Up until now, the value stored in t2 depended on whether we had found
a monotonic clock or not. Fix that by always storing nanoseconds: we
avoid extra calculations and accessing a global variable all the time.
The impact is contained to the actual getting of the time. And we
mitigate by using clock_gettime to get the realtime clock instead of
gettimeofday, if that's available.
Change-Id: Iceef2d050fd9472f1a66e11e7ded79fe5163a132
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
This test is doing a brute-force mapping. Coupled with changes to
QCOMPARE last year, this now allocates and deallocates a lot of memory
per iteration. On my Sandybridge, it takes two minutes to run:
111136.781153 task-clock # 0.999 CPUs utilized
371,692,633,238 cycles # 3.344 GHz
182,641,818,708 stalled-cycles-frontend # 49.14% frontend cycles idle
57,951,552,830 stalled-cycles-backend # 15.59% backend cycles idle
477,216,332,971 instructions # 1.28 insns per cycle
# 0.38 stalled cycles per insn
86,959,637,669 branches # 782.456 M/sec
309,185,237 branch-misses # 0.36% of all branches
111.264868818 seconds time elapsed
Changing the iteration step from 1 to 5 reduces the runtime to about 5
seconds.
Change-Id: I9cad6f85f535f472319da7cd6c4aa28e12ddf1b7
Reviewed-by: Jason McDonald <jason.mcdonald@nokia.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
These projects are expected to fail, but we need to make sure they fail
for the right reason.
Change-Id: I8a7caaa663060712c5c7113ef3b054feba2e2287
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
In the CI system, an environment variable is used to convey
the CMAKE_PREFIX_PATH, but that can not be relied upon.
Change-Id: Ie4fbacaac6ae18f95a3b4d1e796a4b4c91a418c4
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
This test has recently timed out in CI, but appeared to be making
progress. Give it more time to complete.
Change-Id: Ied0fb7aad35ed6d5889dd585a7545687617e5e19
Reviewed-by: Kalle Lehtonen <kalle.ju.lehtonen@nokia.com>
GCC 4.8 warns:
warning: typedef ‘PluginKeyMapConstIterator’ locally defined but not used [-Wunused-local-typedefs]
Change-Id: I0225708b590ac3d72ab3cb8f61e686091c187053
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
They're identical to the ones the compiler will
synthesize and their presence prevents move
constructor and assignment operators from being
synthesized alongside their copy counterparts.
The destructor can go, too.
Change-Id: Ia9cc26ba1ffaa670f03642af457e632f44bd69c2
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
a module's project file may set MODULE_INSTALL_LIBS before loading
qt_module.prf to have an alternative RPATH linked into the users of that
module.
this is relevant only for linking against non-installed -prefix builds
of that module, as otherwise .libs from the module's pri file is used
for rpath.
Change-Id: Ib240e748cf130a71a5991dc643c368a983092ead
Reviewed-by: Simon Hausmann <simon.hausmann@nokia.com>
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
forwarding module pris get rpath_link{,_private} fields, which are
used accordingly by qtAddModule().
Change-Id: I0abc2dc8b1e8744dbf7f439aa7fed9ae159c2c74
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
QMAKE_LIBDIR and QMAKE_FRAMEWORKPATH have the downside that they
always effectively end up in LIBS, which makes for weird prl files.
Change-Id: Iaf61b0038504ff91ae5ec7f9b1255fe3a2d134f1
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
this saves some repeated calculations. also, it's nicer to have most
logic in one place.
Change-Id: Iea362d40f5e6203709ced94d29ca61a4163b8e69
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
the variable name is a function parameter now
Change-Id: I46f3ce37c157312e7f70291a0cef2d666d50664c
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
When incredibuild-xge had been turned on by configure then it was not
ensuring that the generated vcproj files had the right extra comments
needed to make incredibuild run moc and other tools in parallel. By
fixing the prf file to get around a problem in the parser this is now
respected.
Task-number: QTBUG-14482
Change-Id: Ifd24bde2db7d80d3a52719c733d15f228c7b3534
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
Incidentally, this patch shows that Qt isn't consistent in how it
treats empty environment variables used as flags, but that is something
for a separate commit. This patch aims to be behaviour-preserving.
Change-Id: I5390cda314858762b76384291373a6f0167d323a
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Task-number: QTBUG-26793
Change-Id: Ic19cb6581cd5838d26713998e152772a5d12da4f
Reviewed-by: David Faure <faure@kde.org>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Reverting change after request in the developers mailing list.
