Remove SK_SUPPORT_LEGACY_ARCTO_QUADS

Remove SK_SUPPORT_LEGACY_ARCTO_QUADS

TBR=reed@google.com,robertphillips@google.com
BUG=458898
CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg

Committed: https://crrev.com/a5a3bd2a3fbc8ad67f73133ed599e151f0a203c2
Cr-Commit-Position: refs/heads/master@{#318104}

[reed2, 2015-02-15 18:20:12]

reed@chromium.org changed reviewers:
+ reed@chromium.org

[reed2, 2015-02-15 18:20:12]

I looked at the ADDOVAL and ADDRRECT cls, and they are clearly huge! Maybe this
one is worth landing separately, or you could combine it with the ADD ones.

[f(malita), 2015-02-15 18:27:50]

On 2015/02/15 18:20:12, reed2 wrote:
> I looked at the ADDOVAL and ADDRRECT cls, and they are clearly huge! Maybe
this
> one is worth landing separately, or you could combine it with the ADD ones.

The ref-test diffs are troublesome. I was hoping they would be clustered on one
these CLs, but apparently they all touch a few.

I'll need to take a close look to a) make sure the diffs are expected and b)
convert them to regular pixel tests

Also, this CL in particular seems to trigger some text-diffs - essentially
measurements are slightly different (bbox 1px taller). I presume these are
expected?

[f(malita), 2015-02-16 00:43:28]

These look like real failures:

https://storage.googleapis.com/chromium-layout-test-archives/linux_blink_rel/...
https://storage.googleapis.com/chromium-layout-test-archives/linux_blink_rel/...
https://storage.googleapis.com/chromium-layout-test-archives/linux_blink_rel/...
https://storage.googleapis.com/chromium-layout-test-archives/linux_blink_rel/...

[f(malita), 2015-02-16 01:54:09]

On 2015/02/16 00:43:28, f(malita) wrote:
> These look like real failures:
> 
>
https://storage.googleapis.com/chromium-layout-test-archives/linux_blink_rel/...
>
https://storage.googleapis.com/chromium-layout-test-archives/linux_blink_rel/...
>
https://storage.googleapis.com/chromium-layout-test-archives/linux_blink_rel/...
>
https://storage.googleapis.com/chromium-layout-test-archives/linux_blink_rel/...

https://code.google.com/p/skia/issues/detail?id=3428

[reed2, 2015-02-16 03:36:19]

On 2015/02/16 01:54:09, f(malita) wrote:
> On 2015/02/16 00:43:28, f(malita) wrote:
> > These look like real failures:
> > 
> >
>
https://storage.googleapis.com/chromium-layout-test-archives/linux_blink_rel/...
> >
>
https://storage.googleapis.com/chromium-layout-test-archives/linux_blink_rel/...
> >
>
https://storage.googleapis.com/chromium-layout-test-archives/linux_blink_rel/...
> >
>
https://storage.googleapis.com/chromium-layout-test-archives/linux_blink_rel/...
> 
> https://code.google.com/p/skia/issues/detail?id=3428

Hmmm, I'll need to look into those.

[f(malita), 2015-02-20 16:07:41]

fmalita@chromium.org changed reviewers:
+ robertphillips@google.com

[f(malita), 2015-02-20 16:08:03]

After the degenerate arc fix, I think all the remaining diffs are correct:

https://storage.googleapis.com/chromium-layout-test-archives/linux_blink_rel/...

* minor pixel diffs

* text diffs: off-by-one-pixel position/size diffs. I assume these are expected.

* canvas-ellipse-connecting-line.html: minor pixel diffs. Unfortunately this is
a ref test which makes assumptions about addOval impl and will have to be
downgraded to a pixel test

* canvas-scroll-path-into-view.html: text-only diff. I verified that the test
still works correctly - we just need to update its expected values


I'll add suppressions and land, then deal with the Blink rebaselines.

[f(malita), 2015-02-20 16:24:20]

The CQ bit was checked by fmalita@chromium.org

[commit-bot: I haz the power, 2015-02-20 16:25:26]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/928853003/20001

[f(malita), 2015-02-20 16:26:20]

The CQ bit was unchecked by fmalita@chromium.org

[reed1, 2015-02-20 16:47:26]

The CQ bit was checked by reed@google.com

[commit-bot: I haz the power, 2015-02-20 16:48:01]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/928853003/20001

[f(malita), 2015-02-20 16:49:40]

On 2015/02/20 16:48:01, I haz the power (commit-bot) wrote:
> CQ is trying da patch. Follow status at
>  https://chromium-cq-status.appspot.com/patch-status/928853003/20001

Let's wait for a couple of Blink bots to make sure the suppression list is
complete.

[f(malita), 2015-02-20 16:49:43]

The CQ bit was unchecked by fmalita@chromium.org

[f(malita), 2015-02-20 17:00:04]

On 2015/02/20 16:49:40, f(malita) wrote:
> On 2015/02/20 16:48:01, I haz the power (commit-bot) wrote:
> > CQ is trying da patch. Follow status at
> >  https://chromium-cq-status.appspot.com/patch-status/928853003/20001
> 
> Let's wait for a couple of Blink bots to make sure the suppression list is
> complete.

(I guess I could just use CQ_EXTRA_TRYBOTS like the roll bot)

[f(malita), 2015-02-20 17:00:32]

The CQ bit was checked by fmalita@chromium.org

[commit-bot: I haz the power, 2015-02-20 17:01:11]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/928853003/20001

[f(malita), 2015-02-20 18:15:05]

The CQ bit was unchecked by fmalita@chromium.org

[f(malita), 2015-02-20 18:17:28]

Debug builds are crashing on virtual/gpu/fast/canvas/canvas-largedraws.html:
https://storage.googleapis.com/chromium-layout-test-archives/mac_blink_dbg/27...

STDERR:
[29610:1287:0220/084400:32491938234623:INFO:GrAAConvexPathRenderer.cpp(808)]
../../third_party/skia/src/gpu/GrAAConvexPathRenderer.cpp:808: failed assertion
"args.fDevBounds.contains(actualBounds)"
STDERR: 
STDERR:
[29610:1287:0220/084400:32491938283013:FATAL:GrAAConvexPathRenderer.cpp(808)]
SK_CRASH
STDERR: 0   libbase.dylib                       0x00000001059f7f5f
base::debug::StackTrace::StackTrace() + 47
STDERR: 1   libbase.dylib                       0x00000001059f7fb3
base::debug::StackTrace::StackTrace() + 35
STDERR: 2   libbase.dylib                       0x0000000105a58556
logging::LogMessage::~LogMessage() + 70
STDERR: 3   libbase.dylib                       0x0000000105a575a3
logging::LogMessage::~LogMessage() + 35
STDERR: 4   libskia.dylib                       0x00000001167d0c49
SkDebugf_FileLine(char const*, int, bool, char const*, ...) + 521
STDERR: 5   libskia.dylib                       0x00000001167fcdc4
AAConvexPathBatch::generateGeometry(GrBatchTarget*, GrPipeline const*) + 1172
STDERR: 6   libskia.dylib                       0x0000000116857018
GrInOrderDrawBuffer::DrawBatch::execute(GrInOrderDrawBuffer*,
GrInOrderDrawBuffer::SetState const*) + 248
STDERR: 7   libskia.dylib                       0x00000001168580c3
GrInOrderDrawBuffer::closeBatch() + 99
STDERR: 8   libskia.dylib                       0x000000011685657d
GrInOrderDrawBuffer::onFlush() + 93
STDERR: 9   libskia.dylib                       0x000000011684ac97
GrFlushToGpuDrawTarget::flush() + 439
STDERR: 10  libskia.dylib                       0x0000000116827ad0
GrContext::flush(int) + 112
STDERR: 11  libskia.dylib                       0x000000011682ea55
GrContext::readRenderTargetPixels(GrRenderTarget*, int, int, int, int,
GrPixelConfig, void*, unsigned long, unsigned int) + 565
STDERR: 12  libskia.dylib                       0x0000000116966044
SkGpuDevice::onReadPixels(SkImageInfo const&, void*, unsigned long, int, int) +
324
STDERR: 13  libskia.dylib                       0x00000001164e55a9
SkBaseDevice::readPixels(SkImageInfo const&, void*, unsigned long, int, int) +
1161
STDERR: 14  libskia.dylib                       0x00000001164bee96
SkCanvas::readPixels(SkImageInfo const&, void*, unsigned long, int, int) + 230
STDERR: 15  libskia.dylib                       0x000000011678eba6
SkDeferredDevice::onReadPixels(SkImageInfo const&, void*, unsigned long, int,
int) + 102
STDERR: 16  libskia.dylib                       0x00000001164e55a9
SkBaseDevice::readPixels(SkImageInfo const&, void*, unsigned long, int, int) +
1161
STDERR: 17  libskia.dylib                       0x00000001164bee96
SkCanvas::readPixels(SkImageInfo const&, void*, unsigned long, int, int) + 230
STDERR: 18  libblink_platform.dylib             0x0000000107498247
blink::GraphicsContext::readPixels(SkImageInfo const&, void*, unsigned long,
int, int) + 199
STDERR: 19  libblink_platform.dylib             0x00000001074d19f8
blink::ImageBuffer::getImageData(blink::Multiply, blink::IntRect const&,
WTF::ArrayBufferContents&) const + 776
STDERR: 20  libblink_web.dylib                  0x000000010a326007
blink::CanvasRenderingContext2D::getImageData(float, float, float, float,
blink::ExceptionState&) const + 935

