[cleanup] rename var in perf script to match sktest script

NOTRY=true

Change-Id: Ib3c03092045289eb0828f951b1213ecab893d0d9
Reviewed-on: https://skia-review.googlesource.com/10969
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Ben Wagner <benjaminwagner@google.com>

Revert[4] "clean up (partially) colortable api""""

Fixes:
- create temp api for android to pass nullptr
- don't release and access sk_sp<SkData> at the same time in parameters

This reverts commit b14131c1851eea6acbd34cc42a8f860daed36b21.

Bug: skia:
Change-Id: Ic0e4f62520ba9f35455499ed30d306ad19d998a8
Reviewed-on: https://skia-review.googlesource.com/11129
Commit-Queue: Mike Reed <reed@google.com>
Reviewed-by: Matt Sarett <msarett@google.com>

Fix Vulkan memory leak in MSAA

BUG=skia:

Change-Id: Ia32bd6654fb3f5371d8109e04464e1932d4bab52
Reviewed-on: https://skia-review.googlesource.com/11120
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>

Use SkTransferFunctionBehavior in SkImageGenerator

This fixes SkColorSpaceXformCanvas gms that expect
non-linear premuls from the codec.

BUG=skia:

Change-Id: I5dc236d0cd760c23605a26e9c33ddb18955f9231
Reviewed-on: https://skia-review.googlesource.com/10164
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Matt Sarett <msarett@google.com>

Convert all coord transforms to high precision

Testing this to see what the performance/quality tradeoff is. This is
known to fix some gradient bugs on Mali, but we want to observe perf and
see if it slows thigs down.

Bug: skia:5654
Change-Id: I988410478e89576b8e81786f49e627915e224d42
Reviewed-on: https://skia-review.googlesource.com/11128
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>

Support parametric transfer functions in SkImage_Raster::onMakeColorSpace()

BUG=skia:6456
Change-Id: Ib94bba9db669562a9b2b64fff56ebe40a2bc0096
Reviewed-on: https://skia-review.googlesource.com/11122
Commit-Queue: Matt Sarett <msarett@google.com>
Reviewed-by: Mike Klein <mtklein@chromium.org>

move vertex-mode enum into SkVertices

BUG=skia:6366

Change-Id: I3c0bf96cce6d32c9b8d12d16a772aaa6f18981aa
Reviewed-on: https://skia-review.googlesource.com/11062
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Reed <reed@google.com>

Add SkThreadedBMPDevice for Threaded Raster Backend

BUG=skia:

Change-Id: I882b6563c735796f3a4dcd19f6c79e7efd8306ae
Reviewed-on: https://skia-review.googlesource.com/10505
Commit-Queue: Yuqian Li <liyuqian@google.com>
Reviewed-by: Mike Klein <mtklein@chromium.org>

skslc now uses a more recent version of Flex

Bug: 6447
Change-Id: Ie1cb674752442a8992dd247abb2bcace8497c1f4
Reviewed-on: https://skia-review.googlesource.com/11014
Reviewed-by: Ben Wagner <benjaminwagner@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>

Avoid extra bitmap copies in SkColorSpaceXformCanvas

Before:
ColorCanvasDrawBitmap_sRGB_to_sRGB     5.09us
ColorCanvasDrawBitmap_AdobeRGB_to_sRGB 50.7us

After:
ColorCanvasDrawBitmap_sRGB_to_sRGB     2.43us
ColorCanvasDrawBitmap_AdobeRGB_to_sRGB 37.1us

BUG=skia:6456
Change-Id: Ie382936c36fd347b59485882cf8f27f315a5d35f

Change-Id: Ie382936c36fd347b59485882cf8f27f315a5d35f
Reviewed-on: https://skia-review.googlesource.com/11040
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Matt Sarett <msarett@google.com>

Move pipeline handling out of GrMeshDrawOp.

The monolithic GrPipeline is moved to a subclass GrLegacyDrawMeshOp.

The pipeline used to record a GrMesh draw in a GrMeshDrawOp must now be passed rather than implicitly using the op's pipeline.

Change-Id: I50d77e4dcc8d91a523fa7566ce43a9a291174706
Reviewed-on: https://skia-review.googlesource.com/11002
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>

Revert "Revert "Revert "Allow FPs to elevate default precision for the entire fragment program"""

This reverts commit ec53c636b781830cb6146bb32106d4ee44651fcc.

Reason for revert: Mali GMs look bad.

Original change's description:
> Revert "Revert "Allow FPs to elevate default precision for the entire fragment program""
>
> This reverts commit 903c3f70400462e191ccbca63393b2df3fe2dd11.
>
> Reason for revert: Vulkan issue fixed in compiler.
>
> Original change's description:
> > Revert "Allow FPs to elevate default precision for the entire fragment program"
> >
> > This reverts commit 92d7ccafdf896cf19764e025873d13315a76842d.
> >
> > Reason for revert: Vulkan errors.
> >
> > Original change's description:
> > > Allow FPs to elevate default precision for the entire fragment program
> > >
> > > Currently, GrConfigConversionEffect is able to round-trip on many mobile
> > > GPUs because it uses highp for all intermediate variables (including the
> > > texture fetch result). Separating the texture sample into a different
> > > processor breaks that.
> > >
> > > This is a blunt instrument, not to be used lightly.
> > >
> > > Bug: skia:
> > > Change-Id: I2ab365e3da79628069e2eb727c43c2bf45bfd789
> > > Reviewed-on: https://skia-review.googlesource.com/10162
> > > Reviewed-by: Brian Salomon <bsalomon@google.com>
> > > Commit-Queue: Brian Osman <brianosman@google.com>
> > >
> >
> > TBR=bsalomon@google.com,robertphillips@google.com,brianosman@google.com,ethannicholas@google.com,reviews@skia.org
> > NOPRESUBMIT=true
> > NOTREECHECKS=true
> > NOTRY=true
> >
> > Change-Id: Iee5bb409f86a9cabecc76bd1273a5b3cef6af179
> > Reviewed-on: https://skia-review.googlesource.com/10967
> > Reviewed-by: Brian Osman <brianosman@google.com>
> > Commit-Queue: Brian Osman <brianosman@google.com>
> >
>
> TBR=bsalomon@google.com,robertphillips@google.com,reviews@skia.org,brianosman@google.com,ethannicholas@google.com
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
>
> Change-Id: I733a0ecc40b58d8727f0259b5498c8e6610cedce
> Reviewed-on: https://skia-review.googlesource.com/11010
> Reviewed-by: Brian Osman <brianosman@google.com>
> Commit-Queue: Brian Osman <brianosman@google.com>
>

TBR=bsalomon@google.com,robertphillips@google.com,reviews@skia.org,brianosman@google.com,ethannicholas@google.com
# Not skipping CQ checks because original CL landed > 1 day ago.

