QByteArray::assign() re-uses existing unshared capacity(), if any, and
is therefore potentially more efficient than '= QByteArray(~)' (and
never slower).
Task-number: QTBUG-106201
Pick-to: 6.6
Change-Id: Ic51fc57eb0f84d3624ad3447c93a74241ab3612b
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Marc Mutz <marc.mutz@qt.io>
Introduce QT_SUPPORTS_INT128 and QT_NO_INT128 marcos to handle 128-bit
types. These macros allow to undef Qt's own 128-bit types and the
related code, but keep the compiler definitions unchanged.
This is required for Qt Bluetooth, where we need to use
QT_BLUETOOTH_REMOVED_SINCE to get rid of the APIs using
QtBluetooth-specific struct quint128 which clashes with the 128-bit
types. The idea is to use QT_NO_INT128 in Qt Bluetooth's
removed_api.cpp instead of directly undef'ing __SIZEOF_INT128__,
because the latter is UB.
This commit amends befda1acca.
Pick-to: 6.6
Change-Id: Ia2c110b5744c3aaa53eda39fb44984cf5a01fac2
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Commit ca54b741d6 used the internal
has_operator_equal (and commit 01d94760d8
copied that for has_operator_less_than) instead of using the recursive
expander that was being used here. That assumed that the contained type
in std::optional would always be the last final check, which is an
incorrect assumption.
Fixes: QTBUG-115646
Pick-to: 6.6 6.5
Change-Id: Ifbf974a4d10745b099b1fffd177702934bec27ff
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Replicate qvulkaninstance.h's logic. The idea being to allow
to compile code that uses QRhi (so includes qrhi.h) with a pre-built
Qt that was built with Vulkan support, but there is no Vulkan SDK
(and so vulkan.h) on the developer's system. Normally this is not
something we care about, e.g. compiling an application using
QVulkanInstance or any other Vulkan functionality implies one has
to have the SDK installed or at least the headers available in one
way or another.
However, including qrhi.h (and so qrhi_platform.h) happens in some
cases in applications that do not themselves initialize a QRhi,
i.e. they do not care about the QRhiVulkan* structs, because
they only ever use a QRhi retrieved e.g. from the QQuickWindow.
Design Studio is one example. To allow building DS against a Qt
that is Vulkan-enabled (e.g. the official Qt releases) but in
an environment that has no Vulkan headers, we skip the QRhiVulkan*
stuff when there is no vulkan.h available.
Pick-to: 6.6
Change-Id: Ic81250e6c90939a38f79cab1438fbb9f483f56df
Reviewed-by: Miikka Heikkinen <miikka.heikkinen@qt.io>
Use the new Q_NODISCARD_X macro to point users that ignore the QFuture
returned from QtConcurrent::run() to QThreadPool::start(), which does
the same thing, but doesn't return a future, so is better suited for
the fire-and-forget use-case the OP of and commentators on
QTBUG-111875 cited.
Can't pick to older branches, since Q_NODISCARD_X is 6.7+.
Task-number: QTBUG-111875
Change-Id: If0bf920ecc0fb59b9a9a9931ea9dc30f7abff1b7
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
QAndroidPlatformScreen::addWindow() asserted the existence of a backing
store, when a window is added. But QQuickWindow neither needs nor has a
backing store.
This patch removes the assertion, which was introduced in
dbb072eb28
Task-number: QTBUG-97482
Pick-to: 6.6 6.5 6.2 5.15
Change-Id: Ifb43f83d39e16b972754e86d0ff9633f58dac3f8
Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>
The blacklisted test does not crash, as its expected output indeed shows.
Pick-to: 6.5 6.6
Change-Id: I07522a7d065b5f39620975a3546bcd156024c41d
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
In porting the selftest machinery to Catch2 in 24e83de8d1 we
accidentally added an unconditional early return when determining
whether to check for unexpected stderr output, resulting in not
checking error output on any platform.
The return statement has now been moved into the Q_CC_MINGW
condition, but as we now seem to have similar issues on macOS
and Linux with some of the tests outputting "Received signal 6
(SIGABRT)" as we do for QEMU, we need to add a few more explicit
early return conditions to the function.
