On Tue, Jul 30, 2024 at 11:46 AM Yosry Ahmed <yosryahmed@xxxxxxxxxx> wrote: > > > I still think the nr_deferred handling in the shrinker framework does > not seem compatible with zswap. Entries could be double counted, and > previously counted entries could go away during swapins. Unless I am > missing something, it seems like the scanning can be arbitrary, and > not truly proportional to the reclaim priority and zswap LRUs size. We can do some investigation by the side. I did find the trace point that collects deferred numbers, maybe that can illustrate the problem? It'd be an orthogonal issue though. I don't think fixing the nr_deferred part would be enough to handle the various cases Johannes pointed out.
