Summary:
Any call made to a virtual method in constructor/destructor of a base
class won't go to a derived class because the base class may access
uninitialized or destroyed resources.
For example, let's consider the following two classes
class Base {
public:
Base() { foo()->bar(); }
virtual ~Base() { foo()->bar(); }
virtual Foo* foo() const { return nullptr; }
};
class Derived : public Base {
public:
Derived() : mFoo(new Foo) {}
~Derived() override { delete mFoo; }
Foo* foo() const override { return mFoo; }
private:
Foo* mFoo;
};
When an instance of Derived class is created, constructors will run in
the following order:
Base()
Derived()
It's not safe to dispatch foo() method call to Derived class because
constructor of Derived hasn't initialized yet mFoo.
Same story with destructors, they'll run in the following order:
~Derived()
~Base()
It's not safe to dispatch foo() method call in the destructor of Base
class to Derived class because mFoo was deleted.
So, what does that weird C++ behavior has something to do with KWin? Well,
recently Compositor class was split into two classes - WaylandCompositor,
and X11Compositor. Some functionality from X11 doesn't make sense on
Wayland. Therefore methods that implement that stuff were "purified," i.e.
they became pure virtual methods. Unfortunately, when Compositor tears
down it may call pure virtual methods on itself. Given that those calls
cannot be dispatched to X11Compositor or WaylandCompositor, the only
choice that C++ runtime has is to throw an exception.
The fix for this very delicate problem is very simple - do not call virtual
methods from constructors and the destructor. Avoid doing that if you can!
This change moves Compositor::updateClientCompositeBlocking to X11Compositor
so it longer has to be a virtual method. Also, it kind of doesn't make sense
to keep it in base Compositor class because compositing can be blocked only
on X11.
BUG: 411049
Test Plan: KWin no longer crashes when running kwin_x11 --replace command.
Reviewers: #kwin, romangg
Reviewed By: #kwin, romangg
Subscribers: anthonyfieroni, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23098
Summary:
We can simplify the AbstractWaylandOutput code some more by getting name and
refresh rate of an output from the always available output device object.
Test Plan: Tested with DRM, Wayland nested and virtual backends.
Reviewers: #kwin
Subscribers: davidedmundson, apol, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23546
Summary:
KWin blits contents into a QImage raster buffer. Because of this we
don't need our underlying GL surface to be double buffered as it
effectively makes the whole thing tripple buffered.
Test Plan: Ran Aurorae, looked the same, should save some memory
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23331
Summary:
Aurorae has a timer between request update and updating the contents.
This makes some sense.
However on initial creation we want to snap the contents immediately as
otherwise we may be out by a frame between a new window being mapped and
the window decoration being rendered.
Test Plan:
Ran with Auroae
Windows opened immediately seemed to have a deco straight away
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: zzag, romangg, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23332
Summary:
The Wayland and XDG output interfaces are created or destroyed when the
output is en- or disabled. This follows a clear path of dependencies and
we should not check if this is indeed the case but instead assert on it.
Test Plan: Nested Wayland and Drm sessions.
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23540
Summary:
With f013a4369c, cd6b69a4d2, d960be4b3f and edb200f6bc all possible
backends of a Wayland session have been ported to using the internal functions
of AbstractWaylandOutput's for managing outputs.
This removes the alternative code path used before in these backends, what
simplifies the code and is also a prerequisite to removing the Screens global
in the future.
Reviewers: #kwin, apol, zzag
Reviewed By: #kwin, apol, zzag
Subscribers: apol, kwin
Tags: #kwin
Maniphest Tasks: T11459, T11098
Differential Revision: https://phabricator.kde.org/D23485
Summary:
Get the physical size directly from the always available output device
interface instead of saving an additional copy in the abstract wayland
output class.
There is some ambiguity with orientation and naming that needs to be
cleaned up when output orientation is reworked.
Test Plan: Nested Wayland, Drm, virtual backends tested.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: kwin
Tags: #kwin
Maniphest Tasks: T11459
Differential Revision: https://phabricator.kde.org/D23496
Summary:
Get the pixel size directly from the always available output device
interface instead of saving an additional copy in the backends.
Test Plan: Nested Wayland, Drm, virtual backends tested.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: kwin
Tags: #kwin
Maniphest Tasks: T11459
Differential Revision: https://phabricator.kde.org/D23495
Summary:
Let the data saved in OutputDeviceInterface be the single source of truth
and as low hanging fruits first do this for global position and scale.
Test Plan: Nested Wayland, Drm, virtual backends tested.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: davidedmundson, kwin
Tags: #kwin
Maniphest Tasks: T11459
Differential Revision: https://phabricator.kde.org/D23489
Summary:
For every abstract wayland output an output device interface is created
by the backend directly after instance creation. We can therefore rely on
its existence and remove superfluous checks.
Test Plan: Relevant auto tests pass. Wayland nested and DRM sessions tested.
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: kwin
Tags: #kwin
Maniphest Tasks: T11459
Differential Revision: https://phabricator.kde.org/D23486
Summary:
Create output devices in virtual backend. For that the setVirtualOutputs call
can only come after the Wayland server has been initiliazied such that the
display exists to create the output and output device interfaces. Tests have
been adjusted for that.
Test Plan:
```
98% tests passed, 3 tests failed out of 148
Total Test time (real) = 362.97 sec
The following tests FAILED:
33 - kwin-testInternalWindow (Failed)
39 - kwin-testPointerInput (Failed)
101 - kwin-testMoveResize (Failed)
```
Failing of these tests looks unrelated to the change.
Reviewers: #kwin
Subscribers: kwin
Tags: #kwin
Maniphest Tasks: T11459
Differential Revision: https://phabricator.kde.org/D23477
Summary:
Since all Wayland session backends now use the same structure of
AbstractWaylandOutput we can create output devices like in the DRM backend.
First let us do this for Wayland nested sessions.
Test Plan: Manually with output-count 1 and 2. Outputs are correctly shown in KScreen.
Reviewers: #kwin
Subscribers: zzag, kwin
Tags: #kwin
Maniphest Tasks: T11140, T11459
Differential Revision: https://phabricator.kde.org/D23473
Summary:
If the Desktop Grid effect doesn't use the Present Windows effect to
layout windows, windowAt helper can return a window that doesn't belong
to current activity because it doesn't check whether that window belongs
to current activity.
This change addresses that problem by adding corresponding check.
BUG: 301447
FIXED-IN: 5.13.4
Test Plan:
//Unchecked "Use Present Windows effect to layout the windows".//
* Switched to activity #1
* Switched to virtual desktop #1
* Launched Konsole and maximized it(to increate hit area)
* Switched to activity #2
* Switched to virtual desktop #2
* Launched Dolphin and maximized it
* Activated the Desktop Grid effect, clicked on virtual desktop #1 (activity #2 is still active)
* Switched to activity #1
* Activated the Desktop Grid effect, clicked on virtual desktop #2 (activity #1 is still active)
Reviewers: #kwin, davidedmundson
Reviewed By: #kwin, davidedmundson
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D14046
Summary:
The EGL GBM backend must be informed about mode changes so that the surface
gets reset. This fixes mode changes for me. Rotations still not work, this
needs some more structure cleanup.
Test Plan: Manually changed resolution and scale alone and together with KScreen.
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23435
Summary:
It is unnecessary to check if the change-set is null when looking at the
KWayland code. Also the check would be useless at the moment anyway because
before the check we access the change-set already in the DRM backend and at
the check we would not return if it would be null.
Test Plan: Relevant auto tests pass.
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23428
Summary:
With 1a11abc821 X11 standalone plugin uses the AbstractOutput interface for
managing its outputs. In there the refresh rate is stored in 1/μs.
Test Plan: currentRefreshRate() provides correct rate.
Reviewers: #kwin
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23265
Summary:
Qt will position the menu just fine, this is overhead.
No matter how hard I tried, I couldn't get the menu to show up in a
position where it wasn't shown entirely, removing this should be safe.
Suggested by zzag.
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: zzag, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23142
Summary:
Since C++11 alignas is a keyword, so we don't need to create our own.
All relevant compilers (gcc, clang, icc) define __SSE2__, so just use that.
There seem to be frequent crashes in the quad version of this code, but
this change should not make any difference from what I can tell.
Reviewers: #kwin, zzag
Reviewed By: #kwin, zzag
Subscribers: zzag, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23120
Summary: The popup/exec if was duplicated three times, only to calculate the y position.
Reviewers: #kwin, romangg, zzag
Reviewed By: #kwin, romangg, zzag
Subscribers: zzag, romangg, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23102
Summary:
There is no point in using quint32 and casting back and forth in numerous places.
Fix a bunch of compiler warnings that we implicitly cast between signed and unsigned.
This makes things consistent with what we get from libinput.
Reviewers: #kwin, romangg
Reviewed By: #kwin, romangg
Subscribers: zzag, romangg, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23086
Summary:
We have a smart pointer, so be consistent in checking it.
Since there is an operator->, make the code easier to read by removing countless .data()->.
!m_client.isNull() can also be written as m_client since operator bool works.
Reviewers: #kwin, romangg
Reviewed By: #kwin, romangg
Subscribers: romangg, zzag, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23067
Summary:
Having a function with a confusing comment, to iterate over the enum,
is mostly confusing. Make it boring by just moving the cast to the one
place that uses it.
Reviewers: #kwin, romangg, zzag
Reviewed By: #kwin, romangg, zzag
Subscribers: zzag, kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23085
Summary: I don't think it matters, but for completness' sake, undefine all of them.
Reviewers: #kwin, romangg
Reviewed By: #kwin, romangg
Subscribers: kwin
Tags: #kwin
Differential Revision: https://phabricator.kde.org/D23083
