AuroraMiddleware/skia2
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
10,467 Commits 1 Branch 0 Tags 544 MiB
709ca75f03
Commit Graph

699 Commits

Author SHA1 Message Date
commit-bot@chromium.org
3a9ade7f37 Updates bench expectations; adds new platforms. 
BUG=skia:
R=borenet@google.com, robertphillips@google.com

Author: bensong@google.com

Review URL: https://codereview.chromium.org/131343009

git-svn-id: http://skia.googlecode.com/svn/trunk@13179 2bbb7eff-a529-9590-31e7-b0007b416f81
 2014-01-24 21:50:03 +00:00
commit-bot@chromium.org
6dda827913 add leaks flag to show unref'd insts 
Leaks are shown optionally instead of always for tests, gm, and bench.
The current display does not show actual leaks necessarily, but
shows global objects that were not deleted when the test ended.
To enable the end-of-run leak display, pass --leaks or -l.

BUG=skia:
R=mtklein@google.com

Author: caryclark@google.com

Review URL: https://codereview.chromium.org/143943009

git-svn-id: http://skia.googlecode.com/svn/trunk@13151 2bbb7eff-a529-9590-31e7-b0007b416f81
 2014-01-23 17:21:19 +00:00
commit-bot@chromium.org
19382421b9 Convert SkWriter32 to use an SkTDArray for its internal storage. 
This reduces the allocation overhead of a null picture (create, beginRecording(), endRecording) from about 18K to about 1.9K.  (There's still lots more to prune.)

SkPictureFlat can exploit the fact that Writer32 is contiguous simplify its memory management.  The Writer32 itself becomes the scratch buffer.

Remove lots and lots of arbitrary magic numbers that were size guesses and minimum allocation sizes.  Keep your eyes open for the big obvious DUH why we save 16K per picture!  (Spoiler alert.  It's because that first save we issue in beginRecording() forces the old SkWriter32 to allocate 16K.)

Tests passing, DM passing.

bench --match writer: ~20% faster
null bench_record: ~30% faster
bench_record on buildbot .skps: ~3-6% slower, ranging 25% faster to 20% slower
bench_pictures on buildbot .skps: ~1-2% faster, ranging 13% faster to 28% slower

BUG=skia:1850
R=reed@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/137433003

git-svn-id: http://skia.googlecode.com/svn/trunk@13073 2bbb7eff-a529-9590-31e7-b0007b416f81
 2014-01-14 20:51:26 +00:00
commit-bot@chromium.org
c45ece5f7f NEON fast path for box blur 
Calculate 8 channels in parallel by using 16-bits to store each channel. Due to the limitation of VQRDMULH, (int16 * int16 * 2 + 0x8000) >> 16, the fast path can only support kernelSize < 128.
8 significant bits are kept at least in each stage, the final error should less-equal than 1.

Pre-fetching memory for X-direction read. In fact pre-fetching memory doesn't help much for Y direction read, since it is a waste to load a cache line for only read 8 bytes.(I left it there to keep the symmetry. pre-fetch is cheap :) )

bench data on Nexus 10
before:
running bench [640 480]      blur_image_filter_large_10.00_10.00   8888:  cmsecs =  25081.48
running bench [640 480]      blur_image_filter_small_10.00_10.00   8888:  cmsecs =  25038.04
running bench [640 480]        blur_image_filter_large_1.00_1.00   8888:  cmsecs =  25209.04
running bench [640 480]        blur_image_filter_small_1.00_1.00   8888:  cmsecs =  24928.01
running bench [640 480]        blur_image_filter_large_0.00_1.00   8888:  cmsecs =  17160.98
running bench [640 480]       blur_image_filter_large_0.00_10.00   8888:  cmsecs =  17924.11
running bench [640 480]        blur_image_filter_large_1.00_0.00   8888:  cmsecs =  14609.19
running bench [640 480]       blur_image_filter_large_10.00_0.00   8888:  cmsecs =  14625.91

after:
running bench [640 480]      blur_image_filter_large_10.00_10.00   8888:  cmsecs =  14848.42
running bench [640 480]      blur_image_filter_small_10.00_10.00   8888:  cmsecs =  16037.29
running bench [640 480]        blur_image_filter_large_1.00_1.00   8888:  cmsecs =  14819.55
running bench [640 480]        blur_image_filter_small_1.00_1.00   8888:  cmsecs =  14563.69
running bench [640 480]        blur_image_filter_large_0.00_1.00   8888:  cmsecs =  11905.34
running bench [640 480]       blur_image_filter_large_0.00_10.00   8888:  cmsecs =  11883.85
running bench [640 480]        blur_image_filter_large_1.00_0.00   8888:  cmsecs =   9576.51
running bench [640 480]       blur_image_filter_large_10.00_0.00   8888:  cmsecs =   9793.84

