From: Andrea Arcangeli <aarcange@redhat.com>
To: Rik van Riel <riel@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
Hugh Dickins <hugh.dickins@tiscali.co.uk>,
Andrew Morton <akpm@linux-foundation.org>,
Izik Eidus <ieidus@redhat.com>, Chris Wright <chrisw@redhat.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH 2/9] ksm: let shared pages be swappable
Date: Wed, 2 Dec 2009 13:55:01 +0100 [thread overview]
Message-ID: <20091202125501.GD28697@random.random> (raw)
In-Reply-To: <4B15F642.1080308@redhat.com>
On Wed, Dec 02, 2009 at 12:08:18AM -0500, Rik van Riel wrote:
> The VM needs to touch a few (but only a few) PTEs in
> that situation, to make sure that anonymous pages get
> moved to the inactive anon list and get to a real chance
> at being referenced before we try to evict anonymous
> pages.
>
> Without a small amount of pre-aging, we would end up
> essentially doing FIFO replacement of anonymous memory,
> which has been known to be disastrous to performance
> for over 40 years now.
So far the only kernel that hangs in fork is the newer one...
In general I cannot care less about FIFO, I care about no CPU waste on
100% of my systems were swap is not needed. All my unmapped cache is
100% garbage collectable, and there is never any reason to flush any
tlb and walk the rmap chain. Give me a knob to disable the CPU waste
given I know what is going on, on my systems. I am totally ok with
slightly slower swap performance and fifo replacement in case I
eventually hit swap for a little while, then over time if memory
pressure stays high swap behavior will improve regardless of
flooding ipis to clear young bit when there are hundred gigabytes of
freeaeble cache unmapped and clean.
> Having said that - it may be beneficial to keep very heavily
> shared pages on the active list, without ever trying to scan
> the ptes associated with them.
Just mapped pages in general, not heavily... The other thing that is
beneficial likely is to stop page_referenced after 64 young bit clear,
that is referenced enough, you can enable this under my knob so that
it won't screw your algorithm. I don't have 1 terabyte of memory, so
you don't have to worry for me, I just want every cycle out of my cpu
without having to use O_DIRECT all the time.
--
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>
next prev parent reply other threads:[~2009-12-02 12:55 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-11-24 16:37 [PATCH 0/9] ksm: swapping Hugh Dickins
2009-11-24 16:40 ` [PATCH 1/9] ksm: fix mlockfreed to munlocked Hugh Dickins
2009-11-24 23:53 ` Rik van Riel
2009-11-26 16:20 ` Mel Gorman
2009-11-27 12:45 ` Hugh Dickins
2009-11-30 6:01 ` KOSAKI Motohiro
2009-11-30 12:26 ` Hugh Dickins
2009-11-30 21:27 ` Lee Schermerhorn
2009-12-01 11:14 ` Mel Gorman
2009-11-24 16:42 ` [PATCH 2/9] ksm: let shared pages be swappable Hugh Dickins
2009-11-30 0:46 ` KAMEZAWA Hiroyuki
2009-11-30 9:15 ` KOSAKI Motohiro
2009-11-30 12:38 ` Hugh Dickins
2009-12-01 4:14 ` KOSAKI Motohiro
2009-11-30 11:55 ` Hugh Dickins
2009-11-30 12:07 ` Andrea Arcangeli
2009-12-01 0:39 ` KAMEZAWA Hiroyuki
2009-12-01 6:32 ` Chris Wright
2009-12-01 9:11 ` Andrea Arcangeli
2009-12-01 9:28 ` KOSAKI Motohiro
2009-12-01 9:37 ` Andrea Arcangeli
2009-12-01 9:46 ` KOSAKI Motohiro
2009-12-01 9:59 ` Andrea Arcangeli
2009-12-02 5:08 ` Rik van Riel
2009-12-02 12:55 ` Andrea Arcangeli [this message]
2009-12-03 5:15 ` KOSAKI Motohiro
2009-12-04 5:06 ` KOSAKI Motohiro
2009-12-04 5:16 ` KAMEZAWA Hiroyuki
2009-12-04 14:49 ` Andrea Arcangeli
2009-12-04 17:16 ` Chris Wright
2009-12-04 18:53 ` Andrea Arcangeli
2009-12-04 19:03 ` Chris Wright
2009-12-09 0:43 ` KAMEZAWA Hiroyuki
2009-12-09 1:04 ` Chris Wright
2009-12-09 16:12 ` Andrea Arcangeli
2009-12-09 23:54 ` KAMEZAWA Hiroyuki
2009-12-04 14:45 ` Andrea Arcangeli
2009-12-04 16:21 ` Rik van Riel
2009-11-24 16:43 ` [PATCH 3/9] ksm: hold anon_vma in rmap_item Hugh Dickins
2009-11-24 16:45 ` [PATCH 4/9] ksm: take keyhole reference to page Hugh Dickins
2009-11-24 16:48 ` [PATCH 5/9] ksm: share anon page without allocating Hugh Dickins
2009-11-30 0:04 ` KAMEZAWA Hiroyuki
2009-11-30 11:18 ` Hugh Dickins
2009-12-01 0:02 ` KAMEZAWA Hiroyuki
2009-11-24 16:51 ` [PATCH 6/9] ksm: mem cgroup charge swapin copy Hugh Dickins
2009-11-25 14:23 ` Balbir Singh
2009-11-25 17:12 ` Hugh Dickins
2009-11-25 17:36 ` Balbir Singh
2009-11-30 0:13 ` KAMEZAWA Hiroyuki
2009-11-30 11:40 ` Hugh Dickins
2009-11-24 16:54 ` [PATCH 7/9] ksm: rmap_walk to remove_migation_ptes Hugh Dickins
2009-11-24 16:56 ` [PATCH 8/9] ksm: memory hotremove migration only Hugh Dickins
2009-11-24 16:57 ` [PATCH 9/9] ksm: remove unswappable max_kernel_pages Hugh Dickins
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=20091202125501.GD28697@random.random \
--to=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=chrisw@redhat.com \
--cc=hugh.dickins@tiscali.co.uk \
--cc=ieidus@redhat.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox