diff mbox series

[v2] mm: vmscan: reclaim anon pages if there are swapcache pages

Message ID 20230822024901.2412520-1-liushixin2@huawei.com
State New
Headers
Series [v2] mm: vmscan: reclaim anon pages if there are swapcache pages |

Commit Message

Liu Shixin Aug. 22, 2023, 2:49 a.m. UTC 
  When spaces of swap devices are exhausted, only file pages can be reclaimed.
But there are still some swapcache pages in anon lru list. This can lead
to a premature out-of-memory.

This problem can be fixed by checking number of swapcache pages in
can_reclaim_anon_pages(). For memcg v2, there are swapcache stat that can
be used directly. For memcg v1, use total_swapcache_pages() instead, which
may not accurate but can solve the problem.

Signed-off-by: Liu Shixin <liushixin2@huawei.com>
---
 include/linux/swap.h |  6 ++++++
 mm/memcontrol.c      |  8 ++++++++
 mm/vmscan.c          | 12 ++++++++----
 3 files changed, 22 insertions(+), 4 deletions(-)

Comments

Liu Shixin Aug. 23, 2023, 2 a.m. UTC | #1 
On 2023/8/23 0:35, Yosry Ahmed wrote:
> On Mon, Aug 21, 2023 at 6:54 PM Liu Shixin <liushixin2@huawei.com> wrote:
>> When spaces of swap devices are exhausted, only file pages can be reclaimed.
>> But there are still some swapcache pages in anon lru list. This can lead
>> to a premature out-of-memory.
>>
>> This problem can be fixed by checking number of swapcache pages in
>> can_reclaim_anon_pages(). For memcg v2, there are swapcache stat that can
>> be used directly. For memcg v1, use total_swapcache_pages() instead, which
>> may not accurate but can solve the problem.
> Interesting find. I wonder if we really don't have any handling of
> this situation.
I have alreadly test this problem and can confirm that it is a real problem.
With 9MB swap space and 10MB mem_cgroup limit，when allocate 15MB memory,
there is a probability that OOM occurs.
>
>> Signed-off-by: Liu Shixin <liushixin2@huawei.com>
>> ---
>>  include/linux/swap.h |  6 ++++++
>>  mm/memcontrol.c      |  8 ++++++++
>>  mm/vmscan.c          | 12 ++++++++----
>>  3 files changed, 22 insertions(+), 4 deletions(-)
>>
>> diff --git a/include/linux/swap.h b/include/linux/swap.h
>> index 456546443f1f..0318e918bfa4 100644
>> --- a/include/linux/swap.h
>> +++ b/include/linux/swap.h
>> @@ -669,6 +669,7 @@ static inline void mem_cgroup_uncharge_swap(swp_entry_t entry, unsigned int nr_p
>>  }
>>
>>  extern long mem_cgroup_get_nr_swap_pages(struct mem_cgroup *memcg);
>> +extern long mem_cgroup_get_nr_swapcache_pages(struct mem_cgroup *memcg);
>>  extern bool mem_cgroup_swap_full(struct folio *folio);
>>  #else
>>  static inline void mem_cgroup_swapout(struct folio *folio, swp_entry_t entry)
>> @@ -691,6 +692,11 @@ static inline long mem_cgroup_get_nr_swap_pages(struct mem_cgroup *memcg)
>>         return get_nr_swap_pages();
>>  }
>>
>> +static inline long mem_cgroup_get_nr_swapcache_pages(struct mem_cgroup *memcg)
>> +{
>> +       return total_swapcache_pages();
>> +}
>> +
>>  static inline bool mem_cgroup_swap_full(struct folio *folio)
>>  {
>>         return vm_swap_full();
>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
>> index e8ca4bdcb03c..3e578f41023e 100644
>> --- a/mm/memcontrol.c
>> +++ b/mm/memcontrol.c
>> @@ -7567,6 +7567,14 @@ long mem_cgroup_get_nr_swap_pages(struct mem_cgroup *memcg)
>>         return nr_swap_pages;
>>  }
>>
>> +long mem_cgroup_get_nr_swapcache_pages(struct mem_cgroup *memcg)
>> +{
>> +       if (mem_cgroup_disabled() || do_memsw_account())
>> +               return total_swapcache_pages();
>> +
>> +       return memcg_page_state(memcg, NR_SWAPCACHE);
>> +}
> Is there a reason why we cannot use NR_SWAPCACHE for cgroup v1? Isn't
> that being maintained regardless of cgroup version? It is not exposed
> in cgroup v1's memory.stat, but I don't think there is a reason we
> can't do that -- if only to document that it is being used with cgroup
> v1.
Thanks for your advice, it is more appropriate to use NR_SWAPCACH.
>
>
>> +
>>  bool mem_cgroup_swap_full(struct folio *folio)
>>  {
>>         struct mem_cgroup *memcg;
>> diff --git a/mm/vmscan.c b/mm/vmscan.c
>> index 7c33c5b653ef..bcb6279cbae7 100644
>> --- a/mm/vmscan.c
>> +++ b/mm/vmscan.c
>> @@ -609,13 +609,17 @@ static inline bool can_reclaim_anon_pages(struct mem_cgroup *memcg,
>>         if (memcg == NULL) {
>>                 /*
>>                  * For non-memcg reclaim, is there
>> -                * space in any swap device?
>> +                * space in any swap device or swapcache pages?
>>                  */
>> -               if (get_nr_swap_pages() > 0)
>> +               if (get_nr_swap_pages() + total_swapcache_pages() > 0)
>>                         return true;
>>         } else {
>> -               /* Is the memcg below its swap limit? */
>> -               if (mem_cgroup_get_nr_swap_pages(memcg) > 0)
>> +               /*
>> +                * Is the memcg below its swap limit or is there swapcache
>> +                * pages can be freed?
>> +                */
>> +               if (mem_cgroup_get_nr_swap_pages(memcg) +
>> +                   mem_cgroup_get_nr_swapcache_pages(memcg) > 0)
>>                         return true;
>>         }
> I wonder if it would be more efficient to set a bit in struct
> scan_control if we only are out of swap spaces but have swap cache
> pages, and only isolate anon pages that are in the swap cache, instead
> of isolating random anon pages. We may end up isolating pages that are
> not in the swap cache for a few iterations and wasting cycles.
Good idea. Thanks.
>
>> --
>> 2.25.1
>>
> .
>
diff mbox series

