HP-UX Java™ SDK Version 1.3.1.08 Release Notes


» Back to HP-UX Java™ Archived Releases – Release Notes

These Release Notes are provided in the software and as a standalone file on the website http://h20392.www2.hp.com/portal/swdepot/displayProductInfo.do?productNumber=HPUXJAVAHOME. The website has the most recent information.

The HP-UX SDK, for the Java 2 Platform Version 1.3 release provides the tools for developing and deploying Java™ applications on HP-UX 11.0 and 11i PA-RISC and HP-UX 11.20 and 11.22 Itanium-based workstations and servers.

table of contents

features

installation

usage documentation

» supported tools and options
» additional HotSpot option information
» HP specific options and features
» large Java heap sizes

» using WDB to examine backtraces in Java thread stacks
» Asian TrueType fonts and Asian locales
» date/time methods - new defaults
» profiling capability added
» Using JNI - main/primordial thread stack size limits
» Using JNI - non-main/primordial thread stack size limits
» using JNI - dereferencing null pointers
» using JPDA
» manually launching the application VM
» Closing a socket when accept or read is pending (PA-RISC)
» C and C++ libraries
» CLASSPATH
» JAR files
» compatibility between release 1.3 and previous releases
» migrating from Java 1.1 to Java 2
» web sites with more information

problem fixes and known issues

features

SDK version 1.3.1.08 (combined PA-RISC and Itanium)

The 1.3.1.08 version of the HP-UX SDK, for the Java 2 Platform, HotSpot 1.3.1 Edition provides a slight performance improvement on IPF systems and fixes defects, which are described in "Problem Fixes and Known Issues" section at the end of this document.

The 1.3.1.08 version of the HP-UX SDK includes Sun Microsystems' release 1.3.1_05 and 1.3.1_06.

On PA-RISC, large heap sizes are now supported on HP-UX 11.0 as well as 11i, with the installation of patches. See the "Large Java Heap Sizes" section in these release notes for details.

This release continues to support a 100% Java compatible environment.

version numbering

  • In order to create a very high degree of compatibility between your PA and IA platforms, we have merged the sources for the product under one release model and now have a unified release numbering scheme. Beginning with this SDK version 1.3.1.06, source code for both PA and Itanium Processor Family (IPF) systems is included.
  • Starting with HotSpot 1.3.1, HP uses the same version numbering scheme for HotSpot as for the SDK, following JavaSoft's convention.
  • HotSpot version 1.3.1.x server JVM

    This SDK 1.3.1 release includes the HotSpot 1.3.1 server JVM (the default) and the Classic VM. See the note above on version numbering.

    The HotSpot 1.3.1 Server JVM for HP-UX 11.0 and 11i is suitable for both client and server workloads. We invoke the Server VM with configuration options that suit client-side applications. All the -X options that were in HotSpot 1.0.1 are included in HotSpot 1.3.1.

    The HP-UX SDK, for the Java 2 Platform, HotSpot 1.3.1 Edition includes all of the standard Java 2 SDK Tools. The tools include appletviewer (with a graphical user interface), extcheck, jar, java, javac, javadoc, javah, javap, jdb and oldjdb (only for -classic mode), rmic, rmid, rmiregistry, serialver, keytool, jarsigner, policytool, native2ascii, tnameserv, and idlj.

    previous 1.3 releases

    The HotSpot 1.3.1 Server JVM provided a number of improvements over the previous HotSpot 1.01 and Classic runtime environments:

  • Improved performance
  • Full implementation of Java VM Debugging Interface (JVMDI)
  • Support for Java VM Profiling Interface (JVMPI)
  • Support for HP's debugger WDB 3.0.01 for Java stack unwind
  • Support for the -Xeprof option for HPjmeter
  • A new option, -Xrs, which reduces the use of operating system signals by the Java Virtual Machine
  • A new option, -pa that launches the PA driver on an IPF system
  • Large heap size (HP-UX 11i) PA-RISC
  • Full implementation of JVMDI and JPDA allows you to run HotSpot instead of Classic when using the 1.3 SDK with numerous tools such as WebGain Visual Cafe, TogetherSoft Control Center, Code Warrior, Oracle JDeveloper, IBM Visual Age, Netbeans, JDE, MetaMata, Tek. Tools, Elixier IDE, BlueJ, JIG, JSwat Project, Swing Debugger, and the sample debugger jdb that ships with SDK 1.3.

    Support for JVMPI means that it is possible to profile java code with the HotSpot 1.3.1 VM. Therefore, you can extract more accurate runtime profiles. Some of the tools based on JVMPI include: hprof (-Xrunhprof), JProbe, JUM, and OptimizeIt. (Note: Although -hprof provides some stack trace information not found in -Xeprof, the latter can sometimes be more useful for performance tuning.) For more information on JVMDI and JPDA, go to:
    http://java.sun.com/products/jpda/faq.html.
  • You can now use HP's debugger WDB 3.0.01 to examine backtraces containing mixed language frames (Java and C/C++) in Java thread stacks. This will simplify debugging the VM and Java mixed-language applications. For details, see "Using WDB to Debug the JVM" in these release notes.
  • The -Xrs option is fully documented at:
    http://java.sun.com/j2se/1.3/docs/tooldocs/solaris/java.html
  • Large heap size (HP-UX 11i PA-RISC). Please see the section "Large Heap Sizes" in these release notes for further information.
  • For more detailed information on previous 1.3 releases, see the release notes for each release on HP-UX Java™ Archived Releases Downloads and Documentation.

    installation

    patches

    Operating system patches should be installed before you install the software. To determine which patches have been installed on your machine, login as root and check your machine with: /usr/sbin/swlist -l product

    For the most up-to-date list of required and recommended patches, and instructions on where to obtain them, go http://h20392.www2.hp.com/portal/swdepot/displayProductInfo.do?productNumber=HPUXJAVAHOME, and click on the "Patches" in the left hand navigation bar. Please install any dependency patches as well. These will be listed on the IT Resource Center web page from where you download the patch.

    minimum system recommendation

    HP-UX PA-RISC systems: For best performance on HP-UX PA-RISC systems: HP server series rp5400, rp7400, rp8400, and Superdome, and PA-based workstations, running HP-UX 11.0, 11.11 (11i v1), or 11.23 (11i v2). The minimum system for running Java applications is a PA-RISC 2.0 system.

