diff options
| author | Allan Sandfeld Jensen <allan.jensen@theqtcompany.com> | 2015-01-27 14:08:17 +0100 |
|---|---|---|
| committer | Allan Sandfeld Jensen <allan.jensen@theqtcompany.com> | 2015-02-02 11:14:00 +0000 |
| commit | 9be87085d848c9615319d1be7d35a1f9963c3380 (patch) | |
| tree | 1576389efaac9a92f5a10e5d16994e02536daa82 /tests | |
| parent | 1749c1dd425fdba82eb0f1a0761c4c3d71d55319 (diff) | |
Only use 32bit version of qPremultiply
With auto-vectorization enabled in QtGui, the 32bit version of
qPremultiply is faster than the 64bit version since it can be vectorized
wider (4x on 128bit as opposed to 2x). Since all our important 64bit
targets have SIMD, that makes the 64bit version pointless.
Change-Id: I4e9070a3a3c8e2b54f17a95ba0aee0405cbb8ec9
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Diffstat (limited to 'tests')
| -rw-r--r-- | tests/benchmarks/gui/image/qimageconversion/tst_qimageconversion.cpp | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/tests/benchmarks/gui/image/qimageconversion/tst_qimageconversion.cpp b/tests/benchmarks/gui/image/qimageconversion/tst_qimageconversion.cpp index 00565c4f31..9d6bdb28fa 100644 --- a/tests/benchmarks/gui/image/qimageconversion/tst_qimageconversion.cpp +++ b/tests/benchmarks/gui/image/qimageconversion/tst_qimageconversion.cpp @@ -147,6 +147,7 @@ void tst_QImageConversion::convertGeneric_data() QTest::newRow("argb32 -> rgb666") << argb32 << QImage::Format_RGB666; QTest::newRow("argb32 -> argb8565pm") << argb32 << QImage::Format_ARGB8565_Premultiplied; QTest::newRow("argb32 -> argb4444pm") << argb32 << QImage::Format_ARGB4444_Premultiplied; + QTest::newRow("argb32 -> argb32pm") << argb32 << QImage::Format_ARGB32_Premultiplied; } void tst_QImageConversion::convertGeneric() |