1168451 - BSOD when playing Banana Bread game

Status: RESOLVED INCOMPLETE

(Core :: Graphics: CanvasWebGL, defect)

Description

[ssimon]

Actual Results:
After playing Banana Bread for less than a minute, the computer will blue screen.

Expected Results:
Browser game should not make computer blue screen.

Steps to Recreate:
1 Navigate to the Banana Bread game, click Play Demo
2 Click the first level - the arena, then pick the Low Resolution option
3 After game loads, walk around with WASD and fire the weapons with Left Click
4 Play for about 30 seconds, or until the game bluescreens the computer

NOTE: Banana Bread will not run in Chrome due to an issue with missing textures, so I was unable to confirm if this issue occurs in Chrome as well.  This bug happens does not seem to be affected by e10s, the bug happens when e10s is toggled on and when e10s is toggled off.

Comment 1

[ssimon]

Attachment: ff40_onyxia_DxDiag.txt

Comment 2

[ssimon]

Attachment: ff40_Onyxia_aboutsupport.txt

Comment 3

[u279076]

Could you provide a copy of the BSOD error text?

Comment 4

[ssimon]

Attachment: bsod.png

Comment 5

[ssimon]

(In reply to Anthony Hughes, QA Mentor (:ashughes) from comment #3)
> Could you provide a copy of the BSOD error text?

I've attached a screenshot of the bsod as bsod.png

Comment 6

[u279076]

Thanks, based on https://support.microsoft.com/en-us/kb/137539 it looks like that's a divide-by-zero memory corruption error. Do you have another system with a similar configuration (CPU/GPU/OS/Memory) to confirm if this is specific to Onyxia?

Comment 7

[Todd Brais]

Attachment: fx40_Bytor_Eomer_dxdiag.txt

Here is dxdiag info for Bytor system, with Eomer video card

Comment 8

[Todd Brais]

Attachment: fx40_Bytor_Eomer_support.txt

Comment 9

[Todd Brais]

Hi Anthony;
After checking this on another system; I ran into the same issue that was happening when Sean was regressing bug # 1167621: Dead Trigger 2 Demo becomes laggy when multiple assets (enemies) are on screen.
When testing the Banana Bread game on a different system the mouse sensitivity/mouse acceleration is way too high and makes the game unplayable. As a result of this I cannot thoroughly test this. 
 On a side note, I did not encounter the BSOD,
 I've also attached the DXDIAG and the about:support files for the new system that I tested. These files are labeled as:
fx40_Bytor_Eomer_dxdiag.txt
fx40_Bytor+Eomer_support.text   

As always; please let me know if you have any further questions


(In reply to Anthony Hughes, QA Mentor (:ashughes) from comment #6)
> Thanks, based on https://support.microsoft.com/en-us/kb/137539 it looks like
> that's a divide-by-zero memory corruption error. Do you have another system
> with a similar configuration (CPU/GPU/OS/Memory) to confirm if this is
> specific to Onyxia?

Comment 10

[u279076]

Hi Todd, based on the specs I have I'm not sure that creates an equal test-bed to Onyxia:
> Onyxia has Windows XP with 2x nVidia GeForce 7950 GT 512MB in SLI.
> Bytor has Windows XP with nVidia GeForce 6150 256MB integrated
> Eomer has a nVidia GeForce 7800GT 512MB

In addition, the drivers in use are not equal:
> Onyxia's about:support reports driver 6.14.12.6099 (Oct 16, 2010)
> Bytor's about:support reports driver 6.14.13.783 (Jan 31, 2013)

One option for testing might be to test using the same driver in Onyxia and see if the BSOD reproduces. Another is to put the two NVidia 7950 GT cards in SLI into another Windows XP system. Either way the variability in the systems needs to be reduced before this can be reliably debugged. 

My leaning right now is that Onyxia is using a system configuration that should not be supported but we need more information before that can be concluded.

Comment 11

[Todd Brais]

Anthony;
 Unfortunately what was used for testing is the closest we have to Onyxia. We currently do not have any other AMD SLI motherboards. With that being that, I will test re-test with the drivers mentioned in comment 10 

(In reply to Anthony Hughes, QA Mentor (:ashughes) from comment #10)
> Hi Todd, based on the specs I have I'm not sure that creates an equal
> test-bed to Onyxia:
> > Onyxia has Windows XP with 2x nVidia GeForce 7950 GT 512MB in SLI.
> > Bytor has Windows XP with nVidia GeForce 6150 256MB integrated
> > Eomer has a nVidia GeForce 7800GT 512MB
> 
> In addition, the drivers in use are not equal:
> > Onyxia's about:support reports driver 6.14.12.6099 (Oct 16, 2010)
> > Bytor's about:support reports driver 6.14.13.783 (Jan 31, 2013)
> 
> One option for testing might be to test using the same driver in Onyxia and
> see if the BSOD reproduces. Another is to put the two NVidia 7950 GT cards
> in SLI into another Windows XP system. Either way the variability in the
> systems needs to be reduced before this can be reliably debugged. 
> 
> My leaning right now is that Onyxia is using a system configuration that
> should not be supported but we need more information before that can be
> concluded.

Comment 12

[Todd Brais]

Attachment: fx40_Onyxia_dxdiag.txt

Dxdiag info for Onyxia (6/2/2015)

Comment 13

[Todd Brais]

(In reply to Anthony Hughes, QA Mentor (:ashughes) from comment #10)
> Hi Todd, based on the specs I have I'm not sure that creates an equal
> test-bed to Onyxia:
> > Onyxia has Windows XP with 2x nVidia GeForce 7950 GT 512MB in SLI.
> > Bytor has Windows XP with nVidia GeForce 6150 256MB integrated
> > Eomer has a nVidia GeForce 7800GT 512MB

I was able to reproduce the BSOD using the drivers in Onyxia
> 
> In addition, the drivers in use are not equal:
> > Onyxia's about:support reports driver 6.14.12.6099 (Oct 16, 2010)
> > Bytor's about:support reports driver 6.14.13.783 (Jan 31, 2013)
> 
> One option for testing might be to test using the same driver in Onyxia and
> see if the BSOD reproduces. Another is to put the two NVidia 7950 GT cards
> in SLI into another Windows XP system. Either way the variability in the
> systems needs to be reduced before this can be reliably debugged. 
> 
> My leaning right now is that Onyxia is using a system configuration that
> should not be supported but we need more information before that can be
> concluded.

Comment 14

[Todd Brais]

Attachment: Bytor_Palpatine Card__dxdiag_No BSOD.txt

dxdiag for Bytor system w/Palpatine card=No BSOD

Comment 15

[Todd Brais]

Attachment: about support bytor with palpatine.txt

About:Support file for Bytor system w/Palpatine card=No BSOD

Comment 16

[u279076]

Thanks for helping with this bug. Given that this is isolated to a very specific set of circumstances (ie. GPU, motherboard, and demo) I'm reluctant to investigate this further. 

I'm closing this bug as resolved incomplete. However, we should reopen this bug report if we start to see feedback from our users that this is a significant issue.