The internal representation of the min/max width of windows include what
is outside of the window geometry, so when the window geometry changes,
but the min/max size did not in the same commit, we'd be left with an
out of date min/max size, potentially causing windows to shrink when
configured.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3755>
local_error in meta_egl_query_device_string() is using g_autoptr,
meaning that it was getting freed after g_propagate_error(). This then
would result in error->message becoming invalid, causing crashes when
logging the error message later on.
Fixes: 8234f5bc7 ("egl: Return success status from meta_egl_query_device_string")
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3758>
Just shovel the data through our own stdin/stdout, which will end up at
the right place (e.g. /dev/null).
This should hopefully solve `mutter-dist` failing due to a D-Bus method
call timeout in CI.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3757>
An action can be performed both on a signal emitted by a window or by
a well-known signal name, so that one can do:
[window-id]::signal => command
And so these would work:
w/1::position-changed => resize w/2 20 30
::monitors-changed => move w/1 30 40
We only support "monitors-changed" global signal name or connecting to
window signals for now, but this can be easily expanded adding support
for more global signal values or parsing other kinds of object instances
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3701>
Allows to use:
set_strut 0 0 MONITOR_WIDTH*0.5 MONITOR_HEIGHT/4 top [default]
We use the currently focused window as the reference monitor if any or
the primary one if no monitor is explicitly provided.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3701>
We may need to compare monitor size to fractions and while it can be
done using inverse divisions, they were not supported as we didn't parse
floating point values.
So, just add support for both multiplications and divisions, so that we
can easily do stuff like `MONITOR_HEIGHT*3/4`.
Additions would be easy to support too if we don't care about operator
priorities, but that's out of scope for now.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3701>
We just made an half assed attempt to wait, but e.g. when clients were
waiting for a frame event, the attempt was not enough, as the clients
would not ack any configure until the time they were scheduled to paint
again.
Fix this by actually tracking newly pending configurations, and waiting
for them to return.
On X11, still be lazy and hope for the best, as there is no similar
configuration tracking there.
Some tests were updated to use just 'wait' when there was no actual need
for any extra waiting.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3701>
When a window with an input shape on its decoration window becomes
undecorated and meta_window_x11_update_input_region() gets called via
notify::decorated, the buffer_rect of the window has not been updated
yet while the decorated property has. This would lead to us comparing
the input shape of the client window to the buffer_rect which still
includes the decoration window. This would fail to detect the common
case when the client window has no input shape set, leading to the input
region being set to the size of the client window rather than NULL. If
the window is then resized later, the input shape would remain at the
previous size.
This was not a problem before 6bd920b35, because then we were (wrongly)
always comparing to the client_rect.
Fix this by choosing the correct rect for comparison depending on
whether the window is decorated.
Fixes: 6bd920b35 ("x11/window: Use correct bounding rect to determine NULL input region")
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3451
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3720>
There is nothing to allocate for a 0-sized files, and indeed
posix_fallocate() will error out if the passed len isn't greater
than 0.
Now that anonymous files are used to back the memory selection
source, this fixes unsetting the selection when the screen is
locked.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3752>
When a key binding is removed, and a trigger key sequence is dispatched
before the idle callback that resolves and updates the actual binding,
we should handle that gracefully.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3711>
Two new fields: ref_count and removed, are added to MetaKeyHandler, and
it would be freed only if the ref count has reached 0. When handler is
removed from key_handlers GHashTable, key_handler_destroy() would mark
removed as TRUE, and do an unref. handler->removed is checked in
get_keybinding, and binding with handler removed would not be used.
Also in MetaKeyBinding, it now has the ownership of the name field, to
avoid it being freed before logging. Create or copy a binding would
do a ref inc for handler, and free one would unref handler.
Fixes https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/1870.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3711>
When a client resizes on its own, make sure the new size is passed
through the window constraints machinery directly, to trigger any
potential window management rule that might apply.
Fix a couple of tests to make use of this behavior by introducing a new
'wait_size' command that waits until a window has been resized to a
expected size.
This replaces the fix introduced in 0e736af301 ("window: Ensure
constraints after a Wayland client resize").
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3700>
Starting the timeout to move from hidden to suspended before the window
is mapped means we don't have a previous window configufration which we
need to get the new window configuration with the suspended state.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3731>
These functions are specific for X11 windows only so we should check
if the passed window is an X11 one, not just a MetaWindow since we're
casting to the actual type at later point.
