It is possible to interpret the ammount of padding provided to the
*_set_tablet_area functions in two different and incompatible ways. The X11
backend effectively treats them as being input-centric (i.e., the padding
defines the size of the "dead zone" on the tablet) while the native backend
has an output-centric viewpoint (i.e., the padding defines the size of the
"dead zone" on the display) viewpoint. This difference in opinion causes the
cursor offset to change when switching between Xorg and a Wayland sessions.
The calibration utility within g-c-c does its calculations with an input-
centric viewpoint, so this patch modifies the native backend to work
correctly with these values. To change viewpoints, we can simply invert
the scale and negate the offset. It should be noted that this function
also forgot to apply scaling to the offsets (as required by the matrix
transform done by libinput) which would have further compounded the
cursor offset issue under Wayland.
https://bugzilla.gnome.org/show_bug.cgi?id=784009
It would only allow to alternate between the logical monitors, we actually
want to return NULL here so it can cycle to the whole span of monitors.
https://bugzilla.gnome.org/show_bug.cgi?id=782032
Instead of checking all MetaMonitors in the monitor manager, we want to
look (as the function name says) in the MetaMonitors contained in the
given logical monitor.
Otherwise, it will return TRUE for every logical monitor, given we are
querying for an existing EDID.
https://bugzilla.gnome.org/show_bug.cgi?id=782032
Due to the pen/eraser device separation in X11, CLUTTER_TABLET_DEVICE does
not apply there, this device type is only used in native/evdev. Checking
for CLUTTER_PEN/ERASER_DEVICE makes the left-handed mode correctly applied
on tablets.
https://bugzilla.gnome.org/show_bug.cgi?id=782027
Previously, the function only returned `TRUE` if the given surface was
equal to the given pointer's focused surface. This changes the behaviour
to also return `TRUE` if any of the given surface's subsurfaces are
equal to the pointer's focused surface.
https://bugzilla.gnome.org/show_bug.cgi?id=781811
Use the "destroy" MetaWaylandSurface signal instead of the wl_resource
destroy signal for tracking the lifetime of the surface with pointer
focus.
As unsetting the focus may have side effects due to handlers of the
"focus-surface-changed" signal, connect the signal after the default
handler to make sure other clean up facilities have the chance deal with
the surface destruction before we try to unset the focus.
https://bugzilla.gnome.org/show_bug.cgi?id=783113
Disable-while-typing disables the touchpad while the user is typing.
This patch introduces the necessary backend code to implement the
org.gnome.desktop.peripherals.touchpad.disable-while-typing setting of
gsettings-desktop-schemas which was implemented in commit
4c5b1c1df399d6afaaccb237e299ccd1d5d29ddd and released as part of 3.24.
This is known as dwt in libinput.
This patch has been tested on X11 and Wayland.
https://bugzilla.gnome.org/show_bug.cgi?id=764852
When terminating mutter running as a display server, don't try to resize
maximized windows when unmanaging, as at this point, they will have no
MetaWaylandSurface. Originally this was done instead of setting the
net_wm_state to not mess with future window managers, but when we're a
Wayland compositor, this does not matter.
https://bugzilla.gnome.org/show_bug.cgi?id=782156
If a client changes the state of a surface to issue a set_maximize, this
causes apply_pending_state() to be called before mutter has placed the
window.
If the monitor on which the window is to be shown initially is different
from the one where the pointer is placed, this causes the effect to be
played at the wrong location before the window eventually reaches its
location on another monitor.
Force the window to be placed prior to change its state to maximized in
xdg-shell so that mutter won't relocate the window afterwards.
This also avoids sending an xdg_toplevel.configure with a size of 0x0
which would cause the client to initially draw its surface with some
arbitrary size.
https://bugzilla.gnome.org/show_bug.cgi?id=782183https://bugzilla.gnome.org/show_bug.cgi?id=781353
Fixes cogl_texture_get_data() resorting to the wrong conversions when
extracting the texture data. This notably resulted in RGB/RGBA buffers
copied as-is into BGRA buffers, for instance for the fullscreen animation,
or single-window screenshots of such buffers.
https://bugzilla.gnome.org/show_bug.cgi?id=779234
Previously we would bail out early in xdg_toplevel_role_commit() if no
geometry change was set, ignoring the possible min/max size hints
changes.
But setting a min/max size hint without changing the geometry is
perfectly valid, so we ought to apply the min/max changes regardless of
a geometry change.
https://bugzilla.gnome.org/show_bug.cgi?id=782213
If the client doesn't set a geometry using xdg_shell, we'll compute its
geometry based on its surface and subsurfaces.
Yet, we translate that as a window (re)size only when there is a pending
geometry, that we don't have when we computed the geometry by ourself.
Make sure we set the pending new geometry flag when computing the
geometry when it actually changed.
https://bugzilla.gnome.org/show_bug.cgi?id=782213
We will both create and destroy monitors during initialization (when
using the X11 backend), so don't try to access the monitor manager from
the backend, but store a pointer to it instead.
It's stored in MetaMonitor even though only MetaMonitorTiled uses it,
mostly because it makes more sense to store such a pointer there.
https://bugzilla.gnome.org/show_bug.cgi?id=781723
In some circumstances, the origin tile (0, 0) is not the one that
should be used to drive the monitor when using a non-tiled mode. Update
MetaMonitorTiled to support this case. It also seems to be so that the
preferred mode might be some low resolution or bogus mode on these
monitors, so also adapt MetaMonitorTiled to manage to ignore the
preferred mode of a tiled monitor if the preferred mode doesn't use
both tiles.
https://bugzilla.gnome.org/show_bug.cgi?id=781723
Only support suggested monitor positioning if the monitor is non-tiled.
Normally this functionality is used by virtual machines to provide a
hint of how to place the virtual monitors, and they don't tend to use
tiled monitors anyway.
https://bugzilla.gnome.org/show_bug.cgi?id=781723
Commit 0fd9e38175 fixed setting the out parameter for the x coordinate
when using the X11 backend, but broke compilation when the backend is
not available ...
Really fix the issue by running the X11-specific code when the X11
backend is available and in use, and display the one-time warning
otherwise.
https://bugzilla.gnome.org/show_bug.cgi?id=781902
If we translate between text/plain;charset-utf-8 from the wayland side to
UTF8_STRING on the X11 side, we want to continue all further X11 selection
requests using the same translated UTF8_STRING atom than we use in the
first XConvertSelection call.
https://bugzilla.gnome.org/show_bug.cgi?id=782472
Due to an accidental swap of an else statement and a preprocessor #else,
the output x coordinate is currently only set when not using the X11
windowing system, whoops.
https://bugzilla.gnome.org/show_bug.cgi?id=781902
A client can still commit state to a destroyed subsurface. It wont
update anything on the screen, since the subsurface will not be
visible, but mutter should still handle it and not crash.
https://bugzilla.gnome.org/show_bug.cgi?id=781391
In some cases the hardware cursor is invisible when Mutter is launched from the
TTY, due to drmModeSetCursor2 failing without a fallback being set.
This patch captures the return value of drmModeSetCursor2 and in case of an
error, enables the texture based fallback. It adds a `broken` state, that is
checked in should_have_hw_cursor() and
meta_cursor_renderer_native_realize_cursor_from_*() to avoid copying every
cursor into a gbm buffer when we know it will fail every single time.
https://bugzilla.gnome.org/show_bug.cgi?id=770020
Quick motions can come across as too fast (or slow) if it crosses outputs
with different scales. If this happens, rebuild the motion delta applying
the scale that applies to each logical monitor the pointer is crossing.
https://bugzilla.gnome.org/show_bug.cgi?id=778119
To allow for more natural pointer movements from relative pointer
devices (e.g. mouse, touchpad, tablet tool in relative mode, etc), scale
the relative motion from libinput with the scale of the monitor. In
effect, this means that the pointer movement is twice as fast (physical
movement vs numbers of pixels passed) as before, but it also means that
the same physical movement crosses the distance in a GUI no matter if
it is on a HiDPI monitor or not.
https://bugzilla.gnome.org/show_bug.cgi?id=778119
Clutter's evdev input backend has no support for setting double
click timeout set by gnome-settings-daemon. This results in
touchpad click events timing out on wayland, because the
default timeout value wasn't enough.
This patch moves timeout setting to mutter and removes X11
backend specific setting from clutter.
https://bugzilla.gnome.org/show_bug.cgi?id=771576
The code calculating the output scale involves calculations around pixel
and mm sizes, however we do compare post-transformation pixel sizes to
untransformed mm sizes, which breaks the DPI calculations. Fix this by
transforming back pixel sizes back to untransformed.
While we're at it, actually compare the output height to HIDPI_MIN_HEIGHT
instead of its width, it seems right according to the #define name and
comment.
https://bugzilla.gnome.org/show_bug.cgi?id=777687
When a state changed, e.g. a window went from unfullscreen to
fullscreen, always sync the window geometry, otherwise a compositor
application (e.g. gnome-shell) might end up with an unfinished window
state transition effect.
Without always syncing, the compositor plugin will see a 'size-change'
event, as a result of the state change, but if the size didn't change,
it would never see the 'size-changed' event. If an effect, for example
gnome-shell's fullscreen effect, is triggered on 'size-change' it might
rely on the actual size change to not get stuck. This commit allows it
to have this dependency.
This fixes a bug where a fullscreen effect gets "stuck" when a window
goes fullscreen without changing the window geometry.
https://bugzilla.gnome.org/show_bug.cgi?id=780292
The mitigation to avoid missing EDID blob was incorrect; the reason it
sometimes failed to read was a race between different applications all
trying to read the EDID at the same time. E.g. gnome-shell as GDM would
at the same time as the session gnome-shell try to read the EDID of the
same connector at the same time, triggering a race in the kernel,
making the blob reading ioctl occationally fail with ENOENT.
Remove this mitigation, as it didn't really mitigate anything; the race
could just as well happen when doing the actual read later.
https://bugzilla.gnome.org/show_bug.cgi?id=779837
When mutter is paused (i.e. not the DRM master), stop listening on
hotplug events. Instead read the current state and set modes when
resumed.
This avoids a race condition in the drm API which currently only
manages to properly deal with one application querying the EDID state
at the same time when there are multiple mutter instances running at
the same time (e.g. gnome-shell driving gdm at the same time as
gnome-shell as the session instance).
https://bugzilla.gnome.org/show_bug.cgi?id=779837
If the dnd window ends up lower in the overall stack than the window
it's supposed to fence, the drop might end up in some other window
underneath the expected target window.
Maps and raises the dnd window each time it's shown so that it's always
placed above.
Bugzilla: https://bugzilla.gnome.org/show_bug.cgi?id=779800
A MetaOutput is a connector, not exactly a monitor or a region on the
stage; for example tiled monitors are split up into multiple outputs,
and for what is used in input settings, that makes no sense. Change
this to use logical monitors instead of outputs.
https://bugzilla.gnome.org/show_bug.cgi?id=779745
When no output was specified, the screen limit was used to calculate the
aspect ratio. The screen limit, however, is either just an arbitrary
number if no screen limit is applicable, or a hardware graphics buffer
limit, which has nothing to do with anything actually displayed. Change
it to use the screen size instead, to get something that makes more
sense when no output is found.
https://bugzilla.gnome.org/show_bug.cgi?id=779745
Expose via a new API whether the transform on a logical monitor is
handled by the backend. This was previously only exposed only in the
native backend. This will be used to emulate not supporting transforms
in the backend in the nested backend.
https://bugzilla.gnome.org/show_bug.cgi?id=779745
Previously, the size of the logical monitor was derived directly from
the tiling information. This works fine until we add transformations,
or set modes with a dimension different from the resulting resolution
when tiled. Fix this by traversing the assigned CRTC rects, as these
are already transformed by the configuration system.
https://bugzilla.gnome.org/show_bug.cgi?id=779745
By setting the environment variable MUTTER_DEBUG_TILED_DUMMY_MONITORS
to "1", the dummy MetaMonitorManager backend used when running mutter
nested will create tiled monitors instead of single-output/CRTC
monitors. This makes it possible to test tiled monitor configuration.
https://bugzilla.gnome.org/show_bug.cgi?id=779745
Refactor monitor generation by splitting the generation of modes, CRTCs
and outputs into a separate function. A side effect is that each output
will have its own set of possible modes.
https://bugzilla.gnome.org/show_bug.cgi?id=779745
This will be used to invert the transform in the nested mode, making it
possible to test offscreen texture based transform using the nested
backend.
https://bugzilla.gnome.org/show_bug.cgi?id=779745
Split up logical monitor cration into derived (when derived from
current underlying configuration) and non-derived (when creating from a
logical monitor configuration). This avoids that type of logic in the
logical monitor creation function.
https://bugzilla.gnome.org/show_bug.cgi?id=779745
Add support for non-tiled monitor modes on tiled monitors. This is done
by adding all the other supported modes, except the modes with the
same resolution as the tile dimensions.
https://bugzilla.gnome.org/show_bug.cgi?id=779745
When adding a monitor and all its outputs, don't try to set the logical
monitor of the outputs CRTC if none was assigned. This might happen if
a tiled monitor only uses a subset of the connectors it are connected
via.
https://bugzilla.gnome.org/show_bug.cgi?id=779745
Don't set the CRTC rect and screen size at in read_current(), as those
depends on how the configuration is done. Instead, don't set the CRTC
rect at all, and update the screen dimensions when being configured.
https://bugzilla.gnome.org/show_bug.cgi?id=779745
QT apps reject DnD if the timestamp received in the SelectionRequest
event isn't the same it gave in XdndPosition/Drop client messages.
Bookkeeping and using it in XConvertSelection makes it happy again.
https://bugzilla.gnome.org/show_bug.cgi?id=779757
We are keeping accounting of the focus window as seen by the DnD bridge
right here, so use it instead of the MetaWaylandDragGrab focus as it may
lag behind the real focus (i.e. till the drag source notices the window
and sends XdndEnter to it).
This leads to the window trying to be repositioned more often than
necessary when the drag source takes long to send the XdndEnter client
message, and maybe not repositioned at all if the pointer leaves the
surface while no XdndEnter message was received.
https://bugzilla.gnome.org/show_bug.cgi?id=763246
We currently wait for the selection being cleared by the drag source,
which might not happen on not quite educated clients. This may leave
a stuck XDND grab in the compositor side.