BUG=
R=senorblanco@chromium.org, mtklein@google.com, reed@google.com, kevin.petit@arm.com, kevin.petit.arm@gmail.com

Author: zheng.xu@arm.com

Review URL: https://codereview.chromium.org/105893003

git-svn-id: http://skia.googlecode.com/svn/trunk@13036 2bbb7eff-a529-9590-31e7-b0007b416f81
 2014-01-13 08:16:45 +00:00
borenet@google.com
6f0f5b4aad Remove references to Skia's SVN repository 
BUG=
R=epoger@google.com

Review URL: https://codereview.chromium.org/132423002

git-svn-id: http://skia.googlecode.com/svn/trunk@13006 2bbb7eff-a529-9590-31e7-b0007b416f81
 2014-01-09 21:41:39 +00:00
commit-bot@chromium.org
1034ef1af4 Revert of https://codereview.chromium.org/127223004/ 
Reason for revert: VS can't comprehend std::vector without exceptions.

R=bsalomon@google.com, caryclark@google.com, mtklein@chromium.org
TBR=bsalomon@google.com, caryclark@google.com, mtklein@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/129443002

git-svn-id: http://skia.googlecode.com/svn/trunk@12975 2bbb7eff-a529-9590-31e7-b0007b416f81
 2014-01-08 21:31:11 +00:00
commit-bot@chromium.org
506ecc24bf For comparison, add std::vector. 
Where I've tested, it's second-place to SkTDArray, though the closest API equivalent in our code is SkTArray, which it's significantly faster than.  This is probably a good approximation to how fast we can possibly make something like SkTArray that constructs and destroys its values.  In fact, it might be worth trying making SkTArray a thin shim over std::vector.

BUG=
R=bsalomon@google.com, caryclark@google.com, mtklein@google.com

Author: mtklein@chromium.org

Review URL: https://codereview.chromium.org/127223004

git-svn-id: http://skia.googlecode.com/svn/trunk@12969 2bbb7eff-a529-9590-31e7-b0007b416f81
 2014-01-08 21:16:04 +00:00
commit-bot@chromium.org
85facf71ef Add StackBench to measure performance on stack-like (fixed element size) work loads. 
BUG=303282
R=reed@google.com, caryclark@google.com, mtklein@chromium.org

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/110893007

git-svn-id: http://skia.googlecode.com/svn/trunk@12940 2bbb7eff-a529-9590-31e7-b0007b416f81
 2014-01-07 17:03:31 +00:00
commit-bot@chromium.org
8f88117940 Cap memory usage in path_create bench. 
Memory usage grows unbounded in path_create without this patch (growing the
paths).  This bench also somewhat needlessly cycles through 32 paths, so now
we just work with one.

Peak memory usage drops from ~2-3G to ~150M.  This should fix the NexusS crashes,
or at least get us to the next one.

BUG=skia:1687
R=caryclark@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/111283007

git-svn-id: http://skia.googlecode.com/svn/trunk@12925 2bbb7eff-a529-9590-31e7-b0007b416f81
 2014-01-06 20:19:14 +00:00
skia.committer@gmail.com
f8affa2e5c Sanitizing source files in Housekeeper-Nightly 
git-svn-id: http://skia.googlecode.com/svn/trunk@12853 2bbb7eff-a529-9590-31e7-b0007b416f81
 2014-01-02 07:01:35 +00:00
mike@reedtribe.org
ae8f9528fd speedup SkRect::intersect 
git-svn-id: http://skia.googlecode.com/svn/trunk@12851 2bbb7eff-a529-9590-31e7-b0007b416f81
 2014-01-01 20:32:45 +00:00
tfarina@chromium.org
725a64cf89 Add default resource path for skia test cases in gm and bench tools. 
BUG=skia:1765
TEST=./out/Release/gm --match downsamplebitmap_image
R=mtklein@google.com, scroggo@google.com, yunchao.he@intel.com

