Commit Graph

31480 Commits

Author SHA1 Message Date
Michel Dänzer
e95499038d frames: Fix XGetWMNormalHints return value check
It returns non-0 if there are any hints in the WM_NORMAL_HINTS
property, 0 if there are none.

Fixes the mouse cursor changing to the resize shape over the decorations
of non-resizable windows.

Fixes: c7b3d8c607 ("frames: Push error traps around various X11 calls")
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3096>
2023-07-20 10:19:22 +00:00
Daniel van Vugt
db0870b114 cursor-renderer/native: Removed unused HW_CURSOR_BUFFER_COUNT
It stopped being used in e52641c4b6.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3129>
2023-07-20 15:08:49 +08:00
Gergo Koteles
7b04e8be15 wayland: Find touch grab sequence in subsurfaces also
With libdecor, window moving/resizing only works with
the pointer, not with touch.
The meta_wayland_pointer_can_grab_surface checks for subsurfaces,
but the meta_wayland_touch_find_grab_sequence does not.

Add a similar subsurface check to
meta_wayland_touch_find_grab_sequence.

Closes: GNOME/mutter#2872
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3125>
2023-07-19 13:07:51 +00:00
Bilal Elmoussaoui
ffd3aedbf5 i18n: Remove files with no translatable strings
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3101>
2023-07-19 11:33:59 +00:00
Bilal Elmoussaoui
b852bbba47 cleanup: Stop translating nick/blurb for pspecs
As those strings are intended to be used by some UI but nothing uses
that in reality except GStreamer.
So drop them similar to what GTK did at
https://gitlab.gnome.org/GNOME/gtk/-/merge_requests/4717

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3101>
2023-07-19 11:33:59 +00:00
Robert Mader
efe884a97f wayland: Remove DRM_FORMAT_MOD_INVALID defines
In both cases we already unconditionally import drm_fourcc.h and all
distros shipping recent Mutter should by now have versions of that file
containing MOD_INVALID.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3065>
2023-07-19 10:39:55 +00:00
Robert Mader
96778aa8b1 wayland/dma-buf: Make use of meta-cogl-drm-formats
Removing duplication, making it easier to add new formats and ensuring
that the native backend and Wayland clients can use the same formats.

Also improve related build files so the Wayland backend can be build
without the native backend.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3065>
2023-07-19 10:39:55 +00:00
Robert Mader
192d6686cf backend/cogl-utils: Move, rename to cogl-drm-formats and clean up
1. Move into the new 'common' folder and build for Wayland as well
   so we will be able to share the code in follow-up commits.

2. Rename to cogl-drm-formats to make it more obvious that the format
   map is more than an utility these days.

3. Drop the unused CoglTextureComponents part (see also previous
   commit).

4. Move the map to the header, simplifying some future use-cases.

5. Sync formats with MetaWaylandBuffer and MetaWaylandDmaBufBuffer and
   also use newly introduced opaque formats where appropriate.
   This avoids duplicated code, ensures that new drm-formats added to
   the dmabuf protocol have an adequate representation in Cogl from which
   information like alpha support can be easily derived and finally
   ensures we don't crash if the mappings got out of sync.

6. Remove some likely untested formats. In case some of these are
   actually needed on certain hardware, we can test whether we got
   the correct mapping by also adding support for the corresponding
   wl_shm_format in MetaWaylandBuffer by extending the gradient test in
   https://gitlab.freedesktop.org/jadahl/wayland-test-clients

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3065>
2023-07-19 10:39:55 +00:00
Robert Mader
ae88e13c5a wayland/buffer: Clean up format map
Use the newly introduced opaque formats where appropriate, allowing us
to drop the custom components handling.

Also use pre-mult alpha. This may slightly improve upload times
and should generally be more correct.

Tested with `weston-simple-shm`, ensuring the alpha channel still gets
ignored for opaque formats, and `gradient-test` from
https://gitlab.freedesktop.org/jadahl/wayland-test-clients

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3065>
2023-07-19 10:39:55 +00:00
Robert Mader
eb19ac86ba cogl: Add missing RGBfp16 opaque formats
This adds the last missing opaque format variants we support on Wayland.

