GitLab

As requested by Mike, update the E-mail address listed in the README.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>

DRM_FORMAT_MOD_INVALID means to derive the modifier from the dmabuf.
It provides legacy support and makes it easier to replace wl_drm.

v3: DRM_FORMAT_MOD_INVALID must be advertised to be supported (which
    requires a version bump)
v4: no version bump, but a note for now
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.com>
Reviewed-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Daniel Stone <daniels@collabora.com>

Signed-off-by: Sebastian Krzyszkowiak <dos@dosowisko.net>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.com>

Clarify that after zwp_buffer_release_v1 events, otherwise unused
buffers can be reused without any additional implicit synchronization.
This is in contrast to wl_buffer.release, which doesn't guarantee that
implicit synchronization is not required to safely use a buffer after
the event is received.
Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.com>

Graphics APIs are expected to use this protocol under the hood, and
since there can only be one user of explicit synchronization per
surface, warn about using the protocol directly in such cases.
Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.com>

Add opaque EGL buffers to the supported buffer types for use with the
explicit synchronization protocol. Opaque EGL buffers rely on the same
EGL implementation in both the compositor and clients, which makes it
straightforward to manage client expectations about fence support for
such buffers.

Also make it clearer that implementations are free to support other
buffer types beyond the required ones.
Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.com>

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>

This primary selection is similar in spirit to the eponimous
in X11, allowing a quick "select text + middle click" shortcut
to copying and pasting.

It's otherwise very similar to its Wayland counterpart, and
explicitly made consistent with it.
Signed-off-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Carlos Garnacho <carlosg@gnome.org>

This protocol enables explicit synchronization of asynchronous graphics
operations on buffers on a per-commit basis. Support is currently
limited to dmabuf buffers and dma_fence fence FDs.

Explicit synchronization provides a more versatile notification
mechanism for buffer readiness and availability, and can be used to
improve efficiency by integrating with related functionality in display
and graphics APIs.

This protocol is also useful in ChromeOS ARC++ (running Android apps
inside ChromeOS, using Wayland as the communication protocol), where it
can enable integration of the ChromeOS compositor with the explicit
synchronization mechanisms of the Android display subsystem.

Finally, the per-commit nature of the release events provided by this
protocol potentially offers a solution to a deficiency of the
wl_buffer.release event (see
wayland/wayland#46

).
Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
Reviewed-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Daniel Stone <daniels@collabora.com>
[Pekka: dropped Reveman from maintainers]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>

Although it would probably default to the license at the root of the
repository anyway, it's best to be explicit about it, and also be
consistent with the other extensions.

The copyright holders have been assembled from git history and the
README.
Signed-off-by: Johan Klokkhammer Helsing <johan.helsing@qt.io>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>

The wording in xdg-shell's `set_*` requests implies the compositor
*will* honour the client's request.
This would give clients the control over their actual state, while the
general expectation is that clients kindly ask for state changes which
the compositor may follow.
This patch ensures the actual protocol text reflects these expectations.
Reviewed-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>

The xdg-shell documentation had part of the maximized state render
implications in the `set_maximized` request documentation, not the
actual state.
This moves the relevant lines into the state description.
Signed-off-by: Markus Ongyerth <wl@ongy.net>
Reviewed-by: Mike Blumenkrantz <zmike@osg.samsung.com>
Reviewed-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>

This new protocol description is an evolution of v2.

- All pre-edit text styling is gone.
- Pre-edit cursor can span characters.
- No events regarding input panel (OSK) state nor covered rectangle.
  Compositors are still free to handle situations where the keyboard
  focus rectangle is covered by the input panel.
- No set_preferred_language request for clients.
- There is no event to send keysyms. Compositors can use wl_keyboard
  interface instead.
- All state is double-buffered, with specified defaults.
- The compositor can be notified about external changes to the state.
- The client can detect outdated requests.
Signed-off-by: Dorota Czaplejewicz <dorota.czaplejewicz@puri.sm>
Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Simon Ser <contact@emersion.fr>

Pass --strict to wayland-scanner in order to make it exit with failure
if something wasn't correct.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Simon Ser <contact@emersion.fr>

The wayland-scanner sub-commands private-code and public-code replaced
the old code command, so lets use those in the tests instead.

This requires at least wayland-scanner 1.15.0.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Simon Ser <contact@emersion.fr>

Signed-off-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Olivier Fourdan <ofourdan@redhat.com>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>

This adds a new protocol to negotiate server-side rendering of window
decorations for xdg-toplevels. This allows compositors that want to draw
decorations themselves to send their preference to clients, and clients that
prefer server-side decorations to request them.

