Closed
Bug 1404456
Opened 7 years ago
Closed 7 years ago
Crash in igd10umd32.dll on Youtube
Categories
(Core :: Graphics, defect, P1)
Tracking
()
RESOLVED
FIXED
mozilla58
Tracking | Status | |
---|---|---|
firefox-esr52 | --- | unaffected |
firefox55 | --- | unaffected |
firefox56 | + | fixed |
firefox57 | --- | fixed |
firefox58 | --- | fixed |
People
(Reporter: philipp, Assigned: jya, NeedInfo)
References
Details
(Keywords: crash, regression, Whiteboard: [gfx-noted])
Crash Data
[Tracking Requested - why for this release]:
these crashes seem to spring up on firefox 56 for win7 sp1 users in the browser & content process - so far they account for 2.2% of crashes in 56.0.
This bug was filed from the Socorro interface and is
report bp-d284a25d-a186-4e3e-813c-8001e0170929.
=============================================================
Crashing Thread (0)
Frame Module Signature Source
Ø 0 igd10umd32.dll igd10umd32.dll@0x2235b5
Ø 1 igd10umd32.dll igd10umd32.dll@0x22544f
Ø 2 igd10umd32.dll igd10umd32.dll@0x207500
Ø 3 igd10umd32.dll igd10umd32.dll@0x207407
Ø 4 igd10umd32.dll igd10umd32.dll@0x2070a3
Ø 5 igd10umd32.dll igd10umd32.dll@0x943d41
Ø 6 igd10umd32.dll igd10umd32.dll@0x944036
Ø 7 igd10umd32.dll igd10umd32.dll@0x928c39
Ø 8 igd10umd32.dll igd10umd32.dll@0x940328
Ø 9 igd10umd32.dll igd10umd32.dll@0x941784
Ø 10 igd10umd32.dll igd10umd32.dll@0x1f861e
Ø 11 igd10umd32.dll igd10umd32.dll@0x946eb9
user comments suggest this started happening after the 56 update and many are referring to youtube playback - the comment to this report says: "I have gotten crashes all day on youtube ever since installing the latest Firefox update. The youtube video screen goes solid green, then the whole tab, or all tabs if there are multiple tabs open, will scramble, then it crashes."
adapter & driver version facets indicate that this is limited to a particular device family, so maybe some selected blocklisting can help here:
Adapter device id facet
1 0x0116 71 36.22 %
2 0x0126 52 26.53 %
3 0x0102 43 21.94 %
4 0x0106 23 11.73 %
5 0x2a42 4 2.04 %
6 0x2e22 2 1.02 %
7 0x0112 1 0.51 %
Adapter driver version facet
1 8.15.10.2418 162 82.65 %
2 8.15.10.2342 10 5.10 %
3 8.15.10.2353 10 5.10 %
4 8.15.10.1840 4 2.04 %
5 8.15.10.2405 4 2.04 %
6 8.15.10.1808 2 1.02 %
7 8.15.10.2345 2 1.02 %
8 8.15.10.2361 1 0.51 %
9 8.15.10.2372 1 0.51 %
https://crash-stats.mozilla.com/search/?signature=%3Digd10umd64.dll&signature=%3Digd10umd32.dll&product=Firefox&version=56.0&date=%3E%3D2017-09-20&_facets=signature&_facets=user_comments&_facets=adapter_vendor_id&_facets=build_id&_facets=platform_pretty_version&_facets=install_time&_facets=release_channel&_facets=adapter_device_id&_facets=adapter_driver_version&_facets=app_notes#facet-user_comments
Flags: needinfo?(milan)
These are mostly sandybridge, but that could be just the distribution. Any other correlation? These are basically the crashes that don't have any part of the stack in our code, purely in the driver. Not sure what we can do about those.
Jean-Yves, perhaps this is a Intel version of the problems we've seen with other vendors, we just don't get a usable stack? Some crashes are in MediaPDecoder thread (https://crash-stats.mozilla.com/report/index/b3f4d13b-22aa-4549-84da-e66590170928)
One of the users mentions multiple video downloads, as well as "going back to 55.0.2 as 55.0.3 also had problems", but it isn't clear if 55.0.3 has the same problem as we're seeing here.
Flags: needinfo?(milan) → needinfo?(jyavenard)
Priority: -- → P1
Whiteboard: [gfx-noted]
Assignee | ||
Comment 2•7 years ago
|
||
The pardon we had issue earlier on with 56 beta was with the D3D11 dxva decoder, this isn't available on win7.
If it had been 57,i would said it was just a consequence of bug 1393392 which removed all the blacklists.
Can't make sense of those crashes. The main thread crashing too? No decoding is ever done on that thread
Flags: needinfo?(jyavenard)
Assignee | ||
Comment 3•7 years ago
|
||
s/pardon/problem above...
The changes that were made in 56, reverted to an identical behaviour to what was 55.
Or could it be that retrieving the vendor id of those graphic adapters is causing the crash? has that type of things be seen before?