Do not add testing for these as we are missing helper function to pack
half float values.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3065>
2023-07-19 10:39:55 +00:00
Sebastian Wick
ae5512bc05 cogl/tests: Do not blend for texture format paint tests
The default cogl blend string is
`RGBA = ADD (SRC_COLOR, DST_COLOR*(1-SRC_COLOR[A]))` which is alpha
blending with premult fragment results. We do not clear the src
framebuffer and even if we did set alpha to 1 in the src fb, the
resulting alpha would be 1 and we want to check the alpha of the
fragment color.

Just turn off any kind of blending instead and write out the fragment
color to the fb.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3065>
2023-07-19 10:39:55 +00:00
Sebastian Wick
d79e4c182e cogl/tests: Fix float->int, int->float conversions
To obtain a float between 0 and 1 we have to devide the integer by the
highest possible value instead of the number of values.

Fixes off by one errors in the tests on some hardware/driver
combinations.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3065>
2023-07-19 10:39:55 +00:00
Robert Mader
4097cbbb53 cogl: Extend tests and fix RGB2101010 opaque formats
Notably fix several cases where bitmap packing was broken, resulting in
visual corruption with the GLES2 backend. These were reproducible with
the gradient-test from
https://gitlab.freedesktop.org/jadahl/wayland-test-clients

Also extend testing similar to the RGB8888 formats.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3065>
2023-07-19 10:39:55 +00:00
Robert Mader
f04209dcbf cogl: Add RGB8888 opaque format variants
So we can properly handle matching DRM and WL_SHM formats in a unified
manner.

Add extensive testing between these and existing pre-multiplied alpha
formats, i.e. all formats we support on Wayland.

Note that unfortunately for some format combinations  the value in the
alpha channel is not cleared as expected, likely because of fast-paths
in Cogl. If both source and destination format is opaque, it always
works, however. This thereby includes all cases where they are the same.

Co-Authored-By: Jonas Ådahl <jadahl@gmail.com>
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3065>
2023-07-19 10:39:55 +00:00
Sebastian Wick
5696d61a67 cogl: Fix opaque formats on GLES2 getting premultiplied
On GLES2 reading and writing some Cogl formats is not supported
natively. In those cases we use another format to do the reading and
writing. When the internal format and the temporary format differ in
premultiplication, Cogl tries to adjust for it.

Opaque Cogl formats don't have the premult bit set but our internal
format is a premult format. Cogl tries to adjust for it but completely
misses that the opaque format doesn't have an alpha channel and it
should not do so at all.

So skip the premult adjusting when the Cogl format has no alpha channel.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3065>
2023-07-19 10:39:54 +00:00
Robert Mader
42336f69cc clutter/backend: Fix core profile description
Cogl actually requests 3.1.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3065>
2023-07-19 10:39:54 +00:00
Jonas Ådahl
6841153ad7 tests/kms-device: Add test for scheduled no-ops
We can schedule an update from the cursor manager, but that doesn't mean
there will be an actual plane assignment changed at the time of the
update processing, since for example we might have "touched" a CRTC, but
already left it before the processing started, meaning we have nothing
to change after all.

Add a test case that checks that this works properly.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-18 23:51:34 +02:00
Jonas Ådahl
c96e835d56 thread/impl: Set the task source priority to rather high
Tasks should have high, but not highest priority. Roughly the aim is

  page flip feedback > composited updates > cursor updates

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-18 23:51:34 +02:00
Jonas Ådahl
e52641c4b6 cursor-renderer/native: Replace HW cursor with KMS cursor manager
This removes the old hardware cursor management code and outsources it
to MetaKmsCursorManager. What the native cursor renderer still does,
however, is the preprocessing i.e. rotating/scaling cursor that wouldn't
otherwise be fit for a cursor plane.

The cursor DRM buffers are instead of being per cursor sprite now per
CRTC, meaning we don't need to stop doing hardware cursors if part of
the cursor is on an output that doesn't support it. This is why the
whole scale/transform code changed from being per GPU to per CRTC.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-18 23:51:34 +02:00
Jonas Ådahl
ade2aa71ed tests/kvm: Bump timeout to 60 s
It can be quite slow to set up the test environment inside the VM, as
well as outside, leaving very little time for the test itself. While
it'd be nice to not run the mock env etc outside the VM, let's just bump
the timeout for now, to avoid unnecessary timeout failures.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-18 23:51:34 +02:00
Jonas Ådahl
7a821345db tests/kms/device: Test cursor updates just before turning off CRTC
If we turn of a CRTC, we might have invalidated the cursor manager for
the same CRTC, but that should not mean a cursor plane is assigned when
turning off the CRTC.

