[PATCH] memcg: clean up migration

[PATCH] memcg: clean up migration

[Minchan Kim]

This patch cleans up unncessary BUG_ON check and confusing
charge variable.

That's because memcg charge/uncharge could be handled by
mem_cgroup_[prepare/end] migration itself so charge local variable
in unmap_and_move lost the role since we introduced 01b1ae63c2.

And mem_cgroup_prepare_migratio return 0 if only it is successful.
Otherwise, it jumps to unlock label to clean up so BUG_ON(charge)
isn;t meaningless.

Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 mm/memcontrol.c |    1 +
 mm/migrate.c    |   14 ++++----------
 2 files changed, 5 insertions(+), 10 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 2fc97fc..6832926 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -2872,6 +2872,7 @@ static inline int mem_cgroup_move_swap_account(swp_entry_t entry,
 /*
  * Before starting migration, account PAGE_SIZE to mem_cgroup that the old
  * page belongs to.
+ * Return 0 if charge is successful. Otherwise return -errno.
  */
 int mem_cgroup_prepare_migration(struct page *page,
 	struct page *newpage, struct mem_cgroup **ptr, gfp_t gfp_mask)
diff --git a/mm/migrate.c b/mm/migrate.c
index eb083a6..737c2e5 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -622,7 +622,6 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
 	int *result = NULL;
 	struct page *newpage = get_new_page(page, private, &result);
 	int remap_swapcache = 1;
-	int charge = 0;
 	struct mem_cgroup *mem;
 	struct anon_vma *anon_vma = NULL;
 
@@ -637,9 +636,7 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
 		if (unlikely(split_huge_page(page)))
 			goto move_newpage;
 
-	/* prepare cgroup just returns 0 or -ENOMEM */
 	rc = -EAGAIN;
-
 	if (!trylock_page(page)) {
 		if (!force)
 			goto move_newpage;
@@ -678,13 +675,11 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
 	}
 
 	/* charge against new page */
-	charge = mem_cgroup_prepare_migration(page, newpage, &mem, GFP_KERNEL);
-	if (charge == -ENOMEM) {
-		rc = -ENOMEM;
+	rc = mem_cgroup_prepare_migration(page, newpage, &mem, GFP_KERNEL);
+	if (rc)
 		goto unlock;
-	}
-	BUG_ON(charge);
 
+	rc = -EAGAIN;
 	if (PageWriteback(page)) {
 		if (!force || !sync)
 			goto uncharge;
@@ -767,8 +762,7 @@ skip_unmap:
 		drop_anon_vma(anon_vma);
 
 uncharge:
-	if (!charge)
-		mem_cgroup_end_migration(mem, page, newpage, rc == 0);
+	mem_cgroup_end_migration(mem, page, newpage, rc == 0);
 unlock:
 	unlock_page(page);
 
-- 
1.7.1

--
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] memcg: clean up migration

[Johannes Weiner]

On Mon, Feb 28, 2011 at 12:49:25AM +0900, Minchan Kim wrote:
> This patch cleans up unncessary BUG_ON check and confusing
> charge variable.
> 
> That's because memcg charge/uncharge could be handled by
> mem_cgroup_[prepare/end] migration itself so charge local variable
> in unmap_and_move lost the role since we introduced 01b1ae63c2.
> 
> And mem_cgroup_prepare_migratio return 0 if only it is successful.
> Otherwise, it jumps to unlock label to clean up so BUG_ON(charge)
> isn;t meaningless.
> 
> Cc: Johannes Weiner <hannes@cmpxchg.org>
> Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
> Signed-off-by: Minchan Kim <minchan.kim@gmail.com>

Reviewed-by: Johannes Weiner <hannes@cmpxchg.org>

Thanks, Minchan!

	Hannes

PS: Btw, people sometimes refer to commits by hashes from trees other
than Linus's, so it's nice to include the subject as well:

	01b1ae6 memcg: simple migration handling

You get this easily by taking the first line of

	git show --oneline <commithash>

though there are probably other ways.

