There are only 8 MMX registers, each 64-bit wide, and they alias the
x87 registers. The access to the MMX register cannot use the new
VEX-prefix instructions either.
All of the functions being replaced are either present in the
qdrawhelper_sse2.cpp and qdrawhelper_ssse3.cpp files, or the plain C++
function in qdrawhelper.cpp is vectorised when compiled with
-ftree-vectorize (enabled in -O3), if SSE2 support is enabled.
All x86-64 processors have SSE2, so this is a net improvement for
64-bit builds. For 32-bit builds, without further support this will
cause the code to use non-vector or x87 instructions, which aren't the
best. The solution will come in another commit.
Change-Id: I4a22d8a2516b79172867510202d0fd627db54807
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
WA_PaintOutsidePaintEvent is only suggested to be used when porting Qt3 code
to Qt 4 under X11 platform. and it has been broken now.
Change-Id: Ie4297b2a449f1055ca10ada9efb930e6018b1efb
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
The default type when Qt 3 support is enabled. Same as
AutoConnection but will also cause warnings to be output in
certain situations.
Change-Id: I64bf3c39a740afb716820bfd3173936fda213f4a
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
QStandardPaths now knows a "test mode" which changes writable locations
to point to test directories, in order to prevent auto tests from reading from
or writing to the current user's configuration.
This affects the locations into which test programs might write files:
GenericDataLocation, DataLocation, ConfigLocation,
GenericCacheLocation, CacheLocation.
Other locations are not affected.
Change-Id: I29606c2e74714360edd871a8c387a5c1ef7d1f54
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Jason McDonald <jason.mcdonald@nokia.com>
This adds checks to ensure Q_ALIGNOF is returning the desired alignment
for explicitly-aligned types.
The alignment check is now inlined in the test inside QCOMPARE so we get
slightly more informative errors:
FAIL! : tst_Collections::alignment() Compared values are not the same
Actual (quintptr(&it.value()) % Value::PreferredAlignment): 64
Expected (quintptr(0)): 0
Loc: [tst_collections.cpp(3384)]
In this case, this is enough to notice "non-native" alignments are being
requested. Having test parameters otherwise hidden in template arguments
doesn't help the situation.
Change-Id: I05267fd25b71f183cfb98fb5b0a7dfd6c28da816
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
The intent is to force instantiation of template container classes and
semantics or behaviour are otherwise irrelevant in this context.
tst_collections.cpp:3036:15: warning: inequality comparison result unused
tst_collections.cpp:3037:15: warning: equality comparison result unused
tst_collections.cpp:3100:15: warning: inequality comparison result unused
tst_collections.cpp:3101:15: warning: equality comparison result unused
Change-Id: I70ad38b18dcbc43879e36a34b1da460aee5f7b07
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Instead use QMap if we want a stable order.
Task-number: QTBUG-24995
Change-Id: I93f643df236f5078768f539615fa47163e5262e8
Reviewed-by: Giuseppe D'Angelo <dangelog@gmail.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Test that they do expand properly and don't produce errors. This is
templated code, so it doesn't get tested fully unless we instantiate
them.
Also check that the alignments are correct.
Change-Id: I2a8ee2165167f54b652b4227411e209850974b8e
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
QComboBox does in fact have a user property
since b1b87a73012342dc1619a8e907ea9954d59ca564.
Change-Id: I24eb2ef267cec5d8a9f7348954b703fa6df04fa5
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
Reviewed-by: Thorbjørn Lund Martsum <tmartsum@gmail.com>
Reviewed-by: David Faure <faure@kde.org>
Both classes had such components before, but there were issues with
the NOTIFY signal not being in the same class as the Q_PROPERTY.
This patch solves that problem by using a signal of a different name.
Task-number: QTBUG-15731
Change-Id: Ibc7ce4dba8a6b88c05d62a90e14d0101c5cd3082
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Thorbjørn Lund Martsum <tmartsum@gmail.com>
It's almost exactly the same code in both files, so let's have one file
only. That means we need an #ifdef for the special case of 64-bit types
on i386.
Also take the opportunity to add a comment explaining how this works.
Change-Id: I50d274fa026806ae511b1045aa8a5c25daaa0edc
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
This is the first step in merging the i386 and x86-64
architectures. The next commit will bring i386 support into
qatomic_x86.h.
Change-Id: I24105ea70f3fc29b3fb779a70053f99117440573
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
We don't want to be using or trusting partial loads.
Change-Id: I3934d6cf54cd99eaab2fa7aee9a0e9968d9f3c13
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
These avoids repeating code and documents that the underlying format is
compacted big-endian.