Reporter | ||
Comment 4•7 years ago
|
||
when filtering for the signature's characteristics on nightly (win7sp1, sandybridge, 8.15.10 driver)[1] the first occurrence of it was on 56.0a1 build 20170622030208, and it continued fairly frequently thereafter...
this would be the pushlog to the day before: https://hg.mozilla.org/mozilla-central/pushloghtml?startdate=2017-06-21&tochange=edb7e1ddd9b61e2af2a75cfe5baa0f92a54a2716
...or the pushlog to 3 days before: https://hg.mozilla.org/mozilla-central/pushloghtml?startdate=2017-06-19&tochange=edb7e1ddd9b61e2af2a75cfe5baa0f92a54a2716
[1] https://crash-stats.mozilla.com/search/?platform_version=%3D6.1.7601%20Service%20Pack%201&adapter_driver_version=~8.15.10.&signature=%3Digd10umd64.dll&signature=%3Digd10umd32.dll&adapter_device_id=0x0116&adapter_device_id=0x0126&adapter_device_id=0x0102&adapter_device_id=0x0106&release_channel=nightly&product=Firefox&date=%3E%3D2017-03-29T21%3A18%3A16.000Z&date=%3C2017-09-29T21%3A18%3A16.000Z&_sort=-date&_facets=signature&_facets=user_comments&_facets=adapter_vendor_id&_facets=build_id&_facets=platform_pretty_version&_facets=install_time&_facets=release_channel&_facets=adapter_device_id&_facets=adapter_driver_version&_facets=app_notes&_facets=platform_version&_facets=version&_columns=date&_columns=signature&_columns=product&_columns=version&_columns=build_id&_columns=platform#facet-build_id
Reporter | ||
Comment 5•7 years ago
|
||
...and these were just the A/V & Graphics changes landing before the June 22 build: https://mzl.la/2xFEhXa
New version of the cubeb library?
(In reply to Jean-Yves Avenard [:jya] from comment #3)
> s/pardon/problem above...
>
> The changes that were made in 56, reverted to an identical behaviour to what
> was 55.
>
> Or could it be that retrieving the vendor id of those graphic adapters is
> causing the crash? has that type of things be seen before?
Seems all the crashes are content/gpu process. We'd be checking the vendor on the browser process. Ignoring that aspect of it though, we do load dxgi.dll as a part of identifying the graphics setup, but only on dual GPU configurations.
All Nvidia/Optimus crashes are GPU process, which is probably not that useful info.
Comment 8•7 years ago
|
||
Looks like a fairly high rate of crashes on 56, around the #11 top crash on 56.0 release.
Let's track this for now, maybe we will end up with a dot release if we can figure out a fix or quick blocklist.
I'm not sure this should block our turning on updates to 56, though, since it is not crashing the browser. Milan, what do you think ?
Flags: needinfo?(milan)
This signature is a bit of a catch all, so it isn't 100% for sure that it's the same crash. Either way, not much to go on (I'll look at bug 1405110) and not a browser crash, so I wouldn't block.
Flags: needinfo?(milan)
Updated•7 years ago
|
Reporter | ||
Comment 10•7 years ago
|
||
62.90% of crashes happen in the content process, the rest in the main browser process - only a single one in the gpu process on 56
Reporter | ||
Comment 11•7 years ago
|
||
(was basing that on the search link from comment #0)
Flags: needinfo?(milan)
Reporter | ||
Comment 12•7 years ago
|
||
we also got a number of reports about green screens instead of video content on sumo: https://support.mozilla.org/en-US/questions/firefox?tagged=bug1404456&show=all
Most of these devices should be blacklisted for the D3D11 path for drivers <= 8.15.10.4444. Gerald is preparing a patch.
Assignee | ||
Comment 14•7 years ago
|
||
(In reply to Anthony Jones (:kentuckyfriedtakahe, :k17e) from comment #13)
> Most of these devices should be blacklisted for the D3D11 path for drivers
> <= 8.15.10.4444. Gerald is preparing a patch.
The crash is in the content process.
It is *not* a matter of D3D9 decoder VS D3D11 decoder.
Modifying the blacklist to force using the D3D9 decoder will not help.
Assignee | ||
Comment 15•7 years ago
|
||
(In reply to Anthony Jones (:kentuckyfriedtakahe, :k17e) from comment #13)
> Most of these devices should be blacklisted for the D3D11 path for drivers
> <= 8.15.10.4444. Gerald is preparing a patch.
All crashes occur on Windows 7. There's no D3D11 decoder available on Win7
http://searchfox.org/mozilla-central/source/dom/media/platforms/wmf/WMFVideoMFTManager.cpp#485
Assignee | ||
Comment 16•7 years ago
|
||
The full backtrace
0:000> k =005ec3a8
# ChildEBP RetAddr
WARNING: Stack unwind information not available. Following frames may be wrong.
