JDK-4794360 : REGRESSION: HotSpot server core dumps with signal 11
  • Type: Bug
  • Component: hotspot
  • Sub-Component: compiler
  • Affected Version: 1.4.1
  • Priority: P3
  • Status: Resolved
  • Resolution: Fixed
  • OS: solaris_8
  • CPU: sparc
  • Submitted: 2002-12-16
  • Updated: 2003-03-19
  • Resolved: 2003-03-19
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.
Other
1.4.2 betaFixed
Description

Name: rmT116609			Date: 12/16/2002


FULL PRODUCT VERSION :
java version "1.4.1_01"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.4.1_01-b01)
Java HotSpot(TM) Client VM (build 1.4.1_01-b01, mixed mode)

FULL OPERATING SYSTEM VERSION : SunOS lxor 5.8
Generic_108528-17 sun4u sparc SUNW,Ultra-5_10

Additonal Operating Systems: Linux Redhat 6.2(Zoot)


A DESCRIPTION OF THE PROBLEM :
A while(true) loop used in a spawned java.lang.Thread causes
the HotSpot server VM to core dump with a signal 11 error.
Running with -client prevents the problem from occuring.

STEPS TO FOLLOW TO REPRODUCE THE PROBLEM :
1.  javac EntryAutoStart.java
2.  java -server EntryAutoStart

EXPECTED VERSUS ACTUAL BEHAVIOR :
The output when the program works (java -client):

Starting test
 test(): 1
 test(): 2
 test(): 3
 test(): 4
 test(): 5
 test(): 6
 test(): 7
 test(): 8
 test(): 9
 test(): 10

----------------------------------------------------------
The output when the JVM crashes (java -server):

 Unexpected Signal : 11 occurred at PC=0xFE0D4040
 Function=[Unknown. Nearest: JVM_ArrayCopy+0x4F5C]

Library=/home/jharris/java1.4.1_01/jre/lib/sparc/server/libjvm.so

 Current Java thread:

 Dynamic libraries:
 0x10000         ../java1.4.1_01/bin/java
 0xff350000      /lib/libthread.so.1
 0xff390000      /lib/libdl.so.1
 0xff200000      /lib/libc.so.1
 0xff330000
/usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1
 0xfe000000
/home/jharris/java1.4.1_01/jre/lib/sparc/server/libjvm.so
 0xff2d0000      /lib/libCrun.so.1
 0xff1d0000      /lib/libsocket.so.1
 0xff100000      /lib/libnsl.so.1
 0xff0d0000      /lib/libm.so.1
 0xff300000      /lib/libw.so.1
 0xff0b0000      /lib/libmp.so.2
 0xff070000
/home/jharris/java1.4.1_01/jre/lib/sparc/native_threads/libhpi.so
 0xff040000
/home/jharris/java1.4.1_01/jre/lib/sparc/libverify.so
 0xfe7c0000
/home/jharris/java1.4.1_01/jre/lib/sparc/libjava.so
 0xff020000
/home/jharris/java1.4.1_01/jre/lib/sparc/libzip.so
 0xfdf90000
/usr/lib/locale/en_US.ISO8859-1/en_US.ISO8859-1.so.2

 Local Time = Thu Nov 14 14:11:15 2002
 Elapsed Time = 1
 #
 # HotSpot Virtual Machine Error : 11
 # Error ID : 4F530E43505002E6 01
 # Please report this error at
 # http://java.sun.com/cgi-bin/bugreport.cgi
 #
 # Java VM: Java HotSpot(TM) Server VM (1.4.1_01-b01 mixed
mode)
 #
 # An error report file has been saved as hs_err_pid422.log.
 # Please refer to the file for further information.
 #
 Abort (core dumped)


REPRODUCIBILITY :
This bug can be reproduced always.

