1642566 - Crash in [@ mozilla::dom::ContentParent::MinTabSelect]

Status: RESOLVED DUPLICATE of bug 1641211

(Core :: DOM: Navigation, defect, P2)

Description

[Alex Chronopoulos [:achronop]]

This bug is for crash report bp-102d0e8e-0b3c-481d-8044-5bdac0200601.

Top 10 frames of crashing thread:

0 xul.dll static mozilla::dom::ContentParent::MinTabSelect dom/ipc/ContentParent.cpp:847
1 xul.dll static mozilla::dom::ContentParent::GetUsedBrowserProcess dom/ipc/ContentParent.cpp:908
2 xul.dll static mozilla::dom::ContentParent::GetNewOrUsedBrowserProcessInternal dom/ipc/ContentParent.cpp:990
3 xul.dll static mozilla::dom::ContentParent::GetNewOrUsedBrowserProcess dom/ipc/ContentParent.cpp:1087
4 xul.dll static mozilla::dom::ContentParent::CreateBrowser dom/ipc/ContentParent.cpp:1329
5 xul.dll nsFrameLoader::TryRemoteBrowserInternal dom/base/nsFrameLoader.cpp:2581
6 xul.dll nsFrameLoader::TryRemoteBrowser dom/base/nsFrameLoader.cpp:2644
7 xul.dll nsFrameLoader::ShowRemoteFrame dom/base/nsFrameLoader.cpp:1018
8 xul.dll nsFrameLoader::Show dom/base/nsFrameLoader.cpp:889
9 xul.dll nsSubDocumentFrame::ShowViewer layout/generic/nsSubDocumentFrame.cpp:192

Some hints: Bug 1582318 changed this assert from NS_ASSERT to MOZ_DIAGNOSTIC_ASSERT and made it active on release builds for Nightly and Dev. However, it has been silent for about 2 months, till 31 of May. Bug 1602757 landed one day before and modified the way that a context process is reported as dead.

Comment 1

[Alex Chronopoulos [:achronop]]

Opps I had not looked at the OSX crashes. Taking them into account the problem must be there before Bug 1602757. Feel free to redirect accordingly.

Comment 2

[Alex Chronopoulos [:achronop]]

The following is a complete different crash report but it has the same crash reason, oddly. It might be a symbol issue. I add it here so it does not fall through the cracks.

https://crash-stats.mozilla.org/report/index/bbb217a7-e4fe-4484-898b-aaafd0200531

Comment 3

[Julien Cristau [:jcristau]]

There's more signatures with that crash reason:
https://crash-stats.mozilla.org/search/?moz_crash_reason=%3DMOZ_DIAGNOSTIC_ASSERT(!p->IsDead())

Comment 4

[Chris Peterson [:cpeterson]]

Jesup, are these crashes fallout from your process preallocation changes (bug 1602757)? Most of the crashes do not have Fission enabled.

Comment 5

[Randell Jesup [:jesup] (needinfo me)]

These crashes appear to be ongoing (see the crash-stats graphs over time); it may have bumped up a little recently, but that appears to start on 5/27; the bug 1602757 code landed on 5/30 and I don't think were in nightly until 5/31. It's possible that this has slightly increased the occurrence of this crash

Comment 6

[Kris Maglione [:kmag]]

It's possible that this is related to bug 1641211

Comment 7

[Tim Spurway [:tspurway]]

S1 or S2 bugs need an assignee - could you find someone for this bug?

Comment 8

[Chris Peterson [:cpeterson]]

kmag has patches waiting for review in bug 1641211 that should help or fix this crash.

I'm marking this bug as a sec bug because bug 1641211 is a sec bug. We'll

Comment 9

[Andrew McCreight [:mccr8]]

It sounded like you wanted this to be a sec bug, so I'll hide it.

Comment 12

[Chris Peterson [:cpeterson]]

Clearing Fission Milestone for bugs resolved as duplicates. We don't need to track duplicates.