[PATCH 1/4] mempolicies: private pointer in check_range and MPOL_MF_INVERT

[PATCH 1/4] mempolicies: private pointer in check_range and MPOL_MF_INVERT

[Christoph Lameter]

[PATCH] memory policies: Genericize check_range()

This was was first posted at
http://marc.theaimsgroup.com/?l=linux-mm&m=113149240227584&w=2

(Part of this functionality is also contained in the direct migration
pathset. The functionality here is more generic and independent of that
patchset.)

- Add internal flags MPOL_MF_INVERT to control check_range() behavior.

- Replace the pagelist passed through by check_range by a general
  private pointer that may be used for other purposes.
  (The following patches will use that to merge numa_maps into
  mempolicy.c and to better group the page migration code in
  the policy layer)

- Improve some comments.

Signed-off-by: Christoph Lameter <clameter@sgi.com>

Index: linux-2.6.15-rc1-mm2/mm/mempolicy.c
===================================================================
--- linux-2.6.15-rc1-mm2.orig/mm/mempolicy.c	2005-11-21 10:51:46.000000000 -0800
+++ linux-2.6.15-rc1-mm2/mm/mempolicy.c	2005-11-21 12:02:32.000000000 -0800
@@ -88,8 +88,9 @@
 #include <asm/tlbflush.h>
 #include <asm/uaccess.h>
 
-/* Internal MPOL_MF_xxx flags */
+/* Internal flags */
 #define MPOL_MF_DISCONTIG_OK (MPOL_MF_INTERNAL << 0)	/* Skip checks for continuous vmas */
+#define MPOL_MF_INVERT (MPOL_MF_INTERNAL << 1)		/* Invert check for nodemask */
 
 static kmem_cache_t *policy_cache;
 static kmem_cache_t *sn_cache;
@@ -232,11 +233,11 @@ static void migrate_page_add(struct vm_a
 	}
 }
 
-/* Ensure all existing pages follow the policy. */
+/* Scan through pages checking if pages follow certain conditions. */
 static int check_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
 		unsigned long addr, unsigned long end,
 		const nodemask_t *nodes, unsigned long flags,
-		struct list_head *pagelist)
+		void *private)
 {
 	pte_t *orig_pte;
 	pte_t *pte;
@@ -246,6 +247,7 @@ static int check_pte_range(struct vm_are
 	do {
 		unsigned long pfn;
 		unsigned int nid;
+		struct page *page;
 
 		if (!pte_present(*pte))
 			continue;
@@ -254,15 +256,16 @@ static int check_pte_range(struct vm_are
 			print_bad_pte(vma, *pte, addr);
 			continue;
 		}
-		nid = pfn_to_nid(pfn);
-		if (!node_isset(nid, *nodes)) {
-			if (pagelist) {
-				struct page *page = pfn_to_page(pfn);
+		page = pfn_to_page(pfn);
+		nid = page_to_nid(page);
+		if (node_isset(nid, *nodes) == !!(flags & MPOL_MF_INVERT))
+			continue;
+
+		if (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL))
+			migrate_page_add(vma, page, private, flags);
+		else
+			break;
 