This is inspired by a protocol from KDE [1] which has been implemented in
KDE and Sway and was submitted for consideration in 2017 [2]. This patch
provides an updated protocol with those concerns taken into account.
Signed-off-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Drew DeVault <sir@cmpwn.com>
Reviewed-by: David Edmundson <davidedmundson@kde.org>
Reviewed-by: Eike Hein <hein@kde.org>
Reviewed-by: Alan Griffiths <alan.griffiths@canonical.com>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>

[1] https://github.com/KDE/kwayland/blob/master/src/client/protocols/server-decoration.xml
[2] https://lists.freedesktop.org/archives/wayland-devel/2017-October/035564.html

It seems that this was partially done in
a3cf97ff

; this patch just corrects an
oversight.
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>

da331647

 added a compatiblity macro for
old versions of pkg-config. However, the file in which that macro
resides was not included. From the autoconf docs: "Note that if you use
aclocal from Automake to generate aclocal.m4, you must also set
ACLOCAL_AMFLAGS = -I dir in your top-level Makefile.am.".
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>

This adds two events to the protocol. The goal is to allow clients to
give the user the ability to select outputs with the same names the
compositor uses and to identify outputs consistently across sessions.
The output name is a short and stiff identifier with strict limits on
permitted characters, which is suitable for storing in config files,
command line arguments, etc. A warmer "description" event is also
provided to (optionally) provide a more human readable name, and has
much broader restrictions on its form.
Signed-off-by: Drew DeVault <sir@cmpwn.com>
Reviewed-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
[Jonas: Fixed formatting and commit subject]
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>

this adds implementation from a related discussion long ago in which
it was decided that it would be useful for clients to know if/where their
windows were tiled so that various behaviors and visuals could be modified
to improve UX

a window which is e.g., tiled on the right side of the screen would set the
right|top|bottom tiled states in configure
Signed-off-by: Mike Blumenkrantz <zmike@osg.samsung.com>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>
[Jonas: Minor formatting fixes]
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>

Changes since v2: simplified docs
Changes since v1: added since=2 to enum members

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>

Signed-off-by: Mike Blumenkrantz <zmike@osg.samsung.com>
Reviewed-by: Derek Foreman <derekf@osg.samsung.com>

wl_pointer, wl_keyboard and wl_touch events currently use a 32-bit
timestamp with millisecond resolution. In some cases, notably latency
measurements, this resolution is too coarse to be useful.

This protocol provides additional high-resolution timestamps events,
which are emitted before the corresponding input event. Each timestamp
event contains a high-resolution, and ideally higher-accuracy, version
of the 'time' argument of the first subsequent supported input event.

Clients that care about high-resolution timestamps just need to keep
track of the last timestamp event they receive and associate it with the
next supported input event that arrives.
Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>

Signed-off-by: Jonas Ådahl <jadahl@gmail.com>

Having a strict requirement on clients obeying the configured window
geometry for fullscreen toplevel surfaces might have the side effect of
making it harder or impossible to implement various hardware
optimizations on certain system configurations. By softening
requirements on the geometry while loosely defining the border fill, we
remove that restriction.

Clients that still want total control of the surrounding area can
still for example prepare the attached buffers to match the configured
surface size, or use subsurfaces in combination with wp_viewporter to
make up a surface matching the fullscreen window geometry dimensions.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Arnaud Vrac <rawoul@gmail.com>
Reviewed-by: Mike Blumenkrantz <zmike@samsung.com>

The description for xdg_toplevel.unset_fullscreen was completely
missing, so add it.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Mike Blumenkrantz <zmike@samsung.com>

It was not explicitly specified (as it is in set/unset_maximize) that
the compositor will respond with a configure event when a client asks to
be fullscreened, and the meaning of the output parameter was somewhat
awkwardly described.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Mike Blumenkrantz <zmike@samsung.com>

Specify that the maximize/unmaximize state requests only affects the
state a surface will return to if it is currently fullscreen.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Mike Blumenkrantz <zmike@samsung.com>

There is no configure 'request' only configure 'events'.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Mike Blumenkrantz <zmike@samsung.com>

While there is no currently known usages of setting an anchor offset on
the same axis as the 'flip' constraint action is set, it must still be
specified so compositors behave the same.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>

Bitfields allowed for impossible combinations of anchor edges, such as
being on the left and right edge. Use of explicit enumerations means we
don't need to handle that case.
Signed-off-by: David Edmundson <davidedmundson@kde.org>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>

ensure that this is as precise and explicit as possible for all useful
cases and also define previously-unspecified behavior
Signed-off-by: Mike Blumenkrantz <zmike@osg.samsung.com>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>

Allow using some other protocol (custom, or future xdg_* based) to set
up the parent-child relationship of a popup. This allows future
protocols to use xdg_popup when mapping popups over surfaces not based
on xdg_surface.

An example use case for this is the window menu, where a shells UI
client can use xdg_popup to create popup menus over windows it does not
have a xdg_surface of by having a custom protocol setting up the proper
parent-child relationship.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-By: Mike Blumenkrantz <zmike@osg.samsung.com>

Change the semantics of xdg_toplevel.set_parent to allow chaining
multiple parent-child relationships together, while allowing
arbitrarily unmapping parents, while keeping what is left over of the
chain intact.

This makes things easier to manage when parent-child relationships
cross client borders, for example when using xdg_foreign.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Signed-off-by: Mike Blumenkrantz <zmike@osg.samsung.com>

Setting a null-surface as a toplevel parent should unset the
parent-child relationship. This was not specified, so lets do that.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Signed-off-by: Mike Blumenkrantz <zmike@osg.samsung.com>