The `guess_candidates()` function scores each display that an input
device could be mapped to and then uses the `sort_by_score()` comparator
to find the best option. The function expects the list to be sorted from
best to worst, but the comparator currently sorts them in the opposite
order. This causes the function to end up returning the _worst_ match
rather than the the best. This commit reverses the sort order of the
comparator so that the best display can be returned as intended.
Closes: #1889
(Cherry-picked from commit 64ff1f20f81749b3b00ace5a8b73487747a8ed3d)
Right now we damage the stage even if an actor is not mapped, for
example in the overview.
Stop doing so, reducing over-paint significantly in some situations.
Clones will still do stage damage on their own.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2035>
(cherry picked from commit 136caca5d51e0ac8ff5e7b79aca32d6da2306d4f)
This unicode was initially meant to be b/w, but fonts and pango
eventually had another plan and we ended up with color glyphs
being used here, so we get strings like "⚫⚪⚪⚪". Change the
unicode used to ensure these are simpler b/w glyhps like
"● ○ ○ ○" that are easier to read on the OSD (and maybe even a
bit less ugly).
Fixes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/4733
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2064>
(cherry picked from commit 5125f66afa2927a368fc1ef1229b3bf86e81b303)
A clutter event's axes array is indexed by `ClutterInputAxis`.
However, a few lines accidentally use `ClutterInputAxisFlags` as
indices, reading incorrect values from elsewhere in memory. As a
result, broadcasted axis values for the tilt, rotation, and wheel
axes don't reflect actual event data.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1982
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2065>
(cherry picked from commit af6fb2a702fe3a7675b243dea7a210d635403cc2)
On ClutterInputFocus::reset, avoid to unset the preedit text if
none was set earlier. This seems to trick GTK clients into focusing
the cursor position again, even when we are moving away from it.
Fixes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/4647
(Cherry-picked from commit 3b6f9111c7da3fee82d3ffd19c444155f84d86ea)
Quoting the spec for `wl_data_device::drop`:
> If the resulting action is "ask", the action will not be considered
> final. The drag-and-drop destination is expected to perform one last
> wl_data_offer.set_actions request, or wl_data_offer.destroy in order
> to cancel the operation.
We did not respect the action choosen by the drop destination when
it called `wl_data_offer::set_actions` after `wl_data_device::drop`
if a user override was still active. This eventually resulted in
a protocol error in `wl_data_offer::finish`, as the current action
could still be `ask`.
Fix this by only allowing a user override to `ask` before `drop` is
called, thus making sure the final `set_actions` preference is
honored.
Closes https://gitlab.gnome.org/GNOME/mutter/-/issues/1952
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2043>
(cherry picked from commit 33b834c43373b715219d1b911813b59b2d0e2477)
With each wl_ouitput corresponding to a monitor, the logical monitor is
not part of the MetaWaylandOutput anymore.
Previously, send_xdg_output_events() would compare the old logical
monitor against the new one to determine whether the size and/or
position was changed and should be sent along with the xdg_output
events.
But that logic is now defeated as there is no old/new logical monitor
anymore, so the updated size or location would never be sent again.
Xwayland relies on this information to update its X11 clients and its
own internal root size, without this the X11 screen size and XRandR
information would never be updated.
To avoid that issue, always send the xdg_output size and location on
xdg_output events, Xwayland is smart enough to update its X11 clients
with XRandR only when the layout actually change.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1964
Fixes: bf7c3450 - Make each wl_output correspond to one monitor
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2050>
ClutterText implements its own get_paint_volume() with its own cache,
but was not invalidating the actor paint volume when when it has
changed. This sometimes could result in labels, especially quickly
changing ones, using the old paint volume which either would cut off the
label or leave parts of the old label on screen.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1943
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2006>
Easily seeing the buildtype and debug status can be quite useful.
Gnome-Shell already has it.
And while on it, add a 'Directories' section for already existing
entries.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2044>
meta_keymap_x11_replace_keycode currently reports to the X server
that the key types data is changed when adding a key to the keymap.
It's not changed. The number of key types is the same, and none of
them are modified.
This has two bad side effects:
1) It sends all of the key types data into the request
2) It hits a bug in the X server leading to the request getting
rejected entirely. See:
https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/761
Furthmore, the changed structure used to report to the X server
that the key types data is changed doesn't actually need to modified
at all in the function. It's already prepped by libX11 with the
correct state for the changes mutter is doing when
XkbChangeTypesOfKey is called.
This commit addresses the above two problems by just removing the
lines causing the issues.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2039>
Keys in the reserved keycode list are always added for the first group.
Before the previous commit such keycodes were not found unless that was
the current group. But now that we can also find matching keycodes that
are not directly in the current group, this is not necessary anymore.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1955>
Some keys, such as enter or backspace are only bound to a single group,
even if multiple groups are configured. Because the code was previously
only looking for keysyms in the same group as the current one, no
matching keycodes for these would be found if the current group is not
the first group. This was causing those keys to not work on the X11 OSK.
To fix this use the correct action to convert an out of range group for
that key according to its group_info field.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1955>
This effectively changes meta_keymap_x11_get_entries_for_keyval() to
meta_keymap_x11_get_entry_for_keyval() and moves the check if the
keycode maps to the keyval in the current group there. This simplifies
the code a bit and will allow a followup fix.
As a side effect this now also causes the reserved kecodes to be
searched, if no keycodes were found, rather than just when only ones
matching the wrong groups.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1955>
The extra stage update we schedule in `apply_state()` is mainly
needed in two situations:
- a partial update happened only in obscurred or off-screen parts
of a surface
- a surface requests frame callbacks without having done damage,
notably the (in)famous Firefox vsync implementation.
Commit 0330ce1f150f071620 limited the update to cases when the actor
was mapped, breaking it for Firefox in the overview.
Remove the mapped check again and get the stage from the backend,
restoring previous behaviour.
Fixes 0330ce1f150f071620
Closes https://gitlab.gnome.org/GNOME/mutter/-/issues/1957
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2034>
Avoid having laptops suspend or lock as soon as the power cable is
unplugged as the timeout for those actions when on battery are smaller
than the timeouts when on AC.
- laptop is plugged in, and hasn't been used for X minutes
- laptop is unplugged
- the gnome-settings-daemon power plugin sets up its timeouts for
inactivity for the "on battery" case
- those X minutes of inactivity are still counting, and are above
the level of one of the timeouts (say, suspend or lock screen),
mutter fires the timeouts
- gsd-power activates the action associated with the timeout
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1953
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2029>
If the ancestor a window is transient for has already been unmanaged
when the window is activated via meta_window_activate_full while its
transient_for property still points to that ancestor, this will cause
the already unmanaged ancestor to get added to the windows workspace.
This is after the ancestor had its workspace set to NULL when it was
unmanaged, causing this to look like an actual workspace change. Once
the window has been added to the workspace, it will never be removed
again, because the it has already been unmanaged. This confuses things
like the shell window tracker and leads to phantom windows being
considered present for apps that are not even running anymore.
Fixes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/4184
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2003>
When a test failed, an exception would be raised. This meant that the
mocked service would stay alive, and the test case being run eventually
failing due to a timeout, not the failure itself.
Fix this by catching the exception during the test, ensuring that we
tear down properly, then re-raise the same exception again after having
teared down.
This avoids the dead lock, while still printing the appropriate error
message.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2008>
With the introduction of MetaContext, the responsibility for handling
signals was changed to the application (e.g. GNOME Shell) using
libmutter. What wasn't fixed was making the stand-alone mutter do the
equivalent as well. This commit fixes this.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2007>
When the native backend is paused we still process the udev events
even though this isn't needed and may just cause unneeded events to be
triggered afterwards.
Since we'll resume with full changes on such event, we can just block
the signal hander when paused and restore it afterwards.
As per this we can cleanup also a bit the device adding signal handling
given that now we don't have to disconnect/reconnect it again.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1964>
Resume happens after we may have received various events that we've
ignored, so at this point we need to just emit an hotplug event like if
everything changed so that user settings may be re-applied.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1964>
On hotplug events we may get informations about what CRTC or connector
changed a property (and the property itself), so in such case let's just
ignore the changes to the non-affected CRTCs/connectors, and let's read
only the affected one
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1964>
On hotplug events we may receive a "CRTC" or "CONNECTOR" property that
indicates which crtc/connector property ID has changed.
In such case, instead of update data for all the devices, only update the
device containing the relative connector.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1964>
Hotplug events may contain CRTC or CONNECTOR ids to notify a property
change to just one owner, so we need to find its parent device.
Also we may want to update properties directly without having to go through
all the devices, so expose a simple way to find them.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1964>
In case we have no devices, after a KMS update (both because they've
all have been removed or because there were none), we may need to behave
differently compared to the case in which nothing changed, so add a more
specific KMS update change type
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1964>
If only gamma changed on drm CRTC's we don't have to rebuild the whole
monitors, nor to inform the backed about, the only consumer could be the
DBus API, and so we still emit a signal, but nothing else is needed.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1964>
Since we cache already all the KMS parameters we care about let's check at
each device update if anything has really changed and only in such case
emit a resources-changed signal.
In this way we can also filter out the DRM parameters that when changed
don't require a full monitors rebuild.
Examples are the gamma settings or the privacy screen parameters, that
emits an udev "hotplug" event when changed, but we want to register those
only when we handle the changed property.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1964>
A client request for maximizing itself should always be handled by mutter
by emitting a configure event with the native maximized resolution,
regardless of the client's own set limits. This also aligns the behavior by
allowing fixed-sized windows to go into fullscreen or maximized state.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1997>
These match their alpha counterparts, apart from not setting the
alpha bit. This allows our internal mashinery to more easily
distinguish whether we need a slow alpha-pass during rendering or not.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1810>
These match their alpha counterparts, apart from not setting the
alpha bit. This allows our internal mashinery to more easily
distinguish whether we need a slow alpha-pass during rendering or not.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1810>
It works correctly with scanouts, in contrast to
clutter_stage_capture_into. Inspired by
meta_screen_cast_area_stream_src_record_to_buffer.
maybe_paint_cursor_sprite is now unused and thus removed.
v2:
* clutter_stage_paint_to_buffer requires switching to recording from an
idle callback as well. (Jonas Ådahl)
v3:
* Set human readable name for idle source. (Ivan Molodetskikh)
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1940
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1914>
The way wl_seat capabilities work, by notifying clients of capabilities
changes, and clients consequently requesting the relevant interface
objects (pointer, keyboard, touch) is inherently racy.
On quick VT changes for example, capabilities on the seat will be added
and removed, and by the time the client receives the capability change
notification and requests the relevant keyboard, pointer or touch,
another VT switch might have occurred and the wl_pointer, wl_keyboard or
wl_touch already destroyed, leading to a protocol error which kills the
client.
To avoid this, create the objects when requested regardless of the
capabilities.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1797
Related: https://bugzilla.gnome.org/show_bug.cgi?id=790932
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/77>
So far, releases are done locally by invoking `meson dist`.
We can do better and leverage the existing CI infrastructure, to get
to the following release workflow:
- bump version in meson.build, update NEWS etc.
- open merge request for the release
- merge when the pipeline (including dist check) succeeds
- tag the release
- wait for the tag pipeline to spit out the tarball artifact
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1980>
I don't think this ever happened to me, but it can't hurt enforcing
that every release has a corresponding NEWS entry.
(The script has been copied from Polari, thus the metainfo support)
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1980>
Analogous to `get_image()` this returns a `ClutterContent` for a
given `MetaWindowActor`. This can be used to implement window
effects without a roundtrip from GPU to CPU memory.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1893>
In X11 when we switch to another tty all the the signals are blocked (as
the display fd is not replying back to polling, causing the main loop to
stop), and they are all handled once we switch back to the tty.
This is not a problem for most of external events, but in case of
accelerometer changes, once we reactivate a mutter session we'll get
them all together, causing lots of monitor reconfigurations leading to
black screen for some seconds and most of the times to a wrong
configuration being applied.
To avoid this, batch all these events using an idle to only apply the
last one we got in a loop.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1217
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1233>
Create a test system bus and use it to run all the tests, add a mock
SensorsProxy (via dbusmock template) server that implements the
net.hadess.SensorProxy interface.
To make testing easier, the service is created on request of a proxy for
it, whose lifetime controls the mock service lifetime as well.
This is done using a further mock service that is used to manage the
others, using python-dbusmock to simplify the handling.
Add basic tests for the orientation manager.
As per the usage dbusmock, we're now launching all the tests under such
wrapper, so that local dbus environment won't ever considered, and
there's no risk that it may affect the tests results both locally and in
CI.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1233>
When creating the configuration for the builtin monitor we try to get
the panel configuration for the builtin panel, but we don't proceed if
that monitor is currently inactive.
This is fine when adjusting an active configuration to the current
device rotation, but it isn't correct when we want to create a new
configuration based on another where the monitor is configured but not
yet enabled.
So, only find the panel configuration without looking the current state
but ensuring that the passed configuration will enable it.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1233>
When we get an orientation event we don't care about keeping track of the
configuration changes, but actually we can consider the new configuration
just a variant of the previous one, adapted to floating device hardware
events, so we only want to apply it if possible, but we don't want to keep
a record of it for reverting capabilities.
Doing that would in fact, break the ability of reverting back to an actual
temporary or persistent configuration.
For example when device orientation events happen while we're waiting for
an user resolution change confirmation, we would save our new rotated
configuration in the history, making then impossible to revert back to
the original persistent one.
So in such case, don't keep track of those configurations in the history,
but only keep track of the last one as current, checking whether the
new current is child or sibling of the previously one.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1221
Related to: https://gitlab.gnome.org/GNOME/mutter/-/issues/646
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1233>
When creating a configuration taking orientation into account we're using
the sensors orientation even if this is currently not used (for example
when an accelerator is available, but there's no touch screen).
This would cause to have a different behavior when configuration is
created and when we're loading a known configuration on startup.
So always honor whether the monitor's orientation is managed or not.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1233>
All the auto-rotation code is expecting to have a built-in panel, but we
still monitor accelerometer changes if we don't have one (uncommon, but
possible).
Thus manage the panel orientation in such case and update it on monitors
changes.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1233>
These are ClutterInputFocus subclasses, so this will trigger reset of
the input method. As the .done event is possibly deferred in the
zwp_text_input_v3 implementation, ensure the changes caused by the
reset are flushed immediately, before the button press is forwarded
to the client by MetaWaylandPointer.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1940>
This mode is passed along by the ClutterInputMethod, the
ClutterInputFocus will preserve it and ensure it is honored
whenever the IM is being reset.
This mode is immediate. The ClutterInputFocus commits the
text directly without queueing a CLUTTER_IM_COMMIT event.
This is important so events are serialized in the right order
in the wayland implementations (i.e. commit before wl_pointer.press).
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1940>
In line with GTK, the input method context should be reset when clicks
are handled by the ClutterInputFocus user. The reset action can then
either clear or commit the preedit text, as configured by the IM module.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1940>
This commit adds support to atomic KMS backend for optional plane property
prop_fb_damage_clips. Some drivers (e.g. EVDI) take advantage of this
property and process only updated regions of the screen instead of
processing the full frame. This can save system resources.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1879>
Added a function `meta_window_set_inactive_since` it sets
xattr on the cgroup directory for the given MetaWindow.
Resource management daemons can then monitor these changes on xattr
and make allocation decisions accordingly.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1960>
Currently the only way to get cgroup for a MetaWindow is to get it's
PID and perform a bunch of file accesses and string manipulations.
This is especially not feasible if we want to get the cgroup every
time a MetaWindow has gained or lost focus.
A solution to this is to cache the GFile for a cgroup path.
The creation and access of this GFile is handled by
`meta_window_get_unit_cgroup` function.
`meta_window_unit_cgroup_equal` is a utility function which allows
us to compare whether two MetaWindows belong to the same cgroup.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1960>
To utilize the API provided by libsystemd it would be better to
create a separate HAVE_LIBSYSTEMD configuration option instead of
having to rely on HAVE_NATIVE_BACKEND.
For now this will be utilized for getting the control group of a
MetaWindow.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1960>
and the subsurface was not previously detached from it using
`wl_subsurface_destroy()`.
Without 'window-actor/wayland: Remove subsurface actors on dispose' this
test would fail.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1958>
commit c4a73e795020722eda3e2bec0c16d96f9f37333b added
code to cleanup the renderer when the meta backend is
disposed. Unfortunately, this introduced a crash when
the window manager is replaced.
This is because cleaning up the renderer involves talking
to the X server over a display connection that's closed
two lines higher as part of the clutter_backend_destroy
call.
This commit fixes the crash by swapping their order.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1965>
There is very little point in sending an X11 client message to
gnome-panel in case gnome-shell isn't handling the binding. We
can just as well do nothing, so do exactly that.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1886>
Using "framebuffer_discard" in the list refers to an invalid extension.
The extension which introduces glDiscardFramebufferEXT is
"GL_EXT_discard_framebuffer".
With this fix, cogl_gl_framebuffer_fbo_discard_buffers can actually call
glDiscardFramebufferEXT.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1974>
The docu for `Popen.wait()` says:
> This will deadlock when using stdout=PIPE and/or stderr=PIPE and
> the child process generates enough output to a pipe such that it
> blocks waiting for the OS pipe buffer to accept more data.
Fixes 3caa5fea3c66cd860e333524ae1a62a195aeda1a
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1971>
If for some reason uncrustify gets angry at our file and indent on/off
marks, it will result in an error code other than 0. Since in those
cases there is no output to diff with, we misinterpret those situations
as "the whole file should be deleted", which is far from it.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1970>
Borrowed from Tracker, this script has different mode usages:
* --sha allows specifying a commit SHA, HEAD^ by default.
* --dry-run shows suggested changes in stdout, by default
the local tree is changed.
* --rewrite amends the last commit, in addition to performing
the changes.
This allows from simple style checks in the development tree,
automatically rewriting with the suggested changes (e.g. with
`git rebase --exec ./check-style -r`) and CI integration.
The script has per-function granularity, and works by re-styling
individually every commit, and inside every commit every chunk.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1924>
We fetch a frame clock that we schedule update on when queuing
_NET_WM_FRAME_DRAWN events. In some situations this frame clock is the
one from the stage, and if there are multiple hotplugs in a row, we
failed to update it as there were no stage views changes on the window
actor itself. As an actor updates the stage views list on layout, When a
queue_frame_drawn() call was done (typically from an X11 event) after a
second hotplug, it'd attempt to schedule an update on the frame clock
from the previous hotplug, as it didn't get notified about any
stage-views changes since for itself there was none.
Fix this by not caching the frame clock at all and just fetch it every
time.
In the majority of cases, this fetching means iterating over a very
short list (most often a single entry, rarely more), so it's very
unlikely to be of any relevance. The only situations where it might be a
heavier operation is the short time between a hotplug and a layout, as
it will attempt to traverse up to the stage to find a clock, but that's
likely only a few levels, so even that is unlikely to be an issue.
Closes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/4486
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1947>
Make sure that when we've recreated views that we'll actually paint a
new frame for it. This was very rarely a problem, as views tend to
result in getting damage etc being queued as side effects of various
things, like layout, but e.g. when running certain tests, this might not
happen. There is no situation where we want to create a new view that
should remain unpainted, so just make sure we initialize it to become up
to date.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1947>
This is more in line with the protocol, and allows us to remove some
awkward code that tries to "combine" different metadata from different
monitors into one, which sometimes meant picking an arbitrary "main"
monitor, or "and" metadata together to find a common ground.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1712>
Makes workspace transitions in gnome-shell look more seamless, since
both outgoing and incoming workspace have focused windows.
This is only done for click focus mode, since it's not known which
window would be focused for the other modes.
Track the state and recompute it when it changes, to avoid redrawing
the windows needlessly.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/850>
Each workspace has a window that will be focused when switching to that
workspace. Add a function to retrieve that window.
This is only relevant for click-to-focus focus mode, since with the two
other modes no window will be focused upon switching, and will only gain
focus when hovered.
This will be used in the next commit to make this window appear focused.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/850>
This was introduced by accident in commit 1467b6b02a73bca76
y-inverted textures in combination with shape masks appear to
be only commonly used with EGLstreams. However, as we draw the
shape mask ourselves, we don't want to apply the y-invert to it
as testified by the left over `cogl_pipeline_set_layer_matrix()`.
Note that we still allow to apply viemports and buffer transforms,
as the Xwayland mode setting emulation may use it (in fact only
the former, but it probably does not hurt to leave the later as well).
Closes https://gitlab.gnome.org/GNOME/mutter/-/issues/1792
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1937>
Monitor configuration check tests can be very complex and in case of
failures we can't easily catch where a failure happened without entering
in debug mode, something that isn't always an option in CI or external
builders.
So add more debug statements in configuration check functions and use
macros to ensure that we print the caller function and location on more
complex check functions.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/522>
Primary monitor is just the same of the other monitors, but it has a
primary monitor flag. Since the computation of the scaling isn't
dependent anymore on the computed configuration we can now generate the
primary monitor config together with the others.
However, we've to ensure that the primary monitor is the first of the
configs list in order to properly compute the positioning.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/522>
Compute the monitor scaling in a separated function using the primary
monitor (not its config) and pass it to the creation function instead.
This will allow removing the special logic for the primary monitor.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/522>
Factorize the creation of a configuration inside one function that looks for
the primary monitor and the other monitors using the matching rules and
dispose them according to the chosen policy (checking if the result is valid
when using the suggested positioning).
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/522>
Add a find_monitors function that allows to search for monitors that match
the MonitorMatchRule filter and use this to look for the primary monitor and
the other monitors that need to match the requested filter in order to be
configured.
Having just one function doing this kind of checks reduces the possibility
of unexpected results.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/522>
Verify that the suggested monitor configuration contains only adjacent monitors,
and that if this is not the case we fallback to the linear configuration.
This can happen in case of multi-DPI setup, so add a test checking this too.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/522>
It could happen that monitors suggest to use coordinates that don't take
in consideration the scaling applied to one monitor, and such the
generated configuration is not valid because not all the monitors are
adjacent.
So enforce this check before accepting a suggested configuration as it
is.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/522>
We may need to check if rectangles region has adjacent neighbors and
so if there are no gaps in between monitors.
This can be done by checking if each monitor is adjacent to any other in
the same region.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/522>
This code sneaked unconditionally, even though we can disable
tracing code with -Dprofiler=false. Add some COGL_HAS_TRACING
checks so that this code is also optionally built.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1951>
This code sneaked unconditionally, even though we can disable
tracing code with -Dprofiler=false. Add some COGL_HAS_TRACING
checks so that this code is also optionally built.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1951>
This code sneaked unconditionally, even though we can disable
tracing code with -Dprofiler=false. Add some COGL_HAS_TRACING
checks so that this code is also optionally built.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1951>
Will be used to trace a lot more, and with more details, and thus may
have a larger impact on what is actually measured. This potential impact
is the reason for enabling only when needed.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1700>
A trace "anchor" is a trace head (CoglTraceHead) that is placed in a
certain scope (e.g. function scope), but then only triggered in another
scope, e.g. a condition.
This makes it possible to have optional trace instrumentation, that is
enabled only given e.g. a debug flag.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1700>
The failure to allocate was not properly handled, causing crashes later
on due to the offscreen being NULL.
#0 cogl_gl_framebuffer_bind (target=36160, gl_framebuffer=0x0)
#1 _cogl_driver_gl_flush_framebuffer_state (...)
#2 cogl_context_flush_framebuffer_state (read_buffer=0x55f48f386780, draw_buffer=0x55f48f386780, ...)
#3 cogl_framebuffer_clear4f (framebuffer=0x55f48f386780, ...)
#4 clutter_layer_node_pre_draw (...)
#5 clutter_paint_node_paint (...)
...
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1942>
When a selection owner advertises a mime type, but does not provide the
content upon a request for the mime type content, the requesting side
might wait indefinitely on the content.
To avoid this situation, add a timeout source, which will cancel the
selection transfer request after a certain timeout (15 seconds) passed.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1874>
Currently, if g-r-d closes the read end of the pipe for a
SelectionRead() operation, due to realizing that the application, that
should provide the mime type content, does not provide any content,
mutter won't notice that and still assumes that the read() operation
on the pipe in g-r-d is still happening, as mutter never writes to the
pipe in that situation and therefore cannot realize that the pipe is
already closed.
The effect of this is, that if g-r-d aborts a read() operation and
requests a new read() operation via SelectionRead(), mutter will deny
the request since it assumes that the previous read() operation is
still ongoing.
Fix this behaviour by also checking the pipe fd in mutter before
denying a SelectionRead() request.
https://gitlab.gnome.org/GNOME/gnome-remote-desktop/-/issues/60
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1874>
With some resolutions (such as 4096x2160) we may compute duplicated
scale factors because we used a too wide threshold to check for an
applicable value.
In fact, while when we're at the first and last values it's fine to
search applicable values up to SCALE_FACTORS_STEP, on intermediate ones
we should stop in the middle of it, or we're end up overlapping the
previous scaling value domain.
In the said example in fact we were returning 2.666667 both when
looking to a scaling value close to 2.75 and 3.00 as the upper bound of
2.75 (3.0) was overlapping with the lower bound of 3.0 (2.75).
With the current code, the lower and upper bounds will be instead 2.875.
Adapt test to this, and this allows to also ensure that we're always
returning a sorted and unique list of scales (which is useful as also
g-c-c can ensure that this is true).
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1878>
We introduced META_MONITOR_SCALES_CONSTRAINT_NO_FRAC to get global scale
values however, this didn't work properly for some resolutions.
In fact it may happen that for some resolutions (such as 3200x1800) that
we did not compute some odd scaling levels (such as 3.0) but instead
its closest fractional value that allowed to get an integer resolution
(2.98507452 in this case).
Now this is something relevant when using fractional scaling because we
want to ensure that the returned value, when multiplied to the scaled
sizes, will produce an integer resolution, but it's not in global scale
mode where we don't use a scaled framebuffer.
So, take a short path when using no fractional mode and just return all
the applicable values without waste iterations on fractional values.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1878>
Scaling values computation code served us well in the past years but
it's quite delicate and it has some issues in edge cases, so add a test
that verifies that the computed scaling values for all the most common
resolutions (and some that may be common in future) are what we expect
to be.
This may also serve us in future when we'd define a better algorithm to
compute the preferred scale, but this not the day.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1878>
When deriving the global scale from current monitor, we were just checking the
supported value by the primary monitor, without considering weather the current
scale was supported by other monitors.
Resolve this by checking if the picked global scale is valid for all active
monitors, and if it's not the case, use a fallback strategy by just picking the
maximum scale level supported by every head.
Fixes https://gitlab.gnome.org/GNOME/mutter/issues/407
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/336>
In Xrandr we were caching the available scaling modes that were computed just
for the current mode, for each monitor, while we can actually reuse the
default implementation, by just passing the proper scaling constraint.
In monitor we need then to properly filter these values, by only accepting
integer scaling factors that would allow to have a minimal logical monitor
size.
Fixes https://gitlab.gnome.org/GNOME/mutter/issues/407
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/336>
We only listen to it for 2 settings (drag threshold, double click
time), and we already have the stock ClutterSettings object tracking
the source of these. This code is redundant.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1862>
Seems glGetString(GL_RENDERER) in the wild can return NULL, causing
issues with strstr(). Handle this more gracefully by using
g_return_val_if_fail(), that assumes a NULL renderer means software
rendering.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1931>
This will require some symbol exporting, but the benefit is that have
better control of what external test cases can do in terms of creating
more testing specific contexts.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
This makes it possible to declare the type in an installed header (so
that e.g. META_CONTEXT_TEST(context) works), but without having to
expose the MetaContextClass struct.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
Add a method meta_context_destroy() that both runs dispose and unrefs
the context. Tear down is moved to dispose() so that things owned by the
context are destroyed when calling meta_context_destroy(), or when the
last reference is released.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
Before we first created the MetaWaylandCompositor instance, which would
repare Clutter/Cogl so they could initialize and turn on Wayland display
server features, then later to initialize the rest. Now that part is
done by the Wayland infrastructure itself, so we don't need the early
initialization. Simplify things a bit by centralizing it all into a
single meta_wayland_compositor_new() call.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
This is done by keeping around a pointer to MetaContext as
"client_pointer" (which is practically the same as "user_pointer"
elsewhere), as well as creating a `MetaIceConnection` wrapper for ICE
connections.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
As with the compositor type enum, also have the X11 display policy enum,
as it's also effectively part of the context configuration. But as with
the compositor type, move it to a header file for enums only, and since
this is a private one, create a private variant meta-enums.h.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
Since this tests the `--virtual-monitor` command line argument, it uses
the `MetaContextMain` variant of the context, as it's there that command
line argument is handled.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
This object intends to replace the scattered functions that are used to
make up what is effectively a "mutter context". It takes care of the
command line arguments that is now done in main.c, persistant virtual
monitors, and the like.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
The clutter tests neeed to start and stop, thus uses their own main loop
instead of the one in MetaContext. Shouldn't matter, since nothing
in mutter should happen that makes the test self-terminate from inside
mutter.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
Users can add option entries, and it'll be part of the configuration
phase.
Create the main group manually to be able to set a user_data pointer;
this will be required to not have to rely on globals when parsing
options using a callback.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
This intends to replace the call to `meta_register_with_session()` that
deals with X11 session management, and is called when the user is
"ready". In thet test context, doing that makes no sense, so make it a
no-op.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
The start phase creates the MetaDisplay object, and initializes Wayland, and
creates the main loop.
The run phase runs the main loop and handles returning an error if the
context was terminated with an error.
The terminate phase terminates the main loop, with or without an error.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
Configuration is the first step of the lifetime of a context, after
creation; it's here where argc/argv is processed, and it's determined
what kind of compositor, etc, it is going to be.
The tests always run as Wayand compositors, so the configuration is
quite simple, but will involve more steps later on.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
It'll be part of and owned by MetaContext, intending to replace
`meta_is_wayland_compositor()`, but place it in a new file for public
enums so that it can be used from wherever.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
This introduces a MetaContext implementation aimed to be used for test
cases, with as little boiler plate as possible needed in the test.
It currently doesn't do anything, just fills out the GObject boiler
plate and sets a name.
Build it into every core test, for compilation, even though it isn't
used anywhere yet.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
This type is intended to replace the scattered functions used to
configure how the Mutter compositor is run. It currently doesn't do
anything, and only has a human readable name, intended to be set to e.g.
"GNOME Shell".
It's an abstract type, and is intended to be used via either a future
`MetaContextMain` for real display server use cases, and a
`MetaContextTest` for test cases.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1861>
This job does:
1. Download the coverity bundle and untar it
2. Build mutter using clang and the coverity tool
3. Compress the coverity report
4. Upload for analysis
Things to note:
- Analysis are throttled, as per https://scan.coverity.com/faq#frequency
we qualify for 21 weekly builds, 3 daily. Mutter is sometimes a busy
project, so it seems we'd get often those consumed early in the day.
This is something we can resign to, but the times we'll try to upload
a report to have it rejected make the operation kinda pointless and
probably better throttled by ourselves.
- The task is manual, given the restrictions above.
- The task only applies on master, as the envvar holding the coverity
token is protected in gitlab.
- I had to use clang as the coverity tool doesn't seem to work ATM with
gcc as per recent Fedora.
- The coverity tarball is 1.2GB in size, which is a bit too big to have
it downloaded each time. As per their upload instructions, the tarball
gets updated twice yearly, so this is cached to minimize downloads.
- The coverity token for mutter is kept private/hidden in gitlab CI
settings.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1100>
Not sure how to update the damage or redraw clip or something; at least
this works properly when under a constantly-redrawing window, which is
ok for debugging purposes.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1762>
Max render time shows how early the frame clock needs to be dispatched
to make it to the predicted next presentation time. Before this commit
it was set to refresh interval minus 2 ms. This means Mutter would
always start compositing 14.7 ms before a display refresh on a 60 Hz
screen or 4.9 ms before a display refresh on a 144 Hz screen. However,
Mutter frequently does not need as much time to finish compositing and
submit buffer to KMS:
max render time
/------------\
---|---------------|---------------|---> presentations
D----S D--S
D - frame clock dispatch
S - buffer submission
This commit aims to automatically compute a shorter max render time to
make Mutter start compositing as late as possible (but still making it
in time for the presentation):
max render time
/-----\
---|---------------|---------------|---> presentations
D----S D--S
Why is this better? First of all, Mutter gets application contents to
draw at the time when compositing starts. If new application buffer
arrives after the compositing has started, but before the next
presentation, it won't make it on screen:
---|---------------|---------------|---> presentations
D----S D--S
A-------------X----------->
^ doesn't make it for this presentation
A - application buffer commit
X - application buffer sampled by Mutter
Here the application committed just a few ms too late and didn't make on
screen until the next presentation. If compositing starts later in the
frame cycle, applications can commit buffers closer to the presentation.
These buffers will be more up-to-date thereby reducing input latency.
---|---------------|---------------|---> presentations
D----S D--S
A----X---->
^ made it!
Moreover, applications are recommended to render their frames on frame
callbacks, which Mutter sends right after compositing is done. Since
this commit delays the compositing, it also reduces the latency for
applications drawing on frame callbacks. Compare:
---|---------------|---------------|---> presentations
D----S D--S
F--A-------X----------->
\____________________/
latency
---|---------------|---------------|---> presentations
D----S D--S
F--A-------X---->
\_____________/
less latency
F - frame callback received, application starts rendering
So how do we actually estimate max render time? We want it to be as low
as possible, but still large enough so as not to miss any frames by
accident:
max render time
/-----\
---|---------------|---------------|---> presentations
D------S------------->
oops, took a little too long
For a successful presentation, the frame needs to be submitted to KMS
and the GPU work must be completed before the vblank. This deadline can
be computed by subtracting the vblank duration (calculated from display
mode) from the predicted next presentation time.
We don't know how long compositing will take, and we also don't know how
long the GPU work will take, since clients can submit buffers with
unfinished GPU work. So we measure and estimate these values.
The frame clock dispatch can be split into two phases:
1. From start of the dispatch to all GPU commands being submitted (but
not finished)—until the call to eglSwapBuffers().
2. From eglSwapBuffers() to submitting the buffer to KMS and to GPU
work completing. These happen in parallel, and we want the latest of
the two to be done before the vblank.
We measure these three durations and store them for the last 16 frames.
The estimate for each duration is a maximum of these last 16 durations.
Usually even taking just the last frame's durations as the estimates
works well enough, but I found that screen-capturing with OBS Studio
increases duration variability enough to cause frequent missed frames
when using that method. Taking a maximum of the last 16 frames smoothes
out this variability.
The durations are naturally quite variable and the estimates aren't
perfect. To take this into account, an additional constant 2 ms is added
to the max render time.
How does it perform in practice? On my desktop with 144 Hz monitors I
get a max render time of 4–5 ms instead of the default 4.9 ms (I had
1 ms manually configured in sway) and on my laptop with a 60 Hz screen I
get a max render time of 4.8–5.5 ms instead of the default 14.7 ms (I
had 5–6 ms manually configured in sway). Weston [1] went with a 7 ms
default.
The main downside is that if there's a sudden heavy batch of work in the
compositing, which would've made it in default 14.7 ms, but doesn't make
it in reduced 6 ms, there is a delayed frame which would otherwise not
be there. Arguably, this happens rarely enough to be a good trade-off
for reduced latency. One possible solution is a "next frame is expected
to be heavy" function which manually increases max render time for the
next frame. This would avoid this single dropped frame at the start of
complex animations.
[1]: https://www.collabora.com/about-us/blog/2015/02/12/weston-repaint-scheduling/
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1762>
Scanout doesn't go through the usual path of compositing and doing
eglSwapBuffers, therefore it doesn't hit the timestamp query placed in
that path. Instead, get the timings by binding the scanout buffer to an
FBO and doing a timestamp query on the FBO.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1762>
Add utilities that allow getting the current GPU timestamp and creating
a query which completes upon completion of all operations currently
submitted on a framebuffer. Combined, these two allow measuring how long
it took the GPU to finish rendering something to a framebuffer.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1762>
This drops some custom building of various components that are now up to
date. While at it, start using the FDO_DISTRIBUTION_PACKAGES variable to
install packages, as it with the bumped ci-templates version also
doesn't install weak dependencies.
This also requires tweaking the pipewire dead lock work arounds, as it
changed configuration file paths.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1865>
This fixes a warning/error:
In function 'parse_settings',
inlined from 'read_settings' at ../clutter/clutter/x11/xsettings/xsettings-client.c:398:25:
../clutter/clutter/x11/xsettings/xsettings-client.c:202:13: error: 'buffer.byte_order' may be used uninitialized [-Werror=maybe-uninitialized]
202 | if (buffer.byte_order != MSBFirst &&
| ~~~~~~^~~~~~~~~~~
This is needed to bump the CI image from F33 to F34, which includes a
upgraded compiler.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1865>
In order to make it possible to e.g. unload an unused DRM device, we
need to make sure that we don't keep the file descriptor open if we
don't need it; otherwise we block anyone from unloading the
corresponding module.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1828>
We need to call eglBindAPI() with GLES before we setup the secondary
GPU blit. We've been lucky not really needing this, as it has been
GLES default, which is what the secondary blit uses, in order to not
depend on the default, or if we want to create the secondary blit
objects after initializing cogl, we must make sure to bind the right API
at the right time.
As we need to bind the GLES API when setting up the secondary blit, we
need to make sure that cogl gets the right API bound when that's done,
so Cogl can continue working. For this, add a "bind_api()" method on the
CoglRenderer object, that will know what API is correct to bind.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1828>
The DRM buffers aren't really tied to mode setting, so they shouldn't
need to have an associated mode setting device. Now that we have a
device file level object that can fill this role, port over
MetaDrmBuffer and friends away from MetaKmsDevice to MetaDeviceFile.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1828>
Keep a private MetaDeviceFile instance for the GPU's managed by the
renderer. This is a step towards decoupling rendering from mode setting,
as well as on-demand holding of device file descriptors.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1828>
Tags are meant to make it possible for a device file opener to tag a
file if it has affected the state the file descriptor is in; e.g. if it
has enabled a DRM capability.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1828>
Handle open() failing due to being interrupted by trying again until it
either succeeds, or fails due to some other error. This was an error
handling path taken when opening sysfs files; do the same here to not
potentially regress once we open sysfs files with the device pool.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1828>
It's only when we take/release from/to logind we need these two
integers, so only retrieve them when that's done. Making this change
makes it possible to open devices that don't have these parameters.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1828>
This changes the way the KMS backends load; if we're headless, we always
use the dummy one and fail otherwise; in other cases, we first try the
atomic backend, and if that fails, fall back on the simple one.
The aim for this is to have the impl device open and close the device
when needed, using the device pool directly.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1828>
This practically does the same thing as part of MetaLauncher, except
with added thread safety and caching. For example, opening the same file
a second time will return the same MetaDeviceFile, and only once all
acquired MetaDeviceFile's are released, will the file descriptor be
closed and control of the device released.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1828>
ClutterInputDevice's get_group_n_modes() vfunc is meant to return
-1 for groups that are out of the known range, not within. Fix the
early return condition, and let the native backend return correctly
the number of modes for the given group.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1920>
Scanouts are taken away after painting. However, when we're
streaming, what we actually want is to capture whatever is
going to end up on screen - and that includes the scanout
if there's any.
Add a before-paint watch that only records new frames if a
scanout is set.
Inspired by (and commit log mostly copied from) e6a13e5d5774
("monitor-stream-src: Add before-paint watch to catch scanouts").
v2:
* Do not call stage_painted from before_stage_painted (Georges Basile
Stavracas Neto)
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1707
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1902>
When the MetaWindow resize machinery for toplevels ended up in the
Wayland window implementation, we tried to avoid configuring
not-yet-mapped windows that just had its zero sized dimension pass
through the constraint machinery, resulting in a 1x1 sized window.
If we'd properly set up the min size metadata earlier, that 1x1 would
likely be the minimum size set of a window, which makes things harder to
predict when peeking at side effects.
However, what the side effect peeking intends to do, as documented in
the comment, was to figure out when the client hadn't committed any
buffer yet, i.e. during the initial map, and in those cases avoid
sending that nasty 1x1 size, resulting in silly window sizes. A more
robust way to detect this is instead checking when we shouldn't really
try resize things our own way, and in those cases early out as was done
before.
This means that, for a yet to me mapped window, we only ever want to
send an initial non-zero configuration when 1) it's initially maximized,
2) initially fullscreen, or 3) initially tiled in any way, as those are
the situations where the compositor is the one deciding the size.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1912>
When drmModePageFlip() or drmModeAtomicCommit() unexpectedly failed (e.g.
ENOSPC, which has been seen in the wild), this failure was not handled
very gracefully. The page flip listener for the scanout was left in the
MetaKmsUpdate, meaning when the primary plane composition was later page
flipped, two page flip listeners were added, one for the primary plane,
and one for the scanout. This caused the 'page-flipped' event to be
handled twice, the second time being fatal.
Handle this by making 'no-discard' listener flag be somewhat reversed,
and say 'drop-on-error', and then drop all 'drop-on-error' listeners
when a MetaKmsUpdate failed to be processed.
Also for a "preserve" flagged update, don't ever trigger "discard"
callbacks just yet, as preserved updates are used again for the primary
plane composition, in order to not miss e.g. CRTC gamma updates, or
cursor plane updates, which were added separately.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1809
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1910>
A simply wrapper around `CoglTexture`, making it easy to reuse
content without roundtrip from GPU to CPU memory and back.
It optionally takes a clip rectangle which is implemented by
creating a `CoglSubTexture`. A limitation here is that floating
point clips are not supported.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1888>
The cancel phase for swipe gestures was not being handled, hence,
Wayland "end" events where not sent to clients when the gesture was
cancelled.
A swipe gesture is cancelled when extra finger(s) are put down on the
touchpad in the middle of the gesture or when some, but not all, of the
fingers are put up.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1857>
There is an udev rule marking whether a device should be ignored by
mutter or not, but it was only respected on hotplug events not on init,
partly defeating its purpose. Fix this.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1892>
Getting crossing events is necessary between client surfaces while
there is a popup grab in effect (e.g. allow press-drag-release in
menus), we should only stick with the focus surface while the pointer
is outside any client surface.
This partially undoes commit 79050004b06 (or, at least, mutter no
longer fixes the bug it claimed to fix). This will be addressed in
gtk4.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1885>
A view is only a 'CoglOnscreen' if it ends up on a CRTC, thus needs a
mode. Other views are for virtual monitors, and require no mode setting,
so exclude them from the pending mode set list.
This avoids a dead lock when we'll be waiting indefinitely for mode
setting on a virtual monitor.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1891>
When using `CLUTTER_PAINT=damage-region` highlighting was conspicuously
absent during fullscreen animations like entering or leaving the
overview. That was because `queued_redraw_clip` was empty, because it
had been initialized from `redraw_clip == NULL` (full stage redraw).
Now we paint the damage region as the full view (which it is) instead
of nothing at all.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1890>
Just like we do on EGL. Two bits are required because
`cogl-clip-stack-gl.c` needs each stencil buffer element to be able to
count from 0 to 2.
This mistake probably went unnoticed because:
* Drivers usually provide more than 1 anyway; and
* Optimizations in `cogl-clip-stack-gl.c` avoid calling the code that
needs to count past 1 in most cases.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1873>
Previously we were using a mask of 0x1 for the lifetime of the stencil.
This was wrong for two reasons:
* The intersection algorithm needs to count up to a maximum 2, so a
mask of 1 would clamp to 1 instead. Then decrementing all pixels
resulted in all pixels being zero even though we want some to be 1.
So the stencil then blocked some color buffer pixels being rendered.
* The lifetime of the mask was too long. By leaving it non-zero at
the end of the function we could accidentally end up modifying the
stencil contents during our later color buffer paints.
This fixes faulty rendering of some actors seen in gnome-shell with
test case: `env COGL_DEBUG=stencilling`
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1873>
Previously we were using a mask of 0x1 for the lifetime of the stencil.
This was wrong for two reasons:
* The intersection algorithm needs to count up to a maximum 2, so a
mask of 1 would clamp to 1 instead. Then decrementing all pixels
resulted in all pixels being zero even though we want some to be 1.
So the stencil then blocked some color buffer pixels being rendered.
* The lifetime of the mask was too long. By leaving it non-zero at
the end of the function we could accidentally end up modifying the
stencil contents during our later color buffer paints.
This fixes missing rendering of some actors seen in gnome-shell with
test case: `env COGL_DEBUG=stencilling CLUTTER_PAINT=disable-clipped-redraws`
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1873>
The mutter tests require to run in a valid environment where a display
is available and a session bus, however currently we rely on the current
environment, and this may lead to unexpected behaviors.
So let's just ensure that a display is running through xvfb-run and
that a session bus is running in a temporary directory.
We also ensure to use the gsettings memory backend, even because by
setting TestEnvironment we ensure that no other env variable is leaked
to the test.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1876>
The cogl tests need to run with a display server set, however since we
use TestEnvironment, only the listed env variables will be exposed to
the test and so no DISPLAY will be set when launching it with
gnome-desktop-testing-runner.
As per this, just run the tests using xvfb-run so that we match what's
happening in CI and we ensure that the tests are run in a safe
environment.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1876>
When running multiple tests at once (with --all) as in the
installed-tests cases, we may open and close the display multiple times,
this leads to setting the alarm filter each time that the x11 display is
opened (causing a critical error) because we never disconnect from the
::x11-display-opened signal.
So disconnect from the signal on test destruction, to avoid this to be
emitted multiple times.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1876>
We first initialized the Wayland infrastructure, then the display, but
on shutdown, we first teared down the Wayland infrastructure, then the
display.
Make things a bit more symmetric and tear down the display before
Wayland. This however means we need to tear down some things Wayland a
bit earlier than the rest. For now this is a separate function, but
eventually, it can be replaced with a signal shared by the backend's
'prepare-shutdown' signal.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1863>
This way we can initialize without having any way to retrieve it via
some global variable. This isn't needed now, but will be once Wayland
infrastructure initializiation is done in a single step.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1863>
This is especially important as we might end up here when initializing
the Wayland infrastructure. Later that will be done in one step, meaning
the "get_default()" function will not work properly.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1863>
It might not be there when shutting down, so get it from a more managed
place. Note that this isn't strictly needed right now, but eventually,
the MetaWaylandCompositor pointer will be cleared using a g_clear*()
helper, which clears the pointer before freeing the instance, which
wouldn't work here.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1863>
This will allow us less awkward startup, where previously we had to
pre-initialize Wayland very early so Cogl could bind the Wayland display
when it initialized. Move things around so we bind the Wayland EGL
display when initializing the rest of Wayland infrastructure.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1863>
This object takes over the functionality of meta-idle-monitor-dbus.c,
meta-idle-monitor.c and meta-backend.c, all related to higher level
management of idle watches etc.
The idle D-Bus API is changed to be initialized by the backend instead
of MetaDisplay, as it's more of a backend functionality than what
MetaDisplay usually deals with.
It also takes over the work of implementing "core" idle monitors. The
singleton API is replaced with thin wrapper functions on the backend.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1859>
It just complicates things; we can't run them right now, so just get rid
of the runtime variability; just change the macros if you want to tweak
the test, would you be able to get it running.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1833>
Wayland support is not really a "backend" thing, it just lacked a better
place to store its instance pointer. Eventually we'll have a better
place, but prepare for that by initializing it together with the more
similar subsystems.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1833>
The rest of debug flag details are in util.c and util.h, make things
less scattered by moving the rest to util.c too.
While at it, put the coredump:ability setting needed for being suid
there too, so we have a common place for initializing "debug utils".
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1833>
Commit 64c9c9c5b005cf17103d8e1e0a08dec0b1094267 fixed monitor
screencasting, when fractional screencasting is enabled.
For the remote desktop usage, NotifyPointerMotionAbsolute() submits
the new mouse pointer position in addition to the stream, where the
mouse pointer was moved.
When not using fractional scaling, the mouse pointer position is
correct.
With the usage of fractional scaling, the mouse pointer position is
wrong, as the scale of the position is applied two times.
Fix this behaviour, by reverting the second scale by dividing by the
logical monitor scale, when fractional scaling is used.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1808
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1867>
When finalizing, the MetaDisplay instance will already be gone, so to be
able to gracefully tear down the clipboard integration, make sure to
close sessions before the display is closed, i.e. on prepare-shutdown.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1853>
The callback on_displayfd_ready() would unconditionally set the return
value to TRUE, regardless of the condition.
Use the GIO condition to determine if there was data written and adjust
the return value accordingly.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1851>
Some clients such as mpv might set the WM_TITLE as a UTF8_STRING based
on some unconverted/unvalidated metadata that is not actually UTF8. This
would then be set as the title of a MetaWindow (in the absence of a
valid UTF8 _NET_WM_TITLE). The shell then tries to use this window title
for things like the overview or the window switcher where it would
trigger an UTF8 validation error and leave the shell in an unusable
state.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1794
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1850>
This commit adds scaling support to clutter_stage_capture_into, which
is currently used when screencasting monitors. This is supposed to
fix graphical issues that arise when using fractional scaling.
Fixes#1131
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1855>
We initialize, but might not start, e.g. when a test case just needs a
backend and doesn't start mutter. When cleaning up, we'll still try to
clean up Xwayland integration, and this commit handles cleaning up
without having made the mess.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1856>
We make objects inert when disabling the seat, but we requests may still
have effect. This is especially bad if disabling is followed by
destroying, but also means e.g. set_cursor() not doing an early out
after the pointer capability was disabled.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1822>
MetaDisplay does a lot of things, and is a central part to anything
window management. To let Wayland units have an easier time tearing
down, make it so that the Wayland infrastructure is terminated before
MetaDisplay.
This also makes sure that X11 support is turned off, so that we don't
stumble upon Xwayland terminating due to the Wayland socket connection
being broken. Will mitigate that in a better way in a later commit.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1822>
We might not be the only entity holding on to the X11 GdkDisplay,
meaning the X11 connection will stay alive indefinitely, e.g. if the gjs
context has some reference to it.
Avoid running into issues due to X11 connection errors by setting the
libX11 handlers to no-ops, so when we are terminating; that means the
GDK X11 connection can stay "alive" until its too late, and we'll just
silently ignore any connection errors that may happen due to the
lingering GDK display reference.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1822>
It keeps references to cursors, and cursors keep references to DRM
buffers. In order to be able to clean up on exit, explicitly destroy the
cursor tracker on shutdown.
We can't rely on GObject reference counting, as gjs might hold onto a
reference until it's garbage collected.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1822>
Various things, e.g, the renderer, the stage, either directly or
indirectly depends on GPU objects being alive during tear-down. Make it
so, by moving GPU cleanup after the other cleaning. This will allow
tearing down a couple of more objects.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1822>
The first phase happens early, which discards pending page flips,
meaning the references held by those page flip closures are released.
The second phase happens late, after other units depending on the KMS
abstraction, have been cleaned up.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1822>
We already swapped the front buffer, and even if it didn't get
presented, we should still swap our representation of the state, to not
get into a confused buffer tracking state.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1822>
All pointer a11y is a fabrication of Clutter backend-independent
code, with the help of a ClutterVirtualInputDevice and with some
UI on top.
On the other hand, MetaInputSettings is a backend implementation
detail, this has 2 gotchas:
- In the native backend, the MetaInputSettings (and pointer a11y
with it) are initialized early, before the ClutterSeat core
pointer is set up.
- Doing this from the MetaInputSettings also means another dubious
access from the input thread into main thread territory.
Move the pointer a11y into ClutterSettings, making this effectively
backend-independent business, invariably done from the main thread
and ensured to happen after seat initialization.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1765
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1849>
Since commit d2f8a30625e we use Graphene to union paint volumes, it
turns out a quite severe issue snuck in during review of that MR though:
Unioned paint volumes (so paint volumes of any actors with children) now
have negative heights. Once projected to 2d coordinates they luckily are
correct again, which is why everything is still working.
The problem is that obvious once looking closer: For the y coordinates
of the unioned paint volume we confused the maximum and the minimum
points and simply used the wrong coordinates to create the unioned paint
volume.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1827>
The graphene functions used by clutter for picking assume that boxes are
inclusive in both there start and end coordinates, so picking at y
coordinate 32 for an actor with the height 32 placed at y coordinate 0
would still be considered a hit. This however is wrong as 32 is the
first position that is not in the actor anymore.
Usually this would not be much of a problem, because motion events are
rarely ever at exactly these borders and even if they are there will be
another motion event soon after. But since actors in gnome-shell usually
are aligned with the pixel grid and on X11 enter/leave events are
generated by the X server at integer coordinates, this case is much
more likely for those.
This can cause issues with Firefox which when using client side
decorations, still requests MWM_DECOR_BORDER via _MOTIF_WM_HINTS to have
mutter draw a border + shadow. This means that the Firefox window even
when using CSD is still reparented. For such windows we receive among
others XI_RawMotion and XI_Enter events, but no XI_Motion events. And
the raw motion events are discarded after an enter event, because that
sets has_pointer_focus to TRUE in MetaSeatX11. So when moving the cursor
from the panel to a maximized Firefox window the last event clutter
receives is the enter event at exactly integer coordinates. Since the
panel is 32px tall and the generated enter event is at y position 32,
the picking code will pick a panel actor and the focus will remain on it
as long as the cursor does not leave the Firefox window.
Fix this by excluding the bottom and right border of a box when picking.
Fixes https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/4041
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1842>
Compositor behaviour when calling `wl_data_device_manager_get_data_device`
for the same seat multiple times is not very clearly defined in
the spec and both Mutter and Weston currently don't handle
the DnD case properly.
While Weston handles DnD only for the last created data device,
Mutter, because of some internal reshuffling, ends up toggling
between two devices.
Properly handling this case requires some bigger changes. So
in order to behave predictable and in line with Weston,
only take the last created data device into account while
still keeping the previous created ones around.
The main affect client here is Firefox, which gets very
confused by the toggling behaviour and becomes more stable
with this patch.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1841>
Virtual Kernel Mode Setting (vkms) is a virtual /dev/dri/card* device
not backed by any actual hardware. It's intended for testing purposes,
e.g. to run tests suites with a reproducable setup, or in continuous
integration pipelines.
Currently mutter don't have any tests that can run on top of vkms, but
will eventually get that. To prepare for the ability to do that, and
having said kernel module loaded without causing wierd issues with any
active session, add an udev rule that tells mutter to ignore any vkms
device.
Otherwise, when vkms is loaded, mutter would detect it, assume it's a
regular monitor, configure it as such, thus add a region of the stage
that ends up nowhere, which isn't very helpful. It might also conflict
with running actual tests that need to interact with vkms if the active
session has taken control of it.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1740>
With atomic mode setting, commits don't work when CRTCs aren't enabled,
which they aren't when we're power saving. This means the gamma state
fails to being update. To fix night light and for whatever other reason
gamma ramps was changed during power saving by marking the CRTC gamma
state as invalid when leaving power saving, as well as when resuming.
This means that the next frame will append the CRTC gamma state to the
KMS commit.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1755
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1835>
This GSource is not being properly unref nor the variable holding it
cleared. This on one hand leaks the GSource memory, on the other hand
may trigger warnings in keyboard_repeat() as the source may be
(reentrantly) cleared, yet we don't exit early as
seat_impl->repeat_source is never NULL.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1839>
When a viewport source rect or destination size is set, `stex->dst_width`
gives us the the cropped and/or scaled size. At this step, we need the
uncropped/unscaled size however.
Note: this is only ever relevant if buffer transform and viewport are
used together - otherwise the values are the same.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1836>
The shadow size is factored into the paint volume MetaWindowActorX11
returns in its get_paint_volume() vfunc override, so we should
invalidate the paint volume every time that shadow might change.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1829>
Turns out ClutterClones need a bit of extra handling as always, there's
currently nothing that invalidates a clones paint volume when the source
actors paint volume changes.
Since ClutterClones get_paint_volume() implementation simply takes the
source actors paint volume and returns that, we should make sure they
are kept in sync and invalidate the clones paint volume as soon as the
source actor gets its PV invalidated.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1829>
When using buffer transforms and viewports together, we currently
apply the transformation (read: rotation) first, resulting in
wrong buffer coordinates for viewport source rects.
Flip the order in whitch we apply our matrix transformations.
This can be tested e.g. via:
`weston-simple-damage --use-viewport --transform=flipped-180`
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1832>
If a subsurface first gets reordered and afterwards detached from
the parent before the parent surface got commited, we currently
would end up reattaching the subsurface to its previous parent.
While clients should avoid this behaviour, it's legit according
to the spec.
We already prevent similar cases where the subsurface is destroyed -
extend that check to detaching, which includes the destroy case.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1831>
Currently when reordering subsurfaces, we un- and reparent all child
actors of the window actor. This is unnecessarily wasteful and
triggers bugs in clutter. While the underlying issue should be fixed
eventually, simply reorder the actors with the tools clutter provides
us with, avoiding those bugs and likely being faster as well.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1831>
Users who customized their workspace shortcuts before updating to
GNOME 40 are experiencing troubles when trying to use the same
keybinding for the horizontal shortcuts, because of hidden conflicts
with the vertical ones.
We can address this by registering the old shortcuts with Settings
without showing them in the UI, so that they are taken into account
by Settings' conflict resolution.
https://gitlab.gnome.org/GNOME/mutter/-/issues/1740
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1814>
The X server generates a property change notification whenever it processes a
property change request, even if the value of the property is not changing. This
triggers libgdk to probe all display outputs, which can be slow depending on
which display driver and hardware are in use.
#0 0x00007f8e4d5e91a0 in XRRUpdateConfiguration () at /usr/lib/libXrandr.so.2
#1 0x00007f8e505208da in _gdk_x11_screen_size_changed (screen=0x5566e4b7e080, event=0x7ffe0e44bd60) at ../gdk/x11/gdkscreen-x11.c:1199
#2 0x00007f8e505066d1 in gdk_x11_display_translate_event (translator=0x5566e4b5b110, display=0x5566e4b5b110, event=0x7f8dec001b20, xevent=0x7ffe0e44bd60) at ../gdk/x11/gdkdisplay-x11.c:1201
#3 0x00007f8e505135a0 in _gdk_x11_event_translator_translate (translator=0x5566e4b5b110, display=0x5566e4b5b110, xevent=0x7ffe0e44bd60) at ../gdk/x11/gdkeventtranslator.c:51
#4 0x00007f8e50512c97 in gdk_event_source_translate_event (event_source=0x5566e4b764a0, xevent=0x7ffe0e44bd60) at ../gdk/x11/gdkeventsource.c:243
#5 0x00007f8e50512f57 in _gdk_x11_display_queue_events (display=0x5566e4b5b110) at ../gdk/x11/gdkeventsource.c:341
#6 0x00007f8e50497644 in gdk_display_get_event (display=0x5566e4b5b110) at ../gdk/gdkdisplay.c:442
#7 0x00007f8e5051301f in gdk_event_source_dispatch (source=0x5566e4b764a0, callback=0x0, user_data=0x0) at ../gdk/x11/gdkeventsource.c:363
#8 0x00007f8e516ecf9c in g_main_context_dispatch () at /usr/lib/libglib-2.0.so.0
#9 0x00007f8e51740a49 in () at /usr/lib/libglib-2.0.so.0
#10 0x00007f8e516ec503 in g_main_loop_run () at /usr/lib/libglib-2.0.so.0
#11 0x00007f8e508ef5fd in meta_run_main_loop () at ../src/core/main.c:928
#12 0x00007f8e508ef60e in meta_run () at ../src/core/main.c:943
#13 0x00005566e450842a in ()
#14 0x00007f8e50649b25 in __libc_start_main () at /usr/lib/libc.so.6
When GNOME is animating a display fade when the night light feature is toggled
on or off, it sends a lot of change requests for the CTM property in the
process, which triggers a lot of display probes from gdk. In the case of the
night light feature, the CTM itself is not actually changing, so these requests
are redundant. Fix this by caching the CTM value in the MetaOutputXrandr and
skipping the server requests if it's not being changed.
Fixes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/3978
Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1816>
When we set the matrix, we checked the device mapping mode in the main
thread, then passed along the calculated matrix to the input thread for
application. This could however be racy, as the mapping mode is managed
in the input thread. Fix this by sending the unaltered matrix, having
the input thread checking the mapping mode.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1806>
The connector state wasn't properly predicted, as it earlied out if
the connector wasn't part of a mode set connector list.
Instead use the old CRTC to check whether it was used in any mode set,
and whether the connector was part of any new mode set, to predict
whether the connector is inactive or active.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1821>
When a device only had mode sets which turned off monitors, not enabling
anything, there would be no KMS update created and posted, and the
active monitors would remain on.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1821>
On hybrid graphics system, the primary path used to transfer the stage
framebuffer onto the dedicated GPU's video memory preparing for scanout,
is using the dedicated GPU to glBlitFramebuffer() the content from the
iGPU texture onto the scanout buffer.
After we have done this, we reset the current EGL context back to the
one managed by cogl. What we failed to do, however, was to reset the
current EGL context when we inhibited the actual page flip due to having
entered power save mode.
When we later started to paint again, Cogl thought the current EGL
context was still the correct one, but in fact it was the one used for
the iGPU -> dGPU blit, causing various EGL surface errors, and as a side
effect, eventually hitting an assert.
Fix this by making sure we reset to the Cogl managed EGL context also
for this case.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1803>
Destroying the EGLSurface frees the underlying container structs. When
we call gbm_surface_release_buffer() with a gbm_surface the EGLSurface
was created from, doing that after the EGLSurface was destroyed results
in attempts to access freed memory. Fix this by releasing any buffer
first, followed by destroying the EGLSurface, and lastly, the
gbm_surface.
This was not a problem prior to CoglOnscreen turning into a GObject, as
in that case, the dispose-chain was not setup correctly, and the
EGLSurface destruction was done in the native backend implementation.
This also changes a g_return_if_fail() to a g_warn_if_fail(), as if we
hit the unexpected case, we still need to call up to the parent dispose
vfunc to not cause critical issues.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1803>
It's handled by CoglOnscreenEgl's dispose() implementation. It was
failed to be invoked in the past because the old non-GObject web of
vtables were not setup correctly, meaning the old generic EGL layer of
the CoglOnscreen de-init was never invoked.
When the type inheritence was cleaned up, this mistake was not cleaned
up, so do that now.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1803>
Mutter would deny the application the right to resize itself during an
interactive resize, to avoid the user and the client to fight for the
size.
When the client is not allowed to resize, it would use the client rect
rather than the buffer rect.
As a result, the client window with client side decorations would
quickly shrink to its minimum size.
Use the buffer rect instead, so that the size really remains the same.
https://gitlab.gnome.org/GNOME/mutter/-/issues/1674
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1777>
Meson doesn't seem to handle depending on generated headers, at least
when those headers are pulled in indirectly via another header file.
Luckily, we don't actually need to include the generated D-Bus boiler
plate in meta-monitor-manager-private.h, since the MetaMonitorManager
type no longer is based on the D-Bus service skeleton.
So, by moving the inclusion of the generated D-Bus header file into
meta-monitor-manager.c, we should hopefully get rid of the sporadic
build issues.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1682
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1819>
If there was only a single mode, add the common modes to provide options
to select other resolutions than the built in default. This avoids
issues where the connector listed multiple supported modes, but where
the common modes added would exceed the possible bandwidth. We could
probably make an attempt to filter out more modes from the common mode
list to avoid these issues, but it's likely that the driver already
lists suitable modes, meaning there is no point in adding the common
modes.
The common modes were initially added[0] to add modes to connectors with
a single bundled mode, so we shouldn't regress the original bug fix.
[0] https://bugzilla.gnome.org/show_bug.cgi?id=744544
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1232
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1824>
Lets wait until we have better ways to unredirect client buffers before
we start enabling KMS modifiers for amdgpu and nouveau; we end up with
mismatch between client buffer modifiers and primary plane modifier,
which right now needs to be the same for unredirection.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1792>
The intel DRM driver is known for not being able to handle multi head
setups when KMS modifiers are enabled, due to the implicitly selected
modifiers, while being more suitable for single head setups, cause
bandwidth issues when a certain number of monitor times resolution and
refresh rate is configured.
We don't yet support automatically finding a combination of modifiers
that work, and have because of this disabled KMS modifiers unless the
driver actually needs it.
Lets flip this configuration the other way around, changing the current
udev rule to decide wen to *disable* KMS modifier support, as it so that
only the Intel driver has this problem, while on the other hand, there
several drivers that requires modifiers to function at all.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1792>
The input thread is in deep water doing the meta_is_*() check itself,
as that pokes the MetaMonitorManager managed by the main thread. Use
the getter from the MetaViewportInfo instead.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1793>
We need to pass this info from the main thread, as that pokes the
MetaMonitorManager underneath. Store it in the MetaViewportInfo
so that the input thread can use this information.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1793>
Updating the last_paint_volume while painting has proven itself to be
quite prone to issues: First we had to make sure actors painted by
offscreen effects get their last_paint_volumes updated correctly (see
0320649a1c50993f308344e0dd296abaad0ee6d4), and now a new issue turned up
where we don't update the paint volumes while a fullscreen unredirect is
happening.
To stop those issues from happening and to lay the foundation for using
the last_paint_volume for other things, update the last_paint_volume in
a separate step before painting instead of doing it in
clutter_actor_paint().
To save some resources, avoid introducing another traversal of the
scenegraph and add that step into the existing step of updating the
stage_views lists of actors. To properly update the paint volumes, we
need to do that after finishing the queued redraws, which is why we move
clutter_stage_maybe_finish_queue_redraws() to happen before the new
clutter_stage_finish_layout().
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1699
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1773>
The priv->paint_volume field of ClutterActor stores the cached paint
volume in the actors local coordinate system. It consist of the actors
paint volume itself and the union of all children paint volumes.
We want to invalidate those cached paint volumes according to the
following rules:
- If an actors transformation matrix changes, all paint volumes of the
parent-tree need to be invalidated (that's because the parent-volumes
have unioned the actors paint volume). Our own paint volume does not
need invalidation since the transformation matrix is not applied to it.
- If an actors allocation-size changes, its own paint volume and all the
volumes of the parent-tree need to be invalidated. That's because the
allocation-size is used as the size of the paint volume.
- If a clip gets set or clip_to_allocation gets enabled for an actor,
its own paint volume and all the volumes of the parent-tree need to be
invalidated. That's because the clip is factored in when creating the
paint volume.
So far we did this invalidation in various places and the invalidation
up the parent-tree happened inside clutter_actor_real_queue_relayout().
We did not invalidate on changes to the actors transformation matrices
and the invalidation in clutter_actor_real_queue_relayout() was more
like a "big hammer" that probably invalidated unnecessarily a few times.
So introduce proper infrastructure to invalidate those cached paint
volumes of actors only in the cases where they actually need to be
invalidated. To do that, we reuse the transform_changed() function and
introduce a new function queue_update_paint_volume() that invalidates
the paint volumes up the actor tree.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1773>
ClutterActors can override the get_paint_volume() vfunc in case they
draw outside the allocation. That's used by a bunch of actors, for
example ClutterText or StViewport in gnome-shell.
In case of StViewport, the paint volume returned depends on the value of
the StAdjustment, which means when we start to cache paint volumes more
agressively in ClutterActor, we'll need to add API that allows
StViewport to invalidate the paint volume. So introduce
clutter_actor_invalidate_paint_volume() to invalidate the cached paint
volume.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1773>
Use the new API to make sure the shaped texture has a valid size
during the next layout phase.
This is needed here because, quoting the previous commit:
When the texture size is invalidated using `invalidate_size()`, the new
size will only get calculated the next time `update_size()` is
called. This happens e.g. in `meta_shaped_texture_get_preferred_size()`
via `ensure_size_valid()`.
`update_size()` can chain up to `clutter_content_invalidate_size()`
as well as emitting a `size-changed` signal. If this happens during
layout, the result is a 'change the layout conditions during layout'
issue, causing heavy breakage in e.g. the Shell overview.
To fix this, expose `ensure_size_valid()` as API so callers can make
sure the texture has a valid size without creating redundant size
invalidations calls.
Note that if a buffer with a new size is attached we already trigger
`update_size()` explicitely, avoiding such situations.
Closes https://gitlab.gnome.org/GNOME/mutter/-/issues/1718
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1799>
When the texture size is invalidated using `invalidate_size()`, the new
size will only get calculated the next time `update_size()` is
called. This happens e.g. in `meta_shaped_texture_get_preferred_size()`
via `ensure_size_valid()`.
`update_size()` can chain up to `clutter_content_invalidate_size()`
as well as emitting a `size-changed` signal. If this happens during
layout, the result is a 'change the layout conditions during layout'
issue, causing heavy breakage in e.g. the Shell overview.
To fix this, expose `ensure_size_valid()` as API so callers can make
sure the texture has a valid size without creating redundant size
invalidations calls.
Note that if a buffer with a new size is attached we already trigger
`update_size()` explicitely, avoiding such situations.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1799>
If only a viewport destination size is set, the noop viewport has
to take the buffer scale into account.
If a viewport source but no viewport destination size is set, the
destination size is that of the viewport source, not of the whole
texture.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1786>
The action might not have been triggered yet, as per its trigger
threshold. This doesn't mean we shouldn't reset the point(s) accumulated
so far.
This fixes those touchpoints persisting after disable/enable, thus
making gesture recognition fail from there on.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1791>
We might want to perform distance/threshold checks in the ::prepare
vfunc, but we didn't record the last motion event yet. This used to
give a delta of 0/0 between the press and last motion coordinates,
despite the ClutterGestureAction having a trigger threshold. This
happens no longer.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1791>
The usage of clutter_actor_get_preferred_width/height() for building the
pick box can trigger Clutters size negotiation machinery in case the
allocation of the actor is invalidated, with commit 82f3bdd1 we worked
around that by excluding actors with invalidated allocations from
picking.
There's no need to do that though, when picking we always want to
operate on the last known allocation of the actor, since that is what's
actually painted on the screen.
So instead of not picking at all when an actors allocation is
invalidated, just use the size of the last allocation. We still have to
factor in one extra case, that's when an actor hasn't gotten any
allocation yet: In that case we want to exclude the actor from picking
since the actor is not on the screen yet.
This fixes a regression introduced by the commit mentioned above where
picking wouldn't work on windows that have just been resized.
https://gitlab.gnome.org/GNOME/mutter/-/issues/1674
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1784>
If a client is naive enough to assume that it can set the selection while
it has got no surfaces, mutter will simply ignore the request and leave
the selection unchanged.
This is good and the expected behavior, however the poor client that did
this will enter in an inconsistent state where it "claimed" the selection,
but nobody told it that the wl_data_source is not current.
So, when the client is focused the next time, it will receive wl_data_offers
as usual, but it will still think all the time that it is owning the
selection. In the case of GTK, that takes client-side shortcuts, so any
attempted paste will still bring back the client-side aborted selection.
To fix this, cancel the selection right away if it happened while unfocused,
the client will be able to undo its own failed selection, and not assume
that future offers are its own.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1469
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1772>
It seems to be the preferred format of the Mesa V3D driver on
Raspberry Pi 4. If the compositor doesn't advertise it then Mesa will
fallback from `zwp_linux_dmabuf_v1` to `wl_drm`, incorrectly. Meaning
it will keep using a buffer with modifiers on an interface that does
not have modifiers.
Add support for `DRM_FORMAT_ABGR2101010`. It works, and prevents Mesa
from taking its broken fallback path.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1520
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1798>
As documented in g_once_init_enter(): "While @location has a volatile qualifier,
this is a historical artifact and the pointer passed to it should not be
volatile.". And effectively this now warns with modern glibc.
Drop the "volatile" qualifier from these static variables as it's expected.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1785>
As documented in g_once_init_enter(): "While @location has a volatile qualifier,
this is a historical artifact and the pointer passed to it should not be
volatile.". And effectively this now warns with modern glibc.
Drop the "volatile" qualifier from these static variables as it's expected.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1785>
As documented in g_once_init_enter(): "While @location has a volatile qualifier,
this is a historical artifact and the pointer passed to it should not be
volatile.". And effectively this now warns with modern glibc.
Drop the "volatile" qualifier from these static variables as it's expected.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1785>
For Xwayland, mutter creates the sockets in the standard /tmp/.X11-unix
directory.
Yet, if that directory already exists, it may have been created by
another user with full control over the created socket.
To avoid that issue, if the directory /tmp/.X11-unix already exists,
check that the permissions are as we expect, i.e. the directory belongs
to either root or the user herself, is writable and has the sticky bit.
Thanks to fabian@ritter-vogt.de for reporting that issue.
https://gitlab.gnome.org/GNOME/mutter/-/issues/1708
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1787>
Unlike other subsurface state, placement operations need to get
applied in order. As per spec:
```
Requests are handled in order and applied immediately to a pending
state. The final pending state is copied to the active state the
next time the state of the parent surface is applied.
```
Having placement operations being part of the subsurface state
makes it difficult to support arbitrary orderings. Make them
part of the parents surface pending state instead.
Closes https://gitlab.gnome.org/GNOME/mutter/-/issues/1691
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1768>
Some events such as the proximity one requires a device to be set before
we process them, so ensure we process the event details after we've
added the device to the seat.
This may lead to handle a device-removed signal before the clutter event
but it's anyways not different from what we did before commit 012c0a18
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1779>
The CallyStage objects lifetime is tied to the stage, so if we add a
weak pointer to it, we won't be able to remove it, as we would try to do
so not until the stage itself is being disposed, at which point removing
it fails. However, not removing it will make the stage try to clean up
the weak refs, and since it does this more or less directly after
freeing the cally stage, it ends up writing NULL to freed memory,
causing memory corruption.
Fix this by avoiding adding the weak pointer when that pointer is to the
stage.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1775>
MetaInputSettings unref:ed the seat on destruction, but it never ref:ed
it on construction, meaning it "stole" the reference from the rightful
owner. Make MetaInputSettings less of a thief.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1775>
XIQueryPointer allocates the button state mask that we were leaking in
some places. We need to manually free this, because there is no XI
function that would do this for us.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1728>
In case the shell ignores or can't accept the restart request we should
hide the message that has been just requested to be shown.
As per ::show-restart-message signal documentation, this has to be done by
emitting the signal with a NULL message.
So follow the API properly in such case
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1780>
In non-systemd managed session we are unable to start services on
demand. Instead, gnome-session will start everything at login time,
including any X11 related service (i.e. gsd-xsettings).
However, in order to start gsd-xsettings, Xwayland needs to be started
already. Otherwise it will connect to GNOME_SETUP_DISPLAY and login will
hang at that point.
Fix this by detecting whether mutter is running in a systemd unit. If it
is, we assume that we are systemd managed and the machinery to start the
services works fine. If not, we assume that the session management may
unconditionally try to start X11 related services and Xwayland must be
started in order to not block this.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1771>
On Wayland MetaInputSettings is part of the input thread. Connecting
a GSettings binding to the default ClutterSettings could result in the
change notification being emitted on the input thread. This then could
end up triggering the same handler from two different threads at the
same time. In the case of the ClutterText layout cache it was attempting
to unref the same layout twice, leading to a crash.
This can be avoided by simply removing the GSettings bind. This does not
cause changes to this setting to be missed by ClutterSettings because it
itself already sets up a bind.
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1696
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1776>
When deciding if `configure` event should be sent,
`meta_window_wayland_move_resize_internal` compares requested window size
with `window->rect` size. However, `window->rect` is only updated when `commit`
is received. So the following sequence produces incorrect result:
1. a window initially has size `size1`
2. `move_resize_internal` is called with `size2`. `configure` is sent
3. `move_resize_internal` is called with `size1` to restore original size,
but `commit` for `size2` haven't arrived yet. So `window->rect` still has size
`size1`, and thus new `configure` is not sent
4. `commit` for `size2` arrives, window changes size to `size2`
Expected window size in the end: `size1`
Actual: `size2`
To fix the issue, take size from pending `configure` events into account.
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1627
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1755>
With the surfaceless mode in the headless backend, it's now possible to
initiate the headless mode without any mode setting devices, or render
nodes, without any special CI runner privileges.
The native backend tests include screen cast tests, so make them
possible to run by starting pipewire. Testing shows that enabling audio
support (pulseaudio & jack compat layers) makes the tests dead lock and
eventually timeout, so disable those features for now.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
This eliminates the need for any render node or device nodes, thus can
be used without any graphics hardware available at all, or with a
graphics driver without any render node available.
The surfaceless mode currently requires EGL_KHR_no_config_context to
configure the initial EGL display.
This also means we can enable the native backend tests in CI, as it
should work without any additional privileges.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
Tests that creating and starting a virtual screen cast monitor works,
and that at least one one buffer is processed.
Currently the content of the buffer isn't checked more than it can be
mmap():ed. Only MemFd buffers are tested for for now, as DMA buffers
would need a surfaceless EGL context to check properly.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
The new RecordVirtual API creates a virtual monitor, i.e. a region of
the stage that isn't backed by real monitor hardware. It's intended to
be used by e.g. network screens on active sessions, virtual remote
desktop screens when running headless, and scenarios like that.
A major difference between the current Record* API's is that
RecordVirtual relies on PipeWire itself to negotiate the refresh rate
and size, as it can't rely on any existing monitor, for those details.
This also means that the virtual monitor is not created until the stream
negotiation has finished and a virtual monitor resolution has been
determined.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
The area source, window source, and monitor source, currently set up the
stream size up front, given the area, maximum allowed window size or
monitor resolution, but for to be introduced sources, the size will be
negotiated using PipeWire, instead of specified via the D-Bus API. This
commit changes the internal source API to allow for this. There are
currently no users of this new behaviour.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
There may be a race between the ability to turn stream relative input
coordinates and turning them into screen coordinates, due to the future
scenario where the entity backing a stream is created and managed ad-hoc
depending on PipeWire stream negotiations.
When an input event is sent during this time, drop it.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
Make it possible to create persintent virtual monitors using command
line argument. This will not be the only way to create virtual monitors,
the primary way will be using the screen cast API, but using command
line argumenst is convenient for debugging purposes.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
The testing currently done is:
* Creating a virtual monitor succeeds and gets the right configuration
* Painting a few times results in the expected output
* Changing the content of the stage also changes the painted content
accordingly
* Destroying the virtual monitor works as expected
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
This adds a test framework that makes it possible to compare the result
of painting a view against a reference image. Test reference as PNG
images are stored in src/tests/ref-tests/.
Reference images needs to be created for testing to be able to succeed.
Adding a test reference image is done using the
`MUTTER_REF_TEST_UPDATE` environment variable. See meta-ref-test.c for
details.
The image comparison code is largely based on the reference image test
framework in weston; see meta-ref-test.c for details.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
Virtual monitors are monitors that isn't backed by any monitor like
hardware. It would typically be backed by e.g. a remote desktop service,
or a network display.
It is currently only supported by the native backend, and whether the
X11 backend will ever see virtual monitors is an open question. This
rest of this commit message describes how it works under the native
backend.
Each virutal monitor consists of virtualized mode setting components:
* A virtual CRTC mode (MetaCrtcModeVirtual)
* A virtual CRTC (MetaCrtcVirtual)
* A virtual connector (MetaOutputVirtual)
In difference to the corresponding mode setting objects that represents
KMS objects, the virtual ones isn't directly tied to a MetaGpu, other
than the CoglFramebuffer being part of the GPU context of the primary
GPU, which is the case for all monitors no matter what GPU they are
connected to. Part of the reason for this is that a MetaGpu in practice
represents a mode setting device, and its CRTCs and outputs, are all
backed by real mode setting objects, while a virtual monitor is only
backed by a framebuffer that is tied to the primary GPU. Maybe this will
be reevaluated in the future, but since a virtual monitor is not tied to
any GPU currently, so is the case for the virtual mode setting objects.
The native rendering backend, including the cursor renderer, is adapted
to handle the situation where a CRTC does not have a GPU associated with
it; this in practice means that it e.g. will not try to upload HW cursor
buffers when the cursor is only on a virtual monitor. The same applies
to the native renderer, which is made to avoid creating
MetaOnscreenNative for views that are backed by virtual CRTCs, as well
as to avoid trying to mode set on such views.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
The order of which function argument expressions are executed is
undefined, so don't rely on this for setting the background colors, as
it results in different colors on different architectures.
For example, it has been observed that the order of execution is
reversed comparing x86_64 and aarch64, making these two architectures
having different background color.
Fix this confusion, and also reproduceability in future reference tests,
by making the order of execution predictable.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
It's useful to be able to have very very tiny monitors (e.g. 60x60
pixels) when doing reference testing, as tests have reference images
that the output is compared to. Smaller reference images the less
storage they use.
To avoid annoying pointless warnings when this is done, change the
pedantic workspace work area code to be more forgiving if the work area
happens to match the display size.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
When rebuilding the monitors (e.g. during hotplug), make sure to detach
the disposed monitors from any outputs before creating the new monitors.
While this isn't currently needed, as outputs are too being recreated,
with the to be introduced virtual outputs that are created for virtual
monitors, this is not always the case anymore, as these virtual outputs
are not regenerated each time anything changes.
Prepare for this by making sure that cleaning up disposed monitors
detach themself properly from the outputs, so new ones can attach
themself to outputs without running into conflicts.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
This makes it possible to pass custom properties to backends when
constructing tests. This will be used to create "headless" native
backend instances for testing the headless native backend.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
With this commit, it's possible to run mutter without being DRM master.
It's not yet possible to add virtual monitors, but one can for example
already add virtual input devices.
This currently doesn't try to hook up to any logind session, thus will
not have a real seat assigned. Currently it's hard coded to "seat0".
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
Currently our only entry point for DRM devices is MetaKms*, but in order
to run without being DRM master, we cannot use /dev/dri/card*, nor can
we be either of the existing MetaKmsImplDevice implementation (legacy
KMS, and atomic KMS), as they both depend on being DRM master.
Thus to handle running without being DRM master (i.e. headless), add a
"dummy" MetaKmsImplDevice implementation, that doesn't do any mode
setting at all, and that switches to operate on the render node, instead
of the card node itself.
This means we still use the same GBM code paths as the regular native
backend paths, except we never make use of any CRTC backed onscreen
framebuffers.
Eventually, this "dummy" MetaKmsImplDevice will be replaced separating
"KMS" device objects from "render" device objects, but that will require
more significant changes. It will, however, be necessary for e.g. going
from being headless, only having access to a render node, to turning
into a real session, with a seat, being DRM master, and having access to
a card node.
This is currently not hooked up, but will be in a later commit.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
Add a flag to MetaSeatNative and MetaSeatImpl that tells it not to
attempt to create a libinput context. This is intended to be used when
mutter is to run headless, as in without any input devices other than
virtual ones.
Currently not hooked up.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
This leaves only the atomic mode setting cap check before creating the
impl device, aiming to make it possible to create a non-mode-setting
MetaKmsImplDevice implementation.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
Make it possible to pass --headless as a command line argument in order
to turn the native backend "headless". This currently doesn't do
anything, but the intention is that it should not use logind nor KMS,
and work completely headless with only virtual outputs.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1698>
This was added with commit c956193d, when we did connect to the seat
signals in input settings, but this is not the case anymore since commit
2c1558dd, so we can safely remove this call now.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1763>
Input settings requires a valid seat in order to initialize the a11y
settings (since commit 1609d145), however in X11 we never set it and
even if we create the input settings early (as per commit 7547891a) we
never initialize the seat for it.
This leads to startup critical errors on X11:
clutter_seat_get_pointer_a11y_settings: assertion
'CLUTTER_IS_SEAT (seat)' failed
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1763>
This commit adds the events created in the function
`meta_seat_x11_notify_devices` to the clutter events queue, which
are currently only added to the stage queue making the events not
being picked up by the `clutter_seat_handle_event_post` function.
This results in devices not getting added to the device-list of
`MetaInputSettings`.
Fixes the bug in which mouse and touchpad settings are not working in
the settings app during x11 session.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1767>
Since commit 2ceac4a device-related X11 events aren't processed anymore,
causing the input settings not to handle the devices.
This is due to the fact that we may never call clutter_seat_handle_event_post()
for such events.
While this is always happening for the native backend, it doesn't happen in
X11 because the events are removed from the queue as part of
meta_x11_handle_event(), and thus no event was queued to the stage by the
backend events source.
This also makes sure that the event post handler is called after the
event is actually processed, and not before an event is queued.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1564
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1769>
The nested backend may need to have an input setting implementation,
while we don't want to change the host settings (re-using an X11 input
settings) we can add a dummy implementation, until something more
complex is needed.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1769>
The sync ring has an API about "frames", where it is notified about
the end of frames. However, its "insert wait" call is done before
updates, meaning that some "insert waits" will never see the "after
frame" if there was no frame drawn. This will cause mismatching in the
frame counting, causing freezes in the synchronization until something
else triggers an actual frame, effectively "unfreezing" the sync ring.
Fix this by not only notifying the sync ring about frames when there
were actual frames drawn, but also on plain updates which didn't result
in a drawn frame.
Related: https://gitlab.gnome.org/GNOME/mutter/-/issues/1516
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1754>
When a gtk theme uses larger shadows for the unfocused state than for
the focused one, this can cause a crash in meta_frame_left_click_event.
Since whether to call meta_frame_left_click_event is decided based on
the decoration size before focusing and the control that was clicked on
after focusing, this can result in an event handled in
meta_frame_left_click_event being on the client area.
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1668
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1748>
With commit 7d7876880998fe7b414bb38f8094af9822020d1b we switched to
storing pointer coordinates in MetaInputDeviceNative instead of
ClutterInputDevice, and while we had set the coordinates of the
ClutterInputDevice in ClutterStage when queueing an event, we now set
the MetaInputDeviceNative coordinates in new_absolute_motion_event().
Here a small mistake snuck in: new_absolute_motion_event() only
translates the coordinates of the event, but we call
meta_input_device_native_set_coords() using the x and y variables
(which remain untranslated), so now the input device coordinates are no
longer translated.
Fix that by translating the coordinates of the x and y variables in case
we're we handling a tablet/stylus event instead of only translating the
event coordinates.
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1685
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1760>
This fixes the interpolate test to not use the wall clock, but the
monotonic clock. It also cleans up the timestamp granularity naming, so
that the different granularity is clearer, as in the test, different
timestamps have different granularity.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1751>
This more or less rewrites this test so that it explicitly tests the
"interpolation" when a timeline loops, i.e. that if something occupies
the thread when a timeline was supposed to have looped, we end up in the
right place "in the middle" of the next timeline cycle.
The test more or less does this:
* Start a 3 second looping timeline
* Sleep so that we're in the middle of the first cycle
* Sleep again so that we end up in the middle of the next cycle
The semantics checked are that we see the following frames:
* The first frame with timestamp 0
* The second frame in the middle of the first cycle (timestamp ~= 1.5
sceonds)
* The third frame in the end of the first cycle (timestamp == 3.0
seconds)
* The fourth frame, first in the second cycle, with timestamp ~= 1.5
seconds)
This means we can increase the "grace period" to the double (from 0.5 s
to 1 s), while at the same time decrease the time spent running the test
(from 10 s to 4.5 s). This should hopefully make the test less flaky,
especially in slower runners, e.g. aarch64.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1751>
With commit c98575344295b715a0301c7319195b2724dc2bde the support for
multiple hardware cursors broke, but those were never properly supported
anyway as we usually assume there's only one hardware cursor around.
With the introduction of the KMS thread in the future, we'll only have
one KMS cursor that gets updated directly from the input thread. So
apart from the fact that it never really makes sense to have two cursors
visible, in this new model having multiple cursors won't work anyway.
So make the cursor we show for stylii a software cursor again.
Eventually the plan is to make the input device that's driving the KMS
cursor interchangeable, so that we can always use hardware cursors.
This reverts commit 165b7369c88644867a7c0c2791c48826240b63b5.
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1645
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1758>
To make the double buffered shadow buffer damaged tiles detection
feasable, a new EGL extension is needed for creating FBO's backed by
a custom CPU memory buffer, instead of DMA buffers, as DMA buffers can
be very slow to read, much slower than just painting the shadow buffer
directly.
Leave the code there, since such an EGL extension is intended to be
added, but hide it behind an env var so that it isn't enabled by
accident.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1724>
This commit assumes that cursor surfaces work in a "mailbox" fashion. If
they are painted multiple times before a successful flip, all commits
but the last get discarded, and the last commit gets presented after the
flip succeeds. This is more or less how it works in the atomic backend,
and also more or less how it works in other backends, with the exception
that the cursor painting might fail without any way of knowing. This
assumption is still better than unconditionally discarding all cursor
surface feedbacks as if the cursor painting always fails.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1484>
Regarding the sequence = 0 fallback: in some cases (moving a cursor
plane on atomic amdgpu) we get sequence = 0 in the page flip callback.
This seems like an amdgpu bug, so work around it by assuming a sequence
delta of 1 (it is equal to 1 because of the sequence != 0 check above).
Sequence can also legitimately be 0 if we're lucky during the 32-bit
overflow, in which case assuming a delta of 1 will give more or less
reasonable values on this and next presentation, after which it'll be
back to normal.
Sequence is also 0 on mode set fallback and when running nested, in
which case assuming a delta of 1 every frame is the best we can do.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1484>
The presentation-time protocol allows surfaces to get accurate
timestamps of when their contents were shown on screen.
This commit implements a stub version of the protocol which correctly
discards all presentation feedback objects (as if the surface contents
are never shown on screen). Subsequent commits will implement sending
the presented events to surfaces shown on screen.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1484>
This concerns only the cases when the presentation timestamp is received
directly from the device (from KMS or from GLX). In the majority of
cases this timestamp is already MONOTONIC. When it isn't, after this
commit, the current value of the MONOTONIC clock is sampled instead.
The alternative is to store the clock id alongside the timestamp, with
possible values of MONOTONIC, REALTIME (from KMS) and GETTIMEOFDAY (from
GLX; this might be the same as REALTIME, I'm not sure), and then
"convert" the timestamp to MONOTONIC when needed. An example of such a
conversion was done in compositor.c (removed in this commit). It would
also be needed for the presentation-time Wayland protocol. However, it
seems that the vast majority of up-to-date systems are using MONOTONIC
anyway, making this effort not justified.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1484>
KMS and GLX device timestamps have microsecond precision, and whenever
we sample the time ourselves it's not the real presentation time anyway,
so nanosecond precision for that case is unnecessary.
The presentation timestamp in ClutterFrameInfo is in microseconds, too,
so this commit makes them have the same precision.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1484>
A flag indicating whether the presentation timestamp was provided by
the display hardware (rather than sampled in user space).
It will be used for the presentation-time Wayland protocol.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1484>
A flag indicating whether the presentation timestamp was provided by the
display hardware (rather than sampled in user space).
It will be used for the presentation-time Wayland protocol.
This is definitely the case for page_flip_handler(), and I'm assuming
this is also the case for the two instances in the GLX code.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1484>
ClutterText has a bit of a mess around its signalling of changes to the
cursor position: There's the position (deprecated) and cursor-position
property, and there's the cursor-changed and cursor-event (deprecated)
signal. The two properties are supposed to be notified when the cursor
position changes, and the two signals are notified when the cursor
position or size changes.
Now the properties notifications and the signals get fired in two very
different places: The two properties are notified in
clutter_text_set_cursor_position(), while the signals are fired during
the paint cycle when we figured out the final cursor position. The
latter is a pretty bad idea, nobody expects such a signal to be fired
during painting, and also changes to the text that are done in the
signal handler will only be applied on the next paint.
Now StEntry listens to cursor position changes via cursor-changed and
invalidates its text shadow, but since the signal is only notified
during the paint, the old text shadow will still get applied. To fix
this, also emit the cursor-changed signal when we notify the
cursor-position property.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1757>
The old calculation was introduced to improve the precision
with commit c16a5ec1cf14d67ee8100afd31a2e85e7003f9a8.
Here, I call the calculation as "revision 2", and the
calculation even older as "revision 1", and the new
calculation introduced with this commit as "reivion 3".
Revision 2 has two problems:
1. The calculation is mixed with fixed-point numbers and
floating-point numbers.
To overcome the precision loss of fixed-point numbers division,
it first "calculates refresh rate in milliHz first for extra
precision", but this requires converting the value back to Hz.
An extra calculation has performance and precision costs.
It is also hard to understand for programmers.
2. The calculation has a bias.
In the process, it does:
refresh += (drm_mode->vtotal / 2);
It prevents the value from being rounded to a smaller value in
a fixed-point integer arithmetics, but it only adds a small
bias (0.0005) and consumes some fraction bits for
floating point arithmetic.
Revision 3, introduced with this commit always uses
double-precision floating-point values for true precision and
to ease understanding of this code. It also removes the bias.
Another change is that it now has two internal values, numerator
and denominator. Revision 1 also calculated those two values
first, and later performed a division with them, which minimizes
the precision loss caused by divisions. This method has risks of
overflowing the two values and revision 1 caused problems due to
that, but revision 3 won't thanks to double-precision. Therefore,
revision 3 will theoretically have the result identical with
the calculation with infinite-precision.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1737>
Handle the case of a TOUCH_BEGIN event during window dragging separately
instead of treating it like a TOUCH_UPDATE event: Simply return TRUE to
make Clutter stop event propagation if it's the pointer emulating
sequence and let Clutter propagate the event if it isn't.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/427>
Window dragging should be cancelled when the touch sequences we're using
are no longer available. Also listen to TOUCH_CANCEL events if the
window is grabbed and cancel the grab op when a TOUCH_CANCEL event
happens.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/427>
This removes the responsibility of tracking these from the backend to
the base object. The backends are instead responsible for calling the
function to update the values.
For the native backend, it's important that this happens on the correct
thread, so each time either of these states may change, post a idle
callback on the main thread that sets the, at the time of queuing said
callback, up to date state. This means that things on the main thread
will always be able to get a "new enough but not too new" state when
listening on the 'notify::' signals and getting the property value
after.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1739>
In an x11 session, we don't receive motion events from X when the
pointer is above a window. Since commit 734a1859 we only do picking on
motion events though, which means when clicking the mouse to focus a
window, we don't repick and might still think the pointer is hovering
above another window or actor, ending up not focussing the window.
Fix this by always repicking on BUTTON_PRESS events. While this is not
necessary in the wayland session, button presses happen rarely compared
to motion events, so it's not a performance regression to do it in
Wayland sessions, too.
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1660
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1752>
ClutterText allows setting a custom PangoAttrList, and St uses that to
set the text style it's reading from CSS. One style St enforces using
this mechanism is the text color and setting the text color should
obviously not affect the size of the layout. ClutterText does queue a
relayout in that case though because it unconditionally queues a
relayout when updating the PangoAttrList.