Fixes changes part of commit e1e6534eb
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3727>
If a window sends a configure stacking request, we were comparing the
active window with the event window even though they were different
client types (e.g. wayland and x11).
This was leading to a critical error, so let's handle this by ensuring
that the active window is of the same kind of the event window before
doing x11-specific checks. Behaving as different applications in case.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3727>
Because `meta_kms_impl_device_simple_initable_init` is called in the
middle of `meta_kms_device_new`, the crtcs list for `MetaKmsDevice`
has not been populated yet. And thus the loop to detect missing
cursor planes and create fake ones never iterated. But the crtcs list
does already exist in `MetaKmsImplDevice` so iterate over that instead.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3264
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3676>
Prior to commit 5dfed8a431, the MetaWaylandKeyboard would always remember
the last key press serial, and consider it valid after the key was released,
as long as no other key presses/releases happened in between.
That commit improved things so that MetaWaylandKeyboard can track multiple
keys being pressed simultaneously, but also changed so that the serial for
a key press is immediately forgotten after the key press event was received.
This may break in situations like testing or keyboard macros where key
press and release is handled in a quick sucession, so the client reaction
to the key press (e.g. popping up a menu) might arrive too late.
Add a sort of spiritual successor to this handling, and make keyboard
press serials corresponding to the last key up forgotten at the next
key press/release received.
Fixes: 5dfed8a431 ("wayland: Preserve serial for all pressed keys")
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3458
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3721>
We did not track the current surface (i.e. the logical focus) too
thoroughly, so there might be chances that a stale surface pointer
here becomes the focus. Track its destruction (like it's done at e.g.
MetaWaylandPointer) and unset the current surface early, in order
to avoid possible invalid memory access.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3372
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3715>
It seems only the iconcache used to use it, but this is gone since
commit d16ddc42ce.
Even before that, the Xrender usage was removed in commit 556e7694de,
albeit leaving a redundant include <X11/extensions/Xrender.h> in its
place then, which comes from libXrender.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3716>
Fix an obvious copy paste error that slipped through the cracks.
Fortunately it doesn't have a visual impact for well behaving clients
but only makes us not hit direct-scanout paths, assuming no other bugs
in the stack.
Fixes: f21762ea6e (wayland: Add support for preferred_buffer_scale/transform)
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3717>
This call is meant to replace meta_wayland_keyboard_get_focus_client(),
since we will always have a MetaWaylandSeat with an input focus (or not),
but we may or may not have a MetaWaylandKeyboard.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3707>
When there is a new owner but there is no matching mime type we clear
the saved mimetype and the saved clipboard but an outstanding async
meta_selection_transfer_async can set the saved clipboard.
Abort the async transfer when we have a new owner.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3678>
The precondition checks in meta_selection_source_memory_new can return
NULL if the mimetype is NULL but callers expect the error to be set when
NULL is returned.
Let's just make sure we never call it with a NULL mimetype.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3678>
The current code checking keyboard serials for popup/grab
validation is a bit simple, tracking one key press exclusively.
This may break expectations if a client uses a serial
corresponding to a previous key that is still pressed.
Keep track of the serials corresponding to all pressed keys,
and ensure these are reset across focus changes, since the
validity of those serials is already outdated. The code does
still keep track of a single (last) key release serial, since
the validity lifetime is somewhat underdefined with those if
we keep track of multiple keys simultaneously.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3267
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3644>
Instead of initializing to 'suspended', which will send the `SUSPENDED`
xdg_toplevel state, set it to hidden at first. If the window is placed
on an inactive workspace, it'll eventually enter the 'suspended' state,
but will have had some time in non-suspended state to get map, even if
not visibly.
This fixes inital suspended state when mapping a window maximized.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3229
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3475>
This, in contrast to meta_window_should_be_showing() reports whether a
window should be showing despite not being showable. This is useful to
know the intended visibility state that should happen in the immediate
future.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3475>
d991961ae changed this code from client rect to buffer rect to account
for the fact that for SSD windows the decorations are now included in
the bounding region, but it kept the variable name as client_area and
the comment was also still referring to the client rect.
