Commit Graph

3315 Commits

Author SHA1 Message Date
reed
9ce9d6772d update callsites for Make image factories
not forced yet, as we still have the build-guard. waiting on chrome CL

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1810813003

TBR=

Review URL: https://codereview.chromium.org/1810813003
2016-03-17 10:51:11 -07:00
robertphillips
c5035e70cc Add SkSpecialImage::extractSubset & NewFromPixmap
This is calved off of: https://codereview.chromium.org/1785643003/ (Switch SkBlurImageFilter over to new onFilterImage interface)

This now relies on: https://codereview.chromium.org/1813483002/ (ImagePixelLocker now manually allocates SkPixmap) to clean up the uses of SkAutoPixmapStorage in Chromium

GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1787883002

Committed: https://skia.googlesource.com/skia/+/250581493a0859987e482810879e85e5ac2dc002

Review URL: https://codereview.chromium.org/1787883002
2016-03-17 06:58:39 -07:00
senorblanco
bf680c30a2 Image filters: fix the zero-sigma fast path in SkBlurImageFilter.
The fast path was not correctly taking the input srcOffset into account.

BUG=595324
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1807673005

Review URL: https://codereview.chromium.org/1807673005
2016-03-16 16:15:53 -07:00
mtklein
18300a3aa7 detach -> release
The C++ standard library uses the name "release" for the operation we call "detach".

Rewriting each "detach(" to "release(" brings us a step closer to using standard library types directly (e.g. std::unique_ptr instead of SkAutoTDelete).

This was a fairly blind transformation.  There may have been unintentional conversions in here, but it's probably for the best to have everything uniformly say "release".

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1809733002

Review URL: https://codereview.chromium.org/1809733002
2016-03-16 13:53:35 -07:00
robertphillips
19dea94f1d Revert of Add SkSpecialImage::extractSubset & NewFromPixmap (patchset #5 id:80001 of https://codereview.chromium.org/1787883002/ )
Reason for revert:
Need to wean ImagePixelLocker.h off of SkAutoPixmapStorage :(

Original issue's description:
> Add SkSpecialImage::extractSubset & NewFromPixmap
>
> This is calved off of: https://codereview.chromium.org/1785643003/ (Switch SkBlurImageFilter over to new onFilterImage interface)
>
> GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1787883002
>
> Committed: https://skia.googlesource.com/skia/+/250581493a0859987e482810879e85e5ac2dc002

TBR=bsalomon@google.com,reed@google.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/1808833002
2016-03-16 10:39:09 -07:00
mtklein
5f939ab658 Use std::unique_ptr.
TBR=reed@google.com

Committed: https://skia.googlesource.com/skia/+/20c1e3abfc681771f73eb19fde7284196e028940

Committed: https://skia.googlesource.com/skia/+/3dd9ed37c24611af86f0fe374bd3698b63f09450
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1780933003

Committed: https://skia.googlesource.com/skia/+/2691d76a06e1af6282f8b3a3140cc93361be10c4

Review URL: https://codereview.chromium.org/1780933003
2016-03-16 10:28:35 -07:00
mtklein
e7ec417268 Revert of Use std::unique_ptr. (patchset #10 id:170001 of https://codereview.chromium.org/1780933003/ )
Reason for revert:
aww jeez, I dropped ps 8...

Original issue's description:
> Use std::unique_ptr.
>
> TBR=reed@google.com
>
> Committed: https://skia.googlesource.com/skia/+/20c1e3abfc681771f73eb19fde7284196e028940
>
> Committed: https://skia.googlesource.com/skia/+/3dd9ed37c24611af86f0fe374bd3698b63f09450
> GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1780933003
>
> Committed: https://skia.googlesource.com/skia/+/2691d76a06e1af6282f8b3a3140cc93361be10c4

TBR=bungeman@google.com,mtklein@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/1809693002
2016-03-16 10:03:15 -07:00
robertphillips
250581493a Add SkSpecialImage::extractSubset & NewFromPixmap
This is calved off of: https://codereview.chromium.org/1785643003/ (Switch SkBlurImageFilter over to new onFilterImage interface)

GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1787883002

Review URL: https://codereview.chromium.org/1787883002
2016-03-16 09:47:08 -07:00
mtklein
2691d76a06 Use std::unique_ptr.
TBR=reed@google.com

Committed: https://skia.googlesource.com/skia/+/20c1e3abfc681771f73eb19fde7284196e028940

Committed: https://skia.googlesource.com/skia/+/3dd9ed37c24611af86f0fe374bd3698b63f09450
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1780933003

Review URL: https://codereview.chromium.org/1780933003
2016-03-16 08:39:42 -07:00
reed
1a9b964084 Reland of "more shader-->sp conversions (patchset #5 id:80001 of https://codereview.chromium.org/1789633002/ )"
This reverts commit 9283d20afc.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1793303002

