Before: HexRgb: 0.00230 ms per iteration, HexArgb: 0.00290 ms per iteration
After: HexRgb: 0.00051 ms per iteration, HexArgb: 0.00061 ms per iteration
This showed up as a relevant optimization when profiling KIconLoader
which uses QColor::name() as part of the key -- thanks to Mark Gaiser for
the investigation and first suggestion of a solution. I have also seen
customer code writing a replacement for QColor::name() because it was
too slow to be used as a hash key.
Change-Id: I009ccdd712ea0d869d466e2c9894e0cea58f0e68
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Cleaning out the workarounds for the discontinued "Embedded Android"
platform of Boot2Qt.
Change-Id: I0ff9d770e82a43457fb7e5da0428f4597ead4038
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@theqtcompany.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
Math functions are linked in by default on INTEGRITY.
Change-Id: I737ae87c02b2321caca3975f69525731e839d1a7
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
After this change, this was the distribution of calls in
QtGui and QtWidgets when the patch was developed for 5.4:
QtGui QtWidgets
move 23 63
copy 23 36
Change-Id: If3f536e52fc242c585e7fa0662049c0657efcc9c
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@theqtcompany.com>
Reviewed-by: Lars Knoll <lars.knoll@theqtcompany.com>
To calculate the real count we need to use the actual fixed point
increment and can not use the floating point value increment wass based
on since it might round differently.
Includes auto-test by Gabriel de Dietrich.
Task-number: QTBUG-50153
Change-Id: Ia973088f361c90370fa20bac14a4b8f373b5d234
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@theqtcompany.com>
Reviewed-by: Olivier Goffart (Woboq GmbH) <ogoffart@woboq.com>
From Qt 5.7 -> tools & applications are lisenced under GPL v3 with some
exceptions, see
http://blog.qt.io/blog/2016/01/13/new-agreement-with-the-kde-free-qt-foundation/
Updated license headers to use new GPL-EXCEPT header instead of LGPL21 one
(in those files which will be under GPL 3 with exceptions)
Change-Id: I42a473ddc97101492a60b9287d90979d9eb35ae1
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Lars Knoll <lars.knoll@theqtcompany.com>
Move some code (like registrations of meta types) from init() to
initTestCase() in the process.
Change-Id: I57db5156647cfadab554fbed853b2e68b2815f3b
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@theqtcompany.com>
Use QByteArray/QString addition instead in loops and for
test row names.
Change-Id: I31adb60bdaf7ea243143a9244b6c4f66f38b189d
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@theqtcompany.com>
Use QVERIFY2() with QTemporaryDir/File::errorString() consistently.
Attempt to catch issues like the below warning and follow-up issues.
QSYSTEM: tst_QFiledialog::clearLineEdit() QFileSystemWatcher: FindNextChangeNotification failed for "C:\Users\qt\_____aaaaaaaaaaaaaaaaaaaaaa" (Access is denied.)
Task-number: QTBUG-47370
Change-Id: I58a6e87c502627e976efa62ad73c912f3b2d49fa
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@theqtcompany.com>
The keyword no longer has a meaning for the new CI.
Change-Id: Ibcea4c7a82fb7f982cf4569fdff19f82066543d1
Reviewed-by: Simon Hausmann <simon.hausmann@theqtcompany.com>
At the edge of the view, a line segment could end up as not producing
any pixels even if not clipped by the floating-point clip
routine. Make sure the starting point for the next line is still
updated correctly for any significant segment lengths.
Change-Id: I381a4efb81ce6006f3da4c67abf279aea79e4663
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@theqtcompany.com>
In C++, type punning with a union is not allowed. It will result in
compiler defined behavior at best, and undefined behavior at worst.
Specifically, if QRgba64 is passed to a function by value, the different
members of a struct might be passed in separate registers. This means
that any write to the quint64 might not blank out the values of the
struct whenever the compiler looses track with TBAA.
Change-Id: I991b5492fe4bb13a14bb670fef5bf13dacbe6c0a
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@theqtcompany.com>
Just like from RGB32 to RGB30 we must also repremultiply when converting
from RGB64 because the alpha channel loses more precision than the other
color channels.
