Summary:
The idea behind this autotest is inspired by bug 356328 which produced
incorrect rendering results. Also it's inspired by openQA which performs
image reference comparisons.
This test case tries to go further. It creates reference images which
must match the rendering result exactly. So far the test case verifies
the start condition - kwin started and one frame is rendered with default
cursor in the middle of the screen. And it verifies the moving of the
cursor without any windows shown. Whenever the cursor moves a repaint
should be triggered and the old and new area should be properly
repainted.
To support this the test needs some minor changes in KWin:
* Scene provides a frameRendered signal - needed for waiting on frame
* Scene and SceneQPainter are exported
* SceneQPainter provides access to it's Backend, so that we get to the
backbuffer
* ScriptedEffectLoader is exported for getting a list of all scripted
effects - (we don't want fade to manipulate the rendering)
Reviewers: #kwin, #plasma_on_wayland
Subscribers: plasma-devel, kwin
Tags: #plasma_on_wayland, #kwin
Differential Revision: https://phabricator.kde.org/D2046
This change introduces a tree for WindowPixmap. A WindowPixmap can
have a parent WindowPixmap and children. Each child represents a
SubSurface and references the buffer of that SubSurface.
The tree of WindowPixmaps is updated each time updateBuffer is called.
If there are new SubSurfaces new WindowPixmaps are created, for removed
SubSurfaces the old WindowPixmap gets removed. Also the ordering is
updated to reflect the current state.
This is currently rather expensive and should be changed to only rebuild
the tree if it actually changed.
The screen projection matrix is needed if an effect wants to customize
the modelview projection matrix while rendering a window and keeping
the currently applied screen projection and transformation.
With this change a new ctor overload is added to ScreenPaintData
which allows passing a projection matrix through the effects. This
allows effects to put up custom shaders with a shared projection matrix
and without having to calculate it themselves.
The projection matrix is a read-only information for the effects. There
is no way to change or overwrite it.
This describes an additional offset for the client content. On X11
our client content position matches with the window - the window
decoration is part of the overall content coordinate system.
On Wayland the content is an own texture starting at 0/0. Thus a
mapping to texture coordinates will be required when server side
decorations are provided. The new information is used in the scene's
to adjust the rendering and generating of quads.
This change introduces a mechanism for internal windows to be rendered
to a QOpenGLFramebufferObject to be composited using the texture bound
to the FBO. This is useful for in-process rendering (e.g. QtQuick) and
at the same time bypassing the windowing system.
The OpenGL context of the QOpenGLFramebufferObject needs to be sharing
with the compositing OpenGL context.
EglWaylandBackend gains support for creating textures from a
BufferInterface. At the same time it loses the possibility to use
the Xcb shm extension to load the texture. That is Xwayland is
required.
In order to support it in a better way the WindowPixmap is passed
to the Texture for loading and updating. Which is then passed to the
backend specific implementation.
The concept of Buffers do not match WindowPixmap perfectly. With X11
we had a pixmap as long as the size was the sime, then it got discarded.
With Wayland we get a new Buffer whenever the window gets damaged.
Furthermore the Buffer might get destroyed any time (especially if the
client disconnects) or the data becomes invalid (it's a shm section after
all).
This adds some constraints on how the Buffer can be used. It's suggested
that the implementing sub-classes do a deep copy of the Buffer's data
when accessing it. For OpenGL that's rather obvious, for QPainter it
needs a dedicated QImage::copy.
WindowPixmap holds a pointer to the currently used Buffer, but doesn't
guarantee that it stays valid. Every time the window gets damaged, the
pointer needs to be updated.
The QPainter based scene is the first to implement support for Buffers:
on creation a deep copy is performed, on damage the changed parts are
painted into the deep copy.
Move the buffer-swap-pending state from the compositing backends to
the Compositor class. The Compositor is the only class that needs to
access the state, and this way it to do it without calling through
a chain of virtual functions. This commit adds two new functions to
Compositor; aboutToSwapBuffers() and bufferSwapComplete(). The
backends call these functions to set and reset the buffer-swap-pending
state.
