Bug: skia:6916
Change-Id: I16badf80c3b34e517b8baab161150c9434f325aa
Reviewed-on: https://skia-review.googlesource.com/30100
Commit-Queue: Ravi Mistry <rmistry@google.com>
Reviewed-by: Eric Boren <borenet@google.com>
Like _lowp.cpp, it's only meant to be compiled offline.
Change-Id: I0d4f7c1fd8fa880ffd084c1e332f6a33def6e26f
Reviewed-on: https://skia-review.googlesource.com/30262
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
I think we can replace a lot of legacy code with an SkRasterPipeline
backend that works in 8-bit and stays interlaced. Think of this as a
"lowerp" replacement for lowp.
I'm having some trouble getting ARMv8 working.
ARMv7 should be fine, but I want to turn it on separately from x86.
I haven't looked at 32-bit x86 yet, but that's also on the todo list.
Open questions to follow up on:
- is it better to fold every multiply back down to 8-bit
(as seen here), or to allow intermediates to accumulate
in 16-bit and divide by 255 when done/needed?
- is it better pass tightly packed 8-bit vectors between stages (as
seen here), or to keep the 8-bit values unpacked in 16-bit lanes?
- should we make V wider than 1 register?
GMs look good. All diffs invisible and plausibly due to the 15->8 bit
precision drop. A quick bench run showed this running in about 0.75x
the time of the existing lowp backend.
Change-Id: I24aa46ff1d19c0b9b8dc192d5b1821cab0b8843c
Reviewed-on: https://skia-review.googlesource.com/29886
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Add a private API used by Android framework, which writes the clip
into a stencil buffer. This is used by HWUI to clip the WebView.
Bug: 31489986
Change-Id: I94515f1539acd9d069c8aceb3300577feed9c94f
Reviewed-on: https://skia-review.googlesource.com/29521
Commit-Queue: Stan Iliev <stani@google.com>
Reviewed-by: Derek Sollenberger <djsollen@google.com>
Change-Id: I34cf12b2aa16b2441b9e57162cbaee1444f42c52
Reviewed-on: https://skia-review.googlesource.com/29607
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Bug: skia:6857
Change-Id: Ibbbdd70483502d9e21429f86d3183e55bc007dd6
Reviewed-on: https://skia-review.googlesource.com/30101
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
This reverts commit 706a076879.
Reason for revert: this looks less a "fast path" and more "completely essential path" according to Chrome roll tests and some layout tests (most are small diffs, but some are radical).
Original change's description:
> remove another SkConvertPixels "fast path"
>
> SkColorSpaceXform now uses the same mechanism as our default path,
> SkRasterPipeline. There's no real reason to jump out to it as a
> special case any more.
>
> Change-Id: I19490de5b331267209cf117534942fb175c624c9
> Reviewed-on: https://skia-review.googlesource.com/29900
> Reviewed-by: Brian Osman <brianosman@google.com>
> Commit-Queue: Mike Klein <mtklein@chromium.org>
TBR=mtklein@chromium.org,brianosman@google.com
Change-Id: I92b17fe0d44e609d8c06e8fa2933f1f572a98094
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/30160
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Bug: skia:6890
Change-Id: I4534e6d2794e0b66ea8e21f3148772d14ea97de8
Reviewed-on: https://skia-review.googlesource.com/29562
Reviewed-by: Eric Boren <borenet@google.com>
Commit-Queue: Ben Wagner <benjaminwagner@google.com>
Raster pipeline FTW!
Change-Id: Ia2e600cf6c6e780f5a1519374c178197cf229776
Reviewed-on: https://skia-review.googlesource.com/29884
Commit-Queue: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Mike Klein <mtklein@chromium.org>
This reverts commit cc8eb60c48.
Reason for revert: Chrome change landed that should fix chrome roll
Original change's description:
> Revert "Revert "Revert "Add support for semaphores to be inserted on GrContext flush"""
>
> This reverts commit 876aed8758.
