[PATCH] mm: vmscan : pgdemote vmstat is not getting updated when MGLRU is enabled.

[PATCH] mm: vmscan : pgdemote vmstat is not getting updated when MGLRU is enabled.

[Donet Tom]

When MGLRU is enabled, the pgdemote_kswapd, pgdemote_direct, and
pgdemote_khugepaged stats in vmstat are not being updated.

'Commit f77f0c751478 ("mm,memcg: provide per-cgroup counters for
NUMA balancing operations")' moved the pgdemote vmstat update
from demote_folio_list() to shrink_inactive_list(), which is in
the normal LRU path. As a result, the pgdemote stats are updated
correctly for the normal LRU but not for MGLRU.

To address this, we have added the pgdemote stat update in the
evict_folios() function, which is in the MGLRU path. With this
patch, the pgdemote stats will now be updated correctly when MGLRU
is enabled.

Without this patch vmstat output when MGLRU is enabled
======================================================
pgdemote_kswapd 0
pgdemote_direct 0
pgdemote_khugepaged 0

With this patch vmstat output when MGLRU is enabled
===================================================
pgdemote_kswapd 43234
pgdemote_direct 4691
pgdemote_khugepaged 0

Fixes: f77f0c751478 ("mm,memcg: provide per-cgroup counters for NUMA balancing operations")
Signed-off-by: Donet Tom <donettom@linux.ibm.com>
---
 mm/vmscan.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 9a859b7d18d7..b1ec5ece067e 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -4642,6 +4642,9 @@ static int evict_folios(struct lruvec *lruvec, struct scan_control *sc, int swap
 		reset_batch_size(walk);
 	}
 
+	__mod_lruvec_state(lruvec, PGDEMOTE_KSWAPD + reclaimer_offset(),
+					stat.nr_demoted);
+
 	item = PGSTEAL_KSWAPD + reclaimer_offset();
 	if (!cgroup_reclaim(sc))
 		__count_vm_events(item, reclaimed);
-- 
2.43.5

Re: [PATCH] mm: vmscan : pgdemote vmstat is not getting updated when MGLRU is enabled.

[Yu Zhao]

On Wed, Jan 8, 2025 at 11:05 PM Donet Tom <donettom@linux.ibm.com> wrote:
>
> When MGLRU is enabled, the pgdemote_kswapd, pgdemote_direct, and
> pgdemote_khugepaged stats in vmstat are not being updated.
>
> 'Commit f77f0c751478 ("mm,memcg: provide per-cgroup counters for
> NUMA balancing operations")' moved the pgdemote vmstat update
> from demote_folio_list() to shrink_inactive_list(), which is in
> the normal LRU path. As a result, the pgdemote stats are updated
> correctly for the normal LRU but not for MGLRU.
>
> To address this, we have added the pgdemote stat update in the
> evict_folios() function, which is in the MGLRU path. With this
> patch, the pgdemote stats will now be updated correctly when MGLRU
> is enabled.
>
> Without this patch vmstat output when MGLRU is enabled
> ======================================================
> pgdemote_kswapd 0
> pgdemote_direct 0
> pgdemote_khugepaged 0
>
> With this patch vmstat output when MGLRU is enabled
> ===================================================
> pgdemote_kswapd 43234
> pgdemote_direct 4691
> pgdemote_khugepaged 0
>
> Fixes: f77f0c751478 ("mm,memcg: provide per-cgroup counters for NUMA balancing operations")
> Signed-off-by: Donet Tom <donettom@linux.ibm.com>

Acked-by: Yu Zhao <yuzhao@google.com>

Re: [PATCH] mm: vmscan : pgdemote vmstat is not getting updated when MGLRU is enabled.

[Li Zhijian]

> When MGLRU is enabled, the pgdemote_kswapd, pgdemote_direct, and
> pgdemote_khugepaged stats in vmstat are not being updated.
>
> 'Commit f77f0c751478 ("mm,memcg: provide per-cgroup counters for
> NUMA balancing operations")' moved the pgdemote vmstat update
> from demote_folio_list() to shrink_inactive_list(), which is in
> the normal LRU path. As a result, the pgdemote stats are updated
> correctly for the normal LRU but not for MGLRU.
>
> To address this, we have added the pgdemote stat update in the
> evict_folios() function, which is in the MGLRU path. With this
> patch, the pgdemote stats will now be updated correctly when MGLRU
> is enabled.
>
> Without this patch vmstat output when MGLRU is enabled
> ======================================================
> pgdemote_kswapd 0
> pgdemote_direct 0
> pgdemote_khugepaged 0
>
> With this patch vmstat output when MGLRU is enabled
> ===================================================
> pgdemote_kswapd 43234
> pgdemote_direct 4691
> pgdemote_khugepaged 0
>
> Fixes: f77f0c751478 ("mm,memcg: provide per-cgroup counters for NUMA balancing operations")
> Signed-off-by: Donet Tom <donettom@linux.ibm.com>


