Uploaded image for project: 'JDK'
  1. JDK
  2. JDK-4101641

Java caused an invalid page fault in kernel32.dll

XMLWordPrintable

    • x86
    • windows_95



      We have encountered a problem with a graphics application employing both
      the 1.1.3 and 1.1.4 JDK running on Windows 95. We're not sure of the exact
      cause, but at random points in the application, we are informed that "Java
      caused an invalid page fault in kernel32.dll".

      Here is the dump data:

         JAVA caused an invalid page fault in
         module KERNEL32.DLL at 0137:bff9a07c.
         Registers:
         EAX=02ecf920 CS=0137 EIP=bff9a07c EFLGS=00000246
         EBX=00000000 SS=013f ESP=02ecf91c EBP=02ecf970
         ECX=007af230 DS=013f ESI=00000000 FS=300f
         EDX=000000cc ES=013f EDI=007af26c GS=0000
         Bytes at CS:EIP:
         5e 8b e5 5d c2 10 00 64 a1 00 00 00 00 55 8b ec
         Stack dump:
         007af26c c0000005 00000000 00000000 bff9a07c
         00000000 02ecf974 bff9d1fd 00000001 00000000
         00000026 00000016 0000000b 00000064 bff9ac4c
         02ecf964

      Here is some other information about the platform (although we have
      reproduced the problem on several Pentium computers employing Windows95):
         Windows 95 Version 4.00.950
         Pentium 64.0MB RAM
         Java JDK 1.1.3

      Apparently, we're not the only people who have had this problem. I have
      seen other instances of this problem reported in the newsgroups. All seem
      to point to a Windows95 platform. We have been unable to reproduce the
      problem on a Windows NT platform.



      The above problem occurs even when using JDK 1.1.5.




         Ravi,

      Here's the piece of code we are getting the "Java caused an invalid page
      fault error in
      kernel32.dll" in. We finally were able to reproduce the failure by calling
      the following method
      a few times. We're basically adding and removing the same bean to another
      base frame bean
      on demand and the following method is the one that gets called at
      "removing" time.

      The precise line where the page fault occurs is the following:
           container.remove((Component) beanInfo.m_BedrockBeanIfc);

      Let me know if this information is helpful for you to reproduce the error.

      Thanks for your time,

      Jose' Vega
      ###@###.###
      512.491.2625

      -------------------------------------------------------- Sample Code
      Included --------------------------------------------------------

      /* Method causing Page Fault */
       public synchronized void destroyBean(String beanName)
          { // begin destroyBean()
              try
              {
                  BeanInfo beanInfo = (BeanInfo)
      m_hashtableBeans.remove(beanName);
                  if ( beanInfo.m_strBeanAttachedTo != null )
                  {
                      if ( beanInfo.m_strBeanAttachedTo.length() > 0 )
                      {
                          BeanInfo beanInfoAttached =
      (BeanInfo)m_hashtableBeans.get(beanInfo.m_strBeanAttachedTo);
                          Container container = (Container)
      beanInfoAttached.m_BedrockBeanIfc;

      -------------->>>>>>> container.remove((Component)
      beanInfo.m_BedrockBeanIfc);
      -------------->>>> IT NEVER RETURNS FROM THIS CALL AND SOMETIMES IT EVEN
      LOCKS
      -------------->>>> THE MACHINE.
                      }
                  }
                  beanInfo.m_BedrockBeanIfc.destroyBean();
              }
              catch( NullPointerException e)
              {
                  e.printStackTrace();

              }
          }; // end destroyBean()


            dmendenhsunw David Mendenhall (Inactive)
            duke J. Duke
            Votes:
            0 Vote for this issue
            Watchers:
            0 Start watching this issue

              Created:
              Updated:
              Resolved:
              Imported:
              Indexed: