Open Bug 1892222 Opened 5 months ago Updated 2 months ago

Intermittent layout/reftests/svg/smil/style/anim-css-font-1.svg == layout/reftests/svg/smil/style/anim-css-font-1-ref.svg | single tracking bug

Categories

(Core :: SVG, defect, P5)

defect

Tracking

()

People

(Reporter: intermittent-bug-filer, Unassigned)

References

Details

(Keywords: intermittent-failure, intermittent-testcase)

Tier 1 failure here

Summary: Intermittent [tier 2] layout/reftests/svg/smil/style/anim-css-font-1.svg == layout/reftests/svg/smil/style/anim-css-font-1-ref.svg | single tracking bug → Intermittent layout/reftests/svg/smil/style/anim-css-font-1.svg == layout/reftests/svg/smil/style/anim-css-font-1-ref.svg | single tracking bug

At the third instance the outline of the file icon is more accentuated.

(Ignore this note^ from comment 0; it's off-topic for this bug; it's describing a different test failure that happened to occur in the same test-run; that's covered by bug 1894445.)

For this bug here, on anim-css-font-1.svg, the intermittent thing is in fact an unexpected pass. Apparently this test has been fuzzy on Linux for quite a while, but that fuzziness is (intermittently) gone now.

The fuzziness is described in bug 1599619 which is where we added the annotation. We can trivially just change the annotation to allow the test to exactly-match now, I guess.

One observation... in both this bug and bug 1894565, the test-failure-screenshots show that the text is not antialiased at all, which is a bit surprising/concerning I think.

I'm suspicious that these bugs and several others are showing new/unexpected results as of the infra changes in bug 1888460; I wonder if the enw worker machines we're using as part of that change have a different font configuration for some reason?

Depends on: 1895092
You need to log in before you can comment on or make changes to this bug.