Change-Id: Ic3274a0a8b776e811354c3441391ffdc80678292
Reviewed-on: https://skia-review.googlesource.com/11061
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>

Remove kAsIs_CachedFormat

This was confusing and unnecessary. Added an explicit sBGR format to
handle one case. Now the CF always fully describes the color type and
transfer function (or it's legacy).

Bug: skia:
Change-Id: I8f792e8aa18cd2391aa541a9368e345d92bba828
Reviewed-on: https://skia-review.googlesource.com/11013
Reviewed-by: Matt Sarett <msarett@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>

jumper, drop Android NDK dependency

We don't _really_ need the Android NDK.  We just need <arm_neon.h>
(which comes from Clang, not the NDK) and a smattering of <stdint.h>
([u]intN_t), <string.h> (memcpy) and <stddef.h> (size_t).

The idea here is solely to make it easier to run build_stages.py.
If this becomes a pain to maintain, I'm happy to go back to the NDK.

Change-Id: Ic6bb287646b6160ac42ac6e4d5290a66a7e92425
Reviewed-on: https://skia-review.googlesource.com/10980
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>

Update SKP version

Automatic commit by the RecreateSKPs bot.

TBR=update-skps@skia.org
NO_MERGE_BUILDS

Change-Id: If3040a5c77537c3696f1af89ea9ea39dc6938a1e
Reviewed-on: https://skia-review.googlesource.com/11026
Reviewed-by: update-skps <update-skps@skia.org>
Commit-Queue: update-skps <update-skps@skia.org>

Revert "Revert[2] "clean up (partially) colortable api"""

This reverts commit 9920b10f5292838f00600f676c4578cd11705e60.

Reason for revert: trying to get details on w2k failure

https://chromium-swarm.appspot.com/task?id=354345d34ba3b310&refresh=10

Caught exception 3221225477 EXCEPTION_ACCESS_VIOLATION, was running:
unit test  HugeBlurImageFilter
unit test  FontNames
unit test  Codec_PngRoundTrip
unit test  ClampRange
unit test  FontHost
unit test  ColorMatrixFilter
f16 image scaled_codec_premul abnormal.wbmp
565 image brd_android_codec_divisor_0.167 interlaced3.png_0.167
unit test  Codec_png
unit test  ImageFilterBlurLargeImage
unit test  FontObj
unit test  DrawText
unit test  GrShape
565 image brd_android_codec_divisor_0.333 interlaced2.png_0.333
unit test  PathOpsOpCubicsThreaded
unit test  PathOpsOpLoopsThreaded
unit test  FontMgr
unit test  ColorToHSVRoundTrip
unit test  Image_Serialize_Encoding_Failure
Likely culprit:
unit test  Image_Serialize_Encoding_Failure
step returned non-zero exit code: -1073741819

Original change's description:
> Revert[2] "clean up (partially) colortable api""
>
> This reverts commit 1d1165ca6575e082b892c5460492c411618783ad.
>
> Bug: skia:
> Change-Id: Idbc0634ae3cec2e79f592d252de8751b077e6408
> Reviewed-on: https://skia-review.googlesource.com/11024
> Reviewed-by: Mike Reed <reed@google.com>
> Commit-Queue: Mike Reed <reed@google.com>
>

TBR=reed@google.com,reviews@skia.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Change-Id: Ia4e73434b083224baa36092c69526c2f59bb16aa
Reviewed-on: https://skia-review.googlesource.com/11025
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Reed <reed@google.com>

Revert[2] "clean up (partially) colortable api""

This reverts commit 1d1165ca6575e082b892c5460492c411618783ad.

Bug: skia:
Change-Id: Idbc0634ae3cec2e79f592d252de8751b077e6408
Reviewed-on: https://skia-review.googlesource.com/11024
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Reed <reed@google.com>

Revert "Respect kRectsMustMatchForMSAASrc_BlitFramebufferFlag in dst setup"

This reverts commit d58f040532f2f5a63d24bd17d7c588e52c0b99c3.

Reason for revert: tests/BlendTest is failing on the Nexus Player:

https://chromium-swarm.appspot.com/task?id=353ffc638e202210
https://chromium-swarm.appspot.com/task?id=353ff5e35819ab10

Original change's description:
> Respect kRectsMustMatchForMSAASrc_BlitFramebufferFlag in dst setup
>
> Crurently, when preparing a texture for blitFramebuffer, we ignore the
> kRectsMustMatchForMSAASrc_BlitFramebufferFlag, and may attempt to
> copy from one src rect to a different dst rect.
>
> This change updates initDescForDstCopy and setupDstTexture to allocate
> larger textures if necessary and accomodate this flags requirements.
>
> Bug: 658277
> Change-Id: I9f45a03d4055e0ad87c01e1d826287695096e609
> Reviewed-on: https://skia-review.googlesource.com/10941
> Commit-Queue: Brian Salomon <bsalomon@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
>

TBR=bsalomon@google.com,ericrk@chromium.org,reviews@skia.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Change-Id: I0fd6ca95bbc342f21978783b0103073179017795
Reviewed-on: https://skia-review.googlesource.com/11016
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>

Revert of fix size check for drawBitmap fast-path in SkBitmapDevice::drawBitmapRect (patchset #2 id:20001 of https://codereview.chromium.org/2089583002/ )

Reason for revert:
I agree the diffs are not meaningful, but nevertheless their existence is blocking the roll into Chrome.

