A customer provided evidence indicating that a process
was stalled in direct reclaim:
- The process was trapped in throttle_direct_reclaim().
The function wait_event_killable() was called to wait condition
allow_direct_reclaim(pgdat) for current node to be true.
The allow_direct_reclaim(pgdat) examined the number of free pages
on the node by zone_page_state() which just returns value in
zone->vm_stat[NR_FREE_PAGES].
- On node #1, zone->vm_stat[NR_FREE_PAGES] was 0.
However, the freelist on this node was not empty.
- This inconsistent of vmstat value was caused by percpu vmstat on
nohz_full cpus. Every increment/decrement of vmstat is performed
on percpu vmstat counter at first, then pooled diffs are cumulated
to the zone's vmstat counter in timely manner. However, on nohz_full
cpus (in case of this customer's system, 48 of 52 cpus) these pooled
diffs were not cumulated once the cpu had no event on it so that
the cpu started sleeping infinitely.
I checked percpu vmstat and found there were total 69 counts not
cumulated to the zone's vmstat counter yet.
- In this situation, kswapd did not help the trapped process.
In pgdat_balanced(), zone_wakermark_ok_safe() examined the number
of free pages on the node by zone_page_state_snapshot() which
checks pending counts on percpu vmstat.
Therefore kswapd could know there were 69 free pages correctly.
Since zone->_watermark = {8, 20, 32}, kswapd did not work because
69 was greater than 32 as high watermark.
Change allow_direct_reclaim to use zone_page_state_snapshot, which
allows a more precise version of the vmstat counters to be used.
allow_direct_reclaim will only be called from try_to_free_pages,
which is not a hot path.
Suggested-by: Michal Hocko <[email protected]>
Signed-off-by: Marcelo Tosatti <[email protected]>
---
Index: linux-vmstat-remote/mm/vmscan.c
===================================================================
--- linux-vmstat-remote.orig/mm/vmscan.c
+++ linux-vmstat-remote/mm/vmscan.c
@@ -6886,7 +6886,7 @@ static bool allow_direct_reclaim(pg_data
continue;
pfmemalloc_reserve += min_wmark_pages(zone);
- free_pages += zone_page_state(zone, NR_FREE_PAGES);
+ free_pages += zone_page_state_snapshot(zone, NR_FREE_PAGES);
}
/* If there are no reserves (unexpected config) then do not throttle */
On Mon 15-05-23 15:00:16, Marcelo Tosatti wrote:
> A customer provided evidence indicating that a process
> was stalled in direct reclaim:
>
> - The process was trapped in throttle_direct_reclaim().
> The function wait_event_killable() was called to wait condition
> allow_direct_reclaim(pgdat) for current node to be true.
> The allow_direct_reclaim(pgdat) examined the number of free pages
> on the node by zone_page_state() which just returns value in
> zone->vm_stat[NR_FREE_PAGES].
>
> - On node #1, zone->vm_stat[NR_FREE_PAGES] was 0.
> However, the freelist on this node was not empty.
>
> - This inconsistent of vmstat value was caused by percpu vmstat on
> nohz_full cpus. Every increment/decrement of vmstat is performed
> on percpu vmstat counter at first, then pooled diffs are cumulated
> to the zone's vmstat counter in timely manner. However, on nohz_full
> cpus (in case of this customer's system, 48 of 52 cpus) these pooled
> diffs were not cumulated once the cpu had no event on it so that
> the cpu started sleeping infinitely.
> I checked percpu vmstat and found there were total 69 counts not
> cumulated to the zone's vmstat counter yet.
>
> - In this situation, kswapd did not help the trapped process.
> In pgdat_balanced(), zone_wakermark_ok_safe() examined the number
> of free pages on the node by zone_page_state_snapshot() which
> checks pending counts on percpu vmstat.
> Therefore kswapd could know there were 69 free pages correctly.
> Since zone->_watermark = {8, 20, 32}, kswapd did not work because
> 69 was greater than 32 as high watermark.
>
> Change allow_direct_reclaim to use zone_page_state_snapshot, which
> allows a more precise version of the vmstat counters to be used.
