865182 - Element with CSS transitions sometimes "jumping"

Status: RESOLVED FIXED

(Core :: CSS Parsing and Computation, defect)

Description

[Peter Forgacs]

Attachment: progressBar.html

User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.31 (KHTML, like Gecko) Chrome/26.0.1410.64 Safari/537.31



Actual results:

If we add transition to an element, then it sometimes jumps to the new position, then instantly jump back and start the animation from the original position.

The correct behavior would be doing the animation without jumping to the new position.

I have attached an HTML where I can reproduce this issue all time.
The progress bar is jumping forward and back (every run 3-4 times).

A much simpler example is here:
http://jsfiddle.net/RwtHn/5/
If you click to the right bottom area, the red ball follows your mouse cursor.
Playing with this it is possible to reproduce the jumping effect after a few minutes.

I have tested this issue with other browsers too (Chrome, IE10), and they don't have this bug.

Comment 1

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Guessing this could be related to bug 849399.

Comment 2

[Loic]

(In reply to David Baron [:dbaron] (don't cc:, use needinfo? instead) from comment #1)
> Guessing this could be related to bug 849399.

I don't think so because the current issue is present before the regression range of bug 849399.

Anyway I think there is a regression here, but it's hard to track. Indeed, with FF20, it's easy to observe several "jumps" during only one animation of the progress bar in the testcase, but jumps are very rare with older versions (near May 2012). So I would say a 2nd regression has increased the frequency of jumps after the 1st regression.

To track the 1st regression, I needed to run 5 or 6 animations of the progress bar to observe only one jump.

1st possible regression range:
m-c
good=2012-05-19
bad=2012-05-20
http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=642d1a36702f&tochange=0e2cc686b07b

Comment 3

[Alice0775 White]

The remarkable regression range is as follows.

Regression window(m-c)
Good:
http://hg.mozilla.org/mozilla-central/rev/c14e2d5f17de
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/17.0 Firefox/17.0 ID:20120813105346
Bad:
http://hg.mozilla.org/mozilla-central/rev/75cdb3f932c6
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/17.0 Firefox/17.0 ID:20120813110945
Pushlog:
http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=c14e2d5f17de&tochange=75cdb3f932c6


Regression window(m-i)
Good:
http://hg.mozilla.org/integration/mozilla-inbound/rev/9e0799cf1096
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/17.0 Firefox/17.0 ID:20120812220843
Bad:
http://hg.mozilla.org/integration/mozilla-inbound/rev/6a0b53d78465
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:17.0) Gecko/17.0 Firefox/17.0 ID:20120813031248
Pushlog:
http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=ed614ea130c0&tochange=6a0b53d78465

Comment 4

[Loic]

With bug 849399 fixed in Nightly, maybe this one is fixed too.

Comment 5

[(Out Sep 17-23) Ryan VanderMeulen]

Confirmed that the attached testcase started behaving as expected after bug 849399 landed.