-				migrate_page_add(vma, page, pagelist, flags);
-			} else
-				break;
-		}
 	} while (pte++, addr += PAGE_SIZE, addr != end);
 	pte_unmap_unlock(orig_pte, ptl);
 	return addr != end;
@@ -271,7 +274,7 @@ static int check_pte_range(struct vm_are
 static inline int check_pmd_range(struct vm_area_struct *vma, pud_t *pud,
 		unsigned long addr, unsigned long end,
 		const nodemask_t *nodes, unsigned long flags,
-		struct list_head *pagelist)
+		void *private)
 {
 	pmd_t *pmd;
 	unsigned long next;
@@ -282,7 +285,7 @@ static inline int check_pmd_range(struct
 		if (pmd_none_or_clear_bad(pmd))
 			continue;
 		if (check_pte_range(vma, pmd, addr, next, nodes,
-				    flags, pagelist))
+				    flags, private))
 			return -EIO;
 	} while (pmd++, addr = next, addr != end);
 	return 0;
@@ -291,7 +294,7 @@ static inline int check_pmd_range(struct
 static inline int check_pud_range(struct vm_area_struct *vma, pgd_t *pgd,
 		unsigned long addr, unsigned long end,
 		const nodemask_t *nodes, unsigned long flags,
-		struct list_head *pagelist)
+		void *private)
 {
 	pud_t *pud;
 	unsigned long next;
@@ -302,7 +305,7 @@ static inline int check_pud_range(struct
 		if (pud_none_or_clear_bad(pud))
 			continue;
 		if (check_pmd_range(vma, pud, addr, next, nodes,
-				    flags, pagelist))
+				    flags, private))
 			return -EIO;
 	} while (pud++, addr = next, addr != end);
 	return 0;
@@ -311,7 +314,7 @@ static inline int check_pud_range(struct
 static inline int check_pgd_range(struct vm_area_struct *vma,
 		unsigned long addr, unsigned long end,
 		const nodemask_t *nodes, unsigned long flags,
-		struct list_head *pagelist)
+		void *private)
 {
 	pgd_t *pgd;
 	unsigned long next;
@@ -322,7 +325,7 @@ static inline int check_pgd_range(struct
 		if (pgd_none_or_clear_bad(pgd))
 			continue;
 		if (check_pud_range(vma, pgd, addr, next, nodes,
-				    flags, pagelist))
+				    flags, private))
 			return -EIO;
 	} while (pgd++, addr = next, addr != end);
 	return 0;
@@ -344,7 +347,7 @@ static inline int vma_migratable(struct 
  */
 static struct vm_area_struct *
 check_range(struct mm_struct *mm, unsigned long start, unsigned long end,
-	    const nodemask_t *nodes, unsigned long flags, struct list_head *pagelist)
+	    const nodemask_t *nodes, unsigned long flags, void *private)
 {
 	int err;
 	struct vm_area_struct *first, *vma, *prev;
@@ -373,7 +376,7 @@ check_range(struct mm_struct *mm, unsign
 			if (vma->vm_start > start)
 				start = vma->vm_start;
 			err = check_pgd_range(vma, start, endvma, nodes,
-						flags, pagelist);
+						flags, private);
 			if (err) {
 				first = ERR_PTR(err);
 				break;
@@ -461,7 +464,8 @@ long do_mbind(unsigned long start, unsig
 	int err;
 	LIST_HEAD(pagelist);
 
-	if ((flags & ~(unsigned long)(MPOL_MF_STRICT|MPOL_MF_MOVE|MPOL_MF_MOVE_ALL))
+	if ((flags & ~(unsigned long)(MPOL_MF_STRICT |
+				      MPOL_MF_MOVE | MPOL_MF_MOVE_ALL))
 	    || mode > MPOL_MAX)
 		return -EINVAL;
 	if ((flags & MPOL_MF_MOVE_ALL) && !capable(CAP_SYS_RESOURCE))
@@ -499,8 +503,9 @@ long do_mbind(unsigned long start, unsig
 			mode,nodes_addr(nodes)[0]);
 
 	down_write(&mm->mmap_sem);
-	vma = check_range(mm, start, end, nmask, flags,
-	      (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) ? &pagelist : NULL);
+	vma = check_range(mm, start, end, nmask,
+			  flags | MPOL_MF_INVERT, &pagelist);
+
 	err = PTR_ERR(vma);
 	if (!IS_ERR(vma)) {
 		int nr_failed = 0;
@@ -655,7 +660,6 @@ int do_migrate_pages(struct mm_struct *m
 	nodemask_t nodes;
 
 	nodes_andnot(nodes, *from_nodes, *to_nodes);
-	nodes_complement(nodes, nodes);
 
 	down_read(&mm->mmap_sem);
 	check_range(mm, mm->mmap->vm_start, TASK_SIZE, &nodes,

[PATCH 2/4] Fold numa_maps into memopolicies.c

[Christoph Lameter]

Fold numa_maps functionality into mempolicy.c

First discussed at http://marc.theaimsgroup.com/?t=113149255100001&r=1&w=2

- Use the check_range() in mempolicy.c to gather statistics.

