[PATCH] page migration: sys_move_pages(): support moving of individual pages
move_pages() is used to move individual pages of a process. The function can
be used to determine the location of pages and to move them onto the desired
node. move_pages() returns status information for each page.
long move_pages(pid, number_of_pages_to_move,
addresses_of_pages[],
nodes[] or NULL,
status[],
flags);
The addresses of pages is an array of void * pointing to the
pages to be moved.
The nodes array contains the node numbers that the pages should be moved
to. If a NULL is passed instead of an array then no pages are moved but
the status array is updated. The status request may be used to determine
the page state before issuing another move_pages() to move pages.
The status array will contain the state of all individual page migration
attempts when the function terminates. The status array is only valid if
move_pages() completed successfullly.
Possible page states in status[]:
0..MAX_NUMNODES The page is now on the indicated node.
-ENOENT Page is not present
-EACCES Page is mapped by multiple processes and can only
be moved if MPOL_MF_MOVE_ALL is specified.
-EPERM The page has been mlocked by a process/driver and
cannot be moved.
-EBUSY Page is busy and cannot be moved. Try again later.
-EFAULT Invalid address (no VMA or zero page).
-ENOMEM Unable to allocate memory on target node.
-EIO Unable to write back page. The page must be written
back in order to move it since the page is dirty and the
filesystem does not provide a migration function that
would allow the moving of dirty pages.
-EINVAL A dirty page cannot be moved. The filesystem does not provide
a migration function and has no ability to write back pages.
The flags parameter indicates what types of pages to move:
MPOL_MF_MOVE Move pages that are only mapped by the process.
MPOL_MF_MOVE_ALL Also move pages that are mapped by multiple processes.
Requires sufficient capabilities.
Possible return codes from move_pages()
-ENOENT No pages found that would require moving. All pages
are either already on the target node, not present, had an
invalid address or could not be moved because they were
mapped by multiple processes.
-EINVAL Flags other than MPOL_MF_MOVE(_ALL) specified or an attempt
to migrate pages in a kernel thread.
-EPERM MPOL_MF_MOVE_ALL specified without sufficient priviledges.
or an attempt to move a process belonging to another user.
-EACCES One of the target nodes is not allowed by the current cpuset.
-ENODEV One of the target nodes is not online.
-ESRCH Process does not exist.
-E2BIG Too many pages to move.
-ENOMEM Not enough memory to allocate control array.
-EFAULT Parameters could not be accessed.
A test program for move_pages() may be found with the patches
on ftp.kernel.org:/pub/linux/kernel/people/christoph/pmig/patches-2.6.17-rc4-mm3
From: Christoph Lameter <clameter@sgi.com>
Detailed results for sys_move_pages()
Pass a pointer to an integer to get_new_page() that may be used to
indicate where the completion status of a migration operation should be
placed. This allows sys_move_pags() to report back exactly what happened to
each page.
Wish there would be a better way to do this. Looks a bit hacky.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Jes Sorensen <jes@trained-monkey.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Michael Kerrisk <mtk-manpages@gmx.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
diff --git a/Documentation/vm/page_migration b/Documentation/vm/page_migration
index 0a5d5fb1..99f89aa 100644
--- a/Documentation/vm/page_migration
+++ b/Documentation/vm/page_migration
@@ -26,8 +26,13 @@
Manual migration is useful if for example the scheduler has relocated
a process to a processor on a distant node. A batch scheduler or an
administrator may detect the situation and move the pages of the process
-nearer to the new processor. At some point in the future we may have
-some mechanism in the scheduler that will automatically move the pages.
+nearer to the new processor. The kernel itself does only provide
+manual page migration support. Automatic page migration may be implemented
+through user space processes that move pages. A special function call
+"move_pages" allows the moving of individual pages within a process.
+A NUMA profiler may f.e. obtain a log showing frequent off node
+accesses and may use the result to move pages to more advantageous
+locations.
Larger installations usually partition the system using cpusets into
sections of nodes. Paul Jackson has equipped cpusets with the ability to
@@ -62,22 +67,14 @@
It also prevents the swapper or other scans to encounter
the page.
