All of lore.kernel.org
 help / color / mirror / Atom feed
From: Wu Fengguang <fengguang.wu@intel.com>
To: Nikanth Karthikesan <knikanth@suse.de>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>,
	Jens Axboe <axboe@kernel.dk>,
	Andrew Morton <akpm@linux-foundation.org>,
	Jan Kara <jack@suse.cz>, Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Trond Myklebust <Trond.Myklebust@netapp.com>,
	Peter Staubach <staubach@redhat.com>
Subject: Re: [RFC][PATCH] Per file dirty limit throttling
Date: Tue, 17 Aug 2010 10:41:40 +0800	[thread overview]
Message-ID: <20100817024140.GB13916@localhost> (raw)
In-Reply-To: <201008160949.51512.knikanth@suse.de>

On Mon, Aug 16, 2010 at 12:19:50PM +0800, Nikanth Karthikesan wrote:
> When the total dirty pages exceed vm_dirty_ratio, the dirtier is made to do
> the writeback. But this dirtier may not be the one who took the system to this
> state. Instead, if we can track the dirty count per-file, we could throttle
> the dirtier of a file, when the file's dirty pages exceed a certain limit.
> Even though this dirtier may not be the one who dirtied the other pages of
> this file, it is fair to throttle this process, as it uses that file.

Nikanth, there's a more elegant solution in upstream kernel.
See the comment for task_dirty_limit() in commit 1babe1838.

NFS may want to limit per-file dirty pages, to prevent long stall time
inside the nfs_getattr()->filemap_write_and_wait() calls (and problems
like that). Peter Staubach has similar ideas on it.

Thanks,
Fengguang

WARNING: multiple messages have this Message-ID (diff)
From: Wu Fengguang <fengguang.wu@intel.com>
To: Nikanth Karthikesan <knikanth@suse.de>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>,
	Jens Axboe <axboe@kernel.dk>,
	Andrew Morton <akpm@linux-foundation.org>,
	Jan Kara <jack@suse.cz>, Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Trond Myklebust <Trond.Myklebust@netapp.com>,
	Peter Staubach <staubach@redhat.com>
Subject: Re: [RFC][PATCH] Per file dirty limit throttling
Date: Tue, 17 Aug 2010 10:41:40 +0800	[thread overview]
Message-ID: <20100817024140.GB13916@localhost> (raw)
In-Reply-To: <201008160949.51512.knikanth@suse.de>

On Mon, Aug 16, 2010 at 12:19:50PM +0800, Nikanth Karthikesan wrote:
> When the total dirty pages exceed vm_dirty_ratio, the dirtier is made to do
> the writeback. But this dirtier may not be the one who took the system to this
> state. Instead, if we can track the dirty count per-file, we could throttle
> the dirtier of a file, when the file's dirty pages exceed a certain limit.
> Even though this dirtier may not be the one who dirtied the other pages of
> this file, it is fair to throttle this process, as it uses that file.

Nikanth, there's a more elegant solution in upstream kernel.
See the comment for task_dirty_limit() in commit 1babe1838.

NFS may want to limit per-file dirty pages, to prevent long stall time
inside the nfs_getattr()->filemap_write_and_wait() calls (and problems
like that). Peter Staubach has similar ideas on it.

Thanks,
Fengguang

--
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:[~2010-08-17  2:41 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-08-16  4:19 [RFC][PATCH] Per file dirty limit throttling Nikanth Karthikesan
2010-08-16  4:19 ` Nikanth Karthikesan
2010-08-16 11:05 ` Peter Zijlstra
2010-08-16 11:05   ` Peter Zijlstra
2010-08-17  5:09   ` Nikanth Karthikesan
2010-08-17  5:09     ` Nikanth Karthikesan
2010-08-17  8:24     ` Peter Zijlstra
2010-08-17  8:24       ` Peter Zijlstra
2010-08-18  9:22       ` Nikanth Karthikesan
2010-08-18  9:22         ` Nikanth Karthikesan
2010-08-18  9:58         ` Peter Zijlstra
2010-08-18  9:58           ` Peter Zijlstra
2010-08-18 14:08           ` Balbir Singh
2010-08-18 14:08             ` Balbir Singh
2010-08-18 14:25             ` Peter Zijlstra
2010-08-18 14:25               ` Peter Zijlstra
2010-08-18 14:48               ` Balbir Singh
2010-08-18 14:48                 ` Balbir Singh
2010-08-23 12:19           ` Nikanth Karthikesan
2010-08-23 12:19             ` Nikanth Karthikesan
2010-08-16 16:53 ` Bill Davidsen
2010-08-16 16:53   ` Bill Davidsen
2010-08-17  2:53   ` Wu Fengguang
2010-08-17  2:53     ` Wu Fengguang
2010-08-17  2:41 ` Wu Fengguang [this message]
2010-08-17  2:41   ` Wu Fengguang

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=20100817024140.GB13916@localhost \
    --to=fengguang.wu@intel.com \
    --cc=Trond.Myklebust@netapp.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=akpm@linux-foundation.org \
    --cc=axboe@kernel.dk \
    --cc=jack@suse.cz \
    --cc=knikanth@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=staubach@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.