>
> Reason for revert: the bots seem to be unhappily red with this CL
>
> Original change's description:
> > Revert "Revert "Add support for semaphores to be inserted on GrContext flush""
> >
> > This reverts commit 8724b46099.
> >
> > Reason for revert: Creating a test CL to see what happens on the bots
> >
> > Original change's description:
> > > Revert "Add support for semaphores to be inserted on GrContext flush"
> > >
> > > This reverts commit cd1416efbc.
> > >
> > > Reason for revert: speculative, to try to fix roll see gpu_tests.pixel_integration_test.PixelIntegrationTest.Pixel_GpuRasterization_ConcavePaths
> > >
> > > Original change's description:
> > > > Add support for semaphores to be inserted on GrContext flush
> > > >
> > > > This also moves the logic of inserting semaphores down into GrDrawingManager
> > > > and finishFlush on GrGpu. With it being on finishFlush, there should be no
> > > > issues when the DrawingManager starts respecting the proxy passed in assuming
> > > > it always calls finishFlush at the end (which it should).
> > > >
> > > > Bug: skia:
> > > > Change-Id: I925c2a289dcbbb9159b9120878af1d34f21a2dc7
> > > > Reviewed-on: https://skia-review.googlesource.com/25641
> > > > Reviewed-by: Brian Salomon <bsalomon@google.com>
> > > > Commit-Queue: Greg Daniel <egdaniel@google.com>
> > >
> > > TBR=egdaniel@google.com,bsalomon@google.com,robertphillips@google.com
> > >
> > > Change-Id: I9c5b9cf8c060193e1861dbb8f0c10fb11dfb5249
> > > No-Presubmit: true
> > > No-Tree-Checks: true
> > > No-Try: true
> > > Bug: skia:
> > > Reviewed-on: https://skia-review.googlesource.com/25980
> > > Reviewed-by: Mike Reed <reed@google.com>
> > > Commit-Queue: Mike Reed <reed@google.com>
> >
> > TBR=egdaniel@google.com,bsalomon@google.com,robertphillips@google.com,reed@google.com
> >
> > # Not skipping CQ checks because original CL landed > 1 day ago.
> >
> > Bug: skia:
> > Change-Id: I5edbeaa0769670ee58f362f0ccaa78319410aa6c
> > Reviewed-on: https://skia-review.googlesource.com/26160
> > Reviewed-by: Brian Salomon <bsalomon@google.com>
> > Commit-Queue: Greg Daniel <egdaniel@google.com>
>
> TBR=egdaniel@google.com,bsalomon@google.com,robertphillips@google.com,reed@google.com
>
> Change-Id: I22fd6febafe70489a5fdb695c6f4263368eb423d
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Bug: skia:
> Reviewed-on: https://skia-review.googlesource.com/29422
> Reviewed-by: Yuqian Li <liyuqian@google.com>
> Commit-Queue: Yuqian Li <liyuqian@google.com>
TBR=egdaniel@google.com,bsalomon@google.com,robertphillips@google.com,liyuqian@google.com,reed@google.com
Change-Id: Ie3eae818b02599a70f714ef6b6635ce7d171bde6
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:
Reviewed-on: https://skia-review.googlesource.com/30000
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
perf.skia.org showed that reducing the threshold would only harm
path_fill_small_sawtooth, but will benefit many many skps and svgs
playback.
Bug: skia:
Change-Id: I38904548206521d78a7c9c5804c0d989b23dc405
TBR: caryclark@google.com, reed@google.com
Reviewed-on: https://skia-review.googlesource.com/29882
Reviewed-by: Yuqian Li <liyuqian@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
SkColorSpaceXform now uses the same mechanism as our default path,
SkRasterPipeline. There's no real reason to jump out to it as a
special case any more.
