Intermittent accessible/tests/browser/scroll/browser_test_scroll_bounds.js | Timeout while waiting for cache update
Categories
(Core :: Disability Access APIs, defect)
Tracking
()
| Tracking | Status | |
|---|---|---|
| firefox160 | --- | fixed |
People
(Reporter: morgan, Assigned: morgan)
References
(Blocks 1 open bug)
Details
(Keywords: perf-alert, Whiteboard: [a11y-flakytest-fix])
Attachments
(1 external file)
Test: accessible/tests/browser/scroll/browser_test_scroll_bounds.js
Failure message: Timeout while waiting for cache update
This failure occurred 59 times out of 2,039 runs (2.89%) between 2026-09-21 and 2026-10-11.
| Assignee | ||
Comment 1•1 month ago
|
||
Historically, we've queued an update for ScrollPosition in two cases:
- When an accessible's overflow styling changes
- When a scroll is performed
We also explicitly avoid pushing a scroll update when the scroll range
for a given acc is computed to be (0, 0) since we use the existence of
the scroll key in the parent process as an indicator of whether the
given acc is a scroll container.
This causes us to occasionally misrepresent scrollable containers as
non-scrollable in the parent process.
Imagine a document is created with scrollable overflow styling,
but does not initially have enough content to generate a scroll bar
(i.e. its scroll range is 0x0).
Then, the content inside the doc expands and creates a
nonzero scroll range.
We'd continue to assert in the parent process that this doc
isn't scrollable because no style change has happened and no
scroll has occurred.
With this patch, we now queue an additional update to scroll position
when witnessing the MozScrolledAreaChanged event from layout.
This event is fired when the scroll range of the root scroll frame
is altered due to the size of its internal content changing in reflow.
This caused intermittent failures in browser_test_scroll_bounds.js. A
new test case below exercises this functionality explicitly.
Comment 3•1 month ago
|
||
| bugherder | ||
Comment 4•27 days ago
|
||
Perfherder has detected a talos performance change from push f4ee789fe4f3d4a5d695c4acedc1860fe4d2d928.
No action is required from the author; this comment is provided for informational purposes only.
| Improvement | Test | Platform | Options | Absolute values [old vs new] |
|---|---|---|---|---|
| 17% | perf_reftest_singletons id-getter-1.html (doc) | linux2404-64-shippable | e10s fission stylo webrender | 327.38 ms -> 272.88 ms |
Need Help or Information?
If you have any questions, please reach out to acseh@mozilla.com. Alternatively, you can find help on Slack by joining #perf-help, and on Matrix you can find help by joining #perftest.
Details of the alert can be found in the alert summary, including links to graphs and comparisons for each of the affected tests.
Updated•19 days ago
|
Description
•