AuroraMiddleware/skia2
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
4e82cdb90a
skia2/bench
History
commit-bot@chromium.org 0be2d8354b Fix memcpy32_sse2_unalign. 
The whole point of mempcy32_sse2_unalign is that we didn't align dst128
and src128.  So it's not safe at all to cast them back to dst and src.
That tells the compiler that dst/src are 128-bit aligned, and then it
autovectorizes the cleanup while-loop using that (false) knowledge with
aligned SSE instructions.

This leads to crashes on memcpy32_sse2_unalign_10, which is small enough
that we actually get non-16-byte aligned memory.  The larger size
benches could be crashing too, but they're big enough allocations that
they're probably always 16-byte aligned anyway.

BUG=skia:2589
R=fmalita@chromium.org, mtklein@google.com

Author: mtklein@chromium.org

Review URL: https://codereview.chromium.org/291893008

git-svn-id: http://skia.googlecode.com/svn/trunk@14851 2bbb7eff-a529-9590-31e7-b0007b416f81
2014-05-22 18:24:42 +00:00
..
BenchTool/BenchTool.xcodeproj
AAClipBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
bench_compare.py Make bench_compare executable. 2013-11-12 15:25:31 +00:00
bench_util.py gen_bench_expectations to generate bench expectations files. 2014-03-17 21:16:29 +00:00
BenchGpuTimer_gl.cpp
BenchGpuTimer_gl.h
benchmain.cpp Generate bench/Android.mk from gyp. 2014-05-16 13:15:41 +00:00
BenchSysTimer_c.cpp
BenchSysTimer_c.h
BenchSysTimer_mach.cpp
BenchSysTimer_mach.h
BenchSysTimer_posix.cpp
BenchSysTimer_posix.h
BenchSysTimer_windows.cpp Use lowercase windows.h in includes to fix Windows cross compilation using mingw. 2013-12-03 15:23:37 +00:00
BenchSysTimer_windows.h Use lowercase windows.h in includes to fix Windows cross compilation using mingw. 2013-12-03 15:23:37 +00:00
BenchTimer.cpp extend SkBenchmark to allow a bench to return a durationScale, which allows it to perform fewer actual interations, but report a scale factor to account for that. Thus a very slow bench can be compared head-to-head with a faster one, w/o actually forcing the tool to run for the full duration of the slower test. 2013-05-29 15:39:54 +00:00
BenchTimer.h extend SkBenchmark to allow a bench to return a durationScale, which allows it to perform fewer actual interations, but report a scale factor to account for that. Thus a very slow bench can be compared head-to-head with a faster one, w/o actually forcing the tool to run for the full duration of the slower test. 2013-05-29 15:39:54 +00:00
BicubicBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
BitmapBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
BitmapRectBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
BitmapScaleBench.cpp Adding this call simulates seeing a *different* bitmap every time we scale, which changes the code path (places much heavier load on constructing a new filter). This better emulates the kind of environment most embedders see us in, where they try to cache a scaled bitmap rather than frequently rescaling it. 2014-03-06 17:16:37 +00:00
BlurBench.cpp move common blur types into central header 2014-04-28 16:25:35 +00:00
BlurImageFilterBench.cpp Factory methods for heap-allocated SkImageFilter objects. 2014-03-10 10:51:58 +00:00
BlurRectBench.cpp move common blur types into central header 2014-04-28 16:25:35 +00:00
BlurRoundRectBench.cpp move common blur types into central header 2014-04-28 16:25:35 +00:00
ChartBench.cpp Remove global SkRandoms from ChartBench. 2014-05-08 15:12:43 +00:00
check_bench_regressions.py Adds dashboard link for one-line plot on each alert item. 2014-05-13 17:38:17 +00:00
ChecksumBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
ChromeBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
CmapBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
ColorFilterBench.cpp Factory methods for heap-allocated SkColorFilter objects. 2014-02-21 18:46:30 +00:00
ColorPrivBench.cpp Make sure four_byte_interp benches the entire FourByteInterp call. 2013-12-05 16:43:08 +00:00
CoverageBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
DashBench.cpp reduce total number of segments in giant-dash 2014-05-01 15:30:52 +00:00
DecodeBench.cpp Upstream fix for DecodeBench. 2014-02-20 15:41:17 +00:00
DeferredCanvasBench.cpp Reduce internal explicit SaveFlags usage. 2014-04-18 14:19:31 +00:00
DeferredSurfaceCopyBench.cpp Generate bench/Android.mk from gyp. 2014-05-16 13:15:41 +00:00
DisplacementBench.cpp Factory methods for heap-allocated SkImageFilter objects. 2014-03-10 10:51:58 +00:00
FontCacheBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
FontScalerBench.cpp Clean up global gSkSuppressFontCachePurgeSpew. 2014-05-12 15:48:38 +00:00
FSRectBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
GameBench.cpp GameBench: curCell doesn't need to be extra state. 2014-05-07 17:12:55 +00:00
gen_bench_expectations.py Adds the ability to filter bench expectations in builder/config level. 2014-05-12 14:36:43 +00:00
GradientBench.cpp Remove unused (by clients) SkUnitMapper 2014-05-22 12:27:41 +00:00
GrMemoryPoolBench.cpp DM: also run benches once. 2014-02-26 23:01:57 +00:00
GrOrderedSetBench.cpp Add GrSet class built on top of RedBlackTree 2014-02-27 20:23:22 +00:00
GrResourceCacheBench.cpp Factor GrTexture into public GrTexture and private GrTextureImpl. 2014-05-09 20:46:48 +00:00
gUniqueGlyphIDs.h switch array to uint16_t to fix overflows 2013-06-11 19:53:19 +00:00
HairlinePathBench.cpp Fix bench to compile when SK_SUPPORTS_GPU is not provided as a compiler option 2014-02-19 19:30:42 +00:00
ImageCacheBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
ImageDecodeBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
ImageFilterDAGBench.cpp Add a new bench that tests a simple image filter DAG. 2014-04-10 16:16:13 +00:00
InterpBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
LightingBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
LineBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
MagnifierBench.cpp Factory methods for heap-allocated SkImageFilter objects. 2014-03-10 10:51:58 +00:00
MathBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
Matrix44Bench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
MatrixBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
MatrixConvolutionBench.cpp Clean up SkImageFilter. 2014-03-12 16:36:08 +00:00
MemcpyBench.cpp Fix memcpy32_sse2_unalign. 2014-05-22 18:24:42 +00:00
MemoryBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
MemsetBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
MergeBench.cpp Factory methods for heap-allocated SkImageFilter objects. 2014-03-10 10:51:58 +00:00
MorphologyBench.cpp Factory methods for heap-allocated SkImageFilter objects. 2014-03-10 10:51:58 +00:00
MutexBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
PathBench.cpp Cap memory usage in path_create bench. 2014-01-06 20:19:14 +00:00
PathIterBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
PathUtilsBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
PerlinNoiseBench.cpp SkPerlinNoiseShader::CreateTubulenceNoise() is misspelled 2014-04-01 16:09:37 +00:00
PicturePlaybackBench.cpp Staging for cleanup of SkPicture-related headers 2014-04-18 18:04:41 +00:00
PictureRecordBench.cpp Staging for cleanup of SkPicture-related headers 2014-04-18 18:04:41 +00:00
PremulAndUnpremulAlphaOpsBench.cpp Sanitizing source files in Housekeeper-Nightly 2014-03-18 03:02:11 +00:00
QuadTreeBench.cpp Rename benchmark classes, hopefully less confusing to valgrid? 2014-04-23 21:03:45 +00:00
ReadPixBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
RectBench.cpp Move non-trivial work in RectBench() to onPreDraw. 2013-12-05 12:07:44 +00:00
RectoriBench.cpp move common blur types into central header 2014-04-28 16:25:35 +00:00
RefCntBench.cpp Fix compilation with SK_ENABLE_INST_COUNT=1 2013-12-05 12:08:12 +00:00
RegionBench.cpp Sanitizing source files in Housekeeper-Nightly 2014-01-02 07:01:35 +00:00
RegionContainBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
RepeatTileBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
ResultsWriter.h Changed JSON formatting more, workaround bug where SkStringPrintf causes encoding issues 2014-05-20 17:35:10 +00:00
RTreeBench.cpp Rename benchmark classes, hopefully less confusing to valgrid? 2014-04-23 21:03:45 +00:00
ScalarBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
ShaderMaskBench.cpp Remove support for inheriting the paint color from SkColorShader 2014-04-24 16:54:46 +00:00
SkBenchLogger.cpp
SkBenchLogger.h Clean up SkTypes.h. 2013-10-11 18:50:45 +00:00
SkBenchmark.cpp Fix compilation with SK_ENABLE_INST_COUNT=1 2013-12-05 12:08:12 +00:00
SkBenchmark.h DM: also run benches once. 2014-02-26 23:01:57 +00:00
SkGMBench.cpp DM: also run benches once. 2014-02-26 23:01:57 +00:00
SkGMBench.h Allow GMs to be used as benchmarks. Make convex_poly_clip opt in. 2014-02-03 14:48:17 +00:00
SkipZeroesBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
SortBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
StackBench.cpp Revert of https://codereview.chromium.org/127223004/ 2014-01-08 21:31:11 +00:00
StrokeBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
TableBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
TextBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
tile_analyze.py
TileBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
TimerData.cpp Fix handling of aggregate times in TimerData::getResult 2013-12-17 16:32:51 +00:00
TimerData.h Committed: http://code.google.com/p/skia/source/detail?r=10372 2013-07-31 20:00:56 +00:00
VertBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00
WritePixelsBench.cpp Sanitizing source files in Housekeeper-Nightly 2014-03-08 03:02:06 +00:00
WriterBench.cpp Convert SkWriter32 to use an SkTDArray for its internal storage. 2014-01-14 20:51:26 +00:00
XfermodeBench.cpp Simplify benchmark internal API. 2013-12-03 18:17:16 +00:00