linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: mel@skynet.ie (Mel Gorman)
To: Christoph Lameter <clameter@sgi.com>
Cc: akpm@linux-foundatin.org, linux-kernel@vger.kernel.org,
	linux-mm@kvack.org
Subject: Re: [patch 07/23] SLUB: Add defrag_ratio field and sysfs support.
Date: Thu, 8 Nov 2007 15:07:05 +0000	[thread overview]
Message-ID: <20071108150705.GD2591@skynet.ie> (raw)
In-Reply-To: <20071107011228.102370371@sgi.com>

On (06/11/07 17:11), Christoph Lameter didst pronounce:
> The defrag_ratio is used to set the threshold at which defragmentation
> should be run on a slabcache.
> 

I'm thick, I would like to see a quick note here on what defragmentation
means. Also, this defrag_ratio seems to have a significantly different
meaning to the other defrag_ratio which isn't helping my poor head at
all.

"The defrag_ratio sets a threshold at which a slab will be vacated of all
it's objects and the pages freed during memory reclaim."

?

> The allocation ratio is measured in a percentage of the available slots.
> The percentage will be lower for slabs that are more fragmented.
> 
> Add a defrag ratio field and set it to 30% by default. A limit of 30% specified
> that less than 3 out of 10 available slots for objects are in use before
> reclaim occurs.
> 
> Reviewed-by: Rik van Riel <riel@redhat.com>
> Signed-off-by: Christoph Lameter <clameter@sgi.com>
> ---
>  include/linux/slub_def.h |    7 +++++++
>  mm/slub.c                |   18 ++++++++++++++++++
>  2 files changed, 25 insertions(+)
> 
> Index: linux-2.6/include/linux/slub_def.h
> ===================================================================
> --- linux-2.6.orig/include/linux/slub_def.h	2007-11-06 12:36:28.000000000 -0800
> +++ linux-2.6/include/linux/slub_def.h	2007-11-06 12:37:44.000000000 -0800
> @@ -53,6 +53,13 @@ struct kmem_cache {
>  	void (*ctor)(struct kmem_cache *, void *);
>  	int inuse;		/* Offset to metadata */
>  	int align;		/* Alignment */
> +	int defrag_ratio;	/*
> +				 * objects/possible-objects limit. If we have
> +				 * less that the specified percentage of
> +				 * objects allocated then defrag passes
> +				 * will start to occur during reclaim.
> +				 */
> +
>  	const char *name;	/* Name (only for display!) */
>  	struct list_head list;	/* List of slab caches */
>  #ifdef CONFIG_SLUB_DEBUG
> Index: linux-2.6/mm/slub.c
> ===================================================================
> --- linux-2.6.orig/mm/slub.c	2007-11-06 12:37:25.000000000 -0800
> +++ linux-2.6/mm/slub.c	2007-11-06 12:37:44.000000000 -0800
> @@ -2363,6 +2363,7 @@ static int kmem_cache_open(struct kmem_c
>  		goto error;
>  
>  	s->refcount = 1;
> +	s->defrag_ratio = 30;
>  #ifdef CONFIG_NUMA
>  	s->remote_node_defrag_ratio = 100;
>  #endif
> @@ -4005,6 +4006,22 @@ static ssize_t free_calls_show(struct km
>  }
>  SLAB_ATTR_RO(free_calls);
>  
> +static ssize_t defrag_ratio_show(struct kmem_cache *s, char *buf)
> +{
> +	return sprintf(buf, "%d\n", s->defrag_ratio);
> +}
> +
> +static ssize_t defrag_ratio_store(struct kmem_cache *s,
> +				const char *buf, size_t length)
> +{
> +	int n = simple_strtoul(buf, NULL, 10);
> +
> +	if (n < 100)
> +		s->defrag_ratio = n;
> +	return length;
> +}
> +SLAB_ATTR(defrag_ratio);
> +
>  #ifdef CONFIG_NUMA
>  static ssize_t remote_node_defrag_ratio_show(struct kmem_cache *s, char *buf)
>  {
> @@ -4047,6 +4064,7 @@ static struct attribute * slab_attrs[] =
>  	&shrink_attr.attr,
>  	&alloc_calls_attr.attr,
>  	&free_calls_attr.attr,
> +	&defrag_ratio_attr.attr,
>  #ifdef CONFIG_ZONE_DMA
>  	&cache_dma_attr.attr,
>  #endif
> 
> -- 
> 

-- 
-- 
Mel Gorman
Part-time Phd Student                          Linux Technology Center
University of Limerick                         IBM Dublin Software Lab

--
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>

  parent reply	other threads:[~2007-11-08 15:07 UTC|newest]

Thread overview: 77+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-11-07  1:11 [patch 00/23] Slab defragmentation V6 Christoph Lameter
2007-11-07  1:11 ` [patch 01/23] SLUB: Move count_partial() Christoph Lameter
2007-11-07  1:11 ` [patch 02/23] SLUB: Rename NUMA defrag_ratio to remote_node_defrag_ratio Christoph Lameter
2007-11-08 14:50   ` Mel Gorman
2007-11-08 17:25     ` Matt Mackall
2007-11-08 19:16       ` Christoph Lameter
2007-11-08 19:47         ` Matt Mackall
2007-11-08 20:01           ` Christoph Lameter
2007-11-08 21:03             ` Matt Mackall
2007-11-08 21:28               ` Christoph Lameter
2007-11-08 23:08                 ` Matt Mackall
2007-11-08 18:56     ` Christoph Lameter
2007-11-08 20:10       ` Mel Gorman
2007-11-07  1:11 ` [patch 03/23] bufferhead: Revert constructor removal Christoph Lameter
2007-11-07  1:11 ` [patch 04/23] dentries: Extract common code to remove dentry from lru Christoph Lameter
2007-11-07  8:50   ` Johannes Weiner
2007-11-07  9:43     ` Jörn Engel
2007-11-07 18:55       ` Christoph Lameter
2007-11-07 18:54         ` Jörn Engel
2007-11-07 19:00           ` Christoph Lameter
2007-11-07 18:28     ` Christoph Lameter
2007-11-07  1:11 ` [patch 05/23] VM: Allow get_page_unless_zero on compound pages Christoph Lameter
2007-11-07  1:11 ` [patch 06/23] SLUB: Extend slabinfo to support -D and -C options Christoph Lameter
2007-11-08 15:00   ` Mel Gorman
2007-11-07  1:11 ` [patch 07/23] SLUB: Add defrag_ratio field and sysfs support Christoph Lameter
2007-11-07  8:55   ` Johannes Weiner
2007-11-07 18:30     ` Christoph Lameter
2007-11-08 15:07   ` Mel Gorman [this message]
2007-11-08 18:59     ` Christoph Lameter
2007-11-07  1:11 ` [patch 08/23] SLUB: Replace ctor field with ops field in /sys/slab/:0000008 /sys/slab/:0000016 /sys/slab/:0000024 /sys/slab/:0000032 /sys/slab/:0000040 /sys/slab/:0000048 /sys/slab/:0000056 /sys/slab/:0000064 /sys/slab/:0000072 /sys/slab/:0000080 /sys/slab/:0000088 /sys/slab/:0000096 /sys/slab/:0000104 /sys/slab/:0000128 /sys/slab/:0000144 /sys/slab/:0000184 /sys/slab/:0000192 /sys/slab/:0000216 /sys/slab/:0000256 /sys/slab/:0000344 /sys/slab/:0000384 /sys/slab/:0000448 /sys/slab/:0000512 /sys/slab/:0000768 /sys/slab/:0000968 /sys/slab/:0001024 /sys/slab/:0001152 /sys/slab/:0001312 /sys/slab/:0001536 /sys/slab/:0002048 /sys/slab/:0003072 /sys/slab/:0004096 /sys/slab/:a-0000016 /sys/slab/:a-0000024 /sys/slab/:a-0000056 /sys/slab/:a-0000080 /sys/slab/:a-0000128 /sys/slab/Acpi-Namesp ace /sys/slab/Acpi-Operand /sys/slab/Acpi-Parse /sys/slab/Acpi-ParseExt /sys/slab/Acpi-State /sys/ Christoph Lameter
2007-11-07  1:11 ` [patch 09/23] SLUB: Add get() and kick() methods Christoph Lameter
2007-11-07  2:37   ` Adrian Bunk
2007-11-07  3:07     ` Christoph Lameter
2007-11-07  3:26       ` Adrian Bunk
2007-11-07  1:11 ` [patch 10/23] SLUB: Sort slab cache list and establish maximum objects for defrag slabs Christoph Lameter
2007-11-07  1:11 ` [patch 11/23] SLUB: Slab defrag core Christoph Lameter
2007-11-07 22:13   ` Christoph Lameter
2007-11-07  1:11 ` [patch 12/23] SLUB: Trigger defragmentation from memory reclaim Christoph Lameter
2007-11-07  9:28   ` Johannes Weiner
2007-11-07 18:34     ` Christoph Lameter
2007-11-08 15:12   ` Mel Gorman
2007-11-08 19:00     ` Christoph Lameter
2007-11-07  1:11 ` [patch 13/23] Buffer heads: Support slab defrag Christoph Lameter
2007-11-07  1:11 ` [patch 14/23] inodes: Support generic defragmentation Christoph Lameter
2007-11-07 10:17   ` Jörn Engel
2007-11-07 10:31     ` Jörn Engel
2007-11-07 10:35     ` Andreas Schwab
2007-11-07 10:35       ` Jörn Engel
2007-11-07 18:40     ` Christoph Lameter
2007-11-07 18:51       ` Jörn Engel
2007-11-07 19:00         ` Christoph Lameter
2007-11-07  1:11 ` [patch 15/23] FS: ExtX filesystem defrag Christoph Lameter
2007-11-07  1:11 ` [patch 16/23] FS: XFS slab defragmentation Christoph Lameter
2007-11-07  1:11 ` [patch 17/23] FS: Proc filesystem support for slab defrag Christoph Lameter
2007-11-07  1:11 ` [patch 18/23] FS: Slab defrag: Reiserfs support Christoph Lameter
2007-11-07  1:11 ` [patch 19/23] FS: Socket inode defragmentation Christoph Lameter
2007-11-07  1:11 ` [patch 20/23] dentries: Add constructor Christoph Lameter
2007-11-08 15:23   ` Mel Gorman
2007-11-08 19:03     ` Christoph Lameter
2007-11-07  1:11 ` [patch 21/23] dentries: dentry defragmentation Christoph Lameter
2007-11-07  1:11 ` [patch 22/23] SLUB: Slab reclaim through Lumpy reclaim Christoph Lameter
2007-11-07  1:11 ` [patch 23/23] SLUB: Add SlabReclaimable() to avoid repeated reclaim attempts Christoph Lameter
2007-11-08 15:26 ` [patch 00/23] Slab defragmentation V6 Mel Gorman
2007-11-08 16:01   ` Plans for Onezonelist patch series ??? Lee Schermerhorn
2007-11-08 18:34     ` Christoph Lameter
2007-11-08 18:40       ` Mel Gorman
2007-11-08 18:43         ` Christoph Lameter
2007-11-08 20:06           ` Mel Gorman
2007-11-08 20:20             ` Christoph Lameter
2007-11-08 20:29               ` Mel Gorman
2007-11-08 18:39     ` Mel Gorman
2007-11-08 19:39       ` Christoph Lameter
2007-11-08 19:12   ` [patch 00/23] Slab defragmentation V6 Christoph Lameter
2007-11-08 20:24     ` Mel Gorman
2007-11-08 20:28       ` Christoph Lameter
2007-11-08 20:58     ` Lee Schermerhorn
2007-11-08 21:27       ` Christoph Lameter

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=20071108150705.GD2591@skynet.ie \
    --to=mel@skynet.ie \
    --cc=akpm@linux-foundatin.org \
    --cc=clameter@sgi.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    /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