We can actually do a bit better, and clear the grab if:
1) The drag source sent XdndDrop to the wayland drag destination.
2) There's no accepting drag destination and all pointer buttons are
released.
3) As usual, whenever the drag source clears the selection data
https://bugzilla.gnome.org/show_bug.cgi?id=763246
We already do have a texture with an internal format in these paths,
so we should check the required format according to it.
This fixes CoglAtlasTexture (and CoglPangoRenderer indirectly), as
it forces a RGBA format on its texture, but pixel_format_to_gl()
anyway assumed swizzling is performed on the texture, while it is
not the case.
https://bugzilla.gnome.org/show_bug.cgi?id=779234
This is used by the GL driver in order to determine whether swizzling
actually applies given the bitmap and target texture internal format.
If both agree that they store BGRA, then swizzling may apply.
https://bugzilla.gnome.org/show_bug.cgi?id=779234
No XDnD events which notify DnD status change comes in Wayland. To emulate XDnD
behavior, MetaDnd checks whether there is a grab or not when the modal window
starts showing. When there is a grab, it processes the raw events from
compositor, and emits DnD signals for plugin.
https://bugzilla.gnome.org/show_bug.cgi?id=765003
Implement MetaDnd for emitting DnD signals to plugins such as gnome-shell. The
xdnd handling code comes from gnome-shell, and it is hidden behind MetaDnd now.
https://bugzilla.gnome.org/show_bug.cgi?id=765003
When running nested, the pointer can be outside of the stage, meaning
outside of any logical monitor. Handle this when getting the current
logical monitor by falling back to the first logical monitor when the
pointer coordinate is outside of any logical monitor.
https://bugzilla.gnome.org/show_bug.cgi?id=779001
Whenever an EGLOutput consumer is temporary unable to handle
eglStreamConsumerAcquire() operations (e.g. during a VT-switch),
an EGL_RESOURCE_BUSY_EXT error is generated.
This change adds the appropriate error handling to flip_egl_stream() in
order to recover from such errors.
https://bugzilla.gnome.org/show_bug.cgi?id=779112
This change adds descriptions for the following errors to
get_egl_error_str():
- EGL_BAD_STREAM_KHR
- EGL_BAD_STATE_KHR
- EGL_BAD_DEVICE_EXT
- EGL_BAD_OUTPUT_LAYER_EXT
https://bugzilla.gnome.org/show_bug.cgi?id=779112
Ungrabbed pointer motion events over a client window area don't even
reach mutter in X compositor mode, but as a wayland compositor we
process those events which ends up in a call stack like:
- meta_window_handle_ungrabbed_event
- meta_ui_frame_handle_event
- handle_motion_notify_event
- get_control
- meta_ui_frame_calc_geometry
Computing frame geometry is a relatively CPU expensive operation and
doing it on every motion event over a client window is pointless work
since we aren't going to change the cursor or prelight any frame
widget.
This commit special cases the determination of
META_FRAME_CONTROL_CLIENT_AREA using a much faster method. When
continuously moving the pointer over an X (client) window, it results
in a ~40% decrease in mutter cpu usage.
https://bugzilla.gnome.org/show_bug.cgi?id=779436
If the GL implementation/hw supports the GL_*_texture_swizzle extension,
pretend that BGRA textures shall contain RGBA data, and let the flipping
happen when the texture will be used in the rendering pipeline.
This avoids rather expensive format conversions when forcing BGRA buffers
into RGBA textures, which happens rather often with WL_SHM_FORMAT_ARGB8888
buffers (like gtk+ uses) in little-endian machines.
In intel/mesa/wayland, the performance improvement is rather noticeable,
CPU% as seen by top decreases from 45-50% to 25-30% when running
gtk+/tests/scrolling-performance with a cairo renderer.
https://bugzilla.gnome.org/show_bug.cgi?id=779234
Set up things so that if the INTEL_swap_event extension is not present,
but the driver is known to have good thread support, we use an extra
thread and call glXWaitVideoSync() in the thread. This allows idles
to work properly, even when Mutter is constantly redrawing new frames;
otherwise, without INTEL_swap_event, we'll just block in glXSwapBuffers().
https://bugzilla.gnome.org/show_bug.cgi?id=779039
Because the threaded-swap-wait functionality requires XInitThreads(),
and because it isn't clear that it is a win for all applications,
add a API function to conditionally enable it.
Fix the cogl-crate example not to just have a hard-coded dependency
on libX11.
https://bugzilla.gnome.org/show_bug.cgi?id=779039
It's a good guess that the buffer swap will occur at the next vblank,
so use glXWaitVideoSync in a separate thread to deliver a sync event
rather than just letting the client block when frame drawing, which
can signficantly change app logic as compared to the INTEL_swap_event
case.
https://bugzilla.gnome.org/show_bug.cgi?id=779039
When we don't have GLX_OML_sync_control, we still can set the
frame presentation time, but we always use the system monotonic time,
so return that from get_clock_time().
https://bugzilla.gnome.org/show_bug.cgi?id=779039
As previously commented in the code, SGI_video_sync is per-display, rather
than per-renderer. The is_direct flag for the renderer was tested before
it was initialized (per-display) and that resulted in SGI_video_sync
never being used.
https://bugzilla.gnome.org/show_bug.cgi?id=779039
Commits 6dbec6f8, 734402e1 and f041b35b introduced memory leaks by
switching to returning copies instead of the original buffers but
forgetting to free those original buffers.
Some error cases were also not freeing the ->prop buffer as they
should.
https://bugzilla.gnome.org/show_bug.cgi?id=642652
Wayland windows can be zero sized until clients attach a buffer, but
our rectangle code doesn't deal with this case well, in particular,
meta_screen_get_monitor_for_rect() might end up choosing the wrong
monitor for a zero sized rectangle since
meta_rectangle_contains_rect() considers a zero sized rectangle at the
right or bottom edges of another rectangle (the monitor's) to be
contained there.
Since out size limits constraint will enforce a minimum size of 1x1,
we might as well enforce that when setting up the constraint info so
that the correct monitor gets chosen and the single monitor constraint
doesn't move these windows to the wrong one.
https://bugzilla.gnome.org/show_bug.cgi?id=772525
Wayland windows are initially zero sized until clients commit the
first buffer. Despite being invisible, clients are allowed to request
such windows to be fullscreened on a specific output before they
attach the first buffer which means we need to be able to move them.
meta_window_move_to_monitor() doesn't handle this case because these
windows' initial monitor is a placeholder since their initial
coordinates are 0,0+0+0, which results in us using a rectangle as
old_area for meta_window_move_between_rects() that might be to the "right"
of the window causing the move to go further out of the visible
screen's coordinates. This is later "corrected" by the constraints
system but the window might end up in the wrong monitor.
To fix this, we can make meta_window_move_between_rects() accept a
NULL old_area, meaning that we move the window to the new_area without
trying to keep a relative position.
https://bugzilla.gnome.org/show_bug.cgi?id=772525
The reason for the display to be closed may be meta_screen_new()
returning NULL, in which case we don't have a screen to free.
Avoid a segfault on exit by adding a proper check.
https://bugzilla.gnome.org/show_bug.cgi?id=778831
We currently don't have any shadow class for combo box popups,
which means the default shadow of normal windows is used. That's
clearly odd given that the two are very different, and isn't
consistent with GTK+-3's client-side shadows for popups. While
we could add a dedicated shadow class, the designers are fine
with reusing the existing shadow for dropdown-menus, so let's
do that.
https://bugzilla.gnome.org/show_bug.cgi?id=744667
Split up the X11 backend into two parts, one for running as a
Compositing Manager, and one for running as a nested Wayland
compositor.
This commit also cleans up the compositor configuration calculation,
attempting to make it more approachable.
https://bugzilla.gnome.org/show_bug.cgi?id=777800
In order to minimize the amount of breakage, while at the same time
making it easier to make backward incompatible changes needed to
continue turning libmutter into a capable Wayland compositor, make the
libmutter and friends (libmutter-clutter, libmutter-cogl*) parallel
installable by adding a version number to the name. This changes
various filenames, for example what previously was libmutter.so is now
libmutter-0.so (assuming the version for now is 0), and
libmutter-clutter-1.0.so is now libmutter-clutter-0.so. The pkg-config
filenames and GObject introspection has been renamed to reflect this as
well.
This enables a downstream compositor rely on a specific version of the
libmutter API, while gracefully handling API/ABI changes by having to
update to the new version at their own pace.
https://bugzilla.gnome.org/show_bug.cgi?id=777317
This signal provides the necessary information to let gnome-shell trigger
updates of pad leds/oleds whenever a pad group switches mode, and the
actions associated to buttons do too.
https://bugzilla.gnome.org/show_bug.cgi?id=776543
And add specific private methods to notify about tablet mapping and mode
switches. The signal allows the mutter side to trigger OSDs in a generic
way.
https://bugzilla.gnome.org/show_bug.cgi?id=771098
As all the relevant backends are expected to provide
ClutterPadButtonEvents, it makes no sense to split the information,
plus all other event fields are now available and might be needed
in the future.
https://bugzilla.gnome.org/show_bug.cgi?id=771098
Using ClutterInputDeviceEvdev::output-aspect-ratio. This only applies
to devices which are not calibratable, so again we need to implement
this at the toolkit level.
https://bugzilla.gnome.org/show_bug.cgi?id=774115
We couldn't properly merge output-mapping matrix and calibration into
one. Now that libinput calibration matrix is free to use, we can
actually implement tablet calibration with it.
https://bugzilla.gnome.org/show_bug.cgi?id=774115
When != 0, this property will express the W:H ratio of the desired
output area (be it one monitor or the span of all monitors). At
the time of translating coordinates, coordinates will be skewed so
that the input area of the tablet is a rectangle of the given ratio.
Events happening in the input areas that fall outside the output
aspect ratio will be clamped to the nearest coordinate in the rect.
If the ratio is 0, the whole input area of the tablet will be used
and no coordinate skewing will happen.
https://bugzilla.gnome.org/show_bug.cgi?id=774115
And transform absolute events using this matrix. This property is
akin to the "Coordinate Transformation Matrix" property in X11,
and will be used to map tablets/touchscreens to outputs, favoured
over the libinput matrix which is meant for calibration.
https://bugzilla.gnome.org/show_bug.cgi?id=774115
Function "handle_raise_or_lower (src/core/keybindings.c)" is called
when running 'raise-or-lower' on a window. This function iterates
through all the windows in the stack to determine if our window is
already on top or obscured. The problem is that the window stack
includes windows in another workspaces and also windows that are
minimized.
https://bugzilla.gnome.org/show_bug.cgi?id=705200
Different libEGL will do different things for eglGetDisplay since it has
to guess what kind of display it's been handed. Better to just use the
API that makes it explicit.
Signed-off-by: Adam Jackson <ajax@redhat.com>
https://bugzilla.gnome.org/show_bug.cgi?id=772422
The initial state of the hardware cursor is not known, so always force
update it the first time we update the cursor. Do this by changing the
'force' flag of update_hw_cursor() to an 'invalidated' hw cursor state.
https://bugzilla.gnome.org/show_bug.cgi?id=771056
Clutter assumed seat0 which is most usually, but not always correct.
Add an evdev-backend specific function to allow passing the seat
that will be used for ClutterDeviceManager construction, which we
already obtain in MetaLauncher.
https://bugzilla.gnome.org/show_bug.cgi?id=778092
If the meta_window_actor_effect_completed() triggers inconsistent
accounting, there's also high chances that the thaw call will be
unexpected at this time too, which will lead to a g_error().
This makes mutter more lenient to effect_completed() calls of the
right type (i.e. those triggering freeze/thaw) being performed more
times than necessary in the upper parts. A warning will be issued,
but the process won't abort.
https://bugzilla.gnome.org/show_bug.cgi?id=777691
Since both the libinput event source and the key repeat timer have the
same priority, the order in which both handlers are called is
arbitrary if both sources are ready on the same poll return. This
means that sometimes we generate key repeats when there's already a
real key event queued on libinput that would cancel the repeat timer
if only it was processed before.
One solution would be lowering the repeat timer source priority a
notch lower than the libinput source but that would mean that a steady
stream of events from libinput (e.g. pointer device motion) would
prevent any key repeats to happen.
Instead, we can fix this problem by trying to dispatch libinput from
the key repeat timer and checking if the timer source has been
destroyed before generating more key repeats.
https://bugzilla.gnome.org/show_bug.cgi?id=774989
Commit 9214d5029c changed the notify*
API to use microseconds and we already have a microsecond time source
here so we can use the timestamp directly without losing resolution at
this layer.
https://bugzilla.gnome.org/show_bug.cgi?id=774989
Make sure that each logical monitor owns the expected actual monitors.
This currently needs special care when dealing with laptop lid the
configuration, as the MetaMonitorConfigManager path still deosn't
handle restoring the previous configuration, meaning the logical
monitor with the external monitor will continue being primary.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
This is partly redundant given that the current monitor mode is checked
against the CRTC mode, but this also checks the disabled CRTCs. Later
the configured mode position and transform will be checked.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Check that the crtc mode has the correct intra-monitor position. In
effect, this tests that the CRTCs in a tiled monitor are configured with
the correct mode on the correct position.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
The MetaOutput::is_primary state was not correctly managed in two cases:
* for tiled monitors, the primary state got overridden when setting
the preferred resolution
* for laptop lid, it was not set if the laptop panel happened to be
the first output
https://bugzilla.gnome.org/show_bug.cgi?id=777732
The main output of the main (first) monitor of a primary logical
monitor should have the MetaOutput::is_primary field set; all other
outputs should have it not set.
All outputs associated with a logical monitor with presentation set
should have MetaOutput::is_presentation set. No other outputs should
have it set.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Add a flag to the test setup creation function enabling the caller to
specify whether a stored config should be used. This is done by
changing the value of the hotplug_mode_update MetaOutput field,
normally used by VMs to do the same.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Both the monitor unit tests and monitor store unit tests will want to
check whether the config manager is used and set custom configuration
files.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
MetaMonitorConfigStore provides an XML storage mechanism for
MetaMonitorConfigManager. It stores configuration files defined in the
same level as the MetaMonitorsConfig format, i.e. refers to high level
"monitors" and "monitor modes" instead of connectors and CRTCs.
Only reading custom files are implemented and so far unused.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Don't let a dummy option context consume the arguments; just let the
GLib test suite do it. It'll handle the basic command line arguments
and allow doing things such as specifying what test to run.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
This tests only the monitor configuration and basic functionality. It
does not test anything related to window management and Wayland client
interaction.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Handle headless setup gracefully by having no logical monitors. This
commit only makes the monitor management code deal with it; other areas
may still not be able to handle it.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Use the g_assert_cmp(int|uint|...) macros when comparing integers and
unsigned integers. This means that the mismatched numbers are printed
in the test report.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Add a meta_monitors_config_new() helper. It's exposed outside of
meta-monitor-config-manager.c already, as it'll be used externally in a
later commit.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Add a test case that checks that configuration works when the lid is
initialy closed then later opened. This test case is disabled when the
legacy configuration is used as it does not handle that situation.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Handle configuring when the laptop lid is closed. This is so far
handled by creating a linear configuration while ignoring the laptop
panel. Changing the current configuration will come later.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Move the UpClients notify::lid-is-closed signal handling into
MetaMonitorManager, and put the getter behind a vfunc. This means
Placing it behind a vfunc allows custom backends to implement it
differently; for example the test backend can mock the state.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Adds an API to get the position suggested by the backend. This
translates to position advertised by some VM:s, used to hint at a
position making the position more natural (i.e. placed similarly to how
it may be placed on the host desktop).
https://bugzilla.gnome.org/show_bug.cgi?id=777732
The new monitor configuration system (MetaMonitorConfigManager) aims to
replace the current MetaMonitorConfig. The main difference between the
two is that MetaMonitorConfigManager works with higher level input
(MetaMonitor, MetaMonitorMode) instead of directly looking at the CRTC
and connector state. It still produces CRTC and connector configuration
later applied by the respective backends.
Other difference the new system aims to introduce is that the
configuration system doesn't manipulate the monitor manager state; that
responsibility is left for the monitor manager to handle (it only
manages configuration and creates CRTC/connector assignments, it
doesn't apply anything).
The new configuration system allows backends to not rely on deriving the
current configuration from the CRTC/connector state, as this may no longer be
possible (i.e. when using KMS and multiple framebuffers).
The MetaMonitorConfigManager system is so far disabled by default, as
it does not yet have all the features of the old system, but eventually
it will replace MetaMonitorConfig which will at that point be removed.
This will make it possible to remove old hacks introduced due to
limitations in the old system.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Operate on MetaMonitor's instead of MetaOutput's, as the latter may be
only a subset of an actual "monitor" when referring to the physical
computer equipment.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
When a logical monitor constains monitors with different subpixel
ordering, make the wl_output have the subpixel order 'unknown' so that
clients don't make assumptions given only a subset of the monitors of
the given region.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Sometimes we hit a race on hot-plug where we try to read the KMS
resources and the EDID blob is not yet ready. This would normally
result in a ENOENT when retrieving the blob. Handle this by retrying
after 50 milliseconds after a hot-plug event. Do this up to 10 times,
and after that give up trying to get the EDID blob and continue with
best effort.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
The function meta_monitor_manager_read_current_config() was renamed to
meta_monitor_manager_read_current_state() as it does not read any
configuration, but reads the current state as described by the backend.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
In preparation of replacing the configuration system with one working
with high level monitors instead of low level outputs etc, move
configuarion handling code into obviously named function (containing
the word 'legacy'.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
A monitor spec object is meant to be used to identify a certain monitor
on a certain output. The spec is unique per actual monitor and connector,
meaning that a monitor that changes from one connector from another
(e.g. HDMI1 to HDMI2) will not be identified as the same. It is meant
to associate for example a configuration entry with an actual monitor.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Add a "mode spec" concept, meaning to be used as a identifier for an
actual monitor mode. It consists of details making a mode unique, i.e.
the total resolution and refresh rate. This will later be used to get
the actual monitor mode (set of one or more CRTC modes).
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Add "monitor modes" abstracting the modes set on a monitor. On normal
monitors, this directly maps to the CRTC modes, but on tiled monitors,
a monitor mode can consist modes per tiled output.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Don't try to mirror the physical dimension, since that's a property of
one of the monitors, not of the logical monitor. Callers are changed to
deal with choosing the monitor to represent the logical monitor.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Don't deal with adding/removing tiled Xrandr monitors in the generic
backend, but leave it to the Xrandr backend. The tiled monitor will
itself notify the backend when such a monitor is added and removed.
Tiled Xrandr monitors are now based no MetaMonitor instead of
MetaLogicalMonitor. This means that mirrored tiled monitors will now be
represented correctly.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Instead of using crtcs and outputs to generate logical monitors, use
the ready made monitor abstraction that hides irrelevant things such as
monitor tiling etc.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Generate a set of "monitors" abstracting the physical concepts. Each
monitor is built up of one or more outputs; multiple outputs being
tiled monitors. Logical monitors will later be built from these.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
The MetaMonitorMode referred to the mode of a CRTC, and with the future
introduction of a MetaMonitor, theh old name would be confusing.
Instead call it what it is.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Set up the expected result in a declarative way in the same place as
the test case setup is declared. This way we have a completely
declarative way to create test cases.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Add private API for overriding the compositor configuration, i.e. the
compositor type (X11 WM or Wayland compositor) and backend type. This
will make it possible to add a special test backend used by src/tests/.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Put the monitor xinerama index in a separate struct that is attached to
the logical monitor using g_object_set/get_qdata(). Eventually this
should be moved to some "X11 window manager" object, but lets keep it
in MetaScreen until we have such a thing.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Don't store logical monitor specific state in an array where the index
from the monitor manager is used as index locally. Instead just use
table associating a logical monitor with a monitor specific state
holder, and store the state in there. This way we don't have the
workspace implementation relying on implementation details of other
units.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Instead of storing the logical monitors in an array and having users
either look up them in the array given an index or iterate using
indices, put it in a GList, and use GList iterators when iterating and
alternative API where array indices were previously used.
This allows for more liberty regarding the type of the logical monitor.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Refactor the tiled monitor assembly code (that constructs a logical
monitor out of tiling information. Part of the reason is to move away
from array based storage, part is to make the code easier to follow,
and part is to separate logical monitor construction from list
manipulation.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Rewrite check_fullscreen_func to not use indexes (and
offset-index-as-pointer) tricks. This also removes the usage of an API
constructing temporary logical monitor arrays carrying indices.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Change meta_monitor_manager_get_logical_monitor_at() to use floats,
replace users of meta_monitor_manager_get_monitor_at_point() to use the
API that returns a logical monitor and remove the now unused function.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Move the last piece of monitor grid getter API to the monitor manager
away from MetaScreen. The public facing API are still there, but are
thin wrappers around the MetaMonitorManager API.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
The method used for getting the current logical monitor (the monitor
where the pointer cursor is currently at) depends on the backend type,
so move that logic to the corresponding backends.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Turning a rectangle into a logical monitor also has nothing to do with
the screen (MetaScreen) so move it to MetaMonitorManager which has that
information.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Let the backend initialize the cursor tracker, and change all call
sites to get the cursor tracker from the backend instead of from the
screen. It wasn't associated with the screen anyway, so the API was
missleading.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
Instead of keeping around array indexes, keep track of them by storing
a pointer instead. This also changes from using an array (imitating the
X11 behaviour) to more explicit storing.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
To complement the current API which takes an index referencing a
logical monitor in the logical monitor array, add API that takes a
direct reference to the logical monitor itself. The intention is to
replace the usage of the index based API with one that doesn't rely on
internal implementation details.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
This is the current equivalent of looking up the logical monitor in the
logical monitor array using the number, but eventually that will be
deprecated, and before that done differently, so add a temporary helper
for the places that has not been ported yet.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
It was just pointer to the actual list; having to synchronize a list of
logical monitors with the actual monitors managed by the backend is
unnecessary.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
The fullscreen monitors state is set given a set of xinerama monitor
identification numbers. When the monitor configuration changes (e.g. by
a hotplug event) these are no longer valid, and may point to
uninitialized or unallocated data. Avoid accessing
uninitialized/unallocated memory by clearing the fullscreen monitor
state when the monitor configuration changes.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
It checks whether a surface is on a given "logical monitor", not
output. Output here is the Wayland name for the same thing, but should
not be confused with MetaOutput.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
In preparation for further refactorizations, rename the MetaMonitorInfo
struct to MetaLogicalMonitor. Eventually, part of MetaLogicalMonitor
will be split into a MetaMonitor type.
https://bugzilla.gnome.org/show_bug.cgi?id=777732
src/backends/meta-egl.c: In function ‘set_egl_error’:
src/backends/meta-egl.c:144:16: error: format not a string literal and no format arguments [-Werror=format-security]
error_str);
^~~~~~~~~
https://bugzilla.gnome.org/show_bug.cgi?id=777389
Use the proposed EGL_WL_wayland_eglstream EGL extension instead of the
file descriptor hack that was used as a temporary solution.
Note that this results in EGL clients will no longer work if they are
running on a Nvidia driver with a version older than 370.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
Instead of having a way to determine the type of a buffer, add a
realization step that implicitly detects the buffer type. This makes it
possible to both realize (i.e. creating needed objects from the buffer)
and determine the type at the same time, which may be the only possible
way (for example, the only way to know whether a buffer is a EGLStream
is to create the EGLStream from it).
https://bugzilla.gnome.org/show_bug.cgi?id=773629
We need a GLES2 extension macro in cogl-texture-2d-gl.c, but we can't
include GLES2/gl2ext.h because it will conflict with things in
GL/glext.h. We can't rely on cogl including anything for us since it'd
only include GLES2/gl2ext.h if OpenGL support was explicitly disabled.
https://bugzilla.gnome.org/show_bug.cgi?id=774891
When the monitor the surface is on has a scale other than 1, the
coordinate of the window menu popup position needs to be scaled, as it
is reported in logical pixels, while the stage is still in physical
pixels.
https://bugzilla.gnome.org/show_bug.cgi?id=776055
A window manager must select the SubstructureRedirect mask on the root
window to receive the MapRequest from the X11 clients and manage the
windows. Without this event mask set, a window manager won't be able to
map any new window.
The Wayland selection code in mutter can change/clear the event mask on
the requestor window from a XSelectionRequest event when the window is
not managed by mutter/gnome-shell.
A rogue or simply buggy X11 client may send a XConvertSelection() on the
root window and mutter will happily change/clear its own event mask on
the root window, effectively turning itself into a regular X11 client
unable to map any new X11 window from the other X11 clients.
To avoid this, simply check that the requestor window is not the root
window prior to change/clear the event mask on that window.
https://bugzilla.gnome.org/show_bug.cgi?id=776128
Commit 5eb5f72 - wayland: Check surface outputs after mapped state
changes connected the ::mapped signal handler, we need to disconnect it
on destroy to avoid a possible assertion failure in
update_surface_output_state()
https://bugzilla.gnome.org/show_bug.cgi?id=776036
Commit 4295fdb892 made us skip focusing
all xdg_popups instead of just non-grabbing ones as intended. This
means that when unmanaging a window we might select a xdg_popup window
to focus (in meta_stack_get_default_focus_window() ) but then since we
don't actually focus it we go on unmanaging the focused window which
triggers an assertion, as it should.
To avoid this and still fixing bug 771694 we can make use of the
MetaWindow->input property for non-grabbing xdg_popup windows since
their semantics, in this regard, are the same as no input X11 windows.
This way, when unmanaging a focused window while a xdg_popup is up,
we'll either give focus to the xdg_popup or not select the popup at
all to be focused if it's non-grabbing.
https://bugzilla.gnome.org/show_bug.cgi?id=775986
We need to do swap notifications asynchronously from flip events since
these might be processed during swap buffers if we are waiting for the
previous frame's flip to continue with the current.
This means that we might have more than one swap notification queued
to be delivered when the idle handler runs. In that case we must
deliver all notifications for which we've already seen a flip event.
Failing to do so means that if a new frame, that only swaps buffers on
such a swap notification backlogged Onscreen, is started, when later
we get its flip event, we'd notify only an old frame which would hit
this MetaStageNative's frame_cb() early exit:
if (global_frame_counter <= presented_frame_counter)
return;
and we'd never finish the new frame and thus clutter's master clock
would be waiting forever stuck.
https://bugzilla.gnome.org/show_bug.cgi?id=774557
EGLDevice requires a define from GLES2, even when GL is used instead.
As type definitions may conflict between the two, we shouldn't include
both at the same time. Instead, provide the missing define explicitly
when not using GLES2.
https://bugzilla.gnome.org/show_bug.cgi?id=774891
We currently only focus unfocused windows on button press if no
modifiers (or just ignored modifiers) are in effect. This behavior
seems surprising and counter-intuitive so let's do it for any modifier
combination instead.
https://bugzilla.gnome.org/show_bug.cgi?id=746642
There's no reason to keep this ~15 year old piece of code around as
well as the preference handling that would only make sense if this
hunk was actually enabled.
https://bugzilla.gnome.org/show_bug.cgi?id=746642
When flush-swap-notify is already queued, we might end up trying to
requeue it, for example when handling flip callbacks inside
swap-buffers. Actually queuing it there is harmless, since old frames
will be discarded anyway.
https://bugzilla.gnome.org/show_bug.cgi?id=774923
We might still end up in swap-buffer without the previous flip callback
having been invoked. This can happen if there are two monitors, and we
manage to draw before having all monitor flip callbacks invoked.
https://bugzilla.gnome.org/show_bug.cgi?id=774923
A window's unconstrained_rect is essentially just the target rectangle
we hand to meta_window_move_resize_internal() except it's not updated
until the window actually moves or resizes.
As such, for wayland client resizes, since they're async, using
window->unconstrained_rect right after calling move_resize_internal()
to update the grab anchor position on unmaximize doesn't work as it
does for X clients.
To fix this, we can just use the target rectangle for the grab
anchor. Note that comment here was already wrong since it says we
should be taking constraints into account and yet the code used the
unconstrained rect anyway.
https://bugzilla.gnome.org/show_bug.cgi?id=770345
This reverts commit 989ec7fc60.
We now rely on accurately knowing if a window moved and/or resized in
meta_window_move_resize_internal() so the wayland implementation can't
lie any longer.
https://bugzilla.gnome.org/show_bug.cgi?id=770345
In order for the compositor plugin to be able to animate window size
changes properly we need to let it know of the starting and final
window sizes.
For X clients this can be done synchronously and thus with a single
call into the compositor plugin since it's us (the window manager)
who's in charge of the final window size.
Wayland clients though, have the final say over their window size
since it's determined from the client allocated buffer.
This patch moves the meta_compositor_size_change_window() calls before
move_resize_internal() which lets the compositor plugin know the old
window size and freezes the MetaWindowActor.
Then we get rid of the META_MOVE_RESIZE_DONT_SYNC_COMPOSITOR flag
since it's not needed anymore as the window actor is frozen and that
means we can use meta_compositor_sync_window_geometry() as the point
where we inform the compositor plugin of the final window size.
https://bugzilla.gnome.org/show_bug.cgi?id=770345
This will be used to let plugins know when a previous size change
actually becomes effective. This is needed to handle wayland client
resizing properly since, unlike X, it's async.
https://bugzilla.gnome.org/show_bug.cgi?id=770345
Normally textures in OpenGL are inverted on the Y axis, and we only
apply our rotation transform when it is not. To make the common case
work as normal, default to assuming textures are Y inverted.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
This commit adds for a new type of buffer being attached to a Wayland
surface: buffers from an EGLStream. These buffers behave very
differently from regular Wayland buffers; instead of each buffer
reperesenting an actual frame, the same buffer is attached over and
over again, and EGL API is used to switch the content of the OpenGL
texture associated with the buffer attached. It more or less
side-tracks the Wayland buffer handling.
It is implemented by creating a MetaWaylandEglStream object, dealing
with the EGLStream state. The lifetime of the MetaWaylandEglStream is
tied to the texture object (CoglTexture), which is referenced-counted
and owned by both the actors and the MetaWaylandBuffer.
When the buffer is reattached and committed, the EGLStream is triggered
to switch the content of the associated texture to the new content.
This means that one cannot keep old texture content around without
copying, so any feature relying on that will effectively be broken.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
Add API to enable the caller to have a custom method for allocating an
external texture. This will enable the possibility for mutter to
generate a texture from for example an EGLStream without having to add
support for that in Cogl.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
Add support for inverted Y Wayland buffers. OpenGL textures are by
default inverted, so adding support for EGL_WAYLAND_Y_INVERTED_WL
effectively means adding support for non-inverted, which makes the
MetaShapedTexture apply a transformation when drawing only when querying
EGL_WAYLAND_Y_INVERTED_WL resulted in the response "EGL_FALSE".
https://bugzilla.gnome.org/show_bug.cgi?id=773629
Don't rely on the Cogl layer having Wayland specific paths by
determining the buffer type and creating the EGLImage ourself, while
using the newly exposed CoglTexture from EGLImage API. This changes the
API used by MetaWaylandSurface to make the MetaWaylandBuffer API be
aware when the buffer is being attached. For SHM and EGL buffers, only
the first time it is attached will result in a new texture being
allocated, but later for EGLStream's, more logic on every attach is
needed.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
This commit adds support for using a EGLDevice and EGLStreams for
rendering on top of KMS instead of gbm. It is disabled by default; to
enable it pass --enable-egl-device to configure.
By default gbm is first tried, and if it fails, the EGLDevice path is
tried. If both fails, mutter will terminate just as before.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
There is no way to pass any backend specific parameters to a
CoglFramebuffer until after it has been allocated by
cogl_framebuffer_allocate() (since this is where the winsys/platform
fields are initialized). This can make it hard to actually allocate
anything, if the platform depends on some backend specific data.
A proper solution would be to refactor the onscreens and framebuffers to
use a GObject based type system instead of the home baked Cogl one, but
that'll be left for another day. For now, allocate in two steps, one to
allocate the backend specific parts (MetaOnscreenNative), and one to
allocate the actual onscreen framebuffer (via
meta_onscreen_native_allocate()).
So far there is nothing that forces this separation, but in the future
there will, for example EGLDevice's need to know about the CRTC in
order to create the EGLSurface.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
A swap-buffers should never be issued when we are waiting for a flipped
callback, so instead of trying to handle a situation that sholud never
happen, warn instead.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
When a swap failed with EACCES (possibly due to VT switching), don't
mark the framebuffer as 'in use', so that it'll be cleaned up properly
and not set as current.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
For when there is no gbm available, for example when using
EGLDevice/EGLStream's, just fall back to the OpenGL texture based
cursor rendering path.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
Drivers may be bad at guessing what is passed to eglGetDisplay, ending
up return non-functioning EGLDisplay's. Using eglGetPlatformDisplay
avoids this issue.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
Lets use a pbuffer surface as a dummy surface instead of a gbm based
one, so that we don't need to rely on the availability of gbm to create
a dummy surface when there is no need for it.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
Separate gbm initialization from general renderer initialization. Do
this even though no other initialization is done for now; later there
will will be other types of rendering mode, initialized in their own
functions.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
Add proc symbol loading and helper functions for calling them, dealing
with errors etc. So far no extension symbols are loaded, only the
infrastructure is there.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
In another step getting rid of the duplications introduced by Cogl,
use the equivalent GLib types where Cogl types previously used. While
CoglBool is not a typedef to gboolean, they are both typedefs to int,
and we already use GLib's TRUE/FALSE to set them.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
Initialize the GError pointer used when creating the renderer. If an
error occurs, the error is expected to be NULL, otherwise it'll
misinterpreted as already set.
https://bugzilla.gnome.org/show_bug.cgi?id=773629
Similarly to Weston (where this code originated), there were two errors
in the X11 lockfile handling.
Firstly, after reading 11 characters from the lock file (which could
have been placed by any process), there was no guarantee of
NUL-termination, meaning strtol could've theoretically run off the end
of the string.
Secondly, whilst writing the new lock, the trailing NUL byte was not
correctly accounted for. The size passed as an input to snprintf takes
the maximum size of the string including the trailing NUL, whilst the
return (and the input to write) gives the actual size of the string
without the trailing NUL.
The code did attempt to check the return value, however snprintf returns
the size of the _potential_ string written, before snprintf culls it, so
this was off by one, and the LF was not being written.
Signed-off-by: Daniel Stone <daniels@collabora.com>
https://bugzilla.gnome.org/show_bug.cgi?id=774613
In that case, the MetaWindow is created, but it should also be unminimized
to satisfy the MapRequest triggered by the client, otherwise these would
stay minimized until they're shown explicitly by the user.
https://bugzilla.gnome.org/show_bug.cgi?id=774333
Some applications like Wine may choose to juggle the same user time
window across different toplevels, in that case we receive warnings
when trying to register the window a second time, leading to wrong
accounting.
If the window was already used as the user time window for another
toplevel, unset it from the previous MetaWindow owner, and unregister
so the registration with the new MetaWindow is successful.
https://bugzilla.gnome.org/show_bug.cgi?id=774330
This is implemented using Wacom-driver specific properties at
the moment, until libinput becomes the fallback driver handling
tablet and pad management.
Whenever a tool becomes in proximity, a new ClutterInputDeviceToolXI2
will be created (if it wasn't created previously) for the given
serial number. This tool will be set in all events send from the
device.
https://bugzilla.gnome.org/show_bug.cgi?id=773779
The Clutter X11 backend can't drop CLUTTER_PEN_DEVICE and
CLUTTER_ERASER_DEVICE in favor of CLUTTER_TABLET_DEVICE without
losing information (as the driver will create one device for each).
So make MetaInputSettings cater for both sets of device types.
https://bugzilla.gnome.org/show_bug.cgi?id=773779
Its only purpose was caching settings applying to an stylus/tool, this
is now handled through ClutterInputDeviceTool evdev specific API, or
X device properties, so is not needed anymore.
https://bugzilla.gnome.org/show_bug.cgi?id=773779
Stylus configuration (stylus buttons, pressure) was handled
at the very high level, doing the button and pressure translations
right before sending these to wayland clients.
However, it makes more sense to store these settings into the
ClutterInputDeviceTool itself, and have clutter apply the config
at the lower level so 1) the settings actually apply desktop-wide,
not just in clients and 2) X11 and wayland may share similar
configuration paths. The settings are now just applied whenever
the tool enters proximity, in reaction to
ClutterDeviceManager::tool-changed.
This commit moves all handling of these two settings to
the clutter level, and removes the wayland-specific paths
https://bugzilla.gnome.org/show_bug.cgi?id=773779
We do so whenever a tool enters or leaves proximity. We now also
ensure that last_tool is NULL after it leaves proximity, although
the CLUTTER_PROXIMITY_OUT event itself should still contain tool
information.
https://bugzilla.gnome.org/show_bug.cgi?id=773779
And remove the wayland-specific handling. This works for both Wayland and
X11 (provided the compositor receives pad events through a passive grab
there).
https://bugzilla.gnome.org/show_bug.cgi?id=773779
We most notably handle button events (acquired through a passive grab on
all device buttons) which are translated to CLUTTER_PAD_BUTTON* events,
so there is generic handling of pad actions on X11.
https://bugzilla.gnome.org/show_bug.cgi?id=773779
We kind of rely on the ::show-pad-osd handler to destroy the
previous actor. Just prevent the emission of multiple signals
till the actor has been destroyed.
https://bugzilla.gnome.org/show_bug.cgi?id=771067
Wayland popup grabs, unlike other grab types, can be safely cancelled
so there's no reason to deny compositor grab requests if a wayland
popup is on.
In particular, this allows entering the overview via a keybinding or
locking the screen while a wayland popup has a grab which is something
that's been advertised as a wayland improvement over X.
https://bugzilla.gnome.org/show_bug.cgi?id=771235
Enabling edge scrolling before disabling two finger would result in
edge scrolling not actually being enabled because two finger is still
enabled at the time and we bail out.
This patch moves this logic to common code for both the native and X
backends and fixes it by ensuring that both settings are never set at
the same time and still re-checking if edge scrolling should be
enabled after two finger scrolling gets disabled.
We also simplify the code by not checking for supported/available
settings since the underlying devices will just reject those values
and there isn't anything we can do about it here. It's the UI's job to
only show supported/available settings to users.
https://bugzilla.gnome.org/show_bug.cgi?id=771744
Checking for supported methods isn't needed since libinput will just
error out and do nothing itself if a requested method isn't supported
and, in fact, this logic was preventing the enum values 0 from being
set.
https://bugzilla.gnome.org/show_bug.cgi?id=771744
Commit fcc7501eb8 had the side-effect of
stacking fullscreen windows below docks which went unnoticed since we
don't use docks in GNOME anymore.
Instead of re-introducing the fullscreen layer, which we don't need
otherwise, we can fix this issue by ensuring we stack docks below all
other windows when the monitor they're on is marked fullscreen. This
has the added benefit that the visibility rule for 3rd party docks
becomes the same as gnome-shell's chrome.
https://bugzilla.gnome.org/show_bug.cgi?id=772937
The keyboard focus semantics for non-grabbing xdg_shell v6 popups is
pretty undefined.
Same applies for subsurfaces, but in practice, subsurfaces never receive
keyboard focus, so it makes sense to do the same for non-grabbing
popups.
https://bugzilla.gnome.org/show_bug.cgi?id=773210
mutter would remove focus from a toplevel when showing one of its
transient window which is not on top and not focused.
When using xdg_popup without grab as allowed in xdg_shell v6, the popup
wouldn't be focused, and if an intermediate event occurs before the
popup is shown, it's not placed on top either, which could randomly
trigger a loss of focus in the corresponding toplevel window.
Remove that special case, it doesn't make much sense to globally unset
focus when mapping a new window.
https://bugzilla.gnome.org/show_bug.cgi?id=773210
commit e2bfaf0751 does this:
g_hash_table_insert (cards,
g_udev_device_get_name (parent_device),
g_steal_pointer (&parent_device));
The problem is the g_steal_pointer call may happen before the
g_udev_device_get_name call leading to a crash.
This commit does the get_name call on an earlier line
https://bugzilla.gnome.org/show_bug.cgi?id=771442
Right now we accept any character device that matches the glob card*.
That's fine, but we can be a little more specific by checking that
the devtype is what we expect.
This commit does that.
https://bugzilla.gnome.org/show_bug.cgi?id=771442
Despite g_udev_client_new taking a list of subsystems, it doesn't
implicitly filter results to those subsystems.
This commit explicitly adds a subsystem match to make sure sound cards
don't end up in the resulting list of video cards.
https://bugzilla.gnome.org/show_bug.cgi?id=771442
We shouldn't cancel the pointer grab when there is a compositor grab,
since that'd break things like drag-n-drop via the overview and
alt-tabs.
The original reason for cancelling the pointer grab on compositor grabs
was to avoid a re-entry when a compositor grab was activated while
there was an active pointer constraint grab. The re-entry would happen
when the compositor grab cleared the pointer focus. Clearing the focus
would trigger the pointer constraint to be deactivated, which would end
its grab. Ending the grab would reset the grab to the default one, which
could focus the same surface again, triggering the constraint to
re-enable before it finished disabling.
This is now avoided because the default grab handler is now aware of
compositor grabs, and won't override the cleared pointer focus until
the compositor grab ends.
https://bugzilla.gnome.org/show_bug.cgi?id=772914
Teach the default grab about compositor grabs (i.e.
display->event_route) so that it can avoid setting a pointer focus when
after the compositor grab actively unset the pointer focus.
https://bugzilla.gnome.org/show_bug.cgi?id=772914
The frame rect will at this point not be set for Wayland popups, since
the popup is placed and constrained before the actual buffer will be
attached. To still be able to calculate a proper monitor to be used for
constraining, use the ConstraintInfo::current dimensions instead, since
they will have the expected size. This should not cause any issues with
present paths since when a window is otherwise placed, it usually
doesn't change monitor calculation result.
