Summary:
Unfortunately Aurorae decoration engine creates several internal clients
per each decoration. One of those clients represents QOffscreenSurface,
which is not a toplevel. Given that no QWindow object will be found for
such clients, m_internalWindowFlags contains undefined value.
Luckily, QOffscreenSurface sets FramelessWindowHint flag, but because
InternalClient is not able to find matching QWindow object, cached
QWindow flags won't have that hint set.
Thus InternalClient will attempt to decorate QOffscreenSurface. A new
Aurorae decoration will be created, which means a new QOffscreenSurface
will be created, which means a new Aurorae decoration will be created,
and so on.
This change restricts subset of internal clients that can be decorated.
Only clients with valid m_internalWindow can be decorated. If m_internalWindow
isn't null, then m_internalWindowFlags is guaranteed to be valid as well.
BUG: 407612
FIXED-IN: 5.16.3
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D22136
Summary:
GTK clients check whether _GTK_FRAME_EXTENTS is in _NET_SUPPORTED, so we
don't need this script anymore.
Also, because GTK clients don't set shadow extents, this script
is doing nothing.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: ngraham, jackg, broulik, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D21371
Summary:
Use QTRY_COMPARE macro to not fail if signal already sent. Also shorten
timeout on expect to fail test for faster test runs.
Test Plan: Debug console test passes again.
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D22124
Summary:
The internal window test is failing on CI because the client add spy waits
are not triggered. The signal has been emitted already at this point.
Removing them fixes this (the condition is still checked by subsequent
count verify on the spy) in all but one instance. In this case the wait
needs to be guarded.
Is there a more general approach to it? Always guarding is ugly. Also when
was this test regression introduced? In the past we must have had some
slack until the signal was fired to start the wait call.
Test Plan: Internal window test passes with this patch again.
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: davidedmundson, zzag, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D22119
Summary:
Declare it in the source file for internal usage of ApplicationX11 and
Compositor classes. Additionally use modern style connect and do other
code style updates.
Test Plan: Manually in X session and Wayland windowed.
Reviewers: #kwin, zzag
Subscribers: davidedmundson, sbergeron, zzag, kwin
Tags: #kwin
Maniphest Tasks: T11071
Differential Revision: https://phabricator.kde.org/D21655
Summary:
Some of these variables have been declared protected but can apparently
also be private.
Other minor code style improvements.
Test Plan: Compiles, Wayland nested session runs.
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: zzag, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D22109
Summary:
As in 2c088894b fix remaining autotests failing since no borders has become
the new default behavior of Breeze.
Autotests should not rely on this default behavior of Breeze though, but be
run with a faked deco such that changes on Breeze does not directly change
the autotests behavior. That's a goal for the future.
Test Plan: 100% tests passed, 0 tests failed out of 147
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: zzag, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D21746
Summary:
To do so can in some situations lock up as the loop goes through
different positions incrementing by client->width/height.
If this is zero we can get into a stuck state.
This became a more common issue due to my earlier patch that places windows
in ShellClient::finishInit to allow the maximize placement strategy to
set the first configure size.
BUG: 408754
Reviewers: #kwin, broulik
Reviewed By: broulik
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D21997
Summary:
On Placement=Maximized it becomes a problem because we end up sending an invalid size and when we try to recover from it, we recover from the wrong size.
This fixes setting the right size to Plasma Mobile applications.
Reviewers: #kwin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D22054
Summary:
The screenlock fails on X11 if it can't grab the keyboard.
We can't nicely solve the generic case. We can solve the common case of
a kwin effect being active.
It's not critical, arguably not even desirable to have these effects
persist after the screen is locked through an external trigger. We can
just close the effect early.
Key grabs have to be relased early before the close animation completes
so that the locker doesn't have a race based on animation times.
It's not ideal, but no worse than the current state for not much work.
BUG: 234153
Test Plan:
locked screen on a timer
opened various effects
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: ngraham, zzag, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D20890
Summary: readX11ShadowProperty fails without an X connection
Test Plan: Tested by Kai who didn't have xwayland
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: zzag, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D21965
Summary:
Compositing today is ubiquitous. There is no reason to keep compositing
specific functions of Toplevel, Client and Workspace classes in the
composite.cpp source file. Instead let these definitions be separated.
Test Plan: Compiles
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: zzag, kwin
Tags: #kwin
Maniphest Tasks: T11071
Differential Revision: https://phabricator.kde.org/D21654
Summary:
Plasma's Task manager exposes an optional feature whereby the user
can middle-click on a window to close it, but the Present Windows effect
does not do the same.
The presence of a close button you can left-click does not replace the desirable
feature to be able to middle-click on a window to close it, because then the
whole window becomes a click target, so it can be much much faster than
having to aim for the little close button. Also it's off by default, so a user
who goes out of their way to turn it on is signaling that they want to accept the
risk of accidentally closing a window by accident.
Finally, the feature is not allowed for left-click, so people can never accidentally
wreck Present Windows for themselves by assigning it to left-click by accident
and then mistakenly closing their windows.
This reverts commit 55585514f9.
FEATURE: 321190
FIXED-IN: 5.17.0
Test Plan:
Set "Close window" in the Present windows effect, trigger effect, and middle-click on window
{F6815303}
Reviewers: #kwin, davidedmundson, broulik, zzag, #plasma, hein, mart
Reviewed By: #kwin, #plasma, mart
Subscribers: mart, abetts, apol, zzag, luebking, kossebau, graesslin, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D21083
On modern machines session startup takes a few seconds where the 2s might artificially slow down the process by like 50%.
This makes the animation quicker to make the system feel ready quicker.
Differential Revision: https://phabricator.kde.org/D21904
Summary:
QtQuick windows created by KWin currently use the default swap interval
of 1, meaning buffer swaps will block until vblank. However, this
results in a problematic interaction on hybrid graphics systems running
the proprietary NVIDIA driver. VSync on such setups relies on a system
called "PRIME synchronization", where the xf86-video-modesetting driver
controlling the display will signal the NVIDIA driver when vblank has
occurred. The issue is that it will only do so if there has been damage
to the screen.
So, when KWin creates a QtQuick window, compositing will stop waiting on
the window to render, therefore no damage to the screen will occur. But
this means that no vblank notifications will be delivered to the NVIDIA
driver, so the glXSwapBuffers call by the QtQuick window will block
perpetually. The end result is a freeze of the desktop.
To get around this, we can simply disable vsync for QtQuick windows by
setting the swap interval for the default QSurfaceFormat to 0. Since
they are redirected, this shouldn't cause any tearing.
BUG: 406180
FIXED-IN: 5.16.2
Test Plan:
Using the proprietary NVIDIA driver on a hybrid graphics system, with
PRIME synchronization enabled (see
https://devtalk.nvidia.com/default/topic/957814/linux/prime-and-prime-synchronization/),
perform any action causing a QtQuick window to be created by KWin, for
example, triggering the application switcher dialogue with alt + tab.
Ensure the desktop does not temporarily freeze.
Note, this required a Qt build that includes commit
https://code.qt.io/cgit/qt/qtbase.git/commit/?id=0c1831178540462da31fd7a4b6d2e446bc84498b
resolving a bug that prevented the changing the swap interval.
Reviewers: #kwin, davidedmundson, zzag
Reviewed By: #kwin, davidedmundson, zzag
Subscribers: zzag, romangg, alexeymin, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D21808
Summary:
QtQuick windows created by KWin currently use the default swap interval
of 1, meaning buffer swaps will block until vblank. However, this
results in a problematic interaction on hybrid graphics systems running
the proprietary NVIDIA driver. VSync on such setups relies on a system
called "PRIME synchronization", where the xf86-video-modesetting driver
controlling the display will signal the NVIDIA driver when vblank has
occurred. The issue is that it will only do so if there has been damage
to the screen.
So, when KWin creates a QtQuick window, compositing will stop waiting on
the window to render, therefore no damage to the screen will occur. But
this means that no vblank notifications will be delivered to the NVIDIA
driver, so the glXSwapBuffers call by the QtQuick window will block
perpetually. The end result is a freeze of the desktop.
To get around this, we can simply disable vsync for QtQuick windows by
setting the swap interval for the default QSurfaceFormat to 0. Since
they are redirected, this shouldn't cause any tearing.
BUG: 406180
FIXED-IN: 5.16.2
Test Plan:
Using the proprietary NVIDIA driver on a hybrid graphics system, with
PRIME synchronization enabled (see
https://devtalk.nvidia.com/default/topic/957814/linux/prime-and-prime-synchronization/),
perform any action causing a QtQuick window to be created by KWin, for
example, triggering the application switcher dialogue with alt + tab.
Ensure the desktop does not temporarily freeze.
Note, this required a Qt build that includes commit
https://code.qt.io/cgit/qt/qtbase.git/commit/?id=0c1831178540462da31fd7a4b6d2e446bc84498b
resolving a bug that prevented the changing the swap interval.
