Linux-Fsdevel Archive on lore.kernel.org
 help / color / Atom feed
From: David Rientjes <rientjes@google.com>
To: chukaiping <chukaiping@baidu.com>
Cc: mcgrof@kernel.org, keescook@chromium.org, yzaikin@google.com,
	akpm@linux-foundation.org, linux-kernel@vger.kernel.org,
	linux-fsdevel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH v2] mm/compaction:let proactive compaction order configurable
Date: Fri, 16 Apr 2021 14:31:28 -0700 (PDT)
Message-ID: <7efa316c-d39b-59a5-bc52-62325127a917@google.com> (raw)
In-Reply-To: <1618593751-32148-1-git-send-email-chukaiping@baidu.com>

On Sat, 17 Apr 2021, chukaiping wrote:

> Currently the proactive compaction order is fixed to
> COMPACTION_HPAGE_ORDER(9), it's OK in most machines with lots of
> normal 4KB memory, but it's too high for the machines with small
> normal memory, for example the machines with most memory configured
> as 1GB hugetlbfs huge pages. In these machines the max order of
> free pages is often below 9, and it's always below 9 even with hard
> compaction. This will lead to proactive compaction be triggered very
> frequently. In these machines we only care about order of 3 or 4.
> This patch export the oder to proc and let it configurable
> by user, and the default value is still COMPACTION_HPAGE_ORDER.
> 

Still not entirely clear on the use case beyond hugepages.  In your 
response from v1, you indicated you were not concerned with allocation 
latency of hugepages but rather had a thundering herd problem where once 
fragmentation got bad, many threads started compacting all at once.

I'm not sure that tuning the proactive compaction order is the right 
solution.  I think the proactive compaction order is more about starting 
compaction when a known order of interest (like a hugepage) is fully 
depleted and we want a page of that order so the idea is to start 
recovering from that situation.

Is this not a userspace policy decision?  I'm wondering if it would simply 
be better to manually invoke compaction periodically or when the 
fragmentation ratio has reached a certain level.  You can manually invoke 
compaction yourself through sysfs for each node on the system.

> Signed-off-by: chukaiping <chukaiping@baidu.com>
> Reported-by: kernel test robot <lkp@intel.com>
> ---
> 
> Changes in v2:
>     - fix the compile error in ia64 and powerpc
>     - change the hard coded max order number from 10 to MAX_ORDER - 1
> 
>  include/linux/compaction.h |    1 +
>  kernel/sysctl.c            |   11 +++++++++++
>  mm/compaction.c            |   14 +++++++++++---
>  3 files changed, 23 insertions(+), 3 deletions(-)
> 
> diff --git a/include/linux/compaction.h b/include/linux/compaction.h
> index ed4070e..151ccd1 100644
> --- a/include/linux/compaction.h
> +++ b/include/linux/compaction.h
> @@ -83,6 +83,7 @@ static inline unsigned long compact_gap(unsigned int order)
>  #ifdef CONFIG_COMPACTION
>  extern int sysctl_compact_memory;
>  extern unsigned int sysctl_compaction_proactiveness;
> +extern unsigned int sysctl_compaction_order;
>  extern int sysctl_compaction_handler(struct ctl_table *table, int write,
>  			void *buffer, size_t *length, loff_t *ppos);
>  extern int sysctl_extfrag_threshold;
> diff --git a/kernel/sysctl.c b/kernel/sysctl.c
> index 62fbd09..a607d4d 100644
> --- a/kernel/sysctl.c
> +++ b/kernel/sysctl.c
> @@ -195,6 +195,8 @@ enum sysctl_writes_mode {
>  #endif /* CONFIG_SMP */
>  #endif /* CONFIG_SCHED_DEBUG */
>  
> +static int max_buddy_zone = MAX_ORDER - 1;
> +
>  #ifdef CONFIG_COMPACTION
>  static int min_extfrag_threshold;
>  static int max_extfrag_threshold = 1000;
> @@ -2871,6 +2873,15 @@ int proc_do_static_key(struct ctl_table *table, int write,
>  		.extra2		= &one_hundred,
>  	},
>  	{
> +		.procname       = "compaction_order",
> +		.data           = &sysctl_compaction_order,
> +		.maxlen         = sizeof(sysctl_compaction_order),
> +		.mode           = 0644,
> +		.proc_handler   = proc_dointvec_minmax,
> +		.extra1         = SYSCTL_ZERO,
> +		.extra2         = &max_buddy_zone,
> +	},
> +	{
>  		.procname	= "extfrag_threshold",
>  		.data		= &sysctl_extfrag_threshold,
>  		.maxlen		= sizeof(int),
> diff --git a/mm/compaction.c b/mm/compaction.c
> index e04f447..bfd1d5e 100644
> --- a/mm/compaction.c
> +++ b/mm/compaction.c
> @@ -1925,16 +1925,16 @@ static bool kswapd_is_running(pg_data_t *pgdat)
>  
>  /*
>   * A zone's fragmentation score is the external fragmentation wrt to the
> - * COMPACTION_HPAGE_ORDER. It returns a value in the range [0, 100].
> + * sysctl_compaction_order. It returns a value in the range [0, 100].
>   */
>  static unsigned int fragmentation_score_zone(struct zone *zone)
>  {
> -	return extfrag_for_order(zone, COMPACTION_HPAGE_ORDER);
> +	return extfrag_for_order(zone, sysctl_compaction_order);
>  }
>  
>  /*
>   * A weighted zone's fragmentation score is the external fragmentation
> - * wrt to the COMPACTION_HPAGE_ORDER scaled by the zone's size. It
> + * wrt to the sysctl_compaction_order scaled by the zone's size. It
>   * returns a value in the range [0, 100].
>   *
>   * The scaling factor ensures that proactive compaction focuses on larger
> @@ -2666,6 +2666,7 @@ static void compact_nodes(void)
>   * background. It takes values in the range [0, 100].
>   */
>  unsigned int __read_mostly sysctl_compaction_proactiveness = 20;
> +unsigned int __read_mostly sysctl_compaction_order;
>  
>  /*
>   * This is the entry point for compacting all nodes via
> @@ -2958,6 +2959,13 @@ static int __init kcompactd_init(void)
>  	int nid;
>  	int ret;
>  
> +	/*
> +	 * move the initialization of sysctl_compaction_order to here to
> +	 * eliminate compile error in ia64 and powerpc architecture because
> +	 * COMPACTION_HPAGE_ORDER is a variable in this architecture
> +	 */
> +	sysctl_compaction_order = COMPACTION_HPAGE_ORDER;
> +
>  	ret = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
>  					"mm/compaction:online",
>  					kcompactd_cpu_online, NULL);
> -- 
> 1.7.1
> 
> 
> 

  reply index

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-16 17:22 chukaiping
2021-04-16 21:31 ` David Rientjes [this message]
2021-04-19 12:02   ` 答复: " Chu,Kaiping
2021-04-21  7:21 chukaiping
2021-04-22 16:27 ` Nitin Gupta
2021-04-22 20:33 ` Rafael Aquini

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=7efa316c-d39b-59a5-bc52-62325127a917@google.com \
    --to=rientjes@google.com \
    --cc=akpm@linux-foundation.org \
    --cc=chukaiping@baidu.com \
    --cc=keescook@chromium.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mcgrof@kernel.org \
    --cc=yzaikin@google.com \
    /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

Linux-Fsdevel Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-fsdevel/0 linux-fsdevel/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-fsdevel linux-fsdevel/ https://lore.kernel.org/linux-fsdevel \
		linux-fsdevel@vger.kernel.org
	public-inbox-index linux-fsdevel

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-fsdevel


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git