00 005ec3a8 0225c514 igd10umd32+0x2235b5
01 005ec3e0 6004981c 0x225c514
02 005ec3e8 600124b8 d3d11!CD3D11FormatHelper::GetParentFormat+0x19
03 005ec44c 600388fc d3d11!CResource<ID3D11Texture2D>::CResource<ID3D11Texture2D>+0xd2
04 005ec470 6000eafb d3d11!TCLSWrappers<CTexture2D>::CLSFinalConstructFn+0x44
05 005ec4b4 6000b132 d3d11!CLayeredObjectWithCLS<CTexture2D>::FinalConstruct+0x71
06 005ec4e8 60002314 d3d11!CLayeredObjectWithCLS<CTexture2D>::CreateInstance+0xa7
07 005ec7c0 5fff250f d3d11!CDevice::CreateLayeredChild+0x170
08 005ec7d8 60024599 d3d11!CBridgeImpl<ID3D11LayeredDevice,ID3D11LayeredDevice,CLayeredObject<CDevice> >::CreateLayeredChild+0x20
09 005ec800 600245fb d3d11!CD3D11LayeredChild<ID3D11DeviceChild,NDXGI::CDevice,64>::FinalConstruct+0x27
0a 005ec83c 6001d235 d3d11!NDXGI::CResource::FinalConstruct+0x2b
0b 005ec8c4 5fff2682 d3d11!NDXGI::CDevice::CreateLayeredChild+0x1d8
0c 005ec8e0 6002cd1b d3d11!CBridgeImpl<ID3D11LayeredDevice,ID3D11LayeredDevice,CLayeredObject<NDXGI::CDevice> >::CreateLayeredChild+0x22
0d 005ec900 6002cb8e d3d11!NOutermost::CDeviceChild::FinalConstruct+0x1e
0e 005ec92c 6002c975 d3d11!CUseCountedObject<NOutermost::CDeviceChild>::CUseCountedObject<NOutermost::CDeviceChild>+0x49
0f 005ec974 6002c24e d3d11!CUseCountedObject<NOutermost::CDeviceChild>::CreateInstance+0x66
10 005eca0c 600098c7 d3d11!NOutermost::CDevice::CreateLayeredChild+0xd0
11 005eca34 5fff8654 d3d11!CDevice::CreateAndRecreateLayeredChild<SD3D11LayeredTexture2DCreationArgs>+0x2b
12 005ecb6c 5fff844b d3d11!CDevice::CreateTexture2D_Worker+0x200
13 005ecb88 6676f3d6 d3d11!CDevice::CreateTexture2D+0x1f
14 005ecbac 66771207 d2d1!CD3DDeviceCommon::CreateTexture+0x26
15 005ecc14 6676d0f5 d2d1!CD3DSurface::Create+0xb5
16 005ecc84 6676cce2 d2d1!BitmapRealization::CreateSurface+0xd5
17 005eccac 666e08f0 d2d1!BitmapRealization::Initialize+0xa7
18 005ecce0 666dd224 d2d1!CreateRefObject<BitmapRealization,win_scope::no_addref_release<CD3DDeviceLevel1> *,D2D_SIZE_U,void const *,unsigned int,enum BitmapOptionsInternal::Enum,enum DXGI_FORMAT>+0x6d
19 005ecd24 666c9612 d2d1!CHwSurfaceRenderTarget::CreateCopiedBitmapFromMemory+0x42
1a 005ecd84 666bca8b d2d1!DrawingContext::CreateBitmap+0x14c
1b 005ecdec 57ec7ecb d2d1!D2DDeviceContextBase<ID2D1BitmapRenderTarget,ID2D1BitmapRenderTarget,ID2D1DeviceContext>::CreateBitmap+0xa9
1c (Inline) -------- xul!ID2D1RenderTarget::CreateBitmap+0x2d
1d 005ecee8 57ecabc2 xul!mozilla::gfx::CreatePartialBitmapForSurface+0x345 [z:\build\build\src\gfx\2d\helpersd2d.h @ 639]
1e 005ed010 57ecbae3 xul!mozilla::gfx::DrawTargetD2D1::GetImageForSurface+0x103 [z:\build\build\src\gfx\2d\drawtargetd2d1.cpp @ 1947]
1f 005ed224 57ef31b7 xul!mozilla::gfx::DrawTargetD2D1::MaskSurface+0x83 [z:\build\build\src\gfx\2d\drawtargetd2d1.cpp @ 384]
20 005ed2cc 57fd08f8 xul!mozilla::gfx::DrawTargetDual::MaskSurface+0x5d [z:\build\build\src\gfx\2d\drawtargetdual.cpp @ 125]
21 005ed344 58949fbb xul!gfxAlphaBoxBlur::Paint+0x135 [z:\build\build\src\gfx\thebes\gfxblur.cpp @ 214]
22 005ed3c0 588cd40d xul!nsContextBoxBlur::DoPaint+0x9c [z:\build\build\src\layout\painting\nscssrendering.cpp @ 4330]
23 005ed5d0 57a269ce xul!nsTextFrame::PaintOneShadow+0x4d1 [z:\build\build\src\layout\generic\nstextframe.cpp @ 6441]
24 (Inline) -------- xul!nsTextFrame::PaintShadows+0x4dad2e
25 005ed890 5754c54b xul!nsTextFrame::PaintText+0x4db09d [z:\build\build\src\layout\generic\nstextframe.cpp @ 7151]
26 005ed980 5754c318 xul!nsDisplayText::RenderToContext+0x1e8 [z:\build\build\src\layout\generic\nstextframe.cpp @ 5306]
27 005ed9a8 573e4573 xul!nsDisplayText::Paint+0x43 [z:\build\build\src\layout\generic\nstextframe.cpp @ 5183]
28 005eda18 57558295 xul!mozilla::FrameLayerBuilder::PaintItems+0x2d0 [z:\build\build\src\layout\painting\framelayerbuilder.cpp @ 6093]
29 005edb34 57557ac7 xul!mozilla::FrameLayerBuilder::DrawPaintedLayer+0x339 [z:\build\build\src\layout\painting\framelayerbuilder.cpp @ 6265]