-2. Generate a list of newly allocates pages. These pages will contain the
- contents of the pages from the first list after page migration is
- complete.
+2. We need to have a function of type new_page_t that can be
+ passed to migrate_pages(). This function should figure out
+ how to allocate the correct new page given the old page.
3. The migrate_pages() function is called which attempts
- to do the migration. It returns the moved pages in the
- list specified as the third parameter and the failed
- migrations in the fourth parameter. When the function
- returns the first list will contain the pages that could still be retried.
-
-4. The leftover pages of various types are returned
- to the LRU using putback_to_lru_pages() or otherwise
- disposed of. The pages will still have the refcount as
- increased by isolate_lru_pages() if putback_to_lru_pages() is not
- used! The kernel may want to handle the various cases of failures in
- different ways.
+ to do the migration. It will call the function to allocate
+ the new page for each page that is considered for
+ moving.
B. How migrate_pages() works
----------------------------
diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index bcb80ca..32c999f 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -1584,7 +1584,7 @@
data8 sys_keyctl
data8 sys_ioprio_set
data8 sys_ioprio_get // 1275
- data8 sys_ni_syscall
+ data8 sys_move_pages
data8 sys_inotify_init
data8 sys_inotify_add_watch
data8 sys_inotify_rm_watch
diff --git a/include/asm-ia64/unistd.h b/include/asm-ia64/unistd.h
index 632f2ee..bb0eb72 100644
--- a/include/asm-ia64/unistd.h
+++ b/include/asm-ia64/unistd.h
@@ -265,7 +265,7 @@
#define __NR_keyctl 1273
#define __NR_ioprio_set 1274
#define __NR_ioprio_get 1275
-/* 1276 is available for reuse (was briefly sys_set_zone_reclaim) */
+#define __NR_move_pages 1276
#define __NR_inotify_init 1277
#define __NR_inotify_add_watch 1278
#define __NR_inotify_rm_watch 1279
diff --git a/include/linux/migrate.h b/include/linux/migrate.h
index 5b95d65..5dba23a 100644
--- a/include/linux/migrate.h
+++ b/include/linux/migrate.h
@@ -3,7 +3,7 @@
#include <linux/mm.h>
-typedef struct page *new_page_t(struct page *, unsigned long private);
+typedef struct page *new_page_t(struct page *, unsigned long private, int **);
#ifdef CONFIG_MIGRATION
extern int isolate_lru_page(struct page *p, struct list_head *pagelist);
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index bd67a44..7e3f234 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -516,6 +516,11 @@
asmlinkage long sys_migrate_pages(pid_t pid, unsigned long maxnode,
const unsigned long __user *from,
const unsigned long __user *to);
+asmlinkage long sys_move_pages(pid_t pid, unsigned long nr_pages,
+ const void __user * __user *pages,
+ const int __user *nodes,
+ int __user *status,
+ int flags);
asmlinkage long sys_mbind(unsigned long start, unsigned long len,
unsigned long mode,
unsigned long __user *nmask,
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 5433195..5972297 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -87,6 +87,7 @@
cond_syscall(sys_inotify_add_watch);
cond_syscall(sys_inotify_rm_watch);
cond_syscall(sys_migrate_pages);
+cond_syscall(sys_move_pages);
cond_syscall(sys_chown16);
cond_syscall(sys_fchown16);
cond_syscall(sys_getegid16);
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index f432642..05b84ac 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -588,7 +588,7 @@
isolate_lru_page(page, pagelist);
}
-static struct page *new_node_page(struct page *page, unsigned long node)
+static struct page *new_node_page(struct page *page, unsigned long node, int **x)
{
return alloc_pages_node(node, GFP_HIGHUSER, 0);
}
@@ -698,7 +698,7 @@
}
-static struct page *new_vma_page(struct page *page, unsigned long private)
+static struct page *new_vma_page(struct page *page, unsigned long private, int **x)
{
struct vm_area_struct *vma = (struct vm_area_struct *)private;
diff --git a/mm/migrate.c b/mm/migrate.c
index 251a8d1..033a12f 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -25,6 +25,8 @@
#include <linux/cpu.h>
#include <linux/cpuset.h>
#include <linux/writeback.h>
+#include <linux/mempolicy.h>
+#include <linux/vmalloc.h>
#include "internal.h"
@@ -62,9 +64,8 @@
}
/*
- * migrate_prep() needs to be called after we have compiled the list of pages
- * to be migrated using isolate_lru_page() but before we begin a series of calls
- * to migrate_pages().