We can avoid this relayout by reusing an optimization ClutterText has:
clutter_text_queue_redraw_or_relayout() will only queue a relayout if
the requested size of the layout changed.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1750>
Make sure to reset all the state that was set for an interactive grab op
back to the defaults after a grab op has ended.
Especially important here is setting grab_frame_action back to FALSE,
since this will constrain window-titlebars to the panel. We set this to
TRUE on some grabs, for example when resizing, but not when moving
windows. Since this remained being set to TRUE, it would also constrain
non-grab window movements, like calling MetaWindow.move_frame(), which
is used by gnome-shells OSK. By resetting it back to FALSE after a grab,
the OSK can now always move non-maximized windows to the position it
wants.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1736>
The cache had the size 9, which was "big enough" in the past, but when
more ways pipelines could be constructed, the size was not enough. The
need to increase the cache size was hard to spot though, since adding
pipeline flag didn't give any hints about the cache being directly tied
to these flag values.
So, when enough flag bits were set when attempting to retrieve and put a
pipeline in the cache, it'd instead overwrite some arbitrary stack
memory, which would sooner or later result in a memory corruption
induced crash. Valgrind could not detect this particular memory
corruption, as it messed up stack memory, not e.g. freed heap memory, so
it instead got confused and thought plain stack values were unreadable.
Fix these two issues by making the cache size the combination of all
pipeline flags + 1, so that we can safely put any flag combination in
the cache.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1747>
If `meta_xwayland_start_xserver()` returned via an error path, some of
the socket FDs were leaked.
Fix that, and add `steal_fd()` calls to make it clearer that FDs passed
to `g_subprocess_launcher_take_fd()` are transferred to it. There were
no bugs with how `GSubprocessLauncher` was being used.
Spotted while working on
https://gitlab.gnome.org/GNOME/glib/-/issues/2332.
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1738>
Each next and current scanout buffer has a reference on them making sure
they stay alive. When dumb buffers were used on the secondary GPU state,
this didn't happen, leading to crashes due to unref:ing one time too
many, with backtraces such as
0) g_type_check_instance_is_fundamentally_a ()
1) g_object_unref ()
2) secondary_gpu_release_dumb ()
3) import_shared_framebuffer ()
4) update_secondary_gpu_state_post_swap_buffers ()
5) meta_onscreen_native_swap_buffers_with_damage ()
6) cogl_onscreen_swap_buffers_with_damage ()
7) swap_framebuffer ()
8) clutter_stage_cogl_redraw_view_primary ()
9) clutter_stage_cogl_redraw_view ()
10) _clutter_stage_window_redraw_view ()
11) handle_frame_clock_frame ()
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1746>
gnome-shell has this hack where it sets the environment variable
"NO_AT_BRIDGE" to "1" before calling meta_init() and then unsets it
after meta_init() returns.
This variable being set to "1" will then cause the ATK bridge in
at-spi2-gtk to fail to load, which GTK then ignores. This is on purpose,
since accessibility is supposed to be done done by GNOME Shell via
Clutter, not via GTK.
The problem is that, now, by default, setting "NO_AT_BRIDGE" to
"1" during meta_init() only has the desired effect on an X11 session,
where we always connect to the X11 server on startup (i.e. during
meta_init()). With Xwayland on-demand, we do not attempt to create the
GDK display during meta_init(), thus this hack falls apart.
Since there are no real altenatives to this hack, just move it to
mutter, which have a better idea when GDK displays are created or not.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1744>
Commit afa43154 tried to make sure the focus was properly changed when
calling focus_default_window() by checking the focused window just after
trying to set the focus.
However, the X11 “Inter-Client Communication Conventions Manual” version
2.0 (ICCCM 2 for short) states that some X11 client may want to use a so
called “globally active input” model in which case the client expects
keyboard input and set input focus even when it's not one of its own
window.
To comply with this, when dealing with such clients, mutter will not
change the focus and send a WM_TAKE_FOCUS message instead.
That mechanism will defeat the logic introduced by commit afa43154
because the focused window is not changed in this case. As a result, the
input focus will fallback to the no-focus window.
To avoid this, only check that the focus change occurred for windows
using a synchronous focus model.
v2: Split specific test for "globally active input" model (Florian).
v3: Remove the check for window->unmanaging which is useless (Jonas).
Fixes: afa43154 - "core: Make sure focus_default_window() worked"
Close: https://gitlab.gnome.org/GNOME/mutter/-/issues/1620
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1716>
X11 clients can use different models of input handling, of which some
may not result focus being set synchronously.
For such clients, meta_focus_window() will not change the focus itself
but rely on the client itself to set the input focus on the desired
window.
Add a new MetaWindow API to check when dealing with such a window.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1716>
Without these devices, things that depend on the existance of input
device classes won't know about the existance of e.g. pointer devices,
if the only pointer device is from a virtual one.
This requires handling situations where e.g. a device doesn't have a
device node thus can't be matched against a udev device.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1688>
Libinput will queue a few initial events when a seat is assigned to the
udev backend; a result of it probing udev adding detected devices. For
us to see these events, we need to dispatch libinput before going idle,
as nothing will show up on the libinput file descriptor until something
else (e.g. keyboard event or mouse movement) wakes us up.
Do this by adding a prepare() function to the libinput GSource, that
checks whether there are any events in the queue already, and return
TRUE if so is the case, causing us to dispatch before going fully idle.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1688>
The API for NotifyKeyboardKeycode() does not mention what type of
keycode is expected to be submitted.
So, clarify in the API that the keycode to submit is expected to be an
evdev keycode.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1732>
Currently, when a remote desktop user submits a keycode, it will be
interpreted differently, when using the x11 session, instead of a
wayland session.
In a wayland session, submitting a keycode will have the expected
result (as if the key was pressed locally).
In a x11 session, this is not the case. Instead of getting the expected
key, some other key will be pressed (or sometimes even none).
The reason for this is that the native backend interprets the keycode
as evdev keycode and the x11 backend interprets the keycode as xkb
keycode.
To ensure that both backends produce the same behaviour when submitting
a keycode, fix the x11 backend to always interpret the keycode as evdev
keycode, instead of a xkb keycode.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1732>
GObject signals pass the emitting GObject as the first argument to
signal handler callbacks. When refactoring the grab-op-begin/end signals
to remove MetaScreen with commit 1d5e37050df2b8e1db8b5ea301ee0162d77d4b74,
the "screen" argument was replaced with a "display" argument instead of
being removed completely. This made us call the signal handlers with two
identical MetaDisplay arguments, which is very confusing and actually
wasn't handled in a grab-op-begin handler in gnome-shell.
So fix this by not adding the MetaDisplay as an argument to those
signals, GObject will take care of that for us.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1734>
We're going to round the workspace backgrounds in the new overview for
gnome-shell 40.
So far corner-rounding was only possible for StWidgets because the
rounded clipping was done using cairo drawing. We now need rounded
clipping for ClutterActors too because backgrounds are drawn using
ClutterActors (or more specifically a ClutterContent). To implement
that, first a ClutterOffscreenEffect subclass together with a fragment
shader from GSK (see gskSetOutputColor() [1] in the GSK GL renderer
code) was investigated, and while that was generic and worked quite
well, it was extremely slow for the case of drawing wallpapers because
of all the FBOs that had to be allocated.
This is the new, more performant approach: Use the same fragment shader,
but perform the rounded clipping right in MetaBackgroundContent while
we're painting the wallpaper. This has almost no performance impact,
with the downside of not being a generic solution.
To allow for rounded clipping not only at the edges of the wallpaper,
but using any given bounding rectangle, the API exposes not only the
radius, but also a bounding rect.
[1] https://gitlab.gnome.org/GNOME/gtk/-/blob/master/gsk/resources/glsl/preamble.fs.glsl
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1717>
d42f1873fcd0876244eb8468d72ce35459ba94ca introduced a semi circular
reference between the CoglFramebuffer, and CoglJournal, where
CoglJournal would keep a reference on the CoglFramebuffer when there
were any entries in the journal log.
To avoid risking leaking these objects indefinitely, when freeing
objects without doing anything that triggered a flush, CoglFramebuffer
had a "filter" on cogl_object_unref() calls, which knew
about under what conditions CoglJournal had a reference to it. When it
could detect that there were only the journal itself holding such a
reference, it'd flush the journal, effectively releasing the reference
the journal held, thus freeing itself, as well as the journal.
When CoglFramebuffer was ported to be implemented using GObject instead
of CoglObject, this "filter" was missed, causing not only awkward but
infrequent leaks, but also situations where we'd flush journals when
only the journal itself held the last reference to the framebuffer,
meaning the journal would free the framebuffer, thus itself, in the
middle of flushing, causing memory corruption and crashes.
A way to detect this, by asserting on CoglObject reference count during
flush, is by adding the `g_assert()` as described below, which will
assert instead cause memory corruption.
void
_cogl_journal_flush (CoglJournal *journal
{
...
_cogl_journal_discard (journal);
+ g_assert (journal->_parent.ref_count > 0);
...
}
Fix this by making CoglFramebuffer the owner of the journal, which it
already was, and remove any circle referencing that was there before, as
it is not needed given that the CoglFramebuffer pointer is guaranteed to
be valid for the lifetime of CoglJournal as the framebuffer is the owner
of the journal.
However, to not miss flushing before tearing down, which is important as
this flushes painting calls to the driver that is important for e.g.
using the result of those journal entries, flush the journal the first
time cogl_framebuffer_dispose() is called, before doing anything else.
This also adds a test case. Without having broken the circular
reference, the test would fail on g_assert_null (offscreen), as it would
have been "leaked" at this point, but the actual memory corruption would
be a result of the `cogl_texture_get_data()` call, which flushes the
framebuffer, and causes the 'mid-flush' destruction of the journal
described above. Note that the texture keeps track of dependent
framebuffers, but it does not hold any references to them.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1474
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1735>
Currently, the documentation for NotifyPointerAxis() just mentions that
a smooth scroll event is emitted.
However, this is not entirely correct. For each NotifyPointerAxis(),
mutter emits an emulated discrete scrolling event based on the
submitted accumulated smooth scrolling deltas.
Additionally, it doesn't mention how the motion deltas need to be
interpreted.
So, document the NotifyPointerAxis() notification better by mentioning
the emulation of discrete scroll events, how these discrete scroll
events are calculated and how the motion deltas need to be interpreted.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1727>
When a remote desktop user emits a virtual smooth scrolling event, a
smooth scroll event, that is not emulated, is emitted and on occasion
a discrete scroll event, that is emulated, is emitted.
As base for the discrete scrolling event, the smooth scrolling steps
are accumulated.
When the accumulated smooth scrolling steps surpass the
DISCRETE_SCROLL_STEP, the discrete scrolling event is emitted.
Currently, mutter uses for DISCRETE_SCROLL_STEP the value 10, which is
a terrible value to work with, especially for high resolution mouse
wheels.
When a triple resolution mouse wheel is used, each scrolling step will
have the value 3 1/3.
Three of such events won't however surpass the DISCRETE_SCROLL_STEP.
To fix this situation, add DBL_EPSILON to the calculation step, when
checking for the discrete scroll event to ensure that 3 smooth scroll
events, with each having the value 3 1/3, emit a discrete scrolling
event.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1727>
MetaVirtualInputDeviceX11 currently doesn't handle smooth scroll events
at all.
So, if a user of the remote desktop API uses smooth scroll events, then
only the wayland backend handles these events.
The user of the remote desktop API however, might not know which
backend is being used and actually the user should not even have to
care about it.
Actual smooth events cannot be emulated in the X11 events.
What can be done however is accumulating smooth events and then when
the accumulated steps surpass the DISCRETE_SCROLL_STEP value, emit a
discrete scroll event.
So, do exactly that, to make smooth scroll events work when the remote
desktop API is used with the x11 backend.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1727>
With 734a185915464d2da32b6b1251c8af9a631b0f8c an optimization was
introduced to only pick on events which can actually cause the pointer
to move. In case of touch events, the first event (TOUCH_BEGIN) will
already move the touchpoint though, and we'll send our crossing
CLUTTER_ENTER event to the actor this TOUCH_BEGIN happened on.
So fix this embarrassing bug that caused touch input to break by also
picking to find an event-actor on TOUCH_BEGIN events.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1733>
Aside from ENTER/LEAVE, there are only two kinds of events that can move
the pointer, motion events and touch update events. Everything else
keeps the pointer at it's current position.
The reason we pick inside _clutter_process_event_details() is that we
want to set the event actor. Now if an event can't move the pointer, it
also can't change the event actor (well, it can subsequently by
triggering changes to the scenegraph, but that's handled elsewhere), so
there's no need to pick a new event actor when we get those events.
Instead, simply reuse the actor that's already associated with the
current input device as the event actor for non MOTION/TOUCH_UPDATE
events.
Events where a device or a touchpoint goes away (like DEVICE_REMOVED or
TOUCH_END/CANCEL) also affect picking, they don't need a repick, but
instead the actor associated with the device/touchpoint needs to be
unassociated. This is ensured by invoking remove_device_for_event() on
those events and will not be affected by this change.
This should improve performance while scrolling quite a bit, since
scroll events come in unthrottled and we now no longer do a repick on
each one of those.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1729>
Since a9a9a0d1c5 it didn't print anything. Now we reimplement it to
show FPS to a higher precision than in 3.36, add update/render times,
and separate stats per view/CRTC.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/154>
Given X11 nature, the pointer "leaves" the stage anytime it wanders into
a client window, or any other areas that are not deemed part of the
stage input region.
Yet we want to stay correct in those situations, e.g. have the clutter
side reasonably in sync, picking and highlighting to work properly, etc.
In order to achieve that, emulate motion events on XI_RawMotion. These
are as much throttled as our pointer tracking for a11y, in order to avoid
too many XIQueryPointer sync calls. This emulation only kicks in anytime
that X11 notifies us that we are not "on" the stage.
This replaces some sync_pointer() calls in GNOME Shell code that are
there just to compensate for this trait of X11, e.g. in the message tray
code.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1659>
Ensure we issue a motion event for the current pointer position,
as there might be situations where compositor modals get X grabs
from other clients stacked on top, or missed events in between
otherwise.
Ensure the Clutter state is still up-to-date afterwards here. This
replaces some sync_pointer() calls done in GNOME Shell code, always
done after modality changes.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1659>
Do these Wayland operations (that apply on both native and nested backends)
in the MetaCompositorServer subclass. We want to add more backend specific
behavior here in the future.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1659>
In the case that DnD is performed and succeeds, we want to release
the grab early, and let the transfer IPC happen in the back. For
that to happen without a hitch, drag source and offer must be left
related to each other after undoing the grab, even though the default
ungrabbing code does that automatically (indirectly, by unsetting the
drag focus).
In these cases, we used to manually unset the current source, so
this decoupling was skipped. Notably, one missed case is X11 DnD,
so we might end up with the situation there that DnD did succeed,
transfer is ongoing, but the source and offer are already decoupled,
this confused the machinery and missed the finishing XdndFinished
to be emitted to the X11 drag source.
The prior commits prepared for this source/offer decoupling being
a manual operation, this commit avoids doing this automatically
when ungrabbing.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1720>
This object is just being detached, with no code unref'ing it. Do
this whenever the XDnD selection goes unowned, usually a good
indication that the drag source no longer is one.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1720>
g_set_error_literal() asserts that the provided message is not NULL.
If it is NULL, the function is entirely no-op.
This resulted in a NULL dereference of the GError, which remained
NULL in this case, when trying to print a warning in
clutter_stage_cogl_redraw_view().
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1715>
We might have a stage view listener attached to the stage itself if the
actor didn't have a suitable frame clock when the actor was associated
with the timeline. We'd then listen to stage-views-changed signals on
the stage itself to be able to attach to a frame clock when one
appeared.
What went wrong is that if an actor that didn't have a frameclock was
associated with a timeline, but then destroyed, the timeline would
disassociate itself from the actor, but it'd still listen on the
stage-views-changed signal on the stage. This would be in itself
harmless, until the timeline itself is destroyed, as at this point, it
wouldn't clean up the stage-views-changed listener on the stage, as it's
assumed to only be valid when there is an actor attached.
Fix this issue by cleaning up the stage's stage-views-changed listener
when the actor is destroyed, as we wouldn't be able to make use of it by
then anyway.
Closes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/3323
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1719>
PipeWire recently introduced busy buffers, which actually fixes the last remaining
issue that blocked us from downgrading these cogl_framebuffer_finish() calls into
cogl_framebuffer_flush() ones.
Switch to cogl_framebuffer_flush() in all three stream sources.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1701>
The 0.3.21 release includes busy buffers, which are a requirement
for us to switch to cogl_framebuffer_flush() in the next commit.
Bump pipewire dependency to 0.3.21, and adjust CI to build it from
the corresponding tag.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1701>
Before this commit, next presentation time could end up behind now_us or
ahead of now_us depending on how presentation times happened to be
aligned relative to integer multiples of refresh_interval_us. It's not
clear whether this was originally intended because even if it the next
presentation time ends up behind now_us, it is moved ahead by a while
loop down below in this function.
Even though this difference in behavior didn't really matter, it made
reasoning about the subsequent branches more complex. It would also
potentially introduce bugs if the logic was to be modified. So this
commit makes it so next presentation time is always ahead of now_us.
It also adds a comment with a graph detailing the computations, and
adjusts the variable names to drop unfortunate terminology mistakes.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1707>
Last presentation time is mainly used to make sure predicted
presentation time is aligned with display refreshes. Even if it went
back in time, there will be no issue as next presentation time takes
current time into account. Synthetic presentation time is not exactly
aligned with display refreshes, so using it would only result in
inconsistent animations.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1707>
When a transfer request is done to the MetaSelectionSourceRemote source,
it's translated to a SelectionTransfer signal, which the remote desktop
server is supposed to respond to with SelectionWrite.
A timeout (set to 15 seconds) is added to handle too long timeouts,
which cancels the transfer request.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1552>
Nothing is hooked up, it only does basic sanity checking i.e. whether
the clipboard was enabled when interacting with it. No actual clipboard
integration is hooked up yet.
This also syncs org.gnome.Mutter.RemoteDesktop.xml from
gnome-remote-desktop.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1552>
The original implementation of ::touch-mode tested for keyboard
presence to know whether the OSK and other touch-only features were
enabled.
However that didn't pan out, every webcam, card reader and kitchen
sink like to live a second life as EV_KEY devices. This made the
detection of actual external keyboards a much harder task than it
sounds, and was thus removed in commit f8e2234ce59fbb.
Try a different approach here, and test for pointer devices, it
doesn't matter if internal or external devices, the rationales:
- It is significantly easier to get this right, there's virtually
no devices with abs/rel axes that don't try to be a real input
device of some sorts.
- It's not as good as testing for keyboard presence, but it's the
next best thing. These usually come in pairs, except in weird
setups.
- It is better than not having anything for a number of situations:
- Non-convertible laptops with a touchscreen will get touch-mode
disabled due to touchpad presence (plus keyboard). There's
been complains about OSK triggering with those.
- Same for desktop machines with USB touchscreens, the mouse
(and presumably keyboard) attached would make touch-mode
get in the middle.
- Convertible laptops with a broken tablet-mode switch get a
chance to work on tablet modes that do disable input devices
(e.g. detachable keyboards, or via firmware)
- Kiosk machines, tablets, and other devices that have a
touchscreen but will not regularly have a mouse/keyboard
will get the touch-mode enabled.
All in all, this seems to cover more situations the way we expect it,
there's only one situation that the OSK would show where it might
not be desirable, and one that might not show when it better should:
- Tablets and kiosk machines that get one keyboard plugged, but not a
mouse, will still show the OSK, despite being able to type right
away.
- Convertible laptops with broken/unreliable tablet-mode switch (e.g.
ignored by the kernel) rely entirely on the device/firmware
characteristics to work. If after folding into tablet mode the
touchpad remains active, touch-mode will not turn on.
Fixing the tablet-mode switch on these devices should be preferred,
as that'll also make libinput magically disable the touchpad.
The latter can be worked around with the a11y toggle. The former is
merely inconvenient, and nothing prevents the user from plugging a mouse
in addition.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1710>
Technically this is still wrong if the source actor or dnd actor are
transformed in other ways. However geometry scale is the by far most
common case and we currently lack convenience API in Clutter to
easily compute the right values.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1683>
When recording the screen and real time encoding it using a gstreamer
pipeline, that pipeline can stall when the encoder is too slow. Log a
debug message using the new SCREEN_CAST debug topic in that case so we
know when framedrops are happening.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1709>
Since commit c255031b6d6, we allow some modifier+scroll events to
pass through to Clutter to enable gnome-shell to handle them. That
action shouldn't trigger a modifier-only action at the same time, so
reset the corresponding tracking just like we do for modifier+click.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1695>
Since commit c255031b6d6 we pass scroll-events through to
the compositor if the window_grab_modifiers are pressed;
in order to allow gnome-shell to check for those events,
expose the struct member as a MetaDisplay property.
Also take the opportunity to pick a more generic name, now
that the modifier is no longer used exclusively for mouse
clicks (unless we maintain the notion of scroll events as
button 4 and 5 "clicks").
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1695>
The other end of the PipeWire stream can set the buffer data type to a
bitmask of supported buffer types. We should respect this, and not
attempt to allocate a DMA buffer if it isn't asked for.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1697>
Instead of getters, pass the width, height and stride around when
relevant. This also removes the redudant "stream_size" and
"stream_height" variables from the src struct, as they are already part
of the video format.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1697>
Previously we were setting the FBO's viewport to be the same dimensions as
the stage itself for compatibility. This works for most cases, but not if
the actor is larger than the stage. In that case it could cause excessive
clipping if the actor's transformed screen position was negative, which
is seen in https://gitlab.gnome.org/GNOME/gnome-shell/issues/2087
Also if a small actor paints to its negative dimensions (like a box-shadow)
then we might be missing those pixels on the left or top, even though
they're inside the paint volume.
Now we set the viewport dimensions to match the area we're actually
rendering so the FBO contents are never over or under clipped.
Fixes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/3068
Although if you try using shadows larger than that (like in
gnome-shell#1090) then you will also need gnome-shell!1417.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1053>
We had been doing it backwards as far back as e3966882e8 which meant
that we were translating by `fbo_offset / resource_scale` stage units
instead of just `fbo_offset`.
Because `fbo_offset` is in stage units already, it's not scaled and so
needs to be applied after the unscaling from texels to stage units.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1053>
Mutter freezes Xwayland commits when resizing windows, and thaw them in
the window actors' after_paint() for X11.
Yet, after_paint() could be never called, as when a new window is mapped
while the overview is active in gnome-shell.
As a result, the content of the X11 window will remain invisible to the
overview.
Add a new window actor API to tell whether commits can be frozen. For
Wayland window actors, this always return FALSE, whereas for X11 window
actors, it checks whether the Clutter actor is mapped.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1678>
When (un)maximizing, (un)fullscreening, the move/resize action is
flagged with 'ACTION_MOVE' and 'ACTION_RESIZE' , while e.g.
'appears-focus' does not.
When a client misbehaved and didn't immediately reply to a configure
request with a commit with the corresponding ack_configure, the
following commit would trigger a oddly timed move, making the window
appear to move back to a previous position.
Avoid this issue by only carrying over the target window position if the
configuration actually contained a new position.
We cannot only rely on the flags however, as e.g. a new position should
be respected during interactive resize, even though only 'ACTION_RESIZE'
is passed in such scenarios.
Do the same for the size, except if the window state dictates that the
size is fixed to a certain size, e.g. being fullscreen or maximized.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1445>
It's currently only handled by a surface backend framebuffer (assuming
the right GLX extensions are available). While it's theoretically
possible to do the same with the offcreen by having multiple textures,
it's not supported, so leave the FBO variant with a single warning if we
end up there.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1514>
The object was still pretending to be CoglFramebuffer itself, by using
naming and calling conventions making it seem like that. Fix that by
passing around the driver instead of the framebuffer.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1514>
Mutter needs to fetch the X11 Window ID from the onscreen and did that
by using an X11 specific API on the CoglOnscreen, where the X11 type was
"expanded" (Window -> uint32_t). Change this by introducing an interface
called CoglX11Onscreen, implemented by both the Xlib and GLX onscreen
implementations, that keeps the right type (Window), while avoiding X11
specific API for CoglOnscreen.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1514>
Mutter didn't use the APIs for resizeability of CoglOnscreens but
managed the size itself. For the native backend we don't ever resize
onscreens. Thus, remove this unused functionality.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1514>
The framebuffer driver was lazilly initialized on demand in some cases
(onscreen), and up front other (offscreen). Replace this with a more
predictable up front initialization, done at framebuffer allocation.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1514>
The CoglGLFramebuffer (not CoglGlFramebuffer) is a private struct for
keeping track of the framebuffer object. To avoid confusing with
CoglGlFramebuffer, rename it CoglGlFbo.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1514>
This way we can have separate types per modes of operation (e.g. if it's
backed by an EGLSurface or single texture), instead of being dependent
on a certain type (onscreen vs offscreen).
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1514>
Instead of calling "init_onscreen()" on two different separate vtables
from the allocate() funtion, just have the CoglOnscreen sub types
themself implement allocate() and initialize in there.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1514>
Thins means that e.g. MetaOnscreenNative now inherits CoglOnscreenEgl,
which inherits CoglOnscreen which inherits CoglFramebuffer, all being
the same GObject instance.
This makes it necessary to the one creating the onscreen to know what it
wants to create. For the X11 backend, the type of renderer (Xlib EGL or
GLX) determines the type, and for the native backend, it's currently
always MetaOnscreenNative.
The "winsys" vfunc entries related to onscreens hasn't been moved yet,
that will come later.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1514>
To get meta-renderer-native.c down to a bit more managable size, and to
isolate "onscreen" functionality from other (at least partly), move out
the things related to CoglOnscreen to meta-onscreen-native.[ch].
A couple of structs are moved to a new shared header file, as
abstracting those types (e.g. (primary, secondary) render devices) will
be dealt with later.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1514>
The mutter naming convention for types and their instance variables is:
Type name:
[Namespace][BaseName][SubType]
Instance name:
[base_name]_[sub_type]
This means that e.g. CoglOnscreenGLX is renamed CoglOnscreenGlx, and
glx_onscreen is renamed onscreen_glx. This is in preparation for
GObjectification.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1514>
The GLX winsys code had split up the GLX onscreen implementation into an
Xlib part (with a struct name confusingly enough identical to that of
the onscreen in the actual Xlib winsys). To remove some confusion,
combine the two.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1514>
Makes sure that monitor specs which may be read from EDID data do not
contain characters that are invalid in XML. Makes it possible to restore
monitor configs of monitor models with characters such as '&' in them.
To make this change not break any tests, the sample monitor configs need
to be adjusted as well. Apostrophes don't strictly have to be escaped in
XML text elements. However, we now do escape the elements in
`<monitorspec>` specifically.
Closes: <https://gitlab.gnome.org/GNOME/mutter/-/issues/1011>
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1685>
Previously the wl_resource and MetaWaylandGtkSurface corresponding to
any client gtk_surface have been kept around until the exit of the
client due to the client side destroy method not signaling the
destruction to the server. Ideally the protocol would have specified a
destroy request marked as destructor to handle this automatically,
however this is no longer possible due to the destroy method being
implicitly generated in the absence of an explicit request in the
protocol. Adding a destroy request marked as destructor now would
generate a new destroy method that unconditionally would send the
request to the server, which would break clients running on servers not
supporting that request.
So instead of modifying the destroy request add a new "release"
destructor, that indicates to the server that it can release the
resource. This can be optionally be used by clients depending on the
server protocol version.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1307>
The MetaWaylandSurface corresponding to a MetaWaylandGtkSurface can be
destroyed before the MetaWaylandGtkSurface is destroyed. In its destroy
function MetaWaylandSurface however was unsetting the destructor of the
correspnding resource along with the gtk_surface1 interface
implementation. This was done to prevent further gtk_surface1 requests
on a NULLed MetaWaylandSurface, if it has been destroyed before the
MetaWaylandGtkSurface.
It would be enough to just unset the resource implementation, while
keeping the destructor to fix this leak. However the following commit
will rely on the implementation being available after the
MetaWaylandSurface has been destroyed. So instead introduce NULL checks
for all functions that can be called on the gtk_surface1 interface and
do not unset the implementation.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1307>
If the monitor configuration changed, even though the streamed monitor
didn't change, we'd still fail to continue streaming, as we failed to
update the stage watchers, meaning we wouldn't be notified about when
the stage views were painted.
Fix this by reattaching the stage watches, i.e. update the painted
signalling listeners to listen to the right views, when monitor changes
happens.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1691>
We've inherited, and still keep in place, code that reads ini files
at ~/.config/clutter-1.0/settings.ini and /etc/clutter-1.0/settings.ini
to tweak different aspects of Clutter.
Some of these should use GSettings instead, some others are exposed
nowadays differently for our purposes (e.g. envvars, looking glass, ...).
Overall seems like an unexpected entry point nowadays, so remove the
parsing of these .ini files altogether.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1693>
Make the API used more shared and better named.
meta_monitor_manager_on_hotplug() was renamed
meta_monitor_manager_reconfigure(), and meta_monitor_manager_reload()
was introduced to combine reading the current state and reconfiguring.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1689>
It was named "backend_native" and "backend" which is easily confused with
MetaBackendNative and MetaBackend which tends to have those names.
Prepare for introducing the usage of a MetaBackendNative and MetaBackend
pointers here by cleaning up the naming.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1689>
Fullscreen X11 windows that attempt to change the resolution on Wayland
use a surface viewport to achieve this without affecting the resolution
of the display. This however also means that pointer events will be
delivered in the display coordinates while the code handling the window
frame is not aware of any such viewport scaling. So a right click
outside of the area corresponding to the new resolution will not be
considered to be on the client area. And since the only area that is
ignored when determining whether to perform the right click action, such
as opening the context menu, is the client area, this will result in the
action being performed, despite happening on the (scaled) client area.
While it would be possible to scale the event coordinates so that
get_control() correctly determines the frame element the cursor is on,
viewport scaling only affects fullscreen windows. Since fullscreen
windows have no frame, we can always assume that if the window gets
delivered an event for a fullscreen window, it is on the client area
without doing any additional calculations.
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1592
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1661>
We remove pending pings when unmanaging a window, but currently
don't prevent new pings to be scheduled after that.
The previous commit fixed a code path where this did indeed happen,
but as the result of gnome-shell trying to attach a Clutter actor
to a non-existent window actor is pretty bad, also guard can_ping()
against being called for an unmanaging window.
https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/2467
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1676>
This seems to have been the default in the past, but was (accidentally?) modified
by 8adab0275.
For GNOME 40, we'll be returning to our root with horizontal workspaces, so instead
of overriding it in GNOME Shell side, change the default back to what it once was.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1684>
If we are about to replace `redraw_clip` with a clamped version of itself
then we may as well do the same for `queued_redraw_clip`, so you can see
more precisely what the damage of the current frame is.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1554>
Initially we generate the new part of fb_clip_region from the new part
of redraw_clip, scale it up and clamp. But the clamping means the new
part of fb_clip_region might now represent a slightly larger area than
the new part of redraw_clip, by one pixel.
In some rare cases where a foreground actor honours redraw_clip, but
the background actor does not (meaning it might fill all fb_clip_region),
you could find 1px rendering glitches in that gap as the background
actor paints there but the foreground actor does not.
To ensure such glitches can never happen we now regenerate the final
redraw_clip as a clamped superset of the final fb_clip_region. That's
the minimum area we must paint to ensure no gaps appear inside
fb_clip_region.
Although the fix here sounds like the intent of the old code, the old
code forgot to include the new part of fb_clip_region in the clamping
of the final redraw_clip. So the new part of redraw_clip was sometimes
kept too small for the new part of fb_clip_region.
We also move the code to the main path because technically it's also
needed when `has_buffer_age == FALSE`.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1500
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1554>
We'll have two persistent client connections alive for the whole test,
one X11 client, and one Wayland client. So in order to be able to set up
the async waiter, do so after setting up the X11 client, as after that
we know we'll have a MetaX11Display ready to use.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1681>
One for the public channel, and one for the private maintainance
channel. Use the public one for test clients, otherwise tests become
flaky, and the private one for MetaX11Display.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1681>
This adds a MetaKmsImplDevice backend using atomic drmMode* API in constrast to
non-atomic legacy drmMode* API used in MetaKmsImplDeviceSimple.
This has various behavioral differences worth noting, compared to the
simple backend:
* We can only commit once per CRTC per page flip.
This means that we can only update the cursor plane once. If a primary
plane composition missed a dead line, we cannot commit only a cursor
update that would be presented earlier.
* Partial success is not possible with the atomic backend.
Cursor planes may fail with the simple backend. This is not the case
with the atomic backend. This will instead later be handled using API
specific to the atomic backend, that will effectively translate into
TEST_ONLY commits.
For testing and debugging purposes, the environment variable
MUTTER_DEBUG_ENABLE_ATOMIC_KMS can be set to either 1 or 0 to
force-enable or force-disable atomic mode setting. Setting it to some
other value will cause mutter to abort().
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/548
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
In order to reliably manage the reference count of the user data passed
to page flip listeners - being the stage view - make the ownership of
this data travel through the different objects that take responsibility
of the next step.
Initially this is the MetaKmsPageFlipListener that belongs to a
MetaKmsUpdate.
When a page flip is successfully queued, the ownership is transferred to
a MetaKmsPageFlipClosure that is part of a MetaKmsPageFlipData. In the
simple impl device, the MetaKmsPageFlipData is passed to
drmModePageFlip(), then returned back via the DRM event. In the future
atomic impl device, the MetaKmsPageFlipData is stored in a table, then
retrieved when DRM event are handled.
When the DRM events are handled, the page flip listener's interface
callbacks are invoked, and after that, the user data is freed using the
passed GDestroyNotify function, in the main context, the same as where
the interface callbacks were called.
When a page flip fails, the ownership is also transferred to a
MetaKmsPageFlipClosure that is part of a MetaKmsPageFlipData. This page
flip data will be passed to the main context via a callback, where it
will discard the page flip, and free the user data using the provided
GDestroyNotify.
Note that this adds back a page flip listener type flag for telling the
KMS implementation whether to actively discard a page flip via the
interface, or just free the user data. Avoiding discarding via the
interface is needed for the direct scanout case, where we immediately
need to know the result in order to fall back to the composite pipeline
if the direct scanout failed. We do in fact also need active discard via
the interface paths, e.g. in the simple impl device when we're
asynchronously retrying a page flip, so replace the ad-hoc discard paths
in meta-renderer-native.c and replace them by not asking for no-discard
page flip error handling.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Suspending might have changed the CRTC configuration, turning some off,
some on, etc. We need to update our internal representation of this
state, so that we know how to reconfigure upon resuming, e.g. what CRTCs
to turn off again.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Destroying an onscreen destroyes the gbm_surface, the gbm_bo's, and the
fb_id's. Doing this (drmModeRmFB() of the fb_id specifically), may on
some hw implicitly disable the CRTC of the plane that framebuffer was
assigned to. This would cause following atomic commit that attempts to
disable the CRTC to fail as disabling an already disabled CRTC is not
allowed.
It'd also mean we'd always disable the plane before having finished next
mode set, leaving it monitor content potentially empty when not really
necessary.
Solve this by keeping the CoglOnscreens (thus the gbm_surface, gbm_bo
and fb_id) alive until the following global mode set has completed, i.e.
the new state has been fully committed and applied.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
This makes "power save" (i.e. when you make a monitor go into power save
mode, or make it come out of power save mode), a per device action when
turning on power saving (power save being set to 'off'), and implicitly
handled when turning off power saving (power save being set to 'on')
when doing a mode set.
This is needed as with atomic mode setting, the configuration of DPMS
(Display Power Management Signaling), is replaced by directly turning on
or off CRTCs, and via the CRTC drm properties. Thus in order to handle
both with a common API, make that API high level enough for both cases
being covered.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Before we received new gamma updates via D-Bus and posted the update to
KMS directly. This won't be possible with atomic KMS, since one can only
update the state of a CRTC once per cycle.
Thus, to handle this, when configured by D-Bus, only cache the value,
and mark it as invalid. The next frame, the native renderer will pick
up the newly cached gamma value and configure the CRTCs accordingly.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
We cannot switch DPMS state to 'on' first, then mode set later, when
using atomic KMS. So when we're turning it on, just let the eventual
mode set handle DPMS too.
When switching DPMS to 'off', do it directly, synchronously, both by
setting the DPMS state and switching off CRTCs.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Before each frame is maybe redrawn, push any new cursor KMS state to the
pending update. It'll then either be posted during the next page flip,
or when the same frame finishes, in case nothing was redrawn.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
This makes it possible to post a symbolic page flip and frame callback,
meant to be used by immediate symbolic page flip reply when emulating
cursor plane changes using legacy drmMode* functions.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
We will soon need to use CoglOnscreen frame events communicate cursor
plane changes; this means we need to have a way to queue them without
going through any of the current APIs that can do so, i.e. the swap
buffer functions and direct scanout.
Add a function that just adds a frame info to the queue. The one who
adds it is responsible for emitting it too.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Just as with the frame clock, add an API to communicate that a frame did
not result in a presentation. This can't happen yet but will when we
emulate atomic cursor plane changes using legacy drmMode API.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
In constrast to notify_presented(), notify_ready() also returns the
state machine to the idle state, but without providing new frame
information, as no frame was actually presented.
This will happen for example with the simple KMS impl backend will do a
cursor movement, which will trigger a symbolic "page flip" reply in
order to emulate atomic KMS behavior. When this happen, we should just
try to reschedule again.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Don't mode set each CRTC in separate KMS updates, as reconfiguring one
CRTC might cause other CRTCs to be implicitly reset thus as well,
causing KMS return EBUSY if using atomic modesetting.
Prepare for this by compositing each CRTC first including adding steps
to the KMS update, but wait until all views has rendered at least once
before posting the initial update. After this each CRTC is posted
separately.
Using EGLStreams instead of normal page flipping seems to fail when
doing this though, so handle that the old way for the EGLStream case,
i.e. eglSwapBuffers() -> mode set with dumb buffer -> eglStream
"acquire" (resulting in page flip under the hood).
For this we also introduce a new error code so that we don't use client
buffers when doing mode sets, which could accidentally configure the
CRTC in a way that is incompatible with the primary plane buffers.
Do the same also when we're in power save mode, to only have one special
case path for this scenario in the regular swap-buffer path.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Instead of setting the frame result in the most generic layer, have the
backends do it themselves. This is necessary to communicate that a
swap-buffer call didn't really succeed completely to present the swapped
buffer, e.g. errors from KMS.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
This argument is intended to be used by clutter to be able to
communicate with the onscreen backend, that happens to be the native
backend. It will be used to pass a ClutterFrame pointer, where the
result of page flips, mode sets etc can be communicated whenever it is
available.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
ClutterFrame aims to carry information valid during dispatching a frame.
A frame may or may not include redrawing, but will always end with a
result.
A asynchronous page flip, for example, will result in a
CLUTTER_FRAME_RESULT_PENDING_PRESENTED, while a frame that only
dispatched events etc will result in CLUTTER_FRAME_RESULT_IDLE. Instead
of this being implicit, make the ClutterStageWindow implementation
handle this itself.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
The way drm events are handled depends on whether we're using atomic or
not. Lets move the handling to the implementation, so that later the
atomic backend can handle the event they it need to.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
If we reassign e.g. a cursor plane twice before it's updated, we need to
make sure the 'fb-unchanged' flag is correctly handled, so that if we
changed the fb first, then updated the assignment again only changing
the position, the new assignment should not be flagged with
fb-unchanged.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
When we e.g. try to post an direct client buffer scanout update, it
might arbitrarily fail; when this happen we still will want to post the
rest of the update when we try again after having composited the primary
plane. To do this, add a way to preserve the metadata of an update if it
failed, only dropping the failed plane assignments. This involves
unlocking a previously locked MetaKmsUpdate, so that e.g. a new primary
plane can be assigned.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Sealing is a one way operation, but in the next commit, the "seal" will
be broken, so to avoid missusing the "seal" terminology, rename related
methods and variables to use the term "lock" instead. E.g.
meta_update_is_sealed() is now meta_update_is_locked().
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
If a modeset is pending, it's likely that the cursor update will not
work; thus, wait with updating the cursor so that it's applied together
with the mode set update.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Something might want to affect the next update that is going to be
posted, but without posting it immediately. For example, changing the
cursor might need to wait for mode setting. Make it possible to get
feedback from posting the update, in order to gracefully handle any
errors.
Note, the API for notifiying about results take out the result listener
from the update, and notifies them in an open coded for loop. The reason
for this is that in the next commit we'll sometimes reuse updates, and
we only want notify about the results once.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Page flipping shouldn't necessarily be an actively requested action, but
happen implicitly depending on the given state. Thus, change the "page
flip" update into adding listeners for page flip feedback instead.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
This will later make it possible to pass cursor plane assignments,
together with a complete update including the primary plane, but not
failing the whole update if just processing the cursor plane failed.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
If posting an update resulted in an immediate error, don't communicate
this failure using the page flip feedback callbacks, but directly as a
return value.
This makes it possible for the direct client buffer scanout path not to
pass around flags triggering this behavior, meaning we can handle such
direct scanouts better.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Instead of a "post all pending updates", pass an update specific to a
single device. This gets rid of the awkward "combine feedback" function,
and makes it possible to queue updates to a multiple devices without
always posting them together.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Custom page flips are meant to allow using e.g. EGLStream API to
indirectly trigger page flip queueing, when the KMS API cannot be used
directly. This is really something that is specific to a device, so
instead of making part of the page flip API, make it a configuration of
the update itself.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Eventually the type of impl device will depend on the driver details, so
get that information before constructing the impl device. This commit
doesn't introduce any new usage of the details, it just prepares for
the future.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
This commit consolidates DRM buffer management to the MetaDrmBuffer
types, where the base type handles the common functionality (such as
managing the framebuffer id using drmModeAdd*/RMFb()), and the sub types
their corresponding type specific behavior.
This means that drmModeAdd*/RmFB() handling is moved from meta-gpu-kms.c
to meta-drm-buffer.c; dumb buffer allocation/management from
meta-renderer-native.c.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
For now feedbacks from an update are combined, meaning we might lose
error information. The feedback API may have to be reconsidered and
redesigned when planes gets a more front seat position.
This means we need to avoid trying to post updates if we're in power
save mode, as it may be empty.
Note that this is an intermediate state during refactoring that aims to
introduce atomic mode setting support, and we'll stop combining
feedbacks completely in the future.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Instead of a constructor method, use the type directly and handle error
reporting using GInitable.
The DRM capability setting is done before construction, as later it'll
determine what type of impl device should be constructed.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Instead of telling MetaKmsConnector fill a MetaKmsUpdate with connector
property changes, make the update itself aware of the changes, making
the impl side translate that to property changes.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Instead of having MetaKmsPlaneAssignment carry a low level property
list, set the actual state change, and then have the implementation
translate that into the necessary property changes.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
With the aim at always using the property table to fetch and parse
property metadata, move IN_FORMATS handling to the property table, using
the newly introduced parse vfunc.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Values may need to be processed and parsed in custom ways; make this
possible via the property table infrastructure using a callback.
Will be used for e.g. parsing rotation and formats.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Instead of relatively verbosely going through the DRM properties finding
the properties we care about and saving their ID's, add a more
declarative way to fetch property metadata. This'll allow for fetching
more property IDs with relatively less code, which will be useful for
the atomic backend.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
This contains a copy of a drmModeModeInfo, describing a mode. It also
has an unused pointer to the impl device it is associated with. It'll
later be used to get a blob ID for the mode.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1488>
Current Xwayland has marked the command line option "-listen" as
deprecated in favor of "-listenfd".
Use the pkg-config variable "have_listenfd" (if available) from Xwayland
to determine if we should use that option, to avoid a deprecation
warning when spawning Xwayland.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1682>
Mutter listens to two display connections, one for regular X11 clients
and another one for the so called "managed services".
Once an available display number is found for the regular X11 clients,
mutter would then redo the work to find another available display number
for the managed services.
Yet, it does so starting from the same initial display, which is a waste
of time since it just tried all displays to find the first available
one, so all these, including the regular display it just took, are now
in use.
So instead of starting over from the beginning when looking for a
display available for the managed services, continue from the next
display immediately after the one we found precedently.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1680>
Some X11 servers may not always create a lock file, yet mutter uses the
lock file to find a possible display number and then tries to bind to
the socket corresponding to that display number.
If it fails to bind, it will simply bail out. As a result, if an X11
server is already listening on that display but hadn't created a lock
file, mutter won't be able to start Xwayland.
To avoid that possible issue, make mutter retry with another display
for a given number of tries when binding fails even though the display
was supposed to be available based on the lock file presence.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1604
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1669>
The function choose_xdisplay() calls open_display_sockets() which calls
ensure_x11_unix_dir().
We don't need to do that from within the loop though, as the directory
/tmp/.X11-unix is the same regardless of the display number.
Move the call to ensure_x11_unix_dir() from open_display_sockets() to
choose_xdisplay() prior to enter the display loop.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1669>
In case of failure to open the display sockets, we would not propagatre
the error which can cause a crash when trying to show the error message.
Properly propagate the error to avoid the crash.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1669>
In the shutdown paths we check with the X11 display whether there's
remaining clients. However this happens in paths that happen after
the MetaX11Display vanished in the case of Xwayland crash.
Since in that situation the clients are forcibly vanishing too,
skip the client check.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1677>
This means backend implementations can have more control of the order of
how things are destroyed. To be precise, this will, in the next commit,
allow us to destroy the logind integration after the clutter backend
thus the libinput owning seat, that uses the logind integration to
release input devices.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1670>
We can't post tasks to the input thread when cleaning up the
MetaSeatImpl, as that will make the GTask complain about adding
references to a to be purged object. Avoid this by adding an explicit
meta_seat_impl_destroy() function that handles the destruction of the
MetaSeatImpl properly.
This also does more of the cleanup in the input thread, as that is where
it was managed. Will likely not make a difference as before this
happened after tearing down the thread, but lets tear down things in the
thread they were managed for good measure.
This fixes the last log spew I see right now when terminating mutter.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1670>
Closing automatically Xwayland once all relevant X11 clients are gone is
inherently racy, if a new client comes along right at the time we're
killing Xwayland.
Fixing the possible race conditions between mutter, Xwayland and the X11
clients may take some time.
Meanwhile, make that an experimental feature "autoclose-xwayland".
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1673>
<super> is considered a system modifier, and applications cannot use it
for keyboard shortcuts or as button modifier. It doesn't seem too much
of a loss taking <super>+scroll-event away as well, so that it becomes
available to gnome-shell/extensions.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1674>
Currently, mutter checks for the presence of X11 windows to decide
whether or not Xwayland can be terminated, when Xwayland is started on
demand.
Unfortunately, not all X11 clients will map a window all the time, an
X11 client may keep the X11 connection opened after closing all its
windows. In that case, we may terminate Xwayland while there are some
X11 client connected still, and terminating Xwayland will also kill
those X11 clients.
To avoid that issue, check the X11 clients actually connected using the
XRes extension. The XRes extension provides the PID of the (local) X11
clients connected to the Xserver, so we need to match that against the
actual executable names, and compare with a list of known executables
that we can safely ignore, such as ibus-x11 or gsd-xsettings.