Since this is not approximated accurately in the simple blending
functions and the functions are no longer needed now the main render
engine supports higher accuracy, the simple blending routines for RGB30
have been removed.
Change-Id: I2b7b8eb015e330a487848fc4370ad3a1e966be91
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
- Replace Q[TRY]_VERIFY(pointer == 0) by Q[TRY]_VERIFY(!pointer).
- Replace Q[TRY]_VERIFY(smartPointer == 0) by
Q[TRY]_VERIFY(smartPointer.isNull()).
- Replace Q[TRY]_VERIFY(a == b) by Q[TRY]_COMPARE(a, b) and
add casts where necessary. The values will then be logged
should a test fail.
Change-Id: I624deb320c378c18a29b3707f48583d53bfd5186
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@theqtcompany.com>
With two alpha-bits half opacity can not be represented, so we must use
one third (85) instead of half (127).
Change-Id: I2b3f1c983a3034196bf2604840945ad3a81f5b38
Reviewed-by: Liang Qi <liang.qi@theqtcompany.com>
The tests were trying to access one pixel outside of the image.
Change-Id: Ieabdefbbdb76bb736214a0495bec72c881b27d2c
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
Could causedSSE4 instructions to be used on non SSE4 machines
in cases when qUnpremultiplywas not inlined.
This reverts commit 964ccc5853.
Change-Id: Ic676ade8f75129e8d37c4d96cbfb2bdb5b794919
Task-number: QTBUG-45741
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QMatrix and QTransform can be compared for equality,
so qHash should be overloaded, too.
[ChangeLog][QtCore][QMatrix] Added qHash(QMatrix).
[ChangeLog][QtCore][QTransform] Added qHash(QTransform).
Change-Id: I1ce925ebe258c9d7e35b68e5ac5c3373f1460c58
Reviewed-by: Olivier Goffart (Woboq GmbH) <ogoffart@woboq.com>
Merges the SSE4 specific unpremultiply with the normal version, and
adds a SSE2 fallback. There was no reason to split the two since
compile time options will ensure the right version is inlined.
Also adds short-cut for 0 and 255 values.
Change-Id: Ie5aa262f6964219fd3062d4a498f697cf79a4595
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Adds the last two missing source types to rgb64 rendering.
Conical and radial gradients. At the same time linear
gradients are moved to a template form to increase code
sharing.
Change-Id: I30fdd0837b0da03e3447683856ebbe4d7f48df6c
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
Some tests were failing because the color was not read correctly from
a QImage. To make it possibly to read more accurate colors a pixel
accessor returing QColor has been added.
Some tests also had the wrong order of arguments, confusing dest and src
formats. This has been corrected, so they test what they claim to test.
A test for RGB30 linear gradients is also added.
Change-Id: Ic623ae1b8e0bf7383056b641c6e8230a1d7dd0dd
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
This patch updates the internal color precisions of solids and
gradients to 16bit per color. This makes it possible to render
at higher precision on non-premultiplied ARGB32, the RGB30
formats and any other hi-color formats if more are added.
[ChangeLog][QtGui][Painting] Internal precision of solids and gradients
is now up to 16bit per color.
Change-Id: Ieae5468bd6de1f56adfa4cb9fa966faf2ed824fd
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
If a cosmetic polyline contained a 1-pixel segment, the next segment
would be drawn with wrong starting point.
The original fix for QTBUG-26156 had some unwanted side effects
(QTBUG-31579 and now QTBUG-42398). It tried to skip start-point update
if stroke() did not actually draw anything (because the segment was
too small). However, to determine that, it tested for a change in
lastPixel. But that was not failsafe; in some cases (1-pixel segment),
lastPixel could be unchanged even though the segment had been
drawn. With this change, we instead test directly whether stroke()
skipped the segment or not.
Task-number: QTBUG-42398
Change-Id: Id751db69a18cd1af4f45070db9d5698aa532d22a
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@theqtcompany.com>
Fixes the drawn position of end points in drawLine.
