1332586 - The transition is not smooth when enabling/disabling RDM

Status: RESOLVED DUPLICATE of bug 1278757

(DevTools :: Responsive Design Mode, defect, P2)

Description

[Mihai Boldan, Desktop QA [:mboldan]]

Attachment: Wrong transition.gif

[Affected versions]:
- Firefox 52.0a2 (2017-01-19), Firefox 53.0a1 (2017-01-19)

[Affected platforms]:
- Windows 10x64, Ubuntu 16.04x64, Mac OS X 10.12

[Steps to reproduce]:
1. Launch Firefox.
2. Enable RDM.
3. Disable RDM.

[Expected result]:
- The transition is smooth, without any glitches.

[Actual result]:
- The transition is not smooth.

[Regression range]:
- This is a regression.
Last good revision: c4449eab07d39e20ea315603f1b1863eeed7dcfe
First bad revision: 5a4cdb6dfb19b458229c60e0e19f083ba83d0f58
Pushlog:
https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=c4449eab07d39e20ea315603f1b1863eeed7dcfe&tochange=5a4cdb6dfb19b458229c60e0e19f083ba83d0f58

Comment 1

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

This is effectively the same issue as what's tracked in bug 1278757, so I'll dupe over there.  The attachment is nice to have, I'll mention it there.  Thanks for filing!

Comment 2

[Astley Chen (inactive)]

This bug was initiated as regression but the symptom is not that obvious.
Are we still going to keep tracking as regression ? Or should we instead block bug 1278757 and uplift patches accordingly once it's fixed?

Comment 3

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

(In reply to Astley Chen [:astley] (UTC+8) from comment #2)
> This bug was initiated as regression but the symptom is not that obvious.
> Are we still going to keep tracking as regression ? Or should we instead
> block bug 1278757 and uplift patches accordingly once it's fixed?

Well, it's definitely kind of a subjective thing.  On a fast machine, it's less noticeable I imagine.  Also, we landed a few speedups already under bug 1278757 that improve it.

I'd like keep improving it if we find ways to do so.  If there are concrete changes to make that will improve it, let's file those and have them block bug 1278757.

I think the current state is "good enough" that it would not require uplifting future improvements in this area, but there is still room to do better.

Comment 4

[Astley Chen (inactive)]

Attachment: Bug1332586-transition-artifact.mov

This was recorded on MBPR 15 with Firefox 52.0b5 running in parallel Windows 10 client OS. The symptom is able to reproduce on macOS either. It's quite obvious there is a temp flash of RWD window on top left and then re-positioned to center. As a fan of RWD, hope that there is still a change to improve it on FF52.

Comment 5

[Astley Chen (inactive)]

jryans, there is no remaining bugs that block bug 1278757. Is it still a work-in-progress ?
Regarding to this regression bug, I'll suggest it fix-optional and depend on your final call on FF52 cycle.
Thanks.

Comment 6

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

(In reply to Astley Chen [:astley] (UTC+8) from comment #5)
> jryans, there is no remaining bugs that block bug 1278757. Is it still a
> work-in-progress ?
> Regarding to this regression bug, I'll suggest it fix-optional and depend on
> your final call on FF52 cycle.
> Thanks.

(I am assuming the reason this bug is coming up even though it's resolved as a duplicate is for the bug flags, so I'm going to clear them.  If additional follow up is needed on this issue, let's discuss it in bug 1278757.)

I do have more ideas about the perf issue, so I added another bug under bug 1278757.  We have already uplifted some perf fixes to 52.  I think the current state of perf in 52 is good enough.  Could be better, but we can always make future improvements over time.