mtklein [Mon, 26 Sep 2016 13:29:18 +0000 (06:29 -0700)]
These calls to SkRandom are not sequenced.
This is why we see two versions of this GM: some compilers evaluate the
rand.nextRangeScalar() calls left to right, others right to left.
Remember this bug?
BUG=skia:3241
(https://codereview.chromium.org/
1157943006 fixed hittestpath last year.)
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2368263002
Review-Url: https://codereview.chromium.org/
2368263002
caryclark [Mon, 26 Sep 2016 12:36:58 +0000 (05:36 -0700)]
fix tiger b
The tiger tests have uncovered numerous bugs.
This CL fixes the last of them.
If a pair of curves do not intersect, but
have one or both ends very close to the opposite
curve, consider that an intersection.
TBR=reed@google.com
BUG=skia:5131
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2356363003
Review-Url: https://codereview.chromium.org/
2356363003
cblume [Mon, 26 Sep 2016 11:22:59 +0000 (04:22 -0700)]
Dash to use SkIsAlign2
Dash checks for alignment by actively aligning.
It should really just check for alignment.
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2366283002
Review-Url: https://codereview.chromium.org/
2366283002
brianosman [Sun, 25 Sep 2016 20:15:58 +0000 (13:15 -0700)]
Add SkColor4f serialization
Adjusted usage in color shader, and will also be using this
in gradients, soon.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2334123003
Review-Url: https://codereview.chromium.org/
2334123003
update-skps [Sun, 25 Sep 2016 09:15:17 +0000 (02:15 -0700)]
Update SKP version
Automatic commit by the RecreateSKPs bot.
TBR=
NO_MERGE_BUILDS
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2368053002
Review-Url: https://codereview.chromium.org/
2368053002
benjaminwagner [Sat, 24 Sep 2016 00:45:45 +0000 (17:45 -0700)]
Disable msaa on iHD530 bot.
BUG=skia:5792
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2356213005
Review-Url: https://codereview.chromium.org/
2356213005
mtklein [Fri, 23 Sep 2016 20:18:41 +0000 (13:18 -0700)]
GN: ANGLE
Angle's existing GN files only work in Chrome, so I've written a new one.
This won't work on Windows, but our GN build doesn't work on Windows anyway. So this CL is an attempt to get a ahead of that curve on ANGLE. It looks large but fairly straightforward.
Now working on Linux:
$ gn gen angle --args=skia_use_angle=true
$ ninja -C angle
$ angle/dm --config angle-gl --src gm -w dm-out
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2361983002
Review-Url: https://codereview.chromium.org/
2361983002
brianosman [Fri, 23 Sep 2016 20:04:05 +0000 (13:04 -0700)]
Change SkSpecialImage::makeSurface and makeTightSurface to take output
properties (color space), bounds, and (optional) alphaType.
We were being pretty inconsistent before. Raster was honoring all
components of the info. GPU was using the supplied color type, but
propagating the source's color space. All call sites were saying N32.
What we want to do is propagate the original device's color space, and
pick a good format from that. Rather than force all the clients to
jump through hoops constructing an SkImageInfo that meets our criteria,
just have them supply the few bits we care about, and do everything else
internally.
This also lets us always use RGBA on GPU, but N32 on raster.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2349373004
Committed: https://skia.googlesource.com/skia/+/
53c38087949252d27cde668368a3eeb59cc2eb00
Review-Url: https://codereview.chromium.org/
2349373004
Robert Phillips [Fri, 23 Sep 2016 18:23:22 +0000 (14:23 -0400)]
Add validation of RRects to SkValidatingReadBuffer
This comes from the Skia fuzzer where it is inverting the RRect's rect which causes trouble down the line.
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2591
Change-Id: I5c34105a47369492d2df99d39a4e29116060ad37
Reviewed-on: https://skia-review.googlesource.com/2591
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
bsalomon [Fri, 23 Sep 2016 19:09:16 +0000 (12:09 -0700)]
Don't compute path keys for volatile paths in GrShape.
Otherwise, we will compute cache keys for internally transformed paths that don't repeat (e.g. clip paths transformed into device space with a changing view matrix).
BUG=chromium:649562
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2369513002
Review-Url: https://codereview.chromium.org/
2369513002
csmartdalton [Fri, 23 Sep 2016 18:36:11 +0000 (11:36 -0700)]
Add Pixel C knobs to skpbench
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2369533002
Review-Url: https://codereview.chromium.org/
2369533002
jvanverth [Fri, 23 Sep 2016 17:30:04 +0000 (10:30 -0700)]
Some Vulkan memory fixes and cleanup
* Switch back to not setting transfer_dst on all buffers
* Add some missing unit tests
* Add tracking of heap usage for debugging purposes
* Fall back to non-device-local memory if device-local allocation fails
BUG=skia:5031
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2356343003
Committed: https://skia.googlesource.com/skia/+/
c5850e9fdb62cc4ae5ed2b6af51aea92cac07455
Review-Url: https://codereview.chromium.org/
2356343003
Mike Reed [Fri, 23 Sep 2016 15:43:54 +0000 (11:43 -0400)]
change SkXfermodeImageFilter to carry no impl information
BUG=skia:
NOTREECHECKS=True
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2584
Change-Id: Ibf2dfa7722348ae71a99f1c895de0d2996350e58
Reviewed-on: https://skia-review.googlesource.com/2584
Commit-Queue: Mike Reed <reed@google.com>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Reviewed-by: Robert Phillips <robertphillips@google.com>
caryclark [Fri, 23 Sep 2016 16:32:26 +0000 (09:32 -0700)]
fix msan bug in pathops
Msan and Valgrind found an uninitialized memory mistake in
pathops. This also fixes similar bugs where not all parts
of the geometry were covered in the loop iteration.
