1992149 - [armeabi-v7a] startup Crash in [@ split_config.armeabi_v7a.apk] (JNA libjnidispatch.so)

Status: REOPENED

(GeckoView :: General, defect, P1)

Description

[Sebastian Hengst [:aryx] (needinfo me if it's about an intermittent or backout)]

Frequent startup crash with armeabi-v7a architecture, many more reports for 145.0a1 Nightly than 143.0.x

[Tracking Requested - why for this release]:

Crash report: https://crash-stats.mozilla.org/report/index/9b36997c-c4f6-4577-86bc-521a10251002

Reason:

SIGSEGV / SEGV_MAPERR

Top 10 frames:

0  split_config.armeabi_v7a.apk  split_config.armeabi_v7a.apk@0x13040
1  split_config.armeabi_v7a.apk  split_config.armeabi_v7a.apk@0x11c76
2  libsqlite.so  libsqlite.so@0xccee3
3  split_config.armeabi_v7a.apk  split_config.armeabi_v7a.apk@0x1a55a
4  libmegazord.so  <core::iter::adapters::GenericShunt<I, R> as core::iter::traits::iterator::It...  /builds/worker/.rustup/toolchains/1.87.0-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/core/src/iter/adapters/mod.rs:175
5  ?  @0xaa2bfb0a
6  split_config.armeabi_v7a.apk  split_config.armeabi_v7a.apk@0x122a6
7  split_config.armeabi_v7a.apk  split_config.armeabi_v7a.apk@0x63d2
8  split_config.armeabi_v7a.apk  split_config.armeabi_v7a.apk@0x1a4fa
9  split_config.armeabi_v7a.apk  split_config.armeabi_v7a.apk@0x5e32

Comment 1

[Ted Campbell [:tcampbell]]

Doesn't look like a huge nubmer of crashes yet, but I'll take a look

Comment 2

[Ted Campbell [:tcampbell]]

60% of the devices are Samsung with the most common being a tablet model from 2016. I looked into a few dumps and they often ended up executing nonsense instructions. The stacks are a bit working in crashstats, but when I look at raw dumps, they are consistently in libmegazord.

Stacks often include:
uniffi_search_checksum_method_searchengineselector_clear_search_config

I'm wondering if this is related to JNA crashes we recently fixed..

Comment 3

[Ted Campbell [:tcampbell]]

The split_config.armeabi_v7a library reference actually seems to be the libjnidispatch.so from JNA, but doesn't seem to save the name in memory map correctly.

Comment 4

[Ted Campbell [:tcampbell]]

The bulk of crashes seem to go away with the JNA update.

Comment 5

[Ted Campbell [:tcampbell]]

There are also a small number of crashes that exist after the JNA fix such as https://crash-stats.mozilla.org/report/index/1c900cb1-11bc-499a-8410-9c9160250930
This group of crashes has <core::iter::adapters::GenericShunt<I, R> as core::iter::traits::iterator::Iterator>::next in the stack (also built into the megazord)

Comment 6

[Ryan VanderMeulen [:RyanVM]]

Given that this is an ongoing issue with crash volume on Release as well, let's reopen this to track the remaining issue separate from the JNA problems we were having with Nightly.

Comment 7

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

The bug is marked as tracked for firefox145 (nightly). We have limited time to fix this, the soft freeze is in 3 days. However, the bug still isn't assigned and has low severity.

:owlish, could you please find an assignee and increase the severity for this tracked bug? If you disagree with the tracking decision, please talk with the release managers.

For more information, please visit BugBot documentation.

Comment 8

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

The bug is linked to a topcrash signature, which matches the following criterion:

• Top 10 AArch64 and ARM crashes on beta (startup)

:owlish, could you consider increasing the severity of this top-crash bug?

For more information, please visit BugBot documentation.