Fixes: d991961ae ("x11: Use input region from frame window for decorated windows")
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3697>
When updating the input region we check whether the input shape reported
by XShape matches the bounding region of a window to determine when it
was not set by the client. We would then use a NULL input region instead
which always matches the full size of the window. The code however was
using the client rect for this comparison, which does not include the
window frame. Since d991961a the frame is considered part of the input
region.
This meant that for SSD windows where the input region would match the
bounding region, we would not detect that and fail to set the input
region to NULL, but instead set it to the reported input shape.
Usually this would not be the case due to the GTK frame window having
shadows and a resize region, but in the presence of an issue that causes
GTK to wrongly detect _GTK_FRAME_EXTENTS as not supported, GTK would not
draw shadows or set an input shape. And due to GTK not updating its
input shape, there would be no further calls to
meta_window_x11_update_input_region() after the initial one.
The input region would therefore remain at the fixed size from the
initial call. This was causing windows to become click-through outside
of the region corresponding to their initial size after being resized.
Fixes: d991961ae ("x11: Use input region from frame window for decorated windows")
Related: https://gitlab.gnome.org/GNOME/gtk/-/issues/6558
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3404
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3697>
Right now we store touch serials on their according MetaWaylandTouchInfo
entries. These entries are gone as soon as the touchpoint ended though, and
it's not unlikely that clients will respond to that touch-end event after we
removed the touchpoint.
In this case we currently can't match the client provided serial to any of
our known touch sequences, which causes xdg_popup grabs that get requested
shortly after the touch-end to fail.
Let's be a bit more gentle on clients here and store the latest touch-down
serial on the MetaWaylandTouch, so that it continues to be around after the
touch-end and we can match the serial of the xdg_popup_grab() as expected.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2946>
In the rare event that hotplugs destroy and then create a new EGLContext
with the exactly the same ID, this ensures we will forget the old program
which presumably wouldn't work in the new context. It will be recreated.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3304>
The EGL context can only import and blit an EGLImage if the
backing DMA buffer has a format modifier combination that is advertised
as supported and not marked as "external_only".
When the context can't blit the imported image, we can still paint using
it GL_OES_EGL_image_external using the texture target
GL_TEXTURE_EXTERNAL_OES.
However, treat drivers who doesn't support modifiers at all as if they
do support blitting, if the modifier is 'linear', to avoid regressions.
[jadahl: Make shader path a fallback to allow hardware to utilize copy
engines via blitting]
Closes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/6221
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/2247
Related: https://launchpad.net/bugs/1970291
now only falls back if modifiers are supported, and they mark linear as
export only.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3304>
This is a critical part of any OpenGL program. Mesa allowed us to get
away without it and provided a sane default of the full buffer, but
Nvidia seems to default to an empty/zero viewport so would refuse to
paint any pixels.
In the OpenGL ES 2.0 spec this is ambiguous:
> In the initial state, w and h are set to the width and height,
> respectively, of the window into which the GL is to do its rendering.
because the first "window" used is EGL_NO_SURFACE in
init_secondary_gpu_data_gpu. It has no width or height.
In the OpenGL ES 3.0 spec the ambiguity is somewhat resolved:
> If the default framebuffer is bound but no default framebuffer is
> associated with the GL context (see chapter 4), then w and h are
> initially set to zero.
but not entirely resolved because neither spec says whether
EGL_NO_SURFACE should be treated as zero dimensions (Nvidia) or ignored
and not counted as the first "window" (Mesa).
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3304>
As mentioned in the OES_EGL_image_external spec, there is no implicit
sync between the EGLImage producer and consumer. And in this code path
we don't have meta_drm_buffer_gbm_new_lock_front on the primary GPU to
do it for us either. So synchronization has to be done manually or else
the secondary GPU is likely to get an unfinished image.
This problem has only been observed when the secondary GPU is using the
Nvidia proprietary driver.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3304>
When the monitors change meta_window_update_for_monitors_changed is
called which is responsible for updating window->monitor. It however can
go through the entire window placement and constraint machinery before
it's able to do so. In this period window->monitor points to the old
MetaMonitor where the monitor number doesn't reflect the index into the
MonitorManager anymore.
Avoid relying on the window->monitor->number and go through the Monitor
directly.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3402
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3691>
In every other device and circumstance, we update the logical focus
(i.e. the surface that would be focused, if no other circumstances
applied) but let the MetaWaylandInput figure out the surface that
should be effectively focused, if at all. This is where we apply the
actual compositor state (e.g. grabs) and result in an effective focus
window.