R=borenet@google.com
NOTREECHECKS=true
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2366893003
Review-Url: https://codereview.chromium.org/
2366893003
Mike Klein [Fri, 23 Sep 2016 15:10:08 +0000 (11:10 -0400)]
Include GLTestContext_command_buffer.cpp in G3 build.
This is now required to build our GPU testing tools.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2587
NOTREECHECKS=true
Change-Id: I219665fbe50e9bbea9abb88a6d63c34b76ca92e6
Reviewed-on: https://skia-review.googlesource.com/2587
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Ben Wagner <benjaminwagner@google.com>
brianosman [Fri, 23 Sep 2016 15:49:50 +0000 (08:49 -0700)]
Revert of Create special surfaces according to original device (not always in N32) (patchset #9 id:160001 of https://codereview.chromium.org/
2349373004/ )
Reason for revert:
DM crash and/or TSAN failure
Original issue's description:
> Change SkSpecialImage::makeSurface and makeTightSurface to take output
> properties (color space), bounds, and (optional) alphaType.
>
> We were being pretty inconsistent before. Raster was honoring all
> components of the info. GPU was using the supplied color type, but
> propagating the source's color space. All call sites were saying N32.
>
> What we want to do is propagate the original device's color space, and
> pick a good format from that. Rather than force all the clients to
> jump through hoops constructing an SkImageInfo that meets our criteria,
> just have them supply the few bits we care about, and do everything else
> internally.
>
> This also lets us always use RGBA on GPU, but N32 on raster.
>
> BUG=skia:
> GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2349373004
>
> Committed: https://skia.googlesource.com/skia/+/
53c38087949252d27cde668368a3eeb59cc2eb00
TBR=robertphillips@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/
2366723004
Mike Reed [Fri, 23 Sep 2016 14:05:05 +0000 (10:05 -0400)]
use SkCanvas::ClipOp
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2583
Change-Id: I651e2c278bb4d2df08b6e134c734c5403c7e7ff7
Reviewed-on: https://skia-review.googlesource.com/2583
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Mike Reed <reed@google.com>
bungeman [Fri, 23 Sep 2016 15:16:04 +0000 (08:16 -0700)]
Json ide script for gn to output cmake.
Can be run like
gn gen out/debug --ide=json --json-ide-script=../../gn/gn_to_cmake.py
or
gn gen out/debug --ide=json
python gn/gn_to_cmake.py out/debug/project.json
This first pass creates CMakeLists.txt with just enough information for an
ide. Each target is just a static library with sources, include
directories, and defines.
NOTRY=true
This isn't automatically tested.
Review-Url: https://codereview.chromium.org/
2363873002
brianosman [Fri, 23 Sep 2016 15:11:55 +0000 (08:11 -0700)]
Change SkSpecialImage::makeSurface and makeTightSurface to take output
properties (color space), bounds, and (optional) alphaType.
We were being pretty inconsistent before. Raster was honoring all
components of the info. GPU was using the supplied color type, but
propagating the source's color space. All call sites were saying N32.
What we want to do is propagate the original device's color space, and
pick a good format from that. Rather than force all the clients to
jump through hoops constructing an SkImageInfo that meets our criteria,
just have them supply the few bits we care about, and do everything else
internally.
This also lets us always use RGBA on GPU, but N32 on raster.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2349373004
Review-Url: https://codereview.chromium.org/
2349373004
brianosman [Fri, 23 Sep 2016 14:11:34 +0000 (07:11 -0700)]
Make GrRenderableConfigForColorSpace argument const
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2367813002
Review-Url: https://codereview.chromium.org/
2367813002
borenet [Fri, 23 Sep 2016 13:37:57 +0000 (06:37 -0700)]
Add swarming task for nanobench upload
BUG=skia:5719
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2364693003
Review-Url: https://codereview.chromium.org/
2364693003
caryclark [Fri, 23 Sep 2016 12:47:20 +0000 (05:47 -0700)]
split tight quads and conics
Tight quads and conics may nearly fold over on themselves, confusing
coincidence against other curves. Split them at their max curvature
early on to avoid complicating later logic.
TBR=reed@google.com
BUG=skia:5131
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2357353002
Review-Url: https://codereview.chromium.org/
2357353002
brianosman [Thu, 22 Sep 2016 21:50:58 +0000 (14:50 -0700)]
sRGB cleanup in image filters. Ensure we configure paints correctly.
Net effect of the two calls is (basically) the same, but given that we're
propagating from isGammaCorrect on the DC, I figured it makes sense to use
the same-named API.
More importantly, a couple places had slipped through, so those draws would
ignore sRGB-ness of inputs entirely.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2357413004
Review-Url: https://codereview.chromium.org/
2357413004
brianosman [Thu, 22 Sep 2016 21:27:34 +0000 (14:27 -0700)]
It is possible to try and draw to an unsupported format. Let this go.
Squelches assert firing in gradient code on low-end Android bots.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2365703002
Review-Url: https://codereview.chromium.org/
2365703002
halcanary [Thu, 22 Sep 2016 21:13:16 +0000 (14:13 -0700)]
SkPDF: do not assert when openStream return nullptr
BUG=163892
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2332813002
Review-Url: https://codereview.chromium.org/
2332813002
reed [Thu, 22 Sep 2016 21:12:46 +0000 (14:12 -0700)]
document.close from bool to void
Nothing meaningful is returned, it is inconsistent with the bulk of the rest of skia's api. The C api is waiting on this change as well.
Required chrome CL: https://codereview.chromium.org/
2355343003/
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2354403002
Review-Url: https://codereview.chromium.org/
2354403002
jvanverth [Thu, 22 Sep 2016 20:45:24 +0000 (13:45 -0700)]
Revert of Some Vulkan memory fixes and cleanup (patchset #2 id:20001 of https://codereview.chromium.org/
2356343003/ )
Reason for revert:
fHeapIndex is not used in release, need to hide behind SK_DEBUG. Failing on Perf-Android-Clang-NVIDIA_Shield-GPU-TegraX1-arm64-Release-GN_Android_Vulkan.