We also check against our own executable name, considering that the X11
window manager is also an X11 client connected to the Xserver.
Also, XRes returning the PID of local clients only is not a problem
considering that Xwayland does not listen to remote connections.
However, if the user spawns a client remotely on another system using
ssh tunneling (ssh -X), only clients which actually map a window will
be accounted for.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1537
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1671>
Some GSettings schemas moved from g-s-d to gsettings-desktop-schemas
in g-s-d 40.alpha. 40.alpha.1 didn't include the corresponding
adjustment, so do a quick follow-up release to not abort on startup
with the latest g-s-d release.
ClutterClickEvent currently doesn't check if button press and release
happen within the drag threshold, which can be surprising sometimes.
Only emit the "clicked" signal if the button press and release happen
within the area of drag threshold.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1667>
Changes in games between fullscreen and windowed modes may trigger
chaotic situations where the buffer and the frame size temporarily
disagree, producing rectangles with negative width/height. This is
usually followed by other updates that bring the pointer constraint
up to date.
This makes cairo panic and return an "error" empty region, which breaks
deeper down when using the region rectangles to apply the pointer
constraint.
If we hit this situation, ignore the frame rectangle, and just go with
the buffer rectangle.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1655>
We didn't tear down the libinput objects in the right thread when
exiting, but did so after the input thread exited.
We also tried to destroy the libinput devices after the libinput context
was destroyed, which isn't allowed.
Fix these two issues by tearing down the libinput objects in a input
thread task that when done exits the input thread. This effectively
"flushes" the input thread tasks while destroying the libinput objects
just before the thread exits.
While it might fine to tear down libinput objects in an arbitrary (main
in this case) thread while making sure nothing pokes at it in parallel
(e.g. the input thread is gone), libinput is by definition single
threaded, and could theoretically make assumptions about this, and we
shouldn't cause any possible surprises here, so make sure to destroy it
all in the right thread.
This fixes an abort() on exit caused by an assert about invalid object
destruction in libinput.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1663>
The "seat" usually refers to the ClutterSeat (MetaSeatNative) object,
and "seat_impl" to the MetaSeatImpl object, but there were still a few
places where this wasn't adhered to so fix those.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1663>
When supported, this property allows the window system to apply a 3x3 color
correction matrix in order to transform colors from the window system's native
color space to the measured color space of a display device.
Query for this property and set the 'supports-color-transform' property in the
GetResource reply. Add support for the SetOutputCTM DBus method and plumb that
through to the server's CTM property.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1048>
Add a test case to check whether the stage views and frame clocks are
all correctly freed after working with timelines a bit and then
hotplugging and removing all monitors.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1632>
In clutter_timeline_set_actor() we currently always unset the
priv->frame_clock pointer of the old actor when a new actor gets set.
The priv->frame_clock pointer takes a reference on the ClutterFrameClock
though, so we leak ClutterFrameClocks here.
To fix it, simply rely on update_frame_clock(), which will call
set_frame_clock_internal() unconditionally to update the
priv->frame_clock pointer for us.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1632>
Currently, the NotifyPointerAxis method always assumes that the scroll
source is CLUTTER_SCROLL_SOURCE_FINGER.
This is however not always true and in some cases a remote desktop
client needs to submit a PointerAxis event with a custom axis step.
This is for example the case with high resolution mouse wheels, where
the NotifyPointerAxisDiscrete method is unsuitable.
In such cases NotifyPointerAxis needs to be called, but with the
intention that the scroll source is still a mouse wheel.
To solve this situation, don't assume the scroll source always to be
CLUTTER_SCROLL_SOURCE_FINGER.
Instead, add further flag options to NotifyPointerAxis, which allow a
remote desktop client to choose the scroll source.
This way a remote desktop client can choose what scroll source is the
most suitable one for the current scroll event.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1636>
The monitor texture is the final background image. It doesn't need to
have any alpha channel. Cross-fades (which is the process of rendering
*into* the monitor texture) still work just fine.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1665>
Because the framebuffer itself might be backed by a texture, which might
have mipmapping enabled. If so then rendering to the framebuffer will make
those mipmaps out of date.
Technically we are flagging the framebuffer's mipmaps as dirty *before*
they are, because the journal hasn't been flushed yet. But we need to do
it early because ideally the next flush will both write the offscreen
framebuffer contents and then read them for use in rendering to an onscreen
framebuffer. And the `mipmaps_dirty` flag needs to be set before the read,
so therefore we need to do it before the next journal flush.
Fixes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/3146
where the offscreen framebuffer in question is meta-background's
`monitor->fbo`.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1664>
Some effects such as the BrightnessContrastEffect try to skip rendering
by returning early with a FALSE return value in pre_paint() in cases
where the effect would not change the rendering. This stopped working
when effects were ported to paint nodes.
In the case of OffscreenEffects, like BrightnessContrastEffect,
pre_paint() is responsible for setting up the offscreen buffer which is
then used in paint_node(). However if pre_paint() fails, this buffer is
not created and attempting to use it will result in several error
messages and broken rendering.
Instead of trying to call paint_node() of the effect if pre_paint()
failed, just draw the actor.
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1576
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1658>
Unfortunately there's situations where we can't fully rely on only
crossing events here. One such situation is pointer visibility changes
due to touch interaction, or e.g. after closing the lid.
In these situations the pointer position stays the same, picks the
same actor, yet we want to see the right surface as the pointer focus
again in the wayland side.
This used to happen on the first motion event after the pointer
visibility change before commit ad3f2b0b863, use motion events again
for picking so we don't break these usecases.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1657>
These events may be emitted for touchpoints (in which case they contain
an event sequence). Ignore those as they are not relevant for pointer
picking, and shouldn't influence its focus.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1657>
When using alt-tab to switch between windows, on a key handler we
undo the compositor grab, which triggers a repick, which generates
crossing events, which are handled in place, and trigger these
reentrancy checks.
On one hand, we do intend these crossings to take effect in place,
rather than being queued (possibly after a number of already queued
events). On the other hand, we now outright discourage generating
events from random places (and hope to make it just not possible,
eventually) thus we can afford not to protect against reentrancy
caused by API misuse.
So just drop these checks, and let these crossing events be
properly handled.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1657>
This signal may be left dangling when disconnecting a device, and be executed
later on if the device is connected again, and mapped to other output. Make it
sure the signal handler is disconnected when unplugging the device.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1657>
Just because of implementation details, this is only relevant to Wayland,
and is done via ::effects-completed handlers there. Ideally, Clutter should
notice by itself about effects starting, finishing, and affecting picking.
Doing this in generic code seems slightly cleaner in the interim.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1654>
This is the very same code than meta_wayland_pointer_repick(),
made part of Clutter, so triggering repicks on the same pointer
coordinates is made easy.
The intention is to remove meta_wayland_pointer_repick() in favor
of this call.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1654>
In the wayland code we are in synchronization with the events that
the ClutterStage is managing at the moment. Asking the ClutterSeat for
the pointer position gets ahead of the current events, and may result
in imprecise coordinates sent in wl_pointer.enter.
To be in consistence with the motion events that might be already
queued, we should ask the stage for the last known coordinates.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1654>
Updating the state before emitting the CLUTTER_LEAVE event allows its
handlers to query the pointer actor, and avoid getting the same actor
again. Conceptually, this makes picking more "atomic", and the events a
notification of the change.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1654>
To clear a pointer constraint, the Wayland backend passes a NULL
constraint to the native input backend.
The new async API however tries to reference/un-reference the given
object to use it while running in a separate task, which leads to a
warning from GLib trying to g_object_ref()/g_object_unref() a non
GObject pointer.
To avoid that issue, simply set the data only if the given constraints
pointer is not NULL.
Suggested-by: Carlos Garnacho <carlosg@gnome.org>
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1587
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1652>
I messed up and released 40.alpha at the same time as 3.38.2, when it's
supposed to be in January. In order to re-align with the schedule, change
the upcoming version to 40.alpha2 so we don't have to skip a release and
will be back on track in time of 40.beta.
Turns out GObject implicitly notifies all properties by default as soon
as a property setter is called, no matter if the property actually
changed or not. One can opt-out of this behavior by setting the
G_PARAM_EXPLICIT_NOTIFY flag.
So since almost all our properties get notified explicitely (well,
except ClutterActors deprecated show-on-set-parent property), set this
flag for all properties of ClutterActor and ClutterStage now. This
significantly reduces the number of notify:: signals emitted on
ClutterActors, because in gnome-shell javascript we usually set GObject
properties directly, not by going through the extra setter method.
More cleanups can be done in the future, since this flag is suitable for
almost every property in Clutter and even Mutter.
This fixes a crash where we'd hit a newly introduced assertion in
on_device_actor_reactive_changed() of ClutterStage because
notify::reactive got emitted multiple times.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1651>
Due to a few reasons currently the updating of input devices after stage
relayouts isn't working right now. Since we now have a list of pointer
devices in ClutterStage, we can simply use that list and can avoid
asking the input thread. Also we no longer need to check whether the
devices are pointer devices, since our list only consists of pointer
devices.
So switch to ClutterStages private list of pointer devices, which also
includes the core pointer (as opposed to the list returned by
clutter_seat_peek_devices()). This fixes picking after relayouts.
Note that this doesn't catch every possible change that might need a
repick, actors might also need a repick after transformation changes or
in case their custom clip has been changed.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1634>
find_focusable_ancestor() may pick an ancestor window which is not
mapped or hidden, and setting focus on that window will fail.
Be a tad more selective when looking for a focusable ancestor, to reduce
the chance of meta_window_focus() not focusing the happy chosen one.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1643>
The function focus_default_window() optionally takes a MetaWindow
argument denoting a window that should not be focused.
That function calls focus_ancestor_or_top_window() which in turn
calls meta_window_focus() to pass focus to another window.
However meta_window_focus() gives no guarantee that the given window
will end up being the one focused, and can fail in various and creative
ways.
If that fails, we could possibly end up with the focus window being the
one to avoid, while the caller assumes focus was changed, going as far
as asserting that fact like meta_window_unmanage() does.
As a result, mutter may abort simply because meta_window_focus() failed
to set focus on the expected window.
To avoid that issue, check that the focus did not end up on the window
that we explicitly did not want, and if that's the case, simply fallback
to the default focus window.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/862
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1643>
If texture allocation failed for a sliced 2D texture, the alloc()
function would free the slices immediately, but not clear the pointer to
the slices array. When the code attempting to allocate the texture then
freed the texture object, the cleanup functions tried to free the slices
array again, since it wasn't NULL.
Fix this by clearing the slices array after freeing it.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1580
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1648>
As planned and prepared with the last commits, let ClutterStage take
care of tracking input devices and their respective actors. This means
we now can remove the old infrastructure for this from
ClutterInputDevice.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1633>
We're moving keeping track of input devices and their associated actors
out of ClutterInputDevice, this commit basically adds replacements for
clutter_input_device_update() and clutter_input_device_set_actor() to
clutter-main and shuffles the internals of those functions around a bit
for clarity.
clutter_stage_update_device() is made available in clutter-mutter.h
because we need to update the actor of input devices not only from
ClutterStage (when repicking after relayout), but also from
MetaWaylandPointer (inside meta_wayland_pointer_repick()).
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1633>
With the introduction of the input thread, we want to avoid modifying
ClutterInputDevices from the main thread, since they're owned and
updated by the thread.
There's one part of ClutterInputDevice that's still modified from the
main thread though, that is device-actors of pointer devices, and we're
going to move that state-tracking into ClutterStage instead.
So start that by adding the infrastructure to ClutterStage to keep track
of those things. It consists of two hashtables which associate devices
and touch sequences with actors, those hashtables get updated using
clutter_stage_update_device_entry() and
clutter_stage_remove_device_entry(), they can be queried by calling
clutter_stage_get_device_actor(), which will replace
clutter_input_device_get_actor().
clutter_stage_get_device_coords() is added and made available in
clutter-mutter.h because we need to get the coordinates when repicking
in meta_wayland_pointer_repick().
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1633>
The n_steps variable corresponds to the number of *pairs* of
texture lookups that the blur shader does. For example, when
n_steps = 1, the for-loop reads 1 pixel before and 1 pixel
after the current one.
Our blur shader is heavily inspired in WebRender's blur shader,
the biggest difference being that we calculate the gaussian
samples in the fragment shader itself, and not in the vertex
shader. (This could be an improvement in performance for the
future though!)
WebRender's blur shader calculates n_steps differently than what
we currently do, though. It calculates n_step in such a way that
at least 2 steps are performed for evey non-zero sigma value.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1646>
The shader already operates on floating point sigma, and
there's just no reason for us to force it to be an unsigned
integer. It's still important that sigma must be positive
though.
Make sigma a float, and make sure it's a positive number.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1646>
Yet another way to reduce the instruction count of the fragment
shader. Passing a pair of floats once is virtually free, compared
to computing horizontal and vertical on each fragment run.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1646>
New entries indirectly added to `pending_queue_redraw` during the loop
would make our iterator invalid and cause `g_hash_table_iter_next` to
fail without having visited all elements. That was seen as assertion
failures but also likely resulted in incomplete paint clips.
Now we steal the iterator's entry before such corruption can happen,
free it manually, and reset the iterator to the beginning on every
iteration. This is actually safe and efficient because we're removing each
entry we visit. So no time is wasted in resuming from the (new) beginning
of the hash table.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1557
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1615>
Add a new pair of APIs corresponding to CoglFramebuffer's draw_rectangles()
and draw_textured_rectangles(). They're generally more performance compared
to adding multiple single-rect operations. These variants are heavily used
by GNOME Shell's CSS implementation.
The op array is built to match cogl_framebuffer_draw_textured_rectangles()
always, which means it's a series of 8 floats composed (x1 y1 x2 y2) and
(s1 t1 s2 t2). To avoid adding new struct fields to ClutterPaintOperation,
which is a performance and memory sensitive structure, simply divide the
array length by 8 (which is guaranteed to be correct).
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1637>
Power saving changes in laptop panels enable/disable the attached
touchscreen input device, this is an asynchronous operation that
may be happening while the device is disappearing.
In fact, closing the lid is such perfect storm where both things
happen around the same time.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1644>
Currently, the MetaInputDeviceNative owns the libinput_device, with the
small catch that it is eventually finished in the main thread (as the
CLUTTER_DEVICE_REMOVED event keeps the last reference to it).
Make it sure that the libinput_device is destroyed in the input thread,
before giving away the last extra input device references.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1639>
Setting an ortho projection gives us pretty much the same result as
manually calculating the projection matrix. The ortho projection is
actually more "complete" than the custom projection we've been using,
as it also considers z-near and z-far, but in practice the generated
pixels are exactly equal.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1642>
There was an attempt to remove an unnecessary inclusion of a header
file, but only got so far as compile testing after having commented out,
but didn't remove the comment before creating a commit. This commit
fixes that mistake.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1640>
In the purely paint node based rendering future, ClutterEffects
simply add more paint nodes to the tree when painting the actor.
This is the leap to achieve that future.
Add paint nodes to pre_paint, paint, and post_paint, and move the
ClutterEffectNode creation to _clutter_effect_paint().
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1355>
The paint node tree that ClutterOffscreenEffect generates is
simple:
Root
|------------+
| |
Layer Pipeline
|
Actor
Right now, both pre-paint and ClutterLayerNode push the offscreen
to the framebuffer stack. That's harmless, and will go away soon
anyway.
The actor node is created and added in a separate function because
it'll be reused by the next commit.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1355>
Add a new ClutterPaintNode parameter to the paint_target() vfunc.
For now, create a temporary ClutterEffectNode that is passed to
paint_target() and immediately painted; next commits will move
this to upper layers.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1355>
The blur pipeline is still cached on ClutterBlurEffect, and we
simply update the uniforms when asked to create the pipeline.
Now that ClutterOffscreenEffect will use the blur pipeline, it
doesn't need to override the paint_target() vfunc anymore.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1355>
The most annoying aspect of ClutterOffscreenEffect right now, and
the reason for all its subclasses to override pre-paint, is that
the pipeline creation isn't under subclasses' control. That means
all subclasses must ask ClutterOffscreenEffect to run pre-paint
and create the pipeline, then they all create their own pipelines
to paint.
To reduce this complexity, add a new create_pipeline() vfunc to
ClutterOffscreenEffect. Next commits will port effects to use this
new vfunc.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1355>
ClutterLayerNode currently skips pushing the offscreen framebuffer when
no operations are set. This was added at the time because pushing and
popping was a synchronization point in Cogl, slow enough to force the
layer node to have this protective measure. Nowadays, pushing and
popping on the paint context is free.
Make ClutterLayerNode always push and pop in pre and post paint.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1355>
Xwayland in its main development branch now provides a pkg-config file
which can be used to determine Xwayland installation path and various
options enabled at build time.
Change mutter build system to optionally use that mechanism if
available. If not (as with Xwayland from the current stable branch),
keep the fallback mechanisms (basically search for the Xwayland and parse
its command line options to determine if initfd is available).
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1635>
ClutterBlurNode is a paint node based on ClutterLayerNode
that draws all children in an offscreen framebuffer, blurs
this framebuffer, and finally paints the blurred contents
according the the paint operations added.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1627>
Copy-paste error sneaked ClutterTransformNodeClass in
clutter_blit_node_class_init(). It wasn't problematic
because they both typedef to ClutterPaintNodeClass, but
fix it anyway.
Also switch to g_object_ref() in clutter_blit_node_new().
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1627>
Until now we would `clutter_input_focus_set_cursor_location` with
cursor-rectangle-in-physical-pixels + actor-location-in-stage-coordinates.
But those use different scaling factors so it only got the right answer
when the framebuffer scale was 1.0.
This directly determines the geometry of the invisible dummy cursor in
gnome-shell ibusCandidatePopup.js:
```
panelService.connect('set-cursor-location', (ps, x, y, w, h) => {
this._setDummyCursorGeometry(x, y, w, h);
});
```
And because it's invisible it wasn't obvious that it was wrong until you
enable `CLUTTER_PAINT=damage-region` and you can see its ghost at the wrong
offset and scale.
So now we `clutter_input_focus_set_cursor_location` using purely unscaled
stage coordinates. And `CLUTTER_PAINT=damage-region` shows that
gnome-shell's `_dummyCursor` is placed precisely over the visible cursor.
Fixes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/3399
and probably other IBus issues that arise when using framebuffer scaling.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1576>
`priv->cursor_rect` is stored in physical pixels, not local coordinates.
So unscale it before returning from `clutter_text_get_cursor_rect`, which
is explicitly documented as returning "actor-relative coordinates".
This went missed with fractional scaling support, and unnoticed as nobody
uses `clutter_text_get_cursor_rect` yet. But that will soon change.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1576>
Before now it was assumed that the only negative local coordinates would
be those introduced by `_clutter_actor_box_enlarge_for_effects`, and we
used the difference for `fbo_offset_x/y`. But that was misguided (of me)
because gnome-shell can give us paint volumes at negative coordinates too,
like when rendering `box-shadow` on the top or left edge of an actor.
The maximum extents of negative coordinates we might need to render are
in fact the (enlarged) left and top edges of the paint volume. So use
those as the FBO offset. This places the actor's local origin correctly
within the FBO and thus also ensures it's not over-clipped at the edges
of the FBO, which now line up with the enlarged extents of the paint
volume.
This fixes one third of
https://gitlab.gnome.org/GNOME/gnome-shell/issues/1090
The other fixes required are !1053 and gnome-shell!1417.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1538>
Apparently it can happen that a timeline tries to pick a frame clock
from an actor that's on a stage, but the actor still doesn't find a
frame clock and returns NULL.
This probably is the case when starting a timeline right after attaching
an actor to a newly created stage, so before the first stage-update
cycle. In this case clutter_actor_update_stage_views() will not have run
and the stage-actor will have priv->stage_views set to NULL even though
there are stage views.
To prevent this from happening, use the complete list of stage views
maintained by the backend when picking a frame clock for the stage.
This doesn't fix any issue appearing on master, but is correct
nonetheless.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1631>
$CI_COMMIT_REF_NAME can be a branch name or a tag, depending on the
pipeline, but our checkout script only deals with the former at the
moment.
Address this by rather than looking for a remote branch name, just
try to fetch the ref like we do for merge request pipelines.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1630>
We are currently not very good at communicating what's going in,
in particular for non-merge request pipelines where the output
is usually just "Using origin/master instead" (instead of what?).
Improve this by consistently communicate what we are looking for,
whether we found it, and what we end up doing.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1630>
To find XWayland output that should be the primary one, iterate through all
XWayland outputs, and compare their geometry to the geometry of the primary
logical monitor.
To avoid possible race conditions (Mutter's monitor configuration already
updated, but Xrandr not yet), set the output both after Randr notifications and
after 'monitors-changed' signal.
https://gitlab.gnome.org/GNOME/mutter/-/issues/1407
Signed-off-by: Aleksandr Mezin <mezin.alexander@gmail.com>
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1558>
When we're running under a polyinstantiated SELinux environment, we'll
likely start with an isolated and empty /tmp, meannig no /tmp/.X11-unix
directory to add things to. To make it possible to still function in
this kind of setup, make sure said directory exists.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1626>
Instead of using native backend platform data specifically, store
this info in ClutterMotionEvent. This includes time in usec since
it's just used for motion events, in the future it could make sense
to make these general to all events again, but it could make sense
to make ClutterEvent structs private before.
In order to express that a motion event has relative motion info,
the CLUTTER_EVENT_FLAG_RELATIVE_MOTION event flag has been added
for it.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1623>
We had code in both backends that sort of independently associated
sequences to slots. Make both transform slots to sequences the same
way, so they may share the implementation convert those back to slots.
This helper now lives in Clutter API.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1623>
We have this as platform-dependent data in the native backend, and
a bunch of fallback code done in place in the evcode users. Stop
making this platform-dependent data, and move it to the relevant
ClutterEvents.
The fallback code for the X11 backend case is about the same, but
now it is done directly by the X11 backend.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1623>
Making this an event is overly convoluted, accounting that we
emit the event, then convert it to a ClutterStage signal, then
its only consumer (a11y) sets the active ATK state.
Take the event out of the equation, unify activation/deactivation
of the stage in MetaStage, and use it from the X11 backend too.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1623>
It is not possible to express a blit operation using paint
nodes as of now. This is a requirement for GNOME Shell, e.g.,
to implement its blur effect.
Add a new ClutterBlitNode node that takes two framebuffers as
input, and blits the source framebuffer into the current one
according to added rectangles.
Because this paint node uses the rectangles in a different way
compared to all the other nodes, add an auxiliary method to
ensure all blit operations are valid.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1340>
ClutterLayerNode is the "offscreen framebuffer" node, that paints it's
child nodes in a separate framebuffer, and then copies that framebuffer
to the parent one.
It'll be useful to hand ClutterLayerNode which framebuffer and pipeline
to use, as this is a requirement for porting e.g. ClutterOffscreenEffect
and subclasses.
Add a new clutter_layer_node_new_to_framebuffer() API.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1340>
Some effects, such as ShellBlurEffect and ClutterOffscreenEffect, need
to make sure the actor is painted fully opaque. With ClutterActorNode,
however, that is currently not possible.
Add a new 'opacity' parameter to clutter_actor_node_new(). It follows
the opacity override heuristic, where -1 means disable, and anything
else is clamped to [0, 255].
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1340>
Introduce a new paint_node vfunc that, if implemented, allows
the effect to add nodes to a transient paint node that is
immediately painted. This is a transitional step until we
have fully delegated paint node rendering.
The most basic implementation of a ClutterEffect.paint_node
vfunc, and also the default implementation, is with an actor
node, as follows:
```
static void
foo_bar_paint_node (ClutterEffect *effect,
ClutterPaintNode *node,
ClutterPaintContext *paint_context,
ClutterEffectPaintFlags flags)
{
g_autoptr (ClutterPaintNode) actor_node = NULL;
actor_node = clutter_actor_node_new (effect->actor);
clutter_paint_node_add_child (node, actor_node);
}
```
This example gives the exact same behavior of simply calling
clutter_actor_continue_paint(). In the future, the paint node
itself will be a parameter of clutter_actor_continue_paint()
and we'll be able to simplify it event more.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1340>
ClutterEffectNode is a private ClutterPaintNode implementation
that does effectively nothing, but helps organizing the paint
node tree. It also helps debugging, since it can output the
effect class and name to the JSON debugging routines.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1340>
The idea of having a paint node tree is that we don't really need
to retrieve the framebuffer from ClutterPaintContext. For example,
ClutterLayerNode draws into an offscreen framebuffer; if any child
of a layer node needs to retrieve a framebuffer to draw, the layer
node's offscreen framebuffer should be used.
However, clutter_paint_node_get_framebuffer() goes straight to the
root node of the tree, skipping any potential paint nodes with a
custom framebuffer.
Modify clutter_paint_node_get_framebuffer() to walk up the paint
node tree until a node with a custom framebuffer appears. In many
cases, this will end up either in dummy or layer node's custom
framebuffer implementations.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1340>
Ensure that color_ptr gets set, and avoid color_char usage too in
that case. Fixes:
../../../../Source/gnome/mutter/src/backends/native/meta-monitor-manager-kms.c: In function ‘meta_monitor_manager_kms_set_crtc_gamma’:
../../../../Source/gnome/mutter/src/backends/native/meta-monitor-manager-kms.c:370:7: warning: ‘color_char’ may be used uninitialized in this function [-Wmaybe-uninitialized]
370 | g_string_append_printf (string, " %c: ", color_char);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../../../../Source/gnome/mutter/src/backends/native/meta-monitor-manager-kms.c:351:12: note: ‘color_char’ was declared here
351 | char color_char;
| ^~~~~~~~~~
../../../../Source/gnome/mutter/src/backends/native/meta-monitor-manager-kms.c:391:36: warning: ‘color_ptr’ may be used uninitialized in this function [-Wmaybe-uninitialized]
391 | (*color_ptr)[i]);
| ~^~~~~~~~~~~
../../../../Source/gnome/mutter/src/backends/native/meta-monitor-manager-kms.c:350:24: note: ‘color_ptr’ was declared here
350 | unsigned short **color_ptr;
| ^~~~~~~~~
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1624>
Assert that the region is created, thus we passed a valid enum value
to the get_scaled_region() function. Fixes:
../../../../Source/gnome/mutter/src/compositor/meta-surface-actor.c: In function ‘get_scaled_region’:
../../../../Source/gnome/mutter/src/compositor/meta-surface-actor.c:113:10: warning: ‘scaled_region’ may be used uninitialized in this function [-Wmaybe-uninitialized]
113 | return scaled_region;
| ^~~~~~~~~~~~~
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1624>
Fixes a compiler warning with -Wmaybe-uninitialized enabled:
../../../../Source/gnome/mutter/clutter/clutter/clutter-actor.c: In function ‘clutter_actor_paint’:
../../../../Source/gnome/mutter/clutter/clutter/clutter-actor.c:3808:50: warning: ‘result’ may be used uninitialized in this function [-Wmaybe-uninitialized]
3808 | else if (result == CLUTTER_CULL_RESULT_OUT && success)
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~
Which might presumably happen in the unlikely case that there's no clip
frusta.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1624>
It was a test case in the Wayland test client directory, but it wasn't a
Wayland test client but a standalone test linking to libmutter. Since it
uses rlimit to implement certain aspects of the test, it can't be made
part of the regular unit tests, as that means any test running after
being stuck with the rlimit set, thus keep it standalone, but at least
run it as part of the test suite.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1557>
The XIOErrorExitHandler expects (Display *, void *) whereas mutter uses
(Display *, MetaX11Display *).
That causes a warning at build time:
warning: passing argument 2 of ‘XSetIOErrorExitHandler’ from
incompatible pointer type [-Wincompatible-pointer-types]
813 | XSetIOErrorExitHandler (xdisplay, x_io_error_exit, display);
Actually, the MetaX11Display is not even used, so we can just use the
expected API and ignore the value.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1621>
The stack and stack tracker tend to cause missed frames from time to
time, especially when there are many open windows. Add some
instrumentation to make it this easily verifiable when profiling.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1616>
While browsing sysprof profiling reports, I saw surface-commit taking
significant times sometimes; trace attach too, to see whether such
things are due to e.g. texture uploads.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1616>
Constantly manipulating the stack caused severe stalls (several seconds)
with many open windows when switching workspaces. The cause for this was
that each show/hide call dealt with the stack in isolation, meaning if
you hid N windows, we'd manipulate and synchronize the stack N times,
potentially doing synchronous calls to the X server while doing so.
Avoid the most severe stalls by freezing the stack while calculating
showing; this made the worst case go from several seconds to around
10-20 ms, which is still bad, but by far not as bad.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1616>
Figuring out the MetaSeatImpl this much indirectly is fairly awkward when
the keymap is only updated from the MetaSeatImpl, pass instead the seat
impl's xkb_state, as we have it handy in all the places this is called.
This will not break on NULL seats during initialization, should the numlock
state be restored from previous sessions.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1556
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1612>
The API allows for invalid barriers to be created; in an X11 session,
this could result in involutary early exit, so guard against those with
soft asserts. These will be logged in the journal as warnings, but will
avoid the crash unless compiled out.
Note that this doesn't fix the bug, it just makes it more detectable.
Related: https://bugzilla.redhat.com/show_bug.cgi?id=1901610
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1611>
Make it impossible to add individual includes of input thread objects.
This must go through meta-input-thread.h now, which should be enough
to make anyone think it twice.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
They're a dime a dozen. If it gets called exclusively from the
input thread, it got one. Hopefully these breadcrumbs will be
enough so people don't lose their way here.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
This (now) doesn't change anything in regards to the API that the UI
thread should access from the MetaSeatImpl. The MetaInputDeviceNative,
MetaInputSettings and MetaKeymap objects are now considered owned by
the input thread, as well as all of libinput objects.
The MetaEventSource now dispatches events in a GMainContext that is
the thread default to this thread, and all UI-thread-accessible API
(seat and virtual input device API) will be handled in a serialized
manner by that same input thread.
The MetaSeatImpl itself is still considered to be owned by the caller
thread, and all the signals that this object emits will be emitted in
the GMainContext that is default at the time of calling
meta_seat_impl_new().
The MetaInputSettings configuration changes will likewise be handled
in the input thread, close to libinput devices.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
Instead of going through the event queue, stage handling code, and
back to the input device via a vmethod call, do this directly in the
MetaSeatImpl. This is not too different from X11, where everything
happens inside the backend.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
All that is left in the "public" struct is all state that ClutterStage
delegates on ClutterInputDevice. That should move somewhere else, but
not here, not now.
All private fields belong to construct-only properties, with only getter
API, and idempotent vmethods (except keyboard a11y, atm). This should
be enough to make ClutterInputDevice obviously thread safe, outside the
backend.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
This API is the one accessed from different bits of the UI thread,
make it "async" (it's basically one-way setters, so API stays the same
in the surface) and able to run in the MetaSeatImpl main context.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
Simplify the handling of numlock state, so it can be entirely handled
within the input thread. Since the saving/restoring is triggered inside
each backend code, there's no need anymore for meta_backend_set_numlock().
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
Push it a little bit down to the MetaSeatNative. As both the UI thread
and the input thread are interested in dealing with the xkb_keymap and
it is not meant to be used in different threads, keep 2 separate copies
around.
The keyboard map will always be set from the UI thread, so the xkb_keymap
owned by the MetaSeatNative (owned by the UI thread) can be considered
canonical.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
Don't let the vfuncs (meant to be called from the UI thread) deal with
xkb state itself. Instead store the current state in struct fields, which
is then fetched in vfuncs.
This makes the keymap able to be used from the UI thread, while being
maintained by the input thread. Same caveats apply than
clutter_seat_query_state(), you are asking for the most up-to-date state,
but it still may be changing under your feet.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
Wrap all keyboard state updates, and all pointer/stylus/touch cursor
position with a write lock, and ::query_state() (The only entrypoint
to this state from other threads) with a read lock.
The principle is that query_state may be called from different threads
(UI so far, but maybe KMS too in the future), while the input thread
may (or may not) be updating it. This state is fetched "atomically"
(eg. x/y will be consistently old or new, if the input thread were
updating it at the same time).
There's other places deep in backends/native that read this state,
they all will run in the input thread, so they count as "other readers"
to the other thread. Those changes are already mutually exclusive with
updates, so they don't explicitly need the RW lock.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
While barriers will be added from the main thread, the native barrier
manager will sit close to the MetaSeatImpl in its own thread. Add the
necessary locking so that we can pass MetaBarrierImplNative from the
UI thread to the input thread, and ensure the MetaBarrier signals are
still emitted in the UI thread.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
Depending on the backend, we want to integrate this object at different
levels. It will sit close to the MetaBackendX11/MetaSeatX11 in X11, but
it will be put deep down with MetaSeatImpl in the native backend, in a
separate thread.
Since we can't depend on a single object type, nor are able to track
ClutterSeat signals neatly, make this API something to be called
explicitly by backends.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
These changes will happen in the input event management code, so let them
be emitted via the MetaSeatImpl, as that's what we'll have neat access to.
The ClutterSeat signals are now emitted from there.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
Move most of the functional bits (those meant to run on a standalone
thread) to a MetaSeatImpl object. This object is managed by the MetaSeatImpl
and not exposed outside the friend MetaSeatNative/MetaInputDeviceNative/
MetaInputSettings classes.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
Banish MetaInputSettings from MetaBackend "public" API, it's now meant to
spend the rest of its days in the backend dungeons, maybe hanging
off a thread.
MetaInputMapper replaces all external uses.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
Delegate on the MetaInputMapper all matching of inputs and outputs,
including configuration. Other secondary aspects, like output
aspect ratio for tablet letterbox mode, or toggling attached devices
with power saving changes, are also moved here.
This makes MetaInputSettings independent of MetaMonitorManager,
all interaction with it happens indirectly via public API entrypoints.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
Rename the set_tablet_keep_aspect() vfunc into a set_tablet_aspect_ratio()
one that takes an aspect ratio double, instead of leaking monitor info
into subclasses to let them all figure out this number themselves.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
We have 2 sources (this one in MetaSeatNative, and the one in
MetaBackend) dispatching ClutterEvents to the stage. Make the
MetaSeatNative one exclusively about dispatching the libinput
queue, and leave ClutterEvents to the other.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
This will resort to SW rendering if this cursor renderer does not
own the MetaKmsCursorRenderer, so it's pretty much equivalent thus
far, except we may now implement logic to flip the kms cursor renderer
around.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
We actually have a set_send_events() vfunc that can enable or disable
devices at the libinput and X11 input driver level, so use that. A
positive side effect is that those layers will leave the device at
a consistent idle state (as opposed to going mute maybe amid user
input).
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
We are aiming for a split of HW and SW cursor rendering management.
Given the HW plane is a limited resource and the amount of cursor
renderers may be >1 (due to tablets, even though we currently use an
always-software cursor renderer there), it would ideally be able to
switch between renderers.
Being MetaCursorRenderer not really a singleton, having cursor
inhibitor accounting here doesn't pan out. Make it MetaBackend API
so all cursor renderers get the same picture.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
These use now more of a "pull" model, where they receive update
notifications and the relevant input position is queried, instead
of the coordinates being passed along.
This allows to treat cursor renderers all the same independently
of the device they track. This notifying of position changes should
ideally be more backend-y than core-y, a better location will be
figured out in future commits.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
Instead of letting the wayland bits maintain an always-software
cursor renderer, let the cursor renderer be managed by the backend,
and only hook to it (as we do for pointer cursor) in the wayland
bits.
ATM, make the cursor renderer still always-software, although
ideally we should allow moving the HW cursor management between
renderers.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
Different devices may get standalone cursor renderers, add this API
to adapt slowly to this. The meta_backend_get_cursor_renderer() call
still exists, but shortcuts to the mouse pointer's renderer (as it
actually did before).
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
Use a new set in MetaInputDeviceNative, this coexists with
ClutterInputDevice coords for the time being. This API will
eventually be only accessed from the input thread.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
This will query a device state; position and modifiers.
Note that ClutterSeat subclasses don't implement the vmethod yet,
so calling clutter_seat_query_state() may crash ATM.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1403>
While multiple built-in panels isn't actually supported in any
meaningful manner, if we would ever end up with such a situation, e.g.
due to kernel bugs[0], we shouldn't crash when trying to set an
'external only' without any external monitors.
While we could handle this with more degraded functionality (e.g. don't
support the 'switch' method of monitor configuration at all), handle it
by simply not trying to switch to external-only when there are no,
according to the kernel, external monitors available. This would e.g.
still allow betwene 'mirror-all', and 'linear' switches.
The crash itself was disguised as an arbitrary X11 BadValue error, due
to mutter trying to resize the root window to 0x0, as the monitor
configuration that was applied consisted of zero logical monitors, thus
was effectively empty.
[0] https://bugzilla.redhat.com/show_bug.cgi?id=1896904
Related: https://bugzilla.redhat.com/show_bug.cgi?id=1899260
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1607>
xkb recently gained support for user-specified keymaps, which means we
can no longer assume that the configuration data is necessarily fully
complete or correct; and the configuration language is quite a labyrinth,
so it's easy to get wrong. If setting the keymap fails, leave it in
whatever state it previously had, since that seems preferable to crashing
with a NULL pointer dereference.
Resolves: https://gitlab.gnome.org/GNOME/mutter/-/issues/1555
Signed-off-by: Simon McVittie <smcv@debian.org>
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1605>
Testing points and rays against boxes is substantially cheaper - in
fact, almost trivial - compared to triangles. Check if the actor's
paint volume doesn't intersect with the current pick point / ray,
and skip recursing altogether in those cases.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1520>
The cached pick mode never actually cached anything, since it is
always, unconditionally reset when painting. Furthermore, next
commits will cull out actors during pick, which makes the pick
stack uncacheable.
Remove the cached pick mode.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1520>
Users of Debian arm64 (aarch64) report that on at least some GPUs
or screens, after time-based screen blanking has occurred, it is not
possible to unlock the screen. Bisection indicates that this regressed
in commit 209b1ba3, so presumably this is because a refresh rate of 0
is reported while the screen is blanked, leading to the frame clock
pausing forever.
Fixes: 209b1ba3 "clutter/frame-clock: Adapt refresh rate from to frame info"
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1536
Bug-Debian: https://bugs.debian.org/974172
Signed-off-by: Simon McVittie <smcv@debian.org>
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1601>
It's currently possible that some last_paint_volumes don't get updated
during a paint cycle, this can happen when a ClutterOffscreenEffect is
used: The offscreen effect might skip painting the content and the
children of an actor because it uses its own offscreened texture
instead. This means the offscreen effect doesn't call
clutter_actor_continue_paint(), and thus the the last_paint_volumes of
the children won't be updated.
Now one might think that isn't a problem, because as soon as a child
changes it's size or position, the offscreened texture would get
invalidated and clutter_actor_continue_paint() would get called. It's
not that easy though: Because the last_paint_volume includes all the
transformation matrices up to eye-coordinates, it has to be updated on
any changes to matrices, which includes position/transformation changes
to any actor up the hierarchy.
Now that's where get into problems with the offscreen effect: In case of
transformation changes to the offscreened actor or an actor up the
hierarchy, the offscreened texture won't get invalidated (that makes
sense, we can simply paint it transformed) and the last_paint_volumes
won't get updated even though they should.
This leaves us around with outdated last_paint_volumes where
last_paint_volume_valid is still set to TRUE. It can cause issues with
culling and clipped redraws.
So fix that by ensuring that all children that would get painted by
Clutter get their last_paint_volumes updated in case a ClutterEffect
decided not to call clutter_actor_continue_paint().
This ignores the case where a paint() vfunc override does the same and
doesn't call clutter_actor_paint() on children. Let's ignore this case
for now, there shouldn't be any implementation which does that and
ideally in a world that's painted solely by ClutterContent, we can get
rid of that vfunc in the future.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1591>
The first comment isn't really needed anymore since
the is_full_stage_redraw_queued() underneath is quite self-explaining.
Also rephrase the second comment a bit, including that
_clutter_paint_volume_get_stage_paint_box() does the aligning to the
pixel grid.
Finally, the last comment also looks out of date since we do that
rounding inside _clutter_paint_volume_get_stage_paint_box(), so remove
it.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1511>
We currently pass actors a reference to their associated
ClutterStageQueueRedrawEntry when queueing a redraw. This "splitting" of
the ownership of the entry has introduced quite a few bugs in the past
and is hard to follow.
So give up the "splitting" of the ownership and exclusively handle those
entries inside ClutterStage. To still allow removing the entry when an
actor gets unrealized introduce clutter_stage_dequeue_actor_redraw()
similar to what we already have for relayouts.
To be able to efficiently find entries when actors queue redraws, make
pending_queue_redraws a GHashTable, which fits quite nicely and also
allows removing the QueueRedrawEntries actor pointer in favour of the
key of the hashtable.
Since the struct is now private to ClutterStage, we can also rename it
to QueueRedrawEntry.
While at it, also sneak in the removal of the leading underscore from
clutter_stage_queue_actor_redraw().
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1511>
The "queue-redraw" signal is not used anywhere in Clutter and we now
also removed the vfunc implementation of the stage. So stop emitting it
and remove it, but keep the propagate_queue_redraw infrastructure to
make sure clones still get their redraws queued.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1511>
Since we now decoupled the "queue-redraw" signal from creating the stage
clip, we can move signal emission into
_clutter_actor_queue_redraw_full() and emit the signal right away when
queueing a redraw on an actor. With that we now no longer have to
accommodate for the stage pending_queue_redraws list changing while
iterating over it.
To ensure we don't emit the signal too often when multiple redraws are
queued on one actor, use the propagated_one_redraw flag to limit the
number of emissions to a single one for every update cycle.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1511>
Putting together the redraw clip of the stage never really fitted nicely
with the "queue-redraw" signal emission, it forces us to emit the
signals in a batch and we also use a weird trick to get the old paint
volume that's already on-screen into the final redraw clip (we call
_clutter_actor_propagate_queue_redraw() on the stage).
So start breaking up this association by making the stage explicitely
request the redraw clip from the actor and removing the
ClutterPaintVolume argument from _clutter_actor_finish_queue_redraw().
This is done by adding a private function
clutter_actor_get_redraw_clip() which returns our old (currently
visible) paint volume and the new paint volume.
This also allows removing the check whether a full stage redraw has been
queued in clutter_actor_real_queue_redraw() and we can now just stop the
signal emission if a propagation happened at least once.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1511>
This adds 64 bit ARM support to parts of the CI pipeline, more
specifically, we now:
- Build a 64 bit ARM container image alongside the 64 bit x86 one.
- Build test mutter on 64 bit ARM
- Run tests on 64 bit ARM
- Check coverage on 64 bit ARM
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1577>
The most common case for Clutter is 2D axis-aligned actors, which
maintain these properties even after projecting to eye coordinates.
In those cases, we can use a simpler hit test by checking against
boxes.
Not only this is simpler, but this maintains an important aspect
of picking that is a requirement for Clutter: watertightness. Even
though the triangles checks do work on x86_64, they do not guarantee
watertightness. This breaks tests on ARM.
Use graphene_box_t to hit-test axis-aligned 2D actors.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1599>
The image registry must be enabled for the CI to work, as the container
image from the upstream registry is placed (as a reference, more or
less) in the local registry.
Recently, ci-templates got a bit more helpful if this happened, and will
fail up front and generate a JUnit report with a message describing the
issue.
Update to this version.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1587>
clutter_actor_paint() implements a clear preference for custom clips
over clip_to_allocation: If a custom clip is set, clip_to_allocation is
ignored.
Since the paint volume reflects what Clutter is going to paint, we
should handle it the same when putting together our paint volume: So
first handle custom clips, and if one is set, use that. Then handle
clip_to_allocation, and if that's set, use that. And finally, if both
aren't set, union our allocation with the children paint volumes to get
the building volume.
clutter_actor_paint() also doesn't check whether the custom clip is
empty: If that's the case, it will simply not paint anything. Given that
that's allowed by clutter_actor_paint(), the paint volume should also
follow here and return an empty paint volume in case the custom clip is
empty.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1592>
When a custom clip is set for an actor, this actor is not going to allow
any painting outside that clip. That includes the children, which may
also not paint outside that clip.
Now in case clip_to_allocation is set to TRUE, we already already do the
right thing and simply use the allocation as our paint volume, ignoring
the volumes of our children. The same should be done for the custom
clip, so also stop the process of building the paint volume once we see
that a custom clip is set and simply use that clip.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1592>
In case we only have a single view (or there's only one view left to
check and the actor is visible on previous views) we can take a short-
cut, saving a region allocation and some calculations.
While on it, declare float numbers in '.f' style to make them more
recognizable.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1596>
Commit 03c69ed8 ("Do not go past size hints on resize") was meant to
ensure the size hints set by the client would be honored during resize,
as going past those values could cause the window to move on resize.
However, it did so by calling ensure_size_hints_satisfied() which works
with the frame rect rather than the client rect. As a result, the
minimum size enforced would end up being larger than expected with
client-side decorations.
Use meta_window_maybe_apply_size_hints() instead which automatically
adjusts for client size.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1542
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1594>
It's the rules that guard the pipelines, deciding when they are allowed
to run or not. The 'only_default' comes from the fact that in the past
they set the only: field.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1588>
gnome-shell depends on gdm's client library at runtime, but the
new pipeline-built image no longer provides it.
Add it back, but built it from source instead of using the Fedora
package, so we don't draw in all of gdm's runtime dependencies
(which includes a full GNOME session including gnome-shell).
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1590>
This allows profilers to trace the callers of whatever is spending the
most time on the GPU, and to measure render times more accurately.
Previously such information was hidden as it completed in the
background (asynchronously) after we call swap buffers.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1327>
When using CLUTTER_DEBUG_PAINT_DAMAGE_REGION, the blue swap region is
always a superset of the red redraw region. So painting both in full (since
the previous commit) just meant the red region was overdrawn and came out
purple. That doesn't provide enough visual contrast, changes the user
experience unexpectedly and reduces performance.
So just subtract the redraw region from the swap region. This way the
red redraw region is always red, not purple.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1571>
Previously when CLUTTER_DEBUG_PAINT_DAMAGE_REGION was set, that would lead
to has_buffer_age==FALSE, which would lead to use_clipped_redraw==FALSE
which would mean swap_region was always empty. And so the blue region of
CLUTTER_DEBUG_PAINT_DAMAGE_REGION was always empty, *and* fb_clip_region
was always the full view rectangle which is not useful for debugging.
Now when CLUTTER_DEBUG_PAINT_DAMAGE_REGION is set, we don't let that
affect use_clipped_redraw, which means fb_clip_region is calculated
realistically.
But that's not enough. Calculating fb_clip_region properly with
CLUTTER_DEBUG_PAINT_DAMAGE_REGION would still lead to colouring artefacts
left on screen from previous frames that don't apply to the current frame.
So to fix that we also paint_stage for the whole screen every time when
using CLUTTER_DEBUG_PAINT_DAMAGE_REGION.
So now you will only ever see red and blue shading that's applicable to
the current frame, and no artefacts from the previous frames.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1535
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1571>
It was being painted without scaling and offsetting so would only look
right at scale 1.0 and only on the view with origin (0,0). Now we include
the framebuffer scale and view origin it will be painted in the correct
location.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1571>
ci-templates had a bug where it failed to find the ready-built container
images on the upstream repository due to invalid assumptions about the
relationship between upstream repository and image registry. This
resulted in the container image to be rebuilt, wasting valuable CPU
cycles and storage space. Bump the included ci-templates template to
include the fix for this.