Pick-to: 6.5 6.6
Change-Id: I7a25f000843b5f1003a5db883f08285185046b46
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Nothing prevents client code from calling QLocale::setDefault() before
we ever instantiate QLocale::system() - aside from some quirks that
mean setDefault(), currently, does instantiate QLocale::system() to
force initialization of defaultLocalePrivate - so using defaultIndex()
could set the system QLocalePrivate instance's index incorrectly.
In any case, even if the index is initially set correctly, a
subsequent change to the system locale would change the correct index;
and nothing outside QLocale::system() has access to the instance that
would then be remembering an out-of-date index.
Actually tripping over that inconsistency took some deviousness, but
was possible. The index is (currently) only used for month name
lookups and those special-case, for the Roman-derived calendars, the
system locale, to only use the index if the system locale offers no
name for a month. Meanwhile, updateSystemPrivate() uses the fallback
locale's index for its look-up of which CLDR data to copy into the
fallback QLocaleData for the system locale.
None the less, a non-Roman calendar's lookup will go via the index to
get at the CLDR data for that calendar, thereby exposing the system
locale's index to use; and, sure enough, a setDefault() could lead
that to produce wrong answers.
In QLocale::system() there's a cached QLocalePrivate, whose index we
need to ensure stays in sync with the active system locale. So pass
its &m_index to systemData(), which will now (when passed this) ensure
it's up to date. Since we always have called systemData(), to ensure
it is up to date, we can skip that update in the initialization of the
cached private and use m_index = -1 to let systemData() know when it's
in the initial call, thereby making the static cache constinit.
Amended a test to what proved the issue was present.
Change-Id: I8d7ab5830cf0bbb9265c2af2a1edc9396ddef79f
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
It is part of a solution to produce notarizable binaries for
application using Qt WebEngine.
Pick-to: 6.6 6.5
Task-number: QTBUG-99555
Change-Id: Ice0bbcfb12829906328f55158b0c37f03878f17c
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
The member function is never implemented.
Change-Id: Iee93c7d24df40b48b897d937889bb9556b8feb1a
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Was missing for some reason when this got added in 6.4.
Pick-to: 6.6 6.5
Change-Id: I4f8b780c7119649688d8d8a197e8ef59730b0b61
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
If a callback deleteLaters a QRhiResource that should still be
handled and not left unreleased. Swap the order.
Pick-to: 6.6 6.5
Change-Id: I8419a28a9db5bb59f768ab5820dfaf593464d6d2
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
And the ability to deregister.
Going to be required by QRhiWidget.
Change-Id: If185cbed2faa042098ac1f6bb1d6daaffd834377
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
caps.apiVersion is inst->apiVersion() limited by the version the
physical device reports. Therefore, use caps.apiVersion when
setting up the memory allocator, like it is done everywhere else.
This is expected prevent vma init failures on lavapipe in some Linux
distros.
Pick-to: 6.6 6.5
Change-Id: I4e693820c95a5e0174846afb20e42aadd56034d6
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
One may very well force a 3D API not functional. In this case there will
be no QRhi. A bunch of warnings will be printed (which is good), but
then flush() should not do anything, and that was not handled before.
Pick-to: 6.6 6.5
Change-Id: I82139070311152c959d39a553842f4462d8e7811
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
...and add a single toggle that can be used to switch between
regular (auto-retaining) and unretaining command buffers.
Change-Id: I9ef4b149f172415733e590e8320165358e29fa16
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
Activity.getReferrer does not only return app IDs but also URLs if
Intent.EXTRA_REFERRER is set on the Intent. In the case of Chrome the referrer
is set to the website triggering the Intent. To improve the detection of the
calling app we check first if the browser specific
Browser.EXTRAS_APPLICATION_ID is set. If it is not set we fall back to
Intent.getReferrer.
Pick-to: 6.6 6.5
Change-Id: I33d1edd52de98486d9616713e531ea20ada87bcb
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
- Pass the QList by value, no heap allocation and no plain new/delete
- A lambda means that there isn't runtime string-based lookup to find
the member method in QObjectPrivate
The code is only a couple of lines and used in a single place, so might
as well move the code from _q_reregisterTimers to the local lambda.
Modify tst_moc to account for the numer of methods in QObjectPrivate
changing. The test had hardcoded numbers.
