Commit Graph

1269 Commits

Author SHA1 Message Date
mtklein
56da025020 Switch uses of SkChecksum::Compute to Murmur3.
SkChecksum::Compute is a very, very poorly distributed hash function.
This replaces all remaining uses with Murmur3.

The only interesting stuff is in src/gpu.

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/1d024a3c909ae5cefa5e8b339e2b52dc73ee85ac

Committed: https://skia.googlesource.com/skia/+/540e95483d285b555e9b1a73d18c16e7d7c0deba

Review URL: https://codereview.chromium.org/1436973003
2015-11-16 11:16:23 -08:00
mtklein
3c2d32b8e2 Revert of Switch uses of SkChecksum::Compute to Murmur3. (patchset #2 id:20001 of https://codereview.chromium.org/1436973003/ )
Reason for revert:
gotta put back *compute = 0.

Original issue's description:
> Switch uses of SkChecksum::Compute to Murmur3.
>
> SkChecksum::Compute is a very, very poorly distributed hash function.
> This replaces all remaining uses with Murmur3.
>
> The only interesting stuff is in src/gpu.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/1d024a3c909ae5cefa5e8b339e2b52dc73ee85ac
>
> Committed: https://skia.googlesource.com/skia/+/540e95483d285b555e9b1a73d18c16e7d7c0deba

TBR=bsalomon@google.com,mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Review URL: https://codereview.chromium.org/1448023005
2015-11-16 11:01:18 -08:00
mtklein
540e95483d Switch uses of SkChecksum::Compute to Murmur3.
SkChecksum::Compute is a very, very poorly distributed hash function.
This replaces all remaining uses with Murmur3.

The only interesting stuff is in src/gpu.

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/1d024a3c909ae5cefa5e8b339e2b52dc73ee85ac

Review URL: https://codereview.chromium.org/1436973003
2015-11-16 09:08:21 -08:00
msarett
4b0778ec49 Remove dependency on src/android from dm and nanobench
BUG=skia:

Review URL: https://codereview.chromium.org/1443033002
2015-11-13 09:59:11 -08:00
mtklein
0274efbd3d Revert of Switch uses of SkChecksum::Compute to Murmur3. (patchset #2 id:20001 of https://codereview.chromium.org/1436973003/ )
Reason for revert:
I would not have expected this to change any images.

Looks like it's changed both MSAA and non-MSAA GMs:
https://gold.skia.org/search2?blame=1d024a3c909ae5cefa5e8b339e2b52dc73ee85ac&unt=true&head=true&query=source_type%3Dgm

Original issue's description:
> Switch uses of SkChecksum::Compute to Murmur3.
>
> SkChecksum::Compute is a very, very poorly distributed hash function.
> This replaces all remaining uses with Murmur3.
>
> The only interesting stuff is in src/gpu.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/1d024a3c909ae5cefa5e8b339e2b52dc73ee85ac

TBR=bsalomon@google.com,mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Review URL: https://codereview.chromium.org/1445523003
2015-11-13 07:07:47 -08:00
mtklein
1d024a3c90 Switch uses of SkChecksum::Compute to Murmur3.
SkChecksum::Compute is a very, very poorly distributed hash function.
This replaces all remaining uses with Murmur3.

The only interesting stuff is in src/gpu.

BUG=skia:

Review URL: https://codereview.chromium.org/1436973003
2015-11-12 15:44:09 -08:00
msarett
84a8065275 Reorganize BRD code in new tools directory
Rename SkCodecTools.h to SkBitmapRegionDecoderPriv.h

Move BRD code to its own directory in tools.  This
allows us to not need to expose the entire tools
directory in Android.

BUG=skia:

Review URL: https://codereview.chromium.org/1417393004
2015-11-10 15:49:46 -08:00
msarett
0459c9425f Delete dead subset benches from nanobench
This approach to subset decoding is no longer supported.
We have replaced it with an implementation that does not
depend on forked libraries.
https://codereview.chromium.org/1406153015/

BUG=skia:

Review URL: https://codereview.chromium.org/1430493005
2015-11-10 14:52:13 -08:00
halcanary
6950de6c41 Comments Style: s/skbug.com/bug.skia.org/
DOCS_PREVIEW= https://skia.org/?cl=1432503003

Review URL: https://codereview.chromium.org/1432503003
2015-11-07 05:29:00 -08:00
msarett
5cb4885b4c Rename SkBitmapRegionDecoder and Create function
We no longer need to worry about namespace
conflicts SkBitmapRegionDecoder in Android (which
we are replacing).

