Merge rsync://rsync.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6

16 files changed, 277 insertions, 111 deletions

diff --git a/mm/bootmem.c b/mm/bootmem.c
index d213feded10..50353e0dac1 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -29,9 +29,7 @@ unsigned long max_low_pfn;
 unsigned long min_low_pfn;
 unsigned long max_pfn;
 
-EXPORT_SYMBOL(max_pfn);		/* This is exported so
-				 * dma_get_required_mask(), which uses
-				 * it, can be an inline function */
+EXPORT_UNUSED_SYMBOL(max_pfn);  /*  June 2006  */
 
 static LIST_HEAD(bdata_list);
 #ifdef CONFIG_CRASH_DUMP
diff --git a/mm/fadvise.c b/mm/fadvise.c
index 0a03357a1f8..168c78a121b 100644
--- a/mm/fadvise.c
+++ b/mm/fadvise.c
@@ -23,18 +23,6 @@
 /*
  * POSIX_FADV_WILLNEED could set PG_Referenced, and POSIX_FADV_NOREUSE could
  * deactivate the pages and clear PG_Referenced.
- *
- * LINUX_FADV_ASYNC_WRITE: start async writeout of any dirty pages between file
- * offsets `offset' and `offset+len' inclusive.  Any pages which are currently
- * under writeout are skipped, whether or not they are dirty.
- *
- * LINUX_FADV_WRITE_WAIT: wait upon writeout of any dirty pages between file
- * offsets `offset' and `offset+len'.
- *
- * By combining these two operations the application may do several things:
- *
- * LINUX_FADV_ASYNC_WRITE: push some or all of the dirty pages at the disk.
- *
  */
 asmlinkage long sys_fadvise64_64(int fd, loff_t offset, loff_t len, int advice)
 {
@@ -85,7 +73,6 @@ asmlinkage long sys_fadvise64_64(int fd, loff_t offset, loff_t len, int advice)
 		file->f_ra.ra_pages = bdi->ra_pages * 2;
 		break;
 	case POSIX_FADV_WILLNEED:
-	case POSIX_FADV_NOREUSE:
 		if (!mapping->a_ops->readpage) {
 			ret = -EINVAL;
 			break;
@@ -106,6 +93,8 @@ asmlinkage long sys_fadvise64_64(int fd, loff_t offset, loff_t len, int advice)
 		if (ret > 0)
 			ret = 0;
 		break;
+	case POSIX_FADV_NOREUSE:
+		break;
 	case POSIX_FADV_DONTNEED:
 		if (!bdi_write_congested(mapping->backing_dev_info))
 			filemap_flush(mapping);
diff --git a/mm/filemap.c b/mm/filemap.c
index d087fc3d328..b9a60c43b61 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -849,8 +849,6 @@ static void shrink_readahead_size_eio(struct file *filp,
 		return;
 
 	ra->ra_pages /= 4;
-	printk(KERN_WARNING "Reducing readahead size to %luK\n",
-			ra->ra_pages << (PAGE_CACHE_SHIFT - 10));
 }
 
 /**
diff --git a/mm/memory.c b/mm/memory.c
index c1e14c9e67e..109e9866237 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -47,6 +47,7 @@
 #include <linux/pagemap.h>
 #include <linux/rmap.h>
 #include <linux/module.h>
+#include <linux/delayacct.h>
 #include <linux/init.h>
 
 #include <asm/pgalloc.h>
@@ -1549,9 +1550,9 @@ gotten:
 		flush_cache_page(vma, address, pte_pfn(orig_pte));
 		entry = mk_pte(new_page, vma->vm_page_prot);
 		entry = maybe_mkwrite(pte_mkdirty(entry), vma);
+		lazy_mmu_prot_update(entry);
 		ptep_establish(vma, address, page_table, entry);
 		update_mmu_cache(vma, address, entry);
-		lazy_mmu_prot_update(entry);
 		lru_cache_add_active(new_page);
 		page_add_new_anon_rmap(new_page, vma, address);
 
@@ -1853,7 +1854,7 @@ int vmtruncate_range(struct inode *inode, loff_t offset, loff_t end)
 
 	return 0;
 }
-EXPORT_SYMBOL(vmtruncate_range);
+EXPORT_UNUSED_SYMBOL(vmtruncate_range);  /*  June 2006  */
 
 /* 
  * Primitive swap readahead code. We simply read an aligned block of
@@ -1934,6 +1935,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		migration_entry_wait(mm, pmd, address);
 		goto out;
 	}
+	delayacct_set_flag(DELAYACCT_PF_SWAPIN);
 	page = lookup_swap_cache(entry);
 	if (!page) {
  		swapin_readahead(entry, address, vma);
@@ -1946,6 +1948,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 			page_table = pte_offset_map_lock(mm, pmd, address, &ptl);
 			if (likely(pte_same(*page_table, orig_pte)))
 				ret = VM_FAULT_OOM;
+			delayacct_clear_flag(DELAYACCT_PF_SWAPIN);
 			goto unlock;
 		}
 
@@ -1955,6 +1958,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		grab_swap_token();
 	}
 
+	delayacct_clear_flag(DELAYACCT_PF_SWAPIN);
 	mark_page_accessed(page);
 	lock_page(page);
 
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 01c9fb97c61..c37319542b7 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -52,6 +52,9 @@ static int __add_section(struct zone *zone, unsigned long phys_start_pfn)
 	int nr_pages = PAGES_PER_SECTION;
 	int ret;
 
+	if (pfn_valid(phys_start_pfn))
+		return -EEXIST;
+
 	ret = sparse_add_one_section(zone, phys_start_pfn, nr_pages);
 
 	if (ret < 0)
@@ -76,15 +79,22 @@ int __add_pages(struct zone *zone, unsigned long phys_start_pfn,
 {
 	unsigned long i;
 	int err = 0;
+	int start_sec, end_sec;
+	/* during initialize mem_map, align hot-added range to section */
+	start_sec = pfn_to_section_nr(phys_start_pfn);
+	end_sec = pfn_to_section_nr(phys_start_pfn + nr_pages - 1);
 
-	for (i = 0; i < nr_pages; i += PAGES_PER_SECTION) {
-		err = __add_section(zone, phys_start_pfn + i);
+	for (i = start_sec; i <= end_sec; i++) {
+		err = __add_section(zone, i << PFN_SECTION_SHIFT);
 
-		/* We want to keep adding the rest of the
-		 * sections if the first ones already exist
+		/*
+		 * EEXIST is finally dealed with by ioresource collision
+		 * check. see add_memory() => register_memory_resource()
+		 * Warning will be printed if there is collision.
 		 */
 		if (err && (err != -EEXIST))
 			break;
+		err = 0;
 	}
 
 	return err;
@@ -156,7 +166,7 @@ int online_pages(unsigned long pfn, unsigned long nr_pages)
 	res.flags = IORESOURCE_MEM; /* we just need system ram */
 	section_end = res.end;
 
-	while (find_next_system_ram(&res) >= 0) {
+	while ((res.start < res.end) && (find_next_system_ram(&res) >= 0)) {
 		start_pfn = (unsigned long)(res.start >> PAGE_SHIFT);
 		nr_pages = (unsigned long)
                            ((res.end + 1 - res.start) >> PAGE_SHIFT);
@@ -213,10 +223,9 @@ static void rollback_node_hotadd(int nid, pg_data_t *pgdat)
 }
 
 /* add this memory to iomem resource */
-static void register_memory_resource(u64 start, u64 size)
+static struct resource *register_memory_resource(u64 start, u64 size)
 {
 	struct resource *res;
-
 	res = kzalloc(sizeof(struct resource), GFP_KERNEL);
 	BUG_ON(!res);
 
@@ -228,7 +237,18 @@ static void register_memory_resource(u64 start, u64 size)
 		printk("System RAM resource %llx - %llx cannot be added\n",
 		(unsigned long long)res->start, (unsigned long long)res->end);
 		kfree(res);
+		res = NULL;
 	}
+	return res;
+}
+
+static void release_memory_resource(struct resource *res)
+{
+	if (!res)
+		return;
+	release_resource(res);
+	kfree(res);
+	return;
 }
 
 
@@ -237,8 +257,13 @@ int add_memory(int nid, u64 start, u64 size)
 {
 	pg_data_t *pgdat = NULL;
 	int new_pgdat = 0;
+	struct resource *res;
 	int ret;
 
+	res = register_memory_resource(start, size);
+	if (!res)
+		return -EEXIST;
+
 	if (!node_online(nid)) {
 		pgdat = hotadd_new_pgdat(nid, start);
 		if (!pgdat)
@@ -268,14 +293,13 @@ int add_memory(int nid, u64 start, u64 size)
 		BUG_ON(ret);
 	}
 
