linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Mel Gorman <mgorman@suse.de>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: David Rientjes <rientjes@google.com>,
	Han Pingtian <hanpt@linux.vnet.ibm.com>,
	linux-kernel@vger.kernel.org, Michal Hocko <mhocko@suse.cz>,
	linux-mm@kvack.org, Dave Hansen <dave.hansen@intel.com>
Subject: Re: [RFC] mm: show message when updating min_free_kbytes in thp
Date: Wed, 15 Jan 2014 15:22:18 +0000	[thread overview]
Message-ID: <20140115152218.GL4963@suse.de> (raw)
In-Reply-To: <20140114163533.ab191e118e82ca7b4d499551@linux-foundation.org>

On Tue, Jan 14, 2014 at 04:35:33PM -0800, Andrew Morton wrote:
> > Would it be overkill to save the kernel default both with and without thp 
> > and then doing a WARN_ON_ONCE() if a user-written value is ever less?
> 
> Well, min_free_kbytes is a userspace thing, not a kernel thing - maybe
> THP shouldn't be dinking with it.  What effect is THP trying to achieve
> and can we achieve it by other/better means?

It moved logic from hugeadm where few people knew about it to the
kernel. The value is related to anti-fragmentation. With the recommended
setting the probability of mixing pages of different mobility within a
single pageblock is reduced. Very very superficially, it reduces the
number of instances the mm_page_alloc_extfrag tracepoint is triggered
with parameters that are considered to be severely fragmenting.

-- 
Mel Gorman
SUSE Labs

--
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:[~2014-01-15 15:22 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-01-01  0:29 [RFC] mm: show message when updating min_free_kbytes in thp Han Pingtian
2014-01-02 18:05 ` Dave Hansen
2014-01-02 21:58   ` David Rientjes
2014-01-02 22:10     ` Dave Hansen
2014-01-02 23:36       ` David Rientjes
2014-01-02 23:48         ` Dave Hansen
2014-01-03  3:33   ` Han Pingtian
2014-01-03 18:17     ` Dave Hansen
2014-01-05  0:35       ` Han Pingtian
2014-01-06 16:46         ` Michal Hocko
2014-01-08  3:59           ` Han Pingtian
2014-01-08  8:20           ` Han Pingtian
2014-01-08 10:16             ` Michal Hocko
2014-01-09  7:32               ` Han Pingtian
2014-01-09  9:02                 ` Michal Hocko
2014-01-09 21:15                 ` David Rientjes
2014-01-10  8:05                   ` Michal Hocko
2014-01-10  8:13                     ` Andrew Morton
2014-01-10  8:17                       ` Michal Hocko
2014-01-11  3:27                         ` Han Pingtian
2014-01-14 20:07                         ` Han Pingtian
2014-01-14 23:52                           ` Andrew Morton
2014-01-15  0:25                             ` David Rientjes
2014-01-15  0:35                               ` Andrew Morton
2014-01-15  0:52                                 ` David Rientjes
2014-01-15 15:22                                 ` Mel Gorman [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=20140115152218.GL4963@suse.de \
    --to=mgorman@suse.de \
    --cc=akpm@linux-foundation.org \
    --cc=dave.hansen@intel.com \
    --cc=hanpt@linux.vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@suse.cz \
    --cc=rientjes@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
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).