Change-Id: I07906fc7138b8e601e4695f8d2de1b5fdd88449c
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Showing code that cannot possibly compile is bad practice.
(vkAllocateCommandBuffers is in QVulkanDeviceFunctions, not
QVulkanFunctions)
Rather use a member function that is actually in QVulkanFunctions.
Pick-to: 6.6 6.5 6.2
Change-Id: I3063a8a229152144ce866e915eb4c14ecc58e9ce
Reviewed-by: Laszlo Agocs <laszlo.agocs@qt.io>
Ironically enough when the old macOS 13 problem is no longer occurring
(at least on an M1 Mini with 13.2.1), so the OS version based
differentiation could likely be removed, there are now reports about
problems with old Intel hardware and earlier OS versions.
Therefore, get rid of the OS version based logic, and rather add
a global flag that is enabled unconditionally for now, which
disables MTLBinaryArchive usage altogether.
How much we lose is not very clear anyway. The OS performs its
own persistent caching (most likely), hidden from applications.
Thus what we lose really is the fine grained control over the
data (e.g. specifying a custom storage file via
QQuickGraphicsConfiguration), with the possibility of pre-seeding
for the first run. As the performance of subsequent runs is less
likely to change in any significant way, this is seen as not
a big deal for Apple hardware in general so we might just live
with this. (and on macOS 13+ this was already crippled anyway
due to the OS version logic)
Pick-to: 6.6 6.5
Fixes: QTBUG-114338
Task-number: QTBUG-108216
Change-Id: If7b908baea2093f6882674ebfbdc18e770d6503e
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Running a Vulkan-based QRhiWidget (upcoming in 6.7) and QQuickWidget
shows the "...does not support OpenGL-only windows" warning which is
pointless nowadays. Since Qt 6.4 RasterGLSurface is not in use,
whereas the surface type can very well be VulkanSurface too.
Pick-to: 6.6 6.5
Change-Id: I790767e683b2a4cffc99cbc38015aca809cf83c5
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
Since this was written we added support in windeployqt to find these
DLLs similarly to how the old d3dcompiler_*.dll is bundled.
Therefore, rephrase a little bit.
Change-Id: I13eafea67607900307ff84da3323e84e066de9f8
Reviewed-by: Laszlo Agocs <laszlo.agocs@qt.io>
Should be no different from MetalSurface and OpenGLSurface.
Pick-to: 6.6 6.5
Change-Id: I529f2904a43e44376e9f2da2489fac57670d954d
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
This reverts commit f647375275. There are
no signed integer overflows in atomics.
For the non-atomic side, unlike the commit being reverted, we fix the
signed integer overflow by removing the "signed" part instead of the
"overflow" part, and use unsigned integer overflows instead.
Change-Id: I53335f845a1345299031fffd176f5ba479163e44
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
After 449b45ff34 we emit aboutToQuit
before quitting the main event loop, matching the documentation and
semantics of "about to". Clarify this in the documentation, so that
client do not rely on a missing main event loop to trigger processing
of root level deleteLater() calls in a nested event loop.
Pick-to: 6.5 6.6
Change-Id: Ifd116394a1960814edf69f20c5aaff6a8e5cfc24
Reviewed-by: Jarek Kobus <jaroslaw.kobus@qt.io>
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
QConfFileSettingsPrivate::initAccess is not async, as the commit
introducing compile guards around initAccess in ctor of
QConfFileSettingsPrivate suggests. Calling initAccess is necessary
to read stats from a conf file.
This makes 50 more tests in tst_qsettings pass.
Task-number: QTBUG-115509
Change-Id: I99e8970413a14d7b595be23d42bf46faa26b3dd7
Reviewed-by: Lorn Potter <lorn.potter@gmail.com>
We can use plain data() for the recent and imminent that are further
from forLocalMSecs than its UTC equivalent can be: more than 99% of
the time these shall agree and we can save some fiddly details. We can
also use data(recent) as place-holder for the last transition's data,
saving a need to call previousTransition() - which may be expensive
(at least on Darwin). This avoids a quirk around minMSecs(), used by
the MS backend for a fake transition, so marginally simplifies some
initializations (and comments). It also gives us the information we
need for the fall-back (when transitions don't help or aren't known),
that uses the offsetFromUtc from recent and imminent, which the early
data() values supply.
