Semi-automated, just needed ~20 manual fixes:
$ find \( -iname \*.cpp -or -iname \*.h \) -exec perl -pe 's/(\.|->)load\(\)/$1loadRelaxed\(\)/g' -i \{\} +
$ find \( -iname \*.cpp -or -iname \*.h \) -exec perl -pe 's/(\.|->)store\(/$1storeRelaxed\(/g' -i \{\} +
It can be easily improved (e.g. for store check that there are no commas
after the opening parens). The most common offender is QLibrary::load,
and some code using std::atomic directly.
Change-Id: I07c38a3c8ed32c924ef4999e85c7e45cf48f0f6c
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Try to replace all wordings like '.. to 0' with '.. to \nullptr'. Also
checked for 'null pointer' and similar.
Change-Id: I73341f59ba51e0798e816a8b1a532c7c7374b74a
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
FBOs cannot be shared. Streamline the remainder of the sentence.
Change-Id: I654b23f86f4ee4ea2ca8e71958464d9f55394297
Reviewed-by: Martin Smith <martin.smith@qt.io>
Replace null and '\c nullptr' with \nullptr in the documentation.
Change-Id: I58934eea06943309ba895833f1991629870ab45b
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Replace 0 with \nullptr in the documentation.
As a drive-by also replace some 0 with nullptr in the corresponding
code.
Change-Id: Ieffbfffa76e3018257b667a3e8ad6e3b88486cde
Reviewed-by: Paul Wicking <paul.wicking@qt.io>
Qt for Python users reading the documentation assume that int(0) can
be passed for pointer parameters. Use the newly introduced \nullptr to
disambiguate this.
In a follow-up step, the \nullptr macro can be defined as None
when generating the Qt for Python documentation.
Task-number: PYSIDE-903
Change-Id: I3a45f87175a0668ab5f3f95f0aff409f7e3ef027
Reviewed-by: Cristian Maureira-Fredes <cristian.maureira-fredes@qt.io>
Reviewed-by: Christian Tismer <tismer@stackless.com>
Reviewed-by: Venugopal Shivashankar <Venugopal.Shivashankar@qt.io>
Reviewed-by: Paul Wicking <paul.wicking@qt.io>
Instead of doing it up front and then restoring the current context
if the makeCurrent call failed.
By delaying the update, QPlatformOpenGLContext::makeCurrent has the option
to check which context was current before the call.
Change-Id: I458f5c00d7c820acbe2f8552b0513fce99b9b683
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
Reviewed-by: Laszlo Agocs <laszlo.agocs@qt.io>
The case of the platform context failing makeCurrent is the exception.
Change-Id: I606e6fb379f84e886bff9e27355dda9b31ff9ebe
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
Reviewed-by: Laszlo Agocs <laszlo.agocs@qt.io>
This is the squashed diff from wip/webassembly to dev.
Done-with: Peng Wu <peng.wu@intopalo.com>
Done-with: Sami Enne <sami.enne@intopalo.com>
Done-with: Morten Johan Sørvig <morten.sorvig@qt.io>
Started-by: Andrew Knight <andrew.knight@intopalo.com>
Change-Id: I6562433c0a38d6ec49ab675e0f104f2665f3392d
Reviewed-by: Lorn Potter <lorn.potter@gmail.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@qt.io>
Change-Id: Id0118e44e1ed44fecdd803dbc57ca0588a1afec0
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
Reviewed-by: Martin Smith <martin.smith@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Seems this issue is still there with Adreno 5xx and 6xx is suspected to
have it as well (no device to test though), so added both 5xx and 6xx
to cover these. Updated 30x to 3xx in order to cover Adreno 320 and 330
as per https://en.wikipedia.org/wiki/Adreno.
Amends 9ae028f507
Task-number: QTBUG-66702
Change-Id: I6ce3f6499d3ff9da884be45039e5f5e0990f7e1f
Reviewed-by: Laszlo Agocs <laszlo.agocs@qt.io>
To fix QTBUG-52493 we tied the exposed state of a window to the
application being in the foreground. This has the result of a
visible flash of black between hiding the launch screen and showing
the first frame of the application, as the application is still
waiting for UIApplicationStateActive to begin rendering, which
happens after iOS hides the launch screen.