Change-Id: I5a2dc0084945d99368183203a0a9b7c116874620
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
The mode argument (third argument to (_)open) is only used when
(_)O_CREAT is also specified. As we're opening the file read-only it is
unused and unnecessary.
Change-Id: Icc16edec5a7d44c57ad02865048c56114c39d4bc
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
elfHash and friends are used solely from do_translate, which already
checks for null strings. There's no need to do it again here.
Change-Id: I90a16d2623ca753a444e53952539001988568bdb
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
There isn't really a need for the dependency as LanguageChange events can be
caught in QObject::eventFilter, directly.
Change-Id: I39778fbe1663924d97705b514ae399cfd3749776
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
The warning is presented now that we use __assume in every
Q_ASSERT. It presents itself in code that calls functions, like:
tools/qbitarray.h(128): warning #2261: __assume expression with side effects discarded
{ Q_ASSERT(i >= 0 && i < size());
^
Change-Id: I5851b3c35b56c00b9c541c2489233121ef621e04
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
For some reference platforms and SDKs we will need to pass in
extra paths. Currently users have to modify the mkspec to adjust
paths or set environment variables that will be picked up.
This change introduces the -device <name> and -device-option
<key=value> option. The key value pairs will be written to a
qdevice.pri and can be used by the qmake.conf of the device spec.
The reason to not save the key value pairs in qconfig.pri is
becase of the fact that the device spec loads the qdevice.pri
earlier than the qconfig.pri. qdevice.pri allows the mkspec
to set the compiler flags and qconfig.pri allows configure to
add to those compiler flags.
Done-with: Holger Freyther
Change-Id: I931a197b8be72397e1eedfee09502eefc01c9d4f
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
Reviewed-by: Johannes Zellner <johannes.zellner@nokia.com>
Reviewed-by: Donald Carr <donald.carr@nokia.com>
This autotest gives different results on consecutive runs, and is
therefore insignificant for the purpose of detecting regressions.
Task-number: QTBUG-24977
Change-Id: I5c4dfd663ce5df6b60ae47a29d332c06e3c0585f
Reviewed-by: Kalle Lehtonen <kalle.ju.lehtonen@nokia.com>
This commit addresses a long-standing Qt 5 to-do. Whilst a trivial
change, it is binary incompatible.
Task-number: QTBUG-665
Change-Id: I4294233d876dec79eda57113bdf298ce73643e76
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
This commit addresses a long-standing Qt 5 to-do comment. Whilst a
trivial change, it is binary incompatible.
Task-number: QTBUG-259
Change-Id: I2fc7bfda488318dbabbbea9f5ff9d2b1d6ce0784
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
This commit addresses a long-standing Qt 5 to-do. Whilst a trivial
change, it is binary incompatible.
Task-number: QTBUG-1433
Change-Id: I6e31e47fd5791cb6f1373e2696ffc95f7174f0b0
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
This silences warnings from valgrind about a conditional depending on an
uninitialized value.
Change-Id: I819a44ed5dc02e163c00849811870c94f66b6651
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
According to the Intel Optimization Manual section 3.5.1.1 Use of INC
and DEC Instructions, those instructions modify only part of the flags
register, so they mey introduce unnecessary data dependencies on
previous flag-setting operations so that the resulting flags are
computed. Preferring ADD and SUB (rule 33) is recommended.
However, we don't do it for 16-bit integers. The reason is that the
presence of the 0x66 prefix may trigger a slower decoding codepath in
the processor (up to 6 cycles, as opposed to 1). The same Intel manual
talks about Length-Changing Prefix, which applies in particular to
instructions with 16-bit immediates. The assembler generally produces
uses the 8-bit immediate variant of the ADD and SUB instructions, but
to be on the safe side, we prefer to use INC and DEC here.
Change-Id: Ic03236ac600a5b4e087614d21df5d3c666ae064e
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
The assembly output showed that GCC was generating some wrong code in
some conditions, so update the constraints so it will do the right
thing: the expectedValue constraint needs to be in/out with early
clobber. In/out because cmpxchg8b really does produce output and, even
if we don't care about it, GCC needs to be told that the registers
used (EAX:EDX) were modified. The early clobber is necessary so it
won't schedule EAX or EDX to be the same as the EBX_reg (the register
we'll xchg EBX with).
Since EAX and EDX are in/out and EBX can't be used, the only remaining
low register for the "sete" instruction is CL. So use it directly and
set ECX to be in/out too.
