1121164 - Black artifacts still seen when scrolling on Mac Nightly with Late-2013 MacBook Pro Retina

Status: RESOLVED WORKSFORME

(Core :: Graphics, defect)

Description

[(no longer active)]

+++ This bug was initially created as a clone of Bug #1079996 +++

This is a continuation of bug 1079996.

Comment 1

[Benoit Girard (:BenWa)]

We should start fresh here unless we have evidence that the bug you see is the same as bug 1079996. Black squares is a common symptom for graphics bugs.

Do you have any STR, regression-range or affected versions?

Comment 2

[Robert Kaiser]

Bug 1079996 comment #11 is giving some more important insight for this, esp. this one: It happens on 10.9 and 10.10 on "MacBook Pro (Retina, 15-inch, Late 2013)".

Until we know which specific piece of this is behind the issue, let's put that characteristic in the summary.

Comment 3

[(no longer active)]

(In reply to Benoit Girard (:BenWa) from comment #1)
> We should start fresh here unless we have evidence that the bug you see is
> the same as bug 1079996. Black squares is a common symptom for graphics bugs.
> 
> Do you have any STR, regression-range or affected versions?

Nothing more than what I have said in the other bug.

Comment 4

[Liz Henry (:lizzard) (relman/hg->git project)]

It sounds like many cases of this were fixed.  It's good that we have clues to how to reproduce the lingering issues. 

There are some possible steps to reproduce this here: https://bugzilla.mozilla.org/show_bug.cgi?id=1079996#c21   

Florin, can someone on your team try to reproduce this or do exploratory testing around scrolling in the test cases that Ehsan came up with?  Thanks!

Comment 5

[[SV Manager] Florin Mezei, QA (:FlorinMezei)]

We can have a look, but note that this is fairly low on our priority list, and this would very likely happen only sometime next week.

Comment 6

[Lawrence Mandel [:lmandel] (use needinfo)]

Given that this is a clone of bug 1079996, for which 35+ are affected, I'm marking 37 and 38 as affected. Ehsan, as you're still experiencing this issue, how often do you experience black artifacts? Do you think this is critical enough to push for a fix in 37 specifically?

Comment 7

[(no longer active)]

(In reply to Lawrence Mandel [:lmandel] (use needinfo) from comment #6)
> Given that this is a clone of bug 1079996, for which 35+ are affected, I'm
> marking 37 and 38 as affected. Ehsan, as you're still experiencing this
> issue, how often do you experience black artifacts?

I have been offline for the past three weeks, and have just started to use Nightly again as of this morning.  I have not yet experienced this on the latest Nightly since this morning, but I can't be certain whether the issue is fixed without more testing as it happens very intermittently.

> Do you think this is
> critical enough to push for a fix in 37 specifically?

Speaking as a user, yes, as this issue was very visible when using the browser.  But this may or may not be a problem hitting a majority of our users depending on what we have learned from this in the time that I was away.  I suggest asking someone from the Graphics team about our latest understanding of this issue.  Redirecting to Milan.

Comment 8

[Milan Sreckovic [:milan] (needinfo for best results)]

I haven't seen this on Nightly with my "Early-2013 MacBook Pro Retina", and I haven't heard anybody else mention it either (and we do have a majority of the graphics team in Toronto run on OS X.)

At this point, I'd probably stop tracking and wait a bit before WFM, if it doesn't come back.  If it does, we may steal Ehsan's laptop to investigate...

Comment 9

[[SV Manager] Florin Mezei, QA (:FlorinMezei)]

(In reply to Liz Henry :lizzard from comment #4)
> Florin, can someone on your team try to reproduce this or do exploratory
> testing around scrolling in the test cases that Ehsan came up with?  Thanks!

We've tried to reproduce this issue this week, using the latest Nightly (e10s on & off) and latest Aurora on 3 different machines:
- Early-2013 15" MacBook Pro Retina (Mac OS X 10.9.5)
- miniMac (Mac OS X 10.9.5)
- iMac (Mac OS X 10.10)

We've tried various pages (including pages mentioned in bug 1079996: TBPL, GMail, https://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html, https://bugzilla.mozilla.org/page.cgi?id=splinter.html&bug=1018320&attachment=8540683) and different types of scrolling (including <Space> & <Shift>+<Space>) with different window sizes. 

We could not reproduce the issue at all. Everything worked fine during scrolling.

Comment 10

[(no longer active)]

FWIW so far I have not reproduced this at least in e10s mode on the latest Nightly.

Comment 11

[(no longer active)]

Attachment: Screen Shot 2015-03-12 at 11.06.51 AM.png

Just got this again.  This is on a MacBook Pro (Retina, 13-inch, Mid 2014).

Comment 12

[Milan Sreckovic [:milan] (needinfo for best results)]

Latest nightly, I presume, so not the same as bug 1138934...

Comment 13

[(no longer active)]

Yep, latest as of yesterday.

Comment 14

[(no longer active)]

Clicking on the link I'm putting in the URL field reproduces this 2 or 3 times out of 5 times.  The gray rectangle shows up at the bottom left side of the screen.

Comment 15

[Markus Stange [:mstange]]

It turns out that the URL from comment 14 does not actually reproduce the bug. The gray rectangle that it shows is the gray background of a click-to-play plugin, and not a painting bug.

So we still don't have a way of reproducing this.

Comment 16

[Ryan VanderMeulen [:RyanVM]]

Ehsan, have you seen these artifacts recently?

Comment 17

[(no longer active)]

No.

Comment 18

[Ryan VanderMeulen [:RyanVM]]

Let's call it WFM then and file a new bug if it reappears. Thanks!