JDK-4807962 : JVM cores with SIG 11 in Function name=JVM_NewInstance
  • Type: Bug
  • Component: hotspot
  • Sub-Component: runtime
  • Affected Version: 1.3.1,1.3.1_06
  • Priority: P2
  • Status: Closed
  • Resolution: Cannot Reproduce
  • OS: solaris_8
  • CPU: sparc
  • Submitted: 2003-01-24
  • Updated: 2009-06-25
  • Resolved: 2003-10-07
Related Reports
Relates :  
Relates :  
Description
customer was running into bug 4780201 and escalation 543518

They were also seeing another problem that appeared similar but was 
in CMoveNode::Value instead of IfNode::Ideal

After applying the binary fix for bug 4780201 they are still seeing this secondary problem.

core file is in /net/cores.east/cores/
named core-20030123-01

the error log shows 

Unexpected Signal : 11 occurred at PC=0xfecb95f0
Function name=JVM_NewInstance
Library=/home/areiner/1.3.1_06/jre/lib/sparc/server/libjvm.so

Current Java thread:

Dynamic libraries:
0x10000 	/home/areiner/1.3.1_06/bin/../bin/sparc/native_threads/java
0xff360000 	/usr/lib/lwp/libthread.so.1
0xff390000 	/usr/lib/libdl.so.1
0xff280000 	/usr/lib/libc.so.1
0xff350000 	/usr/platform/SUNW,Sun-Fire-880/lib/libc_psr.so.1
0xfec00000 	/home/areiner/1.3.1_06/jre/lib/sparc/server/libjvm.so
0xff230000 	/usr/lib/libCrun.so.1
0xff200000 	/usr/lib/libsocket.so.1
0xfeb00000 	/usr/lib/libnsl.so.1
0xff1d0000 	/usr/lib/libm.so.1
0xff260000 	/usr/lib/libw.so.1
0xff1a0000 	/usr/lib/libmp.so.2
0xff180000 	/usr/lib/librt.so.1
0xff160000 	/usr/lib/libaio.so.1
0xfebd0000 	/home/areiner/1.3.1_06/jre/lib/sparc/native_threads/libhpi.so
0xfead0000 	/home/areiner/1.3.1_06/jre/lib/sparc/libverify.so
0xfea90000 	/home/areiner/1.3.1_06/jre/lib/sparc/libjava.so
0xfea50000 	/home/areiner/1.3.1_06/jre/lib/sparc/libzip.so
0xfe9b0000 	/usr/lib/locale/en_US.ISO8859-1/en_US.ISO8859-1.so.2
0xfb790000 	/home/areiner/1.3.1_06/jre/lib/sparc/libnet.so
0xfb5e0000 	/home/areiner/1.3.1_06/jre/lib/sparc/libioser12.so
0xfb5c0000 	/usr/lib/nss_files.so.1

Local Time = Thu Jan 23 09:35:51 2003
Elapsed Time = 3152
#
# HotSpot Virtual Machine Error : 11
# Error ID : 4F530E43505002BD 01
# Please report this error at
# http://java.sun.com/cgi-bin/bugreport.cgi
#
# Java VM: Java HotSpot(TM) Server VM (1.3.1-internal mixed mode)
#


