This patch implements trivial statistics for the memory resource controller.

Signed-off-by: Balaji Rao <balajirrao@gmail.com>
CC: Balbir Singh <balbir@linux.vnet.ibm.com>
CC: Dhaval Giani <dhaval@linux.vnet.ibm.com>


diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index eb681a6..84f3fe5 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -45,6 +45,8 @@ enum mem_cgroup_stat_index {
 	 */
 	MEM_CGROUP_STAT_CACHE, 	   /* # of pages charged as cache */
 	MEM_CGROUP_STAT_RSS,	   /* # of pages charged as rss */
+	MEM_CGROUP_STAT_PGIN_COUNT,	/* # of pages paged in */
+	MEM_CGROUP_STAT_PGOUT_COUNT,	/* # of pages paged out */
 
 	MEM_CGROUP_STAT_NSTATS,
 };
@@ -196,6 +198,13 @@ static void mem_cgroup_charge_statistics(struct mem_cgroup *mem, int flags,
 		__mem_cgroup_stat_add_safe(stat, MEM_CGROUP_STAT_CACHE, val);
 	else
 		__mem_cgroup_stat_add_safe(stat, MEM_CGROUP_STAT_RSS, val);
+
+	if (charge)
+		__mem_cgroup_stat_add_safe(stat,
+				MEM_CGROUP_STAT_PGIN_COUNT, 1);
+	else
+		__mem_cgroup_stat_add_safe(stat,
+				MEM_CGROUP_STAT_PGOUT_COUNT, 1);
 }
 
 static struct mem_cgroup_per_zone *
@@ -886,6 +895,8 @@ static const struct mem_cgroup_stat_desc {
 } mem_cgroup_stat_desc[] = {
 	[MEM_CGROUP_STAT_CACHE] = { "cache", PAGE_SIZE, },
 	[MEM_CGROUP_STAT_RSS] = { "rss", PAGE_SIZE, },
+	[MEM_CGROUP_STAT_PGIN_COUNT] = {"page_in_count", 1, },
+	[MEM_CGROUP_STAT_PGOUT_COUNT] = {"page_out_count", 1, },
 };
 
 static int mem_control_stat_show(struct cgroup *cont, struct cftype *cft,
_______________________________________________
Containers mailing list
Containers@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers

On Wed, Mar 26, 2008 at 11:18 AM, Balaji Rao <balajirrao@gmail.com> wrote:
>         [MEM_CGROUP_STAT_RSS] = { "rss", PAGE_SIZE, },
>  +       [MEM_CGROUP_STAT_PGIN_COUNT] = {"page_in_count", 1, },
>  +       [MEM_CGROUP_STAT_PGOUT_COUNT] = {"page_out_count", 1, },
>   };
>

Should these be called "pgpgin" and "pgpgout" for consistency with /proc/vmstat?

Paul
_______________________________________________
Containers mailing list
Containers@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers

Paul Menage wrote:
> On Wed, Mar 26, 2008 at 11:18 AM, Balaji Rao <balajirrao@gmail.com> wrote:
>>         [MEM_CGROUP_STAT_RSS] = { "rss", PAGE_SIZE, },
>>  +       [MEM_CGROUP_STAT_PGIN_COUNT] = {"page_in_count", 1, },
>>  +       [MEM_CGROUP_STAT_PGOUT_COUNT] = {"page_out_count", 1, },
>>   };
>>
> 
> Should these be called "pgpgin" and "pgpgout" for consistency with /proc/vmstat?
> 
> Paul

Yes, that is a good name. We also want to instantaneous metrics like
page_in_count per second and page_out_count per second to help the administrator
monitor what is going on within the tasks in the cgroup w.r.t. memory.


-- 
	Warm Regards,
	Balbir Singh
	Linux Technology Center
	IBM, ISTL
_______________________________________________
Containers mailing list
Containers@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers

Balaji Rao wrote:
> This patch implements trivial statistics for the memory resource controller.
> 
> Signed-off-by: Balaji Rao <balajirrao@gmail.com>
> CC: Balbir Singh <balbir@linux.vnet.ibm.com>
> CC: Dhaval Giani <dhaval@linux.vnet.ibm.com>
> 
> 
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index eb681a6..84f3fe5 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -45,6 +45,8 @@ enum mem_cgroup_stat_index {
>  	 */
>  	MEM_CGROUP_STAT_CACHE, 	   /* # of pages charged as cache */
>  	MEM_CGROUP_STAT_RSS,	   /* # of pages charged as rss */
> +	MEM_CGROUP_STAT_PGIN_COUNT,	/* # of pages paged in */
> +	MEM_CGROUP_STAT_PGOUT_COUNT,	/* # of pages paged out */
> 
>  	MEM_CGROUP_STAT_NSTATS,
>  };
> @@ -196,6 +198,13 @@ static void mem_cgroup_charge_statistics(struct mem_cgroup *mem, int flags,
>  		__mem_cgroup_stat_add_safe(stat, MEM_CGROUP_STAT_CACHE, val);
>  	else
>  		__mem_cgroup_stat_add_safe(stat, MEM_CGROUP_STAT_RSS, val);
> +
> +	if (charge)
> +		__mem_cgroup_stat_add_safe(stat,
> +				MEM_CGROUP_STAT_PGIN_COUNT, 1);
> +	else
> +		__mem_cgroup_stat_add_safe(stat,
> +				MEM_CGROUP_STAT_PGOUT_COUNT, 1);
>  }
> 
>  static struct mem_cgroup_per_zone *
> @@ -886,6 +895,8 @@ static const struct mem_cgroup_stat_desc {
>  } mem_cgroup_stat_desc[] = {
>  	[MEM_CGROUP_STAT_CACHE] = { "cache", PAGE_SIZE, },
>  	[MEM_CGROUP_STAT_RSS] = { "rss", PAGE_SIZE, },
> +	[MEM_CGROUP_STAT_PGIN_COUNT] = {"page_in_count", 1, },
> +	[MEM_CGROUP_STAT_PGOUT_COUNT] = {"page_out_count", 1, },
>  };
> 
>  static int mem_control_stat_show(struct cgroup *cont, struct cftype *cft,

Looks simple and nice. Please see Paul's mail on the suggested name change as well

Acked-by:  Balbir Singh <balbir@linux.vnet.ibm.com>


-- 
	Warm Regards,
	Balbir Singh
	Linux Technology Center
	IBM, ISTL
_______________________________________________
Containers mailing list
Containers@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers