bb91787b5f
cursor: stop using set_image() in wlr_cursor_unset_image()
...
set_image() will go away.
2 years ago
bbd53b3b46
screencopy-v1: fix logical->physical region transform
...
This now matches the transformations used for e.g. cursor scissor
regions and damage.
2 years ago
57a0313779
wlr_scene: Don't recompute node location during rendering
...
We already compute it during the render list creation, we should cache
it and re-use the value.
2 years ago
48f2a7a3a1
wlr_scene: Simplify direct scanout handling
...
By adding a sent_feedback bool into the list entry that we can mutate
we no longer need to maintain this `sent_direct_scanout_feedback` variable.
sent_feedback will also be useful for output layers.
2 years ago
9613b3bc8a
wlr_scene: Wrap render_list nodes in struct
...
We will be able to add more state here
2 years ago
53b07b5b72
cursor: set image for new outputs
...
When an output is added to wlr_cursor, update its cursor image.
Closes: https://gitlab.freedesktop.org/wlroots/wlroots/-/issues/2119
2 years ago
d5581e42d4
cursor: add cursor_update_outputs()
...
Sets the cursor image for each output depending on the wlr_cursor
state.
2 years ago
d1d6cd9008
cursor: keep track of XCursor manager and name
...
Will be useful to apply the current cursor to new outputs.
2 years ago
69c44de22f
cursor: track surface globally, instead of per-output
...
Most of the surface-related state does not need to be per-output.
Move it to wlr_cursor_state.
2 years ago
c46fabe45c
cursor: use wlr_cursor_unset_image() in wlr_cursor_set_surface()
2 years ago
f0b8a68654
cursor: store wlr_cursor inline in wlr_cursor_state
...
Removes one allocation, makes this a bit more consistent with the
rest of wlroots.
2 years ago
8243399385
output: Set output mode during main commit
...
Removes duplication across all the backends to finally apply the mode
to the output.
2 years ago
ea3c6fde3a
wlr_scene: Add opaque region blend mode optimization
...
If we are only rendering within a region of the surface that is
advertised as opaque, we can choose no blending
2 years ago
71b57b8d27
fullscreen-shell: set surface role
2 years ago
be05097968
output: add wlr_output_state_init()
...
This changes the semantics of wlr_output_state. Instead of having
fields with uninitialized memory when missing from the committed
bitflag, all fields are always initialized (and maybe NULL/empty),
just like we do in wlr_surface_state. This reduces the chances of
footguns when reading a field, and removes the need to check for
the committed bitfield everywhere.
A new wlr_output_state_init() function takes care of initializing
the Pixman region.
2 years ago
8a5b5e6f28
compositor: listen to role_resource destroy signal
...
Call wlr_surface_destroy_role_object() when the role_resource is
destroyed.
2 years ago
89cb484220
compositor: replace role_data with role_resource
...
This increases type safety, makes it more obvious that role_data
must represent the role object, and will allow for automatic
cleanup when the resource is destroyed.
2 years ago
00f1870d35
input-method-v2: add popup_surface_from_resource()
...
This checks the resource type before casting its user data pointer.
2 years ago
fc9b58e84d
compositor: don't call role unmap hook without an object
2 years ago
0f67580aab
compositor: introduce wlr_surface_set_role_object()
2 years ago
0040c78c0b
compositor: fix wlr_surface_set_role() error messages
2 years ago
d59749aa44
cursor: add support for animated XCursor
2 years ago
6dd8b092e9
output/swapchain: fix error return value types
...
This function returns a bool, not a pointer.
2 years ago
b5eaa47049
scene: Use wlr_output_state_copy for direct scanout path
2 years ago
4b7c526578
output: Introduce wlr_output_state_copy()
2 years ago
1e68e1f661
seat: fix keyboard/pointer/touch resource destroy
...
We need to always unlink the resource, even if it's already inert.
seat_client_destroy_*() may be called multiple times on the same
resource.
Closes: https://gitlab.freedesktop.org/wlroots/wlroots/-/issues/3638
2 years ago
c4d4409deb
output: stop using wlr_output.back_buffer in output_ensure_buffer()
...
This allows us to use the new render pass API, and drops some
implicit state usage.
2 years ago
b625a13156
output: attach buffer to state in output_ensure_buffer()
2 years ago
2ac31fc630
wlr_scene: Introduce wlr_scene_buffer_set_filter_mode
2 years ago
64c0272f81
output_state: Gamma lut should be freed only if committed
...
We consider state that is not committed to be uninitialized.
2 years ago
220402b717
cursor: add wlr_cursor_unset_image()
...
It's pretty awkward to call wlr_cursor_set_image() with 6 zeroes.
Hide that awkwardness in wlroots.
2 years ago
da04b066ea
cursor: add wlr_cursor_set_xcursor()
...
This keeps track of the last set XCursor. If it hasn't changed,
skip the texture upload.
In the future, support for animated XCursors can be added.
Closes: https://gitlab.freedesktop.org/wlroots/wlroots/-/issues/3436
2 years ago
58a2f8bd5c
wlr_scene: Clear damage ring on transform change
2 years ago
61667404ce
wlr_scene: Clear damage ring on scale change
...