-	/* register this memory as resource */
-	register_memory_resource(start, size);
-
 	return ret;
 error:
 	/* rollback pgdat allocation and others */
 	if (new_pgdat)
 		rollback_node_hotadd(nid, pgdat);
+	if (res)
+		release_memory_resource(res);
 
 	return ret;
 }
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index e07e27e846a..a9963ceddd6 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -1176,7 +1176,15 @@ static inline unsigned interleave_nid(struct mempolicy *pol,
 	if (vma) {
 		unsigned long off;
 
-		off = vma->vm_pgoff;
+		/*
+		 * for small pages, there is no difference between
+		 * shift and PAGE_SHIFT, so the bit-shift is safe.
+		 * for huge pages, since vm_pgoff is in units of small
+		 * pages, we need to shift off the always 0 bits to get
+		 * a useful offset.
+		 */
+		BUG_ON(shift < PAGE_SHIFT);
+		off = vma->vm_pgoff >> (shift - PAGE_SHIFT);
 		off += (addr - vma->vm_start) >> shift;
 		return offset_il_node(pol, vma, off);
 	} else
diff --git a/mm/mempool.c b/mm/mempool.c
index fe6e05289cc..ccd8cb8cd41 100644
--- a/mm/mempool.c
+++ b/mm/mempool.c
@@ -238,8 +238,13 @@ repeat_alloc:
 	init_wait(&wait);
 	prepare_to_wait(&pool->wait, &wait, TASK_UNINTERRUPTIBLE);
 	smp_mb();
-	if (!pool->curr_nr)
-		io_schedule();
+	if (!pool->curr_nr) {
+		/*
+		 * FIXME: this should be io_schedule().  The timeout is there
+		 * as a workaround for some DM problems in 2.6.18.
+		 */
+		io_schedule_timeout(5*HZ);
+	}
 	finish_wait(&pool->wait, &wait);
 
 	goto repeat_alloc;
diff --git a/mm/mmap.c b/mm/mmap.c
index c1868ecdbc5..e66a0b524af 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -30,6 +30,10 @@
 #include <asm/cacheflush.h>
 #include <asm/tlb.h>
 
+#ifndef arch_mmap_check
+#define arch_mmap_check(addr, len, flags)	(0)
+#endif
+
 static void unmap_region(struct mm_struct *mm,
 		struct vm_area_struct *vma, struct vm_area_struct *prev,
 		unsigned long start, unsigned long end);
@@ -913,6 +917,10 @@ unsigned long do_mmap_pgoff(struct file * file, unsigned long addr,
 	if (!len)
 		return -EINVAL;
 
+	error = arch_mmap_check(addr, len, flags);
+	if (error)
+		return error;
+
 	/* Careful about overflows.. */
 	len = PAGE_ALIGN(len);
 	if (!len || len > TASK_SIZE)
@@ -1859,6 +1867,7 @@ unsigned long do_brk(unsigned long addr, unsigned long len)
 	unsigned long flags;
 	struct rb_node ** rb_link, * rb_parent;
 	pgoff_t pgoff = addr >> PAGE_SHIFT;
+	int error;
 
 	len = PAGE_ALIGN(len);
 	if (!len)
@@ -1867,6 +1876,12 @@ unsigned long do_brk(unsigned long addr, unsigned long len)
 	if ((addr + len) > TASK_SIZE || (addr + len) < addr)
 		return -EINVAL;
 
+	flags = VM_DATA_DEFAULT_FLAGS | VM_ACCOUNT | mm->def_flags;
+
+	error = arch_mmap_check(addr, len, flags);
+	if (error)
+		return error;
+
 	/*
 	 * mlock MCL_FUTURE?
 	 */
@@ -1907,8 +1922,6 @@ unsigned long do_brk(unsigned long addr, unsigned long len)
 	if (security_vm_enough_memory(len >> PAGE_SHIFT))
 		return -ENOMEM;
 
-	flags = VM_DATA_DEFAULT_FLAGS | VM_ACCOUNT | mm->def_flags;
-
 	/* Can we just expand an old private anonymous mapping? */
 	if (vma_merge(mm, prev, addr, addr + len, flags,
 					NULL, NULL, pgoff, NULL))
diff --git a/mm/mmzone.c b/mm/mmzone.c
index 0959ee1a479..febea1c9816 100644
--- a/mm/mmzone.c
+++ b/mm/mmzone.c
@@ -14,7 +14,7 @@ struct pglist_data *first_online_pgdat(void)
 	return NODE_DATA(first_online_node);
 }
 
-EXPORT_SYMBOL(first_online_pgdat);
+EXPORT_UNUSED_SYMBOL(first_online_pgdat);  /*  June 2006  */
 
 struct pglist_data *next_online_pgdat(struct pglist_data *pgdat)
 {
@@ -24,7 +24,7 @@ struct pglist_data *next_online_pgdat(struct pglist_data *pgdat)
 		return NULL;
 	return NODE_DATA(nid);
 }
-EXPORT_SYMBOL(next_online_pgdat);
+EXPORT_UNUSED_SYMBOL(next_online_pgdat);  /*  June 2006  */
 
 
 /*
@@ -45,5 +45,5 @@ struct zone *next_zone(struct zone *zone)
 	}
 	return zone;
 }
-EXPORT_SYMBOL(next_zone);
+EXPORT_UNUSED_SYMBOL(next_zone);  /*  June 2006  */
 
diff --git a/mm/nommu.c b/mm/nommu.c
index 5151c44a825..c576df71e3b 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -1070,6 +1070,7 @@ int remap_pfn_range(struct vm_area_struct *vma, unsigned long from,
 	vma->vm_start = vma->vm_pgoff << PAGE_SHIFT;
 	return 0;
 }
+EXPORT_SYMBOL(remap_pfn_range);
 
 void swap_unplug_io_fn(struct backing_dev_info *bdi, struct page *page)
 {
@@ -1090,6 +1091,7 @@ void unmap_mapping_range(struct address_space *mapping,
 			 int even_cows)
 {
 }
+EXPORT_SYMBOL(unmap_mapping_range);
 
 /*
  * Check that a process has enough memory to allocate a new virtual
diff --git a/mm/slab.c b/mm/slab.c
index 85c2e03098a..21ba0603570 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -674,6 +674,37 @@ static struct kmem_cache cache_cache = {
 #endif
 };
 
+#ifdef CONFIG_LOCKDEP
+
+/*
+ * Slab sometimes uses the kmalloc slabs to store the slab headers
+ * for other slabs "off slab".
+ * The locking for this is tricky in that it nests within the locks
+ * of all other slabs in a few places; to deal with this special
+ * locking we put on-slab caches into a separate lock-class.
+ */
+static struct lock_class_key on_slab_key;
+
+static inline void init_lock_keys(struct cache_sizes *s)
+{
+	int q;
+
+	for (q = 0; q < MAX_NUMNODES; q++) {
+		if (!s->cs_cachep->nodelists[q] || OFF_SLAB(s->cs_cachep))
+			continue;
+		lockdep_set_class(&s->cs_cachep->nodelists[q]->list_lock,
+				  &on_slab_key);
+	}
+}
+
+#else
+static inline void init_lock_keys(struct cache_sizes *s)
+{
+}
+#endif
+
+
+
 /* Guard access to the cache-chain. */
 static DEFINE_MUTEX(cache_chain_mutex);
 static struct list_head cache_chain;
@@ -1021,8 +1052,7 @@ static void drain_alien_cache(struct kmem_cache *cachep,
 	}
 }
 
-static inline int cache_free_alien(struct kmem_cache *cachep, void *objp,
-				   int nesting)
+static inline int cache_free_alien(struct kmem_cache *cachep, void *objp)
 {
 	struct slab *slabp = virt_to_slab(objp);
 	int nodeid = slabp->nodeid;
@@ -1040,7 +1070,7 @@ static inline int cache_free_alien(struct kmem_cache *cachep, void *objp,
 	STATS_INC_NODEFREES(cachep);
 	if (l3->alien && l3->alien[nodeid]) {
 		alien = l3->alien[nodeid];
-		spin_lock_nested(&alien->lock, nesting);
+		spin_lock(&alien->lock);
 		if (unlikely(alien->avail == alien->limit)) {
 			STATS_INC_ACOVERFLOW(cachep);
 			__drain_alien_cache(cachep, alien, nodeid);
@@ -1069,15 +1099,14 @@ static inline void free_alien_cache(struct array_cache **ac_ptr)
 {
 }
 
-static inline int cache_free_alien(struct kmem_cache *cachep, void *objp,
-				   int nesting)
+static inline int cache_free_alien(struct kmem_cache *cachep, void *objp)
 {
 	return 0;
 }
 