Original issue's description:
> Some Vulkan memory fixes and cleanup
>
> * Switch back to not setting transfer_dst on all buffers
> * Add some missing unit tests
> * Add tracking of heap usage for debugging purposes
> * Fall back to non-device-local memory if device-local allocation fails
>
> BUG=skia:5031
> GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2356343003
>
> Committed: https://skia.googlesource.com/skia/+/
c5850e9fdb62cc4ae5ed2b6af51aea92cac07455
TBR=egdaniel@google.com,brianosman@google.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:5031
Review-Url: https://codereview.chromium.org/
2358123004
msarett [Thu, 22 Sep 2016 20:23:56 +0000 (13:23 -0700)]
G instead of R, thanks Gold
NOTRY=true
TBR=mtklein@google.com
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2362863002
Review-Url: https://codereview.chromium.org/
2362863002
caryclark [Thu, 22 Sep 2016 19:52:20 +0000 (12:52 -0700)]
don't close a contour with a line if nothing's there
R=kjlubick@google.com
BUG=skia:5789
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2360083006
Review-Url: https://codereview.chromium.org/
2360083006
jvanverth [Thu, 22 Sep 2016 19:47:35 +0000 (12:47 -0700)]
Some Vulkan memory fixes and cleanup
* Switch back to not setting transfer_dst on all buffers
* Add some missing unit tests
* Add tracking of heap usage for debugging purposes
* Fall back to non-device-local memory if device-local allocation fails
BUG=skia:5031
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2356343003
Review-Url: https://codereview.chromium.org/
2356343003
hcm [Thu, 22 Sep 2016 19:47:14 +0000 (12:47 -0700)]
Update Chrome section to include Skia roll/sheriff/bug info
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2351593003
NOTRY=true
DOCS_PREVIEW= https://skia.org/?cl=
2351593003
Review-Url: https://codereview.chromium.org/
2351593003
robertphillips [Thu, 22 Sep 2016 19:42:30 +0000 (12:42 -0700)]
Make AALinearizingConvexPathRenderer able to handle stroke and fill
This is intended to catch stroke-and-fill convex paths with potentially small stroke widths (e.g., .1).
It does have the disconcerting side effect of changing bevel-joined stroke-and-filled degenerate-single-line-rects into plain rects (w/o triangular end-caps).
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2301353004
Committed: https://skia.googlesource.com/skia/+/
522bcd99fa65a8abd130880f59b500cf367d0845
Review-Url: https://codereview.chromium.org/
2301353004
bsalomon [Thu, 22 Sep 2016 19:42:11 +0000 (12:42 -0700)]
Change implementation of flush-count based GrGpuResource purging
Change default to approx 30seconds (given some API usage assumptions)
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2361093002
Review-Url: https://codereview.chromium.org/
2361093002
csmartdalton [Thu, 22 Sep 2016 19:37:21 +0000 (12:37 -0700)]
skpbench fixes for linux
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2354273004
Review-Url: https://codereview.chromium.org/
2354273004
msarett [Thu, 22 Sep 2016 19:37:04 +0000 (12:37 -0700)]
Add BGRA as input format to SkColorSpaceXform
This is immediately useful for webp and I think it's a fair guess
that BGRA src formats are not uncommon.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2353363008
Review-Url: https://codereview.chromium.org/
2353363008
bungeman [Thu, 22 Sep 2016 19:32:07 +0000 (12:32 -0700)]
Remove assert that current color is never used.
The COLR table format allows for use of the 'current color'.
Currently the code asserts if this is used, but will use a nearby
luminance grey to this color instead if the assert is ignored.
Remove the assert now that this path is known to be used.
BUG=skia:5788
Review-Url: https://codereview.chromium.org/
2359923004
brianosman [Thu, 22 Sep 2016 19:31:58 +0000 (12:31 -0700)]
Support for color-spaces with multi-stop (texture) gradients
Texture is F16 linear, unless that's not supported. In that
case, we pack down to sRGB.
Added more test patches to the gamut GM with many stops,
to test this case. Now they render correctly.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2343253002
Review-Url: https://codereview.chromium.org/
2343253002
mtklein [Thu, 22 Sep 2016 18:51:24 +0000 (11:51 -0700)]
GN: build skiaserve
I trimmed the libmicrohttpd sources and defines down to the minimum needed to build and run. This builds and runs on Linux and Android for me.
Request.h was missing an include for SkTypes.h, which supplies the default for SK_GPU_SUPPORTED if not otherwise defined.
To build on Android, exit() -> _exit().
build.py was unused.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2367513002
NOTREECHECKS=true
Review-Url: https://codereview.chromium.org/
2367513002
caryclark [Thu, 22 Sep 2016 17:24:59 +0000 (10:24 -0700)]
Reland of ix for conic fuzz (patchset #1 id:1 of https://codereview.chromium.org/
2361473004/ )
Reason for revert:
Landed suppression in Chrome's LayoutTests/TestExpectations
Original issue's description:
> Revert of fix for conic fuzz (patchset #3 id:40001 of https://codereview.chromium.org/
2350263003/ )
>
> Reason for revert:
> See if this fixes the layout tests.
>
> Original issue's description:
> > fix for conic fuzz
> >
> > A fuzzer generates a conic that hangs when drawn.
> > The quads that approximate the conics move up and down
> > in y, confusing the renderer.
> >
> > This fix ensures that the split conic maintains the
> > same y direction as the original conic.
> >
> > R=reed@google.com
> > BUG=647922
> > GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2350263003
> >
> > Committed: https://skia.googlesource.com/skia/+/
ac78863acdef4b428aaf66985b80c76d1be0fdea
>
> TBR=reed@google.com
> # Skipping CQ checks because original CL landed less than 1 days ago.
