1824283 - WPT failure in css/css-masking/mask-image/mask-image-2.html

Status: RESOLVED FIXED

(Core :: Layout, task)

Description

[Daniel Holbert [:dholbert]]

WPT failure:
https://wpt.fyi/results/css/css-masking/mask-image/mask-image-2.html

We have "Found 2400 pixels different, maximum difference per channel 1"

Likely this is just a case where rounding (of intermediate-in-the-gradient) fractional color-channel values works out slightly differently in these two codepaths, for one reason or another, and it's not really visually perceptible.

side note: 2400 pixels does seem like a lot, but it's just due to the structure of the test that makes the mismatching pixels repeat.

Really, there are only 12 differing positions in the gradient (with maxDifference of 1); and those pixels get stretched across a 100px-wide rect (so 1200 mismatching pixels), and then that rect gets repeated (so 2400 pixels).

Comment 1

[Daniel Holbert [:dholbert]]

Attachment: Bug 1824283: Annotate WPT mask-image-2.html with observed fuzziness. r?#layout-reviewers

The 2400px differing pixels may seem like a lot, but really the observed
fuzziness here is pretty minimal; really, it's just 12 distinct points
(stripes) in each of the two 100px-by-100px gradients where we seem to be
rounding differently vs. the reference case, and producing an imperceptible
+/-1 difference in some color channel.

Comment 2

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/7957a017e897
Annotate WPT mask-image-2.html with observed fuzziness. r=emilio

Comment 3

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

Created web-platform-tests PR https://github.com/web-platform-tests/wpt/pull/39188 for changes under testing/web-platform/tests

Comment 4

[Sandor Molnar[:smolnar]]

https://hg.mozilla.org/mozilla-central/rev/7957a017e897

Comment 5

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

Upstream PR merged by moz-wptsync-bot