| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |\
| |
| |
| |
| |
| |
| |
| |
| | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
Fix QStaticText with OpenGL1 engine
tst_qimage: backport fix of commit 0d7e683 into 4.7
qdrawhelper: backport the optimisations in fetchTransformBilinear from master to 4.7
|
| | | |
|
| |\ \
| |/
|/|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-1:
Fix #pragma message(), the message must be enclosed in parentheses
preserve non-standard source references
QSslConfiguration: fix crash when accessing null pointer
properly parse ts contexts in obsolete messages
move the MAGIC_OBSOLETE_REFERENCE hack to the xlf file handler
Fix static build on Windows with MinGW.
Reenable the reporting of CPU features.
make qdoc3 boot-strappable again
fix MinGW cross compilation with -debug-and-release
Updated and new Slovenian translations for Qt 4.7
Designer: Fix crash on unsupported Language/Country combination.
Doc: Fixed qdoc warnings and generalized the date and time descriptions.
Doc: Whitespace fixes.
Doc: Fixed snippets in QML introduction broken in an earlier commit.
doc: Fixed numerous qdoc warnings.
|
| | |
| |
| |
| |
| |
| |
| |
| |
| | |
the PO format doc suggests that all references are in
filename:linennumber format, but some tools seem to deviate from that.
so do our best to preserve such references, even if we can't do anything
with them.
Task-number: QTBUG-9488
|
| | |
| |
| |
| |
| |
| |
| |
| | |
We were accessing the d-pointer of a QSslConfiguration which is
initialized lazily.
Reviewed-by: Markus Goetz
Task-number: QTBUG-13265
|
| | | |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
In QWidget and QGraphicsWidget, propagating fonts blank out the resolve
mask, mainly to help in determining inheritance. This means that when
the font is set on a QPainter with device and therefore resolved against
the device font, none of the attributes in the font will take presedence
and the font will essentially just be ignored. For QWidgets this has not
been a problem due to the fact that the painter is opened on the widget,
getting the propagated font as its device font. For QGraphicsWidget,
however, the QPainter is opened on the viewport and the viewport's font
will therefore override the propagated font when it's passed into
setFont(). To fix this, we need the font returned by
QGraphicsWidget::font() to contain the resolve mask of the inherited
parts of its font as well. This is done last minute to avoid messing
up the rules of inheritance in the QGraphicsWidget tree (problem arises
e.g. in hierarchy A -> B -> C and a font is set on B which should override
the font C has previously inherited from A.)
Task-number: QTBUG-13188
Reviewed-by: Lars
|
| | |
| |
| |
| |
| |
| |
| |
| | |
There are explicit tests in qdatastream for this. IMO it's wrong,
wrong, wrong, but it's currently abused in the SVG module so we can't
change it.
Reviewed-by: Gunnar
|
| |\ \
| |/
|/|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-2: (374 commits)
Add a test for the QPen::brush() != Qt::NoBrush for a Qt::NoPen pen.
Calling QPen::brush() on a Qt::NoPen pen, should return QBrush::NoBrush.
use QFile:map instead of ::mmap
Do not use global static const references to objects
Revert "Refactor blend_transformed_bilinear to simplify the blend type checking"
Don't rely on mapFromGlobal in QGraphicsScenePrivate::itemsAtPosition.
Refactor blend_transformed_bilinear to simplify the blend type checking
Note RTL behavior changes in docs and changelog
Implement qt_memfill32 with Neon.
Implement the composition mode Plus with Neon.
Fix the broken unicode detection of ODBC driver.
Doc: Fixed qdoc warnings.
Doc: updating getting started docs - not finished
Make it possible again to build Qt without webkit
Doc: fixing creator bugs, removing menus and textbox in the header
Fixing qdoc index file...Commit hack to work around the massive amounts of dependencies in the upstream branch.
Doc: Fixing overlapping text in header list
Use Ctrl rather than Alt for switching tabs in the demo browser
Doc: add link to new gettings started to index.html
Doc: Fixing bug involving header misplacement in Creator style
...
|
| | |\
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-2: (374 commits)
Add a test for the QPen::brush() != Qt::NoBrush for a Qt::NoPen pen.
Calling QPen::brush() on a Qt::NoPen pen, should return QBrush::NoBrush.
use QFile:map instead of ::mmap
Do not use global static const references to objects
Revert "Refactor blend_transformed_bilinear to simplify the blend type checking"
Don't rely on mapFromGlobal in QGraphicsScenePrivate::itemsAtPosition.
Refactor blend_transformed_bilinear to simplify the blend type checking
Note RTL behavior changes in docs and changelog
Implement qt_memfill32 with Neon.
Implement the composition mode Plus with Neon.
Fix the broken unicode detection of ODBC driver.
Doc: Fixed qdoc warnings.
Doc: updating getting started docs - not finished
Make it possible again to build Qt without webkit
Doc: fixing creator bugs, removing menus and textbox in the header
Fixing qdoc index file...Commit hack to work around the massive amounts of dependencies in the upstream branch.