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
> BUG=647922
>
> Committed: https://skia.googlesource.com/skia/+/
08b345588414b861af8a55950e7dc21a1bd85a28
TBR=reed@google.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=647922
Review-Url: https://codereview.chromium.org/
2359253002
caryclark [Thu, 22 Sep 2016 14:42:39 +0000 (07:42 -0700)]
Revert of fix for conic fuzz (patchset #3 id:40001 of https://codereview.chromium.org/
2350263003/ )
Reason for revert:
See if this fixes the layout tests.
Original issue's description:
> fix for conic fuzz
>
> A fuzzer generates a conic that hangs when drawn.
> The quads that approximate the conics move up and down
> in y, confusing the renderer.
>
> This fix ensures that the split conic maintains the
> same y direction as the original conic.
>
> R=reed@google.com
> BUG=647922
> GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2350263003
>
> Committed: https://skia.googlesource.com/skia/+/
ac78863acdef4b428aaf66985b80c76d1be0fdea
TBR=reed@google.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=647922
Review-Url: https://codereview.chromium.org/
2361473004
brianosman [Thu, 22 Sep 2016 14:15:37 +0000 (07:15 -0700)]
Add output format properties to SkImageFilter::Context
For now, this is just the color space (of the original
requesting device). This is used when constructing
intermediate rendering surfaces, so that we ensure we
land in a surface that's similar/compatible to the
final consumer of the DAG's output.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2357273002
Review-Url: https://codereview.chromium.org/
2357273002
msarett [Thu, 22 Sep 2016 14:02:24 +0000 (07:02 -0700)]
Make SkColorSpaceXform::New() take bare ptrs
There's no need to take sk_sp if we're not going to
ref the ptr.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2360863003
Review-Url: https://codereview.chromium.org/
2360863003
mtklein [Thu, 22 Sep 2016 13:50:24 +0000 (06:50 -0700)]
Run commandbuffer config on CommandBuffer Perf bots.
Uh, so hey, I just noticed that while the CommandBuffer Test bot runs
with --config commandbuffer, the Perf bot just runs as a vanilla GPU bot.
Shouldn't we pass --config commandbuffer?
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2364483002
CQ_INCLUDE_TRYBOTS=master.client.skia:Perf-Mac-Clang-MacMini6.2-GPU-HD4000-x86_64-Debug-CommandBuffer-Trybot
Review-Url: https://codereview.chromium.org/
2364483002
mtklein [Thu, 22 Sep 2016 13:01:17 +0000 (06:01 -0700)]
Clean up dead code.
SkOSEnvironment was used to set --enable-gl-path-rendering, and is now unused.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2365483002
Review-Url: https://codereview.chromium.org/
2365483002
cblume [Thu, 22 Sep 2016 12:25:26 +0000 (05:25 -0700)]
Is char* or uintptr_t easier to read?
Using a char* instead of uintptr_t allows us to use fewer
reinterpret_casts which may make the code easier to read.
BUG=648512
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2356703002
Review-Url: https://codereview.chromium.org/
2356703002
benjaminwagner [Thu, 22 Sep 2016 12:25:12 +0000 (05:25 -0700)]
Add recipe support for Intel HD Graphics 530.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2358183002
Review-Url: https://codereview.chromium.org/
2358183002
msarett [Thu, 22 Sep 2016 12:20:21 +0000 (05:20 -0700)]
Do not quickReject until virtual drawPatch
TBR=reed@google.com
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2358873003
Review-Url: https://codereview.chromium.org/
2358873003
caryclark [Thu, 22 Sep 2016 12:20:07 +0000 (05:20 -0700)]
fix next kevin fuzz
Fix one more fuzzer crash.
R=kjlubick@google.com
BUG=skia:5775
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2357373002
Review-Url: https://codereview.chromium.org/
2357373002
caryclark [Thu, 22 Sep 2016 12:15:14 +0000 (05:15 -0700)]
fix for conic fuzz
A fuzzer generates a conic that hangs when drawn.
The quads that approximate the conics move up and down
in y, confusing the renderer.
This fix ensures that the split conic maintains the
same y direction as the original conic.
R=reed@google.com
BUG=647922
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2350263003
Review-Url: https://codereview.chromium.org/
2350263003
caryclark [Thu, 22 Sep 2016 12:15:04 +0000 (05:15 -0700)]
speed up debug dm
SkPathMeasure can take minutes with pathological cases.
Limit the debug check to a reasonable number.
R=reed@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2356343004
Review-Url: https://codereview.chromium.org/
2356343004
csmartdalton [Thu, 22 Sep 2016 12:10:02 +0000 (05:10 -0700)]
Add hardware monitoring to skpbench
Adds a Hardware class with hooks for entering and exiting
"benchmarking" mode (e.g. locking clocks, etc.) as well as periodic
polling of hardware to verify the environment is stable.
Adds a partial implementation for generic Android hardware, but
ultimately we will need to write specific classes tailored to each
unique platform we need to test.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2360473002
Review-Url: https://codereview.chromium.org/
2360473002
borenet [Thu, 22 Sep 2016 12:09:56 +0000 (05:09 -0700)]
Add Jobs to tasks.json
BUG=skia:5723
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2363443002
Review-Url: https://codereview.chromium.org/
2363443002
mtklein [Wed, 21 Sep 2016 21:01:32 +0000 (14:01 -0700)]
GN: take over CommandBuffer bot
As you suspected, I see nothing preventing us from building our side of things unconditionally.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2358173002
NOTREECHECKS=true
Review-Url: https://codereview.chromium.org/
2358173002
bsalomon [Wed, 21 Sep 2016 18:23:46 +0000 (11:23 -0700)]
Use sk_careful_memcpy when writing optional conic weights for GrShape path data key.
