Fix access to J9ModronThreadLocalHeap.realHeapAlloc in older core files

[keithc-ca]

This updates the code so it does not rely on the existence of the method realHeapAlloc() when examining older core files.

Fixes: #9944.

[pshipton]

Does this fix the _realHeapTopOffset_ problem? It appears a CorruptDataException could still occur.

[pshipton]

@dmitripivkine

[keithc-ca]

Does this fix the _realHeapTopOffset_ problem? It appears a CorruptDataException could still occur.

Yes, it should fix that problem except for a small window between the merges of #9745 and #9813 where core files don't contain the required algorithm version update.

[pshipton]

I don't understand how, since nothing in this stack is modified.

NoSuchFieldError: com/ibm/j9ddr/vm29/structure/J9ModronThreadLocalHeap._realHeapTopOffset_
	at com.ibm.j9ddr.vm29.pointer.generated.J9ModronThreadLocalHeapPointer.realHeapTop(J9ModronThreadLocalHeapPointer.java:217)
	at com.ibm.j9ddr.vm29.j9.gc.GCObjectHeapIteratorAddressOrderedList_V1.<init>(GCObjectHeapIteratorAddressOrderedList_V1.java:78)
	at com.ibm.j9ddr.vm29.j9.gc.GCHeapRegionDescriptor_V1.objectIterator(GCHeapRegionDescriptor_V1.java:195)
	at com.ibm.j9ddr.vm29.j9.gc.GCObjectHeapIterator.fromHeapRegionDescriptor(GCObjectHeapIterator.java:51)
	at com.ibm.j9ddr.vm29.j9.walkers.HeapWalker.<init>(HeapWalker.java:83)

jenkins compile xlinux jdk11

[pshipton]

Ok I see. I was running an older JVM rather than the latest like I thought.

[pshipton]

lgtm. We should also put this in our release streams. Please create a PR for 0.21.

[pshipton]

@LinHu2016 fyi this change.

[keithc-ca]

lgtm. We should also put this in our release streams. Please create a PR for 0.21.

See #9959.

[pshipton]

jenkins test sanity,extended zlinux jdk11