2021-08-18 15:24:44

by Johannes Weiner

[permalink] [raw]
Subject: [PATCH] mm: vmscan: fix missing psi annotation for node_reclaim()

In a debugging session the other day, Rik noticed that node_reclaim()
was missing memstall annotations. This means we'll miss pressure and
lost productivity resulting from reclaim on an overloaded local NUMA
node when vm.zone_reclaim_mode is enabled.

There haven't been any reports, but that's likely because
vm.zone_reclaim_mode hasn't been a commonly used feature recently, and
the intersection between such setups and psi users is probably
nil. Although, secondary memory such as CXL-connected DIMMS,
persistent memory etc. and the page demotion patches that handle them
(https://lore.kernel.org/lkml/[email protected]/)
could soon make this a more common codepath again.

Reported-by: Rik van Riel <[email protected]>
Signed-off-by: Johannes Weiner <[email protected]>
---
mm/vmscan.c | 3 +++
1 file changed, 3 insertions(+)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 701106e1829c..8ec4412c6116 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -4424,11 +4424,13 @@ static int __node_reclaim(struct pglist_data *pgdat, gfp_t gfp_mask, unsigned in
.may_swap = 1,
.reclaim_idx = gfp_zone(gfp_mask),
};
+ unsigned long pflags;

trace_mm_vmscan_node_reclaim_begin(pgdat->node_id, order,
sc.gfp_mask);

cond_resched();
+ psi_memstall_enter(&pflags);
fs_reclaim_acquire(sc.gfp_mask);
/*
* We need to be able to allocate from the reserves for RECLAIM_UNMAP
@@ -4453,6 +4455,7 @@ static int __node_reclaim(struct pglist_data *pgdat, gfp_t gfp_mask, unsigned in
current->flags &= ~PF_SWAPWRITE;
memalloc_noreclaim_restore(noreclaim_flag);
fs_reclaim_release(sc.gfp_mask);
+ psi_memstall_leave(&pflags);

trace_mm_vmscan_node_reclaim_end(sc.nr_reclaimed);

--
2.32.0


2021-08-19 16:04:28

by Shakeel Butt

[permalink] [raw]
Subject: Re: [PATCH] mm: vmscan: fix missing psi annotation for node_reclaim()

On Wed, Aug 18, 2021 at 8:26 AM Johannes Weiner <[email protected]> wrote:
>
> In a debugging session the other day, Rik noticed that node_reclaim()
> was missing memstall annotations. This means we'll miss pressure and
> lost productivity resulting from reclaim on an overloaded local NUMA
> node when vm.zone_reclaim_mode is enabled.
>
> There haven't been any reports, but that's likely because
> vm.zone_reclaim_mode hasn't been a commonly used feature recently, and
> the intersection between such setups and psi users is probably
> nil. Although, secondary memory such as CXL-connected DIMMS,
> persistent memory etc. and the page demotion patches that handle them
> (https://lore.kernel.org/lkml/[email protected]/)
> could soon make this a more common codepath again.
>
> Reported-by: Rik van Riel <[email protected]>
> Signed-off-by: Johannes Weiner <[email protected]>

Reviewed-by: Shakeel Butt <[email protected]>