On a lot of hardware, using bigger plane sizes than necessary means wasting power.
This is specifically problematic with the cursor plane, where we so far only had a
single fixed size hint through drm caps, even though the hardware often could use
a smaller cursor size.
This adds support for the per-plane SIZE_HINTS property, which allows us to pick a
smaller cursor size when the cursor image fits into it, and should save some power
that way.
Usually, only a small subset of cursors would be used, for example the
default shape, the pointer shape, the text shape, etc.
Another reason why this change makes the KXcursorTheme load cursors on
demand is to prepare the cursor loading machinery to SVG cursors.
Both kwin and plasma have different placement strategies during interactive
move resize. That can result in the window flipping between two positions.
In order to prevent that, this change makes kwin placement take precedence
over the window position specified using the plasmashell protocol. This can
be considered as a workaround.
BUG: 481829
There's generally a lot of problems with higher than 8bpc in docking stations,
especially when multiple monitors are involved. Until these problems are hopefully
eventually fixed on the driver side, limit the bpc to 8 with docks by default
Otherwise, prePaintPass, postPaintPass and possible future overhead like overlay
plane matching aren't accounted for, which can cause frames to be dropped.
CCBUG: 488782
While they are harmless, they can result in the client repainting more
than needed and resulting in the opposite edge "bouncing" when resizing
the window.
Initialize X11Window::m_client geometry with a proper value so, later, the
X11Window::configure() function can properly determine whether the client
window is resized.
NETWinInfo queries the window state and setState() will do nothing if the
proposed new state matches the cached server side state.
On the other hand, given how the test is structured, there can be pending
fullscreen changes on the kwin side that are yet to be sent to the X server
when the NETWinInfo object is created.
This change fixes that race condition by adding an explicit Xcb::sync().
Some clients (like xdotool) can send multiple events for the same
key. In some instances the number of presses and releases can even
be mismatched leading to stuck keys. While we clean up pressed keys
when the client vanishes, this doesnt help when it doesnt immediately
disconnect (for example Xwayland keeps client connections live
longer than the actual x client to reuse them).
Also fixes a copy paste mistake where key presses where also written
to buttons.
This avoids regressing compared to PowerDevil in 6.1 which also
protected against setting internal display brightness to 0.
CCBUG: 430439
CCBUG: 483490
It's completely unused. If brightness modification is desired for night light
for example, it should be tied in with the actual brightness controls for outputs
instead
With programmable LUTs, consecutive transfer functions, inverse transfer functions and
multipliers can all be combined into one LUT. This allows offloading operations in more
situations and makes the operations more efficient too, as potentially fewer LUTs have
to be programmed
This is useful for autotests and some other special cases where we need to calculate
the result of a color pipeline on the CPU. Long term, this should replace
ColorDescription::mapTo
When the GLX or the EGL backend is destroyed, it is going to reset the
RenderLoop state, including the number of frames in flight. It does so
because of the historical reasons. At the time, there was no output frame
object to track the lifecycle of a frame.
After introducing the OutputFrame and hooking it into the RenderLoop,
the pending frame count will be reset automatically in RenderLoop when
the GLX or the EGL backend is destroyed. But we forgot to remove
the invalidate() function calls. So, when the GLX backend goes down, it
resets the pending frame count to zero and then it destroys the pending
OutputFrame object, which would result in decreasing the pending frame
count by 1 and triggering an assert in the RenderLoopPrivate::notifyFrameDropped()
function.
Since there is the OutputFrame helper now, the invalidate() function
can be dropped. Technically, the invalidate function did more than just
reset pendingFrameCount to 0, for example also stop the compositeTime.
But that should be fairly harmless new behavior.
Switching to triple buffering requires dropping a frame, so if we constantly
switch back and forth between double and triple buffering, that can cause
very visible performance issues
CCBUG: 488843
sched_setscheduler() is implemented as a stub in musl that does
nothing because Linux provides no way to set scheduling parameters
per process.
Use pthread_setschedparam() to change the scheduling parameters of
the threads instead.
BUG: 487996
Wayland specification is that the compositor chooses the actions based
on keyboard modifiers rather than the application initiating the drag
being told the modifiers.
Some distributions do not wish to build the KWin X11 backend as
they do not use it, even though they wish to maintain X11 support
for Xwayland when using KWin as a Wayland compositor.
Allow this choice by splitting the build flag and setting it up to
forcibly disable building the backend when building X11 code is
switched off.
These were off by one because the surface tab was removed.
Insetad of numbers look up the indices of the widgets so its
less prone to break in the future.
Xaver Hugl
Vlad Zahorodnii
l10n daemon script
Harald Sitter
David Redondo
Jakob Petsovits
Bruno Filipe
Nicolas Fella
Nicolas Fella
David Edmundson
Neal Gompa