All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mel Gorman <mgorman@suse.de>
To: Cong Wang <amwang@redhat.com>
Cc: Mel Gorman <mel@csn.ul.ie>,
	linux-kernel@vger.kernel.org, akpm@linux-foundation.org,
	Andrea Arcangeli <aarcange@redhat.com>,
	Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Rik van Riel <riel@redhat.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
	linux-mm@kvack.org
Subject: Re: [PATCH 2/3] mm: make the threshold of enabling THP configurable
Date: Wed, 22 Jun 2011 12:15:25 +0100	[thread overview]
Message-ID: <20110622111525.GK9396@suse.de> (raw)
In-Reply-To: <4E01C80F.8070605@redhat.com>

On Wed, Jun 22, 2011 at 06:46:39PM +0800, Cong Wang wrote:
> ??? 2011???06???22??? 17:16, Mel Gorman ??????:
> >
> >What I meant was that there is a rational reason why 512M is the
> >default for enabling THP by default. Tuning it lower than that by any
> >means makes very little sense. Tuning it higher might make some sense
> >but it is more likely that THP would simply be disabled via sysctl. I
> >see very little advantage to introducing this Kconfig option other
> >than as a source of confusion when running make oldconfig.
> >
> 
> The tunable range is (512, 8192), so 512M is the minimum.
> 
> Sure, I knew it can be disabled via /sys, actually we can do even
> more in user-space, that is totally move the 512M check out of kernel,
> why we didn't?
> 

Because the reason why 512M is the default is not obvious and there
was no guarantee all distros would chose a reasonable default for
an init script (or know that an init script was even necessary).
This is one of the few cases where there is a sensible default that
is the least surprising.

> In short, I think we should either remove the 512M from kernel, or
> make 512M to be tunable.
> 

That just hands them a different sort of rope to hang themselves with
where THP gets enabled on small machines or botting with mem=128M
and getting surprised later by the high min_free_kbytes.

At this point, I don't really care if the Kconfig entry exists or
not. I think it gains nothing but additional confusion for people
who write .config files but it's not a topic I want to discuss for
days either.

-- 
Mel Gorman
SUSE Labs

WARNING: multiple messages have this Message-ID (diff)
From: Mel Gorman <mgorman@suse.de>
To: Cong Wang <amwang@redhat.com>
Cc: Mel Gorman <mel@csn.ul.ie>,
	linux-kernel@vger.kernel.org, akpm@linux-foundation.org,
	Andrea Arcangeli <aarcange@redhat.com>,
	Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Rik van Riel <riel@redhat.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
	linux-mm@kvack.org
Subject: Re: [PATCH 2/3] mm: make the threshold of enabling THP configurable
Date: Wed, 22 Jun 2011 12:15:25 +0100	[thread overview]
Message-ID: <20110622111525.GK9396@suse.de> (raw)
In-Reply-To: <4E01C80F.8070605@redhat.com>

On Wed, Jun 22, 2011 at 06:46:39PM +0800, Cong Wang wrote:
> ??? 2011???06???22??? 17:16, Mel Gorman ??????:
> >
> >What I meant was that there is a rational reason why 512M is the
> >default for enabling THP by default. Tuning it lower than that by any
> >means makes very little sense. Tuning it higher might make some sense
> >but it is more likely that THP would simply be disabled via sysctl. I
> >see very little advantage to introducing this Kconfig option other
> >than as a source of confusion when running make oldconfig.
> >
> 
> The tunable range is (512, 8192), so 512M is the minimum.
> 
> Sure, I knew it can be disabled via /sys, actually we can do even
> more in user-space, that is totally move the 512M check out of kernel,
> why we didn't?
> 

Because the reason why 512M is the default is not obvious and there
was no guarantee all distros would chose a reasonable default for
an init script (or know that an init script was even necessary).
This is one of the few cases where there is a sensible default that
is the least surprising.

> In short, I think we should either remove the 512M from kernel, or
> make 512M to be tunable.
> 

That just hands them a different sort of rope to hang themselves with
where THP gets enabled on small machines or botting with mem=128M
and getting surprised later by the high min_free_kbytes.

At this point, I don't really care if the Kconfig entry exists or
not. I think it gains nothing but additional confusion for people
who write .config files but it's not a topic I want to discuss for
days either.