2a 005edbd4 5755787d xul!mozilla::layers::ClientPaintedLayer::PaintThebes+0x13f [z:\build\build\src\gfx\layers\client\clientpaintedlayer.cpp @ 171]
2b 005edc1c 57557750 xul!mozilla::layers::ClientPaintedLayer::RenderLayerWithReadback+0x9f [z:\build\build\src\gfx\layers\client\clientpaintedlayer.cpp @ 273]
2c 005edc40 57413938 xul!mozilla::layers::ClientContainerLayer::RenderLayer+0x83 [z:\build\build\src\gfx\layers\client\clientcontainerlayer.h @ 59]
2d 005edde4 5754e38b xul!mozilla::layers::ClientLayerManager::EndTransactionInternal+0x118 [z:\build\build\src\gfx\layers\client\clientlayermanager.cpp @ 383]
2e 005ede24 57421629 xul!mozilla::layers::ClientLayerManager::EndTransaction+0x3e [z:\build\build\src\gfx\layers\client\clientlayermanager.cpp @ 436]
2f 005ee070 57569282 xul!nsDisplayList::PaintRoot+0x53d [z:\build\build\src\layout\painting\nsdisplaylist.cpp @ 2324]
30 005eed50 5754f051 xul!nsLayoutUtils::PaintFrame+0x4cd [z:\build\build\src\layout\base\nslayoututils.cpp @ 3738]
31 005eee14 575689c4 xul!mozilla::PresShell::Paint+0x1df [z:\build\build\src\layout\base\presshell.cpp @ 6459]
32 005eee5c 57568acd xul!nsViewManager::ProcessPendingUpdatesPaint+0xe2 [z:\build\build\src\view\nsviewmanager.cpp @ 480]
33 005eee90 575397f3 xul!nsViewManager::ProcessPendingUpdatesForView+0xf3 [z:\build\build\src\view\nsviewmanager.cpp @ 408]
34 005eeea8 5756a09b xul!nsViewManager::ProcessPendingUpdates+0x3b [z:\build\build\src\view\nsviewmanager.cpp @ 1094]
35 005ef110 577dd0d6 xul!nsRefreshDriver::Tick+0x4c4 [z:\build\build\src\layout\base\nsrefreshdriver.cpp @ 2052]
36 005ef160 577dd037 xul!nsRefreshDriver::DoTick+0x41 [z:\build\build\src\layout\base\nsrefreshdriver.cpp @ 1498]
37 005ef170 577dcfd9 xul!nsRefreshDriver::FinishedWaitingForTransaction+0x50 [z:\build\build\src\layout\base\nsrefreshdriver.cpp @ 2159]
38 005ef180 57550706 xul!nsRefreshDriver::NotifyTransactionCompleted+0x3e [z:\build\build\src\layout\base\nsrefreshdriver.cpp @ 2225]
39 005ef19c 586972a8 xul!mozilla::layers::ClientLayerManager::DidComposite+0x68 [z:\build\build\src\gfx\layers\client\clientlayermanager.cpp @ 531]
3a 005ef1b4 57a283aa xul!mozilla::dom::TabChild::DidComposite+0x2c [z:\build\build\src\dom\ipc\tabchild.cpp @ 2916]
3b 005ef1d4 57550e52 xul!mozilla::layers::CompositorBridgeChild::RecvDidComposite+0x4d77ff [z:\build\build\src\gfx\layers\ipc\compositorbridgechild.cpp @ 538]
3c 005ef288 5755123f xul!mozilla::layers::PCompositorBridgeChild::OnMessageReceived+0x15a [z:\build\build\src\obj-firefox\ipc\ipdl\pcompositorbridgechild.cpp @ 1495]
3d 005ef298 573c83a4 xul!mozilla::layers::PCompositorManagerChild::OnMessageReceived+0x2b [z:\build\build\src\obj-firefox\ipc\ipdl\pcompositormanagerchild.cpp @ 162]
3e 005ef2b4 573c74e8 xul!mozilla::ipc::MessageChannel::DispatchAsyncMessage+0x56 [z:\build\build\src\ipc\glue\messagechannel.cpp @ 2093]
3f 005ef30c 573c796a xul!mozilla::ipc::MessageChannel::DispatchMessageW+0xc1 [z:\build\build\src\ipc\glue\messagechannel.cpp @ 2021]
40 005ef324 573c78ee xul!mozilla::ipc::MessageChannel::RunMessage+0x6b [z:\build\build\src\ipc\glue\messagechannel.cpp @ 1888]
41 005ef33c 57ba941a xul!mozilla::ipc::MessageChannel::MessageTask::Run+0x37 [z:\build\build\src\ipc\glue\messagechannel.cpp @ 1921]
42 005ef384 5745ae54 xul!mozilla::SchedulerGroup::Runnable::Run+0x5a [z:\build\build\src\xpcom\threads\schedulergroup.cpp @ 388]
43 005ef3ec 57459bcf xul!nsThread::ProcessNextEvent+0x16b [z:\build\build\src\xpcom\threads\nsthread.cpp @ 1452]
44 (Inline) -------- xul!NS_ProcessNextEvent+0x14
45 005ef420 57d2cd20 xul!mozilla::ipc::MessagePump::Run+0x7a [z:\build\build\src\ipc\glue\messagepump.cpp @ 97]
46 005ef43c 57296ebb xul!mozilla::ipc::MessagePumpForChildProcess::Run+0x58 [z:\build\build\src\ipc\glue\messagepump.cpp @ 302]
47 (Inline) -------- xul!MessageLoop::RunInternal+0x8