--
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] memcg: clean up migration

[Minchan Kim]

On Mon, Feb 28, 2011 at 6:09 AM, Johannes Weiner <hannes@cmpxchg.org> wrote:
> On Mon, Feb 28, 2011 at 12:49:25AM +0900, Minchan Kim wrote:
>> This patch cleans up unncessary BUG_ON check and confusing
>> charge variable.
>>
>> That's because memcg charge/uncharge could be handled by
>> mem_cgroup_[prepare/end] migration itself so charge local variable
>> in unmap_and_move lost the role since we introduced 01b1ae63c2.
>>
>> And mem_cgroup_prepare_migratio return 0 if only it is successful.
>> Otherwise, it jumps to unlock label to clean up so BUG_ON(charge)
>> isn;t meaningless.
>>
>> Cc: Johannes Weiner <hannes@cmpxchg.org>
>> Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
>> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
>> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
>> Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
>
> Reviewed-by: Johannes Weiner <hannes@cmpxchg.org>
>
> Thanks, Minchan!
>
>        Hannes
>
> PS: Btw, people sometimes refer to commits by hashes from trees other
> than Linus's, so it's nice to include the subject as well:
>
>        01b1ae6 memcg: simple migration handling
>
> You get this easily by taking the first line of
>
>        git show --oneline <commithash>
>
> though there are probably other ways.
>

I forgot adding the name with old comment's copy & paste. I will
resend with fixing some typo.

Thanks.


-- 
Kind regards,
Minchan Kim

--
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] memcg: clean up migration

[KAMEZAWA Hiroyuki]

On Mon, 28 Feb 2011 00:49:25 +0900
Minchan Kim <minchan.kim@gmail.com> wrote:

> This patch cleans up unncessary BUG_ON check and confusing
> charge variable.
> 
> That's because memcg charge/uncharge could be handled by
> mem_cgroup_[prepare/end] migration itself so charge local variable
> in unmap_and_move lost the role since we introduced 01b1ae63c2.
> 
> And mem_cgroup_prepare_migratio return 0 if only it is successful.
> Otherwise, it jumps to unlock label to clean up so BUG_ON(charge)
> isn;t meaningless.
> 
> Cc: Johannes Weiner <hannes@cmpxchg.org>
> Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
> Signed-off-by: Minchan Kim <minchan.kim@gmail.com>

Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>


> ---
>  mm/memcontrol.c |    1 +
>  mm/migrate.c    |   14 ++++----------
>  2 files changed, 5 insertions(+), 10 deletions(-)
> 
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 2fc97fc..6832926 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -2872,6 +2872,7 @@ static inline int mem_cgroup_move_swap_account(swp_entry_t entry,
>  /*
>   * Before starting migration, account PAGE_SIZE to mem_cgroup that the old
>   * page belongs to.
> + * Return 0 if charge is successful. Otherwise return -errno.
>   */
>  int mem_cgroup_prepare_migration(struct page *page,
>  	struct page *newpage, struct mem_cgroup **ptr, gfp_t gfp_mask)
> diff --git a/mm/migrate.c b/mm/migrate.c
> index eb083a6..737c2e5 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -622,7 +622,6 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
>  	int *result = NULL;
>  	struct page *newpage = get_new_page(page, private, &result);
>  	int remap_swapcache = 1;
> -	int charge = 0;
>  	struct mem_cgroup *mem;
>  	struct anon_vma *anon_vma = NULL;
>  
> @@ -637,9 +636,7 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
>  		if (unlikely(split_huge_page(page)))
>  			goto move_newpage;
>  
> -	/* prepare cgroup just returns 0 or -ENOMEM */
>  	rc = -EAGAIN;
> -
>  	if (!trylock_page(page)) {
>  		if (!force)
>  			goto move_newpage;
> @@ -678,13 +675,11 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
>  	}
>  
>  	/* charge against new page */
> -	charge = mem_cgroup_prepare_migration(page, newpage, &mem, GFP_KERNEL);
> -	if (charge == -ENOMEM) {
> -		rc = -ENOMEM;
> +	rc = mem_cgroup_prepare_migration(page, newpage, &mem, GFP_KERNEL);
> +	if (rc)
>  		goto unlock;
> -	}
> -	BUG_ON(charge);
>  
> +	rc = -EAGAIN;
>  	if (PageWriteback(page)) {
>  		if (!force || !sync)
>  			goto uncharge;
> @@ -767,8 +762,7 @@ skip_unmap:
>  		drop_anon_vma(anon_vma);
>  
>  uncharge:
> -	if (!charge)
> -		mem_cgroup_end_migration(mem, page, newpage, rc == 0);
> +	mem_cgroup_end_migration(mem, page, newpage, rc == 0);
>  unlock:
>  	unlock_page(page);
>  
> -- 
> 1.7.1
> 
> 

--
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] memcg: clean up migration

[Daisuke Nishimura]

On Mon, 28 Feb 2011 00:49:25 +0900
Minchan Kim <minchan.kim@gmail.com> wrote:

> This patch cleans up unncessary BUG_ON check and confusing
> charge variable.
> 
> That's because memcg charge/uncharge could be handled by
> mem_cgroup_[prepare/end] migration itself so charge local variable
> in unmap_and_move lost the role since we introduced 01b1ae63c2.
> 
> And mem_cgroup_prepare_migratio return 0 if only it is successful.
> Otherwise, it jumps to unlock label to clean up so BUG_ON(charge)
> isn;t meaningless.
> 
> Cc: Johannes Weiner <hannes@cmpxchg.org>
> Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
> Signed-off-by: Minchan Kim <minchan.kim@gmail.com>

It looks good to me, but I have one minor comment.