Patch

diff --git a/include/linux/swap.h b/include/linux/swap.h
index 456546443f1f..0318e918bfa4 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -669,6 +669,7 @@  static inline void mem_cgroup_uncharge_swap(swp_entry_t entry, unsigned int nr_p
 }
 
 extern long mem_cgroup_get_nr_swap_pages(struct mem_cgroup *memcg);
+extern long mem_cgroup_get_nr_swapcache_pages(struct mem_cgroup *memcg);
 extern bool mem_cgroup_swap_full(struct folio *folio);
 #else
 static inline void mem_cgroup_swapout(struct folio *folio, swp_entry_t entry)
@@ -691,6 +692,11 @@  static inline long mem_cgroup_get_nr_swap_pages(struct mem_cgroup *memcg)
 	return get_nr_swap_pages();
 }
 
+static inline long mem_cgroup_get_nr_swapcache_pages(struct mem_cgroup *memcg)
+{
+	return total_swapcache_pages();
+}
+
 static inline bool mem_cgroup_swap_full(struct folio *folio)
 {
 	return vm_swap_full();
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index e8ca4bdcb03c..3e578f41023e 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -7567,6 +7567,14 @@  long mem_cgroup_get_nr_swap_pages(struct mem_cgroup *memcg)
 	return nr_swap_pages;
 }
 
+long mem_cgroup_get_nr_swapcache_pages(struct mem_cgroup *memcg)
+{
+	if (mem_cgroup_disabled() || do_memsw_account())
+		return total_swapcache_pages();
+
+	return memcg_page_state(memcg, NR_SWAPCACHE);
+}
+
 bool mem_cgroup_swap_full(struct folio *folio)
 {
 	struct mem_cgroup *memcg;
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 7c33c5b653ef..bcb6279cbae7 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -609,13 +609,17 @@  static inline bool can_reclaim_anon_pages(struct mem_cgroup *memcg,
 	if (memcg == NULL) {
 		/*
 		 * For non-memcg reclaim, is there
-		 * space in any swap device?
+		 * space in any swap device or swapcache pages?
 		 */
-		if (get_nr_swap_pages() > 0)
+		if (get_nr_swap_pages() + total_swapcache_pages() > 0)
 			return true;
 	} else {
-		/* Is the memcg below its swap limit? */
-		if (mem_cgroup_get_nr_swap_pages(memcg) > 0)
+		/*
+		 * Is the memcg below its swap limit or is there swapcache
+		 * pages can be freed?
+		 */
+		if (mem_cgroup_get_nr_swap_pages(memcg) +
+		    mem_cgroup_get_nr_swapcache_pages(memcg) > 0)
 			return true;
 	}