ADDITIONAL SYSTEM INFORMATION :
Alpine Linux v3.20, Intel(R) Xeon(R) Platinum 8375C CPU @ 2.90GHz, 64 cores, 490G RAM. Running in containerized environment (Humio application). JVM configured with G1GC, 376GB heap, compressed class pointers, NUMA enabled.


A DESCRIPTION OF THE PROBLEM :
JVM crashes with SIGSEGV during G1 garbage collection in G1ParScanThreadState::steal_and_trim_queue method. This appears to be the same issue as reported in JDK-8351900 but occurring on Java 24 (Zulu build 24.0.1+9) instead of Java 21. The crash occurs in a GC worker thread during evacuation phase.

Key similarities to JDK-8351900:

Same problematic frame: G1ParScanThreadState::steal_and_trim_queue
Same signal: SIGSEGV with SEGV_MAPERR
Same GC phase: G1 evacuation
Similar stack trace pattern


STEPS TO FOLLOW TO REPRODUCE THE PROBLEM :
Cannot reproduce reliably. Crash occurred during normal operation of Humio application after 17+ hours of runtime during G1 garbage collection.

EXPECTED VERSUS ACTUAL BEHAVIOR :
EXPECTED -
G1 garbage collection should complete successfully without JVM crash.

ACTUAL -
JVM crashes with SIGSEGV (signal 11) at address 0x00000000000e69f9 in G1ParScanThreadState::steal_and_trim_queue method. Signal code SEGV_MAPERR indicates attempt to access unmapped memory.



---------- BEGIN SOURCE ----------
No reproducible test case available. Issue occurs during production workload.


---------- END SOURCE ----------