48 005ef474 57296e79 xul!MessageLoop::RunHandler+0x20 [z:\build\build\src\ipc\chromium\src\base\message_loop.cc @ 320]
49 005ef494 57726128 xul!MessageLoop::Run+0x19 [z:\build\build\src\ipc\chromium\src\base\message_loop.cc @ 300]
4a 005ef4a4 57725eb7 xul!nsBaseAppShell::Run+0x34 [z:\build\build\src\widget\nsbaseappshell.cpp @ 158]
4b 005ef4b4 58f87cde xul!nsAppShell::Run+0x26 [z:\build\build\src\widget\windows\nsappshell.cpp @ 278]
4c 005ef4c8 57d2ccde xul!XRE_RunAppShell+0x26 [z:\build\build\src\toolkit\xre\nsembedfunctions.cpp @ 882]
4d 005ef4d8 57296ebb xul!mozilla::ipc::MessagePumpForChildProcess::Run+0x16 [z:\build\build\src\ipc\glue\messagepump.cpp @ 279]
4e (Inline) -------- xul!MessageLoop::RunInternal+0x8
4f 005ef510 57296e79 xul!MessageLoop::RunHandler+0x20 [z:\build\build\src\ipc\chromium\src\base\message_loop.cc @ 320]
50 005ef530 58f87b71 xul!MessageLoop::Run+0x19 [z:\build\build\src\ipc\chromium\src\base\message_loop.cc @ 300]
51 005ef654 58f8a743 xul!XRE_InitChildProcess+0x46c [z:\build\build\src\toolkit\xre\nsembedfunctions.cpp @ 703]
52 005ef660 0038ad42 xul!mozilla::BootstrapImpl::XRE_InitChildProcess+0x11 [z:\build\build\src\toolkit\xre\bootstrap.cpp @ 65]
53 005ef688 00387750 firefox!content_process_main+0x74 [z:\build\build\src\ipc\contentproc\plugin-container.cpp @ 65]
54 (Inline) -------- firefox!NS_internal_main+0x6079
55 005ef9f8 00386295 firefox!wmain+0x6200 [z:\build\build\src\toolkit\xre\nswindowswmain.cpp @ 115]
56 (Inline) -------- firefox!invoke_main+0x1d
57 005efa40 76a8336a firefox!__scrt_common_main_seh+0xf9 [f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl @ 253]
58 005efa4c 77ae9902 kernel32!BaseThreadInitThunk+0xe
59 005efa8c 77ae98d5 ntdll!__RtlUserThreadStart+0x70
5a 005efaa4 00000000 ntdll!_RtlUserThreadStart+0x1b
Looks to be gfx::DrawTargetD2D1::GetImageForSurface() could the underlying issue as with bug 1405110 I don't know...
:Bas does this give you better information about what's going on?
Flags: needinfo?(bas)
Comment 17•7 years ago
|
||
(In reply to Jean-Yves Avenard [:jya] from comment #16)
> The full backtrace
>
> 0:000> k =005ec3a8
> # ChildEBP RetAddr
> WARNING: Stack unwind information not available. Following frames may be
> wrong.
> 00 005ec3a8 0225c514 igd10umd32+0x2235b5
> 01 005ec3e0 6004981c 0x225c514
> 02 005ec3e8 600124b8 d3d11!CD3D11FormatHelper::GetParentFormat+0x19
> 03 005ec44c 600388fc
> d3d11!CResource<ID3D11Texture2D>::CResource<ID3D11Texture2D>+0xd2
> 04 005ec470 6000eafb
> d3d11!TCLSWrappers<CTexture2D>::CLSFinalConstructFn+0x44
> 05 005ec4b4 6000b132
> d3d11!CLayeredObjectWithCLS<CTexture2D>::FinalConstruct+0x71
> 06 005ec4e8 60002314
> d3d11!CLayeredObjectWithCLS<CTexture2D>::CreateInstance+0xa7
> 07 005ec7c0 5fff250f d3d11!CDevice::CreateLayeredChild+0x170
> 08 005ec7d8 60024599
> d3d11!CBridgeImpl<ID3D11LayeredDevice,ID3D11LayeredDevice,
> CLayeredObject<CDevice> >::CreateLayeredChild+0x20
> 09 005ec800 600245fb
> d3d11!CD3D11LayeredChild<ID3D11DeviceChild,NDXGI::CDevice,64>::
> FinalConstruct+0x27
> 0a 005ec83c 6001d235 d3d11!NDXGI::CResource::FinalConstruct+0x2b
> 0b 005ec8c4 5fff2682 d3d11!NDXGI::CDevice::CreateLayeredChild+0x1d8
> 0c 005ec8e0 6002cd1b
> d3d11!CBridgeImpl<ID3D11LayeredDevice,ID3D11LayeredDevice,
> CLayeredObject<NDXGI::CDevice> >::CreateLayeredChild+0x22
> 0d 005ec900 6002cb8e d3d11!NOutermost::CDeviceChild::FinalConstruct+0x1e
> 0e 005ec92c 6002c975
> d3d11!CUseCountedObject<NOutermost::CDeviceChild>::
> CUseCountedObject<NOutermost::CDeviceChild>+0x49
> 0f 005ec974 6002c24e
> d3d11!CUseCountedObject<NOutermost::CDeviceChild>::CreateInstance+0x66
> 10 005eca0c 600098c7 d3d11!NOutermost::CDevice::CreateLayeredChild+0xd0
> 11 005eca34 5fff8654
> d3d11!CDevice::
> CreateAndRecreateLayeredChild<SD3D11LayeredTexture2DCreationArgs>+0x2b
> 12 005ecb6c 5fff844b d3d11!CDevice::CreateTexture2D_Worker+0x200
> 13 005ecb88 6676f3d6 d3d11!CDevice::CreateTexture2D+0x1f
> 14 005ecbac 66771207 d2d1!CD3DDeviceCommon::CreateTexture+0x26
> 15 005ecc14 6676d0f5 d2d1!CD3DSurface::Create+0xb5
> 16 005ecc84 6676cce2 d2d1!BitmapRealization::CreateSurface+0xd5
> 17 005eccac 666e08f0 d2d1!BitmapRealization::Initialize+0xa7
> 18 005ecce0 666dd224
> d2d1!CreateRefObject<BitmapRealization,win_scope::
> no_addref_release<CD3DDeviceLevel1> *,D2D_SIZE_U,void const *,unsigned
> int,enum BitmapOptionsInternal::Enum,enum DXGI_FORMAT>+0x6d
> 19 005ecd24 666c9612
> d2d1!CHwSurfaceRenderTarget::CreateCopiedBitmapFromMemory+0x42
> 1a 005ecd84 666bca8b d2d1!DrawingContext::CreateBitmap+0x14c
> 1b 005ecdec 57ec7ecb
> d2d1!D2DDeviceContextBase<ID2D1BitmapRenderTarget,ID2D1BitmapRenderTarget,
> ID2D1DeviceContext>::CreateBitmap+0xa9
> 1c (Inline) -------- xul!ID2D1RenderTarget::CreateBitmap+0x2d
> 1d 005ecee8 57ecabc2
> xul!mozilla::gfx::CreatePartialBitmapForSurface+0x345
> [z:\build\build\src\gfx\2d\helpersd2d.h @ 639]
> 1e 005ed010 57ecbae3
> xul!mozilla::gfx::DrawTargetD2D1::GetImageForSurface+0x103
> [z:\build\build\src\gfx\2d\drawtargetd2d1.cpp @ 1947]
> 1f 005ed224 57ef31b7 xul!mozilla::gfx::DrawTargetD2D1::MaskSurface+0x83
> [z:\build\build\src\gfx\2d\drawtargetd2d1.cpp @ 384]
> 20 005ed2cc 57fd08f8 xul!mozilla::gfx::DrawTargetDual::MaskSurface+0x5d
> [z:\build\build\src\gfx\2d\drawtargetdual.cpp @ 125]
> 21 005ed344 58949fbb xul!gfxAlphaBoxBlur::Paint+0x135
> [z:\build\build\src\gfx\thebes\gfxblur.cpp @ 214]
> 22 005ed3c0 588cd40d xul!nsContextBoxBlur::DoPaint+0x9c
> [z:\build\build\src\layout\painting\nscssrendering.cpp @ 4330]
> 23 005ed5d0 57a269ce xul!nsTextFrame::PaintOneShadow+0x4d1
> [z:\build\build\src\layout\generic\nstextframe.cpp @ 6441]
> 24 (Inline) -------- xul!nsTextFrame::PaintShadows+0x4dad2e
> 25 005ed890 5754c54b xul!nsTextFrame::PaintText+0x4db09d
> [z:\build\build\src\layout\generic\nstextframe.cpp @ 7151]
> 26 005ed980 5754c318 xul!nsDisplayText::RenderToContext+0x1e8
> [z:\build\build\src\layout\generic\nstextframe.cpp @ 5306]
> 27 005ed9a8 573e4573 xul!nsDisplayText::Paint+0x43
> [z:\build\build\src\layout\generic\nstextframe.cpp @ 5183]
> 28 005eda18 57558295 xul!mozilla::FrameLayerBuilder::PaintItems+0x2d0
> [z:\build\build\src\layout\painting\framelayerbuilder.cpp @ 6093]
> 29 005edb34 57557ac7
> xul!mozilla::FrameLayerBuilder::DrawPaintedLayer+0x339
> [z:\build\build\src\layout\painting\framelayerbuilder.cpp @ 6265]
> 2a 005edbd4 5755787d
> xul!mozilla::layers::ClientPaintedLayer::PaintThebes+0x13f
> [z:\build\build\src\gfx\layers\client\clientpaintedlayer.cpp @ 171]
> 2b 005edc1c 57557750
> xul!mozilla::layers::ClientPaintedLayer::RenderLayerWithReadback+0x9f
> [z:\build\build\src\gfx\layers\client\clientpaintedlayer.cpp @ 273]
> 2c 005edc40 57413938
> xul!mozilla::layers::ClientContainerLayer::RenderLayer+0x83
> [z:\build\build\src\gfx\layers\client\clientcontainerlayer.h @ 59]
> 2d 005edde4 5754e38b
> xul!mozilla::layers::ClientLayerManager::EndTransactionInternal+0x118
> [z:\build\build\src\gfx\layers\client\clientlayermanager.cpp @ 383]
> 2e 005ede24 57421629
> xul!mozilla::layers::ClientLayerManager::EndTransaction+0x3e
> [z:\build\build\src\gfx\layers\client\clientlayermanager.cpp @ 436]
> 2f 005ee070 57569282 xul!nsDisplayList::PaintRoot+0x53d
> [z:\build\build\src\layout\painting\nsdisplaylist.cpp @ 2324]
> 30 005eed50 5754f051 xul!nsLayoutUtils::PaintFrame+0x4cd
> [z:\build\build\src\layout\base\nslayoututils.cpp @ 3738]
> 31 005eee14 575689c4 xul!mozilla::PresShell::Paint+0x1df
> [z:\build\build\src\layout\base\presshell.cpp @ 6459]
> 32 005eee5c 57568acd xul!nsViewManager::ProcessPendingUpdatesPaint+0xe2
> [z:\build\build\src\view\nsviewmanager.cpp @ 480]
> 33 005eee90 575397f3
> xul!nsViewManager::ProcessPendingUpdatesForView+0xf3
> [z:\build\build\src\view\nsviewmanager.cpp @ 408]
