summaryrefslogtreecommitdiffstats
path: root/src/gui/painting/qdrawhelper.cpp
Commit message (Collapse)AuthorAgeFilesLines
* Updated year in copyright headerKai Koehne2014-03-261-1/+1
| | | | | | | | | | | | | | | | | | find . -path '*/3rdparty/*' -prune -o -type f -print | xargs -L1 sed -i -E 's/Copyright(.*) 2013 Digia/Copyright\1 2014 Digia/g' Manually patched files: demos/spectrum/3rdparty/fftreal/fftreal_wrapper.h demos/spectrum/3rdparty/fftreal/fftreal_wrapper.cpp src/3rdparty/s60/eiksoftkeyimage.h tools/qdoc3/test/qt-project.qdocconf tests/auto/qsharedpointer/nontracked.h tests/auto/qsharedpointer/nontracked.cpp Change-Id: I3f9074923b4d6bd4666258ab04f01476cc6e901c Reviewed-by: Thiago Macieira <thiago.macieira@intel.com> Reviewed-by: Sergio Ahumada <sahumada@blackberry.com> Reviewed-by: Lars Knoll <lars.knoll@digia.com>
* Update copyright year in Digia's license headersSergio Ahumada2013-01-131-1/+1
| | | | | | Change-Id: I52bf8ef0447b701b4ebf7d7d240013a72adb9425 Reviewed-by: Akseli Salovaara <akseli.salovaara@digia.com> Reviewed-by: Sergio Ahumada <sergio.ahumada@digia.com>
* Change copyrights from Nokia to DigiaIikka Eklund2012-09-291-24/+24
| | | | | | | | Change copyrights and license headers from Nokia to Digia Change-Id: I280c0a575987d1770e354b4948f1d4d767d711ea Reviewed-by: Simo Fält <simo.falt@digia.com> Reviewed-by: Sergio Ahumada <sergio.ahumada@digia.com>
* Update contact information in license headers.Sergio Ahumada2012-08-011-2/+2
| | | | | | | | | | | | | - Replace Nokia contact email address with Qt Project website. - Remove "All rights reserved" line from license headers. 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: Ie7ba62011752fcb149b99b26317c54f2a0cfa931 Reviewed-by: Jason McDonald <jason.mcdonald@nokia.com>
* Update year in Nokia copyright messages.Jason McDonald2012-01-111-1/+1
| | | | | | | | | Simple search and replace. This commit doesn't touch 3rd-party files, nor translations (where the change is not so simple and will be handled in a separate commit). Change-Id: I4e48513b8078a44a8cd272326685b25338890148 Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
* Update licenseheader text in source files for qt4.8Jyri Tahtela2011-05-131-17/+17
| | | | | | | Updated version of LGPL and FDL licenseheaders. Apply release phase licenseheaders for all source files. Reviewed-by: Trust Me
* Support gamma correction of text on GLEskil Abrahamsen Blomfeldt2011-05-031-8/+2
| | | | | | | | | | | | | | | | | | | | | | | | If the SRGB framebuffer extension in GL is available, we can support gamma correction of text with a gamma of 2.1. On Mac this is sufficient for gamma correcting subpixel antialiased text. Gray antialiasing should not be gamma corrected on Mac. On Windows, the user can potentially set the gamma value to anything between 1.0 and 2.2 (or something like that). We support anything that resembles 1.0 closely enough by pushing the text out without any correction (like before). We also support anything that resembles 2.1 (the gamma hardcoded in GL's SRGB extension) by turning on the extension before blending the text. In between the two, we'll use gray antialiasing to avoid differing too much from the raster engine (which is our reference in this.) For gray antialiasing on Windows, we use a constant gamma of 2.3 which has been determined by experimentation. Since this is close enough to 2.1 we do gamma correction with SRGB extension. The distance limit of 0.2 is determined by some experimentation. Reviewed-by: Samuel
* Added support for six-parameter radial gradients.Samuel Rødal2011-04-131-9/+33
| | | | | | | | The extended radial gradients conform to the radial gradient specification in HTML 5 canvas. Task-number: QTBUG-14075 Reviewed-by: Andreas Kling
* Improved gradient table generation performance for two-stop gradients.Samuel Rødal2011-04-131-0/+5
| | | | | | | | Two stops is a fairly common case so we gain quite a bit by special casing it. Improves performance by 10 % in parcycle benchmark, and by 90 % in a synthetic benchmark. Reviewed-by: Andreas Kling
* Optimized radial gradient fetch using SSE 2.Samuel Rødal2011-04-131-0/+5
| | | | | | | On an i7 this improves performance by 22 % in parcycle, 107 % in default svgviewer example, and 283 % in a synthetic radial gradient benchmark. Reviewed-by: Andreas Kling