---------- BEGIN SOURCE ----------
/*
 * This example crashes the Solaris Sparc JVM for java 1.4.0 and 1.4.1_01
 * with a core dump.  It works on 1.3.1_06 with -client and -server options.
 * The J2SE_Solaris_8_Recommended patches were all installed.
 * The crash happens with the java -server command line option and can be
avoided
 * by using the -client command line argument or by changing the while(true)
code
 * in the inc() method to be while(!done) and set done to false as shown in the
comments
 * in the code.
 *
 * run with java -server EntryAutoStart to show error
 * run with java -client EntryAutoStart to show working program
 *
 * The output when the program works (-client mode):
 *
 * Starting test
 * test(): 1
 * test(): 2
 * test(): 3
 * test(): 4
 * test(): 5
 * test(): 6
 * test(): 7
 * test(): 8
 * test(): 9
 * test(): 10
 *
 * The output when the JVM crashes (-server mode):
 *
 * Unexpected Signal : 11 occurred at PC=0xFE0D4040
 * Function=[Unknown. Nearest: JVM_ArrayCopy+0x4F5C]
 * Library=/home/jharris/java1.4.1_01/jre/lib/sparc/server/libjvm.so
 *
 * Current Java thread:
 *
 * Dynamic libraries:
 * 0x10000         ../java1.4.1_01/bin/java
 * 0xff350000      /lib/libthread.so.1
 * 0xff390000      /lib/libdl.so.1
 * 0xff200000      /lib/libc.so.1
 * 0xff330000      /usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1
 * 0xfe000000      /home/jharris/java1.4.1_01/jre/lib/sparc/server/libjvm.so
 * 0xff2d0000      /lib/libCrun.so.1
 * 0xff1d0000      /lib/libsocket.so.1
 * 0xff100000      /lib/libnsl.so.1
 * 0xff0d0000      /lib/libm.so.1
 * 0xff300000      /lib/libw.so.1
 * 0xff0b0000      /lib/libmp.so.2
 * 0xff070000
/home/jharris/java1.4.1_01/jre/lib/sparc/native_threads/libhpi.so
 * 0xff040000      /home/jharris/java1.4.1_01/jre/lib/sparc/libverify.so
 * 0xfe7c0000      /home/jharris/java1.4.1_01/jre/lib/sparc/libjava.so
 * 0xff020000      /home/jharris/java1.4.1_01/jre/lib/sparc/libzip.so
 * 0xfdf90000      /usr/lib/locale/en_US.ISO8859-1/en_US.ISO8859-1.so.2
 *
 * Local Time = Thu Nov 14 14:11:15 2002
 * Elapsed Time = 1
 * #
 * # HotSpot Virtual Machine Error : 11
 * # Error ID : 4F530E43505002E6 01
 * # Please report this error at
 * # http://java.sun.com/cgi-bin/bugreport.cgi
 * #
 * # Java VM: Java HotSpot(TM) Server VM (1.4.1_01-b01 mixed mode)
 * #
 * # An error report file has been saved as hs_err_pid422.log.
 * # Please refer to the file for further information.
 * #
 * Abort (core dumped)
 */

public class EntryAutoStart
{
    private static volatile int reg = 0;
    private static volatile boolean go = false;


    public EntryAutoStart()
    {
      RunThread rt = new RunThread(this);
      rt.setDaemon(true);
      rt.start();
    }

    public int test()
    {
      go = true;

      while(go)
      {}

      return(reg);
    }

    public void inc()
    {
      // Swapping in the following two lines prevents the JVM
      // from crashing in -server mode.
      //
      // boolean done = false;
      // while(!done)

      while(true)
      {
        while(!go)
        {}

        reg++;

        go = false;
      }
    }


    public static void main(String[] args)
    {
       EntryAutoStart eas = new EntryAutoStart();

       System.out.println("Starting test");

       for(int i=0; i<10; i++)
         System.out.println("test(): " + eas.test());
    }
}

class RunThread extends Thread
{
   private EntryAutoStart eas;

   public RunThread(EntryAutoStart eas)
   {
      super();

      this.eas = eas;
   }

   public void run()
   {
      eas.inc();
   }
}
---------- END SOURCE ----------

CUSTOMER WORKAROUND :
1. Use the java -client command line option.

OR

2. Don't use while(true) coding style, replace with
    boolean done = false;
    while(!done)

Release Regression From : 1.3.1_06
The above release value was the last known release where this 
bug was known to work. Since then there has been a regression.

(Review ID: 167071) 
======================================================================

Comments
CONVERTED DATA BugTraq+ Release Management Values COMMIT TO FIX: mantis-beta FIXED IN: mantis-beta INTEGRATED IN: mantis-beta tiger tiger-b03
14-06-2004

EVALUATION ###@###.### 2002-12-16 Setup test case in Bugs/4794360 and reproduced with current build. variable 'l' is nil in PhaseIdealLoop::build_loop_tree_impl(), around line "1609 while( is_postvisited(l->_head) ) {" ----- ----- Problem is a combination of code in ciTypeFlow which causes compiler to turn a well structured loop nest into a series of nested irreducible loops, along with some unsafe assumptions in build_loop_tree about incoming CFG shape. ###@###.### 2003-03-05
05-03-2003