linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: linux-kernel@vger.kernel.org
Cc: linux-mm@kvack.org, lee.schermerhorn@hp.com
Subject: [patch 13/19] ramfs pages are non-reclaimable
Date: Wed, 02 Jan 2008 17:41:57 -0500	[thread overview]
Message-ID: <20080102224154.910855018@redhat.com> (raw)
In-Reply-To: 20080102224144.885671949@redhat.com

[-- Attachment #1: noreclaim-02-ramdisk-and-ramfs-pages-are-nonreclaimable.patch --]
[-- Type: text/plain, Size: 4711 bytes --]

V3 -> V4:
+ drivers/block/rd.c was replaced by brd.c in 24-rc4-mm1.
  Update patch to add brd_open() to mark mapping as nonreclaimable

V2 -> V3:
+  rebase to 23-mm1 atop RvR's split LRU series [no changes]

V1 -> V2:
+  add ramfs pages to this class of non-reclaimable pages by
   marking ramfs address_space [mapping] as non-reclaimble.

Christoph Lameter pointed out that ram disk pages also clutter the
LRU lists.  When vmscan finds them dirty and tries to clean them,
the ram disk writeback function just redirties the page so that it
goes back onto the active list.  Round and round she goes...

Define new address_space flag [shares address_space flags member
with mapping's gfp mask] to indicate that the address space contains
all non-reclaimable pages.  This will provide for efficient testing
of ramdisk pages in page_reclaimable().

Also provide wrapper functions to set/test the noreclaim state to
minimize #ifdefs in ramdisk driver and any other users of this
facility.

Set the noreclaim state on address_space structures for new
ramdisk inodes.  Test the noreclaim state in page_reclaimable()
to cull non-reclaimable pages.

Similarly, ramfs pages are non-reclaimable.  Set the 'noreclaim'
address_space flag for new ramfs inodes.

These changes depend on [CONFIG_]NORECLAIM.


Signed-off-by:  Lee Schermerhorn <lee.schermerhorn@hp.com>
Signed-off-by:  Rik van Riel <riel@redhat.com>

Index: linux-2.6.24-rc6-mm1/include/linux/pagemap.h
===================================================================
--- linux-2.6.24-rc6-mm1.orig/include/linux/pagemap.h	2007-12-23 23:45:44.000000000 -0500
+++ linux-2.6.24-rc6-mm1/include/linux/pagemap.h	2008-01-02 13:22:23.000000000 -0500
@@ -30,6 +30,28 @@ static inline void mapping_set_error(str
 	}
 }
 
+#ifdef CONFIG_NORECLAIM
+#define AS_NORECLAIM	(__GFP_BITS_SHIFT + 2)	/* e.g., ramdisk, SHM_LOCK */
+
+static inline void mapping_set_noreclaim(struct address_space *mapping)
+{
+	set_bit(AS_NORECLAIM, &mapping->flags);
+}
+
+static inline int mapping_non_reclaimable(struct address_space *mapping)
+{
+	if (mapping && (mapping->flags & AS_NORECLAIM))
+		return 1;
+	return 0;
+}
+#else
+static inline void mapping_set_noreclaim(struct address_space *mapping) { }
+static inline int mapping_non_reclaimable(struct address_space *mapping)
+{
+	return 0;
+}
+#endif
+
 static inline gfp_t mapping_gfp_mask(struct address_space * mapping)
 {
 	return (__force gfp_t)mapping->flags & __GFP_BITS_MASK;
Index: linux-2.6.24-rc6-mm1/mm/vmscan.c
===================================================================
--- linux-2.6.24-rc6-mm1.orig/mm/vmscan.c	2008-01-02 13:07:09.000000000 -0500
+++ linux-2.6.24-rc6-mm1/mm/vmscan.c	2008-01-02 13:22:23.000000000 -0500
@@ -2237,6 +2237,7 @@ int zone_reclaim(struct zone *zone, gfp_
  *               If !NULL, called from fault path.
  *
  * Reasons page might not be reclaimable:
+ * + page's mapping marked non-reclaimable
  * TODO - later patches
  *
  * TODO:  specify locking assumptions
@@ -2246,6 +2247,9 @@ int page_reclaimable(struct page *page, 
 
 	VM_BUG_ON(PageNoreclaim(page));
 
+	if (mapping_non_reclaimable(page_mapping(page)))
+		return 0;
+
 	/* TODO:  test page [!]reclaimable conditions */
 
 	return 1;
Index: linux-2.6.24-rc6-mm1/fs/ramfs/inode.c
===================================================================
--- linux-2.6.24-rc6-mm1.orig/fs/ramfs/inode.c	2007-12-23 23:45:35.000000000 -0500
+++ linux-2.6.24-rc6-mm1/fs/ramfs/inode.c	2008-01-02 13:22:23.000000000 -0500
@@ -61,6 +61,7 @@ struct inode *ramfs_get_inode(struct sup
 		inode->i_mapping->a_ops = &ramfs_aops;
 		inode->i_mapping->backing_dev_info = &ramfs_backing_dev_info;
 		mapping_set_gfp_mask(inode->i_mapping, GFP_HIGHUSER);
+		mapping_set_noreclaim(inode->i_mapping);
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 		switch (mode & S_IFMT) {
 		default:
Index: linux-2.6.24-rc6-mm1/drivers/block/brd.c
===================================================================
--- linux-2.6.24-rc6-mm1.orig/drivers/block/brd.c	2007-12-23 23:45:43.000000000 -0500
+++ linux-2.6.24-rc6-mm1/drivers/block/brd.c	2008-01-02 13:24:18.000000000 -0500
@@ -373,8 +373,21 @@ static int brd_ioctl(struct inode *inode
 	return error;
 }
 