The only exception to this is `meta_wayland_seat_set_input_focus()`
where the logical focus for keyboards and related devices is set, but
also applied as the effective surface.
We should do the same as everywhere else, and let MetaWaylandInput
focus determine whether the logical focus is also the effective focus.
These replaced set_focus() internal calls will happen through the
default MetaWaylandEventInterface.focus implementation in MetaWaylandSeat.
This resulted in keyboard focus being set on windows in circumstances
it ought not to, like in the overview. Actual key events were never
sent in these circumstances, but changes to modifier state could.
Closes: https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/7528
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3704>
If a modal dialog (i.e. with a "revolves around center of parent window" policy)
becomes fullscreen, we cannot neatly honor both the modal dialog going fullscreen
and the window staying around the center of its parent.
In order to make fullscreening work in this situation, allow fullscreen modal
dialogs to "snap out" of the parent. This rule will become again effective after
the window is unfullscreened.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3425
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3695>
By ignoring obcured surfaces we can generalize the single-pixel-buffer
background check to work with arbitrary surface trees, as long as the
two top ones are the background and content surfaces.
As a nice side effect of this, clients like Firefox now usually take the
shorter `meta_window_is_fullscreen (window) && n_visible_surface_actors == 1`
route.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3699>
Move the code out of cogl_onscreen_egl_swap_buffers_with_damage, and
call the new function from callers of the former.
v2:
* Use early return if the cogl context doesn't support timestamp
queries. (Sebastian Wick)
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3689>
`mtk_rectangle_new()` allocates the object dynamically,
but in the "contains-point" test case the allocated object
was not freed. Fix that by creating the object on the stack.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3679>
Avoids using a static and make sure things are properly freed by
using g_autoptr. Also take the opportunity to add some missing NULL
initialization for auto-pointers variables
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3687>
In consistence with the code style, and in order to fix build errors
with older clang:
../src/backends/native/meta-onscreen-native.c:521:7: error: expected expression
521 | graphene_rect_t src_rect;
| ^
../src/backends/native/meta-onscreen-native.c:529:39: error: use of undeclared identifier 'src_rect'; did you mean 'dst_rect'?
529 | &src_rect);
| ^~~~~~~~
| dst_rect
../src/backends/native/meta-onscreen-native.c:522:20: note: 'dst_rect' declared here
522 | MtkRectangle dst_rect;
| ^
And warnings with newer clang:
../src/backends/native/meta-onscreen-native.c:521:7: warning: label followed by a declaration is a C23 extension [-Wc23-extensions]
521 | graphene_rect_t src_rect;
| ^
This should allow the build for coverity to succeed again.
Fixes: adc776d0d7 ("crtc/kms: Pass on src and dst rects to primary plane assignments")
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3686>
This implements the explicit sync protocol linux-drm-syncobj-v1. This
works by importing a DRM syncobj timeline and importing/exporting fds
to/from the sync points on the timeline corresponding to buffer acquire
and release. We take fds for sync points provided during a surface
commit and use them to delay transaction application, and fetch fds
from Cogl to signal when we are done using a particular buffer.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3300>
This adds the explicit sync wayland protocol to the list of build
dependencies.
This adds a copy of the linux-drm-syncobj protocol that we will use
privately for builds. This avoids the pain of requiring wayland-protocols
1.34 for distros. This commit can be reverted once we want to use
linux-drm-syncobj from the system wayland-protocols.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3300>
On big endian architectures the mapping of drm formats to cogl
formats is significantly trimmed compared to on little endian
architectures.
meta_wayland_init_shm tries a bunch of formats, including some
that just aren't mapped on e.g. s390x.
The code asserts the mapping will exist, however, leading to
crashes when Xwayland starts.
This commit makes failure to find a mapping non-fatal.
Suggested by Jonas Ådahl.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3672>
When logging in from gdm to gnome, the main plane is deactivated, and
leads to the screen going blank before gnome is able to enable it
again.
Using the new CloseFB ioctl, allows to keep the gdm login screen
displayed until gnome-shell replace it.
Signed-off-by: Jocelyn Falempe <jfalempe@redhat.com>
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3659>
This also gets rid of the MetaFrameSyncMode enum and instead issues a
VRR update when the requested state differs from the CRTC state.