(https://storage.googleapis.com/chromium-layout-test-archives/linux_trusty_blink_rel/7376/layout-test-results/results.html)

Original issue's description:
> Fix the size check for the drawBitmap fast-path in SkBitmapDevice::drawBitmapRect. It would fail when the source rectangle had a non-zero offset, in which case it would compare the source rectangle with the offset to the extracted bitmap size, which always fails. The only thing that should matter is that the source rectangle and extract bitmap have the same size, since the offset gets added onto the matrix.
>
> BUG=skia:
> GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2089583002
>
> Review-Url: https://codereview.chromium.org/2089583002
> Committed: https://skia.googlesource.com/skia/+/ea9bc0c07b5dae78a9a449d7d7a07fc79262d41a

TBR=reed@google.com,caryclark@google.com,lsalzman@mozilla.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/2787263005

Respect kRectsMustMatchForMSAASrc_BlitFramebufferFlag in dst setup

Crurently, when preparing a texture for blitFramebuffer, we ignore the
kRectsMustMatchForMSAASrc_BlitFramebufferFlag, and may attempt to
copy from one src rect to a different dst rect.

This change updates initDescForDstCopy and setupDstTexture to allocate
larger textures if necessary and accomodate this flags requirements.

Bug: 658277
Change-Id: I9f45a03d4055e0ad87c01e1d826287695096e609
Reviewed-on: https://skia-review.googlesource.com/10941
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>

switched skslc to_string strategy after repeated problems on Android

BUG=skia:

Change-Id: I9e27d06351e4bfcd503882b2c604878b44b6e5cc
Reviewed-on: https://skia-review.googlesource.com/11015
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>

added missing <inttypes.h> include

Bug: skia:
Change-Id: I24f556c791f5a8ca295f6dc88dbf5532a7a9f688
Reviewed-on: https://skia-review.googlesource.com/11008
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>

Fix perf configs for ANGLE bots.

Don't run non-ANGLE configs on ANGLE bots.

Also, stop perfing angle_gl_es2, since it's not used by Chrome. Add perfing for
ANGLE MSAA, and omit test that fails with MSAA config.

NOTRY=true
BUG=skia:6457

Change-Id: Iafe605a369e6bfddddeee3d30a538ec889eb0952
Reviewed-on: https://skia-review.googlesource.com/10968
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Ben Wagner <benjaminwagner@google.com>

Fix the size check for the drawBitmap fast-path in SkBitmapDevice::drawBitmapRect. It would fail when the source rectangle had a non-zero offset, in which case it would compare the source rectangle with the offset to the extracted bitmap size, which always fails. The only thing that should matter is that the source rectangle and extract bitmap have the same size, since the offset gets added onto the matrix.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2089583002

Review-Url: https://codereview.chromium.org/2089583002

Revert "Revert "Allow FPs to elevate default precision for the entire fragment program""

This reverts commit 903c3f70400462e191ccbca63393b2df3fe2dd11.

Reason for revert: Vulkan issue fixed in compiler.

Original change's description:
> Revert "Allow FPs to elevate default precision for the entire fragment program"
>
> This reverts commit 92d7ccafdf896cf19764e025873d13315a76842d.
>
> Reason for revert: Vulkan errors.
>
> Original change's description:
> > Allow FPs to elevate default precision for the entire fragment program
> >
> > Currently, GrConfigConversionEffect is able to round-trip on many mobile
> > GPUs because it uses highp for all intermediate variables (including the
> > texture fetch result). Separating the texture sample into a different
> > processor breaks that.
> >
> > This is a blunt instrument, not to be used lightly.
> >
> > Bug: skia:
> > Change-Id: I2ab365e3da79628069e2eb727c43c2bf45bfd789
> > Reviewed-on: https://skia-review.googlesource.com/10162
> > Reviewed-by: Brian Salomon <bsalomon@google.com>
> > Commit-Queue: Brian Osman <brianosman@google.com>
> >
>
> TBR=bsalomon@google.com,robertphillips@google.com,brianosman@google.com,ethannicholas@google.com,reviews@skia.org
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
>
> Change-Id: Iee5bb409f86a9cabecc76bd1273a5b3cef6af179
> Reviewed-on: https://skia-review.googlesource.com/10967
> Reviewed-by: Brian Osman <brianosman@google.com>
> Commit-Queue: Brian Osman <brianosman@google.com>
>

TBR=bsalomon@google.com,robertphillips@google.com,reviews@skia.org,brianosman@google.com,ethannicholas@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Change-Id: I733a0ecc40b58d8727f0259b5498c8e6610cedce
Reviewed-on: https://skia-review.googlesource.com/11010
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>

fixed SPIR-V compilation error with unmodified variables

Change-Id: I22981a60c21a2e16d41d69c53e099f221a202a58
Reviewed-on: https://skia-review.googlesource.com/11000
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>

skslc can now be compiled with no Skia dependencies, in preparation for its eventual

This reverts commit 9bd301d640ff63c280b202c7dd00bc00a3315ff4.

Bug: skia:
Change-Id: I5ad3f77ef33aa5ce2fd27fe383c9339c571663a1
Reviewed-on: https://skia-review.googlesource.com/10964
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>

jumper, a couple simple loads and stores

Change-Id: I217d7b562f5fa443978044e17469ba757c061209
Reviewed-on: https://skia-review.googlesource.com/10971
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>

jumper, only ignore red zone on Windows

On Linux and Mac there's always a red zone of 128 bytes of stack space
for us to use without touching the stack pointer.  We'd been generating
stage code as if that's not there because it's not there on Windows.

We have a separate .S file for Windows anyway, so there's no need to
ignore the red zone when we know it's there.

Change-Id: I81a7841020bb8aad68bf35feac851727ef1d0758
Reviewed-on: https://skia-review.googlesource.com/10965
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Mike Klein <mtklein@chromium.org>
Reviewed-by: Herb Derby <herb@google.com>

Update NVidia Shields to NRD90M

BUG=skia:

Change-Id: I9411bfa3a9324224e2dd830a47fa025d750cbebc
Reviewed-on: https://skia-review.googlesource.com/10044
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Commit-Queue: Kevin Lubick <kjlubick@google.com>

Revert "Allow FPs to elevate default precision for the entire fragment program"

This reverts commit 92d7ccafdf896cf19764e025873d13315a76842d.

Reason for revert: Vulkan errors.