TBR=robertphillips@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2359933002
NO_DEPENDENCY_CHECKS=true
Review-Url: https://codereview.chromium.org/
2359933002
bsalomon [Wed, 21 Sep 2016 18:16:05 +0000 (11:16 -0700)]
Add optional sw generated path coverage mask caching
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2335343008
All the bots except the painfully slow windows compiler have finished so,
NOTRY=true
Review-Url: https://codereview.chromium.org/
2335343008
reed [Wed, 21 Sep 2016 18:15:07 +0000 (11:15 -0700)]
allow clip calls w/o op param, remove unnecessary kReplace ops
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2357333002
Review-Url: https://codereview.chromium.org/
2357333002
robertphillips [Wed, 21 Sep 2016 18:00:23 +0000 (11:00 -0700)]
Remove unused SkImage_Base and SkImage_Gpu onNewSurface methods
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2353403002
Review-Url: https://codereview.chromium.org/
2353403002
bsalomon [Wed, 21 Sep 2016 17:53:24 +0000 (10:53 -0700)]
Conditionally insert gl_PointSize into shaders.
BUG=chromium:648816
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2358843002
Review-Url: https://codereview.chromium.org/
2358843002
jvanverth [Wed, 21 Sep 2016 17:20:23 +0000 (10:20 -0700)]
Always use transfer_dst for buffers.
This will allow Adreno to use non-coherent buffers.
BUG=skia:5034
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2361473002
Review-Url: https://codereview.chromium.org/
2361473002
mtklein [Wed, 21 Sep 2016 17:11:25 +0000 (10:11 -0700)]
Build tools on NoGPU bots.
I was just lazy disabling this before. It turns out to not be difficult.
This gets us slightly better testing out of the NoGPU bot, as we now know
not only that Skia compiles, but also that DM and nanobench can link (and
thus that Skia's got enough compiled in it to be a coherent library).
skpbench requires GPU support.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2356283002
Review-Url: https://codereview.chromium.org/
2356283002
brianosman [Wed, 21 Sep 2016 16:46:57 +0000 (09:46 -0700)]
Add a transient image filter cache to SkImage::makeWithFilter & PDF
Added a bench for makeWithFilter (~23 ms -> ~6 ms median locally).
Also fixed indentation.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2360663002
Review-Url: https://codereview.chromium.org/
2360663002
mtklein [Wed, 21 Sep 2016 16:14:19 +0000 (09:14 -0700)]
GN: is_skia_standalone
is_skia_standalone will be defined and set to true iff we're using our
standalone BUILDCONFIG.gn. Any other build (Fuchsia, Chrome, Pdfium)
will have their own BUILDCONFIG.gn, which presumably does not define
is_skia_standalone.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2357963002
Review-Url: https://codereview.chromium.org/
2357963002
caryclark [Wed, 21 Sep 2016 15:46:56 +0000 (08:46 -0700)]
fix skia pathops fuzzers
Add isolated tests.
R=kjlubick@google.com
BUG=skia:5775
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2358043002
Review-Url: https://codereview.chromium.org/
2358043002
bsalomon [Wed, 21 Sep 2016 15:26:57 +0000 (08:26 -0700)]
Make GrShape compute keys for short paths from path data instead of using the gen id.
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2357643002
Review-Url: https://codereview.chromium.org/
2357643002
bsalomon [Wed, 21 Sep 2016 14:47:34 +0000 (07:47 -0700)]
Stop closing filled paths in GrShape
This can force a copy of the path data and isn't really being taken advantage of. It complicates a forthcoming change to key small paths based on their data rather than gen id.
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2355833002
Committed: https://skia.googlesource.com/skia/+/
ad001fdc77b2dfe32c1723cb58636728dfb776d8
Review-Url: https://codereview.chromium.org/
2355833002
egdaniel [Wed, 21 Sep 2016 14:36:14 +0000 (07:36 -0700)]
Add define of SK_ENABLE_VK_LAYERS to gn build
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2357953002
Review-Url: https://codereview.chromium.org/
2357953002
liyuqian [Wed, 21 Sep 2016 14:11:08 +0000 (07:11 -0700)]
Add setBounds to SkLiteDL for Android
The RootRenderNode in Android is changing bounds dynamically. This is a
temporary fix to accomodate that behaviour. (See also b/
31304115)
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2358793002
Review-Url: https://codereview.chromium.org/
2358793002
sdefresne [Wed, 21 Sep 2016 13:51:33 +0000 (06:51 -0700)]
Fix memory leaks reported by clang static analyzer.
The CFURLCopyFileSystemPath & CFURLCopyFileSystemPath methods
respect the "Create Rule" [1] regarding the ownership of the
returned reference. This means that the objects need to be
deallocated explicitly by calling CFRelease.
[1]: https://developer.apple.com/library/content/documentation/CoreFoundation/Conceptual/CFMemoryMgmt/Concepts/Ownership.html#//apple_ref/doc/uid/
20001148-103029
BUG=648210
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2360573002
Review-Url: https://codereview.chromium.org/
2360573002
brianosman [Wed, 21 Sep 2016 13:45:09 +0000 (06:45 -0700)]
Fix some GPU image filter code to preserve precision and color space
On the pure-GPU path, we just have an SkSpecialImage (that's definitely
texture backed), and we need a renderable config for the draw context we
make. Added a helper function to pick - this is basically the high
precision analog of what we were doing before (always using 8888).
The assert that I added catches many other problems in image filter code,
but those fixes are coming in subsequent CLs.
12 GMs render correctly (or more correctly) in gpusrgb and gpuf16
configs. In most cases, they were drawing previously nothing.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2359443003
Review-Url: https://codereview.chromium.org/
2359443003
bungeman [Tue, 20 Sep 2016 22:46:36 +0000 (15:46 -0700)]
Remove SK_ADDING_src_ports_SkFontConfigInterface_cpp
This flag is no longer defined anywhere, so remove it and the code it
was guarding.
TBR=reed
This is just removing dead transitional code.