Fixes: fee33299 ("onscreen/native: Allow requesting frame synchronization")
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3646>
This includes checking the vrr_capable property on the connector as well
as the VRR_ENABLE property on any CRTC the connector might get assigned
to. Also takes into account when a GPU is tagged for broken VRR support.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3646>
As we are getting close to have a Wayland-only build, guard those usages
so we don't 'regress'.
Once Kiosk figures it use case, we can revert both this commit and the
previous one.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3652>
When multiple configure requests sent to the wayland client within a sort period, maximized windows may end up in wrong position and cover struts. To avoid this, queue a resize when the resize event sent by the wayland client results a changed size or position to ensure that the final size and position will be always right.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3601>
Using `meta_wayland_single_pixel_buffer_from_buffer()` can fail if
the buffer resource is already gone. Given that we release the buffer
immediately, that is actually expected - but the client this was tested
with did/does not do so.
This made the single pixel black bar optimization fail for
GstWaylandsink, while working for the mpv Wayland backend.
Access the object directly, which is both faster and works as long as
the buffer is alive.
Fixes: ed50cbbfe4 (window-actor/wayland: Update scanout candidate check)
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3662>
The gnome-shell magnifier listens to the `visibility-changed` signal
and calls meta_cursor_tracker_set_pointer_visible(false) when the
cursor became visible.
This leads to a reentrance in meta_cursor_tracker_set_pointer_visible()
and clutter_seat_uninhibit_unfocus() gets called twice, once from the
meta_cursor_tracker_set_pointer_visible(false) by the magnifier and then
the original meta_cursor_tracker_set_pointer_visible(true) continues,
after the first call has set is_showing to false again. This breaks the
inhibitor counting and the ability to use the cursor while using the
magnifier.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3661>
Near window destruction, there might be cases where the surface
actor does no longer have a surface, yet it's still in the stage
and eligible for picking. In that situation looking for modal
dialogs attached to this surface will evidently fail, so avoid
this check on a NULL surface.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3332
Fixes: 93a9e7f3f ("core: Move code ignoring events on windows with modals to Wayland")
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3660>
Since commit e30eb78891 `ClutterFrameClock` assumes that a valid CPU time
implies timestamp query support, which is also checked in
`cogl_onscreen_egl_swap_buffers_with_damage()`.
Unconditionally setting the CPU time on direct scanout meant that the
compositing path would be stuck on the last (direct scanout optimized)
result on GL implementations without timestamp query support since.
be0aa2976e (clutter/frame-clock: Avoid rapidly toggling dynamic max render time)
Fix that by explicitly marking the gpu rendering duration as valid when
querying the GPU timestamps is supported and check for it ClutterFrameClock.
Fixes: 56580ea7c9 ("backends/native: Assume zero rendering time for direct scanout buffers")
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3655>
This makes sure that xdg-output.logical_size and
xdg-output.logical_position are only sent when they actually changed.
There should be no behavior change in wl_output_transform_from_transform
but it now uses the same technique of tracking the protocol state and
comparing it to the current state to compute which properties have
changed.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3622>
X11 server side focus changes, such as when a focus change was requested
by mutter for a globally active window, did not go through
meta_display_set_input_focus(), which is responsible for emitting the
`focus-window` signal. Since this signal is what triggers the display
server specific code to handle focus changes, this was leading to a
problem on Wayland where the focus remained on the last active Wayland
window when the focus got changed to a globally active XWayland window.
This commit now changes handling X11 server side focus changes to also
go through the code path that emits the signal while making sure to not
trigger another focus change and keeping the same serials as the
previous code to not interfere with future focus changes.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3328
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3651>
MetaSelectionSourceMemory currently uses GBytes for its underlying data.
This can cause memory overhead when large items, such as HD images, are
stored in the clipboard. This commit changes the underlying data
structure to a MetaAnonymousFile object, which writes to memfd instead
of heap. When reading, MetaSelectionSourceMemory will create a
Gio.UnixInputStream from the file descriptor generated by
MetaAnonymousFile. We subclass the UnixInputStream as
MetaUnixInputStream, to override the stream's close_fn function so
that it invokes meta_anonymous_file_close_fd when the stream terminates.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3551>
Which got introduced in wl_compositor version 6.
Note that if the surface is visible on multiple monitors with different
transforms, we pick the transform of the monitor which we choose for the
scale as well. This doesn't really matter at the moment, as the
transform is only really relevant for direct-scanout - which we
currently only support for fullscreen clients.
Once we support direct-scanout for partially visible clients we'll
likely want to introduce a more sophisticated algorithm.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3580>
Some clients - notably the Gstreamer vapostproc element when using Intel
GPUs - only support BGRA, not BGRx. We already assume that we can
support this format for window screen casts, and even in case of failure
we now have a re-negotiation fallback in place. Thus it's pretty safe to
support it for all screen cast types.
The possible duplication in case of window screen casts doesn't seem to
be a problem for either Pipewire or existing clients like OBS.
Note that the implementation lays the foundation to make it easy to add
more formats in the future.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3617>
The POINTER_EMULATED flag was a convenience to filter either
side of smooth/discrete events that we should ignore based on
the source.
This distinction was challenged, first by v120 mice that use
Clutter smooth events to deliver semi-discrete changes, second
by commit e0c4b2b241 ("backends/native: Mark the emulated smooth
scroll event as such") which made the smooth events be flagged
as emulated, and the discrete whole-step events marked as
real.
This distinction feels convenient for the time being, since
upper layers might be confused by real smooth scroll events
without finish flags. Adapt to this change at MetaWaylandPointer
so that we drop the POINTER_EMULATED check, and the events are
perhaps filtered based on their source and the preferred
wl_seat version of the client that we are talking to.
This handles the whole grid of combinations:
- wheel sources with wl_seat >=8 result in wl_pointer.axis_value120
from "emulated" smooth scroll events, with value120 information.
- wheel sources with wl_seat < 8 result in wl_pointer.axis_discrete
from "real" discrete scroll events.
- finger/continuous sources prefer smooth events. Previously, always
non-emulated for those.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3642>
The wl_pointer.axis_discrete axis (applicable to wl_seat <= v8) is
meant to be sent together with wl_pointer.axis events in the same
frame. And the wl_pointer.axis_value120 event replaces it in
wl_seat >= v9, but has the same relation with the other information
available in a frame.
This emission should not be conditional to anything, so drop the
various checks leading to maybe sending wl_pointer.axis or not.
This fixes emission of wl_pointer.axis in conjunction with discrete
events, for some combinations of versions and (non)value120 mice.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3642>
Unveiled by commit e0c4b2b241 ("backends/native: Mark the emulated smooth
scroll event as such"). The sudden "lack" of smooth scroll events (Used by
Clutter to forward v120 events) made it evident we silently ignore Clutter
discrete events, as we don't send wl_pointer.axis_value120 for these.
Fix this by assigning a value120 value to discrete scroll events. This
makes wl_pointer.axis_value120 events actually sent on non-v120 mice.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3642>
meta_window_handle_ungrabbed_event() triggers the show of the window menu in
gnome-shell via meta_window_show_menu() on hold of Meta + right mouse button
click.
Since meta_display_handle_event() was refactored lately and now forwards a
lot more events to Clutter (including the one triggering the window menu),
gnome-shell now sees this event after the menu has opened, figures that the
source-actor is outside of the menu, and immediately closes the menu again.
This is the correct behavior from the PopupMenuManager on the gnome-shell
side, it is the responsibility of the event handler that opens the menu (aka
meta_window_handle_ungrabbed_event()) to return CLUTTER_EVENT_STOP and stop
event propagation.
So fix this issue by adding a return value to
meta_window_handle_ungrabbed_event() and stopping event propagation in case
the event opened the window menu.
While at it, also return CLUTTER_EVENT_STOP for events triggering window
drags, so we can drop the extra check for that in
meta_display_handle_event().
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3623>
For each libinput scroll event we generate two clutter events
(continuous and discrete), one of them marked as emulated. libei
explicitly specifies that emulation of scrolling must be done in the
client (if desired) so drop the emulated one.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3637>
`meta_window_set_user_time()` will not update the window's
user time if it timestamp in the argument is before the
currently saved timestamp. However, when trying to work around
problematic timestamps, this is exactly what needs to be done.
So force the update to happen by setting the "is user time set?"
flag to false.
