Commit Graph

31480 Commits

Author SHA1 Message Date
Bilal Elmoussaoui
8008e365b3 clutter: Remove Settings::backend deprecated property
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3103>
2023-08-07 19:30:34 +00:00
Bilal Elmoussaoui
68b4cc661d clutter: Remove deprecated & unnused enums
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3103>
2023-08-07 19:30:34 +00:00
Bilal Elmoussaoui
185058456f clutter: Remove deprecated BoxLayout pack-start property
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3103>
2023-08-07 19:30:34 +00:00
Bilal Elmoussaoui
8d038789f6 clutter: Remove deprecated macros usage
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3103>
2023-08-07 19:30:34 +00:00
Carlos Garnacho
cd39ba8baf core: Get per-direction pad feature labels for rings/strips
Let the caller specify the directions, so that the pad OSD UI may
assign distinct labels to each direction, instead of showing an
unified one on both directions.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3005>
2023-08-07 18:25:01 +00:00
Carlos Garnacho
3e3660ae5f core: Drop META_PAD_DIRECTION_NONE
Avoid this strange value, and stick with the ones suitable
for rings/strips.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3005>
2023-08-07 18:25:01 +00:00
Carlos Garnacho
3e6f7a9463 core: Drop META_PAD_FEATURE_BUTTON value
This is now handled externally through separate API calls.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3005>
2023-08-07 18:25:01 +00:00
Carlos Garnacho
2da9b67673 core: Separate pad button labels in high-level MetaDisplay API
These "features" are somewhat less featured, it's becoming too ugly
to handle all of them with a single API call. The clear outlier are
buttons, so move them to a separate function.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3005>
2023-08-07 18:25:01 +00:00
Carlos Garnacho
b6a6e9f187 core: Separate MetaPadActionMapper pad button labeling
This will be handled separately in future commits.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3005>
2023-08-07 18:25:00 +00:00
Carlos Garnacho
85ac3a2d1a wayland: Separate pad button labeling from other pad features
This will be fully split in future commits.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3005>
2023-08-07 18:25:00 +00:00
Carlos Garnacho
233e612ef8 core: Rename MetaPadActionType to MetaPadFeatureType
We want to separate pad buttons from strips and rings for
purposes of labeling. Start by renaming the enum "features"
to use a less englobing word.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3005>
2023-08-07 18:25:00 +00:00
Jonas Ådahl
112cca75e2 kms/impl-device: Remove deadline timer arming from queue_update()
queue_update() in a previous iteration was called in two situations:

 * A page flip was already pending, meaning if we would commit an
   update, it'd fail with EBUSY.
 * A update was marked as "always-defer" meaning it should only be
   processed from the deadline callback (would there be one). These were
   used for cursor-only updates.

In the latter, we had to arm the deadline timer when queuing a new
update, if it wasn't armed already, while in the former, we would
currently idle, waiting for the page flip callback. At that callback
would the deadline timer be re-armed again.

Since we're only handling the former now, we'll never need to arm the
timer again, so remove code doing so. The code removed were never
actually executed anymore, after the "always-defer" flag on updates was
removed.

Fixes: 27ed069766 ("kms/impl-device: Add deadline based KMS commit scheduling")
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/2940
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3150>
2023-08-07 15:59:18 +00:00
Danial Behzadi
deaa49e16f Update Persian translation 2023-08-07 14:31:28 +00:00
Daniel van Vugt
bafea13e95 kms/impl-device: Handle empty updates properly
This fixes freezes on cursor movement when:

1. meta_onscreen_native_finish_frame would create an empty update.
2. maybe_update_cursor_plane would leave it empty (!cursor_invalidated).
3. do_process would fail to recognise a non-null empty update as an error.
4. meta_kms_impl_device_handle_update would fail to send feedback on errors
   from do_process, because do_process is meant to do it.
5. Page flip listeners would wait forever for feedback that never comes.

Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/2923,
       https://gitlab.gnome.org/GNOME/mutter/-/issues/2924,
       https://gitlab.gnome.org/GNOME/mutter/-/issues/2926,
       https://gitlab.gnome.org/GNOME/mutter/-/issues/2933

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3138>
2023-08-07 16:51:43 +08:00
Daniel van Vugt
b516b009fa kms/impl-device: Treat handling update as flushing
Primary plane updates were forgetting to do this in OnscreenNative, but
rather than do it for each post there we should simply do it for each
post.

This fixes cursor stutter in the fallback path (not using deadline timers)
where needs_flush_crtcs would remain populated but CRTC_NEEDS_FLUSH would
never be emitted, because handle_flush hadn't been called for the last
post.

This is safe as the current use of scheduled flushing is only for cursor
updates, and since cursor updates happen on the same thread as processing,
and due to the fact that we always use the most up to date cursor position
when flushing, we never risk leaving an old cursor state unflushed.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3138>
2023-08-07 16:50:36 +08:00
Daniel van Vugt
7493ed39ce kms/impl-device: Avoid retrying a failing deadline timer
So as to not fill the log with:
Failed to determine deadline: drmWaitVBlank failed: Operation not permitted

This currently happens on nvidia-drm but hopefully Nvidia will fix that
in future.