This fixes opening a popup menu that would be positioned on the left
edge of a not-left-most monitor, for example a 'File' menu on a window
maximized on a second monitor.
https://bugzilla.gnome.org/show_bug.cgi?id=773141
Using the view's MetaMonitorInfo to find all the crtcs which should be
configured to display a given onscreen doesn't work unfortunately. The
association runs only the other way around, i.e. we need to go through
each crtc and find the ones corresponding to our monitor info.
https://bugzilla.gnome.org/show_bug.cgi?id=773115
If this isn't initialized and an idle watch gets instanced before
meta_idle_monitor_native_reset_idletime() gets called, that idle watch
would get triggered as soon as we hit the main loop.
This was causing gnome-session to go into idle mode at session start
thus making gnome-shell lock the screen.
In the past this bug was being masked by either logind emiting
session active signals or a stray input event making it through at
startup.
https://bugzilla.gnome.org/show_bug.cgi?id=772839
Commit 5fbb479301 was wrong too. What we
really want to do here is getting view relative coordinates given the
view's and the rectangle's global coordinates so we need to subtract
the view's origin from the rectangle's.
https://bugzilla.gnome.org/show_bug.cgi?id=771502
When disabling the device/capability, we can't rely on cancelled events
being emitted timely, because the capability will be already disabled by
then, all touches must be cancelled immediately then.
https://bugzilla.gnome.org/show_bug.cgi?id=772929
When the monitor is scaled (i.e. HiDPI scaling) the placement coordinates
ere still in unscaled xdg_surface window geometry coordinate space when
used to place the window. Fix this by scaling the coordinates by the
monitor scale of the parent toplevel window before using them.
https://bugzilla.gnome.org/show_bug.cgi?id=771841
Always use the monitor of the toplevel surface's window, so that the
popup menu and the parent will always have the same scale. This fixes
the dimensions sent in the xdg_popup configure event.
https://bugzilla.gnome.org/show_bug.cgi?id=771841
When Xwayland confines, the surface dimensions will include the server
side window manager decorations. We don't want the decorations to be
included in the constraint region so intersect the calculated input
region with the parts of the buffer rect that is not part of the window
frame.
https://bugzilla.gnome.org/show_bug.cgi?id=771859
Put the conditions for enabling a pointer constraint in a helper
function, and use that in both maybe_enable() and maybe_remove(). The
constraint region checking is still only done in maybe_enable()
however.
This changes the conditions for maybe disabling the constraint on focus
change and other trigger points, namely it makes constraints by Xwayland
not disable when they shouldn't due to the constraining window being an
override-redirect window.
https://bugzilla.gnome.org/show_bug.cgi?id=771859
When the grab is cancelled, for example because of an Alt-tab, VT
switch etc, disable or remove (depending on the constraint type) the
constraint. This avoids a re-entry issue when the focus is returned and
the focus listener tries to re-enable a disabled constraint.
https://bugzilla.gnome.org/show_bug.cgi?id=771859
Dismiss the popup when the grab is cancelled, so that if the grab is
ended for whatever reason (such as VT switching or the last pointer
being disconnected), it doesn't try to end the grab when it isn't
active.
This fixes a crash when VT switching back and forth while a popup grab
is active.
https://bugzilla.gnome.org/show_bug.cgi?id=771858
Previously a grab could suddenly end without the grabber knowing
anything about it. Some grabs assume they won't suddenly end without
notice, and can use then new 'cancel' vfunc to be notified.
Currently a grab is cancelled when a new one is started (i.e. in
meta_wayland_pointer_grab_start()), when a non-popup compositor wide
event route is initiated, and when the seat looses the pointer
capability.
https://bugzilla.gnome.org/show_bug.cgi?id=771858
Instead of using meta_wayland_pointer_end_grab() which focuses the new
grab, add a new helper mean to be used to reset the grab state without
changing the pointer focus. When using this function, the call site is
supposed to explicitly manage focus.
https://bugzilla.gnome.org/show_bug.cgi?id=771646
Make the caller of focus setting check whether there is a pointer to
update the focus state of. It makes it more obvious what to expect, as
the call would be a no-op in when no pointer is present.
Grabbing is still allowed without the presence of a pointer because it
is used by popups even on touch-only systems.
https://bugzilla.gnome.org/show_bug.cgi?id=771646
Make the caller of focus setting and grab starting check whether there
is a keyboard to update the focus state or start grabbing. It makes it
more obvious what to expect, as the call would be a no-op in when no
keyboard is present.
https://bugzilla.gnome.org/show_bug.cgi?id=771646
The variable name 'l' usually refers to a GList iterator, but here it's
just a short hand for a specific list. Stop using this shorthand, since
it just makes it harder to read what list is used.
https://bugzilla.gnome.org/show_bug.cgi?id=771646
The order doesn't only affect the visual layout, but also which action
cancels the dialog (and therefore responds to Escape). It is completely
surprising that this triggers a destructive action like force-quit, so
swap the actions to wait when the dialog is cancelled.
https://bugzilla.gnome.org/show_bug.cgi?id=737109
GNOME Shell's window matching currently fails frequently with Flatpak
applications, as one of the primary hints used to link windows with
.desktop files - the WM_CLASS - no longer matches when flatpak renames
the exported .desktop file. Luckily, Flatpak provides us with a fail-safe
way to map from the PID to the corresponding application ID, so expose an
appropriate method that allows GNOME Shell to reliably match windows to
the corresponding Flatpak app.
https://bugzilla.gnome.org/show_bug.cgi?id=772614
In order to kill a window, on both X11 and wayland we first try to
kill(3) the corresponding process, so we can add the newly added
get_client_pid() method to share that code.
https://bugzilla.gnome.org/show_bug.cgi?id=772613
It is often useful to identify the client process that created
a particular window, however the existing meta_window_get_pid()
method relies on _NET_WM_PID, which is only available on X11 and
depends on applications to set it correctly (which may not even
be possible when the app runs in its own PID namespace as Flatpak
apps do). So add a get_client_pid() method that uses windowing
system facilities to resolve the PID associated with a particular
window.
https://bugzilla.gnome.org/show_bug.cgi?id=772613
Previously the focus was reset implicitly by a memset() on the whole
MetaWaylandPointer struct. When MetaWaylandPointer was turned into a
GObject, this was not possible any more, and the focus was not updated
properly.
https://bugzilla.gnome.org/show_bug.cgi?id=771646
Going through the global mode pool and then checking if the mode is
available for a given output is pointless work since we can look at
the output's available modes directly.
This implicitly changes how we choose the default mode since, instead
of relying on the sort order of the global modes array, we now rely on
the sort order of the output modes array. Still not ideal, but at
least it makes more sense since the global array is essentially
unsorted.
https://bugzilla.gnome.org/show_bug.cgi?id=772176
This isn't technically needed and, in fact, makes us default to
interlaced modes in some cases which isn't desirable.
Note that X doesn't account for these flags either for its mode
refresh rates.
https://bugzilla.gnome.org/show_bug.cgi?id=772176
As the m format specifier doesn't consume any arguments, the number
of varargs currently doesn't match the number of specifiers; the
failed transform may be relevant, so include it in the message
instead of removing the excess argument.
There may be other windows managing selection whose events are seen in
our GDK event filter, like st-clipboard in gnome-shell, we should in
that case not interfere on Selection/SelectionRequest events that are
not meant for us.
This fixes an odd feedback loop where requesting clipboard contents from
wayland results in a XConvertRequest call and a SelectionRequest event
that is interpreted by mutter as a request from another X11 client, so
the current data source is poked for content, which happens to be the
X11 bridge, which does a XConvertRequest to get contents... This is only
broken after the many nested async operations create enough pipes and
cancellables to run out of fds.
Adding checks to ensure only events meant to our "selection owner"
window are managed prevent this unintended loop to happen in the first
place.
https://bugzilla.gnome.org/show_bug.cgi?id=760745
backends/meta-input-settings.c:1245:27: error: format '%lx' expects
argument of type 'long unsigned int', but argument 3 has type 'guint64
{aka long long unsigned int}' [-Werror=format=]
Commit a4fb7ef5a3 dropped translations of our internal cogl/clutter
forks, which broke the local-based text direction support. Instead
of bringing back translations just for this purpose, we can re-use
GTK's translations which use the same technique.
https://bugzilla.gnome.org/show_bug.cgi?id=771549
Those will be unseen by g-s-d/g-c-c, so no settings will be written on
disk for those. Still, look up an ID correctly in this case instead of
crashing.
https://bugzilla.gnome.org/show_bug.cgi?id=771628
When we mess with a window actor's visibility from the shell side
(yes, I know :-( ), we should at least restore the proper visibility
when we're done with it ...
https://bugzilla.gnome.org/show_bug.cgi?id=771536
A xdg_popup, when active, always has a parent surface. However, a popup
created may immediately become invalid, for example when it is not
granted a grab, in which case it won't be assigned a parent since it
will never be mapped.
This case needs to be handled elsewhere, as one cannot assume a
MetaWaylandXdgPoup that is processed (via wl_surface commit handling
etc) will have a parent_surface.
https://bugzilla.gnome.org/show_bug.cgi?id=771495
If a X11 client would initiate a Xdnd session after it had lost pointer
focus (for example when the Xdnd event starting the drag happens after
the implicit pointer grab is already broken due to the button being
released), just end the drag operation instead of dereferencing the
non-existing focus surface.
Also avoid using a native Wayland surface as a drag origin, as that can
never happen, but allow any arbitrary Xwayland client, since there is
no way to find out the actual drag origin.
https://bugzilla.gnome.org/show_bug.cgi?id=770940
We cannot check 'has-target' or 'has-seat' when verifying a
wl_data_offer.finish request is valid or not, since the source may have
effected 'has-target' or whether theh source has a seat or not when the
finish request was already on the wire.
Instead of checking against the source state, keep track whether the
required operations has been done on the offer in question (i.e.
whether an action has been sent, or a mime type been accepted).
This fixes incorrectly raised error when dragging from gtk+'s testdnd
via Xwayland onto gtk+'s testdnd using Wayland.
https://bugzilla.gnome.org/show_bug.cgi?id=770937
Now with the existance of offscreen view framebuffers the buffer age
damage regions are only valid if the view in question doesn't doesn't
have an intermediate offscreen. So, for views that doesn't have buffer
age, return the dirty pixel (0,0).
https://bugzilla.gnome.org/show_bug.cgi?id=770672
Since wl_surface.set_buffer_transform() is not supported, until it is
added, pretend outputs are never transformed, so that clients are less
likely to attach pre-transformed buffers.
https://bugzilla.gnome.org/show_bug.cgi?id=770672
When blitting an offscreen onto an onscreen, the whole offscreen should
always be drawn on the whole onscreen. Thus, don't try to convert
between coordinate spaces, just draw the whole offscreen on the whole
onscreen.
https://bugzilla.gnome.org/show_bug.cgi?id=770672
Also maybe remove a constraint when the pointer focus changes. This is
needed because when Xwayland has a constraint focus may change, the
constraint object will not receive a 'appears-focused' event on its
window since it never changed.
This happens for example when an override-redirect window (which never
appears focused) holds the constraint, and alt-tab happens. In this case
focus changes, but from the constraint's point of view, none of the
windows it knows about changed its focus appearance.
https://bugzilla.gnome.org/show_bug.cgi?id=771345
Instead of initializing the default grab when the device class is
enabled, initialize it on object initialization. This way other device
classes can still grab the pointer, as if there was one. This may be
useful for example if a touch grab is active and a mouse is connected.
This also makes it possible for popup grabs, which currently use a
pointer grab for controlling, to be triggered by touch devices, while
still holding an active pointer grab.
https://bugzilla.gnome.org/show_bug.cgi?id=771305
Make the device <-> seat association permanent, and move it into
MetaWaylandInputDevice. A device will never be disassociated with a
seat, so there is no point in unsetting it.
https://bugzilla.gnome.org/show_bug.cgi?id=771305
Add a new object class, MetaWaylandInputDevice, and make all device
classes (pointer, keyboard, touch) inherit it. In the future common
functionality may be placed there.
https://bugzilla.gnome.org/show_bug.cgi?id=771305
Meant to replace explicitly checking whether a
MetaWaylandPointer/MetaWaylandKeyboard/MetaWaylandTouch has a seat or
not to determine whether they are supposed to be active or not.
https://bugzilla.gnome.org/show_bug.cgi?id=771305
This is needed to make the wayland backend react to configuration
changes until gnome-control-center is updated to use the
gsettings-desktop-schemas settings.
https://bugzilla.gnome.org/show_bug.cgi?id=771315
Some output devices only advertise their preferred mode even though
they're able to display others too. This means we can include some
common modes in each output's supported list.
This is particularly important for mirroring, since we can only mirror
outputs which are using the same resolution.
https://bugzilla.gnome.org/show_bug.cgi?id=744544
Since Xwayland surface constraints might need to enable not only
because the constrained window appears focused, add a pointer focus
listener and try constrain whenever the pointer focus changes. It's
still required that a Xwayland window is focused to activate.
https://bugzilla.gnome.org/show_bug.cgi?id=771050
Xwayland surfaces are special, because there is no reliable way to
associate a window with its corresponding "application window" (the one
which was given focus). Many games that require pointer warping and
confining pointer grabs may for example create override redirect windows
and make that window receive input even though it will never be the
focus window.
