[Bug 684] New: [regression] zero/shark fail to run the jdk jtreg tests with hs20

[bugzilla-daemon at icedtea.classpath.org]

http://icedtea.classpath.org/bugzilla/show_bug.cgi?id=684

           Summary: [regression] zero/shark fail to run the jdk jtreg tests
                    with hs20
           Product: IcedTea
           Version: 6-hg
          Platform: all
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P5
         Component: Zero
        AssignedTo: unassigned at icedtea.classpath.org
        ReportedBy: doko at ubuntu.com


seen on icedtea6, and icedtea6-1.10 on x86, x86_64, powerpc, didn't check other
platforms.

<doko> Passed: java/io/InputStreamReader/GrowAfterEOF.java
<doko> Passed: java/io/InputStreamReader/NullCreate.java
<doko> Passed: java/io/InputStreamReader/One.java
<doko> Exception in thread "ReadAheadIterator1"
java.util.NoSuchElementException
<doko>         at java.util.LinkedList.remove(LinkedList.java:805)
<doko>         at java.util.LinkedList.removeFirst(LinkedList.java:151)
<doko>         at
com.sun.javatest.TRT_Iterator.nextElement(TRT_Iterator.java:176)
<doko>         at com.sun.javatest.TRT_Iterator.next(TRT_Iterator.java:200)
<doko>         at
com.sun.javatest.util.ReadAheadIterator.readAhead(ReadAheadIterator.java:258)
<doko>         at
com.sun.javatest.util.ReadAheadIterator.access$000(ReadAheadIterator.java:36)
<doko>         at
com.sun.javatest.util.ReadAheadIterator$1.run(ReadAheadIterator.java:192)
<mjw> ah, we are using samevm of course
<mjw> doko, it probably is the last .jtr file touched
<doko> jdk/JTwork/jtData/harness.trace shows:
<doko> Test starting: java/io/InputStreamReader/One.java
<doko> Test finished: java/io/InputStreamReader/One.java: Passed. Execution
successful
<doko> so a new test is not started
<doko> mjw: do you think, omitting -samevm will help?
<mjw> doko, it might because it will start a fresh java process for each test
<mjw> doko, but maybe this is in the test runner framework itself?
<ptisnovs> doko: mjw but it will take eons to complete all jck tests ;-)
<doko> ptisnovs: so the alternative is not to run them?
<ptisnovs> doko: we have to investigate why it throws this exception. To be
honest I have newer seen it on my machines, but it does not mean it's not a bug
in jtreg, of course
<doko> ptisnovs: this is a shark build, configured with the optional hotspot
<ptisnovs> doko: is this exception always thrown on the same test (ie after
One.java passes)
<ptisnovs> doko: ?
<doko> ptisnovs: yes, my last log stops in the same location ...
<ptisnovs> doko: will look at it, probably test One.java broke something in JVM



<doko> ptisnovs: interesting, without -samevm, I get right at the start of the
jdk tests:
<doko> Passed: com/sun/awt/Translucency/WindowOpacity.java
<doko> Exception in thread "ReadAheadIterator0" java.lang.IllegalStateException
<doko>         at
com.sun.javatest.TestResult.setTestDescription(TestResult.java:1882)
<doko>         at
com.sun.javatest.TestResultTable.getCachedResult(TestResultTable.java:2159)
<doko>         at
com.sun.javatest.TRT_TreeNode.processFile(TRT_TreeNode.java:1045)
<doko>         at
com.sun.javatest.TRT_TreeNode.scanIfNeeded(TRT_TreeNode.java:639)
<doko>         at
com.sun.javatest.TRT_TreeNode.getChildCount(TRT_TreeNode.java:113)
<doko>         at
com.sun.javatest.TRT_Iterator$PseudoFrame.nextIndex(TRT_Iterator.java:904)
<doko>         at com.sun.javatest.TRT_Iterator.findNext(TRT_Iterator.java:599)
<doko>         at
com.sun.javatest.TRT_Iterator.nextElement(TRT_Iterator.java:179)
<doko>         at com.sun.javatest.TRT_Iterator.next(TRT_Iterator.java:200)
<doko>         at
com.sun.javatest.util.ReadAheadIterator.readAhead(ReadAheadIterator.java:258)
<doko>         at
com.sun.javatest.util.ReadAheadIterator.access$000(ReadAheadIterator.java:36)
<doko>         at
com.sun.javatest.util.ReadAheadIterator$1.run(ReadAheadIterator.java:192)
<doko> Passed: com/sun/crypto/provider/Cipher/AES/Test4511676.java
<ptisnovs> strange
<doko> again, new hotspot only
<ptisnovs> doko: may I use this HS on i686/x86_64?
<doko> ptisnovs: what do you mean? it works as pure hotspot.
<ptisnovs> doko: you wrote: "configured with the optional hotspot"
<doko> ptisnovs: --with-hotspot-build=hs20 --enable-shark
<ptisnovs> doko: ah great, will try it


-- 
Configure bugmail: http://icedtea.classpath.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.