This commit also renames a number of functions and variables to make
their meaning clear.
The act of promoting the contents of the back buffer to become the
contents of the front buffer is referred to as posting the buffer,
presenting the buffer, or swapping the buffers; rendering the buffer
is what paintScreen() does.
Create and import X sync fences into GL and use them to synchronize
the kwin command stream with the X command stream.
This prevents damaged windows from being composited by kwin before
the rendering that triggered the damage events have finished on
the GPU.
Requires GL_EXT_x11_sync_object.
Tested-by: Marco Martin <notmart@gmail.com>
NOTE: this is not working completely yet, lots of code is still ifdefed
other parts are still broken.
The main difference for the new decoration API is that it is neither
QWidget nor QWindow based. It's just a QObject which processes input
events and has a paint method to render the decoration. This means all
the workarounds for the QWidget interception are removed. Also the paint
redirector is removed. Instead each compositor has now its own renderer
which can be optimized for the specific case. E.g. the OpenGL compositor
renders to a scratch image which gets copied into the combined texture,
the XRender compositor copies into the XPixmaps.
Input events are also changed. The events are composed into QMouseEvents
and passed through the decoration, which might accept them. If they are
not accpted we assume that it's a press on the decoration area allowing
us to resize/move the window. Input events are not completely working
yet, e.g. wheel events are not yet processed and double click on deco
is not yet working.
Overall KDecoration2 is way more stateful and KWin core needs more
adjustments for it. E.g. borders are allowed to be disabled at any time.
Only the X based Scenes need an overlay window, so the Compositor doesn't
need to check for it in the Wayland case.
OverlayWindow is moved from OpenGLBackend to the sub classes which need
to provide it.
The egl wayland backend registers for the callback for a rendered frame.
This allows to throttle KWin's compositor so that we don't render frames
which wouldn't end up on the screen.
For this the Scene provides a method to query whether the last frame got
rendered. By default this returns true in all backends. The Egl Wayland
backend returns true or false depending on whether the callback for the
last frame was recieved.
In case the last frame has not been renderd when performCompositing is
tried to be called, the method returns just like in the case when the
overlay window is not visible. Once the frame callback has been recieved
performCompositing is invoked again.
The pure virtual methods windowAdded, windowClosed, windowDeleted and
windowGeometryShapeChanged had identical implementations in both XRender
and OpenGL scene. They were accessing the hash with Scene::Windows which
is nowhere else used except for creating the stacking order in ::paint.
The implementations are moved to the base class, the only Scene specific
code is a pure virtual factory method to create the Scene window. This
already existed in SceneOpenGL to create either a SceneOpenGL1 or 2
window.
Also the hash of windows is a Scene private member now and the creation
of the stacking order is provided by a method, so that the Scene sub
classes do no longer need to access the stacking order at all.
REVIEW: 111207
Instead of having the Shadow factory method check the compositor type and
do the decision which Shadow sub class to create, a pure virtual method in
Scene is called which returns the specific Shadow sub class instance.
Instead of having the EffectFrameImpl check the compositor type and do
the decision which Scene::EffectFrame to create, a pure virtual method
in Scene is called which returns the specific Scene::EffectFrame.
With QtQuick2 it's possible that the scene graph rendering context either
lives in an own thread or uses the main GUI thread. In the latter case
it's the same thread as our compositing OpenGL context lives in. This
means our basic assumption that between two rendering passes the context
stays current does not hold.
The code already ensured that before we start a rendering pass the
context is made current, but there are many more possible cases. If we
use OpenGL in areas not triggered by the rendering loop but in response
to other events the context needs to be made current. This includes the
loading and unloading of effects (some effects use OpenGL in the static
effect check, in the ctor and dtor), background loading of texture data,
lazy loading after first usage invoked by shortcut, etc. etc.