Additionally, the static Create() function does not
need to repeat the name BitmapRegionDecoder.

BUG=skia:

Review URL: https://codereview.chromium.org/1415243007
2015-11-06 08:56:32 -08:00
scroggo
501b7344f1 Combine native sampling with sampling
In SkSampledCodec, allow the native codec to do its scaling first, then
sample on top of that. Since the only codec which can do native scaling
is JPEG, and we know what it can do, hard-code for JPEG. Check to see
if the sampleSize is something JPEG supports, or a multiple of
something it supports. If so, use JPEG directly or combine them.

BUG=skia:4320

Review URL: https://codereview.chromium.org/1417583009
2015-11-03 07:55:12 -08:00
senorblanco
a8bd38e178 Add cropped-then-expanded test cases to blur_image_filter tests.
These are benches similar to the imagefilterscropexpand GM: an
input filter is cropped to a smaller size, then the blur is re-expanded
out to a larger size.

BUG=skbug:4502

Review URL: https://codereview.chromium.org/1412373004
2015-10-30 13:17:20 -07:00
mtklein
bbba16878f nanobench: fix and simplify --samplingTime
Recent changes to WallTimer broke --samplingTime.  In particular, this idiom became nonsensical:
   WallTimer timer;
   timer.start();
   do {
     ...
     timer.end();
   } while(timer.fWall < ...);

WallTimer started making private use of fWall between when start() and end() were called, so the second time around the loop we end up with nonsense.

If that makes no sense, don't worry.  The code here using now_ms() is just as fast, just as precise, and clearer.

I took the opportunity to simplify --samplingTime <complicated string parsing> to --ms <int>, and to simplify the code that depends on it.

BUG=skia:

Review URL: https://codereview.chromium.org/1419103004
2015-10-28 11:36:30 -07:00
msarett
35e5d1b449 Refactor SkBitmapRegionDecoderInterface for Android
The result SkBitmap, the pixel allocator, and the alpha
preference need to be communicated from the client to
the region decoder.

BUG=skia:

Review URL: https://codereview.chromium.org/1418093006
2015-10-27 12:50:25 -07:00
cdalton
e6d2024c68 Flush GrContext between benchmark draw loops
This change updates a small subset of benchmarks to flush the GrContext
between draw loops (specifically SKP benchmarks, SampleApp, and the
warmup in visualbench). This helps improve timing accuracy by not
allowing the gpu to batch across draw boundaries in the affected
benchmarks.

BUG=skia:

Review URL: https://codereview.chromium.org/1427533002
2015-10-26 13:45:29 -07:00
egdaniel
cb7ba1eafc Make appending default precision be controled by GLSL
BUG=skia:

Review URL: https://codereview.chromium.org/1403373012
2015-10-26 08:38:26 -07:00
tomhudson
3c8ceb7350 Simplify linkages to Android framework internals
We've migrated SkHwuiRenderer into the Android Framework as
android::uirenderer::TestWindowContext in response to an internal
bug; we now delete that class and change our build references here.

R=djsollen@google.com

Review URL: https://codereview.chromium.org/1407053009
2015-10-26 07:21:32 -07:00
senorblanco
7b87ee7a1c Image filters: simplify filterInputGPU().
Remove a call to canFilterImageGPU() / filterImageGPU() from
filterInputGPU(). There's no reason to do this, since
the subsequent filterImage() call will do it for us anyway.
And this call actually defeats caching (as demonstrated by
the attached bench).

BUG=skia:

Review URL: https://codereview.chromium.org/1411013004
2015-10-26 06:55:47 -07:00
egdaniel
472d44e8f8 Add version string and force highp NDS transfrom to GLSLCaps
This also include the use of any() in the shaders.

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/cef4bce8e260b49bf3417eadbac806cf7d39cdc8

Review URL: https://codereview.chromium.org/1417993004
2015-10-22 08:20:00 -07:00
msarett
26ad17b8f8 Implementation of SkBitmapRegionDecoder using SkAndroidCodec
Includes testing in DM and nanobench

BUG=skia:

Review URL: https://codereview.chromium.org/1402863002
2015-10-22 07:29:19 -07:00
egdaniel
899ba37db3 Revert of Add version string and force highp NDS transfrom to GLSLCaps (patchset #3 id:40001 of https://codereview.chromium.org/1417993004/ )
Reason for revert:
breaking bots

Original issue's description:
> Add version string and force highp NDS transfrom to GLSLCaps
>
> This also include the use of any() in the shaders.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/cef4bce8e260b49bf3417eadbac806cf7d39cdc8

TBR=bsalomon@google.com,jvanverth@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Review URL: https://codereview.chromium.org/1420033002
2015-10-22 07:08:37 -07:00
egdaniel
cef4bce8e2 Add version string and force highp NDS transfrom to GLSLCaps
This also include the use of any() in the shaders.

BUG=skia:

Review URL: https://codereview.chromium.org/1417993004
2015-10-22 06:54:15 -07:00
egdaniel
0d3f061262 Move GrGLShaderVar to GrGLSL
BUG=skia:

Review URL: https://codereview.chromium.org/1417123002
2015-10-21 10:45:48 -07:00
egdaniel
f529439fea Move shader precision modifier check onto GLSLCaps
BUG=skia:

Review URL: https://codereview.chromium.org/1414373002
2015-10-21 07:14:17 -07:00
msarett
74deb981d5 Valgrind experiment
I don't really expect this to fix the errors, but I think
it's worth it to try shaking up the valgrind bot overnight.
There's some strange behavior with regard to color type on
the valgrind bot that I can't reproduce and that we aren't
seeing on any of the other bots.

TBR=mtklein,scroggo

BUG=skia:

Review URL: https://codereview.chromium.org/1418723002
2015-10-20 16:45:56 -07:00
mtklein
4dfdbb19ba nanobench: Add some Valgrind failure paranoia.
Trying to figure out

http://build.chromium.org/p/client.skia/builders/Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-Valgrind/builds/586/steps/nanobench/logs/stdio

BUG=skia:4482

Review URL: https://codereview.chromium.org/1418633002
2015-10-20 07:45:29 -07:00
robertphillips
423f6461e9 Add SkTTopoSort
BUG=skia:4094

Review URL: https://codereview.chromium.org/1414503003
2015-10-19 12:15:55 -07:00
bsalomon
2f8ac3538d Misc: Don't run etc1 bench with invalid image and fix VS2015 build warning
Review URL: https://codereview.chromium.org/1409993003
2015-10-19 08:29:16 -07:00
scroggo
860e8a6719 Renable image benchmarking
- Remove --images '' to renable image benchmarking
- Add a flag to disable testing JPEG's buildTileIndex, since it also leaks memory
- Do not run images on GPU
- Do not run large interlaced images on 32 bit bots
- When buildTileIndex is not being used in the subset benches, do not use it for BRD

BUG=skia:3418
BUG=skia:4469
BUG=skia:4471
BUG=skia:4360

Review URL: https://codereview.chromium.org/1396113002
2015-10-15 07:51:28 -07:00
scroggo
fd2b067b9e Remove unnecessary allocation in nanobench
Review URL: https://codereview.chromium.org/1408593002
2015-10-14 09:00:36 -07:00
scroggo
2da1a854b0 Stop benching PNG buildTileIndex by default
It leaks memory and kills the bots. Supply an option to run it for
local comparison.

BUG=skia:4360
BUG=skia:3418

Review URL: https://codereview.chromium.org/1396613007
2015-10-13 13:33:33 -07:00
msarett
fdb47571a3 Add subsetting to SkScanlineDecoder
This CL allows the SkScanlineDecoder to decode partial
scanlines.

This is a first step in efficiently implementing subsetting
in SkScaledCodec.

BUG=skia:4209

Review URL: https://codereview.chromium.org/1390213002
2015-10-13 12:50:14 -07:00
scroggo
e361781bf7 Fix codec memory leaks in nanobench
SubsetTranslateBench.cpp:
Unref the color table, so it gets deleted.

SkBitmapRegionDecoderInterface.cpp:
Delete the stream if it is not used.

BUG=skia:3418

Review URL: https://codereview.chromium.org/1396113003
2015-10-09 12:15:57 -07:00
msarett
e6dd004c1b Fill incomplete images in SkCodec parent class
Rather than implementing some sort of "fill" in every
SkCodec subclass for incomplete images, let's make the
parent class handle this situation.

