EVALUATION

http://hg.openjdk.java.net/jdk7/hotspot-gc/hotspot/rev/883a82d6d41d
                                     

EVALUATION

http://hg.openjdk.java.net/hsx/hsx19/baseline/rev/f6b107f3629a
                                     

EVALUATION

http://hg.openjdk.java.net/jdk7/hotspot-rt/hotspot/rev/883a82d6d41d
                                     

EVALUATION

The test.out in his failure directoy:

( http://jqa.ireland/export/home0/results/1.5.0_24/Build02/Regression/5u24-b02-hsx-solaris10sun4u/issue6626217/failed/ )

..actually looks like a pass: it ends with an exception being thrown (rather than a crash):

  >>Loader2>> Returning clazz sun.misc.Launcher$AppClassLoader@14b7453:bug_21227
Exception in thread "main" java.lang.LinkageError: loader constraint violation: loader (instance of Loader2) previously initiated loading for a different type with name "many_loader"
	at java.lang.ClassLoader.defineClass1(Native Method)
	at java.lang.ClassLoader.defineClass(ClassLoader.java:621)
	at java.lang.ClassLoader.defineClass(ClassLoader.java:466)
	at Loader2.findClass2(Loader2.java:21)
	at Loader2.loadClass(Loader2.java:37)
	at java.lang.ClassLoader.loadClass(ClassLoader.java:252)
	at from_loader2.gen(from_loader2.java:5)
	at many_loader.make(many_loader.java:21)
	at bug_21227.main(bug_21227.java:26)



The test sript Test6626217.sh ends with:

${JAVA} ${BIT_FLAG} -Xverify -Xint -cp . bug_21227 >test.out 2>&1
grep "violates loader constraints" test.out
exit $?

~
~

5.0-rev-latest uses the phrase "Class %s violates loader constraints"

But hsx16 uses the phrase:  "loader constraint violation: " plus more detail.

So the test for 5.0 doesn't recognise the error thrown by hsx16.
                                     

SUGGESTED FIX

I'll propose changing the testcase to:

egrep "violates loader constraints|loader constraint violation" test.out