Original change's description:
> Allow FPs to elevate default precision for the entire fragment program
>
> Currently, GrConfigConversionEffect is able to round-trip on many mobile
> GPUs because it uses highp for all intermediate variables (including the
> texture fetch result). Separating the texture sample into a different
> processor breaks that.
>
> This is a blunt instrument, not to be used lightly.
>
> Bug: skia:
> Change-Id: I2ab365e3da79628069e2eb727c43c2bf45bfd789
> Reviewed-on: https://skia-review.googlesource.com/10162
> Reviewed-by: Brian Salomon <bsalomon@google.com>
> Commit-Queue: Brian Osman <brianosman@google.com>
>

TBR=bsalomon@google.com,robertphillips@google.com,brianosman@google.com,ethannicholas@google.com,reviews@skia.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Change-Id: Iee5bb409f86a9cabecc76bd1273a5b3cef6af179
Reviewed-on: https://skia-review.googlesource.com/10967
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>

jumper, caught up on blend modes

Change-Id: I39538c90cd4c68691c3956e3f51616b77e4c90d1
Reviewed-on: https://skia-review.googlesource.com/10961
Reviewed-by: Mike Klein <mtklein@chromium.org>
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>

Remove compiler bug workaround

The ICE was (indirectly) triggered by the swizzle code, which has been
removed. Verified that this code now compiles on MSVC without incident.

Change-Id: I730ad8d39c02a37daa91c2248bd680339b7c8de2
Reviewed-on: https://skia-review.googlesource.com/10963
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>

Revert "clean up (partially) colortable api"

This reverts commit 2e491a6a113c5e16a3b7bede5fa6f588deeb928d.

Reason for revert: Windows unit tests failing?

Original change's description:
> clean up (partially) colortable api
>
> Needs this to land: https://codereview.chromium.org/2789853002/
>
> Bug: skia:
> Change-Id: I38d916a546b7fa64d000d973e695ddda24a589e7
> Reviewed-on: https://skia-review.googlesource.com/10600
> Commit-Queue: Mike Reed <reed@google.com>
> Reviewed-by: Matt Sarett <msarett@google.com>
>

TBR=msarett@google.com,scroggo@google.com,reed@google.com,reviews@skia.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Change-Id: I626e7edfcea82576a440dcaa851a04cedee6233f
Reviewed-on: https://skia-review.googlesource.com/10966
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>

make build_stages a little less mtklein-centric

Change-Id: I69069aeaefd1c8c90de83eb86bb935e82a74bc9f
Reviewed-on: https://skia-review.googlesource.com/10923
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>

Add tracking of unique proxyID beside unique renderTargetID in auditTrail and clearOp

Split out of: https://skia-review.googlesource.com/c/10284/ (Omnibus: Remove GrSurface-derived classes from ops)

Change-Id: I5845a47d94decc455ec3b1f0a5876b1c82aa32e8
Reviewed-on: https://skia-review.googlesource.com/10750
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>

Revert "skslc can now be compiled with no Skia dependencies, in preparation for its eventual"

This reverts commit f3333c89bf05fc602d9bf8e1e24547668c660383.

Reason for revert: breaking the bots

Original change's description:
> skslc can now be compiled with no Skia dependencies, in preparation for its eventual
> role in Skia's build process.
>
> This reverts commit bcf35f86d50b784b165de703b404998dd4299f6a.
>
> BUG=skia:
>
> Change-Id: Id0a12dfc4d804d69a3c6bf60fed37e89ee130f02
> Reviewed-on: https://skia-review.googlesource.com/10802
> Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
> Reviewed-by: Ben Wagner <benjaminwagner@google.com>
>

TBR=benjaminwagner@google.com,ethannicholas@google.com,reviews@skia.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Change-Id: Ic7b50d391d25b3870acffa9764cbafc7f5c3be89
Reviewed-on: https://skia-review.googlesource.com/10962
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>

Allow FPs to elevate default precision for the entire fragment program

Currently, GrConfigConversionEffect is able to round-trip on many mobile
GPUs because it uses highp for all intermediate variables (including the
texture fetch result). Separating the texture sample into a different
processor breaks that.

This is a blunt instrument, not to be used lightly.

Bug: skia:
Change-Id: I2ab365e3da79628069e2eb727c43c2bf45bfd789
Reviewed-on: https://skia-review.googlesource.com/10162
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>

clean up (partially) colortable api

Needs this to land: https://codereview.chromium.org/2789853002/

Bug: skia:
Change-Id: I38d916a546b7fa64d000d973e695ddda24a589e7
Reviewed-on: https://skia-review.googlesource.com/10600
Commit-Queue: Mike Reed <reed@google.com>
Reviewed-by: Matt Sarett <msarett@google.com>

skslc can now be compiled with no Skia dependencies, in preparation for its eventual
role in Skia's build process.

This reverts commit bcf35f86d50b784b165de703b404998dd4299f6a.

BUG=skia:

Change-Id: Id0a12dfc4d804d69a3c6bf60fed37e89ee130f02
Reviewed-on: https://skia-review.googlesource.com/10802
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: Ben Wagner <benjaminwagner@google.com>

jumper, another batch of blend modes

Change-Id: I0c48ec80eee8b7c7e9fb980efa8ed1dad5ad9768
Reviewed-on: https://skia-review.googlesource.com/10924
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>

jumper, more blend modes

Change-Id: I17ce08a7ec62ef8ffe8ae567079d669a87ef9a9c
Reviewed-on: https://skia-review.googlesource.com/10921
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>

Revert "Respect kRectsMustMatchForMSAASrc_BlitFramebufferFlag in dst setup"

This reverts commit e0ff6ab272dcc5c4d6f3b509da7e0e97acada814.

Reason for revert: a bunch of bots failed.

Original change's description:
> Respect kRectsMustMatchForMSAASrc_BlitFramebufferFlag in dst setup
>
> Crurently, when preparing a texture for blitFramebuffer, we ignore the
> kRectsMustMatchForMSAASrc_BlitFramebufferFlag, and may attempt to
> copy from one src rect to a different dst rect.
>
> This change updates initDescForDstCopy and setupDstTexture to allocate
> larger textures if necessary and accomodate this flags requirements.
>
> Bug: 658277
> Change-Id: I500f10dba5700f5f7a7acad04bcdbc9ac9994835
> Reviewed-on: https://skia-review.googlesource.com/10247
> Commit-Queue: Brian Salomon <bsalomon@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
>

TBR=bsalomon@google.com,ericrk@chromium.org,reviews@skia.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Change-Id: I7fbd6c2652fe71c707d3120b035e0365fbc7fa66
Reviewed-on: https://skia-review.googlesource.com/10920
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>

ok, most features can be vias

This refactors most features out of ok's core into vias:
   -w    --> a .png dumping via, "png", opening the door to other types
   -m/-s --> a filtering via "filter"

Everything now can print a brief help message too.

