From: Arjan van de Ven <arjan@infradead.org>
To: Al Boldi <a1426z@gawab.com>
Cc: Willy Tarreau <willy@w.ods.org>,
Alan Cox <alan@lxorguk.ukuu.org.uk>,
barryn@pobox.com, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] strict VM overcommit accounting for 2.4.32/2.4.33-pre1
Date: Sun, 01 Jan 2006 10:12:33 +0100 [thread overview]
Message-ID: <1136106754.17830.4.camel@laptopd505.fenrus.org> (raw)
In-Reply-To: <200512312036.01351.a1426z@gawab.com>
On Sat, 2005-12-31 at 20:36 +0300, Al Boldi wrote:
> Arjan van de Ven wrote:
> > On Sat, 2005-12-31 at 17:02 +0300, Al Boldi wrote:
> > > Shouldn't it be possible to disable overcommit completely, thus giving
> > > kswapd a break from running wild trying to find something to swap/page,
> > > which is the reason why the system gets unstable going over 95% in your
> > > example.
> >
> > shared mappings make this impractical. To disable overcommit completely,
> > each process would need to account for all its own shared libraries, eg
> > each process gets glibc added etc. You'll find that on any
> > non-extremely-stripped system you then end up with much more memory
> > needed than you have ram.
>
> Are you implying shared maps are implemented by way of overcommitting?
yes. Using only 1 page shared is already overcommitting because in
principle each user can cause a COW on that page and cause a memory
allocation for it in the future. (just like "traditional" overcommit can
cause a pagefault with an allocation)
> Really, overcommit is an add-on feature like swapping, only overcommit is
> free because it's a lier. So removing an add-on feature should not affect
> the underlying system in any way, such as shared mappings or swapping.
then I think you're misunderstanding how the linux VM works
> It should be possible to allow swapping to handle all memory requests
> exceeding physical RAM. OverCommit should be a tuning option for those who
> like to live on the edge, because it really is a gamble.
but it's a worthwhile gamble to not need GOBBLES of memory you're not
using 99.9999999999999% of the time
>
> In the case where swap = physical RAM and overcommit_ratio = 0, the kernel is
> in effect hiding the fact that it is overcommitting.
swap==ram is not really relevant; you'll need a LOT more to cover the
shared maps...
next prev parent reply other threads:[~2006-01-01 9:12 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-12-30 20:06 Al Boldi
2005-12-30 20:18 ` Barry K. Nathan
2005-12-30 20:51 ` Al Boldi
2005-12-30 21:44 ` Willy TARREAU
2005-12-31 0:55 ` Alan Cox
2005-12-31 4:59 ` Al Boldi
2005-12-31 7:38 ` Willy Tarreau
2005-12-31 14:02 ` Al Boldi
2005-12-31 14:18 ` Willy Tarreau
2005-12-31 14:26 ` Arjan van de Ven
2005-12-31 14:58 ` Willy Tarreau
2005-12-31 15:17 ` Arjan van de Ven
2005-12-31 17:36 ` Al Boldi
2006-01-01 9:12 ` Arjan van de Ven [this message]
-- strict thread matches above, loose matches on Subject: below --
2005-12-30 7:44 Barry K. Nathan
2005-12-30 17:48 ` Willy Tarreau
2005-12-30 18:17 ` Arjan van de Ven
2005-12-30 18:33 ` Willy Tarreau
2005-12-30 19:37 ` Barry K. Nathan
2005-12-30 20:45 ` Willy Tarreau
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=1136106754.17830.4.camel@laptopd505.fenrus.org \
--to=arjan@infradead.org \
--cc=a1426z@gawab.com \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=barryn@pobox.com \
--cc=linux-kernel@vger.kernel.org \
--cc=willy@w.ods.org \
/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
all inboxes | Powered by JetHome®