On Wed, 6 Jun 2001, Greg Stein wrote: > The problem can be a little bit harder. Note that the two pools could be on > disjoint branches of some other pool.
Yes, but you can treat that as no different from any other "not an ancestor" case: > P1 > / \ > P2 P3 As you sort of hinted at, P2 and P3 are *not* guaranteed to have the same lifetime at all. Therefore, if P3 is the one being used by the MMAP (for example), and setaside(P2) is called, you have to just move the stuff over to P2 as you said. If and only if setaside is called on P2 or an ancestor of P2 will the setaside say "I'm already fine, no need to move the data". --Cliff -------------------------------------------------------------- Cliff Woolley [EMAIL PROTECTED] Charlottesville, VA