According to the iOS OpenGL ES Programming Guide, it should be safe
to render GL in UIApplicationStateInactive as well, and even in
UIApplicationStateBackground, as long as the rendering finishes
before the UIApplicationDidEnterBackgroundNotification returns.
To ensure that we catch any bugs in this area, checks have been
added that verify that no rendering happens while in the background
state.
Task-number: QTBUG-63229
Task-number: QTBUG-52493
Task-number: QTBUG-55205
Change-Id: Ib42bedbeddd7479ab0fb5e5b7de9f5805658e111
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@qt.io>
The glyph cache resize is clearly doing something that is not actually
legal with OpenGL ES. Until this gets investigated properly, add the
Vivante GC2000 (found in the commonly used i.MX6 quad) to the list since
reports show that the issue occurs there as well.
Task-number: QTBUG-49490
Change-Id: Ia890346d8dbb1691bc113e2ef522713ba6709393
Reviewed-by: Louis Kröger <louis.kroeger@gmail.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@qt.io>
On some ARM devices the font glyph generation is broken
Add an environment variable to enable workaround_brokenFBOReadBack
in QOpenGLContext, to fix font rendering on such devices as
Mali and Adreno
Change-Id: I9cc99ecb8b71a35bc369ec9dd11b877016b1179e
Reviewed-by: Laszlo Agocs <laszlo.agocs@qt.io>
Avoid looking up by key twice in a row in various locations, but instead
using iterators and index lookup.
Change-Id: I61a079115199ab9c041ad3a26d36b45ee3f775e0
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The problem of the check is that it makes Qt non compliant with OpenGL
thread affinity semantics. One is allowed to make a GL context
current on any thread, without the Qt-specific idea of moving
the QOpenGLContext to that thread first. Moreover, the move is
plain impossible if the user needs to handle the context
to 3rd party code which assumes the context to have GL semantics.
Add an application flag to disable that.
Change-Id: I55ca02ee62f8cc171a9a1bddef5331ad0949c061
Reviewed-by: Laszlo Agocs <laszlo.agocs@theqtcompany.com>
The printf-style version of QDebug expands to a lot less code than the
std::ostream-style version. Of course, you pay in type safety (but
compilers warn about it these days), you cannot stream complex Qt
types and streaming QStrings is awkward, but in many cases you
actually improve on readability.
But the main reason is that something that's not supposed to be
executed under normal operation has no business bloating executable
code size.
This is not an attempt at converting all qWarnings() to printf-style,
only the low-hanging fruit.
In this first part, replace
qWarning() << "...";
with
qWarning("...");
In QTransform shared warning strings.
Saves 3KiB in text size on optimized GCC 5.3 AMD64 builds.
Change-Id: I142a8020eaab043d78465178192f2c8c6d1cc4f9
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@theqtcompany.com>
Reviewed-by: Kai Koehne <kai.koehne@theqtcompany.com>
Saves some code, is easier to maintain and will allow for some
more nice refactoring.
Change-Id: Ica7ae8e9d36acbe6586e488bc6aff114336c65bb
Reviewed-by: Laszlo Agocs <laszlo.agocs@theqtcompany.com>
Add an getProcAddress(const char *) overload to QOpenGLContext,
and refactor the QPA interface to take a const char *. Like this
we can avoid lots of mallocs when resoving GL methods.
Change-Id: Ic45b985fbaa0da8d32ba3e3b485351173352ca6f
Reviewed-by: Laszlo Agocs <laszlo.agocs@theqtcompany.com>
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
... by replacing them with C++11 range-for loops.
To avoid detaches of these mutable Qt containers,
wrap the container in qAsConst().