and the stack trace is
t@9 (l@9) terminated by signal ABRT (Abort)
(/opt/SUNWspro/bin/../WS6U1/bin/sparcv9/dbx) where          
current thread: t@9
=>[1] _lwp_kill(0x0, 0x9, 0x0, 0xff33a004, 0xff386000, 0xff33e440), at 0xff31bee0
  [2] raise(0x6, 0x0, 0x0, 0xffffffff, 0xff33e3cc, 0x0), at 0xff2cb738
  [3] abort(0xff33a004, 0xd577e170, 0x0, 0x4, 0x0, 0xd577e191), at 0xff2b5aac
  [4] os::abort(0x1, 0xff0c6000, 0x1, 0xd577e1f0, 0x0, 0xfecb95f0), at 0xfefc68f4
  [5] os::handle_unexpected_exception(0xff33e2bc, 0xff13697c, 0xff0da184, 0xff06f94c, 0xff0c6000, 0xd577e210), at 0xfefc59ec
  [6] JVM_handle_solaris_signal(0x0, 0x202da0, 0xd577ecc0, 0xff0c6000, 0xb, 0xd577ef78), at 0xfee0aa60
  [7] __sighndlr(0xb, 0xd577ef78, 0xd577ecc0, 0xfee0aa78, 0x0, 0x0), at 0xff374c8c
  [8] call_user_handler(0xfebc0e00, 0x9, 0xff3878e0, 0xd577ecc0, 0xd577ef78, 0xb), at 0xff36fadc
  [9] sigacthandler(0xfebc0e00, 0xd577ef78, 0xd577ecc0, 0xff386000, 0xd577ef78, 0xb), at 0xff36fca8
  ---- called from signal handler with signal -21230080 (SIG-21230080) ------
  [10] Type::meet(0x0, 0xd0488, 0xd0488, 0xd04d8, 0x93ebb8, 0x93eb9c), at 0xfecb95f0
  [11] CMoveNode::Value(0x93eb9c, 0xd577f3d4, 0xd577f3d4, 0xff0c6000, 0xd577f3d4, 0x0), at 0xfedd4c2c
  [12] PhaseIterGVN::transform_old(0x0, 0xd577f3d4, 0x93eb9c, 0xd577f4f4, 0x5, 0xd577f3fc), at 0xfecc1428
  [13] PhaseIterGVN::optimize(0x60, 0x0, 0x3f05c8, 0xd577f3a0, 0xd577f390, 0x93e288), at 0xfed73628
  [14] PhaseIdealLoop::PhaseIdealLoop(0xd577f178, 0x3f05f4, 0x1, 0x2e3e80, 0x4000, 0x991928), at 0xfed945e0
  [15] Compile::Optimize(0xd577fa10, 0xd577f848, 0xd577fa10, 0xd577f86c, 0x0, 0xd577f66c), at 0xfedcca94
  [16] Compile::Compile(0x26afb0, 0xd577fa90, 0x58229c, 0xff1276d4, 0xd577fb30, 0xd577fb40), at 0xfedcb898
  [17] C2Compiler::compile_method(0x278b0, 0xd577fd88, 0x58229c, 0x5821b0, 0xffffffff, 0x1), at 0xfedc75e0
  [18] CompileBroker::invoke_compiler_on_method(0x5821b0, 0x58229c, 0xff0e8d0c, 0x0, 0x0, 0x180), at 0xfedc7be4
  [19] CompileBroker::compiler_thread_loop(0x28cb8, 0x202da0, 0xff0c6000, 0xd577ffa0, 0xff0c6000, 0xffffffff), at 0xfee80a98
  [20] JavaThread::run(0xd5580000, 0xff0d1e38, 0xff0c6000, 0x200000, 0x202da0, 0x200000), at 0xfee16334
  [21] _start(0xff0c6000, 0xfebc0e00, 0x0, 0x0, 0x0, 0x0), at 0xfee14064

Comments
EVALUATION ###@###.### 2003-02-07 --------------------------- The offending method was reported to be com/prosrm/forecaster/models/common/ExtendedMNHCModel::updateCompleteRows() in customer's forecaster.jar. 1. Ran CTW test with 1.3.1_08 build's gamma_g on forecaster.jar. But CTW skips compiling this class due to failure while loading the class. Running with -XX:-CompileTheWorldPreloadClasses gets around this problem. 2. But running CTW test on jar doesn't reproduce the problem, either with gamma_g or a fastdebug libjvm.so. 3. Later running CTW with -XX:+VerifyLoopOptimizations -XX:+TraceIterativeGVN options assert at 2 places -- a) assert(fail == 0, "verify loops failed") due to Mismatched control setting for a Conditional move node, CMovePNode. This assert happens in all the versions, 1.3.1 ,1.4.1 & latest 1.4.2 b) assert(add->Opcode() == Op_AddI, ""). This happens in 1.3.1, 1.4.1 but 1.4.2 sees a SIGBUS at the same place. Not sure if these asserts could be false asserts, as none of the problems manifest when CTW test runs without -XX:+VerifyLoopOptimizations. And asserts happen at places other than reported by customer. Logs from runs with 1.3.1 & 1.4.2 are attached. Looking in the stack trace, input value node for the conditional_move node is bad. ----------------------------------- Customer did not come back after a fastdebug build was provided with instrumented hotspot source to produce CMove node data & complete IR data for methods in the failing class. So we are closing this bug for lack for more data to work on. ###@###.### 2003-10-07
07-10-2003