- Improve the numa_maps code in general and fix some comments.

Signed-off-by: Christoph Lameter <clameter@sgi.com>

Index: linux-2.6.15-rc1-mm1/mm/mempolicy.c
===================================================================
--- linux-2.6.15-rc1-mm1.orig/mm/mempolicy.c	2005-11-17 19:19:24.000000000 -0800
+++ linux-2.6.15-rc1-mm1/mm/mempolicy.c	2005-11-17 19:20:40.000000000 -0800
@@ -84,6 +84,8 @@
 #include <linux/compat.h>
 #include <linux/mempolicy.h>
 #include <linux/swap.h>
+#include <linux/seq_file.h>
+#include <linux/proc_fs.h>
 
 #include <asm/tlbflush.h>
 #include <asm/uaccess.h>
@@ -91,6 +93,7 @@
 /* Internal flags */
 #define MPOL_MF_DISCONTIG_OK (MPOL_MF_INTERNAL << 0)	/* Skip checks for continuous vmas */
 #define MPOL_MF_INVERT (MPOL_MF_INTERNAL << 1)		/* Invert check for nodemask */
+#define MPOL_MF_STATS (MPOL_MF_INTERNAL << 2)		/* Gather statistics */
 
 static kmem_cache_t *policy_cache;
 static kmem_cache_t *sn_cache;
@@ -233,6 +236,8 @@ static void migrate_page_add(struct vm_a
 	}
 }
 
+static void gather_stats(struct page *, void *);
+
 /* Scan through pages checking if pages follow certain conditions. */
 static int check_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
 		unsigned long addr, unsigned long end,
@@ -261,7 +266,9 @@ static int check_pte_range(struct vm_are
 		if (node_isset(nid, *nodes) == !!(flags & MPOL_MF_INVERT))
 			continue;
 