Review URL: https://codereview.chromium.org/1793303002
2016-03-13 14:13:58 -07:00
reed
9283d20afc Revert of more shader-->sp conversions (patchset #5 id:80001 of https://codereview.chromium.org/1789633002/ )
Reason for revert:
seems to have changed pictureshadertile

Original issue's description:
> more shader-->sp conversions
>
> BUG=skia:
> GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1789633002
>
> Committed: https://skia.googlesource.com/skia/+/ce563cdd48ec7ce4e0420dd88760df9be9dba50c

TBR=fmalita@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Review URL: https://codereview.chromium.org/1790353002
2016-03-13 13:01:57 -07:00
reed
ce563cdd48 more shader-->sp conversions
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1789633002

Review URL: https://codereview.chromium.org/1789633002
2016-03-13 12:32:36 -07:00
reed
fde05114e6 Make sp variants for SkData
Review URL: https://codereview.chromium.org/1779263003
2016-03-11 13:02:28 -08:00
cdalton
9c3f14327a Add support for vertex and geometry shader textures
Adds a visibility bitfield to GrTextureAccess that controls in which
shaders the texture should be accessible. Also adds caps and
validation to ensure we don't exceed texture limits.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1782583002

Review URL: https://codereview.chromium.org/1782583002
2016-03-11 10:07:37 -08:00
bsalomon
41b952c296 Add deferred texture upload API.
Performs thread-safe decoding of SkImage in order to later create a texture-backed SkImage.

The client allocates storage for the data.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1776693002

Review URL: https://codereview.chromium.org/1776693002
2016-03-11 06:46:33 -08:00
mtklein
218c846ac0 Revert of Use std::unique_ptr. (patchset #8 id:130001 of https://codereview.chromium.org/1780933003/ )
Reason for revert:
Another Android ambiguity due to implicit bool...

frameworks/base/core/jni/android/graphics/Utils.cpp:110:35: error: call of overloaded 'SkMemoryStream(SkAutoTUnref<SkData>&)' is ambiguous
     return new SkMemoryStream(data);

Original issue's description:
> Use std::unique_ptr.
>
> TBR=reed@google.com
>
> Committed: https://skia.googlesource.com/skia/+/20c1e3abfc681771f73eb19fde7284196e028940
>
> Committed: https://skia.googlesource.com/skia/+/3dd9ed37c24611af86f0fe374bd3698b63f09450

TBR=bungeman@google.com,mtklein@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/1785353002
2016-03-11 06:10:30 -08:00
mtklein
3dd9ed37c2 Use std::unique_ptr.
TBR=reed@google.com

Committed: https://skia.googlesource.com/skia/+/20c1e3abfc681771f73eb19fde7284196e028940

Review URL: https://codereview.chromium.org/1780933003
2016-03-11 05:18:48 -08:00
mtklein
977c07dad2 Revert of Use std::unique_ptr. (patchset #7 id:120001 of https://codereview.chromium.org/1780933003/ )
Reason for revert:
Now we remember!  The problem was Clank:

https://build.chromium.org/p/tryserver.chromium.android/builders/android_clang_dbg_recipe/builds/34329

Original issue's description:
> Use std::unique_ptr.
>
> TBR=reed@google.com
>
> Committed: https://skia.googlesource.com/skia/+/20c1e3abfc681771f73eb19fde7284196e028940

TBR=bungeman@google.com,mtklein@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/1782973002
2016-03-10 11:31:27 -08:00
mtklein
20c1e3abfc Use std::unique_ptr.
TBR=reed@google.com

Review URL: https://codereview.chromium.org/1780933003
2016-03-10 10:10:03 -08:00
brianosman
8e47881281 Remove anglesrgb config test - that config was removed.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1778413002

Review URL: https://codereview.chromium.org/1778413002
2016-03-10 09:47:57 -08:00
brianosman
d93c120852 Added (color=8888|f16|srgb) option to gpu configurations, along with gpuf16 and gpusrgb predefined configs. Runs the gpu backend in gamma-correct mode (with either FP16 linear or sRGB 8888 frambuffers).
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1750383002

Committed: https://skia.googlesource.com/skia/+/a6f58194733c1c50e4fe5f98585e42344f29b6f0

Review URL: https://codereview.chromium.org/1750383002
2016-03-10 07:49:08 -08:00
bsalomon
e6d665e24f Fix readback of 565 on Mac and enable tests
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1778143003

Review URL: https://codereview.chromium.org/1778143003
2016-03-10 07:22:25 -08:00
reed
150835e779 sk_sp versions of newWithColorFilter and newWithLocalMatrix
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1782703002

Review URL: https://codereview.chromium.org/1782703002
2016-03-10 06:36:49 -08:00
msarett
4984c3c95f Update Skia's YUV API
We should match the recently designed API in SkCodec.
https://codereview.chromium.org/1549473003/

This requires changes in Chromium as well.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1716523002

Committed: https://skia.googlesource.com/skia/+/095d31c8a0eeb5d491febf064bc3c8a44e22b94f

Review URL: https://codereview.chromium.org/1716523002
2016-03-10 05:44:43 -08:00
bsalomon
0d996868c3 Restore SkImage::NewTextureFromPixmap changes.
6ceeebd37a contained the fix for the issues that prompted the revert (index8 bitmaps weren't drawing).

