[PATCH 2/4]JBD2: Group short-lived and reclaimable kernel allocations

[Mingming Cao <cmm@us.ibm.com>]

JBD2: Group short-lived and reclaimable kernel allocations
From: Mingming Cao <cmm@us.ibm.com>
Ported from JBD to JBD2

--------------------------------
From: Mel Gorman <mel@csn.ul.ie>
Date: Tue, 16 Oct 2007 08:25:52 +0000 (-0700)
Subject: Group short-lived and reclaimable kernel allocations
X-Git-Tag: v2.6.24-rc1~1137
X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=e12ba74d8ff3e2f73a583500d7095e406df4d093

Group short-lived and reclaimable kernel allocations

This patch marks a number of allocations that are either short-lived such as
network buffers or are reclaimable such as inode allocations.  When something
like updatedb is called, long-lived and unmovable kernel allocations tend to
be spread throughout the address space which increases fragmentation.

This patch groups these allocations together as much as possible by adding a
new MIGRATE_TYPE.  The MIGRATE_RECLAIMABLE type is for allocations that can be
reclaimed on demand, but not moved.  i.e.  they can be migrated by deleting
them and re-reading the information from elsewhere.

Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
---

---
 fs/jbd2/journal.c |    4 ++--
 fs/jbd2/revoke.c  |    6 ++++--
 2 files changed, 6 insertions(+), 4 deletions(-)

Index: linux-2.6.24-rc7/fs/jbd2/journal.c
===================================================================
--- linux-2.6.24-rc7.orig/fs/jbd2/journal.c	2008-01-16 15:02:40.000000000 -0800
+++ linux-2.6.24-rc7/fs/jbd2/journal.c	2008-01-16 17:40:24.000000000 -0800
@@ -1975,7 +1975,7 @@ static int journal_init_jbd2_journal_hea
 	jbd2_journal_head_cache = kmem_cache_create("jbd2_journal_head",
 				sizeof(struct journal_head),
 				0,		/* offset */
-				0,		/* flags */
+				SLAB_TEMPORARY,	/* flags */
 				NULL);		/* ctor */
 	retval = 0;
 	if (jbd2_journal_head_cache == 0) {
@@ -2271,7 +2271,7 @@ static int __init journal_init_handle_ca
 	jbd2_handle_cache = kmem_cache_create("jbd2_journal_handle",
 				sizeof(handle_t),
 				0,		/* offset */
-				0,		/* flags */
+				SLAB_TEMPORARY,	/* flags */
 				NULL);		/* ctor */
 	if (jbd2_handle_cache == NULL) {
 		printk(KERN_EMERG "JBD: failed to create handle cache\n");
Index: linux-2.6.24-rc7/fs/jbd2/revoke.c
===================================================================
--- linux-2.6.24-rc7.orig/fs/jbd2/revoke.c	2008-01-06 13:45:38.000000000 -0800
+++ linux-2.6.24-rc7/fs/jbd2/revoke.c	2008-01-16 17:40:24.000000000 -0800
@@ -171,13 +171,15 @@ int __init jbd2_journal_init_revoke_cach
 {
 	jbd2_revoke_record_cache = kmem_cache_create("jbd2_revoke_record",
 					   sizeof(struct jbd2_revoke_record_s),
-					   0, SLAB_HWCACHE_ALIGN, NULL);
+					   0,
+					   SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY,
+					   NULL);
 	if (jbd2_revoke_record_cache == 0)
 		return -ENOMEM;
 
 	jbd2_revoke_table_cache = kmem_cache_create("jbd2_revoke_table",
 					   sizeof(struct jbd2_revoke_table_s),
-					   0, 0, NULL);
+					   0, SLAB_TEMPORARY, NULL);
 	if (jbd2_revoke_table_cache == 0) {
 		kmem_cache_destroy(jbd2_revoke_record_cache);
 		jbd2_revoke_record_cache = NULL;