    HP-UX Itanium-based systems: hp rx4610, rx5670, and rx9610 servers and the hp i2000 workstation.

    installation instructions

    If you download the software from the website, you need approximately 60MB of disk space to download the .depot file. To install the software from the .depot file, you need an additional 100MB disk space. After installing the software, you can remove the .depot file. The HP-UX SDK, for the Java 2 Platform installs under /opt/java1.3. As root user, use the SD-UX swinstall command to install the software:

    /usr/sbin/swinstall

    It will lead you through the installation. Change Source Depot Type to "Local Directory" and Source Depot Path to /tmp/<filename> . (If you used a directory other than /tmp in the previous step, replace /tmp with that directory name.) We recommend you select the "Reinstall filesets" and unselect the "Mount filesystems" option from the options menu.

    WARNING: Do not unarchive rt.jar, il8n.jar, jpda.jar, and tools.jar. These files are needed by the SDK tools and the runtime environment.

    Add the directory /opt/java1.3/bin to your PATH.

    Note: To install the software into an alternate location, see
    "Installing into an alternate location".

    For information on setting important system parameters required for correct execution of Java programs refer to the Programmer's Guide which can be found at Programmer's Guide for Java™ 2.

    Installing into an alternate location

    To install the software into an alternate location, add @/<alternatedirectory> to the end of the swinstall line, and add the alternate directory to your PATH.

    For example:

    /usr/sbin/swinstall -s <download> \* @/<alternatedirectory>

    Java prepends <alternatedirectory> to the default product path. Java will therefore install in /<alternatedirectory>/opt/java1.3.

    If you want your Java home directory in <alternatedirectory> without the /opt/java1.3 directories, you need to install into a temporary directory, and then move the directories to where you want them.

    For example:

    swinstall -s <download> \* @/<temporarydirectory>
    mv <temporarydirectory>/opt/java*/* <finalalternatedirectory>

    Or you can install and link to the directories.

    For example:

    swinstall -s <download> \* @/<actualdirectory>
    ln -s <finalalternatedirectory> /<actualdirectory>/opt/java*

    You will notice that the two files, <alternatedirectory>/etc and <alternatedirectory>/var are created. These may be purged, because they do not apply to a product installed in an alternate location.

    Note that we are testing another method of installing into an alternate location using <tag>. When we are confident that this method works with software depot install, we will post instructions here.

    file structure

    The diagram below displays an abbreviated form of the file structure:
      java1.3
         |
      ___|_____________________________________________________
       |        |        |         |           |          |    
      bin      lib      jre      src.jar      demo     include 
       |        |        |
      java    tools.jar  |
      javac   dt.jar     |
      javadoc ir.idl     |
      javah   orb.idl    |
      javap              | 
      jdb                |
                      ___|__________
                         |          |
                        bin        lib
                         |          |
       __________________|______     __|_____________________________________
       |       |       |      |        |         |     |         |         |
    java PA_RISC PA_RISC2  IA64     rt.jar security PA_RISC PA_RISC2.0   IA64
           ___|___ ___|_______|__   il8n.jar      _____|__  __|_____     ___|_____ 
              |       |       |                    |     |   |   |        |     |    
          native   native  native              classic HS* classic HS* classic HS*
          threads  threads threads                                         
              |       |       |
            java    java    java
    					H*  = hotspot
    

    The tools install under opt/java1.3/bin and the libraries install under opt/java1.3/lib. The tools.jar file contains the classes for supporting the tools and utilities. The file dt.jar contains the DesignTime archive of BeanInfo files. The jre directory includes the runtime environment. The file rt.jar contains the runtime classes for the core API. The file il8n.jar contains the internationalization and localization classes and files. The security directory contains security management files. The PA_RISC and PA_RISC2 directories contain the shared libraries used by the HP-UX platform. The file src.jar contains an archive of source files for the core API for informational purposes. To view the files, enter the command

    $ jar xvf src.jar.

    The include directory contains the header files for supporting JNI and JVMDI.

    usage documentation

    The HP-UX Programmer's Guide for Java 2 provides additional information.

    The Java 2 Platform, Standard Edition, v 1.3 API Specification is provided at http://java.sun.com/j2se/1.3/docs/api/index.html. Note: Only the java.x packages are supported.

    Below are some additional documentation notes.

    supported tools and options

    To run a tool on HP-UX, either use the full path name or add the path to the startup file. For example, for javac, on the command line you could enter: /opt/java1.3/bin/javac yourfile.java. You could alternatively add /opt/java1.3/bin to your PATH statement and on the command line enter: javac yourfile.java

    The HotSpot technology accepts all of the standard tools and options. The standard tools and the standard documentation references are noted below:

    The appletviewer tool is documented at:
    http://www.java.sun.com/products/jdk/1.3/docs/tooldocs/solaris/appletviewer.html

    CLASSPATH is documented at:
    http://www.java.sun.com/products/jdk/1.3/docs/tooldocs/solaris/classpath.html

    The jar tool is documented at:
    http://www.java.sun.com/products/jdk/1.3/docs/tooldocs/solaris/jar.html

    The java tool is documented at:
    http://www.java.sun.com/products/jdk/1.3/docs/tooldocs/solaris/java.html

    The javac tool is documented at:
    http://www.java.sun.com/products/jdk/1.3/docs/tooldocs/solaris/javac.html

    The javadoc tool is documented at:
    http://www.java.sun.com/products/jdk/1.3/docs/tooldocs/solaris/javadoc.html

    The javah tool is documented at:
    http://www.java.sun.com/products/jdk/1.3/docs/tooldocs/solaris/javah.html

    The javap tool is documented at:
    http://java.sun.com/reference/docs/index.html

    The jdb (JPDA) jdb tool is documented at:
    http://java.sun.com/docs/index.html

    The old jdb (old jdb for classic technology only) tool is documented at:
    http://java.sun.com/j2se/1.3/docs/tooldocs/win32/oldjdb.html

    The extcheck tool is documented at:
    http://java.sun.com/reference/docs/index.html

    The rmic tool is documented at:
    http://java.sun.com/reference/docs/index.html

    The rmiregistry tool is documented at:
    http://java.sun.com/reference/docs/index.html

    The rmid tool is documented at:
    http://java.sun.com/reference/docs/index.html

    The serialver tool is documented at:
    http://java.sun.com/reference/docs/index.html

    The native2ascii tool is documented at:
    http://java.sun.com/reference/docs/index.html

    The keytool tool is documented at:
    http://java.sun.com/reference/docs/index.html

    The jarsigner tool is documented at:
    http://java.sun.com/reference/docs/index.html

    The policytool tool is documented at:
    http://java.sun.com/reference/docs/index.html

    The tnameserv tool is documented at:
    http://java.sun.com/reference/docs/index.html

    The idlj tool is documented at:
    http://java.sun.com/reference/docs/index.html

    additional HotSpot option information

    The HotSpot technology accepts all of the standard options as well as the following partial list of non-standard -X options. Non-standard options are subject to change in future releases.