Change-Id: I19490de5b331267209cf117534942fb175c624c9
Reviewed-on: https://skia-review.googlesource.com/29900
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Expand the existing GM with a column "stretching" the [0,1] color stop
domain onto something much wider than [0,360].
Change-Id: Ica4c4b40c2cad20fec12b7d2871354e119a7bcf2
Reviewed-on: https://skia-review.googlesource.com/29880
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Does the transform before handling curves. This way all curves can be
carefully chopped in device space in order two avoid msaa.
Bug: skia:
Change-Id: I0508668142cda3fe3fda41f8475c408288aa4a47
Reviewed-on: https://skia-review.googlesource.com/29720
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Expand the sweep gradient definition to include a color stop angular range ([0, 360] by default).
Color stop positions in [0,1] are mapped to this range, and drawing outside is controlled by a
tile mode param.
This is closer to the CSS gradients spec and allows us to use fewer color stops in Blink conic
gradients.
Impl-wise, the remapping is effected after t calculation, and before tiling.
Change-Id: I5d71be01d134404d6eb9d7e2a904ec636b39f855
Reviewed-on: https://skia-review.googlesource.com/27704
Commit-Queue: Florin Malita <fmalita@chromium.org>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: Mike Klein <mtklein@chromium.org>
This reverts commit 93061b5344.
Reason for revert: bot failures
Original change's description:
> support for 'half' types in sksl, plus general numeric type improvements
>
> Bug: skia:
> Change-Id: Id285262fda8291847f11343d499b5df62ddb4b09
> Reviewed-on: https://skia-review.googlesource.com/28980
> Reviewed-by: Brian Salomon <bsalomon@google.com>
> Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
TBR=bsalomon@google.com,ethannicholas@google.com
Change-Id: Ie8672271d35b9fcdf567f8bc3674084748be66ad
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:
Reviewed-on: https://skia-review.googlesource.com/29600
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
If we were doing this math with real numbers or even just doubles, these
clamps wouldn't be necessary. But we're favoring speed over accuracy
here when we emulate fmod() and some of those inaccuracies end up with
values outside the [0,tile) range, negative!
To keep the spirit of fast over 100% accurate, I've just added a safety
clamp to 0. The case in the unit test now returns 0 where it should
really return something like 7 or 8, but at least we won't try to read
_way_ outside the image buffer.
BUG=chromium:749260
Change-Id: Ifc5cfe69798beccbb2a16547510158576e06eb3a
Reviewed-on: https://skia-review.googlesource.com/29580
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Bug: angle:2098 skia:6857
Change-Id: I502f05cdeb56514db4e144fdbb6d6f5a6f476b2e
Reviewed-on: https://skia-review.googlesource.com/29520
Reviewed-by: Yuqian Li <liyuqian@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
A new fuzzer appears to be complaining about using uninitialized glyph
ids. These uninitilized glyph ids appear to be comming from far up the
stack, but they don't actually get used until much later. If Skia is
passed uninitialized memory in a draw call it will of course eventually
need to use it and be blamed when it does. This change will make it
obvious if the issue is up stack from Skia.
BUG=chromium:750070,chromium:750071,chromium:750072
Change-Id: Ic6ca2f6af3620ad4a31cb017570f42550360891a
Reviewed-on: https://skia-review.googlesource.com/29421
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Ben Wagner <bungeman@google.com>
Change-Id: I8108f48bb55cfb39532ef05e8e263769ff8fb10f
Reviewed-on: https://skia-review.googlesource.com/29420
Reviewed-by: Yuqian Li <liyuqian@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>
This reverts commit 876aed8758.
Reason for revert: the bots seem to be unhappily red with this CL
Original change's description:
> Revert "Revert "Add support for semaphores to be inserted on GrContext flush""
>
> This reverts commit 8724b46099.
>
> Reason for revert: Creating a test CL to see what happens on the bots
>
> Original change's description:
> > Revert "Add support for semaphores to be inserted on GrContext flush"
> >
> > This reverts commit cd1416efbc.
