See previous commit log on the effects of this.
This means the deadline evasion needs to be added in both cases in
clutter_frame_clock_notify_presented.
v2:
* Use meta_kms_update_set_sync_fd. (Jonas Ådahl)
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3958>
If the KMS thread is using the deadline timer, and a valid sync_file
descriptor is passed in:
1. The update is deferred, and the deadline timer is left armed, until
the sync_fd signals (becomes readable).
2. Implicit synchronization is disabled for the KMS update.
This means cursor updates should no longer miss a display refresh
cycle due to mutter's compositing GPU work finishing too late.
v2:
* Use g_autoptr for GSource in meta_kms_impl_device_handle_update.
(Sebastian Wick)
v3:
* Use meta_kms_update_get_sync_fd, don't track sync_fd in
CrtcFrame::submitted_update. (Jonas Ådahl)
v4:
* Clean up CrtcFrame::submitted_update members in crtc_frame_free.
v5:
* Coding style cleanup in meta_kms_impl_device_handle_update.
(Jonas Ådahl)
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3958>
The new_connectors list must be initialized empty, otherwise, connectors
are duplicated.
Fixes: 115b2c124711 ("wayland/drm-lease: Make MetaDrmLeaseManager::connectors a GList")
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3967>
We're already inhibiting real time scheduling when reading new KMS state
after hot plugs, as well as when during mode sets, due to the kernel not
being able to reliably handle these within the 250 ms limit. However, we
didn't do this during initial probing, which meant that occasionally
we'd run into these kind of issues during startup.
Handle this by always inhibiting real time scheduling up front, and
don't uninhibit until all initially discovered device have finished
processing their initial mode set.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3628
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3960>
And take it into account in meta_kms_crtc_get_deadline_evasion.
This uses the same fundamental approach as clutter frame clock scheduling:
Measure the deadline timer dispatch duration, keep track of the longest
duration, and set the timer to fire such that the longest measured
dispatch duration would result in it completing shortly before start of
vblank.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3612
v2:
* Move DEADLINE_EVASION_CONSTANT_US addition from
meta_kms_crtc_determine_deadline to meta_kms_crtc_get_deadline_evasion.
* Calculate how long before start of vblank dispatch completed for
debug output in crtc_frame_deadline_dispatch.
* Shorten over-long lines in crtc_frame_deadline_dispatch.
v3:
* Take VRR into account in crtc_frame_deadline_dispatch &
meta_kms_crtc_update_shortterm_max_dispatch_duration. (Robert Mader)
v4:
* Check if deadline has already passed in meta_kms_crtc_determine_deadline,
set the deadline for one refresh interval later if so.
* Fix indentation in crtc_frame_deadline_dispatch.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3934>
On a hybrid machine with i915 primary and nvidia-drm (470) secondary,
`meta_render_device_egl_stream_initable_init` calls
`meta_kms_inhibit_kernel_thread` to change from the default 'kernel'
thread type to 'user'. And soon after that it would
`meta_render_device_egl_stream_finalize` because I'm not actually
using that GPU, and calls `meta_kms_uninhibit_kernel_thread`.
So during startup Mutter would default to a realtime kernel thread,
switch to a user thread (which doesn't support realtime), and then
switch back to a realtime kernel thread.
In the middle of all that, while the thread type was 'user' and
realtime disabled, something was invoking `ensure_crtc_frame` which
created a `CrtcFrame` without a deadline timer. Soon after that the
thread type changed back to 'kernel' with deadline timers expected, but
our existing `CrtcFrame` has no deadline timer associated with it. And
so it would never fire, causing the cursor to freeze whenever the primary
plane isn't changing. And the problem was permanent, not just the first
frame because each `CrtcFrame` gets repeatedly reused (maybe shouldn't
be called a "Frame"?).
Now we adapt to switching between kernel and user thread types by adding
and removing the deadline timer as required.
Close: https://gitlab.gnome.org/GNOME/mutter/-/issues/3464
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3950>
If we finish compositing in time, the composited result will be
submitted prior to the deadline timer is triggered, and we'll be fine,
and if not, at least the cursor updates will be smooth, which makes it
appear smoother than not.
There is a risk that this can negatively impact composited updates when
moving the cursor, so make it possible to toggle a paint-debug flag for
now until this has been more tested.
This also mean we need to disarm the deadline timer after handling
update, as there might be a scheduled cursor update pending, but we
already handled it, so disarm the timer.
Here is an illustration of the difference.
In the following scenario, with disarming, the composited frame E, and
the cursor movement C gets presented. With this branch, only the cursor
movement C gets presented.
```
* A: beginning of composited frame
* B: begin notification reaches KMS thread
* C: cursor moved
* D: calculated deadline dispatch time (disabled with the branch)
* E: KMS update posted
* F: KMS update reaches KMS thread
* G: actual deadline (and with branch and gets committed)
Compositor thread: --------A---------------E---------
\ \
\ \
KMS thread: -----------B------C----D---F-G----
```
In the following scenario, by not disarming, the cursor update C will be
presented, and the would-be-delayed composited frame E would be delayed
anyway, i.e. fixing cursor stutter.