Add a test case for this.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-18 23:51:34 +02:00
Jonas Ådahl
6d873036e0 Add KMS cursor manager
This new manager object intends to take over management of the cursor
plane from the native cursor renderer. It's API is intended to be used
from the main thread, except for the _in_input() function, but mainly
operates in the KMS context, i.e. the KMS thread.

It makes use of an "update filter" that is called before each
MetaKmsUpdate is turned into a atomic KMS commit or a set of legacy
drmMode*() API calls. When the cursor position has been invalidated,
it'll assign the cursor plane in the filter callback, using an as up to
date as possible pointer position as the source for the cursor plane
position.

Cursor updates from the input thread schedules updates for the affected
CRTCs which will cause the filter to be run, potentially for cursor-only
commits.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-18 23:51:34 +02:00
Jonas Ådahl
5a798989c4 seat/native: Add run impl task helper
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-18 23:51:34 +02:00
Jonas Ådahl
ef64279035 seat/impl: Add signal for pointer motions
This signal is emitted directly in the input thread.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 23:42:38 +02:00
Jonas Ådahl
f4b76727db stage-view: Add API to inhibit cursor overlay painting
This adds some plumbing to get the "default" paint flags for regular
stage painting, where one either wants to paint the overlay, or not.

If inhibited, the 'no-cursors' paint flag is used, otherwise the 'none'
flag. This will be used to allow having a per stage view hw cursor
state.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 23:42:38 +02:00
Jonas Ådahl
5d35138df0 cursor-renderer: Make 'handled_by_backend' state 'needs_overlay'
This will, conceptually, and eventually, allow a backend to both want an
overlay, while still handling it by default.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 23:42:38 +02:00
Jonas Ådahl
3b2f95cb1f kms/update: Add destroy-notify to result listener user data
No thread guarantees, users need to handle it themselves.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 23:42:38 +02:00
Jonas Ådahl
27ed069766 kms/impl-device: Add deadline based KMS commit scheduling
This makes it possible to post KMS updates that will always defer until
just before the scanout deadline. This is useful to allow queuing cursor
updates where we don't want to post them to KMS immediately, but rather
wait until as late as possible to get lower latency.

We cannot delay primary plane compositions however, and this is due to
how the kernel may prioritize GPU work - not until a pipeline gets
attached to a atomic commit will it in some drivers get bumped to high
priority. This means we still need to post any update that depends on
OpenGL pipelines as soon as possible.

To avoid working on compositing, then getting stomped on the feet by the
deadline scheduler, the deadline timer is disarmed whenever there is a
frame currently being painted. This will still allow new cursor updates
to arrive during composition, but will delay the actual KMS commit until
the primary plane update has been posted.

Still, even for cursor-only we still need higher than default timing
capabilities, thus the deadline scheduler depends on the KMS thread
getting real-time scheduling priority. When the thread isn't realtime
scheduled, the KMS thread instead asks the main thread to "flush" the
commit as part of the regular frame update. A flushing update means one
that isn't set to always defer and has a latching CRTC.

