Commit Graph

20594 Commits

Author SHA1 Message Date
ericrk
369e9375a3 Add Histogram Macros to Skia
Adds a set of histogram macros to Skia, modeled after Chrome's
UMA_HISTOGRAM_* macros. These allow logging of high frequency events,
and are useful to analyze real world usage of certain features.

By default, these macros are no-ops. Users can provide a custom
header file which defines these macros if they wish to collect
histogram data. Chrome will provide such a header.

I've currently only added two macros:
- SK_HISTOGRAM_BOOLEAN - logs a true/false type relationship (whether
we are tiling a texture or not on each draw).
- SK_HISTOGRAM_ENUMERATION - logs a set of potential values (which of
a number of choices were selected for the texture upload path).

We could add more unused macros at the moment, but it seems easier to
add these as needed, WDYT?

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1652053004

Review URL: https://codereview.chromium.org/1652053004
2016-02-05 15:32:36 -08:00
msarett
60dcd3cb85 SkPngCodec clean-ups
Use png_read_row() instead of png_read_rows(1).
All png_read_rows() does is call png_read_row() in
a loop.  This comes from Leon's comment in:
https://codereview.chromium.org/1671003004/

Also there is a bit of refactoring.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1675843002

Review URL: https://codereview.chromium.org/1675843002
2016-02-05 15:13:12 -08:00
caryclark
52b64619d9 increase subdivision for very large cubics
Allow 512 subdivisions for hairline cubics instead of 64
to allow very large cubics to be properly approximated with
lines.

R=reed@google.com
BUG=339297
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1674813002

Review URL: https://codereview.chromium.org/1674813002
2016-02-05 13:59:31 -08:00
bsalomon
2fba809120 Fix memory leaks in SurfaceClear_Gpu test
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1672083002

Review URL: https://codereview.chromium.org/1672083002
2016-02-05 13:47:06 -08:00
erikchen
7fec91ce66 skia: Add support for CHROMIUM_image backed textures.
I created a new abstract base class TextureStorageAllocator that consumers of
Skia can subclass and pass back to Skia. When a surface is created with a
pointer to a TextureStorageAllocator, any textures it creates, or that are
derived from the original surface, will allocate and deallocate storage using
the methods on TextureStorageAllocator.

BUG=https://code.google.com/p/chromium/issues/detail?id=579664
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1623653002

Committed: https://skia.googlesource.com/skia/+/92098e691f10a010e7421125ba4d44c02506bb55

Review URL: https://codereview.chromium.org/1623653002
2016-02-05 12:10:55 -08:00
reed
3601f280dc add kRGBA_F16_SkColorType
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1666343002

Review URL: https://codereview.chromium.org/1666343002
2016-02-05 11:18:39 -08:00
cdalton
06604b9562 Add GL indirect drawing APIs
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1666803002

Review URL: https://codereview.chromium.org/1666803002
2016-02-05 10:09:51 -08:00
cdalton
af8bc7d7a4 Add support for GL_EXT_raster_multisample
Updates the GrGLGpu to call glRasterSamplesEXT when multisample is
enabled for a mixed sampled render target, but the stencil test is not
active.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1666563003

Review URL: https://codereview.chromium.org/1666563003
2016-02-05 09:35:20 -08:00
ethannicholas
f6a84bdb47 Added JSON support for TextBlob et al.
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1674643002

Review URL: https://codereview.chromium.org/1674643002
2016-02-05 09:04:45 -08:00
yujieqin
9c7a8a4648 Optimize the SkRawStream when the input is an asset stream
BUG=b/26841494
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1645963002

Review URL: https://codereview.chromium.org/1645963002
2016-02-05 08:21:19 -08:00
bsalomon
e63ffef624 Remove deferred clear from SkGpuDevice
Add combining to GrClearBatch

Fix issue with state tracking in GrGLGpu::createTestingOnlyBackendTexture

Add tests for clearing GPU SkSurfaces and add tests for GrDrawContext::clear().

Add comment that SkCanvas::flush will resolve the RT in the GPU case.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1658823002

Review URL: https://codereview.chromium.org/1658823002
2016-02-05 07:17:34 -08:00
robertphillips
d3b32bf832 Last bit of GrGLSLFragmentProcessor-derived class cleanup
I believe these are the last two that used the dispreferred constructor pattern.

GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1667953005

Review URL: https://codereview.chromium.org/1667953005
2016-02-05 07:15:39 -08:00
benjaminwagner
f49c75a8f1 Move Google3-specific stack limitation logic to template classes.
Remove #ifdefs in other files.

Reapplies https://codereview.chromium.org/1656143003; removing the implicit constructors for GLPtr and GLPtrAlias resolves the build issue on Android.

Also reverts https://codereview.chromium.org/1663013004

Does not change the public API.

TBR=reed
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1666203002

Review URL: https://codereview.chromium.org/1666203002
2016-02-05 07:02:38 -08:00
mtklein
001cc1f41b dont upload to Gold from MSAN bot
NOTRY=true
NOTREECHECKS=true
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1673643002

Review URL: https://codereview.chromium.org/1673643002
2016-02-05 06:16:59 -08:00
mtklein
3f20f511dc update CQ with new bot names
NOTRY=true
NOTREECHECKS=true

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1675643002

Review URL: https://codereview.chromium.org/1675643002
2016-02-05 06:10:38 -08:00
jcgregorio
e7d7f90b59 Revert of skia: Add support for CHROMIUM_image backed textures. (patchset #14 id:260001 of https://codereview.chromium.org/1623653002/ )
Reason for revert:
Seems to causing the DEPS roll to fail:

   https://codereview.chromium.org/1668293002/

Original issue's description:
> skia: Add support for CHROMIUM_image backed textures.
>
> I created a new abstract base class TextureStorageAllocator that consumers of
> Skia can subclass and pass back to Skia. When a surface is created with a
> pointer to a TextureStorageAllocator, any textures it creates, or that are
> derived from the original surface, will allocate and deallocate storage using
> the methods on TextureStorageAllocator.
>
> BUG=https://code.google.com/p/chromium/issues/detail?id=579664
> GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1623653002
>
> Committed: https://skia.googlesource.com/skia/+/92098e691f10a010e7421125ba4d44c02506bb55

TBR=bsalomon@chromium.org,cblume@chromium.org,bsalomon@google.com,robertphillips@google.com,egdaniel@google.com,reed@google.com,erikchen@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=https://code.google.com/p/chromium/issues/detail?id=579664

Review URL: https://codereview.chromium.org/1669213002
2016-02-04 19:51:25 -08:00
msarett
a87d6de6a5 Set sRGB flag for PNGs with an sRGB chunk
BUG=skia:3471
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1667823004

Review URL: https://codereview.chromium.org/1667823004
2016-02-04 15:37:58 -08:00
robertphillips
16aad78602 Fix memory leak in SkSpecialSurface
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1666373002

TBR=jcgregorio@google.com
NOTRY=true

Review URL: https://codereview.chromium.org/1666373002
2016-02-04 14:59:58 -08:00
fmalita
3a94c6c62c Add SkAutoPixmapStorage::detachPixelsAsData()
Allows passing pixels ownership to clients.

BUG=skia:4896
R=reed@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1662353002

Review URL: https://codereview.chromium.org/1662353002
2016-02-04 13:09:59 -08:00
erikchen
92098e691f skia: Add support for CHROMIUM_image backed textures.
I created a new abstract base class TextureStorageAllocator that consumers of
Skia can subclass and pass back to Skia. When a surface is created with a
pointer to a TextureStorageAllocator, any textures it creates, or that are
derived from the original surface, will allocate and deallocate storage using
the methods on TextureStorageAllocator.

BUG=https://code.google.com/p/chromium/issues/detail?id=579664
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1623653002

Review URL: https://codereview.chromium.org/1623653002
2016-02-04 12:03:08 -08:00
reed
f7cdb06d62 extend compose-colorfilter to 4f
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1664663003

Review URL: https://codereview.chromium.org/1664663003
2016-02-04 11:35:27 -08:00
caryclark
eb10ed4090 speculatively comment out chrome mac perf test crasher
The Chrome Mac perf buildbot is mysteriously crashing.
Maybe it's related to this line of code.

Try taking it out to see if the crash goes away.

R=bsalomon@google.com
BUG=580313
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1663673006

Review URL: https://codereview.chromium.org/1663673006
2016-02-04 11:35:06 -08:00
ethannicholas
299200c24b Fixed references to MakeMatrix function
TBR=joshualitt@chromium.org
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1666213002

Review URL: https://codereview.chromium.org/1666213002
2016-02-04 11:19:40 -08:00
robertphillips
b6c65e9995 Add SkSpecialImage & SkSpecialSurface classes
Initial classes.

GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1579323002

Review URL: https://codereview.chromium.org/1579323002
2016-02-04 10:52:42 -08:00
ethannicholas
7471c780d4 JSON API cleanups
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1664263002

Review URL: https://codereview.chromium.org/1664263002
2016-02-04 10:37:50 -08:00
djsollen
6b48984333 Don't print dialog box prior to crashing.
This returns us to our previous behavior when we used 0xbbadbeef
to cause the process to terminate.

BUG=skia:4893
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1667113002

Review URL: https://codereview.chromium.org/1667113002
2016-02-04 08:59:35 -08:00
ethannicholas
0bd103497a Improved support for images/bitmaps in SkJSONCanvas
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1662063003

Review URL: https://codereview.chromium.org/1662063003
2016-02-04 06:45:25 -08:00
scroggo
7b5e5536a1 Add SkAndroidCodec::getPixels
This is a synonym for the version of getAndroidPixels that accepts
only three parameters (i.e. no AndroidOptions). It is very similar
to SkCodec::getPixels, so I think the motivation for naming the
version with options differently does not apply here.

Add comments to the header describing defaults.

Update the test to use a template, and delete a lot of redundant code.

Rename a variable to stop shadowing another variable.

GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1647153002

Review URL: https://codereview.chromium.org/1647153002
2016-02-04 06:14:24 -08:00
robertphillips
bf536af15f Next round of GrGLSLFragmentProcessor-derived class cleanup
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1661143003

Review URL: https://codereview.chromium.org/1661143003
2016-02-04 06:11:53 -08:00
joshualitt
29e5a8932d Modify skia serve to return clip rect and view matrix
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1662003002

Review URL: https://codereview.chromium.org/1662003002
2016-02-04 06:08:33 -08:00
reed
a3d99a515b don't get dismayed by negative scales for HQ
needs https://codereview.chromium.org/1663793003/ to land first

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1668033002

Review URL: https://codereview.chromium.org/1668033002
2016-02-03 19:07:54 -08:00
reed
50b98b424c add note about security and new subclasses
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1667663004

Review URL: https://codereview.chromium.org/1667663004
2016-02-03 16:31:36 -08:00
benjaminwagner
ff172dab92 Revert of Work around exasperating Google3 Android build error. (patchset #1 id:1 of https://codereview.chromium.org/1649493005/ )
Reason for revert:
This workaround no longer seems to be necessary.

Original issue's description:
> Work around exasperating Google3 Android build error.
>
> BUG=skia:
> GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1649493005
>
> Committed: https://skia.googlesource.com/skia/+/900901fb1ba5170549670829fc8f9d9b590ff5f0

TBR=mtklein@google.com,reed@google.com
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=skia:

Review URL: https://codereview.chromium.org/1669663003
2016-02-03 16:29:25 -08:00
msarett
e1daa48b9a Fix Unpremul Index8 tests
TBR=scroggo@google.com

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1663323002

Review URL: https://codereview.chromium.org/1663323002
2016-02-03 15:31:18 -08:00
msarett
095742419d SSE optimizations for GrayAlpha -> RGBA/BGRA Premul/Unpremul
Swizzle Runtime (Dell Venue 8)
Unpremul 0.17x
Premul   0.20x

PNG Decode Runtime on GrayAlpha Encoded PNGs (Dell Venue 8)
Unpremul Regular  0.91x
Unpremul ZeroInit 0.92x
Premul   Regular  0.84x
Premul   ZeroInit 0.85x

BUG=skia:4767
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1666853002
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot

Review URL: https://codereview.chromium.org/1666853002
2016-02-03 15:28:35 -08:00
Brian Salomon
de3a726ad3 Make SkImage::readYUV8Planes const
TBR=reed@google.com

GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1660393003
BUG=skia:

Review URL: https://codereview.chromium.org/1660393003 .
2016-02-03 17:09:00 -05:00
cdalton
e3fafcf7d1 Increase max vertex attrib count in GrPrimitiveProcessor
Increases the max vertex attrib count in GrPrimitiveProcessor from 6 to
8. This is the required minimum by ES 2.0 for GL_MAX_VERTEX_ATTRIBS.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1663513004

Review URL: https://codereview.chromium.org/1663513004
2016-02-03 14:02:50 -08:00
halcanary
72002f28d5 SkMojo: generate on Darwin/x86_64
also clean up generate and skmojo.gyp to work right.

BUG=skia:4891
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1660043004

Review URL: https://codereview.chromium.org/1660043004
2016-02-03 13:54:03 -08:00
cdalton
e62cbdffcf Fix assert accessing pipeline from initBatchTracker
Marks fPipelineInstalled as true before calling initBatchTracker. The
batch subclass ought to be able to access the newly constructed
pipeline object during its init.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1666813002

Review URL: https://codereview.chromium.org/1666813002
2016-02-03 13:29:19 -08:00
msarett
ebf440816b Fix bugs in SkBmpCodec
The decode should not depend on the requested alpha type.

These were exposed by:
https://codereview.chromium.org/1641273003

This should cause the number of untriaged images in Gold
to go to zero.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1663303002

Review URL: https://codereview.chromium.org/1663303002
2016-02-03 13:12:38 -08:00
bungeman
1aeac30809 Revert "Factor out xcode c/cpp settings into variables to apply to cmake builds."
This reverts commit 32235eac4b.
This change should no longer be necessary now that gyp has been updated.

Review URL: https://codereview.chromium.org/1664923002
2016-02-03 13:06:28 -08:00
msarett
9e9444cad8 Test unpremul decodes
This is a workaround until SkCanvas supports drawing kUnpremul sources.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1661913005

Review URL: https://codereview.chromium.org/1661913005
2016-02-03 12:39:11 -08:00
robertphillips
9cdb9920fc Clean up GrGLSLFragmentProcessor-derived classes
In some other patches it was observed that many of the GrGLSLFragmentProcessor-derived classes needlessly pass a parameter to their constructors.

GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1666773002

Review URL: https://codereview.chromium.org/1666773002
2016-02-03 12:25:40 -08:00
benjaminwagner
d32d19a229 Reduce stack usage for Google3.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1663013004

Review URL: https://codereview.chromium.org/1663013004
2016-02-03 12:20:19 -08:00
scroggo
8673714d75 Treat bad values passed to --images as a fatal error
If an option is passed to --images that is either a non-existent path or
a folder with no images matching the supported types, assume this is
an error and exit, so they can supply a valid path instead.

Share code between DM and nanobench in SkCommonFlags.

nanobench now behaves more like DM - it will check a directory for
images that match the supported extensions.

GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1611323004

Review URL: https://codereview.chromium.org/1611323004
2016-02-03 12:19:11 -08:00
halcanary
7a76f9c8f4 SkMojo: test linking Skia against the Mojo SDK
TODO: build on systems other than Linux.

Add mojo_skd to the DEPS.

Add a DM::Via called `mojo-`.

everything is hidden behind the gyp variable `skia_mojo`.
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1644043003

Review URL: https://codereview.chromium.org/1644043003
2016-02-03 11:53:19 -08:00
bungeman
84de5c86f9 Roll third_party/externals/gyp 08429da795..87ac4d0e63fc (63 commits).
08429da795..87ac4d0e63

This is primarily to pick up changes to the CMake generator for Mac.

Review URL: https://codereview.chromium.org/1669503003
2016-02-03 11:25:26 -08:00
mtklein
136baaa927 DM: add uninteresting hashes FYI
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1663223002

Review URL: https://codereview.chromium.org/1663223002
2016-02-03 11:21:45 -08:00
msarett
1e06079b25 NEON optimizations for GrayAlpha -> RGBA/BGRA Premul/Unpremul
PNG Decode Time Nexus 6P (for a test set of GrayAlpha encoded PNGs)
Regular   Unpremul 0.91x
Zero Init Unpremul 0.92x
Regular   Premul   0.84x
Zero Init Premul   0.86x

BUG=skia:4767
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1663623002
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot

Review URL: https://codereview.chromium.org/1663623002
2016-02-03 11:17:44 -08:00
robertphillips
ffad53aa75 Add an avoidXfermode GM to exercise coverage
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1663133003

Review URL: https://codereview.chromium.org/1663133003
2016-02-03 10:46:01 -08:00