Refactor rtree to be lock-free.

Recent huge allocation refactoring associates huge allocations with
arenas, but it remains necessary to quickly look up huge allocation
metadata during reallocation/deallocation.  A global radix tree remains
a good solution to this problem, but locking would have become the
primary bottleneck after (upcoming) migration of chunk management from
global to per arena data structures.

This lock-free implementation uses double-checked reads to traverse the
tree, so that in the steady state, each read or write requires only a
single atomic operation.

This implementation also assures that no more than two tree levels
actually exist, through a combination of careful virtual memory
allocation which makes large sparse nodes cheap, and skipping the root
node on x64 (possible because the top 16 bits are all 0 in practice).

include/jemalloc/internal/jemalloc_internal.h.in

@@ -955,7 +955,7 @@ ivsalloc(const void *ptr, bool demote)
 {
 
 	/* Return 0 if ptr is not within a chunk managed by jemalloc. */
-	if (rtree_get(chunks_rtree, (uintptr_t)CHUNK_ADDR2BASE(ptr)) == 0)
+	if (rtree_get(&chunks_rtree, (uintptr_t)CHUNK_ADDR2BASE(ptr)) == 0)
 		return (0);
 
 	return (isalloc(ptr, demote));