See https://gitlab.freedesktop.org/freedesktop/ci-templates/-/issues/22.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1583>
only/except keywords where deperecated in favor of rules.
Since we started using GNOME/mutter!1525 it introduced
a second pipeline being run for each commit.
Detached pipelines are the only way to access CI_MERGE_REQUEST_*
variables, and if we disable normal pipelines you will need to
create wip/spam MRs in order to run the tests.
This reworked rules makes it so, the normal pipeline needs manual
interaction to be started, and the detached/MR pipleines is always
run.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1578>
Since we schedule frames for each stage view seperately now, surfaces receive
frame callbacks for each stage view they are visible on.
Only emit frame callbacks for the primary stage view.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1468>
Our main use case of `is_obscured()` is frame callback emission.
Since we now support stage views running at differt speeds, we
need to know whether an actor is visible on a specific stage view
in order to schedule frame callbacks accordingly.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1468>
Commit e28c1ab4 added a hints_have_changed() function to only
recalculate windows features when the WM_NORMAL_HINTS change.
That function hints_have_changed() however was merely checking whether
the various XSizeHints flags where flipped, which is not sufficient
because the hints may remain the same while the actual values are
changed.
Not checking for the actual value differences would prevent some windows
from being able to switch fullscreen.
Improve the helper function hints_have_changed() to check not only for
flags being flipped, but also for the values being changed for each
relevant XSizeHints flags being set currently.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1534
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1566>
Sometimes the automatically selected primary GPU isn't suitable with no
way to make an well educated guess to do it better. To make it possible
for the user to override the automatically calculated default, make it
possible to override it using a udev rule.
E.g. to select /dev/dri/card1 as the primary GPU, add a file e.g.
/usr/lib/udev/rules.d/61-mutter-primary-gpu.rules (path my vary
depending on distribution) containing the fellowing line:
ENV{DEVNAME}=="/dev/dri/card1", TAG+="mutter-device-preferred-primary"
Reboot or manual triggering of udev rules to make it take effect may be
required.
Related: https://gitlab.gnome.org/GNOME/mutter/merge_requests/1057https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1562
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1562>
Instead of aborting with an error, display a half transparent gray
square instead of cursors and log a warning in the journal, allowing the
user to fix their system withotu having to rely on switching to a TTY.
It will be immediately obvious the cursor is silly looking, which will
be a better hint than just aborting.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1428
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1563>
If review fails, it'll fail very early in the pipeline, but we won't see
the test case failure until the whole pipeline succeeds, which might be
10 minutes later.
To avoid sitting there wondering why it failed, let the time consuming
jobs wait until the review stage, which tends to take less than 20
seconds, succeeds. This way the review test result will be presented
earlier.
This changes the pipeline to run the check-commit-log job also for
non-merge requests, with the difference being that it will pass
immediately if it's not a merge request.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1549>
Clutter is a "scope", so everything under should be its own
"sub-prefix", e.g. changes to ClutterActor should be prefixed
'clutter/actor', ClutterFrameClock with 'clutter/frame-clock',
CoglFramebuffer with 'cogl/framebuffer' etc.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1549>
So that it can merge things with an even more locked down master branch.
It's given the name "Marge Bot", and the E-mail
"gnome-shell-list@gnome.org" (even though it's being decomissioned) to
make validators happy.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1564
It'd happen that the test runner would get CPU starved, and not see the
frame-clock changed notification before the timeline stopped. Decrease
the risk for this by moving the initial position of the actor having its
position transitioned to be closer to the view edge. This means the
frame clock will be changed earlier, increasing the chance of the
timeline not stopping before the relayout happens.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1555
For X11 grabs, the pattern matching mechanism would simply ignore
applications which have neither WM_CLASS nor WM_NAME set.
When dealing with an override redirect window however, it is not
uncommon that these window have neither value set as these window are
supposed to be ignored by the window manager.
When the WM_CLASS or the WM_NAME is not set by the client, assume the
value is empty so the pattern matching can allow for these.
https://gitlab.gnome.org/GNOME/mutter/-/issues/1249
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1254>
Monitor tile info is possible to fetch when RANDR version 15 is exposed
by the X11 server. We had inverted the check meaning that only if older
versions were advertised would we attempt to init the tile information.
Fix this guard, thus fix monitor tiling on X11.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1524
As explained in https://gitlab.gnome.org/GNOME/mutter/-/issues/1494,
with commit 29caa5bea576ed056aa6c82de192426abe6019ae we stopped queueing
a relayout for the parent of the removed actor in
clutter_actor_remove_child_internal(). This relayout was, as opposed to
the relayout in clutter_actor_real_hide()/clutter_actor_real_unmap(),
queued unconditionally without looking at the parents NO_LAYOUT flag.
Now while that relayout in clutter_actor_remove_child_internal() would
do unnecessary work if the parent had the NO_LAYOUT flag set, it did
also queue a redraw of the parent, which is necessary in any case.
So by removing that relayout in clutter_actor_remove_child_internal(),
we stopped queueing redraws for NO_LAYOUT parents when a child gets
removed from the scenegraph. This caused bugs where the texture of the
child would be left visible on the screen even though the child got
destroyed.
To fix this, make sure again that we always queue a redraw on the parent
when unmapping a child.
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1494
Not calling libinput dispatch in the backend constructor defeats the
logic in post init as the device added events have not been processed
yet.
So instead of trying to guess the cursor initial visibility, simply
update it along when devices get added.
Additional benefit, we do not need to walk the all device list looking
for touchscreens anymore, we just need to check the device being added
since the current logic is to hide the cursor as soon as a touchscreen
is found.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1534
At startup, libinput dispatch is called from the MetaSeatNative
constructed callback.
That means that we may get libinput events even before the default seat
is set.
In turn, processing those events may trigger the use the default seat
while it's still not set yet, and cause a crash of gnome-shell/mutter
at startup.
A simple reproducer for this is to start gnome-shell/mutter with a
tablet connected and the stylus in proximity, the proximity event will
cause gnome-shell/mutter to crash at startup.
To avoid that issue, avoid dispatching libinput events early from the
MetaSeatNative constructed callback, those events will eventually get
processed when the seat and the backend are all setup.
https://gitlab.gnome.org/GNOME/mutter/-/issues/1501https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1534
Commit 8bdd2aa7 would offset the window position by the difference
between the configured window size and the committed size from the
client to prevent the window from drifting while resizing.
This, however, did not take into account the actual geometry scale, so
when using any scale greater than 1, the window would rapidly drift away
due to that offset.
In order to solve this, we need to make sure we store away the pending
window configuration in the stage coordinate space, in order to not
loose precision. When we then calculate the offset given the result from
the client, it'll use the right scalars, while before, one scalar was in
surface coordinates, while the other in stage coordinates.
https://gitlab.gnome.org/GNOME/mutter/-/issues/1490
We're just a "ninja coverage" away from getting the full report,
it is not that much extra to the pipeline. Also, fix bugs that
accumulated since the manual task was added, e.g. let the full
built+tested tree available on the coverity stage.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1535
We want the bounding box so `ceilf` seems more appropriate. It was
only written using `roundf` before as a workaround for inaccuracies
coming out of `clutter_actor_get_transformed_size` that would have
tricked `ceilf` into landing on the wrong integer. But that's since
been fixed by 67cc60cbda so we can use `ceilf` now.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1532
Otherwise we might run into a use-after-free and crash on (virtual)
device removal:
Invalid read of size 8
at clutter_input_device_get_device_type (clutter-input-device.c:811)
by update_last_device (meta-backend.c:1282)
by g_main_dispatch (gmain.c:3325)
by g_main_context_dispatch (gmain.c:4016)
by g_main_context_iterate.constprop.0 (gmain.c:4092)
by g_main_loop_run (gmain.c:4290)
by meta_run_main_loop (main.c:708)
by meta_run (main.c:723)
by main (main.c:550)
Address is 32 bytes inside a block of size 504 free'd
at free (vg_replace_malloc.c:538)
by g_type_free_instance (gtype.c:1939)
by clutter_event_free (clutter-event.c:1420)
by _clutter_stage_process_queued_events (clutter-stage.c:830)
by handle_frame_clock_before_frame (clutter-stage-view.c:1064)
by clutter_frame_clock_dispatch (clutter-frame-clock.c:405)
by frame_clock_source_dispatch (clutter-frame-clock.c:456)
by g_main_dispatch (gmain.c:3325)
by g_main_context_dispatch (gmain.c:4016)
by g_main_context_iterate.constprop.0 (gmain.c:4092)
by g_main_loop_run (gmain.c:4290)
by meta_run_main_loop (main.c:708)
by meta_run (main.c:723)
Block was alloc'd at
at malloc (vg_replace_malloc.c:307)
by g_malloc (gmem.c:106)
by g_slice_alloc (gslice.c:1025)
by g_slice_alloc0 (gslice.c:1051)
by g_type_create_instance (gtype.c:1839)
by g_object_new_internal (gobject.c:1939)
by g_object_new_valist (gobject.c:2264)
by g_object_new (gobject.c:1782)
by meta_input_device_native_new_virtual (meta-input-device-native.c:1365)
by meta_virtual_input_device_native_constructed (meta-virtual-input-device-native.c:705)
by g_object_new_internal (gobject.c:1979)
by g_object_new_valist (gobject.c:2264)
Suggested-by: Carlos Garnacho <carlosg@gnome.org>
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1529
The redraw_pending boolean is used to schedule the first stage update
after starting Clutter. This flag is superfluous because we have the
pending_finish_queue_redraws flag which does the same.
While at it, also remove the redraw_count debug variable, since there
should be better ways to count the number of queued redraws nowadays,
for example Sysprof marks.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1527
Queueing a redraw with a clip is easy enough and this function is
private anyway, so remove it and call _clutter_actor_queue_redraw_full()
instead in the one function using it.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1527
ClutterRedrawFlags are a way to give Clutter additional context
about what it needs to redraw. There currently is only one flag defined,
CLUTTER_REDRAW_CLIPPED_TO_ALLOCATION, this flag would clip the redraw to
the actors current allocation.
Since ClutterActor also provides the clip_to_allocation property (which
affects the paint volume of the actor instead of only one redraw), the
additional CLIPPED_TO_ALLOCATION flag seems unnecessary. It's also only
defined to be used privately in Clutter, which it never is, so let's
remove it.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1527
Turns out the g_signal_has_handler_pending() call in
update_stage_views() is actually more expensive than comparing the
sorted list (which is usually very short), so remove that and simply
always emit the signal.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1524
This commit introduces a few important changes in order to
acommodate graphene_ray_t. Most of them are positive changes,
so don't panic :)
The first very visible change is that neither the actor box
nor the clip rectangles are projected before being pushed.
This required changing the parameters of the related functions
at both ClutterPickContext, and ClutterPickStack, to receive
boxes instead of vertices. These rectangles are projected on
demand now, so in the best case (first actor picked) only
one projection happens; and in the worst case, it projects
as much as it does now.
The second important change is that there are no more checks
for axis-alignment anymore. That's because picking now happens
in 3D space, using triangles.
Talking about triangles in 3D space, this is what is used now
for picking. We break down each actor rectangle in 2 triangles,
and check if the projected pick point is inside any one of them,
of if the ray intersects any one of them. The same check happens
for the clip rectangles.
Checking the projected pick point is both an optimization for the
2D case, and a workaround to graphene_ray_t problems with float
precision, which is specially visible on edges such as the top
bar.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1509
ClutterPickStack is a new boxed type that stores the vertices
and clip rectangles. It is meant to be a byproduct of picking,
and takes over most of what ClutterStage currently does.
It introduces a 'seal' system, inspired by MetaKmsUpdate. After
the pick operation is done, and the rectangles are collected,
the pick stack is sealed, and is not allowed to be externally
modified anymore. Internally, it still can invalidate pick
records when an actor is destroyed.
For now, it handles both the clip rectangles, and the matrix
stack, separatedly. Future commits will rearrange this.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1509
Because clones may not have identical geometry to their source actors.
So we can't use the geometry of the source actor to decide to take the
more optimized (more clipped) path.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1480
The "paint" signal of ClutterActor has been a pain for everyone involved
long enough now, turns out we actually use it nowhere except tests
anymore (which has been handled in the last commits), so get rid of it
for good before anyone starts using it again.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1522
The "paint" signal of ClutterActor is deprecated and will be removed. We
have a good replacement to get notified about stage paints nowadays,
that is "after-paint" on ClutterStage, so switch to that signal where it
makes sense.
I didn't bother to update the few tests (namely Clutters
conform/texture-fbo.c, conform/text-cache.c,
interactive/test-cogl-multitexture.c and Cogls
conform/test-multitexture.c, conform/test-texture-mipmaps.c) where it's
harder to replace the signal since we don't build those anyway.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1522
The paint-wrapper test wraps around the painting process of an actor to
paint its own texture before and after painting, it does that using the
"paint" signal.
This signal is deprecated and will be removed from Clutter, and since
this "use-case" won't be supported anymore afterwards (the proper way is
to use a ClutterEffect for things like this), remove the test.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1522
Rely on the seat stage, or other ways to fetch it. Also rely that
there is actually a single stage, so that we assign the right stage
to all events going out of the seat, in a single place.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1486
This is a bit scattered around, with the setter/getter in Clutter, and
it only being only directly honored in Wayland (it goes straight through
device properties in X11).
Make this private native API, and out of public ClutterInputDevice API.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1486
Make the upper part agnostic about the device being relative in order
to avoid applying keep-aspect. The X11 bits already are, so make it
sure it's also the case for the native backend.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1486
The semantics for libinput events are not as expected here. Besides
it's pointless, as those should arrive per-slot in a burst, and we
cancel on the first event.
We can simply use the Clutter event for this.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1486
As it does seem from a read to libinput code, TOUCH_CANCEL events
actually do contain slot information, and are emitted per-slot.
This means we can avoid iterating over the slots ourselves, they
are still expected to be sent altogether.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1486
We want to coalesce multiple touch events into the same wl_touch.frame
event. Instead of poking internals to peek the touch events (and their
slots) coming at us before we handle them, simplify things by queueing
the event at a slightly lower priority than events, so we are ensured
to handle all pending input events before sending the event.
If there's no pending events, we can just send the frame event. As it
doesn't make sense to hold any longer.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1486
When malformed modes are provided and no valid mode spec is found, mutter
will eventually try to access the last element of an empty list. Warn about
this and exit properly.
https://gitlab.gnome.org/GNOME/mutter/-/issues/1481
We will use a dedicated variable when transitioning to/from fullscreen state
and leave the previously used 'saved_rect' exclusively for transitioning
between floating and maximized state.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/801
Implements the "prior window window geometry dimensions" as described in
the documentation of 'xdg_toplevel' request 'unset_maximized':
"If available and applicable, the compositor will include the window
geometry dimensions the window had prior to being maximized in the
configure event."
and 'unset_fullscreen':
"The compositor may include the previous window geometry dimensions in
the configure event, if applicable."
Fixes https://gitlab.gnome.org/GNOME/mutter/issues/792.
If this call is available, we can turn libX11 IO errors (fatal by definition)
into something we can recover from. Try to dispose all X11 resources and close
the display instead, so the compositor can survive the event.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1447
If the MetaX11Display abruptly closed when X11 windows were present,
we would still try to deal with them while freezing/thawing for the
"fade out" animation.
At the bottom of that, the X server may be gone, just try to cope
with it.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1447
Since commit eb9cd3857d we initialize the allocation of ClutterActors to
an UNINITIALIZED ClutterActorBox. We do that to ensure the actor even
emits notify::allocation in case it got a new valid allocation of
0,0,0,0.
Now there's still the case where an actor gets removed from the
scenegraph and added again to a different parent, in this case we still
don't emit notify::allocation right now in case the new allocation
equals the old one. There's two good reasons to do so though:
1) To Clutter, there's no difference between a newly created actor and
an actor which got removed from the scenegraph, it's not consistent to
always notify the allocation property in the former situation, but not
always notify it in the latter situation.
2) When an allocation changes, Clutter notifies the subtree of that
actor about an absolute geometry change (see the call to
transform_changed() in clutter_actor_set_allocation_internal()). Now
when an actor gets reparented, obviously the absolute geometry might
change, so to make sure transform_changed() is always called in that
case we need to make sure an allocation change happens.
So simply reset the allocation property of the actor to an UNINITIALIZED
ClutterActorBox as soon as it gets unrealized.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1498
We introduced the absolute_origin_changed flag when preparing for the
removal of ClutterAllocationFlags in commit dc8e5c7f8b. Turns out in the
mean-time commit df4eeff6f2 happened, which renders the whole
absolute_origin_changed flag moot.
That's because we now notify the whole subtree about the absolute origin
change by calling transform_changed() when the allocation of an actor
changes. transform_changed() traverses the subtree and calls
absolute_geometry_changed() on every actor immediately, which renders
the whole propagation of the absolute_origin_changed flag obsolete.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1498
Since we now moved the queuing of relayouts into the mapping and
unmapping functions, we no longer need to do it when adding or removing
a child, that's because removing a child always unmaps the child, and
adding it to a stage (if it's visible) will map it.
So remove those calls to queue_relayout() since they're no longer
needed.
With the above we no longer queue a relayout in
clutter_actor_add_child_internal(), that means there's one place where
we need to explicitely queue relayouts now: That's when using the
set_child_at_index/above/below() APIs, those are special because they
avoid unmapping and mapping of actors and would now no longer get a
relayout.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1366
In theory there's no big difference between only handling mapped actors
vs only handling visible actors in clutter_actor_allocate(): The
function is called recursively starting with an actor that is attached
to a stage, so it should only be called on mapped actors anyway.
The behavior of skipping hidden actors was introduced as an optimization
with commit 0eab73dc. Since the last commit, we handle
enable_paint_unmapped a bit better and don't do unnecessary work when
mapping or unmapping, so we can now be a bit stricter enforcing our
invariants and only allow mapped actors in clutter_actor_allocate().
We need to exclude toplevel actors from this check since the stage has a
very different mapped state than normal actors, depending on the
mappedness of the x11 window. Also we need to make an exception for
clones (of course...): Those need their source actor to have an
allocation, which means they might try to force-allocate it, and in that
case we shouldn't bail out of clutter_actor_allocate().
Also moving the clutter_actor_queue_relayout() call from
clutter_actor_real_show() to clutter_actor_real_map() seems to fix a bug
where we don't queue redraws/relayouts on children when a parent gets
shown.
Fixes https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/2973https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1366
We currently support only one case where an actor can get mapped or
unmapped during painting, that is using
_clutter_actor_enable_paint_unmapped() (although we could arguably do a
better job explicitely forbidding it in other cases). This function is
called when painting ClutterClone or MetaWindowActors during
screensharing. It temporarily (fake) realizes and maps the actor and all
its children so it can get painted.
Now a problem will appear when we'll start coupling layout and the
mapped state of actors more closely with the next commit: Since
enable_paint_unmapped() is meant to be enabled and disabled during every
clone paint, we also notify the "mapped" property twice on every clone
paint. That means with the next commit we would queue a relayout for the
source actor on every clone paint.
To avoid this unnecessary work, check whether we're being painted while
unmapped using the new unmapped_paint_branch_counter. Then avoid queuing
relayouts or invalidating paint volumes in that case.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1366
We're going to enforce some invariants a bit stricter and will only
allow allocating if an actor is mapped, not only visible.
Since actors are only mapped if their parent is mapped and stages are
hidden by default, we need to show the stage to ensure the actors are
mapped before we allocate them. So do that and call clutter_actor_show()
on the stage before fake-allocating the test actors.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1366
Add new private API to ClutterActor, returning TRUE in case the actor is
being painted while unmapped. This is useful for implementations of the
paint() vfunc or for signal handlers of the "notify::mapped" signal.
Use this API in CallyActor to properly detect "notify::mapped" emissions
while painting unmapped, this fixes detecting the case where
painting-unmapped is used for screencasting.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1366
Just like the existing in_cloned_branch counter, add a property which
tracks whether the actor is part of a subtree that's being painted while
unmapped. This is going to be useful for a few things, for example
changing the clutter_actor_is_in_clone_paint() API to use
enable_paint_unmapped instead of in_clone_paint.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1366
clutter_actor_queue_relayout() detects whether a parent has the
NO_LAYOUT flag set by itself and then queues a shallow relayout for us.
There's no need to duplicate that logic when showing actors, so simply
call clutter_actor_queue_relayout() and let that handle it.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1366
Apparently clutter_paint_volume_union() has problems building the union
of two paint volumes in eye coordinates, that's probably because of the
negative coordinates that come into play there.
Circumvent that by making even more use of Graphene and letting it take
care of computing the union. We do that by creating two graphene_box_t's
from the axis-aligned paint volumes and intersecting those boxes, then
setting our vertices to the new min and max points of the resulting box.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1507
Bug 448183 fixed an issue with _NET_WM_MOVERESIZE_WINDOW not moving a
window by basing the resize on the current (new) rectangle instead of
the original rectangle.
While this fixes the issue with _NET_WM_MOVERESIZE_WINDOW, this also
causes windows with a size increment to move when the resize also
implies a move, such windows might drift while resizing.
Make sure to use the current rectangle for non-interactive resizes only.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/543
The XSizeHints set by X11 clients give a hint to the window manager
about size increment, aspect ratio, base, minimum and maximum size, etc.
When an X11 client changes those values, there is a good chance that it
will affect the actual window size in some way, and mutter rightfully
queue a window resize in that case.
However, mutter does not check if any of the hints have actually changed
and unconditionally queue a window resize whenever a client changes its
WM_NORMAL_HINTS property.
That can be a problem when a zealous client such as xterm decides to
update its WM_NORMAL_HINTS property on resize, because in return mutter
will queue a non-user driven resize in the middle of user-driven events,
hence defeating the purpose of the META_MOVE_RESIZE_USER_ACTION flag.
To avoid that issue, make mutter a bit smarter and avoid queuing a
window resize if the XSizeHints haven't actually changed.
https://gitlab.gnome.org/GNOME/mutter/-/issues/543
On interactive resize, mutter calculates the difference in size based on
the pointer location and relies on window constraints to ensure the
minimum size is honored.
Wayland however does asynchronous window configuration, meaning that not
checking for size hints early enough may lead to the window moving as
the locations was initially computed on a size which will be invalidate
by the client eventually.
Make sure to respect the client size hint on update_resize() so that we
don't end up with a window moving unexpectedly when the client
eventually acked the configuration.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1495
Many tablets have a native portrait mode panel, yet come with a keyboard dock,
where the device gets docked in landscape mode. To avoid the display being
on its side when mutter starts while the tablet is docked, we need to take
the accelerometer reported orientation into account even if there is a
tablet-mode-switch which indicates that the device is NOT in tablet-mode
(because it is docked).
Add special handling for the first time the "orientation-changed"
signal gets signalled by the orientation-manager, which happens after it
has successfully claimed the accelerometer with iio-sensor-proxy.
The added special handling of the initial "orientation-changed" signal
does a number of checks:
1. panel_orientation_managed is false because of the tablet-mode-switch and not
because of other reasons.
2. The device has a native portrait mode panel (and thus likely needs rotation
to display the image the right way up when docked).
If all these checks succeed then it continues with creating a new
monitors-config based on the orientation ignoring the panel_orientation_managed
value (for the initial/first "orientation-changed" signal only).
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1506
The orientation reported by the orientation_manager may have changed while
panel_orientation_managed was false. So when panel_orientation_managed
changes to true we should re-check the orientation.
This fixes the orientation not being correct when e.g. taking a 360 degree
hinges 2-in-1 in clamshell mode (so landscape orientation) and then folding
it into tablet mode while holding it in portrait orientation.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1506
Add support for the (mostly theoretical) case of an input-device
offering tablet-mode-switch functionality being unplugged.
This makes the has_tablet_switch handling identical to the has_touchscreen
handling, leading to more consistent code.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1506
Detect if a tablet-mode-switch device is already present when mutter
starts by checking for this from meta_seat_native_constructed. This
mirrors how we also set has_touchscreen from meta_seat_native_constructed.
This fixes tablet-mode-switches only being recognized when they are added
at runtime.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1506
Unconditionally setting has_touchscreen to check_touch_mode
when a new device gets added leads to has_touchscreen becoming
false when during runtime e.g. an USB keyboard gets plugged in.
Fix this by setting has_touchscreen to TRUE when check_touch_mode
is TRUE and leaving it alone otherwise.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1506
Add a helper function to determine if a seat has a (physical)
touchscreen associated with it.
Currently src/backends/meta-backend.c has a private version of this
(check_has_physical_touchscreen) and further patches in this patch-set
need the same functionality. So add a generic helper for this to
avoid code duplication.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1506
_cogl_util_get_eye_planes_for_screen_poly() is quite a complicated beast. Ever
since Clutter became a compositor toolkit, and specially after we switched to
graphene_frustum_t on paint volumes, we can brutally simplify this function.
The new code assumes camera is at (0, 0, 0) at world coordinates (i.e. before
applying the projection). We also consider that the redraw clip are at stage
coordinates. That means that converting the clip rectangle to world rectangle
is simply a matter of projecting the corresponding vertices using the "view"
matrix. Furthermore, we only need to project the top-left, and bottom-right
vertices, since top-right and bottom-left can be derived from those two.
The frustum setup still uses triplets of vertices to setup the planes, except
now the first vertex is always the camera (hardcoded to 0, 0, 0), and the other
two vertices are the projected clip rectangle vertices.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
The redraw clip region may contain multiple clip rectangles. We currently
only use the extents of this region, but having multiple frusta for each
rectangle is a better alternative, and will allow us to remove the extra
projection we currently do.
Make the clip frustum an array, with multiple frusta.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
The clip planes / frustum are contextual to painting. In the past, for
the lack of a better place, it was added to ClutterStage, but now we
have an appropriate home for such data: ClutterPaintContext.
Move the frustum to the paint context.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
While refactoring the clipping planes / frustum code, it became more and
more evident that we do not need to update them while picking. Picking
nowadays goes through a completely different code path, that does not
rely on paint volume culling.
While it might be interesting to eventually also cull out based on paint
volumes, it certainly won't go through the painting code anymore.
Remove setting up the view when picking, and rename functions appropriatedly.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
Culling paint volumes don't give this level of detail anymore, and in
fact knowing whether it was partially or fully in was only being used
in a debug path. For the purposes of culling, it doesn't matter if a
given actor is partially or completely inside the frustum; either way,
it must be painted.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
Instead of 4 planes, use a graphene_frustum_t to store the clipping
planes.
The cautious reviewer might noticed that we are now setting up 6
planes: the 4 planes we were doing before, plus 2 extra planes in
the Z axis. These extra planes simulate an "infinite" Z far, and
an "on-camera" Z near.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
ClutterStage defines the 8 vertices of a frustum:
4 ----------------------------- 5
| \ / |
| \ / |
| 0 --------------------- 1 |
| | | |
| | | |
| 3 --------------------- 2 |
| / \ |
| / \ |
7 ----------------------------- 6
Then, it uses triplets of vertices to create each clipping plane.
It only sets up 4 planes (it doesn't clip based on depth), defined
by the following vertices:
* 0 - 4 - 5
* 1 - 5 - 6
* 2 - 6 - 7
* 0 - 7 - 4
The first 3 triplets are selected using the for-loop. However, the
last triplet is different, and is done out of the loop. It could
have been made simpler by using the "3 - 7 - 4" triplet.
Simplify the current code by using the suggested triplet, calculated
inside the for-loop.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
Instead of our own implementation that upscales, then downscales back,
use graphene_matrix_inverse() directly. This is possible after switching
to a z-near value that doesn't have problems with float precision.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
Picking is specially sensitive for float precision, and tests can
easily fail when something changes, even if ever so slightly. A
simple way to workaround this is by adjusting the projected points
using the same procedure described at 67cc60cbda.
Round projected points for picking to 256ths.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1489
A first step towards abandoning the CoglObject type system: convert
CoglFramebuffer, CoglOffscreen and CoglOnscreen into GObjects.
CoglFramebuffer is turned into an abstract GObject, while the two others
are currently final. The "winsys" and "platform" are still sprinkled
'void *' in the the non-abstract type instances however.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
The first argument is the framebuffer operated on, so in order to stay
consistest, rename 'src' to 'framebuffer'. The second is the
destination. The destination is commonly referred to as 'dst' elsewhere,
so rename 'dest' to 'dst'.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1496
The spec states:
```
A sub-surface becomes mapped, when a non-NULL wl_buffer is applied
and the parent surface is mapped. The order of which one happens
first is irrelevant. A sub-surface is hidden if the parent becomes hidden,
or if a NULL wl_buffer is applied. These rules apply recursively
through the tree of surfaces.
```
In the past we relied on Clutter actor behaviour to realize the recursive
part - which then broke in
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/664
when we changed the actor hierachy in regards to subsurfaces.
Explicitly encode the desired behaviour in `MetaWaylandSubsurface`, fixing
the issue and making it future proof.
Closes https://gitlab.gnome.org/GNOME/mutter/-/issues/1384
Aligning windows manually with other windows has become less important
since the advent of tiling. This decreases the usefulness of edge
resistance, which in fact many users perceive as lag nowadays.
Account for that by limiting resistance to screen and monitor edges by
default, and only include windows when the control key is pressed.
https://bugzilla.gnome.org/show_bug.cgi?id=679609
Commit 033f0d11bf added a fallback in case the tile monitor wasn't
set before, but didn't actually check for a previously set value.
As a result, the "fallback" is not set unconditionally, which may
differ from the expected monitor: The tile monitor is determined
by the pointer position, while the window's monitor is the one
where the biggest part of the window resides on.
https://gitlab.gnome.org/GNOME/mutter/-/issues/1389
It's pointless to call into functions that produce information that will
end up nowhere, so lets not. This will generate less angst when doing
more intense data gathering and string generation in debug log calls.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1467
Fedora's mutter is built with --auto-features=enabled which enables
-initfd support. But the current meson code doesn't check whether
-initfd is available in Xwayland, leading to the wayland session not
starting up.
Check for -initfd's availability before enabling it.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1103
The timestamp sent with _NET_WM_FRAME_DRAWN should be in "high
resolution X server timestamps", meaning they should have the same scope
as the built in X11 32 bit unsigned integer timestamps, i.e. overflow at
the same time.
This was not done correctly when mutter had determined the X server used
the monotonic clock, where it'd just forward the monotonic clock,
confusing any client using _NET_WM_FRAME_DRAWN and friends.
Fix this by 1) splitting the timestamp conversiot into an X11 case and a
display server case, where the display server case simply clamps the
monotonic clock, as it is assumed Xwayland is always usign the monotonic
clock, and 2) if we're a X11 compositing manager, if the X server is
using the monotonic clock, apply the same semantics as the display
server case and always just clamp, or if not, calculate the offset every
10 seconds, and offset the monotonic clock timestamp with the calculated
X server timestamp offset.
This fixes an issue that would occur if mutter (or rather GNOME Shell)
would have been started before a X11 timestamp overflow, after the
overflow happened. In this case, GTK3 clients would get unclamped
timestamps, and get very confused, resulting in frames queued several
weeks into the future.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1494
Mutter sends a proximity-in event before the required tablet tool
resource is properly allocated on the client. This is violating the
Wayland protocol. Because libwayland ignores events for objects it
doesn't know yet, this is not noticeable in most applications. However,
if https://gitlab.freedesktop.org/wayland/wayland/-/issues/176 gets
fixed, these applications would likely crash immediately. Therefore this
PR removes the responsible code which, again, shouldn't have any effect
on client applications as they ignore this event anyway.
Relevant part of the spec:
This event can be received when the tool has moved from one surface to
another, or when the tool has come back into proximity above the
surface.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1427
When we resize a window we send it configure requests with size
suggestion. Some clients, e.g. gnome-terminal will limit its size to a
discrete set given the font size resulting in the size often not being
respected completely, but used as a hint to find a size as large as
possible but not larger than the configured size.
When doing an interactive resize dragging the right or top side of a
window, this caused issues with the configured window size not matching
the one used by the client, as the configured position wouldn't be
correct for the actual size. Fix this by offsetting the position given
the size mismatch offset, making the position again in sync with the
size.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1447https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1477
Keeping track of the projected position is costly, and adds quite some complexity
to ClutterOffscreenEffect.pre_paint(). As far as research goes, there's not a
single consumer of this function that uses the position for anything - only size
is used.
Remove clutter_offscreen_effect_get_target_rect(), and drop the annoying position
field from ClutterOffscreenEffect as well. This allows us to stop projecting the
position on pre-paint, and simplify things.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1474
ClutterPipelineNode will be used by GNOME Shell in the future.
Fortunately for us, CoglPipeline is already usable from GJS,
so we don't need to skip the constructor for the pipeline node.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1474
Move unreffing the framebuffer to ClutterOffscreenEffect.pre_paint().
This will allow us to properly chain up ClutterOffscreenEffect.paint()
and not reimplement exactly what ClutterEffect does by default.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1474
Mutter still relies heavily on singletons such as its MetaBackend.
For that, the backend implementation has a meta_init_backend() function
which is called at startup from meta_init(), which creates the desired
backend and sets the singleton which is returned by meta_get_backend().
Unfortunately, that means that the backend is never actually freed, and
all the code from the backend finalize function never actually get
called.
Add a meta_release_backend() to free the backend singleton.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1438
The input settings constructor installs callback functions on device
added/remove and tool-changed.
However, on dispose, those signals are not disconnected, meaning that on
teardown, once the devices get removed eventually, the callback will
still fire and call the callback with freed data, causing a crash.
Make sure we clear the signals on devices on dispose, to avoid the crash
on teardown.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1438
A boring one, with the exception that row and column needed to be
swapped. For the sake of consistency, the variable names were also
synchronized with the values they hold, so e.g. xy → yx, etc.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
This one is a bit tricky. The tl;dr; is that switching from right-hand
multiplication to left-hand multiplication required applying the stack
from left to root. This actually allowed simplifying the code a bit,
since CoglMatrixEntry only stores a pointer to its parent, and that's
all we need to know for left-hand multiplication.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
In this case, since we are building the entire matrix by ourselves,
reverse the order of operations (translate + scale → scale + translate)
and build it using graphene-specific APIs.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
This is another instance of graphene reversing the order of operations (see
the commit notes of how ClutterActor was ported.) The tl;dr; here is that,
in the CoglMatrix past, we used to do:
(actor transforms) → scale
and now, it's the other way round:
scale → (actor transforms)
due to changing from right-handed multiplications (CoglMatrix) to left-handed
ones (graphene_matrix_t).
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
ClutterActor is a particularly heavy user of matrices, and
switching to graphene_matrix_* APIs means we had to change
the order of operations due to left-hand vs right-hand
differences.
When applying the actor transform, there are 2 main branches
that can be followed: the default transforms, and when a
custom transform is set.
To facilitate review, here's the table that I've made to
guide myself:
+--------------- Case 1: Default Transforms --------------+
| CoglMatrix | graphene_matrix_t |
+----------------------------+----------------------------+
| multiply (child transform) | translate (-pivot) |
| translate (allocation)¹ | rotate_x (angle) |
| translate (pivot)¹ | rotate_y (angle) |
| translate (translation)¹ | rotate_z (angle) |
| scale (sx, sy, sz) | scale (sx, sy, sz) |
| rotate_z (angle) | translate (translation)¹ |
| rotate_y (angle) | translate (pivot)¹ |
| rotate_x (angle) | translate (allocation)¹ |
| translate (-pivot) | multiply (child transform) |
+----------------------------+----------------------------+
¹ - these 3 translations are simplified as a single call
to translate(allocation + pivot + translation)
+---------------- Case 2: Custom Transform ---------------+
| CoglMatrix | graphene_matrix_t |
+----------------------------+----------------------------+
| multiply (child transform) | translate (-pivot) |
| translate (allocation)² | multiply (transform) |
| translate (pivot)² | translate (pivot)² |
| multiply (transform) | translate (allocation)² |
| translate (-pivot) | multiply (child transform) |
+----------------------------+----------------------------+
² - likewise, these 2 translations are simplified as a
single call to translate(allocation + pivot)
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
CoglMatrix already is a typedef to graphene_matrix_t. This commit
simply drops the CoglMatrix type, and align parameters. There is
no functional change here, it's simply a find-and-replace commit.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
Ideally, we would use Graphene to do that, however as of now Graphene
lacks these APIs so we still need these helpers. Since we're preparing
to get rid of CoglMatrix, move them to a separate file, and rename them
with the 'cogl_graphene' prefix.
Since I'm already touching the world with this change, I'm also renaming
cogl_matrix_transform_point() to cogl_graphene_matrix_project_point(),
as per XXX comment, to make it consistent with the transform/projection
semantics in place.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
Given that CoglMatrix is simply a typedef to graphene_matrix_t, we can
remove all the GType machinery and reuse Graphene's.
Also remove the clutter-cogl helper, and cogl_matrix_to_graphene_matrix()
which is now unused.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
After the previous commit, the only field in the CoglMatrix structure is
a graphene_matrix_t. That means that CoglMatrix is effectively a graphene
matrix now, and the CoglMatrix struct isn't that much useful anymore.
Remove the CoglMatrix structure and make the CoglMatrix type a typedef to
graphene_matrix_t.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
Remove the cached inverse, and dirty flags, and typedef CoglMatrix to
graphene_matrix_t itself. I preverved the type for this commit to help
reducing the commit size, next commits will remove the CoglMatrix type.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
It turns it to be quite easy to inverse the transform, and doing that
on ClutterActor level means we can actually think about removing
CoglMatrix entirely and using graphene_matrix_t everywhere.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
CoglMatrix doesn't have a 1:1 mapping of graphene functions, and
sometimes it's just not worth adding wrappers over it. It is easier
to expose the internal graphene_matrix_t and let callers use it
directly.
Add new cogl_matrix_get_graphene_matrix() helper function, and
simplify Clutter's matrix progress function.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
Instead of listing all matrix cells as floats, and the inverse
as a 16-length float array, use graphene_matrix_t in the structure
itself.
With this commit, all from/to CoglMatrix conversions are gone. It
is also not possible to initialize a CoglMatrix using the macro
anymore.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
Rename cogl_matrix_get_array() to cogl_matrix_to_float(), and
make it copy the floats to an out argument instead of returning
a pointer to the casted CoglMatrix struct.
The naming change is specifically made to match graphene's,
and ease the transition.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
Internally, a graphene_matrix_t representing the same transform that
of a CoglMatrix is the same matrix but transposed, so in order to get
the same element given a column and row for a matrix as if it would
be located in Cogl, it is necessary to swap the row and column when
retrieving it from the graphene matrix.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
Turns out inverting a matrix was the largest chunk of the CoglMatrix
code. By switching to Graphene, a lot of it can go away. The inverse
is still cached in the CoglMatrix struct itself, to preserve the
optimization.
However, switching to graphene_matrix_t to calculate the inverse has
a challenge: float precision. We had to work around it here, and it
needs an explanation.
The way to detect whether a matrix is invertible or not (i.e.
whether it's not a "singular" matrix, or not) is by checking
if the determinant equals 0. So far, so good.
Both graphene_matrix_t and CoglMatrix use single-precision
floats to store their 4x4 matrices. Graphene uses vectorized
operations to optimize determinant calculation, while Cogl
tries to keep track of the matrix type and has special-purpose
determinant functions for different matrix types (the most
common one being a 3D matrix).
Cogl, however, has a fundamentally flawed check for whether
the matrix is invertible or not. Have a look:
```
float det;
…
if (det*det < 1e-25)
return FALSE;
```
Notice that 1e-25 is *way* smaller than FLT_EPSILON. This
check is fundamentally flawed.
"In practice, what does it break?", the reader might ask.
Well, in this case, the answer is opposite of that: Cogl
inverts matrices that should not be invertible. Let's see
an example: the model-view-projection of a 4K monitor. It
looks like this:
```
| +0,002693 +0,000000 +0,000000 +0,000000 |
| +0,000000 -0,002693 +0,000000 +0,000000 |
| +0,000000 +0,000000 +0,002693 +0,000000 |
| -5,169809 +2,908017 -5,036834 +1,000000 |
```
The determinant of this matrix is -0.000000019530306557.
It evidently is smaller than FLT_EPSILON. In this situation,
Cogl would happily calculate the inverse matrix, whereas
Graphene (correctly) bails out and thinks it's a singular
matrix.
This commit works around that by exploiting the maths around
it. The basis of it is:
inverse(scalar * M) = (1/scalar) * M'
which can be extrapolated to:
inverse(M) = scalar * inverse(scalar * M) = M'
In other words, scaling the to-be-inversed matrix, then
scaling the inverse matrix by the same factor, gives us
the desired inverse. In this commit, the scale is calculated
as 1 / (smallest value in the diagonal).
I'm sorry for everyone that has to read through this :(
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
Next commits, and this patchset in general, will make this patchset
obsolete, since it'll only test graphene types against each other.
If at all useful, the Euler test should be moved to graphene.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
Use dot products to simplify calculations. Because the 'w' column of
the matrix is always summed, use 1.f in the 'w' component of the point
vector.
Because CoglMatrix is column-major and graphene_matrix_t is row-major,
it is necessary to transpose the matrix before retrieving the rows.
When we switch CoglMatrix to be row-major, this transposition will
go away.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
This is pretty similar to the other conversions, except we need to
store the matrix flags before operating on it, and update it using
this old value after. That's because cogl_matrix_init_from_array()
marks the matrix as entirely dirty, and we don't want that.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
At this point, we are still only changing CoglMatrix APIs internally, and
it should still produce the same output as before.
To achieve this, using graphens matrix implementation, we need to exploit
some knowledge about conventions used in Cogl and graphene respectively.
In Cogl, transformation matrices are equivalent to those of affine
transformation matrices. The convention used by graphene, however, is to
operate on matrices that are transposed compared to their affine
counterparts.
So for example, let's say we want to multiply the affine matrices A and B,
to get C.
A × B = C
The first step is to convert A and B to graphene matrices. We do this by
importing the floating point array, importing it directly using graphene.
Cogl exports its matrix to a column major floating point array. When we
import this in graphene, being row major, we end up with the same matrix,
only transposed.
Cogl Graphene
A <===> Aᵀ
B <===> Bᵀ
We then multiply these imported matrices in reverse
Bᵀ × Aᵀ
which in turn, due to ABᵀ = BᵀAᵀ, gives us
Bᵀ × Aᵀ = (A × B)ᵀ
Our original goal was to find C, thus we know that
A × B = C
That means we can shuffle things around a bit.
A × B = C
Bᵀ × Aᵀ = (A × B)ᵀ
Bᵀ × Aᵀ = Cᵀ
With the same conversion as done when going from Cogl to graphene, only
the other way around, we still end up effectively transposing the matrix
during the conversion.
Graphene Cogl
Cᵀ <===> C
Thus when converting Cᵀ to Cogl, we in fact end up with C.
(Explanation authored by Jonas Ådahl)
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
Graphene provides skewing as part of graphene_matrix_t API, and it'll
be easier for the transition to just expose similar API surfaces.
Move the matrix skew methods to CoglMatrix.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1439
During seat initialization, we process early libinput events (adding all known
devices) before the seat gets a stage assigned. This causes warnings when trying
to handle the corresponding CLUTTER_DEVICE_ADDED events, as they are sent
stageless.
As it is definitely too soon to have those events sent meaningfully, filter
those events out and instead handle the CLUTTER_DEVICE_ADDED emission for all
known devices after the seat receives an stage. This makes the events guaranteed
to be emitted early in initialization, but not so soon that they can't be
handled yet.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1472
Mutter itself is versioned now, so passing the version information
to the plugin is redunant now: The version is already determined by
linking to a particular API version (gnome-shell) or by installing
to a versioned plugin path (external plugins).
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1473
A Meta.WaylandClient() object has a GSubprocessLauncher object
passed externally. Currently this object is kept while the
WaylandClient object exists, but is is only needed until the call
to spawn is made.
This patch frees that GSubprocessLauncher just after that call,
thus freeing those resources.
Fix https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1462
This reverts the commits 372d73e2757 and 1d2004524728 - the special
case for alpha-less textures could only happen on Wayland, but now
the opaque region is also set in those cases.
This commit saves us some allocations, simplifies the logic a bit and
makes sure culling uses the same opaque region as our painting paths.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1463
Wayland clients using buffers without alpha channel are not expected to
set an opaque region. However, we rely on the opaque region for the fast
painting path in `MetaShapedTexture`.
Thus, make sure to always set an opaque region internally in those cases.
For X11 clients, wo do so already.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1463
Just like we used to before 30809665d8.
Because in some cases `clip_region` is able to shave off an extra pixel
from the edge of the redraw rectangle(s). And not shaving that off was
making the background rendering inconsistent with shaped-texture, causing
occasional off-by-one artefacts. Now both shaped-texture and
background-content agree on the clip region again that doesn't happen.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1443https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1464
This is essentially a revert of
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/326. This commit
had the unintended side effect that the built sources are actually
rebuilt for every individual user of libmutter_dep. With there being more
tests and generated files, the number of targets to build is increasing
squarely.
Not doing this reduces the number of targets from 2044 to 874, thus
saving man hours and CI burnt cycles in the long run. There's the slight
risk of reintroducing the random build breaks, but mutter is essentially
doing as suggested at https://github.com/mesonbuild/meson/issues/1084
(the only difference being addressed in the previous commit), so meson
ought to behave as expected.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1458
We only update the last device from actual input interaction here,
avoid this pair of events. This is specially nasty with
CLUTTER_DEVICE_REMOVED, since the device we're notifying upon will be
disposed soon after emission.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1460
Like it's done for the pointer in other places. Without a stage assigned,
some bits (like IM handling) may end up with events ignored, and misbehave.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1413
The Meta.WaylandClient constructor receives a GSubprocessLauncher
as a parameter, and stores it internally. Unfortunately, its
refcount value isn't increased, which results in the object being
released twice.
This patch fixes this bug.
Fix https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1454
Writing tests' output to a log file makes them difficult to debug when
the test might be running on an autobuilder or CI system where only
stdout/stderr are recorded. This is particularly troublesome if a
failure is only reproducible on a particular autobuilder.
Recent Automake versions have the convention that detailed output from
failing tests is written to stdout/stderr, not just to log files, when
the VERBOSE environment variable is set; borrow that convention as a
trigger for producing detailed test output.
This was originally cogl!14, but applies equally to mutter's fork of cogl.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1273
Signed-off-by: Simon McVittie <smcv@debian.org>
When a MetaBarrier is first created it allocates a backend
impl object which does the actual heavy lifting.
Unfortunately, that backend object is never freed.
This commit ensures the implementation gets freed when
the barrier object is freed.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1451
meta_barrier_destroy is responsible for removing the extra
reference added in meta_barrier_constructed.
Unfortunately, it fails to do this because of a misplaced early
return statement.
This commit removes the spurious return.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1449
This is already taken care of in meta_backend_monitors_changed(), called
from the same code paths that emit ::monitors-changed-internal. It is
better to leave this up to backend internals.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1448
This resolves a couple of FIXMEs. The FIXME comments were right in
stating that not *all* journals needed flushing, only the one we
are trying to put on screen needs flushing.
However we can't eliminate all flushes because the winsys swap calls
that follow go directly into OpenGL which knows nothing about cogl
journalling. So the journal *must* be flushed before the swap, to give
OpenGL the correct state.
