Fix keepRuntimeAlive for pthreads

[sbc100]

Fixes: #22889

[kripken]

Suggested change

	// However, for pthreads we always always need to allow the runtime to exit
	// However, for pthreads we always need to allow the runtime to exit.

[sbc100]

Maybe I was just wanting to be emphatic! :)

[kripken]

Heh, these double-positives/double-negatives always remind me of the joke,

A teacher told the class, "English has double negatives
that cancel out: 'I don't not want to go' means I want
to go, at least somewhat. But a double positive, unlike
in other languages, is basically the same as a positive."

A student at the back of the class scoffed, "Yeah, right."

[kripken]

Why do we need to always allow the runtime to exit with pthreads? Is that documented somewhere?

And, if so, should we do pthreads => EXIT_RUNTIME=1?

[sbc100]

Because pthreads themselves need to be able to exit. EXIT_RUNTIME=0 means the main thread never exits but it doesn't mean that pthreads can't exit.

Each pthread has its own "runtime" that can be kept alive and stops the thread from exiting. EXIT_RUNTIME has not effect (or should have no effect) on threads.

[sbc100]

I updated the comment.