Now, vm_page_busy_sleep() expects the page's object to be locked.
vm_object_unwire() does some unusual lazy locking of the object chain,
and keeps objects locked until a busy page is encountered or the loop
terminates. When a busy page is encountered, rather than unlocking all
but the "bottom-level" object, we must instead skip the object to which
tm belongs.
To provide a concrete example, suppose we are asked to unwire pages
in the range [0, 1] from object O, with backing object BO with offset 0.
Suppose no page is resident at 0 in O, but an xbusy page is resident at 1
in O. BO must contain a page at 0. vm_object_unwire() will lock
both O and BO to unwire the page at 0. It would then encounter the
busy page at 1 in O, and it must unlock BO, not O, before sleeping on
the page's busy state.