linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Cong Wang <amwang@redhat.com>
To: John Stoffel <john@stoffel.org>
Cc: linux-kernel@vger.kernel.org,
	Andrew Morton <akpm@linux-foundation.org>,
	Cong Wang <xiyou.wangcong@gmail.com>,
	Alexander Viro <viro@zeniv.linux.org.uk>,
	Matthew Wilcox <matthew@wil.cx>,
	linux-fsdevel@vger.kernel.org, linux-mm@kvack.org,
	Keiichi Kii <k-keiichi@bx.jp.nec.com>
Subject: Re: [RFC Patch] fs: implement per-file drop caches
Date: Thu, 31 May 2012 14:28:16 +0800	[thread overview]
Message-ID: <1338445696.19369.27.camel@cr0> (raw)
In-Reply-To: <20422.14538.833061.105058@quad.stoffel.home>

On Wed, 2012-05-30 at 11:12 -0400, John Stoffel wrote:
> Cong> This is a draft patch of implementing per-file drop caches.
> 
> Interesting.  So can I do this from outside a process?  I'm a
> SysAdmin, so my POV is from noticing, finding and fixing performance
> problems when the system is under pressure.  

Yes, sure, we need to write a utility (or patch an existing one) to do
this for you admins.

> 
> Cong> It introduces a new fcntl command  F_DROP_CACHES to drop
> Cong> file caches of a specific file. The reason is that currently
> Cong> we only have a system-wide drop caches interface, it could
> Cong> cause system-wide performance down if we drop all page caches
> Cong> when we actually want to drop the caches of some huge file.
> 
> How can I tell how much cache is used by a file?  And what is the
> performance impact of this when run on a busy system?  And what does
> this patch buy us since I figure the VM should already be dropping
> caches once the system comes under mem pressure...
> 

AFAIK, we don't export such information to user-space, we only have
system-wide statistics.

Keiichi (in Cc) once wrote a patch to implement page cache tracepoint:
http://marc.info/?l=linux-mm&m=131102496904326&w=3

but the patches are still not in upstream.

Thanks!

--
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:[~2012-05-31  6:28 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-05-30 13:38 [RFC Patch] fs: implement per-file drop caches Cong Wang
2012-05-30 15:12 ` John Stoffel
2012-05-31  6:28   ` Cong Wang [this message]
2012-05-30 15:14 ` Pádraig Brady
2012-05-31  6:20   ` Cong Wang
2012-05-31  6:30     ` KOSAKI Motohiro
2012-05-31 12:11       ` Cong Wang
2012-05-31 19:09         ` KOSAKI Motohiro
2012-06-01 11:32           ` Cong Wang
2012-06-01 13:08             ` John Stoffel
2012-06-04  3:28               ` Cong Wang

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=1338445696.19369.27.camel@cr0 \
    --to=amwang@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=john@stoffel.org \
    --cc=k-keiichi@bx.jp.nec.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=matthew@wil.cx \
    --cc=viro@zeniv.linux.org.uk \
    --cc=xiyou.wangcong@gmail.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).