[PATCH v3 1/4] memcg: cleanup useless LRU_ALL_EVICTABLE

[PATCH v3 1/4] memcg: cleanup useless LRU_ALL_EVICTABLE

[Wanpeng Li]

From: Wanpeng Li <liwp@linux.vnet.ibm.com>

Since LRU_ALL_EVICTABLE is useless, just remove it.
Add LRU_ALL_UNEVICTABLE to mask unevictable pages.

Signed-off-by: Wanpeng Li <liwp.linux@gmail.com>
---
 include/linux/mmzone.h |    2 +-
 mm/memcontrol.c        |    4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 68c569f..5873620 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -209,7 +209,7 @@ struct lruvec {
 /* Mask used at gathering information at once (see memcontrol.c) */
 #define LRU_ALL_FILE (BIT(LRU_INACTIVE_FILE) | BIT(LRU_ACTIVE_FILE))
 #define LRU_ALL_ANON (BIT(LRU_INACTIVE_ANON) | BIT(LRU_ACTIVE_ANON))
-#define LRU_ALL_EVICTABLE (LRU_ALL_FILE | LRU_ALL_ANON)
+#define LRU_ALL_UNEVICTABLE (BIT(LRU_UNEVICTABLE))
 #define LRU_ALL	     ((1 << NR_LRU_LISTS) - 1)
 
 /* Isolate clean file */
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 724bd02..ccda728 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -4033,11 +4033,11 @@ static int mem_control_numa_stat_show(struct cgroup *cont, struct cftype *cft,
 	}
 	seq_putc(m, '\n');
 
-	unevictable_nr = mem_cgroup_nr_lru_pages(memcg, BIT(LRU_UNEVICTABLE));
+	unevictable_nr = mem_cgroup_nr_lru_pages(memcg, LRU_ALL_UNEVICTABLE);
 	seq_printf(m, "unevictable=%lu", unevictable_nr);
 	for_each_node_state(nid, N_HIGH_MEMORY) {
 		node_nr = mem_cgroup_node_nr_lru_pages(memcg, nid,
-				BIT(LRU_UNEVICTABLE));
+				LRU_ALL_UNEVICTABLE);
 		seq_printf(m, " N%d=%lu", nid, node_nr);
 	}
 	seq_putc(m, '\n');
-- 
1.7.9.5

Re: [PATCH v3 1/4] memcg: cleanup useless LRU_ALL_EVICTABLE

[Michal Hocko]

Marking this patch as memcg is quite misleading. mm is more suitable.

On Mon 25-06-12 16:43:32, Wanpeng Li wrote:
> From: Wanpeng Li <liwp@linux.vnet.ibm.com>
> 
> Since LRU_ALL_EVICTABLE is useless, just remove it.

It's really strange but I do not see any usage of this except for it's
introduction. So agreed to remove it.

> Add LRU_ALL_UNEVICTABLE to mask unevictable pages.

Does it really make sense to introduce just for two uses in a single
function?

> 
> Signed-off-by: Wanpeng Li <liwp.linux@gmail.com>
> ---
>  include/linux/mmzone.h |    2 +-
>  mm/memcontrol.c        |    4 ++--
>  2 files changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
> index 68c569f..5873620 100644
> --- a/include/linux/mmzone.h
> +++ b/include/linux/mmzone.h
> @@ -209,7 +209,7 @@ struct lruvec {
>  /* Mask used at gathering information at once (see memcontrol.c) */
>  #define LRU_ALL_FILE (BIT(LRU_INACTIVE_FILE) | BIT(LRU_ACTIVE_FILE))
>  #define LRU_ALL_ANON (BIT(LRU_INACTIVE_ANON) | BIT(LRU_ACTIVE_ANON))
> -#define LRU_ALL_EVICTABLE (LRU_ALL_FILE | LRU_ALL_ANON)
> +#define LRU_ALL_UNEVICTABLE (BIT(LRU_UNEVICTABLE))
>  #define LRU_ALL	     ((1 << NR_LRU_LISTS) - 1)
>  
>  /* Isolate clean file */
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 724bd02..ccda728 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -4033,11 +4033,11 @@ static int mem_control_numa_stat_show(struct cgroup *cont, struct cftype *cft,
>  	}
>  	seq_putc(m, '\n');
>  
> -	unevictable_nr = mem_cgroup_nr_lru_pages(memcg, BIT(LRU_UNEVICTABLE));
> +	unevictable_nr = mem_cgroup_nr_lru_pages(memcg, LRU_ALL_UNEVICTABLE);
>  	seq_printf(m, "unevictable=%lu", unevictable_nr);
>  	for_each_node_state(nid, N_HIGH_MEMORY) {
>  		node_nr = mem_cgroup_node_nr_lru_pages(memcg, nid,
> -				BIT(LRU_UNEVICTABLE));
> +				LRU_ALL_UNEVICTABLE);
>  		seq_printf(m, " N%d=%lu", nid, node_nr);
>  	}
>  	seq_putc(m, '\n');
> -- 
> 1.7.9.5
> 
> --
> To unsubscribe from this list: send the line "unsubscribe cgroups" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

-- 
Michal Hocko
SUSE Labs
SUSE LINUX s.r.o.
Lihovarska 1060/12
190 00 Praha 9    
Czech Republic

Re: [PATCH v3 1/4] memcg: cleanup useless LRU_ALL_EVICTABLE

[Kamezawa Hiroyuki]

(2012/06/25 18:33), Michal Hocko wrote:
> Marking this patch as memcg is quite misleading. mm is more suitable.
>
> On Mon 25-06-12 16:43:32, Wanpeng Li wrote:
>> From: Wanpeng Li <liwp-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org>
>>
>> Since LRU_ALL_EVICTABLE is useless, just remove it.
>
> It's really strange but I do not see any usage of this except for it's
> introduction. So agreed to remove it.
>
>> Add LRU_ALL_UNEVICTABLE to mask unevictable pages.
>
> Does it really make sense to introduce just for two uses in a single
> function?
>

I don't think so.

I doubt LRU_ALL_UNEVICTABLE implies there are other UNEVICTABLE LRUS
other than LRU_UNEVICTABLE.

For example, LRU_ALL_UNEVICTABLE means
	if total_swap==0, UNEVICTABLE+ANON
	if total_swap!=0, UNEVICTABLE,

This will have its own meanings...but I don't see there are any users.

Thanks,
-Kame