[PATCH V2 2/2] add stats to monitor soft_limit reclaim

[PATCH V2 2/2] add stats to monitor soft_limit reclaim

[Ying Han]

The stat is added:

/dev/cgroup/*/memory.stat
soft_steal:        - # of pages reclaimed from soft_limit hierarchical reclaim
total_soft_steal:  - # sum of all children's "soft_steal"

Change log v2...v1
1. removed the counting on number of skips on shrink_zone. This is due to the
change on the previous patch.

Signed-off-by: Ying Han <yinghan@google.com>
---
 Documentation/cgroups/memory.txt |    2 ++
 include/linux/memcontrol.h       |    5 +++++
 mm/memcontrol.c                  |   14 ++++++++++++++
 3 files changed, 21 insertions(+), 0 deletions(-)

diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt
index b6ed61c..dcda6c5 100644
--- a/Documentation/cgroups/memory.txt
+++ b/Documentation/cgroups/memory.txt
@@ -385,6 +385,7 @@ mapped_file	- # of bytes of mapped file (includes tmpfs/shmem)
 pgpgin		- # of pages paged in (equivalent to # of charging events).
 pgpgout		- # of pages paged out (equivalent to # of uncharging events).
 swap		- # of bytes of swap usage
+soft_steal	- # of pages reclaimed from global hierarchical reclaim
 inactive_anon	- # of bytes of anonymous memory and swap cache memory on
 		LRU list.
 active_anon	- # of bytes of anonymous and swap cache memory on active
@@ -406,6 +407,7 @@ total_mapped_file	- sum of all children's "cache"
 total_pgpgin		- sum of all children's "pgpgin"
 total_pgpgout		- sum of all children's "pgpgout"
 total_swap		- sum of all children's "swap"
+total_soft_steal	- sum of all children's "soft_steal"
 total_inactive_anon	- sum of all children's "inactive_anon"
 total_active_anon	- sum of all children's "active_anon"
 total_inactive_file	- sum of all children's "inactive_file"
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 01281ac..151ab40 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -115,6 +115,7 @@ struct zone_reclaim_stat*
 mem_cgroup_get_reclaim_stat_from_page(struct page *page);
 extern void mem_cgroup_print_oom_info(struct mem_cgroup *memcg,
 					struct task_struct *p);
+void mem_cgroup_soft_steal(struct mem_cgroup *memcg, int val);
 
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
 extern int do_swap_account;
@@ -356,6 +357,10 @@ static inline void mem_cgroup_split_huge_fixup(struct page *head,
 {
 }
 
+static inline void mem_cgroup_soft_steal(struct mem_cgroup *memcg,
+					 int val)
+{
+}
 #endif /* CONFIG_CGROUP_MEM_CONT */
 
 #if !defined(CONFIG_CGROUP_MEM_RES_CTLR) || !defined(CONFIG_DEBUG_VM)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 67fff28..5e4aa41 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -94,6 +94,8 @@ enum mem_cgroup_events_index {
 	MEM_CGROUP_EVENTS_PGPGIN,	/* # of pages paged in */
 	MEM_CGROUP_EVENTS_PGPGOUT,	/* # of pages paged out */
 	MEM_CGROUP_EVENTS_COUNT,	/* # of pages paged in/out */
+	MEM_CGROUP_EVENTS_SOFT_STEAL,	/* # of pages reclaimed from */
+					/* oft reclaim               */
 	MEM_CGROUP_EVENTS_NSTATS,
 };
 /*
@@ -624,6 +626,11 @@ static void mem_cgroup_charge_statistics(struct mem_cgroup *mem,
 	preempt_enable();
 }
 
+void mem_cgroup_soft_steal(struct mem_cgroup *mem, int val)
+{
+	this_cpu_add(mem->stat->events[MEM_CGROUP_EVENTS_SOFT_STEAL], val);
+}
+
 static unsigned long mem_cgroup_get_local_zonestat(struct mem_cgroup *mem,
 					enum lru_list idx)
 {
@@ -3326,6 +3333,9 @@ unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order,
 						&nr_scanned);
 		nr_reclaimed += reclaimed;
 		*total_scanned += nr_scanned;
+
+		mem_cgroup_soft_steal(mz->mem, reclaimed);
+
 		spin_lock(&mctz->lock);
 
 		/*
@@ -3783,6 +3793,7 @@ enum {
 	MCS_PGPGIN,
 	MCS_PGPGOUT,
 	MCS_SWAP,
+	MCS_SOFT_STEAL,
 	MCS_INACTIVE_ANON,
 	MCS_ACTIVE_ANON,
 	MCS_INACTIVE_FILE,
@@ -3805,6 +3816,7 @@ struct {
 	{"pgpgin", "total_pgpgin"},
 	{"pgpgout", "total_pgpgout"},
 	{"swap", "total_swap"},
+	{"soft_steal", "total_soft_steal"},
 	{"inactive_anon", "total_inactive_anon"},
 	{"active_anon", "total_active_anon"},
 	{"inactive_file", "total_inactive_file"},
@@ -3833,6 +3845,8 @@ mem_cgroup_get_local_stat(struct mem_cgroup *mem, struct mcs_total_stat *s)
 		val = mem_cgroup_read_stat(mem, MEM_CGROUP_STAT_SWAPOUT);
 		s->stat[MCS_SWAP] += val * PAGE_SIZE;
 	}
+	val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_SOFT_STEAL);
+	s->stat[MCS_SOFT_STEAL] += val;
 
 	/* per zone stat */
 	val = mem_cgroup_get_local_zonestat(mem, LRU_INACTIVE_ANON);
-- 
1.7.3.1

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH V2 2/2] add stats to monitor soft_limit reclaim

[KAMEZAWA Hiroyuki]

On Mon, 28 Mar 2011 16:51:10 -0700
Ying Han <yinghan@google.com> wrote:

> The stat is added:
> 
> /dev/cgroup/*/memory.stat
> soft_steal:        - # of pages reclaimed from soft_limit hierarchical reclaim
> total_soft_steal:  - # sum of all children's "soft_steal"
> 
> Change log v2...v1
> 1. removed the counting on number of skips on shrink_zone. This is due to the
> change on the previous patch.
> 
> Signed-off-by: Ying Han <yinghan@google.com>

Hmm...