Great minds think alike, It's excatly same with mine[0] :)

There is another bug[1] as well.
[0] 
https://lore.kernel.org/linux-mm/20250110122133.423481-2-lizhijian@fujitsu.com/ 
[1] 
https://lore.kernel.org/linux-mm/20250110122133.423481-1-lizhijian@fujitsu.com/ 

Feel free to add vvv if your patch is accepted at last.
Tested-by: Li Zhijian <lizhijian@fujitsu.com> Reviewed-by: Li Zhijian 
<lizhijian@fujitsu.com>


> ---
>   mm/vmscan.c  <https://lore.kernel.org/lkml/20250109060540.451261-1-donettom@linux.ibm.com/#Z31mm:vmscan.c>  | 3 +++
>   1 filechanged  <https://lore.kernel.org/lkml/20250109060540.451261-1-donettom@linux.ibm.com/#related>, 3 insertions(+)
>
> diff 
> <https://lore.kernel.org/lkml/20250109060540.451261-1-donettom@linux.ibm.com/#iZ31mm:vmscan.c> 
> --git a/mm/vmscan.c b/mm/vmscan.c index 9a859b7d18d7..b1ec5ece067e 
> 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -4642,6 +4642,9 @@ 
> static int evict_folios(struct lruvec *lruvec, struct scan_control 
> *sc, int swap   		reset_batch_size(walk);
>   	}
>   
> + __mod_lruvec_state(lruvec, PGDEMOTE_KSWAPD + reclaimer_offset(), + 
> stat.nr_demoted); +   	item = PGSTEAL_KSWAPD + reclaimer_offset();
>   	if (!cgroup_reclaim(sc))
>   		__count_vm_events(item, reclaimed);
> -- 

Re: [PATCH] mm: vmscan : pgdemote vmstat is not getting updated when MGLRU is enabled.

[Li Zhijian]

> When MGLRU is enabled, the pgdemote_kswapd, pgdemote_direct, and
> pgdemote_khugepaged stats in vmstat are not being updated.
> 'Commit f77f0c751478 ("mm,memcg: provide per-cgroup counters for
> NUMA balancing operations")' moved the pgdemote vmstat update
> from demote_folio_list() to shrink_inactive_list(), which is in
> the normal LRU path. As a result, the pgdemote stats are updated
> correctly for the normal LRU but not for MGLRU.
> To address this, we have added the pgdemote stat update in the
> evict_folios() function, which is in the MGLRU path. With this
> patch, the pgdemote stats will now be updated correctly when MGLRU
> is enabled.
> Without this patch vmstat output when MGLRU is enabled
> ======================================================
> pgdemote_kswapd 0
> pgdemote_direct 0
> pgdemote_khugepaged 0
> With this patch vmstat output when MGLRU is enabled
> ===================================================
> pgdemote_kswapd 43234
> pgdemote_direct 4691
> pgdemote_khugepaged 0
> Fixes: f77f0c751478 ("mm,memcg: provide per-cgroup counters for NUMA balancing operations")
> Signed-off-by: Donet Tom <donettom@linux.ibm.com>

Great minds think alike, It's excatly same with mine[0] :)

There is another bug[1] as well.

[0] https://lore.kernel.org/linux-mm/20250110122133.423481-2-lizhijian@fujitsu.com/

[1] https://lore.kernel.org/linux-mm/20250110122133.423481-1-lizhijian@fujitsu.com/

Feel free to add vvv if your patch is accepted at last.

Tested-by: Li Zhijian <lizhijian@fujitsu.com>

Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>

> ---
>    mm/vmscan.c  <https://lore.kernel.org/lkml/20250109060540.451261-1-donettom@linux.ibm.com/#Z31mm:vmscan.c>  | 3 +++
>    1 filechanged  <https://lore.kernel.org/lkml/20250109060540.451261-1-donettom@linux.ibm.com/#related>, 3 insertions(+)
> diff <https://lore.kernel.org/lkml/20250109060540.451261-1-donettom@linux.ibm.com/#iZ31mm:vmscan.c> --git a/mm/vmscan.c b/mm/vmscan.c index 9a859b7d18d7..b1ec5ece067e 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -4642,6 +4642,9 @@ static int evict_folios(struct lruvec *lruvec, struct scan_control *sc, int swap   		reset_batch_size(walk);
>    	}
>    + __mod_lruvec_state(lruvec, PGDEMOTE_KSWAPD + reclaimer_offset(), + stat.nr_demoted); +   	item = PGSTEAL_KSWAPD + reclaimer_offset();
>    	if (!cgroup_reclaim(sc))
>    		__count_vm_events(item, reclaimed);
> --