1537610 - Since last Firefox Nigthly update, when having two windows open side by side and having the mouse go over the "inactive" window (the windows that wasn't clicked last) the mouse and the active window act like it's going over the active window

Status: RESOLVED FIXED

(Core :: DOM: Events, defect, P2)

Description

[mastil.daniel]

User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:68.0) Gecko/20100101 Firefox/68.0

Steps to reproduce:

For example, I have youtube on one side and another page on the other, I last clicked on the youtube one and the put my mouse over the other one but it acts like I'm going over the youtube one

Actual results:

If I go over the other page and, for example, I go over the middle maybe the progress bar on the youtube video appear, clicking on it changes the page but it's all about the effects that going over the page does

Expected results:

It should work on the page I go over, not over the active page, it worked that way before, it makes sense for it to work that way

Comment 1

[Alice0775 White]

[Tracking Requested - why for this release]:

I can reproduce the issue on Nightly68.0a1 windows10.

STR

1. Open Firefox [Win1]

2. Open New window (Ctrl+N) [Win2]

3. Place these window like [Win1][Win2]

4. Activate [Win1]

5. Load Youtube video in [Win1] and playback

6. Move the mouse pointer out of the video so that video controller will be hidden

7. Mouse pointer over [Win2] and move mouse pointer at same height of the video

Actual Results:
The video controller appears on the video

Expected results:
Mouse pointer should not affect other window.

Regression window:
https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=2244c803a5d08806b9824af9f2f8700791034cd0&tochange=8cd5c2339ca11b720f8b0e067c463f66b1f0ad5c

Regressed by:
8cd5c2339ca1 Henri Sivonen — Bug 1535541 Use chrome-process TabParent focus instead of APZ focus for keyboard event routing. r=nika

Henri, Your patch seems to cause the regression, Can you please look into this?

Comment 2

[Henri Sivonen (:hsivonen)]

The patch for bug 1536730 should fix this. (I.e. fix-wise this is a duplicate but not symptom-wise.)

Comment 3

[Henri Sivonen (:hsivonen)]

Filed bug 1537680 about a real fix covering out-of-process iframes. (The patch for bug 1536730 will fix this for the common case but not for out-of-process iframes in an unfocused window.)

Comment 4

[Henri Sivonen (:hsivonen)]

Also: Good observation. Thank you for filing and identifying the regression!

Comment 5

[Julien Cristau [:jcristau] (back April 22)]

Per comment 2 this should be fixed now.