> ---
>  Documentation/cgroups/memory.txt |    2 ++
>  include/linux/memcontrol.h       |    5 +++++
>  mm/memcontrol.c                  |   14 ++++++++++++++
>  3 files changed, 21 insertions(+), 0 deletions(-)
> 
> diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt
> index b6ed61c..dcda6c5 100644
> --- a/Documentation/cgroups/memory.txt
> +++ b/Documentation/cgroups/memory.txt
> @@ -385,6 +385,7 @@ mapped_file	- # of bytes of mapped file (includes tmpfs/shmem)
>  pgpgin		- # of pages paged in (equivalent to # of charging events).
>  pgpgout		- # of pages paged out (equivalent to # of uncharging events).
>  swap		- # of bytes of swap usage
> +soft_steal	- # of pages reclaimed from global hierarchical reclaim
>  inactive_anon	- # of bytes of anonymous memory and swap cache memory on
>  		LRU list.
>  active_anon	- # of bytes of anonymous and swap cache memory on active
> @@ -406,6 +407,7 @@ total_mapped_file	- sum of all children's "cache"
>  total_pgpgin		- sum of all children's "pgpgin"
>  total_pgpgout		- sum of all children's "pgpgout"
>  total_swap		- sum of all children's "swap"
> +total_soft_steal	- sum of all children's "soft_steal"
>  total_inactive_anon	- sum of all children's "inactive_anon"
>  total_active_anon	- sum of all children's "active_anon"
>  total_inactive_file	- sum of all children's "inactive_file"
> diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
> index 01281ac..151ab40 100644
> --- a/include/linux/memcontrol.h
> +++ b/include/linux/memcontrol.h
> @@ -115,6 +115,7 @@ struct zone_reclaim_stat*
>  mem_cgroup_get_reclaim_stat_from_page(struct page *page);
>  extern void mem_cgroup_print_oom_info(struct mem_cgroup *memcg,
>  					struct task_struct *p);
> +void mem_cgroup_soft_steal(struct mem_cgroup *memcg, int val);
>  
>  #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
>  extern int do_swap_account;
> @@ -356,6 +357,10 @@ static inline void mem_cgroup_split_huge_fixup(struct page *head,
>  {
>  }
>  
> +static inline void mem_cgroup_soft_steal(struct mem_cgroup *memcg,
> +					 int val)
> +{
> +}
>  #endif /* CONFIG_CGROUP_MEM_CONT */
>  
>  #if !defined(CONFIG_CGROUP_MEM_RES_CTLR) || !defined(CONFIG_DEBUG_VM)
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 67fff28..5e4aa41 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -94,6 +94,8 @@ enum mem_cgroup_events_index {
>  	MEM_CGROUP_EVENTS_PGPGIN,	/* # of pages paged in */
>  	MEM_CGROUP_EVENTS_PGPGOUT,	/* # of pages paged out */
>  	MEM_CGROUP_EVENTS_COUNT,	/* # of pages paged in/out */
> +	MEM_CGROUP_EVENTS_SOFT_STEAL,	/* # of pages reclaimed from */
> +					/* oft reclaim               */
>  	MEM_CGROUP_EVENTS_NSTATS,
>  };
>  /*
> @@ -624,6 +626,11 @@ static void mem_cgroup_charge_statistics(struct mem_cgroup *mem,
>  	preempt_enable();
>  }
>  
> +void mem_cgroup_soft_steal(struct mem_cgroup *mem, int val)
> +{
> +	this_cpu_add(mem->stat->events[MEM_CGROUP_EVENTS_SOFT_STEAL], val);
> +}
> +
>  static unsigned long mem_cgroup_get_local_zonestat(struct mem_cgroup *mem,
>  					enum lru_list idx)
>  {
> @@ -3326,6 +3333,9 @@ unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order,
>  						&nr_scanned);
>  		nr_reclaimed += reclaimed;
>  		*total_scanned += nr_scanned;
> +
> +		mem_cgroup_soft_steal(mz->mem, reclaimed);
> +

Here, you add "the number of reclaimed pages from the all descendants under me".
Could you move this to mem_cgroup_hierarchical_reclaim() ? Then, you can report
the correct stats even with hierarchy enabled.

Even if the value is recorded into hierarchy, total_steal will show total.

BTW, soft_scan and soft_total_scan aren't necessary ?

Thanks,
-Kame

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH V2 2/2] add stats to monitor soft_limit reclaim

[Daisuke Nishimura]

On Mon, 28 Mar 2011 16:51:10 -0700
Ying Han <yinghan@google.com> wrote:

> The stat is added:
> 
> /dev/cgroup/*/memory.stat
> soft_steal:        - # of pages reclaimed from soft_limit hierarchical reclaim
> total_soft_steal:  - # sum of all children's "soft_steal"
> 
> Change log v2...v1
> 1. removed the counting on number of skips on shrink_zone. This is due to the
> change on the previous patch.
> 
> Signed-off-by: Ying Han <yinghan@google.com>
> ---
>  Documentation/cgroups/memory.txt |    2 ++
>  include/linux/memcontrol.h       |    5 +++++
>  mm/memcontrol.c                  |   14 ++++++++++++++
>  3 files changed, 21 insertions(+), 0 deletions(-)
> 
> diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt
> index b6ed61c..dcda6c5 100644
> --- a/Documentation/cgroups/memory.txt
> +++ b/Documentation/cgroups/memory.txt
> @@ -385,6 +385,7 @@ mapped_file	- # of bytes of mapped file (includes tmpfs/shmem)
>  pgpgin		- # of pages paged in (equivalent to # of charging events).
>  pgpgout		- # of pages paged out (equivalent to # of uncharging events).
>  swap		- # of bytes of swap usage
> +soft_steal	- # of pages reclaimed from global hierarchical reclaim
>  inactive_anon	- # of bytes of anonymous memory and swap cache memory on
>  		LRU list.
>  active_anon	- # of bytes of anonymous and swap cache memory on active
> @@ -406,6 +407,7 @@ total_mapped_file	- sum of all children's "cache"
>  total_pgpgin		- sum of all children's "pgpgin"
>  total_pgpgout		- sum of all children's "pgpgout"
>  total_swap		- sum of all children's "swap"
> +total_soft_steal	- sum of all children's "soft_steal"
>  total_inactive_anon	- sum of all children's "inactive_anon"
>  total_active_anon	- sum of all children's "active_anon"
>  total_inactive_file	- sum of all children's "inactive_file"
> diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
> index 01281ac..151ab40 100644
> --- a/include/linux/memcontrol.h
> +++ b/include/linux/memcontrol.h
> @@ -115,6 +115,7 @@ struct zone_reclaim_stat*
>  mem_cgroup_get_reclaim_stat_from_page(struct page *page);
>  extern void mem_cgroup_print_oom_info(struct mem_cgroup *memcg,
>  					struct task_struct *p);
> +void mem_cgroup_soft_steal(struct mem_cgroup *memcg, int val);
>  
>  #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
>  extern int do_swap_account;
> @@ -356,6 +357,10 @@ static inline void mem_cgroup_split_huge_fixup(struct page *head,
>  {
>  }
>  
> +static inline void mem_cgroup_soft_steal(struct mem_cgroup *memcg,
> +					 int val)
> +{
> +}
>  #endif /* CONFIG_CGROUP_MEM_CONT */
>  
Do you use this function outside of memcontrol.c in future, right ?
I'm asking just for clarification, and I'm sorry if I miss some past discussions.

Thanks,
Daisuke Nishimura.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH V2 2/2] add stats to monitor soft_limit reclaim

[Ying Han]

On Mon, Mar 28, 2011 at 6:22 PM, KAMEZAWA Hiroyuki
<kamezawa.hiroyu@jp.fujitsu.com> wrote:
> On Mon, 28 Mar 2011 16:51:10 -0700
> Ying Han <yinghan@google.com> wrote:
>
>> The stat is added:
>>
>> /dev/cgroup/*/memory.stat
>> soft_steal:        - # of pages reclaimed from soft_limit hierarchical reclaim
>> total_soft_steal:  - # sum of all children's "soft_steal"
>>
>> Change log v2...v1
>> 1. removed the counting on number of skips on shrink_zone. This is due to the
>> change on the previous patch.
>>
>> Signed-off-by: Ying Han <yinghan@google.com>
>
> Hmm...
>
>
>> ---
>>  Documentation/cgroups/memory.txt |    2 ++
>>  include/linux/memcontrol.h       |    5 +++++
>>  mm/memcontrol.c                  |   14 ++++++++++++++
>>  3 files changed, 21 insertions(+), 0 deletions(-)
>>
>> diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt
>> index b6ed61c..dcda6c5 100644
>> --- a/Documentation/cgroups/memory.txt
>> +++ b/Documentation/cgroups/memory.txt
>> @@ -385,6 +385,7 @@ mapped_file       - # of bytes of mapped file (includes tmpfs/shmem)
>>  pgpgin               - # of pages paged in (equivalent to # of charging events).
>>  pgpgout              - # of pages paged out (equivalent to # of uncharging events).
>>  swap         - # of bytes of swap usage
>> +soft_steal   - # of pages reclaimed from global hierarchical reclaim
>>  inactive_anon        - # of bytes of anonymous memory and swap cache memory on
>>               LRU list.
>>  active_anon  - # of bytes of anonymous and swap cache memory on active
>> @@ -406,6 +407,7 @@ total_mapped_file - sum of all children's "cache"
>>  total_pgpgin         - sum of all children's "pgpgin"
>>  total_pgpgout                - sum of all children's "pgpgout"
>>  total_swap           - sum of all children's "swap"
>> +total_soft_steal     - sum of all children's "soft_steal"
>>  total_inactive_anon  - sum of all children's "inactive_anon"
>>  total_active_anon    - sum of all children's "active_anon"
>>  total_inactive_file  - sum of all children's "inactive_file"
>> diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
>> index 01281ac..151ab40 100644
>> --- a/include/linux/memcontrol.h
>> +++ b/include/linux/memcontrol.h
>> @@ -115,6 +115,7 @@ struct zone_reclaim_stat*
>>  mem_cgroup_get_reclaim_stat_from_page(struct page *page);
>>  extern void mem_cgroup_print_oom_info(struct mem_cgroup *memcg,
>>                                       struct task_struct *p);
>> +void mem_cgroup_soft_steal(struct mem_cgroup *memcg, int val);
>>
>>  #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
>>  extern int do_swap_account;
>> @@ -356,6 +357,10 @@ static inline void mem_cgroup_split_huge_fixup(struct page *head,
>>  {
>>  }
>>
>> +static inline void mem_cgroup_soft_steal(struct mem_cgroup *memcg,
>> +                                      int val)
>> +{
>> +}
>>  #endif /* CONFIG_CGROUP_MEM_CONT */
>>
>>  #if !defined(CONFIG_CGROUP_MEM_RES_CTLR) || !defined(CONFIG_DEBUG_VM)
>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
>> index 67fff28..5e4aa41 100644
>> --- a/mm/memcontrol.c
>> +++ b/mm/memcontrol.c
>> @@ -94,6 +94,8 @@ enum mem_cgroup_events_index {
>>       MEM_CGROUP_EVENTS_PGPGIN,       /* # of pages paged in */
>>       MEM_CGROUP_EVENTS_PGPGOUT,      /* # of pages paged out */
>>       MEM_CGROUP_EVENTS_COUNT,        /* # of pages paged in/out */
>> +     MEM_CGROUP_EVENTS_SOFT_STEAL,   /* # of pages reclaimed from */
>> +                                     /* oft reclaim               */
>>       MEM_CGROUP_EVENTS_NSTATS,
>>  };
>>  /*
>> @@ -624,6 +626,11 @@ static void mem_cgroup_charge_statistics(struct mem_cgroup *mem,
>>       preempt_enable();
>>  }
>>
>> +void mem_cgroup_soft_steal(struct mem_cgroup *mem, int val)
>> +{
>> +     this_cpu_add(mem->stat->events[MEM_CGROUP_EVENTS_SOFT_STEAL], val);
>> +}
>> +
>>  static unsigned long mem_cgroup_get_local_zonestat(struct mem_cgroup *mem,
>>                                       enum lru_list idx)
>>  {
>> @@ -3326,6 +3333,9 @@ unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order,
>>                                               &nr_scanned);
>>               nr_reclaimed += reclaimed;
>>               *total_scanned += nr_scanned;
>> +
>> +             mem_cgroup_soft_steal(mz->mem, reclaimed);
>> +
>
> Here, you add "the number of reclaimed pages from the all descendants under me".
> Could you move this to mem_cgroup_hierarchical_reclaim() ? Then, you can report
> the correct stats even with hierarchy enabled.
>
> Even if the value is recorded into hierarchy, total_steal will show total.

good point. I will make that change.

>
> BTW, soft_scan and soft_total_scan aren't necessary ?

Hmm, i can look into that.

--Ying
>
> Thanks,
> -Kame
>
>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH V2 2/2] add stats to monitor soft_limit reclaim

[Ying Han]

On Mon, Mar 28, 2011 at 7:32 PM, Daisuke Nishimura
<nishimura@mxp.nes.nec.co.jp> wrote:
> On Mon, 28 Mar 2011 16:51:10 -0700
> Ying Han <yinghan@google.com> wrote:
>
>> The stat is added:
>>
>> /dev/cgroup/*/memory.stat
>> soft_steal:        - # of pages reclaimed from soft_limit hierarchical reclaim
>> total_soft_steal:  - # sum of all children's "soft_steal"
>>
>> Change log v2...v1
>> 1. removed the counting on number of skips on shrink_zone. This is due to the
>> change on the previous patch.
>>
>> Signed-off-by: Ying Han <yinghan@google.com>
>> ---
>>  Documentation/cgroups/memory.txt |    2 ++
>>  include/linux/memcontrol.h       |    5 +++++
>>  mm/memcontrol.c                  |   14 ++++++++++++++
>>  3 files changed, 21 insertions(+), 0 deletions(-)
>>
>> diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt
>> index b6ed61c..dcda6c5 100644
>> --- a/Documentation/cgroups/memory.txt
>> +++ b/Documentation/cgroups/memory.txt
>> @@ -385,6 +385,7 @@ mapped_file       - # of bytes of mapped file (includes tmpfs/shmem)
>>  pgpgin               - # of pages paged in (equivalent to # of charging events).
>>  pgpgout              - # of pages paged out (equivalent to # of uncharging events).
>>  swap         - # of bytes of swap usage
>> +soft_steal   - # of pages reclaimed from global hierarchical reclaim
>>  inactive_anon        - # of bytes of anonymous memory and swap cache memory on
>>               LRU list.
>>  active_anon  - # of bytes of anonymous and swap cache memory on active
>> @@ -406,6 +407,7 @@ total_mapped_file - sum of all children's "cache"
>>  total_pgpgin         - sum of all children's "pgpgin"
>>  total_pgpgout                - sum of all children's "pgpgout"
>>  total_swap           - sum of all children's "swap"
>> +total_soft_steal     - sum of all children's "soft_steal"
>>  total_inactive_anon  - sum of all children's "inactive_anon"
>>  total_active_anon    - sum of all children's "active_anon"
>>  total_inactive_file  - sum of all children's "inactive_file"
>> diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
>> index 01281ac..151ab40 100644
>> --- a/include/linux/memcontrol.h
>> +++ b/include/linux/memcontrol.h
>> @@ -115,6 +115,7 @@ struct zone_reclaim_stat*
>>  mem_cgroup_get_reclaim_stat_from_page(struct page *page);
>>  extern void mem_cgroup_print_oom_info(struct mem_cgroup *memcg,
>>                                       struct task_struct *p);
>> +void mem_cgroup_soft_steal(struct mem_cgroup *memcg, int val);
>>
>>  #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
>>  extern int do_swap_account;
>> @@ -356,6 +357,10 @@ static inline void mem_cgroup_split_huge_fixup(struct page *head,
>>  {
>>  }
>>
>> +static inline void mem_cgroup_soft_steal(struct mem_cgroup *memcg,
>> +                                      int val)
>> +{
>> +}
>>  #endif /* CONFIG_CGROUP_MEM_CONT */
>>
> Do you use this function outside of memcontrol.c in future, right ?
> I'm asking just for clarification, and I'm sorry if I miss some past discussions.

No, you didn't miss the discussion. That is a reasonable change. thanks

--Ying
>
> Thanks,
> Daisuke Nishimura.
>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH V2 0/2] memcg:add the soft_limit reclaim in global direct reclaim

[Ying Han]

We recently added the change in global background reclaim which counts the
return value of soft_limit reclaim. Now this patch adds the similar logic
on global direct reclaim.

We should skip scanning global LRU on shrink_zone if soft_limit reclaim does
enough work. This is the first step where we start with counting the nr_scanned
and nr_reclaimed from soft_limit reclaim into global scan_control.

The patch is based on mmotm-2011-04-14-15-08 plus
0001-check-pageunevictable-in-lru_deactivate_fn.patch from Minchan.

Ying Han (2):
  Add the soft_limit reclaim in global direct reclaim.
  Add stats to monitor soft_limit reclaim

 Documentation/cgroups/memory.txt |   16 +++++++--
 mm/memcontrol.c                  |   68 ++++++++++++++++++++++++++++----------
 mm/vmscan.c                      |   16 ++++++++-
 3 files changed, 76 insertions(+), 24 deletions(-)

-- 
1.7.3.1

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH V2 1/2] Add the soft_limit reclaim in global direct reclaim.

[Ying Han]

We recently added the change in global background reclaim which
counts the return value of soft_limit reclaim. Now this patch adds
the similar logic on global direct reclaim.

We should skip scanning global LRU on shrink_zone if soft_limit reclaim
does enough work. This is the first step where we start with counting
the nr_scanned and nr_reclaimed from soft_limit reclaim into global
scan_control.

no change since v1.

Signed-off-by: Ying Han <yinghan@google.com>
---
 mm/vmscan.c |   16 ++++++++++++++--
 1 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index b3a569f..84003cc 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1959,11 +1959,14 @@ restart:
  * If a zone is deemed to be full of pinned pages then just give it a light
  * scan then give up on it.
  */
-static void shrink_zones(int priority, struct zonelist *zonelist,
+static unsigned long shrink_zones(int priority, struct zonelist *zonelist,
 					struct scan_control *sc)
 {
 	struct zoneref *z;
 	struct zone *zone;
+	unsigned long nr_soft_reclaimed;
+	unsigned long nr_soft_scanned;
+	unsigned long total_scanned = 0;
 
 	for_each_zone_zonelist_nodemask(zone, z, zonelist,
 					gfp_zone(sc->gfp_mask), sc->nodemask) {
@@ -1980,8 +1983,17 @@ static void shrink_zones(int priority, struct zonelist *zonelist,
 				continue;	/* Let kswapd poll it */
 		}
 
+		nr_soft_scanned = 0;
+		nr_soft_reclaimed = mem_cgroup_soft_limit_reclaim(zone,
+							sc->order, sc->gfp_mask,
+							&nr_soft_scanned);
+		sc->nr_reclaimed += nr_soft_reclaimed;
+		total_scanned += nr_soft_scanned;
+
 		shrink_zone(priority, zone, sc);
 	}
+
+	return total_scanned;
 }
 
 static bool zone_reclaimable(struct zone *zone)
@@ -2045,7 +2057,7 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist,
 		sc->nr_scanned = 0;
 		if (!priority)
 			disable_swap_token();
-		shrink_zones(priority, zonelist, sc);
+		total_scanned += shrink_zones(priority, zonelist, sc);
 		/*
 		 * Don't shrink slabs when reclaiming memory from
 		 * over limit cgroups
-- 
1.7.3.1

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH V2 2/2] Add stats to monitor soft_limit reclaim

[Ying Han]

This patch extend the soft_limit reclaim stats to both global background
reclaim and global direct reclaim.