> >
> > Reason for revert: speculative, to try to fix roll see gpu_tests.pixel_integration_test.PixelIntegrationTest.Pixel_GpuRasterization_ConcavePaths
> >
> > Original change's description:
> > > Add support for semaphores to be inserted on GrContext flush
> > >
> > > This also moves the logic of inserting semaphores down into GrDrawingManager
> > > and finishFlush on GrGpu. With it being on finishFlush, there should be no
> > > issues when the DrawingManager starts respecting the proxy passed in assuming
> > > it always calls finishFlush at the end (which it should).
> > >
> > > Bug: skia:
> > > Change-Id: I925c2a289dcbbb9159b9120878af1d34f21a2dc7
> > > Reviewed-on: https://skia-review.googlesource.com/25641
> > > Reviewed-by: Brian Salomon <bsalomon@google.com>
> > > Commit-Queue: Greg Daniel <egdaniel@google.com>
> >
> > TBR=egdaniel@google.com,bsalomon@google.com,robertphillips@google.com
> >
> > Change-Id: I9c5b9cf8c060193e1861dbb8f0c10fb11dfb5249
> > No-Presubmit: true
> > No-Tree-Checks: true
> > No-Try: true
> > Bug: skia:
> > Reviewed-on: https://skia-review.googlesource.com/25980
> > Reviewed-by: Mike Reed <reed@google.com>
> > Commit-Queue: Mike Reed <reed@google.com>
>
> TBR=egdaniel@google.com,bsalomon@google.com,robertphillips@google.com,reed@google.com
>
> # Not skipping CQ checks because original CL landed > 1 day ago.
>
> Bug: skia:
> Change-Id: I5edbeaa0769670ee58f362f0ccaa78319410aa6c
> Reviewed-on: https://skia-review.googlesource.com/26160
> Reviewed-by: Brian Salomon <bsalomon@google.com>
> Commit-Queue: Greg Daniel <egdaniel@google.com>
TBR=egdaniel@google.com,bsalomon@google.com,robertphillips@google.com,reed@google.com
Change-Id: I22fd6febafe70489a5fdb695c6f4263368eb423d
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:
Reviewed-on: https://skia-review.googlesource.com/29422
Reviewed-by: Yuqian Li <liyuqian@google.com>
Commit-Queue: Yuqian Li <liyuqian@google.com>
Bug: skia:
Change-Id: I0d421874e7ab71a39e2708575314f5b3615882e4
Reviewed-on: https://skia-review.googlesource.com/29221
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
cce8965d2c..878c8b1e5e
ANGLE now crashes (on program compilation) if there isn't a flush between uses of different flavors of ANGLE context (e.g., angle_gl_es2 vs. angle_gl_es3).
Change-Id: If59b6ec683e682db5214bb002a70863cee5fe013
Reviewed-on: https://skia-review.googlesource.com/28865
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
This reverts commit 8724b46099.
Reason for revert: Creating a test CL to see what happens on the bots
Original change's description:
> Revert "Add support for semaphores to be inserted on GrContext flush"
>
> This reverts commit cd1416efbc.
>
> Reason for revert: speculative, to try to fix roll see gpu_tests.pixel_integration_test.PixelIntegrationTest.Pixel_GpuRasterization_ConcavePaths
>
> Original change's description:
> > Add support for semaphores to be inserted on GrContext flush
> >
> > This also moves the logic of inserting semaphores down into GrDrawingManager
> > and finishFlush on GrGpu. With it being on finishFlush, there should be no
> > issues when the DrawingManager starts respecting the proxy passed in assuming
> > it always calls finishFlush at the end (which it should).
