More C1 crashing in JRuby

Charles Oliver Nutter charles.nutter at sun.com
Fri Oct 10 07:43:42 PDT 2008


Another JRuby user reported SIGSEGV in C1 with a simple snippit of Ruby 
code:

c = 1

while (true)
   flag = 0
   (1..20).each do |e|
     flag += (c % e)
   end
   break if (flag == 0)
   c += 1
end
c

I was able to reproduce it using Landon Fuller's Java 6 and OpenJDK 7 
builds on OS X, and have attached the dumps to this email (8135 is Java 
6, 8160 is Java 7). It did not appear to crash on Apple Java 5. I've 
also attached the original reporter's dump file (2912), from Windows XP 
+ Java 6.

As you'd expect, switching to C2 makes the problem disappear. And like 
last time, I'm willing to make whatever changes necessary to work around 
the problem, if possible.

The JRuby bug report is here:

http://jira.codehaus.org/browse/JRUBY-3049

Latest JRuby nightly builds are available here (unpack, put bin/ in 
PATH, run 'jruby test.rb' where test.rb is the above script; defaults to 
-client, pass --server to switch):

http://jruby.headius.com:8080/hudson/job/jruby-dist/

Do you want me to file this bug also through the usual channels?

- Charlie
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: hs_err_pid8160.log
Url: http://mail.openjdk.java.net/pipermail/hotspot-dev/attachments/20081010/171ace3a/attachment.ksh 
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: hs_err_pid8135.log
Url: http://mail.openjdk.java.net/pipermail/hotspot-dev/attachments/20081010/171ace3a/attachment-0001.ksh 
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: hs_err_pid2912.log
Url: http://mail.openjdk.java.net/pipermail/hotspot-dev/attachments/20081010/171ace3a/attachment-0002.ksh 


More information about the hotspot-dev mailing list