The following stats are renamed and added:

$cat /dev/cgroup/memory/A/memory.stat
soft_kswapd_steal 1053626
soft_kswapd_scan 1053693
soft_direct_steal 1481810
soft_direct_scan 1481996

changelog v2..v1:
1. rename the stats on soft_kswapd/direct_steal/scan.
2. fix the documentation to match the stat name.

Signed-off-by: Ying Han <yinghan@google.com>
---
 Documentation/cgroups/memory.txt |   16 +++++++--
 mm/memcontrol.c                  |   68 ++++++++++++++++++++++++++++----------
 2 files changed, 62 insertions(+), 22 deletions(-)

diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt
index 0c40dab..575cf33 100644
--- a/Documentation/cgroups/memory.txt
+++ b/Documentation/cgroups/memory.txt
@@ -387,8 +387,14 @@ pgpgout		- # of pages paged out (equivalent to # of uncharging events).
 swap		- # of bytes of swap usage
 pgfault		- # of page faults.
 pgmajfault	- # of major page faults.
-soft_steal	- # of pages reclaimed from global hierarchical reclaim
-soft_scan	- # of pages scanned from global hierarchical reclaim
+soft_kswapd_steal- # of pages reclaimed in global hierarchical reclaim from
+		background reclaim
+soft_kswapd_scan - # of pages scanned in global hierarchical reclaim from
+		background reclaim
+soft_direct_steal- # of pages reclaimed in global hierarchical reclaim from
+		direct reclaim
+soft_direct_scan- # of pages scanned in global hierarchical reclaim from
+		direct reclaim
 inactive_anon	- # of bytes of anonymous memory and swap cache memory on
 		LRU list.
 active_anon	- # of bytes of anonymous and swap cache memory on active
@@ -412,8 +418,10 @@ total_pgpgout		- sum of all children's "pgpgout"
 total_swap		- sum of all children's "swap"
 total_pgfault		- sum of all children's "pgfault"
 total_pgmajfault	- sum of all children's "pgmajfault"
-total_soft_steal	- sum of all children's "soft_steal"
-total_soft_scan		- sum of all children's "soft_scan"
+total_soft_kswapd_steal	- sum of all children's "soft_kswapd_steal"
+total_soft_kswapd_scan	- sum of all children's "soft_kswapd_scan"
+total_soft_direct_steal	- sum of all children's "soft_direct_steal"
+total_soft_direct_scan	- sum of all children's "soft_direct_scan"
 total_inactive_anon	- sum of all children's "inactive_anon"
 total_active_anon	- sum of all children's "active_anon"
 total_inactive_file	- sum of all children's "inactive_file"
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index c2776f1..da1fb2b 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -96,10 +96,14 @@ enum mem_cgroup_events_index {
 	MEM_CGROUP_EVENTS_COUNT,	/* # of pages paged in/out */
 	MEM_CGROUP_EVENTS_PGFAULT,	/* # of page-faults */
 	MEM_CGROUP_EVENTS_PGMAJFAULT,	/* # of major page-faults */
-	MEM_CGROUP_EVENTS_SOFT_STEAL,	/* # of pages reclaimed from */
-					/* soft reclaim               */
-	MEM_CGROUP_EVENTS_SOFT_SCAN,	/* # of pages scanned from */
-					/* soft reclaim               */
+	MEM_CGROUP_EVENTS_SOFT_KSWAPD_STEAL, /* # of pages reclaimed from */
+					/* soft reclaim in background reclaim */
+	MEM_CGROUP_EVENTS_SOFT_KSWAPD_SCAN, /* # of pages scanned from */
+					/* soft reclaim in background reclaim */
+	MEM_CGROUP_EVENTS_SOFT_DIRECT_STEAL, /* # of pages reclaimed from */
+					/* soft reclaim in direct reclaim */
+	MEM_CGROUP_EVENTS_SOFT_DIRECT_SCAN, /* # of pages scanned from */
+					/* soft reclaim in direct reclaim */
 	MEM_CGROUP_EVENTS_NSTATS,
 };
 /*
@@ -640,14 +644,30 @@ static void mem_cgroup_charge_statistics(struct mem_cgroup *mem,
 	preempt_enable();
 }
 
-static void mem_cgroup_soft_steal(struct mem_cgroup *mem, int val)
+static void mem_cgroup_soft_steal(struct mem_cgroup *mem, bool is_kswapd,
+				  int val)
 {
-	this_cpu_add(mem->stat->events[MEM_CGROUP_EVENTS_SOFT_STEAL], val);
+	if (is_kswapd)
+		this_cpu_add(
+			mem->stat->events[MEM_CGROUP_EVENTS_SOFT_KSWAPD_STEAL],
+									val);
+	else
+		this_cpu_add(
+			mem->stat->events[MEM_CGROUP_EVENTS_SOFT_DIRECT_STEAL],
+									val);
 }
 
-static void mem_cgroup_soft_scan(struct mem_cgroup *mem, int val)
+static void mem_cgroup_soft_scan(struct mem_cgroup *mem, bool is_kswapd,
+				 int val)
 {
-	this_cpu_add(mem->stat->events[MEM_CGROUP_EVENTS_SOFT_SCAN], val);
+	if (is_kswapd)
+		this_cpu_add(
+			mem->stat->events[MEM_CGROUP_EVENTS_SOFT_KSWAPD_SCAN],
+									val);
+	else
+		this_cpu_add(
+			mem->stat->events[MEM_CGROUP_EVENTS_SOFT_DIRECT_SCAN],
+									val);
 }
 
 static unsigned long mem_cgroup_get_local_zonestat(struct mem_cgroup *mem,
@@ -1495,6 +1515,7 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem,
 	bool noswap = reclaim_options & MEM_CGROUP_RECLAIM_NOSWAP;
 	bool shrink = reclaim_options & MEM_CGROUP_RECLAIM_SHRINK;
 	bool check_soft = reclaim_options & MEM_CGROUP_RECLAIM_SOFT;
+	bool is_kswapd = false;
 	unsigned long excess;
 	unsigned long nr_scanned;
 