P.S. If this turns out to cause any bugs then the next best answer is
to just remove the FIXME comments. Because flushing is still the right
thing to do.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1362
We only did this if we weren't currently doing an interactive resize,
but since the finish_move_resize() is not the actual interactive resize
but the acknowledgment of the configure event that was emitted as a
result, we shouldn't limit ourself to the same flags used during resize.
This fixes temporarly "stuck" position of attached modal dialogs while
they are being resized.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1163https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1446
In X, buttons 1, 2, 3 are left, middle, right. In evdev, the order is
BTN_LEFT, BTN_RIGHT, BTN_MIDDLE. So setting a scroll button to 2 gave us a
middle button in the X session and a right button in a wayland session.
Fix that by hard-coding the LMR buttons handling.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1433
Even when a direct client buffer has a compatible format, stride and
modifier for direct scanout, drmModePageFlip() may still fail sometimes.
From testing, it has been observed that it may seemingly randomly fail
with ENOSPC, where all subsequent attempts later on the same CRTC
failing with EBUSY.
Handle this by falling back to flipping after having composited a full
frame again.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1410
We already correctly set the font-dpi based on user settings in
MetaSettings at each user change and as part of backend initialization,
so there's no point to set it also during x11 backend post-parsing and
using X11 values, as this may happen at later point and lead to a wrong
clutter font DPI value.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1444
If there is no laptop panel (for example on a desktop PC or a virtual
machine), attempting to put a NULL monitor in the list of matches
will just make mapping_helper_apply() crash.
Mitigates: https://gitlab.gnome.org/GNOME/mutter/-/issues/1414
Signed-off-by: Simon McVittie <smcv@debian.org>
It's enabled by default when using the i915 driver, but disabled
everywhere else until it can be made reliably an improvement. Until
then, for anyone want to force-enable it, add the string
'dma-buf-screen-sharing' to the experimental features list in GSettings.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1442
Seems DMA buffer based interprocess buffer sharing is more broken than
not, so for now only enable it when using the i915 driver.
For example vmwgfx, qxl and radeon, it results in mmap() failing to mmap the
memory region. Other drivers, e.g. amdgpu will function, but may hit
very slow memory download paths, resulting in worse performance.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1442
When the CoglRenderer didn't set the DMA buffer constructor vfunc, we
return NULL. What we didn't do was set the error, meaning the caller
would crash if it tried to look up why DMA buffer construction failed.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1430
We only want the panel autorotation to happen if the laptop has an
accelerometer, and is in tablet mode. Regular laptop mode should
lock the orientation, and let it be configured manually.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1311
So far, we've expected this signal to not happen whenever autorotation
shouldn't apply (no accelerometer is a strong reason). In future commits
we'll add further checks to this policy, so prevent autorotation to
change the display configuration if the MetaOrientationManager signal
happens but it should be ignored.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1311
Instead of having everyone check net.hadess.SensorProxy themselves, have
this all controlled by the MetaOrientationManager, and proxied everywhere
else via a readonly property in org.gnome.Mutter.DisplayConfig.
We want to attach more complex policies here, and it seems better to
centralize the handling of the autorotation feature rather than
implementing policy changes all over the place.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1311
We used to pick the "best" output for each builtin/size/edid categories,
and then pick the "best" (in that order) of those for each input device.
This is most often enough, but is prone to wrong results in some corner
cases (eg. 2 outputs with the exact same dimensions).
Change this to a score mechanism that doesn't leave outputs out. The
weights are the same, but the score is accumulated if an output matches
multiple categories. All outputs are evaluated and sorted by score, and
input devices with the best matches are applied first (as they already
did).
This should break the tie if eg. there's 2 outputs with similar dimensions,
but one of them has some EDID match in addition. The output with multiple
matches will score higher up, while it might have been entirely discarded
with the previous implementation.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1175https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1202
The work at https://gitlab.gnome.org/GNOME/gnome-control-center/issues/239
intended to make integrated devices optionally mappable to other outputs
(in order to allow fix mishandling from our heuristics, or to quickly reach
things in other monitor without changing devices).
This was missed in that plan, we do allow cycling outputs, but we still did
prevent it from doing anything for integrated devices. Fix that, and change
output cycling so we don't allow a "NULL" EDID for integrated devices, this
makes those go through the MetaInputMapper (resulting in one output listed
twice), instead of mapping to the full stage.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1186https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1201
The cursor tracker may give us a valid position, and a
valid cursor sprite, and yet the cursor can be hidden,
meaning we must hide the cursor on the stream as well.
Remove cursor from stream buffer if it's hidden.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1421
Scanouts are taken away after painting. However, when we're
streaming, what we actually want is to capture whatever is
going to end up on screen - and that includes the scanout
if there's any.
Add a before-paint watch that only records new frames if a
scanout is set.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1421
It'll allow subclasses to get notified of the before-paint
signal without having to connect to it. This will allow
MetaStage to have proper watches being fired there without
the cost of the signal handling machinery.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1421
When there's a direct scanout set in the stage view, we
have to use it instead of the view's regular onscreen
framebuffer.
Use the new CoglScanout API to implement blitting to the
stream framebuffer.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1421
This will be used when screencasting monitors so that if
there's scanout in place, it'll still be possible to blit
it to a PipeWire-owned framebuffer, and stream it.
Add a new 'blit_to_framebuffer' vfunc to CoglScanout, and
implement it in MetaDrmBufferGbm.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1421
Just because X11/XI uses a particular terminology doesn't mean we
have to use the same terms in our own API. The replacement terms
are in line with gtk@1c856a208, which seems a better precedent
for consistency.
Follow-up to commit 17417a82a5.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1425
Using opaque painting paths can have a big impact on painting performance.
In order to easily validate whether we use the opaque paths, add a opaque
(green) or blended (purple) overlay over painted areas if the
`META_DEBUG_PAINT_OPAQUE_REGION` `MetaDebugPaintFlag` is set.
You can do so in `lg` via:
`Meta.add_debug_paint_flag(Meta.DebugPaintFlag.OPAQUE_REGION)`
This can be helpful for application developers, as previously it was not
trivial to check whether e.g. Wayland or X11 opaque regions where
properly set.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1372
In certain situations it's desirable to keep pipelines around for
the whole lifetime of the session. In order to not leak them and
properly clean them up on shutdown, introduce a new mechanism to
create named pipelines that are bound to their correstponding
context and may be used across file boundries.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1372
Analogous to `ClutterDrawDebugFlag` but intended for concepts that
are not present in Clutter, such as Wayland/X11 opaque regions.
Also add the first flag for the later.
To set the flag, run:
`Meta.add_debug_paint_flag(Meta.DebugPaintFlag.OPAQUE_REGION)`
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1372
This is no longer directly related to DBus messages, but our own event
queue. Change the wording and use CLUTTER_PRIORITY_EVENTS to make it
bolder, even though it's the same than G_PRIORITY_DEFAULT.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1286
We set an idle to coalesce multiple IM events together, as the protocol
requires us to send them in one frame, and unfortunately there is no
idea about whether more IM events are upcoming.
One good hint though are key events generated from the IM, we want to
apply all IM changes before the key event is processed, so make it sure
that the .done event is flushed before the key event is handled.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1286
These are not given directly to the input focus anymore, instead
queued up as events. This way, all actions triggered by the input
method (commit and preedit buffer ones, but also synthesized key
events) queue up the same way, and are thus processed in the exact
same order than they are given to us.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1286
The clutter_input_focus_filter_key_event() function has been made
a more generic filter_event(). Besides its old role about letting
key events go through the IM, it will also process the IM events
that are possibly injected as a result.
Users have been updated to these changes.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1286
When in the overview culling via `self->clip_region` is unavailable.
The region is `NULL` because the paint call has not originated from a
`WindowGroup`, because the overview does not use `WindowGroup`.
So the main wallpaper was being painted in full while in the overview.
That's a waste of effort because `redraw_clip` is going to be used to
stencil/scissor out only the parts that are changing. We don't need to
paint *most* of the wallpaper, only the parts behind anything changing.
For the overview this reduces GPU power usage (intel_gpu_top) roughly
10% and reduces render times almost as much.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1363
`meta_background_content_paint_content` was mixing two different
coordinate systems in `actor_pixel_rect`. It was initialized with
actor-local coordinates and then `if (self->clip_region)` would be
treated as stage coordinates. This worked because `self->clip_region`
was only non-NULL outside of the overview where both coordinate systems
were the same. So it always got the right answer, possibly by accident.
In order to enhance the function however we will need to know which
coordinate system we're working in, so now we make it explicit.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1363
Previously we only culled actors that didn't intersect the bounding box
of the redraw clip. Now we also cull those whose paint volume bounds don't
intersect the arbitrary shape of the redraw clip.
This was inspired by the activities overview where idle windows and
workspace previews were being needlessly repainted. In that particular
case this yields more than 10% reduction in render time. But it probably
helps in other situations too.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1359
There's no need to update the outputs recursively in case the actor gets
mapped or unmapped. That's because mapping happens recursively itself,
so if a window with multiple subsurfaces is shown, all subsurfaces will
receive a "notify::mapped" signal.
Since this was the only remaining user of
meta_wayland_surface_update_outputs_recursively(), we can now remove
that function.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1358
Since we now listen to the "stage-views-on-changed" signal (which
"catches" all the changes we want) on MetaWaylandActorSurfaces for
updating the wl_outputs the surface is on, we no longer need to call
meta_wayland_surface_update_outputs_recursively() on all geometry
changes, so remove that signal handler.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1358
ClutterActors new "stage-views-changed" signal fits pretty well for the
updating of wl_outputs a MetaWaylandActorSurface is on: With that signal
we get notified if the surface moved to a different CRTC, of which every
output has at least one.
So start listening to that signal, which fixes a bug where the wl_output
of a surface changes, but its allocation remains the same (which means
no signals triggering an update of the outputs will be emitted) and no
enter/leave events for the new wl_outputs are sent to the client. This
can happen when a monitor is hotplugged but the new allocation is
exactly the same as the old one even though it's on a different monitor.
Since the "stage-views-on-changed" signal will also get emitted when a
parent actor of the surface is moved, this means we can now remove the
call to meta_wayland_surface_update_outputs_recursively() on window
position changes or the completion of window-effects.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1358
QXL doesn't support mmap():ing a DMA buffer allocated in mutter inside
the PipeWire stream consumer process. To make screen casting work again
on QXL, disable DMA buffer based screen casting for QXL.
Eventually, it should be the client that renegotiates the supported
buffer types, but until then we need this list.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1318
The X server, including Xwayland, can be compiled with different X11
extensions enabled at build time.
When an X11 extension is built in the X server, it's usually also
enabled at run time. Users can chose to disable those extensions at run
time using the X server command line option "-extension".
However, in the case of Xwayland, it is spawned automatically by the
Wayland compositor, and the command line options are not configurable
by users.
Add a new setting to disable a selected set of X extension in Xwayland
at startup, without needing to rebuild Xwayland.
Of course, if Xwayland is not built with a given extension support in
the first place (which is the default for the security extension for
example), that option has no effect.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1405
Delay the addition and removal of devices using ClutterDeviceEvent's so that
they are processed following the libinput event order, and that we don't
have to flush the events on removal.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1371
When a device is removed from the seat the events that this device may have
emitted just before being removed might still be in the stage events queue,
this may lead a to a crash because:
Once the device is removed, we dispose it and the staling event is
kept in queue and sent for processing at next loop.
During event processing we ask the backend to update the last device
with the disposed device
The device is disposed once the events referencing it, are free'd
The actual last device emission happens in an idle, but at this point
the device may have been free'd, and in any case will be still disposed
and so not providing useful informations.
To avoid this, once a device has been added/removed from the seat, we queue
ClutterDeviceEvent events to inform the stack that the device state has
changed, preserving the order with the other actual generated device events.
In this way it can't happen that we emit another event before that the
device has been added or after that it has been removed.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1345
Clutter device events are special events coming from the backend when an
input device is added or removed.
When such events are processed, we should make the seat to handle them by
calling vfunc that can be implemented by each backend and eventually
emitting the appropriate signal.
If a device is removed, we can also safely dispose it, as it can be
considered stale at this point.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1371
Add clutter device added and removed events to allow processing of them as
it happens in the backends, queuing them and performing actions in order.
This allows not to loose any event that is performed just before removing or
disabling a device, and still process the events in order in the event
queue.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1371
When removing a device that has been just marked as the last in use, we may
try to notify that a NULL device is the last one.
This is not supported, as both update_last_device() and the clients of the
"::last-device-changed" signal are assuming that the last device is always
a valid ClutterInputDevice.
So let's avoid erroring, and stop the idle when clearing the current device.
Related to: https://gitlab.gnome.org/GNOME/mutter/-/issues/1345https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1371
A GAppInfo is not guaranteed to have a filename or an application (or
rather a desktop ID). Add a check for application_id to be non-NULL
before trying to call sn_launcher_set_application_id, which would crash
otherwise.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1392
PipeWire reuses buffers, and buffer metadatas, when streaming. When
the cursor is moved to outside the stream, the cursor meta also needs
to be updated, otherwise it'll use the cursor position of whatever is
in the buffer.
Don't bail out when cursor is outside the stream, and ensure to record
a metadata-only frame. This only applies to metadata streams.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1417
If we returned early in one of the checks but already assigned the
surface role, we'd later run into a double-free and crash. Just do
the checks at the beginning.
Also add a missing return statement that was left out in commit
88ff196fe3ca5 and tighten the parent surface check.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1415
The spec allows `parent_resource` to be `NULL`, requiring the parent
surface to get specified by some other protocol. Send a protocol error
with some meaningful explanation instead of crashing.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1408
Currently, the maximum size for a mouse pointer bitmap for screen
casting is 64x64 pixels.
However, this limit is hit way too often as it is way too low and
results in crashes in either gnome-remote-desktop or mutter.
For example: The a11y settings in g-c-c allow setting a larger pointer
bitmap in order to increase the visibility of the mouse pointer.
With the current limit of 64x64 pixels it is not possible to use the
larger variants of the default mouse pointer bitmap, without
experiencing any crash.
Another way to hit the limit is when display scaling is used or some
game uses a custom (large) mouse pointer bitmap.
The VNC backend in gnome-remote-desktop does not seem to have a maximum
pointer bitmap size.
The RDP backend on the other hand has a maximum pointer bitmap size at
384x384.
Use this size (384x384) as maximum size instead of the current 64x64
size for mouse pointer bitmaps to avoid crashes in mutter and
gnome-remote-desktop and to ensure that bigger mouse pointer bitmaps
can be used.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1414
It is linear config manager created when ensuring configuration.
However, the switch config is not set as LINEAR, but left as UNKNOWN.
This leads switch mode OSD always shows "Join Displays" icon, rather
than the next icon which is "External Only" after connect an external
display and press Super+P once at first time since mutter starts.
This patch moves switch config setting into
meta_monitor_config_manager_create_linear() (and the sibling functions)
to well prepare the monitors config and avoid missing settings.
This is a regression introduced by 149e4d69345a.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1362
Clients can set minimum and maximum to identical values to indicate a
fixed-size window. A compositor can ignore these requests and thus a client
has to ensure these limits.
To support clients that do not ensure these size limits by themselves and
to skip unnecessary function calls, we will prevent resizing requests by
the client if a fixed-size has been requested.
https://gitlab.gnome.org/GNOME/mutter/-/issues/1331
This was from the old clutter-as-application-library days, where it had
to try find a suitable backend. Now we already have a backend selected
(MetaBackend), and the clutter backend is already predecided depending
on that, so we don't need the code that auto detects an appropriate one
anymore.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1364
We're moving towards not supporting stand-alone application style
clutter stages, meaning the stage tests use will be reused instead of
recreated. To make this feasable, tests must clean up after themself.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1364
meta_run() is still left intact and does the same as before; the new
functions are only intended to be used by tests, as they may need to set
things up after starting up. Doing so linearly in the test case is much
easier than adding callbacks, so meta_run() is split up to make this
possible.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1364
There is no reason to use Xsettings for the X11 backend, as it comes
from the GSetting store anyway, so move the font setting reading to
ClutterSettings and read directly from GSettings.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1364
The delete event was used for signalling the close button was clicked on
clutter windows. Being a compositor we should never see these, unless
we're running nested. Remove the plumbing of the DELETE event and just
directly call meta_quit() when we see it, if we're running nested.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1364
We checked if we were using the usig the X11 backend to decide when to
deal with a11y event posting - in order to make the clutter code less
windowing system dependent, make this check a check whether we're a
display server or not, in contrast to a window/compositing manager
client. This is made into a vfunc ot ClutterBackendClass, implemented by
MetaClutterBackendNative and MetaClutterBackendX11.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1364
Test that if a timeline got its frame clock from a parent of the
associated actor, if that actor moves across the stage so that the stage
views changes and thus the would be picked frame clock too, this is
noticed by the timeline so that it also changes to the correct frame
clock.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1404
Timelines where the frame clock was picked from a parent of the
associated actor didn't get notified about any stage views changes, as
it only listened on the associated actor. If that actor didn't actually
get its stage views changed (because it went from empty to empty), we'd
end up with a stale frame clock, leading to crashes.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1404
When we pick the frame clock given the associated actor, that frame
clock in fact comes from a picked actor. In order to not end up with
stale frame clocks, which may happen on e.g. hotplugs, monitor layout
changes, or non-optimal frame clocks, which may happen when the parent
used for picking the clock moves to another view, lets listen to
'stage-views-changed' on the actor used for picking the clock too.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1327https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1404
Flip flop resize, which is the result of respecting ConfigureNotify
makes test annoyingly racy, as one cannot do
clutter_actor_set_size (stage, 1024, 768);
wait_for_paint (stage);
g_assert_assert (clutter_actor_get_width (stage) == 1024);
The reason for this is any lingering ConfigureNotify event that might
arrive in an inconvenient time in response to some earlier resize.
In order to not risk breaking any current behavior in the X11 CM case
(running as a compositing window manager), only avoid changing the stage
size in response to ConfigureNotify when running nested.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1404
Let's not expose that outside of mutter quite yet; it's not used in
gnome-shell, and to avoid future breakage if it starts to be used, lets
move it to clutter-mutter.h so only mutter and clutter itself can use
it.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1404
This aims to make sure a view and its resources are destroyed when it
should. Using references might keep certain components (e.g frame clock)
alive for too long.
We currently don't take any long lived references to the stage view
anywhere, so this doesn't matter in practice, but this may change, and
will be used by a to be added test case.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1404
In order to run gnome-shell as part of the CI pipeline, we need
an additional runtime dependency plus python3 modules to mock
required system services and run gnome-shell-perf-tool.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1401
Just as wl_shm, hook up the Wayland DMA-BUF protocol to the 64 bit half
point pixel formats too. This makes it possible for Wayland EGL clients
to use 64 bit pixel EGL configurations.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/804
Now that cogl understands them, hook wl_shm up so they can be used.
This also bumps the wayland-server version dependency to 1.17.90, which
corresponds to the master branch of wayland. The new formats will be
available in 1.18.0.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/804
In order to support the DRM formats DRM_FORMAT_ABGR16161616F and
friends, as well as the wl_shm formats WL_SHM_FORMAT_ABGR16161616F and
friends, cogl needs to have knowledge about said formats too.
We don't have a software implementation of the half point data types
however, so the pack/unpack methods remain unimplemented. We don't need
them for now, so it's not crucial that we add them.
For the GLES2 driver, currently only two formats are supported, and
since we don't currently have pack/unpack implementations, the other
formats will for now remain unsupported, until we have a half float
implementation.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/804
Without doing this, we'd use the same sprite that was last set by
mutter, most likely a leftptr cursor, and fail to update when e.g.
moving the pointer above a text entry and the displayed cursor updated
to a cursor position marker.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1391
The displayed cursor is the one displayed on the screen, e.g. via the
hardware cursor plane, by Xorg, or using the stage overlay.
When screen recording under X11, we don't get a stream of pointer and
cursor updates, as they might be grabbed by some other client. Because
of this, the cursor tracker or cursor renderer are not kept up to date
with positional and cursor state.
To be able to use the stage overlays when recording, we need to be able
to update the overlay without updating the displayed cursor, as we
shouldn't update the X server with cursor state we just retrieved from
it.
Thus, to achieve this, create a separate overlay cursor pointer. When
being a display server, they are always the same, but when using X11,
during screen recording, the overlay one will be polled at a fixed
interval to get a somewhat up to date state.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1391
Always force-track the cursor position (so that the X11 backend can keep
it up to date), and if the cursor wasn't part of the sampled
framebuffer when reading pixels into CPU memory, draw it in an extra
pass using cairo after the fact. The cairo based cursor painting only
happens on the X11 backend, as we otherwise inhibit the hw cursor.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1391
On X11 we won't always receive cursor positions, as some other client
might have grabbed the pointer (e.g. for implementing a popup menu). To
make screen casting show a somewhat correct cursor position, we need to
actively poll the X server about the current cursor position.
We only really want to do this when screen casting or taking a
screenshot, so add an API that forces the cursor tracker to track the
cursor position.
On the native backend this is a no-op as we by default always track the
cursor position anyway.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1391
Only when the cursor isn't handled by the backend is the overlay made
visible. This is intended to be used when painting the stage to an
offscreen using clutter_stage_paint_to_(frame)buffer() in a way where
the cursor is always included.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1391
Detect displays marked as 'non-desktop' by the kernel and skip them when
creating the outputs. Mutter is not able to render images that are shown
properly on those devices anyway.
This avoids lighting up attached VR HMDs and showing the GDM login
screen between the eyes in a VR HMD instead of on the monitor.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1393
Allowing code from inside mutter to create a child process and
delegate on it some of its tasks is something very useful. This can
be done easily with the g_subprocess and g_subprocess_launcher classes
already available in GLib and GObject.
Unfortunately, although the child process can be a graphical program,
currently it is not possible for the inner code to identify the
windows created by the child in a secure manner (this is: being able
to ensure that a malicious program won't be able to trick the inner
code into thinking it is a child process launched by it).
Under X11 this is not a problem because any program has full control
over their windows, but under Wayland it is a different story: a
program can't neither force their window to be kept at the top (like a
docker program does) or at the bottom (like a program for desktop icons
does), nor hide it from the list of windows. This means that it is not
possible for a "classic", non-priviledged program, to fulfill these
tasks, and it can be done only from code inside mutter (like a
gnome-shell extension).
This is a non desirable situation, because an extension runs in the
same main loop than the whole desktop itself, which means that a
complex extension can need to do too much work inside the main loop,
and freeze the whole desktop for too much time. Also, it is important
to note that javascript doesn't have access to fork(), or threads,
which means that, at most, all the parallel computing that can do is
those available in the _async calls in GLib/GObject.
Also, having to create an extension for any priviledged graphical
element is an stopper for a lot of programmers who already know
GTK+ but doesn't know Clutter.
This patch wants to offer a solution to this problem, by offering a
new class that allows to launch a trusted child process from inside
mutter, and make it to use an specific UNIX socket to communicate
with the compositor. It also allows to check whether an specific
MetaWindow was created by one of this trusted child processes or not.
This allows to create extensions that launch a child process, and
when that process creates a window, the extension can confirm in a
secure way that the window really belongs to that process
launched by it, so it can give to that window "superpowers" like
being kept at the bottom of the desktop, not being listed in the
list of windows or shown in the Activities panel... Also, in future
versions, it could easily implement protocol extensions that only
could be used by these trusted child processes.
Several examples of the usefulness of this are that, with it, it
is possible to write programs that implements:
- desktop icons
- a dock
- a top or bottom bar
...
all in a secure manner, avoiding insecure programs to do the same.
In fact, even if the same code is launched manually, it won't have
those privileges, only the specific process launched from inside
mutter.
Since this is only needed under Wayland, it won't work under X11.
Fixes https://gitlab.gnome.org/GNOME/mutter/issues/741
Intended to be used to pass state from screen cast clients down the
line. The first use case will be a boolean whether a screen cast is a
plain recording or not, e.g. letting the Shell decide whether to use a
red dot as the icon, or the generic "sharing" symbol.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1377
The clutter_actor_get_transformed_position returns the position of the
top left point of the actor, with the actor transformations. That means
that if the actor is rotated 180º it'll return the "screen" position top
right.
Using this to calculate if the actor is in the screen is causing
problems when it's transformted.
This patch adds a new function clutter_actor_get_transformed_extents,
that will return the transformed actor bounding rect.
This new function is used on the update_stage_views so the actor will
get updated. this way rotated actors will be updated if they are on the
screen.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1386
GLib will now be linking against sysprof-capture-4.a. To support that,
sysprof had to remove the GLib dependency from sysprof-capture-4 which
had the side-effect of breaking ABi.
This bumps the dependency and includes a fallback to compile just the
libsysprof-capture-4.a using a subproject wrap.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1352
Commit 510cbef15a changed the logic in `handle_update()` for X11 window
actors to return early if the surface is not an X11 surface.
That works fine for plain Xorg, but on Xwayland, the surface is actually
a Wayland surface, therefore the function returns early before updating
the drop shadows of server-side decorations for X11 windows.
Change the test logic to restore drops shadows with Xwayland windows.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1384
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1358
The memory selection source was only providing the "text/plain" or the
"text/plain;charset=utf-8" mimetype, but not "STRING" or "UTF8_STRING",
which some X11 clients, like wine, are looking for. This was breaking
pasting from the clipboard in wine applications.
Fix this by adding those targets when they are missing and the selection
source provides the corresponding mimetypes.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1369
Wine destroys its old selection window immediately before creating a new
selection. This would trigger restoring the clipboard, which would
overwrite the new selection with the old one. The selection window
however can also be destroyed as part of the shutdown process of
applications, such as Chromium for example. In those cases we want the
clipboard to be restored after the selection window has been destroyed.
Solve this by not immediately restoring the clipboard but instead using
a timeout which can be canceled by any new selection owner, such as in
the Wine case.
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1338https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1369
The new "id" properties for the MetaCrtc* and MetaOuput* objects are 64-bit
values, so take care to pass 64-bit values when calling g_object_new.
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1343.
When using its EGLStream-based presentation path with the proprietary NVIDIA
driver, mutter will use a different function to process page flips -
custom_egl_stream_page_flip. If that fails due to an EBUSY error, it will
attempt to retry the flip. However, when retrying, it unconditionally uses the
libdrm-based path. In practice, this causes a segfault when attempting to
access plane_assignments->fb_id, since plane_assignments will be NULL in the
EGLStream case. The issue can be reproduced reliably by VT-switching away from
GNOME and back again while an EGL application is running.
This patch has mutter also use the custom page flip function when retrying the
failed flip.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1375
Instead of blindly hoping that `$INCLUDE` contains the parent directory
of `gsettings-desktop-schemas`.
Because `gsettings-desktop-schemas.pc` says:
```
Cflags: -I/SOME/DIRECTORY/gsettings-desktop-schemas
```
Which means to include the version that Meson has configured you need
to drop the directory prefix and only `#include <gdesktop-enums.h>`.
This fixes a build failure with local installs triggered by 775ec67a44
but it's also the right thing to do™.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1370
During animation or other things that cause multiple frames in a row
being painted, we might skip recording frames if the max framerate is
reached.
Doing so means we might end up skipping the last frame in a series,
ending with the last frame we sent was not the last one, making things
appear to get stuck sometimes.
Handle this by creating a timeout if we ever throttle, and at the time
the timeout callback is triggered, make sure we eventually send an up to
date frame.
This is handle differently depending on the source type. A monitor
source type reports 1x1 pixel damage on each view its monitor overlaps,
while a window source type simply records a frame from the surface
directly, except without recording a timestamp, so that timestamps
always refer to when damage actually happened.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1361
Now that we don't use the record function to early out depending on
implicit state (don't record pixels if only cursor moved for example),
let it simply report an error when it fails, as we should no longer ever
return without pixels if nothing failed.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1361
Both do more or less the same but with different methods - one puts
pixels into a buffer using the CPU, the other puts pixels into a buffer
using the GPU.
However, they are behaving slightly different, which they shouldn't.
Lets first address the misleading disconnect in naming, and later we'll
make them behave more similarly.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1361
That was obviously always the intention, but it didn't work when the
display was scaled. My 3840x2160 monitor with a 3840x2160 texture was
being rendered with LINEAR filtering.
It seems the `force_bilinear` flag was TRUE when it should be FALSE.
Because a texture area that's an integer fraction of the texture
resolution is still a perfect match when that integer is the monitor
scale. We were also getting:
`meta_actor_painting_untransformed (fb, W, H, W, H, NULL, NULL) == FALSE`
when the display was scaled. Because the second W,H was not the real
sampling resolution. So with both of those issues fixed we now get
NEAREST filtering when the texture resolution matches the resolution it's
physically being rendered at.
Note: The background texture actually wasn't equal to the physical monitor
resolution prior to January 2020 (76240e24f7). So it wasn't possible to do
this before then. Since then however, the texture resolution is always
equal to the physical monitor resolution.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1346
Make clutter_actor_allocate_preferred_size() convenient to use from
layout managers by not "automatically" honouring the fixed position of
the actor, but instead allowing to pass a position to allocate the
actor at.
This way we can move the handling of fixed positions to
ClutterFixedLayout, the layout manager which is responsible for
allocating actors using fixed positions.
This also makes clutter_actor_allocate_preferred_size() more similar to
clutter_actor_allocate_available_size().
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1310
It's currently a bit hard to get the fixed position of an actor. It can
be either done by using g_object_get() with the "fixed-x"/"fixed-y"
properties or by calling clutter_actor_get_position().
Calling clutter_actor_get_position() can return the fixed position, but
it might also return the allocated position if the allocation is valid.
The latter is not the best behavior when querying the fixed position
during an allocation, so introduce a new function
clutter_actor_get_fixed_position() which always gets the fixed position
and returns FALSE in case no fixed position is set.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1310
It doesn't take all children - subsurfaces in this case - into
account, thus creating glitches if subsurfaces extend outside
of the toplevel surface.
Further more it doesn't seem to serve any special purpose - it was
added in f7315c9a36462, a pretty big commit, and no discussion was
started about the code in question. So it was likely just overlooked
in the review process.
Closes https://gitlab.gnome.org/GNOME/mutter/-/issues/873
Closes https://gitlab.gnome.org/GNOME/mutter/-/issues/1316
gnome-shell displays workspace previews at one tenth scale. That's a
few binary orders of magnitude so even using a LINEAR filter was
resulting in visible jaggies. Now we apply mipmapping so they appear
smooth.
As an added bonus, the mipmaps used occupy roughly 1% the memory of
the original image (0.1 x 0.1 = 0.01) so they actually fit into GPU/CPU
caches now and rendering performance is improved. There's no need to
traverse the original texture which at 4K resolution occupies 33MB,
only a 331KB mipmap.
In my case this reduces the render time for the overview by ~10%.
Closes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/1416https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1347
In the case of indirect rendering like the first frame to use mutter's
background wallpaper:
Texture_A -> FBO_B (Texture_B) -> FBO_C (screen)
we would be trying to render the contents of both FBO_B and FBO_C in
the same flush, before the contents of Texture_A had made it to FBO_B.
So when FBO_C wants to use mipmaps of Texture_B they didn't exist yet
and appeared all black. And the blackness would remain for subsequent
frames as cogl has now decided the mipmaps of FBO_B are no longer
"dirty" and don't need refreshing:
FBO_B (Texture_B) (mipmaps_dirty==FALSE but black) -> FBO_C (screen)
We must flush FBO_B before referencing Texture_B for use in rendering
FBO_C. This only happens when Texture_A changes (e.g. when the user
changes their background wallpaper) so there's no ongoing performance
penalty from this flush.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1347
With the introduction of the shallow relayout mechanism another small
but severe regression sneaked into our layout machinery: We might
allocate an actor twice during the same allocation cycle, with one
allocation happening using the wrong parent.
This issue happens when reparenting an actor from a NO_LAYOUT parent to
a non-NO_LAYOUT parent, in particular it triggered a bug in gnome-shell
when DND reparents a child from the NO_LAYOUT uiGroup to the overviews
Workspace actor after a drag ended. The reason the issue happens is the
following chain of events:
1. child of a NO_LAYOUT parent queues a relayout, this child is added to
the priv->pending_relayouts list maintained by ClutterStage
2. child is reparented to a different parent which doesn't have the
NO_LAYOUT flag set, another relayout is queued, this time a different
actor is added to the priv->pending_relayouts list
3. the relayout happens and we go through the pending_relayouts list
backwards, that means the correct relayout queued during 2. happens
first, then the old one happens and we simply call
clutter_actor_allocate_preferred_size() on the actor, that allocation
overrides the other, correct one.
So fix that issue by adding a method to ClutterStage which removes
actors from the pending_relayouts list again and call this method as
soon as an actor with a NO_LAYOUT parent is detached from the stage.
With that in place, we can also remove the check whether an actor is
still on stage while looping through pending_relayouts. In case
something else is going wrong and the actor is not on stage,
clutter_actor_allocate() will warn anyway.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1356
When picking which frame clock to use, we traverse up in the actor
hierarchy until a suitable frame clock is found. ClutterTimeline
also listens to the 'stage-views-changed' to make sure it's always
attached to the correct frame clock.
However, there is one special situation where neither of them would
work: when the stage doesn't have a frame clock yet, and the actor
of the timeline is outside any stage view. When that happens, the
returned frame clock is NULL, and 'stage-views-changed' is never
emitted by the actor.
Monitor the stage for stage view changes when the frame clock is
NULL.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
An actor may be placed without being on any current stage view; in this
case, to get the ball rolling, walk up the actor tree to find the first
actor where a frame clock can be picked from.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
The frame clock owner should be able to explicitly destroy (i.e. make
defunct) a frame clock, e.g. when a stage view is destructed. This is so
that other objects can keep reference to its without it being left
around even after stopped being usable.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
Currently there is a point in between hot plug, and when the stage view
list is up to date. The check also tests for this behaviour; would this
ever change, the test should be adapted to deal with this too.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
Replace the default master clock with multiple frame clocks, each
driving its own stage view. As each stage view represents one CRTC, this
means we draw each CRTC with its own designated frame clock,
disconnected from all the others.
For example this means we when using the native backend will never need
to wait for one monitor to vsync before painting another, so e.g. having
a 144 Hz monitor next to a 60 Hz monitor, things including both Wayland
and X11 applications and shell UI will be able to render at the
corresponding monitor refresh rate.
This also changes a warning about missed frames when sending
_NETWM_FRAME_TIMINGS messages to a debug log entry, as it's expected
that we'll start missing frames e.g. when a X11 window (via Xwayland) is
exclusively within a stage view that was not painted, while another one
was, still increasing the global frame clock.
Addititonally, this also requires the X11 window actor to schedule
timeouts for _NET_WM_FRAME_DRAWN/_NET_WM_FRAME_TIMINGS event emitting,
if the actor wasn't on any stage views, as now we'll only get the frame
callbacks on actors when they actually were painted, while in the past,
we'd invoke that vfunc when anything was painted.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/903
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
This also changes the view construction path used by the renderer view
to use the new 'add_view()' function, meaning we have a common entry
point for views into the renderer, which will be useful later on.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
Before we'd create the view in init(), then continue poking at it in
realize(). Move all of the screen stage view initialization to
realize(), as that's when we have all the dependent state available.
This is possible since there is nothing needing it until realizing.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
The repaint callbacks are not tied to repaint, thus a bit misleading.
What the functionality in the pre/post-paint callbacks here cares about
is when actually painting; the non-painting related parts has already
moved out to a *-update signal.
This also renames the related MetaWindowActorClass vfuncs, to align with
naming convention of the signals that it listens to.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
Instead of going via MetaCompositor to know about when we updated
(confusingly named post-paint), use the new stage signal directly.
Note that this doesn't change the time frame callbacks are dispatched;
it's still not tied to actual painting even though it seemed so before
given the function names.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
The clutter "thread" repaint callback are not tied to painting, but
indirectly to updating. What the cursor renderer cares about is when we
actually painted, as this is related to the OpenGL fallback paths.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
We'd emit multiple "presented" signals per frame, one for "sync" and one
for "completion". Only the latter were ever used, and removing the
differentiation eases the avoidance of cogl onscreen framebuffer frame
callback details leaking into clutter.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
The vfunc was not tied to "paint", but was used by MetaWindowActorX11
as part of the "update" mechanisms. In order to make that more clear,
special case it in MetaWindowActorX11 by type checking the surface
actor, handling the case without MetaSurfacActor abstraction.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
The synchronization must happen no matter the painting, as it in itself
might result in reported damage, making the stage actually painted. Thus
move it out of the "pre-paint" handler, to something explicitly not tied
to the painting itself - ClutterStage::before-update.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
Instead of the 'pre-paint' signal on MetaCompositor, rely directly on
the 'before-update' signal on the stage. A reason for this is that the
callback should not only invoked in connection to painting, but updating
in general. Currently the 'pre-paint' signal is emitted no matter
whether there were any painting or not, but that's both misleading and
will go away.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
Right now the stage only had a signal called 'after-paint' which was not
tied to painting but updating. Change this to offer 4 signals, for the 4
different stages:
* before-update - emitted in the beginning before the actual stage
updating
* before-paint - emitted before painting if there will be any stage
painting
* after-paint - emitted after painting if there was any stage painting
* after-update - emitted as a last step of updating, no matter whether
there were any painting or not
Currently there were only one listener, that should only really have
been called if there was any painting, so no changes to listeners are
needed.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
The mutexes was used by ClutterTexture's async upload and to match GDK's
mutexes on X11. GDK's X11 connection does not share anything with
Clutter's, we don't have the Gdk Clutter backend left, and we have
already removed ClutterTexture, so lets remove these mutexes as well.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
A frame clock dispatch doesn't necessarily result in a frame drawn,
meaning we'll end up in the idle state. However, it may be the case that
something still requires another frame, and will in that case have
requested one to be scheduled. In order to not dead lock, try to
reschedule directly if requested after dispatching, if we ended up in
the idle state.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
The frame clock wouldn't be useable yet, but none the less, add API to
get the frame clock best suited for driving the actor. Currently this
translates to the fastest one, but that might change.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
The frame clock is meant to eventually drive the painting of the view,
in contrast to the master frame clock painting every view on the stage.
Right now it's a useless place holder.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
The native backend had a plain counter, and the X11 backend used the
CoglOnscreen of the screen; change it into a plain counter in
ClutterStageCogl. This also moves the global frame count setting to the
frame info constuctor.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
We currently have mutter set a global frame counter on the frame info in
the native backend, but in order to do this from clutter, change the
frame info construction from being implicitly done so when swapping
buffers to having the caller create the frame info and passing that to
the swap buffers call.
While this commit doesn't introduce any other changes than the API, the
intention is later to have the caller be able to pass it's own state
(e.g. the global frame count) along with the frame info.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
We had time unit conversion helpers (e.g. us2ms(), ns2us(), etc) in
multiple places. Clean that up by moving them all to a common file. That
file is clutter-private.h, as it's accessible by both from clutter/ and
src/.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
Currently unused, but it's intention is to use as a initial refresh rate
for a with the stage view associated frame clock. It defaults to 60 Hz
if nothing sets it, but the native backend sets it to the associated
CRTCs current mode's refresh rate.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
Without an associated actor, or explicit frame clock set, in the future
a timeline will not know how to progress, as there will be no singe
frame clock to assume is the main one. Thus, deprecate the construction
of timelines without either an actor or frame clock set.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
The association is inactive, as in it doesn't do anything yet, but it
will later be used to determine what frame clock should be driving the
timeline by looking at what stage view the actor is currently on.
This also adapts sub types (ClutterPropertyTransition) to have
constuctors that takes an actor just as the new ClutterTimeline
constructor.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
This is so something outside of clutter-stage.c (i.e.
clutter-stage-view.c) can eventually do various things
_clutter_stage_do_update() does now while not redrawing the whole stage.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
The MetaLater functionality needs to make sure an update is scheduled so
that it can run its callbacks etc. This used a ClutterTimeline (which is
an object more or less meant to drive animations markers, frames etc)
just to keep the master frame clock running. We're moving away from a
single master clock, so just schedule updates directly instead, with the
newly exposed API.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
Devices are updated (repicked) as part of the stage update phase, as
their stacking, position and transform might have changed since since
the last update.
The redraw clip was used to avoid unnecessary updating of devices, if
the device in question had it's position outside of the redraw clip. If
the device coordinate was outside of the redraw clip, what was
underneith the device couldn't have changed.
What it failed to do, however, was to update devices if a relayout had
happened in the same update, as it checked the state whether a layout
had happened before attempting to do a relayout, effectively delaying
the device updating to the next update.
This commit changes the behavior to always update the device given the
complete redraw clip caused by all possible relayouts of the same update
as the device update happens in.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
We'd check if there was any queued redraw on the stage, but this is
inappropriate for two reasons:
1) A monitor and area screen cast source only cares about damage on a
subset of the stage.
2) The global pending-redraw is going away when paint scheduling will be
more view centric.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
This will allow anyone to finish any queued redraws making their
corresponding damage end up being posted to the stage views. This will
allow units to check whether, so far, any updates are queued on a
particular stage view.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
Add API to add and remove ClutterTimeline objects to the frame clock.
Just as the legacy master clock, having a timeline added to the frame
clock causes the frame clock to continuously reschedule updates until
the timeline is removed.
ClutterTimeline is adapted to be able to be driven by a
ClutterFrameClock. This is done by adding a 'frame-clock' property, and
if set, the timeline will add and remove itself to the frame clock
instead of the master clock.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
The timestamp comes from the GSource, meaning it's a more accurate
representation of when the frame started to be dispatched compared to
getting the current time in any callback.
Currently unused.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
In certain scenarios, the frame clock needs to handle present feedback
long before the assumed presentation time happens. To avoid scheduling
the next frame to soon, avoid scheduling one if we were presented half a
frame interval within the last expected presentation time.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
This adds a current unused, apart from tests, frame clock. It just
reschedules given a refresh rate, based on presentation time feedback.
The aiming for it is to be used with a single frame listener (stage
views) that will notify when a frame is presented. It does not aim to
handle multiple frame listeners, instead, it's assumed that different
frame listeners will use their own frame clocks.
Also add a test that verifies that the basic functionality works.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1285
When a transition is created for the allocation change, it will delay
the new allocation box getting set depending on transition details.
This, however, means that e.g. the 'needs_allocation' flag never gets
cleared if a transition is created, causing other parts of the code to
get confused thinking it didn't pass through a layout step before paint.
Fix this by calling clutter_actor_allocate_internal() with the current
allocation box if a transition was created, so that we'll properly clear
'needs_allocation' flag.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1345
41130b08eb added a fix for culling subsurfaces with geometry scale.
Unfortunately it only did so for the opaque regions, not for clip and
unobscured regions, as the effect was hidden by bug that was only
fixed by 3187fe8ebc7.
Apply the same fix to clip and unobscured regions and use the chance
to move most of the slightly hackish geometry scale related code
into a single place.
We need to scale slightly differently in the two cases, indicated by
the new `ScalePerspectiveType` enum, as the scale is dependent on the
perspective - once from outside, once from inside of the scaled actor.
Closes https://gitlab.gnome.org/GNOME/mutter/-/issues/1312
Since we now have the neccessary infrastructure to get notified about
changes to the absolute transformation matrix, we can also invalidate
the stage-views list on updates to this matrix.
So rename absolute_allocation_changed() to absolute_geometry_changed()
to make it clear this function is not only about allocations, and call
that function recursively for all children on changes to the
transformation matrix, too.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1343
If we want to invalidate the stage-views list reliably on changes to the
actors transformation matrices, we also need to get notified about
changes to the custom transformations applied using the
apply_transform() vfunc.
So provide a new API that allows invalidating the transformation matrix
for actors implementing custom transformations, too. This in turn allows
us to cache the matrix applied using the apply_transform() vfunc by
moving responsibility of keeping track of the caching from
clutter_actor_real_apply_transform() to
_clutter_actor_apply_modelview_transform().
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1343
For ClutterText, the resource scale the text is drawn with affects the
size of the allocation: ClutterText will choose a font scale based on
the resource scale, and that font scale can lead to a slight difference
in size compared to the unscaled font.
We currently handle that by queuing a relayout inside the
"resource-scale-changed" signal handler. This solution is a bit
problematic though since it will take one more allocation cycle until
the allocation is actually updated after a scale-change, so the actor is
painted using the wrong allocation for one frame.
Also the current solution can lead to relayout loops in a few cases, for
example if a ClutterText is located near the edge on a 1x scaled monitor
and is moved to intersect a 2x scaled monitor: Now the resource scale
will change to 2 and a new allocation box is calculated; if this
allocation box is slightly smaller than the old one because of the new
font scale, the allocation won't intersect the 2x scaled monitor again
and the resource scale switches back to 1. Now the allocation gets
larger again and intersects the 2x scaled monitor again.
This commit introduces a way to properly support those actors: In case
an actors resource scale might affect its allocation, it should call the
private function clutter_actor_queue_immediate_relayout(). This will
make sure the actor gets a relayout before the upcoming paint happens
afte every resource scale change. Also potential relayout loops can
be handled by the actors themselves using a "phase" argument that's
passed to implementations of the calculate_resource_scale() vfunc.
The new API is private because resource scales are not meant to be used
in a way where the scale affects the allocation. With ClutterText and
the current behavior of Pango, that can't be avoid though, so we need it
anyway.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1276
Since we now always return a resource scale, we can remove the boolean
return value from clutter_actor_get_resource_scale() and
_clutter_actor_get_real_resource_scale(), and instead simply return the
scale.
While at it, also remove the underscore from the
_clutter_actor_get_real_resource_scale() private API.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1276
Now that ClutterActor has a convenient API for getting the stage views
an actor is presented on, we can remove a large part of the code for
resource-scale calculation and instead rely on the stage-views list.
The way this works is a bit different from the old resource scales:
clutter_actor_get_resource_scale() always returns a scale, but this
value is only guaranteed to be correct when called from a vfunc_paint()
implementation, in all other cases the value is guessed using the scale
of the parent actor or the last valid scale. Now in case the value
previously reported by clutter_actor_get_resource_scale() turns out to
be wrong, "resource-scale-changed" will be emitted before the next paint
and the actor has a chance to update its resources.
The general idea behind this new implementation is for actors which only
need the scale during painting to continue using
clutter_actor_get_resource_scale() as they do right now, and for actors
which need the resource scale on other occasions, like during size
negotiation, to use the scale reported by
clutter_actor_get_resource_scale() but also listen to the
"resource-scale-changed" signal to eventually redo the work using the
correct scale.
The "guessing" of the scale is done with the intention of always giving
actors a scale to work with so they don't have to fall back to a scale
value the actor itself has to define, and also with the intention of
emitting the "resource-scale-changed" signal as rarely as possible, so
that when an actor is newly created, it won't have to load its resources
multiple times.
The big advantage this has over the old resource scales is that it's now
safe to call clutter_actor_get_resource_scale() from everywhere (before,
calling it from size negotiation functions would usually fail). It will
also make it a lot easier to use the resource scale for complex cases
like ClutterText without risking to get into relayout loops.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1276
Add private API to ClutterBackend to set a fallback resource scale
available to Clutter. This API will be used for "guessing" the
resource-scale of ClutterActors in case the actor is not attached to a
stage or not properly positioned yet.
We set this value from inside mutters MetaRenderer while creating new
stage-views for each logical monitor. This makes it possible to set the
fallback scale to the scale of the primary monitor, which is the monitor
where most ClutterActors are going to be positioned.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1276
We're going to refactor resource scales, making the notification of
changes to the resource scale a lot more important than it is right now
(we won't guarantee queried scales are correct outside the paint cycle
anymore).
