We've been requiring C++17 since Qt 6.0, and our qAsConst use finally
starts to bother us (QTBUG-99313), so time to port away from it
now.
Since qAsConst has exactly the same semantics as std::as_const (down
to rvalue treatment, constexpr'ness and noexcept'ness), there's really
nothing more to it than a global search-and-replace, with manual
unstaging of the actual definition and documentation in dist/,
src/corelib/doc/ and src/corelib/global/.
Task-number: QTBUG-99313
Change-Id: I4c7114444a325ad4e62d0fcbfd347d2bbfb21541
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
If an error is returned from qt_safe_poll inside the event dispatcher
the error is currently printed and ignored.
For most cases this behavior seems to be fine, but when used in critical
systems e.g. automotive or medical, a error might indicate a more severe
problem and the application should be stopped instead.
The system can then decide itself what to do e.g. restarting the
application using a watchdog.
Change-Id: Iaf5abb20bb3941eaeff19d14e41c395c88fa088d
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Replace the current license disclaimer in files by
a SPDX-License-Identifier.
Files that have to be modified by hand are modified.
License files are organized under LICENSES directory.
Task-number: QTBUG-67283
Change-Id: Id880c92784c40f3bbde861c0d93f58151c18b9f1
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Reviewed-by: Jörg Bornemann <joerg.bornemann@qt.io>
Turn an implicit conversion to bool in
* a call to testAnyFlag in one case;
* a comparison against 0 in another case, for consistency with the
surrounding code.
Change-Id: I1eee42ba82c59a72430bf507ea80408c553be889
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Resolve remaining Qt6 TODO
[ChangeLog][QtCore][QAbstractEventDispatcher] The signature of the abstract virtual registerTime function now takes a qint64 value for the interval parameter.
Change-Id: I10166ad5cfb455edc404d465a3731ff094a8977e
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Move away from using 0 as pointer literal.
Done using clang-tidy. This is not complete as
run-clang-tidy can't handle all of qtbase in one go.
Change-Id: I1076a21f32aac0dab078af6f175f7508145eece0
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
QObjectPrivate::threadData used to be a QThreadData *, and was
read and written from multiple threads without proper synchronization.
As an example, it was read from QCoreApplication::postEvent and
written from QObject::moveToThread, therefore causing UB.
Port threadData to a proper atomic, removing the races. Fix all usage
points.
In general, QObject is documented to be simply reentrant,
not thread-safe, and certain bits (e.g. timers, moveToThread)
are not even reentrant. The reasoning therefore is that a given
QObject's threadData is not supposed to be touched by multiple
threads without some synchronization happening elsewhere, and
therefore relaxed loads should be sufficient.
As drive-by change: refactor QCoreApplication::postEvent.
It was particularly subtle, because it had a loop using a volatile
to cope with the possibility of the receiver object switching thread
while we tried to lock its thread's event queue.
However, volatile does not achieve any synchronization, so drop it,
and refactor the algorithm using better locking primitives.
Put this algorithm in a common place, and also reuse it from
removePostedEvents, which was lacking any synchronization.
Change-Id: Icc755f7eb418ff54b33db4bdd87fd8eaf4e82c7a
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
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>
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>
Compilation and link times in CPU seconds with GCC 7, using precompiled
headers (not including moc, rcc, uic, etc. steps or headersclean):
Before After
Debug -O0 198,1 180,3
Debug -Og 240,7 229,2
Release -O3 267,1 249,2
Release LTO 239,4 229,8
QtCore required a little manual adjusting because some files are
bootstrapped into moc itself and into qmake.
Change-Id: I84e363d735b443cb9beefffd14b8b57c10e7da36
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@qt.io>
Much like the QObject::setParent() warning, which helps a bit when debugging.
Change-Id: I2abf277a12aa1ce04bd8b5759f46f8bfdcb25383
Reviewed-by: Robin Burchell <robin.burchell@viroteck.net>
Reviewed-by: Olivier Goffart (Woboq GmbH) <ogoffart@woboq.com>
This is what std::vector implementations usually do,
because it minimizes memory fragmentation and useless
allocations since no user will call clear() unless
she intends to append new data afterwards.
Fix calls to resize(0) that show how existing code
tried to work around the issue.
Adjust test. Port from QVERIFY(==) to QCOMPARE as a
drive-by.
[ChangeLog][QtCore][QVector] clear() now preserves
capacity. To shed capacity, call squeeze() or swap
with a default-constructed QVector object, see the
documentation for an example.
Change-Id: I9cebe611a97e027a89e821e64408a4741b31f1f6
Reviewed-by: Lars Knoll <lars.knoll@theqtcompany.com>
This allows us to support file descriptors >= FD_SETSIZE.
