1914345 - Create a pref to allow using moz2d to determine stroke bounds

Status: RESOLVED FIXED

(Core :: SVG, task)

Description

[Robert Longson [:longsonr]]

At least one of the problems with this approach has gone.

Comment 1

[Robert Longson [:longsonr]]

https://treeherder.mozilla.org/jobs?repo=try&revision=171c7e1bf2b9eb539a0e7a29f436c5ab37289366

Comment 2

[Robert Longson [:longsonr]]

https://treeherder.mozilla.org/jobs?repo=try&revision=b1d60e792a7e4e09d28ca079121e9325e2b2d8f1

Comment 3

[Robert Longson [:longsonr]]

Attachment: Bug 1914345 Part 1 - create a pref to allow using moz2d to determine stroke bounds r=emilio

Comment 4

[Robert Longson [:longsonr]]

Attachment: Bug 1914345 Part 2 - enable moz2d stroke bounds on nightly r=emilio

Comment 5

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/6ffabd992e4e
Part 1 - create a pref to allow using moz2d to determine stroke bounds r=emilio
https://hg.mozilla.org/integration/autoland/rev/c0f325a58cd9
Part 2 - enable moz2d stroke bounds on nightly r=emilio

Comment 6

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Created web-platform-tests PR https://github.com/web-platform-tests/wpt/pull/47791 for changes under testing/web-platform/tests

Comment 7

[Serban Stanca [:SerbanS]]

Backed out for causing Android reftests failures.

• Backout link
• Push with failures
• Failure Log
• Failure line: REFTEST TEST-UNEXPECTED-FAIL | gfx/tests/reftest/1662062-1-no-blurry.html == gfx/tests/reftest/1662062-1-ref.html | image comparison, max difference: 244, number of differing pixels: 4111

Also, please check these reftests failures. They're also on Android platform.

• Failure Log

Comment 8

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Upstream PR was closed without merging

Comment 9

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/3d7bad39c3f0
Part 1 - create a pref to allow using moz2d to determine stroke bounds r=emilio
https://hg.mozilla.org/integration/autoland/rev/92acbb3b404d
Part 2 - enable moz2d stroke bounds on nightly r=emilio

Comment 10

[Robert Longson [:longsonr]]

The other reftest failures do not contain any SVG so can't be due to these patches.

Comment 11

[Serban Stanca [:SerbanS]]

Backed out for causing reftests failures.

• Backout link
• Push with failures
• Failure Log
• Failure line: REFTEST TEST-UNEXPECTED-FAIL | gfx/tests/reftest/1662062-1-no-blurry.html == gfx/tests/reftest/1662062-1-ref.html | image comparison, max difference: 255, number of differing pixels: 20259

Comment 12

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Upstream PR was closed without merging

Comment 13

[Robert Longson [:longsonr]]

I'll just land part 1. The whole change is preffed off so it can't cause any failures.

Comment 14

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/03bd6b90ec89
Part 1 - create a pref to allow using moz2d to determine stroke bounds r=emilio

Comment 15

[Norisz Fay [:noriszfay]]

Backed out for causing reftest failure on 1662062-1-no-blurry.html

Backout link

Push with failures

Failure log

Comment 16

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Upstream PR was closed without merging

Comment 17

[Daniel Holbert [:dholbert]]

bug 1915025 is filed for some of the test failures that look superficially like they were caused by this (but which in fact go back further and just happen-to-be-tests-that-don't-normally-run).

Comment 18

[Robert Longson [:longsonr]]

https://treeherder.mozilla.org/jobs?repo=try&revision=cd1bd16f10ea31eb29915139d4cb0316bca1638a

Comment 19

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/a7d50718034f
Part 1 - create a pref to allow using moz2d to determine stroke bounds r=emilio
https://hg.mozilla.org/integration/autoland/rev/6b6da4bd90cb
Part 2 - enable moz2d stroke bounds on nightly r=emilio

Comment 20

[Robert Longson [:longsonr]]

Attachment: Bug 1914345 Part 3 - Increase fuzziness for Android R-nofis r=emilio

Comment 21

[Robert Longson [:longsonr]]

Attachment: Bug 1914345 Part 4 - Increase fuzziness for Android R-nofis r=emilio

Comment 22

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/b056a6ba9a8b
Part 3 - Increase fuzziness for Android R-nofis r=emilio
https://hg.mozilla.org/integration/autoland/rev/ae6159a806a3
Part 4 - Increase fuzziness for Android R-nofis r=emilio CLOSED TREE

Comment 23

[Iulian Moraru]

https://hg.mozilla.org/mozilla-central/rev/a7d50718034f
https://hg.mozilla.org/mozilla-central/rev/6b6da4bd90cb

Comment 24

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Upstream PR merged by moz-wptsync-bot

Comment 25

[Atila Butkovits]

https://hg.mozilla.org/mozilla-central/rev/b056a6ba9a8b
https://hg.mozilla.org/mozilla-central/rev/ae6159a806a3

Comment 26

[Bas Schouten (:bas.schouten)]

Could this be the cause of a regression to SVG-Stockcharts? (See https://treeherder.mozilla.org/perfherder/graphs?highlightAlerts=1&highlightChangelogData=1&highlightCommonAlerts=0&replicates=0&selected=4670725,85077601&series=mozilla-central,4670725,1,13&timerange=1209600)

Comment 27

[Robert Longson [:longsonr]]

The pref is on for nightly only so it could be preffed off and then you could test whether performance changes. We will have to live with it though because the old way while simpler, was incorrect and overestimated the bounds. That leads to observable issues e.g. bug 1849054