This includes an API change to SkCodec.h

SkCodec::getScanlines() now returns the number of lines it
read successfully, rather than an SkCodec::Result enum.
getScanlines() most often fails on an incomplete input, in
which case it is useful to know how many lines were
successfully decoded - this provides more information than
kIncomplete vs kSuccess.  We do lose information when the
API is used improperly, as we are no longer able to return
kInvalidParameter or kScanlineNotStarted.

Known Issues:
Does not work for incomplete fFrameIsSubset gifs.
Does not work for incomplete icos.

BUG=skia:

Review URL: https://codereview.chromium.org/1332053002
2015-10-09 11:07:34 -07:00
scroggo
ab12c27158 Update Subset benches to support interlacing and fix bugs
Instead of decoding one line at a time, if the ScanlineOrder is kNone,
decode all of the lines in one pass, and then copy the subset into the
output. This will allow us to more realistically test subset decodes
for interlaced png. It also makes running them not take forever.

Do *not* support other modes (besides kTopDown), since they are not
used by the big three we need to replace BitmapRegionDecoder
implementation (skbug.com/4428).

Fix a bug in SubsetTranslateBench and SubsetZoomBench:
When we decode another subset, we need to reset the scanline decode
first. This bug appears to have been present since the introduction of
these tests in crrev.com/1160953002

BUG=skia:4205
BUG=skia:3418

Review URL: https://codereview.chromium.org/1387233002
2015-10-08 12:01:39 -07:00
reed
5ea95df02d Revert "Revert of factories should return baseclass, allowing the impl to specialize (patchset #4 id:60001 of https://codereview.chromium.org/1390523005/ )"
This reverts commit 95376a0dde.

BUG=skia:
TBR=

Review URL: https://codereview.chromium.org/1389083002
2015-10-06 14:05:32 -07:00
schenney
95376a0dde Revert of factories should return baseclass, allowing the impl to specialize (patchset #4 id:60001 of https://codereview.chromium.org/1390523005/ )
Reason for revert:
Breaks Chrome with this link error: ../../third_party/skia/include/effects/SkMorphologyImageFilter.h:75: error: undefined reference to 'SkMorphologyImageFilter::SkMorphologyImageFilter(int, int, SkImageFilter*, SkImageFilter::CropRect const*)'
../../third_party/skia/include/effects/SkMorphologyImageFilter.h:104: error: undefined reference to 'SkMorphologyImageFilter::SkMorphologyImageFilter(int, int, SkImageFilter*, SkImageFilter::CropRect const*)'

Presumably due to code in third_party/WebKit/Source/platform/graphics/filters/FEMorphology.cpp that contains:
#include "SkMorphologyImageFilter.h"

...

    if (m_type == FEMORPHOLOGY_OPERATOR_DILATE)
        return adoptRef(SkDilateImageFilter::Create(radiusX, radiusY, input.get(), &rect));
    return adoptRef(SkErodeImageFilter::Create(radiusX, radiusY, input.get(), &rect));

Original issue's description:
> factories should return baseclass, allowing the impl to specialize
>
> waiting on https://codereview.chromium.org/1386163002/# to land
>
> BUG=skia:4424
>
> Committed: https://skia.googlesource.com/skia/+/80a6dcaa1b757826ed7414f64b035d512d9ccbf8

TBR=senorblanco@google.com,robertphillips@google.com,reed@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:4424

Review URL: https://codereview.chromium.org/1389063002
2015-10-06 12:59:55 -07:00
reed
80a6dcaa1b factories should return baseclass, allowing the impl to specialize
waiting on https://codereview.chromium.org/1386163002/# to land

BUG=skia:4424

Review URL: https://codereview.chromium.org/1390523005
2015-10-06 11:12:23 -07:00
joshualitt
cb54e8ed45 Cleanup timing state machine
BUG=skia:

Review URL: https://codereview.chromium.org/1386933002
2015-10-05 13:58:26 -07:00
scroggo
303fa35012 Specify bench_type & source_type for image benches
This will allow us to use perf filtering for comparing SkImageDecoder
to SkCodec.

BUG=skia:3418

Review URL: https://codereview.chromium.org/1387863002
2015-10-05 11:03:34 -07:00
joshualitt
98d2e2f095 Factor out VisualBench timing code into a helper class
BUG=skia:

Review URL: https://codereview.chromium.org/1375363003
2015-10-05 07:23:30 -07:00
joshualitt
b235168f55 S4 only has 8 texture units on the GLContext we create
BUG=skia:4373

Review URL: https://codereview.chromium.org/1386643002
2015-10-02 06:49:23 -07:00
mtklein
a1ebeb25e9 Remove const from const int loops.
This drives me nuts, and prevents `while (loops --> 0)`.

BUG=skia:

Review URL: https://codereview.chromium.org/1379923005
2015-10-01 09:43:39 -07:00
joshualitt
8a6697af95 Fix for nexus 5 crashing in GL benches
GLBenches do not expect gl state  to change between onPerCanvasPreDraw and *PostDraw, but we do a clear and sometimes we clear as draw.  This causes us to bind vertex objects / programs / etc.

This change creates two new virtual methods which are called right before and immediately after timing.

BUG=skia:

Review URL: https://codereview.chromium.org/1379853003
2015-09-30 12:11:07 -07:00
scroggo
46c5747256 Merge SkCodec with SkScanlineDecoder
Benefits:
- This mimics other decoding APIs (including the ones SkCodec relies
on, e.g. a png_struct, which can be used to decode an entire image or
one line at a time).

- It allows a client to ask us to do what we can do efficiently - i.e.
start from encoded data and either decode the whole thing or scanlines.

- It removes the duplicate methods which appeared in both SkCodec and
SkScanlineDecoder (some of which, e.g. in SkJpegScanlineDecoder, just
call fCodec->sameMethod()).

- It simplifies moving more checks into the base class (e.g. the
examples in skbug.com/4284).

BUG=skia:4175
BUG=skia:4284

=====================================================================

SkScanlineDecoder.h/.cpp:
Removed.

SkCodec.h/.cpp:
Add methods, enums, and variables which were previously in
SkScanlineDecoder.
Default fCurrScanline to -1, as a sentinel that start has not been
called.

General changes:
Convert SkScanlineDecoders to SkCodecs.

General changes in SkCodec subclasses:
Merge SkScanlineDecoder implementation into SkCodec. Most (all?) owned
an SkCodec, so they now call this-> instead of fCodec->.

SkBmpCodec.h/.cpp:
Replace the unused rowOrder method with an override for
onGetScanlineOrder.
Make getDstRow const, since it is called by onGetY, which is const.

SkCodec_libpng.h/.cpp:
Make SkPngCodec an abstract class, with two subclasses which handle
scanline decoding separately (they share code for decoding the entire
image). Reimplement onReallyHasAlpha so that it can return the most
recent result (e.g. after a scanline decode which only decoded part
of the image) or a better answer (e.g. if the whole image is known to
be opaque).
Compute fNumberPasses early, so we know which subclass to instantiate.
Make SkPngInterlaceScanlineDecoder use the base class' fCurrScanline
rather than a separate variable.

CodexTest.cpp:
Add tests for the state changes in SkCodec (need to call start before
decoding scanlines; calling getPixels means that start will need to
be called again before decoding more scanlines).
Add a test which decodes in stripes, currently only used for an
interlaced PNG.

TODO: Add tests for onReallyHasAlpha.

Review URL: https://codereview.chromium.org/1365313002
2015-09-30 08:57:14 -07:00
herb
62a69c26b3 Move Mutexy things to private.
There is no API change.

TBR=reed@google.com

BUG=skia:

Review URL: https://codereview.chromium.org/1368333004
2015-09-29 11:47:45 -07:00
herb
e6e41a8a19 Move SkAtomics.h to private.
There are no API changes.

TBR=reed@google.com

BUG=skia:

Review URL: https://codereview.chromium.org/1369333004
2015-09-28 11:24:13 -07:00
fmalita
2f5891ea64 Remove SkBitmapSource
To avoid breaking existing SKPs, add a deserialization stub which
unflattens SkBitmapSource records to SkImageSources.

R=reed@google.com,mtklein@google.com,robertphillips@google.com

Review URL: https://codereview.chromium.org/1363913002
2015-09-25 09:15:55 -07:00
reed
57b767127f use override to make the new xcode happier
BUG=skia:
TBR=

Review URL: https://codereview.chromium.org/1368673004
2015-09-25 04:49:43 -07:00
joshualitt
8db65a6d0c stop scaling loops and improve warmup bench
BUG=skia:

Review URL: https://codereview.chromium.org/1363663004
2015-09-24 12:42:02 -07:00