1122035 - Content is not shown in the most recent Nightly with e10s enabled if about:newtab wasn't loaded first (win7 on parallels for mac)

Status: RESOLVED DUPLICATE of bug 1107718

(Core :: Graphics, defect)

Description

[Denis]

Attachment: Parallels.png

User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/40.0.2214.69 Safari/537.36

Steps to reproduce:

1. Install latest Nightly (38.0a1 2015.01.15), navigate to some site
2. Restart program (with option to re-open the tabs so it starts from the site loaded earlier)


Actual results:

Site seems to be loaded but instead of content I can see only that "wheel", see attached screenshot.  

Now if I open about:newtab page, navigate to some site picking one from that quick-access table using mouse then it loads fine. Any other tab opened after that also seems working properly.

Content also disappears when I open error console.

I'm not sure if it matters but I use Parallels Desktop on Mac to run Windows 7.  The bug appears only there. I don't have any other Windows machines to test it there. Mac OS X and Linux seem working fine with the latest Nightly.


Expected results:

Pages should be rendered properly.

Comment 1

[:Gijs (he/him)]

Can you reproduce if you create a clean profile?

If so, can you try to find a regression window using mozregression ? ( http://mozilla.github.io/mozregression/ )

Comment 2

[Denis]

Yes, I can reproduce it with clean profile.

Here is the result of mozregression run:

98:59.82 LOG: MainThread Bisector INFO Narrowed inbound regression window from [
20ab622b, 6c673d6a] (3 revisions) to [6c8f62d7, 6c673d6a] (2 revisions) (~1 step
s left)
98:59.82 LOG: MainThread Bisector INFO Oh noes, no (more) inbound revisions :(
98:59.82 LOG: MainThread Bisector INFO Last good revision: 6c8f62d7fd1c
98:59.84 LOG: MainThread Bisector INFO First bad revision: 6c673d6a2fb8
98:59.84 LOG: MainThread Bisector INFO Pushlog:
https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=6c8f62
d7fd1c&tochange=6c673d6a2fb8


Note: It wasn't always as I described earlier. Sometimes, during the regression runs the browser just crashed when I tried to navigate somewhere. Also, I had to set browser.tabs.remote.autostart to true every time manually because it wasn't enabled by default.

How else can I help?

Comment 3

[:Gijs (he/him)]

(In reply to Denis from comment #2)
> Yes, I can reproduce it with clean profile.
> 
> Here is the result of mozregression run:
> 
> 98:59.82 LOG: MainThread Bisector INFO Narrowed inbound regression window
> from [
> 20ab622b, 6c673d6a] (3 revisions) to [6c8f62d7, 6c673d6a] (2 revisions) (~1
> step
> s left)
> 98:59.82 LOG: MainThread Bisector INFO Oh noes, no (more) inbound revisions
> :(
> 98:59.82 LOG: MainThread Bisector INFO Last good revision: 6c8f62d7fd1c
> 98:59.84 LOG: MainThread Bisector INFO First bad revision: 6c673d6a2fb8
> 98:59.84 LOG: MainThread Bisector INFO Pushlog:
> https://hg.mozilla.org/integration/mozilla-inbound/
> pushloghtml?fromchange=6c8f62
> d7fd1c&tochange=6c673d6a2fb8
> 
> 
> Note: It wasn't always as I described earlier. Sometimes, during the
> regression runs the browser just crashed when I tried to navigate somewhere.
> Also, I had to set browser.tabs.remote.autostart to true every time manually
> because it wasn't enabled by default.
> 
> How else can I help?

This is already very helpful!

Fixing that link (thanks, terminal software everywhere...):

https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=6c8f62d7fd1c&tochange=6c673d6a2fb8

Denis, can you also copy/paste the contents of about:support's graphics section here?

Bas, all the bugs in the regression window are yours... can you take a look which one is likely to have caused this issue? :-)

Comment 4

[:Gijs (he/him)]

[Tracking Requested - why for this release]:
e10s regression... dunno if we should track for 37 or 38, but either way.

Comment 5

[Denis]

Here is the content of about:support's Graphics section



Graphics
--------

Adapter Description: Parallels Display Adapter (WDDM)
Adapter Drivers: prl_umdd prl_umdd10
Adapter RAM: Unknown
ClearType Parameters: Gamma: 2200 Pixel Structure: R ClearType Level: 100 Enhanced Contrast: 50
Device ID: 0x4005
Direct2D Enabled: Blocked for your graphics card because of unresolved driver issues.
DirectWrite Enabled: false (6.2.9200.16571)
Driver Date: 12-19-2014
Driver Version: 10.1.28859.0
GPU #2 Active: false
GPU Accelerated Windows: 0/1 Basic (OMTC) Blocked for your graphics card because of unresolved driver issues.
Subsys ID: 04001ab8
Vendor ID: 0x1ab8
WebGL Renderer: Blocked for your graphics card because of unresolved driver issues.
windowLayerManagerRemote: true
AzureCanvasBackend: skia
AzureContentBackend: cairo
AzureFallbackCanvasBackend: cairo
AzureSkiaAccelerated: 0

Comment 6

[Bas Schouten (:bas.schouten)]

This was probably bug https://bugzilla.mozilla.org/show_bug.cgi?id=1107718. Someone who runs parallels on the gfx team might be able to check this.

Comment 7

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

Dan, do you run parallels?

Comment 8

[u480271]

I don't at the moment but I have access to it. Is this bug running windows in a parallels hosted on mac?

Comment 9

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

Not tracking for 37 as e10s is disabled for Aurora 37.

Comment 10

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

(In reply to Dan Glastonbury :djg :kamidphish from comment #8)
> I don't at the moment but I have access to it. Is this bug running windows
> in a parallels hosted on mac?

Correct.

Comment 11

[Brad Lassey [:blassey] (use needinfo?)]

Duping based on Bas's comment. Please reopen if you see this again