> 34 005eeea8 5756a09b xul!nsViewManager::ProcessPendingUpdates+0x3b
> [z:\build\build\src\view\nsviewmanager.cpp @ 1094]
> 35 005ef110 577dd0d6 xul!nsRefreshDriver::Tick+0x4c4
> [z:\build\build\src\layout\base\nsrefreshdriver.cpp @ 2052]
> 36 005ef160 577dd037 xul!nsRefreshDriver::DoTick+0x41
> [z:\build\build\src\layout\base\nsrefreshdriver.cpp @ 1498]
> 37 005ef170 577dcfd9
> xul!nsRefreshDriver::FinishedWaitingForTransaction+0x50
> [z:\build\build\src\layout\base\nsrefreshdriver.cpp @ 2159]
> 38 005ef180 57550706
> xul!nsRefreshDriver::NotifyTransactionCompleted+0x3e
> [z:\build\build\src\layout\base\nsrefreshdriver.cpp @ 2225]
> 39 005ef19c 586972a8
> xul!mozilla::layers::ClientLayerManager::DidComposite+0x68
> [z:\build\build\src\gfx\layers\client\clientlayermanager.cpp @ 531]
> 3a 005ef1b4 57a283aa xul!mozilla::dom::TabChild::DidComposite+0x2c
> [z:\build\build\src\dom\ipc\tabchild.cpp @ 2916]
> 3b 005ef1d4 57550e52
> xul!mozilla::layers::CompositorBridgeChild::RecvDidComposite+0x4d77ff
> [z:\build\build\src\gfx\layers\ipc\compositorbridgechild.cpp @ 538]
> 3c 005ef288 5755123f
> xul!mozilla::layers::PCompositorBridgeChild::OnMessageReceived+0x15a
> [z:\build\build\src\obj-firefox\ipc\ipdl\pcompositorbridgechild.cpp @ 1495]
> 3d 005ef298 573c83a4
> xul!mozilla::layers::PCompositorManagerChild::OnMessageReceived+0x2b
> [z:\build\build\src\obj-firefox\ipc\ipdl\pcompositormanagerchild.cpp @ 162]
> 3e 005ef2b4 573c74e8
> xul!mozilla::ipc::MessageChannel::DispatchAsyncMessage+0x56
> [z:\build\build\src\ipc\glue\messagechannel.cpp @ 2093]
> 3f 005ef30c 573c796a
> xul!mozilla::ipc::MessageChannel::DispatchMessageW+0xc1
> [z:\build\build\src\ipc\glue\messagechannel.cpp @ 2021]
> 40 005ef324 573c78ee xul!mozilla::ipc::MessageChannel::RunMessage+0x6b
> [z:\build\build\src\ipc\glue\messagechannel.cpp @ 1888]
> 41 005ef33c 57ba941a
> xul!mozilla::ipc::MessageChannel::MessageTask::Run+0x37
> [z:\build\build\src\ipc\glue\messagechannel.cpp @ 1921]
> 42 005ef384 5745ae54 xul!mozilla::SchedulerGroup::Runnable::Run+0x5a
> [z:\build\build\src\xpcom\threads\schedulergroup.cpp @ 388]
> 43 005ef3ec 57459bcf xul!nsThread::ProcessNextEvent+0x16b
> [z:\build\build\src\xpcom\threads\nsthread.cpp @ 1452]
> 44 (Inline) -------- xul!NS_ProcessNextEvent+0x14
> 45 005ef420 57d2cd20 xul!mozilla::ipc::MessagePump::Run+0x7a
> [z:\build\build\src\ipc\glue\messagepump.cpp @ 97]
> 46 005ef43c 57296ebb
> xul!mozilla::ipc::MessagePumpForChildProcess::Run+0x58
> [z:\build\build\src\ipc\glue\messagepump.cpp @ 302]
> 47 (Inline) -------- xul!MessageLoop::RunInternal+0x8
> 48 005ef474 57296e79 xul!MessageLoop::RunHandler+0x20
> [z:\build\build\src\ipc\chromium\src\base\message_loop.cc @ 320]
> 49 005ef494 57726128 xul!MessageLoop::Run+0x19
> [z:\build\build\src\ipc\chromium\src\base\message_loop.cc @ 300]
> 4a 005ef4a4 57725eb7 xul!nsBaseAppShell::Run+0x34
> [z:\build\build\src\widget\nsbaseappshell.cpp @ 158]
> 4b 005ef4b4 58f87cde xul!nsAppShell::Run+0x26
> [z:\build\build\src\widget\windows\nsappshell.cpp @ 278]
> 4c 005ef4c8 57d2ccde xul!XRE_RunAppShell+0x26
> [z:\build\build\src\toolkit\xre\nsembedfunctions.cpp @ 882]
> 4d 005ef4d8 57296ebb
> xul!mozilla::ipc::MessagePumpForChildProcess::Run+0x16
> [z:\build\build\src\ipc\glue\messagepump.cpp @ 279]
> 4e (Inline) -------- xul!MessageLoop::RunInternal+0x8
> 4f 005ef510 57296e79 xul!MessageLoop::RunHandler+0x20
> [z:\build\build\src\ipc\chromium\src\base\message_loop.cc @ 320]
> 50 005ef530 58f87b71 xul!MessageLoop::Run+0x19
> [z:\build\build\src\ipc\chromium\src\base\message_loop.cc @ 300]
> 51 005ef654 58f8a743 xul!XRE_InitChildProcess+0x46c
> [z:\build\build\src\toolkit\xre\nsembedfunctions.cpp @ 703]
> 52 005ef660 0038ad42
> xul!mozilla::BootstrapImpl::XRE_InitChildProcess+0x11
> [z:\build\build\src\toolkit\xre\bootstrap.cpp @ 65]
> 53 005ef688 00387750 firefox!content_process_main+0x74
