Re: [PATCH] mm: add Adaptive Memory Pressure Signaling (AMPRESS)

[Andre Ramos]

On Mon, Mar 02, 2026 at 02:00:00PM +0100, David Hildenbrand wrote:
  > Which value does exposing requested_kb really have if there are
  > multiply subscribers to notify? Doesn't quite make sense to even
  > expose that.

  You're right, that's a real design flaw. With N subscribers each
  independently acting on requested_kb, the aggregate reclaim is
  unbounded. The field would need to be either dropped or replaced
  with a hint that scales with subscriber count — neither of which
  was thought through here.

  > If apps have droppable caches, a better solution might be to use
  > something like MAP_DROPPABLE where possible, still leaving the
  > kernel in charge of when and how much memory to reclaim.

  That's a much cleaner approach. I'll look into MAP_DROPPABLE.
  Thank you for the pointer.

  André