+/*
+ * brd_open():
+ * Just mark the mapping as containing non-reclaimable pages
+ */
+static int brd_open(struct inode *inode, struct file *filp)
+{
+	struct address_space *mapping = inode->i_mapping;
+
+	mapping_set_noreclaim(mapping);
+	return 0;
+}
+
 static struct block_device_operations brd_fops = {
 	.owner =		THIS_MODULE,
+	.open  =		brd_open,
 	.ioctl =		brd_ioctl,
 #ifdef CONFIG_BLK_DEV_XIP
 	.direct_access =	brd_direct_access,

-- 
All Rights Reversed


  parent reply	other threads:[~2008-01-02 22:41 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-01-02 22:41 [patch 00/19] VM pageout scalability improvements linux-kernel
2008-01-02 22:41 ` [patch 01/19] move isolate_lru_page() to vmscan.c linux-kernel
2008-01-02 22:41 ` [patch 02/19] free swap space on swap-in/activation linux-kernel
2008-01-02 22:41 ` [patch 03/19] define page_file_cache() function linux-kernel
2008-01-02 22:41 ` [patch 04/19] debugging checks for page_file_cache() linux-kernel
2008-01-02 22:41 ` [patch 05/19] Use an indexed array for LRU variables linux-kernel
2008-01-02 22:41 ` [patch 06/19] split LRU lists into anon & file sets linux-kernel
2008-01-07  9:23   ` KAMEZAWA Hiroyuki
2008-01-02 22:41 ` [patch 07/19] split anon & file LRUs for memcontrol code linux-kernel
2008-01-07 10:04   ` KAMEZAWA Hiroyuki
2008-01-07 14:10     ` Balbir Singh
2008-01-07 15:23     ` Rik van Riel
2008-01-02 22:41 ` [patch 08/19] SEQ replacement for anonymous pages linux-kernel
2008-01-02 22:41 ` [patch 09/19] add newly swapped in pages to the inactive list linux-kernel
2008-01-02 22:41 ` [patch 10/19] No Reclaim LRU Infrastructure linux-kernel
2008-01-02 22:41 ` [patch 11/19] Non-reclaimable page statistics linux-kernel
2008-01-02 22:41 ` [patch 12/19] scan noreclaim list for reclaimable pages linux-kernel
2008-01-02 22:41 ` linux-kernel [this message]
2008-01-02 22:41 ` [patch 14/19] SHM_LOCKED pages are nonreclaimable linux-kernel
2008-01-02 22:41 ` [patch 15/19] non-reclaimable mlocked pages linux-kernel
2008-01-02 22:42 ` [patch 16/19] mlock vma pages under mmap_sem held for read linux-kernel
2008-01-02 22:42 ` [patch 17/19] handle mlocked pages during map/unmap and truncate linux-kernel
2008-01-02 22:42 ` [patch 18/19] account mlocked pages linux-kernel
2008-01-02 22:42 ` [patch 19/19] cull non-reclaimable anon pages from the LRU at fault time linux-kernel
2008-01-02 23:17 ` [patch 00/19] VM pageout scalability - one big patch Rik van Riel
2008-01-03  3:44 ` [patch 00/19] VM pageout scalability improvements Rik van Riel
2008-01-10  2:39   ` Christoph Lameter
2008-01-10  3:14     ` Rik van Riel
2008-01-03 16:52 ` Lee Schermerhorn
2008-01-03 17:00   ` Rik van Riel
2008-01-03 17:13     ` Lee Schermerhorn
2008-01-03 22:00       ` Rik van Riel
2008-01-04 16:25         ` Lee Schermerhorn
2008-01-04 16:34           ` Andi Kleen
2008-01-04 16:55             ` Rik van Riel
2008-01-04 18:07               ` Larry Woodman
2008-01-04 17:06             ` Lee Schermerhorn
2008-01-07 19:07               ` Christoph Lameter
2008-01-07 19:32                 ` Rik van Riel
2008-01-07 10:06     ` KAMEZAWA Hiroyuki
2008-01-07 15:18       ` Rik van Riel
  -- strict thread matches above, loose matches on Subject: below --
2008-01-08 20:59 Rik van Riel
2008-01-08 20:59 ` [patch 13/19] ramfs pages are non-reclaimable Rik van Riel

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=20080102224154.910855018@redhat.com \
    --to=linux-kernel@vger.kernel.org \
    --cc=lee.schermerhorn@hp.com \
    --cc=linux-mm@kvack.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;
as well as URLs for NNTP newsgroup(s).