Change-Id: I90fd517ad542ef92034403c15ebb8300a56ac693
Reviewed-by: Lars Knoll <lars.knoll@theqtcompany.com>
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
From Qt 5.7 -> LGPL v2.1 isn't an option anymore, see
http://blog.qt.io/blog/2016/01/13/new-agreement-with-the-kde-free-qt-foundation/
Updated license headers to use new LGPL header instead of LGPL21 one
(in those files which will be under LGPL v3)
Change-Id: I046ec3e47b1876cd7b4b0353a576b352e3a946d9
Reviewed-by: Lars Knoll <lars.knoll@theqtcompany.com>
This code seems to be hit three times per frame in a hybrid QQ2/Qt3D app,
quickly generating thousands of allocations here via the detach in begin().
Heaptrack confirms they are all gone now.
Change-Id: I7d09f3f5c2b24a2590b701216715055751306e82
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
If, after checking a condition, we issue a qFatal()
or a qCritical(), by definition that check is
unlikely to be true.
Tell the compiler so it can move the error handling
code out of the normal code path to increase the
effective icache size.
Moved conditional code around where possible so that
we could always use Q_UNLIKELY, instead of having to
revert to Q_LIKELY here and there.
In some cases, simplified the expressions newly wrapped
in Q_UNLIKELY as a drive-by.
Change-Id: I67537d62b04bc6977d69254690c5ebbdf98bfd6d
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
Reviewed-by: Olivier Goffart (Woboq GmbH) <ogoffart@woboq.com>
Returning a dangling pointer is no good. Do what QOffscreenSurface does:
just null it out.
Task-number: QTBUG-42803
Change-Id: I01a6db9ae8974a1c78157ebc67097c8dac3a6b6e
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
Using the approach we already do for some GLES 3.0 functions we can provide a cross-platform,
cross-GL-GLES wrapper for ES 3.0 and 3.1 functions too.
Applications only have to take extra care about the version requests (context version
and version directives in shader code), the rest of their code can stay the same across
desktop/mobile/embedded, even when ES 3 functions are used.
The new functions are placed to a new subclass which is placed between QOpenGLFunctions
and the internal QOpenGLExtensions. This is necessary because, unlike with QOpenGLFunctions,
there is no guarantee that these functions are always available in all configurations.
When running on desktop OpenGL, we resolve as usual. If the OpenGL version contains the
function in question, either in core or as an extension, it will all just work. This is
handy because it does not rely on 4.x extensions like GL_ARB_ESx_compatibility, and so
ES 3.0 functions will be functional on OpenGL 3.x systems too by just setting a 3.x version
number in the QSurfaceFormat.
We will no longer qFatal on broken systems where the driver returns a 3.0 or 3.1 context without
the corresponding functions present. Instead, we show a warning and gracefully fall back to
resolving as usual, via eglGetProcAddress or similar. For functions that are available in ES2 as
an extension this may just work fine.
Added also an example that runs identically both with OpenGL and OpenGL ES 3 and utilizes some
ES 3.0 features like instanced drawing.
[ChangeLog] Added QOpenGLExtraFunctions providing OpenGL ES 3.0 and 3.1 function wrappers in
a cross-platform manner.
Task-number: QTBUG-46161
Change-Id: I9f929eb61946c35c415b178c4d6ab2c1c958684e
Reviewed-by: Paul Olav Tvete <paul.tvete@theqtcompany.com>
QOpenGLContext::defaultFramebufferObject() knows nothing about QOpenGLWidget
and QQuickWidget. The problem is that this function (and others that rely on it)
is expected to give the widget's backing FBO in paintGL() and friends.
To overcome this, we have to provide a way for such widgets that indicate what is
the expected "default fbo".
Task-number: QTBUG-43269
Change-Id: I43f439f8609382b9f7004707ab0ef9f091952b4f
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
QOpenGLFunctions allows both deriving from it and getting an instance
via QOpenGLContext::functions().
Unsurprisingly a large number of users attempt to use the versioned
function wrappers in the same way. Unfortunately this approach was
not that well supported.
Besides some potential base class exporting issues the real blocker for
QOpenGLWidget - or any versionfunction subclass whose associated context
changes during its lifetime - is that the functions instances could
only be initialized once. Unlike instances retrieved via
QOpenGLContext::versionFunctions(), instances created "manually" were not
deinitialized upon the destruction of the associated context because
context did not know about them.