Review-Url: https://codereview.chromium.org/
2352163002
fmalita [Tue, 20 Sep 2016 22:45:57 +0000 (15:45 -0700)]
[SVGDom] Opacity optimization
Apply opacity as fill/stroke paint alpha instead of saveLayer, when
possible.
R=robertphillips@google.com,stephana@google.com,reed@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2353503005
Committed: https://skia.googlesource.com/skia/+/
3dbb7b9f196d793fbd16243157ee67638891f5dc
Review-Url: https://codereview.chromium.org/
2353503005
mtklein [Tue, 20 Sep 2016 22:41:29 +0000 (15:41 -0700)]
Blacklist serialize-8888 typefacerendering_pfaMac
BUG=skia:5778
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2356753003
TBR=halcanary@google.com
NOTREECHECKS=true
CQ_INCLUDE_TRYBOTS=master.client.skia:Test-Mac-Clang-MacMini6.2-CPU-AVX-x86_64-Debug-GN-Trybot
Review-Url: https://codereview.chromium.org/
2356753003
mtklein [Tue, 20 Sep 2016 22:11:46 +0000 (15:11 -0700)]
GN: fold :vulkan into :gpu.
Obviously, :vulkan can't be used without :gpu, but more subtly,
:gpu needs to see SK_VULKAN defined if we're using Vulkan.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2352923003
NOTREECHECKS=true
Review-Url: https://codereview.chromium.org/
2352923003
fmalita [Tue, 20 Sep 2016 21:06:57 +0000 (14:06 -0700)]
Revert of Stop closing filled paths in GrShape (patchset #1 id:1 of https://codereview.chromium.org/
2355833002/ )
Reason for revert:
Triggers msaa16 asserts: https://build.chromium.org/p/client.skia/builders/Test-Win10-MSVC-ShuttleA-GPU-GTX660-x86_64-Debug/builds/4236/steps/test_skia%20on%20Windows/logs/stdio
c:\b\work\skia\src\gpu\grmesh.h:76: fatal error: "assert(indexCount)"
Caught exception
2147483651 EXCEPTION_BREAKPOINT, was running:
msaa16 svg cartman.svg
Likely culprit:
msaa16 svg cartman.svg
Original issue's description:
> Stop closing filled paths in GrShape
>
> This can force a copy of the path data and isn't really being taken advantage of. It complicates a forthcoming change to key small paths based on their data rather than gen id.
> GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2355833002
>
> Committed: https://skia.googlesource.com/skia/+/
ad001fdc77b2dfe32c1723cb58636728dfb776d8
TBR=robertphillips@google.com,bsalomon@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/
2357713002
borenet [Tue, 20 Sep 2016 20:39:46 +0000 (13:39 -0700)]
Revert of [SVGDom] Opacity optimization (patchset #4 id:60001 of https://codereview.chromium.org/
2353503005/ )
Reason for revert:
Failing assertion
Original issue's description:
> [SVGDom] Opacity optimization
>
> Apply opacity as fill/stroke paint alpha instead of saveLayer, when
> possible.
>
> R=robertphillips@google.com,stephana@google.com,reed@google.com
> GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2353503005
>
> Committed: https://skia.googlesource.com/skia/+/
3dbb7b9f196d793fbd16243157ee67638891f5dc
TBR=reed@google.com,robertphillips@google.com,stephana@google.com,fmalita@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/
2359603002
mtklein [Tue, 20 Sep 2016 20:36:47 +0000 (13:36 -0700)]
GN: add skia_use_mesa for OSMesa in tools.
I tested by running
$ out/dm --src gm --config mesa
$ out/nanobench --config mesa --match Xfermode
... but I ran out of patience waiting for nanobench to finish. Probably works.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2357703002
Review-Url: https://codereview.chromium.org/
2357703002
borenet [Tue, 20 Sep 2016 20:32:37 +0000 (13:32 -0700)]
Attempt to fix nanobench upload
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2356813003
Review-Url: https://codereview.chromium.org/
2356813003
stani [Tue, 20 Sep 2016 20:24:49 +0000 (13:24 -0700)]
Revert of Clear stencil buffer before using it for drawing (patchset #2 id:20001 of https://codereview.chromium.org/
2342873004/ )
Reason for revert:
performance regression in Android macrobench tests
Original issue's description:
> Clear stencil buffer before using it for drawing
>
> Clear stencil buffer before using it for drawing. This is
> resolving an issue with a failing android CTS test in drawArc.
> GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2342873004
>
> Committed: https://skia.googlesource.com/skia/+/
c326a383e1069497884905f9dd3e720e17d8fcb5
TBR=bsalomon@google.com,reed@google.com
# Not skipping CQ checks because original CL landed more than 1 days ago.
Review-Url: https://codereview.chromium.org/
2355923002
jvanverth [Tue, 20 Sep 2016 20:18:00 +0000 (13:18 -0700)]
Fix VkWrapTests
BUG=skia:
TBR=bsalomon@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2352173002
Review-Url: https://codereview.chromium.org/
2352173002
halcanary [Tue, 20 Sep 2016 20:11:01 +0000 (13:11 -0700)]
GM: add test for type1 font
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2350453002
Review-Url: https://codereview.chromium.org/
2350453002
egdaniel [Tue, 20 Sep 2016 20:00:26 +0000 (13:00 -0700)]
Disable framebuffer fetch on the Adreno5xx devices.
This should help with some of the rendering bugs currently in GL for the
S7.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2348293004
Review-Url: https://codereview.chromium.org/
2348293004
egdaniel [Tue, 20 Sep 2016 19:57:45 +0000 (12:57 -0700)]
Fix possible divide by zero in vulkan granularity adjustments
This should fix the divide by zero crash on the nexus player bots
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2358613002
Review-Url: https://codereview.chromium.org/
2358613002
bsalomon [Tue, 20 Sep 2016 19:44:36 +0000 (12:44 -0700)]
Stop closing filled paths in GrShape
This can force a copy of the path data and isn't really being taken advantage of. It complicates a forthcoming change to key small paths based on their data rather than gen id.
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2355833002
Review-Url: https://codereview.chromium.org/
2355833002
fmalita [Tue, 20 Sep 2016 19:44:31 +0000 (12:44 -0700)]
[SVGDom] Opacity optimization
Apply opacity as fill/stroke paint alpha instead of saveLayer, when
possible.
R=robertphillips@google.com,stephana@google.com,reed@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2353503005
Review-Url: https://codereview.chromium.org/
2353503005
egdaniel [Tue, 20 Sep 2016 19:39:29 +0000 (12:39 -0700)]
Disable failing unit tests on adreno vulkan
As mentioned in the attached bug, these failures all seem to be caused
by driver issues that have no workarounds at this time.
BUG=skia:5777
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2355883002
Review-Url: https://codereview.chromium.org/
2355883002
egdaniel [Tue, 20 Sep 2016 19:39:25 +0000 (12:39 -0700)]
Use STDArray to for tracked resources in vulkan command buffer
This is a minor speed improvement to help with the number of times we
are freeing and re-allocating the tracked resource arrays
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2338963004
Review-Url: https://codereview.chromium.org/
2338963004
borenet [Tue, 20 Sep 2016 19:34:05 +0000 (12:34 -0700)]
Revert of Skia recipe: Use [depot_tools]/gsutil.py instead of [depot_tools]/third_party/gsutil/gsutil (patchset #1 id:1 of https://codereview.chromium.org/
2298293002/ )
Reason for revert:
Broke all nanobench uploads, eg. https://uberchromegw.corp.google.com/i/client.skia.android/builders/Perf-Android-Clang-Nexus6p-CPU-Snapdragon810-arm64-Release-GN_Android/builds/154/steps/Upload%20perf%20results/logs/stdio
Original issue's description:
> Skia recipe: Use [depot_tools]/gsutil.py instead of [depot_tools]/third_party/gsutil/gsutil
>
> The latter is being removed in https://codereview.chromium.org/
2280023003/
> Is also pinned to 3.25, whereas the former is updated more often.
>
> BUG=skia:
> GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2298293002
>
> Committed: https://skia.googlesource.com/skia/+/
f15590f803596f51d0950994b2aeb623d2d1fc24
TBR=hinoka@google.com
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review-Url: https://codereview.chromium.org/
2355943002
mtklein [Tue, 20 Sep 2016 19:28:12 +0000 (12:28 -0700)]
GN: add skia_enable_gpu (default true)
This lets us replace the NoGPU bot.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2356853002
NOTREECHECKS=true
Review-Url: https://codereview.chromium.org/
2356853002
mtklein [Tue, 20 Sep 2016 19:09:12 +0000 (12:09 -0700)]
GN: compiler_prefix -> cc_wrapper
I was a bit mixed up thinking Chrome called this feature compiler_prefix.
They actually call it cc_wrapper. So let's switch to their name before we
get too used to this.
No bots are affected.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2355003002
DOCS_PREVIEW= https://skia.org/?cl=
2355003002
NOTREECHECKS=true
Review-Url: https://codereview.chromium.org/
2355003002
mtklein [Tue, 20 Sep 2016 18:45:11 +0000 (11:45 -0700)]
Snazzier dump log step on Android bots.
(The failing NexusPlayer bot is a demo.)
This should make stack traces more useful, turning this sort of thing
09-20 11:29:39.536 2978 2978 F DEBUG : #00 pc
00970fd0 /data/local/tmp/dm
into something like this
09-20 11:29:39.536 2978 2978 F DEBUG : #00 pc
00970fd0 adjust_bounds_to_granularity(SkIRect*, SkIRect const&, VkExtent2D const&, int, int) at /b/work/skia/out/Build-Ubuntu-Clang-x86-Debug-GN_Android_Vulkan/Debug/../../../src/gpu/vk/GrVkGpu.cpp:1803 /data/local/tmp/dm
Some bots like the S7 already have good enough stack traces, e.g.
09-20 11:35:12.567 936 936 F DEBUG : #00 pc
00000000000bed6c /system/vendor/lib64/hw/vulkan.msm8996.so (_ZN13QglManagedBuf14ConfirmEntriesEv+108)
09-20 11:35:12.567 936 936 F DEBUG : #01 pc
00000000000b098c /system/vendor/lib64/hw/vulkan.msm8996.so (_ZN9QglBltLib6FillHwEPK10QglBltFillPK15QglBltColorFillPK15QglBltDepthFillPjP12QglBltStatusSB_+588)
09-20 11:35:12.567 936 936 F DEBUG : #02 pc
00000000000b23bc /system/vendor/lib64/hw/vulkan.msm8996.so (_ZN9QglBltLib9FillImageEPK10QglBltFillPK15QglBltColorFillPK15QglBltDepthFillPjP12QglBltStatusSB_+348)
09-20 11:35:12.567 936 936 F DEBUG : #03 pc
000000000009bb00 /system/vendor/lib64/hw/vulkan.msm8996.so (_ZN16A5xCommandBuffer22PerformConditionalFillEiiP10QglBltFillPK15QglBltColorFillP15QglBltDepthFillP12QglBltStatusS8_+256)
09-20 11:35:12.567 936 936 F DEBUG : #04 pc
000000000009c0c0 /system/vendor/lib64/hw/vulkan.msm8996.so (_ZN16A5xCommandBuffer23HwWriteSubpassInitClearEP10QglBltFillPK15QglBltColorFillP15QglBltDepthFillP12QglBltStatusS8_+224)
09-20 11:35:12.567 936 936 F DEBUG : #05 pc
0000000000072610 /system/vendor/lib64/hw/vulkan.msm8996.so (_ZN16QglCommandBuffer18WriteSubpassClearsEv+464)
09-20 11:35:12.567 936 936 F DEBUG : #06 pc
0000000000073ae0 /system/vendor/lib64/hw/vulkan.msm8996.so (_ZN16QglCommandBuffer12BeginSubpassEv+32)
09-20 11:35:12.567 936 936 F DEBUG : #07 pc
0000000000063120 /system/vendor/lib64/hw/vulkan.msm8996.so (vkCmdBeginRenderPass+224)
09-20 11:35:12.567 936 936 F DEBUG : #08 pc
