8 files changed, 55 insertions, 41 deletions

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 78ad38ddd01f..dbe1aabf96cd 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -2122,10 +2122,8 @@ static void btrfs_read_locked_inode(struct inode *inode)
          * any xattrs or acls
          */
         maybe_acls = acls_after_inode_item(leaf, path->slots[0], inode->i_ino);
-        if (!maybe_acls) {
-                inode->i_acl = NULL;
-                inode->i_default_acl = NULL;
-        }
+        if (!maybe_acls)
+                cache_no_acl(inode);
 
         BTRFS_I(inode)->block_group = btrfs_find_block_group(root, 0,
                                                 alloc_group_block, 0);
diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c
index edd2ad6416d8..8fcb6239218e 100644
--- a/fs/jffs2/acl.c
+++ b/fs/jffs2/acl.c
@@ -284,8 +284,7 @@ int jffs2_init_acl_pre(struct inode *dir_i, struct inode *inode, int *i_mode)
         struct posix_acl *acl, *clone;
         int rc;
 
-        inode->i_default_acl = NULL;
-        inode->i_acl = NULL;
+        cache_no_acl(inode);
 
         if (S_ISLNK(*i_mode))
                 return 0;       /* Symlink always has no-ACL */
diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c
index f272bf032e1e..91fa3ad6e8c2 100644
--- a/fs/jfs/acl.c
+++ b/fs/jfs/acl.c
@@ -118,15 +118,16 @@ out:
 
 static int jfs_check_acl(struct inode *inode, int mask)
 {
-        if (inode->i_acl == ACL_NOT_CACHED) {
-                struct posix_acl *acl = jfs_get_acl(inode, ACL_TYPE_ACCESS);
-                if (IS_ERR(acl))
-                        return PTR_ERR(acl);
+        struct posix_acl *acl = jfs_get_acl(inode, ACL_TYPE_ACCESS);
+
+        if (IS_ERR(acl))
+                return PTR_ERR(acl);
+        if (acl) {
+                int error = posix_acl_permission(inode, acl, mask);
                 posix_acl_release(acl);
+                return error;
         }
 
-        if (inode->i_acl)
-                return posix_acl_permission(inode, inode->i_acl, mask);
         return -EAGAIN;
 }
 
diff --git a/include/linux/posix_acl.h b/include/linux/posix_acl.h
index 0cdba01b7756..065a3652a3ea 100644
--- a/include/linux/posix_acl.h
+++ b/include/linux/posix_acl.h
@@ -83,6 +83,7 @@ extern int posix_acl_chmod_masq(struct posix_acl *, mode_t);
 extern struct posix_acl *get_posix_acl(struct inode *, int);
 extern int set_posix_acl(struct inode *, int, struct posix_acl *);
 
+#ifdef CONFIG_FS_POSIX_ACL
 static inline struct posix_acl *get_cached_acl(struct inode *inode, int type)
 {
         struct posix_acl **p, *acl;
@@ -146,5 +147,14 @@ static inline void forget_cached_acl(struct inode *inode, int type)
         if (old != ACL_NOT_CACHED)
                 posix_acl_release(old);
 }
+#endif
+
+static inline void cache_no_acl(struct inode *inode)
+{
+#ifdef CONFIG_FS_POSIX_ACL
+        inode->i_acl = NULL;
+        inode->i_default_acl = NULL;
+#endif
+}
 
 #endif  /* __LINUX_POSIX_ACL_H */
diff --git a/kernel/futex.c b/kernel/futex.c
index 80b5ce716596..1c337112335c 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -284,6 +284,25 @@ void put_futex_key(int fshared, union futex_key *key)
         drop_futex_key_refs(key);
 }
 
+/*
+ * fault_in_user_writeable - fault in user address and verify RW access
+ * @uaddr:      pointer to faulting user space address
+ *
+ * Slow path to fixup the fault we just took in the atomic write
+ * access to @uaddr.
+ *
+ * We have no generic implementation of a non destructive write to the
+ * user address. We know that we faulted in the atomic pagefault
+ * disabled section so we can as well avoid the #PF overhead by
+ * calling get_user_pages() right away.
+ */
+static int fault_in_user_writeable(u32 __user *uaddr)
+{
+        int ret = get_user_pages(current, current->mm, (unsigned long)uaddr,
+                                 sizeof(*uaddr), 1, 0, NULL, NULL);
+        return ret < 0 ? ret : 0;
+}
+
 /**
  * futex_top_waiter() - Return the highest priority waiter on a futex
  * @hb:     the hash bucket the futex_q's reside in
@@ -896,7 +915,6 @@ retry:
 retry_private:
         op_ret = futex_atomic_op_inuser(op, uaddr2);
         if (unlikely(op_ret < 0)) {
-                u32 dummy;
 
                 double_unlock_hb(hb1, hb2);
 
@@ -914,7 +932,7 @@ retry_private:
                         goto out_put_keys;
                 }
 