> >
> > Bug: skia:
> > Change-Id: I925c2a289dcbbb9159b9120878af1d34f21a2dc7
> > Reviewed-on: https://skia-review.googlesource.com/25641
> > Reviewed-by: Brian Salomon <bsalomon@google.com>
> > Commit-Queue: Greg Daniel <egdaniel@google.com>
>
> TBR=egdaniel@google.com,bsalomon@google.com,robertphillips@google.com
>
> Change-Id: I9c5b9cf8c060193e1861dbb8f0c10fb11dfb5249
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Bug: skia:
> Reviewed-on: https://skia-review.googlesource.com/25980
> Reviewed-by: Mike Reed <reed@google.com>
> Commit-Queue: Mike Reed <reed@google.com>
TBR=egdaniel@google.com,bsalomon@google.com,robertphillips@google.com,reed@google.com
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: skia:
Change-Id: I5edbeaa0769670ee58f362f0ccaa78319410aa6c
Reviewed-on: https://skia-review.googlesource.com/26160
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
We use several symbols from this library... we currently fail to link.
(The other small changes in here are automatic from gn format.)
BUG=skia:6891
Change-Id: Iec6f5deceecdb61571827ebb502a9f7e7e4a4bef
Reviewed-on: https://skia-review.googlesource.com/29260
Commit-Queue: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Heather Miller <hcm@google.com>
This reverts commit df0e09feac.
Reason for revert: Experimental revert to see if this is blocking the roll
Original change's description:
> Remove origin field from GrSurface
>
> This mainly consists of rm origin from GrSurface and the wrapBackEnd*
> methods and then re-adding an explicit origin parameter to all the
> GrGpu methods that need it.
>
> Change-Id: Iabd79ae98b227b5b9409f3ab5bbcc48af9613c18
> Reviewed-on: https://skia-review.googlesource.com/26363
> Reviewed-by: Brian Salomon <bsalomon@google.com>
> Commit-Queue: Robert Phillips <robertphillips@google.com>
TBR=bsalomon@google.com,robertphillips@google.com
Change-Id: Id606aa01e84e2b83be71d833eefca477c1ad0d01
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/29220
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Note that this does not clean up the users of this function, which
may themselves be subject to other overflow issues.
BUG=chromium:728936
Change-Id: I3eaa7627c3b6ff49296ea2618a0157dacdc1d9cc
Reviewed-on: https://skia-review.googlesource.com/29121
Commit-Queue: Ben Wagner <bungeman@google.com>
Reviewed-by: Mike Klein <mtklein@chromium.org>
Reviewed-by: Ben Wagner <bungeman@google.com>
- add a Debug builder
- add Debug and Release CPU testers on GCE
CQ_INCLUDE_TRYBOTS=skia.primary:Build-Win-Clang-x86_64-Debug,Test-Win2k8-Clang-GCE-CPU-AVX2-x86_64-Debug,Test-Win2k8-Clang-GCE-CPU-AVX2-x86_64-Release
Change-Id: I4f2d7309b36a532683f68fbdde9c7a9aaaa861ea
Reviewed-on: https://skia-review.googlesource.com/29023
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
By default it emulates your installed cl.exe, but the bots don't have
one. I think the fallback is 2013, which causes all sorts of pre-C++11
problems.
CQ_INCLUDE_TRYBOTS=skia.primary:Build-Win-Clang-x86_64-Release
Change-Id: I2556abe68825e58762b4172d067ba6826de5c133
Reviewed-on: https://skia-review.googlesource.com/29021
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Currently just adding support functions to go back and forth between
GrPixelConfigs and MTLPixelFormats.
Bug: skia:
Change-Id: I01a7d6877ebed87b87090ac2b920fee45dc0e856
Reviewed-on: https://skia-review.googlesource.com/29080
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
Remove CanCombine and AreEqual (unused)
Remove isInitialized() now that pipelines are initialized at creation.
Change-Id: Ibacf81d1f879c7ef9ea91a8f471c3d1df3eb2fed
Reviewed-on: https://skia-review.googlesource.com/29020
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
