647942 - [D3D9] Occasional full system restart with HW acceleration turned on

Status: RESOLVED WONTFIX

(Core :: Graphics, defect)

Description

[Justin Lebar (not reading bugmail)]

My brother experienced full system restarts in FF4.  When he disabled hardware acceleration (it was on by default), the problem stopped.

He's running Windows, but I don't know if it's XP or 7.  I'll comment here as soon as he gets back to me.

Here's his about:support graphics section, apparently after he disabled hw acceleration (linebreaks hopefully added in the right places):

Adapter 
  Description  NVIDIA GeForce 7600 GT
  Vendor ID10de
  Device ID0391
Adapter
  RAM Unknown
Adapter
  Drivers nv4_dispDriver Version6.14.12.6099
  Driver Date10-16-2010
Direct2D Enabled false
DirectWrite Enabled false (0.0.0.0, font cache n/a)
WebGL Renderer Google Inc. -- ANGLE -- OpenGL ES 2.0 (ANGLE 0.0.0.541)
GPU Accelerated Windows 0/2

Comment 1

[Justin Lebar (not reading bugmail)]

He's running XP.

Comment 2

[(mostly gone) XtC4UaLL [:xtc4uall]]

Has he opened certain Sites?
Would a Driver Update help?
http://www.nvidia.com/object/winxp-266.58-whql-driver.html

Comment 3

[Justin Lebar (not reading bugmail)]

A driver update probably would help, but I don't want him to do that until we figure out whether we want to blacklist the current driver.

He said it was reproducible logging into blogger.com, but was happening elsewhere too.

Comment 4

[Benoit Jacob [:bjacob] (mostly away)]

If we want to make progress on this one, the only way forward that I can see is to run this in some sort of DirectX debugger that would log all the DirectX calls to disk, so that information would survive the crash.

Comment 5

[Bas Schouten (:bas.schouten)]

(In reply to comment #4)
> If we want to make progress on this one, the only way forward that I can see is
> to run this in some sort of DirectX debugger that would log all the DirectX
> calls to disk, so that information would survive the crash.

Even then that's unlikely to be helpful. We should never be able to trigger this so likely it's just invalid driver behavior leading to this.

Comment 6

[Benoit Jacob [:bjacob] (mostly away)]

Sure, but it could still be interesting to know what is triggering this driver bug.

Comment 7

[timeless]

http://msdn.microsoft.com/en-us/library/bb173355%28v=vs.85%29.aspx sounds like it could be fun.

OTOH you might actually want to try to hook OutputDebugString (depends.exe does) with the debug version of directx:
http://msdn.microsoft.com/en-us/library/ee416587%28v=vs.85%29.aspx

Comment 8

[Justin Lebar (not reading bugmail)]

It seems that our two options are either to have my brother run an instrumented build or just close this INCOMPLETE.

Do you guys think this is worth investigating further?

Comment 9

[Virtual_ManPL [:Virtual] 🇵🇱 - (please needinfo? me - so I will see your comment/reply/question/etc.)]

I'm marking this bug as WONTFIX per bug #1318558 and bug #1343941.