linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Jamie Lokier <jamie.lokier@cern.ch>
To: "William J. Earl" <wje@cthulhu.engr.sgi.com>
Cc: Chuck Lever <cel@monkey.org>, linux-mm@kvack.org
Subject: Re: madvise (MADV_FREE)
Date: Tue, 21 Mar 2000 15:08:34 +0100	[thread overview]
Message-ID: <20000321150834.A5291@pcep-jamie.cern.ch> (raw)
In-Reply-To: <14550.56676.35208.422139@liveoak.engr.sgi.com>; from William J. Earl on Mon, Mar 20, 2000 at 06:24:36PM -0800

William J. Earl wrote:
>      I have been asked by some application people to have free() use
> MADV_DONTNEED or the equivalent in selected cases, specifically when
> the memory allocated is large, in order to free up the physical and
> virtual (swap space) memory for other uses.  If the application uses
> very large chunks of memory, giving it back entirely is a win.  The
> application could be recoded to do its own mmap() of /dev/zero and
> munmap(), but would prefer that this behavior be automatic.  Of course,
> MADV_DONTNEED does not apply in the case of mmap()/munmap() of /dev/zero,
> but it is not implausible to give up virtual memory.  Note that
> I am not claiming one should do anything of the sort for small
> allocations.

Take a look at Glibc's malloc/free, which is the only one we care about
for Linux.  Glibc's malloc uses mmap() of /dev/zero for large
allocations automatically.  You can change the threshold if you like.

However, assuming this was not the case, even your application would
benefit more from MADV_FREE than MADV_DONTNEED.  MADV_DONTNEED forces a
non-trivial minimum recycling cost, whereas MADV_FREE allows the cost to
be balanced between the kernel and the application, according to the
current paging situation.

-- Jamie
--
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.eu.org/Linux-MM/

  reply	other threads:[~2000-03-21 14:08 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20000320135939.A3390@pcep-jamie.cern.ch>
2000-03-20 19:09 ` MADV_SPACEAVAIL and MADV_FREE in pre2-3 Chuck Lever
2000-03-21  1:20   ` madvise (MADV_FREE) Jamie Lokier
2000-03-21  2:24     ` William J. Earl
2000-03-21 14:08       ` Jamie Lokier [this message]
2000-03-22 16:24     ` Chuck Lever
2000-03-22 18:05       ` Jamie Lokier
2000-03-22 21:39         ` Chuck Lever
2000-03-22 22:31           ` Jamie Lokier
2000-03-22 22:44             ` Stephen C. Tweedie
2000-03-23 18:53             ` Chuck Lever
2000-03-24  0:00               ` /dev/recycle Jamie Lokier
2000-03-24  9:14                 ` /dev/recycle Christoph Rohland
2000-03-24 13:10                   ` /dev/recycle Jamie Lokier
2000-03-24 13:54                     ` /dev/recycle Christoph Rohland
2000-03-24 14:17                       ` /dev/recycle Jamie Lokier
2000-03-24 17:40                         ` /dev/recycle Christoph Rohland
2000-03-24 18:13                           ` /dev/recycle Jamie Lokier
2000-03-25  8:35                             ` /dev/recycle Christoph Rohland
2000-03-28  0:48                 ` /dev/recycle Chuck Lever
2000-03-24  0:21               ` madvise (MADV_FREE) Jamie Lokier
2000-03-24  7:21                 ` lars brinkhoff
2000-03-24 17:42                   ` Jeff Dike
2000-03-24 16:49                     ` Jamie Lokier
2000-03-24 17:08                     ` Stephen C. Tweedie
2000-03-24 19:58                       ` Jeff Dike
2000-03-25  0:30                         ` Stephen C. Tweedie
2000-03-22 22:33           ` Stephen C. Tweedie
2000-03-22 22:45             ` Jamie Lokier
2000-03-22 22:48               ` Stephen C. Tweedie
2000-03-22 22:55                 ` Q. about swap-cache orphans Jamie Lokier
2000-03-22 22:58                   ` Stephen C. Tweedie
2000-03-22 18:15       ` madvise (MADV_FREE) Christoph Rohland
2000-03-22 18:30         ` Jamie Lokier
2000-03-23 16:56           ` Christoph Rohland
2000-03-21  1:29   ` MADV_DONTNEED Jamie Lokier
2000-03-22 17:04     ` MADV_DONTNEED Chuck Lever
2000-03-22 17:10       ` MADV_DONTNEED Stephen C. Tweedie
2000-03-22 17:32         ` MADV_DONTNEED Jamie Lokier
2000-03-22 17:33         ` MADV_DONTNEED Jamie Lokier
2000-03-22 17:37           ` MADV_DONTNEED Stephen C. Tweedie
2000-03-22 17:43       ` MADV_DONTNEED Jamie Lokier
2000-03-22 21:54         ` MADV_DONTNEED Chuck Lever
2000-03-22 22:41           ` MADV_DONTNEED Jamie Lokier
2000-03-23 19:13             ` MADV_DONTNEED James Antill
2000-03-21  1:47   ` Extensions to mincore Jamie Lokier
2000-03-21  9:11     ` Eric W. Biederman
2000-03-21  9:40       ` lars brinkhoff
2000-03-21 11:34       ` Stephen C. Tweedie
2000-03-21 15:15         ` Jamie Lokier
2000-03-21 15:41           ` Stephen C. Tweedie
2000-03-21 15:55             ` Jamie Lokier
2000-03-21 16:08               ` Stephen C. Tweedie
2000-03-21 16:48                 ` Jamie Lokier
2000-03-22  7:36                   ` Eric W. Biederman
2000-03-21  1:50   ` MADV flags as mmap options Jamie Lokier

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=20000321150834.A5291@pcep-jamie.cern.ch \
    --to=jamie.lokier@cern.ch \
    --cc=cel@monkey.org \
    --cc=linux-mm@kvack.org \
    --cc=wje@cthulhu.engr.sgi.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