Re: [PATCH] slub Discard slab page only when node partials > minimum setting

["Alex,Shi" <alex.shi@intel.com>]

On Thu, 2011-09-15 at 13:48 +0800, Pekka Enberg wrote:
> On Wed, Sep 7, 2011 at 4:03 AM, Alex,Shi <alex.shi@intel.com> wrote:
> > Unfreeze_partials may try to discard slab page, the discarding condition
> > should be 'when node partials number > minimum partial number setting',
> > not '<' in current code.
> >
> > This patch base on penberg's tree's 'slub/partial' head.
> >
> > git://git.kernel.org/pub/scm/linux/kernel/git/penberg/slab-2.6.git
> >
> > Signed-off-by: Alex Shi <alex.shi@intel.com>
> >
> > ---
> >  mm/slub.c |    2 +-
> >  1 files changed, 1 insertions(+), 1 deletions(-)
> >
> > diff --git a/mm/slub.c b/mm/slub.c
> > index b351480..66a5b29 100644
> > --- a/mm/slub.c
> > +++ b/mm/slub.c
> > @@ -1954,7 +1954,7 @@ static void unfreeze_partials(struct kmem_cache *s)
> >
> >                        new.frozen = 0;
> >
> > -                       if (!new.inuse && (!n || n->nr_partial < s->min_partial))
> > +                       if (!new.inuse && (!n || n->nr_partial > s->min_partial))
> >                                m = M_FREE;
> >                        else {
> >                                struct kmem_cache_node *n2 = get_node(s,
> 
> Can you please resend the patch with Christoph's ACK and a better
> explanation why the condition needs to be flipped. A reference to
> commit 81107188f123e3c2217ac2f2feb2a1147904c62f ("slub: Fix partial
> count comparison confusion") is probably sufficient.
> 
> P.S. Please use the penberg@cs.helsinki.fi email address for now.
> 
>                         Pekka

Is the following OK? Pekka. :) 

==========
From: Alex Shi <alex.shi@intel.com>
Date: Tue, 6 Sep 2011 14:46:01 +0800
Subject: [PATCH ] Discard slab page when node partial > mininum partial number

Unfreeze_partials will try to discard empty slab pages when the slab
node partial number is greater than s->min_partial, not less than
s->min_partial. Otherwise the empty slab page will keep growing and eat
up all system memory.

Signed-off-by: Alex Shi <alex.shi@intel.com>
Acked-by: Christoph Lameter <cl@linux.com>
---
 mm/slub.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/mm/slub.c b/mm/slub.c
index 1348c09..492beab 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1953,7 +1953,7 @@ static void unfreeze_partials(struct kmem_cache *s)
 
 			new.frozen = 0;
 
-			if (!new.inuse && (!n || n->nr_partial < s->min_partial))
+			if (!new.inuse && (!n || n->nr_partial > s->min_partial))
 				m = M_FREE;
 			else {
 				struct kmem_cache_node *n2 = get_node(s,
-- 
1.7.0