-		if (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL))
+		if (flags & MPOL_MF_STATS)
+			gather_stats(page, private);
+		else if (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL))
 			migrate_page_add(vma, page, private, flags);
 		else
 			break;
@@ -1498,3 +1511,132 @@ void numa_policy_rebind(const nodemask_t
 {
 	rebind_policy(current->mempolicy, old, new);
 }
+
+/*
+ * Display pages allocated per node and memory policy via /proc.
+ */
+
+static const char *policy_types[] = { "default", "prefer", "bind",
+				      "interleave" };
+
+/*
+ * Convert a mempolicy into a string.
+ * Returns the number of characters in buffer (if positive)
+ * or an error (negative)
+ */
+static inline int mpol_to_str(char *buffer, int maxlen, struct mempolicy *pol)
+{
+	char *p = buffer;
+	int l;
+	nodemask_t nodes;
+	int mode = pol ? pol->policy : MPOL_DEFAULT;
+
+	switch (mode) {
+	case MPOL_DEFAULT:
+		nodes_clear(nodes);
+		break;
+
+	case MPOL_PREFERRED:
+		nodes_clear(nodes);
+		node_set(pol->v.preferred_node, nodes);
+		break;
+
+	case MPOL_BIND:
+		get_zonemask(pol, &nodes);
+		break;
+
+	case MPOL_INTERLEAVE:
+		nodes = pol->v.nodes;
+		break;
+
+	default:
+		BUG();
+		return -EFAULT;
+	}
+
+	l = strlen(policy_types[mode]);
+ 	if (buffer + maxlen < p + l + 1)
+ 		return -ENOSPC;
+
+	strcpy(p, policy_types[mode]);
+	p += l;
+
+	if (!nodes_empty(nodes)) {
+		if (buffer + maxlen < p + 2)
+			return -ENOSPC;
+		*p++ = '=';
+	 	p += nodelist_scnprintf(p, buffer + maxlen - p, nodes);
+	}
+	return p - buffer;
+}
+
+struct numa_maps {
+	unsigned long pages;
+	unsigned long anon;
+	unsigned long mapped;
+	unsigned long mapcount_max;
+	unsigned long node[MAX_NUMNODES];
+};
+
+static void gather_stats(struct page *page, void *private)
+{
+	struct numa_maps *md = private;
+	int count = page_mapcount(page);
+
+	if (count)
+		md->mapped++;
+
+	if (count > md->mapcount_max)
+		md->mapcount_max = count;
+
+	md->pages++;
+
+	if (PageAnon(page))
+		md->anon++;
+
+	md->node[page_to_nid(page)]++;
+	cond_resched();
+}
+
+int show_numa_map(struct seq_file *m, void *v)
+{
+	struct task_struct *task = m->private;
+	struct vm_area_struct *vma = v;
+	struct numa_maps *md;
+	int n;
+	char buffer[50];
+
+	if (!vma->vm_mm)
+		return 0;
+
+	md = kzalloc(sizeof(struct numa_maps), GFP_KERNEL);
+	if (!md)
+		return 0;
+
+	check_pgd_range(vma, vma->vm_start, vma->vm_end,
+		    &node_online_map, MPOL_MF_STATS, md);
+
+	if (md->pages) {
+		mpol_to_str(buffer, sizeof(buffer),
+			    get_vma_policy(task, vma, vma->vm_start));
+
+		seq_printf(m, "%08lx %s pages=%lu mapped=%lu maxref=%lu",
+			   vma->vm_start, buffer, md->pages,
+			   md->mapped, md->mapcount_max);
+
+		if (md->anon)
+			seq_printf(m," anon=%lu",md->anon);
+
+		for_each_online_node(n)
+			if (md->node[n])
+				seq_printf(m, " N%d=%lu", n, md->node[n]);
+
+		seq_putc(m, '\n');
+	}
+	kfree(md);
+
+	if (m->count < m->size)
+		m->version = (vma != get_gate_vma(task)) ? vma->vm_start : 0;
+	return 0;
+}
+
Index: linux-2.6.15-rc1-mm1/fs/proc/task_mmu.c
===================================================================
--- linux-2.6.15-rc1-mm1.orig/fs/proc/task_mmu.c	2005-11-17 15:17:31.000000000 -0800
+++ linux-2.6.15-rc1-mm1/fs/proc/task_mmu.c	2005-11-17 19:19:57.000000000 -0800
@@ -390,130 +390,12 @@ struct seq_operations proc_pid_smaps_op 
 };
 
 #ifdef CONFIG_NUMA
-
-struct numa_maps {
-	unsigned long pages;
-	unsigned long anon;
-	unsigned long mapped;
-	unsigned long mapcount_max;
-	unsigned long node[MAX_NUMNODES];
-};
-
-/*
- * Calculate numa node maps for a vma
- */
-static struct numa_maps *get_numa_maps(const struct vm_area_struct *vma)
-{
-	struct page *page;
-	unsigned long vaddr;
-	struct mm_struct *mm = vma->vm_mm;
-	int i;
-	struct numa_maps *md = kmalloc(sizeof(struct numa_maps), GFP_KERNEL);
-
-	if (!md)
-		return NULL;
-	md->pages = 0;
-	md->anon = 0;
-	md->mapped = 0;
-	md->mapcount_max = 0;
-	for_each_node(i)
-		md->node[i] =0;
-
- 	for (vaddr = vma->vm_start; vaddr < vma->vm_end; vaddr += PAGE_SIZE) {
-		page = follow_page(mm, vaddr, 0);
-		if (page) {
-			int count = page_mapcount(page);
-
-			if (count)
-				md->mapped++;
-			if (count > md->mapcount_max)
-				md->mapcount_max = count;
-			md->pages++;
-			if (PageAnon(page))
-				md->anon++;
-			md->node[page_to_nid(page)]++;
-		}
-		cond_resched();
-	}
-	return md;
-}
-
-static int show_numa_map(struct seq_file *m, void *v)
-{
-	struct task_struct *task = m->private;
-	struct vm_area_struct *vma = v;
-	struct mempolicy *pol;
-	struct numa_maps *md;
-	struct zone **z;
-	int n;
-	int first;
-
-	if (!vma->vm_mm)
-		return 0;
-
-	md = get_numa_maps(vma);
-	if (!md)
-		return 0;
-
-	seq_printf(m, "%08lx", vma->vm_start);
-	pol = get_vma_policy(task, vma, vma->vm_start);
-	/* Print policy */
-	switch (pol->policy) {
-	case MPOL_PREFERRED:
-		seq_printf(m, " prefer=%d", pol->v.preferred_node);
-		break;
-	case MPOL_BIND:
-		seq_printf(m, " bind={");
-		first = 1;
-		for (z = pol->v.zonelist->zones; *z; z++) {
-
-			if (!first)
-				seq_putc(m, ',');
-			else
-				first = 0;
-			seq_printf(m, "%d/%s", (*z)->zone_pgdat->node_id,
-					(*z)->name);
-		}
-		seq_putc(m, '}');
-		break;
-	case MPOL_INTERLEAVE:
-		seq_printf(m, " interleave={");
-		first = 1;
-		for_each_node(n) {
-			if (node_isset(n, pol->v.nodes)) {
-				if (!first)
-					seq_putc(m,',');
-				else
-					first = 0;
-				seq_printf(m, "%d",n);
-			}
-		}
-		seq_putc(m, '}');
-		break;
-	default:
-		seq_printf(m," default");
-		break;
-	}
-	seq_printf(m, " MaxRef=%lu Pages=%lu Mapped=%lu",
-			md->mapcount_max, md->pages, md->mapped);
-	if (md->anon)
-		seq_printf(m," Anon=%lu",md->anon);
-
-	for_each_online_node(n) {
-		if (md->node[n])
-			seq_printf(m, " N%d=%lu", n, md->node[n]);
-	}
-	seq_putc(m, '\n');
-	kfree(md);
-	if (m->count < m->size)  /* vma is copied successfully */
-		m->version = (vma != get_gate_vma(task)) ? vma->vm_start : 0;
-	return 0;
-}
+extern int show_numa_map(struct seq_file *m, void *v);
 
 struct seq_operations proc_pid_numa_maps_op = {
-	.start	= m_start,
-	.next	= m_next,
-	.stop	= m_stop,
-	.show	= show_numa_map
+        .start  = m_start,
+        .next   = m_next,
+        .stop   = m_stop,
+        .show   = show_numa_map
 };
 #endif

[PATCH 3/4] mempolicies: unexport get_vma_policy()

[Christoph Lameter]

Since the numa_maps functionality is now in mempolicy.c we no longer
need to export get_vma_policy().

Signed-off-by: Christoph Lameter <clameter@sgi.com>

Index: linux-2.6.15-rc1-mm2/mm/mempolicy.c
===================================================================
--- linux-2.6.15-rc1-mm2.orig/mm/mempolicy.c	2005-11-21 12:39:54.000000000 -0800
+++ linux-2.6.15-rc1-mm2/mm/mempolicy.c	2005-11-21 12:40:38.000000000 -0800
@@ -943,8 +943,8 @@ asmlinkage long compat_sys_mbind(compat_
 #endif
 
 /* Return effective policy for a VMA */
-struct mempolicy *
-get_vma_policy(struct task_struct *task, struct vm_area_struct *vma, unsigned long addr)
+static struct mempolicy * get_vma_policy(struct task_struct *task,
+		struct vm_area_struct *vma, unsigned long addr)
 {
 	struct mempolicy *pol = task->mempolicy;
 