Change-Id: I7e4a35333446a587cfd13c077fa5e19fa3d1abc4
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
I wrote a script to help find the files, but I reviewed the
contributions manually to be sure I wasn't claiming copyright for search
& replace, adding Q_DECL_NOTHROW or adding "We mean it" headers.
Change-Id: I7a9e11d7b64a4cc78e24ffff142b506368fc8842
Reviewed-by: Lars Knoll <lars.knoll@theqtcompany.com>
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>
The special case was needed for INTEGRITY 5, but select support has
since long been added.
Change-Id: Ice25c5becc9d7dcdf6c5bc9eeac36b70e1b94df3
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
Reviewed-by: Louai Al-Khanji <louai.al-khanji@theqtcompany.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.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>
The platform is no longer supported or actively maintained, and is
in the way for improvements to the Unix event dispatcher and QProcess
implementations.
Change-Id: I3935488ca12e2139ea5f46068d7665a453e20526
Reviewed-by: Lars Knoll <lars.knoll@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>
Starting/stopping timers from another thread may result in errors that
may not appear until hours, days or weeks after if a release build of
Qt is used with the GLib/UNIX event dispatchers. Such errors may
manifest as warnings such as "QObject::killTimer(): Error: timer id 7
is not valid for object 0x2a51b488 (), timer has not been killed" and
application crashes (e.g. crashes in malloc, realloc and
malloc_consolidate).
Initial-patch-by: Eike Ziller <eike.ziller@digia.com>
Task-number: QTBUG-40636
Change-Id: I2de50d50eb1fc7467fcebb9c73b74d2f85137933
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This is inconsistent (e.g. the glib event dispatcher doesn't do this) and
unnecessary (already done by QThreadPrivate::finish before deleting the
event dispatcher, and by the QCoreApplication destructor).
Change-Id: Ib98147953a5e71ec87aab0ec7ef5063bb7d80134
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
postEvent() accesses it with the postEventList mutex locked, but
processEvent() was checking it without any mutex locked.
Change-Id: I31bbb50f7a1c337067b8e3de16ee7cd11400b517
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
It's read and written by different threads, so this was a race.
Change-Id: Ieffaa169eb67f40dc935291b3994f9ff1c7e05f0
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This avoids an extra division by 1000 when getting the current time.
This can't overflow, under normal circumstances, even on 32-bit: when
adding two values less than 1 billion, the result is less than 2
billion, which is less than 2^31.
Change-Id: I6f8e1aadfe2fcf6ac8da584eab4c1e61aee51cbb
Reviewed-by: David Faure (KDE) <faure@kde.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The uniprocessor taskIdCurrent global variable (declared in taskLib.h)
does not exist in VxWorks SMP, because of concurrent execution on
multiple CPUs. Any uniprocessor code that reads taskIdCurrent should
make calls to taskIdSelf() instead.
Change-Id: I4e0efef32297f339d6121c7d4bca3820e0fc9294
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
eventfd(7) uses less resources than a pipe, as it only needs to store a
single 64-bit integer, as opposed to a full buffer.
It was introduced first on Linux version 2.6.22 and glibc 2.7. However,
both the configure-time test and the runtime usage require the use of
EFD_CLOEXEC for thread-safety, so this code will be enabled only for
Linux 2.6.27 and up as well as glibc 2.9 and up.
Change-Id: Ic7e10b28d7b1d4ca24be614ed84055c4429a68e4
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Remove watchUnixSignal(), unixSignal() and associated code. These are relics
which were being used by QWS to detect virtual console switching. Currently
they are not being used at all. The recommended way to watch for Unix signals
in Qt is http://doc-snapshot.qt-project.org/5.0/unix-signals.html.
Change-Id: Id34207cb8853442302a45b2816356da0f973ebb1
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
Using a pipe for thread wake-ups is inefficient and can introduce significant
latency. Replaced the pipe by directly sending a BPS event.
Refactored the wake-up code in the private class of the UNIX event dispatcher.
Change-Id: Ic073b0b56c3cbf8327fc6bc3c37132cc3583ef86
Reviewed-by: Thomas McGuire <thomas.mcguire@kdab.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Rafael Roquetto <rafael.roquetto@kdab.com>
There are more opportunities in QtCore and the rest of Qt to make signals
private instead of public. This is a test-dart to see if there is any
reason not to do this.
It would be nice to make QObject::destroyed private, but as it has a
default argument it would be source incompatible to anyone connecting
to the SIGNAL(destroyed()) instead of SIGNAL(destroyed(QObject*)).