Review URL: https://codereview.chromium.org/120873002

git-svn-id: http://skia.googlecode.com/svn/trunk@12850 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-31 14:29:52 +00:00
commit-bot@chromium.org
5f6c2c0477 Rebaseline for Nexus7 tile multithreads config. Widens the range for benched calculated from average. 
BUG=
R=robertphillips@google.com

Author: bensong@google.com

Review URL: https://codereview.chromium.org/118593002

git-svn-id: http://skia.googlecode.com/svn/trunk@12772 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-19 16:20:04 +00:00
commit-bot@chromium.org
6cb74f2b86 Adjusts Ubuntu bench multi_4_threads values. Widens range to cope with the new average instead of 25th. 
BUG=
R=robertphillips@google.com

Author: bensong@google.com

Review URL: https://codereview.chromium.org/108653014

git-svn-id: http://skia.googlecode.com/svn/trunk@12736 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-18 14:55:02 +00:00
commit-bot@chromium.org
b6a1c0154d Adjusts Win7 bench tile_256x256_multi_4_threads. Using 8f4d230. 
BUG=
R=robertphillips@google.com

Author: bensong@google.com

Review URL: https://codereview.chromium.org/117523002

git-svn-id: http://skia.googlecode.com/svn/trunk@12721 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-17 20:23:22 +00:00
reed@google.com
e1ca705cac deprecate SkScalarRound (and its ilk), use SkScalarRound[ToInt,ToScalar]. #define SK_SUPPORT_DEPRECATED_SCALARROUND for legacy clients 
BUG=
R=robertphillips@google.com

Review URL: https://codereview.chromium.org/111353003

git-svn-id: http://skia.googlecode.com/svn/trunk@12719 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-17 19:22:07 +00:00
commit-bot@chromium.org
3fb15f4d21 Bench rebaseline for Ubuntu. Also manually adjust one entry for Nexus7. 
BUG= https://code.google.com/p/skia/issues/detail?id=1871
R=robertphillips@google.com

Author: bensong@google.com

Review URL: https://codereview.chromium.org/106593003

git-svn-id: http://skia.googlecode.com/svn/trunk@12718 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-17 18:39:27 +00:00
reed@google.com
8f4d2306fa remove SK_SCALAR_IS_[FLOAT,FIXED] and assume floats 
To keep the CL (slightly) managable, this does not make any changes to
existing macros (e.g. SkScalarMul). Just tackling #ifdef constructs this
time around.

BUG=
R=bsalomon@google.com, caryclark@google.com

Review URL: https://codereview.chromium.org/117053002

git-svn-id: http://skia.googlecode.com/svn/trunk@12712 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-17 16:44:46 +00:00
robertphillips@google.com
bbd893dcc3 Fix handling of aggregate times in TimerData::getResult 
https://codereview.chromium.org/117393002/



git-svn-id: http://skia.googlecode.com/svn/trunk@12709 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-17 16:32:51 +00:00
commit-bot@chromium.org
cea08483de Bench rebaseline for Nexus7 and Win7. Expected values are from versions afada4c and 48d94b8. 
BUG=
R=robertphillips@google.com, rmistry@google.com

Author: bensong@google.com

Review URL: https://codereview.chromium.org/107663007

git-svn-id: http://skia.googlecode.com/svn/trunk@12699 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-16 22:58:40 +00:00
commit-bot@chromium.org
cc6db406e4 Make sure four_byte_interp benches the entire FourByteInterp call. 
Relative bench performance still says switch to Fast.

Desktop:
    four_byte_interp_slow_256  11.68
    four_byte_interp_slow_255  12.11
    four_byte_interp_fast_256   7.17
    four_byte_interp_fast_255   8.17

N5:
    four_byte_interp_slow_256  28.49
    four_byte_interp_slow_255  25.08
    four_byte_interp_fast_256  19.40
    four_byte_interp_fast_255  21.69

BUG=
R=reed@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/102053006

git-svn-id: http://skia.googlecode.com/svn/trunk@12512 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-05 16:43:08 +00:00
commit-bot@chromium.org
ab1c13864d Fix compilation with SK_ENABLE_INST_COUNT=1 
Add INHERITED declarations to class declarations that prevent
compilation with the flag.

Remove SK_DEFINE_INST_COUNT from all class implementations.  Instead,
use function-local static variables in the reference count helper
classes to create the global instances to store the needed info. The
accessor functions are defined inline in the helper classes, so
definitions are not needed. The initialization point of the variables
should be as well defined as previously.

