Has anyone run IcedTea7 jtreg tests recently?

[Omair Majid]

Andrew Haley wrote:
> Omair Majid wrote:
>> Changing ulimit -n to 4096 (from 1024) fixes the too many open files
>> problem, but now I am getting a jvm crash. Here is the relevant part of
>> the output:
>>
>> [snip]
>> FAILED: demo/jvmti/waiters/WaitersTest.java
>> FAILED: demo/nio/ZipFileSystem/sanity.sh
>> FAILED:
>> java/awt/appletviewer/IOExceptionIfEncodedURLTest/IOExceptionIfEncodedURLTest.sh
>>
>> #
>> # A fatal error has been detected by the Java Runtime Environment:
>> #
>> #  SIGSEGV (0xb) at pc=0x0000000002000101, pid=3555, tid=140601982036304
>> #
>> # JRE version: 7.0-b49
>> # Java VM: OpenJDK 64-Bit Server VM (15.0-b02 mixed mode linux-amd64 )
>> # Distribution: Built on Fedora release 10 (Cambridge)
>> # Problematic frame:
>> # C  0x0000000002000101
>> #
>> # An error report file with more information is saved as:
>> [snip]
>>
>> I have attached the error report.
> 
> Care to run that under gdb and tell us where the SEGV occurred?
> 
> Andrew.
> 

The segfault is not directly caused by 
java/awt/Component/isLightweightCrash/IsLightweightCrash.java as the 
hotspot log might indicate. It looks like com/sun/net/httpserver/bugs/ 
tests have to be run before the java/awt/ tests to reproduce the error. 
Infact, I can change where the crash happens by changing when I attach 
gdb to java. gdb backtrace and hotspot error log attached.

Omair
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: gdb-backtrace
Url: http://mail.openjdk.java.net/pipermail/distro-pkg-dev/attachments/20090310/d847bad8/gdb-backtrace.ksh 
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: hs_err_pid24110.log
Url: http://mail.openjdk.java.net/pipermail/distro-pkg-dev/attachments/20090310/d847bad8/hs_err_pid24110.log