linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
* [BUGFIX][PATCH mmotm] memcg: try_get_mem_cgroup_from_swapcache fix
@ 2009-03-22 15:02 Daisuke Nishimura
  2009-03-22 18:40 ` Balbir Singh
  2009-03-23  0:03 ` KAMEZAWA Hiroyuki
  0 siblings, 2 replies; 5+ messages in thread
From: Daisuke Nishimura @ 2009-03-22 15:02 UTC (permalink / raw)
  To: Andrew Morton
  Cc: Balbir Singh, KAMEZAWA Hiroyuki, Daisuke Nishimura, linux-mm

From: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>

css_tryget can be called twice in !PageCgroupUsed case.

Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
---
This is a fix for cgroups-use-css-id-in-swap-cgroup-for-saving-memory-v5.patch

 mm/memcontrol.c |   10 ++++------
 1 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 5de6be9..55dea59 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -1027,9 +1027,11 @@ static struct mem_cgroup *try_get_mem_cgroup_from_swapcache(struct page *page)
 	/*
 	 * Used bit of swapcache is solid under page lock.
 	 */
-	if (PageCgroupUsed(pc))
+	if (PageCgroupUsed(pc)) {
 		mem = pc->mem_cgroup;
-	else {
+		if (mem && !css_tryget(&mem->css))
+			mem = NULL;
+	} else {
 		ent.val = page_private(page);
 		id = lookup_swap_cgroup(ent);
 		rcu_read_lock();
@@ -1038,10 +1040,6 @@ static struct mem_cgroup *try_get_mem_cgroup_from_swapcache(struct page *page)
 			mem = NULL;
 		rcu_read_unlock();
 	}
-	if (!mem)
-		return NULL;
-	if (!css_tryget(&mem->css))
-		return NULL;
 	return mem;
 }
 



--
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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [BUGFIX][PATCH mmotm] memcg: try_get_mem_cgroup_from_swapcache fix
  2009-03-22 15:02 [BUGFIX][PATCH mmotm] memcg: try_get_mem_cgroup_from_swapcache fix Daisuke Nishimura
@ 2009-03-22 18:40 ` Balbir Singh
  2009-03-23  0:13   ` Daisuke Nishimura
  2009-03-23  0:03 ` KAMEZAWA Hiroyuki
  1 sibling, 1 reply; 5+ messages in thread
From: Balbir Singh @ 2009-03-22 18:40 UTC (permalink / raw)
  To: nishimura; +Cc: Andrew Morton, KAMEZAWA Hiroyuki, linux-mm

* Daisuke Nishimura <d-nishimura@mtf.biglobe.ne.jp> [2009-03-23 00:02:38]:

> From: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
> 
> css_tryget can be called twice in !PageCgroupUsed case.
> 
> Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
> ---
> This is a fix for cgroups-use-css-id-in-swap-cgroup-for-saving-memory-v5.patch
> 
>  mm/memcontrol.c |   10 ++++------
>  1 files changed, 4 insertions(+), 6 deletions(-)
> 
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 5de6be9..55dea59 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -1027,9 +1027,11 @@ static struct mem_cgroup *try_get_mem_cgroup_from_swapcache(struct page *page)
>  	/*
>  	 * Used bit of swapcache is solid under page lock.
>  	 */
> -	if (PageCgroupUsed(pc))
> +	if (PageCgroupUsed(pc)) {
>  		mem = pc->mem_cgroup;
> -	else {
> +		if (mem && !css_tryget(&mem->css))
> +			mem = NULL;
> +	} else {
>  		ent.val = page_private(page);
>  		id = lookup_swap_cgroup(ent);
>  		rcu_read_lock();
> @@ -1038,10 +1040,6 @@ static struct mem_cgroup *try_get_mem_cgroup_from_swapcache(struct page *page)
>  			mem = NULL;
>  		rcu_read_unlock();
>  	}
> -	if (!mem)
> -		return NULL;
> -	if (!css_tryget(&mem->css))
> -		return NULL;
>  	return mem;
>  }

How did you detect the problem? Any test case/steps to reproduce the issue?


