test_threaded_nonlocal_return test sometimes fail with an apparent threading
issue
----------------------------------------------------------------------------------
Key: JRUBY-2653
URL: http://jira.codehaus.org/browse/JRUBY-2653
Project: JRuby
Issue Type: Bug
Affects Versions: JRuby 1.1.2
Reporter: Charles Oliver Nutter
Assignee: Thomas E Enebo
Fix For: JRuby 1.1.3
See the failure here:
{noformat}
test_threaded_nonlocal_return(TestThreadedNonlocalReturn):
NoMethodError: undefined method `<<' for nil:NilClass
/export/home/nsieger/bamboo/home/xml-data/build-dir/JRUBY-TRSNAP/maven/jruby/target/test-classes/test/test_threaded_nonlocal_return.rb:16:in
`test_threaded_nonlocal_return'
/export/home/nsieger/bamboo/home/xml-data/build-dir/JRUBY-TRSNAP/maven/jruby/target/test-classes/test/test_threaded_nonlocal_return.rb:16:in
`times'
/export/home/nsieger/bamboo/home/xml-data/build-dir/JRUBY-TRSNAP/maven/jruby/target/test-classes/test/test_threaded_nonlocal_return.rb:16:in
`test_threaded_nonlocal_return'
{noformat}
I believe this is caused by the lazy initialization of the array 'aggs' as in
'aggs[j] = []'. My guess is that when multiple threads are doing this, they may
step on each other by nil'ing a range of values believed to be non-existent
while at the same time another thread is actually populating them. The fix is
to pre-initialize all ten slots in the aggs array so there's no cross-thread
mutation of the aggs array itself.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
