Re: [PATCH v2] memcontrol: only transfer the memcg data for migration

[Johannes Weiner]

On Wed, Oct 04, 2023 at 12:36:22PM -0700, Nhat Pham wrote:
> For most migration use cases, only transfer the memcg data from the old
> folio to the new folio, and clear the old folio's memcg data. No
> charging and uncharging will be done.
> 
> This shaves off some work on the migration path, and avoids the
> temporary double charging of a folio during its migration.
> 
> The only exception is replace_page_cache_folio(), which will use the old
> mem_cgroup_migrate() (now renamed to mem_cgroup_replace_folio). In that
> context, the isolation of the old page isn't quite as thorough as with
> migration, so we cannot use our new implementation directly.
> 
> This patch is the result of the following discussion on the new hugetlb
> memcg accounting behavior:
> 
> https://lore.kernel.org/lkml/20231003171329.GB314430@monkey/
> 
> This should be added as the second prep patch in the following series:
> https://lore.kernel.org/all/20231003001828.2554080-1-nphamcs@xxxxxxxxx/
> (hugetlb memcg accounting)
> 
> and should go right before the following patch:
> hugetlb: memcg: account hugetlb-backed memory in memory controller
> 
> Reported-by: Mike Kravetz <mike.kravetz@xxxxxxxxxx>
> Closes: https://lore.kernel.org/lkml/20231003171329.GB314430@monkey/

These two tags shouldn't be here, but in the fixlet instead. This is
the dependency patch. Otherwise looks good to me:

> Suggested-by: Johannes Weiner <hannes@xxxxxxxxxxx>
> Signed-off-by: Nhat Pham <nphamcs@xxxxxxxxx>

Acked-by: Johannes Weiner <hannes@xxxxxxxxxxx>