1872416 - Crash in [@ gfxFontEntry::GetShmemCharacterMap]

Status: NEW

(Core :: Graphics: Text, defect, P3)

Description

[BugBot [:suhaib / :marco/ :calixte]]

Crash report: https://crash-stats.mozilla.org/report/index/776e4cc7-6e27-4727-b584-4be850231228

Reason: EXCEPTION_ACCESS_VIOLATION_READ

Top 10 frames of crashing thread:

0  xul.dll  gfxFontEntry::GetShmemCharacterMap const  gfx/thebes/gfxFontUtils.h:406
0  xul.dll  gfxFontEntry::HasCharacter  gfx/thebes/gfxFontEntry.h
0  xul.dll  gfxFont::HasCharacter const  gfx/thebes/gfxFont.h:1787
1  xul.dll  gfxFontGroup::GetEllipsisTextRun  gfx/thebes/gfxTextRun.cpp:2905
2  xul.dll  mozilla::css::GetEllipsisTextRun  layout/generic/TextOverflow.cpp:58
3  xul.dll  mozilla::css::TextOverflow::Marker::SetupString  layout/generic/TextOverflow.cpp:916
4  xul.dll  mozilla::css::TextOverflow::ExamineLineFrames  layout/generic/TextOverflow.cpp:575
4  xul.dll  mozilla::css::TextOverflow::ProcessLine  layout/generic/TextOverflow.cpp:699
5  xul.dll  DisplayLine  layout/generic/nsBlockFrame.cpp:7340
5  xul.dll  nsBlockFrame::BuildDisplayList  layout/generic/nsBlockFrame.cpp:7492

By querying Nightly crashes reported within the last 2 months, here are some insights about the signature:

• First crash report: 2023-12-08
• Process type: Content
• Is startup crash: No
• Has user comments: No
• Is null crash: No

Comment 1

[BugBot [:suhaib / :marco/ :calixte]]

The Bugbug bot thinks this bug should belong to the 'Core::Graphics: Text' component, and is moving the bug to that component. Please correct in case you think the bot is wrong.

Comment 2

[Timothy Nikkel (:tnikkel)]

7 total crashes since May. All of them this month. All of them on nightly. All of them on the cpu "GenuineIntel family 6 model 76 stepping 4". Seems like a problem affecting only one machine.

Comment 3

[Bob Hood [:bhood]]

Looping in Jonathan in case he has additional input.

Comment 4

[Jonathan Kew [:jfkthame]]

(In reply to Timothy Nikkel (:tnikkel) from comment #2)

7 total crashes since May. All of them this month. All of them on nightly. All of them on the cpu "GenuineIntel family 6 model 76 stepping 4". Seems like a problem affecting only one machine.

Currently I see two recent reports, both from 123.0 (so beta/release, not nightly) and from different machines (one Win10, one Win11, different CPU models). So it's not a single-machine issue, but it is still an extremely low crash rate.

Comment 5

[Timothy Nikkel (:tnikkel)]

All of the crashes that are on non-nightly have a non-zero bit flip value.

All of the crashes on nightly have the same cpu id and same version of Windows.

So it seems like bitflips and then one computer on nightly.