@@ -1504,6 +1525,9 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem,
 	if (root_mem->memsw_is_minimum)
 		noswap = true;
 
+	if (current_is_kswapd())
+		is_kswapd = true;
+
 	while (1) {
 		victim = mem_cgroup_select_victim(root_mem);
 		if (victim == root_mem) {
@@ -1544,8 +1568,8 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem,
 				noswap, get_swappiness(victim), zone,
 				&nr_scanned);
 			*total_scanned += nr_scanned;
-			mem_cgroup_soft_steal(victim, ret);
-			mem_cgroup_soft_scan(victim, nr_scanned);
+			mem_cgroup_soft_steal(victim, is_kswapd, ret);
+			mem_cgroup_soft_scan(victim, is_kswapd, nr_scanned);
 		} else
 			ret = try_to_free_mem_cgroup_pages(victim, gfp_mask,
 						noswap, get_swappiness(victim));
@@ -3840,8 +3864,10 @@ enum {
 	MCS_SWAP,
 	MCS_PGFAULT,
 	MCS_PGMAJFAULT,
-	MCS_SOFT_STEAL,
-	MCS_SOFT_SCAN,
+	MCS_SOFT_KSWAPD_STEAL,
+	MCS_SOFT_KSWAPD_SCAN,
+	MCS_SOFT_DIRECT_STEAL,
+	MCS_SOFT_DIRECT_SCAN,
 	MCS_INACTIVE_ANON,
 	MCS_ACTIVE_ANON,
 	MCS_INACTIVE_FILE,
@@ -3866,8 +3892,10 @@ struct {
 	{"swap", "total_swap"},
 	{"pgfault", "total_pgfault"},
 	{"pgmajfault", "total_pgmajfault"},
-	{"soft_steal", "total_soft_steal"},
-	{"soft_scan", "total_soft_scan"},
+	{"soft_kswapd_steal", "total_soft_kswapd_steal"},
+	{"soft_kswapd_scan", "total_soft_kswapd_scan"},
+	{"soft_direct_steal", "total_soft_direct_steal"},
+	{"soft_direct_scan", "total_soft_direct_scan"},
 	{"inactive_anon", "total_inactive_anon"},
 	{"active_anon", "total_active_anon"},
 	{"inactive_file", "total_inactive_file"},
@@ -3896,10 +3924,14 @@ mem_cgroup_get_local_stat(struct mem_cgroup *mem, struct mcs_total_stat *s)
 		val = mem_cgroup_read_stat(mem, MEM_CGROUP_STAT_SWAPOUT);
 		s->stat[MCS_SWAP] += val * PAGE_SIZE;
 	}
-	val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_SOFT_STEAL);
-	s->stat[MCS_SOFT_STEAL] += val;
-	val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_SOFT_SCAN);
-	s->stat[MCS_SOFT_SCAN] += val;
+	val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_SOFT_KSWAPD_STEAL);
+	s->stat[MCS_SOFT_KSWAPD_STEAL] += val;
+	val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_SOFT_KSWAPD_SCAN);
+	s->stat[MCS_SOFT_KSWAPD_SCAN] += val;
+	val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_SOFT_DIRECT_STEAL);
+	s->stat[MCS_SOFT_DIRECT_STEAL] += val;
+	val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_SOFT_DIRECT_SCAN);
+	s->stat[MCS_SOFT_DIRECT_SCAN] += val;
 	val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_PGFAULT);
 	s->stat[MCS_PGFAULT] += val;
 	val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_PGMAJFAULT);
-- 
1.7.3.1

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH V2 1/2] Add the soft_limit reclaim in global direct reclaim.

[Ying Han]

On Mon, May 2, 2011 at 9:50 AM, Ying Han <yinghan@google.com> wrote:
> We recently added the change in global background reclaim which
> counts the return value of soft_limit reclaim. Now this patch adds
> the similar logic on global direct reclaim.
>
> We should skip scanning global LRU on shrink_zone if soft_limit reclaim
> does enough work. This is the first step where we start with counting
> the nr_scanned and nr_reclaimed from soft_limit reclaim into global
> scan_control.
>
> no change since v1.
>
> Signed-off-by: Ying Han <yinghan@google.com>
> ---
>  mm/vmscan.c |   16 ++++++++++++++--
>  1 files changed, 14 insertions(+), 2 deletions(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index b3a569f..84003cc 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -1959,11 +1959,14 @@ restart:
>  * If a zone is deemed to be full of pinned pages then just give it a light
>  * scan then give up on it.
>  */
> -static void shrink_zones(int priority, struct zonelist *zonelist,
> +static unsigned long shrink_zones(int priority, struct zonelist *zonelist,
>                                        struct scan_control *sc)
>  {
>        struct zoneref *z;
>        struct zone *zone;
> +       unsigned long nr_soft_reclaimed;
> +       unsigned long nr_soft_scanned;
> +       unsigned long total_scanned = 0;
>
>        for_each_zone_zonelist_nodemask(zone, z, zonelist,
>                                        gfp_zone(sc->gfp_mask), sc->nodemask) {
> @@ -1980,8 +1983,17 @@ static void shrink_zones(int priority, struct zonelist *zonelist,
>                                continue;       /* Let kswapd poll it */
>                }
>
> +               nr_soft_scanned = 0;
> +               nr_soft_reclaimed = mem_cgroup_soft_limit_reclaim(zone,
> +                                                       sc->order, sc->gfp_mask,
> +                                                       &nr_soft_scanned);
> +               sc->nr_reclaimed += nr_soft_reclaimed;
> +               total_scanned += nr_soft_scanned;
> +
>                shrink_zone(priority, zone, sc);
>        }
> +
> +       return total_scanned;
>  }
>
>  static bool zone_reclaimable(struct zone *zone)
> @@ -2045,7 +2057,7 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist,
>                sc->nr_scanned = 0;
>                if (!priority)
>                        disable_swap_token();
> -               shrink_zones(priority, zonelist, sc);
> +               total_scanned += shrink_zones(priority, zonelist, sc);
>                /*
>                 * Don't shrink slabs when reclaiming memory from
>                 * over limit cgroups
> --
> 1.7.3.1
>
>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH V2 2/2] Add stats to monitor soft_limit reclaim

[Ying Han]