Having a separate signal/vfunc for this will make the difference between
the new clutter_actor_get_resource_scale() API (which can return a
guessed value) and the notification of changes to the resource scale
(which will be guaranteed to return an up-to-date value) more obvious.
So replace the "resource-scale" property of ClutterActor with a
"resource-scale-changed" signal that's emitted when the resource scale
is recalculated.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1276
The portal API requires a screencast session only for absolution motion
with remote desktop, other methods including relative motion do not
require a screencast session.
There is no reason to be more strict than the API actually is, check for
a screencast session only when required, like for absolute motion events
and touch events.
Tested with https://gitlab.gnome.org/snippets/1122
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1307
There are a couple of places in gnome-shell where we aren't interested
in which workspace is active, but whether a given workspace is active.
Of course it's easy to use the former to determine the latter, but we
can offer a convenience property on the workspace itself almost for
free, so let's do that.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1336
ClutterBoxLayout calculates the preferred size of the opposite
orientation (so for example the height if the orientation is horizontal)
by getting the preferred size of the real orientation first, and then
the preferred size of the opposite orientation, using the other size as
for_width/height when doing the request.
Right now, for non-homogeneous layouts this for_width/height does not
adjust for the spacing set on the box layout. This leads to children
being passed a slightly larger for_width/height, which in case of
ClutterText might cause the line to not wrap when it actually should.
This in turn means we can end up with an incorrect preferred size for
the opposite orientation, leading to a wrong allocation.
So fix that and adjust for the spacing just as we do for homogeneous
layouts by subtracting the total spacing from the available size that is
distributed between children.
This fixes the wrong height of the checkbox label reported in
https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/2574.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1333
These tests were written (and copy-pasted) before ClutterActor
had an actual background-color property. As a preparation to
the removal of ClutterRectangle, replace all these rectangles
with plain actors and background colors.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1332
The property is deprecated and the current implementation simply
redirects it to ClutterActor::background-color, so remove it.
Also update the tests to set the background color directly.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1332
ClutterStage is the one and only subclass of ClutterGroup, but
it overrides basically everything specific to ClutterGroup to
mimic a ClutterActor. What a waste!
Subclass ClutterActor directly and remove all the now useless
vfunc overrides from ClutterStage. Adapt CallyStage to subclass
CallyActor as well.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1332
It is deprecated in favor of the 'z-position' property, and
the implementation itself redirects to the z-position, so
just drop it and replace all get|set_depth calls to their
z-position counterparts.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1332
We were setting the pipeline colour to all white (1.0, 1.0, 1.0, 1.0)
and so the default layer combine function multiplied each pixel
(R, G, B, A) by all ones. Obviously multiplying by one four times per
pixel is a waste of effort so we remove the colour setting *and* set
the layer combine function to a trivial shader that will ignore whatever
the current pipeline colour is set to. So now we do **zero** multiplies
per pixel.
On an i7-7700 at UHD 3840x2160 this results in 5% faster render times
and 10% lower power usage (says intel_gpu_top). The benefit is probably
much higher for virtual machines though, as they're no longer being
asked to do CPU-based math on every pixel of a window.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1331
The previous commit removed checks for intermediate focus states which
would make tests randomly fail, because of their time dependence. What
can be tested however is that if there is no other window available that
would accept the focus, that the focus remains at 'none', after the
focused window has been closed. This newly introduced test checks the
focus directly after closing the window (and syncing) and after the time
it would have taken for the queue to finish. The first check has a
similar timing issue as the removed focus checks in the other tests, but
the test will never accidentally fail, because regardless of whether the
queue has finished or not, the focus is always expected to be 'none'.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1329
While c3d13203 ensured that the test-client has actually closed the
window before testing for the focus change, it also made another timing
related issue with the tests more likely to happen. Serveral tests
assert that the focus is set to 'none' after the focussed window has
been closed when the window below does not accept focus. This however
can never be reliably tested, because closing the window triggers
timeout based iteration of a queue of default focus candidate windows.
This starts after the window has been closed and might finish before the
clients have finished synchronizing. This issue is more likely to
trigger the shorter the queue is and the more test clients there are
that could delay the synchronization.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1329
This avoids some issues which could happen on some setups[0] due to
meta-native-renderer.c:dummy_power_save_page_flip →
meta_onscreen_native_swap_drm_fb implicitly turning of the primary
plane (by destroying the KMS framebuffer assigned to it):
* drmModeObjectSetProperty could return an "Invalid argument" error
between setting a non-empty cursor with drmModeSetCursor(2) and
enabling the primary plane again:
Failed to DPMS: Failed to set connector 69 property 2: Invalid argument
(This was harmless other than the error message, as we always re-set
a mode on the CRTC after setting the DPMS property to on, which
enables the primary plane and implicitly sets the DRM property to on)
* drmModeSetCursor(2) could return an "Invalid argument" error between
setting the DPMS property to on and enabling the primary plane again:
Failed to set hardware cursor (drmModeSetCursor failed: Invalid argument), using OpenGL from now on
[0] E.g. with the amdgpu DC display code.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1240
I noticed my system would fall back to the slow unclipped (and
uncullable) paint path whenever a window touched the left edge of
the screen. Turns out that was a red herring. Just that
`use_clipped_redraw` was uninitialized so clipping/culling was used
randomly.
So the compiler failed to notice `use_clipped_redraw` was uninitialized.
Weirdly, as soon as you fix that it starts complaining that `buffer_age`
might be uninitialized, which appears to be wrong. So we initialize that
too, to shut up the compiler warnings/errors.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1323
In commit 4c1fde9d MetaCullable related code was moved out of
MetaShapedTexture into MetaSurfaceActor. While generally desirable,
this removed drawing optimizations in MetaShapedTexture for partial
redraws. The common case for fully obscured actors was still supposed
to work, but it was now discovered that it actually did not.
This commit revert parts of 4c1fde9d: it reintroduces clipping
to MetaShapedTexture but leaves all culling and actor related logic
in MetaSurfaceActor.
Thanks to Daniel van Vugt for uncovering the issue.
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/850
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1295https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1326
When trying to find a default focus window, the code iterates through a
queue of candidates with a timeout between each candidate. If the window
the current timeout is waiting for gets destroyed, this process just
stops instead of trying the next window in the queue.
This issue was made more likely to be triggered with the previous change
to the closed-transient-no-input-parents-queued-default-focus-destroyed
test due to the introduction of a wait, which can introduce a
delay between the two destroy commands.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1325
Some tests were not waiting for the test client to actually issue
destroy commands before checking their effect on the window focus.
Similarly when mutter is supposed to change the focus based on a delay
by sending a WM_TAKE_FOCUS to the client, this also could fail without
synchronization with the client before checking the result.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1325
The ClutterBindConstraint will change the preferred size an actor
reports so it returns the same size as the source actor in some cases.
This behavior was introduced recently with 4f8e518d.
This can lead to infinite loops in case the source actor is a parent of
the actor the BindConstraint is attached to, that's because calling
get_preferred_size() on the source will recursively call
get_preferred_size() on the actor again.
So to avoid those loops, check if the source is a parent of the actor
we're attached to and don't update the preferred size in that case.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1282
For ClutterClones we need to apply a scale to the texture of the clone
to ensure the painted texture of the source actor actually fits the
allocation of the clone. We're doing this using the transformation
matrix instead of using the scale_x/scale_y properties of ClutterActor
to allow users to scale ClutterClones using that API independently.
Now it's quite a bad idea to get the allocation boxes for calculating
that scale using clutter_actor_get_allocation_box(), since that method
will internally do an immediate relayout of the stage in case the actor
isn't allocated. Another side effect of that approach is that it makes
it impossible to invalidate the transform (which will be needed when we
start caching those matrices) properly.
So since we eventually allocate both the source actor and the clone
ourselves anyway, we can simply use the allocation box inside
clutter_clone_allocate() (which is definitely updated and valid at that
point) to calculate the scale factor.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1181
It seems wrong to use the scale factor of the X axis on the Z axis and
it looks like this has been accidentally changed in commit 570fa3f044.
So use a factor of 1.0 instead to not scale the Z axis at all because
the layout machinery only works in X and Y coordinates.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1181
There are cases where a layout manager used by an actor also wants to
return a custom size when the actor has no children, for example in case
the layout manager requests a fixed size. This is currently impossible
because we only query the layout manager when calculating the preferred
size if the actor has children.
So fix that and also use the layout managers size negotiation functions
in case the actor has no children.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1322
_cogl_shader_set_source_with_boilerplate and _cogl_shader_compile_real
have enough GL assumptions that it makes sense to push them into the
backend. Taken together their only callers are under driver/gl, so.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1194
This had been an entirely-too-GL-aware collection of renderer queries,
mostly to work around driver bugs and handle software drivers
intelligently. The driver workarounds have been removed (fix your
driver, and if you can't because it's closed-source, fix that first),
and we now delegate the am-i-software-or-not logic to the backend, so
this can all go
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1194
We delegate the answer through CoglDriverVtable::is_hardware_accelerated
since this is properly a property of the renderer, and not something the
cogl core should know about. The answer given for the nop driver is
admittedly arbitrary, yes it's infinitely fast but no there's not any
"hardware" making it so.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1194
The size of the buffer the texture will be written to by
paint_to_buffer() is determined based on
meta_screen_cast_area_stream_src_get_specs() which uses roundf() to
calculate the width and height after scaling. Because the size of the
texture to be written to that buffer is calculated using ceilf(), it
might exceed the allocated buffer when using fractional scaling.
In 3.36 paint_to_buffer() is used from capture_view() which also uses
roundf() to allocate its buffer. Here this leads to a memory corruption
resulting in a crash when taking screenshots of an area.
Fixes https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/2842https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1320
Event delivery with grabbing popups stay essentially the same within
the grabbing client, we still must honor the implicit grab as long as
there is one.
This is however not the case, the popup_grab_focus() function ignores
the button state. The popup_grab_button() function will already indirectly
re-sync the focus surface after the last button is released, so checking
for button state in popup_grab_focus() is sufficient to make the implicit
grab honored with popup grabs involved.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1275https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1270
When a touch sequence was rejected, we'd update the event timestamps of
incoming touch events to help with implementing grabs. This was done by
sending a ClientMessage with a counter, and comparing the counter to
decide whether we're seing a replayed event or not.
This had the unforseen consequence that we would potentially end up
destroying all actors including the stage, since, when mutter receives a
ClientMessage event, it would assume that it's a WM_PROTOCOLS event, and
handle it as such. The problem with this approach is that it would
ignore fact that there might be other ClientMessage types sent to it,
for example the touch synchronization one. What could happen is that the
touch count value would match up with the value of the WM_DELETE_WINDOW
atom, clutter would treat this as WM_PROTOCOLS:WM_DELETE_WINDOW, which
it'd translate to clutter_actor_destroy(stage).
Destroying the stage in such a way is not expected, and caused wierd
crashes in different places depending on what was going on.
This commit make sure we only treat WM_PROTOCOLS client messages as
WM_PROTOCOLS client messages effectively avoiding the issue.
This fixes crashes such as:
#0 meta_window_get_buffer_rect (window=0x0, rect=rect@entry=0x7ffd7fc62e40) at core/window.c:4396
#1 0x00007f1e2634837f in get_top_visible_window_actor (compositor=0x297d700, compositor=0x297d700) at compositor/compositor.c:1059
#2 meta_compositor_sync_stack (compositor=0x297d700, stack=<optimized out>, stack@entry=0x26e3140) at compositor/compositor.c:1176
#3 0x00007f1e263757ac in meta_stack_tracker_sync_stack (tracker=0x297dbc0) at core/stack-tracker.c:871
#4 0x00007f1e26375899 in stack_tracker_sync_stack_later (data=<optimized out>) at core/stack-tracker.c:881
#5 0x00007f1e26376914 in run_repaint_laters (laters_list=0x7f1e2663b7d8 <laters+24>) at core/util.c:809
#6 run_all_repaint_laters (data=<optimized out>) at core/util.c:826
#7 0x00007f1e26b18325 in _clutter_run_repaint_functions (flags=flags@entry=CLUTTER_REPAINT_FLAGS_PRE_PAINT) at clutter-main.c:3448
#8 0x00007f1e26b18fc5 in master_clock_update_stages (master_clock=0x32d6a80, stages=0x4e5a740) at clutter-master-clock-default.c:437
#9 clutter_clock_dispatch (source=<optimized out>, callback=<optimized out>, user_data=<optimized out>) at clutter-master-clock-default.c:567
#10 0x00007f1e27e48049 in g_main_dispatch (context=0x225b8d0) at gmain.c:3175
#11 g_main_context_dispatch (context=context@entry=0x225b8d0) at gmain.c:3828
#12 0x00007f1e27e483a8 in g_main_context_iterate (context=0x225b8d0, block=block@entry=1, dispatch=dispatch@entry=1, self=<optimized out>) at gmain.c:3901
#13 0x00007f1e27e4867a in g_main_loop_run (loop=0x24e29f0) at gmain.c:4097
#14 0x00007f1e2636a3dc in meta_run () at core/main.c:666
#15 0x000000000040219c in main (argc=1, argv=0x7ffd7fc63238) at ../src/main.c:534
and
#0 0x00007f93943c1f25 in raise () at /usr/lib/libc.so.6
#1 0x00007f93943ab897 in abort () at /usr/lib/libc.so.6
#2 0x00007f9393e1e062 in g_assertion_message (domain=<optimized out>, file=<optimized out>, line=<optimized out>, func=0x7f93933e6860 <__func__.116322> "meta_x11_get_stage_window",
#3 0x00007f9393e4ab1d in g_assertion_message_expr ()
#4 0x00007f939338ecd7 in meta_x11_get_stage_window (stage=<optimized out>) at ../mutter/src/backends/x11/meta-stage-x11.c:923
#5 0x00007f939339e599 in meta_backend_x11_cm_translate_device_event (x11=<optimized out>, device_event=0x55bc8bcfd6b0) at ../mutter/src/backends/x11/cm/meta-backend-x11-cm.c:381
#6 0x00007f939339f2e2 in meta_backend_x11_translate_device_event (device_event=0x55bc8bcfd6b0, x11=0x55bc89dd5220) at ../mutter/src/backends/x11/meta-backend-x11.c:179
#7 0x00007f939339f2e2 in translate_device_event (device_event=0x55bc8bcfd6b0, x11=0x55bc89dd5220) at ../mutter/src/backends/x11/meta-backend-x11.c:208
#8 0x00007f939339f2e2 in maybe_spoof_event_as_stage_event (input_event=0x55bc8bcfd6b0, x11=0x55bc89dd5220) at ../mutter/src/backends/x11/meta-backend-x11.c:284
#9 0x00007f939339f2e2 in handle_input_event (event=0x7fff62d60490, x11=0x55bc89dd5220) at ../mutter/src/backends/x11/meta-backend-x11.c:309
#10 0x00007f939339f2e2 in handle_host_xevent (event=0x7fff62d60490, backend=0x55bc89dd5220) at ../mutter/src/backends/x11/meta-backend-x11.c:413
#11 0x00007f939339f2e2 in x_event_source_dispatch (source=<optimized out>, callback=<optimized out>, user_data=<optimized out>) at ../mutter/src/backends/x11/meta-backend-x11.c:467
#12 0x00007f9393e6c39e in g_main_dispatch (context=0x55bc89dd03e0) at ../glib/glib/gmain.c:3179
#13 0x00007f9393e6c39e in g_main_context_dispatch (context=context@entry=0x55bc89dd03e0) at ../glib/glib/gmain.c:3844
#14 0x00007f9393e6e1b1 in g_main_context_iterate (context=0x55bc89dd03e0, block=block@entry=1, dispatch=dispatch@entry=1, self=<optimized out>) at ../glib/glib/gmain.c:3917
#15 0x00007f9393e6f0c3 in g_main_loop_run (loop=0x55bc8a042640) at ../glib/glib/gmain.c:4111
#16 0x00007f9393369a0c in meta_run () at ../mutter/src/core/main.c:676
#17 0x000055bc880f2426 in main (argc=<optimized out>, argv=<optimized out>) at ../gnome-shell/src/main.c:552
Related: https://gitlab.gnome.org/GNOME/mutter/-/issues/338
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/951https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1317
The modifier state of the input device is supposed to be set to the
newest state, while the modifier state detail of the event is set to the
last state before the event (so not including the changes triggered by
the event).
So since the modifier state of the event is the last state anyway, the
state of the ClutterInputDevice is supposed to be set by the backend and
not by the stage while queuing the event, so stop setting the state
here.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1275
Make the clutter_input_device_get_actor() API public and remove
clutter_input_device_get_pointer_actor() in favour of the new function.
This allows also getting the "pointer" actor for a given touch sequence,
not only for real pointer input devices like mice.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1275
Switch from clutter_seat_list_devices() to the new peek_devices() method
of ClutterSeat in cases where we're only looping through the returned
list without manipulating it. This way we don't have to unnecessarily
copy around the list of devices.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1275
Add a method to ClutterSeat that allows peeking the list of input
devices and allow looping through devices a bit faster. The API left is
private so we can make use of peeking the GList internally, but don't
have to expose any details to the outside, which means we'd have to
eventually stick with a GList forever to avoid breaking API.
Since we now have the peek_devices() API internally, we can implement
ClutterSeats public list_devices() API using g_list_copy() on the list
returned by peek_devices().
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1275
While it's strongly discouraged, it is possible to queue a new relayout
of an actor in the middle of an allocation cycle, we warn about it but
don't forbid it.
With the introduction of the "shallow relayout" API, our handling of
those relayouts silently changed: Before introducing "shallow
relayouts", we'd handle them on the next stage update, but with the
priv->pending_relayouts hashtable and the
priv->pending_relayouts_version counter, we now do them immediately
during the same allocation cycle (the counter is increased by 1 when
queuing the relayout and we switch to a new GHashTableIter after
finishing the current relayout, which means we'll now do the newly
queued relayout).
This change in behavior was probably not intended and wasn't mentioned
in the commit message of 5257c6ecc2d2c843ecc1ad5558e6a471eff5c7ee, so
switch back to the old behavior, which is more robust in preventing
allocation-loops. To do this, use a GSList instead of GHashTable for the
pending_relayouts list, and simply steal that list before doing the
relayouts in _clutter_stage_maybe_relayout().
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1267
As explained in the last commits, we'll let gnome-shell take care of
this since freezing and thawing needs to be decoupled from the effect
starting and ending.
So stop freezing the MetaWindowActor when starting the effect and
thawing the actor when ending the effect.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1250
As explained in the last commit, gnome-shell needs to be able to thaw
window actor updates during its size-change effect is active.
So make meta_window_actor_freeze() and meta_window_actor_thaw() public
API, which will allow the shell to freeze and thaw actor updates itself.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1250
The size-change animation in gnome-shell needs to sync the window actors
geometry during the animation, it currently does this by notifying the
compositor that the animation was finished before it actually is.
This causes a few bugs in Mutter though, since it will now emit the
"effects-completed" signal on the window actor even though they aren't
completed.
To fix that, we need to decouple freezing and thawing of actor updates
from window effects and allow gnome-shell to thaw actor updates before
it notifies Mutter that the effect is completed.
The first step for this is allowing to sync the actor geometry while an
effect is active, this should be redundant since effects which actually
need to inhibit those updates will freeze the actor anyway. Also a
geometry change happening while another effect is active will kill the
old effect anyway because MetaPluginManager kills all the active window
effects before starting a new one; so the new size-change effect for any
geometry change is going to kill the current effect.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1250
Otherwise it can't be a GObject inheriting MetaCrtc. This fixes warning
logged as
"invalid uninstantiatable type '(null)' in cast to 'MetaCrtcKms'"
Fixes: 1ce9e379d99a8e15b3d043013062104f9b569574
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1308
ClutterAlignConstraint currently assumes the source actor is positioned
in the same coordinate system as the actor it's attached to and
automatically offsets the adjusted allocation by the origin of the
source actor.
This behavior is only valid though in case the source actor is a sibling
of the constraint actor. If the source actor is somewhere else in the
actor tree, the behavior gets annoying because the constraint actor is
offset by (seemingly) random positions.
To fix this, stop offsetting the constraint actors allocation by the
position of the source.
To still make it possible to align the constraint actors origin with the
origin of the source, no longer override the origin of the allocation
in the AlignConstraint. This allows users to align the origin using a
BindConstraint, binding the actor position to the position of the
source, which is more flexible and also more elegant.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/737
Add a new pivot-point property to the ClutterAlignConstraint, similar to
the pivot point used by ClutterActor, defining the point in the
constraint actor around which the aligning is applied to the actor.
Just as the ClutterActor property, this property is defined using a
GraphenePoint.
By default this property remains set to (-1, -1) and the actor
will always be aligned inside the source actor, preserving the existing
behavior of ClutterAlignConstraint.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/737
Trying to figure out what this comment was actually about, it turns out
that MSC means Media Stream Counter, and as mentioned in an article[0]
is related to DRI3 and the X11 Present extension. Anyway, the comment
has been there raising questions for some years now, I think we can
remove it.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1287
The ID and name are just moved into the instance private, while the rest
is moved to a `MetaCrtcModeInfo` struct which is used during
construction and retrieved via a getter. Opens up the possibility to
add actual sub types.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1287
Just as with MetaOutput, instead of the home baked "inheritance" system,
using a gpointer and a GDestroyNotify function to keep the what
effectively is sub type details, make MetaCrtc an abstract derivable
type, and make the implementations inherit it.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1287
Instead of the home baked "inheritance" system, using a gpointer and a
GDestroyNotify function to keep the what effectively is sub type
details, make MetaOutput an abstract derivable type, and make the
implementations inherit it.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1287
It's used for two things: avoid reading configs, and actual hotplug
update mode. The former requires the suggested position to be (-1, -1)
to trick the monitor configuration generator to skip using the suggested
position even if hotplug update mode is set to TRUE. The latter should
use the actual hotplug mode coordinates.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1287
Now set as a property during construction. Only actually set by the
Xrandr backend, as it's the only one currently not supporting all
transforms, which is the default.
While at it, move the 'ALL_TRANFORMS' macro to meta-monitor-tranforms.h.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1287
The output info is established during construction and will stay the
same for the lifetime of the MetaOutput object. Moving it out of the
main struct enables us to eventually clean up the MetaOutput type
inheritence to use proper GObject types.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1287
MetaCrtcInfo and MetaOutputInfo did not represent information about
MetaCrtc and MetaOutput, but the result of the monitor configuration
assignment algorithm, thus rename it to MetaCrtcAssignment and
MetaOutputAssignment.
The purpose for this is to be able to introduce a struct that actually
carries information about the CRTCs and outputs, as retrieved from the
backend implementations.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1287
That is is_presentation, is_primary, is_underscanning and backlight.
The first three are set during CRTC assignment as they are only valid
when active. The other is set separately, as it is untied to
monitor configuration.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1287
It was used during configuration to ensure that we always dealt with
every output and CRTC. Do this without polluting the MetaOutput and
MetaCrtc structs with intermediate variables not used by the
corresponding types themself.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1287
Now that we have a proper way to mark our allocation as uninitialized,
make use of that and only disallow implicit transitions of the
"allocation" property if that is the case.
This fixes a bug where easing the allocation of an actor is impossible
when someone queued a relayout on it (or a child of it) before.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1290
We currently initialize the ClutterActorBox of the actors allocation to
zero, but there's a difference between a valid zero-allocation and an
actor having never been allocated. Currently it's impossible for us to
detect the latter case in a reliable way and we use the needs_allocation
flag for this, which may also be set in other situations.
So initialize the allocation of actors to the newly added UNINITIALIZED
ClutterActorBox, which will make it easier to detect whether an actor
already got its initial allocation.
This also fixes another issue right now: Actors which get allocated a
(valid) zero allocation, will now notify the "allocation" property in
this case.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1290
Add support for an artificial UNINITIALIZED marking for ClutterActorBox,
done by setting the boxes origin to Infinity and its size to -Infinity.
That is a value that's considered an invalid allocation by Clutter and
which can never be set by sane code.
This will allow setting the allocation of ClutterActors to an
UNINITIALIZED box when creating actors or when removing them from the
scenegraph and makes it possible to explicitely detect uninitialized
allocations, which is useful in a few cases.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1290
The current code assumes that the actor will always have the same
size and position of the background texture, but part of the implicit
contract of being a ClutterContent is being able to render itself
at any given actor, at any given size.
For example, if the current code is given an actor with 0x0+100+100
as geometry, and no clipped region, it'll render not the whole
background, but the 0x0+100+100 rectangle of the background. In
practice, the actor geometry acts like a "clip mask" over the
background texture, due to the assumption that the actor will
always have the same size of the monitor.
Make the calculation of the texture slices relative to the actor
box.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1302
MetaBackgroundActor is still necessary for culling purposes,
but now the actual rendering of the background is delegated
to MetaBackgroundContent, as well as the sizing information.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1302
MetaBackgroundContent is a ClutterContent implementation
that can render a background to any attached actor. Right
now, it preserves all the properties and the rendering
model of MetaBackgroundActor.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1302
X11 window stacking operations are by nature prone to race conditions.
For example, we might queue a "raise above" operation, but before it
actually takes place, the sibling the window was to be rased above, is
withdrawn.
In these cases we'd log warnings even though they are expected to
happen. Downgrade these warnings to debug messages, only printed when
MUTTER_VERBOSE is set.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1300
Test that the stage-views list of ClutterActor is correct when moving an
actor, reparenting it, or hiding an actor up the hierarchy. Also test
that the "stage-views-changed" signal works as expected.
Don't test actor transforms for now because those aren't supported yet.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1196
We currently go through the whole tree of mapped actors on every paint
cycle to update the stage views actors are on. Even if no actors need
updating of their stage views, traversing the actor tree is still quite
expensive and shows up when using a profiler.
So tone down the amounts of full-tree traversals we have to do on every
paint cycle and only traverse a subtree if it includes an actor which
actually needs updating of its stage views.
We do that by setting the `needs_update_stage_views` flag to TRUE
recursively for all parents up to the stage when the stage-views list of
an actor gets invalidated. This way we end up updating a few more actors
than necessary, but can avoid searching the whole actor tree for actors
which have `needs_update_stage_views` set to TRUE.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1196
Add a new signal that's emitted when the stage views an actor being
painted on have changed, "stage-views-changed". For example this signal
can be helpful when tracking whether an actor is painted on multiple
stage views or only one.
Since we must clear the stage-views list when an actor leaves the stage
(actors that aren't attached to a stage don't get notified about the
stage views being changed/replaced), we also emit the new signal when an
actor gets detached from the stage (otherwise there would be an edge
case where no signal is emitted but it really should: An actor is
visible on a stage view, then detached from the stage, and then attached
again and immeditely moved outside the view).
Also skip the comparison of the old stage-views list and the new one if
nobody is listening to the signal to save some resources.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1196
There are certain rendering techniques and optimizations, for example
the unredirection of non-fullscreen windows, where information about the
output/stage-view an actor is on is needed to determine whether the
optimization can be enabled.
So add a new method to ClutterActor that allows listing the stage-views
the actor is being painted on: clutter_actor_peek_stage_views()
With the way Clutter works, the only point where we can reliably get
this information is during or right before the paint phase, when the
layout phase of the stage has been completed and no more changes to the
actors transformation matrices happen. So to get the stage views the
actor is on, introduce a new step that's done on every master clock tick
between layout and paint cycle: Traversing through the actor tree and
updating the stage-views the mapped actors are going to be painted on.
We're doing this in a separate step instead of inside
clutter_actor_paint() itself for a few reasons: It keeps the code
separate from the painting code, making profiling easier and issues
easier to track down (hopefully), it allows for a new
"stage-views-changed" signal that doesn't interfere with painting, and
finally, it will make it very easy to update the resource scales in the
same step in the future.
Currently, this list is only invalidated on allocation changes of
actors, but not on changes to the transformation matrices. That's
because there's no proper API to invalidate the transformation matrices
ClutterActor implementations can apply through the apply_transform()
vfunc.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1196
When the stage views the stage is shown on are changed, ClutterStage
currently provides a clutter_stage_update_resource_scales() method
that allows invalidating the resource scales of all actors. With the new
stage-views API that's going to be added to ClutterActor, we also need a
method to invalidate the stage-views lists of actors in case the stage
views are rebuilt and fortunately we can re-use the infrastructure for
invalidating resource scales for that.
So since resource scales depend on the stage views an actor is on,
rename clutter_stage_update_resource_scales() and related methods to
clutter_stage_clear_stage_views(), which also covers resource scales.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1196
While the layout manager of a ClutterActor does get properly unset when
destroying an actor, we currently forget to disconnect the
"layout-changed" signal from it.
So do that, and while at it, also switch to using the signal id for
disconnecting from the signal instead of
g_signal_handlers_disconnect_by_func(), which caused problems before
because it might traverse the signal handler list.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1281
We currently are confusing g_param_spec_enum and g_param_spec_flags for
the offscreen-redirect property of ClutterActor. Since it's actually a
flag, make it a flag everywhere.
Fun fact: This was already partly done with
d7814cf63eab1cf95625b06face9f914133d7eaa, but that commit missed the
setter.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1292
Just like the ClutterBindConstraint, the ClutterAlignConstraint should
listen to "queue-relayout" of its source actor, not
"notify::allocation". That's because the latter will queue a relayout
during an allocation cycle and might cause relayout loops.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1296
When an app disappears after some data from it has been copied to the
clipboard, the owner of the clipboard selection becomes a new memory
selection source. The initial reference this new selection source is
never unref'ed, which leads to this being leaked on the next clipboard
selection owner change.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1293
Using XDG_CONFIG_HOME allows users to place their keyboard configuration into
their home directory and have them loaded automatically.
libxkbcommon now defaults to XDG_CONFIG_HOME/xkb/ first, see
https://github.com/xkbcommon/libxkbcommon/pull/117
However - libxkbcommon uses secure_getenv() to obtain XDG_CONFIG_HOME and thus
fails to load this for the mutter context which has cap_sys_nice.
We need to manually add that search path as lookup path.
As we can only append paths to libxkbcommon's context, we need to start with
an empty search path set, add our custom path, then append the default search
paths.
The net effect is nil where a user doesn't have XDG_CONFIG_HOME/xkb/.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/936
Hiding a compositor stage is not something that's really supported, but
will still be used by tests, to get closer to a "fresh" stage for each
test case, when the tests eventually start using the mutter provided
stage.
It'll use that stage simply because creating standalone stages isn't
supported.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1289
The script parser only included G_PARAM_CONSTRUCT_ONLY parameters when
constructing objects. This causes issues if an object requires a
parameter to be set during construction, but may also change after. Fix
this by including G_PARAM_CONSTRUCT parameters when constructing script
objects as well.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1289
We would get the MetaDisplay from the backend singleton before creating
the MetaCompositor, then in MetaCompositor, get the backend singleton
again to get the stage. To get rid of the extra singleton fetching, just
pass the backend the MetaCompositor constructors, and fetch the stage
directly from the backend everytime it's needed.
This also makes it available earlier than before, as we didn't set our
instance private stage pointer until the manage() call.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1289
Start follow the convention used in ClutterFrameClock by including the
meaning as well as time granularity in the variable name. The
constructor takes the intended duration of the constructed timeline in
milli seconds, so call the constructor argument `duration_ms`. This is
done in preparation for adding more constuctors.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1289
Also fix a test that dependends on a specific element order in a list
that wasn't defined to have any particular order.
The frames per second is decreased from 30 to 10, to make the test less
flaky when running in CI.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1289
The shadow was disabled for the X11 client as it was far to unreliable
when comparing sizes.
It seems that the Wayland backend has been somewhat unreliable as well,
where some race condition causing incorrect sizes thus a flaky test.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1288
A "show" command calls gtk_window_show() and gdk_display_sync(), then
returns. This means that the X11 window objects are guaranteed to have
been created in the X11 server.
After that, the test runner will look up the window's associated
MetaWindow and wait for it to be shown.
What this doesn't account for is if mutter didn't get enough CPU time to
see the new window. When this happens, the 'default-size' stacking test
sometimes failed after hiding and showing the X11 window.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1288
If a test is not expected to succeed, then running it could be considered
to be a waste of resources, particularly if the failure might manifest
as an indefinite hang (see cogl!11), or if the test is likely to dump core
and trigger "expensive" crash-reporting mechanisms like systemd-coredump,
corekeeper, abrt or apport.
Skip the tests that are expected to fail. They can still be requested via
an environment variable, which can be set after fixing a bug to check which
tests are now passing.
Originally cogl!15, adapted for mutter's fork of cogl to use gboolean
instead of CoglBool.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1272
Signed-off-by: Simon McVittie <smcv@debian.org>
For actors which don't have needs_allocation set to TRUE and where the
new allocation wouldn't be different from the old one, the allocate()
vfunc doesn't have to be called. We still did this in case a parent
actor was moved though (so the absolute origin changed), because we
needed to propagate the ABSOLUTE_ORIGIN_CHANGED allocation flag down to
all actors.
Since that flag is now removed and got replaced with a private property,
we can simply notify the children about the absolute allocation change
using the existing infrastructure and safely stop allocating children at
this point.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1247
With commit 0eab73dc2e we introduced an optimization of not doing
allocations for actors which are hidden. This broke the propagation of
absolute origin changes to hidden actors, so if an actor is moved while
its child is hidden, the child will not get
priv->needs_compute_resource_scale set to TRUE, which means the resource
scale won't be updated when the child gets mapped and shown again.
Since we now have priv->absolute_origin_changed, we can simply check
whether that is TRUE for our parent before bailing out of
clutter_actor_allocate() and if it is, notify the whole hidden sub-tree
about the absolute origin change.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1247
Since clutter_stage_set_viewport() is only used inside clutter-stage.c
anyway, we can make it a static method. Also we can remove the x and y
arguments from it since they're always set to 0 anyway.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1247
When getting the last allocation using
clutter_actor_get_allocation_box(), Clutter will do an immediate
relayout of the stage in case an actor has an invalid allocation. Since
the allocation is always invalid when the allocate() vfunc is called,
clutter_stage_allocate() always forces another allocation cycle.
To fix that, stop comparing the old allocation to the new one to find
out whether the viewport changed, but instead use the existing check in
_clutter_stage_set_viewport() and implement the behavior of rounding the
viewport to the nearest int using roundf() (which should behave just as
CLUTTER_NEARBYINT()) since we're passing around floats anyway.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1247
When manipulating the allocation of a ClutterActor from an allocate()
vfunc override, clutter_actor_set_allocation() is used to let Clutter
know about the changes.
If the actors allocation or its absolute origin did not change before
that, this can also affect the actors absolute_origin_changed property
used by the children to detect changes to their absolute position.
So fix this bug (which luckily didn't seem to affect us so far) and set
priv->absolute_origin_changed to TRUE in case the origin changes inside
clutter_actor_set_allocation_internal(). Since this function is always
called when our allocation changes, we no longer need to update
absolute_origin_changed in clutter_actor_allocate() now.
Since a change to the absolute origin always affects the resource scale,
too, we also need to move that check from clutter_actor_allocate() here
to make sure we update the resource scale.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1247
Since the introduction of the shallow relayout functionality it's
possible to start an allocation cycle at any point in the tree, not only
at the stage. Now when starting an allocation at an actor that's not the
stage, we'd still look at the absolute_origin_changed property of this
actors parent, which might still be set to TRUE from the parents last
allocation.
So avoid using the parents absolute_origin_changed property from the
last allocation in case a shallow relayout is being done and always
reset the absolute_origin_changed property to FALSE after the allocation
cycle.
This broke with the removal of the ABSOLUTE_ORIGIN_CHANGED
ClutterAllocationFlag that was done in commit dc8e5c7f.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1247
When we created the DMA buffer backed CoglFramebuffer, we handed it over
to CoglDmaBufHandle which took its own reference. What we failed to do
was to release our own reference to it, effectively leaking it.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1283
The stream will clean up the buffers, so let it do that before we
destroy them under its feet. Note that it'll only do this after the
following PipeWire commit:
commit fbaa4ddedd84afdffca16f090dcc4b0db8ccfc29
Author: Wim Taymans <wtaymans@redhat.com>
Date: Mon Jun 1 15:36:09 2020 +0200
stream: allow NULL param and 0 buffers in disconnect
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1283
wait_reconfigure ensures that the whole configure back and forth
completes before continuing. Doing this after every state change ensures
that we always end up with the expected state, thus fixes flakyness of
the restore-position stacking test.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1279
This cannot be made to work reliably. Some factoids:
- Internal devices may be connected via USB.
- The ACPI spec provides the _PLD (Physical location of device) hook to
determine how is an USB device connected, with an anecdotal success
rate. Internal devices may be seen as external and vice-versa, there is
also an "unknown" value that is widely used.
- There may be non-USB keyboards, the old "AT Translated Set 2 Keyboard"
interface does not change on hotplugging.
- Libinput has an internal series of quirks to classify keyboards as
internal of external, also with an "unknown" value.
These heuristics are kinda hopeless to get right by our own hand. Drop
this external keyboard detection in the hope that there will be something
more deterministic to rely on in the future (e.g. the libinput quirks
made available to us directly or indirectly).
Fixes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/2378
Related: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/2353https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1277
When r is 128 or more, running tests compiled with the undefined behaviour
sanitizer (ubsan) reports:
test-utils.c:312:45: runtime error: left shift of 128 by 24 places cannot be represented in type 'int'
which indeed it cannot. Force the type to be unsigned 32-bit so that we
get defined behaviour.
Similarly, in test-atlas-migration, the left-shifted guint8 is promoted
to int, which again does not have enough non-sign bits available to
left-shift a value >= 128 by 24 bits. Again, force the shift to be done
in unsigned 32-bit space.
This was originally cogl!22, but applies equally to mutter's fork of cogl.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1271
Signed-off-by: Simon McVittie <smcv@debian.org>
According to the cogl_bitmap_new_for_data documentation, the data is not
copied, so the application must keep the buffer alive for the lifetime
of the CoglBitmap. Freeing it too early led to a use-after-free in the
cogl unit tests. With that fixed, the test passes, so remove the known
failure annotation.
This AddressSanitizer trace is from the original cogl, but the bug and
fix apply equally to mutter's fork of cogl:
==6223==ERROR: AddressSanitizer: heap-use-after-free on address 0x62100001a500 at pc 0x7f3e2d4e7f4e bp 0x7ffcd9c41f30 sp 0x7ffcd9c416e0
READ of size 4096 at 0x62100001a500 thread T0
#0 0x7f3e2d4e7f4d (/usr/lib/x86_64-linux-gnu/libasan.so.5+0x96f4d)
#1 0x7f3e260c7f6b in util_copy_box ../src/gallium/auxiliary/util/u_surface.c:131
#2 0x7f3e268c6c10 in u_default_texture_subdata ../src/gallium/auxiliary/util/u_transfer.c:67
#3 0x7f3e26486459 in st_TexSubImage ../src/mesa/state_tracker/st_cb_texture.c:1480
#4 0x7f3e26487029 in st_TexImage ../src/mesa/state_tracker/st_cb_texture.c:1709
#5 0x7f3e26487029 in st_TexImage ../src/mesa/state_tracker/st_cb_texture.c:1691
#6 0x7f3e2644bdba in teximage ../src/mesa/main/teximage.c:3105
#7 0x7f3e2644bdba in teximage_err ../src/mesa/main/teximage.c:3132
#8 0x7f3e2644d84f in _mesa_TexImage2D ../src/mesa/main/teximage.c:3170
#9 0x7f3e2cd1f7df in _cogl_texture_driver_upload_to_gl driver/gl/gl/cogl-texture-driver-gl.c:347
#10 0x7f3e2ccd441b in allocate_from_bitmap driver/gl/cogl-texture-2d-gl.c:255
#11 0x7f3e2ccd441b in _cogl_texture_2d_gl_allocate driver/gl/cogl-texture-2d-gl.c:462
#12 0x7f3e2ce3a6c0 in cogl_texture_allocate cogl/cogl-texture.c:1398
#13 0x7f3e2ce3e116 in _cogl_texture_pre_paint cogl/cogl-texture.c:359
#14 0x7f3e2cdee177 in _cogl_pipeline_layer_pre_paint cogl/cogl-pipeline-layer.c:864
#15 0x7f3e2cd574af in _cogl_rectangles_validate_layer_cb cogl/cogl-primitives.c:542
#16 0x7f3e2cdd742f in cogl_pipeline_foreach_layer cogl/cogl-pipeline.c:735
#17 0x7f3e2cd5c8b0 in _cogl_framebuffer_draw_multitextured_rectangles cogl/cogl-primitives.c:658
#18 0x7f3e2cd60152 in cogl_rectangle cogl/cogl-primitives.c:858
#19 0x5570a71ed6a0 in check_texture tests/conform/test-premult.c:103
#20 0x5570a71ed946 in test_premult tests/conform/test-premult.c:159
#21 0x5570a71df0d6 in main tests/conform/test-conform-main.c:58
#22 0x7f3e2bcd809a in __libc_start_main ../csu/libc-start.c:308
#23 0x5570a71e0869 in _start (/home/smcv/src/debian/cogl/tests/conform/.libs/test-conformance+0x33869)
0x62100001a500 is located 0 bytes inside of 4096-byte region [0x62100001a500,0x62100001b500)
freed by thread T0 here:
#0 0x7f3e2d5581d7 in __interceptor_free (/usr/lib/x86_64-linux-gnu/libasan.so.5+0x1071d7)
#1 0x5570a71ed58b in make_texture tests/conform/test-premult.c:69
previously allocated by thread T0 here:
#0 0x7f3e2d558588 in malloc (/usr/lib/x86_64-linux-gnu/libasan.so.5+0x107588)
#1 0x7f3e2d384500 in g_malloc ../../../glib/gmem.c:99
This was originally cogl!12.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1274
Signed-off-by: Simon McVittie <smcv@debian.org>
In clutter_text_queue_redraw_or_relayout() we check whether the size
of the layout has changed and queue a relayout if it did, otherwise we
only queue a redraw and save some resources.
The current check for this also queues a redraw if the actor has no
valid allocation. That seems right on the first glance since the actor
will be allocated anyway, but we actually want to call
clutter_actor_queue_relayout() again here because that also invalidates
the size cache of the actor which might have been updated and marked
valid in the meantime.
So make sure the size cache is always properly invalidated after the
size of the layout changed and also call clutter_actor_queue_relayout()
in case the actor has no allocation.
This fixes a bug where getting the preferred width of a non-allocated
ClutterText, then changing the string of the ClutterText, and then
getting the preferred width again would return the old cached width (the
width before we changed the string).
The only place where this bug is currently happening is in the overview,
where we call get_preferred_width() on the unallocated ClutterText of
the window clone title: When the window title changes while the
ClutterText is unallocated the size of the title is going to be wrong
and the text might end up ellipsized or too large.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1150
Move Wayland support (i.e. the MetaWaylandCompositor object) made to be
part of the backend. This is due to the fact that it is needed by the
backend initialization, e.g. the Wayland EGLDisplay server support.
The backend is changed to be more involved in Wayland and clutter
initialization, so that the parts needed for clutter initialization
happens before clutter itself initialization happens, and the rest
happens after. This simplifies the setup a bit, as clutter and Wayland
init now happens as part of the backend initialization.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1218
On X11 we don't update the texture in certain circumstances, such as if
the surface is a fullscreen unredirect, or doesn't have a Pixmap. On
Wayland we only want to avoid updating the texture if there is no
texture, but as this is handled implicitly by MetashapedTexture, we
don't need to try to emulate the X11-y conditions in the generic layer
and instead just have the implementations handle update processing
themself.
This doesn't have any functional changes, but removes a vfunc from
MetaSurfaceActorClass.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1218
It's effectively used by mutter by abusing a ClutterTimeline to scedule
updates. Timelines are not really suited in places that is done, as it
is really just about getting a single new update scheduled whenever
suitable, so expose the API so we can use it directly.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1218
We could call clutter_stage_schedule_update() and it wouldn't actually
schedule anything, as the master frame clock only tries to reschedule if
1) there is an active timeline, 2) there are pending relayouts, 3) there
are pending redraws, or 4) there are pending events. Thus, a call to
clutter_stage_schedule_update() didn't have any effect if it was called
at the wrong time.
Fix this by adding a boolean state "needs_update" to the stage, set on
clutter_stage_schedule_update() and cleared on
_clutter_stage_do_update(), that will make the master clock reschedule
an update if it is TRUE.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1218
We need to use the framebuffer of the view instead of the onscreen
framebuffer when painting the damage region, otherwise the redraw clips
on rotated monitors won't be shown correctly.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1237
Compare, tile by tile, whether actual damage actually changed any
pixels. While this requires mmap():ing DMA buffers and comparing their
content, we should only ever use shadow buffers when we're using the
software renderer, meaning mmap() is cheap as it doesn't involve any
downloading.
This works by making the shadow framebuffer double buffered, while
keeping track of damage history. When we're about to swap the onscreen
buffer, we compare what part of the posted damage actually changed,
records that into a damage history, then given the onscreen buffer age,
collect all actual damage for that age. The intersection of these tiles,
and the actual damage, is then used when blitting the shadow buffer to
the onscreen framebuffer.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1157https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1237
Move the damage history tracking to a new ClutterDamageHistory helper
type. The aim is to be able to track damage history elsewhere without
reimplementing the data structure and tracking logic.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1237
This fixes the last "copy everything" paths when clutter doesn't
directly paint onto the onscreen framebuffer. It adds a new hook into
the stage view called before the swap buffer, as at this point, we have
the swap buffer damag regions ready, which corresponds to the regions we
must blit according to the damage reported to clutter.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1237
The rest didn't change, so only actually paint the part of the offscreen
that was composited as part of the stage painting. In practice, this
means that, unless a shadow buffer is used, we now only paint the
damaged part of the stage, and copy the damage part of the offscreen to
the onscreen.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1237
We failed to use the buffer age when monitors were rotated, as when they
are, we first composite to an offscreen framebuffer, then later again to
the onscreen. The buffer age checking happened on the offscreen, and an
offscreen being single buffered, they can't possible support buffer
ages.
Instead, move the buffer age check to check the actual onscreen
framebuffer. The offscreen to onscreen painting is still always full
frame, but that will be fixed in a later commit.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1237
The manual "cleaning" of the viewport and projection state is removed,
and we only ever try to invalidate the state so that it'll be updated
next time. Change the API used to reflect this.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1237
The stage would fetch the front framebuffer and set the viewport and
projection matrix, but if we are going to more than one front buffer,
that won't work, so let the stage just pass the viewport and projection
matrix to the view and have the view deal with the framebuffer(s).
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1237
Will be used for logging to identify what view a log entry concerns. For
the native and nested backend this is the name of the output the CRTC is
assigned to drive; for X11 it's just "X11 screen", and for the legacy
"X11 screen" emulation mode of the nested backend it's called "legacy
nested".
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1237
There's no reason to notify the surface that its geometry changed when
the visibility of the actor changes. This is only needed to update the
outputs of the surface, so do that directly instead.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1235
We started listening to notify::mapped with commit
5eb5f724347ae8d141ef38b36565169c60c53ab1 in order to emit
wl_surface.leave events consistently when a surface gets hidden. This
caused a problem with the ClutterClones used in the overview, since
those temporarily map and unmap the windows for painting, spamming
wl_surface.leave and enter events to all surfaces.