Therefore, the requirements for enabling a constraint for a wl_surface
from Xwayland needs to be relaxed in order. This commit changes
Xwayland wl_surfaces to not require being focused to be enabled; it'll
be enabled as long as any X11 window is the one with focus.
https://bugzilla.gnome.org/show_bug.cgi?id=771050
Require pointer focus to enable, otherwise we can't guarantee it has
entered the surface, as the focus may have been given to a subsurface,
override-redirect or other sub window covering the surface that was
requested to have o pointer constraint.
https://bugzilla.gnome.org/show_bug.cgi?id=771050
Add a signal that is emitted when the pointer focus surface of the
pointer device changes. This will later be used by the pointer
constraints to maybe enable pointer constraints when a surface receives
pointer focus.
https://bugzilla.gnome.org/show_bug.cgi?id=771050
For Xwayland, a newly created wl_surface and X11 Window pair may not be
immediately associated, but Xwayland may still request a pointer
constraint on some of its wl_surface's. Handle the situation by
postponing maybe enabling the constraint until the window and surface
has been associated.
https://bugzilla.gnome.org/show_bug.cgi?id=771050
Move the MetaWaylandSurface::destroy signal before starting the actual
destruction, in wl_surface_destructor, so that all fields (e.g. surface
role) are intact when the listeners are invoked.
https://bugzilla.gnome.org/show_bug.cgi?id=771050
When the Xwayland wl_surface is created, it may not yet be possible to
associate it with the corresponding X11 Window. Add a signal to the
Xwayland role to communicate with any interested parties.
https://bugzilla.gnome.org/show_bug.cgi?id=771050
Clutter discards any motion event if next event happens to also be a
motion event. This is problematic when the motion event carries
relative motion deltas, since the information about them is completely
lost.
Until we have moved away made the stage stop discarding motion events,
lets work around the issue by compressing them, effectively adding
multiple relative motion deltas together, would one be discarded.
https://bugzilla.gnome.org/show_bug.cgi?id=771049
If a client would attach a buffer to a surface, commit, destroy the
buffer and then later set the surface as a cursor, there will be no
wl_buffer available to be used by the cursor role. Instead of
dereferencing the non-existing wl_buffer resource, handle this situation
by logging a warning and treating a prematurely destroyd wl_buffer as if
no buffer had been attached.
https://bugzilla.gnome.org/show_bug.cgi?id=770992
Not having a surface actor would cause the window actor state to be
considered frozen, thus causing various state (such as geometry, shape
etc) synchronization to be delayed until thawed. If the window actor
was "thawed" due to having a surface set, not all state would be
properly synchronized, causing the thawed window actor to be displayed
incorrectly.
This patch fixes this by putting state synchronization after thawing in
a common function, calling it both from frozen count decreasing and
surface setting.
This fixes for example misplaced menus in Steam.
https://bugzilla.gnome.org/show_bug.cgi?id=770991
The rectangle passed to capture_view() is in stage coordinate space;
thus, to translate to framebuffer coordinate space, the origin need to
be translated by the view layout position.
This fixes capturing views not at position (0, 0).
https://bugzilla.gnome.org/show_bug.cgi?id=770127
Implement min/max size request from xdg-shell-v6 and plug it into the
existing code so that windows with fixed size cannot be tiled/maximized
in Wayland just like in X11.
Bugzilla: https://bugzilla.gnome.org/show_bug.cgi?id=770226
The seat capability updating is synchronous, but input events are
asynchronous (first queued then emitted). This means we may end up in a
situation where we from libinput first may receive a key event,
immediately followed by a device-removed event. Clutter will first
queue the key event, then remove the device, immediately triggering the
seat capability removal.
Later, when the clutter stage processes the queued events, the
previously queued key event will be processed, eventually making it
into MetaWaylandSeat. Before this patch, MetaWaylandSeat would still
forward the key event to MetaWaylandKeyboard, even though it had
'released' it. Doing this would cause referencing potentially freed
memory, such as the xkb state that was unreferenced when the seat
removed the capability.
In order to avoid processing these lingering events, for now, just drop
them on the floor if the capability has been removed.
Eventually, the event queuing etc needs to be redesigned to work better
when used in a Wayland compositor, but for now at least don't access
freed memory.
https://bugzilla.gnome.org/show_bug.cgi?id=770727
When a modal transient is unmanaging, most likely the parent of the
modal transient should be focused.
In Wayland, a MetaWindow is created when a shell surface role (like
xdg_toplevel) is created, but a window cannot be shown until a buffer
is attached. If a client would create two modal transients and make
them both have the same parent, but only one get a buffer attached
(i.e. shown), when unmanaging the modal transient that was showing,
when finding a new focus candidate, the stacking code will ignore the
not-to-be-shown buffer-less modal transient when finding a good
candidate for focusing. In the case described here, this means it will
find the parent of the unmanaging modal transient.
This newly chosen candidate will then be passed to meta_window_focus();
meta_window_focus() will then try to find any modal transient to focus
instead, will find the one without any buffer, then fail to focus it
because it cannot be mapped, thus making meta_window_focus() not focus
anything. Since meta_window_focus() didn't change any focus state, the
assert in meta_window_unmanage() checking that the unmanaging window
isn't focused anymore will be hit, causing mutter to abort.
For now, fix this by checking whether the modal transient can actually
be focused in meta_window_focus(). For X11 client windows, a window
will be defined to be focusable always, but for Wayland client windows,
a window will be determined focusable only if it has a buffer attached.
In the future, we should probably do a more thorough refactorization of
focus handling to get rid of any X11 - Wayland differences.
https://bugzilla.gnome.org/show_bug.cgi?id=757568
We may be assigned multiple times, if the surface is assigned to be a
cursor surface multiple times. Each time e.g. wl_pointer.set_cursor is
called, we'll be assigned.
While the role object exists, we'll handle buffer use count even when
we are not actively assigned, thus we should only handle the initial
assignment use count bump when constructing, so that we don't increase
it when reassigned, where the wl_resource may already have been
released.
https://bugzilla.gnome.org/show_bug.cgi?id=770402
For backend handled cursors, if nothing else changes on the clutter
stage, we end up not sending out frame callbacks since clutter doesn't
draw a new frame.
To fix this, we'll keep cursor surfaces' frame callbacks separate from
other surfaces' and trigger them from the new
MetaCursorRenderer::cursor-painted signal which handles both software
and hardware cursors.
https://bugzilla.gnome.org/show_bug.cgi?id=749913
This signal allows interested parties to be notified of a new cursor
frame being painted regardless of whether it's being painted by the
backend directly or if it's a software rendered cursor frame handled
by clutter.
https://bugzilla.gnome.org/show_bug.cgi?id=749913
Instead of hiding stage views enablement behind MUTTER_STAGE_VIEWS=1,
default to enable it, while making it possible to disable using
MUTTER_STAGE_VIEWS=0 instead.
https://bugzilla.gnome.org/show_bug.cgi?id=770366
Absolute pointer events used the X coordinate as both X and Y. This
caused the pointer cursor to be moved incorrectly for absolute pointer
devices, commonly used in virtual machines.
https://bugzilla.gnome.org/show_bug.cgi?id=770557
Even without a compositor grab, key events may still be expected to
be processed by the compositor and not applications, for instance
when using ctrl-alt-tab to keynav in the top bar. On X11, focus is
moved to the stage window in that case, so that events are processed
before they are dispatched by the window manager. On wayland, we need
to handle this case ourselves, so make sure to not pass key events to
wayland in that case, and move the key focus back to the stage when
appropriate.
https://bugzilla.gnome.org/show_bug.cgi?id=758167
The WL_bind_wayland_display spec says that EGL images should be created
using EGL_WAYLAND_BUFFER_WL as the target and a NULL context. Mesa
seems to be lenient and accept any context, however some other stacks
aren't so forgiving and fail if anything apart from EGL_NO_CONTEXT is
used.
Signed-off-by: Sjoerd Simons <sjoerd.simons@collabora.co.uk>
https://bugzilla.gnome.org/show_bug.cgi?id=769731
Switch to the output naming logic used by the X server's modesetting
driver which, in particular, uses drmModeConnector's connector_type_id
instead of connector_id.
The kernel generates new connector_id's every time there are changes
which means we can't identify the same monitor on the same connector
after an hardware hotplug. Switching to connector_type_id fixes this.
https://bugzilla.gnome.org/show_bug.cgi?id=770338
If cogl fails to create a texture from the client's given buffer,
mutter would raise a fatal error and terminate.
As a result, a broken client might kill gnome-shell/mutter and take the
entire Wayland session with it.
Instead of raising a fatal error in this case, log the cogl error
message and send the client an OOM error, so mutter/gnome-shell can
survive an unsupported buffer size.
https://bugzilla.gnome.org/show_bug.cgi?id=770387
For some reason, when a modal dialog was made an attaching
transient-for, if the window wasn't "constructing", it would be
unmanaged and rely on some side effect to be recreated. This side
effect is not triggered for Wayland clients, thus if one happen to set
a surface as "modal" via gtk_surface.set_modal before
xdg_toplevel.set_parent, it'd be unmanaged and never show up.
Instead, simply just set the tranciency anyway for Wayland clients.
This makes GTK+ clients that set_modal() before set_transient_for()
work.
https://bugzilla.gnome.org/show_bug.cgi?id=770324
Windows from Xwayland still needs to use the Wayland path, but is
represented an MetaWindowX11, thus the abstraction introduced in
"window: Make meta_window_has_pointer() per protocol implemented"
is wrong. Lets turn back time, and reconsider how this can be
abstracted more correctly in the future.
This reverts commit 9fb891d216.
Rely on the actor surface role's commit function for queuing frame
callbacks. This also makes the surface actor state synchronization work
again, which was broken by 'wayland: Sync surface actor state in actor
role commit handler'.
https://bugzilla.gnome.org/show_bug.cgi?id=770131
Don't check whether the surface of the role has a window, but whether
the corresponding toplevel surface has a window. This is necessary to
make subsurfaces not always early out.
https://bugzilla.gnome.org/show_bug.cgi?id=770131
There is nothing stopping a subsurface from commiting its state before
its parents role has been assigned. Thus, we need to handle
meta_wayland_surface_get_toplevel() returning NULL for subsurfaces even
on commit.
Make sure to always call the parent role commit vfunc, so that they can
handle updating their state properly.
This means other places need to handle the situation where
surface->window is NULL on commit. This may for example happen when the
parent of a modal dialog is unmapped or NULL is attached to a
wl_shell_surface.
https://bugzilla.gnome.org/show_bug.cgi?id=769936
Port the xdg_shell implementation to use the unstable v6 protocol. This
includes:
- making xdg_surface a generic base interface for xdg_shell surface
roles
- create a xdg_toplevel role replacing the old xdg_surface
- change the xdg_opup role to be based on xdg_surface
- make xdg_popup not grab by default
- add support for xdg_positioner
https://bugzilla.gnome.org/show_bug.cgi?id=769936
Emit a 'configure' signal before configuring the role. This will enable
extensions to send its own configure events before the role is
configured.
https://bugzilla.gnome.org/show_bug.cgi?id=769936
Add support for assigning a window a custom window placement rule used
for calculating the initial window position as well as defining how a
window is constrained.
The custom rule is a declarative rule which defines a set of parameters
which the placing algorithm and constrain algorithm uses for
calculating the position of a window. It is meant to be used to
implement positioning of menus and other popup windows created via
Wayland.
A custom placement rule replaces any other placement or constraint
rule.
https://bugzilla.gnome.org/show_bug.cgi?id=769936
Allow passing parameters (only GObject parameters supported for now) so
that role assignment can affect the paremeters set during construction.
If a role was already assigned when assigning, the passed parameters
are set using g_object_set_valist().
https://bugzilla.gnome.org/show_bug.cgi?id=769936
Instead of piping the protocol file content to wayland-scanner, pass
the file name as an argument. This enables a new enough wayland-scanner
to print more meaningful error messages.
https://bugzilla.gnome.org/show_bug.cgi?id=769936
We can only honor this properly in the MUTTER_STAGE_VIEWS=1 case. When using
the legacy view, software implemented transforms are only exposed if there is
only one output, as we can only transform the entire stage there.
https://bugzilla.gnome.org/show_bug.cgi?id=745079
The texture is only created if the view is transformed at the software level,
otherwise the texture is NULL, and rendering happens on the onscreen.
https://bugzilla.gnome.org/show_bug.cgi?id=745079
"Blit" the result on the framebuffer after each view is painted.
This of course only applies if there is an offscreen buffer to
perform any blitting. Otherwise the onscreen framebuffer is rendered
to directly and this step is not necessary.
https://bugzilla.gnome.org/show_bug.cgi?id=745079
The offscreen is given through the ::back-buffer property, the ClutterStageView
will set up the the CoglPipeline used to render it back to the "onscreen"
framebuffer.
The pipeline can be altered through the setup_pipeline() vfunc, so ClutterStageView
implementations can alter the default behavior of blitting from offscreen to
onscreen with no transformations.
All getters of "the framebuffer" that were expecting to get an onscreen have
been updated to call the right clutter_stage_view_get_onscreen() function.
https://bugzilla.gnome.org/show_bug.cgi?id=745079
The call to _cogl_framebuffer_winsys_update_size() results in no-op here,
as the framebuffer has already the right size when rebuilding the views.
https://bugzilla.gnome.org/show_bug.cgi?id=745079
Those will need a separate treatment from the modes that we eventually
support through "software", so split those into a separate enum so we
can can do the right thing when applying the configuration.
Also, add a helper function that returns the transform that the software
fallbacks should perform, which should be "normal" if the rotation is
already handled via hw.
The function applying the configuration has been modified to always set
a HW rotation mode (even if normal), when we come to support SW rotation
modes, we'll be relying on a normal transformation, so it will be
necessary to have mixed HW/SW managed transforms.
https://bugzilla.gnome.org/show_bug.cgi?id=745079
This commits adds support for exporting xdg_surface handles via
xdg_exporter and importing them via xdg_importer.
This bumps the required wayland-protocols version to 1.6.
https://bugzilla.gnome.org/show_bug.cgi?id=769786
Meant to be used by users of MetaWaylandSurface's that need to know
when the surface was unmapped. So far only emitted by shell surfaces
(surfaces with MetaWindow's).
https://bugzilla.gnome.org/show_bug.cgi?id=769786
As whether edge scrolling is enabled depends on whether two-finger
scrolling is disabled, make sure to update two-finger scrolling first.
Note that this only fixes the problem on startup. Changing the
settings in GSettings directly might cause an inconsistent state, but
the main UI for this setting, gnome-control-center, makes sure to
update two-finger scrolling before edge scrolling.
https://bugzilla.gnome.org/show_bug.cgi?id=769276
intltool is unmaintained nowadays while upstream gettext gained
support for formats like GSettings schemas and .desktop files,
and offers a mechanism to teach it about other XML formats not
yet supported out of the box which we can use for the rest.