-- 
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  reply	other threads:[~2011-06-22 11:15 UTC|newest]

Thread overview: 83+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-20 16:34 [PATCH 1/3] mm: completely disable THP by transparent_hugepage=never Amerigo Wang
2011-06-20 16:34 ` Amerigo Wang
2011-06-20 16:34 ` [PATCH 2/3] mm: make the threshold of enabling THP configurable Amerigo Wang
2011-06-20 16:34   ` Amerigo Wang
2011-06-20 16:59   ` Dave Hansen
2011-06-20 16:59     ` Dave Hansen
2011-06-20 17:23     ` Cong Wang
2011-06-20 17:23       ` Cong Wang
2011-06-20 16:59   ` Mel Gorman
2011-06-20 16:59     ` Mel Gorman
2011-06-20 17:16     ` Cong Wang
2011-06-20 17:16       ` Cong Wang
2011-06-21  9:36       ` Mel Gorman
2011-06-21  9:36         ` Mel Gorman
2011-06-22  2:41         ` Cong Wang
2011-06-22  2:41           ` Cong Wang
2011-06-22  9:16           ` Mel Gorman
2011-06-22  9:16             ` Mel Gorman
2011-06-22 10:46             ` Cong Wang
2011-06-22 10:46               ` Cong Wang
2011-06-22 11:15               ` Mel Gorman [this message]
2011-06-22 11:15                 ` Mel Gorman
2011-06-22 12:34                 ` Cong Wang
2011-06-22 12:34                   ` Cong Wang
2011-06-20 16:34 ` [PATCH 3/3] mm: print information when THP is disabled automatically Amerigo Wang
2011-06-20 16:34   ` Amerigo Wang
2011-06-20 16:54   ` Andrea Arcangeli
2011-06-20 16:54     ` Andrea Arcangeli
2011-06-20 17:25     ` Cong Wang
2011-06-20 17:25       ` Cong Wang
2011-06-20 17:01   ` Mel Gorman
2011-06-20 17:01     ` Mel Gorman
2011-06-20 17:26     ` Cong Wang
2011-06-20 17:26       ` Cong Wang
2011-06-20 19:37       ` Andrea Arcangeli
2011-06-20 19:37         ` Andrea Arcangeli
2011-06-21  9:40       ` Mel Gorman
2011-06-21  9:40         ` Mel Gorman
2011-06-20 16:50 ` [PATCH 1/3] mm: completely disable THP by transparent_hugepage=never Andrea Arcangeli
2011-06-20 16:50   ` Andrea Arcangeli
2011-06-20 16:55   ` Rik van Riel
2011-06-20 16:55     ` Rik van Riel
2011-06-20 17:01   ` Cong Wang
2011-06-20 17:01     ` Cong Wang
2011-06-20 19:43     ` Andrea Arcangeli
2011-06-20 19:43       ` Andrea Arcangeli
2011-06-21  3:15       ` Cong Wang
2011-06-21  3:15         ` Cong Wang
2011-06-20 16:58 ` Mel Gorman
2011-06-20 16:58   ` Mel Gorman
2011-06-20 17:07   ` Cong Wang
2011-06-20 17:07     ` Cong Wang
2011-06-20 17:10     ` Rik van Riel
2011-06-20 17:10       ` Rik van Riel
2011-06-20 17:19       ` Cong Wang
2011-06-20 17:19         ` Cong Wang
2011-06-20 17:28         ` Rik van Riel
2011-06-20 17:28           ` Rik van Riel
2011-06-20 17:34           ` Cong Wang
2011-06-20 17:34             ` Cong Wang
2011-06-20 17:50             ` Rik van Riel
2011-06-20 17:50               ` Rik van Riel
2011-06-20 18:25               ` Vivek Goyal
2011-06-20 18:25                 ` Vivek Goyal
2011-06-20 19:21                 ` Andrea Arcangeli
2011-06-20 19:21                   ` Andrea Arcangeli
2011-06-21  4:08                   ` Cong Wang
2011-06-21  4:08                     ` Cong Wang
2011-06-21 14:43                     ` Andrea Arcangeli
2011-06-21 14:43                       ` Andrea Arcangeli
2011-06-22  2:56                       ` Cong Wang
2011-06-22  2:56                         ` Cong Wang
2011-06-22 14:22                         ` Andrea Arcangeli
2011-06-22 14:22                           ` Andrea Arcangeli
2011-06-21 20:01                     ` Rik van Riel
2011-06-21 20:01                       ` Rik van Riel
2011-06-21  3:28               ` Cong Wang
2011-06-21  3:28                 ` Cong Wang
2011-06-20 17:58             ` Eric B Munson
2011-06-21  3:36               ` Cong Wang
2011-06-21  3:36                 ` Cong Wang
2011-06-20 17:59           ` Vivek Goyal
2011-06-20 17:59             ` Vivek Goyal

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=20110622111525.GK9396@suse.de \
    --to=mgorman@suse.de \
    --cc=aarcange@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=amwang@redhat.com \
    --cc=benh@kernel.crashing.org \
    --cc=hannes@cmpxchg.org \
    --cc=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mel@csn.ul.ie \
    --cc=riel@redhat.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 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.