0000000000635f60 /data/local/tmp/nanobench (_ZN24GrVkPrimaryCommandBuffer15beginRenderPassEPK7GrVkGpuPK14GrVkRenderPassjPK12VkClearValueRK16GrVkRenderTargetRK7SkIRectb+132)
or
09-20 11:42:24.557 937 937 F DEBUG : backtrace:
09-20 11:42:24.557 937 937 F DEBUG : #00 pc
0000000000069404 /system/lib64/libc.so (tgkill+8)
09-20 11:42:24.557 937 937 F DEBUG : #01 pc
0000000000066b94 /system/lib64/libc.so (pthread_kill+68)
09-20 11:42:24.557 937 937 F DEBUG : #02 pc
0000000000023a28 /system/lib64/libc.so (raise+28)
09-20 11:42:24.557 937 937 F DEBUG : #03 pc
000000000001e358 /system/lib64/libc.so (abort+60)
09-20 11:42:24.557 937 937 F DEBUG : #04 pc
000000000076430c /data/local/tmp/dm (_Z17sk_abort_no_printv+8)
These won't be affected.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2351243002
NOTREECHECKS=true
Review-Url: https://codereview.chromium.org/
2351243002
brianosman [Tue, 20 Sep 2016 16:56:23 +0000 (09:56 -0700)]
Cleanup some sRGB TODOs, mostly around setting GrPaint flags correctly
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2358433004
Review-Url: https://codereview.chromium.org/
2358433004
jvanverth [Tue, 20 Sep 2016 16:20:03 +0000 (09:20 -0700)]
Support use of non-coherent memory allocations in Vulkan.
BUG=skia:5034
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2348523002
Review-Url: https://codereview.chromium.org/
2348523002
bsalomon [Tue, 20 Sep 2016 16:12:47 +0000 (09:12 -0700)]
Stop flattening GrCoordTransforms in parent GrFragmentProcessors.
This changes moves to a model that iterates over GrCTs in a GrFP hierarchy when inserting transformations by GrGLSLPrimitiveProcessors.
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2339203002
Committed: https://skia.googlesource.com/skia/+/
d91237ee051523f439238042674ade99207fe4a6
Review-Url: https://codereview.chromium.org/
2339203002
egdaniel [Tue, 20 Sep 2016 15:54:23 +0000 (08:54 -0700)]
Workaround for Adreno INITIALIZATION_FAILED bug
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2352083002
Review-Url: https://codereview.chromium.org/
2352083002
sdefresne [Tue, 20 Sep 2016 15:53:14 +0000 (08:53 -0700)]
Fix a memory leak reported by clang static analyzer.
The CFStringCreateWithCString method respect the "Create Rule" [1]
regarding the ownership of the returned reference. This means that
the object needs to be deallocated explicitly by calling CFRelease.
[1]: https://developer.apple.com/library/content/documentation/CoreFoundation/Conceptual/CFMemoryMgmt/Concepts/Ownership.html#//apple_ref/doc/uid/
20001148-103029
BUG=648210
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2354963002
Review-Url: https://codereview.chromium.org/
2354963002
reed [Tue, 20 Sep 2016 15:42:38 +0000 (08:42 -0700)]
abstract name of clipping ops, to transtion to a more restricted set
SkRegion::Op --> SkCanvas::ClipOp (alias) --> SkClipOp
pre-CL needed in chrome : https://codereview.chromium.org/
2355583002/
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2355483002
Review-Url: https://codereview.chromium.org/
2355483002
mtklein [Tue, 20 Sep 2016 15:24:34 +0000 (08:24 -0700)]
skia_enable_android_framework_defines in all code, public and private, optional and required.
The bug fixed here can be demonstrated by
$ ninja -C out pdf -v
before and after this CL. Before you won't see the Android defines, afterwards you do. Previously the defines applied only privately to Skia, and only to its non-optional components.
TBR=reed@google.com
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2356783002
Review-Url: https://codereview.chromium.org/
2356783002
brianosman [Tue, 20 Sep 2016 15:06:30 +0000 (08:06 -0700)]
Auto-generate much of the help text for --config to reduce maintenance
The list of configs, and the details about all predefined configs are now
built directly from the gPredefinedConfigs array.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2354943002
Review-Url: https://codereview.chromium.org/
2354943002
borenet [Tue, 20 Sep 2016 14:01:23 +0000 (07:01 -0700)]
Fix Win toolchain asset, fix script
BUG=skia:5764
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2351033004
Review-Url: https://codereview.chromium.org/
2351033004
fmalita [Tue, 20 Sep 2016 13:55:16 +0000 (06:55 -0700)]
Harden LinearGradient4fContext
Three distinct issues exposed by Fuzzer:
1) truncating the flags to 8 bits may trip the assert in SkToU8
Since clients can pass in any garbage, we cannot assume it fits in 8
bits. Just static_cast<> instead.
2) vertical gradients with dx == -0.0 don't trigger interval inversion.
For dx < 0 we want reversed intervals. But, alas, -0.0 < 0 == false
(thanks, Ob^WIEEE 754!). Use signbit() instead.
3) half closed range checking needs to be inverted for reversed
intervals.
Normally we check for [p0, p1), but for reversed intervals (p1 < p0) we
want to check for [p1, p0), and not (p1, p0]. Adjust the logic
accordingly.
BUG=skia:5647
R=reed@google.com,kjlubick@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=
2349153004
Review-Url: https://codereview.chromium.org/
2349153004