On Mon, May 2, 2011 at 9:50 AM, Ying Han <yinghan@google.com> wrote:
> This patch extend the soft_limit reclaim stats to both global background
> reclaim and global direct reclaim.
>
> The following stats are renamed and added:
>
> $cat /dev/cgroup/memory/A/memory.stat
> soft_kswapd_steal 1053626
> soft_kswapd_scan 1053693
> soft_direct_steal 1481810
> soft_direct_scan 1481996
>
> changelog v2..v1:
> 1. rename the stats on soft_kswapd/direct_steal/scan.
> 2. fix the documentation to match the stat name.
>
> Signed-off-by: Ying Han <yinghan@google.com>
> ---
>  Documentation/cgroups/memory.txt |   16 +++++++--
>  mm/memcontrol.c                  |   68 ++++++++++++++++++++++++++++----------
>  2 files changed, 62 insertions(+), 22 deletions(-)
>
> diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt
> index 0c40dab..575cf33 100644
> --- a/Documentation/cgroups/memory.txt
> +++ b/Documentation/cgroups/memory.txt
> @@ -387,8 +387,14 @@ pgpgout            - # of pages paged out (equivalent to # of uncharging events).
>  swap           - # of bytes of swap usage
>  pgfault                - # of page faults.
>  pgmajfault     - # of major page faults.
> -soft_steal     - # of pages reclaimed from global hierarchical reclaim
> -soft_scan      - # of pages scanned from global hierarchical reclaim
> +soft_kswapd_steal- # of pages reclaimed in global hierarchical reclaim from
> +               background reclaim
> +soft_kswapd_scan - # of pages scanned in global hierarchical reclaim from
> +               background reclaim
> +soft_direct_steal- # of pages reclaimed in global hierarchical reclaim from
> +               direct reclaim
> +soft_direct_scan- # of pages scanned in global hierarchical reclaim from
> +               direct reclaim
>  inactive_anon  - # of bytes of anonymous memory and swap cache memory on
>                LRU list.
>  active_anon    - # of bytes of anonymous and swap cache memory on active
> @@ -412,8 +418,10 @@ total_pgpgout              - sum of all children's "pgpgout"
>  total_swap             - sum of all children's "swap"
>  total_pgfault          - sum of all children's "pgfault"
>  total_pgmajfault       - sum of all children's "pgmajfault"
> -total_soft_steal       - sum of all children's "soft_steal"
> -total_soft_scan                - sum of all children's "soft_scan"
> +total_soft_kswapd_steal        - sum of all children's "soft_kswapd_steal"
> +total_soft_kswapd_scan - sum of all children's "soft_kswapd_scan"
> +total_soft_direct_steal        - sum of all children's "soft_direct_steal"
> +total_soft_direct_scan - sum of all children's "soft_direct_scan"
>  total_inactive_anon    - sum of all children's "inactive_anon"
>  total_active_anon      - sum of all children's "active_anon"
>  total_inactive_file    - sum of all children's "inactive_file"
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index c2776f1..da1fb2b 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -96,10 +96,14 @@ enum mem_cgroup_events_index {
>        MEM_CGROUP_EVENTS_COUNT,        /* # of pages paged in/out */
>        MEM_CGROUP_EVENTS_PGFAULT,      /* # of page-faults */
>        MEM_CGROUP_EVENTS_PGMAJFAULT,   /* # of major page-faults */
> -       MEM_CGROUP_EVENTS_SOFT_STEAL,   /* # of pages reclaimed from */
> -                                       /* soft reclaim               */
> -       MEM_CGROUP_EVENTS_SOFT_SCAN,    /* # of pages scanned from */
> -                                       /* soft reclaim               */
> +       MEM_CGROUP_EVENTS_SOFT_KSWAPD_STEAL, /* # of pages reclaimed from */
> +                                       /* soft reclaim in background reclaim */
> +       MEM_CGROUP_EVENTS_SOFT_KSWAPD_SCAN, /* # of pages scanned from */
> +                                       /* soft reclaim in background reclaim */
> +       MEM_CGROUP_EVENTS_SOFT_DIRECT_STEAL, /* # of pages reclaimed from */
> +                                       /* soft reclaim in direct reclaim */
> +       MEM_CGROUP_EVENTS_SOFT_DIRECT_SCAN, /* # of pages scanned from */
> +                                       /* soft reclaim in direct reclaim */
>        MEM_CGROUP_EVENTS_NSTATS,
>  };
>  /*
> @@ -640,14 +644,30 @@ static void mem_cgroup_charge_statistics(struct mem_cgroup *mem,
>        preempt_enable();
>  }
>
> -static void mem_cgroup_soft_steal(struct mem_cgroup *mem, int val)
> +static void mem_cgroup_soft_steal(struct mem_cgroup *mem, bool is_kswapd,
> +                                 int val)
>  {
> -       this_cpu_add(mem->stat->events[MEM_CGROUP_EVENTS_SOFT_STEAL], val);
> +       if (is_kswapd)
> +               this_cpu_add(
> +                       mem->stat->events[MEM_CGROUP_EVENTS_SOFT_KSWAPD_STEAL],
> +                                                                       val);
> +       else
> +               this_cpu_add(
> +                       mem->stat->events[MEM_CGROUP_EVENTS_SOFT_DIRECT_STEAL],
> +                                                                       val);
>  }
>
> -static void mem_cgroup_soft_scan(struct mem_cgroup *mem, int val)
> +static void mem_cgroup_soft_scan(struct mem_cgroup *mem, bool is_kswapd,
> +                                int val)
>  {
> -       this_cpu_add(mem->stat->events[MEM_CGROUP_EVENTS_SOFT_SCAN], val);
> +       if (is_kswapd)
> +               this_cpu_add(
> +                       mem->stat->events[MEM_CGROUP_EVENTS_SOFT_KSWAPD_SCAN],
> +                                                                       val);
> +       else
> +               this_cpu_add(
> +                       mem->stat->events[MEM_CGROUP_EVENTS_SOFT_DIRECT_SCAN],
> +                                                                       val);
>  }
>
>  static unsigned long mem_cgroup_get_local_zonestat(struct mem_cgroup *mem,
> @@ -1495,6 +1515,7 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem,
>        bool noswap = reclaim_options & MEM_CGROUP_RECLAIM_NOSWAP;
>        bool shrink = reclaim_options & MEM_CGROUP_RECLAIM_SHRINK;
>        bool check_soft = reclaim_options & MEM_CGROUP_RECLAIM_SOFT;
> +       bool is_kswapd = false;
>        unsigned long excess;
>        unsigned long nr_scanned;
>
> @@ -1504,6 +1525,9 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem,
>        if (root_mem->memsw_is_minimum)
>                noswap = true;
>
> +       if (current_is_kswapd())
> +               is_kswapd = true;
> +
>        while (1) {
>                victim = mem_cgroup_select_victim(root_mem);
>                if (victim == root_mem) {
> @@ -1544,8 +1568,8 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem,
>                                noswap, get_swappiness(victim), zone,
>                                &nr_scanned);
>                        *total_scanned += nr_scanned;
> -                       mem_cgroup_soft_steal(victim, ret);
> -                       mem_cgroup_soft_scan(victim, nr_scanned);
> +                       mem_cgroup_soft_steal(victim, is_kswapd, ret);
> +                       mem_cgroup_soft_scan(victim, is_kswapd, nr_scanned);
>                } else
>                        ret = try_to_free_mem_cgroup_pages(victim, gfp_mask,
>                                                noswap, get_swappiness(victim));
> @@ -3840,8 +3864,10 @@ enum {
>        MCS_SWAP,
>        MCS_PGFAULT,
>        MCS_PGMAJFAULT,
> -       MCS_SOFT_STEAL,
> -       MCS_SOFT_SCAN,
> +       MCS_SOFT_KSWAPD_STEAL,
> +       MCS_SOFT_KSWAPD_SCAN,
> +       MCS_SOFT_DIRECT_STEAL,
> +       MCS_SOFT_DIRECT_SCAN,
>        MCS_INACTIVE_ANON,
>        MCS_ACTIVE_ANON,
>        MCS_INACTIVE_FILE,
> @@ -3866,8 +3892,10 @@ struct {
>        {"swap", "total_swap"},
>        {"pgfault", "total_pgfault"},
>        {"pgmajfault", "total_pgmajfault"},
> -       {"soft_steal", "total_soft_steal"},
> -       {"soft_scan", "total_soft_scan"},
> +       {"soft_kswapd_steal", "total_soft_kswapd_steal"},
> +       {"soft_kswapd_scan", "total_soft_kswapd_scan"},
> +       {"soft_direct_steal", "total_soft_direct_steal"},
> +       {"soft_direct_scan", "total_soft_direct_scan"},
>        {"inactive_anon", "total_inactive_anon"},
>        {"active_anon", "total_active_anon"},
>        {"inactive_file", "total_inactive_file"},
> @@ -3896,10 +3924,14 @@ mem_cgroup_get_local_stat(struct mem_cgroup *mem, struct mcs_total_stat *s)
>                val = mem_cgroup_read_stat(mem, MEM_CGROUP_STAT_SWAPOUT);
>                s->stat[MCS_SWAP] += val * PAGE_SIZE;
>        }
> -       val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_SOFT_STEAL);
> -       s->stat[MCS_SOFT_STEAL] += val;
> -       val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_SOFT_SCAN);
> -       s->stat[MCS_SOFT_SCAN] += val;
> +       val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_SOFT_KSWAPD_STEAL);
> +       s->stat[MCS_SOFT_KSWAPD_STEAL] += val;
> +       val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_SOFT_KSWAPD_SCAN);
> +       s->stat[MCS_SOFT_KSWAPD_SCAN] += val;
> +       val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_SOFT_DIRECT_STEAL);
> +       s->stat[MCS_SOFT_DIRECT_STEAL] += val;
> +       val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_SOFT_DIRECT_SCAN);
> +       s->stat[MCS_SOFT_DIRECT_SCAN] += val;
>        val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_PGFAULT);
>        s->stat[MCS_PGFAULT] += val;
>        val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_PGMAJFAULT);
> --
> 1.7.3.1
>
>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH V2 1/2] Add the soft_limit reclaim in global direct reclaim.

