Closed Bug 1654179 Opened 4 years ago Closed 4 years ago

Intermittent gfx/layers/apz/test/mochitest/test_group_zoom-2.html | helper_zoom_with_dynamic_toolbar.html | The initial scale value should be restored to 1.0 - got 2.199819564819336, expected 1

Categories

(Core :: Panning and Zooming, defect, P5)

defect

Tracking

()

RESOLVED FIXED
81 Branch
Tracking Status
firefox-esr68 --- unaffected
firefox-esr78 --- unaffected
firefox78 --- unaffected
firefox79 --- unaffected
firefox80 --- fixed
firefox81 --- fixed

People

(Reporter: intermittent-bug-filer, Assigned: hiro)

References

(Regression)

Details

(Keywords: intermittent-failure, regression)

Attachments

(1 file)

Summary: Intermittent TV tier 2 gfx/layers/apz/test/mochitest/test_group_zoom-2.html | helper_zoom_with_dynamic_toolbar.html | The initial scale value should be restored to 1.0 - got 2.199819564819336, expected 1 → Intermittent gfx/layers/apz/test/mochitest/test_group_zoom-2.html | helper_zoom_with_dynamic_toolbar.html | The initial scale value should be restored to 1.0 - got 2.199819564819336, expected 1
Regressed by: 1652190
Has Regression Range: --- → yes
Keywords: regression

This is from bug 1652190 and has an elevated failure level. Please investigate when you find time for it.

Flags: needinfo?(hikezoe.birchill)

Set release status flags based on info from the regressing bug 1652190

Can't yet reproduce locally yet.

Assignee: nobody → hikezoe.birchill
Status: NEW → ASSIGNED
Flags: needinfo?(hikezoe.birchill)

Since it seems to be happening on linux pretty easily in CI you should be able to get a pernosco trace if you can't repro locally.

Yeah, I tried once, but it's not so often. https://treeherder.mozilla.org/#/jobs?repo=try&revision=cd5f553f96f7921f40f1a3d9a0ab22f96f3d4d76

In my understanding, pernosco session is not available on re-runs, right?

Maybe try with the test-verify job? I was hitting it first try on https://treeherder.mozilla.org/#/jobs?repo=try&group_state=expanded&revision=09e8f041a6abb955c0142da92f5e4142a0514fed . You can make a no-op change to test_group_zoom-2.html so the verify job picks it up as "modified". (And I don't actually know if pernosco will work on verify jobs, I haven't tried that myself before)

Okay, I could reproduce the failure locally on Android emulator with MOZ_CHAOSMODE=255 and --run-until-failure option.
(Note that pernosco wasn't triggered in TV runs).

What I observed there is that;

  1. After pinchZoomInWithTouch and promiseApzFlushedRepaints, the pinch-zoom looks still in the middle of the process, so that visualViewport.scale value doesn't get to the expected size (it's around 2.420)
  2. Even if I added a Promise call to wait for the scale value got > 2.40 after the promiseApzFlushedRepaints, it sometimes got stuck there, so presumably native touch events sometimes are dropped

So, basically it seems it's an underlying issue in apz_test_native_event_utils.js (or in apz_test_utils.js).

I am going to do a workaround the issue to use setResolutionAndScaleTo with a relatively small value.

Just pushed a try run with the workaround;
https://treeherder.mozilla.org/#/jobs?repo=try&revision=958dd2a55bec1ce6a8dba2121a5c305e47f49239

So that we can work around the race condition that the test proceeds in the
middle of the zoomin in the content.

Attachment #9167611 - Attachment description: Bug 1654179 - Use nsIDOMWindowUtils.setResolutionAndScaleTo and specifying a reletively small scale value instead of zoom in the content. r?botond → Bug 1654179 - Use nsIDOMWindowUtils.setResolutionAndScaleTo and specifying a relatively small scale value instead of zoom in the content. r?botond
Pushed by hikezoe.birchill@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/f46c4759b50c Use nsIDOMWindowUtils.setResolutionAndScaleTo and specifying a relatively small scale value instead of zoom in the content. r=botond
Status: ASSIGNED → RESOLVED
Closed: 4 years ago
Resolution: --- → FIXED
Target Milestone: --- → 81 Branch

Hit this once during a central-as-beta simulation today which is based on a changeset which has the fix for this bug as ancestor. Shall this bug get reopened or a new one created?

Flags: needinfo?(hikezoe.birchill)

Would you mind opening a new bug for the failure?

Flags: needinfo?(hikezoe.birchill)
See Also: → 1657228

The patch landed in nightly and beta is affected.
:hiro, is this bug important enough to require an uplift?
If not please set status_beta to wontfix.

For more information, please visit auto_nag documentation.

Flags: needinfo?(hikezoe.birchill)

Thank you, Julien and Ryan!

Flags: needinfo?(hikezoe.birchill)
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: