Verify VarHandle class initialization

[JasonFengJ9]

Verify VarHandle class initialization

When a static field hasn't been initialized implicitly and a VarHandle is created to access this field. The field has to be initialized before returning the value via the VarHandle.
This test verifies that a VarHandle returns correct field value when there is no implicit static field initialization.

close #3205
depends #3928

Without dependent PR, this test fails with following stacktrace:

FAILED: testFinalFieldInClassNotInitialized
java.lang.AssertionError: expected [1] but found [0]
	at org.testng.Assert.fail(Assert.java:96)
	at org.testng.Assert.failNotEquals(Assert.java:776)
	at org.testng.Assert.assertEqualsImpl(Assert.java:137)
	at org.testng.Assert.assertEquals(Assert.java:118)
	at org.testng.Assert.assertEquals(Assert.java:652)
	at org.testng.Assert.assertEquals(Assert.java:662)
	at org.openj9.test.varhandle.StaticFieldVarHandleTests.testFinalFieldInClassNotInitialized(StaticFieldVarHandleTests.java:2463)

Manually verified that this test and the test from #3205 pass with #3928.

Reviewer: @DanHeidinga
FYI: @llxia

Signed-off-by: Jason Feng fengj@ca.ibm.com

[DanHeidinga]

@JasonFengJ9 Do the tests pass on the RI as well?

[JasonFengJ9]

@DanHeidinga yes, I verified that RI passes the standalone testcase (same code snippet as this PR).

[DanHeidinga]

Jenkins test sanity zlinux jdk11

[JasonFengJ9]

@DanHeidinga the sanity build failure is a compilation error objs/compiler/z/runtime/Math.o.s:36: Error: junk at end of line, first unrecognized character is % which can't be caused by this PR.
I don't see such error in previous build, maybe try another build?

[DanHeidinga]

Jenkins test sanity zlinux,xlinux jdk11

[JasonFengJ9]

Same failure again, it also occurred in another PR build https://ci.eclipse.org/openj9/job/PullRequest-Sanity-JDK11-linux_390-64_cmprssptrs-OpenJ9/131/console, created #4206.
It is not related to this PR.

[JasonFengJ9]

PullRequest-Sanity-JDK11-linux_x86-64_cmprssptrs-OpenJ9 JarExtensionsTests.BuildInitial2 probably is a machine issue, and is not related this PR.
#4204 is now merged and the compilation seems fine at https://ci.eclipse.org/openj9/job/PullRequest-Sanity-JDK11-linux_390-64_cmprssptrs-OpenJ9/132/.

@DanHeidinga could you start another PR build?

[pshipton]

@JasonFengJ9 you need to rebase first.

[JasonFengJ9]

@pshipton yeah, that's right, done.

[pshipton]

Jenkins test sanity zlinux,xlinux jdk11