Index: linux-2.6.15-rc1-mm2/include/linux/mempolicy.h
===================================================================
--- linux-2.6.15-rc1-mm2.orig/include/linux/mempolicy.h	2005-11-21 12:11:07.000000000 -0800
+++ linux-2.6.15-rc1-mm2/include/linux/mempolicy.h	2005-11-21 12:40:06.000000000 -0800
@@ -144,9 +144,6 @@ void mpol_free_shared_policy(struct shar
 struct mempolicy *mpol_shared_policy_lookup(struct shared_policy *sp,
 					    unsigned long idx);
 
-struct mempolicy *get_vma_policy(struct task_struct *task,
-			struct vm_area_struct *vma, unsigned long addr);
-
 extern void numa_default_policy(void);
 extern void numa_policy_init(void);
 extern void numa_policy_rebind(const nodemask_t *old, const nodemask_t *new);

[PATCH 4/4] Move page migration related functions near do_migrate_pages()

[Christoph Lameter]

Group page migration functions in mempolicy.c

Add a forward declaration for migrate_page_add (like gather_stats()) and use
our new found mobility to group all page migration related function around
do_migrate_pages().

Signed-off-by: Christoph Lameter <clameter@sgi.com>

Index: linux-2.6.15-rc1-mm2/mm/mempolicy.c
===================================================================
--- linux-2.6.15-rc1-mm2.orig/mm/mempolicy.c	2005-11-21 12:14:43.000000000 -0800
+++ linux-2.6.15-rc1-mm2/mm/mempolicy.c	2005-11-21 12:21:00.000000000 -0800
@@ -188,55 +188,9 @@ static struct mempolicy *mpol_new(int mo
 	return policy;
 }
 
-/* Check if we are the only process mapping the page in question */
-static inline int single_mm_mapping(struct mm_struct *mm,
-			struct address_space *mapping)
-{
-	struct vm_area_struct *vma;
-	struct prio_tree_iter iter;
-	int rc = 1;
-
-	spin_lock(&mapping->i_mmap_lock);
-	vma_prio_tree_foreach(vma, &iter, &mapping->i_mmap, 0, ULONG_MAX)
-		if (mm != vma->vm_mm) {
-			rc = 0;
-			goto out;
-		}
-	list_for_each_entry(vma, &mapping->i_mmap_nonlinear, shared.vm_set.list)
-		if (mm != vma->vm_mm) {
-			rc = 0;
-			goto out;
-		}
-out:
-	spin_unlock(&mapping->i_mmap_lock);
-	return rc;
-}
-
-/*
- * Add a page to be migrated to the pagelist
- */
-static void migrate_page_add(struct vm_area_struct *vma,
-	struct page *page, struct list_head *pagelist, unsigned long flags)
-{
-	/*
-	 * Avoid migrating a page that is shared by others and not writable.
-	 */
-	if ((flags & MPOL_MF_MOVE_ALL) || !page->mapping || PageAnon(page) ||
-	    mapping_writably_mapped(page->mapping) ||
-	    single_mm_mapping(vma->vm_mm, page->mapping)) {
-		int rc = isolate_lru_page(page);
-
-		if (rc == 1)
-			list_add(&page->lru, pagelist);
-		/*
-		 * If the isolate attempt was not successful then we just
-		 * encountered an unswappable page. Something must be wrong.
-	 	 */
-		WARN_ON(rc == 0);
-	}
-}
-
 static void gather_stats(struct page *, void *);
+static void migrate_page_add(struct vm_area_struct *vma,
+	struct page *page, struct list_head *pagelist, unsigned long flags);
 
 /* Scan through pages checking if pages follow certain conditions. */
 static int check_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
@@ -448,90 +402,6 @@ static int contextualize_policy(int mode
 	return mpol_check_policy(mode, nodes);
 }
 
-static int swap_pages(struct list_head *pagelist)
-{
-	LIST_HEAD(moved);
-	LIST_HEAD(failed);
-	int n;
-
-	n = migrate_pages(pagelist, NULL, &moved, &failed);
-	putback_lru_pages(&failed);
-	putback_lru_pages(&moved);
-
-	return n;
-}
-
-long do_mbind(unsigned long start, unsigned long len,
-		unsigned long mode, nodemask_t *nmask, unsigned long flags)
-{
-	struct vm_area_struct *vma;
-	struct mm_struct *mm = current->mm;
-	struct mempolicy *new;
-	unsigned long end;
-	int err;
-	LIST_HEAD(pagelist);
-
-	if ((flags & ~(unsigned long)(MPOL_MF_STRICT |
-				      MPOL_MF_MOVE | MPOL_MF_MOVE_ALL))
-	    || mode > MPOL_MAX)
-		return -EINVAL;
-	if ((flags & MPOL_MF_MOVE_ALL) && !capable(CAP_SYS_RESOURCE))
-		return -EPERM;
-
-	if (start & ~PAGE_MASK)
-		return -EINVAL;
-
-	if (mode == MPOL_DEFAULT)
-		flags &= ~MPOL_MF_STRICT;
-
-	len = (len + PAGE_SIZE - 1) & PAGE_MASK;
-	end = start + len;
-
-	if (end < start)
-		return -EINVAL;
-	if (end == start)
-		return 0;
-
-	if (mpol_check_policy(mode, nmask))
-		return -EINVAL;
-
-	new = mpol_new(mode, nmask);
-	if (IS_ERR(new))
-		return PTR_ERR(new);
-
-	/*
-	 * If we are using the default policy then operation
-	 * on discontinuous address spaces is okay after all
-	 */
-	if (!new)
-		flags |= MPOL_MF_DISCONTIG_OK;
-
-	PDprintk("mbind %lx-%lx mode:%ld nodes:%lx\n",start,start+len,
-			mode,nodes_addr(nodes)[0]);
-
-	down_write(&mm->mmap_sem);
-	vma = check_range(mm, start, end, nmask,
-			  flags | MPOL_MF_INVERT, &pagelist);
-
-	err = PTR_ERR(vma);
-	if (!IS_ERR(vma)) {
-		int nr_failed = 0;
-
-		err = mbind_range(vma, start, end, new);
-		if (!list_empty(&pagelist))
-			nr_failed = swap_pages(&pagelist);
-
-		if (!err && nr_failed && (flags & MPOL_MF_STRICT))
-			err = -EIO;
-	}
-	if (!list_empty(&pagelist))
-		putback_lru_pages(&pagelist);
-
-	up_write(&mm->mmap_sem);
-	mpol_free(new);
-	return err;
-}
-
 /* Set the process memory policy */
 long do_set_mempolicy(int mode, nodemask_t *nodes)
 {
@@ -652,6 +522,71 @@ long do_get_mempolicy(int *policy, nodem
 }
 