```
* A: beginning of composited frame
* B: begin notification reaches KMS thread
* C: cursor moved
* D: calculated deadline dispatch time (and with branch will be dispatched)
* E: KMS update posted
* F: actual deadline
* G: KMS update reaches KMS thread (and with branch gets postponed)
Compositor thread: --------A---------------E---------
\ \
\ \
KMS thread: -----------B------C----D-F-G------
```
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3184>
Although we track updates for EGL_DEVICE, they are often empty because
the primary plane has a custom page flip method. That means there's
no CRTC latched yet, but we do know exactly which CRTC is associated
with the flip. Set it so the update can still be processed.
Fixes: 27ed069766 ("kms/impl-device: Add deadline based KMS commit scheduling")
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3939>
The relationship between MetaKmsConnector and MetaDrmLease is already
stored in MetaDrmLeaseManager::leased_connectors.
Change the type of MetaDrmLeaseManager::connectors to a GList.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3922>
As in the protocol definition for wp_drm_lease_connector_v1::withdrawn:
Sent to indicate that the compositor will no longer honor requests
for DRM leases which include this connector. [...] Compositors are
encouraged to send this event when [...] the connector gets leased
to a client.
Withdrawn the leased connectors and, if they are available once the
lease finishes, advertise them again.
Related to: https://gitlab.freedesktop.org/wayland/wayland-protocols/-/merge_requests/322/
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3922>
The `lease` variable is reused in a loop and might contain values from
previous loop iterations.
Fixes: 4a5fcef38de7 ("native/kms-lease: Implement leasing out a set of connectors")
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3922>
And stop passing in the color states from the RendererNative. We also
keep the color states updated by listening for changes in the color
device.
The RendererX11Cm has a single view and no mapping to a specific color
device, so we handle the absense of a color device as well and rely on
ClutterStageView to have the default color states.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3930>
This allows us to destroy and create a new offscreen dynamically, when
the rotation or color state changes.
An idle gsource with priority higher than CLUTTER_PRIORITY_REDRAW is
used to ensure the an offscreen exists when required without having to
allocate in the redraw process.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3930>
When triple buffering, `meta_onscreen_native_prepare_frame` for the next
frame is called before `notify_view_crtc_presented` for the previous frame.
So our booleans were unfortunately still TRUE in the second prepare_frame,
resulting in two frames with the same property updates.
When double buffering, having roughly one frame interval between
`meta_onscreen_native_prepare_frame` and `notify_view_crtc_presented`
meant that property updates signalled between the swap and presentation
wouldn't get attached to a KMS update, and would be forgotten when
`notify_view_crtc_presented` resets the flags to FALSE.
To solve these we now keep a separate flag and counter per property,
tracking invalidation and pending updates respectively. The latter is a
counter rather than a boolean in support of triple buffering where two
updates may be pending concurrently (next and posted).
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3912>
Because if the current theme has exceeded the dimensions of
`DRM_CAP_CURSOR_WIDTH/HEIGHT` then the warning is just going to repeat
every time the cursor changes. We still fall back to software cursors
just fine so it's not important to repeat the warning.
In Mutter 46 the warning was "Invalid theme cursor size". Same problem.
Closes: https://gitlab.gnome.org/GNOME/mutter/-/issues/3597
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3924>
by executing `global.context.get_debug_control().exported = true`.
This makes it possible to get access to the debug service without having
to start with `--enable-debug`.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3902>
It won't be used until later when we flip, and in fact assigning
it early could have led to its own assertion failing on the next frame
in the unlikely event that we return with "Failed to ensure KMS FB ID...
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3891>
Introduced in libinput 1.26 this feature allows restricting the
tablet tool pressure range to a subset of its physical range. The
use-case is either to require some higher-than-usual minimum pressure
before the pen reacts, or lower-than-usual pressure to reach the maximum
logical pressure.
libinput takes a [0.0, 1.0] normalized range which we expose as percent
in the gsettings. The wacom driver doesn't have an exact equivalent but
it has a Threshold setting (range [1, 2048]) that defines when a button
is generated for tip down.
See gsettings-desktop-schemas!84
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3794>
Stylus actions that don't map into LMR or back/forward are now created
as a clutter button event with a button number of zero. Nothing is
actually done with those events for now, they're just discarded later.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3649>
For stylus buttons we apply a button mapping, e.g. secondary button ->
right. This mapping was previously applied to the clutter event's evdev
code only, not the actual clutter button. As a result, gnome-shell would
always treat the BTN_STYLUS as middle and BTN_STYLUS2 as right,
regardless of the mapping.
Move the mapping up so we first adjust our evcode, then proceed with
the usual mappings.
Note that this temporary breaks the stylus mapping to Back/Forward which
will be fixed in a follow-up commit.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3649>
Use the helper function for mapping a stylus tool evdev code to a
clutter button code. This fixes a (theoretical) issue - if a tool were
to send any button other than the one we handled those would likely be
BTN_SIDE and friends and we'd likely end up with negative button
numbers. The BTN_TOOL_PEN range is not predicable enough to do any sort
of calculation conversion because things like BTN_TOOL_DOUBLETAP have
specific meanings that aren't actually buttons.
Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3649>
