The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/e04404cc853af2b6b91744e9d6d8a15403d1d426

commit e04404cc853af2b6b91744e9d6d8a15403d1d426
Author: pdr <pdr@chromium.org>
Date: Sun May 22 04:48:53 2016

Disable antialising for shapes with shape-rendering="optimizeSpeed"

The SVG spec hints that optimizeSpeed may disable antialising[1] and
this matches Gecko's behavior. This patch disables shape anti-aliasing
when optimizeSpeed is used.

The shape rendering test has been improved to cover all shape-rendering
values, and to include both stroked lines and filled paths.

[1] https://svgwg.org/svg2-draft/single-page.html#painting-ShapeRendering

BUG= 603796 

Review-Url: https://codereview.chromium.org/2003493003
Cr-Commit-Position: refs/heads/master@{#395272}

[modify] https://crrev.com/e04404cc853af2b6b91744e9d6d8a15403d1d426/third_party/WebKit/LayoutTests/TestExpectations
[modify] https://crrev.com/e04404cc853af2b6b91744e9d6d8a15403d1d426/third_party/WebKit/LayoutTests/platform/mac/svg/custom/shape-rendering-expected.png
[modify] https://crrev.com/e04404cc853af2b6b91744e9d6d8a15403d1d426/third_party/WebKit/LayoutTests/svg/custom/shape-rendering-expected.txt
[modify] https://crrev.com/e04404cc853af2b6b91744e9d6d8a15403d1d426/third_party/WebKit/LayoutTests/svg/custom/shape-rendering.svg
[modify] https://crrev.com/e04404cc853af2b6b91744e9d6d8a15403d1d426/third_party/WebKit/Source/core/paint/SVGShapePainter.cpp

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/af2d6e1462a9e2cef72e3d501b65560c5a616f76

commit af2d6e1462a9e2cef72e3d501b65560c5a616f76
Author: Rebaseline Bot <blink-rebaseline-bot@chromium.org>
Date: Sun May 22 06:02:17 2016

Auto-rebaseline for r395272

https://chromium.googlesource.com/chromium/src/+/e04404cc8

BUG= 603796 
TBR=pdr@chromium.org

Review URL: https://codereview.chromium.org/2000033002 .

Cr-Commit-Position: refs/heads/master@{#395273}

[modify] https://crrev.com/af2d6e1462a9e2cef72e3d501b65560c5a616f76/third_party/WebKit/LayoutTests/TestExpectations
[add] https://crrev.com/af2d6e1462a9e2cef72e3d501b65560c5a616f76/third_party/WebKit/LayoutTests/platform/android/svg/custom/shape-rendering-expected.png
[add] https://crrev.com/af2d6e1462a9e2cef72e3d501b65560c5a616f76/third_party/WebKit/LayoutTests/platform/android/svg/custom/shape-rendering-expected.txt
[add] https://crrev.com/af2d6e1462a9e2cef72e3d501b65560c5a616f76/third_party/WebKit/LayoutTests/platform/android/svg/custom/use-referencing-nonexisting-symbol-expected.png
[modify] https://crrev.com/af2d6e1462a9e2cef72e3d501b65560c5a616f76/third_party/WebKit/LayoutTests/platform/linux/svg/custom/shape-rendering-expected.png
[modify] https://crrev.com/af2d6e1462a9e2cef72e3d501b65560c5a616f76/third_party/WebKit/LayoutTests/platform/linux/svg/custom/shape-rendering-expected.txt
[modify] https://crrev.com/af2d6e1462a9e2cef72e3d501b65560c5a616f76/third_party/WebKit/LayoutTests/platform/linux/svg/custom/use-referencing-nonexisting-symbol-expected.png
[modify] https://crrev.com/af2d6e1462a9e2cef72e3d501b65560c5a616f76/third_party/WebKit/LayoutTests/platform/mac/svg/custom/use-referencing-nonexisting-symbol-expected.png
[add] https://crrev.com/af2d6e1462a9e2cef72e3d501b65560c5a616f76/third_party/WebKit/LayoutTests/platform/win/svg/custom/shape-rendering-expected.txt
[add] https://crrev.com/af2d6e1462a9e2cef72e3d501b65560c5a616f76/third_party/WebKit/LayoutTests/platform/win7/svg/custom/shape-rendering-expected.png
[add] https://crrev.com/af2d6e1462a9e2cef72e3d501b65560c5a616f76/third_party/WebKit/LayoutTests/platform/win7/svg/custom/shape-rendering-expected.txt
[add] https://crrev.com/af2d6e1462a9e2cef72e3d501b65560c5a616f76/third_party/WebKit/LayoutTests/platform/win7/svg/custom/use-referencing-nonexisting-symbol-expected.png

I filed a WebKit bug as well: https://bugs.webkit.org/show_bug.cgi?id=157981

Going to request a merge to M52 for this patch, due to issue 617658.

[Automated comment] Less than 2 weeks to go before stable on M52, manual review required.

@tpms, this merge is trivial and affects a google property.

Before we approve merge to M52, Could you please confirm whether this change is baked/verified in Canary and safe to merge? (As per https://bugs.chromium.org/p/chromium/issues/detail?id=617658#c22, still failing with this patch applied)

Also is this change applicable to all OS or any specific OS? 

Yes, this patch has baked in Canary and is safe to merge.

This merge does not completely fix the issue in 617658 but papers over the bug. This only affects linux.

Thank you pdr@.

M52 is already in stable and bar is VERY high. As this issue only effects Linux and has been known for a while, do you really think it is worth to merge to M52? I just want to double make sure that it doesn't cause any regression in field before approving.

We just checked Windows and it does not repro there, nor on Mac, so we think this is limited to linux. I'm going to lean towards not merging after all.

I will talk with the docs folks and see if we can live with the regression.

Sure, thank you. Removing "Merge-Review-52" & "Hotlist-Merge-review" labels for now. Please re-request M52 merge if needed.