> ---
>  mm/memcontrol.c |    1 +
>  mm/migrate.c    |   14 ++++----------
>  2 files changed, 5 insertions(+), 10 deletions(-)
> 
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 2fc97fc..6832926 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -2872,6 +2872,7 @@ static inline int mem_cgroup_move_swap_account(swp_entry_t entry,
>  /*
>   * Before starting migration, account PAGE_SIZE to mem_cgroup that the old
>   * page belongs to.
> + * Return 0 if charge is successful. Otherwise return -errno.
>   */
>  int mem_cgroup_prepare_migration(struct page *page,
>  	struct page *newpage, struct mem_cgroup **ptr, gfp_t gfp_mask)
> diff --git a/mm/migrate.c b/mm/migrate.c
> index eb083a6..737c2e5 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -622,7 +622,6 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
>  	int *result = NULL;
>  	struct page *newpage = get_new_page(page, private, &result);
>  	int remap_swapcache = 1;
> -	int charge = 0;
>  	struct mem_cgroup *mem;
>  	struct anon_vma *anon_vma = NULL;
>  
> @@ -637,9 +636,7 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
>  		if (unlikely(split_huge_page(page)))
>  			goto move_newpage;
>  
> -	/* prepare cgroup just returns 0 or -ENOMEM */
>  	rc = -EAGAIN;
> -
>  	if (!trylock_page(page)) {
>  		if (!force)
>  			goto move_newpage;
> @@ -678,13 +675,11 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
>  	}
>  
>  	/* charge against new page */
> -	charge = mem_cgroup_prepare_migration(page, newpage, &mem, GFP_KERNEL);
> -	if (charge == -ENOMEM) {
> -		rc = -ENOMEM;
> +	rc = mem_cgroup_prepare_migration(page, newpage, &mem, GFP_KERNEL);
> +	if (rc)
>  		goto unlock;
> -	}
> -	BUG_ON(charge);
>  
> +	rc = -EAGAIN;
>  	if (PageWriteback(page)) {
>  		if (!force || !sync)
>  			goto uncharge;
How about

	if (mem_cgroup_prepare_migration(..)) {
		rc = -ENOMEM;
		goto unlock;
	}

?

Re-setting "rc" to -EAGAIN is not necessary in this case.
"if (mem_cgroup_...)" is commonly used in many places.

Anyway,

	Acked-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>


Thanks,
Daisuke Nishimura.

> @@ -767,8 +762,7 @@ skip_unmap:
>  		drop_anon_vma(anon_vma);
>  
>  uncharge:
> -	if (!charge)
> -		mem_cgroup_end_migration(mem, page, newpage, rc == 0);
> +	mem_cgroup_end_migration(mem, page, newpage, rc == 0);
>  unlock:
>  	unlock_page(page);
>  
> -- 
> 1.7.1
> 

--
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] memcg: clean up migration

[Minchan Kim]