Change-Id: I9e653aab98fd57182a6d458c7a80052130980284
Reviewed-on: https://skia-review.googlesource.com/10509
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>

Respect kRectsMustMatchForMSAASrc_BlitFramebufferFlag in dst setup

Crurently, when preparing a texture for blitFramebuffer, we ignore the
kRectsMustMatchForMSAASrc_BlitFramebufferFlag, and may attempt to
copy from one src rect to a different dst rect.

This change updates initDescForDstCopy and setupDstTexture to allocate
larger textures if necessary and accomodate this flags requirements.

Bug: 658277
Change-Id: I500f10dba5700f5f7a7acad04bcdbc9ac9994835
Reviewed-on: https://skia-review.googlesource.com/10247
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>

Respect xfer barriers in InstancedRendering::Op

Fixes a bug introduced when this Op stopped storing a GrPipeline.

Change-Id: I1a39814cebe18b321ea369f005bb7759f233cfe6
Reviewed-on: https://skia-review.googlesource.com/10804
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>

tell Google3 we do not need executable stack

Change-Id: I5ff3599448d027fcac43a53e98a801ce672ce5ee
Reviewed-on: https://skia-review.googlesource.com/10861
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Mike Klein <mtklein@chromium.org>

avoid deprecated SkRect::MakeFromIRect

Change-Id: I55a5fd92b4b53b2676983233619cee4875967e01
Reviewed-on: https://skia-review.googlesource.com/10860
Reviewed-by: Mike Klein <mtklein@chromium.org>
Reviewed-by: Matt Sarett <msarett@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>

SkColorSpaceXformCanvas: setMatrix(), clipRect() in constructor

BUG=703363
Change-Id: Ie0053940dd041614b93400e34145118fb0742cae

Change-Id: Ie0053940dd041614b93400e34145118fb0742cae
Reviewed-on: https://skia-review.googlesource.com/10806
Commit-Queue: Matt Sarett <msarett@google.com>
Reviewed-by: Mike Klein <mtklein@chromium.org>

Add more overrides to SkColorSpaceXformCanvas

Number of matching gms is unchanged.

Bug: skia:
Change-Id: I43961c716899ddacc560dd7888d4427563cdf7ca
Reviewed-on: https://skia-review.googlesource.com/10805
Commit-Queue: Matt Sarett <msarett@google.com>
Reviewed-by: Mike Klein <mtklein@chromium.org>

Turn on SkJumper all the time, try 2.

Rebased on top of _win.S change, which I hope lands first.

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel,win_chrome_official,win_chromium_dbg_ng,win_chromium_rel_ng,win_chromium_compile_dbg_ng,win_chromium_compile_rel_ng,win_clang;master.tryserver.chromium.android:android_optional_gpu_tests_rel

Change-Id: I30e97dbb55b8a42251b46f5607096ca12bc670b9
Reviewed-on: https://skia-review.googlesource.com/10462
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Herb Derby <herb@google.com>

fixed skslc SPIR-V memory error

BUG=skia:6446

Change-Id: Ibc55124db60d6a05964ddcd02d285e313379f93e
Reviewed-on: https://skia-review.googlesource.com/10756
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>

Add Linux HD2000 jobs

Bug: skia:6400
NOTRY=TRUE
Change-Id: I253ef2d0b137a4a36f26dcabb8cc4a1919ed9b60
Reviewed-on: https://skia-review.googlesource.com/10759
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Reviewed-by: Ben Wagner <benjaminwagner@google.com>

Remove Galaxy J5 bot.

NOTRY=true

Change-Id: I9288e5928a08fc57545866aad1cc8b70f86eed6d
Reviewed-on: https://skia-review.googlesource.com/10801
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Ben Wagner <benjaminwagner@google.com>

Add glPolygonMode support.

Mainly added because it's useful for rendering in wireframe mode.

Change-Id: I84dab43ce5b56a48d47dd16d4c98fa7648bb1f0b
Reviewed-on: https://skia-review.googlesource.com/10530
Commit-Queue: Jim Van Verth <jvanverth@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>

make _win.S know if it's 64-bit

I think this is the root of my Windows / Chrome problems.
Even on 32-bit builds, Chrome compiles nacl64.exe in 64-bit mode.

So to make things simple, always put _win.S in the sources,
and no-op it away when assembling for 32-bit.

Change-Id: I19f163491739a6c0cbdedd0ce353f1d2289907ae
Reviewed-on: https://skia-review.googlesource.com/10637
Reviewed-by: Matt Sarett <msarett@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>

Revert "skslc can now be compiled with no Skia dependencies, in preparation for"

This reverts commit 7833466da45bfa1e078427c4a6db94d41c5c1535.

Reason for revert: Vulkan assertion failure

Original change's description:
> skslc can now be compiled with no Skia dependencies, in preparation for
> its eventual role in Skia's build process.
>
> Bug: skia:
> Change-Id: Iaa9933f4fc4a64bec60aa897c509a3513f457a78
> Reviewed-on: https://skia-review.googlesource.com/10282
> Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
> Reviewed-by: Ben Wagner <benjaminwagner@google.com>
>

TBR=egdaniel@google.com,benjaminwagner@google.com,ethannicholas@google.com,reviews@skia.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Change-Id: Ic64cac2395abb406116885ddd725f74a434c8c49
Reviewed-on: https://skia-review.googlesource.com/10758
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>

skslc can now be compiled with no Skia dependencies, in preparation for
its eventual role in Skia's build process.

Bug: skia:
Change-Id: Iaa9933f4fc4a64bec60aa897c509a3513f457a78
Reviewed-on: https://skia-review.googlesource.com/10282
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: Ben Wagner <benjaminwagner@google.com>

Make xformer class for SkCSXCanvas, use for draw loopers

Fixes draw looper gms in gbr-8888 config.