This reverts commit 50a5bd5429e0c8b99be81b86d3730737cb7f8514.
Change-Id: Ic4a420d6ad0995810ed61d31edd28e7b603cca5e
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@nokia.com>
QItemDelegate takes the USER property as the value to be set using
model->setData(). The text is what is expected.
follow-up to b1b87a73012342dc1619a8e907ea9954d59ca564 which in Qt 4.8.0
broke the interaction between QComboBox and QItemDelegate used by
QDataWidgetMapper and item views.
Task-number: QTBUG-26501
Change-Id: I005a0a6af5265ee22baf417957e520992d16d6e1
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
The newer CMake version has the POSITION_INDEPENDENT_CODE property which is
what we need here. The CMake 2.8.8 implementation uses awkward and incomplete
string manipulation which I don't want to maintain for any amount of time
when Qt 5.0 is released.
Change-Id: If7ace9c6925ccdbf800f1863fa2368e55fa44d7f
Reviewed-by: Brad King <brad.king@kitware.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
-fixed generated HTML file names
-removed qt-network group in qt/qtdoc and removed \ingroup instances
Change-Id: Ida5ded5da30e6db1978382f6742a5da8036e055e
Reviewed-by: Shane Kearns <shane.kearns@accenture.com>
Reviewed-by: Martin Smith <martin.smith@nokia.com>
Reviewed-by: Laszlo Papp <lpapp@kde.org>
The class used to be called QTimeStamp before.
Change-Id: I6e543f4f131a80299eb5dae78d8852b7899bab8a
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
This code predated the Q_LIKELY macro. Now that we have it, use the
official macro.
Change-Id: I0e43609f8d6827f8fef7b928655a965cba579b8d
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Documentation incorrectly states that it is m13 that is set to 1,
when it is instead m33.
Task-number: QTBUG-10232
Change-Id: If3c7d3eb1c478ebef4f13b4842235445afc48049
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Finder on Mac OS X sends a command line argument of
the form -psn_0_3523420 when launching applications.
Eat the argument like we do in Qt 4.
Change-Id: I09db9dea035e62c0b816a833c9ad6fcd78f83735
Reviewed-by: Teemu Katajisto <teemu.katajisto@digia.com>
Reviewed-by: Simo Fält <simo.falt@nokia.com>
Reviewed-by: Marius Storm-Olsen <marius.storm-olsen@nokia.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
This does the job of keeping the terminal keyboard working when
the application segfaults but is somewhat unsafe because ioctl()
is not async-signal safe.
Change-Id: Ie6728a4252e18c29bba8f8308e6c00d4a1eb6a8e
Reviewed-by: Elvis Lee <kwangwoong.lee@lge.com>
Reviewed-by: Jørgen Lind <jorgen.lind@nokia.com>
These functions are a faster version of {,!}qgetenv().is{Null,Empty}(),
a common pattern in Qt code.
Their main advantage is that they don't need to allocate memory, so
they can be used in noexcept functions, or dynamic initialisation of
namespace-scope statics, because throwing in these contexts invokes
std::terminate().
Change-Id: I651c5bd72f450b5d7df76590f8791572fe992af5
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Using the new getter to access the "test mode" setting.
Change-Id: Id26a350cd3fab4bf2e5f58ba67bc7323f99c9cc3
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Library code might need to know this, e.g. when calling an external
process, to give it the right configuration. (For instance when
ksycoca code calls kbuildsycoca to recreate the DB at the right place).
Change-Id: I343ddefff816586f9d391973c08ff1e1ad86bf0e
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
SRCDIR was not defined for WinCE but it should no longer be used.
Fixed test case to use QFINDTESTDATA instead.
Change-Id: I07cbf7d42790d33e2d205d1682ec10e7577a92bd
Reviewed-by: Miikka Heikkinen <miikka.heikkinen@digia.com>
Reviewed-by: Andreas Holzammer <andreas.holzammer@kdab.com>
QWindow::setParent() sets the parent to zero instead of the desired
parent, if platform window has not yet been created for the parent.
This caused QWindowsWindow::setParent() to skip setting the parent
later, when correct window was specified, as the QWindow parent-child
relationship hadn't changed. Fixed by changing the the check to use
native handles instead.
Task-number: QTBUG-26791
Change-Id: I292a1ddf746583a7268f2d07c20166995c0dd7d6
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
Followup to I5c925ae3e191244c7ab9415e4ba2fe49b93dd2af : touch was not
actually working by default. A major symptom was getting double-clicks
instead of clicks on buttons and such, making it quite hard to
interact with anything. There are some problems remaining.
