Bugzilla Bug 187624

*pmd-3.9 (09 May 2007)

  09 May 2007; Vlastimil Babka <caster@gentoo.org>
  +files/pmd-3.9-build.xml.patch, +files/pmd-3.9-launcher-pre-commands,
  +pmd-3.9.ebuild:
  Version bump, bug #170603. Initially worked on in java-overlay by Miroslav
  Šulc <fordfrog@gentoo.org> and Petteri Räty <betelgeuse@gentoo.org>.
  Migrated to split-ant, no longer installs symlink to
  /usr/share/ant-core/lib.

Created an attachment (id=128020) [edit]
emerge --info

dev-java/backport-util-concurrent has test failures but does not die:

 [java] Time: 180.484
     [java] There were 2 failures:
     [java] 1)
testTimedInvokeAll6(ScheduledExecutorTest)junit.framework.AssertionFailedError
     [java]     at
ScheduledExecutorTest.testTimedInvokeAll6(ScheduledExecutorTest.java:1140)
     [java]     at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
     [java]     at
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
     [java]     at
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
     [java] 2)
testTimedInvokeAll6(ThreadPoolExecutorSubclassTest)junit.framework.AssertionFailedError
     [java]     at
ThreadPoolExecutorSubclassTest.testTimedInvokeAll6(ThreadPoolExecutorSubclassTest.java:1649)
     [java]     at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
     [java]     at
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
     [java]     at
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
     [java]
     [java] FAILURES!!!
     [java] Tests run: 1859,  Failures: 2,  Errors: 0

(In reply to comment #1)
> Created an attachment (id=128020) [edit]
> emerge --info
> 
> dev-java/backport-util-concurrent has test failures but does not die:
> 

Do they work with 3.0?

(In reply to comment #2)
> (In reply to comment #1)
> > Created an attachment (id=128020) [edit]
> > emerge --info
> > 
> > dev-java/backport-util-concurrent has test failures but does not die:
> Do they work with 3.0?

 3.0 is hard-dep of pmd-3.9, so...no.  asm-3.0 is another dep that has to go
stable, I tested them all together, so no real need to file individual bugs I
think.

(In reply to comment #1)
> 
> dev-java/backport-util-concurrent has test failures but does not die:
> 

On my server with java5 use flag and sun-jdk-1.5:

     [java] Time: 156.555
     [java]
     [java] OK (1859 tests)
     [java]

Works fine with 1.4 too. Really need more info about your failures. For example
which VM were you running them with.

USE=test Using: sun-jdk-1.4: failure (see above)
USE="java5 test" Using: sun-jdk-1.5: failure, but different, only one failure)

There was 1 failure:
     [java] 1)
testAwaitLockCount(ReentrantLockTest)junit.framework.AssertionFailedError
     [java]     at
ReentrantLockTest.testAwaitLockCount(ReentrantLockTest.java:1119)
     [java]     at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
     [java]     at
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
     [java]     at
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)

(In reply to comment #5)
> Works fine with 1.4 too. Really need more info about your failures. For example
> which VM were you running them with.

 Do you need anything more?  Are there any prerequisites for the tests? 

(In reply to comment #7)
> (In reply to comment #5)
> > Works fine with 1.4 too. Really need more info about your failures. For example
> > which VM were you running them with.
> 
>  Do you need anything more?  Are there any prerequisites for the tests? 
> 

The junit output files would be nice.

(In reply to comment #8)
> The junit output files would be nice.

 Without FEATURES=userpriv it runs well.  Have a look at
http://dev.gentoo.org/~opfer/backport-util-concurrent-3.0-test.tar.bz2

For the record, cannot reproduce the failures even with/without userpriv, the
tarball has no outputs and we can't even remember if it was result of good or
bad run. The tests are most probably timing stuff and such can fail with
different CPU usage etc. It doesn't die on them, so can be stabilized.

x86 stable, good we talked about it Caster...

ppc stable