Problem Description 	: Javac crashes hotspot VM with -server -Xcomp options on windows platforms

Tested_Java_Release   	: HotSPot Express PIT build (6U3 pre-build b01)
Tested_Build          	: /net/vmsqe/export/weekly/mustang/weeklyjdk/UR/HEb01 (PIT Build)
Operating System	: Windows only
Machine Name 		: EM64t-001.sfbay
PRT Archiver bit are : /net/prt-archiver.sfbay/data/archived_workspaces/1.6/update3/baseline/2007/20070619161220.kvn.6570219_6u3

Test results            : http://gtee.sfbay.sun.com/gtee/results/MUSTANG_UR/PROMOTION/HE-WEEKLY/not-070619220840/vm/WINDOWS-I586-ANY/vm-vm_6.0_server_comp_WINDOWS-I586-ANY2007-06-19-22-50-43/analysis.html#New_Failures
http://gtee.sfbay.sun.com/gtee/results/MUSTANG_UR/PROMOTION/HE-WEEKLY/not-070619220840/vm/WINDOWS-I586-ANY/vm-vm_6.0_server_comp_WINDOWS-I586-ANY2007-06-19-22-28-49/analysis.html#New_Failures

NOTE : 
1.) This happens only with the Hotspot express pit builds & was not reproducible on promoted 6U2 B05 build
2.) I dint find an appropriate Product release for the hotsport express bug, so using 6U3/b01 for now as a place holder

Steps To Reproduce :  

C:\local\common\testbase\jtreg\UR_MM_REGRESSION\test\java\lang\management\CompositeData>Y:\weekly\mustang\weeklyjdk\UR\HEb01\windows-i586\bin\javac -J-Xcomp -server OpenTypeConverter.java
#
# An unexpected error has been detected by Java Runtime Environment:
#
#  EXCEPTION_ACCESS_VIOLATION (0xc0000005) at pc=0x081f68a0, pid=69596, tid=8885
2
#
# Java VM: Java HotSpot(TM) Tiered VM (20070615124853.et151817.baseline compiled
 mode windows-x86)
# Problematic frame:
# V  [jvm.dll+0x1f68a0]
#
# An error report file with more information is saved as:
# C:\local\common\testbase\jtreg\UR_MM_REGRESSION\test\java\lang\management\Comp
ositeData\hs_err_pid69596.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.
#
Posted Date : 2007-06-21 18:24:06.0

N/A

This fails all the way back to jdk1.7.0 b03.  I don't know why the testing for 1.7 doesn't include running this same test.  It sure would be nice if we'd found this earlier.

It looks like it's caused by 20060913073259.nips.unswitch

Running 20060913073259.nips.unswitch
# To suppress the following error report, specify this argument
# after -XX: or in .hotspotrc:  SuppressErrorAt=/loopnode.hpp:586]
#
# An unexpected error has been detected by Java Runtime Environment:
#
#  Internal Error (/net/prt-solx86-q1-6/PrtBuildDir/workspace/src/share/vm/opto/loopnode.hpp, 586), pid=15744, tid=15
#
# Java VM: Java HotSpot(TM) Server VM (20060913073259.nips.unswitch-debug compiled mode)
#
# Error: assert(n != 0,"Bad immediate dominator info.")
# An error report file with more information is saved as hs_err_pid15744.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
#
Current thread is 15
Dumping core ...
Abort - core dumped

Here's a command line to reproduce it:

java -Xcomp -classpath /java/re/jdk/1.7.0/latest/binaries/solaris-i586/lib/tools.jar com.sun.tools.javac.Main /net/smite/never/tmp/OpenTypeConverter.java
Posted Date : 2007-06-21 20:06:30.0

Probably wasn't caught earlier as the issue with JTreg (not passing VM options - to javac) was resolved only recently.
Posted Date : 2007-06-21 23:27:44.0

Smaller test case:

java -server -Xcomp -Xbatch \
 -XX:CompileOnly=com/sun/tools/javac/main/JavaCompiler.generate \
 -XX:CompileOnly=com/sun/tools/javac/tree/JCTree.pos \
 -XX:CompileOnly=com/sun/tools/javac/util/Log.error \
 -classpath /java/re/jdk/1.7.0/latest/binaries/solaris-i586/lib/tools.jar \
 com.sun.tools.javac.Main OpenTypeConverter.java

OpenTypeConverter.java is in attachment.
Posted Date : 2007-06-27 16:33:47.0

This failure is now visible on other platforms too. Here is the list of platforms where similar javac crash were seen using JDK 6U2 b05.

[1] 64BIT Solaris AMD 64
#
# An unexpected error has been detected by Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0xfffffd7ffe8c2fc1, pid=20926, tid=10
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (20070621105710.km88527.6570268_6u3 compiled mode solaris-amd64)
# Problematic frame:
# V  [libjvm.so+0x2c2fc1]
#
# An error report file with more information is saved as:
# /export/local/1847.MUSTANG_UR.PROMOTION.VM+vm_6.0_server_comp_64BITSOLARIS-AMD64_1135523F341/results/java/lang/management/MemoryMXBean/CollectionUsageThresholdConcMarkSweepGC/hs_err_pid20926.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.
#

[2] 64BIT Solaris SPARC
#
# An unexpected error has been detected by Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0xffffffff7dd0edc4, pid=26605, tid=8
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (20070621105710.km88527.6570268_6u3 compiled mode solaris-sparc)
# Problematic frame:
# V  [libjvm.so+0x10edc4]
#
# An error report file with more information is saved as:
# /export/local/1847.MUSTANG_UR.PROMOTION.VM+vm_6.0_server_comp_64BITSOLSPARC_1135523F0A3/results/java/lang/management/MemoryMXBean/LowMemoryTestSerialGC/hs_err_pid26605.log
#
# If you would like to submit a bug report, please visit:
#   http://java.sun.com/webapps/bugreport/crash.jsp
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.
#

I also ran the same set of tests with other VM options which did not result in the crash (Eg: -XX:+UseConcMarkSweepGC -XX:+CompileThreshold=100).
Posted Date : 2007-06-28 06:35:52.0

why does this continue to happen

how can this be fixed

since this has happened it now will not let me play the game it happened in.  What can I do to fix this.

I am not able to play past 2 holes in the mini golf madness game.  It has rebooted itself a total of 9 times this time around.  Earlier when I tried to play this game it rebooted itself a total of 10 times.  Plus this is doing this in a few other games.  Also some games that I used to be able to play I no longer can because it won't load or when it does after a couple of hands played it starts the rebooting again.  This is getting to be a real headache.  All of these games happened in Pogo.   Sincerely yours.  Pam Brown