Bug: skia:
Change-Id: I0a9306cc0523c82f2ad9516752d79c1d86b1e295
Reviewed-on: https://skia-review.googlesource.com/10644
Commit-Queue: Matt Sarett <msarett@google.com>
Reviewed-by: Mike Klein <mtklein@chromium.org>

Use the rounded text size for gasp.

When comparing against gasp ranges always use the rounded (gdi) size.
Previously the truncated size was used, but it appears other
implementations are using the rounded size.

BUG=chromium:706693

Change-Id: I0f545175bf2d5f4e8db610b26c2f3e21a89eeb2a
Reviewed-on: https://skia-review.googlesource.com/10751
Reviewed-by: Dominik Röttsches <drott@chromium.org>
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Ben Wagner <bungeman@google.com>

Minor cleanup (remove unused GrRenderTargetContext::asTexture method)

Change-Id: I719204e9870b3c849a5c60d967359a70138e6c9c
Reviewed-on: https://skia-review.googlesource.com/10753
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>

Use correct tolerance for conic chopping in MSAA and default path renderers

Bug: 706581
Change-Id: I16b3037380508316fc73ab7a8cebf960de66938e
Reviewed-on: https://skia-review.googlesource.com/10752
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>

Update instructions on Skia/Chromium multi try.

Change-Id: I7ac583dbf04d18210847c54f7c69cf8a4f46deea
Reviewed-on: https://skia-review.googlesource.com/10597
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Ben Wagner <bungeman@google.com>

Use last value for axis for variation position.

SkFontArguments::VariationPosition may be over specified. If there are
multiple values for a given axis, ensure the last value specified is
used, since that's what css-fonts-4 requires.

BUG=chromium:674878

Change-Id: I6704c15c520c89efb9ee84659a3e16e0d07691c9
Reviewed-on: https://skia-review.googlesource.com/10513
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Mike Klein <mtklein@chromium.org>

Renames of processor analysis-related classes and method.

GrProcesserSet::FragmentProcessorAnalysis->GrProcessorSet::Analysis
GrPipelineAnalysisColor->GrProcessorAnalysisColor
GrPipelineAnalysisCoverage->GrProcessorAnalysisCoverage
GrMeshDrawOp::getFragmentProcessorAnalysisInputs->GrMeshDrawOp::getProcessorAnalysisInputs

Change-Id: I28ad19dfab5f4ac1788c4eacdec5e1af2a701dd0
Reviewed-on: https://skia-review.googlesource.com/10747
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>

Don't perform analysis twice in InstancedRendering::Op

Change-Id: Ie0aeeb7bf63090b0c27a9b062cce0cdc561a1d7e
Reviewed-on: https://skia-review.googlesource.com/10583
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>

Add GrSurfaceProxy* parameter to GrContext::flush

Split out of: https://skia-review.googlesource.com/c/10284/ (Omnibus: Remove GrSurface-derived classes from ops)

Change-Id: I12c66a6fa826c2363b21ece56391fb352d12d6b3
Reviewed-on: https://skia-review.googlesource.com/10541
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>

[recipes.cfg] Disable autoroller

Bug: skia:
Change-Id: Ie87deb32bf2fb257b1aac25bdef9d5c978d0320b
Reviewed-on: https://skia-review.googlesource.com/10542
Reviewed-by: Ravi Mistry <rmistry@google.com>
Commit-Queue: Ravi Mistry <rmistry@google.com>

Reverse clamp logic in SkColorSpaceXform

Follow-up from https://skia-review.googlesource.com/c/10581/

Bug: skia:
Change-Id: I499f0908e1375854202c1cfd476827500fc89237
Reviewed-on: https://skia-review.googlesource.com/10645
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>

Don't export stage symbols.

Chromium Mac bots are getting tripped up by stages being visible.
.hidden and .private_extern are -fvisibilty=hidden for ELF and MACH-O.

CQ_INCLUDE_TRYBOTS=skia.primary:Build-Mac-Clang-arm-Debug-iOS

Change-Id: I8dbb04f514eead4ab480664f2674db4b57611b84
Reviewed-on: https://skia-review.googlesource.com/10622
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>

565 color xforms: make sure we clamp if necessary

Bug: skia:
Change-Id: I48528a0ff8dd199e0cd1649f973b5722384eacf3
Reviewed-on: https://skia-review.googlesource.com/10581
Reviewed-by: Matt Sarett <msarett@google.com>
Commit-Queue: Matt Sarett <msarett@google.com>

jumper, revert to generating .S files

I went with the unified-in-one-.cpp approach mostly to make it easy to
roll out SkJumper.  I no longer see any difficultly rolling out the
assembly files, and it's possible the unified .cpp approach just makes
things harder.

Let's see if it's any easier to get Chrome's official build to work with
normal assembly files.  It's not going to be a problem to roll out.

This is a partial revert of https://skia-review.googlesource.com/c/9336.

CQ_INCLUDE_TRYBOTS=skia.primary:Test-Win2k8-MSVC-GCE-CPU-AVX2-x86_64-Debug,Test-Mac-Clang-MacMini6.2-CPU-AVX-x86_64-Debug,Test-Ubuntu-Clang-GCE-CPU-AVX2-x86_64-Debug,Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Debug

Change-Id: Idfdbd2d322452b44bc0adaf6dc299cc7649bc51e
Reviewed-on: https://skia-review.googlesource.com/10561
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>

Fix is_convex() check for inset polys.

Change-Id: I1a8a24132b0cf9eb80dd586bf2f1b930adb7b582
Reviewed-on: https://skia-review.googlesource.com/10525
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>

Make analysis optional to GrPipeline::init().

GrXPFactory::createXferProcessor now takes GrPipelineAnalysisColor and GrPipelineAnalysisCoverage rather than GrProcessorSet::FragmentProcessorAnalysis.

This will make it so ops do not have to retain the analysis or rerun it to create pipelines at flush time.

Change-Id: Ib28ba65de425b20c2647329275f209aec168c3df
Reviewed-on: https://skia-review.googlesource.com/10474
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>

Remove blacklist of Vulkan GPUMemoryTest

This was blacklisted on Nvidia Windows. A recent change was made that may
be a possible fix for this test involving the uploading of initial mip map
memory.