If we were trying to change the scale in wlr_scene_output_build_state,
we would be using incorrect damage
2 years ago
96b95bf775
wlr_scene: Reset damage ring bounds every frame
...
This fixes damage issues when a output state attempts to change
the transformed size in some way
2 years ago
4391845910
wlr_damage_ring_set_bounds: Ignore duplicate size requests
2 years ago
3896b6b107
wlr_scene: Fix transforms with non 1 scale
...
wlr_region_transform / wlr_box_transform want unscaled coordinates here
2 years ago
753f3cc4fa
compositor: add wlr_surface_role.no_object
...
This commit allows to make a role as not represented by an object,
which fixes calling role commit handlers for roles like cursor
surfaces.
Fixes: 099b9de752
2 years ago
670915eeea
output: add wlr_output_state_set_layers()
2 years ago
0925a500d1
wlr_scene: Don't use direct scanout if trying to commit a new render_format
...
Found by inspection.
2 years ago
a09bb1314d
wlr_output_state: Consider gamma_lut uninitialized if not committed
2 years ago
b80e67d6e6
wlr_output_state: Consider buffer uninitialized if not committed
2 years ago
6e8fb5509f
cursor: only reset the cursor when the surface changes
...
If the set_cursor request is used with the same surface, don't call
cursor_output_cursor_reset_image(). That function sends
wl_surface.leave and can cause an infinite feedback loop with some
clients (submitting a LoDPI cursor when the surface leaves an HiDPI
output).
Closes: https://gitlab.freedesktop.org/wlroots/wlroots/-/issues/3669
2 years ago
05dd08afe5
scene: introduce wlr_scene_output_build_state()
2 years ago
7e67b8152b
scene: Don't rely on output current state during rendering
...
In the next commit, state such as scale, transform, width/height that
we gather from the output might be outdated and incorrect.
2 years ago
ae7bad86dd
output: introduce wlr_output_state_set_damage()
2 years ago
36376e2ddf
gamma-control-v1: stop applying gamma LUTs implicitly
...
Make the compositor responsible for doing that, so that they can
properly integrate with their output commit sequence.
2 years ago
ec9b79ef19
gamma-control-v1: add wlr_gamma_control_v1_send_failed_and_destroy()
2 years ago
026fc6eda0
gamma-control-v1: add wlr_gamma_control_v1_apply()
2 years ago
5d5cf34486
output: add wlr_output_state_set_gamma_lut()
2 years ago
6e635d2fd3
gamma-control-v1: add wlr_gamma_control_manager_v1_get_control()
2 years ago
70c1a57248
gamma-control-v1: introduce set_gamma event
2 years ago
63f3802e0f
subcompositor: set role_data with wlr_surface_set_role()
2 years ago
24a479a501
drag: don't set icon's role_data
...
Currently, an icon surface's role_data is set manually to a struct
wlr_drag_icon, which is hacky, incorrect (as role_data is supposed
to be the surface's role object, and drag icons don't have them), and
will be disallowed by future changes.
2 years ago
31d958f70a
cursor: fix applying hotspot offset
...
"On surface.attach requests to the pointer surface, hotspot_x and
hotspot_y are _decremented_ by the x and y parameters passed to the
request."
2 years ago
544c0622fa
presentation-time: constify wlr_presentation_feedback_send_presented()
...
We don't need to mutate the event in this function.
2 years ago
25a62a3775
subcompositor: simplify subsurface_consider_map(), call on add
...
Fixes: 49e9be62ae
2 years ago
2acc74a3db
Clear input regions if they're ignored
2 years ago
9f28366b55
wlr_tablet_v2_tool: Tablet tool surfaces are always mapped
2 years ago
d700bd2d7c
wlr_seat_pointer: cursor surfaces are always mapped
2 years ago
9e8947e4d5
add render timer API
...
Based on five calls:
wlr_render_timer_create - creates a timer which can be reused across
frames on the same renderer
wlr_renderer_begin_buffer_pass - now takes a timer so that backends can
record when the rendering starts and finishes
wlr_render_timer_get_time - should be called as late as possible so that
queries can make their way back from the GPU
wlr_render_timer_destroy - self-explanatory
The timer is exposed as an opaque `struct wlr_render_timer` so that
backends can store whatever they want in there.
2 years ago
37b5f8a89e
compositor: unmap subsurfaces too on unmap
...
Similar to 49e9be62ae
2 years ago
15f2f6642f
gamma-control: Read ramps using pread
...
read advances the file description offset, and requires the client to
ensure that it is reset appropriately before the next gamma ramp
submission. As there is no event to indicate that wlroots finished
processing the new gamma ramp, this can result in a race between the
client seeking in the file and wlroots reading its content.
Use pread with a fixed offset instead.
Fixes: https://gitlab.freedesktop.org/wlroots/wlroots/-/issues/3655
2 years ago
b61d5922f1
compositor: notify subsurfaces about a commit in the rendering order
...
This is the default order used across wlroots unless the reversed order
specifically is required (e.g. in wlr_surface_surface_at()).