Relates to: https://gitlab.gnome.org/GNOME/mutter/-/issues/2923,
            https://gitlab.gnome.org/GNOME/mutter/-/issues/2924

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3138>
2023-08-07 16:30:32 +08:00
Daniel Rusek
79e523ccca Update Czech translation 2023-08-06 15:28:56 +00:00
Hugo Carvalho
3440d2e721 Update Portuguese translation 2023-08-04 16:12:39 +00:00
Daniel van Vugt
5a33e98431 kms/update: Remove unused is_flushing function
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3148>
2023-08-04 09:37:33 +00:00
Daniel van Vugt
6c97bd79d7 kms/cursor-manager: Fix backwards FB_UNCHANGED flag
It was causing the simple/legacy backend to never upload cursor buffers.

Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/2927,
       https://gitlab.gnome.org/GNOME/mutter/-/issues/2931

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3144>
2023-08-02 16:15:11 +08:00
Florian Müllner
9b28b7fcce ci: Include tecla
It is now used by gnome-shell to preview the keyboard layout.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3140>
2023-08-01 18:55:58 +03:00
Luming Zh
f425d42dbb Update Chinese (China) translation 2023-07-31 17:41:59 +00:00
Danial Behzadi
4231de380d Update Persian translation 2023-07-30 23:25:26 +00:00
Danial Behzadi
20328ed5b1 Update Persian translation 2023-07-30 21:54:48 +00:00
Michel Dänzer
679457373a tests/kvm: Use git fetch --depth=1
Otherwise it fetches the full history of the target tag, which is a lot
of data for the Linux kernel.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3141>
2023-07-30 16:29:33 +00:00
Asier Sarasua Garmendia
34d0e5b7d6 Update Basque translation 2023-07-30 07:34:44 +00:00
Matej Urbančič
bcd92c1e72 Update Slovenian translation 2023-07-29 14:37:59 +00:00
Artur So
09b1aff5d1 Update Russian translation 2023-07-28 18:15:15 +00:00
Yuri Chornoivan
2b427b3f83 Update Ukrainian translation 2023-07-28 17:51:39 +00:00
Yosef Or Boczko
339d586b44 Update Hebrew translation 2023-07-28 07:15:56 +00:00
Ekaterine Papava
646193142b Update Georgian translation 2023-07-28 05:25:56 +00:00
Ekaterine Papava
804defadbd Update Georgian translation 2023-07-28 05:21:35 +00:00
Balázs Úr
2eb6d05b17 Update Hungarian translation 2023-07-25 23:34:37 +00:00
Robert Mader
7f19739e6b native/cogl-utils: Add YUV formats
As well as required subformats. This makes the YUV formats available in
zwp_linux_dmabuf_v1.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2191>
2023-07-25 21:24:35 +00:00
Niels De Graef
239912cc1f wayland/dma-buf: Add support for YUV formats
Implement importing of multi-plane formats. For now, only support
importing planes individually using "sub-formats". This is the most
commonly driver-supported approach in the moment, used by other
Wayland compositors as well.

In the future we will additionally want to support importing the formats
directly and let the drivers handle conversion internally.

Co-Authored-By: Robert Mader <robert.mader@collabora.com>
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2191>
2023-07-25 21:24:35 +00:00
Robert Mader
b15e14e027 native/cogl-utils: Add MetaMultiTextureFormat section
So they can be derived from the DRM format as well.

While updating the users, ensure we don't announce support for
DRM formats in zwp_linux_dmabuf_v1 if the MetaMultiTextureFormat is
INVALID. This will be used for YUV subformats in following commits.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2191>
2023-07-25 21:24:35 +00:00
Niels De Graef
3dd9f15eba shaped-texture: Start using MetaMultiTexture
To be able to later support more complex YUV formats, we need to make
sure that MetaShapedTexture (the one who will actually render the
texture) can use the MetaMultiTexture class.

Co-Authored-By: Robert Mader <robert.mader@collabora.com>
Co-Authored-By: Daniel van Vugt <daniel.van.vugt@canonical.com>
Co-Authored-By: Sebastian Wick <sebastian.wick@redhat.com>
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2191>
2023-07-25 21:24:35 +00:00
Niels De Graef
5181a826d1 compositor: Add MetaMultiTexture class
In future commits, we want to be able to handle more complex textures,
such as video frames which are encoded in a YUV-pixel format and have
multiple planes (which each map to a separate texture).

To accomplish this, we introduce a new object `MetaMultiTexture`: this
object can deal with more complex formats by handling multiple
`CoglTexture`s.

