720676 - 64-bit crash @ powf

Status: RESOLVED INCOMPLETE

(Core :: Graphics, defect, critical)

Description

[Scoobidiver (away)]

It's #33 top browser crasher in 12.0a1.

It's a new crash signature that first appeared in 11.0a1/20111207.
The regression range might be:
http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=338fae43b9d0&tochange=489f2d51b011

It happens with recent graphics drivers from any vendors.

Signature 	powf More Reports Search
UUID	5be3beea-2720-4620-be9d-ed2442120123
Date Processed	2012-01-23 17:45:51
Uptime	8622
Last Crash	6.5 days before submission
Install Age	12.2 hours since version was first installed.
Install Time	2012-01-23 05:32:43
Product	Firefox
Version	12.0a1
Build ID	20120121121718
Release Channel	nightly
OS	Windows NT
OS Version	6.1.7601 Service Pack 1
Build Architecture	amd64
Build Architecture Info	family 6 model 42 stepping 7
Crash Reason	0x00000879 / 0x00000000
Crash Address	0x7fefdd9cacd
App Notes 	
AdapterVendorID: 0x10de, AdapterDeviceID: 0x0e22, AdapterSubsysID: 00000000, AdapterDriverVersion: 8.17.12.8562
D2D? D2D+
DWrite? DWrite+
D3D10 Layers? D3D10 Layers+
EMCheckCompatibility	True

Frame 	Module 	Signature [Expand] 	Source
0 	KERNELBASE.dll 	RaiseException 	
1 	KERNELBASE.dll 	RaiseException 	
2 	dxgi.dll 	powf 	
3 	dxgi.dll 	CD3D10Device::RemoveDevice 	
4 	crypt32.dll 	_GSHandlerCheck_SEH 	
5 	d3d10_1core.dll 	CDevice::DriverInternalError 	
6 	dxgi.dll 	operator delete[] 	
7 	dxgi.dll 	CDXGISwapChainBase<IDXGISwapChain>::PresentImpl 	

More reports at:
https://crash-stats.mozilla.com/report/list?signature=powf

Comment 1

[Benjamin Smedberg]

Is this on 64-bit windows only?

Comment 2

[Scoobidiver (away)]

(In reply to Benjamin Smedberg  [:bsmedberg] from comment #1)
> Is this on 64-bit windows only?
Yes.

Comment 3

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

Sounds reminiscent of bug 635617: 64-bit specific Windows crash in a math function.

Can the crash be reproduced in a MSVC 2010 build?

Comment 4

[Makoto Kato [:m_kato]]

(In reply to Benoit Jacob [:bjacob] from comment #3)
> Sounds reminiscent of bug 635617: 64-bit specific Windows crash in a math
> function.
> 
> Can the crash be reproduced in a MSVC 2010 build?

No. This exception code is 0x00000879 by CDevice::DriverInternalError.

dup of bug 612120?

Comment 5

[Brian Carpenter [:geeknik]]

I'm seeing this crash in the latest Win64 Nightly build.

STR:
Set layers.offmainthreadcomposition.enabled to True
Restart Browser
Minimize Browser
Maximize Browser
Wait for crash

Crash Reports:
https://crash-stats.mozilla.com/report/index/f24c0266-ccd6-47bf-b0a3-6285d2130826
https://crash-stats.mozilla.com/report/index/113095cc-f74a-4c12-872a-286752130826
https://crash-stats.mozilla.com/report/index/3df3fdbd-fe48-4a7f-a9d6-3b31f2130826

I've attached a log from WinDBG, but it looks like it may be an issue with Avast and not Firefox. Can someone else chime and take a look?

Comment 6

[Brian Carpenter [:geeknik]]

Attachment: Windbg log file

Comment 7

[Makoto Kato [:m_kato]]

(In reply to Brian Carpenter [:geeknik] from comment #5)
> I'm seeing this crash in the latest Win64 Nightly build.
> 
> STR:
> Set layers.offmainthreadcomposition.enabled to True
> Restart Browser
> Minimize Browser
> Maximize Browser
> Wait for crash
> 
> Crash Reports:
> https://crash-stats.mozilla.com/report/index/f24c0266-ccd6-47bf-b0a3-
> 6285d2130826
> https://crash-stats.mozilla.com/report/index/113095cc-f74a-4c12-872a-
> 286752130826
> https://crash-stats.mozilla.com/report/index/3df3fdbd-fe48-4a7f-a9d6-
> 3b31f2130826
> 
> I've attached a log from WinDBG, but it looks like it may be an issue with
> Avast and not Firefox. Can someone else chime and take a look?

I think that your issue is bug 903842 or new bug with OMTC.

Comment 8

[Makoto Kato [:m_kato]]

I filed Brian's issue as bug 909607.

Comment 9

[Anthony Hughes (:ashughes) [QA]]

Closing this bug report as incomplete since this has no recent reports with supported Firefox versions. Please reopen this bug report if you can reproduce this crash.