>
> allow_direct_reclaim will only be called from try_to_free_pages,
> which is not a hot path.
I have asked this patch to contain a note about testing done for future
reference IIRC. You have said that the above scenario is not easy to
reproduce and therefore not much of testing has been done but this is a
valuable information on its own.
> Suggested-by: Michal Hocko <[email protected]>
> Signed-off-by: Marcelo Tosatti <[email protected]>
I have also acked the patch. So let me repeat
Acked-by: Michal Hocko <[email protected]>
Btw. the patch should be routed separately as it is a potential fix.
>
> ---
>
> Index: linux-vmstat-remote/mm/vmscan.c
> ===================================================================
> --- linux-vmstat-remote.orig/mm/vmscan.c
> +++ linux-vmstat-remote/mm/vmscan.c
> @@ -6886,7 +6886,7 @@ static bool allow_direct_reclaim(pg_data
> continue;
>
> pfmemalloc_reserve += min_wmark_pages(zone);
> - free_pages += zone_page_state(zone, NR_FREE_PAGES);
> + free_pages += zone_page_state_snapshot(zone, NR_FREE_PAGES);
> }
>
> /* If there are no reserves (unexpected config) then do not throttle */
>
--
Michal Hocko
SUSE Labs
On Mon, May 15, 2023 at 03:00:16PM -0300, Marcelo Tosatti wrote:
> A customer provided evidence indicating that a process
> was stalled in direct reclaim:
>
> - The process was trapped in throttle_direct_reclaim().
> The function wait_event_killable() was called to wait condition
> allow_direct_reclaim(pgdat) for current node to be true.
> The allow_direct_reclaim(pgdat) examined the number of free pages
> on the node by zone_page_state() which just returns value in
> zone->vm_stat[NR_FREE_PAGES].
>
> - On node #1, zone->vm_stat[NR_FREE_PAGES] was 0.
> However, the freelist on this node was not empty.
>
> - This inconsistent of vmstat value was caused by percpu vmstat on
> nohz_full cpus. Every increment/decrement of vmstat is performed
> on percpu vmstat counter at first, then pooled diffs are cumulated
> to the zone's vmstat counter in timely manner. However, on nohz_full
> cpus (in case of this customer's system, 48 of 52 cpus) these pooled
> diffs were not cumulated once the cpu had no event on it so that
> the cpu started sleeping infinitely.
> I checked percpu vmstat and found there were total 69 counts not
> cumulated to the zone's vmstat counter yet.
>
> - In this situation, kswapd did not help the trapped process.
> In pgdat_balanced(), zone_wakermark_ok_safe() examined the number
> of free pages on the node by zone_page_state_snapshot() which
> checks pending counts on percpu vmstat.
> Therefore kswapd could know there were 69 free pages correctly.
> Since zone->_watermark = {8, 20, 32}, kswapd did not work because
> 69 was greater than 32 as high watermark.
>
> Change allow_direct_reclaim to use zone_page_state_snapshot, which
> allows a more precise version of the vmstat counters to be used.
>
> allow_direct_reclaim will only be called from try_to_free_pages,
> which is not a hot path.
>
> Suggested-by: Michal Hocko <[email protected]>
> Signed-off-by: Marcelo Tosatti <[email protected]>
>
> ---
>
> Index: linux-vmstat-remote/mm/vmscan.c
> ===================================================================
> --- linux-vmstat-remote.orig/mm/vmscan.c
> +++ linux-vmstat-remote/mm/vmscan.c
> @@ -6886,7 +6886,7 @@ static bool allow_direct_reclaim(pg_data
> continue;
>
> pfmemalloc_reserve += min_wmark_pages(zone);
> - free_pages += zone_page_state(zone, NR_FREE_PAGES);
> + free_pages += zone_page_state_snapshot(zone, NR_FREE_PAGES);
> }
>
> /* If there are no reserves (unexpected config) then do not throttle */
>
>
Reviewed-by: Aaron Tomlin <[email protected]>
--
Aaron Tomlin