From: Rik van Riel <riel@redhat.com>
To: linux-kernel@vger.kernel.org
Cc: Andrew Morton <akpm@linux-foundation.org>,
Lee Schermerhorn <lee.schermerhorn@hp.com>,
Kosaki Motohiro <kosaki.motohiro@jp.fujitsu.com>
Subject: [PATCH -mm 15/16] ramfs pages are non-reclaimable
Date: Fri, 23 May 2008 15:55:21 -0400 [thread overview]
Message-ID: <20080523195535.852019704@redhat.com> (raw)
In-Reply-To: 20080523195506.084894989@redhat.com
[-- Attachment #1: rvr-13-lts-noreclaim-ramfs-pages-are-non-reclaimable.patch --]
[-- Type: text/plain, Size: 4996 bytes --]
From: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
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_LRU.
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
---
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.
drivers/block/brd.c | 13 +++++++++++++
fs/ramfs/inode.c | 1 +
include/linux/pagemap.h | 22 ++++++++++++++++++++++
mm/vmscan.c | 5 +++++
4 files changed, 41 insertions(+)
Index: linux-2.6.26-rc2-mm1/include/linux/pagemap.h
===================================================================
--- linux-2.6.26-rc2-mm1.orig/include/linux/pagemap.h 2008-05-23 15:14:03.000000000 -0400
+++ linux-2.6.26-rc2-mm1/include/linux/pagemap.h 2008-05-23 15:19:21.000000000 -0400
@@ -30,6 +30,28 @@ static inline void mapping_set_error(str
}
}
+#ifdef CONFIG_NORECLAIM_LRU
+#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-2.6.26-rc2-mm1/mm/vmscan.c
===================================================================
--- linux-2.6.26-rc2-mm1.orig/mm/vmscan.c 2008-05-23 15:19:11.000000000 -0400
+++ linux-2.6.26-rc2-mm1/mm/vmscan.c 2008-05-23 15:19:21.000000000 -0400
@@ -2301,6 +2301,8 @@ int zone_reclaim(struct zone *zone, gfp_
* lists vs noreclaim list.
*
* Reasons page might not be reclaimable:
+ * (1) page's mapping marked non-reclaimable
+ *
* TODO - later patches
*/
int page_reclaimable(struct page *page, struct vm_area_struct *vma)
@@ -2308,6 +2310,9 @@ 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-2.6.26-rc2-mm1/fs/ramfs/inode.c
===================================================================
--- linux-2.6.26-rc2-mm1.orig/fs/ramfs/inode.c 2008-05-23 15:14:03.000000000 -0400
+++ linux-2.6.26-rc2-mm1/fs/ramfs/inode.c 2008-05-23 15:19:21.000000000 -0400
@@ -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-2.6.26-rc2-mm1/drivers/block/brd.c
===================================================================
--- linux-2.6.26-rc2-mm1.orig/drivers/block/brd.c 2008-05-23 15:14:03.000000000 -0400
+++ linux-2.6.26-rc2-mm1/drivers/block/brd.c 2008-05-23 15:19:21.000000000 -0400
@@ -374,8 +374,21 @@ 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
next prev parent reply other threads:[~2008-05-23 22:10 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-05-23 19:55 [PATCH -mm 00/16] VM pageout scalability improvements (V8) Rik van Riel
2008-05-23 19:55 ` [PATCH -mm 01/16] move isolate_lru_page() to vmscan.c Rik van Riel
2008-05-23 19:55 ` [PATCH -mm 02/16] Use an indexed array for LRU variables Rik van Riel
2008-05-27 16:54 ` Lee Schermerhorn
2008-05-27 17:03 ` Rik van Riel
2008-05-23 19:55 ` [PATCH -mm 03/16] use an array for the LRU pagevecs Rik van Riel
2008-05-23 19:55 ` [PATCH -mm 04/16] free swap space on swap-in/activation Rik van Riel
2008-05-28 9:08 ` Daisuke Nishimura
2008-05-23 19:55 ` [PATCH -mm 05/16] define page_file_cache() function Rik van Riel
2008-05-23 19:55 ` [PATCH -mm 06/16] split LRU lists into anon & file sets Rik van Riel
2008-05-23 19:55 ` [PATCH -mm 07/16] second chance replacement for anonymous pages Rik van Riel
2008-05-28 5:36 ` Daisuke Nishimura
2008-05-28 13:39 ` Rik van Riel
2008-05-28 15:42 ` Daisuke Nishimura
2008-05-28 16:08 ` Rik van Riel
2008-05-28 11:03 ` KOSAKI Motohiro
2008-05-28 13:43 ` Rik van Riel
2008-05-23 19:55 ` [PATCH -mm 08/16] add some sanity checks to get_scan_ratio Rik van Riel
2008-05-23 19:55 ` [PATCH -mm 09/16] fix pagecache reclaim referenced bit check Rik van Riel
2008-05-23 19:55 ` [PATCH -mm 10/16] add newly swapped in pages to the inactive list Rik van Riel
2008-05-23 19:55 ` [PATCH -mm 11/16] more aggressively use lumpy reclaim Rik van Riel
2008-05-23 19:55 ` [PATCH -mm 12/16] pageflag helpers for configed-out flags Rik van Riel
2008-05-23 19:55 ` [PATCH -mm 13/16] No Reclaim LRU Infrastructure Rik van Riel
2008-05-23 19:55 ` [PATCH -mm 14/16] Non-reclaimable page statistics Rik van Riel
2008-05-23 19:55 ` Rik van Riel [this message]
2008-05-23 19:55 ` [PATCH -mm 16/16] SHM_LOCKED pages are nonreclaimable Rik van Riel
2008-05-26 18:24 ` [PATCH -mm 00/16] VM pageout scalability improvements (V8) Balbir Singh
2008-05-26 19:33 ` Rik van Riel
2008-05-27 15:54 ` Lee Schermerhorn
2008-05-27 16:10 ` Balbir Singh
2008-05-28 1:12 ` KAMEZAWA Hiroyuki
2008-05-28 11:04 ` [RFC PATCH] No Reclaim LRU Infrastructure enhancement for memcgroup KOSAKI Motohiro
2008-05-29 2:30 ` Balbir Singh
2008-05-29 11:14 ` Daisuke Nishimura
2008-05-28 11:49 ` [PATCH -mm 00/16] VM pageout scalability improvements (V8) Balbir Singh
2008-05-28 13:33 ` KOSAKI Motohiro
2008-05-28 13:36 ` Balbir Singh
2008-05-29 12:47 ` Carsten Otte
2008-05-29 14:43 ` 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=20080523195535.852019704@redhat.com \
--to=riel@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=lee.schermerhorn@hp.com \
--cc=linux-kernel@vger.kernel.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