1561058 - Crash in [@ mozilla::ipc::WriteIPDLParam<T>]

Status: RESOLVED FIXED

(Core :: IPC, defect, P3)

Description

[Neha Kochar [:neha]]

This bug is for crash report bp-4fb422eb-61e3-425f-917c-2bb7f0190624.

Top 10 frames of crashing thread:

0 XUL void mozilla::ipc::WriteIPDLParam<mozilla::dom::BrowsingContext*&> ipc/glue/IPDLParamTraits.h:53
1 XUL mozilla::dom::PContentParent::SendWindowPostMessage ipc/ipdl/PContentParent.cpp:4986
2 XUL mozilla::dom::ContentParent::RecvWindowPostMessage dom/ipc/ContentParent.cpp:5985
3 XUL mozilla::dom::PContentParent::OnMessageReceived ipc/ipdl/PContentParent.cpp:10867
4 XUL mozilla::ipc::MessageChannel::DispatchMessage ipc/glue/MessageChannel.cpp:2158
5 XUL mozilla::ipc::MessageChannel::MessageTask::Run ipc/glue/MessageChannel.cpp:1970
6 XUL nsThread::ProcessNextEvent xpcom/threads/nsThread.cpp:1225
7 XUL NS_ProcessPendingEvents xpcom/threads/nsThreadUtils.cpp:434
8 XUL nsBaseAppShell::NativeEventCallback widget/nsBaseAppShell.cpp:87
9 XUL nsAppShell::ProcessGeckoEvents widget/cocoa/nsAppShell.mm:440

I got this crash while trying to reproduce Bug 1560106 but the backtrace is not the same.
With fission enabled, I browsed to https://9gag.com/ (as mentioned in that bug) and scrolled resulting in this crash.

Comment 1

[Nika Layzell [:nika] (ni? for response)]

Attachment: Bug 1561058 - Null check ContentParent within ContentParent::RecvPostMessage,

Comment 2

[Andrew McCreight [:mccr8]]

That patch is basically the same thing as I have in bug 1560313, though I don't have the log, but I do add some null checks in a few other places. So maybe this is just a dupe of that other bug?

Comment 3

[Nika Layzell [:nika] (ni? for response)]

(In reply to Andrew McCreight [:mccr8] from comment #2)

That patch is basically the same thing as I have in bug 1560313, though I don't have the log, but I do add some null checks in a few other places. So maybe this is just a dupe of that other bug?

This one happens on the other side of the IPC channel, the check which I'm doing is in the parent. They're probably related, however.

Comment 4

[Andrew McCreight [:mccr8]]

(In reply to :Nika Layzell (busy - slow to reply) from comment #3)

This one happens on the other side of the IPC channel, the check which I'm doing is in the parent. They're probably related, however.

Ah, I missed that. Sorry.

Comment 5

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/72051a14f93e
Null check ContentParent within ContentParent::RecvPostMessage, r=peterv

Comment 6

[Razvan Maries]

https://hg.mozilla.org/mozilla-central/rev/72051a14f93e