* Prepared for SIMD implementation of radial gradients.Samuel Rødal2011-04-131-177/+23
| | | | | | | Made the radial gradient fetch func into a template to be able to optimize the inner loop using SIMD instructions. Reviewed-by: Benjamin Poulain
* Get rid of double conversions and arithmetic when qreal is float.Samuel Rødal2011-04-061-38/+38
| | | | | Task-number: QT-4624 Reviewed-by: Kim
* Fix compilation error.Jason McDonald2011-01-191-0/+2
| | | | | | | | Protect #include of feature-specific header in #ifdef. Task-number: QTBUG-16652 Reviewed-by: Trust Me Acked-by: Jani Hautakangas
* Update copyright year to 2011.Jason McDonald2011-01-101-1/+1
| | | | Reviewed-by: Trust Me
* Improve performance of bilinear upscaling of images with neon.Samuel Rødal2010-12-221-3/+32
| | | | | | | Gives an improvement of around 19.6 % for upscaling a 256x256 pixmap to 480x480 with SmoothPixmapFilter enabled. Reviewed-by: Olivier Goffart
* Improve performance of bilinear downscaling of images with neon.Samuel Rødal2010-12-221-26/+107
| | | | | | | Gives a 36 % performance boost for downscaling a 512x512 pixmap to 480x480 with SmoothPixmapFilter enabled. Reviewed-by: Andreas Kling
* Prevent crash in drawhelper code when the cpu has MMXEXT but no SSE.Samuel Rødal2010-12-151-11/+0
| | | | | | | | | | We can't use the qdrawhelper_sse.cpp or qdrawhelper_sse3dnow.cpp when SSE is not run-time detected, as those are compiled with -msse and MMXEXT is just a subset of SSE. The compiler might choose to use an instruction not in the subset, causing a crash at run-time. Task-number: QTBUG-15693 Reviewed-by: Thiago Macieira
* Prevent out-of-bounds memory access in drawhelper.Samuel Rødal2010-12-081-0/+2
| | | | | | | | | | The coordinates should be modulo the image width and height like for all the other tiled blend functions. Covered by perspectives.qps Task-number: QTBUG-15837 Reviewed-by: Olivier Goffart
* Clean up ARM SIMD drawhelper code and make sure it works.Jani Hautakangas2010-10-151-128/+31
| | | | | | | | | | Blending argb32_pre pixmaps is now ARM SIMD boosted because it's the most crucial and most used feature in applications. ARM SIMD boosted blending is 30% faster than non ARM SIMD boosted blending. Task-number: QTBUG-4891 Reviewed-by: Samuel
* Fixed compilation of qdrawhelper.cpp with XLC 10Rohan McGovern2010-09-161-1/+1
| | | | | | | With XLC 10, calls to static functions within a template function need to be fully qualified. Reviewed-by: Marius Storm-Olsen
* qdrawhelper: Optimisations in fetchTransformedBilinearOlivier Goffart2010-09-081-30/+59
| | | | | | | Scaling down, no rotation, with SSE2. Specialize for ARGB32 not tiled, so the pixelbound can be simplified Reviewed-by: Samuel
* qdrawhelper: use SSE2 for interpolation in fetchTransformedBilinearOlivier Goffart2010-09-081-0/+77
| | | | | | | | when scale down, with no rotations Process the pixel 4 by 4 and do the interpolation using SSE2 Reviewed-by: Samuel
* qdrawhelper: small optimisations in fetchTransformBilinearOlivier Goffart2010-09-081-16/+15
| | | | | | | | Another way to compute the interpolation that does less multiplications. Small inpact on benchmark Made-with: Samuel
* Allow Windows x64 to use SSE2 to speed up blendingliang jian2010-09-061-105/+119
| | | | | | | | | | | Windows 64 does not support MMX with MSVC. This is a problem with the way SSE is currently used because it rely on previous vector instructions being available. This patches fixes that by using the intended functions for SSE2 on Windows. Merge-request: 792 Reviewed-by: Benjamin Poulain <benjamin.poulain@nokia.com>
* qdrawhelper: micro optimisation in fetchTransformBilinearOlivier Goffart2010-09-031-7/+7
| | | | | | move the -1 out of the loop Reviewed-by: Benjamin Poulain
* qdrawhelper: Remove blend_transformed_bilinear_argbOlivier Goffart2010-09-021-179/+4
| | | | | | | With the recent optimisation in fetchTransformedBilinear, the generic path is faster than the 'optimized' path Reviewed-by: Samuel
* Use NEON and preloading for 16 bit small / medium sized image blits.Samuel Rødal2010-09-011-0/+1