Fixes: 8f3da9f68a ("Use meta_window_set_user_time for setting user time consistently")
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3027>
priv->current_drag is the property that
meta_compositor_get_current_window_drag() and therefore also
meta_display_is_grabbed() bases its check on. We use
meta_display_is_grabbed() to select which cursor to use (window cursor
vs root cursor) when updating the cursor in MetaCursorTracker.
Since meta_window_drag_begin() sets a new cursor, and therefore triggers the
cursor tracker to update the current visible cursor, we should set
priv->current_drag before the call to meta_window_drag_begin(). This makes
sure the cursor tracker sees that there's a window drag and changes the
cursor right away when the window drag begins (instead of doing so on
subsequent pointer events).
Fixes: 525ed1166c ("wayland/pointer: Unset current surface during window drags")
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3630>
This is used as the minimum refresh rate in the variable refresh rate
range.
This value is expected to be found in the DisplayPort and eDP EDID of
every monitor that supports variable refresh rate.
It is also found in the HDMI EDID of some monitors that support
variable (FreeSync), but most likely not all of them. The rest require
parsing the AMD vendor extension which libdisplay-info doesn't
support.
No fallback is implemented for cases where libdisplay-info support is
disabled.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3576>
Find a surface actor that meets the criteria to drive the refresh
rate for the view. If a new surface actor is found, add a
reference to it and request frame synchronization to be enabled
for the relevant MetaRendererViewNative.
Whenever the surface actor schedules a repaint or an update, and
in case frame synchronization is enabled for the
MetaRendererViewNative, schedule the update to occur "now". This
effectively makes the surface actor's frame rate drive the refresh
rate of the monitor.
If the actor is frozen or destroyed, it is no longer expected to
schedule repaints or updates and should stop driving the refresh
rate.
When there is no surface actor to drive the refresh rate, request
frame synchronization to be disabled for the MetaRendererViewNative.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1154>
This makes all stage updates that result from applying the pending
state of a Wayland surface emit an "update-scheduled" signal in the
context of the relevant surface actor.
A common case where an "update-scheduled" signal is needed is
when applying "empty" client commits. In this case a
"repaint-scheduled" signal would not be emitted since the commit
doesn't trigger a repaint. However, it is still important to add
handling for such updates with variable refresh rate when the
releavnt actor is also driving the refresh rate.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1154>
This function allows scheduling a stage update in the context of a
surface actor and emit the "update-scheduled" signal. This signal is
similar to "repaint-scheduled", but can be used to be notified of
updates that do not necessarily result in a repaint.
With variable refresh rate, any update potentially affecting the
frame pacing of a surface actor needs to be handled differently
depending on whether that surface actor drives the refresh rate or
not.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1154>
Frame synchronization is enabled for a view as long as it's
applicable to be enabled. It is considered applicable if it's both
requested for the onscreen and if the onscreen uses a CRTC which is
configured with a variable refresh rate mode.
When frame synchronization is enabled, it enables both the the variable
scheduling mode of ClutterFrameClock and the variable refresh rate
property for the CRTC.
Changes in the frame synchronization mode are applied asynchronously,
before the next frame is drawn.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1154>
Add variable refresh rate output modes for connectors that are VRR
capable when VRR is not disabled for the GPU.
Variable refresh rate output modes are sorted before their fixed
refresh rate counterparts. They are also marked as the preferred mode
for the output between the two.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1154>
When VRR is not disabled for a GPU, create two variants of every
display mode: one with fixed refresh rate and another with variable
refresh rate.
The variable refresh rate modes are not used yet. They will be used
in a following commit.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1154>
This can be used to disable VRR in specific drivers and hardware
combinations where it is found to be problematic.
No default rules are added for now to encourage testing.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1154>
Require the "variable-refresh-rate" keyword under the
"experimental-features" gsetting to enable the feature for now.
It would no longer be required once the experience with variable
refresh rate is good enough for general use and handles all common
use cases well.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1154>
Use a sufficiently low refresh rate to calculate the CRTC deadline
when variable refresh rate is enabled. This is done to avoid cursor
updates from driving the monitor refresh rate.
It's not great solution and is sometimes not enough, but it avoids
stutter in the main content as a result of cursor movement in most
cases.
The unfortunate downside of this approach is that cursor movement
would usually only update with the main content and would not be
smooth when the main content updates are not frequent enough.
A better solution may use an approach similar to LFC (Low Framerate
Compensation) to insert cursor-only updates between updates of the
main content, but achieving adequate results with an approach of this
nature requires more research and experimentation.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1154>
It is not trivial to accurately estimate the expected presentation
time with variable refresh rate, and not doing so only affects debug
prints.
No change in behavior for now because the expected presentation time
is always calculated. A following commit will introduce a case where
it is not.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1154>
Add functions to update and monitor the value of the "VRR_ENABLED"
KMS property.
This requires the addition of functions to process CRTC property
updates in both the atomic and the simple KMS backends. The
implementation is similar to the implemention of processing
connector updates.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/1154>
Commit d48129f5ee broke the displaying of our own drag cursor during window
drags, as now the window cursor is always used, even during stage grabs
(window grabs are just a kind of stage grab).
To fix it, while not regressing on the intention of the other commit, unset
the MetaWaylandPointer surface in case a window drag is active (instead of
all kinds of grabs) by checking via meta_display_is_grab().
Fixes: d48129f5ee ("wayland: Fix pointer cursor during Wayland grabs")
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3316
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3624>
Move the bulk of the implementation inside MetaWaylandPointer
files, like it happens in other places (e.g. MetaWaylandTabletSeat).
This avoids MetaWaylandPointer struct peeking from outside.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3627>
The _MUTTER_FRAME_EXTENTS need to have the correct size before the
window is shown, otherwise the client window size will effectively have
the headarbar height subtracted when initially syncing the frame/client
geometries.
This reverts commit f10b3eac62.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3628>
QueryDeviceString can successfully return NULL. The convention however
is that when NULL is returned, the error will be set.
This commit makes the returned string an output parameter which allows
us to return the success status and have the error set accordingly.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3614>
The convention is that when a function returns FALSE or NULL, the error
will be set. In this file we call set_egl_error but it might not set an
error. Code of the form
if (egl_do_thing (..., &error))
use_error (error);
will crash in those cases.
This commit makes sure we always set the error even if EGL doesn't give
us an error.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3614>
We depend on Gio for many introspection features (GSettings, GFile...)
without this we'd end up not having many symbols being exposed.
And errors such as
Warning: Meta: meta_display_add_keybinding: argument settings:
Unresolved type: 'GSettings*'
As per previous commit we now have this dependency implicitly, but still
it's better to make it clearer in case Clutter would drop it.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3620>
This is added as a hint so that the caller can optionally use it, e.g.
in the case that a window drag is driven by the mouse. This will be used
to more accurately follow the drag position, other than starting at the
current pointer position.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3546>
Pointer constraints obey to logical pointer focus changes, and looking
up the current surface in order to maybe enable them is getting ahead
of itself, since the pointer focus might differ from the current surface
due to other factors (e.g. grabs).
Change the code checking whether constraints should be enabled to again
check the pointer logical focus, this will be influenced by the
MetaWaylandEventInterface mechanism, and correctly reflect the logical
state accounting for those factors.
Fixes: 125ba92169 ("wayland: Port pointer constraints to using MetaWaylandEventInterface")
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3303
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3618>
Currently, we don't handle too well the removal of a MetaWaylandEventInterface
during meta_wayland_event_interface_invalidate_all_focus(), since the
MetaWaylandEventInterface may be freed at an intermediate point in that function
while handling the focus change for all input devices.
Turn this invalidate_all_focus() function into a MetaWaylandInput method, and
always ensure to use the currently effective MetaWaylandEventInterface when
resetting the focus for each device.
This fixes the situation through handling reentrancy naturally, a focus
sync (say, triggered by a grab) would reset a device focus (say, pointer), which
would remove an event interface (say, a pointer constraint), which would
invalidate_all_focus() again underneath using the new effective
MetaWaylandEventInterface. When that is done, the initial invalidate_all_focus()
call would re-apply the same focus to the same currently effective
MetaWaylandEventInterface, resulting in a no-op for the remainder of the function.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3618>
Even though the logical focus is typically business that only the
MetaWaylandEventInterface mechanism minds about, there are some pointer
subsystems that want to look this up, as opposed to the current surface.
Add a getter to make this easier, without struct peeking.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3618>
This also removes some g_return_if_fail's because the test suite expects
to be able to create arbitrary KmsUpdates even if they don't make sense
for the real state. We just get lucky that the test suite isn't
constructing updates with color space and hdr changes, yet.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3552>