Task-number: QTBUG-104012
Change-Id: I08403d129c0fec478b09de5b660f4a0a97b17148
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This commit adds some initial support for handling the 'L' suffix after
numbers. This one is especially important given that the __cplusplus
define is using it.
Other suffixes will be handled in some later commit, which should also
unify the already divergent parse behavior between DIGIT and PP_DIGIT
parsing (e.g. when it comes to byte prefixes).
Task-number: QTBUG-83160
Task-number: QTBUG-115558
Pick-to: 6.6 6.5 6.2
Change-Id: Ie61eae49c468abfaee80e7e4f7097917a254dc0e
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
mouseMoveEvent overrides in QColorLuminancePicker and QColorPicker
have triggered value changes unconditionally. This happened under the
assumption that the widget attribute WA_Hover is not set (which is the
default behavior). In that case, mouseMoveEvents are only delivered if
a button is pressed.
If WA_Hover is set - e.g. by applying a style sheet - mouseMoveEvents
get delivered also when no button is pressed. This leads to faulty
behavior: The color and the luminance change, whenever the mouse is
moved into the respective widget. Color/luminance are changed to the
value representing the edge on which the mouse has left the area.
This patch changes both mouseMoveEvent overrides. They return early
to avoid hovering changing the colors of the luminance/color picker,
but ignore() the event in case anything behind the picker needs hover.
Since this is a purely graphical effect, an autotest was not added.
Fixes: QTBUG-115516
Pick-to: 6.6 6.5 6.2 5.15
Change-Id: I000d113a1c81c46799cbb5197bf9acb3849e7d3b
Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>
The PRELOAD variable in wasm_shell.html wouldn't get substituted with
preload list when using qmake. Do that as it is done in cmake.
Fixes: QTBUG-115507
Change-Id: I3c659626dc6fa6f4fdf9e31bd62b87fc6a7d8bbe
Reviewed-by: Lorn Potter <lorn.potter@gmail.com>
Reviewed-by: Piotr Wierciński <piotr.wiercinski@qt.io>
Previously, a target name with a JS special character (like, -, for
example), would lead to an invalid export name being generated for
WASM modules. Sanitize these by replacing any non-alphanumeric
character with underscores, as is done for feature names.
This is a qmake version of 58a47edda1
Fixes: QTBUG-115506
Change-Id: I7c84076be54da91cf0f707c1613afc382acdcb83
Reviewed-by: Lorn Potter <lorn.potter@gmail.com>
Reviewed-by: Piotr Wierciński <piotr.wiercinski@qt.io>
We dropped support for macOS 10.3 in Qt 4.6, 14 years ago.
But the logic to only 'fake' the unloading on Q_OS_DARWIN
remains. Also, add a statement explaining the behavior
in more detail.
Pick-to: 6.5 6.6
Change-Id: I62ec7df7c4b807f84c96619f78b3cef704c51335
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
There is no QT_CONFIG_thread.
Thanks to Alexey Edelev for spotting why I couldn't run the test.
Pick-to: 6.6 6.5
Change-Id: I11c99d9b1ff8fed67b118028b76fba8ee6db3c42
Reviewed-by: Alexey Edelev <alexey.edelev@qt.io>
Qt insight's TP is using plugin type "generic" so that insight plugins
can be deployed for every Qt application. As "generic" plugins are part
of QtGui windeployqt deploys these if a dependency to QtGui is found.
As defaulting to a deployment of insight plugins might cause confusion
among users, we make deployment of these plugins explicit if the module
is available.
Pick-to: 6.6 6.5 6.2 5.15
Change-Id: I9d2a8595373d5a15b7afbeaf7174226563b1cb6f
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Reviewed-by: Timothée Keller <timothee.keller@qt.io>
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Those tests often fail on Android 13 on RHEL 8.6 and 8.8. This patch
skips them to unblock CI while the underlying reasons are investigated
and fixed.
Task-number: QTQAINFRA-5606
Change-Id: If088d69c2160470ef50b2e74cd9b9399451c526d
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Axel Spoerl <axel.spoerl@qt.io>