So there's nothing stopping us, just make the switch ...
https://bugzilla.gnome.org/show_bug.cgi?id=769073
An empty argument list means "unspecified arguments", and not
"no arguments" like it does in C++. If an implementer of Mutter
plugins uses gcc -Wold-style-definition, as configured by
AX_COMPILER_FLAGS_CFLAGS, they will get warnings about this.
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: Florian Müllner <fmuellner@gnome.org>
Bug: https://bugzilla.gnome.org/show_bug.cgi?id=769971
The scale will have been set to 1 no matter what when initializing the
MetaOutput since it at the time didn't have an CRTC assigned to it.
Now, when we assign the CRTC to the output, we need to update the scale.
https://bugzilla.gnome.org/show_bug.cgi?id=769505
This is somewhat gross at the moment, because we're after all mimicking
real keyboard events, we can only lookup keycodes that are available
in the current map, and the control of levels is rather limited.
Eventually, we want to implement the text_input protocol, handle these
events separately to MetaWaylandKeyboard, so event->key.keyval is
is guaranteed to be the final result. Until then, this is the farthest
we can get.
https://bugzilla.gnome.org/show_bug.cgi?id=765009
Evcodes don't cut it when we have something already specifying the
character to be printed, despite the current group/level. This API
allows some more control on the intended output.
https://bugzilla.gnome.org/show_bug.cgi?id=765009
libinput does it for us, but only for physical devices. When we add
virtual devices to the same seat, we need to track button press count
ourself.
https://bugzilla.gnome.org/show_bug.cgi?id=765009
Virtual input devices aim to enable injecting input events as if they
came from hardware events. This is useful for things such as remote
controlling, for example via a remote desktop session.
The API so far only consists of stumps.
https://bugzilla.gnome.org/show_bug.cgi?id=765009
We do some things when binding to a socket fails (closing the fd,
logging, unlinking files, ...) those might affect errno in some
or other way, so it might no longer be EADDRINUSE even if we later
try to make those non fatal.
It seems better to check errno soon after the failure, and don't
rely on it in any way at a later point. All error paths in
bind_to_abstract_socket() also have early logging, which also might
help figure out better the point of failure when the socket fails
to be created.
https://bugzilla.gnome.org/show_bug.cgi?id=769578
Allocate the offscreen stage view framebuffers up front; otherwise they
may get allocated after the viewport calculated by the stage is set,
which would cause the viewport to be incorrect until recalculated.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Support changing the mouse and trackball acceleration profile. This
makes it possible to for example disable pointer acceleration by
choosing the 'flat' profile.
This adds an optional dependency on gudev. Gudev is used by the X11
backend to detect whether a device is a mouse or not. Without gudev
support, the accel profile settings has have effect for mouse devices.
Trackball still uses the "strstr" approach, since udev doesn't support
tagging devices as trackball devices yet.
https://bugzilla.gnome.org/show_bug.cgi?id=769179
Add support for setting edge-scrolling separately from two-finger
scrolling. We now have 2 separate boolean settings for those, with the
Mouse panel in gnome-control-center allowing to set only one of those at
a time, but nothing precludes both being set in the configuration.
We need to handle:
- two-finger-scrolling-enabled and edge-scrolling-enabled settings both
being set.
- those 2 settings being change out-of-order
- two-finger-scrolling being set on a device that doesn't support it
- edge-scrolling-enabled on a device that doesn't support it
And the combinations of one touchpad supporting just one of edge
scrolling and two-finger scrolling and another vice-versa.
https://bugzilla.gnome.org/show_bug.cgi?id=768245
Instead of continuing eventually crashing with a segmentation fault due
to a missing renderer, make MetaBackend an GInitable, and gracefully
handle the failure to fully create the backend with an EXIT_FAILURE.
https://bugzilla.gnome.org/show_bug.cgi?id=769036
We only use a handful of the attributes set, so lets stop pretending
that things are initialized for a reason. Eventually we should stop
using XWindowAttributes in the generic MetaWindow creation path.
https://bugzilla.gnome.org/show_bug.cgi?id=769070
If the compiler cannot figure out that the condition for setting
the dev variable is the same as the condition for accessing it,
it will complain about potential uninitialized use.
We must lookup the mode switch serial for the group where the button
belongs to. Also, avoid the changes if the client requests setting
the feedback for buttons owned by the compositor.
We assumed that each group could only have 1 strip and/or ring, because
accounting is performed per group, so we could not assume the real
index for anything above 1. Get rid of this restriction, now that
MetaWaylandTabletPad does its own accounting of rings/strips, alongside
groups.
This is best for 2 reasons:
- It's feels cleaner doing first creation of rings/strips and then
the group assignment. The other option is making groups iterate
other all rings/strips and selectively skip those not meant for
it, which sounds somewhat redundant.
- Some minimal accounting of rings/strips without group restrictions
is needed for meta_wayland_tablet_pad_get_label().
The rings/strips memory is now owned by MetaWaylandTabletPad instead
of groups, which is sort of meaningless since all are meant to go
at the same time.
There may be external/compositor-specific reasons to trigger the
pad OSD. Expose this call so the pad OSD can be triggered looking
up the right settings, monitor, etc...
This API will be used from the gnome-shell pad OSD implementation, in order
to show the actions that currently apply to every button/ring/strip in the
tablet.
It does nothing at the moment, but can be hooked into MetaWaylandTabletPad
now. For X11, we need to trigger these for the pad events we receive from
the passive pad button grabs.
This function will be useful for the wayland implementation, because buttons
are mapped at the time of sending those through the wire.
As x11/wayland implementations differ here, this function will be useful for
the wayland implementation, as the action is handled lat
Some settings make no sense on external tablets, and others make
no sense in display/system-integrated tablets. Perform those checks
so we don't end up with possibly broken configuration.
Depending on clutter_input_device_get_mapping(), or whether the current
tool is either cursor or lens (those don't make any sense in absolute
mode), relative motions will be reported.
This function call only applies to tablets, and thus will error
out unless it's called with CLUTTER_TABLET_DEVICEs. This will
allow setting absolute/relative mapping on those on the fly, as
this is optional.
Given that information defines largely how such devices are to be
configured, it makes sense to have that information at hand. A getter
has been also added for the places where it could be useful, although
it will require HAVE_LIBWACOM checks in callers too.
All pads will share the same focus than the keyboard, so this means that:
- The focus changes in-sync for keyboard and all pad devices, and
- Newly plugged pads will be immediately focused on that same surface
This object represents the collection of buttons, strips and rings
in a tablet pad. All the objects created (pad, strips and rings)
share a common focus surface and have the same lifetime.
This will only be practical for pads (and maybe generic buttonsets in
the future?), we just need to know the number as the events will also
contain a number as the identificator.
Those map closely what we get from libinput. Button events have
been made its own separate struct, its semantics fall somewhere
in between of ClutterButtonEvent and ClutterKeyEvent, so is better
emitted as its own set.
This is now separated from the generic cursor one. This means that wl_surfaces
can't be shared across wl_pointer and wp_tablet_tool. This is a change in
tablet protocol v2.
This is a simple subclass of MetaWaylandSurfaceRoleCursor, mostly
so we can distinguish by GType, the methods in the parent class
still apply and are useful.
When launching a GNOME session from a text-mode VT, the logind session
type is unlikely to be set to either "wayland" or "x11". We search for a
supported session type first with logind and then with
$XDG_SESSION_TYPE. As a fallback, we also test $DISPLAY in case of a
"tty" logind session to support starting through xinit. Ideally, such
setups should set XDG_SESSION_TYPE=x11.
If no supported session type is found, we throw an error.
https://bugzilla.gnome.org/show_bug.cgi?id=759388
Currently the setter doesn't take ownership of the value, but dispose()
will unref it (and thus release someone else's reference). Fix this by
taking ownership of the property value in the setter.
They are already effectively interchangeable so this should reduce
pointless casts.
Just like in GDK though, we need to keep the old definition for
instrospection to be able to include the struct's fields.
There's no need for a cast for printing an object's type or address,
so we can remove variables that are unused when not building with
CLUTTER_ENABLE_DEBUG.
Mutter (and libmutter users) are the only users of this version of
cogl, and will more or less only use the cogl-1.0, cogl-2.0 and cogl
experimental API variants, and having the possibility of having
different API versions of the same API depending on what file includes
it is error prone and confusing. Lets just remove the possibility of
having different versions of the same API.
https://bugzilla.gnome.org/show_bug.cgi?id=768977
We bypass our build configuration to fetch API from a version which
isn't the one we actually use. Stop bypassing and just admit that the
1.0 API is still there, but still deprecated.
https://bugzilla.gnome.org/show_bug.cgi?id=768977
We didn't include clutter-build-config.h, meaning we included a
different API of cogl than the rest of clutter. This API contains the
function cogl_sqrti which was the only thing used. Lets include the
build config file and stop depending on the API that is no longer
exposed to us.
https://bugzilla.gnome.org/show_bug.cgi?id=768977
By creating a pending gbm/EGL surface pair, only setting it on
swap-buffers, we would draw onto a buffer on the old surface, then swap
the buffer from the new surface, causing the first frame after a
hot-plug always having no content.
This was in the past not very noticable since some non-deterministic but
frequent side effect in gnome-shell caused hot-plugging to always render
two new frames, but after "Introduce regional stage rendering", this
side effect did not occur as often, thus making it more visible.
This commit updates the current gbm/EGL surface pair before painting a
frame, so that when the frame is painted, the surface with the correct
size is used and the buffer from correct surface is swapped.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Being a listener to a signal, it is inconvenient to enforce order of
execution between different signal listeners. If there are things in
the backend that should be updated before various other signal
handlers, make sure so is done by emitting the signal after having
explicitly notified the backend.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
CoglFrameInfo is a frame info container associated with a single
onscreen framebuffer. The clutter stage will eventually support drawing
a stage frame with multiple onscreen framebuffers, thus needs its own
frame info container.
This patch introduces a new stage signal 'presented' and a accompaning
ClutterFrameInfo and adapts the stage windows and past onscreen frame
callbacks users to use the signal and new info container.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Call a CoglContext "cogl_context", CoglDisplay "cogl_display" and
CoglRenderer "cogl_renderer" so that they won't be confused with
ClutterContext, MetaDisplay and MetaRenderer etc.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Make the cogl vfunc functions have names that are globally
discoverable. Calling the same function in every backend the same name
causes code navigation tools to not function properly. Rename the
affected functions to closer correspond to the style mutter uses.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Add support for drawing a stage using multiple framebuffers each making
up one part of the stage. This works by the stage backend
(ClutterStageWindow) providing a list of views which will be for
splitting up the stage in different regions.
A view layout, for now, is a set of rectangles. The stage window (i.e.
stage "backend" will use this information when drawing a frame, using
one framebuffer for each view. The scene graph is adapted to explictly
take a view when painting the stage. It will use this view, its
assigned framebuffer and layout to offset and clip the drawing
accordingly.
This effectively removes any notion of "stage framebuffer", since each
stage now may consist of multiple framebuffers. Therefore, API
involving this has been deprecated and made no-ops; namely
clutter_stage_ensure_context(). Callers are now assumed to either
always use a framebuffer reference explicitly, or push/pop the
framebuffer of a given view where the code has not yet changed to use
the explicit-buffer-using cogl API.
Currently only the nested X11 backend supports this mode fully, and the
per view framebuffers are all offscreen. Upon frame completion, it'll
blit each view's framebuffer onto the onscreen framebuffer before
swapping.
Other backends (X11 CM and native/KMS) are adapted to manage a
full-stage view. The X11 CM backend will continue to use this method,
while the native/KMS backend will be adopted to use multiple view
drawing.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
We were compensating for a clone paint viewport offset even when we
were not in clone paniting mode. This would break painting if we offset
the viewport for some other reason for example as in the future stage
view painting.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Instead of assuming there is a single onscreen framebuffer, use the
helper functions for setting the frame callback and getting the frame
counter.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
In preperation for having allowing drawing onto multiple onscreen
framebuffers, move the onscreen framebuffer handling to the
corresponding winsys dependent backends.
Currently the onscreen framebuffer is still accessed, but, as can seen
by the usage of "legacy" in the accessor name, it should be considered
the legacy method. Eventually only the X11 Compositing Manager backend
will make use of the legacy single onscreen framebuffer API.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Split the stage window implementations into three separate objects: one
for X11 as a compositing manager, one for X11 running as a nested
Wayland compositor, and one for running with the native backend.
The new stage window implementations are only thin shells; this is in
preparation for making the stage windows behave more differently.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
The stage resizing was placed in the generic backend, which was only
run on certain configurations (when running nested or using the native
backend). This commits makes the resizing more explicit thus more
obvious.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
This commit completes the move of monitor logic to the monitor
mangager. The renderer now only deals with framebuffers, asking the
monitor manager to do the crtc flip tracking.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Let MetaMonitorManagerKms manage KMS modes. This lets us pass less
state to MetaRendererNative. Instead let MetaMonitorManager tell the
monitor manager when it should set the mode and with what framebuffer.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Absorb the CoglRendererKMS struct into MetaRendererNative. The gbm
device initialization is moved earlier so that the renderer fails to
initialize if the gbm device creation failed.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Move the KMS interaction from cogl into mutter, where most of the other
KMS interaction already takes place. This also removes dead code which
were only excercised when non-mutter callers used the cogl KMS backend.
The cogl KMS API was updated to pass via MetaRendererNative instead of
via the different cogl objects.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Instead of passing around the KMS file descriptor via clutter to cogl,
just make our own clutter backend create the cogl renderer and set the
KSM fd.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
MetaRenderer is meant to be the object responsible for rendering the
scene graph. It will contain the logic related to the cogl winsys
backend, the clutter backend, and the clutter stage window.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
The actor-offscreen-redirect didn't initialize its state properly, so
it could potentially end up with the "was_painted" state being TRUE
from the start, effectively skipping the whole test.
