Open Bug 1917148 Opened 2 years ago Updated 2 years ago

Crash in [@ js::jit::JitFrameLayout::calleeToken]

Categories

(Core :: JavaScript Engine: JIT, defect, P5)

x86
All
defect

Tracking

()

Tracking Status
firefox132 --- affected

People

(Reporter: release-mgmt-account-bot, Unassigned)

References

(Blocks 2 open bugs)

Details

(Keywords: crash)

Crash Data

Crash report: https://crash-stats.mozilla.org/report/index/f09379e7-a452-48f5-92b7-cf6620240831

Reason: EXCEPTION_ACCESS_VIOLATION_READ

Top 10 frames of crashing thread:

0  xul.dll  js::jit::JitFrameLayout::calleeToken const  js/src/jit/JitFrames.h:252
0  xul.dll  js::jit::JSJitFrameIter::calleeToken const  js/src/jit/JSJitFrameIter.cpp:84
0  xul.dll  js::jit::JSJitFrameIter::script const  js/src/jit/JSJitFrameIter.cpp:120
0  xul.dll  js::JitFrameIter::isSelfHostedIgnoringInlining const  js/src/vm/FrameIter.cpp:117
0  xul.dll  JS::GetScriptedCallerActivationRealmFast  js/src/jsapi.cpp:4752
0  xul.dll  JS::GetScriptedCallerGlobal  js/src/jsapi.cpp:4783
0  xul.dll  mozilla::dom::GetIncumbentGlobal  dom/script/ScriptSettings.cpp:213
1  xul.dll  mozilla::dom::CallbackObject::FinishSlowJSInitIfMoreThanOneOwner  dom/bindings/CallbackObject.cpp:109
2  xul.dll  mozilla::dom::binding_detail::FastFunction::FinishSlowJSInitIfMoreThanOneOwner  dist/include/mozilla/dom/FunctionBinding.h:131
2  xul.dll  mozilla::dom::RootedCallback<mozilla::OwningNonNull<mozilla::dom::binding_detail::FastFunction> >::~RootedCallback  dom/bindings/CallbackObject.h:640

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

  • First crash report: 2024-07-06
  • Process type: Content
  • Is startup crash: No
  • Has user comments: No
  • Is null crash: Yes - 2 out of 5 crashes happened on null or near null memory address

Moving to core: JavaScript Engine: JIT to evaluate

Component: General → JavaScript Engine: JIT
Severity: -- → S3
Priority: -- → P5

Looking at the crash addresses, these looks like values that would be expected on the stack, and thus suggest a bad stack walking offset.
These are likely to happen on off-by-(few bytes offsets).

Given the low volume, I am going to assume this might be some memory corruption, especially given one of the user comment reporting many tabs crashes. No URL to investigate yet.

You need to log in before you can comment on or make changes to this bug.