Doc: Fixing overlapping text in header list
Use Ctrl rather than Alt for switching tabs in the demo browser
Doc: add link to new gettings started to index.html
Doc: Fixing bug involving header misplacement in Creator style
...
|
| | | | |
|
| |\ \ \
| |/ /
|/| |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
* '4.7' of scm.dev.nokia.troll.no:qt/qt-qml:
Update example screenshot in doc.
Handle layoutChanged() properly in QML views.
Fix ListView and GridView tests following off-by-one fixes.
Revert 4bc81bb1cb3cd4a0a3fe071e00556124e770d7ac
Fix off-by-one at end when views scroll to keep currentItem visible.
Make declarative examples a bit more qt-like.
Add quit buttons to declarative demos
ListView.view and GridView.view properties should not be writable.
Avoid Flickable view jumping when drag threashold is exceeded.
Fix ListView.view attached property with VisualItemModel
Remove dead code.
Add test for VisualItemModel layoutChanged() handling
Remove unused code.
Handle QAbstractItemModel layoutChanged() signal
|
| | |\ \ |
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Update all visible items when layoutChanged() is emitted.
Task-number: QTBUG-13146
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Our tests were verifying incorrect positions...
Task-number: QTBUG-12822
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
We assumed that the attached object was created at the same time as
the item.
Task-number: QTBUG-13166
Reviewed-by: Aaron Kennedy
|
| | | |/
| |/|
| | |
| | | |
Task-number: QTBUG-13146
|
| |\ \ \
| |_|/
|/| |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-1: (61 commits)
Autotest: don't use Q_FUNC_INFO for testing which method got called
fix generated makefile dependencies
Cocoa: Demo browser can get stuck after closing modal dialog
Restore default if to system default on session close.
tst_qmake doesn't need QtGui
Use the full path to qmake in the qmake unit test
qdoc: Fixed erroneous links to QML basic types.
Fixed item view background color in Gtk style
scope fixes and clutter reduction for sql driver projects
I don't know why some linkers can't call this function, so comment it out.
QNetworkSession::close() method now send closed() signal while faking disconnection.
Add the missing license headers to the QString benchmark data
Fix building of qsimd.cpp on Windows CE
Use QElapsedTimer for the benchlib tests.
Properly implement the CPU feature disabling in qsimd.cpp.
Report the detected CPU features in the corelib boilerplate
Detect CPU features on ARM by reading the ELF auxvec.
Split the CPU-detection code into multiple functions for readability
Fixed delivering gestures to a toplevel widget.
Unroll the SSSE3 code even more to avoid the need to keep an extra variable for inverting the result
...
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | | |
The string from Q_FUNC_INFO varies according to the compiler, so instead
use an enum.
Reviewed-by: Carlos Duclos
|
| | | |
| | |
| | |
| | | |
Reviewed-By: ossi
|
| | | |
| | |
| | |
| | | |
Reviewed-By: ossi
|
| | |\ \
| | |/
| |/|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-1: (55 commits)
qdoc: Fixed erroneous links to QML basic types.
Fixed item view background color in Gtk style
scope fixes and clutter reduction for sql driver projects
I don't know why some linkers can't call this function, so comment it out.
QNetworkSession::close() method now send closed() signal while faking disconnection.
Add the missing license headers to the QString benchmark data
Fix building of qsimd.cpp on Windows CE
Use QElapsedTimer for the benchlib tests.
Properly implement the CPU feature disabling in qsimd.cpp.
Report the detected CPU features in the corelib boilerplate
Detect CPU features on ARM by reading the ELF auxvec.
Split the CPU-detection code into multiple functions for readability
Fixed delivering gestures to a toplevel widget.
Unroll the SSSE3 code even more to avoid the need to keep an extra variable for inverting the result
Don't try to compile the SSE2 and SSSE3 code with compilers that don't support them (e.g. ARM)
Improve on the SSSE3 with alternate aligning function.
Add the beginnings of a new SSSE3-based aligning algorithm
Small fixup
Update the SSSE3-with-alignment function to use aligned loads.
Add an ucstrncmp that uses SSSE3 with aligning.
...
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Autotests were modified to check for 50% color coverage as well.
Before they were not really testing the bgcolor since they would
accept only the red grid color as a valid test pass.
Task-number: QTBUG-13125
Reviewed-by: ogoffart
|
| | | |\
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-1: (50 commits)
Add the missing license headers to the QString benchmark data
Fix building of qsimd.cpp on Windows CE
Use QElapsedTimer for the benchlib tests.
Properly implement the CPU feature disabling in qsimd.cpp.
Report the detected CPU features in the corelib boilerplate
Detect CPU features on ARM by reading the ELF auxvec.
Split the CPU-detection code into multiple functions for readability
Fixed delivering gestures to a toplevel widget.