For whatever reason, it can't find enough registers in debug mode and
this expansion doesn't work. It looks like a bug though, since this
requires 4 registers and one memory operand and in debug mode it must
have EAX, ECX, EDX, ESI and EDI free for use. One of ESI or EDI is
used to xchg EBX with, which means there must be at least one more
free general register.
Change-Id: I1f11e68d776bf9ad216b34ca316a53129122fabe
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
That compiler is no longer supported
Change-Id: I6ff9003a8b986478850ad2e6e6662a44264236d7
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
If we deprecate the entire class, then we get the warning for every use
of it with GCC, including the three operators defined inside the class.
So instead, mark as only the constructors as deprecated, which should
reduce the warnings to the places where the class was instantiated.
That should be enough to poison the code enough: to fix the warning, you
need to stop using the class there and/or stop calling one of the
deprecated functions that deal with it.
Unfortunately, we still get one more warning inside the inline code, in
QImage::textList. To remove that warning, we disable all deprecation
warnings around that code block.
Change-Id: If3c11647a9d2cab055e15810d7cfe0d8f5e7042f
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Unlike QNX mkspecs, these mkspecs target more specific, BlackBerry-only
devices. So far, the only difference is the definition of Q_OS_BLACKBERRY
and linking with -lbps, but this will likely change with time. It will also
allow conditional compilation to distinguish between qnx and blackberry
devices.
Change-Id: Ia44e9b4059f720c308240e9dac7e90f7ae7920d0
Reviewed-by: Sean Harmer <sh@theharmers.co.uk>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
The static qt_mac_is_sheet() function in qcocoafiledialoghelper.mm was
not called anywhere, remove it.
Change-Id: I88785e15aa17ae9c9ffbc33eba30433c8834b798
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
Silence warnings about unused function parameters and local variables
while building the Cocoa platform plugin.
Change-Id: I6aedc4cb21c5fb48d2d6e501561473d3f7112aed
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
Remove unused (and misspelled) flipedRect() function, as well as several
unused variables.
Change-Id: I48a5bf2eaad67686f60523a5c22262cad5314128
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
clearNSWindow() should also clear the delegate and remove the
contentView from the window, since we are no longer using that window.
Make sure the QCocoaWindow::~QCocoaWindow() doesn't release the
m_contentView until after clearNSWindow(), to avoid crashes while
trying to cleanup the window and view.
Change-Id: Ia081488f629a4fd4cf10fb1053fb8183b1914d35
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
Calling raise() on a hidden window should not show it. The setVisible()
function will ensure that the window is raised (since we use orderFront
or makeKeyAndOrderFront). This fixes the failing
tst_QDockWidget::task169808_setFloating() test as well.
Task-number: QTBUG-24774
Change-Id: If34472ebbcd615c10654efafd54c84c03d10bc8c
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
THe UK still uses the Imperial system at least for distances
and many other things.
Change-Id: I99379de35620114328ad6a7fc9b226a46692bedd
Reviewed-by: Denis Dzyubenko <denis.dzyubenko@nokia.com>
Replaced hard coding as QT_GUI_PASSWORD_ECHO_DELAY with
a style hint.
Change-Id: I0b78ebad723dbe19d9b9496583203e31545874e2
Reviewed-by: Andrew den Exter <andrew.den-exter@nokia.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Apparently there are still issues here and there (e.g. in declarative)
with properly recognizing more complex event sequences (like triple
clicks). The behavior of qtbase is correct but we need a test case
that makes sure the functionality will not regress.
Change-Id: I08c558fcfdde0dd06e194b4f0affc6f6896573bf
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Qt and Cocoa opacity levels are compatible, so we just need to forward
the setOpacity() argument to [NSWindow setAlphaValue]
Change-Id: I5fd5678894fd6949811ad1c4b493e592561f01f6
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
Start from 2.9, fontconfig will reset the result to FcResultNoMatch at
the beginning of FcFontSort().
According to
http://lists.freedesktop.org/archives/fontconfig/2012-March/003857.html
the result value of FcFontSort() can be ignored, checking the nfont
value of the fontset returned is sufficient.
The fix works for pre-2.9 versions as well, since those versions don't
touch the result at all.
Change-Id: Iba6c1157e314088a90867292a4bd970bb873e284
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@nokia.com>
This implementation was disabled and didn't really work
properly, so it doesn't make sense to update or do
bug fixes to it. Removing it completely instead.
Change-Id: I62abe35f0f9a91bb207b87296d25c3c62d58228e
Reviewed-by: Jiang Jiang <jiang.jiang@nokia.com>