> [z:\build\build\src\ipc\contentproc\plugin-container.cpp @ 65]
> 54 (Inline) -------- firefox!NS_internal_main+0x6079
> 55 005ef9f8 00386295 firefox!wmain+0x6200
> [z:\build\build\src\toolkit\xre\nswindowswmain.cpp @ 115]
> 56 (Inline) -------- firefox!invoke_main+0x1d
> 57 005efa40 76a8336a firefox!__scrt_common_main_seh+0xf9
> [f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl @ 253]
> 58 005efa4c 77ae9902 kernel32!BaseThreadInitThunk+0xe
> 59 005efa8c 77ae98d5 ntdll!__RtlUserThreadStart+0x70
> 5a 005efaa4 00000000 ntdll!_RtlUserThreadStart+0x1b
>
> Looks to be gfx::DrawTargetD2D1::GetImageForSurface() could the underlying
> issue as with bug 1405110 I don't know...
>
> :Bas does this give you better information about what's going on?
It's interesting, it seems that igd10umd32.dll@0x2235b5 is indeed the crash location of the initial crash report this bug was filed for and that I can see, for the backtrace you got though, was it also a read exception accessing 0x0 (or some location close to that) or something else? Considering the stack trace, I suspect this is 'something' in the drivers messing up, rather than the caller.. if you go up the stack, are the parameters passed to CreateBitmap valid? (I suspect so, or it'd be crashing in other stacks that I've seen before when we pass invalid parameters)
I wonder what GetParentFormat does and what it accesses, that must be trying to access a 0x0 object... especially since this is under D2D so it's harder to figure out what it's doing..
It doesn't look like it's a race condition, none of the other threads are using D3D (at least in the report in the initial bug post), there's also no indication of a device reset or anything else that indicates the device is getting confused. We should probably ask Intel if they've seen this before, and what that location this point in the code is trying to access as it's hard for me to guess, even with the full backtrace.
It could indeed be that this crash signature encompasses a lot of different issues.
Flags: needinfo?(bas)
Assignee | ||
Comment 18•7 years ago
|
||
Joe, any chances you could get someone at Intel to give us some feedback on what's going on here?
Thanks
Flags: needinfo?(joseph.k.olivas)
Comment 19•7 years ago
|
||
(In reply to Jean-Yves Avenard [:jya] from comment #18)
> Joe, any chances you could get someone at Intel to give us some feedback on
> what's going on here?
>
> Thanks
Jaen=Yves, I've reported the backtrace and the issue to the graphics team at Intel to take a look at it. Hope to have some insight soon.
Thanks
Comment 20•7 years ago
|
||
Jean-Yves*
Comment 21•7 years ago
|
||
Sounds like this is likely to be a driver issue from email discussions.
The crash rate is high, if there is anything we can do here, it is probably a dot release driver.
Reporter | ||
Comment 22•7 years ago
|
||
we will see if bug 1405110 that's going into 57.0b8 has a positive effect on these crashes.
Comment 23•7 years ago
|
||
Could I have the raw dump files to send in for analysis?
Comment 24•7 years ago
|
||
I have contacted several users to ask them if they are OK with us sharing dumps. I will send them to Ken directly if I receive a positive response.
Comment 25•7 years ago
|
||
Ken, I've sent you 7 crash dumps from 2 different users.
Reporter | ||
Comment 26•7 years ago
|
||
the issue seems fixed with bug 1409141 landing in 57.0b10.
Updated•7 years ago
|
Status: NEW → RESOLVED
Closed: 7 years ago
Flags: needinfo?(milan)
Resolution: --- → FIXED
Comment 27•7 years ago
|
||
We will likely include the fix in bug 1409141 in a 56.0.2 dot release.
Updated•7 years ago
|
Assignee: nobody → jyavenard
status-firefox-esr52:
--- → unaffected
Depends on: 1409141
See Also: 1409141 →
Target Milestone: --- → mozilla58
Updated•7 years ago
|
You need to log in
before you can comment on or make changes to this bug.
Description
•