We can easily fix that by also treating mapped clones as mapped, which
means the surface should also be on a wl_output when the overview is
shown.
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1141https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1235
All existing users of clutter_actor_has_mapped_clones() actually want to
know whether the actor is being cloned by a visible clone, it doesn't
matter to them if that clone is attached to an actor somewhere else in
the tree or to the actor itself.
So make clutter_actor_has_mapped_clones() a bit more convenient to use
and also check the clones of the parent-actors in that function.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1235
We started listening to "notify::position" on surface actors with commit
08e4cb54. This commit was done to fix a regression from commit cf1edff9,
which forgot to handle some cases like the actual WindowActor and not
the SurfaceActor (which is a child of the WindowActor) moving (that was
fixed by listening to MetaWindows "position-changed" signal). Also that
commit introduced meta_wayland_surface_update_outputs_recursively(),
which updates the outputs of all (sub-)surfaces in case any position
changed and made sure subsurfaces also get their outputs updated in case
the parent actor moved.
Connecting to the "notify::position" signal, which the above commit also
did is now superflous though because position changes will queue a
relayout and the actors allocation will change during the next
allocation cycle, notifying the "allocation" property which we also
listen to.
So save some resources and stop listening to that signal.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1235
The comment in _clutter_actor_get_allocation_clip() explicitely notices
that it doesn't need the behavior of doing an immediate relayout as
clutter_actor_get_allocation_box() does. The comment is also still valid
since the code calling _clutter_actor_get_allocation_clip() checks for
priv->needs_allocation just before.
So let's just use the allocation directly here instead of going through
that function.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1264
We don't have enough Xlib code in mutter ...
Joking aside, it can be useful to make the cursor invisible
without hiding it, for example for replacing the actual cursor
with an actor in gnome-shell; the real cursor should still
update the focus surface in that case, and not sneak into
screenshots or -casts.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1244
We're iterating inside the PipeWire loop when detecting PipeWire errors,
and shouldn't destroy the PipeWire objects mid-iteration. Avoid this by
first disabling the stream src (effectively stopping the recording),
then notifying about it being closed in an idle callback. The
notification eventually makes the rest of the screen cast code clean up
the objects, including the src and the associated PipeWire objects, but
will do so outside the PipeWire loop iteration.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1251https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1251
In the native backend, the MetaRenderer manages the view by creating one
per CRTC, but until now the MetaStageX11 managed the view for the X11
backend. This caused some issues as it meant meta_renderer_get_views()
not returning anything, and that the view of the X11 screen not being a
MetaRendererView, while in the other backends, all views are.
Fix this by moving the view management responsibility to
MetaRendererX11Cm, and have MetaStageX11 only operate on it via
meta_renderer_x11_cm_*() API. The MetaRendererX11Cm takes care of making
sure the view is always added to the list in the renderer, and turning
X11 screen sizes into "layouts" etc.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1251
"Legacy" is a misleading name, it's just how the native backend and the
X11 backend behaves differently. Instead rename it to 'add_view()' and
add the sanity check to the caller.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1251
A DMA buffer might not be able to scanout, and in that case the import
with GBM_BO_USE_SCANOUT will fail. Handle that by failing to scanout,
effectively falling back to compositing.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1261
Since PIDs are inherently insecure because they are reused after a
certain amount of processes was started, it's possible the client PID
was spoofed by the client.
So make sure users of the meta_window_get_pid() API are aware of those
issues and add a note to the documentation that the PID can not be
totally trusted.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1180
Since the PID of a window can't change as long as the window exists, we
can safely cache it after we got a valid PID once, so do that by adding
a new `window->client_pid` private property.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1180
The shell uses the PID of windows to map them to apps or to find out
which window/app triggered a dialog. It currently fails to do that in
some situations on Wayland, because meta_window_get_pid() only returns a
valid PID for x11 clients.
So use the client PID instead of the X11-exclusive _NET_WM_PID property
to find out the PID of the process that started the window. We can do
that by simply renaming the already existing
meta_window_get_client_pid() API to meta_window_get_pid() and moving
the old API providing the _NET_WM_PID to meta_window_get_netwm_pid().
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1180
ClutterBoxLayout's layout policy of using the generic ClutterActor
align/expand properties for children that are expanded and a custom
meta otherwise is confusing, in particular as the x-fill/y-fill
defaults don't match the default CLUTTER_ACTOR_ALIGN_FILL align.
StBoxLayout's own custom child meta (which was deprecated last
cycle) is probably the only consumer. And luckily, the St meta
uses different x-fill/y-fill default that match the ClutterActor
defaults, so removing it will not affect code that doesn't use
the deprecated properties themselves.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1265
This stuff has been deprecated for a very long time, and given that
ClutterBoxLayout is most commonly used via StBoxLayout, the impact of
removing it should be low. It will however open the door to further
cleanups.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1265
We're going to remove the "allocation-changed" signal from ClutterActor
since it's no longer needed now that ClutterAllocationFlags are gone.
So listen to "notify-allocation" instead, which has been the recommended
thing to do for some time now anyway.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1245
The ABSOLUTE_ORIGIN_CHANGED allocation flag is only really useful to
propagate the information of the absolute origin of an actor having
changed inside Clutter. It wasn't used anywhere else besides for some
debug messages and it probably shouldn't be used in custom layout
implementations anyway since 1) actors shouldn't have to be aware of
absolute allocation changes and 2) it doesn't factor in changes to the
transformation matrix of a parent.
Also the propagation of absolute origin changes using this flag broke
with commit 0eab73dc2e and now hidden actors are no longer notified
about those changes.
Additionally, this flag gets in the way of a few potential optimizations
since it has to be propagated even if the allocation box of the child
hasn't changed, forcing a reallocation of the child.
So replace this flag with a simple new private property of ClutterActor
absolute_origin_changed, but keep the exact same behavior for now.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1245
Since we now only layout the children ourselves in case the actor
implementation doesn't override the allocate vfunc, we can remove
clutter_actor_maybe_layout_children() and move the functionality inside
clutter_actor_real_allocate().
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1245
Now that we no longer have the DELEGATE_LAYOUT we expect all actors
overriding the allocate() vfunc to allocate their children themselves.
Since clutter_actor_set_allocation() is only called from custom
vfunc_allocate() implementations, the condition in
clutter_actor_maybe_layout_children() would always fail, which makes
calling the function useless anyway.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1245
The CLUTTER_DELEGATE_LAYOUT flag is unintuitive and makes the allocation
process inside Clutter unnecessarily complicated. It's very easy for
actors overriding the allocate() vfunc to layout their children
themselves (in fact most of them do this), and it also never made sense
that clutter_actor_set_allocation() does eventually layout children.
There was no ClutterActor implementation in mutter or gnome-shell which
actually used the DELEGATE_LAYOUT flag, but even without it, it's fairly
easy to archive the same behavior now: In the allocate() override,
adjust the allocation as wanted, then chain up to the parent vfunc
without calling clutter_actor_set_allocation().
So remove the CLUTTER_DELEGATE_LAYOUT flag, which will allow making the
relayout code in Clutter a bit easier to follow.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1245
We're going to remove allocation flags, so stop depending on the
DELEGATE_LAYOUT flag in ClutterStage and call
clutter_layout_manager_allocate() directly, which is pretty
straightforward.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1245
They all checked that the remote session service talked with the
correct peer, and some of them did check that there is an associated
screencast session.
Add a new check for the session being started (as it's state is
decoupled with screencast session availability) and move all checks
to a function that is called from all input-oriented DBus methods.
Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1254https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1258
The public API to get the parent actor, clutter_actor_get_parent() does
a type check whether the actor is actually a ClutterActor. In case of
_clutter_actor_apply_relative_transformation_matrix(), which is called
recursively and very often during the paint process, this type check
shows up with almost twice the amount of hits than the actual matrix
multiplication.
So use the parent pointer directly in some code paths that are executed
very often and avoid the expensive type checking there, we can do that
since both places are not public API.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1259
It was removed in 3.34 as part of 6ed5d2e2. And we thought that was the
only thread that might exist and use X11. But the top gnome-shell crasher
in 3.36 seems to suggest otherwise.
We don't know what or where the offending thread is, but since:
1. We used XInitThreads for years already prior to 3.34; and
2. Extensions or any change to mutter/gnome-shell could conceivably use
threads to make X calls, directly or indirectly,
it's probably a good idea to reintroduce XInitThreads. The failing assertion
in libx11 is also accompanied by a strong hint:
```
fprintf(stderr, "[xcb] Most likely this is a multi-threaded client " \
"and XInitThreads has not been called\n");
```
https://bugs.launchpad.net/bugs/1877075
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1252https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1256
Reverting the scale and offset applied to the damage history can be done
in one step, using a few less temporary allocations by passing the
offset right away to a new scale_offset_and_clamp_region() function.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1113
Since the damage history region is tracked per-view, all the regions it
includes should be inside the current view anyway, so don't
unnecessarily intersect that region with the view.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1113
Since we now check for the buffer age before setting up the
fb_clip_region, that region will be set to the full extents of the view
in case the buffer age is invalid. This in turn means we don't have to
do this again later and can simply fill the damage history with the
fb_clip_region that's already set for us.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1113
Since a NULL redraw_clip means that a full view redraw should be done
and an empty redraw clip may never be set (see the width/height checks
in clutter_stage_view_add_redraw_clip()), the fb_clip_region should
always be set to a reasonable region that's either the whole view or
individual regions inside the view.
So make sure that's actually the case by warning and that the
fb_clip_region isn't empty, which allows dropping another few lines of
code.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1113
Right now we're checking for the DISABLE_CLIPPED_REDRAWS debug flag
after creating the fb_clip_region and adjusting the redraw_clip. That
means that if may_use_clipped_redraw was TRUE, the redraw_clip will
still be set to the region and thus cause the stage to only be partially
redrawn. Since we don't push a clip to the framebuffer though
(use_clipped_redraw is now FALSE), parts of the view will get corrupted.
To fix that, disable clipped redraws right away if the debug flag is
set. This also allows removing the may_use_clipped_redraw bool and
replacing it entirely with use_clipped_redraw.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1113
We already have a better way to paint the redraw clip: Painting the
damage region paints the individual rects of the clip region and not
only the bounding rect.
So stop painting an outline around the redraw clip bounding rect when
CLUTTER_DEBUG_REDRAWS is set.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1113
While this is meant as an optimzation to only use the scissor clip and
not the stencil buffer if there's only one clip rectangle, it's not
needed since this optimization is going to be applied to region clips
anyway inside _cogl_clip_stack_gl_flush() (see cogl-clip-stack-gl.c).
So remove the unnecessary optimization here and rely on cogl-clip-stack
to do it for us.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1113
This was introduced with commit 9ab338d7b6 because the clipping of
fractionally scaled redraws caused glitches, it seems like this is no
longer needed nowadays, so let's remove it.
This should make obscured region culling work a bit better for
fractionally scaled framebuffers because because we overdraw a slightly
smaller region than the actually damaged one. We still do overdraw
though since the clipping region is stored using integers and thus
any non-integer values have to be extended to the bounding rect.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1113
It doesn't make sense to set the redraw clip when painting the stage if
clipped redraws are disabled. That's because when visualizing the redraw
clip and any new redraws are clipped, the old visualiziations would
remain visible, leaving multiple confusing rectangles on the screen.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1113
When the wallpaper image is larger than the monitor resolution we already
use mipmapping to scale it down smoothly in hardware. We use
`GL_TEXTURE_MIN_FILTER` = `GL_LINEAR_MIPMAP_LINEAR` for the highest quality
scaling that GL can do. However that option is designed for 3D use cases
where the mipmap level is changing over time or space.
Since our wallpaper is not changing distance from us we can improve the
rendering quality even more than `GL_LINEAR_MIPMAP_LINEAR`. To do this we
now set `GL_TEXTURE_MAX_LEVEL` (if available) to limit the mipmap level or
blurriness level to the lowest resolution (highest level) that is still
equal to or higher than the monitor itself. This way we get the benefits
of mipmapping (downscaling in hardware) *and* retain the maximum possible
sharpness for the monitor resolution -- something that
`GL_LINEAR_MIPMAP_LINEAR` alone doesn't do.
Example:
Monitor is 1920x1080
Wallpaper photo is 4000x3000
Mipmaps stored on the GPU are 4000x3000, 2000x1500, 1000x750, ...
Before: You would see an average of the 2000x1500 and 1000x750 images.
After: You will now only see the 2000x1500 image, linearly sampled.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/1003
It's very useful to have common functions for easily creating a monitor
test setup for all kinds of tests, so move create_monitor_test_setup()
and check_monitor_configuration() and all the structs those are using to
monitor-test-utils.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1243
We're going to move some structs from monitor-unit-tests.c to
monitor-test-utils.h and some names are currently clashing with the
struct names here, so rename those to be specific to the
MonitorStoreUnitTests.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1243
check_monitor_test_clients_state() is a function that's only meant to be
used in the monitor-unit-tests, and since we're going to move the
functions for creating MonitorTestSetups into a common file, this
function is going to be in the way of that. So move the checking of the
test client state outside of check_monitor_test_clients_state().
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1243
We're going to move the functions for building MonitorTestSetups to the
common monitor-test-utils.c file.
To make building test setups a bit more straightforward in case no
TestCaseExpect is wanted, change create_monitor_test_setup() to take a
MonitorTestCaseSetup instead of a MonitorTestCase as an argument.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1243
Commit e06daa58c335b7 changed the tested values to use corresponding valid
enum values instead of negative ones. Unfortunately that made one value
become a duplicate of an existing one and also in part defeated the original
intention of checking the implementation of
`meta_output_crtc_to_logical_transform`.
Use `meta_monitor_transform_invert` to fix both shortcomings.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1242
One of the important classes in Mutter's handling of client textures is
the `MetaShapedTexture`. This commit adds a few gtk-doc comments which
explain its purpose, with special attention to the viewport methods.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1210
Since we're now connecting to one more signal of MetaWaylandOutput, keep
signal connections in one place and move connecting the
"output-destroyed" signal to surface_entered_output() and disconnecting
it to surface_left_output().
This also allows us to use the "outputs_to_destroy_notify_id" as a
simple set and rename it to "outputs".
While at it, also use g_hash_table_destroy() instead of
g_hash_table_unref() since destroy is more clear than unref and does the
same thing in this case.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1230
When hotplugging a new monitor, we recreate all the MetaWaylandOutputs
and need to emit leave events to the surfaces for the old wl_outputs and
enter events for the newly created ones.
There's a race condition though: We might update the monitors a surface
is on (and thus emit enter/leave events for the wl_outputs) before the
Wayland client is registered with the new wl_output (ie. the
bind_output() callback of MetaWaylandOutput was called), which means we
don't send an enter event to the client in surface_entered_output().
Since MetaWaylandSurface now has the MetaWaylandOutput in its outputs
hashtable, it thinks the client has been notified and won't send any
more enter events.
To fix that, make MetaWaylandOutput emit a new signal "output-bound"
when a client bound to the output and make all surfaces which are on
that output listen to the signal. In the signal handler compare the
newly added client to the client the surface belongs to, and if it's the
same one, send an enter event to that client.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1230
The "output-destroyed" signal is used for notifying MetaWaylandSurfaces
that an output they are shown just got invalid (for example because a
monitor hotplug happened).
While we delay the destroying of outputs by 10 seconds since commit
1923db97 because of a race-condition, it doesn't make sense to wait 10
seconds until we let surfaces know that an output was destroyed.
So move the emission of the "output-destroyed" signal to
make_output_inert(), which is called before we start the 10 seconds
delay.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1230
When tiling, we want to set the tile monitor. To not have to do this
from the call site, make meta_window_tile() fall back to the current
monitor if nothing set it prior to the call.
This will make it more convenient for test cases to test tiling
behavior.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1171
The test tests that (for both X11 and Wayland) that:
* The client unmaximizes after mapping maximized to a predictable size
* That the client unmaximizes to the same size after toggling maximize
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1171
This makes sure that a client has properly responded to a configure
event it itself triggered. In practice, this is just two 'wait'
commands, with a 'dispatch' in between, which is needed because a single
one does not reliably include the two way round trip happening when e.g.
responding to a unmaximize configure event triggered by a unmaximize
request.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1171
The 'assert_size' command checks that the size of the window, both
client side and compositor side, corresponds to an expected size set by
the test case.
The size comparison can only be done when the window is using 'csd', in
order for both the client and server to have the same amount of
understanding of the title bar. For ssd, the client cannot know how
large the title bar, thus cannot verify the full window size.
Sizes can be specified to mean the size of the monitor divided by a
number. This is that one can make sure a window is maximized or
fullscreened correctly.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1171
Gtk is quite buggy and "fluid" in how it handles the shadow margins for
windows under X11. The "size" of the window fluctuate between including and
excluding a shadow margin in a way that causes issues, as there are no
atomic update of any state going on.
In order to avoid running into those particular issues now, lets get rid
of shadows so the margins are always zero, when the client is using the
X11 backend.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1171
To get some kind of consistency between what 'resize' means for the
compositor and the client, make the size correspond to the "frame rect"
of the window, i.e. the window geometry in the Wayland case, and the
window size including the titlebar in the X11 case.
This is so that the window size later can be reliably compared both in
the compositor and in the client using the same expected dimensions.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1171
When toying with the test client to try to reproduce issues (e.g.
writing commands on stdin to create and manipulate windows), when you
write a command incorrectly you'll get a warning printed to standard
out. The problem, however, is that it doesn't include a line break in
the end, meaning when you type the correct command, it won't be on a new
line.
Fix this minor annoyance by adding line breaks to all warning messages.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1171
The test client could already understand the resize command, but they
could not be added to metatests as the command was not properly plumbed
via the test runner. Establish the plumbing for the resize command so
that resize tests can be added.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1171
We only test the meson-build job, so there's no point to save artifacts for
the other test-build only builds.
So, only save meson logs artifacts (with a default gitlab expiration time)
for the other build-without-* jobs
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1236
While we don't have an high number of tests, we still have some code
coverage and so we can track this via gitlab CI, given that it supports it
natively.
So add gcovr to the DockerFile dependency, build with -Db_coverage=true
meson native parameter, and add another manual job to make ninja to generate
the coverage reports on requests or in any master or tag ref.
Keep the artifacts around to be able to browse the generated HTML files and
eventually print the text reports so that they can be parsed by gitlab.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1236
This removes ClutterAnimation and related tests. ClutterAnimation has
been deprecated for a long time, and replacements exist and are used by
e.g. GNOME Shell since a while back.
This also disables a few relatively unrelated interactive tests, as they
rely on ClutterAnimation to implement some animations they use to
illustrate what they actually test.
As interactive tests currently are more or less untestable due to any
interaction with them crashing, as well as they in practice means
rewriting the tests using non-deprecated animation APIs, they are not
ported right now. To actually port the interactive tests, it needs to be
possible to fist interact with them.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1192
In the past, it was a odd mix of possible different types, all coalesced
into an unsigned integer. Now, hovewer, it's always a
ClutterAnimationType, so lets change the name of getter, setter and
property to what it really is.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1192
When memfd_create isn't used, the file isn't sealed. Therefore, we
should skip test_readonly_seals on the fallback case. This fixes
compilation error on FreeBSD 12, which does not support memfd_create.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1229
We were iterating through evcodes, but using API that expects Clutter button
numbers. Instead of transforming those to Clutter numbers to have those translated
back, use the inner seat API that already takes evcodes.
Fixes stuck buttons keys after a virtual device is destroyed while those are
pressed.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1239
The CLUTTER_ACTOR_IN_REPARENT and the CLUTTER_IN_REPARENT flag are never
set and the logic for skipping unmap, unrealize and the emission of the
"parent-set" signal during reparents has been solved differently by
leaving out the CHECK_STATE and EMIT_PARENT_SET flags when calling
add_child_internal() and remove_child_internal().
The only place where those REPARENT flags are theoretically still useful
is in the clutter_actor_verify_map_state() debugging function, but that
is never called during reparent anyway, so simply leave the comment
regarding reparent there.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1228
Move to center uses all monitors for calculating work area.
This can lead to an unexpected behaviour on some monitor
configurations resulting in current window being split between
monitors. We should move window to the center of the active display.
Closes https://gitlab.gnome.org/GNOME/mutter/-/issues/1073
Inhibiting remote access means any current remote access session is
terminated, and no new ones can be created, until remote access is
uninhibited. The inhibitation is ref counted, meaning there can be more
than one inhibitor.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1212
When resizing an X11 window with client side decorations, the shadow is
clipped by the frame bounds so that we don't need to paint the shadow
under the opaque areas covered by the window and its frame.
When the X11 client uses the EMWH synchronization mechanism (like all
gtk-3 based clients), the actual window may not be updated so that the
actual window and it frame may be behind the expected window frame
bounds, which gives the impression of de-synchronized shadows.
To avoid the issue, keep a copy of the frame bounds as a cache and only
update it when the client is not frozen so that the clipping occurs on
the actual content.
Closes: https://gitlab.gnome.org/GNOME/mutter/issues/1178https://gitlab.gnome.org/GNOME/mutter/merge_requests/1214
The redraw clip that's painted together with the damage region has to be
copied earlier than we do right now. That's because if
PAINT_DAMAGE_REGION is enabled, buffer age is disabled and thus
use_clipped_redraw is FALSE. That means the redraw_clip is updated and
set to the full view-rect. If we copy the queued_redraw_clip after that,
it's also going to be set to the full view-rect. So copy the redraw clip
a bit earlier to make sure we're actually passing the real redraw clip
to paint_damage_region().
Also keep the queued_redraw_clip around a bit longer so it can actually
be used by paint_damage_region() and isn't freed before that.
While at it, move paint_damage_region() from swap_framebuffer() into
clutter_stage_cogl_redraw_view() so we don't have to pass things to
swap_framebuffer() only for debugging.
Fixes https://gitlab.gnome.org/GNOME/mutter/-/issues/1104https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1208
It takes coordinates in stage coordinate space, and will result in
a screen cast stream consisting of that area, but scaled up by the scale
factor of the view that overlaps with the area and has the highest scale
factor.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1207
Will be used by the stage to not paint the overlays. We skip all
overlays since overlays are only ever used for pointer cursors when the
hardware cursors cannot or should not be used.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1207
These phase callbacks are not intended to be inovked when something
secondary is painting the stage, such as a screen cast stream, or
similar. Thus, only invoke the callbacks when there is a view associated
with the paint context, which will not be the case for offscreen
painting.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1207
Either onto a framebuffer, or into a CPU memory buffer. The latter will
use an former API and then copy the result to CPU memory. The former
allocates an offscreen framebuffer, sets up the relevant framebuffer
matrices and paints part of the stage defined by the passed rectangle.
This will be used by a RecordArea screen cast API. The former to paint
directly onto PipeWire handled dma-buf framebuffers, and the latter for
PipeWire handled shared memory buffers.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1207
If there is a paint context available (i.e. for the phases that are
during the actual stage paint), pass it along the callbacks, so that
the callback implementations can change their operation depending on the
paint context state.
This also means we can get the current view from the paint context,
instead of the temporarily used field in the instance struct.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1207
A paint flag affects a paint operation in ways defined by the flags.
Currently no flags are defined, so no semantical changes are defined
yet. Eventually a flag aiming to avoid painting of cursors is going to
be added, so that screen cast streams can decide whether to include a
cursor or not.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1207
If drmModePageFlip() or custom_page_flip_func fails, process_page_flip() was
forgetting to undo the ref taken for that call. This would leak page_flip_data.
The reference counting works like this:
- when created, ref count is 1
- when calling drmModePageFlip, ref count is increased to 2
- new: if flip failed, ref count is decreased back to 1
- if calling schedule_retry_page_flip(), it takes a ref internally
- if calling mode_set_fallback(), it takes a ref internally
- all return FALSE paths have an explicit unref
- return TRUE path has an explicit unref
This issue was found by code inspection and while debugging an unrelated issue
with debug prints sprinkled around. I am not aware of any end-user visible
issues being fixed by this, as the leak is small and probably very rare.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1209
When testing a laptop with intel and DisplayLink devices, attempting to set the
DL output as the only active output resulted in GNOME/Wayland freezing. The
main event loop was running fine, but nothing on screen would get updated once
the DL output become the only one. This patch fixes that issue.
DisplayLink USB 3 devices use an out-of-tree kernel DRM driver called EVDI.
EVDI can sometimes fail drmModePageFlip(). For me, the flip fails reliably when
hotplugging the DL dock and when changing display configuration to DL only.
Mutter has a workaround for failing flips, it just calls drmModeSetCrtc() and
that succeeds.
What does not work reliably in the fallback path is Mutter keeping track of the
pageflip. Since drmModePageFlip() failed, there will not be a pageflip event
coming and instead Mutter queues a callback in its stead. When you have more
than one output, some other output repainting will attempt to swap buffers and
calls wait_for_pending_flips() which has the side-effect of dispatching any
queued flip callbacks. With multiple outputs, you don't get stuck (unless they
all fail the exact same way at the same time?). When you have only one output,
it cannot proceed to repaint and buffer swap because the pageflip is not marked
complete yet. Nothing dispatches the flip callback, leading to the freeze.
The flip callback is intended to be an idle callback, implemented with a
GSource. It is supposed to be called as soon as execution returns to the main
event loop. The setup of the GSource is incomplete, so it will never dispatch.
Fix the GSource setup by setting its ready-time to be always in the past. That
gets it dispatched on the next cycle of the main event loop. This is now the
default behavior for all sources created by meta_kms_add_source_in_impl().
Sources that need a delay continue to do that by overriding the ready-time
explicitly.
An alternative solution could have been to implement GSource prepare and check
callbacks returning TRUE. However, since meta_kms_add_source_in_impl() is used
by flip retry code as well, and that code needs a delay through the ready-time,
I was afraid I might break the flip retry code. Hence I decided to use
ready-time instead.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1209
According to the XSetSelectionOwner libX11 documentation:
[...] If the owner window it has specified in the request is later
destroyed, the owner of the selection automatically reverts to None,
but the last-change time is not affected.
This is indeed visible through the selection_timestamp field in
XFixesSelectionNotify events.
Use this to check whether the selection time is recent-ish (thus
likely coming from an explicit XSetSelectionOwner request) and honor
the client intent by setting a "NULL" owner. If the selection time
is too old, it's definitely an indication of the owner client being
closed, the scenario where we do want the clipboard manager to take
over.
This fixes two usecases:
- X11 LibreOffice / WPS clear the selection each time before copying
its own content. Mutter's clipboard manager would see each of those
as a hint to take over, competing with the client over selection
ownership. This would simply no longer happen
- Password managers may want to clear the selection, which would be
frustrated by our clipboard manager.
There's a slight window of opportunity for the heuristics to fail
though, if a X11 client sets the selection and closes within 50ms, we
would miss the clipboard manager taking over.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1206
The X11 selection source was being preserved after unsetting its
ownership. This is no leak as it would be eventually replaced by
another source, or destroyed on finalize. But it's pointless to
keep it.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1206
Test the two modes of MetaAnonymousFile, MAPMODE_SHARED and
MAPMODE_PRIVATE and make sure they don't leak data to other FDs when
writing to an FD provided by `meta_anonymous_file_get_fd` even though
the data of both FDs is residing in the same chunk of memory.
We do all the reading tests using mmap instead of read() since using
read() on shared FDs is going to move the read cursor of the fd. That
means using read() once on the shared FD returned by
meta_anonymous_file_get_fd() in MAPMODE_PRIVATE breaks every subsequent
read() call.
Also test the fallback code of MetaAnonymousFile in case `memfd_create`
isn't used for the same issues.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/1012
Add MetaAnonymousFile, an abstraction around anonymous read-only files.
Files can be created by calling meta_anonymous_file_new(), passing the
data of the file. Subsequent calls to meta_anonymous_file_open_fd()
return a fd that's ready to be sent over the socket.
When mapmode is META_ANONYMOUS_FILE_MAPMODE_PRIVATE the fd is only
guaranteed to be mmap-able readonly with MAP_PRIVATE but does not
require duplicating the file for each resource when memfd_create is
available. META_ANONYMOUS_FILE_MAPMODE_SHARED may be used when the
client must be able to map the file with MAP_SHARED but it also means
that the file has to be duplicated even when memfd_create is available.
Pretty much all of this code was written for weston by Sebastian Wick,
see https://gitlab.freedesktop.org/wayland/weston/merge_requests/240.
Co-authored-by: Sebastian Wick <sebastian@sebastianwick.net>
https://gitlab.gnome.org/GNOME/mutter/merge_requests/1012
Instead of having everything clumped at MetaWaylandDataManager,
split the primary selection to its own struct. This manager is
handled separately from wl_data_device_manager and other selection
managers, so they would be able to interoperate between them, even.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1193
This is still an openly defined struct, as we will need accessed
by "subclasses". Same principle applies than with the
MetaWaylandDataSource refactor, this is not meant to introduce
functional changes, so just go with it.
On the bright side, the interactions are now clearer, so it could
be made saner in the future.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1193
The split wasn't 100% clean, and some extra private API had to be
added for it (but well, looking at the API, it's already evident
there's a cleanup/streamlining task due). This is meant to be a
refactor with no functional changes, so just go with it.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1193
We already have a signal callback that translates selection ownership changes to
data_device/primary .selection events. Given both will be run when a data source
is being replaced, and this event emission being deleted is kinda short sighted
in that in only knows about Wayland, rely entirely on MetaSelection::owner-changed
emission.
Fixes spurious .selection(null) events being sent when a compositor-local source
takes over the selection without the focus changing (eg. screenshot to clipboard).
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1160https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1204
We are meant to send a .cancelled event after the drop is performed
in certain situations, but only for version>3 clients. Since this is
all version 3 business, only set the drop_performed flag for v3
clients. This drops the need to perform version checks at the time
of cancelling (which is present for other usecases in v1).
Fixes emission of wl_data_source.cancelled for v1 clients.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1177https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1203
For the cases where we read a fixed size from the selection (eg. imposing
limits for the clipboard manager), g_input_stream_read_bytes_async() might
not read up to this given size if the other side is spoonfeeding it content.
Cater for multiple read/write cycles here, until (maximum) transfer size is
reached.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1198
Flushing the X11 selection output stream may happen synchronously or
implicitly, in which case there is not a task to complete. Check there
is actually a task before returning errors. We additionally set the
pipe_error flag, so future operations will fail with an error, albeit
with a more generic message.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1198
If a write_async() comes up while we are flushing on the background,
the task will be queued, but not deemed a reason on itself to keep
flushing (and finish the task) after a property delete event.
To fix this, do not ever queue up write_async tasks (this leaves
priv->pending_task only used for flush(), so the "flush to end"
behavior in the background is consistent). We only start a
background flush if there's reasons to do it, but the tasks are
immediately finished.
All data will still be ensured to be transfered on flush/close,
this makes the caller in this situation still able to reach to it.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1198
It does not make sense to check for the stream not being closed,
this might happen multiple times during the lifetime of the stream
for a single transfer. We want to notify the INCR transfer just
once.
Check for the explicit conditions that we want, that the remaining
data is bigger than we can transfer at once, and that we are not
yet within the INCR transfer.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1198
The stream automatically flushes after data size exceeds the
size we deem for INCR chunks, but we still try to copy it all.
Actually limit the data we copy, and leave the rest for future
INCR chunks.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1198
INCR transfers are mandated to finish with a final 0-size XChangeProperty
roundtrip after the final data chunk. Actually honor this and ensure we
iterate just once more for this.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1198
This seemed to work under the assumption that a flush() call can
only result in one INCR roundtrip. This is evidently not true, so
we should hold things off until all pending data is actually flushed.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1198
If say we want 32bit data, but have 2 bytes stored, we would simply
ignore flush requests. Allow (and don't clear) the needs_flush flag
if we have less than the element size accumulated.
Instead handle this in can_flush(), so it's triggered whenever we
have enough data to fill 1 element, or if the stream is closing
(seems a broken situation, but triggered by the caller).
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1198
XMaxRequestSize/XMaxExtendedRequestSize are documented to return
the maximum size in 4-byte units, whereas we are comparing this
to byte lenghts. We can afford 4x the data here.
Since I don't know the payload size of the XChangeProperty request,
be generous and allot 400 bytes for it, we have some to spare.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1198
When closing the lid of a laptop, we reconfigure all the monitors in order
to update the CRTCs and (if enabled) the global UI scaling factor.
To do this, we try first to reuse the current configuration for the usable
monitors, but if we have only monitor enabled and this one is on the laptop
lid we just end up creating a new configuration where the primary monitor is
the laptop one (as per find_primary_monitor() in MetaMonitorConfigManager),
but ignoring the user parameters.
In case the user selected a different resolution / scaling compared to the
default one, while the laptop lid is closed we might change the monitors
layout, causing applications to rescale or reposition.
To avoid this, when creating the monitors configuration from the current
current state, in case we have only one monitor available and that one is
the laptop panel, let's just reuse this configuration.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1200
Try to bypass compositing if there is a fullscreen toplevel window with
a buffer compatible with the primary plane of the monitor it is
fullscreen on. Only non-mirrored is currently supported; as well as
fullscreened on a single monitor. It should be possible to extend with
more cases, but this starts small.
It does this by introducing a new MetaCompositor sub type
MetaCompositorNative specific to the native backend, which derives from
MetaCompositorServer, containing functionality only relevant for when
running on top of the native backend.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
We need to coordinate with MetaCompositor during pre-paint so that we
have control over whether MetaLater callbacks happen first, or the
MetaCompositor pre-paint logic.
In order to do so, make MetaLater listen to a new signal "pre-paint" on
MetaCompositor, that is called MetaCompositors own pre-paint handling.
This fixes an issue where the top window actor was calculated after the
MetaCompositor pre-paint handling, meaning the top actor being painted
was out-of-date.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
Since the order of destruction during MetaDisplay tear down is a bit
unordered, there are pieces that try to destruct its compositing
dependent pieces (i.e. queued MetaLater callbacks) after MetaCompositor
has been cleaned up, meaning we need to put some slightly awkward NULL
checks to avoid crashing.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
MetaCompositor is the place in mutter that manages the higher level
state of compositing, such as handling what happens before and after
paint. In order for other units that depend on having a compositor
instance active, but should be initialized before the X11 implementation
of MetaCompositor registers as a X11 compositing manager, split the
initialization of compositing into two steps:
1) Instantiate the object - only construct the instance, making it
possible for users to start listening to signals etc
2) Manage - this e.g. establishes the compositor as the X11 compositing
manager and similar things.
This will enable us to put compositing dependent scattered global
variables into a MetaCompositor owned object.
For now, compositor management is internally done by calling a new
`meta_compositor_do_manage()`, as right now we can't change the API of
`meta_compositor_manage()` as it is public. For the next version, manual
management of compositing will removed from the public API, and only
managed internally.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
While at it, fix some style inconsistencies, for now use a single
singleton struct instead of multiple static variables, and
other non-functional cleanups. Semantically, there is no changes
introduced.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
This will check whether the current backing buffer is compatible with
the primary plane of the passed CoglOnscreen. Since this will extend the
time before a buffer is released, the MetaWaylandBufferRef is swapped
and orphaned if a new buffer is committed before the previous one was
released. It'll eventually be released, usually by the next page flip
callback.
Currently implemented for EGLImage and DMA-BUF buffer types.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
Advertising support for modifiers means we will most likely not not be
able to scan out client buffers directly, meaning it just as likely that
we won't be able to scan out even fullscreen windows without atomic KMS.
When we have atomic support, we should advertise support for modifiers
if atomic is used to drive the CRTCs, as we by then can check whether we
can scan out directly, place in an overlay plane, etc.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
While this is fairly incomplete, as to check things fully we need to use
TEST_ONLY in atomic to try out a complete assignment on the device, but
this works well enough for legacy non-modifier cases.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
Make it possible to cause the next frame to scan out directly from the
passed CoglScannout. This makes it possible to completely bypass
compositing for the following frame.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
Instead of always swapping buffers and flipping the back buffer, make it
possible to scan out a provided buffer directly without swapping any EGL
buffers.
A buffer is passed as an object implementing the empty CoglScanout
interface. It is only possible to do this in the native backend; and the
interface is implemented by MetaDrmBufferGbm. When directly scanned out,
instead of calling gbm_surface_lock_front_buffer() to get the gbm_bo and
fbid, get it directly from the MetaDrmBufferGbm, and use that to create
the page flip KMS update.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
Transitions are used for animating actors when e.g. going from/to
fullscreen, and the like. We need to know such things when deciding
whether to avoid compositing a window actor, so make add API visible to
mutter that checks whether there are any transitions active.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
Better to have the relevant object figure out whether it is a good
position to be unredirectable other than the actor, which should be
responsible for being composited.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
This removes the MetaWindowX11::priv pointer. It is replaced with a
meta_window_x11_get_private() helper function, and another method to get
the client rect without going through MetaWindowX11Private.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
Surface buffers are created with meta_drm_buffer_new_acquire(), taking a
gbm_surface acquiring the gbm itself, and meta_drm_buffer_new_take()
that takes over ownership of a passed gbm_bo.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
Currently a buffer use count always reaches zero before it is replaced.
This is due to the fact that at the point a new buffer is attached, the
last potential user releases it (the stage) since the currently
displayed frame has a composited copy of the buffer.
This may however change, if a buffer is scanned out directly, meaning it
should not be released until the page flip callback is invoked.
Prepare for this by making the buffer reference a heap allocated struct,
enabling us to keep a pointer to it longer than the buffer is attached.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/798
The CRTC level transform (not necessarily the hw transform) must be
taken into account when calculating the position of the CRTC in the
stage coordinate space, when placing the hw cursor, otherwise we'll
place the cursor as if the monitor was not rotated.
This wasn't a problem in the past, as with rotation, we always used the
OpenGL cursor, so the issue newer showed.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1199
The port to per CRTC views was incomplete; we still used the logical
monitor layout as the stage view layout, while still using one view per
CRTC.
This worked fine for most cases, e.g. regular monitors, tiled or
non-tiled, transformed or non-transformed. Where it broke, however, was
when a monitor consists of multiple CRTCs. We already have the layout a
CRTC corresponds to on the stage kept with the CRTC metadata, so use
this directly.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1170https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1199`
The CRTC level transform (i.e. not necessarily the one set on the
hardware) is what is relevant for calculating the layout the CRTC will
have on the stage, so only use the one that can be handled by the
hardware for the CRTC assignment.
This makes the CRTC layout valid for tiled monitors.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1199
Previously the tile coordinate was used to offset a CRTC scanout
coordinate within a larger framebuffer. Since 3.36 we're always
scanning out from (0, 0) as we always have one framebuffer per CRTC; we
instead use the tile coordinate to calculate the coordinate the tile has
in the stage view. Adapt calculation to fulfil this promise instead of
the old one.
This also corrects the tiled custom monitor test case.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1199
test_client_new might return early if conditions are not met, leaving some
allocated data around without freeing it.
Since we're not using the client before, there's no need to initialize it early
and just initialize it when it's going to be returned.
https://gitlab.gnome.org/GNOME/mutter/merge_requests/1195
Dereference the loop variable rather than the original list head. This
fixes a regression introduced in 4413b86a3 ("backends: Replace
ClutterDeviceManager usage in favor of ClutterSeat", 2019-10-04) which
broke button scrolling with trackballs.
Closes:https://gitlab.gnome.org/GNOME/mutter/-/issues/1120
Most visible with xwayland-on-demand, at the time of setting things up
for X11 selections, we don't forward the current state. This makes the
first started X11 app oblivious to eg. the current clipboard.
Syncing selections up at the time of initializing the X11 selection
stuff ensures that doesn't happen.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1186
click_action_query_long_press() can potentially schedule more than
one timeout, since it doesn't clear any already-existing timeout.
Make sure to clear the long press timeout before scheduling a new
one.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1188
Like the click action, it makes sense to cancel the ongoing gesture
when the action is disabled. Do so by overriding our new friend,
ClutterActorMeta.set_enabled, and canceling the gesture when disabling
the action.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1188
ClutterClickAction, like other actions, can potentially be disabled
at any time (that is not during painting). When that happens with
ClutterClickAction, it must release all timeouts and disconnect from
the stage's 'capture-event'.
Override ClutterActorMeta.set_enabled and release the click action
when the action is being disabled.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1188
Various subclasses of ClutterActorMeta need to reacto to being
disabled. Right now, however, the only way to do that is by
overriding GObject's 'notify' vfunc, and doing a string comparison
against "enabled".
Add a new vfunc to ClutterActorMeta in order to replace this bad
practice.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1187
On VT switch, the devices are removed, which means for Wayland disabling
the keyboard.
When the keyboard is disabled, the associated `xkb_state` is freed and
recreated whenever the keyboard is re-enabled when switching back to the
compositor VT.
That means the `xkb_state` for Wayland is lost whereas the same for
clutter is kept, which causes to a discrepancy with locked modifiers on
VT switch.
To avoid that issue, preserve the XKB info only to dispose it when the
keyboard is eventually finalized.
Closes: https://gitlab.gnome.org/GNOME/mutter/issues/344https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1185
In the unlikely case we have multiple rectangles in our selection
(selection spanning several lines, or across LTR/RTL bounds), paint each
of those instead of setting a CoglPath-based clip/fill.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1126
The motion events of tablets for example need to be mapped on the
selected screen area if the input device is configured to use only a
part of the active logical monitor.
To achieve this behavior each motion event is transformed using the
transformation matrix set for the input device.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/1118
At some point we crossed the streams... In a short timespan we had
1f00aba92c32 merged, pushing WacomDevice to a common parent object,
and dcaa45fc0c199 implementing device grouping for X11.
The latter did not rely on the former, and just happened to
merge/compile without issues, but would promptly trigger a crash
whenever the API would be used.
Drop all traces of the WacomDevice internal to MetaInputDeviceX11.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1183
With the introduction of "shallow" relayouts, we are now able to enter
allocation cycles not only at the stage but also deeper down the
hierarchy if we know an actors allocation isn't affected by its children
since the NO_LAYOUT flag is set.
Now that means when queuing relayouts it's possible that
`priv->needs_allocation` gets set to TRUE for some actors down the
hierarchy, but not for actors higher up in the hierarchy. An actor tree
where that happens could look like that:
stage -> container -> container2 (NO_LAYOUT) -> textActor
With that tree, if the "textActor" queues a relayout, "container2" will
be added to the relayout hashtable of the stage and the actors "stage"
and "container" will have `priv->needs_allocation` set to FALSE.
Now if another relayout on the stage actor is queued,
`clutter_stage_queue_actor_relayout()` currently removes all the other
hashtable entries in favour of the stage entry, (wrongly) assuming that
will allocate everything. It doesn't allocate everything because in the
example above "container" has `priv->needs_allocation` set to FALSE,
which makes clutter_actor_allocate() return early before allocating its
children, so in the end "container2" will never get a new allocation.
To fix this, stop flushing the relayout hashtable when queuing a
stage-relayout and still add new entries to the hashtable if a stage
relayout is already queued to make sure we still go through all the
previously queued "shallow" relayouts. That shouldn't hurt performance,
too, because as soon as an actor got allocated once, it doesn't need an
allocation anymore and should bail out in clutter_actor_allocate() as
long as it's absolute position didn't change.
Fixes https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/2538https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1173
We have branched now, time for a shiny new CI image.
Update the Dockerfile to:
- switch to F32
- use a single shared copr
- drop dependencies that are now covered by builddep
- do not include weak deps
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1176
Currently we check whether a window is alive everytime it's focused.
This means that an application that doesn't respond to the check-alive
event during startup always showing the "application froze" dialog,
without the user ever trying to interact with it.
An example where this tends to to happen is with games, and for this
particular scenario, it's purely an annoyance, as I never tried to
interact with the game window in the first place, so I don't care that
it's not responding - it's loading.
To avoid these unnecessary particular "app-is-frozen" popups, remove the
alive check from the focus function, and instead move it back to the
"meta_window_activate_full()" call. To also trigger it slightly more
often, also add it to the path that triggers the window focus when a
user actively clicks on the window.
This means that we currently check whether a window is alive on:
* Any time the window is activated. This means e.g. alt-tab or
selecting the window in the overview.
* The user clicks on the window.
Note that the second only works for an already focused window on
Wayland, as on X11, we don't refocus it. This particular case isn't
changed with this commit, as we didn't call meta_window_focus() to begin
with here.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1182
This fixes an issue where a non-maximized screen casted window would be
stretched to fill the whole screen cast stream, instead of just the crop
that corresponds to the current window size.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1174
It isn't immediately obvious that this is impossible, because there's some
"action at a distance" going on with framebuffers that have their size
set lazily, after their textures get allocated; so let's make this a
critical warning rather than crashing.
In particular, this works around a crash when gnome-shell tries to blur a
background that hasn't yet had any space allocated for it - which it seems
is really an actor layout bug, but more robustness seems good to have.
Workaround for <https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/2538>.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1172
Signed-off-by: Simon McVittie <smcv@debian.org>
A texture with no pixels isn't a useful thing to have, and breaks
assumptions elsewhere. For example, CoglFramebuffer assumes that after
a texture has been allocated, it will have width and height both greater
than 0.
In particular, this works around a crash when gnome-shell tries to blur a
background that hasn't yet had any space allocated for it - which it seems
is really an actor layout bug, but more robustness seems good to have.
Workaround for <https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/2538>.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1172
Signed-off-by: Simon McVittie <smcv@debian.org>
Disabling a click action after a button-press but before a
button-release is captured makes ClutterClickAction connect to
captured-event and never disconnect.
This change fixes it by making sure the captured-event is only
processed if the action is still enabled, otherwise releasing
the action (reset state) and propagating the event.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1170
Nothing should ever disable an actor modifier (e.g. effect) during the
paint sequence, nor should any actor be set or unset on it. If this
would happen, log warnings so that it can be tracked down.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1166
Picking now only happens on allocated actors, but the
callback in the actor-pick test is not waiting for the
stage to run an allocation cycle. Ideally, we'd wait
for this cycle, but for now, forcing an allocation works
as well.
Allocate the overlay actor in the actor-pick test.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1169
When selecting the pick regions for an actor we were not considering
whether the actor was allocated and that was causing issues where the
preferred width/height of the actor was used when deciding whether
the actor should be considered as a pick target.
Check if the actor has a valid allocation, in addition to being mapped
and being in pick mode, in clutter_actor_should_pick_paint().
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1169
Normally we bail out in `sync_actor_geometry()`. The comment there
states:
```
Normally we want freezing a window to also freeze its position; this allows
windows to atomically move and resize together, either under app control,
or because the user is resizing from the left/top. But on initial placement
we need to assign a position, since immediately after the window
is shown, the map effect will go into effect and prevent further geometry
updates.
```
The signal for the initial sync originates in `MetaWindow` though and predates
`xdg_toplevel_set_maximized`, which again calls `meta_window_force_placement`,
triggering the signal too early. As a result, Wayland clients that start up
maximized have a wrong map animation, starting in the top-left corner.
In order to fix this without changing big parts of the geometry logic and risking
regressions, force the initial sync again before mapping.
Solution suggested by Jonas Ådahl.
https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1164
2020-03-30 15:59:48 +00:00
1188 changed files with 105932 additions and 92685 deletions
print("\nIssue the following command in your local tree to apply the suggested changes (needs uncrustify installed):\n\n $ git rebase origin/main --exec \"./check-style.py -r\"\n")
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