Revert "TBR=robertphillips@google.com"

This reverts commit 236640872f.

TBR=
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1775263004

Review URL: https://codereview.chromium.org/1775263004
2016-03-09 18:44:43 -08:00
reed
1cb3646add un-restore test, as bug is not fixed
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1783633002

TBR=

Review URL: https://codereview.chromium.org/1783633002
2016-03-09 15:21:32 -08:00
bsalomon
236640872f TBR=robertphillips@google.com
Revert "Add SkImage::NewTextureFromPixmap"

This reverts commit 045afea979.

Revert "Fix Mac bots because of failing readback of 565"

This reverts commit e344b26a7d.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1777863003

Review URL: https://codereview.chromium.org/1777863003
2016-03-09 14:50:15 -08:00
reed
5671c5b912 SkImage now has makeShader to return sk_sp
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1785473002

Review URL: https://codereview.chromium.org/1785473002
2016-03-09 14:47:34 -08:00
reed
6ceeebd37a unify peekPixels around pixmap parameter
requires this chrome CL to land first
https://codereview.chromium.org/1775393003/

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1784563002

Review URL: https://codereview.chromium.org/1784563002
2016-03-09 14:26:26 -08:00
Brian Salomon
e344b26a7d Fix Mac bots because of failing readback of 565
TBR=robertphillips@google.com

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1784553003

Review URL: https://codereview.chromium.org/1784553003 .
2016-03-09 16:27:49 -05:00
bsalomon
045afea979 Add SkImage::NewTextureFromPixmap
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1776913003

Review URL: https://codereview.chromium.org/1776913003
2016-03-09 11:31:18 -08:00
robertphillips
6ac97b7eb9 Switch SkImageSource image filter over to new onFilterImage interface
This CL relies on: https://codereview.chromium.org/1762013002/ (Swap over to using SkImageFilter::filterImage instead of filterImageDeprecated)

TBR=bsalomon@google.com

GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1772933002

Review URL: https://codereview.chromium.org/1772933002
2016-03-09 05:17:10 -08:00
halcanary
8103a34300 SkPDF: Add sk_sp setters; .release() becomes std::move()
Note to reviewers:  Start with changes to SkPDFTypes.h

Many places that had a bare pointer owning a reference are refactored to
use a sk_sp.

There remain several places where a non-owning pointer `T*` should be
replaced with `const sk_sp<T>&` to eliminate the common pattern
`sk_sp<T>(SkRef(x))`.

Committed: https://skia.googlesource.com/skia/+/9904c9212074279380e21f96575078734dbbd308
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1775043002

Review URL: https://codereview.chromium.org/1775043002
2016-03-08 15:10:16 -08:00
halcanary
51d04d3c17 Revert of SkPDF: Add sk_sp setters; .release() becomes std::move() (patchset #2 id:20001 of https://codereview.chromium.org/1775043002/ )
Reason for revert:
https://build.chromium.org/p/client.skia/builders/Linux%20Builder/builds/6405/steps/compile/logs/stdio

Original issue's description:
> SkPDF: Add sk_sp setters; .release() becomes std::move()
>
> Note to reviewers:  Start with changes to SkPDFTypes.h
>
> Many places that had a bare pointer owning a reference are refactored to
> use a sk_sp.
>
> There remain several places where a non-owning pointer `T*` should be
> replaced with `const sk_sp<T>&` to eliminate the common pattern
> `sk_sp<T>(SkRef(x))`.
>
> Committed: https://skia.googlesource.com/skia/+/9904c9212074279380e21f96575078734dbbd308

TBR=bungeman@google.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/1775143002
2016-03-08 13:03:55 -08:00
reed
647cc84748 Add sk_ref_sp helper function.
Review URL: https://codereview.chromium.org/1773943004
2016-03-08 12:54:48 -08:00
halcanary
9904c92120 SkPDF: Add sk_sp setters; .release() becomes std::move()
Note to reviewers:  Start with changes to SkPDFTypes.h

Many places that had a bare pointer owning a reference are refactored to
use a sk_sp.

There remain several places where a non-owning pointer `T*` should be
replaced with `const sk_sp<T>&` to eliminate the common pattern
`sk_sp<T>(SkRef(x))`.