Currently the function-pointer-based connect syntax does not accept
a functor (or lambda) with a different number of arguments than the
signal. Olivier says a fix for that might come in 5.1, but for now
the qfiledialog2 test is changed to not use that anymore.
Also, the function pointer for a private signal can not be assigned to
a local variable, so the qmetamethod test is changed to not do so
anymore.
Change-Id: Iaf776b822f9ba364f2c184df0c6b23811da56e44
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Change copyrights and license headers from Nokia to Digia
Change-Id: If1cc974286d29fd01ec6c19dd4719a67f4c3f00e
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Sergio Ahumada <sergio.ahumada@digia.com>
It is an extension coming from the use case when you, for instance, need to
implement a countdown timer in client codes, and manually maintain a dedicated
variable for counting down with the help of yet another Timer. There might be
other use cases as well. The returned value is meant to be in milliseconds, as
the method documentation says, since it is reasonable, and consistent with the
rest (ie. the interval accessor).
The elapsed time is already being tracked inside the event dispatcher, thus the
effort is only exposing that for all platforms supported according to the
desired timer identifier, and propagating up to the QTimer public API. It is
done by using the QTimerInfoList class in the glib and unix dispatchers, and the
WinTimeInfo struct for the windows dispatcher.
It might be a good idea to to establish a QWinTimerInfo
(qtimerinfo_win{_p.h,cpp}) in the future for resembling the interface for
windows with the glib/unix management so that it would be consistent. That would
mean abstracting out a base class (~interface) for the timer info classes.
Something like that QAbstractTimerInfo.
Test: Build test only on (Arch)Linux, Windows and Mac. I have also run the unit
tests and they passed as well.
Change-Id: Ie37b3aff909313ebc92e511e27d029abb070f110
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
As in the past, to avoid rewriting various autotests that contain
line-number information, an extra blank line has been inserted at the
end of the license text to ensure that this commit does not change the
total number of lines in the license header.
Change-Id: I311e001373776812699d6efc045b5f742890c689
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
... and deprecate the old registerTimer() functions. The new pure-
virtual registerTimer() breaks source-compatibility. Subclasses cannot
be instantiated anymore, since the pure virtual function signature has
changed.
QAbstractEventDispatcher::TimerInfo is no longer a QPair. It is now a
struct with timerId, interval, and timerType members. This is a source
incompatibility that should only affect subclasses of
QAbstractEventDispatcher, which will need to pass 3 arguments to the
TimerInfo constructor instead of 2. If the subclass used QPair<int,int>
instead of the TimerInfo typedef, the QPair<int,int> declarations will
need to be replaced with TimerInfo.
Call the new registerTimer() function with the type from
QObject::startTimer(). Change all subclasses of QAbstractEventDispatcher
to reimplement the new virtual function. The type argument is unused at
the momemnt, except to ensure that registeredTimers() returns the type
each timer was registered with. Implementations for the various
dispatchers will be done in separate commits.
Author: Thiago Macieira <thiago.macieira@nokia.com>
Change-Id: Ia22697e0ab0847810c5d162ef473e0e5a17a904b
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This makes it easier to see the guts of the unix event dispatcher, and to
experiment with it.
Change-Id: I715bb68c4de6798e10bc55304a128b88e0249c63
Reviewed-by: João Abecasis <joao.abecasis@nokia.com>
When normalizing timevals, we need to bound the tv_usec member to
0-999999 inclusive, otherwise select() may return EINVAL.
When rounding timevals to the nearest millisecond in the UNIX event
dispatcher, pass the timeval through normalizeTimeval() when returning.
As discovered on Mac OS X with tst_QFutureWatcher, starting a 10 second
timer would end up calling select() with timeval = { 9l, 1000000 },
resulting in numerous "select: Invalid argument" warnings being printed
to the console.
Change-Id: Ic539e935bf847e0d4c22a73ad203e3a7a81d0690
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Marc Mutz
Dominik Holland
Lucie Gérard
Fabian Kosmale
Giuseppe D'Angelo
Allan Sandfeld Jensen
Liu Yang
Lars Knoll
Morten Johan Sørvig
Robert Loehning
Thiago Macieira
David Faure
Marc Mutz
Louai Al-Khanji
Jani Heikkinen
Rolland Dudemaine
Matti Paaso
Jonathan Liu
Frederik Gladhorn
Pasi Petäjäjärvi
Sergio Ahumada
Rafael Roquetto
Bernd Weimer
Stephen Kelly
Iikka Eklund
Lars Knoll
Laszlo Papp
Jason McDonald
Bradley T. Hughes
Robin Burchell