Fixing so that the test even run resulted in the test getting stuck in
a dead lock due to the verification that a frame was drawn was done
from a paint callback. A paint callback had the mutex held, so when the
test case tried to run the main loop, the next paint callback caller
path taken would try to re-lock the same mutex, thus dead lock.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Make it possible to force mutter to start as a X11 compositing/window
manager. This is needed when intending to start mutter as an X11 window
manager while running inside a Wayland session, for example when
intending to debug it in Xephyr.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Use the correct pointer types for cogl objects. This avoids warnings
when including the cogl headers doesn't result in all the cogl types
being typedefs to void.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
In cogl use cogl-config.h and in clutter use clutter-build-config.h. We
can't use clutter-config.h in clutter because its already used and
installed.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
If you include a file that might define __INSIDE_COGL_H__, don't
undefine it if it wasn't defined in that file. This makes it possible
to include for example cogl-gles2.h from some other file which defines
__INSIDE_COGL_H__.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Introduce two new clutter backends: MetaClutterBackendX11 and
MetaClutterBackendNative. They are so far only wrap ClutterBackendX11
and ClutterBackendEglNative respectively, but the aim is to move things
from the original clutter backends when needed.
https://bugzilla.gnome.org/show_bug.cgi?id=768976
Using clutter API to transform coordinates is only accurate right
after a clutter layout pass but this function is used e.g. to deliver
pointer motion events which can happen at any time. This isn't a
problem for wayland clients since they don't control their position,
but X clients do and we'd be sending outdated coordinates if a client
is moving a window in response to motion events.
https://bugzilla.gnome.org/show_bug.cgi?id=768039
This layer isn't really being used and in fact, it causes
meta_stack_get_default_focus_window() to return a fullscreen window
even if the naturally topmost window in the stack isn't a fullscreen
one.
Note that commit a3bf9b01aa changed how
we choose the default focus window from the MRU to the topmost in the
stack.
https://bugzilla.gnome.org/show_bug.cgi?id=768221
Since the check for backend->cogl_context was accidentally moved
to clutter_backend_do_real_create_context, the Glib source that
is created at the end of clutter_backend_do_create_context() is
created and added each time create_context() is called, though
create_context() is supposed to be idempotent.
https://bugzilla.gnome.org/show_bug.cgi?id=768243
When restarting (X compositor only, obviously), we want to keep
the same window focused. There is code that tries to do this by
calling XGetInputFocus() but the previously focused window will
almost certainly not still be focused by the time we get to the
point where we call XGetInputFocus(), and in fact, probably was
no longer correct after the previous window manager exited, so
the net result is that we tend to focus no window on restart.
A better approach is to leave the _NET_ACTIVE_WINDOW property
set on the root window during exit, and if we find it set when
starting, use that to initialize focus.
https://bugzilla.gnome.org/show_bug.cgi?id=766243
Emit a signal so that interested parties can recreate their FBOs and
queue a full scene graph redraw to ensure we don't end up showing
graphical artifacts.
This relies on the GL driver supporting the
NV_robustness_video_memory_purge extension and cogl creating a
suitable GL context. For now we only make use of it with the X backend
since the only driver with which this is useful is NVIDIA.
https://bugzilla.gnome.org/show_bug.cgi?id=739178
When using a context with robustness, glGetError() may return
GL_CONTEXT_LOST at any time and this error doesn't get cleared until
the application calls glGetGraphicsResetStatus() . This means that our
error checking can't call glGetError() in a loop without checking for
that return value and returning in that case.
https://bugzilla.gnome.org/show_bug.cgi?id=739178
The call fetching the targets mistakenly used the timestamp meant
to back up the TIMESTAMP atom (hence, it's the timestamp at which
the selection is *owned* by the compositor, on behalf of a wayland
client).
This timestamp is actually only updated when the compositor gets
to own the selection, so it's a randomly late timestamp to retrieve
the TARGETS atom content, which certain clients might end up
ignoring.
https://bugzilla.gnome.org/show_bug.cgi?id=768007
Frames are painted on the frame window according to the GTK+ theme.
Depending on the target's visual, this means either drawing over
a black destination or a fully transparent one. So in cases where
the theme doesn't paint decorations with full opacity, decorations
for windows with an rgba visual look different from those with a
non-rgba visual. Using an rgba visual for all frames independent
from the client's visual can potentially break clients, so our
only option for a consistent appearance is to explicitly initialize
the frame background to black before painting the theme's decoration
on top.
https://bugzilla.gnome.org/show_bug.cgi?id=745060
The GTK+ theme may draw parts of the decorations outside the actual
frame. Since commit f9db65f47f we make sure that the frame is big
enough to account for any overdrawing, however as we still clip the
cairo context to the actual frame before drawing the decorations,
those parts aren't actually painted.
This issue is not very obvious for most frames, as they use a non-rgba
visual where the unpainted parts appear black, which gives the expected
result with many themes once the shape mask is applied (as the mask does
include any overdrawn parts). For frames using an rgba visual however,
unpainted parts are transparent, so any overdrawn decorations are clearly
missing.
Fix this by only clipping out the client area when drawing decorations.
https://bugzilla.gnome.org/show_bug.cgi?id=745060
We ignore all damage while a surface is frozen and queue a full
update instead once it's thawed. While not super efficient, this
isn't overly bad for the intended case of catching up with any
updates that happened during a compositor effect. However when
extended frame sync is used, surfaces are also frozen while the
client is drawing a frame, in which case the current behavior is
pretty damaging (pun intended), as we end up redrawing the entire
window each frame. To address this, keep track of the actual damage
we ignore and apply it when the surface is thawed.
https://bugzilla.gnome.org/show_bug.cgi?id=767798
The result flag needs to be marked as moved even for pending moves,
otherwise the window's unconstrained_rect doesn't get updated in
meta_window_move_resize_internal() and the anchor grab is wrong.
https://bugzilla.gnome.org/show_bug.cgi?id=764180
The X11 backend uses EWMH's _NET_WM_PID to get the PID of an offending
client and kill its PID to force the client to terminate.
The Wayland backend is using a Wayland protocol error, but if the client
is hung, that will not be sufficient to kill the client.
Retrieve the client PID under Wayland using the Wayland client API
wl_client_get_credentials() and kill() the client the same way the X11
backend does.
https://bugzilla.gnome.org/show_bug.cgi?id=767464
xdg-shell allows desktop environments to extend the list of states
within a given range.
Use this possibility to add a new state for tiled so that gtk+ can
benefit from this.
https://bugzilla.gnome.org/show_bug.cgi?id=766860
gjs throws exceptions on non UTF-8 strings which, in some cases, crash
gnome-shell. ICCCM string properties are defined to be Latin-1 encoded
so we can try to convert them to avoid it.
Note that _NET_WM_NAME is defined to be UTF-8 and we already validate
it in utf8_string_from_results() .
https://bugzilla.gnome.org/show_bug.cgi?id=752788
printf string precision counts bytes so we may end up creating invalid
UTF-8 strings here. Instead, use glib's unicode aware methods to clip
the title.
https://bugzilla.gnome.org/show_bug.cgi?id=765535
This makes us behave the same both on bind and when an output
changes. In particular, we were not sending scale and done events on
output changes. We were also unconditionally sending mode events on
output changes even though these should only be sent if there is an
actual mode change.
https://bugzilla.gnome.org/show_bug.cgi?id=766528
Stacking hidden X windows below the guard window is a necessity to
ensure input events aren't delivered to them. Wayland windows don't
need this because the decision to send them input events is done by us
looking at the clutter scene graph.
But, since we don't stack hidden wayland windows along with their X
siblings we lose their relative stack positions while hidden. As
there's no ill side effect to re-stacking hidden wayland windows below
the X guard window we can fix this by just doing it regardless of
window type.
https://bugzilla.gnome.org/show_bug.cgi?id=764844
If we try to send notify event (either from surface_state_changed()
or from meta_window_wayland_move_resize_internal()),
we will crash, because we don't have a sufrace anymore.
There's no reason why to resize the window that is being
unmanaged anyway.
https://bugzilla.gnome.org/show_bug.cgi?id=751847
meta_parse_accelerator() considers 0 length accelerator strings as
valid, meaning that the keybinding should be disabled. Unfortunately,
it doesn't initialize the MetaKeyCombo so if the caller doesn't
initialize it either, we end up using random values and possibly
grabbing random keys.
https://bugzilla.gnome.org/show_bug.cgi?id=766270
The 'select-all' action is currently only bound to <ctrl>a, which makes
it awkward to use when caps-lock is active, and is inconsistent with GTK+.
Just accept both upper- and lower-case variants.
https://bugzilla.gnome.org/show_bug.cgi?id=766326
All the upper layers are prepared for multiple onscreen cursors, but
this. All MetaCursorRenderers created would poke the same internal
MetaOverlay in the stage.
This will lead to multiple cursor renderers resorting to the "SW"
rendering paths (as it can be seen with tablet support) to reuse the
same overlay, thus leading to flickering when a different
MetaCursorRenderer takes over the overlay.
Fix this by allowing per-cursor-renderer overlays, their lifetime
is attached to the cursor renderer, so is expected to be tear down
if the relevant device (eg. tablet) disappears.
meta_wayland_tablet_manager_update()/handle_event() are called before
the MetaWaylandSeat counterparts. If the event comes from a device
managed by MetaWaylandTabletManager, the event will be exclusively handled
by it.
Each tool has its own MetaCursorRenderer instance, which is created/destroyed
upon proximity, and possibly updated through focus and set_cursor calls in
between.
This struct keeps the server side information for the wl_tablet_manager
global resource. It keeps the clients requesting this interface, and
does keep track of the plugged tablet devices, so
wl_tablet_manager.device_added is emitted on the expected clients.
Move into a standalone meta-wayland-surface-role-cursor.[ch], and
make generic enough to work for pointe and additional (eg. tablet)
cursors.
Most notably, the sprite is now kept completely internal to the
cursor role, and updates are routed through the given
MetaCursorRenderer (which may be the default one for the pointer,
or something else).
The way cursor updates after cursor surface destruction has also
been reworked, the pointer will just keep track of the last cursor
surface, so older surfaces being destroyed don't trigger pointer
rechecks/updates.
There's places where it would be convenient to add listeners to this,
so add the signal. The signal is only emitted once during destruction,
it is convenient for the places where we want notifications at a time
the object is still alive, as opposed to weak refs which notify after
the fact.
The device name is something more natural, similar to what's seen
in X11, the sysname is rather the event node basename, which may
also vary depending on device insertion/detection time.
Sadly, GLib's autoptr cleanup macros cannot be detected by the C
pre-processor, because they generate a function. This means that we are
forced to bump up the dependency on GLib 2.49, in order to build against
a newer version of gdbus-codegen.
Starting from GLib 2.49, the gdbus-codegen tool automatically generates
the auto cleanup symbols for the GDBus proxy and skeleton interfaces.
Since we don't depend on a specific version of GLib we need to
conditionally generate the auto cleanup symbols in case an older version
of gdbus-codegen is used when building Mutter.
This commit unbreaks the build under GNOME Continuous, which has been
failing with:
usr/include/glib-2.0/glib/gmacros.h:415:43: error: redefinition of 'glib_autoptr_cleanup_Login1Session'
#define _GLIB_AUTOPTR_FUNC_NAME(TypeName) glib_autoptr_cleanup_##TypeName
^
[...]
/usr/include/glib-2.0/glib/gmacros.h:415:43: note: previous definition of 'glib_autoptr_cleanup_Login1Session' was here
./meta-dbus-login1.h:82:1: note: in expansion of macro 'G_DEFINE_AUTOPTR_CLEANUP_FUNC'
G_DEFINE_AUTOPTR_CLEANUP_FUNC (Login1Session, g_object_unref)
^
Separate "xdg_surface", "xdg_popup" and "xdg_shell" related functions
into three sections. Prior to this, the "xdg_shell" part was a bit all
over the place.
https://bugzilla.gnome.org/show_bug.cgi?id=763431
Dismiss the popup when the parent is destroyed, and do this in the
destructor of the parent object. This makes the parent destory listener
unnecessary, since we already handle the parent child unlinking
explicitly in the object destructor.
https://bugzilla.gnome.org/show_bug.cgi?id=763431
Instead of relying on destroy signals attached to the corresponding
role object, let the roles explicitly dismiss the popup when it should
be dismissed.
https://bugzilla.gnome.org/show_bug.cgi?id=763431
Add a bridge between the MetaWaylandPopup object and the corresponding
popup surface role. This bridge replaces communicating dismissed and
unmapped popup events.
https://bugzilla.gnome.org/show_bug.cgi?id=763431
The only time the surface pointer (priv->surface) may be NULL is when
the surface is unmanaged but still painting, possibly due to a unmap
animation or the like, so only guard handle this situation in the entry
points that may come from the stage painting.
https://bugzilla.gnome.org/show_bug.cgi?id=763431
Before this commit, on Wayland, the buffer rect would have the size of
the attached Wayland buffer, no matter the scale. The scale would then
be applied ad-hoc by callers when a sane rectangle was needed. This
commit changes buffer_rect to rather represent the surface rect (i.e.
what is drawn on the stage, including client side shadow). The users of
buffer_rect will no longer need to scale the buffer_rect themself to
get a usable rectangle.
https://bugzilla.gnome.org/show_bug.cgi?id=763431
A large part of meta_wayland_surface_apply_window_state() was only
relevant for xdg_surface. Make this more obvious by splitting it up,
moving the relevant parts to the relevant roles.
https://bugzilla.gnome.org/show_bug.cgi?id=763431
Move xdg_shell related functionality to a new meta-wayland-xdg-shell.c
and wl_shell related functionality to a new meta-wayland-wl-shell.c,
and adapt role object tree.
Common functionality related to the surface being drawn as a
MetaSurfaceActor was moved to a MetaWaylandSurfaceRoleActorSurface role.
The subsurface role GObject is made to inherit the actor surface GObject.
Shell surface hooks (configure, ping, close, popup done) were added to
a MetaWaylandSurfaceRoleShellSurface GObject which inherits the
surface actor role GObject.
The shell surface roles (xdg_surface, xdg_popup, wl_shell_surface) are
made to inherit the shell surface GObject and implement the relevant
API.
https://bugzilla.gnome.org/show_bug.cgi?id=757623https://bugzilla.gnome.org/show_bug.cgi?id=763431
2016-05-01 21:11:27 +08:00
770 changed files with 52393 additions and 28280 deletions
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
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.