 #endif
 
-static int __devinit cpuup_callback(struct notifier_block *nfb,
+static int __cpuinit cpuup_callback(struct notifier_block *nfb,
 				    unsigned long action, void *hcpu)
 {
 	long cpu = (long)hcpu;
@@ -1393,6 +1422,7 @@ void __init kmem_cache_init(void)
 					ARCH_KMALLOC_FLAGS|SLAB_PANIC,
 					NULL, NULL);
 		}
+		init_lock_keys(sizes);
 
 		sizes->cs_dmacachep = kmem_cache_create(names->name_dma,
 					sizes->cs_size,
@@ -1760,8 +1790,6 @@ static void slab_destroy_objs(struct kmem_cache *cachep, struct slab *slabp)
 }
 #endif
 
-static void __cache_free(struct kmem_cache *cachep, void *objp, int nesting);
-
 /**
  * slab_destroy - destroy and release all objects in a slab
  * @cachep: cache pointer being destroyed
@@ -1785,17 +1813,8 @@ static void slab_destroy(struct kmem_cache *cachep, struct slab *slabp)
 		call_rcu(&slab_rcu->head, kmem_rcu_free);
 	} else {
 		kmem_freepages(cachep, addr);
-		if (OFF_SLAB(cachep)) {
-			unsigned long flags;
-
-			/*
-		 	 * lockdep: we may nest inside an already held
-			 * ac->lock, so pass in a nesting flag:
-			 */
-			local_irq_save(flags);
-			__cache_free(cachep->slabp_cache, slabp, 1);
-			local_irq_restore(flags);
-		}
+		if (OFF_SLAB(cachep))
+			kmem_cache_free(cachep->slabp_cache, slabp);
 	}
 }
 
@@ -3100,16 +3119,7 @@ static void free_block(struct kmem_cache *cachep, void **objpp, int nr_objects,
 		if (slabp->inuse == 0) {
 			if (l3->free_objects > l3->free_limit) {
 				l3->free_objects -= cachep->num;
-				/*
-				 * It is safe to drop the lock. The slab is
-				 * no longer linked to the cache. cachep
-				 * cannot disappear - we are using it and
-				 * all destruction of caches must be
-				 * serialized properly by the user.
-				 */
-				spin_unlock(&l3->list_lock);
 				slab_destroy(cachep, slabp);
-				spin_lock(&l3->list_lock);
 			} else {
 				list_add(&slabp->list, &l3->slabs_free);
 			}
@@ -3135,7 +3145,7 @@ static void cache_flusharray(struct kmem_cache *cachep, struct array_cache *ac)
 #endif
 	check_irq_off();
 	l3 = cachep->nodelists[node];
-	spin_lock_nested(&l3->list_lock, SINGLE_DEPTH_NESTING);
+	spin_lock(&l3->list_lock);
 	if (l3->shared) {
 		struct array_cache *shared_array = l3->shared;
 		int max = shared_array->limit - shared_array->avail;
@@ -3178,14 +3188,14 @@ free_done:
  * Release an obj back to its cache. If the obj has a constructed state, it must
  * be in this state _before_ it is released.  Called with disabled ints.
  */
-static void __cache_free(struct kmem_cache *cachep, void *objp, int nesting)
+static inline void __cache_free(struct kmem_cache *cachep, void *objp)
 {
 	struct array_cache *ac = cpu_cache_get(cachep);
 
 	check_irq_off();
 	objp = cache_free_debugcheck(cachep, objp, __builtin_return_address(0));
 
-	if (cache_free_alien(cachep, objp, nesting))
+	if (cache_free_alien(cachep, objp))
 		return;
 
 	if (likely(ac->avail < ac->limit)) {
@@ -3214,7 +3224,7 @@ void *kmem_cache_alloc(struct kmem_cache *cachep, gfp_t flags)
 EXPORT_SYMBOL(kmem_cache_alloc);
 
