On Fri 25-09-20 19:49:12, zhongjiang-ali wrote:
> After appling the series patches(mm: fix page aging across multiple cgroups),
> cgroup memory reclaim strategy is based on reclaim root's inactive:active
> ratio. if the target lruvec need to deactivate, its children cgroup also will
> deactivate. That will result in hot page to be reclaimed and other cgroup's
> cold page will be left, which is not expected.
>
> The patch will not force deactivate when inactive_is_low is not true unless
> we has scanned the inactive list and memory is unable to reclaim.
Do you have any data to present?
> Signed-off-by: zhongjiang-ali <zhongjiang-ali@xxxxxxxxxxxxxxxxx>
> ---
> mm/vmscan.c | 15 ++++++++++++++-
> 1 file changed, 14 insertions(+), 1 deletion(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 466fc31..77d395f 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -2407,8 +2407,21 @@ static void get_scan_count(struct lruvec *lruvec, struct scan_control *sc,
> case SCAN_FILE:
> case SCAN_ANON:
> /* Scan one type exclusively */
> - if ((scan_balance == SCAN_FILE) != file)
> + if ((scan_balance == SCAN_FILE) != file) {
> scan = 0;
> + } else {
> + /*
> + * Reclaim memory is based on the root's inactive: active
> + * ratio, but it is possible that silbing cgroup has a lot
> + * of cold memory to reclaim rather than reclaim the hot
> + * cache in the current cgroup.
> + */
> + if (!sc->force_deactivate && is_active_lru(lru) &&
> + !inactive_is_low(lruvec, lru - LRU_ACTIVE)) {
> + if (sc->may_deactivate & (1 << file))
> + scan = 0;
> + }
> + }
> break;
> default:
> /* Look ma, no brain */
> --
> 1.8.3.1
>
--
Michal Hocko
SUSE Labs