Based on a patch by Jørgen Lind, and modified so that
it caused no test regressions.
Task-number: QTBUG-38144
Change-Id: I24aa28480cc6ae09abf91d80378970565a29b254
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
This structure is meant to replace QRgb where higher precision is
needed.
Change-Id: I49d441e2133371a8b91c2e6af0c137bcc5fcb9ed
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
Use qmath and cmath methods instead of math.h methods.
Change-Id: I86ee2465c999822bf00a7cefee1642c4c30590a6
Reviewed-by: Tony Sarajärvi <tony.sarajarvi@digia.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>
Adds an SSE4.1 optimized version of qUnpremultiply and uses it in the
most drawing conversions methods. This gives a speed-up of little over
2x.
Change-Id: Ieb858a94ada1eb86d7af715ac1a100f1587f360d
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
We shouldn't create QPixmaps when comparing QBrushes that do not
contain a QPixmap.
This patch extends the comparison logic to comparing QImage cachekeys
if the brushes are QImage based.
Note the comparison still produces false negatives on equal content on
different pixmaps and images, but this is preserving existing behavior.
Task-number: QTBUG-43766
Change-Id: I001b4032172c1e568aad311f7df2eaae6aee8dc6
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
We still have a bunch of Q_WS_ ifdefs in our code, which are easy to
mistake for Q_OS_ ifdefs when quickly scanning the code. By renaming
the ifdefs we make it clear that the code in question is dead.
In incremental follow-ups, we can then selectively either remove, or
port, the pieces that are dead code.
Change-Id: Ib5ef3e9e0662d321f179f3e25122cacafff0f41f
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
If a QBrush has been created without QPixmap the texture() method will
create one. This patch avoids that in several places by checking the
type of the texture brush before accessing it, or not accessing it at
all.
Task-number: QTBUG-43766
Change-Id: If6009fe1d5bd51b239ae2c838e5c3b904b56b11a
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
tst_collections.cpp
tst_collections.cpp(3138) : warning C4305: 'argument' : truncation from 'size_t' to 'bool'
tst_collections.cpp(3190) : see reference to function template instantiation 'void testContainerTypedefs<QVector<int>>(Container)' being compiled
with[Container=QVector<int>]
(repeated)
tst_qringbuffer.cpp(297) : warning C4267: 'argument' : conversion from 'size_t' to 'int', possible loss of data
tst_qringbuffer.cpp(300) : warning C4309: '=' : truncation of constant value
tst_qringbuffer.cpp(306) : warning C4267: 'argument' : conversion from 'size_t' to 'int', possible loss of data
tst_qrawfont.cpp(947) : warning C4309: 'argument' : truncation of constant value
tst_qsslsocket_onDemandCertificates_member.cpp(217) : warning C4189: 'rootCertLoadingAllowed' : local variable is initialized but not referenced
Change-Id: I6143d4ad121088a0d5bdd6dd2637eb3641a26096
Reviewed-by: Kai Koehne <kai.koehne@theqtcompany.com>
QWidget is polished when show() is called, so unless you do this
first, QWidget::font() will return a default constructed QFont,
and not necessarily the default specified in the platform theme.
Comparing this to QPainter::font() later, which has been resolved
against the platform theme font, and not the default constructed
font, you can get a mismatch and a false test failure. This
happened on Android.
Change-Id: I41ef5b10879bbd4bb8ef8d52ecaccf5e8e894075
Reviewed-by: Gunnar Sletta <gunnar@sletta.org>
The testdata needs to be in a qrc to be available on Android.
Note that a single test will still fail. Since this requires
a platform-independent fix, it will be committed separately.
Change-Id: Ib2438a3298d81d77b7f4f240ef045aa0500d8382
Reviewed-by: BogDan Vatra <bogdan@kde.org>
The change 35bc3dc45a moved some padding out
of QTextureGlyphCache into the font engines directly, however this was not
done for the DirectWrite font engine so it caused a buffer overrun.
Task-number: QTBUG-41782
Change-Id: I4e643159036f06c5edd8a742dc6694d517a47826
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@theqtcompany.com>