Hi, Daisuke

On Mon, Feb 28, 2011 at 11:18 AM, Daisuke Nishimura
<nishimura@mxp.nes.nec.co.jp> wrote:
> On Mon, 28 Feb 2011 00:49:25 +0900
> Minchan Kim <minchan.kim@gmail.com> wrote:
>
>> This patch cleans up unncessary BUG_ON check and confusing
>> charge variable.
>>
>> That's because memcg charge/uncharge could be handled by
>> mem_cgroup_[prepare/end] migration itself so charge local variable
>> in unmap_and_move lost the role since we introduced 01b1ae63c2.
>>
>> And mem_cgroup_prepare_migratio return 0 if only it is successful.
>> Otherwise, it jumps to unlock label to clean up so BUG_ON(charge)
>> isn;t meaningless.
>>
>> Cc: Johannes Weiner <hannes@cmpxchg.org>
>> Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
>> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
>> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
>> Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
>
> It looks good to me, but I have one minor comment.
>
>> ---
>>  mm/memcontrol.c |    1 +
>>  mm/migrate.c    |   14 ++++----------
>>  2 files changed, 5 insertions(+), 10 deletions(-)
>>
>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
>> index 2fc97fc..6832926 100644
>> --- a/mm/memcontrol.c
>> +++ b/mm/memcontrol.c
>> @@ -2872,6 +2872,7 @@ static inline int mem_cgroup_move_swap_account(swp_entry_t entry,
>>  /*
>>   * Before starting migration, account PAGE_SIZE to mem_cgroup that the old
>>   * page belongs to.
>> + * Return 0 if charge is successful. Otherwise return -errno.
>>   */
>>  int mem_cgroup_prepare_migration(struct page *page,
>>       struct page *newpage, struct mem_cgroup **ptr, gfp_t gfp_mask)
>> diff --git a/mm/migrate.c b/mm/migrate.c
>> index eb083a6..737c2e5 100644
>> --- a/mm/migrate.c
>> +++ b/mm/migrate.c
>> @@ -622,7 +622,6 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
>>       int *result = NULL;
>>       struct page *newpage = get_new_page(page, private, &result);
>>       int remap_swapcache = 1;
>> -     int charge = 0;
>>       struct mem_cgroup *mem;
>>       struct anon_vma *anon_vma = NULL;
>>
>> @@ -637,9 +636,7 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
>>               if (unlikely(split_huge_page(page)))
>>                       goto move_newpage;
>>
>> -     /* prepare cgroup just returns 0 or -ENOMEM */
>>       rc = -EAGAIN;
>> -
>>       if (!trylock_page(page)) {
>>               if (!force)
>>                       goto move_newpage;
>> @@ -678,13 +675,11 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
>>       }
>>
>>       /* charge against new page */
>> -     charge = mem_cgroup_prepare_migration(page, newpage, &mem, GFP_KERNEL);
>> -     if (charge == -ENOMEM) {
>> -             rc = -ENOMEM;
>> +     rc = mem_cgroup_prepare_migration(page, newpage, &mem, GFP_KERNEL);
>> +     if (rc)
>>               goto unlock;
>> -     }
>> -     BUG_ON(charge);
>>
>> +     rc = -EAGAIN;
>>       if (PageWriteback(page)) {
>>               if (!force || !sync)
>>                       goto uncharge;
> How about
>
>        if (mem_cgroup_prepare_migration(..)) {
>                rc = -ENOMEM;
>                goto unlock;
>        }
>
> ?
>
> Re-setting "rc" to -EAGAIN is not necessary in this case.
> "if (mem_cgroup_...)" is commonly used in many places.
>
It works now but Johannes doesn't like it and me, either.
It makes unnecessary dependency which mem_cgroup_preparre_migration
can't propagate error to migrate_pages.
Although we don't need it, I want to remove such unnecessary dependency.

> Anyway,
>
>        Acked-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
>
>

Thanks!.

-- 
Kind regards,
Minchan Kim

--
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] memcg: clean up migration

[Daisuke Nishimura]

> >> @@ -678,13 +675,11 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
> >> A  A  A  }
> >>
> >> A  A  A  /* charge against new page */
> >> - A  A  charge = mem_cgroup_prepare_migration(page, newpage, &mem, GFP_KERNEL);
> >> - A  A  if (charge == -ENOMEM) {
> >> - A  A  A  A  A  A  rc = -ENOMEM;
> >> + A  A  rc = mem_cgroup_prepare_migration(page, newpage, &mem, GFP_KERNEL);
> >> + A  A  if (rc)
> >> A  A  A  A  A  A  A  goto unlock;
> >> - A  A  }
> >> - A  A  BUG_ON(charge);
> >>
> >> + A  A  rc = -EAGAIN;
> >> A  A  A  if (PageWriteback(page)) {
> >> A  A  A  A  A  A  A  if (!force || !sync)
> >> A  A  A  A  A  A  A  A  A  A  A  goto uncharge;
> > How about
> >
> > A  A  A  A if (mem_cgroup_prepare_migration(..)) {
> > A  A  A  A  A  A  A  A rc = -ENOMEM;
> > A  A  A  A  A  A  A  A goto unlock;
> > A  A  A  A }
> >
> > ?
> >
> > Re-setting "rc" to -EAGAIN is not necessary in this case.
> > "if (mem_cgroup_...)" is commonly used in many places.
> >
> It works now but Johannes doesn't like it and me, either.
> It makes unnecessary dependency which mem_cgroup_preparre_migration
> can't propagate error to migrate_pages.
> Although we don't need it, I want to remove such unnecessary dependency.
> 
I see.
Thank you for your explanation.

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>