Change-Id: I09fa5965036bac242ccc1f4db71f6d912445ca01
Reviewed-by: Laszlo Agocs <laszlo.p.agocs@nokia.com>
All functions in QElapsedTimer are marked Q_DECL_NOTHROW. This code
is often introduced in many places to deal with timeouts and doesn't
need exception handlers. In particular, it's used in QMutex locking.
In addition, mark QDateTime::currentMSecsSinceEpoch as nothrow, as it
can't throw exceptions either and it is needed by the generic
QElapsedTimer.
Q{Date,Time}::current{Date,Time} operate on local time and run into at
least one cancellation point, which we must consider throwing. And
returning a QDateTime allocates memory.
Change-Id: Id776c5ec831fc06d7419a9ff5442d9b35cff1a22
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
For the same reason as the previous commit (about the atomic classes),
mark these functions as never throwing an exception.
Change-Id: Idf46e41b226f54cb8658416efdf985ca40dd2952
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Actually, only the "new" atomics are marked. The old implementation,
based on qoldbasicatomic.h is unchanged, but should still work without a
problem.
The following configurations were tested and do work:
- x86 64-bit
- x86 32-bit
- generic GCC
- generic C++11 std::atomic
- bootstrap
- ARMv6 and 7
- MIPS
- MSVC 2010 32-bit
- MSVC 2010 64-bit
The only two configurations untested are IA-64 and ARMv5. Except for
MSVC, all configurations were tested with GCC 4.6 (MIPS and ARM) and 4.7
(x86 and generics).
Change-Id: Iecbfeacd9d20b535453e91335165e9a221e0b47e
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
These functions are inline (on most architectures) but they contain
inline assembly which the compiler could not understand. In any case,
if it decides not to inline them, it would need to generate exception
handlers.
Change-Id: If9d50793d715c51781e76d0a539da03a6d83e255
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
These functions are not supposed to return, not even by exception.
qt_message() _can_ throw, but we're fine with the compiler calling
std::terminate() then, since the backtrace will still include the
assertion location.
This behaviour is ensured by a new macro, QT_TERMINATE_ON_EXCEPTION,
which expands to something like
try { expr; } catch(...) { std::terminate(); }
if the compiler doesn't support Q_DECL_NOEXCEPT (but maybe
Q_DECL_NOTHROW), and to something like just
expr;
otherwise (including in the QT_NO_EXCEPTION case).
The real macro preserves scopes in all cases, and aims
to work even if <exception> isn't included in the TU it's used in,
so is a little bit more complex than that.
Change-Id: Ie6a2b7776e6aa77e57bd9aea6e184e5fa1cec81c
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The macros have been moved to their respective modules.
Change-Id: I653668b608cd3b79824a25b0e7b1c238330c0007
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
The convention for QMap is that "free" methods simply free memory and
that "destroy" ones also run destructors of types affected. Since this
function calls destroySubTree, it should be more properly called
"destroy".
This also gets rid of another function called simply "free" in our
headers.
Change-Id: I3fa621c78848ede33dd5b78428c704d0423ba4e7
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
QLinkedList does use allocations in the .h, by way of operator new and
operator delete. But for consistency with other classes, don't call our
function free().
Change-Id: I9295f5945619ad4ea82fd1f7f55b34700c54e06a
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
The important change is to use functions not called malloc and
free. On some old systems or under some weird circumstances, they
might be #defined to something different. More importantly, this
allows us to quickly verify whether we're using the stdlib.h functions
in our public headers.
Change-Id: I66b963ac2bc75d64028c04542e62a929e4b36b19
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
The memory is allocated in qlist.cpp, so it should be freed in
qlist.cpp. Freeing it in qlist.cpp ties our hands about future
improvements to the allocator.
In addition, silence the warning by the too-smart-for-its-own-good GCC
that we're trying to free a non-heap object:
qlist.h:763:14: warning: attempt to free a non-heap object "QListData::shared_null" [-Wfree-nonheap-object]
The warning is wrong. It should say "possibly" somewhere because GCC
failed to account for all conditions in the path to free().
Change-Id: I34a6c16bba9a2197fc83eb3c7a63ae06fb25bf15
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Hard float enabled Raspbian image is now primary platform provided by the
Raspberry Pi foundation.
Change-Id: I2d59ad26fac459907b4f51764975a3e6e29c6033
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Reviewed-by: Thomas Senyk <thomas.senyk@pelagicore.com>
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
