Skip to contentSkip to conversation

add migratepage address space op to shmem

PATCH In review Lee Schermerhorn 3 participants 4 messages Apr 20, 2006

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 add migratepage address space op to shmem

Add migratepage address space op to shmem

Basic problem: pages of a shared memory segment can only be migrated once.

In 2.6.16 through 2.6.17-rc1, shared memory mappings do not have a migratepage address space op. Therefore, migrate_pages() falls back to default processing. In this path, it will try to pageout() dirty pages. Once a shared memory page has been migrated it becomes dirty, so migrate_pages() will try to page it out. However, because the page count is 3 [cache + current + pte], pageout() will return PAGE_KEEP because is_page_cache_freeable() returns false. This will abort all subsequent migrations.

This patch adds a migratepage address space op to shared memory segments to avoid taking the default path. We use the "migrate_page()" function because it knows how to migrate dirty pages. This allows shared memory segment pages to migrate, subject to other conditions such as # pte's referencing the page [page_mapcount(page)], when requested.

I think this is safe. If we're migrating a shared memory page, then we found the page via a page table, so it must be in memory.

Can be verified with memtoy and the shmem-mbind-test script, both available at: http://free.linux.hp.com/~lts/Tools/

mm/shmem.c +3 -0
Index: linux-2.6.17-rc1-mm3/mm/shmem.c
===================================================================
@@ -46,6 +46,8 @@
 #include <linux/mempolicy.h>
 #include <linux/namei.h>
 #include <linux/ctype.h>
+#include <linux/migrate.h>
+
 #include <asm/uaccess.h>
 #include <asm/div64.h>
 #include <asm/pgtable.h>
@@ -2165,6 +2167,7 @@ static struct address_space_operations s
 	.prepare_write	= shmem_prepare_write,
 	.commit_write	= simple_commit_write,
 #endif
+	.migratepage	= migrate_page,
 };
 
 static struct file_operations shmem_file_operations = {

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

#1 lore raw
Quoted 47 lines · Lee Schermerhorn

Add migratepage address space op to shmem

Basic problem: pages of a shared memory segment can only be migrated once.

In 2.6.16 through 2.6.17-rc1, shared memory mappings do not have a migratepage address space op. Therefore, migrate_pages() falls back to default processing. In this path, it will try to pageout() dirty pages. Once a shared memory page has been migrated it becomes dirty, so migrate_pages() will try to page it out. However, because the page count is 3 [cache + current + pte], pageout() will return PAGE_KEEP because is_page_cache_freeable() returns false. This will abort all subsequent migrations.

This patch adds a migratepage address space op to shared memory segments to avoid taking the default path. We use the "migrate_page()" function because it knows how to migrate dirty pages. This allows shared memory segment pages to migrate, subject to other conditions such as # pte's referencing the page [page_mapcount(page)], when requested.

I think this is safe. If we're migrating a shared memory page, then we found the page via a page table, so it must be in memory.

Can be verified with memtoy and the shmem-mbind-test script, both available at: http://free.linux.hp.com/~lts/Tools/

Index: linux-2.6.17-rc1-mm3/mm/shmem.c
===================================================================

mm/shmem.c +2 -0
@@ -46,6 +46,8 @@
 #include <linux/mempolicy.h>
 #include <linux/namei.h>
 #include <linux/ctype.h>
+#include <linux/migrate.h>
+
 #include <asm/uaccess.h>
 #include <asm/div64.h>
 #include <asm/pgtable.h>

mm/shmem.c static struct address_space_operations s · line 2173

Quoted 7 lines
mm/shmem.c +1 -0
@@ -2165,6 +2167,7 @@ static struct address_space_operations s
 	.prepare_write	= shmem_prepare_write,
 	.commit_write	= simple_commit_write,
 #endif
+	.migratepage	= migrate_page,
 };
 
 static struct file_operations shmem_file_operations = {

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

#2 lore raw
In reply to Lee Schermerhorn
Quoted 9 lines · Lee Schermerhorn

Add migratepage address space op to shmem

Basic problem: pages of a shared memory segment can only be migrated once.

In 2.6.16 through 2.6.17-rc1, shared memory mappings do not have a migratepage address space op. Therefore, migrate_pages() falls back to default processing.

This sounds to me like a bugfix-for-2.6.17 rather than a "PATCH 2.6.17-rc1-mm3"?

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

#3 lore raw
Quoted 6 lines · Andrew Morton
Quoted 3 lines

In 2.6.16 through 2.6.17-rc1, shared memory mappings do not have a migratepage address space op. Therefore, migrate_pages() falls back to default processing.

This sounds to me like a bugfix-for-2.6.17 rather than a "PATCH 2.6.17-rc1-mm3"?

Correct.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>