It first appeared in 16.0a1/20120716. The regression range is: http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=57abb5f70e01&tochange=4cffe2b37d0c It's likely a regression from bug 598868. Signature nsHTMLMediaElement::EndMediaStreamPlayback() More Reports Search UUID 36691628-1eff-4059-b10a-bb6e02120716 Date Processed 2012-07-16 14:25:11 Uptime 92 Last Crash more than 3 months before submission Install Age 3.1 minutes since version was first installed. Install Time 2012-07-16 14:20:51 Product Firefox Version 16.0a1 Build ID 20120716030546 Release Channel nightly OS Windows NT OS Version 6.1.7601 Service Pack 1 Build Architecture x86 Build Architecture Info GenuineIntel family 6 model 37 stepping 2 Crash Reason EXCEPTION_ACCESS_VIOLATION_READ Crash Address 0x8 User Comments Just testing the WebRTC demo page; Sequence: - start video - stop video - start audio - stop audio => crash App Notes AdapterVendorID: 0x1002, AdapterDeviceID: 0x68c1, AdapterSubsysID: 00000000, AdapterDriverVersion: 8.723.8.2000 D3D10 Layers? D3D10 Layers- D3D9 Layers? D3D9 Layers- EMCheckCompatibility True Adapter Vendor ID 0x1002 Adapter Device ID 0x68c1 Total Virtual Memory 4294836224 Available Virtual Memory 3967201280 System Memory Use Percentage 72 Available Page File 4346949632 Available Physical Memory 1155796992 Frame Module Signature Source 0 xul.dll nsHTMLMediaElement::EndMediaStreamPlayback content/html/content/src/nsHTMLMediaElement.cpp:2579 1 xul.dll nsHTMLMediaElement::AbortExistingLoads content/html/content/src/nsHTMLMediaElement.cpp:581 2 xul.dll nsHTMLMediaElement::Load content/html/content/src/nsHTMLMediaElement.cpp:701 3 xul.dll nsHTMLMediaElement::SetAttr content/html/content/src/nsHTMLMediaElement.cpp:1843 4 xul.dll nsGenericHTMLElement::SetAttr content/html/content/src/nsGenericHTMLElement.h:168 5 xul.dll nsGenericHTMLElement::SetAttrHelper content/html/content/src/nsGenericHTMLElement.cpp:2829 6 xul.dll nsHTMLMediaElement::SetSrc content/html/content/src/nsHTMLMediaElement.cpp:475 7 xul.dll NS_InvokeByIndex_P xpcom/reflect/xptcall/src/md/win32/xptcinvoke.cpp:70 8 xul.dll XPCWrappedNative::CallMethod js/xpconnect/src/XPCWrappedNative.cpp:2382 9 xul.dll XPCWrappedNative::SetAttribute js/xpconnect/src/xpcprivate.h:2815 10 xul.dll XPC_WN_GetterSetter js/xpconnect/src/XPCWrappedNativeJSOps.cpp:1514 11 mozjs.dll js::InvokeKernel js/src/jsinterp.cpp:344 12 mozjs.dll js::Invoke js/src/jsinterp.cpp:387 13 mozjs.dll js::Shape::set js/src/jsscopeinlines.h:296 14 mozjs.dll js::baseops::SetPropertyHelper js/src/jsobj.cpp:4942 15 mozjs.dll js::SetPropertyOperation js/src/jsinterpinlines.h:350 16 mozjs.dll js::Interpret js/src/jsinterp.cpp:2355 17 mozjs.dll js::InvokeKernel js/src/jsinterp.cpp:355 18 mozjs.dll js::Invoke js/src/jsinterp.cpp:387 19 mozjs.dll JS_CallFunctionValue js/src/jsapi.cpp:5564 ... More reports at: https://crash-stats.mozilla.com/report/list?signature=nsHTMLMediaElement%3A%3AEndMediaStreamPlayback%28%29 https://crash-stats.mozilla.com/report/list?signature=nsHTMLMediaElement%3A%3AEndMediaStreamPlayback

Anant - Any ideas?

One of the crash reports had this comment: Just testing the WebRTC demo page; Sequence: - start video - stop video - start audio - stop audio => crash - but I wasn't able to reproduce.

I was able to reproduce using the WebRTC demo page, using the step in comment 2.

|mVideoFrameContainer| is not null-checked, and probably should be. Here is a possible fix. Feel free to forward if you think someone is more appropritate for the review.

Comment on attachment 643499 [details] [diff] [review] Fix a crash in nsHTMLMediaElement::EndMediaStreamPlayback r= Switching review to roc and/or cjones (who reviewed the patch that last changed this line); revision 99337, bug 598868

Comment on attachment 643499 [details] [diff] [review] Fix a crash in nsHTMLMediaElement::EndMediaStreamPlayback r= Review of attachment 643499 [details] [diff] [review]: ----------------------------------------------------------------- ::: content/html/content/src/nsHTMLMediaElement.cpp @@ +2576,5 @@ > if (mPaused) { > GetMediaStream()->ChangeExplicitBlockerCount(-1); > } > + if (mVideoFrameContainer) { > + mVideoFrameContainer->GetImageContainer()->SetCurrentImage(nsnull); Move this up be inside if (container) and use 'container'

Addressed roc's comments.

Can we land the fix for this? I'm blocked on some of my testing due to this bug.

Jason, sorry about that. I'll rebase it tomorrow, and ask someone to land it.

Busted. I tested this with my USB headset and tried to access the audio piece of Anant's page https://people.mozilla.com/~anarayanan/gum_test.html and repeatedly was getting NO_DEVICES_FOUND. I tested this on the 8/1/2012 build before this fix landed, did not get this error, and successfully could hear and could playback sound I generated. Filed a followup bug in bug 781705.

Disregard. I realized that the issue I hit is already a known issue. Will retest now.

We do not have an attached testcase here and the one from the demo page has been changed a lot since then. So I don't think we can have a crashtest here.