Review URL: https://codereview.chromium.org/1775043002
2016-03-08 12:38:22 -08:00
reed
93bb080503 make pm4f be RGBA always, not pmcolor order
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1774523002

Review URL: https://codereview.chromium.org/1774523002
2016-03-08 10:09:18 -08:00
bungeman
c901c11549 Add element_type, swap, operators, fix reset on sk_sp.
The 'element_type' typedef is to play nice with std::pointer_traits.

The full complement of operators and swap to match unique_ptr so that
sk_sp can be properly compared to nullptr and used with standard
containers.

Update to 'reset' so that calling 'unref' is the last operation.

This also adds tests for these changes, and sets the fPtr to nullptr
in debug for easier bug finding.

Review URL: https://codereview.chromium.org/1773453002
2016-03-08 08:35:23 -08:00
msarett
53add95b2e Disable color space test for old versions of libpng
Should fix the Google3 build.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1769953002

Review URL: https://codereview.chromium.org/1769953002
2016-03-07 17:25:12 -08:00
robertphillips
4418dbac33 Swap over to using SkImageFilter::filterImage instead of filterImageDeprecated
This CL relies on https://codereview.chromium.org/1757983002/ (Add SkSpecialImage-based methods to SkImageFilter)

GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1762013002

TBR=bsalomon@google.com

Review URL: https://codereview.chromium.org/1762013002
2016-03-07 12:45:14 -08:00
mtklein
71aca54991 add virtual ~Effect()
Today we make an sk_sp<Effect> from an sk_sp<EffectImpl>.
But when the sk_sp<Effect> dies, it calls ~Effect(), not ~EffectImpl().
Making ~Effect() virtual fixes this.

This should make our Google3 tests sized-delete clean, unblocking those folks.

Review URL: https://codereview.chromium.org/1769093002
2016-03-07 12:28:17 -08:00
herb
c60fee0f15 Rework files and add test cases.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1757193002

Review URL: https://codereview.chromium.org/1757193002
2016-03-07 09:17:47 -08:00
msarett
1590f3b328 Revert of Update Skia's YUV API (patchset #5 id:160001 of https://codereview.chromium.org/1716523002/ )
Reason for revert:
Blimp failures with manual roll in Chrome.

Original issue's description:
> Update Skia's YUV API
>
> We should match the recently designed API in SkCodec.
> https://codereview.chromium.org/1549473003/
>
> This requires changes in Chromium as well.
>
> BUG=skia:
> GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1716523002
>
> Committed: https://skia.googlesource.com/skia/+/095d31c8a0eeb5d491febf064bc3c8a44e22b94f

TBR=scroggo@google.com,reed@google.com,bsalomon@google.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Review URL: https://codereview.chromium.org/1775493002
2016-03-07 09:16:52 -08:00
msarett
095d31c8a0 Update Skia's YUV API
We should match the recently designed API in SkCodec.
https://codereview.chromium.org/1549473003/

This requires changes in Chromium as well.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1716523002

Review URL: https://codereview.chromium.org/1716523002
2016-03-07 08:39:12 -08:00
reed
941da9d661 Fix behavior of sk_sp::reset(T*) and add unittest.
Previously, sk_sp::reset(T* t) did not release its own reference
if its internal pointer was the same as 't'. This leaks a reference.
Now always release the current reference when non-nullptr.

Review URL: https://codereview.chromium.org/1767983002
2016-03-06 13:54:00 -08:00
halcanary
ffa39e02bc sk_sp: fix contravariant constructors
TBR=reed
This is obviously correct and needs to be fixed.

Review URL: https://codereview.chromium.org/1771583002
2016-03-05 08:30:28 -08:00
reed
f70b531daa Move annotations to canvas virtual (patchset #8 id:140001 of https://codereview.chromium.org/1744103002/ )"
Need to land chrome change first https://codereview.chromium.org/1766723003/

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1758023003

TBR=

Review URL: https://codereview.chromium.org/1758023003
2016-03-04 16:36:20 -08:00
msarett
6a738217ea Set SkColorSpace object for PNGs and parse ICC profiles
Code for ICC profile parsing adapted from:
https://codereview.chromium.org/1707033002/

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1726823002

Review URL: https://codereview.chromium.org/1726823002
2016-03-04 13:27:35 -08:00
reed
d882901576 Add support for new bitmapshader context (patchset #5 id:80001 of https://codereview.chromium.org/1757993002/ )"
This reverts commit cd660e1c07.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1760123003

Review URL: https://codereview.chromium.org/1760123003
2016-03-04 11:07:43 -08:00
reed
a0cee5f30e remove align16 calls in skhader context sizes. will handle this elsewhere as needed
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1768433003

TBR=

basically reverts previous CL, but keeps the create --> onCreate change

Review URL: https://codereview.chromium.org/1768433003
2016-03-04 07:38:12 -08:00