A pattern like
initializeOpenGLFunctions();
delete context;
create new context and make it current
initializeOpenGLFunctions();
is working fine in QOpenGLFunctions-derived classes but not with the
versioned ones.
To overcome this, start registering such instances to the context too.
QOpenGLContext::destroy() can then reset the internal state so a
subsequent initializeOpenGLFunctions() will reinitialize properly instead
of bailing out mistakenly thinking that everything is ready to use.
Task-number: QTBUG-45199
Change-Id: Ia1420bcccb33c51508698b7a1b036c7544a66e74
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
We're documenting that this context exists, therefore applications
are allowed to assume that it exists. For the same reason, we
should expose it as public API -- an application should be allowed
to upload GL objects before creating or showing QOpenGLWidgets
that use them.
Change-Id: Iedbba7cd393018d93833fd2a6b02cce0639a89b4
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Reviewed-by: Laszlo Agocs <laszlo.agocs@theqtcompany.com>
Qt copyrights are now in The Qt Company, so we could update the source
code headers accordingly. In the same go we should also fix the links to
point to qt.io.
Outdated header.LGPL removed (use header.LGPL21 instead)
Old header.LGPL3 renamed to header.LGPL3-COMM to match actual licensing
combination. New header.LGPL-COMM taken in the use file which were
using old header.LGPL3 (src/plugins/platforms/android/extract.cpp)
Added new header.LGPL3 containing Commercial + LGPLv3 + GPLv2 license
combination
Change-Id: I6f49b819a8a20cc4f88b794a8f6726d975e8ffbe
Reviewed-by: Matti Paaso <matti.paaso@theqtcompany.com>
Try making it compatible with QOpenGLContext::functions() which returns
an already initialized QOpenGLFunctions.
Unfortunately we cannot make them 100% compatible because functions()
conveniently requires that the context (or a sharing context) is current.
versionFunctions() has no such requirement and we cannot safely introduce
it anymore.
What we can do is to state that as long as the context is the current one,
the initializeOpenGLFunctions() call can be omitted. If another context (or
no context) is current, the call will still be needed, like it is today.
Also, we require that the exact same context is current. Sharing does not
count since the exact behavior of sharing contexts with different versions
is unknown.
[ChangeLog][QtGui] initializeOpenGLFunctions() no longer needs to be called when
querying a versioned function wrapper object via QOpenGLContext::versionFunctions().
Change-Id: I0b4d1ae1f780da3f5dec9fc8dc67255c13faab6e
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Applications have no public API to check for the ThreadedOpenGL flag
reported by the platform plugins. Add a static function to QOpenGLContext
to give interested applications a way to decide if they should do
rendering on separate threads.
Task-number: QTBUG-44208
Change-Id: I629332512ab31fe49f842065a911270cd6cea7aa
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
On platforms like Windows (and presumably on mobile devices too)
the loss of the context (e.g. the underlying D3D device in case of
ANGLE) is an event that can happen randomly and needs sufficient
handling.
Enhance QOpenGLContext::isValid() with the purpose of indicating
context loss.
Currently only the Windows EGL backend (ANGLE) has support for it.
Other platforms may be added later.
Task-number: QTBUG-43263
Change-Id: I8177694c1ee7cebbd5d330e34757fd94c563e6d6
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@theqtcompany.com>
This function is private.
Task-number: QTBUG-35907
Change-Id: I370c0bfd8fda11c68ee76ee42967f117a81b381c
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Giuseppe D'Angelo
Christian Ehrlicher
Qt Forward Merge Bot
Friedemann Kleint
Tor Arne Vestbø
Morten Johan Sørvig
Herman van Hazendonk
Liang Qi
Tero Alamäki
Lars Knoll
Laszlo Agocs
Marco Martin
Allan Sandfeld Jensen
Anton Kudryavtsev
Liang Qi
Laszlo Agocs
Marc Mutz
Lars Knoll
Andy Maloney
Jani Heikkinen
Frederik Gladhorn
Volker Krause
Sérgio Martins
Laszlo Agocs
Sze Howe Koh