[f(malita), 2015-02-20 18:27:08]

I can repro the crash locally with 

blink/tools/run_layout_tests.sh  --debug
virtual/gpu/fast/canvas/canvas-largedraws.html

The relevant C2D snippet:

        ctx = newCanvasContext();
        ctx.fillStyle = '#0f0';
        ctx.fillRect(0, 0, 100, 50);
        ctx.scale(2, -4500000000);
        ctx.arc(25, 50, 56, 0, 2*Math.PI, false);
        ctx.fill();
        ctx.arc(105000000005, -105000000005, 105000000004, 0,
10500000000*Math.PI, false);
        ctx.fill();
        ctx.getImageData(50, 25, 1, 1);

[reed1, 2015-02-20 18:31:56]

reed@google.com changed reviewers:
+ bsalomon@google.com

[reed2, 2015-02-24 23:19:51]

Did we fix the crashers associated with this?

[f(malita), 2015-02-24 23:24:37]

fmalita@chromium.org changed reviewers:
+ egdaniel@google.com

[f(malita), 2015-02-24 23:25:01]

On 2015/02/24 23:19:51, reed2 wrote:
> Did we fix the crashers associated with this?

+Greg

We fixed the initial assert, but I can still trigger a different one with the
Skia repro in https://code.google.com/p/skia/issues/detail?id=3453#c1.

[f(malita), 2015-02-25 16:15:02]

On 2015/02/24 23:25:01, f(malita) wrote:
> On 2015/02/24 23:19:51, reed2 wrote:
> > Did we fix the crashers associated with this?
> 
> +Greg
> 
> We fixed the initial assert, but I can still trigger a different one with the
> Skia repro in https://code.google.com/p/skia/issues/detail?id=3453#c1.

https://chromium.googlesource.com/skia/+/5a23a14b1fbc7503bdeff83e4b45ae7c258c...
rolled, so I think we're good to go. I fired another blink bot series.

[f(malita), 2015-02-25 19:25:13]

The CQ bit was checked by fmalita@chromium.org

[commit-bot: I haz the power, 2015-02-25 19:26:44]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/928853003/20001

[commit-bot: I haz the power, 2015-02-25 20:30:22]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-02-25 20:31:14]

Patchset 2 (id:??) landed as
https://crrev.com/a5a3bd2a3fbc8ad67f73133ed599e151f0a203c2
Cr-Commit-Position: refs/heads/master@{#318104}