JDK-8256371 : Compiler thread crashes in ClassLoaderData::is_alive()
  • Type: Bug
  • Component: hotspot
  • Sub-Component: compiler
  • Affected Version: 15.0.2,16
  • Priority: P2
  • Status: Closed
  • Resolution: External
  • OS: linux
  • CPU: x86_64
  • Submitted: 2020-11-15
  • Updated: 2020-12-14
  • Resolved: 2020-12-14
The Version table provides details related to the release that this issue/RFE will be addressed.

Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.

To download the current JDK release, click here.
JDK 16
16Resolved
Related Reports
Relates :  
JDK-8246487 - misc tests SIGSEGV in ClassLoaderData::is_alive()
Relates :  
JDK-8256572 - assert(n->_left->_parent == n) failed: sanity in share/memory/metaspace/blockTree.cpp:113
Relates :  
JDK-8255087 - Compiler thread crashes in ClassLoaderData::is_alive()
Description
Various tests are failing with SIGSEGV in the Compiler thread.

For the first sighting:

compiler/codegen/TestLongDoubleVect.java

#  SIGSEGV (0xb) at pc=0x00007fa529de1d30, pid=19074, tid=19089
#
# JRE version: Java(TM) SE Runtime Environment (16.0+25) (fastdebug build 16-ea+25-1585)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (fastdebug 16-ea+25-1585, compiled mode, sharing, g1 gc, linux-amd64)
# Problematic frame:
# V  [libjvm.so+0x978d30]  ClassLoaderData::is_alive() const+0x0

Here's the crashing thread's stack:

---------------  T H R E A D  ---------------

Current thread (0x00007fa524318cd0):  JavaThread "C2 CompilerThread0" daemon [_thread_in_vm, id=19089, stack(0x00007fa48ddfe000,0x00007fa48deff000)]

Stack: [0x00007fa48ddfe000,0x00007fa48deff000],  sp=0x00007fa48defdbe8,  free space=1022k
Native frames: (J=compiled Java code, A=aot compiled Java code, j=interpreted, Vv=VM code, C=native code)
V  [libjvm.so+0x978d30]  ClassLoaderData::is_alive() const+0x0
V  [libjvm.so+0xa0652b]  CompileQueue::get()+0x10b
V  [libjvm.so+0xa0a0b5]  CompileBroker::compiler_thread_loop()+0x2f5
V  [libjvm.so+0x1886786]  JavaThread::thread_main_inner()+0x256
V  [libjvm.so+0x188d440]  Thread::call_run()+0x100
V  [libjvm.so+0x156da56]  thread_native_entry(Thread*)+0x116


siginfo: si_signo: 11 (SIGSEGV), si_code: 1 (SEGV_MAPERR), si_addr: 0x000000000000002c
Comments
I am closing this bug as "External", since it has the exact same symptom as the previous occurrence (JDK-8255087)
30-11-2020
ILW = Same as JDK-8255087 = P2
17-11-2020
Another failure mode observed in the same job set with a different test: compiler/gcbarriers/EqvUncastStepOverBarrier.java # Internal Error (/opt/mach5/mesos/work_dir/slaves/47535081-0322-4d83-bb78-f7e11abb86d0-S45751/frameworks/1735e8a2-a1db-478c-8104-60c8b0af87dd-0196/executors/ecf1d32d-2b7e-43df-affc-285adeb9dfed/runs/84b95878-4026-4533-95b8-a43568742224/workspace/open/src/hotspot/share/ci/ciInstanceKlass.cpp:54), pid=19559, tid=19576 # assert(get_instanceKlass()->is_loaded()) failed: must be at least loaded # # JRE version: Java(TM) SE Runtime Environment (16.0+25) (fastdebug build 16-ea+25-1585) # Java VM: Java HotSpot(TM) 64-Bit Server VM (fastdebug 16-ea+25-1585, compiled mode, sharing, g1 gc, linux-amd64) # Problematic frame: # V [libjvm.so+0x8c3636] ciInstanceKlass::ciInstanceKlass(Klass*)+0x6b6 Here's the crashing thread's stack: --------------- T H R E A D --------------- Current thread (0x00007f3ce0317f60): JavaThread "C2 CompilerThread0" daemon [_thread_in_vm, id=19576, stack(0x00007f3c495fe000,0x00007f3c496ff000)] Current CompileTask: C2: 1093 236 b jdk.internal.module.Builder::requires (10 bytes) Stack: [0x00007f3c495fe000,0x00007f3c496ff000], sp=0x00007f3c496fd300, free space=1020k Native frames: (J=compiled Java code, A=aot compiled Java code, j=interpreted, Vv=VM code, C=native code) V [libjvm.so+0x8c3636] ciInstanceKlass::ciInstanceKlass(Klass*)+0x6b6 V [libjvm.so+0x90276d] ciObjectFactory::create_new_metadata(Metadata*)+0x60d V [libjvm.so+0x902e6d] ciObjectFactory::get_metadata(Metadata*)+0x1cd V [libjvm.so+0x8aabd7] ciEnv::get_klass_by_name_impl(ciKlass*, constantPoolHandle const&, ciSymbol*, bool)+0x407 V [libjvm.so+0x8aaedc] ciEnv::get_klass_by_name_impl(ciKlass*, constantPoolHandle const&, ciSymbol*, bool)+0x70c V [libjvm.so+0x91c47b] ciSignature::ciSignature(ciKlass*, constantPoolHandle const&, ciSymbol*)+0x31b V [libjvm.so+0x8dff84] ciMethod::ciMethod(methodHandle const&, ciInstanceKlass*)+0x664 V [libjvm.so+0x9025aa] ciObjectFactory::create_new_metadata(Metadata*)+0x44a V [libjvm.so+0x902e6d] ciObjectFactory::get_metadata(Metadata*)+0x1cd V [libjvm.so+0x8a9a36] ciEnv::get_method_from_handle(Method*)+0xb6 V [libjvm.so+0xa08e86] CompileBroker::invoke_compiler_on_method(CompileTask*)+0x476 V [libjvm.so+0xa0a368] CompileBroker::compiler_thread_loop()+0x5a8 V [libjvm.so+0x1886786] JavaThread::thread_main_inner()+0x256 V [libjvm.so+0x188d440] Thread::call_run()+0x100 V [libjvm.so+0x156da56] thread_native_entry(Thread*)+0x116
15-11-2020
Other sightings of the same issue: JDK-8255087 Compiler thread crashes in ClassLoaderData::is_alive() JDK-8246487 misc tests SIGSEGV in ClassLoaderData::is_alive()
15-11-2020