in cloning since hardware is usually very restricted.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
For the attach on an enabled output to have an effect, we need to go
through drmModeSetCrtc or ATOMIC_ALLOW_MODESET.
v9:
- Add another XXX comment.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
When a head is detached from an enabled output, that output needs to go
through a modeset (drmModeSetCrtc() / ATOMIC_ALLOW_MODESET) so that the
connector is actually removed from the CRTC.
This has not yet been a problem, because an output could only have one
head at a time, and would be automatically disabled on detach. It would
be a problem with clone mode.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
If we are processing a connector that does not have an existing routing,
it is possible we pick a CRTC that was previously routed to a connector
we have not enabled yet. If that happens, the latter connector cannot
preserve its routing.
Check that no other connector we might enable later had this CRTC
before.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
To support shared-CRTC clone mode, the chosen CRTC needs to support
driving all the attached connectors. Replace the old algorithm with a
new one that takes into account all associated connectors.
Ideally it should use possible_clones mask to check which encoders (and
therefore connectors) actually can be in a cloned set. However, the DRM
documentation says about possible_clones and possible_crtcs masks both:
"In reality almost every driver gets this wrong."
- https://01.org/linuxgraphics/gfx-docs/drm/gpu/drm-kms.html#c.drm_encoder
Looking at a target device and its kernel where clone mode is desired,
possible_clones is indeed self-conflicting and would not allow cloning
at all. Therefore the implemented algorithm replaces the checking of
possible_clones with luck. It even goes out of its way to find any CRTC
for a configuration, even if not advertised by the kernel as not
supported.
Libweston would need infrastructure to allow trial-and-error CRTC
allocation: rather than picking one CRTC in advance and do or die, it
should try all available CRTCs one by one. Unfortunately that is not yet
possible, so this patch implements what it can. It is also the DRM
upstream opinion that trial-and-error with ATOMIC_TEST would be the way
to go.
Unlike the old algorithm, the new algorithm prefers routings that were
in place when Weston started instead of when enabling an output. When
you never temporarily disable an output, this makes no difference.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
compositor.c has 'touch', so use 'touch' here as well. It is not a
device to begin with.
Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
v1 Tested-by: Matt Hoosier <matt.hoosier@gmail.com>
We have weston_seat_{init,release}_{pointer,keyboard,touch}() as the
backend-facing API. There is no need to expose the create/destroy
functions which have been for internal use only for quite a long time.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
v1 Tested-by: Matt Hoosier <matt.hoosier@gmail.com>
If an input device is associated to an output that then gets disabled,
there is no case where associating to a different output would be
correct.
The output association is used for absolute positioned input devices,
and an input device like a touchscreen cannot ever be automatically
valid for more than one possible output - the touchscreen display
device.
Therefore do not automatically reassing implicitly associated input
devices to another output. This removes some log spam on shutdown.
In fact, if there can be more than one output at any time, absolute
input devices must be explicitly configured to associate with the
correct output, or the results are essentially undefined in any case.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
The use case driving this change is a clone mode setup, where the user
is hotplugging or unplugging a cloned touchscreen. Even if the output
and head are force-enabled, the touch device should still follow the
connector connection status. If there is no video signal for the
touchscreen (disconnected connector), then the touch input should be
ignored as well.
When the output is force-enabled, we need to trigger
output_heads_changed from connector status changes. If the head or
output are not force-enabled, the compositor will likely attach and
detach the head as appropriate. In clone mode, the attach or detach
needs to trigger output_heads_changed directly. In other cases, it may
be handled through the output getting enabled or disabled which are
different signals.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Associating input devices with weston_outputs by the output name fails
when one output has several heads. We need to match against head names
instead of output names to be able to find all names.
This fixes touchscreen output association in shared-CRTC clone mode when
outputs or input devices appear or disappear.
Even though notify_output_create() is called only when new outputs
appear, the implementation is prepared to also remove output
associations. This will be handy in the future when this function will
handle also head detaching from an output that remains enabled.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Removing the output association from an evdev_device is more than just
setting the pointer to NULL, one also needs to remove the destroy
listener and flag the destroy listener as unused (notify == NULL).
evdev_device_set_output() can already remove associations, so let it
also handle an assignment to NULL output.
Fix notify_output_destroy() to handle removing an association correctly.
Previously, the listener was left "used", which would mean the next call
to evdev_device_set_output() would have wl_list_remove()'d, accessing
freed memory. This could be triggered by having a touchscreen with a
specified output association, and unplugging then re-plugging the
corresponding output.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
In the future evdev_device_set_output() will start getting called more
often, redundantly. Short-circuit the setting if the chosen output is
already set for an input device. This reduces churn in the logs.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Daniel Stone
Ankit Nautiyal
Pekka Paalanen
Guido Günther
nerdopolis
Peter Hutterer
Tomohito Esaki
Louis-Francis Ratté-Boulianne
Matt Hoosier
David Fort
Semi Malinen