Merge commit 'v2.6.36.2' into linux-tegra-2.6.36

10 files changed, 72 insertions, 47 deletions

diff --git a/mm/filemap.c b/mm/filemap.c
index 3d4df44e4221..9701a501f769 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -631,7 +631,9 @@ repeat:
 	pagep = radix_tree_lookup_slot(&mapping->page_tree, offset);
 	if (pagep) {
 		page = radix_tree_deref_slot(pagep);
-		if (unlikely(!page || page == RADIX_TREE_RETRY))
+		if (unlikely(!page))
+			goto out;
+		if (radix_tree_deref_retry(page))
 			goto repeat;
 
 		if (!page_cache_get_speculative(page))
@@ -647,6 +649,7 @@ repeat:
 			goto repeat;
 		}
 	}
+out:
 	rcu_read_unlock();
 
 	return page;
@@ -764,12 +767,11 @@ repeat:
 		page = radix_tree_deref_slot((void **)pages[i]);
 		if (unlikely(!page))
 			continue;
-		/*
-		 * this can only trigger if nr_found == 1, making livelock
-		 * a non issue.
-		 */
-		if (unlikely(page == RADIX_TREE_RETRY))
+		if (radix_tree_deref_retry(page)) {
+			if (ret)
+				start = pages[ret-1]->index;
 			goto restart;
+		}
 
 		if (!page_cache_get_speculative(page))
 			goto repeat;
@@ -817,11 +819,7 @@ repeat:
 		page = radix_tree_deref_slot((void **)pages[i]);
 		if (unlikely(!page))
 			continue;
-		/*
-		 * this can only trigger if nr_found == 1, making livelock
-		 * a non issue.
-		 */
-		if (unlikely(page == RADIX_TREE_RETRY))
+		if (radix_tree_deref_retry(page))
 			goto restart;
 
 		if (page->mapping == NULL || page->index != index)
@@ -874,11 +872,7 @@ repeat:
 		page = radix_tree_deref_slot((void **)pages[i]);
 		if (unlikely(!page))
 			continue;
-		/*
-		 * this can only trigger if nr_found == 1, making livelock
-		 * a non issue.
-		 */
-		if (unlikely(page == RADIX_TREE_RETRY))
+		if (radix_tree_deref_retry(page))
 			goto restart;
 
 		if (!page_cache_get_speculative(page))
@@ -1016,6 +1010,9 @@ find_page:
 				goto page_not_up_to_date;
 			if (!trylock_page(page))
 				goto page_not_up_to_date;
+			/* Did it get truncated before we got the lock? */
+			if (!page->mapping)
+				goto page_not_up_to_date_locked;
 			if (!mapping->a_ops->is_partially_uptodate(page,
 								desc, offset))
 				goto page_not_up_to_date_locked;
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index c03273807182..2697806746d0 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2380,8 +2380,11 @@ retry_avoidcopy:
 	 * When the original hugepage is shared one, it does not have
 	 * anon_vma prepared.
 	 */
-	if (unlikely(anon_vma_prepare(vma)))
+	if (unlikely(anon_vma_prepare(vma))) {
+		/* Caller expects lock to be held */
+		spin_lock(&mm->page_table_lock);
 		return VM_FAULT_OOM;
+	}
 
 	copy_huge_page(new_page, old_page, address, vma);
 	__SetPageUptodate(new_page);
@@ -2665,7 +2668,8 @@ out_page_table_lock:
 		unlock_page(pagecache_page);
 		put_page(pagecache_page);
 	}
-	unlock_page(page);
+	if (page != pagecache_page)
+		unlock_page(page);
 
 out_mutex:
 	mutex_unlock(&hugetlb_instantiation_mutex);
diff --git a/mm/internal.h b/mm/internal.h
index 6a697bb97fc5..dedb0aff673f 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -62,7 +62,7 @@ extern bool is_free_buddy_page(struct page *page);
  */
 static inline unsigned long page_order(struct page *page)
 {
-	VM_BUG_ON(!PageBuddy(page));
+	/* PageBuddy() must be checked by the caller */
 	return page_private(page);
 }
 
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 9be3cf8a5da4..e6aadd65a71a 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -269,13 +269,14 @@ enum move_type {
 
 /* "mc" and its members are protected by cgroup_mutex */
 static struct move_charge_struct {
-	spinlock_t	  lock; /* for from, to, moving_task */
+	spinlock_t	  lock; /* for from, to */
 	struct mem_cgroup *from;
 	struct mem_cgroup *to;
 	unsigned long precharge;
 	unsigned long moved_charge;
 	unsigned long moved_swap;
 	struct task_struct *moving_task;	/* a task moving charges */
+	struct mm_struct *mm;
 	wait_queue_head_t waitq;		/* a waitq for other context */
 } mc = {
 	.lock = __SPIN_LOCK_UNLOCKED(mc.lock),
@@ -4445,7 +4446,7 @@ static unsigned long mem_cgroup_count_precharge(struct mm_struct *mm)
 	unsigned long precharge;
 	struct vm_area_struct *vma;
 
-	down_read(&mm->mmap_sem);
+	/* We've already held the mmap_sem */
 	for (vma = mm->mmap; vma; vma = vma->vm_next) {
 		struct mm_walk mem_cgroup_count_precharge_walk = {
 			.pmd_entry = mem_cgroup_count_precharge_pte_range,
@@ -4457,7 +4458,6 @@ static unsigned long mem_cgroup_count_precharge(struct mm_struct *mm)
 		walk_page_range(vma->vm_start, vma->vm_end,
 					&mem_cgroup_count_precharge_walk);
 	}
-	up_read(&mm->mmap_sem);
 
 	precharge = mc.precharge;
 	mc.precharge = 0;
@@ -4508,11 +4508,16 @@ static void mem_cgroup_clear_mc(void)
 
 		mc.moved_swap = 0;
 	}
+	if (mc.mm) {
+		up_read(&mc.mm->mmap_sem);
+		mmput(mc.mm);
+	}
 	spin_lock(&mc.lock);
 	mc.from = NULL;
 	mc.to = NULL;
-	mc.moving_task = NULL;
 	spin_unlock(&mc.lock);
+	mc.moving_task = NULL;
+	mc.mm = NULL;
 	memcg_oom_recover(from);
 	memcg_oom_recover(to);
 	wake_up_all(&mc.waitq);
@@ -4537,26 +4542,37 @@ static int mem_cgroup_can_attach(struct cgroup_subsys *ss,
 			return 0;
 		/* We move charges only when we move a owner of the mm */
 		if (mm->owner == p) {
+			/*
+			 * We do all the move charge works under one mmap_sem to
+			 * avoid deadlock with down_write(&mmap_sem)
+			 * -> try_charge() -> if (mc.moving_task) -> sleep.
+			 */
+			down_read(&mm->mmap_sem);
+
 			VM_BUG_ON(mc.from);
 			VM_BUG_ON(mc.to);
 			VM_BUG_ON(mc.precharge);
 			VM_BUG_ON(mc.moved_charge);
 			VM_BUG_ON(mc.moved_swap);
 			VM_BUG_ON(mc.moving_task);
+			VM_BUG_ON(mc.mm);
+
 			spin_lock(&mc.lock);
 			mc.from = from;
 			mc.to = mem;
 			mc.precharge = 0;
 			mc.moved_charge = 0;
 			mc.moved_swap = 0;
-			mc.moving_task = current;
 			spin_unlock(&mc.lock);
+			mc.moving_task = current;
+			mc.mm = mm;
 
 			ret = mem_cgroup_precharge_mc(mm);
 			if (ret)
 				mem_cgroup_clear_mc();
-		}
-		mmput(mm);
+			/* We call up_read() and mmput() in clear_mc(). */
+		} else
+			mmput(mm);
 	}
 	return ret;
 }
@@ -4644,7 +4660,7 @@ static void mem_cgroup_move_charge(struct mm_struct *mm)
 	struct vm_area_struct *vma;
 
 	lru_add_drain_all();
-	down_read(&mm->mmap_sem);
+	/* We've already held the mmap_sem */
 	for (vma = mm->mmap; vma; vma = vma->vm_next) {
 		int ret;
 		struct mm_walk mem_cgroup_move_charge_walk = {
@@ -4663,7 +4679,6 @@ static void mem_cgroup_move_charge(struct mm_struct *mm)
 			 */
 			break;
 	}
-	up_read(&mm->mmap_sem);
 }
 
 static void mem_cgroup_move_task(struct cgroup_subsys *ss,
@@ -4672,17 +4687,11 @@ static void mem_cgroup_move_task(struct cgroup_subsys *ss,
 				struct task_struct *p,
 				bool threadgroup)
 {
-	struct mm_struct *mm;
-
-	if (!mc.to)
+	if (!mc.mm)
 		/* no need to move charge */
 		return;
 
-	mm = get_task_mm(p);
-	if (mm) {
-		mem_cgroup_move_charge(mm);
-		mmput(mm);
-	}
+	mem_cgroup_move_charge(mc.mm);
 	mem_cgroup_clear_mc();
 }
 #else	/* !CONFIG_MMU */
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index dd186c1a5d53..6345dfe78d2c 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -659,7 +659,7 @@ static int test_pages_in_a_zone(unsigned long start_pfn, unsigned long end_pfn)
  * Scanning pfn is much easier than scanning lru list.
  * Scan pfn from start to end and Find LRU page.
  */
-int scan_lru_pages(unsigned long start, unsigned long end)
+unsigned long scan_lru_pages(unsigned long start, unsigned long end)
 {
 	unsigned long pfn;
 	struct page *page;
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index f969da5dd8a2..c1002c68d617 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -1588,7 +1588,7 @@ unsigned slab_node(struct mempolicy *policy)
 		(void)first_zones_zonelist(zonelist, highest_zoneidx,
 							&policy->v.nodes,
 							&zone);
-		return zone->node;
+		return zone ? zone->node : numa_node_id();
 	}
 
 	default:
diff --git a/mm/mprotect.c b/mm/mprotect.c
index 2d1bf7cf8851..4c5133873097 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -211,6 +211,7 @@ success:
 	mmu_notifier_invalidate_range_end(mm, start, end);
 	vm_stat_account(mm, oldflags, vma->vm_file, -nrpages);
 	vm_stat_account(mm, newflags, vma->vm_file, nrpages);
+	perf_event_mmap(vma);
 	return 0;
 
 fail:
@@ -299,7 +300,6 @@ SYSCALL_DEFINE3(mprotect, unsigned long, start, size_t, len,
 		error = mprotect_fixup(vma, &prev, nstart, tmp, newflags);
 		if (error)
 			goto out;
-		perf_event_mmap(vma);
 		nstart = tmp;
 
 		if (nstart < prev->vm_end)
diff --git a/mm/nommu.c b/mm/nommu.c
index 88ff091eb07a..acb3bd3c1cb9 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -1668,6 +1668,7 @@ void exit_mmap(struct mm_struct *mm)
 		mm->mmap = vma->vm_next;
 		delete_vma_from_mm(vma);
 		delete_vma(mm, vma);
+		cond_resched();
 	}
 
 	kleave("");
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index f12ad1836abe..2bd6f6da38ea 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -103,19 +103,24 @@ gfp_t gfp_allowed_mask __read_mostly = GFP_BOOT_MASK;
  * only be modified with pm_mutex held, unless the suspend/hibernate code is
  * guaranteed not to run in parallel with that modification).
  */
-void set_gfp_allowed_mask(gfp_t mask)
+
+static gfp_t saved_gfp_mask;
+
+void pm_restore_gfp_mask(void)
 {
 	WARN_ON(!mutex_is_locked(&pm_mutex));
-	gfp_allowed_mask = mask;
+	if (saved_gfp_mask) {
+		gfp_allowed_mask = saved_gfp_mask;
+		saved_gfp_mask = 0;
+	}
 }
 
-gfp_t clear_gfp_allowed_mask(gfp_t mask)
+void pm_restrict_gfp_mask(void)
 {
-	gfp_t ret = gfp_allowed_mask;
-
 	WARN_ON(!mutex_is_locked(&pm_mutex));
-	gfp_allowed_mask &= ~mask;
-	return ret;
+	WARN_ON(saved_gfp_mask);
+	saved_gfp_mask = gfp_allowed_mask;
+	gfp_allowed_mask &= ~GFP_IOFS;
 }
 #endif /* CONFIG_PM_SLEEP */
 
@@ -530,7 +535,7 @@ static inline void __free_one_page(struct page *page,
 	 * so it's less likely to be used soon and more likely to be merged
 	 * as a higher order page
 	 */
-	if ((order < MAX_ORDER-1) && pfn_valid_within(page_to_pfn(buddy))) {
+	if ((order < MAX_ORDER-2) && pfn_valid_within(page_to_pfn(buddy))) {
 		struct page *higher_page, *higher_buddy;
 		combined_idx = __find_combined_index(page_idx, order);
 		higher_page = page + combined_idx - page_idx;
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 6b8889da69a6..d8087f0db507 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -517,6 +517,15 @@ static atomic_t vmap_lazy_nr = ATOMIC_INIT(0);
 static void purge_fragmented_blocks_allcpus(void);
 
 /*
+ * called before a call to iounmap() if the caller wants vm_area_struct's
+ * immediately freed.
+ */
+void set_iounmap_nonlazy(void)
+{
+	atomic_set(&vmap_lazy_nr, lazy_max_pages()+1);
+}
+
+/*
  * Purges all lazily-freed vmap areas.
  *
  * If sync is 0 then don't purge if there is already a purge in progress.