-- 
	Balbir

--
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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [BUGFIX][PATCH mmotm] memcg: try_get_mem_cgroup_from_swapcache fix
  2009-03-22 15:02 [BUGFIX][PATCH mmotm] memcg: try_get_mem_cgroup_from_swapcache fix Daisuke Nishimura
  2009-03-22 18:40 ` Balbir Singh
@ 2009-03-23  0:03 ` KAMEZAWA Hiroyuki
  1 sibling, 0 replies; 5+ messages in thread
From: KAMEZAWA Hiroyuki @ 2009-03-23  0:03 UTC (permalink / raw)
  To: nishimura; +Cc: Daisuke Nishimura, Andrew Morton, Balbir Singh, linux-mm

On Mon, 23 Mar 2009 00:02:38 +0900
Daisuke Nishimura <d-nishimura@mtf.biglobe.ne.jp> wrote:

> From: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
> 
> css_tryget can be called twice in !PageCgroupUsed case.
> 
> Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>

Thank you
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujtsu.com>

> ---
> This is a fix for cgroups-use-css-id-in-swap-cgroup-for-saving-memory-v5.patch
> 
>  mm/memcontrol.c |   10 ++++------
>  1 files changed, 4 insertions(+), 6 deletions(-)
> 
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 5de6be9..55dea59 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -1027,9 +1027,11 @@ static struct mem_cgroup *try_get_mem_cgroup_from_swapcache(struct page *page)
>  	/*
>  	 * Used bit of swapcache is solid under page lock.
>  	 */
> -	if (PageCgroupUsed(pc))
> +	if (PageCgroupUsed(pc)) {
>  		mem = pc->mem_cgroup;
> -	else {
> +		if (mem && !css_tryget(&mem->css))
> +			mem = NULL;
> +	} else {
>  		ent.val = page_private(page);
>  		id = lookup_swap_cgroup(ent);
>  		rcu_read_lock();
> @@ -1038,10 +1040,6 @@ static struct mem_cgroup *try_get_mem_cgroup_from_swapcache(struct page *page)
>  			mem = NULL;
>  		rcu_read_unlock();
>  	}
> -	if (!mem)
> -		return NULL;
> -	if (!css_tryget(&mem->css))
> -		return NULL;
>  	return mem;
>  }
>  
> 
> 
> 
> 

--
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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [BUGFIX][PATCH mmotm] memcg: try_get_mem_cgroup_from_swapcache fix
  2009-03-22 18:40 ` Balbir Singh
@ 2009-03-23  0:13   ` Daisuke Nishimura
  2009-03-23  0:19     ` KAMEZAWA Hiroyuki
  0 siblings, 1 reply; 5+ messages in thread
From: Daisuke Nishimura @ 2009-03-23  0:13 UTC (permalink / raw)
  To: balbir; +Cc: nishimura, Andrew Morton, KAMEZAWA Hiroyuki, linux-mm

On Mon, 23 Mar 2009 00:10:15 +0530, Balbir Singh <balbir@linux.vnet.ibm.com> wrote:
> * Daisuke Nishimura <d-nishimura@mtf.biglobe.ne.jp> [2009-03-23 00:02:38]:
> 
> > From: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
> > 
> > css_tryget can be called twice in !PageCgroupUsed case.
> > 
> > Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
> > ---
> > This is a fix for cgroups-use-css-id-in-swap-cgroup-for-saving-memory-v5.patch
> > 
> >  mm/memcontrol.c |   10 ++++------
> >  1 files changed, 4 insertions(+), 6 deletions(-)
> > 
> > diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> > index 5de6be9..55dea59 100644
> > --- a/mm/memcontrol.c
> > +++ b/mm/memcontrol.c
> > @@ -1027,9 +1027,11 @@ static struct mem_cgroup *try_get_mem_cgroup_from_swapcache(struct page *page)
> >  	/*
> >  	 * Used bit of swapcache is solid under page lock.
> >  	 */
> > -	if (PageCgroupUsed(pc))
> > +	if (PageCgroupUsed(pc)) {
> >  		mem = pc->mem_cgroup;
> > -	else {
> > +		if (mem && !css_tryget(&mem->css))
> > +			mem = NULL;
> > +	} else {
> >  		ent.val = page_private(page);
> >  		id = lookup_swap_cgroup(ent);
> >  		rcu_read_lock();
> > @@ -1038,10 +1040,6 @@ static struct mem_cgroup *try_get_mem_cgroup_from_swapcache(struct page *page)
> >  			mem = NULL;
> >  		rcu_read_unlock();
> >  	}
> > -	if (!mem)
> > -		return NULL;
> > -	if (!css_tryget(&mem->css))
> > -		return NULL;
> >  	return mem;
> >  }
> 
> How did you detect the problem? Any test case/steps to reproduce the issue?
> 
I found this when rebasing my patch onto mmotm and reviewing it.