    -pa

    Launches the PA driver and uses the PA2.0 libraries while executing on an
    IPF system.

    -X

    Prints out a brief usage message describing the non-standard options.

    -Xbatch

    Disables background compilation. If compilation of a large method is taking a long time, the performance engine will revert to interpreting the method. It will compile the method as a background task, running the method in interpreter mode until the background compilation is finished. The -Xbatch flag disables background compilation so that compilation of all methods proceeds as a foreground task until completed, regardless of how long the compilation takes. This flag is provided for users who desire more deterministic behavior of method compilation for purposes such as benchmarking.

    -Xbootclasspath:<bootclasspath>

    Specify a colon-separated list of directories, JAR archives, and ZIP archives to search for boot class files. The specified boot class libraries will be used instead of the boot class files in the jre/lib/rt.jar archive normally used by the Java 2 software.

    -Xincgc

    Enable the incremental garbage collector. The incremental garbage collector, which is off by default, will eliminate occasional garbage-collection pauses during program execution. However, it can lead to a roughly 10% decrease in overall performance.

    -Xint

    Operate in interpreted-only mode. Compilation to native code is disabled, and all bytecodes are executed by the interpreter. The performance benefits offered by the Java HotSpot adaptive compiler will not be present in this mode.

    -Xmn

    Set the Java new generation heap size (for example: -Xmn64m). The "new generation" is the first generation in HotSpot's generational garbage collector. This option replaces the option -XX:NewSize=N.

    -Xms<size>

    Specify the initial size, in bytes, of the memory allocation pool. This value must be a multiple of 1024 greater than 1MB. Append the letter k or K to indicate kilobytes, or m or M to indicate megabytes. The default value is 5248KB. Examples:

    -Xms4194304
    -Xms4096k
    -Xms4m

    -Xmx

    Specify the maximum size, in bytes, of the memory allocation pool. This value must a multiple of 1024 greater than 2MB. Append the letter k or K to indicate kilobytes, or m or M to indicate megabytes. The default value is 64MB. Examples:

    -Xmx83886080
    -Xmx81920k
    -Xmx80m

    -Xnoclassgc

    Disables class garbage collection.

    -Xoptgc

    The optimistic garbage collection flag. Improves garbage collection performance of applications with mostly short-lived objects. A server-side application that creates many short-lived objects for each transaction is likely to benefit greatly with Xoptgc. However this flag should be used with caution. It is not recommended for applications that build up objects quickly during the run time that are not short-lived.

    -Xprof
    (excerpt below from http://java.sun.com/products/hotspot/1.0/README.html)

    Profiles the running program, and sends profiling data to standard output. This option is provided as a utility that is useful in program development and is not intended to be be used in production systems.

    -Xrs (HotSpot 1.3.1 and later)

    Reduces use of operating-system signals by the Java virtual machine (JVM), allowing for orderly shutdown of a Java application. Allows user cleanup code (such as closing database connections) to run at shutdown, even if the JVM terminates abruptly. For more information, see Javasoft's documentation at http://java.sun.com/j2se/1.3/docs/tooldocs/solaris/java.html.

    -Xss<size>

    Specifies the size of stack for each new Java thread. The default Java thread stack size is 512 KB. This flag is appropriate for programs that have small thread stack size requirements and/or create several thousand threads, potentially running out of virtual memory. Program threads that overflow the allocated stack will receive java.lang.StackOverFlowException. Defaults:

    -Xss512k (Java 1.3 and 1.4 32-bit mode)
    -X1m (Java 1.4 64-bit mode)

    -XX:+AggressiveHeap

    This option instructs the JVM to push memory use to the limit: the overall heap is around 3850MB, the memory management policy defers collection as long as possible, and (beginning with J2SE 1.3.1.05) some GC activity is done in parallel. Because this option sets heap size, do not use the -Xms or -Xmx options in conjunction with -XX:+AggressiveHeap. Doing so will cause the options to override each other's settings for heap size.

    Because the -XX:+AggressiveHeap option has specific system requirements for correct operation and may require privileged access to system configuration parameters, it should be used with caution. We have found it to be useful for certain applications that create a lot of short lived objects.

    -XX:+AllowUserSignalHandlers

    Instructs the JVM not to complain if the application installs signal handlers.

    -XX:+DisableExplicitGC

    Disable calls to System.gc(). The JVM still performs garbage collection when necessary.

    -XX:MaxNewSize=<size> (not supported on 1.3)

    Sets the maximum size of new generation (in bytes). Note that in HotSpot 1.0.1 this option took an integer that specified a value in Kbytes. Starting with HotSpot 1.3.1, the integer argument specifies bytes. The arguments can now be followed by either 'k' or 'm' to specify Kbytes or Mbytes.

    -XX:NewSize=<size>

    Sets the default size of new generation (in bytes). Note that in HotSpot 1.0.1 this option took an integer that specified a value in Kbytes. Starting with HotSpot 1.3.1, the integer argument specifies bytes. The arguments can now be followed by either 'k' or 'm' to specify Kbytes or Mbytes.

    -XX:NewSizeThreadIncrease=<sizeInKb>

    Sets the additional size added to desired new generation size per non-daemon thread (in bytes). Note that in HotSpot 1.0.1 this option took an integer that specified a value in Kbytes. Starting with HotSpot 1.3.1, the integer argument specifies bytes. The arguments can now be followed by either 'k' or 'm' to specify Kbytes or Mbytes.

    -XX:MaxPermSize

    Sets the maximum size of permanent generation (in bytes). Note that in HotSpot 1.0.1 this option took an integer that specified a value in Kbytes. Starting with HotSpot 1.3.1, the integer argument specifies bytes. The arguments can now be followed by either 'k' or 'm' to specify Kbytes or Mbytes.

    -XX:+ServerApp

    A set of XX options which, when bundled together, make some applications run faster. For each release, the options as well as the values may be different depending upon the default values of XX options. We recommend that you test to see whether this set enhances the performance of your application before you use the option in production.