Remove SK_DECLARE_INST_COUNT_TEMPLATE and use SK_DECLARE_INST_COUNT
instead. This avoids possible future compilation errors further.

For SK_ENABLE_INST_COUNT=0 compilation, add an empty static member
function to all classes that use SK_DECLARE_INST_COUNT and
SK_DECLARE_INST_COUNT_ROOT macros. The function ensures that classes
contain public INHERITED typedef. This member function seems to be
compiled away. This shouĺd ensure that part of the compilation errors
are caught earlier.

Also adds DSK_DECLARE_INST_COUNT to few SkPDFDict subclasses.

R=robertphillips@google.com, richardlin@chromium.org, bsalomon@google.com

Author: kkinnunen@nvidia.com

Review URL: https://codereview.chromium.org/98703002

git-svn-id: http://skia.googlecode.com/svn/trunk@12501 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-05 12:08:12 +00:00
commit-bot@chromium.org
000d732ca7 Move non-trivial work in RectBench() to onPreDraw. 
Saw this on a profile where rects was filtered out.  ಠ_ಠ

BUG=
R=reed@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/93703003

git-svn-id: http://skia.googlecode.com/svn/trunk@12500 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-05 12:07:44 +00:00
bensong@google.com
967b258065 Fixes bench_util's CONFIG_RE to cope with space flexibility. 
The real "fix" down the road is Joe's ongoing work of switching to JSON.
Review URL: https://codereview.chromium.org/103943004

git-svn-id: http://skia.googlecode.com/svn/trunk@12498 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-05 01:31:56 +00:00
commit-bot@chromium.org
3361471a35 Simplify benchmark internal API. 
I'm not quite sure why I wrote such a convoluted API with setLoops()/getLoops().
This replaces it with a loops argument passed to onDraw().

This CL is largely mechanical translation from the old API to the new one.
MathBench used this->getLoops() outside onDraw(), which seems incorrect.  I
fixed it.

BUG=
R=djsollen@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/99893003

git-svn-id: http://skia.googlecode.com/svn/trunk@12466 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-03 18:17:16 +00:00
commit-bot@chromium.org
e3bb3bce3e Add JSON output option to bench. 
A new command-line flag --outResultsFile takes the filename to write the JSON into.

The human readable output is the same as before with one exception. Previously DEBUG would be printed if in debug mode, or nothing would be printed if in release mode. Now this is reported as a named option:

  build=DEBUG

or

  build=RELEASE

R=djsollen@google.com, bsalomon@google.com

Author: jcgregorio@google.com

Review URL: https://codereview.chromium.org/83863002

git-svn-id: http://skia.googlecode.com/svn/trunk@12465 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-03 18:16:48 +00:00
bungeman@google.com
0d9e3da8bb Use lowercase windows.h in includes to fix Windows cross compilation using mingw. 
While it doesn't matter on Windows, mingw on case-sensitive OSes uses all lower case filenames for platform include files. I found the problem in SkCondVar.h from Mozilla checkout of skia sources, but the patch contains a fix for the whole skia tree.

R=bungeman@google.com

Review URL: https://codereview.chromium.org/99173003

git-svn-id: http://skia.googlecode.com/svn/trunk@12461 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-03 15:23:37 +00:00
mtklein@google.com
4178ec0582 Fix off-by-one in four_byte_interp benches. 
BUG=

Review URL: https://codereview.chromium.org/100563002

git-svn-id: http://skia.googlecode.com/svn/trunk@12451 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-03 00:01:03 +00:00
commit-bot@chromium.org
c25d2219cb Add four_byte_interp benches. 
Desktop:
running bench [640 480] four_byte_interp_slow_256    NONRENDERING: cmsecs =      6.18
running bench [640 480] four_byte_interp_slow_255    NONRENDERING: cmsecs =      6.83
running bench [640 480] four_byte_interp_fast_256    NONRENDERING: cmsecs =      5.02
running bench [640 480] four_byte_interp_fast_255    NONRENDERING: cmsecs =      5.88

N5:
running bench [640 480] four_byte_interp_slow_256    NONRENDERING: cmsecs =     22.84
running bench [640 480] four_byte_interp_slow_255    NONRENDERING: cmsecs =     25.11
running bench [640 480] four_byte_interp_fast_256    NONRENDERING: cmsecs =     18.89
running bench [640 480] four_byte_interp_fast_255    NONRENDERING: cmsecs =     22.32