 /**
- * kmem_cache_alloc - Allocate an object. The memory is set to zero.
+ * kmem_cache_zalloc - Allocate an object. The memory is set to zero.
  * @cache: The cache to allocate from.
  * @flags: See kmalloc().
  *
@@ -3424,7 +3434,7 @@ void kmem_cache_free(struct kmem_cache *cachep, void *objp)
 	BUG_ON(virt_to_cache(objp) != cachep);
 
 	local_irq_save(flags);
-	__cache_free(cachep, objp, 0);
+	__cache_free(cachep, objp);
 	local_irq_restore(flags);
 }
 EXPORT_SYMBOL(kmem_cache_free);
@@ -3449,7 +3459,7 @@ void kfree(const void *objp)
 	kfree_debugcheck(objp);
 	c = virt_to_cache(objp);
 	debug_check_no_locks_freed(objp, obj_size(c));
-	__cache_free(c, (void *)objp, 0);
+	__cache_free(c, (void *)objp);
 	local_irq_restore(flags);
 }
 EXPORT_SYMBOL(kfree);
diff --git a/mm/swap.c b/mm/swap.c
index 8fd095c4ae5..687686a61f7 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -54,6 +54,26 @@ void put_page(struct page *page)
 }
 EXPORT_SYMBOL(put_page);
 
+/**
+ * put_pages_list(): release a list of pages
+ *
+ * Release a list of pages which are strung together on page.lru.  Currently
+ * used by read_cache_pages() and related error recovery code.
+ *
+ * @pages: list of pages threaded on page->lru
+ */
+void put_pages_list(struct list_head *pages)
+{
+	while (!list_empty(pages)) {
+		struct page *victim;
+
+		victim = list_entry(pages->prev, struct page, lru);
+		list_del(&victim->lru);
+		page_cache_release(victim);
+	}
+}
+EXPORT_SYMBOL(put_pages_list);
+
 /*
  * Writeback is about to end against a page which has been marked for immediate
  * reclaim.  If it still appears to be reclaimable, move it to the tail of the
diff --git a/mm/swapfile.c b/mm/swapfile.c
index e70d6c6d6fe..f1f5ec78378 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -442,11 +442,12 @@ int swap_type_of(dev_t device)
 
 		if (!(swap_info[i].flags & SWP_WRITEOK))
 			continue;
+
 		if (!device) {
 			spin_unlock(&swap_lock);
 			return i;
 		}
-		inode = swap_info->swap_file->f_dentry->d_inode;
+		inode = swap_info[i].swap_file->f_dentry->d_inode;
 		if (S_ISBLK(inode->i_mode) &&
 		    device == MKDEV(imajor(inode), iminor(inode))) {
 			spin_unlock(&swap_lock);
diff --git a/mm/truncate.c b/mm/truncate.c
index cf1b015df4a..c6ab55ec688 100644
--- a/mm/truncate.c
+++ b/mm/truncate.c
@@ -68,10 +68,10 @@ invalidate_complete_page(struct address_space *mapping, struct page *page)
 		return 0;
 
 	write_lock_irq(&mapping->tree_lock);
-	if (PageDirty(page)) {
-		write_unlock_irq(&mapping->tree_lock);
-		return 0;
-	}
+	if (PageDirty(page))
+		goto failed;
+	if (page_count(page) != 2)	/* caller's ref + pagecache ref */
+		goto failed;
 
 	BUG_ON(PagePrivate(page));
 	__remove_from_page_cache(page);
@@ -79,6 +79,9 @@ invalidate_complete_page(struct address_space *mapping, struct page *page)
 	ClearPageUptodate(page);
 	page_cache_release(page);	/* pagecache ref */
 	return 1;
+failed:
+	write_unlock_irq(&mapping->tree_lock);
+	return 0;
 }
 
 /**
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 7b450798b45..266162d2ba2 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -340,7 +340,7 @@ void __vunmap(void *addr, int deallocate_pages)
 			__free_page(area->pages[i]);
 		}
 
-		if (area->nr_pages > PAGE_SIZE/sizeof(struct page *))
+		if (area->flags & VM_VPAGES)
 			vfree(area->pages);
 		else
 			kfree(area->pages);
@@ -427,9 +427,10 @@ void *__vmalloc_area_node(struct vm_struct *area, gfp_t gfp_mask,
 
 	area->nr_pages = nr_pages;
 	/* Please note that the recursion is strictly bounded. */
-	if (array_size > PAGE_SIZE)
+	if (array_size > PAGE_SIZE) {
 		pages = __vmalloc_node(array_size, gfp_mask, PAGE_KERNEL, node);
-	else
+		area->flags |= VM_VPAGES;
+	} else
 		pages = kmalloc_node(array_size, (gfp_mask & ~__GFP_HIGHMEM), node);
 	area->pages = pages;
 	if (!area->pages) {
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 73b83d67bab..c1b5f4106b3 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -12,6 +12,7 @@
 #include <linux/config.h>
 #include <linux/mm.h>
 #include <linux/module.h>
+#include <linux/cpu.h>
 
 void __get_zone_counts(unsigned long *active, unsigned long *inactive,
 			unsigned long *free, struct pglist_data *pgdat)
@@ -81,6 +82,7 @@ void all_vm_events(unsigned long *ret)
 {
 	sum_vm_events(ret, &cpu_online_map);
 }
+EXPORT_SYMBOL_GPL(all_vm_events);
 
 #ifdef CONFIG_HOTPLUG
 /*
@@ -113,17 +115,72 @@ EXPORT_SYMBOL(vm_stat);
 
 #ifdef CONFIG_SMP
 
-#define STAT_THRESHOLD 32
+static int calculate_threshold(struct zone *zone)
+{
+	int threshold;
+	int mem;	/* memory in 128 MB units */
+
+	/*
+	 * The threshold scales with the number of processors and the amount
+	 * of memory per zone. More memory means that we can defer updates for
+	 * longer, more processors could lead to more contention.
+ 	 * fls() is used to have a cheap way of logarithmic scaling.
+	 *
+	 * Some sample thresholds:
+	 *
+	 * Threshold	Processors	(fls)	Zonesize	fls(mem+1)
+	 * ------------------------------------------------------------------
+	 * 8		1		1	0.9-1 GB	4
+	 * 16		2		2	0.9-1 GB	4
+	 * 20 		2		2	1-2 GB		5
+	 * 24		2		2	2-4 GB		6
+	 * 28		2		2	4-8 GB		7
+	 * 32		2		2	8-16 GB		8
+	 * 4		2		2	<128M		1
+	 * 30		4		3	2-4 GB		5
+	 * 48		4		3	8-16 GB		8
+	 * 32		8		4	1-2 GB		4
+	 * 32		8		4	0.9-1GB		4
+	 * 10		16		5	<128M		1
+	 * 40		16		5	900M		4
+	 * 70		64		7	2-4 GB		5
+	 * 84		64		7	4-8 GB		6
+	 * 108		512		9	4-8 GB		6
+	 * 125		1024		10	8-16 GB		8
+	 * 125		1024		10	16-32 GB	9
+	 */
+
+	mem = zone->present_pages >> (27 - PAGE_SHIFT);
+
+	threshold = 2 * fls(num_online_cpus()) * (1 + fls(mem));
+
+	/*
+	 * Maximum threshold is 125
+	 */
+	threshold = min(125, threshold);
+
+	return threshold;
+}
 
 /*
- * Determine pointer to currently valid differential byte given a zone and
- * the item number.
- *
- * Preemption must be off
+ * Refresh the thresholds for each zone.
  */
-static inline s8 *diff_pointer(struct zone *zone, enum zone_stat_item item)
+static void refresh_zone_stat_thresholds(void)
 {
-	return &zone_pcp(zone, smp_processor_id())->vm_stat_diff[item];
+	struct zone *zone;
+	int cpu;
+	int threshold;
+
+	for_each_zone(zone) {
+
+		if (!zone->present_pages)
+			continue;
+
+		threshold = calculate_threshold(zone);
+
+		for_each_online_cpu(cpu)
+			zone_pcp(zone, cpu)->stat_threshold = threshold;
+	}
 }
 
 /*
@@ -132,17 +189,16 @@ static inline s8 *diff_pointer(struct zone *zone, enum zone_stat_item item)
 void __mod_zone_page_state(struct zone *zone, enum zone_stat_item item,
 				int delta)
 {
-	s8 *p;
+	struct per_cpu_pageset *pcp = zone_pcp(zone, smp_processor_id());
+	s8 *p = pcp->vm_stat_diff + item;
 	long x;
 
-	p = diff_pointer(zone, item);
 	x = delta + *p;
 