-                ret = get_user(dummy, uaddr2);
+                ret = fault_in_user_writeable(uaddr2);
                 if (ret)
                         goto out_put_keys;
 
@@ -1204,7 +1222,7 @@ retry_private:
                         double_unlock_hb(hb1, hb2);
                         put_futex_key(fshared, &key2);
                         put_futex_key(fshared, &key1);
-                        ret = get_user(curval2, uaddr2);
+                        ret = fault_in_user_writeable(uaddr2);
                         if (!ret)
                                 goto retry;
                         goto out;
@@ -1482,7 +1500,7 @@ retry:
 handle_fault:
         spin_unlock(q->lock_ptr);
 
-        ret = get_user(uval, uaddr);
+        ret = fault_in_user_writeable(uaddr);
 
         spin_lock(q->lock_ptr);
 
@@ -1807,7 +1825,6 @@ static int futex_lock_pi(u32 __user *uaddr, int fshared,
 {
         struct hrtimer_sleeper timeout, *to = NULL;
         struct futex_hash_bucket *hb;
-        u32 uval;
         struct futex_q q;
         int res, ret;
 
@@ -1909,16 +1926,9 @@ out:
         return ret != -EINTR ? ret : -ERESTARTNOINTR;
 
 uaddr_faulted:
-        /*
-         * We have to r/w  *(int __user *)uaddr, and we have to modify it
-         * atomically.  Therefore, if we continue to fault after get_user()
-         * below, we need to handle the fault ourselves, while still holding
-         * the mmap_sem.  This can occur if the uaddr is under contention as
-         * we have to drop the mmap_sem in order to call get_user().
-         */
         queue_unlock(&q, hb);
 
-        ret = get_user(uval, uaddr);
+        ret = fault_in_user_writeable(uaddr);
         if (ret)
                 goto out_put_key;
 
@@ -2013,17 +2023,10 @@ out:
         return ret;
 
 pi_faulted:
-        /*
-         * We have to r/w  *(int __user *)uaddr, and we have to modify it
-         * atomically.  Therefore, if we continue to fault after get_user()
-         * below, we need to handle the fault ourselves, while still holding
-         * the mmap_sem.  This can occur if the uaddr is under contention as
-         * we have to drop the mmap_sem in order to call get_user().
-         */
         spin_unlock(&hb->lock);
         put_futex_key(fshared, &key);
 
-        ret = get_user(uval, uaddr);
+        ret = fault_in_user_writeable(uaddr);
         if (!ret)
                 goto retry;
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index aecc9cdfdfce..5d714f8fb303 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1153,10 +1153,10 @@ again:
                          * properly detect and handle allocation failures.
                          *
                          * We most definitely don't want callers attempting to
-                         * allocate greater than single-page units with
+                         * allocate greater than order-1 page units with
                          * __GFP_NOFAIL.
                          */
-                        WARN_ON_ONCE(order > 0);
+                        WARN_ON_ONCE(order > 1);
                 }
                 spin_lock_irqsave(&zone->lock, flags);
                 page = __rmqueue(zone, order, migratetype);
diff --git a/mm/shmem.c b/mm/shmem.c
index 5f2019fc7895..d713239ce2ce 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -1558,6 +1558,7 @@ static struct inode *shmem_get_inode(struct super_block *sb, int mode,
                 spin_lock_init(&info->lock);
                 info->flags = flags & VM_NORESERVE;
                 INIT_LIST_HEAD(&info->swaplist);
+                cache_no_acl(inode);
 
                 switch (mode & S_IFMT) {
                 default:
@@ -2379,10 +2380,6 @@ static struct inode *shmem_alloc_inode(struct super_block *sb)
         p = (struct shmem_inode_info *)kmem_cache_alloc(shmem_inode_cachep, GFP_KERNEL);
         if (!p)
                 return NULL;
-#ifdef CONFIG_TMPFS_POSIX_ACL
-        p->vfs_inode.i_acl = NULL;
-        p->vfs_inode.i_default_acl = NULL;
-#endif
         return &p->vfs_inode;
 }
 
diff --git a/mm/slub.c b/mm/slub.c
index ce62b770e2fc..819f056b39c6 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1085,11 +1085,17 @@ static struct page *allocate_slab(struct kmem_cache *s, gfp_t flags, int node)
 {
         struct page *page;
         struct kmem_cache_order_objects oo = s->oo;
+        gfp_t alloc_gfp;
 
         flags |= s->allocflags;
 
-        page = alloc_slab_page(flags | __GFP_NOWARN | __GFP_NORETRY, node,
-                                                                        oo);
+        /*
+         * Let the initial higher-order allocation fail under memory pressure
+         * so we fall-back to the minimum order allocation.
+         */
+        alloc_gfp = (flags | __GFP_NOWARN | __GFP_NORETRY) & ~__GFP_NOFAIL;
+
+        page = alloc_slab_page(alloc_gfp, node, oo);
         if (unlikely(!page)) {
                 oo = s->min;
                 /*