On Tue, Jun 20, 2023 at 8:30 PM Yosry Ahmed <yosryahmed@xxxxxxxxxx> wrote:
>
> Evidently, global_reclaim() can be a confusing name. Especially that it
> used to exist before with a subtly different definition (removed by
> commit b5ead35e7e1d ("mm: vmscan: naming fixes: global_reclaim() and
> sane_reclaim()"). It can be interpreted as non-cgroup reclaim, even
> though it returns true for cgroup reclaim on the root memcg
> (through memory.reclaim).
>
> Rename it to root_reclaim() in an attempt to make it less ambiguous, and
> add documentation to it as well as cgroup_reclaim.
>
> Reported-by: Johannes Weiner <hannes@xxxxxxxxxxx>
> Closes: https://lore.kernel.org/lkml/20230405200150.GA35884@xxxxxxxxxxx/
> Signed-off-by: Yosry Ahmed <yosryahmed@xxxxxxxxxx>
Acked-by: Yu Zhao <yuzhao@xxxxxxxxxx>