-	if (unlikely(x > STAT_THRESHOLD || x < -STAT_THRESHOLD)) {
+	if (unlikely(x > pcp->stat_threshold || x < -pcp->stat_threshold)) {
 		zone_page_state_add(x, zone, item);
 		x = 0;
 	}
-
 	*p = x;
 }
 EXPORT_SYMBOL(__mod_zone_page_state);
@@ -171,10 +227,12 @@ EXPORT_SYMBOL(mod_zone_page_state);
  * No overflow check is necessary and therefore the differential can be
  * incremented or decremented in place which may allow the compilers to
  * generate better code.
- *
  * The increment or decrement is known and therefore one boundary check can
  * be omitted.
  *
+ * NOTE: These functions are very performance sensitive. Change only
+ * with care.
+ *
  * Some processors have inc/dec instructions that are atomic vs an interrupt.
  * However, the code must first determine the differential location in a zone
  * based on the processor number and then inc/dec the counter. There is no
@@ -184,13 +242,16 @@ EXPORT_SYMBOL(mod_zone_page_state);
  */
 static void __inc_zone_state(struct zone *zone, enum zone_stat_item item)
 {
-	s8 *p = diff_pointer(zone, item);
+	struct per_cpu_pageset *pcp = zone_pcp(zone, smp_processor_id());
+	s8 *p = pcp->vm_stat_diff + item;
 
 	(*p)++;
 
-	if (unlikely(*p > STAT_THRESHOLD)) {
-		zone_page_state_add(*p, zone, item);
-		*p = 0;
+	if (unlikely(*p > pcp->stat_threshold)) {
+		int overstep = pcp->stat_threshold / 2;
+
+		zone_page_state_add(*p + overstep, zone, item);
+		*p = -overstep;
 	}
 }
 
@@ -203,13 +264,16 @@ EXPORT_SYMBOL(__inc_zone_page_state);
 void __dec_zone_page_state(struct page *page, enum zone_stat_item item)
 {
 	struct zone *zone = page_zone(page);
-	s8 *p = diff_pointer(zone, item);
+	struct per_cpu_pageset *pcp = zone_pcp(zone, smp_processor_id());
+	s8 *p = pcp->vm_stat_diff + item;
 
 	(*p)--;
 
-	if (unlikely(*p < -STAT_THRESHOLD)) {
-		zone_page_state_add(*p, zone, item);
-		*p = 0;
+	if (unlikely(*p < - pcp->stat_threshold)) {
+		int overstep = pcp->stat_threshold / 2;
+
+		zone_page_state_add(*p - overstep, zone, item);
+		*p = overstep;
 	}
 }
 EXPORT_SYMBOL(__dec_zone_page_state);
@@ -238,19 +302,9 @@ EXPORT_SYMBOL(inc_zone_page_state);
 void dec_zone_page_state(struct page *page, enum zone_stat_item item)
 {
 	unsigned long flags;
-	struct zone *zone;
-	s8 *p;
 
-	zone = page_zone(page);
 	local_irq_save(flags);
-	p = diff_pointer(zone, item);
-
-	(*p)--;
-
-	if (unlikely(*p < -STAT_THRESHOLD)) {
-		zone_page_state_add(*p, zone, item);
-		*p = 0;
-	}
+	__dec_zone_page_state(page, item);
 	local_irq_restore(flags);
 }
 EXPORT_SYMBOL(dec_zone_page_state);
@@ -524,6 +578,10 @@ static int zoneinfo_show(struct seq_file *m, void *arg)
 					   pageset->pcp[j].high,
 					   pageset->pcp[j].batch);
 			}
+#ifdef CONFIG_SMP
+			seq_printf(m, "\n  vm stats threshold: %d",
+					pageset->stat_threshold);
+#endif
 		}
 		seq_printf(m,
 			   "\n  all_unreclaimable: %u"
@@ -612,3 +670,35 @@ struct seq_operations vmstat_op = {
 
 #endif /* CONFIG_PROC_FS */
 
+#ifdef CONFIG_SMP
+/*
+ * Use the cpu notifier to insure that the thresholds are recalculated
+ * when necessary.
+ */
+static int __cpuinit vmstat_cpuup_callback(struct notifier_block *nfb,
+		unsigned long action,
+		void *hcpu)
+{
+	switch (action) {
+		case CPU_UP_PREPARE:
+		case CPU_UP_CANCELED:
+		case CPU_DEAD:
+			refresh_zone_stat_thresholds();
+			break;
+		default:
+			break;
+	}
+	return NOTIFY_OK;
+}
+
+static struct notifier_block __cpuinitdata vmstat_notifier =
+	{ &vmstat_cpuup_callback, NULL, 0 };
+
+int __init setup_vmstat(void)
+{
+	refresh_zone_stat_thresholds();
+	register_cpu_notifier(&vmstat_notifier);
+	return 0;
+}
+module_init(setup_vmstat)
+#endif