| | | | | | | | | | | | | | This gives a nice speedup for blitting of small and medium sized images by using preloading and avoiding function call overhead to memcpy for each scanline. For larger image widths memcpy becomes more efficient. Speedups of up to 40 % for 64 pixel wide images were measured. For image widths between 2 and 16 the speedup ranges between 12 % and 28 %. Task-number: QT-3401 Reviewed-by: Benjamin Poulain <benjamin.poulain@nokia.com>
* Undefined SSE symbols when crosscompiling Qt on PPC.Benjamin Poulain2010-08-311-1/+0
| | | | | | | | | | Qt does not build on PowerPC when compiling for both x86 and PPC on Mac. The compiler is invoked only once for both architecture so the defines are there in order to get the optimized path for x86. Those defines needs to be removed from the compilation environment when the target is set to PPC by GCC. Reviewed-by: Kent Hansen
* qdrawhelper: backport the optimisations in fetchTransformBilinear from ↵Olivier Goffart2010-08-301-138/+257
| | | | | | | | | | | | | | | master to 4.7 This backport the following commits: e55b6a3 qdrawhelper: remove code duplication 0d7e683 qdrawhelper: optimize fetchTransformedBilinear 29ef46e Fix compilation with RVCT 6601458 qdrawhelper: Use SSE2 in fetchTransformedBilinear (when scalling up) 398ef0ca Fix nasty copy-paste bug in fetchTransformedBilinear() d585ece qdrawhelper: fix assert in fetchTransformedBilinear Reviewed-by: Benjamin Poulain
* Revert "Refactor blend_transformed_bilinear to simplify the blend type checking"Olivier Goffart2010-08-261-67/+66
| | | | | | | | | This reverts commit d2089600ea247b5d6354e8eee4becf40802c4693. Reverting in order to avoid conflict in the master branch. We are going to consider backporting the changes to Qt 4.7.1 Reviewed-by: Benjamin Poulain
* Refactor blend_transformed_bilinear to simplify the blend type checkingBenjamin Poulain2010-08-251-66/+67
| | | | | | | | | | | The function blend_transformed_bilinear_argb() was checking the blend type at runtime for each pixel in order to clamp the coordinates. This code was duplicated in both branch of the function. This patch factorize the code by doing the clamping in a template function. Reviewed-by: Samuel Rødal
* Implement qt_memfill32 with Neon.Benjamin Poulain2010-08-251-0/+1
| | | | | | | | | | | This patch introduce a implementation of qt_memfill32 with the Neon instructions set from ARMv7. The loop is unrolled 1 time to get better performance. This implementation of memfill is 330% faster on the N900. Reviewed-by: Samuel Rødal
* Implement the composition mode Plus with Neon.Benjamin Poulain2010-08-251-6/+3
| | | | | | | | | On the benchmark tst_QPainter::compositionModes(), this patches gives the following improvements: -300x300:opaque: 390% -300x300:!opaque: 1085% Reviewed-by: Samuel Rødal
* Implement the general blending of ARGB32_pm with SSSE3Benjamin Poulain2010-08-161-0/+11
| | | | | | | | | | | | | | | | | | | SSSE3 provides two tools to improve the blending speed over SSE2: -palignr -byte permutation The alignement is enforced on src and dst with palignr to always make aligned access. The extraction of the alpha mask is done with a byte permutation in order to save two instructions per cycle. On Atom, this patch gives between 0% (aligned src) to 10% of improvement (unaligned 4 and 12 bytes). On Core 2, this patch gives consistently 8% to 10% of improvement for every miss-alignment. Reviewed-by: Samuel Rødal
* Implement comp_Source with SSE2 when there is a const alphaBenjamin Poulain2010-08-041-0/+2
| | | | | | On Atom, comp_Source is 280% faster with the SSE2 implementation. Reviewed-by: Andreas Kling
* Implement the composition mode "Plus" with SSE2Benjamin Poulain2010-07-271-5/+3
| | | | | | | | | | | Implement the composition function for CompositionMode_Plus with SSE2. The macro MIX() can be replaced by a single instruction add-saturate, which increase the speed a lot (13 times faster on the blend benchmark). Reviewed-by: Olivier Goffart Reviewed-by: Andreas Kling
* Clean the CompositionFunction tables of drawhelperBenjamin Poulain2010-07-271-68/+59
| | | | | | | | | | | | | | | | Some instructions sets were defining partial table of composition functions. To work around that, qInitDrawhelperAsm() was resetting the composition function of QPainter::CompositionMode_Destination and anything above QPainter::CompositionMode_Xor. This was a problem because it makes it impossible to implement fast path for those composition mode. This patch export prototypes for the generic functions of each composition mode. The specialized implementations now define a complete table. Reviewed-by: Andreas Kling
