Re: [PATCH v3 6/8] workingset: memcg: sleep when flushing stats in workingset_refault()

[Yosry Ahmed]

On Wed, Apr 5, 2023 at 1:00 AM Michal Koutný <mkoutny@xxxxxxxx> wrote:
>
> On Tue, Apr 04, 2023 at 11:09:02AM -0700, Yosry Ahmed <yosryahmed@xxxxxxxxxx> wrote:
> > IIUC there are multiple places where we can sleep in this path, we can
> > sleep waiting for a page to be read from disk, we can sleep during
> > allocating the page to read into, and IIUC the allocations on the
> > fault path can even run into reclaim, going into the vmscan code. So
> > there are precedents, but I am not sure if that's enough argument.
>
> I expect it'd depend on the proportion of the slow/fast paths.
> OK, let's see how it turns out in wider population.

Agreed. It also depends on the number of memcgs and how much the
periodic flusher is keeping up. I think no amount of testing will
cover all or even most workloads.

>
> Thanks,
> Michal