I suppose this bug can leads to an unremovable directory.


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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [BUGFIX][PATCH mmotm] memcg: try_get_mem_cgroup_from_swapcache fix
  2009-03-23  0:13   ` Daisuke Nishimura
@ 2009-03-23  0:19     ` KAMEZAWA Hiroyuki
  0 siblings, 0 replies; 5+ messages in thread
From: KAMEZAWA Hiroyuki @ 2009-03-23  0:19 UTC (permalink / raw)
  To: Daisuke Nishimura; +Cc: balbir, Andrew Morton, linux-mm

On Mon, 23 Mar 2009 09:13:37 +0900
Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> wrote:

> On Mon, 23 Mar 2009 00:10:15 +0530, Balbir Singh <balbir@linux.vnet.ibm.com> wrote:
> > * Daisuke Nishimura <d-nishimura@mtf.biglobe.ne.jp> [2009-03-23 00:02:38]:
> > 
> > > From: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
> > > 
> > > css_tryget can be called twice in !PageCgroupUsed case.
> > > 
> > > Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
> > > ---
> > > This is a fix for cgroups-use-css-id-in-swap-cgroup-for-saving-memory-v5.patch
> > > 
> > >  mm/memcontrol.c |   10 ++++------
> > >  1 files changed, 4 insertions(+), 6 deletions(-)
> > > 
> > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> > > index 5de6be9..55dea59 100644
> > > --- a/mm/memcontrol.c
> > > +++ b/mm/memcontrol.c
> > > @@ -1027,9 +1027,11 @@ static struct mem_cgroup *try_get_mem_cgroup_from_swapcache(struct page *page)
> > >  	/*
> > >  	 * Used bit of swapcache is solid under page lock.
> > >  	 */
> > > -	if (PageCgroupUsed(pc))
> > > +	if (PageCgroupUsed(pc)) {
> > >  		mem = pc->mem_cgroup;
> > > -	else {
> > > +		if (mem && !css_tryget(&mem->css))
> > > +			mem = NULL;
> > > +	} else {
> > >  		ent.val = page_private(page);
> > >  		id = lookup_swap_cgroup(ent);
> > >  		rcu_read_lock();
> > > @@ -1038,10 +1040,6 @@ static struct mem_cgroup *try_get_mem_cgroup_from_swapcache(struct page *page)
> > >  			mem = NULL;
> > >  		rcu_read_unlock();
> > >  	}
> > > -	if (!mem)
> > > -		return NULL;
> > > -	if (!css_tryget(&mem->css))
> > > -		return NULL;
> > >  	return mem;
> > >  }
> > 
> > How did you detect the problem? Any test case/steps to reproduce the issue?
> > 
> I found this when rebasing my patch onto mmotm and reviewing it.
> 
> I suppose this bug can leads to an unremovable directory.
> 

I think I saw this in the weekend but couldn't find why ;(
Thank you!
-Kame

> 
> 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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2009-03-22 23:27 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-03-22 15:02 [BUGFIX][PATCH mmotm] memcg: try_get_mem_cgroup_from_swapcache fix Daisuke Nishimura
2009-03-22 18:40 ` Balbir Singh
2009-03-23  0:13   ` Daisuke Nishimura
2009-03-23  0:19     ` KAMEZAWA Hiroyuki
2009-03-23  0:03 ` KAMEZAWA Hiroyuki

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox