I believe this is paint?

Able to reproduce the issue on Windows, Mac, below are bisect results :

Last good build : 49.0.2585.0
First Bad build : 49.0.2586.0

Bisect Result :
You are probably looking for a change made after 363785 (known good), but no lat
er than 363812 (first known bad).
CHANGELOG URL:
  https://chromium.googlesource.com/chromium/src/+log/42290cb91a66566637a1715737b1961a8480b934..b5daa8ef58d6f02e446ad19cc44dc2c0d192db98

Suspecting CL : https://chromium.googlesource.com/chromium/src/+/9c60333a01e2cad5fa21318c39c8a330244b3d56

I won't be getting to this stuff anytime soon.

FYI the large background sizes are the best way to illustrate the problem but the bug is present with all background size values

Sorry I mean "background-position" instead of background size.

I'll at least try to figure out what's happening since I'm in this code right now.

Thanks, schenney!!

The attached file demonstrates the problem without window resize.

Deleted: background-large-position-and-size-remains-stable.html 1.7 KB

background-large-position-and-size-remains-stable.html 1.7 KB View Download

I have a fix for background-repeat: repeat, but I suspect we'll also need something for round and space. And we need some serious code cleanup now. It's getting unmanageable.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/41f3d062c90c1a5d67228e6e061c7e35fd4d62fd

commit 41f3d062c90c1a5d67228e6e061c7e35fd4d62fd
Author: schenney <schenney@chromium.org>
Date: Fri Apr 01 22:20:57 2016

Fix for phase in background tiling.

When computing the phase for background image tiling using snapped
coordinates, we get very incorrect results, and unstable results,
when the background-position is large (because the division of
the position by the tile width magnifies the rounding artifacts).
This patch changes the phase calculation to compute the phase using
unsnapped values then re-apply the fractional component to the
snapped values. This results in stable, near-pixel perfect results.
Given very large background-positions, we can't do much better.
There's a new test and some results changed by a pixel or so.

There is some code refactoring to reduce what is becoming a very
unwieldy method. Sorry that makes it harder to review.

R=leviw@chromium.org
BUG= 593079 

Review URL: https://codereview.chromium.org/1847983003

Cr-Commit-Position: refs/heads/master@{#384706}

[modify] https://crrev.com/41f3d062c90c1a5d67228e6e061c7e35fd4d62fd/third_party/WebKit/LayoutTests/TestExpectations
[add] https://crrev.com/41f3d062c90c1a5d67228e6e061c7e35fd4d62fd/third_party/WebKit/LayoutTests/css3/background/background-large-position-and-size-remains-stable.html
[add] https://crrev.com/41f3d062c90c1a5d67228e6e061c7e35fd4d62fd/third_party/WebKit/LayoutTests/css3/background/resources/red-green-red-vertical.png
[add] https://crrev.com/41f3d062c90c1a5d67228e6e061c7e35fd4d62fd/third_party/WebKit/LayoutTests/css3/background/resources/red-green-red.png
[add] https://crrev.com/41f3d062c90c1a5d67228e6e061c7e35fd4d62fd/third_party/WebKit/LayoutTests/css3/background/resources/red-red-green-vertical.png
[add] https://crrev.com/41f3d062c90c1a5d67228e6e061c7e35fd4d62fd/third_party/WebKit/LayoutTests/css3/background/resources/red-red-green.png
[add] https://crrev.com/41f3d062c90c1a5d67228e6e061c7e35fd4d62fd/third_party/WebKit/LayoutTests/platform/linux/css3/background/background-large-position-and-size-remains-stable-expected.png
[add] https://crrev.com/41f3d062c90c1a5d67228e6e061c7e35fd4d62fd/third_party/WebKit/LayoutTests/platform/linux/css3/background/background-large-position-and-size-remains-stable-expected.txt
[modify] https://crrev.com/41f3d062c90c1a5d67228e6e061c7e35fd4d62fd/third_party/WebKit/Source/core/paint/BackgroundImageGeometry.cpp
[modify] https://crrev.com/41f3d062c90c1a5d67228e6e061c7e35fd4d62fd/third_party/WebKit/Source/core/paint/BackgroundImageGeometry.h

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/75a1fce006af0649210716da9c1ed05f6f377a0d

commit 75a1fce006af0649210716da9c1ed05f6f377a0d
Author: Rebaseline Bot <blink-rebaseline-bot@chromium.org>
Date: Sat Apr 02 00:41:42 2016

Auto-rebaseline for r384706

https://chromium.googlesource.com/chromium/src/+/41f3d062c

BUG= 593079 
TBR=schenney@chromium.org

Review URL: https://codereview.chromium.org/1847853006 .

Cr-Commit-Position: refs/heads/master@{#384759}

[modify] https://crrev.com/75a1fce006af0649210716da9c1ed05f6f377a0d/third_party/WebKit/LayoutTests/TestExpectations
[add] https://crrev.com/75a1fce006af0649210716da9c1ed05f6f377a0d/third_party/WebKit/LayoutTests/css3/background/background-large-position-and-size-remains-stable-expected.png
[modify] https://crrev.com/75a1fce006af0649210716da9c1ed05f6f377a0d/third_party/WebKit/LayoutTests/fast/backgrounds/size/backgroundSize12-expected.png
[modify] https://crrev.com/75a1fce006af0649210716da9c1ed05f6f377a0d/third_party/WebKit/LayoutTests/fast/backgrounds/size/backgroundSize18-expected.png
[modify] https://crrev.com/75a1fce006af0649210716da9c1ed05f6f377a0d/third_party/WebKit/LayoutTests/fast/backgrounds/size/backgroundSize19-expected.png
[modify] https://crrev.com/75a1fce006af0649210716da9c1ed05f6f377a0d/third_party/WebKit/LayoutTests/fast/backgrounds/size/backgroundSize21-expected.png
[modify] https://crrev.com/75a1fce006af0649210716da9c1ed05f6f377a0d/third_party/WebKit/LayoutTests/fast/backgrounds/size/backgroundSize22-expected.png
[add] https://crrev.com/75a1fce006af0649210716da9c1ed05f6f377a0d/third_party/WebKit/LayoutTests/platform/mac/css3/background/background-large-position-and-size-remains-stable-expected.txt
[add] https://crrev.com/75a1fce006af0649210716da9c1ed05f6f377a0d/third_party/WebKit/LayoutTests/platform/win/css3/background/background-large-position-and-size-remains-stable-expected.txt
[modify] https://crrev.com/75a1fce006af0649210716da9c1ed05f6f377a0d/third_party/WebKit/LayoutTests/platform/win/fast/inline/inline-box-background-repeat-x-expected.png

Fixed? :)

Why yes, I think it is fixed.