1598680 - Crash in [@ mozilla::layers::ImageBridgeParent::RecvUpdateAsyncPluginSurface]

Status: RESOLVED FIXED

(Core Graveyard :: Plug-ins, defect, P1)

Description

[Marcia Knous [:marcia]]

This bug is for crash report bp-a5c37af9-73f6-43ed-9058-f93520191122.

Seen while looking at Windows nightly crashes - crashes started in 20191121214422: https://bit.ly/2XDhhT2

Code was touched in Bug 1577336. ni on :handyman

Top 10 frames of crashing thread:

0 xul.dll mozilla::layers::ImageBridgeParent::RecvUpdateAsyncPluginSurface gfx/layers/ipc/ImageBridgeParent.cpp:525
1 xul.dll mozilla::layers::PImageBridgeParent::OnMessageReceived ipc/ipdl/PImageBridgeParent.cpp:587
2 xul.dll void mozilla::ipc::MessageChannel::DispatchMessage ipc/glue/MessageChannel.cpp:2130
3 xul.dll mozilla::ipc::MessageChannel::MessageTask::Run ipc/glue/MessageChannel.cpp:2003
4 xul.dll MessageLoop::DoWork ipc/chromium/src/base/message_loop.cc:523
5 xul.dll void base::MessagePumpForUI::DoRunLoop ipc/chromium/src/base/message_pump_win.cc:203
6 xul.dll base::MessagePumpWin::Run ipc/chromium/src/base/message_pump_win.h:79
7 xul.dll void MessageLoop::RunHandler ipc/chromium/src/base/message_loop.cc:308
8 xul.dll MessageLoop::Run ipc/chromium/src/base/message_loop.cc:290
9 xul.dll base::Thread::ThreadMain ipc/chromium/src/base/thread.cc:192

Comment 1

[David Parks [:handyman]]

Looks like it just needs a null check.

Comment 2

[David Parks [:handyman]]

Attachment: Bug 1598680: Check that D3D device exists in async plugin rendering

If the compositor has lost the D3D device, we cannot complete async plugin surface creation/blitting.

Comment 3

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/32b03ac0cbcf
Check that D3D device exists in async plugin rendering r=haik

Comment 4

[Natalia Csoregi [:nataliaCs]]

https://hg.mozilla.org/mozilla-central/rev/32b03ac0cbcf

Comment 5

[Toshihiro Yamada]

I get a crash with same signature, but crash line is different.

Crash report
https://crash-stats.mozilla.org/report/index/a6ed3efd-c65c-409d-8047-069200191126

Seemingly, Null check is needed there, too.

Comment 6

[Marcia Knous [:marcia]]

Hi David - Should we file a new bug for Comment 5?

Comment 7

[David Parks [:handyman]]

Sure, thanks. I'm going to investigate why this is happening but it could just be an occasional failure in getting the draw target that needs a null check.

Comment 8

[Itiel]

Same crash, on the latest Nightly:

Comment 9

[Itiel]

Same crash, on the latest Nightly:
https://crash-stats.mozilla.org/report/index/b2733b71-4fd7-489a-b8d8-d909d0191128

Comment 10

[Toshihiro Yamada]

(In reply to David Parks (dparks) [:handyman] from comment #7)

Sure, thanks. I'm going to investigate why this is happening but it could just be an occasional failure in getting the draw target that needs a null check.

Yes, crash happened occasionally(I encountered only 2 times) when I moved to a tab containing Flash contents.

Additional info:
This and fixed crash have other signatures related to Bug 1600032.
And after crash have happened, crash that has signature only IPC related(Also Bug 1600032) repeated when I opened Flash content.

Comment 11

[Ryan VanderMeulen [:RyanVM]]

This was backed out from Beta along with bug 1577336 to take care of the crashes reported in the other bugs.
https://hg.mozilla.org/releases/mozilla-beta/rev/3af5f7e1bd23