    -XX:SurvivorRatio=<size>

    Ratio of eden/survivor space size. Default for SDK 1.3 is 64MB. Number can include 'm' or 'M' for megabytes, 'k' or 'K' for kilobytes, and 'g' or 'G' for gigabytes. For example, 32k is the same as 32768.

    -XX:+UseCompilerSafepoints (PA-RISC 1.3.1, 1.4 and later, Itanium 1.4.2 and later)

    Enables compiler safe points. In this release, compiler safe points is off by default. Enabling compiler safepoints guarantees a more deterministic delay to stop all running java threads before doing a safepoint operation, namely garbage collection and deoptimization. In this release, a patch is required. See "Known Problems" in these release notes.

    -XX:+UseGetTimeOfDay (HotSpot 1.3.1 and later)

    Instructs the JVM to use the GetTimeOfDay call instead of using the new lightweight mechanism where the number of CPU ticks since the application started is used to calculate the current time. See "Date/Time Methods - New Defaults" in these release notes for more information.

    -XX:+UseOnStackReplacement (PA-RISC 1.3.1, 1.4 and later, Itanium 1.4.2 and later)

    Enables on stack replacement. In this release, on stack replacement is off by default. On stack replacement enables the interpreter to go into compiled code while it is executing the same instance of the method call. If you enable on stack replacement, you should also enable compiler safe points (see the previous option). For this release, a patch is required. See "Known Problems" in these release notes.

    -XX:+UseSIGUSR2 (PA-RISC only)

    Use the java command line option -XX:+UseSIGUSR2 if you want the JVM to use SIGUSR2 for internal operations like Thread.interrupt() calls instead of SIGUSR1, the default. This allows you to better implement third party middleware applications that in some versions want to use SIGUSR1 for similar purposes in their native code.

    HP specific options and features

    Additional HP specific documentation is provided in the HP-UX Programmer's Guide.

    Noteworthy HP specific options and features for Java 2 version 1.3 include the following:

  • -pa
  • -pa11 (PA-RISC only)
  • -Xeprof
  • -Xnocatch
  • -Xoptgc
  • -Xprep
  • -verbosegc
  • -Xverbosegc
  • and the FastSwing feature.
  • These are described below.

    -pa option

    If you have downloaded both the IPF and PA SDK on an IPF system, you can run the PA SDK by specifying this option. Normally, Java detects that you are executing on an IPF system and launches the IPF driver. To launch the PA driver and use the PA2.0 libraries, use this option.

    -pa11 option (PA-RISC only)

    HP's PA-RISC 2.0 architecture offers performance features not compatible with previous architectures. PA1.1 binaries can be run on a PA1.1 as well as PA2.0 based systems; however, a PA2.0 binary can only run on a PA2.0 based system. Starting with the 1.2.2 release of the SDK, HP includes two versions of the shared libraries comprising APIs and VMs. The PA2.0 shared libraries will be default if the user is running on a PA2.0 system. The user can override the use of the PA2.0 version of the shared libraries on a PA2.0 by specifying the -pa11 flag. For example:

    On a PA2.0 based system, invoking Java by typing

    java -version

    results in something similar to:

    java version "1.3.1"
    HotSpot VM (..., mixed mode, PA2.0 build 1.3.1.00-00/04/23-PA_RISC 2.0)

    The generated version string indicates that the PA2.0 version of the VM will be used.

    You can override the use of the PA2.0 version of the VMs and APIs on a PA2.0 system by adding the -pa11 flag as follows:

    java -pa11 -version

    This results in something similar to:

    java version "1.3.1"
    HotSpot VM (..., mixed mode, PA1.1 build 1.3.1.00-00/-4/23-PA_RISC 1.1)

    The version string indicates that the PA1.1 version of the VM in spite of the fact that we are running on a PA2.0 system.

    Note: If you run HotSpot with the -pa11 flag or run on a PA 1.1 system, your heap address space will be restricted to 1G.

    -Xeprof option

    The -Xeprof option generates profile data for HPjmeter. The -Xeprof option enables profiling of Java applications running on HotSpot version 1.2.2.05 or greater and collects method clock and CPU times, method call count, and call graph. (For more information on HPjmeter, see http://h20392.www2.hp.com/portal/swdepot/displayProductInfo.do?productNumber=HPUXJAVAHOME.)

    To profile your application use the following command:

    java -Xeprof:<options> ApplicationClassName

    To profile your applet, use:

    appletviewer -J-Xeprof:<options> URL>

    where <options> is a list of <key>[=<value> ] arguments separated by commas.

    After the profiled applet or application normally terminates execution, the Java Virtual Machine writes the profile data to a file in the current directory.

    We have found the following options useful in most cases:

    For CPU time metrics with minimal intrusion:

    -Xeprof or
    -Xeprof:ie=no

    Exact call count information and object creation profiling:

    -Xint -Xeprof:ie=no

    To see the complete list of available options, use

    java -Xeprof:help

    Here are the supported -Xeprof options:

    file=<filename>

    The profile data will be written to the named file. The default is java.eprof.

    times=quick|thorough

    Collect call graph with inclusive method clock and CPU times and method call count.

    This option uses tracing with reduction and collects the data separately for each thread, throughout the whole execution time of the program.

    The quick value instructs the profiler to use the hardware Interval Timer register for time measurement. This value results in faster profiling runs, but in very rare circumstances can produce inaccurate data. This is the default for PA-RISC 2.0 based machines. If you ever suspect that the profile data generated using the quick value is incorrect then redo the run to see whether the results can be replicated.

    The thorough value is the opposite of quick, disabling the use of the Interval Timer. The profiling runs will be longer, but will provide timing data with the same quality as the system calls used to measure the time. However, the profiling intrusion and overhead also increase. This is the default for PA-RISC 1.1 based machines.

    ie=yes|no

    Enable/disable the profiling intrusion estimation.

    ie=yes, the default value, specifies that the profiler estimate the profiling intrusion and write the estimated values to the profile data file. A future version of HPjmeter will be able to present the timing data as: raw, meaning the data as collected, or compensated, meaning with the estimated intrusion subtracted.

    Disabling intrusion estimation reduces the size of the data files, but will also disable the intrusion compensation feature.

    inlining=disable|enable

    Disable/enable inlining in the HotSpot VM.

    The compiler in the HotSpot VM optimizes Java applications by inlining frequently called methods. Execution of inlined methods does not count as "calls" from the profiler's viewpoint. Instead, the time spent in an inlined method is attributed to its caller.

    The count of created objects cannot be reliably estimated in the presence of inlining, because the calls to the constructors may have been inlined.