It supports shaders for pixel format conversion from YUV to RGBA, as
well as blending. While custom bleding is currently only required for
YUV formats, we also implement it for RGB ones. This allows us to
simplify code in other places and will be needed in the future once
we want to support blending between different color spaces.

Co-Authored-By: Robert Mader <robert.mader@collabora.com>
Co-Authored-By: Sebastian Wick <sebastian.wick@redhat.com>
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2191>
2023-07-25 21:24:35 +00:00
Jonas Ådahl
358c10de14 kms/cursor-manager: Only update cursor plane relevant device on mode set
When we see a mode set, the cursor manager will update all the cursor
planes so they are set correctly as part of the mode set. KMS updates
are always per-device, and what was wrong was that it didn't filter out
CRTCs on devices that wasn't part of the mode set.

Reported-by: Michel Dänzer <mdaenzer@redhat.com>
Tested-by: Michel Dänzer <mdaenzer@redhat.com>
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3130>
2023-07-23 10:16:19 +00:00
Florian Müllner
cf3a5da1d5 ci: *Really* fix using image in toolbox
It turns out that gnome-shell's toolbox image is still broken
after commit 86b77f65e7.

Toolbox' entry point ensures that the calling user exists inside
the container, and makes their home available inside the container.

There are two ways the `useradd` command in our image may interfere
with that:

 - by default, useradd uses the smallest available UID in the
   normal user range (usually 1000); this is highly likely to
   clash with the host user UID

 - if the host's /home is a symlink (for instance to /var/home
   on Silverblue), then toolbox recreates that layout inside the
   container; it cannot do that if /home is already a non-empty
   directory

Luckily we can address both issues without affecting the ability
to build and run tests as user: We can simply create the `meta-user`
with a UID and home directory that are unlikely to clash.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3134>
2023-07-23 09:39:18 +00:00
Robert Mader
c45d030132 ci: Temporarily disable aarch64 test
The runners constantly time out for the last two days.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3135>
2023-07-23 01:31:09 +02:00
Daniel van Vugt
7c0caf5727 clutter/frame-clock: Limit jitter measurements to one frame interval (100%)
Measurements above 100% were originally allowed to show when frame skipping
was occurring so you didn't have to also check the frame rate. But that
also resulted in arbitrarily high jitter values being reported when
returning from idle. And those are frequent enough to look like a bug or
untrustworthy so let's not do that anymore.

Taking the remainder of a high jitter value is still a meaningful jitter
value.

Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/2906
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3123>
2023-07-21 12:32:37 +00:00
Sebastian Keller
7faf4a308e constraints: Delay initial maximization until after reparenting is done
For SSD windows the decoration window from the frames client might have
a different maximization state than the client window and would end up
restoring it once shown. Avoid this issue by waiting until all
reparenting is done before applying the initial maximization (and
minimization).

Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/2579
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3035>
2023-07-21 11:38:17 +00:00
Daniel van Vugt
626498348b stage-impl: Blit damage regions of all ages when using a shadow FB
Original idea by Gert van de Kraats, modified to avoid having an
unused intermediate swap_region.

Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/2602
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3117>
2023-07-21 10:43:18 +00:00
Jonas Ådahl
828bc023d0 cogl/build: Remove unused built header list
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3131>
2023-07-21 07:59:54 +00:00
Jonas Ådahl
e96475bb8d build: Use correct variable for gathering built headers
We added things unused variable, which is a bit useless.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3131>
2023-07-21 07:59:54 +00:00
Daniel van Vugt
f5ff0f732a onscreen/native: Forget the view pointer when it is detached
Otherwise it would linger after a hotplug event, pointing to the old
(destroyed) view. And it's much easier to debug a NULL pointer than
a pointer to freed memory.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3118>
2023-07-20 23:28:49 +00:00
Michel Dänzer
9c021b0bc2 clutter/frame-clock: Clamp recorded maxima to refresh interval
clutter_frame_clock_compute_max_render_time_us clamps to the refresh
interval anyway, so the only effect a higher recorded maximum can have
is to delay it falling below the refresh interval again.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3090>
2023-07-20 22:34:58 +00:00
Michel Dänzer
69afa7a142 clutter/frame-clock: Use single pair of maxima for render time estimate
Instead of separate pairs of short- and long-term maxima for each
measured step of the frame update process.

This should result in the render time estimate rising less often:
Previously it did whenever the measurement of any of at least 3 out of
4 steps reached a new maximum, even if that didn't result in a new
maximum for the whole update duration. Now it's only in the latter
case.

This should also result in a lower render time estimate (and thus
input→output latency) in general, since the variability of
measurements for each 3/4 steps doesn't always add up anymore. The flip
side of this is that it might result in missing a display refresh cycle
more often.

v2:
* Fix coding style in maybe_update_longterm_max_duration_us.
  (Robert Mader)

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3090>
2023-07-20 22:34:58 +00:00
Fabio Tomat
c4c6d17649 Update Friulian translation 2023-07-20 21:49:29 +00:00