1842213 - Locally saved testcase from bug 1360222 will cause continuous memory increase and browser UI becomes unresponsive

Status: NEW

(Core :: Graphics: Canvas2D, defect)

Description

[Mayank Bansal]

1. Use Windows with d2d enabled (which should be default in Nightlies).
2. Download the testcase from bug 1360222 to your local machine (https://bugzilla.mozilla.org/attachment.cgi?id=9342747)
3. Close all Browser Windows (i.e. completely close the browser)
4. Double click on the locally saved testcase (such that this testcase is the only tab)

AR: Rapid continuous memory increase. The browser UI sort of becomes unresponsive.
ER: Not so

Profile: https://share.firefox.dev/3NZsfhv

Regression range:
https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=96c385db59064e3bbffaeeef6c4d9b7e083c5b08&tochange=02d875d4f7d769cd03be3175490377dc600355fe

Confirmed bug 1678652 (verified by setting webgl.disable-fail-if-major-performance-caveat= False and retesting)

Comment 1

[Mayank Bansal]

Attachment: about:support

Comment 2

[Mayank Bansal]

Attachment: bug1360222_standalone_testcase.html

Comment 3

[BugBot [:suhaib / :marco/ :calixte]]

Set release status flags based on info from the regressing bug 1678652

:jgilbert, since you are the author of the regressor, bug 1678652, could you take a look? Also, could you set the severity field?

For more information, please visit BugBot documentation.

Comment 4

[Mayank Bansal]

With webgl.disable-fail-if-major-performance-caveat= True, I get the following regression range:

https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=bcfbf7c9d10876c3e47a3cf44d6540b6e584ccf3&tochange=83f3726c41f9c91a9bd8e146a3422e287c4a9354

Comment 5

[Mayank Bansal]

Latest profile of opening the testcase from disk: https://share.firefox.dev/4aW2lox The profile is only 8s, but the browser was stuck for 1minute
Profile with opening the testcase from the browser: https://share.firefox.dev/3U4H3yY

Comment 7

[Kelsey Gilbert [:jgilbert]]

This is canvas2d and MakeTextRun, not webgl, based on the profile.
I think it's likely that the association with bug 1678652 is unrelated or spurious.

Comment 8

[Dan]

Attachment: Bug 1842213 - "WebDriver:ElementSendKeys" doesn't use implicit wait for element to become keyboard-interactable. r?whimboo

Comment 9

[Phabricator Automation]

Comment on attachment 9423905 [details]
Bug 1842213 - "WebDriver:ElementSendKeys" doesn't use implicit wait for element to become keyboard-interactable. r?whimboo

Revision D221662 was moved to bug 1842214. Setting attachment 9423905 [details] to obsolete.