    To obtain an accurate method call count and to enable the created objects metric, run the VM with inlining=disable.

    -Xnocatch

    The -Xnocatch option disables the Java "catch-all" signal handler. Use this option to generate clean stack traces from native code.

    -Xoptgc

    The Xoptgc, or 'optimistic garbage collection' flag, improves garbage collection performance of applications with mostly short-lived objects. A server-side application that creates many short-lived objects for each transaction is likely to benefit greatly with Xoptgc. However this flag should be used with caution. It is not recommended for applications that build up objects quickly during the run time that are not short-lived.

    -Xprep

    The -Xprep option is used to dynamically preprocess (modify) bytecodes of the classes loaded by the VM. Its syntax is:

    -Xprep <factory_class_name>:<arguments>

    where <factory_class_name> is a qualified name of the class that will be used to create the preprocessor, and <arguments> is any string that will be passed to the method creating the preprocessor. The location of the factory class must be specified in the -Xbootclasspath option passed to the VM, together with the location of the appropriate rt.jar.

    When the -Xprep option is specified, before loading the application classes, the Java VM will load the specified factory class and execute the method in the class declared as:

    class <factory_class_name> implements Preprocessor {
    public static Preprocessor createPreprocessor (String arg)

    where Preprocessor is an interface defined as:

    package hp.javatools.bytecode;
    public interface Preprocessor {
    public abstract byte[] instrument (String name, byte[] klass); }

    The VM will pass the <arguments> specified in the -Xprep option to the createPreprocessor method as its only argument. The Preprocessor object returned by the invocation will be saved by the VM.

    For each subsequently loaded class, the VM will invoke the instrument() method on the Preprocessor object, passing the name of the class being loaded, and the bytecode representation of the class. The returned array of bytes will be used by the VM as the replacement of the original version of the class. If null is returned, the original version of the class will be used.

    -verbosegc or -verbose:gc

    Prints out the result of a garbage collection to the stdout stream. At every garbage collection, the following 5 fields are printed:

    [%T %B->%A(%C), %D]

    %T is "GC:" when the garbage collection is a scavenge, and "Full GC:" when its a full garbage collection. A scavenge collects live objects from the New Generation only, whereas a full garbage collection collects objects from all spaces in the Java heap.

    %B is the size of Java heap used before garbage collection, in KB.

    %A is the size after garbage collection, in KB.

    %C is the current capacity of the entire Java heap, in KB.

    %D is the duration of the collection in milliseconds.

    -Xverbosegc<options>

    To better understand how garbage collection works in the HotSpot JVM, we recommend the article "Improving Java Application Performance and Scalability by Reducing Garbage Collection Times and Sizing Memory Using JDK 1.4.1" (November 2002) by Nagendra Nagarajayya and J. Steven Mayer. The article describes garbage collectors in SDK 1.4.x, however it also covers the older collectors used in 1.3.x.

    In addition, we recommend HP's tool HPjtune, which graphically displays information contained in an Xverbosegc log.

    This option prints out detailed information about the spaces within the Java Heap before and after garbage collection.

    The syntax of the option is:

    -Xverbosegc[:help]|[0|1][:file=[stdout|stderr|<filename>]]

    :help prints this message.

    0|1 controls the printing of heap information:

    0 Print only after each full GC
    1 (default) Print after every Scavenge and Full GC

    :file=[stdout|stderr|<filename>] specifies output file

    stderr (default) directs output to standard error stream
    stdout directs output to standard output stream
    <filename> file to which the output will be written

    At every garbage collection, the following 18 fields are printed:
    <GC: %1 %2 %3 %4 %5 %6 %7 %8 %9 %10 %11 %12 %13 %14 %15 %16 %17 %18 >

    1: Indicates the cause of the garbage collection.
        -1: indicates a scavenge (during a scavenge only objects from
            the New space are collected)
        0-6: indicates a full garbage collection (during a full garbage
             collection objects from all areas are collected)
    The code indicates the reason for the full garbage collection as follows:
        Reason:
        0:  Call to System.gc()
            The call was made by the application.
        1:  Old Generation full
            An object was to be allocated in the Old Generation, but there
            was no room there.
        2:  Permanent Generation full
            The heap area holding the reflection objects (representing Java
            classes and methods) was full.
        3:  Train Generation full
            Reserved for Java Virtual Machine developers
        4:  Old generation expanded on last scavenge
            Due to implementation reasons, if the Old Generation expanded
            on last scavenge, no more scavenges can be performed reliably
            before the next full garbage collection.
        5:  Old generation too full to scavenge
            An object was to be allocated in the New Generation, but there
            was no room there.  However, the VM has determined that the Old
            Generation was likely too full for a scavenge to compete without
            expanding the Old Generation.  Therefore a full garbage collection
            was performed rather than a scavenge.
        6:  FullGCAlot
            Reserved for Java Virtual Machine developers.
    %2: The time elapsed between the Java program start and the start of this
        garbage collection event.
    %3: Garbage collection invocation. This is the sequential number (count) of the
        garbage collection event. Counts of Scavenge and
        Full GCs are maintained separately. 
    %4: Size of the object allocation request that forced the GC, in bytes.
    %5: Tenuring threshold - determines how long the new born object
        remains in the New Generation.
        The report includes the size of each space:
           Occupied before garbage collection (Before)
           Occupied after garbage collection (After)
           Current capacity (Capacity)
    All are in bytes.
    Eden sub-space (within the New Generation)
    %6:  Before
    %7:  After
    %8:  Capacity
    Survivor sub-space (within the New Generation)
    %9:  Before
    %10: After
    %11: Capacity
    Old Generation
    %12: Before
    %13: After
    %14: Capacity
    Permanent Generation (Storage of Reflective Objects)
    %15: Before
    %16: After
    %17: Capacity
    %18: Duration of the garbage collection in seconds.
    

    FastSwing

    FastSwing is an HP feature which provides significant performance improvement for Swing Applications on a Remote X-Server. Remote X-Servers include X-Terminals, PC-XServers like Exceed and Reflection X and remote unix workstations.

    To use this feature invoke java or appletviewer as follows:

    /opt/java1.3/bin/java -Dhp.swing.useFastSwing=true MyApp

    or

    /opt/java1.3/bin/appletviewer -J-Dhp.swing.useFastSwing=true applet.html

    Currently we recommend using this feature only for Remote displays as it has the following caveat:

    Double-buffered Swing Components cannot perform Graphics2D operations with the FastSwing feature turned on. When doing so you will get the following exception:

    java.lang.ClassCastException: sun.awt.motif.X11OffScreenImage

    at BezierAnimationPanel.run(BezierAnimationPanel.java:223)
    at java.lang.Thread.run(Unknown Source)

    large Java heap sizes

    expanding heap size in native applications (hp-ux 11.0, 11i PA-RISC)
    If you embed libjvm in a native application, and wish to use a large Java heap, you need to ensure that enough private data space is enabled. On HP-UX 11.0 and 11i PA-RISC, by using HP-UX's EXEC_MAGIC linked with "-N" you can expand your available memory space from 1GB to around 1.7GB.

    HP-UX 11.0 PA-RISC

    With the installation of the required patches shown below (or their superseded patches), you can get larger Java heap by using the command below.

    Required Patches: PHKL_27278, PHKL_23409, PHKL_27364, PHKL_26136

    For 1500MB to 2400MB of Java heap:

    chatr +q3p enable <executable name>

    HP-UX 11i PA-RISC

    With the installation of the required patch shown below (or its superseded patch), you can get larger Java heap by using the commands below.

    Required Patch: PHKL_27278 (or its superseded patch)

    For 1500MB to 2400MB of Java heap:

    chatr +q3p enable <executable name>

    For 2400MB to 3.8GB of Java heap:

    chatr +q3p enable +q4p enable <executable name>

    See also "Application Dependent Considerations When Using Large Heap Size" in these release notes.

    expanding heap size (hp-ux 11i PA-RISC)
    Hotspot 1.3.1 now supports heaps up to 3.0GB on HP-UX 11i, with the installation of the patch shown below.

    For Java invoked from the command line on HP-UX 11i, Java will automatically choose an appropriate executable.

    Required patch: PHKL_27278 (or its superseded patch)

    For heaps less than 1500MB, the executable is 'java'.

    For heaps greater than or equal to 1500MB, and less than 2400MB the executable is 'java_q3p'.

    For heaps of 2400MB to 3.8GB, the executable is 'java_q4p'.

    You do not need to directly invoke these programs. Just invoke 'java' as usual, and the appropriate program will be run for you.

    In addition, be aware that if you wish to use very large heaps, because of segmentation in the HP-UX virtual address space, when the Java heap is larger than 3000MB, either new space (-Xmn) or old space (-mx minus -Xmn) must be approximately 850MB or less.

    See also the next section, "Application Dependent Considerations When Using Large Heap Size."

    application dependent considerations when using large heap size

    Thread stacks and C heap are allocated from the same address space as the Java heap, so if you set the Java heap too large, new threads may not start correctly. Or some other part of the runtime or native methods may suddenly fail if the C heap cannot allocate a new page. An application may start up correctly with a 1.7GB heap, but this does not necessarily mean it's going to work correctly.

    For example, if you use a 1MB stack size, and have about 80 threads in the process, you will have 80MB for stacks. If you have native libraries, you would probably add another 64MB for C heap. You have now used a total of 144MB of your heap for other things.

    Since all programs have varying C heap requirements and have varying numbers of threads, it's difficult to ascertain what the effect will be of running the application at its limit. It's important to understand the real requirements of your application. We recommend that you perform sizing tests before deployment with a realistic load, while monitoring with the -Xverbosegc option and a tool like GlancePlus.

    For more on the -Xverbosegc option, go to the HotSpot chapter of the HP-UX Programmer's Guide for Java 2.

    For more information on GlancePlus, see HP GlancePlus software .

    For more information on performance tuning Java on HP-UX, refer to our Performance Tuning website.

    using WDB to examine backtraces in Java thread stacks

    You can now use HP's debugger WDB 3.0.01 (the Gnu Debugger gdb) to examine backtraces containing mixed language frames (Java and C/C++) in Java thread stacks. This will simplify debugging the VM and Java mixed-language applications. Set the environment variable GDB_JAVA_UNWINDLIB to the pathname of the Java Unwind Share Library libjunwind.sl (PA) or libjunwind.so (IA), which is in the SDK.

    The default location of the Java Unwind Library in the SDK is:

    /opt/java1.3/jre/lib/PA_RISC/server.libjunwind.sl
    /opt/java1.3/jre/lib/PA_RISC2.0/server/libjunwind.sl
    /opt/java1.3/jre/lib/IA64/server/libjunwind.so

    For example, in ksh, you should set the environment variable like this:

    For PA1.1 machines:

    export GDB_JAVA_UNWINDLIB=/opt/java1.3/jre/lib/PA_RISC/libjunwind.sl

    For PA2.0 machines:

    export GDB_JAVA_UNWINDLIB=/opt/java1.3/jre/lib/PA_RISC2.0/libjunwind.sl

    For Itanium Processor Family machines:

    export GDB_JAVA_UNWINDLIB=/opt/java1.3/jre/lib/IA64/libjunwind.so

    If you installed the SDK in a location other than the default, you would substitute the non-default location for "/opt/java1.3" in the above commands. Then use WDB as usual to debug your Java applications or core files. See the tutorial slides on Debugging Native Code with gdb (WDB) (PDF, 245KB) for help on how to use the new Java stack unwind functionality.

    » Download the latest Adobe® Acrobat® Reader

    Asian TrueType fonts and Asian locales

    The SDK now supports HP-UX Asian TrueType fonts, with the installation of patches.

    For information on which HP-UX patches you need for each language and for additional documentation on fonts, please see the document HP-UX Fonts and the Java™ Runtime Environment.

    The following Asian Locales are now supported by HP's 1.3.1.08 SDK with TrueType fonts.

    ja_JP.SJIS   Japanese with Shift-JIS encoding
         
    ja_JP.eucJP   Japanese with JIS EUC encoding
         
    ko_KR.eucKR   Korean with KSC5601 EUC encoding
         
    zh_CN.gb18030   Simplified Chinese with GB18030 encoding
    (supported only on HP-UX 11i)
         
    zh_CN.hp15CN   Simplified Chinese, with GB2312 EUC encoding
         
    zh_TW.big5   Traditional Taiwan Chinese with Big5 encoding
         
    zh_TW.eucTW   Traditional Taiwan Chinese with CNS11643 encoding (planes 1-3)
         
    zh_HK.hkbig5   Traditional HongKong Chinese with Big5 HK encoding
    In HP-UX Fonts and the Java™ Runtime Environment you will find information on the following topics:

    date/time methods - new defaults

    Since SDK 1.2.2.09 and SDK 1.3.1, there has been a change in the way the HotSpot JVM uses the gettimeofday() system call to obtain date and time information.

