All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andreas Hollmann <hollmann@in.tum.de>
To: Andi Kleen <andi@firstfloor.org>
Cc: linux-numa <linux-numa@vger.kernel.org>,
	linux-mm@kvack.org, mgorman@suse.de, akpm@linux-foundation.org
Subject: Re: Is it possible to disable numa_balance after boot?
Date: Sat, 4 Jan 2014 22:37:08 +0100	[thread overview]
Message-ID: <CAGz0_-0Q0XxvmXZii0MUrgm8dmYYF5xck3398iyZA2dRySuw5w@mail.gmail.com> (raw)
In-Reply-To: <20140104182235.GT20765@two.firstfloor.org>

2014/1/4 Andi Kleen <andi@firstfloor.org>:
> On Sat, Jan 04, 2014 at 06:46:55PM +0100, Andreas Hollmann wrote:
>> Hi,
>>
>> is possible to turn of numa balancing (introduced in 3.8) in a running kernel?
>
>
> I submitted a patch to do it some time ago
>
> https://lkml.org/lkml/2013/4/24/529
>
> But it didn't seem to have made it in. Andrew? Mel?
>
> Yes I agree a disable switch is totally needed for such an intrusive
> feature, if only to isolate problems with it.

That would be great. Additionally it would be nice to do it per application.

Some applications work well with pinning, others don't and it would be
bad to disable numa balancing globally.

>
> -Andi
>
>
>>
>> I'm running a recent arch kernel and numa balancing is enabled by
>> default. I checked
>> several documents and found some sysctl variable which influence the behavior of
>> numa balance, but there is no clear documentation if it's possible to
>> disable it.
>>
>> The only defined way to disable it is using a kernel parameter
>>
>> numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
>> Allowed values are enable and disable
>>
>> Is there any other way?
>>
>> Best regards,
>> Andreas
>>
>>
>> $ uname -a
>> Linux inwest 3.12.6-1-ARCH #1 SMP PREEMPT Fri Dec 20 19:39:00 CET 2013
>> x86_64 GNU/Linux
>>
>> $ cat /usr/src/linux-3.12.6-1-ARCH/.config | grep NUMA_BALANCING
>> CONFIG_ARCH_SUPPORTS_NUMA_BALANCING=y
>> CONFIG_NUMA_BALANCING_DEFAULT_ENABLED=y
>> CONFIG_NUMA_BALANCING=y
>>
>> $ ls -l /proc/sys/kernel | grep numa_bal
>> -rw-r--r-- 1 root root 0 Jan  4 14:23 numa_balancing_scan_delay_ms
>> -rw-r--r-- 1 root root 0 Jan  4 14:23 numa_balancing_scan_period_max_ms
>> -rw-r--r-- 1 root root 0 Jan  4 14:23 numa_balancing_scan_period_min_ms
>> -rw-r--r-- 1 root root 0 Jan  4 14:23 numa_balancing_scan_period_reset
>> -rw-r--r-- 1 root root 0 Jan  4 14:23 numa_balancing_scan_size_mb
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-numa" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>
>
> --
> ak@linux.intel.com -- Speaking for myself only.

--
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>

  reply	other threads:[~2014-01-04 21:37 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-01-04 17:46 Is it possible to disable numa_balance after boot? Andreas Hollmann
2014-01-04 18:22 ` Andi Kleen
2014-01-04 18:22   ` Andi Kleen
2014-01-04 21:37   ` Andreas Hollmann [this message]
2014-01-06 20:33   ` Andrew Morton
2014-01-06 20:33     ` Andrew Morton

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=CAGz0_-0Q0XxvmXZii0MUrgm8dmYYF5xck3398iyZA2dRySuw5w@mail.gmail.com \
    --to=hollmann@in.tum.de \
    --cc=akpm@linux-foundation.org \
    --cc=andi@firstfloor.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-numa@vger.kernel.org \
    --cc=mgorman@suse.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.