BUG=
R=reed@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/93933003

git-svn-id: http://skia.googlecode.com/svn/trunk@12444 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-12-02 22:32:47 +00:00
commit-bot@chromium.org
4b413c8bb1 remove SkFloatToScalar macro 
BUG=
R=reed@google.com, djsollen@google.com

Author: caryclark@google.com

Review URL: https://codereview.chromium.org/85463005

git-svn-id: http://skia.googlecode.com/svn/trunk@12385 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-25 19:44:07 +00:00
robertphillips@google.com
a4662865e3 More Windows 64b compilation warning fixes 
https://codereview.chromium.org/47513017/



git-svn-id: http://skia.googlecode.com/svn/trunk@12337 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-21 14:24:16 +00:00
commit-bot@chromium.org
644629c1c7 Implement a benchmark for GrResourceCache 
Adds "grresourcecache_add" and "grresourcecache_find" bench tests to test
GrResourceCache::add and GrResourceCache::find. The tests work only
with GPU backends, since GrResourceCache needs an GrGpu.

Modifies bench tests to override SkBenchmark::isSuitableFor(Backend)
function that specifies what kind of backend the test is inteded
for. This replaces the previous "fIsRendering" flag that would
indicate test that did no rendering.

Adds SkCanvas::getGrContext() call to get the GrContext that the
canvas ends up drawing to. The member function solves a common
use-case that is also used in the benchmark added here.

R=mtklein@google.com, bsalomon@google.com

Author: kkinnunen@nvidia.com

Review URL: https://codereview.chromium.org/73643005

git-svn-id: http://skia.googlecode.com/svn/trunk@12334 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-21 06:21:58 +00:00
robertphillips@google.com
8c99c9f4a6 Reverting r12315 (More Windows 64b compilation warning fixes) due to compilation failures 
git-svn-id: http://skia.googlecode.com/svn/trunk@12316 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-20 15:56:14 +00:00
robertphillips@google.com
80051d38a3 More Windows 64b compilation warning fixes 
https://codereview.chromium.org/47513017/



git-svn-id: http://skia.googlecode.com/svn/trunk@12315 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-20 15:46:10 +00:00
djsollen@google.com
db490e9971 fix bench so that if SK_SUPPORTS_GPU even when not provided as compiler option 
In the case that SK_SUPPORTS_GPU is not provided to the compiler the value is
eithe defined in in the SkUserConfig.h or SkPostConfig.h.  Prior to this change
those headers were not read prior to trying to include the GPU headers which
resulted in a failed compile.

R=bsalomon@google.com

Review URL: https://codereview.chromium.org/76213004

git-svn-id: http://skia.googlecode.com/svn/trunk@12309 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-20 13:15:40 +00:00
commit-bot@chromium.org
a1a097ee81 increase coverage of SkPath.cpp, remove unused code 
Using Mike Klein's excellent coverage tool, increase the
unit testing of SkPath.cpp from 70% to 95%.

Along the way, determined that these functions were not
maintained or used:

SkPath::pathTo
SkPath::contains

as well as a large block of SkPath::cheapGetDirection().

Changed SkPath::validate() to permit infinities in
the path data points.

Fixed errors in preserving direction.
Fixed error setting direction when convexity is unknown.

Added missing conic to moveTo only detector.

BUG=
R=bsalomon@google.com, reed@google.com

Author: caryclark@google.com

Review URL: https://codereview.chromium.org/65493004

git-svn-id: http://skia.googlecode.com/svn/trunk@12291 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-14 16:53:22 +00:00
jvanverth@google.com
21ca4339a6 Rebaseline bench expectations for tabl_theverge.skp_tile_256x256_grid. 
git-svn-id: http://skia.googlecode.com/svn/trunk@12257 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-12 20:58:41 +00:00
scroggo@google.com
f4dcea6020 Make bench_compare executable. 
(SkipBuildbotRuns)

Review URL: https://codereview.chromium.org/63153005

git-svn-id: http://skia.googlecode.com/svn/trunk@12243 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-12 15:25:31 +00:00
jvanverth@google.com
62454a6603 Update bench expectations. 
git-svn-id: http://skia.googlecode.com/svn/trunk@12213 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-11 14:36:54 +00:00
robertphillips@google.com
6c1e49af8d Removed deprecated API calls that have crept in 
git-svn-id: http://skia.googlecode.com/svn/trunk@12210 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-10 15:08:45 +00:00
senorblanco@chromium.org
27eec46d69 SSE2 implementation of RGBA box blurs. This yields ~2X perf improvement on Xeon ES-2690. 
R=mtklein@google.com

Review URL: https://codereview.chromium.org/61643011

git-svn-id: http://skia.googlecode.com/svn/trunk@12204 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-08 20:49:04 +00:00
commit-bot@chromium.org
11e5b972a9 Add sk_float_rsqrt with SSE + NEON fast paths. 
Current numbers:

N4:
running bench [640 480]      math_fastIsqrt    NONRENDERING: cmsecs =      3.12
running bench [640 480]      math_slowIsqrt    NONRENDERING: cmsecs =      4.82
running bench [640 480] math_sk_float_rsqrt    NONRENDERING: cmsecs =      1.99

Desktop:
running bench [640 480]      math_fastIsqrt    NONRENDERING: cmsecs =      0.89
running bench [640 480]      math_slowIsqrt    NONRENDERING: cmsecs =      0.94
running bench [640 480] math_sk_float_rsqrt    NONRENDERING: cmsecs =      0.09

Haven't found any other benches where this is a significant effect yet.

BUG=
R=reed@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/60083014

git-svn-id: http://skia.googlecode.com/svn/trunk@12203 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-08 20:14:16 +00:00
skia.committer@gmail.com
ab7442c8d7 Sanitizing source files in Housekeeper-Nightly 
git-svn-id: http://skia.googlecode.com/svn/trunk@12186 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-08 07:01:56 +00:00
reed@google.com
126f7f5244 Revert "Revert "speed up A8 by creating a new entry-point in SkDraw that blits the path's coverage directly into an A8 target, regardless of the previous pixel values."" 
This reverts commit 3c77887b3eb2d32ab668ab4e5f2f9e79103956e8.

BUG=

Review URL: https://codereview.chromium.org/50673005

git-svn-id: http://skia.googlecode.com/svn/trunk@12167 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-07 16:06:53 +00:00
scroggo@google.com
8610d2cdbd Fixes for blurroundrect gm/bench. 
Simplify naming/various cases.
Have an atlas of gms, rather than several different images.
In the bench, pull non rendering out of the loop.
Use meaningful enums instead of integers where appropriate.
Add comments.

Addresses comments in https://codereview.chromium.org/52793005/

R=robertphillips@google.com

Review URL: https://codereview.chromium.org/59983004

git-svn-id: http://skia.googlecode.com/svn/trunk@12144 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-05 21:10:58 +00:00
scroggo@google.com
ec1a96c317 Another int to scalar fix. 
git-svn-id: http://skia.googlecode.com/svn/trunk@12138 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-05 17:27:01 +00:00
scroggo@google.com
99e113beb2 And fix the similar bench. 
git-svn-id: http://skia.googlecode.com/svn/trunk@12136 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-05 16:40:15 +00:00
scroggo@google.com
7b05659072 Add gms and benchmarks for drawing blurry round rects. 
Further changes (https://codereview.chromium.org/48623006) will change
the speed at which the bench draws and the drawing of the gm (the gm
change is small).

One of these round rects causes slow drawing in a webpage that we have
observed.

BUG=https://b.corp.google.com/issue?id=11174385

Review URL: https://codereview.chromium.org/52793005

git-svn-id: http://skia.googlecode.com/svn/trunk@12133 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-05 15:57:21 +00:00
reed@google.com
84e922bfb3 Revert "speed up A8 by creating a new entry-point in SkDraw that blits the path's coverage directly into an A8 target, regardless of the previous pixel values." 
This reverts commit d615e839b71f75df895de6850b774a9e1c28ad2a.

Revert "must initialize SkDraw.fClip"

This reverts commit 108e46d29b5f57927fc8b8c403bb52019d8cb16d.

BUG=

Review URL: https://codereview.chromium.org/57883006

git-svn-id: http://skia.googlecode.com/svn/trunk@12122 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-04 20:57:36 +00:00
reed@google.com
6bf614fc23 must initialize SkDraw.fClip 
git-svn-id: http://skia.googlecode.com/svn/trunk@12120 2bbb7eff-a529-9590-31e7-b0007b416f81
 2013-11-04 20:31:40 +00:00