* Fixed bug in drawTiledPixmap when width of pixmap matches target rect.Samuel Rødal2010-07-011-1/+2
| | | | | | | qt_memconvert's duff's device implementation assumes that count is > 0, if count is 0 it will still blit eight pixels. Reviewed-by: Trond
* Improved performance of 16 bit memrotates using NEON instructions.Samuel Rødal2010-07-011-0/+3
| | | | | | | | | | | Make the memrotate functions a function pointer table so that we can replace it with optimized versions, and implement an optimized NEON version for the 90 and 270 rotations. Measured performance improvement for a 400x400 16-bit pixmap was 17 % for 270 degree rotation and 11 % for 90 degree rotation. Reviewed-by: Trond
* Add an implementation of comp_func_solid_SourceOver_neon() with Neon.Benjamin Poulain2010-06-231-0/+1
| | | | | | | | | | The function comp_func_solid_SourceOver_neon() is use extensively by WebKit via the calls to fillRect() of QPainter(). Implementing the function with Neon provides some performance improvement (around 175% of the previous speed). Reviewed-by: Samuel Rødal
* Add a SSE2 implementation of comp_func_solid_SourceOver()Benjamin Poulain2010-06-231-1/+4
| | | | | | | | This function is used quite a lot by WebKit animations, the SSE2 implementation is twice as fast in those uses cases. Reviewed-by: Andreas Kling Reviewed-by: Samuel Rødal
* Add a SSE2 version of comp_func_SourceOver()Benjamin Poulain2010-06-231-1/+6
| | | | | | | | | | | | Implement a version of comp_func_SourceOver() with SSE2. This gives a performance boost of 11% on some WebKit animations. Two new macros were added to simplify the implementation of the different blending primitives: BLEND_SOURCE_OVER_ARGB32_SSE2() and BLEND_SOURCE_OVER_ARGB32_WITH_CONST_ALPHA_SSE2() Reviewed-by: Samuel Rødal
* Fixed pixel-bleeding when stretching subrected pixmaps.Gunnar Sletta2010-04-201-8/+13
| | | | | | | | | When stretching a subrect of a pixmap we need to clamp the sampling to the subrect. This was done for the ARGB32_Premultiplied target format but not for the generic fallback. This patch adapts the code so that the two code paths are equivalent. Reviewed-by: Samuel
* qdrawhelper: fix optim in 2245641baOlivier Goffart2010-04-131-29/+88
| | | | | | | | | | The commit 2245641baa58125b57faf12496bc472491565498 was not correct as the x2 (and y2) were not correct on the first line (the value was 1 instead of 0) Fixes tst_QImage::smoothScale3 Reviewed-by: Samuel Rødal
* Safeguard ourselves against corrupt registry values for cleartype gammaGunnar Sletta2010-04-121-0/+5
| | | | | Reviewed-by: Eskil Task: http://bugreports.qt.nokia.com/browse/QTBUG-7596
* qdrawhelper: optimize the fetch transformed bilinear functionsOlivier Goffart2010-04-121-46/+46
| | | | | | | | | Since we know that x1 and y1 are between the bounds, and than x2 and y2 are bigger, we only need to check the upper bound for x2 and y2 This gives 5% speedup on a trace from the QML samegame demo. Reviewed-by: Samuel Rødal
* QDrawHelper: Reduce code duplicationsOlivier Goffart2010-04-091-215/+61
| | | | | | | And apply same optimisation as in 6a9e6f4780f5fc3aaf34f93c85de575f81c91e48 for duplicated code Reviewed-by: Samuel Rødal
* Speedup fetchTransformedBilinear in the fast_matrix caseOlivier Goffart2010-04-091-7/+9
| | | | | | 3% faster on a trace from the qml samegame. Reviewed-by: Samuel Rødal
* Optimized SourceOver and 16 bit dest fetches, dest stores using NEON.Samuel Rødal2010-03-261-4/+8
| | | | | | | | This makes for example linear gradient blending on top of RGB16 156 % faster (from 20.4 fps to 52.3 fps in my benchmark). Task-number: QTBUG-6684 Reviewed-by: Gunnar Sletta
* Optimized scaled/transformed image blending for ARGB32PM and RGB16 on RGB16.Samuel Rødal2010-03-261-0/+6
| | | | | | | | | | | | | Before: :/traces/qmlphoneconcept.trace, iterations: 5, frames: 48, min(ms): 1207, median(ms): 1212, stddev: 0,165153 %, max(fps): 39,768020 After: traces/qmlphoneconcept.trace, iterations: 3, frames: 48, min(ms): 884, median(ms): 886, stddev: 0,383097 %, max(fps): 54,298643 Task-number: QTBUG-6684 Reviewed-by: Gunnar Sletta
generated by cgit v0.12 at 2025-09-06 15:58:18 (GMT)