New issue
Advanced search Search tips

Issue 898931 link

Starred by 2 users
Status: Assigned
Owner:
senorblanco@chromium.org
Cc:
viswa.karala@chromium.org
senorblanco@chromium.org
bsalo...@google.com
Components:
EstimatedDays: ----
NextAction: ----
OS: Windows , Mac
Pri: 2
Type: Bug-Regression



Sign in to add a comment

Path element with multiple closed paths with arcs has new/different gradient artifacts

Reported by jcherwo...@gmail.com, Oct 25 
UserAgent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/72.0.3591.0 Safari/537.36

Steps to reproduce the problem:
1. View arcs3.svg
2. Also view at various browser zoom levels
3. Repeat steps 1-2 with arcs2.svg and arcs.svg

What is the expected behavior?
All shapes consistently appear with solid grey fill

What went wrong?
Shapes appear with gradient-like artifacts, depending on zoom level and version:

arcs3.svg - Appears with artifacts in v72 at most zoom levels, including slight artifacts at 100%. Appears correct in v70 at all zoom levels.

arcs2.svg - Appears with gradient artifacts at 100% zoom both in v70 (stable) and v72 (canary). Also affects other zoom levels, e.g., 200%

arcs.svg - Appears correct mainly only at 100% zoom in v72, incorrect at 80/90/110/120/300% zoom. In v70 it also appears incorrect at 100% zoom

Did this work before? Yes 70.0.3538.77

Does this work in other browsers? Yes

Chrome version: 72.0.3591.0  Channel: canary
OS Version: 10.0
Flash Version: 

arcs3.svg demonstrates a new regression introduced in v71-v72 not present in v70. It may be related to changes made for  issue 887103 , which issue 882584 was merged into. Issue 882584 was not actually fixed as shown by arcs.svg and arcs2.svg, and may not be a duplicate of  issue 887103 .
arcs3.svg
441 bytes Download
arcs3_good_v70.PNG
18.8 KB View Download
arcs3_bad_v72.PNG
18.7 KB View Download
arcs2.svg
1.0 KB Download
arcs2_good.png
24.2 KB View Download
arcs2_bad.PNG
37.5 KB View Download
arcs.svg
607 bytes Download
arcs_good.PNG
6.5 KB View Download
arcs_bad_v72_zoom110.PNG
14.5 KB View Download
arcs_bad_v70.PNG
11.7 KB View Download

Comment 1 by jcherwo...@gmail.com, Oct 25 

Dev build 71.0.3578.20 is the same as 72/canary.

Comment 2 by f...@opera.com, Oct 25 

This looks like a GPU rasterization bug, can you try testing with GPU rasterization disabled to verify?

Comment 3 by jcherwo...@gmail.com, Oct 25 

Yes, disabling GPU rasterization avoids all of these problems.

Comment 4 by f...@opera.com, Oct 25

Components: -Blink>SVG Internals>GPU>Rasterization
Thanks for checking. Forwarding to a more appropriate component.

Comment 5 by phanindra.mandapaka@chromium.org, Oct 25

Labels: Needs-Triage-M72

Comment 6 by vamshi.kommuri@chromium.org, Oct 26

Labels: Needs-Bisect

Comment 7 by viswa.karala@chromium.org, Oct 26

Cc: viswa.karala@chromium.org
Labels: -Pri-2 -Needs-Bisect RegressedIn-71 ReleaseBlock-Stable Triaged-ET Target-71 Target-72 FoundIn-72 M-71 FoundIn-71 hasbisect OS-Mac Pri-1
Owner: reed@google.com
Status: Assigned (was: Unconfirmed)
Able to reproduce the issue on reported chrome version 72.0.3591.0 using Windows-10, Mac 10.12.6 hence providing Bisect Info

Bisect Info:
================
Good build: 71.0.3368.0
Bad build: 71.0.3369.0

Note: On running per-revision bisect got "RuntimeError: We don't have enough builds to bisect. revlist: []", hence proving below bisect info from chromium bisect

You are probably looking for a change made after 595557 (known good), but no later than 595635 (first known bad).
Change Log: https://chromium.googlesource.com/chromium/src/+log/73a64c107201141b3af296e76fb8c98c14523d12..59aed3764077ffd201c04490e81801d6a24ed9f2
Suspecting: https://chromium.googlesource.com/chromium/src/+/05ee0393e7a7f04f306be79aa5b8c1f4398344a6 from above change log
Change-Id: I330bdfe4dcbfed55a03385043e703af6fe070ffd
Reviewed-on: https://chromium-review.googlesource.com/1254341

@Mike Reed: Please confirm the issue and help in re-assigning if it is not related to your change.
Adding ReleaseBlock-Stable for M-71, feel free to remove it if not applicable.

Thanks!

Comment 8 by schenney@chromium.org, Oct 26

Owner: ----
Status: Untriaged (was: Assigned)

Comment 9 by vmi...@chromium.org, Oct 26

Cc: senorblanco@chromium.org
Owner: bsalomon@chromium.org
Status: Assigned (was: Untriaged)
Assigning to bsalomon@ for triage.

Comment 10 by gov...@chromium.org, Oct 29 

M71 Stable promotion is coming VERY soon. Your bug is labelled as Stable  ReleaseBlock, pls make sure to land the fix and request a merge into the release branch ASAP. Thank you.

Comment 11 by gov...@chromium.org, Oct 29 

M71 Stable promotion is coming VERY soon. Your bug is labelled as Stable  ReleaseBlock, pls make sure to land the fix and request a merge into the release branch ASAP. Thank you.

Comment 12 by bsalo...@google.com, Oct 29

Cc: bsalo...@google.com
Owner: senorblanco@chromium.org
I swear I updated this bug on Friday.. not sure what happened. Anyway, this doesn't reproduce when the tessellating path renderer is disabled. I was able to reproduce on linux with --force-gpu-rasterization and the OP's svgs.

Comment 13 by gov...@chromium.org, Nov 5 

M71 Stable promotion is coming VERY soon. Your bug is labelled as Stable  ReleaseBlock, pls make sure to land the fix and request a merge into the release branch ASAP. Thank you.

Comment 14 by gov...@chromium.org, Nov 8 

M71 Stable promotion is coming VERY soon. Your bug is labelled as Stable  ReleaseBlock, pls make sure to land the fix and request a merge into the release branch ASAP. Thank you.

Comment 15 by senorblanco@chromium.org, Nov 8

Labels: -ReleaseBlock-Stable
I don't think this is a release blocker.

Comment 16 by senorblanco@chromium.org, Today (16 hours ago)

Labels: -Pri-1 Pri-2
►

Sign in to add a comment