To properly handle these cases new methods are added to EffectsHandler
to make the compositing OpenGL context current. These calls delegate down
into the scene. On non-OpenGL scenes they are noop, but on OpenGL they go
into the backend and make the context current. In addition they ensure
that Qt doesn't think that it's QOpenGLContext is current by calling
doneCurrent() on the QOpenGLContext::currentContext(). This unfortunately
causes an additional call to makeCurrent with a null context, but there
is no other way to tell Qt - it doesn't notice when a different context
is made current with low level API calls. In the multi-threaded
architecture this doesn't matter as ::currentContext() returns null.
A short evaluation showed that a transition to QOpenGLContext doesn't
seem feasible. Qt only supports either GLX or EGL while KWin supports
both and when entering the transition phase for Wayland, it would become
extremely tricky if our native platform is X11, but we want a Wayland
EGL context. A future solution might be to have a "KWin-QPA plugin" which
uses either xcb or Wayland and hides everything from Qt.
The API documentation is extended to describe when the effects-framework
ensures that an OpenGL context is current. The effects are changed to
make the context current in cases where it's not guaranteed. This has
been done by looking for creation or deletion of GLTextures and Shaders.
If there are other OpenGL usages outside the rendering loop, ctor/dtor
this needs to be changed, too.
AbstractThumbnailItem inherits from QQuickPaintedItem using QPainter to
do the fallback painting of icons.
The scene is adjusted to get the information from QQuickItem instead of
QDeclarativeItem. Clipping got a little bit more complex as the clip
path does not exist any more. To get it right the ThumbnailItem needs to
specify the parent it wants to be clipped to with the clipTo property.
E.g.:
clipTo: listView
The scene uses this clipTo parent item to correctly calculate the clip
region. Also the ThumbnailItem needs to have clipping enabled.
Note: this commit currently breaks TabBox as the qml and view are not
yet adjusted. In scripting the export of the item is disabled, but any
qml script using a ThumbnailItem would obviously also fail.
Cross fading with previous pixmap is achieved by referencing the old
window pixmap. WindowPaintData has a cross-fade-factor which interpolates
between 0.0 (completely old pixmap) to 1.0 (completely new pixmap).
If a cross fading factor is set and a previous pixmap is valid this one
is rendered on top of the current pixmap with opacity adjusted. This
results in a smoother fading.
To simplify the setup the AnimationEffect is extended and also takes care
about correctly (un)referencing the previous window pixmap. The maximize
effect is adjusted to make use of this new capabilities.
Unfortunately this setup has a huge problem with the case that the window
decoration gets smaller (e.g. from normal to maximized state). In this
situation it can happen that the old window is rendered with parts outside
the content resulting in video garbage being shown. To prevent this a set
of new WindowQuads is generated with normalized texture coordinates in
the safe area which contains real content.
For OpenGL2Window a PreviousContentLeaf is added which is only set up in
case the crass fading factor is set.
REVIEW: 110578
The behavior for creating a pixmap for a window is moved from Toplevel
into a dedicated class WindowPixmap. Scene::Window holds a reference to
this class and creates a new WindowPixmap whenever the pixmap needs to be
discarded. In addition it also keeps the old WindowPixmap around for the
case that creating the new pixmap fails. The compositor can in that case
use the previous pixmap which reduces possible flickering. Also this
referencing can be used to improve transition effects like the maximize
windows effect which would benefit from starting with the old pixmap.
For XRender and OpenGL a dedicated sub-class of the WindowPixmap is
created which provides the additional mapping to an XRender picture and
OpenGL texture respectively.
BUG: 319563
FIXED-IN: 4.11
REVIEW: 110577
either by
- forcing fullrepaints unconditionally
- turning a repaint to a full one beyond a threshhold
- completing the the backbuffer from the frontbuffer after the paint
BUG: 307965
FIXED-IN: 4.10
REVIEW: 107198
Add an option to kcmcompositing in the 'Advanced' tab, to enable or
disable color correction. It is specified that it's experimental and it
needs Kolor Manager.
Before painting for a particular screen, ColorCorrection::setupForOutput
should be called.
A screen property is added for WindowPaintData.
In kwinglutils, The fragment shaders are intercepted before being
compiled and they get a couple of lines of code inserted in order to do
the color correction. This happens only when color correction is enabled, of
course.
For D-Bus communication with KolorServer, everything is async.