    For performance reasons a new mechanism is used that uses the number of cpu ticks since the application started, to calculate the current time.

    As a result, changes to the system date or time using date(1), adjtime(2) or time synchronization utilities such as ntp will not be reflected in the date and time that Java returns, until the process is restarted.

    If your application requires that Java immediately reflects such system time changes, you can use the -XX:+UseGetTimeOfDay option to tell the JVM to use the gettimeofday call instead of the new, lightweight mechanism. However you may notice a drop in performance.

    profiling capability added

    In the SDK version 1.3.1.05, a SIGPROF handler to support future profiling capability is installed automatically. This may cause incompatibilities with other native code or profiling tools which use SIGPROF.

    You can turn off the SIGPROF handler by using the following option:

    -XX:+ReduceSignalUsage

    However you should be aware that using this option also turns off the SIGQUIT handler. Therefore you will not be able to get a Java stack trace.

    Using JNI - main/primordial thread stack size limits

    The primordial thread is the first thread when a process is created. This is the thread that has the main method. It is also called the main thread. The primordial thread stack size is controlled by the kernel parameter maxssiz or maxssiz_64bit. The Java VM (JVM) has two options for controlling the stack size:

    -XX:MainThreadStackSize=n
    -Xss[n][k or m]

    In the Java VM the size of the primordial thread is restricted to the *greater* of MainThreadStackSize (default 2M) or ThreadStackSize (specified by -Xss). For example, if you specify -Xss1m, the JVM still takes 2M for the main thread. And if you specify -Xss4m, the JVM takes 4M for the main thread as well.

    If your application calls JNI_CreateJavaVM or JNI_AttachCurrentThread from the primordial thread, under certain conditions the stack usage could cross the JVM-imposed primordial thread stack size limit of 2M and cause a stack overflow situation in the native code itself, even if no Java code was ever run on the primordial thread.

    One workaround is to use the option -XX:MainThreadStackSize=<value> to increase the primordial thread stack size, However be aware that -XX options are non-standard options, and are liable to change from release to release.

    Other workarounds and examples are provided in our Programmer's Guide along with the effects of using maxssiz on the amount of writeable data space available to the application. Refer to the JNI chapter of our Programmer's Guide for Java 2 at: Using Java™ 2 JNI on HP-UX

    Using JNI - non-main/primordial thread stack size limits

    The default stack size for 1.4 64-bit mode JVM- created threads is 1MB. On PA-RISC 32 and 64-bit systems, the default stack size is 64KB. Therefore, if you are using C language main programs that attach with JNI, you will want to adjust the stack size to avoid overflows.

    Here are some suggestions to work around stack overflow problems in threads other than the main thread:

  • If the thread is created in native code and is attached to Java through JNI_AttachCurrentThread, increase the stack size attribute when creating the thread with pthread_create.
  • If the thread is created inside Java and there is a stack overflow condition, increase the thread stack size with -Xss<n>
  • using JNI—dereferencing null pointers

    In Java 2, JNI code that incorrectly dereferences NULL will result in a SIGSEGV, which may be different behavior than that experienced with Java 1.1 releases. For example, in Java 1.1, JNI methods that dereference NULL pointers like this:

    int *p = NULL;
    return *p;

    will return the value 0.

    With the Java 2 the HotSpot VM, such dereferences result in a SIGSEGV, and java NULL checks can be performed without having to emit explicit code to do so. With Java 2 HotSpot, the signal is caught, and a null pointer exception is thrown if the offending instruction was within the VM (compiled code, or in the interpreter). This method may uncover hidden programming errors.

    Also, if you are including the HP-UX Runtime Environment for the Java 2 Platform in an application and bypassing our standard driver, for example by making calls to JNI_CreatJavaVM from inside the application, link the application with the "-z" option. The -z option will indicate that dereferencing NULL pointers in the application should generate a SIGSEGV instead of the traditional behavior of returning zero. If you do not, you will not be able to take advantage of implicit null pointer checks; null pointer checks will have to be explicit, potentially degrading performance. Linking with -z may also expose existing but quiet bugs in an application. This is because the SIGSEGVs were not being generated before.

    using JPDA

    The JPDA architecture and components are documented at http://java.sun.com/products/jpda/ and http://java.sun.com/reference/docs/index.html. HP's SDK version 1.3.0.01 supports both the HotSpot and Classic VMs. The jdb for HP-UX is similar to the Solaris implementation. For usage information, see "Sun VM Invocation Options" at http://java.sun.com/products/jpda/doc/conninv.html.

    manually launching the application VM
    (excerpted from http://java.sun.com/products/jpda/readme.html)

    If you are running the version of jdb provided in this release, the application VM is launched for you with the debugger back end loaded. However, in the following cases, you will be launching your own application VM, either by hand or in your implementation.
    1. Remote debugging with the -attach or -listen jdb option.
    2. You are implementing a debugger which uses the JDWP directly.
    3. You are implementing a debugger back end which uses JVMDI.

    Currently, the first two cases require a command line like the following:

    java -Xdebug -Xnoagent -Djava.compiler=NONE
    -Xrunjdwp:transport=dt_socket,server=y,suspend=y
    -classpath class-path class-name

    The -Xdebug option enables debugging. The -Xnoagent disables the default sun.tools.debug debug Agent. The -Djava.compiler=NONE disables the JIT compiler.

    For the third case, you must use the same command line options as described above, but you are free to use your own mechanism for loading the JVMDI client into the application VM. You do not need to use -Xrun.

    The Connection and Invocation Details document at http://java.sun.com/products/jpda/doc/conninv.html contains more information on necessary VM invocation options and sub-options of -Xrunjdwp.

    Closing a socket when accept or read is pending (PA-RISC)

    Because of changes to the mechanism by which a socket is closed in the VM, you no longer need to use the -XdoCloseWithReadPending option we recommended in earlier releases. The new mechanism uses an ioctl call for which the following two patches (or their superseded patches) are required for PA-RISC only:

    PHNE_26771 for HP-UX 11.00
    PHNE_28089 for HP-UX 11.11

    C and C++ libraries

    Libraries compiled with the cfront HP C++ compiler will not work with HotSpot. HotSpot requires use of the HP aC++ compiler for any C++ libraries loaded dynamically at runtime.