The verbose KMS debug logging makes the processing take too long, making
us more likely to miss the deadline. Avoid this by increasing the
evasion length when debug logging is enabled. Not the best, but better
than changing the behavior completely.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 23:42:38 +02:00
Jonas Ådahl
48501236ec kms: Move error codes to main header
Some will be used outside of meta-kms*.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
ee87654224 kms/update: Add helper to checking if an update passed
Instead of comparing values, just add a `did_pass()` helper.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
5bed90114d kms/update: Treat NULL main context as default for page flips too
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
40caa6d87c kms/update: Use callback helper to free page flip listener data
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
6059d8f42d kms/page-flip: Make destroy notify function optional
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
587da92b0a kms: Mark mode set updates as such
This is currently unused, but will be later used to distinguish betwen
per frame/CRTC updates and device wide ones.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
a7954f4a75 onscreen/native: Move out timeval to timestamp helper
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
17a78303d0 kms: Make result listener take a vtable too
As with the page flip listener, prepare to make result listeners more
capable by allowing more vfuncs for the same listener instance in the
future.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
f97f0f2781 thread: Document the behavior of meta_thread_post_impl_task()
Specifically in what contexts things are invoked.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
6ed75c6253 thread: Allow passing a NULL callback vfunc
This is useful when one just wants to free something, which can be done
using the passed destroy notify function.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
0d5d1745ce thread: Push the thread-default main context
Makes g_main_context_get_thread_default() do the right thing.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
59ba952661 thread: Always add impl thread context as callback source
This is helpful when we add callbacks that should be dispatched in the
KMS impl thread.

This invalidates an assumption about callbacks not being in the impl
context, so some asserts for that are also removed.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
84ff167328 thread: Add private API to flush callbacks for main context
This will be used to flush out certain callbacks when switching thread
type.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
2617bd8e72 thread/impl: Add 'reset' signal
This signal is emitted before terminating the thread, but also when
resetting the thread type. This is to allow thread implementations to
make sure they have no stale pending callbacks to any old main contexts.

This commit "terminates" the impl thread even if there is no actual
thread; this is to trigger the "reset" signal, also when switching from
a user thread to a kernel thread.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
bedec579b8 tests/kms/updates: Use async update API
We ignored the feedback and entered a temporary main loop, so no need to
do things sync.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
15a91e9fd4 kms/update: Stop sealing updates
This isn't very useful anymore, since posting an update takes ownership
of it. Not sealing it also means e.g. more listeners can be added after
posting.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
54d9fab161 kms: Ask to be real time scheduled
This will allow more time sensitive scheduling strategies.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
6da3e9eb69 thread/impl: Add API to tell whether its realtime scheduled
Will be used by thread impl's to decide whether they can rely on real
time scheduling or not.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
e92e4fe64d thread: Register with profiler
This means we can add COGL_TRACE*() instrumentation that is grouped
correctly in sysprof. If kernel threading is enabled, they will end up
in a "Compositor (KMS thread)" group (ignoring translations).

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
bd2fa92c29 thread: Support making threads real time scheduled
Real time scheduling is needed for better control of when we commit
updates to the kernel, so add a property to MetaThread that, if the
thread implementation uses a kernel thread and not a user thread, RTKit
is asked to make the thread real time scheduled using the maximum
priority allowed.

Currently RTKit doesn't support the GetAll() D-Bus properties method, so
some fall back code is added, as GDBusProxy depends on GetAll() working
to make the cached properties up to date. Once
https://github.com/heftig/rtkit/pull/30 lands and becomes widely
available in distributions, the work around can be dropped.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
276ebbf5ee kms: Use a kernel thread by default
Also add an API to inhibit the kernel thread from being used, and make
MetaRenderDeviceEglStream inhibit the kernel thread from being used if
it's active.

The reason for this is that the MetaRenderDeviceEGlStream is used when
using EGLStreams instead of KMS for page flipping. This means the actual
page flipping happens as a side effect of using EGL/OpenGL, which can't
easily be done off thread.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00
Jonas Ådahl
788ad43e17 thread: Allow switching thread type
This will be necessary in order to default to 'kernel' and then switch
to 'user' if the thread instance can no longer be properly multi
threaded.

To avoid having the same thread impl creating and destroying
GMainContext's, this also means always creating a GMainContext for the
thread-impl. When running in user-thread mode, the GMainContext is
wrapped in a wrapper source and dispatched as part of the real main
thread GMainContext, and when in kernel-thread mode, it runs
independently in the dedicated thread.

This has the consequence that the wrapper source will always have the
priority of the highest impl context GSource, but only after it has
dispatched once. Would we need it earlier than that, we either need a
way to introspect existing sources in a GMainContext and their
priorities, or manually track known sources in MetaThreadImpl.

The wrapper source will never be below 0, as that'd mean it could reach
INT_MAX priority if it had no more sources attached to it, meaning it'd
never be dispatched again.

Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2777>
2023-07-17 21:19:34 +02:00