Should AOT method relocation after local AOT compilation be an InterruptibleOperation?

[cjjdespres]

Clang informs me that in this block of code, responsible for relocating a freshly-compiled AOT method:

openj9/runtime/compiler/control/CompilationThread.cpp

Lines 10417 to 10432 in 110ef95

	TR::CompilationInfoPerThreadBase::InterruptibleOperation(*entry->_compInfoPT);
	// need to get a non-shared cache VM to relocate
	TR_J9VMBase *fe = TR_J9VMBase::get(jitConfig, vmThread);
	TR_ResolvedMethod *compilee = fe->createResolvedMethod(comp->trMemory(), (TR_OpaqueMethodBlock *)method);
	relocatedMetaData = entry->_compInfoPT->reloRuntime()->prepareRelocateAOTCodeAndData(
	vmThread,
	fe,
	aotMCCRuntimeCodeCache,
	cacheEntry,
	method,
	true,
	comp->getOptions(),
	comp,
	compilee
	);
	returnCode = entry->_compInfoPT->reloRuntime()->returnCode();

we create and then immediately destroy the TR::CompilationInfoPerThreadBase::InterruptibleOperation object that is created, because we don't name the object. This is probably an oversight. I think that relocation in other circumstances is considered to be interruptible - the other calls to prepareRelocateAOTCodeAndData (for local AOT loads, and for relocating methods received from the JITServer) appear to occur down the call stack from TR::CompilationInfoPerThreadBase::compile() while that function is holding InterruptibleOperation compilingMethodBody(*this).

Attn @mpirvu.

[cjjdespres]

Fixed by #20737.