BUG=skia:6092

Change-Id: Ie75c4665bea1f5139a674b9e0f53a9aac4a97f98
Reviewed-on: https://skia-review.googlesource.com/10285
Commit-Queue: Greg Daniel <egdaniel@google.com>
Reviewed-by: Ben Wagner <benjaminwagner@google.com>

More GrSurfaceProxy-clean up

Split out of: https://skia-review.googlesource.com/c/10484/ (Omnibus: Push instantiation of GrTextures later (post TextureSampler))

Change-Id: I1ee39a23c749e420dce0ad561ee1c8b09bdcc763
Reviewed-on: https://skia-review.googlesource.com/10485
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>

Use FontWeight symbols when available on Mac.

Newer versions of macOS have NSFontWeightXXX symbols and iOS has
UIFontWeightXXX symbols to provide the conversion between [-1, 1]
CTFontDescriptor weights and [0, 1000] CSS weights. Currently the
values these symbols provide are hard coded (for expediency), but the
symbols should be used when available since the values may change
without notice.

Change-Id: I0875c6a96c5b14e456f91bccf7a84f0306a3d63a
Reviewed-on: https://skia-review.googlesource.com/10163
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Ben Wagner <bungeman@google.com>

Miscellaneous GrSurfaceProxy-related cleanup

This is pulled out of: https://skia-review.googlesource.com/c/10284/ (Remove GrSurface-derived classes from ops)

Change-Id: I083c0beefe4899b3517d0b0569bb25096809f410
Reviewed-on: https://skia-review.googlesource.com/10483
Commit-Queue: Robert Phillips <robertphillips@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>

Remove GrPipelineOptimizations computation from GrPipeline::init and nest in GrMeshDrawOp.

Change-Id: I4a702c83857606c1cb050294c408922eef5769ea
Reviewed-on: https://skia-review.googlesource.com/10414
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>

Add skia_use_egl flag

This allows us to specify it for Linux Arm devices (like chromebooks, and
possibly RPIs)
Bug: skia:6442

Change-Id: I043bc3b3686caf28fedfe1a6ab1dfd300cc4f066
Reviewed-on: https://skia-review.googlesource.com/10469
Reviewed-by: Mike Klein <mtklein@chromium.org>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Kevin Lubick <kjlubick@google.com>

Fix copy offset when uploading initial mip data in Vulkan

This is a fix to an original fix which was reverted

BUG=skia:

Change-Id: I21fcda28a5d342984784a35d2b3e985c6e4199cf
Reviewed-on: https://skia-review.googlesource.com/10290
Commit-Queue: Greg Daniel <egdaniel@google.com>
Reviewed-by: Jim Van Verth <jvanverth@google.com>

Remove SK_IGNORE_DW_GRAY_FIX.

Users have been updated and this define is no longer used.

Change-Id: I0430039978171c16ffe19fb3037897bcdf313e42
Reviewed-on: https://skia-review.googlesource.com/10417
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>

Add unistd.h include to GrVkGpu.cpp on unix for sleep

Fixes Fuchsia build

Change-Id: Ieb3b7092586495f5153ba18d14a94d8dcdab1dbd
Reviewed-on: https://skia-review.googlesource.com/10261
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>

Marker for driver update on Win10 Golo GT610

BUG=chromium:705001
NOTRY=true

Change-Id: I84c888cec3ed644be4ef1bf5f7ea219bf3b8ded3
Reviewed-on: https://skia-review.googlesource.com/10411
Reviewed-by: Ben Wagner <benjaminwagner@google.com>
Commit-Queue: Ben Wagner <benjaminwagner@google.com>

Remove GrXferProcessor::getOptimizations.

This replaces GrXferProcessor::getOptimizations with a new function on GrXPFactory. The results are made available via FragmentProcessorAnalysis.

Bug: skia:
Change-Id: I535985458c9d13ad858cac94e957e2fdbe332036
Reviewed-on: https://skia-review.googlesource.com/10218
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>

Retract GrPipeline from setRenderTargetState calls

Split out of: https://skia-review.googlesource.com/c/10284/ (Remove GrSurface-derived classes from ops)

Change-Id: Ib3dda649875b8da31def20476e4b5a87a4aa7cac
Reviewed-on: https://skia-review.googlesource.com/10320
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>

Remove GrSurface::flushWrites

Change-Id: Ifac5af00ef852ee212964baa113b490a03e0168d
Reviewed-on: https://skia-review.googlesource.com/10293
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>

Update Galaxy S7s to March 2017 patch

Bug: skia:6434
NOTRY=true
NOTREECHECKS=true

Change-Id: I7daaccbf14facb4dec8929e0e819add7c4e41a46
Reviewed-on: https://skia-review.googlesource.com/10398
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Commit-Queue: Kevin Lubick <kjlubick@google.com>

Roll recipe dependencies (trivial).

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/bb49d6609717078440ca257a1960cd34033bd047 build_scan.py: Add scripts/master to path (hinoka@google.com)

TBR=iannucci@chromium.org

Change-Id: I6ccf0620c27c3bbcda4c9093a6e8ef79fbd0f967
Recipe-Tryjob-Bypass-Reason: Autoroller
Bugdroid-Send-Email: False
Reviewed-on: https://skia-review.googlesource.com/10335
Reviewed-by: Recipe Roller <recipe-roller@chromium.org>
Commit-Queue: Recipe Roller <recipe-roller@chromium.org>

Roll recipe dependencies (trivial).

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/bb49d6609717078440ca257a1960cd34033bd047 build_scan.py: Add scripts/master to path (hinoka@google.com)

TBR=iannucci@chromium.org

Change-Id: I10045227a914796ac5cae9f5f638b2b084ad15a6
Recipe-Tryjob-Bypass-Reason: Autoroller
Bugdroid-Send-Email: False
Reviewed-on: https://skia-review.googlesource.com/10334
Reviewed-by: Recipe Roller <recipe-roller@chromium.org>
Commit-Queue: Recipe Roller <recipe-roller@chromium.org>

Roll recipe dependencies (trivial).

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/bb49d6609717078440ca257a1960cd34033bd047 build_scan.py: Add scripts/master to path (hinoka@google.com)

TBR=iannucci@chromium.org

Change-Id: Ia2d6af47decbe6dee57071f5606c898cf517d919
Recipe-Tryjob-Bypass-Reason: Autoroller
Bugdroid-Send-Email: False
Reviewed-on: https://skia-review.googlesource.com/10333
Reviewed-by: Recipe Roller <recipe-roller@chromium.org>
Commit-Queue: Recipe Roller <recipe-roller@chromium.org>

Roll recipe dependencies (trivial).

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/bb49d6609717078440ca257a1960cd34033bd047 build_scan.py: Add scripts/master to path (hinoka@google.com)

TBR=iannucci@chromium.org

Change-Id: I82c58c585d6a3e470aee13b788c32c52cc6e6297
Recipe-Tryjob-Bypass-Reason: Autoroller
Bugdroid-Send-Email: False
Reviewed-on: https://skia-review.googlesource.com/10332
Reviewed-by: Recipe Roller <recipe-roller@chromium.org>
Commit-Queue: Recipe Roller <recipe-roller@chromium.org>

Roll recipe dependencies (trivial).

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/d67cc72477ac9cf51dac9ff5aa56fd5f996f09d9 Roll recipe dependencies (trivial). (recipe-roller@chromium.org)
  https://crrev.com/720ae5cfa1f63afa71fa2e753a3d43cc0b166b28 [recipes.cfg] set autoroll recipe options (iannucci@chromium.org)
depot_tools:
  https://crrev.com/b2e5564353be2974b1241d1bd57b3259100b14ea [recipes.cfg] set autoroll recipe options (iannucci@chromium.org)

TBR=borenet@google.com

Change-Id: I5f5f7f8186224fb35c5fb34df07a79616c08aac7
Recipe-Tryjob-Bypass-Reason: Autoroller
Bugdroid-Send-Email: False
Reviewed-on: https://skia-review.googlesource.com/10331
Reviewed-by: Recipe Roller <recipe-roller@chromium.org>
Commit-Queue: Recipe Roller <recipe-roller@chromium.org>

Roll recipe dependencies (trivial).

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/d67cc72477ac9cf51dac9ff5aa56fd5f996f09d9 Roll recipe dependencies (trivial). (recipe-roller@chromium.org)
  https://crrev.com/720ae5cfa1f63afa71fa2e753a3d43cc0b166b28 [recipes.cfg] set autoroll recipe options (iannucci@chromium.org)
depot_tools:
  https://crrev.com/b2e5564353be2974b1241d1bd57b3259100b14ea [recipes.cfg] set autoroll recipe options (iannucci@chromium.org)

TBR=borenet@google.com

Change-Id: Ie6076adc7f44992cf63ee7ed86f70896a6cec5b7
Recipe-Tryjob-Bypass-Reason: Autoroller
Bugdroid-Send-Email: False
Reviewed-on: https://skia-review.googlesource.com/10330
Reviewed-by: Recipe Roller <recipe-roller@chromium.org>
Commit-Queue: Recipe Roller <recipe-roller@chromium.org>

Add debug helper for finding leaks in GrMemoryPool.h

Change-Id: I1d11769f9e99606af5e7752b3a6d055a5bf6a48d
Reviewed-on: https://skia-review.googlesource.com/10289
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>

Roll recipe dependencies (trivial).

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/d67cc72477ac9cf51dac9ff5aa56fd5f996f09d9 Roll recipe dependencies (trivial). (recipe-roller@chromium.org)
  https://crrev.com/720ae5cfa1f63afa71fa2e753a3d43cc0b166b28 [recipes.cfg] set autoroll recipe options (iannucci@chromium.org)
depot_tools:
  https://crrev.com/b2e5564353be2974b1241d1bd57b3259100b14ea [recipes.cfg] set autoroll recipe options (iannucci@chromium.org)

TBR=borenet@google.com

Change-Id: Ie827fe882fc5ec16a539b590d1b76575766118a1
Recipe-Tryjob-Bypass-Reason: Autoroller
Bugdroid-Send-Email: False
Reviewed-on: https://skia-review.googlesource.com/10329
Reviewed-by: Recipe Roller <recipe-roller@chromium.org>
Commit-Queue: Recipe Roller <recipe-roller@chromium.org>

Roll recipe dependencies (trivial).

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/d67cc72477ac9cf51dac9ff5aa56fd5f996f09d9 Roll recipe dependencies (trivial). (recipe-roller@chromium.org)
  https://crrev.com/720ae5cfa1f63afa71fa2e753a3d43cc0b166b28 [recipes.cfg] set autoroll recipe options (iannucci@chromium.org)
depot_tools:
  https://crrev.com/b2e5564353be2974b1241d1bd57b3259100b14ea [recipes.cfg] set autoroll recipe options (iannucci@chromium.org)

TBR=borenet@google.com

Change-Id: I8e4aea83c0d89e6933d1563bd790bd85cc5e33f2
Recipe-Tryjob-Bypass-Reason: Autoroller
Bugdroid-Send-Email: False
Reviewed-on: https://skia-review.googlesource.com/10328
Reviewed-by: Recipe Roller <recipe-roller@chromium.org>
Commit-Queue: Recipe Roller <recipe-roller@chromium.org>

Fix msan errors in webp decoder

Bug: skia:
Change-Id: I919a24fc61137c4a30160c97ad69fa4a2eadd031
Reviewed-on: https://skia-review.googlesource.com/10325
Reviewed-by: Matt Sarett <msarett@google.com>
Commit-Queue: Matt Sarett <msarett@google.com>

Add extra_asmflags

This makes it easier to cross-compile for non-Android ARM (e.g. Raspberry PI)

Bug: skia:
Change-Id: I4414d933bc38d56009aefcbe10e68f7c70964d9c
Reviewed-on: https://skia-review.googlesource.com/10291
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Mike Klein <mtklein@chromium.org>

[recipes.cfg] set autoroll recipe options

Bug: skia:
Change-Id: Icdc01ffba00dfda37bcb467f2a4a33aef0991f56
Reviewed-on: https://skia-review.googlesource.com/10245
Reviewed-by: Eric Boren <borenet@google.com>
Commit-Queue: Eric Boren <borenet@google.com>