linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@linux-foundation.org>
To: Johannes Stezenbach <js@sig21.net>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
	Thomas Gleixner <tglx@linutronix.de>
Subject: Re: huge debug_objects_cache. swapping but 25% mem free
Date: Fri, 3 Feb 2012 15:12:18 -0800	[thread overview]
Message-ID: <20120203151218.59ceb1c1.akpm@linux-foundation.org> (raw)
In-Reply-To: <20120201182932.GA15518@sig21.net>

On Wed, 1 Feb 2012 19:29:32 +0100
Johannes Stezenbach <js@sig21.net> wrote:

> On Mon, Jan 30, 2012 at 04:40:48PM +0100, Johannes Stezenbach wrote:
> > 
> > According to slabtop debug_objects_cache eats half the
> > memory, is this expected?  The Kconfig help text does not
> > suggest it.
> 
> FWIW, today after running full backup using rsync:
> 
>   OBJS ACTIVE  USE OBJ SIZE  SLABS OBJ/SLAB CACHE SIZE NAME
> 4025450 4025447  99%    0.36K 183977       22   1471816K debug_objects_cache
> 170266  87241  51%    1.74K   9645       18    308640K ext3_inode_cache
>  54855  40215  73%    0.58K   2034       27     32544K dentry
>  38124  12474  32%    0.42K   2119       18     16952K buffer_head
>  17794  17464  98%    0.51K    574       31      9184K sysfs_dir_cache
>  11434   8494  74%    0.87K    713       18     11408K radix_tree_node
>  11400   5570  48%    0.38K    570       20      4560K kmalloc-64
> 
> /sys/kernel/debug/debug_objects/stats:
> max_chain     :870
> warnings      :0
> fixups        :0
> pool_free     :258
> pool_min_free :252
> pool_used     :4025193
> pool_max_used :4025195
> 
> 
> then after "echo 3 >/proc/sys/vm/drop_caches":
> 
>   OBJS ACTIVE  USE OBJ SIZE  SLABS OBJ/SLAB CACHE SIZE NAME
> 4030488 4030485  99%    0.36K 184206       22   1473648K debug_objects_cache
>  17955   9071  50%    0.58K    665       27     10640K dentry
>  17763  17455  98%    0.51K    573       31      9168K sysfs_dir_cache
>   9408   9156  97%    0.33K    392       24      3136K kmalloc-8
> 
> 
> I'm still not sure if this is normal behaviour of DEBUG_OBJECTS or a bug?

Bug.  It appears that you hadn't been using CONFIG_DEBUG_OBJECTS before
3.2 so we don't know if it is a regression.


      reply	other threads:[~2012-02-03 23:12 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-01-30 15:40 huge debug_objects_cache. swapping but 25% mem free Johannes Stezenbach
2012-02-01 18:29 ` Johannes Stezenbach
2012-02-03 23:12   ` Andrew Morton [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20120203151218.59ceb1c1.akpm@linux-foundation.org \
    --to=akpm@linux-foundation.org \
    --cc=js@sig21.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=tglx@linutronix.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).