2 years ago
49e9be62ae
subcompositor: consider mapping on parent map
...
Fixes: https://gitlab.freedesktop.org/wlroots/wlroots/-/issues/3654
2 years ago
a2f94b500c
virtual-keyboard: Handle inert seats
2 years ago
f9f17b6236
virtual-pointer: Handle inert seats
...
Inert seats are handled like inert outputs, so they work as if the
wanted seat was NULL. The seat argument is only a suggestion, so this
is okay.
2 years ago
07e27ba854
xdg-popup: Handle inert seats
...
According to the protocol text, the popup will be immediately dismissed
when the grab is rejected. Because the grab can't be executed with a
destroyed seat, the popup is dismissed when one is provided.
2 years ago
0601813ab6
text-input: Handle inert seats
2 years ago
bcb69b1827
primary-selection: Handle inert seats
2 years ago
247e86dca9
keyboard-shortcuts-inhibit: Handle inert seats
2 years ago
16948c7369
input-method: Handle inert seats
2 years ago
126222884d
wlr_idle: Handle inert seats
2 years ago
f988a75a80
tablet: Handle inert seats
2 years ago
5120addfbb
data-control: Handle inert seats
2 years ago
f97f4496de
data-device: Handle inert seats
2 years ago
0e5f76186e
seat: Allow binding to inert seats
...
If a seat is destroyed while a client is trying to bind it, wlroots
needs to create an inert seat resource instead of crashing.
2 years ago
52b93f7eb4
output/cursor: fix scale and transform
...
Many issues here:
- wlr_output_cursor_set_buffer() takes a buffer already scaled for
the output, not a buffer with scale=1.
- wlr_output_cursor.{width,height,hotspot_x,hotspot_y} are in output
buffer coordinates.
- render_cursor_buffer() had hardcoded no-ops for scale and
transform, instead of using the cursor surface's.
Fixes: b64e7e88bf
2 years ago
09c87cec3f
cursor: fix wl_pointer.set_cursor hotspot updates
...
A client might update the cursor with the same wl_surface as
before, but with a different hotspot. Don't ignore such updates.
Fixes: 9c9e3f6263
2 years ago
8c0eeb6a86
session-lock: fix buffer check
...
If the first commit has no buffer, the unmap hook won't catch it. Check
in the commit hook instead.
2 years ago
e75407fd4d
session-lock: remove unused wlr_session_lock_surface_v1.events.map
2 years ago
d086ee1b9e
compositor: remove wlr_surface_role.precommit
...
It was previously used to handle unmapping properly but is obsolete now.
2 years ago
af4181f388
drag: use unified map logic
2 years ago
743da5c0ae
input-method: use unified map logic
2 years ago
229abfae0c
session-lock: use unified map logic
2 years ago
c63f3659c5
layer-shell: use unified map logic
2 years ago
b0437fc416
xdg-shell: use unified map logic
2 years ago
c590bb600f
subcompositor: use unified map logic
2 years ago
6b40e08148
compositor: introduce unified map logic
2 years ago
6668c822b3
cursor: unset wlr_output_cursor.texture on surface destroy
...
When the surface is destroyed, its current wlr_texture is about to
get destroyed as well. Reset wlr_output_cursor.texture to prevent
use-after-free.
Closes: https://gitlab.freedesktop.org/wlroots/wlroots/-/issues/3652
2 years ago
9c9e3f6263
cursor: ignore wlr_cursor_set_surface() with same surface
...
Avoids damaging the output and rendering the cursor again if it
hasn't changed.
2 years ago
5bcd537ff4
output/cursor: use new rendering API
2 years ago
93a6acae9f
output: add wlr_output_begin_render_pass()
2 years ago
95062904c7
wlr_scene: Introduce wlr_scene_buffer_set_opacity
2 years ago
2346b90a9f
wlr_scene: Make scene_rect_from_node public
...
This is in an effort to be more consistent with both already public
getters: wlr_scene_buffer_from_node and wlr_scene_tree_from_node
2 years ago
88f80c4a7c
wlr_scene: Make scene_tree_from_node public
...
This is to allow for compositors that want to be more implicit about
how their scene is organized. Such a compositor may want to walk up
at a certain scene node to find something such as a surface to focus on.
2 years ago
2c6aeed7d6
output: use pending render format if any
...
If a modeset contains a render format change, use that instead of
the current one stored in wlr_output.render_format.
This fixes render_bit_depth configuration not being applied without
a second modeset in Sway.
2 years ago
d495fb8c04
wlr_{box, fbox}_equal: Consider empty boxes NULL
2 years ago
96f3f3c92e
render/pixel-format: add support for block-based formats
...
Some formats like sub-sampled YCbCr use a block of bytes to
store the color values for more than one pixel. Update our format
table to be able to handle such formats.
2 years ago
a0ebc401d0
linux-dmabuf-v1: fix wlr_drm_format_set leak in feedback_compile()
...
Fixes: 43b25fd34e
2 years ago
Simon Ser
Manuel Stoeckl
Alexander Orzechowski
Alexander Orzechowski
Kirill Primak
Rose Hudson
Kenny Levinsen
Väinö Mäkelä