linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Hugh Dickins <hugh.dickins@tiscali.co.uk>
To: Izik Eidus <ieidus@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>,
	Rik van Riel <riel@redhat.com>, Chris Wright <chrisw@redhat.com>,
	Nick Piggin <nickpiggin@yahoo.com.au>,
	Andrew Morton <akpm@linux-foundation.org>,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: [PATCH 4/12] ksm: break cow once unshared
Date: Mon, 3 Aug 2009 13:12:59 +0100 (BST)	[thread overview]
Message-ID: <Pine.LNX.4.64.0908031311590.16754@sister.anvils> (raw)
In-Reply-To: <Pine.LNX.4.64.0908031304430.16449@sister.anvils>

We kept agreeing not to bother about the unswappable shared KSM pages
which later become unshared by others: observation suggests they're not
a significant proportion.  But they are disadvantageous, and it is easier
to break COW to replace them by swappable pages, than offer statistics
to show that they don't matter; then we can stop worrying about them.

Doing this in ksm_do_scan, they don't go through cmp_and_merge_page on
this pass: give them a good chance of getting into the unstable tree
on the next pass, or back into the stable, by computing checksum now.

Signed-off-by: Hugh Dickins <hugh.dickins@tiscali.co.uk>
---

 mm/ksm.c |    8 ++++++++
 1 file changed, 8 insertions(+)

--- ksm3/mm/ksm.c	2009-08-02 13:49:51.000000000 +0100
+++ ksm4/mm/ksm.c	2009-08-02 13:49:59.000000000 +0100
@@ -1275,6 +1275,14 @@ static void ksm_do_scan(unsigned int sca
 			return;
 		if (!PageKsm(page) || !in_stable_tree(rmap_item))
 			cmp_and_merge_page(page, rmap_item);
+		else if (page_mapcount(page) == 1) {
+			/*
+			 * Replace now-unshared ksm page by ordinary page.
+			 */
+			break_cow(rmap_item->mm, rmap_item->address);
+			remove_rmap_item_from_tree(rmap_item);
+			rmap_item->oldchecksum = calc_checksum(page);
+		}
 		put_page(page);
 	}
 }

--
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:[~2009-08-03 11:53 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-08-03 12:08 [PATCH 0/12] ksm: stats, oom, doc, misc Hugh Dickins
2009-08-03 12:10 ` [PATCH 1/12] ksm: rename kernel_pages_allocated Hugh Dickins
2009-08-03 14:21   ` Izik Eidus
2009-08-03 16:48     ` Andrea Arcangeli
2009-08-03 12:11 ` [PATCH 2/12] ksm: move pages_sharing updates Hugh Dickins
2009-08-03 14:34   ` Izik Eidus
2009-08-03 16:53   ` Andrea Arcangeli
2009-08-03 17:34     ` Hugh Dickins
2009-08-03 12:11 ` [PATCH 3/12] ksm: pages_unshared and pages_volatile Hugh Dickins
2009-08-03 14:54   ` Izik Eidus
2009-08-04 21:49   ` Andrew Morton
2009-08-05 11:39     ` Hugh Dickins
2009-08-05 15:11       ` Andrea Arcangeli
2009-08-03 12:12 ` Hugh Dickins [this message]
2009-08-03 16:00   ` [PATCH 4/12] ksm: break cow once unshared Izik Eidus
2009-08-03 12:14 ` [PATCH 5/12] ksm: keep quiet while list empty Hugh Dickins
2009-08-03 16:55   ` Izik Eidus
2009-08-04 21:59   ` Andrew Morton
2009-08-05 11:54     ` Hugh Dickins
2009-08-03 12:15 ` [PATCH 6/12] ksm: five little cleanups Hugh Dickins
2009-08-04 12:41   ` Izik Eidus
2009-08-03 12:16 ` [PATCH 7/12] ksm: fix endless loop on oom Hugh Dickins
2009-08-04 12:55   ` Izik Eidus
2009-08-03 12:17 ` [PATCH 8/12] ksm: distribute remove_mm_from_lists Hugh Dickins
2009-08-04 13:03   ` Izik Eidus
2009-08-03 12:18 ` [PATCH 9/12] ksm: fix oom deadlock Hugh Dickins
2009-08-04 19:32   ` Izik Eidus
2009-08-25 14:58   ` Andrea Arcangeli
2009-08-25 15:22     ` [PATCH 13/12] ksm: fix munlock during exit_mmap deadlock Andrea Arcangeli
2009-08-25 17:49       ` Hugh Dickins
2009-08-25 18:10         ` Andrea Arcangeli
2009-08-25 18:58           ` Hugh Dickins
2009-08-25 19:45             ` Andrea Arcangeli
2009-08-26 16:18               ` Justin M. Forbes
2009-08-26 19:17               ` Hugh Dickins
2009-08-26 19:44                 ` Andrea Arcangeli
2009-08-26 19:57                   ` Hugh Dickins
2009-08-26 20:28                     ` Andrea Arcangeli
2009-08-26 20:54                     ` Izik Eidus
2009-08-26 21:14                       ` Andrea Arcangeli
2009-08-26 21:49                         ` Izik Eidus
2009-08-27 19:11                           ` Hugh Dickins
2009-08-27 19:35                             ` Izik Eidus
2009-08-26 22:00                         ` David Rientjes
2009-08-26 20:29                   ` Hugh Dickins
2009-08-25 17:35     ` [PATCH 9/12] ksm: fix oom deadlock Hugh Dickins
2009-08-25 17:47       ` Andrea Arcangeli
2009-08-03 12:19 ` [PATCH 10/12] ksm: sysfs and defaults Hugh Dickins
2009-08-04 19:34   ` Izik Eidus
2009-08-03 12:21 ` [PATCH 11/12] ksm: add some documentation Hugh Dickins
2009-08-04 19:35   ` Izik Eidus
2009-08-03 12:22 ` [PATCH 12/12] ksm: remove VM_MERGEABLE_FLAGS Hugh Dickins
2009-08-04 19:35   ` Izik Eidus

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=Pine.LNX.4.64.0908031311590.16754@sister.anvils \
    --to=hugh.dickins@tiscali.co.uk \
    --cc=aarcange@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=chrisw@redhat.com \
    --cc=ieidus@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=nickpiggin@yahoo.com.au \
    --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