[Rik van Riel]

On 05/02/2011 12:50 PM, Ying Han wrote:
> We recently added the change in global background reclaim which
> counts the return value of soft_limit reclaim. Now this patch adds
> the similar logic on global direct reclaim.
>
> We should skip scanning global LRU on shrink_zone if soft_limit reclaim
> does enough work. This is the first step where we start with counting
> the nr_scanned and nr_reclaimed from soft_limit reclaim into global
> scan_control.

Would be nice to see that at some point, but simply counting
the amount reclaimed from the over softlimit groups is a good
start.

> no change since v1.
>
> Signed-off-by: Ying Han<yinghan@google.com>

Acked-by: Rik van Riel<riel@redhat.com>

-- 
All rights reversed

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH V2 2/2] Add stats to monitor soft_limit reclaim

[Rik van Riel]

On 05/02/2011 12:50 PM, Ying Han wrote:
> This patch extend the soft_limit reclaim stats to both global background
> reclaim and global direct reclaim.
>
> The following stats are renamed and added:
>
> $cat /dev/cgroup/memory/A/memory.stat
> soft_kswapd_steal 1053626
> soft_kswapd_scan 1053693
> soft_direct_steal 1481810
> soft_direct_scan 1481996
>
> changelog v2..v1:
> 1. rename the stats on soft_kswapd/direct_steal/scan.
> 2. fix the documentation to match the stat name.

> Signed-off-by: Ying Han<yinghan@google.com>

Acked-by: Rik van Riel<riel@redhat.com>

I expect people to continue arguing over the names a little
longer, but feel free to keep my Acked-by: across the various
name changes :)

-- 
All rights reversed

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH V2 2/2] Add stats to monitor soft_limit reclaim

[Ying Han]

On Tue, May 3, 2011 at 6:54 AM, Rik van Riel <riel@redhat.com> wrote:
> On 05/02/2011 12:50 PM, Ying Han wrote:
>>
>> This patch extend the soft_limit reclaim stats to both global background
>> reclaim and global direct reclaim.
>>
>> The following stats are renamed and added:
>>
>> $cat /dev/cgroup/memory/A/memory.stat
>> soft_kswapd_steal 1053626
>> soft_kswapd_scan 1053693
>> soft_direct_steal 1481810
>> soft_direct_scan 1481996
>>
>> changelog v2..v1:
>> 1. rename the stats on soft_kswapd/direct_steal/scan.
>> 2. fix the documentation to match the stat name.
>
>> Signed-off-by: Ying Han<yinghan@google.com>
>
> Acked-by: Rik van Riel<riel@redhat.com>
>
> I expect people to continue arguing over the names a little
> longer, but feel free to keep my Acked-by: across the various
> name changes :)

Hi Rik:

Thank you for reviewing and acking. Regarding the name, i think we are
reaching some stage. And my
current naming is following that as well. :)

--Ying
>
> --
> All rights reversed
>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>