    CLASSPATH

    CLASSPATH will be automatically set during installation. If you get the error "Unable to initialize threads: cannot find class java1.3/lang/Thread", your CLASSPATH environment variable may be incorrect.

    See also http://java.sun.com/reference/docs/index.html.

    JAR files

    If you run an applet from a JAR file, the classes files should only be available in the JAR file, and the JAR file should not be in the class path.

    compatibility with previous releases

    Compatibility information is provided at: http://java.sun.com/j2se/1.3/compatibility.html.

    migrating from Java 1.1 to Java 2

    Information to assist you in migrating from 1.1 to Java 2 is provided at http://java.sun.com/reference/docs/index.html.

    The SDK 2 java launcher tool replaced the JDK 1.1 jre tool.

    web sites with more information

    The following websites have additional information:

    Java Standard Edition Platform Documentation:
    » http://java.sun.com/docs

    Java 2 SDK version 1.3 features and tools:
    » http://java.sun.com/reference/docs/index.html

    Java 2 SDK version 1.2 features and tools:
    » http://java.sun.com/products/jdk/1.2/docs/index.html

    Java 2 version 1.3 API Specification:
    » http://java.sun.com/j2se/1.3/docs/api/index.html

    Java 2 platforms and APIs - Authorized Books:
    » http://java.sun.com/docs/books/index.html

    Java tutorial:
    » http://java.sun.com/docs/books/tutorial/index.html

    Java Security:
    » http://java.sun.com/security/index.html

    Java Products and APIs:
    » http://java.sun.com/products/
    » http://java.sun.com/j2se/

    Swing Connection newsletter:
    » http://java.sun.com/products/jfc/tsc/index.html

    Performance Tuning Java on HP-UX
    » Java™ Performance Tuning

    problem fixes and known issues

    Known JavaSoft bugs are documented in the Bug Database at http://developer.java.sun.com/

    Information on JavaSoft 1.3 fixed bugs is available at http://java.sun.com/docs/index.html

    problem fixes

    The 1.3.1.08 release includes enhancements and fixes from previous 1.3.1 releases, and in addition includes the following enhancements and defect fixes.

    JAGae43143 HP SR 8606279087 Enhancement: random number enhancements
    JAGae49130 HP SR 8606285189   -Xeprof option generates data files which are rejected by HPjmeter
    JAGae52239 HP SR SR 8606288308 Enhancement: Selective disabling of loop optimizations from 1.4.0
    known issues

    Below is some information on a few problem topics.

    using the C++ (-AA) option (PA-RISC only) 
    If you are running HP-UX 11.0 or 11.11 PA-RISC, you cannot use the ANSI C++ standard (-AA) option in an application that loads Java. Starting with SDK 1.4.1.03, there is a workaround for this problem.

    NOTE: This only affects PA-RISC systems; on Itanium systems, the C++ runtime libraries are designed to support -AA by default.

    using X font server with Asian true type fonts
    Because of a defect, we recommend that you do not run the X font server with Asian true type fonts. This problem is fixed in the SDK 1.4 release.

    using -Xeprof with -Xcomp (Itanium)
    On Itanium-based systems, using the -Xeprof option in conjunction with -Xcomp may result in abrupt VM termination (abort). The workaround is to not specify -Xcomp when using the
    -Xeprof option. This defect will be fixed in a future release.

    shl_load HotSpot libjvm problem due to TLS
    The libjvm library for the HotSpot 1.01 JVM uses thread local storage (TLS). Currently the dynamic loader that is used by shl_load does not support dynamically loading a shared library containing TLS when the library was not included in the link line.

    You may have a need to load a library dynamically (using shl_load or dlopen) that contains TLS, for example libjvm.sl, without having linked your application against it. For example, this might be the case if your application uses plug-ins.

    The current workaround is a new linker feature LD_PRELOAD that is available for HP-UX 11.0 in patch PHSS_26559. For HP-UX 11i the feature is included. For more information on LD_PRELOAD functionality and its limitations, read the man page for dld.sl AFTER you have installed the patch.
    on stack replacement (PA-RISC only)
    In this version, on stack replacement is off by default.
    To enable it, use the -XX:+UseOnStackReplacement option.

    If the VM is executing a method that has a loop with a large number of iterations, an intra-method hotspot will occur. In order to get better performance, the method should run in compiled mode instead of interpreted mode. On stack replacement enables the interpreter to go into compiled code while it is executing the same instance of the method call. If you use on stack replacement, you should also enable compiler safe points (see the next section).

    HotSpot compiler safe points (PA-RISC)
    NOTE: For both HP-UX 11.0 and 11i, using Compiler Safe Points requires a patch. The required patch numbers are shown below. For information on locating and installing the patches, go to the "Installation" section in this document.

    HP-UX 11.0 PHKL_27770
    HP-UX 11i PHKL_24751
    In this version, compiler safe points is off by default. To turn it on, use the -XX:+UseCompilerSafepoints option. Enabling compiler safepoints guarantees a more deterministic delay to stop all running java threads before doing a safepoint operation, namely garbage collection and deoptimization.

    missing property files for Japanese printing
    Beginning with the SDK 1.3 for HP-UX, postscript printing is supported. The property files necessary for printing Japanese fonts are shipped as
    psfont.properties.ja and psfontj2d.properties.ja.

    using linker option +noenvvar on Itanimum and PA-64 systems
    If your application links with libjvm and uses the JNI interface APIs to load the JVM directly, do not use the linker option +noenvvar on Itanium or PA-64 systems. The defect does not exist on 32-bit systems. It is expected to be fixed in a future release.

    running Java with setuid or setgid
    Java requires dynamic loading (SHLIB_PATH, LD_LIBRARY_PATH) which are disabled in setuid or setgid executables. Therefore Java cannot run with setuid or setgid.

    running Aries Itanium emulation on PA2.0
    If you are running the Aries Itanium emulator on a PA2.0 system with HP-UX 11.22 or 11.23, a defect exists which causes Java math calculations to produce incorrect math routine results.

    This problem applies to those applications that do Java math calculations, specifically any that use the constant Float.MAX_VALUE, or a floating point value that is the maximum floating value.

    Because the same Java math calculations are used in the GUI code, applications that use GUIs will also be affected.

    This problem is fixed by installing the following patches:

    HP-UX 11.22: PHSS_29654
    HP-UX 11.23: PHSS_29658

    legal notices
    Copyright © Hewlett-Packard Company 2003

    Java and all Java-based trademarks and logos are trademarks or registered trademarks of Sun Microsystems, Inc. in the U.S. and other countries.