 /*
+ * page migration
+ */
+
+/* Check if we are the only process mapping the page in question */
+static inline int single_mm_mapping(struct mm_struct *mm,
+			struct address_space *mapping)
+{
+	struct vm_area_struct *vma;
+	struct prio_tree_iter iter;
+	int rc = 1;
+
+	spin_lock(&mapping->i_mmap_lock);
+	vma_prio_tree_foreach(vma, &iter, &mapping->i_mmap, 0, ULONG_MAX)
+		if (mm != vma->vm_mm) {
+			rc = 0;
+			goto out;
+		}
+	list_for_each_entry(vma, &mapping->i_mmap_nonlinear, shared.vm_set.list)
+		if (mm != vma->vm_mm) {
+			rc = 0;
+			goto out;
+		}
+out:
+	spin_unlock(&mapping->i_mmap_lock);
+	return rc;
+}
+
+/*
+ * Add a page to be migrated to the pagelist
+ */
+static void migrate_page_add(struct vm_area_struct *vma,
+	struct page *page, struct list_head *pagelist, unsigned long flags)
+{
+	/*
+	 * Avoid migrating a page that is shared by others and not writable.
+	 */
+	if ((flags & MPOL_MF_MOVE_ALL) || !page->mapping || PageAnon(page) ||
+	    mapping_writably_mapped(page->mapping) ||
+	    single_mm_mapping(vma->vm_mm, page->mapping)) {
+		int rc = isolate_lru_page(page);
+
+		if (rc == 1)
+			list_add(&page->lru, pagelist);
+		/*
+		 * If the isolate attempt was not successful then we just
+		 * encountered an unswappable page. Something must be wrong.
+	 	 */
+		WARN_ON(rc == 0);
+	}
+}
+
+static int swap_pages(struct list_head *pagelist)
+{
+	LIST_HEAD(moved);
+	LIST_HEAD(failed);
+	int n;
+
+	n = migrate_pages(pagelist, NULL, &moved, &failed);
+	putback_lru_pages(&failed);
+	putback_lru_pages(&moved);
+
+	return n;
+}
+
+/*
  * For now migrate_pages simply swaps out the pages from nodes that are in
  * the source set but not in the target set. In the future, we would
  * want a function that moves pages between the two nodesets in such
@@ -681,6 +616,77 @@ int do_migrate_pages(struct mm_struct *m
 	return count;
 }
 
+long do_mbind(unsigned long start, unsigned long len,
+		unsigned long mode, nodemask_t *nmask, unsigned long flags)
+{
+	struct vm_area_struct *vma;
+	struct mm_struct *mm = current->mm;
+	struct mempolicy *new;
+	unsigned long end;
+	int err;
+	LIST_HEAD(pagelist);
+
+	if ((flags & ~(unsigned long)(MPOL_MF_STRICT |
+				      MPOL_MF_MOVE | MPOL_MF_MOVE_ALL))
+	    || mode > MPOL_MAX)
+		return -EINVAL;
+	if ((flags & MPOL_MF_MOVE_ALL) && !capable(CAP_SYS_RESOURCE))
+		return -EPERM;
+
+	if (start & ~PAGE_MASK)
+		return -EINVAL;
+
+	if (mode == MPOL_DEFAULT)
+		flags &= ~MPOL_MF_STRICT;
+
+	len = (len + PAGE_SIZE - 1) & PAGE_MASK;
+	end = start + len;
+
+	if (end < start)
+		return -EINVAL;
+	if (end == start)
+		return 0;
+
+	if (mpol_check_policy(mode, nmask))
+		return -EINVAL;
+
+	new = mpol_new(mode, nmask);
+	if (IS_ERR(new))
+		return PTR_ERR(new);
+
+	/*
+	 * If we are using the default policy then operation
+	 * on discontinuous address spaces is okay after all
+	 */
+	if (!new)
+		flags |= MPOL_MF_DISCONTIG_OK;
+
+	PDprintk("mbind %lx-%lx mode:%ld nodes:%lx\n",start,start+len,
+			mode,nodes_addr(nodes)[0]);
+
+	down_write(&mm->mmap_sem);
+	vma = check_range(mm, start, end, nmask,
+			  flags | MPOL_MF_INVERT, &pagelist);
+
+	err = PTR_ERR(vma);
+	if (!IS_ERR(vma)) {
+		int nr_failed = 0;
+
+		err = mbind_range(vma, start, end, new);
+		if (!list_empty(&pagelist))
+			nr_failed = swap_pages(&pagelist);
+
+		if (!err && nr_failed && (flags & MPOL_MF_STRICT))
+			err = -EIO;
+	}
+	if (!list_empty(&pagelist))
+		putback_lru_pages(&pagelist);
+
+	up_write(&mm->mmap_sem);
+	mpol_free(new);
+	return err;
+}
+
 /*
  * User space interface with variable sized bitmaps for nodelists.
  */