Unroll the SSSE3 code even more to avoid the need to keep an extra variable for inverting the result
Don't try to compile the SSE2 and SSSE3 code with compilers that don't support them (e.g. ARM)
Improve on the SSSE3 with alternate aligning function.
Add the beginnings of a new SSSE3-based aligning algorithm
Small fixup
Update the SSSE3-with-alignment function to use aligned loads.
Add an ucstrncmp that uses SSSE3 with aligning.
Add an SSSE3 version of ucstrncmp
Optimise the tail comparison of ucstrncmp
Add a version of ucstrncmp with SSE2 with aligning.
Add an SSE2-optimised version of ucstrncmp
Add the ucstrncmp benchmarks
...
|
| | | | | |
|
| | | | |\
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-1: (48 commits)
Use QElapsedTimer for the benchlib tests.
Properly implement the CPU feature disabling in qsimd.cpp.
Report the detected CPU features in the corelib boilerplate
Detect CPU features on ARM by reading the ELF auxvec.
Split the CPU-detection code into multiple functions for readability
Fixed delivering gestures to a toplevel widget.
Unroll the SSSE3 code even more to avoid the need to keep an extra variable for inverting the result
Don't try to compile the SSE2 and SSSE3 code with compilers that don't support them (e.g. ARM)
Improve on the SSSE3 with alternate aligning function.
Add the beginnings of a new SSSE3-based aligning algorithm
Small fixup
Update the SSSE3-with-alignment function to use aligned loads.
Add an ucstrncmp that uses SSSE3 with aligning.
Add an SSSE3 version of ucstrncmp
Optimise the tail comparison of ucstrncmp
Add a version of ucstrncmp with SSE2 with aligning.
Add an SSE2-optimised version of ucstrncmp
Add the ucstrncmp benchmarks
Update the data generation script to use a non-including .cpp
Update comments in QString about alignment performance.
...
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
If there is only one widget which is a toplevel, then gestures were not
delivered to it because we assumed (wrong) that there is at least one child
widget.
Reviewed-by: Frederik Gladhorn
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
for inverting the result
On 32-bit, we're out of registers already, so this variable ended up in memory
|
| | | | | |
| | | | |
| | | | |
| | | | | |
support them (e.g. ARM)
|
| | | | | |
| | | | |
| | | | |
| | | | | |
Even though this function *only* does aligned loads, it's worse than the other function
|
| | | | | | |
|
| | | | | | |
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
This results in no change on the Core-i7, but another 2.6% on the Atom
(so it's now 8% better than 4-byte loads and 31% better than current
code)
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
The results on i7 are 32% improvement over current code, 13%
improvement over 4-byte loads, 6% over the unaligned SSSE3
loads. However, it's about 2.5% slower than pure SSE2 code due to
complexity.
The results on Atom are 30% improvement over current code, 7% over
4-byte loads, 15% over pure unaligned SSE2 and 9% over unaligned
SSSE3.
|
| | | | | | |
|
| | | | | | |
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
This is a different technique of aligning. Instead of reading some
bytes before the string, we will read some bytes of the string twice.
Best results are only 2% improvement over the unaligned SSE2 on a
Core-i7.
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
First results make it 34% faster than current ucstrncmp, 16% faster than
the 32-bit version.
|
| | | | | | |
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
This speeds up the writing of algorithms, so I don't have to recompile
megabytes of data dumps.
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Also use the real-world data that I collected. The resulting files are
quite large, so I've added to Git only the smallest dump (apparently
Teambuilder scanning the environment for "TEAMBUILDER=").
|
| | | | | | |
|
| | | | | | |
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
It's currently slightly worse than SSE2 with prolog aligning
(i.e., it's no good)
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Instead of using a non-SIMD method for aligning, we instead load more bytes
from p1 and use the PALIGNR instruction to realign to what we want.
The result is that it's bit slower than the non-SIMD comparison, due to the
complexity. For strings over 8 QChars wide, it's only slightly worse than
the non-SIMD comparison.
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
The prolog tries to align p1 to a multiple of 16, so as to run
aligned loads, which are faster. Unfortunately, my tests so far
indicate that the prolog ends up taking longer than the benefit
of having aligned loads.
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
This function uses unaligned loads. I'll try to write one with
aligned loads later.
|
| | | | | |
| | | | |
| | | | |
| | | | | |
This is a copy of qMemEquals from qstring.cpp
|
| | | | | | |
|
| | | | | |\
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-1:
fix build after first webkit self-injection attempt
Fixed touch event delivery in QGraphicsView.
let WebKit inject itself into the qt configuration
add indirect input/output specification capability to QMAKE_SUBSTITUTES
Add test that exercises lupdate warnings for QtScript
Make qsTrId() / QT_TRID_NOOP() accessible from QtScript
Make lupdate's QtScript frontend recognize qsTrId() / QT_TRID_NOOP()
Add support for comments and meta-data in the lupdate QtScript frontend
Streamline lupdate's QtScript frontend's error messaging
Doc: linking up orphant files
Doc: Updating menu links
|
