Re: [PATCH v6 3/6] mm/workingset: extend the workingset detection for anon LRU

[Joonsoo Kim]

2020ë 7ì 2ì (ë) ìì 6:25, Vlastimil Babka <vbabka@xxxxxxx>ëì ìì:
>
> On 6/17/20 7:26 AM, js1304@xxxxxxxxx wrote:
> > From: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx>
>
> Hi,
>
> I would adjust the subject, as it sounds like the patch does the whole
> workingset detection, not just preparation.
> How about:
>
> mm/workingset: prepare the workingset infrastructure for anon LRU

Looks good. I will use it.

> > In the following patch, workingset detection will be applied to
> > anonymous LRU. To prepare it, this patch adds some code to
> > distinguish/handle the both LRUs.
>
> How about:
> To prepare for this, this patch splits workingset event counters for refault,
> activate and restore into anon and file variants, as well as the refaults
> counter in struct lruvec.

Will do.

> > v6: do not introduce a new nonresident_age for anon LRU since
> > we need to use *unified* nonresident_age to implement workingset
> > detection for anon LRU.
>
> Again, v6 update info shouldn't go to changelog. In this case I think it doesn't
> need mentioning at all, at least not in this patch.

Okay. I agree that this should not be included in the changelog. I just want
to notice someone who checked previous patches about that there is an
important change in this version.

> > Acked-by: Johannes Weiner <hannes@xxxxxxxxxxx>
> > Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx>
>
> Acked-by: Vlastimil Babka <vbabka@xxxxxxx>

Thanks.