+ * migrate_prep() needs to be called before we start compiling a list of pages
+ * to be migrated using isolate_lru_page().
*/
int migrate_prep(void)
{
@@ -588,7 +589,8 @@
struct page *page, int force)
{
int rc = 0;
- struct page *newpage = get_new_page(page, private);
+ int *result = NULL;
+ struct page *newpage = get_new_page(page, private, &result);
if (!newpage)
return -ENOMEM;
@@ -642,6 +644,12 @@
* then this will free the page.
*/
move_to_lru(newpage);
+ if (result) {
+ if (rc)
+ *result = rc;
+ else
+ *result = page_to_nid(newpage);
+ }
return rc;
}
@@ -710,3 +718,255 @@
return nr_failed + retry;
}
+#ifdef CONFIG_NUMA
+/*
+ * Move a list of individual pages
+ */
+struct page_to_node {
+ unsigned long addr;
+ struct page *page;
+ int node;
+ int status;
+};
+
+static struct page *new_page_node(struct page *p, unsigned long private,
+ int **result)
+{
+ struct page_to_node *pm = (struct page_to_node *)private;
+
+ while (pm->node != MAX_NUMNODES && pm->page != p)
+ pm++;
+
+ if (pm->node == MAX_NUMNODES)
+ return NULL;
+
+ *result = &pm->status;
+
+ return alloc_pages_node(pm->node, GFP_HIGHUSER, 0);
+}
+
+/*
+ * Move a set of pages as indicated in the pm array. The addr
+ * field must be set to the virtual address of the page to be moved
+ * and the node number must contain a valid target node.
+ */
+static int do_move_pages(struct mm_struct *mm, struct page_to_node *pm,
+ int migrate_all)
+{
+ int err;
+ struct page_to_node *pp;
+ LIST_HEAD(pagelist);
+
+ down_read(&mm->mmap_sem);
+
+ /*
+ * Build a list of pages to migrate
+ */
+ migrate_prep();
+ for (pp = pm; pp->node != MAX_NUMNODES; pp++) {
+ struct vm_area_struct *vma;
+ struct page *page;
+
+ /*
+ * A valid page pointer that will not match any of the
+ * pages that will be moved.
+ */
+ pp->page = ZERO_PAGE(0);
+
+ err = -EFAULT;
+ vma = find_vma(mm, pp->addr);
+ if (!vma)
+ goto set_status;
+
+ page = follow_page(vma, pp->addr, FOLL_GET);
+ err = -ENOENT;
+ if (!page)
+ goto set_status;
+
+ if (PageReserved(page)) /* Check for zero page */
+ goto put_and_set;
+
+ pp->page = page;
+ err = page_to_nid(page);
+
+ if (err == pp->node)
+ /*
+ * Node already in the right place
+ */
+ goto put_and_set;
+
+ err = -EACCES;
+ if (page_mapcount(page) > 1 &&
+ !migrate_all)
+ goto put_and_set;
+
+ err = isolate_lru_page(page, &pagelist);
+put_and_set:
+ /*
+ * Either remove the duplicate refcount from
+ * isolate_lru_page() or drop the page ref if it was
+ * not isolated.
+ */
+ put_page(page);
+set_status:
+ pp->status = err;
+ }
+
+ if (!list_empty(&pagelist))
+ err = migrate_pages(&pagelist, new_page_node,
+ (unsigned long)pm);
+ else
+ err = -ENOENT;
+
+ up_read(&mm->mmap_sem);
+ return err;
+}
+
+/*
+ * Determine the nodes of a list of pages. The addr in the pm array
+ * must have been set to the virtual address of which we want to determine
+ * the node number.
+ */
+static int do_pages_stat(struct mm_struct *mm, struct page_to_node *pm)
+{
+ down_read(&mm->mmap_sem);
+
+ for ( ; pm->node != MAX_NUMNODES; pm++) {
+ struct vm_area_struct *vma;
+ struct page *page;
+ int err;
+
+ err = -EFAULT;
+ vma = find_vma(mm, pm->addr);
+ if (!vma)
+ goto set_status;
+
+ page = follow_page(vma, pm->addr, 0);
+ err = -ENOENT;
+ /* Use PageReserved to check for zero page */
+ if (!page || PageReserved(page))
+ goto set_status;
+
+ err = page_to_nid(page);
+set_status:
+ pm->status = err;
+ }
+
+ up_read(&mm->mmap_sem);
+ return 0;
+}
+
+/*
+ * Move a list of pages in the address space of the currently executing
+ * process.
+ */
+asmlinkage long sys_move_pages(pid_t pid, unsigned long nr_pages,
+ const void __user * __user *pages,
+ const int __user *nodes,
+ int __user *status, int flags)
+{
+ int err = 0;
+ int i;
+ struct task_struct *task;
+ nodemask_t task_nodes;
+ struct mm_struct *mm;
+ struct page_to_node *pm = NULL;
+
+ /* Check flags */
+ if (flags & ~(MPOL_MF_MOVE|MPOL_MF_MOVE_ALL))
+ return -EINVAL;
+
+ if ((flags & MPOL_MF_MOVE_ALL) && !capable(CAP_SYS_NICE))
+ return -EPERM;
+
+ /* Find the mm_struct */
+ read_lock(&tasklist_lock);
+ task = pid ? find_task_by_pid(pid) : current;
+ if (!task) {
+ read_unlock(&tasklist_lock);
+ return -ESRCH;
+ }
+ mm = get_task_mm(task);
+ read_unlock(&tasklist_lock);
+
+ if (!mm)
+ return -EINVAL;
+
+ /*
+ * Check if this process has the right to modify the specified
+ * process. The right exists if the process has administrative
+ * capabilities, superuser privileges or the same
+ * userid as the target process.
+ */
+ if ((current->euid != task->suid) && (current->euid != task->uid) &&
+ (current->uid != task->suid) && (current->uid != task->uid) &&
+ !capable(CAP_SYS_NICE)) {
+ err = -EPERM;
+ goto out2;
+ }
+
+ task_nodes = cpuset_mems_allowed(task);
+
+ /* Limit nr_pages so that the multiplication may not overflow */
+ if (nr_pages >= ULONG_MAX / sizeof(struct page_to_node) - 1) {
+ err = -E2BIG;
+ goto out2;
+ }
+
+ pm = vmalloc((nr_pages + 1) * sizeof(struct page_to_node));
+ if (!pm) {
+ err = -ENOMEM;
+ goto out2;
+ }
+
+ /*
+ * Get parameters from user space and initialize the pm
+ * array. Return various errors if the user did something wrong.
+ */
+ for (i = 0; i < nr_pages; i++) {
+ const void *p;
+
+ err = -EFAULT;
+ if (get_user(p, pages + i))
+ goto out;
+
+ pm[i].addr = (unsigned long)p;
+ if (nodes) {
+ int node;
+
+ if (get_user(node, nodes + i))
+ goto out;
+
+ err = -ENODEV;
+ if (!node_online(node))
+ goto out;
+
+ err = -EACCES;
+ if (!node_isset(node, task_nodes))
+ goto out;
+
+ pm[i].node = node;
+ }
+ }
+ /* End marker */
+ pm[nr_pages].node = MAX_NUMNODES;
+
+ if (nodes)
+ err = do_move_pages(mm, pm, flags & MPOL_MF_MOVE_ALL);
+ else
+ err = do_pages_stat(mm, pm);
+
+ if (err >= 0)
+ /* Return status information */
+ for (i = 0; i < nr_pages; i++)
+ if (put_user(pm[i].status, status + i))
+ err = -EFAULT;
+
+out:
+ vfree(pm);
+out2:
+ mmput(mm);
+ return err;
+}
+#endif
+