The implementation basically manages a set of color lookup tables for
different outputs and for different window regions. These are taken via
D-Bus. Each lookup table has around 700 KB.
This commit reintroduces the changes from the former merge with the
"color2" branch. In this form, it can be easily reverted.
REVIEW: 106141
This merge is incomplete and it does not include the review number of
the associated review request. It should have been pushed as a single
commit, because the merged commits were not intended to be published in
their form.
This reverts commit dcba90263069a221a5489b1915c5cf1ca39d090c, reversing
changes made to 50ae07525c7fde07794e7548c3d6e5a69cb1a89d.
Conflicts:
kwin/scene_opengl.cpp
kwin/scene_opengl.h
Results in cleaner changes.
Put all the color correction stuff from SceneOpenGL in SceneOpenGL2.
Conflicts:
kwin/eglonxbackend.cpp
kwin/glxbackend.cpp
kwin/scene.h
kwin/scene_opengl.cpp
kwin/scene_opengl.h
Replace dynamic_casts to check the type for for Toplevel by isFoo()
calls and use static_casts in such blocks.
Furthermore method shape() returns now a constant reference instead of a
copy of the QRegion.
REVIEW: 106364
The handling for creating and managing the OpenGL context is
split out of the SceneOpenGL into the abstract OpenGLBackend
and it's two subclasses GlxBackend and EglOnXBackend.
The backends take care of creating the OpenGL context on the
windowing system, e.g. on glx an OpenGL context on the overlay
window is created and in the egl case an EGL context is created.
This means that the SceneOpenGL itself does not have to care
about the specific underlying infrastructure.
Furthermore the backend provides the Textures for the specific
texture from pixmap operations. For that in each of the backend
files an additional subclass of the TexturePrivate is defined.
These subclasses hold the EglImage and GLXPixmap respectively.
The backend is able to create such a private texture and for
that the ctor of the Texture is changed to take the backend as
a parameter and the Scene provides a factory method for
creating Textures. To make this work inside Window the Textures
are now hold as pointers which seems a better choice anyway as
to the member functions pointers are passed.
The Scene has always been created and destroyed inside what is
now the split out compositor. Which means it is actually owned
by the Compositor. The static pointer has never been needed
inside KWin core. Access to the Scene is not required for the
Window Manager. The only real usage is in the EffectsHandlerImpl
and in utils.h to provide a convenient way to figure out whether
compositing is currently active (scene != NULL).
The EffectsHandlerImpl gets also created by the Compositor after
the Scene is created and gets deleted just before the Scene gets
deleted. This allows to inject the Scene into the EffectsHandlerImpl
to resolve the static access in this class.
The convenient way to access the compositing() in utils.h had
to go. To provide the same feature the Compositor provides a
hasScene() access which has the same behavior as the old method.
In order to keep the code changes small in Workspace and Toplevel
a new method compositing() is defined which properly resolves
the state. A disadvantage is that this can no longer be inlined
and consists of several method calls and pointer checks.
The implementation consists of a class in libkwineffects.
There are some slight modifications in the compositor. Regions for
different outputs are drawn at different times.
Currently only per output color correction is implemented. However, the
grounds are prepared for implementing per window color correction
easily.
The ColorCorrection class needs to communicate via D-Bus with a KDED
module, KolorServer, which is a part of KolorManager.
The only visible part for the user consists of a check box in the
advanced tab for the compositing KCM.
The actual correction is done by injecting a piece of code in the
fragment shader, code that does a 3D lookup into a special color lookup
texture. The data for these textures is obtained from KolorServer. All
D-Bus calls are async.
The LanczosFilter has a FBO in screen size. When the screen
geometry changes this FBO has to be recreated. To go completely
sure the lanczos filter gets deleted on screen changes.
To achieve this the LanczosFilter is wrapped inside a
QWeakPointer so that we can track when it got deleted. This
brings an additional advantage by delaying the creation of the
shader till it is really needed, that is when for the first
time a window thumbnail with lanczos is rendered.
BUG: 296065
FIXED-IN: 4.9.0
REVIEW: 105479