Reviewers: #kwin, davidedmundson, zzag
Reviewed By: #kwin, davidedmundson, zzag
Subscribers: zzag, romangg, alexeymin, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D21808
Summary:
There are several issues with code of AbstractOutput class:
(a) Some methods are documented, and some are not. In general, we tend
to document all public methods in KWin core. It looks like a very
minor issue, but there are methods that have very ambiguous return
value. One such method is geometry(). It's not obvious whether the
returned geometry is in device independent pixels or not;
(b) There's a mix of methods defined in the cpp file and in the header.
This is not very good because reading such code becomes a bit harder
if you don't use any fancy IDE;
(c) Missing Q_DISABLE_COPY, etc.
This change addresses these issues, so the code is a bit more readable
and easier to work with.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: broulik, cfeck, davidedmundson, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D21874
Summary:
The color correction manager doesn't make any specific assumptions about
underlying platform, e.g. whether it's x11, etc. The platform just
has to be capable of setting gamma ramps. Given that, there are no any
significant technical blockers for making this feature work on x.
Reviewers: #kwin, davidedmundson, romangg
Reviewed By: #kwin, davidedmundson, romangg
Subscribers: romangg, neobrain, GB_2, filipf, davidedmundson, ngraham, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D21345
Summary:
After porting all platforms to using AbstractOutput the BasicScreens
class is not needed anymore.
Test Plan: Compiles
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: kwin
Tags: #kwin
Maniphest Tasks: T11098
Differential Revision: https://phabricator.kde.org/D21800
In 1a11abc821 some function calls in the createXdgOutput function were
accidentally removed resulting in possibility of broken setup on startup.
Add these calls back again.
Summary:
Represent outputs in the X11 session via AbstractOutput. For that we
move all Wayland specific parts of AbstractOutput into a new subclass
AbstractWaylandOutput and let the outputs of our Wayland backends inherit
from there.
This should allow us to get rid of the Screens class later on.
Test Plan: Manually in X session.
Reviewers: #kwin, zzag, davidedmundson
Reviewed By: #kwin, zzag, davidedmundson
Subscribers: ngraham, zzag, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D19208
Summary:
To homogenize our backends and as another step to remove the Screens class
use the AbstractOutput class in the windowed X11 backend.
Test Plan: Manually in X session.
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: davidedmundson, zzag, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D19207
Summary:
Since KSnapshot is no longer used, I modified the text in `effect_builtins.cpp` to reflect that, changing it to a generic "screenshot tools".
BUG: 408407
FIXED-IN: 5.17.0
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: filipf, davidedmundson, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D21700
Since d51b8dc093 the test fails on CI, apparently because we test with
Breeze default settings, which is no borders now and we can move a bit into
the window geometry and still be on the border if borders exist, otherwise
not.
For now fix it by checking if there are borders or not and then test
accordingly. But long-term we should test both cases and besides not rely on
external decorations for our test, instead use a fake specific for our
integration testing.
Summary:
XWayland clients have a surface associated with them asynchronously. In this
case we don't directly want to set focus on this Toplevel, but wait until the
surface is set.
This patch aims in conjunction with an unrelated fix to SDL at improving Steam
Big Picture Mode in our Wayland session.
Test Plan: Steam BPM regains focus on game close.
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: anthonyfieroni, davidedmundson, zzag, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D19262
Summary:
To homogenize our backends and as another step to remove the Screens class
use the AbstractOutput class in the framebuffer backend.
Test Plan: Manually on VT enforcing the framebuffer backend.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: davidedmundson, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D19199
Summary:
If a user has set his border size explicitly in the past also respect this
selection with the new auto border size mode by setting auto borders to
false on update.
For this a kconf_update file and a sed based shell script is added.
Test Plan: Manually.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: zzag, davidedmundson, kwin
Tags: #kwin
Maniphest Tasks: T8707
Differential Revision: https://phabricator.kde.org/D13480
Summary:
This is an alternative solution to T8707 and in comparision to D13276 a less
drastic change to KWin's default behavior.
Instead of changing the border size default for all KDecoration plugins by
switching the default from border size Normal to None introduce new
functionality, which allows a KDecoration plugin to recommend a border size in
its metadata. By default KWin listens for these recommendations and sets the
border size accordingly.
If there is no metadata recommending a border size, KWin falls back to the
current setting of Normal sized borders.
A user is able to override the recommendations from the KCM, which has been
extended accordingly.
Test Plan: Manually with adjusted metadata of Breeze.
Reviewers: #kwin, #plasma, #vdg, ngraham
Reviewed By: #vdg, ngraham
Subscribers: hpereiradacosta, filipf, anemeth, davidedmundson, abetts, graesslin, ngraham, zzag, kwin
Tags: #kwin
Maniphest Tasks: T8707
Differential Revision: https://phabricator.kde.org/D13284
